The majority of patients following emergent AAA repair will remain intubated and sedated upon ICU transfer until they can demonstrate stable hemodynamics and no evidence of clinically significant bleeding that requires aggressive management. Extubation within the first 6 h of ICU arrival correlates with less nosocomial complications and shorter ICU stay and should be strongly considered if no contraindications exist [7]. Endovascular techniques are associated with less time on the ventilator due to less overall physiological insult as well as decreased requirement of sedation for pain control [29].

An ICU admission portable chest radiograph is helpful to both confirm placement and exclude complications from central lines placed emergently in the operating room and to identify potential barriers to early extubation. Most AAA patients have a history of tobacco abuse and are at risk for chronic obstructive pulmonary disease. Patients with clinically significant COPD may have evidence of hyperinflation (more than nine posterior ribs completely visible) or hyperlucency in the upper lung fields due to air trapping and upper lobe predominant emphysema most common in smoking-related lung disease. Chest radiograph can also identify pulmonary infiltrates that may suggest TRALI, TACO, or underlying left ventricular dysfunction and when present should prompt early and aggressive assessment of intravascular volume and examination for myocardial ischemia.

Assist-control ventilation is generally preferred initially to minimize patient work of breathing until hemodynamics stabilize. Typical starting tidal volume and respiratory rate are 6–8 ml/kg and 12–15, respectively, which should be adjusted based on initial pH and initial peak and plateau pressures measured on the ventilator. The vast majority of these patients will have a metabolic acidosis on initial ICU presentation, and management of this finding should focus on continued resuscitation with mechanical ventilation simply serving as an adjunct to stabilize pH and reduce the risk of arrhythmia [31, 32]. Elevated peak pressures with a low plateau pressure should prompt suctioning to clear the endotracheal tube and central airways of mucus plugging, followed by bronchodilator administration if persistent for likely underlying obstructive lung disease. Severe COPD patients may benefit from slower respiratory rates and low tidal volumes to reduce the risk of dynamic air trapping and barotrauma. Elevated peak and plateau pressures combined with diffuse infiltrates on chest radiograph are concerning for problems with lung compliance and should prompt a lung-protective ventilator strategy (tidal volumes of <6 cc/kg ideal body weight, titrated to keep the plateau pressure less than 30 cmH₂O, and consideration of higher levels of PEEP). FiO₂ should initially be set at 100 % and then weaned to a PaO₂ of 70 mmHg or greater, with an initial PEEP of 5 cmH₂O.

Attempts at ventilator liberation should begin as rapidly as possible once hemodynamics have stabilized and pain, bleeding, acidosis, and myocardial ischemia have been addressed. Patients who are alert enough to protect their airway and demonstrate an appropriate rapid shallow breathing index (RSBI, frequency/tidal volume in L) of <105 should be considered for a spontaneous breathing trial using either CPAP or T piece and extubated in 30 min if doing well. Multiple studies have shown that the prophylactic use of noninvasive positive pressure ventilation immediately post-extubation, especially in patients with COPD, can reduce the risk of reintubation [33].

Abdominal compartment syndrome (ACS) is one of the most feared complications among all surgical patients and particularly after emergent abdominal aortic surgery. All rAAA patients regardless of the method of repair should be considered to be at high risk for developing ACS, and we recommend routine focused monitoring postoperatively in this patient population. Although ACS is primarily thought of as a complication after open abdominal surgery, there is a significant risk of ACS even among patients who undergo endovascular repair. Epidemiologic studies have demonstrated an incidence of ACS in 30–50 % of patients after open rAAA repair and in up to 30 % of emergent endovascular repairs [34–36]. The etiology and causes of ACS after rAAA repair are multifactorial and have been related to the amount of fluid resuscitation, the presence and depth of presenting shock, the volume of retroperitoneal hematoma, the duration of ischemia, and the development of postoperative abdominal complications such as ischemic bowel [37]. Understanding these factors and the common causes of ACS allows the ICU physician to anticipate and potentially even prevent the development of ACS. Diagnostic clues to developing or frank ACS can range from subtle physiologic changes to complete cardiorespiratory collapse, and the key to avoiding unnecessary morbidity or death is always earlier recognition and intervention.

The most important diagnostic strategy is to appreciate the potential risk for developing ACS in an individual patient by identifying the presence and number of risk factors as described above. These patients should then be closely monitored for the early physiologic indicators of developing ACS combined with routine serial monitoring of intra-abdominal pressure. Early clues to developing intra-abdominal hypertension include progressive tachycardia, tachypnea (if spontaneously ventilating), decreasing urine output, and decreasing mean arterial pressures. Unfortunately these are all relatively nonspecific signs of ACS, but should prompt at least consideration of the diagnosis. More specific indicators of ACS include worsening abdominal distension and firmness on exam, steadily decreasing pulmonary compliance, decreasing mean arterial pressures (MAP), and sudden oliguria or anuria that is not responsive to volume expansion. It is important to also understand that the signs of decreasing pulmonary compliance will depend on the mode of ventilation, with rising peak or mean airway pressures on a volume-controlled mode or decreasing tidal volumes on a pressure-controlled mode. Early changes in pulmonary compliance may be less obvious in the patient on ventilator modes that automatically compensate for compliance changes (such as pressure release volume-controlled ventilation). Similarly, decreases in blood pressure may be less appreciated in the patient on vasopressor agents that are being titrated to a certain MAP, so the dose of vasopressor should also be followed closely. Systemic markers of perfusion or metabolic acidosis (lactate, base deficit) may be elevated late in the course of ACS, but they are not reliable early indicators. The best indicators of ACS include rising peak airway pressures, decreasing MAP and/or increased vasopressor requirement, and decreased urine output in the setting of increasing abdominal distension and abdominal pressures.

There are now published international consensus guidelines on ACS that have standardized the diagnostic criteria and provide evidence-based recommendations for interventions and therapeutic options [38, 39]. The preferred method of assessment of intra-abdominal pressure (IAP) is via bladder pressures obtained with the patient supine, relaxed, and measured at the midaxillary line. Intra-abdominal hypertension is defined as a sustained intra-abdominal pressure (IAP) ≥12 mmHg, while ACS is characterized by a sustained IAP ≥20 mmHg. Some have proposed that calculating an actual abdominal perfusion pressure (defined as MAP-IAP) is superior to the above definition, with a perfusion pressure <60 mmHg indicating ACS [39]. We recommend routine serial assessments of IAP via the bladder catheter in all patients after emergent repair of a rAAA and that these be continued until the patient is out of the early high-risk period (initial 2–3 days) and has no clinical signs of elevated IAP. If the patient develops elevated IAP, then initial interventions to present progression to ACS include diuresis or dialysis/hemofiltration for volume overload, a trial of intravenous paralytic agents, and assessment for any abdominal complications (such as hemorrhage, bowel ischemia) [38, 40]. There is also a select subgroup of patients that will develop ACS due to massive ascites, and these patients can often be treated successfully by large-volume paracentesis and either repeat paracentesis as needed or placement of a temporary drainage catheter for continuous evacuation of fluid.

Although some of these temporizing maneuvers can delay or even prevent progression to ACS, the majority of patients that develop true ACS will require an emergent decompressive laparotomy. The key technical steps to successful decompression are to widely open the skin and abdominal fascia, to perform a thorough exploration to identify any pathologic process underlying the ACS (such as bleeding, ischemic bowel, bowel obstruction with massive dilation), and to perform a temporary abdominal closure with enough laxity to avoid recurrent ACS. However open abdomen and temporary closure techniques are not without complications. There is a higher risk of infection, fistula formation, skin necrosis, and abdominal wall retraction with loss of domain following decompressive laparotomy. Mortality is also significantly higher among patients who develop ACS after ruptured AAA repair. In one recent series, mortality with ACS was 62 % among patients who had undergone open repair and was a strikingly high 83 % among the endovascular group [34]. The high rates of postoperative ACS and the associated morbidity/mortality have prompted some to propose prophylactically leaving the abdomen open at the time of initial open repair, with delayed closure performed once the patient is out of the high-risk time window. We would recommend strong consideration of this approach in any patient undergoing open repair who is requiring ongoing fluid and vasopressor resuscitation, with significant bowel distension/edema or with undue fascial tension or elevated airway pressures during fascial closure [37, 40, 41]. In the patient with elevated IAP that progresses after an endovascular repair, decompressive laparotomy should be considered before the development of ACS. The results of a study by Choi et al. are shown in Fig. 15.1 and display what factors most commonly weighed into a surgeon deciding to progress to decompressive laparotomy [40]. As the ACS in this patient population is often attributed to the large retroperitoneal hematoma that would otherwise be evacuated with an open approach, an alternative-described treatment modality is the placement of a percutaneous image-guided catheter into the hematoma and infusing thrombolytics to break up and evacuate the clot [42]. Finally, the decision to perform a decompressive laparotomy must also be made with consideration of the patient’s overall status and likelihood of survival, aligned with any known advanced directives and the wishes of the patient or their surrogate decision-makers.

High mortality is also associated with the development of critical limb ischemia following ruptured AAA repair. Development of acute limb ischemia can potentially be caused by multiple different etiologies including postoperative thromboembolic disease, prolonged preoperative and intraoperative ischemia, extremity compartment syndrome from reperfusion, or distal embolization of plaque or clot from aortic clamping and intraoperative manipulation. In addition, aortic manipulation can put the patient at risk for dislodgement of cholesterol particles with resultant cholesterol emboli. These differ from atherothrombotic emboli in that they are typically smaller and cause ischemia more distally in smaller vessels. Virtually any end organ may be affected by cholesterol emboli, and the true incidence is unknown. Physical signs of this phenomenon can go well beyond limb ischemia and may include fever, skin petechiae, and signs of end-organ damage such as renal azotemia, worsening respiratory distress, or even neurologic changes. The presence of livedo reticularis in this setting is strongly associated with cholesterol emboli syndrome. Blue toe syndrome or ischemia to distal extremities following both elective and emergent aortic surgery has also been associated with cholesterol emboli [51]. The importance of rapid identification and intervention for postoperative limb ischemia is well described in the available literature. In one series of 46 emergent ruptured AAA repairs (all done open), there was a 17 % incidence of postoperative critical limb ischemia [52]. These required a variety of interventions including attempts at operative repair, thrombolysis, or catheter embolectomy. Ultimately, 63 % of the patients who developed limb ischemia progressed to frank limb necrosis. The overall mortality due to critical limb ischemia for the entire cohort was 11 %, and among those who developed limb ischemia, the mortality rate was 83 % [52].

As with abdominal compartment syndrome, all patients who undergo emergent repair for a ruptured AAA should have routine postoperative monitoring done to identify any signs of impending or current limb ischemia. This can be complicated by the multiple factors in these patients that can compromise the reliability and accuracy of the physical exam of the extremities. These can include hypotension, vasopressors, preexisting vascular disease, edema, venous stasis changes, and obesity. It is critical for the operative team to assure that there is adequate distal flow to the extremities following the rAAA repair and to establish a new baseline extremity vascular exam that can be reliably compared to subsequent examinations for any significant change. The other key component is to then accurately communicate (and preferably demonstrate) the key parts of that exam to the ICU team who will be involved in the patient’s postoperative care. This should include documentation of both the pulse and Doppler exam for the femoral, popliteal, and all three pedal vessels to characterize which have a palpable pulse, which have only a Doppler signal, and which have neither. Subsequent vascular checks should be done on these patients hourly during the initial resuscitation and stabilization period and should preferably be done by the same person during each shift. There should be clear instructions to notify the ICU and operative team immediately for any significant change in the exam indicating worsening perfusion. This includes the loss of any palpable pulse, the change from palpable to Doppler signal only, and of most concern the complete loss of pulse and Doppler signals in any vessel. It is important to recognize developing limb ischemia as early as possible in order to initiate prompt interventions to improve or restore flow. As the development of limb ischemia is associated with a high mortality in addition to a high rate of limb loss, this monitoring and early intervention process can be lifesaving in addition to limbsaving.

Interventions for the development of limb ischemia can include any of the following alone or in combination: chemical anticoagulation, thrombolysis, catheter-based lysis or thrombectomy, infusion of vasodilators, endovascular angioplasty and/or stenting, operative thrombectomy/embolectomy, revision or repair of anastomoses, and operative revascularization of the affected extremity. A full description of all of these options is beyond the scope of this chapter, but meticulous attention to proper ICU care and management may minimize the time to identification and intervention of this postoperative complication or may even prevent it from developing. Avoidance of both over- and under-resuscitation can help reestablish steady perfusion and avoid repeat periods of relative low-flow or no-flow ischemia. Although correction of coagulopathy is often emphasized in the critically ill postoperative patient, there is a growing appreciation of the multiple phenotypes of coagulation abnormalities during and after shock, including a predisposition to thrombotic complications rather than hemorrhagic complications. Complete normalization of all coagulation parameters should not be used as a target in the non-bleeding patient after a major vascular reconstruction. The increasing use of thromboelastography (TEG) may allow for better appreciation of the current dynamic clotting and clot lysis function in an individual patient and avoid over- or undertreating coagulopathy based off of standard coagulation parameters such as the prothrombin time and partial thromboplastin time. If acute postoperative limb ischemia or threatened limb ischemia develops, then therapeutic anticoagulation with heparin should usually be started immediately for suspected thromboembolism. This may be effective alone in improving perfusion and effecting resolution of any partially obstructing thrombus, but if gross loss of tissue is impending or occurring, then immediate endovascular or operative intervention is typically necessary. In the event that tissue loss is too advanced or extremity revascularization is not possible or will not be tolerated due to the severity of illness, then an amputation should be considered. In rare cases with extremity necrosis in patients who are too unstable to tolerate any operative intervention, a temporizing “medical amputation” can be performed by placement of a proximal extremity tourniquet to occlude both inflow and outflow or by placing the extremity in dry ice until they are stable enough to tolerate a formal surgical amputation [53–55].

Extremity compartment syndrome deserves special mention in any discussion of the ICU and care and monitoring in this patient population. All post-op ruptured AAA patients are at risk for development of compartment syndrome of the lower extremities, and in addition to the limb concerns, it can impact multiple areas of their ICU care including the cardiovascular and renal systems. Patients at particularly high risk include those with prolonged shock, longer durations of preoperative and/or intraoperative limb ischemia (>4–6 h), lack of palpable pedal pulses after surgery, and those with concurrent venous thrombosis and decreased venous return. Regardless of the individual risk factors, all patients should be closely monitored for compartment syndrome following an emergent rAAA repair. The classic early signs of compartment syndrome including severe pain, sensory deficits, loss of toe/foot extension are frequently not reliable in this patient population as the exam is compromised by narcotics, sedation, and often mechanical ventilation. Serial examination of the calf compartments, preferably by the same person or groups of people, should be performed to detect increased swelling and tenseness. Soft compartments with an intact distal vascular exam do not require any further evaluation. If there is increasing concern for a compartment syndrome based on the physical exam, then either compartment pressures should be measured at the bedside or operative fasciotomies should be performed. Similar to abdominal compartment syndrome, compartment pressures above 15 mmHg should raise concern, and those above 20–25 mmHg should prompt intervention. Although several authors have proposed using a perfusion pressure (MAP-compartment pressure) to diagnose compartment syndrome, this has primarily been validated in isolated orthopedic trauma and not in the vascular or ICU patient population. Several others have examined the use of a cutaneous near-infrared spectroscopy monitor to monitor for compartment syndrome and have reported increased sensitivity and specificity when compared to standard clinical criteria and physical examination [56–58]. In addition to providing earlier warnings of impending compartment syndrome, this technology has the advantage of being noninvasive, portable, and continuous. The risk of postoperative extremity compartment syndrome is almost exclusively limited to the calf, although rarely thigh and gluteal compartment syndromes have been reported and should be considered for the patient with evidence of rhabdomyolysis and an unconcerning calf exam. For the patient with failed attempts at restoration of perfusion or irreversible major tissue loss, emergent amputation may be necessary and lifesaving. This is crucial in the setting of a post-op ruptured AAA patient as any increased metabolic demand such as tissue necrosis may be the inciting cause of additional cardiopulmonary stress with resultant major morbidity or mortality [52].

In addition to evaluation of the extremity, any suspected or proven compartment syndrome should prompt immediate assessment for rhabdomyolysis, myoglobinuria, and acute kidney injury with serial laboratory studies (serum CPK, urine myoglobin, and BUN/creatinine). Fluids should be titrated to maintain UOP of at least 30–50 cc/h, and if the CPK is greater than 5000, then we recommend increasing the goal UOP to 80–100 cc/h. Although mannitol and bicarbonate administration are commonly advocated adjuncts for rhabdomyolysis, there is no high-quality evidence demonstrating any benefit to these therapies above standard fluid resuscitation. We reserve these therapies for the patient with a CPK >10,000 and that is rising despite standard fluid resuscitation and positive urine myoglobin. The most common error we see in this area is underestimation of the amount of bicarbonate required to truly alkalinize the urine. We will typically give an immediate bolus of 1–2 ampules (50–100 meq) of sodium bicarbonate (NaHCO3), followed by a continuous infusion of D5W with 100–150 meq NaHCO3/l running at 100–150 cc/h. Confirmation of urine alkalization can be obtained with a simple bedside urine dipstick for pH. Intermittent mannitol boluses may also be given if the patient is failing to adequately respond to the measures above, but should not be given to the anuric patient.

The physiologic stress and relative hemodynamic instability that often accompanies the presentation of a ruptured AAA create a decidedly vulnerable environment for the adrenal gland. The incidence of adrenal insufficiency (AI) is approximately 30 % overall after ruptured AAA repair, and it is reported that up to 67 % of patients with unexplained postoperative hypotension have underlying AI. Table 15.3 highlights the systemic physiologic impact of AI on overall outcomes between those who have AI during recovery after rAAA and those who do not [59]. The stress of the vascular event and major surgery can increase cortisol production tenfold and cease the typical diurnal cycling of cortisol production. The robust blood supply to the adrenal gland is protective from supraceliac clamping; however, the combination of blood loss, mechanical interruption, and microvascular thrombosis or emboli make adrenal ischemia and AI a real possibility that should be considered in all patients [59]. Additional important factors that are critical to elucidate are any history of prior AI, current or recent use of steroid medications, prior adrenal surgery or radiation, and whether any medications have been administered that can interfere with adrenal glucocorticoid and/or mineralocorticoid production. One of the more commonly used medications that can suppress adrenocortical function is etomidate, a commonly used induction agent for rapid sequence intubation. Among patients with septic shock, the incidence of AI was found to be 76 % after etomidate administration versus 51 % with no etomidate [60]. Several other series in shock states (including hemorrhagic shock) have confirmed these findings and also suggest a possible adverse impact on survival [60–62]. However, others have challenged these findings, particularly with a single dose of etomidate [63, 64]. We prefer to avoid etomidate if possible in this patient population and also assume that some degree of adrenal insufficiency is likely in the post-op patient with hypotension despite adequate volume replacement and who received etomidate for intubation or during surgery.

It is well understood that cortisol production is intimately involved in the production and activation of catecholamines that regulate not only vascular tone but cardiac function as well. The classic clinical picture of AI is profound hypotension that is not responsive to intravenous volume expansion and pressor medications. Individuals with unidentified AI are therefore at risk of requiring a larger amount of fluid resuscitation, higher doses of vasopressor medications, and the expected resultant increase in complications, organ failures, and mortality [59]. However, these factors can be difficult to isolate and attribute to AI in the complicated post-op rAAA patient who is hypotensive and requiring large-volume resuscitation. The criteria for diagnosis of AI and for initiation of supplemental low-dose (or “stress dose”) steroids have undergone a number of changes and modifications over the past decade as new high-quality controlled data has become available [65–67]. Previous recommendations for diagnosing AI in critically ill patients focused on laboratory testing to evaluate serum cortisol levels, cortisol response to a corticotropin stimulation test, or both. Although these results have been shown to be predictive of outcomes, they have not been shown to be reliable for guiding initiation or continuation of therapy [67–69]. The most widely accepted current guidelines in critically ill patients come from the Surviving Sepsis Campaign (www.​survivingsepsis.​org), and although they focus on septic shock, their algorithms have been widely adopted among diverse ICU populations with shock from varying etiologies [70, 71]. For suspected AI, therapy should be initiated with low-dose hydrocortisone (200–300 mg/day intravenously) based on clinical assessment alone and should not rely on or be delayed for the results of laboratory testing of cortisol levels or a corticotropin (ACTH) stimulation test. Adrenal insufficiency should be suspected in all patients with hypotension despite adequate or ongoing volume expansion and requiring high dose or increasing doses of vasopressor medications and in the absence of another identified cause of the refractory shock. Empiric steroid therapy should be immediately initiated in these patients using hydrocortisone (or equivalent agent) at a dose of 50–100 mg every 8 h. Specific additional mineralocorticoid supplementation (typically with fludrocortisone) is not recommended unless there is some etiology or concern for severe mineralocorticoid deficiency that is being inadequately supplemented by the hydrocortisone. Exogenous steroid treatment should be implemented for the improved outcomes for these patients when hypotension remains unexplained after the first 24 h. Boluses of hydrocortisone are typically used initially with 100 mg immediately followed by 50–100 mg IV every 8 h. Hydrocortisone is typically preferred over dexamethasone or other formulations because of its added benefit of mineralocorticoid activity, although some intensivists advocate adding a specific mineralocorticoid agent (such as fludrocortisone).

Ischemic colitis is one of the most feared and morbid conditions or complications after abdominal aortic surgery and is particularly well described after both elective and emergent AAA repair. Although the etiology is often assumed to be simple interruption of the colon blood supply, typically in the inferior mesenteric artery distribution, the actual cause is frequently multifactorial. It can include preexisting atherosclerotic or thrombotic disease of the mesenteric vessels, anatomic variation or absence of collateral vessels to the colon, ligation or exclusion of the hypogastric arteries, hypotension with low-flow states, medications, vasopressor use, and infection. The risk of ischemic colitis is substantially higher in ruptured AAA patients in comparison to those undergoing elective repair and will also vary highly depending on the aggressiveness of screening. Clinically significant ischemic colitis is demonstrated in approximately 5 % of elective open AAA repairs, with an increase to 35 % in emergent ruptured AAA repairs [72, 73]. When routine or aggressive endoscopic surveillance is employed, up to 65 % of patients will have some evidence of ischemic colitis after open repair of a ruptured AAA [72, 74]. Much less data is available on ischemic colitis after endovascular repair, but several series have described an incidence of 1.4–1.7 % for all endovascular AAA repairs [75, 76]. The incidence in EVAR for rAAA is undoubtedly higher, but has not been well characterized. Of interest, the presence of atheroemboli as the primary source of colonic ischemia is much more common following endovascular repair and carries an overall poor prognosis. The presentation and clinical significance of postoperative ischemic colitis exists along a broad spectrum, from relatively asymptomatic disease limited to the mucosa and identified on endoscopy to full-thickness colon necrosis with perforation. However, there is a clear and significant increase in overall mortality among patients with clinically evident ischemic colitis to 40–60 % and up to 90 % in the presence of necrosis with perforation [77, 78].