Characterization of tungiasis infection and morbidity using thermography in Kenya revealed higher disease burden during COVID-19 school closures

Abstract

Abstract Background Tungiasis is a neglected tropical skin disease caused by the sand flea Tunga penetrans. Female fleas penetrate the skin, particularly at the feet, and cause severe inflammation. This study aimed to characterize disease burden in two highly affected regions in Kenya, to test the use of thermography to detect tungiasis-associated inflammation and to create a new two-level classification of disease severity suitable for mapping, targeting, and monitoring interventions. Methods From February 2020 to April 2021, 3532 pupils age 8–14 years were quasi-randomly selected in 35 public primary schools and examined for tungiasis and associated symptoms. Of the infected pupils, 266 were quasi-randomly selected and their households visited, where an additional 1138 family members were examined. Inflammation was assessed using infra-red thermography. A Clinical score was created combining the number of locations on the feet with acute and chronic symptoms and infra-red hotspots. Results The overall prevalence of tungiasis among all the school pupils who were randomly selected during survey rounds 1 and 3 was 9.3% [95% confidence interval (CI): 8.4–10.3]. Based on mixed effects logistic models, the odds of infection with tungiasis among school pupils was three times higher in Kwale (coastal Kenya) than in Siaya [western Kenya; adjusted odds ratio (aOR) = 0.36, 95% CI: 0.18–0.74]; three times higher in males than in females (aOR = 3.0, 95% CI: 2.32–3.91) and three times lower among pupils sleeping in a house with a concrete floor (aOR = 0.32, 95% CI: 0.24–0.44). The odds of finding an infected person among the household population during surveys before the COVID-19 pandemic was a third (aOR = 0.32, 95% CI: 0.19–0.53) of that when schools were closed due to COVID-19 restrictions and approximately half (aOR = 0.44, 95% CI: 0.29–0.68) in surveys done after school re-opening (round 3). Infection intensity was positively correlated with inflammation as measured by thermography (Spearman’s rho = 0.68, P < 0.001) and with the clinical score (rho = 0.86, P < 0.001). Based on the two-level classification, severe cases were associated with a threefold higher level of pain (OR = 2.99, 95% CI: 2.02–4.43) and itching (OR = 3.31, 95% CI: 2.24–4.89) than mild cases. Conclusions Thermography was a valuable addition for assessing morbidity and the proposed two-level classification of disease severity clearly separated patients with mild and severe impacts. The burden of tungiasis was considerably higher in households surveyed during COVID-19 restrictions suggesting underlying risks are found in the home environment more than in school.