Dynamic Changes in Portal Vein Flow during Decongestion in Patients with Heart Failure and Cardio-Renal Syndrome

Introduction

Hospitalizations from heart failure have been shown to be preceded by a gradual increase in cardiac filling pressures using invasive ambulatory monitoring measurements [1^, 2]. Noninvasive methods for assessment of congestion are still being developed. While bioelectrical vectorial impedance analysis has been proposed as an aid to evaluate body hydration status [3], this technique does not directly assess congestion. The use of point-of-care ultrasound (POCUS) to evaluate lung congestion has been shown to decrease decompensations and urgent heart failure visits [4^, 5]. However, besides lung congestion, the abdominal compartment contributes significantly to deranged cardiac as well as renal function in congestive heart failure (CHF) [6].

There is increasing recognition that worsening renal function in CHF is related to altered renal blood flow [7^, 8]. Recently, alterations in renal venous flow (IRVF) assessed by Doppler imaging have been associated with worse outcomes in patients with CHF [9^, 10]. Besides IRVF alterations, portal vein flow (PVF) alterations have been proposed as a marker of venous congestion and right ventricular dysfunction [11^, 12]. PVF and IRVF alterations have been shown to correlate with each other and were independently associated with the development of subsequent acute kidney injury (AKI) in patients undergoing cardiac surgery [13^, 14]. Portal vein (PV) Doppler is easily obtainable and less time consuming than intrarenal venous Doppler [13] Given the potential usefulness of evaluating venous congestion via POCUS of PVF, we decided to study the dynamic changes that occur during decongestion in patients presenting with heart failure to the emergency department.

Materials and Methods

Results

A total of 12 patients were included (Table 1). Six patients had left-sided heart failure (heart failure with reduced ejection fraction and heart failure with preserved ejection fraction) (patients 1–6), while 6 had right-sided heart failure (pulmonary arterial hypertension and constrictive pericarditis who were not surgical candidates) (patients 7–12). At the time of diagnosis, all patients had a plethoric IVC and a PV PF >30%. After diuretic treatment, all patients were able to be decongested as evidenced by clinical improvement, negative cumulative fluid balance (mean ± SD) (−8.2 ± 4 L), and a decrease in brain natriuretic peptide (2,085 ± 2,359 pg/mL before decongestion vs. 606 ± 474 pg/mL after decongestion p = 0.032). Three patients had >2 POCUS evaluations and showed progressive improvements in portal flow alterations with decongestion (Fig. 1). All patients showed improvement in PV PF (81.75 ± 46% before decongestion vs. 17.43 ± 8% after decongestion p = 0.0003) (Fig. 2a). While most patients with left-sided heart failure had a nonplethoric IVC after treatment (66.6%), this was not achieved in any of the patients with right-sided heart failure (Fig. 2b). A total of 8 patients (patients 3–10) presented with congestive AKI (Serum Cr [SCr] 3.42 ± 1.94 mg/dL at presentation), and it resolved in all of them after decongestion (SCr 1.87 ± 0.92 mg/dL). Only 6 patients had available urinalysis. None showed increased proteinuria or hematuria. Urine sediment was bland in all cases. Urinary sodium was <20 mEq/L in 4 patients, and >30 mEq/L in 2 patients with recent administration of IV loop diuretic. No other cause of AKI was identified (Table 2).

Open multimedia modal

Open multimedia modal

Open multimedia modal

Open multimedia modal

Discussion

In this study, we present a case series of patients with acute decompensated heart failure and point-of-care ultrasound evidence of abdominal venous congestion evidenced by a mean PV PF of 81.75%. We saw a high incidence of congestive AKI (67%) likely secondary to selection: patients with ADHF and AKI might be more likely to trigger an advanced POCUS consult from ED physicians than patients with ADHF without AKI. The fact that only patients with severe venous congestion were included could have also played a role. Given exclusion of patients with intrinsic AKI, all patients were able to be decongested with diuretic treatment. Decongestion was accompanied by an improvement in PV PF in all patients. A nonplethoric IVC was achieved in most patients with left-sided heart failure but not in those with right-sided heart failure. While previous studies have suggested that a false positive increase in PV PF can be present in healthy individuals with low BMI [15], we aimed to increase the specificity of this exam by adding the evaluation of the IVC to the study protocol. In fact, a previous study has shown that multiple-site evaluation of venous congestion including IVC, hepatic, portal, and intrarenal veins increases the specificity of this assessment [14]. The fact that altered PVF improved considerably with diuretics in association with other signs of decongestion provides evidence that this assessment represents abdominal venous congestion in this patient population.

Previous studies of abnormal venous flow patterns in heart failure have been largely associative, and currently, there is little evidence to demonstrate the reversibility of such altered patterns with any therapeutic strategy [8]. In this series, PVF alterations were completely reversed by decongestion in all patients and PVF normalization coincided with return to baseline of SCr in those with congestive AKI. While there is considerable literature regarding the use of intrarenal venous Doppler in the assessment of renal-congestion [9^, 10^, 16^, 17], PVF and IRVF alterations have been shown to correlate with each other, and PVF is easier to obtain and less time consuming than intrarenal venous Doppler, suggesting that PVF assessment might be more realistically integrated into patient care [13].

Evaluation of PV pulsatility in heart failure might add valuable information to currently established POCUS protocols like IVC and lung ultrasound (LUS). While evaluation of IVC has been proposed as a way to assess congestion in patients with heart failure [18^, 19], the results from this series suggest that PV PF may add additional information about the degree of venous congestion that could be especially valuable when assessing patients with right-sided heart failure. LUS detection of lung congestion has become an established part of the POCUS assessment of congestion, incorporation of LUS has been shown to decrease exacerbations and urgent heart failure visits [4^, 5]. However, LUS cannot accurately distinguish cardiogenic from noncardiogenic pulmonary edema, especially in the setting of primary interstitial lung pathology [20]. Also, LUS does not evaluate abdominal venous congestion directly which is especially relevant for patients with right-sided heart failure. Given that the prevailing pathophysiology of worsening kidney injury in patients with CHF relates to increased CVP and renal venous pressure [7], assessment of PV pulsatility could represent a new bedside tool that allows the clinician to monitor abdominal congestion in real time during diagnosis and treatment.

The limitations of the present study include the lack of prospective screening of all heart failure patients presenting to the emergency department. Also, patients were only included if they had both a plethoric IVC and a PV PF >30%. As such, the findings of this study might not apply to patients with ADHF who present with less severe congestion or who do not have both of these POCUS criteria despite severe clinical congestion.

Conclusions

Evaluation of abdominal venous congestion by POCUS could aid in diagnosis and follow-up of patients with congestive kidney injury. A plethoric IVC and an elevated PV PF in the setting of AKI should alert the clinician to the presence of venous congestion with the potential for end-organ injury. Further prospective studies are needed to evaluate the accuracy and clinical utility of this novel bedside tool.

Statement of Ethics

The Institutional Review Board approved this study with waiver of patient consent (NMM-3393).

Conflict of Interest Statement

The authors declare no conflicts of interest.

Funding Sources

The authors declare no specific funding for this study.

Author Contributions

E.R.A., P.R., and G.G. designed the study and wrote the manuscript. E.R.A. performed point-of-care ultrasound evaluations and collected pertinent patient data.