PTH and FGF-23 are hormones central to calcium and phosphate regulation. Aberrations in calcium and phosphate homeostasis are known to occur in hyperthyroidism, with hyperthyroid cats reported to have elevated plasma PTH concentrations and suppressed plasma FGF-23 concentrations. Elevated PTH and FGF-23 concentrations are independent predictors of survival time in human haemodialysis patients, and therefore assessment of their prognostic significance in hyperthyroid cats is warranted. This study investigated if PTH and FGF-23 were independently associated with survival time (all cause mortality) in hyperthyroid cats.

Cats diagnosed with hyperthyroidism at two London-based first opinion practices between 1999 and 2009 were included in the study. Plasma PTH and FGF-23 concentrations (measured by validated IRMA and ELISA respectively) at the time of diagnosis of hyperthyroidism were entered into a univariable Cox regression model of survival time. FGF-23 concentrations were divided by 100 to allow easier interpretation of hazard ratios (HR). Two multivariable survival models were constructed, the first model included age and baseline plasma concentrations of creatinine, total calcium, phosphate, PTH and FGF-23, and the second model also included urine specific gravity (USG), presence of concurrent hypertension, packed cell volume (PCV) and urine protein:creatinine ratio (UPC). For each model, the variables were included in a backward, stepwise multivariable Cox regression analyses for survival.

Univariable Cox regression analysis indicated that both plasma PTH (HR = 1.006, 95% CI for HR = 1.002–1.009, n = 198; P = 0.001) and FGF-23 concentrations (HR for FGF-23 concentration/100 = 1.026, 95% CI for HR = 1.016–1.036, n = 203; P < 0.001) were negatively associated with survival time. In the first multivariable model (n = 152), survival was negatively associated with both age (P < 0.001) and plasma FGF-23 concentration (P < 0.001), but PTH was not independently associated with survival time. In the second multivariable model (n = 93), survival was positively associated with PCV (P = 0.006), and negatively associated with the presence of hypertension at the time of diagnosis of hyperthyroidism (P = 0.039) and UPC (P = 0.041), however in this model, plasma FGF-23 and PTH concentrations were not independently associated with survival time. If PCV was excluded from the second model, USG was positively associated with survival (P = 0.017), however age (P = 0.017), UPC (P = 0.047) and plasma FGF-23 concentration (P = 0.047) were negatively associated with survival time.

Plasma FGF-23 concentrations may have prognostic significance in hyperthyroid cats, however this does not appear to be independent of changes in PCV. An interaction between FGF-23 and PCV has not been noted previously.