Spontaneous bacterial peritonitis (SBP) is an infection in the peritoneum. It can occur in patients OTHER THAN those with cirrhosis (nephrotic syndrome patients, for example), and has a high mortality. This is where we can actually make a difference - by making the diagnosis early. Every hour of delayed diagnosis (and then delayed treatment) raises mortality by 3%.
So if you are suspicious of it, TAP THAT BELLY. And you have to have a high degree of suspicion because up to 10% of patients with SBP will not have signs and symptoms of it!
Once you get that tap back, look at the poly number: greater than 250? Treat that! Go to that 3rd generation cephalosporin, ESPECIALLY if there has been no nosocomial or beta lactam exposure recently.
Next, after getting the elevated polys on tap, look at 3 things: the BUN, the creatinine, and the T Bili. If the BUN is >30, the Creatinine is >1, or the T Bili is >4, give 1.5g/kg albumin within 6 hours of diagnosis. Why? SBP carries an increased risk of renal failure and this maneuver decreases renal failure and decreases mortality.
Belly taps don’t need to be hard or intimidating; chlorhex the belly, grab your sterile gloves and sterile probe cover for your ultrasound, and use a long PIV to insert the catheter into the peritoneum and draw back fluid from that catheter. Sending more fluid in your culture bottles increases the ability to detect bacteria, so pull off at least 20cc just for the culture bottles. Risk of complications w/ a paracentesis is less than 1%- most commonly bleeding and infection. Look at the skin before going in, be sterile, and look with the ultrasound to make sure there are no blood vessels in your way.
Don’t let INR elevations or low platelets skew you away from doing a paracentesis. Patients with cirrhosis, specifically, have a new balance in their pro/anti coagulant balance, so regardless of an elevated INR, it is STILL SAFE TO TAP.
- Think about SBP and tap early.
- Take off enough fluid to obtain a good culture.
- Give albumin if BUN >30, Cr >1, or T Bili >4.
Sammy Boyd is a current third year resident at Stony Brook Emergency Medicine and Captain in the United States Air Force.
Trauma - we see some version of trauma every shift. We know trauma; we have an approach to it, a rhythm. But the rhythm changes when your trauma patient is pregnant. Imagine EMS calling this in:
“We have an approximately 30-year-old pregnant female - unclear how far along she is. She was the restrained driver in a high-speed motor vehicle collision. Heart rate 110, respirations 22, BP 108/58, O2 sat 99%. 5 minutes out.”
Nervous? Don’t worry. There’s some overlap here with the regular trauma evaluation, but pregnancy does involve some additional nuance. Remember, the best fetal resuscitation is maternal resuscitation.
What is the epidemiology of trauma in the pregnant patient?
Trauma is the most common cause of non-obstetric maternal death in the United States; it complicates 1 in 12 pregnancies. The most common mechanism in these patients is BLUNT TRAUMA, via:
Motor vehicle accidentsFalls (the combination of an altered center of gravity, lightheadedness from a baseline respiratory alkalosis, and laxity of the pelvic ligaments doesn’t bode well)Intimate partner violence (which accounts for 25% of these cases. As a side note, injury prevention is paramount in pregnancy. Always ask about intimate partner violence.)
Pregnancy masks (and predisposes your patient to) badness
Understanding trauma in pregnancy is largely a matter of understanding the anatomy and physiology of pregnancy. The pregnant patient experiences increased intravascular volume, cardiac output (by up to 40% in the third trimester), and heart rate (by up to 10-15 bpm in the third trimester). She also has decreased systemic vascular resistance; both the systolic and diastolic pressures may drop by 10 mm Hg. The increased plasma volume relative to red blood cell expansion creates a dilutional anemia. This patient can lose 30% of her volume before her vitals reflect it. Do not be reassured by normal-ish vital signs. In the second trimester and beyond, the waters become murkier as the uterus moves out of the pelvis. Both the uterus and the bladder become more vulnerable to blunt injury simply due to their positions. As the uterus enlarges, it causes aortocaval compression, decreasing venous return and, in turn, impairing cardiac output. It also pushes upward toward the thoracic cavity, decreasing functional residual capacity and making oxygenation more difficult.
With that in mind, let’s return to our case...
How do you prep?
Call for help. Call OB/GYN, NICU, anesthesia, trauma...rally the troops based on your hospital’s resources. You’re about to receive not one, but two patients, so you need teams for both of them.
The familiar mantra: IV (labs), O2, monitor…+/- Rhogam
Obtain access above the diaphragm - place an IO if need be. Your patient’s venous return is compromised, and you need your meds/fluids/blood to circulate.
Labs must include blood type, Rh status, and a full coagulation profile including fibrinogen. This patient’s fibrinogen is decreased at baseline; a normal fibrinogen may mean she is already moving into DIC.
In the Rh-negative patient, we must also consider Rhogam which can be given up to 72 hours after the trauma. Feto-maternal hemorrhage complicates up to 30% of pregnant trauma cases. In patients up to 12 weeks’ gestation, give 50 mcg (which covers 5 cc of bleeding); in those beyond 12 weeks or with over 30 cc of bleeding, give 300 mcg. Some argue that we should give 300 mcg empirically in these cases, where we may not have accurate information rapidly about gestational age and/or the amount of feto-maternal hemorrhage. What about the Kleihauer-Betke test, you ask? In real life, a negative test does not exclude the diagnosis. Exercise caution.
Expect a difficult airway. Increased circulating progesterone increases total body water, which creates an edematous airway with friable mucosa. Additionally, the combination of decreased lower esophageal sphincter tone, decreased gut motility, delayed gastric emptying, and a gravid uterus pushing upward creates the perfect scenario for emesis and aspiration.
How to approach this: pre-oxygenate well, but be cautious with positive-pressure ventilation; consider sitting your patient upright. When using the bag-valve-mask, use low volumes and go slowly to avoid insufflating the stomach. Have smaller backup endotracheal tubes and multiple suction setups ready.
This patient will desaturate faster than usual due to increased oxygen demand, decreased reserve secondary to the uterus pushing upward, and increased minute ventilation. A normal pCO2 in a pregnant patient is 30.
How to approach this: ventilate faster than usual to avoid acidosis, aiming for an O2 saturation of 94% (which is best for both mom and the fetus). If you need to place a chest tube, aim one or two rib spaces higher than you normally would to avoid intra-abdominal placement.
Understand the concept of aortocaval compression. Remember that your vital signs may be misleading.
How to approach this: tilt the patient onto her left side (15-30 degrees) or have someone manually displace the uterus. Your choice of method depends on the logistics of the resuscitation - for example, it’s easier to do compressions in the second scenario. Use crystalloid liberally and have a low threshold to transfuse blood products.
Disability and Exposure - these evaluations are similar to those in a non-pregnant patient.
The Secondary Survey: additional nuances to consider
When examining the uterus, estimate the gestational age using fundal height. Inspect it for tenderness to palpation, contour, contractions, and palpable fetal parts. Uterine rupture is a real concern. Remember, too, that placental abruption can be complete, partial, or concealed.
Consider a sterile speculum exam to inspect for perineal/vaginal lacerations, urethral injury, ruptured membranes, status of the cervical os (open or closed), and possible fetal presentation. But remember, this exam is contraindicated in the third-trimester patient with vaginal bleeding.
Assess the fetal heart rate (normal is 120-160 bpm). Hopefully the OB/GYN team is setting up external fetal monitoring by now.
Speaking of imaging...
In this patient, your FAST is less sensitive since 1) it’s harder to obtain adequate windows, and 2) the gravid uterus compresses the paracolic gutters, changing how fluid flows in the peritoneum.
How to approach this: don’t be afraid to get CTs. I repeat: if the mechanism is significant or you suspect serious traumatic injury, don’t be afraid to order CTs. Contrast is a category B drug, and gadolinium is a category C drug. With regard to radiation, the American College of Radiology has stated that doses of less than 50 micrograys are not associated with increased rates of fetal anomaly or loss. A pan-CT (i.e. CT head, cervical spine, chest, abdomen, and pelvis) uses 25.2 micrograys. The American College of Obstetrics and Gynecology recommends limiting exposure to less than 50 micrograys throughout pregnancy. So while this is a large chunk of the “allowable” radiation, and does put the fetus at an increased risk of childhood and lifelong cancer, consider the risk/benefit ratio for the patient in front of you.
The resuscitative hysterotomy (previously known as the perimortem C-section) is a last-ditch effort. Consider this in the pregnant patient at 24 weeks gestation or later (i.e. fundus at the umbilicus or higher) who had a witnessed loss of pulses. The first incision should be made as soon as possible after (but ideally within four minutes of) loss of pulses; the procedure should be completed within 5 minutes. Act quickly and decisively; this is not the time for dainty cuts. The best outcomes are associated with procedure completion within five minutes, but neonatal survival has been reported after delays as great as thirty minutes. Your goal: give mom the best chance at ROSC (and possibly also deliver a viable neonate).
In terms of supplies, you don’t need much - just a scalpel and OR scissors to begin. In the meantime, send someone to grab a few clamps and sterile towels. Use the scalpel to make a large vertical incision from xiphoid to pubis. Cut down to the peritoneal wall and then through the peritoneum. Identify and deliver the uterus. Cut through the lower uterine segment vertically as well, taking care to avoid the fetus. Use scissors to extend this incision upward to the fundus, placing your non-dominant hand underneath the scissors to guard the fetus. Deliver the baby, clamp and cut the cord, and hand it off. Lastly, pack the uterus and abdomen with sterile packing or towels and resume your resuscitation.
A word on resuscitation medications in pregnancy
Safe medications include: the tetanus booster, most antibiotics (with the exception of fluoroquinolones and gentamicin), rapid-sequence intubation medications, propofol, and heparin.
Medications to avoid include: NSAIDs (since they can compromise uterine blood flow), narcotics (unless delivery is imminent), and antiepileptics (except in critical cases such as status epilepticus).
The best fetal resuscitation is maternal resuscitation.Do not be reassured by normal-ish vital signs.Remember to displace the uterus.Expect and prepare for a difficult airway.Know the resources available to you.
Sonika Raj is a current second year resident at Stony Brook Emergency Medicine.
https://www.emrap.org/episode/trauma/traumainhttps://coreem.net/podcast/episode-34-0-trauma-in-pregnancy/https://www.emrap.org/episode/january2013/traumainhttps://www.emrap.org/episode/december2004/traumainImaging of Trauma in the Pregnant Patient. Raptis et al, Radiographics 34(3): 748, May-Jun 2014http://www.emdocs.net/resuscitation-of-the-pregnant-trauma-patient-pearls-pitfalls/
A 5 day old male, ex-36.6 weeker, presents to the emergency department with jaundice. The patient’s mother states that he has turned more and more yellow over the last couple days.
The patient has had good oral intake and wet diapers since birth.
The patient was born by vaginal delivery after mother was induced due to polyhydraminos and separation of the amniotic sac from the uterine wall. The patient had a 1 day stay in the NICU and was discharged home with a total bilirubin of 10.
In the ED, the patient was noted to have severe jaundice, lethargy, and sunsetting of the eyes concerning for kernicterus (Image 1). Two IV lines were placed, labs sent, fluid boluses started, phototherapy commenced, and a CT head obtained which was unremarkable. His total/direct bilirubin resulted at 38.5/1.1. The patient was admitted to PICU for further management.
In the PICU, the patient received antibiotics, phototherapy, and an exchange transfusion. He was noted to have unusual eye movements involving a persistent downgaze. Neurology was consulted and EEGs were performed which were normal.
As the bilirubin level decreased, the patient’s clinical condition improved. The patient was ultimately discharged and has been following up with primary care, neurology, and other outpatient providers for intermittent episodes of downgaze and possible neurological delay.
What is this patient’s etiology of jaundice and elevated bilirubin?
Let’s first discuss hyperbilirubinemia in the newborn.
Many newborns have a total bilirubin greater than 1 mg/dl, and as the bilirubin level increases, the patient will look more jaundiced. Severely high bilirubin (>25 mg/dl) can cross the blood-brain barrier, causing kernicterus.1 Kernicterus is damage to the brain caused by deposition of unconjugated bilirubin in the brain stem and basal ganglia. Symptoms of kernicterus include trouble feeding, hypotonia, lethargy, seizures, and oculomotor paresis of upward gaze (sunsetting eyes).2
The spleen and the liver are involved in heme breakdown, which produces bilirubin. Bilirubin clearance begins at the liver, where hepatocytes conjugate the bilirubin, making it more water-soluble. The bilirubin is then secreted in bile and excreted into the digestive tract, where digestive enzymes reduce the bilirubin, aiding in its excretion in the stool. Infants, however, have few gut bacteria, and have beta-glucuronidase in their intestines, which deconjugates the conjugated bilirubin, resulting in bilirubin reabsorption.1,3
Physiological neonatal jaundice results from the fact that: 1) Infants have more red blood cells with a shorter life span, resulting in increased bilirubin production; and 2) Infants lack some of the enzymes and gut bacteria necessary in the clearance and excretion of bilirubin.1,3-4
Breastfeeding jaundice occurs in the first week of life due to insufficient breast milk intake, resulting in insufficient bowel movements to excrete bilirubin. Breast milk jaundice, on the other hand, occurs in the 6th to 14th day of life, and it is hypothesized that the beta-glucuronidase in breast milk might lead to decreased conjugated bilirubin.1,3
Pathological causes of neonatal jaundice can be attributed to increased bilirubin production and decreased clearance. Causes of increased production include isoantibodies (i.e ABO or Rh incompatibility); structural defects (i.e hereditary spherocytosis); enzymatic defects (i.e G6PD deficiency); birth trauma (i.e cephalohematomas), and polycythemia. Causes of decreased clearance can be caused by biliary atresia, Gilbert syndrome and Crigler-Najjar syndrome.1,3-5
Phototherapy should be started based on the patient’s total bilirubin level, age, and risk factors (i.e G6PD, lethargy, sepsis). Phototherapy works by converting bilirubin into lumirubin, which is water-soluble. Exchange transfusion is used in patients with elevated bilirubin levels and signs of acute encephalopathy. This procedure involves taking small volumes of the patient’s blood and replacing it with donor red cells, until the patient’s blood volume has been replaced twice.1,3-6
Now, back to our patient:
Unlikely causes of our patient’s jaundice and elevated bilirubin are hemoglobinopathies, such as HbS, HbC, or beta/alpha thalassemia because the patient presented at 5 days of age and would still have fetal hemoglobin. The patient’s blood type was O+ and the mother’s blood type was A+, so ABO incompatibility is also an unlikely etiology. The patient’s physical exam showed no evidence of bruises or cephalohematoma and the labs did not reveal polycythemia. The bilirubin level is too high for breastfeeding jaundice and the patient is too young for breastmilk jaundice. There is no clinical or laboratory evidence of microangiopathies, such as DIC, TTP, or HUS. The majority of the patient’s hyperbilirubinemia is indirect, decreasing the likelihood of a hepatic etiology. Although Crigler-Najjar syndrome could present with indirect hyperbilirubinemia, it is an inherited disorder and this patient has no family history of Crigler-Najjar.
G6PD and pyruvate kinase deficiency assays were sent during the patient’s hospital stay. The patient was found to have G6PD deficiency. G6PD is an enzyme that plays a crucial role in protecting red blood cells from breaking down prematurely from oxidizing free-radicals. Thus, neonates with G6PD deficiency are at risk for hemolytic jaundice.7,8 If those assays had come back negative, RBC membrane and enzyme panels would be sent at 6 months of age.
Samita Heslin is a current first year resident at Stony Brook Emergency Medicine.
Wong, Ronald J., and Vinod K. Bhutani. "Pathogenesis and etiology of unconjugated hyperbilirubinemia in the newborn." UpToDate. Waltham, MA: UpToDate (2015).
We are extremely fortunate and proud, as a program and a department, to have attracted and matched such a bright, talented, and personable group of new residents. They will all surely fit right in with our SB EM family. Please warmly welcome them all -Kingsley, Daniel, Amanda, Elizabeth, Erin, Benjamin, Facundo, Jeremy, Robert, Kevin, Alison, Timothy, James, Jose, Matthew, and Matthew!
An 8 year old female is skating on a frozen pond in February. She falls through the ice and disappears under the surface of the water. Her friend runs up the street to get her father who arrives within 5 minutes to find her floating amongst the broken ice. He is able to pull her from the water. She is unconscious, not breathing, and is cold to the touch. Pulse is not readily palpable. The ambulance is just arriving...
Drowning, which has been universally defined as “the process of experiencing respiratory impairment from submersion/immersion in liquid” by the World Health Organization1, is widespread throughout the globe and is a significant burden of disease. Although accurate epidemiology is hindered by reporting issues, there are a few striking statistics that highlight the problem: in the United States, 4628 individuals died in 20162, with younger patients disproportionately suffering morbidity and mortality. That’s more than 12 per day!
Although the definition focuses on respiratory dysfunction, it is important to understand that after a drowning event, there are three distinct outcomes: no morbidity, morbidity, or mortality. The morbidity category encompasses a wide spectrum of disease with neurologic sequelae being the primary, and often devastating, manifestation.
The pathophysiology of drowning is intuitive3. Voluntary breath holding is limited by the patient’s panicked and inefficient efforts to stay above water. Eventually, small aspiration occurs when an involuntary inspiration takes place. Laryngospasm ensues, preventing additional breaths even if the patient returns to the surface temporarily. Systemic hypoxemia causes unconsciousness. Breath attempts cease but the already aspirated fluid leads to a ventilation-perfusion mismatch due to direct alveolar damage, surfactant washout, bronchospasm, atelectasis, and pulmonary edema. Hypoxemia continues and eventually leads to irreversible neuronal injury in the brain with additional injury to the cardiovascular and renal systems.
Underlying cardiovascular or neurologic disease (particularly epilepsy) should be considered in every drowning patient, but these rarely impact initial resuscitation. Depending on circumstances (falls, boating or diving accident), the real possibility of trauma must also be taken into account. The Wilderness Medicine Society notes that the prevalence of concurrent cervical spine injury is low and does not recommend routine immobilization unless there is a specific concern for neck trauma, such as focal neurologic deficit4. At no time should immobilization delay or hamper resuscitation.
Rescue, obviously, is the essential intervention in order to prevent or shorten cerebral hypoxia. However, the rescuer is at significant risk of harm or death in many circumstances. Attempts at rescue should only be made by those with the appropriate training and equipment. Lay persons can attempt to extend or throw rescue devices to those actively drowning but should avoid entering the water themselves. The details of proper rescue technique are usually beyond the purview of the emergency physician, but one point worth mentioning is that in-water resuscitation (rescue ventilations) should be considered only if the patient has a pulse and there is a significant delay to extrication.
Upon presentation to the ED, patients can be graded using the Szpilman Classification for Drowning to gauge the severity of illness and predict mortality5. This is based on physical exam only, where patients presenting without cough and having clear lungs score a Grade 1 classification predicting a 0% mortality rate. However, patients with hypotension, diffuse rales, and foam in the airway have a 20% mortality rate. Those who present in cardiac arrest have only 7% chance of survival.
Asymptomatic patients who are well-appearing and normoxic need no diagnostic testing. The Wilderness Medicine Society recommends discharge after 4-6 hours of emergency department observation. Even those initially symptomatic or requiring supplemental oxygen can be discharged if the vitals have normalized within this timeframe. Keep in mind that chest x-rays done in the emergency department are not predictive of clinical course6. Those with respiratory symptoms but otherwise stable do not need an extensive workup though may require admission for prolonged supplemental oxygen requirements with or without non-invasive positive pressure.
Those who are critically ill need aggressive interventions and diagnostics. Initial assessment should include an extended careful pulse check of about 30 seconds because starting compressions on a hypothermic heart that is still beating may initiate a lethal dysrhythmia. Ongoing pulse oximetry and core temperature measurement, in addition to standard cardiac monitoring, is essential. As the primary insult is always hypoxia, focus on reversing this factor is always the primary objective. Immediate supplemental oxygen and positive-pressure ventilation should be initiated as needed. Escalating to intubation in order to both reverse hypoxia and protect the airway from further aspiration may be appropriate. Hyperoxia should be avoided. Mechanical ventilation should follow ARDSNet protocol.
Patients are typically hypotensive. Although this can be multifactorial, immersion diuresis can result in profound hypovolemia by the time the patient is being resuscitated by EMS or in the ED. Crystalloids should be bolused empirically to correct this and support hemodynamics. Refractory hypotension may require vasopressor support.
Hypothermia is almost always a complication of drowning and needs to be reversed rapidly. There is a misconception that cold water immersion is protective. The more severe the accidental hypothermia is, the longer submersion time can be presumed and the worse cerebral hypoxia7. Low core temperature will prevent ROSC, depress cardiac function, and worsen systemic hypotension. Furthermore, standard ACLS medications and defibrillations may be rendered ineffective until normothermia is achieved8. There are many active external and internal techniques for reversing hypothermia: forced-air warming blankets, radiant heaters, warm water bath, warming mattress, warmed IVFs, warmed inhaled oxygen, warmed bladder irrigation, peritoneal lavage, thoracic lavage, hemodialysis, extracorporeal membrane oxygenation, and cardiopulmonary bypass9.
For those found in cardiac arrest, the above considerations should be intervened on simultaneously if ROSC is to be achieved. Patients are usually in PEA after a tachy-brady decompensation and will remain so until hypoxia, hypothermia, and hypotension are reversed. Although no studies have demonstrated benefit in drowning patients, therapeutic hypothermia should be started after ROSC as per standard ACLS protocols as harm has never been demonstrated.
Advance centers may have ECMO available as a resource to use in patients presenting after a drowning event. A recent retrospective review of 247 drowning patients who received extracorporeal life support before, during, or after cardiac arrest showed a survival benefit[x]. Not surprisingly, those who never had an initial arrest before the extracorporal life support was utilized had the best outcomes, suggesting that early intervention should be considered before acute decompensation.
Take Home Points:
Drowning encompasses a spectrum of outcomes from asymptomatic presentations to vegetative states to deathModifiers such as dry, wet, secondary, delayed, salt-water, fresh-water, active, passive are no longer used as they do not exist or meaningfully impact managementResuscitation should focus on reversing the three H’s: Hypoxia, Hypothermia, and Hypotension
Ryan Barnicle is a current second year resident at Stony Brook, former lifeguard, and EMT-B.
A new definition of drowning: towards documentation and prevention of a global public health problem. Beeck et al. Bulletin of the World Health Organization. November 2015.Number of deaths, death rates, and age-adjusted death rates for injury deaths, by mechanism and intent of death: United States, 2016. National Vital Statistics Reports, Vol. 67, No. 5, July 26, 2018Szpilman D, Bierens JJ, Handley AJ, et al. Drowning. N Engl J Med. 2012; 366(22):2102-2110.Practice Guidelines for the Prevention and Treatment of Drowning. Wilderness Medical Society. Wilderness and Environmental Medicine, 27, 236-251. 2016.Near-drowning and drowning classification: a proposal to stratify mortality based on the analysis of 1,831 cases. Szpilman et al. Chest. 1997.Gregorakos L, Markou N, Psalida V, et al. Near-drowning: clinical course of lung injury in adults. Lung. 2009; 187(2):93- 97.Association of water temperature and submersion duration and drowning outcome. Quan et al. Resuscitation 85 (2014) 790–794.Vanden Hoek, TL, Morrison LJ, American Heart Association, et al. 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science. Part 12: Cardiac arrest in special situations. Circulation. 2010; 122(18):S829-S861Drowning In The Adult Population: Emergency Department Resuscitation And Treatment. Schmidt A, Sempsrott J. EB Medicine – Emergency Medicine Practice 17, 5. May 2015.Extracorporeal life support for victims of drowning. Burke et al. Resuscitation 104 (2016) 19-23.
A 31-year-old male with no significant past medical history presents with a fever of 39C and a bifrontal throbbing headache with photosensitivity for 1 day. He has also had nausea and 3 episodes of non-bloody, non-bilious emesis. He denies recent trauma, rash, recent travel, and sick contacts. There is no altered mental status.
On physical exam, he has photophobia but no neurological deficits. He complains of a stiff neck but has negative Kernig and Brudzinski signs. He has normal heart, lung, and abdominal exams.
What’s on the top of your differential diagnoses?
It’s important to not delay meningitis workup and treatment because without treatment, the case fatality rate can be as high as 70% for bacterial meningitis1. Delays in treatment increase adverse events, including in-hospital mortality and neurological deficits at discharge2,3.
There are a few questions that seem to come up each time we think about the diagnosis and treatment of meningitis:
1. When should you get a CT head before performing the lumbar puncture?
The Infectious Diseases Society of America (IDSA) recommends that only the following patients should get a CT head before their LP4:
Immunocompromised state (i.e. HIV, immunosuppressive therapy, transplants)History of CNS disease (mass lesion, stroke, or focal infection)New-onset seizure (within 1 week of presentation)Papilledema*Abnormal Level of ConsciousnessFocal Neurological Deficit
*Using ultrasound to measure papilledema:
To evaluate for increased ICP on ultrasound, measure the optic nerve sheath diameter (ONSD) 3mm distal from the posterior globe (Image 1)5. ONSD > 5mm is indicative of increased ICP. Optic disc swelling is shown in Image 2 along with an enlarged ONSD6.
2. Should you give steroids?
A 2015 Cochrane Meta-Analysis6 provides guidance on this topic. This meta-analysis had 4121 individual patients for 25 randomized trials. 22 studies used dexamethasone and the remaining 3 studies used hydrocortisone or prednisone. The meta-analysis found no difference in overall mortality between patients who received dexamethasone and those who didn’t.
However, in the subgroup analysis, glucocorticoids reduced mortality in patients with meningitis caused by S. pneumoniae (NNT 18). Glucocorticoids were also associated with lower rates of hearing loss (NNT 21) and short-term neurological sequelae. Side effects of dexamethasone were minimum, with a small increase in recurrent fever (NNH 16) but no adverse outcomes7,8.
Take home points:
1. Don’t delay workup and treatment for meningitis.
2. Not everyone needs a CT Head before LP - refer to the IDSA guidelines.
3. Start antibiotics as early as possible, and before the CT Head/LP if they are going to take >1 hour.
4. Give glucocorticoids early - before or with antibiotics.
Samita Heslin is a current first year resident at Stony Brook Emergency Medicine.
“Chapter 2: Epidemiology of Meningitis Caused by Neisseria Meningitidis, Streptococcus Pneumoniae, and Haemophilus Influenza Format:Select One PDF [95K].” Centers for Disease Control and Prevention, Centers for Disease Control and Prevention, 15 Mar 2012, www.cdc.gov/meningitis/lab-manual/chpt02-epi.html.Steven, Peduzzi, & Quagliarello. "Community-acquired bacterial meningitis: risk stratification for adverse clinical outcome and effect of antibiotic timing." Ann Intern Med 129.11_Part_1 (1998): 862-869.Auburtin, Marc, et al. "Detrimental role of delayed antibiotic administration and penicillin-nonsusceptible strains in adult intensive care unit patients with pneumococcal meningitis: the PNEUMOREA prospective multicenter study." Critical care medicine 34.11 (2006): 2758-2765.Tunkel, Allan, et al. "Practice guidelines for the management of bacterial meningitis." Clin Infect Dis 39.9 (2004): 1267-1284.“Keeping an Eye on Intracranial Pressure: Measuring ICP Using Ocular Ultrasound.” UK EMIG QuickHit, 12 Dec. 2012, ukemigquickhit.wordpress.com/2012/10/15/keeping-an-eye-on-intracranial-pressure-measuring-icp-using-ocular-ultrasound/.“Papilledema and the Crescent Sign.” SinaiEM, 13 Dec. 2012, sinaiem.org/papilledema-and-the-crescent-sign/.Brouwer, Matthijs, et al. "Corticosteroids for bacterial meningitis." Cochrane Database of Systematic Reviews (2015). https://www.cochrane.org/CD004405/ARI_corticosteroids-bacterial-meningitisMount, Hillary, and Boyle. "Aseptic and Bacterial Meningitis: Evaluation, Treatment, and Prevention." Am Fam Physician 96.5 (2017).
In 2005, psychologists Fredrickson and Losada published on the critical positivity ratio with the idea that the ratio of positive to negative affect differentiates a person who flourishes from one who languishes1. Through complex mathematics they determined a critical ratio of 2.9: we therefore must have about three positive experiences or emotions to balance each one negative. Though the ratio itself has been challenged, it does seem true that our negative experiences hurt us more than the positive ones lift us up. I have personally found this to be very true when reflecting on the aortic catastrophes I have seen over this past year.
I received a prehospital notification of a male in his 70s with history of aortic root aneurysm who called for chest pain. En route he began to have numbness of his lower extremity and became altered. Upon arrival to the ED, I noted that patient was unresponsive and pulseless and began CPR. We activated our Code Aorta due to concern for dissection leading to arrest. The patient was noted to be in PEA and bedside ultrasound revealed a pericardial effusion. Vascular surgery deemed the patient not a surgical candidate so after attempts at pericardiocentesis without significant change in patient’s cardiac activity, the patient expired.
This was an emotional resuscitation: this patient who was a fairly healthy and active man had been able to call for help himself just minutes before and I had to stand by waiting for him to arrive as I received prehospital notifications which convinced me that he was actively dissecting in the ambulance. I found myself wishing he had come even just minutes earlier- that there was more we could have done. This one negative case weighed on me for weeks as I pondered what we could have done differently with my mind continuously coming up blank. It is only with reflecting upon a few aortic disasters with great outcomes that I can find peace in this loss.
I was walking down the hall of my zone in the ED when the nurse called me to evaluate a patient she was concerned may have just seized. The patient had yet to be seen and I learned that she was a female in her early 70s with history of nephrolithiasis and hypothyroidism who presented complaining of right lower quadrant pain. Her triage vitals had revealed a heart rate of 48 but otherwise she was afebrile in triage with a blood pressure of 139/76, alert and conversational. On my initial evaluation she was pale, diaphoretic, and intermittently responsive to my questions. Fingerstick was 127, repeat vitals essentially unchanged from triage. My attending requested we move the patient to our resuscitation room, at which point repeat vitals revealed a blood pressure of 52 over palp. As the team worked to start central and arterial lines, I began to ultrasound the patient. I was unable to visualize any recognizable structures in my FAST views of the right and left upper quadrants due to body habitus and what seemed like air or something obstructing, but when I placed the probe on the patient’s epigastrium, a 7.7 cm aorta with possible free fluid posteriorly revealed itself. CT which was performed in tandem with vascular surgery notification showed a 7.0 x 6.5 cm infrarenal AAA with rupture into the right sided retroperitoneum. The patient was taken emergently to the OR for endovascular aortic repair and is doing well to this day.
I was in another patient’s room when I heard screaming in the hallway followed by an announcement “new patient critical care room 3” overhead. I walked into the room to see a young appearing 50 year old man with history of hypertension complaining of severe pain of his abdomen. He had been sitting on the toilet attempting to have a bowel movement when he had sudden onset of severe abdominal pain. EMS had astutely recognized decreased pulses of the right lower extremity. The patient was clearly uncomfortable, with a distended abdomen and diaphoresis. An upright chest xray was performed for possible free intraabdominal air and when negative, he was immediately taken to CT which revealed a Type B dissection just distal to the left subclavian, extending to the right common iliac artery and almost completely occluding the SMA. After controlling blood pressure with esmolol, nicardipine, and nitroprusside drips, he was taken emergently to the OR for thoracic endovascular aortic repair with stenting of the iliac and is doing well.
Fredrickson and Losada would tell me I need one more win to get my positivity ratio up, so here it is. This one a bit of a miracle in my opinion…
A woman in her 70s presented to the ED with abdominal pain. As the nurse wheeled her by my desk and into a room I overheard she had recently been here for the same. On quick chart review it seemed that she had presented 5 days earlier and was found to have a 7.6 cm AAA with concern for rupture. This was an incidental finding on a right upper quadrant ultrasound performed for possible biliary colic due to her complaint of intermittent right flank pain, further evaluated subsequently by CT of the aorta. She had refused surgery due to concern for prolonged recovery phase and decided on comfort measures. On the morning of her second presentation, she stated that she was tired of having pain in her abdomen and was amenable to surgery. The patient was pale and we were unable to obtain a blood pressure. Code Aorta was activated, we obtained an arterial line with an initial blood pressure of 61/40 and simultaneously obtained a cordis through which we began transfusing blood. The patient went to the OR emergently and had successful endovascular aortic repair. How she survived for 5 days with her rupturing aorta I will never understand.
If your patient is sweating, you should be sweating.Let kidney stones and biliary colic be a diagnosis of exclusion.
Let’s flourish this year: for each loss we have at least three wins, it’s just a matter of reflecting on the good and not just on the bad.Ashley Mogul is a third year resident and the Academic Chief at Stony Brook.@JustAMogulReferences:
Fredrickson BL, Losada MF. Positive affect and the complex dynamics of human flourishing. Am Psychol 2005. 60(7): 678-86. PMID: 16221001
I recently had the privilege of caring for a 40ish year old man, primarily Spanish speaking, who was obviously uncomfortable and appeared to be in distress. By way of an interpreter and through clenched teeth, he told me that he had been experiencing severe abdominal pain over the last 3 days. He had recently traveled to Texas and been diagnosed with a “liver infection” for which he had been on antibiotics. He had experienced loose, brown stool for the last 2 weeks. He had been started on an antibiotic, the name of which he could not recall, the week prior. He did not otherwise have any medical problems or take medications by his report.
On exam he was alert, in apparent distress and occasionally clutching his abdomen. He had generalized abdominal tenderness, though his abdomen was flat and without peritoneal signs or bruising. His skin was warm and dry, pulses intact in all extremities. He was tachycardic without abnormal heart sounds and his lung were clear.
His vital signs were as follows: 36.3 orally, RR: 28, HR: 123, SpO2: 100%, unable to detect BP with automated cuff. BP was 60/palp manually.
Take a minute to consider what your differential and management might be.
We ordered norepinephrine to be infused peripherally. He was felt to be overall fluid-down, and was empirically given 3L of Lactated Ringers, broad spectrum antibiotics, and laboratories including a lactate, VBG, and blood cultures were drawn and sent. RUSH exam revealed a normal heart without pericardial effusion and hyperdynamic LV function, no intraperitoneal fluid, and a normal sized aorta. His IVC was flat and lung sliding was present bilaterally. An upright chest XR was negative for pneumothorax, nor free peritoneal air. Meanwhile, a CT of the abdomen and pelvis with contrast was ordered.
Armed with a story of recent travel with profound diarrhea and a nebulous “liver infection,” I was prepared to declare this septic shock from a GI source and move along after he was stabilized. One thing continued to strike me as odd, however: how could this man continue to mentate and eloquently express his discomfort even in the setting of profound hypotension, which I had concluded from overwhelming sepsis?
The answer came from another member of the clinical team. A learner in the department who was looking back through the patient’s records while I conducted the H&P came to me and asked if this patient’s known history of Addison’s Disease might be contributing to his presentation today.
A bit on adrenal insufficiency / adrenal crisis:
Categorized as primary (adrenal disease), secondary (most commonly caused by cessation of exogenous steroids, also by pituitary disease), and tertiary (hypothalamic disease)The most common inciting factor for adrenal crisis is fever or GI illness, though it can be brought about by any systemic insult (e.g., trauma, sepsis, surgery, GI bleeding, etc.)The classic feature of adrenal crisis is profound hypotension that can be refractory to volume and vasopressor resuscitation which is due to vasoplegia brought on by mineralocorticoid deficiency. Abdominal pain is also common, along with a list of other general and nonspecific symptoms.
There is no specific diagnostic laboratory study to confirm the diagnosis in the immediate setting; however, if it is suspected then a mineralocorticoid (e.g., hydrocortisone, dexamethasone) is the first line therapy in addition to fluid resuscitation. Early initiation of pressors is indicated in the face of profound hypotension. Glucocorticoids (e.g., methylprednisolone, prednisone) will not treat the hypotension, as they have no mineralocorticoid effects!
Classic findings for adrenal insufficiency are seen on the chem panel: hyperkalemia, hyponatremia, hypoglycemia, though these findings may vary depending on the etiology of the deficiency.
Feeling foolish, I asked the patient when the last time he last took his home steroids. He said that he had stopped taking them earlier in the week when he began taking the antibiotics. We administered hydrocortisone 100 mg IV and over the next 30 minutes the patient’s blood pressure became measurable and lingered around 100 systolic.
In this particular case, his potassium and glucose levels were normal; however, he was profoundly hyponatremic, the initial value at 108 mmol/L. Given that he was mentating normally and was neurologically intact, we simply fluid restricted the patient from receiving further crystalloids. The goal of should be to correct the hyponatremia by 6 mmol/L per day, unless there are severe features (e.g., seizure, focal neurological deficit, coma). In the latter example, increasing the sodium by 6 mmol/L in the first 6 hours and then stopping for the day will treat the severe features safely. The consequence of overshooting the correction is central pontine myelinolysis, a devastating disease.
This patient’s blood pressure continued to improve following hydrocortisone administration. His CT revealed a normal abdomen and pelvis. He had a brief ICU stay for monitoring of his blood pressure and electrolyte resolution. His sodium was gradually corrected over the subsequent 4 days and his blood pressure returned to normal ranges. He was seen by endocrine and was restarted on his home regimen of fludrocortisone, a PO mineralocorticoid. The importance of his life sustaining medication was reinforced and he was discharged to home 3 days after initial presentation.
In the critically ill patient always ask about recent use of systemic steroids.Review recent prior records when available.Resist early diagnostic closure and keep your differentials broad.When giving stress dose steroids, an agent with mineralocortiocoid activity (e.g., hydrocortisone, fludrocortisone) must be used, as it has effect on blood pressure via the RAAS system unlike glucocorticoids (e.g., prednisone, methylprednisolone).In hyponatremia, in general the treatment is: fluid restriction and gradual correction of the sodium.
Alexander Bracey is a third year resident and the QA/QI Chief at Stony Brook.
Brit Long. emdocs - Adrenal Crisis in the ED.. emDocs. Published on May 15th 2015. Accessed on November 28th 2018, Available at [http://www.emdocs.net/adrenal-crisis-in-the-ed]Tintinalli JE et al. Adrenal insufficiency and adrenal crisis. Tintinalli’s Emergency Medicine: A Comprehensive Study Guide. 2011; 7.Scott Weingart. EMCrit Podcast 39 – Hyponatremia. EMCrit Blog. Published on January 17, 2011. Accessed on November 28th 2018. Available at [http://emcrit.org/emcrit/hyponatremia/ ].
A 20 year old female with past medical history of poorly controlled type 1 diabetes, anorexia, and depression presents to the Emergency Department due to abdominal pain, nausea, and shortness of breath. She is noted
to have an elevated blood glucose of 512 and chemistry reveals bicarb <6, beta hydroxybuterate 9.90, and pH 6.88. You have diagnosed diabetic ketoacidosis (DKA) and plan to start treatment for it when you note a beta hCG of 1380.
DKA in pregnancy is considered an obstetrical emergency and is one of the leading causes of fetal loss. 0.5-10% of diabetic women will experience DKA during their pregnancy1,2. While the maternal mortality is low at <1% (though higher than in non-pregnant patients2), there are significant effects on the fetus including increased risk of preterm delivery and a perinatal mortality with reports ranging from 9-35%1,2,3. The hyperglycemia and acidosis of DKA both contribute to these poor outcomes. Maternal hyperglycemia leads to fetal hyperglycemia and due to osmotic forces this leads to a fetal diuresis and therefore fetal volume depletion3. Maternal acidemia combined with maternal diuresis results in decreased uterine blood flow and placental perfusion which therefore decreases oxygen delivery to the fetus1.
The pregnant state predisposes diabetics to worsening glycemic control and therefore DKA1. Pregnancy is a state of relative insulin deficiency due to lower maternal fasting glucose levels. The pregnant woman is in a lipolytic state with an increased baseline of free fatty acids in the blood. Common symptoms such as vomiting during pregnancy can contribute to the development of DKA4. Additionally, the physiologic tachypnea of pregnancy decreases the ability to buffer ketoacids via respiratory compensation and therefore makes the acidosis state less tolerable2,4.
The presentation of DKA is the same in this population as the general population, though euglycemic DKA is more frequent5. It has been reported that up to 36% of pregnant DKA patients have a blood glucose <200 mg/dL3. The treatment is also the same as in the general population with the addition of considering fetal heart rate monitoring and early consultation with a high risk obstetric team and an endocrinologist. Patients often require ICU level of care. Abnormalities in fetal heart rate tracing including repetitive decelerations and absence of variability will often occur but resolve as maternal acidosis improves1. DKA alone is not an indication for delivery as morbidity and mortality can be decreased via in utero resuscitation of the fetus by aggressive treatment of the mother’s DKA. The condition of the mother and fetus as well as the fetal gestational age must all be taken into account when considering early delivery1.
Ashley Mogul is a third year resident and the Academic Chief at Stony Brook.
Ecker, JL. Pregestational diabetes mellitus: Obstetrical issues and management. In: UpToDate, Barss V (Ed), UpToDate, Waltham, MA, 2018.Morton-Eggleston, EB & Seely, EW. Pregestational diabetes: Preconception counseling, evaluation, and management. In: UpToDate, Barss V (Ed), UpToDate, Waltham, MA, 2018.Clardy, PF & Reardon, CC. Critical illness during pregnancy and the postpartum period. In: UpToDate, Finlay G (Ed), UpToDate, Waltham, MA, 2018.Nyce, AL, Lubkin, CL, & Chansky, ME. “Diabetic ketoacidosis”. Tintinalli’s Emergency Medicine: A comprehensive study guide. Ed. Tintinalli, JE. McGraw-Hill Education. 2016.Hirsch, IB & Emmett M. Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Clinical features, evaluation, and diagnosis. In: UpToDate, Mulder JE (Ed), UpToDate, Waltham, MA, 2018.
Note: MOLST forms are specific to New York. Other states have slightly different forms,
though essentially have the same goals.
As part of the workup for our patients, especially those who are critically ill, it’s our responsibility as providers to ask about and respect their treatment goals. Sometimes, it’s a non-issue at the time, for example, a 60 year old being admitted for a moderate COPD exacerbation who is responding well to treatments but just isn’t quite ready to go home yet. But what if their COPD is compounded by multiple significant comorbidities, and they’re getting close to a trial of BiPAP?
Sometimes initiating the goals of care discussion can be very intimidating.
‘Are they going to think that we’ve given up on them?’
‘Are we jinxing them by asking about what they would want?’
‘What if I don’t know what their overall course will be?’
‘I don’t really know how to talk to them without making them upset…’
But let’s face it, the onus is on us to be our patient’s advocates. We need to know what they want and it’s important to take it a step further and document the conversation. Fortunately for us, the documentation part is relatively straight forward- just follow along with the MOLST form.
The MOLST Form (Medical Orders for Life Sustaining Treatment), is a form developed to make a patient’s wishes crystal clear. It’s valid both in hospital and out-of-hospital (which is super important….. in hospital DNR’s aren’t valid in the out-of-hospital environment, so EMS providers can’t honor them without contacting Medical Control for approval). It never expires, doesn’t need to be notarized, and classically is printed on fairly obnoxious pink paper to make it stand out in a chart or taped to someone’s refrigerator. The MOLST form can be adjusted over time- there’s two whole pages at the end that allow modifications to be made without having to fill out a new one. Best yet, it can be filled out in sections. For example, it’s possible that just the “Do not resuscitate” portion is filled out. You could also fill out the “Do not intubate” portion as well. For those patients who have expressed additional wishes, there are sections for what they prefer when it comes to feeding tubes, intravenous fluids, antibiotics, etc. There’s even a section that they can write in absolute specifics, for example they don’t want to go on dialysis.
Let’s go back to our COPD patient with lots of comorbidities, who is getting closer to BiPAP.
Maybe you want to start the conversation off with a quick recap of what has been done so far, and what the immediate next steps are. Perhaps then you transition by saying, “I’m not sure that we’re there yet, or that we’re ever going to get there, but I think we should talk about what you would want if you suddenly got really sick. Sometimes medical conditions change very quickly, and I would rather have this conversation when we have time and everyone is calm.” Maybe next you want to acknowledge the elephant in the room with, “These are really serious questions than can seem scary, but I want to make sure that if something changes, we respect your wishes.”
Depending upon what the clinical situation is, maybe now you transition into the DNR/DNI talk. My personal phrase tends to be, “If your heart were to stop beating, would you want us to attempt to resuscitate you, or would you want us to allow a natural death to occur?” Notice the word “attempt”. I think we do our patients a disservice when we present it as this idea that if you pick option A, we’re definitely going to get you back when this is not necessarily the case. Think about the medical dramas on TV… How often do the TV characters miraculously recover from a cardiac arrest? Way, way more often than reality. And for our patients and their families, that’s probably the only frame of reference that they have. Most of the general public has no idea what goes on when we code a patient…. The chest compressions, intubation, central lines….. whatever else. Part of these goals of care discussions involve us making sure our patients are making an informed decision, which likely isn’t in line with what they picture due to TV and movie portrayal.
After the DNR part, next often comes the DNI follow up. Again, depending on the situation, maybe we need to follow that part up with what the treatment plan is going to be from here. Perhaps you decide to present it as three paths. In Path A, we aggressively treat your medical condition, which likely involves procedures that could cause pain and limit your ability to communicate with your family. In Path B, we shift our focus to aggressively treating your symptoms and making you comfortable. In Path C, we continue what we’re doing now, but don’t escalate care. The conversation can go lots of ways but that’s a general framework you can use to figure out how to present huge decisions to your patients and their families. (It also goes along quite nicely with the Treatment Guidelines section of the MOLST form.)
Notice, the comfort care option isn’t phrased as “we either can do everything to try and save you, or do nothing.” In my opinion, that’s not really a fair thing to say to a patient. By not doing central lines, starting pressors, intubating, and whatever else…. It’s not that we’re giving up; we’re changing our focus to making sure whatever time the patient has left has the very best quality. We may also need to mention that we can’t predict exactly when death will occur if the comfort care option is selected- sometimes it’s minutes, days, or weeks away.
So now that we have a better idea of what the patient wants, I generally say, “I want to make sure that when I’m not your doctor anymore, I can still protect your wishes. To do that, there’s a form that we can fill out together, that makes all these things very clear in case something happens suddenly.” I also reiterate that the decisions aren’t set in stone, and can be adjusted if there is a change in clinical course.
In comes the MOLST form. To complete it, you need to be a licensed provider, so generally this involves an attending signing the form in two places. As a resident, you can act as one of the two witnesses- I generally try and have a family member act as at least one witness. A bit of an oddity to the form is that witnesses only write their names- no signatures are needed.
Once the form is completed and signed, I make a copy of it for the patient or their family, placing the original pink version in the chart. I also educate the patient and their family that when the patient leaves the hospital, the original pink copy always travels with them- it’s their safety blanket of sorts, and should be in a very obvious, well known location at home (I actually responded to a cardiac arrest as a paramedic and arrived to find a patient in bed with their MOLST form taped to the wall above their bed…. A job well done in my book).
The MOLST form, in all reality, should only take 5-10 minutes to complete, once the goals of care discussion has been done. Realistically, it often takes more time for someone to figure out their license number than to actually fill out the form. Most patients and their families really appreciate that their voices are being heard and advocated for, and your medicine colleagues will likely very much appreciate your attention to detail when you admit the patient. Most importantly, you’ve just given the patient a way to have their wishes protected not only during their hospital stay, but also if/when they go home.
Still have questions about goals of care discussions? I highly recommend a really fantastic article that recently came out in Annals of Emergency Medicine, which was written by a resident. Check it out at:
Wang DH. “Beyond Code Status: Palliative Care Begins in the Emergency Department.” Annals of Emergency Medicine 2017;69(4):437-443. PMID:28131488
Lauren Maloney is a third year resident and the Administrative Chief at Stony Brook.