Abstract
Summary
Tuberculosis is a common disease in India. Its prevalence is higher in patients with cirrhosis of the liver. This study was conducted to determine the prevalence of tuberculosis in patients with liver cirrhosis in Western India. The prevalence was fifteen times higher than in the general population. It was significantly higher in alcoholics. The response to treatment and outcome were found to be favourable.
Introduction
An increasing morbidity and mortality is forecast across the world from tuberculosis (TB), with the number of newly detected cases steadily rising from 7.5 million in 1997 to 11.9 million in 2005. 1 India is classified along with sub-Saharan countries as group four in view of the high disease burden of TB and the reported prevalence of all forms in India is 5.05/1000. 1
There is an increase in the incidence of TB in patients with cirrhosis of liver because of the immune system dysregulation and the treatment and outcome is worse in this group because of their compromised liver function.
This study was undertaken in order to determine the prevalence of TB in patients with cirrhosis of liver (Table 1).
Prevalence of tuberculosis in cirrhotic patients based on aetiology
Methods
This is a retrospective study. All patients with cirrhosis of the liver admitted to our centre from 2003 to 2008 were enlisted: all patients who were diagnosed with active TB after the diagnosis of cirrhosis of the liver were included in the study group. Patients with a past history of TB were only included in the study group if they had never previously defaulted on treatment and had been declared cured.
The patients in the study group were analysed on the basis of the aetiology of the cirrhosis, age, sex, site of TB, treatment given and outcome.
Results
A total of 667 (552 males, 115 females) patients were enlisted. Fifty had active TB after the diagnosis of cirrhosis and formed the study group. Thus, the prevalence of TB in the cirrhosis patients was 73.8/1000.
The majority of the patients in the study group were male (46, 92%) and the mean age was 46.82 years (range 7–70 years)
The aetiological factors responsible for cirrhosis of liver included: alcohol (30, 60%); postviral infection – hepatitis Band C (14, 28%); and other causes (6, 12%). None of the patients had HIV co-infection. Two patients were on interferon therapy when they developed reactivation of TB.
The TB sites included: pulmonary (20, 40%); pleural (14, 28%); peritoneal (9, 18%); cervical lymph nodes (3, 6%); liver (2, 4%); intestines (1, 2%); and vertebra (1, 2%). Thus, extrapulmonary TB was more common in the cirrhotic patients.
In 32 patients, TB was confirmed either histopathologically or by fluid analysis. In the remaining 18 patients, indirect indicators such as a history of fever and weight loss, high erythrocyte sedimentation rate (ESR), a positive Mantoux test or typical X-ray appearance were taken into account before starting treatment.
Thirty-five patients were given full dose, four drug anti-TB therapy; 14 were given a modified regimen due to a deranged liver function test. The modified regimen included ethambutol with a flouroquinolone with, or without, streptomycin or clarithromycin. One patient had multidrug resistant TB and was given HRZES (isoniazid, rifampicin, pyrizinamide, ethambutol, streptomycin), ciprofloxacin and clarithromycin.
The duration of treatment ranged from 6–18 months; seven patients (14%) defaulted on treatment; 45 showed improvement in clinical symptoms at the end of 2 months and four (8%) died as a result of cirrhosis-related complications. None had drug-induced liver failure. One patient was lost to follow-up.
Discussion
The prevalence of TB in our study is 73.8/1000, which is 15 times higher than the prevalence of all forms of TB in the general population. This is attributable to the immune dysregulation in cirrhotic patients. The mortality rate of 80/1000 population in cirrhotic patients with TB is also much higher than for the general population where it ranges from 0.5–2.3 per 1000 population. 1 This can be attributed to the complications of chronic liver disease such as spontaneous bacterial peritonitis, variceal bleed and encephalopathy which are prevalent in these patients. Moreover, the chances of drug-induced liver injury and acute liver failure are higher among these patients.
TB was more common among the males (8.36% in males versus 3.5% in females). It was significantly more common among the alcoholics (P value < 0.05) compared to other causes of cirrhosis as estimated by a chi-square test. This may be due to the poor hygiene, poor general care and immunosuppressed state of many chronic alcoholics.
Pulmonary TB was noted in 40% of cases and extrapulmonary TB in 60%. Yoo et al. have also reported a higher incidence of extrapulmonary involvement in cirrhotic patients. 2
Treatment response was favourable with clinical improvement in 90% of cases at two months. However, they were not followed up for a long period and side-effects that may have occurred later were not documented. TB patients with liver cirrhosis have a higher mortality and higher number of side-effects from the antituberculous therapy. Multidrug regimens are tolerated well if the side-effects are properly monitored. 3
A Danish study reported a high incidence of TB in patients with cirrhosis of the liver, especially among the >65 year age group. It has also been suggested that liver cirrhosis is an independent risk factor for TB and that patients with liver cirrhosis who acquire TB have a poor prognosis. 4
The drawback of our study is the small study group size. The study should be done on a larger scale in order to better determine the prevalence.
To conclude, the prevalence of TB is significantly higher in cirrhotic patients, especially alcoholics, extrapulmonary TB is more common and response to treatment is favourable.