Maternal venous Doppler characteristics are abnormal in pre-eclampsia but not in gestational hypertension

Abstract

Objective To compare functional characteristics of maternal thoraco-abdominal arteries and veins in proteinuric and non-proteinuric hypertension in pregnancy. Methods This retrospective study included women with singleton pregnancies during the third trimester, which were either uncomplicated or complicated with different clinical types of hypertension: non-proteinuric gestational hypertension (GH), early-onset pre-eclampsia (PE) diagnosed< 34 weeks or late-onset PE diagnosed >= 34 weeks. Demographic maternal and neonatal data were recorded, together with maternal serum and urine analytes. All women underwent standardized automated blood-pressure measurement, together with non-invasive impedance cardiography (ICG), for measurement of cardiac output (CO), aortic flow velocity index (VI) and aortic flow acceleration index (ACI). A standardized combined Doppler-electrocardiography assessment of maternal venous hemodynamics was performed to measure renal interlobar vein impedance index (RIVI), hepatic vein impedance index (HVI) and venous pulse transit time (VPTT) in liver and kidneys. Finally, resistance index (RI), pulsatility index (PI) and arterial pulse transit time (APTT) were measured in the uterine arcuate arteries. Mann-Whitney U-tests and Fisher's exact tests were used for intergroup comparisons, and linear dependence between variables was assessed using Pearson's correlation coefficient (r). Results A total of 150 pregnancies were evaluated: 22 with uncomplicated pregnancy, 41 GH, 31 early PE and 56 late PE. Aortic VI and ACI were lower in GH, early PE and late PE than in uncomplicated pregnancy. Both early PE and late PE differed from GH by having shorter APTT in the uterine arcuate arteries and higher RIVI. Hemodynamic abnormalities were most pronounced in early PE, during which uterine arcuate artery RI was higher and VPTT in kidneys was shorter than in late PE. There was a significant correlation between degree of proteinuria and RIVI for the left (r = 0.381) and right (r = 0.347) kidney in late PE, but this was not true for early PE. Conclusions There is a gradient of worsening arterial and venous hemodynamic abnormalities from GH to late PE and then to early PE. Venous hemodynamic abnormalities are present only in PE, with a linear correlation between proteinuria and RIVI in late PE. The role of the maternal venous compartment in the pathophysiology and etiology of PE-related symptoms may be much more important than considered at present. Copyright (C) 2014 ISUOG. Published by John Wiley & Sons Ltd.