The synergy of serum SFRP5 levels and the TyG index in predicting coronary artery disease and prognosing major adverse cardiovascular events

Baseline clinical characteristics

Patients were categorized into three groups based on their CAG results and SYNTAX scores: a control group (n = 50), a mild lesion group (< 23 points, n = 175) and a moderate-severe lesion group (≥ 23 points, n = 85). Table 1 outlines the baseline clinical characteristics observed within the study population. No significant differences were observed among the groups (P > 0.001) regarding age, sex, BMI, smoking history, drinking history, hypertension, n-HDL-c, VLDL-C, Scr, and eGFR assay results. Lp(a) and UA levels significantly differed only between the control group and the moderate-severe lesion group. D-D levels significantly differed only between the mild group and the moderate-severe lesion group. LDH and CK-MB values differed significantly between the control group and the mild lesion group and between the mild lesion group and the moderate-severe lesion group. The levels of ApoA1 and ApoB were significantly different between the control group and the mild and moderate-severe lesion groups, whereas no significant difference was observed between the mild group and the moderate-severe group. There was a substantial increase in the concentrations of TC, TG, LDL-C, SFRP5, HbA1c and TyG, and this increase was statistically significant across all groups. In contrast, both HDL-C and EF levels were significantly decreased.

Table 1 Baseline clinical characteristicsSFRP5 and TyG index

Table 2 presents the correlation analysis among serum SFRP5 levels, the TyG index, and various variables within the patient cohort. Notably, serum SFRP5 levels were positively correlated with HDL-C. However, they were negatively correlated with TG, LDL-C, and HbA1c levels and SYNTAX scores, all of which were statistically significant (P < 0.001), and negatively correlated with FBG (r=-0.160, P = 0.005). Conversely, the TyG index was positively correlated with LDL-C, TG, FBG, and HbA1c levels and SYNTAX scores and was negatively correlated with HDL-C levels (P < 0.001). Importantly, the analysis revealed a significant inverse relationship between serum SFRP5 levels and the TyG index (r= -0.312, P < 0.001) (Fig. 1).

Table 2 The correlation between serum SFRP5 levels and the TyG indexFig. 1

Scatter diagram showing the correlation between serum SFRP5 levels and the TyG index

Multiple logistic regression analysis of risk factors of CAD

Figure 2 displays the results of multiple logistic regression analysis, examining the association between serum SFRP5 levels and various relevant variables in patients who enrolled in this research. The forest plot presents the results of the multivariate logistic regression model that examined the relationship between serum SFRP5 levels and various other factors. In the analysis, patients with CAD were considered the dependent variable, while SFRP5, TyG index, LDL-C, and HDL-C were considered independent variables. The logistic multivariate regression analysis revealed that the TyG index and LDL-C were independent risk factors for CAD, while SFRP5 and HDL-C were protective factors for CAD after adjusting for other confounding factors (P < 0.05). These findings suggest that an elevated TyG index, along with increased LDL-C levels, is associated with an increased risk of CAD. Conversely, higher levels of SFRP5 and HDL-C are associated with a reduced risk of CAD.

Fig. 2

Forest plot of the results of the multivariate logistic regression model exploring the association between plasma SFRP5 levels and other related factors in patients with CHD.

Diagnostic value of serum SFRP5 levels and the TyG index

The diagnostic value of serum SFRP5 levels and the TyG index was illustrated in the ROC curve for the occurrence of CAD. The optimal cutoff value for the serum SFRP5 concentration was determined to be 115.58 pg/mL. At this threshold, the sensitivity, specificity, and AUC for predicting coronary heart disease were 84.6%, 77.6% and 0.87 [95% CI (0.82,0.92), P < 0.001], respectively. Similarly, the optimal cutoff value for the TyG index was determined to be 8.49, with corresponding sensitivity, specificity, and AUC values of 72.3%, 69.4%, and 0.74 [95% CI (0.65,0.83), P < 0.001], respectively. Notably, when both serum SFRP5 levels and the TyG index were considered together, the combination exhibited improved predictive performance, with corresponding sensitivity, specificity, and AUC values of 90.8%, 79.4%, and 0.91 for predicting the occurrence of CAD [95% CI (0.87, 0.95), P < 0.001] (Fig. 3 & Table 3).

Fig. 3

Diagnostic value of plasma SFRP5 levels and the TyG index in ROC curve

Table 3 Diagnostic value of serum SFRP5 levels and the TyG index in ROC curveThe SFRP5 levels, TyG index, and MACEs

Table 4 provides the results of the univariate Cox regression analysis, demonstrating the independent roles of variables as predictors for clinical endpoint events. In this analysis, statistically significant independent associations were observed for surgery time, sex, LVD, EF, SFRP5, TyG index, TG, TC, HDL-C, LDL-C, ApoB, Lp(a), FDP, FIB, D-D, LDH, CK-MB, UA, r, BNP, ALT, and AST (P < 0.05). Using the stepwise forward method, the multivariate Cox regression analysis revealed significant associations (Table 5) for SFRP5 (P = 0.01), TyG index (P = 0.001), and LDL-C (P = 0.043). The Kaplan‒Meier analysis, assessing event-free survival, demonstrated a clear trend. As the TyG index increased, the incidence of MACE increased and SFRP5 levels decreased (P < 0.001) (Fig. 4 & Fig. 5). Patients with a concentration of SFRP5 > 115.58 pg/mL and a TyG index < 8.49 exhibit a better prognosis for avoiding MACEs (P < 0.001) (Fig. 6).

Table 4 Univariate Cox regression analysis of MACETable 5 Multivariate Cox regression analysis of MACEFig. 4