I. INTRODUCTION

During the COVID-19 pandemic, part of the individuals infected with SARS-CoV-2 manifested a mild to moderate respiratory illness, however, many elderly people and people with other comorbidities evolved to severe conditions[^1]. Patients in the intensive care unit (ICU) developed secondary complications, such as liver damage, venous thromboembolism, acute kidney injury (AKI), and others[^2]. AKI in this case is correlated with poor prognosis and a higher risk of patient mortality[^3].

Currently, the definition and staging of acute kidney injury are based on the KIDGO (Kidney Disease Improving Global Outcomes) criteria, which unifies the RIFLE (Risk, Injury, Failure, Loss and End-Stage) and AKIN (Acute Kidney Injury Network) criteria. Thus, KDIGO uses changes in serum creatinine and urinary output, requiring at least two serum creatinine values obtained in 48 hours[^5].

Serum creatinine levels increase as a $50\mathrm{\%}$ drop in renal function occurs, that is, it is not directly correlated with the decrease in glomerular filtration rate (GFR), which makes early diagnosis impossible. Given this, in recent years, studies with new biomarkers for AKI have gained emphasis in different clinical settings. Among the new biomarkers tested are: Cystatin C, Lipocalin Associated with Human Neutrophil Gelatinase (NGAL), N-acetyl -B-D-glucosaminidase (NAG), Kidney Injury Molecule-1 (KIM-1), Interleukin-18 (IL -18), Netrin-1 and others[7,8].

Given the difficulty of diagnosing AKI using serum creatinine and the inefficient prognosis in COVID-19, the use of biomarkers such as Cystatin C and interleukin-18 is suggested[^9]. Cystatin C is a cysteine proteinase inhibitor protein, which is related to several pathological processes. It is freely filtered in the glomeruli, as it has a low molecular weight, but is almost completely reabsorbed in the proximal tubules[^10]. Interleukin-18 is a pro-inflammatory cytokine that induces the release of inflammatory cytokines and TNF, thus acting as a mediator in the immune system[^6].

The present work is a systematic review of the comprehensive literature, with the main objective of exploring biomarkers for the early diagnosis of AKI induced by COVID-19. Predicting and/or determining the patient's prognosis, as this is of clinical importance and is extremely necessary to apply as a laboratory practice.

II. METHODS

a) Instrument for selection of studies and inclusion criteria

The instrument for selecting the studies was the Relevance Test (TR), based on Pereira's model (2006)[^11]. This consists of forms containing four selection steps, to include or exclude the articles found with the search strategy.

Initially, TR1 was applied to the references to select studies that meet the following inclusion criteria:

• Publications made from January 2020 to February 2022;
• Published in Portuguese, English and Spanish.

As exclusion criteria, those who did not meet the criteria described above were considered.

b) Definition of descriptors and search in the literature

A literature search was performed by crossing the following descriptors in Health Sciences/Medical Subject Headings (DeCS/MeSH) available on the VHL network: Acute Kidney Injury and COVID-19 "OR" SARS-CoV- two. The articles were searched using the advanced method, using the term "title/abstract/subject" with the combination of descriptors and the Boolean operators "AND" and "OR".

c) Data extraction and analysis

Taking into account the methodological rigor of a systematic literature review (SLR), a form was prepared to extract the following information from the included studies: bibliographic reference, type of study, research objectives, methodology, and results obtained[^12]. To synthesize as much information as possible about the biomarkers for the prognosis of AKI, the following data were collected: characteristics and clinical conditions of the patients, criteria for diagnosing AKI, biomarkers used, the value of the biomarker and serum creatinine on admission and whether there was any change in the glomerular filtration rate (GFR), biomarker performance for AKI prognosis and clinical outcomes.

III. RESULTS

Bearing in mind that new articles are inserted in the databases every day, April 24, 2022, was chosen to carry out the bibliographic research of this RSL. A total of 75,631 articles were obtained, of which TR1 was applied. After applying the first test, a total of 13,531 articles were included, in which the TR2 was applied in the titles and abstracts with the help of the search platform, 12,492 studies were excluded and 1,039 were included. In the next step, TR3 was applied to the abstracts of previously selected articles, which resulted

After analyzing the eight studies selected for this SLR, a summary of the general characteristics was prepared and made available in synoptic tables. In Frame 1, information and characteristics of the articles analyzed during the study are presented.

Frame 1: Information and characteristics of the articles selected during the study

Biomarker predictors of poor prognosis such as ICU stay, renal replacement therapy (RRT), or death were observed for patients with AKI in COVID-19, as shown in Frame 2.

Frame 2: Prognostic predictor biomarkers for AKI in COVID-19

DKK33 Dickkopf-related protein 3, KIM-1 Kidney Injury Molecule 1, AKI Acute Kidney Injury, NAG N-acetyl-β-D-glucosaminidase, NGAL Neutrophil gelatinase-associated lipocalin, MR - ProADM Mid-regional pro-adrenomedullin, TRS Therapy Renal replacement, ICU Intensive care unit.

Two studies evaluated biomarkers that were useful for early diagnosis, these markers are shown in Frame 3.

Frame 3: Biomarkers for early diagnosis

$\beta 2MG$ $\beta 2$ -microglobulin, L-FABP Liver-type fatty acid binding protein.

When analyzing the characteristics of the clinical conditions of the populations of the eight studies, it was observed that among the eight studies, the size of the population with COVID-19 ranged from 18 to 389 patients, with three studies presenting, in addition to the COVID-19 cohort, a control group, such as can be observed in Table 1.

Frame 4: General characteristics of studies with biomarkers for prognosis

Fukao et al. (2021).

β2MG β2-microglobulin, IL-6 Interleucina 6, L-FABP Liver-type fatty acid binding protein, AKI Acute kidney injury, PCT Procalcitonin.

After analyzing the studies, it is observed that the classification used for the detection of AKI was KDIGO (2012), having been used in seven of the eight studies found, and only one study did not present this information. Biomarkers were analyzed in blood and urine samples, with analysis of serum biomarkers being the most prevalent among the eight studies. The dosage of these markers was carried out from the moment of consultation in the emergency department or hospital admission. In summary, this information is available in Frames 6 and 7.

Frame 6: Biomarkers for AKI prognosis in patients with COVID-19.

AUC Area under the curve, $\alpha 1MGCR$ $\alpha 1$ -microglobulin-creatinine ratios, CKD Chronic kidney disease, eGFR Estimated glomerular filtration rate, KIM-1 Kidney Injury Molecule 1, MR - ProADM Mid-regional pro-adrenomedullin, NAG N-acetil- $\beta$ D-glicosaminidase, NGAL Neutrophil gelatinase-associated lipocalin, AKI Acute kidney injury, GFR Glomerular filtration rate, RRT Renal replacement therapy, uα1MG Urinary α1 microglobulin, uDKK3 Urinary Dickkopf-3, ICU Intensive care unit.

$\alpha 1MG\alpha 1$ microglobulin, $\beta 2MG\beta 2$ -microglobulin, L-FABP Liver-type fatty acid binding protein AKI Acute kidney injury, NAG N-acetil- $\beta$ -D-glicosaminidase, PCT Procalcitonin, GFR Glomerular filtration rate.

Regarding the results for prognosis, it was seen that the biomarkers used were predictors of poor prognosis. KIM-1/creatinine and NGAL/creatinine ratios were predictors of death from COVID-19 ${}^{13}$. Serum creatinine and cystatin C were indicators of AKI, worsening of the lesion, and need for RRT. The NAG marker was significant for predicting AKI in patients with severe AKI who required RRT ${}^{14}$. Elevation in the urinary biomarker uDKK33 at six months from hospital admission was suggestive of AKI chronicity by COVID-19 ${}^{15}$. Serum creatinine and urinary biomarkers KIM-1 and NGAL correlated with urinary cytokines of incidence and worsening of AKI ${}^{16}$. Serum creatinine and KIM-1 were significantly elevated in patients who achieved the composite endpoint (acute kidney injury, ICU, and death) and the NAG marker was significantly elevated in patients who suffered from AKI ${}^{17}$.

In Frame 7, the results of two studies with biomarkers for the early diagnosis of AKI in COVID-19 are available.

IV. DISCUSSION

The electronic survey involved studies carried out during the COVID-19 pandemic, published between 2021 and 2022. Given the pandemic context, one can observe agility in carrying out the surveys. In addition, according to the analyzed studies, it is observed that the search for new AKI laboratory biomarkers for the identification and progression of AKI continues to advance.

Despite the size of the sample (18 to 389 participants), no influence was verified on the characteristics and clinical conditions of the patients. It was observed that among the eight studies, the patients who developed AKI were over 50 years old, and were hypertensive or diabetic. In addition, the KDIGO diagnostic criteria, GFR calculation using the CKD-EPI formula, and biomarker analysis (ELISA) were also similar among studies. In addition to this, nephelometric tests were also used, mainly to analyze Cystatin C, which helps in the comparison between them. Although unusual, Cystatin C is considered an early marker of AKI and its dosage has been performed with nephelometric tests, as they are considered quite stable ${}^{18}$.

Among the biomarkers, urinary KIM-1 stood out as the most promising for the diagnosis and prognosis of AKI. Significant elevations were found in the KIM-1/urinary creatinine ratio in patients with AKI and specific deaths from COVID-19 ${}^{13}$. Consistently, a significant increase in KIM-1 levels between 24h and 48h after admission to the ICU in patients with septic AKI due to other diseases who did not survive was also verified ${}^{19}$. In a study by ${}^{16}$, a correlation of urinary KIM-1 with several urinary cytokines related to the incidence and worsening of AKI in COVID-19 was observed. The results of ${}^{17}$ showed elevations in urinary KIM-1 levels in COVID-19-positive patients with a composite outcome of AKI, ICU admission, and death. Similarly, in a study carried out with patients hospitalized with AKI in other diseases, the increase in urinary KIM-1 and NAG levels at the time of consultation with a nephrologist were predictors of the composite outcome of RRT or in-hospital death ${}^{20}$.

The second prominent biomarker was NGAL, both for serum concentrations and urinary levels. Serum NGAL was useful for predicting AKI, need for RRT, ICU stay, and expressly highs in severe AKI requiring RRT ${}^{14}$. NGAL/creatinine was useful as a good predictor of mortality in patients with COVID-19 and AKI ${}^{13}$. Similar results were found in a study with critically ill patients due to other diseases, where $40\mathrm{\%}$ more cases of AKI were detected with the evaluation of NGAL and creatinine than when using creatinine alone, and in these patients, the risk of ICU admission, need for TRS and death was higher ${}^{21,22}$.

Another biomarker with prognostic results for AKI was sCysC, whose elevation exceeded the urinary KIM-1/creatinine and NGAL/creatinine ratio in ICU patients, but was not considered a predictor of specific mortality from COVID-19 ${}^{13}$. Serum and urine creatinine were used in all eight studies, as it is the standard AKI diagnostic marker. On admission, serum creatinine was similar to sCysC for predicting AKI and the need for

• ${\mathsf{RRT}}^{14}$. However, the elevation of sCysC compared to serum creatinine may have been influenced by high doses of corticosteroids[^15].
• In the analysis of urinary dickkopf-3 (uDKK3), a new biomarker of CKD progression. uDKK3 levels remained high 6 months after hospital admission for COVID-19 in patients with AKI and a greater decline in glomerular filtration. The existence of a secondary AKI over an unresolved AKI has been suggested, which may contribute to the transition from AKI to CKD at 6 months post-discharge15. These findings are similar to what is in the literature, since in a study with post-cardiac surgery patients, urinary DKK3 was associated with a risk of severe loss of glomerular filtration after the transition from AKI to CKD during the patients' follow-up period23.
• As for biomarkers for early diagnosis of AKI, the highlights were $\beta 2\mathrm{MG}$ and L-FABP, which were elevated in urine samples as a reflection of the onset of ${\mathrm{AKI}}^{24}$. L-FABP is a protein present in the liver that plays an important role in regulating the metabolism of fatty acids. In addition, it has a high affinity for lipid peroxidation products, which promotes its elimination in the urine[25].
• 26 Evaluated common biomarkers, but included serum PCT dosage in their study, which in turn was useful for early detection of AKI in patients with COVID-1927.
• Although the cost of testing may still be a limiting factor, the results have shown promise for assessing the incidence of AKI with new biomarkers ${}^{16,28,14,17}$, predict AKI severity, help identify patients who need RRT ${}^{16,14}$ or ICU ${}^{28,14,17}$, predict death ${}^{28,17,13}$ and perform early detection of AKI in COVID-19 ${}^{24,26}$, thus minimizing the cost of the test.
• With the use of new biomarkers detected before serum creatinine, it was possible to detail the AKI caused by COVID-19 for the first time in the literature[^13]. The evaluation of biomarkers such as KIM-1 and Cystatin C at admission can provide diagnostic data and help to outline the prognosis of AKI, which can influence medical decisions and thus increase the chances of better outcomes^17,14. In addition, it is considered important to evaluate the renal system in COVID-19 with AKI biomarkers, as well as to consider their continuous evaluation during hospitalization[^13].
• The application of new sensitive biomarkers for AKI proved to be important for the diagnosis and prognosis of the lesion in the COVID-19 scenario.
• Different biomarkers were able to predict AKI severity, need for RRT, hospitalization, and death. The poor prognosis of AKI in COVID-19 observed in the analyzed studies highlights the importance of identifying new biomarkers and applying them in laboratory practice and different clinical scenarios.
• Although the use of some of these biomarkers is limited due to the high cost of the necessary reagents,

Disclosure statement

• No potential conflict of interest was reported by the author(s).

Funding