The Skeptics Guide to Emergency Medicine

Dr. Ken Milne

Meet ’em, greet ’em, treat ’em and street ’em

Episodes

Mentioned books

Nov 22, 2025 • 1h 2min

SGEM#491: Prevalence of violence against health care workers among agitated patients in an urban emergency department

Date: October 30, 2025 Reference: Boes et al. Prevalence of violence against health care workers among agitated patients in an urban emergency department. October 2025 AEM Guest Skeptic: Dr. Suchismita Datta. She is an Assistant Professor and Director of Research in the Department of Emergency Medicine at the NYU Grossman Long Island Hospital Campus. Case: It's 7:34 pm. You just took a sip of your portable espresso machine coffee from your favourite Batman cup. It's been 34 minutes into your Tuesday overnight shift. Things are a little crazy, but you are proud of yourself for getting your day colleague out on time. Just then, your second-year resident walks briskly towards you and sits down next to you, an air of fluster about her. She is trying to keep her cool, but you can see her hands are trembling as she is putting in orders for the patient she just saw. You ask her what's up, and she responds, "The intoxicated patient in bed 12 just threatened to punch me when I told him I couldn't give him more pain medications. He said pretty awful things about me, called me names, and cursed at me... they called over security... but I still feel a little shook up about it... " She looks down and shakes her head, and then looks back at you and asks, "How often does this actually happen? Violence against us?" Background: Violence toward clinicians is not a rare event in the ED but rather a predictable occupational hazard. The Canadian Association of Emergency Physicians (CAEP) has called ED violence “unacceptable,” urging a system-wide, zero-tolerance culture and coordinated mitigation efforts across hospitals, EMS, and law enforcement. Their formal CAEP Position Statement on violence in the ED summarizes scope, risk factors, and policy recommendations for prevention and reporting [1]. In the US, the American College of Emergency Physicians (ACEP) maintains a consolidated resource hub on ED workplace violence, including policy statements and advocacy for federal legislation (OSHA standards and the SAVE Act) to mandate prevention programs and establish penalties for assaults on healthcare workers. ACEP’s 2022-member poll found 85% of respondents reported violence had increased in their ED over the prior five years, with 45% saying it had greatly increased. The 2024 follow-up highlighted that >90% feared threats or attacks in the prior year. These data align with the day-to-day experience of emergency physicians and underscore persistent underreporting and inadequate institutional responses [2]. Both CAEP and ACEP emphasize practical approaches such as environmental design, staffing and security policies, de-escalation training, standardized reporting, and partnerships with law enforcement. At the same time, they reject the idea that violence is “part of the job.” Clinical Question: Among ED patients with, what is the prevalence of violent events against health care workers, and how does that compare with events formally reported to the hospital? Reference: Boes et al. Prevalence of violence against health care workers among agitated patients in an urban emergency department. October 2025 AEM Population: ED patients from a locked observation unit at Hennepin County Medical Center (Minneapolis, MN). Exclusions: Patients known to be in custody at the time of the encounter were excluded from data collection. Exposure: Agitation, defined as an Altered Mental Status Score (AMSS) ≥ +1 (range from −4 to +4). Observers then recorded whether the encounter included verbal abuse, a threat of violence, or a violent act against a health care worker. Comparison: N/A Outcome: Primary Outcome: Assault against any health care worker, defined by Minnesota state statute as an act with intent to cause fear of immediate bodily harm or death, or intentional infliction/attempt to inflict bodily harm. Secondary Outcomes: Verbal abuse of health care workers by agitated patients (distinct from threats), defined as harsh/insulting/derogatory language or gestures intended to frighten, humiliate, or belittle. Type of Study: A secondary analysis of two prospective, observational studies conducted in the ED setting. This is an SGEMHOP, and we are pleased to have the lead author on the episode, Dr. Brian Driver. He is a faculty emergency physician and Director of Clinical Research in the Department of Emergency Medicine at Hennepin County Medical Center. Authors’ Conclusions: “Verbal abuse, threats of assault, and violent acts occurred frequently in ED patients with agitation and were underreported.” Quality Checklist for Observational Study: Did the study address a clearly focused issue? Yes Did the authors use an appropriate method to answer their question Yes Was the cohort recruited in an acceptable way? Yes Was the exposure accurately measured to minimize bias? Yes Was the outcome accurately measured to minimize bias? Yes Have the authors identified all-important confounding factors? Unsure Was the follow-up of subjects complete enough? Yes How precise are the results? Unsure Do you believe the results? Yes Can the results be applied to the local population? Unsure Do the results of this study fit with other available evidence? Yes Funding of the Study. None is stated in the manuscript, while the authors declare no conflicts of interest. Results: Across 17,873 screened encounters, 4,609 (26%) involved agitation (AMSS ≥ +1). Alcohol or drug intoxication was present in 4,108 (89%) encounters. Among agitated patients, the median age was ~36 to 39 years, about 50% were male, and the cohort included a substantial proportion of Black, non-Hispanic patients (40–46% across violence strata). The study took place in a locked ED observation unit within an urban safety‑net hospital with ~100,000 annual visits. Key Result: Among agitated ED encounters, assaults were common, formal reports of verbal abuse were rarely reported, while most violent acts were reported. Only 0.5% (9/1,786) of verbal-abuse events and 61.9% (224/362) of violent acts were formally reported to the hospital. Listen to the SGEM podcast to hear Brian answer our five nerdy questions. External Validity: This was a single-center analysis in a locked ED observation unit that preferentially rooms patients with suspected intoxication. While you suggest similar patients would otherwise be mixed in general ED beds, concentrating intoxicated, agitated patients may change the observed prevalence and risk environment. Measurement Bias and Lack of Inter-Rater Reliability. Trained observers used standardized definitions, but inter-rater reliability was not reported, and observers were embedded in the clinical setting (not blinded), which can introduce observation/expectation bias. This is particularly important for subjective constructs like “verbal abuse.” Comparator Misalignment & Underreporting Bias: The “formal report” comparator aggregated events from anywhere in the ED, not just the studied unit or only agitated patients. Using the same denominator (n=4,609) overestimates reporting percentages and biases toward the null in observed–reported differences. Institutional reporting is also known to be incomplete. Don't these differences limit the interpretation of the magnitude of underreporting? Ascertainment Constraints: The authors note constraints such as language limitations (inability to capture some non-English events), areas not observed (triage), and binary counting at the encounter level (not tallying multiple events per visit). Each can attenuate or distort true prevalence. Could this not systematically shift some of the point estimates? Descriptive Design: As a prevalence study, it does not adjust for potential drivers of violence (staffing, throughput, intoxication level, and de-escalation availability), limiting inference about determinants or modifiable factors. These confounders make it challenging to know how best to address the problem of violence based on this data set. Comment on the Authors’ Conclusion Compared to the SGEM Conclusion: We generally agree with the author that the rate of violence in agitated patients against health care workers is high and under-reported. SGEM Bottom Line: Violence against ED staff is common and substantially underreported, especially for verbal abuse. Dr. Suchi Datta Case Resolution: You tell your resident that, sadly, her experience is not singular and that violence against healthcare workers is a problem, especially amongst agitated patients. You encourage her to report the incident, as there needs to be more documented encounters to help facility advocacy on a systemic level. You also provide her with some resources to help process her trauma. She is very thankful and goes on to talk to you about how the encounter made her feel. You take a walk outside to the trauma bay and, after some breathing exercises, feel strong enough to come back and continue running the ED the rest of the night. Clinical Application: Be careful around agitated patients. Make sure you accurately report incidents of violence. We need to advocate for systemic changes to protect healthcare workers as per ACEP and CAEP. This includes more robust prevention, reporting, and accountability. Violence should not be part of our job. What Do I Tell My Patient? N/A Keener Kontest: Last week’s winner was Dr. Steven Stelts from New Zealand. He knew Magnesium got its name from Magnesia, a region in Thessaly, Greece. This is an area known in ancient times for its deposits of minerals and stones containing magnesium. Listen to the SGEM podcast for this week’s question. If you know, then send an email to thesgem@gmail.com with “keener” in the subject line. The first correct answer will receive a shoutout on the next episode. Now it is your turn, SGEMers.

Oct 25, 2025 • 23min

SGEM#490: These (Removable) Boots are Made for Walking

Reference: Boutin A, et al. Removable Boot vs Casting of Toddler’s Fractures: A Randomized Clinical Trial. JAMA Pediatr. Published April 2025. Date: July 23, 2025 Dr. Andrew Tagg Guest Skeptic: Dr. Andrew (Andy) Tagg is an Emergency Physician with a special interest in education and lifelong learning. He is the co-founder and website lead of Don’t Forget the Bubbles (DFTB). Case: A two-year-old boy presents to your pediatric emergency department (ED) with a limp and refusal to bear weight. His parents aren’t exactly sure what happened. They were at the park and suspected that he might have twisted his leg coming down a slide. On your exam, he does not have a fever. He does not have any swelling or deformity of his bilateral lower extremities. You order X-rays, which confirm a nondisplaced spiral fracture of the distal tibia, a classic toddler’s fracture. After you disclose his diagnosis, his worried parents ask you, “A fracture? Does that mean it’s broken? Will he need to get a cast? He’s such an active little guy, typically.” Background: Toddler’s fractures are subtle, nondisplaced spiral fractures of the tibia. They typically occur in children between the ages of 9 months and 4 years. They often present in children who are just beginning to walk, hence the term “toddler’s fracture.” These kids can come in with inability to bear weight, a limp, or nonspecific pain. Another challenge is that the history you get may or may not help guide you. These injuries usually result from low-energy trauma, such as a twisting injury during a fall. That can happen multiple times a day when you’re learning to walk! The mechanism of injury is so subtle sometimes that parents or caretakers may not recall any inciting event. The findings on X-ray can be quite subtle as well. AP and lateral views may reveal a very small hairline fracture that’s easily missed. In some situations, X-rays will be negative despite clinical exam findings suggestive of a toddler’s fracture. In those situations, repeated X-rays in a week or so can show some evidence of periosteal reaction. The traditional management has included immobilization with a long leg or short leg cast, based on the belief that toddlers are unlikely to limit activity independently and require rigid immobilization to promote healing and pain relief. However, growing concerns about the discomfort, skin breakdown, need for follow-up visits, and potential complications from casting have led to interest in less restrictive treatments. such as removable walking boots. Clinical Question: In children with radiographically confirmed toddler’s fractures, is treatment with a removable walking boot noninferior to a circumferential cast? Reference: Boutin A, et al. Removable Boot vs Casting of Toddler’s Fractures: A Randomized Clinical Trial. JAMA Pediatr. Published April 2025. Population: Children aged 9 months to 4 years with radiograph-visible tibial toddler’s fractures seen in 4 Canadian pediatric emergency departments. Exclusion: Presented more than 5 days after injury or had an increased risk for pathological fracture or delayed healing Intervention: Removable prefabricated walking boot. Parents were told to use the boot for one week, then use it as needed for symptoms for up to three weeks. This was based on caregiver discretion, with no scheduled follow-up. Comparison: Standard circumferential casting. This varied across sites: Two sites did a long-leg splint in the ED, followed by a fiberglass cast. Another site did a short or long leg splint/cast in the ED, followed by a long leg fiberglass cast. The final site placed a short-leg splint in the ED, followed by a short-leg fiberglass cast. The casts were applied within seven days of the ED visit. Two sites had casts that were peelable, meaning they could be removed by caregivers at home. The other two sites had patients return to the clinic for cast removal. Outcome: Primary: Evaluation Enfant Douleur (EVENDOL) pain score at 4 weeks Secondary: Return to activity, complications, caregiver satisfaction, care burden, healthcare utilization. Trial: Pragmatic, multicenter, assessor-masked, noninferiority randomized clinical trial Authors’ Conclusions : In this multicenter randomized clinical trial examining the management of children with TF, a removable boot without physician follow-up was noninferior to circumferential casting with respect to pain recovery. While there was a clinically relevant but not statistically significant trend toward more skin complications in the boot group, there was no difference in caregiver satisfaction, and the boot strategy demonstrated reduced childcare-related challenges Quality Checklist for Randomized Clinical Trials: The study population included or focused on those in the emergency department. Yes The patients were adequately randomized. Yes The randomization process was concealed. Yes The patients were analyzed in the groups to which they were randomized. Yes The study patients were recruited consecutively (i.e. no selection bias). Unsure. The patients in both groups were similar with respect to prognostic factors. Yes All participants (patients, clinicians, outcome assessors) were unaware of group allocation. No All groups were treated equally except for the intervention. Yes Follow-up was complete (i.e. at least 80% for both groups). Yes All patient-important outcomes were considered. Yes The treatment effect was large enough and precise enough to be clinically significant. Yes Financial conflicts of interest. Some authors reported grant funding from healthcare institutions or relevant foundations, but no direct industry sponsorship from Big Boot Results: They enrolled 129 children in the study, with 65 children randomized to the boot group and 64 children randomized to the cast group. The mean age was 2.2 years. Key Results: Removable boot was non-inferior compared to circumferential casting in the management of children with Toddler Fractures. Primary Outcome: This was pain at 4 Weeks on the EVENDOL scale. The non-inferiority margin was defined as 2 points on the EVENDOL scale. The boot group had a EVENDOL mean score of 1.2 (SD 1.5) while the cast group had a mean score of 1.8 (SD 2.1). We mentioned before that they did both an intention-to-treat and per-protocol analysis. The upper bound of the confidence interval was well below the threshold of 2, supporting non-inferiority. Secondary Outcomes: Children treated with a boot were more likely to return to their usual activities by four weeks, with over three-quarters back to normal compared to just 41% in the cast group. This was a difference of 36% (95% CI 9-63%). By 12 weeks, everyone had returned to normal weight bearing and baseline activities. Any other positives for the boot group? They also reported fewer day-to-day challenges. Bathing was easier. Only 41% in the boot group moderately/strongly agreed that bathing the child was a challenge compared to 72% of the cast group (Difference -32%, 95% CI -47 to -18%). There was less need to carry their child around (44% vs 68%; Difference -22%, 95% CI -27 to -15%). Skin complications like mild redness or pressure sores were slightly more common in the boot group, but these were not statistically significant (difference of 22%, 95% CI -6 to 50%). These were generally minor and picked up early. Overall satisfaction was high in both groups; 80% satisfied in the boot group compared to 70% in the cast group (difference 9%, 95% CI -20 to 39%). Though caregivers in the boot group were more likely to say they'd choose the same treatment again mainly because of the care burden and inconvenience of getting the cast placed and removed. There was no difference in unplanned doctor visits or repeat X-rays between the two groups. Radiograph-Visible Fractures Only This study only included radiograph-visible toddler fractures, the clear, obvious ones. But a large chunk of toddler’s fractures don’t show up on the initial X-ray. Those occult injuries may have different trajectories and pain profiles, and we often treat them similarly in practice. So, while the findings are robust for visible fractures, we can’t automatically apply them to every limping toddler with a presumed injury and a normal X-ray. The fracture morphologies included were also interesting. Toddlers' fractures are classically nondisplaced spiral fractures. This study included buckle fracture and transverse fractures too. They even included a few fractures with very minimal displacement, albeit those types of fractures only represented a smaller portion of the population. This feature may add a bit to the generalizability and makes us wonder…could we adopt the less is more approach for even more fracture types? Masking and Bias: The Pragmatic Trade-Off This was a pragmatic trial, which is great for real-world relevance, but only the outcome assessors were blinded. Parents and clinicians knew exactly which treatment was given, which might influence how they reported things like activity levels or skin problems. That’s a potential source of reporting bias, especially when dealing with subjective outcomes. Still, the fact that the EVENDOL scores, assessed blindly via video, aligned with parental reports adds some weight to the findings. Harms vs Benefits: What Matters Most? The boot came out ahead in terms of day-to-day practicality. It’s easier to bathe. There’s less carrying. The kids also returned to their activities quicker. That’s huge for families. Skin complications in the boot group were more common. Most of these (92%) were mild erythema but there were a few minor pressure sores. What’s more meaningful to parents? A slightly higher risk of a rash or being able to get through a week without plastic-wrapping their child’s leg for every bath?

Oct 18, 2025 • 46min

SGEM#489: Smooth Muscle Relaxator – But does Magnesium Work for Renal Colic?

Date: October 10, 2025 Guest Skeptic: Dr. Sergey Motov is an Emergency Physician in the Department of Emergency Medicine, Maimonides Medical Center in New York City. He is also one of the world’s leading researchers on pain management in the emergency department. Case: A 37-year-old man presents to the emergency department (ED) with severe right-sided flank pain. The pain started about eight hours ago as a vague discomfort in his right flank, but it has gotten progressively worse and now is radiating to his groin. Patient reports nausea, an increased urge to urinate and noticing blood in his urine on one occasion. The patient denies prior medical or surgical history. Upon ED arrival, his vital signs are normal. Physical examination revealed a stated age patient in distress due to severe right flank pain, prominent right-sided costovertebral angle tenderness, and absence of abdominal tenderness or guarding. While strongly considering renal colic in differential diagnosis and reaching for the bedside ultrasound, you are wondering if a single dose of a non-steroidal anti-inflammatory (NSAID) will be enough to relieve this patient’s pain, or should you add Magnesium or Lidocaine? Background: Renal colic is a common and extremely painful emergency department (ED) complaint encountered in the ED that frequently recurs. The nonsteroidal anti-inflammatory drugs (NSAIDs) given intravenously or intramuscularly (IM) are frequently used as first-line therapy. However, about 30% of ED patients receiving NSAIDS require rescue analgesia in the form of opioids. Opioid use, though effective, is limited at times due to the potentially dangerous adverse effects. Thus, there might be a role for other non-opioid classes of drugs to be co-administered with NSAIDs for relief of renal colic. Magnesium sulfate (MgSO₄) has been suggested as a possible treatment option. It may blunt ureteral smooth muscle spasm by antagonizing calcium influx in smooth muscle and by N‑methyl‑D‑aspartate (NMDA) receptor antagonism. These are mechanisms that can reduce visceral pain and augment other analgesics. Small ED trials and meta‑analyses suggest MgSO₄ can reduce pain scores and opioid use in renal colic, though the evidence base has been limited and heterogeneous [1]. Another suggested treatment modality for renal colic is intravenous lidocaine. We looked at this treatment on SGEM#202 and were unimpressed with the efficacy. Systemic lidocaine blocks voltage‑gated sodium channels and appears to modulate central sensitization and visceral pain pathways. In ED populations, systematic reviews indicate IV lidocaine offers variable analgesia with a mixed signal for benefit, and renal colic–specific RCTs suggest it may be inferior to ketorolac and best considered (if at all) as part of a multimodal strategy rather than as monotherapy [2]. Clinical Question: In adult ED patients with suspected renal colic receiving IM diclofenac, does adding IV magnesium sulfate or IV lidocaine increase the proportion achieving ≥50% reduction in pain at 30 minutes compared to a saline placebo? Reference: Toumia M, Sassi S, Dhaoui R, et al. Magnesium Sulfate Versus Lidocaine as an Adjunct for Renal Colic in the Emergency Department: A Randomized, Double-Blind Controlled Trial. Ann Emerg Med 2024 Population: The study enrolled adults aged 18 to 65 years with suspected acute renal colic and a pain score of 5 or more on a 10-point numerical rating scale (NRS). Exclusions: Pregnancy/breastfeeding; NSAID, MgSO₄, or lidocaine contraindication or allergy; renal/hepatic dysfunction; analgesic use in prior 6 h; bleeding diathesis or GI hemorrhage; significant CAD/arrhythmia; seizures; peritoneal signs; altered mental status; anticoagulation; hemodynamic instability; morphine allergy. Intervention: All patients received 75 mg IM diclofenac. The intervention groups then received either 1g IV MgSO₄ (10 mL) over 2-4 minutes or 1.5 mg/kg IV lidocaine (10 mL) over 2-4 minutes. Comparison: Placebo(10 mL normal saline) after 75 mg IM diclofenac. Outcome: Primary Outcome: The proportion of participants achieving at least a 50% reduction in the NRS score at 30 minutes after drug administration. Secondary Outcomes: Need for rescue analgesia, time required for 50% pain reduction, proportion of participants with persistent pain (NRS>2) at 90 minutes, frequency of adverse events, and frequency of return visits to the ED for renal colic recurrence. Type of Study: Prospective, multicenter, randomized, double‑blind, placebo‑controlled, 3-arm trial conducted from November 2022 to August 2023 in three academic hospital EDs and one regional hospital ED in Tunisia. Authors’ Conclusions: “Adding intravenous MgSO4, but not lidocaine, to IM diclofenac offered superior pain relief but at levels below accepted thresholds for clinical importance.” Quality Checklist for Randomized Clinical Trials: The study population included or focused on those in the emergency department. Yes The patients were adequately randomized. Yes The randomization process was concealed. Yes The patients were analyzed in the groups to which they were randomized. Yes The study patients were recruited consecutively (i.e. no selection bias). Unsure The patients in both groups were similar with respect to prognostic factors. Yes All participants (patients, clinicians, outcome assessors) were unaware of group allocation. Yes All groups were treated equally except for the intervention. Yes Follow-up was complete (i.e. at least 80% for both groups). Yes All patient-important outcomes were considered. Yes The treatment effect was large enough and precise enough to be clinically significant. No Financial conflicts of interest. This study was conducted without financial support. The authors have no conflict of interest relevant to this article to disclose. Results: They screened 1,321 patients and included 840 who were randomized (280 per arm). The mean age is in the mid 40s with a fairly even male/female split. The mean baseline NRS ~8.5–8.7. Ultrasound showed stones in ~20% and pyelocaliceal dilation in over one-third of patients. Key Result: Adding IV MgSO₄ to IM diclofenac increased the proportion of patients achieving ≥50% pain reduction at 30 min and reduced rescue analgesia use, but pain score differences never reached the 1.3‑point MCID, while IV lidocaine was not superior to diclofenac alone. Primary Outcome: ≥50% NRS reduction at 30 minutes MgSO₄ + diclofenac: 227/280 (81.7%) Lidocaine + diclofenac: 204/280 (72.9%) Placebo + diclofenac: 201/280 (71.8%) MgSO₄ vs lidocaine +8.8% (95% CI 1.89–15.7, p=0.013) MgSO₄ vs control +9.9% (95% CI 2.95–16.84, p=0.004). Secondary Outcomes: Rescue analgesia: Lower need in MgSO₄ group (17.1%) versus lidocaine (22.5%) and control (28.9%). Time to 50% pain reduction: No significant differences between groups (approximately 21 minutes for all groups). Persistent pain at 90 minutes: Similar across groups (20-22.1% with NRS>2). Return ED visits: No differences (22-26%) Adverse events: Significantly more in MgSO₄ group (57.1%) versus control (22.5%), primarily facial flushing (48% MgSO₄ vs 10% Lido vs 14% placebo) Statistical vs. Clinical Significance: This study exemplifies a critical issue in clinical research: achieving statistical significance without clinical importance. While the primary outcome showed statistical significance for MgSO₄ lidocaine and even more so for control, the actual NRS differences were all below 0.5 points at every time point. In addition, no between-group differences reached the generally accepted reduction of 1.3 for being clinically important [3,4]. There are concerns in the literature about what is considered a minimally important difference (MID) or minimally clinically important difference (MCID) [5, 6]. Choice of Primary Outcome Metric: The primary outcome was the proportion of patients achieving ≥50% pain reduction rather than the actual change in pain scores. This dichotomization of a continuous variable has several problems: Loss of Information: Converting continuous NRS data to a binary outcome discards valuable information about the magnitude of pain relief. For example, a patient going from NRS 10 to 5 (50% reduction) is counted the same as a patient going from 10 to 2 (80% reduction), yet these represent vastly different clinical scenarios. Arbitrary Threshold: The 50% reduction threshold, while commonly used, may not represent meaningful relief for all patients. For example, some patients with NRS 10→5 may still require rescue analgesia, while others with NRS 10→6 (40% reduction) might be satisfied. Diagnostic Verification of Kidney Stone: The diagnosis of renal colic was based on emergency physician clinical judgment or imaging with ultrasound showing direct or indirect signs of urinary tract stones in only 58% of participants when performed. This is consistent with current radiology, emergency medicine and urology guidelines not to get imaging on all patients [7]. We cover this on the SGEM Xtra: Come Together, Right Now...Over Renal Colic. While this approach may increase real-world applicability (generalizability), the “suspected renal colic” inclusion without universal confirmatory imaging may reduce internal validity and introduce misclassification bias. A confirmed stone sensitivity analysis would have helped strengthen the findings. Adjunct to IM Rather Than IV NSAID: While acknowledging and respecting the authors’ standard of care practice of using IM Diclofenac, there are several important implications: Delayed NSAID Absorption: IM administration results in slower and more variable absorption compared to IV administration.

Oct 11, 2025 • 0sec

SGEM#488: It’s Just a Minor Stroke – Should We Still Lyse?

Date: October 3, 2025 Reference: Doheim et al. Meta-Analysis of Randomized Controlled Trials on IV Thrombolysis in Patients With Minor Acute Ischemic Stroke. Neurology 2025 Guest Skeptic: Dr. Casey Parker is a Rural Generalist, Evidence-based medicine enthusiast and Ultrasound Nerd. This episode was recorded live, in beautiful Broome, Australia, at the Spring Seminar on Emergency Medicine (SSEM 2025). You can get copies of the slides used in the presentation at this LINK. You can also watch the episode on YouTube. Case: Dani is a recently retired emergency department (ED) doc who has spent the last year travelling the world, playing banjo & sharing time with family and friends. This morning, whilst eating a breakfast of eggs and ham, Dani had a sudden onset of right-hand weakness and difficulty speaking. Dani’s family called 000 (911 in North America), and she was taken to the ED within one hour. On arrival at your medium-sized rural ED, Dani is assessed by the “Stroke Team aka, you” as having mild motor weakness in the right hand and mild dysarthria. Dani is given an NIHSS score of 4. A rapid CT and CTA is quickly reported as “no acute large vessel occlusion” and “No intracranial bleed and no established cortical infarction”. You know that many centers in the city are offering intravenous tPA for patients with acute ischemic stroke. You wonder if Dani should get a dose? Background: Minor ischemic strokes (MIS), often defined by NIHSS ≤5, are very common, with roughly half of all ischemic strokes presenting with mild deficits. Despite the mild presentation, these strokes are not always benign. About 30% of patients with initially minor stroke symptoms end up significantly disabled (unable to walk independently) at 90 days [1]. In short, a small stroke can still have a big impact on a patient’s life if it isn’t effectively treated or if it progresses. Dr. Daniel Fatovich There have been gallons of ink spilled in the discussion of the stroke literature, with much debate on previous SGEM episodes about the relative risks and benefits of IV thrombolytic therapy for acute strokes. Drs. Ken Milne and Danny Fatovich have earned themselves the title of “non-expert EM contrarians” when discussing the literature around acute ischemic stroke management with Neurologists all over the world. IV thrombolysis (tissue plasminogen activator [tPA], or newer Tenecteplase [TNK]) is a well-established therapy for acute ischemic stroke based on some questionable evidence [2-6]. However, its role in mild strokes has been hotly debated. On one hand, treating early might prevent a minor stroke from evolving or causing hidden disability. On the other hand, tPA carries a risk of intracerebral hemorrhage, and many minor stroke patients recover well without aggressive intervention. Guidelines have wrestled with this nuance: current recommendations endorse tPA for mild strokes that have clearly disabling deficits, but advise against tPA for mild non-disabling strokes [7]. The core controversy is whether the potential functional benefit in MIS is worth the bleeding risk if the patient is already doing okay. Things changed 10 years ago after Mr. CLEAN was published. It showed that endovascular interventions (EVT) for acute large vessel occlusions (LVOs) could have impressive results (NNT of 7). However, the role of IV thrombolytics for minor stroke syndromes remains unclear and controversial. Legendary (now-retired) ED Dr. Joe Lex once stated, “If I can kick the syringe outta’ your hand – then don’t give me the tPA!” Was Joe right? Before 2019, practice varied widely. Some neurologists treated almost any stroke within the window, reasoning that “time is brain” even for mild deficits, while others were more conservative. Observational studies yielded mixed signals. Several studies suggested that thrombolysis in mild strokes improves the chance of an excellent outcome at discharge or 90 days, while others showed minimal benefit. The PRISMS trial (2018) was a key randomized study in this area. It compared alteplase vs. aspirin in patients with minor non-disabling strokes (NIHSS ≤5). PRISMS found no difference in 90-day functional outcomes (mRS score 0-1) between the tPA and aspirin groups, but did find an increase in symptomatic intracerebral hemorrhage with tPA [8]. However, that trial was stopped early after only ~1/3 of the planned enrolment (313/948). This was reported due to a lack of funding. There are issues with stopping trials early, which we have discussed on previous SGEM episodes. Stopping PRISMS early meant it lacked the power to definitively settle the question of lysis minor, non-disabling strokes. Consequently, equipoise remained, and actual practice often followed guideline nuance, treating “minor-but-disabling” strokes (for example, isolated aphasia or hemiparesis that significantly limits function) while generally avoiding tPA in trivial or rapidly improving strokes. Viele et al JAMA 2016, Guyatt et al BMJ 2012, Tyson et al Trials 2016 Clinical Question: In adults with minor acute ischemic stroke (generally NIHSS ≤5), does IV thrombolysis (IVT) improve functional outcomes compared with non‑thrombolytic standard care? Reference: Doheim et al. Meta-Analysis of Randomized Controlled Trials on IV Thrombolysis in Patients With Minor Acute Ischemic Stroke. Neurology 2025 Population: Adults (≥18 y) with minor ischemic stroke (NIHSS <6) eligible to receive IVT within 12 hours of onset from RCTs. Excluded: Nonrandomized studies or those without a control arm. Trials included patients with nondisabling and, in some RCTs, mildly disabling symptoms. Intervention: A variety of IV thrombolytic drugs (Alteplase, Tenecteplase, pro-urokinase) given within varying time windows, but most within 3 to 4.5 hours, followed by standard care. Comparison: Non-thrombolytic standard care (NT‑SC), which could include dual or single antiplatelet therapy, anticoagulants, statins, antihypertensives, glucose control, and other risk‑factor–directed treatments. Outcome: Primary Outcome: Excellent functional outcome at ~90 days, defined as mRS 0-1. (For IST‑3 subgroup data, OHS was converted to mRS; where only 6-month data existed, it was used.) Secondary Outcomes: Favourable outcome mRS 0-2, 90-day mortality, recurrent ischemic or hemorrhagic stroke and safety (symptomatic ICH [sICH]and any ICH). Type of Study: Systematic review and meta-analysis or RCTs Authors’ Conclusions: “IVT does not confer improved functional outcomes among patients with minor strokes and can be associated with higher odds of sICH and mortality.” Quality Checklist for Therapeutic Systematic Reviews: (yes/no/unsure) Was the clinical question sensible and answerable? Yes Was the search detailed and exhaustive? Yes Were primary studies of high methodological quality? Yes Were the assessments of studies reproducible? Yes Were the outcomes clinically relevant? Yes Was there low statistical heterogeneity for the primary outcomes? Unsure Was the treatment effect large and precise enough to be clinically significant? No Who funded the review? No targeted funding reported. Conflicts of interest declared? Authors report no relevant disclosures. (though not true for the primary trials - most had conflicts/funding by drug companies) Results: A total of 3,364 patients from four RCTs were included in the primary analysis. The age ranged from ~56 to 80 years across trials. Most had non-disabling deficits. Some RCTs included a minority with disabling symptoms. Trials varied in time windows (≤3 h, ≤4.5 h, ≤12 h) and imaging criteria (TEMPO‑2 required evidence of intracranial occlusion). The typical baseline NIHSS medians ~2 to 4 in the RCT. Key Results: Compared with non-thrombolytic standard care, IV thrombolysis did not improve excellent 90‑day functional outcome (mRS 0-1) and was associated with higher odds of symptomatic ICH and mortality in patients with minor stroke. Primary Outcome: mRS 0-1 at ~90 days: OR 85 (95% CI 0.70–1.03). No significant benefit of IVT vs NT‑SC. Similar null results across Alteplase, Tenecteplase, and prourokinase subgroups. Null also across disabling and nondisabling presentations. Secondary Outcomes: mRS 0–2 at ~90 days: OR 71 (95% CI 0.55–0.91). Lower odds of independence with IVT. It became non-significant when post hoc IST‑3 data were added (OR 0.85, 95% CI 0.58–1.24). Symptomatic ICH: OR 10 (95% CI 2.01–12.96). Increased with IVT. Any ICH: OR 21 (95% CI 1.63–3.01). Increased with IVT. 90‑day Mortality: OR 84 (95% CI 1.18–2.89). Increased with IVT. Recurrent Stroke: OR 01 (95% CI 0.79–1.29). No statistical difference. What Goes Into the Sausage Machine: This meta-analysis includes a range of trials with different inclusion, exclusion criteria, differing baselines, geographic/demographic and “standard care” comparators, so it is a bit of an evidentiary fruit salad. It can be hard to know how this data applies to the patient and the drugs/system of care that you are working with in your hospital. To make this decision, we may need to look back at individual cohorts and see if they represent our patients and system of care. Rural Stroke Application: In large tertiary hospitals where access to immediate imaging, stroke teams, and endovascular “clot retrieval” interventions has become the standard of care, all of the options are available. However, in rural or remote hospitals where we often do not have as much information or access to interventional neuro-radiology, we run the risk of delivering “second-class” care to our patients with acute ischemic strokes. This trial gives us some clarification around the role of tPA in our smaller community. Is it best to transfer patients with large, disabling strokes to a bigger centre,

Oct 4, 2025 • 36min

SGEM#487: Tell Me How I’m Supposed to Breathe with No Air? Nasal High Flow or Standard Care for Pediatric Intubation

Reference: George S, et al. Effectiveness of nasal high-flow oxygen during apnoea on hypoxaemia and intubation success in paediatric emergency and ICU settings: a randomised, controlled, open-label trial. Lancet Respir Med. March 2025 Date: July 10, 2025 Guest Skeptic: Dr. Spyridon Karageorgos is a Pediatric Chief Resident at Aghia Sophia Children’s Hospital, Athens, Greece and faculty of the Pediatric Emergency Medicine MSc at Queen Mary University in London. Case: A two-year-old boy presents in the emergency department (ED) with severe respiratory distress and hypoxemia. You attempt to use some non-invasive forms of respiratory support, but he continues to have significant work of breathing and retractions. His mental status begins to decline, and he appears much sleepier than before. The team makes the decision to intubate him. You follow the steps of the pre-intubation checklist and pre-oxygenate him with 100% FiO2. As the sedative and paralytic for intubation are given, the respiratory therapist asks, “Do you also want to use nasal high flow (NHF) for apneic oxygenation during intubation?” Background: Managing hypoxia in pediatric patients in EDs and intensive care units (ICUs) remains a challenge. Hypoxia can arise from various causes, including bronchiolitis, pneumonia, asthma, or undifferentiated respiratory failure. Ensuring timely and effective oxygenation is critical to stabilizing these patients and preventing progression to respiratory failure or cardiac arrest. NHF oxygenation has gained traction as a respiratory support modality in both ED and ICU settings. NHF delivers humidified and heated oxygen at high flow rates, which typically exceed the patient’s inspiratory flow through nasal cannula. This mechanism not only improves oxygenation but can also help reduce the work of breathing by flushing anatomical dead space and providing some degree of positive end-expiratory pressure (PEEP). The use of NHF is considered less invasive than continuous positive airway pressure (CPAP) or bilevel positive airway pressure (BiPAP) or mechanical ventilation. It is usually better tolerated, especially by children. The use of NHF has expanded into general pediatric practice, especially for treating conditions like bronchiolitis and other forms of acute respiratory distress. But, its comparative efficacy to standard oxygen therapy in various clinical settings and patient populations remains a subject of ongoing investigation. While we would always prefer a controlled intubation, sometimes the patients we see in the ED are unstable and need to be intubated emergently. We often try to pre-oxygenate prior to intubating to maximize oxygen reserves. Previous studies have suggested that the use of NHF may reduce the risk of hypoxemia and improve first-attempt intubation . However, randomized controlled trials evaluating this practice in the pediatric population are limited. Clinical Question: In children requiring emergency intubation, does the use of nasal high-flow oxygen for apneic oxygenation reduce hypoxemia and increase the rate of successful first-attempt intubation compared to standard care? Reference: George S, et al. Effectiveness of nasal high-flow oxygen during apnoea on hypoxaemia and intubation success in paediatric emergency and ICU settings: a randomised, controlled, open-label trial. Lancet Respir Med. March 2025 Population: Children aged 1 month to 15 years with acute hypoxic respiratory failure (SpO₂ <92% on room air) requiring emergency endotracheal intubation in EDs and pediatric and neonatal ICUs across Australia, New Zealand, and Switzerland. Exclusion: Primary nasal intubation, blocked nasal airways, elective endotracheal tube change, intubation required immediately for loss of cardiac output or respiratory arrest, location of intubation outside of ED or ICU, death Intervention: Nasal high-flow oxygen at 2L/kg/min during the apneic phase of intubation. Weight HFNC rate 0-12 kg 2L/kg/min (max 25 L/min) 13-15 kg 30L/min 15-30 kg 35L/min 30-50 kg 40L/min >50 kg 50L/min Comparison: Standard care Outcome: Primary Outcomes: Hypoxemia (SpO2 ≤90% or difference of ≥10% if they were unable to achieve a pre-intubation saturation of 100% or the patient had cyanotic congenital heart disease with a right to left shunt) and first-attempt intubation success without hypoxemia Secondary Outcomes: Total intubation attempts, re-oxygenation needs, duration of ventilation, lowest oxygen saturation throughout intubation period, length of stay, mortality, adverse events. Trial: Randomized controlled, open-label, pragmatic multicenter trial Dr. Shane George Guest Author: Dr. Shane George is a paediatric emergency and critical care physician at Gold Coast University Hospital, Australia. That’s right he’s trained in both emergency medicine and paediatric intensive care. He’s the lead for children’s critical care research for Gold Coast Health which is affiliated with the University of Queensland and Vice Chair of the Paediatric Research in Emergency Departments International Collaborative (PREDICT) Authors’ Conclusions: The use of NHF during emergency intubation in children did not result in a reduction in hypoxaemic events or an increase in the frequency of successful intubation on the first attempt. However, in per-protocol analysis, there were fewer hypoxaemic events but no difference in successful intubation without hypoxaemia on first attempt. Barriers to the application of NHF during emergency intubation and the reasons for abandoning intubation attempts before physiological compromise should be further investigated to inform future research and recommendations for intubation guidelines and clinical practice. Quality Checklist for Randomized Clinical Trials: The study population included or focused on those in the emergency department. No The patients were adequately randomized. Yes The randomization process was concealed. Yes The patients were analyzed in the groups to which they were randomized. Yes The study patients were recruited consecutively (i.e. no selection bias). Unsure The patients in both groups were similar with respect to prognostic factors. Yes All participants (patients, clinicians, outcome assessors) were unaware of group allocation. No All groups were treated equally except for the intervention. Unsure Follow-up was complete (i.e. at least 80% for both groups). Yes All patient-important outcomes were considered. Yes The treatment effect was large enough and precise enough to be clinically significant. Unsure Financial conflicts of interest. The funders (Fisher & Paykel Healthcare) provided equipment that was used during the trial but did not have a role in study design, data collection, data analysis, data interpretation, or writing of manuscript. Three of the authors reported consultancy fees, travel, and accommodations by the same company. Results: They included 969 children with 535 assigned to NHF and 534 assigned to the standard care group. Key Results: There was no statistical difference in hypoxemic events or first attempt intubation success among patients who received nasal high flow compared to standard care. Primary Outcome: Hypoxemia occurred in 61 (12.8%) of the NHF group compared to 77 (16.2%) of the standard care group (aOR 0.74; 97.5% CI 0.46-1.18, p=0.15). First attempt successful intubation without desaturation occurred in 300 (63%) of the NHF group compared to 280 (59.1%) of the standard care group (aOR 1.13; 97.5% CI 0.79-1.62, p=0.43). They also did a per protocol analysis where 45 of the intubations were removed. Hypoxemia occurred in 48 (10.8%) of the NHF group compared to 77 (16.7%) of the standard care group (aOR 0.59; 97.5% CI 0.36-0.97, p=0.017). First attempt successful intubation without desaturation occurred in 284 (64%) of the NHF group compared to 268 (58.1%) of the standard care group (aOR 1.22; 97.5% CI 0.87-1.71, p=0.19). Secondary Outcomes: There were no statistical differences in intubation attempts, need for re-oxygenation in between intubation attempts, duration of mechanical ventilation, length of ICU or hospital stay. The rates of minor and major adverse events were similar as well. Tune in to the podcast to hear Dr. George answer our questions. Biases In this study, approximately 14% of eligible patients were not approached for inclusion due to the treating clinician’s discretion. This is separate from those who were not approached for social or compassionate reasons. Another 14% were missed for unknown reasons. This could have introduced selection bias. Do you have any insights on why the treating clinicians chose not to approach certain patients? How do you think this may have impacted the study results? The open-label design (lack of masking) meant neither clinicians nor evaluators were blinded, introducing potential performance and detection bias. What impact to you think that may have had on the trial? Standard Care When we looked at the protocol deviations in the study. It looks like there was more non-compliance in the NHF group, with most of the reasons being that clinicians seemed to have forgotten to put on the NHF. Did you happen to collect data on what “standard care” meant to clinicians involved in the study? Modified Intention-to-Treat (mITT) vs Per-Protocol (PP) Analysis You performed two different analyses for the data: Modified Intention-to-Treat (mITT) and Per-Protocol (PP) analysis. Each type of analysis has some strengths and weaknesses. Why did you not perform a pure ITT? Your team did find a statistical difference in the rate of hypoxemia between the two groups based on the analysis performed with the per protocol analysis favoring the use of NHF.

Sep 27, 2025 • 42min

SGEM#486: Call Me, on the Line – Telemental Health for Suicide Prevention

Date: September 18, 2025 Guest Skeptic: Dr. Neil Dasgupta is an Emergency Medicine (EM) physician and emergency department (ED) intensivist from Long Island, NY. He is the Vice Chair of the ED and Program Director of the EM residency program at Nassau University Medical Center in East Meadow, NY. Reference: Doupnik et al. Impact of telemental health on suicide prevention care in U.S. emergency departments. AEM Sept 2025 Trigger Warning: The following case scenario discusses suicide and self-harm. If you or someone you know is at risk, seek immediate help (dial 911/999/112 as appropriate, or 988 in the US/Canada for suicide & crisis support). Resources: Substance Abuse and Mental Health Services Administration (SAMHSA) National Alliance on Mental Illness (NAMI) American Foundation for Suicide Prevention (AFSP) Case: It’s 23:15 on a Tuesday in a 10-bed rural emergency department (ED) that serves as the community’s sole hospital. A 37-year-old male arrives with a friend after texting that they “can’t do this anymore.” The friend is concerned because he has access to firearms. Triage vitals are stable. The nurse uses the ED’s standard suicide‑risk screen, which is positive. The nurse activates the service’s 24/7 telemental‑health workflow. A video cart is wheeled into the room, and a remote clinician joins the conversation. Background: Delivery of quality mental health care is one of the major difficulties affecting our EDs. Caring for these patients presents a particular kind of challenge, since establishing rapport with the patient, getting a detailed history, gathering collateral information from others, overcoming possible intoxications or toxidromes, requiring staff for continuous observation and treating physical injuries can require substantial levels of time and skills. In addition, suicide remains a leading cause of death, and EDs are a frequent point of contact for people in crisis. Many of these encounters involve complex psychosocial factors, limited outpatient capacity, and time-sensitive safety planning. The stakes are high, and what happens in the ED can shape risk in the hours to days after discharge. As emergency physicians, we balance therapeutic alliance, thorough risk assessment, and efficient disposition in an environment built primarily for acute medical care, not longitudinal mental health follow-up. Frustration often occurs due to limited resources, high volumes, inpatient boarding and overcrowding; it can seem impossible for an ED physician to provide compassionate, nuanced, complete psychiatric care. Patients utilize the ED for mental health care because they often do not have a choice. Sometimes that lack of choice stems from a report of suicidal thoughts, which in most communities represents a lack of capacity to make medical decisions, and EMS systems are required to transport these patients for emergent psychiatric care. In many communities, especially in areas that have less robust access to health care in general, there are profound administrative, financial and systemic barriers to creating or maintaining a functional level of mental health care infrastructure, profoundly overburdening the services that exist and pushing those needs onto the local emergency departments. Telemedicine (particularly telepsychiatry and broader telemental health) has become a pragmatic way to expand access to mental health expertise. This has accelerated with the pandemic-era virtual care. Programs vary widely, with some providing on-demand psychiatric prescribers, while others lean on social work, psychology, or case management. Integration with the ED team and the electronic health record (EHR) can be excellent in some settings and minimal in others. Despite legislative progress in the US, including the Affordable Care Act and the Addiction Equity Act, reimbursements remain poor for the care of such patients, severely limiting access to care. As with many other challenging issues, for lack of a better option, the ED becomes the entryway to any portion of the healthcare system. For rural and critical‑access hospitals, telemedicine can be the difference between no specialist input and round-the-clock access. But the question that matters to front-line EM clinicians is not just “Is telehealth available?” It’s “Does telehealth meaningfully improve the way we deliver suicide‑prevention care in the ED?”And, even if processes improve, do those changes translate into better patient-oriented outcomes (POO) such as reduced attempts, ED revisits, or suicide deaths? Clinical Question: Among US hospital-affiliated EDs, is having access to telemental health associated with greater routine use of recommended suicide‑prevention practices? Reference: Doupnik et al. Impact of telemental health on suicide prevention care in U.S. emergency departments. AEM Sept 2025 Population: 606 EDs in the US associated with a general medical hospital Excluded: EDs in government hospitals (VA/DoD), specialty hospitals (orthopedic), rehabilitation hospitals, and independent children’s hospitals (surveyed separately). Freestanding EDs not affiliated with a general hospital. Exposure: ED use of telemental health (telepsychiatry/telemental services available to the ED). Comparison: EDs without telemental health. Analyses stratified by critical‑access hospital status and adjusted for ED/hospital characteristics. Outcome: Primary Outcome: Routine use of six recommended suicide‑prevention practices Assessment of current suicidal intent/plans Past suicidal thoughts/behaviours Access to lethal means Standard approach to discharge planning Routinely scheduling follow-up Lethal‑means restriction counselling. Secondary Outcomes: Prevalence and characteristics of ED telemental health programs (staffing, hours, and EHR integration). Type of Study: A National cross-sectional survey with stratified probability sampling and nonresponse weighting. Dr. Stephanie Doupnik This is an SGEMHOP, and we are pleased to have the lead author on the episode, DrStephanie Doupnik is an Assistant Professor of Pediatrics and Health Policy and Director of the Division of Pediatric Hospital Medicine at Vanderbilt University Medical Center. Dr. Doupnik’s research has been funded by the National Institute of Health and focuses on the implementation of mental health services and suicide prevention care in EDs and hospitals, including the use of telehealth. Authors’ Conclusions: “Telemental health care is widely used across all types of EDs, and EDs with telemental health care are more likely to use suicide prevention practices. Critical-access hospitals rely on telemental health care to a great extent and need better access to telehealth psychiatry and EHR information sharing. Quality Checklist for Reporting of Survey Studies (Yes/No/Unsure)? Were hypotheses or aims explicitly stated? Yes Were operational definitions of the predictor (independent) and outcome (dependent) variables provided? Yes Were participant eligibility criteria (inclusion and exclusion) explicitly stated? Yes Were participants recruited using an acceptable recruitment strategy? Yes Were participants selected by a random/probability sampling strategy? Yes Was the sample size appropriate? Unsure Were participants randomly assigned into groups/ conditions? N/A Was the response/participation/recruitment rate provided? Yes Was the attrition rate acceptable? Unsure Was the attrition rate treated appropriately in data analyses? Yes Were the chosen statistical tests appropriate to address hypotheses or research questions? Yes Did the study include a formative research or pilot phase? Yes Were the measures provided in the report (or in a supplement) in full? Yes Were all measures of established validity, or was a validation procedure undertaken by the authors? Unsure Was the study sample described in terms of key demographic characteristics? Yes Was the data collection process described with sufficient detail for it to be replicated? Yes Were generalizations of findings restricted to the population from which the sample was drawn? Yes Was the study approved by a relevant institutional review board or research ethics committee? Yes Did participants provide informed consent (or assent, where relevant)? Yes Were funding sources or conflicts of interest disclosed? Yes Results: Of 977 eligible EDs, 606 responded (62%), weighted to 4,321 EDs nationally. Responders more often were rural, smaller, and critical‑access hospitals with nonresponse weights applied. Overall, 68% of responding EDs reported using telemental health. Lower ED volume, smaller bed size, and critical‑access status were associated with higher telehealth use. Key Result: Over two-thirds of US emergency departments use telemental health. Secondary Outcomes: Availability: More than 80% of telehealth EDs reported 24/7 coverage, with critical‑access EDs reporting they were more likely to have 24/7 access (81% vs 64%). Staffing: Telepsychiatrists/prescribers available in 68% of non-CAH programs vs 54% in CAHs. CAHs relied more on private contractors/other systems. EHR Integration: Telehealth clinicians could view the EHR in 64% of non-CAHs vs 28% of CAHs; documentation privileges 64% vs 35%. Post‑ED Follow‑Up: Only about one quarter of programs provided any follow-up after discharge. Listen to the SGEM podcast to hear Stephanie answer our five nerdy questions. Self‑Report & Social Desirability Bias: Outcomes were ED leaders’ reports of “routine” practice. Without chart audits or direct observation, overestimation is plausible, especially for practices perceived as best‑practice (lethal‑means counselling). How did you mitigate this issue?

Sep 23, 2025 • 38min

SGEM#485: I Want a New Drug – One Not Associated with Neurodevelopmental Disorders

Date: Sept 16, 2025 Reference: Prada et al. Evaluation of the evidence on acetaminophen use and neurodevelopmental disorders using the Navigation Guide methodology. Environ Health. August 2025 Guest Skeptic: Dr. Andrew Martin is an emergency physician practicing in Jacksonville, Florida. Case: A 27-year-old at 24 weeks’ gestation presents to the emergency department (ED) with fever (38.6 °C), myalgias, and sore throat. She took 650 mg of acetaminophen (Tylenol) six hours ago with partial relief. She hesitated to repeat the dose after reading online posts about “Tylenol and autism.” She has no abdominal pain, no vaginal bleeding, and normal fetal movement. Vitals otherwise stable; pharynx erythematous, no exudate. She asks, “Is it safe to take another dose, or could this hurt my baby’s brain later?” Background: Acetaminophen (paracetamol) is the most used analgesic–antipyretic in pregnancy. A recent prospective cohort study suggests ~40 to 65% of pregnant people report using it. They are typically using acetaminophen for headache, myalgias, or fever, with most use being short and intermittent. Alternatives, particularly non-steroidal anti-inflammatory drugs (NSAIDs), carry well-described fetal risks in late gestation. This is one of the reasons why acetaminophen remains the default first-line choice [1,2]. Biologically, acetaminophen crosses the placenta and achieves fetal levels like maternal levels, making the developing brain theoretically exposed during critical windows [3]. This has motivated a large observational literature examining whether prenatal exposure is linked to later neurodevelopmental outcomes such as ADHD and autism. Meta-analyses generally report small associations (summary effects around 1.2 to 1.3) and signal stronger effects with longer duration of use, though heterogeneity in exposure measurement and outcome ascertainment is substantial. Professional bodies, including the American College of Obstetricians and Gynecologists (ACOG) and the Society of Obstetricians and Gynecologists of Canada [SOGC], continue to recommend acetaminophen for appropriate indications at the lowest effective dose and shortest duration. At the same time, they do acknowledge ongoing research and the limitations of observational data (including confounding by indication). For emergency clinicians, the practical tension is familiar. The dilemma is that untreated maternal fever and significant pain can themselves harm pregnancy, yet patients are increasingly asking about possible long-term neurodevelopmental potential harms of using acetaminophen. Clinical Question: Is acetaminophen exposure during pregnancy associated with ADHD, ASD, or other neurodevelopmental disorders (NDDs) in children? Reference: Prada et al. Evaluation of the evidence on acetaminophen use and neurodevelopmental disorders using the Navigation Guide methodology. Environ Health. August 2025 Population: Observational studies assessing children of pregnant individuals for neurodevelopmental outcomes. Excluded: Postnatal exposures, non-human studies for the primary analysis, non-original publications, and duplicate reports from the same cohort. Exposure: Prenatal acetaminophen (maternal self-report, biomarkers such as meconium/cord blood, or medical records/prescription registries). Comparison: Children who were not exposed prenatally to acetaminophen, or those exposed to alternative analgesics. Outcome: Primary Outcome: NDDs (particularly ADHD and ASD) and related symptomatology measured by clinical diagnoses, medication use, or validated behavioural scales. Secondary Outcomes: Timing and dose–response patterns, broader cognitive/behavioural domains (language & executive function), and triangulation across design types. Type of Study: Systematic review using the Navigation Guide methodology with a qualitative synthesis (no meta-analysis) due to substantial heterogeneity. Authors’ Conclusions: “Our analyses using the Navigation Guide thus support evidence consistent with an association between acetaminophen exposure during pregnancy and increased incidence of NDDs. Appropriate and immediate steps should be taken to advise pregnant women to limit acetaminophen consumption to protect their offspring’s neurodevelopment.” Quality Checklist for Systematic Review: The main question being addressed should be clearly stated. Yes The search for studies was detailed and exhaustive. Unsure Were the criteria used to select articles for inclusion appropriate? Yes Were the included studies sufficiently valid for the type of question asked? Unsure Were the results similar from study to study? No Were there any financial conflicts of Interest? Yes Result: They searched PubMed (primary) through Feb 25, 2025, with confirmatory checks in Web of Science and Google Scholar. A total of 46 studies were included. The studies consisted of a mix of prospective cohorts, retrospective/historical cohorts, sibling-controlled cohorts, and two case-control studies. Sample sizes ranged from ~100 to a national registry scale (Sweden’s registry cohort included 2.5 million births). Key Result: The preponderance of studies reported positive associations between prenatal acetaminophen exposure and ADHD/ASD or related symptoms, with dose–response suggested in several biomarker and prospective cohorts. However, some sibling-controlled analyses attenuated associations toward null, and overall heterogeneity was high. Of 46 studies, 27 reported positive associations, 9 null, and 4 inverse (protective) associations. Several higher-quality or biomarker-based studies suggested dose-response relationships. No pooling into a meta-analysis due to heterogeneity 1) Exposure Misclassification: Most contributing cohorts ascertained prenatal acetaminophen exposure via maternal self-report during pregnancy or post‑partum recall, or from prescription/registry data that do not fully capture over-the-counter use. The authors’ own risk‑of‑bias summaries flag the exposure domain as a frequent concern, and they rate retrospective self-report, especially when collected after child diagnosis, as high risk for recall bias. In contrast, biomarker studies (meconium, cord blood, maternal plasma/urine) mitigate recall bias and, in several instances, suggest dose–response, but they remain snapshots that imperfectly reflect timing and cumulative dose. The review also highlights how low-sensitivity exposure measurement can deflate associations. An example is the large Swedish sibling‑analysis cohort that reported 7.5% use based on midwife interviews, despite contemporaneous sources indicating ~50 to 60% use. This implies substantial underascertainment (misclassification bias or ascertainment bias) that likely biases the estimate toward the null and compounds loss of power in within-family models. Together, these exposure‑measurement limitations should make us more cautious in both the direction and magnitude of observed associations in the qualitative synthesis. 2) Residual Confounding: Although many included studies adjusted for a broad set of factors (maternal age, socioeconomic status, smoking, alcohol, illness, fever, and infection) and some used negative‑control exposure periods or propensity approaches, the review acknowledges that unmeasured and residual confounding remain possible. Importantly, the authors did not implement a quantitative bias analysis (E‑values, probabilistic bias analysis) to bound the strength of confounding necessary to explain the findings. They explicitly list this as a limitation of the review. Given that indications for acetaminophen (fever, pain, intercurrent infection) may themselves relate to neurodevelopmental outcomes and can be difficult to measure with sufficient granularity, any qualitative conclusion about “persistence after adjustment” should be interpreted skeptically. 3) Outcome Heterogeneity & Variable Ascertainment: The review pools evidence across disparate outcome definitions: registry-based clinical diagnoses (ICD‑coded ADHD/ASD), ADHD medication use, and multiple validated behavioural scales (CBCL, SDQ), often at different child ages and with parent vs teacher report. This heterogeneity in outcome measurement (plus differences in timing of assessment) creates non-comparability that complicates causal interpretation and precludes valid pooling. Appropriately, the authors do not conduct a meta-analysis, citing the substantial heterogeneity in exposure assessment, outcome measures, and confounder adjustment. Risk‑of‑bias summaries also show concerns in the outcome domain for some outcome types. Mixing clinical diagnoses, proxy markers (medication), and symptom scales across varied ages increases measurement error and between-study variability, limiting the precision and generalizability of the qualitative synthesis. 4) Search Scope & Evidence‑Grading: The search strategy relied on PubMed as the primary database with confirmatory checks in Web of Science and Google Scholar. No additional eligible studies were identified beyond PubMed, and no discussion of searching the grey literature. Heavy reliance on a single primary database risks missing studies indexed elsewhere or gray literature, potentially introducing retrieval bias.[4] In addition, while the Navigation Guide offers a structured approach, the authors note that its numeric domain‑averaging can imply unwarranted precision and assign equal weight to domains even when certain biases (confounding, exposure error) likely dominate. They attempted sensitivity analyses (excluding lowest‑scoring papers and up-weighting confounding) but acknowledge the framework’s default “moderate” rating for observational evidence could skew certainty assessments. These methodological choices reasonably reflect transparency but also represent limitations of the review’s grading and synthesis.

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!

Get the app

The Skeptics Guide to Emergency Medicine

Episodes

Mentioned books

SGEM#493: You Can’t Hold Me Down with Spinal Motion Restrictions

SGEM#492: Give Blood – To Anemic Patients with Acute Brain Injuries

SGEM Xtra: Talkin’ Bout a Revolution…Training Health Communicators

SGEM#491: Prevalence of violence against health care workers among agitated patients in an urban emergency department

SGEM#490: These (Removable) Boots are Made for Walking

SGEM#489: Smooth Muscle Relaxator – But does Magnesium Work for Renal Colic?

SGEM#488: It’s Just a Minor Stroke – Should We Still Lyse?

SGEM#487: Tell Me How I’m Supposed to Breathe with No Air? Nasal High Flow or Standard Care for Pediatric Intubation

SGEM#486: Call Me, on the Line – Telemental Health for Suicide Prevention

SGEM#485: I Want a New Drug – One Not Associated with Neurodevelopmental Disorders

The AI-powered Podcast Player