Endovascular repair of ruptured aneurysm continues to evolve since the advent of endovascular repair of abdominal aortic aneurysm (EVAR) in the 1990s. Prior to 1994 ruptured aneurysm mortality was 45%–60%. There has been an increasing trend to treat ruptures endovascularly with evidence of improved outcomes and reduced length of hospital stays.
In our experience time is of the essence and time to seal is imperative for best possible outcomes. Randomized trials have suggested no difference in long-term mortality. Additionally, more than half of patients presenting with ruptures may not be anatomically amenable to EVAR.
The first step is to develop established protocols and pathways to manage acute aortic emergencies. It is important to have a multidisciplinary center equipped with the necessary skills and armamentarium to be able to handle difficult situations and complex care in the preoperative to postoperative periods in order to minimize morbidity and mortality. Hospitals that do not have the device inventory or staffing required for the urgency of a ruptured aneurysm may consider transferring the patient to a specialty center equipped to deal with magnitude using a multidisciplinary and standardized approach.
Time is of the essence. This presents a major obstacle that is sometimes out of the control of the surgeons. Hinchliffe et al. showed that computer tomography does not delay the time to the operating room (OR) compared with open to endovascular repair. On the contrary, the earlier French randomized trial suggested the delayed time to surgery averaged 1.3 hours for open versus 2.9 hours for EVAR. Our own institution has an average time to OR of 10–35 minutes with an average time of 20 minutes. Lloyd et al. showed that more than 88% of patients died 2 hours after diagnosing a ruptured aneurysm. An ideal management protocol for rupture fast tracks the patient to the CT scanner. Arguably, unstable patients can be taken directly to the OR and a treatment decision made after the initial aortogram; however, alternative and less deadly diagnosis should always be entertained. Additionally, an endovascular first approach allows for endovascular aortic control with an occlusion balloon. If the anatomy is not amenable to endovascular repair, the endovascular aortic control proves to be invaluable during the time required for aortic dissection and clamping.
Permissive hypotension or hypotensive hemostasis is important until aortic control is achieved and should be emphasized to limit rerupture and ongoing hemorrhage.
Arrival in the operating room is usually met with much disarray, and it is important to be prepared with a variety of endografts to allow for rapid deployment. Adjuncts, such as sheaths, wires, and catheters, should already be open on the field as well as the concomitant equipment for possible open repair. We advocate using an endograft with which the operator is familiar and comfortable, to decrease the chance of maldeployment and to decrease operative time. The IMPROVE trial showed that local anesthesia for the procedure may offer mortality benefit. Our practice is to prepare and to drape the patient, and even to establish aortic balloon for control, prior to induction of anesthesia.
Rapid percutaneous femoral access is followed by the placement of a stiff wire into the descending thoracic aorta followed by a 12 French introducer sheath, and an occlusion balloon as described by Li et al. is appropriate for unstable ruptures. In unstable patients, there may not be time for placement of closure devices prior to sheath placement, and the femoral arteries can be exposed and primarily repaired at the end of the case. It is not uncommon for iliac access vessel anatomy to be challenging. A stiff wire can be used to straighten tortuous iliac arteries, and there are a few techniques to cross small or heavily diseased iliac access arteries. Balloon angioplasty to create a passage for the device is usually the first option. Unusually small and calcified iliac vessels are the most problematic. The so-called “pave and crack” technique may allow the easier passage of stent graft after the initial lining of the iliac with balloon expandable covered stent. This will allow the endograft to seal within the covered stent. Access vessel injury is usually identified after the endograft is in place and the sheaths are being removed. Injury to the iliac bifurcation usually occurs as the large sheath is retracted through the external and the common iliac. Classically, resistance and then complete lack of resistance is felt as the sheath tears the external iliac off the common iliac artery at the origin of the hypogastric. Concurrently, a drop in blood pressure will be seen when the retroperitoneal hemorrhage occurs. Most importantly, wire access must be maintained at all times so that an occlusion balloon can be placed. A covered stent graft is used to bridge the gap and seal the hemorrhage. Again, the key is to maintain wire access so that an occlusion balloon can be placed and a stent graft can be used to bridge the gap. If there is the possibility of an iliac injury, placement of an occlusion balloon into the aorta from the contralateral side and imaging of the iliac as the sheath is removed may prevent significant blood loss. Fortunately, endovascular device technology has created lower profile devices and more flexible sheaths, making access vessel rupture a rarity ( Fig. 13.1 ).
Once access is obtained and a stiff wire platform is established, a key maneuver is to place the sheath as far as possible into the aorta to support an occlusion balloon for unstable patients. This prevents downward displacement from improved blood pressure from the resuscitative efforts. The sheath needs to be left in place during the stent graft deployment so that the occlusion balloon can be safely withdrawn. The use of an exchange length wire for occlusion balloon support cannot be overemphasized as much as the fact that access can never be lost. The exchange length Lunderquist wire (Cook Medical, Bloomington, Indiana) will support the balloon in the descending aorta and limit the need for wire exchanges. Another option is to place the occlusion balloon from an upper extremity access to avoid downward displacement; however, the required sheath size increases the risk of upper extremity arterial injury. Disadvantages of a large brachial sheath are (1) the cut-down can deny a potential access point for anesthesia, (2) stiff wire around unknown arch anatomy may cause unnecessary trauma, and (3) the extended arm can impede imaging.
Great care should be taken to avoid the over-inflation of the compliant occlusion balloon, which may cause secondary rupture. Continued unstable hemodynamics after inflation of the occlusion balloon should trigger troubleshooting the occlusion balloon and may provide insight to either under-inflation or over-inflation with secondary rupture. The under-inflated balloon is recognized by mobility with each cardiac cycle. Additionally, confirming balloon apposition to the aortic sidewalls can confirm adequate inflation. When the balloon ruptures, it will obviously not be possible to instill any more half-and-half saline and dye solution, and blood may be retrieved from the port. The balloon should be inflated to the point that the balloon does not extend past the markers on its catheter. If rupture of the segment with the occlusion balloon occurs, additional coverage is necessary with a thoracic stent graft and reinflation of the balloon within the stent graft. Most balloons come with a stopcock in the package to maintain the desired pressure and to free the operator’s hands. The balloon shaft is then securely clamped to the drapes and inspected at intervals to prevent migration. Once the occlusion balloon is appropriately placed, anesthesia may start aggressive resuscitation with blood and blood products to limit coagulopathy postoperatively ( Fig. 13.2 ).
A useful tip and time-saving step is to use the sheath as an injection port with hand injections of contrast. The quality of the images can be adequate to allow the decision for EVAR versus open and angiographic locating of the lowest renal without needing to take the time for setting up power injection equipment. If flow is inadequate, the occlusion balloon can be gently deflated before repeat injection. Multiple earlier studies suggest 20%–47% of cases of ruptured abdominal aorta are amenable to off-the-shelf EVAR within the manufacturer’s indications for use. With the advancing stent graft technology and increasing outcome data for outside indications for use, it is estimated that EVAR can be used to treat up to 80% of ruptures with the application of adjunctive techniques.
Contralateral groin access is achieved and percutaneous methods can be used for larger sheath insertion. A second stiff wire is advanced into the aorta followed by a larger sheath and the endograft main body. If an occlusion balloon is used, it may be necessary to deflate it partially in order to pass the wire to the thoracic aorta. The graft is usually oriented so that the iliac limbs are crossed. Attention must be paid to the contralateral sheath as it is inserted and its relation to the ipsilateral sheath in order to identify where the contralateral limb should open so that cannulation of the contralateral limb is improved. Cannulation can be time-consuming if not performed correctly. Once the device is in the appropriate position, the occlusion balloon must be retracted into the sac to prevent impinging and entrapment, which can be deleterious. Some degree of hypotension is likely with deflation. Making your staff aware and prepared is a key step in allowing the rapid conduct of the case. One approach is to deploy the main body fully and to insert the occlusion balloon into the aortic neck from the main body side. A second option is to continue cannulating the gate and to proceed in the traditional fashion. A useful tip in large aneurysms is to leave the wire in the suprarenal aorta, because it allows for maintained conformation of the aortoiliac anatomy, and then double access the introducer sheath with a glide wire for cannulation of the gate. Another useful tip is to have a second aortic occlusion balloon primed and ready for use. Reintroduction of a previously inflated balloon into the smaller sheaths can be difficult. During the gate cannulation process, the occlusion balloon needs to be monitored to prevent distal migration from improving blood pressure. If the balloon is pushed down, this may cause endograft main body migration and loss of the seal zone. If this occurs, a cuff can be placed to improve the seal.
Gate cannulation is the most feared cause for interruption of the case progress; however, the increased number of elective endograft placements has made most operators more capable than in the past. It is important to have a variety of catheter shapes and tip conformation to assist with cannulation. The contralateral limb can often be cannulated without problem. However, if there is difficulty, a snare can be used to cannulate the gate. This can be done using the up-and-over technique with the wire, placing the snare at the base of the contralateral limb, and using multiple angles to place the wire in the body of the snare. If a snare is needed, it is preferable to finish the ipsilateral iliac stent graft deployment so that at least one aspect of the repair is completed. Alternatively, the gate can be crossed from a brachial puncture and then snared in the iliac. If unable to proceed with gate cannulation despite adjuncts, the bifurcated device can be converted to aorto-uni-iliac configuration by either deploying an aortic cuff over the gate orifice or simply deploying a second main body inside the first main body. The contralateral common iliac artery is then embolized and a crossover bypass is performed. Most gates are usually cannulating in 5 minutes. For unstable ruptures, time of cannulation remains of the essence, and if not feasible with the options described, consideration for conversion to aorto-uni-iliac should be rapid. Reichart suggested that 16% of gate cannulations will take longer than 20 minutes. Early in our experience, 15% of our patients required an aorto-uni endograft, but currently the vast majority can be treated with standard bifurcated devices.
Another important clinical issue is whether bifurcated or aorto-uni endografts are better. AJAX randomized trial used only aorta-uni endografts with the advantage of an easy learning curve and rapid control of abdominal hemorrhage; however, this adds time for the femoral–femoral crossover and increases risk of infection. Additionally, general anesthesia is required to perform a bilateral groin cut-down in an emergency setting. It has also been found that bilateral groin incision may increase pain, therefore increasing pain requirements and contributing to postoperative respiratory issues. Fewer stent-graft resources are needed to support a rupture for aorto-uni devices. The most common device needed is usually the largest. In 2002, Verhoeven et al. showed that bifurcated devices could be deployed under local anesthesia. Post hoc analysis of the IMPROVE trial suggested that bifurcated devices offer a survival advantage.
Once the graft of choice is deployed and the proximal, distal, and connecting segments are ballooned, an aortogram is performed with specific emphasis on endoleak, kinking, or clot formation. For aortic necks outside of those indicated for use or persistent type Ia endoleak, special adjuncts may be useful to create a seal. Aptus endoanchor (Aptus Endosystem, Sunnyvale, California) has been found useful by delivering a stapling device to the site of leak, and it also only requires a 16-French sheath from the main body access site. Additionally, a Palmaz stent (Cordis Endovascular, Miami, Florida) may be used as a last-ditch effort to create a seal. Imaging is of key importance in decision making for open versus endovascular repair, but conversion rates are still 1%–2%. In a last-ditch effort for unstable patients with Type Ia endoleak, the risk and benefit of coverage of a low renal orifice with or without an attempt to snorkel and stent the renal will need to be assessed. For patients with end-stage renal failure, the renal artery orifices can be covered with impunity. Type Ib endoleaks are usually managed with extension to the external iliac with concomitant coiling or plugging the ipsilateral hypogastric artery. Care is taken to prevent coverage of both the hypogastric arteries with the concomitant coverage of patent inferior mesenteric artery. If this occurs, the importance of ensuring the patency of the superior mesenteric artery and celiac artery cannot be overemphasized in order to prevent bowel ischemia. A large persistent endoleak may require a conversion to open repair. Conversion to open is not without consequences, because conversion has higher mortality in an already morbid situation.
Thrombosis of the access vessels secondary to the presence of large-bore sheaths is a concern, especially because use of heparin is variable among surgeons. With ruptures, it is generally our practice not to anticoagulate. Unstable patients with massive transfusion and supraceliac occlusion are usually in a coagulopathic state, and anticoagulation contributes to the ongoing bleeding and worsening abdominal compartment pressures. It is not unreasonable to administer heparin and it is a common practice among many groups, especially after aortic control. There is the opposing school of thought that an occluded femoral can be thrombectomized at the end of the case, rather than to contribute to the continued massive exsanguination from the lumbar vessels and raw surfaces of the autodissected retroperitoneum with heparinization.
Access-site complications are relatively simple to control and to treat. For patients in whom closure devices were not placed prior to sheath insertion, cut-down and repair is safest. Heavily diseased or stenotic femoral arteries occasionally require some type of revascularization after sheath removal. In these situations, simplicity and expediency are best in order to get the patient out of the operating room. Most repairs usually involve only a patch angioplasty or a local endarterectomy.
Significant clinical challenge can occur in the postoperative period. Bowel ischemia after elective EVAR is less than 1%, compared with 6.4% after ruptures. The risk factors for this, apart from age and coverage of the hypogastric artery, are not controllable. The combination of inferior mesenteric artery coverage and systemic hypotension is enough to cause colonic ischemia. If undetected, this has been shown to increase mortality significantly. Bowel movements intraoperatively, or even in the immediate perioperative time, should raise suspicions of colonic ischemia. It is our practice that all ruptured aneurysms undergo sigmoidoscopy at the bedside postoperatively on day 1, or sooner, based on clinical suspicion. Treatment for colonic ischemia is based on the degree of ischemia. Mild ischemia is treated with IV antibiotics and bowel rest, and more severe forms with bowel resection. Bowel ischemia involving the small intestine can prove to be more difficult to diagnose. The combination of preexisting atherosclerosis and hypotension can lead to infarction of bowel in the celiac and superior mesenteric distributions. Use of occlusion balloons can also predispose to embolization or ischemia to the mesenteric vessels. Exploratory laparotomy is often the only way to diagnose the problem, and treatment usually involves bowel resection and, occasionally, arterial reconstruction.
Compared with open repair, EVAR has a higher abdominal compartment syndrome rate, affecting up to 20% of the ruptured abdominal aneurysms. Large retroperitoneal hematoma with accompanying space-occupying effect, large volume fluid resuscitation, and the associated soft tissue edema result in high abdominal compartment pressures. These high pressures can result in reduced cardiac venous return, respiratory, renal, and cardiac dysfunction. Acute abdominal compartment syndrome is not only associated with increased mortality, but delayed diagnosis results in 100% mortality. Coagulopathy, use of occlusion balloon, conversion to aorta-uni device, and massive transfusion are all associated with increased risk of abdominal compartment syndrome. Diagnosis is usually made by a combination of physical and physiological findings. The abdomen becomes distended and tense on examination. Pressures required for adequate tidal volumes on the ventilator increase and urine output decreases. Measuring bladder pressures may give some insight into the abdominal compartment pressure situation, but the diagnosis is usually based on clinical suspicion and a variety of subtle signs. Treatment involves laparotomy and open abdominal dressing without disruption of the retroperitoneum. This will require multiple OR trips for closure. Entry into the retroperitoneal hematoma is to be avoided.
Irrespective of the modality used for ruptured aneurysm, acute renal failure is more pronounced, accounting for up to 40% postoperatively. Use of larger doses of contrast coupled with hemodynamic instability contributes to acute renal impairments. The United States national inpatient sample database review of the years 2005–2012 suggests doubled mortality and incidence of acute renal failure with EVARs compared with open repair. Additionally, ruptured EVAR is estimated to require 2 to 3 times the amount of contrast compared with routine EVAR. Aggressive and timely resuscitation is the only tool we have against contrast-induced nephropathy, because appropriate imaging is critical to appropriate sealing of the rupture. Use of intravascular ultrasound and carbon dioxide angiography in ruptured aneurysms has not been explored in the literature. The issue with using alternative imaging is that it may slow down the procedure in a patient with ongoing bleeding. The authors believe that contrast use should be judicious, but as much contrast as is necessary should be used to allow for necessary fluoroscopic images.
The IMPROVE trial reintervention after ruptured aneurysm showed the midterm interventions to be as much as twice as high for rupture. The majority of these were for endoleaks. Our personal experience was different and suggested that Type II endoleaks were less common and Type 1 endoleaks were similar to elective EVAR. In any event, these are mostly treated with translumbar or transfemoral routes in elective fashion. The bigger picture is that even with the late finding of an endoleak, patients survive the acute event and can now be treated in a more elective and controlled setting.
Unpublished data from our institution have demonstrated that endografts placed for abdominal aortic aneurysm rupture are at a higher risk of infection. In this series, the microbiology was skin flora compared with the endografts placed for nonruptured aneurysms, which grew a wide variety of organisms. This may be the result of preoperative antibiotics not being given during less than ideal skin preparation or device handling.