Skip directly to search Skip directly to A to Z list Skip directly to page options Skip directly to site content

Volume 7, Number 3—June 2001


Spoligotype Database of Mycobacterium tuberculosis: Biogeographic Distribution of Shared Types and Epidemiologic and Phylogenetic Perspectives

Christophe Sola*Comments to Author , Ingrid Filliol*, Maria Cristina Gutierrez†, Igor Mokrousov*, Véronique Vincent†, and Nalin Rastogi*Comments to Author 
Author affiliations: *Institut Pasteur de Guadeloupe, Pointe à Pitre, Guadeloupe; †Centre National de Référence des Mycobactéries, Institut Pasteur, Paris, France

Main Article

Table 1

Source of data for 3,319 spoligotypes of Mycobacterium tuberculosis used to generate the database of 259 shared types

No. of isolates Origina Year Reference
136 Denmark 1999 31
147 Italy 1999 32
157 Cuba 1998 33
1b Philippines 1997 34
3 Peru 1998 35
18 USA Unpublished R. Frothingham
105 France 1997 36
167 United Kingdom 1997 37
296 France Unpublished This study
28 Zimbabwe 1998 38
32 Guinea-Bissau 1999 25
118 The Netherlands 1997 7
68 Various countries 1999 15
58 France Unpublished J. Maïsetti & B. Carbonnelle
62 Russia Unpublished O. Narvskaya
84 West Africa 1999 39
5 Thailand Unpublished P. Palittapongarnpim
14 Romania 1997 40
17 Brazil 1999 41
5b Spain Unpublished S. Semper & C. Martin
1,283 USA 2000 12
1b United Kingdom 1999 42
19 The Netherlands 1998 43
1b The Netherlands 1999 19
69 Far East Asia 1995 44
69 Caribbean 1999 6
356 Caribbean Unpublished This study

aAlthough a potential sampling bias cannot be excluded, the sampling of isolates and their representativeness (in order of description) was as follows: Denmark, of 249 isolates described with a low copy number of IS6110 collected since 1992 (exhaustivity 93%), 24 shared types, representing 136 spoligotypes, were retained (9 other shared types, representing 49 isolates that were found exclusively in Denmark (S1,S2,S4,S19,S22,S23,S27,S30,S33), were not included in the present analysis; Italy, of 158 isolates from 156 patients in Verona collected during 1996-1997, 147 spoligotypes were retained; Cuba, of 160 isolates typed (obtained from a pool of 578 smear-positive sputa collected during 1994-1995), 157 spoligotypes described (exhaustivity 36%) were retained; Philippines, no data except for a single spoligotype available; Peru, of 29 strains isolated during 1995-1996 from the sputa of patients in Lima and Cuzco, only 3 were retained in this study since the remaining isolates shared spoligotypes with patients in Texas (12) and are included in the 1,283 Texan profiles; USA, 18 clinical isolates from the collection of R. Frothingham (representativeness unknown); France, 111 isolates from 105 hospitalized patients in Paris obtained during 1993 (patients were from three major hospitals that represented 5% of the total public hospital beds in Paris); United Kingdom, 167 isolates from all the culture-positive tuberculosis (TB) patients from three large hospitals in northwest London (without any indication of period of recruitment); France, 296 isolates sent for reference purposes during a 3-year period to the Centre National de Référence des Mycobactéries, Institut Pasteur, Paris; Zimbabwe, 28 spoligotypes obtained directly from sputum samples during a 1-month recruitment period (December 1995) of sputum-positive TB cases representing 20% of all cases; Guinea-Bissau, of 229 spoligotypes obtained from samples of 900 patients with suspected TB cases during 1989-1994, only 32 spoligotypes were fully described by the authors, and were retained for the analysis; the Netherlands, 118 isolates of unspecified representativeness from the collection of National Institute of Health (RIVM, Bilthoven); International multicenter study, 68 of 90 isolates from 38 countries representing the five continents; France, 58 isolates during a 1-year (1999) recruitment in the University Hospital of Angers; Russia, 62 isolates representing the St. Petersburg area collected during 1997-1999; West Africa, 84 isolates from Ivory Coast and around Dakar, Senegal, collected during 1994-1995; Thailand, 5 isolates from northern Thailand (unknown representativeness); Romania, 14 isolates of unknown representativeness; Brazil, 17 spoligotypes out of 91 isolates from a São Paulo hospital in 1995 (unknown representativeness); Spain, 5 multidrug-resistant isolates (unknown representativeness); USA, 1,429 clinical isolates from 1,283 patients during 1994-1999 that are part of an ongoing population-based study in Houston, Texas; United Kingdom, a single spoligotype from ancient DNA extracted from a bone sample; the Netherlands, 19 spoligotypes obtained from paraffin-wax embedded tissue samples previously collected during 1983-1993 (unknown representativeness); the Netherlands, a single spoligotype from a previous study (unknown representativeness); Far East Asia, 69 isolates from China and Mongolia obtained during 1992-1994 (unknown representativeness); Caribbean, 425 clinical isolates from a population-based ongoing study that includes all cultures isolated in Guadeloupe, Martinique, and French Guiana since 1994 and covers a 1 million population (exhaustivity 100%). Some isolates in this pool came from patients from other countries (essentially neighboring countries such as Haiti, Dominican Republic, Brazil, Commonwealth of Dominica, Barbados, and Surinam).
bDescription of a given spoligotype without precise number of isolates within this type.

Main Article

1For this purpose, the independent sampling sizes for Europe and the USA were taken as n1 and n2, the number of individuals within a given shared-type "x" was k1 and k2, and in this case, the representativeness of the two samples was p1=k1/n1 and P2=k2/n2, respectively. To assess if the divergence observed between p1 and p2 was due to sampling bias or the existence of two distinct populations, the percentage of individuals (p0) harboring shared-type "x" in the population studied was estimated by the equation p0= k1+k2/n1+n2=n1p1+n2p2/n1+n2. The distribution of the percentage of shared-type "x" in the sample sizes n1 and n2 follows a normal distribution with a mean p0 and a standard deviation of formula imageand formula imagerespectively, and the difference d=p1-p2 follows a normal distribution of mean p0-p0=0 and of variance σd2p12p22 = p0q0/n1+p0q0/n2 or σd2=p0q0 (1/n1+1/n2). The two samples being independent, the two variances were additive; the standard deviation σd= was calculated, and the homogeneity of the samples tested was assessed using the quotient d/σd=p1-p2/formula image. If the absolute value of the quotient d/σd<2, the two samples were considered to belong to a same population (CI 95%) and the variation observed in the distribution of isolates for given shared types could be due to a sampling bias. Inversely, if d/σd>2, then the differences observed in the distribution of isolates for given shared types were statistically significant and not due to potential sample bias.