Factors associated with anastomotic leak following gastrectomy for gastric adenocarcinoma and its effect on long-term outcomes

Abstract

Objective: Gastrectomy for cancer is a technically demanding surgery and anastomotic leak is an important complication of this surgery. This study aimed to identify the factors associated with anastomotic leak following gastrectomy in gastric cancer patients and its long-term effect on outcomes.

Material and Methods: This is an ambispective study of 181 patients who underwent curative gastrectomy for gastric adenocarcinoma over 13 years, at our institution. Groups with and without anastomotic leak were compared using the Mann-Whitney U test (continuous variables) and Chi-square test (categorical variables). A multivariable analysis was performed to determine the factors associated with anastomotic leak.

Results: Out of the 181 patients who underwent curative gastrectomy, 12 (6.6%) patients experienced anastomotic leak. A multivariable analysis revealed that younger age, presence of comorbidities, hypoalbuminemia, tumor location in the proximal stomach, type of reconstruction, and positive margin status were associating factors for anastomotic leak. During a median follow-up of 34 months (ranging from 12 to 130), it was observed that 25 (18.3%) patients developed anastomotic stenosis, but it was not related to anastomotic leak. The incidence of post-operative pulmonary complications, administration of adjuvant therapy, recurrence rates, and mortality due to anastomotic leak did not significantly change. Moreover, neoadjuvant therapy did not increase the incidence of anastomotic leaks.

Conclusion: Factors like younger age, presence of comorbidities, hypoalbuminemia, tumor location in the proximal stomach, type of reconstruction, and positive margin status were associated with increased risk of anastomotic leak, which needs further studies to validate the findings. Thus, preoperative optimization and resection with adequate margins may be of utmost importance in preventing anastomotic leaks.

Keywords: Gastric cancer, gastrectomy, anastomotic leak, long-term effect, gastrojejunostomy, esophagojejunostomy

Introduction

Gastric cancer is the fifth most common cancer and the fourth most common cause of cancer deaths in the world (1). Gastrectomy with D2 lymphadenectomy is the main treatment for gastric cancer (2). Gastrectomy for cancer requires expertise, with complication rates varying from 20%-46% (2). The incidence of anastomotic leak following gastrectomy is 1.5-14.4% (3-7). Mortality due to anastomotic leak is 38.5% of all surgery-related mortality after gastrectomy for gastric cancer (7). Anastomotic leaks lead to poor quality of life, lengthening hospital stay, increased financial burden, and mortality (8).

Post-operative complications can hamper recovery, delaying the initiation of adjuvant chemotherapy, which can adversely affect the overall and recurrencefree survival of patients after curative gastrectomy for gastric cancer. These complications can be disastrous to both short and long-term outcomes (9,10). There is some uncertainty about neoadjuvant therapy causing an increased risk of post-operative anastomotic leak (11).

Most of the studies have assessed risk factors for esophago-jejunal anastomotic leak, but few have addressed the complications following gastrojejunostomy leak. No randomized study is available, and the existing literature has shown conflicting results in terms of determinants of leak and its sequelae. The present study aimed to identify the factors associated with anastomotic leak following gastrectomy in gastric adenocarcinoma patients and its long-term effect on outcome.

Material and Methods

It is an ambispective observational study in which all consecutive patients who underwent potentially curative gastrectomy (total, subtotal/distal, proximal) for gastric adenocarcinoma from 2009 to 2021 at a tertiary care center were included after ethical clearance from institutional ethics committee (IEC code: 2023-93-MCh-EXP-52 PGI/BE/224/2023). Data was retrieved from a prospectively maintained hospital information system and informed consent was taken from the patient/patient’s family. Patients who underwent resection or a bypass procedure for benign pathology, non-adenocarcinoma malignant pathology, or metastatic disease were excluded from the study.

Patients’ demographic characteristics were recorded along with the following clinical, surgical, and pathological characteristics (e.g., pre-operative patient factors, neoadjuvant therapy, surgery-related factors, and tumor-related factors). Early postoperative complications like pulmonary, and cardiac complications were also evaluated.

All patients were divided into two groups as anastomotic leak (AL) group and no anastomotic leak (NAL) group. These groups were compared in terms of the above clinicopathological and surgical factors. All patients were followed up for at least 12 months. Follow-up data was collected from the outpatient records and/or telephonic conversations. Long-term anastomotic complications (anastomotic stenosis, fistula formation) were noted and analyzed.

The decision for neoadjuvant therapy was made after a discussion in a tumor board. Total or proximal gastrectomy or subtotal/distal gastrectomy with D2 lymph node dissection with or without adjacent organ resection was determined by the stage of the disease and location of the tumor, using standard criteria.

Anastomotic leak in the present study was defined as the leak of luminal contents from the anastomotic line that is from Rouxen-Y esophago-jejunal (RYEJ)/esophagogastric (EG) or gastrojejunal (GJ) anastomotic site with clinical manifestations. They presented clinically as luminal contents through a wound or drain or with collection near the anastomosis associated with fever, inflammatory response, metabolic disturbance, and/or multiple-organ failure, confirmed by oral contrast CT or at re-operation.

Anastomotic stenosis in the present study was defined as anastomotic site narrowing post-gastrectomy who presented with features of gastric outlet obstruction at least after one month of surgery confirmed by endoscopy or contrast study and those requiring endoscopic intervention or revision surgery.

Anastomotic leaks containing collections were managed by percutaneous drainage, antibiotics, and enteral feeding through a feeding jejunostomy (FJ) tube (placed during primary surgery). In patients who could not tolerate enteral feeds or in whom enteral feeding access was not available, total parenteral nutrition (TPN) was initiated. Surgical re-exploration was reserved for hemodynamically unstable patients, who demonstrated signs of clinical deterioration on conservative management or in case of free contrast leak into the peritoneal cavity.

Endoscopic balloon dilation was considered a primary treatment for anastomotic stenosis following gastrectomy and surgery was reserved for the failures.

Statistical Analysis

Patients were divided into two groups as anastomotic leak (AL) and no anastomotic leak (NAL) groups. Continuous variables were expressed as median and compared using the MannWhitney U test. The association between categorical data was compared using the Chi-square test. Univariate analysis was done with logistic regression analysis and the variables which had p< 0.05 were considered for multivariable regression analysis. P< 0.05 was considered significant. Statistical analysis was done using the IBM SPSS 26.0 (SPSS Inc., Chicago, IL, United Stated of America) package program.

Results

Of a total of 287 patients who underwent gastrectomy for gastric tumors of all types during this period, a total of 181 patients were included for analysis. Twelve (6.6%) of them had anastomotic leak (six following total gastrectomy; five following distal/subtotal gastrectomy; one following proximal gastrectomy). Nine were managed conservatively with antibiotic upgradation and the drain was kept for a longer time or percutaneous drain placement done in them. The remaining three patients underwent re-exploration due to the persistence of symptoms or deterioration with conservative management. Median time of presentation of the leak was on the 6^th postoperative day (range four to 48 days), and median hospital stay was 18 days (range nine to 23 days).

Median age of the patients was 54 (18-85) years with 75.7% being male. Comorbidities were more commonly seen in the AL group (p= 0.02). Univariate analysis suggested that the following factors were associated with anastomotic leak younger age, presence of comorbidities, pre-operative anemia, blood transfusion, hypoalbuminemia, total gastrectomy, positive margin status, and tumor location. Neoadjuvant therapy, the technique of anastomosis, the extent of lymphadenectomy, en-bloc resection of adjacent organs, and tumor-related pathological factors were not correlated with increased anastomotic leak (Table 1-3). Pulmonary complications were more common in the AL group (33.3% vs.15.4%), although the difference was not statistically significant. There was no difference in the postoperative mortality, administration of adjuvant therapy, or recurrence rates between the two groups (Table 4).

On multivariable analysis, factors that conferred significant adverse impact on anastomotic leak rate were younger age, presence of comorbidities, hypoalbuminemia, proximal tumor location, type of reconstruction, and positive margin status (Table 5).

Of the total of 181 patients, 136 were available for evaluation of long-term anastomotic complications. Out of the 136 patients, 25 (18.3%) developed anastomotic stenosis after a median follow-up of 34 months (12 to 130 months). No chronic fistula was documented. Only two out of 25 patients who developed anastomotic stricture (8%) had a history of AL, and the remaining 23 (92%) had tumor recurrence (local, distant, and both local and distant recurrence). Anastomotic leak did not influence the rate of anastomotic stenosis (8% vs. 9%, p= 0.97). Most of these patients were presented with dysphagia or gastric outlet obstruction, evaluated by upper GI endoscopy and biopsy from stricture. In the AL group, both patients with stenosis had benign stricture, underwent esophageal dilatation, and none required reoperation. In the NAL group, all stenoses were due to the recurrence of the disease. Two patients with local recurrence at the anastomotic site leading to stenosis underwent endoscopic guided SEMS placement, the remaining 21 patients were managed conservatively with best supportive care.

Discussion

In the present study, anastomotic leak rate following curative gastric resection for adenocarcinoma was 6.6%. This is similar to that reported at various centers across the globe. The Japanese National Clinical Database (NCD) of digestive surgery reported an anastomotic leak rate following total gastrectomy as 4.4% (881 of 20011) in 2011. Recent studies from Japan and Korea have shown anastomotic leak incidence ranging from 1.5-4.9%, whereas studies from the Western world ranged from 5.2-14.4% (3-7).

Of the 12 patients in the AL group, nine (75%) were successfully managed conservatively (five required percutaneous drain insertion, four had prolonged surgical drain in situ), while three (25%) patients required re-exploration. In a systematic review in 2015, surgical re-exploration was necessary in 23.7% of patients which is comparable to the present study (12). Mortality following an anastomotic leak in the present study was 8.3%, which was comparable with a large-volume retrospective study by Roh et al. (7).

In our study, younger age, presence of comorbidities, hypoalbuminemia, tumor location, type of reconstruction, and positive margin status were associated with increased risk of anastomotic leak.

We found that the leak rate was higher in younger age patients. Median age of patients in the AL group was nearly 10 years less than that in the NAL group. This is in contrast with other studies that reported more complications in the older age group (7,13). A definitive explanation is not possible, however, in the present study, we found that younger patients had a higher incidence of signet ring cell tumors involving the proximal stomach necessitating more extensive resection (total gastrectomy) and esophagojejunostomy.

The presence of co-morbidities independently increased the risk (odds ratio was 15) of anastomotic leak. Out of 12 patients presenting with leak, nine (75%) were suffering from one or more medical illnesses (diabetes, hypertension, chronic pulmonary disease, or cardiac illness). A Korean study by Roh et al. documented similar findings wherein the patients who developed AL, 61 percent were affected by one or multiple co-morbidity (7). Kim et al. in their report have suggested the presence of cardiovascular disorder as a significant factor influencing the rates of anastomotic dehiscence (odds ratio 1.8) (13). Cardiovascular diseases and diabetes may increase the need for vasopressor support in the peri-operative period, impair the microcirculation, and adversely affect the glucose metabolism thus impeding the process of wound (anastomotic) healing.

Hypoalbuminemia (albumin <3.5 gm/dL) was also found to be an independent factor associated with anastomotic leak with an odds ratio of 10. Malnutrition renders patients more susceptible to infection, increases tissue edema, prolongs wound healing, and increases the risk of post-operative complications (10). It emphasizes the importance of preoperative nutritional optimization. Liu et al., in their report, have documented similar results (14). On the contrary, Migita et al. and Kim et al. fhave ailed to demonstrate any association between serum albumin and anastomotic healing (8,13).

Tumors with microscopic margins positive for tumor cells (R1 resection) had a higher risk (six times) of anastomotic leak. Kim et al. have also documented a significant rise in anastomotic complications (2.3 times) with a positive resection margin (13).

Eight times increased risk of anastomotic leak was documented in patients with tumors involving the whole stomach or most of the body of the stomach necessitating total gastrectomy and esophagojejunostomy. A similar rise has been noted with proximal tumors (proximal third of the stomach) in a study by Kim et al. (13). Regardless of the anastomotic technique (stapled or sutured), RYEJ is considered technically more difficult than GJ. Moreover, the esophagus is a less compliant organ in comparison to the stomach as well as the blood supply is less robust. The proximal gross margin in the case of proximal gastrectomy (1-2 cm) is less than that in distal gastrectomy (usually >5 cm). The proximity to the tumor tissue might be responsible for poor healing capacity.

Conversely, combined organ resection, technique of anastomosis, extent of lymph node dissection, and stage of the disease did not affect the leak rate. Importantly, in our study, 37 of 181 (20.4%) patients received neoadjuvant therapy but this did not lead to an increased rate of anastomotic leak. Studies on the effect of neoadjuvant therapy in colorectal cancer have demonstrated tumor regression to the extent of complete pathologic response leading to improved local control (15). This may counteract the ill effects (tissue edema and inflammation) of neoadjuvant therapy (chemotherapy and/or radiotherapy), and thus may not translate into increased AL.

In the present study, 47/181 (25.9%) patients had signet ring cell adenocarcinoma, however, histopathological type did not significantly increase leak rates. Similarly, the need for multivisceral resection, the extent of lymphadenectomy, the technique of anastomosis, and the stage of the disease did not have a significant effect on anastomotic dehiscence as confirmed by a Japanese study by Migita et al. (8). In the present study, 25/136 (18.3%) patients developed anastomotic stenosis during follow-up. However, post-operative leak (16.7% vs. 18.5%) was not associated with the development of stricture. Fukagawa et al. have documented the stenosis rate of 7.8% following open proximal gastrectomy, and 3.4% following open total gastrectomy (16). Multiple factors have been proposed for the development of mechanical gastric outlet obstruction: Ischemia, tension following tissue approximation, subacute obstruction, use of circular staplers, narrow diameter staplers, and the occurrence of an anastomotic leak (17). However, the present study did not show any difference in terms of anastomotic leaks.

Being an ambispective single-centre study, the limitations of this study include small sample size, the possibility of selection bias, and a low incidence of anastomotic leaks. Multicentre large prospective studies are required to validate the results.

Conclusion

Factors like younger age, presence of comorbidities, hypoalbuminemia, tumor location in the proximal stomach, type of reconstruction, and positive margin status were associated with increased risk of anastomotic leak, which needs further studies to validate these findings. Thus, pre-operative optimization and resection with adequate margins may be of utmost importance in preventing anastomotic leaks.

Cite this article as: Shaganti R, Godara SK, Singh RK, R R, Misra S, Kumar S. Factors associated with anastomotic leak following gastrectomy for gastric adenocarcinoma and its effect on long-term outcomes. Turk J Surg 2024; 40 (2): 111- 118.

This study was obtained from Sanjay Gandhi Post Graduate Institute Medical Sciences of Ethics Committee (Decision no: 2023-93-MCh-EXP-52, Date: 22.06.2023).

Externally peer-reviewed.

Concept - RKS; Design - RKS, SR, RR; Supervision - RKS, SM, SK, RR; Data Collection and/or Processing - RKS, SKG, SR; Analysis and/or Interpretation - RKS, SKG, SR; Literature Search - SR; Writing Manuscript - SR, RKS; Critical Reviews - All of authors.

The authors have no conflicts of interest to declare.

The authors declared that this study has received no financial support.

References

1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA 2021; 71(3): 209-49. https://doi.org/10.3322/caac.21660
2. Li J, Zhang Y, Hu DM, Gong TP, Xu R, Gao J. Impact of postoperative complications on long-term outcomes of patients following surgery for gastric cancer: A systematic review and meta-analysis of 64 follow-up studies. Asian J Surg 2020; 43(7): 719-29. https://doi. org/10.1016/j.asjsur.2019.10.007
3. Katai H, Mizusawa J, Katayama H, Kunisaki C, Sakuramoto S, Inaki N, et al. Stomach cancer study group of japan clinical oncology group. Single-arm confirmatory trial of laparoscopy-assisted total or proximal gastrectomy with nodal dissection for clinical stage I gastric cancer: Japan clinical oncology group study JCOG1401. Gastric Cancer 2019; 22(5): 999-1008. https://doi.org/10.1007/s10120-019-00929-9
4. Kanaji S, Ohyama M, Yasuda T, Sendo H, Suzuki S, Kawasaki K, et al. Can the intraoperative leak test prevent postoperative leakage of esophagojejunal anastomosis after total gastrectomy? Surg Today 2016; 46(7): 815-20. https://doi.org/10.1007/s00595-015-1243-y
5. Carboni F, Valle M, Federici O, Levi Sandri GB, Camperchioli I, Lapenta R, et al. Esophagojejunal anastomosis leakage after total gastrectomy for esophagogastric junction adenocarcinoma: Options of treatment. J Gastrointest Oncol 2016; 7(4): 51-22. https://doi.org/10.21037/ jgo.2016.06.02
6. El-Sourani N, Bruns H, Troja A, Raab HR, Antolovic D. Routine use of contrast swallow after total gastrectomy and esophagectomy: Is it justified? Pol J Radiol 2017; 82: 170-3. https://doi.org/10.12659/ PJR.899951
7. Roh CK, Choi S, Seo WJ, Cho M, Kim HI, Lee SK, et al. Incidence and treatment outcomes of leakage after gastrectomy for gastric cancer: Experience of 14,075 patients from a large volume centre. Eur J Surg Oncol 2021; 47(9): 2304-12. https://doi.org/10.1016/j. ejso.2021.02.013
8. Migita K, Takayama T, Matsumoto S, Wakatsuki K, Enomoto K, Tanaka T, et al. Risk factors for esophagojejunal anastomotic leakage after elective gastrectomy for gastric cancer. J Gastrointest Surg 2012; 16(9): 1659-65. https://doi.org/10.1007/s11605-012-1932-4
9. Lee HJ, Park W, Lee H, Lee KH, Park JC, Shin SK, et al. Endoscopy-guided balloon dilation of benign anastomotic strictures after radical gastrectomy for gastric cancer. Gut Liver 2014; 8(4): 394-9. https://doi. org/10.5009/gnl.2014.8.4.394
10. Kanda M. Preoperative predictors of postoperative complications after gastric cancer resection. Surg Today 2020; 50(1): 3-11. https://doi. org/10.1007/s00595-019-01877-8
11. Haskins IN, Kroh MD, Amdur RL, Ponksy JL, Rodriguez JH, Vaziri K. The effect of neoadjuvant chemoradiation on anastomotic leak and additional 30-day morbidity and mortality in patients undergoing total gastrectomy for gastric cancer. J Gastrointest Surg 2017; 21(10): 1577- 83. https://doi.org/10.1007/s11605-017-3496-9
12. Makuuchi R, Irino T, Tanizawa Y, Bando E, Kawamura T, Terashima M. Esophagojejunal anastomotic leakage following gastrectomy for gastric cancer. Surg Today 2019; 49(3): 187-96. https://doi. org/10.1007/s00595-018-1726-8
13. Kim SH, Son SY, Park YS, Ahn SH, Park DJ, Kim HH, et al. Risk factors for anastomotic leakage: A retrospective cohort study in a single gastric surgical unit. J Gastric Cancer 2015; 15(3):167-75. https://doi. org/10.5230/jgc.2015.15.3.167
14. Liu ZJ, Ge XL, Ai SC, Wang HK, Sun F, Chen L, et al. Postoperative decrease of serum albumin predicts short-term complications in patients undergoing gastric cancer resection. World J Gastroenterol 2017; 23(27): 4978-85. https://doi.org/10.3748/wjg.v23.i27.4978
15. Hu MH, Huang RK, Zhao RS, Yang KL, Wang H. Does neoadjuvant therapy increase the incidence of anastomotic leakage after anterior resection for mid and low rectal cancer? A systematic review and meta-analysis. Colorectal Dis 2017; 19(1): 16-26. https://doi.org/10.1111/ codi.13424
16. Fukagawa T, Gotoda T, Oda I, Deguchi Y, Saka M, Morita S, et al. Stenosis of esophago-jejuno anastomosis after gastric surgery. World J Surg 2010; 34(8): 1859-63. https://doi.org/10.1007/s00268-010- 0609-y
17. Guyton KL, Hyman NH, Alverdy JC. Prevention of perioperative anastomotic healing complications: Anastomotic stricture and anastomotic leak. Adv Surg 2016; 50(1): 129-41. https://doi.org/10.1016/j. yasu.2016.03.011