Inter-hospital transport of neonatal patients on extracorporeal membrane oxygenation: mobile-ECMO.

OBJECTIVE: To describe the equipment, personnel requirements, training, management techniques, and logistic problems encountered in the design and implementation of a mobile extracorporeal membrane oxygenation (ECMO) program. DESIGN: This is a report of a technique for the transport of patients on ECMO and a description of our retrospective case series. SETTINGS: The study was conducted at a regional referral children's hospital and ECMO unit. PATIENTS: Thirteen neonatal medical patients with acute respiratory failure were transported with mobile-ECMO. RESULTS: Over a 24-month period, we transported 13 neonatal patients with mobile-ECMO. The reason for transport with mobile-ECMO was inability to convert from high-frequency ventilation (4 of 13), patient already on ECMO (1 of 13), and patient deemed too unstable for conventional transport (8 of 13). Eleven of the 13 patients were transported from other ECMO centers. Of the 13, 9 survived. No major complications during transport were reported for any of the patients. Follow-up data were available on all nine survivors of neonatal mobile-ECMO. Eight of these had normal magnetic resonance imaging scans of the brain; the ninth had a small hemorrhage in the left cerebellum. CONCLUSION: Our limited series shows that patients can be safely transported with mobile-ECMO. This program does not replace the early appropriate transfer for ECMO-eligible patients to an ECMO center.

Evaluation of the reconstructed carotid artery following extracorporeal membrane oxygenation.

Over a 12-month period, 28 neonatal patients in respiratory failure were supported with extracorporeal membrane oxygenation (ECMO), and 11 of these underwent successful repair of the right carotid artery post-ECMO. Nine of 11 were studied with duplex color-flow Doppler imaging between 14 and 109 days of age and again at 1 year of age. A velocity ratio (A/B) of the peak systolic velocity above the level of the anastomosis to the peak systolic velocity below the anastomosis was measured to assess the degree of stenosis, if any, at the repair site. Antegrade flow through the carotid was detected post-ECMO in 8 of 9 infants, and antegrade and retrograde flow was documented in 1 infant. A/B ratios ranged from 1.00 to 8.60 (A/B ratio of 1 is normal; 2.0 indicates at least a 50% obstruction to flow). Four of 9 patients had ratios greater than 2.0, and 8 of 9 exhibited some evidence of obstruction. Follow-up scans were performed on 8 of 9 infants between 12 and 18 months of age. All infants examined showed marked improvement in A/B ratio and patency of the vessel, documenting that initial narrowing of the vessel is reversible. No infant had evidence of embolic phenomena to the right side of the brain by magnetic resonance imaging. Repair of the common carotid artery post-ECMO is technically feasible without increasing the risk of brain injury.

Congenital mitral valve stenosis in infancy: a different approach to a difficult problem.

Mitral valve replacement and bypass are high-risk operations in the infant with severe mitral stenosis. A palliative approach is presented that obviates the need for valve replacement in infancy.

Pulmonary vascular disease and hypertension after valve surgery for mitral stenosis.

The reduction of pulmonary hypertension that occurs within 24 hours of valve replacement for mitral stenosis is well documented, but patients who die after surgery have not been adequately studied. Clinical and autopsy data for 16 patients who died following mitral valve replacement were reviewed. The emphasis was on preoperative and postoperative pulmonary arterial pressure and pulmonary vascular disease, including arterial, venous, and capillary changes. Morphologic features were graded and summed to obtain an additive histologic assessment (AHA). Patients were divided into three groups: 1) those who had uneventful operations and early postoperative periods but died prior to discharge; 2) those who had postoperative difficulty, with identifiable acute anatomic causes of death; and 3) those who had postoperative difficulty, with no apparent acute anatomic cause of death. In group 1 (n = 4) the preoperative pulmonary arterial pressure was 43 +/- 17 mm Hg, and AHA ranged from 0 to 4; in group 2 (n = 5) the preoperative pulmonary arterial pressure was 60 +/- 15 mm Hg, but AHA ranged only from 2 to 5. In group 3 (n = 7) the preoperative pulmonary arterial pressure was 59 +/- 12 mm Hg; AHA ranged from 6 to 9, significantly higher than that of the other groups (P less than 0.005). Three patients from group 3 had elevated pulmonary arterial pressure (60, 52, and 50 mm Hg three, six, and 15 days after surgery, respectively). Two additional patients had right heart failure with normally contracting left ventricles terminally. It is concluded that some patients with mitral stenosis who die after surgery with persistently elevated pulmonary arterial pressure have sufficiently severe pulmonary vascular disease to account for their persistent pulmonary hypertension and death.

Antibiotic prophylaxis for cardiovascular surgery. Efficacy with coronary artery bypass.

Two hundred twenty patients were randomly assigned to receive either ceforanide or cephalothin as perioperative antibiotic prophylaxis during cardiovascular surgery. More infections were seen among cephalothin recipients (8 deep, 32 total) than among ceforanide recipients (1 deep, 17 total). Among patients who had only coronary artery bypass grafting, more cephalothin recipients had infection than did ceforanide recipients (19 of 82 as opposed to 7 of 83; p = 0.001; relative risk, 2.7; 95% confidence interval, 1.22 to 6.18). The difference between the two regimens was attributable to fewer blood, wound, and urinary tract infections. Among patients who had other procedures, there was no difference in the efficacy of the two regimens. Cephalothin recipients who developed wound or blood stream infections had lower antibiotic levels in their atrial appendages than recipients not developing such infections (p = 0.02). If one assumes that cephalothin does not increase the risk of infection, then these data show that antibiotic prophylaxis prevents infection after coronary artery bypass surgery, and, in the dosages used, that ceforanide is superior to cephalothin.

Determinants of operative survival following combined mitral valve replacement and coronary revascularization.

To determine the operative survival rate following combined mitral valve replacement (MVR) and coronary artery bypass graft (CABG) operation, we evaluated 100 patients, who were seen consecutively at the Peter Bent Brigham and Brigham and Women's Hospital from 1972 to 1982. There were 63 men and 37 women; the mean age was 62 years. Thirty-six patients were in New York Heart Association (NYHA) Functional Class III, and 64 were in functional Class IV. Mitral regurgitation was predominant in 76 patients; mitral stenosis, in 24. Emergency operations were performed in 15 patients, and elective or semielective operations were performed in 85. There were 18 operative deaths (18%): 9 in patients having elective operations (10.5%) and 9 in those having emergency operations (60%; p less than 0.01). Significant preoperative factors related to operative death were NYHA functional class, increased pulmonary vascular resistance, lower cardiac index, and lower ejection fraction in the nonsurvivors. The rate of survival did not differ according to sex, age, or degree of coronary artery disease. In addition, myocardial protection with potassium cardioplegia and complete coronary revascularization significantly reduced operative mortality in the elective group of patients but did not alter the mortality in the emergency group.

The in-hospital risk of rereplacement of dysfunctional mitral and aortic valves.

From January 1, 1972, to January 1, 1981, 1282 consecutive valve replacements were performed at the Peter Bent Brigham Hospital. Fifty-eight of these (4.5%) were rereplacements: 14 for prosthetic valve subacute bacterial endocarditis (SBE), 29 for primary valve dysfunction (PVD) and 15 for perivalvular leak (PVL). Aortic rereplacement was done in 27 patients (eight for SBE, 10 for PVD and nine for PVL) and mitral rereplacement in 31 patients (six for SBE, 19 for PVD and six for PVL). Twenty-six patients were in New York Heart Association functional class III and 32 were in class IV. Ten class IV patients underwent emergency valve replacement (three for SBE, two for PVD and five for PVL). The overall in-hospital mortality rate for valve rereplacement was 14% (eight of 58 patients). For mitral valve rereplacement it was 13% (four of 31), compared with 6.4% (19 of 295) for initial mitral valve replacement. For aortic valve rereplacement, the mortality rate was 15% (four of 27), compared with 5.2% (23 of 440) for initial aortic valve replacement. The cause of death was low cardiac output in six patients, hemorrhage in one patient and sepsis in one patient. All of the patients who died were in functional class IV (p = 0.018). The in-hospital risk for rereplacement of a dysfunctional aortic or mitral valve is no different from that for primary valve replacement unless the patient deteriorates to class IV and requires an emergent operation.

The effect of hemodilution on experimental myocardial infarct size.

Hemodilution is used frequently for patients undergoing elective coronary bypass surgery but relatively little is known about the effects of hemodilution on acute myocardial infarction. Left ventricular infarctions were planimetered after acute coronary occlusion in dogs treated with and without hemodilution. The control group was maintained at a normal hematocrit (40-45) while the intervention group was hemodiluted with 6% dextran in saline 30 min after left anterior descending coronary artery occlusion to a hematocrit less than 50% of control (20). Systemic arterial and left atrial pressure were maintained at control levels as were arterial pO2 and pH. Mean infarct size in the control group was 20.72 +/- 1.38% (SEM) of the left ventricle while the hemodiluted group had a mean infarct size of 18.80 +/- 2.11% of the left ventricle. Reduction of red cell mass to less than 50% of normal does not increase the size of myocardial infarction after acute coronary artery occlusion suggesting that oxygen supply to ischemic areas is maintained if systemic hypoxia is prevented.