Guiding Principles for Physician-Modified Endografts

PMEG Should Be Performed in Specialized Aortic Centers

The European Society for Vascular Surgery (ESVS) guidelines recommend that complex endovascular aortic repair, including FEVAR and BEVAR, should be concentrated in high-volume aortic centers with sufficient procedural volume and multidisciplinary expertise. ¹³ Although no specific volume threshold has been established for PMEG specifically, the same principles should apply. PMEG is a technically demanding procedure that requires extensive expertise planning and fenestration/branch positioning, advanced endovascular techniques (including bail-out strategies) and an intimate knowledge of the endograft platform used for the PMEG. Several studies have demonstrated a substantial learning curve for PMEG and FEVAR procedures.^14,15 Increasing operator and institutional experience has been associated with shorter procedure times, improved technical efficiency, and lower complication rates. Therefore, PMEG should be restricted to centers with established expertise in complex endovascular aortic repair and the capability to manage procedural complications and long-term follow-up.

PMEG Is Not CMD; Dedicated Training and Proctoring Are Essential When Starting a PMEG Program or Changing Techniques

Physician-modified endograft requires distinct skills that are non-transferable from experience gained with CMDs. In addition to procedural planning and implantation, operators must acquire expertise in fenestration positioning, creation, and reinforcing, adding diameter reducing ties, and resheating the endograft. Even desheating prior to the modification may differ from routine in situ deployment depending on the endograft platform. Furthermore, the choice of endograft platform may itself be influenced by the modifications required and position of the fenestrations. Consequently, even in high-volume centers, adequate training is essential when starting a PMEG program. Difficulties encountered during graft modification and deployment are best identified and addressed outside the clinical setting during dry-run rehearsals rather than during a patient procedure, particularly in semi-urgent cases. The same principle applies when introducing changes to an established PMEG protocol. Proctoring by experienced operators, either in person or remotely, may improve patient safety during the learning curve, as unforeseen challenges are inevitable when modifying devices beyond their intended use.

PMEG Planning and Endograft Modifications Should Rely on Reproducible Methods Supported by the Best Available Evidence

Correct fenestration positioning is of utmost importance to achieving a good outcome. Multiple fenestrations, however, make this key step increasingly complex and susceptible to operator-dependent variability. Therefore, we suggest that when dealing with multiple fenestrations, a template should be strongly considered to guide the placement of fenestrations in the endograft. The use of a two-dimensional (2D) or three-dimensional (3D) template based on central lumen line reconstruction measurements derived from high-quality computed tomography angiography (≤1 mm slice thickness) is frequently reported and closely resembles the planning methodology used for CMDs.^11,12,16 Furthermore, the use of a template facilitates optimal positioning of multiple fenestrations while avoiding stent-strut interference, as the template can be adjusted relative to a position that best accommodates all the fenestrations.

Several groups have described the use of 3D printed aortic models to assist in fenestration placement, but results have been conflicting and only small cohorts have been reported.^17,18 Advanced novel techniques, such as laser-fenestration technologies and artificial intelligence assisted planning have been introduced with the aim of improving reproducibility and precision, but are not yet supported by robust clinical data.^16,19 As reproducible and consistent target vessel incorporation and patency must be achieved across operators and institutions, standardized planning methods are needed at the institutional level.

Many different suture types and reinforcement materials are currently used to reinforce fenestrations, in part because of the absence of comparative clinical evidence. Available experimental data suggest that a double-loop of reinforcement materials and a high number of suture passages (>15 per quadrant) improve mechanical resistance to deformation.^20,21 However, no consensus currently exists regarding the optimal reinforcement strategy.

Multiple endograft resheathing techniques have also been described. Experimental studies suggest that the use of a crimper may reduce graft infolding compared with traditional tourniquet-based reconstraining techniques. ²² Consequently, local protocols should be guided by the best available evidence, regularly updated as new data emerge, while complying with existing regulations and taking local resources into account.

Novel PMEG Techniques Should Undergo Appropriate Experimental Validation Before Clinical Implementation

Physician-modified endograft comes with a large degree of freedom. The possibilities are virtually unlimited. At the same time, the field is rapidly evolving and results for different techniques are being published.^9,11,19 Innovation is essential to further improve PMEG. However, novel concepts should not be introduced directly into routine clinical practice without appropriate evaluation. The treating physician has an ethical responsibility to ensure that modifications to established PMEG workflows are supported by experimental testing and, where possible, prospective clinical evaluation before widespread adoption.

Endograft Modifications to the Endograft Platform Should Be Minimized

Physician-modified endograft is based on the modification of commercially available endografts outside their original instructions for use. Although these modifications allow treatment of anatomies that would otherwise be unsuitable for standard endovascular repair, altering endovascular devices in ways outside of the instructions for use might come with unknown consequences. Additional alterations should only be introduced when a clear procedural or clinical benefit is anticipated. For instance, the addition of orientating markers requires additional suture passages through the endograft fabric without clear benefit, and when presented the choice between perforating or non-perforating diameter reducing ties, it is preferential not to perforate the fabric as doing so potentially damages the integrity of the fabric, increasing the risk for (occult) endoleaks, without proven benefit. ²³

PMEG Programs Should Have a Clear Modification Protocol and Prospective Tracking of Outcomes

The range of possible modifications is virtually endless. Therefore, detailed reporting of the modifications performed is essential. High-quality visual depictions of the modification process performed should routinely accompany publications, allowing techniques to be more easily understood, reproduced, and critically evaluated by other centers. Furthermore, because PMEG encompasses a wide variety of planning methods, endograft platforms, modification strategies, and implantation techniques, results from highly heterogeneous cohorts should be interpreted with caution.^8,9,24 Such heterogeneity may lead to overestimation or underestimating of the safety and effectiveness of individual techniques.

When starting a PMEG program, these considerations should be among the foremost points of discussion. Results from retrospective PMEG cohorts may be particularly difficult to reproduce for a variety of reasons. First, PMEG is a technically demanding procedure that requires not only advanced endovascular skills but also extensive experience in planning complex aortic repairs.^14,15 Second, PMEG is an umbrella term encompassing a wide variety of modification strategies, which may result in substantially different procedures being grouped together under a single label. Third, retrospective studies may be susceptible to both publication bias and reporting bias, particularly because responsibility for device modification rests entirely with the treating physician. Finally, PMEG techniques frequently evolve over time, meaning that modifications performed early in a program may differ substantially from those used later in the same series. ¹⁴ This may introduce significant heterogeneity within individual cohorts and potentially obscure technique-specific outcomes. To improve the quality, transparency, and reproducibility of PMEG data, prospective multicenter studies, registries, and standardized reporting frameworks are warranted.^6,7

As PMEG techniques can vary greatly and the literature may suffer from significant selection bias, tracking local results in an institutional registry should be the norm. In this manner, early detection of safety signals is possible. Collaborating in national or international registries should be strongly encouraged to establish more robust data regarding the safety and efficacy of PMEG in a broader context, allowing for the possibility to look for difference in outcomes based on the techniques used and better understand the impact of institutional volume on PMEG outcomes.

Treatment Decisions Should Acknowledge the Limitations of Current Evidence and Be Made Through Shared Decision-Making

Multiple endovascular treatment strategies may be considered when managing complex aortic pathology, particularly in semi-acute patients. These include PMEGs, off-the shelf branched devices, parallel stent-grafting techniques, in situ fenestrations, or delayed repair while awaiting manufacture of a CMD.^2,7,8,25-30 As PMEGs continue to evolve and new devices and techniques become available, questions inevitably arise about the relative role of each treatment strategy. However, it is important to recognize that all comparative data currently available are derived from observational studies.^9,25-30 Therefore, comparisons between treatment options are subject to significant selection bias, leading to a documented poor agreement between observational trials and randomized controlled trials. ³¹ Consequently, no strong conclusions can be drawn from current literature on the relative place PMEG, including the optimal treatment strategy for elective patients who are suitable for multiple approaches and the threshold at which PMEG should be preferred over waiting for a CMD in semi-urgent presentations.^2,3,9

Generating higher-quality evidence, ideally through prospective multicenter studies and randomized comparisons where feasible, should remain a priority. Until such evidence becomes available, uncertainty should be openly acknowledged and incorporated into shared decision-making.