Accuracy of formulas used to predict post‐transfusion packed cell volume rise in anemic dogs

Abstract

To assess the accuracy of published formulas used to guide packed red blood cell (pRBC) transfusions in anemic dogs and to compare the predicted rise in packed cell volume (PCV) to the actual post-transfusion rise in PCV. Prospective observational study from April 2009 through July 2009. A small animal emergency and specialty hospital. Thirty-one anemic client-owned dogs that received pRBC transfusions for treatment of anemia. None Four formulas were evaluated to determine their predictive ability with respect to rise in PCV following transfusion with pRBC. Post-transfusion rise in PCV were compared to calculated rise in PCV using 4 different formulas. Bias and limits of agreement were investigated using Bland-Altman analyses. Accuracy of existing formulas to predict rise in PCV following transfusion varied significantly. Formula 1 (volume to be transfused [VT] [mL]= 1 mL × % PCV rise × kg body weight [BW]) overestimated the expected rise in PCV (mean difference, 6.30), while formula 2 (VT [mL]= 2 mL ×% PCV rise × kg BW) underestimated the rise in PCV (mean difference, -3.01). Formula 3 (VT [mL]= 90 mL × kg BW × [(desired PCV - Patient PCV)/PCV of donor blood]) and formula 4 (VT [mL]= 1.5 mL ×% PCV rise × kg BW) performed well (mean difference 0.23 and 0.09, respectively) in predicting rise in PCV following pRBC transfusion. Agreement between 2 formulas, "VT (mL)= kg BW × blood volume (90 mL) × [(desired PCV - recipient PCV)/Donor PCV]" and "VT (mL)= 1.5 ×desired rise in PCV × kg BW," was found when they were compared to the actual rise in PCV following pRBC transfusion in anemic dogs. Further research is warranted to determine whether these formulas perform similarly well for other species.