|
|
| |||||||||||||||||||||||||||||||||||||||||||||||
Background For decades it has been assumed that postprimary tuberculosis is usually caused by reactivation of endogenous infection rather than by a new, exogenous infection.
Methods We performed DNA fingerprinting with restriction-fragment–length polymorphism analysis on pairs of isolates of Mycobacterium tuberculosis from 16 compliant patients who had a relapse of pulmonary tuberculosis after curative treatment of postprimary tuberculosis. The patients lived in areas of South Africa where tuberculosis is endemic. Medical records were reviewed for clinical data.
Results For 12 of the 16 patients, the restriction-fragment–length polymorphism banding patterns for the isolates obtained after the relapse were different from those for the isolates from the initial tuberculous disease. This finding indicates that reinfection was the cause of the recurrence of tuberculosis after curative treatment. Two patients had reinfections with a multidrug-resistant strain. All 15 patients who were tested for the human immunodeficiency virus were seronegative.
Conclusions Exogenous reinfection appears to be a major cause of postprimary tuberculosis after a previous cure in an area with a high incidence of this disease. This finding emphasizes the importance of achieving cures and of preventing anyone with infectious tuberculosis from exposing others to the disease.
Before the introduction of antituberculous medication, there was little recognition of the distinction between endogenous reactivation and exogenous reinfection in patients who had multiple episodes of tuberculosis, since untreated established tuberculous lesions may be alternately active and dormant.3 Effective treatment regimens made possible the sterilization of pulmonary lesions, but it was accepted that subsequent episodes of tuberculosis were almost invariably caused by endogenous reactivation.4 The complete sterilization of a lesion became possible with improved treatment regimens, especially with the introduction of rifampin, a drug with a potent sterilizing effect. With short-course combination therapy consisting of isoniazid, rifampin, and pyrazinamide, the relapse rate dropped from 21 percent to 1 to 2 percent.5,6,7 In this era of effective treatment regimens, the notion that multiple episodes of tuberculosis in one patient are almost always caused by endogenous reactivation may be questioned. It is now possible to characterize the genotype of Mycobacterium tuberculosis by DNA fingerprinting, which can show whether a new episode of the disease is caused by infection with the same strain that caused a previous episode or by a different strain.
In this study we used DNA fingerprinting to determine the relative frequency of endogenous reactivation and exogenous reinfection in patients with multiple episodes of postprimary tuberculosis. We aimed to determine the importance of this distinction in terms of the definition of cure, the efficacy of current treatment regimens, and the control of tuberculosis.
Methods
All the patients described in this report were treated in two neighboring suburban communities in metropolitan Cape Town, South Africa. These communities have a geographic area of 2.42 km2 and a population of 34,294, of whom 99.7 percent are of mixed race. The number of reported cases of tuberculosis per year in these two communities is very high (1000 cases per 100,000 population per year).8 The birth rate is 29.3 per 1000 population, and the infant mortality rate is 38 per 1000 live births. In general, people live in poor socioeconomic conditions, although most live in houses with running water and electricity. Two primary care clinics and an adjacent tertiary care hospital serve the area.
Before 1996, when all the elements of the World Health Organization (WHO) strategy of directly observed short-course treatment (DOTS) were implemented, all patients were treated at one of the primary care clinics by directly observed therapy with three drugs (patients needing treatment for a first episode [new patients]) or four drugs (patients needing treatment for a subsequent episode [returning patients]). Until 1996, there was no systematic surveillance for treatment failure, although for most patients a sputum sample was examined (directly and by culture) between the fourth and sixth months. With its implementation in 1996, the WHO DOTS strategy required an accurate recording of cases in clinic registers, the administration of directly observed therapy (four drugs for new patients and five drugs for returning patients), and surveillance for treatment outcome, including treatment failure. Except for multidrug-resistant infections, there was no surveillance for relapse.
Patients
Patients included in this study had at least two episodes of postprimary pulmonary tuberculosis within the study period (between September 1992 and May 1998), with cultures from the two most recent successive episodes (referred to as first and second episodes) available for restriction-fragment–length polymorphism (RFLP) analysis and with cure as the outcome of the first episode. Extensive clinical histories of the included patients were obtained and included data on age, sex, medical history, status with regard to infection with the human immunodeficiency virus (HIV), findings on chest radiography, the results of sputum staining and cultures, drug sensitivities, treatment, and outcome. According to WHO criteria, cure was defined as the completion of a course of six to eight months of directly observed combination therapy (with isoniazid, rifampin, and pyrazinamide in a single tablet), compliance (attendance for the course of therapy, with at least 80 percent of prescribed doses taken), and a sputum culture positive for M. tuberculosis at diagnosis and at least one negative sputum culture at the end of treatment. All patients who needed treatment for a subsequent episode but who did not meet the criteria for previous cure were excluded from the study.
M. tuberculosis Isolates
Sputum samples were stained and cultured at the laboratory that routinely served the clinics. Drug-susceptibility testing was performed by the indirect-proportion method in accordance with the guidelines of the South African National Tuberculosis Program. In September 1992, in the Cape Town communities described above, we initiated a prospective study in which all cultures positive for M. tuberculosis from patients residing in these communities were genotyped by RFLP and a data base of the results was established. To evaluate the representativeness of the patients included in the RFLP data base, they were compared with patients who had positive cultures in the district tuberculosis register for the period from January 1996 through May 1998 with regard to the clinical classification. The percentage of cultures obtained for RFLP analysis was calculated for both new and returning patients. These data were also used to calculate the annual number of cases of culture-positive pulmonary tuberculosis per 100,000 population.
Isolates of M. tuberculosis were genotyped by RFLP according to an internationally standardized method,9 and the results were analyzed with GelCompar software (version 4.0, Applied Maths BVBA, Kortrijk, Belgium). For the RFLP data base, the extent of accumulated laboratory error was calculated by comparing successive DNA fingerprints for serial isolates collected during the first two months of treatment, according to the following formula: laboratory error rate equals the number of isolates with DNA fingerprints that failed to match the DNA fingerprints of the subsequent isolate, divided by the total number of serial isolates analyzed. To exclude the possibility of mixed infection, DNA was isolated from the entire culture, and all RFLP patterns were carefully analyzed to identify any background bands reflecting different strains. Analysis of the complete RFLP data base identified only five isolates with background bands. These isolates were excluded from further analysis.
The isolates included in this study were analyzed as follows. Genomic DNA was digested with PvuII or HinfI, electrophoretically fractionated, and transferred to Hybond-N+ membranes (Amersham, Buckinghamshire, United Kingdom). The PvuII digests were hybridized with an IS6110 3' probe (complementary to the IS6110 domain between nucleotides 631 and 875) that had been labeled by enhanced chemiluminescence. The HinfI digests were hybridized with a 32P-labeled MTB484(1) probe.10 The respective RFLP patterns were visualized by autoradiography. The IS6110 3' DNA fingerprints from each patient's isolates were compared with those in the complete RFLP data base (with data gathered from September 1992 through May 1998) by means of the clustering formula known as the "unweighted pair-group method using arithmetic averages" and the Dice coefficient to determine whether the isolates belonged to a cluster (suggesting infection by recent transmission) or were unique (suggesting reactivation of a latent infection) within the communities studied.
A patient whose isolates of M. tuberculosis from the first and second episodes of postprimary tuberculosis were identical on RFLP analysis with each DNA probe was considered to have tuberculosis due to reactivation of endogenous infection. A patient whose isolates from the first and second episodes of postprimary tuberculosis were different was considered to have tuberculosis due to a new, exogenous infection.
Results
The rate of reported cases of culture-positive pulmonary tuberculosis from January 1996 through May 1998 was estimated to be very high (225 cases per 100,000 population per year for new cases and 251 per 100,000 per year for all true incident cases) (Table 1).
|
The median age of these 16 patients was 35 years; 9 were women and 7 were men (Table 2). The median interval between cure and subsequent diagnosis (isolation of a subsequent culture-positive specimen) was 25.5 months. Fifteen patients (94 percent) were tested for HIV infection, and all 15 tested negative. None of the patients had a medical history of diabetes, end-stage renal disease, or cancer or had been treated with immunosuppressive drugs. Chest radiographs revealed evidence of cavitary disease in 11 patients during the first episode of tuberculosis and in 12 during the second episode of tuberculosis.
|
|
Of the four patients with tuberculosis due to reactivation, two were older than the median age and two younger. The interval from cure to reactivation was longer than the median interval in one patient and less than the median in the other three. All four patients were infected with a strain that belonged to a cluster circulating in the community during their respective disease-free intervals.
Discussion
Using DNA fingerprinting, we found evidence that exogenous reinfection can have a dominant role in the pathogenesis of postprimary tuberculosis in adults in an area with a high incidence of the disease.1,2,11 To our knowledge, there have been only two other reports of RFLP analysis of isolates from patients with repeated episodes of tuberculosis who were living in a high-incidence area.12,13 These studies suggested that under conditions of endemic disease, the rate of endogenous reactivation far exceeds the rate of exogenous reinfection. However, these reports do not include a definition of cure or detailed information on the patients' drug regimens, compliance, or immune status. Furthermore, 15 patients (34 percent)12 and 40 patients (49 percent)13 were excluded because only a single culture was positive and the possibility of laboratory contamination was raised. If they had not been excluded, 90 percent of these patients would have been classified as having disease due to exogenous reinfection. Their exclusion might therefore have biased the results toward the importance of endogenous reactivation.
In our study, the rate of recovery of cultures of M. tuberculosis for RFLP analysis was lower for patients needing retreatment after cure than for other patients. The reason for this difference is not clear; it was not the result of any deliberate policy. The results of the study must, however, be viewed with caution because of the possibility of some unintentional bias. Nevertheless, there is no reason to suppose that strains from cases due to reinfection would be preferentially recovered as compared with those from cases due to reactivation, a condition necessary for such a bias.
In our study, we did not exclude patients with only a single positive culture, for the following reasons. It has been shown that tuberculous lesions can yield an isolated positive culture.14 There was no evidence of false positive cultures: all the patients received a diagnosis of active tuberculosis for both episodes on the basis of positive sputum-test results, the findings on chest radiographs, and the clinical history. The complete RFLP data base was analyzed to assess the extent of laboratory contamination, and the rate of accumulated error was found to be only 3.4 percent (17 discordant results among 499 serial isolates). On the basis of this error rate, it is possible that results for 1 isolate (3.4 percent of 32 isolates) may have been erroneous. Therefore, we believe that it is unlikely that the classification of disease as due to exogenous reinfection reflects extensive laboratory error, such as error due to cross-contamination or mixed infection at the first episode. This argument is further supported by the observation that in 11 of the 12 cases of exogenous reinfection, the strain responsible for reinfection was identified as one of a cluster of strains present in the communities. Only one isolate was retrieved from a patient with reinfection for whom there was no matching isolate in the entire data base. This patient appears to have been reinfected outside the communities or by an undetected source within the communities.
All four patients who were found to have disease due to reactivation were infected with individual strains that belonged to a cluster circulating within the communities during these patients' respective disease-free intervals. The molecular definition of exogenous reinfection used in this study excludes the possibility of exogenous reinfection with the same strain. Therefore, it is possible that some of the patients considered to have reactivation actually had new, exogenous infection with the same strain. Our results may thus underestimate the extent of exogenous reinfection.
It has generally been believed that reinfection is more difficult to confirm than primary infection, because of the immune response to M. tuberculosis antigens that develops after primary infection. There are no empirical data on the changes in the level of immunity over time, but it is assumed that in immunocompetent persons, reinfection is rare during the first two to five years after a first infection.2 In our study, cases of reactivation tended to occur soon after the previous episode was cured, but many of the cases due to reinfection also occurred early after a previous cure. It has been shown in persons infected with HIV that reinfection can occur not only years after a previous infection (or episode of disease) but even during treatment for active tuberculosis.15 In our study, in which all the patients tested for HIV infection were negative, reinfection occurred as little as seven and eight months after a previous cure. These results suggest that in immunocompetent persons living in an area where tuberculosis is endemic, reinfection and progression to active disease may occur at any time after treatment has been discontinued.
In patients previously treated and cured, a subsequent episode would be expected to represent endogenous reactivation. We found that exogenous reinfection has a predominant role in this population of patients, who have multiple episodes of active tuberculosis. In areas with a high incidence of tuberculosis, exogenous reinfection might also be a cause of the first episode of postprimary tuberculosis, since the immunity that develops after primary infection followed by a period of latency cannot be expected to confer more protection against exogenous reinfection than the immunity that develops after an episode of active disease.
The controversy with regard to endogenous as opposed to exogenous pathogenesis of tuberculosis is of importance in the planning of clinical trials and national tuberculosis-control programs. If, in areas with a high incidence of the disease, postprimary episodes of pulmonary tuberculosis after previous cure result predominantly from exogenous reinfection, as indicated by our results, the effectiveness of a drug regimen cannot be evaluated on the basis of the relapse rate without the additional information provided by RFLP analysis of bacterial isolates. In the evaluation of national tuberculosis-control programs for areas in which the disease is endemic, RFLP analysis can prove the effectiveness of the currently used treatment regimens. "Cure" in a patient who later has another episode of tuberculosis is not necessarily an incorrect concept. The more likely possibility is that he or she has a new infection after the cure. The emphasis should thus be placed on achieving cure in patients and on prompt case detection to prevent reexposure to sources of active tuberculosis.
Supported by the Collen Research Foundation, Leuven, Belgium, and the Glaxo–Wellcome Action TB Initiative.
We are indebted to the technical, secretarial, and nursing staff of the tuberculosis research team; to Gian van der Spuy for assistance with computer analysis; to Dr. I. Toms, Health Services, Tygerberg, South Africa; and to Joske Trappeniers.
Source Information
From the Department of Pediatrics and Child Health (A.V.R., R.P.G., N.B.) and the Medical Research Council Center for Molecular and Cellular Biology and the Department of Medical Biochemistry (R.W., M.R., T.C.V., P.D.V.H.), University of Stellenbosch, Tygerberg, South Africa; the Department of Pediatrics, Catholic University Leuven, Leuven, Belgium (A.V.R.); and the International Union against Tuberculosis and Lung Disease, Paris (D.A.E.).
Address reprint requests to Dr. van Helden at the Department of Medical Biochemistry, University of Stellenbosch, P.O. Box 19063, Tygerberg 7505, South Africa, or at pvh{at}gerga.sun.ac.za.
References
| |||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Recurrent Tuberculosis Due to Exogenous Reinfection
Stead W. W., Bates J. H., de Boer A. S., van Soolingen D., Borgdorff M. W., van Rie A., Warren R., van Helden P. D.
Extract |
Full Text
N Engl J Med 2000;
342:1050-1051, Apr 6, 2000.
Correspondence
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
