Impact of epinephrine volume on further bleeding due to high-risk peptic ulcer disease in the combination therapy era

Table 1 Baseline characteristics (n = 132)

Demographics	n (%)	mean ± SD	Medications	n (%)	mean ± SD
Age (yr)		70 ± 16	Antiplatelet agents	64 (48)
Sex (male)	86 (65)		Anticoagulants	36 (27)
Race (White)	96 (73)		NSAIDs	28 (21)
Presentation			Medical interventions
In-hospital bleeding	64 (48)		ICU admission	66 (50)
Hematemesis	25 (19)		Hypotension requiring vasopressors	39 (30)
Melena	93 (70)		Blood transfusion (units)		4 ± 4
Hematochezia	29 (22)
Systolic BP (mmHg)		112 ± 22	Endoscopic findings
Diastolic BP (mmHg)		63 ± 14	Time to endoscopy (h)		29 ± 29
Heart rate (BPM)		95 ± 19	Ulcer location (gastric)	54 (41)
Hemoglobin (g/dL)		8 ± 2	Forrest classification
Platelets (103/µL)		275 ± 129	Ia	13 (10)
BUN (mg/dL)		51 ± 29	Ib	47 (36)
Creatinine (mg/dL)		1.6 ± 1	IIa	72 (55)
Glasgow-Blatchford score		15 ± 3	Size (mm)		13 ± 9
Medical history			Endoscopic interventions
Cardiovascular disease	55 (42)		Additional modality
Congestive heart failure	37 (28)		Thermal therapy	60 (45)
Active malignancy	18 (14)		Clipping	53 (40)
Chronic renal dysfunction	59 (45)		Both thermal therapy and clipping	19 (14)
Dialysis use	22 (17)		Epinephrine volume (mL)		5.5 ± 3
Cirrhosis	11 (8)		Large-volume epinephrine use (≥10 mL)	18 (14)

BP: Blood pressure; BPM: Beats per minute; BUN: Blood urea nitrogen; ICU: Intensive care unit; NSAIDs: Non-steroidal anti-inflammatory drugs.

Table 2 Univariable and multivariable logistic regression analyses for factors associated with further bleeding at 7 d

Variable	OR	95%CI	P value
Univariable logistic regression:
Age (≥ 75 yr)	2.47	0.88-7.60	0.09
Admission status (in-hospital)	2.91	1.01-9.63	0.06
Hematochezia	2.96	0.98-8.6	0.04
Creatinine (mg/dL)	1.86	1.31-2.78	< 0.001
Hypotension requiring vasopressors	5.70	1.98-17.88	< 0.01
Cardiovascular disease and/or congestive heart failure	2.71	0.94-8.98	0.08
Antiplatelet therapy, anticoagulants, and/or NSAIDs	0.57	0.20-1.70	0.30
Time to endoscopy (> 24 h)	0.71	0.23-2.00	0.53
Location of ulcer (duodenal)	6.19	1.65-40.43	0.02
Forrest class (Ia and Ib)	2.47	0.88-7.60	0.09
Size of ulcer (> 20 mm)	0.89	0.13-3.59	0.88
Epinephrine volume (mL)	1.06	0.92-1.22	0.38
Multivariable logistic regression:
Hematochezia	1.48	0.41-5.05	0.54
Creatinine (mg/dL)	1.96	1.30-3.20	< 0.01
Hypotension requiring vasopressors	6.34	1.87-25.52	< 0.01
Location of ulcer (duodenal)	3.44	0.81-23.72	0.13

Variables with P < 0.05 in univariable analysis were included in multivariable analysis. aOR: Adjusted odds ratio; BPM: Beats per minute; BUN: Blood urea nitrogen; CI: Confidence interval; NSAIDs: Non-steroidal anti-inflammatory drugs; OR: Odds ratio.

Table 3 Multivariable logistic regression and cox proportional hazards analyses for factors associated with further bleeding at 30 d, need for additional therapeutic interventions, and mortality at 30 d

Variable	aOR or aHR	95%CI	P value
Further bleeding at 30 d1:
Hematochezia	2.83	0.95-8.44	0.06
Creatinine (mg/dL)	1.73	1.18-2.64	< 0.01
Hypotension requiring vasopressors	7.68	2.69-24.38	< 0.001
Epinephrine volume (mL)	1.07	0.93-1.24	0.31
Need for additional therapeutic interventions1:
Admission status (in-hospital)	1.36	0.37-5.18	0.64
Hematochezia	1.49	0.43-4.90	0.52
Creatinine (mg/dL)	1.60	1.06-2.47	0.03
Hypotension requiring vasopressors	8.53	2.51-34.72	< 0.01
Epinephrine volume (mL)	1.09	0.93-1.26	0.27
Mortality at 30 d2:
Creatinine (mg/dL)	1.77	1.36-2.30	< 0.001
Hypotension requiring vasopressors	4.09	1.39-12.09	0.01

¹Logistic regression analysis.

²Cox proportional hazards analysis. Variables with P < 0.05 in univariable analysis were included in multivariable analysis. aHR: Adjusted hazard ratio; aOR: Adjusted odds ratio; BUN: Blood urea nitrogen; CI: Confidence interval.

Table 4 Prospective combination therapy studies incorporating epinephrine for peptic ulcer disease

Ref.	Additional therapy	Mean volume (mL)	PPI	Forrest class	Number	Rebleeding		Follow-up
Karaman et al[14], 2011	Thermal	6	Yes	1a and 1b	78a	4	5%	4 wk
Kim et al[12], 2015	Thermal	6	Yes	1a, 1b, 2a	151	12	8%	30 d
Lin et al[20], 1999	Thermal	7	Yes	1a, 1b, 2a	30	2	7%	14 d
Tekant et al[22], 1995	Thermal	7	No	1b and 2a	48b	3	6%	5 d
Chau et al[18], 2003	Thermal	8	Yes	1a, 1b, 2a	164c	34	21%	10 d
Chung et al[19], 1999	Thermal	10	No	1a, 1b, 2a	41	4	10%	7 d
Lin et al[17], 2003	Thermal and Clipping	10	Yes	1a, 1b, 2a	86	7	8%	14 d
Chung et al[21], 1997	Thermal	10	Some	1a and 1b	135	5	4%	4 wk
Grgov et al[13], 2013	Clipping	11	Yes	1a, 1b, 2a	35	2	6%	8 wk
Bianco et al[16], 2004	Thermal	12	Yes	1a, 1b, 2a	58	5	9%	30 d
Taghavi et al[15], 2009	Thermal and Clipping	21	Yes	1a, 1b, 2a	147c	13	9%	30 d
Total		10			973	91	9%

^aAll patients received between 5 and 6 mL of epinephrine.

^bPatients who received endoscopic therapy for pigmented spots or adherent clots were excluded.

^cPatients who received endoscopic therapy for adherent clots were excluded. PPI: Proton pump inhibitor.