Background: Diagnosis and management of tuberculosis (TB) in patients with cirrhosis remains challenging. We studied the clinical spectrum, diagnosis, and management of TB along with the assessment of the diagnostic utility of various laboratory investigations in this cohort. Methods: A retrospective review of records of patients with cirrhosis (July 2017 and December 2019) was done. Out of 30 patients with cirrhosis and TB, 20 patients with pleural/peritoneal TB (cases) were compared with 20 consecutively selected spontaneous bacterial peritonitis (SBP) controls. Composite of clinical, laboratory, radiologic features and response to antituberculosis therapy (ATT) was taken as the gold standard to diagnose TB. Results: Extrapulmonary TB (EPTB) (n = 23, 76.7%) was more common. Overall, 9 (30%) patients presented with ATT-induced hepatitis. Patients with pleural/peritoneal TB had less severe hepatic dysfunction as compared to SBP group with significantly lower CTP [8 ± 1.5 vs. 9 ± 1.7 (P = 0.01)], MELD [16.3 ± 5.8 vs. 20.2 ± 6.6 (P = 0.02)] and MELD-Na [18.8 ± 5.9 vs. 22.5 ± 7.1 (P = 0.03)] scores. Median ascitic/pleural fluid total protein [2.7 (2.4-3.1) vs. 1.1 (0.9-1.2); P < 0.0001] and adenosine deaminase (ADA) levels [34.5 (30.3-42.7) vs. 15 (13-16); P < 0.0001] were significantly higher in the TB group. Total protein levels had a sensitivity and specificity 81% and 93.3%, respectively, at cut off value of >2 g/dl with an AUROC of 0.89 [(0.79-0.96); P < 0.001] whereas ADA levels at cutoff >26 IU/L showed 80% sensitivity and 90% specificity to diagnose pleural/peritoneal TB with an AUROC of 0.93 [(0.82-0.97); P < 0.001]. Only 11 (36.7%), and 8 (26.6%) patients showed positivity on GeneXpert and mTB-PCR, respectively. Patients with Child-Turcotte-Pugh scores of ≤7 and 8-10 tolerated well two and one hepatotoxic drugs, respectively. Conclusions: EPTB is more frequent in patients with cirrhosis. Relatively lower cutoffs of ascitic/pleural fluid total protein and ADA may be useful to diagnose EPTB in patients with high pretest probability. Individualized ATT with close monitoring and dynamic modifications is effective and well-tolerated.
Background: Diagnosis and management of tuberculosis (TB) in patients with cirrhosis remains challenging. We studied the clinical spectrum, diagnosis, and management of TB along with the assessment of the diagnostic utility of various laboratory investigations in this cohort. Methods: A retrospective review of records of patients with cirrhosis (July 2017 and December 2019) was done. Out of 30 patients with cirrhosis and TB, 20 patients with pleural/peritoneal TB (cases) were compared with 20 consecutively selected spontaneous bacterial peritonitis (SBP) controls. Composite of clinical, laboratory, radiologic features and response to antituberculosis therapy (ATT) was taken as the gold standard to diagnose TB. Results: Extrapulmonary TB (EPTB) (n = 23, 76.7%) was more common. Overall, 9 (30%) patients presented with ATT-induced hepatitis. Patients with pleural/peritoneal TB had less severe hepatic dysfunction as compared to SBP group with significantly lower CTP [8 ± 1.5 vs. 9 ± 1.7 (P = 0.01)], MELD [16.3 ± 5.8 vs. 20.2 ± 6.6 (P = 0.02)] and MELD-Na [18.8 ± 5.9 vs. 22.5 ± 7.1 (P = 0.03)] scores. Median ascitic/pleural fluid total protein [2.7 (2.4-3.1) vs. 1.1 (0.9-1.2); P < 0.0001] and adenosine deaminase (ADA) levels [34.5 (30.3-42.7) vs. 15 (13-16); P < 0.0001] were significantly higher in the TB group. Total protein levels had a sensitivity and specificity 81% and 93.3%, respectively, at cut off value of >2 g/dl with an AUROC of 0.89 [(0.79-0.96); P < 0.001] whereas ADA levels at cutoff >26 IU/L showed 80% sensitivity and 90% specificity to diagnose pleural/peritoneal TB with an AUROC of 0.93 [(0.82-0.97); P < 0.001]. Only 11 (36.7%), and 8 (26.6%) patients showed positivity on GeneXpert and mTB-PCR, respectively. Patients with Child-Turcotte-Pugh scores of ≤7 and 8-10 tolerated well two and one hepatotoxic drugs, respectively. Conclusions: EPTB is more frequent in patients with cirrhosis. Relatively lower cutoffs of ascitic/pleural fluid total protein and ADA may be useful to diagnose EPTB in patients with high pretest probability. Individualized ATT with close monitoring and dynamic modifications is effective and well-tolerated.
Authors: Karen R Steingart; Ian Schiller; David J Horne; Madhukar Pai; Catharina C Boehme; Nandini Dendukuri Journal: Cochrane Database Syst Rev Date: 2014-01-21
Authors: Catharina C Boehme; Pamela Nabeta; Doris Hillemann; Mark P Nicol; Shubhada Shenai; Fiorella Krapp; Jenny Allen; Rasim Tahirli; Robert Blakemore; Roxana Rustomjee; Ana Milovic; Martin Jones; Sean M O'Brien; David H Persing; Sabine Ruesch-Gerdes; Eduardo Gotuzzo; Camilla Rodrigues; David Alland; Mark D Perkins Journal: N Engl J Med Date: 2010-09-01 Impact factor: 91.245