DOI: 10.1148/radiol.2352040624
(Radiology 2005;235:562-567.)
© RSNA, 2005
Lumbar Disk Herniation: Do MR Imaging Findings Predict Recurrence after Surgical Diskectomy?1
Claudio Dora, MD,
Marius R. Schmid, MD,
Achim Elfering, PhD,
Marco Zanetti, MD,
Juerg Hodler, MD, MBA and
Norbert Boos, MD
1 From the Center for Spinal Surgery (C.D., N.B.) and Department of Radiology (M.R.S., M.Z., J.H.), University of Zurich, Balgrist Hospital, Forchstrasse 340, CH-8008 Zurich, Switzerland; and Department of Psychology, University of Bern, Bern, Switzerland (A.E.). Received April 5, 2004; revision requested June 16; revision received July 6; accepted August 5. Address correspondence to M.R.S. (e-mail: marius.schmid@balgrist.ch).
 |
ABSTRACT
|
|---|
PURPOSE: To retrospectively evaluate if the degree of disk degeneration and disk herniation volume at magnetic resonance (MR) imaging are risk factors for recurrent disk herniation.
MATERIALS AND METHODS: The institutional review board permits such retrospective studies, and individual informed consent was not required. MR imaging findings obtained before initial diskectomy in 30 patients (mean age, 42.8 years; 10 women, 20 men) with recurrent disk herniation (study group) and 30 patients (mean age, 42.2 years; nine women, 21 men) without recurrence for at least 2 years after surgery (control group) were compared. Disk degeneration was assessed on T2-weighted sagittal MR images with a five-point grading system (grade I indicated no degeneration; grade II, horizontal hypointense bands within disk; grade III, inhomogeneous disk with intermediate signal intensity; grade IV, no distinction between inner and outer parts of disk; and grade V, collapsed hypointense disk). Disk herniation was classified as representing protrusion, extrusion, or sequestration. The volume of both the affected intervertebral disk and the herniated disk material was measured. Qualitative and quantitative analyses were performed by two readers. The 
2 test was used for comparison of categorical variables. For comparison of disk degeneration and volumes between patients with and those without recurrence, a paired two-tailed t test was used. Odds ratios based on the extent of disk degeneration were calculated for the entire sample.
RESULTS: Advanced disk degeneration (grades IV and V) was significantly less frequent in the study group than in the control group (P < .006). The risk of recurrent disk herniation decreased by a factor of 3.4 for each increase in grade of disk degeneration (odds ratio: 3.58; 95% confidence interval: 1.3, 9.6; P = .011). Mean disk herniation volume as a percentage of intervertebral disk volume was 10.59% ± 3.41 in the study group and 11.56% ± 3.84 in the control group. This difference was not significant (P = .31).
CONCLUSION: Minor disk degeneration but not herniation volume represents a risk factor for the recurrence of disk herniation after diskectomy.
© RSNA, 2005
|
INTRODUCTION
|
|---|
The reported prevalence of unsatisfactory results after diskectomy is between 5% and 20% (1–3). Recurrent herniation is one of the most common reasons for an unsatisfactory outcome and occurs in 5%–11% of surgically treated patients (3–5). Several quite different conditions have been described as "recurrent disk herniation": (a) recurrence at the same disk level and side as the primary herniation (1–6), (b) contralateral disk herniation at the same level (7), and (c) a new herniation at a different level (4,6). Recurrent disk herniation at the same level and side differs from other herniation types because the extent of the initial annular defect and the surgical incision may represent predisposing factors for recurrence. In the present study, we refer to this condition when we use the term recurrent disk herniation.
In a recent article, Carragee et al (8) reported that the size of the annular defect is one of the most important risk factors for recurrent disk herniation. In our experience, recurrent disk herniation occurs more frequently in patients whose nucleus pulposus still has a normal or almost-normal magnetic resonance (MR) imaging appearance. When such disks are surgically explored, gel-like material is commonly found to be protruding through the annular defect. Burns and Young (9) performed surgical re-explorations in 18 patients with recurrent disk herniation and found that disk protrusion with only a slightly bulging annulus fibrosus was the least favorable form of disk herniation owing to the difficulty in sufficiently removing the nucleus pulposus in such disks. Jackson (10) suggested that incomplete degeneration of the disk is related to incomplete removal of disk material during the original operation.
We hypothesized that recurrent disk herniation is less common in advanced disk degeneration and in larger volumes of disk herniation. Thus, the purpose of our study was to retrospectively evaluate if the degree of disk degeneration and disk herniation volume at MR imaging are risk factors for recurrent disk herniation.
|
MATERIALS AND METHODS
|
|---|
Study Group
Thirty consecutive patients (mean age, 42.8 years; age range, 23–62 years; 10 women, 20 men) who underwent surgical treatment of recurrent disk herniation between January 1997 and December 2001 and met our inclusion criteria (absence of leg pain for at least 7 days after primary diskectomy and availability of MR examination results obtained before the first surgical intervention) were retrospectively identified. Exclusion criteria were previous spinal surgery or concomitant spinal disease.
The indication for revision surgery was a clinically relevant recurrent disk herniation with sensory and motor deficits. The motor deficit was considered to be relevant when its Medical Research Council grade (11) was lower than 3 (grade 0 indicates no motor function; grade 1, flicker or trace contraction; grade 2, active movement with gravity eliminated; grade 3, active movement against gravity; grade 4, active movement against resistance; and grade 5, normal strength). Radicular leg pain unresponsive to an adequate trial of conservative treatment was also considered to represent an indication for surgery.
For 22 of the 30 patients, the initial surgery had been performed at our institution (University Hospital Balgrist). The remaining eight patients had been treated elsewhere. On the basis of a general permit issued by the responsible state agency, our institutional review board allows retrospective analysis of patient data relating to standard diagnostic or therapeutic procedures without individual informed consent.
Control Group
Thirty consecutive patients (mean age, 42.2 years; age range, 18–63 years; nine women, 21 men) who underwent primary diskectomy were retrospectively identified between January and June 2001 on the basis of the following inclusion criteria: MR imaging examination performed before surgery, surgically verified disk herniation, no major concomitant spinal abnormalities, failure of adequate (at least 6 weeks) trial of conservative treatment or motor deficit (Medical Research Council grade 3 or worse), and no clinical evidence of recurrent herniation during a 2-year follow-up. The follow-up data were obtained during a routine clinical examination 1 year and during a structured telephone interview 2 years after surgery. These telephone interviews were not part of routine care; all interviews were performed during 2003 by one orthopedic surgeon (C.D.). The heights and weights of both the control group patients and the study group patients were obtained from anesthesia records (n = 52) or during the telephone interview (n = 8). At the time these interviews were conducted, there were no regulations or rulings in our country that restricted their performance; this fact was confirmed with our institutional review board.
Surgical Technique
In all patients (both groups, n = 60), diskectomy was performed through a standard interlaminar approach with limited laminotomy. If required for adequate exposure, a medial facetectomy was performed. Through a rectangular incision into the annulus, herniated tissue and protruding parts of the posterolateral annulus were removed. Loose nucleus pulposus fragments were removed from the intervertebral disk space by using a rongeur. No attempt was made to radically excise the disk. For all patients in the control group (n = 30) and 22 patients in the study group, surgery was performed by one of three staff spine surgeons with 11 (N.B.), 11, and 5 years of experience. For the remaining eight patients, who had undergone surgery at other institutions, the surgical reports were reviewed by one orthopedic surgeon (C.D.) and indicated that surgery was performed in a similar fashion.
MR Imaging Protocol
Almost all of the MR imaging examinations in the control group (n = 28) and a majority of the examinations in the study group (n = 20) were performed at our institution by using a 1.0-T MR unit (Expert; Siemens Medical Solutions, Erlangen, Germany) and a dedicated receive-only spine coil. The imaging protocol for these 48 examinations consisted of a sagittal T1-weighted spin-echo sequence (repetition time msec/echo time msec, 700/12; section thickness, 4 mm; intersection gap, 0.8 mm; field of view, 300 mm; matrix, 512 x 512; number of acquisitions, four), a sagittal T2-weighted turbo spin-echo sequence (5000/130; section thickness, 4 mm; intersection gap, 0.8 mm; echo train length, 15; number of acquisitions, four), and a transverse T2-weighted turbo spin-echo sequence (4000/96; section thickness, 4 mm; intersection gap, 0.8 mm; echo train length, seven; field of view, 150 mm; matrix 256 x 256; number of acquisitions, two). MR imaging examinations in the remaining 12 patients had been performed at other institutions. Comparable MR imaging protocols, including the acquisition of sagittal T1- and T2-weighted and transverse T2-weighted images, had been used.
Qualitative Image Analysis
The MR images obtained before primary diskectomy were assessed in consensus by a radiologist (M.R.S.) with 5 years of experience as a staff musculoskeletal radiologist and a staff orthopedic surgeon (C.D.) with 12 years of experience as an orthopedist. Both observers were blinded to the patients’ clinical findings. Disk degeneration was graded on T2-weighted sagittal images by using a five-point scale according to the method of Pfirrmann et al (12). Grade I indicates the presence of a homogeneously hyperintense nucleus pulposus that is clearly distinct from the hypointense outer annular fibers. In grade II degeneration, the nucleus pulposus is inhomogeneous, and horizontal hypointense bands may be present in a sandwichlike configuration. In grade III degeneration, the inner parts of the disk are inhomogeneous and have intermediate signal intensity. In grade IV degeneration, the distinction between the inner and outer parts of the disk is lost, and the inner parts of the disk have intermediate or low signal intensity. In grade V degeneration, the disk is collapsed.
The extent of disk herniation was rated on T1- and T2-weighted images according to the methods of Masaryk et al (13) and Modic et al (14), with differentiation between protrusion, extrusion, and sequestration. A disk was considered protruded if the greatest plane in any direction between the edges of the disk material beyond the disk space was less than the distance between the edges of the base when measured in the same plane. A disk was considered extruded if, in at least one plane, any one distance between the edges of the disk material beyond the disk space was greater than the distance between the edges of the base measured in the same plane. A sequestrated disk herniation was diagnosed when the herniated disk material was clearly separate from the originating disk or when the signal intensity of the herniated material was different from that of the originating disk. Neural compromise was assessed according to the system of Pfirrmann et al (15) as no compromise, contact, deviation, or compression.
Endplate and adjacent bone marrow changes were graded according to the system of Modic et al (14) as follows: no abnormality, low signal intensity on T1-weighted images, high signal intensity on T2-weighted images in comparison with the intensity of normal fatty bone marrow (high signal intensity type I), or high signal intensity on both kinds of images (high signal intensity type II).
Quantitative Image Analysis
The volumes of the intervertebral disks and the herniated disk parts were measured by two independent readers (M.R.S., C.D.). Both readers were blinded to the clinical outcome after surgery. Initially, all sagittal T2-weighted images were digitized by using a 12-bit scanner (VXR-12 plus film digitizer; Vidar Systems, Herndon, Va). Volumes were measured by using Image Access software (Imagic Imaging Solutions, Glattbrugg, Switzerland). The cross-sectional area of the intervertebral disk, including the hernia, was measured on each image that included any disk material. Additional measurements limited to the cross-sectional area of the herniated part of the disk were performed (Fig 1). The cross-sectional area was multiplied by the sum of the section thickness plus the intersection gap. Total disk volume and herniation volume were then calculated by adding the resulting values from all images.
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Two identical sagittal T2-weighted MR images (5000/130) obtained in 53-year-old male patient show how the cross-sectional area of the intervertebral disk is defined when (a) the entire disk, including the herniated part, and (b) only the herniated part are measured.
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Two identical sagittal T2-weighted MR images (5000/130) obtained in 53-year-old male patient show how the cross-sectional area of the intervertebral disk is defined when (a) the entire disk, including the herniated part, and (b) only the herniated part are measured.
|
|
Morphologic Matching
To exclude potential confounding factors, a paired matching was performed by an orthopedic surgeon (C.D.). For every patient from the study group, one matching control subject was selected from the control group according to the following criteria: same sex and same intervertebral disk level affected (L5-S1 vs L3-4 or L4-5), same extent of disk herniation (ie, protrusion, extrusion, or sequestration), and a similar body mass index (BMI) (differing by no more than ±5 kg/m2) and age (differing by no more than ±10 years). For 10 patients from the study group, no matching control subject (according to the above criteria) could be found, resulting in only 20 perfectly matched pairs whose selected MR imaging results could be compared in pairwise fashion.
Data and Statistical Analyses
The intra- and interobserver reliabilities of the classification system for disk degeneration have been evaluated previously; 
values between 0.69 and 0.9 have been reported (12). Observer agreement in volume measurement was assessed by using intraclass correlation coefficients and by calculating the mean difference between raters (ie, the difference between raters regardless of the sign [positive or negative] of the difference) and the 95% confidence interval. For comparison of categorical variables (disk degeneration, type of herniation, neural compromise, and endplate changes), the 2 test was used. Quantitative variables (age, height, weight, BMI, and volume of intervertebral and herniated disk material) were analyzed by using an unpaired two-tailed t test. For comparison of disk degeneration and volumes between morphology-matched pairs of herniated disks with and herniated disks without recurrence, a paired two-tailed t test was used. All analyses were also performed by using multiple logistic regression analysis or analysis of covariance to check for the influence of age, BMI, and sex as covariates. SPSS version 11.5 (SPSS, Chicago, Ill) was used for all calculations. The level of statistical significance was set at .05 for two-tailed testing.
|
RESULTS
|
|---|
Demographic Data
The study and control groups (Table 1) were not significantly different in terms of age, sex, height, weight, or BMI (P = .63 to P = .87). The mean time between primary diskectomy and revision diskectomy was 7 months ± 8 (standard deviation) (range, 7 days to 23 months). Mean follow-up of patients in the control group was 29 months ± 3 (range, 24–33 months). The time interval between primary diskectomy and revision diskectomy was positively related to the percentage volume of herniation (r = 0.57, P = .001) and negatively related to BMI (r = –0.45, P = .014)—that is, revision diskectomy took place later in patients with more severely herniated disks and patients with lower BMIs.
Qualitative Image Analysis
All patients in both groups were seen to have at least one disk extrusion at preoperative MR imaging. Sequestrated disk herniations were less frequent in the study group (Table 2). This difference, however, was not significant (P < .11). Neural compromise was diagnosed in all cases (n = 60) on the basis of the MR imaging findings. Endplate changes were slightly but not significantly more frequent in the control group (P < .12). Advanced disk degeneration (grades IV and V; Fig 2) was significantly more frequent in the control group than in the study group (P < .006). No or minor disk degeneration (grades I–III, Fig 3) was observed before the initial surgery in 83% of patients in the study group and in 50% of patients in the control group. The risk of recurrent disk herniation was reduced by a factor of 3.4 for each increase in the grade of disk degeneration (odds ratio: 3.58; 95% confidence interval: 1.3, 9.6; P < .011 [controlled for age, BMI, and sex]).
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Grade IV disk degeneration without recurrent disk herniation. T2-weighted sagittal MR image (5000/130) in 39-year-old man in control group with lumbar disk herniation (black arrow) at the L4-5 level before diskectomy. Disk degeneration was rated as grade IV because a homogeneous hypointense disk (white arrow) without collapse of the intervertebral space is seen. The risk of recurrence after diskectomy in such disks is low.
|
|
View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Grade II disk degeneration with recurrent disk herniation. T2-weighted sagittal MR image (5000/130) obtained in 41-year-old man in study group with lumbar disk herniation (long black arrow) at the L4-5 level before primary diskectomy. Disk degeneration was rated as grade II because an inhomogeneous bright nucleus pulposus (white arrow) with hypointense bands (short black arrows) was seen. When compared with the risk of the patient in Figure 2, this patient’s risk of experiencing recurrence after diskectomy was 6.8 fold.
|
|
Analysis of only the 20 matched pairs of disk herniations yielded results that were consistent with those yielded by the overall analysis. In these 20 pairs, advanced disk degeneration was significantly (P < .02) less frequent in the study group than in the control group (Table 3).
Power analysis was performed with 2 testing for both evaluations (n = 60 and n = 40) (16). For the entire study population (n = 60), the effect size was 0.42 and the power was .90 (good). For the matched group (n = 40), the effect size was 0.36 and the power was .63 (fair).
Quantitative Image Analysis
Owing to a problem in data storage after imaging, data from one individual in the study group were lost for quantitative image analysis. The intraclass correlation coefficient between the two observers for the volume measurements was 0.97 (95% confidence interval: 0.95, 0.99) for both groups. The mean interobserver differences for the measurement of disk volume, as well as those for the measurement of absolute and relative herniation volume, were less than 1 mm3 and less than 1%, respectively (Table 4), in the control and in the study groups. No significant differences with regard to volume measurements were observed for the total (Table 5) and matched-pair groups (Table 6).
|
DISCUSSION
|
|---|
The present study results demonstrate that low-grade disk degeneration (grades I–III) is an important risk factor for recurrent disk herniation. This finding is in agreement with the findings of a long-term outcome study of standard diskectomy performed for lumbar disk herniation conducted by Yorimitsu et al (17). In that study, patients with preserved disk height before initial surgery generally had a favorable initial result but a high risk of recurrent disk herniation. This finding is contradicted by the results of a prospective controlled study performed by Cinotti et al (18), who observed that ipsilateral recurrent lumbar disk herniation occurred more often when severe disk degeneration was present. However, their patients had additional abnormalities such as lateral stenosis, narrowed intervertebral foramina, and epidural scarring. Additional confounding factors, such as the morphology of disk herniation, were not included in their investigation. However, different types of herniation have a proved effect on the risk of recurrence. Using a classification that was based on surgically assessed fragment type and annular competence, Carragee et al (8) observed an association between annular defects and recurrence. They found that an extruded fragment with massive posterior annular loss had the highest prevalence of recurrent herniation. This appears plausible because there is nearly no annular healing (19–21).
Assessment of the annular defect in the study of Carragee et al was based on probing with Penfield dissectors. However, we have experienced difficulties with this assessment method because of the bias related to the filling of the annular defect with granulation tissue—a filling that is dependent on the time interval between herniation and surgery. Furthermore, an occluded annular defect can be reopened during surgery with variable force and influences the size of the Penfield dissector that can be used. Owing to the restrictive surgical policy for minor disk protrusions at our institution, all disk herniations in our study were classified as extrusions or sequestrations and had sizeable annular defects. The results for the entire study and control groups were confirmed at evaluation of the 20 matched patient pairs, for which possibly confounding factors such as disk level, type of disk herniation, and other factors were eliminated.
The present study did not reveal any significant differences between patients with and those without recurrence with regard to either disk volume or relative herniated volume. Interestingly, however, we observed significantly greater time intervals between surgery and recurrent symptoms in the study group patients who had a higher relative volume of disk herniation and in those who had a lower BMI. Therefore, it seems that a higher relative herniated disk volume does not prevent recurrence of disk herniation, but, in cases of recurrence, the time interval between primary surgery and the recurrence of the hernia will be longer.
A potential limitation of this study was its retrospective design. However, we believe that our hypothesis can reliably be assessed without a prospective design given the condition that the surgical procedure is standardized. Because of the low incidence of recurrent disk herniation, a prospective study design would have required a substantial amount of time and would have risked the loss of a relevant number of patients during follow-up, thus threatening the power of the study. The power of this study was acceptable (.90 for the whole sample and .63 for the matched pairs). In addition, no follow-up MR imaging examinations were performed in the control group. However, even if some individuals in the control group had been found to have recurrent disk herniation at MR imaging, such herniation would have been asymptomatic and therefore without clinical relevance. One additional limitation could be that recurrent herniation was defined by using a clinical parameter (a pain-free interval of at least 7 days after surgery), and no postoperative imaging was performed to confirm the complete removal of the original disk herniation.
The observation that patients with only minor disk degeneration have an increased risk of recurrent disk herniation is clinically relevant. A normal intervertebral disk (grade II at MR imaging) in an adult has a 6.8-fold increase in the risk of recurrent herniation as compared with the risk of a disk with advanced degeneration (grade IV). This is important to consider when patients are informed about the potential outcome after diskectomy.
We conclude that only minor degeneration—and not herniation volume—represents a risk factor for recurrent disk herniation after surgical diskectomy.
|
FOOTNOTES
|
|---|
Abbreviation: BMI = body mass index
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, C.D., M.R.S., M.Z., J.H., N.B.; study concepts, C.D., M.R.S., N.B.; study design, C.D., M.R.S., M.Z., J.H., N.B.; literature research, C.D., M.R.S.; clinical studies, C.D., A.E., N.B.; data acquisition and analysis/interpretation, all authors; statistical analysis, A.E., C.D.; manuscript preparation, C.D., M.R.S.; manuscript definition of intellectual content and final version approval, all authors; manuscript editing, C.D., M.R.S., J.H., N.B.; manuscript revision/review, C.D., M.R.S., M.Z., J.H., N.B.
|
REFERENCES
|
|---|
- Crock HV. Observations on the management of failed spinal operations. J Bone Joint Surg Br 1976; 58:193-199.
- Greenwood J, McGuire TH, Kimbell F. A study of the causes of failure in the herniated intervertebral disc operation: an analysis of sixty-seven reoperated cases. J Neurosurg 1952; 9:15-20.[Medline]
- O’Sullivan MG, Connolly AE, Buckley TF. Recurrent lumbar disc protrusion. Br J Neurosurg 1990; 4:319-325.[Medline]
- Connolly ES. Surgery for recurrent lumbar disc herniation. Clin Neurosurg 1992; 39:211-216.[Medline]
- Fandino J, Botana C, Viladrich A, Gomez-Bueno J. Reoperation after lumbar disc surgery: results in 130 cases. Acta Neurochir Wien 1993; 122:102-104.[CrossRef][Medline]
- Law JD, Lehman RA, Kirsch WM. Reoperation after lumbar intervertebral disc surgery. J Neurosurg 1978; 48:259-263.[Medline]
- Cinotti G, Gumina S, Giannicola G, et al. Contralateral recurrent lumbar disc herniation. Spine 1999; 24:800-806.[CrossRef][Medline]
- Carragee EJ, Han MY, Suen PW, Kim D. Clinical outcomes after lumbar discectomy for sciatica: the effects of fragment type and annular competence. J Bone Joint Surg Am 2003; 85-A:102-108.
- Burns BH, Young RH. Results of surgery in sciatica and low back pain. Lancet 1951; 1:245-249.[CrossRef][Medline]
- Jackson RK. The long-term effects of wide laminectomy for lumbar disc excision: a review of 130 patients. J Bone Joint Surg Br 1971; 53:609-616.
- Kendall HO, Kendall FP. Muscles: testing and functions Baltimore, Md: Williams & Wilkins, 1974.
- Pfirrmann CW, Metzdorf A, Zanetti M, Hodler J, Boos N. Magnetic resonance classification of lumbar intervertebral disc degeneration. Spine 2001; 26:1873-1878.[CrossRef][Medline]
- Masaryk TJ, Ross JS, Modic MT, Boumphrey F, Bohlman H, Wilbert G. High-resolution MR imaging of sequestered lumbar intervertebral disks. AJR Am J Roentgenol 1988; 150:1155-1162.[Abstract/Free Full Text]
- Modic MT, Steinberg PM, Ross JS, Masaryk TJ, Carter JR. Degenerative disk disease: assessment of changes in vertebral body marrow with MR imaging. Radiology 1988; 166:193-199.[Abstract/Free Full Text]
- Pfirrmann CW, Dora C, Schmid M, Zanetti M, Hodler J, Boos N. Grading of lumbar nerve root compromise with magnetic resonance imaging: a reliability study with surgical correlation. Radiology 2004; 230:583-588.[Abstract/Free Full Text]
- Cohen J. Statistical power analysis for the behavioral sciences San Diego, Calif: Academic Press, 1977.
- Yorimitsu E, Chiba K, Toyama Y, Hirabayashi K. Long-term outcomes of standard discectomy for lumbar disc herniation: a follow-up study of more than 10 years. Spine 2001; 26:652-657.[CrossRef][Medline]
- Cinotti G, Roysam GS, Eisenstein SM, Postacchini F. Ipsilateral recurrent disc herniation: a prospective, controlled study. J Bone Joint Surg Br 1998; 80:825-832.
- Bayliss MT, Urban JP, Johnstone B, Holm S. In vitro method for measuring synthesis rates in the intervertebral disc. J Orthop Res 1986; 4:10-17.[CrossRef][Medline]
- Sato M, Asazuma T, Ishihara M, et al. An experimental study of the regeneration of the intervertebral disc with an allograft of cultured annulus fibrosus cells using a tissue-engineering method. Spine 2003; 28:548-553.[CrossRef][Medline]
- Sumida K, Sato K, Aoki M, Matsuyama Y, Iwata H. Serial changes in the rate of proteoglycan synthesis after chemonucleolysis of rabbit intervertebral discs. Spine 1999; 24:1066-1070.[CrossRef][Medline]
TOP
ABSTRACT
INTRODUCTION
