| Popis: |
BACKGROUND CONTEXT Corticoids have potent anti-inflammatory effects, which may help in relieving pain and dysfunction associated with lumbar stenosis. We assessed the effectiveness of a decreasing-dose regimen of oral corticoids in the treatment of lumbar canal stenosis in a prospective, double-blind, randomized, placebo-controlled trial. PURPOSE Because spinal stenosis most frequently affects the elderly, a patient group with a high surgical complication rate, an oral regimen can dramatically improve overall treatment safety. To this end, we assessed the efficacy of oral corticoids for the treatment of lumbar canal stenosis. STUDY DESIGN/SETTING This prospective, double-blind, randomized controlled study was approved by the institutional research ethics committee. Sixty-one patients were electronically randomized to the drug treatment or placebo control group. Opaque envelopes were used to ensure the secrecy of the allocation. The drug treatment group was administered corticoids at 1 mg/kg/day with a one-third dose reduction per week. The control group was administered placebo for the same period. All patients were assessed at four time points during the study: baseline (T0), at the end of treatment (week 3, T3), and at 6 and 12 weeks after study initiation (T6 and T12, respectively). All patients were permitted paracetamol (750-mg tablets) up to 3 times a day as a rescue analgesic. Total paracetamol intake and response were also assessed. The patient and the assessor were blinded to treatment. PATIENT SAMPLE All patients were assessed by the following instruments: the Short Form (SF)-36 Health Survey, the Roland–Morris Questionnaire, the 6-min walk test, a 10-cm visual analog scale (VAS) for pain, and a 5-point Likert-type scale (“much better”, “slightly better”, “unchanged”, “slightly worse”, and “much worse”). The Likert scale was not applied for the initial assessment (T0). OUTCOME MEASURES The sample size was chosen to yield a statistical power of 80% and a significance of 5% when comparing VAS scores between groups, assuming a group standard deviation of 2 cm and a minimum mean intergroup difference of 2 cm. We used Student's t-test to compare continuous variables with a homogenous distribution across groups. Generalized linear models (MLGs) and unbalanced 2-factor ANOVA were used to test the temporal effect of medication and interaction between the test period and medication. Tukey's test was used for 2 by 2 comparisons between time period and medication for the measurement points. Pearson's correlation analyses were performed to assess the relationships among clinical variables. In all statistical tests, P-values of METHODS Inclusion criteria were the presence of claudication within less than 100 m and the presence of at least two of the following lower limb symptoms: pain, weakness, burning, tingling (associated with or independent of lower back pain), and a vertebral canal area of 10°, degenerative pathologies in the hip or knee that can interfere with gait, and a history of total or partial arthroplasty in the hip and/or knee joint.The canal area was calculated on the basis of maximum anteroposterior (b) and mediolateral {1.5 [EN] (a) diameters measured by the computer program Osirix (2011) {3.4 [EN] using magnetic resonance imaging (MRI). After calculating these values, they were individually divided by 2 and multiplied by π (PI), with π = 3.14:Area = (a/2) × (b/2) × π × constant, where the constant is 0.8 when the canal is circular, 0.7 when the canal is elliptical, 0.6 in the presence of facetary compression, and 0.5 when the compression is caused by the disc and facets. RESULTS All 61 eligible patients, including 31 in the corticoid group and 30 in the placebo group, completed the study. There were no group differences in female: male ratio and there were no significant differences in height and weight between groups for the males and females. There was no significant difference in the use of paracetamol as a rescue analgesic between groups over the 21-day drug/placebo administration period. The Roland–Morris questionnaire was administered at baseline (T0), after the 3-week drug/placebo trial (T3), and at 6 and 12 weeks after study initiation (T6, T12). The scores suggested a slight improvement in the corticoid group at the beginning of treatment; however, they were not significantly different from those for the placebo group. In fact, no significant differences in total scores were observed within or between groups for any individual assessment period. The VAS scores for pain suggested a mild improvement in both groups at T3, as indicated by the lower scores compared with those at baseline (T0); however, these scores increased thereafter, with no significant change across test periods in either group. The 6-min walk test, performed according to the American Thoracic Society standards on a 22-m track with patients walking as fast as possible, also showed no significant improvement in both groups. In fact, the total distance travelled by the corticoid group decreased by approximately 40–50 m between T3 and T6 and T12 (Table 2). Because this assessment depends on muscular structure, the male and female subgroups were separately compared; this analysis also indicated no benefit of corticoids (not shown). The SF-36 questionnaire assesses current health conditions, with higher values corresponding to a better condition. The SF-36 scores for our patients suggested modest differences in some of the eight domains during the study period; however, the final values at T6 and T12 were similar between groups. Separate comparisons by gender also did not reveal significant differences in condition between groups (not shown). The Likert-type scale, wherein patients were instructed to document how they were feeling after treatment (much better, slightly better, unchanged, slightly worse, or much worse) also indicated no significant benefits of corticoids (not shown). Because the results were very similar, we compared the results of the different assessment instruments. There were no significant differences among instruments, which confirmed the results of the study and the reliability of the instruments in question. The body mass index (BMI) of patients correlated with the Roland–Morris Questionnaire findings; in other words, patients with a greater body weight, regardless of the group, gave more unfavorable answers. This indicated that obese patients with lumbar canal stenosis are more symptomatic. CONCLUSIONS This placebo-controlled study indicates that a tapering regimen of oral corticoids, starting at 1 mg/kg daily, is not effective in the treatment of lumbar canal stenosis. We found no direct correlation between the degree of stenosis on MRI and symptom severity or quality of life as revealed by the Roland–Morris Questionnaire and SF-36. Lumbar stenosis at L4/L5 is more symptomatic compared with stenoses at the other evaluated levels. BMI was directly associated with the degree of limitation and inversely associated with quality of life. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs. |
Full Text from ScienceDirect