Neither acute nor long-term complications of COVID-19 infection are well described at this point in the outbreak. It is clear that the overarching clinical pattern largely follows that of other viral respiratory infections, with upper or mild lower respiratory symptoms in mild cases, and pneumonia, acute lung injury, ARDS, multiorgan dysfunction, and septic shock in more severe cases. The during-hospitalization and post-hospitalization sequelae beyond these patterns are poorly described, but below is a brief description by system of COVID-19 complications described in the literature to this point.
Key Points
- Acute cardiovascular
complications appear to be the most common and concerning sequelae:
- Acute myocardial injury (7-17% of hospitalized patients in one study),
- Myocarditis (primary cause of death in 7% of COVID deaths in one study),
- Arrhythmias (16.7% of hospitalized and 44.4% of ICU patients in one study), and
- Venous thromboembolism (incidence unknown) are reported.
- Concerns for sudden cardiac death, even after recovery, have been raised but are not well documented in the literature. Proposed mechanisms include respiratory compromise, myocarditis, malignant tachydysrhythmias, heart failure, and coronary plaque instability (i.e. Type 1 MI) secondary to inflammation
- Co-infection and secondary infection rates are unknown but estimates range from 4.8% to 21%, with higher rates in sicker patients. Viral co-infection is more common than bacterial co-infection, but both may be seen
- Cytokine release syndrome and secondary HLH are both described complications, but their incidence is unknown. The relation of this finding to purported benefits of tocilizumab is unknown.
- Other extrapulmonary complications are relatively typical of sepsis and generalized infections, such as kidney injury, abnormal LFTs, and delirium
Cardiovascular Complications
Known cardiovascular complications include acute myocardial injury (7-17% of hospitalized patients in one study), myocarditis (primary cause of death in 7% of COVID deaths and contributor in 33% of COVID deaths in one study), arrhythmias (16.7% of hospitalized and 44.4% of ICU patients in one study), and venous thromboembolism (incidence unknown). The mechanisms of these effects are not well known, but it is suspected that direct viral injury to the myocardium and inflammatory effects on the coagulation and coronary vascular systems may be involved.
Cardiomyopathy is reported to occur in 33% of COVID infected patients requiring ICU care per one study, and is noted in several other studies. It is unknown whether this represents unmasking of an underlying LV dysfunction, or new damage to cardiac function. Right heart dysfunction and pulmonary hypertension are also reported as would be expected in conditions of massive pulmonary insult and ARDS.
Type 1 myocardial infarctions may occur during or after the infection at a higher rate than baseline. It is suspected that this is due to the role of inflammation on coronary plaques and thrombosis risk. This presents a great diagnostic and therapeutic challenge in the setting of STEMI when COVID infection is also suspected. How to workup these patients, whether to obtain immediate catheterization, and many other questions remain. The relative contribution to myocardial injury from Type 1 insult versus more diffuse processes (e.g. Type 2 effects from vasodilatory shock) is not known.
Multiple studies have noted palpitations to be a presenting symptom in COVID patients, although typically only a minority of patients report palpitations. Routine telemetry monitoring may be considered in hospitalized COVID patients, although the benefit of this is unknown, and it does have the potential to complicate infection control practices.
QT prolongation is a known side effect of certain therapeutics involved in COVID, most notably chloroquine, and risk of torsades de pointes should be considered as well.
There are no data available in SARS-COV2 patients regarding long term cardiac sequelae, but studies of SARS-COV (aka SARS) patients at 12 years of follow up demonstrate increased rates of dyslipidemia, glucose metabolism disorders, and other cardiovascular conditions.
Pulmonary Complications
The primary pulmonary complications are consistent with other viral respiratory infections, and are well described in COVID. Rates of ARDS range widely from 3.4% to 95.2% depending on the population studied (all known infected patients vs ICU admitted patients), and are likely further influenced by variation in testing practices. Among ICU patients, one study from Washington state of 21 patients admitted to a critical care unit reported that 71% of patients required mechanical ventilation. There have been anecdotal reports of higher than expected failure rates of HFNC and NIPPV, but there does not appear to be strong literature documenting this finding.
Renal Complications
As would be expected with a condition that can cause overwhelming sepsis, acute kidney injury and acute renal failure have been reported. Data from China has suggested an AKI rate between 0.5% and 15% in all comers, and the Washington study showed a rate of acute renal failure 0f 19.1% in ICU patients. There does not appear to be good data on whether nephrologic management (e.g. early vs late RRT) should differ in COVID as compared to other septic conditions.
Hematologic Complications
Abnormal coagulation parameters, progressing to full blown DIC in sicker patients, have been reported in multiple studies. Most notably D-dimer and fibrin split products are frequently abnormal in COVID patients requiring ICU care.
Cytokine release syndrome and secondary HLH are reported to be known complications per the Surviving Sepsis guidelines, but an incidence is not reported. Of note, tocilizumab is frequently among the considered therapies for COVID infection, and it is not clear whether the purported benefit relates to these hematologic complications. There do not appear to be good data on the role of ferritin or other hematologic markers in monitoring for these complications.
Venous thromboembolism is also reportedly more common in COVID patients than non-COVID patients, although the relative risk is not clear. There does not appear to be literature discussing whether alternative VTE prophylaxis regimens are needed in COVID patients.
Infectious Disease Complications
The incidence of co-infection and secondary infection are much debated in COVID. This is particularly important as some health systems are using a positive influenza test or multiplex respiratory viral panel to rule out COVID-19 infection. The Washington study of 21 ICU patients reported a 4.8% bacterial co-infection and 14.3% viral co-infection rate. Other studies have reported co-infection rates from 5.8 to 15%.
The risk of reinfection after recovery is unknown. Prior data from MERS-CoV and SARS suggest at least some degree of immunity after recovery, but it is not known if this will apply to COVID-19.
Gastrointestinal and Hepatologic Complications
Generalized GI symptoms (typically nausea, diarrhea, and abdominal cramping) have been reported to be less common, but not infrequent elements of the presentation of COVID infection, similar to other viral illnesses.
Consistent with other shock and septic states, abnormal liver function tests are reported (14.3% of ICU patients in one study, but up to 62% in other datasets), and are likely related to hepatic hypoperfusion or congestive hepatopathy secondary to cardiomyopathy. It is suspected that viral invasion of liver and other organs occurs, but the role of these findings in hepatic injury is unknown.
Neurologic Complications
There is shockingly little data on neurologic sequelae of COVID infection. The Washington study reported a 4.8% incidence in their population of 21 ICU patients, but this is not well documented elsewhere. Anecdotal reports of delirium consistent with other septic states exist and are pathologically plausible.
HEENT Complications
There are scattered reports of anosmia (loss of sense of smell) and dysgusia (abnormal taste sensation) as a presenting complaint in COVID infection. There have reportedly been otolaryngolongy societies that have gone as far as recommending COVID testing in patients presenting with these symptoms but no other complaints. The clinical course and long term prognosis of these HEENT symptoms is not well documented.
Pharmacologic Complications
It should not be forgotten that pharmacotherapies carry their own risks of complications and may be responsible for atypical clinical patterns seen in COVID infection. In particular, keep in mind that chloroquine is associated with QT prolongation as well as hemolytic crisis in G6PD patients and that lopinavir/ritonavir (no longer recommended) has several drug-drug interactions, especially with DOACs, P2Y12 inhibitors, and statins.
References
The annotated list of sources reviewed will be provided in a separate post.
David Gordon, MD