Restoring speech intelligibility for hearing aid users with deep learning.

Almost half a billion people world-wide suffer from disabling hearing loss. While hearing aids can partially compensate for this, a large proportion of users struggle to understand speech in situations with background noise. Here, we present a deep learning-based algorithm that selectively suppresses noise while maintaining speech signals. The algorithm restores speech intelligibility for hearing aid users to the level of control subjects with normal hearing. It consists of a deep network that is trained on a large custom database of noisy speech signals and is further optimized by a neural architecture search, using a novel deep learning-based metric for speech intelligibility. The network achieves state-of-the-art denoising on a range of human-graded assessments, generalizes across different noise categories and-in contrast to classic beamforming approaches-operates on a single microphone. The system runs in real time on a laptop, suggesting that large-scale deployment on hearing aid chips could be achieved within a few years. Deep learning-based denoising therefore holds the potential to improve the quality of life of millions of hearing impaired people soon.

Non-intrusive deep learning-based computational speech metrics with high-accuracy across a wide range of acoustic scenes.

Speech with high sound quality and little noise is central to many of our communication tools, including calls, video conferencing and hearing aids. While human ratings provide the best measure of sound quality, they are costly and time-intensive to gather, thus computational metrics are typically used instead. Here we present a non-intrusive, deep learning-based metric that takes only a sound sample as an input and returns ratings in three categories: overall quality, noise, and sound quality. This metric is available via a web API and is composed of a deep neural network ensemble with 5 networks that use either ResNet-26 architectures with STFT inputs or fully-connected networks with wav2vec features as inputs. The networks are trained and tested on over 1 million crowd-sourced human sound ratings across the three categories. Correlations of our metric with human ratings exceed or match other state-of-the-art metrics on 51 out of 56 benchmark scenes, while not requiring clean speech reference samples as opposed to metrics that are performing well on the other 5 scenes. The benchmark scenes represent a wide variety of acoustic environments and a large selection of post-processing methods that include classical methods (e.g. Wiener-filtering) and newer deep-learning methods.

Safety and Efficacy of Matrix-Associated Autologous Chondrocyte Implantation With Spheroids for Patellofemoral or Tibiofemoral Defects: A 5-Year Follow-up of a Phase 2, Dose-Confirmation Trial.

BACKGROUND: Matrix-associated autologous chondrocyte implantation (ACI) is a well-established treatment for cartilage defects. High-level evidence at midterm follow-up is limited, especially for ACI using spheroids (spherical aggregates of ex vivo expanded human autologous chondrocytes and self-synthesized extracellular matrix). PURPOSE: To assess the safety and efficacy of 3-dimensional matrix-associated ACI using spheroids to treat medium to large cartilage defects on different locations in the knee joint (patella, trochlea, and femoral condyle) at 5-year follow-up. STUDY DESIGN: Cohort study; Level of evidence, 2. METHODS: A total of 75 patients aged 18 to 50 years with medium to large (4-10 cm2), isolated, single cartilage defects, International Cartilage Repair Society grade 3 or 4, were randomized on a single-blind basis to treatment with ACI at 1 of 3 dose levels: 3 to 7, 10 to 30, or 40 to 70 spheroids/cm2 of defect size. Outcomes were assessed via changes from baseline Knee injury and Osteoarthritis Outcome Score (KOOS), International Knee Documentation Committee score, and modified Lysholm assessments at 1- and 5-year follow-up. Structural repair was evaluated using MOCART (magnetic resonance observation of cartilage repair tissue) score. Treatment-related adverse events were assessed up to 5 years for all patients. The overall KOOS at 12 months was assessed for superiority versus baseline in a 1-sample, 2-sided t test. RESULTS: A total of 73 patients were treated: 24 in the low-dose group, 25 in the medium-dose group, and 24 in the high-dose group. The overall KOOS improved from 57.0 ± 15.2 at baseline to 73.4 ± 17.3 at 1-year follow-up (P < .0001) and 76.9 ± 19.3 at 5-year follow-up (P < .0001), independent of the applied dose. The different defect locations (patella, trochlea, and weightbearing part of the femoral condyles; P = .2216) and defect sizes (P = .8706) showed comparable clinical improvement. No differences between the various doses were observed. The overall treatment failure rate until 5 years was 4%. Most treatment-related adverse events occurred within the first 12 months after implantation, with the most frequent adverse reactions being joint effusion (n = 71), arthralgia (n = 14), and joint swelling (n = 9). CONCLUSION: ACI using spheroids was safe and effective for defect sizes up to 10 cm2 and showed maintenance of efficacy up to 5 years for all 3 doses that were investigated. REGISTRATION: NCT01225575 (ClinicalTrials.gov identifier); 2009-016816-20 (EudraCT number).

Research and Development of Self-Contained Water Injection Systems.

Reducing fuel consumption and thus CO2 emissions is one of the most urgent tasks of current research in the field of internal combustion engines. Water Injection has proven its benefits to increase power or optimize fuel consumption of passenger cars. This technology enables knock mitigation to either increase the engine power output or raise the compression ratio and efficiency while enabling λ = 1 operation in the complete engine map to meet future emission targets. Current systems have limited container capacity. It is necessary to refill the water tank regularly. This also means that we cannot get the benefits of an engine with a higher compression ratio. For this reason, the self-contained system was investigated. This article is a methodology for finding the right design of a self-contained water injection system, but also a vehicle test that proves the function.

Safety and efficacy of matrix-associated autologous chondrocyte implantation with spheroid technology is independent of spheroid dose after 4 years.

PURPOSE: The aim of this study was to investigate the effect of product dose in autologous chondrocyte implantation (ACI) for the treatment of full-thickness cartilage defects of the knee and to assess its influence on clinical and morphological mid-term outcome. METHODS: Seventy-five patients were included in this single-blind, randomised, prospective, controlled clinical trial. Patients were assigned randomly to three different dose groups [low (3-7 spheroids/cm2), medium (10-30 spheroids/cm2), or high (40-70 spheroids/cm2)] and assessed using standardised clinical and morphological scoring systems (KOOS, IKDC, MOCART) for 4 years following the intervention. RESULTS: The analysis population comprised 75 patients (22 women, 53 men) aged 34 ± 9 years. Defect sizes ranged from 2 to 10 cm2 following intraoperative debridement. The assessment of the primary variable 'overall KOOS' showed a statistically significant improvement, compared with baseline, for each dose group, i.e., at baseline the mean 'overall KOOS' scores were 60.4 ± 13.6, 59.6 ± 15.4, and 51.1 ± 15.4 for the low-, medium-, and high-dose groups, respectively, and 57.0 ± 15.2 for 'all patients'. After 48 months those values improved to 80.0 ± 14.7, 84.0 ± 14.9, and 66.9 ± 21.5 in the respective dose groups and 77.1 ± 18.6 for 'all patients'. Pairwise comparisons of these dose groups did not reveal any statistically significant differences. Likewise, assessment of the subjective IKDC score revealed no statistically significant differences between the three dose groups up to the 48-month visit. However, between 12 and 48 months there was a low, but steady, improvement in the low-dose group and a substantial amelioration in the medium-dose group. The mean MOCART total scores 3 months after treatment were 59.8 ± 10.9, 64.5 ± 10.3, and 64.7 ± 9.4 for the low-, medium-, and high-dose groups, and 62.9 ± 10.3 for 'all patients'; 48 months after treatment these were 73.9 ± 13.1, 78.0 ± 12.4, and 74.3 ± 14.0 for the respective dose groups and 75.5 ± 13.1 for 'all patients'. CONCLUSIONS: Results of this study confirm the efficacy and safety of the applied "advanced therapy medicinal product"; no dose dependence was found either for the incidence or for the severity of any adverse reactions. All doses applied in the present study led to significant clinical improvement over time and can therefore be regarded as effective doses. The influence of product doses in the range investigated seems to be low and can be neglected. Thus, the authorised dose range of 10-70 spheroids/cm2 confirmed by this clinical trial offers a broad therapeutic window for the surgeon applying the product, thereby reducing the risk of over- or underdosing. LEVEL OF EVIDENCE: I.

Clinical outcome and success rates of ACI for cartilage defects of the patella: a subgroup analysis from a controlled randomized clinical phase II trial (CODIS study).

AIM: Cartilage defects of the patella are considered as a problematic entity. Purpose of the present study was to evaluate the outcome of patients treated with autologous chondrocyte implantation (ACI) for cartilage defects of the patella in comparison to patient with defects of the femoral condyles. PATIENTS AND METHODS: 73 patients with a follow-up of 5 years have been included in this subgroup analysis of the randomized controlled clinical trial (RCT). In dependence of defect location, patients were divided into two groups [patella defects (n = 45) and femoral condyle defects (n = 28)]. Clinical outcome was evaluated by the means of the KOOS score at baseline and 6 weeks, 3, 6, 12, 18, 24, 36, 48 and 60 months following ACI. RESULTS: "Responder rate" at 60 months (improvement from baseline of > 7 points in the KOOS score) in patients with patella defects was 86.2%. All scores showed a significant improvement from baseline. While overall KOOS score at 60 months was 81.9 (SD 18.6) points in femoral condyle defects, a mean of 82.6 (SD 14.0) was observed in patella defects (p = 0.2483). CONCLUSION: ACI seems an appropriate surgical treatment for cartilage defects of the patella leading to a high success rate. In this study, the clinical outcome in patients with patellar defects was even better than the already excellent results in patients with defects of the femoral condyle even though the study included relatively large defect sizes for both groups (mean defect size 6.0 ± 1.7 and 5.4 ± 1.6 for femur and patella, respectively).

Safety of three different product doses in autologous chondrocyte implantation: results of a prospective, randomised, controlled trial.

BACKGROUND: This study was conducted to assess the efficacy and safety of the three dose levels of the three-dimensional autologous chondrocyte implantation product chondrosphere® in the treatment of cartilage defects (4-10 cm2) of knee joints. We hereby report the safety results for a 36-month post-treatment observation period. METHODS: This was a prospective phase II trial with a clinical intervention comprising biopsy for culturing spheroids and their subsequent administration (level of evidence: I). Patients' knee defects were investigated by arthroscopy, and a cartilage biopsy was taken for culturing. Patients were randomised, on a single-blind basis, to treatment at the dose levels 3-7 (low), 10-30 (medium) or 40-70 (high) spheroids per square centimetre. Assessment (adverse events, vital signs, electrocardiography, physical examination, concomitant medication and laboratory values) took place 1.5, 3, 6, 12, 24 and 36 months after chondrocyte implantation. RESULTS: Seventy-five patients were included and 73 treated. The incidence of adverse events, of patients with adverse events and of patients with treatment-related adverse events showed no relevant difference between the treatment groups. There were no fatal adverse events, no adverse events led to premature withdrawal from the trial and none led to permanent sequelae. Two patients experienced serious adverse events considered related to the study treatment: arthralgia 2-3 years after implantation and chondropathy 1 and 2 years after implantation. CONCLUSIONS: The treatment with chondrosphere® was generally well tolerated. No relationship was detected between any safety criteria and the dose level: Differences between the dose groups in the incidence of any adverse events, and in numbers of patients with treatment-related adverse events, were insubstantial. TRIAL REGISTRATION: clinicaltrials.gov, NCT01225575 .

Structural Plasticity Denoises Responses and Improves Learning Speed.

Despite an abundance of computational models for learning of synaptic weights, there has been relatively little research on structural plasticity, i.e., the creation and elimination of synapses. Especially, it is not clear how structural plasticity works in concert with spike-timing-dependent plasticity (STDP) and what advantages their combination offers. Here we present a fairly large-scale functional model that uses leaky integrate-and-fire neurons, STDP, homeostasis, recurrent connections, and structural plasticity to learn the input encoding, the relation between inputs, and to infer missing inputs. Using this model, we compare the error and the amount of noise in the network's responses with and without structural plasticity and the influence of structural plasticity on the learning speed of the network. Using structural plasticity during learning shows good results for learning the representation of input values, i.e., structural plasticity strongly reduces the noise of the response by preventing spikes with a high error. For inferring missing inputs we see similar results, with responses having less noise if the network was trained using structural plasticity. Additionally, using structural plasticity with pruning significantly decreased the time to learn weights suitable for inference. Presumably, this is due to the clearer signal containing less spikes that misrepresent the desired value. Therefore, this work shows that structural plasticity is not only able to improve upon the performance using STDP without structural plasticity but also speeds up learning. Additionally, it addresses the practical problem of limited resources for connectivity that is not only apparent in the mammalian neocortex but also in computer hardware or neuromorphic (brain-inspired) hardware by efficiently pruning synapses without losing performance.

The Effect of Cell Dose on the Early Magnetic Resonance Morphological Outcomes of Autologous Cell Implantation for Articular Cartilage Defects in the Knee: A Randomized Clinical Trial.

BACKGROUND: Although autologous chondrocyte implantation (ACI) has been established as a standard treatment for large full-thickness cartilage defects, the effect of different doses of autologous chondrocyte products on structural outcomes has never been examined. HYPOTHESIS: In ACI, the dose level may have an influence on medium-term magnetic resonance morphological findings after treatment. STUDY DESIGN: Randomized controlled trial; Level of evidence, 1. METHODS: A total of 75 patients who underwent ACI using a pure, autologous, third-generation matrix-associated ACI product were divided into 3 groups representing different doses: 3 to 7 spheroids/cm(2), 10 to 30 spheroids/cm(2), and 40 to 70 spheroids/cm(2). Magnetic resonance imaging was performed at 1.5, 3, 6, and 12 months after ACI and was evaluated by the magnetic resonance observation of cartilage repair tissue (MOCART) score and the Knee injury and Osteoarthritis Outcome Score (KOOS). RESULTS: MOCART scores showed improvements after 3 months, with slight dose dependence, and further improvement after 12 months, although without significant dose dependence. The mean MOCART scores after 3 months (0 = worst, 100 = best) were 59.8, 64.5, and 64.7 for the low-, medium-, and high-dose groups, respectively, and 62.9 for all patients; at 12 months, these were 74.1, 74.5, and 68.8 for the respective dose groups and 72.4 for all patients. Several MOCART items (surface of repair tissue, structure of repair tissue, signal intensity of repair tissue, subchondral bone, and synovitis) showed a more rapid response with the medium and high doses than with the low dose, suggesting a potential dose relationship. No significant correlation between the MOCART (overall and subscores) with clinical outcomes as assessed by the overall KOOS was detected at 3- and 12-month assessments. CONCLUSION: This study reveals a trend toward earlier recovery after treatment with higher spheroid doses in terms of better defect filling for full-thickness cartilage defects of the knee, while outcomes after 12 months were similar in all dose groups. However, a correlation with clinical outcomes or the failure rate at 1 year after ACI was not found. A longer follow-up will be required for more definite conclusions on the clinical relevance of ACI cell density to be drawn. REGISTRATION: NCT01225575 (ClinicalTrials.gov identifier); 2009-016816-20 (EudraCT number).

Environmental and Intrinsic Correlates of Stress in Free-Ranging Wolves.

BACKGROUND: When confronted with a stressor, animals react with several physiological and behavioral responses. Although sustained or repeated stress can result in severe deleterious physiological effects, the causes of stress in free-ranging animals are yet poorly documented. In our study, we aimed at identifying the main factors affecting stress levels in free-ranging wolves (Canis lupus). METHODOLOGY/PRINCIPAL FINDINGS: We used fecal cortisol metabolites (FCM) as an index of stress, after validating the method for its application in wolves. We analyzed a total of 450 fecal samples from eleven wolf packs belonging to three protected populations, in Italy (Abruzzo), France (Mercantour), and the United States (Yellowstone). We collected samples during two consecutive winters in each study area. We found no relationship between FCM concentrations and age, sex or social status of individuals. At the group level, our results suggest that breeding pair permanency and the loss of pack members through processes different from dispersal may importantly impact stress levels in wolves. We measured higher FCM levels in comparatively small packs living in sympatry with a population of free-ranging dogs. Lastly, our results indicate that FCM concentrations are associated with endoparasitic infections of individuals. CONCLUSIONS/SIGNIFICANCE: In social mammals sharing strong bonds among group members, the death of one or several members of the group most likely induces important stress in the remainder of the social unit. The potential impact of social and territorial stability on stress levels should be further investigated in free-ranging populations, especially in highly social and in territorial species. As persistent or repeated stressors may facilitate or induce pathologies and physiological alterations that can affect survival and fitness, we advocate considering the potential impact of anthropogenic causes of stress in management and conservation programs regarding wolves and other wildlife.

Abnormal Auditory Gain in Hyperacusis: Investigation with a Computational Model.

Hyperacusis is a frequent auditory disorder that is characterized by abnormal loudness perception where sounds of relatively normal volume are perceived as too loud or even painfully loud. As hyperacusis patients show decreased loudness discomfort levels (LDLs) and steeper loudness growth functions, it has been hypothesized that hyperacusis might be caused by an increase in neuronal response gain in the auditory system. Moreover, since about 85% of hyperacusis patients also experience tinnitus, the conditions might be caused by a common mechanism. However, the mechanisms that give rise to hyperacusis have remained unclear. Here, we have used a computational model of the auditory system to investigate candidate mechanisms for hyperacusis. Assuming that perceived loudness is proportional to the summed activity of all auditory nerve (AN) fibers, the model was tuned to reproduce normal loudness perception. We then evaluated a variety of potential hyperacusis gain mechanisms by determining their effects on model equal-loudness contours and comparing the results to the LDLs of hyperacusis patients with normal hearing thresholds. Hyperacusis was best accounted for by an increase in non-linear gain in the central auditory system. Good fits to the average patient LDLs were obtained for a general increase in gain that affected all frequency channels to the same degree, and also for a frequency-specific gain increase in the high-frequency range. Moreover, the gain needed to be applied after subtraction of spontaneous activity of the AN, which is in contrast to current theories of tinnitus generation based on amplification of spontaneous activity. Hyperacusis and tinnitus might therefore be caused by different changes in neuronal processing in the central auditory system.

Audiometric characteristics of hyperacusis patients.

Hyperacusis is a frequent auditory disorder where sounds of normal volume are perceived as too loud or even painfully loud. There is a high degree of co-morbidity between hyperacusis and tinnitus, most hyperacusis patients also have tinnitus, but only about 30-40% of tinnitus patients also show symptoms of hyperacusis. In order to elucidate the mechanisms of hyperacusis, detailed measurements of loudness discomfort levels (LDLs) across the hearing range would be desirable. However, previous studies have only reported LDLs for a restricted frequency range, e.g., from 0.5 to 4 kHz or from 1 to 8 kHz. We have measured audiograms and LDLs in 381 patients with a primary complaint of hyperacusis for the full standard audiometric frequency range from 0.125 to 8 kHz. On average, patients had mild high-frequency hearing loss, but more than a third of the tested ears had normal hearing thresholds (HTs), i.e., ≤20 dB HL. LDLs were found to be significantly decreased compared to a normal-hearing reference group, with average values around 85 dB HL across the frequency range. However, receiver operating characteristic analysis showed that LDL measurements are neither sensitive nor specific enough to serve as a single test for hyperacusis. There was a moderate positive correlation between HTs and LDLs (r = 0.36), i.e., LDLs tended to be higher at frequencies where hearing loss was present, suggesting that hyperacusis is unlikely to be caused by HT increase, in contrast to tinnitus for which hearing loss is a main trigger. Moreover, our finding that LDLs are decreased across the full range of audiometric frequencies, regardless of the pattern or degree of hearing loss, indicates that hyperacusis might be due to a generalized increase in auditory gain. Tinnitus on the other hand is thought to be caused by neuroplastic changes in a restricted frequency range, suggesting that tinnitus and hyperacusis might not share a common mechanism.

Unsupervised learning of digit recognition using spike-timing-dependent plasticity.

In order to understand how the mammalian neocortex is performing computations, two things are necessary; we need to have a good understanding of the available neuronal processing units and mechanisms, and we need to gain a better understanding of how those mechanisms are combined to build functioning systems. Therefore, in recent years there is an increasing interest in how spiking neural networks (SNN) can be used to perform complex computations or solve pattern recognition tasks. However, it remains a challenging task to design SNNs which use biologically plausible mechanisms (especially for learning new patterns), since most such SNN architectures rely on training in a rate-based network and subsequent conversion to a SNN. We present a SNN for digit recognition which is based on mechanisms with increased biological plausibility, i.e., conductance-based instead of current-based synapses, spike-timing-dependent plasticity with time-dependent weight change, lateral inhibition, and an adaptive spiking threshold. Unlike most other systems, we do not use a teaching signal and do not present any class labels to the network. Using this unsupervised learning scheme, our architecture achieves 95% accuracy on the MNIST benchmark, which is better than previous SNN implementations without supervision. The fact that we used no domain-specific knowledge points toward the general applicability of our network design. Also, the performance of our network scales well with the number of neurons used and shows similar performance for four different learning rules, indicating robustness of the full combination of mechanisms, which suggests applicability in heterogeneous biological neural networks.

Comparative survey of canine parvovirus, canine distemper virus and canine enteric coronavirus infection in free-ranging wolves of central Italy and south-eastern France.

Diseases likely affect large carnivore demography and can hinder conservation efforts. We considered three highly contagious viruses that infect a wide range of domestic and wild mammals: canine parvovirus type 2 (CPV-2), canine distemper virus (CDV) and canine enteric coronaviruses (CECoV). Infection by either one of these viruses can affect populations through increased mortality and/or decreased general health. We investigated infection in the wolf populations of Abruzzo, Lazio e Molise National Park (PNALM), Italy, and of Mercantour National Park (PNM), France. Faecal samples were collected during one winter, from October to March, from four packs in PNALM (n = 79) and from four packs in PNM (n = 66). We screened samples for specific sequences of viral nucleic acids. To our knowledge, our study is the first documented report of CECoV infection in wolves outside Alaska, and of the large-scale occurrence of CPV-2 in European wolf populations. The results suggest that CPV-2 is enzootic in the population of PNALM, but not in PNM and that CECoV is episodic in both areas. We did not detect CDV. Our findings suggest that density and spatial distribution of susceptible hosts, in particular free-ranging dogs, can be important factors influencing infections in wolves. This comparative study is an important step in evaluating the nature of possible disease threats in the studied wolf populations. Recent emergence of new viral strains in Europe additionally strengthens the need for proactive monitoring of wolves and other susceptible sympatric species for viral threats and other impairing infections.

Fate of 1,4-dioxane in the aquatic environment: from sewage to drinking water.

Potential health effects of 1,4-dioxane and the limited data on its occurrence in the water cycle command for more research. In the current study, mobility and persistence of 1,4-dioxane in the sewage-, surface-, and drinking water was investigated. The occurrence of 1,4-dioxane was determined in wastewater samples from four domestic sewage treatment plants (STP). The influent and effluent samples were collected during weekly campaigns. The average influent concentrations in all four plants ranged from 262 ± 32 ng L(-1) to 834 ± 480 ng L(-1), whereas the average effluent concentrations were between 267 ± 35 ng L(-1) and 62,260 ± 36,000 ng L(-1). No removal of 1,4-dioxane during water treatment was observed. Owing to its strong internal chemical bonding, 1,4-dioxane is considered non-biodegradable under conventional bio-treatment technologies. The source of increased 1,4-dioxane concentrations in the effluents was identified to originate from impurities in the methanol used in the postanoxic denitrification process in one of the STPs. In view of poor biodegradation in STPs, surface water samples were collected to establish an extent of 1,4-dioxane pollution. Spatial and temporal distribution of 1,4-dioxane in the Rivers Main, Rhine, and Oder was examined. Concentrations reaching 2200 ng L(-1) in the Oder River, and 860 ng L(-1) in both Main and Rhine River were detected. The average monthly load of 1,4-dioxane in the Rhine River was calculated to equal to 172 kg d(-1). In all rivers, concentration of 1,4-dioxane increased with distance from the spring and was found to negatively correlate with the discharge of the river. Additionally, bank filtration and drinking water samples from two drinking water facilities were analyzed for the presence of 1,4-dioxane. The raw water contained 650 ng L(-1)-670 ng L(-1) of 1,4-dioxane, whereas the concentration in the drinking water fell only to 600 ng L(-1) and 490 ng L(-1), respectively. Neither of the purification processes employed was able to reduce the presence of 1,4-dioxane below the precautionary guideline limit of 100 ng L(-1) set by the German Federal Environmental Agency.

Percutaneous epidural lysis of adhesions in chronic lumbar radicular pain: a randomized, double-blind, placebo-controlled trial.

BACKGROUND: Chronic radicular pain can occur after disc pathology and failed back surgery. An evidence-based effective therapeutic option is not available nor does a gold standard exist. OBJECTIVES: A randomized controlled trial to analyze the clinical efficacy of percutaneous epidural lysis of adhesions in chronic radicular pain. STUDY DESIGN: Prospective randomized placebo controlled interventional trial. Power calculation based on a feasibility trial. SETTING: Medical university centers. METHODS: Within 4 years a total of 381 patients with chronic radicular pain lasting longer than 4 months which failed to respond to conservative treatments were screened and 90 patients were enrolled. They were randomly assigned to receive either percutaneous neurolysis or placebo with concealed allocation in permuted blocks of 4 to 8, stratified by treatment center. The primary outcome measure was the differences in percent change of Oswestry Disability Index (ODI) scores 3 months after intervention. Secondary outcome measures were difference in percent change of ODI scores and Visual Analog Scale (VAS) 6 and 12 months after intervention and success rates defined as at least 50% reduction in ODI scores and VAS scores (mean change from baseline) at 3, 6, and 12 months after treatment. Explorative, 2-sided group comparisons for baseline characteristics between active treatment and controls were done using the t-test for 2 independent samples for quantitative data and Fisher's exact test for binary data. RESULTS: The ODI and VAS scores as well as the success rates for ODI vs VAS were significantly better 3, 6, and 12 months in the lysis group vs the control group. The ODI in the lysis group improved from 55.3 ± 11.6 to 26.4 ± 10.8 after 3 months. The placebo group improved from 55.4 ± 11.5 to 41.8 ± 14.6 (P < 0.01). VAS improved from 6.7 ± 1.1 to 2.9 ± 1.9 in the active group and from 6.7 ± 1.1 to 4.8 ± 2.2 (P < 0.01) after placebo. Twelve month follow-up shows further improvement, the differences remain significant. In multiple linear regression, forward and backward variable selection methods resulted in the same covariate model confirming the univariate result for group comparison in the primary analysis. No severe side effects occurred but minor transient neurological effects such as partial sensomotoric deficits did. One dura puncture and one catheter displacement were found. LIMITATIONS: Specific effects of single treatment components cannot be specified because there was no imaging examination after treatment. CONCLUSION: Based on the findings of our study as well as other studies, we believe the minimally invasive percutaneous adhesiolysis procedure should be the first choice treatment option for patients with chronic lumbosacral radicular pain who present with clinical history and findings similar to those of the patients enrolled in our study.

Radial extracorporeal shock wave therapy (rESWT) induces new bone formation in vivo: results of an animal study in rabbits.

The aim of this study was to investigate if radial extracorporeal shock wave therapy (rESWT) induces new bone formation and to study the time course of ESWT-induced osteogenesis. A total of 4000 impulses of radial shock waves (0.16 mJ/mm²) were applied to one hind leg of 13 New Zealand white rabbits with the contralateral side used for control. Treatment was repeated after 7 days. Fluorochrome sequence labeling of new bone formation was performed by subcutaneous injection of tetracycline, calcein green, alizarin red and calcein blue. Animals were sacrificed 2 weeks (n = 4), 4 weeks (n = 4) and 6 weeks (n = 5) after the first rESWT and bone sections were analyzed by fluorescence microscopy. Deposits of fluorochromes were classified and analyzed for significance with the Fisher exact test. rESWT significantly increased new bone formation at all time points over the 6-week study period. Intensity of ossification reached a peak after 4 weeks and declined at the end of the study. New bone formation was significantly higher and persisted longer at the ventral cortex, which was located in the direction to the shock wave device, compared with the dorsal cortex, emphasizing the dose-dependent process of ESWT-induced osteogenesis. No traumata, such as hemorrhage, periosteal detachment or microfractures, were observed by histologic and radiologic assessment. This is the first study demonstrating low-energy radial shock waves to induce new bone formation in vivo. Based on our results, repetition of ESWT in 6-week intervals can be recommended. Application to bone regions at increased fracture risk (e.g., in osteoporosis) are possible clinical indications.