A man in his late 50s presented to an ED a while back during the summers months with a 3 day H/O multiple complaints including rhinorhea, sore throat, dry cough, vomiting, loose stool, right frontal headaches, dyspnea, dysuria, and dark urine. So, a mixed picture. Fever of 100.9 was measured at home. No contacts or travel. Pretty healthy in the past: just hypertension, back problems and a 1/2 ppd smoker.
On exam, he looked sick. HR 115 T 37.8 RR 18 BP 105/63. Sat 90 on R/A. Decreased A/E on left base but otherwise negative exam. CXR showed a LLL pneumonia. Normal WBC. Lab abnormalities included Plt 104, Na 120, Cl 84, Lactate 2.9, AST 442, ALT 133, LDH 2027. Bili and ALP normal. AFib on ECG. Negative urine. Ceftriaxone and Zithromax IV were started. Dropped BP and became more tachy. POCUS scans for shock initiated (RUSH, ACES, UHP protocol…pick your acronym!). The IVC was flat. The cardiac scan was done mostly to look for an empty LV, but this was found instead…at the end of the video, an arrow will point out the main abnormality.
Pretty obvious pericardial effusion (PCE), eh? Interestingly, the elective echo was negative for effusion. So, what’s the diagnosis?
LLL pneumonia, hyponatremia, and elevated LFTs suggested a presumptive diagnosis of Legionella pneumonia. The IVC scan guided aggressive fluid resuscitation beyond 4 L. Norepinephrine was added. The patient was admitted to the ICU and ultimately improved. Urine antigen testing later confirmed the diagnosis.
So how does one explain the pericardial effusion? PCEs can occur in a minority of Legionella cases. There are a couple of case reports in the literature (1, 2). A PCE can be an extra data point to rule in this diagnosis. If the PCE accumulates quickly enough or is large enough, it can contribute to shock along with sepsis and hypovolemia.
Last point. The POCUS scan and elective echo showed discrepant results. The PCE is blindingly obvious in this case, so I think we know which scan was correct. That brings up a question. Many POCUS studies assessing the accuracy of scans performed by clinicians use elective ultrasound or echos as a gold standard or comparator. Is that a valid gold standard?
You learn early in medicine that you will never diagnose a disease that you don’t consider. We all suffer from tunnel vision, and this can result in misdiagnosis and ultimately significant morbidity for patients. Currently most of us are aware of the importance of a good travel history because of the recent Ebola virus outbreak that put everyone on heightened awareness for at least one continent hopping virus.
This is important because clinicians often assume that the acute onset of fever, sore throat, myalgias, headaches, and cough all point towards a diagnosis of influenza or an influenza-like illness (ILI). In many places in the United States, this would result in a prescription for Tamiflu and a discharge home regardless of the results of a flu swab, as it was in the case reported in this article. I’ll not discuss the evidence behind oseltamivir, but the fact of the matter is that patients want it and physicians give it. Nonetheless, the majority of the time the clinician would probably be right (or at least not horribly wrong).
This is all well and good, except that this patient didn’t have influenza. She actually had dengue, which she had picked up on a trip to Haiti that ended 7 days prior to presentation, consistent with the normal incubation period of 3-14 days.
Why does this matter? Because while dengue fever may not be necessarily life threatening, if it progresses to dengue hemorrhagic fever or dengue shock syndrome it certainly carries a worse prognosis. Typically this occurs in those previously infected, so a asking if there has been a prior diagnosis of dengue is also important. Then how do you distinguish between the two?
Clinically, one sign that can point in the right direction is that dengue can cause a truncal rash, which is uncommon in flu. Another vaguely nonspecific clinical finding is a bimodal or “saddleback” fever that persists for 3 days, resolves, and peaks again in 1-2 days. Typically most symptoms resolve in a week, barring severe complications.
Also, apart from dengue titers, most labs are nonspecific in distinguishing dengue from influenza. Both can cause elevations in CPK, transaminases, and the acute phase reactants (ESR and CRP). Both can also cause leukopenia and thrombocytopenia. One finding that is relatively specific is markedly elevated serum ferritin, which can also be elevated for Legionnaire’s disease, another ILI.
In the end, treatment of dengue is no different from that of influenza, mainly supportive care. People infected with dengue are more likely to result in hospital admission, so appropriate suspicion and testing may prevent the “bounceback” patient who ends up getting admitted. Proper diagnosis of dengue is also necessary for epidemiologic purposes. And while your febrile returning traveler might not have dengue either, you still need to make sure their ILI isn’t leptospirosis, MERS-CoV, Legionnaire’s, measles, malaria, meningococcal disease, or typhoid. Thus, keep an open mind, as those hoofbeats might be horses, zebras, or unicorns.
During Influenza Season: All Influenza-Like Illnesses Are Not Due to Influenza: Dengue Mimicking Influenza
Additional light reading available at the WHO page on dengue.