Preliminary nomogram model for predicting irreversible organ damage of patients with systemic sclerosis

Abstract

Objective: To investigate predictive factors for irreversible organ damage in systemic sclerosis (SSc) and establish a nomogram model.

Methods: This retrospective study included patients with SSc who were treated at our hospital between March 2013 and March 2023. Irreversible organ damage included heart failure, respiratory failure, renal failure, and gangrene of the hands and feet. Cox and LASSO regression analyses were performed to determine the predictive factors. Based on the results, a nomogram model was developed. The model was evaluated using the C-indices, calibration plots and DCA.

Results: A total of 361 patients with systemic sclerosis were randomly divided into the development (n = 181) and validation (n = 180) groups. Multivariate Cox regression analysis showed that age ≥65 years, weight loss, digital ulcers, mRSS ≥16, elevated creatinine, elevated myoglobin, elevated C-reactive protein, renal involvement and cardiac involvement were independent risk factors. Based on the LASSO analysis, a nomogram model of irreversible organ damage was established. The C-indices of the development group at 24, 60 and 96 m were 96.7, 84.5 and 85.7, whereas those of the validation group at 24, 60 and 96 m were 86.6, 79.1 and 78.5, respectively. The results of the DCA showed that the nomogram can be used as a valuable tool to predict irreversible organ damage in patients with SSc.

Conclusion: We included commonly used clinical indicators. According to the nomogram, the probability of irreversible organ damage can be calculated and high-risk patients can be identified.

Keywords: biomarker; cardiovascular; gastrointestinal; hormone; observational study; renal; skin.

Figure 1.

Kaplan–Meier Survival Curve Analysis of irreversible organ damage in patients with SSc. (A) is the comparison of different age groups, (B) is the comparison of the group with or without weight loss, (C) is the comparison of the group with or without digital ulcers, (D) is the comparison of different mRSS groups, (E) is the comparison of different CRP groups, (F) is the comparison of the group with or without elevated myoglobin, (G) is the comparison of the group with or without cardiac involvement and (H) is the comparison of the group with or without renal involvement. The log-rank test was used to determine significant differences between Kaplan–Meier curves

Figure 2.

Screening of variables and nomogram for irreversible organ damage of patients with systemic sclerosis. (A) The selection process of the optimum value of the parameter λ in the Lasso regression model by cross-validation method. (B) The variation characteristics of the coefficient of variables. (C) Nomogram for irreversible organ damage of patients with systemic sclerosis

Figure 3.

Calibration plots and calibration curve in the development and validation group of the nomogram. Calibration in the development group (A) and validation group (B); calibration curve in the development group (C) and validation group (D)

Figure 4.

ROC and DCA of the nomogram ROC in the development group (A) and validation group (B); DCA in the development group (C) and validation group (D). AUC: area under roc curve; DCA: decision curve analysis; ROC: receiver operating characteristic curve