Abstract
Background
Portal vein puncture (PVP) is a critical step during transjugular intrahepatic portosystemic shunt (TIPS) and correlates to several complications. Techniques guiding PVP are needed.
Purpose
To evaluate the safety, feasibility, and efficiency of digital subtraction angiography (DSA) overlay reference during TIPS creation and compare it with transhepatic portal vein (THPV) guiding.
Material and Methods
The clinical records of 185 patients at three medical centers who underwent TIPS placement were reviewed. Portal vein access was guided by THPV guiding in 120 cases and DSA overlay reference in 60 cases. The number of punctures, portal vein entry time, procedural adverse events, technical and hemodynamic success rate were analyzed to compare the safety, feasibility, and efficiency of the two methods.
Results
The median numbers of punctures in group 1 and group 2 were 2 (1–4) and 2 (1–5), respectively (P = 0.094). There was no statistical difference between two groups in needle passes. The median portal vein entry time of group 1 was 12 min (8–16 min) and 13 min (8–16 min) in group 2. No significant difference was found in the PVP time (P = 0.802). Arterioportal fistula formation occurred in 15 patients in group 1; two patients in group 2 had hepatic artery injury. The patients in group 2 had lower rates of procedural adverse events (P = 0.047). Median dose area product of G1 was lower than G2 statistically (P<0.001). There was no significant difference in total fluoroscopy time (P = 0.856).
Conclusion
DSA overlay reference has lower procedural adverse events rates compared with THPV guiding TIPS. It seems to be a safe and effective method for guiding PVP.
Keywords
Introduction
Transjugular intrahepatic portosystemic shunt (TIPS) was recommended as an effective therapy for the management of portal hypertension by several clinical practice guidelines (1–4). During TIPS creation, portal vein (PV) puncture is a critical and arduous step due to the invisibility of portal venous system on image. However, it is related to several procedural complications, hepatic artery (HA) injury, biliary fistula formation, and non-target organ injury (5–7). Several techniques for portal venous access have been reported, such as CO2 wedged hepatic venography, intravenous ultrasound (IVUS), and 3D roadmap. These techniques rely on specialized venography equipment and software, which limits their application in many centers.
Percutaneous placement of guidewire/catheter into the PV provides the most direct display of portal venous system, but it also increases the complexity of operation and the risk of extra liver puncture, which may lead to higher incidence of complications (8,9). Digital subtraction angiography (DSA) overlay reference could easily select the preferred overlay mode at tableside and has been used in clinical practice (orthogonal rings, fiducial markers, and overlay accuracy when image fusion is used for endovascular aneurysm repair guidance).
The aim of the present study was to determines the safety, feasibility, and efficiency of DSA overlay reference during TIPS creation compared with the placement of catheter into the PV.
Material and Methods
Patients
From January 2018 to November 2020, 185 consecutive patients who underwent TIPS placement at three medical centers were involved in this study, including 115 cases in Zhongshan Hospital, Fudan University 33 patients in Minhang Hospital, Fudan University and 37 cases in Jinshan Hospital, Fudan University. The exclusion criteria were as follows: (i) patients who underwent TIPS creation before being admitted to hospital (n = 1); and (ii) PV cannulation under guidance of transhepatic placement of guidewire into the PV) (n = 4). A total of 180 patients were included in the study.
All patients underwent computed tomography (CT) hepatic angiography and portal venography to reveal the structure of HA and portal venous system before the procedure. In 2020, patients whose HA and portal vein were closely related as shown on CT angiography were tended to be guided by DSA overlay reference. Transhepatic portal vein guiding tended to be conducted in 2018 and 2019 as well as in difficult anatomic cases in 2020 including but not limited to the following: (i) patients who underwent splenectomy; and (ii) extensive PV thrombosis.
All enrolled cases were dived into two groups according to the guiding method: Group 1 (G1) included transhepatic PV guiding TIPS (n=120); and Group 2 (G2) included DSA overlay for targeting PV during TIPS creation (n=60).
The experimental protocol was established, according to the ethical guidelines of the Helsinki Declaration and was approved by the Ethics Committee of our hospital. Written informed consent was obtained from all enrolled patients.
TIPS techniques
Transhepatic portal vein guiding
Under the guidance of ultrasound, the peripheral PV was accessed by a 22-G Chiba needle (Merit, South Jordan, Utah, USA) and NEFF set (Merit, Mexico, USA) was introduced into the main PV (or left/right branch). A 4-F pigtail catheter (Merit, South Jordan, Utah, USA) was advanced into the splenic (or superior mesenteric) vein for direct portal venography to reveal the anatomy of the PV system. The 4F Pigtail catheter (Cordis, Miami, FL, USA) was then placed at the chosen puncture site as a fiducial landmark to guide the puncture of PV (Fig. 1). After the whole TIPS procedure, 0.018-inch 3/140 mm NESTER coils (Nester; Cook, Bloomington, IN, USA) were used to embolize the puncture route.
Transhepatic PV guiding PV access. White arrow shows the tip of RUPS-100; black arrow points to the end of the 4 F pigtail catheter, which was the endo-PV marker. PV, portal vein.
DSA overlay reference
Angiography of celiac artery or superior mesenteric artery (only in cases when ectopic HA originates from superior mesenteric artery) to acquire indirect portal venogram with a 4-F MPA catheter (Cordis, Miami, FL, USA) from radial access. The HA and PV were then overlayed on the same image under reconstruction by DSA overlay software as a puncture navigation map (Fig. 2). A microcatheter (Progreat; TERUMO, Fujinomiya City, Japan) was then intubated into the chosen HA branch with its tip as a PV puncture guiding marker.
DSA overlay reference image. Black arrow points to the marked point of the hepatic artery; white arrow points to the chosen shunt site of PV. DSA, digital subtraction angiography; PV, portal vein.
In both groups, once the PV access was established and confirmed by contrast medium injection, a 0.035-inch, 260-cm guidewire (TERUMO, Fujinomiya City, Japan) was then inserted into the splenic (or superior mesenteric) vein and a 4-F pigtail catheter (Cordis, Miami, FL, USA) was advanced through guidewire into the splenic (or superior mesenteric) vein for direct portal venography (Fig. 3), and initial PV pressure was measured subsequently. Based on the patient's clinical status, a 6-mm or 8-mm balloon (Rival; Bard, Tempe, AZ, USA) was applied to dilate the puncture route. Then an 8-mm TIPS stent (Viatorr; Gore, Phoenix, AZ, USA) was placed, portal pressure was measured subsequently, a 6-mm, 8-mm, or 10-mm balloon (Rival; Bard, Tempe, AZ, USA) was used for re-dilation based on the decline extent of portal pressure. After TIPS creation, post-TIPS portal pressure was measured. Hepatic arteriography was conducted to preclude HA injury and further HA embolization was performed if necessary.
Direct portal venography through splenic vein. Black arrow points to the end segment of the pigtail catheter, the marker of PV. White arrow shows the chosen shunt site of PV. PV, portal vein.
Definitions and outcome collection
Patients’ demographics including age, sex, Child-Pugh score (CPS) and classes, indication for TIPS, and etiology of liver diseases were recorded. Indications were refractory visceral hemorrhage, refractory ascites, and Budd–Chiari syndrome according to the American Association for the Study of Liver Diseases (10) and Society of Interventional Radiology (2). The pre- and post-interventional portosystemic pressure gradient (PSG), PSG reduction, number of punctures, PV entry time (PET), and procedural adverse events including arteriovenous fistula formation and HA injury were recorded from clinical data.
PET was calculated from the beginning of the puncture with RUPS-100 to guidewire into the portal venous system and confirmed by manual injection of contrast medium. Procedural adverse events were defined according to the Proposal of a New Adverse Event Classification by the Society of Interventional Radiology Standards of Practice Committee (2).
Statistical analysis
The Kolmogorov–Smirnov test was used to evaluate the normality of descriptive data. Continuous variables are shown as mean ± standard deviation (SD) or median (25th–75th percentile range), or median (range) whereas categorical variables are reported as frequencies. Mann–Whitney U-test was applied to assess the non-normally continuous variables. Comparison of categorical variables was performed using Pearson's chi-square test or Mann–Whitney U-test. P values <0.05 were considered statistically significant. Statistical software SPSS version 22.0 (IBM, Armonk, NY, USA) was used for statistical analysis.
Results
A total of 180 patients (126 men, 54 women; median age = 57 years; age range = 49–67 years) were included in the present study. There were 120 (66.7%) patients in Group 1 and 60 (33.3%) patients in Group 2. Baseline characteristics of the 180 patients undergoing TIPS are shown in Table 1. The patients’ liver function was mainly graded as CPS B (n=89). The median CPS was 7 (range = 6–8). The most frequent indication was visceral bleeding (n=157). The most common etiology was hepatitis B-related cirrhosis (n=93). PV thrombosis was observed in both groups.
Baseline characteristic of 180 TIPS patients.
Values are given as n or median (range).
HBV, hepatitis B virus; HSOS, hepatic sinusoidal obstruction syndrome; TIPS, transjugular intrahepatic portosystemic shunt.
The procedural characteristics of 180 patients with TIPS are shown in Table 2. In this study, no significant difference was found between the two groups in the median number of punctures (2, range = 1–4 vs. 2, range = 1–5; P = 0.094). The median PET in the group 1 was 12 min (range = 8–16 min) and in group 2 it was 13 min (range = 8–16 min). No significant difference was found in PET (P = 0.802). The technical success rate and hemodynamic success rate in G1 and G2 were both 100% and 95.0%, respectively. There was no significant difference of both technical success rate and hemodynamic success rate in two groups (P = 1, both). Procedural adverse events occurred in 17 patients, including 15 cases of slighter arterioportal fistula formation without additional medical treatment in G1 (all caused by transhepatic PV puncture verified by location) and two cases of HA injury on the PV puncture route in G2 (both patients received embolization of responsible bleeding vessels through a pre-intubated catheter in the HA). G2 has lower rates of procedural adverse events than G1 (P = 0.047). The dose area product (DAP) of G1 was lower than G2 statistically (P<0.001). There was no significant difference in total fluoroscopy time (P = 0.856).
Procedural characteristics of 180 TIPS patients.
Values are given as n, mean ± SD, or median (range).
DAP, dose area product; HBV, hepatitis B virus; HSOS, hepatic sinusoidal obstruction syndrome; PSG, portosystemic pressure gradient; TIPS, transjugular intrahepatic portosystemic shunt.
Discussion
PV access is the most critical and difficult step in the creation of TIPS. Several methods have been practiced guiding needle passes from the hepatic vein into the “blind” PV. In this study, we explored the feasibility and safety of DSA overlay for guiding PV puncture and compared it with transhepatic portal vein guiding TIPS. The results showed that the method seems to be safe, feasible, and easy to implement.
The results of TIPS creation under transhepatic portal vein guiding and DSA overlay reference guiding in our study indicate that both guiding methods had low rates of procedural adverse events rates (12.5% and 3.3% in G1 and G2, respectively). Furthermore, DSA overlay guiding TIPS has significantly lower rates of procedural adverse events compared with transhepatic PV guiding. To our knowledge, percutaneous placement of markers into PV with ultrasound guidance requires expert experience when performing the ultrasound (11). This may limit its use. In addition, direct transhepatic puncture of the PV may increase the risk of accidental penetration of the hepatic arteries and (or) biliary tract. In our study, formation of 15 hepatic arterioportal fistulae (APF) occurred in G1 due to transhepatic puncture. Hepatic arteriography was routinely performed in the present study. The relative higher rates of APF may attributed to the increased detection rate by hepatic arteriography. Considering that extra liver puncture was prone to bleeding complications (12,13), it should be performed with caution in patients with coagulopathy (14).
Percutaneous PV localizer guiding TIPS creation has been reported before; it can directly locate the portal vein. Gipson et al. (15) reported PV marker wire guiding PV access (n = 18): the technique success rate and hemodynamic success rate were both 94.4%. Compared with their cohort, the technical success rate and hemodynamic success rate (100% and 95.0%, respectively) of the transhepatic PV guiding portal vein puncture of this study seem to be better. This may be due to the different experience of their operator. In contrast, all of our interventional radiologists have years of expert experience in PV guiding TIPS creation. In our study, compared with transhepatic PV guiding TIPS, DSA overlay guiding PV access has no statistical difference in both technical success rate and hemodynamic success rate (P = 1, both), indicating that DSA overlay may be a feasible technique for guiding TIPS creation.
HA catheterization guiding TIPS has been reported by Warner et al. (11). With the DSA overlay reference, the relative position of PV and HA can be displayed more intuitively, which could potentially increase the accuracy of PV puncture. Yamagami et al. (14) reported that with HA-targeting guidewire guiding PV puncture, the median number of punctures of their study was 5 (range = 1–14). The present study seems to have a better result in median numbers of needle passes (2; range = 1–5). This could be explained by the overlaid image of the HA and PV providing an intuitive display of the relative structure of the HA and PV as a navigation map, thus possibly increasing the accuracy of PV puncture. Although the radial access is invasive, no radial access-related complications were observed. In the present study, hepatic arteriography was routinely performed in the TIPS procedure. It is known that major complications related to artery injury such as hemoperitoneum, hemobilia, hepatic infarction, and HA injury were rare but real (2). Interventions should be taken opportunely. Two patients in our study had a HA injury and were successfully treated with embolization of responsible vessels. For patients with severe coagulopathy or multiple punctures, hepatic arteriography after shunt creation should be considered for potential HA injury.
There are other techniques visualizing the portal venous system such as 3D roadmap, cone beam computed tomography (CBCT) navigation, and CO2 wedged hepatic venography. 3D roadmap (16–18) and CBCT navigation (16) can visualize the structure of the portal venous system. However, both techniques take a relatively long time for reconstruction and are restricted to respiratory motion (16,18). Due to the additional 3D acquisition procedure, CBCT navigation might increase the DAP. In contrast, the DSA overlay technique is not affected by respiration on account of the relatively fixed structure of the HA and PV. In addition, the image overlay technique is easy to operate. It only requires the patient to hold their breath one single time while celiac artery (or superior mesenteric artery) angiography was performed to obtain the hepatic angiogram and portal venogram in one procedure.
CO2 wedged hepatic venography is another non-invasive method of portography while a special delivery system is required (19). In contrast, no additional medical equipment is needed for DSA overlay guiding. It is reported that the PV opacification rate of CO2 wedged hepatic venography was 87% (20) to 91% (21). All the patients included in the DSA overlay group seem to have good portal vein opacification; however, there were 12 patients with PV thrombosis complications. In addition, CO2 wedged hepatic venography is not an absolutely safe method; complications such as air embolism and vapor lock have been reported (22,23). In the present study, no DSA overlay related technical complications occurred.
In summary, DSA overlay may be an alternative method in situations when other guiding techniques are not available or not suitable, with several advantages including the following: first, DSA overlay is easy to implement and no additional equipment is needed. Although radial access is an invasive technique, no radial access-related complications occurred. More importantly, it could avoid extra liver puncture and its potential risks, preparing for the embolization of the possible HA injury on the parenchyma tract. Furthermore, the position of the arterial catheter could determine the depth of needle pass and reduce the risk of non-target puncture (11).
In some cases, hepatic angiography and indirect portal angiography cannot be obtained in one procedure due to the absence or occlusion of splenic vessels on account of surgical splenectomy and other reasons (12); therefore, hepatic arteriography and indirect portal angiography need to be performed separately, which may affect the accuracy of overlayed image. In addition, when the HA originates from the superior mesenteric artery, intubation to the superior mesenteric artery for arteriography and indirect portal angiography can be an alternative choice.
The limitation of this study is that it is a retrospective study. TIPS canulation in difficult anatomical cases evaluated by preoperative CT scan tends to be guided by transhepatic PV puncture, which may interfere with the results. Further randomized control trials may be needed to prove the results.
In conclusion, DSA overlay reference seems to be a safe and effective technique for guiding PV puncture during TIPS creation. It has advantages in procedural adverse events rates compared with transhepatic PV guiding TIPS placement. This study might offer an alternative method for guiding PV puncture, the most critical and difficult step in TIPS placement.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by The National Key Research and Development Program of China (Grant No. 2017YFC0109205) and Zhongshan Hospital of Fudan University, PR China (Grant No. 2018ZSLC23).
