|
|
||||||||
a Department of Perfusion and Intensive Care, Institut Hospitalier Jacques Cartier, Massy, France
b Department of Pediatric Cardiac Surgery, Institut Hospitalier Jacques Cartier, Massy, France
c Department of Pediatric Pneumology, Institut Hospitalier Jacques Cartier, Massy, France
Accepted for publication August 4, 2008.
* Address correspondence to Dr Durandy, Perfusion and Intensive Care Unit, Institut Hospitalier Jacques Cartier, Avenue du Noyer Lambert, Massy, 91300, France (Email: iciprea{at}icip.org).
| This article has been selected for the open discussion forum on the CTSNet Web Site: http://www.ctsnet.org/sections/newsandviews/discussions/index.html
|
| Abstract |
|---|
|
|
|---|
Methods: This retrospective study included 234 consecutive patients weighing less than 10 kg operated under CPB from August 2006 to November 2007. Patients were divided into two groups: group 1 contained 38 patients with CCT exceeding 90 minutes, and group 2 had 196 patients with shorter CCT. Classic factors were used to analyze outcomes, and outcomes were compared with those from the Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery database.
Results: Results, expressed as values for group 1 vs those for group 2, were mortality rate, 5.3% vs 2%; length of hospital stay exceeding 21 days, 5% vs 0.5%; delayed chest closure, 21% vs 2.6%; epinephrine infusion, 45% vs 11%; organ failure, 13% vs 2%; reoperation due to bleeding, 3% vs 0.5%; heart block, 3% vs 1%; time to extubation, in hours, 64 ± 94 vs 19 ± 48; plasma lactate concentrations after bypass, 2.6 vs 1.9 mmol/L; length of stay in intensive care, in hours, 100 ± 105 vs 52 ± 48.
Conclusions: Despite expected differences between the two groups, our results were within the range of values described in the literature. This led us to conclude that warm pediatric cardiac surgery with a long CCT is safe. A large, multicenter, randomized prospective study comparing normothermic and hypothermic pediatric cardiac surgery is underway.
Although some cardiac surgical pediatric institutions have progressively shifted to moderate hypothermic perfusion [1, 2], classical hypothermic perfusion and circulatory arrest are still advocated [3, 4], and normothermic pediatric cardiac surgery is still under scrutiny in the academic community [5]. However, the adverse effects of hypothermia are well described in the literature, and they counterbalance its protective effects. Compared with normothermia, hypothermia is associated with:
We have previously demonstrated the feasibility of warm open heart surgery in pediatric patients [13], and this technique is currently used in several European institutions. The cumulative experience includes more than 8000 unpublished procedures. However, the safety of warm surgery remains a matter of concern for many cardiac surgeons, especially when a prolonged aortic cross-clamp time (CCT) is needed. Long procedures are often necessary for the surgical treatment of complex congenital abnormalities. In such cases, the risk of suboptimal surgical repair is added to the metabolic risks of prolonged extracorporeal circulation.
The goals of this study were:
| Material and Methods |
|---|
|
|
|---|
A miniaturized bypass circuit was composed of 3/16-inch inner diameter silicon tubing for arterial and venous lines, and connected to the Baby RX oxygenator (Terumo Cardiovascular Tystems, Ann Arbor, MI) or, more recently, to the KidsD100 oxygenator (Sorin Group, Milano, Italy). No arterial filter or hemofilter was added so that the priming volume could be uniformly decreased to 120 mL for patients weighing less than 10 kg.
Patients weighing less than 8 kg had a blood prime composed of packed red blood cells and fresh frozen plasma, whereas patients weighing more than 8 kg received an asanguinous blood prime, with 50% hydroxyl ethyl starch and 50% lactated Ringer's, solution to achieve an estimated hemoglobin level on bypass of at least 8 g/dL. Blood transfusions, whenever necessary, were for the most part performed after discontinuing CPB. The techniques for warm perfusion and intermittent warm blood cardioplegia have been previously described [13–15]. The prime solution was heated to 37°C before the institution of the CPB, and the water heater was set to 37.5°C during the entire bypass period. Full-flow CPB with an index of 2.7 l/min/m2 was used during the entire perfusion time, and the interval of warm blood cardioplegia reinjection was 20 minutes during aortic CCT.
In accordance with the Prophylactic Intravenous Use of Milrinone After Cardiac Operation in Pediatrics (PRIMACORP) study [16], a phosphodiesterase III inhibitor was used prophylactically in neonates and infants to minimize the risk of low cardiac output. Before being weaned from bypass, patients were given a loading dose of 1-mg/kg enoximone followed by a continuous dose of 10 µg/kg/min. When needed, epinephrine was infused at a concentration of 0.05 to 0.2 µg/kg/min.
Patient data from groups 1 and 2 are summarized in Tables 1 and 2,
respectively, which illustrate procedure type, age, weight, cardiopulmonary bypass time, and aortic CCT values.
|
|
Respiratory failure was evaluated using fiberoptic tracheobronchial examination, bronchoalveolar lavage with sampling for cytobacteriologic examination, viral screening and computed tomography scan, whenever necessary.
Data analyses were performed using an unpaired t test for continuous, independent data or a
2 test for categoric data. Differences were considered statistically significant when p < 0.05.
| Results |
|---|
|
|
|---|
|
In group 2, 4 of 196 patients died. Two patients had pulmonary atresia and intact septum with right ventricle hypoplasia and no known right ventricular–dependent coronary circulation. After a one and half attempt repair with bidirectional Glenn, both patients were found to have low cardiac output state, and cardiac arrest occurred within the first postoperative day. A third patient with total pulmonary anomalous venous return experienced a postoperative suprasystemic pulmonary pressure that was unresponsive to optimal therapy. Finally, a fourth patient with a complete AVSD died from a third operation within 6 months, this operation was for a mitral valve replacement for residual mitral insufficiency.
Tables 4, 5 and 6
illustrate additional results. It is noteworthy that only 1 patient in group 1 (2.6%) and 3 patients in group 2 (1.5%) had a prolonged hospital stay (>21 days).
|
|
|
No patients experienced any clinical neurologic complications or renal insufficiency. In this study, organ failure was limited to pulmonary diseases, including airway stenosis and parenchymal abnormalities with or without infection.
The incidence of prolonged ventilation was 18% in group 1 (7 of 38) and 4% in group 2 (8 of 196 patients). The causes of prolonged mechanical ventilation were related to the operation (persistence of a cardiac defect or diaphragmatic paralysis leading to reoperation) or to infections. These causes are listed in Tables 7 and 8.
|
|
| Comment |
|---|
|
|
|---|
We have chosen to enhance the complexity of the cases examined by including only patients weighing less than 10 kg. Because there is no consensus on the definition of a long procedure or long aortic CCT, we arbitrarily chose 90 minutes as a cut off between "long" and "normal" CCT. The choice of 90 minutes was justified because a shorter cutoff time would likely have meant the inclusion of simpler cases and therefore might have masked some of the negative effects of warm surgery.
Preoperative differences in the two groups of patients studied were reflected in postoperative results. Although morbidity and mortality were higher in group 1, the 5.2% mortality rate for group 1 was nonetheless equivalent to or lower than rates reported in the literature [14, 19–21]. Furthermore, the primary complication in the 2 patients who died was critical airway stenosis, which was not related to these patients having undergone warm surgery. The overall rate of prolonged hospitalization stay of 1.3% compares favorably with the 12% incidence reported in the STS-EACTS database [14].
The higher morbidity incidence in group 1 included a longer time to extubation and a longer ICU length of stay; however, further differences between groups were limited. The median duration of mechanical ventilation decreased from 20 hours to 6 hours from group 1 to group 2. The median length of ICU stay varied from 48 hours in group 1 to 45 hours in group 2, which was a noticeable moderate decrease. The incidence of the five morbidity variables measured, including delayed sternal closure, epinephrine infusion, heart block, bleeding, and mechanical ventilation exceeding 4 days, was low and most likely unrelated to warm surgery.
No patients experienced any neurologic complications or renal insufficiency, and we believe that this likely demonstrates a beneficial effect of normothermia. Electroencephalographic monitoring and perioperative magnetic resonance imaging were not used during this study. Neurologic assessments were limited to clinical evaluation by medical staff and parents who recorded temporary or permanent neurologic deficits, and the occurrence of seizures, irritability, or abnormal behavior. Deleterious neurologic effects from deep hypothermia and circulatory arrest have been extensively studied [22–24], and the potential benefits of warm surgery on neurologic function warrant further investigation.
Perfusion adequacy during CPB was mainly assessed by plasma lactate concentration. Plasma lactate levels before and during CPB were equivalent in the two groups. The significant difference between the two groups seen postoperatively was reversible and had disappeared by the time the patients were extubated. Peaking of serum lactate levels is commonly observed during the early postoperative period. Nonetheless, the 2.6-mmol/L median value observed in group 1 was far lower than the published values known to be associated with an increased risk of adverse outcome [25–28].
We defined mechanical ventilation time as prolonged when the duration exceeded 4 days. By that time, 96% of group 2 patients were spontaneously breathing; therefore, in this study, a prolonged time to extubation was uncommon. The two main reasons for prolonged mechanical ventilation were pulmonary disease and persistent cardiac defect. Prolonged CPB time is known to be a risk factor for early pulmonary dysfunction and to be associated with a higher incidence of nosocomial infection [29–30]. In group 1, infections developed in 4 patients with prolonged mechanical ventilation compared with only 1 patient in group 2. The higher infection rate seen with prolonged procedures was probably related to immunomodulation generated by the activation of a systemic inflammatory response. When compared with hypothermia, normothermia showed no obvious inflammatory or systemic adverse effects [1, 2, 31]. The higher incidence of pneumonia in group 1 patients was, therefore, most likely a manifestation of the global increase in risk associated with prolonged procedures rather than a specific side effect of warm surgery.
There has been some concern about lung protection during warm CPB, where the only blood supply to the lungs is the bronchial circulation. We have, in the past, seen acute respiratory distress syndrome and severe postoperative pulmonary bleeding develop in 2 patients after two procedures involving mobilization of the aorta (one arterial switch operation and one interruption of the aortic arch repair) and potential bronchial artery bent. Fiberoptic examinations did not reveal the origin of the bleeding, but they did rule out the presence of a proximal tracheobronchial lesion. Cardiogenic pulmonary edema or perioperative pulmonary wounds could not be excluded. These 2 patients represent the only two cases of hemoptysis in our experience. Therefore, although concern about pulmonary injury is reasonable, we did not observe such complications during this study.
The present study had some limitations, including the relatively small number of patients in group 1 and the absence of transplant procedures or hypoplastic left heart operations due to an extensive policy of antenatal detection and a strategy of compassionate care at our institution. However, this retrospective study demonstrated satisfactory results in both group 1 and 2, without a clear trend toward poorer outcomes in the group with long CCT. Concerns are increasing about the effectiveness of protection provided by hypothermia [9, 32, 33]. The results from this preliminary study are encouraging and warrant continuation of warm pediatric cardiac surgery. In addition, these results strongly suggest that a multicenter, randomized prospective study comparing normothermic and hypothermic pediatric cardiac surgery should be performed.
In conclusion, long aortic CCT during warm surgery remain a major risk factor.
The overall differences between the two groups of patients studied here were limited and within the expected values published in the literature for hypothermic surgery. These preliminary data provide a basis for the implementation of a larger, prospective randomized study, which is, in fact, currently underway.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
F. Lacour-Gayet Invited Commentary Ann. Thorac. Surg., December 1, 2008; 86(6): 1947 - 1947. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |