Malhotra, Atul

ABSTRACT Background Systemic infections are a leading cause of hospitalization and death among patients with cirrhosis. Timely and accurate infection identification is essential for both clinical care and the development of predictive models. However, existing methods such as ICD-10 coding are unreliable, and manual chart review is resource-intensive and difficult to scale. This study aimed to develop and validate an automated large language model (LLM)-based approach for infection classification and subtyping in patients with cirrhosis presenting to the emergency department (ED). Method We developed INFEHR (INfection identification and subtyping using Free-text EHR analysis), an LLM-powered pipeline utilizing Claude 3.5 Sonnet to analyze clinical notes from the first 72 hours of admission. Model outputs were compared against a physician-adjudicated gold standard in a cohort of 1,000 encounters from patients with cirrhosis who presented to the ED. Performance was benchmarked against ICD-10 code–based labeling and CDC Adult Sepsis Event criteria. Results INFEHR achieved 94.7% overall accuracy, with 99.5% sensitivity and 92.8% positive predictive value for identifying infection presence, outperforming ICD-10–based classification across all metrics ( p < 0.0001). The model also demonstrated strong performance in classifying pathogen type and infection site. This pipeline processed notes within seconds, offering improvements in efficiency and scalability over manual review. Conclusion INFEHR offers a scalable, reproducible, and accurate method for infection phenotyping in cirrhosis. By overcoming limitations of traditional coding and manual review, it supports high-throughput infection surveillance, improves cohort construction for clinical research, and enables future integration into real-time decision-support tools in hepatology.

Abstract Background: Obstructive sleep apnea (OSA) is associated with obesity and cardiovascular risk. Results from the SURMOUNT-OSA studies demonstrated significant reduction of a number of cardiometabolic risk measures in participants with OSA and obesity following treatment with tirzepatide. Objectives: Systematically analyze all tirzepatide treatment related changes of cardiometabolic risk measures in SURMOUNT-OSA studies and investigate to what extent these changes were mediated by reductions in weight and/or OSA-related measures. Methods: The SURMOUNT-OSA master protocol comprised two 52-week, randomized, double-blind, placebo-controlled Phase 3 studies. Study 1 (N=234) and 2 (N=235) included participants with moderate-to-severe OSA and obesity, who at screening reported not using, or using positive airway pressure (PAP) therapy, respectively. The design and rationale of the studies have been published previously (1). The current analysis investigated the association of treatment with tirzepatide or placebo with SURMOUNT-OSA pre-specified cardiometabolic risk measures including blood pressure, and levels of high-sensitivity C-reactive protein (hsCRP), HDL-C, non-HDL-C, triglycerides, fasting insulin. Further analyses included change in HOMA-IR with tirzepatide compared to placebo, and mediation analysis to determine the proportion of observed changes attributable to reductions in body weight, apnea-hypopnea index (AHI) and sleep apnea-specific hypoxic burden (SASHB). Results: In Study 1, participants experienced greater difference in change from baseline with tirzepatide compared to placebo in SBP (-7.9 mmHg; 95% CI -11.0 to -4.9; p<0.001), and DBP (-3.2 mmHg; 95% CI -5.4 to -1.0; p=0.005) at 48 weeks, and hsCRP (-28.9±8.2%; p=0.003), HDL-C (7.2±2.1%; p<0.001), non-HDL-C (-13.0±3.1%; p<0.001), triglycerides (-32.2±3.7%; p<0.001), fasting insulin (-41.4±5.0%; p<0.001), and HOMA-IR (-48.0±4.9%; p<0.001) at 52 weeks. In Study 2, participants experienced greater differences in change from baseline following tirzepatide compared to placebo in SBP (-4.3 mmHg; 95% CI -7.3 to -1.2; p=0.007) at 48 weeks, and hsCRP (-45.1±8.0%; p<0.001), HDL-C (10.0±2.8%; p<0.001), non-HDL-C (-14.3±2.5%; p<0.001), triglycerides (-31.5±3.7%; p<0.001), fasting insulin (-45.4±5.4%; p<0.001), and HOMA-IR (-54.8±5.0%; p<0.001) at 52 weeks. Statistically significant independent mediation effect of changes in OSA metrics was observed on hsCRP, HOMA-IR and triglycerides. In addition, the combination of changes in weight and OSA metrics, but not weight or OSA improvements alone, had a significant mediation effect on non-HDL-C. There was no significant mediation effect of weight or OSA metrics observed on DBP. Conclusions: In the SURMOUNT-OSA program trials, tirzepatide appeared to have significant cardiometabolic benefits for patients with moderate-to-severe OSA and obesity. Based on the mediation analysis, treating both sleep-disordered breathing and obesity is likely required to optimize cardiometabolic benefits for patients with OSA and obesity.

Abstract Central sleep apnea, a rare polysomnographic finding in the general population, is prevalent in certain cardiovascular conditions including systolic and diastolic left ventricular dysfunction, atrial fibrillation, coronary artery disease, carotid artery stenosis, stroke, and use of certain cardiac-related medications. Polysomnographic findings of central sleep apnea with adverse cardiovascular impacts include nocturnal hypoxemia and arousals, which can lead to increased sympathetic activity both at night and in the daytime. Among cardiovascular diseases, central sleep apnea is most prevalent in patients with left ventricular systolic dysfunction; a large study of more than 900 treated patients has shown a dose-dependent relationship between nocturnal desaturation and mortality. Multiple small randomized controlled trials have shown mitigation of sympathetic activity when central sleep apnea is treated with nocturnal oxygen, continuous positive airway pressure, and adaptive servoventilation. However, two early randomized controlled trials with positive airway pressure devices have shown either a neutral effect on survival or excess premature mortality in the active treatment arm, compared to untreated central sleep apnea. In contrast, the results of the most recent trial using an advanced adaptive servoventilation device showed improved quality of life and no signal for mortality suggesting that treatment of central sleep apnea was at least safe. In addition to positive airway pressure devices, multiple medications have been shown to improve central sleep apnea, but no long-term trials of pharmacologic therapy have been published. Currently, phrenic nerve stimulation is approved for the treatment of central sleep apnea, and the results of a randomized controlled trial showed significant improvement in sleep metrics and quality of life.