Evaluating effectiveness and safety of combined percutaneous transhepatic gallbladder drainage and laparoscopic cholecystectomy in acute cholecystitis patients: Meta-analysis

Abstract

BACKGROUND

Acute cholecystitis (AC) is a common disease in general surgery. Laparoscopic cholecystectomy (LC) is widely recognized as the "gold standard" surgical procedure for treating AC. For low-risk patients without complications, LC is the recommended treatment plan, but there is still controversy regarding the treatment strategy for moderate AC patients, which relies more on the surgeon's experience and the medical platform of the visiting unit. Percutaneous transhepatic gallbladder puncture drainage (PTGBD) can effectively alleviate gallbladder inflammation, reduce gallbladder wall edema and adhesion around the gallbladder, and create a "time window" for elective surgery.

AIM

To compare the clinical efficacy and safety of LC or PTGBD combined with LC for treating AC patients, providing a theoretical basis for choosing reasonable surgical methods for AC patients.

METHODS

In this study, we conducted a clinical investigation regarding the combined use of PTGBD tubes for the treatment of gastric cancer patients with AC. We performed searches in the following databases: PubMed, Web of Science, EMBASE, Cochrane Library, China National Knowledge Infrastructure, and Wanfang Database. The search encompassed literature published from the inception of these databases to the present. Subsequently, relevant data were extracted, and a meta-analysis was conducted using RevMan 5.3 software.

RESULTS

A comprehensive analysis was conducted, encompassing 24 studies involving a total of 2564 patients. These patients were categorized into two groups: 1371 in the LC group and 1193 in the PTGBD + LC group. The outcomes of the meta-analysis revealed noteworthy disparities between the PTGBD + LC group and the LC group in multiple dimensions: (1) Operative time: Mean difference (MD) = 17.51, 95%CI: 9.53-25.49, P < 0.01; (2) Conversion to open surgery rate: Odds ratio (OR) = 2.95, 95%CI: 1.90-4.58, P < 0.01; (3) Intraoperative bleeding loss: MD = 32.27, 95%CI: 23.03-41.50, P < 0.01; (4) Postoperative hospital stay: MD = 1.44, 95%CI: 0.14-2.73, P = 0.03; (5) Overall postoperative complication rate: OR = 1.88, 95%CI: 1.45-2.43, P < 0.01; (6) Bile duct injury: OR = 2.17, 95%CI: 1.30-3.64, P = 0.003; (7) Intra-abdominal hemorrhage: OR = 2.45, 95%CI: 1.06-5.64, P = 0.004; and (8) Wound infection: OR = 0. These findings consistently favored the PTGBD + LC group over the LC group. There were no significant differences in the total duration of hospitalization [MD = -1.85, 95%CI: -4.86-1.16, P = 0.23] or bile leakage [OR = 1.33, 95%CI: 0.81-2.18, P = 0.26] between the two groups.

CONCLUSION

The combination of PTGBD tubes with LC for AC treatment demonstrated superior clinical efficacy and enhanced safety, suggesting its broader application value in clinical practice.

Key Words: Acute cholecystitis, Laparoscopic cholecystectomy, Percutaneous transhepatic gallbladder drainage, Meta-analysis, Efficacy

Core Tip: Laparoscopic cholecystectomy (LC) is the standard surgical procedure for treating acute cholecystitis (AC), but postoperative complications and patient mortality are relatively high. Percutaneous transhepatic gallbladder drainage (PTGBD) can quickly drain infected bile, reduce gallbladder tension, and is often used in combination with delayed LC in clinical practice, but PTGBD is associated with more adverse long-term outcomes. The meta-analysis results of this study showed that the combination of PTGBD and LC for the treatment of AC has short surgical time, low conversion rate to open surgery, less intraoperative bleeding, and low overall incidence of complications, which is worthy of promotion.

INTRODUCTION

Acute cholecystitis (AC) is a prevalent acute abdominal pathology in general surgery and is predominantly attributed to etiologies such as gallstones, bile stasis, and bacterial infections. With the rapid advancements in laparoscopic technologies, laparoscopic cholecystectomy (LC) has emerged as the standard surgical approach, superseding open cholecystectomy for gallbladder pathologies. Despite this progress, mortality rates in high-risk cohorts remain substantial, ranging between 3.7% and 41.0%[1]. Percutaneous transhepatic gallbladder drainage (PTGBD) is an expedited method for draining infected bile, thereby alleviating gallbladder tension and reducing the risk of rupture due to bile-induced chemical irritation. This approach effectively diminishes toxin absorption, mitigates gallbladder obstruction, and stabilizes the disease trajectory, with some patients attaining complete resolution[2]. LC is the preferred modality for uncomplicated, low-risk patients[3]. However, in moderate AC scenarios, characterized by potential organ failure and significant local inflammatory adhesions, the combination of PTGBD with delayed LC appears more judicious[4]. Despite its utility, PTGBD is not an unequivocal therapeutic modality and is linked to less favorable long-term results, including prolonged hospitalization periods and heightened rates of readmission[5,6]. International guidelines exhibit variability in recommending LC over PTGBD for AC management[3,7,8]. Additionally, most of the current domestic and international research on the combination of PTGBD and LC for treating AC has involved single-center studies with small sample sizes. In this study, we sought to perform a comparative evaluation of the clinical effectiveness and safety of LC and the combination of PTGBD and LC in the management of AC by employing an evidence-based medicine framework. This comparison was intended to provide a theoretical basis for appropriately selecting surgical methods for patients with AC.

MATERIALS AND METHODS

Inclusion and exclusion criteria

The inclusion criteria were as follows: (1) Study type: Clinical randomized controlled trials (RCTs); (2) Subjects: Patients diagnosed with AC; (3) Intervention: Control group receiving LC; Experimental group receiving PTGBD combined with LC; and (4) Outcome measures: Clinical efficacy indicators, including operative time, intraoperative blood loss, conversion to open surgery rate, postoperative and total hospital stay; safety indicators, including overall postoperative complication rate, and incidence of bile leakage, bile duct injury, intra-abdominal hemorrhage, and wound infection. The exclusion criteria were as follows: (1) Basic research, abstracts, conference papers, reviews, meta-analyses, systematic reviews, or case reports; (2) Literature not in Chinese or English; (3) Duplicate publications, including studies that were republished or similar in research direction by the same author; (4) Studies with a sample size of fewer than 20 subjects; and (5) Literature where the full text was unavailable, inappropriate statistical methods were used, data could not be extracted, or data were incomplete.

Literature search strategy

Through computerized searches, we conducted an extensive exploration across a multitude of databases, including PubMed, Web of Science, EMBASE, Cochrane Library, China National Knowledge Infrastructure, and Wanfang, with the aim of identifying RCTs pertaining to the combined management of AC using PTGBD and LC. Our search strategy included specific English terms, including "acute cholecystitis" or "AC", "laparoscopic cholecystectomy" or "LC" and "percutaneous transhepatic gallbladder" or "PTGBD". The corresponding Chinese search terms included: Acute cholecystitis, laparoscopic cholecystectomy, percutaneous transhepatic gallbladder puncture and drainage, alongside others. Our search efforts spanned from the inception of these databases to the present day. This manuscript was meticulously crafted in accordance with the directives set forth in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement. Furthermore, our systematic review has been officially registered with PROSPERO under Registration No. 2023110106.

Data extraction

Two researchers carried out a thorough and independent literature screening process following preestablished inclusion and exclusion criteria. Initially, they meticulously scrutinized the titles and abstracts to exclude studies that were evidently unrelated to the research focus. Following this initial screening, the full texts of potentially relevant studies underwent a comprehensive review to validate their eligibility. Any discrepancies that emerged during this process were resolved through consultation with a third researcher. The data extraction procedure encompassed critical details, including the titles of the included literature, authors, publication journals, publication year, study sample size, inclusion periods for samples, treatment modalities, outcome measures, and other pertinent information.

Quality assessment of the included studies

We utilized the Newcastle-Ottawa scale (NOS) to evaluate the methodological quality of the included studies. The NOS assessment criteria encompass the following dimensions: The selection of study groups (comprising 4 items with a maximum score of 4 points), comparability of the groups (1 item with a potential score of up to 2 points), and assessment of outcomes (comprising 3 items with a potential score of up to 3 points). The highest achievable score was 9 points, with a score of ≥ 6 points indicating high-quality literature.

Statistical analysis

The data were analyzed utilizing RevMan 5.3 software for meta-analysis. For dichotomous variables, odds ratios (ORs) in combination with 95%CI were utilized as the primary outcome measures. For continuous variables, the mean difference (MD) and 95%CI were used. The assessment of study heterogeneity was performed using the I² statistic. A fixed-effects model was adopted when P was ≥ 0.1 and I² was < 50%, while a random-effects model was utilized when P was < 0.1 and I² was ≥ 50%. The sources of heterogeneity were meticulously investigated, and publication bias was evaluated through funnel plot analysis. A P value < 0.05 was considered to indicate statistical significance.

RESULTS

Literature search results

According to the literature search strategy, a comprehensive search across various databases was conducted. Initially, 672 articles were identified. After removal of duplicates, 235 articles were excluded. After scrutinizing the titles and abstracts, 369 articles were subsequently excluded. After thorough examination of the full texts and additional assessment, 43 articles were eliminated due to inappropriate intervention measures or data that could not be extracted. Overall, a total of 24 articles were included in this study. The details of the literature selection process and its outcomes are depicted in Figure 1.

Figure 1 Flow diagram depicting the screening and inclusion process of the studies. CNKI: China National Knowledge Infrastructure.

Basic characteristics of the included studies

In the course of this investigation, of the 24 articles that were included, 7 were published in Chinese, while the remaining 17 were published in English. These articles collectively encompassed a cohort of 2564 patients, with 1371 allocated to the control group and 1193 to the experimental group. Table 1 provides a comprehensive overview of the fundamental characteristics of the studies included in this analysis.

Table 1 Overview of the included study characteristics.

Ref.	Country	LC	Timing of LC after admission	PTGBD + LC	Time of LC after PTGBD	NOS
Duan and Li[9], 2021	China	44	Immediately	48	Within 1-3 months	8
Hao and Fan[10], 2022	China	82	Within 24 h	118	Within 1-3 months	8
Liu[11], 2018	China	40	Not specified	40	Simultaneous	8
Ma et al[12], 2020	China	57	Not specified	57	After 1 month	7
Wu and Kuang[13], 2017	China	20	Within 3 months	14	Within 3 months	8
Zhan et al[14], 2023	China	38	Not specified	38	4-6 wk	8
Zhou et al[15], 2019	China	47	Within 3 h	47	3-9 wk	8
El-Gendi et al[16], 2017	Egypt	75	After 72 h	75	6 wk	8
Chikamori et al[17], 2002	Japan	9	Not specified	31	Not specified	7
Choi et al[18], 2012	Korea	63	Within 72 h	40	Average 7.9 d	8
Hu et al[19], 2015	China	35	Not specified	35	6-10 wk	8
Jung and Park[20], 2017	Korea	166	Not specified	128	2-23	8
Karakayali et al[21], 2014	Turkey	48	Within 72 h	43	4-8 wk	8
Ke and Wu et al[22], 2018	China	47	After 72 h	49	1-1096 d	8
Kim et al[23], 2009	Korea	60	Not specified	35	Within 7 d	8
Kim et al[24], 2011	Korea	147	Mean time interval 42.2 h	97	Mean time interval 188.4 h	8
Lee et al[25], 2017	Korea	41	36 had LC within 24 h, 3 had LC 1-3 d, 2 had LC > 7 d	44	Mean 30 d	8
Liu et al[26], 2020	China	45	57.6 ± 12.2 h	58	62.4 ± 11.5 h	8
Na et al[27], 2015	Korea	77	Not specified	39	Not specified	7
Ni et al[28], 2015	China	33	Not specified	26	Within 1 year	8
Pan et al[29], 2023	Taiwan	67	Immediately	23	Not specified	8
Tsumura et al[30], 2004	Japan	73	Within 24 h	60	Not specified	8
Yamazaki et al[31], 2023	Japan	22	Immediately	13	After 2 months	8
Yang and Tian[32], 2022	China	35	After 7 d	35	After 40 d	8

PTGBD: Percutaneous transhepatic gallbladder puncture drainage; LC: Laparoscopic cholecystectomy; NOS: Newcastle-Ottawa scale.

Clinical efficacy meta-analysis results

Operative time: In this comprehensive analysis, all 24 studies[9-32] included a comparative assessment of operative times across the two groups. The heterogeneity test indicated a highly significant variance (P < 0.00001, I² = 94%), thus indicating the application of a random-effects model for the analysis. The meta-analysis demonstrated a significantly shorter operative duration in the PTGBD + LC group than in the LC group, with an average decrease of 17.51 min (95%CI: 9.53-25.49, P < 0.0001). For a visual depiction of these results, please refer to Figure 2A.

Figure 2 Forest plot depicting comparison between the percutaneous transhepatic gallbladder puncture drainage + laparoscopic cholecystectomy group and the laparoscopic cholecystectomy group. A: Forest plot depicting the operation time comparison between the percutaneous transhepatic gallbladder puncture drainage (PTGBD) + laparoscopic cholecystectomy (LC) group and the LC group; B: Forest plot illustrating the comparison of the conversion rate to open surgery between the PTGBD + LC group and the LC group; C: Forest plot illustrating the comparison of intraoperative bleeding between the PTGBD + LC group and the LC group; D: Forest plot depicting the comparison of overall hospital stay between the PTGBD + LC group and the LC group; E: Forest plot depicting the comparison of postoperative hospital stay between the PTGBD + LC group and the LC group; F: Forest plot illustrating the comparison of postoperative complications between the PTGBD + LC group and the LC group; G: Forest plot illustrating the comparison of bile leakage rates between the PTGBD + LC group and the LC group; H: Forest plot depicting the comparison of wound infection rates between the PTGBD + LC group and the LC group; I: Forest plot depicting intraperitoneal hemorrhage in the PTGBD + LC group compared to the LC group; J: Forest plot comparing bile duct injury incidence between the PTGBD + LC group and the LC group. PTGBD: Percutaneous transhepatic gallbladder puncture drainage; LC: Laparoscopic cholecystectomy.

Conversion to open surgery rate: Twenty-two studies[9,10,12-28,30-32] compared the rate of conversion to open surgery between the two groups. The heterogeneity test results showed P = 0.005 and I² = 50%, indicating significant heterogeneity. Consequently, a random effects model was employed for the analysis. The outcomes of the meta-analysis revealed a statistically notable difference: In comparison to the LC group, the PTGBD + LC group displayed a decreased conversion rate to open surgery, with an average reduction of 2.95%. This reduction was statistically significant, with a 95%CI of 1.90 to 4.58, and the P value was less than 0.00001 (Figure 2B).

Intraoperative blood loss: Seventeen studies[9-16,18,19,22,26-28,30-32] compared intraoperative blood loss between the two groups. The heterogeneity test results showed P < 0.00001 and I² = 97%, necessitating the use of a random-effects model for the analysis. The meta-analysis revealed that the PTGBD + LC group exhibited a significant reduction in intraoperative blood loss compared to the LC group, with an average decrease of 32.27 mL, which was statistically significant (95%CI: 23.03-41.50, P < 0.00001; Figure 2C).

Total hospital stay: Fifteen studies[10,11,15,17-20,23-28,31,32] conducted a comparative assessment of the total hospital stay between the two groups. Heterogeneity testing produced significant results (P < 0.00001, I² = 98%), which necessitated the adoption of a random-effects model for the analysis. The meta-analysis revealed that the PTGBD + LC group had a marginally longer total hospital stay than did the LC group, with an average increase of 1.85 d; however, it is important to note that this difference did not achieve statistical significance (95%CI: -4.86-1.16, P = 0.23; Figure 2D).

Postoperative hospital stay: Thirteen studies[9,12-14,17,20-22,24,25,27,30,31] conducted a comparative analysis of postoperative hospital stays between the two groups. The heterogeneity test results showed P < 0.00001 and I² = 94%, necessitating the application of a random-effects model for the analysis. The meta-analysis revealed a statistically significant difference: The PTGBD + LC group demonstrated a noteworthy reduction in postoperative hospital stay compared to the LC group, with an average decrease of 1.44 d (95%CI: 0.14-2.73, P = 0.03; Figure 2E).

Safety meta-analysis results

Postoperative complication rate: Twenty-two studies[9-24,26-29,31,32] compared the overall postoperative complication rates between the groups. The heterogeneity test showed P = 0.0001 and I² = 61%, which necessitated the adoption of a random-effects model for the analysis. A significant reduction in the overall incidence of complications was observed in the PTGBD + LC group compared to the LC group. This reduction amounted to an average of 1.88 patients (95%CI: 1.45-2.43, P < 0.00001; Figure 2F).

Bile leakage: Seventeen studies[9,13-16,18-23,26,27,29-32] compared the incidence of postoperative bile leakage between the two groups. The heterogeneity test showed P = 0.40 and I² = 5%, confirming the use of a fixed-effects model for the analysis. After conducting the meta-analysis, it was evident that the PTGBD + LC group exhibited a lower incidence of postoperative bile leakage than did the LC group, with an average reduction of 1.33 patients. Nevertheless, it is noteworthy that this disparity did not reach statistical significance, as indicated by the 95%CI: 0.81-2.18, P = 0.26 (Figure 2G).

Wound infection: Sixteen studies[9,10,12,14-16,19-21,23,25,26,28-31] compared the incidence of postoperative wound infection between the two groups. The results of the heterogeneity test demonstrated P = 0.67 and I² = 0%, indicating minimal heterogeneity. Consequently, a fixed-effects model was employed for the analysis. The meta-analysis findings demonstrated a statistically significant reduction in postoperative wound infections in the PTGBD + LC group compared with the LC group, with an average decrease of 2.17 patients (95%CI: 1.30-3.64, P = 0.003; Figure 2H).

Intra-abdominal hemorrhage: Ten studies[9,10,13,15-17,20,26,27,32] compared the incidence of postoperative intra-abdominal hemorrhage between the groups. The heterogeneity test results, with P = 0.67 and I² = 0%, supported the application of a fixed-effects model for the analysis. The meta-analysis revealed a notable difference: The PTGBD + LC group exhibited a significantly lower incidence of postoperative intra-abdominal hemorrhage than did the LC group, with an average reduction of 2.45 patients (95%CI: 1.06-5.64, P = 0.004; Figure 2I).

Bile duct injury: Six studies[12,14,16,17,29,30] compared the incidence of postoperative bile duct injury between the two groups. The heterogeneity test results showed P = 0.67 and I² = 0%, thus confirming the use of a fixed-effects model for the analysis. The meta-analysis demonstrated that the PTGBD + LC group exhibited a significantly lower incidence of postoperative bile duct injuries than did the LC group, with an average reduction of 4.46 patients (95%CI: 1.42-14.02, P = 0.01; Figure 2J).

Publication bias: Publication bias analysis was conducted using funnel plots for various outcomes, which included operative time, conversion to open surgery rate, intraoperative blood loss, total hospital stay, postoperative hospital stay, overall postoperative complication rate, postoperative bile leakage, bile duct injury, intra-abdominal hemorrhage, and incidence of wound infection. The results indicated good symmetry in the funnel plots, suggesting that the study results are minimally influenced by publication bias (Figure 3).

Figure 3 Funnel plots of each outcome. A: Operative time; B: Conversion rate to open surgery; C: Intraoperative bleeding; D: Overall hospital stay; E: Postoperative hospital stay; F: Postoperative complications; G: Bile leakage; H: Wound infection; I: Intraperitoneal hemorrhage; J: Bile duct injury. MD: Mean difference; OR: Odds ratio.

DISCUSSION

A meta-analysis refers to the use of statistical methods to analyze and summarize multiple collected research studies, providing a quantified average effect to answer research questions. Its advantage lies in increasing the credibility of the conclusions by enlarging the sample size. This study included 24 articles and included a meta-analysis aimed at comparing the clinical effectiveness and safety of LC combined with PTGBD and LC for the treatment of AC, providing valuable clinical insights. The meta-analysis revealed that compared with the LC group, the PTGBD + LC group had a shorter surgery time (MD = 17.51, 95%CI: 9.53-25.49, P < 0.01), a lower rate of conversion to open surgery (OR = 2.95, 95%CI: 1.90-4.58, P < 0.01), less intraoperative bleeding (MD = 32.27, 95%CI: 23.03-41.50, P < 0.01), a shorter postoperative hospital stay (MD = 1.44, 95%CI: 0.14-2.73, P = 0.03), and a lower overall postoperative complication rate (OR = 1.88, 95%CI: 1.45-2.43, P < 0.01). There were also lower incidences of postoperative bile duct injury (OR = 14.46, 95%CI: 1.42-14.02, P = 0.01), intra-abdominal bleeding (OR = 2.45, 95%CI: 1.06-5.64, P = 0.004), and wound infection (OR = 2.17, 95%CI: 1.30-3.64, P = 0.003), indicating that PTGBD combined with LC is more effective and safer for treating AC than LC alone.

The short surgical time in the PTGBD + LC group may be related to the relief of local inflammation after PTGBD. As a minimally invasive surgery, PTGBD reduces gallbladder swelling, gallbladder wall edema, and inflammation around the gallbladder[33]. PTGBD can immediately alleviate the clinical symptoms of AC in patients in good preoperative condition[34]. In addition, PTGBD can be used for cholangiography to display the anatomical structure of the biliary tract and provide clear information on the surgical site[35].

Previous studies have concluded that the fundamental reason for transitioning to open surgery is recurrent and progressive inflammation accompanied by gallbladder wall swelling and edema[36,37]. The low rate of conversion to open surgery in the PTGBD + LC group may be attributed to PTGBD, as PTGBD has the ability to alleviate inflammatory gallbladder adhesion. After PTGBD, patients have a reduced gallbladder wall thickness, clearer anatomical structure of the gallbladder triangle, reduced intraoperative bleeding, and a reduced risk of biliary tract injury.

Our analysis of intraoperative bleeding revealed less bleeding in the PTGBD + LC group than in the LC group. PTGBD can immediately alleviate the decompression of swollen gallbladders and inflammation around the gallbladder, preventing the development of fibrosis in the Calot triangle[34]. With the help of PTGBD, the surgical field of view of the Calot triangle is clearer, which facilitates LC and reduces intraoperative bleeding.

Our analysis of hospitalization time revealed that the postoperative hospitalization time of the PTGBD + LC group was significantly shorter than that of the LC group. It is possible that the patient's gallbladder wall congestion, edema, and inflammation caused by gallbladder inflammation gradually subside in the short term after PTGBD, making it easier to distinguish anatomy and thereby reducing surgical difficulty and shortening surgical time[25,33]. Our analysis revealed that the total incidence of postoperative complications in the PTGBD + LC group was low, possibly because PTGBDD was associated with minimal trauma. At the same time, supportive treatment, such as systemic anti-infection therapy, can quickly alleviate clinical symptoms, prevent further deterioration of the condition, and provide sufficient time for the treatment of complications[38].

The results of the meta-analysis indicated that the combined hospital stay in the PTGBD + LC group was slightly longer than that in the PTGBD + LC group, but the difference did not reach statistical significance (MD = -1.85, 95%CI: -4.86-1.16, P = 0.23). The extended total hospital stay in the PTGBD + LC group can be attributed to the requirement for a longer duration or even two hospital admissions to complete the treatment, which may be considered the sole drawback of PTGBD.

The presence of publication bias is indicated by the observed asymmetry in the funnel plot. Significant heterogeneity was also observed in some studies, possibly related to variations in surgical experience, surgical instruments, severity grading of AC, and discharge criteria. The interval between PTGBD tube placement and LC was another important factor contributing to heterogeneity. The optimal timing for delayed LC after PTGBD remains controversial, with different experiences and policies at various centers leading to varied optimal timing. Jia et al[39] noted that patients who underwent LC within 5 d after PTGBD had significantly less surgical time, bleeding, postoperative peritoneal drainage duration, postoperative oral intake time, and postoperative complications than those who exceeded 5 d. Finally, most of the included studies were performed in Asia, and the data regarding LC and PTGBD in Europe and America were unclear. RCTs and multicenter studies with large sample sizes are needed to verify the outcomes of this meta-analysis.

Meta-analysis is an observational study type and inevitably possesses biases in its design. The limitations of this study include the following: (1) The inclusion of a restricted number of research articles with small sample sizes has implications for the generalizability of conclusions and increases the vulnerability to publication bias; (2) Variability in the interval from PTGBD to LC in the included studies; and (3) The focus on only English and Chinese literature, neglecting gray literature and other languages, which might introduce certain biases in the results.

CONCLUSION

In summary, this meta-analysis revealed that the amalgamation of PTGBD combined with LC conspicuously enhances clinical efficacy while concurrently mitigating the frequency of postoperative complications in individuals afflicted with AC.