Postmortem evaluation of renal tubular vacuolization in critically ill dogs

OBJECTIVE: To describe the frequency of renal tubular vacuolization (RTV) as a surrogate of osmotic nephrosis and assess hyperosmolar agents as predictors of RTV severity. DESIGN: Retrospective study (February 2004–October 2014). SETTING: Veterinary teaching hospital. ANIMALS: Fifty‐three client‐owned, critically ill dogs that had a postmortem examination. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The frequency, severity, and location of RTV were determined in small group of critically ill dogs postmortem. Logistic regression was performed to assess cumulative 6% HES (670/0.75) and mannitol dose as predictors for RTV severity with presenting serum creatinine concentration, cumulative furosemide dose, and duration of hospitalization as covariates. RTV was noted in 45 (85%) of 53 critically ill dogs and was most commonly located to the medullary rays (68%). Cumulative 6% HES (670/0.75) dose (P = 0.009) and presenting serum creatinine concentration (P = 0.027) were significant predictors of RTV severity. For every 1 mL/kg increase in 6% HES (670/0.75) dose that a dog received, there was 1.6% increased chance of having more severe RTV (OR 1.016; 95% CI 1.004–1.029). In addition, for every 88.4 μmol/L (1 mg/dL) increase in presenting serum creatinine, there was a 22.7% increased chance of having more severe RTV (OR 1.227; 95% CI 1.023–1.472). Cumulative mannitol (P = 0.548) and furosemide (P = 0.136) doses were not significant predictors of RTV severity. CONCLUSION: In a small group of critically ill dogs, there was a high frequency of RTV identified on postmortem examination. Administration of 6% HES (670/0.75) and presenting serum creatinine concentration were significant predictors of RTV severity. Larger prospective studies are needed to determine the etiology and significance of RTV in dogs.