Anaesthetic care for surgical management of adolescent idiopathic scoliosis

Learning objectives.

After reading this paper, the reader will be able to:

• •Explain the important considerations for perioperative care in adolescent idiopathic scoliosis.
• •Identify anaesthetic techniques that do not interfere with intraoperative neurophysiological monitoring.
• •Formulate an evidence-based multimodal approach to effective pain management after posterior spinal instrumentation for adolescent idiopathic scoliosis.
• •Summarise options for the management of postoperative nausea and vomiting.

Key points.

• •A multimodal approach combining intrathecal opioids, PCA, NSAIDs, and paracetamol can provide effective pain relief for posterior spinal instrumentation.
• •Anaesthetic techniques that do not interfere with intraoperative neurophysiological monitoring should be used.
• •Prophylactic aprepitant can be a useful adjunct to dexamethasone and ondansetron for prevention and treatment of postoperative nausea and vomiting.
• •Preoperative warming decreases the duration of intraoperative hypothermia.

Adolescent idiopathic scoliosis (AIS) is an abnormal curvature of the spine in the coronal, sagittal, and axial planes that affects 1–3% of adolescents.¹ The diagnosis is made in clinical practice with the Adam's forward bending test (positive if more than 7°). This is confirmed on radiographs when the coronal spinal curvature exceeds a Cobb angle (degree of the most tilted vertebrae on radiograph) of 10°.1, 2 For scoliosis less than 10°, there is a relatively equal incidence between males and females.¹ Scoliosis greater than 30° affects 0.15–0.3% of adolescents, with the incidence in females five times higher than in males. Although scoliosis has a number of possible aetiologies, AIS is the most frequent and accounts for 80% of cases.¹

In addition to the cosmetic aspects of the deformity, more severe curves are often accompanied by psychosocial distress, and physical problems such as increased back pain and cardiorespiratory dysfunction.¹ AIS with a curve greater than 45–50° requires surgical management; as such, deformities are likely to progress and potentially impair cardiorespiratory function and produce increasing back pain, often associated with a shift of the vertebrae.¹ At our centre, surgical management of AIS is most frequently undertaken with posterior spinal instrumentation and arthrodesis, also described as posterior spinal fusion (PSF), under intraoperative skull femoral traction (IOSFT). The procedure involves insertion of metal rods adjacent to the spine bilaterally, which are attached to the spine using hooks or screws. Once secured, the rods are rotated to realign the spine, then maintain the new alignment. This improved preservation of normal sagittal alignment results in an incidence of low back pain which is comparable with the age-matched general population. This is in contrast to the high incidence of back pain previously reported in patients whose spine has been fused with a flat back after Harrington rod instrumentation.

Approximately 70 posterior spinal instrumentation surgeries for AIS are completed by two paediatric orthpaedic spine surgeons annually at Alberta Children's Hospital. Our centre has created a care pathway that, similar to enhanced recovery after surgery (ERAS), could lead to ameliorations in early mobilisation, advancement of diet, reduced length of stay, and improved overall patients' care and perioperative experience (Table 1 and Fig. 1).

Table 1.

Enhanced recovery postoperative care plan for patients with AIS.

Postoperative day 0 Pain •PCA hydromorphone: basal infusion: 0 μg h−1; bolus: 15 μg kg−1 every 6 min to a maximum of 150 μg kg−1 h−1 •Ketorolac 0.3 mg kg−1 every 8 h (maximum 20 mg) •Paracetamol 15 mg kg−1 every 6 h Uncontrolled pain •Start PCA basal infusion •Start ketamine infusion 0–150 μg kg−1 h−1 Pruritus •Nalbuphine 0.05–0.1 mg kg−1 every 6 h i.v. as required Postoperative nausea and vomiting •Ondansetron 4–8 mg i.v. every 8 h for 24 h •Dexamethasone 0.15–0.3 mg kg−1 i.v. every 8 h as required •Fluid bolus (10–20 ml kg−1) •Reassess if opioid contributing to PONV •Dimenhydrinate 0.5–1 mg kg−1 i.v. every 6 h when required •Aprepitant 40–80 mg by mouth, once Bowel routine: •Start PCA basal infusion •Senna 2 tablets (8.6 mg per tablet) by mouth twice per day •Polyethylene glycol 17 g (in 12 ml) by mouth daily Postoperative day 1 • Pain o Discontinue PCA basal infusion (if started) o Discontinue ketamine (if started) o Initiate long-acting enteral opioid ⁃ Morphine extended release 10 mg by mouth every 8–12 h until discharge ⁃ Hydromorphone extended release 3 mg by mouth every 8–12 h until discharge o Attempt to transition from PCA to oral morphine or hydromorphone •Discontinue urinary catheter •Two physiotherapy sessions by end of day Postoperative day 2 •Discontinue PCA if still using • Initiate one of the following short-acting opioid: o Morphine 5-10mg by mouth every 3 to 4 hours o Hydromorphone 1-2mg by mouth every 3 to 4 hours •Discontinue ketorolac and initiate Ibuprofen 10mg kg-1 by mouth (max 600mg) every 6 hours •Ambulating on wards

Fig 1.

Perioperative care plan for the patients with AIS.

A variety of issues face anaesthetists caring for these patients including but not limited to: pain management (intra- and postoperative); antibiotic prophylaxis and prevention of surgical site infection (SSI); postoperative nausea and vomiting (PONV) prophylaxis and management; temperature management; blood conservation management; and prevention of neurological injury. In this article, we review the evidence and address controversies in each of these areas relating to surgery for AIS.

Preoperative optimisation

All patients should be seen by a paediatric anaesthetist in the preoperative anaesthetic clinic at least 2 weeks before their scheduled surgery. This allows each patient to be assessed thoroughly and any medical issues to be optimised. A detailed discussion of the perioperative plan including postoperative pain and symptom management and the patient's and family's role in the early recovery period decreases patients' anxiety and can improve overall satisfaction.³

Perioperative care

Intraoperative analgesia can be managed using a multimodal analgesic regimen including preoperative paracetamol, intrathecal (i.t.) morphine, infusions of remifentanil and ketamine, i.v. ketorolac and gabapentin.

One option for intra- and postoperative analgesia is the administration of i.t. opioids after induction of anaesthesia. In a retrospective study, postoperative pain control with morphine PCA, i.t. morphine plus PCA, and patient-controlled epidural analgesia (PCEA) after PSF and segmental spinal instrumentation for AIS were compared.⁴ The combination of intrathecal morphine and PCA provided the most effective pain relief.⁵ The dose of i.t. morphine reported to produce effective analgesia with minimal adverse effects varies between 2 and 19 μg kg⁻¹.4, 6 Because the duration of action of i.t. morphine is about 16 h, analgesia is lost in the morning of the first postoperative day, which often coincides with the initiation of physiotherapy.⁷ The use of a multimodal strategy after i.t. morphine with ketorolac, oxycodone and paracetamol has also been shown to be effective, and allows transition to oral opioids on the first postoperative day without the need for i.v. opioids.⁸

Remifentanil infusions may be a controversial option for anaesthesia because of concerns that it may cause hyperalgesia. However, in one study the differences in pain and sedation scores in AIS undergoing surgical management were not significantly different between patients receiving an intraoperative infusion of remifentanil (0.28 μg kg⁻¹ min⁻¹) and those receiving boluses of morphine i.v.⁹ Although the authors concluded that total requirements for morphine were higher before and after 24 h in the remifentanil infusion group, it is difficult to determine the aetiology of this. The use of remifentanil i.v. allows the provider to reduce the dose of propofol infusions, which will minimise both haemodynamic consequences and the impact on neurophysiological monitoring (see below). Pre-incisional i.t. morphine has been shown to blunt the increased postoperative opioid requirements in adolescents undergoing scoliosis correction with a remifentanil infusion as part of the anaesthetic technique.¹⁰

Ketamine 2 μg kg⁻¹ min⁻¹ i.v. during and continued for 48 h after surgery decreased morphine consumption and requirements for antiemetics during surgical management of AIS.¹¹ Similarly, there was a 29.5% reduction in morphine requirements after scoliosis surgery when patients received i.v. ketamine 150 μg kg⁻¹ h⁻¹ combined with an i.v. infusion of magnesium 8 mg kg⁻¹ h⁻¹ intraoperatively.¹² However, other studies do not seem to demonstrate the same benefits when continued longer into the postoperative period.¹³

In one study, a single preoperative dose of gabapentin was ineffective after scoliosis surgery in modifying pain scores or opioid use, and also increased postoperative sedation.¹⁴ However, gabapentin initiated before surgery and continued after surgery has been shown to reduce both opioid requirements and pain immediately after surgery and on the first postoperative day.¹⁵

PONV is often a major challenge for these patients, and in such situations multiple prophylactic antiemetics are required. Dexamethasone (0.15 mg kg⁻¹ up to a total of 8 mg) is commonly combined with ondansetron (0.1 mg kg⁻¹ up to a total of 8 mg) during surgery.¹⁶ Aprepitant is a neurokinin₁ receptor antagonist with a long half-life (9–13 h) that costs around £17 per dose and can be given at a dose of 40 mg by mouth 30–60 min before the start of surgery. It has been shown to be a potent antiemetic agent in at-risk paediatric oncological patients, and one study found that triple therapy including aprepitant (80 mg) combined with dexamethasone and ondansetron was more effective at preventing PONV than dexamethasone and ondansetron in combination.¹⁷ In addition to a number of studies in paediatric oncology showing the safety and effectiveness of aprepitant as an antiemetic, it has been shown to be one of the most effective agents at prevention of PONV in at-risk adult patients undergoing gynaecological, neurosurgical, and general surgical procedures.¹⁶ Although dose-finding studies for perioperative aprepitant in children are lacking, the available literature in adults suggests that both 40 and 80 mg of oral aprepitant provide similar antiemetic effects to ondansetron.

SSI can have serious morbidity and cost. The incidence of perioperative SSI after surgery for AIS is up to 6.7%.¹⁸ To prevent SSIs, the ASA recommendations for infection control for the practice of anaesthesiology (third edition) recommends administration of ceftazolin 40 mg kg⁻¹ (maximum 2 g) 10–60 min before incision to protect against bacteraemia, with a repeated dose at 4 h until skin closure.¹⁹ The recommendations also suggest that the patient's temperature should be maintained between 36 and 38°C from incision until skin closure to reduce SSIs.¹⁹ The extensive positioning and preparation for neuromonitoring required for posterior spinal instrumentation—including i.t. opioids, placement of invasive and neurophysiological monitors, placement of IOSFT, and final surgical (prone) positioning—means that patients are at relatively higher risk of intraoperative hypothermia as the period between induction of anaesthesia and start of surgery is prolonged. Active prewarming of patients while they await transfer to the operating room using a forced-air warming blanket has been shown to reduce the amount of time patients are hypothermic.²⁰ To further prevent intraoperative hypothermia, the operating room may be warmed before patient arrival and a forced air warming blanket and warmed i.v. fluids continued intraoperatively. Finally, to prevent SSIs, the ASA recommends use of aseptic technique for all cannulae including arterial and central venous catheters.¹⁹

Fluid balance can be a point of contention in the operating theatre as anaesthetists and surgeons try to balance the consequences of hypovolaemia and fluid overload. Restrictive fluid administration strategies are associated with shorter hospital length of stay and reduced incidences of pulmonary oedema and postoperative ileus. Importantly, in the adult literature, fluid overload is associated with reduced local and systemic tissue oxygenation. Therefore a restrictive fluid regimen may be used as part of the strategy to reduce SSIs and improve wound healing.²¹ Conversely, a restrictive fluid regimen contributes to hypovolaemia and hypoperfusion whereby the body compensates by directing blood flow away from non-vital organs. This may lead to end-organ dysfunction, in particular acute kidney injury. Hypovolaemia and fluid overload are both harmful and goal-directed fluid therapy (GDFT) is likely to represent a balanced approach. GDFT using stroke volume variation has been applied in major spine surgery in adults in the prone position with positive results.

Blood loss from PSF is reported as being between 275 and 907 ml. Intraoperative cell salvage is not usually necessary, but intraoperative tranexamic acid (20–30 mg kg⁻¹ loading dose followed by 10 mg kg⁻¹ h⁻¹ infusion) may decrease blood loss.²² There is no clear evidence from current studies on the most effective bolus and infusion dose of tranexamic acid that should be used during AIS surgery. However most dose-finding studies suggest a higher dose.22, 23

Neurological complications are an uncommon but significant complication of scoliosis correction with an incidence of 0.3% after PSF.¹⁸ A neurophysiologist should monitor somatosensory evoked potentials (SSEPs) and motor evoked potentials (MEPs) intraoperatively to enable early identification of and intervention for possible neurological injuries. The role of intraoperative neuromonitoring in paediatric spinal surgery has been reviewed recently in this journal.²⁴ As inhaled anaesthetics may interfere with neurophysiological monitoring, anaesthesia is maintained with infusions of propofol, remifentanil, and ketamine. Nitrous oxide in particular causes a more profound depression in amplitude and longer latency in SSEPs and MEPs than the halogenated volatile agents; however, this effect may be modified by other anaesthetic agents. Nevertheless, all volatile anaesthetics and nitrous oxide should be eliminated before initiating neurophysiological monitoring. Although propofol does affect the accuracy of SSEPs and MEPs, equipotent doses of intravenous anaesthetic agents have less effect on these measures than inhalational agents. Ketamine can enhance the amplitude of SSEPs and MEPs with spinal stimulation, although it has minimal effect on subcortical, peripheral, and myogenic responses. Although small doses of neuromuscular blocking agents (NMBAs) (e.g. rocuronium 0.2–0.4 mg kg⁻¹) may be used to facilitate tracheal intubation, these are expected to be metabolised before the start of surgery. Further doses of NMBAs should be avoided after intubation to facilitate neurophysiological monitoring. Opioids, such as remifentanil, do affect SSEPs but not to a clinically significant degree. A bispectral index (BIS) monitor or unprocessed EEG information obtained from the neurophysiologist can be used to guide the anaesthetist on depth of anaesthesia when total intravenous anaesthesia (TIVA) is used. The evidence to support this in adolescent patients is controversial and has been discussed recently in this journal.²⁴

Anaesthesia in the prone position is associated with a number of complications including posterior ischaemic optic neuropathy, stroke secondary to arterial occlusion in the neck, injury from inhibition of venous drainage, and a variety of peripheral neuropathies.²⁵ A full discussion of the anaesthetic considerations for prone positioning is beyond the scope of this paper, but the anaesthetist must be aware of how to care for these patients safely and mitigate injury.

Although some studies have looked at several of these interventions together as a complete bundle for patients undergoing AIS surgery, none have included all the pain management options outlined here. Furthermore, many protocols that have been studied focus only on the postoperative management (see below) with limited discussion of intraoperative management. Therefore, there is a lack of evidence of how to best combine these inventions into a perioperative care bundle.

Postoperative management

PONV and inadequate pain control are often the primary barriers to discharge after AIS correction. Persistent chronic pain with a neuropathic component is common after AIS surgery; it is associated with both preoperative pain and high early postoperative opioid consumption.²⁶ The specific details of the postoperative protocol vary between institutions, but the principles remain the same. Firstly, early transition from i.v. (usually PCA) to long-acting oral opioids minimises adverse effects, while still providing adequate analgesia when using concurrent multimodal analgesic techniques. A postoperative analgesic plan after intraoperative i.t. morphine 5 μg kg⁻¹ comprising PCA morphine, regular paracetamol, NSAID, and oral opioids on the first 2 postoperative days is effective.⁹

A recent review of perioperative pain management after spinal surgery in children concluded that early return to function is promoted by avoiding and treating opioid-induced adverse effects such as nausea and constipation, and early ambulation with physiotherapy, and that ketorolac does not appear to be implicated in the developed of postoperative pseudoarthrosis.²⁷ Others have demonstrated the benefits of an accelerated protocol in AIS surgery, which included discontinuing both the PCA and the urinary catheter by the afternoon of the first postoperative day and having patients ambulate with physiotherapy.²⁸ The benefits included fewer patients having to return to the operating theatre and earlier discharge from hospital (average 3.7 days) with no difference in complication rates or wound complications.²⁸ In another study, early removal of the urinary catheter together with physiotherapy and the introduction of a solid diet on postoperative day (POD) 1 was associated with a 50% reduction in hospital stay.²⁹ It is difficult to say which aspects of the various protocols contributed most to reduced length of stay, or whether the interventions as a group had a synergistic effect.

After operation, patients receive PCA with hydromorphone in the recovery room along with regular ketorolac i.v. (0.3–0.5 mg kg⁻¹ to a maximum of 30 mg every 8 h) and paracetamol (15 mg kg⁻¹ to a maximum of 1000 mg every 6 h p.o. or i.v.). An option is to start a PCA at a basal infusion rate or add a titrated ketamine infusion for patients who rate their pain as equal to or greater than 8/10, although this is seldom needed. Nalbuphine is a synthetic opioid agonist–antagonist that can be used in doses of 0.05–0.1 mg kg⁻¹ every 6 h as required to prevent and control opioid-induced pruritus without counteracting the analgesic effect. Nausea is managed with regular doses of intravenous ondansetron (4–8 mg every 8 h) for the first 24 h. If there is no improvement, i.v. dexamethasone (0.15 mg kg⁻¹ every 8 h) and i.v. dimenhydrinate (0.5–1 mg kg⁻¹ every 6 h) is offered to the patient. Dimenhydrinate is an anticholinergic and antihistaminergic drug thought to exert its antiemetic properties by decreasing vestibular stimulation. Although rarely needed, a repeat dose of aprepitant can be considered if nausea or vomiting is refractory to the above agents and is limiting diet and ambulation. Finally, a bowel routine of two tablets of senna twice daily and polyethylene glycol 17 g by mouth daily can be added to reduce opioid-induced constipation.

On POD 1, a long-acting oral opioid is started while the patient continues to have the hydromorphone bolus only PCA as rescue analgesia. Patients are expected to participate in two physiotherapy sessions which consist of getting the patient to sit up and, if possible, stand and gently ambulate. The urinary catheter is also removed and the patient encouraged to use the lavatory. Options for a long-acting opioid include morphine extended release (10 mg orally every 8 h) or continuous release hydromorphone (3 mg orally every 8–12 h). By POD 2 the PCA can be discontinued and replaced by short-acting opioids including morphine (5–10 mg orally every 3–4 h), or hydromorphone (1–2 mg orally every 3–4 h). Oral ibuprofen (10 mg kg⁻¹ every 8 h) is initiated and ketorolac discontinued. Stopping all i.v. analgesics enables patients to ambulate freely without i.v. catheters, and they are encouraged to ambulate in the wards. Earlier discontinuation of the PCA and early ambulation reduces the incidence of postoperative ileus, which is known to delay discharge from hospital.²⁸

Conclusions

There is no ideal combination of measures for perioperative management of paediatric patients undergoing AIS surgery. Nevertheless, the current literature does support the following principles: good multimodal analgesia; aggressive prophylaxis and treatment of PONV; careful attention to prevention of infection including maintenance of normothermia; use of tranexamic acid to decrease blood loss; and avoidance of drugs that interfere with neuromuscular monitoring. After surgery, the provision of effective analgesia and management of possible adverse effects is key to allow early return of function. One protocol for managing AIS corrective surgery is presented here, although variations based on the same principles could be equally or more effective.

Acknowledgements

The authors thank Dr Fábio Ferri-de-Barros, Pediatric Orthopaedics and Spine Surgeon at the Alberta Children's Hospital, Clinical Associate Professor, University of Calgary and President-Elect, Canadian Paediatric Spine Society for invaluable advice and help in the preparation of this article.

Declaration of interest

The authors declare that they have no conflicts of interest.

MCQs

The associated MCQs (to support CME/CPD activity) will be accessible at www.bjaed.org/cme/home by subscribers to BJA Education.

Biographies

Christopher Young BHSc (Hons) is a resident anaesthesiologist completing his training at the University of Calgary.

Duncan McLuckie DA FRCPC is an attending paediatric anaesthesiologist at the Alberta Children's Hospital and a clinical assistant professor at the University of Calgary. His interests include the development of protocols and care bundles for paediatric patients undergoing common surgeries.

Adam Spencer MSc FRCPC is an attending paediatric anaesthesiologist at the Alberta Children's Hospital and a clinical assistant professor at the University of Calgary. As the director of acute pain services, he has developed a number of perioperative guidelines with the goal of improving patient care.

Matrix codes: 2A07, 3A08