Mariel Purcell

The metabolic cost of passive walking during robotics-assisted treadmill exercise

BACKGROUND We are investigating the potential of robotics-assisted treadmill technology as a mode of exercise in people with spinal cord injury (SCI). People with incomplete SCI can actively contribute to this form of exercise, but in the clinical setting they often walk passively in the system. It is not known whether in doing so they are meeting the recommended guidelines for increasing cardiopulmonary fitness. OBJECTIVE The aims of this study were twofold: to characterise the intensity of passive walking during robotics-assisted treadmill exercise (RATE) in incomplete SCI; and to determine if this intensity meets the recommended guidelines for cardiopulmonary training in this population. METHODS 10 subjects with incomplete SCI twice performed an exercise test on a robotics-assisted treadmill. The test comprised a period of passive walking and a ramp phase to the limit of tolerance. Oxygen uptake VO(2) heart rate (HR) were continuously measured. RESULTS VO(2) during passive exercise was on average 1.4 times higher than resting VO(2R), but this was only 29% of peak VO(2) (VO(2 peak))(range 16-43%). Relative to rest, passive VO(2) (VO(2P) was only 12% of VO(2 peak). HR did not increase from rest to passive walking (81 ± 10 bpm to 81 ± 13 bpm respectively). The HR associated with passive walking was on average 50% of peak HR (HR(peak)) (161 ± 13 bpm). Test-retest reliability was moderate for VO(2R) (R=0.62) and resting HR (HR(R)) (R=0.68), high for VO(2P) (R=0.81), passive HR (HR(P)) (R=0.87) and HR(peak) (R=0.88), and very high (R=0.95) for VO(2 peak). Only HR(p) differed significantly between tests (p=0.029). CONCLUSIONS The intensity of passive walking during RATE is low and is insufficient to increase cardiopulmonary fitness in people with SCI. Subjects must actively contribute to the exercise in order to achieve the recommended training intensity.

Decreases in bone mineral density at cortical and trabecular sites in the tibia and femur during the first year of spinal cord injury.

BACKGROUND Disuse osteoporosis occurs in response to long-term immobilization. Spinal cord injury (SCI) leads to a form of disuse osteoporosis that only affects the paralyzed limbs. High rates of bone resorption after injury are evident from decreases in bone mineral content (BMC), which in the past have been attributed in the main to loss of trabecular bone in the epiphyses and cortical thinning in the shaft through endocortical resorption. METHODS Patients with motor-complete SCI recruited from the Queen Elizabeth National Spinal Injuries Unit (Glasgow, UK) were scanned within 5weeks of injury (baseline) using peripheral Quantitative Computed Tomography (pQCT). Unilateral scans of the tibia, femur and radius provided separate estimates of trabecular and cortical bone parameters in the epiphyses and diaphyses, respectively. Using repeat pQCT scans at 4, 8 and 12months post-injury, changes in BMC, bone mineral density (BMD) and cross-sectional area (CSA) of the bone were quantified. RESULTS Twenty-six subjects (5 female, 21 male) with SCI (12 paraplegic, 14 tetraplegic), ranging from 16 to 76years old, were enrolled onto the study. Repeated-measures analyses showed a significant effect of time since injury on key bone parameters at the epiphyses of the tibia and femur (BMC, total BMD, trabecular BMD) and their diaphyses (BMC, cortical BMD, cortical CSA). There was no significant effect of gender or age on key outcome measures, but there was a tendency for the female subjects to experience greater decreases in cortical BMD. The decreases in cortical BMD in the tibia and femur were found to be statistically significant in both men and women. CONCLUSIONS By carrying out repeat pQCT scans at four-monthly intervals, this study provides a uniquely detailed description of the cortical bone changes that occur alongside trabecular bone changes in the first year of complete SCI. Significant decreases in BMD were recorded in both the cortical and trabecular bone compartments of the tibia and femur throughout the first year of injury. This study provides evidence for the need for targeted early intervention to preserve bone mass within this patient group.

Comparison of peak cardiopulmonary performance parameters during robotics-assisted treadmill exercise and arm crank ergometry in incomplete spinal cord injury

OBJECTIVE (i) to compare cardiopulmonary performance parameters obtained from incremental exercise tests (IETs) performed using a robotics-assisted treadmill and an arm crank ergometer; (ii) to investigate test-retest reliability during both modes of exercise. DESIGN Each participant performed two IETs to the limit of tolerance on both a robotics-assisted treadmill and an arm crank ergometer. SETTING A Spinal Injuries Unit in the United Kingdom. PARTICIPANTS 10 people with an incomplete spinal cord injury (SCI). INTERVENTIONS Not applicable. MAIN OUTCOME MEASURES Peak oxygen uptake (VO(2peak)), the gas exchange threshold (GET), peak heart rate (HR(peak)) and peak lactate (Lactate(peak) were obtained for each mode of assessment. The mean responses and test-retest reliability of the main outcome measures were determined and compared between modes of assessment. RESULTS VO(2peak) was 16% higher (p = 0.016) and the VO2 at the GET was 40% higher (p = 0.007) during the robotics-assisted treadmill exercise (RATE) IET. There was a trend for HR(peak) to be higher during arm crank ergometry (ACE) (p = 0.058). Lactate(peak) was 46% higher (p = 0.006) during the ACE IET. During robotics-assisted exercise, the test-retest reliability was very high for VO2(peak) (r = 0.95), high for the GET (r = 0.75) and HR(peak) (r = 0.88), and moderate for Lactate(peak) (r =0.58). For ACE, the test-retest reliability was very high for VO(2peak) (r = 0.93), high for HR(peak) (r = 0.81) and Lactate(peak) (r = 0.78), and low for the GET (r = 0.16). CONCLUSIONS The results suggest that, when compared with ACE,RATE can be a highly effective stressor of the cardiopulmonary system, and may be a more appropriate mode of assessment to determine and monitor cardiopulmonary fitness in people with incomplete SCI.

Bone morphology of the femur and tibia captured by statistical shape modelling predicts rapid bone loss in acute spinal cord injury patients

After spinal cord injury (SCI), bone loss in the paralysed limbs progresses at variable rates. Decreases in bone mineral density (BMD) in the first year range from 1% (slow) to 40% (rapid). In chronic SCI, fragility fractures commonly occur around the knee, with significant associated morbidity. Osteoporosis treatments await full evaluation in SCI, but should be initiated early and targeted towards patients exhibiting rapid bone loss. The potential to predict rapid bone loss from a single bone scan within weeks of a SCI was investigated using statistical shape modelling (SSM) of bone morphology, hypothesis: baseline bone shape predicts bone loss at 12-months post-injury at fracture-prone sites. In this retrospective cohort study 25 SCI patients (median age, 33 years) were scanned at the distal femur and proximal tibia using peripheral Quantitative Computed Tomography at <5 weeks (baseline), 4, 8 and 12 months post-injury. An SSM was made for each bone. Links between the baseline shape-modes and 12-month total and trabecular BMD loss were analysed using multiple linear regression. One mode from each SSM significantly predicted bone loss (age-adjusted P<0.05 R(2)=0.37-0.61) at baseline. An elongated intercondylar femoral notch (femur mode 4, +1 SD from the mean) was associated with 8.2% additional loss of femoral trabecular BMD at 12-months. A more concave posterior tibial fossa (tibia mode 3, +1 SD) was associated with 9.4% additional 12-month tibial trabecular BMD loss. Baseline bone shape determined from a single bone scan is a valid imaging biomarker for the prediction of 12-month bone loss in SCI patients.

Investigation of Robotic-Assisted Tilt-Table Therapy for Early-Stage Spinal Cord Injury Rehabilitation

Damage to the spinal cord compromises motor function and sensation below the level of injury, resulting in paralysis and progressive secondary health complications. Inactivity and reduced energy requirements result in reduced cardiopulmonary fitness and an increased risk of coronary heart disease and cardiovascular complications. These risks may be minimized through regular physical activity. It is proposed that such activity should begin at the earliest possible time point after injury, before extensive neuromuscular degeneration has occurred. Robotic-assisted tilt-table therapy may be used during early-stage spinal cord injury (SCI) to facilitate stepping training, before orthostatic stability has been achieved. This study investigates whether such a stimulus may be used to maintain pulmonary and coronary health by describing the acute responses of patients with early-stage (<1 yr) motor-complete SCI (cSCI) and motor-incomplete SCI (iSCI) to passive, active, and electrically stimulated robotic-assisted stepping. Active participation was found to elicit an increased response from iSCI patients. The addition of electrical stimulation did not consistently elicit further increases. Extensive muscle atrophy was found to have occurred in those patients with cSCI, thereby limiting the potential effectiveness of electrical stimulation. Active participation in robotic-assisted tilt-table therapy may be used to improve cardiopulmonary fitness in iSCI patients if implemented as part of a regular training program.

Prediction of risk of fracture in the tibia due to altered bone mineral density distribution resulting from disuse: A finite element study

The disuse-related bone loss that results from immobilisation following injury shares characteristics with osteoporosis in post-menopausal women and the aged, with decreases in bone mineral density leading to weakening of the bone and increased risk of fracture. The aim of this study was to use the finite element method to: (i) calculate the mechanical response of the tibia under mechanical load and (ii) estimate of the risk of fracture; comparing between two groups, an able-bodied group and spinal cord injury patients group suffering from varying degrees of bone loss. The tibiae of eight male subjects with chronic spinal cord injury and those of four able-bodied age-matched controls were scanned using multi-slice peripheral quantitative computed tomography. Images were used to develop full three-dimensional models of the tibiae in Mimics (Materialise) and exported into Abaqus (Simulia) for calculation of stress distribution and fracture risk in response to specified loading conditions – compression, bending and torsion. The percentage of elements that exceeded a calculated value of the ultimate stress provided an estimate of the risk of fracture for each subject, which differed between spinal cord injury subjects and their controls. The differences in bone mineral density distribution along the tibia in different subjects resulted in different regions of the bone being at high risk of fracture under set loading conditions, illustrating the benefit of creating individual material distribution models. A predictive tool can be developed based on these models, to enable clinicians to estimate the amount of loading that can be safely allowed onto the skeletal frame of individual patients who suffer from extensive musculoskeletal degeneration (including spinal cord injury, multiple sclerosis and the ageing population). The ultimate aim is to reduce fracture occurrence in these vulnerable groups.

Spinal Cord Injury Caused by Stab Wounds: Incidence, Natural History, and Relevance for Future Research

Spinal cord injury caused by stab wounds (SCISW) results from a partial or complete transection of the cord, and presents opportunities for interventional research. It is recognized that there is low incidence, but little is known about the natural history or the patients suitability for long-term clinical outcome studies. This study aims to provide population-based evidence of the demographics of SCISW, and highlight the issues regarding the potential for future research. The database of the Queen Elizabeth National Spinal Injuries Unit (QENSIU), the sole center for treating SCI in Scotland, was reviewed between 1994 and 2013 to ascertain the incidence, demographics, functional recovery, and mortality rates for new SCISW. During this 20 year period, 35 patients with SCISW were admitted (97.1% male, mean age 30.0 years); 31.4% had a cervical injury, 60.0% had a thoracic injury, and 8.6% had a lumbar injury. All had a neurological examination, with 42.9% diagnosed as motor complete on admission and 77.1% discharged as motor incomplete. A total of 70.4% of patients with an American Spinal Injury Association Impairment Scale (AIS) level of A to C on admission had an improved AIS level on discharge. Nine (25.7%) patients have died since discharge, with mean life expectancy for these patients being 9.1 years after injury (20-65 years of age). Patients had higher levels of comorbidities, substance abuse, secondary events, and poor compliance compared with the general SCI population, which may have contributed to the high mortality rate observed post-discharge. The low incidence, heterogeneous nature, spontaneous recovery rate, and problematic follow-up makes those with penetrating stab injuries of the spinal cord a challenging patient group for SCI research.

Electroencephalographic predictors of neuropathic pain in subacute spinal cord injury

It is widely believed that cortical changes are a consequence of longstanding neuropathic pain (NP). In this article, we demonstrate that NP in individuals with subacute spinal cord injury (SCI) has characteristic electroencephalography markers (EEG) that precede the onset of pain. EEG was recorded in a relaxed state and during motor imagination tasks in 10 able-bodied participants and 31 patients with subacute SCI (11 with NP, 10 without NP, and 10 who had pain develop within 6 months of EEG recording). All 20 patients with SCI initially without NP were tested for mechanically induced allodynia, but only 1 patient, who later had pain develop, reported an unpleasant sensation. The EEG reactivity to eye opening was reduced in the alpha band and absent in the theta and beta bands in the patients who later developed pain and was reduced in those who already had pain. Alpha band power was reduced at BA7 in both the relaxed state and during motor imagination in patients who either had or later developed pain compared with those without pain. All SCI groups had reduced dominant alpha frequency and beta band power at BA7. EEG reactivity to eye opening and reduced spontaneous and induced alpha activity over the parietal cortex were predictors of future NP, as well as markers of existing NP. Clinical Trial Registration Number: NCT02178917 PERSPECTIVE: We demonstrate that brain activity in patients with subacute SCI reveals both early markers and predictors of NP, which may manifest before sensory discomfort. These markers and predictors may complement known sensory phenotypes of NP. They may exist in other patient groups suffering from NP of central origin.

Home Used, Patient Self-Managed, Brain-Computer Interface for Treatment of Central Neuropathic Pain in Spinal Cord Injury: Feasibility Study

Central Neuropathic Pain (CNP) is a frequent chronic condition in people with spinal cord injury (SCI). In a previous study, we showed that using laboratory brain-computer interface (BCI) technology for neurofeedback training, it is possible to reduce pain in SCI people who suffered from CNP for many years. In this study, we show initial results from 12 people with SCI and CNP who practiced neurofeedback on their own using our portable BCI, consisting of a wearable EEG headset (Emotiv, EPOC, USA) and a computer tablet. Eight participants showed a positive initial response to neurofeedback and seven learned how to use portable BCI on their own at home. In this paper, we present a portable BCI and discuss the main challenges of training lay people, patients and their caregivers, to use a custom designed BCI application at home.

Cortical Predictors of Central Neuropathic Pain (CNP) in Sub-acute Patients with Spinal Cord Injury (SCI)

Background Neuropathic pain remains a challenging condition to manage. Gabapentin is a voltage-gated calcium channel blocker which is licensed and commonly administered orally for the management of neuropathic pain. It may however be associated with intolerable side effects in some patient groups which prevents therapeutic doses from being achieved.Topical drug delivery is considered a safer, viable alternative to oral dosing where lower doses can be delivered directly to the site of action. GabaGel(TM) is a topical presentation of gabapentin which has been developed and manufactured as a “Pharmaceutical Special” and subsequently trademarked, in collaboration with St Mary’s Pharmaceutical Unit, a NHS pharmacy manufacturing unit. Patients who present with neuropathic pain which is refractory to conventional therapies, are treated with GabaGel(TM) if deemed clinically appropriate. To assist with future prescribing of this product, a service evaluation was designed to assess efficacy, tolerability, and patient and disease groups who are responders. Aims Our aim was to assess the demographics of patients prescribed GabaGel(TM), including age, indication for GabaGel(TM) and concomitant analgesia used by patients. We also assessed efficacy, tolerability, duration of efficacy, reasons for discontinuation and impact on quality of life measures. Methods Regulatory approval was obtained in the form of service evaluation where no ethics approval was required. Patients attending the pain clinic and treated with GabaGel(TM) (n=148), between 2012 and 2015 were identified through pharmacy records. A postal questionnaire was developed and posted to patients. This included questions on current use of GabaGel(TM), initial and current efficacy, duration of efficacy, concomitant medications and their effects on pain severity, reasons for discontinuation, side effects experienced, and quality of life impact. Patients were sent a letter alongside the questionnaire informing them of the rationale behind the evaluation. They were also informed of planned telephone follow-ups in 3 months’ time for incomplete or non-returned questionnaires. Data from returned completed questionnaires was uploaded on a spreadsheet and analysed using IBM SPSS Statistics Package. Any queries from patients regarding ongoing GabaGel(TM) supply were referred to the Pain Clinic. Results Return rate was 50% (74/148). Indications included general neuropathic pain (GNP) 35.1% (26/74), postherpetic neuralgia (PHN) 14.9% (11/74), trigeminal neuralgia (TGN) 8.1% (6/69), chronic facial pain (CFP) 5.4% (4/74), and complex regional pain syndrome (CRPS) 10.8% (8/74). Pain relief of greater than 30% at one month was seen in CFP 50% (2/4) patients, CRPS 25% (2/8) patients, Peripheral Diabetic Neuropathy (PDPN) 100% (1/1), GNP 34.6% (9/26) patients, PHN 45.5% (5/11), and TGN 33.3% (2/6). Ongoing pain reduction of over 30% was seen in GNP 30.8% (8/26), PHN 45.5% (5/11), and TGN 33.3% (2/6). Of the patients still using GabaGelTM, 73.7% (14/19) continued to experience between 40-100% pain relief. GabaGel(TM) continues to be used by 25.7% (19/74) respondents while 51.4% (38/74) discontinued due to lack of efficacy. Severe nausea was cited as a reason for stopping in 1.4% (1/74), and local irritation in 5.4% (4/74). Conclusion The interim results are encouraging. A mixed response pattern is emerging, possibly attributable to inclusion of patients with a diverse range of pathologies unresponsive to standard medications. Refractory PHN and TGN appear responsive to GabaGel(TM). The overall incidence of systemic and local side effects is low and GabaGel(TM) may be a useful alternative in patients with comorbidities and the elderly. Incomplete questionnaires will be followed up with telephone interview. Patient case notes will also be examined and relevant data pooled to create a responder profile. This survey exposed a lack of a robust mechanism to evaluate unlicensed medications.Background Complex Regional Pain Syndrome (CRPS) is a chronic and disabling condition in which pain, swelling and other symptoms arise in one or more limb(s). CRPS patients report slowed movements and a sense of detachment from their affected body-part(s). Patients are also slower to process tactile stimuli on whichever hand is positioned on the affected side of space. These symptoms have been likened to the syndrome of Hemispatial Neglect (“neglect”), which results from brain injury and is marked by decreased attention to the contralesional side of the body and space. Understanding the full nature of the spatial attention bias in CRPS patients and the extent to which it resembles neglect following brain injury could provide additional insights into the cortical underpinnings of CRPS and inform new treatments. The current study examined the distribution of visual attention in patients with CRPS of the upper or lower limb for the first time. Aims We aimed to test visual attention in patients with CRPS. We hypothesised that they would attend less to targets appearing on the affected as compared to the unaffected side of space. Furthermore, we hypothesised that patients would show less precision in their judgements of spatial targets as compared to controls. Methods Twelve upper and 12 lower limb CRPS patients and 24 controls completed visual temporal order judgement (TOJ) tasks. Two lights were projected onto a table 9cm to the left and right of a fixation point and at offsets of 120, 60, 30, 15, 5, 5, 15, 30, 60 and 120 ms [negative numbers = light appeared on the affected (patients) / non-dominant (controls) side]. Participants stated which light appeared first. They completed the task three times: 1) with their hands out of sight (“no hands”), 2) with the left and right lights projecting onto the left and right hand, respectively (“hands uncrossed”), and 3) with the left and right lights projecting onto the right and left hand, respectively (“hands crossed”). Points of subjective simultaneity (PSSs) and just noticeable differences (JNDs) were obtained from psychometric functions fitted to individual data, and were analysed using bootstrapped linear mixed models regression (1000 samples). Results Group (patients or controls) was a predictor of PSS (β=22, 95%CI=[-44,-4]) and JND (β=24, 95%CI=[5, 41]). For patients, the light on the affected side of space had to appear, on average, 15ms earlier than the light on the unaffected side for the two to be perceived as simultaneous (95%CI=[-27,-2]). Controls exhibited no lateralized bias (PSS=-6, 95% CI=[-16, 6]). Overall, the JNDs of patients (M=96, 95%CI=[76, 117]) were 22ms larger than those of the controls (M=74, 95%CI=[54, 95]). The patients’ biases were statistically similar in the no hands (PSS=-27, 95%CI=[-45, -8]) and the hands uncrossed conditions (PSS=-15, 95%CI=[ 30,6]), but significantly smaller in the hands crossed condition (PSS=-1, 95%CI=[-17, 15]). The difference between JNDs for controls and patients was comparable between the three hand arrangement conditions. Limb (upper or lower) and side (left or right) did not contribute to the prediction of patient’s PSS or JND scores, but other clinical measures did. Conclusion The results provide the first empirical evidence for visual neglect in CRPS. This bias is not limited to visual information about the body because the PSS bias was largest for lights appearing on the table and did not differ between patients with upper and lower limb CRPS. The absence of a bias in the hands crossed condition may reflect an interaction of different spatial reference frames (i.e., attention to the body versus egocentric space) as has been demonstrated in stroke patients. Larger JNDs in patients indicate decreased certainty when judging the order of visual targets, consistent with impaired body representation.Background Pain is bio-psychosocial in nature. Recognising the inter-relationship of physical and psychological factors in Complex Regional Pain Syndrome (CRPS) has clinical implications for treating this often poorly understood chronic pain condition. Whilst there is little evidence suggesting psychosocial factors have a prognostic role in the development of CRPS, research suggests CRPS has significant consequences for patients, such as poor psychological health, poor functional ability and reduced quality of life (QoL). It is also known that anxiety, pain-related fear, and disability in the early stages of CRPS are associated with poorer outcomes in the first year. However, little is known about how demographic and bio-psychosocial factors may vary and relate to patient-defined recovery and other individual characteristics (including duration of the condition and upper- versus lower-limb CRPS). Understanding self-reported symptoms and psychosocial factors within these contexts may be important in helping healthcare professionals design and evaluate future treatments. Aims To undertake a cross-sectional study, describing the self-reported physical and psychological function of people with CRPS (with varying disease duration), and those who had recovered. Using an international sample to ensure population diversity, to identify variances and/or relationships between recovered and non-recovered populations and condition characteristics (upper/lower limb affected). Methods Following ethical and institutional approvals at study sites, potential participants ≥18yrs, who met, or previously met, Budapest CRPS criteria, were identified from 8 country-specific databases (UK, Germany, USA, Canada, Switzerland, Denmark, Netherlands, Poland) and sent postal questionnaires (translated from English as required). Consent was implied by return of completed questionnaires. Data collected within countries was anonymised and pooled on a single common database (lead centre Bath). Self-reported data included: patient demographics, limb affected, disease stage (early, intermediate, late, recovered/non-recovered); knowing CRPS type (I or II); date of CRPS onset; trigger (trauma or spontaneous); and symptoms (from a provided list) in the prior 48 hours. Participants also completed: McGill Pain Questionnaire (MPQ); EQ-5D; Acceptance and Action Questionnaire (AAQ-II); Radboud Skills Questionnaire (RSQ) for upper-limb CRPS; and Measuring Activity Limitations in Walking Questionnaire (WAQ) for lower-limb. Chi-squared, Welch’s t-tests, odds ratios and regression analyses were used to explore the data. Results N=347 participants (80.4% female; mean age=53yrs; 52.7% disease duration ≥ 3yrs. N=310 reported recovery status: 280 (90.3%) non-recovered. Self-reported recovery and number of symptoms were strongly associated (χ2 = 124.94, df = 15, p<.001). There were no associations with demographics, disease duration, trigger, limb affected but positive associations with knowing CRPS type (p < .05) and having caring responsibilities (p < .05). Non-recovered participants had lower psychological flexibility/higher avoidance (AAQ-II), higher pain (MPQ) and lower QoL (EQ-5D) (p ≤ .001). Non-recovered participants with lower-limb CRPS more frequently reported hyperalgesia, allodynia, hair changes, involuntary muscle movements than those with upper-limb CRPS and had poorer MPQ (p < .01), EQ-5D (p < .05), SF-36 Physical Functioning (p < .001) and Energy/Fatigue (p < .05) scores. Function (RSQ), AAQ-II, and number of symptoms were jointly (p < .001) and individually (p < .05) predictive of EQ-5D scores for non-recovered participants with upper-limb CRPS. Conclusion Unresolved CRPS has negative consequences for mental and physical well-being, with people with lower-limb CRPS having the poorest health outcomes. Disease duration and demographic characteristics did not influence health status but a small “recovered” sample (n=30) should be noted. Lower psychological flexibility/higher avoidance in the non-recovered cohort maybe of relevance for treatment interventions, and supports a multi-disciplinary rehabilitation approach which promotes increased physical function conducted with concurrent psychological support. Education about CRPS and responsibilities for others may support recovery from CRPS.People in pain need to effectively communicate to others they are in need of help. This is particularly the case when the person in pain is unable to verbalise, and so needs to rely on nonverbal signals. Observer accuracy in nonverbal pain cue detection is critical, but can vary, and is effected by a range of individual and contextual factors. Whilst there are known gender differences in the recognition of emotional expressions, it is unclear whether this also extends towards pain cues. Furthermore, most gender-based research focuses on facial expressions, whereas there are other nonverbal channels, such as body postures. Finally, studies tend to utilise simple recognition tasks, whereas there are other, more sophisticated approaches, which allow for deeper investigations into different types of attentional processing. One such approach would be to consider whether there are selective attentional biases toward pain, which in turn occur in a gender-specific way.Background: Sessions of Pain Management Programme in which an element of education of patients is involved are similar in structure to cognitive behaviour therapy (CBT) sessions. The key principles of education sessions primarily begin with establishing good rapport and explaining the rationale for therapy (Thorn et al. 2011). Systematic reviews comparing CBT with other treatments concluded that CBT is superior only to treatment as usual or waiting list in improving pain (Eccleston et al 2012). On the other hand Geneen et al (2015) concluded in a systematic review they couldn’t confidently conclude that education alone is effective in reducing pain intensity or related disability in chronic pain in adults. We have decided to review the literature for studies that compared CBT against Education head to head and evaluate the evidence. Aims: To compare the effectiveness of CBT versus Education of patients (Edu) regardless of its delivery method in reducing pain intensity just after the intervention and at 6 months follow up. Methods: Randomised Controlled Trials (RCTs) studies, including 20 or more participants in CBT and Edu arms for the management non-malignant chronic pain of adult patients (18- 65 years of age) were searched in PubMed, Cochrane, Science direct. Data on improvement of pain (measured as pain intensity using a numerical rating scale from 0-10 with 10 is the most severe pain, immediately post intervention and at after 6 months follow up) were extracted from the relevant studies. A meta-analysis approach was followed to estimate the total effect size of the difference between the two interventions in improving pain. Results: Out of initial eligible 5 studies that compared CBT against Education in the same sample of patients at the same time only 3 was qualified for this meta-analysis. These were Turner et al (2006) (n=72 CBT, 76 Edu), Thorn et al. (2011) (n=49 CBT, 34 Edu) and Carmody et al. (2013) (n=48 CBT, 50 Edu). All studies reported that patients were randomised to either CBT or Edu and pain intensity was similar between the two intervention groups at baseline (P>0.1). The overall effect size of the mean difference between CBT and Edu groups’ pain intensity suggested that there is no differences between the two interventions post-intervention (z=.044, P=0.66) or at 6 months follow up (z=-1.126, P=0.26). There was an intermediate level of heterogeneity (Q=4.22, P=0.12, I squared=52.60) but no publication bias among these three studies was identified. Conclusion: CBT and Education programmes of chronic pain management resulted in the same level of pain intensity reduction post-intervention and after 6 months follow up. The two interventions only slightly reduced the pain intensity score in the 3 samples studied which was statistically significant in one study only.