The effect of postoperative malperfusion after surgical treatment of type A acute aortic dissection on early and mid-term survival

Introduction

Acute type A aortic dissection (AAAD) is a life-threatening condition and one of the most challenging diseases faced by cardiothoracic surgeons. Despite preventative measures including early surgical referrals for patients, preoperative care and improved surgical techniques, in-hospital mortality following surgery remains high, ranging from 10% to 30%.^[1,2]

Malperfusion of systemic organs is a complication of aortic dissection caused by branch-vessel involvement. Occurrences can result in dangerous end-organ ischemic dysfunctions, especially when involving the brain. Clinical diagnoses are critical to the development of effective treatment strategies. Proper diagnoses also have important influence on immediate and long-term outcomes of treatment.^[3-5]

Malperfusion following either type A or type B acute aortic dissection, is a fairly common complication. It can involve several arterial regions including the cerebral, renal, or mesenteric segments, as well as the upper or lower limbs. In the Stanford Classification Type A Dissection (De Bakey Classification Types I and II) surgical treatment can prevent this complication. In the event of malperfusion, distal aortic percutaneous fenestration can allow blood flow to return to the true lumen. Alternatively, when the distal aortic true lumen is completely obstructed, an endovascular stent can be inserted to recanalize the lower arteries (i.e. the celiac trunk, renal, superior mesenteric and iliac arteries). The International Registry of Acute Aortic Dissection (IRAD) does not include malperfusion as an independent predictor of mortality. Nonetheless, several studies have supported the relevance of this complication, given that it increased in-hospital mortality and adversely affected late survival.^[6-9]

The aim of this study was to evaluate the effect of postoperative malperfusion (PM) on in-hospital mortality and long-term survival in patients undergoing surgery for AAAD in a single, high-volume aortic surgery center.

Methods

Between January 2005 and December 2015, 227 patients (mean age 62.5 ± 12.6 years) underwent emergent operations for AAAD. The study was approved by the local Institutional Review Board, which waived the need for patient consent. The preoperative patients’ characteristics are given in Table 1.

The diagnosis of malperfusion was based on clinical symptoms and/or imaging evidence, i.e. absence of organ perfusion as determined by computed tomography (CT) scan angiography.

Malperfusion was classified as: cerebral if there was presence of a focal or global stroke leading to brain function deterioration that persisted more than 24 h, or a transient ischemic attack; renal if there was an impairment of renal function (e.g. anuria requiring continuous veno-venous hemofiltration, or a two-fold increase of creatinine serum level); or mesenteric if there was evidence of tense abdominal or intestinal dysfunction, or increased serum levels of liver and/or pancreatic enzymes. The database queries were completely obtained from 214 patients.

Results

At presentation, 119 (55.6%) patients exhibited preoperative malperfusion: 68 (31.8%) were cerebral with neurological symptoms, 38 (17.7%) were renal, and 13 (6.1%) were mesenteric. The most relevant clinical characteristics of the cohort, stratified according to the presence or absence of PM, were reported in Table 1.

Intraoperative mortality occurred in 14 (6.5%) patients and in-hospital mortality after surgery occurred in 48 (22.4%) patients. Using univariate analyses, significant risk factors for in-hospital mortality included: being at least 75 years old, having a body mass index of more than 30 kg/m², having preoperative overall or cerebral malperfusion, having longer cardiopulmonary bypass and aortic clamping times, needing concomitant coronary artery bypass grafting (CABG), having renal PM requiring continuous veno-venous hemofiltration, and have postoperative mechanical ventilation for up to 24 h. The multivariate analyses revealed that being 75 years old or older at the time of surgery (P < 0.001) and having renal PM (P < 0.01) were independent predictors of in-hospital mortality [Table 3].

Fifty-five (25.7%) patients showed clinical symptoms and/or imaging evidence of PM. In 42 cases, only 1 organ system was affected, including: the brain in 13 (6.5%) patients, the kidneys in 22 (10.3%) patients, and the viscera in 7 (3.3%) patients. In a subgroup of PM patients with just 1 affected organ system, the mortality rate was 40.5%. In 11 patients, PM occurred in 2 organ systems (5 in the brain and kidneys, 3 in the brain and viscera, 3 in the kidneys and viscera); the mortality rate in this subgroup was 63.6%. PM involving all 3 organ systems occurred in just 2 cases and both patients died. Overall, the in-hospital mortality rate in patients affected by any PM was 47.3% (26/55) vs. 22.6% (36/159) in patients without PM (P < 0.001). Univariate analyses identified that the risk factors for any PM included: having a preoperative LVEF of less than 40% (P < 0.01), having preoperative renal dysfunction (P < 0.05) and having an entry tear distal to the ascending aorta or to the proximal aortic arch (P < 0.05). Multivariate analyses determined that having a LVEF value of 40% or less was the only independent predictor of having a PM (P < 0.05) [Table 4].

Preoperative cerebral malperfusion was an independent predictor of cerebral PM [odds ratio (OR): 2.5, 95% confidence interval (CI): 1.0-6.1, P < 0.05]. Interestingly, a LVEF of less than 40% was only a found to be a significant risk factor for renal PM when using univariate analysis techniques (P < 0.0001). Finally, an entry tear distal to the ascending aorta or to the proximal aortic arch requiring extensive repair and longer surgical time was a risk factor of mesenteric PM (P < 0.05, using univariate analyses).

Follow-up results

The mean duration of follow-up was 42.4 ± 23.7 months (median 46 months). All patients were followed until the end of the study period. One- and 5-year overall survival rates were 96.0 ± 1.6% and 90.8 ± 3.2%, respectively [Figure 1]. Cox regression analyses identified that independent predictors of long-term survival were: being at least 75 years old at the time of surgery (OR: 3.7, 95% CI: 0.9-14.0, P = 0.05) and having a renal PM (OR: 28.6, 95% CI: 1.8-462.0, P = 0.01) [Table 5]. When the survival probability was dichotomized by age, (with a threshold of 75 years old at the time of the surgery), the 5-year survival rates were 91.6 ± 3.5% for patients < 75 years old and 65.1 ± 19.5% for patients ≥ 75 years old (P < 0.05). The 5-year survival rate for patients without PM was 93.9 ± 3.4% vs. 78.6 ± 7.8% for those affected by PM (Log-rank test, P < 0.01).

Table 5

Predictors of late mortality

Variable	Univariate	Multivariate
	P value	OR	95% CI	P value
Age ≥ 75 years	0.03	3.7	0.9-14.0	0.05
Renal PM	< 0.0001	28.9	1.8-462.0	0.017
Cerebral PM	0.001			NS
Visceral PM	0.09			NS
Any PM	0.001			NS
Entry tear (distal vs. proximal) of the dissection	0.02			NS

PM: postoperative malperfusion; OR: odds ratio; CI: confidence interval; NS: not significant

Figure 1. Overall late survival

Discussion

Despite improvements in medical management and surgical techniques, acute type A aortic dissections still have high mortality and morbidity rates.^[1,2] The IRAD revealed that the expected mortality rate for patients undergoing AAAD surgery ranges from 20% to 30%.^[9] Our cardiac surgery division has extensive experience in the treatment of acute aortic dissection; we observed an in-hospital mortality rate of 29%. Several studies show that both patient characteristics and multi-organ involvement (i.e. affecting the brain, kidney, and mesenteric organs) play a key role in immediate and post-procedural outcomes. For example, Caus et al.^[12] showed that being at least 70 years old at the time of operation was an independent predictor of worsened outcomes for AAAD treated patients; these authors reported a 5-year survival rate of 30%. These data are similar to the clinical experience reported here, in which patients who were at least 75 years old had lower actuarial survival rates than patients under 75 years of age (65.1 ± 19.5% vs. 91.6 ± 3.5%).

Malperfusion of organ systems remains a severe condition that is frequently associated with adverse outcomes in AAAD patients undergoing surgical procedures. Data from the German Registry for AAAD suggested that the number of organs involved in the malperfusion was associated with immediate outcomes of surgery. In fact, outcomes were substantially worsened in the presence of any type of malperfusion syndrome, which was exacerbated with increased numbers of affected organs. A 12.6% early mortality rate was observed in the absence of malperfusion versus 43.4% mortality in patients with three organ systems affected by malperfusion.^[13] Here, preoperative clinical symptoms and/or imaging evidence of malperfusion occurred in 119 patients. After surgery, 55 (25.7%) patients had malperfusion syndrome. In-hospital mortality was significantly higher (47.3%, 26 patients) in this group compared to patients without PM (22.6%, 36 patients) (P < 0.0001). Despite the small sample size per group, a strong association between the number of malperfused organs and early mortality was observed. In fact, when 2 organs were affected operative mortality elevated to greater than 60%.

Furthermore, these data show that PM-affected patients had lower survival probabilities when compared to those who did not develop this postoperative complication (78.6 ± 7.8% vs. 93.9 ± 3.4%). Correlations between the preoperative presence of malperfusion and mortality have been previously described.^[14,15] Pacini et al.^[14] found that patients presenting with any malperfusion syndrome had a mortality rate of 43.7%, compared to 15% in patients without malperfusion (P = 0.001); strikingly, mortality rates were 34.7%, 61.9% and 85.7% with involvement of 1, 2, or more than 2 organ malperfusions, respectively. Mesenteric malperfusion was identified as an independent predictor of operative mortality. Similarly, Geirsson et al.^[15] reported a 30.5% operative mortality in the presence of any malperfusion syndrome; in this study cerebral malperfusion was detected as a risk factor for in-hospital mortality (P < 0.001) and reduced long-term survival (P = 0.0002).

In the present study, the most important independent risk factor of early and 5-year mortality was presence of a renal PM requiring continuous veno-venous hemofiltration. Previously, we identified in 100 consecutive patients receiving AAAD operations from 1995 to 2006, that renal failure, either chronic (OR: 0.3, P = 0.04) or developed acutely in the postoperative period (OR: 8.9, P = 0.001), was a predictor of operative mortality. However, renal failure was not a predictor of reduced 5-year survival.^[8] In the same group of patients, preoperative LVEF values of less than 50% were also predictors of reduced survival (P = 0.02).

Another important issue is the surgical timing of aortic repairs. Previous authors have suggested delaying acute aortic dissection surgeries when patients experience preoperative malperfusion, particularly in the mesentery. This delayed treatment strategy involved early endovascular treatment with a complete or partial resolution of organ ischemia, followed by timely aortic surgeries.^[16,17] While this management approach may be beneficial in a specific subpopulation, patient AAAD survival outcomes have been shown to relate closely to the length of time between diagnoses and surgeries.^[18,19] Given the high mortality of patients with mesenteric malperfusion (40-100%), initial management with an interventional procedure treating the condition should be considered.^[20,21] In fact, previous data suggested that mesenteric malperfusion was associated with the highest mortality rates when compared to malperfusions occurring in any other organ systems. The surgical strategy presented here, consisting of immediate aortic dissection treatment, showed that incidence of preoperative malperfusion was reduced roughly in half; from 56% preoperatively to 25% in the immediate postoperative period.

Univariate analyses of preoperative variables determined that three risk factors predicted the occurrence of a PM in any organ system. These risk factors were: having a LVEF less than 40%, having renal impairment that required continuous hemofiltration, and having an entry tear distal to the ascending aorta or the proximal aortic arch. However, the only variable that maintained significance in the multivariate model was having a preoperative LVEF of less than 40%. Reduced ejection fraction likely associated with concomitant ischemic coronary disease, which could have increased the risk of a postoperative low cardiac function and subsequent PM. Juxtaposition of intimal tears distal to the ascending aorta or the proximal arch were non-significant factors in the multivariate analyses. However, these factors contributed risk to progression of aortic disease and PM. Patients with a primary entry tear in the descending aorta were at the highest risk of PM. These patients probably required additional extensive repairs compared to patients with primary entry tears in the ascending aorta. Some of these high-risk patients may benefit from a "frozen elephant trunk" procedure to address the entire pathology.^[22]

Analysis of preoperative variables contributing risk for each type of PM revealed that only one variable independently predicted cerebral PM: preoperative cerebral malperfusion (OR: 2.5, 95% CI: 1.0-6.1, P < 0.05). Shortening the length of time between onset of cerebral symptoms and dissected aortic surgery was critical for improved outcomes in this subset of patients. Estrera et al.^[23] reported improved outcomes in AAAD patients who underwent cardiac surgeries within 10 h of neurological symptom onsets.

With regard to arterial cannulation sites, some authors have suggested that cannulation of the axillary artery will ensure better brain protection during surgery. However, the experience reported here did not confirm this evidence. Nonetheless, many surgeons still limit the extent of surgery to the ascending aorta, even though limited repair has a higher probability of re-intervention on the remaining aortic segments at a later date. The primary aim of performing AAAD is as an emergent, life-saving procedure. If a center is only able to perform a limited repair technique, but still saves the life of the patient, then the primary intention of the procedure has been achieved.^[24,25]

In this study, no independent predictors of renal and mesenteric PM were identified. However, using univariate analyses, having a LVEF value less than 40% was statistically relevant (P < 0.0001). Additionally, having entry tears distal to ascending aortic segments that required extensive repairs and longer surgical times was also recognized as a significant risk factor (P < 0.05).

This study had several limitations. First, it was a retrospective analysis of an experience at a single institution. Second, preoperative treatments to address organ malperfusions were not performed. Third, the possible effects of revascularization strategies for the treatment of PM were not explored. Revascularization techniques may improve long-term outcomes.

In conclusion, PM is a severe condition that is frequently associated with adverse immediate and long-term outcomes in surgical AAAD patients. At this institution, the incidence of PM after AAAD surgery was noteworthy, occurring in roughly 10% of patients. AAAD surgical procedures effectively reduced preoperative malperfusions in about half of cases. In fact, repairs to the ascending aorta and proximal arch, as well as removal of primary tears, significantly increased the true lumen flow and allowed treatment of a majority of malperfusion syndromes, including those in the cerebral, mesenteric, and renal systems. Postoperative malperfusion, especially involving the kidneys, was associated with high in-hospital mortality and reduced long-term survival. There was no evidence that the types of surgical techniques undertaken, the sites of cannulation, or the use of more complex interventions (requiring circulatory arrest during cardiopulmonary bypass) were risk factors contributing to PM.