Pleural Fluid Adenosine Deaminase Interpreter
Pleural Fluid Adenosine Deaminase Interpreter
Pleural adenosine deaminase is a good test for excluding tuberculous pleural effusion and a much weaker one for confirming it — and which of those two jobs it is doing depends on how much tuberculosis walks through your door. This page interprets the result against your laboratory’s own cut-off and your own pre-test probability.
Pleural fluid adenosine deaminase
ADA + cut-off + prevalence → interpretationADA 62 IU/L, laboratory cut-off 40 IU/L, low pre-test probability, lymphocyte-predominant, no pus
What the test is, and what the numbers mean
post-test odds = pre-test odds × LR
- adenosine deaminase
- an enzyme of purine metabolism released by activated lymphocytes and monocytes. Its ADA2 isoform predominates in tuberculous effusions, which is why the total activity rises in a lymphocytic tuberculous pleuritis
- 0.92 and 0.90
- the pooled sensitivity and specificity across 174 studies and 27,009 patients — 95% confidence intervals 0.90–0.93 and 0.88–0.91, diagnostic odds ratio 97.4
- 40 ± 4 IU/L
- the most studied threshold: 65 studies, sensitivity 0.93 (0.90–0.95), specificity 0.90 (0.87–0.91). Gives LR⁺ 9.3 and LR⁻ 0.078
- above 65 IU/L
- sensitivity 0.86 (0.61–0.96), specificity 0.94 (0.80–0.99) — but on four studies only, hence the very wide intervals. LR⁺ 14.3 and LR⁻ 0.149: better at confirming, worse at excluding
- method
- Giusti’s colorimetric method and automated assays are not numerically interchangeable. An automated assay reads roughly 30% lower on the same fluid, which is why published cut-offs of 30, 35 and 40 IU/L all coexist
- risk of bias
- the meta-analysis records that every one of its 174 included studies was at high risk of bias and that no firm inference could be drawn about the relative clinical utility of different thresholds. Read the intervals above with that in mind
Worked example
ADA 62 IU/L, laboratory cut-off 40 IU/L, low pre-test probability, lymphocyte-predominant, no pus
62 IU/L is above 40 and outside the ±10% indeterminate band, so this is a positive test
At the 40 ± 4 threshold, LR⁺ = 0.93 ÷ (1 − 0.90) = 9.3
Starting from a pre-test probability of 5%: odds 0.053 × 9.3 = 0.49
Post-test probability = 0.49 ÷ 1.49 = about 33%
So two thirds of patients with this result, in this setting, do not have tuberculosis
Had the same result come from a setting where half the effusions are tuberculous, the post-test probability would be about 90% — the number did not change, the answer did
The same result, read against different pre-test probabilities
| Pre-test probability | ADA above 40 ± 4 IU/L | ADA below 40 ± 4 IU/L | ADA above 65 IU/L | ADA below 65 IU/L |
|---|---|---|---|---|
| 50% | 90% | 7.2% | 94% | 13% |
| 30% | 80% | 3.2% | 86% | 6.0% |
| 20% | 70% | 1.9% | 78% | 3.6% |
| 5% | 33% | 0.4% | 43% | 0.8% |
| 1% | 8.6% | 0.08% | 13% | 0.15% |
What the two thresholds are each good for
| Threshold | Studies | Sensitivity | Specificity | LR⁺ | LR⁻ | Best used to |
|---|---|---|---|---|---|---|
| Pooled, all thresholds | 174 | 0.92 (0.90–0.93) | 0.90 (0.88–0.91) | 9.2 | 0.089 | Describe the test in general — not to make a decision, since it mixes thresholds |
| 40 ± 4 IU/L | 65 | 0.93 (0.90–0.95) | 0.90 (0.87–0.91) | 9.3 | 0.078 | Exclude tuberculous effusion. The authors’ own conclusion is that it is less useful to confirm one |
| Above 65 IU/L | 4 | 0.86 (0.61–0.96) | 0.94 (0.80–0.99) | 14.3 | 0.149 | Confirm — at the price of excluding less well, and on four studies with intervals wide enough to drive a bus through |
Other reasons a pleural adenosine deaminase is high
| Cause | How often it crosses the tuberculosis cut-off | How to tell |
|---|---|---|
| Parapneumonic effusion and empyema | More than 40% of parapneumonic effusions in a 2,117-patient series | Neutrophil-predominant, acute illness, low pH and low glucose. Values above 250 U/L occur in empyema but not in tuberculosis |
| Lymphoma | About half of lymphomatous effusions in the same series | Cytology and flow cytometry. Again, above 250 U/L points here rather than at tuberculosis |
| Rheumatoid pleurisy | Recognised cause, frequency not quantified | Known rheumatoid disease, very low pleural glucose, very high LDH, characteristic cytology |
| Other malignancy | Reported, particularly in older patients | The commonest alternative diagnosis for a lymphocytic exudate in a low-incidence country. Cytology, then biopsy |
| Elderly or critically ill patients with tuberculosis | The opposite problem — falsely LOW | A reported cause of adenosine deaminase-negative tuberculous pleuritis. Do not use the test to exclude in this group |
A rule-out test that is usually read as a rule-in
Pleural adenosine deaminase has an unusually solid evidence base for a body fluid test: 174 studies and 27,009 patients pooled in 2019, giving a sensitivity of 0.92 and a specificity of 0.90 for tuberculous pleural effusion. Those are good numbers, and they are the reason the test is offered on almost every pleural fluid panel. They are also the reason it gets misused, because a sensitivity and a specificity say what the test does to a population and not what a particular result means for the patient in front of you.
What the result means depends on where you are. At the commonest threshold the positive likelihood ratio is 9.3 and the negative likelihood ratio is 0.078 — so the test moves a probability up by about ninefold and down by about thirteenfold, and whether that lands anywhere useful depends entirely on where it started. Where tuberculosis causes half the pleural effusions investigated, a positive result takes the probability from 50% to 90% and a negative one to 7%. Where it causes one in twenty, a positive result gets you to 33% and a negative one to 0.4%. Same assay, same number, opposite conclusions: in the high-incidence setting the test confirms, and in the low-incidence setting it excludes and barely confirms at all. That asymmetry is not this page’s editorial view; it is the meta-analysis authors’ own, who wrote that the 40 ± 4 IU/L threshold “can be used to exclude” tuberculous effusion and is “less useful to confirm” it, and who recommend that values below 36–44 IU/L be evaluated in detail for a non-tuberculous disorder.
There is a second reason to be careful with the raw number, which is that it is not really one number. Giusti’s colorimetric method and modern automated assays correlate closely but do not agree numerically: a validation study against the Diazyme automated assay found it read about 30% lower than Giusti on the same fluid, so the same sample supports a cut-off of 40 IU/L by one method and 30 IU/L by the other. A third widely cited series used 35 U/L on a non-Giusti kinetic assay. Adenosine deaminase has no international reference method and no standardised calibrator, and a threshold copied out of a paper into a different laboratory is a threshold in the wrong units. This page asks for your own cut-off for that reason, and treats anything within a tenth of it as indeterminate.
Finally, adenosine deaminase is an activated-lymphocyte enzyme and not a mycobacterial one, so anything that fills a pleural space with activated lymphocytes and monocytes raises it. In the largest single-centre series more than 40% of parapneumonic effusions and about half of lymphomatous effusions exceeded the tuberculosis cut-off, and values above 250 U/L occurred only in empyema and lymphoma — never in tuberculosis. Rheumatoid pleurisy does it too. The corollary is that the test belongs inside a workup rather than at the end of one: establish first that the fluid is an exudate with Light’s criteria, look at the differential and the pH and glucose, and send cytology and mycobacterial culture whatever the adenosine deaminase says. One last caution the meta-analysis itself insists on: all 174 of its included studies were judged at high risk of bias, and it drew no firm conclusion about which threshold is best.
Frequently asked questions
What is a normal pleural fluid adenosine deaminase?
There is no single figure, because the assay is not standardised. The most studied decision threshold is 40 IU/L by Giusti’s colorimetric method; automated assays read about 30% lower and use cut-offs nearer 30 to 35 IU/L. Use the interval and the cut-off your own laboratory reports, and treat a result within about 10% of that cut-off as indeterminate.
Can a high adenosine deaminase confirm tuberculous pleural effusion?
Only weakly, and only where tuberculosis is common. The 2019 meta-analysis found the 40 ± 4 IU/L threshold more useful for excluding than for confirming, and its authors said so explicitly. At a pre-test probability of 5% a positive result reaches only about 33%. Raising the cut-off above 65 IU/L confirms better, with specificity 0.94, but rests on four studies and sacrifices sensitivity.
Does a low adenosine deaminase exclude tuberculosis?
Nearly, in a low-incidence setting. The negative likelihood ratio at the 40 ± 4 threshold is 0.078, which takes a pre-test probability of 5% down to about 0.4%. Against a high pre-test probability it is much less definitive — 50% falls only to about 7%. The enzyme has also been reported as falsely low in elderly and critically ill patients with tuberculous pleuritis.
What else raises pleural fluid adenosine deaminase?
Empyema and parapneumonic effusion, lymphoma, rheumatoid pleurisy and other malignancies. In a series of over two thousand patients, more than 40% of parapneumonic and about half of lymphomatous effusions exceeded the tuberculosis cut-off, and every value above 250 U/L came from empyema or lymphoma rather than tuberculosis.
Should adenosine deaminase replace pleural biopsy?
No. It shortens the workup where it is negative and the pre-test probability was low. Where the result is positive, or where tuberculosis remains likely despite a negative result, mycobacterial culture and pleural biopsy are still needed — not least because neither adenosine deaminase nor a nucleic acid amplification test on fluid gives you drug susceptibility.
Related calculators
References
- Aggarwal AN, Agarwal R, Sehgal IS, Dhooria S. Adenosine deaminase for diagnosis of tuberculous pleural effusion: a systematic review and meta-analysis. PLoS One. 2019;14(3):e0213728.
- Porcel JM, Esquerda A, Bielsa S. Diagnostic performance of adenosine deaminase activity in pleural fluid: a single-center experience with over 2100 consecutive patients. Eur J Intern Med. 2010;21(5):419–423.
- Feres MC, De Martino MC, Maldijian S, et al. Laboratorial validation of an automated assay for the determination of adenosine deaminase activity in pleural fluid and cerebrospinal fluid. J Bras Pneumol. 2008;34(12):1033–1039.
- Light RW. Pleural effusions. Med Clin North Am. 2011;95(6):1055–1070.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.