Criteria for considering studies for this review
Types of studies
All published and unpublished RCTs comparing diagnostic laparoscopy with open appendicectomy or a wait and see strategy.
No limitation on language or publication status was applied. Open randomised clinical trials were included. RCTs that included women as part of the sample were included. Quasi-RCTs were not included.
Types of participants
RCTs that included premenopausal women who presented with acute lower abdominal pain, non-specific abdominal pain, right-sided pain or suspected appendicitis were included. Studies in which participants had a clear diagnosis of appendicitis were excluded.
Non-specific abdominal pain is defined as pain of less than seven days’ duration for which no immediate cause can be found after initial tests have been performed. Non-specific abdominal pain does not clearly require surgical intervention (Poulin 2000; Sanders 2006).
Changes were made to the published protocol to include women with suspected appendicitis, and studies where at least 75% of the participants were women of premenopausal age.
Types of interventions
Trials were included if they evaluated the management of premenopausal women with non-specific acute pain or suspected appendicitis using:
Laparoscopy is defined as a surgical procedure in which a laparoscope is used through the abdominal wall with the aim of visualising the pelvic and abdominal cavities to diagnose an underlying cause of pain. Typically, this procedure is performed within the first 72 hours of an in-patient stay.
Open appendicectomy is performed when a right iliac fossa incision is made and the appendix is excised after the muscular and peritoneal layers are opened.
A wait and see strategy is defined as close clinical observation combined with the use of laboratory tests or diagnostic imaging. It does not include laparoscopy, but it could include laparotomy.
Changes made to the published protocol included that the conventional strategy was replaced by 'wait and see' or by open appendicectomy.
Types of outcome measures
Outcomes measured were related to the effectiveness and safety of each strategy used in the management of non-specific acute abdominal pain or suspected appendicitis in premenopausal women.
Number of specific diagnoses before discharge: number of cases in which a specific diagnosis was reached before discharge, for each strategy studied.
Adverse events (AEs): any events that, in the opinion of the investigator, may adversely affect the rights, welfare or safety of participants in the study as a result of the application of a management method. Complications could be reported in the short term or over the long term. Although removal of a normal appendix is an AE, it was reported separately in the results section, as each participant could be counted only once within each study.
Total length of in-patient stay.
Mean operating time.
Return to normal activities.
Quality of life.
Cost-effectiveness, taking into account direct medical costs from the point of view of third-party payers or institutions.
Search methods for identification of studies
We followed the Menstrual Disorders and Subfertility Group methodology.
All reports that described RCTs of laparoscopy and acute abdominal pain were sought using the following strategy.
The Menstrual Disorders and Subfertility Group (MDSG) Specialised Register was searched by the Group’s trials search co-ordinator using the following terms: "laparoscopic" or "laparoscopic excision" or "laparoscopic imaging" or "laparoscopic dye" or "laparoscopic imaging" or "laparoscopic procedure" or "laparoscopic surgery" or "laparoscopic techniques" or "laparoscopy" or Title CONTAINS "laparoscopic" or "laparoscopic excision" or "laparoscopic imaging" or "laparoscopic dye" or "laparoscopic imaging" or "laparoscopic procedure" or "laparoscopic surgery" or "laparoscopic techniques" or "laparoscopy" AND the terms "acute" or "abdominal pain" or "pelvic pain" or "Pain-abdominal" or "pain-pelvic" or "ectopic pregnancy" or "pelvic inflammatory disease" or "Ovarian Cysts" or "ovarian cyst" or "acute" or Title CONTAINS "acute" or "abdominal pain" or "pelvic pain" or "Pain-abdominal" or "pain-pelvic" or "ectopic pregnancy" or "pelvic inflammatory disease" or "Ovarian Cysts" or "ovarian cyst" or "acute", in the titles, abstracts and keywords.
This register also contains unpublished trial abstracts, which were found by handsearching of 20 relevant journals and conference proceedings.
The search was updated from January 2010 to October 2013 in the Cochrane MDSG Specialised Register, MEDLINE, EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL) and PsycoINFO using the Ovid platform.
A previous search was carried out in the following databases using the Ovid platform.
MEDLINE (1980 to April 2010) (Appendix 1).
EMBASE (1980 to April 2010) (Appendix 2).
CINAHL (1980 to April 2010) (Appendix 3).
CENTRAL (1998 to October 2013) (Appendix 4); PsycINFO (1980 to October 2013) (Appendix 5); and the Cochrane MDSG Specialised Register (Appendix 6).
LILACS (Appendix 7) and SciELO for studies reported in Portuguese and Spanish (February 2013).
Both indexed and free text terms were used. The RCT filter from the MDSG was used.
We also searched the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/en/) search portal (last search May 2013) using the following search strategy: (Laparoscopic OR laparoscopy) AND women AND (abdominal OR pelvic OR appendicitis OR abdomen) AND pain.
Searching other resources
Citation lists from reviewed articles and other relevant publications were searched.
No restrictions, such as language, were applied.
Other strategies for locating studies included personal communication with manufacturers, experts and specialists working in the field and screening of conference proceedings.
Data collection and analysis
Data were analysed using Review Manager software (RevMan 5).
Selection of studies
Two review authors (from Gaitán H, Reveiz L, Elias VM) independently screened the titles and abstracts of trials identified by the search for inclusion based on the selection criteria outlined above. One review author is a content and methodology expert, and the other is a methodology expert.
The full text of an article was retrieved if there was any doubt as to whether the article should be excluded. Gaitán H obtained copies of the studies selected for inclusion and sent them to Reveiz L. Both review authors then independently assessed whether the studies met the inclusion criteria. If disagreement between the two review authors arose, a third review author (Farquhar C) reviewed the information to decide on inclusion or exclusion of a trial.
Further information was sought from the study authors when papers contained insufficient information to allow a decision regarding eligibility for inclusion in the review.
Data extraction and management
Data from studies that met the inclusion criteria were independently extracted by two review authors (Gaitán H, Reveiz L) using a data extraction form. Discrepancies were resolved by discussion.
Data that were extracted included the following.
Inclusion and exclusion criteria, clearly defined.
Baseline information on participants for comparable intervention and control groups at entry (eligibility criteria, age of women, duration of pain, temperature, white blood cell count, abdominal surgery history).
Location of the study.
Power calculation performed.
Method used to generate random allocation.
Methods used to maintain allocation concealment.
Types of interventions provided: type of diagnostic laparoscopic method (video laparoscopy) and type of conventional diagnostic strategy, such as use of laboratory tests, and accepted time of observation.
Other interventions in the groups under evaluation.
Numbers of women enrolled, randomly assigned, excluded after randomisation and analysed.
Outcomes stated in methods versus outcomes reported in results.
Use of any method of blinding of researchers to the intervention for evaluation of outcomes.
How outcomes such as time of hospitalisation before diagnosis, definitive diagnosis, adverse events, recurrent episodes of pain, length of in-patient stay and cost-effectiveness were defined.
Differences between groups for outcome assessment in terms of methods used to obtain the definitive diagnosis.
Time of follow-up of participants to measure outcomes: evolution in terms of recurrent or chronic abdominal pain.
How adverse event reports were validated.
Numbers of participants lost to follow-up in the two groups.
Use of intention-to-treat analysis.
Funding sources reported.
Ethical issues: use of signed informed consent; ethics approval.
This information was collated and presented in the tables Characteristics of included studies and Characteristics of excluded studies.
Assessment of risk of bias in included studies
Two review authors (Reveiz L, Gaitán H) independently assessed the risk of bias of each trial using a simple form and followed the domain-based evaluation as described in the Cochrane Handbook for Systematic Reviews of Interventions 5.1.0 (Higgins 2011). Review authors discussed discrepancies and achieved consensus on the final assessment.
We assessed the following domains as low, unclear or high risk of bias.
Generation of allocation sequence (checking for possible selection bias)
We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.
We assessed the method as:
low risk (any truly random process, e.g. random number table; computer random number generator); or
unclear risk (the trial was described as randomised, but the method used for allocation sequence generation was not described).
Allocation concealment (checking for possible selection bias)
We described for each included study the methods used to conceal the allocation sequence in sufficient detail and to determine whether the intervention allocation could have been foreseen in advance of, or during, recruitment, or changed after assignment.
We assessed the methods as:
low risk (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);
high risk (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth); or
unclear risk (trial was described as randomised, but the method used to conceal the allocation not described).
Blinding or masking (checking for possible performance and detection bias)
We described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We judged studies at low risk of bias if they were blinded, or if we judged that lack of blinding could not have affected the results. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed blinding methods as:
low risk, high risk or unclear risk for participants;
low risk, high risk or unclear risk for personnel; or
low risk, high risk or unclear risk for outcome assessors.
Incomplete outcome data (checking for possible attrition bias through withdrawals, dropouts, protocol deviations)
We assessed methods on outcome data as:
low risk (any one of the following): no missing outcome data; reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; for dichotomous outcome data, the proportions of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardised difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size; missing data imputed using appropriate methods;
high risk (any one of the following): reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardised difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size; ‘as-treated’ analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation; or
Selective reporting bias (reporting bias due to selective outcome reporting)
We described for each included study how we investigated the possibility of selective outcome reporting bias and what we found.
We assessed reporting methods as:
low risk (any one of the following): The study protocol is available and all of the study’s prespecified (primary and secondary) outcomes of interest in the review have been reported in the prespecified way, or the study protocol is not available, but it is clear that published reports include all expected outcomes, including those that were prespecified (convincing text of this nature may be uncommon);
high risk (any one of the following): Not all of the study’s prespecified primary outcomes have been reported; one or more primary outcomes are reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not prespecified; one or more reported primary outcomes were not prespecified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); one or more outcomes of interest in the review are reported incompletely so that they cannot be entered into a meta-analysis; the study report fails to include results for a key outcome that would be expected to have been reported for such a study; or
Free of other bias (bias due to problems not covered elsewhere in the table)
We described for each included study any important concerns that we have about other possible sources of bias (baseline imbalance, sponsorship bias, differential verification bias, partial verification bias and incorporation bias, bias of the presentation data, etc.).
Low risk of bias: The trial appears to be free of other components that could put it at risk of bias.
Unclear risk of bias: The trial may or may not be free of other components that could put it at risk of bias.
High risk of bias: Other factors in the trial could put it at risk of bias (e.g. no sample size calculation made, academic fraud, industry involvement, extreme baseline imbalance).
Measures of treatment effect
Dichotomous data were expressed as odds ratios (ORs) with 95% confidence intervals (CIs) and, when possible, were combined in a meta-analysis using RevMan 5 software. The OR has mathematically sound properties that are consistent with benefits or harms and work well in small samples with rare events.
For continuous outcome data, such as time of hospitalisation before diagnosis and total length of in-patient stay, results for each study were expressed as differences in means with 95% CIs and were combined for meta-analysis, when appropriate, using the mean difference (MD). If the standard deviation was not available, this was imputed using the technique described in the Cochrane Handbook for Systematic Reviews of Interventions (Section 184.108.40.206).
Primary analysis used the fixed-effect model, and sensitivity analysis (if required) used the random-effects model. We used the random-effects model if the I2 statistic was greater than 50% (Higgins 2011).
Cost-effectiveness and quality of life analyses were summarised in narrative form.
If no data were available for some outcomes, these were described within the review. This potentially indicated the need for further clinical trials in this area.
Unit of analysis issues
When different scales were used to report the same outcomes, and we were not able to convert them, we planned to use standardised mean difference (SMD). This was not necessary, as no outcome data were extracted that required this.
Dealing with missing data
The review authors contacted the lead authors of the trials for which data clarification was required. This contact was made by email.
Assessment of heterogeneity
Statistical analysis was performed in accordance with the guidelines developed by The Cochrane Collaboration (Higgins 2011). Assessment of heterogeneity was possible when two or more primary studies were identified for inclusion in a meta-analysis. Heterogeneity (variation) between results of different studies was evaluated by:
performing empirical evaluation through visual inspection of the overlap of CIs on the forest plot; poor overlap indicates heterogeneity;
using the Chi2 statistical test for heterogeneity (Higgins 2011);
using an I2 statistic, which evaluates variation between studies (Higgins 2011); if a value greater than 50% was found, substantial heterogeneity was assumed; or
using a random-effects model or a fixed-effect model.
Assessment of reporting biases
Publication bias was to be assessed using a funnel plot if 10 or more studies were identified for either of the two comparisons. A gap on either side of the graph would have indicated that some trials had not been found, often as the result of difficulties in locating unpublished trials. No within-study reporting bias was assessed.
The presence or absence of heterogeneity was considered before data from trials were pooled. When it was not appropriate to combine the data, primary studies were summarised in narrative form. Women with suspected appendicitis and women with NSAP were analysed separately.
Subgroup analysis and investigation of heterogeneity
Subgroup analysis was planned, if appropriate and possible, with consideration of the following.
If substantive heterogeneity was seen, the review authors first confirmed the data, then considered:
The prespecified potential sources of heterogeneity were considered to explore possible explanations for variations in effects between trials and to guide interpretation of study findings, that is, location, method used to validate time to definitive diagnosis outcomes, location of pain and time of observation before intervention. We were aware of the limited value that interpretation of the causes of heterogeneity has after heterogeneity has been identified.
Protocol change: One of the planned subgroup analyses was changed to become one of the two comparisons.
A sensitivity analysis was planned to explore whether the results of any meta-analysis were sensitive to the inclusion or exclusion of RCTs with the following characteristics.
Unpublished studies, as these studies may not have been peer-reviewed and thus could be of lower quality (all studies were published manuscripts, and therefore this was not undertaken).
Studies with high risk of bias versus studies with low risk of bias.
Studies with no allocation concealment versus those with allocation concealment (this was not done, as most studies did not report on allocation concealment).
Studies in which men were included or women of postmenopausal age were included.
Studies that included participants who did not strictly meet the inclusion criteria (e.g. right-sided pain) (added as a protocol change).