Hommos, Musab S.

Pancreatic exocrine insufficiency (PEI) is an under-recognised cause of chronic diarrhoea in patients with neuroendocrine tumours (NETs). If left untreated or undertreated, PEI can lead to complications such as oxalate nephropathy (ON), which may result in loss of kidney function and potentially limit oncologic treatment options, including peptide receptor radionuclide therapy (PRRT). We present a man in his late 60s with a small bowel NET who developed persistent, uncontrolled diarrhoea for several years. The diarrhoea was attributed to multiple factors, including secretory NET (ie, carcinoid syndrome), bile acid diarrhoea and PEI secondary to somatostatin analogue therapy. Due to financial constraints, the patient was unable to consistently take pancreatic enzyme replacement therapy (PERT). Years after the diagnosis of NET, the patient developed a new onset worsening of kidney function. Further work-up, including a kidney biopsy, confirmed ON, which significantly limited subsequent treatment options, including PRRT. Routine screening for PEI and early initiation of PERT should be considered in all NET patients with chronic diarrhoea to prevent ON and preserve future treatment options.

Abstract Introduction Cardiovascular disease (CVD) is the leading cause of death among patients with kidney failure on dialysis. Social determinants of health may worsen this risk but are seldom incorporated into prognostic models. ICD-10 Z-codes provide standardized documentation of social determinants of health, yet their prognostic value in dialysis populations remains uncertain. This study examined whether Z-coded social determinants of health are associated with adverse cardiovascular outcomes and mortality among patients receiving dialysis for kidney failure. Methods A retrospective cohort study was conducted using the TriNetX Research Network (2010–2024), encompassing de-identified data from 103 institutions. Adults with kidney failure on chronic dialysis were stratified by the presence of ICD-10 Z-codes (Z55–Z65). Propensity score matching (1:1) balanced demographics, comorbidities, medications, and laboratory values. Primary outcomes were all-cause mortality and hospitalization; secondary outcomes included acute heart failure, myocardial infarction, stroke, cardiac arrest, bradyarrhythmias, and ventricular tachycardia. Results After matching, mean age was 62 years, and 42.4% were female in the Z-code cohort versus 41.2% in controls. Over one year, mortality occurred in 12.5% versus 10.9% (hazard ratio [HR], 1.22; 95% confidence interval [CI], 1.15–1.29) and all-cause hospitalizations in 57.3% versus 53.1% (HR, 1.22; 95% CI, 1.19–1.26). Cardiovascular complications were consistently higher in Z-coded patients, including heart failure (24.3% vs 20.5%; HR, 1.29), myocardial infarction (8.0% vs 7.1%; HR, 1.19), stroke (6.6% vs 6.2%; HR, 1.13), cardiac arrest (3.8% vs 3.1%; HR, 1.29), and bradyarrhythmias (6.9% vs 5.5%; HR, 1.33). Discussion Among patients with kidney failure, documented Z-coded social risk factors were associated with greater mortality and cardiovascular morbidity. Incorporating social determinants of health documentation into nephrology care may improve risk stratification and inform targeted interventions.

Background/Objective: Hypertension is the most prevalent chronic disease in primary care and a leading cause of cardiovascular morbidity and mortality. Despite existing guidelines, therapeutic inertia and suboptimal control persist. Large language models (LLMs) offer a potential valuable addition to augment clinical decision-making, yet their reliability for guideline-driven tasks remains unverified. This study evaluated the accuracy and safety of hypertension management recommendations generated by three LLMs compared to expert responses. Methods: Fifty-one clinical vignettes representing 17 core hypertension management concepts were constructed by hypertension experts. Each case was submitted to three LLMs (GPT-4, Gemini, MedLM) and a hypertension expert also wrote the “gold standard” answers. Three blinded expert reviewers rated each response on a 4-point accuracy scale, a binary safety (safe/unsafe) scale, and attempted to identify the source (LLM vs. expert) providing the response. Ratings were analyzed using mean scores, percentages of accurate and safe responses, and inter-rater agreement. Results: GPT-4 had the highest accuracy (83%) and safety (86%) scores among LLMs but remained inferior to expert responses (92% accuracy, 93% safety). Gemini and MedLM performed significantly worse (accuracy: 64% and 35%; safety: 73% and 39%, respectively). GPT-4 generated the most guideline-concordant responses (46%) among the three LLMs (Gemini 35%, MedLM 14%), but remains lower than experts’ responses (68%). Evaluators misidentified LLM responses as expert-written in 10 to 25% of cases, particularly with GPT-4. Inter-rater reliability for accuracy ratings was highest for expert-generated responses (ICC 0.81), with progressively lower agreement for GPT-4 (0.76), Gemini (0.70), and MedLM (0.68). A similar pattern was observed for safety and source discrimination ratings. The agreement was strongest for safety assessments and weakest for source discrimination. Conclusion: Among three tested LLMs, GPT-4 demonstrated closer agreement to expert decisions thereby showing greater potential for supporting hypertension management. However, current LLMs’ versions frequently produce inaccurate or unsafe recommendations and remain inferior to expert judgment. Human-in-the-loop supervision remains essential when deploying LLMs for clinical decision-making.

BACKGROUND: Hypertension, the leading cause of cardiovascular mortality, remains suboptimally controlled. Large language models (LLMs) could improve hypertension control by augmenting clinical decision-making, but their reliability for guideline-driven tasks is unverified. This study evaluated the accuracy and safety of hypertension management recommendations generated by 3 LLMs. METHODS: Fifty-one vignettes were constructed and submitted to the LLMs (GPT-4, Gemini, Medical Large Language Model [by Google; MedLM]) and a hypertension expert to generate the responses. Three blinded reviewers rated each response on a 4-point accuracy scale, a binary safety (safe/unsafe) scale, and attempted to identify the source (LLM versus expert) providing the response. RESULTS: GPT-4 had the highest accuracy (83%) and safety (86%) scores among LLMs but remained inferior to expert responses (92% accuracy, 93% safety). Gemini and MedLM performed significantly worse (accuracy: 64% and 35%; safety: 73% and 39%, respectively). GPT-4 generated the most guideline-concordant responses (46%) among the 3 LLMs (Gemini 35%, MedLM 14%) but was lower than expert responses (68%). Interrater reliability for accuracy ratings was higher for LLM-generated responses (GPT-4 [intraclass correlation coefficient, 0.30], Gemini [intraclass correlation coefficient, 0.61], and MedLM [intraclass correlation coefficient, 0.58]), with lower agreement for expert responses (intraclass correlation coefficient, 0.23). A similar pattern was observed for safety and source discrimination ratings. The agreement was strongest for safety assessments and weakest for source discrimination. CONCLUSIONS: Among the 3 tested LLMs, GPT-4 demonstrated closer agreement to expert decisions, thereby showing greater potential for supporting hypertension management. Despite their potential, current LLM versions are inferior to expert recommendations. Human-in-the-loop supervision remains essential when deploying LLMs for clinical decision-making.

Background: Cardiovascular disease remains the most common cause of death among kidney transplant recipients, with hypertension being one of the most prevalent risk factors. Understanding the current prevalence of hypertension in this population is critical to help shape post-transplant care policies and guidelines and ensure proper management of CVD risk. Methods: We included all kidney transplant recipients who received their transplant between 2019 and 2023 at our center and who completed 24-hour ABPM at 12 months post-transplant. We reported data in mean (SD) unless noted otherwise. BP was categorized per AHA guidelines. This study was deemed exempt by the IRB. Results: A total of 940 patients were included in this study. Detailed results are seen in Figure 1 and Table 1. Notable to mention that 60% of the patients had blood pressure > 130/80 mmHg at 12 months post kidney transplant, with the majority being stage 1. Patients with more severe hypertension tended to be older, had lower eGFR, and had more comorbidities. They also had lower diastolic blood pressure than stage 1 HTN patients, consistent with older age. Night-time BP was significantly higher among stage 2 HTN patients. Conclusion: Hypertension remains a prevalent condition post-kidney transplant despite the majority of patients having satisfactory kidney allograft function. Ongoing BP monitoring and assessment of out-of-office BP post-kidney transplantation is critical to ensure control of this important cardiovascular risk factor. 24-hour ABPM has several advantages in this group, including accurate measurement of out-of-office BP and detection of nocturnal hypertension, an independent risk factor for cardiovascular events. Additional studies are needed to describe the hypertension subphenotypes (e.g., sustained vs. masked) and to identify predictors of severe uncontrolled hypertension in this patient population.

Purpose of review The frequency of natural disasters and man-made conflicts has risen significantly in the past two decades, coinciding with an increase in kidney transplant recipients globally. This review addresses the critical need for disaster preparedness to mitigate the severe impacts on this vulnerable patient cohort. Recent findings Kidney transplant recipients are highly dependent on robust healthcare infrastructures for ongoing care, including specialized medical staff, advanced diagnostics, and a consistent supply of immunosuppressive medications. Disasters disrupt these essential services, leading to increased risks of organ rejection, infections, and other medical complications. Strategies at various levels, from government to individual patients, can help maintain care continuity during such crises. Summary Effective disaster preparedness plans involving strategic medication stockpiling, emergency communication systems, and patient education are crucial to support kidney transplant recipients. By implementing these measures, healthcare systems can better protect the health and well being of transplant patients during and after disasters.