Changing epidemiology of Pneumocystis pneumonia, Northern Ireland, UK and implications for prevention, 1 July 2011-31 July 2012.

Purpose. There is a lack of consensus about which non-human immunodeficiency virus (HIV) patient groups would benefit from prophylaxis. Here, we analysed an enhanced Pneumocystis jirovecii database to describe the epidemiology of Pneumocystis pneumonia (PCP) and P. jirovecii colonizations in Northern Ireland (NI) with a view to identifying risk groups who may benefit from prophylaxis. Methodology. We prospectively collected information on demographics, clinical severity and clinical features for all hospital inpatients in NI aged >= 18 years with P. jirovecii confirmed in any respiratory tract sample. We defined P. jirovecii colonization or PCP according to clinical symptoms and radiological findings. We compared P. jirovecii colonization to PCP using exact logistic regression and presented the odds ratios (OR), 95% confidence intervals (CI) and likelihood ratio test P-values. Results/Key findings. Overall, 36/49 (73 %) of P. jirovecii detections were categorized as PCP. A total of 28/36 (78 %) were in non-HIV patients, of which 18 (64 %) had cancer. The odds of PCP compared to P. jirovecii colonization were eight times higher in those with current exposure to chemotherapy (OR 8.73; 95% CI 0.84, infinity), 16 times higher for those diagnosed with HIV (OR 16.2; 95% CI 1.71, infinity) and 12 times higher for those ever exposed to another immunosuppressive drug (OR 12.1; 95% CI 1.94, infinity). Conclusion. The greatest burden of PCP is now in the non-HIV group, particularly cancer patients. We recommend increasing clinician awareness of PCP risk and strengthening prevention guidelines in non-HIV patients, and promoting the consideration of prophylaxis on a case-by-case basis.