Botulinum Toxin: An Update on Pharmacology and Newer Products in Development.

Since its introduction as a treatment for strabismus, botulinum toxin (BoNT) has had a phenomenal journey and is now recommended as first-line treatment for focal dystonia, despite short-term clinical benefits and the risks of adverse effects. To cater for the high demand across various medical specialties, at least six US Food and Drug Administration (FDA)-approved formulations of BoNT are currently available for diverse labelled indications. The toxo-pharmacological properties of these formulations are not uniform and thus should not be used interchangeably. Synthetic BoNTs and BoNTs from non-clostridial sources are not far from clinical use. Moreover, the study of mutations in naturally occurring toxins has led to modulation in the toxo-pharmacokinetic properties of BoNTs, including the duration and potency. We present an overview of the toxo-pharmacology of conventional and novel BoNT preparations, including those awaiting imminent translation from the laboratory to the clinic.

Assessment of Pharmacology, Safety, and Metabolic activity of Capsaicin Feeding in Mice.

Capsaicin (CAP) activates transient receptor potential vanilloid subfamily 1 (TRPV1) to counter high-fat diet (HFD)-induced obesity. Several studies suggest that CAP induces the browning of white adipocytes in vitro or inguinal white adipose tissue (iWAT) in vivo. However, there is a lack of data on the dose-response for CAP to inhibit HFD-induced obesity. Therefore, we first performed experiments to correlate the effect of various doses of CAP to prevent HFD-induced weight gain in wild-type (WT) mice. Next, we performed a subchronic safety study in WT mice fed a normal chow diet (NCD ± CAP, 0.01% in NCD) or HFD ± CAP (0.01% in HFD) for eight months. We analyzed the expression of adipogenic and thermogenic genes and proteins in the iWAT from these mice, conducted histological studies of vital organs, measured the inflammatory cytokines in plasma and iWAT, and evaluated liver and kidney functions. The dose-response study showed that CAP, at doses above 0.001% in HFD, countered HFD-induced obesity in mice. However, no difference in the anti-obesity effect of CAP was observed at doses above 0.003% in HFD. Also, CAP, above 0.001%, enhanced the expression of sirtuin-1 and thermogenic uncoupling protein 1 (UCP-1) in the iWAT. Safety analyses suggest that CAP did not cause inflammation. However, HFD elevated plasma alanine aminotransferase and creatinine, caused iWAT hypertrophy and hepatic steatosis, and CAP reversed these. Our data suggest that CAP antagonizes HFD-induced metabolic stress and inflammation, while it does not cause any systemic toxicities and is well tolerated by mice.

Implementation of a new strategy to improve the peri-operative management of neuromuscular blockade and its effects on postoperative pulmonary complications.

Inappropriate dosing of neostigmine for antagonism of neuromuscular blockade has been associated with postoperative pulmonary complications. We evaluated the effects of a quality improvement initiative tailored to optimise the use of neostigmine in antagonising neuromuscular blockade on postoperative pulmonary complications, costs and duration of hospital stay. The quality improvement initiative consisted of: a reduction in available neostigmine aliquot sizes; a cognitive aid; an educational component; and a financial incentive for the intra-operative documentation of train-of-four measurement before administration of neostigmine. We conducted a pre-specified analysis of data obtained in our quality improvement study. Additional analyses were conducted in a propensity-matched cohort. An interrupted time series design was used to discriminate between the intervention and a counterfactual scenario. We analysed 12,025 consecutive surgical cases performed in 2015. Postoperative pulmonary complications occurred in 220 (7.5%) of 2937 cases pre-intervention and 568 (6.3%) of 9088 cases post-intervention. Adjusted regression analyses showed significantly a lower risk of postoperative pulmonary complications (OR 0.73 (95%CI 0.61-0.88); p = 0.001), lower costs (incidence rate ratio 0.95 (95%CI 0.93-0.97); p < 0.001) and shorter duration of hospital stay (incidence rate ratio 0.91 (95%CI 0.87-0.94); p < 0.001) after implementation of the quality improvement initiative. Analyses in a propensity-matched sample (n = 2936 per group) and interrupted time series analysis (n = 27,202 cases) confirmed the findings. Our data show that a local, multifaceted quality improvement initiative can enhance the quality of intra-operative neuromuscular blocking agent utilisation, thereby reducing the incidence of postoperative pulmonary complications.

Longitudinal neurophysiological assessment of intramuscular type-A botulin toxin in healthy humans.

The aim of this study is to assess the neurophysiological abnormalities of type A botulin toxin-infiltrated human muscle, and their evolution over time. Seried cMAP measurements, 3 and 20 Hz repetitive nerve stimulation, EMG, SFEMG over 3 months from toxin injection. Our findings consist in lack of decrement with 3 Hz repetitive nerve stimulation and facilitation with 20 Hz repetitive nerve stimulation; progressive increasing of jitter; early appearance of fibrillations; small and short motor unit action potential in the first 3 weeks, followed by increasing of MUAP amplitude and duration, with polyphasic morphology. Although claimed as highly specific and sensible, neuromuscular junction facilitation is an inconstant finding in human botulism. Therefore, lack of neuromuscular junction facilitation cannot exclude a diagnosis of botulism. Our findings are compatible with a process of acute denervation followed by distal reinnervation, favored by terminal nerve sprouting.

Simulation of the kinetics of neuromuscular block: implications for speed of onset.

BACKGROUND: The onset time for paralysis varies 3-fold among nondepolarizing muscle relaxants. Possible explanations include: (a) pharmacokinetic differences among drugs and (b) buffering of drug molecules by acetylcholine receptors as they diffuse into the neuromuscular junction. Although some pharmacokinetic models consider buffered diffusion, these models do not account for either the high density of receptors or synapse geometry. Here, I used computer simulations to calculate the kinetics of buffered diffusion. The goal was to determine the conditions under which buffered diffusion could account for differences in onset time among nondepolarizing muscle relaxants. METHODS: Monte Carlo simulation was used along with a realistic 3-dimensional model of the rat neuromuscular junction. Simulations determined the time dependence of the number of drug-bound receptors. A 1000-fold range of drug potency was examined. In some simulations, the drug concentration outside the junction was changed instantaneously. In other simulations, the concentration changed according to predictions of pharmacokinetic models assuming time-dependent changes in plasma drug concentration. The rate constant for equilibration of drug between plasma and muscle, keo, was varied between 0.15 and 0.6 min(-1). Twitch amplitude was calculated from receptor occupancy assuming a high safety margin for neuromuscular transmission. Some simulations used a synaptic model with an increased nerve-muscle contact width. RESULTS: Simulations with instantaneous changes in drug concentration at the synapse, indicated that the time to 50% twitch depression (onset time) was 0.1 to 30 seconds and was proportional to drug potency. This corresponds to iontophoretic application of drug to isolated neuromuscular junctions, but is too fast to explain onset times in humans. When pharmacokinetic models were used to calculate the drug concentration outside the synapse, buffered diffusion increased onset times of potent drugs (drugs for which the effective concentration at 50% twitch height is <600 nM). Simulations using keo = 0.6 min(-1) and a model with a 2- to 3-fold wider nerve-muscle contact width indicated that buffered diffusion could account for the differences in clinical onset times among the nondepolarizing muscle relaxants. CONCLUSION: Monte Carlo simulation provides a biophysically appropriate way to incorporate buffered diffusion into pharmacokinetic modeling. The simulations indicated that buffered diffusion could account for differences in onset time among drugs. However, a better understanding of the geometry of the human neuromuscular junction is needed before the magnitude of the effect of buffered diffusion can be quantified.

Evidence-based review and assessment of botulinum neurotoxin for the treatment of adult spasticity in the upper motor neuron syndrome.

Botulinum neurotoxin (BoNT) can be injected to achieve therapeutic benefit across a large range of clinical conditions. To assess the efficacy and safety of BoNT injections for the treatment of spasticity associated with the upper motor neuron syndrome (UMNS), an expert panel reviewed evidence from the published literature. Data sources included English-language studies identified via MEDLINE, EMBASE, CINAHL, Current Contents, and the Cochrane Central Register of Controlled Trials. Evidence tables generated in the 2008 Report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology (AAN) review of the use of BoNT for autonomic disorders were also reviewed and updated. The panel evaluated evidence at several levels, supporting BoNT as a class, the serotypes BoNT-A and BoNT-B, as well as the four individual commercially available formulations: abobotulinumtoxinA (A/Abo), onabotulinumtoxinA (A/Ona), incobotulinumtoxinA (A/Inco), and rimabotulinumtoxinB (B/Rima). The panel ultimately made recommendations on the effectiveness of BoNT for the management of spasticity, based upon the strength of clinical evidence and following the AAN classification scale. While the prior report by the AAN provided recommendations for the use of BoNT as a class of drug, this report provides more detail and includes recommendations for the individual formulations. For the treatment of upper limb spasticity, the evidence supported a Level A recommendation for BoNT-A, A/Abo, and A/Ona, with a Level B recommendation for A/Inco; there was insufficient evidence to support a recommendation for B/Rima. For lower limb spasticity, there was sufficient clinical evidence to support a Level A recommendation for A/Ona individually and BoNT-A in aggregate; the clinical evidence for A/Abo supported a Level C recommendation; and there was insufficient information to recommend A/Inco and B/Rima (Level U). There is a need for further comparative effectiveness studies of the available BoNT formulations for the management of spasticity.

Single-pixel optical fluctuation analysis of calcium channel function in active zones of motor nerve terminals.

We used high-resolution fluorescence imaging and single-pixel optical fluctuation analysis to estimate the opening probability of individual voltage-gated calcium (Ca(2+)) channels during an action potential and the number of such Ca(2+) channels within active zones of frog neuromuscular junctions. Analysis revealed ∼36 Ca(2+) channels within each active zone, similar to the number of docked synaptic vesicles but far less than the total number of transmembrane particles reported based on freeze-fracture analysis (∼200-250). The probability that each channel opened during an action potential was only ∼0.2. These results suggest why each active zone averages only one quantal release event during every other action potential, despite a substantial number of docked vesicles. With sparse Ca(2+) channels and low opening probability, triggering of fusion for each vesicle is primarily controlled by Ca(2+) influx through individual Ca(2+) channels. In contrast, the entire synapse is highly reliable because it contains hundreds of active zones.

Succinylcholine in malignant hyperthermia: evaluation of a novel in vivo model.

INTRODUCTION: Malignant hyperthermia (MH) is a potentially lethal anesthetic complication characterized by muscle hypermetabolism and generalized rigor. The exact mechanism of succinylcholine as an MH trigger cannot be examined in existing in vitro models. Therefore, a novel in vivo model was used to examine the metabolic response to succinylcholine. METHODS: With institutional review board approval, 6 MH susceptible (MHS) and 6 MH non-susceptible (MHN) pigs were anesthetized with hemodynamic and systemic metabolic monitoring. Microdialysis catheters were placed intramuscularly. After equilibration, succinylcholine, halothane, and Ringer solution were injected. Lactate was measured in the dialysate and statistically analyzed by Mann-Whitney U-test (significance level P < 0.05). RESULTS: Hemodynamic and systemic metabolic parameters were not different between the groups throughout the experiment. In the MHS pigs, halothane induced a significant increase of lactate. In MHN pigs, no substance induced a reaction. CONCLUSIONS: Halothane, but not succinylcholine, induced a hypermetabolic reaction in this model. Therefore, the role of succinylcholine as an MH trigger remains questionable.

Assessment of gastrointestinal motility using three different assays in vitro.

The protocols detailed in this unit are designed to assess the motor activity of different gastric and intestinal muscle preparations in vitro and the effects of drugs that modulate gastrointestinal motility. The preparations described are characterized by different contractile behaviors, consisting of spontaneous (duodenum), neurogenic (ileum), and drug-stimulated (fundus, ileum) motility; these reproduce motility patterns occurring in the gut wall in vivo. These protocols document the variety of factors that can influence the responses of isolated tissues and describe how such tissues can be used for testing substances that affect gut movements. These preparations allow evaluation of direct interactions with the processes that control contractile machinery, as well as indirect effects resulting from the modification of neurotransmitter release from myenteric neurons. These models can be exploited to assay novel compounds undergoing preclinical development or to evaluate the functional toxicity exerted by environmental or alimentary pollutants, like xenobiotics and naturally occurring toxins, as well as the mechanisms underlying these effects.

Sugammadex for reversal of neuromuscular block after rapid sequence intubation: a systematic review and economic assessment.

Sugammadex 16 mg kg⁻¹ can be used for the immediate reversal of neuromuscular block 3 min after administration of rocuronium and could be used in place of succinylcholine for emergency intubation. We have systematically reviewed the efficacy and cost-effectiveness and made an economic assessment of sugammadex for immediate reversal. The economic assessment investigated whether sugammadex appears cost-effective under various assumptions about the value of any reduction in recovery time with sugammadex, the likelihood of a 'can't intubate, can't ventilate' (CICV) event, the age of the patient, and the length of the procedure. Three trials were included in the efficacy review. Sugammadex administered 3 or 5 min after rocuronium produced markedly faster recovery than placebo or spontaneous recovery from succinylcholine-induced block. No published economic evaluations were found. Our economic analyses showed that sugammadex appears more cost-effective, where the value of any reduction in recovery time is greater, where the reduction in mortality compared with succinylcholine is greater, and where the patient is younger, for lower probabilities of a CICV event and for long procedures which do not require profound block throughout. Because of the lack of evidence, the value of some parameters remains unknown, which makes it difficult to provide a definitive assessment of the cost-effectiveness of sugammadex in practice. The use of sugammadex in combination with high-dose rocuronium is efficacious. Further research is needed to clarify key parameters in the analysis and to allow a fuller economic assessment.

Sugammadex compared with neostigmine/glycopyrrolate for routine reversal of neuromuscular block: a systematic review and economic evaluation.

The cost-effectiveness of sugammadex for the routine reversal of muscle relaxation produced by rocuronium or vecuronium in UK practice is uncertain. We performed a systematic review of randomized controlled trials of sugammadex compared with neostigmine/glycopyrrolate and an economic assessment of sugammadex for the reversal of moderate or profound neuromuscular block (NMB) produced by rocuronium or vecuronium. The economic assessment aimed to establish the reduction in recovery time and the 'value of time saved' which would be necessary for sugammadex to be potentially cost-effective compared with existing practice. Three trials indicated that sugammadex 2 mg kg⁻¹ (4 mg kg⁻¹) produces more rapid recovery from moderate (profound) NMB than neostigmine/glycopyrrolate. The economic assessment indicated that if the reductions in recovery time associated with sugammadex in the trials are replicated in routine practice, sugammadex would be cost-effective if those reductions are achieved in the operating theatre (assumed value of staff time, £4.44 per minute), but not if they are achieved in the recovery room (assumed value of staff time, £0.33 per minute). However, there is considerable uncertainty in these results. Sugammadex has the potential to be cost-effective compared with neostigmine/glycopyrrolate for the reversal of rocuronium-induced moderate or profound NMB, provided that the time savings observed in trials can be achieved and put to productive use in clinical practice. Further research is required to evaluate the effects of sugammadex on patient safety, predictability of recovery from NMB, patient outcomes, and efficient use of resources.

Comparison of mechanomyography and acceleromyography for the assessment of rocuronium induced neuromuscular block in myotonic dystrophy type 1.

We measured acceleromyography and mechanomyography simultaneously with monitoring of rocuronium-induced neuromuscular block in four patients with myotonic dystrophy type 1. Furthermore, we compared neuromuscular block measures from these patients with those from normal controls from previous studies. In myotonic dystrophy type 1 patients, the dose-response curve obtained with acceleromyography was steeper and right-shifted compared with that obtained using mechanomyography. However, the effective doses to produce 95% neuromuscular block determined with both acceleromyography and mechanomyography were similar to each other and to values found in normal patients. In the three myotonic dystrophy type 1 patients with mild to moderate disease, times to recovery from block were similar to those observed in normal controls. In both patients and normal controls, neuromuscular block recovered faster with acceleromyography. However, in one patient with severe muscle wasting, recovery of neuromuscular block was prolonged. We conclude that mechanomyography and acceleromyography cannot be used interchangeably to monitor neuromuscular block in myotonic dystrophy type 1 patients.

Assessment of neuromuscular dysfunction during poisoning by organophosphorus compounds.

Dysfunction of respiratory muscles is a life-threatening complication in poisoning by organophosphorus compounds (OPs). It is both of central and peripheral origin due to impaired cholinergic signalling upon inhibition of acetylcholinesterase (AChE). The dysfunction at neuromuscular synapses is not amenable to anticholinergics and remains a therapeutic challenge. Thus, a clear understanding of the distinct mechanisms occurring at neuromuscular synapses is decisive for the development and improvement of therapeutic strategies, particularly with nerve agent poisoning, where clinical studies are prevented by ethical considerations. Using red blood cell AChE, the kinetics of OP induced inhibition, aging, and spontaneous and oxime-induced reactivation have been elucidated. In a dynamically working in vitro model with real-time determination of membrane-bound AChE, it was shown that the kinetic constants derived from erythrocyte AChE are comparable to muscle AChE in a given species. To assess, whether kinetic considerations of AChE activity are relevant for the neuromuscular function, organotypic spinal cord-skeletal muscle cocultures have been established. In this model neostigmine and VX affected neuromuscular transmission as anticipated from their known actions on AChE. Also oxime-induced restoration of the neuromuscular transmission was observed. These findings were confirmed by functional studies on diaphragm muscles of various species with determination of muscle force generation upon phrenic nerve or indirect electrical field stimulation techniques. Investigations with human intercostal muscles are in progress to assess the conditions in human tissue. The results obtained with paraoxon favourably correlate with data from clinical findings of parathion-poisoned patients where the correlation of neuromuscular transmission with the activity of erythrocyte AChE could be established. In conclusion, a variety of methods are available to follow the microscopic reactions occurring at the synaptic level. Due to the lack of clinical data with different OPs, e.g. nerve agents, well designed animal experiments, reflecting the human situation as close as possible, are indispensable for the development of new drugs against the deleterious OP effects.

Furthering pharmacological and physiological assessment of the glutamatergic receptors at the Drosophila neuromuscular junction.

Drosophila melanogaster larval neuromuscular junctions (NMJs) serve as a model for synaptic physiology. The molecular sequences of the postsynaptic glutamate receptors have been described; however, the pharmacological profile has not been fully elucidated. The postsynaptic molecular sequence suggests a novel glutamate receptor subtype. Kainate does not depolarize the muscle, but dampens evoked EPSP amplitudes. Quantal responses show a decreased amplitude and area under the voltage curve indicative of reduced postsynaptic receptor sensitivity to glutamate transmission. ATPA, a kainate receptor agonist, did not mimic kainate's action. The metabotropic glutamate receptor agonist t-ACPD had no effect. Domoic acid, a kainate/AMPA receptor agonist, blocks the postsynaptic receptors without depolarizing the muscle. However, SYM 2081, a kainate receptor agonist, did depolarize the muscle and reduce the EPSP amplitude at 1 mM but not at 0.1 mM. This supports the notion that these are generally a quisqualate subtype receptors with some oddities in the pharmacological profile. The results suggest a direct postsynaptic action of kainate due to partial antagonist action on the quisqualate receptors. There does not appear to be presynaptic auto-regulation via a kainate receptor subtype or a metabotropic auto-receptor. This study aids in furthering the pharmokinetic profiling and specificity of the receptor subtypes.

Botulinum neurotoxin: the ugly duckling.

This review presents a brief account of the most significant biological effects and clinical applications of botulinum neurotoxins, in a way comprehensive even for casual readers who are not familiar with the subject. The most toxic known substances in botulinum neurotoxins are polypeptides naturally synthesized by bacteria of the genus Clostridium. These polypeptides inhibit acetylcholine release at neuromuscular junctions, thus causing muscle paralysis involving both somatic and autonomic innervation. There is substantial evidence that this muscle-paralyzing feature of botulinum neurotoxins is useful for their beneficial influence on more than 50 pathological conditions such as spastic paralysis, cerebral palsy, focal dystonia, essential tremor, headache, incontinence and a variety of cosmetic interventions. Injection of adequate quantities of botulinum toxins in spastic muscles is considered as a highly hopeful procedure for the treatment of people who suffer from dystonia, cerebral palsy or have experienced a stroke. So far, numerous and reliable studies have established the safety and efficacy of botulinum neurotoxins and advocate wider clinical therapeutic and cosmetic applications.

Cortical somatosensory evoked potentials and spasticity assessment after botulinum toxin type A injection in children with cerebral palsy.

PURPOSE: The mechanism of Botulinum Toxin Type A (BTX-A) action at the neuromuscular junction is well known. But from the introduction of BTX-A, some authors have suggested a central action of BTX-A and possible side effects far from the site of injection. Some studies demonstrate an improvement of cortical SEPs associated with reduction of spasticity after BTX-A injection. The aim of the present study was to determine the effect of BTX-A treatment on cortical somatosensory potentials (SEP). MATERIAL AND METHODS: A group of twenty nine children ranging from 2 to 17 years old with cerebral palsy were studied. Each patients spasticity level was evaluated before, 2 weeks and 6 weeks after BTX-A injection by the Modified Ashworth Scale and modified Gait Physician's Rating Scale. The SEPs from lower and upper extremities were performed before and between 2 and 6 weeks (19.34 +/- 8.82 days) after BTX-A administration. RESULTS: The mean spasitity level was significantly lower 2 and 6 weeks after BTX-A injection. The gait analysis by modified Physician's Rating Scale (PRS) showed significant improvement two weeks and six weeks after BTX-A injection. SEPs results were abnormal before BTX-A injection in 25 children with cerebral palsy. However we didn't find any significant changes of SEPs latencies after BTX-A injection. CONCLUSIONS: The results of SEP after BTX-A administration in children with cerebral palsy do not confirm the central action of BTX-A on somatosensory pathways. We did not find any significant changes of SEP latencies associated with clinical reduction of spasticity. It seems that SEP results could support the opinion, that BTX-A does not have any direct central effect on sensory pathways. Remote side effects may be explained by an indirect mechanism due to modification of the central loops of reflexes or to hematogenous spread of BTX-A.

Phorbol esters and adenosine affect the readily releasable neurotransmitter pool by different mechanisms at amphibian motor nerve endings.

Phorbol esters and adenosine have been proposed to interact at common sites downstream of calcium entry at amphibian motor nerve endings. We thus studied the actions and interactions of phorbol esters and adenosine using electrophysiological recording techniques in conjunction with both binomial statistical analysis and high-frequency stimulation at the amphibian neuromuscular junction. To begin this study, we confirmed previous observations that synchronous evoked acetylcholine (ACh) release (reflected as endplate potentials, EPPs) is well described by a simple binomial distribution. We then used binomial analysis to study the effects of the phorbol ester phorbol dibutyrate (PDBu, 100 nM) and adenosine (50 microM) on the binomial parameters n (the number of calcium charged ACh quanta available for release) and p (the average probability of release), where the mean level of evoked ACh release (m) = np. We found that PDBu increased m by increasing the parameter n whilst adenosine reduced m by reducing n; neither agent affected the parameter p. PDBu had no effect on either the potency or efficacy of the inhibition produced by adenosine. Subtle differences between these two agents were revealed by the patterns of EPPs evoked by high-frequency trains of stimuli. Phorbol esters increased ACh release during the early phase of stimulation but not during the subsequent plateau phase. The inhibitory effect of adenosine was maximal at the beginning of the train and was still present with reduced efficacy during the plateau phase. When taken together with previous findings, these present results suggest that phorbol esters increase the immediately available store of synaptic vesicles by increasing the number of primed vesicles whilst adenosine acts at a later stage of the secretory process to decrease the number of calcium-charged primed vesicles.

Assessment of a simple artificial neural network for predicting residual neuromuscular block.

BACKGROUND: Postoperative residual curarization (PORC) after surgery is common and its detection has a high error rate. Artificial neural networks are being used increasingly to examine complex data. We hypothesized that a neural network would enhance prediction of PORC. METHODS: In 40 previously reported patients, neuromuscular function, neuromuscular block/antagonist usage and time intervals were recorded throughout anaesthesia until tracheal extubation by an observer uninvolved in patient care. PORC was defined as significant 'fade' (train of four <0.7) at extubation. Neuromuscular function was classified as PORC (value=1) or no PORC (value=0). A back-propagation neural network was trained to assign similar values (0, 1) for prediction of PORC, by examining the impact of (i) the degree of spontaneous recovery at reversal, and (ii) the time since pharmacological reversal, using the jackknife method. Successful prediction was defined as attainment of a predicted value within 0.2 of the target value. RESULTS: Twenty-six patients (65%) had PORC at tracheal extubation. Clinical detection of PORC had a sensitivity of 0 and specificity of 1, with an indeterminate positive predictive value and a negative predictive value of 0.35. Using the artificial neural network, one patient with residual block and one with adequate neuromuscular function were incorrectly classified during the test phase, with no indeterminate predictions, giving an artificial neural network sensitivity of 0.96 (chi(2)=44, P<0.001) and specificity of 0.92 (P=1), with a positive predictive value of 0.96 and a negative predictive value of 0.93 (chi(2)=12, P<0.001). CONCLUSIONS: Neural network-based prediction, using readily available clinical measurements, is significantly better than human judgement in predicting recovery of neuromuscular function.