Severe Dengue Epidemic, Sri Lanka, 2017

Abstract

In 2017, a dengue epidemic of unexpected magnitude occurred in Sri Lanka. A total of 186,101 suspected cases and 440 dengue-related deaths occurred. We conducted a comprehensive analysis of this epidemic by comparing national surveillance data for 2017 with data from the preceding 5 years. In all Sri Lanka districts, dengue incidence in 2017 increased significantly over incidence during the previous 5 years. Older schoolchildren and young adults were more clinically symptomatic than those at extremes of age. Limited virologic surveillance showed the dominant circulating variant was dengue virus type 2 cosmopolitan genotype in the most affected district. One quarter of total annual cases were reported 5 weeks after the southwest monsoon started. Changes in vector abundance were not predictive of the increased incidence. Direct government expenditures on dengue control activities in 2017 were US $12.7 million. The lessons learned from this outbreak are useful for other tropical nations facing increasing dengue incidence.

Keywords: Dengue; Sri Lanka; age group; dengue virus; epidemic; mosquitoes; serotype; surveillance; vector-borne infections; viruses.

Figure 1

Comparison of the weekly mean attack rates of dengue reported in 2017 with the 5-year historical mean (2012–2016), Sri Lanka. Rates are cases per 100,000 population. Source: (7).

Figure 2

Comparison of dengue incidence rates per district in 2017 with the 5-year (2012–2016) average, Sri Lanka. A) Incidence rate in 2017. B) Historical mean incidence rate during 2012–2016. Incidence is cases per 100,000 population. Source: (7).

Figure 3

Dengue incidence rates by age group and sex, Sri Lanka, 2017. Incidence is cases per 100,000 population. Source (7).

Figure 4

Phylogenetic tree for 5 dengue virus 2 (DENV-2) isolates from late 2016 and 2017 dengue epidemic (arrows), Sri Lanka, and reference DENV-2 strains. The tree is based on a 1,485-nt fragment that encodes the envelope protein. Classification and naming of DENV-2 genotypes are based on (16). The evolutionary history was inferred using the neighbor-joining method (17). The optimal tree with the sum of branch length = 0.44012906 is shown. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the maximum composite likelihood method (18) and are in the units of the number of base substitutions per site. This analysis involved 33-nt sequences. All ambiguous positions were removed for each sequence pair (pairwise deletion option). The final dataset comprised 1,485 positions. Evolutionary analyses were conducted in MEGA X (12). Scale bar indicates nucleotide substitutions per site.

Figure 5

Time series plot showing weekly number of reported dengue cases (red line), ONI (orange line), PI (green line), and weekly cumulative rainfall, mm (blue line), Sri Lanka, 2012–2017. A) Colombo district. B) Jaffna district. ONI, Oceanic Niño Index; PI, Premise Index.