Reply to Beck et al. and to Owora.
Reply to Beck et al. and to Owora.
复制标题
回复贝克等人。
DOI：
10.1164/rccm.202211-2130le
复制
发表时间：
2023
影响因子：
24.7
通讯作者：
Custovic A
中科院分区：
文献类型：
--
作者：
Custovic A
我们感谢Beck及其同事和Owora博士对我们关于湿疹、喘息和鼻炎演变的手稿的评论，我们在其中建议，现在可能是重新思考特应性进行曲框架的时候了（1）。这两封信都批评了我们对结果的定义，并建议需要医生的诊断才能进行更准确的估计。我们承认，我们的研究结果仅限于基于知识的定义（1）。然而，大多数最初描述特应性进行曲的基于人群的研究使用了类似的定义。此外，我们对特应性疾病流行病学的大部分了解，例如通过儿童哮喘和过敏国际研究（ISAAC）（2）和最近的全球哮喘网络（GAN）（3），使用了相同的广泛验证的工具。同样值得注意的是，我们之前已经证明，与父母通过有效问卷报告相比，使用医生诊断时对早期湿疹的错误分类更大（4）。在我们的分析中，我们关注的是喘息而不是哮喘。虽然不是所有的喘息都是哮喘，但喘息是其最重要的症状。术语哮喘的使用可能会在合并症时间模式的分析中引入偏倚。首先，学龄前喘息者（即使是那些症状严重的）很少被诊断为哮喘（5，6），造成了哮喘诊断跟随湿疹的印象，即使在那些早期喘息先于湿疹的人中，或者他们同时发展。其次，在喘息模式相同的儿童中，有湿疹病史的儿童可能比没有湿疹病史的儿童更容易诊断为哮喘（7）。Beck及其同事建议，食物过敏应包括在特应性进行曲的定义中，IgE致敏（伴或不伴屏障功能障碍）驱动特应性进行曲，而没有致敏则没有特应性或进行曲。我们最初提出的定义是特应性进行曲，指的是从婴儿期湿疹到儿童后期气道疾病症状的连续发展（8）。我们同意食物过敏是重要的，但作为特应性多发性硬化症的一个组成部分，而不是任何游行。很少有基于人群的出生队列有经口服食物激发证实的食物过敏数据。在我们的一个具有此类数据的队列中，我们观察到在具有多Morphine持续性的患者中，经口服食物挑战证实的花生过敏增加了五倍（1）。在最近的另一项分析中，我们发现花生过敏与机器学习衍生的多发病集群之间存在很强的相关性，但与单一疾病集群无关。然而，大多数患有多发性硬化症的人不是花生过敏，一些花生过敏的儿童没有多发性硬化症（1）。我们同意屏障功能障碍和致敏是重要的，但IgE致敏并非独立于屏障功能障碍，在某些情况下可能是其后果（10）。致敏不是特应性进行曲的原始定义的一部分，它的列入并没有加强的情况下，存在任何特定的序列之间的敏感的个人。虽然湿疹、喘鸣和鼻炎患者经常致敏，但许多致敏个体没有任何症状（11），并且在这些疾病的患者中，致敏是偶然发现的。例如，保守估计多发病率为8%，致敏率为40%，仅凭偶然，3.2%（0.0830。4= 0.032）两者都有。因此，致敏可能与症状无关（即良性），或可能导致不同的临床症状。
We are grateful for the comments by Beck and colleagues and Dr. Owora about our manuscript on the evolution of eczema, wheeze, and rhinitis, in which we suggested that the time may have come to rethink the framework of the atopic march (1). Both letters critique our definitions of outcomes and suggest that physician diagnoses are needed for more accurate estimates. We acknowledge that our findings are limited to questionnaire-based definitions (1). However, most population-based studies, which originally described atopic march, used similar definitions. Furthermore, most of our knowledge about epidemiology of atopic diseases, for example through the International Study of Asthma and Allergies in Childhood (ISAAC)(2) and more recently Global Asthma Network (GAN)(3), used the same extensively validated tools. Also of note, we have previously demonstrated greater misclassification of early-life eczema when using physician diagnosis compared with parental report by validated questionnaires (4). For our analyses, we focused on wheeze rather than asthma. Although not all wheeze is asthma, wheeze is its most important symptom. The use of the term asthma may introduce bias into the analyses of temporal patterns of comorbidities. First, preschool wheezers (even those with severe symptoms) are rarely diagnosed as having asthma (5, 6), contributing to the impression that asthma diagnosis follows eczema, even in those in whom early-life wheeze precedes eczema, or they develop contemporaneously. Second, among children with identical wheezing patterns, those with a history of eczema may be more likely to have asthma diagnosis than those without (7). Therefore, wheeze is better suited to capturing symptom progression.Beck and colleagues suggest that food allergy should be included in the definition of atopic march, that IgE-sensitization (with or without barrier dysfunction) drives atopic march, and that without sensitization there is no atopy or march. We defined atopic march as originally proposed, referring to the sequential development of symptoms from eczema in infancy to airway diseases in later childhood (8). We agree that food allergy is important, but as one component of atopic multimorbidity, rather than any march. Few population-based birth cohorts have data on food allergy confirmed by oral food challenge. In one of our cohorts with such data, we observed a fivefold increase in oral food challenge–confirmed peanut allergy in patients with multimorbidity persistence (1). In another recent analysis, we found a strong association between peanut allergy and machine learning–derived multimorbidity cluster, but not with single-disease clusters (9). However, most individuals with multimorbidity are not peanut allergic, and some peanut-allergic children do not have multimorbidity (1). We agree that barrier dysfunction and sensitization are important, but IgE sensitization is not independent of barrier dysfunction and in some cases may be a consequence thereof (10). Sensitization was not part of the original definition of atopic march, and its inclusion does not strengthen the case for the existence of any specific sequence among sensitized individuals. Although patients with eczema, wheeze, and rhinitis are often sensitized, many sensitized individuals do not have any symptoms (11), and in a proportion of patients with these diseases, sensitization is a chance finding. For example, using conservative estimates of multimorbidity prevalence of 8%, and sensitization of 40%, by chance alone, 3.2%(0.0830. 4= 0.032) will have both. Thus, sensitization may be irrelevant to symptoms (ie, benign) or may be contributing to different clinical …