Prevention of positional brachial plexopathy during surgical correction of scoliosis.

Continuous intraoperative monitoring of spinal cord function using somatosensory evoked potentials (SSEP) has gained nearly universal acceptance as a reliable and sensitive method for detecting and possibly preventing neurologic injury during surgical correction of spinal deformities. In several reports, spinal cord injury was identified successfully based on changes in SSEP response characteristics, specifically amplitude and latency. Less well documented and used, however, is monitoring of peripheral nerve function with SSEPs to identify and prevent the neurologic sequelae of prolonged prone positioning on a spinal frame. The authors describe a patient who underwent surgical removal of spinal instrumentation but was not monitored. A brachial plexopathy developed in this patient from pressure on the axilla exerted by a Relton-Hall positioning frame during spinal surgery. In addition, data are presented from 15 of 500 consecutive pediatric patients who underwent surgical correction of scoliosis between 1993 and 1997 with whom intermittent monitoring of ulnar nerve SSEPs was used successfully to identify impending brachial plexopathy, a complication of prone positioning. A statistically significant reduction in ulnar nerve SSEP amplitude was observed in 18 limbs of the 500 patients (3.6%) reviewed. Repositioning the arm(s) or shoulders resulted in nearly immediate improvement of SSEP amplitude, and all awoke without signs of brachial plexopathy. This complication can be avoided by monitoring SSEPs to ulnar nerve stimulation for patients placed in the prone position during spinal surgery.

Subchondral magnetic resonance imaging changes in early osteoarthrosis associated with tibial osteonecrosis.

SUMMARY: The authors report on 41 patients with acute or subacute knee pain and early or midstage degenerative arthrosis with osteonecrotic lesions in the subchondral and metaphyseal region of the medial proximal tibia. Each lesion was identified only on magnetic resonance images (MRI). These MRI changes are classified and the clinical course is defined during a follow-up period averaging 4.5 years. Radiographically, 22 patients had minimal degenerative changes; 12 had moderate arthritis; and 7 were normal. With MRI, 3 distinct types of lesions were identified. Type A lesions had localized areas of decreased signal in the subchondral area. Type B lesions had diffuse signal changes with extension into the metaphysis. Type C lesions had metaphyseal involvement as well as a marginated serpentine subchondral rim usually associated with advanced osteonecrosis. There were 9 type A lesions, 23 type B, and 9 type C. At the end of 1 year, 33 patients (80%) had no or mild symptoms, and 8 (20%) had persistent moderate pain. At 4.5-year follow-up, most patients had symptoms consistent with progressive osteoarthrosis, 12 patients had severe symptoms (29%), 17 (41%) had mild or moderate symptoms, and only 12 (29%) were asymptomatic or had minimal symptoms. The type of MRI change seen initially was predictive of prognosis. Only 6 (19%) of the 32 patients with type A or B findings had severe symptoms at last follow-up. Six (66%) of the 9 patients with a type C MRI lesion had severe symptoms or had an operation by last follow-up. Twelve patients had follow-up MRI at a mean 15 months (range, 12 to 18 months) after the initial evaluation. The type A and B changes were either absent or significantly reduced. The type C subchondral marginated rim changes remained but metaphyseal involvement was reduced. There appears to be a spectrum of tibial subchondral MRI changes associated with sudden onset of medial knee pain in patients with early osteoarthritis of the knee. These changes may be indicative of osteonecrosis. The initial MRI classification is useful in predicting prognosis. Recognition of this problem may avoid unnecessary intra-articular surgery.

Influence of rigid spinal instrumentation on the neurogenic motor evoked potential.

Three case reports are presented to illustrate how placement of spinal instrumentation obliterates previously normal neurogenic motor evoked potentials (NMEPs) elicited by transosseous electrical stimulation. This results in an unacceptably high false-positive rate for NMEP recordings during scoliosis surgery. The loss of the NMEPs was attributed to shunting of electrical current from JO5 spinous process stimulating needles through the metal rod to ground, thus preventing adequate stimulation to the spinal cord. A modification of the transosseous technique with epidural stimulation to improve test reliability is described.

Posterior spinal fusion for scoliosis in patients with cerebral palsy: a comparison of Luque rod and Unit Rod instrumentation.

The development of the U-shaped Unit Rod for posterior spinal arthrodesis is a recent advance in the treatment of spinal deformity in patients with cerebral palsy. The results of 15 patients who underwent arthrodesis with dual Luque rod instrumentation (group I) are compared with the results of 15 patients in whom Unit Rod instrumentation was used (group II). The two treatment groups were similar with respect to age, gender, major spinal curve, and degree of pelvic obliquity. The Unit Rod instrumentation allowed significantly greater correction of both the major curve and pelvic obliquity, as assessed on postoperative radiographs. The mean postoperative major curve in group I was 44.1 degrees, compared with 31.7 degrees in group II (mean corrections of 48.6 and 61.7%, respectively). The mean angle of postoperative pelvic obliquity in group I was 12.6 degrees, compared with 5.2 degrees in group II, corrections of 49.5 and 79.3%, respectively. Sagittal-plane alignment was improved to a similar degree in both groups in those patients with preoperative imbalance and maintained in the remainder of patients.

Efficacy of spinal cord monitoring in scoliosis surgery in patients with cerebral palsy.

Although spinal cord monitoring is recommended during scoliosis surgery, a review from Rancho Los Amigos Medical Center stated that they were only able to obtain reproducible tracings in 53% of cerebral palsy patients. To ascertain that monitoring is both feasible and reliable in these patients, we reviewed the records of 34 consecutive patients with cerebral palsy who had scoliosis surgery at our institution. Spinal cord function was monitored by recording peripheral nerve, cervical/brainstem, and cortical somatosensory evoked potentials to posterior tibial nerve stimulation. Reproducible tracings were achieved in 31 of the 34 patients. Significant intraoperative changes were recorded in 12 of the 31 monitored patients, usually related to and requiring some modifications of the instrumentation. We conclude that with careful technique, spinal cord monitoring using cervical/brainstem somatosensory evoked potentials can be reliably achieved in most patients with cerebral palsy undergoing scoliosis surgery.

Aspirin and warfarin for thromboembolic disease after total joint arthroplasty.

This study compares the benefits of aspirin and warfarin prophylactic agents for patients with thromboembolic disease after total joint arthroplasty. It is a prospective randomized study of 388 patients having total hip or total knee surgery. All consecutive patients having total hip or total knee surgery were entered into this study and evaluated with preoperative and postoperative ventilation perfusion scans and a postoperative venogram. The aspirin and warfarin treatment groups were compared by size and location of venographically revealed clots and changes in ventilation perfusion scans. The results showed that there was no difference in the size or location of deep venous thrombosis in the aspirin or warfarin treatment groups. The venogram was negative in 44.5% of patients; 28.8% had small ++calf clots, 16% had large calf clots, 3.9% had popliteal clots, and 6.7% had femoral clots. Patients with total knee replacement had a 2.6 times greater incidence of calf deep venous thrombosis than patients with total hip replacement. There was no difference between the aspirin and warfarin groups in the incidence of changes in ventilation perfusion scans (18.9%). There was no difference between the 2 groups in bleeding complications. The results suggest that aspirin and warfarin are equivalent in prophylaxis against thromboembolic disease, as determined by prevention of venographic changes or changes in ventilation perfusion scans.

Occipitocervical arthrodesis in children. A new technique and analysis of results.

A new wiring technique for occipitocervical arthrodesis was used in sixteen consecutive children between 1985 and 1992. The twelve boys and four girls had an average age of nine years and six months (range, two years and five months to nineteen years and three months) at the operation. The arthrodesis was performed between the occiput and the second cervical vertebra in ten patients and between the occiput and the third cervical vertebra in six. The instability was related to congenital anomalies (six patients), decompression for cervical stenosis (four patients), Down syndrome (three patients), trauma (one patient), resection of a tumor (one patient), and neurofibromatosis (one patient). Six patients needed a laminectomy for decompression because of cervical stenosis or for removal of a tumor. All of the patients were managed with an autogenous bone graft from the iliac crest and postoperative immobilization with a halo device. Fusion was achieved in fifteen of the sixteen patients. Complications developed in seven patients. The use of wire fixation, combined with the inherent stability of the bone-graft construct, allowed for removal of the halo device relatively early (range, six to twelve weeks), before the fusion was fully mature. No graft was displaced. All of the patients were followed at least until there was radiographic evidence of fusion (fifteen patients) or until a reoperation was performed (one patient). The average duration of follow-up was thirty-seven months (range, twelve to 108 months).

Osteonecrosis of the medial part of the tibial plateau.

Fifteen elderly patients (sixteen knees) were seen because of acute pain in the knee and tenderness to palpation over the medial aspect of the tibial plateau. Initially, plain roentgenograms showed a radiolucent area at the site of the tenderness in only nine of the sixteen knees. However, radionuclide bone scans showed focal increased uptake at the site of the tenderness in four of the seven remaining knees, and magnetic resonance images showed discrete areas of low signal intensity at the same site in the other three knees. Plain roentgenograms eventually showed the typical lesion in all knees. Progression of the symptoms led to a total knee arthroplasty in nine knees and to a unicompartmental replacement in three; a satisfactory result was obtained in all twelve knees. An operation was recommended for two other knees, but it was refused by the patients. The symptoms resolved spontaneously in the remaining two knees. A degenerative tear in the medial meniscus, which is a common finding in this age-group, was noted at the time of a later operation in the three knees that had not had a radiolucent area on the initial plain roentgenograms but that had had an area of low signal intensity on the magnetic resonance images. If osteonecrosis of the tibial plateau is not considered as a potential cause of pain in the knee, symptoms may be attributed to a tear in the meniscus and an unnecessary and unproductive arthroscopy may be performed.

Significance of deep venous thrombosis in the lower extremity after total joint arthroplasty.

Thromboembolic disease continues to pose a significant risk to patients undergoing total joint arthroplasty, but the clinical significance of a deep venous thrombosis (DVT) in the postoperative period remains controversial. This report reviews the outcome of 920 patients undergoing prospective obtained venograms after total hip and total knee surgery. Six hundred two patients also obtained postoperative ventilation perfusion scans (V/Q). All patients received one of the following forms of prophylaxis: aspirin (79%), warfarin (16%), or low-molecular-weight heparin (5%). The results show that 39% were without any clot, 36% had small calf thrombi, 19% had large calf thrombi, 2.5% had popliteal thrombi, and 3.8% had femoral thrombi. Calf thrombi occurred almost 2.5 times more frequently after total knee arthroplasty than total hip arthroplasty (67% versus 27%), yet the incidence of pulmonary embolism (PE) is the same in both groups. Intermediate or high probability V/Q results occurred in 103 patients (15%). There was a positive association between the size of a clot and the likelihood of an intermediate- or high-probability V/Q scan. Thromboembolic events resulted in five symptomatic PE (0.5%) and two deaths (0.2%), one in the hospital on warfarin, and one as an outpatient on aspirin. From these data, the authors conclude that calf thrombi by themselves do not place a patient at risk and are not accurate markers of a patient being at risk. Larger clots have a greater likelihood to embolize, but the size of the clot that is clinically significant is undetermined. A significant chance of a PE after discharge from the hospital warrants some form of continuing prophylaxis in outpatients.

Complications of posterior arthrodesis of the cervical spine in patients who have Down syndrome.

Ten patients who had Down syndrome and had had a posterior arthrodesis of the upper cervical spine were studied. The mean age at the time of the operation was 8.9 years, and the patients had been followed for three days to forty-nine months. Complications related to the operation occurred in all patients. They included infection and dehiscence at the site of the wound, incomplete reduction of the atlanto-axial joint, instability of the adjacent motion segment, neurological sequelae, resorption of the autogenous bone graft, and death in the postoperative period. Resorption of the bone graft, which occurred in six of the patients, has not previously been reported in patients who have Down syndrome, to our knowledge. Several theoretical mechanisms for this complication are proposed. We recommend non-operative management for patients who have Down syndrome and atlanto-axial instability without neurological signs or symptoms. If the severity of symptoms necessitates a posterior arthrodesis, a high rate of complications must be anticipated.

The use of methylmethacrylate in primary total knee replacements with large tibial defects.

Bone defects in the tibial plateau present technical problems in primary knee replacement surgery. A variety of techniques is available to solve these problems. Of 59 patients treated with a cement fill and followed an average of 7.1 years (range, five to 11 years), there was only one failure requiring revision. Overall, knee scores averaged 78 points and roentgenogram scores averaged 85 points. Radiolucent lines were noted in 43 of the 59 patients, but none were progressive except in the one failure. There was no clinical correlation between the presence of radiolucent lines and clinical symptoms. Good long-term results might be expected from the cement fill when the bone defects are smaller than 20 mm and affect less than 50% of either plateau.

Blood loss after total knee replacement. Effects of tourniquet release and continuous passive motion.

We prospectively studied the cases of 121 patients who were being operated on for insertion of a unilateral total knee prosthesis with cement, and we placed them randomly in four groups. In Group I, the tourniquet was inflated throughout the operative procedure, and we released it postoperatively after a compressive dressing had been applied; a splint was used postoperatively for three days. In Group II, the tourniquet remained inflated throughout the operation, but no splint was applied postoperatively, and continuous passive motion was started immediately in the recovery room. In Group III, the tourniquet was released intraoperatively, and hemostasis was achieved by cauterization; postoperatively, a compressive dressing was applied, and a splint was used for three days. In Group IV, the tourniquet was released intraoperatively, hemostasis was established, and then the tourniquet was reinflated; a compressive dressing was applied, and continuous passive motion was started immediately in the recovery room. Hemoglobin and hematocrit values were monitored in all patients. Blood loss in suction drainage was recorded, and the total blood loss was calculated. The results show that total knee arthroplasty is associated with major loss of blood (mean, 1518 milliliters). The calculated blood loss for Groups I, II, and III averaged 1443 milliliters, while that for Group IV averaged 1793 milliliters. Loss in suction drainage correlated with total estimated blood loss and averaged 511 milliliters. The magnitude of blood loss after total knee arthroplasty should be appreciated, and special attention should be paid to the availability of adequate fluid and blood products, preferably blood donated by the patient preoperatively.

Postoperative care of the total knee patient.

After total knee surgery the three most important problems are thromboembolic disease, disorders of wound healing, and inadequate range of motion. With the use of prophylactic anticoagulation, CPM, autologous blood replacement, good nutrition, and careful wound closure, most patients will have an uneventful postoperative course and reach the maximum potential from their knee arthroplasty.

Computer tomography evaluation of Cotrel-Dubousset instrumentation in idiopathic scoliosis.

In 30 adolescent patients with 36 structural curves undergoing posterior spinal arthrodesis using Cotrel-Dubousset (CD) instrumentation, computer tomography (CT) was performed to evaluate the change in rotation and in the sagittal plane. The average preoperative Cobb angle was 51.7 degrees, and improved to 18.7 degrees postoperatively, or 64.9%. The angle of rotation of the apical vertebra relative to the anterior midline of the body (RAML) improved from 27.9 degrees preoperatively to 21.5 degrees postoperatively (23.9%). The angle of rotation relative to the sagittal plane (RAsag) improved from 15.2 degrees preoperatively to 13.4 degrees postoperatively (14.3%). The kyphotic angle increased an average of 10.6 degrees in the 14 patients whose preoperative angle was less than 25 degrees. As measured by CT, CD instrumentation improves both the rotational and sagittal plane deformities better than that reported for other systems.

Long-term results after total condylar knee arthroplasty. Significance of radiolucent lines.

In a prospective study of 123 total condylar knee arthroplasties, radiolucent lines were noted at the bone-cement interface about the tibial component in 65% of the knees. In 36%, the line was thin (less than 2 mm) and under either the medial or lateral tibial plateau. In 22%, thin lines were seen under both the medial and lateral plateaus. There was no statistically significant correlation between the occurrence of thin radiolucent lines in any location and the eventual postoperative clinical result. In 7%, thick lines (greater than 2 mm) were noted about the central peg of the prosthesis as well as under both plateaus and were associated with poor results. A 1-mm radiolucent interface was attached to various tibial components, which were then cemented into cadaver knees. For the polyethylene total condylar component, flexion or rotation of the prosthesis more than 5 degrees, angulation of the roentgen beam more than 6 degrees, or offsetting the beam more than 2.5 cm prevented visualization of the radiolucent line. With metal-backed components, more than 4 degrees of flexion obscured the lines. Therefore, extremely careful positioning of postoperative radiographs is essential for evaluating radiolucent lines after total knee arthroplasty.