Abstract

Background. In a previous prospective study on pulsed radiofrequency (PRF) treatment adjacent to the lumbar dorsal root ganglion (DRG) for patients with chronic lumbosacral radicular pain, we reported success in 55.4% of the patients at 6 months. Identification of predictors for success after PRF may improve outcome. We assessed the predictors of PRF in patients with chronic intractable lumbosacral radicular pain.

Methods. Patients with monosegmental chronic lumbosacral radicular pain of L5 or S1 first received a selective nerve root block at the corresponding level. Independent of the result of this block a PRF treatment at the same level was performed. At 6 weeks, 3 months, and 6 months after the procedure the outcome was evaluated.

Results. A positive diagnostic nerve root block and age ≥ 55 were predictive factors for successful outcome at 6 months, while disability was a negative predictor.

The use of failed back surgery syndrome, gender, duration of pain, Numerical Rating Scale, level and side of treatment, DN4, and RAND-36 as predictors for success was not supported.

Conclusions. Successful outcome after PRF adjacent to the DRG, in patients with intractable chronic lumbosacral radicular pain, is more likely in patients ≥ 55 years, with limited disability and after a positive diagnostic nerve root block. A combination of all these factors creates a fair predictive value (AUC: 0.73).

Introduction

Lumbosacral radicular pain is an important health care problem. Patients with this type of pain have the lowest quality of life compared to other types of neuropathic pain or compared to other major chronic diseases [ 1 ]. It is the most commonly occurring form of neuropathic pain [ 2 , 3 ], thereby creating an important socioeconomic problem. Although 75% of patients have a spontaneous recovery from symptoms within 3 months [ 4–6 ], if pain persists after 3 months of onset, the prospects for long-term recovery are unfavorable [ 7 , 8 ]. Two years after onset of lumbosacral radicular pain 25% of patients are still unable to return to work [ 9 ]. This high prevalence of chronic severe lumbosacral radicular pain may be attributed to two main factors. First, the recurrence rate after an initial recovery within 3 months of onset is high [ 7 ]. Second, treatment frequently fails, as illustrated by the fact that at 1 and 2 years follow-up respectively 45% and 40% of all patients referred for secondary care to a spine clinic who were treated conservatively or were operated on report non-successful outcome [ 8 ].

Pulsed radiofrequency (PRF) treatment uses intermittent high-frequency current, and its use adjacent to the dorsal root ganglion (DRG) has been suggested for the treatment of radicular pain [ 10–16 ]. This target-specific treatment requires accurate identification of the causative level. Therefore, a diagnostic block of the presumed causative DRG has been suggested to determine the exact level of PRF treatment.

In a retrospective study, at least 50% pain relief after PRF of the lumbar DRG was achieved in 41% of patients at 6 months [ 10 ]. In patients with failed back surgery syndrome (FBSS), the use of PRF resulted in no improvement of pain. We conducted an outcome study to measure the effect of PRF, a treatment option that can be considered in patients suffering chronic lumbosacral radicular pain refractory to conservative treatment. A good clinical outcome, defined as a reduction in pain intensity (Numeric Rating Scale – NRS) of at least 2 points or a global perceived effect reporting “full recovery” or “much improvement” at 6 months was reported by 55.4% of patients. To optimize treatment outcome by improving patient selection, there is a need to identify those factors that may be predictive for the outcome of a PRF treatment. As part of the prospective outcome study, we assessed the non-psychological parameters (age, gender, NRS, duration of pain, response to diagnostic block, level and side of treatment, failed back surgery, DN4 and RAND-36, disability) that could predict a successful outcome of PRF adjacent to the DRG in patients with chronic intractable lumbosacral radicular pain.

Methods

Participants

Between January 2010 and September 2012, 461 consecutive patients with lumbosacral radicular pain were screened in the multidisciplinary pain centers of the University Hospital Maastricht, Maastricht, The Netherlands; Sint-Jozefkliniek, Bornem, Belgium; Ziekenhuis Oost-Limburg, Genk, Belgium; and Klinieken Noord-Antwerpen, Brasschaat, Belgium. Most of the patients were referred by medical specialists (neurologists, neurosurgeons, orthopedic surgeons, and rehabilitation physicians).

The institutional ethics review board of each of the participating centers approved the trial, and all patients signed an informed consent form. The trial was registered as NCT00991237 (ClinicalTrials.gov).

Patients were eligible for the study if they reported a unilateral and monosegmental radiating pain down the leg, into the foot to the hallux (L5) or lateral side of the foot (S1), suggesting involvement of the lumbosacral spinal nerve along the affected nerve root concordant with the MRI or CT-scan findings [ 17–20 ]. The radicular syndrome should have been present for at least 3 months, despite conservative treatment consisting of medication and physical therapy [ 4 , 21 ]. Leg pain should be the primary complaint, with an average pain score of at least 5 on the NRS (11 point scale, where 0 = no pain and 10 = the worst pain imaginable).

Exclusion criteria were as follows: younger than 18 and older than 80 years old, an atypical radiation pattern, bilateral symptoms, or involvement of more than one segment (L5 and S1 radiation). A history of cancer, fractures of the lumbar vertebrae, myelopathy, systemic diseases or connective tissue diseases, diabetes mellitus type I, coagulation disorders and the use of anticoagulants, multiple sclerosis, pregnancy, the presence of a cardiac pacemaker or spinal cord stimulator, and RF or PRF treatment of the DRG in the last year. Patients with a score higher than 45 on the Pain Catastrophizing Scale (PCS) were first referred to a psychologist for further evaluation and management [ 22 ].

Patients first received a nerve root block at L5 or S1, depending on the typical radiation pattern, consisting of lidocaine 2% 0.7 mL. Before performing the diagnostic block, patients were asked to indicate the pain intensity in the leg on a NRS 11 point scale. Thirty minutes after the diagnostic block the patients were asked to rate again the pain intensity on the NRS. A reduction on the NRS of at least two points was considered a positive result. The patients’ global perceived effect (GPE) was also assessed using a 7-point Likert scale [ 23 ]. Patients reporting “full recovery” or “much improvement” were considered to have a positive result.

To allow the assessment of the predictive value, all patients, independently of the result of this block, received a PRF treatment at the same level concordant with clinical symptoms and adjacent to the L5 or S1 DRG within a week. At 6 weeks, 3 months, and 6 months after the procedure the outcome was evaluated. If patients at the 6 week evaluation reported clinically insufficient pain relief, defined as less than 50% pain relief on the GPE or less than a 2 point reduction on the NRS, an additional single repeat PRF procedure was performed at the same level when the radiating pattern was the same as during the initial consultation or an adjacent level when the radiating pattern changed from L5 to S1 or vice versa. Patients who received a second intervention might have experienced treatment success at 6 months. Conventional medical management (e.g., acetaminophen, tramadol, amitryptiline, gabapentin, or pregabalin) was further allowed if pain was not adequately controlled.

Intervention Technique

Diagnostic Block

For the diagnostic block, the C-arm is adjusted in such a way that the X-rays run parallel to the end plates of the relevant level. Thereafter, the C-arm is rotated until the spinous process projects over the contralateral facet column. With the C-arm in this projection, the insertion point is found by projecting a metal ruler over the lateral part of the neuroforamen. This lateral approach reduces excessive overflow to adjacent levels. A 10 cm long, 22 gauge needle is inserted here locally in the direction of the rays. Thereafter, the direction is corrected such that the needle is projected as a point on the screen. The direction of the radiation beam is now modified to a profile (lateral) view, and the needle is inserted until the point is located in the craniodorsal part of the neuroforamen. In an anteroposterior (AP) view, the course of a small amount of contrast agent is followed with so-called real-time imaging; it spreads out laterocaudally along the spinal nerve. If a correct image is not obtained, then the needle is carefully advanced a little further. Finally, a small amount of contrast agent shows the course of the spinal nerve in an AP view. Lidocaine 2% 0.7 mL was then injected (0.2 mL stayed in the needle, 0.5 mL was effectively injected).

Lumbar Percutaneous PRF

The PRF treatment is described elsewhere [ 15 ]. The insertion point for PRF treatment is determined in the same way as for the diagnostic block. After insertion of the cannula (22 gauge 100 mm probe with a 5 mm active tip, SMK, NeuroTherm Inc., Wilmington, MA, USA) in the middle on the neuroforamen in the lateral view. The stylet is removed and exchanged for the RF probe. The probe is connected to the radiofrequency generator (NeuroTherm NT1100). The impedance and then the sensory threshold are checked. The patient should now feel tingling at a voltage of <0.5 V. If the threshold is above 0.5V, the cannula is carefully advanced further until a correct sensory stimulation is found. Thereafter, a pulsed current (20 ms current, 2 Hz) is applied two times for 120 seconds each time with an output of 45 V, with a short interval between both treatments. During this procedure, the temperature at the tip of the electrode may not surpass 42°C.

Outcome Measurements

Patients’ baseline characteristics—age, gender, duration of pain, and prior back surgery—were noted. The affected side, level of pathology, pain intensity in the leg (NRS) measured according to the Jensen method [ 24 ], quality of life (RAND-36) [ 25 ], and disability (Oswestry Disability Index [ODI]) [ 26 ] were noted at baseline. The neuropathic character of pain was judged with DN-4 [ 19 , 27 ] and the LANSS pain scale [ 28 ].

The use of medication at inclusion was scored with the Medication Quantification Scale (MQSIII) [ 29 ].

The primary aim of the analysis was to identify baseline variables that were indicators of success after PRF treatment.

Based on the RAND parameters a mental and a physical component scale were calculated [ 30 ]. Success or failure of the PRF treatment was evaluated at 6 weeks and 6 months. Success was defined as a reduction in the NRS score of at least two points or a GPE of 1 or 2 (fully recovered or much improvement). Drop-outs were classified as failures.

MRI performed in a single center (N = 20) was analyzed by an experienced neuroradiologist (JW) who was blinded to treatment side, level, and effect. All images were evaluated two times in two separate sessions to increase consistency. The following parameters were examined: pathology (nerve root compression, disc protrusion, disc extrusion, stenosis of lateral recess), side, and level of pathology. Afterwards, the correlation between the MRI findings and the level of symptoms (= treatment level/site) and success after PRF was assessed.

Statistics

Possible predictors were the result of the diagnostic block, FBSS, gender, age (dichotomized: <55 vs ≥55 years), duration of complaints, level of PRF, side of PRF, DN4 (dichotomized: <4 vs ≥4), ODI, MQSIII, Physical Component Scale and Mental Component Scale of RAND-36, and use of opioids. Using a univariate logistic regression model, the ORs and their confidence intervals were calculated. Those baseline characteristics were identified as predictors if the P value was 0.1 or less [ 31 ]. In a second step, all selected predictors were entered backward stepwise ( P  = 0.1) in a multivariate logistic regression model. Receiver operating characteristic (ROC) curves were constructed and the area under the curve (AUC) was calculated for the selected predictors and the outcome of the multivariate regression model. To determine the influence of a second PRF on the performance of the diagnostic block, we also calculated the diagnostic OR of the diagnostic block after one PRF treatment at 6 weeks. Data were analyzed using SPSS 20.0 (IBM SPSS statistics).

Results

Out of a total of 461 patients screened with presumed chronic lumbosacral radicular pain, 65 were included. Seven patients had to be referred for back surgery during the study period, and four received another treatment such as surgery for another problem during the trial period and were therefore excluded. Two patients were lost to follow-up at the 6 month evaluation. They were all considered failures. Table 1 presents the baseline characteristics. In this group none of the patients had a score of more than 45 on the PCS.

Table 1

Baseline patient characteristics

Patient characteristic 100% (N = 65) mean (SD) 
Age ≥ 55 43.1 (28) 
Gender  
Male 27.7 (18) 
Female 72.3 (47) 
Duration of pain  
3 to 6 months 27.7 (18) 
6 to 12 months 29.2 (19) 
>1 year 43.1 (28) 
Dermatome  
L5 60.0 (39) 
S1 40.0 (26) 
Side  
Left 55.4 (36) 
Right 45.0 (29) 
Failed back surgery 23.1 (15) 
Age 51.80 (14.3) 
NRS 7.06 (1.3) 
Physical component 33.40 (6.7) 
Mental component 45.00 (12.2) 
ODI 43.11 (14.0) 
LANS 8.80 (5.0) 
DN4 4.50 (2.0) 
MQSIII 20.50 (22.0) 
Patient characteristic 100% (N = 65) mean (SD) 
Age ≥ 55 43.1 (28) 
Gender  
Male 27.7 (18) 
Female 72.3 (47) 
Duration of pain  
3 to 6 months 27.7 (18) 
6 to 12 months 29.2 (19) 
>1 year 43.1 (28) 
Dermatome  
L5 60.0 (39) 
S1 40.0 (26) 
Side  
Left 55.4 (36) 
Right 45.0 (29) 
Failed back surgery 23.1 (15) 
Age 51.80 (14.3) 
NRS 7.06 (1.3) 
Physical component 33.40 (6.7) 
Mental component 45.00 (12.2) 
ODI 43.11 (14.0) 
LANS 8.80 (5.0) 
DN4 4.50 (2.0) 
MQSIII 20.50 (22.0) 

Twenty-four subjects did not have adequate pain relief at 6 weeks, and 5 of them received a second PRF [ 32 ].

The univariate analysis selected the results of the diagnostic block, age, MQSIII, and ODI as predictors. The stepwise multivariate analysis with these predictors selected the diagnostic block, age, and ODI as final predictors for a successful outcome at 6 months. The combination of all these final predictors gives an OR of 83.2 ( P  = 0.02, 95% CI 4.76 to 1454) for a successful outcome at 6 months. In Table 2 the diagnostic OR and the OR of the stepwise multivariate regression analysis with their P values and CIs are described.

Table 2

The diagnostic odds ratios (ORs) of all predictors as the result of the univariate regression analysis and the OR of the result of the stepwise multivariate regression analysis

 OR P 95% CI 
   Lower to Upper 
Univariate model    
Gender 0.52 0.26 0.17 to 1.62 
Duration of symptoms (months) 0.72 0.29 0.40 to 1.32 
Level diagnostic block 0.61 0.33 0.22 to 1.64 
Level PRF 0.70 0.48 0.26 to 1.89 
Side 0.94 0.90 0.35 to 2.54 
FBSS 0.90 0.86 0.28 to 2.86 
NRS 0.82 0.32 0.56 to 1.21 
DN4 0.91 0.44 0.71 to 1.16 
DN4 ≥ 4 1.14 0.81 0.38 to 3.47 
LANSS 1.30 0.62 0.45 to 3.82 
MQSIII 0.97 0.04 0.96 to 1.00 
Opioid 0.48 0.29 0.12 to 1.89 
Physical CS (RAND-36) 1.00 0.91 0.92 to 1.08 
Mental CS (RAND-36) 1.01 0.77 0.96 to 1.05 
Age ≥55 3.28 0.03 1.15 to 9.34 
Diagnostic block 3.26 0.06 0.97 to 11.00 
ODI −0.97 0.08 0.93 to 1.00 
Multivariate model    
Predicted probability based on 83.2 0.02 4.76 to 1454 
Age ≥55 3.78 0.02 1.22 to 11.63 
Diagnostic block 3.51 0.07 0.92 to 13.48 
ODI −0.96 0.07 0.92 to 1.00 
 OR P 95% CI 
   Lower to Upper 
Univariate model    
Gender 0.52 0.26 0.17 to 1.62 
Duration of symptoms (months) 0.72 0.29 0.40 to 1.32 
Level diagnostic block 0.61 0.33 0.22 to 1.64 
Level PRF 0.70 0.48 0.26 to 1.89 
Side 0.94 0.90 0.35 to 2.54 
FBSS 0.90 0.86 0.28 to 2.86 
NRS 0.82 0.32 0.56 to 1.21 
DN4 0.91 0.44 0.71 to 1.16 
DN4 ≥ 4 1.14 0.81 0.38 to 3.47 
LANSS 1.30 0.62 0.45 to 3.82 
MQSIII 0.97 0.04 0.96 to 1.00 
Opioid 0.48 0.29 0.12 to 1.89 
Physical CS (RAND-36) 1.00 0.91 0.92 to 1.08 
Mental CS (RAND-36) 1.01 0.77 0.96 to 1.05 
Age ≥55 3.28 0.03 1.15 to 9.34 
Diagnostic block 3.26 0.06 0.97 to 11.00 
ODI −0.97 0.08 0.93 to 1.00 
Multivariate model    
Predicted probability based on 83.2 0.02 4.76 to 1454 
Age ≥55 3.78 0.02 1.22 to 11.63 
Diagnostic block 3.51 0.07 0.92 to 13.48 
ODI −0.96 0.07 0.92 to 1.00 

In Figure 1 the ROC curves of these predictors are shown and the AUC is calculated. One possible cut-off would lead to a sensitivity of all predictors of 69% with a specificity of 59%. The OR of a positive diagnostic block in predicting the success at 6 weeks after one PRF treatment is 3.2 ( P  = 0.072, CI 0.9 to 11.375).

Figure 1

Receiver Operator Characteristic curves. AUC = Aera under the curve; MQSIII = Medication Quantification Scale; ODI = Oswestry Disability Index.

Figure 1

Receiver Operator Characteristic curves. AUC = Aera under the curve; MQSIII = Medication Quantification Scale; ODI = Oswestry Disability Index.

The MRI analysis of a subgroup of 20 consecutive patients showed for the S1 nerve root no correlation between the level of pathology and the level of clinical symptoms (treatment level). For the L5 and S1 nerve root there was a correlation between the side of the pathology and the side of clinical symptoms (agreement in 10 of 11 treated cases, medial disc pathology excluded).

The side and level of nerve root compression on the MRI were not correlated with success after PRF.

Discussion

Principal Findings

A positive diagnostic block of the corresponding nerve root is a predictor for success at 6 weeks and 6 months after PRF treatment adjacent to the DRG in chronic intractable lumbosacral radicular pain. ORs for a diagnostic block at 6 weeks and 6 months are similar, indicating that the number of PRF treatments has no influence on the predictive value of the diagnostic block.

Older age (more than 55 years old) is also a predictor for clinical success at 6 months. A high disability acts as a negative predictor for a successful outcome. When calculating the AUC of these predictors, each of these parameters showed limited predictive value (AUC of positive diagnostic block, older age, and limited disability resp. 0.60, 0.64, 0.60). The combination of these predictors has a fair predictive value (AUC = 0.73, 95% CI: 0.61 to 0.85) for the 6 month outcome.

These results do not support the use of FBSS, gender, duration of pain, NRS, level and side of treatment, DN4, and RAND-36 as predictors of success.

Strength Study

This is the first prospective study to evaluate potential non-psychological factors that may predict the outcome of PRF adjacent to the lumbar DRG for the management of chronic intractable lumbosacral radicular pain. All patients were included according to strict selection criteria and treated regardless of the outcome of the selective diagnostic block, which allows for judging the value of the block as a predictive factor in a homogeneous patient population with dominant radicular symptoms in only one dermatome.

The standardized inclusion and management also makes it possible to assess other factors that may influence the treatment outcome.

Relation with Other Studies

Up till now only two studies have assessed the influence of prior surgery on the outcome of PRF adjacent to the lumbar DRG. A retrospective analysis found that patients with FBSS derived no beneficial effect from PRF treatment adjacent to the lumbar DRG [ 10 ]. This observation was not confirmed in a clinical audit [ 15 ] or in the present study.

How Accurate Is One Single Diagnostic Block?

There is a growing debate on the physiologic and anatomic selectivity of diagnostic “selective” nerve root blocks. According to the anatomy of the spinal nerve in the foramen, 7 µL of local anesthetic should suffice for a nociceptive blockade [ 33 ]. In practice, 0.5 to 1 mL is used, with a risk of spreading to other structures like the nervus sinuvertebralis (innervation of nearby disci intervertebralis, ligamentum longitudinale posterius, ventral dura mater, and nerve root sleeve) and sensory fibers of ramus dorsalis (local back muscles and facet joints) [ 34 ]. This interferes strongly with the anatomic selectivity of this diagnostic procedure. Furthermore, the position of the needle tip during this procedure is normally located distal to the inflammatory process in case of a herniated disc (e.g., diagnostic block of nerve root L5 for an L4-5 herniated disc). Peripheral nerve blocks have the potential to induce adequate pain reduction even when the etiology is located more proximally [ 35 , 36 ].

Ectopic firing and Wallerian degeneration potentiate afferent signaling in radicular pain [ 37 , 38 ], so it is possible that these distal secondary neurological events are interrupted by a diagnostic block [ 33 ]. The extent of the specificity of a nociceptive blockade of the nerve root distal to the etiology for the diagnosis of radicular pain is therefore unknown.

Limitations

The analysis of the predictive factors for clinical outcome was part of a prospective trial that also aimed at evaluating the effect of PRF treatment adjacent to the DRG. Since the evidence for PRF in radicular pain is limited and ideal parameters for treatment are unknown, the objective was to first evaluate the effect of PRF at a single level. Moreover, it is difficult to evaluate the diagnostic value of a single diagnostic nerve root block if more than one nerve root is affected. For this reason, patients with only one affected nerve root were selected. Since a large portion of the screened patients presented two or more affected nerve roots, this reduced significantly the inclusion rate. This is supported by electromyographical data where many patients have more than one affected nerve root [ 39 ], so it is unclear how the information from our study could be extrapolated to situations involving more than one nerve root. The final position of the needle tip for PRF is based on sensory testing, and a threshold below 0.5 V is used as a benchmark. Owing to the variability in the location of the DRG, the exact position of the needle tip relative to the DRG is, however, uncertain. How this can influence the clinical result is unclear.

Psychosocial factors were used as exclusion criteria but not as predictors for the outcome of PRF treatment adjacent to the DRG; this is another limitation of the present study.

From the wide confidence intervals in Table 2 it can be seen that a non-significant result does not mean that this variable has no predictive value, especially when the distribution of that potential predictor is unbalanced. This may hold for the variable of FBSS. This is in agreement with another study where FBSS showed no significant influence on the outcome [ 15 ].

Only 5 patients, out of the 24 who failed at 6 weeks, received a second PRF. The second PRF was offered to the patients, but a majority declined it. Often patients were satisfied with a minor reduction in pain, so no statement on the value of a second PRF can be made.

Unanswered Questions and Future Research

We noted that when the GPE was used to evaluate the diagnostic block, 14 patients with a negative result still had a good clinical outcome (false negative). This number of patients was reduced to three when the NRS was used as an evaluation measure. The combination of NRS and GPE resulted in five patients having a false negative block. This observation suggests that for future research the NRS, or the combination of NRS and GPE, should be used as an outcome parameter.

It is possible that the number of false negative results could be reduced further by increasing the sensitivity of the diagnostic blockade. Targeting only one level of the spinal nerve may generate false negative results of the diagnostic block. Because of the described divergence and convergence of afferent radicular pain signaling, different segmental levels are involved [ 40–43 ].

Therefore, the value of blocking more than one nerve root to more accurately predict the outcome after PRF treatment is worth exploring.

The correlation between imaging and clinical symptoms for radicular pain is not yet established [ 44–50 ]. The majority of patients included in this study presented disc protrusion or disc extrusion and, to a lesser extent, stenosis of the lateral recess. Owing to the strict inclusion criteria (average pain at least 5/10 for the past 4 days), we did not select patients with central canal stenosis since pain scores tend to drop during rest in this population. No correlation between the MRI findings and treatment effect could be established. The subgroup was too small to draw conclusions, but the observations seem to confirm the unclear relation between imaging and symptoms.

Conclusion

A successful outcome after a monosegmental treatment of single-level (L5 or S1) radicular pain with PRF adjacent to the DRG in patients with intractable chronic lumbosacral radicular pain is more likely in patients at least 55 years of age, whereas younger patients and those with a high disability probably have a lower success rate. The diagnostic nerve root block has a limited predictive value. The combination of all these factors creates a fair predictive value.

Acknowledgments

The authors would like to thank the cooperating physicians A. Balthasar MUMC, The Netherlands, K. Lauwers, (KL) KLINA Brasschaat, and P. Vanelderen ZOL, Belgium, for patient recruitment and referrals. The authors are grateful to Nicole Van den Hecke for administrative support.

Funding sources: The work presented here was performed in collaboration among all the authors, without any financial support.

Conflict of interest: None of the authors has a conflict of interest.

References

1
Doth
AH
Hansson
PT
Jensen
MP
Taylor
RS.
The burden of neuropathic pain: A systematic review and meta-analysis of health utilities
.
Pain
 
2010
;
149
338
44
.
2
Khoromi
S
Patsalides
A
Parada
S
, et al.   .
Topiramate in chronic lumbar radicular pain
.
J Pain
 
2005
;
6
:
829
36
.
3
Dworkin
RH
O’Connor
AB
Backonja
M
, et al.   .
Pharmacologic management of neuropathic pain: Evidence-based recommendations
.
Pain
 
2007
;
132
:
237
51
.
4
Weber
H
Holme
I
Amlie
E.
The natural course of acute sciatica with nerve root symptoms in a double-blind placebo-controlled trial evaluating the effect of piroxicam
.
Spine
 
1993
;
18
:
1433
8
.
5
Tubach
F
Beaute
J
Leclerc
A.
Natural history and prognostic indicators of sciatica
.
J Clin Epidemiol
 
2004
;
57
:
174
9
.
6
Peul
WC
Brand
R
Thomeer
RT
Koes
BW.
Influence of gender and other prognostic factors on outcome of sciatica
.
Pain
 
2008
;
138
:
180
91
.
7
Suri
P
Rainville
J
Hunter
DJ
Li
L
Katz
JN.
Recurrence of radicular pain or back pain after nonsurgical treatment of symptomatic lumbar disk herniation
.
Arch Phys Med Rehabil
 
2012
;
93
:
690
5
.
8
Haugen
AJ
Brox
JI
Grovle
L
, et al.   .
Prognostic factors for non-success in patients with sciatica and disc herniation
.
BMC Musculoskelet Disord
 
2012
;
13
183.
9
Grovle
L
Haugen
AJ
Keller
A
, et al.   .
Prognostic factors for return to work in patients with sciatica
.
Spine J
 
2013
;
13
:
1849
57
.
10
Abejon
D
Garcia-del-Valle
S
Fuentes
ML
, et al.   .
Pulsed radiofrequency in lumbar radicular pain: Clinical effects in various etiological groups
.
Pain Pract
 
2007
;
7
:
21
6
.
11
Teixeira
A
Grandinson
M
Sluijter
M.
Pulsed Radiofrequency for radicular pain due to a herniated intervertebral disc—An initial report
.
Pain Pract
 
2005
;
5
:
111
5
.
12
Munglani
R.
The longer term effect of pulsed radiofrequency for neuropathic pain
.
Pain
 
1999
;
80
:
437
9
.
13
Sluijter
ME
Cosman
ER
Rittman
IIWB
van Kleef
M.
The effects of pulsed radiofrequency field applied to the dorsal root ganglion-a preliminary report
.
Pain Clin
 
1998
;
11
:
109
17
.
14
Simopoulos
TT
Kraemer
J
Nagda
JV
Aner
M
Bajwa
ZH.
Response to pulsed and continuous radiofrequency lesioning of the dorsal root ganglion and segmental nerves in patients with chronic lumbar radicular pain
.
Pain Physician
 
2008
;
11
137
44
.
15
Van Boxem
K
van Bilsen
J
de Meij
N
, et al.   .
Pulsed radiofrequency treatment adjacent to the lumbar dorsal root ganglion for the management of lumbosacral radicular syndrome: A clinical audit
.
Pain Med
 
2011
;
12
:
1322
30
.
16
Van Zundert
J
Patijn
J
Kessels
A
, et al.   .
Pulsed radiofrequency adjacent to the cervical dorsal root ganglion in chronic cervical radicular pain: A double blind sham controlled randomized clinical trial
.
Pain
 
2007
;
127
:
173
82
.
17
Vroomen
PC
de Krom
MC
Knottnerus
JA.
Diagnostic value of history and physical examination in patients suspected of sciatica due to disc herniation: A systematic review
.
J Neurol
 
1999
;
246
:
899
906
.
18
Kerr
RS
Cadoux-Hudson
TA
Adams
CB.
The value of accurate clinical assessment in the surgical management of the lumbar disc protrusion
.
J Neurol Neurosurg Psychiatry
 
1988
;
51
:
169
73
.
19
Attal
N
Perrot
S
Fermanian
J
Bouhassira
D.
The neuropathic components of chronic low back pain: A prospective multicenter study using the DN4 Questionnaire
.
J Pain
 
2011
;
12
:
1080
7
.
20
Murphy
DR
Hurwitz
EL
Gerrard
JK
Clary
R.
Pain patterns and descriptions in patients with radicular pain: Does the pain necessarily follow a specific dermatome?
Chiropr Osteopat
 
2009
;
17
:
9
.
21
Hofstee
DJ
Gijtenbeek
JM
Hoogland
PH
, et al.   .
Westeinde sciatica trial: Randomized controlled study of bed rest and physiotherapy for acute sciatica
.
J Neurosurg
 
2002
;
96
:
45
9
.
22
Severeijns
R
van den Hout
MA
Vlaeyen
JW
Picavet
HS.
Pain catastrophizing and general health status in a large Dutch community sample
.
Pain
 
2002
;
99
:
367
76
.
23
Likert
R.
A technique for the measurement of attitudes
.
Arch Psychol
 
1932
;
22
:
140
.
24
Jensen
MP
McFarland
CA.
Increasing the reliability and validity of pain intensity measurement in chronic pain patients
.
Pain
 
1993
;
55
:
195
203
.
25
Ware
JE
Jr.
SF-36 health survey update
.
Spine
 
2000
;
25
:
3130
9
.
26
Roland
M
Fairbank
J.
The Roland-Morris Disability Questionnaire and the Oswestry Disability Questionnaire
.
Spine
 
2000
;
25
:
3115
24
.
27
Bouhassira
D
Attal
N
Alchaar
H
, et al.   .
Comparison of pain syndromes associated with nervous or somatic lesions and development of a new neuropathic pain diagnostic questionnaire (DN4)
.
Pain
 
2005
;
114
29
36
.
28
Bennett
M.
The LANSS Pain Scale: The Leeds assessment of neuropathic symptoms and signs
.
Pain
 
2001
;
92
:
147
57
.
29
Gallizzi
M
Gagnon
C
Harden
RN
Stanos
S
Khan
A.
Medication Quantification Scale Version III: Internal validation of detriment weights using a chronic pain population
.
Pain Pract
 
2008
;
8
:
1
4
.
30
Hays
RD
Morales
LS.
The RAND-36 measure of health-related quality of life
.
Ann Med
 
2001
;
33
:
350
7
.
31
Hosmer
D
Lemeshow
S.
hoofdstuk titel .
Applied Logistic Regression
  .
New York
:
John Wiley & Sons
;
1989
:
135
175
.
32
Van Boxem
K
de Meij
N
Kessels
A
Van Kleef
M
Van Zundert
J.
Pulsed radiofrequency for chronic intractable lumbosacral radicular pain: A six-month cohort study
.
Pain Med
 
2015
;
16
:
1155
62
.
33
Datta
S
Manchikanti
L
Falco
FJ
, et al.   .
Diagnostic utility of selective nerve root blocks in the diagnosis of lumbosacral radicular pain: Systematic review and update of current evidence
.
Pain Physician
 
2013
;
16
:
SE97
124
.
34
Wolff
A
Wilder-Smith
O.
Diagnosis in patients with chronic radiating low back pain without overt focal neurological deficits: What is the value of segmental nerve root blocks?
Therapy
 
2005
;
2
:
577
85
.
35
Xavier
AV
Farrell
CE
McDanal
J
Kissin
I.
Does antidromic activation of nociceptors play a role in sciatic radicular pain?
Pain
 
1990
;
40
:
77
9
.
36
North
RB
Kidd
DH
Zahurak
M
Piantadosi
S.
Specificity of diagnostic nerve blocks: A prospective, randomized study of sciatica due to lumbosacral spine disease
.
Pain
 
1996
;
65
:
77
85
.
37
Ramer
MS
French
GD
Bisby
MA.
Wallerian degeneration is required for both neuropathic pain and sympathetic sprouting into the DRG
.
Pain
 
1997
;
72
71
8
.
38
Ramer
MS
Thompson
SW
McMahon
SB.
Causes and consequences of sympathetic basket formation in dorsal root ganglia
.
Pain
 
1999
;
Suppl 6
:
S111
20
.
39
Czyrny
JJ
Lawrence
J.
The importance of paraspinal muscle EMG in cervical and lumbosacral radiculopathy: Review of 100 cases
.
Electromyogr Clin Neurophysiol
 
1996
;
36
:
503
8
.
40
Pinto
V
Szucs
P
Lima
D
Safronov
BV.
Multisegmental A{delta}- and C-fiber input to neurons in lamina I and the lateral spinal nucleus
.
J Neurosci
 
2010
;
30
:
2384
95
.
41
Wall
PD
Lidierth
M
Hillman
P.
Brief and prolonged effects of Lissauer tract stimulation on dorsal horn cells
.
Pain
 
1999
;
83
:
579
89
.
42
Lidierth
M.
Long-range projections of Adelta primary afferents in the Lissauer tract of the rat
.
Neurosci Lett
 
2007
;
425
:
126
30
.
43
Li
Y
Dorsi
MJ
Meyer
RA
Belzberg
AJ.
Mechanical hyperalgesia after an L5 spinal nerve lesion in the rat is not dependent on input from injured nerve fibers
.
Pain
 
2000
;
85
:
493
502
.
44
Gorman
WF
Hodak
JA.
Herniated intervertebral disc without pain
.
J Okla State Med Assoc
 
1997
;
90
185
90
.
45
Boden
SD
McCowin
PR
Davis
DO
, et al.   .
Abnormal magnetic-resonance scans of the cervical spine in asymptomatic subjects. A prospective investigation
.
J Bone Joint Surg Am
 
1990
;
72
:
1178
84
.
46
Boos
N
Rieder
R
Schade
V
, et al.   .
Volvo Award in clinical sciences. The diagnostic accuracy of magnetic resonance imaging, work perception, and psychosocial factors in identifying symptomatic disc herniations
.
Spine
 
1995
;
20
:
2613
25
.
47
Jensen
MC
Brant-Zawadzki
MN
Obuchowski
N
, et al.   .
Magnetic resonance imaging of the lumbar spine in people without back pain
.
N Engl J Med
 
1994
;
331
:
69
73
.
48
Modic
MT
Obuchowski
NA
Ross
JS
, et al.   .
Acute low back pain and radiculopathy: MR imaging findings and their prognostic role and effect on outcome
.
Radiology
 
2005
;
237
:
597
604
.
49
van Rijn
JC
Klemetso
N
Reitsma
JB
, et al.   .
Observer variation in the evaluation of lumbar herniated discs and root compression: Spiral CT compared with MRI
.
Br J Radiol
 
2006
;
79
:
372
7
.
50
El Barzouhi
A
Vleggeert-Lankamp
CL
Lycklama
A
, et al.   .
Magnetic resonance imaging in follow-up assessment of sciatica
.
N Engl J Med
 
2013
;
368
999
1007
.