The effect of myxomatous mitral valve disease severity on packed cell volume by Professor Adrian Boswood and colleagues at the RVC
This work identified an inverse relationship between packed cell volume (PCV) and the severity of myxomatous mitral valve disease (MMVD) in dogs. The study was a retrospective analysis of prospectively collected data selected from a large population of dogs (n=289) on more than one occasion (n=1465 visits) between 2004-2017 at a research clinic conducted by the RVC in two primary-care practices in London. A control population of normal, unaffected patients seen at the same research clinic was included in the study. Appropriate statistical techniques were used to account for repeated measures from the same individual.
Study eligibility included an echocardiographic diagnosis of MMVD by a board-certified cardiologist and a measurement of PCV available from the same examination. Dogs underwent repeated examinations approximately every 6 months and at each visit a jugular blood sample was taken. PCV (% whole blood), total solids (TS, g/L) and blood urea nitrogen (BUN, mmol/L) concentration were measured. Standard right parasternal echocardiographic views were obtained, from which the left atrial to aortic root ratio and the normalised left ventricular internal diameter at end diastole were calculated. Heart disease was staged according to American College of Veterinary Internal Medicine (ACVIM) guidelines.
PCV appeared to decrease with increasing severity of preclinical disease, producing a significant difference when stages B1 and B2 were compared. This inverse relationship with ACVIM stage did not continue through to stage C. PCV in stage C was significantly greater than either preclinical stage, lying closer to that of the unaffected dogs. A second analysis indicated that the increased PCV in stage C disease may be due to diuresis reducing plasma volume, as dogs receiving furosemide had higher values of PCV. This suggests that dogs with MMVD may develop small reductions in PCV as a result of haemodilution.
Similar associations between left ventricular internal diameter at end diastole, furosemide administration and TS were observed when TS was analysed as the dependant variable. Haemodilution would not be expected to affect PCV in isolation, so finding these results for another component of blood provides further support for the interpretation. In addition to ACVIM stage, several demographic factors also remained in the model. When grouped as Cavalier King Charles Spaniel (CKCS) or non-CKCS, PCV was significantly lower in CKCS.
This study furthers understanding of the pathophysiology of MMVD by suggesting that plasma volume is expanded as early as stage B1.