| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Southwest Oklahoma MRI, 9901 S. Pennsylvania, Oklahoma City, OK 73159
2
Oklahoma Sports Science and Orthopedics, 2149 S.W. 59th St., Ste. 201,
Oklahoma City, OK 73119.
3
Oklahoma Orthopedic Institue, 1016 S.W. 44th St., Ste. 500, Oklahoma City, OK
73109.
4
San Francisco Magnetic Resonance Center, 3333 California St., Ste. 105, San
Francisco, CA 94118.
Received September 7, 1999;
accepted after revision November 1, 1999.
Address correspondence to O.A. Cvitanic.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We retrospectively reviewed the pre- and postoperative MR images of 109 patients with recurrent radiculopathy, lower back pain, or both 1 year after laser diskectomy of 178 disks. From this group of patients, MR images were also obtained in 11 patients with subchondral marrow changes 5-7 years after surgery. These patients were interviewed regarding residual lower back pain. Thirteen asymptomatic laser diskectomy patients also underwent follow-up MR imaging within 1 year of surgery.
RESULTS. After surgery, subchondral marrow abnormalities were identified in 41 of 109 laser diskectomy patients. The remaining 68 patients had no postoperative subchondral bone marrow changes. Postoperative subchondral marrow changes were not associated with inflammation of the adjacent disk space and did not affect surgical outcome. Bone marrow changes decreased in size in the 11 patients examined 5-7 years after laser diskectomy, and eight of these patients described their lower back pain as improved. In 13 asymptomatic laser diskectomy patients, one new subchondral marrow abnormality was identified.
CONCLUSION. Subchondral marrow abnormalities may occur in the vertebral end plates after holmium:yttrium aluminum garnet laser diskectomy. However, these changes probably do not affect surgical outcomes and appear to resolve over time.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Complications after laser diskectomy include infectious and aseptic diskitis, disk fragmentation, epidural hemorrhage, and injury to the annulus or nerve root [10]. Most radiologists and orthopedists consider MR imaging the technique of choice for recurrent lower back pain after surgery, particularly when spinal infection is a possibility [11, 12]. In the knee, subchondral marrow changes on postoperative MR images represent osteonecrosis at biopsy [13, 14]. Our study was prompted by the discovery that after surgery, signal alterations in the vertebral bone marrow adjacent to the disk not operated on were being variously interpreted as end-plate degeneration, postoperative diskitis, and Schmorl's node formation. Accordingly, we reveal the spectrum of subchondral signal abnormalities on MR imaging after laser diskectomy and determine whether such abnormalities adversely affect surgical outcome.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Between February 1992 and July 1994, one author performed laser diskectomy on 674 patients. We studied 109 patients who underwent follow-up MR imaging of the lumbar spine for postoperative recurrent radiculopathy, lower back pain, or both. We identified 49 men and 60 women to participate in the study (age range, 22-65 years; mean age, 44 years). Patients were included in the study group if they met the following criteria: they underwent MR imaging within 12 months (median, 3.5 months) of laser disk decompression; they had not undergone open lumbar surgery during the interval between laser diskectomy and MR imaging; and their preoperative MR images of the lumbar spine were available for comparison.
The clinical records of our patients did not reveal any patient with postoperative fever or wound infection. WBC, erythrocyte sedimentation rates, and C-reactive protein levels were not routinely checked, and no bone biopsies or disk aspirations were performed. The magnitude of pre- and postoperative lower back pain was determined during a telephone interview that was conducted by a registered nurse. During the interview, patients were asked to rate their lower back pain on a scale of 1-10 (1 = no back pain). Eleven of 41 patients with postoperative subchondral marrow changes on MR imaging agreed to return for MR imaging 5-7 years (mean, 6.1 years) after laser diskectomy. During interviews conducted by one of the authors, these patients were asked to rate their current lower back pain on a scale of 1-10 (1 = no back pain). We focused on lower back pain because we were concerned with symptoms arising from potential vertebral injury. Statistical analysis comparing pre- and postoperative lower back pain was determined using the Student's t test. We considered a p value of less than 0.05 significant.
Follow-up MR imaging of the lumbar spine was also performed on 13 asymptomatic laser diskectomy patients (five men and eight women; age range, 22-55 years; mean age, 41 years). These subjects were randomly selected from a list of individuals with excellent clinical outcomes less than 1 year after laser diskectomy. Informed consent was obtained from all subjects.
We performed a survey of MR images of the lumbar spine from nonsurgical patients with lower back pain or radiculopathy to determine the prevalence of atypical signal abnormalities in the subchondral bone marrow in the general population. A total of 193 MR images were retrospectively reviewed by one radiologist. Signal abnormalities were characterized as flame-shaped, elliptic, or banded, according to criteria outlined in the following text.
Laser Surgery
All laser diskectomy procedures were performed using a holmium:yttrium
aluminum garnet laser system (Omnipulse; Trimedyne, Irvine, CA) with a
side-firing tip. Under fluoroscopic guidance, a needle was posterolaterally
placed in the disk space on the side of the radiculopathy. Once the needle was
positioned centrally and parallel to the horizontal axis of the disk, the
stylet was replaced with the side-firing laser fiber. The laser was operated
at 13 W and 10 Hz to deliver 1.3 J per pulse.
Laser procedures in the first 69 patients were performed with a lateral-vertical firing technique in which laser energy was applied for 10 sec followed by a 10-sec pause at the 12-, 3-, 6-, and 9-o'clock positions until the patient indicated relief of symptoms. Laser diskectomy procedures in the next 40 patients and the 13 asymptomatic follow-up patients were performed using a lateral firing technique in which laser energy was applied for 5 sec followed by a 5-sec pause at the 2-, 4-, 8-, and 10-o'clock positions until symptoms were relieved. The procedure was performed without endoscopic guidance or continuous irrigation, but one or more saline flushes were used at the conclusion of disk decompression to remove debris. The chi-square test was used to determine the incidence of laser-induced subchondral marrow changes using the lateral-vertical compared with the lateral laser firing technique.
MR Imaging
Pre- and postoperative MR images of the lumbar spine were a prerequisite
for inclusion of symptomatic patients in the study. Although the MR imaging
methods varied, all images were obtained with 1.5-T MR systems (Signa or
Horizon; General Electric Medical Systems, Milwaukee, WI) between February
1992 and May 1997. For all patients, T1- and T2-weighted sagittal and
T2-weighted axial images were available at the time of review.
Gadolinium-enhanced T1-weighted axial and sagittal images were available in 47
of 109 patients, including 22 of 41 with new subchondral marrow changes.
For the long-term follow-up of symptomatic patients and for asymptomatic patients, MR imaging was performed on a 1.5-T Horizon scanner. The imaging protocol included sagittal T1-weighted fast spin-echo and sagittal T2-weighted fast spin-echo sequences. The protocol for sagittal T1-weighted fast spin-echo sequences included a TR range/TE range of 450-700/10-20, an echo train length of two, and four signals acquired. The protocol for sagittal T2-weighted fast spin-echo sequences included a TR range/TE range of 2500-3500/65-90, an echo train length of eight or 16, and four signals acquired. The section thickness was 4 mm with a 1-mm gap, matrix size was 512 x 256 or 192, and the field of view was 32 cm.
Image Analysis
MR images of 109 patients were independently interpreted by two
radiologists who were unaware of the level(s) of laser diskectomy and
postoperative clinical course. MR images obtained before laser diskectomy were
available for comparison at the time of interpretation. Kappa values were used
to determine the degree of interobserver variation with regard to the
interpretation of subchondral marrow changes on MR images.
Attention was directed to the vertebral bone marrow on postoperative MR images. Areas of high signal intensity on T1-weighted images (fat deposits, hemangiomas, type II Modic end-plate degeneration [15]) or low signal intensity on T1-weighted images but not subchondral in location were ignored. MR images with areas of low signal intensity in the subchondral bone marrow were compared with preoperative MR images to determine if abnormalities developed during the operative interval. If so, the marrow changes were considered a consequence of laser diskectomy.
Postoperative subchondral marrow abnormalities were recorded if they had both anteroposterior and vertical dimensions greater than 1.0 and 0.5 cm, respectively, on T1-weighted sagittal MR images. We chose this minimum size criterion because of the difficulties in confirming the integrity of an area of end plate less than 1.0 cm in length on MR images. If the subchondral marrow abnormality had an anteroposterior dimension diameter greater than twice the vertical dimension and a curved margin shaped like an ellipse, it was considered elliptic. If the margin of the subchondral marrow abnormality was straight rather than curved, it was considered banded. Abnormalities with a vertical dimension greater than the anteroposterior dimension and a curved margin were considered flame-shaped.
We also examined the postoperative appearance of intervertebral disks. Specifically, we assessed and recorded signal intensity, vertical height, enhancement characteristics, and the presence of recurrent herniations.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
= 0.86). Compared with preoperative MR images, 68 of 109 patients with MR images obtained for postoperative recurrent symptoms did not have new subchondral marrow changes. The remaining 41 patients had new subchondral marrow changes on either side of the treated disk. We divided these patients into three categories on the basis of the configuration of the subchondral marrow change (Fig. 1A,1B,1C).
|
|
|
The most frequently observed configuration of postoperative subchondral marrow abnormality was flame-shaped (Figs. 2A,2B,2C and 3A,3B). We identified 22 flame-shaped abnormalities, all of which were present on only one side of the treated disk. IV gadolinium administration in 13 patients enhanced flame-shaped subchondral marrow abnormalities that were either peripheral (five patients) or homogeneous (eight patients). None of the treated disks enhanced. Two of 109 patients had flame-shaped subchondral marrow changes on preoperative MR images. Neither of these patients was monitored long-term.
|
|
|
|
|
On postoperative MR images, elliptic signal abnormalities were identified in 16 patients; eight were present on one side of the treated disk and eight on both sides (Figs. 4A,4B and 5A,5B,5C). In all seven patients in which IV gadolinium was administered, enhancement of the involved subchondral bone, but not of the treated disks, was noted. None of our patients had elliptic signal abnormalities on preoperative MR images.
|
|
|
|
|
In three patients, banded abnormalities were identified in the subchondral bone on one side of the treated disk. After IV administration of gadolinium, enhancement was identified in each of the three horizontal bands but none of the treated disks. Nineteen of 109 patients had subchondral bands on preoperative MR imaging.
In the 11 patients with subchondral marrow signal abnormalities who were scanned 5-7 years after laser diskectomy, the subchondral marrow had partially or completely returned to normal signal intensity with mild to moderate myeloid depletion. Neither of the patients with flame-shaped subchondral signal abnormalities on preoperative MR images underwent long-term follow-up MR imaging. In the 13 asymptomatic laser diskectomy patients, one flame-shaped lesion was identified that was not present on preoperative MR images. No new elliptic or band-shaped lesions were identified.
Our general survey of lumbar MR images in 193 nonsurgical patients with lower back pain or radiculopathy revealed 20 band-shaped (10%), one flame-shaped (0.5%), and no elliptic (0%) signal abnormalities.
MR Imaging Findings in the Intervertebral Disks
In 109 symptomatic patients, we identified four surgically confirmed
recurrent disk herniations. In patients without recurrent herniation, the
signal intensity and morphology, including vertical height of disks, did not
change on postoperative MR images. We identified no patients with contrast
enhancement in the disks or epidural space suggestive of diskitis. In the 11
patients who underwent imaging 5-7 years after laser diskectomy, all disks
remained hypointense on T1- and T2-weighted MR images, and no recurrent disk
herniations were revealed.
We identified 15 treated disks in the 13 laser diskectomy subjects who remained asymptomatic after surgery. These intervertebral disks were unchanged in terms of signal intensity and disk morphology on postoperative MR images compared with preoperative MR images.
Clinical Assessments
The time between laser diskectomy and the recurrent onset of symptoms was
less than 3 weeks (mean, 12 days). There was no significant difference between
the prevalence of lower back pain in patients with postoperative subchondral
marrow changes (85%; 35/41) and those without (71%; 48/68). Furthermore, when
the prevalence of postoperative subchondral marrow change was compared in
asymptomatic and symptomatic patients whose laser diskectomy procedures were
performed using the same technique (lateral-only firing), the difference, 8%
versus 15%, respectively, was insignificant (p >0.10).
During interviews, eight of 11 long-term follow-up patients described an improvement in lower back pain of two or more points on a 10-point visual analog scale after laser diskectomy (mean improvement, 2.2 points). Statistically, the long-term decrease in lower back pain described by these patients was significant (p <0.02).
Correlation of Subchondral Marrow Abnormalities with Laser Firing
Technique
The incidence of new subchondral marrow changes among symptomatic laser
diskectomy patients in whom the lateral-vertical firing technique was used was
36 (52%) of 69. The incidence of new subchondral marrow changes among
symptomatic laser diskectomy patients in whom the lateral-only firing
technique was used was six (15%) of 40. The difference was significant
(p <0.01).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The postoperative subchondral marrow changes we observed on MR images had sharper margins and more varied configurations than those seen on preoperative MR images. In this regard, our observations concur with those of Schoenenberger et al. [18], who describe elliptic and other complex patterns of energy penetration during real-time MR monitoring of laser diskectomy. These features facilitate the diagnosis of postoperative marrow abnormalities as evidenced by the high level of agreement between the radiologists in this study. Unlike end-plate degeneration, the new subchondral marrow changes we observed on MR images (obtained within 1 year of laser diskectomy) were almost resolved after 5-7 years. Overall, these characteristics and the increased frequency of subchondral marrow abnormalities found when laser fire was intermittently directed at end plates favor iatrogenic end-plate injury over simple end-plate degeneration.
Postoperative diskitis is another important consideration in explaining postoperative subchondral marrow changes. The definitive diagnosis of postoperative diskitis requires aspiration and culturing of the disk, a procedure not performed on any of our patients. However, in contemporary clinical practice, the diagnosis and treatment of diskitis is typically performed on the basis of MR imaging findings alone [19]. These findings include contrast enhancement of the disk and the subchondral bone marrow, disk space narrowing, and end-plate erosion [7, 20]. In our 41 patients, postoperative MR imaging findings were dissimilar to those of diskitis. Specifically, we found no evidence of narrowing or enhancement of the treated disk space or end-plate erosion.
Some of the new subchondral marrow changes we found were originally interpreted as Schmorl's nodes on initial radiology reports. Although Schmorl's nodes can have sharp margins on MR images, similar to the marrow abnormalities observed after laser diskectomy [21], they are always associated with end-plate interruption. End-plate interruption was never observed on MR images or on unenhanced radiographs in the 12 patients with available postoperative radiographs. Differences on MR images between the signal intensity of the involved subchondral marrow and that of the adjacent disk also facilitated the exclusion of Schmorl's nodes.
Clinically, the laser diskectomy patients with new subchondral marrow changes on MR images did not report lower back pain with any greater frequency than the other laser diskectomy patients in our study. Our results also revealed a lack of correlation between the size of postoperative subchondral marrow changes and surgical outcomes. Eight of 11 patients with postoperative subchondral marrow changes that were monitored long-term had decreased lower back pain compared with that of the initial postoperative period.
Postoperative subchondral marrow changes on MR images emphasize the importance of laser firing techniques. Previous reports discuss the vaporization of annulus rather than nucleus pulposus when the laser tip is placed eccentrically in the nucleus pulposus or nonparallel to the horizontal axis of the disk [22, 23]. Our results, correlating firing technique with the frequency of postoperative subchondral marrow changes, suggest a direct relationship with direct end-plate exposure to laser fire.
Although the ability of lasers to damage bone is not in doubt [24, 25], the mechanism of injury is debatable. Because lasers operate by the process of photothermal conversion of tissues, thermal injury to the subchondral bone marrow was immediately considered, but our results contradict this mechanism. Because the blood supply of lumbar vertebrae originates from at least four segmental artery branches [26], some say that the volume of circulating blood may be sufficient to dissipate heat and limit the depth of thermal injury [13, 14].
Pulsed lasers such as the holmium:yttrium aluminum garnet are designed to vaporize the superficial 0.2 mm of the target tissue instantaneously [27], but nontarget tissue heating can still occur by thermal superposition [28]. Specifically, if the time between pulses is shorter than the thermal relaxation time constant (the time for a tissue to dissipate 37% of its heat), then residual heat from prior pulses accumulates in the tissue and may cause thermal injury. In our patients, laser pulses were delivered every 0.10 sec (10 Hz). Because the thermal relaxation time constant for holmium:yttrium aluminum garnet lasers is 0.15 sec, thermal superposition was presumptively in effect. Thus, thermal injury may still be a viable explanation for the postoperative subchondral abnormalities we observed.
Some authors claim a photoaccoustic mechanism for bone damage after laser surgery [14, 24]. This theory suggests that the sudden production of gas in an aqueous environment creates powerful sound waves that damage bone. Because bone transmits sound relatively well, sonic vibrations might be responsible for the deep lesions observed on MR images. However, most reported patients with laser injury in the knee (after arthroscopy) underwent surgery with a continuous wave laser (neodymium:yttrium aluminum garnet) [13, 14]. Because of the comparatively low power and nonpulsatile mode of continuous wave lasers, some question their ability to generate waves capable of damaging bone [29].
One criticism of our study is the absence of bone biopsy specimens obtained from areas of abnormal signal on MR images. The retrospective nature of our study precluded this option because most of the marrow abnormalities were degenerative or inflammatory on original radiology reports. Vertebral bone biopsy would have required a degree of invasiveness unlikely to be accepted by our patient population, most of whom opted for laser surgery because of its decreased invasiveness and lessened morbidity. Other shortcomings inherent in our retrospective study include the availability of enhanced MR images for only 22 of 41 patients and postoperative radiographs for only 12 of 41 patients. However, the consistent absence of disk space inflammation and end-plate destruction in the available images provided compelling evidence against diskitis. Finally, a larger number of asymptomatic laser diskectomy patients would have strengthened our conclusions concerning the clinical significance of subchondral marrow abnormalities in symptomatic patients.
In summary, postoperative subchondral marrow changes are uncommon but should be considered abnormal findings on MR imaging after holmium:yttrium aluminum garnet laser diskectomy. Although at first glance these changes may resemble diskitis, endplate degeneration, or a Schmorl's node, careful analysis of the findings will reveal a pattern of acute onset marrow changes without evidence of disk inflammation. We found evidence that the laser diskectomy technique, specifically the direction of laser fire, influenced the frequency of postoperative subchondral marrow changes. We found no association between lower back pain and postoperative subchondral marrow changes. In all 11 patients followed up, MR imaging of the lumbar spine after 5-7 years revealed a nearly complete resolution of the subchondral marrow abnormalities.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. R. Aiello and O. Cvitanic MR Imaging of Marrow Changes After Laser Diskectomy Am. J. Roentgenol., December 1, 2000; 175(6): 1748 - 1748. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
