Society of Cardiovascular Anesthesiologists

Early Extubation After Cardiac Operations In Neonates And Young Infants

Reviewer: Denise Joffe, M.D.

Children's Hospital of Pittsburgh

Philadelphia, PA

Heinle JS, Diaz LK, Fox LS. J Thorac Cardiovasc Surg 1997;114:413-8

Abstract: The recent trend in cardiac surgery is early extubation and this philosophy is extending to the pediatric and now neonatal populations. The driving force seems to be financial; it is felt that patients who are extubated early have shorter ICU and total hospital stays and therefore lower cost of care.

The study was a retrospective review of all patients 90 days of age or younger who had cardiac surgery between November 1, 1995 and November 1, 1996. The study population consisted of 56 patients; 21 patients were < 7 days old, 10 were 8-30 days old, 9 were 31-60 days, and 16 were 61-90 days. Their diagnoses varied and included all the major forms of congenital heart disease (single ventricles, tetralogy of Fallot (TOF), transposition of the great vessels, coarctation, ventricular septal defects (VSD), total anomalous venous return (TAPVR), atrioventricular canals, and a category of "other lesions"). The anesthetic management was standard except that caudal morphine 50-75 mcg/kg was used and the total dose of narcotic was limited to fentanyl 10 mcg/kg. Features of the surgical technique included a single dose of cold crystalloid cardioplegia for myocardial protection, and the use of circulatory arrest in almost all patients. Low-flow bypass was used to minimize the arrest time to less than 30 minutes when possible. Ultrafiltration was used throughout the bypass period.

Patients were extubated if they were hemodynamically stable, and lacked "severe pulmonary dysfunction" or "severe bleeding", and otherwise fulfilled criteria for extubation. The early extubation group consisted of patients extubated in the operating room or within 3 hours of admission to the ICU. Patients not extubated within this time period form the ventilated group. The latter patients were placed on fentanyl and vecuronium infusions. An additional group called the previous year group was also reviewed. It consisted of a similar group of 63 patients who were operated on the previous year. During that time, it was routine to sedate and paralyze all patients after cardiac surgery. Costs related to hopitalization were also reviewed and were calculated as ICU costs, post operative costs (time of arrival in ICU until discharge (D/C) from hospital and total costs (from admission to D/C from hospital).

Results: Features of the surgical technique as discussed above did not differ between groups. 28 of 56 patients (50%) were extubated early (EE). The remaining 28 patients were left intubated for 3.4+/-2.7 days. In the EE group 3 patients or 11% required reintubation. There were no deaths attributed to EE although there was a serious morbidity as a result of delayed recognition of the need to reintubate a patient. Patients were more likely to require post operative ventilation if they had a higher ASA score, longer CPB time (75+/-26 vs 96+/-31 mins, p=0.009), or required inotropic support (3 vs 14, p=0.001). The majority (50% or more) of patients with TOF, TAPVR, VSD and the group with other lesions were in the EE group. Forty-two percent (13), of neonates (<30 days old) were in the EE. All patients requiring reintubation were in the neonatal group, implying a 23% incidence of reintubation in this group. Sixty percent of patients (15), between 2 and 3 months of age were in the EE group. Patients in the EE group had significantly shorter ICU stays (3.3+/-3.9 vs 6.7+/-2.9 days p=0.0006) and postoperative hospital stays (5.9+/-3.3 vs 13.5 +/-9.7 days, p=0.0004) than patients in the ventilated group. Patients in the previous year group had significantly longer intubations and stays than in the EE group as well. Costs were significantly lower in the EE group.

Discussion: The authors have concluded that EE is economically advantageous and can be accomplished safely in many neonates and young infants undergoing repair of congenital heart disease. Based on their data I find it very hard to conclude the same. In the neonatal group, a reintubation rate close to 25% suggests a problem with extubation criteria or a problem with the choice of patient. I suggest the latter. The authors failed to present their own data fairly. Since all patients requiring reintubation were neonates, the data should reflect the unacceptibly high reintubation rate in that group, and not express the number as a percentage of all EE patients. It is not surprising that neonates would not be amenable to "fast-tracking". Neonates have little cardiorespiratory reserve without the addition of a sternotomy, CPB +/- circulatory arrest and pain. I think this study demonstrates just how fragile such patients can be. The Boston group has shown that the postoperative course of some patients after neonatal cardiac surgery is associated with decreases in cardiac output that delay or require an emergency sternotomy in up to 24% of patients. These are critically ill patients and it doesn't make sense to push their reserve to the limit.

The second major flaw of this study is that groups of patients that are not comparable are studied. Neonates and the older newborns should not be compared since the latter catagory is generally a healthier group with very different heart disease than newborns. Also, the assignment of "extubation criteria" implies that patients who are left intubated for any reason are sicker and more likely to require longer postoperative ventilation and ICU stays than the EE group. Any conclusions about how the time of extubation effects their postoperative course is invalid.

This study did demonstrate that it is possible that the total length of postoperative ventilation and ICU and hospital stay can be reduced in some patients outside of the neonatal period. This deserves further examination in view of the small number of patients studied and the fact that their pathophysiology often suggests vigilance is needed. For example, it has been shown that patients exposed to high pulmonary artery pressures and flows beyond 4 weeks have an increased incidence of pulmonary hypertensive crisis. These crisis are best managed in intubated patients. The use of caudal narcotics has become more frequent and probably has a significant role in this anesthetic technique although it is not clear from this study what adjuvants were used to manage pain, how effectively pain was controlled and if any of the reintubations were thought to be secondary to overnarcotization. It is interesting to note that even patients in the early extubation group had an average of 3 day stays in the ICU. This demonstrates that the length of stay does not depend on the exact moment the extubation occurs, but more on the patient's disease and the philosophy with which other aspects of postoperative care is managed.

Bibliography

Anand K, Hickey P. Halothane-morphine compared with high-dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery. NEJM 361:1, 1992

Anand K, Hansen D, Hickey P. Hormonal-metabolic stress responses in neonates undergoing cardiac surgery. Anesthesiology 73:661-670, 1990

Castaneda A, Jonas R, Mayer J, Hanley FL in Cardiac surgery of the neonate and infant Saunders, Philadelphia, 1994

Newsletter