Updated Date: December 11, 2020
Acute cardiac injury, variably defined as increased circulating troponin levels and/or new abnormalities on ECG or echocardiography, were noted in 7-22% of hospitalized patients in early reports from Wuhan (Ruan et al; Wang et al; Chen et al; Shi et al; Guo et al, Zhou et al). When present, these findings were associated with increased risk of ICU admission and death.
- Direct SARS-CoV-2 infection of cardiac myocytes (myocarditis)
- Demand ischemia, with either large or small vessel thrombosis
- Stress (Takotsubo) cardiomyopathy
- Pathological myocardial response to inflammation or cytokine storm
Specific cardiac pathologies include myocarditis, arrhythmia, and precipitation of an acute coronary syndrome. Hypercoagulability in COVID-19, including its impact on the heart, are discussed in Hematology.
Fulminant SARS-CoV-2 myocarditis was clinically suspected in some early case reports (Ruan et al; Zeng et al; Hu et al; Inciardi et al), based on pre-existing clinical criteria and, in some cases, suggestive findings on cardiac MRI (Inciardi et al, Kim et al). Subsequent examination of myocardial tissue in autopsy series (Fox et al; Elsoukkary et al; Basso et al) found direct evidence for viral myocarditis (e.g., lymphocytic infiltrates) were relatively rare (0-14%).
The clinical significance of direct SARS-CoV-2 myocarditis remains unclear. If a patient has elevated troponins with no evidence of obstructive coronary artery disease, it may be on the differential diagnosis but is unlikely to alter management.
- Provide supportive care for heart failure (Zhang et al.) or cardiogenic shock
- Where possible, discuss with cardiology and/or infectious disease consultants to see if the patient might benefit from antivirals or steroids (benefit is unknown)
- Endomyocardial biopsy is unlikely to be informative.
- See Advanced CV Imaging below regarding uses of cardiac MRI.
Cardiac arrhythmias can occur in COVID-19.
An early case series of 138 patients in Wuhan, China, (Wang et al) found evidence of some arrhythmias in 17% of hospitalized patients with COVID-19, rising to 44% in those transferred to the ICU. Another early study of 189 hospitalized patients noted nearly 6% of inpatients had an episode of ventricular fibrillation or sustained ventricular tachycardia (Guo et al).
- Atrial Fibrillation/Atrial Flutter
- Consider beta-blockers, if no evidence of heart failure or shock.
- If acute heart failure or concern for hypotension, use amiodarone if not otherwise contraindicated.
- If unstable (with a pulse), synchronized DC cardioversion with 200 joules (biphasic).
- Ventricular Tachycardia
- If unstable or without palpable pulses: initiate local advanced life support protocol (e.g., ACLS).
- If stable:
- Involve a cardiologist. If cardiologist is not available, involve a senior clinician.Consider a single IV dose of amiodarone 150mg or lidocaine 100mg
Myocardial infarction in COVID-19 may be triggered by a combination of hypercoagulability, cardiac expression of SARS-CoV-2 entry receptor ACE2, possible direct viral myocardial injury, increased myocardial demand, or toxicity from inflammation. Cardiac markers and ECG changes alone may not be able to determine whether an underlying obstructive lesion exists as evidenced by the fact that up to 45% of hospitalized COVID patients have elevated cardiac markers (Lombardi et al).
The diagnosis of an acute coronary syndrome depends on:
- Symptoms (if able to communicate): worsening shortness of breath, chest pain, or other anginal equivalents
- Regional changes in the ECG or wall motion abnormalities on echocardiography
- Rate of change of troponin changes (rapid rise or fall suggests an acute event)
Tool: Life in the Fast Lane page for a summary.
When the diagnosis is not clear, a cardiologist should be consulted.
If a patient is diagnosed with ACS, management should be coordinated with a cardiologist if at all possible. Medical management typically includes:
- Treatment with full dose aspirin, clopidogrel (if not bleeding), heparin, oxygen (if hypoxemic), high-dose statin, nitrates (if hypertensive), and opioids as needed for symptom control. Beta blockers should be used with caution, given the risk of concomitant myocarditis or decompensated heart failure
- If cardiac catheterization is available, there are no fundamental contraindications for patients with COVID-19 as long as strict infection control precautions are followed.
- If cardiac catheterization is not available or if constrained resources unacceptably prolong door-to-balloon time, thrombolytic medications may be considered in lieu of PCI.
Laboratory markers of cardiac injury (troponins, CK-MB, BNP) and electrocardiography are appropriate for most COVID-19 patients admitted to the hospital. Depending on availability and expertise, point-of-care ultrasound can also be considered, particularly in patients with concerning symptoms, lab values or ECG.
The indications for more advanced cardiac diagnostics are similar to patients without COVID-19, but with additional consideration for infection control. Testing should be limited to cases where the results will alter management to avoid unnecessary risk to providers and other patients.
- Transthoracic echocardiography
- Do not obtain routinely.
- When possible, a bedside provider should assess cardiac function with point-of-care ultrasound for the following indications:
- Marked elevation in troponin or NTproBNP, or decline in ScvO2/MvO2
- New heart failure
- New persistent arrhythmia
- Significant ECG changes
- If abnormalities are identified on point-of-care ultrasound (such as a new decrease in LV ejection fraction to below < 50%) and the patient is stable, a formal echocardiogram should be obtained if possible.
- Evaluate both left and right ventricular function.
- The differential diagnosis for right ventricular dysfunction includes myocarditis, hypoxic vasoconstriction, pulmonary embolus, and cytokine mediated dysfunction.
- The differential diagnosis for left ventricular dysfunction includes myocarditis, acute coronary syndrome, and stress-induced cardiomyopathy.
- Regional wall motion abnormalities with elevated troponins suggest an acute coronary syndrome, though direct myocardial injury by the virus can also result in focal wall motion abnormalities.
- Stress testing:
- Should not be commonly required in patients with active COVID. If needed (and if available), consider pharmacologic nuclear stress testing or coronary CT angiography rather than exercise stress test.
- Transesophageal Echocardiogram (TEE)
- Only request if absolutely necessary. Although unclear whether this generates aerosolized virus, it likely does represent increased risk to the patient and provider.
- Consider alternative noninvasive imaging modalities (such as cardiac CT or PET/CT) if they are available and appropriate for the question being asked.
- Cardiac CT
- Can consider for selected patients with elevated cardiac biomarkers when there is a need to distinguish myocardial injury from acute coronary syndrome. The decision to use CT in this context should be discussed with a cardiologist.
- Consider in selected patients as a substitute for TEE to rule out left atrial appendage clot or to evaluate for endocarditis.
- In appropriate cases, multiphase data acquisition may be used to evaluate both left and right ventricular function while concurrently evaluating the lung parenchyma or pulmonary artery.
- Cardiac MRI
- Consider for selected patients with elevated cardiac biomarkers and concern for myocarditis, if this information will impact patient management.
- In the acute and sub-acute periods, T1 and T2 mapping as well as assessment of extracellular volume fraction (ECV) may improve sensitivity for myocarditis. However, the prognostic significance of such abnormalities (especially in the presence of normal ventricular function or when late enhancement abnormalities are absent) remains unclear.
- In selected patients who have recovered from COVID, cardiac MRI using late gadolinium enhancement may be useful for evaluating for residual scar tissue.
- Nuclear imaging
- In COVID-19 patients who require stress testing, vasodilator stress testing is preferred over exercise testing.
- In selected patients who have recovered from COVID-19, PET MPI using a quantitative assessment of myocardial blood flow may be useful for evaluating microvascular dysfunction.
Preexisting cardiovascular disease and metabolic disorders (including diabetes and hyperlipidemia) worsen prognosis in acute COVID-19 (Izcovich et al).
See Anticoagulation in the Hematology section below.
Patients with pre-existing heart failure have a nearly two-fold increased mortality and over three-fold greater risk of mechanical ventilation when they develop COVID-19 (Alvarez-Garcia et al).
There are currently no specific medication changes recommended for patients with prior heart failure who develop COVID-19, though all medications should be titrated based on other clinical context (e.g., in a patient with fevers and decreased oral intake, home diuretic doses may need adjustment).
Patients with rheumatic heart disease or other conditions requiring surgical intervention should continue to be considered as a potential priority case during the COVID-19 pandemic.
Patients with pre-existing hypertension have a significantly increased risk of developing severe COVID-19 disease or dying (Izcovich et al).
Antihypertensive medications, such as RAAS inhibitors (e.g. ACE inhibitors) were initially suspected to be harmful in COVID-19, but these harms have not been supported by subsequent data (e.g., Baral et al). Other classes, such as calcium channel blockers, were thought to possibly be beneficial, but this also remains unsubstantiated.
Unless new data become available, patients with well-controlled hypertension should continue their current anti-hypertensive medications, unless those drugs need to be stopped for other reasons (e.g. renal issues).
The initial protocol used by Brigham and Women’s Hospital, USA, for patients with heart transplant who develop COVID-19 is available here.
Updated Date: December 11, 2020
COVID-19 can cause hyperglycemic crisis in patients with and without known insulin resistance (summarized in Rubino et al). A systematic review of 110 reported cases of diabetic ketoacidosis (DKA) or combined DKA/hyperglycemic hyperosmolar state (HHS), found that 10% of patients with these severe presentations did not have a prior diagnosis of diabetes (Pal et al).
All patients with moderate to severe COVID-19 should thus be evaluated for hyperglycemia. Patients who do have blood glucose levels over 10 mmol/L (180 mg/dl) are generally managed with insulin rather than oral agents. Targets are similar to the management in patients without COVID-19, while recognizing that insulin requirements may be labile.
Treatment of DKA and/or HHS also has the same goals as patients without COVID-19, while recognizing provider safety concerns and limited resources in a pandemic. An example protocol for mild-to-moderate DKA at Brigham and Women’s Hospital, USA, uses subcutaneous insulin with slightly less frequent monitoring than standard protocols, in an effort to minimize provider exposure and conserve PPE while maintaining patient safety.
People with diabetes who develop COVID-19 have ~2-3 fold increases in both mortality and risk of severe COVID-19 (summarized in Izcovich et al).
This risk in part reflects the chronic effects of diabetes, including increased susceptibility to several infections and an association with other chronic diseases, such as dyslipidemia, hypertension, and obesity.
As noted above, however, Pal et al. systematically reviewed several case reports where patients with diabetes and COVID-19 developed severe hyperglycemic crises, including diabetic ketoacidosis (DKA) and hyperglycemic hyperosmolar syndrome (HHS).
- Seventy-four the 110 cases who presented with DKA or combined DKA/HHS had a prior diagnosis of type 2 diabetes. Since these are case reports, however, they do not define how frequently DKA occurs in severe COVID-19 compared to other severe infections.
- Instruct them to continue their normal oral or insulin regimens and monitor their glucose more frequently than usual. Depending on their glucose, they may need to temporarily increase their regimen.
- This includes those patients with type 2 diabetes not on insulin who may not be accustomed to monitoring their glucose -- they should check twice daily if possible.
- Use caution with oral hypoglycemic agents such as sulfonylurea or SGLT2 inhibitors, which can lead to euglycemic DKA, in patients with decreased caloric intake.
For all patients,
- Frequent blood glucose and/or ketone (blood or urine) monitoring should be performed.
- Target blood glucose remains the same as without COVID-19; for hospitalized patients, the Joint British Diabetes Society recommends 6-10 mmol/L, while the American Diabetes Association targets 140-180 mg/dl (Corsino et al.). Blood ketones should be kept below 0.6 mmol/L
- Do not stop basal insulin even if febrile in those with type 1 diabetes or those with type 2 diabetes and require basal insulin for glycemic control. COVID-19 can significantly increase baseline insulin requirements.
- Monitor and maintain appropriate salt and water balance.
- For patients requiring systemic steroids, appropriate insulin adjustments are required.
- Manage DKA and HHS as discussed above.
A BMI over 25-30 (different cutoffs depending on the study) appears to increase the odds of mortality from COVID-19 by nearly 50% and the odds of severe disease by ~2-4 fold (Izcovich et al, Popkin et al). The latter of these two papers raises the concern that BMI may impact the response to vaccination for COVID-19, based on experience with influenza and other vaccines.
There are no specific recommendations for management of COVID-19 patients with elevated BMI, but providers should be cautious with drug dosing, remain aware of possibly altered respiratory mechanics, and stay vigilant for decompensation.
Updated Date: December 11, 2020
Coagulopathy is common in patients with COVID-19; early reports found 8-10% of hospitalized patients met criteria for disseminated intravascular coagulation (DIC; Tang et al) 16 of 183 hospitalized patients in Wuhan , most commonly in the critically ill or those wiith multi-organ dysfunction (Zhou et al). However, the coagulopathy in COVID-19 differs from DIC in bacterial sepsis and may require different management (Merrill et al; Asakura et al).
Median time to diagnosis of DIC in the series by Tang et al was 4 days into hospital admission, and this diagnosis was associated with worse survival in COVID-19 patients. Out of 183 COVID-19 patients in Wuhan, 71% of non-survivors had ISTH score ≥ 5 compared to 0.6% of survivors..
To diagnose DIC you can use the ISTH DIC score (MDcalc online calculator)
- If score < 5, no DIC; recalculate in 1-2 days
- If the patient develops DIC, measure PT/INR, PTT, D-dimer, fibrinogen every 3 days until discharge or death. Elevated PT/PTT and D-dimer correlate with worse prognosis
Management If Not Bleeding:
- See Blood Products. If fibrinogen < 150 mg/dl: use FFP, cryoprecipitate or fibrinogen concentrate if you are worried about infusion volume of other options and it is available (RiaSTAP or Fibryga)
- If platelets <30 k/mcl, transfuse (Consider holding anticoagulation if the patient requires blood products if benefits outweigh risks)
Management if Bleeding:
- See Blood Products. For elevated PT/PTT and bleeding, use FFP, cryoprecipitate or fibrinogen concentrate if you are worried about infusion volume of other options and it is available
- Hold anticoagulation for active bleeding in most cases. Start systemic anticoagulation only if critical thromboembolism or organ failure due to clot (i.e., purpura fulminans). There has been no mortality benefit of therapeutic anticoagulation in DIC (Levi et al).
Most patients with COVID-19 have either a normal white blood cell count (WBC). A small number may have elevated WBC or low WBC. Leukocytosis (>10,000/µL) in 13% and leukopenia (<4000/µL) in 15.5% (Goyal et al Lymphocytopenia, or lymphopenia, typically defined as an absolute lymphocyte count < 1000/µL, is the most common abnormality on the CBC in COVID-19 and is found in over 80% of hospitalized patients (Guan et al; Huang et al). Low lymphocytes are also associated with poor prognosis, with lymphocyte percentage <10% on the WBC differential is strongly associated with decreased survival. (Ruan et al; Tan et al; Yang).
Numerous possible explanations for lymphopenia in COVID-19 have been proposed including
- Invasion/ destruction of lymphocytes via ACE2 receptor
- Acidemia, nutrition, bone marrow suppression
- Cytokine Storm
- Lymphatic organ damage (thymus, spleen) This possibility still requires pathological evidence and remains speculative (Tan et al, Signal Transduct Target Ther, 2020)
- Host Endothelial function. With age and chronic disease, there is more leukocyte adhesion and extravasation (Bermejo-Martin, Journal of Infection, 2020)
- Sequestration of lymphocytes. Cytokine release leads to movement of the lymphocytes to the site of infection, the lung tissue, which could contribute to peripheral lymphopenia (Rahimmanesh, Preprint, 2020)
Any patient with low lymphocytes should be considered potentially infected with COVID unless there is an alternate explanation.
- Please note that concurrent infection and the use of steroids may skew these results. See secondary infections.
No current treatment regimen management changes based on lymphopenia.
- There is no evidence for giving pneumocystis jiroveci prophylaxis given the transient nature of lymphopenia with COVID-19
Thrombocytopenia affected ~12% of patients with COVID-19 in one large meta-analysis (). Another early report found thrombocytopenia in 72.5% of hospitalized patients (Chang et al, JAMA, 2020), and the degree of thrombocytopenia appears to correlate with worse prognosis (Yang et al). Multiple proposed mechanisms have been proposed (Xu et al; Amgalan et al). Click here for a flow chart summarizing possible mechanisms. Disseminated intravascular coagulation (DIC) may contribute as a related or independent process.
- Consider other potential contributing etiologies of thrombocytopenia. Medication, additional infection(s), liver disease, splenomegaly, heparin-induced thrombocytopenia (HIT), thrombotic microangiopathy (TTP, HUS, DIC), alcohol, malignancy, pregnancy, rheumatologic/autoimmune, bone marrow disorders
- Initial workup can include: Peripheral blood smear, PT, aPTT, fibrinogen, LDH, LFTs, B12, folate
- If concerned for Heparin Induced Thrombocytopenia, the pretest probability of HIT can be calculated by 4Ts score (MDCalc 4Ts calculator)
- Laboratory testing for HIT should typically only be sent in patients with at least intermediate probability of HIT (4 or more points on 4Ts score), although need to consider clinical context.
- If sending PF4, use a non-heparin anticoagulant (e.g. bivalirudin or other direct thrombin inhibitor per institutional protocols) while awaiting PF4 results. Serotonin release assays may be necessary to confirm positive PF4 results.
- If concern for DIC, refer to DIC protocol section
- If not bleeding, transfuse platelets if < 10,000/µL
- If bleeding, transfuse platelets according to clinical situation
- In ICU patients, cumulative incidences range from 9% to 70% in patients on varying levels of prophylactic anticoagulation, and whether patients were screened with compression ultrasonography or imaged for change in clinical status (Klok et al; Middeldorp et al; Klok et al; Llitjos et al; Nahum et al; Moll et al). One study suggests COVID-19 patients at increased risk for thrombosis and bleeding (Xu et al, Pulmonology 2020 [preprint, under review]).
- We found that in 102 COVID-19 positive ICU patients, there were 9 radiographically-confirmed DVT or PE, based on imaging obtained for a change in clinical status; all patients received standard dose prophylactic anticoagulation (enoxaparin 40 mg daily or unfractionated heparin 5000 IU three times daily). No events occured in 108 wards patients (Moll et al). Similar findings were reported in Indianapolis (Maatman et al).
- Higher D-dimer and FDP levels track with multi-organ dysfunction syndrome and poorer prognosis (Wang et al).
- The mechanism for VTE are unknown and likely multifactorial:
- Systemic inflammatory response as seen in sepsis
- Stasis/critical illness
- Possibly direct endothelial damage from viral injury/ACE2 binding
- An autopsy series of 10 patients from New Orleans reported thrombotic and microangiopathic pathology (and diffuse alveolar damage) (Fox et al). Our discussions with pathology colleagues indicate more cellular debris than microthrombi.
- There is a theory from the SARS epidemic that SARS-CoV1 Spike protein can be cleaved by FXa and FIIa. Cleavage of the Spike protein activates it which promotes infectivity (Du). By extension, it is hypothesized that anticoagulation might inhibit SARS-CoV-2 replication, however this remains unproven.
- There is a small case series suggesting dipyridamole may be useful, though anticoagulation and antiplatelet agents require further investigation prior to being used therapeutically (Liu et al; Lin et al).
- There is evidence that A blood type is a risk factor for COVID-19 respiratory failure, and O may be protective. This was based on a genome-wide association study (GWAS) of 835 patients and 1255 control participants from Italy and 775 patients and 950 control participants from Spain. Respiratory failure was defined as a patient requiring supplemental oxygen or mechanical ventilation (Ellinghaus et al).
In patients previously anticoagulated with vitamin K antagonists (e.g. warfarin), PT/INR should be monitored closely. Both fever and acute infection may result in increases in INR. Other medications that may be used in COVID-19, such as antibiotics for suspected coinfection, can also result in increased or decreased warfarin metabolism. If frequent INR measurement for dose-titration is not available, and switching to an alternative, parenteral anticoagulant is not feasible, the warfarin dose may be empirically reduced by approximately 10% (e.g. from 5mg daily to 4.5mg daily) when a patient develops fever.
Anticoagulation should not be stopped for patients with COVID-19 unless there is a different reason to do so.
In sickle cell crisis, patients have to consider the risk of COVID-19 exposure when going to the hospital for management of sickle cell crisis. Based on the local risk of nosocomial infection, providers and patients should discuss in advance regarding the patient-specific criteria which would warrant hospital evaluation versus staying at home with oral pain medications, hydration, and rest.
Updated Date: December 11, 2020
Data on coinfection and secondary infections in COVID-19 are limited.
Rates: There is enormous variance in the rates of viral coinfection depending on location, season, and viral coinfection epidemiology. A study in San Francisco found ~20% of symptomatic COVID-19 patients were also PCR positive for another viral pathogen (Kim et al). A meta-analysis of 1014 hospitalized COVID-19 patients found a viral co-infection rate of 3% (95% CI 1-6%, I2=62.3%), with RSV and influenza being the most common coinfections (Lansbury et al). In contrast, two studies in San Francisco and Wuhan, China where hospitalized COVID-19 patients tested for influenza and RSV found that none of these patients had evidence of viral co-infection (Myers et al; Chen et al).
- The decision to test for concurrent viral panels should be based on availability and local epidemiology. Many COVID testing locations do not have the ability to also test for respiratory viruses
- All respiratory viral infections should be considered COVID until proven otherwise (see testing), even if they present with minimal symptoms.
- Empiric oseltamivir is reasonable in some circumstances where influenza rates are high and the patient has tested negative for COVID infection
Most patients who have COVID do not have concurrent bacterial infections. However, as with other viral infections, impaired mucociliary clearance can make these patients susceptible to secondary bacterial infections.
- Coinfection vs secondary infection: One meta-analysis of 3448 COVID-19 patients broke bacterial infections down into co-infection and secondary infection and found the risk of co-infection on presentation to be 3.5%, while the risk of secondary infection after presentation was 15.5%. In this same cohort, 71.3% of patients received antibiotics, despite only 7.1% of patients overall having a bacterial infection (Langford et al) Other studies of secondary bacterial infections show incidence of around 7-8% of hospitalized patients. One meta-analysis of 2183 hospitalized COVID-19 patients found 7% had a bacterial coinfection (95% CI 3-12%, I2=92.2%) (Lansbury et al) Another meta-analysis of 806 hospitalized COVID-19 patients found 8% developed bacterial and/or fungal infections during admission (Rawson et al).
- Most common infections: pneumonia (32%), bacteremia (24%), and urinary tract infections (22%) (He et al).
- Glucocorticoid treatment was also found to be positively associated with secondary infection (He et al). However, we do not recommend withholding steroids in patients who qualify, even if they have concurrent bacterial infection.
Tool: A group out of the University of Toronto created a living systematic review of the data
Clinical reports indicate that rates of bacterial superinfection with COVID-19 are low, but when present, increase mortality risk. That said, unnecessary antibiotics carry risks of fluid overload and drug-resistance, as well as the possibility that antibiotics may become a limited resource. (Zhou et al; Yang et al; Lippi et al; WHO, COVID-19 Interim guidance, May 2020).The choice about whether to give empiric antibiotics will rely on whether or not secondary bacterial infection can be safely ruled out and the acuity of the patient.
- If laboratory and imaging guidance is available, use this evidence to guide the choice about whether to use antibiotics. There is a disproportionate high use of antibiotics despite paucity of evidence for bacterial secondary infection (He et al; Zhou et al; Rawson et al).
- Workup can include any or all of WBC count, left shift and bandemia, procalcitonin, sputum culture, urine analysis and color, cholangitic picture on liver function tests or RUQ ultrasound, urine strep + legionella antigen, blood cultures, stool or other relevant cultures
- If laboratory and imaging is not available or cannot be used to rule out a concurrent bacterial infection, antibiotics should be considered depending on the clinician’s expectations about risks and benefits
- On the one hand, not treating a bacterial co-infection (depending on the type) could be fatal in some patients. There is a strong association between nosocomial infection and mortality (He et al; Wang et al). On the other hand, in one study 75% of patients who developed secondary infection were already receiving prophylactic antibiotics, suggesting prophylactic agents may not prevent hospital-acquired infections and risk selecting for more drug-resistant pathogens (He et al).
- If a patient has shock or multiorgan failure it is appropriate to give antibiotics for the first 24-48 hours until the source is identified
- If antibiotics are to be used, they should reflect guidelines based on presumed source and multi-drug resistant organism risk factors. Administer oral antibiotics (azithromycin, levofloxacin, ciprofloxacin, etc.) when possible to reduce volume load, unless concerns for poor oral absorption.
- Organisms reported for those with secondary bacterial infections included those commonly seen with hospital-acquired infections including Mycoplasma sp., Haemophilus influenzae, Pseudomonas aeruginosa, Klebsiella sp., Enterobacter sp., Staphylococcus aureus, Acinetobacter sp., and E.coli, and vancomycin-resistant Enterococcus sp. (Langford et al; Lansbury et al)
- For empiric coverage for a presumed pulmonary source of infection, we recommend using your institutions antibiogram if one is available. If not, some possibilities include:
- In patients without risk factors for methicillin-resistant Staphylococcus aureus (MRSA) or Pseudomonas (i.e., those living in community without a history of resistant organisms), initiate ceftriaxone and (azithromycin or doxycycline)
- In patients with risk factors for MRSA or Pseudomonas (i.e., chronic hospitalization, prior resistant infections), obtain a respiratory culture and a MRSA nares screen if available and initiate an antipseudomonal cephalosporin (e.g. cefepime) and vancomycin
Tool: IDSA guidelines
Tool: Sanford guide
- Unnecessary antibiotics should be discontinued as soon as possible (ideally, within 48 hours) upon culture maturation. Clinical judgement should prevail over any specific lab value, but we suggest discontinuing when the following criteria are met:
- Clinical status is not deteriorating
- Cultures do not reveal pathogens at 48 hours and/or procalcitonin and WBC are relatively stable from 0 to 48 hours
Malaria often presents with fever and could be confused with COVID in some patients. Where testing is possible (RDT, blood smear), it is important to test for these and to plan for increased demands on testing (Dittrich et al).
- Where incidence is high and testing is not available, empiric (presumptive) therapy with Artemisinin Combined Therapy or the locally-approved regimen is appropriate (even though in non-COVID times empiric treatment this is generally discouraged by the WHO). Chloroquine may be used, but only if it is part of the preferred regimen for Malaria, and not for COVID (see hydroxychloroquine).
- Corticosteroids in concurrent malaria and COVID infection is not yet studied (Brotherton et al). Despite this we recommend using them for COVID infection as you would if the patient was not co-infected.
Dengue fever, like malaria, should be on the differential for COVID in places where it is prevalent. Treatment is largely supportive (oral rehydration or IV rehydration therapy, analgesics, antipyretics).
Parasitic infections should be treated as they would normally (with normal dose antihelminthic agents). We do not support the use outside of clinical trials of high-dose ivermectin for COVID outside of clinical trials.
Strongyloidiasis is a parasitic infection that is often asymptomatic. However, a life-threatening hyperinfection syndrome can occur with immunosuppression, including the use of corticosteroids. Because corticosteroids are a recommended therapy for COVID with hypoxemia or critical illness, we recommend the following: (Stauffer et al.)
Confirmed COVID with asymptomatic, minimally symptomatic, or mild disease (not a current candidate for corticosteroids)
Birth, residence, or long-term travel in Asia, Oceania, Sub-Saharan Africa, South America, Caribbean, Mediterranean countries, Middle East, North Africa*
Screen for Strongyloides infection and treat with ivermectin if positive
Confirmed COVID and likely candidate for corticosteroid treatment
Birth, residence, or long-term travel in Asia, Oceania, Sub-Saharan Africa, South America, Caribbean, Mediterranean countries, Middle East, North Africa*
Empiric treatment with ivermectin
*These groups are at moderate to high risk for disseminated strongyloides infection with administration of corticosteroids (A K Boggild et al).
Fungal pathogens such as Aspergillus sp., Candida albicans, and Pneumocystis jirovecii have been described in a subset of patients (Lansbury et al; Menon et al). Some case series have reported COVID-19 associated pulmonary aspergillosis rates of 20-35% (Arastehfar et al), while others are as low as 3.8% (Lamoth et al). Unsurprisingly, aspergillus infection appears to be associated with increased mortality (OR 3.53). In a prospective Italian cohort of 108 mechanically ventilated COVID-19 patients, probable pulmonary aspergillosis was diagnosed in 30 patients (27.7%) after a median of 4 days from ICU admission, and these patients had a much higher risk of 30-day mortality (OR 3.53 (95% CI 1.29-9.67, p=0.014). Of note, most patients received tocilizumab or steroids in this cohort (Bartoletti et al).
- At this time we do not recommend screening all patients with galactomannan and beta glucan, but patients who are already immunosuppressed, BMT, or oncologic patients should be screened with weekly
- Treatment, and choices around immunosuppression, in these cases is highly individualized and infectious disease consultation is suggested where available.
The interaction between HIV and SARS-CoV-2 remains poorly defined and is likely complex. It remains unclear if, and how, HIV infection affects risk or severity of COVID-19.
Risk of acquiring COVID infection and outcomes in those infected: Multiple studies from New York City (Richardson et al; Sigel et al; Karmen-Tuohy et al), Spain (Vizcarra et al), and China (Guo et al) have found that HIV-positive patients develop COVID-19 at a similar rate as the general population. However, the patients included in these studies were largely on antiretroviral therapy (ART) with well-controlled HIV. A large Spanish cohort study of people with well-controlled HIV found that the rate of COVID-19 diagnosis and hospitalization in HIV patients was decreased to 30.0 cases per 10,000, compared with 41.7 per 10,000 in the general population (Del Amo et al). There is very limited data on COVID-19 patients with poorly controlled HIV or AIDS. One study examined public healthcare data in South Africa, which has the highest rate of HIV in the world at about 20%, with about ⅔ of those on ART (UNAIDS). In this population, HIV infection conferred an adjusted hazard ratio of 2.75 for risk of death from COVID-19 (Davies M, presentation on behalf of Western Cape Department of Health). Detailed information, including the number of participants, CD4+ T cell counts, HIV viral loads, and ART treatment status, has not yet been made available.
Antiretrovirals : Some antiretroviral therapies used for HIV may be protective against COVID-19, but this is not yet fully supported by the data. For further information about antiretroviral agents under investigation for treatment of COVID-19, please see lopinavir-ritonavir in the therapeutics chapter (this so far has not been shown to benefit patients (Cao et al) or reduce viral shedding (Cheng et al)). Tenofovir has also been hypothesized to have a protective benefit, but data seems to be confounded by age and health of participants. A large Spanish cohort study of over 77,000 people with HIV, 236 of whom were diagnosed with COVID-19, found that patients taking tenofovir disproxil fumarate (TDF)/emtricitabine (FTC) had a significantly decreased risk of COVID-19 diagnosis and hospitalization compared with those taking tenofovir alafenamide (TAF)/FTC or abacavir (ABC)/lamivudine (3TC). This may be an effect of increased blood concentrations of tenofovir with TDF compared with TAF, though may also reflect that patients taking TDF are typically younger and healthier than those on TAF (Del Amo et al). Conversely, a smaller observational Spanish study of 2873 HIV-positive individuals, 51 of whom had COVID-19, found that tenofovir (either TDF or TAF) use was disproportionately enriched among COVID-19 cases (Vizcarra et al).
- Keep in mind that people with HIV may present differently. Fever may be less frequent.
- HIV should not change the role of either NAAT or antigen testing.
- The impact of prior HIV on immune response and development of antibodies is not yet known.
Studies to date suggest that well-controlled HIV does not substantially increase the risk or severity of COVID-19, but data on patients with low CD4+ counts remains sparse. Given the limitations of the existing evidence at this time, we recommend that HIV-positive patients be considered high risk and be counseled on precautions accordingly
- Per existing standard of care, all patients with HIV should remain on a daily ART regimen under the supervision of a trained provider
- There is speculation that lymphopenia and immune dysfunction in HIV-positive individuals may protect from the hyperinflammatory state thought to contribute to severe COVID-19 disease (Mascolo et al), but no evidence currently exists to support this theory. This is not a reason to stop HIV treatment.
- We do not recommend changing an existing ART regimen for the purposes of prophylaxis or treatment of COVID-19 in HIV-positive patients
- HIV-positive patients who develop COVID-19 do not require any change from standard protocol in management or treatment strategies
- Given the high prevalence of malnutrition among patients with TB/HIV, ensuring continued social support including food packages is important for disease control.
- Patients with HIV who present with respiratory symptoms should be evaluated for TB in addition to COVID-19 if clinically indicated.
Not enough is known about the incidence of COVID in patients with tuberculosis. One case-study of 49 patients with tuberculosis (eight with drug resistant TB) showed a high case fatality at 12.3% (Tadolini et al). Some have posited that the increased social distancing measures from COVID will decrease tuberculosis, however it is more likely that any benefit on TB deaths is likely to be outweighed by health service disruption (McQuaid et al).
Tool: This multi-institution consensus statement describes TB public health and treatment plans during the COVID epidemic in great detail. Core management issues described include medications, drug-drug interactions, novel therapies, and principles of infection control and workplace safety.
Patients who are newly diagnosed with viral hepatitis B should initiate hepatitis treatment if they qualify, regardless of COVID-19 status. However, patients who are newly diagnosed with viral hepatitis C should defer treatment until after the COVID-19 infection has cleared. For patients with viral hepatitis B or C who are already on treatment, they should continue treatment while being monitored for drug-drug interactions (Reddy).
Updated Date: May 3, 2020
Incidence of Acute Kidney Injury in COVID-19 varies widely, but estimates have ranged from 0.5% (Guan et al) to 27% (Diao et al). The wide range of estimates of AKI incidence may reflect different populations included in studies. The most likely etiology of AKI is acute tubular necrosis (ATN) based on autopsy series from China, but other findings including interstitial inflammation, thrombotic microangiopathy, complement-mediated injury and direct viral infection of tubular cells and podocytes has also been described (Su et al; Diao et al). Studies find variable onset of AKI, from 7 days (Cheng et al) to 15 days after illness onset (Zhou et al). Onset of AKI more rapid and severe in patients with underlying CKD (Cheng et al)
Role of the renin-angiotensin-aldosterone system and medications that target it on the severity of COVID-19 is a source of much speculation and research, since angiotensin-converting enzyme 2 (ACE2) is used by SARS-CoV-2 as a functional receptor to enter into cells (including type II pneumocytes and kidney tubular epithelial cells). There is no data to support issues with RAAS inhibitors during COVID at this time.
- Monitor serum creatinine and electrolytes at least daily where available
- In patients with AKI, order urine electrolytes (urine Na, urea and Cr) and urinalysis with sediment
- Patients may present with proteinuria (44%), hematuria (26.9%) (Cheng et al). For patients with proteinuria, quantify proteinuria with spot urine protein-to-creatinine and albumin-to-creatinine ratios
- Consider other common etiologies of AKI that can occur in patients who do not have COVID-19 (e.g. volume depletion, ATN from hypotension, contrast-associated nephropathy, acute interstitial nephritis and urinary tract obstruction)
- Discontinue all medications that can contribute to AKI (e.g. NSAIDs, ACE inhibitors, ARBs, and diuretics in volume depleted patients) and avoid using iodinated contrast with CT imaging as much as possible
- Consider a gentle fluid challenge (e.g. 1 liter of isotonic crystalloid fluid) to determine if there is a pre-renal component to AKI, especially in patients with clinical or laboratory signs suggestive of intravascular volume depletion (e.g. hypotension, tachycardia, dry mucous membranes, FENa<1% and/or FEurea<35%).
- Be cautious with fluid administration in patients with severe hypoxemia
- If available, consult nephrology for patients with any of the following:
- Creatinine clearance <30 ml/min/1.73m2
- Oliguria: urine output <500cc/day or <0.5cc/Kg/hour
- Volume overload not improving with diuretics
- Hyperkalemia (>5.5) not responsive to dietary K restriction and diuretics
Estimates for RRT range from 0.8 to 5% of hospitalized patients (Guan et al; Zhou et al) in studies including floor patients. Among critically ill patients in the ICU, need for CRRT has been reported as high as 39% (Chen et al). Few studies have reported outcomes of RRT. One case series reported that out of 191 patients, 10 received CRRT, and all 10 died (Zhou et al). The nephrology consult service will determine the need, timing, and modality of renal replacement on a case-by-case basis. Indications for RRT in COVID-19 patients are the same as the indications for all patients.
Mild creatinine kinase elevation is relatively common with SARS-CoV-2 infection, with muscle pain and elevated CK occuring in 11-45% of hospitalized patients, depending on the study. It is more common in severe disease (23.9%, median CK 525 U/L) vs non-severe disease (5.0%, median CK 230 U/L) (Mao; Wang)
- Up to 10% of patients developed rhabdomyolysis complicated by acute renal failure (Chen). Other case reports of rhabdomyolysis in SARS have been published (Tsai; Huang; Wang)
The cause is likely to be some combination of direct viral myopathy, and critical illness/immobility myopathy (which can be made worse by corticosteroids, though do not discontinue them for this reason alone).
- Mild muscle injury: myalgias, proximal weakness, and/or mildly elevated CK (100s U/L)
- Rhabdomyolysis: myalgias, muscle weakness, myoglobinuria, moderate-marked elevation in CK (> 5x ULN, usually > 1500 U/L)
- Also check: CK, BMP, Phosphate, LFTs, TSH, UA, strength exam. Muscle biopsy is rarely likely to change management
- Mild muscle injury does not require specific intervention if renal function is normal
- Rhabdomyolysis carries risk of AKI, usually associated with CK >15-20k U/L. Increased risk of AKI in the setting of sepsis, dehydration, or acidosis (Bosch)
This section is in progress.
Updated Date: April 30, 2020
Depending on the severity of the COVID outbreak, some surgical centers are reducing or stopping elective cases. The following is based on former BWH Pre-operative decision pathway during severe outbreak and inclused suggestions about managing operating room screening/ testing.
If the case is elective:
- Reschedule for a future date.
If the patient has no COVID-19 symptoms or high-risk features:
- Proceed with standard precautions
- face shield, surgical mask
- double glove
- avoid contamination of work surfaces with secretions
- Concerning symptoms include:
- cough, sore throat
- shortness of breath, respiratory failure
- muscle aches, fatigue
- High-risk features include:
- difficulty differentiating symptoms from baseline in patients with thoracic or upper respiratory disease
- contact with known cases
If the case can not be delayed until COVID-19 test results are positive:
- Proceed with increased precautions
- head cover, face shield, N95
- double glove
- avoid contamination of work surfaces with secretions
- experienced provider intubating
- minimize providers entering and exiting the OR/theatre
- perioperative droplet precautions (patient masked, in isolation room)
If COVID-19 test results are available prior to surgery:
- If the patient has a single negative test,
- Proceed with increased precautions, as above.
- If the patient has a positive test,
- Reconsider if the case needs to be done.
- Consider delay for 14 days or repeat testing 24 hours after symptoms resolve.
- Ok to proceed after 2 negative tests 24 hours apart
- If delay may cause significant morbidity or mortality,
- Proceed with COVID-19 precautions
- all “increased precautions as above”
- patient should be on isolation precautions perioperatively
- all aerosol generating portions of the case should ideally be done in a negative pressure room
- full PPE for all providers in the room
Tracheostomy for patients with COVID is a clinical challenge, as it can expose health care practitioners to significant aerosols.
The timing of tracheostomy in any patient is complex. Patients should have adequate time to recover and avoid patient/provider risk, and the course of COVID-19 is longer than many pneumonias. However, earlier tracheostomy can make it easier to wean sedation and mobilize patients. Different institutions require different timeframes for tracheostomy eligibility. Early in pandemic many institutions were waiting a full 21 days, but now practice patterns are changing.
Tool: this consensus guidance covers timing and patient selection and management of tracheostomy
Both upper and lower endoscopy are considered aerosolizing procedures.
Tool: A comparison of guidance from endoscopic societies worldwide, including Wuhan, Hong Kong, Australia, Canada, US, UK, and European societies, can be found here: Lui et al.
Guidance from the US GI societies (AGA, ACG, ASGE, AASLD) can be found here:
- Joint GI Society, Clinical Insights for Gastroenterologists, March 2020.
- Joint GI Society, Guidance on Endoscopic Procedures, March 2020.
- Joint GI Society Statement, Statement on the Use of PPE in Endoscopy, April 2020.
Guidance from the European Society of Gastroenterology and Endoscopy Nurses and Associates can be found here:
Updated Date: December 11, 2020
In one large retrospective study, asthma appeared less prevalent in those with COVID than in those without it (6.75 vs 9.72%), and hospitalization rates did not appear very different between each group. As with all retrospective studies, these are somewhat limited by confounding (Green et al). One large study of about 44,000 people looking at risk factors for severity and mortality in China did not find asthma as a risk factor for disease severity. (X Li et al). Multiple smaller studies have shown similar results. It is not yet known if or why asthma may be protective against infection and/or not linked to worse outcomes.
This Green et al study also did not find any difference between those using ICS or LABA. ICS use in asthma has been dose-dependently associated with lower ACE2 and transmembrane protease serine 2 mRNA expression, but the impact of this on disease status is unknown. (Peters et al). It has also been posited that ICS may reduce airways inflammation and thus offer some protection (Carli et al). At this time we do not recommend treating patients with comorbid asthma and COVID any differently than you would normally, save for avoiding nebulizers (aerosol generating) and favoring MDIs where possible.
In one large metanalysis of 22 studies involving 11,000 patients, COPD was associated with three-fold higher mortality in COVID infected patients (OR 3.23, 1.59-6.57; P<0.05), but it was not more prevalent (5% of the patients(437/9337) than it is in the global population (9%) (Vankata et al). The severity difference in this study did not appear to be related to active smoking. Currently we we do not recommend treating patients with comorbid COPD and COVID any differently than you would normally save for avoiding nebulizers (aerosol generating) and favoring MDIs where possible.
Nebulizers can increase risk of transmission of COVID-19. Patients should avoid use. If nebulizer use is required, it should be done in isolation from others. COVID-19 virus may persist in droplets in the air for several hours.
The relationship between COVID-19 infection and smoking remains unclear. Metanalysis has shown that in terms of testing positive for disease, there may be a slight risk reduction in current smokers compared with nonsmokers (RR=0.73; 95% CI: 0.73–0.99). However, the evidence is not high-quality and may be confounded by other factors (testing and social behaviors may be very different in these groups).Test positivity risk amongst former smokers and never-smokers was similar (Farsalinos et al; Grundy et al). In most studies smoking is associated with more severe disease (OR rates in many metanalyses around ~2 (Grundy et al). Smoking poses multiple health risks and cannot be considered as a protective habit.
This section is in process
This section is in process
Updated Date June 19, 2020
Patients with febrile neutropenia (ANC < 500 cells/mm3 AND T ≥ 101F or T ≥ 100.5 for 1hr) should be worked up for COVID infection at the same time as they are evaluated for other infections. In patients with heme malignancy or SCT: findings are more subtle or absent in neutropenic and immune suppressed patients.
- Examine catheters (port, CVC, others) daily. Avoid rectal exams and any per-rectum therapies in neutropenic patients, but examine the perirectal area if symptoms or persistent fevers.
Tool: BWH guidance on neutropenic fever (workup, empiric antibiosis, line management, etc)
Based on early descriptive studies from China, patients with cancer - particularly those on active treatment for cancer - appear to have a worse prognosis. This includes higher prevalence, higher risk of severe disease, and higher risk of death from COVID-19 in patients with cancer compared to those without. (WHO-China Joint Mission on COVID-19, Yu et al). Prognosis for various cancers is highly variable, and the patient’s oncologist should be involved in goals of care conversations.
In additional labs to standard workup, if available we recommend also obtaining:
- Weekly glucan/galactomannan in neutropenic/transplant patients.
- Specific patient populations may require additional monitoring (such as CMV, EBV monitoring in transplant patients – ask primary oncologist).
Patients with solid tumors are at very high risk of thrombosis but at lower risk of infection than most heme malignancy patients. Prophylactic anticoagulation is particularly important in this setting.
- Hold pharmacologic prophylaxis if platelet count < 30K, use pneumoboots
Immune Checkpoint Inhibitors (ICIs) Most common ICIs are CTLA-4 inhibitor (ipilimumab) and PD-1/PD-L1 inhibitors (pembrolizumab, nivolumab, durvalumab, atezolizumab and avelumab). are not immunosuppressive when used alone, but the steroid dosages used to treat immune toxicities are often immunosuppressive. If patient develops organ dysfunction, it may be due to immune toxicity and not COVID. Please see BWH guidance for more information.
Updated Date: May 11, 2020
Tool: Please report cases at aad.org/covidregistry
Earliest data out of China with rash in 0.2% of patients, without morphologic description (Guan et al). Later data from Italy collected by dermatologists on the front lines showed rash in 18/88 patients (20.4%), excluding patients on new drugs (Recalcati). Morphology and type are varied. In one case study of 375 cases of rashes suspected to be COVID related (Galvan Casas et al) 47% with “other maculopapules” including perifollicular lesions, pityriasis rosea-like lesions, non-palpable purpura and palpable purpura. Lesions lasted for a mean of 8.6 days, and appeared with the onset of other symptoms. Associated with more severe disease (2% mortality) 19% of cases with pseudo-Chilblains (“COVID toes”) Typically appearing in younger patients. Lesions lasted for a mean of 12.7 days and were found later in the course of disease. Associated with less severe disease 19% with urticaria. Lesions lasted for a mean of 6.8 days, and appeared with the onset of other symptoms. Associated with more severe disease (2% mortality) 9% with vesicular eruption. Middle-aged patients. Lesions lasted for a mean of 10.4 days, and appeared before other symptoms. Associated with intermediate severity. and 6% with livedo/necrosis. Found in older patients with most severe disease (10% mortality). See BWH dermatology guidance for information on morbilliform rashes, urticaria, vasculopathies, livedo, and vesicular eruptions
Clinically “Covid toes” presents as erythematous to violaceous papules over acral surfaces (usually the fingers and toes, less commonly the nose and ears) following exposure to cold (fittingly called “acrocyanosis”). The mechanism is unknown, but likely due to Type I interferon-mediated complement activation and resultant microangiopathy (Kolivras). They often cause pain, itching, or paresthesias. Blistering, crusting, and ulceration can occur in severe cases, and they may be accompanied by livedo-like changes in adjacent skin. They are generally self-resolving within 1-2 weeks, however recurrence is possible with repeated cold exposure (Fernandez-Nieto et al; Mazzatto et al). Generally, these do not need extensive workup if they occur in the setting of known COVID. If they occur without clear explanation or are very severe, consider the following workup: CBC with differential, ANA, RF, Cold agglutinins, Cryoglobulins, C3, C4, CH50, CRP, ESR, D-dimer, Fibrinogen, Antiphospholipid antibodies
Treatment involves avoiding cold exposure, wearing socks, smoking cessation, and aspirin for a possible vasculopathic etiology (with caution in the pediatric population given the risk of Reye’s syndrome). In some cases you can use topical steroids, pentoxifylline, hydroxychloroquine, and calcium channel blockers.
Bed-bound patients are at risk for a variety of pressure injuries including erythema, skin breakdown, ulcerations, gangrene, or frank necrosis. Highest risk areas include sites of repeated pressure. In the context of COVID-19 and proning, facial pressure injuries have frequently been reported. Avoid pressure injury by:
- Minimizing pressure over bony prominences and face with dressings. Massachusetts General Hospital recommends the application of foam dressings to the upper chest/clavicles, shoulders, pelvis, elbows, knees, forehead, and dorsal feet. Gel pads may be used under the cheeks/nose (MGH, 2020)
- Turning the head and reposition arms every 2 hours.
- Frequently assessing for blanchable erythema, an early sign of pressure injury
Management of suspected skin cancer in the ambulatory setting is challenging during COVID (NCCN Covid Resources). If skin cancer is suspected, a high resolution photograph should be taken and dermatology telemedicine referral placed where available. Please see additional guidance from BWH guidelines.
Tool: Skin conditions that warrant referral for in-person evaluation (Medical Dermatology Society Guidelines)
See baseline immunosuppressants
Updated Date: December 11, 2020
This is covered Cytokine Storm Syndrome
This content will be covered in pediatrics shortly
COVID-19 can cause a number of symptoms that may overlap with those seen in rheumatologic diseases as outlined below. For patients with established rheumatologic disease who have confirmed or suspected COVID-19, careful evaluation will be required to determine if their symptoms are due to flare of the disease or are sequelae of viral infection.
- Arthralgias, Myalgias, Myositis
- Myalgia or arthralgia occur in approximately 15% of patients. 14.8% of patients (based on analysis as of 2/20/2020 on 55924 cases. (WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19) 2020
- Rhabdomyolysis is also a potential late complication of Covid-19. (Tong et al; Ronco et al). Please see Muscle Injury under Neuromuscular Disorders
- Lung Disease
- Pericarditis and Myocarditis
- Livedo Reticularis and Pernio- or Chilblain-like lesions of hands and feets (“COVID toes”)
- Coagulopathy, Lymphocytopenia, and thrombocytopenia
- Elevated levels of inflammatory markers. including CRP, ESR, and ferritin as well as elevated levels of cytokines including IL-1 and IL-6 (see Cytokine Storm Syndrome)
Patients with rheumatologic disease are not known to differ from other patients in terms of clinical presentation of COVID-19, though further information will be elucidated from the studies of the Global Rheumatology Alliance Registry. Early reports show that symptoms are similar in patients with rheumatic disease than in patients without. A report of 86 patients with immune-mediated inflammatory disease in New York reported high percentages of patients with fever (84%), cough (42%) and shortness of breath (41%), with lower rates of diarrhea, rhinorrhea and loss of taste and smell. Similar to reports from immunocompetent patients. (Haberman et al, NEJM, 2020) Similar findings were found in smaller studies of patients with rheumatic disease: 13 patients with chronic arthritis in Italy (Monti S et al) and 52 rheumatic disease patients in Boston (D’Silva K et al).
Studies looking at outcomes for COVID patients with rheumatic disease are limited to date, and major conclusions cannot be drawn at this time.
Incidence appears to be similar to the general population. In one prospective case series of 86 patients in New York City with a variety of immune-mediated inflammatory diseases. (Haberman et al). Hospitalization and mortality rates appear similar to the general population. In one comparative cohort study of 52 patients with rheumatic disease (75% on immunosuppressive medications) hospitalization and mortality rates were similar. However, rheumatic patients were more likely to require mechanical ventilation than healthy comparators, though the number of patients was low (11 patients [48%] vs 7 patients [18%], multivariable OR with 95% CI 1.07 to 9.05) (D’Silva K et al).
For management of patients on immunosuppressants see baseline immunosuppressants.
Updated Date: June 10, 2020
About 10% to 60% patients have GI symptoms on presentation incidence appears to be much higher in the US as compared to China (Sultan et al; Redd et al). GI symptoms may be the initial or predominant presenting symptom in 15-20% of patients (Redd et al). Very few patients have only GI symptoms throughout the course of illness (Pan et al). Anorexia is the most common gastrointestinal symptom followed by diarrhea and nausea. Vomiting and abdominal pain are not as common (Luo et al;Pan et al). In an analysis of the presenting symptoms and clinical outcomes of 318 adult patients with COVID-19 who were admitted to 9 hospitals across Massachusetts, 61% reported at least one gastrointestinal symptom (35% w/anorexia, 34% w/diarrhea, 26% w/nausea, 15% w/vomiting, and 15% w/abdominal pain) (Redd et al, Gastroenterology, 2020). Different mechanisms may mediate the broad array of GI symptoms seen in COVID-19. Proposed mechanisms include direct damage to the intestinal mucosa (Xiao et al), microthrombi in the lamina propria and submucosa (Bhayana et al; Ignat et al) and indirect alterations in mucosal immunity via the “gut-lung axis” (Pan et al)
- Diarrhea attributed to COVID-19 is often mild (mean # of bowel movements per day = 3-4, range = 2-10) and usually lasts for <1 week (mean # of days = 4-5, range = 1-14). (Han et al; Jin et al; Lin et al).
- Abdominal pain has been described as “stomachache, epigastric pain, and abdominal discomfort;” further characteristics have not been reported (Sultan et al).
- Treatment is largely supportive for diarrhea, nausea, and vomiting.
- If severe symptoms, a more extensive laboratory evaluation (such as lipase, amylase, lactate, stool studies, C Diff) or imaging (such as KUB, CT abdomen / pelvis, abdominal US +/- dopplers, or pelvic US) may be indicated.
Bowel infarction in COVID-19 appears to be rare, but cases reports are emerging (all in critically ill patients). In one retrospective study, abdominal imaging findings were analyzed in all patients admitted to the ICU (33%) or floor (67%) over a 2-week period. The most common indications to obtain a CT a/p were abdominal pain (33%) and sepsis (29%). Among the CT a/p scans performed in ICU patients (n = 20), the following findings were seen: fluid-filled colon in 65%, colonic or rectal thickening in 20%, small bowel wall thickening in 25%, pneumatosis or portal vein gas in 20% (n = 4), and perforation in 5% (n = 1). Among the CT a/p scans performed in non-ICU patients (n = 22), the following findings were seen: fluid-filled colon in 23% and colonic or rectal thickening in 14%; none of the non-ICU patients had findings of small bowel wall thickening, pneumatosis or portal vein gas, or perforation (Bhayana et al). Consider lactate and abdominal imaging in patients with severe abdominal pain. Management of thrombotic risk is discussed in Thrombosis.
Elevated liver biochemistries are seen in 15-70% of patients on presentation (incidence appears to be higher in the US as compared to China), and >90% of patients during the course of illness. Liver injury (defined as AST or ALT > 3x ULN; ALP and/or bilirubin > 2x ULN) is uncommon in presentation (Bloom et al). Acute liver failure has not been reported, even in those who are severely ill and in those with chronic liver disease (AASLD May 14, 2020). Both the incidence and degree of elevation in liver biochemistries are higher in severe disease (as compared to non-severe-disease) (Lei et al). Liver injury is associated with increased length of admission, need for ICU admission, and mortality (Fan et al; Lei et al; Hajifathalian et al). Abnormalities are predominantly hepatocellular and mild (even in severe disease), often rising during the course of illness. (Lei et al; AASLD May 14, 2020; Bloom et al)
- AST > ALT. AST and ALT elevations are usually 1-2x ULN on presentation. Levels are higher in patients with severe disease. May rise to >3x ULN in 40% of patients.
- GGT is often elevated, but ALP is normal.
Hypothesized pathways for liver injury include: a direct viral cytopathic effect (ACE2 is expressed on cholangiocytes and, to a lesser extent, hepatocytes); altered hepatic perfusion secondary to microthrombi; and cytokine-mediated injury (Zhang et al; Bloom et al). A limited number of post-mortem liver examinations have shown relatively non-specific findings, including: moderate microvesicular steatosis; mild, mixed lobular and portal activity; mild sinusoidal dilation with mildly increased small lymphocytes infiltration in sinusoidal spaces; and multifocal hepatic necrosis (Li and Xiao). In critically ill patients, liver injury may be secondary to ischemic / hypoxic hepatitis (“shock liver”); hepatic congestion; and cholestasis of sepsis. Hepatotoxic medications, such as Remdesivir, Hydroxychloroquine, and Tocilizumab, may also contribute.
In general, extensive workup and hepatic imaging is not needed in patients with asymptomatic, mild, hepatocellular-predominant elevations in liver biochemistries. If the patient has RUQ pain or cholestatic enzymes or AST or ALT >3x ULN; ALP or T-bili >2x ULN consider workup as below, and evaluate for drug toxicity. If the liver injury is mild and self-resolving. No specific therapy is typically needed (AASLD Clinical Insights, Updated May 14, 2020)
Hyperlipasemia, defined as a lipase level above the upper limit of normal, has been seen in 12-18% of patients. However, a lipase level >3 times the upper limit of normal appears to be rare (seen in only 2-3% of patients in whom lipase was checked) and no cases meeting diagnostic criteria for acute pancreatitis (as per the Revised Atlanta Classification) have been reported. Gastrointestinal symptoms are common in patients with hyperlipasemia (11-67% with anorexia, 56% with nausea, 11-57% with diarrhea, 33% with general abdominal discomfort) (McNabb et al; Wang et al).
In patients with elevated lipase and abdominal pain / tenderness, CT Abdomen Pelvis is recommended to clarify the differential, including pancreatitis vs enteritis / colitis vs bowel ischemia or obstruction vs cholecystitis or other hepatobiliary process (all of which can cause elevated lipase and abdominal pain) (Hameed et al). If not available, RUQ ultrasound may be helpful in some circumstances.
Chronic liver disease (including NAFLD) is a risk factor for severe disease and increased mortality. Cirrhotics and liver transplant recipients are at particularly high risk of death (Ji et al; Singh et al; Lee et al). The SECURE-Cirrhosis and COVID-Hep registries are tracking data on patients (throughout the world) with cirrhosis, chronic liver disease, and liver transplant who are infected with COVID-19 (Weekly Update, last updated May 4, 2020). Data thus far shows: Patients w/non-cirrhotic chronic liver disease: 18% required ICU admission and 6% died.Patients s/p liver transplant: 22% required ICU admission and 22% died.
Patients w/cirrhosis: 24% required ICU admission and 37% died. Unfortunately, poorer outcomes in cirrhosis are not unexpected. Among patients with ARDS of any cause, cirrhotic patients are known to have poorer outcomes (increased 90-day mortality) as compared to non-cirrhotic patients (Gacouin et al).
Tool: The American Association for the Study of Liver Diseases (AASLD) has constructed a ‘living document’ on best clinical practices in hepatology during the COVID-19 pandemic: AASLD "Clinical Best Practice Advice for Hepatology and Liver Transplant Providers during the COVID-19 Pandemic"
As of May 8, 2020, there were 1170 reported cases of COVID-19 in IBD patients. Of these, 32% required hospitalization, 6% required ICU admission, 5% required mechanical ventilation, and 4% died. Among the 1170 cases, 58% were in remission. Among those in remission, 28% required hospitalization; 30% and 44% of those with mild and moderate-severe disease activity required hospitalization, respectively. (SECURE-IBD Registry) Patients on prednisone (> 20 mg daily) are likely at increased risk of COVID-19. It is unclear if the risk and severity of infection are increased in patients on thiopurines (azathioprine, 6-mercaptopurine), methotrexate, anti-TNF therapies (infliximab, adalimumab, certolizumab, golimumab), vedolizumab, ustekinumab, and the JAK inhibitor tofacitinib (Rubin et al).
Updated Date July 13, 2020
Neurologic manifestations may occur in 36.4%-69% of hospitalized COVID-19 patients (Mao; Helms). Severely ill patients are more likely to have neurologic symptoms (45.5% severe vs. 30.2% non-severe), stroke (5.7% severe vs. 0.8% non-severe), impaired consciousness (14.8% severe vs. 2.4% non-severe), and skeletal muscle injury (19.3% severe vs. 4.8% non-severe) (Mao) Manifestations can include:
- Delirium, confusion, or executive dysfunction (Helms). 69% of patients displayed agitation; 65% of patients assessed with CAM-ICU had confusion; 33% of discharged patients had inattention, disorientation, or poorly organized movements to command
- Smell or taste abnormalities (see Anosmia)
- Corticospinal tract signs (67%) (Helms)
- Dizziness (16.8%) (Mao)
- Stroke (2.5-5%) (see Stroke).
- GBS, Miller Fisher syndrome (case reports) (see BWH guidance on neuromuscular disorders)
- Encephalitis, acute necrotizing encephalopathy, myelitis, CNS demyelinating lesions (case reports) (see Meningoencephalitis)
Pathophysiology: Illness from SARS-CoV-2 can provoke states that increase risk of neurological disease. The pathophysiology of the various neurological manifestations of COVID-19 is currently unknown, but possible mechanisms include:
- Direct viral invasion of the nervous system, with potential transsynaptic spread
- Theoretical possibility of blood-brain barrier disruption secondary to SARS-CoV-2 binding to angiotensin-converting enzyme 2 (ACE2)
- Autoimmune sequelae
- Ischemic injury via systemic hypoxia or local vascular endothelial inflammation or thrombosis
- Toxic metabolic encephalopathies
- Long term impact of the systemic proinflammatory state
(Zubair et al).
Rates of AMS (including delirium) have been relatively high (7.5-66%) in COVID, with variability likely related to differences in assessment of mental status and definitions of deficits. In a series of patients in Strasbourg, France, 66% displayed agitation, 65% of patients assessed with CAM-ICU had confusion, and 33% of discharged patients showed dysexectutive function (Helms). Delirium, characterized by waxing and waning arousal and impaired attention, is common in hospitalized patients of advanced age and with multiple comorbidities. One study of ICU patients (before the COVID pandemic) showed that 83.3% of patients develop delirium. Delirium was present in 39.5% of easily arousable patients and persisted in 10.4% of patients at discharge (Ely).
Encephalopathy in patients with COVID-19 may be caused by systemic infection, toxic-metabolic derangements (hypoxemia, hypercarbia, renal or hepatic dysfunction, nutritional deficiencies) or medication effects (sedation, cephalosporins/ quinolones), or primary CNS dysfunction (e.g. seizure, stroke). Delirium can happen even in the absence of these conditions due to sleep/wake disturbances and psychological stress.
Workup: Recommend performing a general workup for AMS as below. If these are unrevealing and the patient has significant abnormalities, in some patients it may be worth pursuing MRI brain for structural etiologies such as encephalitis or stroke (see Meningoencephalitis and Stroke sections), EEG for seizure, or LP for signs of meningitis or unexplained neurologic findings.
- Treat specific causes as discovered in work-up
- Treat for Delirium
- Detailed guidelines regarding ICU treatment of sedation, pain, agitation, and delirium can be found in BWH’s guidance.
Changes in smell and taste perception have been reported in many patients with COVID-19. A meta-analysis of 10 studies (1627 patients) demonstrated olfactory dysfunction in 53% and gustatory dysfunction in 44% of COVID-19 patients (Tong). Anosmia may precede COVID-19 diagnosis (Kaye), and when anosmia/ageusia occur they most frequently precede hospitalization (Giacomelli).
Pathophysiology is unknown. There is some evidence that supports direct neural invasion of the virus. Retrograde neuronal transport to CNS through peripheral nerves is documented in other viral illnesses rabies, HSV, murine counterparts to coronavirus (Perlman) and there are case report of MRI FLAIR hyperintensities in the olfactory bulbs and right rectus gyrus in COVID-positive patient presenting with isolated anosmia (Politi). However, ACE-2 is expressed in nasal epithelium, but not in olfactory sensory neurons, indicating epithelium may be entry site (Gengler; Zubair).
Recovery: 66-80% of patients with COVID-19-associated smell impairment report spontaneous improvement or resolution within days to weeks of recovery from clinical illness (Yan; Lechien; Hopkins; Vaira). In a study of 3191 patients, median time to recovery for anosmia and ageusia was 7 days (Lee).
Management: No indication for corticosteroids to treat hyposmia/anosmia, as it frequently recovers without intervention
Encephalitis the inflammation of the brain parenchyma secondary to infection or autoimmune conditions. Diagnostic criteria (Venkatesan): AMS > 24 hours without alternative cause and 2 (possible encephalitis) or 3 (probable encephalitis) of the following: (a) fever > 100.4F within 72 hours of presentation, (b) generalized or partial seizures, (c) CSF leukocyte count > 5, (d) abnormality on imaging that is new and consistent with encephalitis. meningitis is the inflammation of the membranous coverings of the brain secondary to infection or autoimmune conditions, myelitis is the inflammation of the spinal cord parenchyma secondary to infection or autoimmune conditions, acute necrotizing encephalopathy (ANE) a rare type of brain injury that usually follows an acute febrile illness (potentially a post-infectious autoimmune condition) characterized by symmetric multifocal brain lesions, without inflammatory cells in the brain parenchyma (Poyiadji). and acute disseminated encephalomyelitis (ADEM) an immune-mediated inflammatory disorder characterized by wide-spread demyelination of the brain and spinal cord have all been reported with COVID infection.
MRI and CSF findings: MRI brain abnormalities are commonly present in patients with severe COVID with encephalopathy or neurologic symptoms, and display some characteristic findings. Of 13 patients with encephalopathy of unclear etiology, 8 (62%) displayed leptomeningeal enhancement. 100% of 11 patients who had perfusion imaging showed bilateral frontotemporal hypoperfusion (Helms). Of 27 ICU patients with neurologic symptoms, 10 (37%) had cortical FLAIR abnormalities (of these, 7 demonstrated cortical diffusion restriction, 5 had subtle leptomeningeal enhancement, and 3 had subcortical or deep white matter signal abnormalities) (Kandemirli). In a study of 37 patients with severe COVID with neurologic manifestations and abnormal MRI, 3 main patterns were seen: 43% had signal abnormalities in the medial temporal lobe, 30% had nonconfluent multifocal white matter hyperintense lesions with variable enhancement (the majority of these with associated hemorrhagic lesions), and 24% had extensive white matter microhemorrhages (Kremer et al). CSF abnormalities may be present patients with neurologic symptoms requiring LP, typically with elevated protein and variable pleocytosis, rarely positive PCR (see below). In study of 22 children with COVID and encephalitis in Wuhan, 10 had CSF pleiocytosis and 8 had elevated CSF protein (Li et al). In a study of 7 adults patients (unclear clinical presentation), CSF showed no pleocytosis in all patients, there was elevated CSF protein in 1 patient, and CSF RT-PCR for SARS-CoV-2 was negative in all patients (Helms).
Evidence suggestive of direct CNS invasion: Rarely CSF has been noted to be positive by PCR (in 2 of 578 samples in one study, but not at levels that are infectious, (Destras et al). Additional case reports: (Moriguchi; Hanna Huang; Xiang et al) SARS-CoV-2 was also identified in 8/22 patient brains in one series by RT-PCR (Puelles). A case report identified SARS-CoV-2 viral particles on electron microscopy of the frontal lobe, in endothelial cells and neural cell bodies (Paniz-Mondolfi).
Evidence of Autoimmunity: Autoantibodies have been noted suggesting some patients may have an autoimmune meningoencephalitis (Lucchese). These are largely against unidentified neural autoantigens, though there have been a couple case reports of COVID-associated NMDA receptor encephalitis (Panariello et al; Monti et al).
Work-up and management: Consult neurology where possible for guidance.
In early studies, the frequency of seizures appears low (<1%) in COVID-19 patients relative to other coronaviruses (8-9% for MERS-CoV and other HCoV (Saad; Dominguez). In a series of 214 patients, 1 patient had a generalized seizure lasting 3 minutes (Mao). Case reports exist of COVID-19 positive patients developing new-onset seizures (Moriguchi; Xiang; Duong; Zanin; Sohal; Bernard-Valnet; Karimi) though in most of these cases, seizures were presumed to be secondary to unmasking of an underlying seizure disorder. Limited evidence to date does not suggest that patients with epilepsy are at higher risk of COVID-19 infection or severe disease manifestations (French).
Work-up and management:
- General seizure work-up and management regardless of COVID-19 status
- Note that convulsive seizure should be considered aerosol generating, providers should don appropriate PPE
Patients with COVID-19 may have an increased risk of stroke related to a systemic inflammatory and prothrombotic state (Klok), possible endothelial dysfunction, and/or medical comorbidities. A systemic review of 39 studies comprising 135 patients (hospitalized, many with severe COVID) found a pooled incidence of acute ischemic stroke of 1.2%, with onset 10+/-8 days after COVID symptoms began. Most of these strokes were large vessel thrombosis, embolism, or stenosis (62%) or multiple vascular territory embolic (26%), rather than small vessel. 38% of patients died (Tan et al).
Workup and management: Protocols for acute stroke work-up and management remain largely the same as for non-COVID patients: IV tPA, endovascular therapy, and 2018 AHA guidelines. Consult neurology for any acute stroke evaluation.
- tPA increases D-dimer levels and decreases fibrinogen levels for at least 24 hrs (Skoloudik). D-dimer should not be used for COVID-19 prognostication post-tPA.
- Given likely hypercoagulable state (see Thrombotic Disease) in many COVID-19 patients, consider therapeutic anticoagulation for confirmed stroke in a COVID-19 patient if stroke mechanism is unclear (discuss with neurology)
Guillain Barré syndrome (GBS) (inclusive of acute inflammatory demyelinating polyneuropathy [AIDP], acute motor axonal neuropathy [AMAN], and acute motor and sensory axonal neuropathy [AMSAN]) have been reported in COVID-19 positive patients. Rare cases have been reported of neuromuscular symptoms presenting prior to COVID-19 symptoms, or as late as 24 days after COVID-19 symptoms.
Tool: BWH management guide for GBS in COVID-19 positive patients.
Patients with MG may be at higher risk of contracting COVID-19 or developing severe disease because they are often on immunosuppressive therapies and have respiratory muscle weakness. In a series of 15 hospitalized patients in Brazil, 13 developed an exacerbation during their hospitalization and 73% needed mechanical ventilation and 30% died (Camelo-Filho).
- While we do not recommend them in general for COVID treatment, these medications are known to exacerbate MG flares: Chloroquine, hydroxychloroquine, azithromycin
- Discuss whether or not to hold immunosuppressants with the patient’s prescribing physician
Tool: BWH management guidelines for all inpatients with Myasthenia Gravis
It is not known whether patients with MS have increased incidence of COVID-19. Patients with MS may be at higher risk of infection in general, including pneumonia and influenza, but do not appear to be at higher risk of all upper respiratory infections (Wijnands; Brownlee; Willis). It is not known if MS patients with COVID-19 have a more severe course of disease. In a French cohort, 3.5% of 347 patients died and 21% had severe COVID, though the study was registry-based (Loupre et al). Data from 1540 MS patients across 21 countries showed that progressive MS and worse disability score were associated with worse outcomes, and that anti-CD20 therapies were linked to increase risk of artificial ventilation but not death (Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS), First results of the COVID-19 in MS Global Data Sharing Initiative suggest anti-CD20 DMTs are associated with worse COVID-19 outcomes, 22-Sep-2020. See here.
Treatment: Please see baseline immunosuppression for more information about the risk of infection associated with disease modifying therapies. In general, during mild viral infections DMTs are usually continued (Brownlee) but in severe disease consider temporarily suspending or delaying certain high-risk immunosuppressive DMTs. Recommendations by DMT type have been published - see for instance the European Academy of Neurology for management of patients with neurological diseases during the COVID-19 pandemic. Do not change treatment without contacting the prescribing physician.
Tool: Report cases to COVIMS, a de-identified patient data repository.