Journal of Orthopaedic Surgery

To evaluate the outcomes, fusion rates, complications, and adjacent segment degeneration associated with transforaminal lumbar interbody fusion (TLIF).

32 men and 80 women aged 15 to 85 (mean, 57) years underwent 141 fusions (84 one-level, 27 2-level, and one 3-level) and were followed up for 24 to 76 (mean, 33) months. 92% of the patients had degenerative lumbar disease, 15 of whom had had previous lumbar surgery. Radiographic and clinical outcomes were assessed at 2 years. The short-form 36 (SF-36) health survey, visual analogue scale (VAS) for pain, and the modified North American Spine Society (NASS) Low Back Pain Outcome Instrument were used.

Of the 141 levels fused, 110 (78%) were fused with remodelling and trabeculae (grade I), and 31 (22%) had intact grafts but were not fully incorporated (grade II). No patient had pseudoarthroses (grade III or IV). For one-level fusions, poorer radiological fusion grades correlated with higher VAS scores for pain (p<0.01). All components of the SF-36, the VAS scores for pain, and the NASS scores improved significantly after TLIF (p<0.01), except for general health in the SF-36 (p=0.59). Improvement from postoperative 6 months to 2 years was not significant, except for physical function (p<0.01) and role function (physical) [p=0.01] in the SF-36. Two years after TLIF, 50% of the patients reported returning to full function, whereas 72% were satisfied. 26 (23%) of the patients had adjacent segment degeneration, but only 4 of them were symptomatic.

TLIF is a safe and effective treatment for degenerative lumbar diseases.

To review early radiographs of patients with de novo degenerative lumbar scoliosis to determine factors predicting early scoliosis progression.

Standing anteroposterior and lateral radiographs of 7 men and 20 women aged 48 to 83 (mean, 63) years with Cobb angles between >5° and <20° were reviewed. They were followed up for a mean of 12 (range, 10–18) years. Radiographic variables measured included (1) the Cobb angle, (2) the grade of rotation of the apical lumbar vertebra, (3) the presence of a lateral vertebral translation of ≥3 mm, (4) the degree of osteoporosis, and (5) the Harrington factor (the degree of scoliosis divided by the number of vertebrae involved).

During the follow-up period, the mean Cobb angle increased 5.3° from 10.1° to 15.4°, representing an increase of 0.4° per year. In initial radiographs, the apical vertebral rotation was rated as grade 0 in 3 patients, grade 1 in 19, grade 2 in 4, and grade 3 in one. A lateral vertebral translation of ≥3 mm was noted in 9 patients. The degree of osteoporosis was rated as grade 0 in 9 patients, grade 1 in 11, grade 2 in 5, and grade 3 in 2. The mean Harrington factor was 2.4. In the multiple regression analysis, only the grade of apical vertebral rotation was significantly correlated with scoliosis progression (regression coefficient=0.502, p=0.009).

Apical vertebral rotation may help predict the scoliosis progression and determine the timing of surgical intervention in patients with early degenerative lumbar scoliosis.

To evaluate the sensitivity, specificity and accuracy of various magnetic resonance imaging (MRI) features in differentiating vertebral compression fractures caused by malignancy, osteoporosis, and infections.

35 men and 45 women aged 40 to 78 (mean, 59) years underwent MRI to assess the underlying pathology of already diagnosed vertebral compression fractures (n=152). The interval from presentation to imaging ranged from 7 to 95 (mean, 62) days. MRI features of each vertebral compression fracture were assessed. The sensitivity, specificity, and accuracy for each of the MRI features were calculated. Association between each MRI feature and various underlying pathologies (malignancy, osteoporosis, and infections) of vertebral compression fractures was evaluated.

Regarding these 80 patients, the MRI diagnosis was correct in 78 and inconclusive in 2 with malignancy. MRI features suggestive of malignant fractures were a convex posterior border of the vertebral body, pedicle involvement, posterior neural element involvement, an epidural mass, a paraspinal mass, and other spinal metastases. MRI features suggestive of osteoporotic fractures were retropulsion, low signal intensity band, spared normal marrow signal intensity, and the fluid sign. MRI features suggestive of infective fractures were contiguous vertebral involvement, end plate disruption, disc involvement, and paraspinal and epidural abscesses.

Combination of several MRI features can provide clues to differentiate between malignant, osteoporotic, and infective vertebral compression fractures.

To evaluate the biocompatibility of collagen membranes for host-graft integration.

18 Achilles tendons in 9 rats were used. Five tendons were controls and repaired with prolene 3–0 sutures only. The remaining 13 tendons were cases, and a 10-mm section was excised. A 15×5 mm bovine pericardial collagen membrane (Tutomesh; Tutogen Medical, Germany) was laid circumferentially around the tendon and secured with prolene 3–0 sutures. Tendons were harvested after 4 and 6 weeks. Only the repair interfaces (i.e. cut ends and immediate surrounding tissue) were evaluated. Integration of the membranes to the tendons was evaluated using a semi-quantitative wound maturation scale (1–4) based on the presence of inflammatory cells, vascularisation, fibroblasts, and the amount and alignment of collagen fibrils. The presence of fibroblasts and vascularisation were positive parameters, whereas inflammatory cell ratios were regarded as negative parameters. Immunohistochemical study was also performed.

There was no host-graft reaction or wound complication (infection, abscess or seroma). Histological and immunohistochemical assessment confirmed re-approximation of the cut tendon ends and incorporation of the membrane onto the tendon as early as week 4. At week 4, the mean maturation scale scores were 2.7 for controls and 3.3 for cases (p=0.11). At week 6, the corresponding values were 3.8 and 3.0 (p=0.004). A reparative process involving formation of blood vessels and invasion by fibroblasts was noted in both control and case tendons. In controls, T- and B-lymphocytes were present. In cases, inflammatory cells were noted at the junction of the host and membrane without invasion of the graft material. There was no lymphocytic infiltration on the graft. The foreign body reaction was localised and minimal.

The collagen membrane achieved favourable host-graft integration as early as week 4, with a minimal foreign body reaction.

To evaluate host-implant integration with collagen membranes in 14 patients who underwent limb salvage surgery for musculoskeletal oncological disease.

8 females and 6 males aged 10 to 69 (mean, 30) years underwent limb savage surgery with collagen membranes (Tutomesh; Tutogen Medical, Germany) for osteosarcoma (n=7), chondrosarcoma (n=3), giant cell tumour (n=1), malignant fibrous histiocytoma (n=1), arteriovenous malformation (n=1), and pigmented villonodular synovitis (n=1). The procedures performed were proximal humeral resection (n=3), partial scapulectomy (n=1), proximal femoral resection (n=2), total femoral resection (n=2), proximal tibial resection (n=3), and wide resection of soft tissues of the knee (n=3). In addition, 10 patients underwent endoprosthesis reconstruction. Reconstruction of musculoskeletal defects was classified into type I (intercalary, n=2), type II (joint, n=4), and type III (both, n=8). Graft incorporation and local recurrence were monitored. Clinical outcome measures entailed the Short Form-36, Toronto Extremity Salvage Score (TESS), and Musculoskeletal Tumor Society Score (MSTS).

Two patients with proximal tibial resection and one with total femoral resection had wound healing problems. No patient had any infection or any foreign body reaction necessitating implant removal. Eight patients with type II or III reconstruction were followed up for a mean of 11 (range, 1–23) months. Their scores in the Short Form-36, TESS, and MSTS were similar to those who had undergone reconstructions without the membrane, with the exception of type II reconstructions for which the membrane conferred good results.

The Tutomesh membrane facilitated host-implant integration and provided a feasible anatomic reconstruction for ligaments in the shoulder, knee, and hip.

To assess the medium-term outcome of titanium uncemented modular tapered stems in revision total hip arthroplasty (THA).

A questionnaire was mailed to 421 patients who underwent revision THA using a titanium uncemented modular tapered stem. 66 of whom underwent re-revision THA. The questionnaire was composed to reveal the Charnley classification, the modified 12-item Oxford hip score, and the Devane patient activity level.

323 (77%) of the patients responded. The mean follow-up time was 6.6 years. 12 patients underwent bilateral revision THA. The mean Oxford hip score was 35.7, compared to 35.8 for all-component revision at postoperative month 6 in the New Zealand Joint Registry. The median Oxford hip scores for the subgroups of the stem-only revision (n=92), all-component revision (n=215), and re-revision (n=16) were 38, 39.9, and 30, respectively. The difference was significant between all-component revision and re-revision (adjusted p=0.003), and between stem-only revision and re-revision (adjusted p=0.037). Regarding patient distribution according to the Charnley class and the Devane patient activity level, the difference was significant between the mean Oxford hip scores of Charnley classes B and C (adjusted p=0.017), and between the Devane patient activity levels of Charnley classes A and C (adjusted p=0.043).

The medium-term outcome of revision THA using a titanium uncemented modular tapered stem was comparable to that reported for other stems.

To evaluate long-term outcomes of total hip arthroplasty (THA) using the Perfecta cementless system.

73 men and 76 women aged 65 to 88 (mean, 71) years underwent 168 THAs using the Perfecta cementless system and were followed up for a mean of 13 (10 to 15) years. 19 patients had bilateral THA. The diagnoses were idiopathic osteoarthritis (n=121), osteonecrosis of the femoral head (n=25), rheumatoid arthritis (n=14), and post-traumatic osteoarthritis (n=8). Patients were evaluated clinically and radiographically before and after THA.

The mean Harris hip score improved from 40 to 84 (p<0.001); the score was excellent or good in 130 hips, fair in 17, and poor in 21. The mean Merle D'Aubigne score improved from 4 to 10 (p<0.001); the score was excellent or good in 138 hips, fair in 9, and poor in 21. Poor results were due to aseptic revision of the cup (n=16) or stem (n=3) or deep infection (n=2). Thigh pain that limited activities of daily living was noted in 3 hips. 142 patients could walk without a limp, 5 had a slight limp, and 3 had a moderate limp and used a cane. The mean Short Form-12 score for the physical function was 49 and for the mental health was 56; the mean Western Ontario and McMaster Universities Osteoarthritis Index was 39; the mean visual analogue scale score for satisfaction was 8. Two patients developed deep infection at postoperative months 8 and 50 and were treated with 2-staged revision. Five patients had hip dislocation within 2 months; 3 of whom endured recurrent dislocation and underwent revision surgery; they had acetabular malposition with excessive anteversion despite a correct acetabular angle. 13 other cups were revised because of loosening or extensive osteolysis (n=10) and polyethylene wear (n=3) after a mean interval of 6 (range, 4–10) years. Three stems were revised because of aseptic loosening after a mean interval of 7 (range, 6–9) years. The survivorship at 14 years was 99% for stems and 91% for cups.

The long-term outcome of the Perfecta cementless stem was good; most major complications occurred in the cup.

To compare long-term outcomes of total hip replacement (THR) in patients with or without contamination of the femoral head.

After a mean period of 12 (range, 8–17) years, 104 female and 71 male THR patients aged 47 to 96 (mean, 77) years were reassessed via a self-administered questionnaire, and 25 other THR patients were reassessed by review of case notes. The questionnaires comprised the 12-item Oxford hip score and the European Quality Of Life (EuroQOL). 87 and 88 patients had positive and negative cultures in the donated femoral heads, respectively. The 2 groups were compared with respect to the Oxford hip score, the EuroQOL, and rates of complication and revision surgery.

Long-term outcomes of THR patients with or without femoral head contamination were not significantly different. Respectively, the mean Oxford hip scores were 36 and 39 (p=0.4); 16 and 14 patients had the maximum score of 48; 2 and 3 patients scored <10 (mostly owing to aseptic loosening). The respective mean visual analogue scale score of the EuroQOL were 65 and 73 (p=0.07); only the dimension of self care was significantly different between groups (p=0.04). Respectively, 14 and 12 patients had complications (16% vs. 15%, χ²=0.05, p=0.8), whereas 11 and 5 patients had revision surgery (13% vs. 6%, χ²=2.2, p>0.1).

Microbiological screening of donated femoral heads plays no role in predicting long-term failure of THR in the donors.

To compare mid-term outcomes of hemi-resurfacing arthroplasty (HRA) and total resurfacing arthroplasty (TRA).

Seven men and 5 women (16 hips) aged 33 to 72 (mean, 50) years underwent HRA, whereas 7 men and 3 women (16 hips) aged 23 to 52 (mean, 40) years underwent metal-on-metal TRA; all were for osteonecrosis of the femoral head. In the HRA and TRA groups respectively, 10 and 8 hips were steroid-induced, whereas 6 and 8 hips were related to alcohol abuse. Surgery for 12 and 4 hips entailed the posterolateral approach, whereas 4 and 12 hips entailed the Hardinge approach. Usually, HRA was performed for early stages of the disease, and TRA for more advanced stages. 12 and 6 hips were classified as JOA stages 1 to 3A, whereas 4 and 10 hips as stages 3B to 4. All the hips were JOA types C1 or C2.

In the HRA and TRA groups respectively, the mean follow-up period were 6.5 and 5.5 years. The mean total JOA hip scores were 57 and 54 preoperatively, 93 and 97 at one year, and 84 and 96 at the final follow-up (p<0.01). The higher score in the TRA patients was mainly attributed to improvement in the pain score. In the HRA group, 12 hips developed groin pain or groin discomfort while walking, though only 4 of them showed apparent joint-space narrowing. Five patients underwent revision surgeries (conversion to total hip arthroplasty) owing to a femoral neck fracture, acetabular protrusio, osteoarthritic change, and severe groin pain. Patients having TRA had no revision surgery and did not complain of groin pain. Implants in both groups were radiographically stable.

TRA were superior to HRA in terms of pain relief and implant survival, even though the former procedure was used for more advanced cases.

To compare the rate and unit of allogeneic blood transfusion in one-stage bilateral total hip arthroplasty (THA) in patients with and without autologous blood reinfusion.

Records of a consecutive series of 43 men and 33 women aged 25 to 83 (mean, 55) years who underwent one-stage sequential bilateral THA by a single surgeon were reviewed. Their risks of cardiopulmonary complications were minimal. At least 4 weeks prior to surgery, 38 of the patients donated 2 units of autologous blood in 2 stages (one to 2 weeks apart). The remaining 38 patients did not donate blood owing to personal preferences or logistical reasons. All pre-donated autologous blood was transfused back to the patients in the recovery room. Estimated blood loss volume, blood salvaged volume, and complications were recorded, as were pre- and post-operative haemoglobin levels. The 2 groups were compared with respect to the rate and unit of allogeneic blood transfusion.

The mean estimated blood loss was 939 (SD, 448; range, 200–2500) ml. The mean volume of blood salvaged was 302 (SD, 196; range, 0–850) ml, representing a collection rate of 32%. In patients with and without autologous blood reinfusion, 16 (42%) and 33 (87%) patients received allogeneic blood transfusion of 0.9 and 2.4 units, respectively (p<0.0001). Four patients developed major cardiopulmonary complications. None had deep venous thrombosis or pulmonary embolism. No complications resulted from blood transfusion. The 2 groups were not significantly different in complication rate (13% vs. 18%, p=0.54) and discharge haemoglobin levels.

Autologous blood reinfusion was effective in reducing the rate and unit of allogeneic blood transfusion in patients undergoing one-stage bilateral THA.

To evaluate treatment outcomes of minimally invasive plate osteosynthesis (MIPO) for distal femoral fractures in 31 patients.

22 male and 9 female consecutive patients aged 21 to 65 (mean, 42) years underwent minimally invasive plate osteosynthesis using a locking compression plate (LCP) for distal femoral fractures. The causes of injury were vehicular accidents (n=24), falls (n=6), and assault (n=1). According to the AO classification, the fractures were classified as types A1 (n=10), A2 (n=7), and A3 (n=14). Most fractures were closed; 3 were Gustilo type-II fractures. Clinical and functional outcomes were assessed using the Knee Society Scores.

The mean operating time was 70 (range, 60–100) minutes. The mean length of hospital stay was 9 (6–14) days. The mean time to union was 3.7 (range, 2.8–4.6) months. The mean follow-up period was 18 (14–26) months. At the one-year follow-up, 29 of the patients had good or excellent outcomes. No patient had angular or rotational deformity of >10°. One osteoporotic woman with a type-A1 fracture had implant failure at 10 weeks and underwent revision surgery. Another patient with a type-A1 fracture had persistent pain on the lateral aspect of the distal thigh and underwent implant removal at 22 months. No patient developed deep infection, malunion or nonunion.

MIPO using a LCP achieves favourable biological fixation for distal femoral fractures with few complications. Bone grafting is not needed even in cases of metaphyseal comminution. Proper patient selection and preoperative planning are essential to prevent complications.

To report 6 patients with periprosthetic humeral fractures treated with open reduction and internal fixation with plate and strut allograft augmentation.

6 women aged 69 to 79 (mean, 73) years underwent open reduction and internal fixation with plate and strut allograft augmentation for periprosthetic humeral fractures (type C) after a fall. They had undergone reverse shoulder arthroplasty for rotator cuff arthropathy. The mean interval between the initial arthroplasty and the fracture was 17 (range, 11–21) months.

The mean follow-up period was 14 (range, 12–16) months. The mean time to union was 5.4 (range, 4–6) months. All fractures united without complications. The mean Constant score at the last follow-up was 64 (range, 56–80). The range of shoulder movement and patient satisfaction were restored to pre-fracture status in all patients, except for one who had more pain in the lateral area of the arm (probably because of soft-tissue irritation by the plate and wires). Three patients had evidence of graft-to-host union and 3 others had graft resorption.

Internal fixation with plate, cable wires and strut allogaft augmentation achieves satisfactory results for periprosthetic humeral fractures.

To evaluate early functional outcomes of hemiarthroplasty for 3- or 4-part proximal humeral fractures.

16 men and 11 women aged 27 to 83 (mean, 56) years underwent hemiarthroplasty for comminuted 3-part (n=13) or 4-part (n=14) proximal humeral fractures. All the patients reported normal shoulder function prior to injury. The range of shoulder motion and muscle power were evaluated, as were subjective pain and satisfaction (using the UCLA scoring system).

At the final follow-up, the mean maximum abduction was 111° (SD, 47°; range, 30°–180°), and the mean maximum forward flexion was 143° (SD, 41°; range, 45°–180°). All patients had radiographic union of the tuberosities. The mean UCLA score was 28; 21 patients attained good-to-excellent scores (≥27), whereas 6 attained lower (fair-to-poor) scores. The mean UCLA score was higher in patients aged <60 (n=13) than those who were older (n=14) [30 vs. 26, p=0.008] and in patients operated after 7 days (n=6) than those operated before 7 days (n=21) [32 vs. 27, p=0.02], but did not differ significantly in terms of gender and fracture type and side. Patient age and gender, fracture type, and injury-to-surgery interval did not have a significant impact on maximum forward flexion and abduction.

Early functional outcomes of hemiarthroplasty for comminuted proximal humeral fractures is good in medically fit and cooperative patients.

To review records of children referred to ousr hospital for developmental dysplasia of the hip (DDH) and evaluate the effectiveness of hip screening in Hong Kong.

Records of children suspected of DDH and referred to our hospital between 1 January 2005 and 31 March 2010 were reviewed. The diagnosis was based on static and dynamic ultrasonography of the hips and/or radiographic features of hip subluxation, dislocation, and pelvic dysplasia. The age of the children at presentation was recorded. Late diagnoses were those presenting after the age of 6 months. Late presentations were those presenting after the walking age.

Of 213 referred children, 33 were diagnosed to have DDH. Their mean age at presentation was 90.8 (range, 9–559) days. The female-to-male ratio was 2.9:1, and the mean follow-up duration was 25.1 (range, 4.2–50.3) months. 30 of the DDH infants were diagnosed before the age of 6 months, one at 8 months, and 2 were late presenters. Hence, the failure rate of the hip screening was 0.09 per 1000 live births. The remaining 180 were confirmed to be normal and followed up until the walking age. The point prevalence of DDH on Hong Kong Island was 0.87/1000 live births.

In light of the low point prevalence of DDH in Hong Kong born children, the protocol of universal clinical assessment and selective ultrasonographic screening of suspected cases is justified.

To investigate the risks and types of complications associated with the Ilizarov technique of lengthening and then nailing in persons of normal height.

Records of 26 men and 6 women aged 21 to 47 (mean, 27) years with body height of 160 to 176 (mean, 170) cm who underwent tibial and fibular lengthening and then intramedullary nailing were reviewed. They were skeletally mature healthy persons, with no systemic/local bone disease, deformity or limb length discrepancy. Pain was assessed using the visual analogue scale (VAS). Patient satisfaction in terms of the treatment outcomes and expectations was also assessed. Complications encountered during or after treatment were recorded.

The mean lengthening achieved was 7.6 (range, 3.5–12) cm or 26% (range, 10–40%) of the original length. The mean duration of external fixation was 96 (range, 45–135) days. The mean follow-up duration after intramedullary nailing was 38.7 (range, 24–93) months. The mean VAS pain score was 9.3 at week 1, 6.6 at week 4, and 5.7 at week 8. After intramedullary nailing, the mean VAS pain score was 2.6 at week 4, 0.9 at month 6, and 0.3 at year 1. 91% of the patients were satisfied with the outcome at week 6; 81% after intramedullary nailing, and 94% at the one-year follow-up. Four patients had revision operations: one for pin exchange owing to pin bending after a fall, one for adjusting external rotation of the tibia after nailing, one for bone grafting for delayed union, and one for drainage of a haematoma just after nailing.

Most complications related to patient discomfort and psychological stress, which were important issues in this type of patients.

To review records of 108 children with radial neck fractures and develop an algorithm for treatment.

Records of 50 girls and 58 boys aged 2 to 14 (mean, 8.7) years with radial neck fractures were reviewed. The most common injury mechanism was tripping and falling on an outstretched hand while running (n=44), followed by falling from monkey bars (n=11). Fractures were classified into grade 1 (n=25), grade 2 (n=60), grade 3 (n=16), grade 4a (n=6), and grade 4b (n=1). 21 patients had associated fractures involving the olecranon, proximal ulna, and/or the humeral supracondyle. The time from injury to treatment ranged from 0 to 7 days. Treatments included casting without manipulation (n=86), closed reduction and casting (n=8), percutaneous Kirschner wire–assisted reduction and casting (n=7), and open reduction and casting (n=7).

Patients were followed up for a mean of 2.7 (range, 1–5) years. Outcome was excellent in 93 patients, good in 11, and fair in 4. Higher fracture grades correlated positively with poorer outcomes (p=0.001) and more invasive treatment (p=0.001). Nonetheless, the post-reduction angles of all the patients were not significantly different (p>0.05). Older children sustained more severe fractures (p=0.04) and had poorer outcomes, even after correction for fracture grade (p=0.007). Patients with associated fractures had significantly poorer outcomes (p<0.05). Two patients developed synostosis of the proximal radioulnar joint. One of whom had an associated olecranon fracture and underwent open reduction and casting. The other had an associated proximal ulnar fracture and underwent repeated percutaneous Kirschner wire–assisted reduction owing to loss of reduction. Five patients developed heterotopic ossification. Four of whom had associated fractures (3 involved the olecranon and one the proximal ulna). 14 patients developed cubitus valgus deformity of 3° to 10°.

Open reduction should only be performed after more conservative treatments fail to achieve reduction.

To report manifestations and management of tuberculosis of the foot and ankle in 21 children.

Records of 12 girls and 9 boys aged 3 to 14 (mean, 7) years with tuberculosis of the foot and ankle were reviewed. The mean delay in presentation was 4.7 (range, 0.5–14) months. All the patients had local swelling, tenderness, and antalgic gait; 16 were limping; 5 had an antecedent trauma; 11 had an abscess; and 6 had a discharging sinus. The diagnosis was based on a smear positive for acid-fast bacilli (n=2), histopathology (n=15), or clinicoradiological findings (n=4). Nine patients had osteoarticular tuberculosis in other parts of the body. None had evidence of pulmonary tuberculosis. Lesions were classified into synovial (articular) and osseous. All 3 synovial lesions occurred in the ankle, 2 of which were at an advanced stage. Osseous lesions occurred in the calcaneus (n=5), metatarsal (n=5), talus (n=3), cuboid (n=3), medial cuneiform (n=1), and phalanx (n=1), and were sub-classified into stages 1 (n=3), 2 (n=5), and 3 (n=10) according to disease progression. All the patients were treated conservatively with splintage and chemotherapy.

The mean follow-up period was 21 (range, 7–51) months. All the 3 patients with stage 1 osseous lesions showed healing within 6 weeks and had complete recovery of function. Four of the 5 patients with stage 2 lesions also showed complete recovery of function and reformation of bony trabeculations; radiological incorporation of sequestrum was obvious by 6 months in most patients. In patients with stage 3 lesions, healing was delayed and there was residual alteration in bony architecture and joint changes. Sclerotic changes and joint involvement also increased the likelihood of poor outcomes. However, none of our patients had any residual tenderness or foot deformity at the final follow-up.

Outcome after non-operative treatment is good, provided the lesions are treated early.

To review outcomes of 106 patients after complete excision of the giant cell tumour of the tendon sheath of the hand, with or without postoperative radiotherapy.

Records of 77 women and 29 men aged 11 to 61 (mean, 31) years who underwent excision of giant cell tumours of the tendon sheath of the hand were reviewed. All patients presented with gradually progressive swelling; only 3 of them also presented with pain. The lesions were located on the dorsal aspect (n=11), the palmar aspect (n=66), both aspects (n=10), or circumferentially (n=14). No patient had multiple lesions. Five patients had bony erosion and 3 had neurovascular bundle involvement.

The mean follow-up period was 12 (range, 4–22) years. 56 patients with well-encapsulated giant cell tumour in the palmer and/or dorsal aspects were classified as at low risk of recurrence. The remaining 50 patients were classified as at high risk of recurrence and underwent postoperative radiotherapy. None of the patients at low risk had any recurrence; 4 of those at high risk had recurrence despite radiotherapy. No complication was attributable to the irradiation, except that 3 patients had some dark pigmentation around the wound scar.

Postoperative radiotherapy may have a role in reducing recurrence of the giant cell tumour of the tendon sheath of the hand.

To evaluate the mid-term outcome of anterior cruciate ligament (ACL) reconstruction using the medial third of the patellar tendon as the bone-patellar tendon-bone (BPTB) graft.

79 men (34 right knees and 45 left knees) aged 20 to 52 (mean, 28) years underwent ACL reconstruction using the medial third of the patellar tendon with bone plugs and were followed up for a minimum of 3 years. Stability was assessed using the Lachman test and the pivot shift test. The range of motion, donor-site symptoms, Lysholm knee score, and International Knee Documentation Committee (IKDC) score were also recorded.

In terms of stability, only 4 patients had a grade-2 laxity postoperatively, compared to all having a grade-2 or −3 laxity preoperatively. Regarding the range of motion, extension loss was <3° in 76 patients and 3° to 5° in 3, whereas flexion loss was <5° in 73 patients, 5° to 15° in 5, and 16° to 25° in one. No patient had patellar tendon rupture/shortening, patellar fracture, or patellar maltracking. The mean Lysholm knee score improved from 48 to 92 (p<0.0001); scores were excellent (>90) in 46 patients and good (83–90) in 32 patients. The overall IKDC score was grade A (normal) in 9 patients, grade B (near normal) in 66, and grade C (abnormal) in 4. Three of the latter patients had osteoarthritic changes and pain on minimal physical activity. Only one patient had a sensation of partial giving way during moderate physical activity. 20% of the patients had anterior knee pain.

Outcome of ACL reconstruction using the medial third of the patellar tendon was comparable to that using the central third of the patellar tendon. In addition, there were no instances of patellar tendon rupture/shortening, patellar fracture, or patellar maltracking.

To compare mid-term outcomes of a high-flexion prosthesis with a conventional prosthesis.

Records of 107 consecutive patients who underwent total knee arthroplasty (TKA) for primary osteoarthritis by a single surgeon were reviewed. 21 men and 36 women (mean age, 65 years) used a high-flexion prosthesis (NexGen CR-Flex Mobile, Zimmer), whereas 38 men and 12 women (mean age, 67 years) used a conventional prosthesis (Genesis II, Smith & Nephew) that preserves the posterior cruciate ligament, with mobile-bearing polyethylene inlay. The Knee Society knee and functional scores and the range of motion (especially maximum passive flexion) were assessed. Radiographs were evaluated to identify radiolucent lines at the bone cement interface, patella tracking, tibiofemoral alignment, and implant positioning.

The mean follow-up period was 68 (range, 51–70) months. In terms of the mean Knee Society scores and range of motion, the difference between groups was not significant (p>0.05), but the improvement in both groups after TKA was significant (p<0.005). Respectively in the high-flexion and conventional TKA groups, 2 and one of the patients developed deep vein thrombosis and were treated with anticoagulants for 3 months. One and 5 of the patients had an unsatisfactory range of motion (<60°) in week 1, which was resolved with mobilisation under general anaesthesia in combination with a peridural catheter for analgesia. No patient developed implant-specific complications such as aseptic loosening (osteolysis, progressive radiolucent lines, implant deviation) or dislocation of the polyethylene insert.

The high-flexion prosthesis revealed no significant advantages over the conventional prosthesis in terms of the Knee Society scores and range of motion. Long-term studies are needed to determine whether the high-flexion implant is superior to the conventional implant in terms of polyethylene wear and aseptic loosening.

To assess the effect of each step of medial soft-tissue releases on the joint gap angle during posterior-stabilised total knee arthroplasty (TKA).

82 women and 9 men (mean age, 72 years) with medial osteoarthritic knees underwent 100 posterior-stabilised TKAs, in which release of superficial fibres of the medial collateral ligament (MCL) were required using the gap control technique. The order of releases was the superficial MCL, the pes anserinus, and then the semi-membranosus. The superficial MCL was released selectively. The effect of each step of medial soft-tissue releases in full extension and in 90° flexion was compared.

After all medial soft-tissue releases, the mean joint gap angles decreased from 8.7° to 3.8° varus in flexion and from 4.4° to 1.4° varus in extension. The total effect of medial soft-tissue releases was significantly larger in flexion than in extension (4.9°±3.2° vs. 3.0°±2.0°, p<0.0001), except for the release of posterior fibres of the superficial MCL. The effect of release of the semi-membranosus in flexion was largest.

The release effect was significantly greater in flexion than in extension during posterior-stabilised TKA; the joint gap technique may be more reliable in medial osteoarthritic knees with moderate and severe varus instability.

To evaluate the sterility of the sterile surgical helmet system (SSHS) during elective total hip and knee arthroplasty in theatres with (n=20) and without (n=20) laminar flow.

Three surgeons performed 14 total knee arthroplasties (TKAs) and 6 total hip arthroplasties (THAs) in a laminar flow theatre and 15 TKAs and 5 THAs in a non-laminar flow theatre. An SSHS was used in all the procedures. Samples were taken from the hood at 30-minute intervals during surgery. Swabs were then broken into cooked meat broths for cultivation of organisms. The broths were then directly inoculated onto blood agar and fastidious anaerobic agar for culture of aerobic and anaerobic bacteria, respectively. After 24 hours, these plates were reinoculated with broths that had been incubated for 24 hours. Microbial growth was quantified as 0 (none), 1 (mild), 2 (moderate) and 3 (heavy). Bacterial contamination in the 2 groups at 30, 60 and 90 minutes was compared.

Respectively in the laminar and non-laminar flow theatres, 0 and 9 of the SSHSs showed bacterial growth after direct inoculation, and 14 and 18 of the SSHSs grew 18 and 24 types of organisms in the swab cultures after 24 hours of incubation. Respectively at 30, 60, and 90 minutes, the degree of contamination from direct incubation was significant, but the degree of contamination on swabs after 24 hours of inoculation was not significant. The mean time-dependent contamination after direct inoculation was 0 for the laminar flow group and 0.5, 0.75, and 1.0 for the non-laminar flow group, whereas the corresponding values after 24 hours of incubation were 1.8, 1.8, and 2.6, and 2, 2.75, and 2.95. Coagulase negative Staphylococcus aureus was the most common organism in both groups.

80% of SSHSs used were contaminated intra-operatively. Direct contact with the SSHS should be avoided by the operating team during surgery. Routine changing of gloves in case of contact with the SSHS should be practised.