In the fall of 2019, Adam Bernheim, MD, assistant professor of diagnostic, molecular, and interventional radiology, and a team of researchers from Icahn School of Medicine at Mount Sinai in New York, traveled to China to meet with radiology colleagues in Chengdu and formed a research partnership. That meeting laid the groundwork that led Bernheim and his team to the leading edge of research into the previously undocumented illness that emerged from China shortly thereafter.
In the early stages of the COVID-19 outbreak in Wuhan and other Chinese cities, physicians used chest CT routinely to diagnose and track the disease, scanning some patients multiple times. In crisis mode, they lacked the time and the sub-specialty expertise to analyze the images systematically on their own, Bernheim says, and they partnered with his team.
“We were the first in the West to have access to large numbers of COVID-19 chest CTs — well before there were significant cases in Europe and the U.S.,” says Bernheim. The team systematically tabulated the findings and published papers in Radiology that described the characteristic patterns and correlated them with early, intermediate, and late stages of disease based on symptom time course.1,2
“We learned about the evolution and progression of disease and the time course of how coronavirus infection unfolds,” he says. He urged all radiologists to familiarize themselves with the characteristic imaging patterns — ground-glass opacity through the lungs often with a rounded morphology and peripheral and lower lung distribution — of COVID-19 cases on CT. “If you haven’t encountered it yet, you probably will,” he says, even on studies conducted for reasons other than diagnosis of COVID-19.
Preparing for Pandemic
“Awareness, preparation, and planning are key to responding to this crisis,” says Suzanne T. Chong, MD, MS, chair of the ACR Commission on General, Small, Emergency and/or Rural Practice, and associate professor, who recently joined the division of emergency radiology at Indiana University (IU). IU has three Level I trauma centers and 26 additional sites, ranging from tertiary care hospitals to community-based clinics. “The more prepared you are, the more lives you will save and the faster you’ll rebound,” she says. “We have been lucky in the Midwest; we have had more time to prepare than areas hit with early outbreaks.”
As she watched the pandemic hit China, Europe, and the U.S. coasts, Chong says this global pandemic mass casualty incident (MCI) is unlike most trauma MCIs in that healthcare professionals themselves are at risk for getting sick and dying and the timeline could extend for weeks and months.
According to Chong, opening all lines of communication is critical. “We tend to be siloed in medicine, but that doesn’t work with a situation like this,” she says. “Sharing information can save lives.”
Ella A. Kazerooni, MD, MS, FACR, chair of the ACR Lung Cancer Screening Registry® and the Lung-RADS® Committee, agrees. Her institution, the University of Michigan/Michigan Medicine, had recently established a strong system of clinical communication with a tiered huddle structure that feeds information up and down the chain quickly and helps create bonds within and among teams. Put in place in 2019, that enhanced communication has helped the institution prepare for the ongoing pandemic.
“The key thing the institution did — and radiology played a role on the diagnostic side — was to prepare to have capacity set aside very early on,” says Kazerooni. Drawing on plans developed to respond to earlier Ebola and H1N1 outbreaks, they quickly set up a respiratory ICU that was already half full at the end of March as the Detroit area witnessed a surge of cases.
Like most hospitals, Michigan Medicine also immediately worked to cancel non-urgent procedures, including imaging. Kazerooni says the radiology department used a multipronged approach to defer non-urgent procedures, beginning with deferral of screening exams, like mammography and lung cancer screening. Referring clinicians looked for any diagnostic testing that could be deferred. Radiologists listed all tests that could be safely postponed, and radiology staff communicated via their EHRs to referring providers about rescheduling patients into the future, when possible. Approaching deferrals as a partnership between radiology and referring clinicians, along with strong physician leadership, have been key, Kazerooni says.
Adjusting Policies and Procedures
In a crisis, Chong points out, it is not business as usual. In addition to canceling non-urgent appointments, other policies and procedures must be adapted to the current environment.
Most radiology departments are taking steps to reduce risk of exposure by minimizing the number of people in the reading rooms and moving to remote reading when possible. Those not already set up for remote reading have stepped up plans to equip radiologists with home equipment or moved staff to reading rooms in satellite clinics to reduce personnel at the hospital. Even those continuing to work onsite are reducing face-to-face consultations with colleagues or radiologists — opting to videoconference even if the person is in the next room.
At most facilities, staff reporting for shifts are checked for symptoms when they arrive and wear masks and personal protective equipment (PPE) throughout their shifts. Shortages of this safety equipment are a concern, especially at smaller facilities that may have more difficulties obtaining needed supplies.
Because asymptomatic patients can spread the virus, RTs are advised to assume everyone is infected —even someone coming in for an X-ray of a broken leg — and wear PPE, says Chong. “Our RTs are on the frontlines; they are the ones who risk exposure,” she says. Chong has been tracking how other facilities have changed policies and procedures, looking for ideas of how to maintain efficiency and ensure safety in the face of COVID-19.
Kazerooni notes that the University of Michigan Medicine worked with its infection control and environmental services departments to evaluate and reduce the cleaning time for their machines, while still adhering to CDC standards. In some cases they were able to reduce the cleaning time by half, depending on the airflow in the room and the time it takes to recirculate — significantly reducing downtime and increasing throughput, especially for CT scanners and radiography suites.
Adapting Approaches
ACR and other radiological associations have issued guidelines for using CT scans and reporting results in the diagnosis of COVID-19, recommending against routine use of CT and highlighting the importance of using standard language in reports. But sometimes things don’t play out as planned, notes Daniel Ortiz, MD, a general radiologist with Summit Radiology Services, PC, an independent private practice of 20 radiologists serving eight hospitals in rural northwest Georgia.
Ortiz found himself on the frontlines of an early outbreak in one of the communities served by Summit Radiology. After a large church event where people gathered from around the area, patients started showing up at the local ED in early March with a mixture of lower respiratory symptoms and atypical presentations. Several of these patients were imaged and had atypical pneumonia appearance. There was no history of travel to China or northern Italy, making it difficult to connect the dots to COVID-19. Even once physicians made the connection, patients did not meet the then-current CDC criteria for testing. And those who were tested often waited up to two weeks for PCR results.
In this situation, Ortiz says, CT was a viable and accessible means of assessing patients — even though a large portion of patients who have negative scans may still be infected. In the absence of PCR test results, the CT helped physicians stratify patients and adjust their level of suspicion — helping to build a story that pushed the treatment in that direction, Ortiz says. Once the outbreak was recognized, ED staff added an item to their screening questionnaire asking about attendance at large gatherings — and specifically the identified church — to speed up the identification of COVID-19 cases and increase pretest probability.
According to Bernheim, the positive swab reverse transcriptase polymerase chain reaction (RT-PCR) test is the cornerstone of diagnosis at Mount Sinai, with chest imaging serving as a complementary tool. “It’s not practical to scan large numbers,” he says. “Patients are scanned selectively when there are complications or there is suspicion for other processes such as pulmonary embolism in COVID-19 patients.”
Brent P. Little, MD, assistant professor of radiology at Massachusetts General Hospital (MGH), points out that even though PCR testing is the gold standard for diagnosis, the testing can take time and more than one test may be required to confirm the diagnosis. He notes that in mid-March chest radiography was still MGH’s first-line diagnostic tool for selected patients presenting at the hospital with respiratory symptoms, just as it was before the pandemic hit. Although he underscores that normal radiographs cannot exclude infection, findings suspicious for COVID-19 on chest radiography can elevate clinical suspicion and help guide clinical decisions while lab testing is underway. He notes that while COVID-19 can have a range of appearances at radiography, many of the cases have a bilateral, peripheral distribution. Radiographs can also provide valuable information about severity of lung findings, or suggest alternative diagnoses.3
Meanwhile, at Michigan Medicine — which has developed its own in-house rapid turnaround testing for COVID-19 — outpatients who suspect they may have COVID-19 contact a specialized nurse triage line. Depending on the severity of their symptoms, patients are either escalated to video triage or sent to a drive-up testing site set up as an extension of a clinic facility. People with milder symptoms are sent home to quarantine; those with more serious symptoms are sent to the ED. Chest CT scans are not a routine part of the diagnostic process, while chest X-rays may be used up front to differentiate COVID-19 infection from other causes of respiratory syndrome, Kazerooni says. In February, few U.S. radiologists had any experience with COVID-19. Now, many are finding it’s their main focus.4
“This is a fluid situation that is changing rapidly,” Bernheim says. “As radiologists, we have the responsibility to equip ourselves with knowledge and information. It’s critical in taking care of patients.”