Murphy DR, Hurwitz EL, McGovern EE.
Rhode Island Spine Center,
Pawtucket, RI 02860, USA.
OBJECTIVE: This study presents the outcomes of patients with lumbar radiculopathy secondary to disk herniation treated after a diagnosis-based clinical decision rule.
METHODS: A prospective observational cohort study was conducted at a multidisciplinary, integrated clinic that includes chiropractic and physical therapy health care services. Data on 49 consecutive patients were collected at baseline, at the end of conservative, nonsurgical treatment and a mean of 14.5 months after cessation of treatment. Disability was measured using the Bournemouth Disability Questionnaire (BDQ) and pain using the Numerical Rating Scale for pain. Fear beliefs were measured with the Fear-Avoidance Beliefs Questionnaire (FABQ). Patients also self-rated improvement.
RESULTS: Mean duration of complaint was 60.5 weeks. Mean self-rated improvement at the end of treatment was 77.5%. Improvement was described as “good” or “excellent” in nearly 90% of patients. Mean percentage improvement on the BDQ was 60.4%. Numerical Rating Scale improved 4.1 points and FABQ improved 4.8 points. Clinically meaningful improvements in pain and disability were seen in 79% and 70% of patients, respectively. Mean number of visits was 13.2. After an average long-term follow-up of 14.5 months, mean self-rated improvement was 81.1%. “Good” or “excellent” improvement was reported by 80% of patients. Mean percentage improvement in BDQ was 67.4%. Numerical Rating Scale improved 4.2 points and FABQ 4.5 points. Clinically meaningful improvements in pain and disability were seen in 79% and 73% of patients, respectively.
CONCLUSIONS: Management based on the decision rule yielded favorable outcomes in this cohort study. Improvement appeared to be maintained over the long term.
From the FULL TEXT Article
Only a small percentage of patients with LRSHD ever require surgery.  Thus, it is essential that optimum nonsurgical approaches are developed, which bring about improvement in pain and disability as quickly as possible, reduce the likelihood of future problems, and minimize the need for surgical intervention. However, although a wide variety of nonsurgical treatments have been recommended for patients with LRSHD,  no individual treatment has been found to be most effective. Perhaps the largest trial of management of LRSHD is that of Weinstein et al  in which 501 patients were randomized to receive either surgery or nonsurgical intervention. Similar substantial improvements in pain and disability were found in both the surgical and nonsurgical groups. A high crossover rate in both groups limits interpretation of these data. In addition, the nonsurgical management was poorly described. Nonetheless, the study likely underestimated the effectiveness of nonsurgical management because patients were only included if they had already failed at least 6 weeks of initial nonsurgical care. Thus, patients who recovered with 6 weeks of nonsurgical management were not included in the study. An additional weakness of the Weinstein et al  study is that the nonsurgical management was not defined, because the treatment approaches taken were left up to the individual clinics involved in the study. It would be interesting to compare outcomes of surgical intervention with those of a well-defined nonsurgical approach such as what is reported here.
The current study supports the notion that nonsurgical management according to the DBCDR is a viable option for patients with LRSHD. Nearly 90% of patents reported their outcome to be either “excellent” or “good.” Clinically meaningful improvement in disability was seen in more than 70% of patients, and clinically meaningful improvement in pain intensity was seen in 74% of patients. These improvements were maintained 14.5 months after cessation of treatment.
The fact that outcomes were as good or better at long-term follow-up is significant because it suggests that patients treated according to the DBCDR generally do not need ongoing “maintenance” or “supportive” care to maintain functional improvement. This may be due to the emphasis on education regarding beliefs, attitudes, and cognitions about spinal pain or the emphasis on the importance of continued exercise. The improvements in FABQ score at the end of treatment and long-term follow-up support the former; however, because compliance to home exercise was not measured, the latter can only be speculated upon.
When using the DBCDR, the clinician uses 3 questions of diagnosis. The first question relates to whether the patient has any symptoms or signs that may reflect a visceral problem or a serious or potentially life-threatening illness. The second question relates to where the pain is arising from. The third question then allows for investigation of factors that may be serving to perpetuate the pain experience. From this, individual treatment decisions are made based on the most important features in each case.
In this study, all patients had one thing in common with regard to question number 2—LRSHD. Because of this, nearly all patients were treated with distraction manipulation, which has been shown to reduce intradiscal pressure. [ 14, 15] Neural mobilization was applied in all patients because this method attempts to mobilize the involved nerve root.  Eight patients were referred for ESI in an attempt to rapidly reduce nerve root pain. Although ESI generally brings about only temporary improvement,  it was felt in these patients that earlier commencement of more active treatments could begin if quick pain relief could be brought about with ESI. None of these 8 patients were treated solely with ESI. Because individual patients in the cohort may have had various other clinical features that were deemed relevant to each individual case, treatment approaches beyond those aimed at the disk or nerve root varied somewhat. These were based on specific features that were found in each individual patient.
In those patients whose symptoms centralized on examination, end-range loading maneuvers in the direction of centralization were provided.  Centralization of symptoms on end-range loading examination was found in 61.5% of patients. Two other studies assessed the presence of centralization signs in patients with confirmed LRSHD. Kopp et al  found that more than half (35 of 67) of patients with this condition centralized with end-range maneuvers. Alexander et al  found that approximately 42% (73 of 173) of LRSHD patients were centralizers. However, in both these studies, only extension was assessed and not flexion or lateral and rotational movements, as was the case in the present study. The protocol for end-range loading examination used in this study was more similar to that of Werneke and Hart, [44, 45] who found that 77% of acute LBP patients 44 and 46% of chronic LBP patients  centralized on end-range loading examination. However, these studies included general cohorts of acute or chronic LBP patients; the subject populations were not limited to patients with LRSHD. As has been found previously,  peripheralization of symptoms on end-range loading examination appeared to carry a negative prognosis in the study reported here, although the percentage of peripheralizers (7%) was too low to allow for statistical analysis of this observation.
Joint manipulation was used in 6 patients, based on the presence of provocation of these LBP patients with segmental palpation and the absence of centralization on end-range loading examination. Several studies have found joint manipulation to be helpful in patients with LRSHD. [47, 48] However, these studies did not use segmental pain provocation signs in the absence of centralization as the primary indication for this treatment modality.
Stabilization exercise was provided in 40 patients. The need for this was based on 3 clinical tests—the hip extension test,  the segmental instability test,  and the Active Straight Leg Raise test.  The stabilization exercise approach was adapted from that of Richardson et al  and McGill.  A number of studies have evaluated the effectiveness of this approach,  at least one of which involved patients with LRSHD. 
In cases in which CPH, fear, or catastrophizing were found, education and graded exposure [55, 56] were the focus. Central to this process was education regarding the nature of CPH.  It was explained to these patients that the primary reason for their severe pain experience was that peripheral nociceptive signals were being amplified by the central nervous system before the arrival of these signals to the conscious aspect of the brain. Thus, fear of movement and catastrophizing, whereas understandable, were based on false, or exaggerated, information. Once the patient understood this, the graded exposure approach was begun. In the model of care evaluated in this study, if depression is suspected, this is monitored and the patient referred if poor response to treatment occurs, and it is deemed that the depression is relevant to this. As it turned out, this did not occur with any of the patients in this cohort.
The patients in this study improved with regard to pain and disability as well as with regard to fear beliefs. There was a statistically significant relationship between improvement in disability and improvement in fear beliefs. This is interesting in light of the fact that treatment was only provided by somatic-based clinicians—no professional psychologic intervention was provided. This suggests that, although psychologic processes such as fear and catastrophizing have been repeatedly found to play an important role in the perpetuation of spinal pain, it may be that somatic-based clinicians are often capable of managing this aspect of the clinical picture. Other studies have found that psychologic symptoms and signs improve with a purely somatic-based approach, especially when the approach involves self-care,  is focused on active exercise  and, particularly, if it is provided in a cognitive-behavioral context. [26,59-61] This does not preclude the possibility, however, that in some individual patients, the psychologic factors may be recalcitrant enough to require professional psychologic or psychiatric intervention.
It is interesting that improvements at the end of treatment appeared to be maintained at long-term follow-up. This may result from the emphasis of the approach on continuation of exercises and teaching self-management of acute episodes. However, compliance with exercise and self-care recommendations was not measured. Nonetheless, it does suggest that ongoing “maintenance” care is not necessary when treatment according to the DBCDR is used.
Seven patients (11.7%) reported increased pain related to the study treatments. This percentage is substantially less than the 30% to 34% that has been found in other studies of adverse reactions to treatment that involved manipulation. [62-64] Also, one study of short-term adverse reactions to manipulation found that patients with these reactions were less likely to experience a positive clinical outcome and were less satisfied with treatment.  In the study reported here, in those in whom follow-up data were available, the increase in pain did not appear to adversely affect their outcome, even with regard to self-rated improvement. This may be reflective of the emphasis on the use of distraction manipulation in this study or the focus on exercise designed to rapidly centralize and resolve the pain. However, the studies are not comparable enough to draw firm conclusions.
Patients were seen for a mean of 12.6 visits. This included examinations, reexaminations, treatment sessions, and exercise sessions. This number is consistent with other cohort studies of patients with radiculopathy treated according to the DBCDR. [7, 8] It is likely that this represents the number of visits that are typically necessary to manage patients with LRSHD; however, this study design does not allow firm conclusions to be drawn.
There are several important limitations to this study. One is the relatively small sample size, as well as the fact that complete follow-up data were not available on all patients. However, because no clinically meaningful differences were found between the group of patients in whom complete data were available and those in whom follow-up data were not available, it is not expected that the outcome would be different between these groups. The treatment occurred at a single practice setting. Therefore, it is not clear whether the findings are generalizable. Also, because this was a pragmatic study, that is, the patients were treated with a multimodal approach as would occur in usual clinical circumstances, it is impossible to determine the extent to which any individual treatment impacted outcome.
The pragmatic nature of the study is also one of its strengths. Because patients were treated as they would be during the usual course of clinical practice, as opposed to being treated in an experimental setting, the study reported “real world” outcomes, at least with regard to the “real world” environment in which this study took place. However, the findings cannot be generalized to clinical environments in which treatment protocols are used that do not strictly follow the DBCDR used in this study.
Also, the data were collected prospectively to avoid recall bias, and long-term follow-up was used. Strict diagnostic criteria were used to determine inclusion in the study. The inclusion of fear beliefs and complications provides useful clinical information that broadens the benefit of the information to nonsurgical clinicians who treat LRSHD. Although the observational design does not allow firm conclusions to be drawn with regard to efficacy, this design does allow conclusions to be drawn regarding safety.65 However, because of this study’s sample size, rare complications are not likely to be detected. Finally, the DBCDR upon which the management of these patients was based can be applied by any appropriately trained practitioner, increasing the generalizability of the findings.
Our findings suggest that patients with LRSHD who are treated according to a strict DBCDR tend to have favorable outcome to treatment. This favorable outcome appears to be maintained over the long term. Fear beliefs also appear to improve with the approach, and a significant relationship between improvement in disability and improvement in fear beliefs was found. The absence of a control group does not allow firm conclusions to be drawn, but further research in the form of large cohort studies and randomized, controlled trials would be beneficial in determining the efficacy of this treatment approach in patients with LRSHD. The treatments used in the study appear to be safe in this patient population.
Funding Sources and Potential Conflicts of Interest
No funding sources or conflicts of interest were reported for this study.