In this study, the prevalence rate of hyperuricemia in patients with IgAN is 26.27%, which is basically consistent with previous literature reports [3]. Some studies have reported that the degree of clinical indexes and renal pathological damage in patients with hyperuricemia was more serious than that in patients with normouricemia[4–7, 9, 18–19]. Lu et al [18] has confirmed that IgAN with hyperuricemia has a poor prognosis.
Therefore, more attention should be paid to the related risk factors of hyperuricemia, and timely and effective interventions are expected to delay the progress of IgAN and improve the prognosis of IgAN. It is necessary to identify the indicators that can cause hyperuricemia in IgAN patients, so as to carry out early intervention, delay the progress and prevent the occurrence of ESRD. There is an urgent need to combine these predictors to establish a comprehensive risk prediction model of hyperuricemia.
Cai et al. [19] used the data of 109 patients to analyze the related factors affecting the occurrence of hyperuricemia in patients with IgAN. The conclusion showed that male patients and hypertriglyceridemia were related to the occurrence of hyperuricemia, but there was no predictive nomogram and the sample size was small. A retrospective study by Qiu et al. [3] which including 648 patients with IgAN demonstrated that four variables (serum creatinine, hyperglycinemia, hypertension, obesity) were associated with hyperuricemia. This study also failed to establish a complete prediction model. So far, with regard to the probability of hyperuricemia in IgAN, risk prediction model is not available.
In this study, we included 1184 IgAN patients and selected variables based on the results of multivariate logistic regression model. In addition, we developed and validated nomogram with ROC curve and calibration curve. The AUC of the development and validation cohorts were 0.8340 and 0.7873 respectively. All calibration curves showed good consistency between the predicted probability and the actual probability. The result of DCA showed that our nomogram had good discrimination ability.
Compared with previous studies, our study is a complete set of prediction model of IgAN with hyperuricemia, and has been verified, and the sample size of nomogram is larger than previous studies, which provides important clues for the prediction of IgAN patients with hyperuricemia, and is more suitable for clinical practice. In the current nomogram, we first included gender, albuminemia, hypertriglyceridemia, gross hematuria, 24 TP, eGFR, BUN and T according to the results of multivariate logistic regression model. In previous studies, these variables were also identified as independent risk factors [8–9, 13–14]. In addition, according to our analysis, hypertension was not identified as an independent risk factor, which may be attributed to the use of antihypertensive drugs, such as ACEI / ARB (losartan can reduce serum uric acid level. Irbesartan can significantly reduce serum uric acid level and oxidative stress level. Angiotensin receptor blockers can reduce the serum uric acid level). Similarly, Scr indicator was not included in the model too. In the previous literatures [3, 20], Scr index was included in the original form. In order to explain the relationship between Scr level and hyperuricemia more conveniently, we transformed the original numerical variable into a categorical variable. This may be the reason why Scr was not included in the model.
The current study has some potential limitations. First of all, the nomogram is developed and validated by retrospective analysis of clinicopathological data of a single institution, which may lead to some deviations in our conclusions. Therefore, before our nomogram is widely used, we need to use data from multiple centers for further evaluation through external verification. Second, we did not statistically analyze the diet habits, exercise style, body mass index and other indicators of the patients, and some other important parameters may be missed, so the future research needs to be further improved. Finally, the patients included in this study may have an inevitable bias in the results due to different treatment regimens. It expects that future research should be based on integrating data from multiple centers around the world to develop and externally validate ideal risk prediction models. The clinical and pathological predictors included in the model should be reliable and convenient for clinical application.