Abstract
Anaplasma phagocytophilum, which causes granulocytic anaplasmosis, and Borrelia burgdorferi, the agent of Lyme borreliosis, are transmitted in multiple Holarctic regions by the same Ixodes sp. tick vectors and maintained in sylvatic cycles with the same rodent reservoirs. Coinfection of humans, domestic animals, wildlife, and ticks with both B. burgdorferi and A. phagocytophilum appears to be common, yet the pathologic mechanisms and ecology remain poorly understood compared with single-agent infection. We compiled available literature describing experimental and naturally occurring coinfection with B. burgdorferi and A. phagocytophilum in I. ricinus-complex ticks and host species, including wildlife, humans, livestock, and pets; calculated odds ratios for coinfection risks; evaluated the hypotheses that odds ratios for coinfection differ significantly between different tick life stages, ticks and vertebrate hosts, and reservoir and nonreservoir hosts; and finally, calculated the predicted coinfection prevalence based on the probability of each pathogen alone and compared this to the reported coinfection prevalence. Sixty-one manuscripts referenced coinfection or coexposure between these pathogens, with prevalence estimates for coinfection up to 67% in vertebrates and 28% in ticks. Of 61 different reports of coinfection or coexposure, 18 had odds ratios for coinfection significantly higher than 1. The combined odds ratios for coinfection in vertebrates and ticks were 1.9 and 1.28, respectively, with statistically significant variability among host and tick species in odds ratios. The overall predicted coexposure prevalence for vertebrates was 3.1% compared with an observed 8.9% prevalence. Odds ratios were not significantly different among classes of vertebrate hosts, species of ticks, stages of ticks, or ticks compared with hosts. We confirm a considerable level of coinfection and coexposure with these pathogens, although not predictably across host or tick species or geographical region.