Massive lumbar disc herniation with complete dural sac stenosis.

BACKGROUND: Large lumbar disc herniation (LDH) has been reported to have a greater tendency to resolve in clinical and pathomorphological evolutions. However, various definitions of large LDH have been used without validation, and the clinical symptoms of large LDH have not been fully elucidated. We conducted a retrospective analysis to determine the clinical characteristics and treatment outcome of massive LDH with complete dural sac stenosis.
MATERIALS AND METHODS: We retrospectively reviewed 33 cases of LDH with complete dural sac stenosis on magnetic resonance imaging. Complete dural sac stenosis was defined as no recognizable rootlet and cerebrospinal fluid signal on T2-weighed axial MR images. The clinical outcome parameters included back pain, leg pain, Oswestry disability index (ODI), and neurological dysfunction. The paired t-test and Wilcoxon's signed rank test were used to compare serial changes in back pain, leg pain and neurological dysfunction.
RESULTS: Mean duration of followup was 66 months (range 24 - 108 months). There were 24 male and 9 female. The mean age was 37 years (range 20 - 53 years). At presentation, mean visual analogue scales for back pain and leg pain were 75.3 ± 19.1 (range 12 - 100) and 80.2 ± 14.6 (range 0 -100), respectively. Mean ODI was 67.1 ± 18.8 (range 26 - 88). Neurological dysfunction was found in 9 patients (27.3%), and the bowel/bladder dysfunction was found in 2 patients (3.1%). Conservative treatment was performed in 21 patients (63.6%) with satisfactory results. Seven patients underwent decompressive surgery, and 5 underwent posterolateral fusion.
CONCLUSIONS: A massive LDH with complete dural sac stenosis was found to be associated with severe back and leg pain at presentation, however surgical treatment can be deferred unless significant neurological symptoms occur.

INTRODUCTION

The majority of patients with lumbar disc herniation (LDH) show a favorable prognosis with nonoperative treatment.1234 Herniated discal tissue spontaneously decreases in size and even disappears.567 The clinical symptoms usually improve regardless of the resolution of herniated discal tissues.28 Large LDH has been reported to have a greater tendency to resolve in clinical and pathomorphological evolutions.7910111213 The ‘massive’ LDH, which was defined in previous studies as the disc material occupying >50% of the spinal canal on axial magnetic resonance (MR) images, showed a successful outcome with nonoperative treatment.1415 Therefore, operative treatment of massive LDH can be deferred unless significant neurological symptoms are observed.

On the other hand, the severe dural sac compression by a massive LDH often worries clinicians and patients, and pushes them towards operative treatment for fear of cauda equina syndrome (CES) or significant neurological dysfunction.2161718 Discectomy along with spinal fusion rather than discectomy alone is sometimes preferred for a massive LDH.1920

In the present study, we defined a massive LDH through observations of dural sac morphology rather than by quantifying or qualifying the size of herniated discal tissue. Little information is available in the literature regarding the clinical outcome of extremely large LDH with severe dural sac compression. The aim of this study was to present the clinical manifestations and treatment outcomes of massive LDH with complete dural sac stenosis.

MATERIALS AND METHODS

882 patients were registered in the clinical LDH database at the department of orthopaedic surgery in a tertiary hospital between July 2002 and November 2009. Our LDH database involved the patients’ demographics, radiological findings, treatment methods, and clinical assessment at each time of followup. Of these, 38 patients were recognized to have a massive LDH with complete dural sac stenosis, and were included in this study. Complete dural sac stenosis was identified by reviewing dural sac morphologies on lumbar MR images; an LDH with no recognizable rootlet and cerebrospinal fluid signal on T2-weighed axial MR images of the lumbar spine [Figure 1]. Complete dural sac stenosis with ligament flavum hypertrophy and spondylolisthesis were excluded. MR interpretations for inclusion were carried out by three experienced orthopaedic surgeons not involved in the care of the study subjects. Disagreements between interpretations were resolved through discussion and a consensus opinion was reached. Five of the 38 patients were lost to followup before 2 years had elapsed. Accordingly, 33 patients with a > 2-year followup comprised the study cohort.

Figure 1

T2WI mid sagittal and axial cuts showing (a-c) Massive LDH with complete duralsac stenosis was defined as no recognizable rootlet and cerebrospinal fluid signal on T2-weighed axial MR images of the lumbar spine

Information on gender, age, body mass index (BMI), smoking, occupation, prodromal symptoms, neurological function, treatment, and clinical outcome were obtained from medical records. The clinical outcome parameters included back pain, leg pain, Oswestry disability index (ODI), and neurological dysfunction. Pain was scored using a visual analogue scale (VAS) on a 100 mm horizontal line, where 0 represented no pain and 100 represented the maximum imaginable pain. ODI was scored from 0 to 100, where lower scores indicated less severe symptoms. Neurological dysfunctions were categorized into 5 groups as previously described21 with modification as follows. 1) Normal: Neurologically intact. 2) Mild: Any sensory change and motor weakness of Medical Research Council (MRC) grade ≥ 4 in any lower extremity muscle by the manual muscle testing.22 3) Moderate: Any sensory change and motor weakness of MRC grade < 4 in any lower extremity muscle. 4) Incomplete or impending CES: Signs of saddle sensory disturbance, bilateral sciatica, and lower extremity weakness, but intact bowel/bladder function. 5) Complete CES: Any other sign of CES with uncontrolled bowel/bladder function. Outcome parameters were measured at baseline, 1 month, 6 months, 12 months, and at last followup.

Patients with impending or established CES, progressive neurological dysfunction, or intractable pain were treated surgically. Standard surgical procedures involved the removal of herniated tissue and wide neural decompression. Spinal fusion was additionally performed in cases with pre-existing degeneration or instability. All surgeries were performed by the senior author (C.H.J).

Descriptive statistics are summarized as frequencies and percentages for categorical variables, and as means and standard deviations for continuous variables. The paired t-test and Wilcoxon's signed rank test were used to compare serial changes in back pain, leg pain and neurological dysfunction. Interobserver reliability for the interpretation of complete dural sac stenosis was assessed using kappa values, which were interpreted as follows: Moderate (0.41 ≤ κ <0.60), substantial (0.60 ≤ κ <0.80), and almost perfect (0.80 ≤ κ <1.00).23 Statistical analysis was carried out using SPSS version 14.0 software (SPSS Inc, Chicago, IL). A P value of <0.05 was considered significant.

RESULTS

Of the 33 massive LDHs with complete dural sac stenosis, 26 were identified as extrusions, and the remaining 7 were sequestrations. The mean age of the 24 male and 9 female patients was 37 years (range 20-53 years), and the mean BMI was 24.2 m/kg2 (range 19.0-33.7 m/kg2). Eighteen patients (55%) were office workers, and 3 (9%) were engaged in heavy labor. Eleven patients (33%) were smokers. One patient had a herniated disc at L2-3, three at L3-4, nineteen at L4-5, and 10 at L5-S1. The mean duration of followup was 66 months (range 24-108 months).

Twenty nine patients (87.9%) had prodromal back or leg pain with a mean VAS score of 36.4 ± 32.1 (range 0-71) for 26.1 ± 32.2 days (range 3-90 days) before presentation. At presentation, mean VAS scores for back pain and leg pain were 75.3 ± 19.1 (range 12-100) and 80.2 ± 14.6 (range 0-100), respectively. Back and leg pain were aggravated at 4.2 (range 0-15 days) and 3.3 (range 0-7 days) days before presentation, respectively. The mean ODI score was 67.1 ± 18.8 (range 26-88). Neurological symptoms occurred for a mean 3.1 days (range 0-60 days) before presentation. Twenty four patients (72.7%) had normal or mild neurological dysfunction. Nine patients (27.3%) had neurological dysfunction. Three patients (9%) showed moderate dysfunction. Bowel and bladder symptoms were present in 2 patients (6.1%) [Table 1].

Table 1

Clinical presentation

Twenty one patients (63.6%) were treated nonoperatively. Nonoperative treatment included oral medication, physical therapy, and patient education. Commonly used medications were nonsteroidal antiinflammatory drugs, corticosteroids, muscle relaxants, and opioid pain medications. Epidural block was performed in 19 patients. Three patients (9%) underwent decompression and primary posterolateral fusion for concomitant degenerative changes. Both back pain and leg pain decreased significantly within 1 month, and continued to decrease at the time of last followup. The mean VAS score for back pain was 75.3 ± 19.1 at presentation and 35.3 ± 16.4 at last followup (P < 0.001) [Figure 2]. The mean VAS score for leg pain was 80.2 ± 14.6 at presentation and 24.9 ± 19.9 at last followup (P < 0.001) [Figure 3]. ODI decreased significantly within 6 months [Figure 4]. The mean ODI was 67.1 ± 18.8 at presentation and 18.5 ± 9.1 at last followup (P < 0.001). Neurological function also improved significantly within 1 month and continued to improve at last followup (P = 0.015) [Figure 5].

Figure 2

Bar diagram showing mean VAS for back pain during the followup period. Back pain significantly decreased within one month (P < 0.001) and continued to decrease at last followup

Figure 3

Bar diagram showing mean VAS for leg pain during the followup period. Leg pain significantly decreased within one month (P < 0.001) and continued to decrease at last followup

Figure 4

Bar diagram showing mean ODI score during the followup period. ODI score significantly decreased within 6 month (P < 0.001) and continued to decrease at last followup

Figure 5

Bar diagram showing neurological function significantly improved within one month (P= 0.015) and continued to improve at last followup

Three patients (9.1%) underwent discectomy along with posterolateral fusion and 9 patients (27.3%) underwent discectomy. Two of the 9 discectomy patients underwent revision surgery with posterolateral fusion for persistent symptoms. Back pain, leg pain, and disability decreased within 1 month. The mean VAS score for back pain was 68.5 ± 23.4 at presentation and 21.5 ± 22.8 at 1 month (P < 0.001). The mean VAS score for leg pain was 88.7 ± 19.1 at presentation and 19.3 ± 28.6 at 1 month (P < 0.001). The mean ODI was 78.8 ± 25.4 at presentation and 14.6 ± 10.9 at last followup (P < 0.001).

At the last followup, 26 of the 33 patients (78.8%) had normal neurological function. Two patients that had underwent operative treatment showed persistent bladder dysfunction only.

The kappa value of interobserver agreement regarding the interpretation of complete dural sac stenosis was 0.78, which indicated substantial agreement.

DISCUSSION

Many efforts have been made to derive a nomenclature and classification for LDH that accurately describes the morphologies of displaced discal tissues.24252627282930 LDH sizes are measured quantitatively by computing greatest cross-sectional areas on axial computed tomographic or MR images.3132 The joint society committee of North American Spine Society, American Society of Spine Radiology, and American Society of Neuroradiology proposed a qualitative grading scheme based on the following definitions; mild (canal compromise of less than one third of the canal), moderate (between one third and two thirds), or severe (over two thirds).29 However, quantification or estimation of LDH size is rarely performed because these are based on one slice of an LDH and cannot provide information about the clinical significance of lesions.31 Schizas et al.,33 proposed a qualitative grading system for lumbar spinal stenosis based on dural sac morphology rather than dural cross-sectional area; this system was found to be more reliable and valid for the assessment of a patient's symptoms, and more reliable from the perspective of clinician decision-making.34 Pfirrmann et al.,30 graded LDHs based on nerve root morphology; this system was reliable and consistent with surgical findings. Beattie et al.,28 reported a strong association between deformation of the dural sac and symptoms in LDH patients. For similar reasons, we defined a massive LDH based on observations of the dural sac morphology rather than quantifying or estimating the size of herniation.

The present study involved 33 patients with massive LDH with complete dural sac stenosis among 882 patients in our LDH database (3.7%). The reliability of assessments based on dural sac morphology is reportedly moderate to substantial.3233 In this study, the reliability of dural sac morphology was found to be substantial (kappa value was 0.78).

The typical presenting symptom of LDH is acute or chronic intermittent lower back pain with associated leg pain. It is widely accepted that both of these pains resolve faster and more consistently when LDHs are large.710111335 However, the relationship between LDH size and clinical symptom severity has not been fully elucidated.28 In this study, the mean VAS scores of back and leg pain at presentation were 75.3 ± 19.1 and 80.2 ± 14.6, respectively, which are higher than those previously reported values of general LDH.3536373839 This difference suggests that massive LDHs with complete dural sac stenosis cause more severe back pain and leg pain at presentation, which is probably because larger LDHs produce higher levels of proinflammatory mediators and cytokines.40 Neurological dysfunction was observed in 42.4% (n = 14) of our study subjects. However, bowel/bladder symptoms appeared in only 2 patients (6.1%). Furthermore, our results demonstrate that neurological function was not severely affected by complete obliteration of the dural sac, which is consistent with previous reports on massive LDH.1415 These findings indicate that operative treatment can be deferred for patients with massive LDH with complete dural sac stenosis, unless significant neurological symptoms are observed.

Cribb et al.,14 reported 15 massive LDH patients (disc material occupying >50% of the spinal canal) with nonoperative treatment. Repeat MR scanning after a mean 2 years showed a resolution of the herniation in 14 patients. No CES was developed in their series. Benson et al.,15 reported 37 patients with conservatively treated massive LDH (disc material occupying >50% of the spinal canal). Eighty-three percent had a complete and sustained recovery after 2-year followup. Only four patients required a discectomy. The average Oswestry disability index improved from 58% to 15%. In this study, conservative treatment yielded a significant improvement in 63.6% of the study population. In particular, back and leg pain significantly improved within 1 month. Neurological deterioration was not observed in any patient.

Several limitations of this study warrant consideration. The first concerns its retrospective nature. Data associated with the use of medical records, miscoding, and a lack of clinical information can adversely affect results. The second limitation is that we did not compare the demographics, clinical characteristics, and followup results in our cohort with a moderately sized LDH. For this reason, we were unable to determine the clinical characteristics and the effects of nonoperative treatments. Moreover, other radiological parameters such as LDH size, canal area, modic changes, ligament flavum hypertrophy, facet arthritis, and dynamic instability were not analyzed. Additional studies regarding the clinical relevance, treatment outcomes, and subgroup analysis are warranted.

To conclude, although a massive LDH with complete dural sac stenosis is initially associated with severe back and leg pain, nonsurgical treatment yields a favorable outcome. Surgical treatment can be deferred unless significant neurological symptoms occur.