Overview of acute kidney manifestations and management of patients with COVID-19

Abstract

Since the start of the COVID-19 pandemic, several manifestations of kidney involvement associated with infection of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus have been described, including proteinuria, hematuria, and acute kidney injury. A growing body of literature has explored the risk factors and pathogenesis of COVID-19-associated acute kidney injury (AKI), including direct and indirect mechanisms, as well as early postdischarge outcomes that may result from various manifestations of kidney involvement. In this review, we explore the current state of knowledge of the epidemiology of COVID-19-associated AKI, potential mechanisms and pathogenesis of AKI, and various management strategies for patients in the acute setting. We highlight how kidney replacement therapy for patients with COVID-19-associated AKI has been affected by the increasing demand for dialysis and how the postacute management of patients, including outpatient follow-up, is vitally important. We also review what is presently known about long-term kidney outcomes after the initial recovery from COVID-19. We provide some guidance as to the management of patients hospitalized with COVID-19 who are at risk for AKI as well as for future clinical research in the setting of COVID-19 and the significance of early identification of patients at highest risk for adverse kidney outcomes.

EPIDEMIOLOGY

Since December 2019, more than 185 million people from over 190 countries and territories have been infected with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus (1). The impact of COVID-19 on patients has been extremely variable and affected by well-described risk factors, both demographic and patient specific, including patient age and comorbidities such as diabetes and hypertension (2).

Multiple research groups around the world have reported on the epidemiology of COVID-19, including risk factors, incidence, and short- and long-term outcomes after diagnosis. Kidney involvement was recognized as a prominent characteristic of COVID-19, with AKI noted in 20−50% of patients admitted to the hospital. Between 15% and 20% of patients with AKI have required kidney replacement therapy (KRT) (3–6). A recent systematic review and meta-analysis by Silver et al. (7) evaluated over 53 published studies covering over 30,000 hospitalized patients with COVID-19 and found a pooled prevalence of AKI of 28%, with the pooled prevalence for a KRT requirement of 9%.

Among patients with acute respiratory distress syndrome due to COVID-19, up to 50% develop AKI (4, 8). Common risk factors include older age, male sex, baseline medical comorbidities, and a number of clinical characteristics during hospital admission (Fig. 1). Compared with patients without AKI, those who develop AKI also had generally higher levels of inflammatory markers including higher C-reactive protein, ferritin, and D-dimer. Moledina et al. (9) also found that over 30% of hospitalized patients with COVID-19 developed AKI compared with 18.2% of hospitalized patients without COVID-19 during the same time period. This risk of AKI remained significant after adjustment for demographics, comorbidities, vitals, medications, and laboratory results (hazard ratio: 1.40, 95% confidence interval: 1.29–1.53). Among patients with AKI, those with COVID-19 were less likely (58%) to have recovered at the time of discharge compared with patients without COVID-19 (69.8%). Patients with COVID-19 and AKI were also significantly more likely to die or have longer hospital stays than patients without COVID-19 and AKI.

Figure 1.

Risk factors for COVID-19-asociated acute kidney injury (AKI). Risk factors that have been associated with COVID-19-associated AKI include patient-related factors at baseline, including demographics and medical comorbidities, along with genetic factors with linkage to possible apolipoprotein L1 (APOL1) high-risk variants as a risk factor. Clinical and laboratory-based measurements and medications including antibiotic and nonsteroidal anti-inflammatory drug (NSAID) use have been associated with AKI as well. CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CRP, C-reactive protein; SOFA, sequential organ failure assessment.

Inpatient and Postdischarge Outcomes for Patients With COVID-19-Associated AKI

Patients with COVID-19-associated AKI are at much higher risk for in-hospital outcomes. Patients with AKI are more likely to be admitted to the intensive care unit (ICU), to require mechanical ventilation and vasopressor support, and had significantly higher in-hospital mortality (4). Among those with COVID-19, 50% of patients with AKI have been found to suffer in-hospital mortality compared with 8% of patients without AKI (adjusted odds ratio: 9.2, 95% confidence interval: 7.5–11.3). Of those who survive to hospital discharge, ∼65% have recovery of kidney function.

As the COVID-19 pandemic has progressed, a number of studies have investigated the longer-term effects and kidney function trajectory after COVID-19-associated AKI. Nugent et al. (10) investigated the association of COVID-19 both with AKI and with the longitudinal trajectory of the estimated glomerular filtration rate (eGFR) in over 1,600 patients requiring hospitalization for COVID-19 within the Yale New Haven Health System. These authors specifically looked at the change in eGFR from the time of hospital discharge to the time of postdischarge laboratory follow-up in patients with COVID-19-associated AKI compared with patients with AKI who tested negative for SARS-CoV-2. There was a significant difference in the change in eGFR in patients with COVID-19-associated AKI compared with those with AKI who were COVID-19 negative, even after adjusting for baseline demographic factors and comorbidities (slope difference of 14 mL/min/1.73 m² in the fully adjusted model). These authors argued that patients who recover from COVID-19-associated AKI may be at a relatively greater risk for longer-term adverse kidney outcomes and may require closer monitoring post discharge.

Long COVID Syndrome and Kidney Manifestations

Several studies have investigated the long-term consequences of patients who survive COVID-19 hospitalization, specifically looking at patients who develop long-term side effects after initial recovery from COVID-19, variously termed postacute sequelae of SARS-CoV-2 virus, long COVID syndrome, or postacute COVID-19 syndrome (12). Common long-term complications include chronic fatigue, muscle weakness, anxiety, depression, insomnia, and shortness of breath (13). Such symptoms have been reported not only in hospitalized COVID-19 survivors but also among low-risk patients who are relatively younger and healthier (14). Stockmann et al. (15) explored long-term kidney-specific outcomes in survivors of COVID-19 hospitalization in a single center study in Berlin, Germany. Among 74 patients with COVID-19-associated AKI requiring KRT, almost all of whom required mechanical ventilation in an ICU setting, 36 patients (49%) died during hospitalization. Among the 37 patients who had been discharged by the study conclusion, 34 patients (92%) had recovered enough native kidney function to be liberated from KRT, at a median overall duration of KRT being 27 days. Even beyond the first 30 days of COVID-19 among those who required hospitalization, adverse kidney manifestations included urinary tract infections, AKI, and chronic kidney disease (16). Huang and colleagues (13) found that 6 mo after survival from COVID-19 hospitalization, 35% of patients had decreased eGFR (<90 mL/min/1.73 m²). Even among patient without AKI during hospitalization, 13% had decreased eGFR at follow-up.

PATHOGENESIS

The SARS-CoV-2 virus has been shown to enter cells through the angiotensin-converting enzyme (ACE)2 receptor through its spike glycoprotein (17), similar to the severe acute respiratory syndrome coronavirus initially described in 2003 (18) but with 10- to 20-fold higher binding affinity. In contrast to the renin-angiotensin-aldosterone system (RAAS) pathway involving ACE converting angiotensin I to angiotensin II to raise blood pressure, ACE2 is involved in another critical pathway of the RAAS acting through Ang-(1–7) that has an anti-inflammatory, blood pressure-lowering effect (19, 20). The ACE2 receptor has been previously shown to be moderately expressed in the kidney, with low expression in the lung and high expression in the testes, small intestine, and thyroid (21). Therefore, there were early reports of the possible link between direct viral entry of the SARS-CoV-2 virus with COVID-19-associated AKI (22, 23). The potential associations between direct and indirect effects of COVID-19 (24), through proinflammatory pathways leading to multiorgan injury, has become increasingly important targets of scientific investigation.

Between May and June 2020, an expert panel of physician-scientists and researchers met virtually to create a consensus statement regarding the pathophysiology of AKI in the setting of COVID-19 (23). These experts identified both direct and indirect effects of the SARS-CoV-2 virus that may have led to AKI. Possible mechanisms of direct viral effects of injury to the kidney from SARS-CoV-2 include direct endothelial damage from viral entry as well as complement activation, local inflammation, and collapsing glomerulopathy. However, the more common indirect effects that these authors postulated as causative of COVID-19-associated AKI included more common causes of in-hospital AKI in general, such as sepsis, nephrotoxic medications, and systemic inflammation (Table 1). Ng et al. (22) also explored various ways in which SARS-CoV-2 may lead to AKI; they also noted that it is unclear whether viral entry into cells can lead directly to AKI.

Table 1.

Potential etiologies of COVID-19-associated acute kidney injury

Direct	Indirect
• Direct viral invasion • Thrombotic microangiopathy • Collapsing glomerulopathy • Immune dysregulation (complement activation)	• Acute tubular injury: sepsis (fevers and prolonged hypotension) and renal ischemia • Prerenal azotemia: hypovolemia • Rhabdomyolysis • Interstitial nephritis (medication induced) • Oxalate nephropathy (high-dose vitamin C) • Drug toxicity

Descriptive kidney biopsy studies have noted a number of distinct histological findings in patients undergoing kidney biopsy for COVID-19 (25–28). In general, the major pathological feature that is nearly universal is the presence of acute tubular injury (ATI) of varying severity. One of the earliest studies describing histological changes in the kidney attributed to COVID-19, by Su et al. (25) in China, described postmortem histological findings in 26 deceased patients who had COVID-19, 11 of whom had either diabetes mellitus or hypertension. Five of the 26 patients required KRT [specifically, continuous KRT (CKRT)] during hospital admission. The authors noted at least mild ATI in all 26 kidney tissue samples, with moderate to severe ATI in nine patients. One notable limitation of this study, however, was the postmortem sampling of the kidney tissue.

Santoriello et al. (29) further examined postmortem kidney pathology findings in 42 patients with COVID-19 who died while hospitalized in New York City. These investigators also found that the predominant histological feature on light microscopy was ATI, with frequent findings of background mild chronic changes due to diabetic glomerulosclerosis. The authors noted collapsing glomerulopathy in one Black patient on the postmortem examination. There were no significant findings of note on immunofluorescence or electron microscopy. In addition, in situ hybridization studies were performed in a subset of 10 autopsies chosen at random without any definitive positivity for SARS-CoV-2 noted.

Glomerular Disease in COVID-19

A number of case reports have individually described collapsing glomerulopathy associated with COVID-19 in the setting of significant proteinuria, similar to that seen in human immunodeficiency virus (HIV) infection (30, 31). Wu et al. (32) investigated the genetic, histopathological, and molecular features of six Black patients with COVID-19-associated AKI and proteinuria. All six patients had high-risk apolipoprotein L1 (APOL1) genotypes, and none of the kidney injuries were associated with direct viral infection of the kidney. They proposed a two-hit theory of AKI involving genetic predisposition (namely, APOL1 high-risk status) and a cytokine-related host response to the virus.

Nasr et al. (27) summarized the results of 13 kidney biopsies in patients with AKI and proteinuria in the United States. Of the patients undergoing kidney biopsy, 12 of 13 had Kidney Disease: Improving Global Outcomes (KDIGO) stage 3 AKI, with 44% of patients with AKI present on admission. Although all 13 patients had proteinuria by definition, 11 of 13 patients had nephrotic-range proteinuria with at least microhematuria in 10 patients. All 13 patients had at least mild ATI, whereas eight patients had evidence of collapsing glomerulopathy attributed to COVID-19 and characterized by collapse of the glomerular tuft with podocyte hyperplasia and notable protein resorption droplets. All eight patients with collapsing glomerulopathy were Black; one patient was noted to have the APOL1 high-risk alleles present, although it is unclear if this patient had a high-risk genotype (G1/G1, G1/G2, or G2/G2) and if APOL1 testing was performed for others. More recently, Shetty et al. (28) similarly published a single center case series of six patients of recent African descent with confirmed COVID-19, AKI, and proteinuria in Cook County, IL. Patients were found to have either a podocytopathy, collapsing glomerulopathy, or both. Of the three patients who underwent APOL1 genetic testing, all had high-risk genotypes. These authors also noted the significantly higher risk of HIV-associated nephropathy in patients with APOL1 high-risk genotypes.

Given its similarity to HIV-associated nephropathy, the collapsing glomerulopathy associated with HIV, the collapsing glomerulopathy seen in the setting of COVID-19 has appropriately been termed COVID-19-associated nephropathy (COVAN) (33). Chandra and Kopp (34) previously summarized the evidence for collapsing glomerulopathy in the setting of cytomegalovirus, as well as Epstein-Barr virus and parvovirus B19, further supporting a potential shared viral-mediated trigger for kidney injury, although mechanistic studies are needed to explore this further.

Biomarkers in COVID-19-Associated AKI

Biomarkers of systemic inflammation, including IL-6, C-reactive protein (CRP), D-dimer, and ferritin, have been associated with severe COVID-19 infection in general as well as with severe AKI (35). Husain-Syed et al. (36) were the first to report on the association of urinary biomarkers beyond proteinuria and hematuria with AKI in patients with COVID-19. They measured urinary tissue inhibitor of metalloproteinase-2 (TIMP-2) and insulin-like growth factor-binding protein 7 (IGFBP7) in 23 patients admitted to a single center in Germany with COVID-19. These authors found that there was no significant difference in biomarker levels between patients who developed AKI and those who did not, although among patients with AKI, [TIMP-2] × [IGFBP7] levels differed among those who remained at KDIGO stage 1 AKI compared with those who progressed.

Since the early recognition of viral entry into cells via the ACE2 receptor, and with ACE2 receptors expressed prominently in the renal tubular epithelium, there has been continued controversy about direct viral infection of the kidney and what, if any, role the presence of viral RNA has on kidney function or adverse kidney outcomes (37). Khan and colleagues (38) summarized much of the literature to date regarding direct infection of the SARS-CoV-2 virus in the kidney, with the presence of virus defined by several methods, including viral RNA by RT-PCR, viral protein detection by immunohistochemistry, and electron microscopy evidence of viral particles on kidney biopsy. The clinical significance of detectable viral RNA in the urine in patients with COVID-19 remains unclear, with Frithiof and colleagues (39) finding no association between the presence of viral RNA and AKI in critically ill patients with COVID-19. More recently, Caceres and colleagues (40) reported that among 52 patients admitted to the hospital with COVID-19 who underwent RT-PCR testing, 20 patients had detectable SARS-CoV-2 viral RNA in urine samples, among which 17 patients (85%) developed AKI.

MANAGEMENT OF PATIENTS WITH COVID-19-ASSOCIATED AKI

Recommendations for the management of patients with COVID-19-associated AKI incorporate the general guidelines outlined in the KDIGO practice guidelines for AKI (41, 42). These include the discontinuation of all nephrotoxic agents when possible in patients at risk for AKI as well as ensuring proper hemodynamic monitoring and fluid management with monitoring of urine output. To reduce the risk of cytokine burden and hemodynamic instability due to volutrauma and barotrauma, Ronco et al. (42) also recommended lung-protective ventilation and proper fluid management, with close monitoring of intake and output, and initiation of KRT, as indicated. Figure 2 outlines a general approach to the care of patients with COVID-19 who develop or are at high risk for AKI. Patients with progressive AKI, worsening electrolyte abnormalities, and persistent fluid overload as well as those with significant proteinuria or hematuria may require nephrology consultation for management and consideration for kidney biopsy.

Figure 2.

Algorithm for evaluating kidney injury in patients admitted to the hospital with COVID-19. All patients with suspected or confirmed COVID-19 should be evaluated for risk of acute kidney injury (AKI) at the time of diagnosis based on demographics and comorbidities and, throughout the patient’s hospitalization, a thorough and complete evaluation for AKI including the use of clinically available blood and urine studies. Timely consultation to nephrology for refractory electrolyte, acid-base, or other metabolic abnormalities or persistent fluid overload is important. Further management decisions would then include the potential need for kidney replacement therapy and biopsy with postdischarge follow-up. BMP, basic metabolic panel; RBC/HPF, red blood cells per high-power field; UA, urinalysis; UACR, urine albumin-to-creatinine ratio; UPCR, urine protein-to-creatinine ratio.

AKI Requiring KRT

Ng et al. (43) found that among nearly 10,000 patients admitted to hospitals in New York City with COVID-19, those with COVID-19-associated AKI had significantly higher rates of death with increasing AKI severity. Of these patients, 638 patients developed AKI requiring dialysis. In total, of the over 3,000 patients who developed AKI but did not require dialysis, the majority (74%) recovered by the time of discharge. In the 108 patients who required any form of dialysis, 72 patients (66.7%) experienced recovery of kidney function, whereas 33 patients (30.6%) remained dialysis dependent upon discharge. This is in contrast to the 10–30% of survivors of general hospitalized AKI requiring dialysis (44).

Among patients with AKI in the setting of COVID-19, ∼10−20% required KRT (9, 45) compared with ∼3.5% in hospitalized patients without COVID-19. In the Study of the Treatment and Outcomes in Critically Ill Patients with COVID-19 (STOP-COVID Consortium), Gupta et al. (45) showed that among patients admitted to an ICU who required KRT, over 50% died within 28 days of ICU admission, with risk factors including older age and severe oliguria. Approximately 33% of patients started on dialysis remained dialysis dependent upon discharge.

Dialysis Modalities for KRT

In critically ill patients with severe AKI, CKRT is the mainstay of management. In the setting of COVID-19, this has remained the modality of choice, given the lower risk of hemodynamic instability compared with intermittent hemodialysis and the ability to specify ultrafiltration goals. One issue specific to COVID-19, however, is the widely documented increase in circuit clotting (23, 46–48). An additional concern, particularly early in the pandemic, was the limited supplies of both dialysate and personnel to meet the increasing need for dialysis.

Notably, although most acute KRT in hospitalized patients occurs through hemodialysis, either continuous or intermittent, acute peritoneal dialysis (PD) is a less frequently used but viable alternative. Throughout the COVID-19 pandemic, often through necessity, acute PD has been successfully implemented in the United States and internationally (49, 50). Given the immense strain on infrastructure and concerns for supply chain disruption, there was a growing concern in places like New York City regarding the availability of dialysate solution for hemodialysis. Hemodialysis has also been challenging in patients with severe COVID-19 due to issues with thrombosis and clotting, especially in those receiving CKRT (50). Parapiboon et al. (50) has outlined several advantages for the use of acute or urgent-start PD in patients with COVID-19 requiring KRT, including the lack of requirement for anticoagulation and improved safety for health-care staff, with less face-to-face time needed with contagious patients compared with hemodialysis.

Indeed, the increasing use of urgent-start PD has forced nephrology programs to develop a PD infrastructure, both inpatient and outpatient, when such programs may have been smaller or nonexistent in the past, especially in certain areas within the United States (51). Steps to increase the use of PD in response to the COVID-19 pandemic include urgent PD catheter insertion, greater patient and caregiver education on PD, and a reevaluation of PD quality metrics (52). In the early stages of the pandemic in New York City, El Shamy et al. (49) developed an acute PD program at Mount Sinai Hospital to meet the increasing need for KRT, especially with limitations on both personnel and supplies, to handle urgent hemodialysis needs. Specifically, patients who were considered for PD included those with baseline chronic kidney disease G4/5 who were started on hemodialysis acutely but with presumed long-term KRT needs and patients with AKI who had not yet started KRT but without urgent clearance needs such as severe hyperkalemia or uremia.

TREATMENT

The treatment of patients with COVID-19 has evolved throughout the course of the pandemic. Initially guided by expert opinion and case reports, additional studies based on retrospective data analysis, prospective studies, and randomized clinical trials, in particular, have helped to better inform management strategies. Hydroxychloroquine was purported to be a potential treatment for COVID-19, particularly in the first few months of the pandemic. Thrombotic microangiopathy has been noted in patients admitted to the hospital with COVID-19, for which hydroxychloroquine use has been suggested to be a potential contributor (53). Dexamethasone, frequently administered to both inpatients and outpatients with COVID-19, is frequently used in patients with AKI without dose adjustment for kidney function. In a study of 157 patients hospitalized with COVID-19 who received remdesivir, 30 patients with AKI did not demonstrate any worsening of kidney function attributed to the drug (54). Various other treatments have been attempted to improve recovery or prevent infection with adverse consequences, including reports of oxalate nephropathy in the setting of excessive vitamin C administration for COVID-19 (55).

Although there have been reports of adverse effects of a number of COVID-19 vaccines, most recently with use of the Johnson & Johnson vaccine, the predominant feature is cerebral venous sinus thrombosis (56). There have been several case reports of AKI and glomerular disease, including pauci-immune glomerulonephritis and minimal change disease, associated with COVID-19 vaccination, and these associations require much further investigation (57–59).

FUTURE RESEARCH DIRECTIONS

From the start of the COVID-19 pandemic, when adverse kidney outcomes were first recognized, there has been a massive worldwide effort to improve our collective understanding of how COVID-19 affects the kidney. Although clinical outcomes in patients with COVID-19 have been elucidated in the short term, further clinical and translational research is needed to improve timely diagnosis and acute management and to inform long-term management of patients after the diagnosis of COVID-19. The association between APOL1 high-risk genotype status and risk for collapsing glomerulopathy from COVID-19, similar to that seen in HIV and other viral infections, requires further study. Exploring this and other potential genetic risk factors will be important in identifying patients at particularly high risk for COVID-19-associated kidney injury. Biomarkers of kidney injury, inflammation, and repair have been shown to risk-stratify patients to adverse short- and long-term kidney outcomes in a variety of clinical settings, including biomarkers in the urine (albumin-to-creatinine ratio, monocyte chemoattractant protein-1, epidermal growth factor, and YKL-40) and in the plasma (kidney injury molecule-1 and tumor necrosis factor receptor 1) (60–63). Such research should be applied in the context of COVID-19 as well.

Ultimately, with a better understanding of the clinical and genetic risk factors as well as appropriate and timely acute management and proper post-COVID-19 follow-up, we can have a meaningful impact on the long-term kidney health of patients who suffer from COVID-19.

GRANTS

This work was supported by National Institutes of Health Grants HL085757 (to C.R.P.), 1K23DK128538-01 (to S.M.), and DK082185 (to C.R.P.).

DISCLOSURES

C.R.P. has reported serving on the board of RenalytixAI and owning equity in same. He also serves on the advisory board of Novartis and on the Data Safety Monitoring Board committee of Genfit. S.M. has received consulting fees from the Dedham Group.

AUTHOR CONTRIBUTIONS

S.M. prepared figures; S.M. and C.R.P. drafted manuscript; C.R.P. edited and revised manuscript; S.M. and C.R.P. approved final version of manuscript.