Anaesthesia for acute intestinal obstruction associated with cerebral arteriovenous malformation in a child.
Abstract: This case report describes the perioperative management of a child presenting with acute intestinal obstruction secondary to bowel malrotation after a recent intracranial haemorrhage associated with an intracranial arteriovenous malformation. We discuss the anaesthesia planning for this case, where the 'optimal' management strategies for the two conditions present are potentially conflicting. Issues include rapid sequence induction in the presence of a ruptured arteriovenous malformation, maintenance of cerebral perfusion pressure in the face of bowel ischaemia, and the use of epidural anaesthesia in a child with recent intracranial haemorrhage. Written consent was obtained from the patient and parents to publish this case.

Key Words: anaesthesia, acute abdomen, AVM, intracerebral haemorrhage
Article Type: Report
Subject: Arteriovenous malformations (Care and treatment)
Arteriovenous malformations (Complications and side effects)
Anesthesia (Dosage and administration)
Intestines (Obstructions)
Intestines (Complications and side effects)
Intestines (Care and treatment)
Brain (Hemorrhage)
Brain (Care and treatment)
Brain (Complications and side effects)
Authors: Scarth, E.J.
White, M.C.
Pub Date: 01/01/2010
Publication: Name: Anaesthesia and Intensive Care Publisher: Australian Society of Anaesthetists Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2010 Australian Society of Anaesthetists ISSN: 0310-057X
Issue: Date: Jan, 2010 Source Volume: 38 Source Issue: 1
Geographic: Geographic Scope: Australia Geographic Code: 8AUST Australia
Accession Number: 221657255
Full Text: We present our management of a seven-year-old girl with recent cerebral haemorrhage due to a large arteriovenous malformation (AVM) who presented with bowel malrotation and duodenal obstruction. This combination of pathologies presents a number of challenges for anaesthesia which are discussed.

CASE HISTORY

A seven-year-old girl was referred to the regional paediatric surgical unit with bilious vomiting. She was previously asymptomatic until one month before this admission when she developed a severe headache and had a seizure. Cerebral computerised tomography and angiography demonstrated intracranial haemorrhage associated with raised intracranial pressure (ICP) secondary to a temporal lobe AVM (Figure 1). She recovered with no neurological deficit, was discharged home on dexamethasone and neurosurgical repair of her AVM planned for three months time when the oedema had resolved. Four days later she was readmitted with vomiting. Cerebral computerised tomography showed no new pathology and resolution of the radiological signs of raised ICP. The vomiting then became bilious and an upper gastrointestinal contrast study performed which showed bowel malrotation and duodenal obstruction requiring emergency laparotomy. On examination, she was alert and orientated with no neurological deficit. She was haemodynamically stable and required no fluid resuscitation.

[FIGURE 1 OMITTED]

Anaesthesia was induced in the supine position with 30[degree] head-up tilt and full routine monitoring including non-invasive blood pressure cycling every minute. She was pre-oxygenated for three minutes, remifentanil 1 [micro]g/kg/minute was administered until she became drowsy, followed by propofol 2 mg/kg and rocuronium 1 mg/kg. Cricoid pressure was applied as she lost consciousness and the trachea intubated with a size 5.5 cuffed oral endotracheal tube. Anaesthesia was maintained with isoflurane in oxygen-enriched air. The left radial artery was cannulated with a 22 G cannula and an 18 G cannula inserted into a forearm vein. A 4.5 Fr 8 cm central venous catheter was inserted into the right internal jugular vein under ultrasound guidance. A thoracic epidural catheter was easily inserted at [T.sub.11-12] using an 18 G Tuohy needle and loss of resistance to saline technique. A urinary catheter was inserted to accurately monitor fluid balance. All fluids were warmed with a Hotline[TM], the child covered with a warming blanket and nasopharyngeal temperature monitored. Fluid boluses of crystalloid 10 ml/kg were given to maintain a central venous pressure of 10 to 12 cmH2O and blood pressure maintained within 10% of baseline using a dopamine infusion titrated to effect up to a maximum of 7.5 [micro]g/kg/minute. An epidural infusion of bupivacaine 0.1% with fentanyl 2 [micro]g/ml was administered intraoperatively at 12 ml/hour (0.5 ml/kg/hour) and the remifentanil infusion continued at 0.2 [micro]g/kg/minute. We had planned to stop the remifentanil infusion once epidural analgesia was established but on reducing the infusion, an increase in heart rate was observed. Inadequate epidural blockade was suspected and the remifentanil infusion continued. Intraoperative haemoglobin and acid-base balance was monitored using arterial blood gas analysis and no blood transfusion was required.

Surgical findings included a very dilated stomach and discolouration of the proximal intestine which appeared ischaemic but not necrotic. After Ladd's procedure and appendicectomy were performed the bowel colour improved and no resection was necessary.

At the end of surgery, the remifentanil infusion was discontinued, tracheal extubation was uneventful and the dopamine infusion ceased. The patient was moving all four limbs and responding appropriately. In recovery she complained of pain (pain score 8/10) particularly on her right side. A unilateral epidural block was suspected (confirming the intraoperative suspicion of inadequate epidural analgesia) so the epidural catheter was manipulated and a bolus of bupivacaine 0.25 ml/kg 0.25% administered. Her pain score improved to 6/10 and she fell asleep but would wake intermittently and complain of pain on movement. Since the epidural was suboptimal but appeared to be providing some analgesic benefit, the infusion was changed to bupivacaine 0.1% (without fentanyl), morphine 0.05 mg/kg was administered and a morphine patient-controlled analgesia (PCA) commenced. She was admitted to the high dependency unit overnight for continuation of arterial pressure monitoring and accurate fluid management. The next day, the epidural block was clearly unilateral so the catheter was removed and morphine PCA continued. She was transferred to the surgical ward and received total parenteral nutrition for six days. She was discharged home on the ninth postoperative day with no neurological complications.

DISCUSSION

This case presents conflicting requirements for anaesthesia: the need for a rapid sequence induction (RSI) in a child with acute intestinal obstruction who also requires a haemodynamically stable induction because of the significant risk of rebleeding from her cerebral AVM (33% in the first year (1)), with a potential for increases in blood pressure in association with perioperative events potentially adding to that risk; the likely requirement for vasopressor therapy to prevent any drop in baseline blood pressure and subsequent neurological injury during anaesthesia while continuing to provide optimal gut perfusion in view of her volvulus and gut ischaemia; and finally, whether epidural or PCA is the technique of choice for postoperative pain management.

Emergency laparotomy requires RSI tracheal intubation. Suxamethonium is the most commonly used muscle relaxant for RSI because of its fast onset and short duration of action. However, administration of suxamethonium can elevate ICP and hyperkalaemic arrest has been reported following administration to a patient with closed head injury (2). In recent years rocuronium has been proposed as an alternative to suxamethonium for RSI. Rocuronium provides satisfactory intubating conditions within 60 seconds (3), compared with 30 seconds for suxamethonium (4), but the duration of action is far longer unless sugammadex is available for rapid reversal of rocuronium-induced neuromuscular blockade (5). A recent systematic review concluded suxamethonium created superior intubating conditions to rocuronium (6). In the case described, we had to weigh up the risks of aspiration from a 'full stomach' against the uncertainty of the actual risks of hyperkalaemic arrest (rare but reported (2)), neurological damage (effect of suxamethonium-induced raised ICP in the presence of a ruptured AVM is unknown), versus the risk of inferior intubating conditions (in children the clinical significance of which is unknown) if rocuronium modification to the 'classical' RSI is used. The author (MW) routinely uses rocuronium in high-risk children undergoing cardiac catheterisation so is familiar with the intubating conditions encountered after administration of rocuronium 1 mg/kg. Therefore, because of the rare but catastrophic side-effects of suxamethonium and our familiarity with rocuronium, we elected to avoid suxamethonium and administered rocuronium instead.

Since it is difficult to establish invasive arterial pressure monitoring in children prior to administering anaesthesia, frequent non-invasive blood pressure monitoring (every minute) was performed. Careful monitoring is required to detect any hypotension associated with induction and hypertension associated with laryngoscopy and intubation. We used remifentanil 1 [micro]g/kg/minute titrated to effect, to attenuate the haemodynamic response to RSI with the aim of minimising the risk of the AVM rebleeding. In adults, remifentanil 1 [micro]g/kg is reported as superior to lignocaine 1.5 mg/kg in this respect7 but is associated with a fall in mean arterial blood pressure of 18% from baseline. We judged that this degree of hypotension with respect to baseline would be undesirable in this case and is outside the narrow target range of 10% we had set ourselves. However, children tend to have less hypotension than adults in response to induction agents and neuraxial blockade due to their greater parasympathetic drive. We planned to use intravenous phenylephrine, if required, to maintain blood pressure within 10% of baseline until central venous access was established and dopamine commenced. All induction agents with the exception of ketamine have hypotensive side-effects and ketamine may cause elevation of ICP. We chose to use remifentanil and propofol on induction for reasons of familiarity and for minimising changes in both haemodynamics and ICP.

Hypotension is a risk not only on induction but also during maintenance of anaesthesia. Despite adequate filling pressures, the vasodilatory effects of inhalational anaesthetics combined with intraoperative fluid shifts associated with acute intestinal obstruction are likely to result in relative hypotension compared to baseline. In the presence of head injury this could result in significant cerebral hypoperfusion. Noradrenaline and dopamine have been used to maintain cerebral perfusion pressure (CPP) in children with head injury (8). In animal models, dopamine worsens cerebral oedema despite restoring CPP (9). Conversely, although noradrenaline can improve CPP it may theoretically compromise splanchnic perfusion through its alpha-adrenergic action. Adults with isolated head trauma (receiving noradrenaline infusions to maintain a target CPP) show evidence of splanchnic ischaemia (10) which is unrelated to the use of inotropes and may worsen with decreased CPP. We decided to administer dopamine to maintain blood pressure within 10% of baseline aiming to optimise CPP and maximise gut perfusion. In our hospital, dopamine is our firstline vasopressor (noradrenaline is second-line), for maintaining supranormal blood pressure in children undergoing renal transplantation from a live-related adult donor. Central venous pressure monitoring ensured adequate filling pressures were maintained throughout.

There is limited data on the prognosis of patients with AVMs. The absolute risk of haemorrhage from an AVM is 2% but the risk of recurrent haemorrhage may be up to 33% in the first year (1). The estimated mortality rate is approximately 1 to 1.5% (11). How this relates to the risk of haemorrhage secondary to a hypertensive response to laryngoscopy, skin incision or episodes of postoperative pain is unknown. The safe variation in blood pressure from preoperative baseline measurements is also unknown. We arbitrarily elected to keep blood pressure within 10% of baseline and used remifentanil to prevent the haemodynamic response to intubation and skin incision (7) and used dopamine to offset the vasodilatory effects of inhalational agents during maintenance of anaesthesia.

Since the risk of rebleeding from the AVM was also assumed to extend into the postoperative period, we aimed to minimise any hypertensive responses to postoperative pain by providing optimal analgesia in the form of an epidural. The benefits and risks of epidural analgesia in this child are as follows: the evidence in favour of epidural analgesia includes low complication rates reported in children (12) and our own experience, where epidural analgesia is used for majority of children undergoing laparotomy in our hospital and is successfully managed by the ward nursing staff and a specialist acute pain team. Furthermore, in adults there is evidence of attenuation of the stress response, shorter duration of postoperative ileus and improved postoperative pain control and recovery (13). The concerns regarding epidural analgesia in this case involve a number of possibilities: the theoretical risk that infusions of large volumes of local anaesthetic could increase ICP by a secondary compression effect on cerebrospinal fluid; the risk that dural puncture in a patient with recent intracranial haemorrhage may not actually lead to brainstem herniation but may cause brain shifting, which with loss of cisternal cushioning, might tear fragile fibrin plugs and result in renewed intracranial bleeding. Overall, we considered that a rapid recovery from anaesthesia enabling prompt assessment of her neurological status and optimal postoperative analgesia would be best provided by epidural analgesia. Unfortunately, our epidural provided suboptimal analgesia and intravenous morphine was required. In retrospect, since morphine PCA provided satisfactory analgesia perhaps this may have been the technique of choice from the outset. However, patients receiving intraoperative remifentanil are known to have increased analgesic requirements postoperatively14,15. Increased opioid usage is associated with a greater number of side-effects such as respiratory depression, nausea and vomiting. In addition, increased postoperative pain is associated with hypertension which could cause rebleeding. In the absence of clear evidence of the superiority of epidural or opioid analgesia, it is likely that local individual and institutional preference and expertise will rightly dictate the choice of analgesia. The concept of 'the best technique is the one that works best in your hands' is applicable here. Institutional infrastructure is also important, and includes the presence or absence of a pain team to monitor the child on the ward and assess the adequacy of the epidural block; and the level of comfort and expertise of ward nursing staff in pain assessment methods and managing postoperative epidural analgesia. A high level of such skills is required in cases such as the one described, because of the importance of detecting inadequate pain relief early and instituting a change in analgesic management to rapidly bring about satisfactory pain relief, not only for humane reasons, but also to minimise the risk of intracerebral rebleed.

In conclusion, we have discussed the issues surrounding a complex case in which two simultaneous pathologies presented conflicting requirements for anaesthesia. Such cases require more than just good skills and evidence-based knowledge because often the actual balance of risks cannot be accurately estimated. Clinical reasoning, experienced judgement, management of ambiguity and reflective learning is required. Therefore, in most cases, the 'optimal' approach will be appropriately informed by the local expertise and preferences of the professionals involved and the policies and practices of their institution.

Accepted for publication on July 14, 2009.

REFERENCES

(1.) Mast H, Young WL, Koennecke HC, Sciacca RR, Osipov A, Pile-Spellman J et al. Risk of spontaneous haemorrhage after diagnosis of cerebral arteriovenous malformation. Lancet 1997; 350:1065-1068.

(2.) Reynolds SF, Heffner J. Airway management of the critically ill patient: rapid sequence intubation. Chest 2005; 127:1397-1412.

(3.) Sasada M, Smith S. Rocuronium. In: Drugs in Anaesthesia and Intensive Care, 3rd ed. Oxford: Oxford University Press 1997. p. 326-327.

(4.) Sasada M, Smith S. Suxamethonium. In: Drugs in Anaesthesia and Intensive Care, 3rd ed. Oxford: Oxford University Press 1997. p. 352-354.

(5.) Yang LP, Keam SJ. Sugammadex: a review of its use in anaesthetic practice. Drugs 2009; 69:919-942.

(6.) Perry JJ, Lee JS, Sillberg VA, Well GA. Rocuronium versus succinylcholine for rapid sequence induction intubation. Cochrane Database Syst Rev 2008; 2: CD002788.

(7.) Kim JT, Shim JK, Kim SH, Ryu HG, Yoon SZ, Jeon YS et al. Remifentanil vs lignocaine for attenuating the haemodynamic response during rapid sequence induction using propofol: double-blind randomised clinical trial. Anaesth Intensive Care 2007; 35:20-23.

(8.) Biestro A, Barrios E, Baraibar J, Puppo C, Lupano D, Cancela M et al. Use of vasopressors to raise cerebral perfusion pressure in head injured patients. Acta Neurochir Suppl 1998; 71:5-9.

(9.) Beaumont A, Hayasaki K, Marmarou A, Barzo P, Fatouros P, Corwin F. Contrasting effects of dopamine therapy in experimental brain injury. J Neurotrauma 2001; 18:1359-1372.

(10.) Venkatesh B, Townsend S, Boots RJ. Does splanchnic ischemia occur in isolated neurotrauma? A prospective observational study. Crit Care Med 1999; 27:1175-1180.

(11.) Crawford PM, West CR, Chadwick DW, Shaw MD. Arteriovenous malformations of the brain: natural history in unoperated patients. J Neurol Neurosurg Psychiatry 1986; 49:1-10.

(12.) Llewellyn N, Moriarty A. The national pediatic epidural audit. Paediatr Anaesth 2007; 17:520-533.

(13.) Fotiadis RJ, Badvie S, Weston MD, Allen-Mersh TG. Epidural analgesia in gastrointestinal surgery. Br J Surg 2004; 91:828-841.

(14.) Guignard B, Bossard AE, Coste C, Sessler DI, Lebrault C, Alfonsi P et al. Acute opioid tolerance: intraoperative remifentanil increases postoperative pain and morphine requirement. Anesthesiology 2000; 93:409-417.

(15.) Crawford MW, Hickey C, Zaarour C, Howard A, Naser B. Development of acute opiod tolerance during infusion of remifentanil for paediatric scoliosis surgery. Anesth Analg 2006; 102:1662-1667.

Address for correspondence: Dr M. C. White, Department of Paediatric Anaesthesia, Bristol Children's Hospital, University Hospitals Bristol NHS Foundation Trust, 2 St Michael's Hill, Bristol BS2 8BJ, UK

E. J. SCARTH *, M. C. WHITE ([dagger])

Department of Paediatric Anaesthesia, Bristol Children's Hospital, University Hospitals Bristol NHS Foundation Trust, Bristol, United Kingdom

* B.Med.Sci., B.M.B.S., M.R.C.P.(UK), Speciality Registrar in Anaesthesia.

([dagger]) B.M., Ch.B., D.C.H., F.R.C.A., Consultant Paediatric Anaesthetist.
Gale Copyright: Copyright 2010 Gale, Cengage Learning. All rights reserved.