|
Abstract (English)
:
| In the roundtable that follows, clinicians discuss a study published in this issue of the Journal in light of its methodology, relevance to practice, and implications for future research. Article discussed:
Hamilton EF, Smith S, Yang L, et al. Third- and fourth-degree perineal lacerations: defining high-risk clinical clusters. Am J Obstet Gynecol 2011;204:309.e1-6.
Back to Article Outline
Back to Article Outline
Third- and fourth-degree perineal lacerations are associated with short-term orbidity, such as blood loss and pain, and long-term morbidity; specifically, pelvic floor dysfunction. Thus, the ability to predict them would be important in obstetrics, as patients at high risk might be candidates for cesarean section. In a new study, researchers used a novel statistical method to identify women who were most susceptible to serious injury. First, they pinpointed independent risk factors. Then they determined which “toxic constellations” of risk factors most increased the likelihood of these lacerations. While many existing studies examine individual risk factors, few group them, making this work especially noteworthy.
George A. Macones, MD, MSCE,Associate Editor
Back to Article Outline
Macones: How would you classify this study design?
Cahill: I view this as a retrospective cohort study. The authors had access to a computerized database of over 25,000 deliveries in the Washington, DC-Baltimore area. They looked at a number of exposures—many of which have been described previously—and how they may be associated with perineal laceration.
Macones: What information do we have on the hospitals and the integrity of the database?
Cahill: The authors did a good job in describing this. One university hospital and 3 academic community teaching hospitals were included. Three of the 4 hospitals were regional referral centers. I believe the data were retrieved from the electronic medical record that is used in this health system. Generally, there is a difference between the quality of datasets used for clinical and research purposes, although in this case, the data seemed fairly robust. There were some missing data, but as expected, it is for parameters that aren’t of much clinical use—such as height. I think for the most part, the exposures and outcomes were clear and unambiguous.
Macones: How would you describe the analytic approach?
Stamilio: First, I think it is important to understand the basic premise of what the authors did. The introduction really explains this well. Essentially, in observational studies, we often generate models that look for possible associations between an exposure and an outcome. Adjusted odds ratios generally serve as the output for those models. While those are fine, they don’t tell us anything about whether a factor actually predicts what will happen to the patient. This paper really focused on the prediction of 3rd- and 4th-degree lacerations. In many ways, prediction studies like this have much more clinical utility than standard association studies.
The authors started with a standard multivariate association analysis. More interestingly, they next used classification and regression trees (CART), which have been in use for a number of years but are seldom used in obstetrical literature. The CART technique is also known as recursive partitioning. Essentially, CART takes individual risk factors, starting with the most important one, and splits the dataset on the presence or absence of that risk factor. There are no assumptions going in about cut-points for continuous variables; CART also finds the best cut-point and splits the variables accordingly. This process is repeated until no variables with discriminating power are left. It is very different from what we usually do in terms of modeling.
Back to Article Outline
Macones: What do the tables tell us?
Odibo: Table 1 is descriptive and tells us who the patients are in this study. But what I really like about this table is that the authors provide some detail about the degree of missing data. What is important here is that 3-4% of subjects had a 3rd- or 4th-degree laceration. Tables 2 and 3 present standard univariate (Table 2) and multivariate analyses. I think the analysis presented in Table 3 confirms much of what we know already: operative delivery, nulliparity, episiotomy, and birthweight are all associated with 3rd- and 4th-degree lacerations.
Macones: What does the figure tell us?
Odibo: This is really the heart of the CART analysis. You can see that the single most discriminating factor was episiotomy. On the left of that page, if you did not have an episiotomy, the only other factor that was discriminating was the length of the 2nd stage of labor. If you look at the right side of the page where the tree details findings on patients who had an episiotomy, you can see that many other factors were discriminating. These included, in order of importance, vacuum delivery, birthweight, maternal age, length of second stage, and body mass index. I know that we aren’t used to looking at these diagrams, so let’s take an example to the far right. If a patient has an episiotomy, a vacuum delivery, and a birthweight of >4312 g, then 100% will have a 3rd- or 4th-degree perineal laceration (7/7).
Back to Article Outline
Macones: I view this study as a bit more of a methods paper rather than a source of information that can be applied in practice tomorrow. What do you think?
Cahill: I agree, but it certainly opens up some good areas for research. One small issue is related to the cut-points that were selected. Some seem difficult to use in a practical way.
Macones: I totally agree. For example, one variable is birthweight. However, before delivery we have an uncertain estimate of birthweight at best. Yet, this study certainly makes us rethink association and prediction.
Macones: Do other methods exist for predicting outcomes?
Stamilio: There certainly are, and I wrote a review paper with Bill Grobman a number of years back on this exact topic.1 In addition to CART, there are methods such as neural networks, predictive nomograms, and multivariable techniques. These all have pros and cons, and I would refer readers to that article if they want to get into the gory details of these various methods.
Macones: Well, thanks for coming to this session of the Journal Club. Hopefully, this will spur some additional research on how best to predict patient outcomes in obstetrics.
Back to Article Outline
|
![[Valid RSS]](images/icons/valid-rss-rogers.png "Validate my RSS feed"){kind=icon}
Difícil 
