Automated, multiplex assay for high-frequency microsatellite instability in colorectal cancer.

PURPOSE: In a series of hereditary nonpolyposis colorectal cancer (HNPCC) patients, we evaluated the sensitivities of the individual microsatellites recommended by the National Cancer Institute (NCI) consensus workshop for detection of high-frequency microsatellite instability (MSI-H). On the basis of this evaluation, we developed a three-marker assay that assigns microsatellite instability (MSI) in a multiplex polymerase chain reaction. METHODS: Individual marker sensitivity was assessed in 18 HNPCC tumors. Multiplex and NCI assays were then assessed in a series of 120 patients with early-onset colon cancer. RESULTS: The sensitivity of microsatellite markers BAT25, BAT26, D2S123, D5S346, and D17S250 for ASI in HNPCC cancers was 100%, 94%, 72%, 50%, and 50%, respectively. The three most accurate markers were combined and optimized in a multiplex assay that assigned MSI-H whenever at least two of three markers revealed ASI. In early-onset colon cancers, the prevalence of MSI-H determined by the multiplex assay and by the NCI assay was 16% and 23%, respectively. The additional MSI-H tumors and patients with MSI-H identified by the NCI assay lacked the traits characteristic of MSI-H seen in tumors and patients identified by the multiplex assay: retention of heterozygosity (NCI additional 22% v multiplex 84%; P =.003), characteristic tumor morphology (0% v 64%; P =.006), and 5-year cancer survival rate (44% v 100%; P =.0003). CONCLUSION: The multiplex assay identifies colon cancers with MSI-H by assessing three highly accurate microsatellite markers. This assay identifies a smaller MSI-H cohort with more homogeneous clinical features and is superior as a marker of favorable prognosis. It merits prospective evaluation as a marker of prognosis and as a screening test for HNPCC.

Ischemic mitral valve reconstruction and replacement: comparison of long-term survival and complications.

OBJECTIVE: This study reviews the 223 consecutive mitral valve operations for ischemic mitral insufficiency performed at New York University Medical Center between January 1976 and January 1996. The results for mitral valve reconstruction are compared with those for prosthetic mitral valve replacement. METHODS: From January 1976 to January 1996, 223 patients with ischemic mitral insufficiency underwent mitral valve reconstruction (n = 152) or prosthetic mitral valve replacement (n = 71). Coronary artery bypass grafting was performed in 89% of cases of mitral reconstruction and 80% of cases of prosthetic replacement. In the group undergoing reconstruction, 77% had valvuloplasty with a ring annuloplasty and 23% had valvuloplasty with suture annuloplasty. In the group undergoing prosthetic replacement, 82% of patients received bioprostheses and 18% received mechanical prostheses. RESULTS: Follow-up was 93% complete (median 14.6 mo, range 0-219 mo). Thirty-day mortality was 10% for mitral reconstruction and 20% for prosthetic replacement. The short-term mortality was higher among patients in New York Heart Association functional class IV than among those in classes I to III (odds ratio 5.75, confidence interval 1.25-26.5) and was reduced among patients with angina relative to those without angina (odds ratio 0.26, confidence interval 0.05-1.2). The 30-day death or complication rate was similarly elevated among patients in functional class IV (odds ratio 5.53; confidence interval 1.23-25.04). Patients with mitral valve reconstruction had lower short-term complication or death rates than did patients with prosthetic valve replacement (odds ratio 0.43, confidence interval 0.20-0.90). Eighty-two percent of patients with mitral valve reconstruction had no insufficiency or only trace insufficiency during the long-term follow-up period. Five-year complication-free survivals were 64% (confidence interval 54%-74%) for patients undergoing mitral valve reconstruction and 47% (confidence interval 33%-60%) for patients undergoing prosthetic valve replacement. Results of a series of statistical analyses suggest that outcome was linked primarily to preoperative New York Heart Association functional class. CONCLUSIONS: Initial mortalities were similar among patients undergoing prosthetic replacement and valve reconstruction. Poor outcome was primarily related to preexisting comorbidities. Patients undergoing valve reconstruction had fewer valve-related complications. Valve reconstruction resulted in excellent durability and freedom from complications. These findings suggest that mitral valve reconstruction should be considered for appropriate patients with ischemic mitral insufficiency.

Minimally invasive versus sternotomy approaches for mitral reconstruction: comparison of intermediate-term results.

BACKGROUND: This study compares intermediate-term outcomes of mitral valve reconstruction after either the standard sternotomy approach or the new minimally invasive approach. Although minimally invasive mitral valve operations appear to offer certain advantages, such as reduced postoperative discomfort and decreased postoperative recovery time, the intermediate-term functional and echocardiographic efficacy has not yet been documented. METHODS: From May 1996 to February 1999, 100 consecutive patients underwent primary mitral reconstruction through a minimally invasive right anterior thoracotomy and peripheral cardiopulmonary bypass and Port-Access technology (Heartport, Inc, Redwood City, Calif). Outcomes were compared with those for our previous 100 patients undergoing primary mitral repair who were operated on with the standard sternotomy approach. RESULTS: Although patients were similar in age, the patients undergoing the minimally invasive approach had a lower preoperative New York Heart Association classification (2.1 +/- 0.5 vs 2.6 +/- 0.6, P <.001). There was one (1.0%) hospital mortality with the sternotomy approach and no such case with the minimally invasive approach. Follow-up revealed that residual mitral insufficiency was similar between the minimally invasive and sternotomy approaches (0.79 +/- 0.06 vs 0.77 +/- 0.06, P =.89, 0- to 3-point scale); likewise, the cumulative freedom from reoperation was not significantly different (94.4% vs 96.8%, P =.38). Follow-up New York Heart Association functional class was significantly better in the patients undergoing the minimally invasive approach (1.5 +/- 0.05 vs 1.2 +/- 0.05, P <.01). CONCLUSIONS: These findings demonstrate comparable 1-year follow-up results after minimally invasive mitral valve reconstruction. Both echocardiographic results and New York Heart Association functional improvements were compatible with results achieved with the standard sternotomy approach. The minimally invasive approach for mitral valve reconstruction provides equally durable results with marked advantages for the patient and should be more widely adopted.

Surgical debulking and intraperitoneal chemotherapy for established peritoneal metastases from colon and appendix cancer.

BACKGROUND: Aggressive treatment of peritoneal metastases from colon cancer by surgical cytoreduction and infusional intraperitoneal (IP) chemotherapy may benefit selected patients. We reviewed our institutional experience to assess patient selection, complications, and outcome. METHODS: Patients having surgical debulking and IP 5-fluoro-2'-deoxyuridine (FUDR) plus leucovorin (LV) for peritoneal metastases from 1987 to 1999 were evaluated retrospectively. RESULTS: There were 64 patients with a mean age of 50 years. Primary tumor sites were 47 in the colon and 17 in the appendix. Peritoneal metastases were synchronous in 48 patients and metachronous in 16 patients. Patients received IP FUDR (1000 mg/m2 daily for 3 days) and IP leucovorin (240 mg/m2) with a median cycle number of 4 (range, 1-28). The median number of complications was 1 (range, 0-5), with no treatment related mortality. Only six patients (9%) required termination of IP chemotherapy because of complications. The median follow-up was 17 months (range, 0-132 months). The median survival was 34 months (range, 2-132); 5-year survival was 28%. Lymph node status, tumor grade, and interval to peritoneal metastasis were not statistically significant prognostic factors for survival. Complete tumor resection was significant on multivariate analysis (P = .04), with a 5-year survival of 54% for complete (n = 19) and 16% for incomplete (n = 45) resection. CONCLUSIONS: Surgical debulking and IP FUDR for peritoneal metastases from colon cancer can be accomplished safely and has yielded an overall 5-year survival of 28%. Complete resection is associated with improved survival (54% at 5 years) and is the most important prognostic indicator.

Late results of isolated mitral annuloplasty for "functional" ischemic mitral insufficiency.

BACKGROUND: Repair of functional ischemic mitral regurgitation (MR) due to annular deformity and leaflet restriction remains a challenge for the surgeon and lacks well-documented outcomes. We investigated outcomes in the treatment of functional ischemic MR corrected by annuloplasty techniques alone. METHODS: From May 1980 to July 1999, 174 patients underwent repair for functional ischemic mitral insufficiency with annuloplasty alone (128 ring annuloplasty; 46 suture annuloplasty). Acute insufficiency was present in 25 (14.4%). Concomitant procedures included CABG (n = 152; 87.4%). Patients were studied longitudinally with annual follow-up and echocardiograms. RESULTS: Overall hospital mortality was 17.8% and was increased by NYHA Class 4 (23.8% vs. 8.7%; p = 0.011), diabetes (25.0% vs. 13.6%; p = 0.059), and chronic mitral insufficiency (16.4% vs. 8.0%; p = 0.070). Multivariate analysis revealed age (beta = 0.099; p = 0.049) and ejection fraction < 30% (beta = 1.260; p = 0.097) as significant predictors of hospital death. Mean postoperative mitral insufficiency was 0.84 +/- 0.86 (scale of 0-4). NYHA Class 4 (beta = 2.33; p = 0.034) and simple suture annuloplasty (beta = 2.08; p = 0.07) were associated with increased risk of late cardiac death. Cumulative incidence of mitral reoperation was 7.7% at 5 years. At follow-up, 89.7% of patients were in NYHA Class 1 or 2 with 83.4% having none or only mild mitral insufficiency. CONCLUSIONS: Ring annuloplasty is associated with a survival benefit when compared to simple suture repair in ischemic patients who require annuloplasty alone to correct the MR. Mitral reconstruction with a ring annuloplasty offers durable results in this homogeneous subset of functional ischemic MR patients. Ischemic mitral insufficiency is associated with significant late mortality.

Robotic-assisted instruments enhance minimally invasive mitral valve surgery.

BACKGROUND: The potential for totally endoscopic mitral valve surgery has been advanced by the development of minimally invasive techniques. Recently surgical robots have offered instrument access through small ports, obviating the need for a significant thoracotomy. This study tested the hypothesis that a microsurgical robot with 5 degrees of freedom is capable of performing an endoscopic mitral valve replacement (MVR). METHODS: Dogs (n = 6) were placed on peripheral cardiopulmonary bypass; aortic occlusion was achieved with endoaortic clamping and transesophageal echocardiographic control. A small left seventh interspace "service entrance" incision was used to insert sutures, retractor blade, and valve prosthesis. Robotically controlled instruments included a thoracoscope and 5-mm needle holders. MVR was performed using an interrupted suture technique. RESULTS: Excellent visualization was achieved with the thoracoscope. Instrument setup required 25.8 minutes (range 12 to 37); valve replacement required 69.3+/-5.39 minutes (range 48 to 78). MVR was accomplished with normal prosthetic valve function and without misplaced sutures or inadvertent injuries. CONCLUSIONS: This study demonstrates the feasibility of adjunctive use of robotic instrumentation for minimally invasive MVR. Clinical trials are indicated.

Assisted venous drainage presents the risk of undetected air microembolism.

OBJECTIVES: The proliferation of minimally invasive cardiac surgery has increased dependence on augmented venous return techniques for cardiopulmonary bypass. Such augmented techniques have the potential to introduce venous air emboli, which can pass to the patient. We examined the potential for the transmission of air emboli with different augmented venous return techniques. METHODS: In vitro bypass systems with augmented venous drainage were created with either kinetically augmented or vacuum-augmented venous return. Roller or centrifugal pumps were used for arterial perfusion in combination with a hollow fiber oxygenator and a 40-micrometer arterial filter. Air was introduced into the venous line via an open 25-gauge needle. Test conditions involved varying the amount of negative venous pressure, the augmented venous return technique, and the arterial pump type. Measurements were recorded at the following sites: pre-arterial pump, post-arterial pump, post-oxygenator, and patient side. RESULTS: Kinetically augmented venous return quickly filled the centrifugal venous pump with macrobubbles requiring continuous manual clearing; a steady state to test for air embolism could not be achieved. Vacuum-augmented venous return handled the air leakage satisfactorily and microbubbles per minute were measured. Higher vacuum pressures resulted in delivery of significantly more microbubbles to the "patient" (P <.001). The use of an arterial centrifugal pump was associated with fewer microbubbles (P =.02). CONCLUSIONS: Some augmented venous return configurations permit a significant quantity of microbubbles to reach the patient despite filtration. A centrifugal pump has air-handling disadvantages when used for kinetic venous drainage, but when used as an arterial pump in combination with vacuum-assisted venous drainage it aids in clearing air emboli.

Protruding aortic arch atheromas: risk of stroke during heart surgery with and without aortic arch endarterectomy.

BACKGROUND: Stroke occurs in 1% to 7% of heart surgery. Aortic arch atherosclerosis is a risk factor for intraoperative stroke, and endarterectomy has been proposed to prevent stroke during heart surgery in patients with arch atheromas. METHODS AND RESULTS: Intraoperative transesophageal echocardiography was performed in 3404 patients undergoing heart surgery between 1990 and 1996. Use of transesophageal echocardiography was unselected and based on equipment availability. Aortic arch atheromas (>/=5 mm, or mobile) were seen in 268 (8%) patients. They were evaluated for intraoperative stroke (confirmed by a neurologist and cerebral infarction on computed tomography or magnetic resonance imaging). Arch endarterectomy was performed in 43 patients as an adjunct to their cardiac procedure in an attempt to prevent intraoperative stroke. The intraoperative stroke rate in all 268 patients with atheromas was high (15.3%). On univariate analysis, age, previous stroke, and arch endarterectomy were significantly associated with intraoperative stroke. On multivariate analysis, age (odds ratio 3.9, P =.01) and arch endarterectomy (odds ratio 3.6, P =.001) were independently predictive of intraoperative stroke. Mortality rate in all 268 patients was high (14.9%). These patients with atheromas also had a long recovery room, intensive care unit, and total hospital length of stay (48 days). CONCLUSIONS: Patients with protruding aortic arch atheromas are at high risk for intraoperative stroke, significant and multiple morbidity, prolonged hospital stay, and death resulting from heart surgery. Aortic arch endarterectomy is strongly associated with intraoperative stroke; its use should be carefully considered in light of these results.

Choice of mitral prosthesis in the elderly. An analysis of actual outcome.

BACKGROUND: In younger patients requiring mitral valve replacement (MVR), mechanical prostheses (MPs) have been reported to give better freedom from all valve-related complications (VRCs) because of the high incidence of late valve degeneration (VD) associated with bioprostheses (BPs). In older patients, however, the risk of VD may be reduced because of the large competing risk of noncardiac death (NCD). Previous studies on VD in the elderly have used actuarial analysis, which overestimates the risk of VD in this population because it assumes that dead patients are still at risk. In contrast, cumulative incidence (actual) analysis acknowledges that patients who die have no risk of VD. This study compares the results of both "actual" and "actuarial" analyses of the freedom from VD in elderly patients undergoing MVR. METHODS AND RESULTS: From June 1976 through January 1996, 504 patients > or = 70 years of age underwent MVR at our institution. Isolated mitral operations were performed in 159 patients, and 169 had concomitant CABG. Hospital mortality was 59 of 374 (15.9%) for tissue prosthesis versus 24 of 130 (18.5%) for mechanical prosthesis (P = NS). For tissue versus mechanical prosthesis, 10-year freedom from noncardiac death was 75.0% versus 67.6% (P = NS); 10-year actuarial freedom from valve degeneration was 79.8% versus 93.4% (P = NS); 10-year actual freedom from valve degeneration was 92.6% versus 95.4% (P = NS); and 10-year actual freedom from all VRCs was 84.4% versus 92.3% (P = NS). CONCLUSIONS: In elderly patients undergoing MVR, actuarial analysis overestimates the 10-year risk of VD compared with actual analysis (20.2% versus 7.4% for BP, 6.6% versus 4.6% for MP). In these patients, the actual freedoms from VD and all VRCs do not differ significantly between BP and MP. Thus, in this age group, the necessity for anticoagulation or its avoidance may be the predominant factor in choosing a replacement mitral valve.

Valve repair versus replacement for mitral insufficiency: when is a mechanical valve still indicated?

OBJECTIVES: Although many advantages of mitral valve reconstruction have been demonstrated, whether specific subgroups of patients exist in whom mechanical valve replacement offers advantages over mitral reconstruction remains undetermined. METHODS: This study examined the late results of mitral valve surgery in patients with mitral insufficiency who received either a St. Jude Medical valve (n = 514) or a mitral valve reconstruction with ring annuloplasty (n = 725) between 1980 and 1996. RESULTS: Overall operative mortality was 7.2% in the patients receiving a St. Jude Medical mitral valve and 5.4% in those undergoing mitral valve reconstruction (no significant difference); isolated mortality was 2.5% in the St. Jude Medical group and 2.2% in the valve reconstruction group (no significant difference). The follow-up interval was more than 5 years for 340 patients with a mean of 39.8 months (98.5% complete). Overall 8-year freedom from late cardiac death, reoperation, and all valve-related complications was 72.8% for the St. Jude Medical group and 64.8% for valve reconstruction group (no significant difference). For patients with isolated, nonrheumatic mitral valve disease, 8-year freedom from late cardiac death and reoperation was better in the mitral valve reconstruction group (88.3%) than in the St. Jude Medical valve group (86.0%; p = 0.05). Furthermore, Cox proportional hazards regression revealed that mitral valve reconstruction was independently associated with a lesser incidence of late cardiac death (p = 0.04), irrespective of preoperative New York Heart Association class. However, the St. Jude Medical valve offered better 8-year freedom from late cardiac death, reoperation, and all valve-related complications than did mitral valve reconstruction in patients with multiple valve disease (77.0% vs 45.3%; p < 0.01). CONCLUSIONS: Therefore, mitral valve reconstruction appears to be the procedure of choice for isolated, nonrheumatic disease, whereas insertion of a St. Jude Medical valve should be preferred for patients with multiple valve disease.

Selective approach to descending thoracic aortic aneurysm repair: a ten-year experience.

BACKGROUND: A variety of surgical techniques has been developed to attempt to minimize the risk of paraplegia after descending thoracic aortic aneurysm repair. This study reviews our institutional experience with several basic techniques over a period of 10 years. METHODS: Seventy-eight consecutive patients underwent repair of descending thoracic aortic aneurysm between 1983 and 1993. Two basic repair strategies were used: (1) distal perfusion with somatosensory evoked potential monitoring (n = 54) and (2) cross-clamping (n = 24), alone (n = 6) or with controlled distal exsanguination (n = 18). RESULTS: The operative mortality rate was 6.5% for elective repair (n = 62), 25.0% for emergent repair (n = 16), and 10.3% overall. Univariate predictors of increased operative risk were emergent operation, rupture, and shock. Neither death nor paraplegia was related to the operative technique used. The incidence of paraplegia was 3.7% in perfused patients and 4.2% in cross-clamping patients (p > 0.05). Paraplegia did not occur after any elective operation (zero of 62) but occurred in 18.6% of emergent cases (p < 0.01). In perfused patients, paraplegia did not occur when the distal pressure was maintained above 55 mm Hg and somatosensory evoked potentials remained intact. When somatosensory evoked potentials were lost (n = 7) in perfused patients, the operative technique was altered successfully in 5 patients, whereas in 2 patients (28.6%), paraplegia developed. CONCLUSIONS: The risks associated with elective descending thoracic aortic aneurysm repair were extremely low using an operative strategy that was flexible but skewed toward perfusion with somatosensory evoked potential monitoring. In perfused patients, paraplegia did not occur when distal pressure was greater than 55 mm Hg and somatosensory evoked potentials remained intact. However, the risks of death and paraplegia were primarily related to emergent presentation, not to technique, and the technique of cross clamping with controlled distal exsanguination was found to be valuable in unstable or in anatomically complicated subsets of patients.

Evaluating neuroprotective agents for clinical anti-ischemic benefit using neurological and neuropsychological changes after cardiac surgery under cardiopulmonary bypass. Methodological strategies and results of a double-blind, placebo-controlled trial of GM1 ganglioside.

BACKGROUND AND PURPOSE: Many neuroprotective agents (NPAs) are effective in acute experimental cerebral ischemia in animals. None have proven effective in human stroke trials. Even short treatment delays cause substantial efficacy loss. Cardiac surgery under cardiopulmonary bypass (CS-CPB) causes cerebral ischemia with cognitive impairment at a predeterminable time point and should permit efficient screening of NPAs for stroke benefit. We sought to develop sensitive methods to assess dysfunction from CS-CPB in a double-blind trial of the NPA GM1 ganglioside. METHODS: Eighteen GM1 and 11 Control patients received GM1 300 mg or placebo, two doses intravenously, before nonemergency CS-CPB. Independent examiners administered structured neurological examinations and neuropsychological test batteries at Baseline and 1 day (Acute Postop; neurological only), 1 week (Early F/U), and > or = 6 months (Long-term F/U) postoperatively; using defined procedures they employed ordinal Clinical Change Scores (CCSs) to quantify neurological cerebral, neurological noncerebral, and neuropsychological performance changes. Several methods to analyze CCSs and neuropsychological test score changes were evaluated. RESULTS: The most sensitive indicators were the mean Acute Postop Neurologist's CCS-Cerebral (P < 10(-5)) and the mean Early F/U Neuropsychologist's CCS (P < .01), with statistically nonsignificant differences favoring GM1. No significant mean changes in Neurologist's CCS-Noncerebral or any Long-term F/U CCSs occurred. CCS distributions and neuropsychological test score mean changes showed similar temporal patterns, with less sensitivity to change. When, as usual in prior CS-CPB studies, impairment was defined by neuropsychological test score declines (increases ignored), results were spurious. CONCLUSIONS: The strokelike cerebral dysfunction (maximal acutely, with eventual recovery) that occurs after CS-CPB is useful to screen NPAs for clinical efficacy. CCSs based on detailed neurological examination and neuropsychological testing are sensitive measures; refinement of this approach should enhance the efficiency of the CS-CPB model. Further testing of GM1 is warranted.

Perioperative morbidity and mortality in combined vs. staged approaches to carotid and coronary revascularization.

Between 1986 and 1994 we identified 57 patients who underwent carotid endarterectomy (CEA) and coronary artery bypass grafting (CABG) during the same hospitalization. Simultaneous CABG and CEA was performed in 28 patients (mean age 70.5 years, 58% male). Indications for CABG in these patients were myocardial infarction in two crescendo angina in 19, congestive heart failure in two and left main or triple-vessel coronary artery disease noted during carotid preoperative evaluation in five. Indications for CEA were transient ischemic attack (TIA) in 12, crescendo TIA in six, cerebrovascular accident (CVA) in five, and asymptomatic stenosis in five. There were no postoperative myocardial infarctions or perioperative deaths. Two patients developed atrial fibrillation, and four patients had CVAs (two were ipsilateral to the side of CEA). Twenty-nine patients underwent staged procedures (i.e., not performed concomitantly but during the same hospitalization). Indications for CABG and CEA were comparable to those in the group undergoing simultaneous procedures. In 17 patients CEA was performed before CABG. There was a single CVA, the result of an intracerebral hemorrhage. Five of the 17 patients had a myocardial infarction and two died; one patient had first-degree heart block requiring a pacemaker. Four additional patients developed atrial fibrillation, one of whom required cardioversion. The remaining 12 patients had CABG followed by CEA. There were no CVAs, myocardial infarctions, arrhythmias, or deaths in this subgroup. These data demonstrate that the performance of simultaneous CABG and CEA procedures is associated with increased neurologic morbidity (14.3%), both ipsilateral and contralateral to the side of carotid surgery in contrast to staged CABG and CEA (3.4%). In addition, when staged carotid surgery preceded coronary revascularization in those with severe coronary artery disease, the combined cardiac complication and mortality rate was significantly higher than when coronary revascularization preceded CEA. This evidence suggests that when CABG and CEA must be performed during the same hospitalization, the procedures should be staged with CABG preceding CEA.

Surgical repair of type A aortic dissection by the circulatory arrest-graft inclusion technique in sixty-six patients.

During an 8-year period (1984 to 1991) 66 patients (mean age 59 years, range 26 to 84 years) with type A aortic dissection (60 ascending aorta tears, 6 arch tears; 35 acute, 31 chronic) had surgical repair by a continuous suture-graft inclusion technique. Hypothermic circulatory arrest (16 degrees C) was used in 58 patients (35/35 acute, 23/31 chronic; mean arrest time 26 minutes, range 10 to 55 minutes). Fifty-two patients had hemiarch repair and 6 had total arch replacement. Aortic valve disease necessitated treatment in 38 patients (1 valved conduit, 20 valve replacements, 17 valve repairs). Recently 11 patients had valve repair by reconstruction of the native aortic root, by means of techniques similar to those used for homograft valve insertion. Operative mortality was 9% (14% acute, 3% chronic). Stroke occurred in 2 patients (3%) and was fatal in both. Variables suggestive of increased operative risk by univariate analysis were acuteness (p = 0.12), visceral ischemia (p = 0.12), and preoperative shock (p = 0.13). No variable was significant by multivariate analysis. Overall actuarial survival at 48 months was 77%, with 3 late deaths from a ruptured distal aneurysm. Late computed tomography or magnetic resonance imaging scan was done in 28 patients at a mean interval of 33 months. These studies identified 1 patient with a pseudoaneurysm requiring reoperation and 3 patients with contained flow between the graft and the wrap. Three patients required late operation: 1 for pseudoaneurysm, 1 for arch dissection, and 1 for repair of a distal aneurysm.

Aortic dissection complicating cardiac surgery: diagnosis by intraoperative biplane transesophageal echocardiography.

Aortic dissection is a rare but devastating complication of cardiopulmonary bypass. Intraoperative transesophageal echocardiography can be a useful technique to define the anatomy of the dissection, to evaluate its extension and progression, and to detect the presence of aortic insufficiency. We describe two cases in which transesophageal echocardiography helped in making a rapid diagnosis of aortic dissection during cardiac surgery and demonstrate how it may play a role in therapeutic decision making.