Effectiveness of intensive medical therapy in type B aortic dissection: A single-center experience

Results 

The database search identified 148 patients, and 83 patients (55 men) fulfilled the criteria for this study. The mean age was 67 years (range, 38 to 85). Forty-two (50%) patients presented to the emergency department within hours of onset of symptom. The most common symptoms included back pain (72.2%), chest pain (44.5%), abdominal pain (30%), and leg pain (4.5%). Predictably 82% of the patients were known to have hypertension, and 25 (32%) were receiving β-blocker therapy as part of their antihypertensive regimen before onset of their symptoms. At the time of discharge, 25 patients (40%) in group 1 were taking four or more antihypertensive medications. Other comorbid conditions included coronary artery disease (CAD) in 21, diabetes mellitus (DM) in seven, and a history of previous stroke (CVA) in eight. Only three patients (3.6%) had a diagnosis of Marfan syndrome. Fifty-one (62%) of patients were smokers, and 20 (24%) had history of previous abdominal aortic aneurysm repair (Table I).

Table I. Patient characteristics

	Characteristic	Patients (n = 83), n (%)
	Mean age, years (range)	67(38-85)
	M:F	55:28
	Hypertension	68(81.9)
	Marfan syndrome	3(3.6)
	Diabetes mellitus	7(8.4)
	Renal dysfunction	4(4.8)
	CABG	7(8.4)
	CAD	21(25.3)
	CVA	8(9.6)

CABG, coronary artery bypass graft; CAD, coronary artery disease; CVA, cerebrovascular accident.

Intensive medical therapy alone was successfully used to treat 64 patients (77%; group 1), and 19 (23%) required emergency surgery for rupture/impending rupture (n = 12) mesenteric ischemia (n = 4) and lower extremity ischemia (n = 3). Of the 19 patients who required surgery, 14 (6 rupture, 3 symptomatic >6 cm aneurysms, 3 mesenteric ischemia, and 2 extremity ischemia) were operated the day of admission, and five patients requiring emergency surgery were initially treated with intensive medical therapy. Three of these five patients had continued chest pain and enlargement of the aorta, one developed leg ischemia, and one patient developed mesenteric ischemia. No intervention was performed solely for renal dysfunction.

Patients aged >70 years (P = .035) and smokers (P = .03) were more likely to require emergency surgery. In contrast, patients who were taking a β-blocker as part of their antihypertensive regimen before onset of symptoms seemed to be protected from emergency surgery (P = .023).

The overall 30-day mortality was 9.6% (8 deaths). Age >70 years (P = .057), clinical presentation with leg pain (P = .013), previous abdominal aortic aneurysm repair (P = .018), and a history of tobacco use (P = .039) increased significantly the possibility of death. Gender, presentation with chest or abdominal pain, history of hypertension, coronary artery disease, and history of diabetes mellitus were not associated with an increase in the over all mortality (Table II).

Table II. Factors affecting mortality and the need for emergency surgery

	Factors	All patients (n = 83)		Group 2 (n = 19)
		Factors affecting mortality		Factors affecting emergency surgery
		Mortality rate (%)	P	Mortality rate (%)	P
	Age
	<70	2.4	.05	12	.035
	>70	16.6		34
	Gender
	M	9	1	22	.7
	F	10		25
	Leg pain
	Yes	33	.013	45	.1
	No	5.6		19.7
	Previous AAA repair
	Yes	25	.018	26	.7
	No	4.7		22
	Tobacco use
	Yes	14	.039	30.6	.03
	No	0		9.6
	β-blocker
	Yes	8	1	8.3	.025
	No	6		32.6

AAA, Abdominal aortic aneurysm.

The mortality rate was significantly higher in group 2 (P = .00046) than in group 1. Two patients (1.6%) in group 1 died from sudden cardiopulmonary arrest (no autopsy was performed), and six patients died in group 2. Four of these deaths occurred in patients who required emergency surgery for rupture or impending rupture. Their deaths were due to multisystem organ failure (MSOF). One of the remaining two patients who died presented with limb ischemia and died of MSOF, and another who presented with mesenteric ischemia died from hemorrhage (Table III).

Table III. Complications

	Type of complication	Group 1 (n =15)	Group 2 (n=12)
	Paraplegia	2	2
	Mental status changes	7	3⁎
	Respiratory failure	1	2
	Myocardial Infarction	4	2
	Arrhythmia	2
	Sepsis	1	2⁎
	Phrenic nerve injury		1
	Bleeding		2†
	Cerebrovascular accident		1†
	Lower extremity ischemia		1

⁎ Each star represents death. Of the 3 patients with mental status changes, 2 died, both patients with sepsis died as well, and 1 patient with bleeding and 1 with cerebrovascular accident died. There were a total of six deaths. † In-hospital mortality.

The overall morbidity rate was 33% (n = 27), consisting of 15 in group 1 and 12 in group 2. Morbidity in group 2 was significantly higher compared with group 1 (P = .002). Two patients in group 1 presented with paraplegia as part of their initial symptoms. Two patients in group 2 developed paraplegia after surgery. Other morbidities included renal failure, bacteremia, mental status changes, renal dysfunction, myocardial infarction, and arrhythmias.

The overall mean length of hospital stay was 14 days and length of ICU care was 8 days. There was no significant difference in the mean length of hospital stay (P = .1) and ICU stay (P = .12) between the two groups.

The mean follow-up time for the cohort of patients who survived the initial admission was 38.2 months. During follow-up, 18 patients (28%) required surgery for continued aneurysmal growth of the dissected aorta. Of these elective surgery cases, 15 were performed ≤1 year from onset of symptoms. The overall survival for the entire cohort was 75%, 60%, and 49% at 1, 3, and 5 years, respectively. For group 1 vs group 2 patients, survival was 85% vs 54% at 1 year, 64% vs 54% at 3 years, and 59% vs 18% at 5 years (P = .002; Fig).

View Large Image Download to PowerPoint

Fig. Kaplan-Meier curves for long-term survival for group 1 (blue line) and group 2 (red line). EMERSX, Emergency surgery.

A review of available MRA or CT scans, or both, of the 13 patients in group 2 revealed that 11 could have been treated with the currently available Excluder thoracic endovascular graft system (W. L. Gore & Assoc, Flagstaff, Ariz). Mean aortic diameter proximal to the proximal aortic tear was 34 mm (range, 26 to 42 mm), and coverage of the left subclavian artery origin would have been required in five cases to achieve an adequate proximal aortic length for seal.

Discussion 

ADTAD in the United States continues to be a challenging medical condition, with no significant improvement in the treatment outcomes during the past several decades. As reported by several authors, hypertension seems to be the most common predisposing factor, whereas dissection related to collagen vascular disease is relatively rare.3, 4 Hypertension was present in 82% of our patients, and only 3.6% were known to have Marfan syndrome. The clinical presentation in this study was also similar to other reports and consisted of severe chest, back, or abdominal pain in the presence of uncontrolled hypertension.

The gold standard for treating newly diagnosed uncomplicated ADTAD remains medical therapy, whereas surgery is reserved for complicated presentations of ADTAD such as visceral or limb ischemia or symptoms of rupture or impending rupture.3 This is because of the high morbidity and mortality rates associated with emergency surgery in patients with ADTAD.5 Our current intensive medical therapy protocol consists of the intravenous administration of a β-blocker, followed by a vasodilator (nitroglycerine) and other additional oral or parenteral antihypertensive medications as needed.

Most recent reports indicate the overall mortality rate for ADTAD to be about 10%.11 This compares favorably with our result, where the overall in-hospital mortality rate was 9.6%. Predictably, the mortality rate was significantly higher if patients required emergency surgery compared with those uncomplicated patients treated with medical therapy alone. Similar to other reports, age >70 also appeared to adversely affect overall survival and increased the need for emergency surgery.12 Contrary to the data reported by Niehuber et al,13 our study did not identify any gender-related increase in morbidity and mortality. However, other factors that did adversely affected outcomes by increasing overall mortality included previous infrarenal aortic aneurysm repair, initial clinical presentation with leg pain, and tobacco use. We have no explanation why leg pain was associated with an increased risk of mortality, although this may be related to a more extensive dissection.

In our series, the early mortality rate for patients who were successfully treated medically was remarkably low (1.6%). From these data it is difficult to justify endovascular therapy in patients who do not develop complications at their initial presentation. This statement seems to be supported by a report by Kato et al,14 where stent grafting of acute dissection was associated with early and late complication rates of 33% and 36%. In contrast, early and late complication rates of 4% and 0% were reported for similar treatment of chronic dissections. These authors thus recommended delay in stent graft treatment of uncomplicated dissections but also suggested that endovascular therapy may decrease the significant morbidity and mortality associated with emergency surgery.14

Emergency surgery was required in 23% of our patients. As indicated by several authors, mortality rates after emergency surgery are very high.3, 11, 12, 15 The high mortality rate seen in patients requiring emergency surgery for ADTAD demonstrated the need for an alternative, safer approach to this complex problem.

In recent years, several reports have documented results of the use of various endovascular therapies as a less invasive alternative treatment for ADTAD. These include covered stent grafting, percutaneous aortic fenestration, and aortic branch vessel stenting.6, 7, 16, 17, 18, 19 The goals of endovascular therapy using stent grafts in ADTAD include coverage and stabilization of the proximal dissection entry site, thrombosis of the false lumen, preservation of the true lumen and branch artery flow, and ideally, a decrease in the diameter of the aorta.

Although there is no level I evidence on the use of endovascular therapy for ADTAD, several reports of small series are available. Technical success with this approach is up to 100% in most series, and mortality rates are 6% to 10%.6, 7, 17 Xu et al17 recently reported successful proximal dissection entry site sealing in up to 95.2% of patients and a false lumen thrombosis rate of 98.4% at the 1-year follow-up.17 In addition, Gaxotte et al16 as well as Xu reported that when complete thrombosis of the false lumen occurred, it resulted in a decrease in the aortic diameter, whereas partial thrombosis was associated with continued aneurysm enlargement.16, 17 Review of CT scans and MRA of 13 patents in group 2 in our study whose images were available revealed that 11 patients could have been treated with currently the available Excluder thoracic endograft system. In four patients, however, coverage of the origin of the left subclavian artery would have been needed to achieve a sufficient proximal aortic length for endograft sealing.

Patients in group 1 in our study were monitored closely, and 28% required subsequent surgical repair. Most of these procedures, 15 (83%) of 18, were performed ≤1 year from the initial presentation, suggesting the need for very close follow up in the earlier months following ADTAD. According to a report by Gallo et al,15 most medically treated patients eventually require surgery for aneurysmal growth. On the other hand, Estrera et al11 reports that only 4.3% of their patients required surgical intervention after discharge. Regardless, as a result of our findings we currently recommend follow-up imaging with a CT scan or MRA at discharge, 3 and 6 months, and yearly thereafter.

Conclusion 

Most patients with ADTAD can initially be successfully managed medically, and when successful, such management is associated with a very low mortality and morbidity rates. Careful follow-up is required, however, because up to one third of these patients require elective intervention in the future. In contrast, outcomes are significantly worse if patients with ADTAD initially require emergency surgery. From retrospective review of imaging studies of patients who required emergency surgery after ADTAD in this series, most patients could have been treated with endovascular grafts, which might have improved outcomes. Further clinical studies are needed to carefully assess the value of endovascular stent grafting in the treatment of patients requiring emergency intervention for ADTAD and the need for endovascular stent grafting in patients with uncomplicated ADTAD.

Author contributions