Metformin Reduces the Risk of Total Hip Arthroplasty in Elderly Patients with Hip Osteoarthritis and Type 2 Diabetes Mellitus

Figures & data

Figure 1 The flowchart of study population selection.

Table 1 Patients’ Baseline Characteristics Between Metformin Users and Non-Metformin Users

	Before PSM			After PSM
	Metformin Group (n=491)	Non-Metformin Group (n=407)	P value	Metformin Group (n=308)	Non-Metformin Group (n=308)	P value
Demographics
Age, years	75.2±3.6	73.9±2.8	<0.01	73.1±2.3	73.2±2.4	0.37
Female, %	313 (63.6%)	241 (59.4%)	0.21	182 (59.1%)	197 (64.0%)	0.25
BMI, kg/m^2	26.0±2.3	25.3±2.1	<0.01	25.4±2.0	25.4±2.1	0.47
Comorbidities
Obesity, %	108 (22.0%)	57 (14.0%)	<0.01	39 (12.7%)	51 (16.6%)	0.21
Hypertension, %	294 (59.8%)	254 (62.6%)	0.41	195 (63.3%)	199 (64.6%)	0.80
Hyperlipidemia, %	172 (35.0%)	154 (37.9%)	0.37	111 (36.0%)	128 (41.6%)	0.19
COPD, %	57 (11.6%)	46 (11.3%)	0.92	34 (11.0%)	41 (13.3%)	0.46
Hip Pain	3.1±0.7	3.0±0.7	0.70	3.0±0.7	3.0±0.7	0.53
Medications for pain management
Systemic corticosteroids, %	108 (22.0%)	97 (23.9%)	0.52	68 (22.1%)	80 (26.0%)	0.30
NSAIDs, %	215 (43.7%)	183 (45.1%)	0.69	135 (43.8%)	145 (47.1%)	0.47
Glucosamine, %	15 (3.0%)	9 (2.2%)	0.54	9 (2.9%)	9 (2.9%)	1.00
Opioids, %	63 (12.8%)	31 (7.6%)	0.01	34 (11.0%)	28 (9.1%)	0.50
Antihyperglycemic agents
Sulfonylureas, %	74 (15.0%)	68 (16.7%)	0.52	48 (15.6%)	59 (19.2%)	0.29
α-glucosidase inhibitors, %	26 (5.3%)	27 (6.7%)	0.40	15 (4.9%)	21 (6.8%)	0.39
Thiazolidinediones, %	6 (1.2%)	4 (1.0%)	1.00	6 (1.9%)	3 (1.0%)	0.51
Insulin, %	15 (3.0%)	25 (6.2%)	0.03	7 (2.3%)	16 (5.2%)	0.09

Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; NSAIDs, nonsteroidal anti-inflammatory drugs.

Figure 2 Kaplan–Meier curve illustrating the pattern of survival free of total Hip arthroplasty (THA) by metformin using status.

Table 2 Association of Metformin Use with the Risk of Total Hip Arthroplasty

	Before PSM			After PSM
	Non-Metformin Group (n=491)	Metformin Group (n=407)	P value	Non-Metformin Group (=308)	Metformin Group (n=308)	P value
No. of events	85 (20.9%)	23 (4.7%)	<0.01	15 (4.9%)	77 (25.0%)	<0.01
Crude HR (95% CI)	Ref.	0.23 (0.16–0.34)	<0.01	Ref.	0.23 (0.15–0.34)	<0.01
Adjusted HR* (95% CI)	Ref.	0.20 (0.13–0.32)	<0.01	Ref.	0.17 (0.10–0.30)	<0.01
Time-dependent Cox model, adjusted HR^† (95% CI)	Ref.	0.15 (0.08–0.25)	<0.01	Ref.	0.12 (0.06–0.22)	<0.01

Notes: *The covariates include demographic variables, comorbidities, medications for pain management, and other antihyperglycemic agents. ^†Time-dependent Cox model considered time-varying exposure (metformin) and time-dependent covariates (demographic variables, comorbidities, medications for pain management, and other antihyperglycemic agents).

Table 3 Association of Metformin Use Duration and Dosage with the Risk of Total hip Arthroplasty

	Before PSM			After PSM
	HR	95% CI	P value	HR	95% CI	P value
Metformin use duration (Long vs short-term)
Crude model	0.48	0.21–1.18	0.12	0.13	0.03–0.48	<0.01
Adjusted model*	0.46	0.19–1.14	0.09	0.10	0.02–0.46	<0.01
Time-dependent Cox model^†	0.43	0.16–1.16	0.10	0.07	0.01–0.51	<0.01
Metformin use dosage (High vs low dose)
Crude model	0.27	0.11–0.73	<0.01	0.17	0.05–0.62	<0.01
Adjusted model*	0.25	0.10–0.62	<0.01	0.15	0.04–0.57	<0.01
Time-dependent Cox model^†	0.30	0.11–0.84	0.02	0.17	0.04–0.80	0.03

Notes: *The covariates include demographic variables, comorbidities, medications for pain management, and other antihyperglycemic agents. ^†Time-dependent Cox model considered time-varying exposure (metformin) and time-dependent covariates (demographic variables, comorbidities, medications for pain management, and other antihyperglycemic agents).