Ophthalmic Artery Doppler in the Puerperium After Hypertensive Disorders of Pregnancy (OPHELIA-PP)

Overview

Hypertensive disorders of pregnancy frequently fail to resolve at delivery. A substantial proportion of women remain hypertensive at the six-week postnatal review, postpartum eclampsia occurs days to weeks after birth, and a minority show persisting renal dysfunction. Attendance at postnatal follow-up is nonetheless poor and specialist resources are limited, so a simple bedside test that identified the women whose cardiovascular recovery is incomplete would allow surveillance to be targeted. Ophthalmic artery Doppler is non-invasive, takes about five minutes and requires only a linear transducer already present in obstetric units. Its antenatal performance is well characterised, but a systematic review by this group (PROSPERO CRD420261488689) identified only three studies in thirty years addressing the postpartum period, two of which included a normotensive comparator, totalling 131 women. None reported a threshold or any measure of diagnostic accuracy, and none recorded antihypertensive treatment at the time of scanning. This prospective cohort study will measure ophthalmic artery Doppler indices at four timepoints between delivery and three months in women with hypertensive disorders of pregnancy and in normotensive controls, and will determine whether the peak systolic velocity ratio measured 24 to 48 hours after delivery predicts persistent hypertension at the six-week postnatal review.

Design. Prospective observational cohort study with consecutive recruitment and a concurrently enrolled normotensive comparison group, conducted at a coordinating tertiary centre and participating satellite centres. Non-interventional nature. Ophthalmic artery Doppler does not alter clinical management, which follows existing institutional protocols throughout. Doppler results are not communicated to the clinicians responsible for the participant, who remain blinded to them for the duration of the study, so that the measurement cannot influence the outcome it is intended to predict. An alerting procedure is nonetheless in place for incidental findings of clinical relevance. Assessments. Ophthalmic artery Doppler is performed at 24 to 48 hours (T0), 7 to 10 days (T1), six weeks (T2) and three months (T3) after delivery, with a telephone follow-up at six months (T4). Blood pressure is measured at each visit using a standardised protocol with a validated automated device. Estimated glomerular filtration rate and albumin-to-creatinine ratio are obtained at six weeks. Doppler protocol. Participants are examined supine with the eyes closed after five minutes of rest. A 7.5-9 MHz linear transducer is placed on the closed upper eyelid without pressure on the globe. The ophthalmic artery is identified superior and medial to the optic nerve, 10-15 mm from the posterior scleral wall. Pulsed-wave sampling uses an insonation angle below 20 degrees, a 2 mm sample gate, a 50 Hz wall filter and a depth of 3.0-4.5 cm, with mechanical index kept at or below 0.3 and insonation time under one minute per eye in accordance with the ALARA principle. Three to five consecutive waveforms of similar morphology are recorded, two cycles are measured per eye, both eyes are examined, and the mean of the two eyes is analysed. First and second peak systolic velocities, their ratio, end-diastolic velocity, resistance index, pulsatility index and mean velocity are recorded. Antihypertensive exposure. The drug, dose and the interval in minutes between the last administration and the Doppler examination are recorded at every timepoint. This addresses the principal methodological limitation of the existing literature, in which no comparative study documented treatment at the time of scanning although these agents are known to modify the waveform acutely. Reproducibility. A random 10% sample of examinations is independently re-read by a second blinded operator. Sample size. The calculation is driven by the precision of the sensitivity estimate for the primary outcome. Assuming a sensitivity of 80% estimated with a 95% confidence interval half-width of 10%, 62 women with the outcome are required; at an anticipated prevalence of persistent hypertension of 30%, this requires 207 women assessed at six weeks, and 244 to be enrolled allowing for 15% attrition. The comparison group of 122 normotensive women gives 90% power at alpha 0.05 to detect a standardised mean difference of 0.40, well below the effects reported in the existing comparative studies.