Hypophosphatemia in Dogs With Presumptive Sepsis: A Retrospective Study (2008–2018)

Background: In humans with sepsis, hypophosphatemia is a marker of illness severity and a negative prognostic indicator. Hypophosphatemia has not been previously investigated in dogs with sepsis, however. This study aimed to estimate the prevalence of hypophosphatemia in dogs, the prevalence of presumptive sepsis in dogs with hypophosphatemia, the prevalence of hypophosphatemia in dogs with presumptive sepsis and the association between outcome and hypophosphatemia in dogs with presumptive sepsis. Methods: Electronic medical records of the Cornell University Hospital for Animals from 2008-2018 were queried to identify all dogs with hypophosphatemia and all dogs with presumptive sepsis. Hypophosphatemia was defined as a serum phosphate concentration <2.7 mg/dL. Sepsis was presumed where ≥2 of 4 systemic inflammatory response syndrome (SIRS) criteria were satisfied associated with a documented or highly suspected infection. Variables were assessed for normality using the D’Agostino-Pearson test. Continuous variables were compared between groups using the Mann-Whitney U test. Differences in frequency between categorical variables were analyzed using contingency tables, calculation of Fisher’s exact test or Chi2 and estimation of odds ratios. Results: In the study period, 47,992 phosphate concentration measurements from 23,752 unique dogs were identified. After eliminating repeat analyses, the period prevalence of hypophosphatemia on a per dog basis over the 11-year study period was 10.6% (2,515/23,752). The prevalence of presumptive sepsis within dogs with hypophosphatemia was 10.7% (268/2,515). During the 11-year study period, 4,406 dogs with an infection were identified, of which 1,233 were diagnosed with presumptive sepsis and had a contemporaneous phosphate concentration. Hypophosphatemia was more prevalent in dogs with presumptive sepsis than in dogs without 21.7% vs 10.2%; OR 2.44 [95% CI 2.12-2.81]; P<0.0001. The case mortality rate was greater in dogs with hypophosphatemia and presumptive sepsis than in dogs with hypophosphatemia without presumptive sepsis (15.3% vs 3.1%; OR 5.70 [95% CI 3.76-8.52]; P<0.0001), however hypophosphatemia was not associated with outcome in dogs with presumptive sepsis OR 0.87 [95% CI 0.60-1.26]; P=0.518. Conclusions: In dogs with hypophosphatemia, a presumed diagnosis of sepsis was associated with increased mortality compared to other associated disease processes. In dogs with presumptive sepsis, hypophosphatemia was not associated with outcome.