Constipation screening in people taking clozapine: A diagnostic accuracy study

Abstract

ObjectiveClozapine is the favoured antipsychotic for treatment-refractory schizophrenia but its safe use requires careful adverse-effect management. Clozapine-induced gastrointestinal hypomotility (CIGH or ‘slow-gut’) is one of the most common and serious of clozapine's adverse effects. CIGH can lead to paralytic ileus, bowel obstruction, gastrointestinal ischaemia, toxic megacolon, and death. Enquiring about constipation is a simple and commonly used screening method for CIGH but its diagnostic accuracy has not previously been assessed. MethodsFirst, we examined the reliability of asking about constipation compared with asking about Rome constipation criteria in inpatients treated with clozapine (n = 69). Second, we examined the diagnostic accuracy of (1) self-reported constipation and (2) the Rome criteria, compared with the reference standard of gastrointestinal motility studies. ResultsAfter 30 motility tests, it was clear constipation screening had very poor diagnostic properties in this inpatient group and the study was terminated. Although 73% of participants had objective CIGH on motility testing, only 26% of participants self-reported constipation, with sensitivity of 18% (95% CI: 5–40%). Specificity and positive predictive values were higher (95% CI: 63–100% and 40–100%, respectively). Adding in Rome criteria improved sensitivity to 50% (95% CI: 28.2–71.8%), but half the cases were still missed, making this no more accurate than tossing a coin. ConclusionsCIGH is often silent, with self-reported constipation having low sensitivity in its diagnosis. Treating CIGH based on self-reported symptoms questions will miss most cases. However, universal bowel motility studies are impractical. In the interests of patient safety, prophylactic laxatives are suggested for people taking clozapine.