That's usually what happens… I think

It's September, and you are rounding on a pediatric inpatient unit with medical students and residents. Typical for this time of year, there are several patients admitted with an asthma exacerbation. All have received the usual asthma treatments, including systemic corticosteroids. As you move on to the next patient, the medical student asks: “This patient's white blood cell count is going up, is that normal? Could this be an infection?” You respond: “Well they received corticosteroids, so we expect a rise in the patient's white blood cell count.” Student: “Why? And how much of a rise do we expect and for how long?” You: “Well, that's what usually happens. And we expect to see something of a rise for a handful of days…” You then trail off and move onto the next patient… In medicine, certain concepts are passed on from one preceptor to another. Eventually, these concepts become dogma. Although this dogma may reflect clinical reality, we are not sure how to explain their finer details or consider how concepts— such as corticosteroids and leukocytosis—may have changed over time. As medicine changes, does our understanding of this dogma need to be re-evaluated? Sullivan et al. sought to quantify the rise in white blood cell count (WBC) in response to corticosteroids in noninfected, nonsurgical patients.1 This retrospective evaluation was large—a total of 28,425 patients were included—and 1608 of these patients received systemic steroids. Patients were stratified into low, medium, or high groups based on dosing and WBC followed for up to 4 days. The study demonstrated a dose-dependent WBC response. For patients who received steroids, WBC peaked on Day 2 across all groups, with the mean value remaining significantly elevated on Day 4 in the medium and high dose groups. On Day 2, the WBC was on average 4.84 × 109/L higher than baseline for the high dose group, 1.70 × 109 higher for the medium dose group, and 0.32 × 109/L higher for the low dose group. After Day 2, WBC declined and plateaued. This study demonstrates that in noninfected patients who receive steroids, clinicians can expect a rise in the WBC within the first 48 h before plateauing, and generally should not expect a new elevation after Day 3 or 4. Some clinicians may ask why should we research concepts that are considered “common sense.” When should we expend effort and resources re-evaluating dogma? Martin Westphal described in an article five common “established concepts” in critical care medicine that are in practice, but may not be rooted in evidence—or have evidence that actually demonstrates harm.2 Similarly, consider “common sense” decision to obtain blood cultures in children who are hospitalized with community-acquired pneumonia, despite a lack of supporting strong evidence,3 for a test with low positive yield rates and at risk of false positives due to contamination. Clearly, medical practice is rife with opportunities to re-evaluate dogma. How should we decide when such re-evaluation will be helpful? We believe that a researcher should demonstrate clinical equipoise in a concept (i.e., there is not a well-described answer) and that, if found, this answer will inform patient care and/or future research. In this study, despite this being “what usually happens,” the authors have demonstrated this concept in a very large sample size and provided values stratified by steroid dosing, as well as when we expect the WBC to plateau. They address a clinically important question for clinicians as they decide on whether to initiate a further infectious workup or not based on patient's elevated WBC. There is value in questioning the “well, that's usually what happens” in medicine so that we are not perpetuating common misconceptions as accepted dogma. It also provides an opportunity to conduct research that is highly accessible to hospitalists while also addressing clinically important questions. The authors have nothing to report. The authors declare no conflict of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.