C0318 - HAS-BLED Score Is Strongly Associated With a Rebleed in Patients With an Initial Episode of Gastrointestinal Bleeding Following Coronary Stenting

Monday, October 24, 2022

3:00 PM – 5:00 PM ET

Location: Crown Ballroom

Presenting Author(s)

Pedro Cortes, MD

Mayo Clinic
Jacksonville, FL

Pedro Cortés, MD¹, Jennifer Zeng, BA¹, Christian Karime, MD², Michele Lewis, MD³, Maoyin Pang, MD, PhD²
¹Mayo Clinic, Jacksonville, FL; ²Mayo Clinic Florida, Jacksonville, FL; ³Mayo Clinic Jacksonville, Jacksonville, FL

Introduction: The HAS-BLED score was developed as a decision tool for starting anticoagulation in patients with atrial fibrillation. It has not been validated in patients already on dual antiplatelet therapy (DAPT) for coronary stenting. We aimed to determine whether the HAS-BLED score was associated with a recurrence of gastrointestinal bleeding (GIB) after an initial episode. We also sought to determine other factors associated with a rebleed including pre-stenting endoscopy and continuation of P2Y12 inhibitor after the initial GIB.

Methods: We performed a retrospective study of patients who had an index GIB following coronary stenting between January 2015 and June 2021 at a single academic center. Clinical data was collected retrospectively. Kaplan-Meier Estimates and Cox Proportional Hazards Regression Analysis were utilized to determine factors associated with a rebleed. Statistical analysis was performed utilizing BlueSky Statistics software v. 7.0.

Results: A hundred patients were included in the study. Table 1. 36 had a rebleed (median time to rebleed, 53.5 days, 14.8-212.5). Having had a major bleed prior to the coronary stenting (HR=2.82, 95% CI: 1.45-5.49, p=0.002) and labile INR (HR=2.75, 95% CI: 1.43-5.30, p=0.003) were strongly associated with an increased risk of a rebleed. Chronic kidney disease (dialysis-dependent, or creatinine > 2.26 mg/dL) was weakly associated with a rebleed, p=0.0823. Every point increase in the HAS-BLED score was associated with a 46% increased risk of a rebleed, p=0.0014. A HAS-BLED score of greater than 3 was associated with a 187% increased risk of a rebleed. A cutoff greater than 3 had a sensitivity 42.2%, specificity 72.2%, and positive predictive value of 50.9% for a rebleed. The indication for the initial coronary stenting, Figure 1, and platelet count and number of red blood cell transfusions at the time of index GIB were associated with a rebleed at p < 0.05. A pre-stenting endoscopy or being continued on prescribed P2Y12 inhibitor were not associated with a rebleed.

Discussion: In our single-center study, we found the HAS-BLED score at the time of coronary stenting was specific but not sensitive for a rebleed. Regardless, the HAS-BLED score may serve as a predictive tool for a rebleed in this patient population. Of its individual variables, chronic kidney disease, prior major bleeding, and labile INR were associated with a higher rebleed risk. We found that the continuation of P2Y12 inhibitor did not increase the risk of rebleed.

Figure: Figure 1: Kaplan-Meier Estimates for A) Time to Rebleed after Index GIB Following Coronary Stenting for All Patients, B) Time to Rebleed after Index GIB by HAS-BLED Score > 3, and C) Time to Rebleed after Index GIB by Indication of Coronary Stenting. Legend A: Red = All Patients; Legend B: Blue = HAS-BLED > 3, Red = HAS-BLED 3 or less; Legend C: Blue = STEMI, Red = NSTEMI, Yellow = Stable CAD.

Baseline Characteristics Median (IQR) or Fraction (%)		Unadjusted Cox Regression For Recurrent GI Bleeding
	All Patients N=100	HR (95% CI)	p-value
Age at Coronary Stent Placement, per 1 year	70.5 (61.7-77.8)	1.00 (0.75-1.34)	1.00
Male gender	74 (74.0%)	0.76 (0.38-1.51)	0.43
White Race	100 (100%)	NA	NA
Hispanic Ethnicity	1 (1.0%)	NA	NA
Never Smoker	32 (32.0%)	0.52 (0.24-1.16)	0.1093
Body Mass Index, per 1 kg/m²	30.0 (26.4-30.4)	1.00 (0.95-1.05)	0.91
Obesity	50 (50.0%)	1.23 (0.64-2.38)	0.53
Comorbidities – defined as per HAS-BLED
Hypertension	91 (91.0%)	0.70 (0.25-1.98)	0.50
Chronic Kidney Disease	15 (15.0%)	2.04 (0.91-4.57)	0.0823
Liver Disease	15 (15.0%)	1.39 (0.57-3.37)	0.47
History of stroke	12 (12.0%)	0.85 (0.30-2.42)	0.77
Prior Major Bleeding (before LHC)	28 (28.0%)	2.82 (1.45-5.49)	0.0023
Labile INR	29 (29.0%)	2.75 (1.43-5.30)	0.0025
Age > 65	66 (66.0%)	1.64 (0.77-3.50)	0.20
Medication predisposing to bleeding	100 (100.0%)	NA	NA
Alcohol use	26 (26.0%)	1.14 (0.53-2.42)	0.74
HAS-BLED score	4 (3-5)	1.46 (1.16-1.84)	0.0014
· HAS-BLED > 3	53 (53.0%)	2.87 (1.41-6.17)	0.0041
Coronary Catheterization Data
Pre-catheterization Endoscopy Performed	12 (12.0%)	1.59 (0.62-4.12)	0.34
· Delay in LHC due to endoscopy	1 (8.3%)	NA	NA
Indication for coronary catheterization
· Acute Coronary Syndrome	65 (65.0%)	2.29 (1.07-4.92)	0.0332
o NSTEMI	41 (41.0%)	1.07 (0.55-2.07)	0.85
o STEMI	24 (24.0%)	2.52 (1.24-5.12)	0.0110
· Stable CAD	35 (35.0%)	0.44 (0.20-0.94)	0.0332
Number of Stents Placed, per 1 stent	2 (1-2)	1.23 (0.87-1.75)	0.24
· 3 or more stents placed	20 (20.0%0	1.45 (0.66-3.20)	0.36
Hemoglobin prior to catheterization, per 1 g/dL	12.70 (10.5-14.2)	0.90 (0.78-1.03)	0.13
Medications After Catheterization
Proton pump inhibitor	31 (31.0%)	0.74 (0.35-1.57)	0.43
SSRI	12 (12.0%)	1.29 (0.50-3.31)	0.60
NSAIDs	2 (2.0%)	NA	NA
Anticoagulation	42 (42.0%)	1.24 (0.64-2.38)	0.53
· Warfarin	27 (27.0%)	1.92 (0.63-5.88)	0.25
· DOAC	14 (14.0%)	0.57 (0.19-1.74)	0.33
Indication for Anticoagulation
· Atrial Fibrillation	33 (33.0%)	1.92 (0.44-8.39)	0.38
· DVT/PE	6 (6.0%)	0.90 (0.21-3.93)	0.89
Initial Episode of Gastrointestinal Bleed (GIB)
Median time to Index GIB, days	166 (22.8-374.8)	NA	NA
P2Y12 inhibitor taken prior to admission	89 (89.0%)	1.21 (0.43-3.42)	0.72
Presenting sign
· None	10 (10.0%)	1.10 (0.39-3.13)	0.86
· Hematemesis	11 (11.0%)	0.77 (0.23-2.50)	0.66
· Melena	50 (25.0%)	0.67 (0.35-1.30)	0.24
· Hematochezia	32 (32.0%)	1.00 (0.50-2.01)	0.99
Labs at time of GIB
· Hemoglobin, per 1 g/dL	8.4 (6.7-10.1)	0.87 (0.75-1.00)	0.0506
o Hemoglobin drop, per 1 g/dL	3.8 (2.7-5.5)	1.03 (0.90-1.18)	0.65
· Platelets, per 50 x 10⁹/L	207 (167.8-272.8)	1.25 (1.07-1.46)	0.005
· INR, per 1 point	1.2 (1.1-1.9)	1.14 (0.95-1.36)	0.15
Endoscopy Performed	81 (81.0%)
· Inpatient Procedure	74/81 (91.4%)	1.45 (0.34-6.12)	0.61
· Normal endoscopy	9/81 (11.1%)	1.67 (0.58-4.82)	0.34
· Esophageal varices	0	NA	NA
· Esophagitis	10/81 (12.4%)	1.67 (0.57-4.87)	0.35
· Gastritis	11/81 (13.6%)	1.67 (0.68-4.11)	0.26
· Ulcerations	33/81 (40.7%)	0.78 (0.36-1.67)	0.52
Source of Bleeding Found
· Unknown	26 (32.1%)	1.23 (0.58-2.61)	0.59
Intervention Performed	34/81 (42.0%)	0.63 (0.29-1.35)	0.23
Number of pRBCs transfused, per 1 unit	1 (0-3)	1.14 (1.01-1.30)	0.0453
Transfer to ICU during hospitalization	21 (21.0%)	1.17 (0.51-2.68)	0.71
Severe GIB	26(26.0%)	1.49 (0.73-3.04)	0.27
P2Y12 Inhibitor Discontinued due to GIB	19/100 (19.0%)	0.56 (0.73-3.04)	0.27
Endpoint
Recurrence of GI Bleeding	36/100 (36.0%)	NA	NA
· Median time to recurrent GIB, days	53.5 (14.8-121.5)	NA	NA

Table: Table 1: Baseline Characteristics of All Patients with an Index GIB and Cox Proportional Hazard Regression Analysis for the Occurrence of a Rebleed

Disclosures:

Pedro Cortés indicated no relevant financial relationships.

Jennifer Zeng indicated no relevant financial relationships.

Christian Karime indicated no relevant financial relationships.

Michele Lewis indicated no relevant financial relationships.

Maoyin Pang indicated no relevant financial relationships.

Pedro Cortés, MD¹, Jennifer Zeng, BA¹, Christian Karime, MD², Michele Lewis, MD³, Maoyin Pang, MD, PhD². C0318 - HAS-BLED Score Is Strongly Associated With a Rebleed in Patients With an Initial Episode of Gastrointestinal Bleeding Following Coronary Stenting, ACG 2022 Annual Scientific Meeting Abstracts. Charlotte, NC: American College of Gastroenterology.