J Gastrointest Surg DOI 10.1007/s11605-015-2848-6
ORIGINAL ARTICLE
Delayed Gastric Emptying (DGE) Following Total Pancreatectomy with Islet Auto Transplantation in Patients with Chronic Pancreatitis George K. John 1 & Vikesh K. Singh 1 & Pankaj J. Pasricha 1 & Amitasha Sinha 1 & Elham Afghani 1 & Daniel Warren 3 & Zhaoli Sun 3 & Niraj Desai 3 & Christi Walsh 2 & Rita R. Kalyani 4 & Erica Hall 4 & Kenzo Hirose 2,3 & Martin A. Makary 2,3 & Ellen M. Stein 1
Received: 8 February 2015 / Accepted: 30 April 2015 # 2015 The Society for Surgery of the Alimentary Tract
Abstract Background The prevalence and factors associated with delayed gastric emptying (DGE) in patients undergoing total pancreatectomy with islet auto transplantation (TP-IAT) for chronic pancreatitis are unknown. Methods A retrospective study of all patients who underwent TP-IAT at Johns Hopkins Hospital (JHH) from August 2011 to November 2014 was performed. The International Study Group of Pancreatic Surgery (ISGPS) clinical grading of DGE was used in this study. Key Results A total of 39 patients with chronic pancreatitis underwent TP-IAT during the study period. The prevalence of DGE following TP-IAT was 35.9 %. Twenty-five patients (64.1 %) had no DGE, 10 (25.6 %) had grade A, 2 (5.1 %) had grade B, and 2 patients (5.1 %) had grade C DGE. Patients with DGE had 5.7-fold higher odds of having a hospital length of stay (LOS) greater than 14 days (OR 5.70, 95 % CI 1.37–23.76, p=0.02). Patients undergoing laparoscopic TP-IAT had significantly shorter LOS (10.5 vs. 14 days, p=0.02) and lower need for prokinetics (0.01) during the postoperative course. Conclusions and Inferences DGE is common after TP-IAT and can prolong LOS. Laparoscopic TP-IAT lowers LOS and need for prokinetics postoperatively. Further studies are needed to determine if laparoscopic approaches will improve long-term dysmotility. Keywords TP-IAT . DGE . Chronic pancreatitis
The preliminary data for this study was presented as a poster of distinction on 7 November 2014 at the joint meeting of the American Pancreatic Association and Japan Pancreas Society in Kohala Coast, HI. * Ellen M. Stein [email protected] 1
Division of Gastroenterology and Hepatology, Johns Hopkins Hospital, 1830 East Monument St, Suite 429, Baltimore, MD 21287, USA
2
Division of Surgical Oncology, Johns Hopkins Hospital, Baltimore, MD, USA
3
Division of Transplant Surgery, Johns Hopkins Hospital, Baltimore, MD, USA
4
Division of Endocrinology, Diabetes, and Metabolism, Johns Hopkins Hospital, Baltimore, MD, USA
Abbreviations TP-IAT Total pancreatectomy with islet auto transplantation ISGPS International study group of pancreatic surgery DGE Delayed gastric emptying LOS Length of stay
Introduction Total pancreatectomy with islet auto transplantation (TP-IAT) is a promising therapy for patients with chronic pancreatitis that results in an improvement in the quality of life and sustained decrease in narcotic analgesic requirements after surgery.1–3 Historical surgical treatments for chronic pancreatitis such as duodenum preserving pancreatic head resection or pancreaticoduodenectomy have been associated with high rates of delayed gastric emptying (DGE) and frequent readmissions post-surgery.4–6 However, the prevalence of
J Gastrointest Surg
dysmotility or DGE following TP-IAT has not been previously studied. Pancreaticoduodenectomy affects intestinal transit time and postprandial and inter-digestive motility post-surgery.7 Gastric emptying and intestinal motility are complex physiological processes, and the exact cause of dysmotility following surgery is still unclear.7 The enzyme cholecystokinin (CCK), primarily synthesized by I cells in the mucosal epithelium of the small intestine, especially the duodenum, has been shown to inhibit gastric emptying via a vagal capsaicin-sensitive afferent pathway and by stimulation of phasic and tonic pyloric motility.7,8 CCK levels have been shown to decrease following pylorus-preserving pancreatoduodenectomy primarily due to duodenal resection.7 Delayed gastric emptying after total pancreatectomy is also attributed to the physical removal of the duodenum which may disrupt the coordination of gastric and intestinal migrating motor complexes.9 Other proposed mechanisms for chronic dysmotility following surgery include reduced secretion of motilin,10 injury to the vagus nerve during surgery,11 anatomical alteration and torsion created during surgery,7,12 and the occurrence of complications such as gastrojejunal anastomotic ulcers or stenosis, post-surgical fluid collections, and intra-abdominal abscesses. The primary aim of this study was to determine the prevalence of and risk factors for DGE in patients following TPIAT. The secondary aim was to determine the impact of DGE on inpatient length of stay (LOS).
Materials and Methods We reviewed the medical records of all patients who underwent TP-IAT at Johns Hopkins Hospital (JHH) from August 2011 to November 2014. Data regarding demographics, co-morbidities, pre-operative dysmotility, prior non-pancreatic and pancreatic operation(s), smoking, alcohol use, illicit drug use, hospital course, and operative and postoperative course were systematically extracted and entered into a database using CDC EpiInfo™ (Atlanta, GA). Calcific chronic pancreatitis was defined as the presence of parenchymal calcification(s) and/or calculi in the main pancreatic duct on abdominal imaging. Non-calcific chronic pancreatitis was defined as the presence of characteristic pancreatic-type abdominal pain, absence of pancreatic calcification(s) or significant main pancreatic duct dilation, and two out of the following three: five or more of the nine endoscopic ultrasound (EUS) criteria, MRI-demonstrated loss of T1 signal intensity, and/or peak (60 min) bicarbonate of 14 days. Multivariate logistic regression was not performed since none of the potential confounders had a statistically significant association with DGE. All statistical analysis was done using IBM SPSS® Statistics (Version 20.0. Armonk, NY: IBM Corp.). Graphs were created using GraphPad Prism (Version 5.0, GraphPad Software, La Jolla, CA, USA).
Results A total of 39 patients with chronic abdominal pain due to chronic pancreatitis underwent TP-IAT during the study period. The majority of TP-IAT recipients were Caucasian (95 %) and women (59 %) with a mean age of 41 years (range 18– 65). The number of TP-IAT procedures per year increased over time with 6 surgeries (15 %) taking place in 2011, 8 (21 %) in 2012, 14 (36 %) in 2013 as well as 11 surgeries (28 %) completed as of November 2014. The prevalence of
J Gastrointest Surg Table 1 ISGPS consensus definition for clinical grading of DGE following TP-IAT
DGE grade
NGT requirement
Inability to tolerate solid oral intake by POD
Vomiting/gastric distension
Use of prokinetics
A B C
4–7 days or reinsertion >POD 3 8–14 days or reinsertion >POD 7 >14 days or reinsertion >POD 14
7 14 21
± + +
± + +
pre-operative bloating or constipation was 26 %, and 5 (13 %) patients had a documented history of gastroparesis prior to surgery. Nine patients (23 %) were on TPN prior to the surgery. There were 31 (80 %) patients with prior abdominal surgeries among whom 26 patients (67 %) reported a history of open or laparoscopic cholecystectomy and 6 patients (15 %) reported prior pancreatic surgery. Fourteen patients (36 %) underwent prior celiac plexus blockade. Of the 39 TP-IAT procedures, 25 (64 %) were open and 14 (36 %) were laparoscopic. Of the open surgeries, two (5 %) were originally laparoscopic but were later converted. The mean duration of surgery was 9.7 h (range 6–14 h). All Table 2 Postoperative clinical course following TP-IAT
patients were observed in the intensive care unit immediately following surgery with the median length of intensive care being 3 days (range 2–13 days). Of these 39 patients, four (10 %) required re-admission to the intensive care unit. Patients undergoing laparoscopic TP-IAT had significantly shorter hospitalizations with the median length of stay for laparoscopic TP-IAT being 10.5 days compared to 14 days for open surgery (p=0.017). The median opioid dose prior to surgery in the laparoscopic and open groups was not significantly different (50.25 oral morphine equivalents (OMEs) vs. 122.2 OMEs, p>0.05). However, the median postoperative opioid dose at discharge
Time to nasogastric tube removal
POD 0 POD 1 POD 2 POD 3 POD 4 POD 5 Need for nasogastric tube re-insertion No Yes Days of total nasogastric tube requirement 0 days 1 day 2–3 days 4–5 days 6–7 days >7 days Time to initiation of liquid diet ≤3 days 4–6 days ≥7 days Time to initiation of full oral diet 14 days Need for prokinetic medications No Yes
Number of patients
p value
Open TPIAT
Laparoscopic TPIAT
1 (4.0 %) 7 (28.0 %) 7 (28.0 %) 8 (32.0 %) 2 (8.0 %) 0 (0.0 %)
2 (14.3 %) 3 (21.4 %) 6 (42.9 %) 2 (14.3 %) 0 (0.0 %) 1 (7.1 %)
p=0.31
20 (80.0 %) 5 (20.0 %)
13 (92.9 %) 1 (7.1 %)
p=0.39
1 (4.0 %) 6 (24.0 %) 11 (44.0 %) 4 (16.0 %) 1 (4.0 %) 2 (8.0 %)
2 (14.3 %) 2 (14.3 %) 9 (64.3 %) 1 (7.1 %) 0 (0.0 %) 0 (0.0 %)
p=0.47
11 (44.0 %) 12 (48.0 %) 2 (8.0 %)
9 (64.3 %) 5 (35.7 %) 0 (0.0 %)
p=0.34
16 (64.0 %) 6 (24.0 %) 3 (12.0 %)
11 (78.6 %) 3 (21.4 %) 0 (0.0 %)
p=0.37
12 (48.0 %) 13 (52.0 %)
13 (92.9 %) 1 (7.1 %)
p=0.01
J Gastrointest Surg
was significantly higher in the open group compared to the laparoscopic group (332.4 OMEs vs. 123.5 OMEs, p=0.04). Postoperative course and DGE During the postoperative course (Table 2), nasogastric tubes were required for three or fewer days in the majority of patients (92 %) with the median duration of initial nasogastric tube use being 2 days (range 0–5 days). Only two patients in the open TP-IAT group and one patient in the laparoscopic group needed the nasogastric tube for more than 3 days. Six patients (15 %) required nasogastric tube re-insertion with a median total duration of nasogastric tube requirement being 2 days (range 0–16 days). More than half the patients (51 %) were able to tolerate a liquid diet by postoperative day 3 and solid food by postoperative day 7 (69 %). The median time to solid food intake was 6 days (range 4–16 days) with all patients in the laparoscopic group able to resume a full diet by 14 days. Prokinetic medications were required for dysmotility symptoms in 14 patients (36 %) with metoclopramide used in 9 patients (23 %). The need for prokinetic drugs was significantly different between the two groups (p=0.01), with only one patient in the laparoscopic group needing prokinetic drugs. Erythromycin or azithromycin was used in a minority of patients (3 %), and the combination of metoclopramide and erythromycin or azithromycin was used in 8 % (Table 3). The mean time to prokinetic drug initiation was 6.5 days (range 1– 16 days), and seven patients (18 %) needed prokinetic drugs at
Table 3
Evaluation and management of DGE
Use of prokinetic medications Metoclopramide only
Number of patients (%)a 9 (23.1 %)
Erythromycin/azithromycin only 1 (2.6 %) Metoclopramide+erythromycin/azithromycin 3 (7.7 %) Timing of prokinetic medications ≤3 days 2 (5.1 %) 4–6 days 5 (12.8 %) ≥7 days 7 (17.9 %) Discharged on prokinetics No 31 (79.5 %) Yes 7 (17.9 %) Method of dysmotility evaluation CT 14 (35.9 %) Upper GI series 9 (23.1 %) Upper GI endoscopy 3 (7.7 %) Initiation of nutritional support Tube feeds 6 (15.4 %) TPN 5 (12.8 %) Tube feeds+TPN 2 (5.1 %) a
Percentage of total TP-IAT patients
the time of discharge. The main modalities used to evaluate persistent dysmotility symptoms was CT scan followed by upper GI series and upper endoscopy in 36, 23, and 8 % of patients, respectively. Tube feeds were needed to be initiated in six patients (15 %) and total parenteral nutrition in five patients (13 %) and the combination of the two in two patients (5 %). Pain control was a major problem following surgery with all patients receiving an opioid PCA and epidural in the postoperative period. Nine patients (23 %) required ketamine in the postoperative period for uncontrolled pain. Patients remained on intravenous analgesics for a median duration of 7 days (range 3–35 days) after which they were switched to oral agents. The ISGPS clinical grading system was used to classify DGE following TP-IAT surgery (Fig. 1, Table 4). No DGE was seen in 25 patients (64.1 %). Grade A DGE was encountered in 10 patients (25.6 %), grade B in 2 patients (5.1 %), and grade C DGE in 2 patients (5.1 %). The overall prevalence of DGE in chronic pancreatitis patients following TP-IAT was 35.9 %, with the prevalence in the open group being 44 % compared to 21 % in the laparoscopic TP-IAT group (p=0.19). Among the open TP-IAT group, two patients (5 %) developed sustained nausea and vomiting and needed to be maintained nil per os (NPO) after postoperative day 7. The addition of these patients to the analysis as a possible delayed DGE increased the statistical significance with a trend favoring lower dysmotility in the laparoscopic TP-IAT group (p=0.09). Patients undergoing open TP-IAT had threefold higher odds (OR 2.88, 95 % CI 0.64–12.93, p=0.17) for the presence of DGE compared to patients undergoing laparoscopic TP-IAT (Fig. 1); however, this was not statistically significant. There was no association of DGE with age, gender, type of pancreatitis (calcific vs. non-calcific/minimal change pancreatitis), history of prior abdominal or pancreatic surgery, last known pre-operative opioid dose, or degree of fibrosis (all p>0.05). Patients with DGE had 5.7-fold higher odds of having a hospital length of stay (LOS) greater than 14 days (OR 5.70, 95 % CI 1.37–23.76, p=0.02).
Fig. 1 DGE grading by type of surgery
J Gastrointest Surg Table 4 DGE following TP-IAT ISGPS DGE grade
Number of patients Open TPIAT
Total number of patients (%) Laparoscopic TPIAT
No DGE
14 (56.0 %)
11 (78.6 %)
25 (64.1 %)
A B C
7 (28.0 %) 2 (8.0 %) 2 (8.0 %) 25 (100 %)
3 (21.4 %) 0 (0 %) 0 (0 %) 14 (100 %)
10 (25.6 %) 2 (5.1 %) 2 (5.1 %) 39 (100 %)
Discussion Our study provides the first estimate of the prevalence of DGE following the TP-IAT procedure, which was 35.9 % and thereby highlights an important and potentially underrecognized clinical problem. To our knowledge, we are the first authors to characterize the burden of DGE after the TP-IAT procedure using open versus laparoscopic approaches. DGE was found to significantly increase hospital LOS. However, patients undergoing the laparoscopic approach demonstrated a shorter LOS compared to patients in the open TP-IAT group and a lower need for prokinetics during the postoperative course. The presence of gastrointestinal dysmotility or DGE is well described following pancreaticoduodenectomy, although widely varying clinical definitions make an exact estimation difficult.16,17 For studies using the ISGPS definition, the prevalence of DGE following pancreaticoduodenectomy ranges between 12 and 45 %.6,18–22 Using the same ISGPS definition, the overall prevalence of DGE in our study following TP-IAT was similar at 35.9 %. The prevalence of DGE in chronic pancreatitis patients following laparoscopic TP-IAT in our study was 21 % which is slightly higher compared to the rates following laparoscopic pancreaticoduodenectomy in other studies which range between 11 and 14 %.23,24 However, prior studies included patients undergoing surgery for all indications whereas our study focuses only on patients undergoing TP-IAT surgery for chronic pancreatitis. It is possible that chronic pancreatitis patients have a higher baseline rate of gastrointestinal dysmotility which is confounded by longstanding abdominal pain and, in most cases, resultant opioid analgesia which causes delay in gastric emptying and intestinal transit. The role of the pancreas and resultant changes during the course of chronic pancreatitis in the regulation of gut motility is yet to be characterized. Although the difference in the incidence of DGE between the open and laparoscopic groups did not meet statistical significance, no patients in the laparoscopic group had severe DGE (grade B or C DGE) and the need for prokinetics in the postoperative course was significantly lower compared to the open group. Previous studies comparing laparoscopic
and open pancreaticoduodenectomies have demonstrated lower length of hospital stay and intraoperative bleeding but no difference in complications, mortality, or delayed gastric emptying.24–26 In our study, we also encountered a group of patients who, despite being initiated on a diet and not meeting classic criteria for DGE, later developed vomiting and required prokinetics but did not undergo reinsertion of a nasogastric tube. This group of patients required additional evaluation as did patients with DGE, and they may suggest a form of DGE which could be added to future modifications of the ISGPS definition. Our study has several strengths. The use of inpatient data in this study ensured complete data collection and obviates concerns regarding missing data. Our study also used the ISGPS consensus definition to grade DGE, thereby applying a validated clinical grading system in the TP-IAT population. Our study also has some limitations. Our study was limited by the relatively small sample size of patients undergoing the TP-IAT procedure and the retrospective nature of the study, yet we still detected significant differences. In addition, although an institutional algorithm was used for nasogastric tube management and prokinetic administration, it is possible that their initiation and discontinuation was influenced by the preferences of the surgical team on service at that time, thereby raising the possibility of selection bias. However, this would be in both directions and would be unlikely to significantly impact the results of this study. Since this was a retrospective study, there is also the possibility of selection bias during the original decision for TP-IAT approach since it is possible that patients with lower pre-operative risk of dysmotility such as ones without prior abdominal or pancreatic surgeries were chosen to undergo laparoscopic TPIAT. However, we did not find any significant association between prior surgeries and DGE and therefore, this is less likely to significantly impact the results of this study. Further studies are needed to estimate the burden of longterm dysmotility following TP-IAT and its association with DGE. With increasing numbers of laparoscopic TP-IATs, studies are also needed to assess for differences in DGE and long-term dysmotility as well as other clinically significant outcomes between open and laparoscopic TP-IAT in a larger cohort of patients.
J Gastrointest Surg
Conclusion In conclusion, DGE following TP-IAT is common and significantly prolongs hospital stays. A laparoscopic TP-IAT approach lowers length of stay and need for prokinetic medications postoperatively. Acknowledgments GKJ, EMS, VKS, and PJP were involved in the concept and design of this study. GKJ reviewed patient records, extracted data, performed statistical analysis, and wrote the first draft of the manuscript. All authors critically appraised and revised the manuscript. All authors gave final approval for submission of the manuscript. Funding No funding declared. Conflict of Interest No competing interests declared.
References 1.
2.
3.
4. 5. 6.
7.
8.
9.
10.
Beilman G. Pancreatitis: Is TPIAT the answer for treatment of chronic pancreatitis? Nature Reviews Gastroenterology & Hepatology 2014. Morgan K, Owczarski S, Borckardt J, Madan A, Nishimura M, Adams D. Pain control and quality of life after pancreatectomy with islet autotransplantation for chronic pancreatitis. J Gastrointest Surg 2012;16:129-33; discussion 33–4. doi 10. 1007/s11605-011-1744-y. Wilson GC, Ahmad SA, Schauer DP, Eckman MH, Abbott DE. Cost-Effectiveness of Total Pancreatectomy and Islet Cell Autotransplantation for the Treatment of Minimal Change Chronic Pancreatitis. J Gastrointest Surg 2014;6:6. Emick DM, Riall TS, Cameron JL, et al. Hospital readmission after pancreaticoduodenectomy. J Gastrointest Surg 2006;10:1243-52. Fisher AV, Sutton JM, Wilson GC, et al. High readmission rates after surgery for chronic pancreatitis. Surgery;156:787-96. Liu QY, Li L, Xia HT, Zhang WZ, Cai SW, Lu SC. Risk factors of delayed gastric emptying following pancreaticoduodenectomy. ANZ J Surg 2014;13:12850. Paraskevas K, Avgerinos C, Manes C, Lytras D, Dervenis C. Delayed gastric emptying is associated with pylorus-preserving but not classical Whipple pancreaticoduodenectomy: a review of the literature and critical reappraisal of the implicated pathomechanism. World J Gastroenterol 2006;12:5951-8. Holzer HH, Turkelson CM, Solomon TE, Raybould HE. Intestinal lipid inhibits gastric emptying via CCK and a vagal capsaicinsensitive afferent pathway in rats. Am J Physiol 1994;267:G625-9. Suzuki H, Mochiki E, Haga N, Shimura T, Itoh Z, Kuwano H. Effect of duodenectomy on gastric motility and gastric hormones in dogs. Ann Surg 2001;233:353-9. Yeo C, Barry M, Sauter P, et al. Erythromycin accelerates gastric emptying after pancreaticoduodenectomy. A prospective, randomized, placebo-controlled trial. Ann Surg 1993;218:229-37; discussion 37-8.
11.
Kim D, Hindenburg A, Sharma S, et al. Is Pylorospasm a Cause of Delayed Gastric Emptying After Pylorus-Preserving Pancreaticoduodenectomy? Ann Surg Oncol 2005;12:222-7. 12. Ueno TTA, Hamanaka Y, Tsurumi M, Suzuki T. . A proposal mechanism of early delayed gastric emptying after pylorus preserving pancreatoduodenectomy. Hepato-gastroenterology 1995;42:269-74. 13. Walsh TN, Rode J, Theis BA, Russell RC. Minimal change chronic pancreatitis. Gut 1992;33:1566-71. 14. Wallace MB, Hawes RH, Durkalski V, et al. The reliability of EUS for the diagnosis of chronic pancreatitis: interobserver agreement among experienced endosonographers. Gastrointest Endosc 2001;53:294-9. 15. Kloppel G, Maillet B. Pseudocysts in chronic pancreatitis: a morphological analysis of 57 resection specimens and 9 autopsy pancreata. Pancreas 1991;6:266-74. 16. Wente M, Bassi C, Dervenis C, et al. Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS). Surgery 2007;142:761-8. 17. Traverso LW, Hashimoto Y. Delayed gastric emptying: the state of the highest level of evidence. J Hepatobiliary Pancreat Surg 2008;15:262-9. 18. Hashimoto Y, Traverso LW. Incidence of pancreatic anastomotic failure and delayed gastric emptying after pancreatoduodenectomy in 507 consecutive patients: use of a web-based calculator to improve homogeneity of definition. Surgery 2010;147:503-15. 19. Malleo G, Crippa S, Butturini G, et al. Delayed gastric emptying after pylorus-preserving pancreaticoduodenectomy: validation of International Study Group of Pancreatic Surgery classification and analysis of risk factors. HPB 2010;12:610-8. 20. Akizuki E, Kimura Y, Nobuoka T, et al. Reconsideration of postoperative oral intake tolerance after pancreaticoduodenectomy: prospective consecutive analysis of delayed gastric emptying according to the ISGPS definition and the amount of dietary intake. Ann Surg 2009;249:986-94. 21. Welsch T, Borm M, Degrate L, Hinz U, Buchler MW, Wente MN. Evaluation of the International Study Group of Pancreatic Surgery definition of delayed gastric emptying after pancreatoduodenectomy in a high-volume centre. Br J Surg 2010;97:1043-50. 22. Park JS, Hwang HK, Kim JK, et al. Clinical validation and risk factors for delayed gastric emptying based on the International Study Group of Pancreatic Surgery (ISGPS) Classification. Surgery 2009;146:882-7. 23. Kendrick ML, Cusati D. Total laparoscopic pancreaticoduodenectomy: feasibility and outcome in an early experience. Arch Surg 2010;145: 19-23. 24. Asbun HJ, Stauffer JA. Laparoscopic vs open pancreaticoduodenectomy: overall outcomes and severity of complications using the Accordion Severity Grading System. J Am Coll Surg 2012;215:810-9. 25. Croome KP, Farnell MB, Que FG, et al. Pancreaticoduodenectomy with Major Vascular Resection: a Comparison of Laparoscopic Versus Open Approaches. J Gastrointest Surg 2014;2:2. 26. Li YB, Wang X, Wang MJ, Yang ZG, Peng B. [Delayed gastric emptying after laparoscopic versus open pancreaticoduodenectomy: a comparative study]. Zhonghua Wai Ke Za Zhi 2013;51:304-7.
ORIGINAL ARTICLE
Delayed Gastric Emptying (DGE) Following Total Pancreatectomy with Islet Auto Transplantation in Patients with Chronic Pancreatitis George K. John 1 & Vikesh K. Singh 1 & Pankaj J. Pasricha 1 & Amitasha Sinha 1 & Elham Afghani 1 & Daniel Warren 3 & Zhaoli Sun 3 & Niraj Desai 3 & Christi Walsh 2 & Rita R. Kalyani 4 & Erica Hall 4 & Kenzo Hirose 2,3 & Martin A. Makary 2,3 & Ellen M. Stein 1
Received: 8 February 2015 / Accepted: 30 April 2015 # 2015 The Society for Surgery of the Alimentary Tract
Abstract Background The prevalence and factors associated with delayed gastric emptying (DGE) in patients undergoing total pancreatectomy with islet auto transplantation (TP-IAT) for chronic pancreatitis are unknown. Methods A retrospective study of all patients who underwent TP-IAT at Johns Hopkins Hospital (JHH) from August 2011 to November 2014 was performed. The International Study Group of Pancreatic Surgery (ISGPS) clinical grading of DGE was used in this study. Key Results A total of 39 patients with chronic pancreatitis underwent TP-IAT during the study period. The prevalence of DGE following TP-IAT was 35.9 %. Twenty-five patients (64.1 %) had no DGE, 10 (25.6 %) had grade A, 2 (5.1 %) had grade B, and 2 patients (5.1 %) had grade C DGE. Patients with DGE had 5.7-fold higher odds of having a hospital length of stay (LOS) greater than 14 days (OR 5.70, 95 % CI 1.37–23.76, p=0.02). Patients undergoing laparoscopic TP-IAT had significantly shorter LOS (10.5 vs. 14 days, p=0.02) and lower need for prokinetics (0.01) during the postoperative course. Conclusions and Inferences DGE is common after TP-IAT and can prolong LOS. Laparoscopic TP-IAT lowers LOS and need for prokinetics postoperatively. Further studies are needed to determine if laparoscopic approaches will improve long-term dysmotility. Keywords TP-IAT . DGE . Chronic pancreatitis
The preliminary data for this study was presented as a poster of distinction on 7 November 2014 at the joint meeting of the American Pancreatic Association and Japan Pancreas Society in Kohala Coast, HI. * Ellen M. Stein [email protected] 1
Division of Gastroenterology and Hepatology, Johns Hopkins Hospital, 1830 East Monument St, Suite 429, Baltimore, MD 21287, USA
2
Division of Surgical Oncology, Johns Hopkins Hospital, Baltimore, MD, USA
3
Division of Transplant Surgery, Johns Hopkins Hospital, Baltimore, MD, USA
4
Division of Endocrinology, Diabetes, and Metabolism, Johns Hopkins Hospital, Baltimore, MD, USA
Abbreviations TP-IAT Total pancreatectomy with islet auto transplantation ISGPS International study group of pancreatic surgery DGE Delayed gastric emptying LOS Length of stay
Introduction Total pancreatectomy with islet auto transplantation (TP-IAT) is a promising therapy for patients with chronic pancreatitis that results in an improvement in the quality of life and sustained decrease in narcotic analgesic requirements after surgery.1–3 Historical surgical treatments for chronic pancreatitis such as duodenum preserving pancreatic head resection or pancreaticoduodenectomy have been associated with high rates of delayed gastric emptying (DGE) and frequent readmissions post-surgery.4–6 However, the prevalence of
J Gastrointest Surg
dysmotility or DGE following TP-IAT has not been previously studied. Pancreaticoduodenectomy affects intestinal transit time and postprandial and inter-digestive motility post-surgery.7 Gastric emptying and intestinal motility are complex physiological processes, and the exact cause of dysmotility following surgery is still unclear.7 The enzyme cholecystokinin (CCK), primarily synthesized by I cells in the mucosal epithelium of the small intestine, especially the duodenum, has been shown to inhibit gastric emptying via a vagal capsaicin-sensitive afferent pathway and by stimulation of phasic and tonic pyloric motility.7,8 CCK levels have been shown to decrease following pylorus-preserving pancreatoduodenectomy primarily due to duodenal resection.7 Delayed gastric emptying after total pancreatectomy is also attributed to the physical removal of the duodenum which may disrupt the coordination of gastric and intestinal migrating motor complexes.9 Other proposed mechanisms for chronic dysmotility following surgery include reduced secretion of motilin,10 injury to the vagus nerve during surgery,11 anatomical alteration and torsion created during surgery,7,12 and the occurrence of complications such as gastrojejunal anastomotic ulcers or stenosis, post-surgical fluid collections, and intra-abdominal abscesses. The primary aim of this study was to determine the prevalence of and risk factors for DGE in patients following TPIAT. The secondary aim was to determine the impact of DGE on inpatient length of stay (LOS).
Materials and Methods We reviewed the medical records of all patients who underwent TP-IAT at Johns Hopkins Hospital (JHH) from August 2011 to November 2014. Data regarding demographics, co-morbidities, pre-operative dysmotility, prior non-pancreatic and pancreatic operation(s), smoking, alcohol use, illicit drug use, hospital course, and operative and postoperative course were systematically extracted and entered into a database using CDC EpiInfo™ (Atlanta, GA). Calcific chronic pancreatitis was defined as the presence of parenchymal calcification(s) and/or calculi in the main pancreatic duct on abdominal imaging. Non-calcific chronic pancreatitis was defined as the presence of characteristic pancreatic-type abdominal pain, absence of pancreatic calcification(s) or significant main pancreatic duct dilation, and two out of the following three: five or more of the nine endoscopic ultrasound (EUS) criteria, MRI-demonstrated loss of T1 signal intensity, and/or peak (60 min) bicarbonate of 14 days. Multivariate logistic regression was not performed since none of the potential confounders had a statistically significant association with DGE. All statistical analysis was done using IBM SPSS® Statistics (Version 20.0. Armonk, NY: IBM Corp.). Graphs were created using GraphPad Prism (Version 5.0, GraphPad Software, La Jolla, CA, USA).
Results A total of 39 patients with chronic abdominal pain due to chronic pancreatitis underwent TP-IAT during the study period. The majority of TP-IAT recipients were Caucasian (95 %) and women (59 %) with a mean age of 41 years (range 18– 65). The number of TP-IAT procedures per year increased over time with 6 surgeries (15 %) taking place in 2011, 8 (21 %) in 2012, 14 (36 %) in 2013 as well as 11 surgeries (28 %) completed as of November 2014. The prevalence of
J Gastrointest Surg Table 1 ISGPS consensus definition for clinical grading of DGE following TP-IAT
DGE grade
NGT requirement
Inability to tolerate solid oral intake by POD
Vomiting/gastric distension
Use of prokinetics
A B C
4–7 days or reinsertion >POD 3 8–14 days or reinsertion >POD 7 >14 days or reinsertion >POD 14
7 14 21
± + +
± + +
pre-operative bloating or constipation was 26 %, and 5 (13 %) patients had a documented history of gastroparesis prior to surgery. Nine patients (23 %) were on TPN prior to the surgery. There were 31 (80 %) patients with prior abdominal surgeries among whom 26 patients (67 %) reported a history of open or laparoscopic cholecystectomy and 6 patients (15 %) reported prior pancreatic surgery. Fourteen patients (36 %) underwent prior celiac plexus blockade. Of the 39 TP-IAT procedures, 25 (64 %) were open and 14 (36 %) were laparoscopic. Of the open surgeries, two (5 %) were originally laparoscopic but were later converted. The mean duration of surgery was 9.7 h (range 6–14 h). All Table 2 Postoperative clinical course following TP-IAT
patients were observed in the intensive care unit immediately following surgery with the median length of intensive care being 3 days (range 2–13 days). Of these 39 patients, four (10 %) required re-admission to the intensive care unit. Patients undergoing laparoscopic TP-IAT had significantly shorter hospitalizations with the median length of stay for laparoscopic TP-IAT being 10.5 days compared to 14 days for open surgery (p=0.017). The median opioid dose prior to surgery in the laparoscopic and open groups was not significantly different (50.25 oral morphine equivalents (OMEs) vs. 122.2 OMEs, p>0.05). However, the median postoperative opioid dose at discharge
Time to nasogastric tube removal
POD 0 POD 1 POD 2 POD 3 POD 4 POD 5 Need for nasogastric tube re-insertion No Yes Days of total nasogastric tube requirement 0 days 1 day 2–3 days 4–5 days 6–7 days >7 days Time to initiation of liquid diet ≤3 days 4–6 days ≥7 days Time to initiation of full oral diet 14 days Need for prokinetic medications No Yes
Number of patients
p value
Open TPIAT
Laparoscopic TPIAT
1 (4.0 %) 7 (28.0 %) 7 (28.0 %) 8 (32.0 %) 2 (8.0 %) 0 (0.0 %)
2 (14.3 %) 3 (21.4 %) 6 (42.9 %) 2 (14.3 %) 0 (0.0 %) 1 (7.1 %)
p=0.31
20 (80.0 %) 5 (20.0 %)
13 (92.9 %) 1 (7.1 %)
p=0.39
1 (4.0 %) 6 (24.0 %) 11 (44.0 %) 4 (16.0 %) 1 (4.0 %) 2 (8.0 %)
2 (14.3 %) 2 (14.3 %) 9 (64.3 %) 1 (7.1 %) 0 (0.0 %) 0 (0.0 %)
p=0.47
11 (44.0 %) 12 (48.0 %) 2 (8.0 %)
9 (64.3 %) 5 (35.7 %) 0 (0.0 %)
p=0.34
16 (64.0 %) 6 (24.0 %) 3 (12.0 %)
11 (78.6 %) 3 (21.4 %) 0 (0.0 %)
p=0.37
12 (48.0 %) 13 (52.0 %)
13 (92.9 %) 1 (7.1 %)
p=0.01
J Gastrointest Surg
was significantly higher in the open group compared to the laparoscopic group (332.4 OMEs vs. 123.5 OMEs, p=0.04). Postoperative course and DGE During the postoperative course (Table 2), nasogastric tubes were required for three or fewer days in the majority of patients (92 %) with the median duration of initial nasogastric tube use being 2 days (range 0–5 days). Only two patients in the open TP-IAT group and one patient in the laparoscopic group needed the nasogastric tube for more than 3 days. Six patients (15 %) required nasogastric tube re-insertion with a median total duration of nasogastric tube requirement being 2 days (range 0–16 days). More than half the patients (51 %) were able to tolerate a liquid diet by postoperative day 3 and solid food by postoperative day 7 (69 %). The median time to solid food intake was 6 days (range 4–16 days) with all patients in the laparoscopic group able to resume a full diet by 14 days. Prokinetic medications were required for dysmotility symptoms in 14 patients (36 %) with metoclopramide used in 9 patients (23 %). The need for prokinetic drugs was significantly different between the two groups (p=0.01), with only one patient in the laparoscopic group needing prokinetic drugs. Erythromycin or azithromycin was used in a minority of patients (3 %), and the combination of metoclopramide and erythromycin or azithromycin was used in 8 % (Table 3). The mean time to prokinetic drug initiation was 6.5 days (range 1– 16 days), and seven patients (18 %) needed prokinetic drugs at
Table 3
Evaluation and management of DGE
Use of prokinetic medications Metoclopramide only
Number of patients (%)a 9 (23.1 %)
Erythromycin/azithromycin only 1 (2.6 %) Metoclopramide+erythromycin/azithromycin 3 (7.7 %) Timing of prokinetic medications ≤3 days 2 (5.1 %) 4–6 days 5 (12.8 %) ≥7 days 7 (17.9 %) Discharged on prokinetics No 31 (79.5 %) Yes 7 (17.9 %) Method of dysmotility evaluation CT 14 (35.9 %) Upper GI series 9 (23.1 %) Upper GI endoscopy 3 (7.7 %) Initiation of nutritional support Tube feeds 6 (15.4 %) TPN 5 (12.8 %) Tube feeds+TPN 2 (5.1 %) a
Percentage of total TP-IAT patients
the time of discharge. The main modalities used to evaluate persistent dysmotility symptoms was CT scan followed by upper GI series and upper endoscopy in 36, 23, and 8 % of patients, respectively. Tube feeds were needed to be initiated in six patients (15 %) and total parenteral nutrition in five patients (13 %) and the combination of the two in two patients (5 %). Pain control was a major problem following surgery with all patients receiving an opioid PCA and epidural in the postoperative period. Nine patients (23 %) required ketamine in the postoperative period for uncontrolled pain. Patients remained on intravenous analgesics for a median duration of 7 days (range 3–35 days) after which they were switched to oral agents. The ISGPS clinical grading system was used to classify DGE following TP-IAT surgery (Fig. 1, Table 4). No DGE was seen in 25 patients (64.1 %). Grade A DGE was encountered in 10 patients (25.6 %), grade B in 2 patients (5.1 %), and grade C DGE in 2 patients (5.1 %). The overall prevalence of DGE in chronic pancreatitis patients following TP-IAT was 35.9 %, with the prevalence in the open group being 44 % compared to 21 % in the laparoscopic TP-IAT group (p=0.19). Among the open TP-IAT group, two patients (5 %) developed sustained nausea and vomiting and needed to be maintained nil per os (NPO) after postoperative day 7. The addition of these patients to the analysis as a possible delayed DGE increased the statistical significance with a trend favoring lower dysmotility in the laparoscopic TP-IAT group (p=0.09). Patients undergoing open TP-IAT had threefold higher odds (OR 2.88, 95 % CI 0.64–12.93, p=0.17) for the presence of DGE compared to patients undergoing laparoscopic TP-IAT (Fig. 1); however, this was not statistically significant. There was no association of DGE with age, gender, type of pancreatitis (calcific vs. non-calcific/minimal change pancreatitis), history of prior abdominal or pancreatic surgery, last known pre-operative opioid dose, or degree of fibrosis (all p>0.05). Patients with DGE had 5.7-fold higher odds of having a hospital length of stay (LOS) greater than 14 days (OR 5.70, 95 % CI 1.37–23.76, p=0.02).
Fig. 1 DGE grading by type of surgery
J Gastrointest Surg Table 4 DGE following TP-IAT ISGPS DGE grade
Number of patients Open TPIAT
Total number of patients (%) Laparoscopic TPIAT
No DGE
14 (56.0 %)
11 (78.6 %)
25 (64.1 %)
A B C
7 (28.0 %) 2 (8.0 %) 2 (8.0 %) 25 (100 %)
3 (21.4 %) 0 (0 %) 0 (0 %) 14 (100 %)
10 (25.6 %) 2 (5.1 %) 2 (5.1 %) 39 (100 %)
Discussion Our study provides the first estimate of the prevalence of DGE following the TP-IAT procedure, which was 35.9 % and thereby highlights an important and potentially underrecognized clinical problem. To our knowledge, we are the first authors to characterize the burden of DGE after the TP-IAT procedure using open versus laparoscopic approaches. DGE was found to significantly increase hospital LOS. However, patients undergoing the laparoscopic approach demonstrated a shorter LOS compared to patients in the open TP-IAT group and a lower need for prokinetics during the postoperative course. The presence of gastrointestinal dysmotility or DGE is well described following pancreaticoduodenectomy, although widely varying clinical definitions make an exact estimation difficult.16,17 For studies using the ISGPS definition, the prevalence of DGE following pancreaticoduodenectomy ranges between 12 and 45 %.6,18–22 Using the same ISGPS definition, the overall prevalence of DGE in our study following TP-IAT was similar at 35.9 %. The prevalence of DGE in chronic pancreatitis patients following laparoscopic TP-IAT in our study was 21 % which is slightly higher compared to the rates following laparoscopic pancreaticoduodenectomy in other studies which range between 11 and 14 %.23,24 However, prior studies included patients undergoing surgery for all indications whereas our study focuses only on patients undergoing TP-IAT surgery for chronic pancreatitis. It is possible that chronic pancreatitis patients have a higher baseline rate of gastrointestinal dysmotility which is confounded by longstanding abdominal pain and, in most cases, resultant opioid analgesia which causes delay in gastric emptying and intestinal transit. The role of the pancreas and resultant changes during the course of chronic pancreatitis in the regulation of gut motility is yet to be characterized. Although the difference in the incidence of DGE between the open and laparoscopic groups did not meet statistical significance, no patients in the laparoscopic group had severe DGE (grade B or C DGE) and the need for prokinetics in the postoperative course was significantly lower compared to the open group. Previous studies comparing laparoscopic
and open pancreaticoduodenectomies have demonstrated lower length of hospital stay and intraoperative bleeding but no difference in complications, mortality, or delayed gastric emptying.24–26 In our study, we also encountered a group of patients who, despite being initiated on a diet and not meeting classic criteria for DGE, later developed vomiting and required prokinetics but did not undergo reinsertion of a nasogastric tube. This group of patients required additional evaluation as did patients with DGE, and they may suggest a form of DGE which could be added to future modifications of the ISGPS definition. Our study has several strengths. The use of inpatient data in this study ensured complete data collection and obviates concerns regarding missing data. Our study also used the ISGPS consensus definition to grade DGE, thereby applying a validated clinical grading system in the TP-IAT population. Our study also has some limitations. Our study was limited by the relatively small sample size of patients undergoing the TP-IAT procedure and the retrospective nature of the study, yet we still detected significant differences. In addition, although an institutional algorithm was used for nasogastric tube management and prokinetic administration, it is possible that their initiation and discontinuation was influenced by the preferences of the surgical team on service at that time, thereby raising the possibility of selection bias. However, this would be in both directions and would be unlikely to significantly impact the results of this study. Since this was a retrospective study, there is also the possibility of selection bias during the original decision for TP-IAT approach since it is possible that patients with lower pre-operative risk of dysmotility such as ones without prior abdominal or pancreatic surgeries were chosen to undergo laparoscopic TPIAT. However, we did not find any significant association between prior surgeries and DGE and therefore, this is less likely to significantly impact the results of this study. Further studies are needed to estimate the burden of longterm dysmotility following TP-IAT and its association with DGE. With increasing numbers of laparoscopic TP-IATs, studies are also needed to assess for differences in DGE and long-term dysmotility as well as other clinically significant outcomes between open and laparoscopic TP-IAT in a larger cohort of patients.
J Gastrointest Surg
Conclusion In conclusion, DGE following TP-IAT is common and significantly prolongs hospital stays. A laparoscopic TP-IAT approach lowers length of stay and need for prokinetic medications postoperatively. Acknowledgments GKJ, EMS, VKS, and PJP were involved in the concept and design of this study. GKJ reviewed patient records, extracted data, performed statistical analysis, and wrote the first draft of the manuscript. All authors critically appraised and revised the manuscript. All authors gave final approval for submission of the manuscript. Funding No funding declared. Conflict of Interest No competing interests declared.
References 1.
2.
3.
4. 5. 6.
7.
8.
9.
10.
Beilman G. Pancreatitis: Is TPIAT the answer for treatment of chronic pancreatitis? Nature Reviews Gastroenterology & Hepatology 2014. Morgan K, Owczarski S, Borckardt J, Madan A, Nishimura M, Adams D. Pain control and quality of life after pancreatectomy with islet autotransplantation for chronic pancreatitis. J Gastrointest Surg 2012;16:129-33; discussion 33–4. doi 10. 1007/s11605-011-1744-y. Wilson GC, Ahmad SA, Schauer DP, Eckman MH, Abbott DE. Cost-Effectiveness of Total Pancreatectomy and Islet Cell Autotransplantation for the Treatment of Minimal Change Chronic Pancreatitis. J Gastrointest Surg 2014;6:6. Emick DM, Riall TS, Cameron JL, et al. Hospital readmission after pancreaticoduodenectomy. J Gastrointest Surg 2006;10:1243-52. Fisher AV, Sutton JM, Wilson GC, et al. High readmission rates after surgery for chronic pancreatitis. Surgery;156:787-96. Liu QY, Li L, Xia HT, Zhang WZ, Cai SW, Lu SC. Risk factors of delayed gastric emptying following pancreaticoduodenectomy. ANZ J Surg 2014;13:12850. Paraskevas K, Avgerinos C, Manes C, Lytras D, Dervenis C. Delayed gastric emptying is associated with pylorus-preserving but not classical Whipple pancreaticoduodenectomy: a review of the literature and critical reappraisal of the implicated pathomechanism. World J Gastroenterol 2006;12:5951-8. Holzer HH, Turkelson CM, Solomon TE, Raybould HE. Intestinal lipid inhibits gastric emptying via CCK and a vagal capsaicinsensitive afferent pathway in rats. Am J Physiol 1994;267:G625-9. Suzuki H, Mochiki E, Haga N, Shimura T, Itoh Z, Kuwano H. Effect of duodenectomy on gastric motility and gastric hormones in dogs. Ann Surg 2001;233:353-9. Yeo C, Barry M, Sauter P, et al. Erythromycin accelerates gastric emptying after pancreaticoduodenectomy. A prospective, randomized, placebo-controlled trial. Ann Surg 1993;218:229-37; discussion 37-8.
11.
Kim D, Hindenburg A, Sharma S, et al. Is Pylorospasm a Cause of Delayed Gastric Emptying After Pylorus-Preserving Pancreaticoduodenectomy? Ann Surg Oncol 2005;12:222-7. 12. Ueno TTA, Hamanaka Y, Tsurumi M, Suzuki T. . A proposal mechanism of early delayed gastric emptying after pylorus preserving pancreatoduodenectomy. Hepato-gastroenterology 1995;42:269-74. 13. Walsh TN, Rode J, Theis BA, Russell RC. Minimal change chronic pancreatitis. Gut 1992;33:1566-71. 14. Wallace MB, Hawes RH, Durkalski V, et al. The reliability of EUS for the diagnosis of chronic pancreatitis: interobserver agreement among experienced endosonographers. Gastrointest Endosc 2001;53:294-9. 15. Kloppel G, Maillet B. Pseudocysts in chronic pancreatitis: a morphological analysis of 57 resection specimens and 9 autopsy pancreata. Pancreas 1991;6:266-74. 16. Wente M, Bassi C, Dervenis C, et al. Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS). Surgery 2007;142:761-8. 17. Traverso LW, Hashimoto Y. Delayed gastric emptying: the state of the highest level of evidence. J Hepatobiliary Pancreat Surg 2008;15:262-9. 18. Hashimoto Y, Traverso LW. Incidence of pancreatic anastomotic failure and delayed gastric emptying after pancreatoduodenectomy in 507 consecutive patients: use of a web-based calculator to improve homogeneity of definition. Surgery 2010;147:503-15. 19. Malleo G, Crippa S, Butturini G, et al. Delayed gastric emptying after pylorus-preserving pancreaticoduodenectomy: validation of International Study Group of Pancreatic Surgery classification and analysis of risk factors. HPB 2010;12:610-8. 20. Akizuki E, Kimura Y, Nobuoka T, et al. Reconsideration of postoperative oral intake tolerance after pancreaticoduodenectomy: prospective consecutive analysis of delayed gastric emptying according to the ISGPS definition and the amount of dietary intake. Ann Surg 2009;249:986-94. 21. Welsch T, Borm M, Degrate L, Hinz U, Buchler MW, Wente MN. Evaluation of the International Study Group of Pancreatic Surgery definition of delayed gastric emptying after pancreatoduodenectomy in a high-volume centre. Br J Surg 2010;97:1043-50. 22. Park JS, Hwang HK, Kim JK, et al. Clinical validation and risk factors for delayed gastric emptying based on the International Study Group of Pancreatic Surgery (ISGPS) Classification. Surgery 2009;146:882-7. 23. Kendrick ML, Cusati D. Total laparoscopic pancreaticoduodenectomy: feasibility and outcome in an early experience. Arch Surg 2010;145: 19-23. 24. Asbun HJ, Stauffer JA. Laparoscopic vs open pancreaticoduodenectomy: overall outcomes and severity of complications using the Accordion Severity Grading System. J Am Coll Surg 2012;215:810-9. 25. Croome KP, Farnell MB, Que FG, et al. Pancreaticoduodenectomy with Major Vascular Resection: a Comparison of Laparoscopic Versus Open Approaches. J Gastrointest Surg 2014;2:2. 26. Li YB, Wang X, Wang MJ, Yang ZG, Peng B. [Delayed gastric emptying after laparoscopic versus open pancreaticoduodenectomy: a comparative study]. Zhonghua Wai Ke Za Zhi 2013;51:304-7.
