Prognostic Factors Affecting Death in Patients with Rheumatoid Arthritis Complicated by Pneumocystis jirovecii Pneumonia and One-Year Clinical Course: The ANSWER Cohort Study

Severe Outcomes of Pneumocystis Pneumonia: A 10-year Retrospective Cohort Study

BACKGROUND

A considerable knowledge gap exists in predicting severe Pneumocystis pneumonia (PCP) outcomes following PCP diagnosis.

METHODS

In this retrospective cohort, we studied immunocompromised patients with PCP admitted to 5 University Health Network centers in Canada (2011-2022). The study outcome included severe PCP, a composite of 21-day ICU admission or 28-day all-cause mortality. Adjusted odds ratios (aOR) estimated the association between severe PCP and comorbidities as well as clinical and laboratory variables at diagnosis.

RESULTS

A total of 44 out of 182 (24.2%) immunocompromised patients (19 [10.4%] HIV-infected, 55 [30.2%] hematologic malignancies, 32 [17.6%] hematopoietic stem cell transplants, 32 [17.6% solid tumors, 26 solid organ transplants [14.3%], 12 (6.6%) autoimmune diseases, and 6 (3.3%) other immunosuppressive conditions) developed composite outcomes (40 ICU admissions [21.9%], 18 deaths [9.9%]). Patients with composite outcomes more often had acute-onset PCP (< 7 days) (18/34 [52.9%] vs. 38/126 [30.1%], p = 0.013), shortness of breath (39/44 [88.6%] vs. 96/136 [70.6%], p = 0.002), chronic liver disease (15/44 [34.1%] vs. 9/138 [6.5%], p < 0.001), hypoalbuminemia (median [IQR] albumin (g/L): 27 [25-31] vs. 32 [29-35], p < 0.001), elevated lactate dehydrogenase (median [IQR] LDH (U/L): 537 [324-809] vs. 340 [237-475], p < 0.001), lymphopenia (median [IQR] absolute lymphocyte count [(10*9/L),]: 0.4 [0.2-0.6] vs. 0.7 [0.3-1.2], p < 0.001), or required supplemental oxygen (39/44 [88.6%] vs. 60/136 [44.1%], p < 0.001) than those without composite outcomes. In multivariable analysis, chronic liver disease (aOR: 11.6, 95% CI: 2.2-61.3) and requiring supplemental oxygen on admission (aOR: 19.7, 95% CI: 3.0-128.5) were significantly associated with severe PCP.

CONCLUSIONS

Alongside hypoxemia upon admission, chronic liver disease appears to significantly predict severe PCP in immunocompromised patients. This biologically plausible finding warrants further investigation.

Comparison of the outcomes of Pneumocystis jirovecii pneumonia in rheumatoid arthritis patients treated with and without biologics

BACKGROUND

To investigate the outcomes of Pneumocystis jirovecii pneumonia (PCP) between patients with rheumatoid arthritis (RA) treated with and without biologics before PCP onset.

PATIENTS AND METHODS

We retrospectively included rheumatoid arthritis (RA) patients with PCP treated with and without biologics before PCP onset. The primary endpoints were 30-day and 180-day survival rates, and the secondary endpoint was severe PCP, including in-hospital death, intensive care unit admission, and requirement of respiratory support during hospitalization.

RESULTS

Eighty-two patients were enrolled in this study, including the Biologics group (n = 39) and Non-Biologics group (n = 43). There were no significantly differences in the 30-day and 180-day survival rates and severe PCP rate in the Biologics group and the Non-Biologics group before and after adjusting the patient characteristics. Kaplan-Meier survival curves for death showed no significantly differences between the Biologics and Non-Biologics groups. Cox regression hazard analysis revealed that the average daily prednisolone dose within 90 days before PCP onset was weakly associated with mortality after PCP.

CONCLUSIONS

Biologic use before PCP onset did not increase the severity and mortality of PCP compared to non-biologics use in patients with RA.

Machine Learning Models for Prediction of Severe Pneumocystis carinii Pneumonia after Kidney Transplantation: A Single-Center Retrospective Study

BACKGROUND

The objective of this study was to formulate and validate a prognostic model for postoperative severe Pneumocystis carinii pneumonia (SPCP) in kidney transplant recipients utilizing machine learning algorithms, and to compare the performance of various models.

METHODS

Clinical manifestations and laboratory test results upon admission were gathered as variables for 88 patients who experienced PCP following kidney transplantation. The most discriminative variables were identified, and subsequently, Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), K-Nearest Neighbor (KNN), Light Gradient Boosting Machine (LGBM), and eXtreme Gradient Boosting (XGB) models were constructed. Finally, the models' predictive capabilities were assessed through ROC curves, sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), and F1-scores. The Shapley additive explanations (SHAP) algorithm was employed to elucidate the contributions of the most effective model's variables.

RESULTS

Through lasso regression, five features-hemoglobin (Hb), Procalcitonin (PCT), C-reactive protein (CRP), progressive dyspnea, and Albumin (ALB)-were identified, and six machine learning models were developed using these variables after evaluating their correlation and multicollinearity. In the validation cohort, the RF model demonstrated the highest AUC (0.920 (0.810-1.000), F1-Score (0.8), accuracy (0.885), sensitivity (0.818), PPV (0.667), and NPV (0.913) among the six models, while the XGB and KNN models exhibited the highest specificity (0.909) among the six models. Notably, CRP exerted a significant influence on the models, as revealed by SHAP and feature importance rankings.

CONCLUSIONS

Machine learning algorithms offer a viable approach for constructing prognostic models to predict the development of severe disease following PCP in kidney transplant recipients, with potential practical applications.

Risk Factors for Pneumocystis jirovecii Pneumonia in Non-HIV Patients Hospitalized for COVID-19: A Case-Control Study

BACKGROUND

Very few cases of Pneumocystis jirovecii pneumonia (PJP) have been reported in COVID-19 so far, and mostly in patients with concomitant HIV infection or in solid-organ transplant recipients. Despite COVID-19 being associated with lymphopenia and the use of steroids, there are no studies specifically aimed at investigating the risk factors for PJP in COVID-19.

METHODS

A retrospective case-control study was performed. We matched PJP cases with controls with a 1:2 ratio, based on age ± 10 years, solid-organ transplantation (SOT), hematological malignancies, and in the setting of PJP development (ICU vs. non-ICU). A direct immunofluorescence assay on bronchoalveolar lavage fluid was used to diagnose PJP.

RESULTS

We enrolled 54 patients. Among 18 cases of PJP, 16 were diagnosed as "proven". Seven of the eighteen cases were immunocompromised, while the other patients had no previous immunological impairment. Patients with PJP had significantly lower median lymphocyte values (p = 0.033), longer COVID-19 duration (p = 0.014), a higher dose of steroid received (p = 0.026), higher CRP values (p = 0.005), and a lower SARS-CoV-2 vaccination rate than the controls (p = 0.029). Cumulative steroid dose is the independent risk factor for PJP development (OR = 1.004, 95%CI = 1-1.008, p = 0.042).

CONCLUSIONS

PJP develops in COVID-19 patients regardless of immunosuppressive conditions and the severity of disease, and it is correlated to the corticosteroid dose received.