Sage Journals: Discover world-class research

Abstract

Constipation is a common and sometimes fatal side effect of clozapine treatment. In this study, we aimed to identify factors associated with clozapine-induced constipation. Data on 202 outpatients stabilized on clozapine treatment were collected. Of these, 71 patients (35%) had a current prescription for laxatives (a proxy for the presence of constipation). Mean clozapine dose was 400.4 mg/day in those prescribed laxatives and 390.1 mg/day in those not prescribed laxatives (p = 0.67), while mean clozapine plasma concentration was 0.53 mg/l and 0.49 mg/l, respectively (p = 0.29). Patients using laxatives had on average 29% higher norclozapine concentrations (mean = 0.34 mg/l) than those who did not use laxatives (mean = 0.27 mg/l; p = 0.046). Laxative use was more common in female patients (49.1%) than male patients (29.1%; p < 0.01). Prescribers should be vigilant for constipation at any dose or plasma concentration of clozapine and should be mindful that male patients may be undertreated. Norclozapine concentrations may predict clozapine-induced constipation.

Keywords

clozapine constipation laxatives

Introduction

Clozapine is an atypical antipsychotic used for treatment-resistant schizophrenia (TRS). It is currently the only drug shown to have superior efficacy over other antipsychotics in TRS [Kane et al. 1988]. Agranulocytosis is a recognized rare, but sometimes fatal, adverse effect of clozapine. Regular haematological monitoring is mandatory in many countries for clozapine-treated patients in order to prevent deaths secondary to severe neutropenia. Other adverse effects of clozapine include hypersalivation, sedation, weight gain, diabetes, seizures and constipation. Constipation is an often overlooked adverse effect, but it is common and can rarely but rapidly lead to death if left untreated [Hibbard et al. 2009].

Constipation is usually defined as difficult, incomplete or infrequent evacuation of dry, hardened faeces from the bowels. Clozapine has been reported to cause constipation in up to 60% of patients [Hayes and Gibler, 1995] and has been associated rarely with bowel obstruction [Hibbard et al. 2009; Weir and Benton, 2013] and toxic megacolon [Parakkal et al. 2011]. Death resulting from clozapine-induced constipation can occur through three mechanisms: (1) untreated bowel obstruction leading to distension, necrosis, perforation or sepsis; (2) aspiration of faecal vomitus; and (3) faecal stasis leading to infection [Palmer et al. 2008].

Patients with schizophrenia are at increased risk of constipation because of a sedentary lifestyle, obesity, low-fibre diets, and dehydration due to reduced fluid intake or hypersalivation. Reduced pain sensitivity due to psychotropic medicines [Palmer et al. 2008] may delay timely detection of the condition. Constipation in general appears to be more common in female patients, and is associated with older age and living in institutions [Nielsen et al. 2013].

Antagonism of the muscarinic M₃ receptors of the intestinal smooth muscle is known to cause constipation [Abrams et al. 2006]. Clozapine has potent antimuscarinic activity at M₃ receptors and clozapine plasma concentrations may be good predictors of the extent of this receptor blockade [de Leon et al. 2003]. One study found that constipation correlated with high serum antimuscarinic activity [de Leon et al. 2003]. The same study also noted norclozapine (the major metabolite of clozapine) to have potent antimuscarinic activity. Another study found the risk of clozapine-induced paralytic ileus to be dose-related [Nielsen et al. 2013]. One important complexity in determining associations with decreased gastrointestinal (GI) motility is that around 30% of the total clozapine dose is excreted via the bile into the faeces as metabolites [Dain et al. 1997], thereby presenting the possibility of a topical effect on the bowel.

It is unclear to what extent clozapine causes constipation via direct, topical action on muscarinic receptors within the intestine or to what extent constipation is related to serum antimuscarinic activity. This study aimed to explore whether clozapine-induced constipation is a presystemic (i.e. dose-related) or systemic (i.e. plasma concentration-related) effect. We also aimed to identify whether laxative use could be predicted by certain demographics or patient factors, for example, age or female sex.

Methods

We included all outpatients at two hospital sites of the South London and Maudsley NHS Foundation Trust who had started treatment with clozapine before 1 July 2012 and who had been prescribed clozapine at a stable dose for a minimum of 3 months. Patients were identified using pharmacy outpatient prescription charts. Data were collected from electronic patient clinical notes, pharmacy outpatient dispensing records, pathology laboratory results, Zaponex Treatment Access System (ZTAS) records and by directly contacting general practitioner (GP) surgeries. We aimed to interview 20% of our sample to ask about over-the-counter (OTC) laxative use. The study was approved by the trust’s Drug and Therapeutics Committee as an audit of practice.

Patients were categorized as either laxative users or nonlaxative users. Laxative users were defined as patients who had laxatives prescribed either by their mental health prescriber or their GP. Laxative prescriptions could be either for when required use or for daily use. The remaining patients were classed as nonlaxative users.

Two-tailed Student’s t-tests were used for continuous data and chi square values and odds ratios (ORs) were calculated for dichotomous variables using Microsoft® Excel 2010. The two groups were analysed for differences in clozapine dose, clozapine and norclozapine plasma concentrations, whether they smoked or not, and factors suspected to contribute to constipation such as age, gender and additional oral anticholinergic use. Anticholinergics included were tricyclic antidepressants [British National Formulary (BNF) 67 category 4.3.1], antimuscarinic drugs used in Parkinsonism (BNF 67 category 4.9.2), drugs for urinary frequency, enuresis and incontinence (BNF 67 category 7.4.2, with the exception of duloxetine) and the antisialagogues, hyoscine hydrobromide and pirenzepine.

Results

We identified 223 patients who met inclusion criteria. A total of 21 patients were excluded from the analysis because of incomplete clinical records. A total of 202 patients were thus included in the analysis. In all, 71 patients (35%) had current prescriptions for laxatives. The most commonly used laxatives were senna (55%; n = 39) and lactulose (50%; n = 36), followed by macrogols (16%; n = 11), ispaghula husk (11%; n = 8) and docusate (7%; n = 5). Overall, 23 patients (32%) were concurrently prescribed 2 laxatives and 5 patients (7%) were prescribed 3 laxatives.

We interviewed 40 outpatients from our cohort (19.8%) of whom 2 subjects had at some time bought laxatives for their own use (5% of sub-sample; 1% of cohort).

The demographics of our study sample are displayed in Table 1. Smoking status was available for 155 patients, of whom two-thirds (103) were smokers. Full information on coprescribed medication was available for 183 patients. Of these, 79 patients (43%) had prescriptions for anticholinergic drugs, with hyoscine hydrobromide being the most commonly prescribed (68 patients).

Table 1.

Demographics.

	n (%)	Missing values
Average Age (years)	44.6	0
Male	141 (69.8)	0
of whom smoke	82 (58.2)	27
Female	61 (30.2)	0
of whom smoke	18 (29.5)	20
Ethnicity		0
White	82 (40.6)
Non-white	120 (59.4)
Smoking status		47
Yes	100 (49.5)
No	55 (27.2)
Additional Anticholinergic Use		25
Yes	79 (39.1)
No	104 (51.5)

Average clozapine doses were numerically slightly higher in patients using laxatives [400.4 mg/day; standard deviation (SD) = 164.4] compared with patients not using laxatives (390.1 mg/day; SD = 157.2), but this difference was not statistically significant, (t = 0.301, p = 0.668). Mean clozapine plasma concentrations were broadly similar for laxative users (0.533 mg/l; SD = 0.29) and nonlaxative users alike (0.486 mg/l; SD = 0.30, t = 0.211, p = 0.292). Patients using laxatives had on average 29% higher norclozapine concentrations (mean = 0.337 mg/l; SD = 0.19) than those who did not use laxatives (mean = 0.269 mg/l; SD = 0.18); this difference was statistically significant (t = 0.507, p = 0.046). The average clozapine/norclozapine ratio did not differ significantly between the two groups (Table 2). Laxative use was more common in female patients (49.1%) than male patients (29.1%; p < 0.01, χ²). There were no significant differences between laxative users and nonlaxative users with respect to age, smoking status, additional anticholinergic use and ethnicity (Table 2).

Table 2.

Comparison of characteristics in clozapine-treated patients who were prescribed laxatives with those who were not.

	n	Laxative users	Non laxative users	Total sample	p value	Odds ratio (95% CI)
Number (%)	202	71 (35)	131(65)
Average clozapine daily dose (mg)	202	400.4	390.2	393.8	0.6681
Average clozapine concentration (mg/l)	185	0.533	0.486	0.503	0.2920
Average norclozapine concentration (mg/l)	122	0.337	0.269	0.295	0.0459
Average clozapine/ norclozapine ratio	122	1.85	1.76	1.82	0.40
Average age (years)	202	46.13	43.79	44.61	0.1365
Male (%)	141	41 (29)	100 (71)		<0.01	0.42 (0.23-0.79)
Female (%)	61	30 (49)	31 (51)			1 (ref)
White (%)	82	33 (40)	49 (60)		>0.05	0.69 (0.38-1.24)
Nonwhite (%)	120	38 (32)	82 (68)			1 (ref)
Smokers (%)	100	40 (40)	60 (60)		>0.5	1.67 (0.59-2.3)
Nonsmokers (%)	55	20 (36)	35 (64)			1 (ref)
Additional anticholinergic use (%)	79	30 (38)	49 (62)		>0.5	1.02 (0.56-1.87)
No additional anticholinergic use (%)	104	39 (38)	65 (62)			1 (ref)

CI, confidence interval.

Because of the significant effect of gender on laxative prescription, we further analysed our sample population for differences between male and female patients. Female clozapine-treated patients had significantly higher average plasma clozapine levels (0.567 mg/l; SD = 0.24) than male patients (0.473 mg/l; SD = 0.31, p = 0.026), although they were prescribed significantly lower average doses (Table 3). Norclozapine plasma levels did not differ significantly between the two samples. Female patients were much less likely to smoke than male patients [OR = 0.30; 95% confidence interval (CI) = 0.15–0.64] and were also on average older (mean age = 47.2 years old; SD = 11.2) than the male patients in our cohort (mean age = 43.5 years old; SD = 10.1).

Table 3.

Comparison of factors thought to be associated with laxative use in male and female patients.

	n	Male patients	Female patients	Total sample	p value	Odds ratio (95% CI)
Number (%)	202	141 (70)	61 (30)
Average clozapine daily dose (mg)	202	412.2	351.1	393.8	0.013
Average clozapine concentration (mg/l)	185	0.473	0.557	0.503	0.026
Average norclozapine concentration (mg/l)	122	0.279	0.335	0.295	0.095
Average clozapine/ norclozapine ratio	122	1.81	1.83	1.82	0.780
Average age (years)	202	43.5	47.2	44.61	0.027
Laxative use (%)	131	41 (29)	30 (49)		<0.01	2.36 (1.27–4.39)
No laxative use (%)	71	100 (71)	31 (51)			1 (ref)
White (%)	82	57 (40)	25 (41)		>0.5	1.02 (0.56–1.89)
Nonwhite (%)	120	84 (60)	46 (59)			1 (ref)
Smokers (%)	100	82 (58)	18 (30)		<0.001	0.31 (0.15–0.64)
Nonsmokers (%)	55	32 (23)	23 (38)			1 (ref)
Additional anticholinergic use (%)	79	58 (41)	21 (34)		>0.1	0.71 (0.38–1.36)
No additional anticholinergic use (%)	104	69 (49)	35 (57)			1 (ref)

Discussion

The results of this study might be taken to indicate that clozapine-induced constipation is neither a dose-related nor a plasma-concentration related effect. We found no significant difference in mean clozapine dose or plasma concentration between laxative and nonlaxative users. This is in contrast to results from a Danish hospital registry database study [Nielsen et al. 2013], which found that patients admitted to hospital with paralytic ileus were prescribed significantly higher clozapine doses than their matched controls. However, an Australian database study [Vella-Brincat et al. 2011] of hospitalizations due to clozapine-induced gastrointestinal hypomotility (CIGH) found clozapine doses to range from 6.25 mg/day to 600 mg/day (median 425 mg/day) and plasma clozapine concentrations to range from 0.17 mg/L to 0.71 mg/L (median 0.63 mg/l), implying that constipation occurs even at small doses and low concentrations. One explanation of our findings in relation to plasma concentrations is that clozapine absorption from the bowel is almost complete and 30% of the dose is excreted primarily as metabolites back into the GI tract via the bile between days 2 and 6 following regular administration [Dain et al. 1997]. Thus plasma concentrations may not be a reflection of the amount of drug acting on the bowel.

We found norclozapine concentrations to be higher in the laxative-using group. Norclozapine is the desmethyl metabolite of clozapine and may be responsible for the adverse metabolic effects, weight gain [Lu et al. 2004] and sedation [Szegedi et al. 1995] associated with clozapine treatment. Our study suggests that constipation is more common in patients with higher norclozapine concentrations. Clozapine might be more constipating than drugs with similar serum antimuscarinic activity [de Leon et al. 2003] because of clozapine’s antagonist action at serotonin (5-HT) receptors [Palmer et al. 2008]. An alternative explanation, as suggested by our study, is norclozapine’s pharmacological activity. Although norclozapine appears to be an agonist at most types of muscarinic receptors [Weiner et al. 2004], it has some antagonist activity at the M₃ receptor [Lameh et al. 2007] (see Table 4). Norclozapine has also been found to be a more potent agonist at δ opioid receptors than clozapine [Olianas et al. 2009]. Opioid analgesics are known to cause constipation and opioid antagonists are expected to reverse opioid-induced constipation. Reducing norclozapine concentrations (e.g. by inhibiting CYP1A2 activity) could therefore have a positive impact on constipation risk during clozapine therapy. Further studies are needed to confirm this.

Table 4.

Pharmacological actions of clozapine and norclozapine at receptors.

Receptor	Clozapine	pK_i	pEC₅₀	Norclozapine	pK_i	pEC₅₀
D₂	Antagonist	7.1		Partial agonist	7.2
D₃	Antagonist	7.1		Partial agonist	6.8
5HT_1A	Agonist		6.6	Agonist		6.3
5HT_2A	Inverse agonist	8.0		Inverse agonist	8.1
5HT_2C	Inverse agonist	7.4		Inverse agonist	7.7
M₁	Antagonist	7.8		Partial agonist		7.3
M₂	Partial agonist		6.2	Partial agonist		6.5
M₃	Antagonist	8.2		Antagonist / partial agonist	6.8	6.5
M₄	Partial agonist		7.4	Agonist		6.9
M₅	Antagonist	7.5		Agonist		7.6
H₁	Antagonist	9.8		Antagonist	8.1
δ opioid	Agonist		5.9	Agonist		7.2

For each point change in pK_i and pEC₅₀, there is a 10-fold change in affinity and plasma concentration, respectively.

5HT, serotonin; D, dopamine; H, histamine; M, muscarinic; pEC₅₀, log(EC₅₀), where EC₅₀ is the plasma concentration required to achieve 50% maximal effect; pK_i,- log(K_i), where K_i is the binding affinity of the inhibitor.

Male patients were much less likely to be treated with laxatives than their female counterparts (OR = 0.42; 95% CI = 0.23–0.79). A Dutch study reviewing hospital admissions due to GI hypomotility associated with antipsychotic treatment found 65% of admissions to be male patients [De Hert et al. 2011]. Only 29% of our male sample had laxatives prescribed. Men may be less likely to report constipation than women and so might be at increased risk of serious complications secondary to untreated clozapine-induced constipation.

In our sample, marginally fewer nonsmokers were prescribed laxatives compared with smokers (36% versus 40%). Tobacco smoke is a known inducer of hepatic enzymes and smokers have been found to have lower clozapine plasma concentrations than nonsmokers, despite being prescribed higher doses [Rostami-Hodjegan et al. 2004]. Stopping smoking increases plasma concentrations of clozapine within 1–2 weeks of cessation [Lowe and Ackman, 2010]. Past studies have found the risk of sedation, confusion and seizures to be higher after smoking cessation in patients stabilised on clozapine therapy [Lowe and Ackman, 2010]. This study found that the risk of constipation is not significantly related to clozapine plasma concentrations or to smoking status. To our knowledge, no previous studies assessing this relationship have been conducted.

Additional anticholinergic use did not appear to significantly affect laxative use. This is somewhat surprising as antagonism at muscarinic M₃ receptors in the bowel is a common mechanism for drug-induced constipation. The most commonly used anticholinergic in our study was hyoscine hydrobromide (in 68 patients). Constipation is not listed in the manufacturers’ summary of product characteristics for either Kwells® or Scopoderm Patches® [Bayer plc, 2008; Novartis Consumer Health UK Ltd, 2014]. This may indicate a unique advantage of hyoscine hydrobromide over other anticholinergic drugs or simply reflect underreporting of constipation as an adverse effect.

Overall, 35% of our patients taking clozapine were prescribed laxatives. A study in 2011 found a similar proportion of laxative use (36%) in patients with schizophrenia who were prescribed any antipsychotic [De Hert et al. 2011]. Similarly, a study investigating the side effects of clozapine maintenance treatment [Yusufi et al. 2007] found that 34% of outpatients were constipated. Most of these cases were elicited using direct questioning and spontaneous reporting was low. Several reasons for this can be postulated: patients treated for schizophrenia may be more likely to be prescribed pain-modulating medicines (for example, anticonvulsants) and hence may not experience the abdominal discomfort normally associated with constipation [Palmer et al. 2008]; patients may be less likely to report physical health problems to their psychiatrist; and patients may have only limited contact with primary care health workers. In inpatients, the incidence of constipation has been reported to be as high as 60% [Hayes and Gibler, 1995]. This may reflect the more frequent contact with healthcare workers or reduced physical activity.

Senna was the most commonly prescribed laxative. Although effective in the short term, the long-term safety of stimulant laxatives has not been formally studied. The next most commonly prescribed laxative was lactulose, an osmotic laxative requiring adequate fluid intake to ensure its effect. When used in combination with another osmotic laxative (for example, macrogols), dehydration may occur. We found two such combinations in our sample. Fibre-based laxatives (for example, ispaghula husk) were less frequently used (n = 8). In uncomplicated cases, these bulk-forming laxatives can stimulate peristalsis. However, adequate fluid intake must be maintained to prevent intestinal obstruction [British Medical Association and Royal Pharmaceutical Society of Britain, 2015]. Appropriate guidelines on the use of laxatives in clozapine have only recently become available [Bleakley and Taylor, 2013].

Limitations

Laxative use in this analysis was used as a proxy for clozapine-induced constipation, the assumption being that those receiving laxatives had been identified as having clozapine-related constipation. It is possible that some patients suffer constipation without receiving prescribed treatment. Also, we could not guarantee adherence to or efficacy of laxative regimens. As we collected data from psychiatric clinical records and prescriptions, it is possible that patients receive laxatives from their GP or buy them from other sources without the treating psychiatric team being aware (although our subsample revealed very low levels of such use). We did not control for effects of other concomitant medicines on constipation and plasma concentrations of clozapine. Our results are based on single clozapine plasma concentration measurements and time of sample collection in relation to dose was not considered. We did not collect data on the extent or duration of smoking.

Conclusion

The risk of constipation does not appear to be related to clozapine dose or plasma concentration. Laxatives were prescribed for a similar proportion of patients as may be expected from previous reports. Clinicians should be vigilant and specifically ask patients about constipation at regular intervals. Men may be particularly at risk of undertreatment. Effective, long-term treatments should be initiated at the earliest opportunity to prevent serious, and sometimes fatal, bowel obstruction. Our study also adds to the clinical data on norclozapine (which are sparse) and may indicate that norclozapine contributes to clozapine-induced constipation to a greater extent that the parent molecule.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest statement

D.T. has received payments for lectures and advisory boards from Eli Lilly, Lundbeck, BristolMyersSquibb, AstraZeneca, Sunovion and Otsuka. S.G. has received payments for lectures and advisory boards from Sunovion. The other authors declare no conflicts of interest in preparing this article.

References

Abrams

Andersson

Buccafusco

Chapple

de Groat

Fryer

. (2006) Muscarinic receptors: their distribution and function in body systems, and the implications for treating overactive bladder. Br J Pharmacol 148: 565–578.

Bayer plc (2008) Kwells 300 microgram tablets, https://www.medicines.org.uk/emc/medicine/18845.

Bleakley

Taylor

(2013) Clozapine Handbook. Warwickshire, UK: Lloyd-Reinhold Communications LLP.

British Medical Association and Royal Pharmaceutical Society of Britain (2015) British National Formulary, 69th edn. London: BMJ Group and Pharmaceutical Press.

Dain

Nicoletti

Ballard

(1997) Biotransformation of clozapine in humans. Drug Metab Dispos 25: 603–609.

De Hert

Dockx

Bernagie

Peuskens

Sweers

Leucht

. (2011) Prevalence and severity of antipsychotic related constipation in patients with schizophrenia: a retrospective descriptive study. BMC Gastroenterol 11: 17.

De Leon

Odom-White

Josiassen

Diaz

Cooper

Simpson

(2003) Serum antimuscarinic activity during clozapine treatment. J Clin Psychopharmacol 23: 336–341.

Hayes

Gibler

(1995) Clozapine-induced constipation. Am J Psychiatry 152: 298.

Hibbard

Propst

Frank

Wyse

(2009) Fatalities associated with clozapine-related constipation and bowel obstruction: a literature review and two case reports. Psychosomatics 50: 416–419.

10.

Kane

Honigfeld

Singer

Meltzer

(1988) Clozapine for the treatment-resistant schizophrenic. A double-blind comparison with chlorpromazine. Arch Gen Psychiatry 45: 789–796.

11.

Lameh

Burstein

Taylor

Weiner

Vanover

Bonhaus

(2007) Pharmacology of N-desmethylclozapine. Pharmacol Ther 115: 223–231.

12.

Lowe

Ackman

(2010) Impact of tobacco smoking cessation on stable clozapine or olanzapine treatment. Ann Pharmacother 44: 727–732.

13.

Lane

Lin

Chen

Chang

(2004) Adjunctive fluvoxamine inhibits clozapine-related weight gain and metabolic disturbances. J Clin Psychiatry 65: 766–771.

14.

Nielsen

Correll

Manu

Kane

(2013) Termination of clozapine treatment due to medical reasons: when is it warranted and how can it be avoided? J Clin Psychiatry 74: 603–613; quiz 613.

15.

Novartis Consumer Health UK Ltd (2014) Scopoderm 1.5 mg Patch. https://www.medicines.org.uk/emc/medicine/29044.

16.

Olianas

Dedoni

Ambu

Onali

(2009) Agonist activity of N-desmethylclozapine at delta-opioid receptors of human frontal cortex. Eur J Pharmacol 607: 96–101.

17.

Palmer

McLean

Ellis

Harrison-Woolrych

(2008) Life-threatening clozapine-induced gastrointestinal hypomotility: an analysis of 102 cases. J Clin Psychiatry 69: 759–768.

18.

Parakkal

DeLemos

Ehrenpreis

(2011) Rapidly progressive shock and death from clozapine-induced toxic megacolon: lessons still to be learned. Am J Gastroenterol 106(Suppl. 2): S340.

19.

Rostami-Hodjegan

Amin

Spencer

Lennard

Tucker

Flanagan

(2004) Influence of dose, cigarette smoking, age, sex, and metabolic activity on plasma clozapine concentrations: a predictive model and nomograms to aid clozapine dose adjustment and to assess compliance in individual patients. J Clin Psychopharmacol 24: 70–78.

20.

Szegedi

Wiesner

Hiemke

(1995) Improved efficacy and fewer side effects under clozapine treatment after addition of fluvoxamine. J Clin Psychopharmacol 15: 141–143.

21.

Vella-Brincat

McKean

Barclay

(2011) Clozapine-induced gastrointestinal hypomotility. Australas Psychiatry 19: 450–451.

22.

Weiner

Meltzer

Veinbergs

Donohue

Spalding

Smith

. (2004) The role of M1 muscarinic receptor agonism of N-desmethylclozapine in the unique clinical effects of clozapine. Psychopharmacology 177: 207–216.

23.

Weir

Benton

(2013) Clozapine-induced acute colonic pseudo-obstruction. J Gen Intern Med 28(S309): 0884–8734.

24.

Yusufi

Mukherjee