Point of Care Ultrasound in Cirrhosis-Associated Acute Kidney Injury: Beyond Inferior Vena Cava

AKI in patients with cirrhosis poses a diagnostic challenge because of a multitude of etiologic possibilities with overlapping presentations. The shortcomings of conventional physical examination are well recognized, especially in clinical scenarios associated with derangements in fluid status. Even laboratory data are insufficient to discern certain etiologies. For example, low urine sodium can be observed in volume depletion, congestive nephropathy, iodinated contrast exposure, intra-abdominal hypertension (IAH), hepatorenal syndrome (HRS), and some cases of tubular injury. Although careful history taking helps to some extent, physicians often rely on their clinical judgment to guide management in these cases, which is influenced by experience, pattern recognition skills, and internal biases. Further, current blanket recommendation of plasma volume expansion with albumin infusion (1 g/kg body weight per day) in patients with cirrhosis presenting with AKI adds to the problem, potentially leading to iatrogenic fluid overload. In a randomized controlled trial including patients with decompensated cirrhosis, repeated daily infusion of intravenous albumin with a targeted serum level did not improve kidney function compared with standard care; however, it did increase the incidence of pulmonary edema/fluid overload (1). As such, there is an obvious need to augment our bedside clinical assessment using novel diagnostic tools, preferably noninvasive ones. Over the past several years, point-of-care ultrasonography (POCUS) has evolved as a valuable adjunct to physical examination, although it remains largely underutilized in nephrology. We have previously discussed the general principles, scope of practice, and training program development of POCUS (2,3) and therefore will not delve into details here. In the context of AKI in cirrhosis, Velez et al. demonstrated that inferior vena cava (IVC) ultrasound helps to characterize the fluid status better and potentially prevents misclassification as HRS (4). Because of the perceived ease of image acquisition, there is a growing interest among nephrologists to learn IVC ultrasound. However, it is important to understand that isolated use of IVC is laden with pitfalls. In our experience-based opinion, multiorgan POCUS is the key to proper assessment of hemodynamics at the bedside, and herein we will explain the rationale.

Estimation of right atrial pressure (RAP) on the basis of the size and collapsibility of IVC in spontaneously breathing patients is a standard practice in echocardiography. Nevertheless, the correlation between IVC parameters and cardiac catheterization-derived RAP is only modest and is not applicable in mechanically ventilated patients (5). Moreover, as the collapsibility of IVC depends on the strength of breath, relying on an absolute percentage is error prone, especially in frail patients or in those who cannot follow instructions (e.g., patients with hepatic encephalopathy). In cases where the cirrhotic liver is hyperechoic, it can be technically challenging to visualize the vessel, and knowledge of alternate sonographic windows/imaging planes is necessary. It is not uncommon for POCUS users to mistake the adjacent aorta for IVC in such scenarios, which may result in incorrect clinical decision making. In some cases, cirrhosis-related local factors such as caudate lobe hypertrophy may alter the diameter of IVC independent of fluid status (6,7). Additionally, it is well known that raised intra-abdominal pressure can result in collapsed IVC, despite intravascular volume excess (8). Demonstration of a large amount of ascites or a distended bowel on POCUS should alert the physician to consider IAH in the differential diagnosis. Focal narrowing of the upper intrahepatic IVC had been suggested to indicate IAH (9), but we found that to be an unreliable finding in our practice. In settings where intra-abdominal pressure measurement is not readily available, we suggest using internal jugular vein (IJV) ultrasound to assess RAP. The vein can be easily located using POCUS and measuring the height of the collapse point provides an estimate of RAP (analogous to the highest point of venous pulsation when assessing jugular venous pressure by inspection). In patients who can follow instructions, change in the IJV cross-sectional area with the Valsalva maneuver predicts RAP (the vessel is more distensible when the RAP is low) (10). Interestingly, a study in patients undergoing laparoscopic surgery demonstrated that IVC collapsibility correlated well with that of IJV in the preoperative phase (r²=0.86, P<0.01) but lost correlation once the abdomen was insufflated to 15 mm Hg (r²=0.26, P=0.42). The mean IVC diameter was reduced by 5 mm during the insufflation phase (11). On a note of caution, IJV POCUS is subject to errors due to inappropriate head elevation angle, inadvertent application of excess transducer pressure, or limited access to the neck because of the presence of dialysis catheters, tracheostomy collars, or braces.

Although the IVC and/or IJV ultrasound helps to estimate RAP, the downstream effects of elevated RAP on organs such as congestive nephropathy can be gauged by Doppler ultrasound of the abdominal veins, typically hepatic, portal, and renal parenchymal veins. This technique, also known as venous excess ultrasound or VExUS, can be used to monitor the efficacy of decongestive therapy in patients with fluid overload thanks to the dynamic nature of these Doppler waveforms (12). However, in cirrhosis, hepatic and portal venous waveforms may be unreliable due to local factors. As such, we use renal parenchymal vein Doppler and an extra-abdominal vessel such as femoral vein or superior vena cava (accessed via suprasternal, supraclavicular, or parasternal windows) where feasible (13,14) to assess the flow patterns. Femoral vein Doppler is technically easier, and a pulsatile waveform demonstrates good correlation with elevated RAP; however, the sensitivity is relatively low (specificity 94% and sensitivity 46% in one study) (15). Moreover, it is less reliable in cases of severe IAH compromising the venous return from the lower limbs. In addition to IVC and VExUS, assessment of the right ventricular function, relative cardiac chamber size, and tricuspid regurgitation allow better elucidation of the right-sided hemodynamics and monitor response to therapy.

Left heart dysfunction and pulmonary congestion can potentially go unnoticed, resulting in inappropriate management or care delays if solely relying on right-sided parameters. Lung ultrasound is an easy-to-learn skill that allows rapid detection of extravascular lung water often before the onset of symptoms. Its diagnostic superiority compared with chest auscultation is well documented, particularly in patients with heart failure and ESKD (16). In liver intensive care units, lung POCUS is often used to adjust ventilator settings, determine readiness for extubation, guide ultrafiltration, and screen for procedural complications such as pneumothorax after central venous catheter placement (17). In addition, the ability to assess left ventricular (LV) filling pressures using Doppler ultrasonography provides valuable insights into fluid tolerance. This would also help when there is a question about cardiogenic (due to elevated LV filling pressures) versus noncardiogenic pulmonary edema. In a recent study, 62% of patients diagnosed with HRS by clinical criteria were found to have elevated cardiac filling pressures determined by right heart catheterization. Interestingly, when switched from volume loading to diuretic therapy, these patients showed significant reductions in serum creatinine values, suggesting superimposed or misclassified (as HRS) cardiorenal pathophysiology (18). Furthermore, estimation of stroke volume using Doppler ultrasonography aids in distinguishing between hypovolemic and high cardiac output states because both are associated with small, collapsible IVC and a hyperdynamic LV on grayscale ultrasound. This allows early initiation of vasopressors and prevents empirical volume expansion if the cardiac output is already high due to cirrhosis-associated splanchnic vasodilation (19). Being a clinician-performed study, POCUS facilitates individualized patient management by enabling serial, goal-directed assessment of hemodynamics.

As in any AKI, hydronephrosis and bladder outlet obstruction can be quickly ruled out at the bedside using POCUS without having to wait for consultative imaging. Concurrent assessment of intrarenal arterial resistive index (RI) may provide useful information in a carefully selected subset of patients, despite being a nonspecific measure. For example, RI has shown to be elevated in patients with IAH and HRS compared with those with volume depletion (20). Figure 1 provides a visual summary of POCUS applications used in the evaluation and management of cirrhotic patients with AKI.

Figure 1.

Summary of common point-of-care ultrasound parameters we use in the assessment of hemodynamics in patients with cirrhosis and AKI. IJ, internal jugular; RAP, right atrial pressure; TR, tricuspid regurgitation; IVC, inferior vena cava; US, ultrasound; VExUS, venous excess ultrasound; LV, left ventricular; LVOT, left ventricular outflow tract. Normal sonographic images shown for illustration purposes.

Some may perceive multiorgan ultrasound to be overwhelming and time-consuming. However, with proper training and thorough understanding of image acquisition principles, bedside hemodynamic assessment should take no more than 15–20 minutes in our experience. This time can be simultaneously used to elicit history or for patient education. Moreover, not every patient requires a comprehensive evaluation using all the Doppler techniques mentioned above. Figure 2 is a representation of our thought process when approaching hemodynamic AKI in cirrhosis using POCUS as a diagnostic aid. Although POCUS is within the scope of appropriately trained physicians (2,3), performing it with inadequate knowledge and skills can lead to patient harm. Establishing image archiving and a robust quality assessment system at the institutional level is vital for the success of any POCUS program. POCUS should never be considered a substitute for meticulous history taking, urine microscopy, or consultative imaging. Future studies in patients with cirrhosis and AKI should consider exploring the effect of POCUS-guided therapy on safety end points such as fluid overload.

Figure 2.

Proposed diagnostic algorithm in a case of cirrhosis and suspected hemodynamic AKI. Color coding: blue boxes indicate right heart and red boxes left heart-related sonographic parameters. Green outlines indicate volume tolerance phenotype, and the orange outlines signify volume intolerance. POCUS, point-of-care ultrasonography; VTI, velocity time integral; E/e′, ratio of the early diastolic waves of the mitral inflow Doppler and mitral annular tissue Doppler; LA, left atrium; RV, right ventricle; RVSP, right ventricular systolic pressure; TAPSE, tricuspid annular plane systolic excursion; S′, tricuspid annular systolic velocity; SVC, superior vena cava; ARDS, acute respiratory distress syndrome; HV, hepatic vein; PV, portal vein. Incorrect angle of insonation is a frequent source of error when assessing LVOT VTI (surrogate for stroke volume) and other Doppler measurements listed.

Disclosures

A. Koratala reports research funding from KidneyCure and the American Society of Nephrology’s William and Sandra Bennett Clinical Scholars Grant and honoraria from Vave Health, Inc., and Echonous, Inc. The remaining author has nothing to disclose.

Funding

None.

Author Contributions

A. Koratala was responsible for the formal analysis and wrote the original draft of the manuscript; N. Reisinger was responsible for supervision reviewed and edited the manuscript; and both authors were responsible for the conceptualization, data curation, and methodology.