Patients

All pediatric patients scheduled for elective cleft lip and palate surgical repair at Dubrava University Hospital in Zagreb between January 2002 and May 2006 were considered eligible for the study. Those with contraindications for the surgery, infection, anemia, asthma, severe heart defects, or mental retardation, were excluded from the study ( Fig. 1 ). Of 150 consecutive patients in the study period, 140 were included after obtaining written parental consent and study approval from the Ethics Committee of Dubrava University Hospital. The study sample consisted of 81 boys and 59 girls with normal cardiovascular and pulmonary status, hematological parameters in the normal range, and American Society of Anesthesiologists (ASA) score I/II. The children were aged between 3 months and 10 years ( Table 1 ). Body weight ranged from 5.5 kg to 26 kg.

Setting up the study.

Method

All children were premedicated with midazolam 0.05 mg/kg (Dormicum®, F. Hoffman-La Roche Ltd., Basel, Switzerland) administered orally 30 min before being transferred to an operating suite. On arrival at the operating suite, all children were immediately pre-induced with sevoflurane (5–8 vol%) (Sevorane ^® , Abbott Laboratories S.A., Abbott Park, IL, USA) administered via face-mask breathing spontaneously to make the cannulation of a peripheral vein more convenient for the children and the staff. Once intravenous access was established, sevoflurane was switched off. After a few minutes the monitored minimal alveolar concentration of sevoflurane was zero.

The children were randomly assigned to receive either midazolam–fentanyl or sevoflurane–fentanyl (study group) anesthesia. Randomization was performed using a computer-generated random number list. A nurse anesthetist informed the anesthesiologist of the type of anesthesia immediately before administration.

In the midazolam–fentanyl group, which served as a control, anesthesia was induced with midazolam 0.05 mg/kg, fentanyl 0.005 mg/kg (Fentanyl ^® , Janssen Pharmaceutica, Beerse, Belgium), and vecuronium 0.1 mg/kg (Norcuron ^® , N.V. Organon, Oss, The Netherlands). In the same group, anesthesia was maintained with an intravenous bolus of midazolam, fentanyl or vecuronium, and oxygen/air mixture without inhalational anesthetics. An intravenous bolus of midazolam (0.025 mg/kg) was administrated every 45 min during the surgical procedure. An additional intravenous bolus of fentanyl (0.001 mg/kg) was added depending on heart rate. Vecuronium (0.01 mg/kg) was added every 40 min or if there was an increase in end-tidal carbon dioxide. Patients were mechanically ventilated to maintain end-tidal carbon-dioxide tension (P _ET CO ₂ ) to maintain normocapnia (4.5–5.0 kPa). At the end of surgery the effect of the paralytic agent, vecuronium, was reversed by neostigmine (0.04 mg/kg). Atropine (0.01 mg/kg) was co-administrated with neostigmine.

In the sevoflurane–fentanyl group, the study group, anesthesia was induced with sevoflurane 5–8 vol% (Sevorane ^® , Abbott Laboratories S.A., Abbott Park, IL, USA) and maintained with sevoflurane 0.8–1.0 vol% in oxygen/air mixture, supplemented with fentanyl 0.005 mg/kg in an intravenous bolus. An additional intravenous bolus of fentanyl (0.001 mg/kg) was added in the same manner as in the midazolam–fentanyl group.

In both groups, before the surgical incision the surgeon infiltrated 2% lidocaine containing adrenaline solution (2 ml; i.e. 40 mg of lidocaine plus 0.025 mg epinephrine) (Lidokain-Adrenalin ®, Belupo, Croatia). Non-steroid anti-inflammatory drugs were not administered, because the authors avoid administering them routinely intraoperatively.

Anesthesia was always administered by one of two anesthesiologists experienced in both anesthetic techniques. During surgery, the children were monitored using standard three-lead electrocardiography, pulse oxymetry, capnometry and non-invasive blood pressure monitoring.

Data collection

Preoperative, intraoperative and postoperative data on oxygen saturation, heart rate, hemoglobin, and hematocrit were collected by the anesthesiologists who carried out the anesthesia. Preoperative data were collected 5 min before induction. Intraoperative and postoperative oxygen saturation and heart rate values were measured every 5 min during surgery and during the patient’s stay in the postanesthesia care unit (PACU); the average values were used for analysis. Intraoperative hemoglobin and hematocrit were measured at the end of surgery. Postoperative hemoglobin and hematocrit values were measured immediately before discharge from the PACU. Intraoperative and postoperative complications were recorded. Intraoperative complications included difficulty of intubation, presence of cardiac arrhythmias and bronchospasm. Difficulty of intubation was determined according to the number of intubation attempts. Easy intubation involved one attempt, moderately difficult intubation involved 2 attenpts, difficult intubation involved 3 or more attempts. Cardiac arrhythmias were recorded by type (supraventricular/ventricular) and frequency of appearance (once/more than once). Bronchospasm was recorded as present or absent. Postoperative complications were observed over 2 h in the PACU, including emergence agitation (0, child sleeps peacefully; 1, agitation), postoperative nausea and vomiting (0, absent; 1, present), and postoperative pain, which was measured using different scales depending on the child’s age. An objective scale was used in children aged < 3 years, the Smiley scale was used in children aged 3–5 years, and a numerical self-rating scale (from 0 to 10) was used in children aged >5 years. If pain was suspected, either paracetamol 20–30 mg/kg or diclophenac 1 mg/kg were administered per rectum and the degree of agitation reassessed after 30 min. Only cases in which agitation persisted after the administration of analgesics were included in the analysis; agitation that disappeared after the administration of analgesics was considered pain-related. Owing to the different pain-assessment scales used, pain intensity could not be analyzed and compared in more detail.

After 2 h in the PACU, the children were transferred to a department where specialized care is provided for patients after cleft lip and palate repair. Postoperative pain was treated according to a routine protocol including paracetamol 15 mg/kg every 6 h or diclophenac 1 mg/kg every 8 h per rectum.

Statistical analysis

The sample size was calculated from the data from a previous study. With α of 0.001 and a power of 90% to detect the differences, the authors calculated that 63 patients were needed per group. They included more patients per group to allow for possible exclusions. Distribution of continuous variables was determined with the Kolmogorov–Smirnov test of normality. Normally distributed data were expressed as mean ± standard deviation (SD). Non normally distributed data were presented as a median with interquartile range (IQR). Categorical data were expressed as frequencies and percentages. The association of categorical variables was assessed using the χ ² test and Fisher’s exact test. Student’s t-test and the Mann–Whitney U-test were used for comparison of normally distributed variables and skewed data, respectively. Changes over time were tested with a two-way non-parametric repeated-measures analysis. P < 0.05 was considered statistically significant. Statistical analysis was performed with SPSS software for Windows, version 11.0 (SPSS Inc., Chicago, IL, USA).