Zhao-Qing Lu, Han-Yu Zhang, Chen-Fen Su, Yue-Yan Xing,Guo-Xing Wang, Chun-Sheng Li

Abstract

Key Words: Acute cholangitis; Endoscopic retrograde cholangiopancreatography; Mortality; Biliary drainage; Organ dysfunction

lNTRODUCTlON

Acute cholangitis (AC) is a life-threatening condition that occurs in the presence of biliary obstruction.Although growing evidence confirms that biliary decompression can dramatically decrease the mortality rates associated with AC, the optimal time for biliary drainage remains controversial[1-3].Most experts agree that the timeframe for biliary decompression is distinct for different severity grades based on the Tokyo guidelines 2018/2013 (TG18/13), which recommend early biliary drainage for Grade II AC and urgent biliary drainage for Grade III AC; however, the exact timeframe is unclear[4].Moreover, the 2019 European Society of Gastrointestinal Endoscopy guidelines recommend biliary drainage as soon as possible in patients with Grade III AC and within 12 h for patients with septic shock, while within 48-72 h for patients with Grade II AC[2]. Meanwhile, the 2021 American Society of Gastrointestinal Endoscopy (ASGE) guidelines suggest endoscopic retrograde cholangiopancreatography (ERCP) within 48 h in AC patients, regardless of severity[1]. Some studies have demonstrated a“the earlier, the better” approach for the management of AC cases[5-8]. They deemed urgent biliary decompression necessary in severe cases. However, other studies have shown that ERCP can be safely delayed to allow for sufficient resuscitation[9-11]. In our clinical experience, the severity of Grade III AC varies dramatically, and we have observed that the optimal timing of biliary decompression may vary for different organ injuries.

So far, no relevant study has determined whether the clinical outcomes of Grade III AC patients vary with different organ injuries and whether AC patients with different organ dysfunction should undergo biliary drainage at distinct times. Consequently, this single-center retrospective study aimed to investigate the relationship between drainage timing and in-hospital mortality (IHM) in patients with AC of different severities, and attempted to stratify patients with AC to identify the best decompression time.

MATERlALS AND METHODS

Patients

Ethical compliance: This retrospective cohort study was conducted according to the tenets of the Declaration of Helsinki, and approved by the Bioethics Committee of Beijing Friendship Hospital,Capital Medical University (Certification No. 2020-P2-224-01). After the review conducted by the Ethics Committee, the need for informed consent was waived.

Sample size: Based on the pre-experimental results, which revealed a 0.87% and 3% mortality rates for patients who did and did not undergo biliary drainage, respectively, we calculated the sample size using PASS 15.0 with Power = 0.8, Alpha = 0.05, and N1 = N2. The resultant sample size was 1306. On this basis, we expanded the sample size by 15% to exclude the reduction caused by data deficiency or diagnostic errors.

Enrollment: We searched the electronic medical record database of the Beijing Friendship Hospital for all adult (＞ 18 years) non-pregnant patients visiting the Emergency Department (ED) between July 2016 and December 2020 who were discharged with a diagnosis of AC. A total of 1498 patients were included in the study. All the medical records were reviewed and a diagnosis of AC was confirmed according to the TG18 diagnostic criteria[12]. Besides pregnant women and patients aged ＜ 18 years, we also excluded patients transferred to other hospitals or outpatient departments for treatment and patients whose medical data were incomplete or missing. Finally, 1305 patients were enrolled (Figure 1).

Data collection

Demographic data, including age and sex, were obtained for each patient. Clinical data, laboratory characteristics, and imaging findings were obtained from the electronic medical records. Clinical data included vital signs, fluid output, Glasgow coma score, the vasoactive drug dose, history of biliary disease, Charlton comorbidity index (CCI), etiology of AC, registration time at the ED, time and type of drainage, discharge time, complications of drainage, hospitalization cost, and IHM. Laboratory findings included white blood cell (WBC) count, platelet count, total bilirubin, albumin, aspartate transaminase,alanine transaminase, creatinine, amylase, prothrombin time-international normalized ratio, arterial blood oxygen pressure, and lactate. Computed tomography, magnetic resonance cholangiopancreatography, or ERCP findings were reviewed to determine the etiology of cholangitis. The time interval for biliary drainage was calculated as the difference between ED registration time and the time of ERCP or percutaneous transhepatic biliary drainage (PTBD). We used the TG18 severity criteria for grading the severity[12].

Definition and indications

Benign stenosis was defined as stenosis in the biliary ducts without radiological findings of metastasis or malignant cells on histological examination. The indications of ERCP were AC combined with persistent extrahepatic biliary obstruction, and cases of anatomical derangement, failed ERCP,intrahepatic biliary obstruction underwent PTBD. The contraindications of ERCP included acute coronary syndrome, acute heart failure (New York Heart Association class III-IV), acute stroke occurring within 2 wk prior to enrollment, and acute pulmonary embolism complicated with the arterial PaO2＜60 mmHg, while the contraindications of PTBD were platelet count ＜ 50000/mm3or prothrombin activity ＜60%. The adverse effects after drainage were based on the lexicon guidelines of the ASGE[13]. The second treatment included a second ERCP and a second PTBD for stone removal or stent placement.The need for a second therapeutic ERCP was assessed per the discretion of the experienced operators who performed the intervention and based on 2018 ERCP Guidelines for China[14]. All of the endoscopists had independently completed 300 ERCP interventions per year for 3 years.

Study outcomes

Our primary outcome was IHM, and the secondary outcomes were hospital length of stay (LOS) and hospitalization costs. When analyzing the survivors’ LOS and entailed cost, we excluded patients who died or were transferred to other hospitals. We investigated the primary and secondary outcomes relevant to the timing of biliary drainage according to severity grading and some predictors based on TG18 and other previous studies.

Statistical analysis

Figure 1 A flowchart of the study population.

Continuous variables were reported as medians with interquartile ranges (IQR) for not normally distributed; categorical variables were expressed as numbers and percentages. Comparisons between groups for continuous variables were evaluated using the Kruskal-Wallis test and Mann-WhitneyUtest, and for categorical variables, Fisher’s exact test was used. Univariate logistic regression analysis and multivariate logistic regression with stepwise variable selection were used to analyze the relationship between organ dysfunction or predictors and IHM. Odds ratios (ORs) and 95% confidence intervals were calculated. Statistical significance was set atP＜ 0.05. Statistical analyses were conducted using SPSS v22.

RESULTS

The characteristics of the study population are shown in Tables 1 and 2. This cohort included 1305 patients, the majority of whom were men (60%). The median age of the group was 74 years (IQR, 63-82 years), and patients with Grade I AC were significantly younger (P＜ 0.001). The CCI of individuals with Grade III cholangitis was significantly higher than that of patients with Grade I or II cholangitis (P＜0.001). Choledocholithiasis (69.4%) was the most common cause of cholangitis, followed by malignant obstruction (18.5%). Among patients with cholangitis caused by malignancy, the number of patients with Grade II cholangitis was higher than that of patients with Grade I or III cholangitis. More patients with Grade II AC underwent biliary decompression (P= 0.006), while the IHM and hospitalization costs were significantly higher in the Grade III AC group than in the Grade I and II AC groups (IHM:P＜0.001, hospitalization costs:P＜ 0.001). Deaths tended to occur earlier in the disease course in patients with Grade III AC (median time 6 d).

A total of 915 patients in this cohort (70.1%) underwent biliary decompression, most of whom underwent endoscopic interventions (90.7%), while more patients underwent PTBD in the severe AC groups (Grade II: 37 cases, Grade III: 35 casesvsGrade I: 13 cases,P= 0.001). One hundred and sixty patients with Grade III were treated with antibiotics and fluid infusion only (disagreement about procedures = 35 cases, including two of mortality; with contraindications = 24 cases, including five of death; not tolerable conditions = 14 cases, including eight of death; obstruction spontaneous relief = 67 cases, including seven of mortality; not persistent biliary obstruction with high risk to stone removal by ERCP = 15 cases without death; not persistent biliary obstruction with selective cholecystectomy = five cases without death). The median drainage time was 24 h. Nearly half of the patients underwent biliary drainage within 24 h (445/915, 48%), while 89 patients with Grade III underwent drainage within 12 h(29%), which was significantly higher than that in Grades I and II (P= 0.04). Among the patients who underwent biliary drainage, more patients in Grade III underwent biliary drainage outside regular working hours (31.9%) and 478 patients (52.2%) required a second intervention for stone removal or stent placement, which exhibited no relationship with severity (P= 0.688). Post-ERCP pancreatitis was the major adverse event (8.2%), and the incidence of adverse events was not significantly different among the three severity grades.

Table 1 Characteristics data for all patients

The association of the IHM, LOS, and hospitalization costs with the timing of biliary drainage in each severity grade has been presented in Table 3. Biliary drainage significantly decreased the IHM in all AC patients (13/915vs28/390,P＜ 0.001). Among individuals with Grade III cholangitis, the IHM rate was significantly decreased in patients who had biliary drainage within 24 h or 48 h (P= 0.041,P= 0.018),while the LOS was significantly increased among those patients who underwent drainage within 12 h,24 h, or 48 h (P＜ 0.001,P＜ 0.001, andP＜ 0.001, respectively). Hospitalization costs significantly increased for all patients with biliary drainage, regardless of severity criteria. For patients with Grade I AC, IHM was not affected by the timing of drainage and there was no relationship between LOS and the timing of biliary drainage for the survivors, but LOS was significantly extended in all the patients who underwent biliary drainage within 24 h or 48 h (P= 0.004,P= 0.002).

Univariate and multivariate analyses of predictors, including TG18 severity grading factors, lactate,CCI, and etiology of malignant obstruction, are shown in Table 4. In the univariate analysis, organ dysfunction, abnormal WBC count, hypoalbuminemia, CCI ＞ 3, and etiology of malignancy were significantly associated with IHM. In a multivariate analysis of organ dysfunction, only neurological,respiratory, renal, and cardiovascular dysfunction, hypoalbuminemia, and malignant obstruction were significantly positively associated with IHM (OR = 5.32, 2.541, 6.356, 4.021, 5.655, and 7.522;P＜ 0.001,P= 0.016,P＜ 0.001,P= 0.012,P＜ 0.001, andP＜ 0.001, respectively).

Table 2 Characteristics data for patients with drainage

From the data presented in Table 4, the association between IHM and the timing of biliary drainage for each clinical predictor is reflected. Among patients with AC complicated with neurological dysfunction or hyperlactatemia, the IHM rate was significantly decreased in individuals who underwent biliary decompression within 12 h (P= 0.041,P= 0.016), while those patients complicated with neurological dysfunction who had drainage within 24 h or 48 h could not improve prognosis compared with those who underwent drainage after 24 or 48 h, as well as those who did not undergo biliary drainage after admission (P= 0.175,P= 0.304). In patients with renal dysfunction, abnormal WBC count, hyperbilirubinemia, hypoalbuminemia, or hyperlactatemia, who had biliary decompression within 24 h or 48 h, there were significantly lower IHM rates than those of the remaining patients in each subgroup. Among patients with AC complicated with hepatic dysfunction, higher CCI (＞ 3), or malignancy, a lower IHM rate was significantly associated with early drainage (＜ 48 h) (P= 0.016,P=0.006,P= 0.021).

DlSCUSSlON

In this retrospective monocenter cohort analysis including 1305 AC patients stratified according to TG18 severity criteria, early drainage (＜ 24 h) in Grade III AC patients had the greatest benefit, which could significantly reduce the all-cause IHM. It is worth noting that, in this subgroup, patients complicated with neurological dysfunction should be drained as early as possible (＜ 12 h). Moreover, early drainage was also required for AC patients with lactate more than 2 mmol/L, abnormal WBC count, hyperbilirubinemia, or hypoalbuminemia. Although these patients may have no severe cholangitis, early drainage could significantly reduce IHM and benefit the most. However, it should be noted that our results did not prove that drainage reduced hospitalization costs compared to conservative treatment. Furthermore,early drainage increased hospitalization costs and LOS.

AC is a clinically heterogeneous disease with recent mortality rates ranging from 2.7% to 10%, which is considered as an improvement since these rates were as high as 50%-60% in severe cases[15,16].Although it is well-accepted that biliary decompression can significantly decrease the mortality of AC, it is highly debatable as to what the optimal strategy is for performing biliary drainage procedures.Should drainage occur as early as possible or should the procedure be delayed until stabilization after adequate resuscitation, especially for Grade III AC patients? Our study showed that biliary drainage could significantly decrease the IHM of AC patients (1.4%vs7.2%,P＜ 0.001), and AC patients of GradeIII underwent decompression within 24 h or 48 h had lower IHM (3.9%vs9.0%,P= 0.041; 4%vs9.9%,P= 0.018). These findings are in line with the results of a large sample size retrospective study, which showed that drainage within 48 h was associated with improved IHM (3.4%vs10.2%,P= 0.019)[17],although it did not discuss drainage within 24 h. Another retrospective study of 6063 patients with AC conducted by Kiriyamaet al[3] reported a distinct result: decompression within 24 h or 48 h improved the 30-d mortality rate only among Grade II patients (drainage within 24 hvs. after 24 h or no drainage:1.7%vs3.4%,P＜ 0.05; drainage within 48 hvsafter 48 h or no drainage: 2.0%vs3.7%,P＜ 0.05), and the drainage timing did not influence 30-d mortality in patients with Grade I and Grade III. This dissimilarity may be because we reviewed the all-cause IHM, while Kiriyamaet al's study observed 30-d cholangitis-caused mortality, and the mortality rate of Grade III AC patients was lower than that in our study (5.1%vs9.4%), especially for subgroups of delayed decompression in which more patients died from malignancy[3]. In addition, as the number and grade of patients without drainage were not shown in the results of Kiriyamaet al[3], the proportion of patients without drainage in our study may be different from the group studied by Kiriyamaet al[3]. Since then, few researchers have tried to examine the possibility of improving the prognosis of Grade III AC patients by undergoing biliary drainage more immediately[6,8]. Unfortunately, consistent with our results, there was no evidence that earlier decompression (within 12 h) benefited these severe patients.

Table 3 Association of the clinical outcomes in relation to the timing of biliary drainage and severity grading by Tokyo guidelines 2018

Table 4 Univariate and multivariate analysis of predictors of mortality, and relationship of mortality with the timing of drainage for each predictor

Another multicenter retrospective net cohort study reported a novel finding in which delayed biliary decompression (＞ 12 h) was significantly associated with increased mortality in AC patients combined with septic shock [OR 3.40 (1.12-10.31)][7]. These results suggest that it may be necessary to approach each organ’s failure separately and that different types of organ dysfunction should be considered when determining the optimal drainage timing. This was verified by our multivariable analysis which showed that neurological, respiratory, renal, and cardiovascular dysfunction were significantly associated with IHM (OR = 5.32, 2.541, 6.356, 4.021, respectively), whereas hepatic and hematological dysfunction did not influence IHM. In addition to indicators affecting mortality recommended in TG18, we also analyzed CCI, malignancy, and lactate levels, which were previously reported to have an impact on the death rate[5,7,18]. Multivariate regression analysis showed that, in addition to organ dysfunction,hypoproteinemia and malignant obstruction significantly increased IHM. These outcomes were similar to those reported in a previous review[3].

In this study, we first attempted to stratify Grade III AC patients according to organ dysfunction to identify patients whose prognosis could most likely be improved by earlier decompression. According to our clinical experience, neurological dysfunction was the most severe complication and might affect drainage operation; therefore, AC patients with altered mental status require immediate biliary decompression. This was verified by our data that only undergoing drainage within 12 h could greatly improve IHM in this subgroup of individuals (0%vs17.3%,P= 0.041). For AC patients with renal dysfunction, performing drainage within 24 h could decrease the IHM. As a result, the renal injury was not only caused by infection but also induced by hypovolemia, which was provoked by fever, decreased appetite, early initiation of antibiotics, and aggressive fluid resuscitation to improve renal flow were more valuable for these patients. Unlike previous studies[7], we did not observe an association between drainage and IHM in AC patients with cardiovascular dysfunction. This may be because only 39 patients were included in this subgroup. Drainage within 12 h significantly reduced mortality in AC patients with increased lactate levels (≥ 2 mmol/L), suggesting that decompression within 12 h might be beneficial for patients with cardiovascular dysfunction. Performing drainage within 48 h greatly reduced mortality in AC patients with hepatic dysfunction, because the amelioration of biliary stasis after drainage could directly rescue hepatic function and diminish inflammatory reaction, even if no significant association between hepatic injury and IHM was observed.

The influence of abnormal WBC count and hyperbilirubinemia on outcomes in AC has been previously described[9]. The underlying pathophysiologic mechanisms are known to result from sepsis;therefore, abnormal WBC count and hyperbilirubinemia are linked with an increased likelihood of poor prognosis. Hypoalbuminemia refers to malnutrition, which is linked to life-threatening infections and poor outcomes. For individuals in these subgroups, undergoing drainage within 24 h indicated a lower IHM.

Decompression within 12, 24, or 48 h in Grade III AC patients would extend the LOS, which was contrary to the findings of Aboelsoudet al[19] who demonstrated that drainage within 24 h or 48 h could decrease the LOS, compared with drainage after 24 h or 48 h (drainage within 24 hvsafter 24 h:mean 7.71vs13.57 d,P= 0.001; drainage within 48 hvsafter 48 h: mean 8.61vs14.24 d,P= 0.002). This was because their study only reviewed patients who underwent drainage, whereas our study included both patients who underwent drainage and those who did not. In addition, most of the Grade II and Grade III patients in this study received only endoscopic nasobliary drainage or endoscopic retrograde biliary drainage treatment at the first ERCP intervention because of their serious condition, and 52.2% of them received a second endoscopic intervention for biliary stone removal or stent placement when their condition stabilized, thus increasing the LOS. Drainage within 12, 24, or 48 h was associated with increased hospital costs despite the severity of AC since endoscopic operation itself was expensive, and the delayed-drainage group contained non-drainage ones who spent little.

Limits of study

Our analysis had certain limitations. First, it was retrospective in design, and as such, was susceptible to record bias and incomplete data. Second, because it is a single-center study, the study results only reflect the clinical status of our center and cannot represent the situation of all hospitals. Third, due to the limited sample size, especially the limited number of patients with cardiovascular dysfunction who underwent drainage within 48 h, some results were not statistically significant. In addition, the time from onset to initial treatment and the antibiotics and fluid resuscitation before admission to our hospital were not evaluated, which may have an impact on the outcome. A prospective multicenter study will be conducted to further explore a precise stratified model to guide drainage timing rather than TG18.

CONCLUSlON

To our knowledge, this is the first study to stratify patients with Grade III AC according to different organ dysfunctions and to recommend the optimal timing of biliary drainage accordingly. Biliary decompression significantly decreases the all-cause IHM rates in AC, and early biliary drainage (＜ 24 h)was shown to be beneficial for AC patients with Grade III severity. Severe cases with different organ dysfunctions have distinct prognoses, and cases complicated with either neurological or cardiovascular dysfunction should undergo drainage within 12 h of admission. For patients with mild or moderate cholangitis, early drainage increases the probability of a second treatment and increases the patients'injury, hospitalization cost, and LOS. A further multicenter prospective cohort study will be conducted to verify the result and investigate whether the optimal timing of drainage based on different organ dysfunctions can increase the 30-d mortality rates and decrease the readmission rate.

ARTlCLE HlGHLlGHTS

FOOTNOTES

Author contributions:Li CS designed the research study and made critical revisions to the article; Lu ZQ, Zhang HY,Xing YY, Su CF, and Wang GX performed the research and analyzed the data; Lu ZQ drafted the article; all authors have read and approved the final manuscript.

lnstitutional review board statement:This retrospective cohort study was conducted according to the tenets of the Declaration of Helsinki, and approved by the Bioethics Committee of Beijing Friendship Hospital, Capital Medical University (Certification No. 2020-P2-224-01).

lnformed consent statement:Since we conducted a retrospective review of patient data obtained from the electronic medical records of our hospital, and as stated by the ethical standards of China, the Ethics Committee exempted the need for obtaining informed consent.

Conflict-of-interest statement:All the Authors have no conflict of interest related to the manuscript.

Data sharing statement:The datasets used during the current study are available from the corresponding author on reasonable request at lcscyyy@163.com.

STROBE statement:The authors have read the STROBE Statement—checklist of items, and the manuscript was prepared and revised according to the STROBE Statement—checklist of items.

Open-Access:This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BYNC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is noncommercial. See: https://creativecommons.org/Licenses/by-nc/4.0/

Country/Territory of origin:China

ORClD number:Zhao-Qing Lu 0000-0001-8895-1596; Han-Yu Zhang 0000-0001-5181-4619; Chen-Fen Su 0000-0002-7882-6975; Yue-Yan Xing 0000-0002-6162-3098; Guo-Xing Wang 0000-0003-1379-5749; Chun-Sheng Li 0000-0002-3161-4457.

S-Editor:Yan JP

L-Editor:A

P-Editor:Yan JP