| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 5797664 | The Veterinary Journal | 2015 | 6 Pages |
â¢A retrospective analysis of 410 declarations of lightning related fatalities (LRF) in livestock is presented.â¢52.9% of the declarations were not compatible with fatal lightning stroke.â¢Pathognomonic singe lesions were present in only 43.3% of the confirmed cases.â¢Combining the veterinary expert investigation with lightning detection data resulted in the most correct LRF diagnosis.
Although lightning strike is an important cause of sudden death in livestock on pasture and among the main reasons why insurance companies consult an expert veterinarian, scientific information on this subject is limited. The aim of the present study was to provide objective information on the circumstantial evidence and pathological findings in lightning related fatalities (LRF), based on a retrospective analysis of 410 declarations, examined by a single expert veterinarian in Flanders, Belgium, from 1998 to 2012. Predictive logistic models for compatibility with LRF were constructed based on anamnestic, environmental and pathological factors. In addition, the added value of lightning location data (LLD) was evaluated. Pathognomonic singe lesions were present in 84/194 (43%) confirmed reports. Factors which remained significantly associated with LRF in the multivariable model were age, presence of a tree or open water in the near surroundings, tympany and presence of feed in the oral cavity at the time of investigation. This basic model had a sensitivity (Se) of 53.8% and a specificity (Sp) of 88.2%. Relying only on LLD to confirm LRF in livestock resulted in a high Se (91.3%), but a low Sp (41.2%), leading to a high probability that a negative case would be wrongly accepted as an LRF. The best results were obtained when combining the model based on the veterinary expert investigation (circumstantial evidence and pathological findings), together with the detection of cloud-to-ground (CG) lightning at the time and location of death (Se 89.1%; Sp 66.7%).
