- Email: [email protected]
Aortic dissection: Surgical and nonsurgical treatments compared
Aortic Dissection: Surgical and Nonsurgical Treatments Compared An Analysis of Seventy-Four Cases at the University of Virginia
Stacey E. Mills, MD, Charlottesville, Virginia Kuldeep Teja, MD, Charlottesville, Virginia Ivan K. Crosby, #BBS,* Charlottesville, Virginia Benjamin C. Sturgill, MD, Charlottesville, Virginia
Aortic dissection, a condition characterized by hemorrhage into the media and variable extension along the length of the aorta, has long been recognized as a catastrophic cardiovascular event. Recent developments in diagnostic and therapeutic skills have improved the prognosis considerably, but there is still controversy as to how cases should be managed. DeBakey et al [I] in 1965, reporting on the surgical treatment, classified aortic dissection into three types according to the site and extent of dissection. In type I, “the dissecting process, along with the intimal tear, arises in the ascending aorta and extends distally for a variable distance, but usually throughout the remaining aorta.” In type II, “the dissecting process is limited to the ascending and transverse aorta and does not extend distal to the left subclavian artery.” In type III, “the dissecting process originates in the descending aorta, distal to the left subclavian and extends downward for a variable distance.” A different surgical approach was described for each type. DeBakey et al [I] reported an overall operative mortality of 21 per cent in 179 patients. More important, 70 per cent of the patients operated on were alive at one year compared with 7 per cent in a large series of nonoperated patients [2]. Others [3,4], however, reported higher mortalities after surgery. Wheat et al [5], noting the reported efficacy of reserpine in the treatment of dissecting aneurysms in turkeys [6], outlined a medical antihypertensive regimen for the treatment of aortic dissection in human beings. After carefully excluding patients with rapidly progressing dissection, Wheat et al treated six patients with hypotensive agents (trimethaphan, reserpine, and guanethidine). All of the From the Departments of Pathology and Surgery*, University of Virginia Medical Center, Charlottesville, Virginia. Reprint requests should be addressed to Stacey E. Mills, MD, Department of Pathology, University of Virginia, Charlottesville, Virginia 22908.
240
patients were alive at the time of the report, representing survival for nine to fifteen months. During the past eleven years after these two reports, one advocating aggressive surgical management and the other conservative medical management, most patients with aortic dissections seen at this medical center have been treated by one or another of these two methods. The present report compares the results of medical and surgical treatment in patients with each of the three types of aortic dissection. Material and Methods A total of seventy-eight patients with aortic dissection presenting at the University of Virginia Medical Center from January 1965 to March 1976 were reviewed. Data were collected from medical records and autopsy reports. Only cases documented by aortography, surgery, or autopsy were included. Four cases were excluded because of insufficient data, leaving seventy-four cases that comprise this report. Each dissection was classified according to the criteria of DeBakey et al [1] as type I, II, or III. The clinical and epidemiologic features, including average age, male/female ratio, black/white patient ratio, frequency of hypertension, frequency of characteristic pain, and correct clinical diagnosis, were determined for the three types. Three groups were identified according to the mode of therapy. Forty patients were treated surgically, always by the interruption of the origin of dissection and the insertion of a Dacrona prosthesis, thirteen were treated according to the Wheat regimen (reserpine or propranolol, trimethaphan, and guanethidine to maintain a systolic pressure of 100 to 120 mm Hg and to keep patients free of pain), and twenty-one received either no treatment or nonspecific supportive care. The results of the three modes of therapy for the three types of dissection were compared according to survival at 2 hours, one week, one month, and one year after the onset of symptoms in the medically and nonspecifically treated groups or from onset of surgery in The American Journal 01 Surgery
Aortic Dissection
TABLE I
Clinical and Epidemiologic Features
Dissection No. Average age (yr) (mean f SD) Male/female ratio Black/white ratio History of hypertension Correct clinical diagnosis Characteristic pain Treated surgically Treated medically (Wheat) Treated supportively only
Type 1
Type II
Type Ill
33 (45%) 58f 11 2518 4/29 19 (58%) 27 (82%) 26 (79%) 21 (64%) 3 (9%) 9 (27%)
10 (14%) 59 f 10 812 218 6 (60%) 4 (40%) 5 (50%) 5 (50%) 0 (0%) 5 (50%)
31 (42%) 57 f 12 2417 6125 26 (84%) 27 (87%) 30 (97%) 14 (45%) 9 (29%) 8 (26%)
the surgically treated group. Patients treated surgically were operated on in most instances within 24 hours of the onset of symptoms, and survival at one week in this group did not differ significantly, whether determined from the onset of symptoms or from the time of surgery. Fifty-six of the seventy-four patients (62 per cent) died, and forty-two were autopsied. The frequency of cardiomegaly (>350 gm [female] to 400 gm [male]) was determined for patients with each of the three types of dissection and compared with an age- and sex-matched control group (nondissection cases) autopsied during the same period of time as covered by the study. The frequency at autopsy of aortic rupture, tamponade, and reentry tears for each of the three types of dissection was determined. Results
The clinical and epidemiologic characteristics of these seventy-four patients are presented in Table I. Types I and III dissections were about equal in frequency and together constituted the vast majority of the cases (86 per cent). The average age of the patients was approximately sixty years, and this did not differ significantly from type to type. Men were affected four times as frequently as women, but the percentage of black patients (19 per cent) was not significantly different from the percentage of black patients in the hospital population at the University of Virginia Medical Center (19 per cent). Hypertension was frequently associated with dissection (69 per cent overall), particularly in those with type III dissection (84 per cent), confirming an association mentioned in virtually every reported series. The correct clinical diagnosis was established much more frequently in patients with types I and III dissections than in patients with type II, probably because symptoms tended to be more characteristic in the former types. The results of surgical treatment (interruption and Dacron prosthesis) of forty patients are presented in Table II, top. The majority of patients had type I dissections. This group did poorly, with only 19 per cent alive at the end of one year. Patients with type Volume 137, February 1979
Total 74 (100%) 58f 11 57117 12162 51 (69%) 58 (78%) 61 (82%) 40 (54%) 12 (16%) 22 (30%)
II dissections fared much better, with 60 per cent of them alive at one year. The results among the operated group with type III dissections (7 of 14 patients [50 per cent] alive at 1 year) compared favorably with the average survival of 35 per cent for surgically treated patients with all types of dissection. The results of medical (Wheat) therapy of twelve patients are presented in Table II, middle. Most of these had type III dissections. Only three patients with type I dissection and none with type II were treated by the Wheat regimen. Only three were alive at one year, all with type III. The results of nonspecific medical treatment of twenty-two patients are presented in Table II, bottom. This group had the lowest overall survival at one year (18 per cent). Among the group treated nonspecifically, those with type III dissections had the best prognosis (38 per cent alive at 1 year) followed by those with type I (11 per cent) and type II (0). Of the twenty-two patients treated nonspecifically, ten died before specific medical or surgical therapy could be initiated, five were misdiagnosed until au-
TABLE II Survival
Survival of Patients No. of Patients According to Type of Dissection Type II Type III Total Type I
Surgically Treated (n = 40) At 2 hr Initial At 1 wk At 1 mo At 1 yr
21 14 (67%) 11(52%) 4 (19%) 4 (19%)
5 3 3 3
(100%) (60%) (60%) (60%)
Medlcally (Wheat) Treated (n = 12) 3 0 Initial 3 (100%) At 2 hr 1(33%) At 1 wk 0 At 1 mo 0 At 1 yr
:: (93%) 10(71%) 9 (64%) 7 (50%) 9 9 7 3 3
(100%) (78%) (33%) (33%)
Nonsurglcally (excluding Wheat) Treated (n = 22) 9 5 8 Initial 8 (89%) 3 (60%) 8 (100%) At 2 hr 5(56%) 0 6 (75%) At 1 wk 4(44%) 0 4 (50%) At 1 mo l(ll%) 0 3 (38%) At 1 yr
:20 (80%) 24(60%) 16 (40%) 14 (35%) 12 12 (100%) 8 (87%) 3 (23%) 3 (23%) 22 19 11 8 4
(86%) (50%) (36%) (18%)
241
Mills et
al
TABLE III
Comparison of Modes of Therapy, Type of Dissection, and Survival at 1 Year
Treatment
No. of Patients According to Type of Dissection TyDe Type Total TvDe - I - II - Ill
Surgical Nonsurgical Total
4/21(19%) l/12 (6%)
315 015
(60%)
7114 (50%) 6/17 (35%)
14140 (35%) ‘7/34 (21%)
5/33 (15%)
3/10 (33%)
13131 (42%)
21174 (26%)
topsy, and five were diagnosed only after the dissection had stabilized, but before specific therapy had been instituted. Only two were considered inoperable. The survival rates at one year of patients with all three types of dissection are compared according to the mode of therapy in Table III. Those with type I dissections had a low survival regardless of the mode of therapy. No patients with type II dissections were treated by the Wheat regimen. Three of five patients with type II dissections treated surgically were alive at one year, whereas none of the five treated nonspecifically were alive. Seven of fourteen surgically treated patients with type III dissections were alive at one year, whereas only three of nine patients treated by the Wheat regimen and three of eight treated nonspecifically were alive at one year. The findings in forty-two autopsied cases are presented in Table IV. Cardiomegaly was a frequent finding in each of the three types of dissection (83 per cent overall). By comparison, cardiomegaly was present in only 47 per cent of the age-matched autopsy control population. Aortic rupture occurred in approximately 75 per cent of those with types I and II dissections but was present at autopsy in only 36 per cent of those with type III, undoubtedly an important factor in their better prognosis. Cardiac tamponade was present in 48 per cent of patients with type I dissection, in 71 per cent of those with type II, and of course did not occur in any patient with type III. Reentry tears, usually associated with a better prognosis.and with termination of further dissection, were present in 43 per cent of patients with type III dissections, 29 per cent of those with type I, and none of those with type II. Comments
The general characteristics (age, male preponderance, hypertension) of the patients who comprised this series were similar to those previously reported [2,7,8]. Some authors ISJO] have suggested that aortic dissection is relatively more common in the black population. Given the frequency of hy242
TABLE IV
Autopsy Findings in 42 Cases No. of Patients According to Type of Dissection Type II Type Ill Total Type I
Cardiomegaly Aortic rupture Tamponade Reentry tear
21 (50%) 19 (90%) 16 (76%) 10(48%) 6(29%)
7 (17%) 6 (86%) 5 (71%) 5(71%) 0
14 10 5 0 6
(33%) (71%) (36%) (43%)
42 (100%) 35 (83%) 26 (62%) 1’5 (36%) 12 (29%}
pertension among black people, this would not be surprising, but in our series black patients were affected no more frequently than white patients. This suggests that other factors in addition to hypertension are operative in the cause of aortic dissection. The more characteristic the symptoms, the more frequently a correct clinical diagnosis was made, Type II aneurysms were the most difficult to diagnose because the limited dissection (ascending aorta) tended to produce pain, which simulated other conditions, such as myocardial infarction or pulmonary embolism. The prognosis for aortic dissection has been vastly improved in recent years by aggressive surgical management; exemplified by the results of DeBakey et al [I]. Employing a combination of technics, including closure of the dissection for type I and resection with insertion of a Dacron prosthesis for types II and III, DeBakey et al reported 70 per cent survival at one year and 50 per cent survival at six years, compared with a dismal 7 per cent at one year and less than 1 per cent at six years for untreated patients. Using trimethaphan, reserpine, and guanethidine to maintain the blood pressure at 100 mm Hg, Wheat et al [5] obtained a remarkable 100 per cent survival in six patients followed nine to fifteen months. It should be noted that these six were selected to exclude patients with aortic rupture, aortic insufficiency, or severe heart failure. The most obvious fact apparent from these data is that patients with type I dissections, those which involve both the ascending and descending aorta, have a poor prognosis regardless of the mode of therapy. Only live of the thirty-three patients (15 per cent) were alive at the end of one year. Although surgically treated patients with type I dissections appeared to do better than those nonsurgically treated (19 per cent versus 8 per cent 1 year survival), the difference is not statistically significant. Surgically treated patients with type II dissections, on the other hand, clearly did better than the unoperated group. Although the numbers are small, the difference is statistically significant (p < 0.05), and surgery appears to be the treatment of choice. For reasons that are not clear, no patients with type II The American Journal of Surgery
Aortic
dissections in this series were treated by the Wheat regimen. Surgically treated patients with type III dissections also appeared to have done better than unoperated patients, but again the difference is not statistically significant. Wheat treatment, as practiced in this institution, also does not appear to offer any long range therapeutic advantages over nonspecific supportive measures in the treatment of aortic dissection. It would appear to have a place in the early stabilization of dissection; however, the survival rate decreases markedly between one week and one month. These findings indicate that surgically treated patients with aortic dissection have a better survival at one year than unoperated patients whether treated by the Wheat regimen or by nonspecific supportive measures. This is particularly true of type II dissection and to a lesser extent to types I and III. In this institution [II], the usual approach has been to employ medical (Wheat) therapy first and to operate only when there is (1) failure to bring pain or blood pressure under control, (2) propagation or extension of the hematoma, (3) decompensated aortic regurgitation, and (4) occlusion or compromise of major branches of the aorta, or (5) saccular dilatation of the aorta. This may not be the best approach. Apparent early success of medical (Wheat) treatment was eroded with the passage of time, as indicated by the one month survival data. A more rational approach might be to operate electively sometime during the first week or two after dissection. The findings at autopsy in the forty-two patients who died confirm the significant association of hypertension (as manifested by cardiomegaly) with aortic dissection. Cardiomegaly was somewhat less common in patients with type III dissection than in those with the other two types and may be a factor in the better overall survival of patients with type III dissection. Another factor which is probably more important in this better prognosis is the lower incidence of aortic rupture, which may be due to the decreased mechanical trauma experienced by the rather immobile descending aorta as opposed to the torsional “bucket-and-handle” forces imposed on the ascending and transverse aorta with each systole. Reentry tears, long associated with a better prognosis, are more common in type III dissections and lead to the early surgical approach of creation of a reentry site [12]. This procedure was later shown to have little or no effect on prolongation of survival. It is our belief that reentry tears may simply occur in increasing numbers with time and are thus more common in the longer living type III patients, but in and of themselves exert no effect on prolongation of life. Volume 137, February 1879
Dissection
Summary
During the eleven year period from 1965 to 1976, aortic dissection was diagnosed in seventy-eight patients treated at the University of Virginia Medical Center. The seventy-four cases that comprise the present report were confirmed by aortography, surgery, or autopsy. Forty patients (54 per cent) were surgically treated by interruption of the origin of dissection and insertion of a Dacron prosthesis. The remaining thirty-four (46 per cent) were treated according to the Wheat regimen or by nonspecific supportive measures. The overall survival at one year was 28 per cent. One year survival for patients with type I dissections treated surgically was 19 per cent compared with 8 per cent one year survival for nonsurgically treated patients. Sixty per cent of patients with type II dissections treated surgically were alive at the end of,one year, whereas no patients with type II dissection treated nonsurgically survived beyond one year. Half the patients with type III dissections treated surgically were alive at one year compared with 35 per cent of those nonsurgically treated. These data suggest that surgery is the treatment of choice for all types of aortic dissections, but particularly for type II. Patients with type I dissections have a very poor prognosis regardless of therapy. References 1. DeBakey ME, Henly WS, Cooley DA, Morris GC Jr, Crawford ES, Beak AC Jr: Surgical management of dissecting aneurysms of the aorta. J Thorac Cardiovasc Surg 49: 1130, 1965. 2. Hirst AE Jr, Johns VJ Jr, Kime SW Jr: Dissecting aneurysm of the aorta: a review of 505 cases. Medicine 37: 217, 1958. 3. Warren WD, Beckwith J, Muller WH Jr: Problems in the surgical management of acute dissecting aneurysm of the aorta. Am Surg 144: 530, 1965. 4. Hume DM, Porter RR: Acute dissecting aortic aneurysms. Surgery 53: 122, 1963. 5. Wheat MW Jr, Palmer RF. Bartley TD, Seelman RC: Treatment of dissecting aneurysms of the aorta without surgery. J Thorac Cardiovasc Surg 50: 364, 1965. 6. Ringer RK: Influence of reserpine and early growth, blood pressure and dissecting aneurysms in turkeys, p 2 1. Conference on the Use of Serpasil in Animal and Poultry Production, May 7, 1969, Rutgers, The State University, New Brunswick, NJ. Summit, New Jersey, CIBA, 1969. 7. Gore I, Hirst AE Jr: Dissecting aneurysms of the aorta. Cardiovasc C/in 5: 239, 1973. 8. Lindsay J Jr, Hurst JW: Dissecting aneurysms of the aorta. JAMA 217: 1533, 1971. 9. Levinson DC. Edmeades DT, Griffith GC: Dissecting aneurysm of the aorta; its clinical electrocardiographic and laboratory features. Report of 58 autopsied cases. Circulation 1: 360, 1950. 10. Rider JA, Chriss JW, Herrmann CR: Dissecting aneurysm of the aorta; a ten year study. Tex A&d 46: 3 11, 1950. 11. Crosby IK, Golden GT: Acute dissecting aneurysms of the aorta. Va Med 102: 558, 1975. 12. DeBakey ME. Cooley DA, Creech 0 Jr: Surgical considerations of dissecting aneurysm of the aorta. Am Surg 142: 586, 1955.
243
Stacey E. Mills, MD, Charlottesville, Virginia Kuldeep Teja, MD, Charlottesville, Virginia Ivan K. Crosby, #BBS,* Charlottesville, Virginia Benjamin C. Sturgill, MD, Charlottesville, Virginia
Aortic dissection, a condition characterized by hemorrhage into the media and variable extension along the length of the aorta, has long been recognized as a catastrophic cardiovascular event. Recent developments in diagnostic and therapeutic skills have improved the prognosis considerably, but there is still controversy as to how cases should be managed. DeBakey et al [I] in 1965, reporting on the surgical treatment, classified aortic dissection into three types according to the site and extent of dissection. In type I, “the dissecting process, along with the intimal tear, arises in the ascending aorta and extends distally for a variable distance, but usually throughout the remaining aorta.” In type II, “the dissecting process is limited to the ascending and transverse aorta and does not extend distal to the left subclavian artery.” In type III, “the dissecting process originates in the descending aorta, distal to the left subclavian and extends downward for a variable distance.” A different surgical approach was described for each type. DeBakey et al [I] reported an overall operative mortality of 21 per cent in 179 patients. More important, 70 per cent of the patients operated on were alive at one year compared with 7 per cent in a large series of nonoperated patients [2]. Others [3,4], however, reported higher mortalities after surgery. Wheat et al [5], noting the reported efficacy of reserpine in the treatment of dissecting aneurysms in turkeys [6], outlined a medical antihypertensive regimen for the treatment of aortic dissection in human beings. After carefully excluding patients with rapidly progressing dissection, Wheat et al treated six patients with hypotensive agents (trimethaphan, reserpine, and guanethidine). All of the From the Departments of Pathology and Surgery*, University of Virginia Medical Center, Charlottesville, Virginia. Reprint requests should be addressed to Stacey E. Mills, MD, Department of Pathology, University of Virginia, Charlottesville, Virginia 22908.
240
patients were alive at the time of the report, representing survival for nine to fifteen months. During the past eleven years after these two reports, one advocating aggressive surgical management and the other conservative medical management, most patients with aortic dissections seen at this medical center have been treated by one or another of these two methods. The present report compares the results of medical and surgical treatment in patients with each of the three types of aortic dissection. Material and Methods A total of seventy-eight patients with aortic dissection presenting at the University of Virginia Medical Center from January 1965 to March 1976 were reviewed. Data were collected from medical records and autopsy reports. Only cases documented by aortography, surgery, or autopsy were included. Four cases were excluded because of insufficient data, leaving seventy-four cases that comprise this report. Each dissection was classified according to the criteria of DeBakey et al [1] as type I, II, or III. The clinical and epidemiologic features, including average age, male/female ratio, black/white patient ratio, frequency of hypertension, frequency of characteristic pain, and correct clinical diagnosis, were determined for the three types. Three groups were identified according to the mode of therapy. Forty patients were treated surgically, always by the interruption of the origin of dissection and the insertion of a Dacrona prosthesis, thirteen were treated according to the Wheat regimen (reserpine or propranolol, trimethaphan, and guanethidine to maintain a systolic pressure of 100 to 120 mm Hg and to keep patients free of pain), and twenty-one received either no treatment or nonspecific supportive care. The results of the three modes of therapy for the three types of dissection were compared according to survival at 2 hours, one week, one month, and one year after the onset of symptoms in the medically and nonspecifically treated groups or from onset of surgery in The American Journal 01 Surgery
Aortic Dissection
TABLE I
Clinical and Epidemiologic Features
Dissection No. Average age (yr) (mean f SD) Male/female ratio Black/white ratio History of hypertension Correct clinical diagnosis Characteristic pain Treated surgically Treated medically (Wheat) Treated supportively only
Type 1
Type II
Type Ill
33 (45%) 58f 11 2518 4/29 19 (58%) 27 (82%) 26 (79%) 21 (64%) 3 (9%) 9 (27%)
10 (14%) 59 f 10 812 218 6 (60%) 4 (40%) 5 (50%) 5 (50%) 0 (0%) 5 (50%)
31 (42%) 57 f 12 2417 6125 26 (84%) 27 (87%) 30 (97%) 14 (45%) 9 (29%) 8 (26%)
the surgically treated group. Patients treated surgically were operated on in most instances within 24 hours of the onset of symptoms, and survival at one week in this group did not differ significantly, whether determined from the onset of symptoms or from the time of surgery. Fifty-six of the seventy-four patients (62 per cent) died, and forty-two were autopsied. The frequency of cardiomegaly (>350 gm [female] to 400 gm [male]) was determined for patients with each of the three types of dissection and compared with an age- and sex-matched control group (nondissection cases) autopsied during the same period of time as covered by the study. The frequency at autopsy of aortic rupture, tamponade, and reentry tears for each of the three types of dissection was determined. Results
The clinical and epidemiologic characteristics of these seventy-four patients are presented in Table I. Types I and III dissections were about equal in frequency and together constituted the vast majority of the cases (86 per cent). The average age of the patients was approximately sixty years, and this did not differ significantly from type to type. Men were affected four times as frequently as women, but the percentage of black patients (19 per cent) was not significantly different from the percentage of black patients in the hospital population at the University of Virginia Medical Center (19 per cent). Hypertension was frequently associated with dissection (69 per cent overall), particularly in those with type III dissection (84 per cent), confirming an association mentioned in virtually every reported series. The correct clinical diagnosis was established much more frequently in patients with types I and III dissections than in patients with type II, probably because symptoms tended to be more characteristic in the former types. The results of surgical treatment (interruption and Dacron prosthesis) of forty patients are presented in Table II, top. The majority of patients had type I dissections. This group did poorly, with only 19 per cent alive at the end of one year. Patients with type Volume 137, February 1979
Total 74 (100%) 58f 11 57117 12162 51 (69%) 58 (78%) 61 (82%) 40 (54%) 12 (16%) 22 (30%)
II dissections fared much better, with 60 per cent of them alive at one year. The results among the operated group with type III dissections (7 of 14 patients [50 per cent] alive at 1 year) compared favorably with the average survival of 35 per cent for surgically treated patients with all types of dissection. The results of medical (Wheat) therapy of twelve patients are presented in Table II, middle. Most of these had type III dissections. Only three patients with type I dissection and none with type II were treated by the Wheat regimen. Only three were alive at one year, all with type III. The results of nonspecific medical treatment of twenty-two patients are presented in Table II, bottom. This group had the lowest overall survival at one year (18 per cent). Among the group treated nonspecifically, those with type III dissections had the best prognosis (38 per cent alive at 1 year) followed by those with type I (11 per cent) and type II (0). Of the twenty-two patients treated nonspecifically, ten died before specific medical or surgical therapy could be initiated, five were misdiagnosed until au-
TABLE II Survival
Survival of Patients No. of Patients According to Type of Dissection Type II Type III Total Type I
Surgically Treated (n = 40) At 2 hr Initial At 1 wk At 1 mo At 1 yr
21 14 (67%) 11(52%) 4 (19%) 4 (19%)
5 3 3 3
(100%) (60%) (60%) (60%)
Medlcally (Wheat) Treated (n = 12) 3 0 Initial 3 (100%) At 2 hr 1(33%) At 1 wk 0 At 1 mo 0 At 1 yr
:: (93%) 10(71%) 9 (64%) 7 (50%) 9 9 7 3 3
(100%) (78%) (33%) (33%)
Nonsurglcally (excluding Wheat) Treated (n = 22) 9 5 8 Initial 8 (89%) 3 (60%) 8 (100%) At 2 hr 5(56%) 0 6 (75%) At 1 wk 4(44%) 0 4 (50%) At 1 mo l(ll%) 0 3 (38%) At 1 yr
:20 (80%) 24(60%) 16 (40%) 14 (35%) 12 12 (100%) 8 (87%) 3 (23%) 3 (23%) 22 19 11 8 4
(86%) (50%) (36%) (18%)
241
Mills et
al
TABLE III
Comparison of Modes of Therapy, Type of Dissection, and Survival at 1 Year
Treatment
No. of Patients According to Type of Dissection TyDe Type Total TvDe - I - II - Ill
Surgical Nonsurgical Total
4/21(19%) l/12 (6%)
315 015
(60%)
7114 (50%) 6/17 (35%)
14140 (35%) ‘7/34 (21%)
5/33 (15%)
3/10 (33%)
13131 (42%)
21174 (26%)
topsy, and five were diagnosed only after the dissection had stabilized, but before specific therapy had been instituted. Only two were considered inoperable. The survival rates at one year of patients with all three types of dissection are compared according to the mode of therapy in Table III. Those with type I dissections had a low survival regardless of the mode of therapy. No patients with type II dissections were treated by the Wheat regimen. Three of five patients with type II dissections treated surgically were alive at one year, whereas none of the five treated nonspecifically were alive. Seven of fourteen surgically treated patients with type III dissections were alive at one year, whereas only three of nine patients treated by the Wheat regimen and three of eight treated nonspecifically were alive at one year. The findings in forty-two autopsied cases are presented in Table IV. Cardiomegaly was a frequent finding in each of the three types of dissection (83 per cent overall). By comparison, cardiomegaly was present in only 47 per cent of the age-matched autopsy control population. Aortic rupture occurred in approximately 75 per cent of those with types I and II dissections but was present at autopsy in only 36 per cent of those with type III, undoubtedly an important factor in their better prognosis. Cardiac tamponade was present in 48 per cent of patients with type I dissection, in 71 per cent of those with type II, and of course did not occur in any patient with type III. Reentry tears, usually associated with a better prognosis.and with termination of further dissection, were present in 43 per cent of patients with type III dissections, 29 per cent of those with type I, and none of those with type II. Comments
The general characteristics (age, male preponderance, hypertension) of the patients who comprised this series were similar to those previously reported [2,7,8]. Some authors ISJO] have suggested that aortic dissection is relatively more common in the black population. Given the frequency of hy242
TABLE IV
Autopsy Findings in 42 Cases No. of Patients According to Type of Dissection Type II Type Ill Total Type I
Cardiomegaly Aortic rupture Tamponade Reentry tear
21 (50%) 19 (90%) 16 (76%) 10(48%) 6(29%)
7 (17%) 6 (86%) 5 (71%) 5(71%) 0
14 10 5 0 6
(33%) (71%) (36%) (43%)
42 (100%) 35 (83%) 26 (62%) 1’5 (36%) 12 (29%}
pertension among black people, this would not be surprising, but in our series black patients were affected no more frequently than white patients. This suggests that other factors in addition to hypertension are operative in the cause of aortic dissection. The more characteristic the symptoms, the more frequently a correct clinical diagnosis was made, Type II aneurysms were the most difficult to diagnose because the limited dissection (ascending aorta) tended to produce pain, which simulated other conditions, such as myocardial infarction or pulmonary embolism. The prognosis for aortic dissection has been vastly improved in recent years by aggressive surgical management; exemplified by the results of DeBakey et al [I]. Employing a combination of technics, including closure of the dissection for type I and resection with insertion of a Dacron prosthesis for types II and III, DeBakey et al reported 70 per cent survival at one year and 50 per cent survival at six years, compared with a dismal 7 per cent at one year and less than 1 per cent at six years for untreated patients. Using trimethaphan, reserpine, and guanethidine to maintain the blood pressure at 100 mm Hg, Wheat et al [5] obtained a remarkable 100 per cent survival in six patients followed nine to fifteen months. It should be noted that these six were selected to exclude patients with aortic rupture, aortic insufficiency, or severe heart failure. The most obvious fact apparent from these data is that patients with type I dissections, those which involve both the ascending and descending aorta, have a poor prognosis regardless of the mode of therapy. Only live of the thirty-three patients (15 per cent) were alive at the end of one year. Although surgically treated patients with type I dissections appeared to do better than those nonsurgically treated (19 per cent versus 8 per cent 1 year survival), the difference is not statistically significant. Surgically treated patients with type II dissections, on the other hand, clearly did better than the unoperated group. Although the numbers are small, the difference is statistically significant (p < 0.05), and surgery appears to be the treatment of choice. For reasons that are not clear, no patients with type II The American Journal of Surgery
Aortic
dissections in this series were treated by the Wheat regimen. Surgically treated patients with type III dissections also appeared to have done better than unoperated patients, but again the difference is not statistically significant. Wheat treatment, as practiced in this institution, also does not appear to offer any long range therapeutic advantages over nonspecific supportive measures in the treatment of aortic dissection. It would appear to have a place in the early stabilization of dissection; however, the survival rate decreases markedly between one week and one month. These findings indicate that surgically treated patients with aortic dissection have a better survival at one year than unoperated patients whether treated by the Wheat regimen or by nonspecific supportive measures. This is particularly true of type II dissection and to a lesser extent to types I and III. In this institution [II], the usual approach has been to employ medical (Wheat) therapy first and to operate only when there is (1) failure to bring pain or blood pressure under control, (2) propagation or extension of the hematoma, (3) decompensated aortic regurgitation, and (4) occlusion or compromise of major branches of the aorta, or (5) saccular dilatation of the aorta. This may not be the best approach. Apparent early success of medical (Wheat) treatment was eroded with the passage of time, as indicated by the one month survival data. A more rational approach might be to operate electively sometime during the first week or two after dissection. The findings at autopsy in the forty-two patients who died confirm the significant association of hypertension (as manifested by cardiomegaly) with aortic dissection. Cardiomegaly was somewhat less common in patients with type III dissection than in those with the other two types and may be a factor in the better overall survival of patients with type III dissection. Another factor which is probably more important in this better prognosis is the lower incidence of aortic rupture, which may be due to the decreased mechanical trauma experienced by the rather immobile descending aorta as opposed to the torsional “bucket-and-handle” forces imposed on the ascending and transverse aorta with each systole. Reentry tears, long associated with a better prognosis, are more common in type III dissections and lead to the early surgical approach of creation of a reentry site [12]. This procedure was later shown to have little or no effect on prolongation of survival. It is our belief that reentry tears may simply occur in increasing numbers with time and are thus more common in the longer living type III patients, but in and of themselves exert no effect on prolongation of life. Volume 137, February 1879
Dissection
Summary
During the eleven year period from 1965 to 1976, aortic dissection was diagnosed in seventy-eight patients treated at the University of Virginia Medical Center. The seventy-four cases that comprise the present report were confirmed by aortography, surgery, or autopsy. Forty patients (54 per cent) were surgically treated by interruption of the origin of dissection and insertion of a Dacron prosthesis. The remaining thirty-four (46 per cent) were treated according to the Wheat regimen or by nonspecific supportive measures. The overall survival at one year was 28 per cent. One year survival for patients with type I dissections treated surgically was 19 per cent compared with 8 per cent one year survival for nonsurgically treated patients. Sixty per cent of patients with type II dissections treated surgically were alive at the end of,one year, whereas no patients with type II dissection treated nonsurgically survived beyond one year. Half the patients with type III dissections treated surgically were alive at one year compared with 35 per cent of those nonsurgically treated. These data suggest that surgery is the treatment of choice for all types of aortic dissections, but particularly for type II. Patients with type I dissections have a very poor prognosis regardless of therapy. References 1. DeBakey ME, Henly WS, Cooley DA, Morris GC Jr, Crawford ES, Beak AC Jr: Surgical management of dissecting aneurysms of the aorta. J Thorac Cardiovasc Surg 49: 1130, 1965. 2. Hirst AE Jr, Johns VJ Jr, Kime SW Jr: Dissecting aneurysm of the aorta: a review of 505 cases. Medicine 37: 217, 1958. 3. Warren WD, Beckwith J, Muller WH Jr: Problems in the surgical management of acute dissecting aneurysm of the aorta. Am Surg 144: 530, 1965. 4. Hume DM, Porter RR: Acute dissecting aortic aneurysms. Surgery 53: 122, 1963. 5. Wheat MW Jr, Palmer RF. Bartley TD, Seelman RC: Treatment of dissecting aneurysms of the aorta without surgery. J Thorac Cardiovasc Surg 50: 364, 1965. 6. Ringer RK: Influence of reserpine and early growth, blood pressure and dissecting aneurysms in turkeys, p 2 1. Conference on the Use of Serpasil in Animal and Poultry Production, May 7, 1969, Rutgers, The State University, New Brunswick, NJ. Summit, New Jersey, CIBA, 1969. 7. Gore I, Hirst AE Jr: Dissecting aneurysms of the aorta. Cardiovasc C/in 5: 239, 1973. 8. Lindsay J Jr, Hurst JW: Dissecting aneurysms of the aorta. JAMA 217: 1533, 1971. 9. Levinson DC. Edmeades DT, Griffith GC: Dissecting aneurysm of the aorta; its clinical electrocardiographic and laboratory features. Report of 58 autopsied cases. Circulation 1: 360, 1950. 10. Rider JA, Chriss JW, Herrmann CR: Dissecting aneurysm of the aorta; a ten year study. Tex A&d 46: 3 11, 1950. 11. Crosby IK, Golden GT: Acute dissecting aneurysms of the aorta. Va Med 102: 558, 1975. 12. DeBakey ME. Cooley DA, Creech 0 Jr: Surgical considerations of dissecting aneurysm of the aorta. Am Surg 142: 586, 1955.
243