Intraoperative differences between near-infrared fluorescence cholangiography with indocyanine green and conventional white light laparoscopic cholecystectomy: an integrative review of evidence base
摘要
The objective of this study was to provide a thorough synthesis of existing evidence of laparoscopic cholecystectomy assisted by Near-Infrared Fluorescence cholangiography with Indocyanine green vs conventional White Light Imaging-assisted laparoscopic cholecystectomy in patients with gallstone disease and cholecystitis.
MethodsWe used a comprehensive approach in the frame of integrative review, analyzed publications of evidence-based studies and synthesized additional data with diverse methodologies.
Eligibility criteria. We included only randomized controlled trials (RCTs) and systematic review with meta-analyses in this review.
Information sources. We searched the Cochrane Central Register of Controlled Trials, PubMed, Embase, ClinicalTrials.gov, CINAHL and PROSPERO for relevant publications in English before April 1, 2025. The search in Russian was conducted in e-Library before April 1, 2025.
Risk of bias. We assessed the risk of bias in the included randomized trials using the Cochrane RoB 2 tool.
Synthesis of results. Statistical analysis with meta-analysis data visualization was performed in RevMan 5.4. The risks of random errors in the meta-analysis were assessed using a diversity-adjusted required information size (DARIS) estimates and Trial Sequential Analysis (TSA) conducted in Copenhagen Trial Unit software.
ResultsIncluded studies. The results of 6 RCTs and 7 meta-analyses were included in qualitative and quantitative syntheses.
Synthesis of results.
Main findings improved identification rates: Near-Infrared Fluorescence cholangiography with Indocyanine green (NIF ICG) significantly increased the identification rates of the common bile duct (CBD) and common hepatic duct (CHD) compared to White Light Imaging (WLI). The common bile duct identification rate was 78.6% (298/379) for NIF ICG vs 49.7% (188/378) for WLI (RR = 1.68, 95% CI [1.31, 2.15], I2 = 66%). The common hepatic duct identification rate was 59.1% (224/379) for NIF ICG vs 32.8% (124/378) for WLI (RR = 1.81, 95% CI [1.53, 2.13], I2 = 40%).
Safety outcomes: The analysis demonstrated comparable safety profiles between NIF ICG and WLI laparoscopic cholecystectomies in terms of bile duct injury conversion to open surgery and cystic duct identification.
These findings were confirmed in sensitivity analyses using various meta-analytic models, including a Bayesian meta-analysis to handle zero-event trials.
There was not enough data on acute cholecystitis for selective synthesis.
DiscussionLimitations of evidence. The small sample size of the primary trials and moderate certainty of evidence did not allow to confirm the advantage of NIF ICG. A statistically significant reduction in bile duct injury (BDI) was not demonstrated, which is likely attributable to the very low incidence of this complication and insufficient statistical power, despite the improved visualization achieved with fluorescence cholangiography.
Interpretation. Further trials of laparoscopic cholecystectomy with NIF ICG should focus on bile duct identification in different populations of patients. Conducting a multi-centered trial of NIF ICG versus WLI or another clearly defined comparator may be a means of gathering homogeneous data in order to develop definitive conclusions.
ConclusionIn comparison with WLI, the use of NIF ICG enhances visualization of biliary structures during cholecystectomy, without yet proving a reduction in rare major complications. As the evidence is limited by small sample size and moderate heterogeneity of the primary studies, current results suggest a possible benefit but are not conclusive. The observed differences in identification rates for the common bile and hepatic ducts may represent overestimation due to the limited number of studies.
Trial registrationRegistration of the review’s meta-analysis: PROSPERO 2025. Clinical Trial Number-CRD420251086887. Available from https://www.crd.york.ac.uk/PROSPERO/view/CRD420251086887.