Epidemiological and clinical characteristics of immunocompromised patients infected with Pneumocystis jirovecii in a twelve-year retrospective study from Norway

Abstract

Background: Pneumocystis pneumonia (PCP) severely menaces modern chemotherapy and immunosuppression. Detailed description of the epidemiology of Pneumocystis jirovecii today is needed to identify candidates for PCP-prophylaxis.

Methods: We performed a 12-year retrospective study of patients with P. jirovecii detected by polymerase chain reaction in Central Norway. In total, 297 patients were included. Comprehensive biological, clinical and epidemiological data were abstracted from patients' medical records. Regional incidence rates and testing trends were also assessed.

Results: From 2007 to 2017 we found a 3.3-fold increase in testing for P. jirovecii accompanied by a 1.8-fold increase in positive results. Simultaneously, regional incidence rates doubled from 5.0 cases per 100,000 person years to 10.8. A majority of the study population had predisposing conditions other than human immunodeficiency virus (HIV). Hematological (36.0%) and solid cancers (25.3%) dominated. Preceding corticosteroids were a common denominator for 72.1%. Most patients (74.4%) presented with at least two cardinal symptoms; cough, dyspnea or fever. Main clinical findings were hypoxia, cytopenias and radiological features consistent with PCP. A total of 88 (29.6%) patients required intensive care and 121 (40.7%) suffered at least one complication. In-hospital mortality was 21.5%. Three patients (1.0%) had received prophylaxis.

Conclusions: P. jirovecii is re-emerging; likely due to increasing immunosuppressants use. This opportunistic pathogen threatens the life of heterogenous non-HIV immunosuppressed populations currently at growth. Corticosteroids seem to be a major risk factor. A strategy to increase prophylaxis is called for.

Keywords: Immunocompromised; Immunosuppression; PCP; Pneumocystis jirovecii; Pneumonia.

Fig. 1

Flowchart. Flowchart of the study population. Adult patients tested in the regional referral laboratory of St. Olavs hospital and followed up in hospitals comprised by the health region of Central Norway were included in the final study population. Recruitment of alive patients required active consent in accordance with the resolution of the regional ethical committee. Molde hospital, a local hospital in the health region, established PCR detection for P. jirovecii too in 2017. Individuals resulting positive there were included in the regional incidence estimates for 2017, but not in the study population. PCR, polymerase chain reaction

Fig. 2

Pneumocystis pneumonia (PCP)-status by study year. Yearly distribution of patients with I) PCP (PCP⁺) based on i) positive direct immunofluorescence and/or ii) C_T value below 36 (black columns; n = 140, 47.1%), II) presumed colonization (PCP⁻) not fulfilling the criteria for PCP (red columns; n = 116, 39.1) and III) “undetermined PCP-status”; patients without information about C_T value and negative or missing DIF result (dark blue columns; n = 41, 13.8). Criteria were applied in retrospect. *The study period was from 2006 to 2017, though PCR was introduced in late 2006. C_T, cycle threshold; DIF, direct immunofluorescence microscopy; PCP, Pneumocystis pneumonia; PCR, polymerase chain reaction

Fig. 3

Trends in testing for Pneumocystis jirovecii by PCR. Number of respiratory samples referred to the Department of Medical Microbiology of St. Olavs hospital for P. jirovecii detection by PCR during the study period (grey columns) and number of respiratory samples resulting positive (black columns). *PCR was introduced in late 2006, and there was a 3.3-fold increase in testing from 2007 to 2017 in our regional referral laboratory. The mean proportion of positive samples (not depicted) was 20.8% (SD 4.7). **In 2017 Molde hospital, a local hospital in the health region, established PCR detection for P. jirovecii too. That year an additional 70 respiratory samples were tested at their laboratory, and 20 (28.6%), representing 17 patients, resulted positive (not depicted). PCR, polymerase chain reaction

Fig. 4

Regional incidence of Pneumocystis jirovecii detected by PCR in Central Norway. Estimated incidence rates of individuals resulting positive for P. jirovecii by PCR in Central Norway health region (dark columns) with 95% confidence intervals and resulting linear trend (dotted line). PCR was introduced in late 2006 in our regional referral laboratory. Thus, estimates were calculated from 2007. Molde hospital, a local hospital in the health region, established PCR detection for P. jirovecii too in 2017. For completeness, individuals resulting positive there were included in the regional incidence estimates for 2017 (*). Regional population counts from Statistic Norway were used to compute the incidence rates. PCR, polymerase chain reaction