Background and objective: Prevention and control of metabolic syndrome is the key to improving the development of systemic lupus erythematosus. The aim of this study was to analyze the relevant factors regarding metabolic syndrome (MS) in systemic lupus erythematosus (SLE).
Methods: A total number of 1238 SLE patients in Yijishan Hospital of Wannan Medical College, Anhui province, from February 2012 to July 2017, were analyzed retrospectively. SLE patients with MS were grouped to group SLE-MS, the others without MS was grouped to group SLE-nMS. The two groups were compared with respect to general characteristics, clinical signs, and laboratory parameters. Random forest approach and multivariate logistic regression were conducted to analyze the related factors regarding MS in SLE.
Results: The constituent ratio of metabolic syndrome was 27.14% (336/1238). More SLE patients with MS presented with more farmers, more married people, lower education level, and more lupus nephritis, proteinuria, oral ulcers, tubular urine, hematuria than SLE patients without MS (P<0.05). Moreover, eighteen important variables, whose average importance scores were highest and whose error rates were lowest, were selected by random forest method. Data from multivariate logistic regression showed that MS in SLE was related with BMI, diastolic blood pressure, systolic blood pressure, fasting blood glucose, arthritis, urea, triglycerides, high-density lipoprotein, and white blood cells.
Conclusion: MS in SLE was closely related to BMI, blood pressure, blood sugar, blood lipids, arthritis, white blood cells, and urea. Targeted prevention and conclusion measures for the risk factors should be taken as early as possible.
Keywords: Metabolic syndrome; Random forest algorithm; Relevant factors; Systemic lupus erythematosus.
Copyright: © Pakistan Journal of Medical Sciences.