|Laparoscopic versus open gastrectomy with D2 lymph node dissection for gastric cancer: a meta-analysis
|Wei HB, Wei B, Qi CL, Chen TF, Huang Y, Zheng ZH, Huang JL, Fang JF
The review found that laparoscopic gastrectomy with D2 lymph node dissection for gastric cancer caused less blood loss, less pain, fewer postoperative complications and faster recovery of bowel function than open radical gastrectomy. Rates of survival and lymph nodes harvested were similar. Due to major limitations in the review, including lack of randomised evidence, these conclusions require very cautious interpretation.
To evaluate the effectiveness and safety of laparoscopic gastrectomy with D2 lymph node dissection compared with open radical gastrectomy for treating gastric cancer.
PubMed, EMBASE and Currents Contents were searched from 2001 to April 2010. The search was limited to studies published in English. Search terms were reported.
Studies that compared the effectiveness and safety of D2 lymph node dissection versus open radical gastrectomy for gastric cancer were eligible for inclusion. Studies were required to report at least four of the outcomes: safety outcomes (operative time, blood loss, harvested lymph nodes), effectiveness outcomes (duration of analgesic medication, first flatus day, hospital stay), postoperative complications (such as wound infection, duodenal stump leakage, anastomotic leakage, postoperative ileus, total complications) and/or cumulative survival rate. Studies using variations of standard laparoscopic techniques were excluded. Most of the included trials were conducted in Asia (Korea, Taiwan, China, Japan). Follow-up duration ranged from 11 months to five years.
The authors did not state clearly how many reviewers performed study selection.
Assessment of study quality
Aspects of quality apparently assessed included study design and comparability of the two groups at baseline. The assessment was conducted independently by two reviewers.
Weighted mean differences (WMDs) were extracted or calculated for continuous outcomes and Peto odds ratios (ORs) for dichotomous outcomes, each with 95% confidence intervals (CIs).
Data extraction were conducted independently by two reviewers.
Methods of synthesis
Studies were combined to calculate pooled odds ratios and weighted mean differences, each with 95% CIs. Heterogeneity was assessed using the Χ² test. Fixed-effect models were used unless there was significant heterogeneity, in which case random-effects models were used. Funnel plots were used to assess publication bias.
Results of the review
Ten studies were included in the review (1,039 participants, range 51 to 211), one randomised controlled trial (RCT) (211 participants) and nine case-control studies (828 participants). The two groups in individual studies did not differ significantly at baseline in age, sex, body mass index or disease stage.
All analyses apart from cumulative survival included all 10 studies. Operative time was significantly longer in the D2 lymph node dissection group than the open radical gastrectomy group (WMD 57.14 minutes, 95% CI 38.12 to 76.150, Χ² p<0.00001) and blood loss was lower (WMD -114.98, 95% CI -160.44 to -69.52, Χ² p<0.00001). There was no significant difference between the groups in lymph node retrieval rates (Χ² p=0.003).
The D2 lymph node dissection group had a significantly shorter duration of analgesia (WMD -0.89, 95% CI -1.45 to -0.32, Χ² p<0.00001) and hospital stay ( WMD -3.27, 95% CI -4.54 to -2.00, Χ² p<0.00001). Passage of flatus was significantly earlier in this group (WMD -0.84 days, 95% CI -1.25 to -0.43, Χ² p<0.00001).
The D2 lymph node dissection group had a lower rate of postoperative complications (OR 0.56, 95% CI 0.32 to 0.95, Χ² p=0.02), wound infections (OR 0.42, 95% CI 0.22 to 0.83, Χ² p=0.86) and postoperative ileus (OR 0.26, 95% CI 0.09 to 0.74, Χ² p=0.97). There was no significant difference between the groups in rates of duodenal stump or anastomotic leakage. Cumulative survival at the longest duration of follow-up (11 to 60 months) did not differ significantly between the two groups (six studies, Χ² p<0.97).
Most analyses showed significant statistical heterogeneity. There was no strong indication of publication bias.
D2 lymph node dissection for gastric cancer caused less blood loss, less pain, fewer postoperative complications and faster recovery of bowel function than open radical gastrectomy. Rates of survival and lymph nodes harvested were similar.
The objectives and inclusion criteria of the review were clear. The search was restricted to three databases and to articles published in English, so it was possible that some studies were missed. Formal testing did not suggest that the review was subject to publication bias. Steps were taken to limit reviewer bias by having two authors independently extract data and assess study quality; the process used for study selection was not described.
The review was limited by poor reporting, with little or no information about validity assessment methods, study quality, or clinical and demographic characteristics of study participants. Units of analysis were not always clear and there was some discrepancy between the text and forest plots. These factors made it difficult to assess the applicability and reliability of the reported results. It was doubtful that it was appropriate to combine randomised and case-control data in meta-analysis given the relatively poor quality of the case-control design. There was no indication of how treatment was allocated in the case control studies and a strong likelihood of selection bias, which was borne out by the discrepancy in size between the groups in some studies. As the authors noted, the studies were probably too small for baseline differences between participants to be detectable.
Statistical heterogeneity was appropriately assessed, but where high levels of heterogeneity were evident (nearly every analysis), the heterogeneity was neither explored nor explained. As the authors noted, there were only a few trials, most were small, only one was randomised and duration of follow up was inadequate to show differences in long-term survival.
Major limitations in the review, which included a lack of randomised evidence, mean that the authors' conclusions require cautious interpretation.
Implications of the review for practice and research
Practice: The authors did not state any implications for practice.
Research: The authors stated that large well-designed multicentre RCTs with long-term follow-up were needed to compare D2 lymph node dissection and open radical gastrectomy for gastric cancer (especially advanced gastric cancer).
National Natural Science Foundation of China.
Wei HB, Wei B, Qi CL, Chen TF, Huang Y, Zheng ZH, Huang JL, Fang JF. Laparoscopic versus open gastrectomy with D2 lymph node dissection for gastric cancer: a meta-analysis. Surgical Laparoscopy, Endoscopy and Percutaneous Techniques 2011; 21(6): 383-390
Subject indexing assigned by NLM
Analgesics /therapeutic use; Blood Loss, Surgical; Clinical Trials as Topic; Female; Gastrectomy /methods; Humans; Laparoscopy /methods; Length of Stay; Lymph Node Excision /methods; Male; Pain, Postoperative /etiology; Recovery of Function; Stomach Neoplasms /surgery; Survival Rate
Database entry date
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.