EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Objectives: Pneumocystis jiroveci pneumonia is a rare but lethal complication in renal transplant recipients. Dose reduction of immunosuppressive agents in such situations is recommended, but its quantity and safety are unclear.

Materials and Methods: From January 2001 to January 2011, twenty of one thousand forty-six renal transplant recipients in a single center developed Pneumocystis jiroveci pneumonia, which was diagnosed by the Giemsa and Gomori methenamine silver stains from a specimen of bronchoalveolar lavage.

Results: We found that timing of the first immunosuppressant reduction of the Pneumocystis jiroveci pneumonia survivor (mean, 1.4 days after admission) was significantly earlier than that of the deceased patient (mean, 5.1 days after admission). Logistic regression analysis indicated that for those whose immunosuppressants were reduced more aggressively (either 1 of the immunosuppressants was reduced by more than 50% within 2 days of hospitalization) were significantly more likely to survive (mortality risk, OR, 0.074 [95% CI, 0.01-0.84]; P = .035). In addition, none of the survivors developed acute rejection or allograft necrosis during a mean follow-up of 2 years.

Conclusions: Dosage reduction of immunosuppressive agents in renal transplant recipients with Pneumocystis jiroveci pneumonia should be prompt and sufficient. Aggressive immunosuppressant dosage reduction is safe in such circumstance and is associated with minimal risk of in-hospital and long-term acute allograft rejection.

Key words : Pneumocystis jiroveci pneumonia, Immunosuppressants, Outcomes, Rejection, Kidney transplant

Pneumocystis jiroveci is a ubiquitous fungal pathogen worldwide. Immunocompromised patients including renal transplant recipients (RTRs) are at risk for infection with Pneumocystis jiroveci pneumonia (PJP). Although the routine use of PJP chemoprophylaxis with trimethoprim-sulfamethoxazole (TMP-SMX) in posttransplant period of the first 3 to 6 months,^1-4 PJP cannot be completely prevented.^{5, 6}

The disease course of PJP is extremely rapid and overwhelming, and its recovery depends on early diagnosis and prompt treatment.⁷ Pneumocystis jiroveci mortality in RTRs is high despite aggressive treatment, and it has been reported to be between 27% and 50%.^{3, 8} To improve outcomes of RTRs, previous studies have identified PJP risk factors in RTRs including poor renal function, prior allograft rejection, mycophenolate mofetil use, and persistent lymphocytopenia.^9-11 Moreover, to reduce PJP mortality rate in RTRs, it is crucial to identify determinants of PJP mortality that have not been investigated before. Meanwhile, for patients with PJP, KDIGO recommends reducing the amount of immunosuppressive medications.^{12, 13} Nevertheless, optimal timing and quantity of dosage reduction are unknown and deserve investigation.

We reasoned that in such an immunocompromised status, dosage reduction of the immunosuppressant should be prompt and sufficient enough. Therefore, we aimed to quantify the adequate timing and degree of immunosuppressant dosage reduction that leads to better pneumonia survival. Furthermore, we also examined long-term outcomes in PJP survivors.

Materials and Methods

Study protocol and subjects
From January 2001 to January 2011, there were 1146 renal transplant patients followed-up at the Taipei Veterans General Hospital, a tertiary-care referral hospital. Pneumocystis jiroveci chemoprophylaxis (80/400 mg/d) was routinely administered in the first 6 months after kidney transplant. Twenty of them developed PJP infection and the incidence was 1.7%. Pneumocystis jiroveci infection was diagnosed by bronchoalveolar lavage performed by chest physicians. The bronchoalveolar lavage specimens were sent for Giemsa and Gomori’s methenamine silver stains. The demographic features and clinical parameters, including age, sex, comorbidities, symptoms at presentation, concomitant pulmonary infection, infection severity (eg, Acute Physiology and Chronic Health Evaluation II [APACHE II] score on the admission day), laboratory data, and patient outcomes were recorded. Estimated glomerular filtration rate (eGFR) was calculated by the simplified Modification of Diet in Renal Disease formula. Absolute neutrophil count was calculated by total white blood cell count multiplied by neutrophil percentage. Absolute lymphocyte count was calculated by total white blood cell count multiplied by lymphocyte percentage. All protocols were approved by the ethics committee of the institution before the study began, and the protocols conformed with the ethical guidelines of the 1975 Helsinki Declaration. The need for informed consent was waived because of the retrospective nature of the study.

Immunosuppressant regimens
Immunosuppressant regimens of patients consisted of oral corticosteroids, tacrolimus, cyclosporine, mycophenolate mofetil, and sirolimus. Because of the retrospective nature of our study, the choice and dosage of immunosuppressants were at the discretion of each nephrology attending physician. The daily dosage, treatment duration, and dosage adjustment of each immunosuppressant agent were collected.

Management of Pneumocystis jiroveci pneumonia infection
Pneumocystis jiroveci pneumonia patients were diagnosed and treated aggressively. Standard PJP treatment protocol including intravenous high-dose TMP-SMX (TMP 15-20 mg/kg/d, adjusted according to eGFR) and corticosteroids were initiated timely, even before the bronchoalveolar lavage examination when suspicious clinically. Concomitant broad-spectrum antibiotics, mechanical ventilation, and hemodialysis and/or continuous renal replacement therapy were applied timely if indicated.

Statistical Analyses
The chi-square or Fisher exact test was used to compare categorical variables, as appropriate. Continuous variables were compared by the t test. Values of continuous variables are presented as means and standard deviation, unless otherwise specified. The logistic regression analysis was used to determine the significance of variables in predicting short-term outcome and in-hospital mortality owing to PJP infection. Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 15.0, IBM Corporation, Armonk, New York, USA). All probabilities were 2-tailed, and a value for P < .05 was considered significant.

Baseline characteristics of study subjects
Table 1 shows the baseline characteristics of the 20 patients (mean age, 55 y; male, 45%; 30% had diabetes mellitus). The average time of PJP onset was 2.7 years after kidney transplant. A total of 16 patients (80%) had fever at presentation, 14 patients (70%) had dyspnea, 12 patients (60%) had cough, 9 patients (45%) had scanty sputum, and 2 patients (10%) had hemoptysis. There were no differences in symptoms between living or dead patients, except that dead RTRs were significantly more likely to have dyspnea at presentation than those who survived (P = .011).

Among patients with PJP development (n=20), 5 patients had concurrent pulmonary infection with Pseudomonas aeruginosa (25%), 6 had Acinetobacter baumannii (30%), and 3 had Enterobacteriaceae spp. (15%). Meanwhile, Candida spp. was isolated in 6 patients. There were no differences in acute rejection history, cytomegalovirus infection history, and polyomavirus infection history between live and dead patients.

Furthermore, we compared the disease severity between live and dead patients. Our data show that there were no differences between these 2 groups of patients, including APACHE II score on admission day, bacteremia rate, concomitant urinary tract infection rate, serum albumin, creatinine, eGFR, absolute neutrophil count, and absolute lymphocyte count (except for the percentage of inotropic agent use [live vs dead patients, 10% vs 100%]; P < .001).

Immunosuppressant regimens before Pneumocystis jiroveci hospitalization
Table 2 shows the regimens of the immunosuppressant agents. Compared to PJP survivors, deceased patients had a significantly higher daily dosage of corticosteroids and had a higher serum trough level of tacrolimus before PJP hospitalization, which was related to their shorter posttransplant vintage. The mean onset of PJP of the deceased patients (1.8 years posttransplant) was shorter than that of the survivors (3.6 years posttransplant; but this did not achieve the statistical significance).

Immunosuppressant dosage reduction after Pneumocystis jiroveci hospitalization
We compared the immunosuppressant dosage reduction between patients that lived and those that died. As shown in Table 2, timing of immunosuppressant dosage reduction of the survivors was significantly earlier than that of the deceased. Fifty percent of the survivors’ immunosuppressive agents had been reduced on the first day of hospitalization, but none of those who died did (P = .033). Meanwhile, the timing of dosage reduction of the first immunosuppressant of the survivors (1.4 days postadmission) was significantly earlier than those that died (5.1 days postadmission; P = .005).

Determinants of Pneumocystis jiroveci mortality
As shown in Table 3, logistic regression analysis indicated that for each 1 day late in reducing the immunosuppressive agents, there was a 1.966-fold risk of increase in PJP mortality (P = .023). In addition, not only the timing, but also the quantity, of immunosuppressant reduction matters. If either one of the immunosuppressive agent’s dosage was reduced more than 50% within 2 days of hospitalization, there was a 92.6% risk reduction in PJP mortality (P = .035).

Clinical outcomes
Unfortunately, 10 of the 20 PJP patients died. Our PJP patients presented with a high mortality rate (50%) and had a poor prognosis despite aggressive therapies, as evidenced by the rapid initiation of high-dose parenteral TMP-SMX (average initiation time: 2.8 days postadmission). All deceased patients developed respiratory failure that required mechanical ventilation, and all of them developed graft failure that required renal replacement therapy. In contrast, PJP survivors had a significantly lower proportion of intubation requirement (40%; P = .011) and requirement for renal replacement therapy (30%; P = .003).

Table 4 summarizes the long-term outcomes of the PJP survivors. We found that all patients were successfully weaned from the mechanical ventilator during the PJP hospitalization. Mean duration of dosage reduction of immunosuppressants for more than 50% of the PJP survivors was 21.9 ± 4.5 days. Two survivors began to depend on dialysis therapy before discharge. Graft loss occurred in 2 other patients 2.6 and 11.7 months after discharge. The former patient encountered a fatal arrhythmia during hemodialysis therapy 5.8 months after discharge and died. There were no episodes of suspicious clinical acute rejection or allograft necrosis occurred during a mean follow-up of 2 years.

The findings of this study indicate that disease severity is equivalent between those patients that lived and those that died. Aggressive dosage reduction of immunosuppressive agents is beneficial for PJP recovery in RTRs. Furthermore, we did not identify any clinically suspected rejection episodes or any allograft necrosis in the survivors, whose immunosuppressant had been reduced much more drastically. We suggest that prompt and sufficient dosage reduction of immunosuppressants is life-saving in such circumstance and does not increase the risk of long-term acute rejection.

The concern of immunosuppressant dosage reduction in RTRs is mostly related to the fear of future allograft rejection or necrosis. Sileri and associates have proposed a standardized protocol for treating severe pneumonia in RTRs. In their protocol, the calcineurin inhibitors were withheld for the first 2 or 3 days. There were no acute rejection episodes during hospitalization,⁷ which is in accordance with our PJP survivors’ outcomes. In our study, we did not identify any clinically evident acute rejection in the PJP survivors, whose immunosuppressive agents had been significantly reduced for an average duration of 3 weeks. We further followed-up our patients for a mean duration of 2 years, and no acute allograft rejection or necrosis was documented.

The majority of studies has reported the PJP occurs 3 to 6 months after renal transplant.^{1, 4, 14-18} In our study, the average time of PJP onset was 2.7 years after renal transplant. The late onset of our PJP cases may be due to the routine first 6-month PJP chemoprophylaxis in our institute. The incidence of PJP is 1.7% in our RTRs, probably because we did not routinely give PJP chemoprophylaxis after episodes of allograft rejection. Nevertheless, a large scale study has shown that prior acute rejection is not associated with development of PJP disease.⁵ On the other hand, our data show that a prior acute rejection history, low eGFR, and lymphocytopenia are not determinants of PJP mortality.

In our study, the daily dosage of corticosteroids and the serum level of tacrolimus of the dead patients before admission were significantly higher than those who survived. However, the posttransplant duration of the dead patients was shorter. This is because the recommended targets of immunosuppressive drug level were different depending on their posttransplant vintage.¹⁹ However, though the baseline tacrolimus level of the live patients was already lower than that of the dead patients, the immunosuppressive agents of the live patients was tapered more aggressively. Outcome data of the survivors revealed that such aggressive immunosuppressant dosage reduction is life-saving, and its long-term risk of acute allograft rejection is minimal.

Limitations of our study are the small number of PJP patients and its retrospective design. However, the incidence of PJP in RTRs decreases after the introduction of routine PJP chemoprophylaxis. Despite the patient number being limited, we found a significant association between the aggressiveness of immunosuppressant dosage reduction and PJP survival.

In conclusion, this study indicates that immunosuppressant dosage reduction should be prompt and sufficient for immunity recovery. In addition, our results suggest that aggressive immunosuppressant dosage reduction possesses a minimal risk of in-hospital and long-term acute rejection.

Volume: 10 Issue: 4 August 2012

FULL TEXT