15 pages


of 15
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  Association between fluid balance and survival in critically ill patients Joon Lee 1,2 , Emma de Louw 3 , Matthew Niemi 3 , Rachel Nelson 3 , Roger G. Mark 1 , Leo Anthony Celi 1,3 , Kenneth J. Mukamal 3 , and John Danziger 3 1 Lab of Computational Physiology, Division of Health, Sciences and Technology, Harvard-Massachusetts Institute of Technology, Cambridge, Ma 2 School of Public Health and Health Systems, University of Waterloo 3 Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Ma Abstract Objective— Although the consequences of chronic fluid retention are well known, those of iatrogenic fluid retention that occurs during critical illness have not been fully determined. Therefore we investigated the association between fluid balance and survival in a cohort of almost 16,000 individuals who survived an intensive care unit (ICU) stay in a large, urban, tertiary medical centre. Design— Longitudinal analysis of fluid balance at ICU discharge and 90-day post-ICU survival. Measurements— Associations between fluid balance during the ICU stay, determined from the electronic bedside record, and survival were tested using Cox proportional hazard models adjusted for severity of critical illness. Results— There were 1827 deaths in the first 90 days after ICU discharge. Compared to the lowest quartile of discharge fluid balance [median (interquartile range) −1.5 (−3.1, −0.7) L], the highest quartile [7.6 (5.7, 10.8) L] was associated with a 35% [95% confidence interval (CI) 1.13–1.61)] higher adjusted risk of death. Fluid balance was not associated with outcome amongst individuals without congestive heart failure or renal dysfunction. Amongst patients with either comorbidity, however, fluid balance was strongly associated with outcome, with the highest quartile having a 55% (95% CI 1.24–1.95) higher adjusted risk of death than the lowest quartile. Isotonic fluid balance, defined as the difference between intravenous isotonic fluid administration and urine output, was similarly associated with 90-day outcomes. Conclusion— Positive fluid balance at the time of ICU discharge is associated with increased risk of death, after adjusting for markers of illness severity and chronic medical conditions, particularly in patients with underlying heart or kidney disease. Restoration of euvolaemia prior to discharge may improve survival after acute illness. Correspondence: John Danziger 185 Pilgrim Road, Farr 8 Boston, MA 02215 T: (617) 632-9880 F: (617) 632-9890 Conflict of interest statement  None of the authors has any conflicts of interest to declare. HHS Public Access Author manuscript  J Intern Med  . Author manuscript; available in PMC 2016 April 01. Published in final edited form as:  J Intern Med  . 2015 April ; 277(4): 468–477. doi:10.1111/joim.12274. A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t    Keywords diuresis; fluid balance; heart failure; oedema; survival; weight Introduction Chronic fluid retention frequently occurs in individuals with underlying cardiac and renal disease, and is associated with hypertension, pulmonary vascular congestion, and death [ 1-5 ]. Due to abnormal sensation of body fluid volume, chronic activation of the renin–angiotensin–aldosterone axis maintains a fluid overloaded state, and requires treatment with dietary sodium restriction and diuretics to re-establish fluid balance [6]. The presence of pulmonary oedema is a widely accepted indication for diuretic therapy, although management of peripheral oedema is controversial [ 7-10 ].The effect of fluid retention in hospitalized patients is not well known. Since the landmark studies of Rivers et al . outlining the benefit of early goal-directed therapy [11], aggressive intravenous fluid (IVF) administration has become the standard of care in the intensive care unit (ICU). Although spontaneous diuresis occurs following illness in many individuals, such diuresis may not happen in those with an underlying tendency towards fluid retention [12]. Fluid retention has been associated with worse outcomes in some [ 13-18 ] but not all [19] critically ill populations, yet has not been evaluated in a broader context.The current study population was a large cohort of medical, surgical, cardiac and cardiothoracic intensive care patients, admitted to a single medical centre over a 7-year period, for whom fluid balance was carefully recorded. We aimed to further characterize the association between fluid balance during the ICU stay and 90-day mortality after discharge from the ICU, while accounting for demographic characteristics, reason for admission, severity of illness, length of stay, peak fluid balance and medical comorbidities. We also specifically investigated whether fluid balance has a greater effect on survival amongst patients with congestive heart failure (CHF) or acute kidney injury (AKI), or impaired renal function at the time of discharge, as we hypothesized that such patients would be at greatest risk of increased mortality. Methods Study population We used the Multiparameter Intelligent Monitoring in Intensive Care (MIMIC-II) research database, a joint venture managed by researchers from the Laboratory for Computational Physiology at Massachusetts Institute of Technology (MIT), Cambridge, MA, USA, and the Department of Medicine at the Beth Israel Deaconess Medical Center (BIDMC) Boston, MA, USA. 20  BIDMC is a 700-bed, urban, academic medical centre with 77 adult ICU (28 medical, 25 surgical, 16 cardiothoracic and eight cardiac) beds. The database contains high temporal resolution data from clinical systems, including laboratory results, electronic documentation and bedside monitor trends and waveforms, for all patients admitted to a BIDMC ICU between 2001 and 2008. Use of the MIMIC II database has been approved by Lee et al.Page 2  J Intern Med  . Author manuscript; available in PMC 2016 April 01. A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t    the institutional review boards of BIDMC and MIT. We restricted our analysis to those patients with an ICU stay of greater than 24 h, because shorter stays are likely to reflect patients who either died or were transferred to a general ward very shortly after ICU admission.Amongst the 21,694 ICU survivors, there were 17,644 with an ICU stay of 24 h or more. We excluded 342 individuals for whom documentation of fluid balance was not available, 759 who received dialysis (as dialytic fluid removal is not included in the electronic medical record), 1129 for whom renal function was not recorded and 19 with missing basic demographic information; the final sample comprised 15,395 unique first ICU hospitalizations. Outcome The primary outcome was death within 90 days of ICU discharge. Deaths were identified from the Social Security Death Index. Exposures The primary exposure was fluid balance in the ICU as determined on ICU discharge. This was automatically computed by the bedside electronic record, and reflects the net cumulative balance of daily inputs and outputs. These variables include, but are not limited to, IVF, oral fluids, medications and blood products (inputs), as well as urine, stools and other body fluids (outputs). Total fluid balance was winsorized at the 0.5 and 99.5 percentiles to limit the effect of outliers. Isotonic fluid balance, as determined by the difference between total isotonic fluid administration (predominantly saline and lactated Ringer’s solution) and total urine output, was evaluated as a secondary exposure. Covariates Demographic information included age, sex and race (categorized as White, African-American, Asian, Hispanic, other or unknown). Medical comorbidities were determined using Elixhauser codes at discharge [ 21 ]. The ICU types were cardiac, surgical, cardiothoracic and medical units. Predictors of illness severity included the Simplified Acute Physiology Score (SAPS) at admission, as well as the peak SAPS and cumulative SAPS during the ICU stay, use of vasopressors, use of mechanical ventilation, length of ICU stay and the total amount of isotonic intravenous fluids (IVFs) administered. The most positive fluid balance at any point during the ICU stay was considered the peak fluid balance. AKI during the ICU stay was defined as a 100% increase in peak serum creatinine compared to the levels at either ICU admission or discharge. The estimated glomerular filtration rate (eGFR) at the time of ICU discharge was calculated from the Modification of Diet in Renal Disease equation using the serum creatinine level at discharge [ 22 ]. Statistical analysis Baseline characteristics were stratified by quartiles of fluid balance. Cox regression analyses were used to compare survival probabilities within 90 days of discharge from the ICU. Follow-up started on the first day after ICU discharge. Patients who survived for more than 90 days were censored. Binary indicator variables were created for all 30 Elixhauser Lee et al.Page 3  J Intern Med  . Author manuscript; available in PMC 2016 April 01. A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t    comorbidities, rather than a summary score, and for use of mechanical ventilation, vasopressor use, the presence of AKI, eGFR <60 mL/min at discharge and quartiles of fluid balance. Race and ICU type were included as multicategory variables. Age, SAPS, laboratory values, administered IVFs, peak fluid balance and length of stay were all included as continuous variables.Fluid balance at discharge was examined in quartiles and as a continuous variable; the lowest quartile was used as the reference for all analyses. We tested for nonlinearity by the inclusion of a quadratic term in the adjusted model ( P  > 0.5). The proportional hazard assumption was tested by visual inspection of a plot of the integrated hazard versus time.In subgroup analyses, we examined a priori  whether underlying heart failure or renal disease modified the effect of fluid balance on mortality. Sensitivity analyses were performed in three groups: (i) patients with CHF as determined by International Classification of Diseases, Ninth Revision codes at discharge; (ii) patients with AKI, as determined by a100% increase in peak serum creatinine compared to the levels at either ICU admission or discharge ; and (iii) those with an eGFR of <60 mL/min at discharge. We created an indicator variable for the presence of any one of these three clinical risk factors (CHF, AKI or eGFR <60 mL/min) to test the interaction between fluid balance in patients with none versus one or more of these factors. Analyses were also stratified by severity of renal disease, according to the fold change in creatinine level during the ICU stay (1.5 to <2, ≥2 to <3 and ≥3).In a sensitivity analysis, we determined whether restoration of fluid balance prior to ICU discharge in those with the highest peak fluid balance was associated with improved survival. Those in the highest quartile of peak fluid balance were stratified by the threshold for the highest discharge fluid balance quartile (4.5 L), and the adjusted effect of fluid balance on 90-day survival was determined.As fluid balance includes both oral intake and hypotonic fluids, we then assessed whether isotonic fluid balance was associated with survival. Weight at both admission and discharge was recorded in only 44% of patients, precluding the use of body weight changes as a meaningful additional exposure. Next, to assess the presence of survival bias, logistic regression analysis, including 1582 individuals who died during the ICU stay and were not included in the srcinal analysis, was performed to determine whether fluid balance was associated with in-ICU mortality. Finally, because fluid balance is unlikely to affect mortality beyond 90 days from discharge, we explored the association between fluid balance and mortality in those patients who survived the first 90 days after ICU discharge as an internal model control. All analyses were performed using JMP Pro (SAS Institute, Cary, NC, USA). Results Baseline characteristics In the 15,395 critically ill patients included in this study, mean (±SD) total fluid balance was 2.6 (±5.2) L, and differed significantly according to ICU type [cardiac 0.1 (±3.4), Lee et al.Page 4  J Intern Med  . Author manuscript; available in PMC 2016 April 01. A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  A  u t  h  or M an u s  c r i   p t  
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks