Research Article: Clinical, Bacteriologic, and Geographic Stratification of Melioidosis Emerges from the Sri Lankan National Surveillance Program

Date Published: February 08, 2018

Publisher: The American Society of Tropical Medicine and Hygiene

Author(s): Harindra D. Sathkumara, Adam J. Merritt, Enoka M. Corea, Shivankari Krishnananthasivam, Mohan Natesan, Timothy J. J. Inglis, Aruna Dharshan De Silva.


Melioidosis, a potentially fatal tropical infection, is said to be underdiagnosed in low-income countries. An increase in melioidosis cases in Sri Lanka allowed us to analyze the relationship among clinical outcome, bacteriology, epidemiology, and geography in the first 108 laboratory-confirmed cases of melioidosis from a nationwide surveillance program. The additional 76 cases of laboratory-confirmed melioidosis confirmed further associations between Burkholderia pseudomallei multilocus sequence typing (MLST) and infection phenotype; ST1137/unifocal bacteremic infection (χ2 = 3.86, P < 0.05), ST1136/multifocal infection without bacteremia (χ2 = 15.8, P < 0.001), and ST1132/unifocal nonbacteremic infection (χ2 = 6.34, P = 0.02). ST1137 infections were predominantly seen in the Western Province, whereas ST1132, 1135, and 1136 infections predominated in the Northwestern Province. Early participating centers in the surveillance program had a lower melioidosis-associated mortality than later participants (χ2 = 3.99, P < 0.05). The based upon related sequence types (eBURST) algorithm, a MLST clustering method that infers founding genotypes and patterns of descent for related isolates and clonal complexes in an unrooted tree, showed uneven distribution of sequence types (STs). There was spatial clustering of the commonest STs (ST1132, 1136, and 1137) in the Western, Northwestern, and Central provinces. The recent increase in melioidosis in Sri Lanka uncovered by laboratory-enhanced surveillance is likely to be the result of a combination of improved laboratory detection, increased clinician awareness, recruitment of clinical centers, and small outbreaks. Further development of the surveillance program into a national genotyping-supported melioidosis registry will improve melioidosis diagnosis, treatment, and prevention where underdiagnosis and mortality rates remain high.

Partial Text

Melioidosis is a potentially fatal bacterial infection resulting from contaminated soil or water exposure in many parts of the tropics.1 In recent years, sporadic cases and occasional case clusters of melioidosis have been recognized in many places outside the main endemic region in Southeast Asia and northern Australia.2–4 These cases have long been thought to be a fraction of the actual case burden, and most likely reflect underdiagnosis because of the need for a combination of physician awareness, specific clinical laboratory expertise, and infrastructure.5 Key difficulties that hampers the early recognition of melioidosis are the broad range of clinical presentations, spanning subclinical seroconversion with delayed onset bacteremia, single or multifocal soft tissue infections, pyogenic infections of specific organs to pneumonia, and septicemia.6,7 Melioidosis is, therefore, a cluster of clinical presentations caused by a common pathogen, the select agent Burkholderia pseudomallei. Several disease classification schemes have been reported for melioidosis, each with different emphases and priorities.8–10 In our first report on a smaller series of culture-confirmed cases from Sri Lanka, we stratified cases on clinicopathologic grounds by whether focal infection was present in one or more organ systems and whether bacteremia was detected.10 Our classification schema reduced the wide range of clinical presentations to five generic categories and revealed an association between disease category and B. pseudomallei genotype in the unifocal bacteremic group. The recent nationwide increase in culture-confirmed melioidosis cases, coupled with prospective B. pseudomallei genotyping, presented an opportunity to investigate the phenotypic, genotypic, and geographic stratification of melioidosis in more detail. In the present study, we sought to develop a phylogeographic appreciation of disease emergence, understand its progression to different clinical outcomes, and measure the extent of disease class/genotype correlation.

The first culture-confirmed case of melioidosis in the current Sri Lankan series occurred in 2006. Since then, there has been an exponential growth in total and fatal culture-confirmed cases (Figure 1). Building on our initial B. pseudomallei collection, isolates from 76 culture-confirmed cases of melioidosis were identified between February 2014 and December 2015 (Supplemental Table 2). The MLST analysis revealed three new allele sequences: one gmhD (allele 124) and two ndh (alleles 20 and 58). Our collection of new strains (76 isolates) resolved to a total of 36 STs. Twenty-three novel STs were shared between 31 isolates and demonstrated greater diversity. Thirteen STs were already present in the B. pseudomallei MLST database, of which eight STs had been seen before in Sri Lanka.10 ST1137 remained the most common ST in Sri Lanka, representing 13 of 76 isolates.10 There were five shared STs (10 isolates), of which three were exclusively seen in the Southeast Asian region (ST308, ST655, and ST912). Five isolates belonged to ST594, which has been seen among clinical and environmental isolates in both Australia and Thailand. One of our isolates (BPs102) belonged to ST132, a dominant Australian ST seen among a large number of clinical, animal, and environmental isolates (N = 139 as of April 2016). However, this was found to carry the YLF gene cluster found predominantly among isolates of Southeast Asian origin.

The recent expansion in numbers of culture-confirmed cases follows almost a decade of raising clinical awareness and matching it with improved in-country laboratory capability.10 The increase in less than a year from a previous total of 37 to almost three times that number is remarkable. We recognize that this may reflect increased reach of the surveillance program as a result of external funding and reference laboratory support. The proportionate increase in fatal cases demonstrates that improved laboratory support does not always result in faster directed antimicrobial therapy, as observed in our earlier report. Case fatality rates decrease only when clinical awareness is raised above a threshold, where clinicians begin to suspect the infection early and start appropriate therapy while investigations are still in progress. In the early stages of sepsis, the diagnosis is entirely laboratory-based and often comes too late to institute direct therapy, especially when relying on positive blood cultures.




Leave a Reply

Your email address will not be published.