CRD summary

This review concluded that robot-assisted laparoscopic fundoplication was a feasible and safe alternative to surgical treatment for gastro-oesophageal reflux disease, despite longer operating times, increased hospital stay and higher costs. The evidence appeared to support this conclusion, but a cautious interpretation is advised given the risk of missing data and the methodological limitations of the included trials.

Authors' objectives

To determine the feasibility and efficiency of robot-assisted versus conventional laparoscopic fundoplication for the treatment of gastro-oesophageal reflux disease in adults.

Searching

MEDLINE, EMBASE, the Cochrane Library and Science Citation Index were searched for published and unpublished studies in any language up to September 2009. Search terms were reported. Additional searches were carried out using the ‘‘related-articles’’ feature of PubMed and by screening the references in retrieved articles.

Study selection

Randomised controlled trials (RCTs), clinical controlled trials (CCTs) and trials that compared robot-assisted with conventional laparoscopic fundoplication for the treatment of gastro-oesophageal reflux disease in adults were eligible for inclusion in the review. Eligible primary outcomes were operating time, hospital stay, and rate of complications.

The Da Vinci robotic system was used in most of the included trials; other trials assessed the Mona and the Aesop systems. The main fundoplication was Nissen; others were Toupet and Dor. Patients with a history of previous gastric surgery, giant hiatal hernia, general contraindications for laparoscopy, and psychiatric diseases were excluded from the studies. Where reported, the mean age of included patients ranged from 39 to 57 years in the robot-assisted laparoscopic fundoplication groups and 38 to 54 years in the conventional laparoscopic fundoplication group. Participant body mass index ranged from 23.7 to 29.2 kg per m² in the robot-assisted laparoscopic fundoplication group and 24.8 to 30.0 kg per m² in the conventional laparoscopic fundoplication group. Included trials were published between October 2001 and June 2009; they were conducted in Belgium, the Netherlands, USA, Germany and Italy.

Two reviewers assessed each study for inclusion; disagreements were resolved through discussion. Authors were contacted if any missing data were needed for further inclusion assessment.

Assessment of study quality

Trial quality criteria were randomisation, concealment of allocation and blinding, incomplete outcome data, selective outcome reporting, differences in baseline characteristics and confounding bias. Each criterion were judged as yes, no or unclear.

It would appear that two reviewers assessed this information.

Data extraction

Two reviewers independently extracted the data. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported for dichotomous outcomes. Means with standard deviations were extracted for continuous outcomes or calculated from median and ranges if necessary. Data on costs were also extracted. Authors were contacted for missing data if required.

Methods of synthesis

Fixed-effect meta-analyses were used to calculate pooled odds ratios with 95% confidence intervals for complication rates; pooled weighted mean differences (WMD) or standardised mean differences (SMDs) with 95% confidence intervals were used for operating time and length of hospital stay. Statistical heterogeneity was assessed using I² and Χ²; values of I² over 50% were reported as heterogeneous. Where significant heterogeneity was identified, a random-effects analysis was used.

Sensitivity and subgroup analyses were used to investigate potential sources of heterogeneity including difference in trial quality.

The authors reported that there were too few trials to assess publication bias; all of the included trials were published in English.

Results of the review

Eleven trials (n=533) including seven RCTs and four CCTs were reported as meeting the inclusion criteria. Most of the included trials did not report sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other potential threats to validity. Three trials had adequate randomisation methods and allocation concealment. Follow-up ranged from 1.12 to 48 months, where reported.

There was a statistically significant difference in favour of robot-assisted laparoscopic fundoplication compared with conventional laparoscopic fundoplication for postoperative complication rate (OR 0.35, 95% CI 0.13 to 0.93; six trials; I²=0%; fixed-effect model). The total operating time was significantly longer for robot-assisted laparoscopic fundoplication (WMD 24.05 minutes, 95% CI 5.19 to 42.92; eight trials; I²=98%; random-effects model). There was no significant difference between the two groups for perioperative complication rate (seven trials) and length of hospital stay (six trials). Pooled analyses were not performed for set-up time (as only one trial reported appropriate data) and effective operating time (due to the high level of statistical heterogeneity in random-effects analysis, I²=95%).

The results of sensitivity and subgroup analyses were also reported, but in general did not significantly change the findings.

Cost information

Four trials reported direct medical costs for surgery (total operative costs, hospital stay expenses and total costs) in Euros (EUR), but no other direct non-medical or indirect costs were reported.

One trial indicated higher costs for robot-assisted laparoscopic fundoplication than for conventional laparoscopic fundoplication for total operative costs (1,534 EUR versus 763 EUR) and total costs (3,244 EUR versus 2,734 EUR), although hospital stay costs were similar (1,710 EUR for robot-assisted laparoscopic fundoplication versus 1,980 EUR conventional laparoscopic fundoplication).

One trial reported that cost consumption was equal between the two groups with increased experience (4,363.82 EUR versus 3,376.35 EUR; p=0.033).

One trial reported the annual costs per patient for the investment and maintenance, with a five-year paying-off period and a 5% per year depreciation. The costs were similar between the robot-assisted laparoscopic fundoplication and conventional laparoscopic fundoplication groups for hospital stay (2,249 EUR versus 2,242 EUR), pharmacy (167 EUR versus 202 EUR), and surgical procedure (1,553 EUR versus 2,525 EUR). Costs of nurses’ salary, investment, and maintenance were greater for robot-assisted laparoscopic fundoplication, but the costs of disposable and reusable materials were higher for conventional laparoscopic fundoplication.

One trial reported unit costs. A statistical difference was found in disposable instruments and total costs (p<0.001). However, there was no statistic difference in total operative cost and hospital stay expenses; total operative cost was 367 EUR per hour and hospital stay cost was 300 EUR per hour.

Authors' conclusions

Robot-assisted laparoscopic fundoplication was a feasible and safe alternative to surgery for the treatment of gastro-oesophageal reflux disease, but was associated with longer operating times, increased length of hospital stay and higher costs.

CRD commentary

This review assessed a well-defined research question. Although literature searches and study inclusion were not limited by publication status or language, there may still have been a risk of publication and language bias as all of the included trials were published in English. Attempts were made to reduce the risk of reviewer error and bias by the use of two reviewers throughout the review processes.

The methodological quality of the included trials was assessed, but some of the trials were not RCTs and a number had other methodological limitations, which may have impacted on the reliability of the data. The small number of RCTs and often small sample sizes were acknowledged by the authors as potential limitations. Statistical heterogeneity was assessed; there was some evidence of heterogeneity between trials. Similarly, there were some clinical differences between the trials, particularly the study populations and methodological quality, although the authors made attempts to investigate these further using additional analyses.

Overall, the evidence appeared to support the authors' conclusions, but a cautious interpretation is advised given the risk of missing data and the methodological limitations of the included trials.

Implications of the review for practice and research

Practice: The authors stated that robot-assisted laparoscopic fundoplication is a feasible and safe alternative to surgical treatment for the treatment of gastro-oesophageal reflux disease. However, its application is likely to be limited by the prolonged operating time required and associated higher costs.

Research: The authors stated that additional large multi-centre randomised controlled trials (that conform to the standards of the 2001 CONSORT statement) are required to compare the effects of robot-assisted laparoscopic fundoplication and surgical treatment for gastro-oesophageal reflux disease. Trials should particularly focus on long-term follow-up and assess the satisfaction scores of all participants.

Funding

Not stated.

Bibliographic details

Mi J, Kang Y, Chen X, Wang B, Wang Z. Whether robot-assisted laparoscopic fundoplication is better for gastroesophageal reflux disease in adults: a systematic review and meta-analysis. Surgical Endoscopy 2010; 24(8): 1803-1814. [PubMed: 20112116]

Original Paper URL

http://www.springerlink.com/content/0077j1h643517515/

Indexing Status

Subject indexing assigned by NLM

MeSH

Adult; Fundoplication /methods; Gastroesophageal Reflux /surgery; Humans; Laparoscopy; Robotics; Time Factors

AccessionNumber

12010007913

Database entry date

05/10/2011

Record Status

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.