Grand Rounds Recap 10.23.19

/

Morbidity and Mortality WITH Dr. Banning

Case 1: NIHSS and Thrombectomy

The NIHSS was designed to be a standardized and repeatable assessment of stroke patients utilized by large multi-center clinical trials to establish a baseline score and then be repeated at regular intervals or after clinical changes. These changes can be used to monitor the effectiveness of treatments and also to predict functional outcomes at three months after a stroke. Although the scale was designed as a research tool and not intended for clinical use at the bedside, the Joint Commission requires documentation of the NIHSS score within 12 hours of admission.

There are some well known limitations of the NIHSS scoring system. Most prominently, it favors left hemispheric cortical strokes over those in other vascular territories such as the right hemisphere, the posterior circulation, and the deeper structures. This is because of the emphasis placed on language processing and speech production, which is controlled through centers in the cortical areas most often within the left hemisphere. There is also a wide variance in inter-rater reliability of the system, which can lead to serious problems when profoundly disparate scores are calculated by different providers for the same patient at the same time.

NIHSS scores are also used to guide treatment, despite the original design of the instrument as a research tool. According to the 2018 AHA guidelines for early management of acute ischemic strokes there is a level I recommendation for thrombectomy in patients presenting within 6 hours of symptom onset and having a NIHSS score of 6 or greater. Reviewing the major trials that examined thrombectomy for ischemic stroke such as MR CLEAN and DAWN one finds significant heterogeneity in the NIHSS scores used in their criteria. The cutoff point of 6 or greater for the AHA recommendations thus boils down to expert consensus based on the evidence put forth by those trials. Interestingly, within those same 2018 guidelines, the AHA also offers a level IIb recommendation for thrombectomy to be considered in patients with an NIHSS of less than 6 based on the observation that a subset of those lower-scored patients declined and ended up either undergoing later thrombectomy or experiencing a decrease in their functional independence.

What do our own neurocritical care and stroke faculty have to say? Frequent reassessment is critical. The neurologic exam of a stroke patient waxes and wanes and this should make providers concerned that collateral circulation that had been perfusing vulnerable tissue is now failing. Have low threshold to engage your neurointerventional colleagues in these cases. Also there are some key portions of the neuro exam that are often overlooked, but carry huge clinical significance. These include assessment of visual fields, swallowing, and ambulation.

Another key learning point from this case is to perform your NIHSS score together with your nursing staff and anyone else who will be involved in reassessment of your patient. This circumvents the problem of poor inter-rater reliability and makes it far more likely that changes in a patient’s neuro exam are caught early.

Case 2: Ectopic Varices

Esophago-gastric varices are the most common and most studied cause of acute variceal bleeding and their management is well established. Ectopic varices are those that occur in any other part of the gastrointestinal tract including the duodenum, jejunum, ileal, colonic, anorectal, and even stomal sites. Of the various sites where they may occur, anorectal varices are the most common, accounting for 45% of the burden. Despite this, they very rarely lead to clinically significant bleeding.

Fortunately, the treatment of ectopic varices is much the same as for esophago-gastric varices, with a few exceptions. Give blood as indicated, octreotide and ceftriaxone, and in the case of anorectal varices you can even attempt to tamponade the bleeding with a Minnesota or Blakemore tube. Additional treatment modalities for ectopic varices include colonoscopy for sclerotherapy or ligation, TIPS, IR guided embolization, and surgical intervention.

In terms of differentiating upper from lower GI bleeding or further localizing the source of bleeding, what options are available? Traditional teaching has it that nasogastric tube insertion and gastric lavage can be used to separate upper from lower bleeding. Unfortunately this has not been borne out by the literature. A Cochrane review looking at the sensitivity and specificity of NG aspiration and lavage in ED patients with hematochezia or melena without hematemesis found huge variation in these test characteristics to the point that we can no longer consider this test as having clinical utility. On the positive side, other reviews have found that the risk of bleeding due to NG tube placement was overall low, so if there is another compelling reason for placement know that it is probably safe to do.

Urgent colonoscopy can be performed without colonic preparation, but the diagnostic information can be limited and it can also delay time to definitive treatment. Furthermore, complications inherent to colonoscopy occur with higher rates when the procedure is done urgently or emergently. With this in mind, it has been shown that there is actually no advantage to colonoscopy over radiologic studies, so the choice of which should be based on local expertise and available resources. There is also a CT protocol that incorporates a non-contrasted scan with both arterial and venous phased scans which is an option for acutely decompensating, potentially hemodynamically tenuous patients that may be amenable to IR procedures.

Case 3 (a series): Utility of Blood Cultures Drawn in the Emergency Department

Blood cultures are part of the core measures for management of sepsis, for diagnosis or endocarditis, and for guiding targeted antibiotic therapy in critically ill patients. With this in mind, when should we order them in the ED and furthermore is there a population of patients who we anticipate discharging but in whom we should still draw blood cultures prior to their leaving?

Blood cultures drawn in the emergency department have been shown to have a much higher rate of contamination than those drawn elsewhere within the health system. This is likely due to a combination of factors ranging from the fast pace, the variability in the type and experience of the operator actually drawing them, and the time pressures imposed by trying to meet core measures. False positive blood cultures have significant financial impacts on a patient level. These include increased pharmacy charges up to $12k, increased lab charges up to $11k, and increased hospital length of stay up to 22 days.

What factors predict true positive blood cultures? Traditional teaching holds that fever and leukocytosis predict bacteremia, however this is no supported by the most current data. Features that predict true positives include rigors, hypotension, need for vasopressors, neutrophil to lymphocyte ratio > 10, or the presence of SIRS criteria. This comes with the caveat that we acknowledge the more recent evolution of our conception of sepsis beyond SIRS, and that these data were not validated for immunocompromised patients or those suspected of having endocarditis.

As to the question of who should be getting cultures, there are actually fairly straightforward IDSA guidelines based on suspected source of infection. For community-acquired pneumonia there is NNT (or number needed to change management) of about 150, and cultures are no longer recommended for patients who are expected to be discharged or admitted to a med/surg floor level of care. Anyone with severe features (broadly, those intubated or in septic shock), or being treated empirically for MRSA or pseudomonas.

In the case of skin and soft tissue infections, patients who are neutropenic, have severe features, or who have unusual predisposing factors such as immersion injuries or animal bites should have cultures drawn. Otherwise blood cultures only change management in a vanishingly small 0.02% of cases and do not alter treatment in immunocompetent patients. Similarly, blood cultures are not recommended in patients being treated for pyelonephritis unless similar severe features are present because the antibiotic choice in general does not change.

To the original question, who should be getting blood cultures if they are anticipated to be discharged from the ED? Examining the IDSA recommendations it would seem that only those patients who are in septic shock or otherwise have some extenuating feature need them, but in fact there is an exception. Three sets of blood cultures are part of the standard diagnostic workup for suspected endocarditis, and certainly not every patient who is at risk for this disease requires admission. Despite a fairly broad practice variation among our group in terms of ordering cultures, there does appear to be consensus that rule out endocarditis is the one scenario where cultures can be drawn and the patient discharged if otherwise systemically well and will a means of return if the cultures are positive.

Case 4: Factors Associated with Delays in Antibiotics

This year our chiefs and residency leadership decided to take our morbidity and mortality conference back to a more traditional format of identifying errors with the intent of determining root causes and improving our practice patterns and patient care. By that same token there are eternally consistent themes, one of which is timing in antibiotics. This month rather than discussing the impacts and the importance of early antibiotic administration we decided to talk about why antibiotics are delayed as we develop tangible and longitudinal metrics of quality in this area.

Case 5: Inadequate Sedation and Rocuronium

Rocuronium is one of the most familiar drugs in emergency medicine and if there is a drug that an emergency physician could cite the dose in their sleep, roc would be a good candidate. By the same token we should also be able to articulate or sedation plan and verbalize It within the same breath as our RSI orders. When this second part of the equation is overlooked or delayed it can induce a waking nightmare of paralysis without sedation. We need to have a healthy appreciation for how important sedation is and what factors may lead to inadequate or delayed administration.

The doses of rocuronium that we use in the emergency department are in the rage of 1-1.2mg/kg, producing paralysis for 30-90 minutes. The drug is metabolized primarily by the liver with a small, but perhaps not insignificant component of renal clearance. Duration of action is known to be prolonged in elderly patients, in males longer than females, and up to 1.5 times as long in states of liver dysfunction.

Even the baseline longer duration of paralysis induced by rocuronium over succinylcholine has been shown to lead to longer time to post-intubation sedation of 27 minutes vs 15, which may be because a paralyzed patient can’t tell us that they are fully awake in any obvious way. We need to be observant and pick up on the subtle signs such as elevated heart rate or blood pressure, lacrimal tearing, preserved eyelash reflexes, or any indication that a patient is responding to verbal stimulus.

In the case that we need to reverse rocuronium such as for stroke or severe TBI patients who were intubated with the drug, we have the option to give sugammadex. This medication works by binding up rocuronium and rendering it unable to interact with the acetylcholine receptor. The dose of this medication is guided by response to train-of-four and it has a very low rate of any adverse events.

Case 6 (a series): Indications for ECMO in the Emergency Department

Extra-corporeal Membrane Oxygenation (ECMO) is a relatively newer technology that was first used in operating rooms in the 1950s. Use outside of that setting was rare, limited to pediatric populations and ARDS, and literature is somewhat limited to case reports until more recently. ECMO-capable centers have sprung up all over the world and now we are expanding the indications for the procedure to the point that streamlining operations to cannulate for ECMO in the ED is an area of active interest.

One such indication is for cardiac arrest. To date, there have been several studies that have shown impressive improvement in neurologically intact survival at one and three months after cardiac arrest compared to traditional manual CPR. This is theorized to be due to superior augmentation of cardiac index from 0.6 with traditional CPR to CI of 2 with ECMO, which functionally means better cerebral perfusion. In fact, as of 2015 the AHA incorporated recommendations to consider ECMO as an alternative to traditional CPR for select patients with a suspected reversible cause of their arrest.

What factors predict survival and better outcomes with ECMO? Simply stated, an initial shockable rhythm, shorter time to starting high-quality CPR, and shorter time to cannulation and starting on the circuit. This final point underscores the current interest in optimizing our operational procedures to make cannulation in the ED as fast and efficient as possible.

Inclusion and exclusion criteria are institutionally-dependent, and at UC we have a clearly defined set, but our CCAT team, who are the ECMO experts, has asked that we call them early and often when we think we have a patient who would benefit. In essence, we should consider ECMO and call the team for all young patients in cardiac arrest who have been down for a short amount of time. There are additional situations outside of cardiac arrest where patients may benefit from ECMO, but our institutional criteria for these are not as well defined, and so the ECMO team reiterates their desire for us to have a low threshold to engage them.

Closing Thoughts

September was an exceptionally high-acuity month in our department. We had a higher than usual number of traumatic arrests, cannulated several patients for ECMO, placed two emergent EVDs in the SRU, and activated massive transfusion protocol numerous times. We delivered our usual excellent standard of care to the sickest of the sick in the region, and did so with professionalism and confidence. This encapsulates what most of us thought of as the essence of emergency medicine when we started: the fast pace, the life-or-death decisions made quickly and with limited information, the heroic saves, and the tragic losses.

While it’s true this is part of what we do, as we become more experienced we begin to see that what happens in the SRU is a unique distillation of experience that we will never encounter again, and the true essence of our calling is found in the subtleties of the pods and minor care. We incorporated more families into code discussions. We went out of our way to call neighbors to look in on pets for patients who otherwise would have left AMA. We approached new cancer diagnoses with empathy and reached out to facilities to ensure that patients who needed to be admitted would not lose their spot in rehab. After a time we realize that the Medicine, complex as it may be, is not the hard part of our job. Rather it is the social, economic, and psychologic aspects that are as unique to each patient as anything else that present the greater challenges and where the art of what we do becomes most prominent.

In Cincinnati we train Emergency Physicians who truly are prepared for anything and we are proud to stand together.

QIKT - Acute, Decompensated Pulmonary Hypertension WITH Drs. Laurence and Wolochatiuk

Pulmonary hypertension is defined by mean pulmonary artery pressure greater than 25 by right-heart catheterization or greater than 35 by transthoracic echocardiography. Impeding heart failure is predicted by cardiac index less than 2.5 and right atrial pressures greater than 8, and this is the patient population that is the focus of this installment of the QIKT series.

The right ventricle is a relatively thin-walled structure that pumps blood into a low-pressure, high-capacitance system of the pulmonary vasculature. In states of chronic pulmonary hypertension the load against which the right ventricle works is increased and produces first hypertrophic changes followed by dilation as the muscle becomes unable to compensate. The dilated right ventricle compresses the left ventricle and decreases filling and ejection fraction. It also widens the tricuspid annulus, leading to regurgitation and further dilation, which in turn limits right ventricular perfusion and sets off the negative feedback loop of worsening ischemia and RV function.

Patients may present nonspecifically with slow progression of symptoms typically involving some constellation of dyspnea, chest pain, peripheral edema, or syncope. Key portions of the physical exam to note are the patients overall appearance, vitals, cardiovascular exam, and signs of congestion. Degrees of hypoxia need to be interpreted within the context of the patient’s baseline saturation that will hopefully be documented in outpatient notes. With respect to the cardiovascular exam, you should expect cooler extremities as a rule of thumb and if warm then consider other etiologies for their symptoms such as sepsis.

Diagnostics in the ED should include CBC, BMP, BNP or NT-proBNP, troponin, VBG and lactate for every patient. In addition to helping facilitate their inpatient treatment, BNP also serves as an independent predictor of mortality. Acute kidney injury is common in patients with pulmonary hypertension and is also associated with short term risk for death. Noting this on a BMP has important implications for patients who may also need contrasted CT scans to look for pulmonary emboli, or diuresis for volume overload. Additional testing such as inflammatory markers, thyroid studies, and cultures should be considered on a case-by-case basis.

Echocardiography also offers important information beyond measurement of PA pressure. TAPSE less than 1.8cm is associated with RV dysfunction, and studies have shown up to a 17% increase in risk for death for every 1mm less than that. By extension we can then use TAPSE as a surrogate marker for therapy, targeting greater than 2cm of excursion.

Chest xray may have some characteristic findings such as an enlarged hilar shadow, but CT offers much more information including severity of RV enlargement and hypertrophy. The ratio of either the PA diameter or area to that of the ascending aorta correlates with degree of elevated PA pressure and can also add valuable information for the inpatient team. Finally, you should consider CTPA in any case when there is suspicion for an acute PE.

Characteristic EKG changes include right axis deviation, right ventricular hypertrophy, incomplete right bundle branch block, large p waves in the inferior leads, or ST depression in the inferior leads. The most common dysrhythmias in pulmonary hypertension are supraventricular tachycardia, AV-nodal reentrant tachycardia, and atrial fibrillation or flutter.

Management of decompensated pulmonary hypertension should focus on treating identifiable triggers, managing tachydysrhythmias, correction of hypotension, augmenting RV function, and addressing hypoxia. As to identifiable triggers, keep sepsis in mind as patients are at risk due to indwelling lines or bowel edema and bacterial translocation. Transfuse for hemoglobin greater than 8 if necessary, and know that patients requiring continuous infusion medications on a home pump need those medications continued at all costs. Run them through a peripheral line if you are concerned for a pump malfunction.

Managing tachydysrhythmias should be done with care. Strongly consider electrical cardioversion if possible and avoid calcium channel blockers and beta blockers as they both have negative inotropic effects. If medications are required then amiodarone is a good choice and digoxin can be considered in select patients (usually those that are already on this medication per their pulmonologist).

Correcting hypotension must also be done with a conscientious approach due to very fragile physiology. Volume status is a delicately fine balance - too little and the patient is hypotensive; too much and you risk worsening their PA pressure. Ideally you should target the patient’s dry weight as documented in clinic with targeted with 250cc boluses of crystalloid. As for vasopressors, there is no ideal single drug, but norepinephrine is our recommended first line. Take care that norepi can increase pulmonary vascular resistance at higher doses. For second line and adjunctive augmentation vasopressin is an excellent choice that both increases peripheral vascular resistance with minimal effect on pulmonary vascular resistance.

Dobutamine is our recommended first line inotrope under doses of 10mcg/kg/min. Higher than this and you risk arrhythmogenesis, so milrinone is the recommended second line if needed. Have low threshold to start norepinephrine to counteract hypotension induced by milrinone.

The key principles of correcting hypoxia are to ensure adequate oxygenation, avoid hypercapnia, and minimize acidosis. Epoprostenol is an inhaled prostacyclin vasodilator that relaxes smooth muscles and improves V/Q mismatch. It has a quick onset and a short half life. Inhaled nitric oxide is another choice with similar properties with the additional benefit of some anti-inflammatory effects in sepsis complicated by ARDS. With nitric oxide be aware that it can induce methemoglobinemia and cause rebound vasoconstriction.

If forced to act and intubate, avoid lying the patient supine. Optimize with NIPPV and anticipate hypotension by having vasopressors ready at bedside. Also consider establishing central and arterial access before induction to facilitate intervention and monitoring if necessary. Etomidate is a good choice for induction as it has the least risk for cardiac or vascular side effects of all possible choices. The goal should be to minimize apneic time and so for all of the above reasons we believe that this patient population is the ideal candidate for an “awake” fiberoptic intubation. Ventilator management should then focus on low tidal volumes, low PEEP, and low plateau pressures.

Ultimately these patients will require admission, likely to an ICU setting. We also endorse engaging the critical care anesthesia team to consider ECMO.

Global Health Grand Rounds - Fever in the Returning Traveler WITH Dr. Wright

 Evaluation of fever in the returning traveler should start, as with any patient, with a detailed history. Important elements to consider include where have they gone, their itinerary, what airports did they pass through, and were they taking prophylactic medications as indicated. You’ll want to know if they received the recommended immunizations and what potential exposures they may have had including sexual encounters. Another critical piece of information in much of the world is any contact with water - not only drinking, but also swimming, bathing, wading, or even washing their clothes in local water supplies.

The differential diagnosis of fever in a returning traveler can be conceptualized in three broad types with some important specific diseases under each category. These are gastrointestinal, dermatologic, and nonspecific fever.

With predominantly gastrointestinal symptoms consider traveler’ diarrhea and cholera. Traveler’s diarrhea is usually self-limited but can follow a prolonged course. If due to some specific organisms such as Shigella, Salmonella, or Campylobacter (to name only a few), then the diarrhea may have blood or mucus mixed in. Cholera is associated with watery diarrhea, vomiting, and muscle cramps. Treatment is mostly supportive, but if your patient particularly unwell you may consider antibiotics and antimotility agents. For travelers to Africa or Latin America choose a fluoroquinolone, and for those coming from Southeast Asia use a macrolide.

Fever with dermatologic manifestations has a broad differential. Mealses is becoming more common due to falling vaccination rates globally. Dengue fever is widely distributed in tropical and subtropical zones and can be difficult to differentiate from Zika and Chikungunya, but once method is the so-called tourniquet test. To do this, apply a blood pressure cuff to the arm, inflate to between the systolic and diastolic blood pressure for five minutes and then remove. If a petechial rash has developed under the cuff then this is considered pathognomonic for Dengue. Schistosomiasis may also present with “swimmer’s itch” in the early stages after exposure. This is common in the Lake Victoria region of Africa and in later stages may present with an acute systemic illness called Katayama fever, or with liver failure and cirrhosis after prolonged, untreated infection. Typhoid fever is another disease that can present with rash and fever in the returning traveler that may also mimic appendicitis with right lower quadrant pain.

The third category, nonspecific fever, also has some important entities to consider. Malaria accounts for about one third of illness in returning travelers from Africa. You should also consider HIV and an acute retroviral syndrome. In certain parts of the world hemorrhagic fevers such as ebola, Marburg, and yellow fever need to be on the differential.

Every patient who you are working up should receive a CBC, BMP, LFTs, UA, and a chest xray. Eosinophilia is usually due to atopic diseases in the developed world, but in low-income countries it is more likely due to an invasive parasite. In some cases you will want a thick and thin smear to evaluate for malaria, quantiferon gold for tuberculosis, and serologic tests for other specific pathogens as guided by the detailed history. Stool studies can also be sent: ova and parasites, giardia antigen, fecal leukocytes and stool cultures.

In some cases patients may require admission both for further evaluation and management, but also for case control and public health reasons. At our hospital the laboratory personnel report cases to the public health department when indicated.