Acute kidney injury (AKI) occurs commonly in critically ill patients and is associated with poor prognosis. Recent large, multinational multicenter epidemiologic studies indicate AKI is associated with increased mechanical ventilation (MV) duration, prolonged hospital stay, and higher rates of death in neonates, children, and adults (1–3). Simultaneously, parallel data indicates significant positive net fluid balance or fluid accumulation, termed fluid overload (FO), is also associated with worse patient outcome, particularly in children (4). Higher degrees of FO are associated with adverse outcomes and complications including increased rates of infection, delayed wound healing, prolonged MV, utilization of renal replacement therapy, and mortality (5–17).
The timing of AKI and FO is important. Consensus statements now recognize day 3 of hospitalization as the time point for which creatinine elevation is damage associated AKI (18). Creatinine elevation prior to this time may be functional damage and not be indicative of meaningful AKI (19). Evidence indicates day 3 of ICU stay (day3)-AKI is associated with an increased duration of MV, longer ICU length of stay (LOS), and higher mortality rates (20). Early fluid accumulation may be less deleterious on patient outcome than late fluid accumulation (21,22). Although volume restoration is central to initial resuscitation and stabilization efforts in critically ill patients (23), distinct phases of fluid balance likely exist in ICU patients (24).
Although AKI and FO each seem to be associated with adverse outcomes, it remains unclear if they carry independent effects. The interplay between the two phenomena is complicated as they are physiologically linked; AKI is a risk factor for fluid accumulation, and fluid accumulation is a risk factor for AKI (25). However, despite the bidirectional risk, not all AKI patients suffer excessive fluid accumulation and not all patients with significant fluid accumulation suffer AKI. Recent evidence from both adult and pediatric populations suggest AKI diagnosed by creatinine elevation and oliguria is associated with worse outcome than either criteria independently (26,27). Unfortunately, even though AKI and FO are frequently assessed, very few studies have attempted to delineate the independent and synergistic impact of both factors. A paucity of data specifically discusses the deleterious associations of FO in the absence of AKI. The relationship of AKI and fluid balance is further complicated by the dilutional effect of fluid on serum creatinine (SCr) concentration. Several recent reports detail how correction of SCr concentration for net fluid balance refines AKI diagnosis, ultimately changing the epidemiology of both AKI prevalence and outcome associations (28–31).
An understanding of the individual and potentially synergistic effects of fluid accumulation in the context of AKI is needed. We studied a population of critically ill pediatric patients to test the hypothesis that FO is a distinct phenotype of critical illness, associated with poor downstream patient outcome both in isolation and in combination with AKI.
MATERIALS AND METHODS
We conducted a secondary analysis of data from the Acute Kidney Injury in Children Expected by Renal Angina and Urinary Biomarkers (AKI-CHERUB, NCT01735162) study, a prospective observational study conducted in a single-center, quaternary care PICU (32). AKI-CHERUB served as the pilot study for the AKI-Assessment of Worldwide AKI in Pediatrics, Renal Angina and Epidemiology (AWARE, NCT01987921) (3). The study was approved by the Institutional Review Board with waiver of the need for informed consent. Study procedures were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national), and with the Helsinki Declaration of 1975, as revised in 2000.
Primary inclusion criteria for AKI-CHERUB study were all children admitted to the PICU 3 months to 25 years old predicted to have a ICU LOS greater than 48 hours with an indwelling urinary catheter. Primary exclusion criteria included a history of end-stage renal disease and admission immediately following renal transplant. Only patients with full data on day 3 (inclusive of SCr, urine output [UOP], and net fluid balance) were included in this analysis.
Demographic and initial available laboratory variables, including Pediatric Risk of Mortality (PRISM)–III score were collected on the day of PICU admission. Sepsis presence or absence was determined on the day of admission to ICU based on consensus pediatric sepsis criteria. Data were collected for 7 consecutive days (days 1–7 denoted as day1, day2, and so on). Daily collected variables were assessed at 8:00 am for each patient on days1–7 and included vital signs, serum and urine laboratory values, MV support level, UOP, and total ICU fluid balance. Day3 was used to assess the principal outcome variables of AKI and FO.
Daily fluid input and output (L) were recorded for each day of hospital admission. Daily FO was calculated as a percentage (%FO) relative to admission weight as previously described (9):
%FO on day3 was determined for each patient and for this analysis, %FO greater than or equal to 20 was used as the cutoff for “FO.”
Acute Kidney Injury
AKI was classified on day3. Kidney Disease Improving Global Outcomes SCr and UOP based AKI criteria were used to stage AKI (33). The worse of either criterion were used to denote AKI stage, and all stages of AKI (1–3) were denoted as “AKI” for this analysis. The baseline SCr was defined as the lowest value in the 3 months before ICU admission. If no baseline SCr was available, a baseline SCr was imputed to an estimated creatinine clearance of 120 mL/min/1.73 m2, as previously validated (3,34).
Fluid Overload/AKI Phenotypes
Four unique patient phenotypes were created based on the achievement of FO and AKI on day 3. These phenotypes were classified as follows: FO–/AKI–, FO+/AKI–, FO–/AKI+, and FO+/AKI.
Fluid Corrected Creatinine
Fluid balance was used to correct creatinine as previously described by the following formula (28,30):
*Total body water by age classification:
The correction was performed on day3 using the day3-SCr and day3 net fluid balance data. AKI classified using corrected SCr was denoted “AKIcorr.”
Short-term and long-term outcomes were studied. Both AKI and FO were used as outcome variables for initial comparison (and creation of phenotypes) and then also used as independent variables for the modeling of longer-term outcomes. Additionally, they were included in bivariate analyses when analyzing patient outcome on day28: duration of MV, ICU, and hospital LOS (both assessed on review of chart at 1 yr following enrollment [data censored at this time if patient still hospitalized], time of discharge from hospital, or time of death), use of continuous renal replacement therapy, and 28-day mortality.
Descriptive statistics for continuous variables were summarized as mean and sd or median and interquartile range when determined skewed by a Shapiro-Wilk test. Categorical variables were summarized as counts and percentages. Summary statistics are presented in tabular form. Differences in ordinal outcome such as LOS and PRISM III score were compared with unpaired t tests or Wilcoxon rank-sum tests, and dichotomous outcomes by chi-square or Fisher exact tests. Logistic regression models assessed bivariate associations between dichotomous outcomes and FO as a continuous variable, Spearman correlation assessed bivariate associations between ordinal variables. Multiple general linear regression models were used to derive estimated associations between AKI, FO, and the outcomes of duration of MV and LOS (hospital and ICU). The regression models incorporated the model terms of age, FO, AKI, and PRISM III scores—to a priori limit the effects of collinearity but also to use practical, highest relevance of importance. Pairwise differences between the four FO/AKI phenotypes were analyzed yielding model estimated means. Statistical significance level was set at alpha less than 0.05 level. The data analysis was performed using SAS software 9.4 (SAS Institute, Cary, NC) and JMP (SAS Institute).
From 184 enrolled patients in CHERUB, 149 had complete day3 and day28 data and were included in this analysis (Table 1). A slight majority of patients were male, most were school age, and came from an evenly balanced proportion of admission sources. Mortality remained lower than mortality risk (MR) predicted by severity of illness (6% vs PRISM III 8 [4–14], MR 10–15%).
Acute Kidney Injury and Fluid Overload
Twenty-nine patients (19.5%) had AKI on day 3 and 36 (24.2%) had FO. Ten patients (6.7%) had both AKI and FO (Table 2). The median age of patients with AKI was 12.9 (5.0–17.3), but 2.9 (1.4–12.5) for patients with FO. Sepsis was more prevalent and the median PRISM III score was higher in patients with AKI than those without AKI (p = 0.008 and 0.0004, respectively). Conversely, the PRISM III score was not significantly different in patients with or without FO. AKI was significantly associated with longer LOS (ICU and hospital), whereas FO was significantly associated with ICU LOS. All associations lost significance on multivariate regression (Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/PCC/B60). Both AKI and FO were associated with mortality on bivariate analysis (p = 0.002 and 0.05, respectively) (Supplemental Table 2, Supplemental Digital Content 1, http://links.lww.com/PCC/B60), but lost significance on adjustment for severity of illness (data not shown).
Separation of AKI and Fluid Overload
Separation into four FO/AKI phenotypes demonstrated nearly equal distribution of both conditions in isolation with 6.7% patients suffering both (Table 3). Paired analyses to control for AKI was used to determine bivariate associations of FO with outcome. In patients without AKI, severity of illness scores (PRISM III) were not significantly different in patients with FO compared with those without FO. In patients without AKI, mortality was higher in FO patients (p = 0.001). Although the median duration of MV and ICU LOS were longer in patients with FO, these differences were not significant (Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/PCC/B60). In patients with AKI, there were no significant differences in severity of illness or ICU outcomes, including mortality, in patients with or without FO. In regression modeling, the median of ICU LOS was significantly longer in the FO+/AKI+ phenotype than any of the other three phenotypes (17.4; 95% CI, 11.0–23.7; p = 0.05) (Fig. 1, top). Additionally, mortality prevalence increased through the phenotype groups as did the odds of death. The odds of death were 23.3 times greater in FO+/AKI+ compared with patients FO–/AKI– (95% CI, 1.9–285.2; p = 0.014) (Fig. 1, bottom). The relative risk to mortality conferred by the exposure of FO (in AKI negative patients) was 14.5 (95% CI, 1.69–123.9; p = 0.015). The relative risk to mortality conferred by the exposure of AKI (in FO negative patients) was (9.9; 95% CI, 0.94–103.7; p = 0.056).
Correction of Creatinine Strengthens Outcome Associations With AKI
After correction of SCr on day 3 for net fluid balance, corrected AKI classifications were determined. “Class switching” between AKI stages occurred in 21 patients (Fig. 2). AKIcorr demonstrated stronger bivariate associations with both MV duration and ICU LOS (p = 0.003 and 0.0005, respectively) compared with the AKI without correction (p = 0.079 and 0.028) (Table 4). When AKIcorr was used to identify FO/AKI phenotypes, the conferred increased risk of mortality with FO (in AKI+ patients) seen in the uncorrected phenotype separation was reduced, and there was no longer a significant increase in odds of mortality (Supplemental Table 3, Supplemental Digital Content 1, http://links.lww.com/PCC/B60).
This study attempts to delineate the independent and synergistic effects of AKI and significant fluid accumulation in a heterogeneous population of critically ill children. Our findings suggest multiple associations of FO with patient outcome: 1) FO may increase ICU resource utilization (longer MV duration and ICU LOS), 2) independent of AKI status, FO may increase mortality, and 3) correction of creatinine for net fluid balance may improve the precision of both AKI diagnosis and associated sequelae, including the associations with FO.
The adjudication and precision of the diagnosis of AKI continues to evolve. The importance of accounting for UOP in AKI diagnosis has been highlighted (3). In this study, we operationalize the directives from the 16th Acute Dialysis and Quality Initiative (determination of AKI by persistent SCr change on day 3) (18). Our study defined %FO as the assessment made at day 3. This time window represents an epoch of ICU care almost assuredly outside the resuscitation phase of critical illness and likely beyond the stabilization phase. Theoretically, this time frame is a period for titration of therapy and was intentionally chosen to match the day of AKI assessment. Additionally, in addition to incorporating the correction of SCr for fluid balance to adjust AKI diagnosis, this study separates patients into phenotypes, stratifying patients by AKI and FO status. Importantly, we incorporated an adjustment in the correction factor by age for total body water composition (the correction factor increases by up to 33% in young children and infants).
Although many recent publications indicate excessive fluid accumulation is associated with poor patient outcome in multiple patient populations (4,6,35–37), very few attempt to extricate the independent effects of fluid balance from AKI. Physiologically, the two diagnoses of AKI and FO are linked as AKI patients with oliguria are at high risk of fluid accumulation. Data from adults and children suggest oliguria alone is an independent risk factor for poor outcome and if oliguria is assumed to be a significant factor (in addition to iatrogenic fluid accumulation and fluid creep) for FO, our data mirror these reports. Separation of patients into FO and AKI “phenotypes” facilitated the identification of potential contribution of FO in the context of controlling for AKI status. Our findings are not conclusive but suggest independent risks with FO and a significant deleterious effect of FO and AKI in concert. This finding also parallels the aforementioned data about oliguric-AKI; patients with AKI by both creatinine elevation and oliguria suffered a significantly worse outcome than either classification alone (26,27). This finding may be impactful. Although a singular (or multimodal) therapy for AKI has yet to be discovered, fluid balance is addressed regularly in the ICU and offers a range of management options. Importantly, net fluid balance is modifiable. Decisions regarding fluid management are made on critically ill patients multiple times daily—pertaining to fluids delivered and fluids removed either independently, using medicine such as diuretics, or even mechanical fluid removal. Notably, on reclassification of the four phenotypes using corrected SCr concentration, the increased conferred risk of mortality by FO seen in the uncorrected phenotype was reduced—suggesting that routine correction of SCr for fluid balance (when diagnosis of AKI is made) may be a first step to identifying the true associations with FO. Together, management of fluid balance by taking into consideration the relationship of FO with AKI diagnosis and downstream outcomes may represent practical opportunities to improve patient care.
Refining the precision of creatinine concentration may be the most feasible novel biomarker for many providers. Even though biomarkers of AKI are gaining more attention in the medical literature, access at the bedside to these diagnostics is not commonplace (particularly in the United States and even more particularly in pediatrics). Analyzing the change in creatinine over time, translated into the kinetic estimate of glomerular filtration rate is a practical method (38). Correction of creatinine for fluid balance is simple and may “unmask” AKI (28–30). Liu et al (28) reported that the correction of SCr for FO uncovered an additional 15% of cases of AKI and further solidified the association of AKI with adverse outcomes. Two studies analyzing children following cardiopulmonary bypass reported parallel findings in two different populations surgery (30,31). This is particularly concerning given the recent data suggesting FO may predate AKI. Hassinger et al (11) recently reported in a cohort of 98 patients following cardiopulmonary bypass that the development of FO (≥ 5%) preceded the development of AKI prior to meeting SCr based definitions.
This report has several weaknesses. Although we prospectively studied a heterogeneous population of critically ill patients, the sample size was relatively small. The requirement of an indwelling urinary catheter and expected stay greater than 48 hours (inclusion/exclusion criteria) skewed enrollment to a subset of PICU patients. The urinary catheter on enrollment criteria also missed patients who did not actually have a catheter on day 0 of ICU course and subsequently had one placed. Many of the associations uncovered by bivariate analysis failed to hold significance on multivariate regression, potentially due to the smaller sample size. There were patterns of AKI and FO classifications based on age and size in addition to the severity of illness that we did not incorporate into our regression models. Additionally, we did not perform dichotomization of continuous variables for the purposes of linear or logistic regression as we felt the additive analyses would miss the exploratory findings of this preliminary report. Finally, our data are limited to a single-center and subject to the practice consistencies inherently associated with providers of a single institution. The findings of this report were derived from a pilot study (AKI-CHERUB) which was halted as the larger, multicenter AWARE study was initiated (3).
Both FO and AKI are associated with poor outcomes. Our findings suggest excessively positive net fluid balance, now termed FO, may confer increased risk for poor outcome for critically ill patients, irrespective of severity of illness or AKI status. Correcting creatinine for net fluid balance provides a simple way of improving AKI diagnostic precision and also of identifying additional patients at risk for poor outcomes. Mitigation of fluid accumulation before day 3 of ICU course may, if possible, be associated with improved patient outcomes. These findings warrant analysis in larger populations.
1. Jetton JG, Boohaker LJ, Sethi SK, et al.; Neonatal Kidney Collaborative (NKC): Incidence and outcomes of neonatal acute kidney injury
(AWAKEN): A multicentre, multinational, observational cohort study. Lancet Child Adolesc Health 2017; 1:184–194
2. Hoste EA, Bagshaw SM, Bellomo R, et al. Epidemiology of acute kidney injury
in critically ill patients: The multinational AKI-EPI study. Intensive Care Med 2015; 41:1411–1423
3. Kaddourah A, Basu RK, Bagshaw SM, et al.; AWARE Investigators: Epidemiology of acute kidney injury
in critically ill children and young adults. N Engl J Med 2017; 376:11–20
4. Alobaidi R, Morgan C, Basu RK, et al. Association between fluid balance and outcomes in critically ill children: A systematic review and meta-analysis. JAMA Pediatr 2018; 172:257–268
5. Alkandari O, Eddington KA, Hyder A, et al. Acute kidney injury
is an independent risk factor for pediatric intensive care
unit mortality, longer length of stay and prolonged mechanical ventilation in critically ill children: A two-center retrospective cohort study. Crit Care 2011; 15:R146
6. Arikan AA, Zappitelli M, Goldstein SL, et al. Fluid overload
is associated with impaired oxygenation and morbidity in critically ill children. Pediatr Crit Care Med 2012; 13:253–258
7. Flori HR, Church G, Liu KD, et al. Positive fluid balance is associated with higher mortality and prolonged mechanical ventilation in pediatric patients with acute lung injury. Crit Care Res Pract 2011; 2011:854142
8. Foland JA, Fortenberry JD, Warshaw BL, et al. Fluid overload
before continuous hemofiltration and survival in critically ill children: A retrospective analysis. Crit Care Med 2004; 32:1771–1776
9. Goldstein SL, Currier H, Graf Cd, et al. Outcome in children receiving continuous venovenous hemofiltration. Pediatrics 2001; 107:1309–1312
10. Goldstein SL, Somers MJ, Baum MA, et al. Pediatric patients with multi-organ dysfunction syndrome receiving continuous renal replacement therapy. Kidney Int 2005; 67:653–658
11. Hassinger AB, Wald EL, Goodman DM. Early postoperative fluid overload
precedes acute kidney injury
and is associated with higher morbidity in pediatric cardiac surgery patients. Pediatr Crit Care Med 2014; 15:131–138
12. Hazle MA, Gajarski RJ, Yu S, et al. Fluid overload
in infants following congenital heart surgery. Pediatr Crit Care Med 2013; 14:44–49
13. Li Y, Wang J, Bai Z, et al. Early fluid overload
is associated with acute kidney injury
and PICU mortality in critically ill children. Eur J Pediatr 2016; 175:39–48
14. Selewski DT, Cornell TT, Blatt NB, et al. Fluid overload
and fluid removal in pediatric patients on extracorporeal membrane oxygenation requiring continuous renal replacement therapy. Crit Care Med 2012; 40:2694–2699
15. Sinitsky L, Walls D, Nadel S, et al. Fluid overload
at 48 hours is associated with respiratory morbidity but not mortality in a general PICU: Retrospective cohort study. Pediatr Crit Care Med 2015; 16:205–209
16. Sutherland SM, Zappitelli M, Alexander SR, et al. Fluid overload
and mortality in children receiving continuous renal replacement therapy: The prospective pediatric continuous renal replacement therapy registry. Am J Kidney Dis 2010; 55:316–325
17. Selewski DT, Askenazi DJ, Bridges BC, et al. The impact of fluid overload
on outcomes in children treated with extracorporeal membrane oxygenation: A multicenter retrospective cohort study. Pediatr Crit Care Med 2017; 18:1126–1135
18. Chawla LS, Bellomo R, Bihorac A, et al.; Acute Disease Quality Initiative Workgroup 16: Acute kidney disease and renal recovery: Consensus report of the Acute Disease Quality Initiative (ADQI) 16 Workgroup. Nat Rev Nephrol 2017; 13:241–257
19. McCullough PA, Shaw AD, Haase M, et al. Diagnosis of acute kidney injury
using functional and injury biomarkers: Workgroup statements from the tenth Acute Dialysis Quality Initiative Consensus Conference. Contrib Nephrol 2013; 182:13–29
20. Blinder JJ, Asaro LA, Wypij D, et al. Acute kidney injury
after pediatric cardiac surgery: A secondary analysis of the safe pediatric euglycemia after cardiac surgery trial. Pediatr Crit Care Med 2017; 18:638–646
21. Murphy CV, Schramm GE, Doherty JA, et al. The importance of fluid management in acute lung injury secondary to septic shock. Chest 2009; 136:102–109
22. Silversides JA, Fitzgerald E, Manickavasagam US, et al.; Role of Active Deresuscitation After Resuscitation (RADAR) Investigators: Deresuscitation of patients with iatrogenic fluid overload
is associated with reduced mortality in critical illness. Crit Care Med 2018; 46:1600–1607
23. Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis campaign: International guidelines for management of sepsis and septic shock: 2016. Intensive Care Med 2017; 43:304–377
24. Hoste EA, Maitland K, Brudney CS, et al.; ADQI XII Investigators Group: Four phases of intravenous fluid therapy: A conceptual model. Br J Anaesth 2014; 113:740–747
25. Ostermann M, Straaten HM, Forni LG. Fluid overload
and acute kidney injury
: Cause or consequence? Crit Care 2015; 19:443
26. Kaddourah A, Basu RK, Goldstein SL, et al.; Assessment of Worldwide Acute Kidney Injury
, Renal Angina and, Epidemiology (AWARE) Investigators: Oliguria and acute kidney injury
in critically ill children: Implications for diagnosis and outcomes. Pediatr Crit Care Med 2019; 20:332–339
27. Kellum JA, Sileanu FE, Murugan R, et al. Classifying AKI by urine output versus serum creatinine level. J Am Soc Nephrol 2015; 26:2231–2238
28. Liu KD, Thompson BT, Ancukiewicz M, et al.; National Institutes of Health National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network: Acute kidney injury
in patients with acute lung injury: Impact of fluid accumulation on classification of acute kidney injury
and associated outcomes. Crit Care Med 2011; 39:2665–2671
29. Macedo E, Bouchard J, Soroko SH, et al.; Program to Improve Care in Acute Renal Disease Study: Fluid accumulation, recognition and staging of acute kidney injury
in critically-ill patients. Crit Care 2010; 14:R82
30. Basu RK, Andrews A, Krawczeski C, et al. Acute kidney injury
based on corrected serum creatinine
is associated with increased morbidity in children following the arterial switch operation. Pediatr Crit Care Med 2013; 14:e218–e224
31. SooHoo MM, Patel SS, Jaggers J, et al. Acute kidney injury
defined by fluid corrected creatinine in neonates after the Norwood procedure. World J Pediatr Congenit Heart Surg 2018; 9:513–521
32. Menon S, Goldstein SL, Mottes T, et al. Urinary biomarker incorporation into the renal angina index early in intensive care unit admission optimizes acute kidney injury
prediction in critically ill children: A prospective cohort study. Nephrol Dial Transplant 2016; 31:586–594
33. Kellum JA, Lameire N; KDIGO AKI Guideline Work Group: Diagnosis, evaluation, and management of acute kidney injury
: A KDIGO summary (part 1). Crit Care 2013; 17:204
34. Zappitelli M, Parikh CR, Akcan-Arikan A, et al. Ascertainment and epidemiology of acute kidney injury
varies with definition interpretation. Clin J Am Soc Nephrol 2008; 3:948–954
35. Valentine SL, Sapru A, Higgerson RA, et al.; Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network; Acute Respiratory Distress Syndrome Clinical Research Network (ARDSNet): Fluid balance in critically ill children with acute lung injury. Crit Care Med 2012; 40:2883–2889
36. Mah KE, Hao S, Sutherland SM, et al. Fluid overload
independent of acute kidney injury
predicts poor outcomes in neonates following congenital heart surgery. Pediatr Nephrol 2018; 33:511–520
37. Flores-González JC, Valladares CM, Yun Castilla C, et al.; Bronquiolitis en la Unidad de Cuidados Intensivos Pediátricos (BRUCIP) WorkGroup: Association of fluid overload
with clinical outcomes in critically ill children with bronchiolitis: Bronquiolitis en la Unidad de Cuidados Intensivos Pediátricos (BRUCIP) Study. Pediatr Crit Care Med 2019; 20:e130–e136
38. Chen S. Retooling the creatinine clearance equation to estimate kinetic GFR when the plasma creatinine is changing acutely. J Am Soc Nephrol 2013; 24:877–888