Management of post-neurosurgical Acinetobacter infections: experiences obtained during an outbreak

Abstract

Sir, We read with interest the comprehensive article by LopezAlvarez et al. [1]. The emergence of multi-drug-resistant Acinetobacter species as major causes of nosocomial meningitis and post-neurosurgical infections has posed a significant challenge to all centers of neurosurgery around the world [1–3]. Carbapenems are the first choice of antibiotics for severe cases, and systemic and topical colistin (polymixin E) have been reintroduced as a last resort for treatment of multi-drug-resistant Acientobacter species [1–4]. During an outbreak in the neurosurgical intensive care unit in our center, six patients developed multi-drug-resistant Acinetobacter meningitis or ventriculitis post-operatively over a 2-month period. Based on the newest recommendations in the literature [1–3], treatment with intravenous colistin was started. Topical administration of colistin (10 mg, equivalent to 125,000 U/day) was performed for all six patients, which included intraventricular administration through external ventricular drains (EVDs) in five, intrathecal administration through external lumbar drains in four, and intrathecal administration via repeat lumbar punctures in three of the patients. We had to change the route of administration in five patients due to failed therapy, from EVDs to lumbar drains, and from lumbar drains to lumbar punctures, whenever possible. We observed four episodes of successful elimination of the pathogen in four patients, including the three who received the drug via lumbar punctures. All three patients underwent a ventriculoperitoneal shunt surgery afterwards, and were discharged in good conditions. The fourth patient, who was being treated through an EVD, developed a recurrence of the Acinetobacter infection and died. The other two patients, one receiving colistin through a lumbar drain and the other through an EVD, died. It seems that successful elimination of the pathogen necessitates an environment free of any catheter or foreign bodies, which may be otherwise needed for administration of topical antibiotics. Kim et al. [4] and Kaufman [3] have suggested shunt removal and external drainage for administration of topical colistin. Cascio et al. [5] prefer intraventricular route to the intrathecal route due to probably higher drug concentrations. Although our results are not still applicable to clinical settings and require further study, we suggest that whenever possible, especially in cases of failed primary therapy, EVDs or lumbar drains be avoided, and the drug be administered via repeat lumbar punctures.