High-dose etoposide and melphalan, and autologous bone marrow transplantation for patients with advanced Hodgkin's disease: importance of disease status at transplant.

PURPOSE: To evaluate an intensive therapy regimen of high-dose etoposide and melphalan and autologous bone marrow transplantation (ABMT) in advanced Hodgkin's disease; and to determine possible prognostic factors that predict for long-term disease-free survival (DFS). PATIENTS AND METHODS: Seventy-three patients with advanced Hodgkin's disease who had failed to achieve remission with front-line chemotherapy (n = 16) or who had relapsed (n = 57) were treated with high-dose etoposide 60 mg/kg and melphalan 160 mg/m2 and ABMT. Previous therapy included mechlorethamine, vincristine, procarbazine, and prednisone (MOPP) alternating with doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD), or hybrid MOPP/ABV. All patients received pretransplant cytoreduction with conventional-dose salvage chemotherapy and 40 also received pretransplant extended-field radiation to areas of bulky nodal disease (> 5 cm). RESULTS: Response to high-dose etoposide and melphalan was determined at 3 months post-ABMT. The complete response (CR) rate was 75% (95% confidence interval [CI], 64% to 84%), including 35 of 50 patients with measurable disease before ABMT (70%; 95% CI, 60% to 86%). There were three early deaths (septicemia) and four late deaths (three interstitial pneumonitis, one intracerebral hemorrhage). Actuarial DFS is 38.6% at 4 years. Multivariate regression analysis showed that disease status at the time of ABMT (no evidence of disease [NED], nonbulky residual disease [NBRD], or bulky disease) was the most important factor determining DFS: 68% of those transplanted with NED versus 26% for patients with NBRD and 0% for bulky disease (P = .0002, log-rank test). Relapse in a previous radiation field was the only other significant prognostic factor. CONCLUSION: Etoposide and melphalan is an effective and well-tolerated intensive therapy regimen in advanced Hodgkin's disease. Patients in complete remission after conventional-dose salvage therapy transplanted with this regimen enjoy superior long-term DFS.

Surgical experience with left ventricular outflow tract obstruction in patients with complete transposition of the great arteries and essentially intact ventricular septum undergoing the Mustard operation.

During a 24-year period (1963-1987), 46 infants and children with complete transposition of the great arteries (CTGA) and an essentially intact ventricular septum (IVS) have undergone surgical relief of left ventricular outflow tract obstruction (LVOTO) concurrently with the Mustard operation (MO). The obstruction was valvar in 2 patients, subvalvar in 43 and combined valvar and subvalvar in 1; the ventricular septum was intact in 42, and 4 patients had a small VSD at the time of repair. Neonatal and pre-Mustard cardiac catheterization (CC) data were analyzed to obtain LV to right ventricular peak systolic pressure ratio (LV/RVPSP) and LVOT peak systolic gradient (PSG). The mean neonatal LV/RVPSP (n = 31) was 0.79 +/- 0.04; pre-Mustard LV/RVPSP (n = 44) was 0.75 +/- 0.05; pre-Mustard LVOT PSG (n = 34) was 46.5 +/- 3.5 mmHg. Neonatal (n = 29) and pre-Mustard (n = 30) cardiac angiograms were retrospectively reviewed to analyse morphologic substrates of LVOTO. In the neonatal period, LVOTO was present in 7 patients (anatomic, n = 4; dynamic, n = 2; combined, n = 1). Immediately prior to the MO, LVOTO was present in 29 (anatomic, n = 20; dynamic n = 2; combined, n = 7). Techniques to correct LVOTO at the initial MO included pulmonary valvotomy (n = 3), ventriculomyectomy (n = 41), resection of an windsock aneurysm (n = 1), LV to pulmonary artery valved conduit (VC) (n = 1). Two patients underwent VC early postoperatively (PO) for severe residual LVOTO (both died).(ABSTRACT TRUNCATED AT 250 WORDS)

Relationship between cerebrospinal fluid dynamics and reversible spinal cord ischemia during experimental thoracic aortic occlusion.

This study was designed to assess the effects of hemodynamic changes and cerebrospinal fluid dynamics on spinal cord function during experimental thoracic aortic occlusion. We investigated the effects of dopamine, sodium nitroprusside, and sodium thiopental in this model. Proximal and distal aortic pressures and cerebrospinal fluid pressure were measured during occlusion in 12 adult mongrel dogs under control conditions and during drug interventions. Spinal cord function was assessed by spinal somatosensory evoked potentials recorded during 3-minute intervals of reversible spinal cord ischemia. By multiple regression analysis, the degree of spinal cord ischemia was positively related to the cerebrospinal fluid pressure (p = 0.0092) and negatively related to the percent change in cerebrospinal fluid pressure (p = 0.028); there were no significant drug effects on cerebrospinal fluid pressure or on the degree of spinal cord ischemia. This study indicates that cerebrospinal fluid pressure is an important factor in determining the degree of spinal cord ischemia during aortic occlusion and suggests that measures to reduce cerebrospinal fluid pressure will mitigate the degree of spinal cord ischemia.

Repair of tricuspid atresia: utility of right ventricular incorporation.

During a 10-year period, 62 patients underwent the following modifications of the Fontan operation for repair of tricuspid atresia: direct atriopulmonary connection (N = 15), atriopulmonary connection using a conduit (N = 5), direct atrioventricular (AV) connection (N = 22), and AV connections with a valved conduit (N = 20), including 2 with combined Fontan-arterial switch procedures. The overall hospital mortality was 16.1% (10/62) (70% confidence limits, 11.2 to 22.4%). By multivariate analysis, the risk factors for early and late death included increasing right atrial pressure after repair, use of an atriopulmonary connection, and previous pulmonary artery banding (all variables, p less than 0.05). Postoperative catheterization was performed in 22 patients including 15 with AV valved-conduit connections. Right ventricular (RV) work based on pulmonary artery pressure minus right atrial pressure was correlated with the preoperative RV to left ventricular volume ratio computed from the four-chamber angiographic projection (p = 0.025), and was appreciable only with ratios exceeding about 30%. In 6 of 19 eligible patients, severe conduit obstruction has developed. Considering the survival data, the risk of reoperation, and postoperative hemodynamic findings, analysis of our experience supports the preferential use of nonvalved AV connections in most patients with tricuspid atresia and ventriculoarterial concordance.

Uterine artery flow velocity waveforms in normal and growth-retarded pregnancies.

Uterine artery flow velocity waveforms were measured by continuous-wave Doppler ultrasound in 15 normal pregnant women studied from midgestation until term. Results were analyzed by calculation of the pulsatility index and the systolic/diastolic blood velocity [corrected] ratio. Both indices decreased from 16 to 20 weeks (indicating a lowering of resistance) and thereafter remained stable until term. Resistance was lower when waveforms were recorded directly over the placenta or from the uterine artery close to the placenta. A group of women with severe intrauterine growth retardation were also studied. In preeclampsia, uterine artery resistance was increased in almost all patients. In pregnancies complicated by intrauterine growth retardation of non preeclamptic origin, a wide range of results was obtained.

Loss of the somatosensory evoked response as an indicator of reversible cerebral ischemia during hypothermic, low-flow cardiopulmonary bypass.

We assessed somatosensory evoked response (SSER) as a monitor of cerebral protection during nonpulsatile, hypothermic cardiopulmonary bypass (CPB). In 13 dogs under CPB, extracorporeal flow rate (EFR) thresholds for loss of SSER were determined by stepwise reduction of the EFR from 2.0 to 0.25 L/min/m2 at perfusion temperatures of 35 degrees C, 30 degrees C, 25 degrees C, and 20 degrees C. Testing began at 35 degrees C in Group 1 (N = 6) and at 20 degrees C in Group 2 (N = 7). Immediately on loss of SSER (denoted as a decrease of 80% or more in the amplitude of the somatosensory evoked potentials), EFR was restored to 2.0 L/min/m. Thresholds for loss of SSER ranged between 0.75 and 0.25 L/min/m2. SSER was always restored on return of EFR to 2.0 L/min/m2; thus loss of SSER was a reversible ischemic change. Both groups had similar threshold values at 35 degrees C, but at lower temperatures, Group 1 thresholds were significantly higher than those in Group 2. Since 35 degrees C was the first test temperature for Group 1 but the last for Group 2, EFR reduction at 35 degrees C apparently caused neurophysiological changes (depletion of cortical energy reserves), which diminished subsequent tolerance to ischemia, but EFR reduction at 20 degrees C did not. Our findings show that loss of SSER warns of reversible cerebral ischemia, and support SSER monitoring as a useful measure of cerebral function during low-flow, hypothermic CPB.

Surgical experience with reparative techniques in patients with congenital mitral valvular anomalies.

Our review of an entire institutional experience with primary repair of congenital and acquired mitral valve (MV) anomalies in children with concordant atrial-ventricular-arterial connections but without atrioventricular septal defects included 48 patients (1962 to September 1986). Mitral stenosis (MS) was the predominate lesion in 24 patients, mitral incompetence in 22, and mixed in two. Most patients with MS had so-called typical congenital MS (Van Praagh) with abnormalities of all valvular components; virtual or complete absence of chordal development was a consistent finding. Other anatomic substrates (not mutually exclusive) producing MS included supramitral ring with Shone's complex (n = 5), parachute mitral valve (n = 2), hypoplastic mitral ring (n = 3), and rheumatic lesions (n = 3). The anatomic substrates producing mitral incompetence included annular dilatation (n = 16; isolated in seven), leaflet prolapse (n = 7), cleft leaflet (n = 6), leaflet deficiency (n = 3), and rheumatic lesions (n = 2). Concurrent repair of associated lesions was performed in 29 (60%) patients for relief of left ventricular outflow tract obstruction (n = 16) and closure of ventricular septal defects (n = 9; double-outlet right ventricle in three). Six (12.5%) patients had previous repair of coarctation of the aorta. The operative mortality rate was 18.7% (70% confidence limits [CL]; 12.8% to 24.6%), with one death in 35 patients since 1975 (2.9%; 70% CL; 0% to 5.8%). The actuarial freedom from early or late death and reoperation (+/- SEM) was 44.0% +/- 14.8% at 8 years postoperatively.(ABSTRACT TRUNCATED AT 250 WORDS)

Identification of risk factors for spinal cord ischemia by the use of monitoring of somatosensory evoked potentials during coarctation repair.

The infrequency of spinal cord infarction and paraplegia after occlusion of the descending thoracic aorta has effectively precluded statistical identification of risk factors. Reversible spinal cord ischemia (SCI), however, is more common, can be detected by intraoperative neurophysiologic monitoring, and can lead to irreversible spinal cord damage. Spinal somatosensory evoked potentials (SEPs) were monitored intraoperatively in 38 patients (18 days to 18 years) undergoing coarctation repair (1982-1986). Although no patients sustained perioperative neurologic dysfunction, 10 of 38 (26%) patients developed reversible SCI, as reflected by greater than 75% loss of SEP N1-P1 interpeak amplitude during aortic occlusion (mean clamp time, 29.1 +/- 1.1 min). During occlusion, seven of 38 (18%) sustained complete loss of the SEP; uniform and prompt (1 to 6 min after clamp release) recovery of the signal occurred in these patients with reperfusion following completion of the repair (n = 6), or temporary institution of partial occlusion (n = 1). By multiple regression analysis the degree of SCI was negatively related to the distal aortic pressure (mean 32.4 +/- 2.4 mm Hg, p = .03), and the occlusion PCO2 (mean 33.1 +/- 1.1 mm Hg; p = .013), and positively related to the change in proximal systolic pressure with aortic occlusion (mean 19.8 +/- 3 mm Hg, p = .003). We conclude that: (1) distal hypotension and SCI commonly occur during aortic occlusion for coarctation repair, and (2) intraoperative interventions that can potentially influence distal aortic perfusion and/or PCO2 should be used judiciously.

Surgical experience with the modified Fontan procedure.

During a 10 year period 109 patients (3 months to 47 years old) underwent modifications of the Fontan procedure for repair of classic tricuspid atresia (TA) (n = 58), univentricular atrioventricular connection (UVH) (n = 38), or other complex malformations (CM) (n = 13). Among patients with TA, an atriopulmonary connection was used in 19 (33%) and incorporation of the right ventricle with the Björk modification and with a right atrial-to-right ventricular valved conduit was used in 20 (34%) and in 19 (33%), respectively. Three of the latter 19 also underwent a combined Fontan-switch procedure. The hospital mortality rate was 13.8% (70% confidence limits, 9.3% to 18.3%) for patients with TA, 28.9% (70% confidence limits, 21.3% to 37.0%) for patients with UVH, and 7.7% (70% confidence limits, 0% to 15.4%) for patients with CM. Multivariate analysis identified with the following variables as risk factors for both early and late deaths: diagnosis of UVH, previous pulmonary artery banding (PAB), and postrepair right atrial pressure, and, in patients with TA, the use of a direct atriopulmonary connection (all variables, p less than .05). Morphometric lung biopsy scores were not different in patients with PAB, implicating the role of ventricular hypertrophy rather than pulmonary vascular disease as the mechanism for the adverse effect of PAB. Right atrial pressure was a predictor of serious late cardiac symptoms, which were present in 10% of eligible patients (p = .032). This review demonstrates a survival advantage with modifications of the Fontan procedure that incorporate the hypoplastic right ventricle in patients with TA.(ABSTRACT TRUNCATED AT 250 WORDS)