High incidence of healthcare facility-acquired Clostridium difficile infections in chronic opioid users.

Abstract

To the Editor, Chronic opioid use impedes intestinal motility [1], impairs gut epithelial integrity [2] and induces intestinal dysbiosis [3]. Intestinal hypomotility promotes intestinal bacterial overgrowth and increases the transit time for secondary bile acids such as deoxycholate and lithocholate. Both deoxycholate and lithocholate inhibit Clostridium difficile growth [4, 5]. These opposing factors for Clostridium difficile infection (CDI) may affect the susceptibility of chronic opioid users to an infection that potentially carries fatal complications. The intestinal pathology of dysbiosis, increased permeability and bacterial overgrowth in chronic opioid users is very similar to that in sickle cell disease (SCD) [6, 7]. Since we previously found a very low incidence of HCFA CDI in SCD patients [8], we hypothesized that the incidence of HCFA CDI in chronic opioid users would be similarly low. In this study, we set out to determine the susceptibility of chronic opioid users to healthcare facility-acquired (HCFA) CDI. This information is currently lacking despite increasing number of patients who are chronic opioid users being admitted to the hospital for acute medical and surgical problems. We carried out a retrospective cohort study of chronic opioid users admitted to an acute teaching medical centre from January 2014 to July 2019. Patients were identified based on ICD codes for Opioid Use Disorders. Cepheid Xpert C. difficile real-time PCR assay targeting the toxigenic B gene was applied to patients with> 2 loose stools/24 hours in the absence of laxative use. A diagnosis of CDI was made if the PCR was positive in the absence of other pathology that could account in the diarrhoeal illness. Incidence rates, confidence intervals (CIs) and statistical differences were calculated using the MedCalc statistical software. This study was approved by the Institutional Review Board. Two hundred and seventy patients with 357 consecutive hospital admissions were included. There were 124 females and 146 males, and a total of 4291 hospital-days. Only 30 (8%) admissions were opioid-related; the other admissions were for acute medical and surgical complaints. Eighty-seven (24%) were re-admissions. The median age was 49 years (range 17–72). The average hospital length-of-stay was 12 days (median 6, range 1–207). Systemic antibiotics were used in 226 (63%) admissions. Eight episodes of CDIs were identified. HCFA CDI incidence rate in chronic opioid users was 18.6 per 10 000 hospital-days (Table 1). In contrast, HCFA CDI incidence rates in hospital-wide populations at our hospital were only 7–10.8 per 10 000 hospital-days between 2014 and 2018, and the combined incidence rate of 8.76 per 10 000 hospital-days. Chronic opioid users were, therefore, twice more susceptible to HCFA CDI than hospital-wide populations (p < 0.0001). We next determined whether there were any particular clinical factors that increased the susceptibility to HCFA CDI in those chronic opioid users who developed the infection compared to those who did not. With the exception of hospital length-of-stay, chronic opioid users who developed HCFA CDI were not different from those who did not in age, gender, transfusion, use of proton-pump inhibitor or immunosuppressive agents, CDI history, or concomitant diabetes mellitus or cancer diagnosis. Chronic opioid users with a longer hospital length-of-stay showed a trend towards significance in HCFA CDI risks (median 25 days (range 6–40) vs 6 days (range 1-207), p = 0.08) in univariate but not in the multivariate analysis. Opioid partial agonists/antagonists (e.g. Suboxone, Subutex) rather than pure agonists (e.g. heroin, oxycodone, hydromorphone, morphine) did not protect against HCFA CDI. The results of the current study rejected our hypothesis. Unlike SCD patients, chronic opioid users are not protected against HCFA CDI. In fact, they are more susceptible to the hospital acquired infection, not only compared to SCD patients but also to hospital-wide population. The reasons for the differences in incidence rates are unclear. Since HbS shifts the oxygen dissociation curve to the right and improves oxygen delivery, it may disfavour the proliferation of anaerobic bacterial species such as C. difficile. However, most chronic opioid users do not have SCD. Yet, the intestinal dysbiosis induced by chronic opioid user renders them to increased risks for the development of HCFA CDI compared to hospital-wide population. In summary, our results raise alarm on the high incidence rate of HCFA CDI in chronic opioid users. Healthcare providers should be made aware of such a potential risk in this patient group so that early suspicion and intervention can be carried out appropriately. All authors declare no conflict of interest. We also confirm that there is no external funding or collaboration associated with this work. None. Michael Lichtbroun: Investigation (equal); Project administration (equal). Firas Jafri: Investigation (equal); Project administration (equal). Richard Chaudhary: Investigation (supporting). Saleha Batool: Investigation (supporting). Jibran Ahmed: Investigation (supporting). Seah H Lim: Conceptualization (lead); Data curation (lead); Formal analysis (lead); Validation (lead).