Outcomes of Medical Management of Acute Type B Aortic Dissection
Background— Currently, the optimal treatment of acute type B aortic dissection remains controversial. The purpose of this study was to report early clinical outcomes of medical management for acute type B aortic dissection.
Methods and Results— Between January 2001 and March 2005, 129 consecutive patients with the confirmed diagnosis of acute type B aortic dissection were studied. Mean age was 61 years (range, 29 to 94), with 33.3% (43/129) female. Acute type B aortic dissection protocol was instituted with the intent to manage all patients medically. Indications for surgical intervention included rupture, aortic expansion, malperfusion, and intractable pain. All patients were followed-up after discharge. Hospital mortality was 10.1% (13/129), 19% (4/21) when vascular intervention was required, and 8.3% (9/108) when medical management was maintained. Early intervention was required in 21 cases (16.2%), 19 (14.7%) open vascular/aortic cases and 2 cases (1.6%) of percutaneous aortic interventions. Morbidity included rupture (4.7%), stroke (4.7%), paraplegia (8.5%), bowel ischemia (7%), acute renal failure (21%), dialysis requirement (13%), and peripheral ischemia (4.7%). Late vascular-related procedures were performed in 5.2% (6/116) of cases. Univariate risk factors for early mortality were rupture (P<0.0001), need for laparotomy (P<0.008), acute renal failure (P<0.0001), need for dialysis (P<0.0001), and lower extremity ischemia (P<0.0004). The only independent risk factors for hospital mortality by multiple logistic regression was rupture (P<0.0009), and independent risk factors for midterm death were history of chronic obstructive pulmonary disease (P<0.002) and low glomerular filtration rate (<57 mL/min; P<0.0001).
Conclusions— Medical management for acute type B aortic dissection is associated acceptable outcomes. Outcomes of other management strategies, eg, endovascular stenting, for acute type B aortic dissection need to be compared with these results.
- aortic dissection
- cerebral infarction
- cerebral ischemia
Currently, the primary treatment of acute descending thoracic aortic dissection (Stanford type B) remains medical. Most of the guiding tenets for the management of acute aortic dissection were derived from the early experiences of Wheat and Austen suggesting better outcomes with medical management.1,2 The Stanford group further classified acute aortic dissection, provided recommendations for treatment, and reported long-term outcome data.3–5 Generally, surgical intervention for acute type B aortic dissection has been reserved for complications such as aneurysmal expansion, end organ malperfusion, and failure of medical management with refractory pain. Recently, however, it has been suggested that certain patients may benefit from earlier surgical intervention, ie, thoracic aortic replacement, to eliminate the potential for late complications with the goal of improving long-term survival.6
Because current outcomes of medical management of acute type B aortic dissection still remain significant with early mortality ranging from 10% to 12%,7–9 in addition to the fact that open surgical repair is associated with significant mortality and morbidity, interest in endovascular approaches is growing. For these reasons, the optimal initial management of acute type B aortic dissection has become debatable. Thus, the purpose of this study was to report both early and intermediate clinical outcomes for acute type B aortic dissection with initial medical management.
Between January 2001 and March 2005, we prospectively collected information on 129 consecutive patients with the confirmed diagnosis of acute type B aortic dissection. Analysis was retrospective. Mean patient age was 61 years (range, 29 to 94), with 33.3% (43/129) women. Diagnosis was made using computed tomographic angiograms in 98% (126/129), transesophageal echocardiography in 95% (122/129), magnetic resonance imaging angiography in 5% (6/129), and contrast angiography in 3% (4/129).
Aortic dissection was classified according to the Stanford classification as type B if the dissection did not involve the ascending aorta. The dissection was considered acute if it presented within 2 weeks of the initial onset of symptoms, eg, pain. All cases of classic dissection, ie, dissecting membrane with some degree of patency of both the true and false lumens, and intramural hematoma were included. Classic dissection was noted in 92% (119/129) with isolated intramural hematoma in 10 cases (8%).
On admission, an acute aortic dissection protocol was initiated for all patients with the intent to manage these patients medically. This protocol included intensive care unit admission with continuous arterial pressure monitoring, central venous access for administration of intravenous antihypertensive medications, and urine output monitoring via a bladder catheter. The initial goals were to halt the progression of dissection by decreasing the Δp/Δt or “impulse force” and to control the pain. β-Blockers, calcium-channel blockers, nitroglycerin, and sodium nitroprusside were used, in that respective order, to provide anti-impulse therapy with the goal of keeping systolic blood pressure <120 mm Hg and mean arterial pressure <80 mm Hg. The patient remained in the cardiovascular intensive care unit until blood pressure and pain were controlled with the patient also having been transitioned from intravenous to oral antihypertensive medications. Aggressive pulmonary therapy, deep venous thrombosis prophylaxis, nutritional support, and patient mobilization were undertaken. Deep venous thrombosis prophylaxis is obtained using low-molecular-weight heparins or sequential compression stockings when low-molecular-weight heparins were contraindicated. The presence of acute dissection did not contraindicate anticoagulation. The patient was continually monitored for pain and blood pressure control and well as change in clinical condition. For the purposes of the study, acceptable reduction in blood pressure was defined as systolic blood pressure <140 mm Hg. If pain persists, however, then a systolic blood pressure of <120 mm Hg becomes the intended pressure.
Indications for vascular surgical intervention included rupture, aortic expansion (aortic diameter >5 cm), retrograde dissection into the ascending aorta, malperfusion (visceral, peripheral), and intractable pain despite optimal medical management. Patients who did not require surgical intervention were discharged when blood pressure and pain were controlled on an oral regimen. Imaging using computed tomography scan is obtained in all patients before discharge to establish a baseline. All patients were followed-up after discharge with serial clinical and radiographic examinations at 6 weeks, 3 months, 6 months, and yearly thereafter. Continued surveillance was maintained through direct patient or referring physician contact.
In-hospital mortality referred to death that occurred during hospitalization. Life-threatening morbidity was defined as any neurologic (stroke, intracranial hemorrhage, and paraplegia), pulmonary (respiratory failure), cardiac (myocardial infarction), visceral (ischemia, necrosis), renal (failure), or peripheral complications. Chronic obstructive pulmonary disease was defined by a history of chronic bronchitis and emphysema or, while on bronchodilators, <60% of predicted forced expired volume in 1 second. A serum creatinine level >2.0 mg/dL or the need for dialysis defined renal dysfunction. Glomerular filtration rate was calculated using the Cockcroft-Gault formula.
Data collection and analysis were approved by the University of Texas Houston Medical School Committee for the Protection of Human Subjects. Analysis was retrospective. Survival was ascertained by direct patient contact (telephone or letter) and by searching the social security death index. Data were collected from chart reviews by a trained nurse abstractor and were entered into a dedicated Microsoft Access database. Data were exported to SAS for analysis, and all computations were performed using SAS version 6.12 running under Windows 2000. Patients were followed until death or until follow-up reached the study end date (March 31, 2005). Surviving patients were right-censored from the denominator when their maximum follow-up time was reached or on the date they were lost to follow-up. Univariate risk factor effects on survival were evaluated using the product-limit method of Kaplan and Meier. Continuous variables were stratified by quartile. Hypothesis tests of homogeneity over strata were computed using the log-rank test. Adjusted effects of risk factors on survival were evaluated using Cox proportional-hazards regression analysis, using stepwise and best-subsets model selection techniques. Continuous variables were left continuous for the Cox analyses. The null hypothesis was rejected at P<0.05.
Statement of Responsibility
The authors had full access to the data and take full responsibility for their integrity. All authors have read and agree to the manuscript as written.
Overall hospital mortality was 10.1% (13/129), 19% (4/21) when vascular intervention was required during the initial hospitalization, and 8.3% (9/108) if medical management was maintained. Six patients (4.7%) presented with rupture and had an associated mortality of 66% (4/6). The 2 survivors were the ones that were able to undergo surgical repair of the descending thoracic aorta. Variables at presentation were listed in Table 1.
Proximal extent of the dissection flap was at level of the innominate artery in 0.8% (1/129), left common carotid in 4.7% (6/129), left subclavian artery in 22.5% (29/129), just distal to left subclavian artery in 60% (78/129), distal descending thoracic aorta in 1.6% (2/129), and abdominal aorta in 1.6% (2/129)
With regards to the medical regimen, 98% (127/129) of patients required at least one intravenous antihypertensive drug during hospitalization with most (80%, 103/129) requiring a multidrug regimen. Table 2 lists the class of intravenous antihypertensive drug used and the duration required to achieve the acceptable reduction in arterial blood pressure (<140 mm Hg) until transitioned to oral medications. Median time to achieve the blood pressure of <140 mm Hg from the time of admission was 48 hours (range, 0 to 720 hours) and median time to control primary pain from time of admission was 48 hours (range, 0 to 264 hours). All patients were discharged on oral antihypertensive medications. Mean hospital length of stay was 15 days (1 to 88 days) and mean intensive care unit stay was 8 days (1 to 58 days).
Life-threatening morbidity was most commonly the result of organ malperfusion and occurred in 47.5% of all patients who presented with acute type B aortic dissection. Patient morbidities are listed in Table 3. Neurological complications occurred in 15.5% of patients (20/129). The 6 patients who experienced stroke were not among those who developed retrograde type A aortic dissection (2 patients). The etiology of cerebral hemorrhage during acute type B aortic dissection was unclear. Two cases were described as intraparenchymal and 1 case as subarachnoid hemorrhage. It was uncertain whether these were the result of previous trauma or the result of hemorrhagic conversion. Paraplegia occurred in 8.5% (11/129) of patients with most (9/11) involving bilateral lower extremities. Interestingly, 55% of the cases of paraplegia (6/11) resolved within 24 hours of presentation. Of note, early in our experience, cerebrospinal fluid drainage catheter was inserted in 2 patients who presented with paraplegia with no improvement in neurologic function.
Of the 6 patients who presented with lower extremity ischemia (6/129, 4.7%), 50% (3/6) of these required a surgical revascularization procedure, with the other one-half of patients noting resolution of peripheral arterial ischemia.
Early vascular intervention (occurring during primary hospitalization) was required in 21 cases (16.3%): 19 (14.7%) open vascular cases and 2 cases (1.6%) of percutaneous aortic interventions. In 16 of the 21 (76%) vascular procedures, open aortic procedures were performed (Table 4). Two cases of retrograde dissection occurred with subsequent replacement of the ascending aorta under profound hypothermic circulatory arrest. Table 5 lists noncardiovascular procedures performed.
Median follow-up was 18.5 months (range, 0 to 54.7 months). One-year and 4-year survival was 81.6% and 72.3%, respectively. Intermediate vascular related procedures were performed in 5.2% (6/116) of cases (5 aortic, 1 peripheral vascular). All the aortic related procedures were performed within 6 months of discharge, with 4 cases because of expansion and 1 case because of recurrence of pain.
Univariate risk factors for hospital mortality were listed in Table 6. The only independent risk factor by multiple logistic regression was rupture (P>0.0009) (Table 7). Independent risk factors for intermediate-term mortality were a history of chronic obstructive pulmonary disease (P<0.002) and low glomerular filtration rate (P<0.0001) (Table 8).
What remains consistent with acute type B aortic dissection is the fact that the mortality persisted at &10% in our series. This remains similar to other series reporting on acute type B aortic dissection including the multicenter International Registry of Acute Dissection.7–12 Predictors of early mortality by univariate analysis were rupture, development of renal failure, peripheral ischemia, and the need for laparotomy (most often to exclude visceral ischemia) with rupture as the only independent predictor of early mortality. This series essentially identified rupture and end organ malperfusion as risk factors, similar to International Registry of Acute Dissection that identified the “deadly triad” of hypotension, organ malperfusion, and lack of chest or abdominal pain as the independent predictors of early death.8,9 Interestingly, intermediate-term survival was related to intrinsic medical conditions of the patient, ie, chronic obstructive pulmonary disease and renal insufficiency (as defined by reduced glomerular filtration rate) and not to late complications of aortic dissection requiring intervention. This may be the result of aggressive follow-up and surgical intervention as indications arose.
The results and observations of this study underscore the unpredictable nature of acute aortic dissection. Early mortality and morbidity are primarily the result of expansion with rupture or end organ malperfusion. However, as noted previously by Wheat, rupture is relatively rare at initial presentation.2 This was a consistent observation in this series with an incidence of rupture at presentation of 4.7% with associated mortality of 66% (4/6). Because aortic rupture is less common at the initial presentation, end-organ of malperfusion is likely the more common cause of significant morbidity during the acute presentation of type B aortic dissection. This is presumably the result of malperfusion leading to thrombosis, ischemic-reperfusion injury, or a systemic inflammatory response syndrome. Unpredictable variability in blood flow patterns occurring in the dissected aorta further complicates acute aortic dissection. Interestingly, the changes in aortic blood flow patterns also allow for resolution of malperfusion as depicted in this series with spontaneous resolution of paraplegia and peripheral ischemia in 55% and 50% of cases, respectively.
Classification of acute aortic dissection that extends into to the arch remains a dilemma. Our current classification system based on the Stanford’s system remains clinically relevant and practical.3,5 Like all classification systems, however, it has limitations. This current scheme does not account for dissection or tears involving the transverse arch. If any involvement of the ascending aorta is noted, then type A dissection is declared and urgent surgical intervention is performed. However, management of type B aortic dissection that extends into the transverse arch is less clear, although some have attempted to address this issue.13 In our current protocol, we use transesophageal echocardiography to further characterize the location of the dissection flap and aortic tear. This allows us with relative certainty to exclude ascending aortic involvement. We currently manage acute type B aortic dissection with arch involvement medically, but we did observe cerebral complications in 7% of patients in that subpopulation. Although the exact etiology of these complications was unclear, transverse arch dissection leading to malperfusion or thrombosis must be considered. Acute type B aortic dissection with extension into the transverse arch needs further study.
The treatment protocol for acute type B aortic dissection has evolved to our current protocol presented in this study. The primary goal of treatment has remained decreasing the Δp/Δt or “impulse force” by controlling blood pressure. Although the goal of “anti-impulse” treatment attempts to reduce systolic blood pressure to <120 mm Hg, we defined systolic blood pressure <140 mm Hg as the level that control of blood pressure was successfully achieved. Whether a more aggressive regimen needs to be attained in the acute setting is yet to be determined. Although the goal of anti-impulse therapy has remained the same, changes have occurred in other aspects of our management protocol as our experience has expanded. One specific alteration in our protocol has been that we now use central venous access in most patients to facilitate the delivery of concentrated intravenous antihypertensives. Early in our experience, many patients would require multiple intravenous antihypertensives necessitating excess delivery of intravenous fluids if given peripherally. The use of central venous access, allowing for the concentrated delivery of antihypertensives, ie, double and triple concentrating the medications, prevented overadministration of intravenous fluid and the development of pulmonary edema and respiratory compromise.
Our follow-up protocol has also been modified as the result of our observations from this study. In the intermediate follow-up, we observed that the need for aortic intervention after discharge was infrequent (4.3%), but that when aortic intervention was required it occurred within 6 months of discharge, all but one from aneurysmal aortic dilatation to >5 cm. This has led to our current radiographic follow-up protocol of CT scan or MRI at discharge, 6 weeks, 3 months, 6 months, and yearly thereafter.
In general, initial medical management has been the consensus for the treatment of acute type B aortic dissection unless associated with life-threatening complications. Early mortality remains significant despite aggressive medical management and diminished long-term survival has been reported in these patients.14 For these reasons, endovascular treatment of acute aortic dissection has gained increased interest as an initial treatment option. The appeal arises from the potential to address malperfusion syndromes as well as exclude the tear and obliterate the false lumen. In a recent meta-analysis, however, the mortality of the subgroup of 248 patients with acute type B aortic dissection treated with endovascular stenting was still significant at 10% with paraplegia of 2% to 3%.15 Early outcomes with endovascular thoracic stenting appear no better than medical treatment for descending thoracic aortic dissection and the durability of endovascular stenting has yet to be determined; however, further studies with this treatment modality need to be performed.
Our study should be viewed with certain limitations. In general, although it is a prospective database, the data were analyzed retrospectively and thus are associated with inherent biases. In addition, our protocol for the management of acute type B dissection evolved, to a certain degree, in the first 2 years of its implementation. As noted, alterations in fluid management occurred early in the series. In addition, maintenance of the anti-impulse therapy and blood pressure control was performed by the cardiovascular surgical team in conjunction with the cardiologist. Although general principles of management were maintained, variations among patients with regards to transitioning to oral antihypertensives existed. These limitations do not detract, however, from the specific results of the actual medical protocol reported in this study. Previous studies have reported some details in the protocol used, but none mention the degree to which drug therapy was effective.11,16
Another bias in our experience was related to the indications for surgical intervention. Balanced between the dismal prognosis when aortic rupture occurs and our relative success with aortic surgical intervention,17,18 we actively intervene when the thoracic aortic diameter exceeds 5 cm. Because this differs from other centers, differences in outcomes exists based on our indications for intervention.19,20 In addition, unless rupture was identified, early aortic intervention was performed in a delayed selective fashion, often times allowing the patient to become “subacute” (>2 weeks from onset of pain but <6 weeks) and then performing the repair during the initial hospitalization. Although in the subacute phase of aortic dissection, these patients were still classified as having an early vascular intervention as opposed to those patients who were discharged and returned with acute expansion.
In conclusion, although acute aortic dissection remains associated with significant mortality and morbidity, medical management is associated with acceptable outcomes. Further studies evaluating other management strategies, eg, endovascular stenting, for acute type B aortic dissection need to be performed and compared with these results.
We thank Kirk Soodhalter for his editorial assistance and for preparation of this manuscript.
Presented at the American Heart Association Scientific Sessions, Dallas, Tex, November 13–16, 2005.
Miller DC, Mitchell RS, Oyer PE, Stinson EB, Jamieson SW, Shumway NE. Independent determinants of operative mortality for patients with aortic dissections. Circulation. 1984; 70 (3 Pt 2): I153–164.
Nienaber CA, Eagle KA. Aortic dissection: new frontiers in diagnosis and management: Part I: from etiology to diagnostic strategies. Circulation. 2003; 108: 628–635.
Suzuki T, Mehta RH, Ince H, Nagai R, Sakomura Y, Weber F, Sumiyoshi T, Bossone E, Trimarchi S, Cooper JV, Smith DE, Isselbacher EM, Eagle KA, Nienaber CA. Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the International Registry of Aortic Dissection (IRAD). Circulation. 2003; 108 (Suppl 1): II312–II317.
Cambria RP, Brewster DC, Gertler J, Moncure AC, Gusberg R, Tilson MD, Darling RC, Hammond G, Mergerman J, Abbott WM. Vascular complications associated with spontaneous aortic dissection. J Vasc Surg. Feb. 1988; 7: 199–209.
Fann JI, Smith JA, Miller DC, Mitchell RS, Moore KA, Grunkemeier G, Stinson EB, Oyer PE, Reitz BA, Shumway NE. Surgical management of aortic dissection during a 30-year period. Circulation. 1995; 92 (9 Suppl): II113–II121.
Eggebrecht H, Nienaber CA, Neuhauser M, Baumgart D, Kische S, Schmermund A, Herold U, Rehders TC, Jakob HG, Erbel R. Endovascular stent-graft placement in aortic dissection: a meta-analysis. Eur Heart J. 2006; 72: 489–498.
Marui A, Mochizuki T, Mitsui N, Koyama T, Kimura F, Horibe M. Toward the best treatment for uncomplicated patients with type B acute aortic dissection: A consideration for sound surgical indication. Circulation. 1999; 100 (19 Suppl): II275–II280.
Estrera AL, Miller CC, 3rd, Chen EP, Meada R, Torres RH, Porat EE, Huynh TT, Azizzadeh A, Safi HJ. Descending thoracic aortic aneurysm repair: 12-year experience using distal aortic perfusion and cerebrospinal fluid drainage. Ann Thorac Surg. 2005; 80: 1290–1296;discussion 1296.