Umbilical artery Doppler revisited: pathophysiology of changes in intrauterine growth restriction revealed

Reduced placental stem artery number. The first mechanism proposed was based on pathological examination of placentae from pregnancies with IUGR and abnormal UA Doppler findings, and suggested that such cases were associated with a reduced number of placental villus stem arteries9. Subsequently, other authors appeared to confirm these findings10, 11, but more recent studies with systematic sampling techniques have been unable to confirm such results12-14. Primary villus maldevelopment. Pathological studies of placentae with IUGR and abnormal UA Doppler findings reported that the terminal villi are often small, hypovascular and fibrotic. This prompted the hypothesis that primary villus maldevelopment may be the underlying event in such cases13-15. Placental stem vessel vasoconstriction. Several authors have reported the pathological observation that stem vessels in placentae from cases of IUGR demonstrate apparent luminal reduction, wall hypertrophy and luminal herniation, consistent with marked, longstanding placental stem villus vasoconstriction16-20. The vast majority of affected pregnancies also demonstrate abnormal Doppler flow velocity waveforms in the uterine arteries, many of which are abnormal long before the UA waveforms deteriorate. This suggests that in most cases of IUGR the underlying event is a reduction in uteroplacental blood flow21, 22. This is in accordance with the pathological finding of defective endovascular trophoblast invasion and conversion of uteroplacental vessels in such cases23, 24. The most common macroscopic placental lesions associated with severe IUGR are placental infarcts, which are areas of villus ischemic necrosis secondary to severe reduction in oxygen delivery due to severe localized impairment of uteroplacental intervillus blood flow25. There is progression of degree of abnormality in the UA Doppler waveform with time, with eventual development of absent or reversed end-diastolic flow3. UA Doppler indices may change rapidly following experimental occlusion of maternal uteroplacental flow26. Resistance to flow in almost all other organs is controlled at the level of the small arteries/arterioles, not at the level of the capillary bed itself since large changes in capillary pressures will lead to tissue fluid balance alterations. These features suggest that the underlying mechanism of abnormal UA Doppler waveforms in most cases of IUGR is likely to be secondary to significant reductions in maternal uteroplacental flow. Therefore, the hypothesis of primary placental/villus maldevelopment is not consistent with the associations with abnormal uteroplacental flow and finding of placental infarcts in such cases. Similarly, the findings of clinical progression, temporal variability and potential for rapid changes secondary to rapid changes to uteroplacental flow all suggest that the mechanism of UA Doppler change is potentially dynamic and rapidly acting, making primary defective growth or intrinsic defective number of stem arteries unlikely. Furthermore, the anatomical terminal villus changes previously described could be a consequence of reduced local fetoplacental flow rather than a primary event and, indeed, similar alterations are described in cases of placental stem villus thrombosis25. The suggestion that placental stem villus vasoconstriction is the underlying mechanism in IUGR with abnormal UA Doppler waveforms has increasing supportive evidence. Anatomically, the vascular smooth muscle of placental stem villus arteries and veins is well developed suggesting an important role in control of placental hemodynamics, most likely mediating control of intraplacental blood flow to minimize ‘ventilation–perfusion mismatch’, in a similar way to that described in the postnatal lung27, 28. The details of such a mechanism, and its secondary effects on placental blood flow, Doppler indices and fetomaternal water balance, have been previously described27-29. The current theory is that prolonged reduction in maternal intervillus flow, and hence oxygen delivery, would therefore result in prolonged vasoconstriction with secondary reduction in stem villus luminal diameter, increased flow resistance and vascular medial hypertrophy. Histopathological evidence for such changes compatible with prolonged vasoconstriction have now been well described in stem vessels from placentas in pregnancies with IUGR by several groups (Figure 1)16-20. (a) Photomicrograph of a histological section from a pregnancy with intrauterine growth restriction and abnormal umbilical artery Doppler indices, demonstrating a parenchymal stem villus with marked stem artery vasoconstriction and wall hypertrophy (hematoxylin and eosin, original magnification ×100). (b) Photomicrograph of a stem villus from an uncomplicated term pregnancy (hematoxylin and eosin, original magnification × 100). The precise mechanism of control of stem villus vessel tone has not previously been determined with certainty, although release of vasoactive mediators from the distal terminal villi, such as nitric oxide (NO), has been suggested30. Recently, however, a detailed study examining hypoxic fetoplacental vasoconstriction using perfused isolated human placental cotyledons, fetoplacental vascular rings in tissue baths and microdissected isolated fetoplacental vessels using electrophysiology, reverse transcriptase-polymerase chain reaction and immunoblotting provides significant new information31. In vitro hypoxia of the cotyledonary perfusate caused significant fetoplacental vasoconstriction in all cases with markedly increased fetoplacental flow resistance. Furthermore, inhibition of NO synthesis reduced basal perfusion flow resistance but did not affect the marked hypoxic vasoconstrictive response31. The vasoconstrictive response was mediated purely at the level of the small, intraplacental stem vessels rather than the larger stem arteries, and was due to hypoxic inhibition of potassium channels within their vascular smooth muscle31. (Such a mechanism has indeed been suggested as part of the oxygen-sensing system of other tissues.) Furthermore, patch-clamping and molecular studies confirmed marked differences in both potassium channel concentration and hypoxic reactivity between small and large fetoplacental vessels31. Examination of the Doppler flow velocity waveform pattern in the UAs provides useful information regarding the fetal condition in pregnancies affected by severe IUGR. In a very small number of cases with severe IUGR in the second trimester there may possibly be a primary defect in placental development as the underlying abnormality, but for the vast majority of cases of IUGR the underlying defect is the result of poor trophoblastic invasion of maternal spiral arteries and reduced uteroplacental flow. This reduction in intervillus flow results in local, intraplacental stem vessel vasoconstriction to minimize intervillus/intravillus flow mismatch; deleterious shunting is thereby initially prevented and fetal oxygenation maintained. However, when widespread, such a response results in increased fetoplacental flow resistance and abnormal UA Doppler waveforms. Progressive reduction in intervillus flow would result in progressive fetoplacental vasoconstriction and worsening of the UA Doppler indices. Local placental NO production appears important for control of basal fetoplacental vascular tone but the marked hypoxic vasoconstrictive response in pathological pregnancies is mediated by inhibition of potassium channels in small intraplacental stem vessels. This recent improved understanding of the pathophysiology and hemodynamics of fetoplacental flow in IUGR may allow development of novel techniques of fetal surveillance and/or intervention.