Analysis of bailout procedure in laparoscopic cholecystectomy for acute cholecystitis

Background: The Tokyo Guidelines 2018 recommend a bailout procedure consisting of fundus-rst cholecystectomy, subtotal cholecystectomy, or open conversion to prevent serious complications in cases of dicult laparoscopic cholecystectomy (LC). Methods: The hospital records of patients with acute cholecystitis who underwent LC from October 2014 to April 2019 were retrospectively analyzed. The clinical data were compared between the standard and bailout groups. A subgroup analysis was performed to compare the fundus-rst and subtotal cholecystectomy techniques versus open conversion. Results: In total, 160 of 416 Japanese patients who underwent LC were diagnosed with acute cholecystitis. Standard LC was performed in 125 (78%) patients, and a bailout procedure was performed in 35 (22%). The duration from onset to surgery was signicantly longer (P = 0.04) and the C-reactive protein (CRP) concentration was signicantly higher (P = 0.001) in the bailout than standard group. The surgical outcomes were worse in the bailout group. In the multivariate analysis, a high CRP concentration at diagnosis was an independent predictor of bailout (P = 0.004). In the subgroup analysis, the open group had a signicantly longer duration from onset to surgery (P = 0.04) and a signicantly higher incidence of preoperative drainage (P = 0.002). With respect to surgical outcomes, the open group had signicantly greater blood loss (P = 0.02) and longer hospital stays (P = 0.002). Conclusion: A high CRP concentration is a risk factor for a bailout procedure. Early LC should be performed for patients with acute cholecystitis and a high CRP concentration.


Background
Improvements in surgical devices and techniques have led to an increase in the use of laparoscopic cholecystectomy (LC) to treat acute cholecystitis (AC). However, bile duct injury (BDI), a life-threatening complication, has also been increasingly reported 1,2,3 . Strasberg et al. 4 introduced the critical view of safety (CVS), which is formed by Calot's triangle and the cystic duct, to avoid BDI. The CVS is now applied worldwide 5,6 . However, it is di cult to establish the CVS in patients with severe in ammation or adhesion after AC. The Tokyo Guidelines 2018 (TG18) recommend a bailout procedure consisting of the fundus-rst approach, subtotal cholecystectomy, or open conversion to prevent BDI when the CVS cannot be secured 5 . However, the bailout procedure is a relatively new concept, and few real-world data have been accumulated. In this single-center retrospective study, we investigated the risk factors for a bailout procedure and performed a subgroup analysis of patients who underwent a bailout procedure.

Methods
The hospital records of patients who underwent LC from October 2014 to April 2019 were retrospectively investigated. In total, 160 of 416 patients who underwent LC met the diagnostic criteria for AC. These 160 patients were divided into the standard LC group (n = 125) and bailout group (n = 35) (Fig. 1). The patients' backgrounds and surgical outcomes such as the operative time, blood loss, postoperative hospital stay, and number of complications were compared between the two groups.
The study protocol is performed in accordance with the relevant guidelines.

Surgical technique
All procedures were performed by a surgeon who had performed more than 100 LCs.

Standard LC
With the patient under general anesthesia and in the supine position, the rst umbilical 12-mm camera port was inserted into the preperitoneal cavity, and insu ation pressure of 10 mmHg was used. Three 5mm ports were inserted from the right hypochondrium and ank. After exposing the CVS at the neck of the gallbladder (GB), the GB artery was cut using a single clip and the cystic duct was cut using a double clip. The whole GB was dissected from the liver bed and removed from the umbilical wound.

Fundus-rst cholecystectomy
Beginning at the fundus (i.e., in a retrograde manner), the GB was separated from the liver without initially visualizing the CVS.

Subtotal cholecystectomy
An incision was made in the GB, the contents were aspirated, as much of the GB wall as possible was removed, and the stump was sutured.

Open conversion
An incision was made to connect each 5-mm port hole, and cholecystectomy was performed. Statistical analysis All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modi ed version of R Commander designed to add statistical functions frequently used in biostatistics. The χ 2 test or Fisher's exact test was used to compare categorical variables, and the Mann-Whitney U test was used to evaluate continuous variables.
Univariate and multivariate logistic regression analyses were used to identify the risk factors for a bailout procedure. Receiver operating characteristic curves were used to determine the accuracy of the perioperative C-reactive protein (CRP) concentration and duration from onset to surgery, with values expressed as the area under the curve (AUC) with 95% con dence intervals. A P value of < 0.05 was considered statistically signi cant.

Page 4/12
The patients' characteristics and operative outcomes are shown in Table 1. In total, 160 patients with AC were treated by LC. Standard LC was performed in 125 (78%) patients, and a bailout procedure was performed in 35 (22%) patients. There were no signi cant differences in age, sex, body mass index (BMI), incidence of preoperative drainage, white blood cell (WBC) count, or Charlson comorbidity index between the two groups. The duration from onset to surgery was signi cantly longer in the bailout group than in the standard group (39.7 ± 40.5 vs. 28.2 ± 46.8 days, respectively; P = 0.04). The CRP concentration was signi cantly higher in the bailout group than in the standard group (10.47 ± 9.5 vs. 5.39 ± 7.5 mg/dL, respectively; P = 0.001). There was no signi cant difference in the incidence rate of complications, de ned as Clavien-Dindo grade ≥ II complications. In the bailout group, the operative time was signi cantly longer (206 ± 61.5 vs. 127 ± 38.7 min, P = 0.04), the blood loss volume was signi cantly higher (211 ± 226 vs. 37 ± 71.8 mL, P = 0.01), and the postoperative hospital stay was signi cantly longer (7.5 vs. 6.5 days, P = 0.04). Data are presented as mean ± standard deviation, n, or n (%).
WBC, white blood cells; CRP, C-reactive protein; CCI, Charlson comorbidity index; SSI, surgical site infection ※Clavien-Dindo grade ≥ II Next, we performed univariate and multivariate analyses of risk factors for a bailout procedure ( Table 2).
The cut-off value of age was the median of all patients, and the cut-off value of the BMI was 22 kg/m 2 (the standard value for adults). The cut-off value of the WBC count was 18,000 cells/µg, which is one of the criteria for moderate cholecystitis. A receiver operating characteristic curve analysis showed that the cut-off value of the preoperative duration (onset to surgery) was 35 days (AUC, 0.61; 45% sensitivity, 77% speci city) ( Fig. 2A) and that the cut-off of the preoperative CRP concentration was 9.4 mg/dL (AUC, 0.66; 51% sensitivity, 80% speci city) (Fig. 2B). The univariate analysis showed that male sex, preoperative drainage, duration from onset to surgery, WBC count, and CRP concentration were signi cant risk factors. The multivariate analysis showed that a high CRP concentration was an independent risk factor for bailout (odds ratio, 4.16; 95% con dence interval, 1.56-11.1; P = 0.004).

Discussion
The present study demonstrated that a high CRP concentration can predict a di cult LC procedure requiring bailout. Some previous studies have also revealed a relationship between a high CRP concentration and open conversion 7,8 . Measurement of the CRP concentration as a predictor is simple and easy to perform. However, CRP is not included in the severity classi cation of the TG18 9 ; instead, the WBC count is included 9 . This is probably because several studies have demonstrated that the WBC count can predict the life prognosis of patients with AC, and the severity classi cation attaches importance to the life prognosis. Therefore, we believe that predictors of the life prognosis are not the same as predictors of surgical di culty and that the CRP concentration should be noted as a predictor of di cult LC. In our comparison of surgical outcomes, the bailout group showed worse outcomes than the standard LC group, as expected. This may re ect the presence of in ammatory adhesions and postoperative pain. However, there were no signi cant differences in complications, especially BDI; therefore, we consider that the purpose of the bailout procedures was achieved.
In our subgroup analysis, the open group had poor surgical outcomes (Table 3) 20 . Therefor the many patients in Open group might have repeated recurrences during the duration unknowingly. Those subclinical recurrence may be an another strong risk factor of di cult LC. In this study, the recurrence was not assessed correctly and in any case, it will be better to perform early LC.
This study had two main limitations. First, this was a retrospective observational study with a small sample size. Second, a speci c protocol of selecting the bailout procedure was not established, and confounding was not removed in the subgroup analysis.

Conclusion
In conclusion, a high CRP concentration is a risk factor for a bailout procedure. Early LC should be performed for patients with AC and a high CRP concentration.

Declarations
Ethics approval and consent to participate This study was conducted in accordance with the Declaration of Helsink, and the protocol was approved by the Ethic Committee of the Nakatsu municipal hospital, Oita Japan. Figure 1 Flow chart of patient enrollment