Proceedings of the 153rd Meeting of the Society of British Neurological Surgeons: Oral Abstracts of the Platform Presentations

Click to increase image sizeClick to decrease image size

T2-04: Safety and efficacy of onyx in treatment of difficult cerebral aneurysms

S. K. Prakash, K. S. Manjunath Prasad, R. J. Bore, J. E. Dervin & F. P. Nath (James Cook University Hospital, Middlesbrough, UK)

Objective: To assess the efficacy and safety of Onyx liquid embolic agent in the treatment of select group of patients with cerebral aneurysms.

Methods: A retrospective, observational study by case note review and post onyx embolisation angiograms of all patients.

Results: This series comprised nine patients with 10 aneurysms. Eight patients were females and the mean age group was 53 years. Five aneurysms were large or giant, four aneurysms had recanalised following coil embolisation and one aneurysm was deemed not suitable for coiling and was also difficult to access surgically. Out of the 10 aneurysms studied, 9 aneurysms were embolised completely. One of the aneurysms was not embolised as it did not satisfactorily pass the seal test. Median time of the first follow up angiogram was at 3 months and of the second follow up angiogram 12 months post embolisation. Seven patients have had two follow up angiograms after embolisation with Onyx. Two patients are yet to have follow up angiograms post Onyx embolisation. In six patients out of the seven patients who have had angiograms, complete obliteration of the aneurysms were noted on both the angiograms, where as in one patient recanalisation of the previously Onyx embolised aneurysm was noted at 12 months. This patient underwent further onyx injection and the aneurysm was secured completely. No neurological deficits were noted at the time of both the follow up angiograms in any of these patients.

Conclusion: Onyx is an ideal agent to treat recanalised aneurysms following coiling which are not suitable for repeat coiling and also in large aneurysms where surgical access can be dangerous. Aneurysm occlusion rates are also superior to that of coiling.

T2-05: Prevalence of recurrence and re-treatment of ruptured intracranial aneurysms treated with endovascular coil occlusion is higher in terminal carotid region aneurysms and in younger patients. A review of 349 cases

R. Corns, M. J. Tait, B. Zebian, S. Little, D. Walsh, T. Hampton, N. Deasy & C. Tolias (King's College Hospital, London, UK)

Objective: Ruptured intracranial aneurysms may be treated either by clipping or coiling. Although coiling may be safer in the short term, coiled aneurysms have a greater risk of aneurysm recurrence and are more likely to require re-treatment. The initial advantage of coiling over clipping may be less in younger patients. We aim to establish the recurrence and re-treatment rates in coiled aneurysms at a single institution and to identify which aneurysms are more prone to recurrence or re-treatment.

Design: A retrospective analysis of 349 patients treated with coiling for ruptured intracranial aneurysms at a single centre between November 2003 and July 2007. Angiography was performed 6 months after initial treatment. Those with significant recurrences were re-treated.

Results: Five hundred and seventy nine endovascular procedures were performed during the study period. Three hundred and forty nine patients underwent primary endovascular treatment of a ruptured aneurysm. Sixty-five percent were completely occluded at six month follow up angiography. The overall re-treatment rate following initial check angiography was 9.7%. Thirty-nine percent of incompletely occluded aneurysms were treated with further coiling, 38% conservatively and 1% with surgery. Recurrence and re-treatment rates both varied by aneurysm site. Anterior communicating artery aneurysms were less likely than terminal carotid region aneurysms to be incompletely occluded: (18/139 (13%) vs. 45/109 (41%); p < 0.0001) and less likely to require re-treatment (6/139 (4%) vs. 17/109 (16%); p < 0.003). Patients under the age of 40y were more likely to have recurrences (30/81(37%) vs. 54/268(20%); p < 0.002) and more likely to require re-treatment (13/81(16%) vs. 21/268(8%); p < 0.02).

Conclusion: The overall outcome in our study is comparable with other large series. Both aneurysm recurrence and the need for re-treatment varied by aneurysm location. Patients aged under 40 were more likely to have incomplete occlusion and to require re-treatment at initial follow up angiography.

T2-06: Aneurysm clippings after the establishment of an endovascular service: An audit of the practice of a regional neurosurgical unit

A. G. Kolias, P. Sinha, Y. Z. Al-Tamimi, R. Ramirez, S. A. Ross, N. I. Phillips & A. K. Tyagi (Department of Neurosurgery, Leeds General Infirmary, Leeds, UK)

Objective: Recent years have seen a shift from clipping to endovascular techniques for the obliteration of cerebral aneurysms. Our objective was to look into the reasons for choosing clipping over coiling for ruptured and unruptured aneurysms treated in our department. Another objective was to look into other parameters of our clipping practice after establishing an endovascular service.

Methods: Retrospective analysis of all patients who underwent clipping of aneurysms in our department from January 2006 to September 2008. The departmental database was used to identify eligible patients. Data were extracted from case notes, operation notes, discharge summaries and radiological reports.

Results: One hundred and two eligible patients were identified (66 females, 36 males). Median age was 50.5 years (Q1-Q3: 45–59.7; range 2–73). Thirty-three patients underwent an elective procedure. WFNS grade for patients with a SAH was 1 (n = 26), 2 (n = 11), 3 (n = 11), 4 (n = 11), 5 (n = 6), missing (n = 4). Two operations were abandoned without clipping (basilar termination, cavernous carotid aneurysms). In 16 of the remaining 100 operations, two aneurysms were clipped. The location of the 116 clipped aneurysms was as follows: MCA (n = 55), ACoA (n = 30), PCoA (n = 9), carotid termination (n = 8), pericallosal (n = 3), ACA (n = 2), ICA (n = 2), ophthalmic (n = 2), SCA (n = 2), AChoA (n = 1), basilar termination (n = 1) and PICA (n = 1). Temporary clipping was employed in 39 operations (n = 94); more than 1 clips were used in 26 aneurysms (n = 94); intraoperative rupture occurred in 15 instances (n = 95). Additional procedures carried out were: evacuation of haematoma (n = 16), bone flap removal (n = 3), EC-MCA bypass (n = 2), muslin wrapping (n = 1). The reasons quoted for choosing clipping were the following: wide neck (n = 51), sac incorporating arteries/branches (n = 19), presence of haematoma (n = 16), giant/large aneurysm (n = 13), unfavourable location/orientation (n = 12), coiling attempted but failed (n = 8), irregular/complex aneurysm (n = 6), young age/patient choice (n = 4), partial coiling only feasible (n = 4), recurrence after coiling (n = 3), no neck (n = 1) and missing (n = 3).

Conclusions: Unfavourable anatomy appears to be the predominant reason for choosing clipping over coiling. The above indications will probably change, as complex endovascular techniques are introduced. We propose that frequent and close auditing of clipping practice is critical in order to maintain standards despite the decrease in caseload.

T2-07: Good grade subarachnoid haemorrhage treatment delays in London: A collaborative management strategy

J. N. Lamb, M. J. Tait, M. J. N. Crocker, B. A. Bell & M. C. Papadopoulos (St George's Hospital, University of London, London, UK)

Objectives: To evaluate treatment delays after admission of good grade subarachnoid haemorrhage (SAH) patients in London and to propose a collaborative network for rapid best practice treatment[1].

Methods: WFNS grade 1 and 2 SAH admissions to one London neurosurgical centre were surveyed for 1 year and the time to definitive treatment was measured. Local incidence was extrapolated to estimate the pan-London caseload of good grade SAH admissions to the seven London neurosurgical centres and consultant availability within these centres was assessed.

Results: In 72 good grade patients, treatment delay varied markedly according to day of admission, ranging from 28 h on Sunday to 63 h on Friday. Delays for endovascular treatment reflected the 5 day a week service provision. The annual incidence of good grade SAH in our study was 3.2 per 100000. The estimated catchment population of the seven London neurosurgical centres is 13 million, yielding a caseload of 425 patients annually with an average of 2.3 presenting each weekend. Up to 20 consultant neurosurgeons and 16 interventional neuroradiologists could participate in a collaborative rota.

Conclusions: There are appreciable delays in definitive treatment of SAH patients because of subspecialist availability at weekends. A collaborative weekend rota by London neurosurgical centres may be a practical approach to reduce delay and the risk of rebleeding[2].

T2-08: Is a delayed catheter angiogram warranted after treatment of ruptured intracranial aneurysms with detachable coils?

J. Tailor, P. Goetz, H. Chandrashekar, T. Stephen, J. Grieve, L. Watkins, F. Robertson, S. Brew & N. Kitchen (The National Hospital for Neurology and Neurosurgery, London, UK)

Objective: The optimal strategy for monitoring the stability of ruptured intracranial aneurysms following coil embolisation is unclear. We aimed to determine the value of a further delayed angiogram in detecting new recurrences or progression of residual lesions visualised on an earlier angiographic study.

Subjects and methods: One hundred and forty-two patients with 148 ruptured aneurysms treated with detachable coils between 2000 and 2006 were included. The presence of a residual neck or aneurysm was assessed on catheter angiograms performed at 6 month and 2 years using the Raymond criteria.

Results:Follow-up angiograms were obtained for 109 aneurysms in 102 patients. Ninety-eight percent class I (complete occlusion), 90% of class II (minor neck residual) and five out of eight class III (aneurysm sac filling) aneurysms remained unchanged at 2 years compared with 6 months. Only 2% of class I aneurysms developed minor neck recurrences, whereas 10% of class II and three out of eight class III aneurysms were re-treated by 2 years.

Conclusion: In our series, the vast majority of coiled ruptured aneurysms that were class I or II at 6 months remained stable at 2 year follow-up. In the absence of a residual lesion in the early angiographic study, a further delayed catheter angiogram may not be warranted.

T2-09: Outcome after subarachnoid haemorrhage – A SAH cohort study

A. T. King¹, C. McMahon¹, P. Tyrrell¹, S. Hulme¹, S. Hopkins², A. Vail² & N. Rothwell² (¹Salford Royal Foundation Trust, Manchester UK and ²University of Manchester, Mnchester, UK)

Objectives: (1) To identify clinical predictors of delayed cerebral ischaemia (DCI) in the initial 15 days following Subarachnoid haemorrhage (SAH). (2) To determine the relationship between these variables and outcome. (3) To determine the relationship between the development of DCI and outcome.

Methods: A prospective cohort study examining 179 patients with aneurysmal SAH January 2004–August 2006 at a single institution. DCI is defined according to strict predetermined criteria. Using these criteria would exclude patients with possible ischaemic events. Rather than exclude these patients the definition of an event was revisited which led to four categories of ‘cerebrovascular event (CVE)’:

• DCI

• Multifactorial deterioration with presumed ischaemia component

• Infarction post-procedure

• Rebleed or rupture peri-procedure.

A range of risk factors were recorded, and outcome assessed at 6 months.

Results: Twenty-seven percent had DCI according to the predetermined definition but 41% had a CVE, as defined above. Seventy-eight percent had a favourable outcome, 22% an unfavourable outcome. Of those who had a CVE, 55% had an unfavourable outcome. Statistically significant factors for a cerebrovascular event at 15 days included female sex, hypertension, statins, infection, poor WFNS and Fisher grade, an EVD in situ and transcranial doppler > 120 cm/sec. Statistically significant predictive factors for death included age, smoking, poor WFNS grade, an EVD in situ and a CVE.

Conclusions: A range of factors influence outcome after SAH. DCI is poorly defined in the literature, whilst a tight definition excludes patients who may have had an ischaemic event. DCI and other CVE's are highly significantly associated with poor outcome.

T2-10: Subarachnoid haemorrhage; Do the elderly fare worse?

C. Barrett, C. Dubois, D. Holliman & P. Mitchell (Newcastle General Hospital, Newcastle upon Tyne, UK)

Objective: To determine whether elderly patients who present with subarachnoid haemorrhage have a poorer outcome.

Methods: An analysis of all patients admitted with SAH to the neurosurgical unit during the calendar years 2006 and 2007. Demographics, WFNS grade on presentation, aneurysm location, mode of treatment and outcome (Modified Rankin Scale at discharge) were recorded for each patient. Four hundred and thirty six patients were admitted with SAH. For analysis patients were divided into younger (<70 years) and older ((greater than)70 years) cohorts.

Results: There were 381 patients, <70 years at the time of presentation. Three hundred and fifteen (82.7%) of these younger patients had a good outcome (MRS 0-3) at the time of discharge. Sixty-six (17.3%) patients in this younger cohort had a poor outcome (MRS 4-6). There were 55 patients (greater than) 70 years admitted to the unit. Thirty-seven (67.3%) of these had a good outcome (MRS 0-3), whereas 18 (32.7%) patients had a poor outcome (MRS 4-6). Although this data appears to support the hypothesis that the elderly have a worse outcome in subarachnoid haemorrhage, further analysis reveals a more complex picture. Seventeen out of 18 elderly patients who were WFNS grade I at presentation were MRS 1 or 2 at the time of discharge.

Conclusions: The elderly subgroup of patients appears to have a poorer outcome after spontaneous SAH. However, the case mix is different to the younger sub-group as more elderly patients are poor grade (WFNS IV, V) at the time of presentation. The vast majority of elderly patients who are good grade at presentation remain so after treatment. Therefore, we conclude that treatment should not be denied to the elderly on grounds of age alone.

T2-11: Psychosocial impact of subarachnoid haemorrhage: A survey of patients with good grade subarachnoid haemorrhage

A. Tarnaris, J. Ekanayake, K. Li, V. Apok, B. Davies & H. L. Brydon (University Hospital North Staffordshire, Stoke-on-Trent, UK)

Objective: Subarachnoid haemorrhage is a condition with high morbidity. Even though the physical sequelae are well documented the study of the psychological outcomes as well as the social impact of this condition is rather limited.

Methods: Prospective follow-up of 98 consecutive patients with good grade (WFNS 1 & 2) SAH via postal questionnaire. All patients were operated by the same surgeon between 2001 and 2008 using the same method. Categorical variables were used to describe psychological status and correlate with clinical presentation and demographic data.

Results: Sixty-six patients returned the questionnaire to date. There were 22 males and 44 females. Fifty-five patients were Grade I and 11 patients were grade 2 at the time of operation. Mean follow-up was 47 (14–84) months. Fifty-six (85%) were employed pre ictus and 40 (61%) were able to return to employment on follow-up. Thirty-two (48%) were able to return to the same work. The mean time required to return to employment was 5.8 (0.25–30) months. Thirteen (20%) patients had to retire on medical grounds. The mean time for excessive tiredness to resolve was 8.9 (0–74) months. Twenty-five (38%) patients reported none, 35 (53%) mild and six (9%) severe problems with memory. Thirty-four (51%) reported no problems, 27 (41%) mild and four (6%) patients severe, reported problems controlling their emotions. Nineteen (29%) patients reported no problems, 39 (59%) mild and 8 (12%) severe problems with making decisions. Eighteen (27%) patients reported none of the above problems. Nine (14%) reported unresolved headaches on follow-up. Twenty (30%) patients reported their mood as ‘happy’, 13 (20%) as ‘neutral’ and 23 (35%) as ‘down sometimes’, whereas none reported their mood as ‘low’.

Conclusions: More than half of the patients suffering from SAH are able to return to the same work within an average of 6 months. However, it might take an average of 9 months for tiredness to resolve. About one fifth of patients will have to retire on medical grounds. Less than one third of patients do not exhibit any memory, emotional problems or signs of executive dysfunction. Recommendations regarding investigations on follow-up will be discussed.

T2-12: Higher levels of high-density lipoprotein cholesterol are associated with favourable outcome following aneurysmal subarachnoid haemorrhage: A post-hoc study of two randomised controlled trials

M. Y. Tseng, P. J. Hutchinson & P. J. Kirkpatrick (Addenbrooke's Hospital, Cambridge, UK)

Objective: Higher levels of high-density lipoprotein (HDL) cholesterol are known to have cardio-protective effects. We aimed to investigate whether similar neuroprotective effects can be seen following aneurysmal subarachnoid haemorrhage (aSAH).

Methods: Outcome measured with modified Rankin scale (mRS) and data of full lipid profile of 160 aSAH patients were prospectively collected as part of two randomised controlled trials (Pravastatin, erythropoietin). The lipid profile included total, low-density lipoprotein (LDL), and HDL cholesterol, triglyceride and LDL/HDL ratio (LHR). Fifty-four patients taking statins were excluded from analyses. The lipid profile between 106 patients with favourable (mRS 1–3) or unfavourable outcome (mRS 4–6) at discharge and at 6 months were compared using repeated measurement ANOVA and Dunnett's correction.

Results: Except for HDL cholesterol and THR, no difference was seen in the lipid profile between patients with favourable/unfavourable outcome. Patients with favourable outcome at discharge had higher levels of HDL cholesterol on Days 3 (difference 0.21 ± 0.07 mmol/l, p = 0.027), 6 (difference 0.29 ±0.10 mmol/l, p < 0.001), 9 (difference 0.28 ±0.08 mmol/l, p < 0.001), 12 (difference 0.23 ±0.08 mmol/l, p = 0.027) and 15 (difference 0.22 ±0.08 mmol/l, p = 0.043) following aSAH. They also had a lower THR on day 9 (difference −2.03 ± 0.73, p = 0.043). Likewise, patients with favourable outcome at 6 months had higher levels of HDL cholesterol on Day 3 (difference 0.22 ± 0.07 mmol/l, p = 0.027), 6 (difference 0.27 ±0.08 mmol/l, p = 0.015), and 9 (difference 0.26 ±0.08 mmol/l, p = 0.025).

Conclusions: Higher levels of HDL cholesterol are associated with favourable outcome at discharge and at 6 months following aSAH, which may be related to the anti-atherogenic, anti-inflammatory and anti-oxidative properties.

T2-13: Inflammation and thrombosis – an alternative mechanism for delayed cerebral ischaemia after SAH

C. J. McMahon¹, A. T. King¹, N. J. Rothwell² & P. Tyrrell¹ (¹Salford Royal Foundation Trust, Manchester UK and ²University of Manchester, Manchester, UK)

Introduction: It has become increasingly recognized that vasospasm is not the sole cause for delayed cerebral ischaemia (DCI) after SAH. Inflammatory and prothrombotic pathways have also been implicated. We investigated whether circulating markers of inflammation, (C-reactive protein (CRP), interleukin-6 (IL-6), interleukin-1 receptor antagonist (IL-1RA), white cell count (WCC) and erythrocyte sedimentation rate (ESR)) predict the development of DCI after SAH.

Methods: Single-centre case–control study nested within a prospective cohort. Two hundred and sixteen patients recruited between January 2004 and August 2007. Plasma concentrations of IL-1RA, IL-6, WCC, CRP and ESR measured within 7 days of SAH before angiography and alternate days post-angiography until the development of DCI or 15 days post-symptom onset. Forty-six patients developed DCI. Two controls targeted per case. Initial, peak, average, final and rate of change of each inflammatory marker analysed. Results expressed as odds ratios (OR) derived from conditional logistic regression (controlling for age and sex).

Results: The development of DCI was strongly associated with ESR, with initial (OR 2.4 (1.3–4.6); p = 0.006) average (OR 2.3 (1.3–4.2); p = 0.006), peak (OR 2.1 (1.1–3.9); p = 0.02) and final ESR (OR 2.0 (1.2–3.3); p = 0.009) showing statistical significance. Rate of change of IL-6 (OR 2.3 (1.1–5); p = 0.03), final (OR 1.2 (1–1.3); p = 0.01) and rate of change WCC (OR 1.3 (1–1.6) p = 0.05) also showed predictive value for the subsequent development of DCI.

Discussion: Elevated ESR, leucocytosis and increasing IL-6, reflects impending DCI after SAH. It is likely that elevated ESR reflects a prothrombotic tendancy in these patients and analysis of fibrinogen, the key protein determining ESR, is awaited.

T2-14: Is there a time window with surgery for intracerebral haemorrhage (ICH)?

A. D. Mendelow, P. Mitchell & B. Gregson (Newcastle University, Newcastle upon Tyne, UK)

Introduction: The Surgical Trial for Intracerebral Haemorrhage (STICH) did not demonstrate an overall benefit from surgery in patients with spontaneous supratentorial ICH. Subgroup analysis of that trial and meta-analysis of the 12 other trials suggested a benefit from surgery in the anatomical group of superficial lobar haematomas. This has led to the STICH II trial of early surgery v initial conservative treatment for this subgroup of patients. Over 100 patients have been randomised so far in STICH II. In parallel with the anatomical issue of STICH II is the issue of timing. Data on the time delay between onset and surgery are being collected in STICH II to allow it to be analysed with delays up to 60 h being included. Many STICH II collaborators have requested that the timing issue be evaluated on existing data so we have performed subgroup analysis of STICH and meta-analysis of the 12 other trials.

Results: Five have sufficient data to analyse the timing of surgery. The STICH dataset shows a non significant trend towards improved outcome from surgery in those operated on within 6 h of the ictus. When a meta-analysis is performed of the 5 trials with timing data, this trend becomes significant with p < 0.05.

Conclusions: This observation should encourage collaborators to randomise patients in STICH II as early as possible.

T3-02: Paediatric brain tumours presenting as intracerebral haemorrhage: 5-year review of incidence and outcome

M. K. Hossain-Ibrahim, M. English, A. Kay, R. Walsh & G. Solanki (Birmingham Children's Hospital, Birmingham, UK)

Objective: No reports in the literature evaluate the frequency of intracranial haemorrhage (ICH) in children with brain tumours. Empyrically some paediatric brain tumours are more likely to haemorrhage than others. This study aims to characterise those tumours, and evaluate whether these children are at greater risk of worsened neurological outcome.

Methods: Prospectively collected data (April 2003– May 2008) for the Regional Tumour Registry database was reviewed to analyse tissue diagnosis, any recurrence, age at diagnosis, gender, symptomatology and survival information. Pathological and radiological features, tumour size and morphology were also reviewed. Statistically significant conclusions regarding frequency of haemorrhage in different intracranial tumour groups were analysed using measures of central tendency, variance, correlation and survival curves.

Results: Paediatric brain tumours of 10.6% (22/208) presented with ICH, including two recurrences. Of the 22 that presented with ICH, 12 were supratrentorial (55%), six were infratentorial (27%) and four were pineal region (18%). Sixty-seven percent(4/6) of PNETs (all supratentorial), 40%(2/5) of pineoblastomas, 40%(4/10) of GBMs and 11%(5/43) of pilocytic astrocytomas (PAs), but no patients with medulloblastomas (n = 37, all infratentorial), ependymomas (n = 12) or DNETs (n = 7) presented with ICH. Mortality rates were highest in PNETs (compared to pineoblastomas, GBMs and PAs). Median survival times of PNET and GBM patients significantly decreased if presenting with ICH. Age, sex and tumour size did not affect ICH risk.

Conclusion: The increased mortality in children presenting with ICH because of PNETs and lower median survival with PNETs and GBMs requires further study and may result in a change in monitoring these patients.

T3-03: Paediatric subdural cerebrospinal fluid effusions: When do they occur? A review of 830 scans

G. A. Solanki, N. Abbas & S. Chapman (Birmingham Children's Hospital, Steelhouse Lane, Birmingham, UK)

Objectives: Subdural CSF effusion may be an early phenomenon in traumatic events but can also occur in infectious cases. In an acute setting, their low density can be easily mistaken for chronic subdural haematomas and thus sometimes raise suspicion of varying age trauma of an inflicted kind. We report on 830 scans positive for subdural collections to evaluate frequency of cerebrospinal fluid (CSF) effusions in the subdural space in children under the age of 16 years.

Methods: A retrospective search of all children with radiological evidence of subdural collections was performed using our PACS database. Three hundred and nine children (119 girls, 190 boys) underwent CT or MR imaging for a variety of reasons. Frequency was correlated with diagnostic and demographic factors.

Results: Thirty cases of CSF effusions were identified in 830 (408 CT and 422 MRI) Scans. Nineteen cases occurred in boys and 11 in girls. Two-thirds (19/30) occurred in traumatic cases, one-third (11/30) with infections. Four in the trauma group occurred in suspected cases of NAHI. Sixteen cases were unilateral and 14 had bilateral effusion.

Conclusion: Traumatic effusions are rare but can cause uncertainty in diagnosis and suspicion of NAHI. In our series, we detail a less than 10% occurrence. Therefore, its recognition by neurosurgeons in this age group is important to avoid serious misdiagnosis. Awareness of potential CSF effusions in this subset of children is important. We confirm that they are overwhelmingly associated with traumatic cases and rarely with NAHI. A significant subset can be found in post infectious cases.

T3-04: Do Baclofen pumps influence the development of scoliosis in children?

S. C. Burn & J. M. Drake (The Hospital for Sick Children, Toronto, Canada)

Objective: Intrathecal baclofen is an effective treatment for spasticity in patients with cerebral palsy. There has been increasing concern however that intrathecal baclofen may accelerate the development of scoliosis in this population. To this end we reviewed our population of paediatric patients with baclofen pumps to assess rate of development of scoliosis.

Methods: This was a retrospective chart and radiology review of all paediatric patients with baclofen pumps. Cobb angles were measured preoperatively and on follow up imaging.

Results: Thirty-eight patients were identified, 19 with cerebral palsy (CP), 8 with dystonic CP, six with head injury and five ‘other’ diagnoses. Mean age at pump insertion was 10.6 years old and post operative follow up was for a mean of 34 (range 1–242 months). Mean annual Cobb angle progression was 18.15° (range 0–67.68), median 11.88°.

Conclusions: In our group of patients, there was notable development and progression of scoliosis at a greater than previously reported rate for the same patient population, and also greater than previously reported patients with IT baclofen pumps. The largest possible confounding factor in this study is the insertion of the pump before skeletal maturity and therefore coinciding with the time where scoliosis may be developing naturally. A prospective study is recommended to gather further data on the development of scoliosis in this particular group with IT baclofen pumps.

T3-05: Management of osteoblastoma and osteoid osteoma of the spine in childhood

S. C. Burn, R. Zeller & J. M. Drake (The Hospital for Sick Children, Toronto, Canada)

Objective: Osteoid osteomas and osteoblastomas of the spine are rare lesions in childhood and management strategies have changed. We reviewed our recent experience of these two lesions to illucidate current treatment modalities and outcomes.

Method: Case records and radiographic images of all patients with the diagnosis of osteoid osteoma and osteoblastoma from 1993 to 2008 were retrospectively reviewed including those managed non-operatively.

Results: Thirty cases were identified, 22 surgical, 8 non-surgical management. Mean age at presentation was 13 years (range 3–17 years). Patients 29/30 (97%) presented with pain. Patients 7/30 (23%) had scoliosis at presentation. Patients 12/30 (40%) experienced relief with NSAIDs. Of the patients who underwent surgery, 7/22 (32%) were diagnosed with osteoid osteoma and 15/22 (68%) with osteoblastoma. Patients around 9/22 (41%) underwent fusion (bone onlay or instrumentation) at the time of surgery. Pain relief off medication was achieved in 16/22 (73%) of the surgical group t a mean of 28 months follow up (range 2–75 months). Patients of 4/8 (50%) of the non-surgical group were pain free at a mean follow up of 33 months (range 24–51 months).

Conclusions: Osteoid osteomas and osteoblastoma ca be challenging management problems in paediatric patients. In those where conservative therapy fails, or where pathological diagnosis is required, surgery using modern intraoperative imaging and spinal instrumentation can provide symptom relief and tumour control in the majority of patients.

T3-06: Non-accidental head injury in paediatric neurosurgery

D. Ferguson, J. McMullan & H. Zaki (Sheffield Children's Hospital, Sheffield, UK)

A retrospective case study assessing the investigation of potential non-accidental head injury to children referred to neurosurgery with head or brain injury at a local children's hospital. Standards were derived from the Child Protection Companion April 2006 which contains guidelines for the history, examination and investigation of non-accidental injury. Children were under the age of five and admitted via the accident and emergency department. Over a 1 year period, 92 patients were identified as potential cases and of these, 54 case notes were retrieved and coded appropriately and so included.

Results: The children were admitted under various specialties. Twenty percent of the patients were assessed by a neurosurgeon. One hundred percent of patients admitted with a head or brain injury had a full history and examination. Neurosurgeons appropriately utilised the expertise of other necessary specialties. Ophthalmology assessed 100% of suspected cases of non-accidental injury. Sixty-one percent of all suspected cases of non-accidental injury had full blood investigations performed and 0% had urine sent for toxicology. Of the five cases with an initial suspicion of non-accidental injury one child was referred to another hospital to be investigated. Of the remaining four, three cases were confirmed by further investigations which included a skeletal survey and CT head.

Conclusion: Only five out of eight standards were met indicating that investigation of non-accidental head injury needs to be addressed.

Recommendations: Ensure full investigations are performed, address the accuracy of coding and audit other departments in the hospital to ensure they are also following guidelines.

T5-01: Alcohol and traumatic brain injury: The understated vector of a silent epidemic

D. Shooman, M. Gardner, J. M. A. Sinclair, J. Hell & A. Belli (Wessex Neurological Centre, Southampton, UK)

Objective: Few studies have analysed the role of alcohol consumption in traumatic brain injury (TBI), with conflicting reports on its effect on outcome. Our aim was to investigate the incidence of alcohol intoxication in our TBI population and its relationship to the severity of injury, intracranial pathology and outcome.

Methods: From the TBI database, demographic, injury and clinical details (including alcohol intoxication) were prospectively recorded for 362 patients aged >16 years and admitted to Neurosurgical ITU between September 2005 and September 2008. Data, including mortality and dichotomized GOS (favourable = 4–5; unfavourable = 1–3) at discharge, 6 and 12 months, were analysed using χ², t-test and ANOVA after normalisation where appropriate.

Results: Intoxication was present in 36% of patients who were slightly younger (mean 37.6 vs. 40.1 years) and predominantly males (83%vs. 76%) when compared with non-intoxicated. There was no significant difference between these groups in GCS (median 8 vs. 10), length of stay and outcome. At 6 months, 67% had a favourable outcome (71% at 12 months) and mortality was 8.4%. There was a strong association between alcohol and mechanism of injury (p = 0.02), with assaults being significantly more common in this subgroup.

Conclusions: To our knowledge, this is the largest single-centre study on alcohol consumption and TBI. Despite no apparent effect on outcome, intoxication plays a role in over a third of the admissions. Given the disease burden of TBI, this further highlights public health concerns about alcohol in the UK. The focus should perhaps be on the rise in consumption in young people, a group for which TBI remains the largest cause of death and disability.

T5-02: A retrospective study of paediatric head trauma admissions in a regional neurosurgical unit from 1998 to 2007

I. S. K. Phang, I. Sexton, C. Mathieson, S. Forsyth, E. J. St. George & J. Brown (Institute of Neurological Sciences, Glasgow, UK)

Objectives: To examine the trends of paediatric head trauma admissions to a neurosurgical unit over ten years.

Methods: Admission details, TBI severity and GOS were obtained from the neurosurgical register for patients aged between 0 and 16 admitted from regional accident and emergency departments between 1998 and 2007.

Results: Five hundred and forty admissions were identified from 539 patients. Seventy-two percent were male and the mean age was 8.8 years. 6.1% of admissions were in the 0 to 1 age group, 16.1% in the 1 to 5 age group, 32.8% in the 5 to 10 age group and 44.9% in the 10 to 16 age group. There was an increase in admissions for patients aged 10 to 16 and a decrease in the number of patients presenting with severe TBI over 10 years. The proportion of admissions in the severe, moderate and mild TBI groups were 11.1%, 15.7% and 73.2%, respectively. Alcohol intoxication was a factor in 2.2% of all admissions and 8.1% of assaults. The three commonest mechanisms of injury across the age groups were because of falls (52.8%), pedestrian versus car traffic accidents (18.0%) and assaults (16.1%).

Conclusions: We report a trend for increasing admissions in the older age group. There is a statistically significant difference in worse outcomes between the severe and mild TBI groups. We conclude by discussing possible factors that contribute to these trends.

T5-03: Clinical study of neurosurgical casualties in sieged Sarajevo: Data from the Bosnian War

K. Dizdarevic, V. Apok & K. Custovic (Department of Neurosurgery, University of Sarajevo, Bosnia)

Objective: Neurosurgical training in the UK is based on access to modern equipment and electricity. In the context of war, such equipment may not be available and neurosurgeons may be expected to adopt uncommon strategies. This may be a reality in the UK as we face a constant threat of war.

Design: This retrospective study analysed the working conditions of Bosnian neurosurgeons and the outcome of their patients (Apr 1992–Jul 1994). The data are presented with the hope of drawing lessons from their experience. Subjects: 1728 neurosurgical patients (Sarajevo population at the time – 361 179) were admitted with an almost equal divide of civilians and military personnel.

Method: Review of hospital data and interviews with neurosurgeons who operated during the Bosnian War.

Results: Eighty-four percent of patients were admitted within 3 h of injury. Diagnosis was largely dependant on clinical examination (34.7% of cases) and plain radiography (47.3%). CT was only used 17.9% of the time because of lack of availability. There was no electricity or water supply. Operating was carried out with battery-powered headlights and ice blocks were used for scrubbing. The majority (80.15%) of admissions were head injuries (average mortality: 29.9%). Seventy percent of these were operated on. The most common surgical strategy was craniectomy with reduced debridemen. The highest mortality was amongst thosed aged 31–40 (24.3%). Spinal injuries (comprising 5.8% of patients) had an average mortality of 18.8%. The infection rate was 1.61%. It was noted that patients admitted later than 6 h post-injury had a higher infection rate (8.1% compared with 1.4%).

Conclusion: The results shown here reflect the ability of the neurosurgeons involved to improvise. Field training was also a key factor. It is worthwhile to consider if UK trainees would benefit from exposure to neurosurgical field equipment. Such training could enhance the functional ability of a neurosurgical unit during wartime or an emergency.

T5-04: UK NHS neurosurgical experience of battlefield injuries

J. S. Dhir, G. W. Jones, D. McAleer, K. Phillips & G. Flint (Queen Elizabeth Hospital, Birmingham, UK)

Introduction: Wars have opened up a different type of experience for all surgeons including neurosurgeons. To provide a high level surgical care to injured soldiers, we need a competent trained team of surgeons in all disciplines, not only out in the field but also at base hospitals in the UK.

Methods: Retrospective study of records of 23 patients between 2004 and 2007 admitted in our Trust. Neurosurgical management and complications were audited. Outcome was based on Glasgow Outcome Score at discharge.

Results: The median patient age was 25 yrs (range 18–53), all male. In nine out of 10 cases, injuries were sustained because of blasts. Immediate aid was at the field hospitals with median of 2 days (range 1–7) before being transferred back to UK. Half of the patients had isolated head trauma. Three out of four had mild- moderate head injuries and remaining one-quarter severe/critical. Two out of three had closed head injury. Brain contusion was the commonest CT finding. Total of 39 neurosurgical procedures were carried out, with one of three being scalp debridements. Neurosurgical complications were brain abscess in two and infected wound in one. Average length of stay at RCDM was 34 days. GOS at discharge was four or five in eight out of 10 cases.

Conclusions: Two of three patients were transferred to RCDM within 48 h and one out of three required neurosurgical procedure at RCDM. Timely aid at scene and at field hospitals with early transfer to RCDM is crucial for good outcome for these young individuals.

T5-05: Hierarchical log linear analysis of admission blood parameters in the prediction of clinical outcome following traumatic brain injury

A. Helmy, I. Timofeev, C. Palmer, A. Gore & P. J. Hutchinson (Addenbrooke's Hospital, Cambridge, UK)

Objective: Multiple regression analysis has been used to identify a wide range of predictors of outcome following traumatic brain injury: the most comprehensive of which is the IMPACT study[1]. A novel method of statistical analysis, hierarchical log linear analysis, can be employed to identify the statistical inter-relationships of these various predictors. We applied this methodology to admission blood parameters in order to investigate the relative impact of these predictors on each other and on mortality.

Methods: Retrospective data collection of admission (within 24 h) serum parameters in patients with traumatic brain injury in a single neurosurgical centre (n = 325). Other parameters that have previously been identified to impact on outcome such as age, admission GCS and injury severity were also collected. Follow up data was collected on all patients at 6 months post-injury and hierarchical log linear analysis was used to identify predictors of mortality.

Results: The only predictors to independently influence mortality were, in order of statistical significance: haemoglobin, age, GCS and white cell count. Multiple interactions exist between these various predictors and other secondary predictors (CRP, abnormal clotting, serum albumin, injury severity and albumin) that can be represented diagrammatically as a hierarchy.

Conclusions: Hierarchical log linear analysis has the ability to illustrate, pictorially, the intuitive inter-relationships between outcome predictors identified by conventional multiple regression. This homogenous data set (single centre) is concordant with other larger series of outcome predictors following traumatic brain injury.

T5-06: Outcomes on the time of insertion of cranioplasty and infection

D. Thavarajah, P. De Lacy, A. Hussain & A. Sugar (Morriston Hospital, Swansea, UK)

Objective: Cranioplasty is carried out for cosmesis, protection and also for neurological improvement for cranial defects. Infection post cranioplasty is problematic. We look at the outcome from 10 years of cranioplasty patients, and aim to see if there is a correlation between infection, time to insertion of cranial plate and reason for craniectomy.

Methods: A retrospective audit identifying all cranioplasty patients between 1998 and 2008 using the theatre logs, the clinical coding system and the maxillofacial laboratory data log of all plates that were made during this time. A proforma then yielded the data from clinical case notes and the microbiology database.

Results: There were 82 cases: 57 males, 25 females. Mean age range 21–40 (33 patients). Reason for craniectomy: Tumour (meningioma) = 19, Trauma (extradural, decompressive) = 46, Craniosyntosis =9, Bone cyst = 8. Time from craniectomy to cranioplasty(months): 0–6 (months) = 22 patients, 7–12 = 26, 13–24 = 20, greater than 25 = 14. Infected cranioplasties = 9 (Seven titanium two acrylic).

Conclusions: An 11% infection rate of cranioplasties. Craniectomy because of trauma, with cranioplasty occurring between 0 and 6 months had the greatest number of infective cases of eight out of nine occuring in this period. Thus suggests that cranioplasty should be done at greater than 6 months. The remainder infective cranioplasties occurred at 13–24 months (n = 1/9). A comparison with other neurosurgical/maxillo-facial units would prove useful, so that an optimum period between craniectomy and cranioplasty in the trauma cases could be defined.

T5-07: Randomised controlled trial of the use of drains versus no drains after burr hole evacuation of chronic subdural haematoma

T. Santarius, P. J. Kirkpatrick, D. Ganesan, H. L. Chia, I. Jalloh, P. Smielewski, H. K. Richards, H. Marcus, S. J. Price, R. W. Kirollos & P. J. Hutchinson (Addenbrooke's Hospital, Cambridge, UK)

Introduction: Chronic subdural haematoma (CSDH) is treatable using relatively simple techniques. The lack of Class I evidence guiding the treatment of CSDH was highlighted in a recent meta-analysis. The aim of this study was to provide Class I evidence for the role of drains in burr hole evacuation of CSDH.

Methods: Between November 2004 and 2007, 215 patients were randomised, 108 were assigned to drain and 107 to no drain groups. The primary outcome was recurrence requiring re-drainage. The secondary outcomes included mortality at 30 days and 6 months, modified Rankin scale (MRS) score at discharge and at six months, and the duration of hospital stay.

Results: The recurrence rate was 10/108 (9.3%) in the drain group and 26/107(24%) in the non-drain group [p = 0.003]. At 6 months, the mortality was 10/106 (9.4%) and 19/105 (18.1%) in the drain and no drain groups, respectively [p = 0.042]. A greater proportion had favourable MRS (0–3) in the drain group at discharge and at 6 months, 81/97 (84%) versus 64/95 (67%) [p = 0.009] and 64/76 (84%) versus 60/85 (71%) [p = 0.040], respectively. Also a significantly greater proportion of patients with drain than without was discharged with GCS 15 and no gross neurological deficit: 76/94 (81%) versus 62/97 (63%) [p = 0.007] and 47/93 (51%) versus 63/96 (66%)[p = 0.036], respectively. There was no difference in the rate of medical or surgical complications between the study groups.

Conclusions: The use of drain with burr hole drainage is associated with lower recurrence rate, better neurological status and MRS at discharge, and better MRS and lower mortality 6 months.

T5-08: Olfactory enstheathing cells cultured from biopsies of human nasal mucosa

C. Kachramanoglou, S. Law, D. Li, C. East, P. Andrews, P. Raisman & D. Choi

Objectives:Olfactory ensheathing cells are the most promising candidate for cell mediated repair of spinal cord/CNS injury. The yield of human olfactory ensheathing cells from the olfactory mucosa has been the limiting factor for clinical translation of autologous transplantation studies. Our aim is to map the distribution of olfactory mucosa in patients and study the yield of olfactory ensheathing cells from biopsies of different regions.

Methods:Nasal mucosa was obtained from different regions of the nasal mucosa endoscopically during routine sinus surgery. Biopsies were histologically processed and cultured in the laboratory and the yield of olfactory ensheathing cells was compared with biopsy location, specimen characteristics, disease severity and patients age.

Results: We obtained 59 specimens from 49 patients. So far, histology has been performed in 29 samples and culture in 26 samples. All 31 specimens contained olfactory mucosa. There was a statistically significant negative correlation between olfactory ensheathing cell yield and age (ρ = −0.483, ρ = 0.01) and a weak correlation between yield and disease severity (ρ = −0.25) and smoking (ρ = −0.22). Better yields were obtained from upper regions of the superior turbinate.

Conclusion: Olfactory ensheathing cells can be obtained from both the lateral and medial wall of the nasal cavity. Obtaining OECs from patient mucosa, however, is much more difficult than in rats. Upper regions of the nose are associated with a higher culture yield in our control patient group. We are further investigating the possible detrimental effect of nasal antiseptic and decongestant preparations on OEC viability, and methods of OEC amplification in vitro.

F2-02: PD SURG: The largest, randomised trial to date comparing surgery with medical therapy in Parkinson's disease

R. Mitchell, A. Williams, S. Patel, N. Ives, C. Rick, J. Daniels, C. Jenkinson, S. Gill, T. Varma & K. Wheatley (On behalf of the MRC PDSURG trial investigators)

Objective: PD SURG is a randomised trial designed to determine whether surgery, using deep brain stimulation (DBS) to the subthalamic nucleus (STN), is more effective than medical therapy (with surgery delayed for at least 1 year) for PD patients whose symptoms are not well controlled by medical therapy. The primary endpoint is the patient's self-evaluation of their quality of life (QoL) using the Parkinson's disease questionnaire (PDQ-39), with Unified Parkinson's Disease Rating Scale (UPDRS) being a secondary endpoint to provide a clinical assessment.

Methods: Mean changes between baseline and 1 year were calculated for the PDQ-39 and UPDRS in the on state. Differences between arms were compared using t-tests.

Results:Three hundred and sixty-six patients were randomised (183 per arm): average age 59 years; 71% male; average time from PD diagnosis 11 years. Main reasons for considering surgery were dyskinesias and severe off-periods. Overall QoL and on state UPDRS were significantly improved with surgery mainly related to motor and ADL domains.

Conclusions: At 1 year, patient assessed QoL is improved following surgery. These differences are likely to be meaningful to patients though there were surgical side effects in a minority.

F2-03: Neuropsychiatric sequelae of bilateral subthalamic nucleus stimulation for Parkinson's disease

A. M. Harries, D. E. Ngoga, J. Kausar, H. S. Pall & R. D. Mitchell (Queen Elizabeth Hospital, Birmingham, UK)

Objective: To determine whether a tendency for adverse neuropsychiatric sequelae could be determined by preoperative psychological assessment of patients whom underwent bilateral subthalamic nucleus (STN) placement for Parkinson's disease.

Methods: A retrospective review of case notes from 1999 to 2008. The pre operative psychological assessment was analysed and compared to outcome.

Results: Sixty-five patients underwent bilateral STN placement during this period. There were 51 men, 14 women. The age range was from 35 to 69 years. All patients had medication reduction (three patients off all medication). All patients passed their dementia rating score (DRS 2) and were considered suitable candidates for surgery. Four patients (6%) developed adverse neuropsychiatric sequelae (one admitted to psychiatric hospital, two patients had suicide attempts and one patient was considered for electro convulsive therapy). Six patients did not have a good outcome post surgery (one Parkinson's Disease plus, three postural drop in BP resulting in poor mobility).

Conclusions: Pre operative psychological assessment provides no indication of which patients will develop adverse neuropsychiatric behaviour post operatively. We recommend that psychological assessment be performed only for patients in whom concern is raised at their preoperative visit.

F2-04: MRI abnormalities in post-anoxic patients evaluated for GPi deep brain stimulation

F. Vergani¹, N. Burger², C. Nicholson¹, L. Cif, X. Vasques² & P. Coubes² (¹Regional Neurosciences Centre, Newcastle General Hospital, Newcastle Upon-Tyne, UK and ²CHRU Montpellier, Service de Neurochirurgie, Montpellier, France)

Title: MRI abnormalities in post-anoxic patients evaluated for GPi deep brain stimulation.

Objective: to compare the different patterns of lesions seen on MRI in post-anoxic patients with the clinical presentation and to evaluate the role of MRI in selecting post-anoxic patients for globus pallidus internus (GPi) deep brain stimulation (DBS).

Methods: Thirty-seven patients with hypoxic encephalopathy were divided into three subgroups according to clinical patterns: Group 1 – dystono-dyskinetic syndrome (DDS) (7); Group 2 – mixed DDS and pyramidal motor deficit (24); Group 3 – pyramidal motor deficit (6). Only patients with pure DDS and mixed syndromes underwent bilateral GPi DBS. MRIs of all patients were reviewed for structural and signal abnormalities.

Results: Abnormalities were seen in the putamen in 35 patients, in the globus pallidus externus in 29, in the GPi in 11, in the thalamus in 15 and in the white matter in 14. The pattern of lesions varied between groups. In Group 1 the GPi was spared and the ventrolateral thalamus was rarely involved. In Groups 2 and 3, lesions were seen in the putamen, GPe and GPi and lesions in the ventrolateral thalamus were more frequent.

Conclusions: Lesions in the basal ganglia and thalamus were observed on MRI in post-anoxic patients. The pattern of lesions correlates with the clinical presentation, with more diffuse lesions being seen in patients with predominantly motor deficits. MRI evaluation is essential in selecting patients for surgery.

F2-05: Indications for and results of pallidal stimulation in post-anoxic dystonia

C. L. Nicholson, N. Burger, F. Vergani, L. Cif, P. Coubes & A. Jenkins (Regional Neurosciences Centre, Newcastle-Upon-Tyne, UK. University Hospital of Montpellier, Montpellier, France)

Objective: To report the results of and to determine the criteria for pallidal deep brain stimulation (DBS) in post-anoxic dystonia, and to report the MRI findings.

Methods: A prospective study was carried out between 1 January 2000 and 12 January 2007. The inclusion criteria were as follows: a history of cerebral hypoxia; limited structural damage of the basal ganglia on MRI; preservation of pyramidal tract function. Outcome was assessed by using the Burke-Fahn-Marsden Dystonia Rating Scale (BFMDRS).

Results: Thirty-one patients (16 children and 15 adults) were entered into the study, of whom one was lost to follow-up. Patients (7/31) had pure extrapyramidal tone and movement disorder, 18/31 patients had a mixed pyramidal and extrapyramidal tone and movement disorder with predominantly extrapyramidal signs and 6/31 patients had a mixed presentation with predominantly pyramidal signs. All MRIs showed abnormalities. The globus pallidus internus was structurally normal in patients with purely extrapyramidal signs. Lesions of the thalamus (11/31) and of the primary motor cortex and underlying white matter (3/31) were most prevalent in patients with predominantly pyramidal signs. There were no serious complications. Overall BFMDRS clinical scores improved by 32% at 1 year and 43% at 3 years (p < 0.001). There was no significant improvement in the disability score of the BFMDRS. The greatest degree of improvement was seen in patients with purely extrapyramidal signs.

Conclusions: Pallidal stimulation is an important advance in the treatment of post-anoxic dystonia. Outcome is strongly correlated with clinical presentation and radiological findings.

F2-06: Assessment of outcome following surgical treatment of paediatric dystonia with deep brain stimulation

P. Epaliyanage, J. P. Lin & R. P. Selway (King's College Hospital, London, UK and Evelina Childrens Hospital Complex Motor Disorders unit, London, UK)

Background: Surgical outcome is often difficult to assess reliably. Many approaches are subjective and assessed by the surgeons themselves. Objective scoring systems may not address the issues from the patients perspective. Children with severe dystonia provide a particularly challenging group in this respect. Multiple disabilities, communication difficulties and inter-patient symptom variability make single scoring system hard to use for qualitative data collection.

Methods: Consecutive patients with severe dystonia underwent comprehensive assessment before insertion of pallidal deep brain stimulators. Before surgery a variety of assessments were tailored to individual patients including a disability and dystonia scale, pain, range of movement, communication, activities of daily living measurements and caregiver priorities. Having documented specifically about family and patient concerns and expectations before agreeing treatment aims and grading their importance, a list of key goals was drawn up with outcomes that can be measured.

Outcome: Thirty patients underwent assessment, agreement of surgical goals and then surgery. The global dystonia scale (BFMDRS) improved in all patients but use of this was hampered by disabilities which exceeded the maximum score. Goals in each patient varied widely but recurring themes included reduction in pain, improved sleep and improved ability to transfer and be seated in a wheelchair, most of which are not considered directly in BFMDRS.

Conclusions: In complex conditions with variable disability, single outcome measuring tools are often inappropriate to address patient, carer and physician perspectives. Individually tailored outcome measures and agreeing goals in advance of surgery allows rigorous quality measurement. However such an approach is time consuming and requires meticulous multidisciplinary setup.

F2-07: Deep brain stimulation for pain after stroke

E. A. C. Pereira, S. R. Qadri, L. Moir, A. L. Green, & T. Z. Aziz (Department of Neurological Surgery, The West Wing, Oxford, UK)

Objective: Central post-stroke pain (CPSP) affects 8% of all stroke patients, often despite medication. Here, we evaluate outcomes from deep brain stimulation (DBS) of the ventroposterolateral thalamus and periventricular grey matter in its treatment.

Methods: Prospective case series. Subjects 21 patients with CPSP treated by DBS Methods Evaluation was undertaken at a single centre over seven years with up to 5 years of follow-up. Visual Analogue Scale (VAS), McGill Pain Questionnaire (MPQ) and quality of life assessments (SF-36 and EQ5D) were completed before and after surgery, at 1–3 months, 6 months and yearly thereafter. DBS parameters were recorded when changed. Changes over time in analgesia and their relationship to stimulation parameters of frequency, voltage and pulsewidth were analysed.

Results: Efficacy was demonstrated in 15 patients (71%). Mean VAS scores showed an initial improvement of 43%, reducing to 19% at 1 year and then with time. Stimulation parameters were altered with time to maintain efficacy and overcome tolerance. Average frequency and voltage both decreased significantly with time (p < 0.05) with average pulsewidth unchanged. Strong positive correlation between frequency and voltage was found (p < 0.005). MPQ showed greater improvements than VAS, as did quality of life indices.

Conclusions: DBS is efficacious for CPSP, but VAS is limited as an assessment tool, suggesting tolerance. CPSP pain is typically a burning allodynia that improves and alters with DBS. Frequency and voltage decreases reflect lower energy requirements over time unlike movement disorders. Tolerance may be overcome by deramping frequency and voltage.

F2-08: Surgical complications of occipital nerve stimulation

A. Ghosh, A. Haridas & L. D. Watkins (National Hospital for Neurology & Neurosurgery, London, UK)

Objective: To review the indication for insertion of and incidence of surgical complications after insertion of occipital nerve stimulators.

Methods: Retrospective audit of electronic and/or paper notes for patients who underwent insertion of occipital nerve stimulators between 2002 and 2007 at the National Hospital for Neurology & Neurosurgery, Queen Square.

Results: Notes were available and reviewed for 46 patients. The indication for insertion was chronic migraine in 44% and chronic cluster headache in 28%, with the remainder being inserted for other primary headache disorders. The hospital length of stay was 4.8 ± 1.9 (mean ± SD) days. Twenty patients (43%) experienced unanticipated complications requiring re-operation; these complications included component migration (6/46), discomfort at insertion site (8/46) and ulceration over leads/electrodes (2/46). A further 12/46 patients required planned stimulator unit revision for battery depletion. There were no systemic life-threatening complications nor any complications with long-term sequelae.

Conclusions: Occipital nerve stimulation is a safe treatment for the management of selected primary headache disorders, although there is a significant incidence of surgical complications requiring further operation.

F2-09: Chronic injury reduces the production of new neurons from endogenous stem cells in the adult human brain: a microenvironment effect

M. Zaben & W. P. Gray (Wessex Neurological Centre, Southampton, UK)

Objective: Despite overwhelming evidence for the presence of stem cells in the adult brain, these cells do not produce neurons after brain injury. This is true even in the permissive niche of the hippocampus where new neurons, important for memory consolidation, are normally produced throughout life. We wished to examine the hypothesis that the injured environment reprograms these stem cells and see if this was reversible.

Methods: We isolated neural stem cells from normal cortex and from the sclerotic hippocampus in five patients undergoing epilepsy surgery. Stem cells were either grown as free-floating spheres in an ideal environment or within a 3D tissue (Hi-Spots) generated from the hippocampus or cortex, and their ability to generate neurons quantified using BrdU and Neu-N immunohistochemistry.

Results: We show that stem cells from the uninjured cortex and the sclerotic (injured) hippocampus generate new neurons with equal efficiency when grown under ideal free-floating conditions. Interestingly, the numbers of newly-born neurons generated by matched stem cells grown in sclerotic hippocampal Hi-Spots were significantly lower (5.9 ± 1.4 cells/mm²) than those of the cortical Hi-Spots (23.7 ± 3.9 cells/mm²) (p < 0.001).

Conclusions: These results strongly implicate the microenvironment in defective hippocampal stem cell function in areas of chronic brain injury, and importantly from the viewpoint of brain repair, show that the effect is reversible.

F4-01: Can a mobile CT head scanner comply with UK ionising radiation regulations?

N. Barua, G. Stevens, N. Rowles, R. T. Foy, R. Loader & J. Palmer (Derriford Hospital, Plymouth, UK)

Objective: Mobile CT head scanning has a potentially important place in the neuro-intensive care (NICU). Implementing the Ceretom® mobile CT scanner requires compliance with Ionising Radiation Regulations 1999, and the requirement to ensure exposures of staff and public are restricted as low as reasonably practicable (ALARP). The aim of this study was to confirm and quantify dose estimates based on the manufacturer's iso-dose maps and produce a workable radiation policy for use in NICU.

Method: To ensure that exposures are ALARP, novel shielding was designed and incorporated into Ceretom®. The iso-dose scatter distribution was measured using an ion chamber at sequential 0.5 m and 1 m distances from the unit's bore to build a dose matrix over 8 m×8 m². All measurements were taken for a single axial scan at highest rated tube current with Perspex CT head and chest phantoms used as scatter centres. Results were simulated to obtain scatter distributions throughout the ward assuming a maximum 60 day patient stay and random scan distribution of patients.

Results: The dose matrix was used with a planning constraint of 300μSv/year, to all exposed persons in the vicinity of the scanner (consistent with UK guidance for a single radiation source). The shielding altered the radiation scatter profile behind and lateral to the scanner. With a 4 m×4 m² controlled area, a maximum of 42 scans per bed per year could be permitted.

Conclusion: Novel shielding designed for the Ceretom® allows use in ward environments whilst ensuring compliance with UK ionising radiation regulatory requirements.

F4-02: Do pre-assessment clinics work? Training and resource implications

R. K. Mathew, K. Abid & D. Crimmins (Leeds General Infirmary, Leeds, UK)

Objectives: To identify the proportion of elective neurosurgical patients that need surgical optimisation pre-operatively. To assess if pre-admission clinic (PAC) is an efficient patient management resource.

Methods: A retrospective analysis of 100 successive patients taken from pre-admission clinic registers and operative records over a time period 1st August – 30th November 2008.

Results: There were 49 females and 51 males, ranging from 17 to 88 years old (mean 55.2 years). Over a quarter (26%) of patients had medical issues that needed pre-surgical optimisation. Of these, 91.4% could be identified through the initial neurosurgical consultation, whereas 8.6% were identified through blood tests. PAC identified 4% of patients with irresolvable medical problems leading to them being indefinitely delayed for surgery. Only half the patients seen in PAC had surgery within 90 days; the remainder required re-assessment on admission.

Conclusions: PAC has a role to play in the pre-operative optimisation of elective patients, not least in avoiding the incidence of on-the-day cancellations due to abnormal blood results. The vast majority of patients however would not need to attend PAC if a detailed past medical history were to be obtained in the initial neurosurgical consultation. In an environment of long waiting lists, the value of PAC diminishes if patients do not undergo surgery within a timely fashion. Furthermore, PAC does not streamline the workload of junior doctors as half of the patients require re-clerking and repeat investigations on admission.

F4-03: Prospective audit of consent in neurosurgery

C. Heaver, S. Kuruvath & J. Wasserberg (Department of Neurosurgery, Queen Elizabeth Hospital, Birmingham, UK)

Aim: A prospective audit of the consent procedure and levels of patient satisfaction in adult neurosurgical patients.

Methods: Prospective audit of all patients consented for neurosurgery in a 7-day period. Data were gathered from case note and from a structured patient questionnaire.

Results: A total of 47 patients underwent surgery. 25% had craniotomies, 40% were spinal procedures, 13% shunt operations and others accounted for 26%. Mean age was 52 years (Range 18–83). Twenty-six were males and 21 females. Seventy-five percent of consents were completed by registrars. Ninety-two percent of patients had the side marked. In the consent form, intended benefits of surgery, general and specific risks were filled in 96, 83 and 81% cases, respectively. Copy of consent was given to 16 patients. From the patient questionnaire, 97% felt satisfied why they needed surgery, 79% on what would happen during the surgery, 79% on the risks involved, 53% on the alternatives to surgery, and 71% on what would happen if left untreated and 76% on the post-operative care. Ninety-one percent were satisfied with the opportunity to ask questions. Ninety-one percent expressed satisfaction with the overall process.

Conclusions: The majority of patients were satisfied with the consent process. An unexpectedly high number were consented by trainees. We conclude that there is room for further improvement in line with the latest GMC and DOH guidelines.[1,], [2]

F4-04: Consent: an event or a memory in lumbar spinal surgery? A multi-centre, multi-specialty prospective study

E. A. C. Pereira¹,², W. B. Lo³, M. J. Gillies¹, P. J. Grover⁴, H. Chandran⁴, D. Nandi⁴, S. Bojanic¹, J. Wilson-MacDonald², J. C. T. Fairbank², S. P. Harland³, & T. A. D. Cadoux-Hudson¹ (¹Department of Neurological Surgery, The West Wing, Oxford, UK; ²Nuffield Orthopaedic Centre, Oxford, UK; ³Department of Neurosurgery, Queen Elizabeth Hospital, Birmingham, UK and ⁴Department of Neurosurgery, Charing Cross Hospital, London, UK)

Objective: To assess written consent for lumbar spinal surgery and its concordance with patient recall of risks at time of surgery.

Methods: A multi-centre, prospective patient interview, casenote and consent form review at time of surgery. Consent forms of all patients receiving elective lumbar spinal decompression procedures (including revision and multi-level surgery) without instrumentation for degenerative conditions at three neurosurgical centres and one orthopaedic centre. During three months from September to November 2008 were reviewed. Available patients were interviewed within 48 h of surgery and their recollection of surgical benefits, risks and alternatives recorded.

Results: One hundred and fifty three consent forms were reviewed (mean 38 per centre, S.D. 11) and 111 patients interviewed (mean 28 per centre, S.D. 8). Eighty-three percent of procedures were single-level. Common risks documented included infection (99%), dural tear (95%), bleeding (88%), recurrence or failure to relieve symptoms (55%), paralysis (54%) and sphincter damage or cauda equina syndrome (54%). Ten percent were operated upon by a different surgeon from that performing written consent. Peri-operative patient recall was poor for risks (paralysis 37%, sphincter damage 29%, infection 18%, dural tear 16%, bleeding 14%, recurrence 14%). Seniority of consenter did not affect patient recall significantly.

Conclusions: Informed consent obtained for elective lumbar spinal surgery is sometimes not completely documented independent of surgeon seniority. Patient recall is poor for risks. A positive correlation between poverty of risk recollection and time from written consent to surgery favours the obtaining of consent as near to time of surgery as possible.

F4-05: A study of impact of the European Working Time Directive (EWTD) on neurosurgical training: A quantitative assessment of reduction in trainee operative numbers, clinic attendances and ward rounds

A. A. Kamat, N. A. Farooqi & T. A. Carroll (Sheffield, UK)

Objective: The impact of the EWTD on surgical training has been documented in other specialities. The impact of EWTD has yet to be quantitatively explored in neurosurgery. We set out to quantitatively examine the effect of EWTD in a UK NHS neurosurgical unit in the context of recent qualitative statements from UK neurosurgical trainees and the SBNS.

Methods: The 6 month periods August–December 2007 and August–December 2008 were compared re trainee operative exposure and clinic attendances. Operative data, collected prospectively, was retrieved from the Trust theatre management database system. Clinic attendance data, also prospective, was obtained from the departmental medical staff scheduler. Trainee diaries provided supplemental data.

Results: Work patterns moved from non-resident 1-in-7 on-call (72 h compliant) to 1-in-8 full shift (56 h compliant) from August 2008. Two thousand five hundred forty six cases were performed over the two 6 month periods. There was a significant reduction in the number of trainee-involved cases during working hours following August 2008 with 42% of such cases being performed without a trainee present. To maintain a minimum of 225 cases per ST trainee per year (ST3 to ST8, as per core curriculum), mandatory non-specialist clinic attendance was reduced by 85% from August 2008.

Conclusions: We show that EWTD implementation has resulted in significant reduction in clinical exposure, including operations occurring during working hours. EWTD implementation with maintenance of operative numbers to meet core curriculum requirements was in particular at the expense of clinic attendance. We also present a quantitative 48 hour model that explores operative numbers and the further clinic attendance attrition required for EWTD August 2009 implementation. Our data questions the appropriateness of the August 2009 48 hour EWTD in neurosurgery.

F4-06: Incidence and management of postoperative intracranial haematomas following elective cranial procedures

J. Tailor, M .Schiariti, H. El-Maghraby & N. Kitchen (The National Hospital for Neurology and Neurosurgery, Queen Square, London UK)

Objective: Postoperative intracranial haematomas are serious complications with potentially fatal consequences after cranial surgery. The rate of postoperative haematoma requiring surgical evacuation is quoted between 0.8 and 2%. However, the true incidence of CT scan confirmed postoperative intracranial haematomas and the need for subsequent surgical evacuation remain unknown. This study was designed to assess the incidence of CT scan confirmed postoperative intracranial haematoma following neurological deterioration in elective cranial procedures and its management in respect to the need for surgical evacuation.

Methods: All elective cranial procedures performed at our institution from 1st August 2007 to 31st July 2008 were enrolled in this study. All postoperative CT scans performed during the same inpatient stay were reviewed with respect to postoperative intracranial haematomas and the need for further surgical intervention was assessed.

Results: From a total of 2794 neurosurgical procedures performed in 1 year, 1040 were elective cranial operations. Of these, 311 patients had postoperative CT scans during the same inpatient stay. CT scan confirmed postoperative haematomas were reported in 104 patients (10%). These were most commonly seen following meningioma and glioma debulking and the commonest site of haematoma were intraparenchymal and subdural. Of the 104 patients, 16 cases (15%) required surgical evacuation. The overall incidence of postoperative haematoma requiring surgical evacuation in elective cranial procedures is 1.5%

Conclusion: In our institution, the incidence of CT scan confirmed post-operative intracranial haematoma following elective cranial procedures is 10%. Only 15% of these patients required further surgical evacuation.

F5-01: Spinal multidisciplinary meetings: Reforming spinal services to achieve 18 week targets

S. Hettige, K. Kok, M. Whitton, S. Sadler, R. P. Selway & D. Bell (King's College Hospital, London, UK)

Objective: High demand spinal services require innovative solutions to achieve 18 week targets. We previously described the introduction of a novel multidisciplinary team meeting (MDM) to triage all spinal referrals to the most appropriate out-patient speciality. Bi-weekly MDMs consisting of two neurosurgeons, a neuroradiologist, a pain specialist and a physiotherapist assessed all new spinal referrals in order to decide the most appropriate treatment. Our objective was to assess the decision making and impact on waiting times of a novel spinal multidisciplinary meeting.

Methods: One hundred fifty-five patients discussed in the MDM were assessed and compared with a similar group managed prior to the introduction of the MDM. Outcomes included appointments with neurosurgery, pain clinic, physiotherapy or rheumatology.

Results: The percentage of patients given neurosurgical clinic appointments was halved to 50.9% of all new spinal referrals with new MDM system. The number of patients being offered surgery from clinic increased from 23% to 45%. The mean waiting time for those patients who were listed for surgery was reduced from 176 days to 76 days.

Conclusions: A multidisciplinary spinal meeting has halved the number of patients being seen in the neurosurgical outpatient department and reduced waiting times for new clinic appointments and surgery. Further evaluation is required to assess the impact upon other specialities and to ensure patient and referrer satisfaction.

F5-02: Specialist primary care triage of spinal patients – a 1300 case experience

D. R. Sandeman & K. Greenfield (Department of Neurosurgery, Frenchay Hospital, Bristol, UK)

Objective:To review the efficacy of a ‘One stop’ spinal assessment service where all GP referrals with back pain underwent Lumbar MRI scanning and immediate specialist assessment.

Subjects: One thousand three hundred ninety five consecutive referrals from North Bristol GP's of patients with lumbar spine pathology referred as part of a pilot service.

Methods: All patients were assessed in a ‘One Stop’ assessment clinic following full sequence lumbar MRI scanning by either a consultant neurosurgeon or experienced spinal physiotherapist, before being triaged to the appropriate care pathway.

Results: Of the 1395 patients referred, 1301 were assessed, 640 by a physio and 661 by a neurosurgeon. One thousand two hundred fifteen patients had MRI scans to inform the assessment (96%). Five hundred nineteen had surgical pathology (40%) – Disc prolapse (295), Stenosis (148), Lithesis (49) being the main surgical pathology groups. Eight patients had previously undisclosed tumours. The majority of the remainder had age related degenerative change (676). Seventy-one (5%) had normal scans. In these the scan, result was useful in identifying an important group of patients with complex back pain. Six hundred thirty seven patients were discharged following the assessment and managed in primary care. Six hundred forty patients were referred to secondary care – 161 had surgery (12%), 406 (31%) were referred to pain clinic. Only 149 (11%) required follow up, producing closure in a single assessment of 89% of patients.

Conclusion: Early specialist assessment of patients with spinal pain with MRI scanning is an efficient way of managing this large group of patients, but requires the involvement of secondary care expertise.

F5-03: A review of C1/ C2 fusions

K. A. Abid, R. K. Mathew, P. Sinha, R. Rodriguez, J. Timothy, S. Ross & G. Towns (Leeds General Infirmary, Leeds, UK)

Objective: To examine outcomes and associated morbidity and mortality of patients undergoing C1/C2 fusions over the previous decade in a neurosurgical unit.

Methods: This is a retrospective case series. A hospital database was used to identify all patients who had C1/C2 fixation/fusion operations between 1998 and 2008. Eighty-five available notes of a possible cohort of 106 records were examined and data collated.

Results: A wide range of surgical procedures were employed. The majority (94%n = 35) experienced either equivocal (42.9%) or a noticeable decrease (51.4%) in their pain score post-procedure. Ninety percent (n = 20) experienced either equivocal (60%) or a noticeable improvement (30%) in objective neurology post-procedure. Surgical complication rates were found to include 5.9% of patients with injury to the vertebral artery, 2.4% of patients with dural injury and 11.8% wound infection rate (n = 85 in all cases). There was one death (1.2%) directly related to surgery. The majority (78.9%n = 19) of vascular injuries and intra-operative difficulties occurred in rheumatoid or trauma patients.

Conclusions: The results elucidate the outcomes of one of the largest C1/C2 fixation series. Successful fusion is the primary endpoint however the majority of patients additionally experienced either no change or a diminution of symptomatic pain or neurology after the procedure. Complication rates in this series are higher than those quoted in earlier studies although the inclusion criteria for these procedures remain undetermined and we present a particularly complex caseload. Our results indicate that these procedures should only be undertaken by specialist cervical spinal surgeons after appropriate cadaveric training.

F5-04: Anterior cervical discectomy: Does it work for patients?

P. Goetz, N. Dawe, T. Ibrahim & D. Choi (The National Hospital for Neurology and Neurosurgery, London, UK)

Objective: Patients undergoing anterior cervical decompression and cage fusion (ACDF) for cervical spondylosis are often perceived by surgeons as making good neurological recovery. However, patient health perception following surgery may differ.

Methods: One hundred twenty two patients who underwent ACDF in a single centre between January 2004 and November 2007 were sent a postal questionnaire to assess post-operative health perception. Outcome measures included: SF-36, myelopathy disability index (MDI) and neck pain disability index (NDI). Patients were dichotomised into early (surgery 0–2 years prior) and late follow-up (2–4 years). Mann-Whitney test was used to compare outcomes in the two groups. Spearman test was used to determine whether correlations existed between outcome measures (significance p < 0.05).

Results: Ninety patients responded. Mean values were 38 (SF-36 physical), 46 (SF-36 mental), 50 (NDI as %) and 28 (MDI as %). Late follow-up SF-36 physical scores were significantly better than early follow-up (34 vs. 42, p = 0.003). None of the other outcome scores differed significantly with time: 46 vs. 46 (SF-36 mental), 49 vs. 51 (NDI) and 25 vs. 31 (MDI). There was a significant negative correlation between NDI and both components of SF-36 (p = 0.002 and p < 0.001 respectively) but no correlation between MDI and SF-36.

Conclusions: When compared to published data, this patient group had better health perception than for end-stage rheumatoid disease but worse than chronic asthma. Although clinicians feel that patients do well after ACDF, patients often do not agree. This may be in part because of the tendency for clinicians to use clinical rather than functional outcome measures.

F5-05: BMP-7 (OP-1®) Safety in anterior cervical fusion surgery

J. C. D. Leach & R. G. Bittar (The Alfred Hospital, Melbourne, Australia & The John Radcliffe Hospital, Oxford, UK)

Objective: To evaluate the safety of the use of bone morphogenetic protein-7 (BMP-7) in anterior cervical fusion surgery in the light of recent FDA warnings.

Methods: Over a 12-month period, 132 patients underwent an anterior cervical fusion using cages, with or without an anterior cervical plate. One hundred twenty four patients had a combination of BMP-7 (one half to one unit; 1.75–3.5 mg BMP-7 and 0.5–1.0 gm collagen,) and tri-calcium phosphate placed within the cage; eight patients had tri-calcium phosphate alone. Patients were evaluated prospectively for early adverse outcomes, including respiratory difficulty, need for re-intubation, dysphagia requiring further investigation or treatment, re-operation, readmission and mortality. Patients who underwent X-rays 24–72 h post-operatively had these films examined retrospectively by a blinded party. Pre-vertebral soft-tissue swelling at the level of the C6 vertebral body was measured and compared with a historical post-operative anterior cervical fusion cohort from the literature.

Results: Patients of 3/124 (2.4%) experienced post-operative complications: one case of transient recurrent brachalgia, one case sudden onset of dysphagia and dysphonia and one case of persistent dysphagia. Mean pre-vertebral soft-tissue measurement at C5/6 in the 20 patients from the BMP-7 group was 20.9 mm (16–27 mm). This compared with 18.7 mm (15–25 mm) in the non-BMP-7 group, and 18mm in the historical control group.

Conclusions: BMP-7 can be used safely in anterior cervical fusion surgery. A slight increase in post-operative pre-vertebral swelling in such patients is not clinically significant. The effect of BMP-7 on the rate and timing of fusion, as well as clinical outcome, is yet to be elucidated.

F5-06: Cervical laminectomy or laminoplasty. Who cares?

B. Nurboja, C. Kachramanoglou & D. Choi (The National Hospital for Neurology and Neurosurgery, London, UK)

Objectives: To compare the outcome and quality of life after cervical laminectomy and laminoplasty.

Methods: Retrospective study of 101 patients (Jan 2004 to Sep 2008) who received either cervical laminectomy (56) or laminoplasty (45) for the treatment of cervical spondylotic myelopathy was performed. The choice of operation was determined usually by the surgeon's experience or preference. We compared the quality of life, pain and radiological features before and after laminectomy and laminoplasty to determine any differences or associations. t-test (p = 0.05) was used to compare groups.

Results: Patient epidemiological data were comparable, except that the laminoplasty group generally involved more lower cervical levels than laminectomy. Surgical time was longer for laminoplasty (120 min) when compared with laminectomy (90 min) but this was non-significant (p = 0.065). Radicular arm pain did not improve in both groups with surgery. However, neck pain did improve in both the laminectomy and laminoplasty groups, with no significant difference between groups. EQ5D best health score improved and EQ5D quality of life improved with laminectomy but not significantly after laminoplasty, but this may be explained by the greater extent of laminoplasty operations (median, three cervical levels) when compared with laminectomy (median, two cervical levels), (p = 0.009). Surprisingly there was more tendency to post-operative kyphosis in the laminoplasty group (p = 0.026).

Conclusions: Laminectomy and laminoplasty are both effective methods of treating cervical spondylotic myelopathy and result in improved quality of life and neck pain. However, radicular arm pain is not treated well by these operations. Our results show that progressive kyphotic deformity is not prevented by laminoplasty.

F5-07: Benefits of minimally invasive spinal fusion

N. A. Hamid, N. Abbas, A. Harries & S. Ushewokunze (Queen Elizabeth Hospital, Birmingham, UK)

Objectives: The advent of new techniques and technology and a better understanding of surgical anatomy have resulted in newer minimally invasive techniques for many spinal operations. We present 24 cases of spinal fusion and pedicle screw fixation using different minimally invasive techniques to describe the merits of minimally invasive spinal fusion and to discuss the potential pitfalls of limited exposure. To describe an overview of the surgical techniques used in different operations.

Methods: During a period of 30 months (May 2006 to November 2008), 24 cases of interbody fusion and pedicle screw fixation using the new minimally invasive techniques were performed by a single surgeon. A tubular retractor was used for discectomy and cage insertion. Pedicle screws and rods were placed percutaneously or through the tubular retractor system. The different techniques used included minimally invasive transforaminal interbody fusion (TLIF), minimally invasive posterior interbody fusion (PLIF) and mini-open single and multi level PLIF. Outcome measures comprised operative time, complications, blood loss and length of post-operative stay. The data were extracted from hospital records and radiological reports.

Results: There was a significant reduction in length of hospital stay. The average blood loss was 150 ml and no patient required blood transfusion. Wound drains were not required in any case. There was no dural tear, no infection and no patient required revision surgery. Though the operative time was slightly longer that the conventional open approach, this appeared to be justified considering the excellent cosmetic result, the minimal muscle damage and a decrease in the post-operative pain.

Conclusion: Our initial experience using the minimally invasive approach has shown many advantages without compromising clinical outcome.

F5-08: Results of an observational cohort of 100 consecutive patients undergoing transforaminal lumbar interbody fusion

M. Al-Mukhtar, S. Kazzaz, J. Allibone & A. T. H. Casey (Spinal Surgery Unit, Royal National Orthopaedic Hospital, Stanmore, London, UK and The National Hospital for Neurology and Neurosurgery, Queen Square, London, UK)

Objective: To assess the clinical outcome and patient satisfaction following Transforaminal Lumbar Interbody Fusion (TLIF) for Spondylolisthesis and Degenerative Disc Disease.

Methods: Observational prospective study, longitudinal cohort on 100 patients, (Male = 45; Female = 55), with mean age of 50 (range 24–81 years) undergoing TLIF (2004–2007).

Outcome measures: Visual Analogue Score (VAS) for back and leg pain, Oswestry Disability Index, Zurich Claudication scale and SF-36. Follow up was performed using a standardised proforma which included clinical outcome and X-ray at 6 weeks, 3 months, 6 months, 1 year and 2 years with mean follow up of 18 months. CT scan was performed if fusion status was uncertain.

Results: Statistical analysis using parametric and non-parametric analysis with significance accepted at p < 0.05. Pre operative VAS for back pain median of 9.0, (mean 9.0; SD 1.63 and 95% CI 0.44). Pre operative VAS for leg pain median 9.0, (mean 7.5; SD 3.6 and 95% CI 0.98). Post operative VAS for back pain median of 5.0, (mean 4.7; SD 3.14 and 95% CI 0.86). Post operative VAS for leg pain median of 0, (mean 2.7; SD 3.39 and 95% CI 0.93).

Conclusions: We have had favourable experience with pedicle screw and TLIF for lumbar instability with a low failure rate and good patient satisfaction judged by halving of back and leg pain median value of zero. Oswestry, Zurich claudication scale and SF-36 showed significant changes.

F5-09: Three level lumbar disc replacement – 2 year follow up

J. C. Sutcliffe & A. Dadds (The London Spine Clinic, London, UK)

Objectives: To study the clinical outcomes and complications of patients undergoing three-level lumbar disc replacement at a single institution.

Methods: Over a 5 year period, 21 patients presented to the senior author with disabling discogenic back pain, unresponsive to conservative management and in whom discography demonstrated low pressure, concordant pain reproduction at three contiguous lumbar levels, with healthy neighbouring discs. Fourteen of these patients have 2 year follow up and are presented here. Patients aged 31 to 52 years at surgery, four female and 10 male. Ten patients underwent surgery utilising the Charite disc prosthesis and four the Prodisc^†. All surgeries were performed via a left retro-peritoneal approach, through a horizontal incision in 13 and a vertical one in 1. The recti were parted and retracted bilaterally in 13 and the left rectus was divided in one.

Results: Average operating time was 135 min; average blood loss 1180 mls. Average hospital stay was 8 days. There were no major complications. Most patients had transient paralytic ileus longer than 24 h, none requiring additional treatment. There were no wound infections or hernia. There was one case of retrograde ejaculation which recovered after 3 months and 1 case of transient sympathectomy. There was no ill effect from division of the rectus muscle.

Oswestry scores fell from 49 to 4.4 at minimum 2 years. Thirteen of these 14 patients (93%) were pain free at 2 year follow-up. Average VAS scores fell from 7.5 to below 1 in the same period.

Conclusions: Multi-level disc replacement in the lumbar spine is increasingly becoming advocated in patients for whom surgical treatments had previously seemed inappropriate. Three level surgery remains a very major procedure, but has the advantage, in cases as outlined here, of dealing with all the pathology and hence resulting in generally good clinical outcomes. The complication rate is no higher than in the single or double level cases performed during the same time frame and the recovery, with appropriate intensive rehabilitative support, is similar, but more time consuming to manage.

* DePuy Spine Inc, Raynham, MA USA.

^† Aesculap, B Braun.

F5-10: X STOP interspinous distraction device in lumbar canal stenosis: Is early success really sustained?

R. Dubey, K. S. M. Prasad & F. P. Nath (James Cook University Hospital, Middlesbrough, UK)

Objective: To report the long term outcome of X STOP interspinous device in lumbar canal stenosis (LCS) at our centre.

Methods: Retrospective case note review was done and success was evaluated using post operative clinic review and outcome measures in the form of the Swiss Spinal Stenosis Questionnaire (SSSQ) and Oswestry Disability Index (ODI) completed retrospectively by 39 out of 41 patients. Operative time, hospital stay, complications and follow-up duration were also assessed.

Results: Over a period of 3.5 years, 41 patients underwent insertion of X STOP interspinous device with a clinical and radiological diagnosis of LCS. Mean age was 71.6 years (M:F::23:18). L4-5 was the commonest level operated. Majority had single level insertion (29), nine had two levels and three patients had insertion of implant at three levels. The 12 mm size implant was used most commonly. Mean operative time was 55 min and the average hospital stay was 4.3 days. Complications were seen in the form of two superficial wound infections, one serous wound collection, one intra-operative CSF leak and one episode of displacement of device needing repositioning. Mean follow-up duration was 34 months (range 17–58 months). Seven patients had removal of X STOP and formal decompression and another two are on waiting list. Using the SSSQ the overall rate of good outcome was 33% in all component domains together. Using ODI, pre-operatively 10 patients (26%) were minimally or moderately disable (scores 10–20% or 20–40% respectively). After surgery, this number increased to 22 (56%), an improvement of 30%, still leaving 17 patients (44%) with either severe disability (score 40–60%), crippled (score 60–80%) or bedbound (score 80–100%).

Conclusions: The overall rate of good outcome from SSSQ (33%) & from ODI (30%) complement each other and is significantly lower to the 86% early success rate using SSSQ quoted in our study done on 22 patients in 2006. These results are significantly less favourable to the published figures (64% good results) for lumbar decompression.

Notes

**DePuy Spine Inc, Raynham, MA USA

^#Aesculap, B Braun