Pneumocystis jirovecii is a transmissible and uncultivable micromycete that causes severe acute pneumonia (ie, Pneumocystis pneumonia, PCP) in immunosuppressed patients. Pneumocystis spp are host specific, and no exosaprophytic form of Pneumocystis sp has been identified so far. Thus, humans may represent the reservoir of P. jirovecii and potential infectious sources for susceptible individuals.Reference Stringer, Beard, Miller and Wakefield ¹

Pneumocystis jirovecii DNA has been detected and quantified using quantitative polymerase chain reaction (qPCR) in the air surrounding PCP patients, suggesting exhalation and spread of P. jirovecii from infected patients within their environment.Reference Choukri, Menotti and Sarfati ^{2 ,} Reference Le Gal, Pougnet and Damiani ³ This finding emphasizes the risk of patient-to-patient transmission of P. jirovecii via the airborne route, which was also prompted by investigations of PCP case clusters in hospitals (see the review by Yiannakis et alReference Yiannakis and Boswell ⁴ ). Taken together, these data support the maintenance of prevention measures based at least on patient treatment and isolation.Reference Siegel, Rhinehart, Jackson and Chiarello ⁵ Nonetheless, there are no arguments to delineate their duration. In this context, we investigated longitudinal P. jirovecii air exhalation by a renal transplant recipient who developed PCP and was efficiently treated with cotrimoxazole (2,880 mg/day). The diagnosis of PCP was based on a positive result of P. jirovecii detection in a bronchoalveolar lavage (BAL) sample using microscopy and a qPCR assay targeting the gene encoding the mitochondrial large subunit ribosomal RNA (mtLSUrRNA), as previously reported.Reference Choukri, Menotti and Sarfati ^{2 ,} Reference Le Gal, Pougnet and Damiani ³ During 5 consecutive days, 5 air samples were collected after treatment initiation in patient’s room at 1 m from the patient’s head. The samples consisted of 3 m³ of air collected using the Coriolis μ air sampler (Bertin Technologies, Montigny-le-Bretonneux, France), which concentrates aerial particles in a liquid medium. The DNA extraction of air samples was performed according to the recommendations of Choukri et al.Reference Choukri, Menotti and Sarfati ² The P. jirovecii burden was determined in the BAL and air samples using a qPCR assay amplifying the mtLSUrRNA gene as described by Choukri et al.Reference Choukri, Menotti and Sarfati ² Samples and a negative control were run in triplicate. The quantity of P. jirovecii DNA in the samples was determined against a standard curve. The genotyping of P. jirovecii from the BAL and air samples was performed by examining cytochrome b (CYB) and mtLSUrRNA genes, as we described elsewhere.Reference Vindrios, Argy and Le Gal ⁶

The P. jirovecii DNA load was evaluated at 2.97×10⁶ copies/mL of native sample in the BAL specimen (2.97×10⁴ copies/µL of extracted DNA). The P. jirovecii DNA load was 1.18×10⁷ copies/m³ of air in the sample collected 1 day after treatment initiation; 2.39×10⁵ copies/m³ in the sample collected 2 days after treatment initiation; 4.48×10³ copies/m³ in the sample collected 3 days after treatment initiation; and <1.3×10³ copies/m³ in the samples collected 4 and 5 days after treatment initiation. Briefly, a sharp decrease of P. jirovecii DNA load was observed between the first and the third air samples (Table 1).

TABLE 1 Results of Quantification and Genotyping of Pneumocystis jirovecii DNA in Pulmonary Samples and 1-m Air Sample From Patients With Pneumocystis Pneumonia in the Present Study and Published Elsewhere

^a Genotypes at the ITS locus described by Damiani et alReference Damiani, Choukri and Le Gal ⁸ for the patients initially studied by Choukri et al.Reference Choukri, Menotti and Sarfati ²

^b Not determined.

^c Partial match.

^d 2 pairs of pulmonary and 1-m air samples during PCP treatment in patient P10 by Le Gal et al.Reference Le Gal, Pougnet and Damiani ³ Considering only the results with perfect genotype matching in pairs of pulmonary and air samples, P. jirovecii was effectively detected at 1-m from 2 patients (patients P4 and P8 initially studied by Choukri et al.Reference Choukri, Menotti and Sarfati ² ) who had been treated for 2–3 days.

A CYB2 allele was identified in the BAL and the first 3 air samples, but typing at this locus did not give positive results in the last 2 air samples. MtLSUrRNA allele 4 was identified in the BAL and in the 5 air samples. Thus, a perfect match of P. jirovecii genotypes was observed. These results are consistent with the fact that P. jirovecii detected in air samples was from the patient’s source and had been exhaled in his environment.

We obtained the first data on P. jirovecii exhalation by a patient during the course of PCP treatment, specifically during the first 5 days but not beyond because the patient was moved from the nephrology unit. The exhalation of P. jirovecii decreased dramatically during the first 3 days of treatment. The low fungal burden in the last 2 samples (<1.3×10³ copies/m³) may be explained by a genuine P. jirovecii exhalation but at a low level or by a residual presence of the fungus in the patient’s room, even though the patient may no longer have been exhaling P. jirovecii. This latter hypothesis may be consistent with the observations by Bartlett et al,Reference Bartlett, Vermund and Jacobs ⁷ who detected P. jirovecii DNA in empty hospital rooms within infectious disease and AIDS units.

We compared our results to those previously reported by Choukri et al,Reference Choukri, Menotti and Sarfati ² Le Gal et al,Reference Le Gal, Pougnet and Damiani ³ and Damiani et al,Reference Damiani, Choukri and Le Gal ⁸ even though their objective was not the analysis of PCP treatment effect on P. jirovecii exhalation (Table 1). Considering only the results with perfect genotype matching in pairs of pulmonary and air samples, P. jirovecii was effectively detected at 1 m from 2 patients who had been treated for 2–3 days, which is consistent with the results of our present study.Reference Choukri, Menotti and Sarfati ^{2 ,} Reference Le Gal, Pougnet and Damiani ^{3 ,} Reference Damiani, Choukri and Le Gal ⁸

Available data on PCP case clusters in hospitals combined with recent studies of P. jirovecii exhalation by infected patients (the present study, Choukri et al,Reference Choukri, Menotti and Sarfati ² Le Gal et al,Reference Le Gal, Pougnet and Damiani ³ Yiannakis et al,Reference Yiannakis and Boswell ⁴ and Damiani et alReference Damiani, Choukri and Le Gal ⁸ ) support the hypothesis of patient-to-patient transmission of P. jirovecii via the airborne route and nosocomial infection occurrence. Although chemoprophylaxis remains essential, the implementation of collective measures is necessary. The putative transmissible stage of P. jirovecii is the ascus.Reference Cushion, Linke and Ashbaugh ⁹ Unfortunately, its median size is 5 µm, precisely the threshold value for choosing droplet versus air precaution measures. Pneumocystis jirovecii detection in the surrounding air at 5 m reported elsewhereReference Choukri, Menotti and Sarfati ^{2 ,} Reference Le Gal, Pougnet and Damiani ³ provides arguments in favor of air precautions, while the infectivity and viability of P. jirovecii in this context remain unknown. At present, the CDC recommends applying standard precautions and avoiding placement of a PCP patient in the same room with an immunocompromised patient.Reference Siegel, Rhinehart, Jackson and Chiarello ⁵ Considering knowledge acquisition on Pneumocystis transmission for the past 10 years, these recommendations should be updated.

Finally, our study shows that PCP treatment dramatically decreased P. jirovecii exhalation and supports maintaining preventive measures, whatever they may be, over at least 5 days after PCP treatment initiation.