Read Author and Year text version

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model Measures Results and Conclusions Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Pharmacologic Intervention for Opioid Induced Constipation: Methylnaltrexone ­ Recommended for Practice

Portenoy, Thomas, Boatwright, Tran, Galsasso, Stanmbler, Von Gunten, & Israel, 2008 Methylnaltrexone 1 mg, 5 mg, or 12.5 mg subcutaneously. Randomized in ratio of 1:1:1. Sample Characteristics: Adult men and women with advanced illness including cancer patients receiving palliative care and chronic opioid therapy for pain. Subjects: 33 patients (39 screened). Eleven patients discontinued study during double blind phase. 18 patients entered open label phase Mean age: 61 years (range 2087). Mean body weight 64 kg Race: Caucasian (79%) Primary Diagnosis: Cancer (85%) Patients receiving laxatives had to be on a stable regimen for at least 4 days and remain on regimen during study. Laxative at baseline 85%: Stool softeners/emollients (33%) Osmotic agents (27%) Median Time to Laxation: Inclusion: First week of study: Subcutaneously injections on Day 1, 3, and 5. (1) Receiving opioid and stable for two weeks and remain stable for 4 weeks (2) No bowel movements despite conventional laxative therapy (3) Life expectancy at least 4 weeks and stable VS. >48 hours for the 1 mg dose group and 1.72, 0.48, and 6.75 hours in the 5, 12.5, and 20 mg dose groups, respectively. The median time to laxation was 1.26 hrs for all patients dosed 5 Opioid adverse events: graded in terms of severity on a 4 point categorical scale . Primary End Point Laxation response: BM within 4 hours or initial dose. Subjective Outcomes obtained prior to each dose and approximately 3 hours post dose. Constipation Assessment: Severity and distress graded on 5 point categorical scale. Opioid Withdrawal scale: modified Himmelsbach scale Patient Satisfaction: 7 point scale All other dose levels > 1mg : 11 of 23 patients (48%) responded p= 0.05. Blinded Phase Twenty two subjects completed blinded phase of study and fourteen completed open label phase 1. Small sample size less than 100 2. High drop out rate (11 of 22 subjects withdrew during double blind phase of study, 4 declined open label phase, and 4 withdrew from open label phase.

After 22 patients dose range extended to 20 mg. Randomized in ratio of 1:1:3 to 1 mg, 12.5 mg, or 20 mg dose groups

Laxation within 4 hours: Day 1: 1 mg dose 1/10 patients (10%); 5 mg dose 3/ 7 (43%); 12.5 mg dose 6/10 (60%); 20 mg dose: 2/6 (33%); Day 2

Received study medication if no bowel movement for at least two days, and had a score of >3 on a five point scale assessing constipation related distress.

No dose response relationship across the three highest doses compared to the 1 mg dose.

Following first week of double blind study: Patients given option for open

1

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Exclusions: (4) fever or unstable VS (5) elevated liver function test or creatinine (6) Platelet count <50 K (7) History GI obstruction (8) Active peritoneal cancer (e.g.ovarian) (9) History peritoneal catheter (10) History hypersensitivity to methylnaltrexone, naltrexone, or naloxone Investigational agent within last 30 days Negative pregnancy test OPEN LABELED PHASE Response rate between 49% and 64% for doses between 5 and 12. 5 mg. Secondary outcomes not evaluated due to small sample size 1 mg dose group required laxative rescue approximately twice as often as other groups.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

label for max of 3 weeks. Initial dose was 5 mg subcutaneously as often as every other day. Max dose 15 mg in first 22 patients and 20 mg for remaining 11 subjects. Dose could be increased or decreased by investigator.

mg, and was statistically significant compared to the 1 mg group (P <0.0003).

Median time to laxation patients dose >5 mg compared to 1 mg dose group significant (p= 0.0003).

No trend in worsening pain control over time

(11) (12)

Setting Characteristics: Multi-center

Study Design: RCT, parallel-group, repeated dose, dose-ranging trial. Included double blind phase for one week followed by an open-labeled phase for a maximum of three weeks.

Purpose: To assess the efficacy and safety of subcutaneous methylnaltexone in patients with advanced illness and opioid induced constipation (OIC).

Conclusion: Doses > 5 mg in patients with advanced illness relieved opioid induced constipation (OIC) without decreased analgesia or withdrawal symptoms.

2

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

To clarify whether a dose response relationship could be identified. Slatkin, Thomas, Lipman et al, 2009 Methylnaltrexone Double Blind Phase Subjects randomized to a single dose of drug or placebo administered subcutaneously (SC). Groups were: 0.15mg/kg 0.3mg/kg Placebo Design was double blind randomized placebo-controlled single dose study, followed by an open label phase, followed by an open label extension phase to determine safety and efficacy of SC methlynaltrexone in Opiod Induced Constipation (OIC) 154 patients initially randomized Median age = 66 81.2% with cancer diagnosis >65% had WHO performance status of 3 or 4 95% had baseline laxative use. Recruited from hospice and palliative care settings 17 treatment sites Inclusion criteria: Advanced illness, life expectancy 1-6 months Stable opiod regimen at least 3 days prior to study entry No significant laxation 48 hours prior Stable vital signs Efficacy/Primary Outcomes Percent patients who defecated within 4 hours of dose 24 hour rescue-free laxation rates Median time to rescuefree laxation Global Clinical Impression of Change (GCIC) scale ( 7 point rating scale from "much worse" to "much better") Improvement in stool consistency ( patient report, 1 = very hard, to 6 = watery) Constipation distress (5 point scale from 1= none to 5 = very much) Methylnaltrexone 0.15mg/kg group 61.7% laxation within 4 hours 64.4% reported improvement in constipation distress 58.7% reported improvement in GCIC Patients who responded to SC dosing often defecated soon after drug administration, with 50% responding within 30 min. No description of average time to defecation or timing in all of respondents to the medication. Timing of response may be important for planning patient care related to toileting. 19 patients reported as having severe AEs. Not clear when in open label part of study these occurred and what percentage of cases were affected. Investigators noted subjects were fail and that rapid reversal of constipation in such cases can be associated with AEs. Data for AEs in control and study groups not provided. Not clear what impact continued laxative use had on results. Many patients were on multiple types of laxatives as well as the study drug. This study done in patients with short life expectancy and advanced disease. Findings may not be applicable to other populations

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Methylnaltrexone 0.3mg/kg group Placebo 13.5% laxation within 4 hours 34% with improvement in constipation distress 21.6% with improvement in GCIC 58.2% laxation within 4 hours 63.5% with improvement in constipation distress 58.8% with improvement in GCIC

Subjects randomly assigned in 1:1:1 ratio to each study group Baseline laxative regimens could be continued. Rescue laxatives ( laxatives administered on as needed (PRN) basis were allowed except within 4 hours before or after administration of the dose Open Label Phase 28 days with 1 dose per 24 hours PRN. Initial dose 0.15mg/kg, could be decreased to 0.075mg/kg or increased to 0.3mg/kg based on response. Protocol Extension Patients completing open label, could enter 3 month extension. Initial dose was the same as in open label phase, with dosing adjusted to 0.075mg/kg,

Secondary Outcomes Pain ­ 10 point scale from 0 (none) to 10 ( worst possible) Modified Himmelsbach opiod withdrawal scale ( composite summed score for 7 symptoms, each scored on 4 point scale from 1= none to 4= severe)

Comparison between each methlynaltrexone group and placebo were significant ( p.<.00001) Laxation within 4 hours in open label phase ( n = 72) 55.8% in week 1 and 2

Exclusion criteria: Previous methylnaltrexone use

3

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Previous natrexone or naoloxone Process suggestive of gastrointestinal obstruction Non-opiod constipation, diverticular disease , fecal impaction, acute abdomen or peritoneal catheter

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

0.15mg/kg, or 0.3mg/kg by investigator discretion

Adverse events(AE)using National Cancer institute common toxicity scale version 2. World Health Organization (WHO) Performance Status

61.7% in week 3 and 4 63.7% at > 8 weeks

19 patients had AEs. Serious AEs reported by 3 patients in open label phase A few patients experienced abdominal pain and watery stool No change in pain or symptoms of opiod withdrawal 4 hours after dosing In open label phase, the 0.3mg/kg dose was associated with more abdominal pain than the 0.15mg/kg dose Study Conclusions SC Methylnaltrexone is effective in treatment of OIC and generally tolerated well No relationship between dose and laxation response, suggesting that optimal dose I 0.15mg/kg

Double Blind Phase Age: Range 21-100, similar median age across all groups Gender: 54.5% Male, 45.5% Female, similar across groups WHO Performance status: 67% 3 or greater. Oral morphine equivalents: Range 8-33130 0.15mg/kg group ­ 207mg/d 0.3mg/kg group ­ 188mg/d Placebo ­ 150 mg/d No significant differences in age, constipation distress, mean pain score across study groups Baseline laxative use: 95% used laxatives 83% used stimlulants 56% used osmotic laxative

4

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

27% used stool softener Open Label Phase 147 elected open label phase 72 patients completed 36 withdrew and 39 died Extension 27 elected extension 9 completed extension

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Opioid Induced Constipation: Opioid Rotation ­ Likely to be Effective

Ahmedzai, 1997 Comparison of transdermal fentanyl and sustained released morphine in palliative cancer patients measuring efficacy, tolerance and quality of life. Sample Characteristics: Palliative cancer adult patients with a life expectance greater than a month and have required are receiving a strong stable dose of opioids. = N = 202, 110 completed the study, mean age of 61.5 years, 55% male. Setting Characteristics: 38 different palliative care centers in the UK. Study Design: Randomized, open, two-period, crossover study. Patients received one treatment for 15 days then the other for 15 days to compare which provided the greatest efficacy, QOL, and least side effects. Quality of Life assessment according to the World Health Organization (WHO) ­ a performance status scale (measured pre, post and daily during the study) and the European Organization for Research and Treatment Center (EORTC) ­ measuring side effects, convenience, and effects on ADLs/careers (measured after each arm of the study.) Constipation and diarrhea were reported through the EORTC pre study, and days 8, 16, 23, 31. These tools have been validated for advanced cancer patients. Daily Diaries recorded estimated sleep length, night awakening, daytime drowsiness, and use of rescue medications. Memorial Pain Assessment card was used BID for assessment of pain and mood. Skin assessment when removing patches. Null hypothesis was 50% would prefer fentanyl and 50% would prefer morphine. The number one reason patients withdrew from the study was due to adverse events. Bowel Function: Fentanyl was associated with significantly less constipation than morphine (p< 0.001). This was confirmed through multiple assessment mechanisms. The remaining results do not pertain to constipation (see reference for other results. Patients were allowed short acting morphine for breakthrough pain, which could alter the results for fentanyl, but the results were still significant Unethical to have a "washout" period. Some of the dropouts may have been due to morphine withdrawal Not blinded therefore, bias may exist according to preference of delivery mechanism. Fentanyl and morphine doses varied amongst individuals and were titrated, as needed, throughout the study.

5

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Sample characteristics: N = 256 patients (aged 26-82 yrs, mean 51 yo) with chronic noncancer pain (mean duration of chronic pain 9.3 yrs) who had been treated with opioids. 64% of patients had nociceptive pain; 33% had neuropathic pain and 31% had a combination of the two. Over 70% of patients had used morphine prior to the study. Setting Characteristics: 35 specialty pain clinics in Belgium, Canada, Denmark, Finland, the United Kingdom, the Netherlands and South Africa. Study Design: randomized crossover study. Patients were randomized to treatment groups using the central randomization minimization technique. There were no significant differences between the sample characteristics of the two treatment groups. One group was randomized to four weeks of treatment with sustained release oral morphine followed by transdermal fentanyl for four weeks. The second group received the same treatments but in reverse order. Purpose: To test the relationship between opioid dose, bowel function and ADLs in patients with advanced cancer admitted to a hospice. Sample Characteristics: N = 50

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Limitations: Subjects were almost all white. No discussion of strategies used for dose conversion Lack of placebo control and open label nature of study was discussed at length. The study was not blinded (i.e. the study could have been improved by having patients use a placebo of the other drug simultaneously to remove bias)

Allan, Hays, Jenson, LePolain de Waroux , Bolt, Donalds, & Kalso, 2001

Purpose: To compare patients preference for transdermal fentanyl or sustained release oral morphine, their level of pain control, and their quality of life after treatment Interventions: Sustained release oral formulation of morphine (10, 30, 60, 100, or 200 mg tablets) Fentanyl, a lipid soluble synthetic opioid, delivered in a transdermal controlled release formulation (25, 50, 75, or 100 mcg/hr patches).

Outcome measures: Patient preference for transdermal fentanyl or sustained release oral morphine; pain control; quality of life; and safety assessments.(adverse effects) Tools: Pain control was assessed at each visit using a global treatment assessment at the end of each treatment period. QOL assessment (SF-36) and pain intensity (0 being low -100 high) were assessed at baseline and at the end of each treatment period. Bowel function questionnaire (no details provided) Safety evaluation (PE, VS, disease progression) on first visit, crossover, and end of study. Statistical analysis: Primary efficacy variable was analyzed with a binomial test; Differences in personal variables was analyzed with the Van Elteren test and the Cochran-Mantel Haensxel test. The other variables were compared with the Koch nonparametric paired analysis for crossover designs

Final N = 212 and 39 could not be assessed. The number of withdrawals in the first treatment group was 16% in the fentanyl group compared with 9% in the morphine group. This was attributed to the fact that most patients (76%) had taken morphine for six weeks before the study and were accustomed to its side effects and the possibility of incomplete cross-tolerance leading to greater than anticipated potency of the fentanyl. This incomplete crosstolerance may also explain the improved pain control perceived with fentanyl. Preference: 7% (15) expressed no preference; 65% (135) preferred fentanyl vs. 28% (59) who preferred morphine (p<0.001) Results related to constipation showed overall incidence of treatment related adverse events was similar in both groups (7074%) Fentanyl was associated with more nausea (26% vs 18%) than morphine but less constipation (16% vs.22%). Reduced constipation was confirmed by the bowel function questionnaire (29% fentanyl vs 48% morphine; p<0.001) Results: Reduced bowel score (particularly stool frequency) in pts taking opioids but not dosedependent relationship. No relationship between bowel score & physical functioning. For

This study was not designed to evaluate the effect of opioid rotation on constipation. Although there is reference to a bowel function questionnaire, results describing/ comparing laxative use, frequency of bowel movements etc... were not reported.

Bennett & Cresswell, 2003

No intervention was evaluated Bowel care protocol used by the hospital was as follows: All constipated pts. and/or those on opioids were started

Measures: 1. Mean daily opioid dose in prior 3 days (calculated as equivalent daily dose of morphine. Fentanyl 1:150, dilaudid 1:7.5, SQ diamorphone 1:2, codeine 10:1)

Limitations: 1. Small sample size; not randomized; short follow up time 2. Pts on opioids (compared those off opioids) are 10 years younger

6

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

(24 male, 26 female). Mean age 71.4 yrs. Median MBI was 54.8; 15 pts. were not on opioids, 35 were. Exclusions: Patients with duration of admission less than one week, patients with communication difficulties, or lack of supporting information in clinical notes. Setting Characteristics: St, Gemmas Hospice Leeds, England. Study Design: Prospective study. Over a two month period, all patients admitted to St, Gemmas Hospice Leeds, with advanced cancer who had been admitted for at least one week were interviewed about their bowel movements. Purpose: Authors challenged the idea of a dose-dependent relationship between opioids and constipation and suggested that some patients may become tolerant to the constipating effects of these drugs. Radbruch, Sabatowski, Loick, G., Kulbe, Kasper et al., 2000 Intervention: Switching patients from long-acting morphine to fentanyl patches. Fentanyl doses were calculated with a conversion table based on a 100:1 dose ratio. If the calculated fentanyl dose was higher than 2.4mg/day = 100ug/h (more 270 mg/day slow release morphine), more than one patch was used. Sample Characteristics: N = 46 (29 males, 17 females); median age 57.5 yrs (31-83 y.); median wt. 62.5 kg; median ht. 172 cm. Sites of primary ca included GI, head and neck, GU, respiratory, breast and hematologic cancers. Setting Characteristics: June 1995-January 1996, Germany (unclear if in-pt. or out-pt. Researcher based at University of 1. Patient diary recording: intensity on a 0-10 visual analogue scale 3 x day; frequency of breakthrough pain and use of PRN medications; use of laxatives; self-assessment of frequency and consistency of defecation. 2. 30 item Quality of life questionnaire of the European Organization for Research and Treatment of Cancer (EORTC-

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

3. Study assessed pts. after they had been titrated on laxatives 4. Overflow diarrhea might have been inaccurately assessed against the bowel score used in the study 5. Pain severity and prevalence of uncontrolled pain not examined independently.

on combination laxative as follows: 37.5 mg dantron titrated from 1 cap/day up to 3 cap BID. If no response, polyethylene glycol was added starting with 1 packet BID and titrated up to 8 packets/day. Rectal measures (glycerin or bisacodyl suppositories or a stimulant enema) were used if stool was present in the rectum.

2. Mean daily dose of dantron in prior 3 days, 3. Modified Barthel Index (MBI) measures 10 ADL with 5 levels of dependence maximum score is 100 points, representing independence in daily living 4. Bowel score (number of times bowels open and stool consistency) over prior 3 days. Calculated as follows: A= # times bowels opened in last 3 days B= Stool consistency (1= hard, 2= normal, 3= soft) Bowel score = A+B; Constipation = <3

any given dose of opioid, fitter pts were treated with larger doses of laxatives. Conclusions: The authors concluded that it is likely that constipation in advanced cancer patients is more strongly related to other variables i.e. food, fluid intake, abdominal tumor involvement, depression, cognitive impairment, use of antimuscarinic drugs. Clinicians should not base laxative prescribing on opioid dose per se nor degree of dependency but should titrate laxatives according to bowel function. Lower doses of opioids, or weaker opioids, may just as likely cause constipation as higher doses & increasing opioid dose may not be associated with more constipation.

Results: 46 patients treated with slow release morphine, (7 excluded in morphine phase because stable analgesia could not be achieved), 39 switched to transdermal fentanyl. Twentythree completed the study, two patients died from basic disease, 12 excluded for various reasons, and not enough data for evaluation was available for two pts.

Open study, so prejudices from staff or patients may have biased the results. The site of primary tumor was situated in the GI tract for a quarter of the patients, and in these patients constipation may easily have been caused by tumor growth. It may be questioned whether the conversion from morphine to

7

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Cologne. Study Design: Open sequential multi-center study. Patients were treated with oral slow-release morphine for at least 6 days (morphine phase), until they reported stable pain intensity scores of 40 or less on a VAS (0= no pain, 100= worst pain imaginable) for at least 2 days. Analgesic therapy then was switched from oral morphine to transdermal fentanyl (fentanyl phase). Fentanyl patches were changed regularly after 3 days. Fentanyl doses were increased when patients reported inadequate pain relief or had to take more than six rescue medications per day. Study terminated after 30 days of transdermal therapy. Patients who completed the study until day 17 or longer were included in an intra-individual comparison of laxative intake using the Wilcoxon rank test Purpose: Investigated constipation and the use of laxatives in patients with chronic cancer pain treated with oral morphine and transdermal fentanyl

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

fentanyl really was equianalgesic. Thus, less constipation may have been the consequence of lower equianalgesic opioid dosage. Results may have been influenced by the high number of patients who dropped out of the study. The difference in the degree of constipation between the two analgesics regimens should be confirmed in a randomized doubleblind study that takes into account both constipation and use of laxatives. Short acting morphine was used for breakthrough pain in both arms of the study, and as a result the pts on fentanyl also had morphine on board. No standardization of the laxative used, so it is unclear whether this influenced the results. The study was supported by a research grant from a pharmaceutical company.

QOL-C30). Assesses nine symptoms and 6 dimensions of QOL. 3. BP, HR, RR and skin reaction at the site of fentanyl application was documented by the treating physician on day 0,6,12,18,24 and 30

The frequency of bowel movements did not change significantly but the use of laxatives was reduced in 23 of 28 pts on transdermal fentanyl and increased in 2 of 28 patients on transdermal fentanyl. No significant changes in vital signs vital signs noted Mild to moderate skin reaction noted in 5 pts. The EORTC quality-of-life questionnaire symptom scores showed significant decrease for constipation only. Conclusions: The use of laxatives was reduced significantly with transdermal fentanyl.

Pharmacologic Intervention: Mineral Oil - Benefits Balanced With Harm

Gal-Ezer & Shaoul, 2006 Mineral oil 30 ml twice daily initially lowered to 25 ml twice daily Sample Characteristics: 17 Year old girl Study Design: Case Report Presented additional case Serum levels of Vitamin A and E were obtained. Vitamin A and E and calcium, phosphorus, and alkaline phosphatase levels, reflecting vitamin D status, and prothrombin time level, reflecting vitamin K status, were within normal limits. Single case report Non cancer patient

8

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

examples of prolonged use of large doses of mineral oil and its effect on fat-soluble vitamin absorption. 30 ml of mineral oil twice daily with enemas as needed. Result was daily defecation with some encopresis. Dose was lowered to 25 ml twice daily. Patient on her own decided to increase her dosage to 40 ml daily for at least five months, resulting in two soft daily stools.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Mineral oil is limited to patients older than 1 year. Recommended dose of 15­45 ml twice daily; it should be given between meals, preferably in midafternoon and at bedtime so as not to interfere with absorption of fat-soluble vitamins. Baker recommended 1­3ml/kg per day for children. Clark studied 25 children with chronic constipation during mineral oil therapy. He monitored serum beta-carotene, retinol (vitamin A1) and alpha-tocopherol (vitamin E) levels. Monitored patients for as long as four months of therapy. A modest increase in serum retinol levels occurred, especially after three months. The researchers concluded that a short course of mineral oil can induce a reduction in the serum level of beta-carotene but has no effect on serum levels of retinol and alpha-= tocopherol. They found no significant differences in prothrombin time or serum retinol and alpha-tocopherol levels after six years of therapy.

Even massive overdosage of mineral oil for a period as long as five months does not have any significant impact on vitamin ADEK levels in an adolescent. Mineral oil is safe in regular doses; however, in smaller, rapidly growing children, a prolonged overdose of mineral oil might be detrimental.

9

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model Measures Results and Conclusions Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Zanetti, Marchiori, Gasparetto, Escuissato, & Souza 2007.

Safety of mineral oil in the treatment of constipation--a lesson from prolonged overdose

Sample Characteristics: Boys: N = 9 Girls: N = 8 Age range: 2 months to 9 years Inclusion: children who had highresolution computed tomography (CT) scans Setting Characteristics: Three hospitals in Brazil Study Design: Retrospective descriptive study Purpose: To evaluate the highresolution CT findings of 17 children with exogenous lipoid pneumonia following aspiration of mineral oil.

High resolution CT scans assessed for presence, distribution, and extension of the following findings: air-space

consolidation, ground-glass

attenuation, and air-space nodules. Criteria for these findings are defined in the Fleischner Societys Glossary of Terms Images were reviewed by two radiologists. Medical record review to identify risk factors and relationship between severity of symptoms and the time using mineral oil, the time between onset of symptoms and chest radiography and the time between chest radiography and diagnosis.

This was just one case study, but the findings were not what has been believed traditionally. Most common symptoms were cough (N = 13), mild fever (N = 1), and progressive dyspnea (n = 9). The main CT findings were air space consolidation (100%), usually with areas of fatty attenuation (70.6%), predominantly bilaterally in the posterior and lower regions of the lungs. Conclusions: High-resolution CT findings with presence of air space consolidation with fat attenuation and history of mineral oil ingestion are suggestive of lipoid pneumonia.

Retrospective study Small sample size Multi site with different protocols Lacked high resolution CT pathology correlation with parenchymal lesions

Pharmacologic Intervention: Peripheral Opioid Antagonists ­ Benefits Balanced with Harms

Meissner, Schmidt, Hartmann, Kath, & Reinhart, 2000 Purpose: To evaluate the use of oral naloxone to treat opiate associated constipation in cancer patients Sample Characteristics: N = 22 Setting Characteristics: Hospital in Germany Study Design: Controlled study with a control period not a control group. Patients were observed for 6 days without intervention then oral naloxone was titrated as follows: 3 mg x 3 ­ day 1 Measures: 1. Himmelsbach withdrawal scale was used to monitor possible systemic withdrawal signs as shivering/ piloreaction, yawning, perspiration, nausea/vomiting, tremor/ restlessness, and lacrimation/ rhinorhea. Each of these signs were quantified with 0 (none), 1 (mild), 2 (moderate), or 3 (severe). Results: Final N = 17. 1. Nausea, restlessness, & sweating were most common side effects 2. Laxation increased in 14/17, laxative use decreased in 9/17, 3. No difference in pain rating between the periods. 4. Other laxatives used: lactulose, paraffin, raglan, glycerol, castor oil, natruimpicosulfate Limitations: 1. Only six days allocated for measurement periods, 2. Subjective Likert scale 0-3 used 3. No control for other laxative use (all participants were on at least one) 4. Study kept overdosed (via medication error) patient in study

Intervention: Naloxone is a competitive agonist of opiod receptors inside and outside the CNS. After systemic administration, it reverses both

10

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

6 mg x 3 ­ day 2 9 mg x 3 ­ day 3 12 mg x 3 ­ day 4 ­ max Titration stopped with laxation or increased peristalsis.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

5. Diet, exercise, fluid intake were not controlled for. 6. P value was not tested

centrally and peripherally mediated opiod effects

2. Laxation frequency 3. Pain intensity on 11 step pain scale 4. Laxatives received 5. Ratio of morphine dose to naloxone dose

5. Naloxone dose was not based on morphine dose, but on opiod tolerance level

Conclusion: Start with low dose naloxone and titrate up

Pharmacologic Intervention in Adults: Polyethylene Glycol (PEG) with or without electrolyte: PEP Weight of Evidence Category: Effectiveness Not Established

Attar A, Lemann, Ferguson et al., 1999 PEG 3350 is an osmotic laxative which opposes the colons normal drying action on the feces, increasing fecal bulk which causes stretching of the circular muscle fibers in the bowel wall and triggers myogenic peristalsis. PEG plus electrolytes (PEG+E) provides electrolyte depletion and dehydration that can occur with other laxatives. Sample Characteristics: N = 115 Inclusion Criteria: Age 18-90 years old, had at least one of the following two symptoms present for at least 3 months. (1) fewer than three stools a week and (2) difficult evacuation. Patients with secondary constipation; those taking concomitant medication which might modify bowel transit; severe liver, renal or cardiac impairment; those not likely to comply with treatment; those not able to give informed consent and women of childbearing age not using effective contraception were excluded. Measures: The efficacy of PEG+E was evaluated in terms of the number of stools per day, evacuation score (dyschezia index, graded 0-3) and global satisfaction index (analogue scale 0-10). Results: Part A: At the end of the one month period, PEG+E showed consistently and significantly better results in terms of number of stools per day, ease of evacuation, and global satisfaction. Sixty percent of PEG+E recipients did not use all their sachets, compared with 40% in the lactulose group. In the lactulose group, 23% increased their dosage to over 2 sachets per day, while 64% of lactulose recipients used 2-3 sachets per day. The use of suppositories and micro-enemas was significantly lower in the PEG+E group than the lactulose group (16% versus 34%) Part B: Daily dosage of PEG+E remained stable with most patients using 1-2 sachets per day. Lactulose group tended to increase to 3 sachets per day. At the end of 3 months, the efficacy of Well designed study.

Lactulose is metabolized to lactic acid by bacteria in the colon. These bacteria exert a local osmotic effect, drawing water and electrolytes into the colon from the surrounding tissues to bulk feces.

Study Design: Single blind, randomized, multi-center study. Part A: Patients were randomized to receive either PEG+E (n = 60)

11

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

or lactulose (n = 55) for 1 month. Part B: Patients >65 continued on the treatment for another 2 months. Patients <= 65 received PEG+E irrespective of the laxative they received at the start of part A

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

PEG+E (pt >65) was significantly better than that produced by lactulose in terms of number of stools per day and ease of evacuation.

Conclusions: using PEG+E instead of lactulose, doubles the percentage of patients successfully treated at 3 months. PEG+E was found to be a superior treatment compared to lactulose to treat idiopathic constipation. Subjects returned to their study centers after 2,4,6,9, and 12 months of treatment; blood and urine samples were collected, and adverse events were reviewed. At each visit, subjects were queried for ROME constipation criteria, and they rated their overall improvement using a global efficacy scale. N = 184 who completed 12 weeks of therapy 1. 2. No clinically significant changes were found in hematology or blood chemistry, particularly electrolytes, for the study population as a whole or the elderly group. With respect to the Global Efficacy Assessment, depending on the month of observation, 80%-88% of enrolled patients and 84%-94% of the elderly were treated successfully. Similar results were obtained from secondary efficacy measures that assessed individual ROME constipation criteria at each visit. The response to treatment was durable over time. Over the one-year course of study representing 218 patient-years at the labeled dose, medicationassociated adverse effects were gastrointestinal complaints of None identified in the discussion of the results This study was not related to cancer or to opioid-induced constipation.

DiPalma, Cleveland, McGowan, & Herrera, 2006

Polyethylene glycol laxative as a single daily dose of 17 g for 12 months.

Sample Characteristics: 311 patients with chronic constipation including 117 aged 65 and older

184 completed all 12 months of treatment. Study subjects who met defined criteria for chronic constipation were enrolled. Setting Characteristics: Multicenter study including 50 centers in the United States

Study Design: A randomized, open-label, single-treatment study

Purpose: The aim of the study was to extend the safety data of polyethylene glycol used for

12

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

chronic treatment of chronic constipation.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

diarrhea, loose stool, flatulence, and nausea. The effects were generally mild or moderate in severity.

Conclusion: Polyethylene glycol laxative is safe and effective for treating constipation in adult and elderly patients for periods up to 12 months, with no evidence of tachyphylaxis. DiPalma, J.A., Cleveland, M.B., McGowan, J., & Herrera, J.L. (2007b). Polyetheylene glycol 3350 (PEG 3350, MiraLAX) laxative 17 g or placebo administered daily for six months Sample Characteristics: N = 204 treatment group N = 100 placebo group Mean age = 53 years (range = 20 to 92) Older subgroup (65 years or older) N = 75 Females: N = 258 (85%) Race/ethnicity: Caucasian 84%; African American 13%; Hispanic or Latino 6.3% Average duration of constipation: 23 years Inclusion: Three-month history when not taking laxatives of fewer than three bowel movements (BMs) per week and one or more remaining ROME symptom criteria. Average of three or fewer satisfactory BM per week during 14-day observational period after Treatment period: Data collection interactive voice response system daily to report BM experience for day. Answered daily and weekly questions (i.e., number of BMs, straining, lumpy/hard stools, evacuation complete, satisfaction, cramping) on Likert scale 0 to 4 (none to extreme) amount of gas on Likert scale 0 to 4 (no gas to extreme gas) 14-day observational period/ qualification period: Baseline constipation status confirmed N = 609 screened, 300 did not meet criteria or failed 14-day observational period. N = 306 enrolled and randomized; 2 excluded N = 175 completed study Primary analysis based on intention to treat (ITT) Efficacy of primary variable, treatment success defined as relief of modified ROME criteria for constipation for 50% or more of subjects treatment weeks. Calculated success rate for treatment period: PEG group achieved significant benefit (p < 0.001) versus placebo in 11 of 12 primary and secondary measures 1. Focus on chronic constipation; no focus on cancer or opioidinduced constipation

Total sample: 41% difference in treatment response (PEG 52%, placebo 11%, p < 0.001)

13

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

eligibility criteria were met. Exclusions: Loose stools Sufficient criteria for irritable bowel syndrome Currently or previously treated with PEG Setting Characteristics: 50 centers in United States

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Elderly: 46% difference favoring PEG Higher total BMs per week for treatment group

Total compliance with study medication > 86%

Study Design: Double-blind placebo-controlled, parallel randomized, controlled trial

Subject withdrawal equivalent for each reason category except lack of efficacy. Twice as many subjects in the placebo group withdrew secondary to lack of efficacy.

Subjects were randomly assigned in a 2:1 ratio PEG to placebo.

Conclusion: PEG is safe and effective in management of constipation in adults and older adults up to six months.

Subjects mixed study Medication in 8 oz of juice or other beverage

Bisacodyl 10 mg (5 mg tablets) used as rescue medication for severe discomfort related to constipation.

Fiber prohibited. Other nonconstipating medications were allowed. Purpose: To compare the safety

14

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

and efficacy of PEG versus placebo over six-month treatment period in patients with chronic constipation. DiPalma, Cleveland, McGowan, & Herrera, 2007a Polyetheylene glycol 3350 (PEG 3350, MiraLAX) laxative 17 g or placebo administered daily for 28 days N = 100 enrolled and randomized within each site by computer N = 93 completed study Treatment versus control Use of nonconstipating medications were allowed. The use of fiber or other laxatives was not allowed. N = 50 treatment group N = 50 placebo group Mean age = 58 years (range = 31 to 84) Older subgroup (65 years or older) N = 28 Females: N = 74 (74%) Race/ethnicity: Caucasian 84%; African American 16% Average duration of constipation: 18 years Constipating medications: analgesics, anticholinergics, antidepressants, chemotherapy agents, anticonvulsants, anithistamines, and resins Inclusion: Met constipation as defined by ROME II criteria Taking medication with > 3% chance incidence of constipation Exclusions: Allergy or sensitivity to study medication Treatment success: relief of ROME II criteria for constipation during last seven days of treatment period Laboratory evaluation repeated at study conclusion Treatment Period Primary responder for treatment success: 39% difference in favor of PEG (78% PEG vs Placebo 39%, p>0.001) 3. Although sample did include opioid-induced constipation and chemotherapy-induced constipation, such patients were not included in sufficient numbers to provide dependable data. More study is needed. Laboratory testing Analgesics associated constipating med in PEG (= N = 17) and placebo (= N = 15) group. Demographic Baseline Evaluation: History and physical Patient withdrawal equivalent for each reason category except lack of efficacy. Three in placebo group withdrew and none in treatment group 1. Lacked observational period prior to treatment initiation to confirm constipation status 2. Some patients in study experienced constipation prior to receiving constipating medication. It is possible medication-induced constipation may respond differently.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Daily bowel diary including sharing BM experiences and answering questions related to study efficacy and safety criteria

Elderly subgroup had similar results. Findings not statistically significant due to small sample size.

Measures:

Baseline constipation (less than three BMs perweek) evident in 58% (n = 29) of PEG group versus 46% (n = 23) of the placebo group. N = 609 screened, 300 did not meet criteria or failed 14-day observational period.

Total number of BM

Average BM per week final week: 8.1 (PEG) versus 5.4 (placebo) p < 0.001

15

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Prior GI surgery Known or suspected GI obstruction, ileus, heart failure, renal failure, ascites, other known chronic bowel, or cardiopulmonary condition Pregnant or lactating Loose stool present and met criteria for IBS Currently or previously taking PEG Setting Characteristics: Four centers in the United States Conclusion: PEG is safe and effective in treating constipation in patients taking constipating medications. No difference in gas, bloating, or cramps

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Number of patient-reported satisfactory BM

2/3 BM rated complete or satisfactory for both treatments

Symptoms of gas, cramping, straining at stool, stool consistency, and bloating using a 10 cm visual analog scale

Treatment difference for straining and stool consistency statistically significant p < 0.001(PEG better)

Study Design: Double-blind placebo-controlled parallel RCT

Subjects received a 4 L jug containing Crystal Light with or without study medication also added. Reconstituted by unblinded study personnel; investigators remained blinded. Purpose: To compare the safety and efficacy of PEG 3350 (MiraLAX) versus placebo for treatment of constipation in subjects with chronic constipation

Szojda, Mulder, Chris, Felt-

Poor taste is an important factor for non-compliance and

Sample Characteristics: N = 100 healthy volunteers

Ratings of taste on a 5-point scale were obtained for both

The taste score of PEG 4000 (mean = 23.9, SD = 0.7) was

Sample was not random but

16

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

recruited by advertisement Males: N = 27 Females: N = 73 Randomized crossover taste comparison of two commonly used PEG preparations, PEG 4000 and PEG 3350 Mean age: 36 years Age range: 18 to 65 years Inclusion: healthy volunteers Exclusions: not described No difference in gender or age was observed.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

volunteers.

Bersma, & Richelle, 2007

is one of the leading causes of therapy failure.

preparations (1 = very bad taste to 5 = very good taste).

significantly better than for PEG 3350 (mean = 2.7, SD = 0.7) (p < 0.0001, Wilcoxon matched pairs test).

The volunteers who tasted PEG 3350 appreciated the taste significantly more when it was given first rather than when receiving it second, after PEG 4000 (p<0.0001)

Conclusion: PEG 4000 tastes better than PEG 3350, which has implications for patient compliance and effectiveness of treatment in patients with chronic constipation. This study only demonstrated a preference for PEG 4000 during a single intake of 25 ml of fluid. Whether this difference is apparent after months or years is unknown. Mixing with other fluids such as juice might also compensate for poor taste.

Setting Characteristics: not described

Study Design: double-blind, crossover randomized trial All subjects tasted first 25 ml of both preparations without swallowing. After tasting each of the preparations, they rinsed their mouths.

Pharmacologic Intervention in Pediatrics: Polyethylene Glycol (PEG) with or without electrolyte: PEP Weight of Evidence Category: Effectiveness Not Established 17

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Sample Characteristics: Seventy-eight children with chronic constipation for greater than three months were enrolled. Age: 2-11 years Mean age: 4.9 +2.6 years Boys: N = 34 (44%) Inclusion Children 7­11 years: one sachet twice daily on days 1 and 2, and two sachets twice daily on days 3, 4 and 5 of the study. Chronic constipation definition: Fewer than three complete bowel movements per week over previous 14 days, in association with either straining or passage of hard stools in at least a quarter of bowel movements. The existing constipation was either untreated or inadequately treated by laxatives.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

The study does not mention which electrolytes were added and in what amount.

,Hardikar, Cranswick, & Heine, 2007

All children received Macrogol 3350 plus electrolytes for 12 weeks.

Primary Outcome: number of spontaneous defecations per week. Secondary Outcome: fecal form, abdominal pain, rectal bleeding, pain on defecation, straining, soiling, amount of stool, stool withholding and assessments, of efficacy by the investigators and parents

Sxity five (80%) completed the study.

Primary Outcome: The mean number of spontaneous defecations per week increased from 1.4 + 0.55 (SD) at baseline to 6.8 +3.45 after 14 days, and 7.1 + 3.45 at 12 weeks (p < 0.001). This study design did not allow determination of a true treatment effect from spontaneous improvement. Open-label studies are subject to bias in assessments of efficacy but the findings in this study are supported by results from several non-randomized and randomized, controlled studies of PEG 3350 based preparations in children.

Children 2­6 years: one sachet daily on days 1 and 2, one sachet twice daily on days 3 and 4, and one sachet three times daily on day 5

Safety assessments: adverse events, laboratory tests (full blood examination, urea and electrolytes and liver function tests) and changes in vital signs. Adverse events were monitored throughout the study; venous blood samples for laboratory safety were taken at visits 1, 3 and 5.

Stool frequency remained unchanged from Visit 2 until the final visit (ANOVA: F = 0.81, p = 0.518).

Study results not relevant to oncology Secondary Outcome: Similar improvements were found in the secondary efficacy variables. There was a significant reduction in reported abdominal pain from 53 (69%) children at baseline to 3 (4%) at the final visit (p < 0.0001). Similarly, 61 (79%) children had pain on defecation at baseline, compared with 7 (9%) at the final visit (p < 0.0001). Treatment was well tolerated. Of 318 adverse events, 262 (82%) were considered mild, and 241 (76%) were deemed unrelated to treatment. Only 3 (4%) children were withdrawn because of poor compliance.

After day 5 and until the end of the study, the dosage was titrated according to the fecal form.

Exclusion: Treated for fecal impaction with bowel washouts within the previous 2 months, or if they had a past history of intestinal perforation or obstruction, Hirschsprungs disease, paralytic ileus, toxic megacolon, severe inflammation of the intestinal tract, urinary tract infections, uncontrolled renal, hepatic or cardiac diseases, endocrine disorders, or any other severe unstable coexisting disease within the previous 30 days.

The dose was increased by one sachet per day in the event of continued hard stools or no bowel movement, and decreased by one to two sachets per day in the event of loose stools or diarrhea.

Compliance: a 4-point scale from very poor (<25% of prescribed dose taken) to very good (>75% of prescribed dose taken).

18

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Setting: Royal Childrens Hospital, Melbourne, Australia,

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Compliance: mean duration of treatment was 75.5 days, during which time the participants took an average of 1.3 sachets (6.9 g) of Macrogol plus electrolytes per day. At each visit, most children had compliance ratings of ,,excellent (> 75% intake of study medication), 86%, 81%, 76%, and 79% at visits 2, 3, 4 and 5, respectively.

Study Design: Open-label, non-randomized The mean duration of treatment was 75.5 days, during which time the participants took an average of 1.3 sachets (6.9 g) of macrogol plus electrolytes per day. At each visit, most children had compliance ratings of ,,excellent (>75% intake of study medication), 86%, 81%, 76% and 79% at visits 2, 3, 4 and 5, respectively.

Conclusion: Longitudinal studies are needed to determine the long-term outcome of successful treatment of chronic constipation in childhood.

Purpose: To evaluate the safety and efficacy of Macrogol 3350 plus electrolytes in the treatment of chronic constipation in children.

Loening-Baucke & Pashankar (2006)

Polyetheylene Glycol 3350 without electrolytes (PEG 3350, MiraLAX) or milk of magnesia (MOM)

Sample Characteristics: N = 117 children asked to participate Total sample enrolled and randomized: N = 79

Improvement > 3 BM per week

ITT analysis

1.

High drop out rate (loss to follow up) Did not measure biochemical profiles for all children

Initially received 0.7 g/kg PEG daily or 2 ml/kg MOM daily.

> 2 episodes of fecal incontinence per month

Compliance rates were 95% for PEG and 65% for MOM significant.

2.

Randomly assigned to intervention groups by drawing sealed envelope Baseline characteristics not

No abdominal pain with or without laxative

After 12 months 62% PEG and 43% MOM group exhibited improvement and 33% PEG and 23% MOM group recovered (not

19

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

significant p > 0.07 PEG group (N = 39) Male/female = 31/8 Recovery > 3 BM per week Conclusion: PEG and MOM effective and safe in long-term treatment > 2 episodes of fecal incontinence per month PEG better acceptance No abdominal pain with or without laxative

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

significant)

Milk of Magnesia group (N = 40) Male/female = 34/6 Inclusion: New referrals (> 1 week) > 4 years Presence functional constipation with fecal incontinence Exclusions: Stool toileting refusal, fecal incontinence without constipation, previously refused one of the study medications, Hirschsprung disease, previous surgery involving colon, chronic intestinal pseudo obstruction, second opinion Setting Characteristics: Childrens Hospital of Iowa

Children who received additional senna were counted as not improved or recovered.

Study Design: RCT

Purpose: To compare two laxatives (PEG without electrolytes and milk of magnesia) evaluating efficacy, safety, acceptance, and one-year outcome

20

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model Measures Results and Conclusions Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Pharmacologic Intervention: Bisacodyl ­ Effectiveness Not Established

Kienzle-Horn, Vix, Schuitjt, Peil, Jordan, & Kamm, 2006 Bisacodyl 10 mg (two 5 mg tablets) or placebo (two tablets) orally once daily on three successive days following a three-day run-in period Sample Characteristics Total sample: N = 55 patients with idiopathic constipation age 19 to 89 years\\ All patients Caucasian Bisacodyl Group (N =28 ) Three-day baseline period, and three treatment days Male/female= 8/19 Placebo Group (N = 27) Male/female = 7/20 Inclusion: Patients > 18 years with documented history of constipation well known to investigator and in otherwise good health Exclusions: Patients with constipation associated with drug treatment, organic disease such as tumors, strictures, inflammatory disease, obstructive conditions, other GI disorders, or history of GI surgery, and pregnancy. Setting Characteristics: Eight primary care practices in Germany Study Design: RCT Phase IV multicenter, double-blind, randomized, placebo-controlled, parallel group design study. Patients were requested to not Primary endpoints during threeday treatment period: Mean number of stools per day Constipation definition: < 3 BM per week on average during the last three months and/or excessive need for straining, hard stool, low stool weight, or sensation of incomplete evacuation in more than a quarter of the evacuations 27 subjects evaluable in each group for efficacy Two patients excluded secondary to taking restricted concomitant medications. One subject lost to follow-up before study medication 1. Power analysis 28 patients per group. Twenty-seven per group completed study. 2. All patients were Caucasian. 3. Predominantly female, which is reflective of higher proportion of females experiencing constipation in general population 4. Although adverse events are reported, there is no evidence that patients were specifically asked about abdominal cramping or flatulence which are listed as objective, not subjective symptom data by these authors. 5. Subjects in study were "in good health"; thus oncology patients excluded.

Patient demographics comparable

Mean stool consistency

Primary endpoints: Mean number of stools per day was significantly greater in the bisacodyl group (1.8 +1.5 /day) compared to placebo (0.95 +0.60 per day) over the treatment phase (p = 0.0061).

Bowel Diary: Patients recorded stool frequency and consistency and adverse events.

Lab values: Blood specimens (complete blood counts, chemistries)

Mean stool consistency score improved from hard at baseline for both groups to between soft and well formed (score 2.8 +1.1, mean + SD) during bisacodyl treatment, remaining moderately hard to hard for placebo (score 4.2) p < 0.0001.

Global assessment of efficacy made by investigator based on evaluation of severity of constipation compared with baseline on 4-point Likert scale

Both treatments were well tolerated.

Global assessment: 19 of 27 in bisacodyl group compared to 14 of 27 in placebo group assessed (score 2.8 +1.1, mean + SD) assessed as significantly or

21

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

drink milk or take antacids at same time as study medications because they could dissolve the enteric coating. Randomization 1:1. Unused medication was collected at end of treatment period. Purpose: To determine the effect and safety of oral bisacodyl on stool frequency and consistency Kienzle-Horn, Vix, Schuitjt, Peil, Jordan, & Kamm, 2007 Bisacodyl sugar-coated tablets 5 to 10 mg daily or 5 to 10 mg sodium picosulphate drops daily over four weeks Sample Characteristics: Total Sample: N = 144 patients with chronic constipation (defined as < 3 stools/week for at least 6 months and/or pain/straining with bowel movements for 6 months) Age: >18 yrs. Old (23-94 yrs) Bisacodyl group ( N = 70) 21 males: 49 females Sodium picosulphate group (N = 74) 19 males: 55 females Secondary efficacy: bisacodyl group(N = 70) 1. Degree of straining using 0-5 scale (absent-severe) 2. Daily stool consistency using 5point scale (liquid-hard) Study schedule: 4 visits ­ initial screening visit followed by a 7-day baseline period, a randomization visit (1:1 basis) and two follow-up visits on days 15 and 29. 1. Absence of placebo Demographic distribution and baseline characteristics of the two treatment groups was similar. 2. Secondary efficacy parameters are subjective but generally accepted. 3. Study evaluated safety over 28 days and did not address potential long-term changes in GI mucosa. 4. No patients with cancer or opioid-induced constipation included in study.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

somewhat improved; not statistically significant (p = 0.423)

Conclusion: Bisacodyl is effective and safe in improving stool frequency and consistency.

Primary efficacy: 1. # of bowel movements/day

Compared to baseline there were significant improvements in stool frequency and consistency as well as in occurrence of straining after 14 and 28 days for both groups (p < 0.0001). Physician global rating: improvements in 74.6% bisocodyl group and 79.2% in sodium piscosulphate at day 29.

sodium picosulphate group (N = 74) Inclusion: 18 years of age with a confirmed diagnosis of chronic constipation (i.e., fewer than three stools per week for at least 6 months and/or

2. Physician global efficacy assessment.

Conclusions: bisacodyl & sodium piscosulphate are equally tolerated & effective in the treatment of chronic constipation

Safety assessments included adverse event monitoring, tolerability, changes in laboratory

22

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

a preponderance of painful stools requiring straining for the past 6 months). Exclusion: History of organic disease of the colon, ileus, any acute surgical abdominal conditions or organic diseases of the rectum and anus. Presence of active gastrointestinal disease, obstruction or dehydration, as well as ingestion of any drug affecting gastrointestinal motility or hypersensitivity to triarylmethane compounds were also excluding factors. Recent (within the past 7 days) use of bisacodyl or sodium picosulphate was also prohibited.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

parameters. Physical exam,and blood samples completed on Day 1, 15, 29.

Setting Characteristics: 15 centers in Germany (Outpatient clinics and specialist gastroenterology units Study Design: Phase IV open label randomized parallel group study

Purpose: Compare safety and efficacy of bisacodyl & sodium picosulphate

Pharmacologic Intervention: Enteral Naloxone ­ Effectiveness Not Established

Tofil, Benner Faro,& Winkler Enteral naloxone for opioids induced constipation versus Sample Characteristics: Daily stool output Mean stool output before enteral naloxone was 0.14 + 0.38 stools 1. Retrospective

23

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Total sample: N = 46 pediatric patients Treatment (N = 23) Mean age= 6.4 yrs (range 5 month to 18 yrs) Males= 14 (61%) Morphine equivalent= 86+83 (range 9 to 197 mg/day) Average LOS in PICU: 15 +8 days (range 7-43) Member Matched Control group (N = 23) Mean age= 6.1 yrs (range 4 month to 17 yrs) Males= 14 (65%) Morphine equivalent= 20+17 (range 1.3 to 59mg/day) Average LOS in PICU: 10 +7 days (range 3-31) Significant inverse relationship between dose of opioids and increase in stool output after naloxone initiation (r2= 0.17; p< 0.5) Adjuvant laxative use Eighteen study patients (78%) received other bowel agents in addition to enteral naloxone (13 MiraLax, 7 bisacodyl, 4 glycerin suppositories, and 1 milk and molasses enema). Eight control subjects received other agents (4 MiraLax, 4 bisacodyl, 1 glycerin suppositories, and 1 milk and molasses enema 3. Possible other patients had unrecognized mild withdrawal symptoms 4. Stool size or weight would have been a more objective outcome measure but were not available 5. Significant difference in Morphine equivalent/day between treatment and control groups (p= 0.1). Nutrition Mean stool output for control was 0.53 + 1.21 stools per day

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

2. Not able to control several variables that could be confounders (i.e. dose, duration, and type of opioids, use additional bowel stimulants and amount of nutrition.

(2006)

control group

per day; after initiation 1.60 +1.14 stools per day (p<0.001) Daily opiate usage

Side effects

Average Length of stay in PICU 5 days before enteral naloxone initiated (range 0 to 13 days) Inclusion: Treatment group: Received opioids and enteral naloxone for suspected opioids induced constipation. Control Group: received opioids. Randomly chosen from patient receiving opioids during study period

Two patients experienced withdrawal symptoms. One patient due to medication error.

Conclusion: Enteral naloxone may be effective in increasing stool output in opioid induced constipation (OIC) but carries risk of withdrawal symptoms. Further study needed.

24

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Exclusions: Setting Characteristics: Pediatric Intensive Care Childrens Hospital Alabama

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Study Design: Retrospective January 2003 to February 2004

Purpose: To describe the effects of enteral naloxone used to treat opioids induced constipation

Pharmacologic Intervention: Osmotic Laxatives ­ Effectiveness Not Established

Agra et. al., 1998 Purpose: Compare senna vs. lactulose in relation to efficacy and adverse events in terminal cancer patients. Interventions: Lactulose - an osmotic laxative, a synthetic disaccharide Senna - stimulate intestinal peristalsis, act mainly in the large intestine, directly stimulating the myenteric plexus and increasing water and electrolyte secretions, thus stimulating peristaltic activity. Their action extends over 6-12 hrs. Side effects are described as abdominal pain, nausea, vomiting and diarrhea. Sample Characteristics: 91 terminal cancer patients; > 18 year old in palliative care program between July 1, 1993 and July 1, 1995. All had clearly documented terminal disease with a life expectancy of < 6 mo. (30% lung ca, 11% breast ca, 8.8 % stomach ca etc.) and caregivers in the home. Mean age 67.8 yrs. (41-93 y.o.) Exclusions: all patients with contraindications to the experimental laxatives were excluded, including those with colectomy, steatorrhea, or aphagia; Karnofsy <10%; pts. who had taken opioids or laxatives during the 72 hr. prior to the initiation of the study. Patients on opioids were on codeine or morphine. Outcome measures: Defecationfree intervals of 72 hours; number of days with defecation events; and general state of health (5-pt Likert scale). Results: Limitations: 1. Patients could differentiate medications by taste and texture. 2. Data based on self-report which is not considered very reliable. 3. Study done in Spain so results may not be transferable to a US population. 4. N < 100 in final analysis 5. Baseline effect of diet, age, tumor type, mobility not discussed in relation to study results

Of the 91 subjects, 75 completed the study for at least 7 days and were evaluated in the analysis.

Defecation days as a function of opioid dose and treatment cost were also examined.

16 subjects abandoned within 4 days and were not included in analysis (1 due to diarrhea, 4 noncompliance, 4 died within first 24 hours, 5 because of permanent hospitalization, and 2 moved out of the jurisdiction of the home care area).

Laxative efficacy was analyzed by t-test and analysis of variance.

25

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Setting Characteristics: Palliative Care Unit (PCU) in Madrid Health Care District #4, Madrid, Spain. The PCU assists patients released from the local hospital and is in charge of the home care follow-up protocols. Study Design: comparative study - randomized open, parallel group design. Randomization schedule stratified for age and gender (limit 8 per stratum). 43 patients assigned to senna arm and 48 to lactulose arm. Study period: 7 days to assess laxative efficacy on defecation days and at variable opioid dosage and 27 days to assess mean morphine dose at which a laxative was necessary. Both laxative and opioid treatment was initiated simultaneously. Prescribers were blinded (single doses of identical volume in closed opaque flasks).

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Strengths: Very precise, detailed description of study design

Group A = treated with senna BID starting at 0.4 ml (12 mg) Group B = treated with lactulose BID starting at 15 ml Based on clinical response doses were increased in increments of 0.4 ml and 15 ml respectively, every 3 days. Max doses were 1.6 ml (48 mg) for senna and 60 ml (40 g) for lactulose. When a pt. reached ceiling of his respective laxative and had 3 days without defecation, he was maintained on that dose and in absence of side effects, he was started on initial dose of other laxative which could then be increased at 3-day intervals until reaching the experimental maximum. Enema and/or mechanical bowel evacuation was prescribed after 3 day period without defecation (for ethical reasons) and this was recorded as a failure with increase in laxative dose. If no results from mechanical evacuation x 6 hr. the patient was held on stand-by outside the study until defecation. Harris & Jackson, 1977 Purpose: This study measured the efficacy of lactulose in oncology patients receiving vincristine, as a treatment for vincristine-induced intractable

Conclusions: 1. No difference between senna and lactulose in efficacy as measured by defecation-free intervals; days with defecation; or in adverse effects. Recommend use of senna based on lower cost. 2. Opioid dose did not determine laxative efficacy. 3. 37.5% of patients required both laxatives by the end of the study.

Sample Characteristics: N = 8 lymphoma and leukemia patients ages 22-67.

Measures: time to first bowel movement after vincristine treatment and lactulose therapy started.

Results: All patients obtained relief of constipation within 2 days of initiating lactulose.

Limitations: Small sample size, no randomization, no mention of side effects, or duration of use.

26

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Six patients had protracted problems with constipation 5 females and one male Intervention: Lactulose was given 20 ml BID to 25 ml TID

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

constipation or for prevention of vincristine associated constipation.

Two patients received prophylactic treatment with lactulose immediately after vincristine therapy was initiated 2 male patients

Setting Characteristics: Inpatient at Royal Perth Hospital in Perth, Western Australia.

Study Design: descriptive study Lederle, Busch et al., 1990 Purpose: To evaluate the use of sorbitol as an inexpensive alternative to lactulose for treating constipation in the elderly. Intervention: Lactulose and 70% sorbitol (0 to 60 ml daily) were given for 4 weeks. During treatment period, patients were instructed to begin taking 30 ml of the study laxative at bedtime, thereafter adjusting the dose as needed between 0 & 60ml. Patients were instructed to maintain high dietary fiber and to avoid sources of free fructose such as apples and pears. Lactulose: nonabsorbable Sample Characteristics: 40 ambulatory men (see setting characteristics below only 30 patients accounted for) aged 65-86 (mean 72) with a history of constipation of at least 1 years duration were eligible. Designed to recruit the most severely constipated ambulatory patients not taking narcotics. Twenty-one took bulk-forming agents throughout the study (psyllium in all cases). Setting Characteristics: Minneapolis Veterans Affairs Medical Center from Sept. 1988 to July 1989. Five pts. were residents of Minnesota Veterans Home and the other 25 were outpatients. The primary endpoints of the study were the average number of bowel movements per week and the average number of days per week on which bowel movements occurred. Patient diary was used to record bowel consistency (hard, soft, loose); daily dose of study laxative; use of any other laxative or enema; and occurrence and severity of seven symptom categories (bloating, cramping, excessive flatulence, nausea, diarrhea, fecal incontinence and other) Statistical analysis: t-test method for cross-over trials (2-sided ) Results: 30 patients completed the study; 9 withdrew prior to randomization due to intolerance of lactulose. One pt. dropped out after randomization. There were no significant differences between sorbitol and lactulose in any outcome measured except nausea, which increased with lactulose (p<0.05). The average # of bowel movements per week was 6.71 with sorbitol and 7.02 with lactulose (95% confidence interval of the difference ranged from -0.43 to 1.06) and the average number of days per week with bowel movements was 5.23 with sorbitol and 5.31 with lactulose (95% Study had a short duration and was confined to ambulatory older men. Caution should be used in extrapolating these findings to women, younger patients or those who are bedridden. The withdrawal of patients because they did not tolerate lactulose may have biased the study. Sample size smaller than the 32 required to provide sufficient power for statistical significance of the results.

Participants dropped out related to intolerance to lactulose

27

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Study Design: a randomized, double-blind, cross over study. Study began with a two week leadin period during which patients were given lactulose in a singleblind fashion. This was followed sequentially by a washout period A (2 weeks), treatment period A (4 weeks) washout period B (2weeks), and treatment period B (4 weeks). The purpose of the lead-in period was to ensure (1) that the patient tolerated lactulose, (2) That the patient understood how to fill out the diary, and (3) that the conditions preceding the two treatment periods were similar. At the end of the washout period A, patients were randomly assigned to receive one of the two study laxatives in treatment period A, with the other laxative being used in treatment period B.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

sugar that has an osmotic laxative effect Sorbitol: nonabsorbable sugar that has an osmotic laxative effect Length of wash-in and washout periods were based on previous studies showing that up to 3-4 days are required for lactulose to take effect and that the carry-over effect after cessation of lactulose is about 6-7 days.

confidence interval of the difference ranged from -0.32 to 0.48) Conclusions: These results support the hypothesis that sorbitol and lactulose have no clinically or statistically significant difference in laxative effect. Sorbitol can be recommended as a cost effective alternative to lactulose for the treatment of constipation in elderly men

Pharmacologic Intervention: Methylcellulose­ Effectiveness Not Established

Newey & Goetzi, 1949 The study was to observe the value methyl cellulose as a laxative , differentiating between the various types of constipation (Spastic Colon, Atonic Bowel & Dyschezia) These observations were made on a group of 70 outpatients whose chief complaint was chronic constipation. The group comprised 25 male and 55 female individuals ranging in age from 18 to 70 years. The duration of constipation ranged from 4 months to 50 years (average:16.2 years). The cause of the constipation were: 39 suffering from spasticity of the colon, 25 from dyschezia . The remaining 6 patients suffered from atony of the bowel. The weekly laxation rate was recorded and the percentage of change in laxation rate calculated. In the spastic colon patients there was a positive 53% change in laxation over the 5th to the 8th week period. No cancer patients included in sample

The amount of methylcellulose taken was recorded.

Doses vary from 2.5 grams per day to 3.75 grams per day.

In the patients with dysczeia there was a positive 34.8% change in laxation over the 5th to 8th week period.

Conclusions

28

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

The period of observation for the different patients ranged from 1 to 6 months (average: 2.5 months). The type of constipation was diagnosed by history and physical exam. This article gives a good review of etiology of and clinical classification of constipation. And points out the need for evidence based studies related to the causative factors thought empirically in the past.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

fMethylcellulose helped patients with spasticity and rectal constipation but was not helpful in patients with atony because they required more stimulation.

The drug was nontoxic.

Temporary treatment on Methylcellulose re-established normal stool habits.

Pharmacologic Intervention: Alvimopan, Peripheral Opioid Antagonist ­ Effectiveness not Established

Taguchi A, Sharma N, Saleem, Sessler, Carpenter, Seyedsad, & Kurz, 2001 ADL 8-2698, an investigational opioid antagonist with limited oral absorption that does not readily cross the blood-brain barrier and so acts on the peripheral opioid receptors in the GI tract without affecting analgesia in patients taking opioids. Doses used in the study were 1 mg and 6 mg PO. Sample Characteristics: N = 78 patients generally healthy or with well-controlled systemic disease, ages 18-78 and undergoing abdominal surgery (15 ­ partial colectomy; 63- total abdominal hysterectomy) with general anesthesia. Exclusion criteria: concomitant use of corticosteroids or immunosuppressive drugs in 2 wks of before surgery; opioid analgesics within 4 wks before surgery; likely to receive NSAIDs after surgery; Chrohns disease; hx of abdominal radiation therapy; hx of treatment with vinca alkaloids. Enrollment of 26 patients per group provided 95% power to identify significant difference of one day in time to fitness for discharge between the group given the 6 mg dose and the placebo group at an alpha level of 0.05 with a two-tailed log-rank test. Time to events (time in hours since end of surgery) was compared among groups using the log-rank test. Other statistical analyses are described in detail. For patient who withdrew administration of the drug or placebo stopped, however, evaluation of the patient continued and all available data were entered into the analysis. Measures: Demographic data, type and duration of surgery were Total of 12 patients withdrew (4 placebo and 8 from 1 mg group) Twelve patients withdrew from the study, however none of these were from the 6 mg arm of the study Time to recovery of GI function was significantly shorter in patients given the 6 mg dose of ADL 82698 than those given placebo. Median time to first passage of flatus decreased from 70 to 49 hours (p= 0.03); time to first bowel movement decreased from 111 to 70 hours (p= 0.01, and the time until patients were ready for discharge decreased from 91 to 68 hours (p= 0.03). PO consumption and actual discharge occurred significantly earlier for the 6 mg dose of ADL 8-2698.Visual analogue scores for pain, itching, and abdominal cramping were Study was funded by the pharmaceutical company Adolor, a grant from NIH, and two other local funds. Two of the authors were employees of Adolor Corporation and contributed to the study design and statistical analysis.

Inclusion: Patients receiving opioids postoperatively for pain

29

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Exclusion: Patients receiving epidural administered analgesia Setting Characteristics: Washington University, St. Louis, MO between January 4, 2000 and July 22, 2000. Study Design: Randomized, placebo controlled study. On day of surgery, patients were randomly assigned in equal proportions to one of three arms using computergenerated randomization stratified according to the type of surgery. Three arms were: 1 mg or 6 mg of ADL 8-2698 or an identical appearing placebo. Patients took drug or placebo two hours before surgery and then BID postoperatively until the first bowel movement, until discharge from the hospital, or for a maximum of 7 days. Patients were seen twice daily by the research team, between 6-8 am and then 4-6 pm and questioned each visit re: time of first passage of flatus and first bowel movement. Oral intake was measured until patients could tolerate regular meals. Subjects were considered ready for discharge if they had adequate PO intake to discontinue IV fluids, when GI function had returned (defined as passage of flatus), they were afebrile, and free of major complications. Purpose: To evaluate the effects of alvimopan (ADL 8-2698) on recorded. Primary efficacy outcomes: Time to first passage of flatus; time to first bowel movement; time until patient was ready for discharge Secondary outcomes: time to first ingestion of solids, time until actual discharge; and visual analogue scores for nausea, abdominal cramping, itching and pain Severity of nausea, abdominal cramping, pain and itching were also recorded using 100 mm visual analogue scales. Total daily use of opioids was also recorded.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

similar in the three groups. In contrast, nausea scores were significantly reduced in the 6 mg dose group compared with the 1 mg and placebo groups (p= 0.003). There was no vomiting in the 6 mg group compared with 23% and 26% in the placebo and 1 mg groups respectively (p= 0.03) Conclusions: The 6 mg dose of ADL 8-2698 improved all major outcomes, with or without correction for the type of surgery. Analgesia efficacy of opioids was not affected by the study drug and there were no adverse events.

30

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

postoperative gastrointestinal function and length of hospitalization. Sample Characteristics: 522 patients taking opioids for persistent non-cancer pain for more than 6 years. Mean age 50. Included individuals with back pain (50%), neurologic (42%) fibromylagia (8%), and arthritis (7%). Setting: Not described Study Design: Randomized double-blind placebo controlled study

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Webster, Jansen, Peppin, Lasko, Snidow, Pierce, et al., 2006

Group 1: Alvimopan 0.5mg twice a day (BID) or Group 2: Alvimopan 1 mg. once a day (QD) or Group 3: Alvimopan 1 mg. twice a day (BID) or Placebo for 6 weeks.

Number of bowel movements in a week. Health related quality of life (HRQOL) was assessed using the patient Assessment of Constipation Quality of Life (PACQOL) Instrument, which consists of 28 questions on physical, psychosocial discomfort, worries/concerns and satisfaction.

The average increase in spontaneous bowel movements (defined as a bowel movement in the absence of laxatives within the previous 24 hours) per week was approximately: 3.5 in groups 1 and 2 (p<0.001) and 4.3 in group 3 (p<0.001) compared with 1.7 in placebo group. The increase in spontaneous bowel movements for each alvimopan dose group was double that of placebo, became apparent within 1 week and was sustained throughout the 6 week treatment period, returning towards baseline promptly after discontinuation of study therapy. Alvimopan showed statistically significant improvement on the satisfaction subscale (p<0.05), physical discomfort and psychosocial discomfort subscales (p,0.05, alvimopan twice daily only) and the worries and concerns subscale (p<0.05).

Cancer patients were not studied. This study was conducted by GlaxoSmithKline.

Webster, L., Jansen, J. P, Peppin, Lasko, etal., (2007).

Alvimopan Vs. Placebo administered orally

Sample Characteristics: N = 522 eligible subjects with noncancer and opioid induced bowel dysfunction randomized 1:1:1:1 ratio to oral alvimopan 0.5 mg BID, 1 mg QD, 1 mg BID or placebo capsules.

OBD Definition History decreased BM frequency since initiating opioids, and 1 or more of following symptoms: incomplete evacuation, hard stools, or straining in >25% of bowel movements. ITT analysis

1. Exclusion of subjects with cancer related pain may limit generalizability of results to this population. 2. Additional Phase III studies are needed. comparing it with other laxatives with known efficacy.

Two week baseline period prior to drug administration, six week treatment period, and two week follow up period with no drug.

No clinically relevant differences in demographic characteristics. 17% of randomized subjects did not complete study. There were no significant differences for

31

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Alvimopan 0.5 mg BID Group (N = 130) Male/female= 41%/59% Alvimopan 1 mg QD Group ( N = 133) Male/female= 37%/63% Alvimopan 1 mg BID Group (N = 130) Male/female= 32%/68% Placebo Group (N = 129) Group 4: Placebo Male/female= 35%/65% (1108 subjects screen; = n = 465, 79% ineligible) 17% of randomized subjects did not complete study. Inclusion: Male or females >18 yrs if had bowel dysfunction related to chronic opioid treatment for non malignant pain Stable dose of full opioid agonist equivalent to 30 mg oral morphine, administered orally, transdermally, IV or subcutaneously for > 1 month prior to first screening. Exclusions: Pregnancy or lactation Opioids for cancer related pain for management of addiction Unwillingness to discontinue laxatives or manual maneuvers to Responders for Opioid Induced Bowel Dysfunction Global Improvement : moderately or significant improvement on 7 point Likert scale compared with baseline. Secondary Endpoint

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

3. Primarily conducted in Caucasian population 4. Higher proportion of females.

Primary Endpoint

withdrawal between groups.

Group 1: Alvimopan 0.5mg twice a day (BID)

Group 2: Alvimopan 1 mg. once a day (QD)

Change in weekly spontaneous bowel movement (SBM) during first 3 weeks of treatment

Primary Endpoint: Compared with placebo there was a statistically and clinically significant increase in mean weekly SBM frequency over the initial 3 weeks of treatment with alvimopan 0.5 mg BID (+1.71 mean SBM/wk), alvimopan 1 mg QD (+1.64 mean SBM/wk), and alvimopan 1 mg BID (+2.52); p<0.001 for all comparisons.

Group 3: Alvimopan 1 mg. twice a day (BID)

Sustained over 6 weeks: Increase SBM frequency, improvement in symptoms such as straining, stool consistency, incomplete evacuation, abdominal bloating/discomfort, and decreased appetite.

Weekly changes in subjective bowel movement symptom ratings scores

Changes in Patient Assessment of Constipation Symptoms (PACSYM) questionnaire scores

PAC-SYM: similar across all treatment groups at baseline. Alvimopan 0.5 mg BID and 1 mg BID significantly improved total score compared with placebo all time points (p<0.05)

Changes in weekly number of rescue laxative doses used Adverse Events: Abdominal pain, nausea, and diarrhea occurred most frequently with higher doses of alvimopan.

32

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

facilitate defecation during baseline period Unmanaged severe constipation GI or pelvic disorders Setting Characteristics: 113 centers in nine countries Study Design: Phase IIb RCT double blind double dummy placebo controlled parallel group study Purpose: To Investigate the efficacy and safety of alvimopan in subjects with non-cancer and opioid induced bowel dysfunction.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Alvimopan 0.5 mg BID regimen demonstrated the best benefit to risk profile.

Pharmacologic Intervention: Lubirprostone ­ Effectiveness not Established

Baker, 2007 Summary of Phase III studies of lubiprostone. Phase III placebo controlled studies: Lubiprostone 24 µg twice daily compared to placebo. Summary of 3 Phase III Studies Study 1 and 2 Inclusion Criteria: Constipation defined as fewer than 3 spontaneous bowel movement per week plus a 6 month history of at least 1 of the other ROME II criteria for functional constipation. Study Design: Double blind placebo controlled study. Two week drug free period followed by randomization treatment with lubiprostone 24 µg twice daily for 4 weeks followed by to continue lubiprostone or placebo. Study 1 and 2 Primary endpoint: Spontaneous bowel movements (SBM) frequency after initiation of double blind treatment. Study 1 and 2 Spontaneous Bowel Movement (SBM) increased during all 4 weeks of lubiprostone in both studies. Study 1: frequencies ranged from 5.1 to 5.7 in lubiprostone group and 2.8 to 3.5 in placebo group (p< .002 at all weeks). Similar results for study 2. Time to SBM shorter in lubiprostone group. Study 1: SBM occurred in 56.7% of patients within 24 hours of first dose compared to 36.9% of placebo group (p= .0024) 1. Absence of information regarding duration of constipation 2. Lack of information related to history of previous types of therapies used to treat constipation. 3. No control group. 4. Data from publications of abstracts only. No peer journal articles. 5. No studies included cancer patients.

Open labeled studies: Lubiprostone 24 µg twice daily

33

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Study 1 Phase III study ( n = 242 patients). Sample characteristics: Female: 90% Mean Age: 48.6 yrs. Caucasian: 86% Study 2 Phase III study ( n = 237 patients) Sample characteristics Predominantly female and Caucasian. Study 3 Phase III study (n = 128) with chronic constipation. Inclusion Criteria: Constipation defined as fewer than 3 spontaneous bowel movement per week plus a 6 month history of at least 1 of the other ROME II criteria for functional constipation. Study Design: Placebo controlled RCT. Two week drug free period followed by treatment with lubiprostone 24 µg twice daily for 4 weeks followed by randomization to continue lubiprostone or placebo. Open Labeled Studies Three studies of long-term clinical safety assessed lubiprostone Study 3 Spontaneous BM (SBM) increased from 1.36/wk at baseline to 6.25, 5.94, 5.52, and 6.2 per week at Weeks 1, 2, 3, and 4 respectively (p<.0001 at all weeks). During randomization phase SBM progressively declined for placebo group. Week 7: SBM 5.59 for lubiprostone group (p= .0223 ) vs. 3.04 in placebo group (p= .0223) compared to baseline. Open Label Studies Lubiprostone reduced abdominal bloating, abdominal discomfort, and constipation severity over 6 to 12 months of treatment and significant improvements were reported for constipation severity, bloating, and abdominal discomfort scores (p<.001). Adverse Events: Although not a high incidence overall there was a higher incidence of nausea.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Study 2: SBM occurred in 69.9% of patients within 24 hours of first dose compared to 31.9% of placebo group (p<.0001)

Study 3 Responder defined as having 3 or more spontaneous BM per week during lubiprostone therapy and converted to non-responder status (fewer than 3 spontaneous BM/week) during the randomized phase for placebo group.

34

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

administered for 12 months in 871 patient with chronic idiopathic constipation. Female = 86.1% Caucasian n = 87% African American n = 7.3% 18.4% Age: Mean age 51 years Age 65 or older = 18.4% ( n = 163)

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Elderly had fewer adverse events. Conclusions: 1. Placebo controlled RCT demonstrated lupipostone well tolerated and not associated with severe adverse effects. There was a higher incidence of nausea. 2. Comparative studies with other therapies needed.

Non-pharmacologic Intervention: Activity/Increased Mobility ­ Effectiveness not Established

Bennett & Cresswell, 2003 No intervention was evaluated Bowel care protocol used by the hospital was as follows: All constipated pts. and/or those on opioids were started on combination laxative as follows: 37.5 mg dantron titrated from 1 cap/day up to 3 cap BID. If no response, polyethylene glycol was added starting with 1 packet BID and titrated up to 8 packets/day. Rectal measures (glycerin or bisacodyl suppositories or a stimulant enema) were used if stool was present in the rectum. Purpose: To test the relationship between opioid dose, bowel function and ADLs in patients with advanced cancer admitted to a hospice. Sample Characteristics: = N = 50 (24 male, 26 female). Mean age 71.4 yrs. Median MBI was 54.8; 15 pts. were not on opioids, 35 were. Exclusions: Patients with duration of admission less than one week, patients with communication difficulties, or lack of supporting information in clinical notes. Setting Characteristics: St, Gemmas Hospice Leeds, England. Study Design: Interview. Over a two month period, all patients admitted to St, Gemmas Hospice Leeds, with advanced cancer who had been admitted for at least one Measures: 1. mean daily opioid dose in prior 3 days (calculated as equivalent daily dose of morphine. Fentanyl 1:150, dilaudid 1:7.5, SQ diamorphone 1:2, codeine 10:1) 2. mean daily dose of dantron in prior 3 days, 3. Modified Barthel Index (MBI) measures 10 ADL with 5 levels of dependence maximum score is 100 points, representing independence in daily living 4. Bowel score (number of times bowels open and stool consistency) over prior 3 days. Calculated as follows: A= # times bowels opened in last 3 days B= Stool consistency (1= hard, 2= normal, 3= soft) Bowel score = A+B; Constipation = <3 Results: Reduced bowel score (particularly stool frequency) in pts taking opiods but not dosedependent relationship. No relationship between bowel score & physical functioning. For any given dose of opioid, fitter pts were treated with larger doses of laxatives. Conclusions: The authors concluded that it is likely that constipation in advanced cancer patients is more strongly related to other variables i.e. food, fluid intake, abdominal tumor involvement, depression, cognitive impairment, use of antimuscarinic drugs. Clinicians should not base laxative prescribing on opioid dose per se nor degree of dependency but should titrate laxatives according to bowel function. Limitations: 1. Small sample size; not randomized; short follow up time 2. Pts on opiods (compared those off opiods) are 10 years younger 3. Study assessed pts. after they had been titrated on laxatives 4. Overflow diarrhea might have been inaccurately assessed against the bowel score used in the study 5. Pain severity and prevalence of uncontrolled pain not examined independently.

35

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

week were interviewed about their bowel movements Purpose: Authors challenged the idea of a dose-dependent relationship between opioids and constipation and suggested that some patients may become tolerant to the constipating effects of these drugs.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Lower doses of opioids, or weaker opioids, may just as likely cause constipation as higher doses & increasing opioid dose may not be associated with more constipation.

Non-pharmacologic Intervention: Biofeedback ­ Effectiveness not Established

Chiarioni, Salandini, & Whitehead, 2005 Purpose: This study compared the efficacy of biofeedback in pelvic function dysenergia (PFD)-induced constipation vs. slow transit (ST) constipation. Intervention: All subjects received 5 weekly biofeedback lessons and were assessed at 1, 6, 12 and 24 months. Sample Characteristics: 75 consecutive patients referred to the Gastroenterology Clinic Hospital for refractory, long standing constipation and met inclusion criteria for the study. N = 52, all with delayed whole gut transit issues. 34 with PFD, and 12 with ST. Setting Characteristics: Gastroenterology Clinic of the Division of GI Rehabiliation of the Unvierstiy of Verona at Valeggio s/M Study Design: Quasi experimental Chiarioni, Whitehead, Pezza, Morelli, & Bassot, 2006 Purpose: to assess long term (24 months) efficacy of biofeedback vs. laxatives and bowel education. Interventions: Biofeedback training: five 30 minute classes Sample Characteristics: N = 109 men and women with a history of chronic/severe pelvic floor dysenergia (PFD) and who were non-responders to previous therapy Measures: Straining, sensation of incomplete evacuation, anorectal blockage, use of enemas/suppositories, abdominal pain, balloon defecation test Results: At 6 months, major improvement was reported by 43 of 54 (80%) of patients using biofeedback vs. 12 of 55 (22%) laxative treated patients (p<.001). Limitations: 1. Study design: Not blinded 2. Restricted to patients with normal whole gut transit times, patients with delayed whole gut Measures: Anorectal manometry at 1, 6 months Questionnaire and balloon defecation test at 1, 6, 12 and 24 months. Results: At 6 months greater improvements were seen in PFD compared with those with ST constipation only. 71% vs. 8% reported improved satisfaction (p= .001). 76% versus 8% reported at or above 3 bowel movements per week (p= <.001). Basically, positive results >70% in all areas for PFD and lesser for ST. Conclusions: PFD patients get greater benefit from biofeedback than ST patients Limitations: 1. Small sample size less than 100 2. Study design: not randomized; two group comparison without controls

36

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Setting Characteristics: GI Rehabilitation Hospital at Verona, Italy, University of North Carolina, Chapel Hill, N.C. & Gastroenterology and Hepatology Section, Dept. of Clinical and Experimental Medicine at University of Perugia in Perugia, Italy. Study design: Randomized control trial, Biofeedback arm N = 54 Laxative/class arm N = 55 (53 patients per group were needed to validate significance)

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

transit times were excluded. 3. Authors insist that success of biofeedback greatly depends on skill of the instructor.

over 5 weeks. Laxative plus bowel education: five 30 minute bowel training classes over 5 weeks. These patients also took 14.622.9g/kg PEG for the first 6 months. After 6 months they took the dose BID

Biofeedback benefits were sustained at 12 & 24 months in all measures but not significantly. Biofeedback showed greater reductions in straining, sensations of incomplete evacuation and anorectal blockage, use of enemas and suppositories, and abdominal pain (p= <.01). Stool frequency increased with the laxative/education also. Conclusions: Biofeedback should be the treatment of choice for PFD-induced constipation.

Non-pharmacologic Intervention: Dietary Fiber ­ Effectiveness not Established

Griffenberg, Morris, Atkinson, & Levenback, 1997 Purpose: Evaluate the effect of fiber on bowel function. Compared results of increased fiber intake with instruction as compared to those on regular diet to outcome bowel function Intervention: Dietary counseling with the treatment group included instructions to increase dietary intake to 3040 g/day. Given All-Bran cereal (unmarked) with 15g fiber per bowl. Participants were also encouraged to increase insoluble fibers i.e. whole grain whole wheat, and pumpernickel breads, All Bran cereal, butter, lima, pinto, and white beans, split and blackSample Characteristics: N = 35 women with cervical cancer who had a radical hysterectomy (Type II or III) Setting Characteristics: MD Anderson Hospital Study Design: Randomized controlled trial. Women were randomized into two groups: high fiber plan or regular diet. Used subjective questionnaires and log of exercise (7days), food diary (3days) & bowel function (7days) at intervals. Both groups were evaluated at one, four and 7 months after surgery. Measures: Measured use of medications to achieve regularity; straining; & pain with elimination. Results: After analysis of food intake, the treatment group took an average of 22.4 g of dietary fiber daily and the control group took 12. 4 grams. Insoluble fiber intake for the treatment and control groups was 16.2g and 8.1 g, respectively. Caloric intake is also discussed. There was not a significant change in the pre and post op between either groups, except the control group had a significant increase in the amount of medications used to achieve regularity. Conclusions: Those with more fiber had significantly less crampy abdominal pain, reports of Limitations: 1. Sample size was small less than 100. Although this is a RCT, the sample size of 35 could be considered a major flaw and thus decrease impact level of evidence. 2. Women only 3. Post surgical only, pre-surgical fiber for control group (10g) is significantly higher than intervention group (7g, p= 0.005)

37

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model Measures Results and Conclusions Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

eyed peas, blackberries, boysenberries, raspberries, dried fig and prunes, artichokes, asparagus, Brussel sprouts, corn, parsnips, spinach, winter squash, turnip greens. McEligot, Gilpin, Rock, Newman, Hollenbach, Thomson, & Pierce, 2002 Purpose: To examine if maintenance of a high dietary fiber intake increases GI discomfort. This study also examined whether an increased fiber intake would impact breast cancer recurrence Intervention: Treatment group received 35grams/day dietary fiber for 12 months Sutton, Dumbleton & Allway, 2007 Increase fiber intake by 6 to 12 grams per day Sample Characteristics: N = 38 women Setting Characteristics: Multi-center in Texas, Arizona, Oregon, and California through the Womens Healthy Eating and Living (WHEL) study Study Design: Double blind, cross-over randomized trial. Study period: 12 months. Measures: 1. "thoughts and feelings" questionnaire of GI s/s; 2. pre and post dietary assessments; 3. Demographic information

straining, BM retention, more BMs with gas and made in less than three minutes.

Results: Fiber intake was not a predictor of self-reported bloating, diarrhea, and upset stomach. There is a 70% increase in heartburn and 40% increase in constipation noted in those taking >40grams of fiber/day than in those taking 35-40 grams of fiber/day

Limitations: 1. Sample size: based on sample size <100. 2. It does not examine the effect of a high fiber diet on constipation. BUT 3. dIt is a study measuring the adverse events of a high fiber diet (35g/d x 12 months) on gastrointestinal discomfort

Sample Characteristics: N = 23 peritoneal dialysis patients using laxatives. Inclusion: PD at home for minimal of 3 months. Study Design: Descriptive study A three stage audit and intervention project. Stage 1: establish current bowel habits, and laxative use. Stage 2 (N= 23): Increase fiber intake by 6 to 12 grams per day using dietary fiber supplement. Stage 3 (N =17): modify daily diet

Stool and laxative diary measuring number of bowel movements/day

Stage 1: Recording diary sent to 126 patients with PD. 70 patients returned diary and 46 subjects reported using laxatives. Stage 2: 23 of the 46 subjects entered the intervention stage. 17 succeeded in replacing prescribed laxatives with the fiber supplement. All 23 patients successfully achieved a change over between increasing fiber and reducing laxatives within 4 week period.

Sample size too small Only 8 patients tried to increase dietary fibre. This limited the value of the study. The fibre supplement used was used as a thickening and stabilizing agent. It helped renal patients on fluid restriction since psyllium and methylcellulose require lots of fluid to get results. Also PHGG has negligible potassium and phosphate. It is not transferable to the oncology patient except if they are in renal compromise. There were flaws in the research design. They did not compare use of bulk forming fiber, even dietary with

38

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

to include foods naturally high in fiber aiming for 6 to 12 grams per day more than current intake and continue monitoring bowel habits and laxative use. Purpose: to explore in more depth the extent to which PD patients on our unit are affected by constipation, how many laxatives on a regular basis and what prevents them from dealing with the problem by increasing the fiber in the their diet.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

soluble fiber. A patient preferred questionnaire was administered which was subjective and no tool was presented in the study for patient outcomes assessed which can be very subjective for the clinicians evaluating the study.

Stage 3: 17 patients asked to try an increase dietary fiber by modifying daily food intake. 16 tried to increase their intake of high fibre foods, with 8 of these succeeding in improving their dietary intake of fibre, however only 2 were able to reduce their intake of fibre supplement. Conclusions The results of the study suggest that a fibre supplement can be as effective as current laxative treatment in preventing constipation, ] as well as being the preferred choice of patients as many felt it improved bowel habits while not having the side effects of stimulant laxatives. The cost of fibre supplements is much higher than standard laxatives.

A stool and laxative recording diary was sent to 126 PD patients. Forty six reported using laxatives. All respondents using laxatives were invited to use a soluble dietary fibre supplement for 4 weeks, followed by dietary advice to see if they could achieve the same effect using high fibre foods. 23 patients entered the intervention stage of the study: 1) to establish bowel habits and laxative use 2) to increase dietary fiber by 6-12 g/day using a supplement , partially Hydrolysed Guar Gum (PHGG). 3) to modify daily diet to include foods naturally high in fiber aiming for 6-12 g/day above usual intake. Study Setting: UK and approved by the Wessex Renal and Transplant Research Group.

39

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model Measures Results and Conclusions Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Limitations: This is a summary of the study (one page) ­ No mention of baseline bowel movements to prove there was not at least a small improvement in the cow milk group. It did discuss the health of the soy milk responders, who had anal fissures, erythema or edema, positive immunological tests, and symptoms of cows milk intolerance. Parents were asked to avoid food with cows milk in it for the study.

Non-pharmacologic Intervention: Soy Milk ­ Effectiveness not Established

Iacono, Cavataio, Montalto, Florena, Tumminello, Soresi, et al.,1998 Purpose: Compare cows milk vs soy milk in relieving symptoms of chronic constipation in young children Sample Characteristics: Very young children, N = 65 Study Design: Randomized control trial, double blind, crossover trial. Participants randomized, 33 to cows milk and 32 to soy milk. Children drank the fluid for two weeks and bowel function was monitored. After 2 weeks, all children had a one week washout period then were crossed over to the other intervention. Those with a positive response (> 8 BM/2 wks) were re-randomized and tested again for 2 weeks (challenge test). Measures: Number of bowel movements (a positive response is 8 or more bowel movements in a two week period) Fecal score (1= mushy liquid stools, 2= soft, 3= hard, difficult to pass) Results: 68% of the children responded to the soy milk with improved bowel scores (1-2) and a mean of 10 BMs in 2 weeks. None of the children in the cows milk group had a response (4 BMs/2 weeks) Conclusions: Soy milk alleviates constipation, only in children who are lactose intolerant or intolerant of cows milk.

Pharmacologic Intervention: Tegaserod for general constipation: PEP Weight of Evidence Category: Not Recommended for Practice

Tegaserod Maleate (Zelnorm) for IBS with constipation. (2002). Medical Letters on Drugs and Therapeutics, 44(1139), 79-80 FDA approval for use of Tegaserod for IBS was based on results of 3 studies (only one published) 3 randomized, double blind trials 2470 women taking tegaserod 6 mg. bid or placebo for 12 weeks Sample Characteristics: 2470 woman with IBS Study Design: Combination of 3 double blind randomized clinical trial Pts. Rated well-being and relief of abdominal discomfort, pain and altered bowel habits. Tegaserod may improve constipation, abdominal discomfort & bloating in some women with constipationpredominant IBS but results not impressive Long- term efficacy and safety have not been established and IBS is a chronic disease. Results: The study populations were similar with regards to demographic distribution. Comments: This was a letter announcing the FDA approval for Tegaserod. No detailed information regarding the studies or the results was included.

Johanson, 2004.

This article was a summary and review of two double-blind, randomized, placebo-controlled trials of identical design (same duration,

Sample characteristics: Across both studies the total n = 2,612 with the majority being white females. All subjects were > 18 years with a minimum 6-month

Symptoms were recorded in a patient diary throughout the study as follows: 1. Time of bowel movements and

Both studies were well-designed outcome studies.

40

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

history of constipation , defined as < 3 complete spontaneous bowel movements (CSBM)/week. Patients with constipation caused by other primary colon disease, pelvic floor dysfunction, metabolic or neurological disturbances or concomitant medication were excluded. Setting characteristics: Europe and Asia (study #1, n = 1264); North and South America (study #2, n = 1348) Study Design: Combination of two double-blind, randomized, placebocontrolled, 3-armed trials of identical design (same duration, efficacy endpoints, patient population). After a two week baseline period, patients were randomized to treatment with Tegaserod 2 mg BID, Tegaserod 6 mg BID or placebo for 12 weeks. Conceptual model: None Note: For details of study #1, refer to Kamm et all. below.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Implications: Unclear what, if any implications this has in the oncology population as patients with co-morbidities such as cancer, organic disease, GI surgery, ano-rectal malformations were excluded from the study.

efficacy endpoints, and patient population) reported in the literature: one in Europe and Asia and the other in North and South America. Purpose: to review the efficacy of Tegaserod in the management of chronic constipation. "Tegaserod, a 5 HT4 receptor partical agonist which stimulates GI motility and intestinal secretion" has demonstrated efficacy in relieving constipation in patients with irritable bowel syndrome.

2.

3.

4.

5.

associated characteristics: completeness of evacuation (yes/no); straining (3 point scale); and stool form (Bristol Stool Form 7- point scale) were recorded on a daily basis Use of other medications including laxatives were recorded daily Satisfaction with bowel habits & "bothersomness" of constipation were each evaluated weekly using a 5point ordinal scale Duration & severity of adverse events reported by the patient or discovered on history and physical were recorded. Safety evaluations included standard laboratory tests, measurement of vital signs, physical exam and electrocardiogram evaluation.

Primary efficacy variable: Both studies showed a significantly greater frequency of CSBM during weeks 1-4 for the Tegaserod groups compared with placebo (p< 0.05 vs. placebo). Secondary efficacy variables: A similar trend was seen over weeks 1-12, however, the 2 mg dose of Tegaserod did not achieve a statistical significance in Study #1. Similarly, when responder rates in study #2 were defined using more stringent efficacy criteria (>3 CSBMs per week) responder rateds remained significant for tegaserod 6 mg during the first four weeks as well as the entire 12week study period. In addition, the quality of bowel movements as measured by diverse secondary efficacy variables, significantly favored tegaserod. Of note, no rebound effect was observed following treatment withdrawal among patient who participated in the second study. Overall, the effect of tegaserod on bowel movements was rapid and predictable. In both studies, patients treated with tegaserod experienced their first SBM and first CSBM statistically significantly faster than patients receiving placebo for both doses of tegaserod.. Safety and tolerability: No consistent difference in the frequency of adverse events was noted between both Tegaserod doses and placebo. Mild to moderate diarrhea was reported as follows: placebo 3.8%, Tegaserod 2 mg 4.5% and 6 mg 7.3%. Of note the median number of spontaneous bowel movements per week by patient history was significantly different

The primary efficacy variable was the response rate of CSBM during the first 4 weeks of treatment, defined as a mean increase of one or more CSBM per week compared to baseline after a minimum of 7 days exposure to the study medication. The secondary efficacy variables were the increase in CSBMs over the entire 12-week treatment period; patient evaluation of symptoms associated with constipation and bowel habits.

41

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model Measures Results and Conclusions Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

from the median number per week recorded in diaries, suggesting that patients do not always accurately recall the number of weekly spontaneous bowel movements. Conclusions: Tegaserod is an effective treatment for chronic constipation. Tegaserod provides more rapid, predictable and consistent improvement in bowel frequency when compared to placebo in patients with a history of chronic constipation. In addition, tegaserod provided relief of accompanying symptoms of chronic constipation including straining, incomplete evacuation etc... It was also safe and well tolerated in this population. Kamm, MullerLissner et al., (2005 Tegaserod: Selective agonist that acts at 5-HT4 receptors to increase small bowel transit, stimulate intestinal secretion and inhibit visceral afferent responses. Intervention: During 12 wk treatment phase, pts. received tabs of tegaserod 2 mg, tegaserod 6 mg or placebo, each taken BID within 30 min. of morning and evening meals. The 2mg and 6 mg tegaserod tablets were different sizes, so in order to maintain blinding of the study two sizes of placebo were made. Patients took one large and one small tablet for each dose (i.e. 6mg tegaserod with small placebo etc.) Purpose: evaluate the efficacy, safety and tolerability of tegaserod in patients with chronic constipation Sample Characteristics: N = 1246, >18 yo (19 yo in Australia) men and women with constipation for 6 mo. or more. Pts with primary organic disease of the colon or pelvic floor dysfunction, metabolic disorders, neuro disorders or other significant disease were excluded. Majority of patients. were female (86%), Caucasian (98%); mean age 46. 85% of patients. had previously used at least one treatment for constipation in the 6 mo. prior to the study. Setting characteristics: study conducted between July 2001 ­ June 2002 across 128 centers in Europe, South Africa, and Australia. Number of spontaneous bowel movements; completeness of defecation. Pts. recorded data in a paper diary including: 1. Time of any BM, 2. Whether BM was associated with straining or a feeling of incomplete evacuation; 3. Stool form using the 7-point Bristol Stool Form Scale, 4. Use of bisacodyl as rescue medication; 5. Weekly assessment of patient satisfaction with bowel habits. Quality of life and general health were evaluated using the Shortform 36 questionnaire completed at baseline, 4 wks and 12 wks. Adverse events; labs, ECG and vital signs and physical exam were completed at baseline and 12 wks. Results: N = 1048 pts (82.9%) completed the double blind treatment period. Reasons for discontinuation were adverse events, unsatisfactory efficacy, consent withdrawal or other. Responder rates (defined as increase of >1 complete spontaneous BM/wk compared with baseline 14 days prior to study as long as have been on treatment 7 days) during weeks 14 was significant for tegaserod over placebo; this effect was sustained over the 12-week treatment period for the 6 mg bid dose. Onset of action significantly faster for patients receiving tegaserodthe higher dose of tegaserod was associated with faster onset of action (98 hr. vs. 174 hr for the 2 mg BID group and 286 hr in the placebo group) Strengths: large sample size; randomization process; statistical methods and data analysis Limitations: Majority of sample was female and Caucasian. Compared efficacy of different dosages of Tegaserod. Did not compare efficacy vs. other drugs Study funded by Novartis Pharma who makes the drug

42

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

Study design: randomized, double-blind, placebo-controlled study with a parallel-group design. Eligible pts. entered a 2 wk washout period without study medications, followed by a 12 wk treatment phase. Pts. seen by the investigator on day 1, and following 4,8, &12 wk of treatment for collection of efficacy data and info on adverse events. Pts. who did not have a BM for more than 96 hrs. were instructed to use bisacodyl as rescue medication with a max dose of 15 mg/day.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

Stool form, straining and other bothersome symptoms all improved with tegaserod. During the double-blind treatment period, the mean number of days/week with additional laxative use was significantly lower (p = 0.01) for both doses of tegaserod vs. placebo. Safety and tolerability: Incidence of diarrhea in tegaserod 6 mg bid compared with placebo was significantly higher; not true of tegaserod 2 mg vs. placebo. Adverse effects was the primary reason for discontinuation from the study for all groups. No clinically relevant changes in hematology, biochemistry, urinalysis, vital signs or ECG parameters noted. Conclusions: Tegaserod is safe, well-tolerated and effective in relieving chronic constipation symptoms. Greater efficacy with the 6 mg. BID dose especially in pts with irritable bowel syndrome. Supports study findings of Johanson JF et al. 2003 (clinical trial of tegaserod in chronic constipation conducted in North and South America) NA Use of Fleets enema is widespread and common. Reports of complications are scarce. This case emphasizes potential hazard of fleets enema in patients with compromised renal function. Magnesium and Phosphate salts

Pharmacologic Intervention: Magnesium and Phosphate Salts ­ Not Recommended for Practice

Nir-Paz,R. Cohen, R., & Haviv, Y.S. (1999). Acute Hyperphosphate mia caused by sodium This was a case study of one patient with acute hyperphosphatemia secondary to rectal administration of a fleet enema. Study Design: Case study (N=1)

43

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model Measures Results and Conclusions Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

phosphate enema in a patient with liver dysfunction and chronic renal failure. Renal Failure, 21(5), 541-544 Wenk, Bertolino, Ochoa, Cullen, Bertucelli et al, 2000 Purpose: Does the consumption of fresh bakers yeast reduce constipation in opioid-treated patients with advanced cancer? Intervention: Fresh bakers yeast (FBY) is a fungus, most commonly used in the kitchen for raising dough. In Argentina, it is a substance available in paste form that can be mixed with liquids and meals and is used both as a food supplement to provide vitamins and as a bowel regulator. Pts. received daily dose of FBY in the AM for 15 days. FBY was dissolved in water at room temp. and with the option of mixing it with the patients favorite food. Initial dose: 6 g once a day in the morning.; doses adjusted according to effects on bowel function. (i.e. dose doubled 12, 25, and 50g. once per day) Max daily dose 50 g. Sample Characteristics: N = 17 median age 59 year old.; 9 male, 8 female; ECOG functional status 1 (4); 2 (6); 3 (4); 4 (3);tumor type lung (6), breast (3), colon (3), laryngeal (2), melanoma, penal, adrenal each (1) Setting Characteristics: Argentina; 15 patients were treated at home, 2 as inpatients Study Design: Patients initiated opiod therapy and concurrently ingested 6 grams/day x 15 days. If no defecation, the dose was doubled daily until laxation occurred. Maximum dose was 50mg. If no results within 3 days a stimulant laxative was added to the treatment. If still no results within two days, an enema was ordered. Results were assessed every day over the phone. Conceptual Model: None Bowel function, fecal consistency, presence of diarrhea, n/v, and colic were recorded. Demographic information and EDDOM, previous bowel function were also recorded.

are harmful in patients with renal failure

Non-pharmacologic Intervention: Fresh Baker's Yeast ­ Not Recommended for Practice

Results: Two patients withdrew from the study. Results were confusing, reported as follows: 8 pts. (just over 50%) had BM daily; 6 pts. QOD and 1 pt. achieved defecation on the 3rd day. 4 required use of stimulant laxative in combination, 2 needed enemas and 1 required a glycerin suppository. Fecal consistency was normal in 9 pts, soft in 4 and hard in the remaining 2. Conclusions: FBY was effective. Limitations: Small sample size; taste of FBY was not tolerated by some; FBY needs to be refrigerated and has a 30-day lifespan; short follow-up (15 days); study uncontrolled; lack of information concerning the mechanism of action of FBY; the long-term efficacy and the side effects of FBY are not defined. Cost: low $2.50/15 days

Pharmacologic Intervention: Benefits Balance With Harm--Enemas

Expert Opinion

Walker, Warner, Brilli, Jacobs, Purpose: This is a retrospective review of 5 Sample Characteristics: 140 pediatric patients were identified Side effects and adverse outcomes with milk and molasses Results: Five children developed significant hemodynamic Case study retrospective review only.

44

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model

who were given milk and molasses enema for the treatment of constipation. Of those 43 records could not be accessed. Five had adverse effects and these charts were analyzed. All five had serious underlying co-morbidities. Study Design: retrospective chart review only. enemas.

Measures

Results and Conclusions

Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

No RCT or Double Blind studies available

2003.

medical records of children with constipation who had adverse effects related to milk and molasses enema. Intervention: Milk and molasses enema work because the sugar in the molasses is irritating to the intestinal lining, and the milk and sugar together produce gas which distends the intestines, causes pressure, peristalsis and evacuation. Milk and molasses enema administration can be associated with significant hemodynamic compromise. These risks appear similar to those risk observed with the use of other enema preparations. Multiple reports describe metabolic derangements in serum phosphate, magnesium, sodium calcium and potassium with the use of sodium phosphate, saline and magnesium phosphate enemas. Cardiac arrhythmias have been reported. Colonic perforation has been reported with over-the-counter enemas, barium and saline enemas and colitis has been associated with the use of soapsuds enemas. Allergic and anaphylactoid reactions have been observed when saline, barium and herbal preparations have been used as enemas. Transient

deterioration after receiving milk and molasses enemas. One died and the others recovered after aggressive resuscitation. Adverse events with children included: Acute respiratory and/or hemodynamic decompensation soon after receiving a milk and molasses enema. The effects of the milk and molasses could have led to increased stool output and hypovolemic shock. Conclusions: Risks for serious adverse reactions from milk and molasses enemas in children are increased when patients have serious underlying medical conditions. Clinical deterioration may also be associated with electrolyte abnormalities included hypokalemia, hypocalcemia, hypophosphatemia, acidosis and hypernatremia or the significant fluid shifts associated with osmotic enemas. When enema therapy is unavoidable, low-volume enemas with short retention times may reduce the risk of complications. Intravascular volume status should be evaluated before using enemas with hypertonic solutions.

45

ONS PUTTING EVIDENCE INTO PRACTICE

Constipation Evidence Table (Comprehensive Literature search completed through May 2008, Evidence on Methylnaltrexone updated November 2009)

Author and Year Characteristics of the Intervention Sample Characteristics, Setting Characteristics, Study Design, and Conceptual Model Measures Results and Conclusions Limitations, Major and Minor Flaws, Cautions and/or Contraindications, Special Training Needs, and Costs

bacteremia has been noted following enema administration and may result from either direct mucosal injury or increased intraluminal pressure and bacterial translocation.

46

Information

Author and Year

46 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

416377

You might also be interested in

BETA
Author and Year
ACOG Practice Bulletin, Number 37, August 2002 (Reaffirmed 2010) (Replaces Practice Bulletin Number 32, November 2001)
001 I-IV