Portal pressure and blood nitric oxide levels as predictors of outcome in biliary atresia.

AIM: To evaluate the incidence of portal hypertension (PHT) in biliary atresia (BA) patients and to monitor its progress after Kasai portoenterostomy (KP) by measuring nitric oxide (NO) levels in peripheral blood.
MATERIALS AND METHODS: A prospective cross-sectional study conducted over a period of 2 years. Intraoperative portal pressure (PP) and blood NO levels at presentation, 1-month, 3-month, and 6-month follow-up, were correlated with clinical and biochemical parameters in BA patients. The mean NO level in age-matched control group was 4.64 ± 2.32 μmol/L.
RESULTS: Thirty-four BA patients underwent KP over a period of 2 years. The mean age of presentation was 2.7 months (range 1-4 months). The mean intraoperative PP was 21.3 ± 5.4 mmHg. The mean PP in patients aged <60 days, 61-90 days, and >90 days was 18.53 ± 4.45 mmHg, 20.33 ± 3.07 mmHg, and 26.5 ± 5.01 mmHg, respectively. The mean PP in the patients who underwent successful KP was 16.75 ± 3.54 mmHg while for those who continued to have jaundice it was 23.94 ± 4.63 mmHg (P < 0.001). NO levels closely followed the PP as shown by the regression equation NO = 4.79 + 0.64 PP mmHg, R (2) = 0.69. The mean NO level at presentation was 18.48 ± 4.17 μmol/L and at 1-month, 3-month, and 6-month follow-up was 11.94 ± 5.62 μmol/L, 10.79 ± 6.02 μmol/L, and 9.93 ± 6.53 μmol/L, respectively (P < 0.001). The difference in NO levels was also statistically significant between the patients who cleared jaundice and those with persisting jaundice.
CONCLUSION: All BA patients had PHT at presentation. PHT worsens with age and has an adverse effect on outcome of KP. NO levels in blood closely follow PP and higher levels are associated with poor outcome.

INTRODUCTION

Extrahepatic biliary atresia (BA) is the most common cause of neonatal obstructive jaundice with an incidence of 1 in 10,000 live births.[12] Progressive damage of extrahepatic and intrahepatic bile ducts secondary to inflammation may occur, leading to fibrosis, biliary cirrhosis, and eventual liver failure. Kasai portoenterostomy (KP) is the standard operation for the treatment of BA.[3] A degree of hepatic fibrosis is invariably present at initial surgery and can be graded histologically. Intrahepatic disease may progress despite successful KP.[4] Portal hypertension (PHT) has been reported in 34-76% of BA patients, and it is the most serious late complication.[56] It manifests as ascites (63%) and esophageal varices (49%) with bleeding complications in 20-60%.[7] PHT is associated with a hyperdynamic state characterized by splanchnic vasodilation and increased blood volume as a result of the increase in local and systemic vasodilators. Nitric oxide (NO) has been recognized as the most important vasodilator molecule that mediates the excessive arterial vasodilation observed in PHT.[814] NO levels in peripheral blood have been utilized to monitor PHT.[9] This study aims to identify the incidence and severity of PHT in BA patients and to monitor its progress after KP using NO levels in peripheral blood.

MATERIALS AND METHODS

It is a prospective cross-sectional study conducted over a period of 2 years (June 2012 to June 2014). Ethical clearance for the study was obtained from the Institute Ethics Committee vide letter number IESC/T-212/01.06.2012. All children presenting with BA were enrolled in the study after taking informed consent from the parents/legal guardians. They were evaluated clinically, biochemically, and radiologically at presentation. Blood levels of NO and liver function tests were done at presentation and at 1-month, 3-month, and 6-month follow-up. Blood concentration of NO is measured as its stable metabolite nitrate and nitrite spectrophotometrically at 540 nm using Griess reaction. Blood levels of NO in age-matched patients admitted for diseases other than BA were taken as control after taking consent. The patients also underwent upper gastrointestinal endoscopy (UGIE) at presentation and at 6-month follow-up to look for presence/progress/resolution of esophageal varices. Esophageal varices were graded as follows:

Grade 1: small straight varices.

Grade 2: enlarged tortuous varices occupying less than one-third of the lumen.

Grade 3: large coil-shaped varices occupying more than one-third of the lumen.

Minilaparotomy was done and if the gallbladder was not atretic, peroperative cholangiogram was performed and diagnosis was confirmed. Portal pressure (PP) measurement was done after surgical access and before any mobilization of the liver and biliary tree. The system used for pressure measurement was Datex Ohmeda S/5™ monitor (range of measurement = –40 to 320 mmHg, accuracy: ±5% or ±2 mmHg) attached to a pressure transducer with continuous flush device and stop cock mounted at the level of the abdomen to avoid any error due to height (10 mmHg for every 13.6 cm difference). A 24-G intravenous cannula, attached to saline-filled pressure tubing, was inserted directly into one of the tributaries of gastroepiploic vein and PP was measured using the setup. KP was performed and wedge biopsy of liver was taken for histopathological examination. The patients were followed at 1 month, 3 months, and 6 months post KP. The patients were declared jaundice free when they achieved bilirubin levels <1.5 mg/dL. The data analysis was done using STATA software version 11 (StataCorp LP, Texas, USA). Blood levels of NO in BA patients were compared with the controls and within themselves using paired t-test. The correlation between the NO levels and PP was calculated using Spearman rank correlation coefficient. The P-value <0.05 was considered statistically significant.

RESULTS

A total of 34 BA patients were included in this study. There were 25 (74%) males and 9 (26%) females. Six patients were lost to follow-up, one in the first month and the remaining five at 3-month follow-up. The mean age of presentation was 2.7 months (range 1-4 months). There were 14 (41%) patients who presented before 60 days of life, 10 (29.5%) patients presented between 61 days and 90 days, and the remaining 10 (29.5%) patients presented after 90 days of age. All patients were clinically jaundiced and they passed high colored urine and acholic stools. On clinical examination, 15 (44%) patients had soft hepatomegaly, 19 (56%) patients had firm hepatomegaly, and 15 (44%) patients had splenomegaly. At 6-month follow-up, 12 (43%) patients were jaundice free while 16 (53%) patients continued to have jaundice. The mean bilirubin level at presentation was 10.10 ± 3.08 mg/dL (range 6.2-23.4 mg/dL). Post KP, there was a statistically significant fall in the mean bilirubin level from the preoperative levels. At 1 month, it was 7.16 ± 3.3 mg/dL, at 3 months it was 6.3 ± 3.9 mg/dL, and at 6 months it was 5.8 ± 4.8 mg/dL (P = 0.0001, 0.0003 and 0.0005 respectively). There was no significant difference in the mean preoperative bilirubin level of patients who were jaundice free at 6 months and those who continued to have jaundice (P = 0.2). The mean intraoperative portal venous pressure was 21.3 ± 5.4 mmHg (range 9-34 mmHg). The mean preoperative blood levels of NO in the study group and the control group were 18.48 ± 4.17 μmol/L and 4.64 ± 2.32 μmol/L, respectively. The difference was statistically significant (P < 0.001). NO levels at 1-month, 3-month, and 6-month follow-up were 11.94 ± 5.62 μmol/L, 10.79 ± 6.02 μmol/L, and 9.93 ± 6.53 μmol/L, respectively [Figure 1]. The fall in the NO levels in the immediate postoperative period was statistically significant (P < 0.005). However, there was no significant fall in NO levels during subsequent follow-up, i.e., between 1 month and 3 months and 3 and 6 months.

Figure 1

Blood NO levels in BA patients: Preoperative and after KP at 1-month, 3-month, and 6-month follow-up (Original)

On UGIE at presentation, 19 (56%) patients had esophageal varices. Grade 1 varices were present in 12 (35%) patients, while 7 (20%) patients had grade 2 esophageal varices. The mean PP in patients with absent varices was lower (17.46 mmHg) than the mean PP in patients with grade 1 varices (23.58 mmHg) and grade 2 varices (25.71 mmHg), and the difference was statistically significant between the groups with and without varices (P = 0.002). However, the difference was not statistically significant between the groups with grade 1 and grade 2 varices (P = 0.8). The mean preoperative NO levels in the patients who did not have any varices were 15.37 ± 3.42 μmol/L and in those with grade 1 and grade 2 varices were 20.05 ± 2.90 μmol/L and 22.45 ± 2.30 μmol/L, respectively [Figure 2]. The difference in blood NO level was again statistically significant between patients with and without varices (P = 0.001) but not among the patients with grade 1 and grade 2 varices (P = 0.33).

Figure 2

Preoperative blood NO levels in BA patients with respect to grades of esophageal varices at presentation (Original)

UGIE was also performed at 6-month follow-up in 28 patients. Three patients with grade 1 varices on initial UGIE showed regression after 6 months. One patient who continued to be clinically jaundiced at 6-month follow-up also had resolution of varices. UGIE at presentation and 6-month follow-up did not reveal any varices in two other patients who were still icteric at 6-month follow-up. These patients showed a trend toward normalizing bilirubin levels (mean bilirubin level at 6 months = 3.5 mg/dL) but were clinically jaundiced. Five patients with persisting jaundice had progression of varices at the end of 6 months. One developed new grade 1 varices, two had progression to grade 2, and two others had progression to grade 3 varices. The mean NO levels at 6-month follow-up in patients with absent varices were 4.85 ± 1.50 μmol/L and in those with grade 1, grade 2, and grade 3 varices were 14.78 ± 7.21 μmol/L, 15.72 ± 2.92 μmol/L, and 19.35 ± 1.08 μmol/L, respectively [Figure 3]. At 6-month follow-up, the difference in mean blood NO levels between groups with and without varices was also statistically significant (P < 0.001).

Figure 3

Blood NO levels in BA patients with respect to grades of esophageal varices at 6-month follow-up after KP (Original)

There was no significant correlation between intraoperative PP with preoperative bilirubin level (P = 0.28), serum glutamic oxaloacetic transaminase (SGOT) (P = 0.648), serum glutamic-pyruvic transaminase (SGPT) (P = 0.517), and alkaline phosphatase (P = 0.282). There was also no significant correlation between preoperative NO levels with bilirubin level (P = 0.1), SGOT (P = 0.372), SGPT (P = 0.452), and alkaline phosphatase (P = 0.254) at presentation.

DISCUSSION

BA is characterized by fibrotic obliteration of the lumen of all or a part of the extrahepatic biliary tree. Bile flow is impaired and liver inflammation, fibrosis, and cirrhosis sets in resulting in PHT. PHT is defined as pressure in the portal venous bed that exceeds 8 mmHg.[10] The mean intraoperative portal venous pressure in this study was 21.32 ± 5.4 mmHg and the median pressure was 21 mmHg (range 9-34 mmHg). Thus, all patients in the cohort had intraoperative PPs greater than 8 mmHg. However, clinical manifestations of PHT were not present in all; 19 (56%) patients had firm hepatomegaly, 15 (44%) patients had splenomegaly, and 19 (56%) patients had esophageal varices at presentation. None had variceal bleed at presentation. Although PHT is present in all BA patients, it varies in severity and manifests in some of the patients only. Variceal bleed, ascites, and hepatic decompensation are late manifestations of PHT and may not be present in the initial months. If untreated, patients will continue to develop liver fibrosis and cirrhosis. Thus, the PHT would worsen with age. The mean PP in patients <60 days old was 18.53 ± 4.45 mmHg, in patients aged between 61-90 days, it was 20.33 ± 3.07 mmHg and in those >90 days old, it was 26.5 ± 5.01 mmHg. The difference in PP was statistically significant between patients aged <90 days and those >90 days (P < 0.001), but it was not significant between patients who were <60 days old and those between 61-90 days (P = 1). Duche et al. also showed a positive correlation between age at surgery and PP index (PPI). Median PPI in 113 BA patients at Kasai operation was 15 cm of water (11 mmHg).[11] The higher median PP finding in our study may be due to the older age of presentation as compared to the above study that is 2.7 months versus 55 days. Shalaby et al. also measured portal venous pressure in BA patients at the time of KP and found a significant correlation between portal venous pressure and age at surgery.[12] In our study, the patients who underwent successful KP and were jaundice free at 6-month follow-up and had considerably lower mean intraoperative PP (16.75 ± 3.54 mmHg) compared to those who continued to have jaundice (23.94 ± 4.63 mmHg) (P < 0.001) [Figure 4]. This finding corroborates with Duche et al. study which states that BA patients with elevated portal pressure at the time of Kasai operation have lower chances of success of this procedure. They concluded that the measurement of the PPI during Kasai operation is simple and safe and better predicts the postoperative outcome than do the histological scores of liver fibrosis. However, Shalby et al. did not find any statistical difference in the median portal venous pressure between the group who cleared jaundice and those who did not clear their jaundice after KP.

Figure 4

Relation between intraoperative PP and outcome at 6-month follow-up in BA patients after KP (Original)

In the cirrhotic model with PHT, the circulatory state is hyperdynamic in the splanchnic bed and this is initiated by arterial vasodilatation. As demonstrated by Niederberger, the levels of NO in the peripheral venous blood of rats with PHT were elevated and these levels come down with the normalization of portal hemodynamics.[12] This increased NO production is responsible for the hemodynamic disturbances and sodium retention observed in the cirrhotic model. A study by Vejchapipat et al. showed that NO production was elevated in BA patients as compared to normal controls.[13] Goel et al. also demonstrated increased NO levels in peripheral blood of patients with extrahepatic portal vein obstruction, which declined after a successful lienorenal shunt surgery and remained low till the shunt was patent. A shunt block was indicated by rising NO levels.[9] This study shows that all BA patients had PHT and the mean NO levels in peripheral blood were higher in the study group as compared to the controls. This suggests that NO might be the vasoactive agent that is directly responsible for the hyperdynamic portal venous circulation. On further analysis, there seems to be a direct correlation between intraoperative PP and pre-operative NO levels that is increased NO levels are seen with higher PP (r = 0.89, P < 0.001) [Figure 5].

Figure 5

Relation between intraoperative PP and pre-operative blood NO levels in BA patients (Original)

Thus, NO levels in peripheral blood closely follow the PP. The regression equation NO = 4.79 + 0.64 PP mmHg, R2 = 0.69 suggest that a rise of 0.64 mmHg in PP would lead to an increase in NO levels by 1 μmol/L, with co-efficient of determination 0.69 [Figure 5]. In this study, the patients who were jaundice free at 6-month follow-up had lower mean NO levels preoperatively (P < 0.001) and during subsequent 1-month (P = 0.0014), 3-month (P = 0.0004), and 6-month (P = 0.0003) follow-up than those who still had jaundice [Table 1]. This is consonant with the outcome correlation between PP and successful KP. Thus, NO levels in peripheral blood can be used to monitor progress of PHT in the follow-up period. Patients with severe liver damage and an advanced degree of PHT and hence a higher NO levels, are more likely to have a failed KP.

Table 1

Pre-operative mean blood NO levels and at 1-month, 3-month, and 6-month follow-up in patients with and without jaundice at 6 months following KP. (Original)

CONCLUSIONS

All BA patients had varying severity of PHT at presentation. PHT worsens with age, and it has an adverse effect on the outcome of KP. Portal venous pressure measurement is a simple and safe method reflecting the degree of liver damage and the severity of PHT in BA patients. Peripheral blood NO levels closely follow PP, and the higher levels are associated with poor outcomes.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.