The tuberculin skin test in confirmed pulmonary tuberculosis in the state of Qatar: where we stand?

OBJECTIVE: This is the first paper to evaluate the potency of the tuberculin skin test (TST) results in confirmed pulmonary tuberculosis in the developing country.
METHOD: Data was collected retrospectively from the tuberculosis (TB) treatment unit during the period from 1998 to 2004. All charts diagnosed as active tuberculosis based on positive sputum smear or culture with documented TST were reviewed. The standard TST was done by injecting o.1 ml of 5 international units subcutaneous RT 23 purified protein derivative (PPD) on volar surface of the right arm.
RESULTS: There were 306 patients with confirmed active pulmonary tuberculosis, of which 58% were smear positive and 42% were smear negative but culture positive. Expatriates accounted for 81% (247) and male for 74% (225) of the patients. The mean TST was 18.5 mm with standard deviation of 7.54 mm. TST was less than 5 mm in 8.2% (25), 5-10 mm in 1.6% (5), 10-15 mm in 11.1% (34) and more than 15 mm in 79.1% (242). False negative (reaction less than 10 mm) was significantly higher in Qatar nationals (17% vs 8% in expatriates) and those with associated diseases (16.3% vs 7% without associated diseases).
CONCLUSION: Although 9.8% of confirmed pulmonary tuberculosis had false negative TST, it remains a potent aide for epidemiological and diagnostic purposes and periodic assessment of this is highly recommended. In our community with BCG vaccination a reaction more than 10 mm should be considered positive.

Introduction

The tuberculin discovered by Robert Koch after he had discovered the tubercle bacillus. He believed that this was a cure for tuberculosis. The old tuberculin (OT) is a concentrated filtrate from broth culture of Mycobacterium tuberculosis that had been killed by heat. Thereafter diluted with saline and/or glycerine and preserved with phenol. In 1934, Florence Seibert prepared precipitates of OT, first with trichloracetic acid and later with ammonia sulfate. She termed this “purified protein derivative” (PPD).

A large batch of PPD was produced by Seibert and Glenn in 1941, which served as the standard reference material in the United States and was designated PPD-S. A 5 tuberculin unit (TU) dose was the standard for use in the United States.

Early studies in tuberculosis sanitoria demonstrate high sensitivity to graduated skin testing with tuberculin and a negative result rule out tuberculosis. However tuberculin reaction of less than 10 mm have been reported in 4 to 5.6% of patients with confirmed pulmonary tuberculosis. On other hand as high as 13–36% have been reported. This has lead to a great deal of controversy in recent years about the role of tuberculin testing for diagnosis and prevalence surveys.There was a great debate about the potency of tuberculin test over time, tuberculin products and method of delivery eg. Mantoux vs multiple punctures.

This is the first paper that examines the reaction and/or potency of PPD in confirmed pulmonary tuberculosis and provides a reference for any future survey in the State of Qatar where BCG is given routinely at birth.

Methods

Data was collected retrospectively from the Tuberculosis treatment unit during five years period from January 1998. This is unique, as there is only one registry for the whole country, which was well kept with patient's files. We reviewed the files for patients with confirmed tuberculosis for age, nationality, sex, associated comorbid conditions, smear status, and drug resistance.

Each patient file was reviewed, and TB was confirmed if smear or culture of sputum was positive for mycobacterium tuberculosis. The TST was done by the nurses at the TB treatment unit and read by the nurses and confirmed by the doctor. We used 0.1 cc of 5 ITU of puriefed protein derivative (PPD) CT68 injected intradermally. This tuberculin PPD (Mantoux) tubersol was manufactured by the Cannaught Laboratories Limited in Toronto- Ontario of Canada, and distrubited by the Pasteur Merieux Cannaught (Rhode-Poulenc group) in Pennsylvania, USA. The size of the reaction (induration) was measured after 48 and 72 h. Less than 10 mm was considered negative in this study.

Analysis and significance testing were performed by the x2 test testing with 95% confidence intervals (Cis) and Contingency coefficient (C) as measure of correlation, using EpiInfor version 6 and SPSS version 7.5.

Results

There were 306 patients with confirmed pulmonary tuberculosis, of which 177 (58%) were smear positive and the rest were smear negative but culture positive for Mycobacterium tuberculosis (MTB) (Table 1).

Table 1.

The Demographic data and Mantoux reaction.

Demographic data	Negative (%) < 10 mm	Positive (%) 10 mm or more	Total (% of 306)	P value	Odd ratio	Confidence interval Cis 95%
Age
Less than 25 year	5 (10%)	43 (90%)	48 (16%)	0.87
25 to 44 year	12 (8%)	148 (92%)	160 (52%)	0.16
45 to 65 year	11 (13%)	74 (87%)	85 (28%)	0.25
more than 65 year	2 (15%)	11 (85%)	13 (4%)	0.37	Fisher exact
Gender
Male	25 (11%)	200 (88%)	225 (73%)	0.2
Female	5 (6%)	76 (94%)	81 (26%)
Nationality
Qatar nationals	10 (17%)	49 (83%)	59 (19%)	0.04 SIG	2.32	0.94–5.61
Expatriates	20 (8%)	227 (92%)	247 (81%)		0.43	0.18–1.1
Associated Disease
Associated with disease	15 (16%)	77 (84%)	92 (30%)	0.012 SIG	2.58	1.13–5.9
Diabetes Mellitus	12 (15%)	66 (85%)	78 (25.5%)	0.055	2.12	0.91–493
Maligancy	1	0	1 (0.3%)	0.1
*Other (HIV, OAD)	2 (15%)	11 (85%)	13 (4.2%)	0.37
No associated disease	15 (7%)	199 (93%)	214 (70%)	0.012 SIG	0.39	0.17–0.88
Smear status
Smear Positive	12 (6.8%)	166 (93.2%)	177 (58%)	0.035	0.45	0.19–1.02
Smear Negative but Culture Positive	18 (14%)	111 (86%)	129 (42%)		2.24	0.98–5.18
Chest radiography
Cavitary	9 (7%)	121 (93%)	130 (42%)	0.15
Non cavitary infiltration	14 (13%)	91 (87%)	105 (34%)	0.13
**Other finding	2 (14%)	12 (86%)	14 (5%)	0.4	Fisher exact
Normal	1 (6%)	17 (94%)	18 (6%)	0.45
Not available	4 (10%)	35 (90%)	39 (13%)
Drug resistance
Sensitive	17 (7%)	246 (93%)	263 (86%)	0.39
Resistance	3 (9%)	30 (91%)	33 (14%)

* Other Diseases including one HIV = Human immunodeficiencey Virus, OAD = chronic Obstructive diseases such as asthma and COPD. Other such as chronic liver disease and bronchiectasis.

** Other Finding such as old pulmonary changes pleural effusion etc.

The mean TST (PPD) was 18.5 mm with standard deviation of 7.54 mm. TST was less than 5 mm in 8.2% (25), 5–10 mm in 1.6% (5), 10–15 mm in 11.1% (34) and 15 mm or more in 79.1% (242).

The mean age of the patients was 38 years with standard deviation of 13.7 (range 6 to 86 years), of which 52% of them were in the 25 to 44 year age group. There was no significant difference between the age of the patients and the TST reaction.

Males accounted for 74% of the patients with a mean TST of 17.8 mm (standard deviation (SD) 7.5 mm), and the rest were females with a mean TST of 20.8 mm (SD 7.2 mm). Although the female had higher mean, this was not statistically significant (p = 0.22).

Qatar nationals accounted for 19% (59) of the patients with mean TST of 16 mm (SD 8.3 mm) and expatriates accounted for the rest with a mean TST of 19 mm (SD 7.mm). The Qatar nationals had higher false negative TST (17%) than the expatriates (8%) that was statistically significant (p = 0.04).

Those with comorbid condition (associated disease) had higher false negative TST (16%) than those patients without comorbid disease (7%), that was statistically significant (p = 0.012). Those comorbid diseases including diabetes mellitus, maligancy and other diseases such as Human Immunodefficiency Virus Infection (HIV), obstructive airways diseases.

Smear positive patients had less negative TST (6.8%) than those with negative smear (14%). This difference was statistically significant (p = 0.035).

Although the negative TST rate varied for the chest radiography findings in those patients, which included; normal (6%), cavitary (7%), noncavitary infiltrates (13%), and other findings (14%). This difference was not statistically significant (p = 0.13).

The rate of negative TST was 7%, 9% for sensitive and resistant isolates respectively. This was not statistically significant.

Discussion

There are two factors contribute to the false-negative tuberculin skin tests in patients with confirmed tuberculosis. First of all is the host factors which include acute or overwhelming tuberculosis, HIV infection, Immunosuppressive diseases (Lymphoma etc.), viral infections (measles, mumps, varicella), live virus vaccination, renal failure and malnutrition. Second factor is that related to the testing procedure such as improper storage of PPD, improper dilution, delayed injection after filling the syringe, subcutaneous injection, lack of experience in interpretation and/or bias in interpretation.

In this paper, 9.8% of all confirmed pulmonary tuberculosis showed reactions of less than 10 mm to 5 international tuberculin unit (ITU) of PPD, which was intermediate between the lowest (4%) reported and the highest (34%) reported false negative TST. There was no significant association between the TST and the age groups, even though there was a small yet inconsistent increase in negative TST with increased age groups. This was not statistically significant as 52% (160) of the patients were in the 45 years age group and 4% (13) were in the 65 years of age or older age group. However there was a significant correlation of TST size and age of the patients (p = 0.006), giving the following formula (TST size = 21 − 0.1 × Age in years ± 7.7). This supports the association of false negative PPD with advanced age.

Qatar nationals had higher false negative (17%) than the expatriates (8%), that was significant. This because Qatar nationals were older (above 65 year accounted for 22% compare to 0.4%) and had more associated disease especially diabetes (38% compare to 28%), rather than a racial difference.

Those with associated disease or comorbid conditions had significantly higher rate of negative TST than those without comorbid conditions. The diabetes was the commonest disease in those patients and followed by chronic pulmonary obstructive disease (COPD) and asthma. This low reaction to tuberculin skin testing in diabetics deserves further evaluation.

Smear negative had a significantly higher rate of negative TST than those with smear positive pulmonary tuberculosis, probably releated to fact that they had a significant higher comorbid conditions (39% vs 25%), also a significantly higher noncavitary infiltrates (60% vs 22%), normal (13% vs 1.4%) and other (9.7% vs 0) chest radioloical findings than the smear positive which is suggestive of possible disseminated tuberculosis. This was similar to recent report from Pakistan.

The rate of negative TST varied with the chest radiography which included cavitary (7%), noncavitary infiltrates (13%), normal (6%) and other (14%) findings. This was not statistically significant.

There was no significant association between the PPD false negative and drug resistance or gender.

In conclusion, tuberculin skin test is a diagnostic aide and clinical judgement remains the essential determinant as 9.8% of confirmed pulmonary tuberculosis had false negative PPD. However, Mantoux test remains a potent aide in our community for epidemiological and diagnostic purposes.