Large gastrointestinal stromal tumours of the stomach: Is laparoscopy reasonable?

BACKGROUND: Laparoscopic resection (LR) offers significant advantages compared to open resections for gastric gastrointestinal stromal tumours (GISTs). We aimed to evaluate whether LR outcomes jeopardised short and long-term outcomes of patients with large GISTs. PATIENTS AND METHODS: Among 50 patients undergoing surgery for gastric GISTs, 12 underwent LR for large GISTs (>5 cm). Their characteristics, perioperative results and survival were retrospectively compared to those of 22 patients who underwent LR for 'small GIST'.
RESULTS: The two groups were similar regarding demographics, rate of wedge resection and mean blood loss. No patient required transfusion or conversion. Operative time was significantly increased in the 'large GIST' group (160 min vs 112 min, P = 0.001). Mean tumour size was significantly lower in the 'small GIST' group (8.4 cm vs 2.4 cm, P = 0.0001). Resection margins were negative. The mortality rate was nil and the overall morbidity rates was similar in both groups. Median length of hospital stay was significantly increased in the 'large GIST' group (7 days vs 5 days, P = 0.004). Median follow-up was 47 months and one patient in the 'small GIST' group developed recurrence and died during follow-up 11 years after surgery. No patient died during follow-up.
CONCLUSIONS: LR for large GISTs is safe and technically feasible and does not negatively influence the oncologic course. Prospective randomised trials should be performed before using this approach in routine surgical care.

INTRODUCTION

Gastrointestinal stromal tumours (GISTs) can arise from the gastrointestinal tract, with the highest incidence in the stomach (50-60%). All gastric GISTs >1 cm should be considered as potentially malignant, and the recommended procedure for non-metastatic forms includes the removal of the visible tumour without spillage and with macroscopic free margins of resection (1-2 cm).[12345] With comparable oncological results, the laparoscopic approach offers significant advantages in terms of post-operative pain and length of stay compared to open surgery[23456] and is increasingly adopted in many centres.[7] However, there is a controversy about the impact of tumour size on both the technical feasibility of laparoscopic resection for gastric GIST and its oncological adequacy in terms of tumour rupture and resection margins.[8] While the size limit for laparoscopic GIST resection is continuously being modified, larger tumours have generally been approached through an open approach and frequently involve more extensive resections. The perceived advantages or equivalence of the laparoscopic approach when compared with the open approach may therefore be a result of this size selection bias, and improved outcomes associated with laparoscopy may be surrogates for less extensive resections.

Therefore, the aim of the present study was to evaluate whether the laparoscopic approach jeopardised the short- and long-term outcomes of patients with large (>5 cm) gastric GISTs when compared to smaller ones.

PATIENTS AND METHODS

Patient Selection

Between January 2003 and September 2013, all consecutive pathologically confirmed gastric GISTs operated on with a curative intent were selected. Preoperative investigations included complete blood function tests as well as routine oesophagogastroduodenoscopy, computed tomography (CT) scan and endoscopic ultrasonography. CT scans were performed to assess both tumour characteristics and extension work-up. Initially, preoperative biopsy of the tumour was performed on a case-by-case basis but was carried out routinely during the final 5 years of the study. In accordance with Privette et al.,[5] GISTs were divided according to site into type I (greater curvature), type II (prepyloric and antrum), and type III (lesser curvature and near the gastro-oesophageal junction).

Surgical Procedures

All resections for gastric GISTs were performed with curative intent. The choice between open or laparoscopic approach was made after careful examination of the previous medical and surgical history of each patient as well as the type and the site of the tumour. In the laparoscopic group, each patient was placed in the supine position with anti-Trendelenburg of 15° and right tilt. Carbon dioxide was insufflated through the first port with a pressure setting of 12 mmHg. A 0° camera was introduced through the umbilical port, and diagnostic laparoscopy performed. A second and third 15 mm port were inserted in the right and left upper abdominal quadrants, respectively, under direct vision. Three or four port sites were generally used to facilitate the gastric resection. Occasionally, identification of the tumour and its extent were assisted with gastroscopy. The gastric wedge resection was achieved using an endo-GIA stapler, and the tumour specimen was extracted using an Endocatch bag. No tumour enucleation was performed.

Studied Criteria

Clinical and histopathologic data were obtained from patient medical records. All intraoperative parameters including estimated blood loss with subsequent blood transfusion and duration of surgery were recorded. Mitotic rate was defined as the number of mitoses per 50 high-power fields (HPF), and tumour size was defined as the maximal tumour dimension in the resected specimen. Patients undergoing gastric resection for GISTs larger than 5 cm constituted the ‘large GIST group’. Post-operative complications were stratified according to the Clavien-Dindo classification,[9] which defines major complications by a score of 3 or higher. Both complications and operative mortality were considered as those occurring within 90 days of surgery or at any time during the post-operative hospital stay.

After discharge, patients were seen at the outpatient clinic for clinical examination, and morphological evaluation (including mainly computed tomography) 1 month after discharge, every 3 months for the first 2 years and every 6 months until the fifth postoperative year. Thereafter, patients were seen yearly, and follow-up was discontinued in the absence of recurrence after the 10th year.

Actuarial overall survival (OS) and disease-free survival (DFS) were calculated from the date of curative surgery to the first recurrence.

Statistical Analysis

Patient baseline characteristics are expressed as median (range) for continuous data (mean ± standard error of the mean are also indicated where appropriate), and as numbers with percentages for categorical data. Preoperative, operative and post-operative characteristics as well as long-term survival of patients were compared according to the tumour size (≥ vs <5 cm). Fisher's exact test or chi-square test was used to compare differences in categorical variables, and the Mann-Whitney test for continuous variables. Cumulative overall survival was determined using the Kaplan-Meier method and compared with the log-rank test. All tests were two-sided. P < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS® version 18.0 for Windows® software (IBM, Armonk, New York, USA).

RESULTS

Preoperative Characteristics

Fifty patients underwent gastrectomy for GIST including 26 females (52%) and 24 males (48%) with a median age of 63 years (range: 17 to 83). Sixteen (32%) gastric GIST patients underwent open resection while the 34 (68%) remaining were treated by laparoscopy [Table 1]. The preoperative characteristics of the laparoscopic group are detailed in Table 2. Among them, 12 (35%) patients with tumours >5 cm formed the “large GIST” group. Tumour was diagnosed incidentally in 12 (35%) patients, whereas the most common symptoms were abdominal pain (50%), bleeding (23%) and anaemia (18%). The tumours were most frequently located in greater (29%) and lesser curvatures (29%), and 19 (56%) patients had preoperative biopsy. There was no significant difference between the two groups regarding demographics. One patient had received imatinib mesylate as neoadjuvant chemotherapy.

Table 1

Demographic characteristics and preoperative data of patients undergoing open and laparoscopic gastric GIST resection

Table 2

Demographic and preoperative data of laparoscopic ‘small GIST’ and ‘large GIST’ groups in the selected population

Surgical procedures and post-operative outcomes

Fifteen (44%) patients had previous abdominal surgery without significant difference in the two groups (41% vs 50%, P = 0.72). Gastric resection was atypical in 28 (82%) patients without difference in the two groups. Intraoperative results are detailed in Table 3. No patient required conversion to open surgery. Although the median operative time was significantly longer in the ‘large GIST’ group (112 min vs 60 min, P = 0.001), the median estimated blood loss ranged 5-1300 mL, without any difference in the two groups (P = 0.86). No patient required transfusion intraoperatively. Tumour rupture or spilling did not occur in any of the two groups. Extraction of tumours was performed without morcellation, in a plastic bag. All tumours in the ‘large GIST’ group were extracted through a Pfannenstiel incision, the length of which varied according to the size of each tumour.

Table 3

Operative results, pathological data and perioperative outcomes of laparoscopic ‘small GIST’ and ‘large GIST’ groups in our patient population

The mortality rate was nil and 3 (9%) patients experienced post-operative complications, mainly pulmonary complications. All post-operative complications are detailed in Table 3. The overall median length of hospital stay in the laparoscopic group was 6 days (range 3-16 days), and it was significantly higher in the ‘large GIST’ group (7 days vs 5 days, P = 0.004).

Tumour characteristics

The overall mean tumour size in our study was 4.5 cm, 7.9 cm in the open resection group and 4.3 cm in the laparoscopic group (P = 0.003). Mean tumuor size was significant higher in the laparoscopic ‘large GIST’ group compared to the ‘small GIST’ group (8.4 cm vs 2.4 cm, P = 0.0001). On microscopic analysis, 71% of tumours had a mitotic rate of fewer than 5 mitoses per 50 HPF, 12% had a mitotic rate between 5 and 10 mitoses per 50 HPF, and 18% had more than 10 mitoses per 50 HPF. There was a statistically significant difference between the laparoscopic ‘large GIST’ and the ‘small GIST’ groups regarding a mitotic rate higher than 10 mitoses per 50 HPF (41.7% vs 4.5%, P = 0.013). There was no statistically significant difference between the two groups regarding histologic type (P = 0.14) [Table 3]. All patients had negative resection margins (R0).

Long-term outcomes

The median follow-up in the laparoscopic group was 47 months (range 6-124 months). Two (4%) patients were lost to follow-up. One (3%) patient with large-GIST recurrence developed recurrence with peritoneal carcinomatosis 6 months after laparoscopic resection and was subsequently treated by both laparoscopic resection and chemotherapy; the patient is still alive 7 years after surgery. One ‘small GIST’ group patient died during follow-up 11 years after gastric surgery. Actuarial 1-year, 5-year and 10-year OS was 100% in both groups; 1-year, 5-year and 10-year DFS was 95%, 95% and 95%, respectively.

DISCUSSION

Laparoscopic gastric resection has the potential advantage of requiring smaller incisions and less bowel manipulation compared with open surgery. This reduces pain, facilitating earlier recovery of bowel function and decreased hospital stay.[10] To our knowledge, this is the first comparison of laparoscopic gastric resection in large and small gastric GISTs. Our data demonstrate that laparoscopic resection is safe and technically feasible in large GISTs and does not negatively influence the oncologic course or oncological safety.

The absence of conversion to open surgery in the present series and a conversion rate of 4.7% in other large series confirms the feasibility of the laparoscopic approach for large lesions.[10] Despite the fact that the sample size of our study was small, there was no significant difference in the incidence of overall or major complications between the large- and small-GIST groups. The few complications observed were essentially pulmonary complications mainly related to a high gastric localisation of the GIST (near the oesogastric junction or greater curvature). All complications were promptly treated without endangering the patient's life at any moment. However, in order to obtain similar results, the resection of such large lesions has to be performed by surgeons experienced in both laparoscopic and upper gastrointestinal tract surgery. At our department, surgeons are well trained in laparoscopic surgery, such as bariatric, hepatobiliary, and oesophageal/cardia cancer procedures. Finally, a prerequisite for performing gastric GIST surgery is an advanced knowledge of intracorporeal suturing, which is often necessary in such procedures. Laparoscopic resection of large gastric GISTs appears to be a reasonable surgical technique for appropriately skilled surgeons performing minimally invasive gastric resections.

Large gastric GISTs and tumours located in difficult locations, such as the oesogastric junction and posterior wall, are technically more difficult to resect, especially via the laparoscopic approach, with an increased risk of tumour spillage. In some difficult situations, endogastric procedures can be useful. For this reason, the Guidelines for Optimal Management of Patients with GIST suggested that laparoscopic resection should be limited to tumours less than 2 cm.[1] However, the mean tumour size for laparoscopic gastrectomy ranged 2.7-6.0 cm in recent meta-analysis[11] and was 8.4 cm in our large-GIST group. Given the low rate of both positive margins and local recurrence, it appears that this recommendation is not based on strong evidence. In the present series, the oncological safety of LR was confirmed, as shown by the comparable overall survival rates of the two groups. Hence, even though safe oncologic resection is the primary concern when considering LR for large GIST, the present study suggests that a laparoscopic approach for gastric GIST should not be limited to tumours smaller than 2 cm and could also be considered for tumours larger than 5 cm. Likewise, the oncological outcome in the present series is comparable to previous reports.[461112]

The present study has several important limitations, which should be taken into account when considering the results. First, although the patients were entered into the database in a prospective fashion, some of the outcome data were gathered through a retrospective review of the medical records and are therefore subject to bias. Indeed, these patients were not randomly selected with regard to the surgical approach, and gastric GISTs were significantly larger in the open group compared to the laparoscopic group. In addition, the sample size was limited to a 10-year study period. Therefore, these results need to be confirmed by high-quality randomised controlled trials with larger sample sizes that compare gastric resection for GISTs smaller and larger than 5 cm.

CONCLUSION

In conclusion, this study suggests that laparoscopic resection for large GISTs is safe and technically feasible and does not negatively influence the oncologic course or oncological safety. Nevertheless, prospective randomised trials should be performed before incorporation of these new data into routine surgical care.