Diagnostics specialists’ take on the accuracy of CT scans

The use of computed tomography in outpatient computed tomography centers (OCTC) is entirely consistent with the clinical data and global best practices. This is also an effective tool for diagnosing the coronavirus infection COVID-19

In April the Moscow clinics had seen a surge in the number of patients suspected for COVID-19, including those in critical condition. A need had emerged to employ mass screening techniques to facilitate early diagnosis of the COVID-19 infection.

To accomplish this, the Moscow outpatient clinics were converted into 48 OCTCs. That way the clinics joined the fight against the coronavirus so that treatment could be initiated at the earliest possible time. To reduce the hospital burden the patients with mild symptoms were urged to maintain self-isolation at home.

General practitioners referred patients to a series of clinical examinations, such as PCR (polymerase chain reaction), blood tests to evaluate signs of inflammation, pulse oximetry to assess blood oxygen saturation, and ECG (electrocardiography) The symptomatic patients were referred for CT examinations. Early diagnosis contributed to the rapid isolation of asymptomatic carriers and better tailoring of treatment regimens for critical patients.

In total, more than 175,000 CT scans have been performed in the CT centers as of today. However, this was never about sending every suspected COVID-19 patient to a CT examination.

What do we know about the COVID-19 diagnosis?

In order to diagnose clinical symptoms accurately, first, the clinicians need to have access to reliable and objective data. Throughout the world, this part is called the learning curve. “Every country reacted differently on the emergence of the new coronavirus infection. The UK relied on general practitioners, South Korea deployed mobile laboratories on parking lots, whereas China put up stadium hospitals to accommodate asymptomatic patients. Doctors were not aware of the etiology, pathogenesis, clinical symptoms, diagnosis, or treatment of the new coronavirus infection,” said Sergey Morozov, the Chief Specialist in Radiation and Instrumental Diagnostics of Moscow and the Central Federal District of Russia, and the Director of the Moscow Center for Diagnostics & Telemedicine.

The World Health Organization (WHO) built their diagnostic guidelines based on the two ICD-10 emergency medical codes. The first code U07.1 necessitated confirmation of the coronavirus disease with laboratory tests. Under the second code U07.2, the diagnosis was supposed to be established according to the clinical symptoms.

Why do I need a CT scan?

“All patients with the coronavirus symptoms and positive PCR tests have changes in their lungs on CT scans,” Sergey Morozov notes. The UpToDate international recommendations classify patients into only two groups based on the degree of lung involvement, which is either less or more than 50%. However, to monitor changes in the lungs, clinicians use a different classification that splits patients into four groups: 25%; 25-50%; 50-75%; 75-100% of lung involvement. The area of damage can be determined by radiography and CT scans.

99% of the patients with severe (CT 3 50-75%) and critical (CT 4 75-100%) lung damage are treated in the hospital setting. Patients with mild-to-moderate lung damage (CT 2 25-50%) are referred to a hospital only if they have hypoxemia.

The Moscow Center of Diagnostics & Telemedicine with the help of the leading experts in radiation diagnostics, anesthesiology, intensive care, and infectious diseases, has published guidelines for COVID-19 radiation diagnostics. The document includes a detailed description of the semiotics of viral lung damage, differential diagnosis, and a protocol for interpretation of chest CT scans. The guidelines were developed to help radiation specialists and are aimed at improving the safety and efficiency of the diagnostic units.

How accurate is the diagnostic procedure?

The domain of instrumental, laboratory and radiation diagnostics utilizes the concept of diagnostic accuracy for revealing signs typical for particular conditions. This is also true for COVID-19.

The accuracy of CT methods for the diagnosis of the coronavirus infection amounts to 97-98% and is based on three indicators: sensitivity, specificity, and the area under the receiver operating characteristic curve (the ROC curve) International studies offer broad evidence for the high sensitivity of computed tomography: 98% (Radiology, Fang et al.), 97.2% (European Journal of Radiology, Long, et al.), 97% (Radiology, Ai et al., Caruso et al.), 67-97 % (Radiology, Bai et al.), 80% (Invest Radiol, Li et al.) Computed tomography is highly specific: 93-100% (Radiology, Bai et al.), 82.8% (Invest Radiol, Li et al.) The area under the H and D curve is 0.77-0.92 (Japanese Journal of Radiology, Himoto, et al.)

According to the publications from China, Italy, and Japan, 91-100% of the patients with positive PCR results have typical CT findings such as ground-glass opacities, 93% of the patients have multilobular injuries, and posterior lesions, whereas bilateral involvement was registered in 80-91% of the patients. This is a typical clinical pattern of the coronavirus pneumonia (Radiology, Caruso D et al., Bai HX et al.; Japanese Journal of Radiology, Himoto Y, et al.; European Radiology, Ye Z et al., Canadian Association of Radiologists Journal, Dai, et al.)

CT study is recommended for patients with typical clinical and epidemiological symptoms, even if they present negative PCR test results (Radiology: Fang Y et al., Ai T et al.) 60-93% the CT findings in COVID-19 appear before PCR assay turn out positive (Radiology: Ai T et al.) CT imaging is capable of showing typical abnormalities in 83% in the asymptomatic (subclinical) disease phase (Inui S et al.)

Treatment of COVID-19

It is worth mentioning that regardless of the condition, the diagnosis is never established based on imaging, instrumental, or laboratory findings. In other words, CT is a diagnostic tool rather than a treatment method. Health organizations across the entire world take full advantage of the at-home and outpatient treatment of COVID patients. The US system relied heavily on home care with remote patient monitoring: (Reviews in cardiovascular medicine, McCullough et al.), China (Investigative Radiology, Li et al.), Italy (Giornale Italiano di Cardiologia, Tarantini et al.; Head Neck, Vaira et al.), the United Kingdom (Clin Radiol, Nair A.), Canada (CMAJ, Glauser), India (Indian Journal of Pediatrics, Singhal T.), South Africa (South African Family Practice, Mash) On the other hand, South Korea concentrated on the establishment of dedicated outpatient centers (Journal of Korean Medical Science, Park et al.)

###