Gaining an Understanding of Pneumocystosis in Wales

Abstract

Pneumocystis pneumonia (PcP) is a serious complication of many significant immunocompromising conditions. Prior incidence estimates in Wales are based on PcP's presentation in the HIV and transplant populations. The objectives were to describe the incidence of PcP in Wales using laboratory reporting measures and assess the impact of underlying immunosuppression cause on mortality. All positive PCR results for PcP between 2015 and 2018 were identified. The total number of unique positives with clinical and radiological correlation was 159 patients, a mean of 39.75 annually. The healthcare records of these patients were reviewed. The mortality at one month was 35.2% and 49.1% at one year. HIV remains the commonest cause of immunosuppression but has lower mortality than non-HIV conditions (12% vs. 59% at one year, p < 0.00001). Non-HIV conditions were categorised as life-threatening and non-life threatening but had a non-significant mortality (66% vs. 54%; p = 0.149), highlighting the negative impact of PcP. An incidence of PcP in Wales of 1.23-1.26 cases per 100,000 has been identified, 32-35% greater than the upper limit previously estimated. There is high mortality in non-HIV patients regardless of immunosuppression cause. A heightened awareness of PcP in these groups will hasten diagnosis and potentially improve mortality.

Keywords: HIV; PcP; Pneumocystis; Wales; incidence; non-HIV.

Figure 1

Diagnostic pathway for reviewing patients with a positive PcP PCR that does not reach established Cq cut-off values. When testing respiratory throat swabs, 126 out of 132 samples from patients without PcP generated a negative PcP PCR result (Specificity: 95.5%, 95% CI: 88.0–98.5. Positive likelihood ratio: 18.5). However, only two of these six false-positive results generated a cycle threshold (Ct value) < 38 cycles generating a specificity and positivity likelihood ratio of 98.5% (95% CI: 92.3–99.8) and 52.1, respectively, indicating that a PcP PCR-positive results with a Ct < 38 cycles on a throat swab was highly predictive of PcP. Similarly, when testing bronchoalveolar lavage fluid (n = 41), the specificity of PcP PCR when applying a positivity threshold of <36 cycles was 95.1% (39/41), and the positive likelihood ratio was 18.0. In relation to combination testing, involving reviewing patients who were deemed weakly positive by PcP PCR, the Fungitell (1–3)-β-D-Glucan threshold to exclude the need for clinical/radiological review was selected based on data from a recent study, where patients, who were PcP PCR-positive and had serum (1–3)-β-D-Glucan >200 pg/mL, the subsequent specificity of a PcP diagnosis was 100% [18]. In an attempt to minimise the impact of using different PcP PCR assays between studies and to improve our certainty of PcP diagnosis, it was decided to increase the (1–3)-β-D-Glucan positivity threshold to ≥300 pg/mL.

Figure 2

Flowchart of PcP PCR positivity and association with potential Pneumocystosis.

Figure 3

PcP mortality by year, separated into 1 month and 1 year.