Renal Negative Pressure Treatment As A Novel Therapy For Heart Failure-induced Renal Dysfunction

For more information Contact Tina tina.xiang@wecistanche.com

ABSTRACT

Congestion is the primary pathophysiological lesion in most heart failure (HF) hospitalizations. Renal congestion increases renal tubular pressure, reducing glomerular filtration rate (GFR) and diuresis. Because each nephron is a fluid-filled column, renal negative pressure therapy (rNPT) applied to the urinary collecting system should reduce tubular pressure, potentially improving kidney function. We evaluated the renal response to rNPT in congestive HF. Ten anesthetized 80-kg pigs underwent instrumentation with bilateral renal pelvic JuxtaFlow catheters. GFR was determined by iothalamate clearance (mGFR) and renal plasma flow (RPF) by para-aminohippurate clearance. Each animal served as its own control with randomization of left versus right kidney to 30 mmHg rNPT or no rNPT. mGFR and RPF were measured simultaneously from the rNPT and no rNPT kidney. Congestive HF was induced via cardiac tamponade maintaining central venous pressure at 20–22.5 mmHg throughout the experiment. Before HF induction, rNPT increased natriuresis, diuresis, and mGFR compared with the control kidney (P < 0.001for all). Natriuresis, diuresis, and mGFR decreased following HF (P < 0.001 for all) but were higher in rNPT kidney versus control (P < 0.001 for all). RPF decreased during HF (P < 0.001) without significant differences between rNPT treatments. During HF, the rNPT kidney had similar diuresis and natriuresis (P > 0.5 for both) and higher fractional excretion of sodium (P = 0.001) compared with the non-rNPT kidney in the no HF period. In conclusion, rNPT resulted in significantly increased diuresis, natriuresis,

and mGFR, with or without experimental HF. rNPT improved key renal parameters of the congested cardiorenal phenotype.

cardiorenal syndrome; diuretic; heart failure; renal negative pressure

Click for more information on improving kidney function

INTRODUCTION

Affecting over 1 million Americans annually, acute decompensated heart failure(ADHF) is the most common hospital discharge diagnosis among Medicare beneficiaries and accounts for more than half of all heart failure (HF)-related expenditures (1-3).On a population level, ADHF is primarily a disease of congestion rather than low cardiac output(4-7). In addition to causing classic HF symptoms, congestion can lead to organ dysfunction with notable adverse effects on the kidneys. In experiments dating back to the 1860s, it was shown that partial occlusion of the renal vein led to a prompt decline in renal blood flow, glomerular filtration rate (GFR), and sodium excretion, with a resolution of these abnormalities after relief of the congestion (8, 9). Since the kidneys are encapsulated and thus can only minimally expand, congestion increases intrarenal and tubular pressures with an associated worsening of GFR and sodium avidity(10-15). Renal venous congestion also diminishes diuretic response, creating a vicious cycle where congestion impedes treatment response and intensifies renal sodium avidity(13.14.16). Unfortunately, loop diuretic therapy also exacerbates aberrations in intratubular pressure, potentially intensifying this process (17). Therefore, strategies to reduce intratubular pressure and arrest this positive-feedback cycle warrant investigation. Because each nephron is a fluid-filled column connecting the urinary collecting system to the glomerulus, the negative pressure applied to the urine collection system might reduce intratubular pressure throughout the nephron. We sought to investigate the effects of renal negative pressure therapy (rNPT) on natriuresis and kidney function during experimental heart failure.

METHODS

The animal study was conducted in accordance with approval by the Yale University Institutional Animal Care and Use Committee. 

Animal Experiments

Ten Yorkshire farm female pigs 18-20wk of age(~80 kg)were used to investigate the effects of rNPT from the JuxtaFlow catheter system (3ive Labs, Roswell, GA). The JuxtaFlow catheter is a memory polymer catheter that expands into a three-dimensional helix when deployed in the renal pelvis allowing application of negative pressure(-30 mmHg)without causing collapse(Supplemental Figure; see https://doi. org/10.6084/m9.figshare.14503062.v1).

After an overnight fast, pigs were anesthetized with a combination of intramuscular ketamine and tiletamine-zolazepam (Telazol), intubated, and maintained on inhaled isoflurane. An intrapericardial catheter was placed via a left lateral thoracotomy, a Swan-Ganz catheter was placed via a right internal jugular vein cut down, and an arterial line for continuous hemodynamic monitoring was placed in the carotid or femoral artery by either the Seldinger technique or an arterial cutdown. Large bore central ve-nous access was similarly placed in either the contralateral jugular or a femoral vein for fluid and tracer infusions. To catheterize the ureters, the bladder was retracted caudally through a small suprapubic incision and each ureter was isolated and directly cannulated through a small incision. The JuxtaFlow catheters were advanced into the renal pelvis under fluoroscopic guidance. They were either drained passively at atmospheric pressure or attached to a vacuum-assisted negative pres-sure pump (-30 mmHg).

Since loop diuretics are used almost universally in patients during the treatment of decompensated heart failure, experiments were undertaken during ongoing furosemide infusion. The effects of rNPT were studied during maximal furosemide diuresis in the absence of HF and in a state of HF characterized by venous congestion and pre-served cardiac output. During the two phases, each animal served as its own control with randomization of either the left or right kidney to rNPT versus no rNPT. The animal received intravenous bolus and continuous infusion of iothalamate(IOT,120 mg bolus with 0.3 mg/min infusion, Guerbet), para-aminohippurate (PAH 800mg bolus with 8.4 mg/min infusion, Millipore Sigma, St.Louis, MO), and furosemide (400 mg bolus with an infusion at 80 mg/h) that were maintained for 2.5h. A 4-L intravenous infusion of normal saline was administered at baseline, followed by a maintenance intravenous infusion matched 1:1 to urine output. The 4-L infusion was utilized 1) to increase urine output facilitating the timed urine collection and 2) because, during the development of this model of acute heart failure, this amount of fluid led to relative stability in filling pressures over time (18). Blinding of rNPT was not feasible during these experiments given that a specific device was used to perform renal negative pressure. We did not specify inclusion and exclusion criteria for animal models either before or after the study, as the experiments were designed for healthy animals.

Experimental Periods

After equilibration of tracers (~2.5h), the right and left kidneys were randomized to-30mmHgrNPT or to drainage at atmospheric pressure. Measurements of RPF and mGFR were taken simultaneously from the rNPT kidney and the control contralateral kidney (no rNPT). To ensure that the tamponade (no HF) and tamponade(HF)experimental periods had similar background fluid status, a rapid large volume normal saline infusion of 20%-25% body weight occurred at this point. After a 10-min equilibration, the ani-mals underwent two 15-min"post fluid" clearance periods. Next, cardiac tamponade was induced by pericardial installation of~200 mL of 6% hydroxyethyl starch (18). Pericardial hydroxyethyl starch and additional intravenous normal saline infusion were titrated to maintain a hemodynamic pro-file relative to baseline pre fluid readings, with preservation of cardiac output and mean arterial pressure and goal central venous pressure at 20-22.5 mmHg throughout the experiment. After stabilization and a 10-min equilibration, two 15-min study periods were repeated.

Assays and Calculations

A Randox Imola automated clinical chemistry analyzer was used to measure urine or serum chemistry parameters (Randox Laboratories, Crumlin, UK). Creatinine measurements were standardized to IDMS traceable National Institute of Standards and Technology reference material (SRM 967). Urine and plasma iothalamate were measured on an Agilent 6490 QTOF equipped with Agilent 1290 UHPLC. Briefly, a stock solution of iothalamate was serially diluted in 0.1% formic acid containing deuterated iothalamate to create the calibration curve(1-2,000 ng/mL). Plasma samples （100 μL） were deproteinized by adding 300 μL of 100%methanol containing deuterated iothalamate(1,000 ng/mL; Cambridge Isotope Laboratories, Inc., Tewksbury, MA), vortexed, and centrifuged at 12,000rpm for 10min. Then.200 μL， of the supernatant, were transferred to glass vials，and10μLof the sample was injected into UHPLC-MS/MS system. Urine samples were diluted 10-fold with 0.1%formic acid containing the internal standard， and 10 μL of the sample were injected. Separation was achieved using AgilentZorbax Eclipse plus RP2.1×50-mm 1.8-μm column with a constant flow rate of 400 μL/min and an instrument-controlled gradient. Also, 0.1% formic acid and 100% methanol were used as Buffer A and B, respectively. Quantitation was achieved using the Agilent MassHunter Quantitative analysis software. Urine and plasma PAH were measured using PAH colorimetric assay kits from Abcam(Cambridge, UK). Urine neutrophil gelatinase-associated lipocalin (NGAL) was measured with a porcine NGALkit from Alpco(Alpco, Salem, NH). Urine cGMP concentrations were measured by competitive enzyme-linked immunosorbent assay according to the manufacturer's guidelines (Parameter cGMP Assay, R&D Systems Inc, Minneapolis, MN).

Creatinine clearance was calculated as urine creatinine ×volume of urine per minute/plasma creatinine. Measured GFR was calculated as urine iothalamate× volume of urine per minute/plasma iothalamate. Renal plasma flow was calculated as (urine PAH× volume of urine per minute/plasma PAH)/0.9;0.9 is the adjustment for the 90% extraction of PAH. Filtration fraction was calculated as GFR/renal plasma flow(19,20). Fractional excretion of sodium (FENa)was calculated as (Naurne/Naerum)×(Crerum/Crurine)×100%. Statistical Analysis

Continuous data are shown as means± standard deviation or median (quartile 1-quartile 3) according to observed distribution. Categorical data are shown as frequency (percentage). Variables with skewed distribution were log-transformed to approximate normal distribution. Changes in systolic blood pressure, mean arterial pressure, cardiac output, central venous pressure, pulmonary capillary wedge pressure, heart rate, NGAL, and cGMP from baseline to post fluid (no HF)or to HF model of venous congestion were analyzed using linear mixed models accounting for correlations within animals; time was included as the main factor with three levels, and com-parisons were undertaken for each pair of time points. Likewise, changes in urine output, urine sodium output, FENa, renal plasma flow, iothalamate clearance, and filtration fraction during the experiments were analyzed by linear mixed models accounting for correlations within animals. rNPT and HF model of venous congestion were included as main factors (binary variables)in a full factorial model. Then, comparisons were undertaken for each pair, i.e."no rNPT

versus"rNPT"in each HF period ("no HF period"and"HIF period"). In addition, the effect of the HF model on venous congestion was analyzed by comparing the"no HF" measurements versus the"HF" measurements. Statistical significance was defined as two-tailed P <0.05. Statistical analysis was performed with IBM SPSS Statistics version 26 (IBM Corp, Armonk, NY) and Stata SE version 16.0 (StataCorp, College Station, TX).

RESULTS

Induction of cardiac tamponade was successful in producing a"warm and wet"HF phenotype with preserved cardiac output and blood pressure but severely elevated right-sided filling pressures (Fig.1).A cardiorenal phenotype also emerged,as evidenced by substantial decreases in urine output(37%), renal sodium excretion(40%), measured GFR (27%),and renal plasma flow(50%)(P<0.001 for all),leading to an increase in filtration fraction (42±18% ys.56±21%, P=0.001). Furthermore, urine cGMP decreased substantially (P< 0.001, Fig.1)and NGAL(a marker of tubular injury)tended to increase with induction of HF, but this did not reach significance (P= 0.14; Fig.1).

During furosemide diuresis, the rNPT kidney substantially increased natriuresis (2.4±0.6mmol/min vs.1.5±0.5mmol/min;P<0.001)and diuresis(19.7±4.5 mL/min vs.11.8±3.7 mL/min; P< 0.001) compared with the control kidney (Fig. 2).rNPT increased iothalamate clearance(79±28 mL/min vs. 62±23 mL/min;P<0.001, Fig.2) and creatinine clearance (105±38 mL/min vs.85±30 mL/min,P= 0.001). Renal plasma flow(P=0.13) did not differ significantly between rNPT and control(Fig. 2). The increased natriuresis with rNPT was not solely driven byincreased sodium filtration,as the FENa also was higher with rNPT(15.9± 3.3% ys.12.0±4.2%,P<0.001).

The beneficial effect of rNPT on GFR was similar in the HF and no HF periods(i.e., similar increments in GFR with rNPT; Pinteraction=0.23 for a different effect of rNPT in HF and no HF). rNPT did not significantly change renal plasma flow in either the HF or the no HF period (P=0.47 for the interaction).During HF,rNPT yielded greater urine output(276±113 mL vs.167±55 mL,P<0.001)and urine sodium excretion(33.0±14.5mmolvs.19.5±6.8mmol;P<0.001)compared with the control kidney (Fig. 2). FENa was also higher with rNPT(14.5±3.0%vs.10.9±2.7%,P<0.001). Renal plasma flow did not change significantly with rNPT in HF(P= 0.58). Filtration fraction increased during rNPT(55.6±24.8% vs.49.0±20.8%;P=0.034)with a similar effect between HF and no HF periods(P=0.70 for the interaction). Urine NGAL was not influenced by rNPT both during the no HF and HF periods (P= 0.37). Urine cGMP increased significantly with rNPT(P= 0.015), which was similar between the no HF and HF periods(P interaction=0.25). Importantly,during HF, the rNPT kid-ney had similar urine output (P= 0.52) and sodium excretion(P=0.87) yet higher FENa(14.5±3.0% vs.12.0±4.2%;P= 0.001) compared with the non-rNPT kidney in the no HF period.

DISCUSSION

The primary observation in this study is that negative pressure applied to the renal pelvis significantly improves a wide range of renal parameters. During high-dose furosemide therapy, rNPT increased GFR, urine output, and sodium output. The increase in cumulative urine sodium excretion could not be ascribed solely to an increase in GFR since the FENa also increased. Importantly, the benefit appeared to be of potential clinical significance since urine output and sodium excretion during experimental HF with rNPT were similar to the non-rNPT kidney during the no HF period.

In ADHF, elevated central venous pressure is transmitted to the renal veins and tubules in the fixed space of the encapsulated kidney (13,15,16,21). Alterations in renal venous blood flow patterns normalize with decongestion, suggesting that interventions to reduce intrarenal pressure may improve diuretic response and thereby ADHF outcomes(15). Indeed, our studies are the first, to our knowledge, to demonstrate that a strategy designed to reduce intrarenal pressure can increase urine output and sodium excretion. These findings are consistent with observations in humans with ADHF; abnormal measurements of renal venous impedance and flow have been associated with higher sodium avidity, diminished diuretic response, and worse HF outcomes independent of central venous pressure(14,16,21-23). In patients with ADHF, the inverse relationship between diuretic response and elevated renal venous impedance is independent of GFR (16). Similarly, we found that improved natriuresis with renal negative pressure therapy was not driven only by improvement in GFR since FENa improved with rNPT.

Furosemide has been reported to induce several detrimental effects on the kidneys. It reduces GFR by up to 25%, although the mechanism is complex (24,25).In addition, furosemide causes an abrupt increase in proximal tubular pressure that should reduce the net ultrafiltration pressure and a rise in interstitial pressure that should obstruct capillary blood flow (17). Renal negative pressure treatment improved natriuresis and GFR during high-dose furosemide infusion, even in the absence of HF. This is an interesting finding since increased filtration should activate tubuloglomerular feedback (TGF), thereby reducing filtration and bringing GFR back to baseline. Thus, favorable effects of rNPT on GFR and natriuresis with loop diuretics could prove to be a significant benefit for many patients with worsening renal function and diuretic resistance. Given that rNPT improved natriuresis and GFR in the absence of HF, it is likely that rNPT may be effective in other pathological situations that share the phenotype of venous congestion and/or fluid overload, such as renal or liver disease. However, we note that rNPT did not improve renal plasma flow, a feature that would have been desirable in HF because reduced renal plasma flow is also a characteristic of severely congested kidneys. In addition, the impact of rNPI on congestion could not be measured since congestion was an independent variable in the study, with central venous pressure in the HF condition maintained constant at 20 mmHg by titration of fluids and intrapericardial pressure. Therefore, and by design, hemodynamics were identical during rNPT versus control therapy as each kidney served as its own control. As such, the current study demonstrates the effect of rNPT on renal parameters in the setting of congestive HF, and further studies are needed to assess the effect of rNPT on congestion itself since decongestion is the ultimate goal of diuresis in congestive HF.

The current observations from our congestion predominant HF model shed light on the human literature on kidney dysfunction in human HF. The majority of contemporary human studies have not found a meaningful association between cardiac output and kidney function(14,26-29). Several human studies have noted an association between central venous pressure(CVP) and renal function(30,31). In the current experiments, with the severe elevation of CVP induced by cardiac tamponade, we observed a substantial reduction in natriuresis, renal plasma flow, GFR, and urinary cGMP. This was despite a cardiac output and blood pressure that were overall similar to baseline values. These observations reinforce the human observational literature that has found congestion to be a more important driver of cardiorenal interactions than poor forward flow. Therefore, the effect of rNPT in the treatment of hypervolemia in patients with ADHF will be investigated, and a first-in-human study is currently ongoing (NCTO4227977).In humans, rNPT will be carried out through ureteral catheters deployed endoscopically to deliver mild controlled negative pressure into the renal pelvis.

We acknowledge several limitations. First, although the acute cardiac tamponade model provided a stable, predictable, and titratable"warm and wet"HF phenotype, acute tamponade is a rare human HF presentation, and thus, findings may not extrapolate to acute or chronic decompensated human HF. Second, we did not study the effect of rNPTin the absence of diuretics; in addition, even in the setting of HF, our model did not demonstrate poor diuretic response, and thus, we cannot demonstrate that rNPT addresses diuretic resistance. Third, although the presumed mechanism underlying the improved renal function with rNPT is the reduction of intratubular and interstitial pressure, this was not directly measured. Fourth, experiments were undertaken in animals anesthetized with isoflurane, which is known to increase renal sympathetic nerve activity and thus could have influenced the observed responses. Fifth, although it is not likely that reducing cardiac output from supranormal to normal values (no HF to HF period) was a driving factor, we cannot negate some potential effect on the observed results. Finally, although utilizing the JuxtaFlow catheter for both negative and atmospheric pressure provided control for any mechanical effects of instrumenting the renal pelvis, we were unable to measure delivered pressure at the level of the renal pelvis through this single-lumen catheter. Thus, it cannot be determined if the actually delivered pres-sure deviated from -30mmHg in the rNPT group and OmmHg in the non-rNPT group. As such, these proof-of-concept results should be viewed as hypothesis-generating and motivate future studies in patients with HF. 

Perspectives and Significance

Initiation of rNPT during high-dose loop diuretic infusion increased diuresis, natriuresis, and mGFR. Importantly, the benefits were of potential clinical significance, as urine output and sodium excretion during HF and rNPT were restored to levels of control kidney. Additional research into the efficacy of the JuxtaFlow system in human decompensated HF is warranted.

Veena S. Rao,

Veena S. Rao,1 Christopher Mauleon,1 Jennifer L. Asher,2 Juan B. Ivey-Miranda,1,3 Zachary L. Cox,4,5Julieta Moreno-Villagomez,1,6 Devin Mahoney,1 Jeffrey M. Turner,7 F. Perry Wilson,8 Christopher S. Wilcox,9and Jeffrey M. Testani1 

1Department of Internal Medicine, Section of Cardiovascular Medicine, Yale University School of Medicine, New Haven, Connecticut; 

2Department of Comparative Medicine, Yale University School of Medicine, New Haven, Connecticut; 

3Hospitalde Cardiologia, Instituto Mexicano del Seguro Social, Mexico City, Mexico; 

4Department of Pharmacy Practice, LipscombUniversity College of Pharmacy, Nashville, Tennessee; 

5Department of Pharmacy, Vanderbilt University Medical Center, Nashville, Tennessee; 

6Facultad de Estudios Superiores Iztacala, Universidad Nacional Autonoma de Mexico, Mexico City,Mexico; 

7Department of Medicine, Division of Nephrology, Yale University School of Medicine, New Haven, Connecticut; 

8Clinical and translational research accelerator, Yale University School of Medicine, New Haven, Connecticut; 

9Divisionof Nephrology and Hypertension Center, Georgetown University, Washington, District of Columbia

REFERENCES

1. Bradley SM, Levy WC, Veenstra DL. Cost-consequences of ultrafiltration for acute heart failure: a decision model analysis. Circ Cardiovasc Qual Outcomes
2: 566–573, 2009. doi:10.1161/ CIRCOUTCOMES.109.853556. 2. Fonarow GC, Corday E; ADHERE Scientific Advisory Committee. Overview of acutely decompensated congestive heart failure (ADHF): a report from the ADHERE registry. Heart Fail Rev 9: 179–185, 2004. doi:10.1007/s10741-005-6127-6.
3. Giamouzis G, Kalogeropoulos A, Georgiopoulou V, Laskar S, Smith AL, Dunbar S, Triposkiadis F, Butler J. Hospitalization epidemic in patients with heart failure: risk factors, risk prediction, knowledge gaps, and future directions. J Card Fail 17: 54–75, 2011. doi:10.1016/j.cardfail.2010.08.010.
4. Gheorghiade M, Filippatos G, De Luca L, Burnett J. Congestion in acute heart failure syndromes: an essential target of evaluation and treatment. Am J Med 119: S3–S10, 2006. doi:10.1016/j.amjmed.2006. 09.011.
5. Gheorghiade M, Follath F, Ponikowski P, Barsuk JH, Blair JE, Cleland JG, Dickstein K, Drazner MH, Fonarow GC, Jaarsma T, Jondeau G, Sendon JL, Mebazaa A, Metra M, Nieminen M, Pang PS, Seferovic P, Stevenson LW, van Veldhuisen DJ, Zannad F, Anker SD, Rhodes A, McMurray JJ, Filippatos G; European Society of Cardiology; European Society of Intensive Care Medicine. Assessing and grading congestion in acute heart failure: a scientific statement from the acute heart failure committee of the heart failure association of the European Society of Cardiology and endorsed by the European Society of Intensive Care Medicine. Eur J Heart Fail 12: 423–433, 2010. doi:10.1093/eurjhf/hfq045.
6. Gheorghiade M, Pang PS. Acute heart failure syndromes. J Am Coll Cardiol 53: 557–573, 2009. doi:10.1016/j.jacc.2008.10.041.
7. Gheorghiade M, Vaduganathan M, Fonarow GC, Bonow RO. Rehospitalization for heart failure: problems and perspectives. J Am Coll Cardiol 61: 391–403, 2013. doi:10.1016/j.jacc.2012.09.038.
8. Ludwig C. Physiologi des Menschen, Leipzig and Heidelberg: Wintersche Verlagshanglung. 1861.
9. Joseph FG, von Haehling S, Anker SD, Raj DS, Radhakrishnan J. The relevance of congestion in the cardio-renal syndrome. Kidney Int 83: 384–391, 2013. doi:10.1038/ki.2012.406.
10. Cruces P, Salas C, Lillo P, Salomon T, Lillo F, Hurtado DE. The renal compartment: a hydraulic view. Intensive Care Med Exp 2: 26, 2014. doi:10.1186/s40635-014-0026-x.
11. Dilley JR, Corradi A, Arendshorst WJ. Glomerular ultrafiltration dynamics during increased renal venous pressure. Am J Physiol 244: F650–F658, 1983. doi:10.1152/ajprenal.1983.244.6.F650.
12. Firth JD, Raine AE, Ledingham JG. Raised venous pressure: a direct cause of renal sodium retention in edema? Lancet 1: 1033–1035, 1988. doi:10.1016/S0140-6736(88)91851-X.
13. Husain-Syed F, Gr€ one HJ, Assmus B, Bauer P, Gall H, Seeger W, Ghofrani A, Ronco C, Birk HW. Congestive nephropathy: a neglected entity? Proposal for diagnostic criteria and future perspectives. ESC Heart Fail 8: 183–203, 2021. doi:10.1002/ ehf2.13118.