Kappa – It’s Greek to Me

The Gist:  Many junior physicians use clinical decision instruments as an objective means of risk stratification or clinical decision making; however, these have subjective components.  Kappa, a measure of interrater agreement, is a commonly expressed statistic in medical literature, particularly in clinical decision aids.  Understanding the use, strengths, and weaknesses of kappa may help with application of decision aids and appraisal of literature.

The Case: A 13 year old boy presented to the Janus General ED after being struck in the head with a baseball bat.  He had a slight headache, no vomiting, normal mental status, and unremarkable physical exam except a hematoma over his left parietal region.
  • I presented the case as low-risk by PECARN with ~<0.05% chance of a clinically significant injury.  An attending inquired as to how I determined that the mechanism was "not severe."  Would my assessment change if Mark McGuire swung the bat that hit my patient?  Similarly, where was my threshold with the 18 month old that fell off a bed? Did the precise number of feet matter? The truth was, probably not, not because it wasn't listed in the objective criteria of the decision aid, but because after my assessment of the patient, I already estimated that the likelihood of a clinically significant injury was minimal. I wondered:  How did they come up with these variables (was there really a difference between falls from 3 ft and 4 ft)? How frequently would other people disagree with my seemingly "objective" determinations?
I found a paper by Nigrovic et al the next day that evaluated the agreement between nurses and physicians in the application of PECARN to mild blunt head injury pediatric patients.  This study demonstrates the differential level of agreement, or reliability, between elements of the PECARN predictors - with notable differences between subjective and objective components.*  For example, everyone agreed on vomiting, but anything containing the word "severe" was a little more nebulous.
  • History of vomiting - 97% agreement between nursing and physician assessment, with an outstanding kappa of 0.89 (95% CI 0.85-0.93). 
  • Severe injury mechanism - 76% agreed, kappa 0.24 (95% CI 0.13-0.35) in the age<2 cohort and kappa = 0.37 (95% CI 0.29-0.45) in the age 2-18 group.
Wait, what is this kappa (k) business?
  • It quantifies interrater reliability - a measure of the degree of agreement between observers that is greater than chance alone.
    • Sometimes, even in medicine, clinicians and trainees guess.  For example, when reading a radiograph and deciding on atelectasis versus infiltrate, a physician may hedge and choose one.  This may seem straightforward, but imagine a variable such as severity of headache.  Suppose one clinician has a terrific headache and rates headaches encountered that day as non- or less severe.  The cases when that clinician and another agree would therefore be based on chance.  
  • Calculation: (Observed Agreement - Agreement Expected by Chance)/(1-Agreement Expected by Chance) - Ok, so, the actual calculation is more complicated and is explained here.
  • Assesses precision/reliability
    • Using the aforementioned study, one can see that nurses and physicians reliably detected the presence of vomiting but less reliably agreed on the presence of a severe mechanism of injury or severe headache.
What does the value mean?
  • -1.0 = perfect disagreement, +1.0 = perfect agreement


What are the limitations of kappa?
  • The expected agreement is affected by abnormal prevalence.  In a skewed sample, the observed agreement may be markedly different than the relative agreement (1).  This is referred to as the kappa paradox, and there are various ways to compensate for this issue.
    • Rare findings - agreement between observers may not be as reliable and will be reflected by a lower kappa.  Looking at the Nigrovic et al paper, the kappa for palpable skull fracture is abysmal at 0.00, yet the proportion of physicians and nurses in agreement was 98%.  This exists as a product of the rarity of the finding, as 1/434 and 7/434 physician and nursing assessments were positive, respectively.  Similarly, signs of basilar skull fracture was fair at 0.37 with an enormous confidence interval (95% CI 0.07-0.67).  
  • Generalizability. Diversity of skill/experience may affect kappa.
    • Are the raters emergency physicians? medical students? specialized radiologists?
    • This was ostensibly what Nigrovic et al sought to determine - do clinicians at various levels of expertise agree?  The answer - it depends.  
What now? As a junior trainee, the ways I evaluate patients and objective data is different than that of a senior clinician.   Thus, I'm armed with this knowledge to acknowledge the limitations of the clinical decision instruments I use, understand why and how the variables are not hard and fast "rules," and use both to better patient care.
    *Note: The developers of PECARN (original study) only selected criteria with a minimum kappa of 0.5 (with a lower bound of the confidence interval of 0.40).

    References
    1.  de Vet HC, Mokkink LB, Terwee CB, Hoekstra OS, Knol DL.  Clinicians are right not to like Cohen’s κ 2013;346:f2125
    2.  Nigrovic LE, Schonfeld D, Dayan PS, Fitz BM, Mitchell SR, Kuppermann N. Nurse and Physician Agreement in the Assessment of Minor Blunt Head Trauma. Pediatrics. 2013. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23979081. Accessed August 29, 2013.

    A Disunion of the Literature – NSAIDs and Fracture

    The Gist:  Currently, practice and literature are divisive with regard to issues of non-steroidal anti-inflammatory medications (NSAIDS) and bone healing in fractures; however, analgesic dosed NSAIDs are probably ok in the acute fracture process in most patients.  Check out this open access article by Yates et al.

    The Case:  A 11 y/o male presented to Janus General with left forearm edema and deformity after falling during a basketball game.  Imaging demonstrated a non-displaced, transverse radius fracture with less than ten degrees of volar angulation. The patient received ibuprofen, was placed in a splint, and follow up with orthopedics was arranged.  Later, the orthopedist instructed me not to give NSAIDs to patients with fractures "because they delay healing."  I was marginally acquainted with the controversy from reading these bits of FOAM (Free Open Access Medical education) previously: The Trauma Professional and Jacobi EM's take, but I was startled enough by the physician's rigid statement of fact to review the literature myself.

    Why the worry?
    • Non-union is a feared, delayed complication of fracture and is estimated to occur in ~1% of long bone fractures although this rate is higher in other fractures and influenced by individual factors (1).
    • Studies of NSAIDs in fracture healing conducted in rats tend to show delayed fracture healing, delayed callus formation, and increased prevalence of non-union -see Table 1 in this article (2,3).  
    • The biologic plausibility - The initially inflammation generated by a fracture plays a role in creating the environment for bone healing (5).  NSAIDs inhibit cyclo-oxygenase (COX), decreasing the presence of some prostaglandins (2). Prostaglandins such as PGF2A and PGE2 aid in bone healing by stimulating osteoblasts.  These prostaglandins are also thought to play a role in angiogenesis necessary for bone repair/regeneration as well as bone metabolism (3).  Thus, COX inhibition and interference with these cytokines may interrupt the body's natural way of repairing itself.
    The Human Literature:  Oftentimes, animal studies and in vitro experiments don't translate into how things work in the human body and it appears this is the case with NSAIDs and fracture, as the data is a wash.  Why might this be?
    • No prospective RCTs with sufficient power have been conducted that reflect NSAIDs dosed for analgesia in patients with acute fracture.  
    • Clinical practice contains far more variables than a well designed laboratory study, thus confounding the data.  For example, mechanism of injury, smoking and glycemic control can affect fracture healing rates but these vary individually (1).  
    • Studies look at various kinds of fractures that have disparate expected rates of non-union (ex: post-op spinal surgery has 10-15% rates of non-union vs 1-5% of long bone fractures) (4).
    • Retrospective studies examining NSAIDs and delayed healing often involved routine, protracted dosing of NSAIDs or may have been more common in patients with more complicated fractures or signs of complications, as they consumed more NSAIDs.  
    This free, full text article by Wheeler et al summarizes the human studies with NSAIDs in fracture healing in Table 2 on the second page.  Below are some of the highlights from the literature: 

    Burd et al Retrospective, increased incidence of non-union in NSAID group (26% vs 7%, p = 0.004) and those with worse fractures.
    • Note: The patients in this study received indomethacin 25 mg TID x 6 weeks, which is not consistent with our acute pain dosing (they were looking at inhibiting heterotopic bone formation).
    Bhattacharyya et al  Retrospective database review, looking at patients who filled prescriptions for an NSAID in the 90 days after humeral shaft fracture.  The study only showed a significant difference in non-union in patients exposed to NSAIDS between 61-90 days (RR 3.9, 95%CI 2.0–6.2)
    • Limitation: NSAID use was neither quantified nor specified by type.  Of note, this cohort of 9995 patients predominantly included older adults (mean age = 77)
    Adolphson et al.  n=42, RCT with piroxicam in post-menopausal women with distal radius fractures with dorsal displacement (Colles), no delay in fracture healing.

    Dodwell et al - meta-analysis, inconclusive.  The pooled odds ratio for nonunion was significantly higher in patients exposed to NSAIDs when the heterogeneous studies of variable quality were analyzed together (OR 3.0, 95% CI 1.6-5.6). The higher quality studies alone did not demonstrate an increased odds of non-union (OR 2.2, 95% CI 0.8-6.3).
    • Limitations: Case control and cohort studies only, long bone and spine surgeries pooled together, variable NSAID and dosing.
    What I've taken away for my practice:
    • Ensure adequate immobilization of fractures to mitigate pain (5).
    • Provide analgesia to patients using my pain algorithm, which will continue include NSAIDs (such as ibuprofen 400 mg) or acetaminophen as needed for pain in patients with fracture. 
    • Consider the individual patient and factors that may put them at risk for impaired healing (high risk fracture, osteoporosis, smoking status, diabetes, etc).
    • Encourage patients to use the lowest effective dose of NSAIDs for the shortest period of time as the data suggests that higher doses of NSAIDs may be associated with delayed healing (4*,6).
    • Avoid aggravating colleagues and continuing dialogue on this evolving topic, when possible, by factoring in which orthopedist is on-call when prescribing NSAIDs. 
    References:
    2.  Pountos I, et al.  Do Nonsteroidal Anti-Inflammatory Drugs Affect Bone Healing? A Critical Analysis. ScientificWorldJournal. 2012; 2012: 606404
    3.  Kurmis et al. The effect of nonsteroidal anti-inflammatory drug administration on acute phase fracture-healing: a review. J Bone Joint Surg Am. 2012 May 2;94(9):815-23. doi: 10.2106/JBJS.J.01743. 
    4.  Reuben SS.  High dose nonsteroidal anti-inflammatory drugs compromise spinal fusion. Can J Anaesth. 2005 May;52(5):506-12.  *note: this study is authored by an individual who had 21 other publications retracted due to falsified studies, so findings should be taken with a grain of salt.
    5.  Menkes J.  Tintinalli's Emergency Medicine.  7th ed. New York: McGraw Hill Medical. 2011.  p.1789
    6. Dodwell ER, et al.  NSAID exposure and risk of nonunion: a meta-analysis of case-control and cohort studies.
    Calcif Tissue Int. 2010 Sep;87(3):193-202. 

    Tackling The Most Commonly Abnormal ED "Vital Sign"

    The Gist:  Pain is subjective and ubiquitous in the emergency department (ED).  Treatment of pain in the ED varies widely between practitioners and is often not formally taught as part of the curriculum, lending this aspect of EM to often be characterized as "art."  I came up with a rough pain control algorithm for acute pain or recurrent pain as I get started in Emergency Medicine, roughly based on an evidence based template from Lipp et al (note: I'm not an expert and this is far less useful for complex pain complaints such as migraine and does not address procedural sedation).
    Check out this article by Dr. Leon Gussow of The Poison Review on the darker side of the emphasis of pain as a Joint Commission mandated vital sign in the ED.  

    Regional Analgesia - It appears that, with the aid of ultrasound, nerve blocks are increasingly common in the ED. These are great for orthopedic and soft tissue injuries, and there seems to be a movement to increase utilization for hip fractures (Haslam et alHaines et al)
    Upsides:   Patient may not need sedation or sedating analgesics, thereby preserving mentation, respiratory drive, and hemodynamics.  Great, targeted analgesia 
    Downsides: Potentially time consuming, invasive, disrupts neuro exam, potential for systemic toxicity (use ultrasound!).  
    Pearls
      • Practice these on an anesthesia rotation. 
      • Use ultrasound:  Ultrasound Guided Nerve Block EducationPart IPart IIFemoral Nerve BlocksSonoguide.com.  Sonospot.com has many useful posts on publications and tipstricks for ultrasound guided nerve blocks (searchable).
      • Perform a solid neurovascular exam before performing the block.  Monitor patients afterward. 
      • Work with consultants (ex: ortho for blocks for femur and hip fractures).
    Non-opioid Analgesics - Acetaminophen (APAP) and Nonsteroidal anti-inflammatory drugs (NSAIDs) are effective for mild to moderate pain (2).
    Upsides:  NSAIDs are especially useful in prostaglandin-potentiated conditions, such as dysmenorrhea, gout, rheumatoid arthritis, and ureteral colic (1, 2).  APAP is safe in pregnancy (1,3).
    Downsides: 
      • Dose-limiting adverse effects on the gastrointestinal tract (NSAIDS). Must be cognisant of total APAP load across all medications so 4g/day is not exceeded  (1,2).
      • NSAIDS may have untoward effects in patients on other medications including: warfarin (bleeding), diuretics/lithium/ACE-I/methotrexate (renal insufficiency and subsequent therapeutic index issues) (1).
      • NSAIDS may induce or worsen renal insufficiency in elderly patients, those with pre-existing renal disease, or in the setting of dehydration (3).
    Pearls:
      • Ceiling dose for analgesia from ibuprofen lies in a 400mg dose (higher ceiling dose for anti-inflammatory effect), discussed in this blog post by Dr. Chris Bond. 
      • Parenteral keterolac offers no better analgesia compared with ibuprofen or naproxen, and costs much more (Dr. Bond's postArora et al, 2, 4).
      • IV APAP, used internationally with success, has made an appearance in the US over the past couple of years, but comes at a cost of more than $10 USD per 1g dose without drastically better analgesia.  The focus on IV APAP is the potential opioid sparing effect (Kwiatkowski et al)
    Opioid Analgesia - These get the most attention in the ED, partially due to the focus on abuse.  There are many drugs in this category but they tend to fall into groups:  
    • Opiates (derived from the opium plant) - morphine, codeine
    • Semi-synthetics - hydrocodone, oxycodone, hydromorphone, oxymorphone, buprenorphine, heroin 
    • Synthetics - fentanyl, alfentanil, sufentanil, remifentanil, methadone, tramadol, propoxyphene, meperidine
    Upsides
      • Good analgesia for visceral and somatic pain
      • Many opioids are titratable
    Downsides:  
      • Respiratory depression
      • Histamine release - pruritus, flushing, hives, and/or hypotension.
      • Released from mast cells, may be mistaken for allergic reaction (which has an immunologic component). 
        • Opiates tend to cause more histamine release than the semi-synthetics and synthetics (6).
      • Potential for abuse/dependence
    Pearls:
      • IV morphine starting dose for moderate-to-severe pain = 0.1 - 0.15 mg/kg. 
        • Patients in severe pain or opioid-tolerant, will not obtain adequate analgesia with a single 0.1-mg/kg dose and may require additional doses.
      • Oral opioids (hydrocodone) may take 30-60 min to achieve analgesia (3). 
        • This may be off-set by time to establish an IV, etc (Miner et al)
      • Oxycodone starting dose 0.125 mg/kg (Miner et al) in most patient populations.
      • IV hydromorphone starting dose 1mg, with an additional 1mg dose if analgesia not achieved (5).
        • less renal clearance so may be better in patients with renal insufficiency (2,4)
        • less histamine release (6).
      • When discharging a patient home with opioids, consider checking the state monitoring program online (ACEP clinical policy, Level C).  This can be time consuming and is generally limited to a single state.  Listen to Dr. Scott Weingart's Practical Evidence podcast on this.
      • When prescribing, prescribe a short course (ex: 3 days).
      • It's widely accepted that analgesia doesn't "mask" surgical pathology, which was a historic concern in patients with suspected intra-abdominal pathology (1-4).
    Poppies in Turkey
    References:
    1.  Miner J, Paris P, Yealy D.  "Pain Management."  Rosen's Emergency Medicine.  7th ed. p 2410-2428.  
    2.  Lipp C, Dhaliwal R, Lang E.  Analgesia in the emergency department: a GRADE-based evaluation of research evidence and recommendations for practice.  Critical Care 2013, 17:212 
    3.  Ducharme, J.  "Acute Pain Management in Adults."  Tintinalli's Emergency Medicine.  7th ed.  p 259-265.
    4.  Heins, Alan. Focus On: Effective Acute Pain Management.  October 2005.  ACEP. 

    #InternLife – Setting Up an RSS Feed Aggregator

    The Gist:  RSS feed aggregators act as a one-stop shop and bring information from blogs, websites, etc to you.  Setting one up is easy and most have free applications that allow seamless interfacing between web browsers, tablets, and smart phones...check out the video below for how I do it.

    • Yes, I realize the irony of posting this on a blog and is not useful for most folks, but I wanted a single link to refer people to when there's a paucity of time to help people set things up.

    Getting Started:

    1.  Pick an RSS Aggregator (Feedly, Flipboard, etc)
    2.  Add content.  The steps will be essentially the same as above.  Here are some of my favorite blogs to follow. You can also get articles from journals as they are published (or before, by following the articles in press feeds).  Here's how I approach that (note: uses GoogleReader but steps are the same in Feedly).
    3.  Read.

    Pearls:

    • Start slowly and steadily, as it's easy to get overwhelmed by information overload (see Dr. Chris Nickson's talk on FOAM overload).  For example, I often start people out with 3-4 blogs.
    • You can add journals (anything from Annals of EM to Journal of Medical Toxicology) and videos, too!  Examples include Amal Mattu's EKG videos, EM in 5 (core EM distilled into digestible 5 minute chunks), and HQMEDED (an assortment of excellent EM cases and talks, most of which are brief)

    LOC – How Much Does It Matter in Neuroimaging Decision-making?

    The Gist:  Loss of consciousness (LOC) in mild traumatic brain injury (mTBI) in adults isn't easily established as we think but often marks a decision point for imaging.  Use the patient's clinical status and risk factors to determine the need for neuroimaging and follow up in cases of mTBI.  Dr. Chad Kessler has a great free open access (FOAMtalk on the subject.

    The Cases:  Initially, as a medical student confronted with a patient with mild head injuries in the emergency department, I spent a good portion of the history attempting to establish the presence and exact duration of loss of consciousness and post-traumatic amnesia.  A few attendings harped on this point early on in my training, thus I felt it was crucial to establish with accuracy.  Take, for example, the following cases of GCS 15, neurologically intact patients with normal vital signs and otherwise asymptomatic. Which one needs a head CT?
    • 16 year old volleyball player fell at practice, certain she was out for "minutes."  Pain free, doing homework in exam room.
    • 20 year old restrained driver in an MVC with a tree, unable recall the accident or the few minutes following the accident. Abrasions on legs, no vomiting.
    • 68 year old restrained driver on ASA 81 mg daily with facial trauma requiring sutures on the forehead, denies LOC, recalls entire accident, conversant.
    • 3 year old fell from the sofa unwitnessed, hit her head with epistaxis that stopped spontaneously at home, and was "possibly" out for a moment before the baby sitter ran into the room
    Major decisions in these mild head injury patients:
    • Who needs a CT scan to identify clinically important injuries?
      • Or, in Europe or the future in the US, we may need to ask who needs an increasingly popular sensitive but poorly specific S100-B biomarker (Study: Zongo et al, Responses: Graham et al, Taira et al)
    • Who is at risk for deterioration? 
    • Who can be sent home after a normal CT scan?  Who needs intervention?  What follow up/precautions are advised?
    Why do we care about LOC/retrograde amnesia when we work up these patients?
    • It's scary to the patient and family/bystanders.  It can be important for building rapport and shared decision making to recognize that the experience of LOC/post-traumatic amnesia can be frightening.
    • Witnessed LOC or retrograde amnesia over 30 minutes before the trauma marks an entry point into clinical decision aids such as the Canadian CT Head Rule and New Orleans Criteria.  Here is my post on a comparison between the two decision aids.  
    • LOC also serves as a high risk criterion in PECARN (for pediatric patients).
    • Questionable indicator of severity of injury.  
      • Longer durations of LOC (> 5 minutes) associated with higher likelihood of significant injuries.  These are, perhaps, more likely to be witnessed or documented.
      • Some cite a study by Owings et al to demonstrate the preponderance of badness associated with patients with only a transient LOC; however, this study selected for really sick patient   (Inclusion criteria in one phase was admission to OR or ICU).
    How do we determine LOC after mTBI?
    • Self-reporting is unreliable.  
      • Mayou et al conducted a study in which patients with blunt head trauma in a motor vehicle accident were asked whether they definitely had LOC, definitely did not have LOC, or were uncertain.  
        • n=124 patients reported definite LOC.  15% of those definitely had LOC and 23% (n=29) probably had LOC.  
        • Essentially, 2/3 of patient's reporting LOC did not actually have LOC.  
      • Confounders:  alcohol, drugs, PTSD, pain, perception of time, pediatric patients.
        • Pediatric patients may not be able to communicate or identify LOC.
      • For neuroimaging purposes, the data suggest that we probably don't need to split hairs over whether someone had LOC for a few seconds or a minute.  Estimates of time are likely inaccurate - was it many minutes or just 1-2?
    • Talk to paramedics, bystanders, parents, etc to better establish whether LOC occurred. 
    How much does LOC matter for the things we care about?
    • 15% of mild TBI patients have intracranial pathology (Jagoda et al).
    • 0.5-1% have clinically significant injuries requiring neurosurgical intervention (Jagoda et al).
    • Do not have to have LOC to have mild TBI or significant cognitive dysfunction.
      • Normal neuroimaging may be falsely reassuring to a patient with mTBI.
      • Good follow up, precautions, and discharge instructions should still be given to these patients (CDC recommendations)
    • LOC alone is not predictive of need for neurosurgical intervention.  Smits et al noted that LOC/PTA plus other risk factors were associated with more significant injuries (note: this cohort was a little higher risk than some other mTBI studies, including patients with GCS 13-14 or high risk criteria + GCS 15).
    What do we do in practice?
    • Wide variation between emergency physicians with regard to ordering head CTs in the context of head trauma (Stiell et alPrevodello et al).  
      • The figure below demonstrates variations between providers in one academic ED in ordering a head CT in patients with head trauma (Limitation: may be confounded by variable injury severity level.
    Prevedello LM, Raja AS, Zane RD, et al.Variation in use of head computed tomography by emergency physicians.  Am J Med. 2012 Apr;125(4):356-64. 
      • Lowest ordering rates not necessarily associated with greater pathology or "misses" requiring further neurosurgical intervention (Stiell et al).
    • Probably don't use clinical decision aids like the CCHR properly.  Reductions in CT scanning in these populations have not panned out as expected, likely secondary to gaps in knowledge translation (Morton et al, Curran et al)
    What can we do?  
    • Educate ourselves with excellent free, open access resources (FOAM): Life in the Fast Lane has a great review of Mild TBI workup and disposition and EBMedicine - evidence based review.  
    • Know how to use decision aids properly and use one that suits one's community and risk tolerance as providers: PECARN, CCHR, NOC
    • Engage in shared decision making with patients.  In patients who don't need neuroimaging, discuss the limitations of scans with regard to concussions and emphasize measures they can take (follow up, precautions, rest, etc).
    • ACEP's clinical policy can also serve as a guide: Order a non-contrast head CT if the patient looks like they have badness or at high risk (following clinical features - Level A if they also have LOC, Level B without LOC):
      • Focal neurologic deficit
      • GCS <15
      • Coagulopathy/on anticoagulants
      • Vomiting
      • Headache (Severe Headache without LOC)
      • >60-65 years old 
    Or, if the patient has LOC/PTA AND: 
      • Drug or alcohol intoxication
      • Deficits in short-term memory
      • Physical evidence of trauma above the clavicle
      • Post-traumatic seizure
    Or, no LOC/PTA AND
      • Physical signs of a basilar skull fracture
      • Dangerous mechanism of injury: ejection from a motor vehicle, a pedestrian struck, or fall from a height of greater than 3 feet or five stairs
    Case Resolution:  The 68 year old and 20 year old received CT scans although family members in the other cases wanted scans.  The 20 year old didn't meet NOC or CCHR criteria and the scan was negative.  The 68 year old patient without LOC had a subdural hemorrhage on CT which was expansile and required neurosurgical drainage within a few hours.

    The FOAM Forecast – Better Than The Weather Guy/Gal

    The Gist:  Old news in the Free Open Access Medical Education (FOAM) world often predates alterations in guidelines and bedside practice.  Thus, while we may regurgitate things solely for test taking purposes or to sate the local practice patterns, it may behoove us to acquaint ourselves with changing schools of thoughts prior to these changes trickling down in standard ways.  FOAM can serve as a means of preconditioning one's way of thinking and actions to mitigate the knowledge translation window.

    The example: I recently completed my second iteration of ACLS testing with the 2010 guidelines and I encountered the following:
    LBBB is a STEMI equivalent?  I was told this by an attending as a third year medical student but, through Twitter, I recalled recently noting that the 2013 AHA Guidelines state that new or presumed new LBBB, in isolation, is not an indication for a patient to proceed directly to PCI or thrombolysis (I.e. no other concerning features on ECG or sick patient status).   As I marked up my algorithm, wandering whether this distinction had clinically relevance, I recalled the following case from nearly a year prior...

    The Case:  A 48 year old obese male presented to the Janus General ED with some new chest pain and mild dyspnea at 0300.  His vital signs were within normal limits and he appeared uncomfortable and anxious but in no distress.  A 12-lead ECG demonstrated a LBBB that failed to meet any of the Sgarbossa criteria.  My supervising physician instructed me to page the cardiology fellow to see the patient and push for PCI after my database search failed to produce an old ECG to look for a prior LBBB.  The cardiology fellow came in and, after seeing the ECG and the patient, stated he would take him to PCI later, perhaps the following day.  The patient had no bump in troponin and ultimately left the hospital without a diagnosis of an MI.
    • Interestingly just one week prior, I watched this brief tutorial from Dr. Amal Mattu on LBBB and MI, one of several times I'd heard this same trope on podcasts.  I followed the directives and waited until after shift to check into LBBB and MI a little further, knowing I shouldn't blindly accept something I watched on the internet.
    How FOAM forecasted the guidelines and provided updated education: 
    • 2011an EMCrit episode featured a discussion with Dr. Stephen Smith regarding reading STEMI on  ECGs with an LBBB, introducing the ST/S ratio 
      • A ratio of the amplitude of the ST segment to the S wave >-0.25 is suggestive of ischemia/STEMI and increases the sensitivity of the Sgarbossa criteria (full text article), awaiting further validation 
    • July 2012: Dr. Mattu foretold the changing guidelines in 20 minute video on LBBB and MI.
    • December 2012:  EMS 12-lead Blog educates EMS providers on the removal of LBBB as a STEMI equivalent in the 2013 AHA guidelines
    • June 2013: Dr. Mattu's ECG tutorial of the week re-emphasized the ability to read MI on an ECG with a LBBB.   
    These created a reading list on LBBB and MI - reinforcing life-long learning skills of reading:
    • Kontos et al
      • n=401 patients with LBBB undergoing AMI rule out
      • No difference in incidence of MIs between chronic and new LBBB
      • Concordant ST changes (n=14) were the most important predictor of AMI (OR 17, 95% CI 3.4-81, P < .001
      Chang et al
      • Observational cohort of ED patients with ACS symptoms
      • No difference in the rate of AMI between the 3 groups: 
        • new or presumed new LBBB - 7.3% (RR 1.1; 95% CI, 0.47-2.84)
        • old LBBB - 5.2% (RR, 0.84; 95% CI, 0.41-1.69)
        • no LBBB -6.1% 
      Jain et al
      • Retrospective single-center analysis. N =36 patients with new/presumed new LBBB and ACS symptoms (~1-2% of AMI population in this system), 12 diagnosed with AMI
      • 30/36 underwent emergent cath
      • Sgarbossa criteria performed poorly in terms of sensitivity (Score of at least 5, sensitivity ~14%)
      Neeland et al - Excellent review of LBBBs.
    How may this translate into clinical impact?  
    • Cath lab activation.  Garvey et al demonstrated that among 14 cath centers, 72% of cancellations were due to ECG reinterpretations.
      • Prehospital setting: As noted in the EMS 12-lead Blog, prehospital ECGs and providers can directly activate the cath lab.  Thus, the use of the modified Sgarbossa criteria and knowledge that a new LBBB alone does not necessarily qualify a patient for cath lab activation. 
      • ED: Improve the ways in which one can read an MI on a LBBB on an ECG (modified Sgarbossa with ST/S ratio)
    • Potentially spare patient unnecessary emergent revascularization. 
    • Ease communication with consultants.  Many cardiologists (such as the one in the above case) have stopped taking patients to PCI based solely on a new LBBB.  
    • So, it's not really about the LBBB, rather the diffusion of knowledge.  Training courses such as ACLS are updated every five years. Thus, if a provider takes a course towards the end of that five years, it may be a few years before they provider is updated.  It often takes even longer to unlearn practices, especially in non-teaching centers. Sites such as TheSGEM and RuralDoctors.net specifically target knowledge translation that may allow for more broad information sharing/collaboration and, perhaps, ultimately improved patient care.
    Limitations - There's a fine balance in being an early adopter, a cautious practitioner, a diplomat, a pot stirrer, and a skilled test taker.  Preparing for shifts in thought/practice, "dogmalysis" (word credit to Dr. Cliff Reid) through digestible bits of FOAM, may at least induce greater discussion and consideration in order to achieve this balance.