Search

The management of syringomyelia in association with hydrocephalus can be a challenging pathology to treat with durable results. We present a durable treatment of a patient with symptomatic, idiopathic hydrocephalus, Chiari type I malformation, and syringomyelia with an endoscope-assisted placement of a single syringo-panventricular multiperforated shunt catheter via a transfrontal transaqueductal approach with simultaneous endoscopic third ventriculostomy (ETV). We discuss the technical details required for this single-shunt approach to a complex pathology, and indications for appropriate use of this novel approach in experienced hands. The patient rapidly improved radiographically and symptomatically, with a functioning shunt twelve years postoperatively. For surgeons well-versed in endoscopic techniques, this approach can be safe, durable, and obviate the need for an otherwise complex shunting system with multiple catheters, or a more traditional invasive craniovertebral decompression, in a patient with idiopathic hydrocephalus and a syrinx.

Deep brain stimulation (DBS) of the ventral intermediate nucleus (VIM-DBS) is a highly successful treatment for medication-refractory essential tremor (ET). Clinical outcomes are dependent on accurate targeting. Here, we aim to develop a framework for connectivity-guided DBS targeting by evaluating probabilistic tractography and clinical response at both initial programming (IP) and clinical follow-up (CF).Magnetic resonance imaging and clinical outcomes were evaluated in 23 ET patients who were treated by VIM-DBS at the University of California Los Angeles (20 at IP, 18 at CF, 14 at both). Lead-DBS was used to model the volume of tissue activated tissue (VTA) based on programming configurations at both IP and CF. Probabilistic tractography, calculated in FSL, was used to evaluate 1) clinically weighted whole brain connectivity of VTA; 2) connectivity between VTA and freesurfer-derived target regions of interest (ROI) including primary motor, premotor, and prefrontal cortices, and cerebellum; and 3) volume of intersection between VTA and probabilistic tractography-based segmentation of the thalamus. At IP, individual contacts were scored as high or low efficacy based on acute tremor improvement. At CF, clinical response was measured by percent of change of the Clinical Rating Scale for Tremor (CRST) compared to preoperative scores. Contributions from each target ROI to clinical response was measured using logistic regression for IP and linear regression for CF.The clinically weighted map of whole brain connectivity of VTA shows preferential connectivity to precentral gyrus and brainstem/cerebellum. The volume of intersection between VTA and thalamic segmentation map based on probabilistic connectivity to primary motor cortex was a significant predictor of contact efficacy at IP (OR = 2.26 per 100 mm3 of overlap, p = .04) and percent change in CRST at CF (β = 14.67 per 100 mm3 of overlap, p = .003).Targeting DBS to the area of thalamus most connected to primary motor cortex based on probabilistic tractography is associated with superior outcomes, providing a potential guide not only for lead targeting but also therapeutic programming.

Introduction: The amygdala is known to play a role in mediating emotion and possibly addiction. We used probabilistic tractography (PT) to evaluate whether structural connectivity of the amygdala to the brain reward network is associated with impulsive choice and tobacco smoking.Methods: Diffusion and structural MRI scans were obtained from 197 healthy subjects (45 with a history of tobacco smoking) randomly sampled from the Human Connectome database. PT was performed to assess amygdala connectivity with several brain regions. Seed masks were generated, and statistical maps of amygdala connectivity were derived. Connectivity results were correlated with a subject performance both on a delayed discounting task and whether they met specified criteria for difficulty quitting smoking.Results: Amygdala connectivity was spatially segregated, with the strongest connectivity to the hippocampus, orbitofrontal cortex (OFC), and brainstem. Connectivity with the hippocampus was associated with preference for larger delayed rewards, whereas connectivity with the OFC, rostral anterior cingulate cortex (rACC), and insula were associated with preference for smaller immediate rewards. Greater nicotine dependence with difficulty quitting was associated with less hippocampal and greater brainstem connectivity. Scores on the Fagerstrom Test for Nicotine Dependence (FTND) correlated with rACC connectivity.Discussion: These findings highlight the importance of the amygdala-hippocampal-ACC network in the valuation of future rewards and substance dependence. These results will help to identify potential targets for neuromodulatory therapies for addiction and related disorders.

BackgroundParkinson disease (PD) patients have difficulty with self-initiated (SI) movements, presumably related to basal ganglia thalamocortical (BGTC) circuit dysfunction, while showing less impairment with externally cued (EC) movements.ObjectivesWe investigate the role of BGTC in movement initiation and the neural underpinning of impaired SI compared to EC movements in PD using multifocal intracranial recordings and correlating signals with symptom severity.MethodsWe compared time-resolved neural activities within and between globus pallidus internus (GPi) and motor cortex during between SI and EC movements recorded invasively in 13 PD patients undergoing deep brain stimulation implantation. We compared cortical (but not subcortical) dynamics with those recorded in 10 essential tremor (ET) patients, who do not have impairments in movement initiation.ResultsSI movements in PD are associated with greater low-beta (13–20 Hz) power suppression during pre-movement period in GPi and motor cortex compared to EC movements in PD and compared to SI movements in ET (motor cortex only). SI movements in PD are uniquely associated with significant low-beta pallidocortical coherence suppression during movement execution that correlates with bradykinesia severity. In ET, motor cortex neural dynamics during EC movements do not significantly differ from that observed in PD and do not significantly differ between SI and EC movements.ConclusionThese findings implicate low beta BGTC oscillations in impaired SI movements in PD. These results provide a physiological basis for the strategy of using EC movements in PD, circumventing diseased neural circuits associated with SI movements and instead engaging circuits that function similarly to those without PD.

Introduction: The relationship between Parkinson Disease (PD) pathology, dopamine replacement therapy (DRT), and impulse control disorder (ICD) development is still incompletely understood. Given the sensorimotor-lateral substantia nigra (SN) selective degeneration associated with PD, we posit that a relative sparing of the limbic-medial SN in the context of DRT drives impulsive, reward-seeking behavior in PD patients with recent history of severe impulsivity. Methods: Impulsive and control participants were selected from a consecutive list of PD patients receiving pre-operative deep brain stimulation (DBS) planning scans including 3T structural MRI and 64 direction diffusion tensor imaging (DTI). Using previously identified substantia nigra (SN) subsegment network connectivity profiles to develop classification targets, split-hemisphere target-based SN segmentation with probabilistic tractography was performed. The relative subsegment volumes and strength of connectivity between the SN and the limbic, associative, and motor network targets were compared. Results: Our results show that there is greater probability of connectivity between the SN and limbic network targets relative to motor and associative network targets in PD patients with recent history of severe impulsivity as compared to PD patients without impulsivity (P = 0.0075). We did not observe relative volumetric subsegment differences across groups. Conclusion: Firstly, our results suggest that fine-grained, atlas-derived classification targets may be used in PD to parcellate and classify functionally distinct subsegments of the SN, with the apparent preservation of previously reported topographical limbic-medial SN, associative-ventral SN, and sensorimotor-lateral SN orientation. We suggest that relative, as opposed to absolute, degeneration amongst SN-associated dopaminergic networks relates to the impulsivity phenotype in PD.

Background and Aim of the Study: Functional ischemic mitral regurgitation (MR) can occur secondary to coronary artery disease. Controversy exists regarding management of these patients. Mitral valve annuloplasty in conjunction with coronary artery bypass grafting (CABG), accepted as the best treatment for severe MR, has been disputed for lesser degrees of regurgitation due to higher mortality. The results of a combined procedure approach were reviewed., Methods: Between February 1992 and June 1999, 100 consecutive patients (mean age 67+/-11 years) with functional ischemic MR underwent mitral valve repair + CABG. The repair was limited to a Duran flexible annuloplasty ring. Among patients, 72% had a preoperative myocardial infarction and 51% required perioperative intra-aortic balloon pump. NYHA functional class was III-IV in 72%; preoperative MR by transesophageal echocardiography (TEE) was grade 3-4+ in 80% and grade 2+ in 20%., Results: Intraoperative completion TEE indicated 0-1+ MR in 98%. Early mortality was 12% and late mortality 14%, for an overall survival of 74%. The mean follow up was 35.8 months. Follow up TEE on 82% of patients showed zero to trivial MR in 42% of patients, grade 1+ MR in 29%, 2+ MR in 24%, and 3-4+ MR in 5%. Follow up NYHA class was I-II in 81% of patients and III-IV in 19%. A significant correlation was found between recurrent MR and declining left ventricular function on follow up only, as well as the occurrence of preoperative myocardial infarction., Conclusion: Functional ischemic MR remains a difficult problem to treat, and has a poor long-term outcome. Ring annuloplasty for functional ischemic MR with coronary artery disease achieves immediate valve competence. However, a significant number of patients develop recurrent MR at intermediate follow up.

Background: The pulmonary autograft procedure for the treatment of aortic valve disease was developed and performed by Ross in 1967. The results he published in 1987 prompted increasing interest in the procedure. The International Registry of the Ross Procedure was established in 1993 to further examine longitudinal clinical outcomes., Methods: The results from the Ross registry document the continued and growing interest in the procedure with 2,523 patients currently enrolled, representing 122 centers and 166 surgeons worldwide., Results: Mortality (1987 to present) reported in the registry is 2.5%. It should be noted that follow-up stands at 70%. The most important issues for the registry to track are the incidence of reoperation for autograft failure and the fate of the pulmonary homograft. Reoperation for all valve-related problems is low (5.4%), with an autograft explant rate of 1.9%. Overall registry data indicate that the right ventricular outflow tract revision rate is 2.8%, with this decreasing by half to 1.3% in the 1987 to present subgroup., Conclusions: Rigorous analysis of outcomes is difficult with registry follow-up currently at 70%; however, the general conclusions derived from the registry are supported by other individual series with excellent follow-up. Success of the registry depends on judicious efforts by all participating surgeons and coordinators in documenting long-term patient results and reporting them to the registry.

The issue of bioprosthetic valve durability has become of critical importance as the number of elderly patients requiring valve operation has continued to increase. Our previous study showed bioprosthetic valve durability to be in excess of 83% at 13 years for patients 70 years of age and older at the time of implantation. There is limited follow-up data in the literature beyond this time point, however. Accordingly a retrospective analysis was conducted of all patients with bioprosthetic valves who were 70 years of age and over at the time of implantation. From September 1974 to April 1994, 1007 patients 70 years of age and over underwent valve replacement using a porcine bioprosthesis. The patients ranged in age from 70 to 104 years (mean, 75.6 +/- 4.3 years). There were 549 men (54.5%) and 458 women (45.5%). Preoperatively 98.8% of the patients were in New York Heart Association functional class III or IV. Operation was performed as an emergency in 66 patients (6.6%). The hospital mortality was 10.9% (110 patients), with 897 hospital survivors. There were 961 valves at risk. Follow-up extended from 1 month to 18.8 years (mean, 56.6 months). The cumulative follow-up is 4232.3 patient-years. A total of 31 valves failed, 12 in the aortic position and 19 in the mitral position (p < 0.0024). The causes of valve failure have included structural deterioration (16 valves), prosthetic endocarditis (7 valves), nonstructural dysfunction (5 valves), prosthetic thrombosis (1 valve), and other (2 valves).(ABSTRACT TRUNCATED AT 250 WORDS)

The number of patients undergoing valve replacement and concomitant coronary artery bypass grafting (CABG) is increasing. To further evaluate the indications for the use of the porcine bioprosthesis, this retrospective comparative analysis of valve structural deterioration was conducted in patients with and without concomitant CABG. From September 1974 to October 1993, 1,567 patients underwent valve replacement using a porcine xenograft. The series was divided into two groups: patients with isolated valve replacement (VR; n = 876) and those with VR and CABG (VR + CABG; n = 691). Aortic valve replacement was performed in 938 patients, mitral valve in 518, tricuspid in 2, and multiple valve replacement in 109 patients. The mean age for the series was 70.7 years (range, 50 to 104 years). The hospital mortality was 8.8% (138 patients). The hospital mortality for the VR group was 7.4% (65 deaths) and the VR + CABG group, 10.6% (73 deaths) p = 0.0365. There were 1,429 patients discharged from the hospital with 1,489 valves at risk. Follow-up extended from 1 month to 17.9 years with a mean of 66.9 months and was 98.3% complete. The cumulative follow-up was 7,927.1 patient-years. Structural deterioration was found to be significantly greater in the VR group for the age category 50 to 59 years (p = 0.0121) and the 60 to 69 years (p = 0.0230). No significant difference in the rate of structural deterioration was found for the two groups for the age category 70 years and older.(ABSTRACT TRUNCATED AT 250 WORDS)

The internal mammary artery has become the conduit of choice in myocardial revascularization. The expanded use of this ideal conduit for sequential grafting has enhanced its application. Between March 1985 and June 1993, 245 consecutive patients underwent revascularization of the myocardium with internal mammary artery bypass grafts with at least one sequential anastomosis. There were 186 men and 59 women, with a mean age of 65.1 (range 40-82) years. Unstable angina was present in 141 patients (57.6%) and 36 patients (14.7%) had left main coronary artery stenosis (> 50%). Before surgery, five patients (2.0%) were in New York Heart Association (NYHA) class II, 113 (46.1%) in class III, and 127 (51.8%) in class IV. There were a total of 1041 coronary artery grafts, mean 4.2 (range 2-7) grafts per patient and 528 sequential left internal mammary artery anastomoses, mean 2.2 per patient. Hospital mortality rate was 2.4% (six patients). Almost two-thirds of the patients experienced no hospital complications. The most frequent complication included arrhythmia in 36 patients (14.7%), respiratory insufficiency in 15 (6.1)% and temporary left phrenic nerve palsy in ten (4.1%). Mean follow-up was 37.0 (range 1-94.2) months. The mean(s.e.m.) actuarial survival rate for patients discharged from hospital was 94.8(1.6)% at 36 months and 82.4(5.0)% at 72 months. At follow-up of 222 patients, 185 (83.3%) were symptom-free in NYHA class I and 27(12.2%) were in class II. Though technically demanding, multiple sequential internal mammary artery grafting is feasible and can be accomplished with low hospital mortality and morbidity.(ABSTRACT TRUNCATED AT 250 WORDS)

Aortic valve replacement is the treatment of choice for elderly patients with aortic stenosis. It can be accomplished with excellent immediate and long-term results with significant functional improvement. Unfortunately, the literature is replete with enthusiastic reports of aortic catheter balloon valvotomy. Initial findings were controversial and the long-term results have been poor due to early valve restenosis. A retrospective analysis of our surgical experience with aortic valve replacement in the elderly seems appropriate in an effort to put this issue in proper perspective. From January 1973 to June 1993, 200 consecutive patients 70 years of age and older with severe aortic stenosis underwent surgical correction. There were 105 men (52.5%) and 95 women (47.5%), with a mean age of 76.2 years (range 70 to 89). Preoperatively, 195 patients (97.5%) were in New York Heart Association (NYHA) Class III or IV. Over one half (61.0%) of the patients experienced no hospital complications. The hospital mortality was 9.0% (18 patients). This included 14 patients in NYHA Class IV. The mean follow-up was 69.3 months and ranged from 1 to 215 months. The actuarial survival for 182 patients discharged from the hospital was 70.8% +/- 4.0% (SEM) at 72 months (73 patients at risk) and 35.2% +/- 5.4% at 144 months (20 patients at risk). Considering the advanced age and preoperative functional classification in this patient group, the results of aortic valve replacement have been excellent. The survival of patients discharged from the hospital compares favorably with a normal population matched for age and sex. The results of aortic balloon valvotomy have been disappointing.(ABSTRACT TRUNCATED AT 250 WORDS)

The expanded use of autografts for aortic valve replacement has focused attention on developing an improved implantation technique with the aim of reducing the incidence of early insufficiency. While the technique for total root replacement with nonstented tissue valves is a proven and established method, it is, however, not generally accepted for use in all clinical circumstances. Through the use of a partial inclusion technique, we propose to modify the standard method of extended aortic root replacement and pedicle coronary implantation. This approach also has the advantage of leaving the recipient aortic root intact. Unlike traditional scalloped subcoronary homograft implantation, it does not enclose the transplanted valve totally within the recipient aorta. Thus, there is preservation of the patient's aorta without distortion of the transplanted valve which occurs when it is forced completely into the closed recipient aortic root. While this partial inclusion method is not as simple as total root replacement, preservation of the recipient root is generally more acceptable to implanting surgeons. Either perfection of this proposed method or the acceptance of a traditional extended aortic root replacement will result in correction of the persistent problem of early aortic insufficiency following the use of autograft valves. Decreasing valve incompetence will have the desired effect of increasing the indication for the use of autografts for aortic valve replacement in young patients.

The problem of early onset aortic insufficiency as seen with the scalloped, subcoronary homograft aortic valve replacement is reduced with the use of a total root replacement. In addition, the naturally competent aortic root is more durable. From September 1985 to April 1991, 26 consecutive patients underwent aortic root replacement with 10 autografts, 14 homografts, and 2 xenografts using a modified implantation method. Twenty-five patients were discharged from the hospital. This partial inclusion root technique for implanting unstented valves in the aortic position decreases the probability of early failure secondary to technical malalignment at the time of implantation. In contrast to total root replacement, it avoids the need to destroy the recipient aortic root. A longitudinal aortotomy is performed to the aortic annulus in the mid-portion of the noncoronary sinus. The proximal suture line is interrupted with the valve oriented in the anatomical position. Circumferential running monofilament side-to-side anastomoses approximate the donor coronary ostia to the recipient. A running medial and lateral posterior suture line to the lateral superior portions of the aortotomy completes the integrity of the anterior wall of the implantation. One autograft attempt failed and one homograft patient died postoperatively. Follow-up ranges from 1 to 6 years in 24 patients. Postoperative aortic insufficiency was significant in one case due to inappropriate sizing of the proximal aortic suture line. There has been no evidence of progressive aortic insufficiency detected by the early onset of diastolic murmurs or echocardiograms as was our previous experience with the scalloped subcoronary method.

Since the advent of homograft aortic valve replacement surgery in the early 1960s, this procedure has been plagued by early aortic insufficiency. This problem has not been observed with total aortic root replacement. From September 1985 to April 1991, a modified method was used in 25 of 39 consecutive patients having aortic root replacement--seven having autografts, 30 having homografts, and two having xenografts. This technique is a new approach for implanting unstented valves in the aortic position. It decreases the probability of early failure resulting from technical malalignment of the valve during implantation. Further, it avoids the need to destroy the recipient aortic root. Whether the valve being used is an autograft, homograft, or xenograft, this method standardizes the insertion technique regardless of the anatomy or disease. Salient features of the method include the following: a longitudinal aortotomy to the aortic anulus in the midportion of the noncoronary sinus; a proximal interrupted suture line with the valve oriented in the anatomic position; and circumferential running monofilament side-to-side approximation of the donor coronary ostia to the recipient coronary ostia. The two anterior commissures are left untethered by nonclosure of the recipient aortotomy. Preliminary results have been impressive, with follow-up ranging from 1 to 3 years in 30 of these patients. Postoperative aortic insufficiency was significant in only two cases. There has been no progression of aortic insufficiency detectable by diastolic murmur or echocardiogram. Late complications have been minimal, and the overall clinical results have been excellent.

Stent mounting of homograft valves was first reported by our group in 1968. Since then, there has been question as to whether or not stent mounting of bioprostheses adversely affects the incidence of structural deterioration in aortic valve replacement. Between November 1967 and July 1988, 571 consecutive patients underwent valve replacement with a stented or unstented homograft. There were 351 men (61.5%) and 220 women (38.5%). The mean age of the group was 49.2 years (range 18 to 79 years). Five hundred thirty-four patients left the hospital (30-day overall mortality 6.5%). Follow-up extends from 6 months to 22 years with a mean of 7.6 years. The cumulative follow-up for the series was 4,095.9 patient-years. Hospital mortality, early technical failure, and prosthetic valve endocarditis were considered censoring events and excluded from this study. Actuarial analysis revealed a significant difference (p less than 0.02) in the freedom from structural valve deterioration for unstented and stended isolated aortic valve replacement. Age (50 and under, and over 50) does not appear to be a factor in structural deterioration in unstented homografts but does influence the rate of failure in stented homografts (p less than 0.05). These results clearly indicate that stent mounting adversely affects tissue valve durability with aortic valve replacement. Moreover, age correlates with structural deterioration if valves are stented and does not if they are unstented. Based on these results, the use of unstented bioprostheses should be reevaluated, along with the design of porcine valve stents.

With an increasing number of elderly patients requiring cardiac valve surgery, the topic of bioprosthetic durability becomes critically important. Previous reports have shown expected survival of bioprosthetic valves to be in excess of 95% at 9 years. However, primary tissue failure appears to accelerate at the end of the first decade and there is limited data into the second decade. With this in mind, we proceeded to analyze all bioprosthetic valves implanted in patients 70 years of age and older. From September 1974 to December 1990, 781 patients underwent valve replacement using a bioprosthesis. Ages ranged from 70 to 88 years with a mean of 75.1. There were 423 males (54.2%) and 358 females (45.8%). Preoperatively, 99.0% of the patients were in either New York Heart Association functional Class III or IV. Fifty-nine patients (7.6%) were done as emergencies. Six hundred ninety-four patients left the hospital (30-day overall mortality 11.1%). In this cohort, there were 733 valves at risk. Follow-up extended from 1 to 186.0 months with a mean of 52.9, which resulted in 3,059.9 patient-years of cumulative follow-up. Bioprosthetic Survival: A total of 23 valves failed in the series; 15 primary tissue failures, seven from endocarditis, and one perivalvular leak. Actuarial survival at 7 years was 94.5% +/- 1.4% standard error of the mean (SEM; 168 valves at risk) and at 13 years, 83.7% +/- 4.8% SEM (11 valves at risk). This analysis provides further documentation of the long-term favorable durability of the bioprosthesis when utilized in patients 70 years of age and over.

Two cases of cardiogenic shock and pulmonary edema due to acute, severe, silent mitral regurgitation are discussed. The mechanism for the mitral regurgitation was papillary muscle rupture in the setting of acute myocardial infarction. Echocardiography established the presence, severity, and cause of the mitral regurgitation and the associated hyperdynamic left ventricular function in the setting of cardiogenic shock. Transesophageal echocardiography is excellent for assessing the mitral valve in critically ill patients in whom transthoracic echocardiography may be inadequate or misleading. This allowed for emergency mitral valve replacement without prolonged attempts at medical stabilization.

From November, 1972, through December, 1986, 219 consecutive patients 70 years of age and older with aortic stenosis (AS) underwent aortic valve replacement. One hundred seven of them had isolated pure AS, and 112 had AS and coronary artery disease (AS + CAD). The mean age of the AS group was 75.4 years (range, 70 to 88 years) and of the AS + CAD group, 74.8 years (range, 70 to 86 years). The mean aortic valve gradient in the AS group was 87.7 +/- 30.6 mm Hg and in the AS + CAD group, 68.0 +/- 51.3 mm Hg (p less than 0.001). Hospital mortality for the AS group was 12.1% (13 patients) and for the AS + CAD group, 8.9% (10 patients). The long-term survival at seven years was 77.2 +/- 5.5% (+/- the standard error of the mean) for the AS group and 57.0 +/- 6.9% for the AS + CAD group (p less than 0.006). Postoperative assessment reveals substantial functional improvement. These early and long-term favorable results provide a much needed reference point when valvuloplasty is being considered. Aortic valve replacement is the treatment of choice in elderly patients with symptomatic AS.

From July 1972 through July 1983, 360 consecutive patients 70 years of age and older (mean age 74.1; range 70-88 years) underwent 362 valve replacement procedures. There were 122 isolated aortic valve replacements (33.7%; Group I); 70 isolated mitral valve replacements (19.3%; Group II); and 170 patients had combined procedures (47.0%; Group III), which included the replacement of at least one valve. Eighteen patients (5.0%) had previous cardiac surgery. Thirty-two patients (8.8%) were operated as emergencies. Three hundred and thirty-two (86.5%) of all valves implanted were porcine heterografts. Pre-operatively, over one-half (53.6%) of the patients were in New York Heart Association Functional Class IV. The overall hospital mortality was 13.8% (50 patients). The aortic valve mortality was 11.5%, the mitral valve mortality was 15.7%, and the combined procedures 14.7%. The follow-up period for hospital survivors (312 patients) extended from 2 weeks to 127.2 months, with a mean of 38.7 months or a total of 1,006 patient-years. The long term survival computed up to six years shows a 65 +/- 3.8% (standard error of the mean) for the entire group. The aortic valve group survival was 71 +/- 5.6%; the mitral valve group 60 +/- 8.2%; and the combined procedures group was 64 +/- 5.3%. Postoperative functional improvement was significant with 71.8% of the survivors in Class I and 19.6% in Class II. Based on these results, advanced age can no longer be considered a deterrant to cardiac surgery. The porcine heterograft appears to be the valve substitute of choice for this age group.(ABSTRACT TRUNCATED AT 250 WORDS)

Bioprosthetic valve durability and the significance of patient age at implantation have received much attention recently. Indications and/or contraindications for implantation of the bioprosthesis in the very young and in the elderly have been reasonably well defined. Patients in the middle years (sixth and seventh decades) present a special problem in the choice of a prosthesis. To better elucidate the failure rate of the Carpentier-Edwards bioprosthesis in middle-aged patients, a comparative study of value failure rates was conducted using the Wilcoxon (Breslow) statistical technique. From September 1978 to December 1986, 502 patients underwent valve replacement with a Carpentier-Edwards bioprosthesis. All patients were operated on by a single surgical team using precisely the same method of valve implantation and myocardial preservation. The overall 30-day mortality was 8.4%. PATIENT SURVIVAL: Follow-up was obtained on all 460 hospital survivors and extends to 109.2 months with a mean of 36.8 months. The cumulative survival is 1,410.6 patient-years. VALVE SURVIVAL: The 481 patients that left the hospital were divided into two subgroups. Group I included patients aged 55 to 69 years; group II, 70 years and older. There were 8 valve failures in group I. The percent of valves free of failure plotted by the actuarial method is 95.4% at 5 years (SEM 1.7, 81 valves at risk) and 95% at 7 years (SEM 1.7, 23 valves at risk). In group II (age 70 and above), there were only two valve failures.(ABSTRACT TRUNCATED AT 250 WORDS)