1981
Quantitative MRI detects impaired vascular reactivity in women after preeclamptic pregnancy1Radiology, University of Pennsylvania, Philadelphia, PA, United States, 2University of Pennsylvania, Philadelphia, PA, United States, 3Maternal Fetal Medicine, University of Pennsylvania, Philadelphia, PA, United States
Synopsis
Large body of evidence suggests maternal endothelial dysfunction (EDF) has a central role in the development of preeclampsia, the most serious hypertensive pregnancy disorder that significantly increases risk for future cardiovascular diseases. Because pathophysiology of preeclampsia remains within 72 hrs of delivery of the placenta, the quantification of surrogate MRI markers of EDF was performed after birth in women with and without hypertensive pregnancy. The quantitative MRI protocol evaluates peripheral micro- and macrovascular reactivity and central arterial stiffness in a single scan session. Preliminary results show a trend of impaired vascular reactivity after hypertensive pregnancy relative to normotensive pregnancy.
INTRODUCTION
Preeclampsia (PE) is the most serious hypertensive pregnancy disorder that accounts for considerable perinatal morbidity and mortality1, 2. The hallmark of PE disease process is maternal endothelial dysfunction (EDF) initiated by abnormalities in early placental development leading to the release of antiangiogenic factors and inflammatory activators into the maternal circulation3, 4. Manifestation of EDF include increased arterial stiffness and peripheral vascular resistance5. Corresponding negative vascular consequences are elevated pulse-wave velocity (PWV) and compromised ability for arterioles to accommodate transient increase in blood flow. Toward establishing a systemic EDF evaluation we had previously integrated multiple quantitative MRI (qMRI) techniques to interrogate the central and peripheral vascular reactivity and successfully applied it to assess EDF associated with aging, cigarette smoke exposure and peripheral artery disease6, 7. The goal of the pilot study was to apply qMRI to evaluate the underlying changes in endothelial function in relation to PE in major vascular territories.METHODS
The qMRI protocol comprises dynamic venous oximetry8, time-resolved arterial velocimetry9 and real-time aortic arch pulse-wave velocity quantification10.
Quantification of peripheral vascular reactivity with dynamic oximetry and velocimetry
Dynamic oximetry and velocimetry were developed to quantify reactive hyperemia in the lower extremity in response to a 5 min cuff-induced ischemia. Only one cuff paradigm is needed to assess micro- and macrovascular function of the peripheral vascular territory by merging the methods as a multi-echo GRE pulse sequence that samples velocity-encoded projections (velocimetry) and full-image echoes for field mapping (MR susceptometric oximetry11) (Fig 1). In dynamic oximetry blood in the capillary bed serves as an endogenous tracer that is “tagged” via oxygen extraction during the suspension of arterial resupply via cuff-occlusion. Upon cuff release or restoration of arterial flow, the kinetics of the oxygen-depleted blood in the capillary bed and its subsequent replacement by normally oxygenated venous blood is then temporally resolved at the imaging location every 2s (Fig 2a). The blood flow velocity in the femoral artery is quantified simultaneously with the velocity-encoded projections to characterize hyperemia (Fig 2b).
Assessment of aortic arch stiffness with non-cardiac triggered pulse-wave velocity quantification
The stiffness of the central artery is assessed in terms of aortic arch pulse-wave velocity (PWV), which represents the rate at which the blood motion is transmitted to downstream locations. For example, in a rigid vessel PWV approaches infinite because proximal pressure increase will displace all blood (incompressible) at the same time. Thus, instantaneous transmission of the motion would result in no phase shift between the velocity waveforms at two locations. In our method, the “foot” of the velocity waveform is time-resolved without gating via complex difference signal intensity computed with velocity-encoded projections (Fig 3). In this manner, gating errors are eliminated compared to conventional MRI methods12, 13 and velocity waveforms over multiple heartbeats can be acquired in real-time. The time-resolved “velocity” waveforms are plotted jointly to determine the transit time via the “foot-to-foot” method as routinely performed in tonometry14. All MR procedures were performed on a 1.5T Siemens Avanto imager (Siemens Medical Solutions) using standard RF coils (CP extremity coil for cuff paradigm studies at the location of the femoral artery and vein, body matrix/spine coil for PWV of the aortic arch). The MRI examination involving quantification of vascular reactivity of femoral artery and vein via 5 min-long cuff-induced ischemia and reperfusion was followed by image-guided quantification of the aortic arch PWV. Subjects were recruited after the delivery because the clinical pathophysiology of PE remains for at least 72 hours postpartum. The profile of the postpartum cohorts is summarized in Table 1. The purpose of this pilot study is to evaluate the hypothesis that hallmark of preeclampsia is impaired endothelial function.
RESULTS
The bar graphs of Fig 4 represent the group averages for each MRI surrogate marker of EDF. All parameters support a trend of impaired vascular reactivity (e.g. elevated PWV, and compromised micro- and macrovascular function: reduced upslope and overshoot, shorter TFF, lower peripheral flow reserve (see Fig 2), etc.) in women diagnosed with preeclampsia during pregnancy relative to normal pregnancies.CONCLUSION
The preliminary data suggests that qMRI can detect negative effects of preeclamptic pregnancy on vascular reactivity but a larger cohort size is needed for a more comprehensive investigation. Studies that examine maternal vascular reactivity associated with the profound cardiovascular system adaptations at 2nd and 3rd trimesters is also warranted to better understand the normal changes in the during pregnancy. Lastly, because there is little information available on women before their preeclamptic pregnancy, the observed EDF in the present pilot study does not imply that PE causes EDF.Acknowledgements
NIH Grants K25 HL111422, RO1 HL109545 and RO1 HL139358, UL1 TR001878, U01 HD087180.References
[1] Hutcheon et al, Best Pract Res Clin Obstet Gynaecol. 2011; [2] Lisonkova et al, Obstet Gynecol. 2014; [3] Roberts JM. Semin Reprod Endocrinol. 1998; [4] Roberts et al, Am J Obstet Gynecol. 1989; [5] Osol et al, Curr Hypertens Rep. 2017; [6] Langham et al, JCMR 2013; [7] Langham et al, JCMR 2015; [8] Langham et al, JACC 2010; [9] Langham et al, MRM 2010; [10] Langham et al, MRM 2011; [11] Langham et al, JCMR 2011; [12] Mohiaddin et al, J Appl Physiol. 1993; [13] Rogers et al, JACC 2001; [14] Van Bortel et al. Am J Hypertens. 2002.Figures
