Outpatient surgery accounts for a significant percentage of anesthetics delivered annually in the United States. Many pediatric procedures, such as hernia repair, circumcision, endoscopy, and heel cord tenotomy, are performed in infants and may occur on an outpatient basis.
Apnea is the most common serious adverse event after general anesthesia in an infant. Premature and former premature infants are at higher risk of apnea than healthy term babies; furthermore, there is little evidence regarding apnea risk in term patients. In addition, infants (younger than 1 year) are at higher risk of intraoperative anesthetic cardiac arrest and other complications and require careful anesthetic management by practitioners with training and ongoing experience in this population.
Apnea of prematurity is found in 50% of premature infants and is almost universal in infants who are 1000 g at birth. Clinically significant apnea in infants is defined as breathing pauses of 20 seconds or pauses of 10 seconds with bradycardia or oxygen desaturation. However, no consensus exists as to what is pathologic in terms of the duration of apnea, degree of change in oxygen saturation, and severity of bradycardia, and the relationship with conditions such as gastroesophageal reflux is unclear.
In the perioperative setting, 1982 brought Steward’s publication of a small series of infants having herniorrhaphy, which showed that preterm infants were more prone to apnea and other airway complications. A larger prospective study of infants having general anesthesia for a variety of procedures found that a much higher proportion of premature infants required postoperative ventilation. The authors postulated that “anesthetics may unmask a defect in ventilatory control of prematurely born infants younger than 41 to 46 weeks conceptual age with preanesthetic history of idiopathic apnea.” Apnea of prematurity and postoperative apnea are primarily central in nature, although a minority of children have an obstructive or mixed pattern.
Overall Risk in Pediatric Anesthesia
Few studies specifically address risk in infants for outpatient surgery. Patel and Hannallah assessed anesthetic complications in a large series of pediatric outpatients and did not note any specific issues in approximately 350 patients younger than 6 months.
Further evaluation of overall risk requires extrapolation from studies of particular patient populations or from adverse outcomes in infants who are not necessarily outpatients. Several studies have demonstrated an increased incidence of complications in infants (younger than 1 year) compared with other pediatric age groups. A prospective survey of 40,240 anesthetics in infants and children from 1978 to 1982 found an overall complication rate of 4.3% in infants compared with 0.5% in children 1 to 14 years of age; the cardiac arrest rate was 1.9% in infants compared with 0.2% in the older patients. Risk increased with increasing American Society of Anesthesiologists (ASA) status and in emergency procedures; the majority of “accidents” in the infant group occurred during the maintenance of anesthesia and were initiated by respiratory events. Analysis of anesthetics conducted in more than 29,000 children from 1982 to 1987 found a high incidence of adverse events in very small infants (younger than 1 month), but patients were more likely to have a higher ASA status or be undergoing major cardiac or intra-abdominal surgery. A large prospective French audit reflecting currently available drugs and monitoring techniques showed that respiratory events accounted for 53% of all intraoperative events and that there remains a higher risk of adverse events in infants compared with older children.
Analysis of closed claims information as published in 1993 showed that pediatric claims were more often related to respiratory events and that the mortality rate was greater than in adults. The complications in pediatric cases were more frequently thought to have been preventable with better monitoring. Analysis of pediatric closed claims from 1990 to 2000 showed a decrease in the proportion of respiratory claims, particularly those for inadequate oxygenation and ventilation, compared with pediatric claims from the earlier period.
The initial observations from the closed claims data led to the creation of the Pediatric Perioperative Cardiac Arrest (POCA) Registry. Basic demographic information from participating institutions was submitted along with case reports of cardiac arrest. Although overall denominator data are available, more specific information such as breakdown of anesthetic agents in all cases or qualifications of the anesthesia caregivers is not. The incidence of cardiac arrest for the institutions studied for the first report (1994 to 1997) was 1.4 per 10,000 anesthetics, with a mortality rate of 26%. Cardiac arrest occurred most often in patients less than 1 year of age and in patients with severe underlying disease. Patients with concurrent diseases and those having emergency surgery were most likely to have fatal outcomes. In patients whose ASA status was 1 or 2, 64% of the cardiac arrests were medication related; two thirds of the medication-related arrests were due to cardiovascular depression from halothane alone or in combination with other drugs. Cases from the POCA registry for the years 1998 to 2004 demonstrated a declining proportion of cardiac arrests related to medications, in parallel with the transition from halothane to sevoflurane in clinical practice.
There is relatively little specific evidence about apnea risk after anesthesia in term infants. Some evidence exists for individual procedures, but it is not generalizable; however, it may help in setting limits for outpatient surgery. Infants with pyloric stenosis require admission because of the need for preoperative fluid resuscitation and the risk of postoperative apnea (related to metabolic abnormalities). Data from 60 full-term neonates and infants undergoing pyloromyotomy showed a significant incidence of apnea (27% preoperatively and 16% postoperatively), and some instances were in patients with normal preoperative pneumograms. Although currently cleft lip repair is not considered appropriate for outpatient surgery because of associated airway concerns, Stephens and colleagues reported a retrospective analysis of 50 neonates (3 to 56 days old; 11 former premature infants of less than 45 weeks’ postconceptual age) having cleft lip repair who had minimal respiratory complications. Ongoing reassessment of practice and refinement of techniques, however, continue to lead to additional procedures being done in a short-stay or day-surgery setting in selected patients: 23-hour admission has been described for otherwise healthy, nonsyndromic patients having primary cleft palate surgery at ages from 6 to 20 months. Large population studies are needed to truly evaluate the risk.
The bulk of evidence regarding apnea risk after anesthesia relates to former premature infants rather than term babies. A number of small case series tried to more accurately define risk; the data from several of these were pooled into a “combined analysis” in 1995 by Coté and colleagues. The combined series contains data from 255 former preterm infants having general anesthesia for inguinal hernia repairs; infants receiving caffeine were excluded. Using a standardized definition of apnea (greater than 15 seconds without bradycardia or less than 15 seconds when accompanied by bradycardia), they looked for associated risk factors to better define the population at risk. Variation was considerable between institutions in the reported incidence, which was thought to be related to differences in monitoring techniques. The combined analysis showed that apnea was strongly and inversely related to both gestational age and postconceptual age and that continuing apnea at home and anemia were also risk factors. No association was found with a number of other historical factors or anesthetic variables, but this may have been due to the relatively small numbers.
The Coté combined analysis does not define a strict cutoff age for all patients but rather defines confidence intervals for the risk of apnea at various combinations of gestational and postconceptual ages. For nonanemic infants free of recovery room apnea, the probability of apnea was not less than 1% until postconceptual age 56 weeks with gestational age 32 weeks or postconceptual age 54 weeks with gestational age 35 weeks. The authors note that individual clinicians must decide on acceptable risk in a given practice setting.
Some question the clinical relevance of apnea detected only by sophisticated monitoring techniques. One group has published a series of 124 former preterm infants, including 67 patients younger than 46 weeks of postconceptual age; those having uncomplicated anesthetics were discharged after an average recovery room stay of 94 minutes with no apparent adverse consequences. One episode of apnea, responsive to stimulation, was noted in an infant on an apnea monitor at home. A retrospective review of respiratory complications in 57 former premature infants having hernia repair noted that all instances of postoperative apnea/bradycardia and laryngospasm occurred within the first 4 hours postoperatively. Caution is urged in generalizing these findings without larger studies to demonstrate the safety of outpatient care in this patient population. A recent “classification and regression tree analysis” identified five factors as predictive of postanesthesia care unit duration of stay after herniorrhaphy in infants: postconceptual age younger than 45 weeks, prematurity, general anesthesia, postoperative opioid administration, and the use of intraoperative regional analgesia.
A prospective randomized trial of caffeine versus placebo for apnea of prematurity in 2006 infants with birth weights of 500 to 1250 g showed that fewer caffeine-treated infants required supplemental oxygen (36% versus 47%) and that treated infants had positive airway pressure discontinued, on average, 1 week earlier. The follow-up phase of the same study showed a modest improvement in survival rate and a modest decrease in the incidence of cerebral palsy and cognitive dysfunction in caffeine-treated very-low-birth-weight infants at 18 months but not at 5 years of age. Economic analysis suggests that caffeine treatment leads to both improved outcomes and a slightly lower cost of care.
Caffeine has been shown to decrease the risk of apnea in former premature infants undergoing general anesthesia, but studies are relatively small. Welborn and colleagues randomly assigned 32 former preterm infants (37 to 44 weeks’ postconceptual age) to receive either 10 mg/kg caffeine or placebo in conjunction with general anesthesia for inguinal hernia repair. No patients in the caffeine group had postoperative bradycardia, prolonged apnea, periodic breathing, or postoperative oxygen saturation less than 90%; 81% of patients in the control group had prolonged apnea at 4 to 6 hours postoperatively. Systematic review of the available studies concluded that evidence supports that caffeine reduces apnea risk but that, because of small numbers and questionable clinical significance of apneic episodes in clinical trials to date, caution should be used in applying these results to routine clinical practice ( Table 68-1 ).
|Study, Year||No. of Trials||No. of Subjects||Intervention||Control||Outcomes|
|Henderson-Smart, 2001||3||78||Caffeine (10 mg/kg in two studies, 5 mg/kg in one)||Placebo||Apnea/bradycardia occurred in fewer treated infants. In two studies, oxygen desaturation was evaluated; fewer episodes occurred in the treatment group.|
Anesthetic Technique and Apnea Risk.
In a prospective comparison by Welborn and colleagues, spinal anesthesia alone had a lower incidence of postoperative apnea and bradycardia in former preterm infants when compared with spinal anesthesia plus sedation or general anesthesia. Other studies have confirmed a lower incidence of oxygen desaturation and bradycardia, although Krane and colleagues did not find a difference in the incidence of central apnea, suggesting that airway obstruction may also play a role in postoperative clinical events. The incidence of apnea after unsupplemented spinal anesthesia in former premature infants is low ; however, cardiopulmonary events occur frequently enough in this population to warrant postoperative observation similar to general anesthesia. A Cochrane review analyzed four small trials comparing spinal with general anesthesia in the repair of inguinal hernia in former preterm infants ( Table 68-2 ). The authors found no significant difference in the proportion of infants having postoperative apnea/bradycardia or oxygen desaturation. Meta-analysis supported a reduction in postoperative apnea in infants having spinal anesthesia without sedation, as well as a borderline significant decrease in the use of postoperative assisted ventilation.