Polymer leak with the Ovation Abdominal Stent Graft System: Early recognition and treatment

Introduction

Endovascular aneurysm repair (EVAR) has become the treatment of choice for patients with abdominal aortic aneurysm (AAA) and has been shown to decrease the short- and midterm morbidity and mortality associates with abdominal aortic repair.^1,2 Multiple stent graft systems are currently available and advances in technology allow for the treatment of more complex and varied anatomy. However, the use of these varied and advanced stent grafts requires an advanced endovascular skill set and an awareness of the unique device-specific complications. The Ovation Abdominal Stent Graft System (Endologix, Inc., Irvine, CA) is a relatively newer device that can potentially accommodate a wider range of aortoiliac anatomies and provide proximal seal in a short length proximal seal zone.^3–5 Proximal seal is achieved via a series of inflatable rings that are filled with a low-viscosity biocompatible polyethylene glycol (PEG)-based polymer at the time of deployment. Leakage of the polymer outside the fill channels can lead to system exposure to the polymer and, in rare circumstances, a hypersensitivity reaction and anaphylactic shock. ⁶ We report a case of anaphylactic shock during graft deployment and describe the worldwide incidence of polymer leak and anaphylactic reaction associated with deployment of the Ovation stent graft system, review patient demographics and outcomes as well contributing factors and strategies to avoid complications related to polymer leak.

Case report

A 79-year-old woman was incidentally found to have a 5.4 cm AAA. Co-morbidities included diabetes mellitus, chronic obstructive pulmonary disease, obesity, hyperlipidemia, hypertension and atrial fibrillation. Cardiac evaluation revealed an ejection fraction of 56% and mild coronary artery disease. She had a 50 pack year history of smoking and was an active smoker. The patient reported no known drug allergies, and laboratory data were within normal ranges. Computed tomography (CTA) revealed a 5.4 cm AAA with no evidence of rupture (Figure 1). Bilateral iliac arteries were noted to be small in caliber and diffusely calcified, and the infra-renal neck was reverse-taper in configuration. The patient underwent percutaneous EVAR under general anesthesia with a 16 F sheath placed via right common femoral approach and deployment of a 34 mm Ovation device per the instructions for use. After deployment of the supra-renal stents, polymer was injected via the auto-injector, and approximately 7 cc of polymer was noted to have remained in the syringe. The contralateral gate was cannulated. Approximately 15 min after polymer injection, the patient suddenly became severely hypotensive (from 120 to 80 mm Hg and then loss of tracing based on invasive monitoring) with a drop in end-tidal CO₂ to 18, desaturation of SaO₂ to 80% and bilateral wheezing noted on chest auscultation. The polymer syringe was noted to be empty. An iStat hemoglobin was reported as 10, and an aortogram revealed an intact aneurysm sac (Figure 2(a)). The patient was simultaneously treated for an anaphylactic reaction with administration of intravenous Benadryl, Solu-medrol, epinephrine and inhaled albuterol resulting in return of vital signs and monitoring indices to normal parameters. The procedure was completed without further hemodynamic instability and completion angiogram in the operating room revealed no evidence of endoleak (Figure 2(b)). The patient was extubated in recovery room and spent a night being monitored in the ICU before being discharged home on post-operative day 1. A 30-day CTA showed a well-placed endograft with no evidence of type I or II endoleak (Figure 3).

OPEN IN VIEWER

Figure 1.

Preoperative CTA demonstrating a 5.4 cm abdominal aortic aneurysm with a reverse-taper neck and severely calcified iliac arteries.

OPEN IN VIEWER

Figure 2.

(a) Intraoperative angiogram 15 min post polymer injection showing no evidence of contrast extravasation and (b) Completion angiogram demonstrating proper inflation of polymer filled rings and exclusion of the aneurysm sac with no evidence of type I or III endoleak.

OPEN IN VIEWER

Figure 3.

30-Day postoperative CTA demonstrating proper graft placement and exclusion of the aneurysm sac with no evidence of type I, II or III endoleak.

Methods

Results

Twenty-six patients (69% male, 19% female and 12% not available) with mean age of 77 years (range 61–100 years) were identified with polymer leak by either an empty fill polymer syringe and/or a compromise of the fill channel. In this time period, approximately 10,000 Ovation grafts were deployed worldwide, yielding an incidence of polymer leak of approximately 0.26% (Table 1). Anaphylactic reactions were reported in 24 cases with two patients exhibiting no symptoms. All 24 patients with anaphylaxis exhibited transient hypotension and three of them also had a rash. All of these patients were treated in accordance with standard recommendations for patients with contrast allergies. One patient exhibited no symptoms but was prophylactically treated for anaphylaxis. None of the patients died, and the aortic aneurysm was noted to be successfully excluded in 23 cases at the end of the procedure. One patient had a type IA endoleak identified on the one-month imaging study, and this resolved post re-intervention. In three cases, the aneurysm sac was not excluded at the end of the case because the physician involved elected not to deploy the iliac limbs (Table 2). Of these patients, one went on to successful re-intervention with placement of the iliac limbs and exclusion of the aneurysm. In one case, the decision was made to monitor the patient and one patient was converted to open surgical repair.

OPEN IN VIEWER

Table 1.

Worldwide distribution of reported cases of anaphylaxis due to polymer leak.

Country	N (%)
Belgium	1 (3.8)
Germany	3 (11.5)
Greece	4 (15.4)
Italy	4 (15.4)
Spain	2 (7.7)
Turkey	1 (3.8)
UK	3 (11.5)
USA	8 (30.8)

OPEN IN VIEWER

Table 2.

Patient and procedure characteristics.

Age (years)	77 (range 61–100)
Gender	N (%)
Female	5 (19)
Male	18 (69)
Unknown	3 (12)
Reaction
Hypotension	21 (80.8)
Hypotension + Rash	3 (11.5)
None	2 (7.7)
Aneurysm excluded
Yes	23 (88.5)
No	2 (7.7)
Excluded at re-intervention	1 (3.8)
Contributing factor/cause
Manufacturing process	10 (38.5)
Early ballooning	5 (19.2)
Low body temperature	4 (15.4)
Excessive device manipulation	4 (15.4)
No cause determined	11 (42.3)

Discussion

The Ovation stent graft is a modular two-docking limb device that includes anchoring struts and polymer-filled sealing rings that function to exclude the aneurysm sac. There are 14 F outer diameter aortic body and 13 F outer diameter iliac limb delivery systems. Ovation Prime’s polymer kit includes the PEG-based fill polymer, contrast agent and a buffer. PEG-based polymers are used in a variety of vascular and medical products. This low viscosity biocompatible polymer is designed to be soluble and non-embolic if accidentally spilled into the vasculature after mixing. In animal studies, fill polymer was injected directly into animal vasculature with no evidence of embolization, toxic response or end-organ abnormality at 30 days. ⁷ However, immediate-type PEG hypersensitivity has been reported, and a literature review documented 37 clinical cases of hypersensitivity related to PEG exposure via various routes.^8,9 Prior to use, the two valves on the fill polymer kit are opened, and the fill polymer is mixed. The fill syringe is then connected to the fill polymer injection port on the aortic body delivery system. The syringe plunger is inserted into the Autoinjector, which is locked in place. The Autoinjector applies controlled force to the syringe plunger to inject the fill polymer into the graft. Injection of the polymer may be associated with systemic exposure in rare circumstances, and this can result in an anaphylactic reaction. One year outcomes of the Ovation graft system for EVAR ⁵ demonstrated a technical success of 100%, with a 30-day major adverse event rate of 2.5%. At one year, the AAA-related and all-cause mortality rates were 0.6% and 2.5%, respectively, with a treatment success rate of 99.3% and no stent graft migration or type I, III or IV endoleaks reported. However, the study does mention a 76-year-old patient who developed an anaphylactoid response when the fill tube became disconnected, and the polymer was injected intravascularly. The procedure was completed, but the patient subsequently went on to develop multi-system organ failure (MSOF), required a subtotal colectomy and died on postoperative day 17 from the MSOF and disseminated intravascular coagulopathy.

Siani et al. ⁶ described a case report of anaphylactic reaction during Ovation stent graft implantation resulting in profound hypotension, but the patient recovered after resuscitation efforts and treatment for an anaphylactic reaction, and the procedure was successfully completed with exclusion of the aneurysm sac noted at the completion of the procedure and the patient suffered no long-term adverse events. The authors did not elaborate on the potential cause of the polymer leak. One additional case of an anaphylactic reaction during Ovation graft implantation was reported in the literature, where the endovascular procedure was abandoned without placement of the iliac limbs and the patient refused subsequent treatment for the aneurysm. ¹⁰

This review was undertaken to specifically attempt to shed light on the potential causes of polymer leak with the Ovation graft as well as increasing physician awareness of this complication and how to anticipate and treat it. While no root cause was established in a significant number of patients (11), a combination of likely factors may be involved (Table 2). In up to 10 patients, the initial manufacturing process may have played a role. In September of 2014, the company recognized that there was a possibility that the manufacturing process for loading the graft may have resulted in damage to the graft and fill channels, allowing for polymer leak. This led to the September 2014 Field Safety Corrective Action. Another contributing factor is ballooning prior to the polymer having enough time to cure (five patients). If the graft is ballooned less than 15 min post mix, the polymer may not have hardened sufficiently increasing the likelihood of polymer leak. When the patient’s body temperature is low (less than the tested range 35°C) (four patients), the polymer cure rate is slowed down by 1 min per degree Celsius. This increases the likelihood of ballooning prior to complete polymer cure. In addition, excessive manipulation of the device (four patients) prior to aortic body de-mating can potentially result in polymer leak. This could occur, for example, if a graft is deployed high and excessive force is applied to reposition the graft after the anchors have engaged the aortic wall.

Following several reports of anaphylactic reaction, the manufacturer made changes to the instructions for use (IFU) ¹¹ for Ovation including cautioning against ballooning of the graft prior to 20 min (compared to 15 min in the prior IFU) after polymer mix and waiting at least an additional minute per degree body temperature below 35°C before disconnection. Although it may seem intuitive that polymer leak with incomplete filling of the proximal sealing rings might result in a type IA endoleak, none of the patients in this series that underwent complete graft deployment were noted to have a type IA endoleak at the completion of the procedure. One of these patients had a type IA endoleak identified at one month follow-up and this resolved post re-intervention. The greater concern appears to be the three patients where the iliac limbs were not deployed. One of these patients did undergo re-intervention with placement of iliac limbs and successful exclusion of the aneurysm sac, but one patient was just monitored, possibly because of the refusal to undergo further treatment and one patient underwent conversion to open repair, which is a major undertaking and can be associated with significant morbidity.

Prompt recognition and treatment of this rare complication can restore the patient to a hemodynamically stable state allowing for completion of the procedure and avoiding need for re-intervention or leaving the patient at risk of aneurysm rupture. Some have suggested that the syringe should be loaded with only the exact amount of required polymer, and polymer infusion should be monitored under continuous fluoroscopy. ³ Awareness of the potential for polymer leak with this device by both the operating surgeon and the anesthesia team can allow for appropriate medications to be drawn up and ready for administration. In addition, we believe that following the manufacturer’s IFU guidelines and recognizing factors that increase the likelihood of polymer leak reduce the occurrence of this relatively rare complication.

Newer stent graft systems for the treatment of AAAs continue to become available on the market. Some devices include some form of polymer that fills a space and the polymer could potentially be exposed to the systemic circulation. It is unclear if exposure to any of these polymers could result in an anaphylactic reaction but being able to recognize the event and treat it promptly could avoid unnecessary intra-operative hemodynamic instability, potential complications and need for re-intervention.

Conclusion

Polymer leak and associated anaphylactic reaction with the Ovation stent graft system are quite rare, but awareness and early recognition of this complication can allow prompt treatment. It may be prudent to discuss this potential complication with patients as well as with members of the care team including the OR staff and anesthesiologist. In addition, several precautionary technical steps can potentially reduce the risk of this rare but serious complication.

The concern with delayed recognition and subsequent treatment of this complication is that the procedure would be aborted if the patient remains unstable and a re-intervention would be required to exclude the aneurysm sac. Some patients may decide not to undergo re-intervention, as was the case with one patient in this series, or may possibly be lost to follow-up. In one case in this series, the patient underwent conversion to open surgical repair, and it is unclear if this decision was made by the patient, physicians taking care of the patient, or both, and what factors led to this treatment plan.