Nomogram Based on Risk Factors for Type 2 Diabetes Mellitus Patients with Coronary Heart Disease

Figures & data

Figure 1 The flow chart presents the entire process of patient follow-up, data collection and statistical analysis in this study.

Table 1 Characteristics of the Participants in Different Groups

	Total (n=3214)	Training Set (n=2252)	Validation Set (n=962)	P-value
Age (years old)	65.00 (60.00,69.00)	65.00 (60.00, 69.00)	65.00 (60.00,68.00)	0.355
Sex				0.975
Male	1605 (49.94%)	1125 (49.96%)	480 (49.90%)
Female	1609 (50.06%)	1127 (50.04%)	482 (50.10%)
Diagnosed CHD	529 (16.46%)	367 (16.30%)	162 (16.84%)	0.704
Diagnosed HTN	1937 (60.27%)	1348 (59.86%)	589 (61.23%)	0.468
Diagnosed HUA	540 (16.80%)	392 (17.41%)	148 (15.38%)	0.160
Diagnosed DR	665 (20.69%)	447 (19.85%)	218 (22.66%)	0.071
Diagnosed DN	1238 (38.52%)	863 (38.32%)	375 (38.98%)	0.725
Smoking	955 (29.71%)	668 (29.66%)	287 (29.83%)	0.995
Alcohol	884 (27.50%)	627 (27.84%)	257 (26.72%	0.616
T2DM duration (years)	8.00 (4.00,13.00)	8.00 (4.00,13.00)	8.00 (4.00, 14.00)	0.955
BMI (kg/m^2)	25.50 (23.40,27.70)	25.40 (23.40,27.60)	25.60 (23.40, 27.70)	0.531
SBP (mmHg)	140.00 (127.00, 154.00)	140.00 (127.00, 153.00)	140.00 (127.00, 154.00)	0.636
DBP (mmHg)	80.00 (73.00,87.00)	80.00 (73.00,86.00)	80.00 (73.00, 87.00)	0.910
FBG (mmol/L)	7.20 (6.00,8.90)	7.20 (6.00,8.90)	7.20 (6.10, 8.90)	0.896
PBG (mmol/L)	11.80 (8.50,15.30)	11.70 (8.50,15.30)	11.90 (8.70,15.20)	0.892
HbA1c (%)	6.94 (6.24,8.00)	6.94 (6.24,8.00)	7.04 (6.34, 7.94)	0.694
TC (mmol/L)	4.58 (3.87,5.32)	4.58 (3.89, 5.31)	4.58 (3.86, 5.33)	0.652
TG (mmol/L)	1.64 (1.20,2.30)	1.65 (1.21,2.31)	1.62 (1.19, 2.27)	0.431
LDL-C (mmol/L)	1.50 (1.20,1.81)	1.51 (1.21,1.81)	1.48 (1.16, 1.81)	0.363
HDL-C (mmol/L)	1.56 (1.34,1.81)	1.56 (1.34, 1.81)	1.56 (1.33, 1.81)	0.892
BUN (mmol/L)	5.46 (4.48,6.50)	5.46 (4.50,6.50)	5.42 (4.45, 6.49)	0.450
SCR (μmol/L)	66.00 (55.00,77.00)	66.00 (55.00,77.00)	66.00 (56.00,77.00)	0.795
UA (umol/L)	313.00 (263.00, 365.00)	315.00 (267.00, 366.00)	306.00 (255.00, 360.00)	0.019
UCR (umol/L)	9.52 (6.40,12.00)	9.48 (6.34,12.00)	9.57 (6.75, 12.00)	0.520
UMA (mg/L)	22.00 (10.00,57.00)	22.00 (10.00,57.30)	22.00 (10.00, 56.00)	0.539
eGFR (mL/min)	64.60 (43.50, 87.70)	64.50 (42.50, 87.80)	64.60 (45.20, 87.50)	0.667
ACR (kg/m^2)	20.70 (10.20,54.30)	20.80 (10.30,55.00)	20.10 (9.98, 53.20)	0.953

Notes: Data are presented as n (%), median (IQR).

Table 2 Coefficients and Lambda.min Value of the LASSO Regression Based on the Training Set

Factors	Coefficients	Lambda.min
Age (years old)	0.036	0.021
T2DM duration (years)	0.012
HTN (%)	0.304
HUA (%)	0.067
BMI (kg/m2)	0.046
HbA1c (%)	0.050
HDL-C (mmol/L)	−0.551
LDL-C (mmol/L)	0.681

Table 3 Model Established by Logistic Regression Analysis Based on the Training Set

Factors	β-Coefficient	Wald-Test	P-value	OR (95% CI)
Age (years old)	0.060	5.916	<0.001	1.062 (1.041, 1.084)
T2DM duration (years)	0.034	3.554	<0.001	1.035 (1.015, 1.054)
HTN (%)	0.605	4.289	<0.001	1.831 (1.393, 2.423)
HUA (%)	0.410	2.740	<0.05	1.507 (1.120, 2.016)
BMI (kg/m2)	0.074	4.072	<0.001	1.077 (1.039, 1.116)
HbA1c (%)	0.145	3.324	<0.001	1.156 (1.061, 1.259)
HDL-C (mmol/L)	−1.208	−6.229	<0.001	0.299 (0.203, 0.435)
LDL-C (mmol/L)	1.182	8.575	<0.001	1.019 (1.014–1.024)

Figure 2 Demographic and clinical feature selection using the LASSO binary logistic regression model in T2DM patients with CHD based on the training set.

Notes: (A) Optimal parameter (lambda) selection in the LASSO model used fivefold cross-validation based on minimum criteria. The partial likelihood deviance (binomial deviance) curve was plotted versus log(lambda). Dotted vertical lines were drawn at the optimal values by using the minimum criteria and the 1 SE of the minimum criteria (the 1-SE criteria). LASSO coefficient profiles of the 8 features. (B) A coefficient profile plot was produced against the log(lambda) sequence. A vertical line was drawn at the value selected using fivefold cross-validation, where optimal lambda resulted in 8 features with nonzero coefficients.

Abbreviations: LASSO, least absolute shrinkage and selection operator; SE, standard error.

Figure 3 Developed nomogram for CHD.

Notes: (A) The nomogram for CHD in T2DM patients was developed in the cohort by integrating age, T2DM duration, HTN, HUA, BMI, HbA1c, HDL-C and LDL-C. (B) An example of nomogram for CHD in T2DM patients. Logistic regression results showed that there were corresponding P values for each index, and the indicators with statistical significance level P ≤ 0.05 were included in the nomogram. “***” means P<0.001, “**” means P<0.05.

Table 4 C-Index in the Array Based on Training Set and Validation Set

Groups	C-Index(95% CI)	Dxy	aDxy	SD	Z	P	n
Training set	0.750(0.724–0.776)	0.499	0.499	0.027	18.63	0	2252
Validation set	0.767(0.726–0.808)	0.534	0.534	0.042	12.81	0	962

Figure 4 The pooled AUC of the ROC curve.

Notes: (A) Training set and (B) Validation set: The pooled AUC of the ROC curve. The y-axis indicates the true positive rate of the risk prediction. The x-axis indicates the false positive rate of the risk prediction. The blue line represents the performance of the nomogram.

Figure 5 The calibration curves of the CHD incidence risk prediction in the array.

Notes: (A) Training set and (B) Validation set: Calibration curves of the CHD incidence risk prediction in the array. The x-axis represents the predicted incidence risk. The y-axis represents the actual diagnosed CHD. The diagonal dotted line represents a perfect prediction by an ideal model. The solid line represents the performance of the nomogram; a closer fit to the diagonal dotted line represents a better prediction.

Figure 6 Decision curve analysis for the incidence risk nomogram of CHD.

Notes: (A) Training set and (B) Validation set: The y-axis indicates the net benefit. The dotted line represents the incidence risk nomogram of CHD. The thin solid line represents the assumption that all patients are diagnosed with CHD. The thin thick solid line represents the assumption that no patients are diagnosed with CHD.