Acute Pulmonary Edema: Admit Everyone?

A 52-year-old man with diabetes, hypertension, end-stage renal disease on dialysis, and CAD presents to the ED with sudden onset of shortness of breath. He endorses missing his last dialysis session because he left the country for a week and was forced to pay for his last session out-of-pocket in Jamaica. Your history is otherwise limited due to extreme respiratory distress. He shakes his head ‘no’ when asked if he has any chest pain.


His vital signs are: BP: 194/96. HR: 84. RR: 26. O2 saturation: 85% on room air. Temp: 97.5 F. He has diffuse rales in all lung fields. A bedside ultrasound shows diffuse B-lines and poor cardiac contractility without pericardial effusion. You place him on BiPAP and intravenous nitroglycerin, while you call the renal service for emergent dialysis. You obtain an ECG, which shows tachycardia but no signs of acute ischemia. You obtain blood for testing.



Does this patient’s clinical presentation necessitate inpatient admission?

If you would consider discharging this patient after dialysis, which parameters would you use to reassess?
Are there any factors that you can use to predict this patient’s need for admission or candidacy for potential discharge after dialysis?


Post an answer/comment and receive points toward asynchronous learning credit (if you have have contributed to other “Cases, Controversy, and Conversation” posts, your points do accumulate).

10 points = 1 hour credit

1 point for relevant answer/comment

5 points for also citing a reference to a relevant online resource(s) or journal article(s)

The following two tabs change content below.


Kyle Kelson, Downstate/Kings County Emergency Medicine resident. @kelsonmd

Latest posts by kkelson (see all)

3 comments for “Acute Pulmonary Edema: Admit Everyone?

  1. January 11, 2017 at 3:52 pm

    This is a fascinating question without a clear answer. However, I would argue that the patient does not necessarily have to be admitted.

    To summarize the case we have an ESRD patient who missed dialysis presenting with respiratory distress secondary to acute pulmonary edema. He is stabilized with non-invasive positive pressure ventilation (NIPPV) and a preload-reducing vasoactive agent (nitroglycerin). The patient shows clinical improvement and is awaiting dialysis.
    To determine a plan, we need to consider the differential diagnosis of etiologies for his pulmonary edema AND ALSO what we need to do to address each one. Once each possibility is addressed, the patient can go home. The question is whether this can be done during an ER stay or whether he needs to be admitted for an adequate evaluation.

    Here is a differential:

    1. Volume overload caused by missed dialysis. He may have several liters of excess total body water which needs removal.
    –> He is about to receive dialysis which is the definitive therapy.

    2. Volume redistribution secondary to CHF exacerbation secondary to neurogenic factors. A sympathetic nervous system surge caused venoconstriction, pulmonary arterioconstriction, and systemic arterioconstriction thereby driving fluid through his alveolar membranes into the lung. There is no excess total body water; the fluid is just in the wrong place. Sympathetic tone driven pulmonary edema (neurogenic pulmonary edema) would have a self-propagating positive feedback loop here. As the patient becomes more dyspneic, the sympathetic tone increases, throwing fuel on the fire.
    –> The sympathetic feedback loop has already been disrupted with NIPPV and nitroglycerin.

    3. Cardiogenic pulmonary edema. Did he have a myocardial infarction? I don’t know. Maybe he did. He’s going to be staying in the ER for several hours no matter what to get dialyzed.
    –>Repeat cardiac biomarkers and EKG during dialysis. The patient is on de facto obs. If he rules out for MI and is already being treated for his CAD there’s no need for cardiology to see him. More history qualifying his CAD would be necessary since he is potentially too high risk to be ruled out with repeat EKG and troponin.

    4. Unlikely causes of the patient’s presentation. Acute valvular insufficiency, ARDS, severe anemia, sepsis, thyrotoxicosis, dysrhythmia, cardiac toxins, myocarditis, SLE flare, pulmonary embolus.
    –>These can be risk-stratified adequately during an ER stay. Based on the history & physical and ER studies, and that the patient was stabilized with NIPPV and nitroglycerin, these potential diagnoses can be stratified as very low risk.
    There is limited evidence addressing the specific question of discharge, but there is evidence we can use to guide our decision making. For example, a 2010 study looked at 25,291 patients with 41,699 episodes of emergent or urgent hemodialysis found that the primary diagnoses were heart failure (83%), fluid overload (11%), and pulmonary edema (6%).

    If at the end of the patient’s dialysis he is:

    1. Clinically at baseline
    2. Without evidence of heart failure / volume overload / pulmonary edema
    3. Without evidence of acute MI
    4. Without evidence of uncommon etiologies for his presentation
    5. With appropriate follow up care

    then I believe he may be safely discharged.

  2. kkelson
    January 14, 2017 at 12:31 pm

    All good points, Adam. We’ve all seen how bad these patients can look when they present, but also how quickly some of them can turn around if managed appropriately. In a survey sent out to our faculty, 23 people responded: 17% stated they would admit or place all of these patients on observation, and 83% stated they would reassess and potentially discharge the patient after dialysis. When asked which parameters would be useful in making a disposition, hemodynamics, continued respiratory distress, continued pulmonary edema, signs of ischemia or infection, and patient reliability were all common responses.

    There’s been no formal study on how to predict which patients will do well if sent home after emergent hemodialysis. The best data comes from a 1999 retrospective chart review of patients that presented to the ED for any chief complaint and who required dialysis. The authors reported that dyspnea was the most common presenting complaint of 69% of patients, and “heart failure” was the most common diagnosis in 65% (in agreement with the data cited above by Dr. Blumenberg). The authors reported that after dialysis, 68% of patients were discharged. There’s no mention of any predictors for patient discharge, but they do state that even among the subset of patients who were initially hemodynamically unstable, 61% were able to avoid admission. Authors did not report readmission rates, mortality, or any other outcomes, but it does suggest that discharge after dialysis is common.

    Predicting Disposition

    Again, there are no good studies here to help stratify patients on presentation. However, as pointed out by our own Dr. Martindale in a recent paper published in the American Journal of Emergency Medicine, lung ultrasound is a very useful tool for evaluating pulmonary edema and can detect improvement in patients post-dialysis. In fact, a study of patients actively undergoing dialysis found that b-lines resolved over the course of dialysis in real time. At least in the inpatient setting, a reduction in amount of b-lines corresponds with symptomatic improvement. A smaller pre-hospital study in patients treated with CPAP also found b-lines to correlate with improvement in hemodynamic parameters. Additionally, two studies in the inpatient setting found an increasing amount of B-lines to be independently associated with both mortality and hospital re-admission rates. Lastly, a prospective study compared bioimpedance data (used by nephrologists as a measure of both volume status and mortality prognosis) with lung ultrasound in dialysis patients. It found that lung ultrasound data not only correlated with bioimpedance, but that lung ultrasound was independently a better predictor of survival time. These studies were conducted over the course of days, rather than hours as would be more typical in the ED. These data all INDIRECTLY suggest that surveillance with lung sonography that demonstrates reduction in b-lines may be predictive of candidacy for discharge after receiving dialysis.

    Hospital course

    Due to staffing issues, the NTG infusion was tapered, paste was given, and the patient went urgently to the dialysis clinic. Upon return from dialysis, his BP remained significantly elevated and wheezing was observed. He was given prednisone and bronchodilation agents for possible COPD (he did not have a known diagnosis of COPD – it was presumed), and he was admitted for fluid overload and possible CHF. He was maintained on his dialysis schedule for management of his volume status. He improved after three days (two dialysis sessions) and was discharged on a steroid taper and continued dialysis as scheduled prior to his presentation.


    Sacchetti A, Stuccio N, Panebianco P, Torres M. ED hemodialysis for treatment of renal failure emergencies. Am J Emerg Med. 1999 May;17(3):305-7.

    Martindale JL. Resolution of sonographic B-lines as a measure of pulmonary decongestion in acute heart failure. Am J Emerg Med. 2016 Jun;34(6):1129-32.

    Noble VE, Murray AF, Capp R, Sylvia-Reardon MH, Steele DJ, Liteplo A. Ultrasound assessment for extravascular lung water in patients undergoing hemodialysis. Time course for resolution. Chest. 2009 Jun;135(6):1433-9.

    Cortellaro F, Ceriani E, Spinelli M, Campanella C, Bossi I, Coen D, Casazza G, Cogliati C. Lung ultrasound for monitoring cardiogenic pulmonary edema. Am J Emerg Med. 2016 Jun;34(6):1129-32.

    Strnad M, Prosen G, Borovnik Lesjak V. Bedside lung ultrasound for monitoring the effectiveness of prehospital treatment with continuous positive airway pressure in acute decompensated heart failure. Eur J Emerg Med. 2016 Feb;23(1):50-5

    Gargani L, Pang PS, Frassi F, Miglioranza MH, Dini FL, Landi P, Picano E. Persistent pulmonary congestion before discharge predicts rehospitalization in heart failure: a lung ultrasound study. Cardiovasc Ultrasound. 2015 Sep 4;13:40.

    Coiro S, Rossignol P, Ambrosio G, Carluccio E, Alunni G, Murrone A, Tritto I, Zannad F, Girerd N. Prognostic value of residual pulmonary congestion at discharge assessed by lung ultrasound imaging in heart failure. Eur J Heart Fail. 2015 Nov;17(11):1172-81.

    Siriopol D, Hogas S, Voroneanu L, Onofriescu M, Apetrii M, Oleniuc M, Moscalu M, Sascau R, Covic A. Predicting mortality in haemodialysis patients: a comparison between lung ultrasonography, bioimpedance data and echocardiography parameters. Nephrol Dial Transplant. 2013 Nov;28(11):2851-9.

  3. Ian deSouza
    January 15, 2017 at 3:10 pm

    Great comments. The term “heart failure” is really thrown around haphazardly. I think “pulmonary edema” is a more accurate (even if simplified) term here. One possible cause of pulmonary edema, fluid overload, would be addressed with hemodialysis. But without directly measuring LV, central arterial and peripheral arterial pressures, we cannot definitively diagnose an exacerbation of heart failure with preserved ejection fraction (HFpEF) an another/additional etiology. One can only use hemodynamics including BP, bedside US for persistent edema to SUSPECT it. (Advanced US techniques using flow may be used to evaluate valves and measure pressures as well.) An MI would be unlikely to result in an exacerbation of HFpEF (diastolic heart failure), as there is actually increased coronary perfusion (still, one could use serial troponins x 2 to rule this out anyway). Instead, a recent MI with significant infarction would be expected to present with heart failure with REDUCED ejection fraction (systolic heart failure), and this illness would present very differently than acute diastolic heart failure.

    I suspect that this patient returned from dialysis with his usual amount of fluid removed but his “vascular failure” (i.e. elevated afterload) insufficiently treated. He was given NTG paste prior to going to dialysis unit, but may have needed continued infusion with dialysis performed in the ED – there was a systems issue as well as a change of shift/transfer of care. This manifested as a “cardiac wheeze”; a repeat US would likely have demonstrated persistent b-lines. The vascular failure may have resolved had appropriate afterload reduction been resumed. Although admission MAY have been avoided, in the end, a few days of admission with an extra dialysis session and reinstatement of HTN medications eventually achieved this goal.

    It may seem crazy to some, but based on recent clarification of the pathophysiology of hypertensive heart failure, patients who do not have sepsis, MI, dysrhythmia, or other significant etiologies nicely summarized by Blumenberg should be considered for discharge if appropriately treated in the ED with adequate improvement.

    This is a great review of hypertensive heart failure:
    Viau DM, et al. Heart 2015;0:1–7. doi:10.1136/heartjnl-2015-307461

Leave a Reply

Your email address will not be published.