Peri‐operative management of an adult with POLG‐related mitochondrial disease

Summary

POLG‐related mitochondrial disease is a rare mitochondrial disorder that is potentially associated with anaesthetic complications such as propofol‐related infusion syndrome. A 19‐year‐old man with mitochondrial DNA deletions and POLG‐related disorders presented for an elective robotic Heller‐Dor myotomy for the treatment of oesophageal pseudo‐achalasia associated with severe gastro‐oesophageal reflux. The fasting period was minimised to reduce the risk of metabolic stress. The anaesthetic technique included a rapid sequence induction with propofol and rocuronium, a remifentanil and sevoflurane‐based general anaesthesia with multimodal monitoring and peri‐operative lactate‐free intravenous fluids with added dextrose. The patient did not experience propofol‐related infusion syndrome but did have delayed tracheal extubation due to residual neuromuscular blockade requiring a second dose of sugammadex. This report demonstrates the safety of single‐use, low‐dose propofol in this patient group. Patients with POLG‐related mitochondrial disease may be at risk of prolonged neuromuscular blockade, and appropriate dosing of neuromuscular blocking agents with monitoring of neuromuscular blockade is strongly encouraged.

Introduction

Mitochondrial diseases (MDs) are a group of multi‐systemic and heterogeneous disorders caused by impairment of the respiratory chain involved in the generation of adenosine triphosphate (ATP) by oxidative phosphorylation [1]. Numerous proteins involved in oxidative phosphorylation are derived from the mitochondrial genome. Deoxyribonucleic acid (DNA) polymerase γ (Pol γ) is a protein involved in the replication of the mitochondrial DNA (mtDNA). Mutations in the POLG catalytic subunit of Pol γ are associated with various clinical presentations ranging from infantile‐onset epilepsy to late‐onset progressive ophthalmoplegia and ataxia. POLG mutations are the most common cause of inherited MD, with the three most prevalent POLG mutations (A467T, W748S and G848S) having a combined carrier frequency of more than 1% in northern Europe [1]. There is no disease‐modifying therapy, randomised controlled clinical trials are lacking and symptomatic therapies are currently the mainstay of treatment.

Mitochondria are potential action sites for anaesthetic agents, and general anaesthesia may be deleterious in patients with MD. These patients may have an increased sensitivity to anaesthetic agents or a predisposition towards malignant hyperthermia and metabolic complications such as lactic acidosis. Insights about peri‐operative management of patients with MD are scarce, with evidence based on case reports and short series [2, 3]. In this report, we describe the peri‐operative management of an adult patient undergoing robotic Heller‐Dor myotomy for the treatment of POLG‐related oesophageal pseudo‐achalasia. This article adheres to the Anaesthesia Case Report (ACRE) checklist.

Report

A 19‐year‐old man weighing 69 kg and 188 cm in height, with a diagnosis of POLG‐related disorders, was scheduled for robotic Heller‐Dor myotomy for the treatment of oesophageal pseudo‐achalasia responsible for severe gastro‐oesophageal reflux disease (GORD) with major dysphagia.

His family history included consanguinity, autism, epilepsy, dyspraxia, dyslexia, ptosis, depression, hypertension and GORD. The childhood of the patient was marked by developmental delay, anaemia, hypotonia and frequent falls. He had no history of seizures or hepatic failure. The patient had POLG mutations (G248S and R309C) and GAG deletion at position 946 in DYT1. Previous general anaesthesia for osteosynthesis, endoscopic gastrostomy, oesophageal dilatation and intravenous access was uneventful.

Pre‐operative clinical examination showed a mild scoliosis, progressive external ophthalmoplegia, cerebellar ataxia, signs of dorsal root ganglionopathy and limb‐girdle muscular dystrophy‐like myopathy. He had GORD with episodic vomiting and major dysphagia despite two endoscopic dilatations of the lower oesophageal sphincter and therefore relied on parenteral nutrition. His vital signs and cardiac and respiratory examinations were normal. Haemoglobin, platelets, urea, creatinine, electrolytes, liver function tests and haemostasis were all within normal limits. Pre‐operative electrocardiogram and transthoracic echocardiography were normal. Oesophageal manometry showed the absence of oesophageal peristalsis and a normal relaxation of the lower sphincter of the oesophagus. Pre‐operative serum lactic acid was 1.3 mmol.l^‐1 (normal values: 0.5‐1.5 mg.l^‐1) and creatine phosphokinase was 133 IU.l^‐1 (normal values: 39‐308 IU.l^‐1).

The pre‐operative fasting period was 6 h for solids and 2 h for water. The patient received 500 ml 10% dextrose intravenously and sweetened water orally 2 h before surgery. Monitoring included electrocardiography, pulse oximetry, capnography, non‐invasive blood pressure, oesophageal temperature, bispectral index (BIS) and a peripheral nerve stimulator for train‐of‐four (TOF) monitoring. General anaesthesia was initiated with a rapid sequence induction (RSI) with propofol 3 mg.kg^‐1 and rocuronium 1.2 mg.kg^‐1.

Intubation of the patient’s trachea was performed without difficulty. A remifentanil infusion (effect site target‐controlled infusion, target: 2.0 to 4.4 ng.ml^‐1) was commenced. After induction, the patient received ketamine 0.2 mg.kg^‐1 to prevent hyperalgesia. Anaesthesia was maintained with sevoflurane (1.5%‐3.5%), and the patient received a second dose of rocuronium 0.6 mg.kg^‐1 to facilitate the surgical procedure. Fluid balance was maintained with intravenous administration of 1500 ml of dextrose 5% and 1000 ml of sodium chloride 0.9%, infused throughout the procedure. Paracetamol 1 g, nefopam 20 mg and morphine 9 mg were administered i.v. for postoperative analgesia. The patient received droperidol 1.25 mg, ondansetron 4 mg and dexamethasone 8 mg i.v. to prevent postoperative nausea and vomiting.

The procedure lasted 167 min, including 118 min of robotic Heller‐Dor myotomy. The patient experienced moderate hypotension requiring norepinephrine (5 μg.ml^‐1, at up to 30 ml.h^‐1). The patient did not have hypothermia or significant blood loss. Following the last rocuronium dose 110 m in prior, the patient had four twitches on neuromuscular TOF assessment and received sugammadex 2 mg.kg^‐1 i.v. Due to a TOF ratio of less than 0.9, the patient received a repeat dose of sugammadex 2 mg.kg^‐1 i.v. and his trachea was extubated after a 30 min spontaneous breathing test.

The patient was admitted to the intensive care unit, and a dextrose infusion was continued until oral feeding was commenced on the second postoperative day. Serum lactic acid concentration was 2.3 mmol.l^‐1 postoperatively and 1.5 mmol.l^‐1 on the following day. The postoperative course was uneventful.

Discussion

Evidence‐based anaesthetic strategies for MD are lacking; the available data are scarce and contradictory. Footitt et al reported the outcome of general anaesthesia in paediatric patients with MD using a retrospective case review study of 38 patients who had undergone 58 anaesthetics between 1989 and 2005 [3]. Various different anaesthetic agents were used, without episodes of malignant hyperthermia or intra‐operative events attributable to the anaesthesia. Despite theoretical concerns, metabolic decompensation after general anaesthesia occurred in only one case. However, no patient had documented POLG‐related MD, and it is likely that pre‐existing POLG mutations predispose for propofol‐related infusion syndrome [4], a rare but severe complication of propofol use which includes metabolic acidosis, rhabdomyolysis, arrhythmias, myocardial failure, renal failure, hepatomegaly and death [5]. Propofol has been shown to affect mitochondrial function through different mechanisms, including disruption of the permeability transition pore, inhibition of multiple electron transport chain complexes and reduction of fatty acid transport into the mitochondrion (Fig. 1) [4, 6]. Propofol‐related infusion syndrome appears to be dose dependent and strongly associated with propofol infusion over a substantial period of time (more than 48 h). Consequently, it has been suggested that patients presenting with POLG mutations should not receive a dose of more than 4 mg.kg^‐1 propofol or a prolonged infusion of propofol, and that appropriate monitoring for PRIS is warranted [6]. Our experience suggests it may be safe to use a low‐dose of propofol at induction.

Figure 1.

Deleterious effects of propofol on oxidative phosphorylation. Propofol has been shown to depress mitochondrial function by inhibiting Complex I (cI), Complex IV (cIV), Cytochrome c (Cyt c) and the acylcarnitine transferase, as well as acting as an uncoupling agent in oxidative phosphorylation and interrupting the electron flow in the inner membrane (IM) at the site of Coenzyme Q (Q). ADP, adenosine diphosphate; ATP, adenosine triphosphate; cII, Complex II; cIII, Complex III; e‐, electron; H+, proton; IMS, intermembrane space; NAD, nicotinamide adenine dinucleotide.

Patients with POLG‐related MD require an individual and meticulous pre‐operative assessment [6, 7]. The anaesthetist should investigate family genetic and phenotypic spectra as well as previous anaesthesia records. Full clinical assessment should include a multidisciplinary evaluation, including neurological, cardiac, respiratory, hepatic, renal and gastrointestinal functions.

Patients with POLG‐related MD may have a clinical phenotype similar to mitochondrial neurogastrointestinal encephalopathy syndrome, which includes the symptoms of persistent diarrhoea and cachexia related to gastrointestinal dysmotility [1]. Other patients with POLG‐related MD have been reported to present with prominent GORD that increases the risk of pulmonary aspiration. A prolonged fasting period of more than 8 hours and RSI with tracheal intubation are standard techniques to prevent aspiration in such patients. Fast‐acting depolarising neuromuscular blocking agents (NMBAs) such as suxamethonium have been reported to trigger malignant hyperthermia in patients with MD [8]. Nevertheless, evidence‐based data regarding this issue are lacking; other patients suffering from MD have received various NMBAs without complication [3].

While prolonged fasting may reduce the risk of pulmonary aspiration, metabolic stress due to prolonged fasting, combined with surgical inflammatory stress, increases the risk of metabolic decompensation and lactic acidosis. Therefore, pre‐operative fasting should be minimised, and MD patients should have dextrose added to their peri‐operative lactate‐free intravenous fluids, unless they are on a ketogenic diet or have presented adverse reactions to higher glucose intake [6]. Normoglycemia is necessary to avoid excessive glycolytic oxidation of glucose and an increase in the plasma lactate concentration [6]. Peri‐operative normoglycemia, normothermia, haemodynamic stability and optimal oxygenation are necessary to preserve energy production by the impaired mitochondria [6, 7]. The anaesthetist should assess the baseline serum lactic acid level and serial blood gas analysis may be useful.

Neuromuscular blockade is often required in laparoscopic and robotic procedures. Deep neuromuscular blockade may improve operating conditions by allowing adequate exposure and preventing iatrogenic accidents related to trocar insertion. However, in patients with MD, increased sensitivity to NMBAs (including rocuronium) with prolonged recovery and inadequate reversal by anticholinesterases have been reported [9, 10]. Similarly, we observed prolonged neuromuscular blockade at the end of the procedure requiring a second administration of sugammadex. In the series of Footitt et al [3], a variety of NMBAs were used, including atracurium, pancuronium, rocuronium and suxamethonium. Interestingly, no significant intra‐operative adverse events were attributable to anaesthetic techniques, and no episodes of prolonged neuromuscular blockade have been reported. However, only one patient received rocuronium. Any patient with MD should have appropriate dosing of NMBAs and adequate monitoring of the neuromuscular blockade.

Our balanced anaesthetic included the administration of different agents together to avoid sole reliance on propofol for general anaesthesia maintenance and to reduce the likelihood of PRIS. Processed electroencephalogram monitoring is helpful in safely minimising the dose of hypnotic agents required. In addition, patients with MD are considered at risk for malignant hyperthermia [7, 8]. Some therefore advise that volatile anaesthetic agents and suxamethonium should not be used. However, data regarding this issue are contradictory, and we report here the safety of sevoflurane administration. In the series of Footitt et al [3], sevoflurane use was frequently reported (21 times in 58 episodes) with no episodes of malignant hyperthermia. Moreover, the excretion of modern volatile anaesthetic agents does not rely on hepatic and renal elimination pathways, which may be compromised in patients with MD.

Finally, analgesics and agents that prevent postoperative nausea and vomiting should be used precociously in these patients. POLG mutations account for various major syndromes associated with epilepsy, such as Alpers–Huttenlocher syndrome, myocerebrohepatopathy spectrum and myoclonic epilepsy myopathy sensory ataxia. It may therefore be wise to avoid drugs that lower the seizure threshold such as nefopam and droperidol. Because the patient had no history of convulsive disorders, we used nefopam as a part of the multimodal analgesia given. Droperidol should also be avoided in POLG patients presenting with parkinsonism, as well as in the case of prolonged QT interval.

This report describes the peri‐operative care of a patient undergoing robotic Heller‐Dor myotomy for the treatment of POLG‐related oesophageal pseudo‐achalasia. This suggests the safety of short‐term application of low‐dose propofol in POLG patients benefiting from sevoflurane‐based balanced general anaesthesia with multimodal monitoring. However, patients presenting with POLG mutations should have appropriate dosing of NMBAs and adequate monitoring of residual neuromuscular blockade.