Objective: To determine the detection rate and the reproducibility of the first (K1), the fourth (K4) and the fifth (K5) phases of the Korotkoff sounds in pregnant women.
Study design: In 77 pregnant women receiving antenatal care in a tertiary referral centre, two observers took 231 simultaneous blood pressure measurements with a shared mercury sphygmomanometer and a multi-aural stethoscope. Detection rates, percentages of observer agreement within 5 mmHg and kappa values were calculated for K1, K4 and K5.
Results: Both observers were able to detect K1, K4 and K5 in 98, 24 and 98% of all 231 measurements, respectively. In 46% of measurements, and in 23% of women, neither observer was able to detect K4. The observers agreed within 5 mmHg in 98 (kappa = 0.99), 69 (kappa = 0.42), and 96% (kappa = 0.99) of measurements for K1, K4 and K5, respectively.
Conclusion: K4 cannot be detected and reproduced accurately in a significant proportion of pregnant women, and therefore should be abandoned as diastolic endpoint in pregnancy.