Abstract
Abstract
Background and objectives
Craniosynostosis surgery is considered a very painful procedure due to extended scalp and periosteal detachment, and is associated with prolonged postoperative consumption of opioids and their side effects. In this observational descriptive case series study, we investigated perioperative opioid consumption in children undergoing craniosynostosis repair under general anesthesia when scalp nerve block with levobupivacaine was involved.
Methods
After standard anesthesia induction, scalp nerve block with levobupivacaine 2 mg/kg plus epinephrine 1:800,000 was performed. Hemodynamic parameters and opioid consumption were noted. Patients were monitored in the recovery room. Requirements of additional analgesia, indicated by the Children's Hospital of Eastern Ontario Pain Scale (CHEOPS) pain score of >9, and incidence of side effects (sedation, nausea, and vomiting) were recorded during the first 24 hours.
Results
A total of 32 patients were recruited in this study; 88% of them needed morphine rescue in the recovery room because they had high CHEOPS scores. Trigonocephaly was the most frequent type of craniosynostosis (37.5%), requiring 50% more opioids in the postoperative period than other forms of craniosynostosis.
Conclusion
Scalp nerve block can be proposed as a complement to the routine craniosynostosis anesthetic protocol, because it is easy to perform, seems to reduce the need for supplementary opioids during the perioperative period, and can reduce the risk of developing acute opioid tolerance and chronic pain. In the event of trigonocephaly or craniofacial reconstruction, a complementary infraorbital nerve block can be added.
Keywords
analgesics, opioid: remifentanil; anesthetics, local: levobupivacaine; craniosynostoses; nerve block; pain, postoperative; scalp;
1. Introduction
Surgery for correction of premature fusing of cranial sutures (craniosynostosis) is considered a very painful procedure because of extended scalp and periosteal detachment, and is associated with prolonged postoperative opioid consumption and the resulting side effects.1, 2 A large number of studies have been conducted on craniosynostosis, but there are few papers concerning the postoperative pain associated with craniofacial syndromes.3, 4
In our clinical practice, a remifentanil–sevoflurane anesthesia technique is used, to promote faster awakening and improve the evaluation of neurological status; however, there is the risk of inadequate postoperative analgesia.1, 5, 6
Prior to the introduction of the scalp nerve block (SNB) technique in our department, the postoperative pain management protocol for craniosynostosis surgery included a standard anesthesia protocol (sevoflurane plus remifentanil infusion). The pain management protocol included administration of intravenous (IV) paracetamol at the end of surgery, followed by titration of morphine in the postanesthesia care unit (PACU) as needed, and then administration of IV nalbuphine hydrochloride in the surgical ward, if necessary, depending on pain scores. This standard approach is insufficient for controlling postoperative pain and is associated with opioid-related side effects (nausea, vomiting, and sleepiness).
To our knowledge, there are no papers on the use of SNB in craniosynostosis surgery; only the use of lidocaine or bupivacaine scalp infiltration has been reported.7, 8, 9 In a previous letter, we reported the first three cases of levobupivacaine SNB for a different kind of cranial surgery in pediatric patients, with a long-lasting analgesic effect of levobupivacaine SNB associated with paracetamol and/or nonsteroidal anti-inflammatory drugs.10
The aim of the present study was to demonstrate a new clinical practice in our hospital and perioperative opioid consumption in children undergoing craniosynostosis surgery under general anesthesia with associated levobupivacaine/epinephrine SNB.
2. Methods
After obtaining parental consent, according to the Helsinki Declaration, the anesthesia data and postoperative pain reports of 32 patients (5–93 months old, American Society of Anesthesiologists status I–II, and scheduled for craniosynostosis repair) were prospectively evaluated. The exclusion criterion was previous scalp incisions or allergy to local anesthetics. Patients were premedicated with oral or intrarectal midazolam 0.5 mg/kg 30 minutes prior to anesthesia induction. The anesthesia protocol was standardized for all patients, consisting of sevoflurane (FiO2 = 1) associated with IV remifentanil 0.5 μg/kg/minute (over 3 minutes) to facilitate orotracheal intubation. No muscle relaxant was administered. Anesthesia was maintained with sevoflurane (1 minimum alveolar concentration) in O2/N2O (50%/50%) and IV remifentanil (0.25 μg/kg/minute). The remifentanil infusion was titrated as required, to avoid hemodynamic variation greater than 20% during pain stimuli. Hemodynamic variables, such as heart rate, systolic blood pressure, diastolic blood pressure (DBP), and mean blood pressure, were continuously monitored and recorded at baseline, prior to, and after the SNB procedure, and prior to and after skin incision. The SNB procedure was performed by an anesthesiologist after orotracheal intubation and 10 minutes prior to skin incision at both sides of the head, using the modified Pinosky10, 11 technique (refer to the next subsection), according to the craniosynostosis type, with levobupivacaine 0.125% 2 mg/kg plus epinephrine 1:800,000 (1.25 μg/mL). Epinephrine concentration was chosen according to the patient's weight.12The neurosurgeon did not perform any skin infiltration to the scalp.
2.1. SNB technique
Anesthetic infiltration was made with a 23-gauge needle. The supraorbital and supratrochlear nerves were blocked with 0.75–1 mL of the local anesthetic solution introduced 0.75–1 cm above the supraorbital notch, perpendicular to the skin and then parallel to the eyebrow. The auriculotemporal nerves were blocked bilaterally with 0.75–1 mL of solution injected 1 cm anterior and superior to the tragus, perpendicular to the skin and then parallel to the ear, avoiding the superficial temporal artery. Posterior branches of the greater auricular nerves were blocked with 0.75–1 mL of solution between the skin and bone, 1 cm posterior to the ear, at the level of the tragus. The greater and third occipital nerves were blocked at each side of the occipital artery with 1.5–2 mL of the local anesthetic solution; lesser occipital nerves were blocked with 1 mL of the local anesthetic solution (Fig. 1, Fig. 2).10, 11, 12, 13, 14, 15
Download full-size image
Download full-size image
2.2. Pain control
At the end of surgery, remifentanil infusion was stopped and the total intraoperative dose was recorded. All patients received 15 mg/kg of IV paracetamol 30–45 minutes prior to skin closure, and then every 6 hours. Postoperative pain scores, morphine/nalbuphine cumulate doses, and associated side effects (sedation, nausea, and vomiting) were assessed and recorded by nurses on arrival at the PACU immediately after orotracheal extubation, every hour for the first 2 hours after awakening, and then for 24 hours in the postoperative ward. Postoperative pain was assessed using the Children's Hospital of Eastern Ontario Pain Scale (CHEOPS) score16, 17, based on crying, facial expression, verbal statements, position of torso, touching of the wound, and movement of legs. The degree of emergence agitation (EA) of the patient was measured using Aono's four-point scale. EA was defined as an Aono's four-point scale score of 3 or higher.18 After the patient had opened his or her eyes in response to language stimulation, CHEOPS and Aono's scale scores were measured. In the PACU, if the CHEOPS score was ≥8 and/or Aono's score was higher than 2, IV morphine (0.02 mg/kg) was administered every 10 minutes, in order to achieve a CHEOPS score of <8 and/or an Aono's four-point scale score of lower than 3. Once in the surgical ward, if the CHEOPS score was ≥8 and/or the Aono's four-point scale score was higher than 2, IV nalbuphine 0.2 mg/kg was administered. Values were recorded in a special chart designed for the study.
2.3. Statistical analysis
The nonparametric Friedman test and Wilcoxon matched pairs test corrected for multiple comparisons were performed to compare changes in hemodynamics and opioid consumption between periods. Fisher exact test was performed to compare nominal data. All results are expressed as median ± median absolute deviation. The median absolute deviation is a variation of the average absolute deviation that is even less affected by outlying values because these values have less influence on the calculation of the median than on the mean. In general, for data with extreme values, the median absolute deviation or interquartile range can provide a more stable estimate or variability than the standard deviation.
A p value of <0.05 was considered statistically significant. All statistical analyses were performed using StatView software for Windows (version 4.57; Abacus Concepts Inc., Berkeley, CA, USA).
On the basis of a previous pilot study, we calculated the mean remifentanil dose in 12 patients who underwent craniosynostosis correction surgery, using the same anesthesia protocol without SNB. The mean remifentanil dose was 0.31 μg/kg/minute. Considering an α value of 0.05 and a β value of 0.2, we estimated that 23 patients would be necessary to demonstrate a 30% reduction in the intraoperative remifentanil dose using SNB.
3. Results
Thirty-four patients were recruited in the study. Only two patients were excluded prior to the completion of the protocol because of intraoperative bleeding requiring long-term endotracheal intubation. Thirty-two patients remained for analysis. Demographics, clinical data, duration of anesthesia, and median time from the end of surgery to awakening are listed in Table 1. Three patients also had Crouzon syndrome.
All patients were operated by the same surgeon, with 13 of them being in the prone position and the others in the supine position. All surgical approaches had a bitragal scalp incision. The trigonocephaly group was treated with fronto-orbital strip craniectomy plus an orbital reconstruction performed by a maxillofacial surgeon. The scaphocephaly group underwent strip craniectomy for sagittal treatment, and fronto-orbital reconstruction for brachycephaly.
3.1. Hemodynamics
Patients showed great hemodynamic stability when heart rate, systolic blood pressure, and mean blood pressure were evaluated prior to and after skin incision, with a slight tendency of DBP to increase during skin incision (Fig. 3A and B). Heart rate increased only 1.5% at skin incision and 6% at skin closure. Blood pressure increased less than 5% at skin incision and closure. DBP showed a median increase of 20% after skin incision. No statistical difference was noted in hemodynamics associated with the performance of SNB.
Download full-size image
3.2. Opioids: rescue doses and side effects
The median total remifentanil dose was 0.23 ± 0.11 μg/kg/minute (range 0.10–0.44 μg/kg/minute). The anesthesiologist did not need to increase remifentanil doses prior to and after skin incision (Table 2).
The average CHEOPS pain score on awakening was 10 points (range 4–13 points). Twenty-six patients who had a CHEOPS score of ≥8 and an Aono's score of higher than 2 on awakening received IV morphine. These scores were persistently elevated during the 1st postoperative hour, and then reduced to <8 (CHEOPS score) and lower than 3 (Aono's score) during the 2nd postoperative hours, 3rd postoperative hours, and 4th postoperative hours (Fig. 4). Only six children had nausea over the first 24 hours, and half of them vomited, requiring IV ondansetron 0.1 mg/kg once. Rudkin's scores were 1 or 2 after morphine administration.19
Download full-size image
Once in the surgical ward, 17 patients needed IV nalbuphine once or twice over the first 24 postoperative hours. No patient had an Aono's score of higher than 2. Eight of them had been diagnosed with trigonocephaly, three with brachycephaly, three with pachycephaly or posterior plagiocephaly, and three with scaphocephaly. Fifteen patients did not require any IV nalbuphine over the first 24 hours. No children had nausea or vomited. Consumption of morphine in PACU and nalbuphine in the surgical ward is presented in Table 3.
4. Discussion
Surgical correction of craniosynostosis is a very painful procedure, but postoperative management of this kind of surgery had not been standardized.1, 5 Anesthetic teams administer a combination of intrarectal paracetamol and codeine, a combination of paracetamol and nonsteroidal anti-inflammatory drugs, or a postoperative infusion of remifentanil.1, 5, 20
Painful procedures cause significant stimulation, which increases blood pressure, heart rate, and intracranial pressure. This can be attenuated by the use of a scalp block, which can be performed simply and quickly.11, 21 It reduces the risk of hypotension associated with an increase in the depth of anesthesia, which is sometimes required to attenuate the noxious stimulation.11, 21 In our study, hemodynamic variables were highly stable during the SNB procedure, skin incision, and closure (Fig. 3A and B). The slight variations in hemodynamic variables suggest that SNB is effective in attenuating sympathetic response to pain stimuli (skin incision and craniotomy) without increasing the requirement for anesthetics or antihypertensive drugs. In a previous letter to the editor, we described hemodynamic stability during skull-pin placement and skin incision when SNB was associated with general anesthesia in pediatric neurosurgery procedures.10
The use of new anesthetic short-acting agents, such as sevoflurane and remifentanil, enables faster neurological evaluation upon awakening, but increases the problem of postoperative analgesia. Several authors have described acute opioid tolerance after remifentanil infusion, increasing postoperative requirement of opioids probably as a result of hyperalgesia, especially when large doses are administered during surgical procedures.6Deficient postoperative analgesia can be prevented with transitional analgesia at the end of the surgery.22 Morphine as transitional analgesia, when remifentanil was perfused intraoperatively, resulted in better neurological recovery in adult neurosurgical patients compared to fentanyl.23 In our study, we did not administer morphine as transitional analgesia so that we could evaluate postoperative pain, demonstrating the efficacy of SNB. SNB after remifentanil-based anesthesia in adult neurosurgery provides transitional analgesia with similar quality and postoperative hemodynamic profile to those of morphine.23
Eighty-two percent of patients had high CHEOPS scores on awakening, which could be controlled by IV morphine, and 46% of patients remained “pain free” for the first 24 hours after surgery. Even though IV morphine was required to lower CHEOPS scores, the median dose administered during the first 2 hours after wakening was very low (median 0.08 mg/kg, range 0.02–0.2 mg/kg), suggesting that high CHEOPS scores were not related only to pain and demonstrating the efficacy of SNB. However, the CHEOPS score has been described as a behavioral measure rather than a pain intensity score per se, and crying itself as well as hunger, anxiety, restraints, or frustration can increase CHEOPS scores.16, 17 Therefore, high CHEOPS scores at awakening can also be explained by factors other than pain, such as EA, already observed in patients receiving sevoflurane and remifentanil infusions during anesthesia.18, 24, 25, 26 In addition, Aono's score was higher than 2 on awakening, suggesting that postoperative excitement may explain this high CHEOPS scores at awakening.
Remifentanil may cause EA that can persist for at least 60 minutes, as already described in previous studies.5, 27 In this report, more than 50% of cases that presented EA required opioids. In fact, opioids were used to reduce EA rather than to relieve pain. It would be interesting to administer a small dose of morphine (0.08 mg/kg) prior to the end of anesthesia, to prevent EA, even if SNB has been performed.
Despite high CHEOPS scores being registered at awakening, CHEOPS scores remained below 8 and Aono's score was lower than 3 for the first 24 hours, and only one or two IV nalbuphine-hydrochloride administrations were required to maintain CHEOPS scores below 8, as observed in a previous report.10 In general, additional analgesia was required only during nursing care (mobilization, washing, and dressing).
The scalp comprises five layers, and is richly innervated by a double sensory system: trigeminal nerve branches (supratrochlear, supraorbital, zygomatic, and auriculotemporal) and cervical plexus branches (posterior branch of the great auricular, 3rd, great, and lesser occipital),13 which can cause severe pain when large scalp incisions such as bitragal incisions are made. Scalp pain responses can be blocked with a local anesthetic administered as an SNB8, 10, 11; however, few studies have showed the benefits of SNB in children. Suresh and Bellig8 have reported a case of regional anesthesia of the scalp nerves, with 0.125% bupivacaine (0.8 mg/kg), in a premature neonate of 700 g who was implanted with an Omaya reservoir. The locoregional technique enabled the use of low doses of opioids (fentanyl), and hemodynamic stability was maintained throughout the procedure, with no changes during incision and skin closure.
Trigonocephaly and brachycephaly seem to be the most painful procedures, requiring 50% and 30% more nalbuphine, respectively, during the postoperative period. This could be explained by temporal muscle detachment, fronto-orbital strip, canthopexia, and bilateral orbital reconstruction, which differ from other craniosynostosis approaches. Orbital reconstruction cannot be blocked completely by SNB. The authors suggest an additional infraorbital block to improve pain control in children with orbital reconstruction. Even when these patients needed more postoperative IV nalbuphine rescue doses (0.2–0.4 mg/kg) over the first 24 hours, the remifentanil doses administered did not differ from those used in other patients (n = 32, median 0.23 ± 0.11 μg/kg/minute), suggesting that SNB could block pain during surgery, but the effect was limited over time. We also observed that remifentanil doses prior to and after skin incision remained unchanged in general, suggesting the efficacy of SNB (Table 2). Meanwhile the postoperative periorbital swelling in these patients could aggravate pain and cause discomfort. The postoperative treatment protocol included furosemide 1 mg/kg and methylprednisolone 2 mg/kg.
Prior to the introduction of SNB, neurosurgeons used to infiltrate the scalp with lidocaine/epinephrine solutions, which had two functions: hemostasis and anesthesia. Levobupivacaine/epinephrine solution provided a long-lasting anesthetic effect on scalp tissues, together with additional vasoconstriction useful to reduce scalp bleeding during incision and detachment. SNB is a regional anesthetic technique that is easy to perform and has no significant side effects. In our study, the surgeon observed only two cases of hematoma at the infiltration site during scalp detachment. Moreover, the SNB technique offers the advantage of limiting the injection to the site of specific nerves, thus reducing the volume of local anesthetic needed in comparison with circular infiltration used by some surgeons.
One of the study limitations was nonexistence of a control group. The authors suggest further studies to establish the real benefits of SNB in craniosynostosis surgery.
5. Conclusion
In conclusion, our study demonstrated that patients undergoing craniosynostosis surgery, a very painful procedure, can benefit from the SNB technique as a complement to the routine analgesic protocol. Levobupivacaine/epinephrine SNB may provide better hemodynamic stability at skin incision and closure, and possibly reduce the need for supplementary opioids.
The SNB technique did not show any significant side effects, and it is easy to perform. In addition, in theory, it will reduce the risk of developing acute opioid tolerance and chronic pain.
The administration of a reduced IV dose of morphine prior to the end of surgery (0.08 mg/kg) is suggested to improve awakening.
A complementary infraorbital nerve block may be useful in patients requiring fronto-orbital strip, canthopexia, or bilateral orbital reconstruction (trigonocephaly or craniofacial reconstruction), and this should be evaluated in a future study.
Acknowledgments
The authors wish to acknowledge the nurses from the PACU and the postoperative ward (pediatric neurosurgery unit), Neurologic and Neurosurgical Hospital Pierre Wertheimer, Bron, France. They thank Anne-Marie Cervera and Lynn Richardson for revising the English text. No funding was received for this work.