Association of Volume Overload With Kidney Function Outcomes Among Patients With Acute Decompensated Heart Failure.

Association of Volume Overload With Kidney Function Outcomes Among Patients With Acute Decompensated Heart Failure - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Am J Kidney Dis . Author manuscript; available in PMC: 2026 Jun 6. Published in final edited form as: Am J Kidney Dis. 2025 Jun 6;86(3):324–334.e1. doi: 10.1053/j.ajkd.2025.03.026 Association of Volume Overload With Kidney Function Outcomes Among Patients With Acute Decompensated Heart Failure Amanda I Moises 1 , Hocine Tighiouart 1, 2, 3 , Jeffrey M Testani 4 , Marcelle Tuttle 1, 2, 3 , Run Banlengchit 5 , Tatsufumi Oka 1 , Ki Jung Lee 1 , Katie Ferguson 1 , Hannah Sarnak 1 , Callum Harding 2, 3 , Michael S Kiernan 1 , Mark J Sarnak 1 , Wendy McCallum 1 Author information Article notes Copyright and License information 1. Tufts Medical Center, Boston, MA 2. Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, Boston, MA 3. Tufts Clinical and Translational Science Institute, Tufts University, Boston, MA 4. Yale School of Medicine, New Haven, CT 5. University of California San Francisco, San Francisco, CA Authors’ Contributions: Research idea and study design: AM, MT, RB, MJS, TO, WM; data analysis/interpretation: AM, HT, JMT, MT, RB, TO, KJL, KF, HS, CH; statistical analysis: HT, WM; supervision or mentorship: MSK, JMT, MJS, WM. Each author contributed important intellectual content during manuscript drafting or revision and agrees to be personally accountable for the individual’s own contributions and to ensure that questions pertaining to the accuracy or integrity of any portion of the work, even one in which the author was not directly involved, are appropriately investigated and resolved, including with documentation in the literature if appropriate. ✉ Corresponding Author: Wendy McCallum, MD, MS, Tufts Medical Center, 800 Washington Street, Box 391, Boston, MA 02111, wmccallum@tuftsmedicalcenter.org Issue date 2025 Sep. PMC Copyright notice PMCID: PMC12354306  NIHMSID: NIHMS2088592  PMID: 40484339 The publisher's version of this article is available at Am J Kidney Dis Abstract Rationale & Objective: Reduced kidney function, as evaluated by estimated glomerular filtration rate (eGFR), is a powerful risk factor for adverse outcomes among patients with decompensated heart failure (ADHF). However, evidence that volume overload is a risk factor for declines in eGFR has been inconsistent. This study examined this association among adults with ADHF. Study Design: Retrospective observational study. Setting & Participants: Adult patients admitted to a quaternary referral hospital for a primary diagnosis of ADHF requiring right heart catheterization between 2015 and 2021. Exposures: Initial central venous pressure (CVP), pulmonary capillary wedge pressure (PCWP), and change in CVP and PCWP. Outcomes: eGFR when CVP and/or PCWP monitoring began, in-hospital eGFR slope, and initiation of dialysis through March 2022. Analytical Approach: Restricted cubic splines and linear mixed models to examine the association of CVP and PCWP with baseline eGFR and in-hospital eGFR slope. Cox proportional hazard regression models were used to examine the association of CVP, PCWP, and changes in CVP and PCWP with initiation of dialysis. Sensitivity analyses treated death as a competing risk. Results: Among 753 patients, higher CVP and PCWP were significantly associated with lower eGFR (β [95%CI]: −5.01 [−6.68, −3.35] and −2.84 [−4.55, −1.13] ml/min/1.73m 2 per 1 SD higher CVP and PCWP, respectively), and lower in-hospital eGFR slope (−1.13 [−1.57, −0.69] and −0.59 [−1.02, −0.15] ml/min/1.73 m 2 /week per 1 SD higher). Over a median of 33 (IQR 13, 58) months, 62 (8.2%) required dialysis and 264 (35.1%) died. Higher CVP and PCWP were associated with increased risk for dialysis: aHR (95%CI) per 1 SD higher CVP was 1.49 (1.17, 1.90) and PCWP 1.30 (1.04, 1.62). Associations remained consistent when treating death before dialysis as a competing risk. Change in CVP and PCWP were not associated with concomitantly assessed eGFR or risk of dialysis. Limitations: Observational design. Conclusions: Volume overload was associated with lower baseline kidney function, greater declines in eGFR, and increased risk of starting dialysis. Changes in the degree of volume overload were not associated with concomitantly assessed changes in eGFR or the risk of dialysis over follow-up. Keywords: decompensated heart failure, congestion, dialysis, cardiorenal syndrome Graphical Abstract Plain-Language Summary Studies evaluating volume overload as a risk factor for kidney function decline have been inconsistent. This study included adult patients admitted to the hospital with acute heart failure (2015–2021) who required invasive monitoring of heart pressures. Statistical models analyzed how volume overload on both the right and left sides of the heart were associated with kidney outcomes, independent of other patient factors. Patients with greater volume overload had lower baseline kidney function, greater decline in kidney function during the hospitalization, and an increased risk of requiring dialysis over approximately 3-years of follow-up. However, changes in the degree of volume overload were not associated with changes in kidney function measured over the same period. Introduction Reduced kidney function is a strong risk factor associated with mortality and other adverse outcomes among patients admitted for acute decompensated heart failure (ADHF). 1 Upwards of one-third of patients admitted for ADHF have acute declines in kidney function, as estimated by glomerular filtration rate (eGFR). Despite the high prevalence of low baseline eGFR and acute eGFR declines, the mechanisms remain unknown. Volume overload has been suggested to contribute to decreased kidney function; however, data have been inconsistent and primarily focused on cross-sectional associations or short-term kidney outcomes. 2 – 4 Invasively obtained measures of central venous pressure (CVP) and pulmonary capillary wedge pressure (PCWP) are often considered the gold standard measurements of right- and left-sided volume overload, respectively. Previous studies have observed a cross-sectional association between higher CVP and lower eGFR among patients admitted for ADHF. 2 , 3 , 5 However, longitudinal changes in eGFR as well as clinically meaningful kidney outcomes—such as need for dialysis—have not been examined. How a decrease in volume overload—in other words, decline in CVP—is related to kidney function outcomes is also unknown. We hypothesized that volume overload, in particular, right-sided overload is associated with both short-term and longer-term declines in kidney function, including initiation of dialysis. We also sought to investigate whether decreases in volume overload are associated with improvements in short-term kidney function and reduced need for dialysis. We aimed to examine the association between volume overload and changes in volume overload with kidney function outcomes using data collected from patients admitted for ADHF to a quaternary academic center requiring right heart catheterization. Methods Study Population The study included all adult admissions to a quaternary academic center with a primary diagnosis of ADHF requiring hemodynamic monitoring via pulmonary artery (PA) catheter between January 1, 2015 and May 27, 2021, with follow-up data collected through March 31, 2022. Partnering with the Tufts Clinical and Translational Science Institute Informatics service allowed for cohort identification derived from their standardized Research Data Warehouse. Eligible patients were identified using standardized Systematized Nomenclature of Medicine Clinical Terms (SNOWMED CT) for ADHF and placement of a PA catheter ( Table S1 ). While indications for PA catheter placement were not available for each individual, generally they were placed for hemodynamic assessment for candidacy for advanced cardiac therapies such as left ventricular assist device (LVAD) or heart transplant (HT), or for inotrope titration. If patients had more than one hospitalization during the study period, the first hospitalization was selected. Patients with a mechanical circulatory support device and/or kidney failure requiring dialysis prior to PA catheter placement were excluded, given challenges in interpreting cardiac hemodynamics and changes in eGFR. Confirmation of the eligibility criteria was performed by manual review ( Figure S1 ). The majority of data variables, including demographic data, comorbid conditions, admission medications, and laboratory values were mapped and merged using the Observational Medical Outcomes Partnership data model based on SNOWMED CT codes, in order to maintain data standardization and high level of fidelity. 6 , 7 Invasive cardiac hemodynamics, and ejection fraction (EF) obtained from echocardiogram were collected manually by members of the research team. Extraction was completed by four trained members of the research team following a standardized extraction algorithm, with each entry confirmed by a second member of the team. The study was deemed exempt from review by the Tufts Medical Center Institutional Review Board and given the retrospective nature, individual-level informed consent was not obtained. Exposure Measures of Volume Overload—Initial Invasive measures of right (CVP)- and left (PCWP)-sided volume overload were obtained according to standard bedside cardiac nursing protocols. The first available measurements were recorded. If measurements obtained at the bedside were unavailable, measurements obtained in the catheterization lab were used. Measures of Volume Overload—Change Change in CVP and change in PCWP were examined as the percent change using the initial CVP (or PCWP) and the last measurement before the PA catheter was removed. For the patients who underwent a major clinical event—initiation of dialysis, placement of an LVAD, or HT—between the first and last hemodynamic measurement, the hemodynamic measurement just before the day of the clinical event was used to define the change in CVP and PCWP as these three clinical events may have impacted their values. Percent change in CVP and PCWP were grouped into quartiles, with quartile 1 being the greatest decline in CVP and quartile 4 being the smallest decline. Outcomes Kidney Outcomes When focusing on the initial measurement of volume overload as the primary exposure of interest, there were several kidney outcomes of interest, including 1) eGFR at the time of PA catheter placement; 2) in-hospital eGFR slope, starting from PA catheter placement to the day of discharge; and 3) initiating dialysis through March 31, 2022. Creatinine measured either on the day of the PA catheter placement (or the day before if not available) was used to estimate eGFR using the CKD-EPI 2021 formula. 8 The eGFR in-hospital slope was calculated by use of linear mixed models incorporating each daily eGFR for each subject from the day of PA catheter placement to the day of discharge; the first eGFR of the day was selected if more than one value was available. The slope was represented as change in eGFR per week (ml/min/1.73 m 2 /week). For the dialysis outcome, we defined initiating dialysis as starting any form of dialysis at any time after PA catheter placement, including during the index hospitalization as well as after discharge. Post-discharge dialysis use was ascertained through linkage with the US Renal Data System. 9 Death, Heart Transplant and Left Ventricular Assist Device outcomes We also evaluated death and a composite endpoint of death, LVAD placement, or HT. Death dates are maintained in the Tufts data warehouse, which were ascertained through linkage with the Social Security death registry. All LVAD and HT dates were confirmed with the advanced heart failure program’s internal registry. Given the complexity of factors influencing whether patients undergo LVAD or HT, these were modeled as a composite outcome of death or LVAD placement or HT. Covariates Several covariates were selected as potential confounding variables, including age, sex, EF, diabetes, home medication use (angiotensin-converting enzyme inhibitors [ACEI] or angiotensin II receptor blockers [ARB], mineralocorticoid receptor antagonists [MRA]), and need for inotrope therapy at the time of the invasive hemodynamic measurements. The eGFR at the time of PA catheter placement was included as an adjustment variable in the time to event outcomes. For baseline models of volume overload, we further adjusted for PCWP for models examining CVP as the exposure variable and vice versa to evaluate whether there is a differential association between right-sided vs left-sided volume overload with outcomes. For change in models of volume overload, we adjusted for the number of days between the first and last measurement, and the baseline measurement of volume overload; in other words, models examining the change in CVP included adjustment for initial CVP, and models examining the change in PCWP included adjustment for initial PCWP. Statistical Analysis Baseline characteristics were compared by quartiles of initial CVP and initial PCWP using analysis of variance and Kruskal-Wallis tests, as well as by χ 2 and Fisher’s exact tests for categorical and continuous variables, as appropriate. We modeled CVP and PCWP with each outcome of interest using restricted cubic splines using 5 knots placed at the default locations of the rcs function in the rms package in R, which are the following percentiles: 5%, 27.5%, 50%, 72.5% and 95%. Linear regression models were used to evaluate the association between CVP and PCWP and eGFR at PA catheter placement and in-hospital eGFR slope. Multivariable Cox proportional hazards (PH) regression models were used to evaluate the association between initial CVP and PCWP with time to initiation of dialysis, death and the composite outcome of death, LVAD or HT. Analyses were performed with CVP and PCWP treated as continuous variables and divided into quartiles. Time at risk for each outcome began at the time of PA catheter placement for models examining initial CVP and PCWP. Associations were examined using unadjusted models, then included adjustments for baseline covariates as listed above (Model 1). Then, we further adjusted for PCWP in models which examined CVP as the exposure variable and vice versa (Model 2). In terms of change in CVP or PCWP, we examined whether change in CVP or change in PCWP was associated with concomitant changes in kidney function and with initiation of dialysis, death and the composite outcome of death, LVAD or HT. We used linear regression models to evaluate the association between change in CVP and PCWP with eGFR slope, which for this analysis was defined as the eGFR slope in between the initial and last CVP (or PCWP). Multivariable Cox PH regression models were used to evaluate the association between change in CVP or PCWP with time to initiation of dialysis, as well as time to death and the composite outcome of death, LVAD or HT. Time at risk for each outcome began at the time of the last CVP (or last PCWP). Associations were first examined using unadjusted models, then included adjustments for covariates as above (Model 1). Then, we further adjusted for initial CVP in models which examined change in CVP as the exposure variable and adjusted for initial PCWP in models examining change in PCWP (Model 2). As a sensitivity analysis, models for the dialysis outcome were repeated with death as a competing event using Fine and Gray subdistribution hazard models. All analyses were performed using SAS Enterprise Guide (Version 8.1, Cary, NC) and R language (version 4.3.1, R Foundation for Statistical Computing, Vienna, Austria). Results There were 753 unique patient admissions for ADHF requiring placement of a PA catheter, with initial CVP measurements and eGFR data available. Median (25 th , 75 th ) follow-up was 33 (13, 58) months. Baseline Characteristics Baseline characteristics by quartiles of CVP are presented in Table 1 , and by quartiles of baseline PCWP in Table S2 . Mean (SD) age was 62 (14) years, fewer patients were women 238 (31.6%). Mean (SD) eGFR at the time of hospital admission was 57 (27) ml/min/1.73 m 2 and mean eGFR at the time of PA catheter placement was 58 (27) ml/min/1.73 m 2 . Median number of days from hospital admission to PA catheter placement was 2 (IQR 1, 4) days. Table 1. Baseline characteristics according to quartile of initial CVP All N=753 Quartile 1 N=189 Quartile 2 N=185 Quartile 3 N=186 Quartile 4 N=193 Age, years 62.2 (14.4) 61.6 (15.2) 61.3 (15.5) 63.5 (13.2) 62.6 (13.7) Female 238 (31.6%) 60 (31.8%) 59 (31.9%) 56 (30.1%) 63 (32.6%) White Race 612 (84.5%) 152 (83.5%) 154 (87.5%) 152 (84.0%) 154 (83.2%) Hypertension 280 (37.2%) 71 (37.6%) 74 (40.0%) 70 (37.6%) 65 (33.7%) Diabetes 300 (39.8%) 69 (36.5%) 73 (39.5%) 76 (40.9%) 82 (42.5%) Ejection Fraction, % 28.8 (19.4) 27.7 (18.7) 29.8 (19.8) 28.5 (20.2) 29.3 (18.9) EF ≤ 40% 536 (72.3%) 140 (75.3%) 131 (72.0%) 132 (72.1%) 133 (70.0%) Home medications ACEI/ARB 393 (52.2%) 114 (60.3%) 90 (48.7%) 96 (51.6%) 93 (48.2%) MRA 315 (41.8%) 80 (42.3%) 80 (43.2%) 78 (41.9%) 77 (39.9%) Loop Diuretic 575 (76.4%) 126 (66.7%) 147 (79.5%) 140 (75.3%) 162 (83.9%) Inotrope use * 229 (30.4%) 55 (29.1%) 50 (27.0%) 68 (36.6%) 56 (29.0%) Mean Arterial Pressure * 83.8 (12.8) 83.3 (12.6) 84.4 (12.3) 84.0 (14.9) 83.5 (11.4) Days from Admission to PA catheter placement 2 (1, 4) 2 (1, 4) 2 (1, 4) 2 (1, 3) 2 (1, 4) Hemodynamic Parameter CVP, mmHg 13.4 (6.6) 5.8 (2.0) 10.5 (1.2) 14.9 (1.3) 22.2 (4.2) PCWP, mmHg 23.4 (8.6) 16.9 (6.6) 21.2 (6.2) 25.6 (8.0) 29.7 (7.8) CI, L/min/m 2 2.2 (0.7) 2.3 (0.6) 2.2 (0.7) 2.2 (0.8) 2.0 (0.8) eGFR at admission, ml/min/1.73 m 2 57.3 (26.6) 62.3 (26.6) 61.9 (26.6) 54.1 (25.0) 51.1 (26.7) eGFR at PA catheter placement, ml/min/1.73 m 2 58.0 (27.1) 64.5 (27.1) 62.6 (26.7) 55.3 (25.4) 49.8 (26.7) Open in a new tab Values presented as either n (%), mean (standard deviation), median (interquartile rage). The eGFR was calculated using the creatinine-based eGFR cr CKD-EPI 2021 equation. 8 Quartiles of CVP were organized with Quartile 1 being lowest CVP and Quartile 4 being highest CVP. * At the time of the initial invasive hemodynamic measurements CVP: central venous pressure; BMI: body mass index; ACEI:angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blocker; MRA: mineralocorticoid receptor blocker; PCWP: pulmonary capillary wedge pressure; CI: cardiac index; eGFR: estimated glomerular filtration rate; PA: pulmonary artery Initial Measurements of Volume Overload and Outcomes eGFR at the Time of PA Catheter placement Adjusted linear regression using restricted cubic splines curves of initial CVP and PCWP associated with mean eGFR at the time of PA catheter placement are shown in Figure 1 . Higher initial CVP was significantly associated with lower eGFR (β [95%CI] −5.02 [−6.69, −3.35] ml/min/1.73 m 2 per 1 SD higher CVP; Figure 1 ). Higher initial PCWP was also significantly associated with lower eGFR ( Figure 1 ). Figure 1. Open in a new tab Adjusted linear regression curves of initial CVP (top) and initial PCWP (bottom) in association with mean baseline eGFR. The two p-values p global and p linear are for testing the statistical significance of the CVP or PCWP overall using restricted cubic splines and the non-linearity in the adjusted model. Both CVP and PCWP are significantly associated with the mean baseline eGFR ( p global <0.001 and 0.02 for CVP and PCWP, respectively) without any evidence of non-linearity ( p linear 0.5 and 0.6 for CVP and PCWP, respectively). Distribution of CVP and PCWP are shown in gray histogram boxes with percentage prevalence along the right-side y-axis; standard deviations were 6.6 mmHg and 8.6 mmHg for CVP and PCWP, respectively. Models include adjustment for age, sex, ejection fraction, diabetes, home medication use (ACEI/ARB, MRA), inotrope use at time of initial hemodynamic measurements. Abbreviations: CVP: central venous pressure; PCWP: pulmonary capillary wedge pressure; eGFR: estimated glomerular filtration rate; ACEI:angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blocker; MRA: mineralocorticoid receptor blocker In-hospital eGFR Slope Adjusted linear regression using restricted cubic splines curves of initial CVP and PCWP associated with in-hospital eGFR slope are shown in Figure 2 . A median number of 8 (IQR 5, 15) values of eGFR were available for incorporation into estimation of eGFR slope. Higher initial CVP was significantly associated with greater in-hospital eGFR decline (−1.13 [−1.57, −0.69] ml/min/1.73 m 2 /week per 1 SD higher CVP; Figure 2 ). Likewise, higher initial PCWP was associated with greater in-hospital eGFR decline ( Figure 2 , Table 2 ). Figure 2. Open in a new tab Adjusted linear regression curves of initial CVP (A) and initial PCWP (B), and percent change in CVP (C) and percent change in PCWP (D) in association with in-hospital mean eGFR slope per week. The two p-values p global and p linear are for testing the statistical significance of the CVP or PCWP overall using restricted cubic splines and the non-linearity in the adjusted model. While CVP is significantly associated with the mean eGFR in-hospital slope when evaluating the functional relation with restricted cubic splines ( p global <0.001) without evidence of non-linearity ( p linear =0.2), the functional relation between PCWP and eGFR in-hospital slope does not meet statistical significance ( p global =0.05). However, without any evidence of non-linearity ( p linear =0.9), the PCWP was significantly associated with greater in-hospital eGFR decline when examined in a linear regression (−0.70 [−1.14, −0.25]) ml/min/1.73 m 2 . In contrast, neither change in CVP nor change in PCWP was associated with slope of eGFR. Distribution of CVP and PCWP and change in CVP and change in PCWP are shown in gray histogram boxes with percentage prevalence along the right-side y-axis. The mean eGFR slope was calculated by linear mixed model incorporating every eGFR data point (limited to one data point per day of hospitalization). Models include adjustment for age, sex, ejection fraction, diabetes, home medication use (ACEI/ARB, MRA), inotrope use at time of hemodynamic measurements, eGFR. Abbreviations: CVP: central venous pressure; PCWP: pulmonary capillary wedge pressure; ACEI:angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blocker; MRA: mineralocorticoid receptor blocker; eGFR: estimated glomerular filtration rate Table 2. eGFR slope estimates overall and by quartile of initial CVP and PCWP Exposure n Mean (SD) slope, ml/min/1.73 m 2 per week Unadjusted slope per 1 SD higher β (95% CI) Adjusted slope per 1 SD higher β (95% CI) Initial CVP Continuous 753 1.21 (6.26) −0.75 (−1.20, −0.31) −1.13 (−1.57, −0.69) Quartile 1 (Lowest) 189 2.41 (5.29) Reference Reference Quartile 2 185 1.02 (5.77) −1.39 (−2.65, −0.12) −1.44 (−2.65, −0.23) Quartile 3 186 1.15 (6.89) −1.26 (−2.52, 0.00) −1.83 (−3.05, −0.61) Quartile 4 (Highest) 193 0.27 (6.81) −2.14 (−3.39, −0.89) −3.10 (−4.32, −1.87) Initial PCWP Continuous 741 1.24 (6.22) −0.32 (−0.77, 0.13) −0.59 (−1.03, −0.15) Quartile 1 (Lowest) 179 1.62 (5.21) Reference Reference Quartile 2 199 1.36 (5.98) −0.26 (−1.51, 1.00) −0.60 (−1.82, 0.63) Quartile 3 177 1.49 (7.14) −0.13 (−1.43, 1.16) −0.59 (−1.86, 0.67) Quartile 4 (Highest) 186 0.52 (6.41) −1.09 (−2.37, 0.18) −1.77 (−3.01, −0.52) Open in a new tab The eGFR slope was calculated from the first CVP (or PCWP) to discharge from the hospitalization using linear mixed model regression. A median 8 (IQR 5, 15) values of eGFR per patient were available for incorporation into the slope estimates. The eGFR values were estimated using the CKD-EPI 2021 formula based on creatinine. The first eGFR of each day was selected if more than one value was available. Adjusted models included adjustment for age, sex, ejection fraction, diabetes, home use of angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, home use of mineralocorticoid receptor antagonists, and need for inotrope therapy at the time of initial pulmonary artery catheter placement. Abbreviations: CVP, central venous pressure; PCWP, pulmonary capillary wedge pressure Initiation of Dialysis Over the course of 33 (IQR 13, 58) months of follow-up, 62 (8.2%) needed dialysis, of whom 14 (1.9%) initiated dialysis during the index hospitalization. In multivariable Cox PH models (Model 1), higher initial CVP and PCWP were each associated with a significantly increased risk of dialysis ( Table 3 ). After including PCWP as an adjustment covariate in models of initial CVP (Model 2), the association between initial CVP and need for dialysis continued to meet statistical significance (HR 1.43 [95% CI 1.06, 1.93] per 1 SD higher CVP) but was attenuated and no longer met statistical significance when models of initial PCWP were adjusted for CVP (HR 1.07 [95% CI 0.80, 1.44] per 1 SD higher PCWP). Table 3. Hazard ratios for association of initial CVP and initial PCWP with the need for dialysis N N Events Follow-up (years) Event Rate per 100-py Unadjusted HR (95% CI) Model 1 HR (95% CI) Model 2 HR (95% CI) Initial CVP Continuous (per 1 SD higher) 753 62 2159 2.9 1.68 (1.34, 2.10) 1.49 (1.17, 1.90) 1.43 (1.06, 1.93) Quartile 1 (Lowest) 189 <11 * 617 1.3 Reference Reference Reference Quartile 2 185 14 580 2.4 1.84 (0.77, 4.39) 1.68 (0.70, 4.04) 1.56 (0.64, 3.77) Quartile 3 186 17 508 3.4 2.45 (1.06, 5.67) 2.07 (0.89, 4.84) 1.74 (0.71, 4.28) Quartile 4 (Highest) 193 23 455 5.1 3.59 (1.61, 8.04) 2.41 (1.07, 5.44) 1.82 (0.71, 4.66) Initial PCWP Continuous (per 1 SD higher) 741 62 2135 2.9 1.45 (1.15, 1.83) 1.30 (1.04, 1.62) 1.07 (0.80, 1.44) Quartile 1 (Lowest) 179 <11 * 486 1.7 Reference Reference Reference Quartile 2 199 18 622 2.9 1.97 (0.86, 4.53) 2.06 (0.88, 4.81) 1.79 (0.76, 4.21) Quartile 3 177 13 514 2.5 1.73 (0.72, 4.17) 1.75 (0.72, 4.28) 1.26 (0.50, 3.21) Quartile 4 (Highest) 186 23 512 4.5 3.00 (1.34, 6.71) 2.70 (1.20, 6.09) 1.47 (0.56, 3.86) Open in a new tab Cox proportional hazards regression modeling for initial CVP and initial PCWP, both as a continuous variable as well as by quartiles. For quartiles of initial volume overload measurement, Quartile 1 is the lowest and Quartile 4 is the highest. Hazard ratios are per every 1 standard deviation higher for continuous models. Event rates are calculated per 100 patient-years (100-py) Model 1: Adjusted for age, sex, ejection fraction, diabetes, home medication use (ACEI/ARB, MRA), inotrope use at time of initial hemodynamic measurement, eGFR Model 2: Adjusted for everything in Model 1, with addition of: adjustment for PCWP in the initial CVP models or adjustment for CVP in the initial PCWP models * Actual N suppressed in accordance to USRDS publication policy Abbreviations: CVP, central venous pressure; PCWP, pulmonary capillary wedge pressure; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; MRA, mineralocorticoid receptor antagonist; eGFR, estimated glomerular filtration rate Death, LVAD, and Heart Transplant Outcomes Over the course of follow-up, 264 (35.1%) died and 406 (54.0%) either died or underwent LVAD implantation or HT. In multivariable Cox PH models (Model 1), higher initial CVP and PCWP were each associated with increased risk of death ( Table 4 ). After including PCWP as an adjustment covariate in models of CVP (Model 2), the association between initial CVP and death continued to meet statistical significance (HR 1.30 [95% CI 1.11, 1.52] per 1 SD higher initial CVP). However, associations were attenuated and no longer met statistical significance when models of PCWP were adjusted for CVP (HR 1.00 [95% CI 0.85, 1.18] per 1 SD higher initial PCWP). In terms of the composite outcome, associations were similar ( Table 4 ). Table 4. Hazard ratios for death and composite outcome of death, left ventricular assist device implantation and heart transplant Death Composite of Death, LVAD, and HT N N Events Event Rate per 100-py Unadjusted HR (95%CI) Model 1 HR (95%CI) Model 2 HR (95%CI) N Events Event Rate per 100-py Unadjusted HR (95%CI) Model 1 HR (95%CI) Model 2 HR (95%CI) Initial CVP Continuous (per 1 SD) 753 264 11.6 1.30 (1.17, 1.46) 1.30 (1.15, 1.47) 1.30 (1.11, 1.52) 406 24.8 1.14 (1.03, 1.25) 1.12 (1.02, 1.24) 1.03 (0.91, 1.17) Quartile 1 189 47 7.4 Reference Reference Reference 93 21.9 Reference Reference Reference Quartile 2 185 53 8.7 1.19 (0.80, 1.76) 1.18 (0.79, 1.75) 1.16 (0.77, 1.74) 86 18.5 0.92 (0.69, 1.23) 0.93 (0.69, 1.24) 0.88 (0.65, 1.19) Quartile 3 186 74 13.6 1.79 (1.24, 2.58) 1.55 (1.07, 2.25) 1.53 (1.03, 2.29) 110 29.2 1.34 (1.01, 1.76) 1.27 (0.97, 1.68) 1.14 (0.84, 1.55) Quartile 4 193 90 18.5 2.31 (1.63, 3.29) 2.03 (1.41, 2.90) 1.97 (1.28, 3.02) 117 31.8 1.43 (1.09, 1.87) 1.37 (1.04, 1.81) 1.14 (0.82, 1.59) Initial PCWP Continuous (per 1 SD) 741 259 11.5 1.14 (1.01, 1.29) 1.17 (1.03, 1.33) 1.00 (0.85, 1.18) 400 24.8 1.19 (1.08, 1.31) 1.18 (1.06, 1.30) 1.15 (1.02, 1.31) Quartile 1 179 51 10.1 Reference Reference Reference 82 21.3 Reference Reference Reference Quartile 2 199 66 10.1 1.10 (0.76, 1.59) 0.94 (0.65, 1.36) 0.82 (0.56, 1.20) 102 21.6 1.13 (0.85, 1.52) 1.02 (0.76, 1.37) 0.99 (0.74, 1.34) Quartile 3 177 75 13.9 1.51 (1.06, 2.15) 1.43 (1.00, 2.04) 1.07 (0.72, 1.60) 108 28.2 1.47 (1.10, 1.95) 1.36 (1.02, 1.82) 1.29 (0.95, 1.77) Quartile 4 186 67 12.2 1.30 (0.90, 1.87) 1.35 (0.93, 1.95) 0.90 (0.57, 1.39) 108 29.1 1.47 (1.10, 1.96) 1.38 (1.03, 1.85) 1.27 (0.89, 1.80) Open in a new tab Cox proportional hazards regression modeling for baseline CVP and PCWP, both as a continuous variable as well as by quartiles (with Quartile 1 being lowest and Quartile 4 being highest). Hazard ratios are per every 1 standard deviation higher for continuous models. Event rates are calculated per 100 patient-years (100-py) Model 1: Adjusted for age, sex, ejection fraction, diabetes, home medication use (ACEI/ARB, MRA), inotrope use at time of initial hemodynamic measurements, eGFR Model 2: Adjusted for everything in Model 1, with addition of adjustment for PCWP in the CVP models or adjustment for CVP in the PCWP models Abbreviations: CVP, central venous pressure; PCWP, pulmonary capillary wedge pressure; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; MRA, mineralocorticoid receptor antagonist; eGFR, estimated glomerular filtration rate Change in Measurements of Volume Overload and Outcomes Concomitant eGFR Slope There were 546 patients with both an initial and last measurement of CVP, and 516 patients available with both an initial and last measurement of PCWP ( Figure S1 ), with median 5 (IQR 3, 8) values of eGFR available for incorporation into slope estimation. Baseline characteristics according to quartiles of percent change in CVP and PCWP are shown in Table S3 and S4 , respectively. Adjusted linear regression using restricted cubic splines curves of percent change in CVP and PCWP associated with concomitant eGFR slope are shown in Figure 2 and Table 5 . There was no clear tendency between degree of decline in CVP with eGFR slope, nor between degree of decline in PCWP and eGFR slope ( Figure 2 , Table 5 ). Table 5. eGFR slope estimates overall and by quartile of % change in CVP and PCWP Exposure n Initial level, mmHg Final level, mmHg % Change in volume overload marker Mean (SD) slope, ml/min/1.73 m 2 per week Unadjusted slope per 1 SD higher β (95% CI) Adjusted slope per 1 SD higher β (95% CI) % Change in CVP Continuous 546 13.9 (6.4) 10.54 (5.0) −20.0 (−50.0, 12.5) 2.71 (10.67) 0.40 (−0.45, 1.23) 0.39 (−0.62, 1.40) Quartile 1 (Greatest Decrease) 138 17.2 (6.2) 6.5 (2.8) −60.0 (−68.2, −53.3) 3.81 (12.16) Reference Reference Quartile 2 132 16.5 (5.7) 10.5 (3.7) −35.7 (−41.2, −30.4) 1.60 (9.79) −2.22 (−4.76, 0.33) −1.94 (−4.38, 0.51) Quartile 3 142 13.4 (5.4) 12.5 (5.0) −7.7 (−12.5, 0.0) 1.86 (8.39) −1.96 (−4.46, 0.55) −1.71 (−4.22, 0.80) Quartile 4 (Least Decrease/Increase) 134 8.5 (4.0) 12.6 (5.3) 38.8 (25.0, 71.4) 3.60 (11.87) −0.22 (−2.75, 2.32) −0.31 (−3.14, 2.51) % Change in PCWP Continuous 516 24.4 (8.3) 20.0 (6.7) −16.7 (−35.7, 5.20) 2.92 (10.63) −0.15 (−1.08, 0.77) −0.38 (−1.47, 0.70) Quartile 1 (Greatest Decrease) 132 29.2 (8.0) 14.8 (4.7) −46.2 (−57.3, −39.3) 2.26 (11.26) Reference Reference Quartile 2 128 26.7 (6.6) 19.7 (4.9) −25.5 (−31.3, −20.0) 3.20 (9.13) 0.94 (−1.66, 3.54) 0.92 (−1.58, 3.42) Quartile 3 127 23.8 (6.7) 22.1 (6.4) −7.7 (−12.5, 0.0) 3.08 (8.46) 0.82 (−1.78, 3.43) 1.07 (−1.52, 3.66) Quartile 4 (Least Decrease/Increase) 129 17.8 (6.2) 23.8 (6.6) 28.6 (14.3, 52.4) 3.16 (13.09) 0.90 (−1.69, 3.49) 0.92 (−1.99, 3.82) Open in a new tab Values for initial and final CVP or PCWP are presented as mean (SD), and % change as median (IQR). The eGFR slope was calculated between the first and final CVP (or PCWP) using linear mixed models. A median 5 (IQR 3, 8) values of eGFR per patient were available for incorporation into the slope estimates. Adjusted models included adjustment for age, sex, ejection fraction, diabetes, home use of angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, home use of mineralocorticoid receptor antagonists, and need for inotrope therapy at the time of initial pulmonary artery catheter placement. Initiation of Dialysis Over the course of follow-up from the final CVP date, 46 (8.4%) initiated dialysis. In multivariable Cox PH models (Model 1), neither greater increase in CVP nor PCWP were associated with risk of dialysis (HR [95%CI] per 1 SD percent higher CVP or PCWP were 1.25 [0.85, 1.84]; 1.13 [0.75, 1.71] respectively; Table 6 ). Table 6. Hazard ratios for association of change in CVP and change in PCWP with the need for dialysis N N Events Follow-up (years) Event Rate per 100-py Unadjusted HR (95% CI) Model 1 HR (95% CI) Model 2 HR (95% CI) % Change in CVP Continuous (per 1 SD greater increase) 546 46 1593 2.9 0.84 (0.57, 1.22) 0.90 (0.59, 1.38) 1.25 (0.85, 1.84) Quartile 1 (Greatest Decrease) 138 12 403 3.0 Reference Reference Reference Quartile 2 132 <11 * 366 3.0 0.99 (0.44, 2.25) 0.89 (0.39, 2.04) 1.04 (0.45, 2.42) Quartile 3 142 13 432 3.0 1.05 (0.48, 2.30) 0.86 (0.38, 1.92) 1.33 (0.57, 3.13) Quartile 4 (Least Decrease) 134 <11 * 393 2.6 0.87 (0.37, 2.01) 0.92 (0.38, 2.19) 2.18 (0.75, 6.34) % Change in PCWP Continuous (per 1 SD greater increase) 516 45 1520 3.0 1.04 (0.77, 1.39) 1.00 (0.73, 1.39) 1.13 (0.75, 1.71) Quartile 1 (Greatest Decrease) 132 <11 * 397 2.3 Reference Reference Reference Quartile 2 128 13 405 3.2 1.48 (0.63, 3.47) 1.57 (0.65, 3.80) 1.79 (0.71, 4.48) Quartile 3 127 12 358 3.4 1.48 (0.63, 3.52) 1.20 (0.49, 2.92) 1.47 (0.56, 3.84) Quartile 4 (Least Decrease) 129 <11 * 359 3.1 1.33 (0.55, 3.22) 1.28 (0.51, 3.17) 1.85 (0.61, 5.59) Open in a new tab Cox proportional hazards regression modeling for change in CVP and change in PCWP, both as a continuous variable as well as by quartiles. For quartiles of percent change in volume overload measurement, Quartile 1 is the greatest degree of percent decline and Quartile 4 is the least degree of decline. Hazard ratios are per every 1 standard deviation greater increase for continuous models. Event rates are calculated per 100 patient-years (100-py) Model 1: Adjusted for age, sex, ejection fraction, diabetes, home medication use (ACEI/ARB, MRA), inotrope use at time of initial hemodynamic measurement, eGFR Model 2: Adjusted for everything in Model 1, with addition of adjustment for initial CVP in change in CVP models or adjustment for initial PCWP in change in PCWP models * Actual N suppressed in accordance to USRDS publication policy Sample sizes for the % change in CVP and PCWP are limited to the patients with a first and last CVP (or PCWP) available. Follow-up time starts at the time of the last CVP (or PCWP) measurement. Abbreviations: CVP, central venous pressure; PCWP, pulmonary capillary wedge pressure; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; MRA, mineralocorticoid receptor antagonist; eGFR, estimated glomerular filtration rate Death, LVAD, and Heart Transplant Outcomes Results examining the change in CVP and PCWP with death and the composite outcome of death, LVAD or HT are shown in Table S5 . An increase in CVP was not associated with risk of death (HR 1.19 [95% CI 0.98, 1.43] per 1 SD greater increase in CVP) nor the composite outcome (HR 1.05 [95% CI 0.93, 1.19] per 1 SD greater increase in CVP). There was an association observed between increase in PCWP with risk of death (HR 1.24 [95% CI 1.01, 1.52] per 1 SD greater increase in PCWP) and the composite outcome (HR 1.16 [95% CI 1.01, 1.34] per 1 SD greater increase in PCWP). Sensitivity Analyses We repeated models treating death prior to dialysis as a competing risk. Subdistribution hazard ratios were similar for initial CVP and PCWP (sHR 1.34 [1.02, 1.77] for CVP; sHR 1.23 [1.00, 1.52] for PCWP; Table S6 ). There was no significant association between change in CVP or change in PCWP with dialysis when treating death prior to dialysis as a competing risk ( Table S6 ). Discussion Among patients admitted with ADHF, invasively obtained measures of volume overload—elevated CVP and PCWP—were associated with higher risk of several adverse kidney outcomes. Higher initial CVP and PCWP were each associated with lower eGFR, greater in-hospital eGFR decline and higher risk of starting dialysis. While the associations for PCWP lost significance after multivariable adjustment, the associations with CVP remained significant even after adjustment for baseline PCWP. The change in CVP and PCWP, however, did not appear to be associated with either concomitant changes in eGFR or risk of starting dialysis. These findings suggest that greater degree of volume overload on initial measurement, in particular right-sided overload, may be a marker for higher risk of adverse kidney outcomes among patients with ADHF. Reduced kidney function remains among one of the strongest risk factors for adverse clinical outcomes among patients with ADHF. 1 , 10 However, despite low kidney function being such an important marker, the pathophysiology underlying reductions in eGFR in HF remains unclear. In terms of potential mechanisms, animal models have shown evidence of tubular injury from volume overload in the kidney, 11 and ultrasound studies have suggested transmission of volume overload into changes in renal vein flow patterns, 12 evidence to suggest that perhaps overload itself can contribute to declines in kidney function. However, most of the prior literature examining invasive measures of volume overload in relation to kidney function among patients with ADHF has largely been cross-sectional in nature or only included acute fluctuations in kidney function over the span of a few days, 2 , 3 , 5 making it challenging to draw any longitudinal interpretations regarding the relation between volume overload and clinically important longer term changes in kidney function. In terms of examining short-term changes in kidney function, one study showed that higher CVP was associated with in-hospital rise of creatinine by ≥0.3 mg/dl. 2 However, other studies using data from the Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE) trial have found no association between elevated baseline CVP or PCWP and increase in creatinine by ≥0.3 mg/dl 3 or with eGFR decline by >20%. 13 This current study has expanded upon these prior results, through a larger sample size of 753 patients, use of linear regression models for a more refined understanding of eGFR slope, and incorporation of dialysis events. We observed a significant linear association between initial CVP and PCWP with in-hospital eGFR slope, whereby patients with higher initial CVP and PCWP measurements had significantly greater in-hospital eGFR declines. Furthermore, measures of volume overload were related to longitudinal declines in eGFR and to requiring dialysis over a median follow-up of nearly 3-years, a clinically meaningful outcome that has not been extensively evaluated in the HF literature. For the kidney outcomes, the associations between PCWP and kidney function decline attenuated after adjustment for the CVP, but not the reverse. There are several potential explanations for this result. First, patients with ADHF with greater right-sided volume overload may represent the patients with right-sided cardiac dysfunction and with the most overall advanced cardiac disease (including left-sided disease) and therefore are already at higher risk for progressive kidney disease. Our covariates included EF and need for inotropes at the time of PA catheter placement as two indices for severity of cardiac disease and while associations remained similar after adjustment, there is possibility of residual confounding. The decision to initiate dialysis may have been influenced in part by the degree of CVP elevation; however, initiation of dialysis for fluid removal would not necessarily be related to the differential association between CVP and PCWP and kidney outcomes that we observed. Second, high CVP tends to reflect systemic venous congestion and may more directly cause congestive nephropathy through higher pressure in the renal veins. This potential mechanism would be consistent with prior studies showing features of right-sided volume overload —such as higher intra-abdominal bladder pressures and changes in intra-abdominal and intra-renal venous flow patterns from fluid overload—being associated with reduced eGFR among patients with ADHF. 12 , 14 One prior study has also similarly observed that indices of worse right ventricular function were associated with adverse kidney outcomes whereas indices of left ventricular dysfunction were less consistently associated, albeit in a very different HF patient population (ambulatory patients with heart failure with preserved ejection fraction). 15 On the other hand, decrease in volume overload—as quantified in our study as percent decline in CVP and PCWP—showed no association with either concomitant eGFR slope or with risk of dialysis. It remains challenging to infer directionality between changes in volume overload and changes in eGFR (i.e., we cannot assume that decrease in volume overload leads to increases or decreases in eGFR, or vice versa), since they are occurring in tandem. Regardless, we also did not observe any association between change in volume overload and either increased or decreased risk of requiring dialysis over follow-up. One interpretation could be that the degree of initial overload is simply a strong marker for the highest risk patient who then develops worse kidney outcomes including decline in eGFR and needing dialysis. Our results, however, cannot directly answer the question of whether intervening on the volume overload to reduce the CVP or PCWP is associated with an improvement or a decline in eGFR in individual patients. Major strengths of this study include the large cohort size of patients with invasive hemodynamic data and length of follow-up time of nearly 3-years. We have evaluated clinically important kidney endpoints, namely need for dialysis, which is lacking in many studies of ADHF. In terms of limitations, the study was a retrospective, observational study. Indications for placement of a PA catheter were not ascertained, and in some cases the decision could have been influenced by declines in eGFR; however, on average, the placement of the PA catheter occurred 2-days into the hospitalization and overall the eGFR from the time of admission to the time of PA catheter placement was 57 vs 58 ml/min/1.73 m 2 , suggesting that for the most part, PA catheters were placed for reasons outside of a decline in eGFR. Some clinical outcomes, for example the decision to start dialysis, could have been driven by the data obtained from the PA catheter. Lastly, the best approach for evaluation of the composite cardiovascular outcome among patients with such advanced HF is challenging, as there are numerous measured and unmeasured factors related to the decision for LVAD placement or timing of a HT. Among patients admitted for ADHF, greater degree of volume overload, particularly right-sided overload, was associated with lower baseline eGFR, more in-hospital declines in eGFR, and an increased risk of requiring dialysis. However, changes in volume overload were not associated with concomitant changes in eGFR or with increased or decreased risk of dialysis. Supplementary Material 1 Figure S1: Consort diagram of eligible patients included Table S1: Diagnostic concept codes for cohort identification Table S2: Baseline characteristics according to quartile of initial PCWP Table S3: Baseline characteristics according to quartile of percent change in CVP Table S4: Baseline characteristics according to quartile of percent change in PCWP Table S5: Hazard ratios for mortality and composite outcome based on change in CVP and change in PCWP Table S6: Sub-distribution hazard ratios for need for dialysis outcome based on CVP and PCWP using Fine and Gray competing risk model NIHMS2088592-supplement-1.pdf (346.6KB, pdf) Support: The work was supported by the NIH CTSA grant (CTSI UM1TR004398), NIH NIDDK training grant (K23DK128657 to WM), and the American Kidney Fund Clinical Scientist in Nephrology Fellow Program (to MT). Funders had no role in study design, data collection, analysis, reporting, or the decision to submit the manuscript for publication. Financial Disclosure: J.M.T. reports grants and/or personal fees from 3ive labs, Bayer, Bristol Myers Squibb, Astra Zeneca, Novartis, Cardionomic, MagentaMed, Reprieve inc., FIRE1,W.L. Gore, Sanofi, Sequana Medical, Otsuka, Abbott, Merck, Windtree Therapeutics, Lexicon pharmaceuticals, Precardia, Relypsa, Regeneron, BD, Edwards life sciences, and Lilly. In addition, J.M.T. has a patent for Treatment of diuretic resistance issued to Yale and Corvidia Therapeutics Inc, a patent methods for measuring renalase issued to Yale, and a patent Treatment of diuretic resistance pending with Reprieve inc. MJS attended an advisory board for Boehringer Ingelheim and is a member of the Akebia Steering Committee Trials. MJS’s spouse in an employee of Eli Lilly. MT reports royalty funds from a patent from Genentech. CH reports current employment with and equity ownership in Holmusk Technologies, Inc. The remaining authors declare that they have no relevant financial interests. Footnotes Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Disclaimer: The data reported here have partially been supplied by the United States Renal Data System. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy or interpretation of the U.S. government. Data Sharing: Deidentified data created for the study will be available upon request to the principal investigator after proposals for collaboration or data sharing have been reviewed. References 1. Heywood JT, Fonarow GC, Costanzo MR, et al. High prevalence of renal dysfunction and its impact on outcome in 118,465 patients hospitalized with acute decompensated heart failure: a report from the ADHERE database. J Card Fail. 2007;13(6):422–430. doi: 10.1016/j.cardfail.2007.03.011 [ DOI ] [ PubMed ] [ Google Scholar ] 2. Mullens W, Abrahams Z, Francis GS, et al. Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J Am Coll Cardiol. 2009;53(7):589–596. doi: 10.1016/j.jacc.2008.05.068 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Nohria A, Hasselblad V, Stebbins A, et al. Cardiorenal interactions: insights from the ESCAPE trial. J Am Coll Cardiol. 2008;51(13):1268–1274. doi: 10.1016/j.jacc.2007.08.072 [ DOI ] [ PubMed ] [ Google Scholar ] 4. Testani JM, Khera AV, St John Sutton MG, et al. Effect of right ventricular function and venous congestion on cardiorenal interactions during the treatment of decompensated heart failure. Am J Cardiol. 2010;105(4):511–516. doi: 10.1016/j.amjcard.2009.10.020 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Damman K, van Deursen VM, Navis G, Voors AA, van Veldhuisen DJ, Hillege HL. Increased central venous pressure is associated with impaired renal function and mortality in a broad spectrum of patients with cardiovascular disease. J Am Coll Cardiol. 2009;53(7):582–588. doi: 10.1016/j.jacc.2008.08.080 [ DOI ] [ PubMed ] [ Google Scholar ] 6. Voss EA, Makadia R, Matcho A, et al. Feasibility and utility of applications of the common data model to multiple, disparate observational health databases. J Am Med Inform Assoc. 2015;22(3):553–564. doi: 10.1093/jamia/ocu023 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Hripcsak G, Duke JD, Shah NH, et al. Observational Health Data Sciences and Informatics (OHDSI): Opportunities for Observational Researchers. Stud Health Technol Inform. 2015;216:574–578. [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Inker LA, Eneanya ND, Coresh J, et al. New Creatinine- and Cystatin C–Based Equations to Estimate GFR without Race. N Engl J Med. 2021;385(19):1737–1749. doi: 10.1056/NEJMoa2102953 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. U.S. Renal Data System. 2022 USRDS annual data report: Epidemiology of kidney disease in the United States. National Institutes of Health, National Institute of Diabetes and DIgestive and Kidney DIseases, Bethesda, MD. Published online 2022. [ Google Scholar ] 10. Hillege HL, Nitsch D, Pfeffer MA, et al. Renal function as a predictor of outcome in a broad spectrum of patients with heart failure. Circulation. 2006;113(5):671–678. doi: 10.1161/CIRCULATIONAHA.105.580506 [ DOI ] [ PubMed ] [ Google Scholar ] 11. Shimada S, Hirose T, Takahashi C, et al. Pathophysiological and molecular mechanisms involved in renal congestion in a novel rat model. Sci Rep. 2018;8(1):16808. doi: 10.1038/s41598-018-35162-4 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Nijst P, Martens P, Dupont M, Tang WHW, Mullens W. Intrarenal Flow Alterations During Transition From Euvolemia to Intravascular Volume Expansion in Heart Failure Patients. JACC Heart Fail. 2017;5(9):672–681. doi: 10.1016/j.jchf.2017.05.006 [ DOI ] [ PubMed ] [ Google Scholar ] 13. Testani JM, McCauley BD, Kimmel SE, Shannon RP. Characteristics of Patients With Improvement or Worsening in Renal Function During Treatment of Acute Decompensated Heart Failure. American Journal of Cardiology. 2010;106(12):1763–1769. doi: 10.1016/j.amjcard.2010.07.050 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Mullens W, Abrahams Z, Skouri HN, et al. Elevated intra-abdominal pressure in acute decompensated heart failure: a potential contributor to worsening renal function? J Am Coll Cardiol. 2008;51(3):300–306. doi: 10.1016/j.jacc.2007.09.043 [ DOI ] [ PubMed ] [ Google Scholar ] 15. Lu H, Chatur S, Lee S, et al. Relationship Between Cardiac Structure and Function With Renal Function Trajectory and Outcomes in Patients With Heart Failure: Insights From the PARAGON-HF Trial. Circ Heart Fail. 2024;17(10):e011942. doi: 10.1161/CIRCHEARTFAILURE.124.011942 [ 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 1 Figure S1: Consort diagram of eligible patients included Table S1: Diagnostic concept codes for cohort identification Table S2: Baseline characteristics according to quartile of initial PCWP Table S3: Baseline characteristics according to quartile of percent change in CVP Table S4: Baseline characteristics according to quartile of percent change in PCWP Table S5: Hazard ratios for mortality and composite outcome based on change in CVP and change in PCWP Table S6: Sub-distribution hazard ratios for need for dialysis outcome based on CVP and PCWP using Fine and Gray competing risk model NIHMS2088592-supplement-1.pdf (346.6KB, pdf) Data Availability Statement Deidentified data created for the study will be available upon request to the principal investigator after proposals for collaboration or data sharing have been reviewed. ACTIONS View on publisher site PDF (707.7 KB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Cardiorenal Interactions During Treatment of Acute Decompensated Heart Failure: Diuretics Versus Ultrafiltration#848383
• Readily Available Urinary Sodium Analysis in Patients With Acute Decompensated Heart Failure#864438
• Boston Medical Center Ultrasound Decongestion Study in Heart Failure#826843
• Predicting acute kidney injury in adults after cardiac surgery with a novel clinical metric: The cardiac diuretic-responsiveness index.#458589
• The Efficacy and Safety of Dapagliflozin in Improving Heart Failure in Dialysis Patients#1057518