RESUMEN
OBJECTIVE: To determine the risk factors for blood contamination during cerebrospinal fluid (CSF) collection in dogs. STUDY DESIGN AND METHODS: This is a prospective study of 170 CSF samples. Data collected included signalment of the patient, body condition score, site of CSF collection (cerebellomedullary cistern (CMC) or lumbar cistern (LC)), number of attempts, clinician expertise, final diagnosis, time of day, skull conformation and day of the week. Analysis of the CSF samples was then performed, and the presence of blood contamination (red blood cells >500/µl) was recorded. Logistic regression was used to quantify the association of potential risk factors of the procedure. Multivariate analysis was performed on the variables that were statistically significant. RESULTS: Of the 170 CSF samples, 53 per cent were collected from the CMC (n=90) and 47 per cent from the LC (n=80). Blood contamination was seen in 20 per cent (n=34) of the samples, 8.9 per cent (n=8) in CMC and 32.5 per cent (n=26) in LC samples. Increased odds of obtaining a contaminated CSF sample were associated with lower level of clinician expertise (odds ratio: 2.5; 95 per cent confidence interval: 0.9-6.7; P=0.046) and with LC versus CMC collection site (odds ratio: 8.1; 95 per cent confidence interval: 2.1-12.9; P=0.001). CLINICAL SIGNIFICANCE: There is increased likelihood of blood contamination when collecting CSF from the LC compared with the CMC site. Increased clinician experience reduced the risk of CSF blood contamination, but none of the other variables examined significantly influenced this.