Does AI Hold the Key to Better Gout Management?

TOPLINE:

The application of an artificial intelligence (AI)–based gout management system in patients with gout and normal or near-normal kidney function is associated with a lower risk of developing chronic kidney disease (CKD) stage ≥ 3 and a greater likelihood of reaching target serum urate levels than that of a traditional electronic medical record system.

METHODOLOGY:

• Researchers conducted an observational cohort study to compare the effect of an AI-based gout management system with that of a traditional electronic medical record system on CKD outcomes and the effectiveness of urate-lowering therapy (ULT) in patients with gout and an estimated glomerular filtration rate (GFR) > 60 mL/min/1.73 m².
• The AI-based system cohort included 1780 patients (mean age, 43.9 years; 97.2% men) who were matched and compared with 890 patients in the traditional system cohort (mean age, 44.4 years; 97.3% men).
• The AI-based system incorporated data from a mobile application comprising a data fusion interface and modules for expert consultation and laboratory results management.
• The primary outcome was either the development of GFR 2 on at least two occasions more than 90 days apart within 1 year or hemodialysis or peritoneal dialysis or kidney transplant; the secondary outcomes included changes in serum urate concentrations, use of medications, adherence, and the incidence of gout flares in the AI-based system group.
• Patients were monitored from the date of diagnosis of gout for up to 4 years; the median follow-up duration was 2.2 years in the traditional system–based cohort and 2.3 years in the AI-based one.

TAKEAWAY:

• During the follow-up period, participants in the AI-based system group showed a lower cumulative incidence of CKD stage ≥ 3 than those in the traditional group (hazard ratio, 0.57; P
• Across all 4 years of follow-up, a higher proportion of participants in the AI-based group achieved target serum urate concentrations P
• In the AI-based group, high adherence to ULT increased from 55.1% in year 1 to 65.2% in year 4, and the incidence of at least one gout flare decreased from 70.3% to 40.8% over the study period.
• Furthermore, in the AI-based group, a consistent reduction was observed in the proportion of patients requiring treatment with nonsteroidal anti-inflammatory drugs, colchicine, and glucocorticoids (Ptrend 

IN PRACTICE:

“Here [In this study], advantages of AI-assisted management likely were at play, by factors such as delivery of real-time health and metabolic information, promotion of patient engagement in monitoring treatment response, and making continuity of care easier by reducing travel to in-person appointments,” the authors wrote.

SOURCE:

The study was led by Han Qi, the Affiliated Hospital of Qingdao University, Qingdao, China. It was published online on November 6, 2024, in Rheumatology.

LIMITATIONS:

Study limitations included the potential for residual bias due to some unmeasured baseline characteristics. The study also lacked relevant information on how comorbidities were managed in patients with gout. The possibility of surveillance bias cannot be ruled out because patients in the AI-based gout management system had more frequent clinic visits and assessments of renal function decline.

DISCLOSURES:

The study was supported by the National Key R&D Program of China, National Natural Science Foundation of China, Taishan Scholar Program of Shandong Province, and Shandong Provincial Natural Science Foundation. One author reported serving as a consultant for various sources. Another author reported receiving consulting fees, research grants, and speaker fees from various pharmaceutical companies outside the submitted work.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.