Pain Selective Anesthesia for Labor Mobility (PSALM Study): a Randomized Controlled Trial of Combined Lidocaine - Chlorprocaine for Labor Epidural Analgesia

Overview

The current "gold standard" epidural analgesia for labor is associated with unwanted side effects, such as motor block (affecting ambulation and the ability to push) and sympathetic blockade (causing maternal hypotension). Here, using direct clinical application of preclinical findings, the investigator assess a novel method for prolonged pain-selective anesthesia during labor avoiding motor block, sympathetic block or other neural side effects. The approach is based on the pain-selective anesthesia the investigator developed in preclinical models, where the exclusive presence of TRPV1 channels on pain fibers (nociceptors) was exploited. The investigator used polarized, membrane-impermeable local anesthetics, together with the co-administration of direct TRPV1 agonists to open TRPV1 channels, so allowing these impermeable local anesthetics to gain selective access to nociceptors. Early preclinical studies used capsaicin as the TRPV1 agonist and QX-314 as the polarized local anesthetic, but neither of these drugs are appropriate for clinical use. Here the investigators use lidocaine as the TRPV1 activator and chloroprocaine as the polarized local anesthetic and hypothesize that this co-administration will elicit pain-selective (nociceptor-specific) anesthesia for labor without side effects. Positive findings in this study will be the first evidence in humans of pain-selective local anesthesia and will be the harbinger for future studies of systemic and regional administration of these drugs for pain-selective local anesthesia for chronic and other pain states.

The holy grail of pain management is providing highly effective pain relief without any side effects. The most common analgesic stratagems include a) systemic analgesia or b) local anesthesia using either peripheral nerve blockade or epidural/spinal anesthesia. Current approaches are not specific for pain blockade and are limited by side effects. The overall goal of the study is to achieve pain-specific local anesthesia, with selective nociceptive block, without either motor block, sensory block or sympathetic block. While the implications apply to many clinical pain states, here the team focuses on labor analgesia, where this strategy will allow mothers to feel no pain, while being able to feel the need to push, retaining the ability to push and to ambulate in labor, and avoiding hypotension. This goal is achieved by targeting nociceptors directly via a novel neural target. The team developed a novel platform for nociceptor-selective local anesthesia. Using the transient receptor potential vanilloid receptor-1 (TRPV1 channel) expressed only by pain fibers, as an innate nociceptor-targeted drug delivery system for selectively introducing membrane-impermeable local anesthetics into nociceptors. TRPV1 is a transmembrane ion channel protein activated by noxious stimuli5 and by chemical irritants such as capsaicin. Capsaicin was used as the TRPV1 agonist and a lidocaine derivative, QX-314, as a polarized, membrane-impermeable, and hence generally ineffective local anesthetic. The preclinical co-investigator (Prof Alex Binshtok) previously showed that the activated TRPV1 channel is large enough to allow entry of QX-314 into nociceptors, so blocking their electrical activity in vitro, while not affecting the excitability of non-nociceptor neurons not expressing TRPV1 channels. Similarly, the co-injection of QX-314 and capsaicin in-vivo selectively abolished the response of animals to noxious stimuli without any detectable motor or tactile deficits, whereas the injection of either capsaicin or QX-314 alone did not produce any antinociceptive effect. Importantly, the team demonstrated that QX-314 permeates the human orthologue of TRPV1, leading to the inhibition of human sodium channel isoforms. The team proposed that this approach for selectively shuttling sodium channel blockers to nociceptive neurons through nociceptor-specific TRPV1 channels could be exploited clinically to produce long-lasting, pain-selective, regional analgesia while preserving motor and autonomic function, for example during childbirth or after surgery. Furthermore, the team proposed that either systemic or regional administration of these combinations could also be used to treat other clinical conditions such as acute postoperative pain, intractable cancer pain, inflammatory or neuropathic pain and itch. However, the clinical application of the "capsaicin-QX-314" approach encountered two main obstacles, both of which this current study is designed to address. 1. Capsaicin-induced pain: The activation of TRPV1 channels by capsaicin leads to severe "injection" pain before QX-314 enters nociceptors to induce analgesia. The team solves this problem by activating TRPV1 channels with lidocaine instead of capsaicin. The team and others have demonstrated that lidocaine activates human TRPV1 channels at 5mM doses (or 0.12%). The team co-applied QX-314 with lidocaine as the TRPV1 agonist instead of capsaicin and showed that lidocaine-induced activation of TRPV1 was sufficient to allow QX-314 to enter nociceptive neurons. Co-application of QX-314 and lidocaine induced a prolonged differential nociceptive block without "injection" pain. 2. QX-314 neurotoxicity: Several groups have demonstrated that QX-314 is neurotoxic in rats. Here the team avoids neurotoxicity by replacing QX-314 by standard local anesthetics with established safety records. The team's approach was based on their earlier observations showing that activation of TRPV1 channels by capsaicin prolongs the duration of nociceptive blockade by other local anesthetics, by facilitating entry for their polarized (membrane-impermeable) forms selectively into nociceptive neurons, thus leading to differential pain blockade with minimal motor block. In the current approach the team uses the local anesthetic chloroprocaine, with the highest pKa (9.1) of all the local anesthetics in clinical use, in order to exploit the relatively high polarized versus non-polarized form for this drug, which makes it a relatively membrane-impermeable local anesthetic. It is important to note that chloroprocaine elicits rapid onset anesthesia in clinical practice despite its high polarization and relative membrane impermeability. This is because the current clinical dose is particularly high (3%), which creates the diffusion gradient required to drive rapid onset. Despite these high doses, toxicity is avoided because chloroprocaine is rapidly mobilized by cholinesterase, a feature of chloroprocaine that allows clinicians to use such high doses; accordingly, this drug has 70 years of established clinical safety in both spinal and epidural anesthesia; The "lidocaine-chloroprocaine" strategy is a complete reverse of the strategy using high dose chloroprocaine to force the drug indiscriminately across neuronal cell membranes. By co-administering chloroprocaine with a TRPV1 agonist (capsaicin or lidocaine) the team can shuttle polarized chloroprocaine selectively into nociceptors while using sub-clinical doses, in order to elicit nociceptor-specific anesthesia, but without other neural side effects. Neurones lacking TRPV1 channels (e.g. sensory, sympathetic and motor neurons) will only be exposed to low doses of the relatively minor non-polarized fraction of chloroprocaine, as it is only the non-polarized form that is able to cross cell membranes. The team's preclinical results show that activation of TRPV1 channels together with anesthetic doses of chloroprocaine leads to enhanced inhibition of sodium currents in nociceptive neurons in vitro and to differential nociceptor-specific anesthesia in vivo. Moreover, the team shows that sub-anesthetic doses of chloroprocaine, when combined with TRPV1 agonists, leads to electrophysiological and behavioral manifestations of profound and prolonged nociceptor-specific anesthesia, with no other neural effects. Importantly, the team shows that chloroprocaine neither activates TRPV1 channels nor induces neurotoxicity, suggesting lack of neurotoxic effect when combined with a TRPV1 agonist. Furthermore, the team has demonstrated that while low concentrations of chloroprocaine do not inhibit sodium currents in nociceptive neurons when administered alone, identical doses do cause significant inhibition of sodium currents when co-administered with capsaicin. Injection near the sciatic nerve of chloroprocaine in sub-anesthetic doses, together with the TRPV1 channel activator capsaicin, produced prolonged analgesia in vivo, without motor deficits. Importantly, this peri-sciatic injection of chloroprocaine with capsaicin did not induce any injection pain or any long-term changes in the sensitivity of the animals to noxious stimuli, suggesting that the "capsaicin-chloroprocaine" approach is devoid of neurotoxic effects. Finally, the team shows the co-application of lidocaine (instead of capsaicin), together with chloroprocaine, produced inhibition of the response to noxious stimuli that was significantly longer than the sum of the effects of lidocaine or chloroprocaine when administered alone, suggesting synergism. Importantly, the co-administration of lidocaine and chloroprocaine did not prolong the motor blockade induced by lidocaine alone, suggesting that the "lidocaine-chloroprocaine" combination only affects nociceptive neurons. Notably, the motor block was short lasting (about 30 min) and was substantially shorter than the nociceptive block. While both chloroprocaine and lidocaine are in routine clinical use for 70 and 80 years respectively, they have never been described in co-administration before. The preclinical data supports using chloroprocaine as a substitute for QX-314 for co-administration with lidocaine as the TRPV1 agonist, for safe and efficient pain-selective anesthesia without side effects. Positive findings in this study will be the first evidence in humans of pain-selective local anesthesia and will be the harbinger for future studies of regional and/or systemic (intravenous or oral) administration of these drug combinations for pain-selective local anesthesia for chronic and other pain states. Patients and epidural insertion (described below) and randomization, blinding, ethics and multicenter sites apply to all patients in both Aim 1 and Aim2. In all studies in this proposal, epidural is performed only after patient consent is obtained and once VAPS is measured during a uterine contraction at baseline. Epidural analgesia is performed exactly as in routine cases; the patient is placed in the sitting or lateral position; the epidural is inserted at the L3/4 interspace using loss of resistence to saline (limited to 2mL) and a multiport epidural catheter is threaded 3-5 cm into the epidural space. Aim 1: the team uses an isobolographic study to determine if there is synergy between lidocaine and chloroprocaine. Determining the EC50: The endpoint in these studies is the EC50, or median local anesthetic concentration (MLAC), which is a robust pharmacodynamic measure as it lies on the linear portion of the dose-response curve. The most widely used approach to measure EC50 for analgesic drugs is based on the Dixon-Massey method and was introduced to anesthesia research by Columbia (statistical co-investigator in this proposal) in the 1990's. This uses an up-down sequential study design, where success or failure in one patient leads to respectively lowering or raising the dose for the next patient in that drug group. The methodology has since been used extensively in anesthesia research, including by us. In this study, 20 mL of the study drug is administered as an incremental test dose over the course of 5 minutes. No other test doses are used. Determining analgesia success: Assessment is made during a uterine contraction ≥20 minutes after completing the epidural bolus dose. The team uses a 5-point Likert scale in English, Hebrew and Arabic. Success is defined as "I am now completely pain free" or "there is great reduction in pain, it is now very tolerable". All other options on the Likert scale are considered a failure. A failure is defined as a "reject" if one of three mitigating conditions exist: i) vaginal exam at the time of analgesic failure shows cervical dilation ≥8cm; ii) epidural rescue medication of up to 15 ml of 1% lidocaine fails to relieve labor pain within a further 20 min; iii) decision to progress to cesarean delivery is made prior to the 20 min assessment. Based on success/failure, the epidural dose for the next patient in the group is reduced/increased by 0.05%wt/vol. Following "reject", the next patient receives the same dose. Isobolographic analysis is a well-established pharmacological method to assess whether drug interactions are synergistic. Using the EC50 as the common endpoint, and calculate the interaction index26, γ = a/A + b/B, where A and B are the EC50 of drugs A and B alone, and (a,b) is a drug combination that elicits that same effect (i.e. the EC50). If γ=1 the combination is additive; if γ\<1 the combination is supra-additive or synergistic. The isobolographic approach and interaction index is an efficient and reliable measure of synergism and has been utilized in many pharmacological and analgesic studies, including previous research by Columbia. Drug groups vary. The first stage (Aim 1a) randomizes patients to either epidural lidocaine alone or epidural chloroprocaine alone and determines the EC50 for each. The second stage (Aim 1b) randomizes patients to one of three combined lidocaine-chloroprocaine groups: fixed lidocaine dose group: 50% of lidocaine EC50 (from Aim 1a) with variable (up-down sequential) dose of chloroprocaine; fixed chloroprocaine dose group: 50% of chloroprocaine EC50 (from Aim 1a) with a variable (up-down sequential) dose of lidocaine; bivariate fixed ratio group: where the lidocaine-chloroprocaine ratio is determined by the ratio of their individual EC50s (from Aim 1a), and where the fixed-ratio is maintained while using the up-down sequential method to measure the EC50 of the combined dose. The Dixon-Massey method uses nominal sample sizes, but typically 30 subjects are needed in order to obtain an EC50 with narrow 95% confidence intervals. Hence 30 subjects will be needed in each of the five drug groups (total 150 patients). EC50 is estimated using Dixon-Massey method (primary) and Wilcoxon-Litchfield probit regression (secondary). The interaction index is calculated for each of the 3 combination groups; these values are then averaged (±95%CI) to give the overall interaction index (the primary outcome for Aim 1). After the determination of success/failure, and after rescue medication if relevant, patients receive the hospital standard protocol for maintenance of epidural analgesia. There is no follow up. Aim 2: The team describes a randomized controlled trial (RCT) of three local anesthesia groups: lidocaine (L), chloroprocaine (C), and a lidocaine-chloroprocaine combination (L-C). When comparing side effects of analgesic medication it is critical to compare equianalgesic doses; ensuring equianalgesia in two ways: i) epidural drug concentrations are all based on the EC50 from Aim 1 (Groups L and C use the EC50 for lidocaine and chloroprocaine from Aim 1a, and Group L-C uses the EC50 of the bivariate fixed ratio group from Aim 1b); ii) self-administered epidural PCA is used for all patients. Epidural medication: The initial bolus is 20mL for each study group, with an additional 50µg fentanyl administered as an incremental test dose over 5 min. No other test dose is used. Maintenance dose: Pharmacist prepares 250mL bags of lidocaine, chloroprocaine and the combination based on their EC50s from Aim 1; bags are number coded and blinded. Fentanyl 500µg is added (2µg/mL) to each bag as in routine care. Epidural solution is administered as PCA (0 mL/hr background infusion, 10mL bolus dose, 10min lockout time, 60mL maximum/hour). The primary endpoint is the modified Bromage motor score. Secondary outcomes thermal imaging of feet and hands, maternal blood pressure, sensory assessment using to ice sensation, and ability to walk in labor and to push in delivery. Outcome measures to confirm equianalgesia are: i) VAPS (0-100mm, measured on an ungraded 10cm line where anchors are "no pain" and "worst pain imaginable"), ii) PCA analgesic requirement, and iii) need for rescue analgesia. Timing of endpoint measures: VAPS is assessed at the peak of uterine contraction; modified Bromage score, thermal imaging, ice sensation, and blood pressure are tested between uterine contractions. Assessments are made twice at baseline prior to the epidural, and at approximately 10 min intervals for the first 30 mins, and at 15 min intervals thereafter until the first epidural top-up request (administered via PCA). After delivery, anesthesia requirement is measured from the PCA pump, and walking ability in labor and pushing ability in delivery are recorded by the blinded midwife and by the patient.