Association between admission hemoglobin level and prognosis in patients with type 2 diabetes mellitus

Table 1 Baseline characteristics of study participants according to hemoglobin levels

Variable	Hemoglobin tertile-Total	Hemoglobin tertile-Tertile 1 (< 11.97)	Hemoglobin tertile-Tertile 2 ( 11.97-< 13.90 )	Hemoglobin tertile-Tertile 3 ( 13.90-< 19.40 )	P value
Male, n (%)	166 (62.64)	34 (38.64)	49 (59.04)	83 (88.30)	< 0.001
Female, n (%)	99 (37.36)	54 (61.36)	34 (40.96)	11(11.70)	< 0.001
Age, years	58.08 ± 12.37	61.67 ± 11.89	59.23 ± 10.35	53.69 ± 13.22	< 0.001
BMI, kg/m2	25.18 ± 3.22	24.55 ± 3.40	24.94 ± 2.97	25.98 ± 3.10	0.010
Duration of diabetes, months	33.54 ± 63.81	47.14 ± 79.51	37.60 ± 66.98	17.22 ± 35.07	0.005
eGFR, mL/min per 1.73 m2	90.88 ± 46.71	60.63 ± 47.85	95.80 ± 41.82	114.85 ± 32.26	< 0.001
History of hypertension	171 (64.53%)	64 (72.73%)	57 (68.67%)	50 (53.19%)	0.014
SBP, mmHg	141.78 ± 23.10	148.42 ± 26.85	141.14 ± 22.07	136.12 ± 18.33	0.001
DBP, mmHg	79.11 ± 12.51	75.66 ± 13.83	79.69 ± 10.50	81.84 ± 12.21	0.003
ALB, g/L	39.54 ± 5.59	35.98 ± 5.61	40.03 ± 5.20	42.46 ± 3.80	< 0.001
BUN, mmol/L	7.47 ± 6.24	9.79 ± 5.99	7.61 ± 8.49	5.15 ± 1.73	< 0.001
SUA, μmol/L	401.18 ± 250.95	449.22 ± 403.60	384.23 ± 106.65	370.82 ± 109.02	0.085
Potassium, mmol/L	4.30 ± 1.00	4.55 ± 1.26	4.31 ± 1.04	4.05 ± 0.50	0.003
Total calcium, mmol/L	2.31 ± 0.47	2.28 ± 0.56	2.35 ± 0.49	2.30 ± 0.34	0.619
Phosphate, mmol/L	1.25 ± 0.31	1.32 ± 0.34	1.28 ± 0.32	1.16 ± 0.24	0.002
Fasting glycaemia, mmol/L	7.95 ± 3.26	7.30 ± 3.09	7.65 ± 3.30	8.83 ± 3.22	0.004
HbA1c, %	11.40 ± 28.88	15.10 ± 49.59	10.22 ± 12.89	9.21 ± 2.13	0.015
TG, mmol/L	2.00 ± 1.71	1.83 ± 1.50	1.69 ± 1.24	2.41 ± 2.12	0.014
TC, mmol/L	4.52 ± 2.28	4.45 ± 1.53	4.70 ± 3.64	4.42 ± 0.96	0.703
24-h urinary protein, mg/d	1179.17 ± 2346.72	2345.71 ± 3176.33	966.77 ± 2086.06	274.63 ± 450.60	< 0.001
ACEI/ARB use, n (%)	228 (86.04)	58 (65.91)	77 (92.77)	93 (98.94)	< 0.001
CKD stage, n (%)					< 0.001
1	148 (55.85)	23 (26.14)	51 (61.45)	74 (78.72)
2	41 (15.47)	10 (11.36)	14 (16.87)	17 (18.09)
3	38 (14.34)	25 (28.41)	12 (14.46)	1 (1.06)
4	19 (7.17)	13 (14.77)	4 (4.82)	2 (2.13)
5	19 (7.17)	17 (19.32)	2 (2.41)	0 (0.00)

Continuous variables are presented as the mean (SD) or median (IQR), and categorical variables are presented as the number (percentage). BMI: Body mass index; eGFR: Estimated glomerular filtration rate; SBP: Systolic blood pressure; DBP: Diastolic blood pressure; ALB: Serum albumin; BUN: Blood urea nitrogen; SUA: Serum uric acid; TC: Total cholesterol; TG: Triglycerides; ACEI: Angiotensin-converting enzyme inhibitors; ARB: Angiotensin receptor blockers; CKD: Chronic kidney disease.

Table 2 Threshold effect analysis of hemoglobin levels on the composite endpoint using piecewise linear regression

Cutoff point of hemoglobin level (K)	Hazard ratio (95%CI)1	P value
< 13.3 g/dL	0.58 (0.42-0.79)	0.0007
≥ 13.3 g/dL	1.63 (0.83-3.20)	0.1585

¹Adjusted for (model 4): Age, gender, body mass index, baseline estimated glomerular filtration rate, history of hypertension, systolic blood pressure, diastolic blood pressure, 24-h urinary protein, angiotensin-converting enzyme inhibitors and/or angiotensin receptor blockers use, albumin, total cholesterol, triglycerides, duration of diabetes, fasting glycaemia, and glycosylated hemoglobin.

A 13.3 g/dL threshold for the hemoglobin level existed for risk of the composite endpoint.