The novel risk score model for predicting the poor anticoagulation control in patients with atrial fibrillation taking warfarin

Methavigul, Komsing; Krittayaphong, Rungroj

doi:10.2478/abm-2025-0013

Figures & Tables

Flow diagram of study patients. INR, international normalized ratio; TTR, time in therapeutic range.

ROC curve illustrating discrimination performance between (A) SHOB-D2AF score and SAMe-TT2R2 score and (B) predicted probability derived from SHOB-D2AF score and SAMe-TT2R2 score. AF, atrial fibrillation; AUC, area under the curve; ROC, receiver operating characteristic.

NRI of SHOB-D2AF compared with SAMe-TT2R2 model. AF, atrial fibrillation; IDI, Integrated Discrimination Index; NRI, Net Reclassification Index; TTR, time in therapeutic range.

Mean TTR stratified by SHOB-D2AF score and SAMe-TT2R2 score

Risk scoring system	No. of patients	Mean TTR ± SD
SHOB-D₂AF score
0	10	71.9 ± 28.2
1	160	59.0 ± 27.8
2	423	57.9 ± 25.4
3	660	55.2 ± 25.6
4	544	51.8 ± 26.9
5	297	48.5 ± 25.8
6	106	43.1 ± 25.3
7	32	40.9 ± 23.5
8	1	47.4 ± 0.0
Total	2,233	53.6 ± 26.4
SAMe-TT₂R₂ score
2	526	53.8 ± 25.4
3	1,041	53.9 ± 27.2
4	583	53.2 ± 25.8
5	75	50.8 ± 26.0
6	8	45.7 ± 24.1
Total	2,233	53.6 ± 26.4

Definition of SHOB-D2AF score

Acronym	Description	Coefficient	Points
S	Symptomatic AF	0.258021	1
H	Heart failure	0.264288	1
O	Omit obesity (BMI < 25 kg/m²)	0.200332	1
B	Bleeding history	0.306595	1
D₂	Diabetes	0.396291	2
AF	Duration from AF diagnosis <2 years	0.313393	1
	Non-paroxysmal AF	0.255041	1
	Constant	0.531131
Maximum points			8

Baseline characteristics of the study population

Variables	All patients (n = 2,233) n (%)	TTR ≥ 65% (n = 801) n (%)	TTR < 65% (n = 1,432) n (%)	P
Age (years)	68.9 ± 10.6	68.6 ± 10.6	69 ± 10.5	0.44
Female sex	980 (43.9)	351 (43.8)	629 (43.9)	0.96
BMI (kg/m²)	25.3 ± 4.8	25.5 ± 4.6	25.1 ± 4.9	0.08
Duration from AF diagnosis (years)	3.6 ± 4.4	3.8 ± 4.7	3.4 ± 4.2	0.04*
AF pattern				0.09
Paroxysmal AF	631 (28.3)	249 (31.1)	382 (26.7)
Persistent AF	421 (18.9)	145 (18.1)	276 (19.3)
Permanent AF	1,181 (52.9)	407 (50.8)	774 (54.1)
Symptomatic AF	1,720 (77.0)	592 (73.9)	1,128 (78.8)	0.01*
Medical history
Diabetes	610 (27.3)	184 (23)	426 (29.7)	<0.01*
Hypertension	1,641 (73.5)	574 (71.7)	1,067 (74.5)	0.14
Dyslipidemia	1,320 (59.1)	485 (60.5)	835 (58.3)	0.30
CAD	356 (15.9)	124 (15.5)	232 (16.2)	0.66
MI	130 (5.8)	41 (5.1)	89 (6.2)	0.29
PAD	29 (1.3)	12 (1.5)	17 (1.2)	0.53
Heart failure	628 (28.1)	192 (24)	436 (30.4)	<0.01*
Previous ischemic stroke/TIA	485 (21.7)	193 (24.1)	292 (20.4)	0.04*
Liver cirrhosis	48 (2.1)	16 (2.0)	32 (2.2)	0.77
Pulmonary disease	23 (1.0)	9 (1.1)	14 (1.0)	0.74
CKD	1,226 (54.9)	421 (52.6)	805 (56.2)	0.10
Dementia	20 (0.9)	8 (1.0)	12 (0.8)	0.70
Hyperthyroidism	87 (3.9)	29 (3.6)	58 (4.1)	0.62
Hypothyroidism	56 (2.5)	17 (2.1%)	39 (2.7)	0.38
Anemia	899 (40.3)	285 (35.6)	614 (42.9)	<0.01*
History of bleeding	241 (10.8)	73 (9.1)	168 (11.7)	0.06
Smoking status				0.80
Current smoking	51 (2.3)	17 (2.1)	34 (2.4)
Ex-smoking	363 (16.3)	135 (16.9)	228 (15.9)
Alcohol use	73 (3.3)	22 (2.7)	51 (3.6)	0.30
CHA₂DS₂-VASc score	3.4 ± 1.6	3.3 ± 1.6	3.4 ± 1.5	0.08
HAS-BLED score	1.6 ± 1.0	1.5 ± 1.0	1.7 ± 1.0	<0.01*
LVEF (%)	59.7 ± 14.2	60.5 ± 13.5	59.2 ± 14.5	0.04*
TTR (%)	53.6 ± 26.4	81.5 ± 11.2	38 ± 18.3	<0.01*
Concomitant medications
Antiplatelet	277 (12.4)	86 (10.7)	191 (13.3)	0.07
NSAIDs	37 (1.7)	14 (1.7)	23 (1.6)	0.80
Amiodarone/dronedarone	104 (4.7)	37 (4.6)	67 (4.7)	0.95
Flecainide/propafenone	18 (0.8)	6 (0.7)	12 (0.8)	0.82
Digitalis	371 (16.6)	126 (15.7)	245 (17.1)	0.40

Predictors of poor anticoagulation control

Predictors	Univariate analysis		Multivariate analysis
Predictors	OR (95% CI)	P	OR (95% CI)	P
Age ≥75 (years)	1.10 (0.91–1.33)	0.32
Female sex	1.00 (0.84–1.20)	0.96
BMI ≥25 (kg/m²)	0.86 (0.73–1.03)	0.10	0.82 (0.68–0.98)	0.03*
Duration from AF diagnosis ≥2 years	0.76 (0.64–0.91)	<0.01*	0.73 (0.61–0.88)	<0.01*
AF pattern
Paroxysmal AF	1.24 (1.03–1.50)	0.03*	0.78 (0.64–0.94)	0.01*
Persistent AF	1.08 (0.87–1.35)	0.497
Permanent AF	1.14 (0.96–1.35)	0.14
Symptomatic AF	1.31 (1.07–1.60)	0.01*	1.29 (1.05–1.59)	0.01*
Diabetes	1.42 (1.16–1.73)	<0.01*	1.49 (1.21–1.83)	<0.01*
Hypertension	1.16 (0.95–1.40)	0.14
Dyslipidemia	0.91 (0.76–1.09)	0.30
CAD	1.06 (0.83–1.34)	0.66
MI	1.23 (0.84–1.80)	0.29
PAD	0.79 (0.38–1.66)	0.54
Heart failure	1.39 (1.14–1.69)	<0.01*	1.30 (1.07–1.59)	0.01*
Previous ischemic stroke/TIA	0.81 (0.66–0.99)	0.04*
Liver cirrhosis	1.12 (0.61–2.06)	0.71
Pulmonary disease	0.87 (0.37–2.02)	0.74
CKD	1.16 (0.97–1.38)	0.10
Dementia	0.84 (0.34–2.06)	0.70
Hyperthyroidism	1.12 (0.71–1.77)	0.62
Hypothyroidism	1.29 (0.73–2.30)	0.39
Anemia	1.36 (1.14–1.63)	<0.01*
History of bleeding	1.33 (0.99–1.77)	0.06	1.36 (1.01–1.83)	0.04*
Smoking status	0.96 (0.77–1.19)	0.69
Alcohol use	1.31 (0.79–2.17)	0.30
LVEF < 50%	1.28 (1.02–1.60)	0.03*
Antiplatelet	1.28 (0.98–1.68)	0.07
NSAIDs	0.92 (0.47–1.79)	0.80
Amiodarone/dronedarone	1.01 (0.67–1.53)	0.95
Flecainide/propafenone	1.12 (0.42–3.00)	0.82
Digitalis	1.11 (0.87–1.40)	0.40

The sensitivity, specificity, predictive value, and AUC of the risk score system in predicting poor anticoagulation control

Risk scoring system	Sensitivity (95% CI)	Specificity (95% CI)	PPV (95%CI)	NPV (95% CI)	AUC (95% CI)
SHOB-D₂AF score ≥4	47.9(45.3–50.5)	63.3(59.9–66.6)	70.0(67.1–72.8)	40.5(37.8–43.2)	0.584(0.563–0.604)
SAMe-TT₂R₂ score ≥3	76.3(74.0–78.4)	23.1(20.3–26.2)	64.0(61.7–66.2)	35.2(31.3–39.4)	0.506(0.485–0.527)

The novel risk score model for predicting the poor anticoagulation control in patients with atrial fibrillation taking warfarin

Figures & Tables

Figure 1.

Figure 2.

Figure 3.

Mean TTR stratified by SHOB-D2AF score and SAMe-TT2R2 score

Definition of SHOB-D2AF score

Baseline characteristics of the study population

Predictors of poor anticoagulation control

The sensitivity, specificity, predictive value, and AUC of the risk score system in predicting poor anticoagulation control

Paradigm

My account