IMAGING AFTER ENDOVASCULAR ANEURYSM SEALING AND CONSEQUENCES FOR ENDOLEAK CLASSIFICATION
Leo van den Ham is a PhD student in the department Multi-Modality Medical Imaging. (Co)Promotors are prof.dr. M.M.P.J. Reijnen from the faculty S&T and prof.dr. C.J.A.M. Zeebregts from University Medical Center Groningen and dr. S. Holewijn from Rijnstate.
In the past decades treatment of abdominal aortic aneurysms has shifted from major open abdominal surgery to minimal invasive endovascular surgery with a huge reduction in early morbidity and mortality. The Nellix EVAS system was developed with the aim to reduce complications at the long term with subsequent lower re-intervention rates,. It promised to be a valuable addition to the selection of stent graft devices available for the treatment of abdominal aortic aneurysms with a novel approach to both device fixation and aneurysm sealing. In Chapter 2 the surgical procedure and technical aspects of Nellix EVAS are thoroughly described. Furthermore, a comprehensive list is given in which Nellix EVAS compares to other devices in terms of usage and how it could have been clinically applied. Successful treatment of the first patient groups treated with Nellix EVAS is also described in this chapter. In theory, the device had the possibility to adequately seal off the aneurysm from the patient’s blood flow and therefore, a high chance of a successful aneurysm exclusion with potentially less long term complications. Long term and large scale research was however needed to confirm this theory and at the same time look for possible down sides. In Chapter 3 is described how to pre-calculate the needed polymer volume based on pre-procedural CT-imaging in planning a Nellix EVAS procedure. If there would be a reliable method for pre-calculating the polymer volume, this could have a positive impact in terms of surgery time and optimal endobag filling. Both a manual volume calculation and computer automated volume measurement were studied in this chapter. Both calculations were proven similar without significant differences (75.3 mL versus 75.7 mL). There was however an average underestimation of the true polymer volume used during surgery (-11.3 ± 9.9 mL). Pre-operative volume calculations were therefore considered to be not useful and an intra-procedural test fill using saline, would still be advised to estimate the total polymer volume needed to properly fill the endobags.
A cornerstone in the successful follow up of patients treated with either EVAR or EVAS is imaging. Chapter 4 describes how Nellix EVAS could be assessed on CT-angiography and what changes could occur in terms of anatomy and morphology of the abdominal aorta and the aortic side branches. A minimal progression of the total aortic diameter was observed in short term follow up, which recovered after one year. Additionally, a slight increase in polymer volume was observed from 30 days to one year after implantation (68.2 mL versus 71.9 mL). In contrast, the average volume of thrombus inside the aneurysm after one year decreased from 85.6 mL to 78.8 mL. Lastly, a change in polymer density was observed, as the polymer became less dense after one year compared to imaging at the first follow up after 30 days.
Conventional CT-angiography makes a static image of the aorta during a single moment in the cardiac cycle which can be a potential limiting factor. Chapter 5 addresses the changes of the abdominal aorta and the Nellix device during the cardiac cycle using ECG-gated CT-angiography (referred to as dynamic CT). The study showed in a small patient population treated with Nellix EVAS, that during the cardiac cycle no significant changes were observed in aortic anatomy and its’ side branches. The aortic neck remained stable and no change in aortic angulation were seen. There was however a significant increase in diameter and surface of the aortic neck. During follow up, one patient experienced a device migration more than 5 mm without visible changes or movement of the surrounding anatomy.
Not only do patients treated with either EVAR or EVAS receive follow up CT-scans, the use of nuclear imaging has increased significantly in the past decades. Especially for patients with a suspected graft infection or in diagnosis of malignancies. In Chapter 6 FDG PET-CT was used to compare the changes in the abdominal aorta before and after Nellix EVAS treatment. One of the objectives of the study was to differentiate physiological changes in FDG uptake after Nellix EVAS from possible pathological or infectious changes. A comparison was made by performing whole body FDG PET-CT on patients before and after Nellix EVAS comparing the uptake values. The study showed that after Nellix EVAS a normal to decreased FDG uptake was observed at the level of the abdominal aorta. No increase in FDG uptake was seen in the, albeit small, patient population suggesting that increased FDG uptake has a pathological origin like a peri-prosthetic infection.
The altered fixation mechanism of the Nellix device has brought a different view on the origin and occurrence of endoleaks. The current endoleak classification system and detection of complications is based on the conventional fixation mechanism of EVAR, mainly with proximal fixation using barbs and hooks with which the device is fixated on the aortic wall. Observation of Nellix EVAS patients during follow up showed different blood/contrast leakage routes on imaging which is the main talking point in Chapter 7. Different types of endoleaks were described and how they can appear on different image modalities. Especially type Ia endoleaks were more prevalent in Nellix EVAS patients compared to EVAR in which type II and Ib endoleaks were more prevalent. Early signs of these type Ia endoleaks on CT-angiography include leakage of contrast between both the endobags or between the aortic wall and the endobag. This chapter suggested using a different type of classification for these endoleaks in order to associate them with the different methods of repair. As a consequence a large multi-center study was performed (Chapter 8) to evaluate the occurrence of type Ia endoleaks, describe the possible cause of the endoleak and how to properly classify these endoleaks using a new classification system. Fifteen Nellix EVAS experienced centers included 58 patients (3.1% of a total of 1851 treated patients) with a proximal endoleak. A panel of international experts, both radiologists and vascular surgeons, evaluated the available imaging of each patient. Most often a low stent positioning, unsuitable anatomy or a combination of both was described as a possible cause of device failure. Four different types of proximal endoleaks were described: Leaking of contrast between the endobags in the aortic neck (type Is1), contrast between aortic wall and endobag at the level of the aneurysm (type Is2), contrast between the endobags (type Is3) and growth of the aneurysm without an observable endoleak (type Is4). It was presumed that non-visible endoleaks with visible migration are at-risk for rupture. In 40 patients (69%) treatment was performed at time of endoleak discovery. Revision to open surgery and embolization using coils or onyx were most prevalent with a high technical success rate (96.5%). Additionally, a comparison in treatment for proximal types of endoleak after Nellix EVAS compared to treatment of endoleaks after EVAR is described in Chapter 9. In conventional EVAR, an extension of the proximal seal is key in prevention of a type Ia endoleak, where in Nellix EVAS prevention of migration is the key to success. The use of a (fenestrated) cuff in EVAR is popular ensuring an adequate seal starting from a healthy part of the aorta. With the added use of so called endoanchors extra fixation can be achieved by fixating the cuff directly to the aortic wall. In patients with Nellix EVAS in the early phase of proximal endoleaks embolization using coils or Onyx can be sufficient to prevent further leakage and migration, although the evidence is limited. In case of a progressive endoleak and migration, especially in a fit patient, conversion to open reconstruction can be considered. If a patient is not fit for major surgery, a proximal extension using two Nellix devices (with chimneys) can be used to exclude the aneurysm.