Spatio-Temporal Distribution of Mycobacterium tuberculosis Complex Strains in Ghana

Abstract

Background: There is a perception that genomic differences in the species/lineages of the nine species making the Mycobacterium tuberculosis complex (MTBC) may affect the efficacy of distinct control tools in certain geographical areas. We therefore analyzed the prevalence and spatial distribution of MTBC species and lineages among isolates from pulmonary TB cases over an 8-year period, 2007-2014.

Methodology: Mycobacterial species isolated by culture from consecutively recruited pulmonary tuberculosis patients presenting at selected district/sub-district health facilities were confirmed as MTBC by IS6110 and rpoß PCR and further assigned lineages and sub lineages by spoligotyping and large sequence polymorphism PCR (RDs 4, 9, 12, 702, 711) assays. Patient characteristics, residency, and risks were obtained with a structured questionnaire. We used SaTScan and ArcMap analyses to identify significantly clustered MTBC lineages within selected districts and spatial display, respectively.

Results: Among 2,551 isolates, 2,019 (79.1%), 516 (20.2%) and 16 (0.6%) were identified as M. tuberculosis sensu stricto (MTBss), M. africanum (Maf), 15 M. bovis and 1 M. caprae, respectively. The proportions of MTBss and Maf were fairly constant within the study period. Maf spoligotypes were dominated by Spoligotype International Type (SIT) 331 (25.42%), SIT 326 (15.25%) and SIT 181 (14.12%). We found M. bovis to be significantly higher in Northern Ghana (1.9% of 212) than Southern Ghana (0.5% of 2339) (p = 0.020). Using the purely spatial and space-time analysis, seven significant MTBC lineage clusters (p< 0.05) were identified. Notable among the clusters were Ghana and Cameroon sub-lineages found to be associated with north and south, respectively.

Conclusion: This study demonstrated that overall, 79.1% of TB in Ghana is caused by MTBss and 20% by M. africanum. Unlike some West African Countries, we did not observe a decline of Maf prevalence in Ghana.

Fig 1. Map of sampling sites and study area.

Sputum samples were obtained from fifteen sampling sites (health facilities) all located within three regions in Ghana; Greater Accra, Central and Northern regions. During the period September 2012 –December 2014, samples were obtained mainly from all the 13 diagnostic centres within the Accra metropolis, (serving more than 46% of the Greater Accra region populace) and the two health facilities located in the northern region (Tamale Teaching Hospital and BMC Nalerigu). The ArcMap program in ArcGIS v. 10.0 was used to create the map.

Fig 2. Spatial distribution and prevalence of identified mycobacterium tuberculosis lineages.

Diagram shows the spatial distribution of (A) 2551 Mycobacterium tuberculosis complex (MTBC) strains; (B) 2535 human adapted MTBC; (C) regional prevalence of 1883 Lineage 4 sub lineages; (D) regional prevalence of 516 Mycobacterium africanum (Maf) isolates from the geographical regions served by the health facility where sampling was carried out in Ghana. Animal strains were found to be associated with the North (p = 0.0389). Similarly, Lineage 2 was found to be associated with the North (p = 0.0006). The most dominant Lineage 4 sub-lineage in the North is Ghana (p = 0.0000) whereas in the South is Cameroon, even though the association is not statistically significant. The unknown sub-type of Lineage 4 is associated with South (p = 0.0001).

Fig 3. Temporal distribution and prevalence of human adapted mycobacterium tuberculosis complex (MTBC).

Figure displays a stacked graph showing the temporal distribution of human adapted MTBC (left y-axis) and a linear graph showing the prevalence of Mycobacterium tuberculosis sensu stricto (MTBss) and Mycobacterium africanum (Maf) (right y-axis) over the entire 8-year study period.

Fig 4. Spatial distribution of human adapted MTBC lineages and major sub-lineages within the eight-year study period.

The figure shows the distribution of MTBC species/lineages/sub-lineages within the 13 districts where participants resided. The blue coloured panel shows the distribution of all the tuberculosis cases recruited with well-defined residential status. The red, brown and green coloured panels show the distribution of lineage 4, lineage 5 and lineage 6 respectively. All other sub-lineages/species have been indicated in the respective legends. This figure was created using the ArcMap program in ArcGIS v. 10.0. Abbreviations: MTBC, Mycobacterium tuberculosis complex; MTBss, Mycobacterium tuberculosis sensu stricto; Maf, Mycobacterium africanum; L4, Lineage 4; L5, Lineage 5; L6, Lineage 6; Ghana, Ghana genotypes (Ghana sub-lineage); Cam, Cameroon sub-lineage; MamE, Mamprusi East district; TamM, Tamale Metropolis; AgWM, Agona West Municipal; GomE, Gomoa East; AwuS, Awutu Senya; GaSM, Ga South Municipal; GaWM, Ga West Municipal; GaCM, Ga Central Municipal; GaEM, Ga East Municipal; AMA, Accra Metropolis; LaNM, La-Nkwantanang Madina Municipal; AdeM, Adenta Municipal; Kpes, Kpeshie Municipal.

Fig 5. Spatial distribution of isolated MTBC within selected districts (2012–2014).

This figure shows the; (A) Sum and case notification rate of all TB cases from September 2012 to December 2014, (B) Sum and normalized distribution of MTBss cases, (C) Sum and normalized distribution of Maf. The total number of cases per year was used as the denominator for normalization. Sampling from TamM did not meet our criteria for being included in analyses for case notification rate and so was excluded in all columns of panel A. Likewise we also recorded no cases in 2012 as such TamM was excluded from 2012 analysis (panel B and C). This figure was created using the ArcMap program in ArcGIS v. 10.0. Abbreviations: MTBss, Mycobacterium tuberculosis sensu stricto; Maf, Mycobacterium africanum; MamE, Mamprusi East district; TamM, Tamale Metropolis; AshK, Ashiedu Keteke; Ayaw, Ayawaso; Able, Ablekuma; OsuK, Osu Klottey; Okai, Okaikoi; Kpes, Kpeshie.

Fig 6. Spatial distribution of lineage 4 and major lineage 4 sub-lineages within selected districts (2012–2014).

The figure shows the (A) Sum and normalized distribution of lineage 4 cases, (B) Sum and normalized distribution of Ghana sub-lineages cases, (C) Sum and normalized distribution of Cameroon sub-lineage cases. The total number of cases per year was used as the denominator for normalization. This study recorded no TB cases for TamM in 2012, consequently TamM was excluded from all analysis carried out using 2012 data. This figure was created using the ArcMap program in ArcGIS v. 10.0. Abbreviations: L4, Lineage 4; Ghana, Ghana genotypes (Ghana sub-lineage); Cam, Cameroon sub-lineage; MamE, Mamprusi East district; TamM, Tamale Metropolis; AshK, Ashiedu Keteke; Ayaw, Ayawaso; Able, Ablekuma; OsuK, Osu Klottey; Okai, Okaikoi; Kpes, Kpeshie.