A strategy to prevent post-endoscopic retrograde cholangiopancreatography bleeding in end-stage renal disease patients

doi:10.21203/rs.3.rs-2439116/v1

Download PDF

Research Article

A strategy to prevent post-endoscopic retrograde cholangiopancreatography bleeding in end-stage renal disease patients

https://doi.org/10.21203/rs.3.rs-2439116/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background:

End-stage renal disease (ESRD) is a risk factor of post-endoscopic retrograde cholangiopancreatography (ERCP) bleeding, which may prolong hospitalization or increase the mortality rate. Hemodialysis is known to correct platelet dysfunction from ESRD but whether it can decrease post-ERCP bleeding is still uncertain. Endoscopic papillary balloon dilation (EPBD) had been reported to decrease post-ERCP bleeding but lacks evidence on whether replacing endoscopic sphincterotomy (EST).

Aims:

To explore the predictive factors of post-ERCP bleeding and if hemodialysis could decrease post-ERCP bleeding in ESRD patients. Furthermore, to investigate if the post ERCP bleeding rate, the severity of hemorrhage, and difficulty of treatment differ between EST and EPBD.

Materials and Methods:

This retrospective study acquired patient data from Chang Gung Memorial Hospitals by using Chang Gung Research Database (CGRD) from January 1, 2010, to June 30, 2020. We enrolled ESRD patients who underwent EST or EPBD on the naïve major papilla. First, we analyze the protective factors to prevent post-ERCP bleeding. And to compare the severity of post-ERCP bleeding and difficulty of hemostatic treatment between EST or EPBD with bleeding.

Results:

Total 580 patients enrolled, and we found hemodialysis significantly decreased post ERCP bleeding in the group of eGFR < 15 mL / min / 1.73m2 (p = 0.039, OR = 0.65, 95% CI: 0.43–0.98). Although the bleeding incidence was similar between EST and EPBD groups, there were significantly more patients in the EST bleeding group who required invasive hemostasis procedures (p = 0.001, N = 30/41, 73.17%). Patients who required two or more hemostasis procedures, such as trans-arterial embolization or endoscopic hemostasis, were exclusively in the EST bleeding group.

Conclusion:

Hemodialysis can decrease post ERCP bleeding rate in patients with eGFR < 15 mL / min / 1.73m². Although the bleeding incidence is similar between EST and EPBD in ESRD patients, hemostatic treatments were more complicated in EST with bleeding than in EPBD.

Previous studies have found that ESRD is a risk factor for post-ERCP bleeding [1–3]. Hemostasis consists of primary hemostasis, coagulation, and fibrinolysis, which is a concurrent process that induces the formation of a hemostatic plug at the damaged wound of the major papilla after EST or EPBD [4]. Several factors contribute to the pathogenesis of uremic bleeding. First, platelet abnormalities occur with reduced dense granule content, intracellular ADP and serotonin. The release of the platelet a-granule protein and β-thromboglobulin occurs in addition to an enhanced intracellular c-AMP. Defective cyclooxygenase activity and abnormalities with the immobilization of platelet Ca⁺⁺ are found. Furthermore, arachidonic acid metabolism and platelet aggregation occur. Secondly, defects in platelet–vessel wall interaction, including abnormal platelet adhesion, along with an altered von Willebrand factor and increased formation of vascular PGI2 is found. Thirdly, other abnormalities include the production of nitric oxide, altered blood rheology, defective platelet diffusivity, decreased release of ADP by erythrocytes, and erythropoietin deficiency.Abnormalities in platelet–platelet and platelet–vessel wall interaction are the main contributors to the pathogenesis of uremic bleeding. Cryoprecipitate and desmopressin acetate(DDAVP) had been reported to improve platelet dysfunction but there lacks evidence of using these agents before ERCP to prevent bleeding [5]. Thus, we would like to explore whether the use of DDAVP or cryoprecipitate before ERCP can prevent post-ERCP hemorrhage in ESRD patients.

In general, hemodialysis can decrease uremic bleeding risk by activating platelets and increasing NO inducers, including tumor necrosis factor-α and interleukin-lβ. [6, 7] However, heparin used for hemodialysis may increase the bleeding risk. Rabih Nasr and Sridhar Chilimuri suggested that hemodialysis should be done before surgery to reduce bleeding risk [8]. There is lack of studies on the use of hemodialysis specifically for ERCP. Therefore, we will investigate whether hemodialysis, on the day or one day before ERCP, in ESRD patients with maintenance hemodialysis, could decrease bleeding risk. In addition, we will examine whether hemodialysis is helpful in preventing post-ERCP bleeding in ESRD patients who have yet to undergo hemodialysis.

The comparison of post-ERCP bleeding between EST and endoscopic papillary balloon dilation/endoscopic papillary large balloon dilation (EPBD/EPLBD) has been debated for decades, although the success rate is adequate for both procedures. EST was developed in 1973 for the removal of bile duct stones initially [9, 10] and became a standard procedure of therapeutic ERCP for naïve major papilla[11]. However, some studies reported disadvantages associated with EST. Martin L. Freeman et al. enrolled 2347 patients performing EST and found complications of pancreatitis in 127 (5.4%) and hemorrhage in 48 (2.0%) [12]. A large, single-center study in Japan reported that 2.7% patients had delayed hemorrhage that occurred 24 hours after EST. Hemorrhage was more frequently found in hemodialysis patients (OR 6.44, 95% CI:1.67–24.8; p = 0.007) and the use of heparin in replacement of antithrombotic agents (OR 3.76, 95% CI:1.42–9.98; p = 0.008) [13]. Another retrospective cohort study showed a similar result in which 5.3% patients had post-EST bleeding with risk factors of hemodialysis (OR 8.4, 95% CI: 2.7–26.4) and prothrombin time prolongation [1]. Furthermore, a retrospective case-control study conducted in Europe compared the bleeding risk after EST in both ESRD patients receiving hemodialysis and renal transplant recipients, which showed a significantly increased risk of post-EST bleeding in the ESRD group up to 25% [2].

On the other hand, EPBD has been an alternative treatment to EST due to lower difficulty and relatively lower bleeding risk [14–16]. Abdul Haseeb et al suggested that EPBD without EST is most useful to reduce the risk of bleeding in patients with coagulopathy [17]. In addition, the European Society of Gastrointestinal Endoscopy guideline in 2016 recommended EPBD as an alternative to EST in the presence of coagulopathy or altered anatomy [18]. Few studies compared EST to EPBD on bleeding risk in ESRD patients and inconsistent results with a limited number of patients were found in studies that did. To the best of our knowledge, three English studies discussed the bleeding risk between EST and EPBD in ESRD patients with hemodialysis. J. Park et al. recruited 61 ESRD patients on hemodialysis who underwent ERCP and reported that the EST group had a significantly higher bleeding risk (16.7%, 5/30) than the EPBD group (0%, 0/23), p = 0.02 [19]. On the contrary, Ming‑Chang Tsai et al. enrolled 74 ESRD patients who underwent ERCP but showed no significant difference in the incidence of bleeding between the EST and EPBD groups (p = 0.484). Another recent study that recruited 80 patients showed EST with EPBD can reduce bleeding compared to only undergoing EST, but no difference in bleeding rate between only EST and only EPBD was found [20, 21]. Post-ERCP bleeding is a serious complication associated with morbidity, prolonged hospital days and even mortality, especially in patients with coagulopathy. In addition to the bleeding risk of EST and EPBD in ESRD patients, we were also interested in the severity of post-ERCP bleeding and the success rate of salvage, hemostatic treatments. This includes endoscopic hemostasis, trans-arterial embolization (TAE) or surgery, which have not been explored.

Since the risk of post-EST bleeding cannot be completely avoided after EST or EPBD in ESRD patients, a strategy to prevent or decrease the risk of post-ERCP bleeding in ESRD patients is needed. Endoscopic hemostasis, TAE or surgery could then be avoided to decrease days of hospitalization and the need for ICU admission, as well as to improve morbidity and mortality.

This is a retrospective, multi-center, cohort study. Patients were enrolled from Chang Gung Memorial Hospitals, and data was retrieved from Chang Gung Research Database (CGRD) between January 1, 2010, and June 1, 2020. CGRD contains patient data from eight Chang Gung Memorial Hospitals throughout Taiwan. Five of these eight hospitals perform ERCP, specifically two university-affiliated medical centers and three regional hospitals. The advantage of using CGRD is that it comprises patients’ characteristics, diagnostic codes, examinations, drugs, and laboratory data, which make results more reliable. Another advantage of using CGRD is that data is obtained throughout Taiwan, which allows a more accurate representation of this country. This study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and reviewed by the Institutional Review Board (IRB)/ethical committees of Chang Gung Memorial Hospital Foundation (IRB number: 202101130B0).

All ESRD patients were defined as eGFR less than 15 mL / min / 1.73m² or on long-term hemodialysis, who underwent ERCP with either EST or EPBD on naïve major papilla were included. We excluded patients less than 18 years old. Patients with hematological disease or liver cirrhosis were excluded because of bleeding tendencies. We did not exclude the patients using anti-coagulant agents because these agents would be discontinued for an appropriate duration before ERCP in clinical practice. All data collection was conducted in CGRD statistical center.

Patient characteristics include age, gender, laboratory examination, anti-coagulant agent use (aspirin, clopidogrel, novel oral anticoagulant (NOAC), warfarin), hemodialysis status (on the day of ERCP, one day before ERCP, two days before ERCP or peritoneal dialysis), cryoprecipitate before ERCP, and DDAVP before ERCP. We defined post-ERCP bleeding as whether patients had endoscopic hemostasis, trans-arterial embolization for bleeding and/or surgery for papillary bleeding. Patients who received blood transfusion with packed red blood cells (pRBC) for more than 2 units or whose hemoglobin (Hb) decreased over 2 g/dL within two weeks after ERCP without obvious bleeding sources were enrolled as post-ERCP bleeding according to previous literature [13, 22]. We also enrolled patients with gastrointestinal hemorrhage by discharge diagnostic codes after ERCP. We defined early bleeding as bleeding noted during ERCP or bleeding within 24 hours, and delayed bleeding as bleeding occurrence 24 hours after the procedure, according to previous studies [13].

An EST is performed using a duodenoscope (Olympus EVIS JF-260V / Olympus TJF-260V) under fluoroscopy to cut the ampullary sphincter and intraduodenal sphincter of the common bile duct or pancreatic duct following selective biliary or pancreatic cannulation. The tip of the sphincterotome is bent for the cutting wire to contact the roof of the papilla and is oriented towards the biliary sphincter or the pancreatic sphincter. EPBD is performed by using high-pressure hydrostatic balloons (Boston Scientific, CRE™ balloon dilatation catheter) to disrupt the sphincter of the major papilla after successful biliary or pancreatic cannulation. The size of EPBD is chosen according to the diameter of the distal common bile duct or pancreatic duct.

In the first part of our study, we investigated the bleeding effects of predictive factors, which include laboratory data, hemodialysis, procedures with EST or EPBD, stent insertion during ERCP, cryoprecipitate transfusion, and DDAVP or proton pump inhibitor (PPI) usage before ERCP by multivariate regression test. Then, we explored whether ESRD patients who received hemodialysis would have a decreased incidence of post-ERCP bleeding compared to those without hemodialysis, and whether hemodialysis performed a day before ERCP could prevent bleeding. In the second part, we classified patients into EST with bleeding and EPBD with bleeding groups to compare the severity of post-ERCP bleeding. This was determined by time of bleeding, amount of Hb drop, units of pRBC transfusion, inotropic agent use, ICU admission, times of endoscopic hemostasis, TAE, duration of hospitalization and mortality. In addition, the success rate of endoscopic hemostasis is analyzed to estimate the difficulty of treating post-ERCP bleeding by endoscopy of the two groups as the sites of bleeding were different according to previous reports.

Statistical analysis:

Data obtained from CGRD was analyzed at the database center of Chang-Gung Memorial Hospital-Linkou medical center after IRB approval. Statistical analysis was performed by SAS software. Independent t-test was used to compare continuous variables, and Chi-square test was used for categorical variables between patient groups. Fisher’s exact test was performed when more than 20% of data points presented an expected frequency of < 5. Univariable and multivariable Cox regression analyses were used to assess the hazard ratio of various risk factors for the prediction of post-ERCP bleeding.

Acknowldegement:

The authors thank the statistical assistance and wish to acknowledge the support of the Maintenance Project of the Center for Big Data Analytics and Statistics (Grant CLRPG3D0049) at Chang Gung Memorial Hospital for study design and monitor, data analysis and interpretation.

This study is based in part on data from the Chang Gung Research Database provided by Chang Gung Memorial Hospital. The interpretation and conclusions contained herein do not represent the position of Chang Gung Memorial Hospital.

A total of 14240 patients underwent EST or EPBD between January 1, 2010, and June 30, 2020, in Chang Gung Memorial Hospital Group. We enrolled 580 patients with end-stage renal disease with naïve major papilla after excluding patients younger than 18 years old, liver cirrhosis, and hematologic disease. In the second part of our study, these 580 patients were divided into 410 patients in the EST group and 170 patients in the EPBD group (Fig. 1).

Patient characteristics are shown in Table 1. There is no significant difference in gender and age between EST and EPBD groups. Laboratory data, related to bleeding tendency, including platelet count, prothrombin time, and serum creatinine level, were similar. There is no difference in the usage of anti-coagulant agents and indications for ERCP. Most patients received ERCP with indications of acute cholangitis due to choledocholithiasis (N = 377, 65.0%) and biliary pancreatitis (N = 65, 11.2%). Near half of our patients (302 patients, 52.2%) received maintenance hemodialysis, and most of the hemodialysis was performed within two days before ERCP (N = 289, 95.7%). No patient received peritoneal dialysis in this study. Blood transfusion with cryoprecipitate before ERCP was only performed in one patient in the EST group. In addition, a patient prescribed with DDAVP injection before ERCP was in the EST group. The hospital days were similar.

The post-ERCP bleeding incidence is 31.8% (N = 185/580) in our ESRD patients, with 32.9% (N = 135/410) in EST and 29.4% (N = 50/170) in EPBD groups respectively. Our analysis by multivariate regression showed no difference in the predictive factors of post-ERCP bleeding, which includes laboratory data, hemodialysis, procedure with EST or EPBD, stent insertion during ERCP, cryoprecipitate transfusion, and DDAVP or PPI usage before ERCP (Table 2). However, if we categorized ESRD patients into eGFR < 15 or > 15 mL / min / 1.73m² for analysis, hemodialysis could significantly decrease post-ERCP bleeding incidence in the group of eGFR < 15 mL / min / 1.73m² (p = 0.039, OR = 0.65, 95% CI: 0.43–0.98) (Table 3 and Table 3A.).

As the mechanism of sphincter injury between EST and EPBD is different, we investigated the severity of post-ERCP bleeding and the success rate of hemostatic treatment between EST and EPBD with bleeding occurrence, although there is no significant difference in the post-ERCP bleeding incidence. Our results showed no difference in the interval of bleeding, need for blood transfusion, amount of blood loss, inotropic agent usage, rate of intensive care unit (ICU) admission, length of hospitalization, post-ERCP pancreatitis and mortality between EST and EPBD groups (Table 4). However, only patients in the group of EST with bleeding occurrence received TAE for hemostasis during admission though no statistical significance (N = 6, 4.44%) was found. Additionally, we compared endoscopic hemostatic treatment between groups of EST and EPBD with bleeding. There are significantly more patients in the EST group who received hemostatic procedures in endoscopy than in the EPBD bleeding group (p = 0.001, N = 30/41, 73.17%). Although no difference was found in the number of endoscopic treatment for post-ERCP bleeding between groups, four patients in the EST bleeding group required more than twice endoscopic hemostasis, while none in the EPBD group (Table 5.).

Table 1
	Total (n = 580)	EST (n = 410)	EPBD (n = 170)	p value
Age	71.00 ± 12.01	70.73 ± 12.26	71.65 ± 11.38	0.342
Gender				0.363
Female	314 (54.14)	217 (52.93)	97 (57.06)
Male	266 (45.86)	193 (47.07)	73 (42.94)
Lab examination
Platelet count	196.2 ± 86.63	200.0 ± 87.46	186.9 ± 84.10	0.340
INR	1.16 ± 0.27	1.15 ± 0.25	1.18 ± 0.32	0.998
Creatinine	6.10 ± 3.33	5.91 ± 3.26	6.57 ± 3.47	0.058
Anti-coagulant agent use
Aspirin	105 (18.10)	81 (19.76)	24 (14.12)	0.108
Plavix	49 (8.45)	32 (7.80)	17 (10.00)	0.387
NOAC	6 (1.03)	4 (0.98)	2 (1.18)	1.000
warfarin	6 (1.03)	2 (0.49)	4 (2.35)	0.064
Indication				0.060
Choledocholithiasis	377 (65.00)	256 (62.44)	121 (71.18)
Biliary pancreatitis	65 (11.21)	50 (12.20)	15 (8.82)
Malignant obstruction	35 (6.03)	31 (7.56)	4 (2.35)
Benigh stricture	46 (7.93)	32 (7.80)	14 (8.24)
Bile leakage	8 (1.38)	8 (1.95)	0 (0)
Others	49 (8.45)	33 (8.05)	16 (9.41)
Hemodialysis				0.221
No	278 (47.93)	206 (50.24)	72 (42.35)
Yes
In two days before ERCP	289 (49.83)	195 (47.56)	94 (55.29)
Over two days before ERCP	13 (2.24)	9 (2.20)	4 (2.35)
Peritoneal dialysis	0 (0)	0 (0)	0 (0)	-
Cryoprecipitate before ERCP	1 (0.17)	1 (0.24)	0 (0)	1.000
DDAVP before ERCP	1 (0.17)	1 (0.24)	0 (0)	1.000
Post ERCP pancreatitis	96 (16.55)	61 (14.88)	35 (20.59)	0.092
Hospital days	16.79 ± 15.95	16.59 ± 15.86	17.27 ± 16.19	0.621

Table 2
	No Bleeding	Bleeding	Univariable		Multivariable
	(n = 395)	(n = 185)	OR (95%CI)	p-value	OR (95%CI)	p-value
Age	71.06 ± 12.18	70.88 ± 11.66	1.00 (0.98–1.01)	0.862
Gender
Female	203 (51.39)	111 (60.00)	reference
Male	192 (48.61)	74 (40.00)	0.70 (0.49-1.00)	0.053
Lab examination
Platelet count	197.3 ± 82.49	193.7 ± 94.94	1.00 (1.00–1.00)	0.644
INR	1.14 ± 0.23	1.21 ± 0.34	2.58 (1.16–5.75)	0.070
Creatinine	6.03 ± 3.43	6.25 ± 3.12	1.02 (0.97–1.08)	0.475
Anti-coagulant agent use
Aspirin	68 (17.22)	37 (20.00)	1.20 (0.77–1.88)	0.417
Plavix	29 (7.34)	20 (10.81)	1.53 (0.84–2.78)	0.164
NOAC	4 (1.01)	2 (1.08)	1.07 (0.19–5.89)	0.939
Warfarin	4 (1.01)	2 (1.08)	1.07 (0.19–5.89)	0.939
Hemodialysis
No	185 (46.84)	93 (50.27)	reference		reference
In wo days before ERCP	201 (50.89)	88 (47.57)	0.87 (0.61–1.24)	0.443	0.85 (0.59–1.21)	0.365
Over two days before ERCP	9 (2.28)	4 (2.16)	0.88 (0.27–2.95)	0.841	0.82 (0.25–2.76)	0.752
Dialysis
No	185 (46.84)	93 (50.27)	reference
Yes	210 (53.16)	92 (49.73)	0.87 (0.61–1.24)	0.440
Procedure
EST	275 (69.62)	135 (72.97)	reference		reference
EPBD	120 (30.38)	50 (27.03)	0.85 (0.58–1.25)	0.409	0.86 (0.58–1.27)	0.457
Stent insertion
No	375 (94.94)	173 (93.51)	reference		reference
Yes	20 (5.06)	12 (6.49)	1.30 (0.62–2.72)	0.485	1.31 (0.62–2.74)	0.481
Cryoprecipitate before ERCP	1 (0.54)	0 (0)
DDAVP before ERCP	1 (0.54)	0 (0)
PPI used before ERCP	121 (30.63)	66 (35.68)	1.26 (0.87–1.82)	0.226	1.28 (0.88–1.86)	0.192

Table 3
	eGFR < 15		p-value	eGFR ≥ 15		p-value
	No Bleeding	Bleeding		No Bleeding	Bleeding
	(n = 276)	(n = 145)		(n = 94)	(n = 28)
Hemodialysis			0.118			1.000
No	90 (32.61)	62 (42.76)		83 (88.30)	25 (89.29)
In two days before ERCP	178 (64.49)	79 (54.48)		11 (11.70)	3 (10.71)
Over two days before ERCP	8 (2.90)	4 (2.76)		-	-
Dialysis			0.039			1.000
No	90 (32.61)	62 (42.76)		83 (88.30)	25 (89.29)
Yes	186 (67.39)	83 (57.24)		11 (11.70)	3 (10.71)
Procedure			0.200			0.954
EST	185 (67.03)	106 (73.10)		71 (75.53)	21 (75.00)
EPBD	91 (32.97)	39 (26.90)		23 (24.47)	7 (25.00)
Stent insertion			0.656			0.620
No	260 (94.20)	135 (93.10)		90 (95.74)	26 (92.86)
Yes	16 (5.80)	10 (6.90)		4 (4.26)	2 (7.14)
Cryoprecipitate before ERCP	0 (0)	1 (0.69)	0.344	0 (0)	0 (0)	-
DDAVP before ERCP	0 (0)	1 (0.69)	0.344	0 (0)	0 (0)	-
PPI used before ERCP	82 (29.71)	51 (35.17)	0.252	35 (37.23)	11 (39.29)	0.844
Post ERCP pancreatitis	49 (17.75)	22 (15.17)	0.502	14 (14.89)	5 (17.86)	0.768

Table 3A
eGFR < 15	No Bleeding	Bleeding	Univariable
eGFR < 15	(n = 276)	(n = 145)	OR (95%CI)	p-value
Dialysis
No	90 (32.61)	62 (42.76)	reference
Yes	186 (67.39)	83 (57.24)	0.65 (0.43–0.98)	0.040

Table 4
	EST with bleeding	EPBD with bleeding	p-value
	(n = 135)	(n = 50)	p-value
Time of bleeding			0.910
Early	26 (19.26)	10 (20.00)
Delay	109 (80.74)	40 (80.00)
Hb drop	2.55 ± 1.67	2.42 ± 1.48	0.708
PRBC transfusion unit	4.25 ± 7.80	3.86 ± 8.29	0.903
Inotropic agent usage	26 (19.26)	9 (18.00)	0.846
ICU admission rate	29 (21.48)	13 (26.00)	0.515
TAE	6 (4.44)	0 (0)	0.193
Hospital days	22.50 ± 16.08	23.13 ± 18.83	0.980
Post ERCP pancreatitis	22 (16.30)	9 (18.00)	0.783
Mortality	15 (11.11)	7 (14.00)	0.590

Table 5
	EST with bleeding	EPBD with bleeding	p-value
	(n = 41)	(n = 14)	p-value
Endoscopy			0.001
Hemostasis	30 (73.17)	3 (21.43)
No hemostasis	11 (26.83)	11 (78.57)
Endoscopic hemostasis times	1.13 ± 0.35	1.00 ± 0.00	0.507
Endoscopy greater than or equal to 2 times	9 (21.95)	5 (35.71)	0.313
Endoscopic hemostasis greater than or equal to 2 times	4 (9.76)	0 (0)	0.562

ESRD is a risk factor of post-ERCP bleeding. Rabih Nasr and Sridhar Chilimuri suggested that hemodialysis should be done before surgery to reduce bleeding risk[8]. Previous studies on post-ERCP bleeding only enrolled ESRD patients under long-term hemodialysis and did not investigate if hemodialysis can decrease post-ERCP bleeding. Our analysis of the effect of hemodialysis on post-ERCP bleeding did not find a correlation with decreased bleeding incidence in our patients. Hemodialysis may partially improve platelet dysfunction but cannot eliminate it completely. Moreover, hemodialysis may contribute to bleeding by heparin use[23–25]. Whether hemodialysis could decrease post-ERCP bleeding is still unclear. Heparin use during hemodialysis may eliminate the effect of reducing bleeding in our ESRD patients. A prospective study of avoiding heparin use in hemodialysis before ERCP is indicated to clarify the effect of heparin use. According to one previous study, the time to ERCP from the last hemodialysis may not impact bleeding risk[19]. There was no difference between hemodialysis within two days or beyond two days before ERCP in our study, consistent with the previous study.

According to our sub-group analysis, maintenance hemodialysis could decrease post-ERCP bleeding in patients with an eGFR less than 15 mL / min / 1.73m² (p = 0.039, OD = 0.65). However, we cannot directly conclude that the bleeding risk can be decreased if hemodialysis was done before ERCP in patients with eGFR < 15 mL / min / 1.73m² without hemodialysis done before. A randomized control trial is suggested to clarify the reducing bleeding effect of hemodialysis in ESRD patients with eGFR < 15 mL / min / 1.73m² but without dialysis treatment yet.

Although previous studies suggested that EPBD has less post-ERCP bleeding, and ESGE guidelines recommend EPBD as an alternative treatment to EST with coagulopathy, there is a lack of studies that compares these two procedures in ESRD patients [17, 18, 26]. Studies comparing bleeding rate of EST and EPBD in ESRD patients with hemodialysis showed inconsistent results with the disadvantage of limited patient enrollment. In our research, we enrolled 580 ESRD patients who underwent ERCP in multi-centers across Taiwan within 10 years. To the best of our knowledge, it has the largest patient number compared to similar studies. The post-ERCP bleeding rate is 31.8% (N = 185/580) in ESRD patients; 32.9% (N = 135/410) in EST and 29.4% (N = 50/170) in the EPBD group, respectively. In our study, we found no significant difference in post-ERCP bleeding occurrence between groups although the bleeding rate seemed higher in the ES group.

Our post-ERCP bleeding rate is higher than in previous reports. Nelson et al. and Dhar et al. reported up to 25% and 26% post-ERCP bleeding rate [27, 28] and Kim et al. reported 23% [3]. However, Tsai et al. reported 8%, J. Park et al. reported 13.1% and N. Takahara reported 5.4% post-ERCP bleeding rate respectively [19, 20, 29]. All studies recruited less than 100 patients. The variation of post-ERCP bleeding rates may be attributed to the different number of patients enrolled in these studies. Another cause of a higher post-ERCP bleeding rate in our study may be due to the enrollment of ESRD patients without hemodialysis in contrast to previous studies where only hemodialytic patients were included.

Because the mechanism of papilla injury is different in EST and EPBD, we questioned if the severity of post-ERCP bleeding may be different between groups. To the best of our knowledge, no previous study examined this. EST cuts the sphincter of Oddi by sphincterotome by using a high frequency generator with mixed cutting and coagulation mode, whereas EPBD disrupts the sphincter of Oddi by hydrostatic pressure from the dilatation balloon, which results in edema with hemorrhage and inflammation around the sphincter [30, 31]. Therefore, we compared the time of bleeding, degree of Hb drop, amount of pRBC transfusion, inotropic agent usage, ICU admission, need of TAE, duration of hospitalization, post ERCP pancreatitis, and mortality rate between ES and EPBD with bleeding. However, no difference was found between EST and EPBD bleeding groups in these variables. Most post-ERCP bleeding were delayed hemorrhage and occurred in a similar rate in EST (80.7%) and EPBD (80.0%). These delayed, hemorrhage patients showed major bleeding to the degree of Hb dropping up to on average 2.5 mg/dL in EST and 2.4 mg/dL in EPBD, respectively. There were also similar rates in patients that needed inotropic agent usage (19.2% vs 18.0%) and ICU admission (21.4% vs 26%) between EST and EPBD with bleeding.

However, with the difficulty of hemostatic treatment between ES and EPBD bleeding groups, significantly more patients received endoscopic hemostasis in the EST bleeding group (p = 0.001). Only patients in the EST bleeding group received endoscopic hemostasis greater than or equal to two times (N = 4/41, 9.7%) although no statistical significance was found. We also found that 6 patients treated with TAE were in the EST bleeding group, but none were found in the EPBD group. It seems that even though the bleeding risk is comparable between EST and EPBD groups, treatment of bleeding after EST by endoscopy is much more troublesome, and in some patients, TAE was required.

DDAVP and cryoprecipitate transfusion are effective in treating uremic bleeding in ESRD patients, but the duration of effect is only a few hours. Conjugated estrogen is an alternative treatment with a longer effect of 2 weeks according to previous studies [32, 33]. However, no studies investigating these drugs in preventing post-ERCP bleeding have been done yet. There was only one patient using DDAVP and another patient receiving cryoprecipitate transfusion before ERCP in our study, thus analysis could not be done. In addition, most patients encountered delayed hemorrhage after EST or EPBD in our study, thus the duration of these drugs may not be long enough to evaluate. A prospective study is needed for further evaluation.

Recent studies showed that metal stent could treat post-ERCP bleeding by compression [34–36]. Plastic stent insertion could also be an effective, alternative option for post-ERCP bleeding in patients, shown in one previous study [37]. Thus, we tried to analyze if temporary biliary stent insertion during ERCP could prevent post-ERCP bleeding in our ESRD patients. However, no clinical significance was found with stent insertion to prevent post-ERCP bleeding. Further investigation is needed to examine if different sizes of stents, stent number, and whether metal or plastic stents could impact post-ERCP bleeding rate in ESRD patients.

There were several advantages in our study, 1. It has the largest patient sample size; 2. It is a multi-center research across Taiwan; 3. We recruited ESRD patients both with or without hemodialysis thus we were able to analyze hemodialysis effects on the post-ERCP bleeding incidence; 4. We were the first to compare the severity of post-ERCP bleeding between EST and EPBD and the difficulty in management. There were some limitations, 1. The size of EST or EPBD could not be identified according to CGRD database, but data shows a relation to bleeding; 2. Minor bleeding could be missed by our enrolled criteria although it is less important; 3. The number of patients that used DDAVP or cryoprecipitate before ERCP is too small to be analyzed.

In conclusion, our study revealed that post-ERCP bleeding rate is comparable between EST and EPBD in ESRD patients, but treatment through endoscopic hemostasis showed more difficult in EST with bleeding. In addition, hemodialysis may decrease post-ERCP bleeding in patients with eGFR < 15 mL / min / 1.73m².

Acknowldegement:

Ethical Approval

This study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and reviewed by the Institutional Review Board (IRB)/ethical committees of Chang Gung Memorial Hospital Foundation (IRB number: 202101130B0).

Competing interests

We declare that there’s no competing interests.

Authors' contributions

S.W. and K.S. wrote the main manuscript, C.W. and M.L. prepared the tables, Y.T., C.L.,N.L. and C.Y. modified the manuscript, Y.W. analysed the data.

Funding

We declare that there’s no funding.

Availability of data and materials

We declare that our data and materials could be accessed.

Nelson, D.B. and M.L.J.J.o.c.g. Freeman, Major hemorrhage from endoscopic sphincterotomy: risk factor analysis. J Clin Gastroenterol, 1994. 19(4): p. 283–287.
Dhar, A., et al., S1437: Incidence Rates of Post-ERCP Complications: A Case Control Study in Renal Transplant Recipients and Dialysis Patients. Gastrointest Endosc 2010. 71(5): p. AB161-AB162.
Kim, S.B., K.H. Kim, and T.N. Kim, Safety and Efficacy of Endoscopic Retrograde Cholangiopancreatography for Choledocholithiasis in Long-Term Dialysis: A Propensity Score Analysis. Dig Dis Sci, 2018. 63(11): p. 3141–3146.
Stassen, J.M., J. Arnout, and H.J.C.m.c. Deckmyn, The hemostatic system. Curr Med Chem, 2004. 11(17): p. 2245–2260.
Galbusera, M., G. Remuzzi, and P. Boccardo, Treatment of bleeding in dialysis patients. Semin Dial, 2009. 22(3): p. 279–86.
Noris, M., et al., Enhanced nitric oxide synthesis in uremia: implications for platelet dysfunction and dialysis hypotension. Kidney Int, 1993. 44(2): p. 445–450.
Herbelin, A., et al., Influence of uremia and hemodialysis on circulating interleukin-1 and tumor necrosis factor α. Kidney Int, 1990. 37(1): p. 116–125.
Nasr, R. and S. Chilimuri, Preoperative Evaluation in Patients With End-Stage Renal Disease and Chronic Kidney Disease. Health Services Insights, 2017. 10.
Freeman, M.L., et al., Complications of endoscopic biliary sphincterotomy. Gastrointest Endosc, 1996. 335(13): p. 909–919.
Sohma, S., et al., Endoscopic Papillotomy: A new approach for extraction of residual stone.‡ V Congreso Internacional de Endoscopia Gastrointestinal. Gastroenterol Endosc, 1974. 22.
Ikeda, S., et al., Endoscopic sphincterotomy: long-term results in 408 patients with complete follow-up. Endoscopy, 1988. 20(01): p. 13–17.
Freeman, M.L., et al., Complications of endoscopic biliary sphincterotomy. N Engl J Med, 1996. 335(13): p. 909–919.
Ikarashi, S., et al., Factors associated with delayed hemorrhage after endoscopic sphincterotomy: Japanese large single-center experience. J Gastroenterol, 2017. 52(12): p. 1258–1265.
Weinberg, B., W. Shindy, and S.J.C.D.o.S.R. Lo, Endoscopic balloon sphincter dilation (sphincteroplasty) versus sphincterotomy for common bile duct stones. Cochrane Database Syst Rev., 2006(4).
Kozarek, R.J.E., Balloon dilation of the sphincter of Oddi. Endoscopy, 1988. 20(S 1): p. 207–210.
DiSario, J., et al., Endoscopic balloon dilation compared to sphincterotomy (EDES) for extraction of bile duct stones: preliminary results. Gastrointest Endosc, 1997. 45(4): p. AB129.
Haseeb, A. and M.L. Freeman, Endoscopic Papillary Large Balloon Dilation Versus Endoscopic Sphincterotomy for Treatment of Bile Duct Stones. Curr Treat Options Gastroenterol, 2019. 17(2): p. 221–230.
Testoni, P.A., et al., Papillary cannulation and sphincterotomy techniques at ERCP: European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline. Endoscopy, 2016. 48(7): p. 657–83.
Park, J.S., et al., Clinical outcome of endoscopic retrograde cholangiopancreatography for choledocholithiasis in end-stage renal disease patients on hemodialysis. Turk J Gastroenterol, 2020. 31(7): p. 538–546.
Tsai, M.C., et al., Major bleeding risk of endoscopic sphincterotomy versus endoscopic papillary balloon dilatation in hemodialysis patients. Saudi J Gastroenterol, 2019. 25(2): p. 106–112.
Wu, J.H., et al., Comparison of Different Endoscopic Methods Used for Managing Choledocholithiasis in Patients with End-Stage Renal Disease Undergoing Hemodialysis. Dig Dis Sci, 2022.
Cotton, P.B., et al., Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc, 1991. 37(3): p. 383–93.
Kaw, D. and D. Malhotra, Platelet dysfunction and end-stage renal disease. Semin Dial, 2006. 19(4): p. 317–22.
Boccardo, P., G. Remuzzi, and M. Galbusera, Platelet dysfunction in renal failure. Semin Thromb Hemost, 2004. 30(5): p. 579–89.
Kaw, D. and D. Malhotra. Hematology: issues in the dialysis patient: platelet dysfunction and end-stage renal disease. in Seminars in dialysis. 2006. Wiley Online Library.
Feng, Y., et al., Comparison of endoscopic papillary large balloon dilation and endoscopic sphincterotomy for retrieval of choledocholithiasis: a meta-analysis of randomized controlled trials. J Gastroenterol, 2012. 47(6): p. 655–63.
Nelson, D.B. and M.L. Freeman, Major hemorrhage from endoscopic sphincterotomy: risk factor analysis. Journal of clinical gastroenterology, 1994. 19(4): p. 283–287.
Dhar, A., et al., S1437: Incidence Rates of Post-ERCP Complications: A Case Control Study in Renal Transplant Recipients and Dialysis Patients. Gastrointestinal Endoscopy, 2010. 71(5): p. AB161-AB162.
Takahara, N., et al., Endoscopic papillary balloon dilation for bile duct stones in patients on hemodialysis. J Gastroenterol, 2012. 47(8): p. 918–23.
Ryozawa, S., et al., Japan Gastroenterological Endoscopy Society guidelines for endoscopic sphincterotomy. Digestive Endoscopy, 2018. 30(2): p. 149–173.
Kogure, H., et al., Multicenter randomized trial of endoscopic papillary large balloon dilation without sphincterotomy versus endoscopic sphincterotomy for removal of bile duct stones: MARVELOUS trial. Endoscopy, 2020. 52(9): p. 736–744.
Weigert, A.L. and A.I. Schafer, Uremic bleeding: pathogenesis and therapy. Am J Med Sci, 1998. 316(2): p. 94–104.
Couch, P. and J.L. Stumpf, Management of uremic bleeding. Clin Pharm, 1990. 9(9): p. 673–81.
Conthe, A., et al., Usefulness of fully covered self-expandable biliary metal stents for the treatment of post-sphyncterotomy ERCP bleeding. Rev Esp Enferm Dig, 2019. 111(12): p. 931–934.
Blero, D., V. Huberty, and J. Devière, Novel biliary self-expanding metal stents: indications and applications. Expert Rev Gastroenterol Hepatol, 2015. 9(3): p. 359–67.
Canena, J., et al., Short-term stenting using fully covered self-expandable metal stents for treatment of refractory biliary leaks, postsphincterotomy bleeding, and perforations. Surg Endosc, 2013. 27(1): p. 313–24.
Schmitz, D., et al., Is insertion of a plastic stent better and safer than epinephrine injection in post sphincterotomy bleeding? Eur J Gastroenterol Hepatol, 2020. 32(4): p. 484–489.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

A strategy to prevent post-endoscopic retrograde cholangiopancreatography bleeding in end-stage renal disease patients

Status:

Version 1

Abstract

Background:

Aims:

Materials and Methods:

Results:

Conclusion:

Figures

Introduction

Materials And Methods

Statistical analysis:

Results

Discussion

Declarations

References

Additional Declarations

Status:

Version 1