Middle segment pancreatectomy: Does it deserve a second chance? Report of three cases and review of the literature

Abstract

Introduction and importance

Meddle segment pancreatectomy (MP) is a parenchyma-sparing surgical procedure that has recently been proposed for treatment of benign or attenuated malignant tumors. However, this procedure is not fully recognized.

Case presentation

We herein report 3 patients undergoing MP for tumors of pancreas body and tail. The first patient was a 38-year-old woman with a neuroendocrine tumor, the second was a 42-year-old woman with a serous cystic neoplasm and the 3rd patient was 57 years old, with mucinous cystadenoma.

A MP with spleen preservation was performed in the 3 patients, with ligatures of splenic vessels in the first patient. Only one patient developed a pancreatic fistula, which was managed medically. In our 3 patients, no endocrine or exocrine insufficiency was observed, but the first patient showed a disease recurrence with liver metastasis 3 years after surgery.

Clinical discussion

Middle pancreatectomy can not only avoid the pancreatic effects of extensive resections, but it is also a technique that has a very low operative or postoperative mortality rate.

Conclusion

MP is a feasible, safe procedure with several advantages, but which unfortunately remains rarely practiced.

Highlights

• •Medium pancreatectomy is currently a well codified intervention that allows pancreatic parenchyma preservation
• •Medium pancreatectomy has benefit in preserving the pancreas endocrine and exocrine functions.
• •Medium pancreatectomy has a very low operative or postoperative mortality rate.

1. Introduction

Modern medical imaging developments have made it easier to detect pancreatic tumors [5]. This increased detection rate has led to more pancreatic resections in recent years. In the past, cephalic duodenopancreatectomy (DP) or distal pancreatectomy (PD) was performed, even for indolent or benign lesions, resulting in excessive resection of normal pancreatic tissue, and causing impairment of the endocrine and exocrine functions of the pancreas [1]. Studies grouping PD and DP showed that 17 %–24 % of patients are diagnosed with pancreatic endocrine insufficiency after surgical resection, and the incidence of exocrine pancreatic insufficiency following distal pancreatectomy and a pancreatico-duodenectomy are documented at 19–80 % and 56–98 % respectively. Recently, limited resections have been proposed as alternatives to radical ones in view of cumulative clinical advantages of these types of resections [1]. From a practical point of view, preserving the pancreatic function for benign or reduced malignancy tumors located in the isthmus or the pancreatic body and which are not able to be treated by enucleation is very difficult for surgeons. In such cases where extensive resection must be avoided a middle segment pancreatectomy (MP) is indicated [2], [3]. The surgical technique starts with a wide opening of the gastrocolic ligament preserving the arch of the greater gastric tuberosity is necessary. The lower edge of the pancreas is freed in order to find the right dissection plane. At this stage the retroisthmic passage is performed. Pancreas will be sectioned by mechanical stapling, then the dissection is done from the right to the left with traction on the piece to release the lower and upper edges of the pancreas respectively. The pancreas is sectioned to the left of the lesion. A pancreatico-gastric or pancreatic-jejunal anastomosis must be performed, before finishing with an epiplooplasty and wide drainage. The objective of this rare intervention is to avoid a left pancreatectomy, in order to preserve the pancreatic function as much as possible, in particular endocrine function [2].

The aim of this study is to assess the feasibility and safety of this intervention by sharing the experience of our center through three cases of medial pancreatectomy, done in the visceral surgery department I of the military hospital of instruction Mohammed V from Rabat between January 2007 and January 2021, by using the PROCESS 2020 Guideline model [20].

A literature review will be included, with the purpose of better defining MP indications, showing its advantages, identifying the main drawbacks and limitations, and finally comparing our results with data from other studies.

2. Presentation of cases

2.1. First case

A 38-year-old patient with no particular medical and surgical history. The clinical history began a year before the diagnosis with vague pain associated with nausea and sometimes vomiting evolving without general condition alteration. The clinical examination was unremarkable apart from sensibility upon deep palpation of the epigastric region. The gastric fibroscopy was unremarkable. Ultrasound and Computed Tomography (CT) scan (Fig. 1) showed a well-limited, roughly rounded 4 cm pancreatic mass moderately enhanced by contrast agent. The mass has grown at the expense of the tail and body of the pancreas. The patient received an endoscopic ultrasound (Fig. 2) which showed a well-limited tumor of the body of the pancreas which overgrown the splenic vessels without invading them. There was no peri-pancreatic, celiac or superior mesenteric lymphadenopathy. Per echo-endoscopic biopsy showed a neuroendocrine tumor classified T3N0 whose non-secreting character was confirmed by biology. Surgical treatment was decided. The approach was a left subcostal incision widened towards the right. Faced with the very limited nature of the tumor to the pancreas body, we decided to make an MP. The splenic vein being intimately adherent to the mass imposed on us its ligature and its section, as well as the splenic artery. A pancreatojejunostomy anastomosis (PJ) on a Y (Fig. 3) was performed with a closure of the cephalic slice of the pancreas. The resection piece pathological examination revealed a well-displaced neuroendocrine tumor. The postoperative was simple, the drain was removed after five days, with the start of oral feeding, and the octreotide was also stopped after 5 days.

Fig. 1.

CT image showing a very limited tumor in the body of the pancreas.

Fig. 2.

Echo endoscopy showing pancreatic tumor (1) as well as splenic vein (2).

Fig. 3.

Operative view showing proximal pancreas (1) and anastomosis of the distal pancreas on Y-loop (2).

A postoperative angio scan showed good revascularization of the spleen by the short vessels and the left gastro omental artery (Fig. 4, Fig. 5). The length of hospital stay was 11 days. Regular checks during the first two years did not find any symptomatic exocrine pancreatic insufficiency defined by the absence of diarrhea or Steatorrhea, and the blood sugar levels never exceeded normal values. During the follow-up of the patient, a recurrence was revealed three years after the intervention by a control abdominal CT that showed the presence of metastases involving almost all liver segments (II, V, VI, VII, VIII) without signs of local pancreatic recurrence.

Fig. 4.

Postoperative angioscan showing splenic revascularization by splenic artery collaterals.

Fig. 5.

Postoperative angioscan showing venous drainage of the spleen by short vessels (1) and left gastroepiploic vein (2).

After having progressed under a 1st line of chemotherapy based on Sunitinib (Sutent®), the patient is currently placed on a 2nd line based on 5FU, Doxorubicin, Bevacizumab (Avastin®) with hepatic lesions stabilization.

2.2. Second case

A 42-year-old woman with chronic depression. She is also on long-term corticosteroid therapy for rheumatoid arthritis. She had abdominal pain with vomiting episodes. The CT scan revealed a cystic formation measuring 49.2 × 44.9 mm of the pancreas body with lipasemia. MP was indicated in view of the very lesion nature. A left subcostal incision widened towards the right was made. The pancreas consistency was normal, and the Wirsung was not dilated, and there was no evidence of invasion. The excision piece measured 8 × 5 × 2 cm with several cysts, weighing 50 g. Pathology concluded with a serous cystadenoma. On the 3rd postoperative day, the patient presented a respiratory discomfort with abdominal pain without fever. The amylase concentration in the drainage fluid was 8 times higher than normal. The CT scan showed partial anastomotic leakage of the PJ anastomosis with a perianastomotic collection and a heterogeneous appearance of the caudal pancreas (Fig. 6). The diagnosis of pancreatic fistula was retained and the patient was placed on sandostatin® for 5 days with a good clinical and radiological outcome. The hospital stay length was 21 days. The patient showed an adrenal insufficiency one month later because of cessation of corticosteroid therapy, which progressed well under medical treatment.

Fig. 6.

CT image showing PJ anastomosis leakage with peri anastomotic collection.

2.3. Third case

A 57-year-old patient, diabetic on insulin for 20 years, hypertensive for 15 years and followed for valvular heart disease under treatment. For 3 years, she has had chronic epigastralgia accompanied by vomiting, with general condition decline and a weight loss estimated at 10 kg. Abdominal ultrasound noted a multi-compartmental cystic formation measuring 42 × 28 mm in the pancreas body. Gastric fibroscopy showed antral erythematous gastritis which would explain the clinical symptomatology. The abdominal CT scan showed a cystic mass of the pancreas body with Wirsung dilation suggesting a papillary intraductal mucinous tumor or a mucinous cystadenoma (Fig. 7). Given the nature of the lesion, which was limited to the body without invasion signs of the splenic vessels, MP with conservation of the spleen was decided. The approach was midline supraumbilical incision. The continuity restoration was ensured by a PJ anastomosis. The pancreas had a firm consistency with Wirsung Dilatation. The excisional piece measured 5.7 × 3 × 3.6 cm and weighed 45 g. The pathologist concludes with a mucinous cystadenoma. The postoperative was simple, and the patient was discharged from the hospital 06 days after. The patient is still on insulin for her diabetes and does not present clinical signs of exocrine pancreatic insufficiency. She was reoperated 10 months after for an incisional hernia, and her preoperative blood test showed a fasting blood glucose level of 1.2 g/l, and her glycated hemoglobin was around 7 %.

Fig. 7.

CT image showing a tumor of the pancreas body with Wirsung dilatation.

3. Discussion

When looking at the pancreatic surgery progress, there is a current trend towards parenchymal sparing rather than radical resection. A middle segment pancreatectomy is indicated in the case of small (less than 5 cm) pancreatic benign and low malignancy risk lesions, including solitary metastases, located in the neck or adjacent body of the gland, not needing extensive lymphadenectomy, and untreatable with enucleoresection [4]. This procedure has shown its benefit in preserving the pancreas endocrine and exocrine functions, without compromising oncological results for serous or mucinous cystadenomas, solid pseudopapillary tumors and neuroendocrine tumors [4]. Furthermore, MP is a good indication for pancreatic trauma, and chronic pancreatitis.

Pancreatic parenchyma preservation enabled by MP prevents more than 95 % of endocrine insufficiencies causing insulin-requiring diabetes, and more than 92 % of exocrine insufficiency [5]. During DP, exocrine insufficiency percentage is around 60 % of cases regardless of continuity restoration mode [6] and diabetes is observed in 0 to 7 % of cases Likewise, PD has about 10 % risk of diabetes [6], [7], [8].

Muller et al. compared 40 patients who had undergone central pancreatectomy with 2 other samples who had Pylorus-preserving pancreaticoduodenectomy Whipple or distal pancreatectomy, noticed no de-novo diabetes in the first group, against 16 and 27 % in the two other groups respectively. Exocrine insufficiency was also less (16 %), compared to the pp-Whipple 61 % and distal pancreatectomy 26 % group [9].

In 2013, Cheng et al. grouped in its review all the patients (22 patients) who had MP between 1999 and 2011 in the USA. Six patients presented an exocrine insufficiency (27 %), and 31 % of this population worsened their old diabetes or completely developed new diabetes [4].

Crippa et al. with his 100 patients' series that had a MP, noticed an exocrine insufficiency not exceeding 5 %, and an endocrine insufficiency of 4 % [10]. Another series of 100 patients studied by Goudard et al. in 2014, showed similar results with a 6%exocrine insufficiency rate, and 7 % endocrine insufficiency (2 % de novo diabetes and 5 % worsening of the diabetic condition) [11]. Xiao et al. conducted a systematic review and meta-analysis, in aim to compare the clinical outcomes of central pancreatectomy with distal pancreatectomy and pancreaticoduodenectomy. In this analysis, the percentage of endocrine and exocrine insufficiency was 4 % and 5 % respectively, compared to 17 % and 29 % after PD, and 24 % and 17 % after DP [12].Robotics introduction for this kind of resection reduced the rate of novo diabetes to 0.3 % [13]. In our 3 patients, no endocrine or exocrine insufficiency was observed.

Middle pancreatectomy not only avoids the pancreatic effects of extensive resections, but it is also a technique that has a very low operative or postoperative mortality rate of less than 1 %, which is identical to PD but much lower than DP which has a death rate of almost 5 % [14]. This rate of less than 1 % was reported by several teams, and confirmed by multiple reviews of the literature: Crippa et al. 0 %, Iacono et al. 0.8 %, Xiao et al. 0.5 %, Santangelo less than 1 %, dragomir 0.6 % [15], [16], [17].

Goudard et al. and Muller et al. observed somewhat higher rates, 3 % and 2.5 % respectively, but still lower than the death rates observed during DP [16], [18].

In robotics air, Rompianesi reported in his review a negligible mortality rate for more than 256 MP achieved by robot [13]. In our series, no patient died.

MP also makes it easier to preserve the spleen. Although this preservation is envisaged in patients proposed for a DP, it becomes technically more delicate, and increases the operating time, and consequently it is not always carried out. Splenectomy can cause multiple complications, including generalized infection, and it also increases the thrombotic risk and eliminates the spleen immune role [9].

Splenic vessels preservation also allows the prevention of the pancreatojejunal anastomosis ischemic complications, and the tail of the pancreas infarction [17].

Some recent studies also suggest a role of spleen cells in the onset of a novo diabetes. These studies noted a higher rate of onset diabetes in patients who had pancreatic resection associated with splenectomy, compared to other patients in whom the spleen was preserved [4].

In our series, only one patient underwent ligation of the splenic vessels, given their intimate contact with the mass, but without splenectomy.

Like any other surgical technique, MP has complications directly attributed to it [17], [18].

MP presents more pancreatic fistulas (PF), between 0 and 65 %, given the presence of 2 anastomoses, and therefore more morbidity by comparing it to PD (0 to 20 %) and DP (2 to 32 %) [15], [18]. We observed a single pancreatic fistula in our context, which represents 33.33 % of our series. This figure is comparable to the literature. Cheng et al., described a morbidity rate of 40.9 %, with 22.7 % pancreatic fistulas [4]. Rates somewhat lower than those were observed by Xiao et al., who noted that half of their population experienced postoperative morbidity, with a FP rate of 35 % [12]. Goudard et al. found an even higher rate, with morbidity reaching 72 %, the main cause of which was anastomotic leak in 63 % of the population [11]. Laparoscopic and robotic surgery development had no impact on this rate of complications linked to MP, which is still very high, more than that of DP [13].

MP also has some limitations, which are seen above all in the excision of certain tumors. This surgical technique is indicated only in case of benign or reduced malignancy tumors [19]. Indeed lymph node dissection is insufficient in this intervention, to be able to extend the indications to malignant tumors such as adenocarcinomas. For benign tumors, extemporaneous histological examination is essential to verify the benignity of the lesion; in its absence, a re-intervention may be necessary if the definitive histological examination shows the existence of an invasive cancer [19]. At surgical level, one of the difficulties is to adapt the excision to the type of tumor in question, when we know that excessive pancreatectomy would risk unnecessarily exposing sequelae linked to a pancreatic parenchyma deficit, the main one being insulin-dependent diabetes.

The pancreatic surgery gold standard is still dominated by DP and PD, despite the existence of different series in the literature that have been performed over long periods, due to the relative rarity of MP indications.

4. Conclusion

Multiple series as well as our experience have shown that MP is a feasible and safe procedure with several advantages, but unfortunately it remains rarely practiced.

MP is currently a well codified intervention that allows pancreatic parenchyma preservation and therefore permits the pancreas to ensure its different functions, in particular endocrine. MP also avoids adverse post-splenectomy effects. MP could be considered for benign and low-grade malignant lesions of the neck and body of the pancreas in case of sufficient surgical experience. Multicentric prospective studies with large population are much needed to identify the true benefits of MP compared to PD and DP, especially in the mini-invasive area.

Abbreviations list

MP

Middle pancreatectomy

DP

Duodenopancreatectomy

PD

Distal pancreatectomy

CT

Computed Tomography

PJ

Pancreaticojejunostomy

PF

Pancreatic fistula

Consent of publication

Written informed consent was obtained from the three patients for publication of those cases and accompanying images. Copies of the written consents are available for review by the Editor-in-Chief of this journal on request.

Ethical approval

Ethical approval is exempt/waived at our institution.

Funding

N/A.

Guarantor

Hamza Sekkat.

CRediT authorship contribution statement

AAA designed the study and participated in drafting the manuscript. HS, KA and JEH collected the data and wrote the first draft of the manuscript. AM and NN participated in the study design and critically reviewed the manuscript. MY, AE and AAA critically reviewed the manuscript. All authors approved the final version of the manuscript.

Conflict of interest

N/A.