Serial evaluation of loop diuretic efficiency following left ventricular assist device implantation

Abstract

More than 50% of heart failure (HF) patients require diuretic therapy after left ventricular assist device (LVAD). Although few data related to diuretic response (DR) exist in stage D patients, tubular sodium reabsorption may be clinically prognostic independent of estimated glomerular filtration rate (eGFR) and proteinuria within this cohort. We aimed to characterize DR serially before and after LVAD implantation in a stage D population. We conducted a prospective, observational cohort study of HF patients receiving diuretics with plans to undergo LVAD implantation. We measured urine sodium (U_Na) and creatinine (U_Cr) at three points after diuretic therapy: pre-LVAD, post-LVAD prior to discharge, and as an outpatient. Prior to LVAD, patients (N = 19) had an average eGFR of 54.0 ± 18.0 mL/min/1.73 m², spot U_Na of 74.8 ± 28.0 mmol/L, and fractional excretion of sodium (FE_Na) of 3.1 ± 2.7%. Pre-LVAD, eGFR did not correlate with spot U_Na nor FE_Na (p > 0.05 for both). LVAD implantation did not improve DR post-LVAD (mean change FE_Na per 40 mg IV furosemide 0.5 ± 1.0%; p = 0.84), and 90% of patients required loop diuretics at 90 days post-surgery. Improved hemodynamics following LVAD may not improve DR or tubular function; larger studies are needed to confirm our results and assess the utility of DR to predict post-LVAD outcomes.

Keywords

Congestive heart failure ventricular assist devices (VAD)renal physiology cardiac drugs circulatory support right heart failure hemodynamic effects

Introduction

Risk stratification for patients undergoing left ventricular assist device (LVAD) implantation is needed to improve prognostication.^1–5 Despite improvement in cardiac output, more than one-half of patients still struggle with congestion and require chronic loop diuretic therapy.^6,7 Kidney function, measured by estimated glomerular filtration rate (eGFR), after LVAD may change substantially and can potentially influence resulting morbidity and mortality.⁸ The eGFR measures filtration function of the kidney but not the kidney’s tubular sodium handling. Diuretic response (DR), also called diuretic efficiency, is a validated measure of tubular sodium handling expressed as the DR relative to the diuretic dose.^9–13 DR has significant interpatient variability at a given eGFR and is independently correlated with adverse heart failure (HF) outcomes, including increased mortality.^14
–18 Furthermore, DR appears unrelated to cardiac output and may not change after LVAD.⁹ Evaluation of DR could improve risk prognostication post-LVAD, predict those requiring diuretic therapy post-LVAD to manage congestion, and provide insight into changes in tubular sodium handling post-LVAD. In this study, we aimed to serially describe DR in a Stage D HF population, both before and after LVAD implantation.

Materials and methods

Study design

This trial was a prospective, observational cohort trial conducted at Vanderbilt University Medical Center from February 2020 to July 2021. Adult patients were eligible for enrollment if they had stage D HF on scheduled loop diuretic therapy and had a planned LVAD implantation. Exclusion criteria included the need for renal replacement therapy or right ventricular mechanical support prior to LVAD; ongoing evaluation for liver or kidney transplantation; pregnancy or breastfeeding; or current incarceration. The study was approved by the Vanderbilt Institutional Review Board, and all data were collected and managed using REDCap electronic data capture tool.^19,20

Patients were administered oral torsemide 100 mg, or alternative loop diuretic regimen if deemed necessary per the primary provider, prior to each urine sample. Urine samples were collected 2 ± 0.5 h following loop diuretic administration to assess spot urine sodium and spot urine creatinine. The urine samples were collected at three serial time points throughout the study: (1) during the index hospitalization prior to LVAD implantation; (2) post-LVAD prior to hospital discharge during index hospitalization when clinically stable on oral loop diuretics; and (3) post-LVAD at the earliest available outpatient visit. Patients were evaluated prospectively throughout study enrollment to determine appropriate urine collection timing per study criteria. All patients received a 2000 mL fluid restriction and 2000 mg/day sodium diet.

Study endpoints

We assessed unadjusted DR metrics of spot urine sodium concentration, fractional excretion of sodium (FE_Na), and calculated 6-h cumulative sodium output using the Natriuretic Response Prediction Equation (NPRE). FE_Na was calculated as: (Na_urine/Na_serum) x (Cr_serum/Cr_urine) x 100%.NRPE was calculated as: Na output (mmol) = eGFR × $(\frac{B S A}{1.73}) \times (\frac{S e r u m C r}{U r i n e C r}) \times 60 \times 3.25 h \times (\frac{U r i n e N a}{1000 m L})$ .^21,22 To account for changes in diuretic dose over the study period, the DR metrics were assessed relative to 40 mg of furosemide equivalent (FE) as previously described using a log-linear relationship.²³ All loop diuretic doses were converted to FEs using the conversion: 80 mg oral (PO) furosemide = 40 mg intravenous (IV) furosemide = 20 mg PO torsemide = 1 mg IV/PO bumetanide.^24,25 Other outcomes included change in hemodynamics, change in eGFR via Modification of Diet in Renal Disease (MDRD) equation, and descriptive post-LVAD outcomes at urine sample 3.

Statistical analysis

Baseline characteristics were calculated for all patients, and outcomes were analyzed for all patients actively enrolled at the time of the sample. Descriptive statistics for categorical variables were reported as counts with percentages, and continuous variables were reported as means with standard deviations (SD). Spearman’s correlation coefficients and corresponding p-values were reported to describe the relationship between continuous variables. For pairwise comparison of variables at different timepoints in the same patient, we used a one-sample Wilcox signed-rank test to test the differences of measurements between two timepoints. All statistical analyses were conducted using R software version 4.13 and IBM SPSS 28. Scatter plots were created using R package ggplot2.^26–28

Results

A total of 19 patients were enrolled in the trial (Figure 1). Seven patients were not included at urine sample 2 due to deferred/declined LVAD (n = 3) or death (n = 4) and two additional patients at urine sample 3 were lost to follow-up. Baseline characteristics before LVAD implantation are summarized in Table 1. The mean age was 53 ± 13 years, and the majority of patients received a HeartMate 3™ LVAD device (79%). The baseline cardiac index (CI) by Fick prior to LVAD was 1.8 ± 0.5 L/min/m² with an eGFR of 54 ± 18 mL/min/1.73 m². Other baseline hemodynamic parameters included: central venous pressure (CVP) 10.6 ± 4.9 mmHg, pulmonary capillary wedge pressure (PCWP) 25.1 ± 9.4 mmHg, and mean arterial pressure (MAP) 80.4 ± 11.6 mmHg (Table 2). The cohort required an average daily oral furosemide dose of 352 ± 322 mg prior to enrollment.

Figure 1.

Study timeline.

Table 1.

Baseline demographics.

Characteristic	N = 19
Age (years)	52.6 (12.8)
Male	14 (73.7)
Etiology of heart failure
Ischemic	9 (47.4)
Non-ischemic	10 (52.6)
INTERMACS profile
1	1 (5.3)
2	4 (21.1)
3	9 (47.4)
4	5 (26.3)
5–7	0 (0)
Pre-LVAD ejection fraction (%)	17.3 (6.6)
Pre-LVAD cardiac index, Fick (L/min/m²)	1.8 (0.5)
Pre-LVAD BNP (pg/mL)	750.2 [38–2880]
Pre-LVAD eGFR (mL/min/1.73 m²)	54.0 (18.0)
Home daily loop diuretic dose in FE (mg)	352 (321.9)
LVAD indication
Bridge to transplant	3 (15.8)
Bridge to decision	8 (42.1)
Destination therapy	8 (42.1)
LVAD device^a
HeartWare™	1 (5.3)
HeartMate™ 3	15 (78.9)

INTERMACS: Interagecy Registry for Mechanically Assisted Circulatory Support; LVAD: left ventricular assist device; BNP: brain natriuretic peptide; eGFR: estimated glomerular filtration rate; FE: furosemide equivalents.

Values displayed in the right column are mean ± SD, median [IQR], or N (%).

Three LVAD devices were not implanted.

Table 2.

Serial hemodynamic and eGFR measurements.

	Pre-LVAD urine sample 1	Post-LVAD urine sample 2	Post-LVAD urine sample 3
Cardiac index, Fick (L/min/m²)	1.8 ± 0.5	2.5 ± 0.5	–
CVP (mmHg)	10.6 ± 4.9	9.6 ± 3.2	–
PCWP (mmHg)	25.1 ± 9.4	14.3 ± 6.4	–
MAP or Doppler (mmHg)	80.4 ± 11.6	78.2 ± 9.6	80.2 ± 9.5
eGFR (mL/min/1.73 m²)	54.0 ± 18.0	64.6 ± 31.4	61.4 ± 15.6
Loop diuretic dose in FE (mg)	169.5 ± 95.5	86.7 ± 73.4	92.0 ± 55.1

CVP: central venous pressure; PCWP: pulmonary capillary wedge pressure; MAP: mean arterial pressure; eGFR: estimated glomerular filtration rate; FE: furosemide equivalents; LVAD: left ventricular assist device.

Values are mean ± SD. Urine sample 1 was drawn pre-LVAD; urine sample 2 post-LVAD while on stable doses of oral loop diuretics; urine sample 3 at an early outpatient appointment.

Diuretic response

Spot urine samples were collected in patients at three time points shown in Figure 1 and used to calculate FE_Na and 6-h sodium output. Measures of DR and changes in DR pre- and post-LVAD are reported in Table 3. Baseline CI demonstrated a negative correlation trend with urine sodium concentration (r = −0.34; p = 0.149), FE_Na (r = −0.42, p = 0.074), and cumulative urine sodium excretion (r = −0.37, p = 0.120), none of which reached statistical significance. eGFR also did not correlate with DR metrics at baseline (Figure 2). Neither baseline CVP nor PCWP correlated with any measure of DR (Supplemental Table 1). Post-LVAD CI and eGFR improved by 0.7 ± 0.6 L/min/m² and 5.7 ± 32.5 mL/min/1.73 m², respectively (Table 2). Despite improvements in CI and eGFR, LVAD implantation did not improve DR relative to baseline with a mean change FE_Na per 40 mg IV furosemide of −0.5 ± 1.0% at timepoint 2 (p = 0.84) and −0.3 ± 1.1% at timepoint 3 (p = 0.45). Across metrics of DR, most patients demonstrated a lack of change in DR post-LVAD even after adjusting for diuretic dose changes (Figure 3).

Table 3.

Serial measurements of diuretic response.

	Pre-LVAD urine sample 1	Post-LVAD urine sample 2	Post-LVAD urine sample 3
Spot urine sodium, mmol/L	74.8 ± 28.0	44.2 ± 17.7	58.0 ± 25.8
FE_Na, %	3.1 ± 2.7	1.1 ± 1.2	1.5 ± 1.0
6-h sodium output, mmol	58.0 ± 54.6	20.3 ± 16.1	33.2 ± 24.2
FE_Na/40 mg IV furosemide	1.2 ± 1.2	0.6 ± 0.4	0.9 ± 0.7
6-h sodium output/40 mg IV furosemide	23.0 ± 22.4	12.0 ± 9.3	19.8 ± 19.7

FE_Na: fractional excretion of sodium; IV: intravenous; LVAD: left ventricular assist device.

Values are mean ± SD. Urine sample 1 was drawn pre-LVAD; urine sample 2 post-LVAD while on stable doses of oral loop diuretics; urine sample 3 at an early outpatient appointment.

Figure 2.

Correlation of eGFR and CI with diuretic response.

Figure 3.

Change in diuretic response.

Post-LVAD outpatient outcomes

Outpatient study visit 3 occurred an average of 50.5 ± 30.9 days after LVAD implant. At this time, 90% of patients required scheduled maintenance loop diuretics and 1 (10%) patient required as needed dosing. Most patients were on torsemide (60%) with 30% on bumetanide and 10% on furosemide. The average home oral loop diuretic dose at study visit 3 was 92 ± 55 mg of FE. 50% of patients had right ventricular failure defined as requirement of right ventricular assist device support at any point post-implant and/or the requirement of continuous parenteral inotrope or inhaled selective pulmonary vasodilator >14 days post-implant. None of the cohort required readmission for HF, one patient necessitated hemodialysis while inpatient, and four patients had died by 90 days.

Discussion

To our knowledge, this is the first prospective study evaluating serial change in DR in stage D HF after LVAD implantation. The main findings of this study are: (1) DR, and therefore tubular sodium handling, did not correlate with eGFR at baseline and (2) DR did not improve in most patients following LVAD despite improvements in eGFR and cardiac index. DR remained low despite implantation of an LVAD with 90% of patients requiring scheduled loop diuretic therapy to prevent hypervolemia approximately 2 months after LVAD implantation.

Changes in renal function are common post-LVAD implant with some patients having stabilization or improvement in eGFR. Proteinuria and eGFR are metrics of renal filtration utilized for risk prognostication in LVAD recipients.^1
–3,7,9 Yet, these measures have poor correlation with DR as they are markers of renal filtration and fail to provide insight into tubular sodium handling, the predominant mechanism for sodium regulation and congestion in HF.^{3,4,16,23,29,30} At the same eGFR, a patient requiring twice as much daily diuretic to maintain euvolemia has different cardio-renal pathology and/or ingrained dietary habits that likely increase the risk of adverse post-LVAD outcomes relative to a patient with lower chronic diuretic requirements. Incorporation of DR into LVAD risk scores may provide additional risk discrimination for potential LVAD recipients.²⁹ These data can help to preemptively inform the degree to which patients may struggle with hypervolemia post-LVAD implantation. As DR metrics are inexpensive, validated, and easily obtained, the barriers to widespread implementation into clinical practice are predicted to be minimal.

Our study findings mimicked those of stage C HF where DR did not correlate with eGFR or hemodynamic measurements including CI, providing independent prognostic value.^9–13 Even at the severe hemodynamic derangements found in Stage D HF, tubular sodium handling and DR remained largely unaffected by baseline or post-LVAD improved cardiac filling pressures and cardiac output. These findings can help explain why more than half of patients require diuretic therapy to prevent congestion following LVAD implementation. Most HF hospitalizations result from symptoms of congestion, not low cardiac output.³¹ Therefore, utilizing DR pre-LVAD as an assessment of the patient’s likelihood of continued issues with hypervolemia post-LVAD may inform patient decisions as to how LVAD implantation will improve quality of life.

Future research is needed to investigate the clinical and prognostic importance of DR in a population with Stage D HF receiving LVAD implantation. Additionally, research on the changes in DR following LVAD implantation will enhance our understanding of the mechanisms of diuretic resistance in this population and the changes in cardio-renal interactions once cardiac output is improved. Knowledge of pre-existing DR and the driving mechanisms can inform clinicians on the diuretic regimens to employ in the pre-LVAD and post-LVAD care period to overcome resistance and potentially improve patient outcomes.

Limitations

Several limitations warrant discussion. The small cohort size, observational design, and intermediate follow-up time period limit definitive conclusions. Given the small number of patients with a high drop-out rate, the lack of correlations observed could be due to a type II error. These data should be regarded as hypothesis-generating. Second, this is a single center study evaluating diuretic resistance and DR may vary within the Stage D HF populations at other medical centers, thereby limiting generalizability. Lastly, there is no universal definition of DR or diuretic resistance. However, we used multiple, validated metrics of DR focusing on measured and calculated sodium excretion rather than urine output, as urine output measurements are significantly limited by measurement issues in usual care.

Conclusion

Our findings suggest that DR did not change commensurate with improvements in both eGFR and CI post-LVAD implantation. Larger studies are needed to understand the cardiorenal interactions in patients with Stage D HF undergoing LVAD implantation and DR’s prognostic value in post-LVAD outcomes.

Supplemental Material

sj-docx-1-jao-10.1177_03913988231193446 – Supplemental material for Serial evaluation of loop diuretic efficiency following left ventricular assist device implantation

Supplemental material, sj-docx-1-jao-10.1177_03913988231193446 for Serial evaluation of loop diuretic efficiency following left ventricular assist device implantation by Sarah M Beargie, Lindsey Tolbert, Robert K Tunney, Zachary L Cox, Wu Gong and Sandip Zalawadiya in The International Journal of Artificial Organs

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The project described was supported by the Vanderbilt Institute for Clinical and Translational Research (VICTR) and CTSA award No. UL1 TR002243 from the National Center for Advancing Translational Sciences. Its contents are solely the responsibility of the authors and do not necessarily represent official views of the National Center for Advancing Translational Sciences or the National Institutes of Health.

ORCID iD

Sarah M Beargie

Supplemental material

Supplemental material for this article is available online.

References

Cowger

Sundareswaran

Rogers

, et al. Predicting survival in patients receiving continuous flow left ventricular assist devices: the HeartMate II risk score. J Am Coll Cardiol 2013; 61(3): 313–321.

Stewart

. Learning from our predictions: the HeartMate risk score in INTERMACS*. JACC Heart Fail 2016; 4(12): 959–961.

Suzuki

Yoshihisa

Sato

, et al. Clinical significance of get with the guidelines-heart failure risk score in patients with chronic heart failure after hospitalization. J Am Heart Assoc 2018; 7(17): e008316.

Topkara

Coromilas

Garan

, et al. Preoperative proteinuria and reduced glomerular filtration rate predicts renal replacement therapy in patients supported with continuous-flow left ventricular assist devices. Circ Heart Fail 2016; 9(12): e002897.

Adamo

Tang

Nassif

, et al. The HeartMate Risk score identifies patients with similar mortality risk across all INTERMACS profiles in a large multicenter analysis. JACC Heart Fail 2016; 4(12): 950–958.

Ross

Stevens

Wanchoo

, et al. Left ventricular assist devices and the kidney. Clin J Am Soc Nephrol 2018; 13(2): 348–355.

Kido

George

Charnigo

, et al. Chronologic changes and correlates of loop diuretic dose in patients with left ventricular assist device. ASAIO J 2017; 63(6): 774–780.

Kirklin

Naftel

Kormos

, et al. Quantifying the effect of cardiorenal syndrome on mortality after left ventricular assist device implant. J Heart Lung Transplant 2013; 32(12): 1205–1213.

Testani

Brisco

Turner

, et al. Loop diuretic efficiency a metric of diuretic responsiveness with prognostic importance in acute decompensated heart failure. Circ Heart Fail 2014; 7(2): 261–270.

10.

ter Maaten

Valente

MAE

Damman

, et al. Combining diuretic response and hemoconcentration to predict rehospitalization after admission for acute heart failure. Circ Heart Fail 2016; 9(6): e002845.

11.

Valente

MAE

Voors

Damman

, et al. Diuretic response in acute heart failure: clinical characteristics and prognostic significance. Eur Heart J. 2014; 35(19): 1284–1293.

12.

ter Maaten

Dunning

E Valente

, et al. Diuretic response in acute heart failure-an analysis from ASCEND-HF. Am Heart J 2015; 170(2): 313–321.

13.

Voors

Davison

Teerlink

, et al. Diuretic response in patients with acute decompensated heart failure: characteristics and clinical outcome–an analysis from RELAX-AHF. Eur J Heart Fail 2014; 16(11): 1230–1240.

14.

Mullens

Damman

Harjola

, et al. The use of diuretics in heart failure with congestion — a position statement from the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail 2019; 21(2): 137–155.

15.

Testani

Hanberg

Cheng

, et al. Rapid and highly accurate prediction of poor loop diuretic natriuretic response in patients with heart failure. Circ Heart Fail 2016; 9(1): e002370.

16.

Nohria

Hasselblad

Stebbins

, et al. Cardiorenal interactions: insights from the ESCAPE trial. J Am Coll Cardiol 2008; 51(13): 1268–1274.

17.

Hodson

Griffin

Mahoney

, et al. Natriuretic response is highly variable and associated with 6-month survival insights from the ROSE-AHF trial. JACC Heart Fail 2019; 7(5): 383–391.

18.

Cox

Lenihan

. Loop diuretic resistance in heart failure: resistance etiology-based strategies to restoring diuretic efficacy. J Card Fail 2014; 20(8): 611–622.

19.

Harris

Taylor

Thielke

, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009; 42(2): 377–381.

20.

Harris

Taylor

Minor

, et al. The REDCap consortium: building an international community of software platform partners. J Biomed Inform 2019; 95: 103208.

21.

Rao

Ivey-Miranda

Cox

, et al. Natriuretic equation to predict loop diuretic response in patients with heart failure. J Am Coll Cardiol 2021; 77(6): 695–708.

22.

Testani

. Diuretic response calculator, http://www.cardiorenalresearch.net/ (accessed 9 February 2020).

23.

ter Maaten

Rao

Hanberg

, et al. Renal tubular resistance is the primary driver for loop diuretic resistance in acute heart failure. Eur J Heart Fail 2017; 19: 1014–1022.

24.

Brater

Day

Burdette

, et al. Bumetanide and furosemide in heart failure. Kidney Int 1984; 26(2): 183–189.

25.

Vargo

Kramer

Black

, et al. Bioavailability, pharmacokinetics, and pharmacodynamics of torsemide and furosemide in patients with congestive heart failure. Clin Pharmacol Ther 1995; 57(6): 601–609.

26.

R Core Team. R: a language and environment for statistical computing, https://www.r-project.org/ (2014, accessed 25 July 2022).

27.

Wickham

. Ggplot2: elegant graphics for data analysis. New York, NY: Springer-Verlag, 2009.

28.

IBM Corporation. IBM SPSS Statistics for Windows. 2021.

29.

Daimee

Wang

Papernov

, et al. Renal function changes following left ventricular assist device implantation. Am J Cardiol 2017; 120(12): 2213–2220.

30.

Gottlieb

Stebbins

Voors

, et al. Effects of nesiritide and predictors of urine output in acute decompensated heart failure results from ASCEND-HF (acute study of clinical effectiveness of nesiritide and decompensated heart failure). J Am Coll Cardiol 2013; 62(13): 1177–1183.

31.

Hollenberg

Lynne Warner Stevenson

Chair Tariq Ahmad

, et al. Expert consensus decision pathway 2019 ACC expert consensus decision pathway on risk assessment, management, and clinical trajectory of patients hospitalized with heart failure a report of the American College of Cardiology Solution Set Oversight Committee Writing Committee. J Am Coll Cardiol 2019; 74(15): 1966–2011.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB