Purpose: To report long-term results (with a minimum follow-up of 13 years) of GR construct [Luque-trolley (LT)] in EOS, to identify factors predictive of curve progression and to establish the timing of definitive fusion.

Method: The study cohort consisted of 37 patients (22M & 15F) who had primary LT between 1983–1995 were reviewed. Group I: 7 patients had LT alone and Group II: 30 had LT with convex fusion. Cobb at initial presentation, after first surgery, before definitive fusion and at the latest follow-up was recorded. Other radiological curve parameters recorded were rib spinal angle difference (RSAD), end vertebral tilts (EVT), apical vertebral rotation (AVR) and T1-S1 length. Complications with respect to development of junctional/apical kyphosis, implant failure, pseudoarthrosis (PA), sagittal/coronal profile and instrumented spinal segment growth at maturity were evaluated.

Results: The mean age at definitive fusion for study cohort was 12.5 years. Group I: Mean age at first surgery was 7.4 years (3.3–9.5y). Mean pre-op Cobb angle of primary curve was 600 (310–710) which was corrected to 280 (200–360). They underwent definitive segmental spinal instrumentation(SSI) with fusion at 13.9 years (9.8–15.1y) when the curve had worsened to 480 (400–650). Group II: Mean age at index surgery was 3.6 years (1.6–8.8y). Mean pre-op Cobb of primary curve was 580 (300–900) which corrected to 300 (100–620). 16/30 patients underwent definitive SSI with fusion at 11.5 years (8.5–14.2y) when the curve deteriorated to 600 (530–770). Instrumented segmental spinal growth was 3.2cms (SD±1.45; range 1–5cms). 14/30 maintained their correction till skeletal maturity. JK was observed in 8 cases [proximal(3), distal(2) & apical(3)] which were corrected at the time of definitive SSI. There was a linear relationship between Cobb angle at definitive fusion with concaveRSA and upperEVT.

Conclusion: Correlation and regression statistics revealed predictive factors of curve progression to be concave RSA (®=0.91 & p=0.001) and upper EVT (®=0.81 & p=0.0004). Patients with high concave RSA and upper EVT should be closely monitored for deterioration. Spinal growth that exceeds the capacity of LT to elongate leads to apical kyphosis. Timing of definitive fusion is influenced by growth velocity, clinico-radiological factors and complications.

Purpose: Intra-op localisation of small nidus in Osteiod osteoma and Osteoblastomas is often difficult resulting in failed excision with persistent pain. We report two year follow-up results of the efficacy and reliability of using an intra-operative gamma probe in conjunction with fluoros-copy to aid resection in primary and revision surgeries.

Method: Eight patients (6M; 2F) with a diagnosis of osteoid osteoma (7) and osteoblastoma (1) were seen at our centre. The mean age at presentation was 20.9 years (9–31y). The tumour was localised to cervical (2), thoracic(4) and lumbar (2) posterior elements. All had back or neck pain of varying duration with a mean of 20 months (6–48mo). Three patients had failed treatments including CT-guided radiofrequency ablation in one and surgical excision under fluoroscopy in two. No case had previously utilised an intra-op gamma probe for localisation. All patients had work-up with plain X-rays, CT, MRI and 99 m Technetium bone scan to identify and localise the lesion. A pre-requisite for use of intra-op gamma probe was a positive pre-op bone scan. On the day of surgery, 600 MBq Tech HMDP (hydroxy-methylene-di-phosphate) was administered IV 3 hours prior to surgery. Fluoroscopy was used to confirm anatomical level, permanent mark made on skin and area exposed surgically. A 5 mm cadmium telluride (Cd Te) probe (which converts gamma radiation into electrical signal) and rate meter were used to scan the area containing lesion and counts per second(cps) recorded. The tumour nidus was then excised and cps from tumour bed and excised specimen recorded.

Results: The mean follow-up was 5.85 years (2–12.3y). The mean cps for osteoid osteoma pre-excision was 203.8 (60–515), which fell to 72.5 (10–220) post-excision. The cps reduced from 373 to 40.5 post-operatively for Osteoblastoma. Complete excision was recorded every time and all patients reported characteristic disappearance of pre-operative pain. All had discontinued analgesic medication and returned to normal activity by three months. All patients were followed-up regularly when they filled NDI, ODI and SF-36.

Conclusion: Gamma probe guided surgical excision facilitates accurate localisation of lesion, is less invasive and most importantly confirmation of complete excision of the tumour nidus consistently every time.

We describe the clinical outcome of a technique of surgical augmentation of chronic massive tears of the rotator cuff using a polyester ligament (Dacron) in 21 symptomatic patients (14 men, seven women) with a mean age of 66.5 years (55.0 to 85.0). All patients had MRI and arthroscopic evidence of chronic massive tears. The clinical outcome was assessed using the Constant and Murley and patient satisfaction scores at a mean follow-up of 36 months (30 to 46).

The polyester ligament (500 mm × 10 mm) was passed into the joint via the portal of Neviaser, medial to the tear through healthy cuff. The two ends of the ligament holding the cuff were passed through tunnels made in the proximal humerus at the footprint of the insertion of the cuff. The ligament was tied with a triple knot over the humeral cortex.

All the patients remained free from pain (p < 0.001) with improvement in function (p < 0.001) and range of movement (p < 0.001). The mean pre-operative and post-operative Constant scores were 46.7 (39.0 to 61.0) and 85.4 (52.0 to 96.0), respectively (p < 0.001). The mean patient satisfaction score was 90%. There were two failures, one due to a ruptured ligament after one year and the other due to deep-seated infection. The MR scan at the final follow-up confirmed intact and thickened bands in 15 of 17 patients.

This technique of augmentation gives consistent relief from pain with improved shoulder movement in patients with symptomatic massive tears of the rotator cuff.

Background: First MTP joint arthrodesis is a well established and very common procedure for painful arthrosis. Plate fixation method has been used with successful outcome (97–100%) by few authors but it is yet to be accepted universally for fear of complications.

Aim: To evaluate clinical and radiological outcome of first MTP fusion using low profile Acumed plate.

Methods: We retrospectively reviewed 125 patients who had 1st MTP arthrodesis (over 6 year period) for painful Hallux Rigidus not relieved by conservative means and for rheumatoid forefoot reconstruction. The preoperative evaluation included a subjective questionnaire, physical exam, AOFAS hallux score and radiographic measurements. Post-operatively, all patients were mobilised with heel weight bearing shoes for six weeks. All patients had follow up of minimum 6 months(range 6 months to 6 years). At the final follow-up all patients had answered a questionnaire which evaluated any limitations of daily activity and restrictions in footwear. Radiological measurements included union of the arthrodesis and various angles (valgus, intermetatarsal and dorsiflexion).

Results: Of the 125 patients we had final reviews for 103 patients. The mean AOFAS improved from 40 to 82. The individual components of AOFAS i.e. pain, walking ability and alignment improved significantly. All patients but one had radiological evidence of fusion at mean of 6 weeks (range 6–8weeks) allowing them to walk with normal footwear. The mean dorsiflexion angle was 15° (range 13 °–18 °). The patient with non-union had re-arthrodesis with bone grafts using the revision plate. Two patients with rheumatoid arthritis required removal of plate for infection and wound breakdown. No plate failure occurred in any of the patients.

Conclusion: The plate fixation is a reliable method for 1st MTP joint fusion that allows for a predictable fusion in a satisfactory alignment with low complication rate. The stability of the fixation allows for early mobilization without need for plaster immobilization and early return to functional activities.

In order to treat painful subluxation or dislocation secondary to cerebral palsy, 11 patients (12 hips) underwent combined femoral and Chiari pelvic osteotomies with additional soft-tissue releases at a mean age of 14.1 years (9.1 to 17.8). Relief of pain, improvement in movement of the hip, and in sitting posture, and ease of perineal care were recorded in all, and were maintained at a mean follow-up of 13.1 years (8 to 17.5). The improvement in general mobility was marginal, but those who were able to walk benefited the most.

The radiological measurements made before operation were modified afterwards to use the lateral margin of the neoacetabulum produced by the pelvic osteotomy. The radiological migration index improved from a mean of 80.6% (61% to 100%) to 13.7% (0% to 33%) (p < 0.0001). The mean changes in centre edge angle and Sharp’s angle were 72° (56° to 87°; p < 0.0001) and 12.3° (9° to 15.6°; p < 0.0001), respectively. Radiological evidence of progressive arthritic change was seen in one hip, in which only a partial reduction had been achieved, and there was early narrowing of the joint space in another. Painless heterotopic ossification was observed in one patient with athetoid quadriplegia. In seven hips the lateral Kawamura approach, elevating the greater trochanter, provided exposure for both osteotomies and allowed the construction of a dome-shaped iliac osteotomy, while protecting the sciatic nerve.

Introduction: Early non progressive horizontal RLLs (< 2mm) under the tibial component following cemented TKR have been noted to be due to poor cement injection into cancellous bone. They may facilitate the entry of joint fluid and wear debris into the interface, which may proceed to ballooning osteolysis. At present, there is no consensus on the preferred cementing technique (single mix versus dual mix cementation) in TKR.

Purposes of the study: To assess RLLs in immediate postoperative radiographs in cemented TKRs at the cement-bone and cement-implant interface.

To compare the RLLs following single mix and dual mix cementation techniques.

Study Design: Prospective, consecutive radiographic analysis.

Material and Methods: 53 consecutive cemented TKRs in 39 patients (12: 27, F: M) with mean age of 72.5 years (range 50–90 y) who were operated on between 2001 to 2004 by the senior author (NMG). 27 had single mix and 26 had double mix cementation. Immediate postoperative radiographic assessment (AP and LAT standing view) was blinded for single mix versus dual mix cementation. All the radiographs were independently assessed by two of the authors for the presence of RLLs using the zonal pattern of the Knee Society scoring system. We have evaluated the RLLs in the cement-implant interface in a similar manner as described for the bone-cement interface.

Results: Most common TKR used was the Maxim (31) followed by the PFC (9). 29(54.7%) TKRs had RLLs (11in single mix : 18 in dual mix). There were more RLLs at the Cement-implant interface (29), than the bone-cement interface (10). In AP view, Zones 1(medial) and 4 (lateral) were the common sites for RLLs in both groups (< 2mm). In Lateral view, Zone 1 (Anterior) followed by Zone 2 (Posterior) were the common sites of RLLs (< 2mm). In AP view, there was no significant difference in the number of RLLs (Zone 4) between the two groups. The total number of RLLs in all zones (Zone 1–4) was significantly low in the single mix group (p< 0.05). There was no difference between the two groups in the lateral view.

Conclusions: Single mix cementation technique reduces the incidence of RLLs in the immediate postoperative radiographs following cemented TKRs.

Introduction The Zweymuller tapered cementless titanium femoral stem has been in widespread use since 1986.1 Stress shielding of the proximal femur has been a concern with pressfit cementless femoral stems. Radiolucent lines (RLLs) are signs of stress shielding and possibly over time, may lead to aseptic loosening.2

Aim To evaluate the longterm radiographic bone response in the femur following Zweimuller total hip replacements.

Study Design Consecutive case study (serial radiographic analysis)

Material and Methods 49 Total Hip Replacements in 42 patients (M:F=25:17) with a mean age of 59 years (range 49–70 years), were included in the study. All patients were operated on by the senior surgeon (JRW). 28 Alloclassic and 21 Endoplus stems were implanted. AP and Lateral radiographs were assessed. A gap of 1mm or more at the bone prosthesis interface was recorded as positive for RLLs in the Gruen zones. The mean duration of follow-up was 46 months (range 24–140 months). 17 patients had follow up of more than 5 years. Other measurements included subsidence, bone remodelling and heterotropic ossification.

Results After 2 years there were distinct radiological changes (RLLs), mainly in Gruen zones 1 and 7, in 18/49(36%) femora. Though there was evidence of RLLs in zones 2 and 6, the numbers were insignificant. Subsidence of more than 3mm was noted in 16 stems (33%). Heterotropic ossification was found in 4 patients (8%), one of whom required excision. Persistent pain due to trochanteric bursitis was noted in 10 patients. 4 patients needed revision due to reasons other than aseptic loosening. There were two dislocations, which needed revision of the acetabular component.

RLLs were more common in the Endoplus group (10/21) compared to the Alloclassic group (8/28). There was no clinical compromise (all had pain free mobility) in these patients and no progression of RLLs was noted.

Conclusion Proximal femoral stress shielding following Zweimuller femoral stem implantation is observed in significant number of patients. The RLLs do not correlate with symptoms and patient satisfaction.

We studied prospectively 22 young athletes who had undergone surgical treatment for lumbar spondylolysis. There were 15 men and seven women with a mean age of 20.2 years (15 to 34). Of these, 13 were professional footballers, four professional cricketers, three hockey players, one a tennis player and one a golfer. Preoperative assessment included plain radiography, single positron-emission CT, planar bone scanning and reverse-gantry CT. In all patients the Oswestry disability index (ODI) and in 19 the Short-Form 36 (SF-36) scores were determined preoperatively, and both were measured again after two years in all patients. Three patients had a Scott’s fusion and 19 a Buck’s fusion.

The mean duration of back pain before surgery was 9.4 months (6 to 36). The mean size of the defect as determined by CT was 3.5 mm (1 to 8) and the mean preoperative and postoperative ODIs were 39.5 (sd 8.7) and 10.7 (sd 12.9), respectively. The mean scores for the physical component of the SF-36 improved from 27.1 (sd 5.1) to 47.8 (sd 7.7). The mean scores for the mental health component of the SF-36 improved from 39.0 (sd 3.9) to 55.4 (sd 6.3) with p < 0.001. After rehabilitation for a mean of seven months (4 to 10) 18 patients (82%) returned to their previous sporting activity.

To determine how long after injury a single photon emission computed tomography (SPECT) scan may remain positive in cases of symptomatic posterior element lumbar stress injuries.

SPECT scans can identify posterior element lumbar stress injuries earlier than other imaging modalities. As these lesions evolve and the spondylolysis becomes chronic, the SPECT scan tends to revert to normal even though healing of the defect has not occurred. The aim of this study was to determine how long after initial injury a SPECT scan might remain positive.

One hundred and sixty-five patients (85 male, 80 female) between the ages of 8 and 38 years with suspicion of lumbar spondylolysis or posterior element lumbar stress injuries were investigated. All patients underwent plain radiographs, planar bone scintigraphy and SPECT imaging. The duration of symptoms at clinical assessment was recorded. The age, sex, symptom reproduction on flexion or extension, level of sporting activity, and the Oswestry Disability Index both pre- and post-treatment were also recorded. SPECT positive images (hot scans) were depicted as cases and SPECT negative images as controls. Univariate and multivariate analysis was performed.

Eighty-five patients (63 male, 22 female) had positive SPECT scans (cases); eighty had negative scans (controls). The mean age at onset of symptoms was 20.2 years for cases and 17.4 years for controls. Bilateral increased uptake on SPECT scan was more common than unilateral. The commonest site for increased uptake was the posterior elements of the fifth lumbar vertebra. Low back pain in extension was more common in SPECT positive cases. The mean time from injury / onset of symptoms to a positive SPECT scan was 7.1 months (range 5.2–9.2 months) and to a negative SPECT scan was 22.5 months (range 16.8–28.4 months).

Intense scintigraphic activity in the posterior elements of the lumbar spine was associated with a more recent injury and was concordant with the patient’s history and physical examination. Chronic, un-united spondylolysis was often scintigraphically occult. There was a window of approximately six months from the onset of symptoms to investigation after which the sensitivity of SPECT imaging diminished.

Recent years have seen the popularization of minimally invasive approaches to the spine.

However, the use of the balloon assisted retroperitoneal approach has not been widely described, moreover there has been no direct comparison between this mini-ALIF (anterior lumbar interbody fusion) and the conventional open method in the literature.

Comparison of peri and intra-operative parameters between the rnini-ALIF (using the balloon assisted dissector and Synframe retractor system) and the open midline approach for single and double level anterior lumbar interbody fusions in order to assess the efficacy of this procedure.

An independent retrospective evaluation of 35 patients who underwent single or double level ALIF under the care of the senior author at the University Hospital, Nottingham during the period from 1997 to 2000. The patients were split between those undergoing a mini-ALIF (balloon assisted retroperitoneal dissection) or the conventional approach via a larger midline incision. The groups were matched for age, sex and number of levels. Data was collated from the medical notes with regards to intra-operative blood loss, operative time, intra-operative complications, PCA requirements, time to mobilisation and length of hospital stay.

A statistically significant (p=0. 01) reduction in time to mobilisation (mean 2. 1 days vs 3. 9 days) and operative time (mean 175mins vs 265mins) was found for the single level mini-ALIF. This reflects the greater number of L5/SI fusions in this group. The number of vascular injuries was also greater in the approach to L4/5.

No difference was found between the two groups for double level procedures.

The immediate advantages of a less invasive approach both to the patient and hospital do not appear to be borne out by this study. Cosmesis was not assessed and the long term functional outcome awaits later confirmation.

A radiological and histological analysis of five knee joints after a minimum of 15 years following the implanting of carbon fibre which had been used as a treatment for knee instability was undertaken. All patients underwent Total Knee replacement for secondary osteoarthritis. Histological analysis demonstrated a variable amount of macroscopically visible carbon particles in the synovium, hyaline cartilage and menisci showed . At microscopy these particles were found enveloped by giant cells and lying quiescent with no active inflammatory changes. No intact carbon fibre ligament was noted within the joint, small portion of the old ligament were covered with a thin fibrous layer but there was no evidence of any structure resembling neo-ligament.

Extra articularly the carbon fibre was covered with a thick fibrous sheath with no active inflammatory changes inflammation. In the bone tunnels the carbon fibre- bone interface showed an apposition of the bone to the carbon fibre without any interposing fibrous sheath.

The histology suggests that carbon fibre bonds directly with the bone without fibrous interposition and that there is no evidence of synovitis changes related to the carbon fibre material.

The study suggest that although carbon fibre failed structurally as a ligament replacement it did not cause any significant long term inflammatory pathology.

Objectives: Comparison of peri and intraoperative parameters between mini-ALIF (using balloon assisted dissector and Synframe retractor) and open midline approach for single and double level ALIF.

Methods: Independent, retrospective evaluation of 35 patients split between those undergoing the mini-ALIF or the conventional approach via larger midline incision. Groups matched for age, sex and number of levels. Operations performed at University Hospital, Nottingham between 1997 and 2000.

Outcome measures: Data collated for operative time, intraoperative blood loss, complications, PCA requirements, time to mobilisation and hospital stay.

Results: Statistically significant (p=0.01) reduction in operative time (175 vs 265mins) and time to mobilization (2.1 vs 3.9 days) found for single level mini-ALIF. Complications namely vascular injuries were almost equal in both groups. No difference was found between the two groups for double level procedures.

Conclusion: The immediate advantages of a less invasive approach both to the patient and the hospital do not appear to be borne out by this study. Cosmesis was not assessed and long term functional outcome awaits later review.

Objective: To assess the clinical outcome and return to sport following surgical treatment of spondylolysis in young sporting individuals.

Design: A prospective outcome analysis of consecutive surgically treated cases of lumbar spondylolysis in young sporting individuals.

Subjects: Twenty-two young sports persons (15M: 7F) with a mean age of 20.2 years (range 15–34 years) were surgically treated for radiographically confirmed spondylolysis between 1994 and 1999. Eleven patients were professional footballers and four were professional cricketers. Pre operative assessment included plain X-rays, SPECT imaging with planar bone scan and reverse gantry CT scans. All subjects had pre-operative Oswestry Disability Index (ODI) and SF36 scores recorded. Eighteen patients underwent Buck’s fusion and four patients underwent Scott’s fusion. A graduated exercise regime was commenced at 12 weeks. At two year follow-up nineteen patients had ODI and SF36 scores recorded.

Outcome Measures: The clinical outcome in individual patients supported by statistical analysis of the pre operative and post-operative data was performed using SPSS (ver 10). Return to the sporting activity at the previous level was regarded as a successful outcome.

Results: Eleven patients had bilateral spondylolysis at L5. Twenty patients had positive uptake on SPECT imaging and the remaining two were diagnosed to have lysis on CT scans alone. The average duration of back pain before the patients underwent surgery were 8.4 months (range 3–36 months). The mean lysis defect determined by CT was 3.5 mm (range 1–8 mm). The mean pre-operative and post-operative ODIs were 40.5 and 12.4 respectively (SEpreop = 2.06 and SEpostop = 3.05). The mean scores of physical health component of SF36 improved from 27.1 to 47.8 (SEmean = 1.1 and 1.7 respectively). The mean scores of mental health component of SF36 improved from 39.1 to 55.3 (SEmean = 0.9 and 1.4 respectively) [P < 0.001]. Eighteen patients returned to their previous active sporting career following an average of seven months of rehabilitation (range 4–10 months).

Conclusions: The surgical repair of bilateral spondylolysis with Buck’s fusion in professional sportsmen and women results in a significant improvement in Oswestry Disability scores (P< . 001) and in all domains of SF36 health questionnaire (P< . 001). 90% return to active sport seven months following surgery.

Objective: To asses the outcome of hermivertebrectomy and fusion for symptomatic thoracic disc herniation.

Design: A retrospective case analysis

Subjects: Between 1993 and 1999, ten patients (M5, F5) were treated surgically for thoracic disc herniation by the two senior authors (JKW & SHM). The average age of patients at presentation was 5Oyears (range 32–77years). Two patients had two level disc herniations (total 12 disc herniation). The most common sites of disc herniation were at T10/11(4 patients). Duration of diffuse mid thoracic hock pain in eight patients varied from one week to six months. The initial neurological evaluation demonstrated weakness and spasticity of varying grades in eight patients, of which five had paraplegia and three had monoparesis. Sensory changes below the level of the lesion were found in eight patients. Sphincter dysfunction was noted in seven patients. Hemivertebrectomy followed by discectomy and fusion was carried out in all patients. Instrumentation with cages was performed in eight patients and only bone grafting in two patients. Spinal cord monitoring was used in all cases.

Outcome Measures: The average follow up was 24 months (range 13–36 months). Pre-operative and postoperative neurological grading was done using MRC grading for motor and sensory deficit. Asymptomatic patients with full activity were regarded as a successful outcome.

Results: Three patients had excellent, three had good, three had fair and one had poor outcome. Seven out of eight patients with cages had radiological fusion. The cage stabilises the segment and maintains the spinal height till bony fusion takes place. One patient with hone graft alone had recurrence of symptoms and had a re-surgery with a poor outcome. Six patients had residual back pain of varying degrees. One patient had atelectasis, which recovered within two days of surgery. One patient had suffered from complete paraplegia immediately after surgery detected by SSEPs. She underwent a MRI scan within the hour and was reoperated. She had complete corpectomy and instrumented fusion. At two years she was walking with a support.

Conclusion: Exposure of the norrnal tissue above and below herniated disc by hemivertebrectomy facilitates the safe removal of the disc and reduces the risk of further neurological damage. Cages were found to have advantages over autogenous strut only grafts. However, persistent back pain in some cases remains an unsolved problem.

Objective: To report a complete neurological recovery following cervical laminectomy and drainage of an extradural panspinal abscess in a patient with quadriplegia~

Design: A retrospective case analysis

Subject: Case report – A 63 year-old male was admitted to the hospital fever of 102.5F and four days old neck and back pain. On admission he was drowsy and short of breath. He was treated successfully with intravenous Penicillin for proven Pneumococcal meningitis three weeks prior to this admissjon On examination he had respiratory distress and quadripegia and upgoing plantar reflex with a sensory level below T2. He had urinary incontinence but his anal tone was preserved with intact bulbocavernosus reflex. He was ventilated for five days. The CSF culture grew Streptococcus pneumoniae. Once his breathing became normal i. e. seven day later a MRI scan revealed a diffuse extradural abscess extending from the cranio-cervical junction to the lower thoracic region posteriorly There were associated oedema and ischemic changes in the cord. He underwent a cervical lamitomy and decompression from C3–7. The intra-operative findings were pus and granulation tissue in the epidural space. The pus was drained and the infected granulation tissue was removed. He was continued postoperatively on intravenous Benzyl Penicillin The patient showed signs of neurological recovery from the third day onwards.

Outcome measures: Ravicovitch and Spallone (1982)¹ suggested a grading system to indicate the post-operative neurological outcome: 0-only signs of infection, 1-Root involvement, 2-Mild Spinal cord synptoms, 3-Severe spinal cord involvement, 4- Functional tresection.

Result: The patient was discharged three weeks following the surgery and was under a rehabilitation team. At six months follow-up he was walking without support and had MRC grade 5/5 power in both his upper and lower limbs. It has been recorded in literature that the duration of neurological symptoms has been shown to influence the functional outcome. If the neurologic grade 4 is present for more than 36 hours, little or no return of function could be expected. ^2,3

Conclusion: This case is unique. The reasons are: 1) Pneumococcal extradural abscess are extremely rare, 2) Ahscess posterior to the cord is also very rare, 3) A full neurological recovery even though the decompression was performed seven days following the episode.