Delayed gastric emptying after pancreaticoduodenectomy in

diabetes mellitus

Takatsugu Oida,* Kenji Mimatsu,* Hisao Kano,* Atsushi Kawasaki,* Youichi Kuboi,* Nobutada Fukino,*Kazutoshi Kida* and Sadao Amano†*Department of Surgery, Social Insurance Yokohama Central Hospital, Naka-ku, Yokohama, Japan and†Department of Surgery, Nihon University School of Medicine, Itabashi-ku, Tokyo, Japan

Key words

delayed gastric emptying, diabetes mellitus, modifiedsubtotal–stomach-preserving pancreaticoduodenectomy,pancreaticoduodenectomy, pylorus-preservingpancreaticoduodenectomy.

Correspondence

Dr Takatsugu Oida, Department of Surgery, SocialInsurance Yokohama Central Hospital, 268Yamashita-cho, Naka-ku, Yokohama 231-8553, Japan.Email: ooida.takatsugu@yokochu.jp

T. Oida MD, PhD; K. Mimatsu MD, PhD; H. Kano MD;A. Kawasaki MD, PhD; Y. Kuboi MD, PhD; N. Fukino

MD; K. Kida MD; S. Amano MD, PhD.

Accepted for publication 15 July 2012.

doi: 10.1111/j.1445-2197.2012.06242.x

Abstract

Introduction: Delayed gastric emptying (DGE) is one of the most troublesomecomplications of pancreaticoduodenectomy (PD). Diabetes mellitus (DM) is one ofthe risk factors for pancreatic cancer. Moreover, several studies have shown thatdiabetic patients tend to have a high incidence of upper gastrointestinal symptoms suchas nausea, vomiting and DGE. Here, we compared the influence of DM on theincidence of DGE after PD.Methods: We retrospectively analysed 67 cases of PD with pancreaticogastrostomy.These patients were categorized into the following two groups: the DM group includedpatients with DM, and the NDM group included patients without DM. The incidenceof DGE was determined and compared between the two groups.Results: In the DM group, 76.5%, 5.9% and 17.6% of the subjects developed classesA, B and C DGE, respectively; the corresponding values in the NDM group were 58%,22%, and 20%. The incidence of DGE did not differ between the two groups(P < 0.2771).Conclusions: DM does not accelerate DGE in patients who have undergone PD.Preoperative DM does not appear to play a key role in post-operative DGE after PD.

Introduction

Pancreaticoduodenectomy (PD) has been performed in patientswith malignant tumours in the peri-ampullary region and the headof the pancreas. Furthermore, pylorus-preserving PD (PPPD)has been reported to yield better results than the Whippleresection with regard to operative mortality and morbidity andpost-operative nutrition.1–3 However, delayed gastric emptying(DGE) is one of the most troublesome complications of thisprocedure.2,4–7 The mechanism of DGE is unclear, and its patho-physiology has not yet been elucidated. Several causes have beensuggested, including local ischaemia of the antrum, the absence ofduodenal hormones, inflammation from pancreaticoenterostomy,oedema from duodenojejunostomy and gastric atony caused byvagotomy.8–12

Diabetes mellitus (DM) is a risk factor for pancreatic cancer.13–15

Moreover, several studies have showed that diabetic patients tend tohave a high incidence of upper gastrointestinal symptoms such asnausea, vomiting and DGE.14–18 Of note, the prevalence of DGE inpatients with type 1 DM is about 50%.16–18

We previously reported that modified subtotal–stomach-preserving PD (MSSPPD) has been performed as an alternative toPPPD for preventing DGE and that the incidence of DGE decreasedafter this procedure.19,20

This is a retrospective study investigating whether DM is associ-ated with a higher incidence of DGE after PD.

Patients and methods

Patients

Between May 1999 and September 2011, 67 patients (51 men and 16women) underwent PD. Their mean age was 65.9 � 5.3 years(range, 54–78 years). Carcinoma of the pancreatic head, lower bileduct and ampulla of Vater was diagnosed in 48, 13 and 6 patients,respectively. PPPD was performed for 25 (37%) patients, andMSSPPD was performed for 42 (63%) patients. Pancreaticogastros-tomy was performed in all patients. Patients were categorized intothe following two groups: the DM group included patients with DM,and the NDM group included patients without DM. The incidence ofDGE after surgery was compared between the two groups.

HEPATOBILIARYANZJSurg.com

© 2012 The AuthorsANZ Journal of Surgery © 2012 Royal Australasian College of Surgeons ANZ J Surg 83 (2013) 973–977

Surgical technique

PancreaticogastrostomyFor all patients in this series, pancreaticogastrostomy was performedas follows. Approximately 3 cm of the pancreatic remnant was freedfrom the retroperitoneal space. A stent was inserted and fixed into thepancreatic duct. Posterior gastrostomy was performed such that theopening was a suitable size for the stump of the pancreatic remnant,anterior gastrostomy was performed subsequently. The pancreaticstump was inserted into the gastric lumen via the posterior gastros-tomy. Pancreaticogastrostomy was performed in the gastric lumenvia the anterior gastrostomy. The stent was brought out through theanterior gastric wall, and the anterior gastrostomy was closed. Thestent was removed more than 2 weeks after the surgery.

Pylorus-preserving PDWhen the hepatogastric ligament and the right part of the gastrocolicligament are resected, the vessel arch around the stomach should bepreserved. The right gastric artery and gastroduodenal artery weredivided and ligated at their origins to preserve the distal arcade. Theduodenum was transected 2–3 cm below the pylorus. Gastrointesti-nal reconstruction after PPPD was performed as follows. Pancreati-cogastrostomy, end-to-side choledochojejunostomy, and end-to-endduodenojejunostomy were performed, in that order. In our recon-struction, the proximal jejunum was brought to the hepatic hilum ina retrocolic position. An end-to-side choledochojejunostomy wasperformed using a single layer of interrupted 4-0 absorbable mono-filament sutures with or without inserting a biliary drainage tube.Finally, end-to-end duodenojejunostomy was performed.

Modified subtotal–stomach-preserving PDAn initial, subtotal–stomach-preserving PD (SSPPD) was per-formed. It involved division of the stomach 2–3 cm proximal to thepyloric ring along with the resection of the entire duodenum distal tothe transection site, resection of the gall bladder and the commonbile duct, and removal of the head of the pancreas. Our modifiedSSPPD technique involved re-resection of the distal part of theremnant stomach preserved after subtotal-stomach resection accord-ing to the angle from the choledochojejunostomy to the gastrojeju-nostomy. Gastrointestinal reconstruction after modified SSPPD wasperformed as follows. Pancreaticogastrostomy, end-to-side choledo-chojejunostomy, and end-to-side gastrojejunostomy were per-formed, in that order. In our reconstruction technique, the proximaljejunum was brought to the hepatic hilum in a retrocolic position. Anend-to-side choledochojejunostomy was then performed, using asingle layer of interrupted 4-0 absorbable monofilament sutures withor without inserting a biliary drainage tube. The proximal stomachwas stretched to the infracolic space. Subsequently, according to theangle from the choledochojejunostomy to the gastrojejunostomy, thedistal stomach was re-resected. The resected part was small andtriangular. Finally, the stomach was anastomosed end-to-side to thejejunum in the infracolic space (Fig. 1).19,20

Definition of DM

Preoperative DM was diagnosed based on documented clinicalhistory or retrospective review of laboratory studies with application

of diagnostic criteria outlined by the American Diabetes Associa-tion.21 The definition and classification of DM have been previouslydescribed.22

Definition of DGE

DGE was defined according to the International Study Group guide-lines (Table 1).23 The pancreas was defined as soft pancreas if theresults of the preoperative pancreatic function tests were withinnormal limits and when the intraoperative assessment revealed non-fibrosed parenchyma with the main pancreatic duct measuring�3 mm in diameter.

Statistical analysis

Univariate analysis was performed using Student’s t-test for con-tinuous variables and Fisher’s exact test and chi-square test forcategorical variables. A P-value of less than 0.05 was consideredstatistically significant.

Results

Table 2 presents preoperative diabetic status. Twelve patients(70.6%) in the DM group received routine treatment with oral anti-diabetic agents, and the remaining five (29.4%) were treated withinsulin. Table 3 summarizes patient characteristics and preoperativevariables. No differences were observed between the two groupswith respect to mean age, sex ratio and preoperative clinical data.

Fig. 1. Our modification of the subtotal–stomach-preserving pancreati-coduodenectomy technique. The stomach was resected along a line2–3 cm proximal to the pyloric ring. The distal stomach was stretchedto the infracolic space. Subsequently, depending on the angle from thecholedochojejunostomy to the gastrojejunostomy, the distal stomachwas re-resected. Gastrointestinal reconstruction after pylorus-preservingpancreaticoduodenectomy was performed as follows. Pancreati-cogastorostomy, end-to-side choledochojejunostomy, and end-to-side gas-trojejunostomy (retrocolic route) were performed, in that order. , resectedarea.

974 Oida et al.

© 2012 The AuthorsANZ Journal of Surgery © 2012 Royal Australasian College of Surgeons

In both groups, the indications for PD were similar. However, preop-erative serum haemoglobin A1c (HbA1c) levels were significantlygreater in the DM group (5.9% � 0.3%) than in the NDM group(5.3% � 0.3%) (P < 0.0001), and preoperative diameter of thepancreatic duct was significantly smaller in the NDM group (4.5 �

1.3 mm) than in the DM group (6.4 � 1.4 mm), (P < 0.0001).Table 4 summarizes intraoperative and post-operative variables. Nodifferences were observed between the two groups with respect tothe mean operative time and blood loss. No deaths occurred in eithergroup. The mean time during which nasogastric suction wasrequired was 2.6 � 2.6 days in the DM group and 3.2 � 2.3 days inthe NDM group; thus, no significant difference was observedbetween the two groups (P < 0.3540). The mean time before theinitiation of a diet after surgery was 5.5 � 2.0 days in the DM groupand 5.7 � 1.5 days in the NDM group; thus, no significant differencewas observed between the two groups (P < 0.59330). Moreover, themean time after which oral ingestion of solid food could be safelyresumed was 11.9 � 4.4 days in the DM group and 12.7 � 3.9 daysin the NDM group; thus, no significant difference was observedbetween the two groups (P < 0.4866). In terms of DGE, 76.5%, 5.9%and 17.6% of the cases were classified as A, B and C, respectively,in the DM group. The corresponding values for the NDM groupwere 58%, 22% and 20%. There were no significant differencesbetween the two groups with respect to DGE (P < 0.2771). Withregard to post-operative complications in the DM group, one (5.9%)patient had pneumonia and two (11.8%) had wound infections. Inthe NDM group, two (4.0%) patients had pneumonia, three (6.0%)had wound infections and one (2.0%) had cholangitis. There were nosignificant differences between the two groups with respect to post-operative complications. Pancreatic leakage was not observed ineither group. The post-operative mean serum albumin level was 3.75� 0.13 g/mL in the DM group and 3.82 � 0.2 g/mL in the NDMgroup; thus, no significant difference was observed between the twogroups. The duration of hospitalization after surgery was 26.2 � 7.6days in the DM group and 26.3 � 4.4 days in the NDM group; thus,no significant difference was observed between the two groups.

Discussion

PPPD has been reported to yield better results than the Whippleresection with regard to operative mortality and morbidity and post-operative nutrition.1–3 However, DGE is frequently observed afterPPPD. The incidence of DGE after PPPD ranges between 25% and70%.2,4–7,24–31 Several causes of DGE have been suggested, includinglocal ischaemia of the antrum, absence of duodenal hormones,inflammation caused by pancreaticoenterostomy, oedema caused byduodenojejunostomy and gastric atony caused by vagotomy.8–12

However, the mechanism of DGE remains unclear, and its patho-physiology has not yet been elucidated. In contrast, several studieshave shown that diabetic patients tend to have a high incidence ofupper gastrointestinal symptoms such as nausea, vomiting andDGE.16–18,32,33 Notably, the prevalence of DGE in patients with DM isabout 50%.16–18 Autonomic neuropathy, especially of the vagalnerves, is regarded as one of the major factors in the pathogenesis ofabnormal gastrointestinal motility in patients with DM.16,34 Previousmanometric studies identified postprandial antral hypomotility,pyloric spasm and intestinal dysmotility in patients with DM.35–37

Therefore, we studied the relationship between DM and DGE in theearly post-operative period after PD with PG.

In general, pancreatic tissue in patients with DM is relatively hardand fibrotic compared with pancreatic tissue in patients without DM.In addition, cancer involvement and inflammation cause pancreaticduct dilatation. In our study, the pancreas was significantly harder inthe DM group than in the NDM group (P < 0.0068); however, therewere no significant differences between the groups with regard topost-operative complications such as pneumonia, wound infectionsand cholangitis. Moreover, pancreatic leakage was not observed ineither group.

In addition, the mean time required for nasogastric suction, themean time before the initiation of diet after surgery, and the meantime after which oral ingestion of solid food could be safely resumeddid not differ between the two groups (P < 0.3540, P < 0.59330 andP < 0.4866, respectively); moreover, there were no significant dif-ferences between the two groups with respect to the DGE grade(P < 0.2771). Chu et al.38 reported that DM and non-DM patientshave similar frequencies of DGE, wound infection, intra-abdominalabscesses, and cardiovascular and pulmonary complications, aswell as length of hospitalization and mortality. Thus, Chu et al.38

concluded that patients with DM were not at increased risk fordevelopment of DGE (DM patients, 38.5%; non-DM patients,41.5%; P < 0.6).

Previously, we reported that MSSPPD has been performed as analternative to PPPD for preventing DGE and that the incidenceof DGE decreased after this procedure.19,20 Our procedure, i.e. a

Table 1 International Study Group of Pancreatic Surgery definition of delayed gastric emptying after pancreatic surgery24

DGE grade Nasogastic tube required Unable to tolerate solid oral intake by POD Vomiting/gastric distension Use of prokinetics

A 4–7 days or reinsertion > POD 3 7 � �B 8–14 days or reinsertion > POD 7 14 + +C >14 days or reinsertion > POD 14 21 + +

DGE, delayed gastric emptying; POD, postoperative day.

Table 2 Preoperative diabetic status

DM group (n = 17) NDM group (n = 50)

Non-treatment 0 (0%) 50 (100%)Treated with OAD 12 (70.6%) 0 (0%)Treated with insulin 5 (29.4%) 0 (0%)

DM, diabetes mellitus; NDM, non-diabetes mellitus; OAD, oral anti-diabeticagents.

DGE after PD in DM patients 975

© 2012 The AuthorsANZ Journal of Surgery © 2012 Royal Australasian College of Surgeons

combination of MSSPPD with vertical stomach reconstruction,aimed to ensure a straight dietary route to the jejunum, therebyenabling gastric contents to easily reach the jejunum and preventingDGE. In the present study, PPPD was performed for 23.5% of thepatients and MSSPPD was performed for 76.5% of the patients inthe DM group, and corresponding values were 42% and 58% in theNDM group; however, no significant difference was observedbetween the two groups (P < 0.1737). Although our sample size ofDM group is small, we think that MSSPPD reconstruction maycontribute to prevention of DGE in patients with DM after surgery.

Among non-surgical DM patients, symptom severity of gas-troparesis correlates poorly with objective measures of DGEseverity;39–41 however, PD is an invasive surgical procedure andan adequate reconstruction procedure is required for preventingpost-operative complications. Schvarcz et al.42 reported that hyper-glycaemia induced in patients with or without DM has the rapidphysiologic effect on reducing gastric emptying time. Thus, glycae-mic control is a necessity of perioperative management.

In conclusion, DM does not accelerate DGE in patients who haveundergone PD. Satisfactory glycemic control is important, and preop-erative DM does not appear to play a key role in post-operative DGEafter PD. However, our study has the limitations of a retrospectivestudy, an adequate reconstruction procedure is important for prevent-ing DGE. Further investigations focusing on surgical reconstructivemethods should be undertaken using randomized controlled study.

References1. Traverso LW, Longmire WP Jr. Preservation of the pylorus in pancrea-

ticoduodenectomy a follow-up evaluation. Ann. Surg. 1980; 192: 306–10.

2. Klinkenbijl JHG, van der Shelling GP, Hop WCT, van Pel R, BruiningHA, Jeekel J. The advantage of pylorus-preserving pancreatoduodenec-tomy in malignant disease of the pancreas and periampullary region.Ann. Surg. 1992; 216: 142–5.

3. Zerbi A, Balzano G, Patuzzo R, Calori G, Braga M, Dicarlo V. Com-parison between pylorus-preserving and Whipple pancreatoduodenec-tomy. Br. J. Surg. 1995; 82: 975–9.

Table 3 Characteristics of the patients

DM group (n = 17) NDM group (n = 50) P-value

Age 67.7 � 4.1 65.4 � 5.6 0.1279Sex ratio (man : woman) 15:2 36:14 0.3225DiagnosisPancreatic cancer 13 (76.5%) 35 (70%) 0.6091Low bile duct cancer 3 (17.6%) 10 (20%) 0.8322Vater carcinoma 1 (5.9%) 5 (10%) 0.6075Initial diseaseHypertension 4 (23.5%) 10 (20%) 0.7572Preoperative pancreatic duct diameter (mm) 6.4 � 1.4 4.5 � 1.3 0.0001†Texture of pancreas 0.0068†

Soft 0 (0%) 16 (32%)Hard 17 (100%) 34 (68%)Preoperative serum albumin (g/mL) 3.81 � 0.2 3.84 � 0.2 0.6529

PTBD 0.2211Yes 13 (76.5%) 30 (60%)No 4 (23.5%) 20 (40%)

Preoperative HbA1c% 5.9 � 0.3 5.3 � 0.3 0.0001†

†Significant difference. DM, diabetes mellitus; NDM, non-diabetes mellitus; PTBD, percutaneous transhepatic bile duct drainage.

Table 4 Outcomes

DM group (n = 17) NDM group (n = 50) P-value

Operative procedure 0.1737PPPD 4 (23.5%) 21 (42%)MSSPPD 13 (76.5%) 29 (58%)

Operative time (min) 430 � 51 434 � 41 0.7351Blood loss (mL) 547 � 176 584 � 183 0.4769N-G removal 2.6 � 2.6 3.2 � 2.3 0.3540Initial diet (days) 5.5 � 2.0 5.7 � 1.5 0.5933Days after which solid diet could be resumed (d) 11.9 � 4.4 12.7 � 3.9 0.4866Delayed gastric emptying 0.2771

A 13 (76.5%) 29 (58%) 0.1737B 1 (5.9%) 11 (22%) 0.1343C 3 (17.6%) 10 (20%) 0.8322

ComplicationsPneumonia 1 (5.9%) 2 (4.0%) 1.00Wound infection 2 (11.8%) 3 (6.0%) 0.5946Cholangitis 0 (0%) 1 (2.0%) 1.00Post-operative serum albumin (g/mL) 3.75 � 0.1 3.82 � 0.2 0.0955Duration of hospitalization (days) 26.2 � 7.6 26.3 � 4.4 0.9346

DM, diabetes mellitus; MSSPPD, modified subtotal–stomach-preserving pancreaticoduodenectomy; N-G, nasogastric tube ; NDM, non-diabetes mellitus; PPPD,pylorus-preserving pancreaticoduodenectomy.

976 Oida et al.

© 2012 The AuthorsANZ Journal of Surgery © 2012 Royal Australasian College of Surgeons

4. Warshaw AL, Torchiana DL. Delayed gastric emptying after pylorus-preserving pancreaticoduodenectomy. Surg. Gynecol. Obstet. 1985; 160:1–4.

5. Roder JD, Stein HJ, Huttl W, Siewert JR. Pylorus-preserving versusstandard pancreaticoduodenectomy: an analysis of 110 pancreatic andperiampullary carcinomas. Br. J. Surg. 1992; 79: 152–5.

6. Williamson RCN, Bliouras N, Cooper MJ, Davies ER. Gastric emptyingand enterogastric reflux after consecutive and conventional pancrea-toduodenectomy. Surgery 1993; 114: 82–6.

7. Krejs GJ. Pancreatic cancer. Epidemiology 2010; 28: 355–8.8. Park YC, Kim SW, Jang JY, Ahn YJ, Park YH. Factors influencing

delayed gastric emptying after pylorus-preserving pancreatoduodenec-tomy. J. Am. Coll. Surg. 2003; 196: 859–65.

9. Kim DK, Hindenburg AA, Sharma SK et al. Is pylorospasm a cause ofdelayed gastric emptying after pylorus-preserving pancreaticoduodenec-tomy? Ann. Surg. Oncol. 2005; 12: 222–7.

10. Tanaka A, Ueno T, Oka M, Suzuki T. Effect of denervation of thepylorus and transection of the duodenum on acetaminophen absorptionin rats; possible mechanism for early delayed gastric emptying afterpylorus preserving pancreatoduodenectomy. Tohoku J. Exp. Med. 2000;192: 239–47.

11. Riediger H, Makowiec F, Schareck WD, Hopt UT, Adam U. Delayedgastric emptying after pylorus-preserving pancreatoduodenectomy isstrongly related to other postoperative complications. J. Gastrointest.Surg. 2003; 7: 758–65.

12. Horstmann O, Markus PM, Ghadimi MB, Becker H. Pylorus preserva-tion has no impact on delayed gastric emptying after pancreatic headresection. Pancreas 2004; 28: 69–74.

13. Jemal A, Siegel R, Ward E, Murray T, Thun MJ. Cancer statistics, 2007.CA Cancer J. Clin. 2007; 57: 43–66.

14. Wang F, Herrinton M, Larsson J, Permert J. The relationship betweendiabetes and pancreatic cancer. Mol. Cancer 2003; 2: 4.

15. Noy A, Bilezikian JP. Clinical review 63: diabetes and pancreaticcancer: clues to the early diagnosis of pancreatic malignancy. J. Clin.Endocrinol. Metab. 1994; 79: 1223–31.

16. Kassander P. Asymptomatic gastric retention in diabetics (gastroparesisdiabeticorum). Ann. Int. Med. 1958; 48: 797–812.

17. Horowitz M, Harding M, Maddox A et al. Gastric and esophageal emp-tying in insulin-dependent diabetes mellitus. J. Gastroenterol. Hepatol.1986; 1: 97–113.

18. Keshavarzian A, Iber FL, Vaeth J. Gastric emptying in patients withinsulin-requiring diabetes mellitus. Am. J. Gastroenterol. 1986; 82:29–35.

19. Oida T, Mimatsu K, Kawasaki A et al. Vertical stomach reconstructionwith pancreaticogastrostomy after modified subtotal-stomach-preserving pancreaticoduodenectomy for preventing delayed gastricemptying. Hepatogastroenterology 2009; 56: 565–7.

20. Oida T, Mimatsu K, Kano H et al. Preventing delayed gastric emptyingin pancreaticogastrostomy by a modified subtotal-stomach-preservingpancreaticoduodenectomy: Oida modification. Hepatogastroenterology2011; 58: 1384–8.

21. American Diabetes Association. Standards of medical care in diabetes –2009. Diabetes Care 2009; 32: S13–61.

22. Chu CK, Mazo AE, Goodaman M et al. Preoperative diabetes mellitusand long-term survival after resection of pancreatic adenocarcinoma.Ann. Surg. Oncol. 2010; 17: 502–13.

23. Wente MN, 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.

24. Berge Henegouwen MI, van Gulik TM, DeWit LT et al. Delayed gastricemptying after standerd pancreaticoduodenectomy versus pylorus-preserving pancreaticoduodenectomy: an analysis of 200 consecutivepatients. J. Am. Coll. Surg. 1997; 185: 373–9.

25. Kobayashi I, Miyachi M, Kanai M et al. Different gastric emptying ofsolid and liquid meal after pylorus-preserving pancreatoduodenectomy.Br. J. Surg. 1998; 85: 927–30.

26. Geer RJ, Richards WO, O’Dorisio TM et al. Efficacy of octreotideacetate in treatment of severe postgastrectomy dumpling syndrome. Ann.Surg. 1990; 212: 678–87.

27. Ingram DM, Sheiner HJ. Postoperative gastric emptying. Br. J. Surg.1981; 68: 572–6.

28. Hocking MP, Harrison WD, Sninsky CA. Gastric dysrhythmiasfollowing pylorus-preserving pancreaticoduodenectomy: possiblemechanism for early delayed gastric emptying. Dig. Dis. Sci. 1990; 35:1226–30.

29. Takada T, Yasuda H, Shikata J, Watanabe S, Shiratori K, Takeuchi T.Postprandial plasma gastric and secretin concentrations after a pancrea-toduodenectomy: a comparison between a pylorous-preserving pancrea-toduodenectomy and the Whipple procedure. Ann. Surg. 1989; 210:47–51.

30. Hunt DR, Mclean R. Pylorus-preserving pancreatectomy: functionalresults. Br. J. Surg. 1987; 76: 173–6.

31. Yeo CJ, Barry MK, Sautr PK et al. Erythromycin accelerates gastricplacebo-controlled trial. Ann. Surg. 1993; 218: 229–37.

32. Wegener M, Börsch G, Schaffstein J, Leverkus F. Gastrointesitinaltransit disorders in patients with insulin-treated diabetes mellitus. Dig.Dis. Sci. 1980; 8: 23–6.

33. Palmer DW, Soergel KH, Kalbfleish JH, Wood CM. Gastric emptying inpatients with diabetes mellitus. Gastroenterology 1984; 86: 485–94.

34. Rundles RW. Diabetic neuropathy: general review with reports of 125cases. Medicine 1945; 24: 111–60.

35. Cmilleri M, Malageleda JR. Abnormal intestinal motility in diabeteswith gastroparesis syndrome. Eur. J. Clin. Invest. 1984; 14: 420–7.

36. Jebbink HJA, Bravenboer B, Akkermans LMA, van Berge-HenegouwenGP, Smout AJPM. Relationship between dyspeptic symptoms and gas-trointestinal motility in patients with type I (insulin-dependent) diabetesmellitus. Diabetologia 1993; 36: 948–54.

37. Mearin F, Camilleri M, Malageleda JR. Pyloric dysfunction in diabeticswith recurrent nausea and vomiting. Gastroenterology 1986; 90: 1919–25.

38. Chu CK, Mazo AE, Sarmiento JM et al. Impact of diabetes mellitus onperioperative outcomes after resection for pancreatic adenocarsinoma. J.Am. Coll. Surg. 2010; 210: 463–73.

39. Jones KL, Russo A, Stevens JE et al. Predictors of delayed gastricemptying in diabetes. Diabetes Care 2001; 24: 1264–9.

40. Samsom M, Vermeijden JR, Smout AJ et al. Prevalence of delayedgastric emptying in diabetic patients and relationship to dyspeptic symp-toms: a prospective study in unselected diabetic patients. Diabetes Care2003; 26: 3116–22.

41. Horowitz M, Maddox AF, Wishart JM et al. Relationships betweenoesophageal transit and solid and liquid gastric emptying in diabetesmellitus. Eur. J. Nucl. Med. 1991; 18: 229–34.

42. Schvarcz E, Palmer M, Aman J et al. Physiological hyperglycemia slowsgastric emptying in normal subjects and patients with insulin-dependentdiabetes mellitus. Gastroenterology 1997; 113: 60–6.

DGE after PD in DM patients 977

© 2012 The AuthorsANZ Journal of Surgery © 2012 Royal Australasian College of Surgeons