Diagnostic Value of Adenosine Deaminase (ADA) in Ascitic Fluid for Peritoneal Tuberculosis: an Evidence-based Case Report

Author Dr. Juliyanti NPM: 1106024464

Chief of Ward Dr. Mulia

Supervisor Dr. Nadia Ayu Mulansari, SpPD

Department of Internal Medicine Faculty of Medicine University of Indonesia Jakarta, September 2012

Letter of Approval

Case Report
Diagnostic Value of Adenosine Deaminase (ADA) in Ascitic Fluid for Peritoneal Tuberculosis: an Evidence-based Case Report

Author
Dr. Juliyanti NPM: 1106024464

Has been approved for presentation at Cipto Mangunkusumo General Hospital in September 2012

Supervisor

Chief of Ward

Dr. Nadia Ayu Mulansari, SpPD

Dr. Mulia

Abstract Background: Adenosine deaminase (ADA) activity of ascitic fluid has been proposed as a useful non-culture method of detecting peritoneal tuberculosis. The aim of this case study is to determine the usefulness of ascitic fluid ADA level as a diagnostic tool for peritoneal tuberculosis (PTB) and define its cut-off point. Methods: This is evidence-based case report. Our clinical question is: what is the value of ADA in ascitic fluid for the diagnosis of PTB? To answer the question, we search PubMed Clinical Queries for the diagnostic evidence with keywords: ascites AND adenosine deaminase AND peritoneal tuberculosis. Inclusion criteria were prospective study, written in English, full text availability and focused on answering our clinical question. Results: At the end of searching, 5 relevant studies were critically appraised for validity, importance and applicability (VIA). All study showed good VIA criteria. An ascitic fluid ADA level cut-off point of 35 to 39 IU/L has sensitivity and specificity more than 90 %. Conclusion: Based on the evidence, our patient who had high ADA level (107 IU/L) in her ascitic fluid was then diagnosed with PTB. Anti-tuberculosis drugs were subsequently given. After 8 weeks of treatment she showed improvement on clinical condition. She was planned to have a 9 month course of anti-tuberculosis drugs. Key words: adenosine deaminase, ascites, peritoneal tuberculosis, evidence-based case report focus.6 PTB is manifested clinically as ascites of insidious onset, abdominal pain and fever.7 The non-specific symptoms associated with PTB and its challenging clinical course can interfere with a definitive diagnosis, and TB peritonitis is often confused with other intra-abdominal diseases. Delayed diagnosis increases the morbidity and mortality of PTB.8,9 Diagnostic tests for PTB are difficult and time-consuming because of paucity of M. tuberculosis in peritoneal fluid.10 A diagnosis of peritoneal TB requires histological confirmation of caseous granulomas. Bacteriological confirmation can be done by using ascitic fluid-derived acid-fast bacilli (AFB) smears as well as cultures for M. tuberculosis. Cultures for M. tuberculosis require 4 weeks of culture time, and AFB smears are too insensitive to meet current needs.10 A laparoscopy-guided biopsy can be used to obtain a rapid diagnosis of PTB; however, this method is associated with risks related to anesthesia and potential injury and bleeding.11 New methods with high specificity and sensitivity for the rapid diagnosis of extra-pulmonary TB would improve clinical

Introduction Tuberculosis (TB) causes some 2 million deaths per year world wide1 and is increasing in incidence in developed and developing countries. Although most persons present with pulmonary symptoms, presentation with extra-pulmonary symptoms is also possible. About 1015% of patients who are immune competent and 5070% of patients with acquired immunodeficiency syndrome (AIDS) present with extra-pulmonary symptoms. 2 Abdominal TB, which may involve the gastrointestinal tract, peritoneum, lymph nodes or solid viscera, constitutes up to 12% of extra pulmonary TB and 1-3% of the total.3 The peritoneum is the sixth most common extra pulmonary site.4 The reported incidence of peritoneal TB among all forms of TB varies from 0.1% to 0.7% worldwide5 and it can be expected to increase with the increasing incidence of TB worldwide. Peritoneal tuberculosis (PTB) usually results from the reactivation of latent TB in peritoneal foci that were established after hematogenous spread from a primary lung

outcomes of patients with extra-pulmonary TB12 A recent major advance in the diagnosis of peritoneal TB is the determination of adenosine deaminase (ADA) activity in the ascitic fluid. ADA is a purine-degrading enzyme that catalyses the deamination of adenosine in an irreversible manner, resulting in the production of inosine as a by-product. ADA is widely distributed in tissues and body fluids, and levels of ADA can be used to differentiate T cells from B cells; ADA levels are 1012 times higher in T cells than in B cells. ADA levels vary with the proliferative status and the maturity of cells.13 ADA is increased in tuberculous ascitic fluid because of the stimulation of T cells by the mycobacterial antigens. Levels in body fluids can be measured rapidly, and they might provide an alternative for the diagnosis of TB.14 Case Resume Female, 51 years old, with chief complaint of progressive abdominal swelling since 2 months prior to admission. She also complained of abdominal pain occasionally, malaise, anorexia and weight loss but had no history of night sweat or fever. She had no history of prior liver disease or alcoholic abuse. Symptoms of dyspnea, cough, chest pain or leg swelling were denied. She was diagnosed with diabetes mellitus and hypertension since 3 years before and was regularly on lisinopril and metformin. She was non-smoker. She was menopause 6 years ago and had no complaint of vaginal bleeding. Her urinary and bowel habit were fine. Physical examination revealed normal vital sign, dullness on percussion of left lower thorax with decreased breath sound, positive shifting dullness on abdomen, no stigmata of hepatic cirrhosis and no peripheral edema. Laboratory result showed leukocytosis (13.000 cell/mm3), 3 thrombocytosis (605.000 cell/mm ), elevated erythrocyte sedimentation rate (ESR, 40 mm/hour), moderate hyponatremia (128

mEq/L), hyperglycemia (256 mg/dl), normal kidney and liver function test, elevated CA 125 level (537 U/mL), normal albumin (3.5 g/dl) and increased globulin (4.81g/dl). Serological test for HIV, hepatitis B and C were negative. Fecal and urine analysis were normal Chest x-ray revealed left sided pleural effusion with no sign of parenchymal involvement. Abdominal ultrasounds showed free peritoneal fluid with no significant finding in abdominal viscera. An abdominal CT scan showed abundant free fluid but no other sign of abnormality. Gynecological study revealed normal result except adhesive fibrin on ascitic fluid. A diagnostic paracentesis was carried out, and a clear yellowish ascitic fluid was obtained that contained 1650 cells/ul, 99.9 % of which were mononuclear cells (lymphocyte). LDH was very high, proteins were increased, serum-asites albumin gradient was < 1,1 g/dl (0,5 g/dl) and A Ziehl-Neelsen stain was negative. Ascitic ADA level was high (107,1 U/L). In practice, we are facing with diagnostic problem of this patient with exudative asites and lymphocyte predominant pleiocytosis her ascitic fluid that suggestive a peritoneal tuberculosis but ascitic fluid revealed negative for acid fast bacilli staining. Gold standard for diagnosis of PTB is positive acid fast bacilli staining on ascitic fluid or positive culture for M. tuberculosis from ascitic fluid or specimen of peritoneum or histological confirmation of caseous granulomas from peritoneum biopsy. But the definitive diagnostic procedure is either time consuming or invasive. We tried to find other rapid diagnostic tool for this patient. We were informed that ADA level can be used to determine whether her ascites was due to TB infection or not, but we still did not know about the accuracy of this assay. Therefore, we formulate our clinical question and try to find the answer about effectiveness of an ADA assay as rapid method for the diagnosis of PTB. The aim of the this study

were (1) to assess the accuracy of ADA in diagnosing peritoneal tuberculosis and differentiating it from other etiologies (2) initiate early anti-tuberculosis treatment in order to improve patients prognosis and reduce the spread and sequel of this infection. Clinical question What is value of ascitic fluid ADA for the diagnosis of peritoneal tuberculosis? Methods 1. Search strategy and selection We searched Pubmed (see Figure 1) to identify all published studies on 27 june 2012 using various terms within the domain (asites), determinant (ADA) and outcome (peritoneal tuberculosis) with clinical queries about diagnosis and narrow filter. Search details was: Diagnosis/Narrow[filter] AND (("ascites"[MeSH Terms] OR "ascites"[All Fields]) AND "adenosine

deaminase"[All Fields] OR ADA[All Fields] AND "peritoneal tuberculosis"[All Fields] OR "tuberculosis peritonitis"[All Fields] OR "tuberculous peritonitis"[All Fields] OR tubercular peritonitis[All Field] Thirty seven articles were retrieved initially (See Fig. 1 for an overview of the selection of articles). The first selection was based on language and abstract availability. The remaining articles were reviewed further by applying inclusion and exclusion criteria. Seven articles relevant to our clinical question were left, out of which 2 retrospective studies were excluded. The remaining 4 articles were prospective study and 1 article with the highest evidence level (meta-analysis). 2. Critical appraisal We appraised meta-analysis of diagnostic research using Q-FAST tools. Other articles included in this review were each appraised using standardized validity criteria for diagnostic research.15,16 Table 1 and Table 2 details our assessment of validity of studies.

37 articles

10 articles was excluded (not written in English or not provide abstract ) 11 articles were excluded (only provide abstract) 9 articles were excluded (not relevant to our clinical question)

27 articles selected

16 articles have full text 7 articles relevant to our clinical question 5 articles included in the critical appraisal
Figure 1. Flow chart of searching strategy

2 articles were excluded (retrospective study)

Table 1. Rapid critical appraisal of meta-analysis using Q-FAST tool Item Summary of Riquelme et al, 2006 Q-question The study was based on clear clinical question and it is similar to ours F-finding Inclusion criteria and search methods are stated in the methods section, inclusion criteria were based on the clinical question. A primary search of the literature was conducted including MEDLINE (PubMed) and hand searching, no limits regarding language 4 prospective studies were included in the meta-analysis A-appraise Individual prospective studies were critically appraised by two independent reviewers, agreement between reviewers for article selection was measured as kappa coefficient of only 0.43 (moderate), probable due to small number of selected articles. S-synthesis The paper includes a clear summary table of the included studies and heterogeneity analysis was done. Four individual studies were homogen. T-transferability The result of study was applicable on our patient because similarity in the domain, determinant and outcome. Table 2. Validity assessment of prospective study Study Independent and Appropriate blind comparison spectrum of patient Saleh et al, 2012 Sharma et al, 2006 Burgess et al, 2001 Sathar et al, 1995 Result We found 1 meta-analysis and other 4 prospective cohort studies that were not included in the meta-analysis. See table 3 that summarizes the importance of study and table 4 shows applicability of the valid and important result. Riquelme et al conducted a systematic review of the literature and a meta-analysis of 4 prospective studies to determine usefulness of ADA levels in diagnosing peritoneal TB and the optimal cut-off point of ADA levels to guide clinical practice.2 In that study, out of 264 patients with ascites, 50 (18.9 %) had ascites caused by TB that confirmed by bacteriologic or histologic methods and all of them had ADA value 39 IU/L measured by the Giusti method. Only 6 patients with an ADA value over this cut-off point had other diagnoses, while ADA values 56 IU/L were Yes Yes Yes Yes Yes Yes Yes Yes

Reference standard ascertained regardless of the index test result Yes Yes Yes Yes

associated only with peritoneal TB. ADA levels showed high sensitivity (100%) and specificity (97%) with optimal cut-off point 39 IU/L and the likelihood ratios (LRs) were 26.8 and 0.038 for values above and below this cut-off. In a more recent study by Saleh et al, ascitic fluid and blood samples were collected from each of 41 patients with a presumptive diagnosis of PTB with ascites who were admitted to Mansoura University Hospital.18 Fluid samples were examined biochemically, cytologically (cell count) and microbiologically (ZiehlNeelsen stain and TB culture in LowensteinJensen media), and ADA levels were determined by using Giusti method. A final clinical diagnosis of PTB was established in fourteen (34%) patients; these patients were subclassified as definite (positive culture for Mycobacterium tuberculosis; 8 patients), highly probable (4 patients) and probable (2 patients) for PTB.

A receiver operating characteristic (ROC) curve showed that a cut-off value of 35 IU/L for the ADA level produced the best results as a diagnostic test for PTB, yielding sensitivity 100 %, specificity 92.6 %, positive predictive value (PPV) 87.5 % and negative predictive value (NPV) 100 %. Other study evaluated the diagnostic accuracy of ascitic fluid interferon-gamma (IFN-gamma) and adenosine deaminase (ADA) assays in the diagnosis of tuberculous ascites. 19 Ascitic fluid from patients with proven tuberculosis (TB) (n = 31) and nonTB ascites (n = 88) was analyzed for IFNgamma and ADA levels. Areas under the ROC curves (AUCs) for the two biologic markers were compared. Levels of ascitic fluid IFN-gamma, median (range): 560 (1041600) pg/mL vs. 4.85 (0-320) pg/mL (p <0.001), and ADA, median (range): 58 (16331) IU/L vs. 10 (0-59) IU/L (p = 0.001), were significantly different between TB and non-TB groups. IFN-gamma and ADA assays showed equal sensitivity (0.97) and differed marginally in specificity (0.97 vs. 0.94). Difference in AUCs was not significant (0.99 vs. 0.98, p <0.62). For differentiating TB from non-TB ascites,

optimal cutoff points were 112 pg/mL for IFN-gamma and 37 IU/L for ADA. Another study by Burgess et al was done on 178 paired ascites and serum specimens from adult patients.20 Specimens were evaluated for biochemistry, ADA, microbiology and cytology. Diagnoses were made according the pre-determined criteria. The median (range) ADA activity in the tuberculous group was 61.6 (17.5-115.0) U/L and was significantly higher than in any other diagnostic group (p<0.05). Best ROC curves cut-off level of 30 U/L for the diagnosis of PTB was found to yield sensitivity 94 % and specificity 92%. One study evaluated the adenosine deaminase (ADA) activity in 30 patients with PTB, 21 patients with ascites due to a malignant disorder, and 41 patients with cirrhosis.21 The ADA activity was significantly (p < 0.0001) higher in the PTB group (mean: 101.84 U/l) than in the control groups (cirrhosis (mean:13.49 U/l) and malignancy (mean: 19.35 U/l)). A cut off value of > 30 U/l gave the ADA test a sensitivity of 93% (26 of 28) a specificity of 96% (51 of 53), a PPV of 93%, a NPV of 96%, and a test accuracy of 95%.

Table 3. Importance of study Study N Best ROC curve cut-off ADA value (IU/L) Riquelme, et al 264 39 Saleh et al, 2012 41 35 Sharma et al, 2006 119 37 Burgess et al, 2001 178 30 Sathar et al, 1995 92 30

AUC

Sn (%) 100 100 97 94 93

Sp (%) 97 92,6 94 92 96

LRs (+) 26.8 N/A N/A N/A N/A

PPV (%) N/A 87,5 N/A 57 93

NPV (%) N/A 100 N/A 99 96

0.99 0.99 0.98 0.92 0.95

N/A: not available Sn: sensitivity; Sp: specificity; LRs: likelihood ratios; PPV: positive predictive value; NPV: negative predictive value.

Table 4. Applicability of study Study The test is available, affordable, accurate and precise in our setting Saleh et al, 2012 Yes Sharma et al, 2006 Yes Burgess et al, 2001 Yes Sathar et al, 1995 Yes

Ability to generate an estimate of patients pre-test probability Yes Yes Yes Yes

The resulting post-test probabilities will affect management and help patient Yes Yes Yes Yes

Discussion After searching, scanning and appraising the studies found in PubMed, there were five studies left that particularly fitted our research question. The authors agreed that ADA is increased in tuberculous ascitic fluid because of the stimulation of T cells by the mycobacterial antigens. The results mentioned above show some variation but sensitivity and specificity levels over 90% have been reported in all studies included. (see Table 3). With respect to our search strategy, we admit that we might have missed relevant articles. Moreover, we excluded all hits that were not in English publication and not available with full text on the Internet, which can be a negative contributor. When applying the results of best available evidence to our patient mentioned above, we have to consider that the metaanalysis included population from Chile and India, where a higher frequency of TB was observed at the time when the studies performed. This setting was similar to ours, where TB is endemic. Therefore, we can extrapolate the result on our patient. Hence, our patient with clinical features suggestive of PTB, from speculation (we didnt have the reported prevalence in diabetic population), she has a 50% pretest probability of having PTB. If the ascites fluid study shows an ADA level of 107 IU/L,

she therefore has a 96 % posttest probability of having PTB. This result was obtained applying the LR to the nomogram (see figure 2) adapted from Fagan.22 The clinical relevance of this test is evident ever because it may avoid the need to confirm the diagnosis with invasive techniques such as laparoscopy. Then, we decide to treat our patient with a 4-drug combination of antituberculosis in accordance with the recommendation of the Center for Disease Control (CDC). However, the CDC in United State insists the need of a sensitivity in vitro test in every patient with PTB to determine the epidemiology of Mycobacterium tuberculosis.23 Importantly, it is known that even using the most advanced techniques to isolate the Mycobacterium, the rate of positive test is still low and we consider ADA activity is a practical and useful approach to make therapeutic decision in patients with suspected PTB.2 After 8 weeks of anti-tuberculosis treatment, on the follow up of our patient, she recovered from constitutional symptoms and her abdominal circumference was reduced from 112 cm to 102 cm. She was planned to have a 9-month course of antituberculosis drugs that consist of 2 months of initial phase and 7 months of continuation phase.

Figure 2. Nomogram for interpreting diagnostic test results, adapted from Fagan.22 Posttest probability (96%) for our patient with pretest probability 50% and LRs 26.8. Conclusion and recommendation From the evidence, we found that measurement of an ascitic ADA activity yields a high rate of diagnostic accuracy for PTB and it is an easy available test. It may help clinicians to decide whether or not to treat patients with ascites who clinically suspicious for PTB. To start empirical treatment on a patient with high ADA value in ascitic fluid seems to be a good approach while waiting for the results of mycobacterial cultures or biopsies. We do recommend increasing the utilization of this test in the diagnostic work-up of patients with suspected PTB. Finally, we have demonstrated that individual clinical questions of patients can be answered fairly accurately using relatively simple evidencebased methodology. Reference 1. WHO. Tuberculosis Fact Sheet No. 104 Global and Regional Incidence, March 2006. Geneva: World Health Organization. Riquelme A, Calvo M, Salech F, Valderrama S, Pattillo A, Arellano M, Arrese M, et al. Value of adenosine 6

2.

deaminase (ADA) in ascitic fluid for the diagnosis of tuberculous peritonitis: a meta-analysis. J Clin Gastroenterol. 2006;40:70510. 3. Sheer TA, Coyle WJ. Gastrointestinal tuberculosis. Curr Gastroenterol Rep. 2003;5:273-8 4. Hillebrand D J, Runyon B A, Yasmineh W G and Rynders G P. Ascitic :uid adenosine deaminase insensitivity in detecting tuberculous peritonitis in the United States. Hepatology. 1996;24:1408-12. 5. Chow KM, Chow VCY, Szeto CC. Indication for peritoneal biopsy in tuberculous peritonitis. Am J Surg. 2003;185:56773. 6. Mehta JB, Dutt A, Harvill L, Mathews KM. Epidemiology of extrapulmonary tuberculosis. A comparative analysis with pre-AIDS era. Chest. 1991;99:11348. 7. Chow KM, Chow VCY, Hung LC, Wong SM, Szeto CC. Tuberculous peritonitis-associated mortality is high among patients waiting for the results of mycobacterial cultures of ascitic fluid samples. Clin Infect Dis. 2002;35:409 13. 8. Ak n Atac alis mer A, Iskender S. Diagnostic tools for tuberculous peritonitis. Turk J Gastroenterol. 2000;11:1625. 9. Sanai FM, Bzeizi KI. Systematic review: tuberculous peritonitis presenting features, diagnostic strategies and treatment. Aliment Pharmacol Ther. 2005;22:685700. 10. Inadomi JM, Kapur S, Kinkhabwala M, Cello JP. The laparoscopic evaluation of ascites. Gastrointest Endosc Clin N Am.2001;11:7991. 11. Vogel Y, Bous JC, Winnekendonk G, Henning BF. Tuberculous peritonitis in a German patient with primary biliary cirrhosis: a case report. J Med Case Reports. 2008;2:32. 12. Portillo- o me L orris L anduro A. Rapid and efficient detection of extra-pulmonary Mycobacterium

13.

14.

15.

16.

17.

18.

19.

20.

21.

tuberculosis by PCR analysis. Int J Tuberc Lung Dis. 2000:4:36170. Dwivedi M, Misra SP, Misra V, Kumar R. Value of adenosine deaminase estimation in the diagnosis of tuberculous ascites. Am J Gastroenterol. 1990;85:11235. Giusti G. Adenosine deaminase. In: Bergmeyer HU, editor. Methods of Enzymatic Analysis. New York: Academic Press;1974. p. 1092-9. Bhandari M, Guyatt GH. How to appraise a diagnostic test. World J. Surg. 2005;29:561-6. Battaglia M, Bucher H, Egger M, Grossenbacher F, Minder C, Pewsner D, et al. The Bayes Library of Diagnostic Studies and Review. 2nd edition. Switzerland: University of Basel Press;2002. Whiting P, Rutjes AWS, Reitsma JB, Bossuyt PMM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Medical Research Methodology. 2003;3:25. Saleh MA, Hammad E, Ramadan MM, Abd El-Rahman A, Enein AF. Use of adenosine deaminase measurements and QuantiFERON in the rapid diagnosis of tuberculous peritonitis.J Med Microbiol. 2012;61(4):514-9. Sharma SK, Tahir M, Mohan A, SmithRohrberg D, Mishra HK, Pandey RM. Diagnostic accuracy of ascitic fluid IFNgamma and adenosine deaminase assays in the diagnosis of tuberculous ascites. J Interferon Cytokine Res. 2006;26(7):484-8. Burgess LJ, Swanepoel CG, Taljaard JJ. The use of adenosine deaminase as a diagnostic tool for peritoneal tuberculosis.Tuberculosis. 2001;81(3):243-8. Sathar MA, Simjee AE, Coovadia YM, Soni PN, Moola SA, Insam B, et al. Ascitic fluid gamma interferon concentrations and adenosine deaminase activity in tuberculous peritonitis. Gut.

1995;36(3):419-21. 22. Fagan TJ. Letter: nomogram for Bayes theorem. N Engl J Med. 1975;293:257. 23. Center for disease control (CDC). National action plan to combat multidrug resistant tuberculosis. JAMA. 1993;270:694.