Aortic dissection: anatomic types and surgical approaches

Our classification system of acute dissection of the aorta is based on the site of the main intimal tear: Type A: on the ascending aorta; type B: on the transverse aortic arch; type C: on the descending aorta. The extension of the dissecting process is classified as "antegrade" or "retrograde". Acute dissection involving the ascending aorta is an absolute surgical urgency. Any delay in referring the patient to a proper surgical institution or to the operating room increases the risk of death. Fifty per cent of patients, indeed, either untreated or medically supported, die within 48 hours after the onset of symptoms. Surgical therapy is mainly aimed at preventing the patient from dying from intrapericardial rupture of the aorta or from acute massive aortic regurgitation. In type A, it is necessary to replace the ascending aorta with a bloodtight Dacron prosthesis after resecting the entry site, if possible. Downstream, joining the two dissected cylinders by two running sutures and the aid of GRF glue, seals the false lumen. Upstream, the reconstruction of the aortic root and the resuspension of the aortic valve, also by means of running sutures and GRF glue, suppress the aortic valve insufficiency in 90% patients. However, in case of pre-existing annulo-aortic ectasia, the ascending aorta must be replaced by a composite tube according to the Bentail technique. The use of GRF glue since the beginning of 1977, has dramatically improved the immediate and long-term results, accounting for a hospital mortality rate of 10%, in patients less than 65 years old. In type B, resecting the entry site requires that the transverse arch be partially or totally replaced. It is, therefore, mandatory to protect the Central Nervous System. In our experience this is best achieved by perfusing the carotid arteries with cold blood (6 degrees C) during circulatory arrest at moderate core hypothermia (28 degrees C). With this technique of "Cerebroplegia", the hospital mortality rate has been lowered to 28%, higher, though, than in patients undergoing isolated replacement of the ascending aorta. In type C, only the dissections demonstrating symptoms of major complications (rupture or deleterious ischemia) require urgent surgical treatment. In the remaining cases, medical treatment, based on permanent and accurate control of the patient's blood pressure, lead to a good long-term survival rate. Close survey at regular intervals, by means of CT scan or MNR is mandatory to detect any aneurysmal evolution, which may require surgery.(ABSTRACT TRUNCATED AT 400 WORDS)

Coronary surgery can be an alternative to heart transplantation in selected patients with end-stage ischemic heart disease

Patients with ischemic heart disease (IHD) low ejection fraction (EF), and congestive heart failure (CHF), are usually referred for orthotopic heart transplantation (OHT). This study reports our experience with coronary artery bypass grafting (CABG) in patients initially referred for OHT, and suggests guidelines to facilitate the choice of procedure (OHT or CABG). Between January 1990 and December 1991, 32 patients with IHD, proposed for OHT, underwent CABG 31/32 patients were male, the mean age was 58 +/- 12 years (40 to 70). Congestive heart failure was present in all patients and was the main symptom. The mean EF was 23 (14 to 31%), mean cardiac index (CI) 2.4 l/min per m2 (1.6 to 3.1 l/min per m2), mean pulmonary artery mean pressure (MPAP) 26 (20 to 37 mmHg) and mean pulmonary wedge pressure 16 (12 to 22 mmHg). Every patients underwent a myocardial viability study by thallium scintigraphy (n = 32) and/or by positron emission tomography (n = 10). The perioperative mortality was 9.3% (3/32). All long-term survivors (n = 27) are in NYHA Class II with a complete follow-up (mean 18 +/- 6 months). Ejection fraction control either by angiography (n = 15) or by single photon emission computed tomography (n = 12) showed an increase of up to 38% (22%-46%). Three determinant factors influenced the choice of CABG. 1) CI > 21/min per m2, 2) MPAP < 35 mmHg. 3) Detection of myocardial viability.(ABSTRACT TRUNCATED AT 250 WORDS)

Specification and implementation of a digital Hopfield-type associative memory with on-chip training

The definition of the requirements for the design of a neural network associative memory, with on-chip training, in standard digital CMOS technology is addressed. Various learning rules that can be integrated in silicon and the associative memory properties of the resulting networks are investigated. The relationships between the architecture of the circuit and the learning rule are studied in order to minimize the extra circuitry required for the implementation of training. A 64-neuron associative memory with on-chip training has been manufactured, and its future extensions are outlined. Beyond the application to the specific circuit described, the general methodology for determining the accuracy requirements can be applied to other circuits and to other autoassociative memory architectures.

Total orthotopic heart transplantation: an alternative to the standard technique

An alternative technique of total orthotopic heart transplantation is described. Complete excision of the recipient's atria is undertaken. Ventricular as well as atrial orthotopic transplantation is performed.

Valve extension with glutaraldehyde-preserved autologous pericardium. Results in mitral valve repair

Preliminary experimental studies in our laboratory have shown that autologous pericardium treated with glutaraldehyde prevents late deterioration and calcification of the tissue. For this reason, glutaraldehyde-treated autologous pericardium has been used in a series of 64 patients who underwent operations for leaflet extension of the mitral valve between 1980 and 1989. Ages ranged from 2.5 to 60 years (mean 19 +/- 15). The causes of mitral valve insufficiency were rheumatic fever (69%), bacterial endocarditis (17%), congenital (8%), endomyocardial fibrosis (4.5%), and trauma (1.5%). The autologous tissue was fixed in a 0.62% glutaraldehyde solution for 15 minutes and rinsed in saline for an additional 15 minutes. Patching techniques varied depending on the site and the extent of the lesion. Associated mitral valve repair techniques (Carpentier's techniques) were mandatory in all patients. The period of follow-up extended from 6 months to 9 years (mean 3.1 +/- 2.5 years). There were no operative deaths in this series, and there was one late death (2%). In the six patients (12%) who underwent reoperation, there has been no case of calcification of the pericardial patch. Postoperative mitral valve function was assessed by bidimensional color Doppler echocardiographic techniques. Mitral valve insufficiency was trivial or absent in 80% of the patients. This experience permits us to conclude that leaflet extension is a simple and safe technique in valve reconstruction, allowing repair of mitral valves that otherwise would need to be replaced. It permits use of an adult-size prosthetic ring in children. Glutaraldehyde-treated autologous pericardium is the material of choice for this type of repair.

Intra-operative transoesophageal echocardiography after mitral repair--specific conditions and pitfalls

The aim of this study was to determine the specific conditions and pitfalls of immediate assessment of the reconstructed mitral valve in open chest patients. Sixty-two patients had a mitral reconstruction controlled by transoesophageal echocardiography and colour Doppler (TEE) from the moment when cardiac activity restarted under extracorporeal circulation (ECC), until complete rewarming of the patient and weaning off the by-pass. The following conditions altered the imaging quality: (1) dried probe in the oesophagus or air bubbles, (2) non-aspirated air from the stomach, (3) small or empty left atria, (4) invagination of the left atrial appendage, (5) trapped air in the posterior pericardium, (6) spontaneous contrast. Transient mitral regurgitation disappeared completely in 12/62 patients: five had temporary impairment of LV function, two had reduced filling of the heart, one had left ventricular outflow obstruction exacerbated by isoproterenol and nitroglycerin, two had ventricular ectopic rhythms, and two epicardial pacing. Only one of the 62 patients had persistent significant MR which required a second run of ECC. We conclude that mitral regurgitation after mitral valve repair is closely related to the quality of LV function. A decision to reoperate should eliminate the possible pitfalls, and take into consideration the specific conditions of immediate postoperative cardiac function.

Cold cerebroplegia. A new technique of cerebral protection during operations on the transverse aortic arch

Profound hypothermia associated with circulatory arrest is the commonest method of cerebral protection during operations on the aortic arch. This technique allows a limited time to perform the aortic repair, however. It also necessitates prolonged cardiopulmonary bypass to rewarm the patient. This may be the cause of coagulation disorders or infection. Selective perfusion of the carotid arteries can also be used. When the perfusion is derived from the main arterial line, however, the repair of the aorta requires that the vessel be crossclamped, and cannot be performed in an "open, bloodless" manner. To avoid the disadvantages of both techniques, we have developed a new technique of cerebral protection. After a regular cardiopulmonary bypass has been established, the carotid arteries are cannulated and perfused with blood cooled at 6 degrees to 12 degrees C, through a separate heat exchanger, while the core temperature is maintained at moderate hypothermia (25 degrees to 28 degrees C, rectal). To perform the "open" distal repair, the cardiopulmonary bypass is discontinued while the carotid perfusion is maintained (250 to 350 ml/min). When the distal repair is completed, cardiopulmonary bypass is resumed and the carotid perfusion is discontinued. Between 1984 and June 1989, 54 patients (mean age 55 years) were operated on with this method (45 elective operations, 9 emergency procedures). Mean duration of cardiopulmonary bypass was 121 minutes (65 to 248), and mean duration of circulatory arrest was 22 minutes (10 to 51). The electroencephalogram, routinely recorded, showed return of the cerebral activity after a mean time of 12 minutes and normal activity after a mean time of 66 minutes. There was no intraoperative death. Hospital mortality rate was 13% (7/54). One death was related to neurologic disorders. All patients but one awakened normally within 8 hours after operation. Two patients (4.3%) experienced a transient neurologic episode (lateral hemianopia) 9 and 11 days postoperatively. There was no hemorrhagic complication (24-hour average blood loss: 840 +/- 540 ml). In our experience the technique of "cold cerebroplegia" has been demonstrated to provide excellent cerebral protection. It requires no prolonged cardiopulmonary bypass and does not limit the time necessary to perform the aortic repair. It may be considered as a safe alternative to profound hypothermia associated with circulatory arrest.

Interaction between roxithromycin and cyclosporin in heart transplant patients

Cyclosporin is an immunosuppressive agent commonly used in transplant patients. It is actively metabolised by the cytochrome P450 system and interactions with drugs metabolised by the same system are predictable. This is particularly relevant since cyclosporin has a low therapeutic index and its renal toxicity is concentration-related. Roxithromycin, a new, well-tolerated macrolide with a weak interactive profile, uses the same isoenzyme of the P450 system as cyclosporin. To evaluate its interaction potential in clinical practice, 8 heart transplant recipients treated with cyclosporin for at least 1 month received roxithromycin for 11 days (150 mg twice daily). Bi-weekly controls of plasma cyclosporin concentrations and creatinine levels were carried out before, during and after roxithromycin treatment. A slight nonsignificant rise in cyclosporin concentrations was observed, but creatinine levels remained stable during roxithromycin treatment. Values of cyclosporin concentrations diminished after withdrawal of roxithromycin. Cyclosporin dosage adjustment was not necessary. There was a minor pharmacokinetic interaction, which can be considered safe for the usual therapeutic dosage of roxithromycin used.

The native mitochondrial F1-inhibitor protein complex carries out uni- and multisite ATP hydrolysis

The rate of ATP hydrolysis under multi- and unisite conditions was determined in the native F1-inhibitor protein complex of bovine heart mitochondria (Adolfsen, R., MacClung, J.A., and Moudrianakis, E.N. (1975) Biochemistry 14, 1727-1735). Aurovertin was used to distinguish between hydrolytic activity catalyzed by the F1-ATPase or the F1-inhibitor protein (F1.I) complex. We found that incubation of aurovertin with the F1.I complex, prior to the addition of substrate, results in a stimulation of the hydrolytic activity from 1 to 8-10 mumol min-1 mg-1. The addition of aurovertin to a F1.I complex simultaneously with ATP results in a 30% inhibition with respect to the untreated F1.I. In contrast, if the F1.I complex is activated up to a hydrolytic activity of 80 mumol min-1 mg-1, aurovertin inhibits the enzyme in a manner similar to that described for F1-ATPase devoid of the inhibitor protein. The native F1.I complex catalyzes the hydrolysis of ATP under conditions for single catalytic site, liberating 0.16-0.18 mol of Pi/mol of enzyme. Preincubation with aurovertin before the addition of substrate had no effect under these conditions. On the other hand, if the F1.I ATPase was allowed to hydrolyze ATP at a single catalytic site, catalysis was inhibited by 98% by aurovertin. In F1-ATPase, the hydrolysis of [gamma-32P]ATP bound to the first catalytic site is promoted by the addition of excess ATP, in the presence or absence of aurovertin. Under conditions for single site catalysis, hydrolysis of [gamma-32P]ATP in the F1.I complex was not promoted by excess ATP. We conclude that the endogenous inhibitor protein regulates catalysis by promoting the entrapment of adenine nucleotides at the high affinity catalytic site, thus hindering cooperative ATP hydrolysis.

[Enoximone in the hemodynamic evaluation before heart transplantation. A new test]

A better selection of patients referred for cardiac transplantation should lead to better surgical results. The presence of severe and irreversible pulmonary hypertension is an important factor and its pretransplantation assessment requires pharmacodynamic testing with vasodilators and dobutamine. The aim of this study of 10 patients in congestive cardiac failure referred for cardiac transplantation was to evaluate enoximone in this indication by comparing it with sodium nitroprussiate (3 micrograms/kg/mn). Intravenous enoximone (total dose of 1.5 mg/kg) increased the cardiac index (+49%; p less than 0.01), slightly reduced the mean systemic blood pressure (-8%; p less than 0.05) whilst inducing a greater reduction in systemic arterial resistances (-36%; p less than 0.01); the fall in mean blood pressure was less than with sodium nitroprussiate (-23%; p less than 0.01). Myocardial oxygen consumption (rate-pressure product) did not increase in contrast to the effect of dobutamine (+21%; p less than 0.01). There was a significant reduction in pulmonary arteriolar resistances (p less than 0.01) with all three drugs but the interpretation of this response and its prognostic significance in patients with a low cardiac output and persistent pulmonary hypertension are discutable even when pulmonary arteriolar resistances are less than 6 Wood units. The value of using an inotropic agent such as Dobutamine or Enoximone is to unmask fixed pulmonary hypertension which may be missed in patients with low cardiac output even with vasodilator drugs, and also to mimic the haemodynamic result of transplantation. In this indication Enoximone may be used like Dobutamine but with the advantage of not increasing myocardial oxygen consumption and being probably less arrhythmogenic.

Kinking of the pulmonary artery: a treatable cause of acute right ventricular failure after heart transplantation

Kinking of the pulmonary artery has been reported only once as a cause of acute right ventricular failure after heart transplantation. We report two cases of kinking of the pulmonary artery encountered during a 3-year period (120 heart transplants). An analysis of the causes of this complication and a diagnostic and therapeutic approach are presented.

[Effect of hypertension on the renal and cardiac functions of patients with heart transplantation]

To evaluate the incidence, risk factors and effects of systemic hypertension on renal function and left ventricular hypertrophy after heart transplantation, 85 recipients under triple drug low low dosage immunosuppressive therapy were studied. After a mean follow-up of 12.5 +/- 8.7 months, high incidence of hypertension was observed in 67% of the patient and 71% had developed de novo hypertension. None of the pre-transplant nor post-transplant cardiovascular risk factors were significantly associated with post-transplant hypertension. Faster deterioration of renal function, as assessed by intraindividual variations of serum creatinine values, was demonstrated in hypertensive patients and appeared as an early indicator of cyclosporine nephrotoxicity in patients at risk for hypertension. [table: see text] Serial (early, intermediate, late) echocardiographic evaluations demonstrated early increase in left ventricular mass and in fractional shortening in both hypertensive and normotensive heart transplant recipients with sustained enhanced contractility in hypertensive patients. [table: see text] Further studies will help to determine the exact relationship between cyclosporine dosages and hypertension, and their respective roles in the development of renal insufficiency and left ventricular hypertrophy after heart transplantation.

Valve repair with Carpentier techniques. The second decade

Among 206 consecutive patients having undergone mitral valve repair with a prosthetic ring between 1972 and 1979 in our institution, the 195 patients (94.5%) who survived the operation were studied to assess the long-term function of this method of repair. Patients' ages ranged from 18 to 79 years (mean age 48.7 years). Mitral valve insufficiency was due to degenerative disease in 113 patients (58%), rheumatic disease in 74 (38%), ischemia and other causes in eight patients (4%). A total of 188 patients (9.7%) were in New York Heart Association class III or IV preoperatively and 94 (48%) had atrial fibrillation. The patients were divided into three functional groups: type I (normal leaflet motion), 35 patients (18%); type II (leaflet prolapse), 147 patients (75%); and type III (restricted leaflet motion), 13 patients (7%). The techniques included prosthetic ring annuloplasty (185 patients), leaflet resection (158 patients), chordal shortening (89 patients), leaflet mobilization (10 patients) and papillary muscle reimplantation (2 patients). Long-term follow-up was available in 189 patients (96.8%), for a rate of 2316 patients per year. The 15-year actuarial and valve-related survival rates were 72.4% and 82.8%, respectively. At 15 years, 93.9% of the patients were free from thromboembolism, 96.6% free from endocarditis, 95.6% free from anticoagulant-related hemorrhage, and 87.38% free from reoperation. Actuarial rate of freedom from reoperation was higher in the group with degenerative disease (92.7%) than in the group with rheumatic disease (76.12%). Among the 157 survivors, 117 (74%) were in New York Heart Association class I and class II and 105 (66%) were in sinus rhythm. Doppler echocardiographic studies showed normal ventricular contractility in 134 patients (84.5%), absence of mitral regurgitation in 112 (74%), trivial regurgitation in 27 (17%), and significant regurgitation in 4 patients (2.5%).

Heart transplantation for corrected transposition of the great vessels

Orthotopic heart transplantation was performed in a patient with corrected transposition of the great vessels. Technical modifications were necessary to be able to transplant a normal heart in this patient. The surgical technique is illustrated.

Valve repair in acute endocarditis

Forty patients were operated on in the early phase of active endocarditis between 1980 and 1988. Indications for operation were heart failure (30 patients), severe valvular regurgitation (4), uncontrolled sepsis (2), septic emboli (3), and other (1 patient). Time between onset of endocarditis symptoms and operation ranged from 12 to 45 days (mean, 30 days). The aortic valve was involved in 3 patients; the mitral valve, in 28; both valves, in 7; and the tricuspid valve, in 2. There was no previous underlying valve pathology in 40%. Lesions found were cusp perforation (17 patients), annular abscess (4), vegetation (13), and chordal rupture (22). Positive blood cultures were found in 30 patients (75%). Bacterial findings were Streptococcus in 12 patients (30%), Staphylococcus in 15 (37.5%), gram-negative in 3 (7.5%), and unknown in 10 (25%). Criteria to perform valve repair were adequate antibiotic therapy for at least 1 week and large excision of all macroscopically involved tissues. In all cases, Carpentier's reconstructive techniques were used. Perioperative mortality was 2.5% (1 patient). Reoperation was necessary in 1 patient. Late mortality was 2.5% (1 patient). Repair was assessed either by angiography or by Doppler echocardiography before hospital discharge: 32 patients showed no regurgitation, whereas 7 had mild regurgitation (3 aortic, 4 mitral). Mean follow-up of 30 months was achieved in all survivors. There was no recurrence of endocarditis and no reoperation for valvular insufficiency. We conclude that valve repair in acute endocarditis is possible and effective in most instances.

Cardiac assistance by atrial or ventricular cardiomyoplasty

Dynamic cardiomyoplasty was conceived to enhance cardiac performance by assisting myocardial contraction. Technically, this procedure consists of placing a pedicled latissimus dorsi muscle flap around the heart and subsequent muscle electrostimulation in synchrony with ventricular systole. Three types of dynamic cardiomyoplasty can be considered. (1) Atrial or ventricular reinforcement is accomplished by wrapping the latissimus dorsi muscle flap around the heart to support hypokinetic or akinetic areas secondary to congenital or acquired diseases. The atrial reinforcement may be performed to improve atrial output after Fontan-type procedures. (2) Ventricular substitution is performed to replace a portion of the ventricular wall. Autologous pericardium is used to create a neoendocardium and facilitate hemostatic closure of the ventricle. The pedicled latissimus dorsi is then secured to replace the resected myocardium. (3) The two previous techniques of ventricular substitution and reinforcement are combined. This reconstructive procedure, which normalizes the ventricular geometrical shape, is particularly useful after extended cardiac resections, such as is done in treatment of large ventricular aneurysms, cardiac tumors, or echinococcal cyst formations. At present, improvement in ventricular function has been obtained in 12 patients at our institution. Preoperative severe cardiac dysfunction was present in all of these patients (New York Heart Association functional class III or IV). Postoperative echocardiography, multigated acquisition scan, and hemodynamic studies demonstrate an improvement in ventricular function and no impairment of ventricular compliance by the muscle flap. After a mean follow-up period of 18 months, all patients are in functional class I or II. We believe that dynamic cardiomyoplasty prolongs and improves the quality of life of patients suffering from severe chronic and irreversible myocardial dysfunction by improving ventricular contraction and limiting cardiac dilatation.

Effect of systemic hypertension on renal function and left ventricular hypertrophy in heart transplant recipients

To evaluate the incidence, risk factors and effects of systemic hypertension on renal function and left ventricular hypertrophy after cardiac transplantation, 85 transplant recipients on triple drug, low dosage, immunosuppressive therapy were studied. After a mean follow-up period of 12.5 +/- 8.7 months, a high incidence of hypertension was observed in 57 (67%) of the patients, and 42 (71%) of the 57 had developed new hypertension. None of the pretransplant and posttransplant cardiovascular risk factors were significantly associated with posttransplant hypertension. Faster deterioration of renal function, as assessed by intraindividual variations of serum creatinine values, was demonstrated in hypertensive patients and appeared as an early indicator of cyclosporine nephrotoxicity in patients at risk for hypertension. Serial echocardiographic evaluations demonstrated an early increase in left ventricular mass and fractional shortening in both hypertensive and normotensive heart transplant recipients. Fractional shortening further diminished significantly in normotensive patients but remained elevated in hypertensive patients, demonstrating sustained enhanced contractility in this group. Further studies will help to determine the exact relation between cyclosporine dosages and hypertension and their respective roles in the development of renal insufficiency and left ventricular hypertrophy after cardiac transplantation.

[Mid-term echographic evaluation of systolic function after cardiac transplantation]

The aim of this study was to analyse variations in left ventricular mass (LVM) and systolic function after cardiac transplantation and to evaluate eventual changes induced by the onset of systemic hypertension (HT). In a retrospective study, we selected the echocardiographic examinations of 42 patients performed at the end of the first month (M1), sixth month (M6), and at the end of follow-up (ME) after cardiac transplantation. The patients were divided into two groups (NT: n = 14; HT: n = 28) depending on whether hypertension occurred during follow-up. The average duration of follow-up was 12.8 +/- 6 months; this was comparable in the two groups. Mild left ventricular hypertrophy was observed from the first postoperative examination: M1 = 193 +/- 50 g; M6 = 199 +/- 62 g; ME = 197 +/- 45 g (NS). The hypertrophy was constant with time and related to wall thickening; it was associated with an increased left ventricular fractional shortening (FS) which decreased with time especially in the NT group (average/42 patients: M1 = 0.38 +/- 0.09; M6 = 0.34 +/- 0.08; ME = 0.35 +/- 0.05; p = 0.03 between M1 and ME). The influence of hypertension on the development of LVH and wall thickening was negligible. The role of transplant rejection should be considered: the repetition of episodes of rejection was less marked than the histological severity of rejection. The role of other factors (ischaemia, persistent haemodynamic abnormalities) is a matter of discussion.

Serum lipid abnormalities in heart transplant recipients: predominance of HDL2-like particles in the HDL pattern

Accelerated coronary atherosclerosis is a major risk limiting long-term survival after heart transplantation and is commonly associated with dyslipoproteinemia even in subjects who were not dyslipoproteinemic before intervention. The purpose of this study was to analyse the abnormalities in the lipid profiles of 2 different groups of heart-transplanted males: 18 subjects with underlying ischemic heart disease (IHD) and 19 subjects with non-obstructive cardiomyopathy of unknown aetiology (CM). Both groups were compared to 33 healthy males. All patients were under immunosuppressive therapy including prednisone, cyclosporin A and azathioprine. A moderate hyperlipidemia was found in all transplant recipients, associated with high HDL-cholesterol concentrations in the CM group (1.80 +/- 0.37 vs. 1.29 +/- 0.23 mmol/l) and normal HDL-cholesterol levels in the IHD group (1.40 +/- 0.23 mmol/l). HDL subfractionation showed a marked increase in HDL2-cholesterol (CM: 1.12 +/- 0.32; IHD: 0.69 +/- 0.28; control: 0.40 +/- 0.17 mmol/l) while HDL3-cholesterol was significantly lower than in the control group. Analysis of HDL particle sizes showed in all transplant subjects an increase of an intermediate size particle HDL2a (diameter 9.0 +/- 0.10 nm) which is a minor form in control subjects. In the CM group, both the common HDL2b (10.2 +/- 0.13 nm) and HDL2a were abundant in 13 of 17 patients. The pattern was more heterogeneous in the IHD group but witnessed to a high frequency of HDL2a particles either alone (5/14) or associated with larger HDL2b (4/14) or with small HDL3 (4/14).(ABSTRACT TRUNCATED AT 250 WORDS)