THE EFFECT OF PATIENT POSITION ON THE REPRODUCIBILITY OF CARDIAC-OUTPUT MEASUREMENTS

Objective: To determine the effect of patient position on the reproduc ibility of cardiac output measurements. Design: Prospective, two-group quasi-experimental design. Convenience sample. Setting: The study inv olved two intensive care units in two adult acute care hospitals. Pati ents: Thirty patients admitted to the intensive care unit who had a th ermodilution pulmonary artery catheter in place. Ages ranged from 39 t o 80 years (mean of 66.4 +/- 11.3 years). Outcome Measures: Thermodilu tion cardiac output measurements. Intervention: The subjects were plac ed in one of two groups, initially by flipping a coin then into altern ate groups. Group A subjects were placed supine, and after 5 minutes h ad cardiac output measurements performed. They were then placed in the 45-degree upright position, and after an additional 5 minutes had car diac output measurements performed. Group B subjects were first placed in the 45-degree upright position, and after 5 minutes had cardiac ou tput measurements performed. They were then placed in the supine flat position, and after an additional 5 minutes had cardiac output measure ments performed. Results: Seventy percent (n = 30) of the sample popul ation displayed a lower cardiac output in the 45-degree upright positi on than that obtained in the supine position, with the decrease rangin g from 1% to 32% (mean decrease 11%). Forty percent (n = 30) of cardia c output measurements obtained in the 45-degree upright were greater t han or equal to 10% less than those obtained in the supine flat positi on. The differences in cardiac output were analyzed with the paired t test, which produced a 95% confidence interval from -0.539 to -0.083. The two-group Wilcoxon test was used to analyze the mean cardiac outpu t with the patient in the supine, flat position and in the 45-degree u pright position. The mean cardiac output at 0 degrees was found to be statistically significant higher (p = 0.0083) than the mean cardiac ou tput at 45 degrees. The effect of coexisting variables were analyzed w ith the Kruskal-Wallis. The use of vasoconstrictors was the only varia ble that had a statistically significant change in cardiac output asso ciated with a change in position. Conclusions: These results indicate that cardiac output measurements are affected by alterations in patien t position. To ensure accurate comparisons between consecutive cardiac output measurements, the researchers recommend that the position in w hich the cardiac output measurements are performed be documented and t he cardiac output measurements be conducted in a uniform position.