TUESDAY MORNING, APRIL 27, 1993

7:30 a.m. FORUM SESSION I - Grand Ballroom

CARDIAC SURGERY

Moderators: Aldo R. Castaneda, M.D.

Bruce A. Reitz, M.D.

F1. Six-Year Experience With e-PTFE Chordal Replacement in Floppy Mitral Valve

CLAUDIO ZUSSA, M.D.*, ELVIO POLESEL, M.D.*,

UBERTO DA COL, M.D.* and CARLO VALFRE, M.D.*

Treviso, Italy

Sponsored by: Robert W.M. Prater, M.D., Bronx, New York

Conventional reparative techniques yield better functional results when compared with valve replacement in mitral position, but they are not always completely satisfactory or not adequate at all in cases of diffuse chordal degeneration or rupture. After two years of animal experiments, in 1986 we started the clinical use of 5-0 expanded polytetrafluoroethylene (e-PTFE) sutures (W.L. Gore & Assoc., Flagstaff, AZ) for chordal replacement in selected cases of mitral repair (86 patients so far). A floppy mitral valve with severe insufficiency was present in 61 cases (39 males, 22 females) with a mean age of 52.6 years (range 15-70). The insertion of 2 to 14 e-PTFE chordae was associated with other procedures which alone were not adequate to achieve a satisfactory result: suture annuloplasty reinforced with autologous pericardium in 54 cases, Carpentier ring in 3, posterior leaflet quadrangular resection in 46, chordal shortening in one. Intraoperative transesophageal echocardiography (TEE) was utilized before cardiopulmonary bypass to evaluate the functional valve anatomy, and after the procedure to assess the result (two valves were replaced during the same operation). All patients were discharged from the hospital (in 56 cases with short term -3 months- oral an-ticoagulation) and followed every six months with TEE. At the last follow-up (2-72 months, mean 19.3) the mitral gradient ranged from 2 to 4 mmHg, the valve area from 2.3 to 3.5 sq. cm, mitral insufficiency was absent in 46 cases, trivial in 10, mild in 2, and severe in the last (at the 18-month follow-up). At redo-operation this patient showed rupture of first order chordae shortened at operation, while artificial chordae were still in place, partially covered by host tissue. No deaths or other valve-related events have been reported so far. This innovative procedure has allowed us to expand the indications for mitral valve repair to most of the cases in which, because of diffuse degenerative pathology, the usual techniques are not adequate.

*By Invitation

F2. Prevention of Experimental Postoperative Pericardia! Adhesions Using a Hyaluronic Acid Coating Solution

JOHN D. MITCHELL, M.D.*, RAYMOND LEE, M.D.*,

KAZUO NEYA, M.D.*, GEORGE T. HODAKOWSKI, M.D.*,

WOLFGANG HARRINGER, M.D.* AND

GUS J. VLAHAKES, M.D.

Boston, Massachusetts

Postoperative pericardial adhesions complicate reoperative cardiac procedures. Prior work on their prevention has focused primarily on pericardial patch substitutes, but difficulties with patch migration and infection, epicar-dial scarring, and late calcification have limited their general use. Recently, topical application of solutions containing hyaluronic acid (HA) have been shown to reduce adhesions following abdominal, pelvic and tendon surgery. The mechanism by which HA solutions prevent adhesion formation is unknown, but is thought to be due to a cytoprotective effect of mesothelial surfaces, thus limiting intraoperative injury. We hypothesized that the intra-pericardial use of a HA solution might reduce the severity of pericardial adhesion formation and epicardial reaction frequently seen at cardiac reoperations.

To test this hypothesis, eighteen mongrel dogs underwent median ster-notomy and pericardiotomy followed by a standardized two hour protocol of forced warm air dessication and abrasion of the pericardial and epicardial surfaces with a dry gauze sponge. Group 1 (n = 6) served as untreated controls with subsequent pericardial and sternotomy closure. Group 2 (n = 6) received topical administration of 0.4% HA in phosphate-buffered saline at the time of pericardiotomy, at twenty minute intervals during the dessica-tion/abrasion protocol, and at pericardial closure. Group 3 (n = 6) served as a vehicle control, receiving phosphate-buffered saline as a topical agent in a fashion identical to Group 2. All animals underwent resternotomy eight weeks after the initial surgery; the intrapericardial adhesions were graded on a 0 to 4 severity scale at seven different areas covering the ventricular, atrial, and great vessel surfaces, producing a mean adhesion score for each animal and for each experimental group. In both the untreated control (Group 1, mean score 3.2 ± 0.4) and vehicle control (Group 3, mean score 3.3 ± 0.2) animals, dense adhesions were encountered with frequent obliteration of anatomical dissection planes. In contrast, animals treated with the HA solution (Group 2, mean score 0.8 ± 0.3) characteristically had no adhesions or filmy, transparent adhesions graded significantly less severe than either the untreated control (Group 2 vs. Group 1, p<0.001, t-test) or vehicle control (Group 2 vs. Group 3, p<0.001, t-test) animals.

HA solutions are efficacious in the prevention of postoperative pericardial adhesions in this model. Further studies investigating the mechanism by which these solutions prevent adhesions, their optimal dose and method of application, and documentation of their safe use in the cardiopulmonary bypass environment are warranted.

*By invitation

F3. Oxygen Radical Mediated Vascular Injury Selectively Inhibits Receptor-Dependent Release of Nitric Oxide from the Canine Coronary Artery Endothelium

JOHN F. SECCOMBE, M.D.*, PAUL J. PEARSON, M.D., Ph.D.*

and HARTZELL V. SCHAFF, M.D.

Rochester, Minnesota and Seattle, Washington

Reperfusion following global cardiac ischemia may injure coronary artery endothelium and lead to vasospasm and thrombosis. Superoxide anion (O²*) has been implicated as a mediator of this process, but the precise mechanism of injury is unknown. We hypothesized that (O²*) impairs coronary en-dothelial production of nitric oxide, a potent endogenous vasodilator and inhibitor of platelet adhesion. To test this theory, we developed an in vitro model of reperfusion injury in which segments of epicardial canine coronary artery were suspended in organ chambers (physiologic salt solution, 37 °C, 95% O₂, 5% CO₂) and exposed to (O²*) (generated by adding xanthine [10-⁴M] and xanthine oxidase [100/µU/ml] to the bathing solution for 70 min). Following (O²*) exposure, epicardial coronary artery smooth muscle exhibited normal contraction to potassium ions (20mM) and prostaglandin F₂a (4x10-⁶); also, the rings relaxed normally upon exposure to isoproterenol and sodium nitroprusside (10^-0 10^-4M) (n = 6). In contrast, endothelium-dependent vasodilation to receptor-dependent agonists acetylcholine and adenosine diphosphate (10^-9 10^-4M) was impaired as compared to control vessels not exposed to (O²*) (n = 9, P<0.05). Importantly, receptor-independent, endothelium-dependent relaxation to the calcium ionophore A23187 was normal (n = 6). Further, endothelium-dependent vasodilation to receptor-dependent agonist bradykinin (non-nitric oxide pathway), was normal following (O²*) exposure. This is the first study to demonstrate that (O²*) selectively impairs receptor-dependent nitric oxide production by the coronary endothelium. Diminished nitric oxide production is a likely mechanism of vasospasm and thrombosis following reperfusion of the ischemic heart.

*By Invitation

F4. Inhomogeneous and Complimentary Delivery of Antegrade and Retrograde Cardioplegia in the Absence of Coronary Artery Obstruction

GABRIEL S. ALDEA, M.D.*, DIHOI], M.D.*,

JAMES D. FONGER, M.D. * and

RICHARD J. SHEMIN, M.D.

Boston, Massachusetts

Optimal myocardial protection relies on adequate delivery of cardioplegia to all areas of the heart. Cardioplegia delivery may have an inhomogeneous distribution throughout the myocardium, placing some areas at risk for ischemic injury, even in the absence of coronary artery obstruction. This study compares flow and distribution of antegrade (AC) and retrograde (RC) cardioplegia delivery in hearts with unobstructed coronary vessels. Cardioplegia delivery (flow) to individual myocardial regions weighing on average 1 gram was accurately measured with radioactive microspheres. In-homogeneity of distribution to these areas was expressed as the coefficient of variation (CV = SD/meanx100).

Six pigs were placed on cardiopulmonary bypass and underwent warm blood cardioplegic arrest. AC and RC were delivered at 150 ml/min and flow to 1152 individual myocardial regions was determined twice for each route of delivery using four different radiolabeled microspheres. Patterns of flow to individual myocardial regions were matched and analyzed by linear regression. Using each region as its own control an R² value was determined to assess reproducibility of flow patterns. Reproducibility and comparison of AC an RC flows in a typical animal are represented in the figures below:

Data is reported as mean ± SD. Statistical comparisons were made by paired

t-test analyses.

	AC	RC
Flow (ml/gm/min)	1.37 ± 0.31	0.39 ± 0.09*
CV	48% ± 17%	106% ± 16%*

*p<0.001 vs. AC-rc

	AC vs. AC	RC vs. RC	AC vs. RC
R2	0.88 ± 0.12**	0.84 ± 0.10**	0.03 ± 0.04

**p<0.001 vs. AC-RC

In unobstructed coronary vessels AC delivered greater flow to each gram of myocardium than RC. Flow delivery to individual myocardial regions was significantly inhomogeneous for both AC and RC, but much more so for RC. Patterns of flow with AC and RC were different and therefore complementary. These findings support the routine combined use of AC and RC to enhance cardioplegia delivery to all regions of the heart and minimize the potential risk of post ischemic myocardial dysfunction.

*By Invitation

F5. Controlled Reoxygenation in Hypoxemic Immature Hearts: A Cardioprotective Strategy that Avoids Reoxygenation Injury and Maintains the Benefits of Cardioplegia

KIYOZO MORITA, M.D.*, KAIA. IHNKEN, M.D.*,

GERALD D. BUCKBERG, M.D., HELEN H. YOUNG, Ph.D.*

and MICHAEL P. SHERMAN, M.D. *

Los Angeles, California

We test the hypothesis that abrupt uncontrolled reoxygenation of cyanotic immature hearts when starting cardiopulmonary bypass (CPB) produces a reoxygenation injury that; a) nullifies the Cardioprotective effects of blood cardioplegia, and b) is avoidable by controlling pO₂ during induction of CPB and blood cardioplegia; controlled reoxygenation (reO₂).

METHODS: Eighteen immature piglets (< 3 weeks old) underwent 30 minutes of aortic clamping with hypocalcemic, blood cardioplegia. Six piglets remained normoxemic (control). Twelve piglets were made hypoxic (pO₂ 20-30mmHg) for up to 2 hours by decreasing ventilator FIO₂ to 6-7% before CPB. Six of these piglets underwent 5 minutes of abrupt uncontrolled reoxygenation by instituting CPB at pO₂ 400 mmHg, before receiving pO₂ 400 mmHg blood cardioplegia. Six others underwent controlled reoxygenation by starting CPB at ambient pO₂ (20-30mmHg), followed 5 minutes later by blood cardioplegia at pO₂150 mmHg. Post CPB myocardial function was evaluated from endsystolic elastance (Ees, conductance catheter), oxidant damage was assessed by measuring myocardial conjugated diene production (lipid peroxidation) and antioxidant reserve capacity determined by measuring malondialdehyde (MDA) produced from myocardium incubated in the oxidant, t-butyl hydroperoxide (t-BHP).

RESULTS: Blood cardioplegia preserved myocardial function, and produced no oxidant damage in normoxemic piglets. Uncontrolled reoxygenation offset these Cardioprotective effects and severe functional depression, lipid peroxidation, and reduced antioxidant reserve capacity occurred despite blood cardioplegia. Conversely, controlled reoxygenation allowed near complete recovery of function and avoided oxidant damage with the same cardioplegia protocol.

Group	% Ees (recovery)	Conjugated Dienes (A233/min/100g)	Antioxidant Reserve (MDA @ 4.0mM t-BHP)
No Hypoxia	102 ± 7	3 ± 3	943 ± 88
Uncontrolled ReO2	21 ± 2*	42 ± 9*	1321 ± 79*
Controlled ReO2	83 ± 8**	4 ± 3**	982 ± 88**

*:p<0.05 vs No Hypoxia, **:p<0.05 vs Uncontrolled ReO₂ (ANOVA)

CONCLUSION: Abrupt uncontrolled reoxygenation causes cardiac damage and nullifies the cardioprotective effects of blood cardioplegia. Controlled reoxygenation avoids these detrimental effects. These findings imply that controlling pO₂ during induction of cardiopulmonary bypass and blood cardioplegia can be used to surgical advantage in cyanotic patients.

*By Invitation

F6. Prevention of Complement Induced Pulmonary Hypertension and Improvement of Right Ventricular Function by Selective Thromboxane Antagonism

WENDEL J. SMITH, M.D. *, MICHAEL P. MURPHY, M.D.*,

ROBERT F. APPLEYARD, Ph.D.*,

ROBERT J. RIZZO, M.D.*, LISHAN AKLOG, M.D.*

and LAWRENCE H. COHN, M.D.

Boston, Massachusetts

Despite advances in myocardial preservation and post operative management, acute reactive pulmonary hypertension can occur following open heart operations and heart transplantation. Complement activation following car-diopulmonary bypass may lead to pulmonary hypertension that causes right ventricular dysfunction in hearts with poor contractile reserve. We hypothesize that reactive pulmonary vasoconstriction is caused by complement induced thromboxane (TXA₂) production.

Sheep (n = 12) were anesthetized, instrumented for baseline measurements and randomized to receive either placebo or SQ30741 (10 mg/kg/hr), a thromboxane receptor antagonist. All sheep then received 30 units/kg of Cobra Venom Factor -Naje Haje (CVF), a known activator of complement, to produce a model of acute pulmonary hypertension. Simultaneous pulmonary artery (PA) and left atrial (LA) thromboxane levels were measured by radioimmunoassay to determine pulmonary production.

	Baseline	15 min.	30 min.	1 hr.	2 hr.
Control PAP SQ PAP (mm Hg)	8.9 ± 1.7 10.1 ± 1.6	19.4 ± 1.6 8.6 ± 1.6*	17.8 ± 1.6 8.0 ± 1.6*	9.8 ± 1.6 7.5 ± 1.6	9.4 ± 1.6 9.2 ± 1.6
Control PVR SQ PVR (dyne sec/cm5)	427 ± 109 352 ± 101	1030 ± 101 299 ± 101*	912 + 101 268 ± 101*	593 ± 101 290 ± 101	442 ± 101 395 ± 101
Control RVSW SQ RVSW (ergs x 105)	2. 66 ± .75 3.48 ± .71	4.67 ± .70# 3.28 ± .71	4.17 ± .71# 3.02 ± .71	2.31 ± .71 2.85 ± .71	2.57 ± .71 3.02 ± .71
values: Mean±SEM; PAP = pulmonary artery pressure; PVR = pulmonary vascular resistance; RVSW = right ventricular stroke work; *p<0.05 vs Control; #p<0.05 vs Baseline (ANOVA)

In all animals there was a marked increase in TXA₂ production in the pulmonary bed during the first 30 minutes of complement activation. In controls this was accompanied by significant increases in pulmonary pressure and resistance causing increased right ventricular stroke work. These changes in PAP, PVR and RVSW were completely prevented by thromboxane receptor blockade with SQ30741, suggesting that complement activation produces an increase in pulmonary vascular resistance that is specifically mediated by thromboxane production in the pulmonary bed. Systemic vascular resistance was not affected by complement activation nor thromboxane receptor blockade.

Select thromboxane receptor antagonism may effectively prevent postoperative complement mediated pulmonary vasospasm and thereby provide protection against RV dysfunction in patients with limited RV contractile reserve.

*By invitation

F7. Cyclosporine and Splenectomy Prolong Cardiac Xenograft Survival by Suppressing IL-6 Production and MHC Class II Expression

STEVEN M. PETERSON, M.D.*,

CHRISTOPHER T. STRZALKA, M.D. *,

JOHN A. JOHNKOSKI, M.D.*, BERNICE NOBLE, Ph.D.*,

EDDIE L. HOOVER, M.D. and

JACOB BERGSLAND, M.D.*

Buffalo, New York

Cyclosporine (CsA) and splenectomy (Spx) synergisticly prolong xenograft survival in the hamster to rat model. Our objective was to identify actions of this therapy on the cellular and humoral response compared to untreated controls. Inbred male Lewis rats received 15 mg/kg/day CsA and were splenectomized 2 days after heterotopic heart transplantation from Golden Syrian hamsters (group 1). Control groups included isografts (group 2) and untreated xenografts (group 3). Plasma IL-6 activity was determined using proliferation of IL-6 dependent 7TD1 hybridoma cells. Deposition of IgM, IgG, and Complement factor 3 (C3) were demonstrated using FITC-conjugated antibodies. Graft infiltrating cells (GIC) expressing MHC Class II antigen were identified with immunoperoxidase staining and OX6 (antibody for la antigen). In a random fashion, OX6+ cells were counted in a minimum of lOhpf per specimen. Results of IL-6 plasma activity in U/ml (mean ± sem):

	Day 1	Day 3	Day 10
Group 1		391 ± 98	35 ± 8
Group 2	113 ± 52	279 ± 88	152 ± 43
Group 3	1470 ± 223	1251 ± 294

Group 1 had a significant decrease in IL-6 activity at day 3 compared to untreated xenografts (p < 0.01). At day 3, light microscopy showed severe rejection in group 3. Group 1 hearts were beating strongly and only a mild cellular infiltrate was seen at day 10, similar to group 2 (isografts). Group 3 (control xenografts) displayed dense accumulation of IgM and C3 along the en-dothelium and interstitial capillaries, whereas groups 1 and 2 had minimal deposits. OX6 + GIC were markedly reduced in group 1 (2.8 ± 0.4) compared to group 3 (9.5 ± 0.6), p < 0.0005. The data indicate a direct correlation between IL-6 and xenograft rejection. The induction of B-cell maturation and stimulation of Ig production is a property of IL-6, and may play a pivotal role in the initiation of the humoral response in concordant rejection. The significant decreased deposition of humoral components (IgM and C3), decreased IL-6 plasma levels, and decreased QIC's expressing MHC Class II antigen within the nonrejecting grafts suggests that the synergy of CsA and Spx depends on the attenuation of cellular and humoral mechanisms involved in xenograft rejection.

*By Invitation

F8. Complete Prevention of Post-Ischemic Spinal Cord Injury Using Regional Perfusion With Hypothermic Saline and Adenosine

JEFFREY A. HEROLD, M.D.*, IRVING L. KRON, M.D.,

LEE BUTTERFIELD, M.D.*, LORNE BLACKBOURNE, M.D.*,

SCOTT LANGENBURG, M.D.* and

CURTIS G. TRIBBLE, M.D.*

Charlottesville, Virginia

Spinal cord injury following operations on the descending thoracic and thoracoabdominal aorta remains a persistent clinical problem. Previous attempts to decrease the risk of this devastating complication by lowering the rate of metabolism of the spinal cord have met with only varying success. We hypothesized that the tolerance of the spinal cord to an ischemic insult could be improved by using adenosine. Twenty New Zealand white rabbits underwent 40 minutes of isolated infrarenal aortic occlusion with heparin an-ticoagulation. Clamps were placed both below the left renal vein and above the aortic bifurcation. In ten rabbits (Group A), a bolus of adenosine (100 mg) was infused into the isolated aortic segment immediately after cross-clamping and this bolus was followed by a flush of hypothermic saline (8 °C, 30 ml/kg) over the first 10 minutes of ischemia time. In one control group of five.animals (Group B), cross-clamping of the descending infrarenal aorta was performed without infusion of adenosine or saline. In another control group of five animals (Group C), the aortic segment was flushed with nor-mothermic saline (37 °C) in a fashion similar to the study group. The aortic clamps were removed after 40 minutes, the abdomen closed, and the animals allowed to recover from anesthesia. Spinal cord function was assessed 12, 24, 48, 72, and 96 hours after operation by the Tarlov scale (0-no movement, 1-slight movement, 2-sits with assistance, 3-sits alone, 4-weak hop, 5-normal hop). All animals were sacrificed at 96 hours after operation and spinal cords harvested for histologic analysis. The spinal cord function of all Group A animals was fully intact with Tarlov scores of 5, while Group B and Group C animals were all paraplegic with Tarlov scores of 0 (p<0.001, ANOVA). Histologic examination of spinal cords from Group A revealed no evidence of cord injury, while Group B and Group C spinal cords had evidence of cord injury with central grey necrosis, axonal swelling, dissolution of Nissl substance, and astrocyte and macrophage infiltration. Regional infusion of the cross-clamped infrarenal rabbit aorta segment with hypothermic saline and adenosine completely prevented paraplegia in our model despite a 40 minute ischemic insult.

*By Invitation

F9. Effects of Pharmacologic Modification of Cerebroplegia Solution on Acute Recovery of Cerebral Blood Flow and Metabolism After Hypothermic Circulatory Arrest

MITSURUAOKI, M.D.*, FIMUKAZUNOMURA, M.D.*,

MICHAEL E. STROMSKI, Ph.D.*, MILES K. TSUJI, M.D.*,

PA UL R. MICKEY, M.D. *, DA VID HOLTZMAN, M.D.*

and RICHARD A. JONAS, M.D.

Boston and Cambridge, Massachusetts

Previous studies have suggested that a simple crystalloid "cerebroplegic" solution may prolong the safe duration of hypothermic circulatory arrest (CA). We explored pharmacologic manipulation of cerebroplegia solutions including an organ preservation solution containing adenosine and glutathione (University of Wisconsin U.W.) and the excitatory neuro-transmitter antagonist MK801 which has demonstrated efficacy in decreasing cerebral ischemic injury.

Forty-six 4-week old minipigs underwent core cooling to 15°C naso-pharyngeal temperature and 2 hours of CA, 45 minutes of rewarming and 3 hours of normothermic reperfusion. Group CPS (n = 12) had 50 ml/kg of saline infused into the carotid artery system at the onset of CA and repeat doses 10 ml/kg every 30 minutes. Group CPU (n = 11) received the same dose of UW. Group CPM (n = 10) received UW with 7.5 mg/1 of MK-801. Group CNT (n = 13) served as control (2 hours CA at 15°C). All solutions were delivered at 4°C. Recovery of cerebral ATP and intracellular pH (pHi) was assessed by magnetic resonance spectroscopy in half the animals in each group, and cerebral blood flow (CBF) by microspheres, cerebral metabolic rate of oxygen (CMRO2), vascular resistance reactivity to acetylcholine (ACh) and nitroglycerin (TNG) in the carotid circulation, and brain temperature were measured in the rest. Brain water content was assessed at the end of the experiment in all animals.

Results are given as men±SEM. *: p<0.05 vs. group CNT by Student-Newman-Keuls test.

	time alter reperfusion	CNT	CPS	CPU	CPM	ANOVA p values
ATP	45 min	34.1 ± 6.9	38.4 ± 4.0	62.7 ± 5.3*	63.2 ± 7.0*	0.004
(% recovery)	225 mm	28. 5 ± 9. 3	57. 3 ± 2.6*	72.2 ± 3.0*	72.9 ± 4.6*	0.000
pHi (Unit)	45 min	6. 53 ± 0.06	6.43 ± 0.10	6.54 ± 0.08	6.71 ± 0.08	0.108
(baseline = 7.12 ± 0.001)	225 min	6.57 ± 0.11	6.90 ± 0.14	7.14 ± 0.02*	7.15 ± 0.01*	0.000
CBF	45 min	33.0 ± 6.1	52.1 ± 4.5'	75.5 ± 4.7*	42.5 ± 4.3	0.000
(% recovery)	225 min	78.3 ± 6.8	96.4 ± 5.6	128.3 ± 7.6*	93. 5 ± 6.8	0.000
CMRO2	45 min	16.3 ± 9.1	27.6 ± 8.6	52.2 ± 4.7*	26.4 ± 4.9	0.000
(% recovery)	225 min	43.6 ± 5.0	59.2 ± 4.4	87.3 ± 13.1	0.002
Brain water content (%)		82.10 ± 0.22	80.74 ± 0.25*	80.53 ± 0.43*	80.56 ± 0.31*	0.000

Brain temperature at the onset on CA was 15.0±0.1°C and dropped to 13.0±0.3°C after cerebroplegia infusion and stayed lower than group CNT throughout the CA. The groups CPS and CPU showed better vascular resistance response (vasodilation) to ACh (p = 0.008) and TNG (p = 0.048) at 60 minutes of reperfusion relative to groups CNT and CPM.

Cerebroplegia improves acute recovery after 2 hours circulatory arrest. Pharmacological modification with UW further improves the recovery of cerebral blood flow and metabolism. MK-801 does not augment the protective effects of UW and reduces the recovery of cerebral blood flow, presumably by a direct vascular action. Modified cerebroplegia may provide a novel approach to improved cerebral protection when circulatory arrest is necessary.

*By invitation

TUESDAY MORNING, April 27, 1993

9:00 a.m. SCIENTIFIC SESSION - Grand Ballroom

Moderators: James L. Cox, M.D.

Martin F. McKneally, M.D.

12. Effects of Diltiazem on Perioperative Ischemia, Arrhythmias and Myocardial Function in Patients Undergoing Elective Coronary Bypass Grafting

RAINALD SEITELBERGER, M.D.*,

WALTRAUD HANNES, M.D. *, MARK GLEICHAUF, M.D.*,

HORST ZAJONC, M.D., VOLKER SCHLOSSER, M.D.*

and ROLAND FASOL, M.D.*

Freiburg, Germany

Sponsored by: Prof. Dr. Ernst Wolner, Vienna, Austria

Apart from the technical quality of the surgical procedure itself, the outcome of patients undergoing coronary bypass surgery highly depends on the quality of perioperative myocardial protection. Recent studies have shown that the continuous, perioperative infusion of calcium antagonists effectively reduces incidence and extent of myocardial ischemia. However, little is known about the influence of calcium antagonists in the prevention of perioperative arrhythmias and the preservation of myocardial function.

Consequently, a prospective, randomized study was performed on 120 patients undergoing elective coronary bypass grafting. The patients received a continuous, perioperative infusion of either the calcium antagonist diltiazem (O.lmg/kg/h, n = 60) or nitroglycerin (control group, 1µg/kg/min, n = 60) over a period of 24 hours. Perioperative monitoring included hemodynamic measurements and 3-channel Holter monitoring up to 24 hours post-operatively, repeated assessment of 12-lead ECG and analysis of ischemia-specific laboratory parameters (CK, CK-MB, CK-MB-mass, troponin-T and myoglobin) until the 6th postoperative day. In addition, regional and global myocardial function was assessed preoperatively and 1 and 4 hours after car-diopulmonary bypass by means of transesophageal echocardiography (monoplane 5MHz faced array transducer). In order to assess the diastolic compliance of the left ventricle, transmitral flow-velocity profiles were performed in the transesophageal long-axis view.

The two groups did not differ with respect to preoperative and operative (number of distal anastomoses, aortic cross clamp and extracorporal circulation time, etc.) data. Except for a significant reduction in perioperative heart rate by an average of 9 beats/min, diltiazem had no influence on hemodynamic parameters. The antiischemic efficacy of diltiazem lead to a reduction of the number (17 ± 9 vs. 25 ± 5, p<0.05) and duration (69 ± 47 vs. 104 ± 87 min, p<0.05) of transient ischemic events and a lower incidence of perioperative myocardial infarction (3.3 vs. 6.7%) as compared to the nitroglycerin group. In addition, peak values of all assessed laboratory parameters were significantly lower in the diltiazem group. With regard to perioperative arrhythmias, patients treated with diltiazem had a lower incidence of perioperative atrial fibrillation (5 vs. 18%, p<0.05) and lower numbers of ventricular premature beats/hour (10±8 vs. 19±22, p<0.05). Perioperative regional and global systolic myocardial function did not significantly differ between both groups. However, the incidence of diastolic compliance disturbances at 4 hours after cardiopulmonary bypass was 53% in the nitroglycerin-, but only 26% in the diltiazem group (p<0.05).

It is concluded that perioperative infusion of the calcium antagonist diltiazem does not adversely affect perioperative hemodynamics and systolic myocardial function and provides potent antiischemic and antiarrhythmic protection in patients undergoing aortocoronary bypass grafting. In addition, diltiazem markedly improves the postoperative diastolic compliance of the left ventricle.

*By Invitation

13. Chordal Preserving Mitral Valve Replacement: The Hemodynamic Study in the Early and Mid-Term Period

YUTAKA OKITA, M.D.*, SHIGEHITO MIKI, M.D.*,

YUICHI UEDA, M.D. *, TAKAFUMI TAHATA, M.D. *,

TETSURO SAKAI, M.D. * and

KATSUHIKO MATSUYAMA, M.D. *

Nara, Japan

Sponsored by: Tadaomi Miyamoto, M.D., Kitakyushu, Japan

The clinical significance of the chordae tendinae in regards to postoperative left ventricular (LV) performance was evaluated in patients with mitral regurgitation (MR) or mitral stenosis (MS) who underwent either mitral valve replacement (MVR) using St. Jude Medical valve with complete chordae preservation (P) or with conventional MVR (C), or valve repair (R). [Patients and methods] (P group): MVR preserving autologous chordae tendinae (n = 37) or replacing chordae with Gore-Tex sutures (n = 13) was performed in 50 patients, 30 with MR and 20 with MS. The valve pathology was rheumatic in 17 patients, degenerative in 10, and infective endocarditis in 3. (C group): conventional MVR involving 25 with MR and 28 with MS. (R group): valve repair in 24 with MR or commissurotomy in 27 patients with MS. The LV performance was analyzed by cineangiography (CAG) in the early (mean 1.2 months), and multiple gated blood scintigraphy (MUGA), or echocardiography (UCG) in the late postoperative periods (mean 5.4 years) in 3 groups of patients. Statistical analysis was performed using Student T-test. [Results: Table] C patients with MVR for MR had a significantly greater end-systolic volume index (ESVI), the LV ejection fraction (EF) was unchanged in the P group, but was decreased in C and R group. The LV contractility index [end-systolic circumferential left ventricular wall stress (ESS)-ESVI-'] was better in P than C group. Both the LVEF by MUGA, and LV fractional shortening (FS) by UCG were significantly higher in P and R than in C group patients. P group exhibited superior LV performance than C group especially in those with MR and depressed preoperative left ventricular function (EF<0.50). There were no significant difference, except for better postoperative EF (CAG) in P group, between the 3 groups in patients with MS. [Conclusion] The results support the concept that maintenance of continuity between the mitral annulus and the papillary muscle has a beneficial effect on postoperative left ventricular performance, especially in patients with MR and depressed preoperative left ventricular function, but has no major effect in patients with MS.

MR
Group		P(n = 30)	C(n = 25)	R(n = 24)
EDV1	pre	118 ± 46	126 ± 37	140 ± 42
(ml-m-2)	post	96 ± 16*	94 ± 20*	113 ± 20
ESVI	pre	53 ± 20	58 ± 18	58 ± 25
(ml-m-2)	post	42 ± 13	50 ± 15	52 ± 14
EF (CAG)	pre	0.53 ± 0.11	0.53 ± 0.08	0.59 ± 0.12
	post	0.56 ± 0.21*	0.49 ± 0.10	0.54 ± 0.09*
ESS/ESVI	pre	4375 ± 1572*	3622 ± 579	3809 ± 138
($)	post	4063 ± 1027*	3224 ± 914	3837 ± 667*
EF (MUGA)	pre	n.a.	n.a.	n.a.
	post	0.57 ± 0.08*	0.51 ± 0.09	0.61 ± 0.07*
FS	pre	0.38 ± 0.08	0.34 ± 0.08	0.36 ± 0.06
	post	0.36 ± 0.07*	0.29 ± 0.08	0.33 ± 0.08*
MS
Group		P (n = 20)	C(n = 28)	R(n = 27)
EDVI	pre	75 ± 22	85 ± 28	78 ± 18
(ml-m-2)	post	91 ± 23	81 ± 22	72 ± 16
ESVI	pre	41 ± 15	42 ± 16	39 ± 13
(ml-m-2)	post	40 ± 15	39 ± 14	33 ± 8
EF (CAG)	pre	0.47 ± 0.08	0.52 ± 0.08	0.50± 0.11
	post	0.56 ± 0.10*	0.48 ± 0.09	0.54 ± 0.08
ESS/ESVI	pre	4685 ± 1 102	4412 ± 1369	4799 ± 1564
($)	post	4787 ± 827	4526 ± 1587	4880 ± 528
EF (MUGA)	pre	n.a.	n.a.	n.a.
	post	0.57 ± 0.08	0.55 ± 0.11	0.56 ± 0.09
FS	pre	0.31 ± 0.09	0.30 ± 0.07	0.32 ± 0.08
	post	0.31 ± 0.08	0.34 ± 0.07	0.34 ± 0.09
Legends; All numerical shows Mean ± SD. n.a.: not available. ($): kdyn-cm-1ml-1-m2 Asterisk (*) shows statistically significant differences (p<0.05) compared with other groups.

*By Invitation

14. Management Needs Revision Following Valve Replacement by St. Jude Medical Prosthesis: Unexpected Results after 10 Years of Prospective Follow-Up of 600 Consecutive Patients

DIETER HORSTKOTTE, M.D.*, H.D. SCHULTE, M.D.*,

WOLFGANG BIRCKS, M.D. and B.E. STRAUER, M.D.*

Dusseldorf, Germany

Between 1978 and 1982, SJM-prostheses were implanted in 600 consecutive patients (P) averaging 50.7 ± 9.6 (14-81) years of age in the aortic (n = 298), mitral (n = 215) position or as multiple valve replacement (n = 87). P had in-hospital examinations every 3 to 6 months. Follow-up was complete and averaged 122.2 ± 1.1 months for late survivors. By random, three different anticoagulation regiments had been used with a target INR 3.0-4.5, INR 2.5-3.2 and INR 1.8-2.7. For these 3 regiments the incidences (events/100 patient-years) for thromboembolic (TE) and bleeding complications (BL) were calculated:

	INR 3.0-4.5	INR 2.5-3.2	INR 1.8-2.7
Aortic	n = 125	n = 86	n = 87
Severe TE	n = 1 (0.08)	n = 2 (0.24)	n = 1 (0.24)
Other TE	n = 21 (1.76)	n = 5 (0.61)	n = 6 (0.73)
Severe BL	n =9 (0.75)	n = 4 (0.49)	-
Other BL	n = 59 (4.94)	n = 31 (3.80)	n = 16 (1.95)
All	n = 90 (7.54)	n = 42 (5.14)	n = 24 (2.93)
Mitral	n = 94	n = 62	n = 59
Severe TE	n = 2 (0.24)	n = 3 (0.52)	n = 3 (0.53)
Other TE	n = 28 (3.30)	n = 28 (4.81)	n = 33 (5.83)
Severe BL	n = 9 (1.06)	n = 3 (0.52)	-
Other BL	n = 80 (9.43)	n = 35 (6.02)	n = 16 (2.83)
All	n = 119 (14.02)	n = 69 (11.86)	n = 52 (9.19)
Multiple	n = 35	n = 29	n = 23
Severe TE	n = 1 (0.31)	n = 1 (0.37)	n = 1 (0.46)
Other TE	n = 11 (3.39)	n = 13 (4.86)	n = 15 (6.97)
Severe BL	n = 4 (1.23)	n = 2 (0.75)	n = 1 (0.46)
Other BL	n = 29 (8.94)	n = 19 (7.11)	n = 7 (3.25)
All	n = 45 (13.87)	n = 35 (13.09)	n = 24 (11.16)

The optimal efficacy risk relation for P with SJM aortic prostheses were calculated as a target INR 2.4-2.6; for SJM mitral prostheses P with atrial contraction, left atrial diameter < 26 mm/m², an INR 2.4-2.6; for SJM mitral prosthesis P without atrial contraction, left atrial diameter > 30 mm/m², an INR 2.7-3.1; and for SJM multiple valve replacement INR 2.8-3.2. These findings give further evidence of the low thrombogenicity of the SJM pros-theses and supports arguments for a much lower intensity of anticoagulation than generally recommended.

10:00 a.m. INTERMISSION - VISIT EXHIBITS

*By Invitation

10:45 a.m. SCIENTIFIC SESSION - Grand Ballroom

Moderators: John L. Ochsner, M.D.

Martin F. McKneally, M.D.

15. Surgical Treatment of Barrett's Carcinoma. Correlation Between Morphologic Findings and Prognosis

TONI LERUT, M.D.*, WILLY COOSEMANS, M.D.*,

DIRK VAN RAEMDONCK, M.D.*, PAUL DELEYN, M.D.,

JEAN MARC MARNETTE, M.D.* and

KAREL GEBOES, M.D.*

Leuven, Belgium

Sponsored by: Tom R. DeMeester, M.D., Los Angeles, California

Barrett carcinoma occurred in 66 of 331 patients with adenocarcinomas of the oesophagus or GE junction. Thirty-two (46%) of these patients had a positive history for gastroesophageal reflux. A history of alcohol and tobacco was absent in 50% and 47.5%, respectively. The mean length of Barrett metaplasia was 7.37 cm. Operability was 98.5%, and resectability 95.5%. There was no postoperative or hospital mortality. Pathological staging was: stage I: 38.3%, stage II: 20.6%, stage III: 22.2% and stage IV: 19%. Overall survival is 80.5% at 1 year, 62.7% at 2 and 58.2% at 5 years. Five-year survival for stage I is 100%, for stage II: 87.5%, for stage III: 22.2% and for stage IV: 0. Thirty-four (51.5%) patients were under surveillance for a related or unrelated condition before diagnosis of their carcinoma, only 9 (26.5%) had positive lymph nodes. Thirty-two had their diagnosis made at their first medical contact, 78% had positive lymph nodes. Five-year survival in patients without nodal metastasis was 85.3% and significantly better than those with metastasis, 38.3% (p = 0.0033).

Out of 66 patients, 19 (28.7%) had a biopsy proven history of Barrett metaplasia before malignancy occurred. Mean time interval between diagnosis of metaplasia and malignant degeneration was 3.8 years (89.5% > 1 yr). Over the surveillance period, the length of metaplastic Barrett remained unchanged in all patients. Barrett ulceration was present from the beginning in 14 patients, 3 patients had never had an ulcer. Intestinal metaplasia was seen in 18 patients. Combining pre- and postoperative pathology revealed severe dysplasia in 16 patients. 13.7% of the 19 patients had stage I disease.

Conclusions: 1) Half of the patients with Barrett carcinoma have no history of reflux or tobacco and alcohol use. Early diagnosis was possible in patients under surveillance with 73.7% diagnosed with stage I disease and 100% five-year survival. Barrett ulcer, intestinal type of metaplasia, were common findings in patients under surveillance with a biopsy-proven history of Barrett metaplasia before carcinoma.

*By Invitation

16. Risk of Replacement of Descending Aorta With a Standardized Left Heart Bypass Technique

HANS G. BORST, M.D., MICHAEL JURMANN, M.D.*,

BEA TE BUHNER * and JOACHIM LAAS, M.D.*

Hannover, Germany

From 4/1986 to 11/1992, 130 patients (pts) aged 48 (15 to 73) years underwent replacement of the descending aorta (DA) using centrifulgal pump (Bio-Medicus) left heart bypass for the following conditions: Aneurysm = 66 (50.8%), dissection = 63 (48.5%), tumor = one. There were 55 aortic redos. The bypass was routed from the left atrium to peripheral vessels in 87, to the downstream aorta in 43 pts. Replacement extended to the 6th intercostal space (ICR) in 59 (45.4%), to the 9th ICR in 53 (40.8%), and was total/subtotal in 21 (16.2%) pts. Aortic clamp time was 45 (16 to 98) minutes. Intercostal arteries down to and including the 6th ICR routinely were sacrificed. Staged aortic clamping with or without intercostal artery reattachment was employed in 39 patients.

Thirty-day mortality was 3%. There were no bypass-related complications. Six patients (4.6%) showed spinal cord sequelae: paraplegia was seen in 4 pts which disappeared in one and improved in 2. Fully reversible paraparesis was noted in 2 pts. The rate of permanent neurological deficit therefore was 2.3%. These complications occurred only in aortic replacement to or beyond the 9th ICR and could not be totally prevented by staged aortic clamping and intercostal artery reconnection despite a significant reduction of spinal cord ischemia time thus pointing to yet uncontrolled factors.

We conclude that full exhaustion of the methodology of left heart bypass is associated with a low risk of early death and spinal complications. It therefore is considered the method of choice in replacement of the DA.