Abstract
Introduction
Urea is a toxin present in acute kidney injury (AKI). We hypothesize that reduction in serum urea levels might improve clinical outcomes. We examined the association between the reduction in urea and mortality.
Methods
Patients with AKI admitted to the Hospital Civil de Guadalajara were enrolled in this retrospective cohort study. We create 4 groups of urea reduction ratio (UXR) stratified by their decrease in urea from the highest index value in comparison to the value on day 10 (0%, 1–25%, 26–50%, and >50%), or at the time of death or discharge if prior to 10 days. Our primary endpoint was to observe the association between UXR and mortality. Secondary observations included determination of which types of patients achieved a UXR >50%, whether the modality of kidney replacement therapy (KRT) effected changes in UXR, and if serum creatinine (sCr) value changes were similarly associated with patient mortality.
Results
A total of 651 AKI patients were enrolled. The mean age was 54.1 years, and 58.6% were male. AKI 3 was present in 58.5%; the mean admission urea was 154 mg/dL. KRT was started in 32.4%, and 18.9% died. A trend toward decreased risk of death was observed in association with the magnitude of UXR. The best survival (94.3%) was observed in patients with a UXR >50%, and the highest mortality (72.1%) was observed in patients achieving a UXR of 0%. After adjusting for age, sex, diabetes mellitus, CKD, antibiotics, sepsis, hypovolemia, cardio-renal syndrome, shock, and AKI stage, the 10-day mortality was higher in groups that did not achieve a UXR of at least 25% (OR: 1.20). Patients achieving a UXR >50% were most likely initiated on dialysis due to a diagnosis of the uremic syndrome or had a diagnosis of obstructive nephropathy. Percentage change in sCr was also associated with increased mortality risk.
Conclusions
In our retrospective cohort of AKI patients, the percent decrease in UXR from admission was associated with a stratified risk of death. Patients with a UXR >25% had the best associated outcomes. Overall, a greater magnitude in UXR was associated with improved patient survival.
Keywords: Acute kidney injury, Urea, Uremic syndrome, Mortality
Introduction
One of the primary functions of the kidney is the excretion of waste products generated during metabolism, maintaining the “internal milieu” and homeostasis. One of the main waste products of metabolism is urea (also known as carbamide). Urea is a volatile organic compound with the chemical formula CH₄N₂O, which is the end-product of protein and nitrogen metabolism. Urea is found in very high concentrations in the blood of patients with kidney disease [1]. Due to its elevation during kidney dysfunction, the metabolic abnormalities that occur are referred to as the “uremic syndrome” or “urine in the blood” [2].
The uremic syndrome includes the development of complications such as encephalopathy, pericarditis, bleeding, endothelial dysfunction, protein carbamylation, disruption of the intestinal barrier, atherosclerosis, and death [3]. Previous studies showed that urea itself induced molecular changes related to insulin resistance, free radical production, apoptosis, and disruption of the protective intestinal barrier. In addition, urea is at the origin of the generation of cyanate, ammonia, and carbamylated compounds, which have all been linked to biological changes. Carbamylation has been held responsible for post-translational protein modifications that are involved in atherogenesis, and other functional changes. In observational clinical studies, these carbamylated compounds were associated with cardiovascular morbidity and overall mortality. Yet, urea cannot be held responsible for all the complex metabolic and clinical changes seen in the uremic syndrome [2]. It is important to note that the clinical associations of urea, and its related compounds, have not always been consistently shown to be tied to patient outcomes [4].
During acute kidney injury (AKI), an elevated urea level is almost universal [5], although it is not always secondary to kidney dysfunction. In critically ill patients, it can be due to multiple causes such as upper intestinal bleeding, catabolism, hypovolemia, or medications [6]. “Uremia” during AKI has negative multiorgan impacts due to various mechanisms. These include intestinal dysbiosis promoting systemic inflammation due to the translocation of intestinal bacteria to the portal circulation, generation of toxins by the colon, insulin resistance and hyperglycemia, hypercarbamylation of cardiac proteins that may induce myocardial inflammation, increased infiltration of activated monocytes, and post-translational modification-derived products as important pathogenic mediators of acute diseases [6]. Due to the correlation between elevated urea and clinical complications in patients with AKI, it is reasonable to consider that decreasing levels of urea could be accompanied by improved clinical outcomes. For this reason, therapies for AKI aim to substantially decrease urea levels [7]. In fact, new evidence even suggests that there is a mortality benefit when starting kidney replacement therapy (KRT) during AKI in the face of high blood urea nitrogen values [8].
To contribute additional data to this highly controversial topic, we explored the association between the reduction in urea values and mortality in hospitalized patients with AKI. We hypothesize that a greater magnitude in the reduction of blood urea might be beneficial and associated with improved patient survival.
Materials and Methods
Patients with AKI who had available admission serum urea levels and at least three additional urea measurements during hospitalization were included. We chose the highest peak of urea during the first 10 days of hospitalization upon which to calculate our urea reduction ratio (UXR), defined as the percentage of urea reduction between the highest peak and the last urea value (death, discharge or at day 10). UXR was calculated as follows: (peak urea – last urea/peak urea) × 100. We created quartile divisions of UXR and divided by the magnitude of their percentage decrease: UXR 0%, UXR 1–25%, UXR 26–50%, and UXR >50%.
The definition of universally known reduction in urea (URR <65%) has been used in patients with CKD on hemodialysis (HD) to evaluate the effectiveness of a single session in reducing pre-HD urea values compared to post-HD. This definition does not capture the spectrum contemplated by our study for the following reasons: a) Our cohort included only AKI patients who were treated with different modalities of KRT (HD and peritoneal dialysis [PD]) and even without them. b) The definition of UXR that we use is during a period of 10 consecutive days, from the moment it is known by nephrology until day 10, which is completely different from the one mentioned above. c) We have designed the definition of UXR based on quartiles of percentage reduction; with this, we try to simplify into groups easily identifiable by the clinician.
Our KRT prescriptions were individualized according to clinical needs. We prescribed intermittent hemodialysis (iHD) sessions every 48 h with 1.7 filter dialyzers, 350 cc blood flow, and 500 cc dialysis flow during a 3.5-h session. Ultrafiltration was determined according to volume status. The prescription of PD was between 18 and 30 exchanges per day (adjusted for weight in kilograms), with a 2-L exchange of various glucose concentrations depending on volume status. The withdrawal of KRT was determined by the operators according to patient needs and clinical trajectory. Additionally, we analyzed the trajectory of serum creatinine (sCr) and divided its trajectory into quintiles: −100% to −50%, −49% to 0%, 1–50%, 51–100%, and more than 100% increase from baseline. AKI was diagnosed by sCr according to KDIGO criteria [9]. We excluded patients with chronic kidney disease (CKD) stage 5 (defined as an estimated glomerular filtration rate of <15 mL/min/1.73 m2, using the 4-variable Modification of Diet in Renal Disease Study equation, MDRD-4) [9], patients on chronic dialysis, patients with <48-h hospital stay, transplant patients, pregnant patients, and those with missing data. The study was approved by the Hospital Civil de Guadalajara Fray Antonio Alcalde Institutional Review Board (HCG/CEI-0550/15) and conducted in adherence with the Declaration of Helsinki. The prospective review of patient data did not require written informed consent from participants in accordance with local/national guidelines. The protocol followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [10].
Data Collection
Clinical characteristics, demographic information, and laboratory data were collected prospectively using automated retrieval from the institutional electronic medical record (EMR). UXR was calculated based upon urea value reduction during the study follow-up. For those who had more than one urea measurement on the same calendar day, the mean urea level was calculated to represent the urea value for the index day. Demographic and clinical variables were collected including age, diabetes, hypertension, hypothyroidism, CKD grade, smoking, cerebrovascular disease, and ischemic heart disease. Baseline sCr level was defined as the most recent value within a year before admission. Biochemical data were prespecified. KRT indications were determined by the consulting physician and based upon the usual criteria of fluid overload with resistance to diuretics, severe hyperkalemia, severe metabolic acidosis, and uremic manifestations including encephalopathy, pericarditis, and convulsions. Patients were evaluated and followed for the first 10 days after AKI diagnosis. We chose a 10-day follow-up because most AKI patients who require dialysis during hospitalization start KRT within this time frame [9]. We also wanted to evaluate the patient’s kidney function within the first few days of an ICU admission, given the high risk of mortality associated with AKI.
Study Outcomes
The primary outcome was in-hospital mortality, by UXR percentages, within 10 days of admission. Additional observations included the characteristics of patients who had the greatest magnitude of UXR and by which KRT modality (HD, PD, or neither/conservative treatment [C]) this goal was achieved. Additionally, we observed the association that existed between the trajectory of sCR changes and mortality.
Statistical Analysis
Continuous variables are summarized as mean ± SD unless otherwise specified. Categorical variables were summarized as numbers with percentages. Survival was compared using the log-rank test, censoring for dead during the follow-up. Patients were followed until death, hospital discharge, or up to 10 days after hospital admission. Multivariable analysis was performed to assess the independent association between in-hospital sCr and urea value (mg/dL), UXR groups, and 10-days mortality. To perform the multivariate analysis, a logistic regression was made; the model 1 was a univariate analysis, and in model 2, we adjusted for age, sex, diabetes mellitus, CKD, antibiotics, sepsis, hypovolemia, cardio-renal syndrome, shock, and AKI stage. We performed a stratified subgroup analysis by age (<65 or ≥65 years), sex, the presence of diabetes mellitus, CKD, hypovolemia, sepsis, shock, cardio-renal syndrome, and AKI KDIGO 2–3. The univariate analysis was performed, estimating the dependence of the 10-days risk of death on urea (mg/dL) reduction as a continuous variable. A two-tailed p value of <0.05 was considered statistically significant. Statistics output was generated by a software package (R Studio, 1.3.1093).
Results
Patients were identified in the EMR from August 2017 to December 2021 who required a nephrology consult because of AKI. 202 patients were excluded for having CKD G5 (56) or a hospital stay <48 h (92), and 54 were excluded because of lack of data. A total of 651 patients were included for analysis (Fig. 1).
Fig. 1.
Flowchart of study population.
The baseline characteristics of the four UXR groups are shown in Table 1. Half of the patients were in the UXR 0% group, and the minority (13.3%) were in the UXR >50% group. The average age was 54.1 (SD ± 18.5) years, which was similar in all groups. Males prevailed in 59.2% of cases, and 35% had diabetes and hypertension. Compared to patients with an UXR 0%, patients with an UXR >50% had less baseline hypertension and CKD G1–4. They also used fewer diuretics and statins. More than half of the patients had AKI 3 (58%), and the most common etiologies were sepsis (45.9%) and hypovolemia (23%). One out of 4 patients had shock as the etiology of their AKI. The mean admission urea value was 154 mg/dL. Compared to the UXR 0% group, those with UXR >50% had a lower urea concentration at baseline (129.7 mg/dL) and required KRT more often (37.9%). The most common indication for starting KRT was fluid overload. The total mortality was 18% and was significantly lower in the UXR >50% group (5.7%).
Table 1.
Baseline clinical characteristics by urea reduction percentage in hospitalized AKI patients
| Variable | Total | UXR 0% |
UXR 1–25% |
UXR 26–50% |
UXR >50% |
|---|---|---|---|---|---|
| Total, n (%) | 651 (100) | 330 (50.7) | 137 (21.0) | 97 (15.0) | 87 (13.3) |
| Age, years, mean (SD) | 54.1 (18.5) | 54.0 (18.6) | 53.1 (19.4) | 53.9 (17.9) | 56.6 (17.3)) |
| Male, n (%) | 389 (59.2) | 196 (59.4) | 82 (59.9) | 60 (61.9) | 51 (58.6) |
| Diabetes, n (%) | 234 (35.9) | 126 (38.9) | 50 (36.5) | 38 (39.2) | 20 (23.0) |
| Hypertension, n (%) | 231 (35.5) | 132 (40.0) | 48 (35.0) | 30 (30.9) | 21 (24.1) |
| Smoker, n (%) | 69 (10.6) | 30 (9.1) | 15 (10.9) | 13 (13.4) | 11 (12.6) |
| Hypothyroidism, n (%) | 19 (2.9) | 9 (2.7) | 1 (0.7) | 6 (6.2) | 3 (3.4) |
| Chronic kidney disease grade 1–4, n (%) | 171 (26.3) | 92 (27.9) | 43 (31.4) | 20 (20.6) | 16 (18.4) |
| Cerebrovascular disease, n (%) | 31 (4.8) | 14 (4.2) | 7 (5.1) | 3 (3.1) | 7 (8.0) |
| Ischemic heart disease, n (%) | 31 (4.8) | 20 (6.1) | 5 (3.6) | 3 (3.1) | 3 (3.4) |
| Medications during hospitalization | |||||
| NSAIDs, n (%) | 192 (29.5) | 95 (28.8) | 43 (31.4) | 31 (32.0) | 23 (26.4) |
| Antibiotics, n (%) | 479 (73.6) | 245 (74.2) | 101 (73.7) | 71 (73.2) | 62 (71.3) |
| Diuretics, n (%) | 252 (38.7) | 148 (44.8) | 51 (37.2) | 32 (33.0) | 21 (24.1) |
| Statins, n (%) | 94 (14.4) | 63 (19.1) | 18 (13.1) | 6 (6.2) | 7 (8.0) |
| Acetylsalicylic acid, n (%) | 58 (8.9) | 38 (11.5) | 11 (8.0) | 4 (4.1) | 5 (5.7) |
| AKI stages | |||||
| KDIGO-1, n (%) | 41 (6.3) | 19 (5.8) | 12 (8.8) | 6 (6.2) | 4 (4.6) |
| KDIGO-2, n (%) | 105 (16.1) | 56 (17.0) | 17 (12.4) | 15 (15.5) | 17 (19.5) |
| KDIGO-3, n (%) | 381 (58.5) | 186 (56.4) | 73 (53.3) | 61 (62.9) | 61 (70.1) |
| Acute on chronic kidney disease, n (%) | 124 (19.1) | 69 (20.9) | 35 (25.5) | 15 (15.5) | 5 (5.7) |
| Etiology of AKI | |||||
| Sepsis, n (%) | 299 (45.9) | 156 (47.3) | 64 (46.7) | 38 (39.2) | 41 (47.1) |
| Hypovolemia, n (%) | 150 (23.0) | 66 (20.0) | 34 (24.8) | 29 (29.9) | 21 (24.1) |
| Cardiorenal syndrome, n (%) | 99 (15.2) | 61 (18.5) | 23 (16.8) | 12 (12.4) | 3 (3.4) |
| Nephrotoxic drugs, n (%) | 40 (6.1) | 20 (6.1) | 7 (5.10) | 10 (10.3) | 3 (3.4) |
| Shock, n (%) | 170 (26.1) | 105 (31.8) | 26 (19.0) | 20 (20.6) | 19 (21.8) |
| Obstructive nephropathy, n (%) | 78 (12.0) | 29 (8.8) | 14 (10.2) | 15 (15.5) | 20 (23.0) |
| Serum urea, mg/dL, mean (SD) | 154.6 (81.9) | 166.6 (90.7) | 148.4 (75.6) | 145 (73.0) | 129.7 (54.5) |
| KRT, n (%) | 211 (32.4) | 102 (30.9) | 45 (32.8) | 31 (31.9) | 33 (37.9) |
| Cause of KRT | |||||
| Hyperkalemia, n (%) | 54 (8.3) | 30 (9.1) | 10 (7.3) | 9 (9.3) | 5 (5.7) |
| Metabolic acidosis, n (%) | 53 (8.1) | 29 (8.8) | 10 (7.3) | 10 (10.3) | 4 (4.6) |
| Fluid overload, n (%) | 85 (13.1) | 49 (14.8) | 14 (10.2) | 13 (13.4) | 9 (10.3) |
| Uremic syndrome, n (%) | 63 (9.7) | 25 (7.6) | 9 (6.6) | 12 (12.4) | 17 (14.9) |
| Hospital length of stay, days, mean (SD) | 5.7 (2.9) | 5.2 (3.1) | 5.6 (2.7) | 6.6 (2.8) | 6.6 (2.5) |
| Mortality, n (%) | 123 (18.9) | 92 (27.9) | 18 (13.1) | 8 (8.2) | 5 (5.7) |
Continuous data are presented as mean ± SD unless otherwise indicated; categorical data are presented as count (%).
NSAIDs, non-steroidal anti-inflammatory drugs; UXR, urea reduction.
We analyzed the change in urea in mg/dL and its association with mortality, separated into groups with 20 units of change (Table 2; Fig. 2). We found that a decreased urea value within the range of 61– 80 mg/dL was associated with a reduction in the risk of mortality by 15% (OR: 0.85, p = 0.03). Increased urea values in the ranges between 40–59 mg/dL and 60–80 mg/dL were associated with increased mortality of 27% and 14%, respectively (OR: 1.27 and OR: 1.14, p ≤ 0.05 for both).
Table 2.
Change in serum urea (mg/dL) and mortality risk
| Urea change, mg/dL | OR (95% CI) | p value |
|---|---|---|
| 80 to 60 | 1.14 (1.01–1.30) | 0.03* |
| 59 to 40 | 1.27 (1.10–1.45) | <0.001* |
| 39 to 20 | 0.99 (0.88–1.12) | 0.99 |
| 19 to 0 | 1.02 (0.94–1.12) | 0.50 |
| −1 to −20 | 0.94 (0.84–1.04) | 0.24 |
| −21 to −40 | 0.93 (0.93–1.03) | 0.20 |
| −41 to −60 | 0.89 (0.78–1.01) | 0.07 |
| −61 to −80 | 0.85 (0.73–0.99) | 0.05* |
Fig. 2.
Relationships between urea reduction and 10-days risk of death.
In Table 3, we describe the relationship between the UXR, KRT, and survival. The overall survival at 10 days was 81.1%; a trend toward lower mortality was observed in patients with greater UXR. The highest survival (96.7%) was achieved in the UXR >50% and KRT groups, and the lowest survival (53.9%) was seen in those patients who, despite having received KRT, had no reduction in urea levels.
Table 3.
Relationship between urea reduction and survival in aki patients
| N | Events | Survival, (%) | ||
|---|---|---|---|---|
| 10-day overall survival | 651 | 123 | 81.1 | 0.78–0.84 |
| 10-day survival by urea reduction percent | ||||
| >50% | 87 | 5 | 94.3 | 0.89–0.99 |
| 26–50% | 97 | 8 | 91.8 | 0.86–0.97 |
| 1–25% | 137 | 18 | 86.9 | 0.81–0.92 |
| 0% | 330 | 92 | 72.1 | 0.67–0.77 |
| 10-day survival by urea reduction percent with KRT or C | ||||
| >50% KRT | 33/87 | 1 | 96.9 | 0.91–1.00 |
| >50% C | 54/87 | 4 | 92.6 | 0.85–0.99 |
| 26–50% KRT | 31/97 | 4 | 87.1 | 0.76–0.99 |
| 26–50% C | 66/97 | 4 | 93.3 | 0.88–0.99 |
| 1–25% KRT | 45/137 | 10 | 77.8 | 0.66–0.90 |
| 1–25% C | 92/137 | 8 | 91.3 | 0.85–0.97 |
| 0% KRT | 102/330 | 47 | 53.9 | 0.45–0.64 |
| 0% C | 228/330 | 45 | 80.3 | 0.75–0.85 |
| 10-day survival by KRT modality or C | ||||
| C | 440 | 61 | 86.1 | 0.83–0.89 |
| HD | 182 | 53 | 70.9 | 0.64–0.77 |
| Peritoneal dialysis | 29 | 9 | 69.0 | 0.54–0.88 |
Log-rank test, p < 0.001.
C, conservative treatment; KRT, kidney replacement therapy.
We assessed the impact of KRT on mortality according to the magnitude of UXR (Table 4). In the unadjusted model, when compared to the UXR >50% and C, those with UXR >50% with KRT, UXR 26–50% with C, and UXR 1–25% with C were associated with a ∼15% decrease in the risk of death. In the same analysis, we observed an increased risk of death in those groups with small magnitudes of urea reduction: UXR 1–25% with KRT, UXR 0% with C, and the highest risk of death in the UXR 0% with KRT group (OR: 1.38, p ≤ 0.001).
Table 4.
The association between urea reduction with KRT or C and 10-day mortality
| Model 1a | Model 2b | |||
|---|---|---|---|---|
| OR (95% CI) | p value | OR (95% CI) | p value | |
| >50% C | 1 (reference) | – | 1 (reference) | – |
| >50% KRT | 0.84 (0.73–0.97) | 0.01 | 0.95 (0.81–1.11) | 0.57 |
| 26–50% C | 0.86 (0.78–0.95) | 0.004 | 1.01 (0.88–1.15) | 0.84 |
| 26–50% KRT | 0.93 (0.81–1.08) | 0.38 | 1.06 (0.90–1.24) | 0.44 |
| 1–25% C | 0.88 (0.81–0.96) | 0.006 | 1.04 (0.92–1.18) | 0.49 |
| 1–25% KRT | 1.15 (1.02–1.30) | 0.01 | 1.20 (1.04–1.39) | 0.009 |
| 0% C | 1.10 (1.03–1.17) | 0.001 | 1.15 (1.03–1.28) | 0.009 |
| 0% KRT | 1.38 (1.27–1.49) | <0.001 | 1.42 (1.26–1.28) | <0.001 |
aModel 1: unadjusted.
bModel 2: adjusted for age, sex, gender, diabetes mellitus, chronic kidney disease, antibiotics, sepsis, hypovolemia, cardio-renal syndrome, shock, and AKI stage.
KRT, kidney replacement therapy; C, conservative treatment.
The multivariate analysis (Table 4) adjusted for age, sex, diabetes mellitus, CKD, antibiotics, sepsis, hypovolemia, cardio-renal syndrome, shock, and AKI stage. An increased risk of death was associated with a low magnitude of urea reduction, UXR 1–25% with KRT, UXR 0% with C, and higher risk in UXR 0% with KRT (OR: 1.42, 1.26–1.28, p ≤ 0.001).
Because the greatest associated benefit in survival was observed in patients with a UXR >50%, we sought to determine the causes of initiation of KRT in this subgroup (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000530237). The only dialysis initiation indication associated with this reduction was the uremic syndrome. A urea reduction >50% was similarly achieved with either PD or HD (online suppl. Table 2).
To evaluate whether there was an association between the etiology of AKI and achieving UXR >50% according to KRT or C (online suppl. Table 3), we identified that only obstructive nephropathy treated with KRT was associated with achieving this outcome.
To identify subgroups of patients where the greatest UXR and mortality benefit was obtained, we performed an analysis according to age, sex, presence of diabetes, CKD, hypovolemia, sepsis, shock, cardiorenal syndrome, and severity of AKI. In Figure 3, KRT was associated with a decreased risk of mortality in younger male patients without diabetes, CKD, hypovolemia, sepsis, shock, or cardiorenal syndrome. The number of events separating these subgroups was small. To analyze whether the group with a UXR >50% had any association with mortality according to the subgroups that were managed with C, we noted that those who did not have sepsis, shock, or cardiorenal syndrome had a lower risk of mortality (Fig. 4).
Fig. 3.
Subgroup analysis of patients with urea reduction >50% with KRT and mortality.
Fig. 4.
Subgroup analysis of patients with urea reduction >50% with C and mortality.
In addition, we analyzed the trajectory of the percentage change in sCr and its relationship with mortality (online suppl. Table 4). We identified a significant association of decreased risk among those who reduced sCr (−100% to −50% and −49% to 0%) with an OR: 0.35 and 0.58, p ≤ 0.05 for both. Additionally, an increased risk was observed in patients with an increase in their sCr percentage (1–50%, 51–100%) with an OR of 2.75, 4.36, p < 0.05 in both, and for those patients with a >100% an OR of 1.54, p = 0.52. We performed a ROC curve analysis of change in sCr (%) and mortality, with a cut off value of >13.33, sensitivity 76.5, specificity 63.6, and AUC 0.714, p ≤ 0.001 (online suppl. Fig. 1). In order to identify if there was a linear relationship with the percentages of change in sCr and mortality, we performed a LOESS curve and observed that there was no such relationship. For example, a 50% increase in sCr showed inconsistent risk and actually the same risk as a decrease of −50% (online suppl. Fig. 2).
Discussion
In this retrospective cohort study of AKI patients, we observed that half of them did not have a sustained change from the initial urea value during their hospitalization. A serum urea value that was equivalent to admission values, after 10 days of hospitalization, was associated with higher mortality. In those patients with a reduction of urea values of at least 25% from admission, there was an associated lower probability of death during hospitalization. In fact, the mortality risk reduction associated with UXR was more evident when the magnitude of the UXR was greater.
We have defined UXR based on quartiles of reduction. With this, we tried to simplify identifiable groups to the clinician. As far as we are aware, this definition has never been used before, and we could not therefore compare it with other studies. The ELAIN trial showed that the patients with the best survival were those who reached urea values ∼82.5 mg/dL (∼13.6 mmol/L) [11]. Based on this finding, we concluded that achieving a UXR >50% (mean urea value at baseline of 154 mg/dL [25.6 mmol/L]) would represent a urea of ∼80 mg/dL (∼13.3 mmol/L), and that this level of urea clearance might be associated with a decreased risk of death.
High urea values have been linked to the uremic syndrome that complicates cases with AKI. Historically, lowering urea values has been one of the main objectives in the management of these patients, despite scarce evidence to support this practice. The KDIGO guidelines consider a high urea value as a potential reason to initiate KRT, although there is no specific urea target value [12]. Different studies [8, 11] have suggested urea values of ∼40 mmol/L (∼240 mg/dL) as thresholds to consider starting KRT in AKI, without direct evidence of improved outcomes versus starting KRT at higher levels [5]. Even a study that evaluated the probability of dying according to different urea values at the beginning of KRT in AKI patients, ranging from 75 mg/dL to 336 mg/dL, described areas under the curve considered poor in predicting mortality. However, it could be argued that this information is from studies conducted nearly 2 decades ago and may not represent current practices or management of AKI [13].
It is reasonable to think that in AKI, the reduction in urea could have positive effects such as limiting pericarditis [14], treating encephalopathy and increasing neurological capacity with the possibility of extubation [15], changing intestinal dysbiosis, decreasing systemic inflammation [16], and treating uremic bleeding [17]. All these circumstances may coexist in critically ill patients, and currently there are no specific treatments for these complications outside of KRT in the uremic patient. In line with this hypothesis, we found that there is symmetry between the magnitude of the decrease in urea and the magnitude of the risk of dying. This further suggests that defining a value for the magnitude of urea reduction as a goal for therapy may be beneficial. In our cohort, we found that a >25% reduction was advantageous to patient outcomes. It is important to consider that none of the recent clinical trials in KRT initiation have described urea reduction in their protocols, nor its association with mortality [8, 11].
We observed a higher survival (96%) in patients with >50% urea reduction with KRT. In contrast, almost half of patients with no reduction in urea with KRT died during follow-up. This reflects a sicker population with a greater hypercatabolic state, sarcopenia, persistent AKI, or hypovolemia. Additionally, this group of patients might represent those who had received inadequate KRT, increasing the risk of death [18].
We observed that the uremic syndrome was the only indication for the initiation of KRT that was associated with a higher probability of achieving UXR >50% (online suppl. Table 1). The decision to categorize these patients as uremic could be influenced by the existing historical dogma [2] and the recommendations of the KDIGO guideline to start KRT in the presence of this complication [12].
In determining which modality of KRT was the most effective to achieve UXR, we found that, compared to C, UXR >50% was obtained similarly with PD and HD (online suppl. Table 2). This is consistent with previous publications that have shown that both modalities are equally effective in lowering urea [19, 20]. It could suggest that in the face of AKI with uremia, any form of KRT can be used to decrease urea values and can be equally effective.
We noted that there are certain groups of patients who, according to the etiology of AKI, achieve UXR 50% (online suppl. Table 3). In fact, only obstructive nephropathy treated with KRT was associated with achieving this goal. AKI due to obstructive nephropathy is known as a diagnosis with high potential to be resolved effectively and quickly after reversal of urinary obstruction [21]. It is possible that in these cases, the timely correction of the urinary tract obstruction, together with KRT, had a synergistic effect on UXR.
Interestingly, patients who achieved UXR >50% with C had a lower risk of dying (Fig. 4). Significantly, these patients did not have sepsis, shock, or cardiorenal syndrome as etiologies of their AKI. This finding is expected and agrees with the hypothesis that renal recovery, without the need for KRT, occurs in less severe illness, and is thus associated with lower mortality [22]. In this context, it is intuitive to think that a kidney with better baseline function has a faster recovery capacity and greater effectiveness to excrete urea.
It is worth mentioning that in the setting of ESKD patients on dialysis, there was no benefit in lowering urea. Intensive dialysis in ESKD patients, as studied in the randomized control trials [23–25], significantly improved urea clearance but failed to improve mortality. However, ESKD patients differ physiologically from patients with AKI, and those conclusions should not necessarily be extrapolated to the AKI population. Our findings, which establish the association between greater urea reduction and reduction in mortality in this AKI cohort, could influence treatment optimization in certain clinical scenarios.
Regarding the trajectory of sCr and its relationship with mortality, we noted a decrease in the mortality risk with higher percentage reductions in sCr, although this trend did not have a linear relationship. Intuitively, it could be assumed that recovery of renal function by sCr is associated with a better prognosis, but there are limitations to estimating renal function with sCr in hospitalized patients with AKI.
Our study has several limitations. As a prospective cohort study, we can establish associations but not a causal relationship between the primary variables of interest and outcomes. Even though we adjusted for numerous variables, residual confounding parameters may still exist. The number of patients studied was relatively small, and we did not recruit variables that could contribute to the generation or reduction of urea such as nutrition, intestinal bleeding, and catabolism rates. Finally, we did not study the magnitude of the percentage of UXR with KRT therapies.
There were some strengths in our cohort, including the prospective follow-up. We were able to capture percentages of UXR that discriminated against the risk of death. We identified subgroups of patients who could receive more benefit than others with UXR. Finally, we were able to distinguish between those who received KRT with different modalities and patients who received C, a strategy that had not been demonstrated before.
Conclusions
We conclude that in patients with AKI, those who had no reductions in blood urea values despite treatment were associated with an increased risk of mortality. Those patients who achieved a urea reduction of at least 25% appeared to nullify this risk. Our results support the implementation of future clinical trials in which to evaluate the association of urea reduction in AKI with mortality.
Statement of Ethics
The study was approved by the Local Ethics Committee at Hospital Civil de Guadalajara (protocol HCG/CEI-0550/15). This prospective review of patient data did not require written informed consent from participants in accordance with local/national guidelines.
Conflict of Interest Statement
The authors declare no conflicts of interest.
Funding Sources
There were no funding sources.
Author Contribution
Jonathan S. Chávez-Íñiguez, Pablo Maggiani-Aguilera, and David González-Barajas were responsible for the design, analysis, and interpretation of the data. Lilia Rizo-Topete, Pablo Galindo, Brian Rifkin, Guillermo Navarro-Blackaller, Ramón Medina-González, Luz Alcantar-Vallin, and Guillermo García-García reviewed the manuscript. Gael Chávez-Alonso, Alma I. Martínez-Aguilar, Cristina Pérez-Hernández, Karla Hernández-Morales, Jahir R. Camacho-Guerrero, Miguel A. Pérez-Venegas, Alexa N. Oseguera-González, Cesar Murguia-Soto, and Karina Renoirte-López were responsible for recruiting data. All authors have read and approved the manuscript.
Funding Statement
There were no funding sources.
Data Availability Statement
Data are not publicly available due to ethical reasons. Further inquiries can be directed to the corresponding author.
Supplementary Material
Supplementary Material
Supplementary Material
Supplementary Material
Supplementary Material
Supplementary Material
References
- 1. Almeras C, Argilés A. The general picture of uremia. Semin Dial. 2009 Jul-Aug;22(4):329–33. 10.1111/j.1525-139X.2009.00575.x. [DOI] [PubMed] [Google Scholar]
- 2. Vanholder R, Gryp T, Glorieux G. Urea and chronic kidney disease: the comeback of the century? (in uraemia research). Nephrol Dial Transpl. 2018 Jan 1;33(1):4–12. 10.1093/ndt/gfx039. [DOI] [PubMed] [Google Scholar]
- 3. do Nascimento GVR, Balbi AL, Ponce D, Abrão JMG. Early initiation of dialysis: mortality and renal function recovery in acute kidney injury patients. J Bras Nefrol. 2012;34(4):337–42. 10.5935/0101-2800.20120022. [DOI] [PubMed] [Google Scholar]
- 4. Duranton F, Cohen G, De Smet R, Rodriguez M, Jankowski J, Vanholder R, et al. Normal and pathologic concentrations of uremic toxins. J Am Soc Nephrol. 2012 Jul;23(7):1258–70. Erratum in: J Am Soc Nephrol. 2013 Dec;24(12):2127-9. 10.1681/ASN.2011121175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Mackenzie J, Chacko B. An isolated elevation in blood urea level is not “uraemia” and not an indication for renal replacement therapy in the ICU. Crit Care. 2017 Nov 13;21(1):275. 10.1186/s13054-017-1868-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Verdier V, Soulage CO, Koppe L. New clinical evidence for urea toxicity. Nephrol Dial Transpl. 2021 Dec 31;37(1):1–4. 10.1093/ndt/gfab269. [DOI] [PubMed] [Google Scholar]
- 7. Gibney N, Hoste E, Burdmann EA, Bunchman T, Kher V, Viswanathan R, et al. Timing of initiation and discontinuation of renal replacement therapy in AKI: unanswered key questions. Clin J Am Soc Nephrol. 2008 May;3(3):876–80. 10.2215/CJN.04871107. [DOI] [PubMed] [Google Scholar]
- 8. Gaudry S, Hajage D, Martin-Lefevre L, Lebbah S, Louis G, Moschietto S, et al. Comparison of two delayed strategies for renal replacement therapy initiation for severe acute kidney injury (AKIKI 2): a multicentre, open-label, randomised, controlled trial. Lancet. 2021 Apr 3;397(10281):1293–300. 10.1016/S0140-6736(21)00350-0. [DOI] [PubMed] [Google Scholar]
- 9. Kellum JA, Lameire N, Aspelin P, Barsoum RS, Burdmann EA, Goldstein SL, et al. Kidney disease: improving global outcomes (KDIGO) acute kidney injury work group. KDIGO clinical practice guideline for acute kidney injury. Kidney Int Supplements. 2012;2(1):1–138. [Google Scholar]
- 10. Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Rev Esp Salud Publica. 2008;82(3):251–9. 10.1590/s1135-57272008000300002. [DOI] [PubMed] [Google Scholar]
- 11. Zarbock A, Kellum JA, Schmidt C, Van Aken H, Wempe C, Pavenstadt H, et al. Effect of early vs delayed initiation of renal replacement therapy on mortality in critically ill patients with acute kidney injury: the ELAIN Randomized Clinical Trial. JAMA. 2016;315(20):2190–9. 10.1001/jama.2016.5828. [DOI] [PubMed] [Google Scholar]
- 12. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Workgroup . KDIGO clinical practice guideline for acute kidney injury. Kidney Int Suppl. 2012;2:1–138. [Google Scholar]
- 13. De Corte W, Vanholder R, Dhondt AW, De Waele JJ, Decruyenaere J, Danneels C, et al. Serum urea concentration is probably not related to outcome in ICU patients with AKI and renal replacement therapy. Nephrol Dial Transpl. 2011 Oct;26(10):3211–8. 10.1093/ndt/gfq840. [DOI] [PubMed] [Google Scholar]
- 14. Chugh S, Singh J, Kichloo A, Gupta S, Katchi T, Solanki S. Uremic- and dialysis-associated pericarditis. Cardiol Rev. 2021 Nov-Dec 01;29(6):310–3. 10.1097/CRD.0000000000000381. [DOI] [PubMed] [Google Scholar]
- 15. Seifter JL, Samuels MA. Uremic encephalopathy and other brain disorders associated with renal failure. Semin Neurol. 2011 Apr;31(2):139–43. Epub 2011 May 17. 10.1055/s-0031-1277984. [DOI] [PubMed] [Google Scholar]
- 16. Rydzewska-Rosołowska A, Sroka N, Kakareko K, Rosołowski M, Zbroch E, Hryszko T. The links between microbiome and uremic toxins in acute kidney injury: beyond gut feeling-A systematic review. Toxins. 2020 Dec 11;12(12):788. 10.3390/toxins12120788. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Galbusera M, Remuzzi G, Boccardo P. Treatment of bleeding in dialysis patients. Semin Dial. 2009 May-Jun;22(3):279–86. 10.1111/j.1525-139X.2008.00556.x. [DOI] [PubMed] [Google Scholar]
- 18. Gunst J, Kashani KB, Hermans G. The urea-creatinine ratio as a novel biomarker of critical illness-associated catabolism. Intensive Care Med. 2019 Dec;45(12):1813–5. 10.1007/s00134-019-05810-y. [DOI] [PubMed] [Google Scholar]
- 19. Gabriel DP, Caramori JT, Martim LC, Barretti P, Balbi AL. High volume peritoneal dialysis vs daily hemodialysis: a randomized, controlled trial in patients with acute kidney injury. Kidney Int. 2008 Apr;73(108):S87–93. 10.1038/sj.ki.5002608. [DOI] [PubMed] [Google Scholar]
- 20. Ponce D, Berbel MN, Abrão JMG, Goes CR, Balbi AL. A randomized clinical trial of high volume peritoneal dialysis versus extended daily hemodialysis for acute kidney injury patients. Int Urol Nephrol. 2013 Jun;45(3):869–78. 10.1007/s11255-012-0301-2. [DOI] [PubMed] [Google Scholar]
- 21. Chávez-Iñiguez JS, Navarro-Gallardo GJ, Medina-González R, Alcantar-Vallin L, García-García G. Acute kidney injury caused by obstructive nephropathy. Int J Nephrol. 2020 Nov 29:2020:8846622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Forni LG, Darmon M, Ostermann M, Oudemans-van Straaten HM, Pettilä V, Prowle JR, et al. Renal recovery after acute kidney injury. Intensive Care Med. 2017 Jun;43(6):855–66. 10.1007/s00134-017-4809-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Depner T, Beck G, Daugirdas J, Kusek J, Eknoyan G. Lessons from the Hemodialysis (HEMO) Study: an improved measure of the actual hemodialysis dose. Am J Kidney Dis. 1999 Jan;33(1):142–9. 10.1016/s0272-6386(99)70272-6. [DOI] [PubMed] [Google Scholar]
- 24. Paniagua R, Amato D, Vonesh E, Correa-Rotter R, Ramos A, Moran J, et al. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, controlled trial. J Am Soc Nephrol. 2002 May;13(5):1307–20. 10.1681/ASN.V1351307. [DOI] [PubMed] [Google Scholar]
- 25. FHN Trial Group, Chertow GM, Levin NW, Beck GJ, Depner TA, Eggers PW, Gassman JJ, et al. In-center hemodialysis six times per week versus three times per week. N Engl J Med. 2010 Dec 9;363(24):2287–300. Erratum in: N Engl J Med. 2011 Jan 6;364(1):93. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are not publicly available due to ethical reasons. Further inquiries can be directed to the corresponding author.