Effect of transient obstructive cholestasis on liver histology: a cross-sectional study

ABSTRACT BACKGROUND: The role of transient obstructive cholestasis on liver histology remains undetermined. OBJECTIVE: To investigate whether transient cholestasis impairs liver histology. DESIGN AND SETTING: Cross-sectional study at a public university hospital (UNICAMP), Brazil. METHODS: 169 individuals undergoing cholecystectomy, with or without cholestasis. were enrolled. Histopathological findings were correlated with clinical and biochemical characteristics. RESULTS: Biliary hepatopathy was more frequent in individuals with resolved cholestasis than in those with active obstruction or no jaundice (P < 0.01), as also were fibrosis and ductular proliferation (P = 0.02). Cholestasis was commoner in individuals with resolved obstruction than in those with no history (P < 0.01) or active cholestasis (P < 0.05). Biliary hepatopathy was associated with longer duration of cholestasis (P < 0.001) and higher bilirubin levels (P = 0.02) in individuals with active obstruction; with lower body mass index (P = 0.02) and longer cholestasis (P < 0.001) in individuals with resolved obstruction; and with longer cholestasis (P < 0.001) and longer interval between endoscopic retrograde cholangiopancreatography and surgery (P = 0.03) overall. In individuals with active obstruction, duration of cholestasis (R = 0.7; P < 0.001) and bilirubin levels (R = 0.6; P = 0.004) were independently correlated with cholestasis severity. Duration of cholestasis (R = 0.7; P < 0.001) was independently correlated with ductular proliferation severity. CONCLUSIONS: Transient cholestasis was associated with significant histopathological changes, even after its resolution. Longer duration of obstruction correlated with greater severity of histopathological changes, especially cholestasis and ductular proliferation. This emphasizes the need for early treatment of obstructive cholestasis.


INTRODUCTION
Cholestasis is an impairment of bile formation and/or bile flow that may present with fatigue, pruritus and, in its most suggestive form, jaundice. It can be classified as intrahepatic or extrahepatic. [1][2][3][4] It may assume various histological patterns, which present different clinical and diagnostic connotations. The two main patterns, which are canalicular cholestasis and chronic cholestasis, constitute general categories, more suggestive of the progress and degree of cholestasis than of any exact cause. The other two patterns, ductular cholestasis and ductal cholestasis, usually develop within a context of canalicular cholestasis, but are differentiated because of their narrower clinical contexts. Basically, the main histological features observed in cholestasis are occurrences of ductular proliferation and bilirubinostasis. 5 Biliary obstruction is caused by mechanical impairment of bile flow through large ducts, mainly extrahepatic bile ducts. Its structural correspondent is the parenchymal cholestasis with biliary stasis located in zone 3 (perivenular). In cases of incomplete obstruction, a ductular reaction may occur without clear evidence of cholestasis. 6 Several studies have shown severe consequences of biliary obstruction on liver histology, with changes such as microscopic cholestasis and cholangitis, liver fibrosis and inflammatory changes. Prolonged maintenance may even lead to biliary cirrhosis. [5][6][7] The vast majority of cases of biliary obstruction are transient, since several treatments can usually be employed to correct the obstructive factor. Treatment may, in most cases, be either surgical or endoscopic. There are cases in which the obstruction resolves spontaneously, such as when small gallstones migrate through the duodenal papilla after a period of obstruction. Commonly, the signs and symptoms of cholestasis cease gradually after the obstructive factor resolves. However, there is no conclusive

OBJECTIVE
The aims of this study were to investigate whether occurrence of transient cholestasis might lead to significant changes in liver histology and to analyze the influence of the duration of cholestasis on liver histology. then accordingly divided between these groups, as follows: 1 (n = 115); 2 (n = 54); 2A (n = 25); and 2B (n = 29). Figure 1 shows the flowchart of the study population and the subdivision according to groups. Additionally, individuals were divided into groups according to their bilirubin levels (< 3 mg/dl, 3-7 mg/dl or ≥ 7 mg/dl).

Study design
Those who presented cholestasis were also divided into two groups according to the duration of obstruction (< 15 days or ≥ 15 days).

Variables
The following clinical and demographic variables were analyzed: age (in years); gender (male or female); cholestasis status (active, resolved, overall or no history of cholestasis); endoscopic retrograde cholangiopancreatography (performed or not performed); duration of cholestasis (estimated in days from clinical history information in the medical records); interval between ERCP and surgery (in days); and body mass index (BMI; expressed in kg/m 2 ).
The histopathological variables considered were the present or absence of biliary pattern liver disease and the following specific histological characteristics: fibrosis, ductular proliferation, cholestasis, portal inflammation, steatosis and cholangitis. These characteristics were classified dichotomously as absent or present and were also stratified ordinally according to their degree of severity, as absent (0), mild (1), moderate (2) or severe (3). 5 The histopathological analyses were all performed by the same pathology team and followed the same parameters as defined by Crawford et al. 8 All specimens were fixed in formalin, embedded in paraffin and sectioned using a microtome at a thickness of 5 µm. Routine specimen processing involved staining the slides with hematoxylin and eosin (15 levels), Masson trichrome (10 levels) and reticulin (5 levels), for a total of 30 levels per specimen. All levels were screened to ensure absence of histological abnormalities. 8

Statistical analysis
The descriptive analysis consisted of presenting frequency tables

Demographic characteristics and nutritional status
There was a predominance of females in the study population (65.7%). There were no differences in gender distribution between the groups (P = 0.9). The average age was 47.2 ± 17.3 years; there were no significant differences in age between the groups studied (P = 0.6). The mean BMI was 27.4 ± 5.2 kg/m 2 ; there were no significant differences between the groups (P = 0.08). The complete comparison among the groups is presented in Table 1.

Biochemical variables
Regarding biochemical variables, group 2A showed significantly higher levels of bilirubin (P < 0.001), AST (P < 0.001), ALT (P < 0.001), ALP (P < 0.001) and GGT (P < 0.001), compared with groups 1 and 2B. Mean INR was lower in group 1 than in groups 2A and 2B (P < 0.001) and ALB was lower in group 2A than in groups 1 and 2B (P = 0.01). The complete comparison among biochemical variables is presented in Table 1.

Distribution of histopathological variables
It was observed that the presence of biliary hepatopathy was more frequent in group 2B than in groups 1 and 2A (P < 0.01) and also in group 2A compared with group 2 (P < 0.05). Fibrosis was more common in group 2B than in groups 1 and 2A (P = 0.02), as also was ductular proliferation (P = 0.02). Cholestasis was more common in group 2B than in groups 2A (P < 0.01) and 1 (P < 0.05) and was also more frequent in group 2A than in group 1 (P < 0.05). There were no significant differences in the distribution of portal inflammation (P = 1.0), cholangitis (P = 0.6) or steatosis (P = 0.3). The complete distribution of histopathological variables is presented in Table 2.

Biliary hepatopathy
In the analysis on the presence of biliary hepatopathy, it was observed that, in the overall population, its presence was associated with lower BMI (P = 0.01) and higher bilirubin levels (P = 0.01) and INR (P = 0.04). In a subgroup analysis, biliary hepatopathy was associated with longer duration of cholestasis (P < 0.001), higher bilirubin levels (P = 0.02) and higher INR (P = 0.02) in group 2A; with lower BMI (P = 0.02) and longer duration of cholestasis (P < 0.001) in group 2B; and with longer duration of cholestasis (P < 0.001) and longer interval between ERCP and surgery (P = 0.03), along with higher levels of ALP (P = 0.01) and GGT (P = 0.01) in group 2. In group 1, there was no variable that differed between the groups with or without biliary hepatopathy. The complete comparison between the variables analyzed according to the presence of biliary hepatopathy is presented in Table 3.       coefficients and multiple regression in the different subgroups and in the entire study population are presented in Table 4.

Liver histology, bilirubin levels and duration of cholestasis
Cholestasis was more frequent in individuals with bilirubin levels over 7 mg/dl (P = 0.04). Other histological features did not differ according to bilirubin levels ( Table 5).

DISCUSSION
The presence of biliary hepatopathy in a population without a clinical history of jaundice may indicate occurrence of subclinical and spontaneously resolved cholestatic events (gallstone migration, for example), but with maintenance of the histopathological alteration indefinitely. Rangaswamy et al. previously observed, among     In individuals with resolved cholestasis who underwent ERCP, a lengthier time elapsed between the procedures, and cholecystectomy was associated with the presence of biliary hepatopathy.
There was also a significant correlation between this interval and the severity of ductular proliferation.
Hence, it seems that performing surgery as early as possible may be beneficial. This may be due to the possibility of undetected  [13][14][15] Similarly, Moody et al., in a meta-analysis study, observed that early cholecystectomy reduced the frequency of readmissions due to biliary complications, among individuals with previous episodes of acute biliary pancreatitis. 16 The duration of obstruction was significantly associated with the severity of both cholestasis and ductular proliferation. This highlights the progressive aspect of these changes in relation to the duration of the causative condition and, of course, emphasizes the importance of early treatment in order to avoid the development of chronic disease. Previous studies have shown that the severity of functional and structural changes in the liver is time-dependent, but with varying individual susceptibility. Functional recovery after decompression is not immediate in animal models, with persistent short-term hepatocytic insufficiency. [17][18][19] Steatosis and portal inflammation were significantly associated with age in the general population and did not differ between groups. This suggested that non-alcoholic fatty liver disease (NAFLD) had a greater influence on these changes than did biliary conditions. Older age is a recognized risk factor for occurrence of NAFLD and also for progression to the most deleterious components of its histopathological spectrum, especially steatohepatitis and fibrosis/cirrhosis. 20,21 In the study population, treatment and resolution of the obstruction did not lead to full reversal of the histopathological changes associated with cholestasis, in a significant number of patients. There is evidence of mitigation of fibrosis and other liver histological abnormalities later on, after resolution of complete biliary obstructions, but complete resolution does not occur in all affected individuals. 22,23 Olguín et al. demonstrated in a murine model that the time elapsed between the obstruction and its correction is determinant for the possibility of histological reversal. 24 In addition, the correlation between the observed changes and laboratory tests was not appropriate, thus demonstrating the importance of histopathological examination in these cases.