Understanding the Role of CXR for Ambulatory Patients in the Era of COVID-19

emDOCs.net is proud to announce a new partnership with Journal of Urgent Care Medicine (@TheJUCM). Together, we hope to provide information on bread and butter emergency medicine and urgent care conditions. For the first in this series, Josh Russell (@UCPracticeTips) covers a study from Weinstock et al. on chest x-ray findings in COVID-19.

Take Home Message

The majority of ambulatory patients with COVID-19 have normal, or minimally abnormal, chest x-rays (CXR).  Use of virtual visits/telemedicine as a default, rather than in person, in clinic visits, is a reasonable strategy for patients without respiratory compromise.

Reference:  Weinstock MB, Echenique A, Russell JW, et al. Chest x-ray findings in 636 ambulatory patients with COVID-19 presenting to an urgent care center: a normal chest x-ray is no guarantee. J Urgent Care Med. April 13, 2020 . [Epub ahead of print]

Background

COVID-19 has forced clinicians in all specialties to rethink their practice and approach, specifically in regards to the management of patients with possible coronavirus infection.  In the Urgent Care (UC) and low acuity ED (ie: “Fast Track”) settings, these concerns are perhaps most salient because this is where patients with low acuity respiratory illnesses commonly present historically.  However, in the era of quarantine and social distancing, telemedicine has gained rapid traction in the initial triage of most complaints.  Many patients and organizations have found that preferentially encouraging such virtual video evaluations offers a safer and more convenient alternative for everyone involved.

Now that billing and patient privacy protection concerns have been mitigated through the H.R. 6074 act, the primary dilemma providers performing telehealth care face involves the inability to obtain objective data such as vital signs, labs, and imaging studies.¹  However, mounting evidence and expert opinion among EM and ICU clinicians suggests that a patient’s overall appearance and level of comfort seem to be the most meaningful variables in determining the need for intervention in patients with COVID-19, namely in the form of administration of supplemental oxygen and initiation of mechanical ventilation (as there are, to date, no proven pharmacotherapies that affect disease severity or progression).²  Thankfully, monitoring over telehealth innately provides this version of respiratory assessment and does not involve removing a patient from their home or put providers at risk of infection.

Even if a “from-the-door” style, observation-only physical exam seems to give us most of the necessary clinical data for triaging most patients with possible COVID-19, imaging of the chest must play some role, right?  Well, the CXR has long been the mainstay, and often only study ordered, when evaluating patients with respiratory complaints and possible COVID-19.³  However, what’s the value of a CXR in a non-toxic, well-appearing, ambulatory patient with suspected COVID-19? The JUCM study by Weinstock et al. may help to shed some light on this question.

The Study

This was the largest study to date of CXR findings in patients with COVID-19 and unlike many earlier studies of chest radiography, focused on plain films and ambulatory patients. The study examined 636 patients with COVID-19 infection, confirmed by nasopharyngeal (NP) swab PCR, at the time of CXR in a NYC UC center.  Approximately ~70% of the CXRs among these patients were initially read as normal.  All CXRs were re-read by a second radiologist for the purposes of the study (who was aware that the patient had confirmed COVID-19) and still 58% were read as normal and 89% of CXRs were read as either normal or only mildly abnormal.

Among the abnormalities identified, lower lobe involvement was most common and the findings were most often bilateral and interstitial or ground-glass in appearance. Importantly, lymphadenopathy and pleural effusions were very uncommon (see table).

Discussion:

Estimating the sensitivity of CXR for COVID-19 using the results of this study should be undertaken with caution because it assumes NP swab PCR to be the “gold standard,” although it has been shown to have many false negatives itself.  PCR, however, is highly specific for COVID-19 and, therefore, we can assume that nearly all 636 patients included actually had the disease.⁴  What is clear from this data is that, among ambulatory COVID-19 patients, significant CXR abnormalities are the exception and not the rule.

Unfortunately, it is important to note that the data on clinical outcomes of the patients in this study were unavailable and, therefore, it is unclear if certain CXR abnormalities may have prognostic implications. Based on current data, CXR abnormalities, even when present, will generally not affect immediate management in non-hypoxemic patients without respiratory distress, especially if COVID-19 has been confirmed by PCR.

Thankfully, if you’re using this approach, you have top cover.  The College of Urgent Care Medicine (CUCM) and Urgent Care Association (UCA) have collaborated with ACEP to produce joint guidelines for who to refer to the ED versus home management in an effort to support UC clinical decision making without diagnostic testing.  Based on current evidence, in generally healthy patients without hypoxemia, these organizations do not include CXR findings (or any other testing) in their risk stratification criteria for UC patients with suspected COVID-19. ^5,6

As we rush to understand this novel disease, our curiosity as clinicians is at an all time high.  The temptation exists to order labs, imaging, viral studies etc. because we are hungry for data to help us to better understand COVID-19. However, as always we must remember the axiom that studies are only helpful when their results will change management. During the COVID-19 pandemic, population health considerations and health care worker safety are more prominent concerns than ever before in the modern era.  For healthy patients who appear well and in no distress when evaluated via video telemedicine,  it does not appear chest radiography offers clinical data that would affect management sufficiently to warrant the exposure risk associated with recommending that likely infectious patients be evaluated in person.

References:

1. HR 6074. https://congress.gov/bill/116th-congress/house-bill/6074/
2. EM:RAP, COVID-19 Update, 4/7/2020; https://www.youtube.com/watch?v=klnVOjE1O-I
3. Wong HYF, Lam HYS, Fong AH-T et al. Frequency and distribution of chest radiographic findings in COVID-19 positive patients. Radiology. March 27, 2020. [Epub ahead of print]
4. Fang Y et al.  Sensitivity of Chest CT for COVID-19: Comparison to RT-PCR. Radiology. Feb, 19 2020 [Epub ahead of print]
5. ACEP COVID-19 Field Guide – CXR and CT Imaging: Chest X-Ray and CT
6. CUCM/ACEP Joint Statement on COVID-19 Risk Stratification in UC: CUCM/UCA/ACEP Risk Stratification