Editorial
The Rookie Goes to Bat: Is there a “July Effect” on Stroke Treatment at Teaching Hospitals?
The Neurohospitalist 2018, Vol. 8(1) 5-6 ª The Author(s) 2017 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1941874417740055 journals.sagepub.com/home/NHO
Tigran Kesayan, MD1, Juan Ramos-Canseco, MD1, and David Z. Rose, MD1
Historically, July 1st is the infamous date when, each year, newly graduated doctors (formerly medical students a few hours ago on June 30) start their internship and junior residents advance into senior roles. Along with the excitement and nervousness of students morphing into resident doctors emerges the “July Effect (JE),” a concept that patient outcomes are worse during this transition.1 The mantra of “Don’t get sick in July” emanates from this annual rite of passage—as seasoned, graduating residents are suddenly replaced by relatively inexperienced newbies. Inside most academic teaching hospitals and university medical centers, neurology residents are among the first-line responders to a “stroke alert” emergency. They must identify the precise time the patient was last known normal; interpret physical examination findings, such as limb weakness, numbness, aphasia, and the like; address abnormal vital signs, such as markedly high blood pressure; correct abnormal laboratory test results, such as hyperglycemia and coagulopathy; and recognize images on various brain scans, which can be a hyperdense clot in an artery or subtle blurring of the graywhite matter. In order to appropriately diagnose, time, and treat a stroke, it is critical to know how various stroke syndromes present and their usual etiologies, how to manage the hospital’s triage logistics and personnel, and how to order intravenous (IV) thrombolytic and intra-arterial (IA) clot extraction therapies. Predictably, brand new neurology residents may lack these skills, and therefore, it seems unlikely that patients with hyperacute stroke have optimal outcomes, door-to-needle times (DTNTs), and IV or IA treatment rates in July. Moreover, clinical presentation of stroke varies significantly from one patient to another and can flux from 1 time point to another in the same patient. Incoming residents have not had much (if any) exposure to dynamic, hyperacute stroke outside of textbooks. Residents’ decisions for diagnosis and treatment in this intense setting, where “time lost is brain lost,” is limited by a paucity of real-world experience. Hence, we read with interest the article “Door-to-needle time in acute stroke treatment and the ‘July Effect’” by
Dr Hawkes and colleagues from Argentina in this month’s Neurohospitalist.2 In a span of over 13 years, they retrospectively studied 101 acute ischemic stroke cases, divided into 3-month intervals, to assess differences between DTNT, thrombolytic-related intracranial hemorrhage, and 3-month outcomes using modified Rankin Scale and National Institute of Health Stroke Scale. They found no detrimental JE with their neurology residents; however, the sample size was small, as only 27 and 31 patients were studied in the June to August and March to May time frames, respectively, over 13 years. This equates to about 2 to 3 patients per year, per time block, and only about 1 thrombolysis per month. These sparse numbers, from only 1 center, lack generalizability and cannot confirm the presence or absence of a JE, especially vis-a`-vis DTNT. Moreover, any difference in thrombolytic rates is unclear between the June to August and the March to May groups, as only about 5% of those who presented with an ischemic stroke eventually received thrombolytic there. Possibly, in Argentina versus the United States, there are differences in thrombolytic inclusion/ exclusion criteria, which importantly, were updated around the time of this project. It seems likely the more complex revisions may impact JE also. To better evaluate JE, the methods section should describe, more specifically, the role of junior residents, senior residents, attendings, and radiologists. It is unclear what the junior residents do in this study—are they making decisions themselves or is there intensive oversight because of a self-fulfilling
1
Department of Neurology, Morsani College of Medicine, University of South Florida, Tampa, FL, USA Corresponding Author: David Z. Rose, Cerebrovascular Division, Department of Neurology, Morsani College of Medicine, University of South Florida, 2 Tampa General Circle, Suite 6087, Tampa, FL 33606, USA. Email:
[email protected]
6 prophecy of the JE itself? Is there a “July Hypervigilance” perhaps? At our institution, all thrombolytic decisions are made by attending vascular neurologists, based on information presented to them by the stroke fellow. A “green” trainee may take longer to gather, process, and present the clinical gestalt to their fellow and, by proxy, to the attending. This may lead to longer DTNT in July than in other months. Authors cite a 5% reduction in in-hospital mortality for every 15-minute reduction in DTNT2 but utilized a 1-hour cutoff; perhaps significance would have been achieved with ordinal minutes instead. Worse than DTNT delays, however, would be a missed stroke, untreated with thrombolytic that should have been (ie, due to misread neuroimaging, inaccurately timed onset, misunderstood symptoms, or alternate diagnosis, and so forth). Yet a third concern is overtreatment with thrombolytic; however, studies have shown stroke mimics given thrombolytic had no symptomatic intracranial hemorrhages.3 Another study of JE found that patients with ischemic stroke admitted in July were less likely to receive clotbusting drugs or be admitted to stroke units.4 Nevertheless, they concluded that new house staff did not influence mortality or functional outcomes of patients with stroke.4 Presumably, this was because stroke teams are typically interdisciplinary, involving more than just neurologists but also emergency medicine physicians, radiologists, cardiologists, neurosurgeons, physical/occupational and speech therapists, dietitians, and crews of nurses and technicians—so if one link in the chain is weak, others pick up the slack. However, in our opinion, it seems unlikely that there is no JE if less thrombolytic is used, and fewer strokes are admitted to the stroke-specific intensive care unit. Other studies have examined JE on heart surgery, orthopedic surgery, and in premature neonates—showing 4% to 12% higher mortality in July.4 Finally, it would have been interesting to look for differences in daytime versus nighttime DTNT, as the new neurology resident may have less nocturnal supervision or feel uncomfortable waking up superiors. When senior residents run stroke alerts in July (instead of new residents experiencing “trial by fire”), JE may be mitigated. Relatedly, punctilious stroke lectures in a “stroke boot camp” also helps.5 At our institution, the University of South Florida, we have constructed an all-day training program of virtual stroke patient simulation at the Center for Advanced Medical Learning and Simulation in downtown Tampa. Three acute stroke lectures by vascular neurology faculty are followed by 10 simulated
The Neurohospitalist 8(1) stroke alerts using paid actors and electronic robots that stroke and seize. Pre- and postcase debriefing sessions address participants’ knowledge gaps and boost confidence. Since program inception in 2012, we have found that mean scores improved for both declarative and procedural “cognitive” knowledge (P < .001) as well as with self-reported “confidence” (P < .001) scales.5 This course is mandatory each July, and feedback is uniformly positive. Despite the aforementioned limitations, our Argentine colleagues crafted a thought-provoking study that increases awareness of the important, underrepresented topic of JE in acute systems of care. Despite the low numbers, it is still entirely possible that they have jettisoned their JE with fastidious preparation, resident oversight, and superlative care. We agree with the authors that higher powered studies are needed to evaluate JE, DTNT targets that are continuously narrowing, and morbidity and mortality associated with new junior residents managing a stroke alert. Until then, however, don’t have a stroke in July. Authors’ Note Dr David Z. Rose is on the speaker’s bureau for BoehringerIngelheim and Boston Scientific. T. Kesayan, J. Ramos-Canseco, and D. Rose contributed to the drafting and revising the manuscript for content. This editorial is not industry sponsored. As this is an editorial, IRB approval or waiver of consent is not required.
References 1. Young JQ, Ranji SR, Wachter RM, Lee CM, Niehaus B, Auerbach AD. “July effect”: impact of the academic year-end changeover on patient outcomes. Ann Intern Med. 2011;155(5):309-315. 2. Hawkes M, Carpani F, Farez M, Ameriso S. Door-to-needle time in acute stroke treatment and the “July Effect” The Neurohospitalist. 2017. 3. Chernyshev OY, Martin-Schild S, et al. Safety of tPA in stroke mimics and neuroimaging-negative cerebral ischemia. Neurology. 2010;74(17):1340-1345. 4. Park TH, Redelmeier DA, Li S, Pongmoragot J, Saposnik G. Academic year-end changeover and stroke outcomes. J Stroke Cerebrovasc Dis. 2015;24(2):500-506. 5. Gangadhara S, Craig D, Burgin WS, et al. Virtual patient stroke simulation for residents and medical students improves knowledge, skill sets and confidence in acute stroke care. Poster presented at: International Stroke Conference (ISC); February 17-19, 2016; Los Angeles, California.
