Acute kidney injury in hospitalized COVID-19 patients in Iran; a systematic review and meta-analysis

Implication for health policy/practice/research/medical education: In this systematic review and meta-analysis, we showed that 24% of hospitalized COVID-19 patients in Iran developed acute kidney injury. The geographical dissimilarities in the proportion of acute kidney injury among COVID-19 patients suggest a role for ethnical and racial differences in the tendency to develop renal involvement. Please cite this paper as: Saghafi A, Aghaali M, Saghafi H. Acute kidney injury in hospitalized COVID-19 patients in Iran; a systematic review and meta-analysis. J Renal Inj Prev. 2021; 10(2): 09. doi: 10.34172/jrip.2021.09. R ev ie w


Introduction
On 31 December 2019, 27 cases of pneumonia of unknown etiology were identified in Wuhan city, Hubei province in China. These patients most notably presented with clinical symptoms of dry cough, dyspnea, fever, and bilateral lung infiltrates on imaging. The causative agent was identified from throat swab samples conducted by the Chinese Centre for Disease Control and Prevention (CCDC) on January 7, 2020, and was subsequently named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The disease was named COVID-19 by the World Health Organization (WHO) (1). On March 11, 2020, the COVID-19 situation was characterized as a pandemic by WHO (2). Although the majority of patients presented with mild symptoms so far, a substantial number of them experienced fatal complications like various end organ failures, septic shock, pulmonary edema, and acute respiratory distress syndrome (ARDS). Acute kidney injury (AKI) is one of the complications in COVID-19 patients, which is reported with widely varied incidence rates in different studies and is known to have a major impact on prognosis and outcome of the disease (3)(4)(5)(6)(7)(8)(9)(10). In a study in Wuhan City, China, during the first few months of the outbreak, the incidence of AKI and aggravation of chronic kidney disease (CKD) among hospitalized COVID-19 patients was concluded to be uncommon. No patient in the study developed AKI based on the diagnostic criteria, reportedly (3). However, the results of Acute kidney injury (AKI) is one of the complications in COVID-19 patients, which is reported with widely varied incidence rates in different studies and is known to have a major impact on prognosis and outcome of the disease. It is noticed that there are considerable differences in AKI rates between different countries. Rates in China are generally much lower than in Western Europe and the United States. One of the potential explanations is heterogeneity along racial and ethnic lines. This study aims to systematically investigate the scientific resources regarding AKI prevalence among hospitalized COVID-19 patients in Iran, and run a meta-analysis on currently published data. Web of Science, PubMed, Embase, Scopus, and Google Scholar databases were searched to identify the articles discussing the occurrence of AKI in hospitalized patients with COVID-19 in Iran. All observational and interventional studies with English full-text providing necessary data for analysis were included with no limitation in time of release or peer-review. Around, 4069 confirmed cases (age; 10-94) from 22 studies were included in the pooled outcome measurement. The proportion of hospitalized patients with COVID-19 in Iran who developed AKI was 24% (95% CI: 17-31%). To the best of our knowledge, this is the first systematic review and meta-analysis to measure the prevalence of AKI in hospitalized COVID-19 patients in Iran. The geographical dissimilarities in the proportion of AKI among COVID-19 patients suggest a role for ethnical and racial differences in the tendency to develop renal involvement.
other studies support the probability of renal involvement in COVID-19. A study from China reported a 0.5% AKI incidence rate (5), while another study from New York City, the United States, calculated AKI development in hospitalized COVID-19 patients to be 36.6% (10). Understanding the pathophysiology of COVID-19 associated AKI is evolving, with no determined theory to date. It is suggested that the cause of kidney involvement might be multifactorial, with cardiovascular comorbidity and predisposing factors (e.g., sepsis, hypovolemia, and nephrotoxins) as important contributors (11). One study described the incidence of AKI as a feature of a thrombotic microangiopathy (TMA)-like phenomenon, secondary to SARS-CoV-2 infection (12). Moreover, virus particles were reported to be present in renal endothelial cells, indicating viremia as a possible cause of endothelial damage in the kidney and a probable contributor to AKI. Other possible mechanisms of injury are direct infection of the renal tubular epithelium and podocytes through an angiotensin-converting enzyme 2 (ACE2)-dependent pathway by SARS-CoV-2 which can cause mitochondrial dysfunction, acute tubular necrosis, the formation of protein reabsorption vacuoles, collapsing glomerulopathy, and protein leakage in Bowman's capsule (11). It is understood that complication of COVID-19 with AKI is associated with significantly higher in-hospital death rate (9), respiratory failure, and need for invasive mechanical ventilation (10); and thus is considered a marker of disease severity and a negative prognostic factor for survival (11). A hypothesis even suggests the possibility of CKD development in COVID-19 survivors complicated with AKI during the course of the disease (13). The difference in AKI prevalence between studies reporting hospitalized COVID-19 patients might partly be explained by different hospital admission policies applied to healthcare systems, and the quality of hospital care (14). However, that is unable to rationalize the substantial variation in AKI occurrence rate among COVID-19 patients ranging from 0% to 46% (3,15). It is noticed that there are considerable differences in AKI rates between different countries. Rates in China are generally much lower than in Western Europe and United States (14,15). One of the potential explanations is heterogeneity along racial and ethnic lines (15). To the best of our knowledge, no publication discussed the issue independently to date, and hence, no validated analysis confirmed the role of ethnicity. This study aims to systematically investigate the scientific resources regarding AKI prevalence among hospitalized COVID-19 patients in Iran, and run a meta-analysis on currently published data. To date, this is the first metaanalysis study specifically concerned with the matter of race and hospital care by limiting the data inclusion to records from Iran. Results of this study can hint at the possible role of diversities in race and ethnicity and trigger further research in AKI predisposing factors, specifically in COVID-19 patients.

Search strategy
This analysis was conducted in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (16). Web of Science, PubMed, Embase, Scopus, and Google Scholar databases were searched to identify the articles discussing the occurrence of AKI in hospitalized patients with COVID-19 in Iran. The whole process of identification was conducted in December 2020. The search terms were selected through the MeSH terms evaluation by two authors independently. Boolean operators were used for an advanced search unless not applicable. The main search string was as follows with minor differences applied to each of the search engines: ("COVID-19" OR "SARS-COV-2" OR "2019-nCoV" OR "coronavirus") AND ("acute kidney injury" OR "acute renal injury" OR "acute kidney failure" OR "acute renal failure" OR "Oliguria" OR "elevated creatinine" OR "elevated serum creatinine" OR "elevated BUN" OR "elevated serum BUN" OR "Reduced GFR" OR "Reduced glomerular filtration rate") AND ("Iran" OR "name of prominent Iranian provinces and cities"). Further details and exact search query for each database are provided in Supplementary file 1.

Inclusion and exclusion criteria
All observational and interventional studies with English full-text providing necessary data for analysis were included with no limitation in time of release or peerreview. Exclusion of the articles based on each of the following criteria was conducted; review articles and other types of publications with no patient enrollment, case reports and case series with less than 10 patients, studies with data from outside Iran or out-patient population, studies limited to COVID-19 patients with renal presentation, articles which specifically focus on pediatric population, pregnant women, end-stage renal disease or chronic dialysis patients, and articles with insufficient data on AKI occurrence.

Data extraction
Search results were evaluated by two independent authors (AS and HS) and study selection was performed assessing the title and the abstract. Among nominated publications, articles' full-texts were reviewed to judge the eligibility for inclusion. A third reviewer (MA) decided on to resolve the disagreements at any phase. The following information was extracted from the studies using an organized checklist: name of the first author, date of the publication, name of the journal, study location, study design, peer-review status, the definition of AKI, type of population (general or special), type of admission ward (general or intensive care), sample size, age range, gender ratio, and proportion of AKI occurrence. General information of the included articles came under scrutiny to prevent duplication and data overlap.

Statistical analysis
We used the random-effects model for investigating pooled prevalence with their specific 95% confidence intervals (CIs). Study characteristics and estimated total proportion were graphically displayed by forest plots. To dissipate any heterogeneity, subgroup analyses with a random-effects model were carried out. Heterogeneity was measured using Cochran's Q test and I 2 method. Publication bias was evaluated using funnel plot and egger's test. The effect of each of the studies on the total proportion and confidence interval was assessed by cumulative graph. A meta-regression was performed to determine the impact of moderator variables on the percentage of AKI occurrence in hospitalized patients with COVID-19. Stata 16.0 (StataCorp, College Station, TX, USA) was used for statistical analysis. P< 0.05 was considered statistically significant.

Results
A total of 948 articles were identified through the systematic search of the databases as previously explained. After removal of all duplications, 746 records remained and were carefully gone through, which returned 125 results. Evaluation of the full-texts to assess eligibility was done and 100 publications were excluded due to either not fulfilling the inclusion criteria or being subject to any of the exclusion criteria. From 25 remained articles, 3 were excluded due to data overlap. Finally, 22 studies  were selected and included in the review and metaanalysis. The search process and results are outlined in Figure 1.
Pooled data from deceased patients showed a significantly higher proportion of AKI compared with the patients who were discharged from hospital (62%, 95% CI: 42-82%, versus 28%, 95% CI: 7-49%, P = 0.02) ( Figure 6 and Figure 7). It should be taken into consideration that the majority of included studies did not report segregated data regarding survival. Hence, these percentages do not represent the whole proportion. AKI was found in 64% (95% CI: 46-83%) of kidney transplant patients, data from two independent studies indicated ( Figure 8).
Evaluating the variables in terms of their impact on AKI occurrence, it turned out that kidney transplantation and death outcome are the two predictors. Details are summarized in Table 2.

Discussion
Since December 2019, when COVID-19 emerged in Wuhan city and rapidly spread throughout China, to date, when it still is the most prominent health issue throughout the world, the data on the disease's presentation, clinical course, and outcome was rapidly growing (5). Because of the novelty of the disease, our knowledge about it is subject to evolution, and the combination of reports yields a more comprehensive understanding as time goes by. We conducted a systematic review and metaanalysis on the studies which reported the occurrence of AKI in hospitalized COVID-19 patients in Iran. We found out that 24% of patients developed AKI, which     indicated kidney damage to be a common complication of the disease. Compared to previous reports, our study returned a higher percentage of AKI. Two previously published meta-analyses, calculated the prevalence of AKI 4.5% (40), and 17% (41). The noteworthy point is that in those meta-analyses, the majority of data came from China, which had published most of the cited documents due to the precedence of disease emergence there. A primary report from china had even concluded that AKI is not a complication of COVID-19 (3), although the following information modified the presumption (14). However, several researchers noticed lower rates of AKI in reports from China than elsewhere (10,15). It means that the conclusion of previous meta-analysis might have been affected by the Chinese reports, while to date, only about 0.1% of COVID-19 cases were from China (42). Hirsch et al conducted a retrospective cohort study on admitted COVID-19 patients in the New York metropolitan area, which showed 36.6% AKI occurrence (10). Another multi-center observational study from New York reported AKI development in up to 46% of admitted COVID-19 patients (43). Although few European data are currently available on this topic, published data presented generally higher rates than Chinese studies (15). These geographical differences can potentially be explained by the differences in admission criteria, hospital care, and ethnic predisposition to kidney involvement. Our study indicated the development of AKI in about a quarter of patients admitted in more than 18 hospitals in various regions of Iran, remarkably higher than most of the Chinese reports and close to reports from Western Europe and United States (14,15). In our study, ICU admitted patients had a higher rate of AKI development than general ward admissions (30% versus 23%). However, the difference was not statistically significant. Moreover, general AKI prevalence in ICU patients was close and even sometimes higher in previous reports from Iran (44)(45)(46). An explanation might be that although AKI is a common complication, it is not a key contributor to disease severity in COVID-19 patients.
In another subgroup study, we analyzed the data from studies which defined AKI either in line with KDIGO criteria (39) or GFR estimation and studies that either did not specify or used serum creatinine to assess AKI. The proportion was higher in the first group (32% versus 21%), although not statistically significant. It is possible that applying a more precise definition to AKI diagnosis leads to a higher occurrence rate.
We conducted an analysis to determine if there is a difference in AKI percentage in deceased compared to discharged patients. Seven articles had reported the necessary data, and analysis revealed that the prevalence of AKI was significantly higher in patients who did not survive than the ones who discharged (62% versus 28%, P = 0.02). Richardson et al performed an observational study which returned 62.7% AKI percentage in deceased patients compared with 8.5% in discharged ones (4). However, due to the scarcity of data, it is yet to develop a conclusion.
Two studies that specifically included kidney transplant recipients who were admitted due to COVID-19 were analyzed. It was shown that 64% of patients developed AKI. AKI is a critical complication in kidney transplant patients, because of its possible contribution to allograft rejection.
Several variables were assessed to determine if they had an impact on AKI occurrence by meta-regression. Two predictors, death outcome and kidney transplantation, were revealed. We assume that the high percentage of AKI development in deceased patients might partly be contributed to multiple organ failure in the terminal phase of severe disease. Further research should be performed to evaluate the impact of AKI as a variable on the survival of COVID-19 patients, limiting confounders effect.
Our review faced some limitations. The data on COVID-19 patients was limited due to the novelty of the disease, and more data might influence the proportion. There was heterogeneity regarding type of study, definition of AKI, and study population. The quality of included studies was varied. Some of the included articles did not mention specific data on AKI occurrence. A serum creatinine cut-off was set to determine AKI in those reports.

Conclusion
To the best of our knowledge, this is the first systematic review and meta-analysis to measure the prevalence of AKI in hospitalized COVID-19 patients in Iran. We showed that 24% of patients developed AKI. The geographical dissimilarities in the proportion of AKI among COVID-19 patients suggest a role for ethnical and racial differences in the tendency to develop renal involvement. Further research is warranted to determine the significance of ethnicity, while other factors like the quality of hospital care might also play a role. COVID-19 needs a comprehensive clinical and research approach, and our understanding of this disease is still insufficient. Large-scale studies are recommended to unfold more accurate characteristics of COVID-19.
Authors' contribution AS, MA, and HS contributed equally to prepare the study concept and design. AS and HS carried out the acquisition of data. MA performed the statistical analysis and interpreted data. AS wrote the primary manuscript. HS critically revised the manuscript. All authors read and signed the final manuscript.

Conflicts of interest
The authors declare no conflict of interest.

Ethical considerations
The study was approved by the ethics committee of Qom University of Medical Sciences (Code: IR.MUQ. REC.1399.226). Ethical issues (including plagiarism, data fabrication, double publication) have been completely observed by the authors.

Funding/Support
None.

Supplementary files
Online Supplementary file 1 shows search strategies used in this study.