If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Cairns Clinical School, College of Medicine and Dentistry, James Cook University, QLD, 4870, AustraliaResearch and Development Unit, Primary Health Care and Dental Care, Narhalsan, Southern Älvsborg County, Region Västra Götaland, SwedenDepartment of Public Health and Community Medicine, Institute of Medicine, The Sahlgrenska Academy, University of Gothenburg, Sweden
Methodological quality of the studies was mostly heterogeneous.
•
Several studies developed prediction models for risk of conversion.
•
Independent risk factors appeared to have additive effects.
•
Only one study was regarded as high quality.
•
High-quality studies should be conducted to externally validate prediction models.
Abstract
Background
The study aims to evaluate the methodological quality of publications relating to predicting the need of conversion from laparoscopic to open cholecystectomy and to describe identified prognostic factors.
Method
Only English full-text articles with their own unique observations from more than 300 patients were included. Only data using multivariate analysis of risk factors were selected. Quality assessment criteria stratifying the risk of bias were constructed and applied.
Results
The methodological quality of the studies were mostly heterogeneous. Most studies performed well in half of the quality criteria and considered similar risk factors, such as male gender and old age, as significant. Several studies developed prediction models for risk of conversion. Independent risk factors appeared to have additive effects.
Conclusion
A detailed critical review of studies of prediction models and risk stratification for conversion from laparoscopic to open cholecystectomy is presented. One study is identified of high quality with a potential to be used in clinical practice, and external validation of this model is recommended.
and that conversion is known to increase perioperative time, complication rates, perioperative costs, the length of hospital stay, and hospital charges.
It is, therefore, essential to identify risk factors for conversion to allow for safer procedures and better surgical planning. A systematic assessment of these factors pre-operatively allows determination of whether OC surgery should be performed initially, avoiding the potential complications brought through an intraoperative conversion from LC to OC. Further, effective conversion prediction models allow patients the right to be better informed of such risks before they give consent.
Several factors encourage a new review of the literature on LC to OC conversion. Two previous reviews agreed on only two important risk factors when considering conversion, namely, male gender and old age
Preoperative risk factors for conversion of laparoscopic cholecystectomy to open surgery - a systematic review and meta-analysis of observational studies.
Preoperative risk factors for conversion of laparoscopic cholecystectomy to open surgery - a systematic review and meta-analysis of observational studies.
recent study focused on only 10 articles for inclusion in their meta-analysis. Further, literature searches in previous reviews were conducted in 2005 and in 2014 respectively. The gap between those studies and current date, and the presence of new critical studies published in recent years, warrant a new review.
study, but not evaluated in Rothman et al.’s review, hence, other significant factors may have been overlooked in Rothman et al.’s meta-analysis. Finally, a comprehensive quality assessment of methodological rigour was not conducted in either review by Tang and Cuschieri
This systematic literature review aims to evaluate risk factors associated with LC–OC conversion as well as to consider the methodological quality of the included studies.
2. Methods
2.1 Protocol and registration
This systematic review was conducted according to guidelines set by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA).
The review was registered in the PROSPERO database, with the reference number CRD42016039195.
2.2 Eligibility criteria
Only English language full-text articles with quantitative studies were included. Inclusion criteria for the systematic review were studies reporting risk factors or prediction models regarding LC–OC conversion. Retrospective and prospective studies including observations from more than 300 patients which presented their own unique observations, and with multivariate statistical techniques adjusting for covariates, were included. No restrictions were made regarding the time of publication.
2.3 Search strategy
Two investigators (AH and RM) performed the systematic search through PubMed (Table 1) and Scopus (Table 2) on June 8, 2016. The two investigators screened publication titles and abstracts independently. Discrepancies were adjudicated by a third review author (AdeC). Full-text articles were then reviewed. An additional snowball search found the two review articles as discussed earlier.
Table 1Search strategy for PubMed.
((((“epidemiology”[MeSH Terms] OR “models, statistical”[MeSH Terms] OR “nomogram”[MeSH Terms] OR “risk factor”[MeSH Terms] OR risk* OR “risk assessment”[MeSH Terms]))) AND ((((“Cholecystectomy”[MeSH Terms]) OR “Biliary Tract Surgical Procedures”[MeSH Terms]) OR “Cholecystectomy, Laparoscopic”[MeSH Terms])) AND (conversion to open surgery[MeSH Terms] OR “conversion to open surgery” OR open surgery[MeSH Terms] OR conver*))
(INDEXTERMS(“epidemiology”) OR INDEXTERMS(“models, statistical”) OR INDEXTERMS(“nomogram”) OR INDEXTERMS(“risk factor”) OR risk* OR INDEXTERMS(“risk assessment”)) AND (INDEXTERMS(“Cholecystectomy”) OR INDEXTERMS(“Biliary Tract Surgical Procedures”) OR INDEXTERMS(“Cholecystectomy, Laparoscopic”)) AND (INDEXTERMS(“conversion to open surgery”) OR “conversion to open surgery” OR INDEXTERMS(“open surgery”) OR conver*)
Eight measures were developed for this review to assess methodological quality (Table 3).
Table 3Criteria for estimating risk for bias.
Low risk
Intermediate risk
High risk
All analysed independent variables are defined
All independent variables analysed are described/defined
At least 70% of all independent variables are described/defined AND the number of non-described variables are stated (Hence you know the total number of independent variables analysed)
The number of independent variables remains unclear OR less than 70% of all independent variables are clearly described/defined.
Sample size calculation
Sample size calculation done AND it is described how it was done AND The estimated sample size (or more) was recruited.
Sample size calculation done AND it is described how it was done AND not able to recruit estimated sample size
No sample size calculation OR no description of how it was done
Data extraction procedure described
Medical chart or reliable database; manually read charts or that they had some mechanism to ensure the quality of their database
Data extracted from a database and no mentioning of a mechanism to ensure quality in that database
No mentioning of how data is extracted
Statistical analysis described
Clearly described what is being used to analysed data
Analysis described but not in detail
Analysis not discussed
Multivariate analysis
(Multivariate stepwise regression + entry & removal) OR (Multivariate non-stepwise stating which variables were entered or if all variables entered)
Stating multivariate regression and stating if it was logistic or Cox but no more details
Multivariate analysis not mentioned or not using multivariate regression
Missing data presented
Give exact numbers of missing data + explain why there are more missing data for some variables
Give exact numbers of missing data but no explanation for why some variables have more missing data
Number of missing data for each variable not provided
Missing data discussed
Missing data discussed on how they should be managed in statistical analysis and final interpretation.
Mentions missing data in discussion but unclear how they managed this
No discussion about how they managed missing data
Validation of model (internal or external) presented
External validation of model presented either as sensitivity and specificity OR Area under curve (AUC).
Internal validation of model presented either as (Cox&Snell R Square OR Naegelkirke R-square) OR Area under curve (AUC).
Outcome of internal or external validation of their final model is not described
attempted to perform an “external validation”, however, it is worth pointing out the methodology utilised was poor. Gholipur et al. used nine cases to calculate sensitivity and specificity, with a point estimate of 67% and a wide confidence interval of 30%–93%, hence making the validation insignificant. Consequently, this study failed to meet a “low risk” for external validation of model.
3.2 Risk factors and their association with highest possible quality of the studies
Several risk factors for LC–OC conversion were identified from the extracted studies (Table 4). Variables considered significant in the high quality study
This variable was reported to increase the of conversion between 1 and 6 times in the 15 of the 20 studies which determined this risk factor as significant.
31053-4&id=gr1.jpg){kind=link}
31053-4&id=gr2.jpg){kind=link}