IL-1beta in provoked blood cells failed to flag gastric ulcers in 77 horses

The paper's title calls the method an ex vivo toll-like receptor agonist response, and that is a capacity test ^[3]. Blood is drawn, the cells in it are provoked with agonists, and the readout is how much interleukin-1 beta those cells make in a tube ^[2]. Interleukin-1 beta was the target because it has been linked to pain and inflammation in other species ^[4]. Whether that link carries over to the horse was part of what was on trial.

Thirteen of the 77 sampled horses do not appear in the final analysis ^[18]. Of the 64 that do, 59 had gastric lesions, so the comparison group without lesions was five animals ^[6][19]. Five on one side can exclude a large difference between groups and not much else. The narrow reading of this null is that there is no systemic interleukin-1 beta signal big enough to show up against a five-animal reference ^[5].

The ulcer-free group was small for a reason: controls are the scarce sample in this population. Gastric ulcers affect a large proportion of horses, and previous studies have reported gastric disease in 80% to 100% of racehorses for one major form of the disease ^[17][9].

The quarter of horses that responded at all is a separate problem from the group comparison. Counted from the 64 analysed, that is about 16 responders and 48 non-responders ^[20]. An assay that reads flat in three horses out of four will not grade pain in an individual animal even if the group difference had reached significance.

One comparison in the study does not lean on the five controls. The researchers split the horses by whether they responded to the stimulation and compared facial pain scores between those two groups, and found no significant difference ^[8]. That test had roughly 16 animals against 48 ^[20].

Gustavo Ferlini Agne, of Adelaide University's School of Animal and Veterinary Sciences ^[21], described the clinical problem in terms of the patient. "Horses can't tell us when they are in pain, and signs of discomfort can be subtle, inconsistent or easily overlooked," Agne said ^[13]. He also said the negative result has value of its own. "This is an important step because knowing what doesn't work is just as valuable as identifying a promising test," he said ^[15].

The researchers are explicit that none of this shows gastric ulcers are painless ^[10]. Their explanation for the null is anatomical: the immune response measured in circulating blood cells may not be sensitive enough to detect the effects of the ulcers, particularly if the inflammation or pain is localized to the stomach ^[11]. Agne said larger studies and horse-specific testing methods will be needed ^[12]. "The next challenge is to develop approaches that are better suited to horses and combine biological measures with careful assessment of pain-related behavior and inflammation," he said ^[23].