Docusate calcium

We did not find any studies on the general efficacy and effectiveness of docusate calcium that met our eligibility criteria.

Docusate sodium

We did not find any studies on the general efficacy and effectiveness of docusate sodium that met our eligibility criteria.

Lactulose

We did not find any studies on the general efficacy and effectiveness of lactulose that met our eligibility criteria.

Lubiprostone

We did not find any evidence on the efficacy of lubiprostone published as full text articles. The literature search, however, detected 12 published abstracts.^20–30 Most trials were of relatively short durations (3 to 4 weeks). In general, lubiprostone had a statistically significant treatment benefit compared with placebo. Consistently higher percentages of patients on lubiprostone than on placebo had spontaneous bowel movements within 24 hours. Only one open-label study over 24 weeks suggested a durable response of lubiprostone.²² Because these abstracts did not provide enough information to critically appraise methods of individual studies, we do not report findings in detail.

Polyethylene Glycol

Three RCTs determined the general efficacy of PEG 3350.^31–33 The largest trial, a fair double-blinded RCT, enrolled 151 patients with chronic constipation who had fewer than three stools during a 7 day run-in period.³¹ Treatment success was defined as a frequency of more than three stools during a 7 day period. After 2 weeks of treatment, significantly more patients on PEG 3350 (17g/d) achieved treatment success than patients on placebo (65.8% vs. 47.8%; P < 0.001). The mean number of bowel movements was 4.5 for patients on PEG 3350 compared with 2.7 for patients on placebo (P < 0.001) The other two studies were cross-over RCTs and reported similar results after 5 days and 2 weeks of treatment, respectively.^{32, 33}

An uncontrolled before-after study³⁴ did not meet our formal eligibility criteria for efficacy; however, because it was the only study with a post-treatment follow-up, we are briefly summarizing its findings. This study enrolled 50 patients with chronic constipation and treated them with PEG 3350 for 14 days. At the end of the active treatment period, 83.3% of patients had more than three bowel movements per week and no longer met Rome II criteria for functional constipation. During the post treatment follow-up (mean 38.4 days), however, no lasting relief of symptoms could be detected. Overall, 61.7% of patients needed new treatment for constipation during this time period.

Psyllium

Two studies provide consistent evidence on the efficacy and effectiveness of psyllium for the treatment of chronic constipation.^{35, 36} Both studies, however, have methodological limitations. The larger study (n = 201) was a poor, single-blinded RCT.³⁵ This study was rated poor primarily because of the lack of an intention-to-treat (ITT) analysis. Furthermore, it remained unclear whether the study population consisted of patients with chronic constipation or a mixed population of acute and chronic constipation. This trial was conducted by 17 general practitioners in the United Kingdom (UK) and funded by a manufacturer of a psyllium product. After 2 weeks of treatment, most parameters of bowel function (stool consistency, frequency of stools, ease of defecation, abdominal pain/discomfort, straining) employed in this study were statistically significantly more improved in patients on psyllium (10.8g/d) than on placebo. For example, more patients on psyllium than on placebo reported improvement of straining (data not reported, P = 0.003). The second study was of fair methodological quality; however, only 22 patients were enrolled in this RCT.³⁶ Therefore chance findings (random error) cannot be ruled out. Results are consistent with findings from the open-label RCT. After 8 weeks of treatment, patients on psyllium (10g/d) had a statistically significantly higher stool frequency than patients on placebo (3.8 vs. 2.9; P < 0.05). Nevertheless, given the methodological limitations of both studies, results must be interpreted cautiously.

Tegaserod

Tegaserod, a 5-HT4 serotonin receptor agonist, has been FDA used for the treatment of chronic constipation in men and women under the age of 65. Five RCTs provide good evidence on the general efficacy of tegaserod for the treatment of chronic constipation.^37–41 These studies are listed in Table 8.

Table 8

Summary of trials assessing the general efficacy of tegaserod for the treatment of chronic constipation in adults.

Docusate sodium vs. psyllium

A double-blinded RCT randomized 187 patients with chronic constipation to docusate sodium (200 mg/d) or psyllium (10.2 g/d).⁴² This study received a poor quality rating because of a high rate of post-randomization exclusions (9%) and the lack of an ITT analysis. After 2 weeks of treatment no significant differences between treatment groups in subjective outcomes (straining, pain with bowel movement, evacuation completeness, constipation) were apparent. Patients on psyllium had more bowel movements (3.51 vs. 2.87/week) and a higher stool water content (73.89% vs. 71.58%) than patients on docusate sodium. These differences reached statistical significance. However, statistical testing was exclusively based on one-sided tests and absolute differences might not be clinically relevant.

Lactulose vs. PEG 3350

One open-label, head-to-head RCT randomized 115 patients to lactulose (10 – 30 g/d) or PEG 3350 (13 – 39 g/d) for the treatment of chronic constipation.⁴³ Thirty-eight percent of participants were geriatric patients. This study, however, was rated as poor because no ITT analysis was conducted. More than 13% of patients dropped out prior to the study endpoint. A completers only analysis indicated that after 4 weeks patients on lactulose had fewer weekly stools (1.3 vs. 0.9; P = 0.005) and more straining (score for straining: 0.5 vs. 1.2; P = 0.0001) than patients on PEG 3350. The overall visual analogue scale (VAS) for improvement was lower in patients on lactulose than on PEG 3350 (5.2 vs. 7.4; P = 0.0001). Although these differences achieved statistical significance, the clinical relevance remains unclear.

PEG 3350 vs. psyllium

The only available evidence comparing PEG 3350 (25g/d) with psyllium (7g/d) was an open-label RCT enrolling 126 Chinese patients with chronic constipation.^{44, 45} This study was funded by a producer of a PEG 3350 formulation. Both treatment groups increased in mean weekly defecation rates. Statistically significantly more patients on PEG 3350 than on psyllium, however, experienced improvement after 2 weeks of treatment with respect to a composite outcome including defecation frequency, stool form, and difficulty of defecation (92% vs. 73%, P = 0.005).

PEG 3350 vs. lactulose

A double-blinded RCT randomized 100 pediatric patients with constipation to PEG 3350 with electrolytes or lactulose. Patients under 6 years of age received PEG 3350 (2.95 g/sachet) or lactulose (6 g/sachet) while children 6 years or older started with 2 sachets/day.⁴⁶ This study was rated as poor quality because of a lack of ITT analysis and a high rate of post-randomization exclusions (9%). After 8 weeks of treatment, both groups showed a significant increase in mean defecation frequency per week (PEG 3350: 3 pre vs. 7 post treatment; lactulose: 3 pre vs. 6 post treatment) and a significant decrease in mean encopresis frequency per week (PEG 3350: 10 pre vs. 3 post; lactulose: 8 pre vs. 3 post treatment). There was no significant difference between treatment groups with respect to either of these parameters at 1, 2, 4, and 8 weeks of the study. Authors defined overall treatment success as three or more bowel movements per week and one or fewer encopresis episodes every 2 weeks. According to this parameter, a significantly higher number of patients in the PEG group were successfully treated compared with the lactulose group (56% vs. 29%, P = 0.02).

Docusate calcium

We did not find any studies on the general harms of docusate calcium that met our eligibility criteria.

Docusate sodium

We did not find any studies on the general harms of docusate sodium that met our eligibility criteria.

Lactulose

We did not find any studies on the general harms of lactulose that met our eligibility criteria.

Lubiprostone

We did not find any evidence on the safety of lubiprostone published as full text articles. The literature search detected 12 published abstracts addressing safety/harms for patients with chronic constipation or IBS-C.^{21, 23–29, 52, 54–56} Most studies were conducted in patients with chronic constipation; only one abstract enrolled patients with IBS-C.⁵² Most trials were of relatively short durations (3 to 4 weeks), but two were long-term studies of 24 and 48 weeks.^{21, 25} The incidence of nausea was consistently higher in lubiprostone than in placebo in controlled studies. The most common adverse events reported were nausea, headache, diarrhea, and bloating. Discontinuations due to adverse events ranged from 3% to almost 20%. These abstracts did not provide enough information to critically appraise the methods of individual studies. Thus, we cannot report findings in detail.

The FDA CDER medical review of lubiprostone⁵⁷ assessed safety data for 1,113 subjects from phase 2 and 3 clinical trials. The most common adverse events reported were headache and gastrointestinal events (nausea, diarrhea, abdominal distention or pain). Gastrointestinal events were the most common events leading to medication withdrawal. There was no evidence that lubiprostone causes adverse events on heart rate, cardiac conduction, cardiac repolarization, or bone mineral density.

Polyethylene Glycol 3350 (PEG 3350)

Three RCTs^31–33 and one open-label observational study³⁴ examined the general harms of PEG 3350. The largest trial, a fair double-blinded placebo-controlled RCT, enrolled 151 patients with chronic constipation and found no significant differences between PEG and placebo for laboratory measurements or adverse events.³¹ The PEG 3350 patients had lower rates of severe cramping and severe gas. The other two RCTs were cross-over studies^{32, 33} that were poor quality. They reported minor adverse events for subjects taking PEG including nausea, gas, cramps, and diarrhea. All four studies were funded by the makers of PEG formulations.

The fair double-blinded placebo-controlled RCT³¹ enrolled 151 adult subjects with a history of constipation and randomized them to PEG 3350 without electrolytes or placebo. Patients were required to have less than two bowel movements during a 7 day qualification period. The groups were similar at baseline for age (mean 46.7 for PEG group and 45.8 for placebo) and gender. They also had similar rates of severe cramping and severe gas during the 7 day pretreatment qualification period. Over the 2 week treatment period, patients treated with PEG had lower rates of severe cramping (12.0% vs. 22.6%; P = 0.001) and severe gas (24% vs. 40.2%; P = 0.001) than those treated with placebo. There were no statistically or clinically significant differences between groups for laboratory measurements (complete blood count [CBC], blood chemistry, and urinalysis after 14 days of treatment) or other adverse events between the groups (data not reported).

The first cross-over RCT³² compared PEG 3350 with electrolytes (8 ounces or 16 ounces) with placebo in 37 adults with chronic constipation. Nausea was reported in 8.3% of subjects in the 8 ounce PEG group and 0% in the other groups. Gas/cramps were reported in 16.7% of the 8 ounce PEG group, 75% of the 16 ounce PEG group, and 0% of the placebo group (P-value not reported). The study was rated as poor quality for adverse events.

The other cross-over RCT³³ randomized 23 patients in a private practice to 2 weeks of treatment with PEG 3350 without electrolytes followed by 2 weeks of placebo or vice versa. Subjects were 18 or older, had a history of constipation, and 3 or fewer BMs during a 7 day placebo run-in. The mean age of subjects was 47.7 years and over 95% were female. Thirteen percent of subjects reported diarrhea while taking PEG (not reported for placebo). There were no significant differences in nausea or heartburn. The authors report that there were no clinically significant differences in blood chemistry, CBC, or urinalysis between the active treatment and placebo patients (numbers not reported). While taking PEG, subjects reported lower scores (0–4 scales rated by patients) for cramping (0.6 vs. 0.9; P < 0.001) and rectal irritation (0.4 vs. 0.6; P = 0.001) compared to placebo. There was no difference in flatus (1.9 vs. 2.0; P = 0.25). The study was rated as poor quality mainly due to high attrition, as 56% of the study population requested termination (44% during placebo and 11% during PEG treatment).

The open-label observational study³⁴ was an uncontrolled before-after study that included a post-treatment follow-up. The study enrolled 50 adults with chronic constipation from a university gastroenterology practice using local advertising. All subjects were treated with PEG 3350 without electrolytes 17 g/d for 14 days. The mean age of patients was 52.1 years, 94% were female, and 60% were Caucasian. The mean duration of constipation was about 22 months. After 14 days, the following adverse events were reported: nausea (2%), constipation (2%), chest congestion (2%), high blood pressure (2%), and headache (4%). The study was rated poor quality for numerous reasons including the lack of a comparison group and no blinding.

The FDA CDER medical review of PEG and the resulting drug labeling note that nausea, abdominal bloating, cramping, and flatulence may occur. In addition, they state that high doses may produce diarrhea and excessive stool frequency, particularly in elderly nursing home patients.

Psyllium

We did not find any good or fair quality evidence on the general harms of psyllium. Two poor quality RCTs examined the general harms of psyllium.^{35, 36, 58} Both studies enrolled subjects with constipation and were funded by the makers of psyllium preparations. Psyllium was well tolerated in both trials. Neither of the studies reported significant increases in adverse events between psyllium and placebo and neither reported any serious adverse events. Given the poor quality of these studies, results should be interpreted cautiously.

The first RCT^{36, 58} compared 11 subjects treated with psyllium (Metamucil®) to 11 treated with placebo for 8 weeks. They enrolled adults aged 19–85 with chronic constipation. After a 4 week run-in, 22 subjects were confirmed by stool diaries to demonstrate constipation and were randomized. The psyllium group had more females (72.7% vs. 54.5%) and a longer mean duration of constipation (33.7 vs. 19.6 months). Psyllium was well tolerated as no patients withdrew from the study due to adverse events. All 22 subjects completed the study. There were no statistically significant differences in the adverse events reported, but there was a trend toward more abdominal pain in the psyllium group (abdominal pain: 18% psyllium vs. 0% placebo; P-values not reported). These results should be interpreted with caution due to the poor quality of the study for evaluating adverse events. Adverse events were not prespecified or defined, ascertainment techniques were not adequately described, and there was no statistical control for potential confounders.

The second RCT³⁵ randomized 201 adults with functional constipation to psyllium (Regulan, ispaghula 3.6 grams three times daily) or placebo for 2 weeks. It was a multi-site study in the UK involving 17 general practitioners. The groups were similar at baseline and had median durations of constipation of 2 (psyllium) and 3 years (placebo). Five subjects in each treatment group named side effects as reason for withdrawal from study. There were no serious adverse events reported.

Tegaserod

Fifteen studies, including 9 RCTs,^{37–39, 47–51, 59} 1 systematic review,⁶⁰ 2 pooled analyses,^{40, 61} 2 open-label prospective cohort studies,^{62, 63} and 1 uncontrolled extension of an RCT⁶⁴ report data on the general safety and harms of tegaserod for the treatment of chronic constipation and IBS-C in adults. These are summarized in Table 22. Most report a greater incidence of diarrhea with tegaserod than placebo. The cardiovascular events reported in these studies for patients treated with tegaserod are included in Table 22.

Table 22

Summary of trials assessing the general safety and harms of tegaserod for the treatment of chronic constipation and IBS-C in adults.

Lactulose vs. PEG 3350

We found just one poor quality open-label, head-to-head RCT that randomized 115 patients to lactulose (10 – 30 g/d) or PEG 3350 (with electrolytes, 13–39 g/d) for the treatment of chronic constipation.⁴³ The study was rated poor primarily because there was no ITT analysis; results should be interpreted cautiously. There were no significant differences in median daily scores for symptoms reflective of tolerance including: liquid stools, abdominal pain, flatulence, bloating and rumbling. However, the number of days with scores greater than 1 (0 to 3 scale) was lower in the PEG group for flatus (3.8 vs. 9.2; P = 0.01) and abdominal pain (3.9 vs. 6.8; P = 0.08). For the 4 week duration of the study, the mean number of liquid stools was higher in the PEG group (2.4 vs. 0.6; P = 0.001). There were 16 premature withdrawals from the study. Three were due to adverse events (2 PEG, diarrhea/vomiting/fever and abdominal pain vs. 1 lactulose, depression). For laboratory assessments, the only statistically significant change was a slight decrease in sodium in the lactulose group from 140 to 139 (P = 0.02). A mild hypokalemia (values not reported) was reported in two patients, one in each group, that were concurrently being treated with diuretics. A total of 61 of the 65 subjects treated with PEG completed an additional 2 months of follow up. There were no significant changes in adverse symptoms or laboratory results during this period. Four adverse events led to drug withdrawal during the additional 2 months: acute diarrhea with fever (1), abdominal pain (2), and vomiting (1).

Lactulose vs. psyllium

We found only 2 poor quality open-label RCTs from the UK comparing the harms or tolerability of lactulose and psyllium.^{65, 66} One RCT funded by the makers of psyllium⁶⁵ reported numerically lower rates of diarrhea and abdominal pain with psyllium. The other RCT⁶⁶ reported no differences in abdominal pain or straining and better tolerance with lactulose due to palatability. The results of these studies should be interpreted with caution due to the poor quality.

The first open-label RCT⁶⁵ randomized 394 subjects to 4 weeks of treatment with psyllium (ispaghula husk, n = 224), lactulose (n = 91), or other laxatives (n = 79). This study included adult patients presenting to general physicians with simple constipation, defined as a change in bowel habits resulting in straining or passage of hard stools. The majority (63%) were female. The duration of constipation ranged from 7 days or less in 37 patients to more than 90 days in 36 patients. The reported incidences of diarrhea (1.5% of days vs. 2.2% of days vs. 4.4% of days; P-values not reported) and abdominal pain or griping during weeks 3 to 4 (15.1% vs. 22.0% vs. 29.5%; P-values not reported) were numerically lower in the psyllium group than the lactulose group or the other laxative group. The study was rated poor quality for numerous reasons including no ITT analysis, no blinding, and adverse events were not pre-specified or defined.

The second open-label RCT⁶⁶ randomized 124 adult patients with chronic constipation to treatment with psyllium (ispaghula 3.5g twice daily) or lactulose (15 ml twice daily up to 60 ml as needed) for 4 weeks. Subjects entered the study via 21 general practitioners. There were no significant differences between the groups for abdominal pain or straining (P-value not reported). For tolerability, there was a statistically significant difference favoring the palatability of lactulose at 7 days (18.5% said psyllium was unpalatable vs. 5.7% for lactulose; P = 0.04). The trend continued at 28 days, but the difference was no longer statistically significant (15.6% vs. 4.2%; P = 0.063). The study was rated poor quality primarily for attrition of almost 26%.

PEG 3350 vs. psyllium

The only available evidence comparing PEG 3350 plus electrolytes (25 g/d) with psyllium (7 g/d) was an open-label RCT enrolling 126 Chinese patients with chronic constipation.^{44, 45} This study was funded by makers of a PEG 3350 formulation. There were no significant differences in adverse events between the groups. The most common adverse events in the PEG 3350 group were dizziness (5%) and fatigue (3.3%); the most common in the psyllium group was dry mouth (5%).

Docusate calcium, Docusate sodium, Lactulose, Lubiprostone, and Psyllium

We did not find any studies on the general harms of these medications in children that met our eligibility criteria.

Polyethylene glycol

We found no studies reporting the general safety of PEG that included a placebo comparison group. Three poor quality studies reported safety or tolerability information without a comparison group.^67–69 Two studies^{67, 69} were funded by the makers of PEG without electrolytes. The other study⁶⁸ did not report a source of funding or any conflicts of interest, but was by the same group of authors as the prospective cohort study. The most common adverse events reported were diarrhea in 10–13%, bloating/flatulence in 6–18%, and pain/cramping in 2–5%. They found no significant laboratory abnormalities. PEG 3350 was well tolerated by children. Results of these studies should be interpreted with caution due to the poor quality.

One prospective cohort study⁶⁷ included 83 children over the age of 2 treated with PEG without electrolytes (mean required dose 0.75 g/kg/d) at pediatric clinics at a referral center for a mean of 8.7 months (range 3 to 30 months). The mean age of subjects was 7.4 (range 2.0–16.9). Previous therapies for constipation had been attempted in 82% of subjects prior to enrollment. For safety and adverse events, the study reported diarrhea in 10%, abdominal pain in 2%, bloating or flatulence in 6%, elevated alanine aminotransferase (ALT) in 11%, and elevated aspartate transaminase (AST) in 4%. No abnormalities in electrolytes were found. Of the 9 patients with abnormal ALTs during treatment, 8 had repeat values 8 weeks later. Seven of the 8 were still on PEG therapy. Seven of the 8 had normal repeat values; one subject had a level 1.2 x normal (28 U/L). The 3 elevated ASTs were <1.5 times normal and all had normal repeat values 8 weeks later while still receiving PEG. The duration and dose of PEG was not different between those with elevated liver function tests (LFTs) and those with normal labs. No major adverse events were reported in the study. For tolerability, PEG was liked by 93% of children. All children (n = 68, 82%) who had used other therapies in the past preferred PEG to other laxatives.

One retrospective chart review⁶⁸ examined the safety of PEG without electrolytes in 75 infants and toddlers with functional constipation under the age of 2 over a 3.5 year period examined. Although they were not required to have chronic constipation, the mean duration of constipation was 10 months (range 0.5 to 23 months). Diarrhea was reported in 7% of 71 subjects followed for up to 4 months and in an additional 2% of 47 subjects followed for over 6 months. Parents did not report increased flatus, abdominal distention, vomiting, or new onset abdominal pain in any subjects. None stopped PEG due to adverse events. Lab tests (CBC, electrolytes, and LFTs) were occasionally done in some subjects and all those checked were normal. The study was rated poor quality for several reasons including: no comparison group, adverse events were not defined, adverse events were not clearly pre-specified, and high attrition.

One dose response study⁶⁹ was a prospective, double-blind, parallel trial that randomized children aged 3 to 18 years with chronic constipation to 4 doses of PEG 3350 without electrolytes (Miralax®, 0.25 g/kg/d, 0.50 g/kg/d, 1 g/kg/d, or 1.5 g/kg/d). All groups were treated for 3 days and evaluated 5 days after beginning treatment. They enrolled forty-one subjects referred to a pediatric gastroenterology clinic for evaluation of chronic constipation with evidence of fecal impaction. For all subjects, the following adverse events were reported: diarrhea (13%), nausea (5%), vomiting (5%), bloating/flatulence (18%), and pain/cramping (5%). Diarrhea was more prevalent in the high dose groups than the low dose groups (25% vs. 10%; P < 0.02). No patients had clinically significant abnormal laboratory values after the use of PEG 3350. For tolerability, 95% of children took the medication on the first attempt. In addition, all children said that they would repeat a 3-day regimen of PEG 3350 to help treat a future fecal impaction. The results of the study should be interpreted with caution due to poor quality (no control group).

Tegaserod

As described in the tegaserod section for general harms in adults (see above), the FDA issued a public health advisory to inform patients and health care professionals that the sponsor of tegaserod agreed to stop selling the medication because of cardiovascular adverse events.¹² We found one RCT that reported on the safety and harms of tegaserod for the treatment of postpubertal adolescents with constipation predominant IBS.⁵³ The study reported that no adverse events were observed in any patient, including diarrhea, dehydration, vomiting, rectal bleeding, weight loss, or headache. In addition there were no dropouts. This study is summarized in Table 27.

Table 27

Summary of trials assessing the general safety and harms of tegaserod for the treatment of chronic constipation and IBS-C in children.

PEG 3350 vs. lactulose

We found one poor quality RCT⁴⁶ meeting our inclusion criteria that compared PEG 3350 with lactulose in children. This study did not report any serious adverse events; it reported more abdominal pain, pain at defecation, and straining at defecation in those treated with lactulose and worse palatability with PEG.⁴⁶ The results should be interpreted cautiously due to the poor quality of this study.

The RCT⁴⁶ was a multicenter head-to-head trial from the Netherlands that randomized 100 patients to PEG 3350 with electrolytes (Transipeg) (2.95–11.8 g/d) or lactulose (6–24 g/d) for 8 weeks of treatment. The trial enrolled children from the ages of 6 months to 15 years (mean 6.5 years) with constipation. Stimulant laxatives were prescribed during the treatment phase if the treatment they were randomized to was unsuccessful. The authors report that 20% of both groups required stimulant laxatives during the study. Adverse events were assessed on a 3 point scale by patients. There were more patients with a weekly score > 1 for abdominal pain, pain at defecation, and straining at defecation in the lactulose group (values not reported, P < 0.05), and more patients had a weekly score > 1 for bad palatability in the PEG group (values not reported, P < 0.05). There were nine premature withdrawals between the two groups, with 4 in the PEG group (2 lost to follow-up, 1 unknown reason, and 1 bad palatability) and 5 in the lactulose group (2 lost to follow-up, 2 helicobacter positive, and 1 unknown). There were no serious adverse events reported. However, the authors did not define serious adverse events or how these were assessed. For tolerability, more patients reported "bad palatability" in the PEG group (%s not reported, P < 0.05). The study was rated poor for several reasons including: lack of an ITT analysis and adverse events were not pre-specified and defined.