Pneumocystis jiroveci pneumonia in kidney and simultaneous pancreas kidney transplant recipients in the present era of routine post-transplant prophylaxis: risk factors and outcomes

Pneumocystis jirovecii pneumonia (PJP), previously known as Pneumocystis carinii pneumonia, is an opportunistic fungal infection observed in immunocompromised individuals. Prior to institution of routine prophylaxis in kidney transplant recipients, the incidence was reported at 5 to 15% and was the highest in the first 6 months after transplantation [1–6]. Prophylaxis with trimethoprim-sulfamethoxazole (TMP-SMZ) is highly effective and is now routinely administered for the initial 6 to 12 months post-transplantation at most transplant centers in the United States [7]. This has drastically reduced the incidence of PJP during the initial highest risk period immediately following transplantation. However, the epidemiology of PJP in the modern era of universal prophylaxis and risk factors for late onset infection in the renal transplant population have not been fully elucidated. There is no consensus on which high-risk individuals may benefit from prolongation or re-initiation of prophylaxis [8, 9].

At our center, administration of PJP prophylaxis is protocolized for the initial 6 to 12 months immediately following transplant. The primary objectives of this study were: 1) to identify risk factors for PJP that, in turn, could provide guidance regarding when prolongation or re-initiation of prophylaxis may be warranted, and 2) to describe outcomes in kidney transplant and simultaneous pancreas kidney recipients whose post-transplant course has been complicated by PJP.

The present study from a high-volume transplant center spanning the course of over two decades provides several insights into the epidemiology of PJP in kidney and simultaneous kidney and pancreas transplant recipients in the present era of routine post-transplant prophylaxis. First, the overall incidence of PJP in the setting of protocolized universal prophylaxis in the initial 6–12 months is low, at less than 1 %, compared to the historically reported statistic from pre-prophylaxis era of 5 to 15% [1–6]. A recent retrospective analysis of all solid organ transplant recipients from the Swiss Transplant Cohort Study reported an incidence of PJP of 1.4%; the incidence was higher at 3.2% in the 13.3% of the recipients who did not receive primary prophylaxis vs. 1.2% in those who received primary prophylaxis [11]. In our study, only three cases of PJP were reported in the first year following transplantation. In each of these, prophylaxis with TMP-SMZ had been stopped prematurely due to medication adverse effects. Additionally, consistent with recent studies, the immediate post-prophylaxis period, i.e. the second year post-transplant, has replaced the first year post-transplant as the highest risk period for developing this opportunistic infection with almost a third of the cases diagnosed during this time [8, 11–13]. Interestingly, this trend is similar to what is observed in the CMV literature, where routine prophylaxis has postponed the highest risk period [11, 14]. It should be highlighted that this change does not reflect just a shift in the period of peak PJP risk, and that the incidence of PJP (overall as well as in the second year) is extremely low compared to the pre-prophylaxis era. Also, PJP can occur very late after transplantation, with a quarter diagnosed a decade after transplant in our study.

Secondly, CMV viremia is a clinically strong and statistically significant predictor of subsequent PJP infection [8, 9, 11–13, 15–18]. Our study was not designed to identify which patients with CMV viremia, a common infection after transplantation, subsequently develop PJP, a rare complication [3]. Nonetheless, it is noteworthy that over 90% of the transplant recipients with PJP who had history of prior CMV viremia had been diagnosed with CMV in the year preceding diagnosis of PJP, and almost all of these were still viremic and/or receiving treatment at the time of admission for PJP. Median time from diagnosis of CMV viremia to diagnosis of PJP was 3.4 months, which argues against simple detection of stress-induced viremia in the setting of impending PJP syndrome. Previous literature echoes our findings; Iriart and colleagues previously reported a 52% incidence of CMV viremia in the year preceding PJP in a study of all solid organ transplants, including kidney, heart and liver [8]. Lee and colleagues reported a median time between CMV viremia and PJP infection of almost 2 months [9]. These results from our and prior studies provide an opportunity to identify this high-risk cohort; we suggest consideration of extending or re-initiating prophylaxis for at least 6 months in recipients whose course is complicated by CMV viremia. Lastly, while history of BK viremia and invasive fungal infections were more common in transplant recipients who developed PJP, the significance of relationship between these and PJP was lost on multivariate analysis. However, extension or re-initiation of prophylaxis should also be considered in recipients with CMV viremia and BK or fungal co-infections.

The association between rejection and risk of subsequent PJP remains unclear. Contrary to what might be expected and findings from some other investigations [13, 18], rejection did not prove to be significantly associated with subsequent diagnosis of PJP. This may be a result of confounding by the protocolized use of 3 months of PJP prophylaxis in the setting of rejection treatment at our center, and is consistent with some recent studies [8, 9]. Similarly, older age was not a predictor of PJP in our study. Findings from previous studies are divided, with the Iriart and Neofytos studies describing age greater than 65 years as a risk factor, and the Lee study demonstrating no significant difference [8, 9, 11].

In addition to rejection and older age, several other variables frequently used as proxies for the degree of immunosuppression, including T-cell depleting induction and BK viremia, were not significant risk factors for subsequent diagnosis of PJP. These results suggest that the increased risk of PJP in recipients with CMV viremia may be a consequence of the substantial immunomodulatory effects of the virus, rather than a reflection of the overall level of immunosuppression alone [19].

Finally, despite a reduction in overall incidence in the modern era, patient as well as graft outcomes continue to be poor among recipients whose course is complicated by PJP [20]. Over half of our patients with PJP died within 2 years. This underscores the importance of identifying patients at high risk for PJP who are likely to benefit from highly effective PJP prophylaxis with TMP-SMZ or alternate agents.

This study has all the limitations of being a small series from a single center. However, data on all our transplant patients is collected prospectively, we analyzed matched controls and our database is one of the few in the country that would be large enough to provide this series. Secondly, individual maintenance immunosuppressive regimens were not analyzed. However, immunosuppressive regimens have been fairly consistent throughout the study time period with a triple drug regimen of a calcineurin-inhibitor, antimetabolite and corticosteroid being our standard of care. Thirdly, preceding lymphopenia has been previously shown to be a risk factor for PJP [8]. Corresponding data was missing for many of the PJP cases in our study and therefore could not included in our analysis. Lastly, the number of PJP cases may appear small, however, PJP is now a rare complication of kidney transplantation, and to the best of our knowledge, this is the largest study of kidney transplant recipients from the present era of routine TMP-SMZ prophylaxis.

In conclusion, in our review of over 6000 transplant recipients, routine PJP prophylaxis has drastically reduced the incidence of PJP to less than 1 %. The peak incidence time period has also been shifted from the initial 6 months to the second year after transplant. Patient survival in affected patients remains dismal, with 2-year mortality rate approaching 60% with treatment. CMV has consistently been shown to be a strong predictor of subsequent PJP and because of this, we suggest extension or re-initiation of PJP prophylaxis for at least 6 months in recipients with CMV viremia. Future studies are needed to elucidate the impact of this intervention on rates of PJP in those with CMV viremia, as well as the feasibility of this approach in this population from tolerability from a myelosuppression standpoint.