COVID-19

For a quick reference guide, please see our COVID-19 summary article.

COVID-19 (coronavirus disease 2019) is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a strain of coronavirus. The first cases were seen in Wuhan, China, in December 2019 before spreading globally, with more than 4 million deaths and 185190 million cases now confirmed. The current outbreak was officially recognized as a pandemic by the World Health Organisation (WHO) on 11 March 2020.

Definitive diagnosis of COVID-19 requires a positive RT-PCR test. Current best practice advises that CT chest is not used to diagnose COVID-19, but may be helpful in assessing for complications. The non-specific imaging findings are most commonly of atypical or organizing pneumonia, typically with a bilateral, peripheral, and basal predominant distribution.

No specific treatment exists although dexamethasone, a steroid agent, has been shown to markedly improve outcomes in ventilated or oxygen-dependent patients.

Several vaccines have now received regulatory approval in multiple territories.

Terminology

The official name of the illness is COVID-19 (a shortening of COronaVIrus Disease-2019) ¹⁵ and it is caused by the "severe acute respiratory syndrome coronavirus 2" (SARS-CoV-2) ^16,22,46. The names of both the disease and the virus should be fully capitalized, except for the letter 'o' in the viral name, which is in lowercase ^16,22,41. 

The official virus name is similar to SARS-CoV-1, the virus strain that caused epidemic severe acute respiratory syndrome (SARS) in 2002-2004, potentially causing confusion ³⁸. The World Health Organisation (WHO) has stated it will use "COVID-19 virus" or the "virus that causes COVID-19" instead of its official name, SARS-CoV-2 when communicating with the public ⁴⁵.

Epidemiology

As of July 2021, the number of cases of confirmed COVID-19 globally is over 185190 million, having affected virtually every territory, other than a few isolated South Atlantic and Pacific island states ¹⁵⁶, according to an online virus tracker created by the medical journal, The Lancet, and hosted by Johns Hopkins University ⁵. 

The R₀ (basic reproduction number) of the original wild type SARS-CoV-2 has been estimated between2.2 and 3.28 in a non-lockdown population, that is each infected individual, on average, causes between 2-3 new infections ^12,33. Several variants have emerged, and early research indicates some variants are more transmissible ^{211, 212}.

The incubation period for COVID-19 was initially calculated to be about five days, which was based on 10 patients only ¹². An American group performed an epidemiological analysis of 181 cases, for which days of exposure and symptom onset could be estimated accurately. They calculated a median incubation period of 5.1 days, that 97.5% became symptomatic within 11.5 days (CI 8.2 to 15.6 days) of being infected, and that extending the cohort to the 99^th percentile results in almost all cases developing symptoms in 14 days after exposure to SARS-CoV-2 ⁹².

As of July 2021, the number of deaths from COVID-19 exceeds 4 million globally ⁵. The case fatality rate is ~2-3% ^5,93. It is speculated that the true case fatality rate is lower than this because many mild/asymptomatic cases are not being tested, which thus skews the apparent death rate upwards ⁹³.

A paper published by the Chinese Center for Disease Control and Prevention (CCDC) analysed all 44,672 cases diagnosed up to 11 February 2020. Of these, ~1% were asymptomatic, and ~80% were classed as "mild" ²⁵. 

Another study looked at clinical characteristics in COVID-19 positively tested close contacts of COVID-19 patients ⁸¹. Approximately 30% of those COVID-19 positive close contacts never developed any symptoms or changes on chest CT scans. The remainder showed changes in CT, but ~20% reportedly developed symptoms during their hospital course, none of them developed severe disease ⁸¹. This suggests that a high percentage of COVID-19 carriers are asymptomatic.

In the Chinese population, 55-60% of COVID-19 patients were male; the median age has been reported between 47 and 59 years ^12,93.

As of July 2021, more than three billion vaccine doses have now been administered globally ⁵.

Paediatric

Children seem to be relatively unaffected by this virus, or indeed other closely-related coronaviruses ^31,47,90 with large cohort studies reporting that 1-2% of COVID-19 patients are children ^59,90,91. However, there have been cases of critically-ill children with infants under 12 months likely to be more seriously affected ⁵⁹. A very low number of pediatric deaths has been reported ^90,91. In children, male gender does not seem to be a risk factor ⁵⁹. The incubation period has been reported to be shorter than in adults, at about two days ⁹⁰. In a small proportion of the patients however a severe, delayed complication termed multisystem inflammatory syndrome in children (MIS-C) can develop, which is characterized by systemic shock with multiorgan involvement. 

NB: Surveillance methods and capacity vary dramatically between countries. Epidemiological parameters are likely to change as larger cohorts of COVID-19 positive cases are studied and new variants emerge, although this will only to some extent reflect a true change in the underlying reality of disease activity (as a disease is studied and understood people will be simultaneously changing their behaviours to alter transmission or prevalence patterns). 

Clinical presentation

COVID-19 typically presents with systemic and/or respiratory manifestations ^93,170. Some also experience mild gastrointestinal or cardiovascular symptoms, although these are less common ^18,50. Others may present solely with a gastroenteritis-like illness, which may not initially be recognised to be COVID-19 ¹⁷¹.

A significant minority of individuals infected with SARS-CoV-2 remain asymptomatic throughout the course of their illness acting as carriers ^70,113,164. 

The full spectrum of clinical manifestations of COVID-19 is broad ^1,13. Symptoms and signs are non-specific ⁶⁸:

Common

• fever (85-90%)
• cough (65-70%)
• anosmia (and other taste and/or smell disturbances) (40-50%) 
• fatigue (35-40%)
• sputum production (30-35%)
• shortness of breath (15-20%)

Less common

• myalgia/arthralgia (10-15%)
• headaches (10-36%) ¹²¹
• sore throat (10-15%)
• chills (10-12%)
• pleuritic pain
• diarrhoea (3-34%) ¹⁷¹
• splenomegaly ²¹⁰

Rare

• nausea, vomiting, abdominal pain, GI bleeding ^93,171
• nasal congestion (<10%) ⁹³
• palpitations, chest tightness ⁵⁰
• haemoptysis (<5%) ¹³⁴
• confusion ¹³⁷, seizures, paraesthesia, altered consciousness ^121,210
• stroke (6%) ^149-151,210 (most commonly cryptogenic ^150,151)

COVID-19 sufferers have reported high rates of disturbances of smell and taste, including anosmia, hyposmia, ageusia, and dysgeusia. The numbers of patients affected vary and current evidence points more towards a neurological than a conductive cause of the olfactory dysfunction ^{79,98,105-107,139}. 

Various reports suggest patients with the disease may have symptoms of conjunctivitis, and those affected, may have positive viral PCR in their conjunctival fluid ^103,104. However, a meta-analysis of over 1,100 patients found that conjunctivitis was only present in 1.1% cases ¹⁴⁰. A small case series found isolated conjunctivitis as the only visible manifestation in some COVID-19 patients ¹⁴¹. Epiphora has also been reported ²¹⁰.

Cutaneous lesions may also be seen, similar to many other viral infections. In a cohort of 88 patients, 20% developed skin disease, most commonly an erythematous rash, although vesicular/pustular eruptions, urticaria, maculopapular lesions and livedo reticularis are all seen ^100,210. Pseudochilblain is a distinctive skin manifestation characterised by erythematous tips of the digits accompanied by vesicles or pustules, and also known as COVID toes; the physical findings mimic those of frostbite ²¹⁰. Most of the skin abnormalities were self-limiting, resolving in a few days ¹⁰⁰. 

Paediatric

In the main, the clinical presentation in children with COVID-19 is milder than in adults ^59,90,210. Symptoms are similar to any acute chest infection, encompassing most commonly pyrexia, dry cough, sore throat, sneezing, myalgia and lethargy. Wheezing has also been noted ^59,90. Other less common (<10%) symptoms in children included diarrhoea, lethargy, rhinorrhoea and vomiting ⁹¹.

The reason for the less severe disease in children remains unclear. Some literature has suggested that viral loads may be lower in children than adults ²¹⁰.

Diagnosis

The definitive test for SARS-CoV-2 is the real-time reverse transcriptase-polymerase chain reaction (RT-PCR) test. This also allows searching for specific sequences from the viral genome: E (envelope protein gene), N (nucleocapsid protein gene) and RdRP (RNA-dependent RNA polymerase gene) ²⁰⁷. It is believed to be highly specific, but with sensitivity reported as low as 60-70% ³² and as high as 95-97% ⁵⁶. Meta-analysis has reported the pooled sensitivity of RT-PCR to be 89% ¹¹⁶. Thus, false negatives are a real clinical problem, and several negative tests might be required in a single case to be confident about excluding the disease.

Its sensitivity is predicated on time since exposure to SARS-CoV-2, with a false negative rate of 100% on the first day after exposure, dropping to 67% on the fourth day. On the day of symptom onset (~4 days after exposure) the false negative rate remains at 38%, and it reaches its nadir of 20% three days after symptoms begin (8 days post exposure). From this point on, the false negative rate starts to climb again reaching 66% on day 21 after exposure ¹³⁸.

Serological tests

The place of antibody tests in the investigation of present and previous cases of COVID-19 remains unclear and contentious. It may have a complementary role in diagnosis, especially when the RT-PCR test is negative in those whom remain suspected to have COVID-19 ¹⁶⁰.

CT as diagnostic test

Multiple radiological organisations and learned societies have stated that CT should not be relied upon as a diagnostic/screening tool for COVID-19 ^{52,57,87,88,116}. On 16 March 2020, an American-Singaporean panel published that CT findings were not part of the diagnostic criteria for COVID-19 ⁵⁶. However, CT findings have been used controversially as a surrogate diagnostic test by some ^2,32,89. 

Laboratory tests

The most common ancillary laboratory findings in patients, including a study of 61,742, were the following ^13,89:

• lymphopenia
• thrombocytosis
• increased prothrombin time (PT)
• increased lactate dehydrogenase

Other commonly identified abnormalities include:

• mild elevated inflammatory markers (CRP ⁸⁹ and ESR)
• elevated D-dimer
• mildly elevated serum amylase: 17% patients (study of 52 cases) ¹⁴⁵
  • acute pancreatitis has been reported, unclear if causal link with SARS-CoV-2 or an epiphenomenon ^186,187
• mildly deranged liver function tests are common, primarily elevated alanine aminotransferase (ALT) and aspartate aminotransferase (AST) ^145,146,210
  • bilirubin rise is generally mild ¹⁴⁶
  • alkaline phosphatase (ALP) and gamma‐glutamyl transferase (GGT) levels remain normal ¹⁴⁶

Complications

In one of the largest studies of hospitalised patients, reviewing 1,099 individuals across China, the admission rate to the intensive care unit (ICU) was 5% ⁹³. In this same study, 6% of all patients required ventilation, whether invasive or non-invasive. ICU patients tend to be older with more comorbidities ^13,93.

Commonly reported sequelae are:

• acute respiratory distress syndrome (ARDS): ~22.5% (range 17-29%) ⁸⁹
• acute thromboembolic disease ¹³⁰
  • pulmonary embolism ^{114,117,131-133}
  • deep vein thrombosis (DVT) ¹²³
• acute cardiac injury: elevated troponin levels ²¹⁰
  • up to 30% of COVID-19 cases that have been hospitalised, with rates exceeding 50% in those with a prior cardiac history
    • myocardial ischaemia
    • myocarditis ¹²⁰
    • arrhythmias
    • cardiomyopathy
    • cardiogenic shock
    • cardiac arrest
    • heart failure ²¹³
• CNS
  • delirium (15%) ^137,210
  • viral encephalitis ¹²¹
  • diffuse leukoencephalopathy ¹⁵⁷
  • microhaemorrhage ^157-159
    • juxtacortical white matter and corpus callosum (particularly of the splenium)
  • stroke (6%): cryptogenic/ischaemic ^150,151,210
    • higher mortality and greater severity of stroke in context of COVID-19 ¹⁵⁰
  • ^​Guillain-Barré syndrome (GBS): rare ¹⁸²
• secondary infections, e.g. bacterial pneumonia, mucormycosis ²¹⁵
• sepsis
• renal ²¹⁰
  • acute kidney injury (AKI) (20-40% of hospitalised patients) ²¹⁰
  • interstitial and/or haemorrhagic cystitis
• coagulopathy ¹⁴⁴
  • disseminated intravascular coagulation (DIC)
• hepatobiliary: liver is the most frequently affected organ after the lungs, although serious sequelae are rare ²¹⁰
  • fulminant liver failure is rare
  • portal vein thrombosis
  • hepatic vein thrombosis
  • acute cholecystitis: due to biliary stasis
• acute pancreatitis ²¹⁰
• gastrointestinal ²¹⁰
  • gastritis
  • enteritis and/or colitis
  • intestinal ischaemia: prothrombotic viral effects
  • small/large bowel perforation
• infarctions of solid abdominal viscera (18%) ²¹⁰
  • liver
  • spleen
  • kidney
• gonads ²¹⁰
  • autoimmune orchitis
• musculoskeletal
  • rhabdomyolysis (late sequela, case reports only) ²¹⁰
• multiorgan failure (MOF) ⁶⁶

In a small subgroup of severe ICU cases:

• secondary haemophagocytic lymphohistiocytosis (a cytokine storm syndrome) ⁴⁹

Risk factors for pulmonary thromboembolism

In a multivariate analysis, an elevated risk of developing PE was associated with ¹³³:

• obesity
• elevated D-dimer
• elevated CRP
• rising D-dimer over time

Paediatric complications

In April 2020, reports started to appear of critically-ill children presenting with a severe inflammatory state like atypical Kawasaki disease and toxic shock syndrome ^126,127. This is now called paediatric multisystem inflammatory state (PMIS). 

Other sequelae have also been reported in children, including ²¹⁰:

• necrotising enterocolitis (NEC)

Pathology

Aetiology

On 9 January 2020, the World Health Organisation (WHO) confirmed that SARS-CoV-2 was the cause of COVID-19 (2019-nCoV was the name of the virus at that time) ^14,37. It is one of the two strains of the SARS-CoV species known to cause human disease, the other being the original severe acute respiratory syndrome coronavirus (SARS-CoV-1), the cause of SARS. It is a member of the Betacoronavirus genus, one of the genera of the Coronaviridae family of viruses. Coronaviruses are enveloped single-stranded RNA viruses that are found in humans, mammals and birds. These viruses are responsible for pulmonary, hepatic, CNS, and intestinal diseases. 

As with many human infections, SARS-CoV-2 is zoonotic. The closest animal coronavirus by genetic sequence is a bat coronavirus, and this is the likely ultimate origin of the virus ^11,19,26. The disease can also be transmitted by snakes ²⁴.

Six other coronaviruses are known to cause human disease. Two are zoonoses: the severe acute respiratory syndrome coronavirus (SARS-CoV-1) and Middle East respiratory syndrome coronavirus (MERS-CoV), both of which may sometimes be fatal. The remaining four viruses all cause the common cold. 

Pathophysiology

The SARS-CoV-2 virus, like the closely-related MERS and SARS coronaviruses, effects its cellular entry via attachment of its virion spike protein (a.k.a. S protein) to the angiotensin-converting enzyme 2 (ACE2) receptor. This receptor is commonly found on alveolar cells of the lung epithelium, underlying the development of respiratory symptoms as the commonest presentation of COVID-19 ⁵⁰. It is thought that the mediation of the less common cardiovascular effects is also via the same ACE2 receptor, which is also commonly expressed on the cells of the cardiovascular system ⁵⁰.

Variants

Although the SARS-CoV-2 virus, like all viruses, continually mutates, therefore creating new variants perpetually, certain variants, have been found to be particularly worrisome to scientists. They might be more transmissible or pathogenic. These have been termed "variants of concern" (VOC). Originally these were given formal scientific designations, however the media and general public assigned them easy to remember names based on their site of first identification. This led to geographic-related stigmatisation, and thus the variants were renamed by the WHO on 31 May 2021, using letters from the start of the Greek alphabet ²¹⁷:

• alpha (B.1.1.7 - Kent variant, discovered Sep 2020)
• beta (B.1.351 - South African variant, discovered May 2020)
• gamma (P.1 - Brazilian variant, discovered Nov 2020)
• delta (B.1.617.2 - Indian variant, discovered Oct 2020)

If further variants of concern arise, then they will be given designations from "epsilon" onwards.

Transmission

Although originating from animals, COVID-19 is now considered to be an indirect zoonosis, as its transmission is now primarily human-to-human. It was initially thought to be predominantly transmitted in a similar way to the common cold, via contact with droplets of infected individuals' upper respiratory tract secretions, e.g. from sneezing or coughing ¹⁹.

It is now thought that aerosol transmission of SARS-CoV-2, i.e. airborne transmission, also occurs ^{58,161,162,172}. In the early stages of the pandemic aerosol risk was thought to be limited to so-called aerosol-generating procedures (AGPs) in healthcare facilities. However, there is an increasing evidence base to show that aerosols are also produced by talking, singing, coughing and expiration ¹⁶².

Fomites transmission is also seen, explaining the emphasis on regular handwashing and regular cleaning of surfaces that might have been exposed to droplets/aerosols containing virion particles ^{58,162,172-174}. The SARS-CoV-2 has a short half-life on some non-porous surfaces, including copper and latex, but is more persistent on other materials such as glass, plastics, stainless steel and porous fabrics ¹⁷⁴. 

Orofaecal spread was seen with the SARS epidemic, and although it remains unclear if SARS-CoV-2 can be transmitted in this way, there is some evidence for it ^19,43.

Sexual transmission has not been seen in the field but remains possible, not least because the SARS-CoV-2 virus has been found in all bodily secretions including seminal and vaginal fluids ¹⁴³.

It remains unclear if COVID-19 could be transmitted through a blood transfusion although no cases have yet be seen. Nevertheless, many national bodies have instituted controls to reduce the chance of this happening including advising that potential donors do not give blood until 28 days after recovering from COVID-19 ¹⁴².

Cohort studies have been unable to rule out the possibility of vertical transmission (mother-to-fetus transfer of COVID-19), but it seems to be a rare event if it does occur ^{21,82,94,152,210}. A large prospective cohort study of 427 pregnant women from all 194 birth units across the UK found that 5% of 265 live births were confirmed as COVID-19 on RT-PCR ¹⁵².

Presymptomatic carriers are likely present in many communities and presymptomatic transmission has been documented ¹⁶³. 

Asymptomatic carriage and transmission are also thought to occur ^163,164.

Considerations for medical imaging departments

Imaging indications

The threshold for the imaging of patients with potential/confirmed COVID-19 demonstrates a degree of variation globally due to local resources, the published guidelines of individual learned bodies and sociocultural approaches to imaging.

The use of CT as a primary screening tool is discouraged, not least because these studies tended to suffer from selection bias ^{52,57,87,88,115}, with a meta-analysis, in April 2020, reporting a pooled sensitivity of 94% and specificity 37% ¹¹⁶. In low prevalence (<10%) countries, the positive predictive value of RT-PCR was ten-fold that of CT chest ¹¹⁶.

According to a Fleischner Society consensus statement published on 7 April 2020 ¹⁰¹:

• imaging is not indicated in patients with suspected COVID-19 and mild clinical features unless they are at risk for disease progression
• imaging is indicated in a patient with COVID-19 and worsening respiratory status
• in a resource-constrained environment, imaging is indicated for medical triage of patients with suspected COVID-19 who present with moderate-severe clinical features and a high pretest probability of disease

Moreover performing CT routinely for large cohorts of patients carries additional risks ¹¹⁵:

• depletion of finite resources, especially PPE due to excessive usage
• increased risk of viral transmission (to staff, patients and carers) as COVID-19 positive and negative patients come into close proximity in the radiology department
• additional ionising radiation exposures

Infection precautions

Given that the staff in a medical imaging department are often in the frontline when dealing with COVID-19 patients, clear infection control guidelines are imperative. At the time of writing (August 2020) droplet-type precautions are in place for COVID-19 patients, that is, medical mask, gown, gloves, and eye protection (aerosol-generating procedures require N95 masks and aprons). A study of 420 clinical staff found that personal protective equipment (PPE) protected the development of SARS-CoV-2 infection ^39,155.

Patients requiring general radiography should receive it portably (to limit transporting patients) or in dedicated auxiliary units. Patients that require transport to departments must wear a mask to and from the unit. Machines, including any ancillary equipment used during examinations, should be cleaned after examinations ⁴⁰. It is recommended that any imaging examinations have two radiographers in attendance using the 'one clean, one in contact with the patient' system to minimize cross-contamination ⁸⁹. The causative organism, SARS-CoV-2, can survive on surfaces for up to 72 hours, reinforcing the need for protection of equipment with barriers such as covers and thorough cleaning of equipment between patients ⁵⁸.

There are case studies of portable chest x-rays performed through the glass window of the patient's room to decrease both staff exposure and amount of personal protective equipment ^102,165, although departmental protocols will vary significantly.

Please follow your departmental policies on personal protective equipment (PPE).

Non-urgent care

Both the American College of Radiology (ACR) and the Centers for Disease Control and Prevention (CDC) in the United States advise that non-urgent outpatient appointments should be rescheduled ^83,84. The British Society of Skeletal Radiologists has advised that intra-articular, soft tissue and perineural steroid injections may reduce viral immunity and therefore should not be performed unless they are unavoidable ⁸⁵. This decision has been criticised for its uncompromising approach, especially when the underlying evidence is far from being clear-cut, with a suggestion that for many the risk-benefit calculation is in favour in performing the joint injection ¹⁸¹.

CT protocol

Patients requiring CT should receive a non-contrast chest CT (unless iodinated contrast medium is indicated), with reconstructions of the volume at 0.625 mm to 1.5 mm slice thickness (gapless) ⁵⁷. If iodinated contrast medium is indicated, for example a CT pulmonary angiogram (CTPA), a non-contrast scan should be considered prior to contrast administration, as contrast may impact the interpretation of ground-glass opacification (GGO) patterns ⁸⁹.

Currently there is no diagnostic benefit to performing a CTPA examination on initial presentation ²⁰³. Although the risk of pulmonary thrombosis is higher in severe cases of COVID-19, it is recommended that D-dimer values are used to guide clinical pathways to justify a CTPA ^203-205.

Radiographic features

The primary findings of COVID-19 on chest radiograph and CT are those of atypical pneumonia ^40,175 or organising pneumonia ^32,34.

However imaging has limited sensitivity for COVID-19, as up to 18% demonstrate normal chest radiographs or CT when mild or early in the disease course, but this decreases to 3% in severe disease ^89,93. Bilateral and/or multilobar involvement is common ^6,78.

The current recommendation of the vast majority of learned societies and professional radiological associations is that imaging should not be employed as a screening/diagnostic tool for COVID-19, but reserved for the evaluation of complications ¹¹⁵.

Plain radiograph

Although less sensitive than chest CT, chest radiography is typically the first-line imaging modality used for patients with suspected COVID-19 ⁹⁷. For ease of decontamination, use of portable radiography units is preferred ⁵².

Chest radiographs may be normal in early/mild disease. In those COVID-19 cases requiring hospitalization, 69% had an abnormal chest radiograph at the initial time of admission, and 80% had radiographic abnormalities sometime during hospitalization ⁹⁷. Findings are most extensive about 10-12 days after symptom onset ⁹⁷.

The most frequent findings are airspace opacities, whether described as consolidation or, less commonly, GGO ^89,97. The distribution is most often bilateral, peripheral, and lower zone predominant ^89.97. In contrast to parenchymal abnormalities, pleural effusion is rare (3%) ⁹⁷.

Although cardiac manifestations of COVID-19 are well-recognised, there is no published evidence of cardiac disease on chest radiographs ¹⁷⁵.

CT

{{youtube:https://youtu.be/UTj3f06Mvwo}}

The primary findings on CT in adults have been reported as ^{13,17,27,28,36}:

• ground-glass opacities (GGO): bilateral, subpleural, peripheral
• crazy paving appearance (GGOs and inter-/intra-lobular septal thickening)
• air space consolidation
• bronchovascular thickening in the lesion
• traction bronchiectasis

The ground-glass and/or consolidative opacities are usually bilateral, peripheral, and basal in distribution ^2,32.

A retrospective study of 112 patients found 54% of asymptomatic patients had pneumonic changes on CT ⁶⁷.

The following chest CT findings have been reported to have the highest discriminatory value (p<0.001) ⁵¹:

• peripheral distribution
• ground-glass opacity
• bronchovascular thickening (in lesions)

A small number of patients have shown a pulmonary target sign ¹⁹⁶, which has only been reported in COVID-19 patients so far. At present it is unclear if this new sign is pathognomonic or simply newly recognized.

Atypical CT findings

These findings only seen in a small minority of patients should raise concern for superadded bacterial pneumonia or other diagnoses ^2,32,89:

• mediastinal lymphadenopathy ¹⁷
• pleural effusions: may occur as a complication of COVID-19
• multiple tiny pulmonary nodules (unlike many other types of viral pneumonia)
• tree-in-bud
• pneumothorax
• cavitation
• atoll sign
• pneumomediastinum ²⁰⁶

Temporal CT changes

Four stages on CT have been described ^17,24,32,86:

• early/initial stage (0-4 days): normal CT or GGO only
  • up to half of patients have normal CT scans within two days of symptom onset
• progressive stage (5-8 days): increased GGO and crazy paving appearance
• peak stage (9-13 days): consolidation
• absorption stage (>14 days): with an improvement in the disease course, "fibrous stripes" appear and the abnormalities resolve at one month and beyond

Paediatric CT

In a small study of five children that had been admitted to hospital with positive COVID-19 RT-PCR tests and who had CT chest performed, only three children had abnormalities. The main abnormality was bilateral patchy ground-glass opacities, similar to the appearances in adults, but less florid, and in all three cases the opacities resolved as they clinically recovered ⁴⁸.

On 18 March 2020, the details of a much larger cohort of 171 children with confirmed COVID-19, and evaluated in a hospital setting was published as a letter in the New England Journal of Medicine. Ground-glass opacities were seen in one-third of the total, whereas almost 16% of children had no imaging features of pneumonia ⁹¹.

Ultrasound

Initial work on patients in China suggests that lung ultrasound may be useful in the evaluation of critically ill COVID-19 patients ⁵⁵. The following patterns have been observed, tending to have a bilateral and posterobasal predominance:

• multiple B-lines
  • ranging from focal to diffuse with spared areas ⁶⁴
  • representing thickened subpleural interlobular septa
    • may also manifest as a light beam sign, an evanescent, broad-based vertical reverberation artifact arising from a regular pleural line ¹²⁸
• irregular, thickened pleural line with scattered discontinuities ⁶³
• subpleural consolidations
  • can be associated with a discrete, localized pleural effusion
  • relatively avascular with colour flow Doppler interrogation
  • pneumonic consolidation typically associated with preservation of flow or hyperemia ⁶⁵
• alveolar consolidation
  • tissue-like appearance with dynamic and static air bronchograms
  • associated with severe, progressive disease 
• restitution of aeration during recovery
  • reappearance of bilateral A-lines

Nuclear medicine

PET-CT

FDG uptake is increased in ground-glass opacities in those with presumed/confirmed COVID-19 ^42,75,167. It has been hypothesised that those with higher SUVs in lung lesions take longer to heal ⁷⁷. 

Radiology report

Plain radiograph

The British Society of Thoracic Imaging (BSTI) have published a reporting proforma for the plain chest radiographic appearances of potential COVID-19 cases ¹⁶⁸. 

• classic/probable COVID-19
  • lower lobe and peripheral predominant multiple opacities that are bilateral (>> unilateral)
• indeterminate for COVID-19
  • does not fit classic or non-COVID-19 descriptors
• non-COVID-19
  • pneumothorax / lobar pneumonia / pleural effusion(s) / pulmonary oedema / other
• normal
  • COVID-19 not excluded

CT

The Radiological Society of North America (RSNA) has released a consensus statement endorsed by the Society of Thoracic Radiology and the American College of Radiology (ACR) that classifies the CT appearance of COVID-19 into four categories for standardised reporting language ⁹⁹:

• typical appearance
  • peripheral, bilateral, GGO +/- consolidation or visible intralobular lines (“crazy paving” pattern)
  • multifocal GGO of rounded morphology +/- consolidation or visible intralobular lines (“crazy paving” pattern)
  • reverse halo sign or other findings of organizing pneumonia
• indeterminate appearance
  • absence of typical CT findings and the presence of
    • multifocal, diffuse, perihilar, or unilateral GGO +/- consolidation lacking a specific distribution and are non-rounded or non-peripheral
    • few very small GGO with a non-rounded and non-peripheral distribution
• atypical appearance
  • absence of typical or indeterminate features and the presence of
    • isolated lobar or segmental consolidation without GGO
    • discrete small nodules (e.g. centrilobular, tree-in-bud) 
    • lung cavitation
    • smoother interlobular septal thickening with pleural effusion
• negative for pneumonia:no CT features to suggest pneumonia, in particular, absent GGO and consolidation

A study evaluating the RSNA chest CT classification system for COVID-19 against RT-PCR results found moderate interobserver agreement. Using a cohort of 96 patients, it reported that 76.9-96.6% of "typical" scans, 51.2-64.1% of "indeterminate" scans, 2.8-5.3% "atypical" scans and 20-25% of "negative" scans returned a RT-PCR confirming COVID-19 ^99,147.

{{youtube:https://www.youtube.com/watch?v=fEiGj5z2dLw}}

CO-RADS

In March 2020, the "COVID-19 standardized reporting working group" of the Dutch Association for Radiology (NVvR) proposed a CT scoring system for COVID-19. They called it CO-RADS (COVID-19 Reporting and Data System) to ensure CT reporting is uniform and replicable. This assigns a score of CO-RADS 1 to 5, dependent on the CT findings. In some cases a score of 0 or 6 may need to be assigned as an alternative. If the CT is uninterpretable then it is CO-RADS 0, and if there is a confirmed positive RT-PCR test then it is CO-RADS 6 ^109,124.

The first study investigating the use of CO-RADS found a reasonable level of interobserver variation, with a Fleiss' kappa score of 0.47 (cf. 0.24 for PI-RADS and 0.67 for Lung-RADS) ¹²⁴. Another study found that CO-RADS is sensitive and specific for diagnosing COVID-19, using RT-PCR as the gold standard ¹⁸⁴. CO-RADS-AI, a system using deep machine learning has been validated for diagnosing COVID-19 on CT chest ¹⁸³.

COVID-RADS

In April 2020, American radiologists based at the University of Southern California proposed the COVID-19 imaging reporting and data system (COVID-RADS), which has a confusingly similar name to CO-RADS (see above) ¹²⁵.

CO-RADS vs COVID-RADS

A 2020 study comparing the head to head performance of the CO-RADS and COVID-RADS reporting systems for 200 cases of COVID-19, found that interobserver consensus was similarly moderate to good, however CO-RADS demonstrated a higher median intraobserver consensus ²¹⁶.

Treatment and prognosis

Treatment

No specific treatment exists for COVID-19 (January 2021). Therefore resources have been concentrated on public health measures to prevent further interhuman transmission of the virus. This has required a multipronged approach and for individuals includes ¹¹:

• meticulous hand-washing
• wearing of face masks
• social distancing
• avoidance of large crowds/crowded environments
• self-isolation

There is now increasing concern that aerosol transmission has a more significant role in viral spread than initially thought and this means that additional public health measures will need to be introduced, these will include ^161,162:

• better ventilation, especially in indoor public spaces, workplaces, educational establishments, healthcare facilities, and community residential centres for the elderly, in many cases this will be as simple as opening windows
• specific aerosol infection control systems, such as:
  • high throughput and effective air filtration
  • virucidal UV (ultraviolet) lighting

In healthcare facilities, concerted efforts are required to effect rapid diagnosis, quarantine infected cases and provide effective supportive therapies. This will encompass empirical treatments with antibiotics, antivirals, and supportive measures.

Mechanical ventilation, both invasive and non-invasive, and extracorporeal membrane oxygenation (ECMO) have also been used where clinically necessary.

Proning

Historical studies have demonstrated a net benefit for patients with moderate to severe ARDS being turned prone ¹¹⁸. Many health care facilities have adopted the practice of turning the sicker COVID-19 patients into a prone position, so-called "proning" to improve their lung oxygenation ¹¹⁹. 

Antiviral therapy

Whilst specific antiviral therapies for SARS-2-CoV do not currently exist, the combination of the protease inhibitors, ritonavir, and lopinavir, or a triple combination of these antiviral agents with the addition of ribavirin, showed some success in the treatment of SARS ²⁰, and early reports suggested similar efficacy in the treatment of COVID-19 ²³. However, a more recent randomized, controlled open-label trial failed to demonstrate any added benefit of lopinavir-ritonavir combination therapy ⁶⁶.

Combination of remdesivir and baricitinib shown to be effective against MERS-CoV and SARS-CoV, showed promising in vitro results against SARS-CoV-2 ^29,197. A preliminary trial in May 2020 showed a small decrease in time to recovery, from 15 to 11 days, in those treated with remdesivir ¹⁵³. However there are concerns about this evidence, and remdesivir has not been shown to decrease mortality nor reduce the length of hospital stay ¹⁶⁹. Other antivirals in phase III trials include oseltamivir, ASC09F (HIV protease inhibitor), lopinavir, ritonavir, darunavir, and cobicistat ⁸⁰.

Dexamethasone, a synthetic steroid, was demonstrated in the large RECOVERY (Randomised Evaluation of COVid-19 thERapY) randomised controlled trial, in June 2020 to decrease deaths by a third in those on mechanical ventilation (p=0.0003), and by a fifth those non-ventilated patients requiring oxygen (p=0.0021). No benefit was seen in those not needing respiratory support ^148,198.

In early 2020, published reports showed that two antimalarial drugs, chloroquine, and its close chemical derivative, hydroxychloroquine, had strong anti-SARS-2-CoV activity in vitro. An initial open-label, randomized clinical trial, demonstrated a significant reduction of viral carriage, and a lower average carrying duration in patients treated with hydroxychloroquine. Furthermore, a combination with the antibiotic azithromycin resulted in a synergistic effect ⁶⁹. However, this trial was later strongly criticised for methodological flaws and questionable conclusions. Later studies have failed to replicate the beneficial effects of these agents and also highlight potential side-effects ¹³⁵.

Passive immunity

Treatment with convalescent plasma (plasma from patients who have recovered from COVID-19 which therefore contains anti-SARS-CoV-2 antibodies) or hyperimmune immunoglobulin (purified antibodies prepared from convalescent plasma) has shown some success in some critically ill patients. Reports are still preliminary and about a small number of patients ^110-112,136. A Cochrane review in May 2020 failed to find convincing evidence that convalescent plasma was an effective treatment, but this will be kept under active review ¹³⁶.

Vaccines

The primary target in developing coronavirus vaccines has been the spike protein (S protein) which is on the surface of the virion particle, and in vivo is the most important antigen for triggering an immune response ⁷⁵. Human vaccines for coronaviruses have been under development since the SARS outbreak 2002-2004, none have been approved for immunisation against SARS or MERS ^11,26. Over 125 vaccine candidates are now in preclinical/clinical trials ¹⁵⁴.

In December 2020, a novel mRNA vaccine originally designated BNT162b2, and now with the generic name tozinameran ²⁰², and co-developed by BioNTech and Pfizer, demonstrated 95% efficacy in protecting adult (>16 years old) subjects, inoculated with two 3-week apart doses, from COVID-19 infection ²⁰⁰. The vaccine - made up of molecules of messenger ribonucleic acid (mRNA) and inserted into lipid nanoparticles (LNPs: ALC-0315 and ALC-0159) to facilitate their entry into human cells - contains the instructions for human cells to synthesize spike protein. The proteins thus produced will stimulate the immune system to produce specific antibodies ²⁰⁰. This messenger RNA, compared to that of viral origin, presents changes to consider, the most important of which is the replacement of uracil with the 1-methylpseudouridine base; modifications that allow it to evade the immune system to be used correctly by ribosomes ^208,209. This vaccine's side-effects were mild including pain at the injection site, headaches and fatigue. Anaphylaxis has been reported in two vaccinees in the UK ²⁰¹. This vaccine has received regulatory approval in multiple territories ²⁰¹.

NSAIDs

In early 2020 there were concerns raised about the safety of non-steroidal anti-inflammatory drugs (NSAIDs) in those with COVID-19. This was based upon anecdotal reports and expert opinion rather than published scientific evidence ^61,166. There remains no evidence that the use of NSAIDs increases the risk of developing COVID-19 or worsens established disease ¹⁶⁶.

ACE inhibitors

Since the SARS-CoV-2 virus acts via the angiotensin converting enzyme 2 (ACE2) receptor, there have been questions raised about the effect that being on ACE inhibitors or angiotensin receptor blockers might have on COVID-19. A randomised controlled trial in Brazil could find no difference in survival between those stopping or continuing taking the agents ²¹⁴. 

Prognosis

In Italy, a document detailing 6801 patients who died from COVID-19 discovered that the median time from the first symptoms to hospital admission was five days, and nine days from symptom onset to death ¹⁷⁸.

Progressive deterioration of imaging changes despite medical treatment is thought to be associated with poor prognosis ²⁷. There is an increased risk of death in men over the age of 60 years old ⁶². The mortality rate is estimated to be 3.6% ⁸⁹.

Early reports show that in some well patients, the RT-PCR test remains falsely positive despite an apparent clinical recovery. This raises the concern that asymptomatic carriage may occur ³⁵.

In a study of 101 patients with COVID-19 who were referred for palliative care, breathlessness, delirium, drowsiness and agitation were the most frequently found symptoms ¹⁸⁰.

Risk factors for severe illness or poor outcome

Studies have shown that the most important risk factors for poor outcome are increasing age, male sex and obesity ^68,95,188. Indeed obesity seems to trump all other comorbidities as a risk factor and its risk is not limited to inpatient populations ¹⁸⁸.

• general
  • vitamin D deficiency ^190-195
  • old age
  • male gender
  • people in a long-term care facility or nursing home
• comorbidities ^68,95,188
  • obesity: 4x increased deaths if body mass index >45 kg/m^{2 188}
  • cardiovascular disease
  • diabetes mellitus
  • hypertension
  • chronic respiratory disease, e.g. COPD, asthma
  • cancer
  • chronic liver disease
  • chronic renal disease
  • immunosuppression
  • organ transplantation ¹⁸⁸
• patient condition and laboratory values at hospital admission ⁹⁶
  • high sequential organ failure assessment (SOFA) score on admission
  • D-dimer levels greater than 1 µg/mL on hospital admission
  • elevated levels of IL-6, troponin I, lactate dehydrogenase
  • lymphopenia

Pregnancy

There is no evidence that pregnant women are more likely to contract COVID-19 or more likely to experience complications from it ^152,210. In a cohort of 427 women in the UK, 10% required admission to critical care for respiratory support and 1% succumbed to the disease ¹⁵². 

The imaging of pregnant COVID-19 patients has found the same typical CT findings as other infected adults ²¹⁰.

Long covid

"Long covid" has been used to refer to the condition whereby those who have recovered from COVID-19 still experience persistent manifestations or those with COVID-19 suffer symptoms for a longer time than normal ¹⁷⁷. 

As of August 2020 there is a dearth of published evidence of the long term effects following COVID-19 infection. An Italian study of 143 patients who had been discharged from hospital following a recovery from COVID-19 reveals that almost 90% people were still complaining of at least one symptom 60 days after the illness had begun ¹⁷⁶. Common symptoms included lethargy, breathlessness, arthralgia and chest discomfort ¹⁷⁶.

History and etymology

The first cases were seen in Wuhan, China, in early December 2019 before spreading globally ^1,2,10. 

The first mention in the medical press about the emerging infection was in the British Medical Journal (BMJ) on 8 January 2020 in a news article, which reported "outbreak of pneumonia of unknown cause in Wuhan, China, has prompted authorities in neighbouring Hong Kong, Macau, and Taiwan to step up border surveillance, amid fears that it could signal the emergence of a new and serious threat to public health" ⁵⁴. On 9 January 2020, the World Health Organisation confirmed that SARS-CoV-2 was the cause of the new disease ^14,37.

The first scientific article about the new disease, initially termed 2019‐new coronavirus (2019‐nCoV) by the World Health Organisation (WHO), was published in the Journal of Medical Virology on 16 January 2020 ⁵³.

On 13 January 2020, the first confirmed case outside China was diagnosed, a Chinese tourist in Thailand ¹⁰. On 20 January, the first infected person in the United States was confirmed to be a man who had recently returned from Wuhan ⁹. The infection was declared a Public Health Emergency of International Concern (PHEIC) on 30 January 2020 by the WHO ⁷. On 28 February 2020, the WHO increased the global risk assessment of COVID-19 to “very high” which is the highest level. On 11 March 2020, COVID-19 was declared a pandemic by the WHO ⁴⁴.

On 27 March 2020, the USA surpassed China as the country with the most confirmed cases ⁵. The number of confirmed cases globally exceeded one million on 3 April 2020, two million on 15 April, five million on 21 May, 10 million on 28 June, 15 million on 23 July, 20 million on 11 August, 25 million on 30 August, 50 million on 7 November, and 100 million on 26 January 2021 ⁵. The number of global deaths surpassed 100,000 on 10 April, 200,000 on 26 April, 500,000 on 28 June, 750,000 on 13 August, one million on 29 September 2020, two million on 15 January 2021, three million on 17 April 2021 and four million on 9 July 2021 ⁵.

The WHO originally called this illness "novel coronavirus-infected pneumonia (NCIP)" and the virus itself had been named "2019 novel coronavirus (2019-nCoV)" ¹. On 11 February 2020, the WHO officially renamed the clinical condition COVID-19 (a shortening of COronaVIrus Disease-19) ¹⁵. On the same day, the Coronavirus Study Group of the International Committee on Taxonomy of Viruses renamed the virus "severe acute respiratory syndrome coronavirus 2" (SARS-CoV-2) ^16,22,46. 

Differential diagnoses

• viral pneumonia including ^71,72:
  • influenza pneumonia A and B
    • distribution more along the bronchovascular bundles
    • bronchial wall thickening
  • paramyxovirus pneumonia
  • cytomegalovirus (CMV) pneumonia
  • adenovirus pneumonia ^71,72
  • SARS-CoV pneumonia
  • MERS coronavirus
  • HSV pneumonia
    • in immunocompromised patients
    • often shows pleural effusions
  • respiratory syncytial virus (RSV) pneumonia
• atypical bacterial pneumonia
  • mycoplasma pneumonia
    • mainly children and adolescents
    • bronchial wall thickening
    • centrilobular nodules
  • chlamydia pneumonia
  • legionella pneumonia
  • Coxiella burnetii pneumonia
• community-acquired pneumonias ¹⁷⁵
  • usually unilateral opacities (cf. bilateral for most COVID-19)
  • consolidation mainly, ground glass opacity unusual (on radiographs)
  • ^​Streptococcus pneumoniae
  • Haemophilus influenzae
  • Klebsiella pneumoniae
• ​pulmonary oedema ⁷¹
• interstitial lung disease ⁷³
  • cryptogenic organizing pneumonia
  • chronic eosinophilic pneumonia (CEP) 
  • acute fibrinous organising pneumonia
  • rheumatoid arthritis-associated pneumonia ⁷¹
• certain drug-induced pneumonitides
  • immune checkpoint inhibitor therapy-related pneumonitis
• neutropenic sepsis
• aspiration pneumonitis
• for a "crazy paving" predominant manifestation also consider conditions such as 
  • pulmonary alveolar proteinosis

Clinical differential diagnoses

The clinical differential diagnosis is very similar to the imaging differential when patients present with typical symptoms, e.g. cough and fever. However, some divergence might be seen if there are less typical presentations, e.g. acute breathlessness, which might raise suspicion for acute pulmonary embolism which is not an imaging differential in many cases ¹³⁴.

Resources

These lists are in alphabetical order:

• general information
  • COVID-19 global cases by Johns Hopkins
  • World Health Organisation information page 
• research publications and datasets
  • ARRS - AJR Open Access COVID-19 Collection
  • British Medical Journal COVID-19 collection
  • COVID-19 Open Research Dataset (CORD-19)
  • European Radiology COVID-19 - latest articles
  • New England Journal of Medicine COVID-19 resources
  • RSNA COVID-19 research
  • The Lancet COVID-19 resource centre
  • World Health Organisation Database of publications on COVID-19
• government information
  • Australia (Department of Health)
  • Australia (Smartraveller)
  • Canada (Government of Canada)
  • Canada (Infection Prevention and Control Canada)
  • Europe (European Centre for Disease Prevention and Control)
  • Israel (Ministry of Health)
  • Italy (Dipartimento della Protezione Civile)
  • Mexico (Secretaría de Salud)
  • New Zealand (Ministry of Health)
  • Singapore (Ministry of Health)
  • United Kingdom (National Health Service UK)​
  • United States of America (Centers for Disease Control and Prevention)

See also

• acute respiratory distress syndrome (ARDS)
• human coronaviruses
• Middle East respiratory syndrome coronavirus (MERS-CoV)
• secondary haemophagocytic lymphohistiocytosis
• severe acute respiratory syndrome coronavirus (SARS-CoV)
• World Health Organisation (WHO)