Is preterm labor a valid endpoint in perinatal research?
Is preterm labor a valid endpoint in perinatal research?
复制标题
早产是围产期研究的有效终点吗？
DOI：
10.1093/aje/155.4.302
复制
发表时间：
2002
影响因子：
5
通讯作者：
Macones,GeorgeA
中科院分区：
文献类型：
--
作者：
Macones,GeorgeA
在这一期的《美国流行病学杂志》上，达扬等人介绍了他们在一项队列研究中的发现，该研究的目的是评估母亲焦虑、抑郁和早产之间的关系。解开母亲心理事件和妊娠结局之间的关系是非常重要的，特别是当我们开始了解这种联系的生物学合理性时。在这方面，Daan等人。对现有的关于这一主题的文献做出了宝贵的贡献。然而，他们的研究为观察性和干预性围产期研究提出了一个重要的问题：早产本身是否应该被用作终点？我认为在回答这个问题时有几个相关的问题需要考虑。首先，早产本身是临床上重要的健康后果吗？如果不是，它是另一个临床上重要结果的合理替代品吗？Daan等人。认为早产本身是临床上重要的结局，主要基于产妇的考虑，如产妇住院的必要性和宫缩药物的使用。当然，考虑到与怀孕相关的下肢静脉淤滞和高凝状态，卧床休息的孕妇存在血栓形成的风险。此外，最常用的酵解剂(β-激动剂和硫酸镁)都与严重的心肺疾病有关(2-4)。幸运的是，溶栓药物引发血栓和心肺并发症的风险非常低。鉴于早产引起的母体并发症的发生率很低，早产的临床重要性在很大程度上取决于它与早产的关系(早产婴儿患许多并发症的风险大大增加，包括死亡、慢性肺部疾病、脑室出血、坏死性小肠结肠炎等)。如果你同意这一观点，那么下一个问题就是早产是否可以合理地替代早产。这个问题的答案无疑是“不”。正如大燕等人所说。指出，被诊断为早产并接受治疗的妇女通常会在足月分娩。这一事实在许多临床试验中得到了证实(其中使用了相对严格的诊断标准)，在这些试验中，大约50%的妇女在37周或更长时间随机服用安慰剂(5)。在临床实践中，可能会使用不那么严格的早产诊断标准，被诊断为早产的妇女最终在足月分娩的比例可能大大高于临床试验中的比例。了解这项研究中被诊断为早产的女性实际在足月分娩的比例肯定是很有趣的。不过，很明显，早产并不是早产的替代品。那么，为什么早产与早产的关联度如此之低呢？这个问题的答案与定义早产的方法有关。传统上，早产的诊断是基于观察到“频繁的”子宫收缩(使用了不同的定义)和宫颈“变化”(定义为宫颈扩张的变化和/或先前检查的消失)。Daan等人。在这项研究中，使用了类似的定义。不幸的是，测量和解释子宫收缩和宫颈扩张都充满了困难。评估子宫收缩有三个组成部分，即频率、持续时间和强度。在羊膜破裂开始之前，用体外测力仪测量宫缩，这是…
In this issue of the American Journal of Epidemiology, Dayan et al.(1) present their findings from a cohort study whose purpose was to assess the relation between maternal anxiety and depression and preterm labor. Unraveling the relation between maternal psychological events and pregnancy outcome is of great importance, especially as we begin to understand the biologic plausibility of such a link. In this regard, Dayan et al. have made a valuable contribution to the existing body of literature on this subject. Their study, however, raises an important question for both observational and interventional perinatal research: Should preterm labor itself be used as an endpoint? I believe there are several relevant issues to consider in answering this question. First, is preterm labor itself a clinically important health outcome? If not, is it a reasonable surrogate for another clinically important outcome? Dayan et al. make the argument that preterm labor is itself a clinically important outcome, based mainly on maternal considerations, such as the need for maternal hospitalization and the use of tocolytic medications. Certainly, there is a risk of thrombosis in pregnant women placed on bed rest, given the lower extremity venous stasis and hypercoagulability associated with pregnancy. In addition, the most commonly used tocolytic agents (beta agonists and magnesium sulfate) have both been associated with serious cardiopulmonary morbidity (2–4). Fortunately, the risk of thromboses and the risk of cardiopulmonary complications with tocolytic agents are exceedingly low. Given the low rate of maternal complications arising from preterm labor, the clinical importance of preterm labor is largely determined by its relation with preterm birth (infants born preterm are at substantially increased risk for many complications, including death, chronic lung disease, intraventricular hemorrhage, necrotizing enterocolitis, and others). If one agrees with this thinking, the next question becomes whether preterm labor is a reasonable surrogate for preterm birth. The answer to this is unquestionably “no.” As Dayan et al. point out, women diagnosed and treated for preterm labor often will deliver at term. This fact is borne out in the many clinical trials of tocolytic agents (where relatively strict diagnostic criteria were used), in which approximately 50 percent of women randomized to placebo delivered at or beyond 37 weeks (5). In clinical practice, where less stringent diagnostic criteria for preterm labor are likely used, the proportion of women “diagnosed” with preterm labor who ultimately deliver at term may be substantially higher than that seen in the clinical trials. It would certainly be interesting to know what proportion of women diagnosed with preterm labor in this study actually went on to deliver at term. Still, it is clear that preterm labor is not a surrogate for preterm birth.Why, then, is preterm labor so poorly associated with preterm birth? The answer to this has to do with the methods by which preterm labor is defined. Traditionally, the diagnosis of preterm labor has been based on the observation of “frequent” uterine contractions (various definitions have been used) and the presence of cervical “change”(defined as a change in cervical dilation and/or effacement from a prior examination). Dayan et al. in this study use a similar definition. Unfortunately, the measurement and interpretation of both uterine contractions and cervical dilation are fraught with difficulty. There are three components of the assessment of uterine contractions, namely, frequency, duration, and intensity. Prior to the onset of ruptured amniotic membranes, contractions are measured with an external tocodynamometer, which …