Long-term follow-up (12 to 35 weeks) after dynamic cardiomyoplasty.

OBJECTIVES: To obtain information on the long-term effects of dynamic cardiomyoplasty on hemodynamics and muscle histology, this surgical method was evaluated in goats. BACKGROUND: Dynamic cardiomyoplasty has been introduced as a new method to treat patients with severe cardiac failure. METHODS: In 24 goats, the left latissimus dorsi muscle was wrapped around the heart. The muscle was then subjected to progressive electrical stimulation. In 16 goats, invasive transesophageal Doppler echocardiographic measurements and histologic evaluation of the latissimus dorsi muscle were performed at > or = 12 weeks after the wrapping. RESULTS: Only two goats showed an increase in aortic and left and right ventricular pressures concomitant with increased aortic flow during latissimus dorsi muscle stimulation both before and after induction of cardiac failure using imipramine. This was accompanied by a preserved latissimus dorsi muscle structure and nearly complete transformation to type I muscle fibers. The remaining 14 goats showed extensive lipomatosis in the latissimus dorsi muscle, with severe intimal hyperplasia and proliferation of smooth muscle cells in the walls of the thoracodorsal artery and its branches. An increase in endoneural and endomysial connective tissue was observed, with some goats showing destroyed nerve branches near the electrodes. These findings differed from those observed after long-term electrical stimulation of goat latissimus dorsi muscle in situ. CONCLUSIONS: Dynamic cardiomyoplasty is of use in the treatment of severe heart failure if the histologic structure of the wrapped latissimus dorsi muscle remains intact. Long-term results in goats suggest that the current approach used in dynamic cardiomyoplasty may lead to deterioration of the wrapped muscle.

An experimental skin sandwich flap on an independent vascular supply for the study of percutaneous absorption.

Further insights into the composite interactive processes of topically applied agents and percutaneous absorption and metabolism by functional skin in vivo have been hampered by the lack of a model system wherein the blood flow to and from the skin is independent but experimentally accessible. Utilizing microsurgical techniques, split-thickness skin grafting with syngeneic skin grafts, and the congenitally athymic (nude) rat, a skin sandwich flap system has been generated that has an independent but accessible vasculature and thus fills this void. We describe the methodology that has been developed to create the flap and present experiments that: demonstrate a lack of significant collateral circulation; quantify the microcirculation of the skin sandwich flap, host side, and graft side at various times during and after the flap has been generated, and note that blood flow to the flap is basically unchanged from host skin; demonstrate the utility of the system in measuring the amount of [14C]benzoic acid that appears in the flap when deposited on the surface in volatile and nonvolatile vehicles as a function of time; and demonstrate the fact that the flap can be reused, and that the total amount of [14C]benzoic acid absorbed across skin does not change in a substantial way as the flap ages.

Considerations for reconstruction after surgery for recurrent cancer.

Refinements in our knowledge of the anatomy and physiology of myocutaneous and fasciocutaneous flaps has served as the impetus to apply these modalities to achieve healed, stable wounds in areas undergoing wide exenterative resection for recurrent rectal carcinoma. The adverse effects on tissue from adjuvant or therapeutic radiation therapy is greatly ameliorated by these healthy nonirradiated tissues, which serve as a conduit for oxygen as well as humeral and cellular factors to enhance the immune response as well as to aid wound healing. Moreover, refinements in surgical technique have allowed the reconstructive surgeon to Radical ablative surgery is possible because of the advances over the past half century in our understanding of cancer, surgical principles, anesthesia, preoperative preparation, and postoperative support. It can be argued that radical ablation without reconstruction promises increased quantity of life, which is only half of the equation. Only quality of existence can make even a few more months of life truly meaningful for a patient. This is the real contribution of reconstruction. It is with the expectation that form and function can be restored toward the premorbid state that the surgical oncologist can confidently recommend radical salvage procedures that provide both quantity and quality of life.

Full-thickness dynamic cardiomyoplasty of the left ventricle with free revascularized latissimus dorsi myografts. An experimental feasibility study.

Dynamic cardiomyoplasty, with use of a free latissimus dorsi myograft revascularized by the internal thoracic artery and vein, was performed in eight dogs subjected to electrical preconditioning for 8 to 12 weeks (group I) and in six unconditioned dogs (group II). The procedure was performed after the resection of the anterior wall of the left ventricle. Cardiac output and left ventricular stroke work were augmented by 23.7% +/- 9.4% and 44.1% +/- 15.9% after graft pacing with 50 Hz burst stimulation at a 1:1 synchronization ratio, while left atrial pressure ranged from 8 to 12 mm Hg. Analysis of the left ventricular function curve showed that graft pacing at rates of 1:1, 2:1, and 3:1 augmented global left ventricular function. Hemodynamic benefit by continuous pacing at a 3:1 ratio was seen for 1.97 +/- 1.90 hours (0.5 to 6.1 hours) in group I until complications unrelated to the graft terminated the study, while it lasted for only 0.19 +/- 0.09 hour in group II. During the stimulation, the ratio of the lactate output to the oxygen consumption of the graft in group I, a possible indicator of metabolic shift, was significantly less than in group II, (0.46 +/- 0.58 and 6.34 +/- 1.73; p < 0.01). We conclude that free grafts of transformed latissimus dorsi muscle can augment global left ventricular performance, with a physiologic preload by oxidative metabolism, and provide a viable option in full-thickness dynamic cardiomyoplasty.

Skeletal muscle-powered ventricle. Effects of size and configuration on ventricular function.

The optimal size and configuration of skeletal muscle-powered ventricles are still undetermined. This study was aimed at comparing three types of skeletal muscle-powered ventricle: (A) a small size (15 ml capacity) double-layered pump, (B) a small size (15 ml capacity) single-layered pump, and (C) a large size (40 to 60 ml capacity) single-layered pump constructed sequentially with the same untrained latissimus dorsi muscle of 12 mongrel dogs. The skeletal muscle-powered ventricle was connected to a mock circulation system, the stroke volumes against 40 to 160 mm Hg of afterload at 5 to 60 mm Hg of preload were measured, and the stroke work was computer analyzed on line. Raising the preload from 5 to 60 mm Hg increased the peak isovolumic developed pressure (A) from 91.3 +/- 11.0 to 215.6 +/- 26.1 mm Hg, (B) from 92.8 +/- 12.0 to 166.3 +/- 19.0 mm Hg, and (C) from 32.3 +/- 5.2 to 121.4 +/- 15.5 mm Hg (p < 0.05, C versus A and B). Similarly, the stroke volume (stroke work) against an afterload of 120 mm Hg increased (A) from 3.8 +/- 0.5 ml (0.22 +/- 0.04 x 10(6) ergs) to 14.5 +/- 1.1 ml (1.05 +/- 0.11 x 10(6) ergs), (B) from 4.5 +/- 0.7 ml (0.30 +/- 0.08 x 10(6) ergs) to 10.7 +/- 0.9 ml (0.63 +/- 0.08 x 10(6) ergs), and (C) from 1.8 +/- 0.5 ml (0.09 +/- 0.04 x 10(6) ergs) to 24.0 +/- 3.6 ml (1.94 +/- 0.41 x 10(6) ergs) (p < 0.05, C versus B at 5 mm Hg of preload; p < 0.05, C versus A and B at preloads > or = 30 mm Hg). At low preloads (5 to 15 mm Hg) both small pumps generated a significantly larger stroke volume (stroke work) than the large pump, whereas at high preloads (> or = 30 mm Hg) the large pump generated a significantly larger stroke volume (stroke work) than the small pumps. It is concluded that under physiologic preload, B (small single-layered pump) performs better than or at least as well as A (small double-layered pump) and C (large single-layered pump), despite being constructed with only one half of the muscle mass used for either A or C.

New configuration of double cardiomyoplasty based on studies of the length-tension properties of the latissimus dorsi muscle.

A new configuration of double cardiomyoplasty was designed according to studies of the length-tension properties of the linear latissimus dorsi muscle. Four dogs had both their right and left latissimus dorsi muscles dissected from the chest wall and attached to a tensiometer to measure force of contraction. The maximum active tension obtained with stimulation of the linear latissimus dorsi muscle was observed when the muscle was at its resting anatomic length and up to 5% above this length. Eight dogs had a double cardiomyoplasty in which the resting anatomic length of both muscles was maintained. Control hemodynamic parameters obtained with the muscles at rest were compared with stimulated muscle protocols. In a normal heart state, stimulation of the double cardiomyoplasty increased the cardiac output 32% (p < 0.05), the stroke volume 39% (p < 0.05), and the left ventricular pressure 42% (p < 0.05). When acute heart failure was induced with high-dose intravenous propranolol (5 mg/kg), stimulation of the double cardiomyoplasty increased the cardiac output 32% (p = 0.01), the stroke volume 32% (p < 0.05), rate of pressure rise 39% (p < 0.01), and myocardial thickening 39% (p < 0.01). The study demonstrated that this configuration of double cardiomyoplasty provides significant hemodynamic assistance in the normal and acutely failing canine heart.

Risk of recoarctation in neonates and infants after repair with patch aortoplasty, subclavian flap, and the combined resection-flap procedure.

Between September 1980 and August 1988, 47 patients younger than 12 months (27 neonates and 20 infants) underwent repair of aortic coarctation. Three surgical techniques were used: patch aortoplasty (group A: 5 neonates and 3 infants, mean age 4.5 months), subclavian flap (group B: 11 neonates and 8 infants, mean age 3.1 months), and the combined resection-flap procedure (group C: 11 neonates and 9 infants, mean age 2.7 months). There were 8 early deaths (3 in group A, 3 in group B, and 2 in group C) and 2 late deaths (both in group B), all of which occurred in patients with complex coexisting anomalies. Follow-up included 37 patients (5 in group A, 14 in group B, and 18 in group C) and ranged from 28 to 108 months (mean 68.0, 60.8, and 51.7 months, respectively). Residual arm-leg pressure gradients greater than 10 mm Hg were detected in 4 of 5 patients in group A, 11 of 14 patients in group B, and 1 of 18 patients in group C (p less than 0.005). There were no reoperations in any infant operated on after 1 month of age in any group (0% risk). However, 4 patients who underwent repair during their neonatal period, with recurrent gradients greater than 20 mm Hg, have required reoperation: in group A, 1 of 5 patients (20% risk) (1 of 2 neonates or 50% risk); in group B, 3 of 14 patients (21% risk) (3 of 9 neonates or 33% risk); and none in group C (0% risk in infants and neonates) (p less than 0.001). In summary, residual gradients and risk of recoarctation are significantly higher when a patch or a subclavian flap had been used for repair. The combined resection-flap procedure (an end-to-end anastomosis enlarged with a subclavian flap) is more effective in avoiding gradients and preventing recoarctation, especially in neonates.

Dynamic cardiomyoplasty for hemodynamic support during acute pulmonary hypertension. An experimental study.

The efficacy of dynamic cardiomyoplasty for hemodynamic support during acute pulmonary hypertension was studied. Five dogs underwent a right latissimus dorsi cardiomyoplasty. Each dog was later studied in a short-term experiment. A graded acute pulmonary hypertension was induced by infusion of glass microspheres into the pulmonary artery. This resulted in decrease in pulmonary artery flow, systemic pressure, and systemic flow. The cardiomyoplasty was then stimulated with a new R wave synchronous rate-responsive pulse-train stimulator (Prometheus system). This pacemaker delivers a pulse train with the duration of stimulation determined as a proportion of the RR interval. At an optimal level of hemodynamic impairment, the dynamic cardiomyoplasty was able to immediately improve pulmonary artery flow 26.4% +/- 5.84% (standard error of the mean) (p less than 0.005, paired t test), mean systemic arterial pressure 11.6% +/- 3.7% (p less than 0.05), and thoracic aortic flow 15.7% +/- 6.3% (p less than 0.05). The degree of improvement in hemodynamic variables could be correlated with the magnitude of hemodynamic impairment present (e.g., r = 0.78; p less than 0.005 for pulmonary blood flow). We conclude that a significant beneficial effect of dynamic cardiomyoplasty on hemodynamics in short-term canine pulmonary hypertension is demonstrated in this study. Thus cardiomyoplasty may be useful in patients with right heart failure associated with increased pulmonary vascular resistance.

Single-stage treatment of sternal wound complications in heart transplant recipients in whom pectoralis major myocutaneous advancement flaps were used.

Treatment of sternal wound complications is controversial, particularly in immunosuppressed heart transplant recipients. Regardless of the severity of infection, we combine immediate, aggressive débridement with bilateral pectoralis major myocutaneous advancement flaps in a single procedure. Compared with management with pectoralis major turnover flaps or distant pedicled muscle flaps, treatment of these sternal wounds with pectoralis major myocutaneous advancement flaps is simpler and quicker and provides better aesthetic results. Furthermore, because pectoralis major myocutaneous flaps are based on the thoracoacromial arteries, whether or not the internal mammary arteries have previously been harvested for coronary grafts is irrelevant. Twenty consecutive heart transplant recipients with sternal wound complications were treated with this technique. No intraoperative or perioperative deaths occurred. The morbidity rate was 30%, with seroma treated by needle aspiration in four patients (20%) being the most common complication. Only one patient had a postoperative wound infection. All patients had excellent functional and aesthetic results.

Protection and revascularization of bronchial anastomoses by the intercostal pedicle flap.

We used an improved method for preparation of the intercostal pedicle flap for encircling bronchial anastomoses, and we studied its vascular supply after the operation. The flap was used in 56 patients undergoing various types of sleeve resection and in three patients undergoing single lung transplantation. The technique is simple, fast, and causes neither extra surgical trauma nor complications. It allows satisfactory isolation and sealing of the bronchial anastomosis. Even if complete anastomotic dehiscence occurs (one case), the flap preserves the continuity of the airway, thus avoiding bronchopleural fistulas or other complications. The postoperative arteriographic study of the intercostal artery supplying the flap (performed in 14 patients) demonstrated the full patency of the vessel in all cases. It also showed that a fine vascular network develops around the anastomosis early in the postoperative period.

Effects of dynamic cardiomyoplasty on indices of left ventricular systolic and diastolic function in a canine model of chronic heart failure.

The effects of cardiomyoplasty were evaluated with multiple-gated equilibrium radionuclide angiocardiography and catheterization in a canine model of chronic heart failure. Doxorubicin was administered to 12 dogs at a dose of 1 mg/kg/wk intravenously for 10 weeks. Left ventricular ejection fraction was reduced from a mean of 53.6% +/- 3.4% to 33.5% +/- 2.3% preoperatively. Two dogs died of presumed arrhythmia during this period. Cardiomyoplasty with the left latissimus dorsi muscle was performed on 10 dogs. The muscle was wrapped around both the left and right ventricles. Five dogs died of infection or arrhythmia after the operation. Postoperatively the muscle remained unstimulated for 2 weeks to allow adhesion to the heart. After this period, the latissimus dorsi muscle was conditioned by a progressive stimulation protocol. After the muscle was conditioned, multiple-gated equilibrium radionuclide angiocardiography studies showed that left ventricular global ejection fraction was 18.4% +/- 7.2% at 0 volts (nonstimulation), 26.2% +/- 3.7% at 5-volt stimulation (p less than 0.05), and 31.0% +/- 5.4% at 10-volt stimulation (p less than 0.05). Regional ejection fractions in low lateral, apical, and low septal regions at 5 volts and 10 volts were higher than those at 0 volts (p less than 0.05). Regional wall motion (percent radial shortening) of the low lateral region was higher than that during nonstimulation (p less than 0.05). Peak emptying rate was 2.07 +/- 0.95 end-diastolic counts per second at 0-volt, 3.10 +/- 0.67 at 5-volt, and 3.34 +/- 0.89 at 10-volt stimulation (p less than 0.05). Peak filling rate was 1.81 +/- 0.52 end-diastolic counts per second at 0-volt, 2.67 +/- 1.18 at 5-volt, and 3.11 +/- 0.65 at 10-volt stimulation (p less than 0.05). Cardiac catheterization data showed a nonsignificant increase in left ventricular rate of pressure rise with increasing voltage (1302 +/- 355 mm Hg/sec at 0 volts, 1450 +/- 413 mm Hg/sec at 5 volts, and 1568 +/- 455 mm Hg/sec at 10 volts). Left ventricular systolic pressures were unchanged. End-diastolic pressures decreased (11.2 +/- 1.48 mm Hg at 0 volts, 10.4 +/- 2.30 mm Hg at 5 volts, and 9.6 +/- 1.52 at 10 volts; p less than 0.05). These data show that cardiomyoplasty can improve indices of systolic and diastolic function in a canine model of chronic heart failure.

Intrathoracic tracheal reconstruction with a collagen-conjugated prosthesis: evaluation of the efficacy of omental wrapping.

Reconstructions of the intrathoracic trachea in 24 dogs were done with the use of 50 mm long collagen-conjugated tracheal prostheses. Omental wrapping was also done in 14 of the dogs (omentopexy group) to evaluate the efficacy of this option in comparison with results in the other 10 dogs (control group). All 24 dogs had uneventful postoperative courses and were killed at 4 weeks or 3, 6, or 12 months after the operation. Better epithelialization and fewer complications, such as mesh exposure and luminal stenosis, were observed in the omentopexy group than in the control group. Angiography and analysis of regenerated blood vessels revealed that vessel ingrowth had started within 4 weeks and that vessel formation reached its maximal point within 6 to 12 months in the omentopexy group. In contrast, revascularization of the subepithelial region in the control group was poor even after 3 months, and vessel formation continued for as long as 12 months. The differences between the two groups were considered to be mainly a result of the speed of blood vessel ingrowth into the regenerated mucosa. We conclude that our prosthesis can be used safely for intrathoracic tracheal reconstruction and that omental wrapping is a useful supplementary method that reduces the occurrence of complications.

Obliteration of empyema tract with deepithelialized unipedicle transverse rectus abdominis myocutaneous flap.

Four patients with chronic empyema after pneumonectomy have undergone successful obliteration of the empyema tract with a deepithelialized transverse rectus abdominis myocutaneous flap. The deepithelialized skin island has provided sufficient bulk for tract obliteration. Rotation of the skin island into the long axis of the rectus muscle has added considerable length to this flap, allowing it to reach the apex of the thoracic cavity. A recurrent loculation developed 4 months after the obliteration procedure in one patient. This was successfully treated with open pleural drainage and a second Clagett procedure. Over a mean follow-up period of 35 months, all four patients are free of further infectious sequelae. Chest roentgenograms have confirmed eradication of the tracts in all four patients.

Treatment of bronchopleural fistula after pneumonectomy by using an omental pedicle.

Dehiscence of the bronchial stump, particularly following pneumonectomy, is one of the most serious complications in lung surgery. Various approaches to this problem have been tried. Two patients with postpneumonectomy bronchopleural fistula were treated successfully by using a pedicle of the greater omentum.

Long-term follow-up comparing subclavian flap angioplasty to resection with modified oblique end-to-end anastomosis.

The definitive surgical procedure for correction of aortic coarctation remains controversial. Therefore, we retrospectively reviewed a total of 56 children under 4 years of age with coarctation repair between 1977 and 1986. Thirty-four had the subclavian flap angioplasty technique and 22 had resection with oblique end-to-end anastomosis. The group was further subdivided to include only the 23 infants less than 3 months of age--eight infants with resection with oblique end-to-end anastomosis (less than or equal to 3ETE) and 15 infants with subclavian flap angioplasty (greater than or equal to 3SFA). The remaining 33 patients older than 3 months of age were divided into 14 patients with resection and oblique end-to-end anastomosis (greater than 3ETE) and 19 patients with the subclavian flap angioplasty technique (greater than 3SFA). The overall mortality was not significantly different between techniques. Postoperative hypertension was significantly more prevalent with end-to-end anastomosis than with the subclavian flap angioplasty technique (p less than 0.01). Seven patients had recurrent coarctation. The 6-year actuarial freedom from recoarctation was 93% +/- 6% in the less than or equal to 3SFA group compared with 53% +/- 20% in the less than or equal to 3ETE group (p less than 0.02), but there was no significant difference in those children operated on at a later age regarding the type of coarctation repair. Therefore, we recommend subclavian flap angioplasty in patients less than 3 months of age. In those older than 3 months either procedure is safe and the risk of recoarctation is similar.

Dynamic cardiomyoplasty at seven years.

Since January 1985, the date of the first dynamic cardiomyoplasty, until April 1992, 52 patients with end-stage heart disease were operated on in our institution. Mean preoperative New York Heart Association functional class was 3.3 and ventricular ejection fraction 16% +/- 3%. Associated procedures in 23 patients comprised ventricular aneurysm resection (10), valve surgery (9), coronary artery bypass (8), and tumor resection (3). Thirty-eight patients had a ventricular reinforcement, 13 a ventricular substitution, and 1 an atrial reinforcement using the left latissimus dorsi muscle. Preassist mortality rate before full latissimus dorsi muscle stimulation was 7 of 13 patients (54%) in the 1985 to 1987 period and 5 of 39 (12%) in the 1988 to 1992 period. The causes of death were heart failure (4), multiorgan failure (4), septicemia (2), ventricular fibrillation (1), and sudden death (1). Multivariate analysis of factors influencing hospital mortality showed that age, cardiac suture technique, associated surgical procedures, biventricular heart failure, and hemodynamic instability plus inotropic drug support were predictors of unfavorable outcome. All patients were followed up for from 2 months to 7 years (mean 21 months). Postassist mortality rate was 8 of 40 (20%). Causes of death included heart failure (5), ventricular fibrillation (1), myocardial infarction (1), and gastric bleeding (1). Preoperative risk factors influencing long-term mortality were permanent New York Heart Association functional class IV, biventricular heart failure, atrial fibrillation, cardiothoracic ratio greater than 60%, and ejection fraction less than 15%. Actuarial survival at 7 years was 70.4% (preassist mortality excluded). Surviving patients were in a mean New York Heart Association functional class of 1.8 (preoperatively 3.3, p < 0.05). The average ejection fractions (rest/stress) were 25%/28% at 1 year, 26%/30% at 2 years, and 23%/28% at 3 years. Average cardiothoracic ratios were 57% +/- 3% at 1 year, 56% +/- 2% at 2 years, and 57% +/- 2.5% at 3 years. Catheterization obtained in 20 patients showed no significant changes at rest in capillary wedge pressure, pulmonary artery pressure, and diastolic left ventricular pressure when compared with preoperative pressures. Average ejection fractions increased from 24% to 30.6%. Maximal oxygen consumption increased from 12.8 +/- 3.5 to 18.6 +/- 4 ml/min per kilogram. The number of rehospitalizations resulting from congestive heart failure was reduced to 0.4 hospitalizations per patient per year (preoperatively 2.4, p < 0.05). In 62% of the patients, pharmacologic therapy was diminished after the operation. Three patients required orthotopic heart transplantation 6 months, 4 years, and 5 years after cardiomyoplasty.

Latissimus dorsi cardiomyoplasty. Perioperative management and postoperative evolution.

Twelve men aged 45 to 69 years, NYHA class 3 or 4 with low isotopic ejection fraction (18 +/- 7 percent), underwent cardiomyoplasty. Eight required cardiopulmonary bypass to treat an associated cardiac lesion. Preoperatively, all patients needed inotropic support with dobutamine and half of them vasodilators, increasing cardiac index by nearly 100 percent. The SvO2 remained over 67 percent during the different stages of the surgical procedure. The mean operating time was 438 +/- 75 min. None of the patients required intra-aortic balloon counterpulsation. Inotropic and vasodilator support was continued in the ICU and appeared especially important during weaning from mechanical ventilation. The average stay in ICU was 6.8 +/- 4.0 days. Three patients died of cardiac failure respectively 8, 11 and 15 months after CMP. One patient underwent transplantation. The eight other surviving patients showed clinical improvement from the third month, but objective criteria for hemodynamic improvement were noted only after one year. Cardiomyoplasty can be an alternative treatment for selected cases of cardiomyopathy.

Pedicled pericardium for repair of right ventricular outflow tract and pulmonary arterial stenoses in tetralogy of Fallot: a six-year experience.

From June 1988 through June 1994, 20 children with symptomatic tetralogy of Fallot, associated with hypoplastic or stenotic pulmonary arteries in 19 cases, were operated on in our institution. Mean age at operation was 49.5 +/- 43 months (ranging from 10 months to 12.5 years). Mean weight was 13.5 +/- 6.5 kg (range 6.2 to 30 kg) and mean body surface area was 0.6 +/- 0.2 m2. Mean preoperative hematocrit value was 47.6% +/- 11.1%, and mean preoperative arterial oxygen saturation ratio was 75.7 +/- 9.5. Six patients (30%) had prior systemic-pulmonary arterial shunts. Pulmonary arterial stenoses were congenital or iatrogenic (or both) and were located principally on the left pulmonary artery (63%), on the pulmonary bifurcation (15%), or separately on the left and right pulmonary arteries (21%). In all children the pulmonary arterial tree was repaired with the patient's own pedicled pericardium. Mean follow-up is 36 +/- 25.2 months (range 1 to 71 months). Hospital mortality rate was 0%. There was one late death, and two children were lost to long-term follow-up. No child required reoperation. Seventeen children returned for evaluation, consisting of physical examination, electrocardiogram, chest roentgenogram, and Doppler echocardiogram, and 11 underwent late catheterization or magnetic resonance imaging of the chest to evaluate the pulmonary arterial repair. All children were in New York Heart Association class I or II. Right ventricular function was normal by echocardiography in 100% with a mean right ventricular systolic pressure of 39.37 +/- 8.4 mm Hg. Mild to moderate pulmonary regurgitation was present in the majority of patients. The results of pulmonary arterial repair were good in 100%. There was no residual stenosis, and we observed uniform enlargement of the repaired pulmonary arteries over a 5-month to 6-year follow-up period. These results are of particular interest inasmuch as other materials previously used for repair of pulmonary arteries do not grow and may even shrink, resulting in residual stenosis.

Oxidized cellulose mesh. II. Using hydroxy-apatite bone grafting material in the treatment of infrabony defects.

Biodegradable oxidized cellulose mesh has been used in conjunction with hydroxyapatite synthetic bone grafting material to enhance retention of the graft material and to improve post-operative healing of infrabony defects treated surgically. Initially a good healing response was achieved, but there was a tendency for the sites to relapse in the 8 wk following surgery. A tendency to relapse in reconstructive dentistry is well known, and the possible factors which limit the amount of regeneration of tissues are considered. Although there appear to be limitations on the regeneration possible in tissues, the factors that determine these limitations are poorly understood at present.

Rectus femoris flap repair of recalcitrant inguinal lymphoceles after heart transplantation.

For access to cardiopulmonary bypass in orthotopic heart transplantation, cannulation of the femoral artery is reserved for situations in which aortic cannulation cannot be performed safely. Femoral cannulation, however, is associated with multiple complications including lymphatic leakage manifested as either lymphocutaneous fistula or lymphocele. Although lymphatic leakage can often be treated conservatively, the length of time required for resolution, the delayed healing, and the risk of secondary infection makes more rapid and sure treatment essential for the immunocompromised and steroid-dependent heart transplant recipient. The purpose of this article is to describe a technique with the use of a rectus femoris muscle flap for operative treatment of inguinal lymphocele in transplant recipients refractory to both standard conservative and operative management and demonstrate its applicability in three cases. The use of a rectus femoris flap provides elimination of dead space, increased vascularity to eliminate infection and enhance wound healing, and, potentially, a conduit for lymphatic drainage.