Epidemiological profile, referral routes and diagnostic accuracy of cases of acute cholangitis among individuals with obstructive jaundice admitted to a tertiary-level university hospital: a cross-sectional study

ABSTRACT BACKGROUND: Obstructive jaundice may lead to ominous complications and requires complex diagnostic evaluations and therapies that are not widely available. OBJECTIVE: To analyze the epidemiological profile, referral routes and diagnostic accuracy at admittance of cases of acute cholangitis among patients with obstructive jaundice treated at a referral unit. DESIGN AND SETTING: Cross-sectional study at a tertiary-level university hospital. METHODS: Patients with obstructive jaundice who were treated by means of endoscopic retrograde cholangiopancreatography, resection and/or surgical biliary drainage were evaluated. The main variables analyzed were epidemiological data, referral route, bilirubin levels and time elapsed between symptom onset and admittance and diagnosing of acute cholangitis at the referral unit. The accuracy of the clinical diagnosis of acute cholangitis was compared with a retrospective analysis on the medical records in accordance with the Tokyo criteria. RESULTS: Female patients predominated (58%), with an average age of 56 years. Acute cholangitis was detected in 9.9% of the individuals; application of the Tokyo criteria showed that the real prevalence was approximately 43%. The main referral route was direct contact (31.8%) and emergency care (29.7%); routing via official referral through the public healthcare system accounted for 17.6%, and internal referral from other specialties, 20%. The direct route with unofficial referral was the most important route for cases of neoplastic etiology (P < 0.01) and was the fastest route (P < 0.01). CONCLUSIONS: There is a deficiency in the official referral routes for patients with obstructive jaundice. The accuracy of the clinical diagnosis of acute cholangitis was poor. Wider dissemination of the Tokyo criteria is essential.


INTRODUCTION
Jaundice is a clinical sign characterized by abnormal yellow coloring of the skin, mucous membranes and sclera. It is caused by increased bilirubin levels in the blood. 1 Most bilirubin is produced when hemoglobin is metabolized to indirect bilirubin, which then binds to albumin and is transported in the plasma to the liver, where it is conjugated with glucuronic acid to become water-soluble and is referred to as direct bilirubin. This is excreted in bile, in the duodenum. In the intestine, bacteria metabolize bilirubin to form urobilinogen.
Part of this urobilinogen is eliminated in feces and part is reabsorbed, reprocessed and excreted in bile (enterohepatic cycle). [2][3][4] Cholestasis is the condition in which the conjugated bile in the liver encounters an obstacle to its elimination in the duodenum. This may be due to disturbances of excretion such as hepatocellular injury (drug or viral hepatitis, pregnancy or sepsis) or abnormalities of the flow between the hepatocyte and the ampulla of Vater, such as gallstones of the main bile duct, periampullary neoplasm or pancreatitis. 4,7,8 The clinical sign of jaundice has a broad spectrum of etiologies and can range in severity from asymptomatic cases that do not require intervention, to others in which there may even be an imminent risk of death. 4  weight loss or itching. Jaundiced patients presenting acute diseases, often of infectious causes, may seek medical care for treatment of fever, chills, abdominal pain or flu-like symptoms. 4,9 Regarding imaging methods in the context of obstructive jaundice, ultrasound of the abdomen presents sensitivity of 46% and specificity of 96% for diagnosing dilatation of the common bile duct; and sensitivity of 38% and specificity of 100% for diagnosing gallstones. It has the advantages of being an inexpensive and accessible examination. However, it is operator-dependent and may be impaired through occurrences of distension of intestinal loops, agitation and obesity, which are common findings in these patients.
Magnetic resonance imaging is the test with the best accuracy for evaluation of the bile ducts, both for benign and malignant diseases, with sensitivity and specificity of up to 98%. [10][11] Contamination associated with infection of stagnant bile leads to inflammation of the biliary tract, and this condition character- Thus, an analysis on the referral routes of patients with obstructive jaundice who are admitted to referral services is essential. This makes it possible to define the main weaknesses observed in the initial management of these patients, ascertain the average time taken to provide referral care and define ways to optimize specialized care for these patients, based on the most prevalent etiologies and the clinical state that these patients present upon admission to the tertiary-level unit.

OBJECTIVE
The aim of this study was to critically analyze the referrals of patients with obstructive jaundice who were treated by means of ERCP or surgery at a tertiary-level hospital in Brazil. This analysis included identifying the main referral routes among patients with obstructive jaundice; identifying the main causes of referral and the time that elapsed between the initial care and admission to this hospital; and reviewing the diagnosis of cases with acute cholangitis at admission to the tertiary-level unit.

METHODS
A cross-sectional, retrospective and descriptive study was carried out to evaluate all consecutive adult patients with obstructive jaundice who were treated by means of ERCP and/or resection or biliary drainage surgery at Hospital de Clínicas, UNICAMP, Campinas, Brazil.
The study participants were identified and selected through the electronic scheduling systems of the outpatient unit for biliary tract surgery, digestive endoscopy unit and main surgical center.
These patients underwent ERCP or pancreatic or biliary resection surgery, and internal biliary drainage operations, between September 2017 and August 2018.
The inclusion criteria were that the patients needed to present: 1. bile duct obstruction with radiological evidence (computed tomography [CT], magnetic resonance imaging [MRI] or ERCP); or 2. jaundice defined by the presence of total bilirubin ≥ 2.5 mg/ dl. The exclusion criteria were: 1. absence of laboratory jaundice reported in this hospital service or recorded in medical records; or 2. indication of ERCP or biliary drainage for other nonobstructive causes (stent replacement, primary sclerosing cholangitis or liver transplantation). A flowchart of the study population is presented in Figure 1.
The variables analyzed were the following: 1. age (expressed in years); 2. sex (male or female); 3. etiology (neoplastic or non-neoplastic); 4. referral routes, which could be through SUS regulations

Statistical analysis
The chi-square test, or Fisher's exact test when necessary, was used to compare proportions. Diagnostic accuracy tests (sensitivity, specificity, positive predictive value, negative predictive value and overall accuracy) were used to evaluate the diagnostic methods. The diagnosis of cholangitis in accordance with the Tokyo protocol was considered to be the gold standard. 18 Normality was assessed using the Shapiro-Wilk test.
To compare continuous measurements, analysis of variance (ANOVA) was used for variables with normal distribution and the Mann-Whitney test was used for those with non-Gaussian distribution. To compare continuous measurements between three or  more groups, the Kruskal-Wallis test was used and, when a significant difference was observed, Tukey's post-test was used to determine the groups between which the difference was significant.
The significance level was taken to be 5% (P < 0.05). The Statistical Analysis System (SAS) for Windows, version 9.2 (SAS Institute Inc., 2002-2008, Cary, NC, USA) was used for the calculations of the statistical analysis.

RESULTS
The demographic and clinical characteristics of the study population are presented in Table 2. Female patients predominated in the sample of this study (58%). The patients were in their fifth decade of age onwards, with a mean age of 56 years. Benign etiologies were slightly more prevalent (52% of the cases).
Acute cholangitis was originally recorded in 9.9% of the cases of obstructive jaundice, but the review of the cases with application of the 2018 Tokyo criteria showed that the prevalence was approximately 43%. The overall accuracy of the clinical diagnosis of cholangitis at admission was estimated to be 67%, with sensitivity of 23% and specificity of 100% ( Table 3).
Regarding the referral routes among the patients with obstructive jaundice, the main routes were through direct contact (31.8%) and emergency care (29.7%). Routing via SUS regulations accounted for 17.6% of the cases and internal referral from other specialties, 20% ( Table 2).
The mean bilirubin levels of the patients with referrals was 15.8 mg/dl. The direct route through unofficial referral was the most important route for cases of neoplastic etiology (P < 0.01) and was the fastest route (P < 0.01). However, there was no statistical difference in bilirubin levels among the routes (Figures 2 to 4).

DISCUSSION
The present sample had a predominance of female patients (58%), with ages from the fifth decade onwards (mean of 56 years).
Benign etiologies were slightly more prevalent (52% of the cases). The great volume of cases and demand for care at our hospital service stems from the structuring of the healthcare service and the scarcity of specialized services providing high-complexity biliopancreatic surgery in our region. The regional health division of the Metropolitan Region of Campinas belongs to Regional Health Department (Departamento Regional de Saúde, DRS) VII, which has an estimated total population of three million inhabitants. 30 Regarding the diagnostic imaging methods available in this DRS, it has been estimated that there are 8.3 CT and 3.8 MRI machines in the public system per million inhabitants. This is slightly higher than the average for the state of São Paulo, which is 7.9 CT and 3.4 MRI machines per million inhabitants. 31 Within the worldwide context, these rates reach 107 CT and 56 MRI machines per million inhabitants in Japan. In emerging South American countries like Chile (14 CT and 9.43 MRI machines per million), these rates are also higher than those of our region. Regarding the main therapy, ERCP, it has been estimated that in Brazil there is an average of 2.98 procedures per 100,000 inhabitants, a rate that is far below the ideal. 32 In the United States, for example, this rate is 74 procedures per 100,000 inhabitants, and in China, an emerging country with a larger population, this rate is 14 per 100,000 inhabitants. 33,34 Regarding surgical procedures for resection of neoplasms, which are the main cause of extra-official referral routes, our hospital is the only public one in this region that has a sufficiently periodic and steady flow of patients for it to be possible to accomplish such procedures. Thus, the low availability of diagnostic examinations and therapeutic procedures (whether surgical or endoscopic) in other healthcare facilities ends up creating a flow of referrals of these patients.
Moreover, the clinically stigmatizing sign of jaundice, the rapid evolution of periampullary neoplasms and the already-discussed low availability of referral centers in this region that can resolve these cases, combined with the slowness and inefficiency of the official referral pathways, have led to potential alternative referral routes. Given the situation outlined above, these alternative referral routes are important for enabling rapid attendance of these patients. However, their existence attests to the organizational failure of the regional healthcare system in relation to allowing individuals access to the necessary resources in a timely manner through official channels.
This study presents some limitations that need to be taken into consideration. Its retrospective design can possibly be correlated with lower quality among the data collected, along with potential loss of participants due to incomplete medical reports. Because this was a single-center study, the findings are not immediately reproducible in other hospital services. In addition, the small sample precludes ultimate conclusions. A multicenter study involving more hospital services that receive referrals of individuals with obstructive jaundice would be more appropriate.

CONCLUSION
There is a deficiency in the official referral route for patients with obstructive jaundice, since less than 20% of the patients were found to have arrived through the regulatory system. However, this deficiency was not found to significantly impact the time taken to attend patients, or their bilirubin levels, because of the existence of alternative routes and emergency care. The frequencies of referrals for neoplastic and non-neoplastic causes were similar. The accuracy of clinical diagnoses of acute cholangitis was found to be poor and, therefore, greater dissemination of the criteria of the Tokyo protocol is essential.