Knemidokoptic mange was first observed on two Hawai‘i ‘Amakihi (Hemignathus virens) mist netted in Manuka Natural Area Reserve (NAR) on the Island of Hawai‘i in June 2007. Microscopic examination of skin scrapings from lesions of the infested individuals revealed the scaley-leg mite, Knemidokoptes jamaicensis. Continued surveillance at Manuka NAR (2007-2009) documented a 24% (15/63) prevalence of mange among Hawai‘i ‘Amakihi distributed from coastal habitat to 1,500 m above sea level (asl). From 2012-2014, we conducted an island-wide survey of wild passerine birds from several leeward sites (Manuka NAR, Kahuku Unit of Hawai‘i Volcanoes National Park (HAVO), Pu‘u Wa‘awa‘a Forest Bird Sanctuary, and Kipahoehoe NAR) and windward sites (Hakalau Forest National Wildlife Refuge, ‘Ᾱinahou Ranch of HAVO, Malama Ki Forest Reserve, and Keauohana Forest Reserve) to determine the current distribution and host range of knemidokoptic mange. We also determined the prevalence of malaria in Hawai‘i ‘Amakihi populations where mange was present and treated a subset of infested Hawai‘i ‘Amakihi mange with a single, topical dose of moxidectin. We mist netted and examined a total of 1,734 passerines, including 738 Hawai‘i ‘Amakihi. Mange was present in Hawai‘i ‘Amakihi at Manuka NAR (595 and 305 m asl), Kahuku Ranch Unit of HAVO (Glover site: 1,201 m asl and Kipuka Akala site: 1,532 m asl), Malama Ki Forest Reserve and Keauohana Forest Reserve (293 m asl). No other passerine birds (n = 995) were infected. Mange prevalence ranged from a high of 69% (40/58) in Keauohana Forest Reserve to a low of 2% (1/65) in the Kahuku Ranch Unit of HAVO (Kipuka Akala). At Manuka NAR prevalence had decreased from 26% in 2010 to 10% (7/81) in 2012–2014. We found no significant relationship between the prevalence of mange and the prevalence of avian malaria in mesic habitats at Manuka NAR (P = 0.59 (FET, n = 81)), but there was a significant association between the prevalence of mange and the prevalence of malaria in lowland wet forests in Puna Forest Reserves (P < 0.01 (FET, n = 72)). This apparent association may be a reflection of the high prevalence of malaria (>80%) in these areas. There was no difference in the frequency of recapture of birds that were infested versus un-infested at first capture at our long-term sites (Manuka NAR and Puna sites) (χ2(1, n = 227) = 1.51, P = 0.22, but when all sites with mange present were pooled, there was a significant difference in the frequency of recaptures between infested and un-infested birds (χ2(1, n = 424) = 7.13, P = 0.01). There was a significant association between parasitemia level (per 10,000 RBCs) and the ranked stage of mange present in infested individuals. We treated 24 Hawai‘i ‘Amakihi with moxidectin and upon recapture (n = 2), found a reduction in both the size and stage of mange lesions, such that a single dose, topical treatment of moxidectin appears to be an effective treatment for knemidokoptic mange in wild populations. Our results suggest that knemidokoptic mange is currently limited to Hawai‘i ‘Amakihi and prevalent in low elevation sites on both the windward and leeward sides of the island.
