Recommendations

Key Recommendations

Management predominantly depends on disease severity, and focuses on the following principles: isolation at a suitable location; infection prevention and control measures; symptom management; prevention of disease progression; optimised supportive care; and organ support in severe or critical illness.

Consider whether the patient can be managed at home. Generally, patients with asymptomatic or mild to moderate disease can be managed at home.[93] Provide symptom relief as necessary, including treatments for fever or cough.[93][523]Consider antiviral or monoclonal antibody treatment in patients with non-severe disease who are at highest risk of hospitalisation.[459][523][726]

Admit patients with severe disease to an appropriate healthcare facility. Assess adults for frailty on admission. Patients with critical disease require intensive care; involve the critical care team in discussions about admission to critical care when necessary. Monitor patients closely for signs of disease progression.[93][523]

Start supportive care according to the clinical presentation. This might include symptom relief, oxygen therapy, intravenous fluids, venous thromboembolism prophylaxis, high-flow nasal oxygen (HFNO), non-invasive or invasive mechanical ventilation, or extracorporeal membrane oxygenation. Sepsis and septic shock should be managed according to local protocols.[93]

Consider empirical antibiotics if there is clinical suspicion of a secondary bacterial infection. Antibiotics may be required in patients with moderate, severe, or critical disease. Give within 1 hour of initial assessment for patients with suspected sepsis or if the patient meets high-risk criteria. Base the regimen on the clinical diagnosis, local epidemiology and susceptibility data, and local treatment guidelines.[93][523]

Consider systemic corticosteroid therapy in patients with severe or critical disease. Moderate-quality evidence suggests that systemic corticosteroids probably reduce 28-day mortality in patients with severe and critical disease, and probably reduce the need for invasive ventilation.[459][523][726]

Consider the antiviral remdesivir in patients with severe disease. Low-certainty evidence suggests that remdesivir possibly reduces mortality, and moderate-certainty evidence suggests that remdesivir probably reduces the need for mechanical ventilation.[459][523][726] 

Consider an interleukin-6 inhibitor (tocilizumab or sarilumab) and/or a Janus kinase inhibitor (baricitinib) in patients with severe or critical disease. High-certainty evidence suggests that interleukin-6 inhibitors reduce mortality and the need for mechanical ventilation. High-certainty evidence suggests that baricitinib reduces mortality, and moderate-certainty evidence suggests that it probably reduces the duration of mechanical ventilation.[459][523][726]

Assess whether the patient requires any rehabilitation or follow-up after discharge. Discontinue transmission-based precautions (including isolation) and release patients from the care pathway according to your local guidance.[93]

Full recommendations

Implement local infection prevention and control procedures when managing patients. For patients in home isolation, advise patients and household members to follow appropriate infection prevention and control measures:

Guidance on when to stop isolation varies widely across locations.

  • Isolation periods may depend on various factors including vaccination status, circulating severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants, and patient factors (e.g., immunocompetent/immunocompromised, asymptomatic/symptomatic, disease severity).

  • The World Health Organization recommends discontinuing transmission-based precautions (including isolation) and releasing patients from the care pathway 10 days after positive test (asymptomatic patients), or 10 days after symptom onset plus at least 3 days without fever and respiratory symptoms (symptomatic patients).[93]

  • However, some countries now recommend isolation periods as short as 5 days to 7 days.[727]

  • Consult your local public health guidance for more information.

Treatment options depend on factors including disease severity, availability, and local guidelines.

The World Health Organization strongly recommends corticosteroids, interleukin-6 inhibitors, or Janus Kinase inhibitors (baricitinib) for patients with severe or critical disease, and nirmatrelvir/ritonavir for patients with non-severe disease. It makes weak or conditional recommendations in favour of molnupiravir or remdesivir for non-severe disease, and remdesivir in those with severe disease.[726][728][729]

[Figure caption and citation for the preceding image starts]: Summary of therapeutic options from the WHOBMJ. 2022 Sep 15;378:o2224 [Citation ends].Summary of therapeutic options from the WHO

The management recommendations in this topic are based on guidelines from the World Health Organization, as well as key guidance from the UK National Institute for Health and Care Excellence, the US National Institutes of Health, and the Infectious Diseases Society of America. Consult your local guidelines for more detailed or specific information.

Patients with suspected or confirmed mild disease (i.e., symptomatic patients meeting the case definition for COVID-19 without evidence of hypoxia or pneumonia) or moderate disease (i.e., clinical signs of pneumonia but no signs of severe pneumonia) and asymptomatic patients should be isolated to contain virus transmission (follow advice from your local public health authority).[93]

  • Approximately 80% of patients have mild illness that does not warrant medical intervention or hospitalisation, depending on the circulating SARS-CoV-2 variant.[459]

  • The pooled proportion of non-severe illness in people infected with the Omicron variant was 98%, and the pooled proportion of asymptomatic infection was 25% (proportions varied depending on vaccination status).[499]

Location of care

  • Manage patients in a healthcare facility, in a community facility, or at home. Home isolation can be considered in most patients, with telemedicine or remote visits as appropriate. Manage patients at high risk of deterioration in a healthcare facility.[93][459]

    • Observational evidence suggests that implementation of an early home treatment algorithm/remote patient monitoring programme reduced the risk of hospitalisation, intensive care unit admission, and length of hospital stay.[730][731]

  • This decision requires careful clinical judgement and should be informed by an assessment of the patient’s home environment to ensure that: infection prevention and control measures and other requirements can be met (e.g., basic hygiene, adequate ventilation); the carer is able to provide care and recognise when the patient may be deteriorating; the carer has adequate support (e.g., food, supplies, psychological support); the support of a trained health worker is available in the community.[732]

Symptom management

  • Fever and pain: paracetamol or ibuprofen are recommended.[93][523] Ibuprofen should only be taken at the lowest effective dose for the shortest period needed to control symptoms.

  • Cough: advise patients to avoid lying on their back as this makes coughing ineffective. Use simple measures first (e.g., a teaspoon of honey in patients aged 1 year and older) to help cough.[523]

    • A meta-analysis found that honey is superior to usual care (e.g., antitussives) for the improvement of upper respiratory tract infection symptoms, particularly cough frequency and severity.[733]

  • Olfactory dysfunction: consider treatment (e.g., olfactory training, intranasal corticosteroids) if olfactory dysfunction persists beyond 2 weeks. Often it improves spontaneously and does not require specific treatment.[734][735]

    • A Cochrane review found there is very limited evidence regarding the efficacy of different interventions at preventing persistent olfactory dysfunction following infection. The only evidence available is for intranasal corticosteroids, and this is of very low certainty, so no conclusions could be drawn.[736]

    • A systematic review and meta-analysis found that there were no significant differences in the improvement of olfactory scores with either intranasal or oral corticosteroids plus olfactory training compared with olfactory training alone. Olfactory function was significantly improved after olfactory training.[737]

Supportive care

  • Advise patients about adequate nutrition and appropriate rehydration. Advise patients to drink fluids regularly to avoid dehydration. Fluid intake needs can be higher than usual because of fever. However, too much fluid can worsen oxygenation.[93][523]

  • Advise patients to improve air circulation by opening a window or door.[523]

  • Provide basic mental health and psychosocial support for all patients, and manage any symptoms of insomnia, depression, or anxiety as appropriate.[93]

  • Most children with mild or moderate disease can be managed with supportive care alone and will not require any specific therapy (unless considered high risk for progression to severe disease).[459]

Antivirals

  • Antiviral agents are approved or have an emergency-use authorisation in some countries. Options include:

    • Nirmatrelvir/ritonavir: nirmatrelvir is an oral SARS-CoV-2 protease inhibitor. It is administered with a low dose of ritonavir to slow the hepatic metabolism of nirmatrelvir and increase the plasma concentration of nirmatrelvir to a therapeutic level

    • Molnupiravir: an oral SARS-CoV-2 nucleoside analogue

    • Remdesivir: an intravenous RNA polymerase inhibitor.

  • The World Health Organization strongly recommends nirmatrelvir/ritonavir, and conditionally recommends remdesivir or molnupiravir, in patients with non-severe disease who are at highest risk of hospitalisation.[726][728][729] 

    • Nirmatrelvir/ritonavir is a superior choice to other treatments for non-severe disease because it may have greater efficacy in preventing hospitalisation compared with the alternatives, has fewer concerns with respect to harms than does molnupiravir, and is easier to administer than intravenous remdesivir. However, it does have significant and complex drug-drug interactions.

    • The recommendation for remdesivir is based on moderate-certainty evidence that suggests remdesivir probably reduces hospital admission, and low-certainty evidence that suggests it may have little or no impact on mortality. The balance between benefits and potential harms favours treatment, but only in the highest risk group.

    • The recommendation for nirmatrelvir/ritonavir is based on moderate-certainty evidence that suggests nirmatrelvir/ritonavir likely reduces hospital admission, and low-certainty evidence that suggests it may have little or no impact on mortality.

    • The recommendation for molnupiravir is based on moderate-certainty evidence that suggests molnupiravir probably reduces hospital admission and time to symptom resolution, and low-certainty evidence that suggests it may reduce mortality.

  • In the UK, the National Institute for Health and Care Excellence recommends nirmatrelvir/ritonavir or molnupiravir or remdesivir for patients who do not need supplemental oxygen, and are thought to be at high risk of progression to severe disease.[523] 

  • In the US, the National Institutes of Health guidelines panel recommends nirmatrelvir/ritonavir and remdesivir as preferred treatments, and molnupiravir as an alternative treatment (i.e., when preferred therapies are not available, feasible to use, or clinically appropriate), for non-hospitalised patients with mild to moderate disease who are at high risk of clinical progression.[459]

    • The Infectious Diseases Society of America supports the use of antivirals in these patients.[460]

  • Treatment should be initiated as soon as possible after diagnosis, ideally within 5 days of symptom onset for nirmatrelvir/ritonavir or molnupiravir, or within 7 days of symptom onset for remdesivir.[459][523][726]

    • If a patient requires hospitalisation after starting treatment, the full treatment course can be completed at the healthcare provider’s discretion.

    • Logistical or supply constraints may make patient triage for antiviral treatment necessary. Therapy should be prioritised for patients who are at the highest risk of progressing to severe disease.

    • Logistical constraints may make it difficult to administer remdesivir in some outpatient settings as it requires administration via intravenous infusion.

  • Antivirals are not generally recommended in children <12 years of age, pregnant women, or breastfeeding women.[459][523][726] However, recommendations vary, and you should consult your local guidelines.

  • Cases of viral rebound (i.e., recurrent positive polymerase chain reaction result) and the recurrence of symptoms have been reported 2 to 8 days after recovery in patients who have completed a 5-day course of nirmatrelvir/ritonavir, including patients who have been vaccinated.[738][739][740][741]

    • The recurrence of symptoms may represent part of the natural history of infection, or may be related to other factors (e.g., reinfection, emergence of treatment-resistant mutations).

    • Symptoms appear to be milder than those experienced during the primary infection and are unlikely to lead to hospitalisation.

    • There is currently no evidence that additional treatment is needed, and patients should follow their local public health guidelines.

    • The frequency and clinical implications of these events are not yet known, and further research is required.

    • The US Food and Drug Administration has instructed the manufacturer to evaluate an additional course of therapy in people with rebound infections by September 2023.

  • Evidence for the use of antivirals in patients with non-severe disease is limited.

    • Nirmatrelvir/ritonavir was found to reduce the risk of hospitalisation or death by 89% (within 3 days of symptom onset) and 88% (within 5 days of symptom onset) compared with placebo in non-hospitalised high-risk adults in the phase 2/3 EPIC-HR trial.[742]

    • Molnupiravir was found to reduce the risk of hospitalisation or death by 31% (absolute risk reduction from 9.7% to 6.8%) in the 29 days after use compared with placebo in non-hospitalised at-risk adults in the phase 3 MOVe-OUT trial.[743]

    • Remdesivir was found to reduce the risk of hospitalisation or death by 87% compared with placebo in non-hospitalised high-risk adults in a randomised, double-blind, placebo-controlled trial.[744]

    • Evidence of clinical efficacy for nirmatrelvir/ritonavir and molnupiravir was initially based on interim analyses of data from single placebo-controlled trials in unvaccinated adults conducted before the emergence of the Omicron variant. Since then, observational studies conducted during periods when the Omicron variant (and subvariants) were dominant suggest that treatment with nirmatrelvir/ritonavir or molnupiravir was associated with a reduced risk of progression to severe disease, hospitalisation, or death.[745][746][747]

    • All three antivirals appear to retain activity against the Omicron variant and its subvariants in vitro.[748][749]

Monoclonal antibodies

  • Monoclonal antibodies are recommended for patients with mild to moderate disease who are at high risk of clinical progression. However, availability and indications may vary between countries.[750] Consult your local guidance for more information.

    • Antibodies bind to non-overlapping epitopes of the receptor-binding domain of the spike protein to block virus entry into host cells.

    • Outpatient administration in specialised clinics is required, which may limit the feasibility of these treatments.[726] 

    • Logistical or supply constraints may make patient triage for monoclonal antibody treatment necessary. Therapy should be prioritised for patients who are at the highest risk of progressing to severe disease.

  • Choice of monoclonal antibody depends on availability, as well as clinical and contextual factors including information about efficacy with different SARS-CoV-2 variants and subvariants.[459][523][726]

    • Options may include bebtelovimab, tixagevimab/cilgavimab, casirivimab/imdevimab, sotrovimab, bamlanivimab/etesevimab, and regdanvimab, depending on your location.

    • Preclinical evidence suggests that casirivimab/imdevimab and bamlanivimab/etesevimab lack neutralisation activity against the Omicron variant and its subvariants in vitro. Sotrovimab and bebtelovimab appear to retain activity against the Omicron variant; however, sotrovimab is not active against the Omicron subvariants. Bebtelovimab appears to retain activity against the Omicron BA.2, BA.2.75, BA.4, and BA.5 subvariants.[749][751][752]

  • In the UK, the National Institute for Health and Care Excellence recommends offering a neutralising monoclonal antibody to patients ≥12 years of age who are not in hospital and are thought to be at high risk of progression to severe disease.[523]

    • Consult local guidance for monoclonal antibodies with current UK access.

  • In the US, the National Institutes of Health guidelines panel recommends bebtelovimab for the treatment of non-hospitalised patients aged ≥18 years with mild to moderate disease who are at high risk of progressing to severe disease. However, it is only recommended when preferred therapies (i.e., oral antivirals) are not available, feasible to use, or clinically appropriate as it has not been evaluated in patients at high risk of progression in placebo-controlled trials.[459]

    • Treatment should be started as soon as possible and within 7 days of symptom onset.

    • There is insufficient evidence to recommend either for or against the use of bebtelovimab in children and adolescents with mild to moderate disease.

    • Use may be considered in patients with mild to moderate disease who are hospitalised for a reason other than COVID-19, provided that they otherwise meet the criteria for outpatient treatment.

    • The panel currently recommends against the use of casirivimab/imdevimab, bamlanivimab/etesevimab, and sotrovimab because Omicron is the dominant variant in the US, and these monoclonal antibodies are predicted to have markedly reduced susceptibility to Omicron and its subvariants. These monoclonal antibodies are not currently authorised for use in the US.[753][754]

    • The Infectious Diseases Society of America supports the use of monoclonal antibodies with activity against the predominant regional variants within 7 days of symptom onset in ambulatory patients with mild to moderate disease who are at high risk for progression to severe disease.[460]

  • The World Health Organization strongly recommends against the use of sotrovimab and casirivimab/imdevimab for patients with non-severe disease, as in vitro data demonstrate that they do not neutralise currently circulating variants of SARS-CoV-2 and their subvariants. The agency makes no recommendations for other monoclonal antibodies, and recommends the use of antivirals instead (see above).[726][728][729] 

  • Evidence for the use of monoclonal antibodies in non-hospitalised patients is uncertain.

    • A Cochrane review found that the evidence is insufficient to draw meaningful conclusions about any specific monoclonal antibody, and the disease stage in which it should be used. Information on outcomes in non-hospitalised patients such as mortality, quality of life, and serious adverse events is either inconclusive or entirely lacking, although casirivimab/imdevimab, sotrovimab, bamlanivimab (alone or in combination with etesevimab), and regdanvimab may reduce the occurrence of hospital admission or death (low-certainty evidence).[755]

    • A systematic review and meta-analysis of 27 randomised controlled trials found that monoclonal antibodies had limited effects on most of the outcomes in non-hospitalised patients with the certainty of evidence ranging from very low to moderate for most outcomes. Monoclonal antibodies reduced hospitalisation, but there were no effects on mortality.[756] 

    • There is emerging evidence to support the use of tixagevimab/cilgavimab for the treatment of COVID-19.[757]

Antimicrobials

  • Consider empirical antibiotics in patients with moderate disease only if there is clinical suspicion of secondary bacterial infection.[93][459][523]

    • Start treatment as soon as possible, and refer to local guidelines for choice of regimen.

    • Do not offer an antibiotic for preventing secondary bacterial pneumonia.

  • Antibiotics may also be considered in older people (particularly those in long-term care facilities) and children <5 years of age to provide empirical antibiotic treatment for possible pneumonia.[93]

  • Advise patients to seek medical help without delay if their symptoms do not improve, or worsen rapidly or significantly.[523]

    • Reconsider whether the person has signs and symptoms of more severe disease on reassessment, and whether to refer them to hospital, other acute community support services, or palliative care services.

Monitor

  • Closely monitor patients with risk factors for severe illness, and counsel patients about signs and symptoms of deterioration or complications that require prompt urgent care (e.g., difficulty breathing, chest pain).[93][459]

  • Pulse oximetry monitoring at home is recommended in symptomatic patients with risk factors for progression to severe disease who are not hospitalised. Patient education and appropriate follow-up are required.[93]

  • If the patient is being managed in hospital, monitor patients closely for signs of clinical deterioration using medical early warning scores (e.g., National Early Warning Score 2 [NEWS2]), and respond immediately with appropriate supportive care interventions.[93] 

    • A systematic review and meta-analysis found that the NEWS2 score had moderate sensitivity and specificity in predicting the deterioration of patients with COVID-19. The score showed good discrimination in predicting the combined outcome of the need for intensive respiratory support, admission to the intensive care unit, or in-hospital mortality.[529]

Corticosteroids

  • Guidelines do not recommend systemic corticosteroids in patients with non-severe disease, unless there is another medical indication to do so, as they may increase the risk of mortality in these patients.[459][523][726]

Antithrombotic therapy

  • Guidelines recommend against the use of anticoagulants and antiplatelet therapy for the prevention of venous thromboembolism or arterial thrombosis in non-hospitalised patients without evidence of venous thromboembolism, unless the patient has other indications for therapy or is participating in a clinical trial.[459]

Highest-risk clinical groups

  • A UK advisory group has generated a list of conditions or cohorts who are at highest risk of serious illness from COVID-19 in the community, and who would therefore benefit from new treatments (e.g., antivirals, monoclonal antibodies). This list may be used when considering the use of these treatments in adults, and includes the following: Down's syndrome and other genetic disorders; solid cancers; haematological diseases and recipients of haematological stem cell transplants; renal and liver diseases; solid organ transplant recipients; immune-mediated inflammatory disorders; immune deficiencies; HIV/AIDS; and rare neurological and severe complex life-limiting neurodisability conditions.[758] Definitions may vary across other guidelines.

Patients with suspected or confirmed severe disease are at risk of rapid clinical deterioration.[93]

  • Severe disease in adults is defined as having clinical signs of pneumonia plus at least one of the following:

    • Respiratory rate >30 breaths/minute

    • Severe respiratory distress

    • SpO₂ <90% on room air

  • Severe disease in children is defined as having clinical signs of pneumonia plus at least one of the following:

    • Central cyanosis or SpO₂ <90%

    • Severe respiratory distress

    • General danger signs: inability to breastfeed or drink, lethargy or unconsciousness, or convulsions

    • Fast breathing (<2 months: ≥60 breaths per minute; 2-11 months: ≥50 breaths per minute; 1-5 years: ≥40 breaths per minute).

The median time from onset of symptoms to hospital admission is around 7 days.[40][759]

  • The estimated length of hospital stay is >10 days (mean 15 days). However, the duration depends on various factors including age, country/region, and available resources.[760]

  • Children are less likely to require hospitalisation, but, if admitted, generally only require supportive care.[32]

Location of care

  • Manage patients in an appropriate healthcare facility under the guidance of a specialist team.[93]

  • Use the Clinical Frailty Scale (CFS) to assess baseline health and inform discussions on treatment expectations when appropriate and within an individualised assessment of frailty. Clinical Frailty Scale Opens in new window Do not use the CFS for younger people, or for people with stable long-term disabilities (e.g., cerebral palsy), learning disabilities, or autism. Make an individualised assessment of frailty in these people, using clinical assessment and alternative scoring methods.[523]

    • Hospitalised frail patients are at higher risk of all-cause mortality compared with non-frail hospitalised patients, regardless of the frailty score/assessment tool used.[761]

    • A meta-analysis found that an increase in CFS was associated with an increase in mortality (each 1-point increase in CFS was associated with a 12% increase in mortality).[762]

    • Patients with a score between 4-9 had significantly increased mortality compared with those with a score of 1-3.[763]

    • However, one systematic review and meta-analysis found that there was no difference in short-term mortality between frail and non-frail patients.[764] Some studies suggest that a more nuanced understanding of frailty and outcomes is needed, and you should exercise caution in placing too much emphasis on the influence of frailty alone when discussing prognosis in older people.[765]

Oxygen

  • Start supplemental oxygen therapy immediately in any patient with emergency signs (i.e., obstructed or absent breathing, severe respiratory distress, central cyanosis, shock, coma and/or convulsions), or any patient without emergency signs and SpO₂ <90%.[93][459]

    • There is no evidence of benefit for oxygen therapy in patients with COVID-19 in the absence of hypoxaemia.[766]

  • Target SpO₂ to ≥94% during resuscitation in adults and children with emergency signs who require emergency airway management and oxygen therapy. Once the patient is stable, a target SpO₂ >90% in children and non-pregnant adults, and ≥92% to 95% in pregnant women, is recommended. Nasal prongs or a nasal cannula are preferred in young children.[93]

    • Some guidelines recommend that SpO₂ should be maintained no higher than 96%.[767]

    • Some centres may recommend different SpO₂ targets in order to support prioritisation of oxygen flow for the most severely ill patients in hospital.

  • Consider positioning techniques (e.g., high supported sitting), and airway clearance management to optimise oxygenation assist with secretion clearance in adults. Consider awake prone positioning (for 8-12 hours/day, broken into shorter periods over the day) in severely ill patients who require supplemental oxygen.[93][459]

    • Awake prone positioning of non-intubated patients was associated with improvement in oxygen variables (PaO₂/FiO₂, PaO₂, and SpO₂), respiratory rate, rate of intubation, and mortality. However, evidence is limited.[768][769][770][771]However, one small non-randomised controlled trial found that prone positioning offered no clinical benefit among patients with hypoxaemia who are not on mechanical ventilation, with evidence of worsening clinical outcomes at day 5.[772]

  • Monitor patients closely for signs of progressive acute hypoxaemic respiratory failure. Patients who continue to deteriorate despite standard oxygen therapy require advanced oxygen/ventilatory support.[93][459]

  • The World Health Organization recommends HFNO, continuous positive airway pressure [CPAP], or non-invasive ventilation (helmet or face mask interface) in hospitalised patients with severe disease and acute hypoxaemic respiratory failure not needing emergent intubation, rather than standard oxygen therapy. Choice depends on factors such as availability of devices and the supply of oxygen, personal comfort and experience, and patient-specific considerations (e.g., claustrophobia with CPAP or non-invasive ventilation masks, nasal discomfort with HFNO).[93]

Symptom management and supportive care

  • Fluids and electrolytes: use cautious fluid management in adults and children without tissue hypoperfusion and fluid responsiveness as aggressive fluid resuscitation may worsen oxygenation.[93] Correct any electrolyte or metabolic abnormalities, such as hyperglycaemia or metabolic acidosis, according to local protocols.[773]

  • Fever and pain: paracetamol or ibuprofen are recommended.[93][523] Ibuprofen should only be taken at the lowest effective dose for the shortest period needed to control symptoms.

  • Cough: short-term use of a cough suppressant may be considered in select patients (e.g., if the cough is distressing to the patient) provided there are no contraindications.[523]

  • Breathlessness: keep the room cool, and encourage relaxation, breathing techniques, and changing body positions. Identify and treat any reversible causes of breathlessness (e.g., pulmonary oedema, pulmonary embolism, COPD, asthma).[523]

  • Anxiety, delirium, and agitation: identify and treat any underlying or reversible causes (e.g., offer reassurance, treat hypoxia, correct metabolic or endocrine abnormalities, address co-infections, minimise use of drugs that may cause or worsen delirium, treat substance withdrawal, maintain normal sleep cycles, treat pain or breathlessness).[93][523]

    • Low doses of haloperidol (or another suitable antipsychotic) can be considered for agitation.[93]

    • Non-pharmacological interventions are the mainstay for the management of delirium when possible, and prevention is key.[774]

  • Mouth care: an important part of overall patient care in hospitalised patients who are ventilated or non-ventilated and those undergoing step-down or end-of-life care.[775]

  • Provide basic mental health and psychosocial support for all patients, and manage any symptoms of insomnia or depression as appropriate.[93]

Venous thromboembolism prophylaxis

  • Assess the risk of bleeding as soon as possible after admission, or by the time of the first consultant review, using a suitable risk assessment tool.[523]

  • Start venous thromboembolism (VTE) prophylaxis in acutely ill hospitalised adults and adolescents, provided there are no contraindications.[93][459][776][777]

    • In the UK, the National Institute for Health and Care Excellence recommends starting as soon as possible (within 14 hours of admission) in young people and adults who need low-flow oxygen and who do not have an increased bleeding risk, and continuing for a minimum of 7 days including after discharge.[523]

    • In the US, the National Institutes of Health guidelines panel recommends prophylactic anticoagulation in all children ≥12 years of age, unless contraindicated.[459]

  • Low molecular weight heparin, unfractionated heparin, or fondaparinux are the recommended options for standard thromboprophylaxis.[93]

    • In the UK, the National Institute for Health and Care Excellence recommends low molecular weight heparin first-line, with fondaparinux or unfractionated heparin reserved for patients who cannot have low molecular weight heparin.[523]

    • In the US, the National Institutes of Health guidelines panel recommends parenteral over oral anticoagulants and, when heparin is used, low molecular weight heparin is preferred over unfractionated heparin. The panel recommends against the use of therapeutic-dose oral anticoagulants, except in the context of a clinical trial.[459]

    • Unfractionated heparin is contraindicated in patients with severe thrombocytopenia and patients with a history of heparin-induced thrombocytopenia. Fondaparinux is recommended in patients with a history of heparin-induced thrombocytopenia. Mechanical thromboprophylaxis (e.g., intermittent pneumatic compression devices) is recommended if anticoagulation is contraindicated or not available.[777][778]

    • A meta-analysis found that low molecular weight heparin was associated with better outcomes (i.e., decreased intensive care unit admission, mechanical ventilation, hospital stay, and mortality) compared with unfractionated heparin in hospitalised patients. However, there was no difference in the incidence of bleeding between the two.[779]

    • Avoid direct oral anticoagulants in the absence of an evidence-based indication for oral anticoagulation. An open-label, multicentre, randomised controlled trial found that in-hospital therapeutic anticoagulation with rivaroxaban or enoxaparin followed by rivaroxaban until day 30 did not improve clinical outcomes and increased bleeding compared with prophylactic anticoagulation among hospitalised patients with an elevated D-dimer level.[780]

  • The optimal dose is yet to be determined. Standard prophylaxis doses are generally recommended across most guidelines over intermediate- or full treatment-dose regimens in patients without an established indication for higher-dose anticoagulation.[781] However, this recommendation varies and you should consult your local guidelines.

    • The World Health Organization recommends standard thromboprophylaxis dosing of anticoagulation rather than therapeutic or intermediate dosing in patients without an established indication for higher-dose anticoagulation.[93]

    • In the UK, the National Institute for Health and Care Excellence recommends a prophylactic dose of a low molecular weight heparin for a minimum of 7 days (including after discharge) in young people and adults who need low-flow oxygen and who do not have an increased bleeding risk. A treatment dose of a low molecular weight heparin for 14 days or until discharge (whichever is sooner) may be considered in young people and adults who need low-flow oxygen and who do not have an increased bleeding risk; however, this is a conditional recommendation only. The decision should be carefully considered, and choice of the most appropriate dose regimen should be guided by bleeding risk, clinical judgement, and local protocols. For those who do not need supplemental oxygen, follow standard VTE prophylaxis guidelines.[523]

    • In the US, the National Institutes of Health guidelines panel recommends therapeutic-dose heparin for patients who have a D-dimer level above the upper limit of normal, require low-flow oxygen, and have no increased bleeding risk, unless a contraindication exists. Treatment should continue for 14 days or until hospital discharge, whichever comes first. The panel recommends using prophylactic-dose heparin for patients who are not administered therapeutic heparin, unless a contraindication exists.[459]

    • Dose adjustments may be required in patients with extremes of body weight or renal impairment.[523]

  • Evidence supports the use of lower-dose anticoagulant regimens.

    • A Cochrane review found that higher-dose regimens resulted in little to no difference in all-cause mortality compared with lower-dose regimens in hospitalised patients; however, higher-dose regimens were associated with an increased risk of minor bleeding up to 30 days (high-certainty evidence). Higher-dose anticoagulants probably reduce pulmonary embolism and slightly increase major bleeding compared with lower-dose regimens up to 30 days (moderate-certainty evidence). Higher-dose anticoagulants may result in little or no difference in deep vein thrombosis, stroke, major adverse limb events, myocardial infarction, atrial fibrillation, or thrombocytopenia compared with lower-dose regimens up to 30 days (low-certainty evidence). Anticoagulants may reduce all‐cause mortality compared with no anticoagulants, but the evidence is very uncertain.[782]

  • For patients who are already on an anticoagulant for another underlying condition, continue the patient’s current medication and therapeutic dose unless contraindicated by a change in clinical circumstances.[459][523] Consider switching to low molecular weight heparin if the patient’s clinical condition is deteriorating and the patient is not currently on low molecular weight heparin.[523]

    • A systematic review and meta-analysis found that the use of oral anticoagulation prior to hospital admission appeared to be ineffective at reducing the risk of intensive care unit admission and mortality; however, the review acknowledged that further trials are needed.[783]

    • A population-based cohort study found that people with atrial fibrillation and a low baseline risk of stroke who were taking oral anticoagulation had a marginally lower risk of COVID-19-related mortality compared with people not taking oral anticoagulation. However, it was not clear whether the association was causal or due to other differences between the groups.[784]

  • Monitor patients for signs and symptoms suggestive of thromboembolism and proceed with appropriate diagnostic and management pathways if clinically suspected.[93]

    • If the patient’s clinical condition changes, assess the risk of VTE, reassess the bleeding risk, and review VTE prophylaxis.[523]

  • Continue anticoagulation until hospital discharge.[93] Routine post-discharge VTE prophylaxis is not generally recommended, except in certain high-risk patients, in the context of a clinical trial, or if another indication for VTE prophylaxis exists.[459][776][777] Ensure patients who require VTE prophylaxis after discharge are able to use it correctly or have arrangements made for someone to help them.[523]

    • A cohort study of nearly 3000 patients found that patients who had a history of venous thromboembolism, peak D-dimer >3 micrograms/mL, and pre-discharge C-reactive protein >10 mg/dL were at high risk of experiencing new-onset venous thromboembolism post discharge, and these patients may benefit from post-discharge anticoagulation.[785]

    • A randomised controlled trial found that rivaroxaban for 35 days after hospital discharge improved clinical outcomes (reduction in thrombotic events) in high-risk patients compared with no extended thromboprophylaxis; however, further research is required.[786]

  • There is currently insufficient evidence to determine the risks and benefits of prophylactic anticoagulation in hospitalised patients with COVID-19.[787]

    • A systematic review and meta-analysis found that the pooled odds of mortality between anticoagulated and non-anticoagulated hospitalised patients were similar, but lower in the standard prophylactic-dose group. Prophylactic-dose anticoagulation significantly decreased the odds of in-hospital death by 17% compared with no anticoagulation. Mortality increased in the intermediate- to therapeutic-dose group with an increased risk of major bleeding.[788]

    • Clinicians should rely on pre-COVID-19 evidence-based principles of anticoagulation management combined with rational approaches to address clinical challenges.[776] VTE prophylaxis in patients with COVID-19 and cancer is similar to those with cancer and without COVID-19.[789]

Antimicrobials

  • Do not offer antibiotics for preventing or treating pneumonia if SARS-CoV-2, another virus, or a fungal infection is likely to be the cause.[523] Guidelines recommend against empirical broad-spectrum antibiotics in the absence of a proven or suspected bacterial infection.[459]

  • Consider empirical antibiotics if there is clinical suspicion of secondary bacterial infection. Give within 1 hour of initial assessment for patients with suspected sepsis or if the patient meets high-risk criteria (or within 4 hours of establishing a diagnosis of secondary bacterial pneumonia); do not wait for microbiology results. Base the regimen on the clinical diagnosis (e.g., community-acquired pneumonia, hospital-acquired pneumonia, sepsis), local epidemiology and susceptibility data, and local treatment guidelines.[93][459][523]

  • Consider seeking specialist advice for people who: are immunocompromised; have a history of infection with resistant organisms; have a history of repeated infective exacerbations of lung disease; are pregnant; or are receiving advanced respiratory or organ support. Seek specialist advice if there is a suspicion that the person has an infection with multidrug-resistant bacteria and may need a different antibiotic, or there is clinical or microbiological evidence of infection and the person's condition does not improve as expected after 48 to 72 hours of antibiotic treatment.[523]

  • Reassess antibiotic use daily. De-escalate empirical therapy on the basis of microbiology results and clinical judgement. Regularly review the possibility of switching from intravenous to oral therapy. Duration of treatment should be as short as possible (e.g., 5 to 7 days). Antibiotic stewardship programmes should be in place.[93]

    • A meta-analysis found that the prevalence of antibiotic prescribing in patients with COVID-19 was 75%, which is significantly higher than the estimated prevalence of bacterial co-infection. Therefore, unnecessary antibiotic use is likely to be high in these patients.[790]

  • Treat laboratory-confirmed co-infections (e.g., malaria, tuberculosis, influenza) as appropriate according to local protocols.[93] The treatment of influenza is the same in all patients regardless of SARS-CoV-2 co-infection. Start empirical treatment with oseltamivir in hospitalised patients who are suspected of having either or both infections as soon as possible without waiting for influenza test results. Antiviral therapy can be stopped once influenza has been ruled out.[459]

Corticosteroids

  • The WHO strongly recommends systemic corticosteroid therapy (low-dose intravenous or oral dexamethasone or hydrocortisone) for 7 to 10 days in adults with severe disease.[726][728][729]

    • It is unclear whether these recommendations can be applied to children or those who are immunocompromised.

    • This recommendation is based on moderate-quality evidence that suggests systemic corticosteroids probably reduce 28-day mortality in patients with severe disease. There is no evidence directly comparing dexamethasone and hydrocortisone. The harms of treatment in this context are considered to be minor.

  • In the UK, the National Institute for Health and Care Excellence recommends offering dexamethasone (or an alternative such as hydrocortisone or prednisolone when dexamethasone cannot be used or is unavailable) to people who need supplemental oxygen to meet their prescribed oxygen saturation levels, or who have a level of hypoxia that needs supplemental oxygen but who are unable to have or tolerate it.

    • Treatment is for up to 10 days unless there is a clear indication to stop early.[523]

  • In the US, the National Institutes of Health guidelines panel recommends dexamethasone, either alone or in combination with remdesivir, in hospitalised adults who require supplemental oxygen.

    • Alternative corticosteroids may be used in situations where dexamethasone is not available.

    • It is not routinely recommended for paediatric patients who require only low levels of oxygen support (i.e., via a nasal cannula only). Use of dexamethasone for the treatment of severe disease in children who are profoundly immunocompromised has not been evaluated, may be harmful, and therefore should be considered only on a case-by-case basis.[459]

    • The Infectious Diseases Society of America supports the use of dexamethasone in hospitalised patients with severe disease.[460]

  • Evidence supports the use of corticosteroids in hospitalised patients.

    • A Cochrane review found that systemic corticosteroids probably slightly reduce all-cause mortality in hospitalised patients with symptomatic disease (moderate-certainty evidence). Most participants in the studies were treated with non-invasive or invasive mechanical ventilation. Low-certainty evidence suggests that there may also be a reduction in ventilator-free days; however, the current evidence remains uncertain due to methodological limitations. Evidence of an increased risk of mortality in symptomatic hospitalised patients without any need for additional oxygen was limited by a lack of statistical significance. It is unknown which systemic corticosteroid is most effective.[791]

    • A living systematic review and network meta-analysis found that corticosteroids probably reduce mortality compared with standard care.[792][793]

  • Monitor patients for adverse effects (e.g., hyperglycaemia, secondary infections, psychiatric effects, reactivation of latent infections) and assess for drug-drug interactions.[459]

Antivirals

  • The World Health Organization conditionally recommends the intravenous antiviral remdesivir in adults with severe disease.[726][728][729]

    • Remdesivir should be administered as soon as possible after onset of symptoms. The treatment course for this indication is 5 to 10 days.

    • There is insufficient evidence to make a recommendation around use in children, and the trials did not include pregnant or breastfeeding women.

    • This recommendation is based on low-certainty evidence that suggests remdesivir possibly reduces mortality, and moderate-certainty evidence that suggests it probably reduces the need for mechanical ventilation. Moderate-certainty evidence suggests that remdesivir probably has little or no impact on time to symptom improvement.

  • In the UK, the National Institute for Health and Care Excellence recommends considering remdesivir in hospitalised adults and children ≥12 years of age (weighing ≥40 kg) who require low-flow supplemental oxygen.[523]

    • This recommendation is based on moderate-certainty evidence that suggests remdesivir probably reduces the risk of mortality in hospitalised patients who need low-flow supplemental oxygen. This is likely because it is being given early in the disease course.

  • In the US, the National Institutes of Health guidelines panel recommends remdesivir in hospitalised children and adults who require supplemental oxygen. It may be given alone (e.g., for patients who require minimal supplemental oxygen) or in combination with dexamethasone (e.g., for patients who require increasing amounts of supplemental oxygen).[459]

    • The panel also recommends remdesivir alone in hospitalised children aged 12 to 17 years who have risk factors for severe disease but do not require supplemental oxygen.

    • The Infectious Diseases Society of America supports the use of remdesivir in hospitalised patients with severe disease who require oxygen.[460]

  • The recommended treatment course is 5 to 10 days or until hospital discharge, whichever comes first.[459][726]

    • Evidence does not suggest any greater benefit with a 10-day course of remdesivir compared with a 5-day course, but suggests an increased risk of harm.[523] However, some experts may recommend a 10-day course in patients who have not shown substantial clinical improvement by day 5.[459]

    • There may be no benefit in completing the full course of remdesivir if the patient progresses.[523] However, US guidelines recommend completing the full treatment course if the patient progresses to requiring high-flow oxygen, non-invasive ventilation, mechanical ventilation, or extracorporeal membrane oxygenation.[459]

  • Despite guidelines recommending the use of remdesivir in patients with severe disease, evidence for its use is conflicting.

    • A Cochrane review found that remdesivir probably has little or no effect on 28-day all-cause mortality in hospitalised patients compared with placebo or usual care (moderate certainty). Effects on clinical improvement or worsening were uncertain. There were insufficient data available to examine the effect of remdesivir on mortality across subgroups defined by respiratory support at baseline.[794]

Interleukin-6 (IL-6) inhibitors

  • The WHO strongly recommends an IL-6 inhibitor (tocilizumab or sarilumab) in patients with severe disease.[726][728][729]

    • IL-6 inhibitors are typically administered as a single intravenous dose; however, a second dose may be administered 12 to 48 hours after the first dose if the clinical response is inadequate.

    • IL-6 inhibitors may be administered in combination with corticosteroids and Janus kinase inhibitors, and should be initiated at the same time as systemic corticosteroids.

    • The applicability of this recommendation to children is currently uncertain.

    • This recommendation is based on high-certainty evidence that shows IL-6 inhibitors reduce mortality and the need for mechanical ventilation, and low-certainty evidence that suggests that IL-6 inhibitors may also reduce the duration of mechanical ventilation and hospitalisation. The evidence regarding the risk of severe adverse events is uncertain.

  • In the UK, the National Institute for Health and Care Excellence recommends a single dose of tocilizumab in hospitalised adults.[523]

    • Patients must meet the following conditions: they are having or have completed a course of corticosteroids such as dexamethasone (unless they cannot have corticosteroids); they have not had another IL-6 inhibitor during this admission; there is no evidence of a bacterial or viral infection (other than SARS-CoV-2) that might be worsened by tocilizumab; AND they either need supplemental oxygen and have a C-reactive protein level of ≥75 mg/L, OR they are within 48 hours of starting HFNO, continuous positive airway pressure, non-invasive ventilation, or invasive mechanical ventilation.

    • Consider tocilizumab for children and young people who have severe disease or paediatric inflammatory multisystem syndrome only if they are aged 1 year and over, and only in the context of a clinical trial.

    • Sarilumab may be considered an alternative option in adults only if tocilizumab cannot be used or is unavailable (use the same eligibility criteria as those for tocilizumab).

  • In the US, the National Institutes of Health guidelines panel recommends tocilizumab (or sarilumab if tocilizumab is not available or not feasible to use) for patients on a corticosteroid with rapidly increasing oxygen needs and systemic inflammation.[459]

    • The Infectious Diseases Society of America recommends considering tocilizumab in hospitalised adults with progressive severe disease who have elevated markers of systemic inflammation, in addition to standard of care (i.e., corticosteroids), rather than standard of care alone. Sarilumab may be used if tocilizumab is not available.[460]

  • Evidence supports the use of IL-6 inhibitors.

    • A Cochrane review found that tocilizumab reduced all-cause mortality at day 28 (high-certainty evidence), and probably resulted in slightly fewer serious adverse events (moderate-certainty evidence) compared with standard care alone or placebo. The evidence suggests uncertainty around the effect on mortality after day 60. However, tocilizumab probably results in little or no increase in clinical improvement at day 28 (i.e., hospital discharge or improvement measured by triallist-defined scales). The impact of tocilizumab on other outcomes is uncertain. Evidence for an effect of sarilumab is uncertain.[795]

    • A living systematic review and network meta-analysis found that IL-6 inhibitors (with corticosteroids) probably reduce mortality (moderate-certainty evidence), are likely to reduce the need for mechanical ventilation (moderate-certainty evidence), and may reduce the duration of hospitalisation (moderate-certainty evidence) compared with standard care.[792][793]

    • A meta-analysis of approximately 20,000 patients from 45 randomised trials found that tocilizumab (in combination with corticosteroids) probably reduces mortality in people with severe or critical disease, and sarilumab (in combination with corticosteroids) could reduce mortality. The available evidence suggests that tocilizumab and sarilumab could be similarly effective. These drugs may not be beneficial when used without corticosteroids.[796]

Janus kinase (JAK) inhibitors

  • The WHO strongly recommends a JAK inhibitor (baricitinib) in patients with severe disease.[726][728][729]

    • JAK inhibitors are administered orally. The treatment duration is 14 days or until hospital discharge, whichever is first.

    • Baricitinib may be administered in combination with corticosteroids and IL-6 inhibitors, and should be initiated at the same time as systemic corticosteroids.

    • The applicability of this recommendation to children is currently uncertain.

    • This recommendation is based on high-certainty evidence that baricitinib reduces mortality, and moderate-certainty evidence that baricitinib probably reduces the duration of mechanical ventilation and the length of hospital stay.

    • Other drugs in this class include tofacitinib and ruxolitinib. The WHO recommends against using these drugs unless baricitinib or IL-6 inhibitors are not available as the effects of tofacitinib or ruxolitinib on mortality, need for mechanical ventilation, and hospital length of stay remain uncertain and more trial evidence is needed.

  • In the UK, the National Institute for Health and Care Excellence recommends baricitinib in hospitalised adults who: need supplemental oxygen, and are having or have completed a course of corticosteroids (unless contraindicated), and have no evidence of infection (other than SARS-CoV-2) that might be worsened by baricitinib.[523]

    • Baricitinib may also be considered in children ≥2 years of age provided they meet the same criteria.

  • In the US, the National Institutes of Health guidelines panel recommends baricitinib (or tofacitinib if baricitinib is not available or not feasible to use) in patients on a corticosteroid with rapidly increasing oxygen needs and systemic inflammation.[459]

    • The Infectious Diseases Society of America suggests baricitinib (in combination with a corticosteroid) in hospitalised adults with severe disease. The guideline panel suggests baricitinib with remdesivir, rather than remdesivir alone, in patients who cannot receive a corticosteroid because of a contraindication. The panel suggests tofacitinib in hospitalised adults with severe disease who are not on non-invasive or invasive mechanical ventilation.[460]

  • Evidence supports the use of JAK inhibitors.

    • A Cochrane review found that JAK inhibitors probably reduced all-cause mortality up to day 28 (moderate-certainty evidence) and up to day 60 (high-certainty evidence). They probably make little or no difference in improvement in clinical status or the rate of adverse events (moderate-certainty evidence). Baricitinib was the most often evaluated JAK inhibitor.[797]

    • A living systematic review and network meta-analysis found that JAK inhibitors probably reduce mortality (high-certainty evidence), reduce the duration of mechanical ventilation (high-certainty evidence), and reduce length of hospital stay (high-certainty evidence) compared with standard care.[792][793]

Monoclonal antibodies

  • Recommendations for monoclonal antibodies in patients with severe disease differ from the recommendations for patients with mild to moderate disease. Key international guidelines do not currently recommend monoclonal antibodies for patients with severe disease.

    • The World Health Organization strongly recommends against the use of casirivimab/imdevimab for patients with any disease severity. The agency makes no other recommendations either for or against the use of other monoclonal antibodies in patients with severe disease.[726][728][729]

    • In the UK, the National Institute for Health and Care Excellence recommends not offering casirivimab/imdevimab to patients who are known or suspected to have infection caused by an Omicron variant.[523] 

    • In the US, the National Institutes of Health guidelines panel states that monoclonal antibodies are not currently authorised for use in hospitalised patients with severe disease. However, they may be available through expanded access programmes for patients who are hospitalised with severe disease and who are immunocompromised.[459]

  • Evidence for the use of monoclonal antibodies in hospitalised patients is uncertain.

    • A Cochrane review found that casirivimab/imdevimab probably has no effect on mortality, progression to invasive mechanical ventilation, and 30-day hospital discharge in hospitalised patients (moderate-certainty evidence). Bamlanivimab may have little to no effect on efficacy outcomes when compared with placebo, but it may increase the occurrence of severe symptoms and adverse events (low-certainty evidence).[755]

    • A systematic review and meta-analysis of 27 randomised controlled trials found that monoclonal antibodies had limited effects on most of the outcomes in hospitalised patients with the certainty of evidence ranging from very low to moderate for most outcomes. Monoclonal antibodies slightly reduced mechanical ventilation and bacteraemia, but there were no effects on mortality.[756] 

Monitor

  • Monitor patients closely for signs of clinical deterioration, and respond immediately with appropriate supportive care interventions.[93]

Discharge and rehabilitation

  • Routinely assess older patients for mobility, functional swallow, cognitive impairment, and mental health concerns, and based on that assessment determine whether the patient is ready for discharge, and whether the patient has any rehabilitation and follow-up requirements.[93]

Palliative care

  • Palliative care interventions should be made accessible at each institution that provides care for patients with COVID-19. Identify whether the patient has an advance care plan and respect the patient’s priorities and preferences when formulating the patient’s care plan.[93]

    • There is a lack of data on palliative care in patients with COVID-19. However, a rapid systematic review of pharmacological strategies used for palliative care in these patients, the first international review of its kind, found that a higher proportion of patients required continuous subcutaneous infusions for medication delivery than is typically seen in the palliative care population. Modest doses of commonly used end-of-life medications were required for symptom control. However, these findings should be interpreted with caution due to the lack of data available.[798]

  • Follow local palliative care guidelines.

Patients with critical disease (i.e., presence of acute respiratory distress syndrome, sepsis, or septic shock) should be admitted or transferred to an intensive/critical care unit. Use existing care bundles (i.e., three or more evidence-informed practices delivered together and consistently to improve care), chosen locally by the hospital or intensive care unit and adapted as necessary for local circumstances.[93]

  • The rate of intensive care admission varies between studies; however, a meta-analysis of nearly 25,000 patients found that the admission rate was 32%, and the pooled prevalence of mortality in patients in the intensive care unit was 35% to 39%.[799][800]

  • The most common reasons for intensive care unit admission were hypoxaemic respiratory failure leading to mechanical ventilation and hypotension.[801]

  • Patients admitted to intensive care units were older, were predominantly male, and had a median length of stay of 23 days (range 12-32 days).[802]

  • The strongest risk factors for critical illness were oxygen saturation <88%; elevated serum troponin, C-reactive protein, and D-dimer; and, to a lesser extent, older age, body mass index >40, heart failure, and male sex.[803] The most common risk factors for intensive care unit mortality were invasive mechanical ventilation, acute kidney injury, and acute respiratory distress syndrome.[804]

  • Risk factors for intensive care admission in children include age <1 month, male sex, pre-existing medical conditions, and presence of lower respiratory tract infection signs or symptoms at presentation.[805] The majority of children who required ventilation had underlying comorbidities, most commonly cardiac disease.[806]

Location of care

  • Manage patients in an intensive/critical care unit under the guidance of a specialist team.[93]

  • Discuss the risks, benefits, and potential outcomes of treatment options with patients and their families, and allow them to express preferences about their management. Take their wishes and expectations into account when considering the ceiling of treatment. Use decision support tools if available. Put treatment escalation plans in place, and discuss any existing advance care plans or advance decisions to refuse treatment with patients who have pre-existing advanced comorbidities.[523]

HFNO or non-invasive ventilation

  • The World Health Organization recommends HFNO, CPAP, or non-invasive ventilation (helmet or face mask interface) in hospitalised patients with critical disease and acute hypoxaemic respiratory failure not needing emergent intubation, rather than standard oxygen therapy.[93]

    • Choice depends on factors such as availability of devices and the supply of oxygen, personal comfort and experience, and patient-specific considerations (e.g., claustrophobia with CPAP or non-invasive ventilation masks, nasal discomfort with HFNO).

    • Consider awake prone positioning (for 8-12 hours/day, broken into shorter periods over the day) in severely ill patients who require HFNO or non-invasive ventilation.

  • In the UK, the National Institute for Health and Care Excellence recommends CPAP in patients with hypoxaemia that is not responding to supplemental oxygen with a fraction of inspired oxygen of ≥0.4 (40%), and escalation to invasive mechanical ventilation would be an option but it is not immediately needed or it is agreed that respiratory support should not be escalated beyond CPAP.[523]

    • Ensure there is access to critical care providers for advice, regular review, and prompt escalation of treatment if needed, and regular assessment and management of symptoms alongside non-invasive respiratory support.

    • Consider using HFNO for people when: they cannot tolerate CPAP but need humidified oxygen at high flow rates; maximal conventional oxygen is not maintaining their target oxygen saturations and they do not need immediate invasive mechanical ventilation or escalation to invasive mechanical ventilation is not suitable, and CPAP is not suitable; or they need a break from CPAP (e.g., mealtimes, skin pressure relief, mouth care), need humidified oxygen or nebulisers (or both), or need weaning from CPAP.

    • Do not routinely offer HFNO as the main form of respiratory support for people with respiratory failure in whom escalation to invasive mechanical ventilation would be appropriate.

    • Optimise pharmacological and non-pharmacological management strategies in people who need non-invasive respiratory support.

    • Consider awake prone positioning for hospitalised patients who are not intubated and have higher oxygen needs.

  • In the US, the National Institutes of Health guidelines panel recommends HFNO over non-invasive ventilation in adults with acute hypoxaemic respiratory failure despite conventional oxygen therapy.[459]

    • The panel recommends a closely monitored trial of non-invasive ventilation in adults if HFNO is not available. A trial of awake prone positioning is recommended in adults with persistent hypoxaemia who require HFNO and for whom endotracheal intubation is not indicated.

    • A time-limited trial of either non-invasive ventilation or HFNO is recommended in infants and children with persistent respiratory failure despite conventional oxygen therapy who have no indicators for endotracheal intubation. There is insufficient evidence to recommend either for or against a trial of awake prone positioning in children.

    • The panel recommends against using awake prone positioning as a rescue therapy for refractory hypoxaemia to avoid intubation in patients who otherwise meet the indications for intubation and invasive mechanical ventilation.

  • Evidence for non-invasive ventilation is limited.

    • There is no certain evidence that non-invasive respiratory support increases or decreases mortality in patients with COVID-19 acute respiratory failure.[807]

    • Limited evidence suggests that non-invasive ventilation reduces the need for intubation, improves resource utilisation, may be associated with better outcomes, and is safe.[808]

    • Indirect and low-certainty evidence suggests that non-invasive ventilation probably reduces mortality, similar to invasive mechanical ventilation, but may increase the risk of viral transmission. HFNO may reduce mortality compared with no HFNO.[809][810]

    • HFNO was superior to non-invasive ventilation for acute respiratory failure in terms of decreasing mortality. However, there was no significant difference in intubation rates and length of hospital stay between the two groups.[811][812]

    • The RECOVERY-RS trial (an open-label, multicenter, adaptive randomised controlled trial) found that CPAP reduced the need for invasive mechanical ventilation in adults admitted to hospital with acute respiratory failure. Neither CPAP nor HFNO reduced mortality when compared with conventional oxygen therapy.[813]

    • The HELMET-COVID trial (a multicentre randomised clinical trial) found that helmet non-invasive ventilation did not significantly reduce 28-day mortality compared with usual respiratory support (alternate use of mask non-invasive ventilation, HFNO, or standard oxygen according to clinical response) among patients with acute hypoxaemic respiratory failure. However, there were several important limitations to the study, and interpretation of the findings is limited by imprecision in the effect size estimate.[814]

    • The SOHO-COVID trial (a randomised clinical trial) found that HFNO did not significantly reduce 28-day mortality compared with standard oxygen therapy among patients with respiratory failure.[815] However, another randomised controlled trial found that treatment with HFNO reduced the likelihood of invasive mechanical ventilation and decreased the time to clinical recovery compared with conventional low-flow oxygen therapy in patients with severe disease.[816]

    • Awake prone positioning of non-intubated patients was associated with improvement in oxygen variables (PaO₂/FiO₂, PaO₂, and SpO₂), respiratory rate, rate of intubation (particularly among those who required advanced respiratory support and those in intensive care unit settings), and mortality. However, evidence is limited.[768][769][770][771]

  • Airborne precautions are recommended for these interventions (including bubble CPAP) due to uncertainty about the potential for aerosolisation.[93]

    • CPAP and HFNO do not appear to be associated with significant additional air or surface viral contamination compared with supplemental oxygen.[817]

  • Patients with hypercapnia, haemodynamic instability, multi-organ failure, or abnormal mental status should generally not receive HFNO, although emerging data suggests that it may be safe in patients with mild to moderate and non-worsening hypercapnia. Patients with hypoxaemic respiratory failure and haemodynamic instability, multi-organ failure, or abnormal mental status should not receive these treatments in place of other options such as invasive ventilation.[93]

  • Monitor patients closely for acute deterioration. If patients do not improve after a short trial of these interventions, they require urgent endotracheal intubation.[93][767]

  • More detailed guidance on the management of ARDS in COVID-19 is beyond the scope of this topic; consult a specialist for further guidance.

Mechanical ventilation

  • Consider endotracheal intubation and invasive mechanical ventilation in patients who are acutely deteriorating despite advanced oxygen/non-invasive ventilatory support measures.[93][459]

    • Use of mechanical ventilation in COVID-19 patients carries a high risk of mortality. Mortality is highly variable across studies, ranging between 21% and 100%. An overall in-hospital mortality risk ratio of 0.70 has been reported based on random-effect pooled estimates. However, it is important to note that outcomes appear to have improved as the pandemic has progressed.[818]

    • Early intubation may be associated with lower all-cause mortality compared with patients undergoing late intubation.[819]

  • Endotracheal intubation should be performed by an experienced provider using airborne precautions.[93] Intubation by video laryngoscopy is recommended if possible.[459] Young children, or adults who are obese or pregnant, may desaturate quickly during intubation and therefore require pre-oxygenation with 100% FiO₂ for 5 minutes.[93] Cuffed endotracheal tubes are preferred over uncuffed endotracheal tubes in children.[459]

  • Mechanically ventilated patients with ARDS should receive a lung-protective, low tidal volume/low inspiratory pressure ventilation strategy (lower targets are recommended in children). A higher positive end-expiratory pressure (PEEP) strategy is preferred over a lower PEEP strategy in moderate to severe ARDS. However, individualisation of PEEP, where the patient is monitored for beneficial or harmful effects and driving pressure during titration with consideration of the risks and benefits of PEEP titration, is recommended.[93][459][767]

    • Although some patients with COVID-19 pneumonia meet the criteria for ARDS, there is some discussion about whether COVID-19 pneumonia is its own specific disease with atypical phenotypes. Anecdotal evidence suggests that the main characteristic of the atypical presentation is the dissociation between well-preserved lung mechanics and the severity of hypoxaemia.[820][821][822][823][824][825] However, this approach has been criticised.[826][827] Results from three large observational cohort studies with data from critically ill patients with acute respiratory failure found that COVID-19-related ARDS had no consistent respiratory subphenotype at baseline (start of invasive ventilation). However, time-dependent analysis showed that two subphenotypes developed during the first 4 days of mechanical ventilation. Patients with an upward trajectory of ventilatory ratio had a higher risk of venous thrombotic events, more frequently developed acute kidney injury, required longer invasive mechanical ventilation, and had higher mortality.[828]

    • It has been argued that an evidence-based approach extrapolating data from ARDS not related to COVID-19 is the most reasonable approach for intensive care of COVID-19 patients.[829] As a consequence of this, some clinicians have warned that protocol-driven ventilator use may be causing lung injury in some patients, and that ventilator settings should be based on physiological findings rather than using standard protocols. High PEEP may have a detrimental effect on patients with normal compliance.[820]

    • PEEP should always be carefully titrated.[830]

  • Consider prone ventilation in patients with severe ARDS for 12 to 16 hours per day. Pregnant women in the third trimester may benefit from being placed in the lateral decubitus position. Caution is required in children.[93][459][767] Longer durations may be feasible in some patients.[831]

  • Lung recruitment manoeuvres are suggested, but staircase recruitment manoeuvres are not recommended.[459][767]

  • More detailed guidance on the management of ARDS in COVID-19, including sedation and the use of neuromuscular blockade during ventilation, is beyond the scope of this topic; consult a specialist for further guidance.

Inhaled pulmonary vasodilator

  • Consider a trial of an inhaled pulmonary vasodilator in adults and children who have severe ARDS and refractory hypoxaemia despite optimising ventilation. Taper off if there is no rapid improvement in oxygenation.[459][767]

  • A systematic review and meta-analysis found that inhaled pulmonary vasodilators may improve oxygenation, but showed no mortality benefit, compared with standard therapy.[832]

Extracorporeal membrane oxygenation

  • Consider extracorporeal membrane oxygenation (ECMO) according to availability and expertise if the above methods fail.[93][767][833]

  • There is insufficient evidence to recommend either for or against the routine use of ECMO.[459]

    • A systematic review and meta-analysis found that in-hospital mortality in adults receiving ECMO was 39%, and the risk of mortality was higher when compared with influenza patients on ECMO (44% versus 38%).[834]

    • Another systematic review and meta-analysis found that the pooled mortality rate was 48.8%, and the rate increased as the pandemic progressed.[835]

    • A registry-based cohort study found that ECMO was associated with a 7.1% reduction in mortality in selected adults (i.e., PaO 2/FiO 2 <80 mmHg) with COVID-19-associated respiratory failure, compared with conventional mechanical ventilation without ECMO. It was most effective in patients aged <65 years and those with a PaO 2/FiO 2 <80 mmHg or with driving pressures >15 cm H 2O during the first 10 days of mechanical ventilation.[836]

    • Single-access, dual-stage venovenous ECMO with early extubation appears to be safe and effective in patients with COVID-19 respiratory failure.[837]

    • There is a risk of neurological complications (e.g., intracranial haemorrhage, ischaemic stroke, and hypoxic ischaemic brain injury) in patients on ECMO.[838]

Management of septic shock/sepsis

  • The management of sepsis and septic shock in patients with COVID-19 is beyond the scope of this topic. See  Complications

Symptom management and supportive care

  • Consider fluid and electrolyte management, antimicrobial treatment, and symptom management as appropriate (see Severe COVID-19 above).

Venous thromboembolism prophylaxis

  • Recommendations for patients with critical disease may differ from those for severe disease (see above). Consult your local guidelines.

    • In the UK, the National Institute for Health and Care Excellence recommends a prophylactic dose of a low molecular weight heparin to young people and adults who need HFNO, CPAP, non-invasive ventilation, or invasive mechanical ventilation, and who do not have an increased bleeding risk. An intermediate or treatment dose of a low molecular weight heparin is only recommended in these patients as part of a clinical trial.[523]

    • In the US, the National Institutes of Health guidelines panel recommends prophylactic-dose heparin (low molecular weight heparin preferred over unfractionated heparin) for patients who are receiving intensive care unit level of care (including patients receiving high-flow oxygen), unless there is a contraindication. The panel recommends against the use of intermediate-dose and therapeutic-dose anticoagulation in these patients, except in the context of a clinical trial. Patients who start on therapeutic-dose heparin while in a non-intensive care unit setting and then transfer to the intensive care unit should be switched from therapeutic to prophylactic-dose heparin unless venous thromboembolism is confirmed. There is insufficient evidence for the panel to recommend either for or against antiplatelet therapy in critically ill patients.[459]

    • Some guidelines recommend that escalated doses can be considered in critically ill patients.[776][839]

  • Evidence is limited in patients with critical disease.

    • A systematic review and meta-analysis of nearly 28,000 hospitalised patients found that both intermediate-dose and therapeutic-dose anticoagulation decreased the risk of thrombotic events in critically ill patients in the intensive care unit compared with prophylactic-dose anticoagulation, but these regimens were associated with an increased bleeding risk and unchanged in-hospital mortality.[840]

Corticosteroids

  • The WHO strongly recommends systemic corticosteroid therapy (low-dose intravenous or oral dexamethasone or hydrocortisone) in adults with critical disease.

    • This recommendation is based on moderate-quality evidence that suggests systemic corticosteroids probably reduce 28-day mortality in patients with critical disease. They also probably reduce the need for invasive ventilation.[726]

  • In the US, the National Institutes of Health guidelines panel recommends dexamethasone (or a suitable alternative corticosteroid) in combination with baricitinib or tocilizumab (or dexamethasone alone if a second immunomodulator cannot be obtained), in hospitalised adults who require high-flow oxygen or non-invasive ventilation.[459]

    • Remdesivir may be added in certain situations.

    • In adults who are on mechanical ventilation or ECMO, the panel recommends dexamethasone in combination with baricitinib or tocilizumab (or dexamethasone alone if a second immunomodulator cannot be obtained) for patients who are within 24 hours of admission to the intensive care unit.

    • The panel recommends using dexamethasone (with or without remdesivir) in hospitalised children who require high-flow oxygen or non-invasive ventilation, or dexamethasone alone in hospitalised children who require invasive mechanical ventilation or extracorporeal membrane oxygenation.

  • A meta-analysis found an increased risk of VTE with corticosteroid administration in patients with critical disease. However, no definite findings were available due to the differing corticosteroid regimens and the heterogeneity of the studies.[841]

  • See the corticosteroids section under Severe COVID-19 above for more information.

Antivirals

  • There are conflicting recommendations across international guidelines about the use of the intravenous antiviral agent remdesivir in patients with critical disease. Remdesivir may increase the risk of death in critically ill patients, and currently only US guidelines recommend its use in select patients. Consult your local guidance for more information.

  • In the US, the National Institutes of Health guidelines panel recommends remdesivir, in combination with dexamethasone, in hospitalised children and adults who require high-flow oxygen or non-invasive ventilation.[459]

    • The panel does not recommend starting remdesivir in patients who require invasive mechanical ventilation or ECMO. However, the panel does recommend completing the full treatment course of remdesivir if the patient is started on it when they are on supplemental low-flow oxygen (see Severe COVID-19 above) and then progress to requiring high-flow oxygen, non-invasive ventilation, mechanical ventilation, or extracorporeal membrane oxygenation.

  • The recommended treatment course is 5 days or until hospital discharge, whichever comes first.[459]

    • Evidence does not suggest any greater benefit with a 10-day course of remdesivir compared with a 5-day course, but suggests an increased risk of harm.[523] However, some experts may recommend a 10-day course in patients who have not shown substantial clinical improvement by day 5.[459]

  • In the UK, the National Institute for Health and Care Excellence recommends against the use of remdesivir in hospitalised patients on HFNO, continuous positive airway pressure, non-invasive mechanical ventilation, or invasive mechanical ventilation, except as part of a clinical trial.[523]

    • This recommendation is based on moderate-certainty evidence that suggests remdesivir may increase the risk of mortality in people who are on these interventions.

  • The World Health Organization conditionally recommends against the use of remdesivir in patients with critical disease as it possibly has little or no effect on mortality or need for mechanical ventilation, and has an uncertain effect on time to symptom improvement.[726][728][729]

  • See the antivirals section under Severe COVID-19 above for more information.

IL-6 inhibitors

  • The WHO strongly recommends an IL-6 inhibitor in patients with critical disease.[726][728][729]

  • In the US, the National Institutes of Health guidelines panel recommends adding tocilizumab to dexamethasone (or a suitable alternative corticosteroid) or dexamethasone plus remdesivir in children ≥2 years of age and adults who require non-invasive mechanical ventilation or HFNO and have been recently hospitalised (e.g., within 3 days) with rapidly increasing oxygen needs and systemic inflammation.[459]

    • In patients who are on mechanical ventilation or ECMO, the panel recommends adding tocilizumab to dexamethasone for patients who are within 24 hours of admission to the intensive care unit.

    • Sarilumab may be used as an alternative if tocilizumab is not available or it is not feasible to use it.

    • The Infectious Diseases Society of America recommends considering tocilizumab in hospitalised adults with critical disease who have elevated markers of systemic inflammation, in addition to standard of care (i.e., corticosteroids), rather than standard of care alone. Sarilumab may be used if tocilizumab is not available.[460]

  • See the IL-6 inhibitors section under Severe COVID-19 above for more information.

Janus kinase (JAK) inhibitors

  • The WHO strongly recommends a JAK inhibitor (baricitinib) in patients with critical disease.[726][728][729]

  • In the UK, the National Institute for Health and Care Excellence recommends baricitinib in hospitalised adults who: need supplemental oxygen (or other respiratory support including HFNO, CPAP, non-invasive ventilation, or mechanical ventilation), and are having or have completed a course of corticosteroids (unless contraindicated), and have no evidence of infection (other than SARS-CoV-2) that might be worsened by baricitinib.[523]

    • Baricitinib may also be considered in children ≥2 years of age provided they meet the same criteria.

  • In the US, the National Institutes of Health guidelines panel recommends adding baricitinib to dexamethasone (or a suitable alternative corticosteroid) or dexamethasone plus remdesivir in children ≥2 years of age and adults who require non-invasive mechanical ventilation or HFNO and have been recently hospitalised with rapidly increasing oxygen needs and systemic inflammation.[459]

    • In patients who are on mechanical ventilation or ECMO, the panel recommends adding baricitinib to dexamethasone for patients who are within 24 hours of admission to the intensive care unit.

    • The panel recommends against the use of baricitinib in combination with tocilizumab except in the context of a clinical trial.

    • Tofacitinib may be used as an alternative if baricitinib is not available or it is not feasible to use it.

  • See the JAK inhibitors section under Severe COVID-19 above for more information.

Discharge and rehabilitation

  • Routinely assess intensive care patients for mobility, functional swallow, cognitive impairment, and mental health concerns, and based on that assessment determine whether the patient is ready for discharge, and whether the patient has any rehabilitation and follow-up requirements.[93]

Palliative care

  • Palliative care interventions should be made accessible at each institution that provides care for patients with COVID-19. Identify whether the patient has an advance care plan and respect the patient’s priorities and preferences when formulating the patient’s care plan.[93]

    • There is a lack of data on palliative care in patients with COVID-19. However, a rapid systematic review of pharmacological strategies used for palliative care in these patients, the first international review of its kind, found that a higher proportion of patients required continuous subcutaneous infusions for medication delivery than is typically seen in the palliative care population. Modest doses of commonly used end-of-life medications were required for symptom control. However, these findings should be interpreted with caution due to the lack of data available.[798]

  • Follow local palliative care guidelines.

Pregnant women should be managed by a multidisciplinary team, including obstetric, perinatal, neonatal, and intensive care specialists, as well as midwifery and mental health and psychosocial support. A woman-centred, respectful, skilled approach to care is recommended.[93] In women with severe or critical disease, the multidisciplinary team should be organised as soon as possible after maternal hypoxaemia occurs in order to assess fetal maturity, disease progression, and the best options for delivery.[842] There is significant heterogeneity in several aspects of management of pregnant women across clinical practice guidelines, especially regarding follow-up after infection and timing of delivery. However, there is a general agreement in the criteria for maternal hospitalisation and mode of delivery.[843]

Location of care

  • Manage pregnant women in a healthcare facility, in a community facility, or at home. Women with suspected or confirmed mild disease may not require acute care in a hospital unless there is concern for rapid deterioration or an inability to return to hospital promptly.[93] Follow local infection prevention and control procedures as for non-pregnant people.

  • Consider home care in women with asymptomatic or mild illness, provided the patient has no signs of potentially severe illness (e.g., breathlessness, haemoptysis, new chest pain/pressure, anorexia, dehydration, confusion), no comorbidities, and no obstetric issues; the patient is able to care for herself; and monitoring and follow-up is possible. Otherwise, manage pregnant women in a hospital setting with appropriate maternal and fetal monitoring whenever possible.[591][844]

  • Postpone routine antenatal or postnatal health visits for women who are in home isolation and reschedule them after the isolation period is completed. Delivery of counselling and care should be conducted via telemedicine whenever possible. Counsel women about healthy diet, mobility and exercise, intake of micronutrients, smoking, and alcohol and substance use. Advise women to seek urgent care if they develop any worsening of illness or danger signs, or danger signs of pregnancy.[93]

Antenatal corticosteroids

  • Consider antenatal corticosteroids for fetal lung maturation in women who are at risk of preterm birth (24 to 37 weeks’ gestation). Caution is advised because corticosteroids could potentially worsen the maternal clinical condition, and the decision should be made in conjunction with the multidisciplinary team.[591][844][845]

    • The WHO recommends antenatal corticosteroids only when there is no clinical evidence of maternal infection and adequate childbirth and newborn care is available, and in women with mild COVID-19 after assessing the risks and benefits.[93]

  • As corticosteroid therapy is recommended as part of the treatment strategy for COVID-19, the specific corticosteroid and dose may depend on whether corticosteroids are indicated for fetal lung maturity.[846]

Treatments

  • There are limited data available on the management of pregnant women with COVID-19; however, pregnant women can generally be treated with the same supportive therapies detailed above, taking into account the physiological changes that occur with pregnancy. It is important that pregnant women are not denied treatment inappropriately.[93][847]

  • Most clinical trials to date have excluded pregnant women. However, potentially effective treatments should not be withheld from pregnant women due to theoretical concerns about the safety of these therapeutic agents in pregnancy. Decisions should be made with a shared decision-making process between the patient and the clinical team.[459]

  • Treatment with corticosteroids should be modified to use non-fluorinated glucocorticoids. Il-6 inhibitors and monoclonal antibodies may be considered. VTE prophylaxis is important in pregnant women. They may also require respiratory support with oxygen, non-invasive ventilation, ventilation in a prone position, intubation and ventilation, and ECMO.[847] Safety of COVID-19 antivirals (e.g., remdesivir, nirmatrelvir/ritonavir, molnupiravir) in pregnancy has not been established. Report any pregnancies that occur during use of these drugs, including paternal use, to the UK COVID-19 Antivirals Pregnancy Registry.

VTE prophylaxis

  • The National Institutes of Health recommends prophylactic dose anticoagulation in pregnant women who are hospitalised with severe disease, provided there are no contraindications to its use. There is currently insufficient evidence to recommend either for or against therapeutic anticoagulation for pregnant women in the absence of known venous thromboembolism. Anticoagulation during labour and delivery requires specialised care and planning, and should be managed in a similar way as for pregnant women with other conditions.[459]

  • The Royal College of Obstetricians and Gynaecologists (RCOG) has also published guidance on the prevention of VTE in pregnant women.[846]

  • There is insufficient evidence to recommend either for or against the use of therapeutic anticoagulation in pregnant women who do not have evidence of VTE.[459]

Labour and delivery

  • Implement local infection prevention and control measures during labour and delivery. Screen birth partners for COVID-19 infection using the standard case definition.[93]

  • Individualise mode of birth based on obstetric indications and the woman’s preferences. Vaginal delivery is preferred in women with confirmed infection to avoid unnecessary surgical complications. Induction of labour, interventions to accelerate labour and delivery, and caesarean delivery are generally only recommended when medically justified based on maternal and fetal condition. COVID-19 positive status alone is not an indication for caesarean section.[93][591][844]

  • Delayed umbilical cord clamping (not earlier than 1 minute after birth) is recommended for improved maternal and infant health and nutrition outcomes. The risk of transmission via blood is thought to be minimal, and there is no evidence that delayed cord clamping increases the risk of viral transmission from the mother to the newborn.[93]

  • Consider babies born to mothers with suspected or confirmed infection to be a person under investigation and isolate them from healthy newborns. Test them for infection 24 hours after birth, and again 48 hours after birth.[848]

Newborn care

  • Experts are divided on separating mother and baby after delivery; make decisions on a case-by-base basis using shared-decision making.

    • A retrospective cohort analysis, the largest series to date, found no clinical evidence of vertical transmission in 101 newborns born to mothers with suspected or confirmed SARS-CoV-2 infection, despite most newborns rooming-in and direct breastfeeding practices. This suggests that separation may not be warranted and breastfeeding appears to be safe.[849]

  • The WHO recommends that mothers and infants should remain together unless the mother is too sick to care for her baby. Breastfeeding should be encouraged while applying appropriate infection prevention and control measures (e.g., performing hand hygiene before and after contact with the baby, wearing a mask while breastfeeding).[93]

    • The WHO advises that the benefits of breastfeeding outweigh the potential risks for transmission.[850]

    • Mother-to-infant transmission appears to be rare during rooming-in, provided that adequate droplet and contact precautions are taken.[851]

  • The RCOG recommends that mothers with confirmed infection and healthy babies are kept together in the immediate postnatal period, provided they do not meet criteria for maternal critical care or additional neonatal care.[846]

    • Breastfeeding should be recommended to all women in line with usual guidance, and women should be supported to make an informed decision about how they feed their baby. Appropriate preventive precautions are recommended to limit transmission to the baby.

  • The American Academy of Pediatrics (AAP) recommends that mothers and newborns may room-in, with appropriate infection prevention and control precautions, according to usual centre practice.[848]

    • However, it may be appropriate to temporarily separate the mother and newborn (or to have the newborn cared for by non-infected carers in the mother’s room) when the mother is acutely ill with COVID-19 and cannot care for the infant in a safe way.

    • The AAP strongly supports breastfeeding as the best choice for feeding. Breast milk can be expressed after appropriate hygiene measures and fed by an uninfected carer. If the mother chooses to breastfeed the infant themselves, appropriate prevention measures are recommended.

    • After discharge, advise mothers with COVID-19 to practice prevention measures (e.g., hand hygiene, respiratory hygiene/mask) for newborn care until: they are afebrile for 24 hours without the use of antipyretics; at least 10 days have passed since symptoms first appeared (or 10 days since a positive test in asymptomatic women); and symptoms have improved.

    • A newborn with documented infection but no symptoms requires close outpatient follow-up after discharge for 14 days after birth

Best Practice has published a separate topic on the Management of co-existing conditions in the context of COVID-19.

Use of this content is subject to our disclaimer