Intrasynovial triamcinolone treatment is not associated with incidence of acute laminitis

Abstract

Background Intrasynovial corticosteroid injections are commonly used in the treatment of equine orthopaedic disease, but corticosteroid administration is widely considered a risk factor for the development of laminitis. Despite a list of putative mechanisms and a number of case reports of steroid‐induced laminitis, no case‐control or cohort studies investigating the association between use of intrasynovial corticosteroids and acute laminitis have been published. Objectives To quantify the risk of laminitis posed by intrasynovial triamcinolone acetonide (TA) administration in a mixed population of horses. Study design Retrospective observational cohort study. Methods Clinical records of horses registered with one large UK equine practice were reviewed retrospectively to identify all horses receiving intrasynovial TA treatment between 1 January 2007 and 31 December 2017. A total of 1510 horses were selected and records investigated for incidence of laminitis over a 4‐month period following treatment. For each TA‐treated horse, an untreated horse, individually matched by age, sex, date of treatment and client type, was selected from the clinical records. Untreated horses were then investigated for laminitis over the same 4‐month period. Data were analysed in a 2 × 2 contingency table using Fisher's exact test. Results A total of 489 horses were lost to follow‐up and 55 horses were excluded, leaving 966 treated and matched, untreated horses. The incidence of laminitis over the 4‐month study period in both groups was identical: 3/966 horses (0.31%) (95% C.I. [0.08%, 0.91%]), equivalent to 0.93 cases per 100 horses per year (P > .9). Main limitations Retrospective study; large proportion (489/1510) of horses lost to follow‐up; large proportion of study population were racehorses; selection method resulted in disproportionate selection of horses born before 2013; similar incidence between groups may reflect existing risk‐based selection by clinicians. Conclusions intrasynovial triamcinolone acetonide administration does not increase the risk of laminitis in this study population.

Systematic review of the association between intrasynovial corticosteroid use and laminitis—What is the evidence?

“…In this study, controls presented same incidence of laminitis as treated horses. This was also a feature in the study of Haseler et al (2020) where incidence of laminitis in the control and in the corticosteroid‐treated groups was the same in a 4‐month follow‐up period (0.31% – 3/966). In this study, two horses in control group and one in the treated group had concurrent risk factors.…”

“…All of the included papers were observational studies. A retrospective observational (cohort) design was employed in two articles (Haseler et al, 2020; Welsh et al, 2016), and Potter et al (2019) performed one study as a retrospective cohort and one as a prospective investigation. One article was designed as a case series (McCluskey & Kavenagh, 2004).…”

“…After applying the MINORs tool, three studies were classified as high risk of bias (McCluskey & Kavenagh, 2004; Potter et al, 2019; Welsh et al, 2016), while one was classified as low risk of bias (Haseler et al, 2020; Figure 1). This was mainly due to their retrospective design, unavailable or unclear information about possible concurrent risk factors of horses included in the studies or unclear description of the employed follow‐up time or loss to follow‐up.…”

“…Only one article investigated the relationship between the incidence of laminitis with corticosteroid therapy administered via the intrasynovial route alone (Haseler et al, 2020). The other three studies encompassed several different routes of the administration of the therapy but also provided enough information regarding intrasynovial routes to allow their inclusion (McCluskey & Kavenagh, 2004; Potter et al, 2019; Welsh et al, 2016).…”

“…Despite this, the period for laminitis diagnosis after corticosteroid injection was not clear in many included articles. As for the ones including such information, great variation in the considered period was found (14 days to 4 months; Haseler et al, 2020; Potter et al, 2019).…”