- Email: [email protected]
Diagnostic utility of the estimation of mycobacterial Antigen A60 specific immunoglobulins IgM, IgA and IgG in the sera of cases of adult human tuberculosis
Tubercleand LungDisease(1995)76, 418-424 © 1995PearsonProfessionalLtd
Tubercleand LungDisease
Diagnostic utility of the estimation of mycobacterial Antigen A60 specific immunoglobulins IgM, IgA and IgG in the sera of cases of adult human tuberculosis S. Gupta, S. Kumari, J. N. Banwalikar, S. K. Gupta
Department of Microbiology, National Institute of Communicable Diseases, Delhi, India, R.B.T.B. Hospital, Kingsway Camp, Delhi, India S U M M A R Y . Setting: An ELISA assay based on mycobacterial Antigen A60 for the estimation of Mycobacterium-specific immunoglobulins in the serum has been used successfully for the rapid diagnosis of tuberculosis in studies done in the Western countries. There are hardly any similar large scale studies in India. Objective: To evaluate the utility of this ELISA test for rapid diagnosis of different clinical forms of tuberculosis in an adult Indian population. Design: ELISA test based on mycobacterial antigen A60 (Anda Biologicals, France) was used to estimate specific IgM, IgA and IgG antibodies in the sera obtained from 337 cases of tuberculosis and 131 controls in the population of Delhi (India). Results: Of the 131 controls, only 9.9% were positive for IgM, 7.6% for IgG, 6.1% for IgA and 9.9% when an IgA and IgG combination was considered. Of 122 cases of active pulmonary tuberculosis, 41% were positive for IgM, 86.8% for IgA, 88.5% for IgG and a very high positivity (98.3%) was seen when IgA and IgG estimations were combined. A relatively low seropositivity was observed in 25 cases of pleural tuberculosis. The corresponding figures in 130 cases of extrapulmonary tuberculosis were 22.3%, 68.4%, 73.8% and 86.15%. When 60 cases of pulmonary tuberculosis who had been successfully treated with antituberculosis drugs were analyzed the rates of seropositivity fell to 11.6%, 46.6%, 58.3% and 66.6% respectively. Conclusion: Our findings point to a very good sensitivity (91.6%) and specificity (90.0%) of the test when combined IgA and IgG antibody titres are considered, to detect cases of adult tuberculosis. The role of IgM estimation can be restricted to the detection of cases of reactivation of tuberculosis.
RE, S UMI~. Cadre: Au cours des 6tudes effectu6es dans les pays de l'Ouest, un essai ELISA bas6 sur l'antig~ne mycobact6rien A60 pour 6valuer les immunoglobulines sp6cifiques du mycobacterium dans le s6rum a 6t6 utilis6 avec succ~s dans le diagnostic rapide de la tuberculose. Peu d'6tudes ont 6t6 effect6es en Inde h une aussi grande 6chelle. Objet: Evaluer l'utilit6 de l'essai ELISA pour le diagnostic rapide des diff6rentes formes cliniques de la tuberculose dans la population indienne adulte. Schema: Un essai ELISA bas~ sur l'antig~ne mycobact~rien A60 (ANDA Biologicals, France), a ~t~ utilis6 afin d'6valuer les anticorps sp6cifiques IgM, IgA et IgG dans les s6rums provenant de 337 cas de tuberculose et de 131 t6moins parmi la population de Delhi, Inde. R(sultats: Sur les 131 t6moins, seulement 9,9% 6taient positifs pour IgM, 7,6% pour IgG, 6,1% pour IgA et 9,9% lorsqu'une combinaison d'IgA et d'IgG a 6t6 test6e. Sur 122 cas de tuberculose pulmonaire active, 41% 6taient positifs pour IgM, 86,8% pour IgA, 88,5% pour IgG et un taux 61ev6 de positivit6 (98,3%) a 6t6 observ6 lorsque les 6valuations pour IgA et IgG 6taient combin6es. Une positivit6 relativement faible a 6t6 remarqu6e dans 25 cas de tuberculose pleurale. Les chiffres correspondants pour 130 cas de tuberculose extrapulmonaire 6taient de 22,3%, 68,4%, 73,8% et 86,15%. Une analyse de 60 case de tuberculose pulmonaire ayant regu avec succbs un traitement antituberculeux a d6montr6 des taux de positivit6 de 11,6%, 46,6%, 58,3% et 66,6% respectivement. Correspondence to: Dr. Sunil Gupta, Assistant Director, Department of Microbiology, National Institute of Communicable Diseases, 22, Shamnath Marg, Delhi-110054. INDIA. Paper received 14 October 1994. Final version accepted 16 February 1995. 418
Role of A60 serologyin diagnosis of adult tuberculosis 419 Conclusion: Ces observations d~montrent une bonne sensibilit~ (91,6%) et sp~cificit~ (90,0%) pour ce test dans la d6tection de cas de tuberculose adulte, utilisant une combinaison de titres des anticorps IgA et IgG. Le r61e de la d6termination d'IgM peut 6te r~serv~ ~ la d6tection des cas de reprise de la tuberculose. R E S U M E N. Marco de referencia: En estudios efectuados en paises occidentales, para el diagn6stico r~ipido de la tuberculosis, se ha utilizado con 6xito el test ELISA basado en el antigeno micobacteriano A60, para evaluar las inmunoglobulinas especificas del mycobacterium en el suero. En India se han efectuado escasos estudios de una amplitud similar. Objetivo: Evaluar la utilidad del test ELISA para el diagn6stico r~ipido de las differentes formas clinicas de tuberculosis en la poblaci6n adulta de India. M(todo: Se utiliz6 el test ELISA basado en el antigeno micobacteriano A60 (Anda Biologicals, Francia), para evaluar los anticuerpos especificos IgM, IgA e IgG, en el suero proveniente de 337 casos de tuberculosis y de 131 controles en la poblaci6n de Delhi, India. Resultados: En los 131 controles, s61o el 9,9% era positivo para IgM, 7,6% para IgG, 6,1% para IgA y 9,9% cuando se evalu6 una combinaci6n de IgA y de IgG. En los 122 casos de tuberculosis pulmonar activa, el 41% era positivo para IgM, 86,8% para IgA, 88,5% para IgG y se observ6 una tasa elevada de positividad (98,3%) cuando se evalu6 IgA e IgG en forma combinada. Se constat6 una positividad relativamente baja en 25 casos de tuberculosis pleural. Las cifras respectivas para 130 casos de tuberculosis extrapulmonar eran 22,3%, 68,4% y 86,15%. El amilisis de 60 casos de tuberculosis pulmonar tratados con 6xito con medicamentos antituberculosos mostr6 una disminuci6n de las tasas de positividad: 11,6%, 46,6%, 58,3% y 66,6%, respectivamente. Conclusi6n: Estos hallazgos demuestran una buena sensibilidad (91,6%) y una buena especificidad (90,0%) del test cuando se combinan los titulos de anticuerpos IgA e IgG para la detecci6n de casos de tuberculosis en adultos. La evaluaci6n de IgM puede ser reservada a la detecci6n de casos de reactivaci6n de tuberculosis.
INTRODUCTION Acid-fast staining of sputum or another appropriate clinical specimen, followed by confirmation by culture, remains the cornerstone of laboratory diagnosis of tuberculosis (TB), and yet these techniques are either slow or lack sensitivity? The disadvantages of tuberculin testing as a diagnostic tool are well known: in addition to the problem of lack of specificity, the patient needs to be seen on two occasions. 2 Attempts have been made to improve the sensitivity and the speed of detection of tubercle bacilli or their components by techniques such as radiometric determination of bacterial growth, gas chromatography, DNA hybridization and polymerase chain reaction. However, all of these have met with problems of either sensitivity or specificity or have proved too technically demanding. 3 Recently, ELISA techniques employing various mycobacterial antigens, e.g. glycolipids from Mycobacterium bovis BCG, antigens 5 and 6 from M. tuberculosis, 64kDa protein of M. boris BCG and a 32kDa protein antigen called P32, have all been attempted for serological diagnosis of tuberculosis, with variable success. 4-10 A renewed attempt to develop a serodiagnostic test was initiated when Antigen 60 became available? I Antigen 60 is a major component of mycobacterial cytoplasm. Prepared from the cytoplasm of M. boris BCG and purified by exclusion gel chromatography, it is a macromolecular (106-107 daltons) lipopolysaccharide protein complex. ~2 Various scientists have made use of this antigen in the ELISA test for the serodiagnosis of tuberculosis by estimating IgM and IgG antibodies in
the serum of patients, with encouraging results. 12-22 There are some unpublished reports demonstrating the role of anti-A60 IgA serology in the diagnosis of tuberculosis. Considering the seriousness of the tuberculosis situation in our country, the present study was designed to evaluate the serodiagnostic potential of Antigen 60 in different clinical forms of tuberculosis in Delhi, India involving the adult (/> 18 years) age group, by measuring the A60-specific IgM, IgA and IgG antibody levels in the serum.
MATERIALS AND METHODS Study population: serum samples were obtained from 337 cases of adult tuberculosis, mainly from the RBTB Hospital (a 2000-bed tuberculosis hospital in Delhi) and other hospitals in Delhi. The diagnosis of TB was based on clinical and radiological criteria, tuberculin (Mantoux [Mx]) reactivity, histopathology, presence of acid-fast bacilli (AFB) and clinical response to antituberculosis treatment (ATT). The TB patients consisted of 147 cases of post-primary active pleuro-pulmonary TB and 130 cases of extra-pulmonary TB. 60 cases of inactive (treated) pulmonary TB were also included in the study.
Post primary active pleuro-pulmonary TB (147 cases) These included: 1. 122 patients with purely pulmonary TB (symptomatic at the time of the study), of whom 50 had received
420
Tubercle and Lung Disease
ATT for more than one month after several weeks of evolution of the disease, while 35 patients had interrupted the treatment after a few weeks (6-8 weeks). 37 patients had not received ATT at all. 67 of the cases were AFB-positive at some stage of the disease. 2. 25 patients suffering from pleural TB (symptomatic at the time of the study), of whom 8 had received ATT for a duration of less than 2 months. 15 cases had been AFB positive at some stage of the disease.
Post primary extra-pulmonary TB (130 cases) This group comprised: 1. 26 cases of miiiary TB; all were radiologically proved, symptomatic and untreated; 8 were AFB positive. 2. 22 cases of cerebrospinal TB, confirmed on the basis of CSF cytology biochemistry, AFB positivity or response to ATT. 3. 18 cases of osteoarticular TB, of whom 5 were AFB positive. 4. 26 cases of gastrointestinal TB, of whom 4 were AFB positive. 5. 18 cases of genitourinary TB, of whom 5 were AFB positive. 6. 20 cases of lymphoreticular TB, of whom 6 were AFB positive.
Post primary inactive pulmonary TB (60 cases) These subjects comprised cases of pulmonary TB who had been given ATT for 6-12 months at an early stage of the disease. They were all clinically healed and were negative for AFB at the time of the study.
Controls 131 age-, sex- and socioeconomically-matched controls were also included in the study. These consisted of 60 healthy adults, 37 of whom had been BCG-vaccinated in the past, 53 diseased controls (subjects suffering from various ailments but showing no evidence of TB), 10 contacts of cases of AFB-positive pulmonary TB, and 8 laboratory technicians handling mycobacterial cultures.
Serodiagnosis of TB by ELISA technique Anti A60 mycobacterial antibodies were estimated in the Table 1.
sera of the subjects under study employing indirect ELISA technique at the National Institute of Communicable Diseases, Delhi, as per the recommendations of the manufacturer (Anda Biologicals, Strasbourg, France). Each time, the positive as well as the negative reference sera provided with the diagnostic kit were included in the test along with the test sera. In the case of IgM antibodies a threshold value was determined from the optical density (OD) values obtained with the positive reference serum. For IgA and IgG determinations, the curves were constructed by plotting the OD values of different reference sera. Thereafter, concentration of the antibodies in test serum in terms of units/ml was determined by extrapolating the OD value of test serum against the reference curve. 14
Tuberculin (Mx) test This test was performed on all the subjects under study using ITU PPD RT-23 with Tween-80 in 0.1 ml dose. A diameter of induration of > 10 mm after 48 hours was considered a positive test? RESULTS
Determination of baseline setting for serological analysis Since the sensitivity (ST) and specificity (SP) of any serological test is dependent on the appropriate cut-off values chosen in the test, in this study we initially attempted to determine the most suitable cut-off values for the 3 types of A60 specific immunoglobulins, i.e. IgM, IgA and IgG, to differentiate the tuberculous cases from the controls. 1. Cut-off value of IgM: in the case of IgM antibodies an optical density index (ODI) was first determined by dividing the OD obtained from the test sample by the OD obtained with the positive reference serum. 23 Thereafter, 3 cut-off values were chosen arbitrarily, i.e. 1.00, 1.50 and 2.00, to segregate the cases of active TB from the controls, the respective values of sensitivity and specificity for the 3 cut-offs being 62.2% and 79.1%, 32.1% and 92.3%, and 13.1% and 95.8% (Table 1). We preferred to choose a cut-off value of 1.50 as it gave a very high specificity (92.3%), although the sensitivity was compromised (32.1%). However this could be countered later on by including IgG and IgA determinations.
Statistical evaluation of individual antibodies. Statistical evaluation of A60 ELISA at different cut-off points for adult tuberculosis
Cut-off value O.D.I. >
IgM ST %
SP %
Statistical evaluation of individual antibodies. IgA Cut-off value ST % SP % Cut-off value > Units/nil > Units/ml
IgG ST %
SP %
1.00
62.2
79.l
100
95.4
69.4
150
94.2
69.4
1.50
32.1
92.3
150
79.8
95.4
2.00
13.1
95.8
200
67.2
96.8
200 250 300
91.8 83.0 72.0
80.5 92.3 94.5
O.D.I. = Optical Density Index, ST = Sensitivity, SP = Specificity.
Role of A60 serology in diagnosis of adult tuberculosis 421 2. Cut off value for IgG: as can be seen in Table 1, 4 different cut-off values were considered i.e. 150, 200, 250 and 300 units/ml. However a titre of 250 units/ml was chosen as the cut-off point as this resulted in a very good specificity (92.3%) and a reasonably good sensitivity (83.0%). 3. Cut off value for IgA: 3 cut off values i.e. 100, 150, 200 units/ml were considered. The cut-off value of 150 units/ml gave a high specificity (95.4%) and a reasonable sensitivity (79.8%), and was accordingly chosen (Table 1). In order to further improve the specificity and sensitivity of the test, different combinations of the 3 immunoglobulins at the above cut-off points were tried. It was found that when both IgA and IgG values were used in conjunction (presence of at least one of these antibodies) a very good specificity (90.0%) mad sensitivity (91.6%) was obtained (Table 2). In serological analysis of controls with regard to A60 specific antibodies of IgM, IgA and IgG classes (Table 3), it was seen that among the healthy controls, very few subjects yielded a positive serology: only 5% were positive for IgM and IgA and slightly more (8.3%) were positive for the IgG antibodies. There was not much serological difference between BCG vaccinated and non-vaccinated controls. However, when the 8 techniTable 2. Statistical evaluation of combinations of Antibodies IgM, IgA and IgG at the cut-off points of _>1.50, 150 and 250 respectively
Type of Antibody Combination
ST %
SP %
IgM and IgA and IgG IgM or IgA or IgG IgA and IgG IgA or IgG IgM and IgG IgM or IgG IgM and IgA IgM or IgA
22.1 100.0 72.5 91.6 27.0 89.3 24.5 92.6
100.0 75.0 98.4 90.0 95.8 79.1 100.0 80.5
ST = Sensitivity, SP = Specificity.
cians were analysed, 25% were found to be positive for IgM and IgA and 50% for IgG antibodies. On the contrary, 10 contacts of cases of TB depicted a very low positivity. None of these subjects were found positive for either IgM or IgG, whereas only 1 (10%) had adequate titres of IgA antibodies. Among the 53 diseased controls, 11.3% were positive for IgM, and 1.9% for IgG antibodies, whereas none had adequate IgA antibody titres. A detailed analysis of different types of diseased controls is given in Table 3. Taking the overall picture of the 131 controls, 9.9% were positive for IgM, 6.1% for IgA, 7.6% for IgG and only 9.9% were found to be positive when IgA and IgG antibodies were considered in conjunction. On the other hand, a high degree of tuberculin (Mx) reactivity (38.9%) was observed (Table 3). The mean values of the 3 immunoglobulins (Igs) were 1.110 (ODI), 124.04 units/ml and 133.83 units/ml for IgM, IgA and IgG antibodies (Table 4).
Serological analysis of active p l e u r o - p u l m o n a r y TB (147 cases)
1. Pulmonary TB (122 cases) Of these, 41% were positive for IgM, 86.8% for IgA, 88.4% for IgG, whereas a very high positivity (98.3%) was seen when IgA and IgG estimations were combined. On the other hand, Mx positivity was only 53% (Table 5). A good correlation was seen between IgA and IgG positivity (75% of cases.) Among the 41% of cases who were positive for IgM antibodies the majority (85%) were cases of reactivation. The corresponding mean titres for the 3 antibodies were 1.376 (ODI), 898 units/ml and 1320 units/ml for IgM, IgA and IgG respectively (Table 4). 2, Pleural TB (25 cases) 40% positivity was seen for IgM, 56% for IgA, 72% for IgG and 88% for combined (IgA, IgG) es-
Table 3. Serologicalanalysis of non-tuberculous individuals (controls)
Subjects No of cases A. Healthy BCG vaccinated non BCG-vaccinated
Total B. Non-tuberculous patients. Asthma Chronic bronchitis Viral pyrexia Urinary tract infection Pneumonia (viral/bacterial) Rheumatoid arthritis Hepatocellular disease Total C. Contacts of cases of TB D. Subjects handling Mycobacteria Grand Total Mx = Tuberculin test (Mantoux).
A60 ELISA positivity in serum IgM IgA IgG No. (%) No. (%) No. (%)
Mx positivity IgA or IgG No. (%)
No. (%)
37 23
2 1
2 1
3 2
4 4
12 8
60
3(5.0)
3(5.0)
5(83)
7(11.6)
20(33.3)
9 8 12 6 10 4 4 53 10 8 131
1 1 0 0 1 1 2 6 (11.3) 0 (0) 2 (25.0) 13 (9.9)
0 0 0 0 1 0 0 1 (1.8) 1 (10.0) 4 (50.0) 13 (9.9)
3 5 4 1 3 4 1 21 (39.6) 5 (50.0) 5 (63.3) 51 (38.9)
0 0 0 0 0 0 0 0 (0) 1 (10.0) 4 (50.0) 8 (6.1)
0 0 0 0 1 0 0 1 (1.9) 0 (0) 4 (50.0) 10 (7.6)
422 Tubercle and Lung Disease Table 4.
A60 specific antibody titres in controls and cases of TB
Type of subject
NO
Controls
131
Cses of active Pulmonary TB Cases of active Pleural TB Cases of active Pleuro-pulmonary TB Cases of active Extra-pulmonary TB Cases of inactive Pulmonary TB
122 25 147 130 60
IgM (O.D.I)
Mean antibody titres (range) IgA (Units/ml)
IgG (Units/ml)
1.110 (0.365-2.670) 1.376 (0.672-3.120) 980 (0.478-2.056) 1.178 (0.478-3.120) 1.040 (0.536-2.780) 1.072 (0.380-1.960)
124.04 (< 10-240) 898 (100-3500) 152 (100-1600) 525 (100-3500) 222.30 (80-2200) 438 (50-800)
133.83 (20-350) 1320 (150-3600) 375 (125-2200) 847.50 (125-3600) 482.77 (80-2500) 865 (50-1800)
O.D.I = optical density index.
Table 5.
Serological analysis of cases of Tuberculosis (T.B.)
Subjects
A60 ELISA positivity in serum IgA IgG No. (%) No. (%)
Mx* positivity
No. of cases
IgM No. (%)
122 25 147
50 (41.0) 10 (40.0) 60 (40.8)
106 (86.8) 14 (56.0) 120 (81.6)
108 (88.5) 18 (72.0) 126 (85.7)
120 (98.3) 22 (88.0) 142 (96.6)
65 (53.0) 10 (40.0) 75 (51.0)
26 22 18 26 18 20
13 (50.0) 6 (27.2) 4 (22.2) 2 (7.6) 2 (11.1) 2 (10.0)
23 (88.4) 12 (54.5) 9 (50.0) 16 (61.5) 14 (77.7) 15 (75.0)
20 (76.0) 16 (72.7) 12 (66.6) 19 (73.0) 14 (77.7) 15 (75.0)
23 (88.4) 18 (81.8) 15 (83.3) 23 (88.4) 16 (88.8) 17 (85.0)
12 (46.1) 13 (59.0) 10 (55.5) 12 (46.2) 9 (55.0) 11 (55.0)
Total (B)
130
29 (22.3)
89 (68.4)
96 (73.8)
112 (86.1)
67 (51.5)
Total (A+B)
277
89 (32.1)
209 (75.4)
222 (80.1)
254 (91.6)
142 (51.2)
Inactive pulmonary
60
7 (11.6)
28 (46.6)
35 (58.3)
40 (66.6)
40 (66.6)
GRAND TOTAL
337
A. Active pleuro pulmonary Pulmonary Pleural Total (A) B. Active extrapulmonary Miliary Cerebrospinal Osteo-articular Gastro-intestinal Genito-urinary Lymphoreticular
IgA or IgG No. (%)
No. (%)
Mx* = Tuberculin test (Mantoux).
timations. A low Mx reactivity (40%) was observed. (Table 5). In 50% of cases, both, IgA and IgG antibodies were found. The mean titres of 3 antibodies, were 0.980 (ODI), 182 units/ml and 375 units/ml for IgM, IgA and IgG respectively (Table 4). On analysing all the 147 cases of active pleuropulmonary TB, it was observed that a very high percentage (96.6%) were positive for either IgG or IgA antibodies, whereas Mx reactivity was displayed in 51% cases only (Table 5). The mean titres of the 3 types of antibodies were 1.178 (ODI), 525 units/ml and 847.50 units/ml for IgM, IgA and IgG respectively (Table 4).
Serology of extra-pulmonary TB (130 cases) The disease-wise break up of cases is given in Table 5. Overall, 73.8% of cases displayed a positive IgG serology, 68.4% were positive for IgA and very few (22.3%) were positive for IgM antibodies. However, a
relatively high IgM positivity (50%) was seen in cases of miliary tuberculosis. The overall positivity came up to 86.1% on taking into account the combined (IgA, IgG) estimations, whereas tuberculin posifivity was displayed in 51.5% of cases only (Table 5). In 60% of cases, the presence of both IgA as well as IgG antibodies was seen. The mean values for the 3 types of antibodies were 1.040 (ODI), 222.3 units/ml and 482.7 units/ml for IgM, IgA and IgG respectively (Table 4).
Serology of inactive pulmonary TB (60 cases) An extremely low IgM positivity (11.6%) was seen followed by 46.6% for IgA and 58.3% for IgG antibodies, a reasonably high Mx reactivity (66.6%) was observed (Table 5). The mean titres for the 3 types of antibodies were 1.072 (ODI), 438 units/ml and 865 units/ml for IgM, IgA and IgG respectively. (Table 4)
Role of A60 serology in diagnosis of adult tuberculosis
Overall picture Considering all the cases of active tuberculosis and the controls, the test sensitivity of 91.6% and a specificity of 90.0% were obtained when combined (IgA, IgG) estimations were ,o taken into account. On the other hand, tuberculin reactivity was seen in 51.2% cases of active TB compared to 38.9% in the control population.
DISCUSSION The present study was performed to evaluate the role of A60 antigen in the serological diagnosis of tuberculosis (TB) in adults from the city of Delhi, India using the ELISA technique for the detection of A60 specific IgM, IgA and IgG antibodies. In this study, for IgM antibodies, a cut-off value of 1.500 of optical density index (ODI) appeared to give the best results. In the earlier studies undertaken in Western countries, a relatively lower cut-off value of 1.00 for ODI was chosen, which could have been due to the lower endemicity of TB in those countries] 5,19 As regards IgG antibodies, a cut-off value of 250 units/ml was chosen in this study to differentiate active TB cases from controls. In earlier studies performed in Italy and France, a cut-off value of 200 serounits/ml was chosen, 18'19 while in another study carried out in Belgium, the baseline was set at 100 units. H In a limited number of studies performed in India, mainly in the western zone, an IgG cut-off of 200 units was chosen. 16,2°,zz For IgA antibodies a cut-off value of 150 units/ml was selected to differentiate cases of active TB from controls. In the control group very few subjects yielded a positive serology against A60 antigen: these were mainly the laboratory staff who had been routinely handling the mycobacterial cultures. Only 9.9% were positive for IgM, 6.1% for IgA and 7.6% for IgG antibodies. Earlier reports on the seropositivity against A60 antigen in the control population ranged between 0 to 50% for IgG and 2.5 to 40% for IgM antibodies. The mean antibody levels in the control population in this study were appreciably lower compared to those in cases of active and inactive TB. Similar results were shown in another Indian study as regards IgG antibodies, z° Surprisingly, none of the human contacts of cases of TB showed a positive A60 serology. Moreover, no appreciable serological difference was observed between the BCG vaccinated and non-vaccinated controls. These findings are more or less in agreement with earlier reports.13,14,i6,17,19
A very good serological response was seen in cases of active pulmonary TB, especially with regard to IgA and IgG antibodies depicting a positivity of 86% and 88.5% respectively, the IgM positivity being rather low (41% only). Moreover, combinations of IgA and IgG antibodies showed a positivity as high as 98.3%. There was a
423
good correlation between IgA and IgG seropositivity (75%), however in a few cases (23%) either of the antibodies could be detected thereby necessitating the value of combined IgA and IgG serology for a better sensitivity. Further, it was observed that IgA response was better in the relatively recent infections. The IgM positivity was seen mainly in cases of reactivation as observed by others. The mean levels of all the 3 antibodies were appreciably higher than in the controls. In studies carried out earlier, an IgG positivity ranging from 48-100% and IgM positivity ranging from 5-76% has been documented. 15,~6,18-21 In one of these studies, 15 combined IgG and IgM estimation gave a sensitivity of 68%. These wide variations could be due to different age groups studied, geographical areas as well as severity of disease in different studies. 15,19 So far, we have not come across any published studies regarding anti A60 IgA serology in TB although good serum IgA responses were shown against p32 and M. bovis BCG sonicated antigen in some studies. 4,z5 In the present study, relatively low seropositivity was observed in cases of pleural TB compared to those of pulmonary TB as shown in a previous study. 18 As far as serology in extra pulmonary TB is concerued, although a high IgA (68.4%) and IgG (73.8%) positivity was seen, this was not the case with IgM (22.3% only). The positivity came to 86.1% when combined IgA and IgG values were considered. Again, a good correlation was observed in IgA and IgG positivity (60%) whereas in the rest of 26% of cases either IgA or IgG was present, indicating the importance of combined IgG and IgA determinations for a better sensitivity. The mean antibody levels were lower than in active pulmonary TB cases. In a few studies done in Western countries, an IgG positivity of 100% and IgM positivity of 0-15% have been determined in cases of extrapulmonary TB, I9,al whereas in Indian studies an IgG positivity of only 65-80% has been shownJ 6,2z On analysis of cases of inactive/(treated) pulmonary tuberculosis, quite a large number were found to be positive for IgA (58.3%) and IgG (46.6%), though this was not the case with IgM antibodies (11.6% positivity only). Moreover, the mean antibody levels were also much higher than those of the controls, although the levels were slightly lower than those in active pulmonary tuberculosis. Few available reports in this aspect show a substantial lowering of anti A60 antibodies with clinical recovery after treatment. 13,~9 Overall, fairly satisfactory results were obtained by estimating A60 specific IgA and IgG antibodies for rapid diagnosis of adult tuberculosis (ST = 91.6%, SP = 90.0%). The role of IgM estimation was found to be restricted to detecting cases of reactivation or to some extent cases of miliary tuberculosis. On the other hand, the tuberculin test could hardly differentiate the cases of active tuberculosis from the controls and there was hardly any correlation between tuberculin reactivity and A60 serology, as shown by others also. m9 In a previous study done in the West, a test sensitivity of 98.6% and a
424
Tubercle and Lung Disease
specificity of 86.7% was demonstrated in post primary TB by estimating A60 specific IgG antibodies only. 19 Isolated reports have shown good results in ELISA tests using other mycobacterial antigens for the diagnosis of TB, e.g 'antigen 6' (ST = 94%, SP = 100%), 4 'antigen 5' (ST = 48.8%, SP = 91-100%) 8, and a dual ELISA system using A60 and antigen cytosolic fraction of M. tuberculosis (ST = 97%, SP = 100%). 17 However, the results are still to be substantiated by other researchers.
10.
CONCLUSION
11.
The present study has conclusively proven the value of the estimation of A60 specific IgA and IgG antibodies for the rapid diagnosis of adult tuberculosis, especially the extrapulmonary type, where one does not have many other diagnostic modalities to choose from. It was also found to be useful to detect paucibacillary (AFB negative) cases of pulmonary tuberculosis.
8.
9.
12.
13.
14.
15.
Acknowledgements The authors are extremely grateful to the Director of the National Institute of Communicable Diseases, Delhi for providing the necessary facilities and to Shri Aalok Gupta and Shri Amar Jit Singh for providing secretarial assistance for the study. The technical assistance provided by Shri Mahabir Prasad is also acknowledged.
16. 17. 18.
19.
References 1. Daniel T M. The rapid diagnosis of tuberculosis: a selective review. J Lab Clin Med 1990; 116: 277-282. 2. Crofton J. et al ed. Appendix E. Tuberculin testing. In: Clinical tuberculosis, Macmillan Publishers Ltd, London, 1992; pp 187-192. 3. Grange J M, Laszlo A. Serodiagnostic test for tuberculosis: a need for assessment of their operational predictive accuracy and acceptability. Bull WHO 1990; 68(5): 571-576. 4. Turneer M, Van Vooren J P, Bmyn J et al. Humoral immune response in human tuberculosis: immunoglobulins G, A and M directed against the purified P32 protein antigen of Mycobacterium boris, bacillus Calmette-Gu6rin. J. Clin. Microbiol 1988; 26: 1714-1719. 5. Reggiardo Z, Vazquez E, Schnaper L. ELISA test for antibodies against mycobacterial lipids. J Immunol Methods 1980; 34: 55-60. 6. Reggiardo Z, Vazquez E. Comparison of enzyme-linked imunnosorbent assay and hemaggulutination test using mycobacterial glycolipids. J Clin Microbiol 1981; 13: 1007-1009. 7. Stroebel A B, Daniel T M, Lau J H K, Leong J C V, Richardson
20.
21.
22.
23.
24.
25.
H. Serologic diagnosis of bone and joint tuberculosis by an enzyme linked immunosorbent assay. J Infect Dis 1982; 146: 280-283. Benjamin R G, Daniel T M. Serodiagnosis of tuberculosis using the enzyme-linked immunosorbent assay of antibody to Mycobacterium tuberculosis antigen 5 Am Rev Respir Dis 1982; 126(1): 1013-1016. Benjamin R G, Debanne S M, Ma Y, Daniel T M. Evaluation of mycobactefial antigens in an enzyme-linked immunosorbent assay (ELISA) for the serodiagnosis of tuberculosis. J Med Microbiol. 1985; 18: 309-318. Thole J E R, Keulen W J, De Bruyn J e t al. Characterization, sequence determination and immunogenicity of a 64 kilodalton protein of Myeobacterium boris BCG expressed in Escherichia coli K-12. Infect Immun 1987; 55: 1466-1475. Cocito C, Vanlinden F. Preparation and properties of antigen 60 from Mycobaeterium boris BCG. Clin Exp Immunol 1986; 66: 262-272. Cocito C G. Properties of the mycobacterial antigen complex A60 and its application to the diagnosis and prognosis of tuberculosis. Chest 1991; 100: 1687-1693. Baelden M C, Vanderelst B, Dieng M e t al. Serologic analysis of human tuberculosis by an ELISA with mycobacterial antigen 60. Scand J Inf Dis 1990; 22: 63-76. Maes R, Homasson J P, Kubin M e t al. Development of an enzyme immunoassay for the serodiagnosis of tuberculosis and mycobacteriosis. Med Microbiol Immunol 1989; 178: 323-335. Charpin D, Herbanlt H, Gevaudan M J et al. Value of ELISA using A60 antigen in the diagnosis of active pulmonary tuberculosis. Am Rev Respir Dis 1990; 142 (2): 380-384. Patel A B, Rawat M S, Grover S. ELISA for diagnosis of tuberculosis. Indian Paediatr 1990; 27: 584-589. Reheman S F, Wagner S, Manch H et al. Evaluation of dual antigen ELISA test for the serodiagnosis of tuberculosis. Bull. WHO 1988; 66(2): 203-209. Fadda G, Gfillo R, Ginesu F et al. Serodiagnosis and follow up of patients with pulmonary tuberculosis by ELISA. Eur J Epidemiol 1992; 8(1): 810-817. Gevaudan M J, Bollet C, Charpin D, Mallet M N. Serological response of tuberculosis patients to antigen 60 of BCG. Eur J Epidemiol 1992; 8(5): 666-676. Gokhale N T, Ranadive S N, Joshi A M e t al. Antibody and immune complex ELISA in the immunodiagnosis of tuberculosis. Med J Arm Force India 1993; 49(3): 159-163. Maes R. Clinical usefulness of serological measurement obtained by Antigen 60 in mycobactefial infections: development of a new concept. Kiln Wochenschr 1991; 69: 696-709. Munshi M M, Chiddarwar S, Patel A, Grover S. Serodiagnosis of extrapulmonary tuberculosis by ELISA. Indian J Pathol Microbiol 1993; 36(4): 356-360. Daniel T M, Dabanne S M. The serodiagnosis of tuberculosis and other mycobacterial disease by enzyme linked immunosorbent assay. Am Rev Respir Dis 1987; 135:1137-1151 Wirrmann C. Public health application of a serological test for tuberculosis: study of the incidence of inapparent infections among the employees of an Alsatian supermarket. Eur J Epidemiol 1990; 6(3): 304-308. Grange J M, Gibson J, Nassau E et al. ELISA, a study of antibodies to Mycobacterium tuberculosis in the IgG, IgA and IgM classes in tuberculosis, sarcoidosis and Crohn's disease. Tubercle Lung Dis 1980; 61: 145-152.
Tubercleand LungDisease
Diagnostic utility of the estimation of mycobacterial Antigen A60 specific immunoglobulins IgM, IgA and IgG in the sera of cases of adult human tuberculosis S. Gupta, S. Kumari, J. N. Banwalikar, S. K. Gupta
Department of Microbiology, National Institute of Communicable Diseases, Delhi, India, R.B.T.B. Hospital, Kingsway Camp, Delhi, India S U M M A R Y . Setting: An ELISA assay based on mycobacterial Antigen A60 for the estimation of Mycobacterium-specific immunoglobulins in the serum has been used successfully for the rapid diagnosis of tuberculosis in studies done in the Western countries. There are hardly any similar large scale studies in India. Objective: To evaluate the utility of this ELISA test for rapid diagnosis of different clinical forms of tuberculosis in an adult Indian population. Design: ELISA test based on mycobacterial antigen A60 (Anda Biologicals, France) was used to estimate specific IgM, IgA and IgG antibodies in the sera obtained from 337 cases of tuberculosis and 131 controls in the population of Delhi (India). Results: Of the 131 controls, only 9.9% were positive for IgM, 7.6% for IgG, 6.1% for IgA and 9.9% when an IgA and IgG combination was considered. Of 122 cases of active pulmonary tuberculosis, 41% were positive for IgM, 86.8% for IgA, 88.5% for IgG and a very high positivity (98.3%) was seen when IgA and IgG estimations were combined. A relatively low seropositivity was observed in 25 cases of pleural tuberculosis. The corresponding figures in 130 cases of extrapulmonary tuberculosis were 22.3%, 68.4%, 73.8% and 86.15%. When 60 cases of pulmonary tuberculosis who had been successfully treated with antituberculosis drugs were analyzed the rates of seropositivity fell to 11.6%, 46.6%, 58.3% and 66.6% respectively. Conclusion: Our findings point to a very good sensitivity (91.6%) and specificity (90.0%) of the test when combined IgA and IgG antibody titres are considered, to detect cases of adult tuberculosis. The role of IgM estimation can be restricted to the detection of cases of reactivation of tuberculosis.
RE, S UMI~. Cadre: Au cours des 6tudes effectu6es dans les pays de l'Ouest, un essai ELISA bas6 sur l'antig~ne mycobact6rien A60 pour 6valuer les immunoglobulines sp6cifiques du mycobacterium dans le s6rum a 6t6 utilis6 avec succ~s dans le diagnostic rapide de la tuberculose. Peu d'6tudes ont 6t6 effect6es en Inde h une aussi grande 6chelle. Objet: Evaluer l'utilit6 de l'essai ELISA pour le diagnostic rapide des diff6rentes formes cliniques de la tuberculose dans la population indienne adulte. Schema: Un essai ELISA bas~ sur l'antig~ne mycobact~rien A60 (ANDA Biologicals, France), a ~t~ utilis6 afin d'6valuer les anticorps sp6cifiques IgM, IgA et IgG dans les s6rums provenant de 337 cas de tuberculose et de 131 t6moins parmi la population de Delhi, Inde. R(sultats: Sur les 131 t6moins, seulement 9,9% 6taient positifs pour IgM, 7,6% pour IgG, 6,1% pour IgA et 9,9% lorsqu'une combinaison d'IgA et d'IgG a 6t6 test6e. Sur 122 cas de tuberculose pulmonaire active, 41% 6taient positifs pour IgM, 86,8% pour IgA, 88,5% pour IgG et un taux 61ev6 de positivit6 (98,3%) a 6t6 observ6 lorsque les 6valuations pour IgA et IgG 6taient combin6es. Une positivit6 relativement faible a 6t6 remarqu6e dans 25 cas de tuberculose pleurale. Les chiffres correspondants pour 130 cas de tuberculose extrapulmonaire 6taient de 22,3%, 68,4%, 73,8% et 86,15%. Une analyse de 60 case de tuberculose pulmonaire ayant regu avec succbs un traitement antituberculeux a d6montr6 des taux de positivit6 de 11,6%, 46,6%, 58,3% et 66,6% respectivement. Correspondence to: Dr. Sunil Gupta, Assistant Director, Department of Microbiology, National Institute of Communicable Diseases, 22, Shamnath Marg, Delhi-110054. INDIA. Paper received 14 October 1994. Final version accepted 16 February 1995. 418
Role of A60 serologyin diagnosis of adult tuberculosis 419 Conclusion: Ces observations d~montrent une bonne sensibilit~ (91,6%) et sp~cificit~ (90,0%) pour ce test dans la d6tection de cas de tuberculose adulte, utilisant une combinaison de titres des anticorps IgA et IgG. Le r61e de la d6termination d'IgM peut 6te r~serv~ ~ la d6tection des cas de reprise de la tuberculose. R E S U M E N. Marco de referencia: En estudios efectuados en paises occidentales, para el diagn6stico r~ipido de la tuberculosis, se ha utilizado con 6xito el test ELISA basado en el antigeno micobacteriano A60, para evaluar las inmunoglobulinas especificas del mycobacterium en el suero. En India se han efectuado escasos estudios de una amplitud similar. Objetivo: Evaluar la utilidad del test ELISA para el diagn6stico r~ipido de las differentes formas clinicas de tuberculosis en la poblaci6n adulta de India. M(todo: Se utiliz6 el test ELISA basado en el antigeno micobacteriano A60 (Anda Biologicals, Francia), para evaluar los anticuerpos especificos IgM, IgA e IgG, en el suero proveniente de 337 casos de tuberculosis y de 131 controles en la poblaci6n de Delhi, India. Resultados: En los 131 controles, s61o el 9,9% era positivo para IgM, 7,6% para IgG, 6,1% para IgA y 9,9% cuando se evalu6 una combinaci6n de IgA y de IgG. En los 122 casos de tuberculosis pulmonar activa, el 41% era positivo para IgM, 86,8% para IgA, 88,5% para IgG y se observ6 una tasa elevada de positividad (98,3%) cuando se evalu6 IgA e IgG en forma combinada. Se constat6 una positividad relativamente baja en 25 casos de tuberculosis pleural. Las cifras respectivas para 130 casos de tuberculosis extrapulmonar eran 22,3%, 68,4% y 86,15%. El amilisis de 60 casos de tuberculosis pulmonar tratados con 6xito con medicamentos antituberculosos mostr6 una disminuci6n de las tasas de positividad: 11,6%, 46,6%, 58,3% y 66,6%, respectivamente. Conclusi6n: Estos hallazgos demuestran una buena sensibilidad (91,6%) y una buena especificidad (90,0%) del test cuando se combinan los titulos de anticuerpos IgA e IgG para la detecci6n de casos de tuberculosis en adultos. La evaluaci6n de IgM puede ser reservada a la detecci6n de casos de reactivaci6n de tuberculosis.
INTRODUCTION Acid-fast staining of sputum or another appropriate clinical specimen, followed by confirmation by culture, remains the cornerstone of laboratory diagnosis of tuberculosis (TB), and yet these techniques are either slow or lack sensitivity? The disadvantages of tuberculin testing as a diagnostic tool are well known: in addition to the problem of lack of specificity, the patient needs to be seen on two occasions. 2 Attempts have been made to improve the sensitivity and the speed of detection of tubercle bacilli or their components by techniques such as radiometric determination of bacterial growth, gas chromatography, DNA hybridization and polymerase chain reaction. However, all of these have met with problems of either sensitivity or specificity or have proved too technically demanding. 3 Recently, ELISA techniques employing various mycobacterial antigens, e.g. glycolipids from Mycobacterium bovis BCG, antigens 5 and 6 from M. tuberculosis, 64kDa protein of M. boris BCG and a 32kDa protein antigen called P32, have all been attempted for serological diagnosis of tuberculosis, with variable success. 4-10 A renewed attempt to develop a serodiagnostic test was initiated when Antigen 60 became available? I Antigen 60 is a major component of mycobacterial cytoplasm. Prepared from the cytoplasm of M. boris BCG and purified by exclusion gel chromatography, it is a macromolecular (106-107 daltons) lipopolysaccharide protein complex. ~2 Various scientists have made use of this antigen in the ELISA test for the serodiagnosis of tuberculosis by estimating IgM and IgG antibodies in
the serum of patients, with encouraging results. 12-22 There are some unpublished reports demonstrating the role of anti-A60 IgA serology in the diagnosis of tuberculosis. Considering the seriousness of the tuberculosis situation in our country, the present study was designed to evaluate the serodiagnostic potential of Antigen 60 in different clinical forms of tuberculosis in Delhi, India involving the adult (/> 18 years) age group, by measuring the A60-specific IgM, IgA and IgG antibody levels in the serum.
MATERIALS AND METHODS Study population: serum samples were obtained from 337 cases of adult tuberculosis, mainly from the RBTB Hospital (a 2000-bed tuberculosis hospital in Delhi) and other hospitals in Delhi. The diagnosis of TB was based on clinical and radiological criteria, tuberculin (Mantoux [Mx]) reactivity, histopathology, presence of acid-fast bacilli (AFB) and clinical response to antituberculosis treatment (ATT). The TB patients consisted of 147 cases of post-primary active pleuro-pulmonary TB and 130 cases of extra-pulmonary TB. 60 cases of inactive (treated) pulmonary TB were also included in the study.
Post primary active pleuro-pulmonary TB (147 cases) These included: 1. 122 patients with purely pulmonary TB (symptomatic at the time of the study), of whom 50 had received
420
Tubercle and Lung Disease
ATT for more than one month after several weeks of evolution of the disease, while 35 patients had interrupted the treatment after a few weeks (6-8 weeks). 37 patients had not received ATT at all. 67 of the cases were AFB-positive at some stage of the disease. 2. 25 patients suffering from pleural TB (symptomatic at the time of the study), of whom 8 had received ATT for a duration of less than 2 months. 15 cases had been AFB positive at some stage of the disease.
Post primary extra-pulmonary TB (130 cases) This group comprised: 1. 26 cases of miiiary TB; all were radiologically proved, symptomatic and untreated; 8 were AFB positive. 2. 22 cases of cerebrospinal TB, confirmed on the basis of CSF cytology biochemistry, AFB positivity or response to ATT. 3. 18 cases of osteoarticular TB, of whom 5 were AFB positive. 4. 26 cases of gastrointestinal TB, of whom 4 were AFB positive. 5. 18 cases of genitourinary TB, of whom 5 were AFB positive. 6. 20 cases of lymphoreticular TB, of whom 6 were AFB positive.
Post primary inactive pulmonary TB (60 cases) These subjects comprised cases of pulmonary TB who had been given ATT for 6-12 months at an early stage of the disease. They were all clinically healed and were negative for AFB at the time of the study.
Controls 131 age-, sex- and socioeconomically-matched controls were also included in the study. These consisted of 60 healthy adults, 37 of whom had been BCG-vaccinated in the past, 53 diseased controls (subjects suffering from various ailments but showing no evidence of TB), 10 contacts of cases of AFB-positive pulmonary TB, and 8 laboratory technicians handling mycobacterial cultures.
Serodiagnosis of TB by ELISA technique Anti A60 mycobacterial antibodies were estimated in the Table 1.
sera of the subjects under study employing indirect ELISA technique at the National Institute of Communicable Diseases, Delhi, as per the recommendations of the manufacturer (Anda Biologicals, Strasbourg, France). Each time, the positive as well as the negative reference sera provided with the diagnostic kit were included in the test along with the test sera. In the case of IgM antibodies a threshold value was determined from the optical density (OD) values obtained with the positive reference serum. For IgA and IgG determinations, the curves were constructed by plotting the OD values of different reference sera. Thereafter, concentration of the antibodies in test serum in terms of units/ml was determined by extrapolating the OD value of test serum against the reference curve. 14
Tuberculin (Mx) test This test was performed on all the subjects under study using ITU PPD RT-23 with Tween-80 in 0.1 ml dose. A diameter of induration of > 10 mm after 48 hours was considered a positive test? RESULTS
Determination of baseline setting for serological analysis Since the sensitivity (ST) and specificity (SP) of any serological test is dependent on the appropriate cut-off values chosen in the test, in this study we initially attempted to determine the most suitable cut-off values for the 3 types of A60 specific immunoglobulins, i.e. IgM, IgA and IgG, to differentiate the tuberculous cases from the controls. 1. Cut-off value of IgM: in the case of IgM antibodies an optical density index (ODI) was first determined by dividing the OD obtained from the test sample by the OD obtained with the positive reference serum. 23 Thereafter, 3 cut-off values were chosen arbitrarily, i.e. 1.00, 1.50 and 2.00, to segregate the cases of active TB from the controls, the respective values of sensitivity and specificity for the 3 cut-offs being 62.2% and 79.1%, 32.1% and 92.3%, and 13.1% and 95.8% (Table 1). We preferred to choose a cut-off value of 1.50 as it gave a very high specificity (92.3%), although the sensitivity was compromised (32.1%). However this could be countered later on by including IgG and IgA determinations.
Statistical evaluation of individual antibodies. Statistical evaluation of A60 ELISA at different cut-off points for adult tuberculosis
Cut-off value O.D.I. >
IgM ST %
SP %
Statistical evaluation of individual antibodies. IgA Cut-off value ST % SP % Cut-off value > Units/nil > Units/ml
IgG ST %
SP %
1.00
62.2
79.l
100
95.4
69.4
150
94.2
69.4
1.50
32.1
92.3
150
79.8
95.4
2.00
13.1
95.8
200
67.2
96.8
200 250 300
91.8 83.0 72.0
80.5 92.3 94.5
O.D.I. = Optical Density Index, ST = Sensitivity, SP = Specificity.
Role of A60 serology in diagnosis of adult tuberculosis 421 2. Cut off value for IgG: as can be seen in Table 1, 4 different cut-off values were considered i.e. 150, 200, 250 and 300 units/ml. However a titre of 250 units/ml was chosen as the cut-off point as this resulted in a very good specificity (92.3%) and a reasonably good sensitivity (83.0%). 3. Cut off value for IgA: 3 cut off values i.e. 100, 150, 200 units/ml were considered. The cut-off value of 150 units/ml gave a high specificity (95.4%) and a reasonable sensitivity (79.8%), and was accordingly chosen (Table 1). In order to further improve the specificity and sensitivity of the test, different combinations of the 3 immunoglobulins at the above cut-off points were tried. It was found that when both IgA and IgG values were used in conjunction (presence of at least one of these antibodies) a very good specificity (90.0%) mad sensitivity (91.6%) was obtained (Table 2). In serological analysis of controls with regard to A60 specific antibodies of IgM, IgA and IgG classes (Table 3), it was seen that among the healthy controls, very few subjects yielded a positive serology: only 5% were positive for IgM and IgA and slightly more (8.3%) were positive for the IgG antibodies. There was not much serological difference between BCG vaccinated and non-vaccinated controls. However, when the 8 techniTable 2. Statistical evaluation of combinations of Antibodies IgM, IgA and IgG at the cut-off points of _>1.50, 150 and 250 respectively
Type of Antibody Combination
ST %
SP %
IgM and IgA and IgG IgM or IgA or IgG IgA and IgG IgA or IgG IgM and IgG IgM or IgG IgM and IgA IgM or IgA
22.1 100.0 72.5 91.6 27.0 89.3 24.5 92.6
100.0 75.0 98.4 90.0 95.8 79.1 100.0 80.5
ST = Sensitivity, SP = Specificity.
cians were analysed, 25% were found to be positive for IgM and IgA and 50% for IgG antibodies. On the contrary, 10 contacts of cases of TB depicted a very low positivity. None of these subjects were found positive for either IgM or IgG, whereas only 1 (10%) had adequate titres of IgA antibodies. Among the 53 diseased controls, 11.3% were positive for IgM, and 1.9% for IgG antibodies, whereas none had adequate IgA antibody titres. A detailed analysis of different types of diseased controls is given in Table 3. Taking the overall picture of the 131 controls, 9.9% were positive for IgM, 6.1% for IgA, 7.6% for IgG and only 9.9% were found to be positive when IgA and IgG antibodies were considered in conjunction. On the other hand, a high degree of tuberculin (Mx) reactivity (38.9%) was observed (Table 3). The mean values of the 3 immunoglobulins (Igs) were 1.110 (ODI), 124.04 units/ml and 133.83 units/ml for IgM, IgA and IgG antibodies (Table 4).
Serological analysis of active p l e u r o - p u l m o n a r y TB (147 cases)
1. Pulmonary TB (122 cases) Of these, 41% were positive for IgM, 86.8% for IgA, 88.4% for IgG, whereas a very high positivity (98.3%) was seen when IgA and IgG estimations were combined. On the other hand, Mx positivity was only 53% (Table 5). A good correlation was seen between IgA and IgG positivity (75% of cases.) Among the 41% of cases who were positive for IgM antibodies the majority (85%) were cases of reactivation. The corresponding mean titres for the 3 antibodies were 1.376 (ODI), 898 units/ml and 1320 units/ml for IgM, IgA and IgG respectively (Table 4). 2, Pleural TB (25 cases) 40% positivity was seen for IgM, 56% for IgA, 72% for IgG and 88% for combined (IgA, IgG) es-
Table 3. Serologicalanalysis of non-tuberculous individuals (controls)
Subjects No of cases A. Healthy BCG vaccinated non BCG-vaccinated
Total B. Non-tuberculous patients. Asthma Chronic bronchitis Viral pyrexia Urinary tract infection Pneumonia (viral/bacterial) Rheumatoid arthritis Hepatocellular disease Total C. Contacts of cases of TB D. Subjects handling Mycobacteria Grand Total Mx = Tuberculin test (Mantoux).
A60 ELISA positivity in serum IgM IgA IgG No. (%) No. (%) No. (%)
Mx positivity IgA or IgG No. (%)
No. (%)
37 23
2 1
2 1
3 2
4 4
12 8
60
3(5.0)
3(5.0)
5(83)
7(11.6)
20(33.3)
9 8 12 6 10 4 4 53 10 8 131
1 1 0 0 1 1 2 6 (11.3) 0 (0) 2 (25.0) 13 (9.9)
0 0 0 0 1 0 0 1 (1.8) 1 (10.0) 4 (50.0) 13 (9.9)
3 5 4 1 3 4 1 21 (39.6) 5 (50.0) 5 (63.3) 51 (38.9)
0 0 0 0 0 0 0 0 (0) 1 (10.0) 4 (50.0) 8 (6.1)
0 0 0 0 1 0 0 1 (1.9) 0 (0) 4 (50.0) 10 (7.6)
422 Tubercle and Lung Disease Table 4.
A60 specific antibody titres in controls and cases of TB
Type of subject
NO
Controls
131
Cses of active Pulmonary TB Cases of active Pleural TB Cases of active Pleuro-pulmonary TB Cases of active Extra-pulmonary TB Cases of inactive Pulmonary TB
122 25 147 130 60
IgM (O.D.I)
Mean antibody titres (range) IgA (Units/ml)
IgG (Units/ml)
1.110 (0.365-2.670) 1.376 (0.672-3.120) 980 (0.478-2.056) 1.178 (0.478-3.120) 1.040 (0.536-2.780) 1.072 (0.380-1.960)
124.04 (< 10-240) 898 (100-3500) 152 (100-1600) 525 (100-3500) 222.30 (80-2200) 438 (50-800)
133.83 (20-350) 1320 (150-3600) 375 (125-2200) 847.50 (125-3600) 482.77 (80-2500) 865 (50-1800)
O.D.I = optical density index.
Table 5.
Serological analysis of cases of Tuberculosis (T.B.)
Subjects
A60 ELISA positivity in serum IgA IgG No. (%) No. (%)
Mx* positivity
No. of cases
IgM No. (%)
122 25 147
50 (41.0) 10 (40.0) 60 (40.8)
106 (86.8) 14 (56.0) 120 (81.6)
108 (88.5) 18 (72.0) 126 (85.7)
120 (98.3) 22 (88.0) 142 (96.6)
65 (53.0) 10 (40.0) 75 (51.0)
26 22 18 26 18 20
13 (50.0) 6 (27.2) 4 (22.2) 2 (7.6) 2 (11.1) 2 (10.0)
23 (88.4) 12 (54.5) 9 (50.0) 16 (61.5) 14 (77.7) 15 (75.0)
20 (76.0) 16 (72.7) 12 (66.6) 19 (73.0) 14 (77.7) 15 (75.0)
23 (88.4) 18 (81.8) 15 (83.3) 23 (88.4) 16 (88.8) 17 (85.0)
12 (46.1) 13 (59.0) 10 (55.5) 12 (46.2) 9 (55.0) 11 (55.0)
Total (B)
130
29 (22.3)
89 (68.4)
96 (73.8)
112 (86.1)
67 (51.5)
Total (A+B)
277
89 (32.1)
209 (75.4)
222 (80.1)
254 (91.6)
142 (51.2)
Inactive pulmonary
60
7 (11.6)
28 (46.6)
35 (58.3)
40 (66.6)
40 (66.6)
GRAND TOTAL
337
A. Active pleuro pulmonary Pulmonary Pleural Total (A) B. Active extrapulmonary Miliary Cerebrospinal Osteo-articular Gastro-intestinal Genito-urinary Lymphoreticular
IgA or IgG No. (%)
No. (%)
Mx* = Tuberculin test (Mantoux).
timations. A low Mx reactivity (40%) was observed. (Table 5). In 50% of cases, both, IgA and IgG antibodies were found. The mean titres of 3 antibodies, were 0.980 (ODI), 182 units/ml and 375 units/ml for IgM, IgA and IgG respectively (Table 4). On analysing all the 147 cases of active pleuropulmonary TB, it was observed that a very high percentage (96.6%) were positive for either IgG or IgA antibodies, whereas Mx reactivity was displayed in 51% cases only (Table 5). The mean titres of the 3 types of antibodies were 1.178 (ODI), 525 units/ml and 847.50 units/ml for IgM, IgA and IgG respectively (Table 4).
Serology of extra-pulmonary TB (130 cases) The disease-wise break up of cases is given in Table 5. Overall, 73.8% of cases displayed a positive IgG serology, 68.4% were positive for IgA and very few (22.3%) were positive for IgM antibodies. However, a
relatively high IgM positivity (50%) was seen in cases of miliary tuberculosis. The overall positivity came up to 86.1% on taking into account the combined (IgA, IgG) estimations, whereas tuberculin posifivity was displayed in 51.5% of cases only (Table 5). In 60% of cases, the presence of both IgA as well as IgG antibodies was seen. The mean values for the 3 types of antibodies were 1.040 (ODI), 222.3 units/ml and 482.7 units/ml for IgM, IgA and IgG respectively (Table 4).
Serology of inactive pulmonary TB (60 cases) An extremely low IgM positivity (11.6%) was seen followed by 46.6% for IgA and 58.3% for IgG antibodies, a reasonably high Mx reactivity (66.6%) was observed (Table 5). The mean titres for the 3 types of antibodies were 1.072 (ODI), 438 units/ml and 865 units/ml for IgM, IgA and IgG respectively. (Table 4)
Role of A60 serology in diagnosis of adult tuberculosis
Overall picture Considering all the cases of active tuberculosis and the controls, the test sensitivity of 91.6% and a specificity of 90.0% were obtained when combined (IgA, IgG) estimations were ,o taken into account. On the other hand, tuberculin reactivity was seen in 51.2% cases of active TB compared to 38.9% in the control population.
DISCUSSION The present study was performed to evaluate the role of A60 antigen in the serological diagnosis of tuberculosis (TB) in adults from the city of Delhi, India using the ELISA technique for the detection of A60 specific IgM, IgA and IgG antibodies. In this study, for IgM antibodies, a cut-off value of 1.500 of optical density index (ODI) appeared to give the best results. In the earlier studies undertaken in Western countries, a relatively lower cut-off value of 1.00 for ODI was chosen, which could have been due to the lower endemicity of TB in those countries] 5,19 As regards IgG antibodies, a cut-off value of 250 units/ml was chosen in this study to differentiate active TB cases from controls. In earlier studies performed in Italy and France, a cut-off value of 200 serounits/ml was chosen, 18'19 while in another study carried out in Belgium, the baseline was set at 100 units. H In a limited number of studies performed in India, mainly in the western zone, an IgG cut-off of 200 units was chosen. 16,2°,zz For IgA antibodies a cut-off value of 150 units/ml was selected to differentiate cases of active TB from controls. In the control group very few subjects yielded a positive serology against A60 antigen: these were mainly the laboratory staff who had been routinely handling the mycobacterial cultures. Only 9.9% were positive for IgM, 6.1% for IgA and 7.6% for IgG antibodies. Earlier reports on the seropositivity against A60 antigen in the control population ranged between 0 to 50% for IgG and 2.5 to 40% for IgM antibodies. The mean antibody levels in the control population in this study were appreciably lower compared to those in cases of active and inactive TB. Similar results were shown in another Indian study as regards IgG antibodies, z° Surprisingly, none of the human contacts of cases of TB showed a positive A60 serology. Moreover, no appreciable serological difference was observed between the BCG vaccinated and non-vaccinated controls. These findings are more or less in agreement with earlier reports.13,14,i6,17,19
A very good serological response was seen in cases of active pulmonary TB, especially with regard to IgA and IgG antibodies depicting a positivity of 86% and 88.5% respectively, the IgM positivity being rather low (41% only). Moreover, combinations of IgA and IgG antibodies showed a positivity as high as 98.3%. There was a
423
good correlation between IgA and IgG seropositivity (75%), however in a few cases (23%) either of the antibodies could be detected thereby necessitating the value of combined IgA and IgG serology for a better sensitivity. Further, it was observed that IgA response was better in the relatively recent infections. The IgM positivity was seen mainly in cases of reactivation as observed by others. The mean levels of all the 3 antibodies were appreciably higher than in the controls. In studies carried out earlier, an IgG positivity ranging from 48-100% and IgM positivity ranging from 5-76% has been documented. 15,~6,18-21 In one of these studies, 15 combined IgG and IgM estimation gave a sensitivity of 68%. These wide variations could be due to different age groups studied, geographical areas as well as severity of disease in different studies. 15,19 So far, we have not come across any published studies regarding anti A60 IgA serology in TB although good serum IgA responses were shown against p32 and M. bovis BCG sonicated antigen in some studies. 4,z5 In the present study, relatively low seropositivity was observed in cases of pleural TB compared to those of pulmonary TB as shown in a previous study. 18 As far as serology in extra pulmonary TB is concerued, although a high IgA (68.4%) and IgG (73.8%) positivity was seen, this was not the case with IgM (22.3% only). The positivity came to 86.1% when combined IgA and IgG values were considered. Again, a good correlation was observed in IgA and IgG positivity (60%) whereas in the rest of 26% of cases either IgA or IgG was present, indicating the importance of combined IgG and IgA determinations for a better sensitivity. The mean antibody levels were lower than in active pulmonary TB cases. In a few studies done in Western countries, an IgG positivity of 100% and IgM positivity of 0-15% have been determined in cases of extrapulmonary TB, I9,al whereas in Indian studies an IgG positivity of only 65-80% has been shownJ 6,2z On analysis of cases of inactive/(treated) pulmonary tuberculosis, quite a large number were found to be positive for IgA (58.3%) and IgG (46.6%), though this was not the case with IgM antibodies (11.6% positivity only). Moreover, the mean antibody levels were also much higher than those of the controls, although the levels were slightly lower than those in active pulmonary tuberculosis. Few available reports in this aspect show a substantial lowering of anti A60 antibodies with clinical recovery after treatment. 13,~9 Overall, fairly satisfactory results were obtained by estimating A60 specific IgA and IgG antibodies for rapid diagnosis of adult tuberculosis (ST = 91.6%, SP = 90.0%). The role of IgM estimation was found to be restricted to detecting cases of reactivation or to some extent cases of miliary tuberculosis. On the other hand, the tuberculin test could hardly differentiate the cases of active tuberculosis from the controls and there was hardly any correlation between tuberculin reactivity and A60 serology, as shown by others also. m9 In a previous study done in the West, a test sensitivity of 98.6% and a
424
Tubercle and Lung Disease
specificity of 86.7% was demonstrated in post primary TB by estimating A60 specific IgG antibodies only. 19 Isolated reports have shown good results in ELISA tests using other mycobacterial antigens for the diagnosis of TB, e.g 'antigen 6' (ST = 94%, SP = 100%), 4 'antigen 5' (ST = 48.8%, SP = 91-100%) 8, and a dual ELISA system using A60 and antigen cytosolic fraction of M. tuberculosis (ST = 97%, SP = 100%). 17 However, the results are still to be substantiated by other researchers.
10.
CONCLUSION
11.
The present study has conclusively proven the value of the estimation of A60 specific IgA and IgG antibodies for the rapid diagnosis of adult tuberculosis, especially the extrapulmonary type, where one does not have many other diagnostic modalities to choose from. It was also found to be useful to detect paucibacillary (AFB negative) cases of pulmonary tuberculosis.
8.
9.
12.
13.
14.
15.
Acknowledgements The authors are extremely grateful to the Director of the National Institute of Communicable Diseases, Delhi for providing the necessary facilities and to Shri Aalok Gupta and Shri Amar Jit Singh for providing secretarial assistance for the study. The technical assistance provided by Shri Mahabir Prasad is also acknowledged.
16. 17. 18.
19.
References 1. Daniel T M. The rapid diagnosis of tuberculosis: a selective review. J Lab Clin Med 1990; 116: 277-282. 2. Crofton J. et al ed. Appendix E. Tuberculin testing. In: Clinical tuberculosis, Macmillan Publishers Ltd, London, 1992; pp 187-192. 3. Grange J M, Laszlo A. Serodiagnostic test for tuberculosis: a need for assessment of their operational predictive accuracy and acceptability. Bull WHO 1990; 68(5): 571-576. 4. Turneer M, Van Vooren J P, Bmyn J et al. Humoral immune response in human tuberculosis: immunoglobulins G, A and M directed against the purified P32 protein antigen of Mycobacterium boris, bacillus Calmette-Gu6rin. J. Clin. Microbiol 1988; 26: 1714-1719. 5. Reggiardo Z, Vazquez E, Schnaper L. ELISA test for antibodies against mycobacterial lipids. J Immunol Methods 1980; 34: 55-60. 6. Reggiardo Z, Vazquez E. Comparison of enzyme-linked imunnosorbent assay and hemaggulutination test using mycobacterial glycolipids. J Clin Microbiol 1981; 13: 1007-1009. 7. Stroebel A B, Daniel T M, Lau J H K, Leong J C V, Richardson
20.
21.
22.
23.
24.
25.
H. Serologic diagnosis of bone and joint tuberculosis by an enzyme linked immunosorbent assay. J Infect Dis 1982; 146: 280-283. Benjamin R G, Daniel T M. Serodiagnosis of tuberculosis using the enzyme-linked immunosorbent assay of antibody to Mycobacterium tuberculosis antigen 5 Am Rev Respir Dis 1982; 126(1): 1013-1016. Benjamin R G, Debanne S M, Ma Y, Daniel T M. Evaluation of mycobactefial antigens in an enzyme-linked immunosorbent assay (ELISA) for the serodiagnosis of tuberculosis. J Med Microbiol. 1985; 18: 309-318. Thole J E R, Keulen W J, De Bruyn J e t al. Characterization, sequence determination and immunogenicity of a 64 kilodalton protein of Myeobacterium boris BCG expressed in Escherichia coli K-12. Infect Immun 1987; 55: 1466-1475. Cocito C, Vanlinden F. Preparation and properties of antigen 60 from Mycobaeterium boris BCG. Clin Exp Immunol 1986; 66: 262-272. Cocito C G. Properties of the mycobacterial antigen complex A60 and its application to the diagnosis and prognosis of tuberculosis. Chest 1991; 100: 1687-1693. Baelden M C, Vanderelst B, Dieng M e t al. Serologic analysis of human tuberculosis by an ELISA with mycobacterial antigen 60. Scand J Inf Dis 1990; 22: 63-76. Maes R, Homasson J P, Kubin M e t al. Development of an enzyme immunoassay for the serodiagnosis of tuberculosis and mycobacteriosis. Med Microbiol Immunol 1989; 178: 323-335. Charpin D, Herbanlt H, Gevaudan M J et al. Value of ELISA using A60 antigen in the diagnosis of active pulmonary tuberculosis. Am Rev Respir Dis 1990; 142 (2): 380-384. Patel A B, Rawat M S, Grover S. ELISA for diagnosis of tuberculosis. Indian Paediatr 1990; 27: 584-589. Reheman S F, Wagner S, Manch H et al. Evaluation of dual antigen ELISA test for the serodiagnosis of tuberculosis. Bull. WHO 1988; 66(2): 203-209. Fadda G, Gfillo R, Ginesu F et al. Serodiagnosis and follow up of patients with pulmonary tuberculosis by ELISA. Eur J Epidemiol 1992; 8(1): 810-817. Gevaudan M J, Bollet C, Charpin D, Mallet M N. Serological response of tuberculosis patients to antigen 60 of BCG. Eur J Epidemiol 1992; 8(5): 666-676. Gokhale N T, Ranadive S N, Joshi A M e t al. Antibody and immune complex ELISA in the immunodiagnosis of tuberculosis. Med J Arm Force India 1993; 49(3): 159-163. Maes R. Clinical usefulness of serological measurement obtained by Antigen 60 in mycobactefial infections: development of a new concept. Kiln Wochenschr 1991; 69: 696-709. Munshi M M, Chiddarwar S, Patel A, Grover S. Serodiagnosis of extrapulmonary tuberculosis by ELISA. Indian J Pathol Microbiol 1993; 36(4): 356-360. Daniel T M, Dabanne S M. The serodiagnosis of tuberculosis and other mycobacterial disease by enzyme linked immunosorbent assay. Am Rev Respir Dis 1987; 135:1137-1151 Wirrmann C. Public health application of a serological test for tuberculosis: study of the incidence of inapparent infections among the employees of an Alsatian supermarket. Eur J Epidemiol 1990; 6(3): 304-308. Grange J M, Gibson J, Nassau E et al. ELISA, a study of antibodies to Mycobacterium tuberculosis in the IgG, IgA and IgM classes in tuberculosis, sarcoidosis and Crohn's disease. Tubercle Lung Dis 1980; 61: 145-152.