Prevalence and Mortality Rates of Acute Kidney Injury among Critically Ill Patients: A Retrospective Study

Acute kidney injury (AKI) poses a significant challenge in critically ill patients. To determine the prevalence, risk factors, and mortality rate of AKI among nonsurgical critically ill patients in Jordan University Hospital, we conducted a retrospective study using a consecutive sampling method, including 457 nonsurgical critically ill patients admitted to the medical intensive care unit (MICU) from January to June 2021. The mean age was 63.8 ± 18 years, with 196 (42.8%) developing AKI during their stay in the MICU. Among AKI nonsurgical patients, pulmonary diseases (n = 52; 34.5%) emerged as the primary cause for admission, exhibiting the highest prevalence, followed by sepsis (n = 40; 20.4%). Furthermore, we found that older age (adjusted OR (AOR): 1.04; 95% confidence interval (CI): 1.04–1.06; p = 0.003), preadmission use of diuretics (AOR: 2.12; 95% CI: 1.06–4.25; p = 0.03), use of ventilators (2.19; 95% CI: 1.12–2.29; p = 0.02), and vasopressor use during MICU stay (AOR: 4.25; 95% CI: 2.1308.47; p = 0.001) were observed to have higher mortality rates. Prior utilization of statins before admission exhibited a significant association with reduced mortality rate (AOR: 0.42; 95% CI: 0.2–0.85; p = 0.02). Finally, AKI was associated with a higher mortality rate during MICU stay (AOR: 2.44; 95% CI: 1.07–5.56; p = 0.03). The prevalence of AKI among nonsurgical patients during MICU stay is higher than what has been reported previously in the literature, which highlights the nuanced importance of identifying more factors contributing to AKI in developing countries, and hence providing preventive measures and adhering to global strategies are recommended.


Introduction
Acute kidney injury (AKI) poses a signifcant challenge in critically ill patients, leading to substantial morbidity and mortality rates while also heightening the risk of both shortterm and long-term complications [1].Te prevalence of AKI reported in critically ill patients varies substantially [2].Te absence of a consensus on AKI has led to a broad spectrum of estimated prevalence within the intensive care unit (ICU), spanning from 1% to 70%.Tese variations in prevalence are contingent upon the specifc criteria employed for AKI diagnosis including creatinine serum level and urine output [2].In Canada, the prevalence of AKI among critically ill patients has been reported to be 64.2%[3].Many developing countries may have restricted resources for conducting research on AKI.As a result, AKI may be underdiagnosed or not well tracked in developing countries, making it difcult to estimate its prevalence and outcomes [4].
Numerous risk factors have been identifed for the onset of AKI, encompassing conditions like heart and pulmonary diseases, liver failure, sepsis, advanced age, preexisting renal disease, and various medications including angiotensinconverting enzyme inhibitors (ACEIs), vasopressors, aminoglycosides, and nonsteroidal anti-infammatory drugs (NSAIDs) [5].Signifcant diferences exist in the prevalence, risk factors, and mortality related to AKI among diferent countries, as well as among diferent areas with varying socioeconomic statuses and diferent hospital facilities [6].Regarding this, length of stay in ICU [7], older age above 60 years old [8], female gender [9], and other chronic diseases [10] were found to be associated with AKI.Tey were poorly investigated in Arab countries, particularly in Jordan.
People in developing countries (i.e., Jordan) may develop more AKI than people in developed countries because of several factors such as medication nonadherence [11], higher poverty rates [12], and higher anxiety levels [13].Due to the scarcity of literature on AKI among nonsurgical critically ill patients in our region, it is noteworthy that only one study conducted in Jordan has investigated the prevalence of AKI in the northern district of the country.Furthermore, it is important to consider that the data collection for that study took place some time ago and may not accurately refect the current state of AKI development [14].Terefore, the aim of our study is to determine the prevalence of AKI in nonsurgical critically ill patients admitted to the medical intensive care unit (MICU) at Jordan University Hospital (JUH).Additionally, we aim to assess the related risk factors and mortality rate in this patient population.

Study Design.
To accomplish the aim of our study, we employed a retrospective study design involving nonsurgical critically ill patients.

Ethical Approval.
Te JUH institutional review board (no.10/2022/5678) approved this study.All procedures conducted in this study adhered to the ethical standards set by the institutional and/or national research committee and followed the principles outlined in the World Medical Association Declaration of Helsinki.Due to the retrospective nature of the study, the requirement for informed consent from patients or their families was waived.We kept all data in a secure place to ensure the anonymity of their identity.

2.3.
Setting.We conveniently selected the JUH, which is the largest hospital in Amman, Jordan, to represent the study population, as it is considered the largest tertiary hospital and has the largest number of ICU beds in Jordan.It provides medical services to population across the central, eastern, and southern regions of the country.Te inclusion criteria for patient selection encompassed patients aged 18 years and older who were admitted to ICUs with medical conditions, including pulmonary, gastroenterology, and neurological diseases.Additionally, patients with infections and cardiovascular conditions not requiring surgical intervention were also considered eligible for inclusion.Te study excluded patients who were under 18 years of age and had surgical conditions due to the inherent risk of AKI associated with surgery.Furthermore, patients with AKI at ICU admission, end-stage renal disease requiring dialysis, recent myocardial infarction, or those necessitating cardiac interventions were also excluded to maintain the scope and focus of the study.

Participants.
Te study included the entirety of nonsurgical critically ill patients who were admitted to MICUs during the period from January to June 2021.Te selected hospital has several ICUs such as neuro, surgical, and medical.However, we only considered the MICU as it is the biggest unit in Jordan with a capacity of 36 beds [15].
Figure 1 illustrates the inclusion and exclusion criteria.Figure 1 illustrates that the total count of admitted patients in the MICU at JUH amounted to 746 patients.Utilizing G * Power software, it was determined that a minimum sample size of 254 participants was deemed sufcient.Out of these, 457 patients met the inclusion criteria, indicating a commendable response rate of 61.2%.It is worth noting that this number of participants surpasses the initially calculated sample size requirement.Consequently, this sample size was deemed robust and capable of providing a 95% confdence interval [16].

Data Collection and Defnitions.
At the time of admission to the MICU, we collected data on baseline characteristics such as demographic factors and comorbidities such as diabetes mellitus (DM), hypertension (HTN), cardiovascular diseases, chronic pulmonary disease, chronic liver diseases, chronic kidney disease (CKD), active malignancy, neurological disease, and endocrinological disease.Furthermore, data on all medication lists before MICU admission were collected, such as those for angiotensin receptor blockers (ARBs) or ACEIs, aspirin, diuretics, statins, proton pump inhibitors (PPI), and NSAIDs.Moreover, baseline vital signs were recorded at admission, including heart rate, systolic blood pressure, and diastolic blood pressure.Te main causes of admission to MICU were identifed as follows: (1) acute pulmonary diseases including any lung pathology that indicates admission to the ICU without major hemodynamical changes or signs of sepsis as an acute exacerbation of baseline lung disease as chronic pulmonary disease, asthma exacerbation, lung fbrosis, and so on, and the new onset of lung pathology as pneumonia without causing hemodynamical changes, (2) acute gastroenterology including medical causes of gastrointestinal bleeding, decompensated liver failure, acute liver failure, etc. Sepsis is defned as meeting the international sepsis defnitions using surviving sepsis campaign [17], (3) acute neurological disease is defned as any neurological disease, such as ischemic or hemorrhagic stroke, or acute neurological illness including myasthenia gravis, Guillain-Barré syndrome, or other neuropathic diseases that requires ICU admission, (4) endocrinological diseases, including diabetic ketoacidosis, myxedema coma, Addisonian crisis, and hypercalcemia or hypocalcemia, and (5) acute cardiac causes, including acute decompensated heart failure associated with 2 Critical Care Research and Practice infections or cardiac arrhythmia associated with infections.Some patients reported illnesses that did not ft the above defnitions.Baseline measurements of renal function were obtained through the assessment of serum creatinine levels and estimated glomerular fltration rate (eGFR), which was calculated using the chronic kidney disease epidemiology collaboration formula.Furthermore, patients with CKD and a GFR below 60 mL/min/1.75m 2 were also considered for inclusion in the study.
2.6.Statistical Analysis.All statistical analyses were conducted using SPSS, version 23.Before conducting analysis, we assessed the normality of data using the Shapiro-Wilk test.In the descriptive analysis of the demographic distribution, categorical variables were presented as percentages, while continuous variables were expressed as mean-± standard deviation.AKI was defned in accordance with the Kidney Disease: Improving Global Outcomes (KDIGO) criteria.Given the unavailability of urine output data for all patients, the diagnostic criterion for AKI relied on serum creatinine levels.Our proposed KDIGO criteria for diagnosing AKI included an abrupt (within 48 hours) decline in kidney function, an absolute increase in serum creatinine of 0.3 mg/dL or more (≥26.4μmol/L), or a percentage increase in serum creatinine of 50% or more (1.5-foldfrom baseline).For patients with stage 3 or higher CKD, a 1.5-fold increase from baseline was considered a diagnostic criterion [18].Missing data were appropriately addressed through imputation techniques.We conducted imputation to ensure that the analysis and results are based on a complete dataset, minimizing the potential biases associated with missing information.Multicollinearity is assessed by calculating the variance infation factor (VIF) to evaluate the interaction between variables.
To examine the potential predictor factors for AKI, including baseline characteristics, medications, and laboratory data, bivariate and multivariate logistic regression analyses were conducted.Additionally, to identify the most signifcant risk factor for mortality in MICU patients, a multivariate model was developed using predictors that exhibited a p value of 0.05 or less in the univariate models.Te adjusted odds ratio (AOR) and its corresponding 95% confdence interval (CI) were then calculated for the multivariate model.

Demographic Factors.
Te Shapiro-Wilk test (W � 0.59, p � 0.16) indicated that our data did not signifcantly deviate from a normal distribution.We present the characteristics of the study sample in Table 1.Te average mean age was 63.8 ± 18 years (range: 18-97 years).More than half of the participants (50.5%) were men.Acute pulmonary diseases were considered the most common indication for admission to the ICU, followed by sepsis, with a percentage of 34.4% and 24.3%, respectively.Out of several comorbidities, HTN was the most prevalent disease with a percentage of 64.5%, while the prevalence of DM and chronic pulmonary diseases was 52.3% and 25.4%, respectively.Te mean serum creatinine level at time of admission was 1.1 ± 1.1 mg/dl, and that in the AKI group was 1.2 ± 1.4 compared to 0.9 ± 0.8 with a p value of 0.03.Te average eGFR at the time of admission was 67.3 ± 23.6 mL/min/1.73m 2 .Approximately 33.6% (140 of the patients) had CKD (GFR < 60 mL/min/ 1.75 m 2 ).Besides, 75.2% of patients were using PPI medications, while 23.9%, 53.2%, and 15.1% of patients were using ACEIs/ARBs, diuretics, and NSAIDs, respectively.
Among the 457 nonsurgical patients, a total of 196 patients (42.8%) developed AKI during their stay in the MICU.Notably, acute pulmonary diseases (n � 52, 34.5%) and sepsis (n � 40, 20.4%) were identifed as the primary causes of admission in AKI patients.To provide further insights, we compared the percentages of total patients and the number of patients who developed AKI in relation to comorbidities.Te fndings are depicted in Figure 2.

Prevalence and Risk Factors of AKI.
Te prevalence of AKI among MICU patients is shown in Figure 3.

Interventions and Outcomes of AKI.
Te analysis illustrated a signifcant association between AKI and the use of vancomycin (p < 0.001), while the use of other antibiotics (colistin and aminoglycosides) did not show such an association (Table 3).Moreover, AKI was signifcantly associated with a higher mortality rate, with 35.7% in the AKI group compared to 26.9% in the non-AKI group (p < 0.001).Approximately 18.7% (34 patients) in the AKI group required renal replacement therapy.No signifcant association was found between the use of vasopressors and ventilators and AKI.Using a multivariate binary logistic regression model, vancomycin use was found to increase the risk of AKI among nonsurgical critically ill patients (AOR: 10; 95% CI: 4.7-24.0;p < 0.001).

Discussion
Tis study reveals AKI prevalence and risk factors in critically nonsurgical patients admitted to JUH with medical conditions.We found that the prevalence of AKI is approximately 42.8%.Patients with AKI are at a higher risk of morbidity and mortality, have a longer hospital stay, and have an increased risk of developing CKD [1].Tis may be related to the selected cutof-point for the diagnostic criteria of developing AKI.A systematic review found that AKI among critically ill patients ranges from 14.4 to 39% [19].
Consistent with the fndings of Melo et al. [20], in a recent systematic review encompassing studies from both developed and developing countries, the prevalence of AKI was also approximately 40%.Tis similarity in AKI prevalence further supports the existing body of literature across different regions.Another study, which published the results of a multination study, showed the overall AKI prevalence as 57.3% [21].In diferent international multicenter studies, the occurrence rate of the condition in patients from both developing and developed countries was approximately 19% 6 Critical Care Research and Practice [22].Results revealed that the mortality rate among all nonsurgical patients is 26.9% and signifcantly increases to 35.7% in patients who developed AKI during their MICU stay.Notably, this is considerably less than results reported in a previous study, which found the mortality rate to be 50-70% [23], and higher than the one in the previous study of Hashemian et al. [24], which reported a percentage of 19.7%.Other factors, such as the patient's condition, the severity of the illness, and older age, contribute to the mortality rate.Also, we found that comorbidities including DM, HTN, pulmonary diseases, and CKD are associated with a higher risk of developing AKI [25].Previous studies showed an association between chronic pulmonary diseases and acute exacerbations of chronic obstructive pulmonary disease and the risk of AKI, which may be related to using positive pressure ventilators, developing hypoxemia and systemic infammation, and diferent cytokines [26,27].Tis study concurs with a previous study by Cao et al. [27] and challenges that by Hamid et al. [28], which found that sepsis (74.3%) was the primary source of developing AKI among nonsurgical hospitalized critically ill patients in Malaysia.PPIs are widely prescribed for gastrointestinal disorders that are predominant among older adults, including dyspepsia, gastric ulcer, and gastroesophageal refux disease.Despite a high prevalence of PPI usage, we did not fnd an increased risk of AKI.Tis observation aligns with a study conducted by Lee et al. [29], where they also reported no signifcant association between preadmission PPI use and AKI in critically ill patients, even after accounting for demographic factors, illness severity, and the indication for PPI use.
We found that use of ACEI/ARB drugs is associated with AKI among nonsurgical patients.Tis study concurs with a previous study by Neyra and Leaf [30], which found that the patients who were taking ACEIs/ARBs at admission were at a higher risk of developing AKI.Tis can be explained by an increase in the mechanism of action of ACEI/ARB by decreasing intraglomerular pressure, which may contribute  Critical Care Research and Practice to AKI risk.Another study found that previous use of ACEI/ ARB may be related to in-hospital mortality rate among MICU patients [31].Long-term and continued use of ACEI/ ARB after MICU discharge and during renal recovery is recommended, with cautious evaluation.It is widely known that NSAIDs directly interfere with renal function and increase the risk of AKI.We found that vancomycin is associated with a higher risk of AKI especially when exceeding the level of 20 mg/l.Other previous studies indicated that vancomycin drug increased AKI among ICU patients [8,32].
Regarding mortality, we found that older age, diuretic drug use prior to hospital admission, and ventilator and vasopressor use during MICU stay are linked to an increased risk of death.Most antibiotic drugs used in critically ill patients are not associated with mortality rate, hospital length of stay, or malignancy.Te risks of AKI and mortality rates were noted in a previous study [32].Surprisingly, the use of diuretics is linked to a two-fold (2.12 times) increased mortality rate.In 2019, they found that diuretic drugs are useful tools in the management of AKI-associated volume overload, and they are improbable to be inherently toxic when used appropriately; however, they are also improbable to have a therapeutic role in AKI as a "stand-alone" therapy by preventing fuid overload [33].Vasopressors are linked to an increased risk of death but did not cause AKI.Tis result requires more investigation because it may be related to early death even before developing AKI.Alternatively, this could be attributed to the clinical approach adopted in our institution, which involves the early initiation of vasopressors.Tis approach may result in reduced utilization of resuscitation fuids, decreased fuid accumulation, and potentially shorter durations of hypotension.Tese fndings are consistent with those reported by Ospina-Tascón et al. [34], who demonstrated that the early initiation of vasopressors was not signifcantly associated with an increased risk of acute renal failure or the need for renal replacement therapy.A previous study found that inotrope use contributes to both AKI and mortality rates [35].Invasive mechanical ventilator use is associated with a threefold increase in the odds of AKI in critically ill patients [36].Tis may indicate that patients die early in the MICU, even before developing AKI.
Statin use was found to have a protective efect on critically ill patients.Tese fndings align with a meta-analysis conducted by Yu et al. [37], which demonstrated that preadmission statin use was associated with favorable outcomes in critically ill patients, including lower short-term mortality rates.

Limitation.
Te utilization of a retrospective study conducted in a single-center hospital represents a limitation of this study.Insufcient data in medical records prevented the assessment of additional variables such as body mass index, urine output volume, and fuid management strategies, which also constitutes a potential limitation.We also did not include factors like nutritional status, genetics, and drug interactions in this study.Future research should explore these variables to enhance our understanding of AKI.Moreover, studying the quality of life among AKI patients is needed [38].Nonetheless, this study benefted from a comprehensive evaluation of various preadmission and postadmission factors within the MICU, with minimal missing data points, highlighting the study's strengths.Moreover, this study is the frst in the region to employ AKIN criteria for diagnosing AKI, adding novelty to the fndings.However, considering the aforementioned limitations, a multicenter prospective study with a larger sample size is warranted to further explore these fndings.

. Conclusion
Tis study highlights that AKI is one of the most severe complications (42.8%) among nonsurgical patients during MICU stay in JUH and is associated with short-and longterm morbidity and mortality outcomes (35.7%).Identifed risk factors for AKI are DM, HTN, CKD, pulmonary causes, smoking, and home medications, including ACEIs/ARBs and NSAIDs.Older patients using diuretics, vasopressors, and ventilators are at higher risk of mortality.Te use of statins has a protective efect with a lower mortality rate.Further prospective studies are required to identify the risk factors for AKI in nonsurgical critically ill patients and provide certain preventive measures.

Figure 2 :
Figure 2: Comparison between comorbidities among total participants and those who developed AKI.

Table 1 :
Te overall mortality rate for nonsurgical patients during their MICU stay was 26.9%.We determined the risk factors for mortality by univariate analysis, including age, baseline comorbidities, medications at admission, cause of admission, interventions during MICU stay such as ventilator and vasopressor use, antibiotics, AKI during MICU stay, and laboratory results.Baseline characteristics and factors contributing to the occurrence of AKI.

Table 4 :
Multivariate binary logistic regression analysis of factors associated with MICU mortality (N � 457).