Wolff, André P. MD*†‡; Wilder Smith, Oliver H. G. MD, PhD†; Crul, Ben J. P. MD, PhD†; van de Heijden, Marc P.†; Groen, Gerbrand J. MD, PhD‡
Selective segmental nerve blocks with local anesthetics are applied for diagnostic purposes in patients with chronic back pain to determine the segmental level of the pain (1–5). Most patients with chronic pain are treated in day-care and must therefore have sufficient motor control to be discharged. In many pain clinics, lidocaine is used because of its intermediate-acting effect. Ropivacaine, a long-acting local anesthetic, provokes a less intense motor deficit than bupivacaine (6) and possibly also less than lidocaine. These pharmacokinetic and pharmacodynamic characteristics of ropivacaine might be of benefit in spinal segmental nerve blocks. No reports have appeared on the use of ropivacaine in segmental nerve blocks, and, specifically, on its effects on segmental motor function.
In one study, the sensory effects of lumbosacral (L1 to S1) segmental nerve blocks by local anesthetics were found to exhibit (7) a large variability in size and location of hypesthetic dermal areas, but with less variability for elicited paraesthesias. It was concluded that for a proper segmental diagnosis, one should also consider the overlapping of neighboring dermatomes (6,8–11). Determination of loss of motor function is equally part of the diagnostic segmental evaluation in invasive pain treatment. Because myotomes are also innervated multisegmentally (8), overlap of muscular innervation and the recruitment of inactive motor units (12) could theoretically mask motor deficits after segmental nerve blocks by local anesthetics. A decrease in pain might further contribute to an increase in motor function, because motor function can be inhibited by pain (13,14).
This study focused on the motor effects induced by L4 spinal nerve blocks with lidocaine 1% versus ropivacaine 0.25%, assuming a ratio of 1:4 for the anesthetic relative potencies for lidocaine and ropivacaine (6). The fourth lumbar segmental nerve was chosen because it is the main motor-supplying nerve for two muscles whose force can easily be measured (i.e., the quadriceps femoris and tibialis anterior muscles). The aims of this study were 1) to establish whether there is also large variability in motor effects after L4 spinal nerve blocks by local anesthetics; 2) to compare the motor effects of the standard local anesthetic lidocaine versus equipotent doses of ropivacaine; and 3) to evaluate the relationship between pain intensity and the effect of a segmental nerve block on motor function. Assessment of duration of effect was not a study goal.
We consecutively recruited patients with unilateral chronic low back pain radiating to the leg who were referred to the pain clinic for symptomatic invasive pain treatment. All patients were examined extensively, including computed tomography, magnetic resonance imaging, and electromyelography, and were diagnosed by a neurologist or an orthopedic surgeon as having lumbosacral radicular syndrome without neurological deficit (n = 44). Inclusion criteria were pain present for at least 6 mo and a verbal numeric rating scale score for pain (VNRS; 0 = no pain; 10 = unendurable pain) of at least 5 at the moment of inclusion in the study. According to the hospital’s standard protocol, these patients were candidates for a series of lumbosacral segmental nerve test blocks, including L4 (n = 22). Exclusion criteria were availability of causal therapy, known hypersensitivity to amino-amide-type local anesthetics or iodide, presence of blood coagulopathy, or mental disorders. Patients were recruited in accordance with the rules of the Declaration of Helsinki. The study was approved by the Hospital Ethics Committee, and each patient gave written, informed consent. Two patients refused inclusion in the study. Twenty patients were ultimately recruited, of which 19 patients finished the complete protocol. One patient was excluded because of too much pain during the measurement procedures, as a result of which the muscle force data became unreliable.
Patients always started their test series with a test block at a level other than L4, which was used for patient instruction and practice according to the study protocol. All other blocks were done on separate days. The second test block was performed at L4 with a randomly assigned local anesthetic—Xylocaine® 1% (lidocaine) or Naropin® 0.25% (ropivacaine)—and vice versa for the third block. Randomization for starting with ropivacaine (Group RL) or lidocaine (Group LR) was done by the hospital pharmacist, who provided closed, numbered envelopes. At the end of the study period, 20 patients had undergone 40 diagnostic segmental nerve blocks at L4 according to the experimental protocol. The blocks were performed by three experienced anesthesiologists specialized in invasive pain treatment. All motor function measurements were performed by a research fellow, who was blinded for the type of local anesthetic, the painful side, and the side of the nerve block.
After patients arrived in the pain treatment room, 5 min before the test block, the VNRS score was recorded and the patient was positioned prone with a pillow under the abdomen to decrease lordosis. Under fluoroscopic guidance, the insertion site of the needle was marked on the skin on the painful side by a pencil. The opposite side was also marked symmetrically to prevent unblinding of the observer. After lidocaine infiltration anesthesia of the skin, a 10-cm electrode (23-gauge, Top®; Sanofi Santé) was inserted in the dorsocranial quadrant of the intervertebral foramen at L4-5 (lateral view) and advanced until the tip was positioned one third to halfway along the pedicle column (anteroposterior view). To confirm the position of the needle point, an electric current generated by a radiofrequency pulse- and lesion-generator system (RFG-3B; Radionics, Burlington, MA) was applied, thereby stimulating the spinal nerve or its dorsal root ganglion (DRG). Paraesthesias were evoked by stimulating with a frequency of 50 Hz, and muscular contractions were evoked by stimulating at 2 Hz with a motor stimulation threshold of at least 1.5 times the sensory stimulation threshold (0.2–2 V). The anesthesiologist recorded the voltages (0–2 V, as displayed on the lesion generator system) necessary to evoke the paraesthetic sensations (experienced by the patient) and muscular contractions (observed by the anesthesiologist by seeing and feeling the muscle contractions by hand). Subsequently, 0.3 mL of contrast dye (Omnipaque® 180 mg/mL; Nycomed Ireland, Ltd., Cork) was injected to confirm the adjacency of the DRG, and radiographs were taken in the anteroposterior and lateral direction. Finally, 0.7 mL of the lidocaine study solution (lidocaine 1% and Omnipaque 0.15%) or 0.7 mL of the ropivacaine study solution (ropivacaine 0.25% and Omnipaque 0.15%) was injected. Forty-five minutes after the test block, the VNRS was recorded again.
Baseline maximal muscular contraction forces were assessed at least 30 min before injection. The maximum voluntary muscle forces (MVMF) of the tibialis anterior and quadriceps femoris muscles were both measured during a maximal effort of 10 s and repeated twice at 5-min intervals to enable muscle recovery. After 30 min, the entire procedure was repeated on the other leg.
To measure MVMF, a special device was developed to perform the exercises in a standard way. This device enabled the patient to have a half-sitting and half-lying posture, which was necessary to optimize the test exercise (Fig. 1). The patient’s head was supported by a pillow in a neutral position. The trunk was held at an angle of 45° to the horizontal and supported by an adaptable support for the back. Hips and knees were both flexed to an angle of 90°. The Microfet handheld Dynamometer™ (Hoggan Health Industries, South Droper, UT) was fixed in this device to measure the MVMF (in newtons). To test the MVMF of the quadriceps muscle, the dynamometer was attached just proximal to the ankle, in the midline between the medial and lateral malleolus. The patient had to extend the lower part of the leg at the knee at maximal strength against the fixed dynamometer for 10 s. To test the MVMF of the tibialis anterior muscle, the dynamometer was attached just proximal to the head of the first metatarsal bone. The patient had to flex the foot at the ankle at MVMF for 10 s.
The data were processed by using Statistica for Windows (Release 4.5; Statsoft Inc., Tulsa, OK). A minimal decrease of 2 points in the VNRS score was considered as clinically significant (15). Between-group demographic data differences were tested by using Student’s t-test. Intervention order effect (first versus second treatment), drug effect (ropivacaine versus lidocaine), side effect (treatment side versus control side), site effect (motor effect on quadriceps femoris or tibialis anterior muscle), and dermatome effect (whether the maximum pain was in the L4 dermatome) on the median of MVMF were evaluated by analysis of variance. For post hoc analysis, Tukey testing was used.
Nonparametric testing (Mann-Whitney U-test and Wilcoxon’s signed rank test) was used to evaluate preblock versus postblock pain. Spearman regression analysis was used to assess the various relationships between preintervention VNRS, change in VNRS, change of MVMF, and sensory and motor thresholds (r > 0.3 was considered relevant with 95% confidence; P < 0.05 was considered statistically significant).
Nineteen (total of 38 segmental nerve blocks) of the 20 patients finished the study. In all but two cases, hypoesthesia was found by pinprick in the corresponding dermatome after local anesthetic injection. Independent assessment of the radiographs revealed no epidural spread of radiocontrast dye.
Group RL contained five male and five female patients with a mean age of 48.7 yr (range, 36–66 yr; SD, 10.9 yr); Group LR consisted of five male and four female patients with a mean age of 55.7 yr (range, 40–79 yr; SD, 13.0 yr). The differences in age between groups were not statistically significant (P = 0.23; Student’s t-test). Patient characteristics are described in Table 1.
Table 2 shows the effects of segmental nerve block on motor function and pain, expressed as relative changes in MVMF and absolute change in VNRS scores, respectively. Some patients did not show any decrease in VNRS after segmental nerve block, but for the entire group, the median VNRS decrease was 4.0 (before versus after:P < 0.00001; Wilcoxon’s signed rank test). The median pain VNRS (interquartile range; range) was 7.0 (6.0–8.0; 0–10) preblock and 3.0 (0.5–6.0; 0–8.5) postblock. In one case in the RL group, the patient underwent a segmental nerve block with lidocaine without any pain at that moment. This case was excluded from further data analysis. There were no significant differences in preblock VNRS, postblock VNRS, and change in VNRS between ropivacaine and lidocaine (Mann-Whitney U-test). Figure 2 shows the pre- and postinjection median MVMFs on the affected side and on the control side.
Analysis of variance demonstrated no significant effect on MVMF for the following independent factors: first versus second intervention (i.e., order effect), drug injected (ropivacaine or lidocaine), site (quadriceps femoris or tibialis anterior muscle), and whether the maximum pain intensity was in the L4 dermatome or not (both for the standard and adapted dermatomal map) (7). A statistically significant difference in effect on MVMF was found for affected versus control side (P = 0.016; Tukey test). When the results were pooled (both muscles together) for the affected side, a decrease of the MVMF was found for median (mean, 4.2%; range, −65% to 71%; SD, 24.4%) values. For the control side, an increase was observed for median (mean, 4.5%; range, −37% to 115%; sd, 24.3%) values.
The data were pooled for multiple regression to assess correlations between factors on the affected side. A significant negative correlation was found for change in VNRS score versus change in median MVMF (Fig. 3; Spearman R = −0.48; P = 0.00001): the larger the decrease in VNRS score, the larger the increase in MVMF on the affected side. This phenomenon was found despite the overall decrease in MVMFs for all cases on the affected side. No statistically significant correlations were found for preblock VNRS score versus change in median and sums of MVMF. Further significant correlations were found for the affected side for preblock VNRS versus postblock VNRS (Spearman R = +0.53; P = 000001) and the change in VNRS score versus postblock VNRS (Spearman R = +0.69; P = 0.0000001).
No significant correlations were found either for change in VNRS or for change in MVMF for the control side. For both sides, there was no significant correlation for preblock VNRS score versus sensory and motor electrostimulation thresholds or for the pre-block sensory or motor electrostimulation threshold versus change in the median or sums of MVMF.
The regression results of this study demonstrate that alleviation of pain by segmental nerve block with local anesthetics can be accompanied by an increase of MVMF in the musculature innervated by that segmental nerve. The larger the decrease in pain, the larger the increase in force. This is the first time that this phenomenon has been described in human subjects undergoing segmental nerve blocks with local anesthetics. This result was unexpected, because one would anticipate segmental nerve block by local anesthetics to decrease MVMF. A plausible explanation is that pain causes inhibition of motor function (16) and that when pain decreases, motor inhibition is reduced, in accordance with our results. However, for the group taken as a whole, the median MVMF on the affected side decreased after segmental nerve block. This was probably the effect of the cases in which nerve block did not decrease pain.
Le Pera et al. (16) demonstrated that tonic muscle pain can inhibit the motor system by using motor evoked potentials from the right abductor digiti minimi by trans-cranial magnetic stimulation of the left primary motor cortex in human subjects. To provoke pain, hypertonic (5%) saline was injected into the right and left abductor digiti minimi, into the right first dorsal interosseus, and into the subcutaneous region of the right abductor digiti minimi. Motor evoked potentials were significantly reduced in amplitude during pain induced in the right abductor digiti minimi and right first dorsal interosseus, but not during pain in the left abductor digiti minimi or during subcutaneous pain. An ipsilateral muscular and possibly a myotomal relationship was suggested. Paik et al. (13) studied the effect of conditioning stimulation of a peripheral nerve on responses of spinal dorsal horn cells and motor neurons in 16 decerebrate-spinal cats. Noxious mechanical and noxious thermal stimuli applied to the receptive fields reduced the activity of dorsal horn cells and motor neurones recorded from a filament of ventral rootlet divided from either the L7 or S1 root. One of the conclusions the authors reached was that conditioning stimulation of a peripheral nerve produced a powerful inhibition of the responses elicited by noxious stimuli, suggesting that inhibition is an antinociceptive effect. Le Bars et al. (14) described diffuse noxious inhibitory controls in animals and humans, where neurons in the dorsal horn of the spinal cord can be strongly inhibited by a nociceptive stimulus applied to any part of the body distinct from their excitatory receptive fields. Diffuse noxious inhibitory control is most likely sustained by a loop that includes supraspinal structures and is therefore different from segmental inhibition. In both of the phenomena described above, nociceptive signals can be modulated by powerful controls. All the above-described mechanisms have in common that pain can inhibit central and/or motor neuron activity and probably represent natural mechanisms meant to protect a subject from further harm. A segmental nerve block potentially decreases pain-induced motor inhibition. This might explain the unexpected increase of MVMF after local anesthetic segmental nerve block-induced pain reduction in patients with lumbosacral radiating pain in our study.
The dynamometer we used has a high intraobserver reliability (17,18), and the specially developed device made it possible to measure MVMF for 10 seconds. Submaximal efforts, avoided by instruction, increase the risk of not using all motor units and, hence, the possibility of recruiting motor units of other myotomes (12). When the effort was not done in the proper way, the measurement was excluded from further data processing, which happened in one patient. We paid special attention to keep a steady muscle temperature by covering the legs under wool blankets, thus preventing changes in stimulus conduction (19).
This study could not demonstrate differences in the effects on MVMF and pain between lidocaine and ropivacaine. In all but one case, the local anesthetics were applied at the correct location. In only one case (after injection with ropivacaine), the DRG and its segmental nerve could not be visualized with contrast dye. It is possible that lidocaine and ropivacaine are equally potent in segmental nerve blocks in the applied dosages. However, the number of patients might have been too small to demonstrate such an effect. There seems to be, except for a theoretical difference in duration of effect (not a study goal), no advantage from one local anesthetic over the other.
We would suggest that, on the basis of the results of this study, the value of local anesthetic segmental nerve blocks in diagnosing and predicting interventional outcomes must be treated with caution. Many factors play a role in causing this uncertainty in this context, including our lack of knowledge on the precise mechanisms of pain involved, the many variants of neurophysiology and pathophysiology, placebo effects, and technical aspects (20). In this study, we have attempted to illuminate motor aspects of this complex interaction. Further investigation is needed to achieve a better understanding of the underlying complexity of the effects of local anesthetic-induced segmental nerve blocks.
We conclude that in patients with unilateral chronic low back pain radiating to the leg, pain reduction induced by nerve (L4) block is associated with increased MVMF of the quadriceps femoris and tibialis anterior muscles. There are no differences in effect on MVMF between lidocaine and ropivacaine. The more the pain decreases, the more the MVMF increases. This study is the first to report these phenomena, which can be important for the interpretation of segmental nerve blocks.
1. Stolker RJ, Vervest ACM, Groen GJ. The management of chronic spinal pain by blocks: a review. Pain 1994;58:1–20.
2. Krempen JF, Smith BS. Nerve-root injection: a method for evaluating etiology in sciatica. J Bone Joint Surg Am 1974;56:1435–44.
3. Tajima T, Furukawa K, Kuramochi E. Selective lumbosacral radiculography block. Spine 1980;5:68–77.
4. Herron LD. Selective nerve root block in patient selection for lumbar surgery: surgical results. J Spinal Disord 1989;2:75–9.
5. Dooley JF, McBroom RJ, Taguchi T, Macnab I. Nerve root infiltration in the diagnosis of radicular pain. Spine 1988;13:79–83.
6. Spencer SL. Local anesthetics and analgesia. In: Ashburn MA, Rice LJ, eds. The management of pain. New York: Churchill Livingstone Inc., 1998:141–69.
7. Wolff AP, Groen GJ, Crul BJP. Diagnostic lumbosacral segmental nerve blocks with local anesthetics: a prospective double-blind study on the variability and interpretation of segmental effects. Reg Anesth Pain Med 2001;26:147–55.
8. Bolk L. De segmentale innervatie van romp en ledematen bij den mensch. Haarlem, The Netherlands: De Erven, F. Bohn, 1910.
9. Sherrington CS. Experiments in examination of the peripheral distribution of the fibers of the posterior roots of some spinal nerves. I. Philos Trans R Soc Lond B Biol Sci 1893;184:641–764.
10. Sherrington CS. Experiments in examination of the peripheral distribution of the fibers of the posterior roots of some spinal nerves. II. Philos Trans R Soc Lond B Biol Sci 1898;190:45–86.
11. Fletcher TF, Kitchell RL. The lumbar, sacral and coccygeal tactile dermatomes of the dog. J Comp Neurol 1966;128:171–80.
12. Sale DG. Influence of exercise and training on motor unit activation. Exerc Sport Sci Rev 1987;15:95–151.
13. Paik KS, Nam SC, Chung JM. Differential inhibition produced by peripheral conditioning stimulation on noxious mechanical and thermal responses of different classes of spinal neurons in the cat. Exp Neurol 1988;99:498–511.
14. Le Bars D, Villanueva L, Bouhassira D, Willer J. Diffuse noxious inhibitory controls (DNIC) in animals and man. Patol Fiziol Eksp Ter 1992;4:55–65.
15. Farrar JT, Berlin JA, Strom BL. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain 2001;94:149–58.
16. Le Pera D, Graven-Nielsen T, Valeriani M, et al. Inhibition of motor system excitability at cortical and spinal level by tonic muscle pain. Clin Neurophysiol 2001;112:1633–41.
17. Aa JHCG, van der Elvers JWH, Oostendorp RAB. Hand Held Dynamometrie: een pleidooi voor het klinisch gebruik van ratio’s. de Bruijn GSTJ. Ned T Fysiotherapie 1996;6:167–77.
18. Bohannon RW. Test-retest reliability of hand-held dynamometry during a single session of strength assessment. Phys Ther 1986;66:206–9.
19. Rome LC. Influence of temperature on muscle recruitment and muscle function in vivo. Am J Physiol 1990;259:R210–22.
20. Hogan QH, Abram SE. Neural blockade for diagnosis and prognosis: a review. Anesthesiology 1997;86:216–41.