WEDNESDAY MORNING, APRIL 20, 1977

8:30 A.M. Scientific Session

Grand Ballroom

23. Extracorporeal Membrane Oxygenation (ECMO) in Newborn Respiratory Failure

ROBERT H. BARTLETT, ALAN B. GAZZAN1GA, ROBERT F. HUXTABLE*,

HOUKJE C. SCHIPPERS*, MELODY O'CONNOR* and

MICHAEL R. JEFFERIES*,Orange, California and Minneapolis, Minnesota

Respiratory failure in the newborn is caused by: Respiratory Distress Syndrome (RDS), Meconium Aspiration Syndrome (MAS), Persistent Fetal Circulation (PFC), Congenital Diaphragmatic Hernia with PFC, or Streptococcal Pneumonitis. All of these disorders are potentially reversible, but high Fi0₂ and ventilator pressure necessary for treatment may add to existing pulmonary damage causing death or chronic bronchopulmonary dysplasia. Temporary life support with ECMO offers the advantages of gaining time, lowering Fi0₂ and pressure, and permitting unusual techniques of pulmonary management.

After extensive laboratory investigation, we have used ECMO in thirteen newborn infants who were moribund from respiratory failure. Indications and results include: RDS (3, 1 Survived), MAS (6, 3 S), PFC (1,1 S), PPC with CDH (2, 1 S), Streptococcal Pneumonitis (1). Venoarterial bypass at flow rates approaching total cardiac output was used for 1-8 days. Bleeding was not a major problem despite heparinization and thrombocytopenia. Patient's ductus arteriosus was ligated during ECMO in four cases. VA bypass permitted low Fi0₂ and pressure; lung function improved rapidly in ten cases. Pulmonary and neurologic function, growth and development are normal in the six survivors.

This study, which includes the first six successful cases, demonstrates that ECMO is a useful method in the treatment of newborn respiratory failure, and suggests that high Fi0₂ and barotrauma are major contributors to the pathogenesis of pulmonary damage.

24. Pneumothoraces in Children with Chronic Pulmonary Disease

SUSAN R. LUCK*, JOHN G. RAFFENSPERGER, HENRY J. SULLIVAN*

and LEWIS E. GIBSON*, Chicago, Illinois

Spontaneous pneumothorax is a frequent complication of chronic lung disease in children. In most instances closed thoracotomy tube drainage will rapidly re-expand the lung. However, prolonged drainage for continued air leak contributes to morbidity and increases mortality of the disease.

Pneumothoraces occurred in two percent of 140 children with cystic fibrosis over a five year period. An immediate ten to twenty percent mortality has been reported. These pneumothoraces are often under tension and further compromise limited respiratory reserve. Chest tube treatment interferes with postural drainage and further increases associated mortality because the child can't raise his copious secretions. Two patients were treated with bilateral thoracotomy, bleb resection, and mechanical pleurodesis. One boy recovered promptly despite severely compromised pulmonary function. Another child had continuous tube drainage for six weeks prior to thoracotomy. She died one week later with respiratory failure although her pneumothorax did not recur. We feel that a more prompt operation should be considered in this group of children. With a limited incision, intercostal nerve blocks, and vigorous tracheal toilet they tolerate an operation well.

One five year old child with histiocytocis X had bilateral pneumothoraces requiring tube drainage for eight weeks. He promptly recovered after bilateral thoracotomy, control of air leaks, and pleurodesis. A fourth child, a fourteen year old girl had tube drainage for three weeks. At operation she had typical apical blebs.

Open thorocotomy will yield good results even in patients with far advanced pulmonary disease. Postoperative complications and mortality are not increased over that seen without operation. Length of hospitalization and relative confinement is decreased.

*By invitation

25. Hypertension and the Renin-Angiotensin System Following Open-Heart Surgery

KENNETH M. TAYLOR*, IAN MORTON*, JOHOIDA J. BROWN*,

WILLIAM H. BAIN* and PHILIP K. CAVES*,

Glasgow, Scotland, U.K.

Sponsored by: NORMAN SHUMWAY, Stanford California

Systolic hypertension and peripheral vasoconstriction are frequently encountered for the first few hours after cardiopulmonary bypass (CPB). Their aetiology is poorly understood, although it has been suggested that the remn-angiotensin system may play an important role. Angiotensin II (AII) - the end product of stimulation of the renin-angiotensin system, is the most powerful naturally occurring vasopressor agent yet identified. We have, therefore, studied plasma AH levels in 10 patients during and after open-heart surgery.

Plasma AH was measured by radio-immunoassay (normal range 5-35 pg/ml). The patients studied (Group 1) were submitted to elective cardiac surgical procedures utilising CPB with non-pulsatile flow at a mean arterial blood pressure = 49 mmHg (± 8.3 S.E.M.). Patients submitted to closed mitral valvotomy were used as controls (Group 2). Serial samples were obtained before, during and for 2 hours post-operatively in both groups.

In both groups, pre-operative plasma AH levels were normal and showed a small rise after the chest had been opened (Mean plasma AH levels <60 pg/ml in both groups). In Group 1, plasma All levels rose markedly during CPB to a mean level of 170 pg/ml ± 28 S.E.M. A very high level was maintained at 2 hours post-operatively (155 pg/ml ± 23 S.E.M.). Thereafter, All levels fell rapidly and had returned to normal 4-24 hours post-operatively in all but one patient. This patient showed a progressive rise of AH post-operatively to a plasma All level of 385 pg/ml at 24 hours, and died after 48 hours from low cardiac output associated with subendocardial necrosis. By contrast, control patients (Group 2) showed no further rise in All levels during or post-operatively.

This study shows that marked elevation in the plasma AH level occurs during and for several hours after non-pulsatile CPB, the period when peripheral vasoconstriction is most marked. We have concluded that the renin-angiotensin system plays an important role in the occurrence of peripheral vasoconstriction and hypertension in the early post-operative period and may also be of critical importance in the production of the low-output syndrome.

*By invitation

26. Prospective Analysis and Treatment of Perioperative Hypertension Related to Coronary Artery Surgery.

ARTHUR J. ROBERTS*, STEPHEN D. HERMAN*, RONALD M. ABEL*, WILLIAM A. GAY, JR.* and VALAVANUR A. SUBRAMANIAN*,

New York, New York

Sponsored by: ARTHUR J. OKINAKA, New York, New York

Forty consecutive patients undergoing saphenous vein bypass graft (SVBG) operations were studied to determine the incidence of systemic hypertension (HYPT) associated with coronary bypass surgery. In 15 of the 40 patients (37.5%) HYPT developed as defined by systolic blood pressure (SBP) >160 or diastolic blood pressure (DBF) >100. HYPT developed within two hours postoperatively (PO) in 11 patients during induction of anesthesia in one patient, and intra-operatively in three patients. A comparison of hemodynamic status two hours PO between patients with HYPT (N=15) and then without HYPT (N=12) showed only a significantly higher systemic vascular resistance in the HYPT group. Plasma catacholamines (nor-epi, dopamine) were also elevated significantly in the HYPT group. Nitroprusside (NP) infusion was given to control BP in 10 patients. The dose was adjusted to decrease SBP <^ 140 with a dose range of 40-100 mgm/min. Hemodynamic data was recorded prior to NP and 20 minutes after infusion.

	Control ← 20 minutes→	Nitroprusside	P-Value
HR	86.75 ± 4.31	94.88 ± 3.80	<.0025
Mean BP mm. Hg.	131.0 ± 3.89	89.75 ± 3.21	<.0005
SBP mm. Hg.	180.0 ± 5.40	129.17 ± 5.52	<.0005
DBP mm. Hg.	96.5 ± 4.60	66.74 + 3.16	<.0005
MPAP mm. Hg.	16.83 ± 2.39	13.17 ± 1.95	<.0025
PWP mm. Hg.	10.25 ± 2.01	7.59 ± 1.75	<.0025
RAP mm. Hg.	9.6 ± .94	7.18 ± .89	<.0005
SVR dynes sec. cm. 5	3509.17 ± 215	2126. ± 165	<.0005
PVR dynes sec. cm.5	229.07 ± 23	161.8 ± 17	<.01
LVSWI 9 m/m2	59.63 ± 4.9	38 ± 3.4	<.0025
RVSWI 9 m/m2	5.01 ± .84	3.55 ± .78	<.005
SVIml/m2	32.75 ± 2.2	32.86 ± 1.8	<.2
CI 1/min/m2	2.83 ± .19	3.16 ± .18	<.025

The data suggest that the incidence of systemic hypertension related to SVBG surgery is high and that elevated catacholamines is associated with elevated systemic vascular resistance and may play a role in the development of post operative hypertension. Nitroprusside rapidly and effectively reduces blood pres sure. Increases in cardiac index from nitroprusside administration appears to be due to an increase in stroke volume when LVFP >7, but reflex tachycardia when LVFP <7.

INTERMISSION - VISIT EXHIBITS

*By invitation

27. The Long-Term Outlook for Valve Replacement in Active Endocarditis

BENSON R. WILCOX, GORDON F. MURRAY and

PETER J. K. STAREK*, Chapel Hill, North Carolina

Treated with antibiotics alone, the prognosis is poor in infective endocarditis associated with congestive failure (>75% mortality). Viewing surgical intervention as the only alternative to this dismal outlook, we have replaced 23 valves in 22 patients over the past ten years during the active phase of infective valvular endocarditis. Short term results reported from our institution (90% thirty day survival) and by other groups have been gratifying. This report details the long-term outlook for such patients.

In our series there have been five deaths, four occurred within six months of initial operation. Three died at reoperation for paravalvular leak and severe heart failure; one died secondary to renal failure complicating prolonged preoperative heart failure. Two of the deaths were in patients with annular abscess. A fifth patient died three years postop with metastatic lung carcinoma. Significant postoperative complications occurred in eleven patients. These complications can be separated into four groups, I - paravalvular problems, 5; II - congestive failure, 4; III - heart block, 2; and, IV - systemic emboli, 5. Many of these complications were transient as evidenced by the fact that 17 of the 22 patients have been followed an average of four years (1 year toll years) and all are leading active lives.

Except in patients with annular abscess, the long-term outlook is as good in patients with active endocarditis as in other individuals under-going valve replacement. These results clearly establish surgery as an important mode of treatment in active infective endocarditis.

*By invitation

28. Re-Replacement of Prosthetic Heart Valves: A 15 Year Experience

JOSEPH G. SANDZA*, RICHARD E. CLARK, JOHN P. CONNORS*,

THOMAS B. FERGUSON and CLARENCE S. WELDON,

St. Louis, Missouri

Since the beginning of prosthetic valve replacement at the Barnes Hospital, 773 patients have received 884 prosthetic valves of various types between 1962 and October, 1976. Sixty-eight (8.8%) required re-replacement with 79 valves (79/884 8.9%). Ten patients (12.6%) had a prosthesis in the same position changed twice. The incidence of re-replacement by position was mitral 13.3%, aortic 5.8%, and tricuspid 4.6%. The re-replacement valves were classified by the predominant problem: (a) periprosthetic leak (L) 48% (38/79), (b) prosthetic valve infection (I) 27.8% (22/79), (c) prosthetic valve wear (W) 16.4% (13/79), and valve thrombosis (T) 7.6% (6/79). Subset incidences by position for mitral, aortic and tricuspid re-replacements were 63%, 34% and 2.5% respectively. Mortality was related to the disease category as follows L=23.7%, I=50%, W=23%, T=33%. Importantly, mortality was related to valve position, mitral 36%, aortic 26%, tricuspid 0%. The worst clinical combination was an infected mitral prosthesis with a mortality of 55% and the lowest mortality was found in the aortic valve leak group (15.3%) if tricuspid re-replacement is not considered. The operative mortality for the entire group was 36% and there was a late mortality of 18.6%. The primary cause for operative death for re-replacement valve surgery was profound left heart failure immediately after cessation of cardiopulmonary bypass when 80% of early deaths occurred. Mortality was also related to the number of re-replacements. Those having a prosthetic valve replaced more than once had an overall mortality of 50% (5/10), but patients having a second re-replacement for an infected prosthesis had a mortality of 83% (5/6). There were no deaths among the patients who had second re-replacement for clotted or leaking valves. No significant improvement in mortality has occurred with time. If it is considered that valve wear and thrombosis are related to valve manufacture and that valve leak and infection are surgical complications, it is seen that 80% of the deaths in the re-operated patients (20/25) are preventable, and that the incidence of re-replacement (76%) can be reduced by meticulous surgical technique and medical care.

*By invitation

29. Factors Influencing Long-Term Survival After Isolated Aortic Valve Replacement

JACK G. COPELAND*, RANDALL B. GRIEPP*,

EDWARD B. STINSON* and NORMAN E. SHUMWAY, Stanford, California

Followup information was obtained for 1,131 patients having had isolated aortic valve replacement between May, 1963 and April, 1976. The mean followup period was 4.4 yrs. and the total study included 4,125 patient years. Valves utilized included Starr-Edwards series 1,000 (83 pts.), 1,200 (204 pts.), 1,260 (435 pts.), 2,320 (49 pts.), fresh allografts (103 pts.), and porcine xenografts (251 pts.). The effects of preoperative variables upon long-term survival were evaluated using actuarial analysis. Five significant non-invasive preoperative determinants of long-term survival were identified (five-year survival rate and P value shown in parentheses): Age 31-64 vs. ≥.64 years (70% vs. 57%, p <.0001), NYHA functional Class II or III vs. IV (80% or 67% vs. 50%, p <.003), radiographic cardiac enlargement vs. no enlargement (64% vs. 75%, p = .007), congestive heart failure vs. no CHF (63% vs. 76%, p = .002), and remote MI vs. no MI (56% vs. 69%, p = .004). Invasive preoperative determinants were also evaluated, pure AS or AS-AR vs. pure AR (72% or 70% vs. 50%, p = .002), left atrial mean pressure <16 mmHg vs. ≥ 16 mmHg (76% vs. 62%, p = .005), pulmonary artery mean pressure < 30 mmHg vs. ≥ 30 mmHg (74% vs. 57%, p = .006), coronary arteriogram normal vs. CAD (at 3'/z years 80% vs. 54%, p = .0002), and left ventriculogram normal vs. LV dysfunction (at 3Vi years 74% vs. 56%, p = .02).

Overall survival was 68% for all patients and 77% for discharged patients at 5 years after operation. Analysis of combined cumulative risk for thromboembolism, valve failure and prosthesis-related death for each valve type showed the percentage of patients free of prosthesis-related complications at 2 ^½postoperative to be as follows: porcine xenograft 78%, SE 1200 and 1260 68%, SE 1000 59%, SE 2320 57%, and allograft 50%. Differences for xenograft vs. SE 1000, 1200, 1260, 2320, and for allograft vs. SE 1200 and 1260 were significant (p <.03).

*By invitation

30. Evaluation of Computer Aided Monitoring of Postoperative Cardiac Patients

L. H. EDMUNDS, JR., H. MAC VAUGH, III,

J. M. STEVENS* and A. E. WECHSLER* Philadelphia, Pennsylvania

The unique economic, medical and educational benefits of a "turn key" computer aided patient monitoring system (Roche Medical Electronics) were evaluated in a prospective, randomized, controlled study of 300 postoperative cardiac patients. Using predetermined ranges of 6 to 11 measurements ("limits"), systematized care was common to both control patients (C) and computer system patients (S). ‘‘Limits" were written on single bedside sheets for C and were entered into the computer for S. Criteria related to "smoothness" of convalescence and work sampling studies were recorded.

Time to reach milestones (e.g. extubation), time outside of each of the 6 to 11 "limits", crises, complications, deaths, duration of intensive care and hospital stay did not differ between groups. During the first year S required more staff care (15.8 hours) than did C (12.6 hours). Subsequently 2.04 hours were saved in S in making measurements, charting and communication, and direct patient care increased 0.56 hours (S 11.2 hrs.; C 12.6 hrs.; P, ns). Downtime of the computer system averaged 16.4 hours per week during the first year and most recently 14.2 hours per week. Frequent failures necessitate on-site technical help and backup front-end units. The fluid infusion system is less versatile and reliable than the manual system.

Current computer aided monitoring systems do not provide discernible medical benefits after cardiac surgery and only small time savings in making, recording and displaying measurements. Downtime and need for backup units and technical help are serious medical and economic disadvantages. The benefit/cost ratio is low.

*By invitation