In this prospective study, the relationships among the modified left ventricular myocardial performance index (Mod-MPI), fetal cardiac timing parameters, and perinatal outcomes were evaluated in fetuses diagnosed with fetal growth restriction (FGR) grouped according to the presence or absence of absent end-diastolic flow (AEDF) in the umbilical artery, as well as in a control group. The findings revealed that in the AEDF (+) group, isovolumetric phases were significantly prolonged, the ejection time was shortened, and both the E- and A-wave velocities were decreased, resulting in an overall increase in Mod-MPI values. The significant reduction in both E- and A-wave velocities, particularly in combination with prolonged isovolumetric phases and elevated Mod-MPI values, indicates impaired diastolic function of the fetal left ventricle. This finding suggests dysfunction in both the passive and active phases of fetal cardiac filling. This group was also characterized by elevated umbilical artery PI, a reduced cerebroplacental ratio (CPR), lower Apgar scores, and increased NICU admission rates. Although the difference in MPI between the groups did not reach statistical significance, the strong correlations observed between MPI and timing parameters such as IRT, ICT, and ET suggest that the Mod-MPI may reflect the fetal heart’s pathophysiological response at an early stage.
Fetal cardiac functions can be influenced not only by maternal factors such as hypertension, diabetes, and infections but also by fetal hemodynamic and structural conditions. Among these, intrauterine growth restriction (FGR) is considered one of the most significant conditions leading to impairment of the fetal cardiovascular system. In fetuses with FGR, the primary cause of ventricular remodeling is increased afterload resulting from elevated placental vascular resistance. However, increased afterload during fetal life is not unique to FGR; similar effects may also be observed in other conditions, such as in the recipient twin of twin-to-twin transfusion syndrome (TTTS), where hypervolemia leads to comparable hemodynamic stress. In such cases, the increased volume load and concomitant placental resistance can result in significant deterioration of the myocardial performance index (MPI). Taken together, these findings suggest that both increased placental vascular resistance and hypervolaemia can significantly affect fetal cardiac function [11].
It has been demonstrated that as the severity of fetal growth restriction (FGR) increases—particularly in cases of early-onset and severe FGR—MPI values also increase in parallel, reflecting prenatal cardiac dysfunction [12]. In particular, it has been suggested that in FGR cases with abnormal umbilical artery Doppler findings (such as absent end-diastolic flow), left heart MPI values may be greater than those in FGR fetuses with normal Doppler flow. However, the findings in the literature remain inconsistent; some researchers have reported no significant difference in MPI even in the presence of AEDF and have suggested that the clinical utility of MPI is still not clearly established [13]. In our study involving 217 pregnancies, we evaluated the relationships among the left ventricular myocardial performance index (Mod-MPI), fetal cardiac timing parameters, and perinatal outcomes.
In FGR cases resulting from uteroplacental insufficiency, the fetal cardiovascular system attempts to adapt to this chronic stress through compensatory mechanisms. Increased placental bed resistance leads to elevated afterload and reduced oxygen supply, causing subclinical changes that primarily affect diastolic function in the fetal heart. Prolonged hypoxic exposure impairs myocardial relaxation capacity and paves the way for the development of diastolic dysfunction. This condition may manifest early as reduced longitudinal myocardial motion and a delayed filling process. The redistribution of fetal circulation (brain-sparing mechanism) prioritizes flow to vital organs, while the heart working under increased hemodynamic load necessitates a comprehensive assessment of global cardiac function [14]. In our study, the FGR group presented significantly prolonged isovolumetric phase durations, shortened ejection times, and markedly reduced E- and A-wave velocities. These findings indicate that under chronic hypoxia, the fetal myocardium develops early diastolic dysfunction, which can be objectively assessed through the components of the modified MPI.
In the study by Bhorat et al., which evaluated 43 fetuses with FGR between 28 and 34 weeks of gestation, the left myocardial performance index (MPI) values measured by spectral Doppler significantly increased in parallel with the worsening of Doppler parameters. In cases of early-onset FGR, the mean MPI threshold was 0.59, whereas in the control group consisting of appropriate-for-gestational-age (AGA) fetuses, the value was 0.37. An MPI threshold of 0.54 provided 87% sensitivity and 75% specificity for predicting adverse perinatal outcomes, whereas a threshold of 0.67 was reported to yield 100% sensitivity and 81% specificity in predicting perinatal mortality. These findings suggest that MPI may serve as a potential prognostic marker reflecting fetal cardiac load and increased perinatal risk [15]. Similarly, in our present study, the mean MPI value was calculated as 0.51 ± 0.07 (median: 0.51; min– max: 0.36–0.70) in the 103 fetuses constituting the FGR group. In the control group, consisting of 114 fetuses with normal growth appropriate for gestational age, the mean MPI value was 0.50 ± 0.07 (median: 0.49; min–max: 0.41–0.73). The difference between the two groups was statistically significant (p = 0.036). These results suggest that the modified MPI may be a sensitive and clinically applicable cardiac marker for detecting subclinical myocardial dysfunction in FGR-induced cases.
While the MCA-PI is used to assess fetal cerebral perfusion, the cerebroplacental ratio (CPR), calculated as the MCA-PI/UA-PI ratio, is utilized to predict the risk of fetal hypoxia. Low CPR values have been associated with adverse perinatal outcomes such as emergency delivery and low Apgar scores. Although the prognostic power of the MCA-PI alone is limited, the DV-PI reflects the risk of intrauterine death, particularly in early FGR; however, its predictive value is diminished in late FGR [16]. In our study, the CPR values were significantly lower in the FGR group than in the control group. The reduction in CPR indicates the development of redistribution in fetal circulation toward the cerebral region, reflecting the activation of the classical “brain-sparing” adaptive response. Although this physiological mechanism aims to preserve cerebral perfusion in response to a reduced oxygen supply, it imposes increased hemodynamic stress on other organ systems, particularly cardiac structures. Indeed, the significantly elevated MPI values observed in the same group represent the fetal heart’s response to this adaptive process. The increase in MPI, accompanied by prolonged ICT and IRT durations and shortened ET, suggests that the fetal myocardium is experiencing heightened diastolic pressure and is approaching its compensatory limit. Therefore, the presence of elevated MPI values in conjunction with low CPR indicates diminished fetal systemic reserve and cardiac tolerance; when these two parameters are evaluated together, they may enable early identification of the risk for adverse perinatal outcomes.
In our study, significantly lower 1- and 5-minute Apgar scores were observed in the FGR group than in the control group, particularly among patients with AEDF (+). This finding supports the notion that fetal cardiac performance influences postnatal adaptive capacity. Since an increase in MPI reflects impairment in both systolic and diastolic functions, elevated MPI values are associated with reduced postnatal cardiac reserve and inadequate circulatory response. Similarly, the literature reports that FGR fetuses with high prenatal MPI values are more likely to have lower Apgar scores and an increased need for resuscitation [15,16,17]. However, when interpreting this relationship, it is important to consider that the AEDF (+) group delivered at earlier gestational ages, and other prematurity-related factors may also influence the Apgar score. In this context, MPI may serve not only as an indicator of intrauterine cardiac stress but also as an indirect marker of early neonatal cardiac function. The negative association observed between MPI and the Apgar score suggests that the modified MPI could be a clinically useful adjunctive parameter for the early identification of adverse perinatal outcomes.
Although, theoretically, right ventricular MPI measurement is considered more physiologically appropriate owing to right ventricular dominance in the fetal period, the left ventricle is generally preferred in clinical practice owing to technical challenges. Assessment of the right ventricle requires two separate imaging planes through the pulmonary and tricuspid valves, making it time-consuming and technically difficult to perform. In contrast, since the aortic valve (AoV) and mitral valve (MV) can be visualized within the same plane in the left ventricle, Doppler measurements have become more practical and reproducible. Indeed, the literature shows that MPI values derived from the left ventricle are more frequently reported than those from the right ventricle. However, the clinical utility of left ventricular MPI in the fetal period remains controversial. This inconsistency largely stems from technical variations in MPI measurement. Factors such as the device characteristics, insonation angle, and sweep speed can directly influence the measurement results. For example, Hernandez-Andrade et al. performed modified MPI measurements using an insonation angle of 0–30° and a sweep speed of 15 cm/s, whereas other studies have used narrower angles and different sweep speeds, making cross-study comparisons difficult [10, 18]. In our study, modified MPI measurements were performed from the left ventricle using an insonation angle of 0–15° and a sweep speed of 5 cm/s, aligning calipers at the peak points of the AoV and MV clicks. This method aims to reduce subjective variability in caliper placement and enhance measurement reproducibility. Owing to this standardized technique, MPI values across different gestational ages could be reliably compared.
Amniotic fluid volume primarily depends on fetal kidney filtration and urine production. Fetal renal artery Doppler measurements reflect renal perfusion, and these parameters have been reported to be associated with amniotic fluid levels [19]. In our study, when the FGR group [AEDF (+), AEDF (–)] was compared with the control group, significant differences were observed not only in MPI values but also in perinatal parameters such as mode of delivery, amniotic fluid index (AFI), and NICU admission rates. In particular, the AEDF (+) group had a higher rate of primary cesarean section, lower AFI levels, and a significantly increased frequency of oligohydramnios. These findings suggest that fetal circulatory dysfunction is not limited to the placental level but also adversely affects physiological processes such as renal perfusion, urine production, and fetal breathing movements. The increase in oligohydramnios may indicate reduced fetal urine output due to decreased renal perfusion, leading to the development of hypovolemia.
In studies investigating the relationship between polyhydramnios and Mod-MPI, it has been reported that, in the presence of isolated polyhydramnios, the Mod-MPI values are significantly greater than those in control groups, and this increase may be associated with adverse perinatal outcomes [20]. In our study, polyhydramnios were detected in 2 patients (4.3%) in the AEDF (+) FGR group, and the Mod-MPI values in these patients were notably above the group average. In both cases, no structural anomalies or infectious etiologies were identified apart from isolated polyhydramnios; however, respiratory distress and a need for neonatal intensive care developed during the neonatal period. These findings suggest that increased amniotic fluid volume may be associated with fetal cardiac loading and myocardial dysfunction. Consistent with the literature, Mod-MPI may serve as a useful indicator for evaluating fetal cardiac reserve in cases of isolated polyhydramnios. However, larger-scale studies are needed to strengthen this conclusion.
Fetal circulation is a dynamic system that is highly dependent on the oxygen and nutrient supplies from the uteroplacental bed, as well as the capacity of the fetoplacental system to transport these resources effectively. Any disruption within this system is first reflected by an increase in vascular resistance, followed by changes in blood flow patterns [21]. In our study, when the PI values of the umbilical artery, middle cerebral artery, and bilateral uterine artery were compared among the AEDF (+), AEDF (–), and control groups, statistically significant differences were detected (p < 0.001). The mean umbilical artery PI (UA-PI) in the AEDF (+) group was 2.03 ± 0.23, which was markedly greater than that in the AEDF (–) and control groups (1.01 ± 0.19 and 1.01 ± 0.14, respectively). This finding indicates a significant increase in placental vascular resistance in AEDF (+) patients and, in line with MPI, suggests increased loading in the fetal circulation. Similarly, middle cerebral artery PI (MCA-PI) values were lower in the AEDF (+) group (1.54 ± 0.45), which is consistent with the “brain-sparing response” that reflects preferential blood flow to the brain. The reduction in MCA-PI demonstrates that, as a compensatory response to fetal hypoxia, the fetal cardiovascular system prioritizes perfusion to central organs. The right and left uterine artery PI values were also significantly greater in the AEDF (+) group (1.74 ± 0.57 and 1.95 ± 0.71, respectively), indicating maternal-origin uterine perfusion insufficiency. These findings reveal that both fetoplacental and uteroplacental circulation are impaired in the AEDF (+) group and that this disruption adversely affects fetal hemodynamics and cardiac function, thereby increasing perinatal risk. In conclusion, elevated UA-PI and uterine artery PI values along with decreased MCA-PI reflect the hemodynamic signs of fetal distress, whereas the concurrently elevated MPI values support the notion that this stress also mirrors the level of cardiac function.
In our cohort, the E/A ratio—calculated as an additional index of left ventricular diastolic function—was similar across groups, despite significant reductions in both E and A velocities in the AEDF (+) subgroup. This pattern suggests proportional declines in early and late diastolic filling, accompanied by shortened systolic ejection time, rather than selective impairment of one diastolic phase. While the composite Mod-MPI did not differ significantly between groups, individual components such as ET and IRT, when interpreted alongside conventional Doppler indices (UA PI, MCA PI, CPR), may offer incremental value in identifying fetuses at risk for adverse perinatal outcomes. Although ROC analyses in our dataset yielded AUC values below 0.50 for these individual parameters, the concept of a combined predictive model remains clinically appealing. Future multicenter prospective studies could integrate UA Doppler abnormalities, CPR<1.0, and MPI component alterations (ET shortening, prolonged IRT, reduced E/A ratio) into a multivariate algorithm to establish clinically applicable cut-offs and improve antenatal surveillance strategies in FGR.
Limitations
Among the strengths of our study are its prospective design and the fact that all Doppler measurements were performed in a single center by an experienced perinatologist via a standardized protocol. However, being a single-center study may limit the generalizability of the findings to other populations or institutions. Modified left ventricular MPI measurements were obtained via the pulsed-wave Doppler method, which can be affected by factors such as fetal position, breathing movements, variability in heart rate, and operator-dependent technical variables. Although all the measurements were performed by an experienced specialist, small deviations in the timing components of the MPI—namely, IRT, ICT, and ET—may lead to significant differences in the total index. Long-term neurodevelopmental and cardiac follow-up data were not available, which limits the ability to correlate prenatal Doppler findings with later functional outcomes.