Emergency Radical Gastrectomy with Pancreatico-duodenectomy for a Recent Onset Perforation of Locally Advanced Gastric Cancer with Pancreatic Head Involvement—Exceptional but a Definite Option

Abstract

Gastric cancer perforations are rare events with management options ranging from lavage and perforation closure, to resection. Usual aim is to perform a damage control procedure, and very few patients are suitable for a curative resection. We report the first case of emergency gastrectomy with pancreatico-duodenectomy performed in emergency for a perforated stomach cancer with pancreatic head invasion. The patient was a 32-year-old gentleman who presented with a perforated antro-pyloric cancer with infiltration of pancreatic head. Emergency radical gastrectomy with en-bloc pancreatico-duodenectomy was performed with due considerations to the patient and disease factors. He had an uneventful postoperative recovery and remains disease free at 18 months of follow-up after having received adjuvant chemotherapy. Curative resections should be selectively offered in advanced (T4b) gastric cancers in patients without multiple adverse factors. In an emergency situation with perforation peritonitis, if the magnitude of resection is deemed unlikely to add to significant morbidity of the surgery, taking multiple factors into consideration, an R0 resection can offer a large survival benefit in such settings.

Introduction

Perforation in gastric cancer is a rare event occurring in less than 1% of the patients [1]. The pancreas is most frequent organ involved in T4b gastric cancers, especially in distal tumors [2, 3]. The combination of perforated gastric cancer with pancreatic invasion is a much rare event, and to the best of our knowledge, no reports mention of the same being treated in a curative manner with combined gastrectomy and pancreatico-duodenectomy. We report the first such case with favorable outcomes and provide rationale for management.

Case Description

A 32-year-old gentleman with no co-morbidities presented with complaints of epigastric pain and vomiting of 1-month duration associated with weight loss and anorexia. Performance status of the patient in the Eastern Co-operative Oncology Group (ECOG) scale was 1. Other than epigastric fullness, physical examination was unremarkable. Upper GI endoscopy was performed 4 days after presentation that showed an ulcerative growth in the pylorus extending into 1st part of duodenum. A naso-jejunal tube was placed during the same procedure for nutritional rehabilitation as the patient had partial gastric outlet obstruction. Biopsy from the growth was reported as moderately differentiated adenocarcinoma. A contrast-enhanced CT scan was performed of abdomen and pelvis which revealed diffuse wall thickening of distal body and pylorus of stomach with extension into 1st and 2nd part of duodenum, with loss of fat planes with pancreatic head. Maximum tumor thickness was 2.1 cm extending over a length of 6.3 cm. There were no significant regional nodes or distant metastases. In view of the locally advanced nature of the disease, he was planned for staging laparoscopy. However, before he could undergo the planned procedure, he developed sudden onset acute abdominal pain and visited emergency services of our hospital 7 days after the endoscopy procedure. His pulse rate was 110/min with diffuse abdominal guarding and tenderness, ECOG 2 status. He was resuscitated with IV fluids, and broad-spectrum antibiotics were administered. Contrast CT of the abdomen was performed which showed free fluid in the abdomen and extra luminal air with tumor infiltration of the head of pancreas (Fig. 1a) and the distal bile duct (Fig. 1b). He was planned for emergency exploration with a diagnosis of gastric tumor perforation in the same evening. On exploration, a 2 × 2-cm tumor perforation at antrum was identified (Fig. 2); also, the tumor was found to infiltrate the pancreatic head and distal bile duct, behind the 1st part of duodenum (Fig. 3). Based on the limited amount and degree of contamination, lack of bowel edema, and absence of pus pockets or flakes formation, the perforation appeared to be a very recent event. In view of his young age, absence of co-morbidities, lack of hemodynamic instability, and the absence of peritoneal or gross lymph nodal disease, decision was taken to proceed with D2 radical distal subtotal gastrectomy with pancreatico-duodenectomy (PD). Surgery lasted for 260 min with a blood loss of 600 ml. He was monitored in the intensive care unit for a day after surgery and was discharged after an uneventful course on post-operative day 9. Final pathology report showed signet ring cell adenocarcinoma with involvement of pancreatic head and peri-pancreatic soft tissues with none of the 23 excised nodes involved—pT4N0. He received 8 cycles of adjuvant chemotherapy (capecitabine–oxaliplatin) beginning 2 weeks after discharge and is disease free at 18 months of follow-up.

Fig. 1.

a Circumferential antropyloric tumor with pancreatic head invasion (red arrow). Free fluid and extra luminal air suggestive of perforation. Naso-jejunal tube seen in the antrum at the level of the tumor. b Distal bile duct invasion by tumor (blue arrow)

Fig. 2.

Tumor site perforation in antrum measuring 2 × 2 cm

Fig. 3.

Growth infiltrating the head of pancreas

Discussion

Gastric cancer (GC) is associated with a poor prognosis with reported overall 5-year survival rates between 20 and 25% across all stages [4]. Surgical resection is the principal therapy for GC, as it offers the only potential for cure. The aim of surgery is to offer a resection with negative margins and adequate D2 lymph node dissection [5].

Pancreatic head involvement by gastric cancer is considered as locally advanced, inoperable disease, and patients are often offered only palliative chemotherapy. In exceptional situations, few selected patients may have a chance of receiving curative treatment. Ideally, such patients should be evaluated for a radical surgery at high volume centers, with or without multivisceral resection when there is likelihood of achieving R0 resection, which can lead to improvement in survival.

There is retrospective data to suggest that selected group of distal gastric cancers with isolated pancreatic head involvement in absence of peritoneal and gross lymph nodal disease can be offered curative surgery, i.e. radical gastrectomy with pancreatico-duodenectomy (PD) to achieve R0 resection, and this can lead to improvement in survival. Roberts et al. [6] have reviewed all case series documenting PD for gastric cancers. Even though surgical morbidity was increased with PD, some case series within the review have documented an improvement in survival compared with retrospective cohorts of non-curative resections. Due to the non-homogenous nature of cases, identification of factors associated negatively with survival in extended surgery could not be identified from this study. In another series of extended resections for GC, multi-organ resection was not a predictor for survival compared with GC not requiring multi-visceral resections [7].

In the study by Fukuchi et al. [8], non-curative factors included adjacent organ invasion (T4b), peritoneal disease or positive peritoneal cytology, and liver metastasis. Unresectable GC patients initially exhibiting only one non-curative factor, who underwent R0 resections after good response to chemotherapy, were shown to have a significant improvement in overall survival (OS). Another series has shown 34% 5-year survival in such patients [9]. Pancreatico-duodenectomy itself is fraught with multitude of complications, which justifiably increases when combined with radical gastrectomy. In the review by Makuuchi et al. [10], incidence of postoperative complications ranged from 23 to 74% when PD was carried out for gastric cancers. The most common morbidity was pancreatic leak. Mortality figures in various studies ranged from 0 to 12.5% in the same review.

The case in discussion here had two adverse factors, i.e., locally advanced gastric adenocarcinoma with duodenal and pancreatic head involvement and tumor perforation, both of which could have prevented a curative resection.

Patients with GC presenting with perforation require emergency exploration. The intraoperative management depends on patient’s co-morbidity profile, hemodynamic condition, and extent of disease. Based on these factors, surgical management can range from peritoneal lavage and closure of perforation with placement of drains to palliative or curative resections. Tumor perforation does portend a poor prognosis; however, patients who undergo curative resection do get improvement in OS. In 155 perforated gastric cancers, 30% 5 year survival was demonstrated in stage III tumors [11]. Despite perforation indicating poorer outcome, it should not detract one from curative surgery as many perforations do occur in early stage disease as well [1, 12].

The combination of perforation in GC with pancreatic head invasion being managed by PD has not been reported yet. Emergency PD is performed in extenuating circumstances of pancreatic trauma, non-trauma bleeding, and endoscopic perforations [13, 14]. However gastric cancer has not been listed in the performed indications. Amongst malignancies, emergency PD was carried out for lymphoma in the case reported by Stratigos et al. [15] for a bleeding duodenal lesion.

Our patient could have been managed differently by exploration and lavage with perforation closure or palliative gastric resection, as the mortality and morbidity of performing radical gastrectomy with pancreatico-duodenectomy in an emergency situation with perforation peritonitis is considerable. However, it would have directed him for a palliative treatment with inferior long-term outcome. Considering his young age, overall health status, hemodynamic stability, and absence of other adverse factors apart from pancreatic head involvement (peritoneal disease and high lymph node burden) and with our high volume pancreatic surgery experience [16], we took a calculated risk of doing radical surgery in emergency which benefited the patient as he is doing well and is disease free at 18 months of follow-up. However, such extended and multi-visceral resections for locally advanced GC should exceptionally be offered, only to well-selected patients at high volume centers, so as to get the desired benefit in improving their long-term survival while keeping the operative morbidity and mortality to minimum.

In effect, one has to balance the improved survival with curative resection versus the increased morbidity associated with extended and multi-visceral resections especially in emergent settings. In carefully selected patients, this balance can be skewed towards a survival benefit.

Conclusion

Pancreatic head involvement in the absence of other factors that preclude resection viz. heavy nodal burden, vascular involvement, or peritoneal disease, should be considered for curative resection after response to neoadjuvant therapy in elective setting. Also, radical gastrectomy with pancreatico-duodenectomy remains a viable option in emergency setting in perforated gastric cancer with pancreatic involvement, although in exceptional cases; as it can lead to improvement in survival in carefully selected cases at high volume centers.

Authors’ Contributions

a. Mufaddal Kazi: Design and writing of the manuscript

b. Manish Bhandare: Performed the surgery. Involved in design and review of the manuscript

c. Vikram Chaudhari: Critical review and final approval of the draft

d. Shailesh Shrikhande: Critical review and final approval of the draft

e. Sagar Kurankar: Involved in surgery and decision making.

Availability of Data and Material

All data was retrieved from electronic medical records of the hospital.

Compliance with Ethical Standards

All procedures followed were in accordance with the ethical standards of the responsible committee (institutional and national).

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

Ethics committee approval was not required for the case report.

Consent for Publication

Consent for publication from the patient was obtained.