Efficacy of a mixture of neem seed oil (Azadirachta indica) and coconut oil (Cocos nucifera) for topical treatment of tungiasis. A randomized controlled, proof-of-principle study

Abstract

Background: Tungiasis is a neglected tropical skin disease caused by the female sand flea (Tunga penetrans), which burrows into the skin causing intense pain, itching and debilitation. People in endemic countries do not have access to an effective and safe home treatment. The aim of this study was to determine the efficacy of a traditionally used and readily available mixture of neem and coconut oil for treatment of tungiasis in coastal Kenya.

Methodology: Ninety-six children aged 6-14 years with at least one embedded viable flea were randomized to be treated with either a mixture of 20% neem (Azadirachta indica) seed oil in coconut oil (NC), or with a 0.05% potassium permanganate (KMnO4) foot bath. Up to two viable fleas were selected for each participant and monitored for 6 days after first treatment using a digital microscope for signs of viability and abnormal development. Acute pathology was assessed on all areas of the feet using a previously established score. Children reported pain levels and itching on a visual scale.

Results: The NC was not more effective in killing embedded sand fleas within 7 days than the current standard with KMnO4, killing on average 40% of the embedded sand fleas six days after the initial treatment. However, the NC was superior with respect to the secondary outcomes of abnormal development and reduced pathology. There was a higher odds that fleas rapidly aged in response to NC compared to KMnO4 (OR 3.4, 95% CI: 1.22-9.49, p = 0.019). NC also reduced acute pathology (p<0.005), and there was a higher odds of children being pain free (OR 3.5, p = 0.001) when treated with NC.

Conclusions: Whilst NC did not kill more fleas than KMnO4 within 7 days, secondary outcomes were better and suggest that a higher impact might have been observed at a longer observation period. Further trials are warranted to assess optimal mixtures and dosages.

Trial registration: The study was approved by the Kenya Medical Research Institute (KEMRI) Scientific and Ethical Review Unit (SERU), Nairobi (Non-SSC Protocol No. 514, 1st April 2016) and approved by and registered with the Pharmacy and Poisons Board's Expert Committee on Clinical Trials PPB/ECCT/16/05/03/2016(94), the authority mandated, by Cap 244 Laws of Kenya, to regulate clinical trials in the country. The trial was also registered with the Pan African Clinical Trial Registry (PACTR201901905832601).

Fig 1

A. Fortaleza stages of embedded Tunga penetrans (stage 3, 4b, 5) and typical pathology (hyperkeratosis, desquamation, fissure, micro-abscess). B. Old and new manipulated lesions.

Fig 2. Viability signs of Tunga penetrans.

A. Stage 3b flea expelling an egg, B. Stage 3a flea excreting a dark red/black faecal thread, C. Stage 4a flea excreting faecal liquid. Photographs taken with the digital handheld video microscope with 200x magnification.

Fig 3. Flow chart of patients and observed sand fleas through the trial.

NC = neem-coconut-oil mixture; KMnO₄ = potassium permanganate foot bath.

Fig 4. Changes in appearance of embedded fleas 7 days after treatment.

Photographs taken with the digital handheld video microscope with 200x magnification. (A). Viable, embedded sand flea at stage 3a with characteristic convex surface, expelling a faecal thread on Day 1 (baseline). (B). The same flea as in A but non-viable at day 7: loss of abdominal cone structure, surface now concave, desiccated appearance, Stage 4b.(C). Viable, embedded flea stage 3a with convex surface, expelling a faecal thread on day 1 (baseline). (D). Rapidly Aged: same flea as in C, still viable on day 7 with clear abdominal cone, but now stage 4a, desiccated appearance, concave surface, brown colour.

Fig 5. Median severity score and interquartile range for acute tungiasis (SSAT) at baseline and day 7 follow up in patients treated with NC and KMnO₄.

Fig 6. Effect of KMnO₄ and NC treatment on tungiasis-associated pain and itching at day 7 follow up.

Bars show modelled mean percentages including 95% confidence intervals. Statistical significance tested with binary logistic regression testing association between treatment and reduction/increase in pain/itching.

Fig 7. Percentage of patients being pain-free and itch-free in the KMnO₄ and NC treatment groups at baseline and at day 7 follow up.

Bars show modelled means including 95% confidence intervals.