| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Anesthesia, Centre of Head and Orthopedics 4231, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark; and Centre of Cleft Lip-Palate, Rygaards Allé 45, Denmark.
Address correspondence and reprint requests to Jacob Steinmetz, MD, Department of Anesthesia, Centre of Head and Orthopedics 4231, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark. Address e-mail to jacobsteinmetz{at}dadlnet.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: In total, 39 infants 4–6-mo-old were prospectively enrolled and randomized to receive either a combination of propofol and remifentanil (n = 17) or sevoflurane and fentanyl anesthesia (n = 22) for surgical repair of cleft lip-gum-palate. Postoperative observations were blinded. The parents kept a sleep diary for 2 wk before admission and 2 wk after returning home. The diary included information about how many times the infant awoke during the night and was difficult to comfort and the longest duration of continuous sleep during the night.
RESULTS: Longest continuous sleep was significantly longer in the sevoflurane group (median 7.2 h) compared with the propofol-remifentanil group (median 5.1 h, P < 0.05). No other significant difference was found between groups. Sleep pattern was impaired after surgery in both groups compared with that before surgery (P < 0.01), but it was considered by the parents to be back to normal after a median of 10 days, with no significant difference between groups.
CONCLUSION: Postoperative sleep disturbances occur in infants after both propofol-remifentanil and sevoflurane anesthesia. Sevoflurane seems to be associated with less impairment of postoperative sleep than propofol-remifentanil in the first weeks after repair of cleft lip and palate in infants.
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In Denmark, specialist nurses visit all children born with a cleft lip and/or palate from early infancy in order to inform, educate, and support the parents. In severe cases this close contact is maintained all through childhood during which time the child undergoes several operations.
Some years ago, these nurses became aware of an increase in sleep disturbances among infants after surgery for complete cleft lip and palate repair. This observation coincided with both implementation of a new surgical technique and the introduction of sevoflurane instead of halothane.
Some studies have shown that sevoflurane (compared with propofol) increased the risk of emergence agitation in children (7,9), whereas postoperative sleep disturbances in infants have not yet been evaluated.
We hypothesized that sleep disturbances were more common after sevoflurane anesthesia. The present study was designed to compare propofol-remifentanil and sevoflurane anesthesia in infants.
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
All children were given cisatracurium 0.1 mg·kg–1 IV once to facilitate tracheal intubation and thereafter their lungs were ventilated with an oxygen-air mixture. The surgeon injected bupivacaine 0.25% with epinephrine as an infraorbital block and as infiltration of the palate, in total, 0.5–1 mL·kg–1 in all infants. Anesthesia was discontinued after the last suture, and tracheal extubation was performed with the infant spontaneously breathing 100% oxygen, when airway reflexes were restored.
After extubation, the infants were transferred to the postanesthesia care room located in the pediatric ward, where monitoring was continued. Postoperative management was uniform in the two groups and according to our department standards. Postoperative pain relief was provided by paracetamol four times a day and morphine when needed. The specific type of anesthesia was blinded to the postoperative caretakers and the parents. The anesthetic sheet was kept in a closed envelope, deposited in the patient's record and remained unopened until the infant was discharged from hospital 1 wk after surgery.
A self-report questionnaire was designed to assess changes in the infant's sleep pattern and was filled in by the parents for 2 wk until the day of admission, during the 1 wk admission as well as 2 wk after returning home (second and third week after surgery). If the parents did not consider the sleep pattern of the child to be "normalized" by then, they continued the sleep diary until 4 wk after returning home. The diary was filled in every morning and included information about the infant's sleep patterns, such as number of episodes the infant awoke during the night, was difficult to comfort, and the longest continuous sleep in hours during the night. The parents were also asked to consider when they felt that the child's sleep pattern was fully recovered. All parents were blinded to the infant's anesthesia group assignment throughout the observation period. Intraoperative data have been reported in a previous paper (11).
In addition, a control group of infants aged 4 mo with isolated cleft palate was used. These are not operated until the age of 18 mo, and the specialist nurses follow them as closely as our patients. Their parents filled in a similar questionnaire for 28 days, which was divided into two 14-day periods.
The primary end point was the awakening episodes during nights at home after surgery.
Statistical analysis
Demographic data and sleep data are reported as median with 25%–75% percentile, proportions with 95% confidence interval. Groups were compared using Mann-Whitney ranked sum test and paired t-test was used for changes within each group. Data analysis and statistical evaluation was performed using a commercial statistical package (SAS version 9.1, Cary, NC).
We estimated that a sample size of 50 would allow us to detect a difference of three awakening episodes per week between the two groups with a power of 0.80 at the 0.05 significance level. The estimated standard deviation was four awakening episodes per week.
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Sleep data
The parents of six infants (four in the sevoflurane group and two in the TIVA group) did not complete the postoperative sleep registration. Sleep data from one patient (TIVA group) were excluded from analysis, since this infant had a second operation due to hemorrhage (Fig. 1).
Compared with the sevoflurane group, the longest sleep duration was significantly shorter in the TIVA group at home after surgery (Table 1). There was no statistically significant difference between the TIVA and the sevoflurane group in the number of "awakening episodes" or "difficult to comfort episodes" during nights at home after surgery, although these episodes tended to be more common in the TIVA group.
|
In both groups, the evening sleep pattern was impaired at home during the weeks after surgery as compared with the situation before surgery (Table 1). All these changes, except for awakening episodes in the sevoflurane group, were significant (P < 0.01). The sleep pattern was considered by the parents to be back to normal after a median of 10 days, with no significant difference between groups (10 (7–17) vs 10 (6–15) for TIVA and sevoflurane, respectively, P = 0.80).
No difference between groups was found during the first week in the hospital (Table 1).
No significant change in sleep quality between the first and second observation period was seen in the control group (Table 2).
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sleep disorders have been reported in older children, and they seem to be common postoperatively (2,12). Postoperative behavioral changes are common in children, and the Post Hospitalization Behavior Questionnaire (PHBQ) is considered the "gold standard" for assessing these disorders in preschool children (13–15). Nevertheless the PHBQ is not suitable for the patients in our study, since categories such as separation anxiety, aggression against authority and apathy do not apply to infants aged 4–6 mo. Consequently, we had to develop a new questionnaire which attempted to achieve a quantitative assessment of sleep quality. A limitation of the study is that the questionnaire has not been validated. We did, however, include a comparable control group of infants, who did not have operations, and no significant sleep changes were found in the period of observation.
No correction for multiple testing was applied, and a statistically significant difference was seen in only one of three end points between the groups. On the other hand, the general pattern showed the most sleep impairment in the TIVA group, including an increase in episodes of awakening as compared with the preoperative weeks.
Our primary end point was episodes of awakening, and the difference of the median values was 0.7 episodes per day, corresponding to five episodes per week, which was not statistically significant between groups. It became clear during our analyses that the variability in sleep pattern was much larger than expected, and the standard deviation was at least twice as large as estimated. The calculated standard deviation for awakening episodes was approximately two episodes per day (14 per week) as opposed to the estimated value of four per week. The detection of a difference of three episodes per week would require inclusion of at least 500 patients with a power of 80%. Therefore, this variable was not well chosen as an end point.
In our previous study (11), agitation did not appear to be a problem among the same patients; only one infant in each group appeared to be agitated. It has been suggested by Kain et al. (4,5) that preoperative anxiety, emergence delirium, and postoperative maladaptive behaviors are closely related. We cannot exclude the relation, but we did not find a correlation between emergence agitation and postoperative sleep disorders. We did not record preoperative anxiety, so the relations are still open to debate.
One of our most important findings is that postoperative sleep quality is impaired. This finding emphasizes the importance of informing the parents beforehand in order to reduce their concern after surgery. Although both types of anesthesia generated postoperative sleep changes, we found the infants in the sevoflurane group to have a significantly longer continuous sleep at home after surgery. The difference was approximately 2 h, which must be considered clinically important, but no significant difference was found in the two other end points. Therefore, we do not regard our finding as clear-cut enough to recommend sevoflurane anesthesia above TIVA, although the difference deserves attention.
A few methodological issues related to this study have to be addressed. First, the two groups were not equal in size. This was due to a modification in surgical procedures, which challenged our inclusion procedure, and thereby terminated the investigation prematurely. The allocation sequence was computer-generated beforehand, thereby making the size of the two groups unequal. Second, anesthesia in the propofol-remifentanil group was induced with sevoflurane, which, in strict terms, excludes the term "TIVA." The median duration of sevoflurane induction was only 4 min, making the impact of volatile gas on the infants minimal.
One can speculate why sleep quality is impaired in infants several weeks after surgery. It is doubtful that persistent pain is the cause, and it seems unlikely that intraoperative awareness should occur so commonly as to affect the findings in a small study like ours.
The reliability of the parents' observations is unknown, but since the anesthesia was blinded to the parents, we do not believe there was a difference in bias between groups. In addition, we did not find any difference in the control group between the first and second fortnight, suggesting that the observations are reliable.
The purpose of the study was to see whether there was a difference in postoperative sleep patterns between two types of anesthesia. We expected sevoflurane to have a worse impact than propofol-remifentanil, but we found the opposite—that sevoflurane caused longer sleep duration in the postoperative fortnight.
In conclusion, postoperative sleep disturbances appear in infants after both propofol-remifentanil and sevoflurane anesthesia. It seems that anesthesia with sevoflurane compared with propofol-remifentanil may be associated with less impairment of sleep in the first weeks at home after surgery for repair of cleft lip and palate in infants.
|
Footnotes |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
