Transjugular intrahepatic portosystemic shunt (TIPS) represents a major advance in the treatment of complications of portal hypertension. Despite continued refinements in technique and devices, hepatic encephalopathy (HE) remains a major drawback of TIPS with a reported incidence of 20% within the first year following intervention [6, 7].
Apart from liver function, the degree of portosystemic gradient reduction has an impact on the incidence of HE [8]. In this regard, it has been shown that following TIPS placement the magnitude of the portosystemic gradient negatively correlates with the shunt diameter of the placed stent [9]. Accordingly, overshunting along with HE are more likely to occur in larger shunt diameters. Therefore, attempts have been made to regulate shunting by means of shunt diameter adaptation. The use of small-diameter (6–8 mm) stents has been advocated for the prevention of HE [10] but showed inconsistent results in terms of efficacy for controlling complications of portal hypertension [11]. Implantation of a large-diameter (ie, 10 mm) subtotally expanded stent (6–8 mm) has been proposed as an alternative approach [9–10]. In theory this approach allows for gradient adapted TIPS creation with the prospect of sequential balloon assisted shunt expansion in cases of insufficient clinical response. However, the success of this approach is dependent on the physical properties of the implanted stent.
The results of this in vitro study demonstrate that the previous generation VIATORR has the tendency to expand completely and, thus does not exhibit the physical properties necessary for individual portosystemic shunt adaptation. The results are in line with previous studies that investigated VIATORR configuration in in-vivo settings and found that the stent tends to passively expand to its nominal diameter following subtotal initial dilatation [2, 3, 11–13]. The duration of passive expansion was neither unanimous nor predictable in the various studies.
The current findings deliver the physical explanation for the continuous passive expansion of the previous VIATORR, as well as the promising results regarding diameter stability of VCX. The previous VIATORR shows gradually decreasing radial forces until full expansion (outer diameter of 10 mm; Figs. 4a-d, black line) without a relevant decrease of forces in the clinically relevant diameter range of 8 to 10 mm. Passive expansion to the nominal diameter of 10 mm will presumably occur, unless the resistance of the adjacent liver tissue exceeds COF in vivo. Only expansion with a 10 mm balloon results in a change in radial forces by raising COF slightly. This is probably due to the fact that the VIATORR is primarily normalized to an outer diameter of 10 mm, so that the inner diameter does not reach a full 10 mm. Modulation with a 10 mm balloon, minimally strains the stent beyond its nominal value.
The clinical experience described above ultimately indicated an update of the VIATORR stent design with the goal of controlled expansion. The predecessor VIATORR stent is a nitinol based stent with an uncovered 2-cm-long self-expanding chain-linked portion and an ePTFE-covered spiral nitinol portion, constrained by a suture. The VIATORR CX (VCX) is similar in design to the original VIATORR with an additional outer constraining balloon expandable sleeve on the lined region of the stent graft. This lined region was added to allow for controlled balloon assisted adjustment of the stent diameter in a range between 8 and 10 mm (inner diameter) [14]. The ability to incrementally expand the VCX and durably maintain a diameter of 8–10 mm provides a framework for this device to potentially calibrate post-TIPS portosystemic pressure gradients such that portal hypertension is sufficiently alleviated while curtailing the occurrence of post-procedure HE. Controlled underdilation of the VCX stent has shown beneficial effects in clinical use with reduced rates of HE, cardiac decompensation and overall outcome [15].
In the current study COF and RRF were examined. While the RRF is the force a stent generates while resisting compression, the COF is the force a self-expanding stent exerts while expanding. In TIPS creation COF is of higher interest, as customized gradient adaptation can only be achieved if surplus passive expansion is not an issue. All VCX stents display a decrease in COF at an outer stent diameter of 8.3 mm to values below the previous VIATORR version. This effect is reinforced by a 50% RRF reduction, which occurs at an outer diameter of 8.5 mm. Thus, it can be assumed that after implantation the VCX will open to a diameter between 8.3 and 8.5 mm without subsequent balloon modulation in most cases. Out of the box the VCX passively expanded to an outer diameter of 8.2 mm at the narrowest point and to a maximum of 8.85 mm in the covered part. The uncovered segment expands to almost 9.3 mm outer diameter, as a result the adjacent covered segment expands to 8.9 mm in vitro when no outer pressure is applied. It is to be suspected that the residual COF measured beyond an outer diameter of 8.3 mm are caused by the expansion of the proximal uncovered segment. In vivo, the duration and extent of passive expansion depends on numerous factors (i.e. degree of liver stiffness, diameter of pre-dilatation), thus the final diameter of the VCX is difficult to predict if post-dilatation is not performed.
In this way, it can be assumed that the stent will persist at a diameter between 8.5 and 8.6 mm after modulation with an 8 mm balloon, and measurements of altered portosystemic pressures can be recorded immediately.
Following modulation of the VCX with a 9 mm balloon only slightly alters the physical properties. COF reduction is noted at a diameter of 8.6 mm following re-crimping; RRF reduction occurs at 8.8 mm. Thus, a certain amount of recoil may occur in-vivo after modulation with a 9 mm balloon, depending on the degree of liver stiffness. Following 10 mm balloon modulation both COF and RRF increase to a larger extent than following 9 mm modulation: COF decrease is noted at 9.3 mm, RRF at 9.6 mm.
Further studies with a long follow-up are necessary to investigate whether the VCX can durably maintain subtotal expansion diameter in vivo.
The current study is limited by the experimental character. COF and RRF were measured after re-crimping; a process which typically does not take place in vivo.