Identifying antibiotics associated with Clostridium difficile infection
The increase in Clostridium difficile infections (CDI) in the early 1990s was associated with the widespread use of clindamycin, cephalosporins and quinolones. A nationwide drive to implement strict antimicrobial stewardship guidelines has changed the nature of the antibiotics associated with C. difficile. As a result, in NHS England and the Heart of England Foundation Trust (HEFT), C. difficile incidence has decreased by over 75% from 2007 to 2017, because of changes in infection control and restricted antibiotic prescribing of quinolones, cephalosporins and clindamycin.
However, the evidence now suggests that broad spectrum antibiotics, long duration of treatment and multiple concurrent antibiotics are associated with CDI. We conducted a study which looked at the antibiotics now associated with CDI, in a setting of stringent antimicrobial stewardship.
This study was done in a large three-hospital NHS Trust, where total antibiotic use is low, through strict antibiotic stewardship (2016/2017 data from Public Health England Health Profiles: 3,625 defined daily dose [DDD] per 1,000 admissions, which is much less than the England average of 4,537 DDDs per 1,000 admissions).
Between April 2014 and March 2015, for each patient at HEFT who was found to be Clostridium difficile toxin (CDT) positive 48 hours after admission, an antibiotic root cause analysis (RCA) structured report was completed by the antimicrobial stewardship team which is comprised pharmacists, microbiologists and infection control nurses.
This involved a detailed review of patients’ electronic and paper records, for primary and secondary care antibiotics prescribed within the three months preceding development of CDI. Each antibiotic prescription was assessed against local prescribing guidelines and infection specialist advice for appropriateness in relation to the clinical indication.
From April 2014 to March 2015, 76 patients at the HEFT were found to be post-48 hour CDT positive. In total, 559 antibiotic courses were prescribed for these patients (12% of the antibiotics in primary care and 88% in secondary care). The median number of antibiotic courses prescribed was six (range=1 to 31). Overall, 95% of prescriptions were clinically compliant with guidelines and infection specialist advice.
Of the antibiotic courses, 28 out of 559 (5%) were deemed inappropriate and non-compliant to the trust’s antibiotic guidelines. Among these, the three antibiotics most commonly prescribed were: co-amoxiclav (n=11), trimethoprim (n=3) and metronidazole (n=3).
Co-amoxiclav was the most frequently prescribed inappropriate antibiotic both in secondary care and primary care, 37% and 45%, respectively. In primary care, the most common indications for inappropriately prescribing co-amoxiclav were: cellulitis (n=2), upper respiratory tract infections (n=1) and urinary catheter removal prophylaxis (n=1). In secondary care the most common indications for inappropriately prescribing co-amoxiclav were respiratory tract infections (n=3) and upper gastrointestinal bleed prophylaxis (n=1).
A commonly used broad spectrum antibiotic in secondary care, intravenous piperacillin/tazobactam, was inappropriately prescribed for community onset cholecystitis.
Excluding the treatment of CDI in secondary care, the mean duration of inappropriate antibiotic courses was longer in primary care (11 days) compared with secondary care (mean=4.25 days).
The study did not show any cases of inappropriate prescribing with antibiotics commonly blamed for CDI, quinolones or cephalosporins. This is because the use of these ‘high-risk antibiotics is rigorously restricted in primary and secondary care across England, but there has been a corresponding increase in use of combination penicillins, co-amoxiclav and piperacillin-tazobactam, which may be now more important agents driving C. difficile infection post-restrictive prescribing guidelines. Our results show that CDI association with antibiotics is heterogeneous and complex. It appears independent of the number and type of antibiotics.
Further antimicrobial stewardship strategies to achieve further reductions in CDI should focus on commonly inappropriate prescriptions (e.g. co-amoxiclav).
Lead Pharmacist, Anti-infectives
Das Pillay, Consultant Microbiologist
Heart of England Foundation Trust, Birmingham
Citation: Clinical Pharmacist DOI: 10.1211/CP.2017.20202680