Abstract
A prospective population-based molecular and conventional epidemiological study of 65.4% of bacteriologically confirmed cases of tuberculosis was carried out on the island of Hong Kong from May 1999 to Oct 2000 by the IS6110-based restriction fragment length polymorphism technique. Eleven of the isolates had five or fewer bands; 24.5% of the remaining 691 isolates belonged to clusters. The estimated proportion of recently transmitted disease was 15 to 20%.
DNA fingerprinting of Mycobacterium tuberculosis has been used in population-based studies to determine the proportion of recently transmitted disease and to assess risk factors associated with recent transmission (1, 4, 5, 6, 7, 8, 10, 13, 14, 17, 19, 20). These studies were carried out mostly in countries with a low prevalence of tuberculosis (1, 4, 5, 6, 7, 10, 14, 17, 20), while a few were done in countries with a high prevalence of tuberculosis and human immunodeficiency virus (HIV) infection (13, 16, 19). There has been no population-based study in places with an intermediate burden of tuberculosis and a low prevalence of HIV infection. We conducted a prospective molecular and conventional epidemiological study of tuberculosis on the island of Hong Kong, which had a tuberculosis notification rate of 108 per 100,000 persons in 2001 (3) and a low prevalence of HIV infection (2), to estimate the proportions of reactivated and recently transmitted disease. This is a report of the results of the first 18 months of the study.
The study was approved by the Ethics Committee of The University of Hong Kong. All persons with culture-positive cases of tuberculosis residing on the island of Hong Kong from 1 May 1999 to 30 October 2000 were recruited for this study. Public health nurses carried out contact investigations and collected the following information: birth place, current and past residences, school or work address, travel history, and contact history in the past 2 years. Information about any past history of tuberculosis and treatment, type and extent of disease, concomitant medical illnesses, and substance abuse was obtained from program forms completed by physicians with the Hong Kong Tuberculosis and Chest Service.
The Tuberculosis Reference Laboratory, Public Health Laboratory Centre, Department of Health, Hong Kong, performed all mycobacteriological investigations. IS6110-based restriction fragment length polymorphism (RFLP) analysis was carried out on isolates in accordance with standardized methods with a 245-bp right-sided probe and internal molecular weight standards (11). Analysis of gel images was done with Molecular Analyst Software Fingerprinting DST, version 1.6 (Molecular Bioscience Group, Bio-Rad Laboratories, Hercules, Calif.). Lanes found to have similar patterns were compared visually and classified as having matching RFLP patterns if the numbers and molecular weights of all of the bands were identical.
Patients were divided into two groups, those with clustered isolates and those with unique isolates. The proportion of unique clusters and potential bias was calculated by the n and n − 1 methods as suggested by Murray and Alland (15). Differences in the distribution of demographic and clinical variables between the two groups were examined by bivariate analysis. Risk ratios (RR) and 95% confidence intervals (CI) were calculated. A definite epidemiological link was considered to be present when clustered patients shared a home, workplace, or the same residential building, while a probable epidemiological link was considered to be present when clustered patients lived in the same housing estate or within 2 city blocks. Statistical analyses were carried out with the SPSS statistical package (SPSS for Windows, version 10).
There were 1,075 culture-positive cases of tuberculosis during the first 18 months of the study. Seven hundred two (65.3%) isolates were available for RFLP analysis. Eleven (1.6%) isolates with five or fewer bands were excluded from the present analysis. Five hundred twenty-two of the 691 isolates had unique patterns, while the remaining 169 belonged to 63 clusters. The proportion of clustered samples was 24.5% by the n method and 15.3% by the n − 1 method, with potential underestimation biases due to incomplete sampling estimated at 6.2 and 4.0%, respectively. A majority of the clusters (39 of 64) had two patients; only 2 clusters had seven patients.
The demographic and clinical characteristics of patients with clustered and unique M. tuberculosis DNA fingerprints are shown in Tables 1 and 2. Clustered isolates were significantly more likely to come from permanent residents (in Hong Kong for ≥7 years) than new residents or nonresidents (RR, 3.89; 95% CI, 1.16 to 13.1) and were significantly more likely to come from those who had traveled to mainland China in the past 2 years than from those who had not (RR, 1.49; 95% CI, 1.01 to 2.20). There was a trend for the unemployed (retired persons, housewives, and students) to be less likely to belong to clusters than the employed. Other characteristics, as shown in the tables, did not differ between the two groups. The demographic and clinical characteristics of patients whose sputum isolates were subjected to RFLP analysis were not different from those whose sputum isolates were not analyzed.
TABLE 1.
Demographic characteristics of patients by clustered and unique groups
| Characteristic | Available no. | Clustered (n = 169) | Unique (n = 522) | RR (95% CI) |
|---|---|---|---|---|
| Male sex | 691 | 121 (71.6) | 364 (69.7) | 1.09 (0.75-1.60) |
| Age (yr) | 691 | |||
| <40 | 49 (29.0) | 135 (25.9) | 1 | |
| 40-60 | 51 (30.2) | 147 (28.2) | 0.96 (0.61-1.51) | |
| >60 | 69 (40.8) | 240 (46.0) | 0.79 (0.52-1.21) | |
| Birthplace | 687 | |||
| Hong Kong | 68 (40.5) | 181 (34.9) | 1 | |
| China | 94 (56.0) | 301 (58.0) | 0.83 (0.58-1.19) | |
| Other | 6 (3.6) | 37 (7.1) | 0.43 (0.17-1.07) | |
| Residency | 689 | |||
| Permanent | 165 (98.2) | 481 (92.3) | 4.57 (1.39-14.9) | |
| New residents, nonresidents | 3 (1.8) | 40 (7.7) | 1 | |
| Institution | 684 | |||
| No | 152 (93.3) | 468 (93.4) | 1 | |
| Yes | 11 (6.7) | 33 (6.6) | 1.03 (0.51-2.08) | |
| Employment | 684 | |||
| Employed | 78 (46.4) | 177 (34.3) | 1 | |
| Unemployed | 24 (14.2) | 88 (17.1) | 0.62 (0.37-1.05) | |
| Retired | 50 (29.6) | 176 (34.1) | 0.64 (0.43-0.97) | |
| Housewife or student | 16 (9.5) | 75 (14.5) | 0.48 (0.27-0.88) | |
| Travel to mainland | 689 | |||
| China | ||||
| No | 68 (40.5) | 282 (54.1) | 1 | |
| Yes | 100 (59.5) | 239 (45.9) | 1.74 (1.22-2.47) | |
| Asia | ||||
| No | 144 (85.7) | 461 (88.5) | 1 | |
| Yes | 24 (14.3) | 60 (11.5) | 1.28 (0.77-2.13) | |
| Contact history | 691 | |||
| No | 157 (94.1) | 490 (93.9) | 1 | |
| Yes | 10 (5.9) | 32 (6.1) | 0.90 (0.43-1.89) |
TABLE 2.
Clinical characteristics of patients by clustered and unique groups
| Characteristic | Available no. | Clustered (n = 169) | Unique (n = 522) | RR (95% CI) |
|---|---|---|---|---|
| Smoking | 686 | |||
| Nonsmoker | 69 (41.1) | 229 (44.2) | 1 | |
| Exsmoker | 60 (35.7) | 176 (34.0) | 1.13 (0.76-1.68) | |
| Current smoker | 39 (23.2) | 113 (21.8) | 1.15 (0.73-1.80) | |
| Imprisonment | 679 | |||
| No | 157 (96.3) | 492 (95.3) | 1 | |
| Yes | 6 (3.7) | 24 (4.7) | 0.78 (0.31-1.95) | |
| Drug abuse | 683 | |||
| No | 157 (95.2) | 498 (96.1) | 1 | |
| Yes | 8 (4.8) | 20 (3.9) | 1.27 (0.55-2.94) | |
| Alcohol abuse | 683 | |||
| No | 145 (87.9) | 465 (89.8) | 1 | |
| Yes | 20 (12.1) | 53 (10.2) | 1.21 (0.70-2.09) | |
| HIV status | 637 | |||
| No | 94 (55.6) | 258 (49.4) | 1 | |
| Yes | 0 | 3 (0.6) | ||
| Not done | 75 (44.4) | 261 (50.0) | 0.79 (0.56-1.12) | |
| Comorbid illness | 691 | |||
| No | 104 (61.5) | 311 (59.6) | 1 | |
| Any | 65 (38.5) | 211 (40.4) | 0.92 (0.65-1.31) | |
| Site of disease | 691 | |||
| Pulmonary only | 157 (92.9) | 456 (87.5) | 1 | |
| Extrapulmonary | 12 (7.1) | 66 (12.5) | 0.54 (0.28-1.02) | |
| Case category | 691 | |||
| New case | 139 (82.4) | 420 (80.5) | 1 | |
| Relapse | 24 (14.2) | 84 (16.1) | 0.86 (0.53-1.41) | |
| Previous default | 3 (1.8) | 5 (1.0) | 1.81 (0.43-7.68) | |
| Other | 3 (1.8) | 13 (2.5) | 0.70 (0.20-2.48) | |
| Drug resistance | 668 | |||
| None | 148 (88.6) | 444 (88.6) | 1 | |
| Single drug | 15 (9.0) | 51 (10.2) | 0.89 (0.49-1.63) | |
| Multidrug | 4 (2.4) | 6 (1.2) | 2.01 (0.56-7.23) | |
| Sputum smear | 675 | |||
| Positive | 99 (59.3) | 269 (53.0) | 1 | |
| Negative | 68 (40.7) | 239 (47.0) | 0.77 (0.54-1.10) |
Only 51 (30.2%) of the patients with clustered isolates had epidemiological links, 13 (7.7%) in the definite category and 38 (22.5%) in the probable category.
This is the first population-based study in a place with an intermediate burden of tuberculosis and a low prevalence of HIV infection to determine the proportion of cases due to recent transmission of disease. We found that only 24.5% of the patients with active tuberculosis on the island of Hong Kong had mycobacterial isolates in their sputum that belonged to clusters. The estimated proportion of clustered samples was 24.5% by the n method and 15.3% with the n − 1 method. Even after adjustment for potential bias due to underestimation because of incomplete sampling, the proportion of clustered isolates in Hong Kong is considerably lower than that reported in both low-prevalence (1, 4, 5, 6, 7, 17) and high-prevalence countries (13, 19).
The factors associated with clustering reported in low-prevalence countries, such as younger age groups, alcohol and drug abuse, imprisonment, and HIV infection, were not important in Hong Kong. Those with isolates belonging to clusters were more likely to have resided locally for ≥7 years and had a history of travel to mainland China in the past 2 years. The significance of this finding is unclear.
Only 30.2% of the clustered cases in Hong Kong had identifiable epidemiological links, which is consistent with other reports in the literature. The low concordance between molecular and conventional epidemiological findings may be due to the low specificity of IS6110-based DNA fingerprinting, especially among isolates with fewer than five copies of IS6110. However, only 1.6% of the isolates in this study had five or fewer bands. The low frequency of isolates with five or fewer bands is not uncommon in some Asian countries (8, 12, 18, 21), although other Asian countries (8, 9, 16, 22) have reported a high proportion of isolates with a single copy of IS6110 (12 to 40%). It has also been postulated that molecular clustering may not be due to recent transmission, as identical pattern could arise independently for isolates of different origins because of convergence of banding patterns (8).
In summary, if the proportion of molecular clustering of isolates truly reflects recent transmission of disease, then about 15 to 20% of the active cases of tuberculosis in Hong Kong are due to recent transmission of the disease.
Acknowledgments
This work was supported by a grant from the Hong Kong Research Grant Council (HKU7272/00M).
We thank the nursing staff of the Hong Kong Government Tuberculosis and Chest Clinic for their hard work in contact investigation, Sandra Luen for data entry, and Anne Dy Buncio for data analysis.
REFERENCES
- 1.Alland, D., G. E. Kalkut, A. R. Moss, R. A. McAdam, J. A. Hahn, W. Bosworth, E. Drucker, and B. Bloom. 1994. Transmission of tuberculosis in New York City. An analysis by DNA fingerprinting and conventional epidemiologic methods. N. Engl. J. Med. 330:1710-1716. [DOI] [PubMed]
- 2.Anonymous. 1999. HIV/AIDS situation in Hong Kong. Report August 1998 to July 1999, p 3-4. Hong Kong Advisory Council on AIDS, Hong Kong, China.
- 3.Anonymous. 2001. Annual report. Hong Kong Government Tuberculosis and Chest Service, Hong Kong, China.
- 4.Barnes, P. F., Z. Yang, S. Preston-Marin, J. M. Pogoda, B. E. Jones, M. Otaya, D. Eisenach, L. Knowles, S. Harvey, and D. Cave. 1997. Patterns of tuberculosis transmission in central Los Angeles. JAMA 278:1159-1163. [PubMed] [Google Scholar]
- 5.Bishal, W. R., N. M. H. Graham, S. Harrington, D. S. Pope, N. Hooper, J. Astemborski, L. Sheely, D. Vlahov, G. Glass, and R. E. Chaisson. 1998. Molecular and geographic patterns of tuberculosis transmission after 15 years of directly observed therapy. JAMA 280:1679-1684. [DOI] [PubMed] [Google Scholar]
- 6.Braden, C. R., C. L. Templeton, M. D. Cave, S. Valway, I. N. Onorato, K. G. Castro, D. Moers, Z. H. Yang, W. W. Stead, and J. H. Bates. 1997. Interpretation of restriction fragment length polymorphism analysis of Mycobacterium tuberculosis isolates from a state with a large rural population. J. Infect. Dis. 175:1446-1452. [DOI] [PubMed] [Google Scholar]
- 7.Burman, W. J., R. R. Reves, A. P. Hawkes, C. A. Rietmeijer, Z. H. Yang, H. El-Hajj, J. H. Bates, and M. D. Cave. 1997. DNA fingerprinting with two probes decreases clustering of Mycobacterium tuberculosis. Am. J. Respir. Crit. Care Med. 155:1140-1146. [DOI] [PubMed] [Google Scholar]
- 8.Dale, J. W., R. M. Nor, S. Ramayah, T. H. Tang, and Z. F. Zainuddin. 1999. Molecular epidemiology of tuberculosis in Malaysia. J. Clin. Microbiol. 37:1265-1268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Das, S., C. N. Paramasivan, D. B. Lowrie, R. Prabhakar, and P. R. Narayanan. 1995. IS 6100 restriction fragment length polymorphism typing of clinical isolates of Mycobacterium tuberculosis from patients with pulmonary tuberculosis in Madras, South India. Tuber. Lung Dis. 76:550-554. [DOI] [PubMed] [Google Scholar]
- 10.Friedman, C. R., M. Y. Stoeckle, B. N. Kreiswirth, W. D. Johnson, Jr., S. M. Manoach, J. Berger, K. Sathianathan, A., Hafner, and L. W. Riley. 1995. Transmission of multi-drug resistant tuberculosis in a large urban setting. Am. J. Respir. Crit. Care Med. 152:335-359. [DOI] [PubMed] [Google Scholar]
- 11.Hermans, P. W. M., D. van Soolingen, J. W. Dale, A. R. J. Schuitema, R. A. McAdam, D. Catty, and J. D. A. van Emden. 1990. Insertion element IS986 from Mycobacterium tuberculosis: a useful tool for diagnosis and epidemiology of tuberculosis. J. Clin. Microbiol. 28:2051-2058. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Huh, Y. J., D. I. Ahn, and S. J. Kim. 1995. Limited variation of DNA fingerprints (IS6110 and IS1081) in Korean strains of Mycobacterium tuberculosis. Tuber. Lung Dis. 76:324-329. [DOI] [PubMed] [Google Scholar]
- 13.Lockman, S., J. D. Sheppard, C. R. Braden, M. J. Mwasekaga, C. L. Woodley, T. A. Kenyon, N. J. Binkin, M. Steinman, F. Montsho, M. Kesupile-Reed, C. Hirschfeldt, M. Notha, T. Moeti, and J. W. Tappero. 2001. Molecular and conventional epidemiology of Mycobacterium tuberculosis in Botswana: a population-based prospective study of 301 pulmonary tuberculosis patients. J. Clin. Microbiol. 39:1042-1047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Moss, A. R., D. Alland, E. Telzak, D. Hewlett, Jr., V. Sharp, P. Chiliade, V. LaBombardi, D. Kabus, B. Hanna, L. Palumbo, K. Brudney, A. Weltman, K. Stoeckle, K. Chirgwin, M., Simberkoff, S. Moghazeh, W. Eisner, M. Luffey, and B. Kreiswirth. 1997. A city-wide outbreak of a multiple-drug-resistant strain of Mycobacterium tuberculosis in New York. Int. J. Tuber. Lung Dis. 1:115-121. [PubMed] [Google Scholar]
- 15.Murray, M., and D. Alland. 2000. Methodological problems in the molecular epidemiology of tuberculosis. Am. J. Epidemiol. 155:565-571. [DOI] [PubMed] [Google Scholar]
- 16.Palittapongarnpim, P., P. Luangsook, C. Tansuphaswadikul, R. Chuchottawaorn, R. Prachaktam, and B. Sathapatayavong. 1997. Restriction fragment length polymorphism study of Mycobacterium tuberculosis in Thailand using IS6110 as probe. Int. J. Tuber. Lung Dis. 1:370-376. [PubMed] [Google Scholar]
- 17.Small, P. M., P. C. Hopewell, S. P. Singh, A. Paz, J. Parsonnet, D. C. Rushton, G. F. Schecter, C. L. Daley, and G. K. Schoolnik. 1994. The epidemiology of tuberculosis in San Francisco. A population-based study using conventional and molecular methods. N. Engl. J. Med. 330:1703-1709. [DOI] [PubMed] [Google Scholar]
- 18.Van Soolingen, D., L. S. Qian, P. E. W. De Haas, J. T. Douglas, H. Traore, F. Portaels, H. Z. Qing, D. Enkhsaikan, P. Nymadawa, and J. D. A. van Embden. 1995. Predominance of a single genotype of Mycobacterium tuberculosis in countries of East Asia. J. Clin. Microbiol. 33:3234-3238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wilkinson, D., M. Pillay, C. Lombard, G. R. Davies, and A. W. Sturm. 1997. Molecular epidemiology and transmission dynamics of Mycobacterium tuberculosis in rural Africa. Trop. Med. Int. Health 2:747-753. [DOI] [PubMed] [Google Scholar]
- 20.Yang, Z. H., P. E. W. de Haas, C. H. Wachmann, D. van Soolingen, J. D. A. van Embden, and A. B. Andersen. 1995. Molecular epidemiology of tuberculosis in Denmark in 1992. J. Clin. Microbiol. 33:2077-2081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yeh, R. W., P. De Leon, C. B. Agasino, J. A. Hahn, C. L. Daley, P. C. Hopewell, and P. M. Small. 1998. Stability of Mycobacterium tuberculosis DNA genotypes. J. Infect. Dis. 177:1107-1111. [DOI] [PubMed] [Google Scholar]
- 22.Yuen, L. K. B., B. C. Ross, K. M. Jackson, and B. Dwyer. 1993. Characterization of Mycobacterium tuberculosis strains from Vietnamese patients by Southern blot hybridization. J. Clin. Microbiol. 31:1615-1618. [DOI] [PMC free article] [PubMed] [Google Scholar]