There is no evidence that the addition of antimicrobials reduce the risk of sepsis after intra- articular corticosteroids in horses with arthritis

PICO question 
In horses with arthritis, does the treatment with intra-articular antimicrobials concurrently with intra-articular corticosteroids reduce the risk of sepsis compared to intra-articular corticosteroids alone? 
Clinical bottom line 
The three studies identified did not demonstrate a reduction of risk when antimicrobials were used. However, the strength of evidence provided by the studies was weak. The power of the studies to detect an effect of antimicrobials was low due to the small number of sepsis cases recorded. Further studies are therefore required to draw conclusions. 


Study design: Retrospective cohort study

PICO question
In horses with arthritis, does the treatment with intra-articular antimicrobials concurrently with intraarticular corticosteroids reduce the risk of sepsis compared to intra-articular corticosteroids alone?

Clinical bottom line
The three studies identified did not demonstrate a reduction of risk when antimicrobials were used. However, the strength of evidence provided by the studies was weak. The power of the studies to detect an effect of antimicrobials was low due to the small number of sepsis cases recorded. Further studies are therefore required to draw conclusions. In one of the sepsis cases there was a combination of triamcinolone acetonide with amikacin, one sepsis case was triamcinolone without amikacin, and two sepsis cases was polysulphated glycosaminoglycan. Hyaluronate was used in all four of the sepsis cases.
Limitations: Due to the low incidence of cases statistical analysis was not performed. Non-random treatment allocation.

Gillespie et al. (2016)
Population: Equine veterinarians The number of septic joints following intra-articular injection was 67 joints out of 319,760 intra-articular injections, giving an incidence of 2.1 septic joints per 10,000 intra-articular injections. Intra-articular antimicrobial usage as an adjunct for all intra-articular injections was used by 46.5% of veterinarians (112/241) and additional 39.4% of veterinarians (95/241) used antimicrobials some of the time. The data from the 64 veterinarians providing data from medical records did not show a reduction in joint sepsis associated with the use of antimicrobial administration.

Limitations:
There is no way of determining validity.
Data obtained from surveys may be inexact. Septic arthritis was diagnosed if bacterial culture of synovial fluid was positive or if synovial fluid analysis was consistent with sepsis. Assessment potential risk factors septic arthritis following intra-articular medication:  using corticosteroids  repeated joint injections  using amikacin sulphate with the intra-articular medication

Main findings: (relevant to PICO question):
Septic arthritis was diagnosed following intra-articular medication in 13 joints from 13 horses of the 16,624 injections -a risk of 7.8 per 10,000 joints injected. Statistically significant risk factors:  veterinarian  type of corticosteroid Septic arthritis was diagnosed in 12 of the 15,934 joints injected with a corticosteroid (risk of 7.5 per 10,000 injections, 95% CI 3.9-13.1). Betamethasone injection had a lower risk of septic arthritis than dexamethasone (P=0.024). None of the 824 joints in which amikacin sulphate was injected developed septic arthritis (risk of 0.0 per 10,000 injections, 95% CI 0.0-44.7), but 13 of the 15,800 joints injected without amikacin sulphate or any antimicrobial did develop sepsis (risk of 8.2 per 10,000 injections, 95% CI 4.4-14.1). However, this was not statistically significant.

Appraisal, application and reflection
Intra-articular injections of corticosteroids are employed in horse practice for the treatment of noninflammatory synovitis and osteoarthritis. The risk of developing iatrogenic septic arthritis after intra-articular injections is well known. The aim of this knowledge summary was to critically appraise published evidence where the addition of intra-articular antimicrobials is to reduce the risk of sepsis. The search strategy did not include the terms corticosteroid, antibiotic, and their derivatives, since specific drug names may have been used in publications. The search gave a lot of results, as it was not very specific, but it makes sure that relevant publications were not missed. However, only two publications related to the PICO question were found and one publication (Smith et al. 2018) was mentioned by a reviewer but was not picked up in the searches due to search query restrictions. One paper was a cross-sectional study and the other two were retrospective cohort studies, one of which includes a case-control design. None of these publications respond directly to the PICO question, but the authors looked at the result of using antimicrobials in addition to intra-articular injections of corticosteroids. In general, the risk of sepsis after intra-articular injections is very low -7.8 cases per 10,000 injections (Steel et al. 2013)  . These observations should be interpreted with caution -due to the low incidence of sepsis after intra-articular injections, the statistical power of the studies may not be sufficient to reflect true results. Thus, the use of antimicrobials is a very common practice -46% (Gillespie et al. 2016) to 93% (Smith et al. 2018) of veterinarians combine intra-articular corticosteroids with intra-articular antimicrobials to reduce the risk of septic inflammation of the joint, but there is no evidence of efficiency of this approach and this treatment is pure empirical. Prospective randomised controlled trials using standardised treatment protocols could answer this clinical question. Smith at al. (2018) estimated that approximately 12,500 medication sessions would be needed to provide a power of 80% for relative risk estimation. But such clinical trials are unlikely to be of practical value given the low incidence of sepsis.