1 INTRODUCTION

Inflammation, tissue injury, and nerve injury all contribute to postoperative pain mechanisms.¹ Opioids are efficacious but many patients suffer from side effects including postoperative nausea and vomiting (PONV) and respiratory depression.^1-4 These can subsequently increase perioperative morbidity and lengthen hospital stay. Intravenous ketamine and alpha agonists are common analgesic adjuncts used in the perioperative setting, however, adverse effects including hallucinations, hemodynamic instability, and excessive sedation can limit utility in some patients. Intravenous infusion of lidocaine, an amide-type local anesthetic, is an emerging technique and is increasingly used for its anti-neuropathic, anti-hyperalgesic, and anti-inflammatory actions.¹ Lidocaine is on the essential drug list of World Health Organization for both its multiple anesthetic and anti-arrhythmic indications and is considered efficacious, versatile, readily available, safe, and cost-effective.^{5, 6} Through use-dependent inhibition of individual voltage-gated sodium channels (Na_v), and thus nerve fiber impulse generation, subsequent signaling transmission to and within the central nervous system is impeded.^{7, 8} Lidocaine can interact with multiple targets, but its primary clinical site is the local anesthetic binding site on the 9 mammalian Na_v isoforms, with subtypes Na_v1.3, Na_v1.7, Na_v1.8, and Na_v1.9 key players in nociceptive signaling. Na_v1.7 channels are of particular importance as they are expressed in peripheral terminals of sensory neurons, within the dorsal route ganglion and on the sensory afferents in the superficial laminae of the spinal cord. They are also found in trigeminal and sympathetic ganglion neurons and visceral sensory neurons.⁹

The half-life of lidocaine is 90–120 min, however, the analgesic benefit of intravenous lidocaine commonly appears to persist the following cessation of the infusion.⁷ This is thought to be due to reduction in the release of inflammatory cytokines and complement thereby reducing peripheral and central sensitization and modulating a pain windup effect.^{7, 10, 11} Intravenous lidocaine also has the potential benefit of attenuating intraoperative sympathetic responses to surgery, limiting airway reactivity, and also reducing anesthetic requirements.^{10, 12-15} It may also allow for earlier postoperative mobilization.¹⁵ In adults, despite off-label use, intravenous lidocaine is considered a simple intervention and it carries minimal risk when correctly administered.^{11, 16} It is particularly useful when the regional local anesthetic techniques are not feasible.

Previous systematic reviews and meta-analyses on perioperative intravenous lidocaine in adult patients have demonstrated a reduction in pain scores up to 24 h postoperatively, reduced incidence of ileus and time to first bowel activity, reduced rates of PONV, as well as shortening hospital length of stay.^{1, 13, 17-24} Most of this data is from abdominal surgery with suggestions of differing analgesic efficacy and gastrointestinal impact depending on specific surgical procedure.¹⁰ For example, open abdominal surgeries appear to receive a greater pain inhibition response and reduced rates of ileus compared with the laparoscopic surgery.^{10, 13} There is also increasing evidence of the efficacy in spinal surgical patients however there have been contrasting results with one study failing to find analgesic benefits.^{24, 25} Intravenous lidocaine has been shown to be beneficial outside the operating theater in the treatment of pediatric opioid refractory cancer pain, headaches, and neuropathic pain states.^{7, 26} While no meta-analysis or systematic reviews are available in the pediatric sphere, clinical benefits of perioperative intravenous lidocaine have been demonstrated in smaller individual studies and these are summarized in Table 1. Within this educational review, we assess the perioperative pediatric use of intravenous lidocaine and aim to and provide guidance on its specific implementation in this setting and also review some of the history of its early use.

2 CLINICAL BENEFITS OF PERIOPERATIVE INTRAVENOUS LIDOCAINE

Included as an early report prior to 2000, the first described use of intravenous lidocaine as an analgesic technique in children was by Wallace et al (1997) who conducted an open-label study in five children (4–7 years of age) with neuroblastoma who were undergoing immunotherapy with anti-GD2 antibody. Children were given lidocaine (2 mg/kg) or morphine (0.1 mg/kg) over 30 min prior to the commencement of therapy, followed by an infusion of lidocaine (1 mg/kg/h) or morphine (0.05–0.1 mg/kg/h). The infusion was maintained for the duration of the therapy and discontinued 2 h after its completion (7 h) or when the pain score had decreased to 2 or less. Breakthrough pain (pain score >2) was managed with IV morphine (0.05 mg/kg) every 2 h or as needed. Pain scores were ascertained using the faces scale and breakthrough morphine consumption was measured. Investigators found no significant differences in pain scores or breakthrough morphine consumption between the morphine and lidocaine subjects. However, three subjects required continuation of morphine infusion until day 4 of therapy (pain score persistently >2) compared with no subjects requiring continuation of lidocaine infusion beyond the 7 h protocol.²⁸ Since this time, intravenous lidocaine for the purpose of perioperative analgesia has been used in a variety of contexts.

3 DOSING AND PHARMACOKINETICS OF INTRAVENOUS LIDOCAINE

At clinically relevant infusion doses (1–2 mg/kg/h) intravenous lidocaine usually results in the plasma concentrations that remain below toxic levels of 5 mcg/ml with lidocaine serum levels at 1–2 mcg/ml appearing to effectively impact postoperative pain.^{6, 11} It also possesses a steep dose-response curve with minimal dose increases resulting in the large increases in pain relief.⁷ About 90% undergoes hepatic elimination by cytochrome P450 and less than 10% is excreted renally unchanged.¹⁵ After a bolus injection or up to a 12 hour infusion, the half-life is around 100 min and this remains linear.¹⁵ There may be age-related changes in the pharmacokinetics but children older than 6 months appear to distribute and eliminate intravenous lidocaine in the same manner as adults and therefore, the applicability of weight-based adult dosing regimens to healthy children appears reasonable.^{16, 34} Due to an increase in absolute volume of distribution as a result of higher body weight, loading doses should be calculated based on total body weight and continuous infusion rates on ideal body weight.³⁵ The most commonly used dose of intravenous lidocaine in adults is a 1.5 mg/kg bolus and 1 to 1.5 mg/kg/h infusion whereas in the absence of a loading dose, it takes more than 60 min and up to 8 h to achieve a therapeutic plasma steady state.^{2, 6, 35} Notably, higher adult dosing (1.5 mg/kg bolus followed by 2 mg/kg/h infusion) has demonstrated plasma concentrations well below the toxic level (5 mcg/ml) even after 48 h of dosing.^{1, 10} Furthermore, in a past study in 10 un-anesthetised volunteers, a rate of 30 mg/kg/h of intravenous lidocaine without a bolus was run for over 10 min. This resulted in muscle fasciculations and twitching with one convulsion episode and even with this 30 times dose, plasma levels at the end of the infusion were 5.29 ± 0.55 mcg/ml further indicating that conventional doses are likely to remain under toxic levels.³⁶ The regimens reported in the pediatric studies use a similar regimen from 1 mg/kg to 1.5 mg/kg bolus with subsequent infusion rates ranging from 1.5 mg/kg/h to 2 mg/kg/h of varied duration. In one study, patients dosed at the higher range (1.5 mg/kg bolus and 2 mg/kg/h) plasma levels were found to be below the toxic concentration, with mean lidocaine plasma concentration measured at 3.91 mcg/ml at the cessation of the infusion (41 min mean time).³¹ Other studies which measured levels after initial bolus, at completion of surgery, 6 h postoperatively and subsequent morning also demonstrated ranges within 1.2–4.2 mcg/ml or less than 4 mcg/ml at a 1.5 mg/kg bolus regimen with a 1 mg/kg/h infusion when used for upto 6 h postoperatively.^{15, 25, 32} This work suggest that these regimens are unlikely to lead to toxic levels in the healthy pediatric patients without co-morbidities. This review will also discuss dosing and use in special populations subsequently. The ideal duration of infusion for optimal effect remains unknown but the intraoperative period is probably the most important for analgesia with postoperative continuation usually dependent on monitoring resources in individual institutions.¹ If infusions are continued postoperatively or used outside of the operating theater, resources such as a higher level of nursing care and physiological monitoring for example in a unit able to provide continuous telemetry such as postoperative care unit or intensive care unit are recommended.^{4, 7, 26}

4 TOXICITY AND SIDE EFFECT PROFILE OF INTRAVENOUS LIDOCAINE

Safety data remains limited in children with systemic analysis of the adult literature demonstrating uncommon and often clinically insignificant adverse effects.¹ This is potentially a result of vigilant monitoring within the operative setting and a short elimination half-life.^{1, 3} It is very rare to encounter hypersensitivity or idiosyncratic reactions.⁶ Plasma levels of greater than 5 mcg/ml are thought to be when mild systemic adverse effects start to occur.³⁷ Typically, the first minor symptoms include perioral numbness, metallic taste, tongue paraesthesia, dizziness, tinnitus, and blurred vision.³⁵ Higher serum levels of free local anesthetic potentially cause increasing sodium channel blockade in the central nervous and cardiovascular systems leading to local anesthetic systemic toxicity (LAST).³⁷ This manifests as seizures, myocardial depression, severe cardiac arrhythmias possibly leading to cardiac arrest. Despite its very low incidence, with less than 100 reported cases in the last 30 years, LAST can be a result of any local anesthetic agent, with bupivacaine use the most often associated with cardiotoxicity.³⁸ Most data from lidocaine is sourced from its use as an antiarrhythmic agent with plasma concentrations between 1.4 mcg/ml and 6 mcg/ml considered safe.³⁵ Furthermore, lidocaine is considered to have a more favorable ratio compared with other local anesthetic agents as demonstrated by the so called CC/CNS ratio (ratio of dose causing cardiovascular collapse to the dose causing seizures). Bupivacaine has a CC/CNS ratio of 2.0 compared with 7.1 for lidocaine indicating that the progression from CNS signs and symptoms to cardiovascular collapse can occur more readily with bupivacaine than with lidocaine.³⁹ Reassuringly, toxic symptoms generally appear to develop in a reasonably sequential and thus predictable manner based on serum lidocaine levels with major neurotoxicity occurring at plasma levels of about 15 mcg/ml and cardiotoxicity at plasma levels of 21 mcg/ml, however, this is not always the case with a spectrum of presentation.^{6, 35, 38} For example, intraoperative use might remove the signs of central nervous system toxicity, it is prudent to note that cardiac toxicity may be the first warning sign of LAST.³⁸ In addition, susceptible patients may exhibit LAST with a lower dosing regimen with factors that influence the plasma concentration of free drug including acid–base status, hypercapnia and hypoxia.⁶ Finally, risk of LAST or other adverse events is higher in the context of bolus dosing which is standard practice prior to commencing lidocaine infusions given steady state is only achieved after 8 h of infusion without a bolus.³ Clinical vigilance and a high threshold for LAST diagnosis is seen as an essential step in improving associated adverse events with local anesthetic agents. Furthermore, given local anesthetic toxicity is additive, intravenous lidocaine should not be used at the same time as, or within the period of action of other major local anesthetic interventions, for example, not within 4 h of nerve block administration.³

The incidence of adverse effects related to postoperative intravenous lidocaine for analgesia has been studied in spinal, orthopedic, and urology patients.⁴ In the absence of a bolus, the infusions were run between 0.8–1.2 mg/kg/h with the majority (80%) commencing intra-operatively (mean duration 30.6 ± 22 h). A total of 50 patients (aged 2–17 years) were analyzed with 24% of infusions associated with adverse effects which were primarily neurologic in nature including limb paraesthesia and visual disturbances. The average onset time of adverse events was 16.2 h (±15.2), of those tested, none had levels over 5 mcg/ml.⁴ Symptoms were deemed to be mild with quick resolution on either cessation or dose reduction. There were no reported cardiovascular events. In another study of 20 participants a higher dose regimen was used (1.5 mg/kg bolus and 2 mg/kg/h) for pediatric colonoscopy with a total dose of 1.58 mg/kg. Only one patient suffered dizziness and vomiting postprocedure which was mild, self-limiting and did not require intervention.¹⁴ The mean length of these procedures however was quite short (15.25 min SD 1.81). In a study which compared intravenous lidocaine to placebo in multilevel spinal surgery, a 1.5 mg/kg bolus dose followed by 1 mg/kg/h was used until 6 h postoperatively. With a mean operation time of 260 min (170–285), the only adverse effects of lidocaine observed were transient skin sensory disturbances at the site of drug administration.²⁵ In the mixed adult and adolescent group, no patient receiving lidocaine reported subjective symptoms of LAST but plasma levels were not recorded.³³ None of the other pediatric studies (listed in Table 1) reported signs of serious side effects during or in the 24 h following administration including neurological impairment or circulatory disturbance and of those that measured plasma levels, toxic serum concentrations were all less than 5 mcg/ml.^{7, 12, 16, 25, 30, 31}

Outside of the perioperative setting, intravenous lidocaine has also been used in cancer patients with opioid-refractory pain. Infusions, with 10 patients receiving a 1 mg/kg loading dose, were titrated up to 3 mg/kg/h to maximal pain relief or development of intolerant side effects including paraesthesia and visual disturbances. All of these symptoms resolved with either no change in range or a reduction in infusion rate.⁷ Of note, in those that experienced adverse effects and had measured serum lidocaine concentrations all were below the supposed toxic level of 5 mcg/ml level. In a cohort of 50 patients receiving postoperative infusions, without a bolus, at a mean starting rate 0.8 mg/kg/h (SD 0.2 mg/kg/h) and maximum of 0.9 mg/kg/h (SD 0.4 mg/kg/h) for a mean duration of 30.6 h(SD 22) less than one quarter (22%) were associated with adverse effects with the majority noting twitching, tingling or numbness in the perioral region, hands or feet, visual disturbances, tinnitus. All of these were noted to be mild-to-moderate severity with no severe adverse events noted and quickly resolved. In all of these cases but one, the serum level was less than 5 mcg/ml and in all cases, discontinuation or reduction of dosing resulted in resolution of all adverse effects. The average time to onset was 16.2 h(SD 15.2) perhaps suggesting a longer infusion time is associated with the incidence of adverse effects.⁴

In contrast to this, and primarily demonstrated in an outpatient setting, in a small cohort of patients undergoing intravenous lidocaine for refractory cancer pain, serum lidocaine levels were not correlated to infusion rates with authors postulating that organ function, drug interaction, and comorbid conditions all may impact the serum concentration in individual patients.⁷ In another group of 15 patients, undergoing a total of 58 outpatient infusions at a rate of 2.4–3.6 mg/kg/h (no bolus), for headache and neuropathic pain, the majority (79%) described side effects.²⁶ These included dizziness, numbness/tingling, and visual changes. There appeared to be no correlation between the rate of infusion and occurrence of side effects and cessation or dose reduction of the infusion was rarely required. No severe symptoms (defined as altered conscious state, seizures, or ECG changes) were experienced. No symptom was deemed significant enough to lead to assessment of serum lidocaine concentrations. These heterogenous findings highlight the need for rigorous dosing studies.

5 SPECIAL POPULATIONS

Within special populations, there is unpredictability with respect to plasma lidocaine concentrations based on the interactions between patient factors and drug pharmacokinetics which lead to important clinical consideration.³ For example, clearance is limited in those with heart failure, hepatic or renal impairment which can result in accumulation of the metabolite monoethylglycinexylidide leading to seizures as a direct and secondary effect.^{3, 6, 11} In addition, those who have immature metabolic clearance, for example, neonates would be at increased risk of drug and metabolite accumulation.⁷ Furthermore, the alpha-1-acid glycoprotein (AAG) fraction is lower in neonates and infants, with the AAG at birth about half of that of an adult, resulting in a comparative higher unbound fraction of lidocaine and an increased elimination half-life and thus increased risk of accumulation with continuous infusions.^{35, 39} In fact, LAST has the highest incidence in infants less than 6 months of age and is associated with bolus dosing.³⁷

Those with mitochondrial diseases and diseases with carnitine deficiency also require additional caution. There are reports of cardiac dysrhythmias, due to inhibition of carnitine-acylcarnitine translocase, in vitro, particularly with bupivacaine and less with lidocaine.^{40, 41} The risk–benefit decision to use intravenous lidocaine therefore requires clinicians to carefully consider intravenous lidocaine pharmacology in the context of individual patient conditions and comorbidities.

6 CONCLUSION

While larger scale trials, inclusive of safety data and a particular focus on pediatric pharmacokinetics, are warranted, the presented data has demonstrated potential positive effects of perioperative intravenous lidocaine. It appears to be well-tolerated and benefits include improved pain, reduced postoperative opioid use, sparing of anesthesia agents, enhanced gastrointestinal function, and stress response attenuation albeit with a variable degree of benefit. Based on the current literature, and in those whom regional or neuraxial analgesia is not appropriate, the use of perioperative intravenous lidocaine in pediatric patients undergoing open abdominal surgery and spinal surgery appears to be reasonable. Concluding from available evidence, a suggested protocol is dosing of 1.5 mg/kg bolus on induction followed by 1 mg/kg/h (ideal body weight) infusion rate for the intraoperative period in children aged over 2 years without comorbid disease. Although the optimal infusion time remains unknown, the intraoperative infusion appears to be the most efficacious component particularly based on the potential risk–benefit ratio and also allows for constant monitoring. Side effects may occasionally occur but are mostly mild and resolve with discontinuation or dose reduction. If utilized perioperatively, intravenous lidocaine should be the sole local anesthetic intervention.

FUNDING INFORMATION

BSvUS is partly funded by the Stan Perron Charitable Foundation and has an NHMRC Investigator Grant (APP 200932). No other conflicts declared for authors.

CONFLICT OF INTEREST

BSvUS is a section editor for the Pediatric Anesthesia.