Characteristics, epidemiology and outcomes of post-transplant lymphoproliferative disorders associated hospitalizations: A 2016-2022 national inpatient sample analysis

Abstract Introduction: Post-transplant lymphoproliferative disorders (PTLD) are rare but serious lymphoid proliferations following solid organ or allogenic transplantation, driven by chronic immunosuppression. Degree of immunosuppression and positive Epstein-Barr virus (EBV) serologies are known risk factors for PTLD development. Given that trends and outcomes of PTLD-associated hospitalizations remain poorly characterized, we aimed to use the National Inpatient Sample (NIS) to investigate the demographics, risk factors, and hospitalization outcomes associated with PTLD-related hospitalizations. Methods: The NIS (2016-2022) data were queried for PTLD-associated hospitalizations using the International Classification of Diseases 10th version (ICD-10) codes. Categorical variables were compared using chi-squared tests and continuous values using Wilcoxon rank sum tests. Multivariable logistic regression models and linear regression were used for outcome analysis. Unsupervised clustering using Euclidean distance and Ward’s D2 method was applied based on 25 comorbidities, with the Davies–Bouldin index determining optimal cluster number. Cox-Proportional Hazards models were used for survival analysis. Results: From 2016-2022 there were 4694 PTLD-associated hospitalizations, with a median age of 52 years. The most common primary diagnosis was kidney transplant complications (5.9%). Most of the hospitalizations were among patients with solid organ transplant (N = 4054, 86%), with the top being renal transplant (N = 2029, 43%), followed by liver (N = 1015, 22%), and heart (740, 16%). With respect to demographics, most PTLD hospitalizations occurred in ages 45-64 (N = 1565, 33%) and 65+ (N = 1112, 24%), and among male patients (N = 2830, 60%). A total of 162 (3.5%) of hospitalizations were associated with tumor lysis syndrome per ICD-10 codes. Subgroup analysis revealed that TLS-associated PTLD hospitalizations were correlated with age > 45 (p=0.049), male (p<0.001), acute kidney injury (p<0.001), heart transplant (p = 0.029), sepsis (p = 0.007) and dialysis (p<0.001). PTLD hospitalizations with a TLS ICD-10 code exhibited increased mortality (N = 30, 19%, p<0.001) and longer length of stay (median 13 vs. 6 days, p<0.001). In a multivariate analysis, length of stay (LOS) was positively associated with dialysis (ꞵ=6.4, p<0.001) and bowel transplantation (ꞵ=3.6, p=0.015) but was negatively associated with chronic kidney disease (ꞵ=-2.9, p=0.013) and heart transplantation (ꞵ=-1.7, p=0.043). With respect to hospital charges, dialysis was associated with significantly higher total hospital charges (ꞵ= 87420, p < 0.001), while chronic kidney disease (ꞵ=-61148, p = 0.026), kidney transplant (ꞵ=-96273, p <0.001), and liver transplant (ꞵ=-68389, p <0.001) were associated with significantly lower charges among PTLD hospitalizations. Unsupervised analysis generated 12 clusters and indicated distinct phenotypes and survival patterns based on comorbidity burden. Specifically, cluster 9 (enriched for diabetes; HR 9.41, 95% CI 1.26–70.3), cluster 11 (characterized by hypertension; HR 18.6, 95% CI 2.6–134, p = 0.004), and cluster 12 (enriched for cytopenias; HR 9.54, 95% CI 1.25–72.7, p = 0.03) all exhibited decreased survival. Among PTLD associated admissions, both bowel (HR: 0.24, 0.089, 0.68, p = 0.007) and heart transplantation (HR: 0.57, 0.378, 0.86, p = 0.007) were associated with improved survival probabilities while TLS decreased survival (HR: 1.49, 1.00, 2.23, p = 0.049). When this model was age-adjusted, bowel transplantation (HR 0.45, 0.16, 1.2) and heart transplantation (HR 0.73, 0.48, 1.1) displayed protective survival hazards ratios but were no longer statistically significant. Discussion: This analysis captures a multi-year national sample of PTLD inpatient hospitalizations and highlights specific risk factors for tumor lysis syndrome, length of stay, and mortality. Heart transplant patients with PTLD demonstrate an association with TLS but interestingly improved survival probabilities. Our findings indicate differences in risk due to transplant type and co-morbidities among patients with PTLD. These results can inform risk stratification strategies and clinical guidelines for TLS monitoring among high-risk PTLD populations. Limitations of this analysis include incomplete EBV serostatus and lack of temporal data. Future research can offer more precise risk stratification and quantify TLS risk among PTLD patients.