Application of ChatGPT to Support Nutritional Recommendations for Dialysis Patients - A Qualitative and Quantitative Evaluation.

OBJECTIVES: The diversity of food preferences and the need for personalized care can be challenging for renal dietitians. We explored ChatGPT-4 to support nutritional advice given to dialysis patients. DESIGN: In this simulation study, we tasked ChatGPT-4 with generating a personalized daily meal plan, including nutritional information. SETTING: The study used virtual "patients" generated through Monte Carlo simulation. SUBJECTS: Data from a randomly selected virtual patient were presented to ChatGPT. INTERVENTION: We provided to ChatGPT patient demographics, food preferences, laboratory data, clinical characteristics, and available budget, to generate a one-day sample menu with recipes and nutritional analyses. The generated content was rated by a renal dietitian and compared with a USDA-approved nutrient analysis software. ChatGPT also analyzed nutrition information of two recipes published online. We also requested a translation of the output into Spanish, Mandarin, Hungarian, German, and Dutch. MAIN OUTCOME MEASURE: The outcome measures were the accuracy of ChatGPT's nutritional analysis, the quality of recipe and cooking instructions on a five-point Likert scale, and a quantitative nutritional analysis. RESULTS: ChatGPT generated a daily menu with five recipes. The renal dietitian rated the recipes at 3 (3,3) [median (Q1, Q3)], the cooking instructions at 5 (5,5), and the nutritional analysis at 2 (2,2) on the five-point Likert scale. ChatGPT's nutritional analysis underestimated calories by 36% (95% CI: 44-88%), protein by 28% (25-167%), fat 48% (29-81%), phosphorus 54% (15-102%), potassium 49% (40-68%), and sodium 53% (14-139%). The nutritional analysis of online available recipes differed only by 0 to 35%. The translations were rated as reliable by native speakers (4 on the five-point Likert scale). CONCLUSION: While ChatGPT-4 shows promise in providing personalized nutritional guidance for diverse dialysis patients, improvements are necessary. This study highlights the importance of thorough qualitative and quantitative evaluation of AI-generated content, especially regarding medical use cases.

Independent and joint associations of nutritional status indicators with mortality risk among chronic hemodialysis patients in the Dialysis Outcomes and Practice Patterns Study (DOPPS).

OBJECTIVE: To consider the Kidney Disease Outcomes Quality Initiative recommendation of using multiple nutritional measurements for patients on maintenance dialysis, we explored data for independent and joint associations of nutritional indicators with mortality risk among maintenance hemodialysis patients treated in 12 countries. SETTING: Dialysis units in seven European countries, the United States, Canada, Australia, New Zealand, and Japan. MAIN OUTCOME: Mortality risk. METHODS: We conducted a prospective cohort study of 40,950 patients from phases I to III of the Dialysis Outcomes and Practice Patterns Study (1996-2008). Independent and joint effects (interactions) of nutritional indicators (serum creatinine, serum albumin, normalized protein catabolic rate, body mass index [BMI]) on mortality risk were assessed by Cox regression with adjustments for demographics, years on dialysis, and comorbidities. RESULTS: Important variations in nutritional indicators were seen by country and patient characteristics. Poorer nutritional status assessed by each indicator was independently associated with higher mortality risk across regions. Significant multiplicative interactions (each p < or = 0.01) between indicators were also observed. For example, by using patients with serum creatinine 7.5-10.5 mg/dL and BMI 21-25 kg/m(2) as referent, BMI <21 kg/m(2) was associated with lower mortality risk among patients with creatinine >10.5 mg/dL (relative risk = 0.68) but with higher mortality risk among those with creatinine <7.5 mg/dL (relative risk = 1.38). The association of lower albumin concentration with higher mortality risk was stronger for patients with lower BMI or lower creatinine. CONCLUSION: The joint effects of nutritional indicators on mortality indicate the need to use multiple measurements when assessing the nutritional status of hemodialysis patients.

Peripheral arterial disease in patients with end-stage renal disease: observations from the Dialysis Outcomes and Practice Patterns Study (DOPPS).

BACKGROUND: Patients with end-stage renal disease are at high risk for cardiovascular morbidity and mortality. The aims of the present study were to describe the prevalence of peripheral arterial disease (PAD) and its effects on prognosis and health-related quality of life (HRQOL) in an international cohort of patients on hemodialysis. METHODS AND RESULTS: Data from the Dialysis Outcomes and Practice Patterns Study (DOPPS), a prospective, international, observational study of hemodialysis patients (n=29,873), were analyzed. Associations between baseline clinical variables and PAD were evaluated by logistic regression analysis. Cox regression models were used to test the association between PAD and risk for all-cause mortality, cardiac mortality, and hospitalization. PAD was diagnosed in 7411 patients (25.3%) with significant geographic variation. Traditional cardiovascular risk factors including age, male sex, diabetes, hypertension, and smoking were identified, together with the duration of hemodialysis, as significant correlates of PAD. Diagnosis of PAD was associated with increased all-cause mortality (hazard ratio [HR]=1.36; P<0.0001), cardiac mortality (HR=1.43; P<0.0001), all-cause hospitalization (HR=1.19; P<0.0001), and hospitalization for a major adverse cardiovascular event (HR=2.05; P<0.0001). HRQOL questionnaires revealed physical health scores that were significantly lower in PAD compared with non-PAD patients (P<0.0001). CONCLUSIONS: PAD is common in hemodialysis patients and is associated with increased risk of cardiovascular mortality, morbidity, and hospitalization and reduced HRQOL.

Intradialytic Hypoxemia and Clinical Outcomes in Patients on Hemodialysis.

BACKGROUND AND OBJECTIVES: Intradialytic hypoxemia has been recognized for decades, but its associations with outcomes have not yet been assessed in a large patient cohort. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: Our retrospective cohort study was conducted between January of 2012 and January of 2015. We recorded blood oxygen saturation every minute during hemodialysis in patients with arteriovenous access. A 6-month baseline period with at least 10 treatments with oxygen saturation measurements preceded a 12-month follow-up. Patients were stratified by the presence or absence of prolonged intradialytic hypoxemia defined as oxygen saturation <90% for at least one third of the treatment time. Demographic, laboratory, and treatment data and hospitalization and mortality rates were compared between the groups. Multivariate Cox regression analysis was used to assess baseline predictors of all-cause mortality during follow-up. RESULTS: In total, 100 (10%) of 983 patients had prolonged intradialytic hypoxemia. These patients were older (+3.6 years; 95% confidence interval, 0.8 to 6.3), had longer dialysis vintage (+1.2 years; 95% confidence interval, 0.3 to 2.1), and had higher prevalence of congestive heart failure (+10.8%; 95% confidence interval, 1.6 to 20.7) and chronic obstructive pulmonary disease (+13%; 95% confidence interval, 5 to 21.2). They also resembled an inflammatory phenotype, with lower serum albumin levels (-0.1 g/dl; 95% confidence interval, -0.2 to 0) and higher neutrophil-to-lymphocyte ratios (+1; 95% confidence interval, 0.5 to 1.6). They had lower hemoglobin levels (-0.2 g/dl; 95% confidence interval, -0.4 to 0) and required more erythropoietin (+1374 U per hemodialysis treatment; 95% confidence interval, 343 to 2405). During follow-up, all-cause hospitalization (1113 hospitalizations; univariate hazard ratio, 1.46; 95% confidence interval, 1.22 to 1.73) and mortality (89 deaths; adjusted hazard ratio, 1.98; 95% confidence interval, 1.14 to 3.43) were higher in patients with prolonged intradialytic hypoxemia. CONCLUSIONS: Prolonged intradialytic hypoxemia was associated with laboratory indicators of inflammation, higher erythropoietin requirements, and higher all-cause hospitalization and mortality.

Kidney Disease Outcomes Quality Initiative (K/DOQI) and the Dialysis Outcomes and Practice Patterns Study (DOPPS): nutrition guidelines, indicators, and practices.

BACKGROUND: Nutritional markers are important predictors of morbidity and mortality in dialysis patients. The Clinical Practice Guidelines for Nutrition in Chronic Renal Failure provides guidelines for assessing nutritional status that were evaluated using data from the Dialysis Outcomes and Practice Patterns Study (DOPPS). METHODS: The level of various nutritional markers (serum albumin, modified subjective global assessment, serum creatinine, normalized protein catabolic rate [nPCR], and body mass index) were described for representative samples of patients and facilities from 7 countries (France, Germany, Italy, Spain, Japan, United Kingdom, and United States) participating in the DOPPS. RESULTS: A strong inverse association was observed between mortality and serum albumin, with a mortality risk 1.38 times higher for patients with serum albumin concentration less than 3.5 g/dL (35 g/L). There were significant differences by country in the proportion of moderately and severely malnourished patients as determined by the modified subjective global assessment score. In the US sample, severely and moderately malnourished patients had a higher mortality risk compared with those not malnourished, 33% and 5% higher, respectively. An inverse relationship exists between serum creatinine concentration and mortality, with a mortality risk 60% to 70% higher in the lowest quartile group compared with the highest quartile group in Europe and the United States. Levels of nPCR varied significantly between European countries, and there was no association between mortality and nPCR in US data. After adjustment for demographic and comorbidity factors, the mortality risk decreased as body mass index increased in both US and European samples. CONCLUSION: DOPPS data highlight the importance of routine assessment of nutritional status, using multiple parameters, in clinical practice to improve patient care.

Lack of appetite in haemodialysis patients--associations with patient characteristics, indicators of nutritional status and outcomes in the international DOPPS.

BACKGROUND: Identification of haemodialysis patients with problems related to lack of appetite should help prevent adverse outcomes. We studied whether a single question about being bothered by lack of appetite within the prior 4 weeks is related to nutritional status, inflammation and risks of death and hospitalization. Additionally, we assessed associations of lack of appetite with depression, dialysis dose and length of haemodialysis. METHODS: This study is an analysis of baseline and longitudinal data from 14 406 patients enrolled in the Dialysis Outcomes and Practice Pattern Study. Cox regression was used to assess whether the degree (not, somewhat, moderately, very much, extremely) that patients were bothered by lack of appetite is an independent predictor of death and hospitalization. Logistic regression was used to identify baseline characteristics associated with being bothered by lack of appetite. RESULTS: The risk of death was more than 2-fold higher [relative risk (RR) = 2.23; 95% confidence interval (CI) = 1.90-2.62] and the risk of hospitalization 33% higher (RR = 1.33; 95% CI = 1.19-1.48) among patients extremely bothered, compared with not bothered, by lack of appetite. These associations followed a dose-response fashion and remained statistically significant after adjustments for 14 comorbidities. Depression, shorter haemodialysis session, hypoalbuminaemia, lower concentration of serum creatinine and normalized protein catabolic rate, lower body mass index and higher leucocyte and neutrophil counts were independently associated with higher odds of being bothered by lack of appetite. CONCLUSIONS: The data suggest that a single question about lack of appetite helps identify haemodialysis patients with poorer nutritional status, inflammation, depression and higher risks of hospitalization and death. The study calls attention to a possible beneficial effect of longer haemodialysis on appetite.