Traumatic sacral pseudomeningocele
Takahiro Naruse, Yukihiro Matsuyama and Naoki Ishiguro
Cyclooxygenase-2 (COX-2), also known as prostaglandin endoperoxide synthase, has been reported to play an important role in the tumorigenicity of many types of tumors. The expression of COX-2 in spinal ependymomas, however, has not been studied. The authors evaluated COX-2 expression in ependymoma of the spinal cord.
Sixteen ependymoma samples obtained in patients undergoing surgery between 1995 and 2004 were utilized for immunohistochemical studies to evaluate COX-2 and vascular endothelial growth factor (VEGF) expression. Intratumoral microvessels were also stained immunohistochemically using anti–human von Willebrand factor antibody and were quantified to determine the microvessel density (MVD). The clinical features were reviewed and recorded and the association with COX-2 expression was assessed.
Seven (43.8%) of the 16 ependymoma specimens expressed COX-2. All three of the myxopapillary-type ependymomas exhibited COX-2–positive staining. Excluding the three myxopapillary-type cases, COX-2 expression was identified in four (30.8%) of 13 cellular-type ependymomas. The COX-2–positive samples exhibited a significant increase in VEGF-positive staining cells and MVD compared with COX-2-negative samples. The clinical features were not associated with COX-2 expression.
The results of the present study indicate that COX-2 expression may promote angiogenesis through VEGF expression in ependymomas of the spinal cord. It is suggested that the use of selective COX-2 inhibitors may provide a new therapeutic strategy for spinal cord ependymomas due to their inhibition of the COX-2-mediated angiogenesis.
Yukihiro Matsuyama, Kazuhiro Chiba, Hisashi Iwata, Takayuki Seo and Yoshiaki Toyama
Chemonucleolysis with condoliase has the potential to be a new, less invasive therapeutic option for patients with lumbar disc herniation (LDH). The aim of the present study was to determine the most suitable therapeutic dose of condoliase.
Patients between 20 and 70 years of age with unilateral leg pain, positive findings on the straight leg raise test, and LDH were recruited. All eligible patients were randomly assigned to receive condoliase (1.25, 2.5, or 5 U) or placebo. The primary end point was a change in the worst leg pain from preadministration (baseline) to week 13. The secondary end points were changes from baseline in the following items: worst back pain, Oswestry Disability Index (ODI), SF-36, and neurological examination. For pharmacokinetic and pharmacodynamic analyses, plasma condoliase concentrations and serum keratan sulfate concentrations were measured. The safety end points were adverse events (AEs) and radiographic and MRI parameters. Data on leg pain, back pain, abnormal neurological findings, and imaging parameters were collected until week 52.
A total of 194 patients received an injection of condoliase or placebo. The mean change in worst leg pain from baseline to week 13 was −31.7 mm (placebo), −46.7 mm (1.25 U), −41.1 mm (2.5 U), and −47.6 mm (5 U). The differences were significant at week 13 in the 1.25-U group (−14.9 mm; 95% CI −28.4 to −1.4 mm; p = 0.03) and 5-U group (−15.9 mm; 95% CI −29.0 to −2.7 mm; p = 0.01) compared with the placebo group. The dose-response improvement in the worst leg pain at week 13 was not significant (p = 0.14). The decrease in the worst leg pain in all 3 condoliase groups was observed from week 1 through week 52. Regarding the other end points, the worst back pain and results of the straight leg raise test, ODI, and SF-36 showed a tendency for sustained improvement in each of the condoliase groups until week 52. In all patients at all time points, plasma condoliase concentrations were below the detectable limit (< 100 μU/ml). Serum keratan sulfate concentrations significantly increased from baseline to 6 hours and 6 weeks after administration in all 3 condoliase groups. No patient died or developed anaphylaxis or neurological sequelae. Five serious AEs occurred in 5 patients (3 patients in the condoliase groups and 2 patients in the placebo group), resolved, and were considered unrelated to the investigational drug. Severe AEs occurred in 10 patients in the condoliase groups and resolved or improved. In the condoliase groups, back pain was the most frequent AE. Modic type 1 change and decrease in disc height were frequent imaging findings. Dose-response relationships were observed for the incidence of adverse drug reactions and decrease in disc height.
Condoliase significantly improved clinical symptoms in patients with LDH and was well tolerated. While all 3 doses had similar efficacy, the incidence of adverse drug reactions and decrease in disc height were dose dependent, thereby suggesting that 1.25 U would be the recommended clinical dose of condoliase.
Clinical trial registration no.: NCT00634946 (clinicaltrials.gov)
Report of two cases and review of the literature
Zenya Ito, Yoshimitsu Osawa, Yukihiro Matsuyama, Takaaki Aoki, Atsushi Harada and Naoki Ishiguro
✓Hypertrophic spinal pachymeningitis (HSP) is a comparatively rare disease characterized by hypertrophic inflammation of the dura mater and clinical symptoms that progress from local pain to myelopathy. The authors report two cases of recurrent HSP and review the English- and Japanese-language literature focusing on the recurrence of HSP.
In the first case, a man who presented at 67 years of age with lower-extremity numbness, gait disturbance, and bladder dysfunction experienced two recurrences of HSP during the 11 years of follow up after his initial laminectomy. Both recurrences were successfully treated with laminoplasty and duraplasty. Three years after his last surgical procedure, he was still able to walk with the aid of a walker. In the second case, a man who presented at 62 years of age with lower-extremity numbness and gait disturbance was initially treated successfully with steroid pulse therapy. Approximately 8 months after his initial presentation, his symptoms recurred. He underwent laminoplasty and duraplasty. At the 2.5-year follow-up examination, he had only mild neurological deficits and was still able to walk unaided.
To explore possible causes of recurrence, the authors searched the English- and Japanese-language literature for cases of HSP. Of the 96 cases identified, 11 were recurrent. Data on the presence or absence of inflammatory signs were available for 84 patients. A chi-square analysis revealed a significantly increased rate of recurrence for patients who had at least one positive inflammatory sign before surgery (six [20%] recurrent cases of 30) compared with those who had no positive inflammatory signs before surgery (two [3.7%] recurrent cases of 54) (p < 0.05). The authors conclude that HSP recurrence occurs because of active inflammation of the dura before surgery and the influence of chronic inflammation, including residual arachnoiditis.
Masahiko Watanabe, Daisuke Sakai, Daisuke Matsuyama, Yukihiro Yamamoto, Masato Sato and Joji Mochida
The purpose of this study was to identify risk factors for surgical site infection after spine surgery, noting the amount of saline used for intraoperative irrigation to minimize wound contamination.
The authors studied 223 consecutive spine operations from January 2006 through December 2006 at our institute. For a case to meet inclusion criteria as a site infection, it needed to require surgical incision and drainage and show positive intraoperative cultures. Preoperative and intraoperative data regarding each patient were collected. Patient characteristics recorded included age, sex, and body mass index (BMI). Preoperative risk factors included preoperative hospital stay, history of smoking, presence of diabetes, and an operation for a traumatized spine. Intraoperative factors that might have been risk factors for infection were collected and analyzed; these included type of procedure, estimated blood loss, duration of operation, and mean amount of saline used for irrigation per hour. Data were subjected to univariate and multivariate logistic regression analyses.
The incidence of surgical site infection in this population was 6.3%. According to the univariate analysis, there was a significant difference in the mean duration of operation and intraoperative blood loss, but not in patient age, BMI, or preoperative hospital stay. The mean amount of saline used for irrigation in the infected group was less than in the noninfected group, but was not significantly different. In the multivariate analysis, sex, advanced age (> 60 years), smoking history, and obesity (BMI > 25 kg/m2) did not show significant differences. In the analysis of patient characteristics, only diabetes (patients receiving any medications or insulin therapy at the time of surgery) was independently associated with an increased risk of surgical site infection (OR 4.88). In the comparison of trauma and elective surgery, trauma showed a significant association with surgical site infection (OR 9.42). In the analysis of surgical factors, a sufficient amount of saline for irrigation (mean > 2000 ml/hour) showed a strong association with the prevention of surgical site infection (OR 0.08), but prolonged operation time (> 3 hours), high blood loss (> 300 g), and instrumentation were not associated with surgical site infection.
Diabetes, trauma, and insufficient intraoperative irrigation of the surgical wound were independent and direct risk factors for surgical site infection following spine surgery. To prevent surgical site infection in spine surgery, it is important to control the perioperative serum glucose levels in patients with diabetes, avoid any delay of surgery in patients with trauma, and decrease intraoperative contamination by irrigating > 2000 ml/hour of saline in all patients.
Tomohiro Banno, Tsuyoshi Ohishi, Daisuke Suzuki, Yosuke Honda, Sho Kobayashi and Yukihiro Matsuyama
Pseudomeningocele arises after spinal fracture and nerve root avulsion or after complications of spine surgery. However, traumatic pseudomeningocele with spina bifida occulta is rare. In this report, a traumatic pseudomeningocele in a patient with spina bifida occulta that required surgical treatment is documented. This 37-year-old man presented to the authors' hospital with headache and a fluctuant mass in the center of his buttocks. A CT scan with myelography and MR imaging of the sacral region revealed a large subcutaneous area of fluid retention communicating with the intradural space through a defect of the S-2 lamina. Because 3 months of conservative treatment was unsuccessful, a free fat graft was placed with fibrin glue to seal the closure of the defect, followed by 1 week of CSF drainage. This is the first report on traumatic pseudomeningocele with spina bifida occulta successfully treated in this manner.
Hiroaki Nakashima, Shiro Imagama, Yasutsugu Yukawa, Tokumi Kanemura, Mitsuhiro Kamiya, Makoto Yanase, Keigo Ito, Masaaki Machino, Go Yoshida, Yoshimoto Ishikawa, Yukihiro Matsuyama, Nobuyuki Hamajima, Naoki Ishiguro and Fumihiko Kato
Postoperative C-5 palsy is a significant complication resulting from cervical decompression procedures. Moreover, when cervical degenerative diseases are treated with a combination of decompression and posterior instrumented fusion, patients are at increased risk for C-5 palsy. However, the clinical and radiological features of this condition remain unclear. Therefore, the purpose of this study was to clarify the risk factors for developing postoperative C-5 palsy.
Eighty-four patients (mean age 60.1 years) who had undergone posterior instrumented fusion using cervical pedicle screws to treat nontraumatic lesions were independently reviewed. The authors analyzed the medical records of some of these patients who developed postoperative C-5 palsy, paying particular attention to their plain radiographs, MRI studies, and CT scans. Risk factors for postoperative C-5 palsy were assessed using multivariate logistic regression analysis. The cutoff values for the pre- and postoperative width of the intervertebral foramen (C4–5) were determined by receiver operating characteristic curve analysis.
Ten (11.9%) of 84 patients developed postoperative C-5 palsy. Seven patients recovered fully from the neurological complications. The pre- and postoperative C4–5 angles showed significant kyphosis in the C-5 palsy group. The pre- and postoperative diameters of the C4–5 foramen on the palsy side were significantly smaller than those on the opposite side in the C-5 palsy group and those bilaterally in the non–C5 palsy group. Risk factors identified by multivariate logistic regression analysis were as follows: 1) ossification of the posterior longitudinal ligament (relative risk [RR] 7.22 [95% CI 1.03–50.55]); 2) posterior shift of the spinal cord (C4–5) (RR 1.73 [95% CI 1.00–2.98]); and 3) postoperative width of the C-5 intervertebral foramen (RR 0.33 [95% CI 0.14–0.79]). The cutoff values of the pre- and postoperative widths of the C-5 intervertebral foramen for C-5 palsy were 2.2 and 2.3 mm, respectively.
Patients with preoperative foraminal stenosis, posterior shift of the spinal cord, and additional iatrogenic foraminal stenosis due to cervical alignment correction were more likely to develop postoperative C-5 palsy after posterior instrumentation with fusion. Prophylactic foraminotomy at C4–5 might be useful when preoperative foraminal stenosis is present on CT. Furthermore, it might be useful for treating postoperative C-5 palsy. To prevent excessive posterior shift of the spinal cord, the authors recommend that appropriate kyphosis reduction should be considered carefully.
Hiroki Ushirozako, Go Yoshida, Sho Kobayashi, Tomohiko Hasegawa, Yu Yamato, Tatsuya Yasuda, Tomohiro Banno, Hideyuki Arima, Shin Oe, Yuki Mihara, Daisuke Togawa and Yukihiro Matsuyama
Intraoperative neuromonitoring may be valuable for predicting postoperative neurological complications, and transcranial motor evoked potentials (TcMEPs) are the most reliable monitoring modality with high sensitivity. One of the most frequent problems of TcMEP monitoring is the high rate of false-positive alerts, also called “anesthetic fade.” The purpose of this study was to clarify the risk factors for false-positive TcMEP alerts and to find ways to reduce false-positive rates.
The authors analyzed 703 patients who underwent TcMEP monitoring under total intravenous anesthesia during spinal surgery within a 7-year interval. They defined an alert point as final TcMEP amplitudes ≤ 30% of the baseline. Variations in body temperature (maximum − minimum body temperature during surgery) were measured. Patients with false-positive alerts were classified into 2 groups: a global group with alerts observed in 2 or more muscles of the upper and lower extremities, and a focal group with alerts observed in 1 muscle.
False-positive alerts occurred in 100 cases (14%), comprising 60 cases with global and 40 cases with focal alerts. Compared with the 545 true-negative cases, in the false-positive cases the patients had received a significantly higher total propofol dose (1915 mg vs 1380 mg; p < 0.001). In the false-positive cases with global alerts, the patients had also received a higher mean propofol dose than those with focal alerts (4.5 mg/kg/hr vs 4.2 mg/kg/hr; p = 0.087). The cutoff value of the total propofol dose for predicting false-positive alerts, with the best sensitivity and specificity, was 1550 mg. Multivariate logistic analysis revealed that a total propofol dose > 1550 mg (OR 4.583; 95% CI 2.785–7.539; p < 0.001), variation in body temperature (1°C difference; OR 1.691; 95% CI 1.060–2.465; p < 0.01), and estimated blood loss (500-ml difference; OR 1.309; 95% CI 1.155–1.484; p < 0.001) were independently associated with false-positive alerts.
Intraoperative total propofol dose > 1550 mg, larger variation in body temperature, and greater blood loss are independently associated with false-positive alerts during spinal surgery. The authors believe that these factors may contribute to the false-positive global alerts that characterize anesthetic fade. As it is necessary to consider multiple confounding factors to distinguish false-positive alerts from true-positive alerts, including variation in body temperature or ischemic condition, the authors argue the importance of a team approach that includes surgeons, anesthesiologists, and medical engineers.
Yu Yamato, Tomohiko Hasegawa, Sho Kobayashi, Tatsuya Yasuda, Daisuke Togawa, Go Yoshida, Tomohiro Banno, Shin Oe, Yuki Mihara and Yukihiro Matsuyama
Despite the significant incidence of rod fractures (RFs) following long-segment corrective fusion surgery, little is known about the optimal treatment strategy. The objectives of this study were to investigate the time course of clinical symptoms and treatments in patients with RFs following adult spinal deformity (ASD) surgery and to establish treatment recommendations.
This study was a retrospective case series of patients with RFs whose data were retrieved from a prospectively collected single-center database. The authors reviewed the cases of 304 patients (mean age 62.9 years) who underwent ASD surgery. Primary symptoms, time course of symptoms, and treatments were investigated by reviewing medical records. Standing whole-spine radiographs obtained before and after RF development and at last follow-up were evaluated. Osseous union was assessed using CT scans and intraoperative findings.
There were 54 RFs in 53 patients (mean age 68.5 years [range 41–84 years]) occurring at a mean of 21 months (range 6–47 months) after surgery. In 1 patient RF occurred twice, with each case at a different time and level, and the symptoms and treatments for these 2 RFs were analyzed separately (1 case of revision surgery and 1 case of nonoperative treatment). The overall rate of RF observed on radiographs after a minimum follow-up of 1 year was 18.0% (54 of 300 cases). The clinical symptoms at the time of RF were pain in 77.8% (42 of 54 cases) and no onset of new symptoms in 20.5% (11 of 54 cases). The pain was temporary and had subsided in 19 of 42 cases (45%) within 2 weeks. In 36 of the 54 cases (66.7%) (including the first RF in the patient with 2 RFs), patients underwent revision surgery at a mean of 116 days (range 5–888 days) after diagnosis. In 18 cases patients received only nonoperative treatment as of the last follow-up, including 17 cases in which the patients experienced no pain and no remarkable progression of deformity (mean 18.5 months after RF development).
This analysis of 54 RFs in 53 patients following corrective fusion surgery for ASD demonstrates a relationship between symptoms and alignment change. Revision surgeries were performed in a total of 36 cases. Nonoperative care was offered in 18 (33.3%) of 54 cases at the last follow-up, with no additional symptoms in 17 of the 18 cases. These data offer useful information regarding informed decision making for patients in whom an RF occurs after ASD surgery.
Hiroki Ushirozako, Go Yoshida, Tomohiko Hasegawa, Yu Yamato, Tatsuya Yasuda, Tomohiro Banno, Hideyuki Arima, Shin Oe, Tomohiro Yamada, Koichiro Ide, Yuh Watanabe, Tadayoshi Kurita and Yukihiro Matsuyama
Transcranial motor evoked potential (TcMEP) monitoring may be valuable for predicting postoperative neurological complications with a high sensitivity and specificity, but one of the most frequent problems is the high false-positive rate. The purpose of this study was to clarify the differences in the risk factors for false-positive TcMEP alerts seen when performing surgery in patients with pediatric scoliosis and adult spinal deformity and to identify a method to reduce the false-positive rate.
The authors retrospectively analyzed 393 patients (282 adult and 111 pediatric patients) who underwent TcMEP monitoring while under total intravenous anesthesia during spinal deformity surgery. They defined their cutoff (alert) point as a final TcMEP amplitude of ≤ 30% of the baseline amplitude. Patients with false-positive alerts were classified into one of two groups: a group with pediatric scoliosis and a group with adult spinal deformity.
There were 14 cases of false-positive alerts (13%) during pediatric scoliosis surgery and 62 cases of false-positive alerts (22%) during adult spinal deformity surgery. Compared to the true-negative cases during adult spinal deformity surgery, the false-positive cases had a significantly longer duration of surgery and greater estimated blood loss (both p < 0.001). Compared to the true-negative cases during pediatric scoliosis surgery, the false-positive cases had received a significantly higher total fentanyl dose and a higher mean propofol dose (0.75 ± 0.32 mg vs 0.51 ± 0.18 mg [p = 0.014] and 5.6 ± 0.8 mg/kg/hr vs 5.0 ± 0.7 mg/kg/hr [p = 0.009], respectively). A multivariate logistic regression analysis revealed that the duration of surgery (1-hour difference: OR 1.701; 95% CI 1.364–2.120; p < 0.001) was independently associated with false-positive alerts during adult spinal deformity surgery. A multivariate logistic regression analysis revealed that the mean propofol dose (1-mg/kg/hr difference: OR 3.117; 95% CI 1.196–8.123; p = 0.020), the total fentanyl dose (0.05-mg difference; OR 1.270; 95% CI 1.078–1.497; p = 0.004), and the duration of surgery (1-hour difference: OR 2.685; 95% CI 1.131–6.377; p = 0.025) were independently associated with false-positive alerts during pediatric scoliosis surgery.
Longer duration of surgery and greater blood loss are more likely to result in false-positive alerts during adult spinal deformity surgery. In particular, anesthetic doses were associated with false-positive TcMEP alerts during pediatric scoliosis surgery. The authors believe that false-positive alerts during pediatric scoliosis surgery, in particular, are caused by “anesthetic fade.”