emDocs a group of seven emergency physicians who want to provide a truly current resource for the rapidly developing field of emergency medicine. Their goal is to inform the global EM community with timely and high yield content about what providers like you are seeing and doing everyday in your local ED.
Positive Blood Culture (ignored), ESR 62 (misinterpreted) Diagnosis of SEA missed despite 10 MD encounters in 10 weeks; patient suffers stroke from septic embolus
An adult diabetic male with a bovine artificial aortic valve and prior CABG presents to the ED with gradually increasing generalized pleuritic chest pain, cough, and fever. ROS is negative for N/V, diaphoresis, or SOB. His exam, CXR, EKG, and chest CT are unremarkable, D-dimer is slightly elevated at 0.7, and troponin is normal. Blood cultures are ordered, but no differential supporting the test is recorded. The patient is treated empirically with levofloxacin for the fever and discharged to follow up with his primary care provider (PCP) in 3 days. A copy of the ED record is sent to the PCP.
The blood cultures grow viridans (gordonii). The lab informs the on-duty ED physician, but no action is taken. The lab also faxes the report to the patient’s PCP. There it is simply date-stamped and filed in the chart. The patient does not see his PCP as instructed.
Three weeks later he again presents to the ED with back pain. He is seen by the same ED physician who received the earlier phone report of the positive BC, but no mention is made of that in the record. Possibly suspecting a spinal epidural abscess, an ESR is ordered. Despite the ESR being 62, the patient is treated with prednisone and discharged.
Over the ensuing week he sees his PCP twice, but the ESR and positive blood cultures are not mentioned. He is treated with more prednisone and Vicodin.
Five weeks after onset, he presents to the ED (twice actually) with weakness and back pain. This time the elevated ESR is noted, but the positive blood culture is not. He is diagnosed with polymyalgia rheumatica (PMR).
He sees his PCP 4 more times for his persistent back pain. On his 6th visit, the PCP repeats the ESR. It is now 80. The possibility of endocarditis is raised, and blood cultures are again ordered. The PCP only discovers the prior report when the new results also grow Strep viridans.
The patient is finally admitted, and an MRI shows osteomyelitis. He is finally given antibiotics.
Ten weeks after the first visit to the ED, he suffers a stroke from a presumptive septic embolus from subacute bacterial endocarditis (SBE) of his bovine aortic valve and is left with a left hemiparesis.
A lawsuit is filed against the ED physicians, hospital, and PCP.
Plaintiff: I had a positive blood culture in February. Over the next 2 months I saw my PCP 9 times and was in the ED 3 times, and you paid no attention to this. You even did a sed rate – twice – and didn’t connect the dots. You also treated me with prednisone, a risk factor for infection. I didn’t have PMR but osteomyelitis and SEA. You failed to consider either diagnosis when you had 12 chances to do so. If you’re going to do blood cultures, do something when they come back positive. Because you didn’t, I am now paralyzed from my stroke – and I need a new heart valve.
Defense: No defense arguments are available for this case.
Result: Settlement for undisclosed amount against the ED physicians and the PCP. The hospital paid nothing.
Form a differential diagnosis, seek clues to rule out the worst-case scenario (this appears to have been started but never finished), and document accordingly.
Why order tests if you pay no attention to the results? When you are given a critical result, assure that you have a system in place to acknowledge that and act on the information.
Document the communication from lab to doctor and from doctor to patient.
Don’t anchor to a diagnosis, especially when a patient is a bounceback seen on multiple visits.
Author: James Dazhe Cao, MD (@JamesCaoMD, Assistant Professor of EM, Medical Toxicology Fellowship Director, UTSW / Parkland Memorial Hospital)// Edited by: Cynthia Santos, MD (Assistant Professor, Emergency Medicine, Medical Toxicology, Rutgers NJMS); Alex Koyfman, MD (@EMHighAK, EM Attending Physician, UTSW / Parkland Memorial Hospital); and Brit Long, MD (@long_brit, EM Attending Physician, San Antonio, TX)
As a medical toxicologist, I have a different take on the world of overdoses. Paracelsus has said “All things are poisons, for there is nothing without poisonous qualities. It is only the dose which makes a thing poison.” The deadly dihydrogen monoxide has its merits – https://www.dhmo.org/facts.html. Paracelsus’ adage is something that all medical toxicologist live by; however, we understand there is much more depth to the art of toxicology than the dose makes the poison.
For emergency medicine physicians, overdose patients present regularly, especially in the current opioid epidemic. In 2017, U.S. poison centers recorded 656,235 human exposures managed at a healthcare facility – a bulk of which initially presented to the emergency department.1 Other than calling the poison center for recommendations and advice (which I encourage everyone to do – 1-800-222-1222), a framework for the initial assessment and management of overdose patients is critical.
In this writeup, I will describe my approach to assessing and managing overdose patients in the emergency department through my lens as a medical toxicologist. I will not discuss the foundational knowledge of toxicology such as toxidromes and will not discuss envenomations. Many other finer details on the initial vital signs and considerations for treating overdose patients can be found in the Goldfrank’s Toxicologic Emergencies 11thedition chapters 3 and 4.2 The textbook is also a valuable resource to read on some of the toxins mentioned in this post.
A 13-year-old girl presents to the children’s emergency department after an ingestion. She is tearful and withdrawn. She states that she has been feeling depressed and, in an attempt to harm herself, has taken 2 bottles worth of her grandmother’s medications. She has normal vital signs and appears stable.
As an emergency physician, the tenant of sick versus not sick is vitally important. This patient appears stable, but depending on the drug, this may be transient. Dr. Sahaphume Srisuma and I had published a study on potentially preventable deaths reported to the National Poison Data System (NPDS) from US poison centers. We showed a commonality between these toxins leading to preventable deaths – minimal early symptoms followed by precipitous deterioration3:
Fluoride containing products
By no means is this a comprehensive list, but remember these drugs to keep that BAFERD Spidey sense sharp. I guarantee you that these drugs haunt my dreams and keep me up at night.
For those patients who present acutely ill, remember the “ABCDE” of medical toxicology:
Dextrose and Decontamination (see below)
Exposure – remove potentially transdermal absorbed substance (i.e. fentanyl patches).
Enhanced elimination (see below)
Extracorporeal membrane oxygenation (ECMO) – becoming more common in emergency departments. Despite its complexity and risks, overdose patients may be perfect candidates for ECMO given the transient effects of most overdoses, so think about initiating it if you have ECMO available.
Manage each step as necessary as one would resuscitate any acutely ill patient. Remember that the basic ACLS management is often more important than the targeted management for a given toxin. Yes, antidotes may help, but without oxygen and blood circulation, antidotes will not work.
As you may have already noticed in my discussion thus far, I did not reveal what the patient has actually taken. I have my reasons for this. Keep your differential open when approaching toxicology cases. Even for my toxicology fellows, it is very easy to anchor and close your differential by “knowing” what the reported ingestion is. Many of our consults start with “the patient took drug X.” However, the patient may have been exposed to something other than what was reported because:
the patient is not truthful about the actual ingestion,
the patient falsely remembered the ingestion or remembered only part of the ingestion, or
the patient thought they took drug X but the contents were actually drug Y.
I encourage my fellows to discuss this point when leaving a recommendation – does the reported ingestion makes sense in the context of the patient’s history, exam, and laboratory analysis?
In my opinion as with most emergency department patients, history is one of the most important aspects of managing an overdose patient. For toxicology patients, I strongly encourage obtaining collateral from multiple sources to collaborate the patient story:
Assisted living staff
Law enforcement personnel
Available pill bottles
Collateral can help confirm or deny the patient’s history. Alternatively, other sources can help piece together the clinical picture that the patient either does not remember or will not admit to. For example, time of ingestion is one of the most common questions that we ask. Bracketing an unknown time is a useful technique especially for something that is time critical such as determination of starting N-acetylcysteine (NAC). Ask the collateral source when they last saw the patient normal, when they received a text message from the patient, or when the patient’s social media post was timestamped.
EMS and law enforcement who were on the scene are important allies in obtaining an accurate history. Most EMS and law enforcement providers know to bring pill bottles and available paraphernalia. If they are available at the hospital, ask them what they saw and what was the environment around the patient when they retrieved him/her.
Pharmacy history is typically an adjunct that is done by the toxicology consult service, as it is time consuming. Knowing what pills are available at home may be important to the clinical picture. Sometimes if the patient is out of your electronic medical record system, calling the patient’s pharmacy to obtain a medication list can be useful.
Pill bottles hold important clues. Remember that the label does not necessarily indicate what is inside the bottle. Patients commonly reuse old prescription bottles and combine multiple pills into a single bottle. Many bottle labels describe the intended pills that are in there. Match the description on the label with the pills inside. Counting pills is a staple of a good toxicologic history. Note when the bottle was filled and how many pills are left in the bottle. As with bracketing a time, this may give you a sense of how many tablets a patient may have taken. The maximum number of pills is if one assumes the patient never took his/her pills on a daily basis and consumed all missing pills in this one ingestion. The minimum number of pills is reflected by daily use of the pills since the refill day and estimating the number of missing pills above that. For unknown pills, many online sources such as Epocrates can help identify pills. The poison center also has resources to help you identify unknown pill.
Airway: Knowing when to protect airway is challenging even for a toxicologist. Have a low threshold to do so, especially if you anticipate clinical deterioration.
Be aware that end-tidal CO2 may be a useful adjunct but understand its limitations and low sensitivity.4
Remember to reserve naloxone, flumazenil, and endotracheal intubation for respiratory depression and not isolated central nervous system (CNS) depression.
Toxidromes: Know them and look for them. Remember that toxidromes typically only apply if a patient ingested substances from a single class of agents. In the 2017 NPDS report, 12.14% or 256,783 human exposure cases involved more than one substance.1
Elements of the physical exam to focus on:
Bowel sounds – helps suspect anticholinergic toxidrome and can direct decontamination measures (see below). So yes, you do have to use your stethoscope below the diaphragm!
Neuromuscular exam – needed to identify potential serotonin syndrome and neuroleptic malignant syndrome.
Toxicologist’s handshake – feel under the axilla for sweat – will be dry in anticholinergic toxidromes or in a really dehydrated patient. Yes, it is gross. Yes, I use gloves to do this.
Know about it, but largely this practice is going away. Decontamination as a percentage of total human exposures reported to U.S. poison centers from 1985-20171:
Do not make administering decontamination modalities a knee-jerk response to an overdose patient. Think long and hard about the toxicity of the drug, the benefit of decontamination, and the risk of the procedure.
However, for large ingestions of a toxic drug, decontamination may be one of the only modalities to potentially save a patient.
Start with the basics as needed including but not limited to comprehensive metabolic panel, acetaminophen concentration, and an EKG. Acetaminophen ingestion may be asymptomatic, and a screening concentration may change management. Order concentrations for other drugs ingested and that your laboratory can turn around in a couple of hours, such as salicylates. No need to order send out testing in the emergency department, as this will not help initial management.
Please have your nurses send some extra tubes of blood and urine down to lab for future testing, as the clinical course may change.
Determine the most likely drug or drugs in the ingestion and the lethality of potential toxic exposure. Consider three key aspects of the case:
Drug: How dangerous is the drug? If the patient is exposed to ibuprofen, the risk of toxicity is relatively minimal with almost no risk of mortality. However, if the patient ingested just 0.5 mg/kg of colchicine, then chances of death are high with little directed therapy.
Dose: Different drugs have different toxic ranges. Colchicine as mentioned becomes dangerous above 0.5 mg/kg, whereas we tolerate acute acetaminophen ingestions of 200 mg/kg. I do not try to remember all of the dose ranges of toxicity. Use your pharmacology references. I use Micromedex to get me in the ballpark: https://www.micromedexsolutions.com/home/dispatch(will need institution access if you have it).
Patient: Consider the patient’s co-morbidities – less physiologic reserve, the more dire the potential outcome.
Excellent supportive care! While antidotal therapy can have dramatic affects, recommendations for antidotes are relatively rare. Including the use of NAC in 2017, only 59,619 instances of antidotes were given to the 2.1 million human exposures calls reported to U.S. poison centers. Even if only one antidote was administered to one exposure call, that represents only 2.8% of calls to the poison center.1
In 2019, the ExTrip workgroup will provide evidence-based recommendations on: Baclofen, Ethylene glycol, Methotrexate, Isoniazid, Calcium Channel Blockers, Beta-Blockers, Dabigatran, Gabapentin/Pregabalin, Quinine/Chloroquine, Amatoxins.
A 49-year-old patient with no known past medical history presents with encephalopathy. The patient is obtunded, receives naloxone without improvement, and is intubated for airway protection. Initial laboratory evaluation and CT of the brain are unremarkable. The patient’s family state that he has been gradually worsening over the past two weeks.
This case represents a class of consults that I endearingly term “could this be tox?” These are cases of diagnostic uncertainty with some element of history reported to the primary provider suggesting a possible toxicologic etiology. We as toxicologists will further investigate. I do not expect emergency providers to have the time to dive into these histories and investigative work, but I wanted to share a glimpse of what we do.
History of Present Illness:
As with the case above, the history is paramount. In this case, collateral is important since without it, you are searching for a needle in a haystack. The world of toxins is vast!
If the patient is awake, start from the beginning of symptoms and explore circumstances leading to symptoms and surrounding potential exposures.
Investigate all medications the person may have been exposed to – patient meds, family member medications. Ask about herbals and supplements, especially if they are from a foreign country! Even prescription medications from other countries can lead to toxicity as the medication may have been withdrawn from the U.S. markets by the FDA.
Anyone else sick in the home or at work? Targeting environmental exposures.
Are any pets sick at home? Infectious disease can affect multiple humans, but environmental exposure effect multiple species. This is figuratively the canary in the coal mine.
Any travel history or camping with consumption of wild flora?
What is your occupation? Occupational exposures may have chronic health effects.
What is the home environment like? For example, if the house was built before 1978, lead paint may have been used. If one is concerned about methemoglobinemia, ask about well water consumption.
Same as above except here I am looking for more subtle toxicologic findings (i.e. Mees’ lines for heavy metal poisoning) to provide clues.
Here we may be sending out more specific toxicologic testing. At our children’s hospital, we have access to a comprehensive urine drug screen using advanced methods like liquid chromatography–mass spectrometry (LC-MS). LC-MS can be helpful to confirm or exclude differential diagnoses, but because of the delay in results, testing is not useful in the initial management.
I frequently get asked about management of unexplained metabolic acidosis. I use the mnemonic KULTO:
K = Ketones (diabetic ketoacidosis, alcoholic ketoacidosis, and starvation ketoacidosis)
T = Toxins (salicylates, toluene, and toxic alcohols)
O = Organic acids (inborn error of metabolism and 5-oxoprolinuria from chronic acetaminophen)
Differential Generation: Use the available history, physical, and laboratory findings to narrow the differential as much as possible. This process tends to be organ based depending on the presentation. Goldfrank’s textbook chapters 15-29 are organized by organ system and has tables listing toxins that cause those organ system dysfunctions.2
Also includes LactMed – drugs and lactation database
Developmental Toxicology Literature – reproductive toxicology database
Summary – Key Take Home Points:
Avoid premature closure by verifying through history and exam that the patient’s presentation is consistent with what the reported ingestion was.
A thorough toxicology history with relevant social history elements take time, but learn to start incorporating some of these questions if diagnostic uncertainty is present – i.e. ask if pets are also sick at home.
Manage and stabilize patients as you would any critically ill patient with attention to ABCs and supportive care. Add in considerations of antidote, decontamination, and enhanced elimination based on suspected exposure.
Know your toxicology resources listed above and call your local poison center or medical toxicologist for clinical assistance.
Gummin DD, Mowry JB, Spyker DA, Brooks DE, Osterthaler KM, Banner W. 2017 Annual Report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 35th Annual Report.Clinical toxicology. 2018;56:1213-1415.
Nelson L. Goldfrank’s toxicologic emergencies. Eleventh edition. ed. New York: McGraw-Hill Education; 2019.
Srisuma S, Cao D, Kleinschmidt K, Heffner AC, Lavonas EJ. Missed opportunities?: an evaluation of potentially preventable poisoning deaths. Clinical toxicology. 2016;54:441-446.
Viglino D, Bourez D, Collomb-Muret R, et al. Noninvasive End Tidal CO2 Is Unhelpful in the Prediction of Complications in Deliberate Drug Poisoning. Annals of emergency medicine. 2016;68:62-70 e61.
Author: Shelly D. Saha, MD (EM Resident, UTSW / Parkland Memorial Hospital, Dallas, TX) // Edited by: Alex Koyfman, MD (@EMHighAK) and Brit Long, MD (@long_brit)
Welcome to EM@3AM, an emDOCs series designed to foster your working knowledge by providing an expedited review of clinical basics. We’ll keep it short, while you keep that EM brain sharp.
A 37-year-old G4P3003 female at 27 weeks GA presents via EMS for vaginal bleeding. She began having brisk vaginal bleeding after coitus about 30 minutes ago and has saturated 1 pad since onset. She denies associated abdominal pain, cramping, or prior episodes of vaginal bleeding during this pregnancy. She has not had regular prenatal care; however, she did have a transabdominal ultrasound in her first trimester showing an IUP. She has a history of 3 C-sections and denies complications during her other pregnancies.
On exam, her vitals are BP 102/65, HR 112, RR 20, SpO2 100% on RA, T 36.8°C. Her abdomen is gravid, soft, and non-tender. On external pelvic exam, she has no visible lesions, and the maternal pad she is sitting on is about 50% saturated with blood. You defer a bimanual or speculum exam at this time.
What’s the diagnosis? What’s the next step in your evaluation and treatment?
Answer: Placenta previa
Placenta overlying the os to any degree is classified as placenta previa.
Placental edge within 2 cm of but not overlying os is classified as low-lying placenta.
Previous terminology of marginal, partial, complete is no longer used with TVUS providing more precise measurements.1
Most placenta previas are diagnosed on routine prenatal transabdominal ultrasonography.
However, early transabdominal ultrasound can be normal, so consider previa for all 2nd/3rd trimester vaginal bleeding.
Incidence: 1 in 200 pregnancies (increasing due to more C-sections being performed).1
Many previas diagnosed in mid-pregnancy will resolve by time of birth due to placental “migration” away from cervix as pregnancy progresses.
Poor maternal outcomes: increased need for blood transfusions, hysterectomy, ICU admission, sepsis, DIC, postpartum hemorrhage, mortality.5
Poor fetal-neonatal outcomes: increased risk of prematurity, lower birth weight, lower Apgar score, mortality (increased 3- to 4-fold).1
Placenta accreta (placental attachment to myometrium) occurs in approximately 10% of patients with placenta previa.6
Vasa previa is more common in patients with placenta previa or low-lying placenta.
Vasa previa – unprotected fetal vessels run over or near cervical os.
High risk of fetal-neonatal hemorrhage with fetal vessel damage during labor.
Stable placenta previa without bleeding or other complications is optimally delivered at 36 – 37 6/7 weeks of gestation.7
ED Presentation & Evaluation
Painless vaginal bleeding after 18 weeks gestation is placenta previa until proven otherwise.
Classic presentation is painless 2nd or 3rd trimester vaginal bleeding.
However, uterine contractions occur in 15-20% of first bleeding episodes, so previa should always be on the differential for late term bleeding, even when abdominal pain/contractions are present.8 Must also consider placental abruption with pain and bleeding.
Can have sentinel bleed which is rarely life-threatening; subsequent bleeding episodes tend to be more severe and occur more frequently.
Ask about prior placenta previa, uterine surgery, social risk factors, prenatal ultrasound abnormalities.
Quantify amount of bleeding (number of pads saturated, passage of clots) and symptoms associated with bleeding (chest pain, shortness of breath, etc.).
Identify vital signs concerning for hemodynamic instability.
Maternal tachypnea, tachycardia, hypotension, hypoxia are signs of hypovolemia.
Use visual aids to monitor blood loss (regularly check maternity pad saturation).
Never perform digital exam for late term bleeding until placental location is known (high chance of disrupting a clot if there is a previa).
Obstetricians debate over whether a speculum exam is safe – can be performed to exclude other causes of vaginal bleeding (laceration, infection, etc), but it is important to avoid direct cervical manipulation.
Diagnosed by transvaginal ultrasound (95% sensitivity, 93% specificity).9
Transabdominal ultrasound is less sensitive (86% sensitivity, 93% specificity).9
Early involvement of obstetrics team – these patients will range from needing admission for observation to emergency C-section based on severity of bleed.
Continuous maternal and fetal monitoring.
Administer RhoGAM to Rh negative women.
Resuscitate with fluids and blood transfusion to achieve and maintain maternal hemodynamic stability. In patients with massive bleed or actively bleeding placenta previa, prioritize giving blood products.
Some sources argue for a transfusion goal of hemoglobin > 10 g/dL in anticipation of a rapid Hgb drop in the actively bleeding placenta.10
Continue blood transfusion until patient is hemodynamically stable and bleeding has slowed.
Most women presenting with actively bleeding placenta previa do not require emergency C-section and stabilize with the above measures.11
Indications for C-section:
Nonreassuring fetal heart rate tracing or maternal hemodynamic instability unresponsive to resuscitation.
Severe bleeding after 34 weeks – maternal risk of hemorrhage > fetal risk of preterm birth after 34 weeks.
>24 weeks: admission to L&D for observation vs emergency C-section if unstable after resuscitation.
<24 weeks: can potentially be discharged home if small amount of bleeding which has resolved and patient has reliable follow-up.
Many women have recurrent bleeding episodes requiring multiple hospitalizations.
A 23-year-old G1P0 woman at 30 weeks gestation presents to the emergency department for vaginal bleeding. She has had no prior prenatal care. She reports that she is soaking through one pad every eight hours. She denies abdominal cramping or a gush of water and reports good fetal movement. Her vital signs are T 37.2°C, HR 100, BP 105/70 mmHg, and RR 18. Her abdomen is soft, nontender, and consistent with a 30 week pregnancy. Which of the following is the most likely cause of this patient’s symptoms?
A) Abruptio placentae
B) Placenta previa
C) Threatened abortion
D) Velamentous cord insertion
Placenta previa is the presence of placental tissue overlying the internal cervical os and is a common cause of painless vaginal bleeding in women beyond 20 weeks gestation. Patients classically present with painless bleeding in the second or third trimester. Most cases of placenta previa are not seen on first-trimester ultrasound, and any woman without a second- or third-trimester ultrasound that presents with antepartum vaginal bleeding needs to have an abdominal ultrasound to assess the location of the placenta. Complications of placenta previa include hemorrhage, preterm labor and rupture of membranes, and fetal malpresentation. Type and screening for blood in patients with placenta previa serve two important purposes. First, mothers that are Rh negative should receive Rho (D) immune globulin after any bleeding episodes, including from delivery, previa, abruption, or threatened abortion, to prevent alloimmunization. Second, since hemorrhage is an important and potentially life-threatening complication of placenta previa, knowledge of maternal blood type is important for giving cross-matched blood.
Abruptio placentae (A) describes the separation of the placenta from the uterine wall prior to delivery. It is the second most common cause of vaginal bleeding after 20 weeks of gestation. In contrast to placenta previa, abruptio placenta is typically associated with abdominal pain. Threatened abortion (C) describes vaginal bleeding before 20 weeks gestation. The patient in the question above is in her third trimester and, therefore, cannot be diagnosed with threatened abortion. Velamentous cord insertion (D) is an uncommon cause of late-trimester bleeding caused by exposed fetal umbilical cord vessels. It is a far less common cause of late-trimester bleeding than placenta previa.
Silver, R. M. (2015). Abnormal Placentation. Obstetrics & Gynecology, 126(3), 654-668. doi:10.1097/aog.0000000000001005
Ananth CV, Wilcox AJ, Savitz DA, Bowes WA Jr, Luther ER. Effect of maternal age and parity on the risk of uteroplacental bleeding disorders in pregnancy. Obstet Gynecol 1996;88:511–6.
Macones GA, Sehdev HM, Parry S, Morgan MA, Berlin JA. The association between maternal cocaine use and placenta previa. Am J Obstet Gynecol 1997;177:1097–100.
Ananth CV, Demisse K, Smulian JC, Vintzileos AM. Placenta previa in singleton and twin births in the United States, 1989 through 1998: a comparison of risk factor profiles and associated conditions. Am J Obstet Gynecol 2003;188:275–81.
Crane JM, Van den Hof MC, Dodds L, Armson BA, Liston R. Maternal complications with placenta previa. Am J Perinatol 2000;17:101–5.
Miller, D., Chollet, J., and Goodwin, T. Clinical risk factors for placenta previa-accreta. American Journal of Obstetrics and Gynecology, 1997(1): 210-214.
Gyamfi-Bannerman, C. (2018). Society for Maternal-Fetal Medicine (SMFM) Consult Series #44: Management of bleeding in the late preterm period. American Journal of Obstetrics and Gynecology, 218(1). doi:10.1016/j.ajog.2017.10.019
Winn, H. N., MD, & Hobbins, J. C., MD. (2000). Clinical Maternal-Fetal Medicine (1st ed.). New York: The Parthenon Publishing Group.
Petpichetchian, C., Pranpanus, S., Suntharasaj, T, et al. (2018). Comparison of transabdominal and transvaginal sonography in the diagnosis of placenta previa. J Clin Ultrasound, 46(6), 386-390. doi:10.1002/jcu.22600
Lockwood, C. J., MD, MHCM, & Russo-Stieglitz, K., MD. (2019, January 30). Placenta Previa: Management. Retrieved from https://www.uptodate.com/contents/placenta-previa-management?search=placentaprevia&source=search_result&selectedTitle=1~102&usage_type=default&display_rank=1
Cotton DB, Read JA, Paul RH, Quilligan EJ. The conservative aggressive management of placenta previa. Am J Obstet Gynecol 1980; 137:687.
Just one case of missed pediatric physical abuse I consider a travesty. The sad state of affairs is that thousands of cases of pediatric physical abuse are missed on initial presentation to EDs across North America. And a small but significant minority of these children die. In fact, 20-30% of children who died from abuse and neglect had previously been evaluated by medical providers for abusive injuries that were not recognized as abuse. Every child that presents to the ED with a suspicious injury gives the treating physician an opportunity to intervene. We have to get better at identifying these kids when there’s still something we can do to protect them, before it’s too late. In this EM Cases main episode podcast on Pediatric Physical Abuse Recognition and Management Dr. Carmen Coombs and Dr. Alyson Holland discuss the 6 B’s of child abuse, the TEN-4 FACE decision rule, the Pittsburgh Infant Brain Injury Score, disclosure tips, screening tests, reporting responsibilities and more…
Podcast production by Anton Helman. Editing and sound design by Richard Hoang & Anton Helman
Written Summary and blog post by Shaun Mehta & Alex Hart, edited by Anton Helman February, 2018
Cite this podcast as: Helman, A, Coombs, C, Holland, A. Pediatric Physical Abuse Recognition and Management. Emergency Medicine Cases. March, 2018. https://emergencymedicinecases.com/pediatric-physical-abuse/. Accessed [date].
Why pediatric physical abuse matters
For those children who survive their physical trauma, there are both short term and long term effects. Short term effects include risk of escalating abuse with increased risk of morbidity and mortality. Long term effects include developmental delays, mental health issues, chronic disease and substance abuse according to the ACEs Study.
Emergency providers’ responsibility in suspected pediatric physical abuse
The unfortunate truth is, if we don’t think about abuse, we’ll definitely miss it. So, the most important first step is always simply to think about it. Then our responsibility turns to reporting it, considering the differential diagnosis, investigating what is medically necessary and ensuring appropriate follow up or admission if necessary.
Sentinel injuries are minor injuries with major significance
Much as with sentinel bleeds of subarachnoid hemorrhages, even seemingly trivial injuries of abuse can be viewed as “sentinel events” for much worse outcomes in the future. According to Sheets et al in 2013, as many as 25% of abused infants had prior sentinel injuries.
One reason we miss physical abuse in these kids is that most sentinel injuries are minor injuries such as bruises, intraoral injuries, or simple fractures and we overlook the real significance of these injuries, which is that they provide an opportunity for us to intervene before more serious injury occurs. As EM providers, recognizing and responding appropriately to sentinel injuries protects kids. Failure to do so can result in repeated abuse and subsequent morbidity and mortality.
Historical Indicators of Pediatric Physical Abuse
No/vague explanation for a significant injury
Important detail of the explanation changes dramatically
Explanation given is inconsistent with the child’s physical and/or developmental capabilities
Different witnesses provide different explanations
Injury occurred as a result of inadequate supervision
Delay in seeking medical care without reasonable explanation
Remember though that no child is immune from abuse and an absence of the above risk factors does not preclude one from investigating suspicious injuries further. Indeed, it is in this population that abuse is often missed.
Physical Exam Findings Suggestive of Abuse: The 6 B’s
As children learn to run, walk and explore new environments, they are bound to hurt themselves. Teasing apart which injuries are due to normal childhood rambunctiousness and those that are a result of abuse requires vigilance and a thorough assessment. Remember to fully undress your patients in order to complete a head to toe exam (Teeuw et al 2012) that won’t miss subtler injuries.
Signs of disciplinary abuse are usually found on areas of the body that are concealed by clothing such as the back and buttocks.
Injuries at different stages of healing
Any injury in a young infant.
Other clues include
Poor child hygiene
A child who appears, anxious, withdrawn or fearful of a person in the room
The most common abusive injury is a bruise. Kids bruise often, but certain scenarios should raise alarm bells.
A) Bruising in the pre-mobile infant.
“If you don’t cruise, you don’t bruise.”
Pierce et al found that only 1.3% of infants less than 5 months old had bruising.
Feldman et al found that over 50% of pre-mobile infants with bruising were victim of abuse.
Make certain you base your assessment on the child’s developmental stage and not absolute age.
Bruising in infants is very rare. Not all of these babies are abused but they almost always (ie, unless there was significant trauma in a public setting with multiple witnesses) warrant further evaluation for child abuse.
Any bruise found in any of the following locations should trigger the possibility of pediatric physical abuse:
Any bruise in a child younger than 4 months old
Angle of Jaw
Pearl: Think of a subconjunctival hemorrhage in an infant as a bruise on the eyeball and frenulum injury as a bruise to the frenulum. These injuries are highly suggestive of abuse in the infant.
C) Patterned Bruises
Linear bruises to buttock (whipping, spanking, paddling)
Linear bruising to the pinna
Hand prints or oval marks
Belt Marks – U-shaped end or associated buckle inflicted puncture wounds
Loop marks (rope, wire, electric cord)
Ligature marks, circumferential rope burns to neck, wrists, ankles and gag marks to comers of the mouth
D) Too Many Bruises
It is helpful when consider whether or not a patient has too many bruises to compare with bruise prone areas such as the shins.
Don’t try to age/date bruises as physicians are unreliable at determining the age of bruises.
While there are no fractures that are pathagnomonic for abuse, any fracture in any age group can be abusivedepending on how the fracture was sustained. Most abusive fractures occur in children < 18 months of age so again, our highest level of suspicion should be in young children. An important exception is that of the Toddler’s fracture which requires little force and can occur from simple falls.
1. Any fracture in a nonambulatory infant or child
2. Femur fracture in an infant < 12-18 months of age
Baldwin 2011: The odds of a femur fracture being abuse rather than accidental trauma was 19 times greater for children < 18 mo of age.
3. Humerus fractures in an infant < 18 months of age
Pandya 2010: Children < 18 mo with a humerus fracture had a 32 greater odds of being the victim of abuse.
Location: Proximal and mid shaft humeral fractures are more likely due to abuse whereas distal humerus/supracondylar fractures are less likely to be due to abuse.
4. Multiple fractures and/or an unexpected healing fracture
5. Skull fractures, especially if complex or bilateral
Deye 2013: A small (~5%) but not insignificant number of infants < 12 months old with apparently isolated skull fracture undergoing abuse evaluation were found to have additional unexpected fractures on skeletal surveys.
Most acute fractures are symptomatic but there are some fractures (metaphyseal fractures and rib fractures) which are often occult and only identified on imaging and are highly concerning for child abuse. These fractures may be found “incidentally” when getting imaging for a different reason (ie, a rib fracture seen on a chest xray obtained in an infant with suspected pneumonia) but need to be addressed the same as any abusive injury.
6. Classical metaphyseal fractures (bucket handle fractures) from being shaken violently back and forth
7. Rib fractures, especially posterior rib fractures (highest probability for abuse)
Every injury can be caused by abuse. Nothing is pathognomonic for abuse.
Signs of abusive head trauma can be subtle and non-specific. The Pittsburgh Infant Brain Injury Score (PIBIS) by Berger et al (see below) can help decide which patients warrant head imaging.
Skull fractures are most often accidental, but a small proportion are associated with abuse according to Deye et al in 2013. Increase your suspicion for abuse if skull fractures are complex, bilateral, depressed, open, presenting with suture diathesis or occipital fractures.
Human bites have typical, stereotyped pattern.
Baby Blues (Irritability)
Some severe injuries (see head trauma) can present with very non-specific symptoms. This is why it is so important to fully examine all concerning patients.
Screening for Medical Conditions Mimicking Abuse
The two most common conditions that mimic bruising are Mongolian spots and hemangiomas.
Compared to a bruise, Mongolian spots (congenital dermal melanocytosis) are bluish-green and located on the buttocks/lower back, present at birth, are non-tended, and fade over months to years. Hemangiomas are also non-tender, subcutaneous, and proliferate over time. If in doubt, use POCUS!
Many conditions may predispose a child to bruising and bleeding including HSP, ITP, leukemia and hemophilia, and consider osteogenesis imperfecta and rickets in children who present with fractures as a result of a minimal force mechanism.
Screening for occult injuries: The workup in suspected pediatric physical abuse
The basic medical evaluation for suspected physical abuse in a young child includes a skeletal survey, head imaging, and trauma labs.
Which children suspected of abuse require a skeletal survey?
Detailed guidelines for skeletal survey in young children can be found here.
As a general guide: Any child < 2 yo in whom you suspect abuse should undergo a skeletal survey.
Pediatric abusive head trauma
Head trauma in young children may present with non-specific symptoms. Signs and symptoms of abusive head trauma in an infant or young child can be subtle or non-specific (sleepy, vomiting, fussy, lethargic) or even completely absent and these are the cases we often to fail to recognize.
Pittsburgh Infant Brain Injury Score (PIBIS) for Abusive Head Trauma
The 5-point PIBIS
Abnormality on dermatologic examination (2 points),
Age ≥3.0 months (1 point),
Head circumference >85th percentile (1 point), and
Serum hemoglobin <11.2g/dL (1 point).
At a score of 2, the sensitivity and specificity for abnormal neuroimaging was 93.3% (95% confidence interval 89.0%–96.3%) and 53% (95% confidence interval 49.3%–57.1%), respectively.
3 Ds – Documentation, Disclosure & Disposition in pediatric physical abuseDocumentation in pediatric physical abuse
Proper documentation can be challenging but is extremely important in these scenarios. The chart is a medicolegal document that may be called upon if an investigation takes place.
Who is providing the history
What, when, who
Use quotations to document exact statements from child and caregiver
Any pain that the child is experiencing
Activities that may affect forensic evidence recovery (eg. bathing)
Review of systems – changes in behaviour, non-specific symptoms
The usual (past medical history, social history, meds, allergies)
Fully expose the child – this is a trauma patient
Describe, draw or even photograph any injuries
If comfortable, offer an interpretation of the findings in the context of the history
Disclosure tips in pediatric physical abuse
Be direct and professional. “As a physician, I worry when I see X, Y and Z and it makes me concerned that someone may have hurt your child.”
Refrain from being accusatory. “It’s not my role to say who hurt your child but it is my obligation to report my concern.”
Encourage the family to focus on the child. “Right now, we need to make sure that your child gets the medical care that he/she needs.”
Call for help. Discuss the case with social work, child protective services, a child abuse consultant (eg. SCAN team), and the primary care physician.
Disposition in pediatric physical abuse
To admit or not to admit? If you are concerned about the child’s safety, then admit to hospital for observation and further work up. Discharge from ED only if the medical evaluation is complete and safe disposition can be arranged by child protective services from the ED.
What is our responsibility in reporting of suspected pediatric physical abuse?
The specific rules around reporting abuse vary across the jurisdictions, but the fundamental principle is the same: health care providers need only reasonable suspicion that a child is being harmed to report, not “proof”. We are penalized for failing to report suspicion, not for reporting unsubstantiated suspicion. Reporting to the provincial or state child welfare office is legally mandated. Reporting to pediatric specialists and healthcare centers specializing in child abuse pediatrics is often useful but does not fulfill that legal mandate.
Take home points for Detection and Management of Pediatric Physical Abuse
Sentinel injuries are minor injuries with major significance. These may appear minor, but are often missed and provide an opportunity for us to intervene before serious injury.
No child is immune from abuse. While risk factors for abuse may be helpful in triggering a suspicion for abuse, many children who suffer from abuse have no identifiable risk factors.
The most common abusive injury is a bruise. If you don’t cruise, you don’t bruise. Keep an eye out for the 6 B’s of child abuse.
Head trauma in young children may present with non-specific symptoms. Be suspicious in the right clinical context.
Concerning features for abuse are subtle. These include delay in seeking care, vague or changing history of trauma, mismatch of history and injury or developmental stage of child, and multiple or patterned injuries.
Certain injury patterns should always raise concern for abuse. These include posterior rib fractures, subdural hematomas, immersion burns, patterned bruises and others. None are pathognomonic. Use the TEN-4 FACE decision tool.
The basic medical evaluation for suspected physical abuse in a young child includes a skeletal survey, head imaging, and trauma labs. There are recommendations when and when not to investigate.
Report abuse when you are suspicious – it is your legal responsibility. Regional guidelines vary, but report based on suspicion, not on proof.
Author: Jeremy Carter, DO RVT (Chief Resident, Rutgers New Jersey Medical School) // Editors: Stephen Alerhand, MD (@SAlerhand; Assistant Professor, Rutgers New Jersey Medical School); Brit Long, MD (@long_brit); and Manpreet Singh, MD (@MprizzleER)
Several prior studies have examined the role of point-of-care ultrasound (POCUS) for the evaluation of suspected acute lower extremity deep vein thrombosis (DVT). To begin with, when performed by emergency physicians, POCUS has been shown to decrease emergency department (ED) time to disposition when compared to scans performed by imaging specialists.1 In addition, 2-point (common femoral and popliteal vein) and 3-point (2-point plus sapheno-femoral junction) compression have been shown to be very accurate when compared to both ultrasound of the full lower extremity venous system (performed by trained imaging specialists) and CT venography.2-9 However, in not scanning the length of the femoral and deep femoral vein, about 6% of DVT’s may be missed.10 Many studies also cite the need for a significant amount of training and practice to become proficient in these exams.2-5,8
During the years before medical school, I trained and worked as a vascular sonographer. This has given me additional insight into the potential obstacles and pitfalls encountered when attempting to accurately rule out lower extremity DVT at the bedside. Colleagues have shared their clips and experiences with me, and these have demonstrated several themes. Below are five potential pitfalls accompanied by teaching pearls that may be of assistance:
The groin area is sensitive to compression due to pain, feeling ticklish, or both. As a result, the patient may hold their breath and perform Valsalva, making it difficult to compress the vein. This gives the false appearance of a non-compressible right common femoral vein.
Contrast this with a “true positive” non-compressible femoral vein secondary to an acute DVT.
Instruct the patient to take a deep inhalation and then exhale. Time your compression to exhalation.
See the fully compressible right common femoral vein when timing compression with exhalation.
The femoral vein in the mid-to-distal thigh dives through Hunter’s Canal. The vessel’s depth makes it difficult to visualize and compress, giving the false appearance of a DVT.
Have the patient straighten the leg and internally rotate the hip, from the usual externally rotated position. Slide the transducer anteriorly on the patient’s thigh, angling straight down towards the bed. Place the free non-scanning hand under the thigh, and squeeze up towards the transducer for easier compression.
Adjust the depth so that the vein to be imaged lies in the center of the screen. If the machine allows, adjust the focus to about the center or just below the vein to be imaged. Use the curvilinear transducer as an alternate option.
Making incorrect assumptions without using adjunctive or confirmatory techniques.
Turn the transducer to view the vessel in long-axis. Look for internal echoes from this other perspective.
In addition, insonate the vessel with color Doppler and pulsed wave Doppler .
The Doppler equation includes the cosine of the angle between the ultrasound beam and the flow in a vessel. If the angle between the ultrasound beam and blood flow is 90 degrees, no flow will be seen, because the cosine of 90 degrees is zero.
Thus, for the best Doppler signal, orient the vessel in long-axis and rock the transducer to make the vein cross the screen at an angle instead of horizontally across, so that the sound waves no longer insonate the vessel at a perpendicular angle.
If feasible on your machine, “steer” or change the orientation of the color box to keep the angle between the US beam and vessel as small as possible.
By rocking the transducer superiorly or inferiorly, the transverse orientation of the vessel can also be evaluated using color Doppler.
Theodoro D, Blaivas M, Duggal S, Snyder G, Lucas M. Real-time B-mode ultrasound in the ED saves time in the diagnosis of deep vein thrombosis (DVT). The American Journal of Emergency Medicine. 2004;22(3):197-200. doi:10.1016/j.ajem.2004.02.007
Frazee BW, Snoey ER, Levitt A. Emergency Department compression ultrasound to diagnose proximal deep vein thrombosis. The Journal of Emergency Medicine. 2001;20(2):107-112. doi:10.1016/s0736-4679(00)00302-4.
Crisp JG, Lovato LM, Jang TB. Compression Ultrasonography of the Lower Extremity With Portable Vascular Ultrasonography Can Accurately Detect Deep Venous Thrombosis in the Emergency Department. Annals of Emergency Medicine. 2010;56(6):601-610. doi:10.1016/j.annemergmed.2010.07.010.
Crowhurst TD, Dunn RJ. Sensitivity and specificity of three-point compression ultrasonography performed by emergency physicians for proximal lower extremity deep venous thrombosis. Emergency Medicine Australasia. 2013;25(6):588-596. doi:10.1111/1742-6723.12155.
Mulcare MR, Lee RW, Pologe JI, et al. Interrater reliability of emergency physician-performed ultrasonography for diagnosing femoral, popliteal, and great saphenous vein thromboses compared to the criterion standard study by radiology. Journal of Clinical Ultrasound. 2016;44(6):360-367. doi:10.1002/jcu.22338.
Fulvio P, Dentali F. Accuracy of real time B-mode ultrasonography in the diagnosis of deep vein thrombosis in hip surgery. Thrombosis Research. 1991;61:70. doi:10.1016/0049-3848(91)90542-5.
Burnside PR, Brown MD, Kline JA. Systematic Review of Emergency Physician–performed Ultrasonography for Lower-Extremity Deep Vein Thrombosis. Academic Emergency Medicine. 2008;15(6):493-498. doi:10.1111/j.1553-2712.2008.00101.x.
Shiver SA, Lyon M, Blaivas M, Adhikari S. Prospective comparison of emergency physician–performed venous ultrasound and CT venography for deep venous thrombosis. The American Journal of Emergency Medicine. 2010;28(3):354-358. doi:10.1016/j.ajem.2009.01.009.
Pedraza García J, Valle Alonso J. Comparison of the Accuracy of Emergency Department-Performed Point-of-Care-Ultrasound (POCUS) in the Diagnosis of Lower-Extremity Deep Vein Thrombosis. J Emerg Med. 2018 May;54(5):656-664.
Adhikari S, Zeger W, Thom C, Fields JM. Isolated Deep Venous Thrombosis: Implications for 2-Point Compression Ultrasonography of the Lower Extremity. Annals of Emergency Medicine. 2015;66(3):262-266. doi:10.1016/j.annemergmed.2014.10.032.
Special thanks to Mikhail Blyakher MD and Michael Anana MD for assistance in the production of this post’s media.
Welcome back to the EM Educator Series. These posts provide brief mini-cases followed by key questions to consider while working. The featured questions provide important learning points for those working with you, as well as vital items to consider in the evaluation and management of the specific condition discussed.
This week has another downloadable PDF document with questions, links and answers you can share with learners as educators in #MedEd. Please message us over Twitter and let us know if you have any feedback on ways to improve this for you. Enjoy!
A 53-year-old male presents with left lower leg pain and pallor. The pain worsened suddenly this morning. He has a history of atrial fibrillation, coronary artery disease, hypertension, and hyperlipidemia.
Which patients are at risk for an ischemic leg?
What etiologies should you consider? Thrombosis, embolism, and what else?
What is the differential for patients with pain out of proportion to exam?
How do you diagnose an acute ischemic limb? What roles do ABI, US, and CT with IV contrast play in diagnosis?
What is the ED management, and who can assist you in the ED (IR, vascular surgery)?
Authors: Andrew Lin, MD (Emergency Medicine Resident, Johns Hopkins Emergency Medicine) and Drew Clare, MD (Attending Physician, Johns Hopkins Emergency Medicine) // Edited by: Alex Koyfman, MD (@EMHighAK) and Brit Long, MD (@long_brit)
Picture this: you’re in the Emergency Department, nearing the end of your shift, when you’re called to see an eye complaint. It sounds like a typical low-acuity eye complaint: this 57-year-old patient presents with a vague chief complaint of having difficulty focusing on distant objects. She notices it has been present for a number of days and notes no other pertinent complaints. She has a history of hypertension and diabetes and has otherwise had no health issues. She sees her doctor regularly. You do your due diligence and perform a thorough review of systems – asking about headache, weakness, and hearing loss. You ask about diplopia, and to your surprise, the patient mentions that she has noticed that she does intermittently see double – especially when she is looking in certain directions.
Diplopia is most often taught as a red flag during the comprehensive workup of many of our most common chief complaints – generalized weakness, headache, and almost any kind of visual complaint. This is with good reason, as diplopia can often be the herald symptom for many debilitating conditions that we are tasked with diagnosing in the emergency department. It is critical for the emergency physician to evaluate this finding thoughtfully and thoroughly to rule out serious pathology. A straightforward way to break down the etiology of diplopia is to first characterize the finding into either monocular diplopia or binocular diplopia, which is the first step in your evaluation. This is determined by having the patient cover each individual eye to see if the diplopia persists. If the diplopia persists when only the unaffected eye is covered, but resolves when only the affected eye is covered, this is monocular diplopia. If the diplopia resolves when either eye is covered, this is binocular diplopia. Monocular diplopia often occurs because of an abnormality in the affected eye itself. This is often due to refractive error, though CNS pathology and retinal disease are also possibilities.  Binocular diplopia, on the other hand, is often due to neuromuscular dysfunction causing misalignment of visual fields. Causes commonly include cranial nerve palsy, ocular muscle dysfunction, or brainstem pathology. Ergo, monocular diplopia can be appropriately managed with an ophthalmological referral and typically does not require neuroimaging. Binocular diplopia, on the other hand, requires further evaluation in the emergency department to rule out acute neurological insult.
Your next step is to determine if the diplopia is isolated or associated with neurologic signs/symptoms. In better building our differential and subsequent workup, the most important part of evaluation is thus a comprehensive neurological exam. The presence or absence of additional neurological symptoms and/or findings in addition to diplopia (vertigo, ataxia, motor or sensory deficits, or bulbar symptoms such as dysphagia or dysarthria) should prompt expedited neuroimaging and consultation with neurology for evaluation for stroke . For example, if the patient presents with diplopia and a thorough neurological exam suggests anything beyond an isolated nerve palsy (of the extraocular muscles – 3, 4, and 6), activation of the stroke protocol should be the first step. Neuroimaging is needed, preferably MRI to evaluate for early signs of stroke and investigate the posterior circulation.
Beyond this crucial branch path in our evaluation of diplopia, our next task lies with localizing the lesion that best explains the diplopia and evaluating for obvious abnormality in ocular motility. Careful evaluation of 3rd and 6th nerve palsies is necessary. Testing the extraocular muscles is a critical step in evaluating where the problem may lie; the physician can often reproduce the diplopia by asking the patient to look in both vertical and horizontal directions. This can help you identify which nerve (or extraocular muscle) is abnormal and serves as the basis of the rest of our workup. In addition to extraocular movements, attention should also be paid to the eye itself, particularly the pupils. A schematic below reviews stereotypical findings for extraocular palsies :
Physical exam findings of various palsies; image courtesy of the BMJ
At this point, one has identified exactly what neurological deficits our patient has, paying close attention to cranial nerves 3 and 6. This will guide further management.
Isolated nerve palsies are the most common cause of diplopia and are typically caused by microvascular insult. This ischemia leads to dysfunction of the extraocular muscles, which causes a gaze palsy. These patients often have many risk factors for vascular disease and require further risk factor modification, but they typically do not require acute therapeutic or imaging. However, the diagnosis of binocular diplopia secondary to microvascular pathology is one of exclusion: there are a myriad of conditions the emergency physician needs to consider. Ultimately, the need for neuroimaging in diplopia will depend on the palsy in question as well as the presence of other abnormalities in the exam or history.
First and foremost, if there is any suspicion of trauma in the history, imaging should be ordered for further evaluation. Direct trauma to the orbit (including the surrounding muscles and nerves) or involvement of the orbital apex (a space posterior to the orbit where CN 3, 4, 6 pass) is worrisome. An orbital floor fracture causing entrapment of extraocular muscles or a retrobulbar hematoma are also possibilities that warrant further imaging. A CT of the head and orbits is appropriate in this setting to rule out structural abnormalities. Contrast should be included if infection is suspected.
Second, an isolated third nerve palsy should always be further imaged with a non-contrast head CT as well as a CTA. While third nerve palsies are often caused by ischemia, approximately 6% of third nerve palsies are caused by a compressive aneurysm. This is often at the junction of the posterior communicating artery and internal carotid artery . Traditionally it was thought that clinical exam could distinguish between a compressive and ischemic etiology. The fibers responsible for constriction of the pupil run on the outer portion of the nerve. An ischemic insult, affecting the central portion of the nerve, would therefore spare pupil constriction. External compression from an aneurysm on the other hand directly impacts the peripheral portion of the nerve, which is responsible for pupillary constriction. Clinical exam is not sufficient, however, as a large population study found that almost 36% of compressive third nerve palsies were found to be pupil-sparing. This supports the need for CT and CTA of the head in any presentation of third nerve palsy . Palsies of the 4th and 6th cranial nerves, in the absence of any other additional concerning findings (especially focal neurological deficits), can be referred to neurology without further imaging. Both 4th and 6th nerve palsies are largely micro-vascular in etiology, which support no need for further evaluation. A few caveats: 6th nerve palsies, by virtue of their anatomic location, have a tendency to be associated with increased intracranial pressure, necessitating a good funduscopic exam to evaluate for papilledema. Additionally, the prevalence of malignancy over microvascular disease in children suggest that even in isolated 6th nerve palsies, children should always receive emergent imaging .
Finally, there are a few additional conditions that require additional workup in the Emergency Department. Cavernous sinus thrombosis is a veno-occlusive episode of the cavernous sinus, which houses many cranial nerves, most importantly 3, 4, and 6 . These patients often present acutely ill, with swelling and pain in the affected eye (the impetus of the infection), possible visual acuity deficits, headache, and multiple risk factors for thromboembolic disease. These patients should obtain CT venogram immediately, along with broad spectrum antibiotics, and blood cultures. CT venogram is very specific, however, it is not sensitive enough to exclude the diagnosis. If the CT venogram is negative, emergent MRV is warranted. Other pathologies that uncommonly present with palsy are thyroid eye disease and Giant Cell Arteritis. Thyroid disease can lead to enlargement and fibrosis of the extraocular musculature, while Giant Cell Arteritis can cause microvascular ischemia from vasculitis. These conditions typically present with typical associated symptoms of their disease process (i.e., you would also expect to see systemic signs of thyroid disease, or unilateral headache in vasculitis). If these conditions are suspected, a TSH or ESR/CRP, respectively, is appropriate. Patients over 50 should have inflammatory markers sent. Myasthenia Gravis is another condition that can present with diplopia, though these patients often present with other symptoms (proximal muscle weakness, respiratory issues), and the diplopia may be intermittent and likely fatigable rather than persistent.
Once the aforementioned acute pathologies have been ruled out, an urgent referral to an ophthalmologist or neurologist is usually warranted; many lesions require further evaluation with non-emergent MRI or advanced ophthalmological techniques. This is especially the case in the setting of complex motility disorders: these are eye movement disorders that present with neurologic findings not localizable to a single cranial nerve. These often require specialty evaluation and imaging to better characterize the deficit involved.
In summary, the general approach to diplopia in the emergency department is as follows:
Delineate if the diplopia is Monocular or Binocular; the latter requires further evaluation in the emergency department. The former can be referred to ophthalmology.
Once Binocular Diplopia has been established, if there are any other focal neurologic findings outside of Diplopia, CVA should be heavily considered and the patient should be evaluated for stroke.
Perform an extraocular exam to ascertain which cranial nerves in question are malfunctioning.
Isolated 4th nerve palsies and 6th nerve palsies can be referred to a specialist for evaluation and do not require imaging in the ED.
The presence of 3rd nerve palsy, multiple concomitant palsies, evidence of papilledema, infection, trauma, or cavernous sinus thrombosis require urgent imaging in the ED.
To conclude the opening case, the patient in question had an impressive sixth nerve palsy; her diplopia was easily reproduced on lateral gaze to the right. However, on physical exam, she demonstrated some difficulty with alternating movements on the right side. In light of these additional neurologic findings, she was sent for a CT followed by an MR/MRA, which demonstrated a left non-hemorrhagic pontine infarct.
In summary, diplopia is a complaint we see infrequently in the emergency department, but one that requires a thorough neurologic and ophthalmologic assessment to rule out debilitating neurological disease. The patient that presents with diplopia should not be imaged routinely, unless a specific underlying etiology is suspected.
Take Home points
Diplopia is best characterized as monocular or binocular; the latter is generally more concerning and warrants further workup in the ED.
Diplopia, in addition to any other neurologic deficit, is grounds for a stroke activation and assessment.
Diplopia does not always require imaging; however certain associated findings warrant specific evaluation:
Suspected Cavernous Sinus Thrombosis: CTV, followed by MRV if CTV negative
Patients over age 50 with diplopia should have inflammatory markers ordered.
Diplopia merits urgent referral to an ophthalmologist or neurologist once immediate, intervenable causes have been ruled out.
 Gerstenblith, Adam T.; Rabinowitz, Michael P., May 27, 2012, Wills Eye Manual: Office and Emergency Room Diagnosis and Treatment of Eye Disease Wolters Kluwer Health, Philadelphia, ISBN: 9781451186024
 Nazerian P, Vanni S, Tarocchi C, Portaccio E, Vannucci N, Para O, Giannazzo G, Gigli C, Grifoni S. Causes of diplopia in the emergency department: diagnostic accuracy of clinical assessment and of head computed tomography. Eur J Emerg Med 2014;21:118–124
 Low Liying, Shah Waqaar, MacEwen Caroline J. Double vision BMJ 2015; 351:h5385
 Danchaivijitr C, Kennard C. Diplopia and eye movement disorders. J Neurol Neurosurg Psychiatry. 2004;75 Suppl 4(Suppl 4):iv24–iv31. doi:10.1136/jnnp.2004.053413
 Dotan G, Rosenfeld E, Stolovitch C, Kesler A: The role of neuroimaging in the evaluation process of children with isolated sixth nerve palsy: Child’s nervous system. Childs Nerve Syst 2013;29:1:89-92.
Use of insulin/glucose to treat hyperkalemia works, but hypoglycemia is a common side effect. Here are some pearls to give this treatment more safely.
Why does this matter?
Hyperkalemia is a life-threatening condition that requires prompt management in the ED. One of the most common treatment options is the administration of insulin and glucose to help shift potassium into the cell temporarily. Usually this is ordered as 10 units of regular insulin IV and 1 ampule of D50. This article explores some common myths and debunks them.
“Pour some sugar on me!”
Consider decreasing insulin dose (5 units or 0.1 U/kg) or increasing dextrose load (50 g) in patients with one or more of the following risk factors:
Pretreatment blood glucose (BG) < 150 mg/dL
Acute kidney injury/chronic kidney disease
No history of diabetes mellitus
Weight < 60 kg
An insulin dose of 5 units (or 0.1 U/kg) has similar efficacy as 10 units and may be safer. The exception to this was seen in a subset of patients with initial potassium levels > 6 mmol/L, where the 10 U group was superior to the 5 U group (K decreased by 1.08 mmol/L vs 0.83 mmol/L, respectively, p=0.018).
A dose of 25 g of dextrose (1 amp of D50) may be inadequate for hypoglycemia prevention, especially in patients with kidney disease and decreased clearance of insulin. Consider a higher dose of dextrose (2 amps, 50 g) or giving it with the insulin as a 4h infusion to decrease the incidence of hypoglycemia.
You should monitor blood glucose levels hourly for 4-6h to match the 4-6h duration of regular insulin when given IV. D50 IV boluses only last around one hour.
Spoon Feed Intranasal tranexamic acid (TXA) with nasal compression was as effective as Merocel nasal packing for bleeding cessation in anterior epistaxis, had a lower rebleeding rate, and was better tolerated than nasal packing.
Why does this matter? TXA is an antifibrinolytic agent used to improve hemostasis and decrease bleeding. Several studies have demonstrated the utility of TXA for management of anterior epistaxis, usually by soaking nasal packing materials in TXA before insertion. However, since nasal packing has several disadvantages, could topical intranasal TXA be used in lieu of nasal packing for anterior epistaxis management?
Can TXA help stop the bleeding? Only the nose knows… This was a single center, 135-patient randomized controlled trial that compared the effectiveness of external nasal compression after application of topical intranasal TXA, external nasal compression after application of topical intranasal saline (placebo-controlled group), and Merocel anterior nasal packing for stopping spontaneous anterior epistaxis with a secondary outcome of rebleeding within 24 hours. TXA was administered as atomized 500 mg diluted in 5 mL normal saline solution sprayed into both nostrils.
The success rate for stopping anterior epistaxis within 15 minutes was 91.1% (41 of 45 patients) in the nasal compression with TXA group, 93.3% (42 of 45 patients) in the nasal packing group, and 71.1% (32 of 45 patients) in the nasal compression with normal saline group. There was no statistically significant difference between the tranexamic acid and nasal packing groups.
Rebleeding within 24 hours occurred in 13.3% of patients in the TXA group, 26.7% in the nasal packing group, and 40% in the compression with saline solution group. Furthermore, 7 out of 45 patients in the nasal packing group had severe pain during nasal packing and requested the procedure be terminated. None of the patients in the TXA group complained of severe pain.
This study had several limitations and importantly only one nasal packing material (Merocel) was used for comparison. However, this study suggests that TXA application with nasal compression was as good as standard nasal packing at stopping bleeding, better at preventing rebleeding within 24 hours, and may be more comfortable and cost-effective than ready-made commercial anterior nasal packing.
Infants ≤60 days of life with hypothermia (< 36 °C) had serious bacterial infection (SBI) 2.8% of the time. There is a suggestion that those 15-28 days with high absolute neutrophil count (ANC) and thrombocytopenia had even greater odds of SBI.
Why does this matter?
We know fever is often bad news in young infants. We know hypothermia is as well. But sometimes infants are exposed too long after a bath and have a transient drop in temperature. Are there other clinical factors that would raise our suspicion for SBI in hypothermic infants ≤60 days?
Hypothermia isn’t cool
This was a retrospective review at a single center over 12 years that identified 360 hypothermic (< 36 °C) infants ≤60 days, of which 2.8% (10/360) had SBI. Using multiple logistic regression, they found three variables that were statistically significant (though a little iffy).
Infants 15-28 days had greater odds of SBI than 0-14 days. aOR 7.60 (95%CI, 1.81-31.86)
“Higher” ANC had greater odds of SBI. aOR 1.25 (95%CI, 1.04-1.50). It was not clear in the full text – “higher” ANC than what? The median ANC in the SBI patients was 2.96 vs. 2.90 in the non-SBI infants; average in the SBI group was 5.19. But there were 7/10 with ANCs 1.6-4.7, which is very reassuring. So, I don’t know how practical this is. We will learn more about ANC cutoffs tomorrow.
Finally, “lower” platelet count also increased odds of SBI. aOR 0.99 (95%CI, 0.99-1.00). Again, I’m not sure – “lower” platelet count than what? Median was 213 in those with SBI; 298 without SBI.
What’s the take home here? Infants ≤60 days with hypothermia had SBI nearly 3% of the time. Although this is a lower rate than in febrile infants, I think we probably need to treat them similarly. If they are over 2 weeks, ANC is really high, or platelet count really low, that’s even more concerning.
Although the podcast won’t drop until next Sunday July 7, Rob Orman has a pretty comprehensive ERcast episode on infant fever you won’t want to miss. This link gets you a discount and JF an Amazon gift card. Nice!
1. Protocolized daily US rounds in the ICU may lead to modest improvements in outcomes such as length of stay and duration of mechanical ventilation.
2. Larger studies are needed to confirm these results, determine if mortality benefit, and suggest the mechanism for these benefits.
We know that severe/sepsis and/or septic shock are bad and those with either are at increased risk of death. We also know the POCUS use is increasing in the ICU and can be helpful in obtaining information to aid in medical decision making. Most trials using ultrasound in these settings use it as an on-demand tool, i.e. used only when there are signs or symptoms suggesting hemodynamic or respiratory change. US is also more prevalently used during the initial or unstable stage of care and less so when the patient is more stable. The authors goal was to see if routine use of US during morning rounds on all patients would improve patient outcomes.
Does routine use of POCUS during morning rounds improve patient outcomes?
Single University Hospital in Hangzhou China (Sir Run Run Shaw)
Patients with sepsis (defined as infection + signs of organ dysfunction)
Advanced stage of malignancy
Younger than 18 years old
Missing values on mortality outcome
Prospective case control study
Three medical teams in the ICU (Attending, two fellows, and three residents).
Two served as control and used POCUS as needed and third was intervention group that did daily US rounds.
They used a checklist for their scans developed under the international guideline of critical care ultrasonography
Treatment changes were recorded and reviewed.
Primary outcome was hospital mortality. Secondary outcomes were ICU length of stay, duration of mechanical ventilation, duration of vasopressors.
Multivariable logistic regression taking into account APACHE II score, age, mechanical ventilation, vasopressors, infection site, and fluid balance through day 3.
Who did the ultrasounds?
Providers in the intervention group.
Had a one week course endorsed by WINFOCUS, attendings, fellows, residents
They used all transducers, different ones for the various scans below
No significant differences in patient population between Intervention (n=41) and Control (n=88)
No statistically significant difference
Mortality trended lower in the intervention group (0.27 vs 0.43, p=0.113)
Shorter duration of mechanical ventilation in intervention (POCUS) group
5.7 ±1 vs 4.5 ±1.2, p= 0.034
Vasopressors and Length of Stay
No significant different in vasopressor days
No significant difference in LOS in ICU (7.2 control Vs 6.9 intervention, p = 0.055)
Sig difference in fluid balance on day 3
Control 48 ml vs Intervention -143 ml)
Unclear because also stated the difference was “a few hundred milliliters” in discussion
POCUS used more in intervention group
283 times in Intervention group
53 times in control
Modified diagnosis in 8 patients (19.5%) confirmed suspicion in 21 (51.2%).
Changed fluid management 81 times (28.6%), vasoactive agents 53 times (18.7%), added diuretics 71 times (25.1%)
In logistic regression model POCUS significantly reduced the likelihood of prolonged ICU stay (OR 0.39). There was also a significant difference in fluid balance.
Mechanical ventilation and Vasopressors were associated with mortality and ICU stay.
Outside of US, single center – may have other factors affecting care. This limits external validity.
Not randomized or blinded. Did the intervention group try harder to wean MV or pressors? The differences between groups seen in this data could be due to practice variability between the intervention attending vs the control attendings. They did try to mitigate this through protocols.
Use of routine POCUS did not show a mortality benefit as intended in the study but did show a shortened duration of mechanical and vasopressor use. Why? This could be because small sample size or that it may not matter. Only larger, more well-designed studies will tell.
Fluid balance was significantly less in intervention group on days 2 and 3. Could this be the cause of the difference in length of stay or vasopressor use? On the other hand, the balance was different by a few hundred milliliters – unlikely of clinical significance. Intuitively, having a POCUS protocol could be a few benefits to the patients – 1) early diagnosis and treatment of causes or sepsis or complications 2) Fluid management. Perhaps there is a certain patient population that would benefit most from this protocol – those with an unclear diagnosis or unclear fluid status. Perhaps isolating these patients would show a more robust benefit.
Take Home Points
1. Protocolized daily US rounds in the ICU may lead to modest improvements in outcomes such as length of stay and duration of mechanical ventilation.
2. Larger studies are needed to confirm these results, determine if mortality benefit, and suggest the mechanism for these benefits.
Authors: Mary Rometti, MD (EM Resident Physician, Rutgers-Robert Wood Johnson Medical School) and Laryssa Patti, MD (EM Attending Physician, Rutgers-RWJMS) // Edited by: Tim Montrief, MD (@EMinMiami), Alex Koyfman, MD (@EMHighAK), and Brit Long, MD (@long_brit)
A 29-year-old female presents to the ED with shortness of breath for two days, which is worse with exertion or laying down. About one week ago, she delivered a healthy, full-term newborn after an uneventful cesarean section due to non-reassuring fetal heart tones. Exam reveals decreased breath sounds in bilateral lung bases, bilateral lower extremity edema, and SpO2 in 80’s on room air. Given these findings, what’s your approach?
Described in the patient presentation above, peripartum cardiomyopathy (PPCM) is a type of dilated cardiomyopathy with left ventricular (LV) dysfunction in the immediate peripartum period during the last month of pregnancy or in the postpartum period within the first 5 months after delivery.1-4 Most cases of peripartum cardiomyopathy (PPCM) present with signs and symptoms of heart failure, such as orthopnea, paroxysmal nocturnal or exertional dyspnea, and fatigue.1,3-7 Pregnant patients with pre-existing cardiovascular conditions that precede heart failure may present with similar symptoms, making a diagnostic conundrum for clinicians. To be diagnosed with PPCM, patients must not have a previous diagnosis of heart failure or reduced cardiac function. According to one study, approximately 56% of PPCM cases will present within the first week after delivery, and up to 75% of patients will be diagnosed within the first month postpartum.1,8 Within the U.S., 1,000 to 1,300 women will develop PPCM each year,2 and up to 1 in 1,000-4,000 U.S. live births per year result in the mother developing PPCM.1,3 Risk factors include older maternal age, African American race, presence of maternal pre-eclampsia or hypertension (HTN) during pregnancy, and a multiple gestation pregnancy.1,3,5 More than half of PPCM cases occur in southern U.S, possibly due to more African American women living in the South.1
Why does PPCM happen?
As a form of left heart failure, in PPCM the heart chambers dilate, thereby leading to lower LV ejection fraction (LVEF) and subsequent poor perfusion.1,3 The exact cause of peripartum cardiomyopathy is unknown, but current theories suggest it is most likely related to genetics and hormones present in late pregnancy. During late pregnancy, the placenta releases multiple chemical signals, including prolactin and tyrosine kinase, which may increase oxidative stress on cardiac vasculature, endothelium, and myocytes, ultimately decreasing cardiac function.1,3,6,9 Angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) potentially help combat this hormonal response by afterload reduction.1,5 Other less likely causes include viral myocarditis, nutritional deficiencies, autoimmunity, fetal microchimerism, and hemodynamic stresses.1,4-6
What else should you consider?
Other diagnoses to consider include:1,3-7,15
Acute coronary syndrome (ACS)
Other non-dilated cardiomyopathies, including Takotsubo cardiomyopathy
Amniotic fluid embolism
How do you work it up?
The main diagnostic criteria of PPCM, per the National Heart, Lung, and Blood Institute, is the presence of heart failure within the last month of pregnancy to up to five months following delivery, as defined by a LV EF < 45%, with no other rationale for a reduced EF and history of pre-existing heart disease.1-4,6,7,9 With these diagnostic criteria in mind, the ED evaluation initially involves excluding other possibilities for the patient’s presentation. An electrocardiogram (EKG) usually shows normal sinus rhythm or sinus tachycardia, but it may also show nonspecific ST segment or T wave changes.1,3 Having these non-specific ST and T wave changes may correlate with myocardial ischemia risk profiles.9 An EKG is most useful to aid in ruling out other cardiac etiologies, such as ACS.1,3,7,9 A chest x-ray typically demonstrates findings of heart failure, including cardiac enlargement or pulmonary congestion. There are no pathognomonic EKG or CXR findings that definitively rule in or out PPCM.1,3,7
Although there is no specific biomarker for PPCM, an elevated brain natriuretic peptide (BNP) can be used to point to cardiac chamber distension and fluid overload, as can an elevated troponin.1,3,15 In healthy pregnancies, BNP values are typically within normal limits or mildly elevated, so a significant elevation of BNP indicates cardiac chamber distension and fluid overload.3,16,17 In the immediately post-partum population, obtaining a D-dimer is unlikely to be helpful since a D-dimer is commonly elevated in normal pregnancy.5 If there is a high suspicion for a PE, a CTA chest should be performed. To differentiate PPCM from pre-eclampsia, additional lab work should be performed, such as a complete blood count (CBC) or liver function tests (LFTs) to screen for thrombocytopenia or transaminitis that can be found with severe pre-eclampsia.5 Lactate dehydrogenase (LDH), uric acid, and fibrinogen are also recommended to assess for preeclampsia in the peripartum and early postpartum periods.
In PPCM, a cardiac point of care ultrasound (POCUS) can show LV dilation and LV systolic dysfunction. Echo can also reveal right ventricular and biatrial enlargement, mitral and tricuspid valve regurgitation, pulmonary HTN, or an intracardiac thrombus, due to decreased forward movement of blood through the heart.1,3 Up to 6% of patients with PPCM experience a thromboembolic event.1,3 An echo is vital for the diagnosis of PPCM, while the other evaluation tools described above serve primarily to rule out other diagnoses on the differential.7
What do you do about it?
ABCs + IV/O2/monitor: In the event of severe respiratory distress, consider non-invasive positive pressure ventilation (NIPPV), which will assist in optimizing preload and afterload. If NIPPV fails, intubate. Since pregnant patients have lower functional residual capacity and a higher basal oxygen consumption rate, they are at higher risk for becoming hypoxic when apneic.5,11 Apneic oxygenation is advised during intubation, including during pregnancy.18 Bag-mask ventilation can aid in maintaining adequate oxygenation saturations during rapid sequence intubation.19
Call your friends: Cardiology and obstetrics should be consulted to assist in definitive management of new-found heart failure in an obstetrical patient and the establishment of outpatient follow up for future monitoring.5
Control volume status: Consider loop diuretics and nitrates. Loop diuretics help remove excess fluid from the body and thereby decrease preload.7,20 Consider afterload reduction to assist with poor LV function with vasodilators, like hydralazine or nitrates.1,5,7 Assess for hypotension and poor uterine perfusion as a result of over-diuresis in the pre-partum patient.1,3,5,6,20 Fetal monitoring can assist in evaluating uterine perfusion.10,21 If the patient becomes hypotensive or there is concern for low cardiac output, consider starting vasopressors or inodilators.5,20 Limited human data exists on the best agent to initiate during pregnancy. Dopamine and ephedrine increase maternal blood pressure and blood flow to the uterus. If possible, avoid phenylephrine and norepinephrine due to increased vasoconstriction of uterine arteries leading to placental insufficiency.5,11,20
Neutralize hormonal responses: ACE inhibitors or ARBs are a viable option for postpartum patients.1,5,20 They should not be given to pregnant patients due to potential teratogenicity.3 Some small trials have proposed using bromocriptine, which acts as a dopamine receptor agonist to inhibit prolactin.9
Avoid thromboembolic events: Since patients are in a hypercoagulable state during pregnancy and in the immediate peripartum period, consider anticoagulation for any patient with an EF < 30% during pregnancy and within 2-3 months following delivery.1,6 Heparin, including low molecular weight heparin, is compatible for use during pregnancy or while breastfeeding. Warfarin and direct-acting oral anticoagulants (DAOCs) should be avoided during pregnancy due to potential harm to a developing fetus.3,9
Prevent arrhythmias: Beta-blockers have shown to improve survival in dilated cardiomyopathies and are safe in pregnancy, but should not be given in acute decompensated heart failure.1,5 Beta-blockers are usually continued for at least six months after symptoms of PPCM resolve.7,20
What happens after the ED?
Further management generally depends on the severity of LV impairment.
In patients with moderately diminished LVEF, conservative management with exogenous oxygen, diuresis, ACE inhibitors, beta-blockers, and anti-coagulation may be sufficient to treat symptoms and limit life threatening complications.
In patients with significantly diminished LVEF with signs of cardiogenic shock, support should be initiated with inotropes and evaluation should be made for the possibility of a cardiac assist devices (e.g. intra-aortic balloon pumps, LV and biventricular assist devices, or extracorporeal membrane oxygenation (ECMO)).1 Arrhythmias, including ventricular tachycardia, are common sequelae of PPCM, occurring in about 4% of PPCM cases.22 Evaluation for a cardiac defibrillator may be indicated, especially at LVEF<35%.3 A wearable cardioverter/defibrillator may be indicated for 3-6 months, regardless of LVEF, due to potential for arrhythmias and sudden cardiac death.23 However, implantable cardiac defibrillators are usually not recommended due to the high recovery rate of PPCM, and wearable defibrillators are recommended in indicated.3,6,21
The likelihood of further PPCM complications were increased by having a LVEF<25%, delayed diagnosis, and subsequent thromboembolic disease.9 A majority (50-80%) of patients with PPCM will recover in 3-6 months of onset of the condition, but subsequent pregnancies have a high risk of disease recurrence, risk of incomplete return to baseline cardiac function, and increased maternal mortality.1,3 Maternal mortality rates have reached about 15% worldwide.9
So, what happened with our patient?
The patient was placed on 3-4 liters of supplemental oxygen via nasal cannula, which increased her oxygen saturation to the mid 90’s. She was admitted to the hospital with obstetrics and cardiology consults. An echocardiogram revealed an LV EF of 40% and dilated LV. Furosemide was given, and her volume status was monitored. She was started on an ACE inhibitor and beta-blocker. Due to her improving symptoms with medical management, further surgical interventions were not pursued.
Take Home Points
The highest risk for PPCM is in the month prior and the five months following delivery. Diagnosis includes heart failure within this timeframe with no other known underlying etiologies.
Consider this diagnosis in patients who are presenting with dyspnea on exertion or other signs of heart failure. Be wary of confusing these with common symptoms of late pregnancy.
Initial management should evaluate and support the patient’s respiratory status, with oxygen supplementation and consideration of non-invasive or invasive ventilation as dictated by the stability of the patient, as well as consideration of nitroglycerin (preload) and diuretics (systemic congestion). Patients in cardiogenic shock require resuscitation with vasopressors, inotropes, and consideration of ventricular assist devices.
In the still pregnant patient, consider early fetal monitoring in order to evaluate for uterine perfusion.
In the pregnant patient, avoidACE-Is, ARBs, warfarin, and DOACs for concern for teratogenicity. These are acceptable in the post-partum patient.
Arany Z and Elkayam U. Peripartum cardiomyopathy. Circulation American Heart Association Journal. 2016; 133(14): 1397-1409. PMID: 27045128. DOI: 10.1161/CIRCULATIONAHA.115.020491.
Honigberg MC and Givertz MM. Peripartum cardiomyopathy. BMJ. 2019; 364:k5287. PMID: 30700415. DOI: 10.1136/bmj.k5287.
Sliwa K, et al. Current state of knowledge on aetiology, diagnosis, management, and therapy of peripartum cardiomyopathy: a position statement from the Heart Failure Association of the European Society of Cardiology Working Group on peripartum cardiomyopathy. European Journal of Heart Failure. 2010; 12(8): 767-778. PMID: 20675664. DOI: 10.1093/eurjhf/hfq120.
Egan DJ, Bisanzo MC, and Hutson HR. Emergency department evaluation and management of peripartum cardiomyopathy. The Journal of Emergency Medicine. 2009; 36(2): 141-147. PMID: 17976813. DOI: 10.1016/j.jemermed.2007.04.006.
Arany Z. Understanding peripartum cardiomyopathy. Annual Review of Medicine. 2018; 69: 165-176. PMID: 28813232. DOI: 10.1146/annurev-med-041316-090545.
Sonmez EM, et al. What’s going wrong with this postpartum woman? American Journal of Emergency Medicine. 2018; 36(4): 737.e1-737.e4. PMID: 29395758. DOI: 10.1016/j.ajem.2018.01.058.
Elkayam U, et al. Pregnancy-associated cardiomyopathy: clinical characteristics and a comparison between early and late presentation. Circulation. 2005; 111(16): 2050-2055. PMID: 15851613. DOI: 10.1161/01.CIR.0000162478.36652.7E.
McGregor AJ, et al. The pregnant heart: cardiac emergencies during pregnancy. American Journal of Emergency Medicine. 2015; 33(4): 573-579. PMID: 25782736. DOI: 10.1016/j.ajem.2015.02.046.
Martin S and Arafeh J. Cardiac disease in pregnancy. American Association of Critical Care Nurses – Advanced Critical Care. 2018; 29 (3): 295-302. PMID: 30185496. DOI: 10.4037/aacnacc2018615.
Troiano NH. Physiologic and hemodynamic changes during pregnancy. American Association of Critical Care Nurses – Advanced Critical Care. 2018; 29(3): 273-283. PMID: 30185494. DOI: 10.4037/aacnacc2018911.
Ouzounian JG and Elkayam U. Physiologic changes during normal pregnancy and delivery. Cardiology Clin. 2012; 30(3): 317-329. PMID: 22813360. DOI: 10.1016/j.cci.2012.05.004.
McKenzie H and Pulley DD. The pregnant patient: assessment and perioperative management. Anesthesiology Clin. 2016; 34(1): 213-222. PMID: 26927749. DOI: 10.1016/j.anclin.2015.10.016.
Wang WW and Wang Yu. Peripartum women with dyspnea in the emergency department: is it peripartum cardiomyopathy? Medicine (Baltimore). 2018; 97(31): e11516. PMID: 30075517. DOI: 10.1097/MD.0000000000011516.
Balaceanu A. B-type natriuretic peptides in pregnant women with normal heart or cardiac disorders. Medical Hypotheses. 2018; 121: 149-151. PMID: 30396469. DOI: 10.1016/j.mehy.2018.09.034.
Elkayam U. Clinical characteristics of peripartum cardiomyopathy in the United States: diagnosis, prognosis, and management. Journal of the American College of Cardiology. 2011; 58(7): 659-670. PMID: 21816300. DOI: 10.1016/j.jacc.2011.03.047.
Pillai A, et al. Apnoeic oxygenation in pregnancy: a modelling investigation. Anaesthesia. 2016; 71(9): 1077-1080. PMID: 27440389. DOI: 10.1111/anae.13563.
Casey JD, et al. Bag-mask ventilation during tracheal intubation of critically ill adults. New England Journal of Medicine. 2019; 380(9): 811-821. PMID: 30779528. DOI: 10.1056/NEJMoa1812405.
Bhattacharyya B, et al. Peripartum cardiomyopathy: a review. Texas Heart Institute Journal. 2012; 39(1): 8-16. PMID: 22412221.
Witcher P. Promoting fetal stabilization during maternal hemodynamic instability or respiratory insufficiency. Critical Care Nursing Quarterly. 2006; 29(1): 70-76. PMID: 16456364.
Mallikethi-Reddy S, et al. Burden of arrhythmias in peripartum cardiomyopathy: analysis of 9841 hospitalizations. International Journal of Cardiology. 2017; 235: 114-117. PMID: 28268089. DOI: 10.1016/j.ijcard.2017.02.084.
Duncker D, et al. Risk for life-threatening arrhythmia in newly diagnosed peripartum cardiomyopathy with low ejection fraction: a German multi-centre analysis. Clinical Research in Cardiology. 2017; 106(8): 582-589. PMID: 28275862. DOI: 10.1007/s00392-017-1090-5.