Precision Nutrition and Personalized Diet Plan for Kidney Health and Kidney Disease Management

Workforce needs and estimated costs/savings for nutrition care in chronic kidney disease-stage 3 through maintenance dialysis

The role of nutrition in chronic kidney disease (CKD) is well known. However, controversies, misconceptions, and gaps in the literature exist regarding the workforce required to provide nutrition care in CKD. This paper reviews the existing literature on this topic, focusing primarily on the United States. Topics covered in this review include the qualifications and services of the registered dietitian nutritionist (RDN), to which specific patients nutrition care should be provided in CKD and on dialysis, barriers to the utilization of nutrition services, the quantity of nutrition care needed to improve patient outcomes, and the risks of provider burnout. Controversies include whether more staffing is associated with better care and the best measures of staffing. The topics are supported with estimations for the economic impact and practicality of workforce decisions.

Evaluating the Effectiveness of a Generative Pretrained Transformer-Based Dietary Recommendation System in Managing Potassium Intake for Hemodialysis Patients

OBJECTIVE

Despite adequate dialysis, the prevalence of hyperkalemia in Chinese hemodialysis (HD) patients remains elevated. This study aims to evaluate the effectiveness of a dietary recommendation system driven by generative pretrained transformers (GPTs) in managing potassium levels in HD patients.

METHODS

We implemented a bespoke dietary guidance tool utilizing GPT technology. Patients undergoing HD at our center were enrolled in the study from October 2023 to November 2023. The intervention comprised of two distinct phases. Initially, patients were provided with conventional dietary education focused on potassium management in HD. Subsequently, in the second phase, they were introduced to a novel GPT-based dietary guidance tool. This artificial intelligence (AI)-powered tool offered real-time insights into the potassium content of various foods and personalized dietary suggestions. The effectiveness of the AI tool was evaluated by assessing the precision of its dietary recommendations. Additionally, we compared predialysis serum potassium levels and the proportion of patients with hyperkalemia among patients before and after the implementation of the GPT-based dietary guidance system.

RESULTS

In our analysis of 324 food photographs uploaded by 88 HD patients, the GPTs system evaluated potassium content with an overall accuracy of 65%. Notably, the accuracy was higher for high-potassium foods at 85%, while it stood at 48% for low-potassium foods. Furthermore, the study examined the effect of GPT-based dietary advice on patients' serum potassium levels, revealing a significant reduction in those adhering to GPTs recommendations compared to recipients of traditional dietary guidance (4.57 ± 0.76 mmol/L vs. 4.84 ± 0.94 mmol/L, P = .004). Importantly, compared to traditional dietary education, dietary education based on the GPTs tool reduced the proportion of hyperkalemia in HD patients from 39.8% to 25% (P = .036).

CONCLUSION

These results underscore the promising role of AI in improving dietary management for HD patients. Nonetheless, the study also points out the need for enhanced accuracy in identifying low potassium foods. It paves the way for future research, suggesting the incorporation of extensive nutritional databases and the assessment of long-term outcomes. This could potentially lead to more refined and effective dietary management strategies in HD care.

Adaptive Nutrition Intervention Stabilizes Serum Phosphorus Levels in Hemodialysis Patients: A Multicenter Decentralized Clinical Trial Using Real-World Data

OBJECTIVE

This study aims to evaluate the effect of an adaptive nutritional and educational intervention for patients on hemodialysis (HD) in a routine care setting, using real-world data from electronic health records.

METHODS

Decentralized clinical trial of seven HD facilities recruited patients who have been on HD for over 3 months (N = 153) for an 8-week adaptive intervention protocol. Patients were divided into four groups: (1) control (2) education intervention (3) meal intervention (4) education and meal interventions. Educational contents were digitally delivered via mobile phones and premade meals tailored on laboratory findings were home-delivered. Changes in serum electrolytes and malnutrition inflammation score (MIS) were analyzed.

RESULTS

Meal intervention statistically significantly stabilized serum phosphorus level (β = -0.81 mg/dL, 95% confidence interval = [-1.40, -0.22]) at week 8, with increased likelihood of being within target serum value range (odds ratio = 1.21, 95% confidence interval = [1.04, 1.40]). Meal group showed better nutritional status (MIS = 3.65) than the education group (MIS = 5.10) at week 8 (adjusted p < .05). No significant changes were observed in serum potassium level, depression, and self-efficacy.

CONCLUSION

It was demonstrated that an adaptive meal intervention in a real-world care setting may benefit serum phosphorus control and nutritional status of patients on HD, without negative effect on depression levels or self-efficacy. More work is needed to develop an effective educational intervention.

AI-Powered Renal Diet Support: Performance of ChatGPT, Bard AI, and Bing Chat

Patients with chronic kidney disease (CKD) necessitate specialized renal diets to prevent complications such as hyperkalemia and hyperphosphatemia. A comprehensive assessment of food components is pivotal, yet burdensome for healthcare providers. With evolving artificial intelligence (AI) technology, models such as ChatGPT, Bard AI, and Bing Chat can be instrumental in educating patients and assisting professionals. To gauge the efficacy of different AI models in discerning potassium and phosphorus content in foods, four AI models-ChatGPT 3.5, ChatGPT 4, Bard AI, and Bing Chat-were evaluated. A total of 240 food items, curated from the Mayo Clinic Renal Diet Handbook for CKD patients, were input into each model. These items were characterized by their potassium (149 items) and phosphorus (91 items) content. Each model was tasked to categorize the items into high or low potassium and high phosphorus content. The results were juxtaposed with the Mayo Clinic Renal Diet Handbook's recommendations. The concordance between repeated sessions was also evaluated to assess model consistency. Among the models tested, ChatGPT 4 displayed superior performance in identifying potassium content, correctly classifying 81% of the foods. It accurately discerned 60% of low potassium and 99% of high potassium foods. In comparison, ChatGPT 3.5 exhibited a 66% accuracy rate. Bard AI and Bing Chat models had an accuracy rate of 79% and 81%, respectively. Regarding phosphorus content, Bard AI stood out with a flawless 100% accuracy rate. ChatGPT 3.5 and Bing Chat recognized 85% and 89% of the high phosphorus foods correctly, while ChatGPT 4 registered a 77% accuracy rate. Emerging AI models manifest a diverse range of accuracy in discerning potassium and phosphorus content in foods suitable for CKD patients. ChatGPT 4, in particular, showed a marked improvement over its predecessor, especially in detecting potassium content. The Bard AI model exhibited exceptional precision for phosphorus identification. This study underscores the potential of AI models as efficient tools in renal dietary planning, though refinements are warranted for optimal utility.

Effectiveness of a Nutritional Mobile Application for Management of Hyperphosphatemia in Patients on Hemodialysis: A Multicenter Open-Label Randomized Clinical Trial

This study aims to determine the effectiveness of a phosphate mobile app (PMA), MyKidneyDiet-Phosphate Tracker ©2019, on hemodialysis (HD) patients with hyperphosphatemia. A multicenter, open-label, randomized controlled trial design allowed randomization of patients with hyperphosphatemia to either the usual care group (UG; receiving a single dietitian-led session with an education booklet) or the PMA group (PG). Thirty-three patients in each intervention group completed the 12-week study. Post-intervention, serum phosphorus levels were reduced in both groups (PG: −0.25 ± 0.42 mmol/L, p = 0.001; UG: −0.23 ± 0.33 mmol/L, p < 0.001) without any treatment difference (p > 0.05). Patients in both groups increased their phosphate knowledge (PG: 2.18 ± 3.40, p = 0.001; UG: 2.50 ± 4.50, p = 0.003), without any treatment difference (p > 0.05). Dietary phosphorus intake of both groups was reduced (PG: −188.1 ± 161.3 mg/d, p < 0.001; UG: −266.0 ± 193.3 mg/d, p < 0.001), without any treatment difference (p > 0.05). The serum calcium levels of patients in the UG group increased significantly (0.09 ± 0.20 mmol/L, p = 0.013) but not for the PG group (−0.03 ± 0.13 mmol/L, p = 0.386), and the treatment difference was significant (p = 0.007). As per phosphate binder adherence, both groups reported a significant increase in Morisky Medication Adherence Scale scores (PG: 1.1 ± 1.2, p < 0.001; UGa: 0.8 ± 1.5, p = 0.007), without any treatment difference (p > 0.05). HD patients with hyperphosphatemia using the PMA achieved reductions in serum phosphorus levels and dietary phosphorus intakes along with improved phosphate knowledge and phosphate binder adherence that were not significantly different from a one-off dietitian intervention. However, binder dose adjustment with meal phosphate content facilitated by the PMA allowed stability of corrected calcium levels, which was not attained by UC patients whose binder dose was fixed.