C-reactive protein in canine babesiosis caused by Babesia rossi and its association with outcome

Abstract

Acute phase proteins (APP) are ideal biomarkers for inflammation due to their stability, relative ease of assay and apparent relation between their concentration and the extent of the insult to tissue. C-reactive protein (CRP) is a positive major APP in dogs and can be used as a predictive marker for risk of disease and to monitor the response to treatment. Increased concentrations in certain diseases are associated with poor outcome. This cross-sectional, observational study of 75 dogs naturally infected with Babesia rossi, a cause of virulent canine babesiosis, was designed to examine the association of CRP concentration at admission and the magnitude of CRP change 24 hours after admission with outcome. Dogs were excluded if there was evidence of concurrent inflammatory diseases at the time of admission, infection with subtypes other than B. rossi, concurrent Ehrlichia canis infections or euthanasia for reasons other than poor prognosis. Diagnosis was confirmed by polymerase chain reaction and reverse line blot. CRP concentrations were determined by an automated human CRP Turbidometric Immunoassay (TIA), previously validated for use in dogs (Bayer CRP TIA, Newbury, UK), on serum samples collected by jugular venipuncture on admission, prior to any therapy, and thereafter daily until discharge or death. There was no significant difference in admission CRP concentration between survivors (n = 57; median = 97.4 mg/l; mean ± SD = 107.5 ± 49.5), and non-survivors (n = 11; median = 101.4 mg/l; mean ± SD = 122.1 ± 64.6) (p = 0.39). After elimination of non-significant predictors, a multiple exact logistic regression model for predicting mortality contained glucose and CRP. Mortality was associated with decreased glucose levels (p = 0.0002) and increased CRP levels (p = 0.045) on admission. Multiple regression analysis failed to show a significant relationship between admission CRP concentration and number of days of hospitalization in the survivors, adjusting for age and sex (p = 0.65). No significance was found in the relationship between the magnitude of change in CRP concentration 24 hours after admission, and the number of days of hospitalization in survivors, (p = 0.34). Using an admission CRP concentration cut-off of 60 mg/l, survival proportions between the two groups were no different (p = 0.34) and when applied to the group of dogs that survived, it was not associated with length of hospitalization (p = 0.25). In corroboration with previous reports glucose was identified as a major prognostic marker for mortality, but additionally the pro-inflammatory marker CRP was identified as a significant co-prognosticator. Copyright