Oesophageal varices in patients with liver cirrhosis attending a major tertiary hospital in Ghana

Introduction Oesophageal variceal bleeding is a potentially fatal consequence of portal hypertension in cirrhotic patients. In Ghana, bleeding oesophageal varices (OV) are a significant cause of acute upper gastrointestinal bleeding with comparatively high mortality. This study was to determine the prevalence of OV and its clinical correlate in cirrhotic patients. Methods This was a cross sectional hospital based study of 149 subjects with liver cirrhosis from 5thNovember, 2015 to 4th November, 2016. Demographic and other clinical data were collected using standardized questionnaire. Liver function, full blood count, HBsAg and anti-HCV Ab tests were done for all patients. All patients underwent an abdominal ultrasound to assess liver and document ascites. Upper GI endoscopy (UGIE) was done to screen for and grade varices. Results A total of 149 patients with a mean age of 45 ± 12.28 years were evaluated. There were 77.85% and 22.15% men and women respectively, with a male to female ratio of 3.5:1. By Child-Pugh Classification, 12 (8.16%) patients were in class A, 64 (43.54%) in class B and 71 (48.3%) in class C at presentation. On UGIE, 135 (90.60%) had varices and 14 patients (9.40%) had no varices. One hundred and eleven of the varices (82.22%) were large varices and the rest (17.78%) small varices. Conclusion Majority of cirrhotic patients present late with advance disease to this referral centre. Most have large varices on their first screening endoscopy. Prophylactic treatment should be considered for all cirrhotics especially patients with decompensated liver cirrhosis when UGIE cannot be done immediately.


Introduction
Liver diseases, especially liver cirrhosis, are common in Ghana and Africa due to the high prevalence of chronic viral hepatitis [1]. Portal hypertension is one of the most significant complications of liver cirrhosis. Gastroesophageal varices are the most important clinical manifestation of portal hypertension. Oesophageal varices (OV) are present at diagnosis in approximately 50% of cirrhotic patients, being more common in Child-Pugh class C patients compared to Child-Pugh class A patients (85% versus 40%) [2,3]. De novo formation of varices occurs at a rate of 8% per year [3,4] and the strongest predictor for development of varices in those with cirrhosis who have no varices at the time of initial endoscopic screening is an hepatic venous pressure gradient (HVPG) > 10mmHg [4]. Once varices form, they progress from small to large at a rate of 5-12% per year [5].
Rupture and bleeding of varices portends a poor outcome. Once OV have been identified in a patient with cirrhosis, the risk of variceal bleeding is 25-35% and accounts for 80-90% of bleeding episodes in these patients [5][6][7]. Bleeding caused by rupture of the OV is associated with a mortality rate of 20% when patients are treated optimally in hospital [8]. The 6 week mortality with each episode of variceal hemorrhage is approximately 15 to 20%, ranging from 0% among patients with Child class A disease to approximately 30% among patients with Child class C disease [9,10] and up to 70% of untreated patients die within 1 year of the initial bleeding episode [11]. The degree of hepatic decompensation (Child class) is the most important determinant of long term survival after a variceal haemorrhage [12,13]. Survivors of an episode of active bleeding have a 70% risk of recurrent hemorrhage within one year [14]. The poor outcome of variceal bleeding makes identification of those at high risk and prevention of a bleeding episode critically important [15]. Early diagnosis of OV before the first bleed is essential as studies of primary prophylaxis with nonselective beta blockers have clearly shown that the risk of variceal bleeding can be reduced by 50% to about 15% for large oesophageal varices [16]. In Ghana, bleeding OV are significant cause of acute upper gastrointestinal bleeding with comparatively high mortality [17]. Deaths from OV bleeds, the leading cause of death from acute upper gastro-intestinal haemorrhage rose from 46% in 2010 to 76% in 2013 [17]. Outcomes following acute upper gastro-intestinal bleeding were dismal with some 38% of fatalities occurring within the first 24 hours [17]. There are no published studies describing the prevalence of OV in liver cirrhosis among Ghanaians. This study is therefore to determine the prevalence of OV and its clinical correlates in cirrhotic patients.

Methods
A formal approval of this study was obtained from the Ethical and for it were excluded from the study. Those who refuse to consent to the study were also excluded. Patients who met the criteria above were selected using the convenience sampling method After thoroughly explaining the study to patients, those who gave their consent were administered appropriate questionnaire (sociodemographic data and clinical history including alcohol use of the patients were obtained). Alcoholic aetiology was made when patient's declared alcohol consumption was more than 21 units of alcohol for men or 14 units for women per week when measurable or local alcohol beverage consumption was three times per week in the past five years and correlated with biological abnormalities related to alcohol consumption [18]. Ten (10) mls of blood was taken on a single occasion for haematological, biochemical and serological workup. Full blood count and liver chemistry and function test were done. For each patient, a modified Child-Pugh score was calculated [19]. All patients were tested for HBsAg and anti-HCV Ab to determine the cause of liver cirrhosis. Serum antinuclear antibodies (ANA), anti-smooth muscles antibodies (ASMA), serum IgG and anti-liver kidney microsomal (LKM) tests were carried out for patients suspected to have autoimmune hepatitis. Viral aetiology of cirrhosis was considered when one of these serological tests of HBV (HBsAg) or ultrasound scan after overnight fast and the following details were recorded: maximum vertical span of the liver; nodularity of liver surface; spleen size (Length of its longest axis); and presence of ascites. UGIE was done for assessment of OV. The presence and size of OV were recorded if present. The sizes of the varices were subdivided into two classes-small and large. Small OV defined as varices that flatten with insufflation or minimally protrude into the oesophageal lumen, while large OV defined as varices that protrude into the oesophageal lumen and touch each other (presence of confluence), or that fill at least 50% of the esophageal lumen [20].
Optic and video endoscopes (GIF XQ 10 Olympus) were used.
Statistical Package for Social Sciences (SPSS) version 18 data entry template was used for statistical analysis. Descriptive statistics was run for all the variables and data presented in appropriate graphs and tables. Chi square test and T test statistics were used to determine the level of association. A multivariate logistic regression analysis was done for selected binary variables to determine if any of them was a predictor of oesophageal varices. All p-values for this work was twotailed with p < 0.05 as significant.

Results
A total of 149 cirrhotics were evaluated. This consisted of 33 female and 116 male patients with a male to female ratio of 3.5:1. The mean age of the patients was 45 ± 12.28 years (Table 1) Table 1. Jaundice, haematemesis, melena stools, weight loss and anorexia were found to be significantly associated with presence of OV. However, no significant association was found between pedal oedema, ascites, encephalopathy and the presence of varices. In multivariate analysis none of these clinical features were found to be statistically significant ( Table 2). The following laboratory parameters; AST, ALP, GGT were found to be associated with presence of OV among the cirrhotic patients. However none of these were statistically significant on multivariate analysis (Table 2).

Discussion
The mean age of the respondents was 45 ± 12.28 years. This is worrying but not surprising; worrying because this age bracket constitutes the working population who unfortunately are bearing the brunt of liver cirrhosis with implications on productivity and burden on the society. As expected however, this is the period when complications of common afflictions of the liver begin to occur. Similar age cohorts have been reported from similar studies in this country [21] and other Africa countries in patients with liver cirrhosis [18,22].
However, the mean age in other studies done in western countries were higher than this study [23,24]. Variations in the mean age in different geographic regions are likely to be related to differences in the aetiologies and especially prevalence of hepatitis viruses in the populations, as well as the timing of the spread of the viral infection and the age of individuals at the time of the infection. Chronic hepatitis B was either alone or in combination with other risk factors was identified as the leading cause of cirrhosis whilst HCV was a less important (6.71%) cause. This is consistent with the high prevalence of HBV (13%) and relatively low prevalence of HCV(3.0%) in Ghana [25,26]. This finding is in parallel with reports in the literature from West Africa and other hepatitis B endemic countries [21,27]. Alcohol abuse was the second common cause of chronic liver disease in this study which implies that alcohol is a significant cause of liver cirrhosis in patients attending clinic at KBTH. This is a public health concern; therefore society should be educated on the harmful effect of alcohol abuse on the liver. A study by Abel et al. in Ethiopia [28], found HBV and HCV as the major causes of liver cirrhosis whiles in Sudan [29] alcohol abuse and HBV were the common causes. However in Brazil

Authors' contributions
Amoako Duah, the principal investigator of the project, was involved in concept design, patient recruitment, data analysis and drafted the manuscript. Kofi Nyaako Nkrumah and Kenneth Tachi guided the principal investigator in concept design, assisted in the analysis and interpretation of the data, critically revised the article and provided final approval of the article.     0.577 Data are presented as frequencies and percentages. Categorical data were analyzed using Chi-squared and continuous data were analyzed using T-test statistics all at 95% confidence intervals. *Data expressed as Mean ± SD