HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Himanshu J. Patel Yoshikazu Suzuki G. Michael Deeb Permission Requests Google Scholar Articles by Patel, H. J. Articles by Deeb, G. M. PubMed Articles by Patel, H. J. Articles by Deeb, G. M. Related Collections Great vessels Peripheral vascular

J Thorac Cardiovasc Surg 2008;135:1288-1296
© 2008 The American Association for Thoracic Surgery

Surgery for Acquired Cardiovascular Disease

Operative delay for peripheral malperfusion syndrome in acute type A aortic dissection: A long-term analysis

^a Department of Surgery, University of Michigan Hospitals, Ann Arbor, Mich
^b Department of Radiology, University of Michigan Hospitals, Ann Arbor, Mich

Received for publication June 21, 2007; revisions received December 27, 2007; accepted for publication January 29, 2008.

^_* Address for reprints: Himanshu J. Patel, MD, Assistant Professor of Surgery, Section of Cardiac Surgery, CVC Room 5144, 1500 E. Medical Center Drive SPC 5864, Ann Arbor, MI 48109-5864. (Email: hjpatel{at}med.umich.edu).

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Background: We previously reported an improvement in early mortality for patients presenting with acute type A dissection with malperfusion using a strategy of initial percutaneous intervention to restore end-organ perfusion and delayed operative repair after resolution of the malperfusion syndrome. This study evaluates the late outcomes with this approach.

Methods: A total of 196 patients were admitted with acute type A dissection (1997–2007). Seventy patients with ischemic end-organ dysfunction underwent percutaneous fenestration or branch vessel stenting. Operative therapy was planned after resolution of the reperfusion injury. Outcomes were compared for patients with (MP) and without (UC) dissection with ischemic end-organ dysfunction.

Results: The mean age of the patients was 57.1 years, and 173 patients underwent operative repair (n = 126 UC group; n = 47 MP group). The remaining 23 patients in the MP group died before repair from complications of malperfusion (11) or aortic rupture (12) while awaiting resolution of the malperfusion syndrome. Operative mortality was seen in 9.2% of all patients (9.5% in UC group vs 8.5% in MP group; P = 1.0). On analysis of the entire cohort (n = 196), the mean survival was higher for the uncomplicated group (95.9 months for UC group vs 53.7 months for MP group; P < .001). A subgroup analysis of patients who underwent operation (n = 173) revealed similar mean survival (95.9 months for UC group vs 80.5 months for MP group; P = .45).

Conclusion: A strategy of immediate reperfusion, stabilization, and planned operative repair for acute type A dissection with malperfusion still carries a significant risk for early and late mortality. However, those patients who survive the initial malperfusion and undergo repair have a similar operative and late survival when compared with those patients presenting with uncomplicated dissection.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Aortic dissection remains the most frequent and lethal complication of thoracic aortic disease. Although several groups, including the multicenter International Registry of Acute Dissection, have demonstrated improving outcomes with operative therapy, malperfusion syndrome associated with acute type A dissection remains a significant adverse risk factor for survival.^1-3 Reported in-hospital mortality rates for patients presenting with malperfusion with end-organ dysfunction have varied from 4% to 90%.^4-10

In 1997, we described our early results for a strategy of operative delay for those patients presenting with acute type A dissection, malperfusion, and ischemic end-organ dysfunction.⁴ In that study, a historical cohort of patients presenting with similar ischemic end-organ dysfunction from malperfusion taken directly for open repair were compared with a cohort managed with initial percutaneous fenestration, selective branch vessel stenting, and delayed operative repair after resolution of the reperfusion injury. In contrast with an in-hospital mortality of 89% in the historical group, those patients who underwent operative delay had an overall 25% mortality, including a 15% mortality from rupture. These early results led to a consistent strategy at our institution of restoring end-organ perfusion before operation for all those patients presenting with acute type A dissection with significant end-organ ischemia from malperfusion. The current study reports the long-term results of this approach and describes the outcomes for the single largest cohort of patients with acute type A dissection, malperfusion, and ischemic end-organ dysfunction reported in the literature.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
This study was approved by the University of Michigan Hospitals Institutional Review Board (no. 2003–0128). Informed consent requirements were waived for this study.

We performed a retrospective analysis of data from all patients who were admitted to the University of Michigan Hospitals with a diagnosis of acute type A aortic dissection from 1997 to April of 2007. Acute dissection for this study was defined as its occurrence within 14 days of presentation to the hospital. A combination of clinic and hospital records, imaging studies, and query of the National Death Index was used to obtain long-term information. A total of 196 consecutive patients who were evaluated for operative repair constituted the study cohort. Diagnosis of acute type A dissection was made by the combination of clinical factors, including history and physical examination, and either the use of dynamic computed tomography (CT) or transesophageal echocardiography.

Acute type A dissection without severe ischemic end-organ dysfunction (UC) was suspected in 126 patients, and they proceeded directly to operative repair. This group included patients without evidence of angiographic or clinical malperfusion, and patients who had angiographic malperfusion but did not display evidence of ischemic end-organ dysfunction (eg, absent femoral pulses but adequate sensation and motor function of legs, elevated creatinine with adequate perfusion on CT scan, or compromised true lumen on CT scan without clinical evidence of ischemia). In these patients, the standard operative procedure consisted of a median sternotomy with the rapid institution of cardiopulmonary bypass. An alpha-stat pH strategy was used on bypass. The extent of proximal resection generally included native aortic valve preservation and resuspension, unless contraindicated by the presence of root aneurysm (4.5 cm) or leaflet pathology. In patients who had a dissection present in the arch, resection of the proximal hemiarch under deep hypothermic circulatory arrest was initially performed as a standard procedure. Extended arch resection was reserved for those patients in whom the entry tear was identified within the arch, or if an arch aneurysm of 4.5 to 5.0 cm or more was identified. With increasing experience with arch resection during the course of the study period, extended arch repair was also preferentially performed in younger patients (<65 years) or patients with connective tissue disease in the presence of dissected aorta alone. The ascending aorta was resected in all patients. Deep hypothermic circulatory arrest was used as a surgical adjunct in all patients in whom the dissection extended into the distal ascending aorta or beyond. Neuroprotective adjuncts during circulatory arrest consisted of retrograde cerebral perfusion only or a combination of antegrade and retrograde cerebral perfusion depending on the operating surgeon preference. However, in all cases of extended arch resection, antegrade cerebral perfusion was used to prevent prolonged cerebral ischemic times. In addition, the head was packed in ice in all patients during the period of cerebral ischemia to prevent rewarming during longer periods of circulatory arrest. In these situations, cerebral ischemic times were calculated as the period of time that the patient did not receive antegrade cerebral flow (ie, either straight hypothermic circulatory arrest or hypothermic circulatory arrest with retrograde cerebral perfusion time). Additional pharmacologic adjuncts included use of intravenous thiopental (1 g), methylprednisolone sodium succinate (Solu-Medrol; Pfizer Inc, New York, NY) (1 g), and mannitol (25 g) administered during the cooling phase when the bladder temperature reached 19°C.

On admission to the hospital, 70 patients (35.7%) had suspected neurologic (stroke or spinal cord ischemia), visceral (abdominal pain or tenderness, acidosis, hyperamylasemia, elevated transaminases, or diarrhea), renal (elevated creatinine and CT evidence of malperfusion), or limb (absent pulses with limb compromise) malperfusion with end-organ dysfunction (MP). This group underwent immediate angiography with percutaneous fenestration and aortic true lumen stenting with or without branch vessel stenting where appropriate. Operative therapy was then planned after resolution of malperfusion and its associated reperfusion syndrome. It is important to note that the decision to delay operative therapy for acute type A dissection for the reasons of malperfusion was only made after full evaluation of the entire clinical picture, incorporating information from the patient's history, physical examination results, and laboratory and imaging data. The mere presence of imaging data suggesting branch vessel compromise in the absence of either symptomatic or physical findings and abnormal laboratory evaluation results did not lead to a delay in operation. The details of the initial clinical scenario prompting delay in aortic repair are listed in Table 1. Presentation in cardiogenic shock was also an indication to proceed directly for operative repair of the type A dissection, regardless of the presence of potential branch vessel compromise. The subsequent intraoperative strategy for the entire group of patients who underwent a delay strategy, including the need for root or extended arch aorta resection or use of deep hypothermic arrest, was similar to that used in the UC group. An outcomes analysis on an intent-to-treat basis was performed, comparing patients presenting with uncomplicated (UC, n = 126) versus suspected malperfusion (MP, n = 70) acute dissection. Follow-up was 100% complete at a mean of 39 ± 39 months.

Data were analyzed using the Statistical Package for the Social Sciences (SPSS Inc, Chicago, Ill). All data are expressed as mean ± standard deviation where applicable. Dichotomous variables were evaluated using chi-square analysis; continuous variables were evaluated using analysis of variance (1-way analysis of variance). Multivariate models (logistic regression for dichotomous variables) were constructed using a backward conditional process to identify factors that were independently associated with each of the outcomes of interest. The factors used in multivariate analysis for both the entire cohort and the operated subgroup included using those with P .1 significance on univariate analysis. Survival analysis was analyzed by Kaplan–Meier methods.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
The mean age of the study cohort was 57.1 ± 14.2 years (72.4% were male). The comorbidities are listed in Table 2 and were similar between groups. Nine patients (5.2%) presented in cardiogenic shock from tamponade, and 22 patients (11.2%) had previous cardiac surgery. Of the 70 patients presenting with suspected malperfusion and undergoing an initial percutaneous approach, 23 died before attempted operative repair. While awaiting resolution of the malperfusion syndrome, 12 of these patients had sudden hemodynamic collapse and death, consistent with aortic false lumen rupture. The remaining 11 patients died before operation from complications from malperfusion syndrome (multisystem organ failure [n = 5] and dense neurologic deficit without recovery [n = 6]). The remaining 47 patients underwent operative repair.

The overall late mortality rate was also significantly lower for the uncomplicated group (UC n = 31 [24.6%] vs MP n = 36 [51.4%]; P < .0001). By univariate analysis, vital status at last follow-up correlated with age (P = .01), preoperative ejection fraction (P = .004), a history of stroke (P = .004), presentation in cardiogenic shock (P = .009), or suspected peripheral malperfusion syndrome (P < .0001). However, on multivariate analysis, only age (P = .009), preoperative ejection fraction (P = .041), or presentation in cardiogenic shock (P = .023) were independently associated with long-term mortality. Although there was no independent effect of suspected peripheral malperfusion on the overall long-term mortality rate, a survival curve analysis demonstrated important time-dependent effects of malperfusion syndrome. This survival analysis is displayed in Figure 1 and demonstrates a significant decrease in survival for the suspected malperfusion group (P < .0001). As is evident from the curves, the predominant difference in mortality is in the early time period (<3 months). The curves remain relatively parallel thereafter.

View larger version (14K): [in this window] [in a new window]   Figure 1. Entire cohort survival distribution function. This graph demonstrates that presentation with acute dissection with malperfusion and ischemic end-organ dysfunction is an important adverse risk factor for long-term survival. Mean survival times were higher in the uncomplicated group (UC 95.9 months vs MP 53.7 months; P < .001). Note that these curves are relatively parallel beyond the first 3 months, thus emphasizing the predominant early effects on survival in patients with malperfusion.

Operated Subgroup Analysis
A separate analysis was conducted comparing outcomes for all those who underwent operative repair. This cohort consisted of the 126 patients undergoing immediate repair, as well as the 47 (of 70 total) patients presenting with suspected malperfusion who survived to operation. Operative therapy was delayed by a median of 4 days for the operated MP subgroup. The preoperative comorbidities, procedural details, and postoperative complications are listed in Table 4. Of note, 4 patients were thought to have renal malperfusion postoperatively and underwent intervention. Of this group, 3 patients required dialysis, and 1 patient (with a prolonged postoperative course) died of upper gastrointestinal hemorrhage.

View larger version (14K): [in this window] [in a new window]   Figure 2. Survival distribution function for operated subgroup. For those patients surviving to operative repair, survival is similar for both patients initially presenting with uncomplicated acute dissection and those initially presenting with malperfusion and end-organ ischemia syndrome (UC 95.9 months vs MP 80.5 months; P = .45).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Traditional therapy for acute type A dissection entails immediate operative repair to prevent rupture and death. Cambria and colleagues⁵ and Lauterbach and colleagues⁶ called to attention the high rates of morbidity and mortality associated with malperfusion syndrome, particularly for the subgroups with renal or mesenteric malperfusion. They suggested that the optimal approach in that setting consisted of initial operative revascularization with subsequent central aortic repair. In contrast, Fann and colleagues⁷ have suggested that immediate aortic repair with proximal false lumen obliteration and reestablishment of true lumen flow may restore peripheral arterial perfusion without the need for associated peripheral arterial revascularization. This group also noted the incidence of mortality to be significantly elevated (>50%) in those subgroups with renal and mesenteric ischemia. Others have also suggested similar outcomes with preferential aortic repair.⁸

In our first study in 1997, we described a comparative analysis of patients treated with immediate operation regardless of peripheral vascular involvement and patients treated with a selective operative delay.⁴ That study suggested that there was a significantly higher risk of mortality (89% vs delay 25%, P = .003) with immediate central aortic repair in those patients with malperfusion syndrome and ischemic end-organ dysfunction. On the basis of this previous work, our algorithm during the last decade for managing patients with acute type A dissection is to proceed with immediate operative repair unless there is evidence of ischemic end-organ dysfunction and malperfusion. This group includes those with 1) no CT angiographic evidence of malperfusion; 2) CT angiographic malperfusion without severe end-organ dysfunction; 3) absent peripheral pulses with intact sensation and motor function of the affected limb; 4) evidence of elevated creatinine without CT evidence of malperfusion; and 5) evidence of a compromised true lumen and no clinical evidence of end-organ ischemia and dysfunction. In contrast, those patients who presented with evidence of neurologic, visceral, renal, or limb malperfusion with end-organ damage proceeded directly for percutaneous intervention with aortic fenestration, true lumen, or branch vessel stenting where appropriate. The purpose of this report was not to ascertain the more suitable strategy for patients presenting with malperfusion (ie, immediate repair for all vs selective operative delay). We believe that early operative repair is indicated in the majority of patients to decrease the risk of dissection-related mortality. Rather, this study was undertaken to describe long-term results associated with a consistent application of a selective operative delay approach in those patients presenting with significant ischemic end-organ dysfunction as a consequence of prolonged branch vessel compromise. In doing so, we assessed the outcomes for the largest group of patients with type A dissection with malperfusion and ischemic end-organ dysfunction reported in the literature. In addition, although most previous studies have focused on early complication rates, few have determined the long-term outcomes for this cohort.¹¹

The long-term results from this study suggest that malperfusion syndrome is an important adverse risk factor for long-term survival. The importance of mesenteric malperfusion syndrome as an independent adverse risk factor for early and late mortality has been noted in this study and by others.^4-8 The early mortality rate of 40.5% for patients with mesenteric malperfusion in our report is consistent with the contemporary results reported by Lauterbach and colleagues,⁶ who also espouse operative delay for this subset of patients with acute type A dissection.

In contrast, the long-term results in our study differ when the analysis is conducted on those who underwent operative repair. In this subgroup, the survival curves of patients with malperfusion who eventually proceeded to operative repair were restored to the level of those who presented with uncomplicated dissection. Although this is admittedly a highly selected group, it suggests the effects of malperfusion syndrome are primarily seen early. Indeed, in the survival analysis of the entire cohort, the predominant difference in survival is seen in the first 3 months with relatively parallel survival curves thereafter, further supporting the notion of predominantly early effects of malperfusion.

The recent report by Geirsson and colleagues¹¹ from the University of Pennsylvania deserves mention. They suggested that although an approach espousing early repair is associated with a higher in-hospital mortality for the group with malperfusion (30.5% vs 6.2% in non-malperfusion), long-term survival may be similar between groups. Although their data suggest an early mortality similar to that of our strategy of operative delay for malperfusion (38.7% for all malperfusion in this report), their group had a significantly different profile than in our study. In particular, although the prevalence of both renal and mesenteric malperfusion with end-organ dysfunction was high in our study (57.1% and 52.1%, respectively), their report had a significantly lower incidence of malperfusion of these beds (renal = 4.1%; mesenteric = 1.4%). Because we and others have suggested a lower survival for these groups, the results from their study may not be suitable for comparison with our data.

A final important point should be noted. Because of our previous work with acute dissection and malperfusion, we have noted a dramatic increase in referrals of patients presenting with this complication. Geographic considerations and referral patterns often result in patients presenting to other institutions (1) before transfer to the University of Michigan. These transfers increase the time interval between the onset of dissection with malperfusion to the time of definitive treatment and worsen the degree of ischemic end-organ dysfunction. These considerations may explain some differences between other series and ours, and may also account for the relatively high proportion of patients with malperfusion seen in our study compared with others.

Limitations of this study include the lack of a comparison group (ie, a cohort undergoing immediate operation regardless of malperfusion status). However, this was not the intent of our study; rather, the focus was to analyze the long-term results of an adopted strategy of operative delay for acute dissection and ischemic end-organ dysfunction. Operative delay while awaiting resolution of the malperfusion syndrome in these patients is associated with a significant risk for early mortality from either aortic false lumen rupture or complications of the malperfusion syndrome. A future study could therefore evaluate whether earlier operative therapy (ie, operation immediately after fenestration and stenting) could improve results by ameliorating the risk for early rupture and subsequent mortality. It is possible that the beneficial effects of avoidance of rupture may be offset by a higher operative mortality in the setting of significant ischemia-reperfusion injury and its deleterious consequences.⁴

Conclusions
Presentation with acute dissection, malperfusion, and ischemic end-organ dysfunction is an important adverse risk factor for survival, particularly in the setting of mesenteric ischemia. A strategy of immediate percutaneous reperfusion and delayed operative repair to allow resolution of malperfusion and associated reperfusion injury still carries a significant risk for early mortality. However, in those patients who undergo operative repair with this approach, survival times are restored to those who present without evidence of ischemic end-organ dysfunction.

	Footnotes

 
Presented at the 33rd Annual Meeting of the Western Thoracic Surgical Association June 27–30, 2007, Santa Ana Pueblo, New Mexico.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Ehrlich MP, Ergin MA, McCullough JN, et al. Results of immediate surgical treatment of all acute type A dissections. Circulation 2000;102(Suppl III):248-252.
   
2. Bavaria JE, Pochettino A, Brinster DR, et al. New paradigms and improved results for the surgical treatment of acute type A dissection. Ann Surg 2001;234:336-343.[Medline]
   
3. Trimarchi S, Nienaber CA, Rampoldi V, et al. Contemporary results of surgery in acute type A aortic dissection: the International Registry of Acute Aortic Dissection experience. J Thorac Cardiovasc Surg 2005;129:112-122.[Abstract/Free Full Text]
   
4. Deeb GM, Williams DM, Quint LE, et al. Surgical delay for acute type A dissection with malperfusion. Ann Thorac Surg 1997;64:1669-1677.[Abstract/Free Full Text]
   
5. Cambria RP, Brewster DC, Gertler J, et al. Vascular complications associated with spontaneous aortic dissection. J Vasc Surg 1988;7:199-209.[Medline]
   
6. Lauterbach SR, Cambria RP, Brewster DC, et al. Contemporary management of aortic branch compromise resulting from acute aortic dissection. J Vasc Surg 2001;33:1185-1192.[Medline]
   
7. Fann JI, Sarris GE, Mitchell RS, et al. Treatment of patients with aortic dissection presenting with peripheral vascular complications. Ann Surg 1990;212:705-713.[Medline]
   
8. Girardi LN, Krieger KH, Lee LY, Mack CA, Tortolani AJ, Isom OW. Management strategies for type A dissection complicated by peripheral vascular malperfusion. Ann Thorac Surg 2004;77:1309-1314.[Abstract/Free Full Text]
   
9. Okita Y, Takamoto S, Ando M, Morota T, Kawashima Y. Surgical strategies in managing organ malperfusion as a complication of aortic dissection. Eur J Cardiothorac Surg 1995;9:242-247.[Abstract]
   
10. Slonim SM, Miller DC, Mitchell RS, Semba CP, Razavi MK, Dake, MD. Percutaneous balloon fenestration and stenting for life-threatening ischemic complications in patients with acute aortic dissection. J Thorac Cardiovasc Surg 1999;117:1118-1127.[Abstract/Free Full Text]
    
11. Geirsson A, Szeto WA, Pochettino A, et al. Significance of malperfusion syndromes prior to contemporary surgical repair for acute type A dissection: outcomes and need for additional revascularization. Eur J Cardiothorac Surg 2007;32:255-262.[Abstract/Free Full Text]
    

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Himanshu J. Patel Yoshikazu Suzuki G. Michael Deeb Permission Requests Google Scholar Articles by Patel, H. J. Articles by Deeb, G. M. PubMed Articles by Patel, H. J. Articles by Deeb, G. M. Related Collections Great vessels Peripheral vascular