JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 2004, p. 3435–3437 0095-1137/04/$08.00⫹0 DOI: 10.1128/JCM.42.8.3435–3437.2004 Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Vol. 42, No. 8
Evaluation of Immunoglobulin M (IgM) and IgG Rapid Cassette Test Kits for Diagnosis of Melioidosis in an Area of Endemicity Vanaporn Wuthiekanun,1 Premjit Amornchai,1 Wirongrong Chierakul,1 Allen C. Cheng,2 Nicholas J. White,1,3 Sharon J. Peacock,1,3 and Nicholas P. J. Day1,3* Faculty of Tropical Medicine, Mahidol University, Bangkok 10400, Thailand1; Menzies School of Health Research and Northern Territory Clinical School, Flinders University, Darwin, Australia2; and Nuffield Department of Clinical Medicine, John Radcliffe Hospital, University of Oxford, Oxford OX3 9DU, United Kingdom3 Received 2 February 2004/Returned for modification 12 March 2004/Accepted 23 March 2004
An enzyme-linked immunosorbent assay-based rapid cassette immunoglobulin G (IgG) and IgM immunochromogenic test kit was compared to the indirect hemagglutination test (IHA) for the diagnosis of acute melioidosis in northeastern Thailand. Admission sera from 70 culture-confirmed septicemic melioidosis patients and 30 patients with localized infections were tested. As a control group, 80 patients with other acute febrile illnesses (other bacterial infections, leptospirosis, or scrub typhus) and 119 healthy individuals were tested. The diagnostic sensitivity of the IgG and IgM tests and the IHA test were 79, 67, and 72%, respectively, with corresponding specificities of 90, 80, and 68%. This kit represents an improvement over IHA for the diagnosis of melioidosis an area of endemicity although, as with other serological tests, it has reduced diagnostic utility in a population with high background seropositivity. for IgM and IgG and found that the detection of IgG was a better indicator of disease and had potential clinical utility (5). Reliable, simple rapid tests with high sensitivity for melioidosis would be a great advance, particularly in rural settings where bacteriology facilities are not available. We have evaluated here a new rapid immunochromogenic test for melioidosis on retrospective patient sera from an area of endemicity in Thailand.
Melioidosis is an infectious disease caused by the saprophytic gram-negative bacterium Burkholderia pseudomallei, which is found widely in soil and surface water in East Asia and northern Australia. Disease manifestations are broad and range from an acute fulminant septicemia to a chronic localized infection that may be difficult to differentiate from other infections (15). Infection requires treatment with expensive and specific antibiotics (ceftazidime or carbapenems) that are not generally used as empirical treatment for sepsis in areas of endemicity; thus, delays in microbiological confirmation of the infection may have fatal consequences. In northeastern Thailand, the mortality rate for disease as a whole is up to 50%, with death occurring in 80% of those presenting with septic shock. The definitive diagnosis of melioidosis is made by the isolation of B. pseudomallei from infected sites or bodily fluids, but cultures often take 2 days or more to become positive and are not available in small rural hospitals (10). Direct immunofluorescence microscopy can detect B. pseudomallei from sputum, urine, or pus with a specificity of 99% and sensitivity of 73% compared to culture (13) and require ⬍2 h. However, this test is not commercially available, and it requires both suitable specimens and specialized microscopy facilities. In Thailand, the most widely use method for serodiagnosis is the indirect hemagglutination assay (IHA), which detects both immunoglobulin M (IgM) and IgG antibodies (2). Since most of the population in the areas of the country where the disease is endemic are seropositive by the IHA test after 4 years of age as a result of repeated environmental exposure to the organism, this test is useful only for excluding melioidosis in areas of endemicity (8). Chenthamarakshan et al. reported the development of an enzyme-linked immunosorbent assay (ELISA)
MATERIALS AND METHODS Study patients and serum samples. Admission sera from unselected patients with culture-proven melioidosis were prospectively obtained from 100 patients admitted between 1994 and 2002 to Sapprasitiprasong Hospital, a major referral center in Ubon Ratchathani in northeastern Thailand. Of these, 70 patients were bacteremic; 29 of these individuals had involvement of more than one site (“disseminated infection”), and 30 patients had localized (single site, nonbacteremic) melioidosis. The sera were drawn and stored at the same time as specimens were taken for initial diagnostic culture. Duration of symptoms prior to admission was recorded. Overall, the in-patient mortality in this group was 28%. Control sera were obtained prospectively from patients with acute febrile illnesses admitted to the same hospital during the same period (n ⫽ 80). Thirty patients were suspected initially of having melioidosis, but blood cultures were positive for other bacterial pathogens, including Escherichia coli, Staphylococcus aureus, and Klebsiella pneumoniae. Another 50 control sera were collected from patients with serologically or culture-confirmed leptospirosis or scrub typhus from a different research laboratory in the area of melioidosis endemicity. These patients were diagnosed on the basis of blood culture or microscopic agglutination test for leptospirosis, and immunofluorescent antibody assay for scrub typhus. Control sera from healthy individuals (n ⫽ 119) were obtained in 1999 from blood donors who were either rice farmers or their relatives (n ⫽ 112) or from staff at Sapprasitiprasong Hospital (n ⫽ 12). This was done to define the background positivity and therefore estimate the likely lower limit of specificity of the test. Bacterial cultures. Blood culture was performed by inoculating 3 to 5 ml of blood into standard media, which was incubated aerobically at 37°C in air. Swabs or samples were taken from any suspected site of infection. Swabs from nonsterile sites were preincubated in a selective broth. Blood and other samples were cultured on horse blood agar and Ashdown’s selective media; positive cultures were identified as reported previously (14). Serological testing. Melioidosis IgM and IgG Rapid Cassette Test kits were kindly supplied by PanBio, Ltd., Windsor, Queensland, Australia. The test was
* Corresponding author. Mailing address: Faculty of Tropical Medicine, Mahidol University, 420/6 Rajvithi Rd., Bangkok, 10400 Thailand. Phone: 662-3541395. Fax: 662-3549169. E-mail:
[email protected]. 3435
3436
WUTHIEKANUN ET AL.
J. CLIN. MICROBIOL.
TABLE 1. Sensitivity and specificity of the IgG and IgM immunochromogenic assays compared to IHAa
D⫺ T⫹
D⫺ T⫺
% Sensitivity (95% CI)
% Specificity (95% CI)
P
79 67 76
21 33 24
20 39 69
179 160 130
79 (70–87) 67 (57–76) 76 (66–84)
90 (85–94) 80 (74–86) 65 (58–72)
⬍0.001 ⬍0.001 ⬍0.001
IgG RCT IgM RCT IHA
79 67 76
21 33 24
7 16 20
73 64 60
79 (70–87) 67 (57–76) 76 (66–84)
91 (83–96) 80 (70–88) 75 (66–84)
⬍0.001 ⬍0.001 ⬍0.001
IgG RCT IgM RCT IHA
79 67 76
21 33 24
13 23 49
106 96 70
79 (70–87) 67 (57–76) 76 (66–84)
89 (70–87) 81 (72–87) 59 (49–68)
⬍0.001 ⬍0.001 ⬍0.001
Test
Melioidosis patients vs all controls
IgG RCTc IgM RCT IHA
Melioidosis patients vs acute febrile illness controls Melioidosis patients vs blood donor controls a b c
No. of samplesb D T⫺
Comparison
⫹
D T
⫹
⫹
95% CI, 95% confidence interval. D⫹ and D⫺, positive and negative for disease, respectively; T⫹ and T⫺, positive and negative test result, respectively. RCT, rapid cassette test.
performed as described previously (6). In brief, 5 l of serum was added to the cassette with 3 drops of buffer, and the results were read after 15 min. Any trace of a pink or purple line was interpreted as a positive result. The results of all cassette tests were independently agreed upon by four investigators, with no discrepancies between observers. The IHA assay was performed according to the method described previously. A positive test was regarded as a titer of 1:160 or more (9, 11). Statistical analysis. Sensitivities and specificities were calculated with exact 95% confidence intervals by using the Stata 8.1 statistical software package (Stata Corp., College Station, Tex.).
RESULTS Sensitivity. Of the 100 patients with culture-confirmed melioidosis, the IgG cassette test was positive in 79 cases (sensitivity, 79%), the IgM cassette test was positive in 67 (sensitivity, 67%), and IHA was positive in 72 cases (sensitivity, 72%) (Table 1). In patients with bacteremic melioidosis (n ⫽ 70), the IgG, IgM, and IHA sensitivities were 77, 66, and 71%, respectively. For patients with localized infection (n ⫽ 30), the sensitivities were 83, 70, and 73%, respectively. Of the 24 patients with negative IgG results, 10 patients had a positive IgM; thus, the combined sensitivity of either a positive IgG or IgM cassette test was 86%. However, only 57 patients had both positive IgG and positive IgM tests (sensitivity, 57%). Specificity. Of the 199 nonmelioidosis controls in the present study (consisting of 80 acute febrile control patients and 119 healthy blood donors), the IgG cassette test was negative in 179 (specificity, 90%), the IgM cassette test was negative in 160 (specificity, 80%), and the IHA was negative in 134 (specificity, 68%). Of the 20 patients that were IgG positive, 7 patients were IgM negative and 13 were IgM positive; thus, the specificity of having both of the two tests positive was 93%. When sera from controls with acute febrile illnesses such as leptospirosis, scrub typhus, and other bacterial sepsis were considered (n ⫽ 80), the specificities of the IgG, IgM, and IHA tests were 91, 80, and 75%, respectively. If only healthy controls (n ⫽ 119) were taken into account, the sensitivities of the IgG, IgM, and IHA test were 89, 81, and 71%. Within the group of patients with culture-proven melioidosis, there was no relationship between the duration of symptoms prior to the diagnostic testing and the rates of positivity of the IgG cassette test, the IgM cassette test, or of the IHA. Similarly, the rates of positivity for all three tests did not vary
significantly with the type of disease (localized, bacteremic, and disseminated) or with in-patient mortality. DISCUSSION Screening tests for melioidosis are required to have a high sensitivity and specificity, and at the same time be rapid, simple, and affordable. Since the clinical diagnosis is difficult, the antibiotic treatment is specific and expensive, and bacteriology laboratories are not available in many areas of endemicity, a rapid serological test would be a major advance with considerable public health benefits. In the past several serological tests for the diagnosis of melioidosis with reportedly high specificities and sensitivities have been described, but difficulties in interpretation, variable reproducibility, and the lack of commercial availability have meant that these tests are not widely available or used (1, 7). We found the current ELISA-based tests used in the present study to be rapid (requiring approximately 15 min to perform) and easy to use and interpret. Our results are comparable to those of a previous study using a quantitative ELISA for the detection of IgG antibodies against B. pseudomallei (5). In the present study, the sensitivity of the test was somewhat higher, possibly because sera were obtained late in the disease course in a number of patients. We chose here to use only admission samples, since they more closely reflect the “real-life” situation in which the test is most likely to be used and to be of value in determining patient management. Previous studies of the humoral response in melioidosis have demonstrated that IgG levels (particularly IgG1 and IgG2 subclasses) may be a good indicator of active disease (5) and may also be useful in monitoring response to therapy (12). In contrast, the IgM antibody response is lower in acute disease and is prevalent in the sera of the healthy but exposed population in areas of endemicity (4). The difference in specificity between the IgG assay and the IHA, thought to detect mainly IgM but also IgG antibodies, may be due to these differences in antibody class response (2). The retrospective design of the present study did not allow us to define the positive and negative predictive values of test results. However, the cases and controls chosen for study reflect approximately the prevalence of melioidosis and other
VOL. 42, 2004
RAPID SEROLOGICAL DIAGNOSIS OF MELIOIDOSIS
causes of febrile illness in the patient group in which this test is most likely to be applied. In previous studies we have noted that the proportion of community-acquired bacteremias due to B. pseudomallei ranged from ⬍10% in the dry season to up to 50% during the monsoon season (3). In addition, other common causes of febrile illness in this region of Thailand include leptospirosis and scrub typhus, both diseases caused by organisms not routinely cultured. We estimate that the proportion of febrile patients ill enough to require in-patient evaluation who have melioidosis during the monsoon season is between 10 and 30%. Given this pretest probability, the IgG assay may be useful in excluding melioidosis. Based on this estimated prevalence, the IgG assay would be expected to have a negative predictive value of between 91 and 98%, and the IgM assay would be expected to have a negative predictive value of between 85 and 96%, compared to the IHA range of 86 to 96%. Despite improvements, the immunochromogenic tests evaluated here fall well short of the sensitivity needed to confidently diagnose melioidosis. At 10 to 30% prevalence, the positive predictive values of the IgG test are between 49 and 79%, those for the IgM assay are between 27 and 59%, and those for the IHA are between 24 and 55%. This suggests that the IgG assay may be useful if combined with clinical rules that increase the pretest probability of melioidosis. A prospective study to validate this is under way. ACKNOWLEDGMENTS We thank our colleagues in the Wellcome Trust-Mahidol University-Oxford Tropical Medicine Research Programme and in Sappasitiprasong Hospital, Ubon Ratchatani, especially Wipada Chaowagul for advice. We also thank Serm Surapinit and Pisit Sirisuwan, students from the Faculty of Allied Health Sciences, Chulalongkorn University, for assistance with this study. Test kits for this study were provided by PanBio, Ltd., Windsor, Queensland, Australia. S.J.P. holds a Wellcome Trust Career Development Award in Clinical Tropical Medicine. This study was part of the Wellcome Trust-Mahidol UniversityOxford Tropical Medicine Research Programme, funded by the Wellcome Trust of Great Britain.
3437
REFERENCES 1. Anuntagool, N., P. Rugdech, and S. Sirisinha. 1993. Identification of specific antigens of Pseudomonas pseudomallei and evaluation of their efficacies for diagnosis of melioidosis. J. Clin. Microbiol. 31:1232–1236. 2. Ashdown, L. R. 1987. Indirect haemagglutination test for melioidosis. Med. J. Aust. 147:364–365. 3. Chaowagul, W., N. J. White, D. A. Dance, Y. Wattanagoon, P. Naigowit, T. M. Davis, S. Looareesuwan, and N. Pitakwatchara. 1989. Melioidosis: a major cause of community-acquired septicemia in northeastern Thailand. J. Infect. Dis. 159:890–899. 4. Chenthamarakshan, V., M. V. Kumutha, J. Vadivelu, and S. D. Puthucheary. 2001. Distribution of immunoglobulin classes and IgG subclasses against a culture filtrate antigen of Burkholderia pseudomallei in melioidosis patients. J. Med. Microbiol. 50:55–61. 5. Chenthamarakshan, V., J. Vadivelu, and S. D. Puthucheary. 2001. Detection of immunoglobulins M and G using culture filtrate antigen of Burkholderia pseudomallei. Diagn. Microbiol. Infect. Dis. 39:1–7. 6. Cuzzubbo, A. J., V. Chenthamarakshan, J. Vadivelu, S. D. Puthucheary, D. Rowland, and P. L. Devine. 2000. Evaluation of a new commercially available immunoglobulin M and immunoglobulin G immunochromatographic test for diagnosis of melioidosis infection. J. Clin. Microbiol. 38:1670–1671. 7. Dharakul, T., S. Songsivilai, N. Anuntagool, W. Chaowagul, S. Wongbunnate, P. Intachote, and S. Sirisinha. 1997. Diagnostic value of an antibody enzyme-linked immunosorbent assay using affinity-purified antigen in an area endemic for melioidosis Am. J. Trop. Med. Hyg. 56:418–423. 8. Kanaphun, P., N. Thirawattanasuk, Y. Suputtamongkol, P. Naigowit, D. A. Dance, M. D. Smith, and N. J. White. 1993. Serology and carriage of Pseudomonas pseudomallei: a prospective study in 1000 hospitalized children in northeast Thailand. J. Infect. Dis. 167:230–233. 9. Leelarasamee, A. 1985. Diagnostic value of indirect hemagglutination method for melioidosis in Thailand. J. Infect. Dis. Antimicrob. Agents 2:213–214. 10. Leelarasamee, A., S. Trakulsomboon, M. Kusum, and S. Dejsirilert. 1997. Isolation rates of Burkholderia pseudomallei among the four regions in Thailand. Southeast Asian J. Trop. Med. Public Health 28:107–113. 11. Naigowit, P., W. Petchkanchanapong, W. Mannebungyung, N. Vetchsprsit, and D. Chaeosil. 1989. Indirect hemagglutination test for the diagnosis of melioidosis. Bull. Fac. Med. Tech. Mahidol Univ. 13:17–25. 12. Vasu, C., J. Vadivelu, and S. D. Puthucheary. 2003. The humoral immune response in melioidosis patients during therapy. Infection 31:24–30. 13. Walsh, A. L., M. D. Smith, V. Wuthiekanun, Y. Suputtamongkol, V. Desakorn, W. Chaowagul, and N. J. White. 1994. Immunofluorescence microscopy for the rapid diagnosis of melioidosis. J. Clin. Pathol. 47:377–379. 14. Walsh, A. L., and V. Wuthiekanun. 1996. The laboratory diagnosis of melioidosis. Br. J. Biomed. Sci. 53:249–253. 15. White, N. J. 2003. Melioidosis. Lancet 361:1715–1722.
