Role of Some Serum Biomarkers in the Early Detection of Diabetic Cardiomyopathy

Figures & data

Table 1. Clinical characteristics of the studied Type 2 diabetes mellitus patients and the two control groups.

Variables	DM-N n = 34 n (%)	DM-DD n = 47 n (%)	DM-SD n = 32 n (%)	DD n = 33 n (%)	Controls n = 31 n (%)	p-value
Age (mean ± SD)	54.7 ± 4.1	56.1 ± 5.7	55.7 ± 8.1	53.7 ± 5.4	55.9 ± 3.6	0.125
Gender
Males	16 (47.1)	21 (44.7)	15 (46.9)	15 (45.5)	15 (48.4)	0.173
Females	18 (52.9)	26 (55.3)	17 (53.1)	18 (54.5)	16 (51.6)
Smoking
Nonsmokers	30 (88.2)	42 (89.4)	18 (56.3)	26 (78.8)	23 (74.2)	0.23
Smoker/exsmokers	4 (11.8)	5 (10.6)	14 (43.8)	7 (21.2)	9 (25.8)
BMI	34.6 ± 5.9	33.5 ± 7.2	29.7 ± 5.1	32.1 ± 4.8	31.7 ± 8.4	0.085
History of hypertension
Yes	16 (47.1)	34 (72.3)	12 (37.5)	16 (48.5)	5 (16.1)	<0.001 ^†
No	18 (52.9)	13 (27.7)	20 (62.5)	17 (51.5)	26 (83.9)

† p < 0.01 is considered highly significant.

AGE: Advanced glycation end-product; DD: Diastolic dysfunction; DM: Diabetes mellitus; N: Normal echocardiography; SD: Systolic dysfunction.

Table 2. Laboratory analysis of the laboratory biomarkers among Type 2 diabetes mellitus patients and the two control groups.

Biomarkers	DM-N n = 34 Mean ± SD	DM-DD n = 47 Mean ± SD	DM-SD n = 32 Mean ± SD	DD n = 33 Mean ± SD	Controls n = 31 Mean ± SD	p-value
TNF-α (pg/ml)	1.8 ± 2.1	5.2 ± 3.2 ^¶ ^#	5.8 ± 3.6 ^‡‡ ^§§	2.1 ± 2.4 ^¶¶	1.4 ± 0.5	<0.001 ^‡
IL-6 (pg/ml)	2.0 ± 0.7	18.3 ± 26.2 ^¶ ^#	24.1 ± 17.9 ^‡‡ ^§§	2.2 ± 1.6 ^¶¶	1.5 ± 0.3	<0.001 ^‡
Insulin (uIU/ml)	15.9 ± 8.7	54.5 ± 57.2 ^¶ ^#	56.9 ± 57.3 ^‡‡ ^§§	14.8 ± 8.6 ^¶¶	6.7 ± 1.4	<0.001 ^‡
AGEs (ng/ml)	9.1 ± 1.3	12.9 ± 5.4 ^¶ ^# ^††	15.6 ± 4.7 ^‡‡ ^§§	7.7 ± 2.5 ^¶¶	7.8 ± 1.0	<0.001 ^‡
IGFBP-7 (ng/ml)	3.5 ± 1.1	4.6 ± 2.4 ^# ^††	6.8 ± 6.2 ^‡‡ ^§§	2.7 ± 0.9 ^¶¶	2.8 ± 0.8	<0.001 ^‡
Creatinine (mg/dl)	0.8 ± 0.2	1.1 ± 0.4 ^¶ ^# ^††	1.3 ± 0.5 ^‡‡ ^§§	0.8 ± 0.2 ^¶¶	0.8 ± 0.1	<0.001 ^‡
Cholesterol (mg/dl)	195.2 ± 49.5	168.9 ± 44.6 ^¶	149.9 ± 42.3 ^‡‡ ^††	211.6 ± 42.9 ^¶¶ ^##	182.2 ± 30.9	<0.001 ^‡
Triglyceride (mg/dl)	121.1 ± 59.8	145.7 ± 57.1	126.3 ± 45.8	143.2 ± 63.7 ^##	90.2 ± 24.2	<0.001 ^‡
HDL-C (mg/dl)	53.9 ± 16.5 ^§	36.0 ± 8.9 ^¶	31.9 ± 7.1 ^‡‡ ^§§	45.5 ± 7.1 ^¶¶ ^##	40.5 ± 14.4	<0.001 ^‡
LDL-C (mg/dl)	117.1 ± 41.9	103.5 ± 36.8	94.1 ± 37.6 ^‡‡	121.5 ± 35.6 ^¶¶	106.5 ± 14.5	0.01 ^†
FBS (mg/dl)	161.8 ± 64.1 ^§	217.1 ± 84.7 ^¶ ^# ^††	187.0 ± 72.0 ^§§	95.6 ± 12.3 ^¶¶	89.4 ± 12.9	<0.001 ^‡
HbA1c%	7.6 ± 1.7 ^§	7.8 ± 1.6 ^#	8.0 ± 1.9 ^§§	5.1 ± 0.6 ^¶¶	5.0 ± 0.3	<0.001 ^‡

† p < 0.05 is considered significant.

‡ p < 0.01 is considered highly significant.

§ Significant difference between DM-N and controls.

¶ Significant difference between DM-DD and DM-N.

# Significant difference between DM-DD and Controls.

†† Significant difference between DM-DD and DM-SD.

‡‡ Significant difference between DM-SD and DM-N.

§§ Significant difference between DM-SD and controls.

¶¶ Significant difference between DD and DM-SD, DM-DD.

## Significant difference between DD and controls.

AGE: Advanced glycation end-product; DD: Diastolic dysfunction; DM: Diabetes mellitus; HDL-C: High density lipoprotein cholesterol; LDL-C: Low density lipoprotein cholesterol; N: Normal echocardiography; SD: Systolic dysfunction.

Table 3. Logistic regression analysis for predicting the risk of different types of diabetic cardiomyopathy.

Variable	Logistic co-efficient	Adjusted odds ratio	95% CI		p-value
			Lower	Upper
DM-DD vs DM-N
Insulin ≥22.7	2.625	13.807	3.160	60.333	<0.001 ^‡
TNF-α ≥3.9	1.593	4.918	1.173	20.630	0.029 ^†
AGE ≥11.4	1.934	6.914	1.026	46.595	0.047 ^†
Creatinine ≥1.1	2.066	7.895	1.247	49.979	0.028 ^†
Constant	-2.142	0.117			<0.001 ^‡
DM-SD vs DM-DD
AGE ≥14.2	1.188	3.281	1.135	9.485	0.028 ^†
IL6 ≥6.4	2.189	8.925	2.298	34.661	0.002 ^‡
Constant	-2.507	0.082			<0.001 ^‡
DCM vs DM-N
TNF-α ≥1.7	1.783	5.945	1.666	24.749	0.014 ^†
AGE ≥11.4	2.217	9.177	1.428	50.564	0.011 ^†
IL6 ≥3.5	1.947	7.009	1.540	31.901	0.012 ^†
Constant	-1.667	.189			<0.001 ^‡

† p < 0.05 is considered significant.

‡ p < 0.01 is considered highly significant.

AGE: Advanced glycation end-product; DCM: Diabetic cardiomyopathy; DD: Diastolic dysfunction; DM: Diabetes mellitus; N: Normal echocardiography; SD: Systolic dysfunction.

Figure 1. ROC curve and AUC for laboratory biomarkers for prediction of Type 2 diabetes mellitus patients with diastolic dysfunction versus Type 2 diabetes mellitus normal cardiac function.

AGE: Advanced glycation end-product; AUC: Area under the curve; ROC: Receiver operating characteristic.

Figure 2. ROC curve and AUC for laboratory biomarkers for prediction of Type 2 diabetes mellitus patients with systolic dysfunction versus diastolic dysfunction.

AGE: Advanced glycation end-product; AUC: Area under the curve; ROC: Receiver operating characteristic.

Figure 3. ROC curve and AUC for laboratory biomarkers for prediction of Type 2 diabetes mellitus patients with diabetic cardiomyopathy versus Type 2 diabetes mellitus normal cardiac function.

AGE: Advanced glycation end-product; AUC: Area under the curve; ROC: Receiver operating characteristic.

Table 4. Biomarkers cutoff levels with sensitivity, specificity, positive and negative predictive values for differentiation between diabetic cardiomyopathy and normal cardiac function among Type 2 diabetes mellitus patients.

Biomarker cutoff level	DM-DD n = 47 n (%)	DM-N n = 34 n (%)	p-value ^†	Youden index	Sensitivity	Specificity	PPV	NPV
TNF-α ≥3.9	33 (70.2)	6 (17.6)	<0.001	0.526	70.2%	82.4%	84.6%	66.7%
Insulin ≥22.7	31 (66.0)	4 (11.8)	<0.001	0.542	66.0%	88.2%	88.6%	65.2%
AGEs ≥11.6	23 (48.9)	2 (5.9)	<0.001	0.489	48.9%	94.1%	92.0%	57.1%
Creatinine ≥1.1	24 (51.1)	2 (5.9)	<0.001	0.468	51.1%	94.1%	92.3%	58.2%
Panel of biomarkers (probability 0.709)	37 (78.7)	0 (0.0)	<0.001	0.787	78.7%	100.0%	100.0%	77.3%
Biomarker cutoff level	DM-SD n = 32 n (%)	DM-DD n = 47 n (%)	p-value ^†	Youden index	Sensitivity	Specificity	PPV	NPV
IL-6 ≥6.4	29 (90.6)	21 (44.7)	<0.001	0.459	90.6%	55.3%	58.0%	89.7%
AGEs ≥14.2	21 (65.6)	13 (27.7)	0.001	0.380	65.6%	72.3%	61.8%	75.6%
Panel of biomarkers (probability 0.316)	29 (90.6)	21 (44.7)	<0.001	0.459	90.6%	55.3%	58.0%	89.7%
Biomarker cutoff level	DCM n = 79 n (%)	DM-N n = 34 n (%)	p-value ^†	Youden index	Sensitivity	Specificity	PPV	NPV
TNF-α ≥1.7	71 (89.9)	8 (23.5)	<0.001	0.663	89.9%	76.5%	89.9%	76.5%
IL-6 ≥3.5	65 (82.3)	4 (11.8)	<0.001	0.823	82.3%	88.2	94.2%	68.2%
AGEs ≥11.4	46 (58.2)	2 (5.9)	<0.001	0.582	58.2%	94.1%	95.8%	49.2%
Panel of biomarkers (probability 0.760)	67 (84.8)	4 (11.8)	<0.001	0.730	84.8%	88.2%	94.4%	71.4%

† p < 0.01 is considered highly significant.

AGEs: Advanced glycation end-products; DCM: Diabetic cardiomyopathy; DD: Diastolic dysfunction; DM: Diabetes mellitus; N: Normal echocardiography; NPV: Negative predictive value; PPV: Positive predictive value; SD: Systolic dysfunction.