Hypertensive disorders of pregnancy complicate 5%-10% of pregnancies and are a leading cause of maternal and perinatal mortality and morbidity. Treatment with antihypertensive medications is intended to prevent adverse maternal and infant outcomes. However, there is no clear consensus regarding the benefit of treatment for mild to moderate gestational hypertension. The maternal/fetal risks of no treatment, such as possible progression to severe hypertension and its associated consequences, have not been shown to clearly outweigh the fetal risks of treatment with antihypertensive medications, which may include intrauterine growth restriction and other neonatal complications.

A recent study published on-line in May in the BJOG: An International Journal of Obstetrics & Gynaecology suggests that the decision to treat on treatment of mild to moderate hypertension should include consideration of possible long-term neurobehavioral consequences for the child (BJOG 2010 [doi: 10.1111/j.1471-0528.2010.02568.x]). In an hypothesis-generating historical cohort study conducted in the Netherlands, the authors identified 202 singleton children born in 1 of 12 hospitals between 1983 and 1987, whose mothers had developed pregnancy-induced or pregnancy-aggravated hypertension and were treated with either methyldopa (61), labetalol (58), or bed rest (83). The children underwent a battery of tests to measure IQ, gross motor development, fine motor development, and memory between approximately ages 4 and 9 years. In addition, parents and teachers were asked to evaluate the child’s behavior.

Overall, mean scores on most areas of functional development did not differ significantly between the groups. However, children prenatally exposed to labetalol were about four times more likely to exhibit characteristics of ADHD than were children in the bed rest group based on a standardized Dutch version of the Teacher Report Form (OR 4.1). Children in the labetalol group were also more likely to exhibit these behaviors than were children in the methyldopa group but not significantly so (OR 2.3). Odds ratios were not adjusted for other factors because of the small number of children in each group who were classified as ADHD. The authors suggest that there is biological plausibility for the effect of prenatal exposure to labetalol on subsequent attention and hyperactivity in primary school children, and that this effect could be mediated by drug-induced fetal growth restriction and neonatal beta blockade.

This interesting study illustrates two critically important points: The first is the difficulty in conducting observational studies of prenatal medication exposure and long-term neurobehavioral outcomes, and the second is the importance of doing these studies in the first place. With respect to the former, even under the best of circumstances, without a randomized controlled trial it is very difficult to account for differences inherent in the three groups in the Dutch study. These include differences between groups in maternal overweight or obesity, tobacco use, preterm or very preterm delivery, infants born small for gestational age, maternal stress, other drug use, etc., all of which may contribute to risk for ADHD. Severity of the underlying maternal condition as measured by highest diastolic blood pressure, as well as gestational age at which treatment was initiated, varied by group.

Furthermore, differences in age at which the child was tested could have influenced the prevalence of ADHD-like symptoms that were likely to be identified by teacher report. And finally, the study was conducted during a period in time when standards of clinical practice were in transition in terms of which medication the obstetrician chose to use for treatment, if any. This common occurrence can lead to “channeling” of patients with certain characteristics to treatment with one or the other drug, which can carry with it inherent underlying differences in patients that are potentially confounding with respect to the outcome.

Nevertheless, these kinds of studies need to be done. Just as there is a need for systematic postmarketing studies for drug safety with respect to risk for birth defects, there is an equally important need for systematic surveillance for neurobehavioral outcomes. Improved efforts are needed to carefully match comparison groups on key maternal and child characteristics and to address the growing number of potential environmental factors that accumulate the longer the period of time to follow-up developmental assessment. Study designs that involve sufficient sample size to generate enough power to evaluate the outcomes of interest, although difficult to come by, are needed. All of these issues call for a systematic coordinated approach to evaluating long-term functional outcomes following prenatal exposures, which in the end may have the most potential public health importance.

This article appeared as the column “Drugs, Pregnancy and Lactation” in Ob.Gyn. News, an Elsevier publication. Dr. Chambers is associate professor of pediatrics and family and preventive medicine at the University of California, San Diego. She is director of the California Teratogen Information Service and Clinical Research Program. Dr. Chambers is a past president of the Organization of Teratology Information Specialists (www.otispregnancy.org) and past president of the Teratology Society. She had no conflicts to disclose related to the topic of this column. To comment, e-mail her at .

Copyright (c) 2009 Elsevier Global Medical News. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.

约有5%~10%的孕妇患有妊娠期高血压，这是导致孕产妇及围产儿死亡及患病的一个重要因素。通常可采用降压药治疗来预防产妇和婴儿出现不良情况。不过，目前对于轻中度妊娠高血压是否需进行药物治疗并未达成一致意见。不予治疗会对孕产妇/胎儿带来一定风险，如可能发展为严重高血压，并导致一些不良后果；而采用降压药治疗则会对胎儿带来一定风险，可能会抑制胎儿在子宫内的发育，导致其他新生儿并发症。两者均无明显优势。

 

五月份的《国际妇产科学杂志》(BJOG: An International Journal of Obstetrics & Gynaecoloy)在线发表的一项最新研究表明，在决定对轻中度高血压进行药物治疗前，应考虑到治疗可能对胎儿产生的长期神经行为影响(BJOG 2010 [doi: 10.1111/j.1471-0528.2010.02568.x])。在荷兰的一项历史队列假说推理研究中，研究者对1983~1987年间于12家医院出生的202名独生子女进行研究，这些孩子的母亲曾有妊娠期高血压或妊娠合并高血压，分别接受甲基多巴(61例)、拉贝洛尔(58例)及卧床休息(83例)治疗。大约在孩子4~9岁时，对他们进行一系列的IQ、粗大运动功能发育、精细运动功能发育及记忆力测试。另外，请父母及老师对孩子的行为进行评估。

 

总体而言，各组之间大部分功能发育的平均得分没有显著差异。不过，根据标准的Dutch教师评估量表结果，出生前接触过拉贝洛尔的孩子更容易出现多动症(ADHD)特征，发生率是卧床休息组孩子的4倍(OR 4.1)。同样出生前接触拉贝洛尔的儿童比甲基多巴组孩子出现这些行为的几率要高，只是并不如此显著(OR 2.3)。由于每组出现ADHD的孩子数目很小，所以未对研究中的优势比进行其他因素校正。作者认为，产前接触拉贝洛尔对于孩子以后注意力、多动的影响存在一定的生物学机理，而且这种影响可能是通过药物诱导的胎儿发育抑制及新生儿β受体阻断产生的。

 

该研究说明了两个关键点：第一，对产前服药及其长期神经行为后果影响进行观察性研究存在难度；第二，优先进行这些研究很重要。对于前者，即使是最好的环境，没有随机对照试验，也很难评估Dutch研究中3个组固有的差别。这些组间差别包括妊娠期是否超重或肥胖、是否吸烟、是否早产、相对于妊娠期而言婴儿出生时体重大小、母亲是否紧张、是否服用其他药物等等，所有这些都可能会增加孩子的ADHD风险。应严格记录母亲的状况，如测量最大舒张压、怀孕第几周开始治疗等，并列出组间差别。

 

另外，孩子接受测试时的年龄差异也会对ADHD症状的评估产生影响，有些可能被老师误报告为ADHD症状。最后，在该研究进行的时间段内，产科医生治疗用药的标准也在发生变化。这些常见的情况会使得存在某些特定症状的患者接受某种药物治疗或其他药物治疗，而这种病人之间的差异可能会混淆结果。

 

不过，仍需要进行这些研究。因为考虑到致畸风险，需要对药物安全性进行系统的上市后研究，同时也需要对药物的神经行为影响进行系统的监测。需要对对照组孕产妇及婴儿的重要特征进行详细比对，并关注不断增加的潜在环境因素数目，注意这些因素在长时期内对发育评估的影响。虽然增加样本量后研究难度会增大，但研究设计时仍要包括足够的样本量，从而产生足够的结果评估。需要对所有因素进行系统协调，以评估妊娠期药物治疗后的长期功能发育影响，这些研究最终会对公共卫生产生重大影响。

 

爱思唯尔  版权所有