- Email: [email protected]
Stenting is unnecessary in duct-to-mucosa pancreaticojejunostomy even in the normal pancreas
Original Paper Received: June 18, 2001 Accepted: October 17, 2001
Pancreatology 2002;2:116–121
Stenting Is Unnecessary in Duct-to-Mucosa Pancreaticojejunostomy Even in the Normal Pancreas Toshihide Imaizumi Nobuhiko Harada Takashi Hatori Akira Fukuda Ken Takasaki Tokyo Women’s Medical University, Tokyo, Japan
Key Words Duct-to-mucosa pancreaticojejunostomy W Nonstented pancreaticojejunostomy W Reconstruction of the normal pancreas
Abstract Background: There is a high risk of anastomotic leakage after pancreaticojejunostomy after pancreaticoduodenectomy (PD) in patients with a normal pancreas because of the high degree of exocrine function. These PD are therefore generally performed using a stenting tube (stented method). In recent years, we have performed pancreaticojejunostomy with duct-to-mucosa anastomosis without a stenting tube (nonstented method) and obtained good results. Methods: The point of this technique is to preserve adequate patency of the pancreatic duct by carefully picking up the pancreatic duct wall with a fine atraumatic needle and monofilament thread. The results of end-to-side pancreaticojejunostomy of the normal pancreas were compared between the nonstented method (n = 109) and the stented method (n = 39). Results: There were no differences in background characteristics between the groups, including age, gender and disease. The mean duration to complete pancreaticojejunostomy was 26.6 min in the nonstented group and 29.2 min in the stented group. The mean durations
ABC
© 2002 S. Karger AG, Basel and IAP 1424–3903/02/0022–0116$18.50/0
Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com
Accessible online at: www.karger.com/journals/pan
of surgical procedure and intraoperative blood loss were also similar in the groups. Morbidity rates due to early postoperative complications were 20.2 and 23.1%, with pancreatic leakage occurring in 7.3 and 7.7% of patients, respectively. These differences were not statistically significant. One patient in the stented group died of sepsis following leakage of pancreaticojejunostomy. There were also no significant differences in the mean time to initiation of solid food intake or postoperative hospital stay. Conclusion: We conclude that a stenting tube is unnecessary if the duct-to-mucosa anastomosis is completely performed. This operative technique can be considered a basic procedure for pancreaticojejunostomy because of the low risk. Copyright © 2002 S. Karger AG, Basel and IAP
Introduction
Since the first description by Kausch [1] and the application by Whipple et al. [2], pancreaticoduodenectomy (PD) has become a standard operation for patients with periampullary disease. In recent years, due to remarkable advances in diagnostic techniques, small periampullary lesions with mostly normal soft parenchyma have become detectable and surgically treatable. Although postoperative morbidity and mortality have been reported to have
Toshihide Imaizumi, MD Department of Gastroenterological Surgery, Tokyo Women’s Medical University 8-1, Kawada-cho, Shinjuku-ku, Tokyo, 162-8666 (Japan) Tel. +81 3 3353 8111, ext 25213, Fax +81 3 5269 7435 E-Mail [email protected]
Table 1. Background characteristics of
patients undergoing pancreaticojejunostomy in the normal pancreas (September 1992 to December 2000)
Age, years Gender (M:F) Disease Invasive ductal adenocarcinoma of pancreatic head Mucin-producing pancreatic tumor Islet cell tumor Serous cyst adenoma Cancer of papilla of Vater Bile duct cancer Duodenal cancer Others
decreased [3, 4], complications in patients with a normal pancreas are still high. The most frequent early complication after PD is anastomotic leakage of the pancreaticojejunostomy resulting in serious complications, such as intra-abdominal abscess or subsequent bleeding. The normal pancreas with a nondilated pancreatic duct poses a particularly high risk for these complications. This procedure is therefore generally carried out using a stenting tube, especially for the normal pancreas [5, 6]. However, we have seen several complications associated with the stenting tube; acute pancreatitis due to subsequent occlusion or bending of the stenting tube or late anastomotic stenosis following iatrogenic injury sustained when withdrawing the external stenting tube. Since September 1992, we have performed pancreaticojejunostomy by duct-to-mucosa anastomosis without stenting tube, even in patients with a normal pancreas with a nondilated duct. In this study, we focused on the normal soft pancreas with a nondilated duct and compared the surgical results after stented and nonstented duct-to-mucosa anastomosis.
Patients and Methods In a consecutive series of 148 patients with a normal, soft pancreas who underwent pancreaticojejunostomy with duct-to-mucosa anastomosis, a historical study was performed to assess the necessity of a stenting tube. Depending on the choice of the surgeon, the anastomosis was performed without stenting tube in 109 of 148 patients (nonstented group) and an internal or external stenting tube was used in the remaining 39 patients (stented group). In both groups, ductto-mucosa end-to-side pancreaticojejunostomy was carried out in the manner described below. The pancreas was defined as normal if the preoperative pancreatic function test was within normal limits and, by means of intraoperative assessment, the pancreatic parenchyma
Stenting Is Unnecessary in Pancreaticojejunostomy
Nonstented (n = 109)
Stented (n = 39)
p value
60.2B2.3 56:53
60.6B3.2 25:14
NS NS
21 22 10 7 28 9 6 6
8 2 5 1 6 11 1 5
NS NS NS NS NS NS NS NS
Table 2. Operative procedure of patients undergoing pancreaticoje-
junostomy in the normal pancreas (September 1992 to December 2000) Operative procedure
Nonstented (n = 109)
Stented (n = 39)
p value
PpPD or PD DpPHR SR
89 3 17
33 2 4
NS NS NS
PpPD = Pylorus-preserving pancreaticoduodenectomy; PD = pancreaticoduodenectomy with gastrectomy; DpPHR = duodenumpreserving pancreatic head resection; SR = segmental resection of the pancreas.
was without fibrosis and the main pancreatic duct was 3 mm in diameter or less. As this is a historical study, the patients were not randomized. However, there was no significant difference in the age and sex distribution, the disease frequency or the prevalence of PD; duodenum preserving pancreatic head resection (DpPHR) or segmental resection of the pancreas (SR) (tables 1, 2). The operative results, including mean duration of the surgical procedure, duration to complete pancreaticojejunostomy, intraoperative blood loss, early morbidity rate, frequency of anastomotic leakage at the pancreaticojejunostomy, mortality rate, time to initiation of solid food intake and postoperative hospital stay were compared between the two groups. Operative Techniques The pancreas, together with the pancreatic duct, is sharply transected with a scalpel, taking care not to leave long parts of the ductal wall of the pancreas. Any bleeding points on the pancreatic cut end are coagulated by electrocautery or repaired with 4–0 or 5–0 nonabsorbable sutures, but no other treatment of the stump (such as mat-
Pancreatology 2002;2:116–121
117
Fig. 1. Transection of the pancreas and treatment of the pancreatic stump. a The pancreas is sharply transected with a scalpel. b Any bleeding points on the pancreatic stump are coagulated by electrocautery or sutured using 4–0 or 5–0 proline sutures. No mattress or fish-mouth shaped suturing is performed.
Fig. 2. Duct-to-mucosa anastomosis of pancreaticojejunostomy (posterior wall). a Stay stitches are placed at the middle anterior wall of the pancreatic duct and the jejunum. b Five stitches are placed at both sides and every quarter of the posterior wall.
Fig. 3. Duct-to-mucosa anastomosis of pancreaticojejunostomy (anterior wall). a Three stitches are placed at every quarter of the anterior wall. b Duct-to-mucosa anastomosis without stenting tube is completed.
tress or fish-mouth shaped suturing) is performed. A small hole is made on the jejunal wall, but no excision of the jejunal seromuscularis or scarification is performed. The end-to-side anastomosis is created in two layers of sutures; the outer layer is composed of the capsular parenchyma of the pancreatic stump and the jejunal wall, and the inner layer is composed of the pancreatic duct and the whole jejunal muscularis. A posterior row of interrupted sutures and an anterior row of continuous sutures are placed in the outer layer using
118
Pancreatology 2002;2:116–121
4–0 nonabsorbable sutures, while interrupted sutures are placed in the inner layer using 6–0 absorbable sutures. These procedures are performed with a fine atraumatic needle and monofilament sutures. The point of this anastomotic technique is to preserve adequate patency of the pancreatic duct by carefully picking up the pancreatic duct wall, in order to avoid suture injury or constriction of the pancreatic duct. It is advantageous to place a fine stay suture at the middle of the anterior wall of the pancreatic duct. Pulling it slightly to the
Imaizumi/Harada/Hatori/Fukuda/Takasaki
left expands the duct enough to allow easy needle suturing. At least 5 stitches are placed on the posterior row and 3 stitches on the anterior row. In this technique, the patency of the anastomosis is preserved adequately without the need for a stenting tube, but the decision of whether a stenting tube was used or not was left to the staff surgeon’s preference (fig. 1–4). Statistics Differences between the groups were analyzed using Student’s t test and the ¯2 test. p values less than or equal to 0.05 were considered significant. Numeric data are expressed as the mean B SE.
Results
Fig. 4. Scheme of a cross-section of the pancreaticojejunostomy. The
outer layer of anastomosis consists of the capsular parenchyma of the pancreas and the seromuscular layer of the jejunum. The inner sutures from a duct-to-mucosa anastomosis between the pancreatic duct (including the pancreatic parenchyma) and the whole jejunal wall. Two-layer anastomosis is completed without stenting.
As shown in table 3, there were no significant differences in the mean duration of the surgical procedure, mean duration to complete pancreaticojejunostomy or mean intraoperative blood loss between the two groups. The overall morbidity rates were 20.2% in the nonstented group and 23.1 % in the stented group. Major anastomotic leakage of the pancreaticojejunostomy was observed in 8 of 109 patients (7.3%) and in 3 of 39 patients (7.7%), respectively. These differences were not statistically significant. One patient in the stented group died of sepsis due to pancreaticojejunostomy. The mean time to initiation of solid food intake and mean postoperative hospital stay were similar in both groups (table 4).
Table 3. Operative results of pancreaticojejunostomy in the normal
pancreas (September 1992 to December 2000) Nonstented Stented (n = 109) (n = 39) Mean duration of surgery, min 277B18 303B25 Mean duration to complete pancreaticojejunostomy, min 26.6B1.3 29.2B2.1 Mean blood loss, ml 1,029B144 1,236B240
p value
NS NS NS
Discussion
Since PD was established by Whipple [2], various innovations have been made and the procedure has been improved by many modifications. Following the development of various diagnostic techniques, the indications for PD have increased and it has become widely used with
Table 4. Outcome of pancreatico-
jejunostomy in the normal pancreas (September 1992 to December 2000) Morbidity rate, % Leakage of pancreaticojejunostomy, % Mortality rate, % Mean time to iniciate solid food, days Mean hospital stay, days
Stenting Is Unnecessary in Pancreaticojejunostomy
Nonstented (n = 109)
Stented (n = 39)
p value
20.2 7.3 0 15.6B1.4 29.1B1.5
23.1 7.7 2.6 13.7B1.9 27.7B2.7
NS NS NS NS NS
Pancreatology 2002;2:116–121
119
little risk. While a radical operation is required in cases of malignant diseases, various PD techniques ranging from an extended operation to surgery aimed at preserving the function of the whole stomach or pyloric ring [7], have been developed in order to preserve the ability to ingest a normal diet. In general, the morbidity rate due to complications after PD ranges from 18 to 52%. In particular, pancreatic fistula has been reported in 8–19% of patients [3, 4, 8, 9]. Serious complications such as intra-abdominal abscesses and bleeding are associated with anastomotic leakage of the pancreaticojejunostomy, and these complications directly influence the results of the operation. Since a normal, soft pancreas is very vulnerable to ischemia and actively produces exocrine secretions, pancreatic fistula, bleeding, autolysis or necrosis can easily occur after surgical treatment. Therefore, the various serious complications associated with pancreaticojejunostomy differ from those due to leakage of a gastrointestinal tract anastomosis, and pancreaticojejunostomy is the most difficult technique performed during reconstruction after PD. Various ideas on how to avoid pancreatic fistula and improve safety have been reported, such as abandoning one-step reconstruction involving pancreaticojejunostomy [10], preventing exocrine function by means of ligation plugging of the pancreatic duct [11] or total pancreatectomy. However, while these would make it possible to avoid leakage at the pancreaticojejunostomy site, it would be nonsense to leave no postoperative pancreatic function. Moreover, the quality of life after total pancreatectomy is extremely poor. Besides the prevention of pancreatic fistula, it is important to preserve pancreatic function by ensuring long-term patency of the anastomosis at the pancreaticojejunostomy site. For this reason, stenting tubes are commonly utilized when performing pancreaticojejunostomy [12, 13]. However, stenting tubes have been reported to be harmful for some patients receiving a duct-to-mucosa anastomosis during pancreaticojejunostomy, especially when the pancreatic duct is less than 3 mm in diameter [14]. We have also encountered several complications associated with stenting tubing. In two patients with external pancreatic stenting tubes, late stenosis of the pancreatic duct occurred due to iatrogenic injury sustained when withdrawing the external stenting tube. A number of patients also suffered damage due to pancreatic leakage, caused by bending or stenosis of the stenting tube at the point fixed to the skin. Similar complications have been reported at other medical institutions. Among patients with an internal stenting tube, we have encountered late pancreatic insufficiency due to subsequent obstruction of the tube or long-term placement of
120
Pancreatology 2002;2:116–121
the tube. Our experience with these complications associated with the stenting tube led to the conclusion that stenting tubes are unnecessary if duct-to-mucosa anastomosis is performed satisfactorily, because the patency of the pancreatic duct is adequately preserved without. It is recently reported that leakage of the pancreaticojejunostomy is markedly reduced by use of stenting tube [5]. In this kind of discussion, the degree of pancreatic fibrosis and the type of pancreaticojejunostomy are very important. In the report, more than half of those patients that underwent the nonstented method and in whom leakage of the pancreaticojejunostomy occurred had undergone the invagination technique. When one performs pancreaticojejunostomy using the invagination technique without stenting tube, leakage or stenosis of the anastomosis naturally occurs. If the usefulness of the stenting tube is insisted upon, anastomosis should be carried out in the same fashion and the degree of pancreatic fibrosis should be standardized. In the present study, patients were not randomized but the study was limited to patients with a normal, soft pancreas with nondilated duct and the procedure was limited to duct-to-mucosa anastomosis. In the results obtained over the last 8 years, the rates of overall morbidity and anastomotic leakage of the pancreaticojejunostomy were about 20 and 7%, respectively, showing no significant difference and no patients died as a result of the operation. There were also no significant differences in the mean time before resuming food intake or the mean postoperative hospital stay. The point of this operation is to preserve satisfactory patency of the pancreatic duct by carefully picking up the pancreatic duct wall to prevent injury and stenosis of the pancreatic duct, and this appears to have been achieved. In conclusion, this operative technique can be considered a basic procedure for pancreaticojejunostomy because of the low risk involved. This procedure is useful for several types of reconstruction after partial pancreatectomy such as pancreaticojejunostomy involving a Roux-Y jejunal loop after segmental resection or pancreaticoduodenostomy after duodenum-preserving pancreatic head resection.
Imaizumi/Harada/Hatori/Fukuda/Takasaki
References 1 Kausch W: Das Carcinoma der Papilla duodeni and seine radikale Entfernung. Betrag Z Klin Chir 1912:78:439–486. 2 Whipple AO, Parsons WB, Mullins CR: Treatment of carcinoma of the ampulla of Vater. Ann Surg 1935:102:763–779. 3 Crist DW, Sitzmann JV, Cameron JL: Improved hospital morbidity, mortality, and survival after the Whipple procedure. Ann Surg 1987;206:358–365. 4 Grace PA, Pitt HA, Tompkins RK, DenBesten L, Longmire WP Jr: Decreased morbidity and mortality after pancreatoduodenectomy. Am J Surg 1986;151:141–149. 5 Roder JD, Stein HJ, Bottcher K, Busch R, Heidecke CD, Siewert JR: Stented versus nonstented pancreaticojejunostomy after pancreatoduodenectomy: A prospective study. Ann Surg 1999;229:41–48.
Stenting Is Unnecessary in Pancreaticojejunostomy
6 Yoshimi T, Ono H, Asato Y, Ohta T, Koizumi S, Amemita R, Hasegawa H: Internal stenting of the hepaticojejunostomy and pancreaticojejunostomy in patients undergoing pancreatoduodenectomy to promote earlier discharge from hospital. Surg Today Jpn J Surg 1996;26: 665–667. 7 Traverso LW, Longmire WP Jr: Preservation of the pylorus in pancreaticoduodenectomy. Surg Gynecol Obstet 1978;146:959–962. 8 Cameron JL, Pitt HA, Yeo CJ, Lillemoe KD, Kaufman HS, Coleman J: One hundred and forty-five consecutive pancreaticoduodenectomies without mortality. Ann Surg 1993;217: 430–435. 9 Trede M, Schwall G, Saeger H: Survival after pancreatoduodenectomy –118 consecutive resections without an operative mortality. Ann Surg 1990;211:447–458.
10 Miyagawa S, Makuuchi M, Kawasaki S: Second-stage pancreatojejunostomy following pancreatoduodenectomy in high-risk patients. Am J Surg 1994;168:66–68. 11 Idezuki Y, Goetz FC, Lillehai RC: Late effect of pancreatic duct ligation on beta cell function. Am J Surg 1969;117:33–39. 12 Biehl T, Traverso LW: Is stenting necessary for a successful pancreatic anastomosis? Am J Surg 1992;163:530–532. 13 Grace PA, Pitt HA, Longmire WP: Pancreatoduodenectomy with pylorus preservation for adenocarcinoma of the head of the pancreas. Br J Surg 1986;73:647–650. 14 Kingsnorth AN: Duct to mucosa isolated Roux loop pancreaticojejunostomy as an improved anastomosis after resection of the pancreas. Surg Gynecol Obstet 1989;169:451–453.
Pancreatology 2002;2:116–121
121
Pancreatology 2002;2:116–121
Stenting Is Unnecessary in Duct-to-Mucosa Pancreaticojejunostomy Even in the Normal Pancreas Toshihide Imaizumi Nobuhiko Harada Takashi Hatori Akira Fukuda Ken Takasaki Tokyo Women’s Medical University, Tokyo, Japan
Key Words Duct-to-mucosa pancreaticojejunostomy W Nonstented pancreaticojejunostomy W Reconstruction of the normal pancreas
Abstract Background: There is a high risk of anastomotic leakage after pancreaticojejunostomy after pancreaticoduodenectomy (PD) in patients with a normal pancreas because of the high degree of exocrine function. These PD are therefore generally performed using a stenting tube (stented method). In recent years, we have performed pancreaticojejunostomy with duct-to-mucosa anastomosis without a stenting tube (nonstented method) and obtained good results. Methods: The point of this technique is to preserve adequate patency of the pancreatic duct by carefully picking up the pancreatic duct wall with a fine atraumatic needle and monofilament thread. The results of end-to-side pancreaticojejunostomy of the normal pancreas were compared between the nonstented method (n = 109) and the stented method (n = 39). Results: There were no differences in background characteristics between the groups, including age, gender and disease. The mean duration to complete pancreaticojejunostomy was 26.6 min in the nonstented group and 29.2 min in the stented group. The mean durations
ABC
© 2002 S. Karger AG, Basel and IAP 1424–3903/02/0022–0116$18.50/0
Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com
Accessible online at: www.karger.com/journals/pan
of surgical procedure and intraoperative blood loss were also similar in the groups. Morbidity rates due to early postoperative complications were 20.2 and 23.1%, with pancreatic leakage occurring in 7.3 and 7.7% of patients, respectively. These differences were not statistically significant. One patient in the stented group died of sepsis following leakage of pancreaticojejunostomy. There were also no significant differences in the mean time to initiation of solid food intake or postoperative hospital stay. Conclusion: We conclude that a stenting tube is unnecessary if the duct-to-mucosa anastomosis is completely performed. This operative technique can be considered a basic procedure for pancreaticojejunostomy because of the low risk. Copyright © 2002 S. Karger AG, Basel and IAP
Introduction
Since the first description by Kausch [1] and the application by Whipple et al. [2], pancreaticoduodenectomy (PD) has become a standard operation for patients with periampullary disease. In recent years, due to remarkable advances in diagnostic techniques, small periampullary lesions with mostly normal soft parenchyma have become detectable and surgically treatable. Although postoperative morbidity and mortality have been reported to have
Toshihide Imaizumi, MD Department of Gastroenterological Surgery, Tokyo Women’s Medical University 8-1, Kawada-cho, Shinjuku-ku, Tokyo, 162-8666 (Japan) Tel. +81 3 3353 8111, ext 25213, Fax +81 3 5269 7435 E-Mail [email protected]
Table 1. Background characteristics of
patients undergoing pancreaticojejunostomy in the normal pancreas (September 1992 to December 2000)
Age, years Gender (M:F) Disease Invasive ductal adenocarcinoma of pancreatic head Mucin-producing pancreatic tumor Islet cell tumor Serous cyst adenoma Cancer of papilla of Vater Bile duct cancer Duodenal cancer Others
decreased [3, 4], complications in patients with a normal pancreas are still high. The most frequent early complication after PD is anastomotic leakage of the pancreaticojejunostomy resulting in serious complications, such as intra-abdominal abscess or subsequent bleeding. The normal pancreas with a nondilated pancreatic duct poses a particularly high risk for these complications. This procedure is therefore generally carried out using a stenting tube, especially for the normal pancreas [5, 6]. However, we have seen several complications associated with the stenting tube; acute pancreatitis due to subsequent occlusion or bending of the stenting tube or late anastomotic stenosis following iatrogenic injury sustained when withdrawing the external stenting tube. Since September 1992, we have performed pancreaticojejunostomy by duct-to-mucosa anastomosis without stenting tube, even in patients with a normal pancreas with a nondilated duct. In this study, we focused on the normal soft pancreas with a nondilated duct and compared the surgical results after stented and nonstented duct-to-mucosa anastomosis.
Patients and Methods In a consecutive series of 148 patients with a normal, soft pancreas who underwent pancreaticojejunostomy with duct-to-mucosa anastomosis, a historical study was performed to assess the necessity of a stenting tube. Depending on the choice of the surgeon, the anastomosis was performed without stenting tube in 109 of 148 patients (nonstented group) and an internal or external stenting tube was used in the remaining 39 patients (stented group). In both groups, ductto-mucosa end-to-side pancreaticojejunostomy was carried out in the manner described below. The pancreas was defined as normal if the preoperative pancreatic function test was within normal limits and, by means of intraoperative assessment, the pancreatic parenchyma
Stenting Is Unnecessary in Pancreaticojejunostomy
Nonstented (n = 109)
Stented (n = 39)
p value
60.2B2.3 56:53
60.6B3.2 25:14
NS NS
21 22 10 7 28 9 6 6
8 2 5 1 6 11 1 5
NS NS NS NS NS NS NS NS
Table 2. Operative procedure of patients undergoing pancreaticoje-
junostomy in the normal pancreas (September 1992 to December 2000) Operative procedure
Nonstented (n = 109)
Stented (n = 39)
p value
PpPD or PD DpPHR SR
89 3 17
33 2 4
NS NS NS
PpPD = Pylorus-preserving pancreaticoduodenectomy; PD = pancreaticoduodenectomy with gastrectomy; DpPHR = duodenumpreserving pancreatic head resection; SR = segmental resection of the pancreas.
was without fibrosis and the main pancreatic duct was 3 mm in diameter or less. As this is a historical study, the patients were not randomized. However, there was no significant difference in the age and sex distribution, the disease frequency or the prevalence of PD; duodenum preserving pancreatic head resection (DpPHR) or segmental resection of the pancreas (SR) (tables 1, 2). The operative results, including mean duration of the surgical procedure, duration to complete pancreaticojejunostomy, intraoperative blood loss, early morbidity rate, frequency of anastomotic leakage at the pancreaticojejunostomy, mortality rate, time to initiation of solid food intake and postoperative hospital stay were compared between the two groups. Operative Techniques The pancreas, together with the pancreatic duct, is sharply transected with a scalpel, taking care not to leave long parts of the ductal wall of the pancreas. Any bleeding points on the pancreatic cut end are coagulated by electrocautery or repaired with 4–0 or 5–0 nonabsorbable sutures, but no other treatment of the stump (such as mat-
Pancreatology 2002;2:116–121
117
Fig. 1. Transection of the pancreas and treatment of the pancreatic stump. a The pancreas is sharply transected with a scalpel. b Any bleeding points on the pancreatic stump are coagulated by electrocautery or sutured using 4–0 or 5–0 proline sutures. No mattress or fish-mouth shaped suturing is performed.
Fig. 2. Duct-to-mucosa anastomosis of pancreaticojejunostomy (posterior wall). a Stay stitches are placed at the middle anterior wall of the pancreatic duct and the jejunum. b Five stitches are placed at both sides and every quarter of the posterior wall.
Fig. 3. Duct-to-mucosa anastomosis of pancreaticojejunostomy (anterior wall). a Three stitches are placed at every quarter of the anterior wall. b Duct-to-mucosa anastomosis without stenting tube is completed.
tress or fish-mouth shaped suturing) is performed. A small hole is made on the jejunal wall, but no excision of the jejunal seromuscularis or scarification is performed. The end-to-side anastomosis is created in two layers of sutures; the outer layer is composed of the capsular parenchyma of the pancreatic stump and the jejunal wall, and the inner layer is composed of the pancreatic duct and the whole jejunal muscularis. A posterior row of interrupted sutures and an anterior row of continuous sutures are placed in the outer layer using
118
Pancreatology 2002;2:116–121
4–0 nonabsorbable sutures, while interrupted sutures are placed in the inner layer using 6–0 absorbable sutures. These procedures are performed with a fine atraumatic needle and monofilament sutures. The point of this anastomotic technique is to preserve adequate patency of the pancreatic duct by carefully picking up the pancreatic duct wall, in order to avoid suture injury or constriction of the pancreatic duct. It is advantageous to place a fine stay suture at the middle of the anterior wall of the pancreatic duct. Pulling it slightly to the
Imaizumi/Harada/Hatori/Fukuda/Takasaki
left expands the duct enough to allow easy needle suturing. At least 5 stitches are placed on the posterior row and 3 stitches on the anterior row. In this technique, the patency of the anastomosis is preserved adequately without the need for a stenting tube, but the decision of whether a stenting tube was used or not was left to the staff surgeon’s preference (fig. 1–4). Statistics Differences between the groups were analyzed using Student’s t test and the ¯2 test. p values less than or equal to 0.05 were considered significant. Numeric data are expressed as the mean B SE.
Results
Fig. 4. Scheme of a cross-section of the pancreaticojejunostomy. The
outer layer of anastomosis consists of the capsular parenchyma of the pancreas and the seromuscular layer of the jejunum. The inner sutures from a duct-to-mucosa anastomosis between the pancreatic duct (including the pancreatic parenchyma) and the whole jejunal wall. Two-layer anastomosis is completed without stenting.
As shown in table 3, there were no significant differences in the mean duration of the surgical procedure, mean duration to complete pancreaticojejunostomy or mean intraoperative blood loss between the two groups. The overall morbidity rates were 20.2% in the nonstented group and 23.1 % in the stented group. Major anastomotic leakage of the pancreaticojejunostomy was observed in 8 of 109 patients (7.3%) and in 3 of 39 patients (7.7%), respectively. These differences were not statistically significant. One patient in the stented group died of sepsis due to pancreaticojejunostomy. The mean time to initiation of solid food intake and mean postoperative hospital stay were similar in both groups (table 4).
Table 3. Operative results of pancreaticojejunostomy in the normal
pancreas (September 1992 to December 2000) Nonstented Stented (n = 109) (n = 39) Mean duration of surgery, min 277B18 303B25 Mean duration to complete pancreaticojejunostomy, min 26.6B1.3 29.2B2.1 Mean blood loss, ml 1,029B144 1,236B240
p value
NS NS NS
Discussion
Since PD was established by Whipple [2], various innovations have been made and the procedure has been improved by many modifications. Following the development of various diagnostic techniques, the indications for PD have increased and it has become widely used with
Table 4. Outcome of pancreatico-
jejunostomy in the normal pancreas (September 1992 to December 2000) Morbidity rate, % Leakage of pancreaticojejunostomy, % Mortality rate, % Mean time to iniciate solid food, days Mean hospital stay, days
Stenting Is Unnecessary in Pancreaticojejunostomy
Nonstented (n = 109)
Stented (n = 39)
p value
20.2 7.3 0 15.6B1.4 29.1B1.5
23.1 7.7 2.6 13.7B1.9 27.7B2.7
NS NS NS NS NS
Pancreatology 2002;2:116–121
119
little risk. While a radical operation is required in cases of malignant diseases, various PD techniques ranging from an extended operation to surgery aimed at preserving the function of the whole stomach or pyloric ring [7], have been developed in order to preserve the ability to ingest a normal diet. In general, the morbidity rate due to complications after PD ranges from 18 to 52%. In particular, pancreatic fistula has been reported in 8–19% of patients [3, 4, 8, 9]. Serious complications such as intra-abdominal abscesses and bleeding are associated with anastomotic leakage of the pancreaticojejunostomy, and these complications directly influence the results of the operation. Since a normal, soft pancreas is very vulnerable to ischemia and actively produces exocrine secretions, pancreatic fistula, bleeding, autolysis or necrosis can easily occur after surgical treatment. Therefore, the various serious complications associated with pancreaticojejunostomy differ from those due to leakage of a gastrointestinal tract anastomosis, and pancreaticojejunostomy is the most difficult technique performed during reconstruction after PD. Various ideas on how to avoid pancreatic fistula and improve safety have been reported, such as abandoning one-step reconstruction involving pancreaticojejunostomy [10], preventing exocrine function by means of ligation plugging of the pancreatic duct [11] or total pancreatectomy. However, while these would make it possible to avoid leakage at the pancreaticojejunostomy site, it would be nonsense to leave no postoperative pancreatic function. Moreover, the quality of life after total pancreatectomy is extremely poor. Besides the prevention of pancreatic fistula, it is important to preserve pancreatic function by ensuring long-term patency of the anastomosis at the pancreaticojejunostomy site. For this reason, stenting tubes are commonly utilized when performing pancreaticojejunostomy [12, 13]. However, stenting tubes have been reported to be harmful for some patients receiving a duct-to-mucosa anastomosis during pancreaticojejunostomy, especially when the pancreatic duct is less than 3 mm in diameter [14]. We have also encountered several complications associated with stenting tubing. In two patients with external pancreatic stenting tubes, late stenosis of the pancreatic duct occurred due to iatrogenic injury sustained when withdrawing the external stenting tube. A number of patients also suffered damage due to pancreatic leakage, caused by bending or stenosis of the stenting tube at the point fixed to the skin. Similar complications have been reported at other medical institutions. Among patients with an internal stenting tube, we have encountered late pancreatic insufficiency due to subsequent obstruction of the tube or long-term placement of
120
Pancreatology 2002;2:116–121
the tube. Our experience with these complications associated with the stenting tube led to the conclusion that stenting tubes are unnecessary if duct-to-mucosa anastomosis is performed satisfactorily, because the patency of the pancreatic duct is adequately preserved without. It is recently reported that leakage of the pancreaticojejunostomy is markedly reduced by use of stenting tube [5]. In this kind of discussion, the degree of pancreatic fibrosis and the type of pancreaticojejunostomy are very important. In the report, more than half of those patients that underwent the nonstented method and in whom leakage of the pancreaticojejunostomy occurred had undergone the invagination technique. When one performs pancreaticojejunostomy using the invagination technique without stenting tube, leakage or stenosis of the anastomosis naturally occurs. If the usefulness of the stenting tube is insisted upon, anastomosis should be carried out in the same fashion and the degree of pancreatic fibrosis should be standardized. In the present study, patients were not randomized but the study was limited to patients with a normal, soft pancreas with nondilated duct and the procedure was limited to duct-to-mucosa anastomosis. In the results obtained over the last 8 years, the rates of overall morbidity and anastomotic leakage of the pancreaticojejunostomy were about 20 and 7%, respectively, showing no significant difference and no patients died as a result of the operation. There were also no significant differences in the mean time before resuming food intake or the mean postoperative hospital stay. The point of this operation is to preserve satisfactory patency of the pancreatic duct by carefully picking up the pancreatic duct wall to prevent injury and stenosis of the pancreatic duct, and this appears to have been achieved. In conclusion, this operative technique can be considered a basic procedure for pancreaticojejunostomy because of the low risk involved. This procedure is useful for several types of reconstruction after partial pancreatectomy such as pancreaticojejunostomy involving a Roux-Y jejunal loop after segmental resection or pancreaticoduodenostomy after duodenum-preserving pancreatic head resection.
Imaizumi/Harada/Hatori/Fukuda/Takasaki
References 1 Kausch W: Das Carcinoma der Papilla duodeni and seine radikale Entfernung. Betrag Z Klin Chir 1912:78:439–486. 2 Whipple AO, Parsons WB, Mullins CR: Treatment of carcinoma of the ampulla of Vater. Ann Surg 1935:102:763–779. 3 Crist DW, Sitzmann JV, Cameron JL: Improved hospital morbidity, mortality, and survival after the Whipple procedure. Ann Surg 1987;206:358–365. 4 Grace PA, Pitt HA, Tompkins RK, DenBesten L, Longmire WP Jr: Decreased morbidity and mortality after pancreatoduodenectomy. Am J Surg 1986;151:141–149. 5 Roder JD, Stein HJ, Bottcher K, Busch R, Heidecke CD, Siewert JR: Stented versus nonstented pancreaticojejunostomy after pancreatoduodenectomy: A prospective study. Ann Surg 1999;229:41–48.
Stenting Is Unnecessary in Pancreaticojejunostomy
6 Yoshimi T, Ono H, Asato Y, Ohta T, Koizumi S, Amemita R, Hasegawa H: Internal stenting of the hepaticojejunostomy and pancreaticojejunostomy in patients undergoing pancreatoduodenectomy to promote earlier discharge from hospital. Surg Today Jpn J Surg 1996;26: 665–667. 7 Traverso LW, Longmire WP Jr: Preservation of the pylorus in pancreaticoduodenectomy. Surg Gynecol Obstet 1978;146:959–962. 8 Cameron JL, Pitt HA, Yeo CJ, Lillemoe KD, Kaufman HS, Coleman J: One hundred and forty-five consecutive pancreaticoduodenectomies without mortality. Ann Surg 1993;217: 430–435. 9 Trede M, Schwall G, Saeger H: Survival after pancreatoduodenectomy –118 consecutive resections without an operative mortality. Ann Surg 1990;211:447–458.
10 Miyagawa S, Makuuchi M, Kawasaki S: Second-stage pancreatojejunostomy following pancreatoduodenectomy in high-risk patients. Am J Surg 1994;168:66–68. 11 Idezuki Y, Goetz FC, Lillehai RC: Late effect of pancreatic duct ligation on beta cell function. Am J Surg 1969;117:33–39. 12 Biehl T, Traverso LW: Is stenting necessary for a successful pancreatic anastomosis? Am J Surg 1992;163:530–532. 13 Grace PA, Pitt HA, Longmire WP: Pancreatoduodenectomy with pylorus preservation for adenocarcinoma of the head of the pancreas. Br J Surg 1986;73:647–650. 14 Kingsnorth AN: Duct to mucosa isolated Roux loop pancreaticojejunostomy as an improved anastomosis after resection of the pancreas. Surg Gynecol Obstet 1989;169:451–453.
Pancreatology 2002;2:116–121
121