The results of this study suggest that the combination of DM and NM may be useful for patients with LSS. Interpretation of the results must be made with caution, however, as the absence of randomization and control does not eliminate the possibility of treatment bias and does not allow one to distinguish these outcomes from those that would result from natural history. However, in a natural history study by Johnsson [32
], 70% of patients were found to be unchanged over 49 months, and 15% improved. So it would appear from this that, while deterioration is not the norm in patients with SS, most patients would not be expected to improve over time.
Nonetheless, clinically meaningful improvement in disability was seen in over two-thirds of the patients, and the improvement appeared to be maintained over an average of 16.5 months after cessation of treatment. The sustained improvement over the long term may relate to the emphasis in the management strategy on treatments that are designed to specifically address the known pathophysiology of LSS and on exercise that was designed to compliment the DM and NM, with continuous monitoring of compliance with home exercise throughout the treatment process. However, this study's design does not allow for firm conclusions to be drawn regarding this. Only two patients went on to require surgery, suggesting that the treatment approach studied here may be an effective alternative to surgery for patients with LSS.
This study can be examined in the light of other studies that have looked at the effectiveness of non-surgical management of LSS, some of which have compared it to surgery. In the Maine Lumbar Spine Study [33
], Atlas, et al compared non-surgical management, which was not strictly defined, with surgery in 141 patients with LSS. They found that at one year FU, the non-surgically treated group had only improved by an average of 1.6 points on the RM, compared to the surgical group, which improved by an average of 8.4 points. At four years, the non-surgical group improved an average of 3.5 points on the RM, while the surgical group had improved by an average of 8.5 points [34
]. This is contrasted with the 5.2 point improvement seen in the present study as a result of the non-surgical approach taken here. Simotas, et al [35
] followed 49 patients treated with a non-surgical approach that included oral non-steroidal or steroidal medication and epidural steroid injection along with postural instruction and mobilization and stabilization exercises. While RM data was collected at baseline, no FU data on the RM are provided. As such, it is difficult to directly compared the Simotas, et al study with the present one. However, they found that nine out of 49 patients (18.5%) required surgery, compared to two out of 44 patients (4.5%) in the present study.
From this, it is reasonable to conclude that the treatment approach in this study is a viable alternative to surgery for patients with LSS, and compares favorably with other non-surgical approaches that have been studied. As the efficacy of surgery does not appear to decrease if it is delayed in favor of a non-surgical trial [36
], most patients with LSS should be treated non-surgically for a period of time before considering operation. DM and NM may be one non-surgical option that can be offered to patients.
The management strategy in this study focused on two treatment modalities, DM and NM. DM may have benefits that relate to biomechanical effects on the stenotic segment, such as reduction of intradiscal pressure [20
] or widening of the space in the region of the nerve root [21
]. DM may also have neurophysiological effects that may be helpful in patients with LSS, such as facilitation of afferent input from mechanoreceptors [37
], possibly helping to improve proprioception, which has been shown to be impaired in patients with LSS [38
], or hypoalgesia [39
]. While most of the studies on the neurophysiological effects of spinal manipulation do not specifically assess DM, evidence suggests that DM has similar neurophysiological effects as other forms of manipulation [40
NM is hypothesized to gently move both the anatomical structures proximate to the neural elements that are being compromised, as well as the neural elements themselves [22
]. This may help patients with LSS by releasing perineural adhesions, thus decreasing traction strain on the nerve root, especially with walking. A great deal more basic science research into the effects of both DM and NM is needed to investigate these proposed mechanisms.
No major complications were seen in any patient, and transient, mild increase in symptoms was seen in 12 patients (21.8%). This is actually less than the 34–55% rate of transient pain related to manipulative treatment in general that has been reported in the literature [41
]. However, rare complications may not be detected in a sample size such as this one, so larger samples will be required to further investigate the safety of this approach to patients with LSS.
These patients were treated an average of 13.3 times. This study does not allow one to draw conclusions about the optimum number of treatments for patients with LSS. However, it can reasonably be said that the 13.3 treatments likely reflects an adequate number of treatments. This may help the treating clinician in decision making regarding how long to continue to treat a patient with LSS using this approach. It would appear that a decrease of three points on the RM instrument is a good indicator of clinically meaningful improvement in patients with LSS [28
]. Thus, if, after 13 or so treatments, a minimum improvement of three points on the RM questionnaire is not seen, further treatment with DM and NM is not warranted. It must be noted that a relatively wide range of treatment visits (2–50) was seen in this study, suggesting that individual differences in patient responses to treatment exists which necessitates greater or fewer than the mean number of treatments Nonetheless, 49 of 54 patients (91%) required between seven and 18 visits. So, from the data presented here, it can be concluded that the number of treatments required should be within six visits of the mean in the majority of patients.
This study is useful in that it assesses the outcome of a treatment approach in a "real world" environment. That is, the patients in this practice-based study were treated as they would be under normal circumstances according to the protocol for patients with LSS utilized at the Rhode Island Spine Center. There was no attempt to control the number of visits or the types of treatments that were provided in addition to DM and NM. As such, however, there is no way to tell the extent to which any particular treatment contributed to the outcome in each case. Also, interpretation of the results with regard to efficacy is not possible because of the absence of randomization and appropriate control group(s). The data presented here do suggest, however that the combination of DM and NM may be a useful approach for patients with LSS, and that further investigation in the form of randomized, controlled trials is warranted.
Limitations of this study include, as was stated earlier, absence of randomization and appropriate control groups. But this study was designed as an observational study to undertake preliminary investigation of the use of DM and NM in the management of patients with LSS. Proper randomized, controlled trials are feasible in this area. Also, twelve patients from this series did not have imaging confirmation of the presence of LSS and the diagnosis was made based on the presence of low back pain and/or leg pain in an older person with a clear history consistent with neurogenic claudication. It is felt that including these patients in the study is appropriate in that neurogenic claudication is so characteristic of LSS [5
], that it is very unlikely that any other condition would be causing the pain in these patients. Finally, the average pain intensity from baseline to the end of treatment was only 1.6 points. A change in 2 points on the NRS is generally considered to be the threshold for clinically meaningful improvement [31
]. However, in spite of this, a clinically meaningful change in disability was seen from baseline to the end of treatment. Also, there was a mean 3 point change in average pain intensity from baseline to final FU. This change can be considered to be clinically meaningful.