Endoscopic ultrasound-guided portal pressure gradient measurement: a systematic review and meta-analysis

Introduction

Cirrhosis is now the ninth leading cause of mortality in the United States (US) [1]. The economic burden of cirrhosis is substantial; the annual direct costs are estimated to be $2.5 billion, with an additional $10.6 billion in indirect costs [2]. These costs increase substantially for patients with decompensated cirrhosis, with annual costs of over $2,400 for diuretic-responsive ascites, $24,480 for diuretic-refractory ascites, $16,400 for hepatic encephalopathy treatment, and $25,600 for treatment of patients with variceal hemorrhage [3]. Portal hypertension (PH) is the major driver of these complications; therefore, the proper diagnosis and management of PH is paramount [4,5].

Measurement of hepatic venous pressure gradient (HVPG) via a transjugular approach using interventional radiology (IR) is the current standard method for determining PH [4,6,7]. It is calculated by measuring the difference between the wedged (indirectly estimated pressure in the portal system) and free hepatic venous pressure, with the help of a catheter in the hepatic vein [4]. An HVPG of >5 mmHg is consistent with mild sinusoidal PH and compensated cirrhosis, while HVPG values of >10 mmHg represent clinically significant PH (CSPH) and are predictive of decompensation, and complications such as variceal hemorrhage and spontaneous bacterial peritonitis [8,9]. Measuring the HVPG by IR approaches is a technically challenging and invasive procedure that requires fluoroscopy; it is therefore rarely performed outside of tertiary centers. Additionally, HVPG cannot be used for accurate measurement in pre-hepatic or pre-sinusoidal causes of PH [10].

Direct endoscopic ultrasound-guided portal pressure gradient (EUS-PPG) measurement has emerged in recent years as a promising alternative method of evaluating PH. Lai et al, in 2004, first demonstrated its use in a porcine model, and in 2014 Fujii-Lau et al reported the first use in a single human subject [10,11]. In 2019, the US Food and Drug Administration approved a novel device called the EchoTip^® Insight^™ Portosystemic Pressure Gradient Measurement System (Cook Endoscopy, Bloomington, US) [12]. Under direct visualization with EUS, the portal vein and hepatic vein are identified. A 25-G needle is directly inserted into these vessels and 3 separate measurements are performed with an attached compact manometer/pressure transducer. These measurements are then averaged to determine the PPG. This is the first meta-analysis that aimed to evaluate the efficacy and safety of the use of EUS-PPG for the routine evaluation of PH.

Materials and methods

We performed an extensive literature search in several major databases—including PubMed, Medline, EMBASE, CINAHL, Cochrane, Web of Science and Google Scholar—from inception to December 2022. We used the following keywords to identify studies reporting the use of endoscopic guided ultrasound to measure portal pressure gradients: “endoscopic”, “ultrasound”, “portal hypertension”, “pressure”, “liver”, “gradient”, and “needle”. We followed the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines [13] to identify studies that used EUS for portal pressure gradient measurements (Supplementary Fig. 1 and Supplementary Table 1).

Results

Search results and population characteristics

From an initial pool of 415 studies, 8 were included in the final analysis. A schematic diagram of the study selection according to the PRISMA guidelines is illustrated in Supplementary Fig. 1. The analysis included a total of 178 patients, the majority of whom were males (52.3%). The patients’ mean age was 55.8 years (range 33.1-64). Six studies were from the US, 1 was from China, and 1 was from Australia. The etiology of liver disease was described in 5 studies. The most common etiology was nonalcoholic fatty liver disease (31.5%) followed by hepatitis C virus (14.2%). Table 1 describes the characteristics of the included studies.

Table 1 Demographic characteristics of the included studies

Meta-analysis outcomes

Primary outcomes

The calculated pooled rate of technical success (Fig. 1) was 94.6% (95%CI 88.5-97.6%; I²=0), while the pooled rate of clinical success (Fig. 2) was 85.4% (95%CI 51.5-97.0%; PI=2-100%; I²=70).

Figure 1 Forest plot showing the technical success of endoscopic ultrasound-guided portal pressure gradient measurement

CI, confidence interval

Figure 2 Forest plot showing the clinical success of endoscopic ultrasound-guided portal pressure gradient measurement

CI, confidence interval

Secondary outcomes

The calculated pooled rate of total AEs (Fig. 3) was 10.9% (95%CI 6.5-17.7%; PI=5-23%; I²=4). The most common AE was abdominal pain (11.0%, 95%CI 6.4-18.3%; PI=4-25%; I²=8.4), followed by bleeding (3.6%, 95%CI 1.4-8.8%; I²=0). No cases of perforation or death were reported in our study. As per the ASGE Lexicon, 93.7% (95%CI 66.3-99.1%; I²=0) of AEs were mild in nature [15].

Figure 3 Forest plot showing overall adverse events associated with endoscopic ultrasound-guided portal pressure gradient measurement

CI, confidence interval

Discussion

This is the first meta-analysis to investigate the EUS-guided measurement of PPG. Our analysis demonstrated high rates of clinical and technical success, with a low AE rate. The technical success rate of EUS-guided measurement of PPG was 94.6%. Since EUS has already been routinely used to identify intra-abdominal structures and vasculature, and to obtain biopsy specimens, all included studies were able to introduce the needle into the hepatic and portal veins with little difficulty [20-28]. There are certain factors that can contribute to a failure to gain access to the vessels, such as a narrow hepatic vein diameter in obese patients and sinusoidal obstruction syndrome or obstruction of a vessel in Budd-Chiari syndrome [22,23,25]. While the inferior vena cava could be an alternative point of entry, it too could be compressed in cases of hepatomegaly [23].

The clinical success rate of EUS-PPG was somewhat lower, at 85.4%. Variabilities in the protocol and the technical expertise of the endoscopic sonographers may have contributed to the lower clinical success rate [22]. Furthermore, the type of sedation, sedation depth, mechanical ventilation parameters, patient positioning and fluctuations in abdominal pressure can affect the accuracy of pressures measured in the hepatic and portal veins [22]. Among the 5 studies analyzed for clinical success, 4 demonstrated near-perfect clinical success rates [21,23,25,26] while the study by Hajifathalian et al was an extreme outlier, with a 44% success rate [22]. The pooled clinical success was 92.7% if the Hajfathalian study was removed from our analysis. In that study, there were 13 of 23 patients whose EUS-PPG results were incongruent with fibrosis on histology. The authors attributed this to Type II error secondary to a small sample size, with only a few cases having clinically significant PH, while there was also slightly greater variability in the hepatic and venous pressure measurement as compared to previous reports [22]. Zhang et al [23] and Lim et al [25] used a transjugular approach as control, while Radlinski et al [26] used CSPH; all 3 studies demonstrated 100% clinical success. A retrospective analysis by Choi et al showed a significant correlation between EUS-PPG ≥5 mmHg and histological hepatic fibrosis [29]. Although the transjugular technique has been shown to correlate well with EUS-PPG, the relationship between the 2 methods is not yet fully understood and could be influenced by patient-specific conditions such as presinusoidal PH [24].

The rate of AEs for EUS-PPG was 10.9%. Most AEs (93.7%) were considered mild as per the ASGE lexicon and were managed conservatively. The sole incidence of bleeding manifested as a hepatic hematoma and was managed conservatively [27]. In order to reduce the risk of bleeding, certain areas can be targeted for introduction of the needle, such as the trunk of the portal vein, where if bleeding occurs it can more easily be noticed and treated [23]. No incidence of perforation, infection or death related to PPG was reported. Therefore, EUS-PPG can be considered a safe procedure.

There are certain disadvantages to EUS-PPG. Given its invasiveness and cost, the widespread use of EUS-PPG may not be feasible in settings with limited resources. The accuracy of EUS-PPG also depends on the technical expertise of the clinician. Moreover, as it is an emerging technique, the relevant evidence and studies available in the literature are limited.

Nonetheless, EUS-PPG also has certain advantages. It is an excellent alternative in circumstances where the transjugular technique is not available or is inaccurate, such as thrombosis or occlusion of the hepatic vein [23]. As EUS-PPG measures portal vein pressure directly, the results will not be compromised by shunt vessels of the hepatic vein [23]. With EUS-PPG, the diagnosis and treatment of PH can be performed by and consolidated under the same gastroenterologist/endoscopist team to optimize the quality of care, with simultaneous variceal screening/surveillance/treatment and liver biopsy if needed [24].

Our study had several limitations, most of which are inherent to any systematic review and meta-analysis. First, we were not able to remove any confounding variables that may have been present in the studies. We did not use gray literature databases in our literature search, as defined by the Cochrane handbook. Second, there were no randomized controlled trials available in the literature to be included in the analysis, but only retrospective or prospective cohort studies. Third, substantial heterogeneity was noted in the pooled clinical success, probably due to different etiologies for elevated PPG and variations in operator expertise, since this is a novel technique. Last, since EUS-PPG is a new technique, the results may be less robust over time, as larger studies on this topic become available and the technique is employed more widely.

Our systematic review and meta-analysis demonstrated that EUS-PPG is a technically feasible and safe procedure for the evaluation of PH. Randomized controlled trials are needed to further evaluate its clinical efficacy. Future studies may consider comparative analyses between EUS-PPG and the transjugular approach, with biopsy as the control, to further elucidate the relationship between the 2 methods of assessing portal pressure. Trials that investigate the relationship between EUS-PPG and noninvasive tests such as fibrosis-4 index are also warranted.