First trimester screening for pre-eclampsia and targeted aspirin prophylaxis: a cost-effectiveness cohort study

OBJECTIVE

Investigate cost-effectiveness of first trimester pre-eclampsia screening using the Fetal Medicine Foundation (FMF) algorithm and targeted aspirin prophylaxis in comparison with standard care.

DESIGN

Retrospective observational study.

SETTING

London tertiary hospital.

POPULATION

5957 pregnancies screened for pre-eclampsia using the National Institute for Health and Care Excellence (NICE) method.

METHODS

Differences in pregnancy outcomes between those who developed pre-eclampsia, term pre-eclampsia and preterm pre-eclampsia were compared by the Kruskal-Wallis and Chi-square tests. The FMF algorithm was applied retrospectively to the cohort. A decision analytic model was used to estimate costs and outcomes for pregnancies screened using NICE and those screened using the FMF algorithm. The decision point probabilities were calculated using the included cohort.

MAIN OUTCOME MEASURES

Incremental healthcare costs and QALY gained per pregnancy screened.

RESULTS

Of 5957 pregnancies, 12.8% and 15.9% were screen-positive for development of pre-eclampsia using the NICE and FMF methods, respectively. Of those who were screen-positive by NICE recommendations, aspirin was not prescribed in 25%. Across the three groups, namely, pregnancies without pre-eclampsia, term pre-eclampsia and preterm pre-eclampsia there was a statistically significant trend in rates of emergency caesarean (respectively 21%, 43% and 71.4%; P < 0.001), admission to neonatal intensive care unit (NICU) (5.9%, 9.4%, 41%; P < 0.001) and length of stay in NICU. The FMF algorithm was associated with seven fewer cases of preterm pre-eclampsia, cost saving of £9.06 and QALY gain of 0.00006/pregnancy screened.

CONCLUSIONS

Using a conservative approach, application of the FMF algorithm achieved clinical benefit and an economic cost saving.

Chronic hypertension and superimposed preeclampsia: screening and diagnosis

Superimposed preeclampsia complicates about 20% of pregnancies in women with chronic hypertension and is associated with increased maternal and perinatal morbidity compared with preeclampsia alone. Distinguishing superimposed preeclampsia from chronic hypertension can be challenging because, in chronic hypertension, the traditional criteria for the diagnosis of preeclampsia, hypertension, and significant proteinuria can often predate the pregnancy. Furthermore, the prevalence of superimposed preeclampsia is unlikely to be uniformly distributed across this high-risk group but is related to the severity of preexisting endothelial dysfunction. This has led to interest in identifying biomarkers that could help in screening and diagnosis of superimposed preeclampsia and in the stratification of risk in women with chronic hypertension. Elevated levels of uric acid and suppression of other renal biomarkers, such as the renin-angiotensin aldosterone system, have been demonstrated in women with superimposed preeclampsia but perform only modestly in its prediction. In addition, central to the pathogenesis of preeclampsia is a tendency toward an antiangiogenic state thought to be triggered by an impaired placenta and, ultimately, contributing to the endothelial dysfunction pathognomonic of the disease. In the general obstetrical population, angiogenic factors, such as soluble fms-like tyrosine kinase-1 and placental growth factor, have shown promise in the prediction of preeclampsia. However, soluble fms-like tyrosine kinase-1 and placental growth factor are impaired in women with chronic hypertension irrespective of whether they develop superimposed preeclampsia. Therefore, the differences in levels are less discriminatory in the prediction of superimposed preeclampsia compared with the general obstetrical population. Alternative biomarkers to the angiogenic and renal factors include those of endothelial dysfunction. A characteristic of both preeclampsia and chronic hypertension is an exaggerated systemic inflammatory response causing or augmenting endothelial dysfunction. Thus, proinflammatory mediators, such as tumor necrosis factor-α, interleukin-6, cell adhesion molecules, and endothelin, have been investigated for their role in the screening and diagnosis of superimposed preeclampsia in women with chronic hypertension. To date, the existing limited evidence suggests that the differences between those who develop superimposed preeclampsia and those who do not are, as with angiogenic factors, also modest and not clinically useful for the stratification of women with chronic hypertension. Finally, pro-B-type natriuretic peptide is regarded as a sensitive marker of early cardiac dysfunction that, in women with chronic hypertension, may predate the pregnancy. Thus, it has been proposed that pro-B-type natriuretic peptide could give insight as to the ability of women with chronic hypertension to adapt to the hemodynamic requirements of pregnancy and, subsequently, their risk of developing superimposed preeclampsia. Although higher levels of pro-B-type natriuretic peptide have been demonstrated in women with superimposed preeclampsia compared with those without, current evidence suggests that pro-B-type natriuretic peptide is not a predictor for the disease. The objectives of this review are to, first, discuss the current criteria for the diagnosis of superimposed preeclampsia and, second, to summarize the evidence for these potential biomarkers that may assist in the diagnosis of superimposed preeclampsia.

A comparison of technicques to disimpact the fetal head on a second stage caesearean simulator

Background

The rate of second stage caesarean section (CS) is rising with associated increases in maternal and neonatal morbidity, which may be related to impaction of the fetal head in the maternal pelvis. In the last 10 years, two devices have been developed to aid disimpaction and reduce these risks: the Fetal Pillow (FP) and the Tydeman Tube (TT). The aim of this study was to determine the distance of upward fetal head elevation achieved on a simulator for second stage CS using these two devices, compared to the established technique of per vaginum digital disimpaction by an assistant.

Methods

We measured elevation of the fetal head achieved with the two devices (TT and FP), compared to digital elevation, on a second stage Caesearean simulator (Desperate Debra ™ set at three levels of severity. Elevation was measured by both a single operator experienced with use of the TT and FP and also multiple assistants with no previous experience of using either device. All measurements were blinded

Results

The trained user achieved greater elevation of the fetal head at both moderate and high levels of severity with the TT (moderate: 30mm vs 12.5mm p<0.001; most severe: 25mm vs 10mm p<0.001) compared to digital elevation. The FP provided comparable elevation to digital at both settings (moderate: 10 vs 12.5mm p=0.149; severe 10 vs 10mm p=0.44).

With untrained users, elevation was also significantly greater with the TT compared to digital elevation (20mm vs 10mm p<0.01). However digital disimpaction was significantly greater than the FP (10mm vs 0mm p<0.0001).

Conclusion

On a simulator, with trained operators, the TT provided greater fetal head elevation than digital elevation and the FP. The FP achieved similar elevation to the digital technique, especially when the user was trained in the procedure.

X-linked sideroblastic anaemia in a female fetus: a case report and a literature review

Background

X-linked sideroblastic anaemia (XLSA) is commonly due to mutations in the ALAS2 gene and predominantly affects hemizygous males. Heterozygous female carriers of the ALAS2 gene mutation are often asymptomatic or only mildly anaemic. XLSA is usually characterized by microcytic erythrocytes (reduced mean corpuscular volume (MCV)) and hypochromia, along with increased red cell distribution width. However, in females with XLSA the characteristic laboratory findings can be dimorphic and present with macrocytic (elevated MCV) in addition to microcytic red cells.

Case presentation

We report a case of fetal anaemia, presenting in the early third trimester of pregnancy, in a female fetus. Ultrasound findings at 29 weeks were of cardiomegaly, prominent umbilical veins, a small rim of ascites, and mean cerebral artery peak systolic velocity (PSV) value above 1.5 Multiples of the Median (MoM). She underwent non-invasive prenatal testing that determined the rhesus genotype of the fetus to be rhesus B negative. No red blood cell antibodies were reported. Other investigations to determine the underlying cause of fetal anaemia included microarray comparative genomic hybridization, serology to exclude congenital infection and a peripheral blood film and fetal bilirubin to detect haemolysis. The maternal grandmother had a history of sideroblastic anaemia diagnosed at the age of 17 years. The mother had mild macrocytic anaemia with haemoglobin of 10.4 g/dl and MCV of 104 fl. The fetal anaemia was successfully treated with two in utero transfusions (IUTs), and delivery occurred via caesarean section at 37 weeks of gestation. The red cell gene sequencing in both the mother and fetus were heterozygous for an ALAS2 mutation causing in utero manifestations of XLSA. The haemoglobin on discharge to the local hospital at five days of age was 19.1 g/dl. Subsequently, the infant became anaemic, requiring regular 3–4 monthly blood transfusions and demonstrating overall normal development. Her anaemia was unresponsive to pyridoxine.

Conclusions

This is one of four cases reporting multiple female members presenting with discordant clinical features of XLSA from being entirely asymptomatic to hydropic in utero. Our report is novel in that there are no previous cases in the literature of anaemia in a female fetus heterozygous for ALAS2 mutation.

First trimester angiogenic and inflammatory factors in women with chronic hypertension and impact of blood pressure control: a case-control study

OBJECTIVES

To assess first trimester serum placental growth factor (PLGF), soluble fms-like tyrosine kinase-1 (sFLT-1), interleukin-6 (IL-6), tumour necrosis factor-α (TNF-α), endothelin and vascular cell adhesion molecule (VCAM) in women with chronic hypertension (CH) stratified according to blood pressure (BP) control.

DESIGN

Case-control.

SETTING

Tertiary referral centre.

POPULATION

650 women with CH, 142 normotensive controls.

METHODS

In the first trimester, patients with CH were subdivided into four groups. Group 1 included women without pre-pregnancy CH presenting with BP ≥140/90 mmHg. Groups 2-4 had pre-pregnancy CH; in group 2 the BP was <140/90 mmHg without antihypertensive medication, in group 3 the BP was <140/90 mmHg with antihypertensive medication, and in group 4 the BP was ≥140/90 mmHg despite antihypertensive medication. PLGF, sFLT-1, IL-6, TNF-α, endothelin and VCAM were measured at 11⁺⁰ -13⁺⁶  weeks' gestation and converted into multiples of the expected median (MoM) using multivariate regression analysis in the controls.

MAIN OUTCOME MEASURE

Comparisons of MoM values of PLGF, sFLT-1, endothelin, IL-6, TNF-α and VCAM between the entire cohort of women with CH and the control group were made using Student's t-test or Mann-Whitney U-test. Comparisons between the four CH groups were made using analysis of variance or Kruskal-Wallis tests.

RESULTS

Compared with the control group, women with CH had significantly lower MoM of PLGF, sFLT-1 and IL-6 and a significantly higher MoM of endothelin. Between the four groups of women with CH, there were no significant differences in the MoM of sFLT-1, PLGF, sFLT-1/PLGF ratio, endothelin, IL-6 or VCAM, or in the levels of TNF- α.

CONCLUSION

In women with CH, differences exist in first trimester angiogenic and inflammatory profiles when compared with normotensive pregnancies. However, these differences do not assist in the stratification of women with CH to identify those with more severe underlying disease and worse pregnancy outcomes.

TWEETABLE ABSTRACT

First trimester blood pressure control impacts on serum PLGF, sFLT-1, endothelin and IL-6 in women with chronic hypertension.

Chronic hypertension in pregnancy stratified by first-trimester blood pressure control and adverse perinatal outcomes: A prospective observational study

INTRODUCTION

The aim of this study was to assess perinatal outcomes in women with chronic hypertension (CH) stratified into four groups according to their blood pressure (BP) control in the first trimester of pregnancy.

MATERIAL AND METHODS

This was a prospective cohort study between January 2011 and June 2017, based in a university hospital in London, UK. The population consisted of four groups: group 1 included women without history of CH, presenting in the first trimester with BP >140/90 mmHg (n = 100). Groups 2-4 had prepregnancy CH; group 2 had BP <140/90 mmHg without antihypertensives (n = 234), group 3 had BP <140/90 mmHg with antihypertensives (n = 272), and group 4 had BP ≥140/90 mmHg despite antihypertensives (n = 194). The main outcome measures were: fetal growth restriction, admission to neonatal (NNU) or neonatal intensive care unit (NICU) for ≥2 days, composite neonatal morbidity, and composite serious adverse neonatal outcome. Outcomes were collected from the hospital databases and for up to 6 weeks postnatally. Differences between groups were assessed using chi-squared test and multivariate logistic regression was used to assess the independent contribution of the four groups to the prediction of pertinent outcomes, after controlling for maternal characteristics.

RESULTS

There was a higher incidence of fetal growth restriction in groups 3 (17.6%) and 4 (18.2%), compared with groups 1 (10.0%) and 2 (11.1%) (P = .04). There were more admissions to the NNU for ≥2 days in groups 3 (23.2%) and 4 (25.0%), compared with groups 1 (17.0%) and 2 (13.2%) (P = .008); and more admissions to NICU for ≥2 days in groups 3 (9.2%) and 4 (9.4%), compared with groups 1 (3.0%) and 2 (3.4%) (P = .01). Composite neonatal morbidity was higher in groups 3 (22.4%) and 4 (21.4%), compared with groups 1 (17.0%) and 2 (11.5%) (P = .009). Composite serious adverse postnatal outcome was higher in groups 3 (3.3%) and 4 (4.2%), compared with groups 1 (1.0%) and 2 (0.9%) but the difference did not reach statistical significance (P = .09). These results were also observed when values were adjusted for maternal characteristics.

CONCLUSIONS

In CH adverse perinatal outcomes are worse in women who are known to have CH and need antihypertensives in the first trimester of pregnancy. Women with newly diagnosed CH in the first trimester have similar outcomes to those with known CH who have antihypertensive treatment discontinued.

First trimester inflammatory mediators in women with chronic hypertension

INTRODUCTION

Chronic hypertension complicates 1%-2% of pregnancies and is one of the most significant risk factors for the development of preeclampsia. Inflammatory mediators, such as interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), vascular cell adhesion molecule (VCAM) and endothelin have been implicated in the endothelial dysfunction that is pathognomonic of preeclampsia and may serve as useful first trimester biomarkers for the prediction of preeclampsia. The objectives of this study are: first, to investigate differences in serum levels of IL-6, TNF-α, VCAM and endothelin at 11⁺⁰ to 13⁺⁶ weeks' gestation in women with chronic hypertension who developed superimposed preeclampsia with those who did not and normotensive controls and, second, to evaluate the performance of these biomarkers in the prediction of preeclampsia.

MATERIAL AND METHODS

The study population was comprised of 650 women with chronic hypertension, including 202 who developed superimposed preeclampsia and 448 who did not, and 142 normotensive controls matched to the chronic hypertension group for storage time and racial origin. Serum concentrations of IL-6, TNF-α, VCAM and endothelin were measured and the values were converted into multiples of the expected median using multivariate regression analysis in the control group. The multiples of the median values of the biomarkers between the two groups of women with chronic hypertension and the controls were compared, and the receiver operating characteristic curve (ROC) was used to assess the performance of these variables for the prediction of preeclampsia.

RESULTS

In women with chronic hypertension, compared with the normotensive controls, there was a significantly higher first trimester median concentration of endothelin but not of VCAM, IL-6 or TNF-α. Within the cohort of women with chronic hypertension, those who developed superimposed preeclampsia, compared with those who did not, had higher first trimester serum concentration of VCAM but not of endothelin, IL-6 or TNF-α. However, serum VCAM provided a poor prediction of superimposed preeclampsia (area under the ROC curve 0.537, 95% CI 0.487-0.587).

CONCLUSIONS

Women with chronic hypertension have increased serum endothelin in the first trimester of pregnancy and those who develop superimposed preeclampsia have higher levels of VCAM. None of the inflammatory mediators performed well in the first trimester in the prediction of preeclampsia.

First trimester serum angiogenic and anti-angiogenic factors in women with chronic hypertension for the prediction of preeclampsia

BACKGROUND

An imbalance between angiogenic and antiangiogenic factors is thought to be a central pathogenetic mechanism in preeclampsia. In pregnancies that subsequently experience preeclampsia, the maternal serum concentration of the angiogenic placental growth factor is decreased from as early as the first trimester of pregnancy, and the concentration of the antiangiogenic soluble fms-like tyrosine kinase-1 is increased in the last few weeks before the clinical presentation of the disease. Chronic hypertension, which complicates 1-2% of pregnancies, is the highest risk factor for the development of preeclampsia among all other factors in maternal demographic characteristics and medical history. Two previous studies in women with chronic hypertension reported that first-trimester serum placental growth factor and soluble fms-like tyrosine kinase-1 levels were not significantly different between those who experienced superimposed preeclampsia and those who did not, whereas a third study reported that concentrations of placental growth factor were decreased.

OBJECTIVE

The purpose of this study was to investigate whether, in women with chronic hypertension, serum concentrations of placental growth factor and soluble fms-like tyrosine kinase-1 and soluble fms-like tyrosine kinase-1/placental growth factor ratio at 11⁺⁰-13⁺⁶ weeks gestation are different between those women who experienced superimposed preeclampsia and those who did not and to compare these values with those in normotensive control subjects.

STUDY DESIGN

The study population comprised 650 women with chronic hypertension, which included 202 women who experienced superimposed preeclampsia and 448 women who did not experience preeclampsia, and 142 normotensive control subjects. Maternal serum concentration of placental growth factor and soluble fms-like tyrosine kinase-1 were measured by an automated biochemical analyzer and converted into multiples of the expected median with the use of multivariate regression analysis in the control group. Comparisons of placental growth factor and soluble fms-like tyrosine kinase-1 levels and soluble fms-like tyrosine kinase-1/placental growth factor ratio in multiples of the expected median values between the 2 groups of chronic hypertension and the control subjects were made with the analysis of variance or the Kruskal-Wallis test.

RESULTS

In the group of women with chronic hypertension who experienced preeclampsia compared with those women who did not experience preeclampsia, there were significantly lower median concentrations of serum placental growth factor multiples of the expected median (0.904 [interquartile range, 0.771-1.052] vs 0.948 [interquartile range, 0.814-1.093]; P=.014) and soluble fms-like tyrosine kinase-1 multiples of the expected median (0.895 [interquartile range, 0.760-1.033] vs 0.938 [interquartile range, 0.807-1.095]; P=.013); they were both lower than in the normotensive control subjects (1.009 [interquartile range, 0.901-1.111] and 0.991 [interquartile range, 0.861-1.159], respectively; P<.01 for both). There were no significant differences among the 3 groups in soluble fms-like tyrosine kinase-1/placental growth factor ratios. In women with chronic hypertension, serum placental growth factor and soluble fms-like tyrosine kinase-1 levels provided poor prediction of superimposed preeclampsia (area under the curve, 0.567 [95% confidence interval, 0.537-0.615] and 0.546 [95% confidence interval, 0.507-0.585], respectively).

CONCLUSION

Women with chronic hypertension, and particularly those who subsequently experienced preeclampsia, have reduced first-trimester concentrations of both placental growth factor and soluble fms-like tyrosine kinase-1.

Severe hypertension, preeclampsia and small for gestational age in women with chronic hypertension diagnosed before and during pregnancy

OBJECTIVES

To compare rates of severe hypertension (SH), preeclampsia (PE) birth of small for gestational age (SGA) neonates between women with chronic hypertension (CH) diagnosed during the first trimester of pregnancy and those with pre-pregnancy CH.

STUDY DESIGN

Prospective cohort study of women with CH and singleton pregnancies referred to an Antenatal Hypertension Clinic at 8-14 weeks' gestation. At presentation the patients were subdivided into four groups based on blood pressure (BP) control. Group 1 included women without a preceding history of CH presenting with BP of ≥140/90 mmHg (n = 86). Groups 2-4 had pre-pregnancy CH; in group 2 the BP was <140/90 mmHg without antihypertensive medication (n = 200), in group 3 the BP was <140/90 mmHg with antihypertensive medication (n = 231) and in group 4 the BP was ≥140/90 mmHg despite antihypertensive medication (n = 173).

MAIN OUTCOME MEASURES

PE, SH (BP ≥ 160/110 mmHg), SGA (birthweight < 10th percentile).

RESULTS

In group 1, the rate of SH (15.1%), was similar to that in group 2 (10.5%) and group 3 (23.8%) but significantly lower than in group 4 (52.6%). In group 1, the rate of PE (12.8%) and SGA <10th centile (18.6%) were similar to those in group 2 (16.5% and 21.0%) and significantly lower than in group 3 (26.0 and 30.7%) and group 4 (26.6% and 31.8).

CONCLUSION

In women diagnosed with CH in the first trimester of pregnancy, the rates of SH, PE and SGA are similar to those with pre-pregnancy CH who present with BP below 140/90 without the need for antihypertensive medication.

Chronic hypertension: first-trimester blood pressure control and likelihood of severe hypertension, preeclampsia, and small for gestational age

BACKGROUND

There is extensive evidence that prepregnancy chronic hypertension is associated with a high risk of development of severe hypertension and preeclampsia and birth of small-for-gestational-age neonates. However, previous studies have not reported whether antihypertensive use, blood pressure control, or normalization of blood pressure during early pregnancy influences the rates of these pregnancy complications.

OBJECTIVE

The purpose of this study was to stratify women with prepregnancy chronic hypertension according to the use of antihypertensive medications and level of blood pressure control at the first hospital visit during the first trimester of pregnancy and to examine the rates of severe hypertension, preeclampsia, and birth of small-for-gestational-age neonates according to such stratification.

STUDY DESIGN

We conducted a prospective study of 586 women with prepregnancy chronic hypertension, in the absence of renal or liver disease, that was booked at a dedicated clinic for the management of hypertension in pregnancy. The patients had singleton pregnancies and were subdivided according to findings in their first visit: group 1 (n=199), blood pressure <140/90 mm Hg without antihypertensive medication; group 2 (n=220), blood pressure <140/90 mm Hg with antihypertensive medication; and group 3 (n=167), systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg, despite antihypertensive medication. In the subsequent management of these pregnancies, our policy was to maintain the blood pressure at 130-140/80-90 mm Hg with the use of antihypertensive medication; antihypertensive drugs were stopped if the blood pressure was persistently <130/80 mm Hg. The outcome measures were severe hypertension (systolic blood pressure ≥160 mm Hg and/or diastolic blood pressure ≥110 mm Hg), preterm and term preeclampsia (in addition to hypertension at least 1 of renal involvement, liver impairment, neurologic complications, or thrombocytopenia), and birth of small-for-gestational-age neonates (birthweight <5th percentile for gestational age). The incidence of these complications was compared in the 3 strata.

RESULTS

The median gestational age at presentation was 10.0 weeks (interquartile range, 9.1-11.0 weeks). In groups 2 and 3, compared with group 1, there was a significantly higher body mass index, incidence of black racial origin, and history of preeclampsia in a previous pregnancy. There was a significant increase from group 1 to group 3 in the incidence of severe hypertension (10.6%, 22.2%, and 52.1%), preterm preeclampsia with onset at <37 weeks of gestation (7.0%, 15.9%, and 20.4%), and small for gestational age (13.1%, 17.7%, and 21.1%), but not term preeclampsia with onset at ≥37 weeks of gestation (9.5%, 9.1%, and 6.6%).

CONCLUSIONS

In women with prepregnancy chronic hypertension, the rates of development of severe hypertension, preterm preeclampsia, and small for gestational age are related to the use of antihypertensive medications and the level of blood pressure control at the first hospital visit during the first trimester of pregnancy.

An inaccurate automated device negatively impacts the diagnosis and treatment of gestational hypertension

OBJECTIVES

Automated blood pressure devices are frequently introduced in maternity care without prior validation for their accuracy in pregnancy. Our objectives were to, firstly, establish the accuracy in pregnancy of a locally used device (Welch Allyn 300) and, secondly, to audit its impact on the diagnosis and treatment of hypertension.

STUDY DESIGN

Validation study: The device was evaluated using the grading criteria of the European Society of Hypertension International Protocol (ESH-IP) (2010). Two observers took nine same-arm measurements alternating between the Welch Allyn and the mercury sphygmomanometer. Thirty-three women of any gestation were included. Clinical audit: One observer took three same-arm measurements alternating between the Welch Allyn and the mercury sphygmomanometer. One hundred women of any gestation referred with suspected hypertension were included. The main outcome measures were the proportion diagnosed with hypertension or commenced on anti-hypertensive treatment on the presenting visit when using either the manual or the automated device.

MAIN OUTCOME MEASURES

Grading criteria of the ESH-IP (2010) and proportion of women diagnosed with hypertension or commenced on antihypertensive therapy at the presenting visit when using either manual sphygmomanometry or the Welch Allyn device.

RESULTS

The Welch Allyn 300 series failed to meet the criteria of the ESH-IP (2010) for pregnancy. Compared to the mercury device, it under diagnosed hypertension by 48% and need for treatment by 80%.

CONCLUSIONS

The Welch Allyn 300 cannot be recommended for the measurement of blood pressure in pregnancy. Its use leads to the under-diagnosis and under-treatment of gestational hypertension.

Validation of the Withings BP-800 in pregnancy and impact of maternal characteristics on the accuracy of blood pressure measurement

OBJECTIVE

Firstly, to validate the Withings BP-800 automated device for use in pregnancy and, secondly, to assess the impact of maternal somatometric and demographic variables on the accuracy of the device.

DESIGN

Prospective observational study.

SETTING

Kings College Hospital, London, UK.

POPULATION

Forty-seven women of any gestation.

METHODS

Validation: The British Hypertension Society (BHS) Protocol (1993) was used for the validation of the Withings BP-800. Two trained observers took nine sequential same arm measurements alternating between the Withings BP-800 and the mercury sphygmomanometer. Assessment of factors affecting the disagreement between the two devices: The associations between discrepancies in the measured systolic and diastolic blood pressure by the two devices and potential predictors of discrepancy and/or possible confounders of associations including age, gestational age, ethnicity, body mass index and arm circumference were investigated using two-level mixed effects models to take into account the repeated measurements.

MAIN OUTCOME MEASURES

Accuracy of the Withings BP-800 based on the grading criteria of the BHS Protocol (1993).

RESULTS

The Withings BP-800 failed to meet the validation criteria of the BHS protocol for pregnancy and preeclampsia. Inter-device discrepancy was significantly associated with larger arm circumferences and was more pronounced with diastolic blood pressure. This relationship was independent of other maternal characteristics.

CONCLUSIONS

The Withings BP-800 cannot be recommended for the measurement of blood pressure in pregnancy or preeclampsia. The inaccuracy of the Withings BP-800 increases when used in patients with larger arm circumferences with a propensity to over-read.