can be attributed to improvements in diagnostic

Transcription

1 Cardiac Intensive Care Early postoperative outcomes in a series of infants with hypoplastic left heart syndrome undergoing stage I palliation operation with either modified Blalock-Taussig shunt or right ventricle to pulmonary artery conduit* Clifford L. Cua, MD; Ravi R. Thiagarajan, MBBS, MPH; Kimberlee Gauvreau, ScD; Lillian Lai, MD; John M. Costello, MD; David L. Wessel, MD; Pedro J. del Nido, MD; John E. Mayer, Jr, MD; Jane W. Newburger, MD; Peter C. Laussen, MBBS Objective: Previous publications using nonconcurrent series of patients indicate improved survival for patients with hypoplastic left heart syndrome (HLHS) undergoing stage I palliation with a right ventricle to pulmonary artery conduit (NW-RVPA) vs. a modified Blalock-Taussig shunt (NW-BT). We compared postoperative outcomes in a concurrent series of patients with HLHS undergoing an NW-BT procedure vs. NW-RVPA procedure. Design: Perioperative data from 66 consecutive patients who underwent NW-BT (n 37) or NW-RVPA (n 29) procedures were retrospectively analyzed. Setting: Cardiac intensive care unit in a tertiary pediatric hospital. Patients: Charts were reviewed for all patients with the diagnosis of HLHS undergoing the NW-BT or NW-RVPA procedure between January 2002 and December Results: Cardiopulmonary bypass time was longer in the NW-BT group than in the NW-RVPA group ( vs mins; p.04). Postoperative diastolic pressures were higher and the PaO 2 to FIO 2 ratio profiles were lower for the NW-RVPA group over the first 72 hrs. Time to sternal closure (2 [1 6] vs. 4 [2 41] days; p.01), duration of mechanical ventilation (113 [49 386] vs. 136 [84 764] hrs; p.01), time to establish enteral feeds (4 [2 8] vs. 5 [3 22] days; p.01), length of intensive care unit stay (11 [7 55] vs. 15 [8 90] days; p.04), and length of hospital stay (16 [11 67] vs. 27 [12 126] days; p.01) were shorter in the NW-RVPA group. Postoperative mortality was not significantly different between the NW-RVPA group (7%) and NW-BT group (11%). Conclusion: At an experienced institution with low stage I palliation mortality for HLHS, there were no differences in early morbidity and mortality between the NW-RVPA and NW-BT procedures. The primary advantage of the NW-RVPA procedure may be faster recovery following surgery and earlier discharge from the hospital. (Pediatr Crit Care Med 2006; 7: ) KEY WORDS: congenital heart disease; Norwood operation; hypoplastic left heart syndrome Survival of infants with hypoplastic left heart syndrome (HLHS) and related singleventricle cardiac defects undergoing stage I palliation operation with a modified Blalock-Taussig shunt (NW- BT) has improved over recent years. This See also p From the Departments of Cardiology (CLC, RRT, KG, LL, JMC, DLW, JWN, PCL) and Cardiac Surgery (PJdN, JEM), Children s Hospital, Boston; and the Departments of Pediatrics (CLC, RRT, KG, LL, JMC, DLW, JWN, PCL) and Surgery (PJdN, JEM), Harvard Medical School, Boston, MA. The authors report no conflict of interest. Presented in part at the Annual Meeting of the American Heart Association, Copyright 2006 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies DOI: /01.PCC can be attributed to improvements in diagnostic techniques; preoperative, intraoperative, and postoperative care; and better understanding of the risks factors associated with mortality among these patients (1 7). Despite an overall decrease in mortality with the stage I palliation operation, early mortality with this procedure varies considerably between institutions (1, 4, 6, 8, 9). A recent modification of the stage I palliation procedure, using a right ventricle to pulmonary artery conduit (NW- RVPA) to provide pulmonary blood flow rather than the modified Blalock-Taussig shunt, has been reported by several congenital heart centers to improve early mortality among patients with singleventricle defects undergoing stage I palliation (10 16). The improved mortality among patients undergoing the NW- RVPA operation has been explained in part by the absence of diastolic blood flow from the systemic circulation into the pulmonary circulation, resulting in better coronary and end-organ perfusion (10 12, 14, 15, 17). Whereas reports of diminished early mortality with the NW-RVPA procedure are compelling, the studies have not been randomized and have relied on historical controls. Although the improved early mortality has been ascribed to the use of the RV-PA conduit, the results of these studies are confounded by the use of historical controls, which may have been subjected to different perioperative management strategies, and higher mortality in an earlier era. We hypothesized that patients with HLHS undergoing the NW- RVPA procedure would have better survival and outcomes than those undergo- 238 Pediatr Crit Care Med 2006 Vol. 7, No. 3

2 ing the NW-BT procedure in a contemporaneous time period. Our study population comprised infants with HLHS undergoing either the NW-BT or NW- RVPA procedure at a single institution during the same time period and exposed to similar operative management strategies. MATERIALS AND METHODS Subjects. We identified all consecutive patients with single-ventricle physiology who underwent stage I palliation with either NW-BT or NW-RVPA procedure over a 2-yr period between January 2002 and December For purposes of inclusion in this study, we selected patients with a diagnosis of HLHS, defined as aortic atresia/mitral atresia, aortic atresia/mitral stenosis, aortic stenosis/mitral stenosis, or aortic stenosis/mitral atresia. Patients with single left ventricle, unbalanced atrioventricular canal, double-outlet right ventricle and hypoplastic left ventricle, or tricuspid atresia who underwent the stage I palliation were excluded. Data collection and chart review for this study was approved by the Investigational Review Board of Children s Hospital, Boston. Surgical and Postoperative Care. During the time period of this study, the NW-BT or NW-RVPA procedure was performed, in large part, according to the preference of the surgeon and/or cardiologist rather than specific anatomical or physiologic determinants. The surgical techniques for the two procedures have been previously published (10 15, 17). After division of the main pulmonary artery, all patients underwent identical neo-aorta reconstruction, homograft augmentation of the aortic arch, and atrial septectomy. As sources of pulmonary blood flow, patients undergoing the NW-BT had a modified Blalock-Taussig shunt constructed between the innominate artery and the right pulmonary artery, and those undergoing the NW-RVPA had a conduit placed from the infundibulum of the right ventricle to the pulmonary artery confluence. All patients received perioperative care in the cardiac intensive care unit at Children s Hospital, Boston. Their care was managed according to established preoperative and postoperative practices for patients with HLHS admitted to our institution. Data Collection. Data were collected through retrospective chart review. Demographic data were recorded, including gender; weight and age at time of surgery; prenatal diagnosis; findings on postnatal echocardiogram (e.g., size of the ascending aorta, status of the interatrial septum, ventricular function, and degree of atrioventricular regurgitation); and preoperative events such as endotracheal intubation, inotropic support, need for cardiopulmonary resuscitation (defined as chest compressions or bolus-dose epinephrine administration), arterial ph 7.30, documented infection, renal insufficiency (defined as creatinine level 1.5 mg/dl), hepatic insufficiency (defined as alanine or aspartate aminotransferase level 500 IU/L), and interventions such as balloon dilation of the atrial septum in the catheterization laboratory. Intraoperative data included type of procedure; size of shunt or conduit; total cardiopulmonary bypass time, cross-clamp time, and circulatory arrest time; precardiopulmonary bypass lactate and hematocrit; and postcardiopulmonary bypass lactate levels. In the immediate postoperative period, defined as the first 72 hrs following transfer from the operating room to the cardiac intensive care unit, data on hemodynamic and respiratory variables were recorded at 6-hr intervals. These data were then averaged at 12-hr intervals for statistical analysis. Hemodynamic variables included heart rate; mean systolic and diastolic pressures; pulse width; volume replacement; and inotrope score [(dobutamine ( g/ kg/min) dopamine ( g/kg/min) epinephrine ( g/kg/min) 100 milrinone ( g/kg/min) 10)] (18). Respiratory variables included FIO 2, mean airway pressure (mpaw), arterial blood ph, PaO 2 and PaCO 2, and the ratio of PaO 2 to FIO 2 (P/F ratio). Other postoperative data collected included daily urine output, fluid balance, days till first negative fluid balance, daily chest drain losses, blood product and colloid use, and maximum daily lactate levels. Postoperative outcome data collected included length of intensive care and hospital stay; duration of mechanical ventilation; presence of an open sternum and time to sternal closure; presence of renal or liver insufficiency; time to establish enteral nutrition; need for peritoneal dialysis catheter use; occurrence of neurologic events (seizures); need for extracorporeal membrane oxygenation support; and death before discharge from the hospital. Statistics. The two operative groups were compared with respect to demographic, preoperative, intraoperative, postoperative, and outcome data. The Shapiro-Wilk test was used to assess the distribution of continuous data. Normally distributed continuous data were summarized as mean SD and compared with Student s t-test, whereas nonnormal data were compared with the Mann-Whitney U test. Categorical data were analyzed with the chisquared test, and Fisher s exact test was used where expected counts in at least one category were 5. Changes in hemodynamic and respiratory function data over time (first 72 postoperative hrs) for the two operation groups were compared by repeated-measures analysis of variance. In a multivariable logistic regression model, the association of type of operation (NW-RVPA and NW-BT) and mortality was sought after adjusting for factors previously known to influence mortality and preoperative and intraoperative variables that were significantly different between the two operative groups. Statistical analysis was performed with statistical software (SPSS 12.0, SPSS, Chicago, IL). RESULTS Demographics and preoperative variables. Sixty-six patients with HLHS underwent stage I palliation operation during the study period, 30 during the year 2002 and 36 during the year 2003 (Fig. 1). Of these, 37 (56%) underwent the NW-BT and 29 (44%) underwent the NW-RVPA procedure. The operative groups in most respects had similar demographic and preoperative variables (Table 1). In comparison with the NW- RVPA group, more patients in the NW-BT group had moderate to severe preoperative atrioventricular valve regurgitation (22% vs. 0%; p.007) and at least one preoperative ph value 7.3 (49% vs. 24%; p.04). Figure 1. Procedures performed each year during the study period: Stage I palliation with a right ventricle to pulmonary artery conduit (NW-RVPA) vs. a modified Blalock-Taussig shunt (NW-BTS). Pediatr Crit Care Med 2006 Vol. 7, No

3 Table 1. Demographic and preoperative details (n 66) Variable NW-BT NW-RVPA p Value Patients 37 (56%) 29 (44%) Male 18 (49%) 18 (62%).28 Age at surgery, days 4 (2 17) 6 (1 36).99 Weight at surgery, kg 3.1 ( ) 3.2 ( ).13 Prenatal diagnosis 18 (49%) 18 (62%).28 Ventricular dysfunction a 2 (5%) 3 (10%).65 b AVVR a 8 (22%) 0 (0%).007 b CPR 5 (14%) 2 (7%).45 b Lowest arterial ph (49%) 7 (24%).04 Mechanical ventilation 23 (62%) 19 (66%).78 Inotrope support 22 (59%) 15 (52%).53 Preoperative intervention 7 (19%) 3 (10%).49 b Intervention type BAS 6 1 BAS ECMO 1 0 Fetal aortic valve dilation 0 2 NW-BT, Norwood-Blalock Taussig shunt; NW-RVPA, Norwood right ventricle to pulmonary artery conduit; AVVR, atrioventricular valve regurgitation; CPR, cardiopulmonary resuscitation; BAS, balloon atrial septostomy; ECMO, extracorporeal membrane oxygenation. a Patients with moderate to severe ventricular dysfunction and atrioventricular valve regurgitation; b Fisher s exact test. Data are number (%) of patients or median (range). Table 2. Anatomical and intraoperative support details Variable NW-BT NW-RVPA p Value HLHS type.7 AA/MA 10 (27%) 9 (31%) AA/MS 10 (27%) 5 (17%) AS/MA 4 (11%) 2 (7%) AS/MS 13 (35%) 13 (45%) Ascending aorta size, cm, 0.35 ( ) 0.34 ( ).85 a median (range) CPB time, min a CA time, min a Hematocrit post-cpb, % Lactate pre-cpb, mmol/l a Lactate post-cpb, mmol/l a NW-BT, Norwood-Blalock Taussig shunt; NW-RVPA, Norwood right ventricle to pulmonary artery conduit; HLHS, hypoplastic left heart syndrome; AA, aortic atresia; MA, mitral atresia; MS, mitral stenosis; AS, aortic stenosis; CPB, cardiopulmonary bypass; CA, circulatory arrest. a Mann-Whitney test. Data are number (%) of patients or mean SD, except as noted. Intraoperative Variables. Anatomical details are shown in Table 2. Most patients undergoing the NW-BT procedure received a 3.5-mm shunt (n 35; 95%); a 3-mm shunt was placed in two patients weighing 2.5 kg. In the NW-BT group, the proximal end of the shunt was anastamosed to the innominate artery, and no patient had an anomalous right subclavian artery. A 5-mm right ventricle to pulmonary artery conduit (n 26; 90%) was most commonly used in patients undergoing the NW-RVPA operation; smaller, 4-mm conduit was used in three patients weighing 2.5 kg. Intraoperative support time, indices of intraoperative tissue perfusion, and hematocrit values following cessation of cardiopulmonary bypass for the two operative groups are shown (Table 2). Hemodynamic and Respiratory Variables. Changes in heart rate, systolic blood pressure, inotrope score, PaCO 2, and mpaw were not significantly different between the two operative groups (Table 3). Patients undergoing the NW-BT procedure had lower diastolic blood pressures and the diastolic blood pressure decreased more significantly over the first 72 postoperative hrs than in the NW- RVPA group (Fig. 2). The P/F ratio was lower for the NW-RVPA group and did not change over time, whereas in the NW-BT group the P/F ratio increased over the time period (Fig. 3). A histogram depicting the spread and variability of peripheral oxygen saturation over the first 72 hrs following surgery for the two operative groups is shown in Figure 4. Postoperative Course and Outcomes. Postoperative variables indicating the cardiac intensive care unit course during the first 72 hrs after surgery and outcome variables for the two operative groups are shown in Table 4. The two operative Table 3. Postoperative hemodynamic and respiratory variables during the first 72 hours after surgery Variable Operation Baseline 12-hr 24-hr 36-hr 48-hr 60-hr 72-hr p Value a Heart rate, beats/min NW-BT NW-RVPA Systolic BP, mm Hg NW-BT NW-RVPA Inotrope score NW-BT NW-RVPA PaCO 2,mmHg NW-BT NW-RVPA mpaw, cm H 2 O NW-BT NW-RVPA NW-BT, Norwood-Blalock Taussig shunt; NW-RVPA, Norwood-right ventricle to pulmonary artery conduit; BP, blood pressure; mpaw, mean airway pressure. a p value for interaction of group and time. Data shown do not include those for the patient undergoing extracorporeal membrane oxygenation during the first 72 postoperative hrs. 240 Pediatr Crit Care Med 2006 Vol. 7, No. 3

4 Figure 2. Changes in mean diastolic blood pressure during the first 72 hrs after postoperative admission in the two groups: Stage I palliation with a right ventricle to pulmonary artery conduit (NW-RVPA) vs. a modified Blalock-Taussig shunt (NW-BT). Error bars represent 95% confidence intervals. Figure 3. Changes in mean PaO 2 /FIO 2 (P/F) ratio during the first 72 hrs after postoperative admission in the two groups: Stage I palliation with a right ventricle to pulmonary artery conduit (NW-RVPA) vs. a modified Blalock-Taussig shunt (NW-BT). Error bars represent 95% confidence intervals. groups did not differ with regard to the proportion of patients admitted to the cardiac intensive care unit with an open chest. Daily fluid intake, colloid and Pediatr Crit Care Med 2006 Vol. 7, No. 3 blood products used for volume expansion, and chest drain output during the first 3 postoperative days did not differ between the two groups. Initial serum lactate levels in the cardiac intensive care unit following surgery were not significantly different between the two groups; however, lactate levels on postoperative day2(2 1 vs. 4 3; p.02) and postoperative day 3 (1 1 vs. 2 1; p.004) were significantly lower in the NW- RVPA vs. the NW-BT group. Although urine output on postoperative day 2 was higher in the NW-RVPA group, days to negative balance was not significantly different for the two groups. Seven patients (19%) in the NW-BT group and two patients (7%) in the NW- RVPA group were placed on extracorporeal membrane oxygenation (p.28; Table 4). Extracorporeal membrane oxygenation was initiated within 72 postoperative hrs for five patients in the NW-BT group and for one patient in the NW-RVPA group. Liver insufficiency was not seen in any patient in this cohort. There were no differences in the incidence of renal insufficiency, need for peritoneal dialysis, nosocomial infections, and neurologic complications in the cardiac intensive care unit between the two operative groups (Table 4). Among patients admitted to the cardiac intensive care unit with an open sternum, those undergoing the NW- RVPA procedure had sternal closure sooner than in the NW-BT group (2 [1 6] vs. 4 [2 41] days; p.01) (Table 4). The median duration of mechanical ventilation (113 [49 386] vs. 136 [84 764] hrs; p.01) and time to establish enteral feeds (4 [2 8] vs. 5 [3 22] days; p.01) was significantly shorter in the NW-RVPA group than the NW-BT group. Furthermore, median length of cardiac intensive care unit stay (11 [7 55] vs. 15 [8 90] days; p.04) and length of hospital stay (16 [11 67] vs. 27 [12 126] days; p.01) were significantly shorter in the NW- RVPA than the NW-BT group. The overall mortality rate for this group of patients undergoing stage I palliation for HLHS was low (9.1%). Unadjusted mortality was not significantly different for the NW-RVPA (7%) and the NW-BT groups (11%; p.67). In an adjusted multivariable logistic regression model controlling for patient weight; size of ascending aorta; presence of preoperative atrioventricular valve regurgitation; and need for balloon atrial septostomy, duration of CPB and preoperative ph 7.3 and type of operation (NW-BT and NW-RVPA) was not significantly associ- 241

5 Figure 4. Histogram depicting the average peripheral oxygen saturation for each patient during the first 72 hrs following stage I palliation with either a right ventricle to pulmonary artery conduit (NW-RVPA) or a modified Blalock-Taussig shunt (NW-BT). Five patients in the NW-BT group were not included in the histogram because they were undergoing extracorporeal membrane oxygenation during this period. ated with mortality (odds ratio for mortality after NW-BT vs. NW-RVPA, 1.2; 95% confidence interval, ; p.96). Three patients who underwent the NW-BT procedure died after discharge from the hospital, whereas none in the NW-RVPA group died in the interstage period while awaiting the bidirectional Glenn operation. The median time to bidirectional Glenn operation was significantly shorter in the NW-RVPA group than in the NW-BT group because of early onset of cyanosis (5 1 vs. 6 2 months; p.01). DISCUSSION The use of a right ventricle to pulmonary artery conduit during stage I palliation for hypoplastic left heart syndrome has been accepted in many centers in an effort to reduce postoperative morbidity and mortality. In our concurrent, uncontrolled, nonrandomized series of infants with hypoplastic left heart syndrome at a single institution undergoing stage I palliation, there was no difference in mortality between the NW-BT shunt and NW- RVPA conduit groups. The NW-RVPA infants required less time to establish enteral feeding and had a shorter duration of open sternotomy, mechanical ventilation, and cardiac intensive care unit and hospital stay, suggesting that the NW- RVPA operation is followed by a faster postoperative recovery. The potential physiologic advantages of the NW-RVPA conduit reported by several authors include higher postoperative diastolic blood pressure, smaller pulse pressure, and by inference, improved coronary blood flow and systemic perfusion (10 17). These hemodynamic improvements have in turn been ascribed as the primary reason for the reported lower early mortality in some series and the simplified early postoperative course. Our analysis also demonstrated a higher postoperative diastolic blood pressure, maintained through the first 72 postoperative hrs, in those undergoing the NW-RVPA procedure vs. the NW-BT procedure, yet there was no difference in mortality. There is wide variation in the recently reported mortality rates associated with the NW-BT, ranging from 10% to 40% (1, 2, 4 6, 8, 19, 20). The geometry of the BT shunt and reconstruction of the systemic outflow without compromising coronary blood flow are important surgical factors that could impact outcome. Centers that initially reported improved postoperative hemodynamics and lower mortality after the NW-RVPA procedure had predominantly used a 4.0-mm BT shunt as part of the Norwood operation before changing to the NW-RVPA (12, 13, 17). This is an important distinguishing factor because in our series of patients undergoing the NW-BT procedure, a 3.5-mm BT shunt was predominantly used. The balance between systemic perfusion and pulmonary blood flow was not measured in our patients, and although this is speculative, the smaller shunt size in our NW-BT group could have been associated with less blood flow into the pulmonary circulation during diastole and preservation of end-organ and coronary perfusion. Studies are necessary to fully determine the coronary flow reserve and autoregulation after the stage I palliation relative to diastolic blood pressure. In earlier reports, the outcome following NW-RVPA was compared with that for historical controls of NW-BT patients. Patients undergoing the NW-RVPA or NW-BT procedure in our contemporary, nonrandomized series had mostly similar demographic variables and preoperative and intraoperative risk factors and thus were a comparable group of patients. 242 Pediatr Crit Care Med 2006 Vol. 7, No. 3

6 Table 4. Intensive care unit course for the first 72-post-operative hours for the two operative groups Variable NW-BT NW-RVPA p value Open sternum (n, %) 34 (92%) 23 (79%) 0.17 Total fluids day 1 (ml/kg/day) Total fluids day 2 (ml/kg/day) a Total fluids day 3 (ml/kg/day) a Colloid volume day 1 (ml) Colloid volume day 2 (ml) a Colloid volume day 3 (ml) PRBC volume day 1 (ml) PRBC volume day 2 (ml) a Chest drain output day 1 (ml/kg) Chest drain output day 2 (ml/kg) a Urine output day 1 (ml/kg/hr) Urine output day 2 (ml/kg/hr) Urine output day 3 (ml/kg/hr) Maximum lactate day 1 (mmol/l) Maximum lactate day 2 (mmol/l) a Maximum lactate day 3 (mmol/l) a ECMO use 7 (19%) 2 (7%) 0.28 b Renal insufficiency 3 (8%) 0 (0%) 0.25 b Peritoneal catheter use 3 (8%) 0 (0%) 0.25 b Nosocomial infection 9 (24%) 11 (38%) 0.23 Neurologic complications 2 (5%) 1 (3%) 1.0 b Days till negative balance c 3 (1 5) 2 (1 9) 0.21 a Days to sternal closure c 4 (2 41) 2 (1 6) 0.01 a Days till enteral feeding c 5 (3 22) 4 (2 8) 0.01 a Duration of ventilation (hr) c 136 (84 764) 113 (49 386) 0.01 a Duration of ICU stay (days) c 15 (8 90) 11 (7 55) 0.04 a Duration of hospital stay (days) c 27 (12 126) 16 (11 67) 0.01 a Mortality prior to discharge 4 (11%) 2 (7%) 0.67 Mortality post-discharge prior to 3 (8%) 0 (0%) 0.25 BDG (interstage death) Age at BDG (mo) ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; BDG, bi-directional Glenn operation; patients placed on ECMO during the first 72 postoperative hours excluded; PRBC, packed red cells. a Mann-Whitney test; b Fisher s exact test; c data shown as median (range). Comparing the NW-RVPA to a contemporary NW-BT group serves to minimize variability in surgical and postoperative management practices that could contribute to differences in outcomes over different time periods. Postoperative morbidity was similar with the two procedures, with no differences in the incidence of complications, end-organ injury, or extracorporeal membrane oxygenation use. Although cardiac output was not measured directly, other indices that provided an indirect measure of the adequacy of systemic perfusion were assessed (Table 4). There were no differences in the intraoperative and postoperative day 1 serum lactate levels, and although the serum lactate level was significantly lower at postoperative days 2 and 3 in the NW-RVPA group, the mean lactate level was within the normal range for both groups by postoperative day 3. There were no differences in the time to achieve a negative fluid balance or biochemical indices of renal or hepatic insufficiency for the two groups. Pediatr Crit Care Med 2006 Vol. 7, No. 3 In some reports, patients who underwent the NW-RVPA were described as having favorable postoperative hemodynamics and a trend toward fewer ventilator manipulations, and these findings were used to suggest that the NW-RVPA procedure was associated with a more stable and possibly easier postoperative course (21). In our series, this postoperative stability could be reflected by shorter time to sternal closure, fewer days on mechanical ventilation, and shorter stay in the cardiac intensive care unit in the NW-RVPA group; in addition, although this was not statistically significant, the inotrope score was lower in the NW-RVPA group. However, in our experience, considerations for the postoperative management of care for patients undergoing the NW- RVPA procedure are different and complex on the basis of their underlying physiology. For example, the NW-RVPA group had significantly lower P/F ratios than the NW-BT patients. There also was greater variability in oxygenation levels following the NW-RVPA procedure. The wider spread or variability in peripheral oxygen saturation in the NW-RVPA group possibly reflects variable pulmonary blood flow across the conduit, and the lower P/F ratios may indicate a lower net pulmonary blood flow in the NW-RVPA than in the NW-BT group. Indeed, NW- RVPA patients in our series presented at a younger age for second-stage palliation with the bidirectional Glenn procedure (5 1 vs. 6 2 months; p.01) (Table 4). If pulmonary blood flow is in fact diminished in the NW-RVPA group, there may be an adverse effect on pulmonary artery growth. The size and length of the conduit, the size of the ventriculotomy, and amount of resection of ventricular muscle at the proximal take-off of the conduit in the NW-RVPA procedure may influence pulmonary blood flow. A conduit 5 mmin size or additional muscle resection of the right ventricle may have resulted in a higher P/F ratio in our patients. Other institutions have described their experience with a 6-mm conduit, reporting no significant difference in oxygen saturations in the NW-RVPA and NW-BT groups (17, 18). Follow-up data are necessary to determine whether the size of the ventriculotomy affects long-term right ventricular function. The reduced hospital length of stay in the NW-RVPA group is an important difference, and although we did not calculate costs in our study, earlier hospital discharge is likely to be advantageous. Although this is speculative, the shorter time to establish enteral feeding in the NW- RVPA group may be a factor in reducing the number of days in the hospital and may be due to improved splanchnic perfusion because of the higher diastolic pressure. We use an algorithm to advance feeding in our cardiac intensive care unit for all neonates and infants following cardiac surgery, and therefore variable feeding practices are less likely to contribute to this difference. There are several limitations of this study, including its retrospective nature and lack of randomization. There were no set criteria as to which patient would undergo the NW-BT or NW-RVPA procedure, with the choice made by the primary cardiologist and surgeon. Although there were five cardiac surgeons at our institution over this study period, three surgeons performed 90% of the procedures, one preferring the NW-BT and the other two the NW-RVPA. Thus, the results may be confounded by surgeon preference. Further- 243

7 more, since there were no standardized postoperative management protocols for patients in this cohort, they were subject to interphysician practice variability. Preoperative variables such as atrioventricular valve regurgitation and a lower ph may have influenced the decision as to which surgical procedure was performed. We also cannot rule out the possibility that the increased incidence of moderate to severe atrioventricular valve regurgitation in the NW-BT group contributed to the longer hospital stay. In addition, given the small sample size and low overall mortality rate, the study may have been underpowered to detect a difference between groups. In conclusion, in this single-center retrospective, nonrandomized series, mortality was not improved by the NW- RVPA operation vs. the NW-BT operation for patients undergoing stage I palliation for hypoplastic left heart syndrome. In institutions where mortality with the NW-BT procedure is already low, there may be no survival advantage in performing the NW-RVPA procedure. The NW- RVPA procedure appears to have the advantages of shorter time to extubation, early enteral nutrition, and shorter length of cardiac intensive care unit and hospital stay. The lower and variable P/F ratio in the NW-RVPA could reflect lower pulmonary blood flow, which may have consequences for the timing of subsequent procedures and pulmonary vascular development. A longitudinal, multicentered, randomized, controlled trial with a larger sample size is currently being conducted to determine whether the NW-RVPA is a preferable source of pulmonary blood flow in comparison with the NW-BT procedure for patients with HLHS undergoing single-ventricle palliation. REFERENCES 1. Forbess JM, Cook N, Roth SJ, et al: Ten-year institutional experience with palliative surgery for hypoplastic left heart syndrome: Risk factors related to stage I mortality. Circulation 1995; 92: Jacobs ML, Rychik J, Murphy JD, et al: Results of Norwood s operation for lesions other than hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 1995; 110: ; discussion, Bove EL: Current status of staged reconstruction for hypoplastic left heart syndrome. Pediatr Cardiol 1998; 19: Mahle WT, Spray TL, Wernovsky G, et al: Survival after reconstructive surgery for hypoplastic left heart syndrome: a 15-year experience from a single institution. Circulation 2000; 102: Azakie T, Merklinger SL, McCrindle BW, et al: Evolving strategies and improving outcomes of the modified Norwood procedure: A 10-year single-institution experience. Ann Thorac Surg 2001; 72: Tweddell JS, Hoffman GM, Mussatto KA, et al: Improved survival of patients undergoing palliation of hypoplastic left heart syndrome: Lessons learned from 115 consecutive patients. Circulation 2002; 106: Mosca RS, Bove EL, Crowley DC, et al: Hemodynamic characteristics of neonates following first stage palliation for hypoplastic left heart syndrome. Circulation 1995; 92: Daebritz SH, Nollert GD, Zurakowski D, et al: Results of Norwood stage I operation: Comparison of hypoplastic left heart syndrome with other malformations. J Thorac Cardiovasc Surg 2000; 119: Ashburn DA, McCrindle BW, Tchervenkov CI, et al: Outcomes after the Norwood operation in neonates with critical aortic stenosis or aortic valve atresia. J Thorac Cardiovasc Surg 2003; 125: Mahle WT, Cuadrado AR, Tam VK: Early experience with a modified Norwood procedure using right ventricle to pulmonary artery conduit. Ann Thorac Surg 2003; 76:1084 8; discussion, Mair R, Tulzer G, Sames E, et al: Right ventricular to pulmonary artery conduit instead of modified Blalock-Taussig shunt improves postoperative hemodynamics in newborns after the Norwood operation. J Thorac Cardiovasc Surg 2003; 126: Malec E, Januszewska K, Kolcz J, et al: Right ventricle-to-pulmonary artery shunt versus modified Blalock-Taussig shunt in the Norwood procedure for hypoplastic left heart syndrome: Influence on early and late haemodynamic status. Eur J Cardiothorac Surg 2003; 23:728 33; discussion, Pizarro C, Malec E, Maher KO, et al: Right ventricle to pulmonary artery conduit improves outcome after stage I Norwood for hypoplastic left heart syndrome. Circulation 2003; 108(Suppl 1): Pizarro C, Norwood WI: Right ventricle to pulmonary artery conduit has a favorable impact on postoperative physiology after stage I Norwood: Preliminary results. Eur J Cardiothorac Surg 2003; 23: Sano S, Ishino K, Kawada M, et al: Right ventricle-pulmonary artery shunt in firststage palliation of hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 2003; 126: ; discussion, Hughes ML, Shekerdemian LS, Brizard CP, et al: Improved early ventricular performance with a right ventricle to pulmonary artery conduit in stage 1 palliation for hypoplastic left heart syndrome: Evidence from strain Doppler echocardiography. Heart 2004; 90: Maher KO, Pizarro C, Gidding SS, et al: Hemodynamic profile after the Norwood procedure with right ventricle to pulmonary artery conduit. Circulation 2003; 108: Wernovsky G, Wypij D, Jonas RA, et al: Postoperative course and hemodynamic profile after the arterial switch operation in neonates and infants: A comparison of low-flow cardiopulmonary bypass and circulatory arrest. Circulation 1995; 92: Jonas RA, Hansen DD, Cook N, et al: Anatomic subtype and survival after reconstructive operation for hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 1994; 107: ; discussion, Kern JH, Hayes CJ, Michler RE, et al: Survival and risk factor analysis for the Norwood procedure for hypoplastic left heart syndrome. Am J Cardiol 1997; 80: Azakie A, Martinez D, Sapru A, et al: Impact of right ventricle to pulmonary artery conduit on outcome of the modified Norwood procedure. Ann Thorac Surg 2004; 77: Pediatr Crit Care Med 2006 Vol. 7, No. 3