Ambulatory blood pressure monitoring

Pregnancy-induced hypertension is the second cause of maternal mortality worldwide (14% of all deaths) [1], the first in Europe [2] as well as one of the principal causes of fetal or neonatal complications such as intrauterine growth restriction, small for gestational age neonate, preterm birth and intrauterine and perinatal death [3]. The principal pathway leading to these complications is through the development of preeclampsia, which in some cases evolves into eclampsia and hemolysis, elevated liver enzymes and low platelet count syndrome (HELLP), the latter being, potentially life-threatening conditions. Ethical and practical issues make research on hypertensive disorders in pregnancy a difficult task, and, consequently, the data available to support recommendations regarding diagnosis, follow-up and treatment of hypertensive disorders in pregnancy are quite limited. In fact, even the blood pressure (BP) thresholds for diagnosing hypertension in pregnancy and for initiating antihypertensive treatment remain still largely debated [4,5]. The scenario is made even more complex by the ongoing discussion on the actual role in this setting of out-of-office BP monitoring methods including 24-h ambulatory BP monitoring (ABPM) and home BP monitoring (HBPM). These techniques have been demonstrated to improve the diagnostic approach and the risk prediction in a general hypertensive population, and evidence is available also to support their usefulness in the management of hypertensive disorders in pregnancy [6,7]. However, the available studies do not provide specific indications for use of either approach in these conditions and do not clarify which of them should be preferred and in what situation. One of the major implications of out-of-office BP measurement in clinical practice is related to the possibility to identify hypertension phenotypes characterized by a discrepancy between office and out-of-office BP levels. The more common situation is that of elevated office BP with normal out-of-office values (known as white-coat or isolated office hypertension), which seems to be associated with better outcomes both in the general hypertensive population and in pregnancy as compared with sustained (both in office and in out-of-office conditions) hypertension [8]. Indeed, ABPM has been specifically recommended for the identification of white-coat hypertension in pregnancy. However, an opposite situation may also occur when out-of-office values are elevated but office BP is normal. This condition, known as masked hypertension, has been shown to be associated with elevated cardiovascular risk in nonpregnant women, either if untreated or under antihypertensive treatment (in the latter case the more appropriate denomination is masked uncontrolled hypertension), the risk level being comparable with sustained (both office and out-of-office) hypertension [9]. In particular, when identified by means of ABPM, the presence of masked hypertension may be related to the elevation in daytime BP, night-time BP or both. Although previous studies did already suggest that nocturnal BP may be relevant in pregnancy [10], the overall relevance of masked hypertension phenomenon in pregnancy is largely unknown. The article by Salazar et al.[11] in the current issue of the Journal of Hypertension sheds some more light on this complex issue. The authors performed ABPM in 87 women with high-risk pregnancy at 20 or more weeks of gestation in whom conventional office BP was normal (<140/90 mmHg). The women were then classified according to the ABPM result as being normotensive (both daytime and night-time BP within normal limits), with isolated diurnal hypertension, isolated nocturnal hypertension or isolated diurnal and nocturnal hypertension. The latter three conditions could be hence classified as masked hypertension. The outcome of pregnancy was then assessed, with one out of four participants who developed preeclampsia or eclampsia. This elevated prevalence should not be surprising, when considering the high-risk profile of the participants at entry. The authors found that despite their normal office BP, women with elevated ambulatory BP values had an increased risk of adverse pregnancy outcome: almost three-fold for diurnal hypertension, almost five-fold for nocturnal hypertension and over seven-fold when both diurnal and nocturnal BP were elevated. The value of nocturnal BP in this regard is supported by a greater area under the receiver-operating characteristics curves (AUC: 0.77 for SBP and 0.80 for DBP), whereas AUC for daytime BP was lower (0.72 and 0.71 for SBP and DBP, respectively) and similar to that of office BP. These findings appear particularly relevant considering that nondipping circadian BP profiles are common in preeclampsia [12]. The conclusions of the study are, however, in evident contrast with previous statements, according to which ‘recognizing sleep hypertension in women with gestational hypertension or preeclampsia appears to be of little practical clinical value, because even though these women have a tendency to more adverse clinical outcomes, this risk is apparent through conventional daytime BP measurement’ [13]. These discrepant views might be related to the fact that in the study from which this conclusion was drawn [10], the population was unselected – it included women with preeclampsia, gestational hypertension and essential hypertension, as well as those with elevated office BP. Thus, at variance with the study of Salazar et al.[11], the effect of nocturnal hypertension could not be easily discriminated from that of diurnal or office hypertension. The pathophysiological reasons for selective elevation of nocturnal BP in participants prone to preeclampsia are unclear, and their assessment was beyond the scope of the present study. However, the analogy to other conditions, in which a similar pattern is frequently observed, such as renal failure, autonomic dysfunction, obstructive sleep apnea or hypoxia exposure, suggests the involvement of sympathetic overactivity during sleep and possibly of alteration in renal sodium handling [14] associated with fluid retention, the latter known to be present also in preeclampsia. Several aspects of the study by Salazar et al. merit additional comment. Despite the apparently modest sample size, the number of adverse outcomes was high, owing to the inclusion of high-risk participants, which made the power of the study adequate. ABPM was performed with a device validated in a normal population and according to a high-quality protocol (elevated frequency of measurements and day/night definition based on patient diary), although it should be considered that discrepant results regarding the validity of this device were obtained in pregnancy [15,16]. Furthermore, the prevalence of masked hypertension and isolated nocturnal hypertension in this article is much higher than in nonpregnant women, possibly due to the particular inclusion criteria used in this study (high-risk pregnancy combined with normal office BP) [11]. In spite of the interesting results, however, this study also has some limitations that should be mentioned. First, hypertension status was defined on the basis of a single session of office measurements. Although this approach largely reflects clinical practice, still, the findings may have been different if office BP were measured, according to current guidelines [17], on at least two occasions, considering how variable office BP is. This is relevant as office BP values in the study participants with masked hypertension, albeit within normal limits, were higher than in participants with normal ambulatory BP values (not surprisingly, considering that the distribution of BP values is continuous and higher ambulatory BP is typically associated also with higher office BP). Hence, it cannot be excluded that some of the participants, whose office BP was close to the threshold for office hypertension diagnosis, could have had office BP elevated at a further visit. Even in such a case, however, the conclusion of the authors on the usefulness of ABPM would not be undermined, as this technique would in any case allow an earlier detection (and treatment) of the elevated risk of preeclampsia, compared with office BP. Second, the quality of sleep was not reported. Therefore, it is not clear whether the high frequency of night-time measurements could have contributed to the nocturnal hypertension through sleep disturbance. Third, no information is available on the presence of some potential confounders such as obesity or sleep-related breathing disorders, in particular obstructive sleep apnea, either of which could selectively increase nocturnal BP while, at the same time, affecting pregnancy outcome [18,19]. Notwithstanding these relatively minor limitations, the article by Salazar et al. may have important implications in the management of hypertensive disorders in pregnancy, should its results be reproduced in other studies. The finding that nocturnal hypertension is a strong predictor of preeclampsia development in high-risk women, despite normal office BP, may boost a routine use of ABPM in the critical phase of pregnancy (around the 30th week) to identify women likely to progress to preeclampsia. In particular, based on this finding, ABPM should be preferred to HBPM, an approach which, despite having several studies to support its prognostic value in pregnancy, is at present inherently unable to provide information on nocturnal BP, at least with the commonly available device technology. Clearly, a demonstration that earlier initiation of antihypertensive treatment (or other therapeutic measures) in these women may improve maternal and/or fetal outcomes would be the next step required to change current practice. Obviously, considering the characteristics of the participants of the present study, the conclusions remain restricted to high-risk pregnancies unless further evidence becomes available also in normal pregnancies. In any case, the study by Salazar et al. is a much welcome contribution to improve clinical management in a difficult but clinically highly relevant field such as hypertensive disorders in pregnancy. ACKNOWLEDGEMENTS Conflicts of interest There are no conflicts of interest.