Evaluation of an automated fetal myocardial performance index.

OBJECTIVE: To compare automated measurements of the fetal left myocardial performance index (MPI) with manual measurements for absolute value, repeatability and waveform acceptability. METHODS: This was a multicenter international online study using images from uncomplicated, morphologically normal singleton pregnancies (16-38 weeks' gestation). Single Doppler ultrasound cardiac cycle images of 25 cases were selected, triplicated and randomized (n = 75). Six senior observers, unaware of the repetition of images, manually calculated MPI for each waveform and the results were compared with automation. Intraobserver repeatability and interobserver reproducibility were assessed using intraclass correlation coefficients (ICCs) and 95% CI. The agreement between each observer's manual MPI measurements and corresponding automated measurements was evaluated using Bland-Altman plots and ICCs with 95% CI. The degree of variation between experts in the classification of fetal MPI waveform quality was assessed using individual cardiac cycle left MPI images previously classified by two authors as 'optimal', 'suboptimal' or 'unacceptable', with 30 images selected for each quality group. Ten images in each category were duplicated and the resulting 120 images were randomized and then classified online by five observers. The kappa statistic (κ) was used to demonstrate interobserver and intraobserver agreement and agreement of classifications by the five observers. RESULTS: The automated measurement software returned the same value for any given image, resulting in an ICC of 1.00. Manual measurements had intraobserver repeatability ICC values ranging from 0.69 to 0.97, and the interobserver reproducibility ICC was 0.78. Comparison of automated vs manual MPI absolute measurements for each observer gave ICCs ranging from 0.77 to 0.96. Interobserver image quality classification agreement gave k = 0.69 (P < 0.001), and the intraobserver agreement was variable (κ ranging from 0.40 to 0.81). CONCLUSIONS: Automated fetal MPI provides superior repeatability and reproducibility to manual methodology. Additionally, experts vary significantly when classifying suitability of fetal MPI waveforms. Automated MPI may facilitate clinical translation by removing human subjectivity. Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.

Reproducibility of constituent time intervals of right and left fetal modified myocardial performance indices on pulsed Doppler echocardiography: a short report.

OBJECTIVE: To assess reproducibility of constituent time intervals measured by pulsed Doppler echocardiography for calculation of the right and left fetal myocardial performance indices (MPIs). METHODS: This was a prospective study of 30 normal singleton pregnancies (19-36 weeks). In each, five different time intervals were measured from the Doppler waveform and four repeated measures were taken for each time interval. Three were from the left heart: isovolumetric contraction time (ICT), isovolumetric relaxation time (IRT) and ejection time (ET); two were from the right heart: 'a-interval' and 'b-interval'. The left and right modified MPIs (Mod-MPIs) generated by these constituent time intervals were also evaluated. Two operators evaluated the reproducibility of all measures. RESULTS: There was generally good intra- and interobserver reproducibility for all time intervals and resultant Mod-MPIs: ICT, range, 19-43 ms with intraclass correlation coefficient (ICC), 0.91 (95% CI, 0.85-0.95); ET, range, 160-184.8 ms with ICC, 0.90 (95% CI, 0.84-0.95); IRT, range, 35.8-48.5 ms with ICC, 0.67 (95% CI, 0.52-0.81); a-interval, range 208-265 ms with ICC, 0.89 (95% CI, 0.82-0.94); b-interval, range, 163.1-188.3 ms with ICC, 0.82 (95% CI, 0.71-0.90); left Mod-MPI, range, 0.33-0.48 with ICC, 0.84 (95% CI, 0.74-0.91); right Mod-MPI, range, 0.21-0.49 with ICC, 0.82 (95% CI, 0.71-0.90). The 95% limits of agreement showed no statistically significant difference in measurements between the two examiners for all time intervals. CONCLUSION: This is the first study to evaluate the reproducibility of the component time intervals of both right and left fetal Mod-MPIs in the second and third trimesters and supports the reproducibility of fetal functional cardiac assessment. The IRT is the main source of variation for the left Mod-MPI so should be the focus of further investigation. Even using two separate pulsed-wave Doppler gates and therefore separate cardiac cycles, the right Mod-MPI is reproducible.

Fetal left modified myocardial performance index: technical refinements in obtaining pulsed-Doppler waveforms.

OBJECTIVE: To assess the influence of machine settings (wall motion filter (WMF), angle of insonation, Doppler aliasing) and phase of valve clicks on repeatability of measurement of the fetal left modified myocardial performance index (Mod-MPI). METHODS: Fetal left Mod-MPI was evaluated prospectively in 157 morphologically normal fetuses at 19-36 weeks' gestation. In a baseline cohort, a previously published technique and settings were used for measurement of Mod-MPI. In a second cohort, the influence of WMF, angle of insonation, Doppler aliasing and selection of the phase of the valve clicks on repeatability of measurement of Mod-MPI was assessed. RESULTS: The intraclass correlation coefficient (ICC) for measurement repeatability in the baseline cohort was 0.22. Increase in WMF to 300 Hz or 500 Hz was associated with an increase in ICC to 0.60 and 0.55, respectively. An angle of insonation of < 15° was associated with an increase in ICC to 0.79 and 0.78 at a WMF of 300 and 500 Hz, respectively. A WMF of 300 Hz, angle of insonation of < 15° and absence of aliasing was associated with an increase in ICC to 0.85 and 0.87 at the beginning and peak of the valve click, respectively. Mod-MPI ranged from 0.35 to 0.48. An increase in calculated MPI was associated with increasing WMF, selection of the beginning vs. peak of the valve clicks and increase in angle of insonation. The presence or absence of aliasing had no effect. CONCLUSIONS: Refinement of machine settings improves repeatability of Mod-MPI, as does selection of the peak of the valve click. We suggest a consensus be reached as to the precise measurement of MPI, but for the moment would suggest: WMF, 300 Hz; angle of insonation, < 15°; avoiding Doppler aliasing; and selection of valve click peak. Systematic variation in measurement of time intervals may be responsible for the widely varying published normal ranges for Mod-MPI.

Spatiotemporal image correlation-derived volumetric Doppler impedance indices from spherical samples of the placenta: intraobserver reliability and correlation with conventional umbilical artery Doppler indices.

OBJECTIVE: Volumetric impedance indices derived from spatiotemporal image correlation (STIC) power Doppler ultrasound (PDU) might overcome the influence of machine settings and attenuation. We examined the feasibility of obtaining these indices from spherical samples of anterior placentas in healthy pregnancies, and assessed intraobserver reliability and correlation with conventional umbilical artery (UA) impedance indices. METHODS: Uncomplicated singleton pregnancies with anterior placenta were included in the study. A single observer evaluated UA pulsatility index (PI), resistance index (RI) and systolic/diastolic ratio (S/D) and acquired three STIC-PDU datasets from the placenta just above the placental cord insertion. Another observer analyzed the STIC-PDU datasets using Virtual Organ Computer-aided AnaLysis (VOCAL) spherical samples from every frame to determine the vascularization index (VI) and vascularization flow index (VFI); maximum, minimum and average values were used to determine the three volumetric impedance indices (vPI, vRI, vS/D). Intraobserver reliability was examined by intraclass correlation coefficients (ICC) and association between volumetric indices from placenta, and UA Doppler indices were assessed by Pearson's correlation coefficient. RESULTS: A total of 25 pregnant women were evaluated but five were excluded because of artifacts observed during analysis. The reliability of measurement of volumetric indices of both VI and VFI from three STIC-PDU datasets was similar, with all ICCs ≥ 0.78. Pearson's r values showed a weak and non-significant correlation between UA pulsed-wave Doppler indices and their respective volumetric indices from spherical samples of placenta (all r ≥ 0.23). VOCAL indices from specific phases of the cardiac cycle showed good repeatability (ICC ≥ 0.92). CONCLUSION: Volumetric impedance indices determined from spherical samples of placenta are sufficiently reliable but do not correlate with UA Doppler indices in healthy pregnancies.

Development of Australian reference ranges for the left fetal modified myocardial performance index and the influence of caliper location on time interval measurement.

OBJECTIVES: To construct gestational age-adjusted reference ranges of the left fetal modified myocardial performance index (Mod-MPI) in the Australian population and assess the influence of valve click caliper position on constituent time intervals and the Mod-MPI. METHODS: This is a prospective longitudinal study of 117 normal singleton fetuses undergoing 318 ultrasound scans at 4-6 weekly intervals between 18 and 38 weeks of gestation. The isovolumetric contraction time (ICT), isovolumetric relaxation time (IRT), and ejection time (ET) were measured at 3 different caliper positions in each fetus: beginning of the original valve clicks ('original'), beginning of the reflected valve clicks ('reflected'), and peak of valve clicks ('peak'). The Mod-MPI was calculated as (ICT + IRT)/ET. RESULTS: The Mod-MPI increased throughout gestation with means ± SD of 0.42 ± 0.05 'reflected' and 'peak' versus 0.49 ± 0.03 'original' at 19 weeks, and means of 0.46 ± 0.05 'reflected and peak' versus 0.51 ± 0.08 'original' at 36 weeks. Throughout gestation, ICT remained fairly constant and IRT increased, while ET decreased with 'original' click and remained constant for 'reflected' and 'peak' clicks. A modest increase in Mod-MPI was seen with increasing fetal heart rate. Analysis of repeatability for the 3 methods showed the following ICCs: 'original', 0.797 (95% CI 0.762-0.829); 'reflected', 0.809 (95% CI 0.775-0.839), and 'peak', 0.799 (95% CI 0.764-0.831). CONCLUSIONS: Detailed exploration of the morphology of mitral and aortic valve closure and opening clicks shows how selection of different phases of these clicks may significantly influence the Mod-MPI. We recommend that the peak of the valve clicks be standardized between research groups.

Immediate outcome of twin-twin transfusion syndrome following selective laser photocoagulation of communicating vessels at the NSW Fetal Therapy Centre.

OBJECTIVE: To audit the outcome for laser photocoagulation for twin-twin transfusion syndrome (TTTS) as managed by the New South Wales Fetal Therapy Centre (NSW FTC). METHODS: A retrospective cohort study. Outcome data were reviewed for referrals between June 2003 and June 2008.The outcome measures included the severity of TTTS at presentation, delivery details (gestational age at delivery, birth weight and Apgar score at 5 min) and perinatal outcome (spontaneous miscarriage, premature rupture of membranes, intrauterine death, placental abruption and neonatal death). RESULTS: Seventy-nine patients were treated with laser therapy for stage I-IV TTTS (median stage III). Median gestational age at treatment was 20 weeks (range 16-25). Median gestational age at delivery was 32 weeks (range 24-40). Survival of at least one baby in this study was 90.7% (88.9% for anterior and 92.1% for posterior placenta), and of both babies was 60.0%. Median birth weight was 1788 g (range 490-3695). Median Apgar score was nine at 5 min. Three women required repeat laser treatment for persistent TTTS. CONCLUSIONS: Selective laser photocoagulation of communicating vessels remains the treatment of choice for TTTS. Referrals to the NSW FTC have increased from five cases in the last half of 2003, to 18 cases in the first half of 2008. Local outcome figures at least equal any in the published international literature and support a continued policy of centralised care in Australia. A two-year follow-up study on neonatal outcome for survivors is underway.

Long term developmental outcomes of pre-school age children following laser surgery for twin-to-twin transfusion syndrome.

BACKGROUND: Laser therapy is now a well recognised treatment for twin-to-twin transfusion syndrome (TTTS). We investigated the early childhood neurodevelopmental outcome of children post laser treatment for TTTS in our centre. METHODS: Children of women who had laser therapy for TTTS between March 2006 and June 2008 were assessed at 30-69 months of age with WPPSI-III and a general health questionnaire. Major neurodevelopmental impairment (NDI) was reported as IQ<70 or cerebral palsy (CP). Borderline cognitive impairment was defined by IQ 70-79. RESULTS: Amongst the 37 pregnancies treated, 62 infants were discharged home and the overall foetal survival rate was 84%. A total of 50 children (84%) from 31 pregnancies were assessed. Average age at assessment was 47 months. Two children with late treatment of congenital hypothyroidism were excluded. The majority of pregnancies were Quintero Stage III (74%). There was a significant trend for worse outcome with higher Quintero stage. The average gestational age at birth was 32 weeks. The majority (39, 78%) of children were found to be neurodevelopmentally normal; 9 (18%) had borderline cognitive development; and 2 (4%) had a major NDI, including one with cerebral palsy (2%). CONCLUSIONS: There was a modest level of neurocognitive impairment post laser therapy for TTTS, mainly borderline cognitive development, lesser so major NDI. There was a low incidence of cerebral palsy. Routine developmental and neurological follow-up of these children is recommended.