Faculty Bibliography

BACKGROUND:Hurricanes are severe weather events that can disrupt power, water, and transportation systems. These disruptions may be deadly for patients requiring maintenance dialysis. We hypothesized that the mortality risk among patients requiring maintenance dialysis would be increased in the 30 days after a hurricane. METHODS:Patients registered as requiring maintenance dialysis in the United States Renal Data System who initiated treatment between January 1, 1997 and December 31, 2017 in one of 108 hurricane-afflicted counties were followed from dialysis initiation until transplantation, dialysis discontinuation, a move to a nonafflicted county, or death. Hurricane exposure was determined as a tropical cyclone event with peak local wind speeds ≥64 knots in the county of a patient's residence. The risk of death after the hurricane was estimated using time-varying Cox proportional hazards models. RESULTS:The median age of the 187,388 patients was 65 years (IQR, 53-75) and 43.7% were female. There were 27 hurricanes and 105,398 deaths in 529,339 person-years of follow-up on dialysis. In total, 29,849 patients were exposed to at least one hurricane. Hurricane exposure was associated with a significantly higher mortality after adjusting for demographic and socioeconomic covariates (hazard ratio, 1.13; 95% confidence interval, 1.05-1.22). The association persisted when adjusting for seasonality. CONCLUSIONS:Patients requiring maintenance dialysis have a higher mortality risk in the 30 days after a hurricane.

A recent study concluded that SARS-CoV-2 mRNA vaccine responses were improved among transplant patients taking mTOR inhibitors (mTORi). This could have profound implications for vaccine strategies in transplant patients; however, limitations in the study design raise concerns about the conclusions. To address this issue more robustly, in a large cohort with appropriate adjustment for confounders, we conducted various regression- and machine learning-based analyses to compare antibody responses by immunosuppressive agents in a national cohort (n = 1037). MMF was associated with significantly lower odds of positive antibody response (aOR = _0.09 0.13_0.18 ). Consistent with the recent mTORi study, the odds tended to be higher with mTORi (aOR = _1.00 1.45_2.13 ); however, importantly, this seemingly protective tendency disappeared (aOR = _0.47 0.73_1.12 ) after adjusting for MMF. We repeated this comparison by combinations of immunosuppression agents. Compared to MMF + tacrolimus, MMF-free regimens were associated with higher odds of positive antibody response (aOR = _2.39 4.26_7.92 for mTORi+tacrolimus; _2.34 5.54_15.32 for mTORi-only; and _6.78 10.25_15.93 for tacrolimus-only), whereas MMF-including regimens were not, regardless of mTORi use (aOR = _0.81 1.54_2.98 for MMF + mTORi; and _0.81 1.51_2.87 for MMF-only). We repeated these analyses in an independent cohort (n = 512) and found similar results. Our study demonstrates that the recently reported findings were confounded by MMF, and that mTORi is not independently associated with improved vaccine responses.

BACKGROUND:Solid organ transplant recipients (SOTRs) are less likely to mount an antibody response to SARS-CoV-2 mRNA vaccines. Understanding risk factors for impaired vaccine response can guide strategies for antibody testing and additional vaccine dose recommendations. METHODS:Using a nationwide observational cohort of 1031 SOTRs, we created a machine learning model to explore, identify, rank, and quantify the association of 19 clinical factors with antibody responses to 2 doses of SARS-CoV-2 mRNA vaccines. External validation of the model was performed using a cohort of 512 SOTRs at Houston Methodist Hospital. RESULTS:Mycophenolate mofetil use, a shorter time since transplant, and older age were the strongest predictors of a negative antibody response, collectively contributing to 76% of the model's prediction performance. Other clinical factors, including transplanted organ, vaccine type (mRNA-1273 versus BNT162b2), sex, race, and other immunosuppressants, showed comparatively weaker associations with an antibody response. This model showed moderate prediction performance, with an area under the receiver operating characteristic curve of 0.79 in our cohort and 0.67 in the external validation cohort. An online calculator based on our prediction model is available at http://transplantmodels.com/covidvaccine/. CONCLUSIONS:Our machine learning model helps understand which transplant patients need closer follow-up and additional doses of vaccine to achieve protective immunity. The online calculator based on this model can be incorporated into transplant providers' practice to facilitate patient-centric, precision risk stratification and inform vaccination strategies among SOTRs.