Impact of heat treatment on Dirofilaria immitis antigen detection in shelter dogs

Brian A. DiGangi, Carly Dworkin, Jason W. Stull, Jeanette O’Quin, Morgan Elser, Antoinette E. Marsh, Lesli Groshong, Wendy Wolfson, Brandy Duhon, Katie Broaddus, Elise N. Gingrich, Emily Swiniarski, Elizabeth A. Berliner

The diagnosis and management of canine heartworm disease is a growing concern for shelter veterinarians. Although the accuracy of commercial antigen test kits has been widely studied, recent reports have renewed interest in antigen blocking as a causative factor for false "no antigen detected" results. The objectives of this study were to determine the prevalence of false "no antigen detected" results in adult dogs entering shelters in northern, southern, and western regions of the country and to identify historical and clinical risk factors for such results. Serum samples were evaluated for Dirofilaria immitis antigen using a commercially available point-of-care ELISA; samples in which no antigen was detected underwent a heat treatment protocol and repeat antigen testing. Whole blood samples underwent Knott testing to identify the presence of microfilariae. Historical and clinical findings were analyzed using exact logistic regression. A total of 616 samples were analyzed. Overall prevalence of positive antigen test results (prior to heat treatment) was 7.3% and frequency of false "no antigen detected" results due to antigen blocking (ie, samples with no antigen detected prior to heat treatment and positive after heat treatment) was 5.2%. Among dogs that had no detectable antigen on the initial tests, dogs that had microfilariae detected via modified Knott testing (OR = 32.30, p-value = 0.013) and dogs that previously received a heartworm preventive (OR = 3.81, p-value = 0.016) had greater odds of antigen blocking than dogs without these factors. Among dogs that were heartworm positive, those without microfilariae detected had greater odds of antigen blocking than dogs with this factor (OR = 11.84, p-value = 0.0005). Geographic region of origin was significantly associated with occurrence of antigen blocking (p = 0.0036); however, blocking occurred in all regions sizably contributing to heartworm diagnoses. Of the 74 dogs found to be infected with heartworms in this study, 39.2% (29) had no detectable antigen prior to heat treatment. Heat treatment of serum samples should be considered to improve diagnostic test accuracy, particularly in dogs that reportedly received a heartworm preventive prior to antigen testing regardless of region of origin.