We conducted a molecular epidemiologic study of tuberculosis (TB) transmission during the years 1996--98 on the Island of Montreal. By combining public health data on the 528 reported cases with IS6110 DNA fingerprints for 430, we detected an overall low frequency of transmission manifesting as secondary cases of active disease. We also identified an important sub-group of TB patients who harbour isolates with matching patterns. Depending on the matching criterion used we attributed between 6% (95% CI: 4, 9%) and 22% (95% CI: 18, 27%) of TB cases to recent transmission; the vast majority of active TB disease reflects infection acquired at an earlier time and/or a different place. However, Haitian-born TB patients yielded a disproportionately high frequency of isolates belonging to matching "clusters" (21%; 95% CI: 13, 32%), while, other foreign-born patients have disproportionately low numbers of clustered isolates (5%; 95% CI: 3, 9%). / The classical interpretation of such results is that there is more ongoing transmission within this immigrant sub-group. We explored an alternative hypothesis: that M. tuberculosis isolates from Haitian-born patients demonstrate reduced genetic diversity reflecting TB transmission patterns in their previously isolated country of origin---hence that a bacterial founder effect accounts for the higher frequency of matching fingerprints. Using a recently introduced measure of fingerprint similarity, genetic distance, we assessed the extent of pattern diversity. The median nearest genetic distance (NGD) was 130 months (inter-quartile range (IQR): 98--201 months) among the 47 distinct isolates from Haitian-born patients; among the non-Haitian foreign-born, the median NGD for the 191 distinct isolates was 128 months (IQR: 103--170 months). Hence the overall genetic heterogeneity of M. tuberculosis organisms among Haitian-born Montrealers was as great as that among a group of patients born in 70 other countries. Local transmission among the Haitian-born remains the most likely scenario. / We demonstrated that a continuous measure, such as genetic distance, may also permit researchers to address a challenge to the interpretation of M. tuberculosis molecular typing results: how to determine whether highly similar, non-identical fingerprint patterns in fact reflect underlying "matches." The distribution of NGD for isolates initially classified as identical (10--27 months), similar (15--108 months) and unique (40--244 months) suggested a possible cut-point of 40 months. Use of this cut-point labelled 19% of isolates as "clustered", suggesting that 14% of Montreal TB cases reflected transmission during the study period.
| Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.84278 |
| Date | January 2004 |
| Creators | Kulaga, Sophie |
| Contributors | Schwartzman, Kevin (advisor) |
| Publisher | McGill University |
| Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
| Language | English |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Format | application/pdf |
| Coverage | Doctor of Philosophy (Department of Epidemiology and Biostatistics.) |
| Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
| Relation | alephsysno: 002141322, proquestno: AAINQ98298, Theses scanned by UMI/ProQuest. |
Page generated in 0.0023 seconds