Abstract

Aims To review all laboratory-confirmed cases of nocardiosis at a tertiary referral hospital over an extended period (2000–2018; 216 months) with regard to microbiological and epidemiological characteristics, risk factors, clinical management, morbidity and mortality. Methods The medical records and microbiological data of all laboratory-confirmed cases of nocardiosis, identified by culture (with reference laboratory confirmation) or identified in a reference laboratory only, were included and analysed retrospectively. Results 18 cases of nocardiosis were identified; 72% (n = 13) were male; all were UK resident. Median age at presentation was 56 years (range 6–83 years). Most had underlying pathology or risk factors including cancer in 39% (n = 7) and immunosuppression in 33% (n = 6). Alcohol and acid fast bacilli (AAFB) microscopy performed in 8/18 cases was negative. Routine 48-hour bacterial culture of 18 isolates was positive in 15; 3 culture-negative specimens were subsequently confirmed positive in a reference laboratory. Four patterns of clinical presentation were observed: cerebral 39% (n = 7), disseminated 28% (n = 5), pulmonary 17% (n = 3), and isolated cutaneous/articular (both n = 1). In addition one case of bacteraemia was noted. Nocardia farcinica accounted for half (n = 9) of all nocardia species identified. 55% (n = 10) required surgical intervention. One co-trimoxazole resistant isolate was identified. Morbidity and mortality were high: 78% (n = 14) required critical care. More than half of patients (55%; n = 10) died from refractory infection, including all of those with disseminated disease (n = 5). Conclusions Nocardia spp should never be regarded as a contaminant or commensal organism in clinical specimens. Correlation of clinical and radiology findings plus risk factors are imperative for nocardiosis to be considered in the differential diagnosis in order to guide appropriate laboratory processing of specimens. Although rare, recognition of nocardiosis is important because of its high mortality. Routine 48-hour bacterial culture does not always identify Nocardia spp and isolates should also be sent to a reference laboratory.