The complement system is an important component of innate immunity that is essential to respond rapidly to infection. During inflammation, both acute and chronic, activation of the complement system promotes the elimination of pathogens. Dysregulation of the complement system may lead to acute lung disease after a highly pathogenic viruses infections (19).

Complement activation through the lectin pathway has already been described in SARS-CoV-2 infected patients. Gao et al. (20) showed that blocking this pathway ameliorates lung injury induced by SARS-CoV and MERS-CoV in mice. Autoptic lung samples and skin biopsies from patients with severe COVID-19 showed a deposition of mannose binding lectin (MBL)-associated serine protease (MASP)2, C4d, and C5b-9 (MAC component), suggesting an activation of the complement system through the alternative and the lectin pathways (21). Preliminary data from patients treated with an anti-complement C5a blocking antibody also suggested a potential benefit of complement targeting therapies in COVID-19 patients with severe lung injuries (20). Due to the implicated role of complement in the pathogenesis of acute lung injury and ARDS, ongoing clinical trials are testing the hypothesis that blocking the complement cascade ameliorates disease severity in COVID-19 patients.

In general, altered hemostasis due to viral infections often leads to vascular complications like thrombosis and or hemorrhage. Vascular and endothelial changes by the virus as well as inherited host factors help define the clinical presentation. Some viral contributions to the ischemic and thrombotic environment include procoagulant factors, hemodynamic changes, and pro-inflammatory cytokines. The pro-inflammatory response also directly induces plaque rupture (22–25). This can be seen with many respiratory viruses like H1N1 influenza (26) and is of increasing concern for patients with severe COVID-19. The majority of severe COVID-19 patients show signs of increased coagulation activity, resulting in consumption of coagulation factors and disseminated microvascular thrombosis. Hypoxia associated with COVID-19 pneumonia increases blood viscosity both directly and through hypoxia-inducible transcription factor-dependent signaling pathway thereby promoting thrombosis (27). Laboratory results indicate a prolongation of the prothrombin time and international normalized ratio (PT/INR) and partial thromboplastin time (PTT), elevation of D-dimer, decreased fibrinogen levels, thrombocytopenia, and schistocytes in peripheral blood smears. In one study of 449 COVID-19 patients, 71% of non-survivors had diffuse intravascular coagulation (DIC) compared to 0.4% of survivors (27). Zhou et al. (25) showed that death is associated with D-dimer <1 μg/mL in a retrospective study of 191 COVID-19 patients. Elevated D-dimer at admission and markedly increasing D-dimer levels during the disease were associated with high mortality (27).

Recently, anti-phospholipid antibodies and infarcts in multiple vascular territories have been reported in three COVID-19 patients with thrombocytopenia. It is known that these antibodies can increase during several infections, and critical illness and may lead to thrombotic events (28).

Based on this data, a prophylactic dose of low molecular weight heparin has been advised to hospitalized COVID-19 patients, despite abnormal coagulation tests, and the thrombotic risk associated with prolonged bed resting (27).

The exact mechanism of kidney involvement is unclear and likely multifactorial. Kidney disease may be caused by SARS-CoV-2 binding to the ACE2 receptor on kidney cells that allows the virus to enter (13, 52, 53). Moreover, normal kidney and intestinal tract have higher ACE2 expression than lung tissue (54). Detection of coronavirus in the kidneys and urine of patients with SARS-CoV and SARS-CoV-2 supports the theory that the virus can directly damage the kidneys (5). However, in another study, no urine tested positive for viral RNA in 72 samples (55). Preliminary evidence in postmortem examinations of kidney tissue from six patients found severe acute tubular necrosis and lymphocyte infiltration. Additionally, SARS-CoV-2 nucleocapsid protein (NP) has been detected through immunohistochemistry in kidney tubules (56). Moreover, in one autopsy of a kidney transplant patient who died of COVID-19, viral inclusion structures were found in the endothelial cells of the kidney (57). Viral infection could induce tubular damage through the deposition of the MAC complex (the final step of the complement cascade) on tubules and infiltration of CD68⁺ macrophages in the tubule-interstitium (56). Diffuse damage in proximal tubules with the loss of brush border, vacuolar degeneration, and even necrosis were observed in a study of 26 autopsies of patients with COVID-19 (58). In the peritubular and glomerular capillary of these autopsies, diffuse erythrocyte aggregation with endothelial damage, and obstruction without fibrin thrombi or distinct fragmentation of erythrocytes or platelets were observed (58). Clusters of SARS-CoV-2 was found with electron microscopy in the tubular epithelium and podocytes (58). However, this finding could be non-specific as the presence of viral proteins may not represent direct damage mediated by the virus and instead indicate clathrin-coated vesicles (59, 60). Puelles et al. (61) showed the presence of SARS-CoV-2 RNA and proteins in all kidney areas, especially in glomerular cells in autopsies of three of six COVID-19 patients.

Kidney biopsies in two Afro-American patients with high-risk APOL1 genotype and COVID-19 infection showed a collapsing focal segmental glomerulosclerosis (62, 63). However, SARS-CoV-2 RNA was not detected in the kidney tissue raining the intriguing hypothesis that cytokine storm increased APOL1 expression leading to podocyte injury (62, 63). Other indirect mechanisms that potentially lead to tubular injury are sepsis, cytokine storm syndrome, shock/hemodynamic instability, rhabdomyolysis, and hypoxia of kidney tissue (5, 48).

Kidney biopsies would be helpful to better understand the histologic pattern of injury (tubular, glomerular, and vascular) and the pathogenesis that could led to AKI. Unfortunately, this is very difficult to obtain given the respiratory and hemodynamic instability of AKI patients and the use of anticoagulation which increases the risk of bleeding. In addition, non-essential procedures in infected patients are not being done in most hospitals given significant risk of exposure to personnel.

Patients with chronic kidney disease (CKD) are known to have a higher risk of upper respiratory tract infection and pneumonia due to their persistent proinflammatory state with functional defects in innate and adaptive immunity (5). No study so far has found that chronic kidney disease is statistically correlated with severe COVID-19. However, a significant association of CKD with severe COVID-19 was observed when data of different studies were combined (64).

Chronic immunosuppression is a well-known risk factor for viral and bacterial infections, but it is also crucial to prevent graft rejection and to contrast the uncontrolled antiviral inflammatory response. Therefore, the transplant community is puzzled in trying to understand the best therapeutic approach, in the absence of any strong clinical data (8).

So far, initial presentation in transplant recipients has been reported heterogeneous to other hosts and many patients did not report contact with infected individuals. Common symptoms at disease onset have been fever, cough, asthenia, myalgias, and diarrhea (65). In a study of 36 transplant recipients, however, fever was less common than in general COVID-19 patients (66). In multiple series, transplant patients show numerous radiopacity and patchy shadows on chest radiographs often at presentation (67–71). However, in one case series of 15 kidney transplant patients, 33% had no acute radiographic findings (72). Laboratory exams often showed lymphopenia with lower CD3, CD4, and CD8 T cells especially in those patients who had received antithymocyte globulin in the weeks before the infection (66).

Until more data is available, the rules to prevent viral infection in the general population apply to transplant patients (hand hygiene, sanitization, social distancing, and avoiding areas where infected patients could be present) (73). Transplant patients with potential COVID-19 infection should not access the transplant center due to risk of viral spread.

According to the European Renal Association—European Dialysis and Transplant Association (ERA-EDTA) guidelines (74), in patients with COVID-19 and without pneumonia, complete withdrawal of immunosuppressants -particularly calcineurin inhibitors (CNI)- is discouraged. Reduction of the dosage of CNI, and withdrawal of mycophenolate, azathioprine, or mTOR-inhibitors should be individualized considering the severity of the disease (74). The concurrent use of antivirals and anti-inflammatories should be carefully considered with attention to drug-drug interactions that may affect the half-life of immunosuppressant drugs (74).

In critically ill patients, withdrawal of immunosuppression could be done while converting those patients to hydrocortisone/solumedrol. This approach may improve viral clearance but could lead to immune reconstitution and kidney's rejection (69, 75). It should be considered that reducing immunosuppression may exacerbate inflammation, so this approach should be cautioned in the absence of anti-inflammatory agents (see below).

In some patients, tacrolimus reduction may be preferred over complete withdrawal (67, 69) because of direct alleged CNI antiviral properties or CNI anti-inflammatory action (76, 77). Pending the results of clinical studies, CNI withdrawal vs. reduction may be established on a case-by-case basis depending on the severity of pneumonia.

Although it is known that diabetic nephropathy is an important comorbidity and AKI is one of the main risk factors for poor outcome during COVID-19 infection, the impact of the infection on other kidney diseases, like end-stage renal disease, is still unclear (8).

Wang et al. (78) described the outbreak of COVID-19 in the hemodialysis (HD) center of Renmin Hospital, Wuhan University. They identified COVID-19 in 37 individuals among 230 HD patients (16.09%) and 4 individuals among 33 staff (12.12%). They presented mostly mild symptoms, and no one required admission to the ICU. During the observation, 7 HD patients died −6 with COVID-19 and 1 without- but deaths were not directly related to pneumonia. The causes of death, in fact, were heart failure, hyperkalemia, and cerebrovascular disease (not peer-reviewed) (79). In a report of five HD patients, diarrhea was the most common symptom, whereas fever, cough, and dyspnea were not present, thus making the diagnose harder (80). In one retrospective multicenter study of 7,154 hemodialyzed patients, about 2% were confirmed having COVID-19 infection and only about 50% of them presented fever while about 20% of patients were asymptomatic (81). The mortality rate among these patients is greater than the general population with COVID-19 peaking at about 31% (81). The same mortality rate was reported in another retrospective study of 59 dialyzed patients (2 on peritoneal dialysis and 57 on hemodialysis) (82).

Circulating CD4 and CD8 T cells, NK cells, and proinflammatory cytokines are significantly lower in COVID-19 HD patients, compared to non-HD COVID-19 individuals (79). Consistently, HD patients infected by SARS-CoV-2 are more likely to present mild symptoms with lower risk of developing ARDS compared to COVID-19 patients not on HD (79). However, the reduced inflammatory response in HD patients suggests that they may be at higher risk of being infected with SARS-CoV-2. Therefore, additional prevention measures are essential in managing the epidemic in HD centers (8). The impaired immune response in HD patients correlates to longer time to clear the virus requiring longer time in isolation corresponding with the outbreak of SARS in 2003 (79, 83).

An interim guidance for outpatient HD facilities has been recently released by the Centers for Disease Control and Prevention (CDC) (84). Early recognition and isolation of individuals with respiratory infection, isolation of infected patients from other hemodialyzed patients, and the use of personal protective equipment are high priority (85). During routine clinical visits, face masks, and eye shields are sufficient, while during high-risk procedures, N95 respirators and other respiratory protection devices are required. Chinese Society of Nephrology and the Taiwan Society of Nephrology have recently published detailed guidelines for managing COVID-19 outbreaks in dialysis units (8, 48).

Patients on peritoneal dialysis should be managed from home, using telemedicine assistance or other systems for communication whenever possible (86).

Early reports, mainly from China, suggested that over 60% of severely ill COVID-19 patients presented with additional organ dysfunction syndromes (119). This has been, at least in part, related to a sepsis-like syndrome induced by high levels of circulating cytokines. Cytokine storm may be induced by a superimposed septic syndrome or by the direct effect of the virus on the infected host (119). It has been suggested that anti-inflammatory drugs may ameliorate COVID-19 infections.

The receptor for SARS-CoV-2, angiotensin-converting enzyme 2 (ACE2), is a membrane-bound aminopeptidase whose function is to cleave angiotensin I and angiotensin II into two peptides, angiotensin-(1–9), and angiotensin-(1–7) (8). Angiotensin-(1–9) has cardiovascular-protective effects and angiotensin-(1–7) acts as vasodilator and has anti-proliferative, and anti-fibrotic function (8, 125, 126).

Experimental animal models showed that, although ACE inhibitors and ARBs do not directly affect ACE2 activity, these agents can upregulate the expression and activity of the receptor in heart and kidney tissue (127). These drugs are commonly prescribed for patients with diabetes or cardiovascular disease who have higher risk of severe COVID-19 disease which has raised concern for the use of these drugs during the infection (8). However, the increased expression of ACE2 could limit the viral spread because it is not accompanied by an increase of TMPRSS2, so the virus could be tied to the receptor but could not enter the cells (8). Another possible protective role of ARBs in the lungs is mediated by angiotensin-(1–7) that has anti-proliferative and anti-fibrotic activity (8). Experimental animal models of acute lung injury, including a model of SARS-CoV infection, showed that ARBs reduce Ang II-mediated lung damage and therefore attenuate COVID-19 infection (127). In a retrospective study of 6,272 patients with COVID-19 and matched COVID-19 negative controls, infected patients were associated with increased use of RAAS inhibitors, but these drugs did not correlate with a more severe disease after multivariable adjustment (105). The study by Reynolds et al. (106) did not find any correlation between the use of anti-hypertensive drugs, such as RAS inhibitors (both angiotensin-receptor blockers and ACE inhibitors), calcium-channel blockers, beta-blockers or thiazide diuretics and the risk of infection or of having a severe disease in 5,894 patients.

The European and American Cardiology Societies, remarked that there is no evidence of a noxious effect of RAS inhibitors in patients affected by COVID-19 (128, 129). Moreover, discontinuation of RAS inhibitor both in healthy and infected individuals could be dangerous, especially in high-risk patients, so they suggest to maintain the ongoing therapy (130).