Gastroenterol Res

Gastroenterology Research, ISSN 1918-2805 print, 1918-2813 online, Open Access

Article copyright, the authors; Journal compilation copyright, Gastroenterol Res and Elmer Press Inc

Journal website http://www.gastrores.org

Original Article

Volume 10, Number 6, December 2017, pages 339-346

Rockall Score Larger Than 7 as a Reliable Criterion for the Selection of Indications for Preventive Transarterial Embolization in a Subgroup of High-Risk Elderly Patients After Primary Endoscopic Hemostasis for Non-Variceal Upper Gastrointestinal Bleeding

**Table 1.** Patient Characteristics
Basic data	PE+ (N = 37)	PE- (N = 74)	P value
Age (IQR)	67 (81 - 58)	65 (76 - 55)	0.155
Gender/male, n (%)	23 (62%)	48 (65%)	0.78
Comorbidities, n (%)	27 (73%)	51 (70%)	0.826
Heart disease, n (%)	18 (49%)	45 (62%)	0.224
Kidney disease, n (%)	6 (16%)	8 (11%)	0.546
Liver disease, n (%)	4 (11%)	5 (7%)	0.481
Cancer, n (%)	2 (5.4%)	3 (4.1%)	1.000
Metabolic disease, n (%)	5 (14%)	13 (18%)	0.77
Respiratory disease, n (%)	3 (8.1%)	5 (6.8%)	1.000
Cerebral disease, n (%)	6 (16%)	15 (21%)	0.804
Forrest IA, n (%)	5 (13.5%)	9 (12.2%)	0.842
Forrest IB, n (%)	6 (16.2%)	8 (10.8%)	0.842
Forrest IIA, n (%)	17 (45.9%)	39 (52.7%)	0.842
Forrest IIB, n (%)	9 (24.3%)	18 (24.3%)	0.842
Ulcer size, mm², IQR	151 (360 - 79)	100 (180 - 50)	0.179
Hemoglobin, g/dL, IQR	8.19 (9.9 - 5.5)	8.86 (11.2 - 7.2)	0.148
Erythrocytes, × 10¹²/L, IQR	2.68 (3.3 - 1.9)	2.93 (3.7 - 2.3)	0.095
Prothrombin international normalized ratio, IQR	1.13 (1.3 - 1.0)	1.12 (1.3 - 1.0)	0.948
Shock index, IQR	1.09 (1.4 - 0.8)	1.17 (1.5 - 0.8)	0.33
Rockall score, IQR	7 (5 - 8)	6 (5 - 8)	0.37

Table 1. Patient Characteristics

Basic data

PE+ (N = 37)

PE- (N = 74)

P value

Age (IQR)

67 (81 - 58)

65 (76 - 55)

0.155

Gender/male, n (%)

23 (62%)

48 (65%)

0.78

Comorbidities, n (%)

27 (73%)

51 (70%)

0.826

Heart disease, n (%)

18 (49%)

45 (62%)

0.224

Kidney disease, n (%)

6 (16%)

8 (11%)

0.546

Liver disease, n (%)

4 (11%)

5 (7%)

0.481

Cancer, n (%)

2 (5.4%)

3 (4.1%)

1.000

Metabolic disease, n (%)

5 (14%)

13 (18%)

0.77

Respiratory disease, n (%)

3 (8.1%)

5 (6.8%)

1.000

Cerebral disease, n (%)

6 (16%)

15 (21%)

0.804

Forrest IA, n (%)

5 (13.5%)

9 (12.2%)

0.842

Forrest IB, n (%)

6 (16.2%)

8 (10.8%)

0.842

Forrest IIA, n (%)

17 (45.9%)

39 (52.7%)

0.842

Forrest IIB, n (%)

9 (24.3%)

18 (24.3%)

0.842

Ulcer size, mm², IQR

151 (360 - 79)

100 (180 - 50)

0.179

Hemoglobin, g/dL, IQR

8.19 (9.9 - 5.5)

8.86 (11.2 - 7.2)

0.148

Erythrocytes, × 10¹²/L, IQR

2.68 (3.3 - 1.9)

2.93 (3.7 - 2.3)

0.095

Prothrombin international normalized ratio, IQR

1.13 (1.3 - 1.0)

1.12 (1.3 - 1.0)

0.948

Shock index, IQR

1.09 (1.4 - 0.8)

1.17 (1.5 - 0.8)

0.33

Rockall score, IQR

7 (5 - 8)

6 (5 - 8)

0.37

**Table 2.** Distribution of Rockall Score
Variables of the clinical Rockall score	PE+ (N = 37)	PE- (N = 74)
SBP: systolic blood pressure.
Age < 60, n (%)	10 (27%)	28 (38%)
Age 60 - 79, n (%)	17 (46%)	34 (46%)
Age > 80, n (%)	10 (27%)	12 (16%)
No major comorbidity, n (%)	13 (35%)	20 (27%)
Other comorbidities, n (%)	15 (41%)	38 (52%)
Liver, kidney cancer, n (%)	9 (24%)	15 (21%)
Hemodynamically stable, n (%)	10 (27%)	37 (50%)
Tachycardia, n (%)	17 (46%)	22 (30%)
Low SBP and tachycardia, n (%)	10 (27%)	15 (20%)

Table 2. Distribution of Rockall Score

Variables of the clinical Rockall score

PE+ (N = 37)

PE- (N = 74)

SBP: systolic blood pressure.

Age < 60, n (%)

10 (27%)

28 (38%)

Age 60 - 79, n (%)

17 (46%)

34 (46%)

Age > 80, n (%)

10 (27%)

12 (16%)

No major comorbidity, n (%)

13 (35%)

20 (27%)

Other comorbidities, n (%)

15 (41%)

38 (52%)

Liver, kidney cancer, n (%)

9 (24%)

15 (21%)

Hemodynamically stable, n (%)

10 (27%)

37 (50%)

Tachycardia, n (%)

17 (46%)

22 (30%)

Low SBP and tachycardia, n (%)

10 (27%)

15 (20%)

**Table 3.** Need for Transfusion Support and Re-Bleeding Rate
	PE+ (N = 37)	PE- (N = 74)	P value	R	P value
RBC: packed red blood cells; FFP: fresh frozen plasma.
Transfusion, n (%)	32 (87%)	63 (85%)	1.000	0.018	0.85
PRBC, units (IQR)	4 (5 - 2)	3 (5 - 2)	0.503	0.071	0.506
FFP units (IQR)	2 (2 - 2)	2 (4 - 2)	0.06	-0.223	0.06
Indications for surgical intervention
Re-bleeding, n (%)	1 (2.7%)	17 (23%)	0.017
High risk ulcer, n (%)	2 (5.4%)	6 (8.1%)	1.000
Surgery, n (%)	3 (8%)	23 (31%)	0.008	-0.297	0.011
Postoperative re-bleeding, n (%)	4 (11%)	20 (27%)	0.053	-0.19	0.047

Table 3. Need for Transfusion Support and Re-Bleeding Rate

PE+ (N = 37)

PE- (N = 74)

P value

RBC: packed red blood cells; FFP: fresh frozen plasma.

Transfusion, n (%)

32 (87%)

63 (85%)

1.000

0.018

0.85

PRBC, units (IQR)

4 (5 - 2)

3 (5 - 2)

0.503

0.071

0.506

FFP units (IQR)

2 (2 - 2)

2 (4 - 2)

0.06

-0.223

0.06

Indications for surgical intervention

Re-bleeding, n (%)

1 (2.7%)

17 (23%)

0.017

High risk ulcer, n (%)

2 (5.4%)

6 (8.1%)

1.000

Surgery, n (%)

3 (8%)

23 (31%)

0.008

-0.297

0.011

Postoperative re-bleeding, n (%)

4 (11%)

20 (27%)

0.053

-0.19

0.047

**Table 4.** Outcome Data
	PE+ (N = 37)	PE- (N = 74)	P value	R	P value
ICU, n (%)	36 (97%)	59 (81%)	0.018	0.224	0.018
ICU stay, days (IQR)	3/(4 - 2)	3/(5 - 2)	0.981	-0.009	0.932
Hospital stay, days (IQR)	7(12-6)	8 (12 - 7)	0.067	-0.175	0.066
Mortality, n (%)	2 (5%)	8 (11%)	0.49	-0.091	0.343

Table 4. Outcome Data

PE+ (N = 37)

PE- (N = 74)

P value

ICU, n (%)

36 (97%)

59 (81%)

0.018

0.224

0.018

ICU stay, days (IQR)

3/(4 - 2)

3/(5 - 2)

0.981

-0.009

0.932

Hospital stay, days (IQR)

7(12-6)

8 (12 - 7)

0.067

-0.175

0.066

Mortality, n (%)

2 (5%)

8 (11%)

0.49

-0.091

0.343

**Table 5.** Binomial Multivariate Logistic Regression-Dependent Factor Re-Bleeding
Risk factor	Odds ratio	Confidance interval for OR	P value
Rockall score	0.967	0.705 - 1.327	0.835
HGB	0.853	0.703 - 1.036	0.109
Endoscopic treatment	0.741	0.418 - 1.316	0.307
Forrest classification	1.247	0.697 - 2.233	0.457
Ulcer size	0.507	0.997 - 1.001	0.350
INR	0.507	0.138 - 1.863	0.306

Table 5. Binomial Multivariate Logistic Regression-Dependent Factor Re-Bleeding

Risk factor

Odds ratio

Confidance interval for OR

P value

Rockall score

0.967

0.705 - 1.327

0.835

HGB

0.853

0.703 - 1.036

0.109

Endoscopic treatment

0.741

0.418 - 1.316

0.307

Forrest classification

1.247

0.697 - 2.233

0.457

Ulcer size

0.507

0.997 - 1.001

0.350

INR

0.507

0.138 - 1.863

0.306

**Table 6.** Treatment Results in Patient Group with Median Rockall Score ≥ 7
	PE+ with Rockall score ≥ 7 (N = 21)	PE- with Rockall score ≥ 7 (N = 34)	P value
Surgery, n (%)	0 (0%)	11 (32%)	0.004
Re-bleeding, n (%)	1 (4.8%)	11 (33%)	0.018
Hospital stay, days (IQR)	7 (11 - 5)	9 (13 - 7)	0.063
Mortality, n (%)	2 (9.5%)	6 (18.2%)	0.461

Table 6. Treatment Results in Patient Group with Median Rockall Score ≥ 7

PE+ with Rockall score ≥ 7 (N = 21)

PE- with Rockall score ≥ 7 (N = 34)

P value

Surgery, n (%)

0 (0%)

11 (32%)

0.004

Re-bleeding, n (%)

1 (4.8%)

11 (33%)

0.018

Hospital stay, days (IQR)

7 (11 - 5)

9 (13 - 7)

0.063

Mortality, n (%)

2 (9.5%)

6 (18.2%)

0.461

**Table 7.** Risk Scoring Systems
Score	Efficacy	Authors
Glasgow-Blatchford	Superior in predicting mortality and re-bleeding rate	Camus et al, 2016 [8] Martinez-Cara et al, 2016 [24]
Rockall	Superior in predicting mortality and re-bleeding rate	Camus et al, 2016 [8] Martinez-Cara et al, 2016 [24] Rockall et al, 1996 [28]
AIMS65	Similar to Glasgow-Blatchford and clinical Rockall score	Ratra et al, 2016 [30]
PNED	Superior in the prediction of mortality in comparison with the Rockall score, but was not validated externally	Stanley et al, 2017 [25] Marmo et al, 2010 [29]

Table 7. Risk Scoring Systems

Score

Efficacy

Authors

Glasgow-Blatchford

Superior in predicting mortality and re-bleeding rate

Camus et al, 2016 [8]
Martinez-Cara et al, 2016 [24]

Rockall

Superior in predicting mortality and re-bleeding rate

Camus et al, 2016 [8]
Martinez-Cara et al, 2016 [24]
Rockall et al, 1996 [28]

AIMS65

Similar to Glasgow-Blatchford and clinical Rockall score

Ratra et al, 2016 [30]

PNED

Superior in the prediction of mortality in comparison with the Rockall score, but was not validated externally

Stanley et al, 2017 [25]
Marmo et al, 2010 [29]