Antibiotic consumption in psychiatric hospitals is low compared to general hospitals when the consumption is measured in defined daily doses (DDD) per 100 patient days. Measured in DDD per admission, however, the consumption of antibiotics in general and psychiatry hospitals may almost be the same.Reference Cizman ¹

Antibiotics have several drug–drug interactions (DDIs) with psychotropic drugs that can lead to adverse events or treatment failure and significantly increase the costs of treatment. The safest drugs for psychiatric patients are penicillin, cephalosporin, aminoglycosides, and nitrofurantoin; macrolides and quinolones should be used with great caution.Reference Hines and Murphy ^{2 –} Reference Hines and Murphy ⁴ Mortality and morbidity are increased in patients experiencing DDIs and inappropriate prescribing.Reference Hajjar, Cafiero and Hanlon ⁵

Very little is known about antibiotic prescribing patterns in psychiatric hospitals, including the frequency of potential DDIs between antibiotics and psychotropic drugs.

All Slovenian psychiatric departments were invited to participate. One hospital did not respond. The study included only patients who were prescribed antibiotics in psychiatric hospitals within the study duration. Patients already undergoing antibiotic treatment prior to the study period were excluded. The study was conducted in two 14-day periods: the first in December 2012 and the second in May 2013. The study excluded patients with antibiotics prescribed before admission. Psychiatrists were in charge of diagnosing infections and prescribing antibiotics. All included patients were monitored throughout the whole course of antibiotic treatment. In each hospital, a dedicated person (psychiatrists and clinical pharmacist) collected the data using a survey form we sent to them.

The incidence of antibiotic prescriptions was expressed as the number of new antibiotic prescriptions per 100 admissions and the number of new antibiotic prescriptions per 100 bed days. The data regarding the number of admissions and bed days during the study periods was obtained from the hospitals’ statistical services. The DDI software program Lexicomp 2012 (version 1.9.1; Lexi-Comp, Hudson, OH), was used to identify potential DDIs. Lexicomp categorizes each potential DDI according to clinical significance in 5 groups: X, D, C, B, and A. DDIs of level D and X were considered clinically significant and were included in this study. This study was approved by the National Medical Ethics Committee of the Republic of Slovenia in 2012.

A total of 74 patients met the inclusion criteria. The incidences of antibiotic prescriptions (expressed as the number of new antibiotic prescriptions per 100 admissions) were 12.9 per 100 admissions in December 2012, 7.58 per 100 admissions in May 2013, and 10.4 per 100 admissions in total. The incidences of new antibiotic prescriptions per 100 bed days were 0.21 in December 2012, 0.33 in May 2013, and 0.28 in total.

The average patient age was 58.1 years (standard deviation [SD], 18.4 years). Most patients were treated for urinary tract infections (62.2%) followed by respiratory tract infections (13.5%). The most commonly prescribed antibiotic was coamoxiclav (36.5%) and cotrimoxazole (25.7%), followed by ciprofloxacin (12.2%), and nitrofurantoin (6.80%).

Patients included in the study who were concomitantly treated with antibiotics received 2.96 psychotropic medications on average (SD 1.59, range 0–8, 95% confidence interval [CI], 2.59–3.33). Potential class D DDIs were identified in 17 patients (22.9% of all patients and 4.48 per 100 hospital admissions). Among these patients, 2 patients had 2 potential class D DDIs simultaneously (2.70% of the total patients and 0.530 per 100 hospital admissions, shown in Table 1). Potential contraindicated X class DDIs were not identified in this study. The most frequent potential antibiotic/psychotropic DDIs occured between cotrimoxazole and quetiapine (2.10 cases per 100 hospital admissions) followed by ciprofloxacin and olanzapine (0.526 cases per 100 hospital admissions).

TABLE 1

Antibiotics and Psychotropic Drugs Received by Patients with Potential Drug–Drug InteractionsFootnote ^a

^a The percent prevalences and number of potential interactions are given according to the total number of patients with infections.

In our study, antibiotics were prescribed to approximately 10% of patients during hospitalization in psychiatric hospitals. In a more recent French study, the mean antibiotic consumption in 116 psychiatric French hospitals was approximately 6 DDDs per 100 patient days.Reference Amadeo, Dumartin, Parneix, Fourrier-Réglat and Rogues ⁶

To our knowledge, this is the first study in which potential DDIs between psychotropics and antibiotics were studied in psychiatric hospitals. In patients receiving antibiotics, the prevalence of potential DDIs in our study was 22.9%, which is lower than the results of previous studies of nonpsychiatric wards.Reference Rosas-Carrasco, García-Peña, Sánchez-García, Vargas-Alarcón, Gutiérrez-Robledo and Juárez-Cedillo ⁷ In our study, 82.3% of the potential DDIs were pharmacodynamic (prolonged QT intervals) and 17.7% pharmacokinetic (via the cytochrome system). Even though pharmacokinetic interactions are more predictable in nature, pharmacodynamic interactions should be avoided or minimized, and they occurred more frequently in our study.Reference Davies, Green, Mottram and Pirmohamed ⁸ In patients treated with antibiotics and psychopharmaceuticals together, we recommend routine electrocardiographic testing, especially during escitalopram and quetiapine therapy in combination with ciprofloxacin and cotrimoxazole. Patients with a long QT syndrome history should avoid taking these drugs. Other pharmacokinetic DDIs occurred only in a few patients. Their clinical effects differed from case to case, depending on the psychopharmaceutical used with cripofloxacin. Ciprofloxacin use should be avoided in patients who are being treated with CYP1A2 substrates (eg, mirtazapine, duloxetine, or olanzapine).

The most frequent potential DDIs noted in this study, between cotrimoxazole and quetiapine, should be avoided; cotrimoxazole may be replaced with nitrofurantoin in most cases. If the dose of duloxetine, mirtazapine or olanzapine is temporarily adjusted, their pharmacokinetic interactions (cytochrome P450 1A2) with ciprofloxacin can be avoided.

Our study has several limitations. A small number of patients were included. The identification and classification of DDIs may also depend on the Lexicomp version being used. Patient habits (eg, smoking) were also excluded from the study, which may have affected the results.Reference Mangerud, Bjerkeset, Holmen, Lydersen and Indredavik ⁹ In addition, the study included only potential DDIs, which do not necessarily occur.

Antibiotic treatment is a common event during hospitalization in adult psychiatric hospitals and poses a risk of important potential DDIs to almost a quarter of all psychiatric patients treated with antibiotics.