REVIEW For reprint orders, please contact: [email protected]
Pharmacologically enhanced spinal cord stimulation for pain: an evolving strategy
Practice Points
Göran Lind†1 & Bengt Linderoth1
Clinical trials have been performed on the effect of combining spinal cord stimulation (SCS) with intrathecal drug administration.
Patients with neuropathic pain syndromes with insufficient benefits from SCS alone may be helped by the combined therapy of SCS plus a drug.
The standard requirements for an optimal SCS response have earlier been a reduction of the visual analog scale by at least 50% and a poststimulatory duration of >45 min.
Lately, however, several authors have challenged these criteria, questioning whether or not a 30% reduction would be valuable and if continuous low intensity SCS (as used by many patients by themselves) can be regarded as a beneficial therapy.
If the clinician and the patient deem the outcome of SCS treatment suboptimal, trials with SCS plus a drug can be performed according to the algorithms given in this review.
At present this begins with intrathecal baclofen in a bolus trial with the doses 25–75 µg.
If this proves beneficial, implantation of a pump could be considered.
Otherwise further options are given in the algorithms.
As indicated by animal data further drugs for this type of combined therapy are just around the corner.
Thus, we are only witnessing the beginning of drug-enhanced spinal stimulation.
SUMMARY
Spinal cord stimulation (SCS) as treatment for chronic neuropathic pain has developed into an important therapeutic strategy. However, several studies indicate that as many as 30–50% of patients do not respond sufficiently to technically well-functioning SCS. Experimental studies have revealed some of the possible neuronal systems and transmitters involved in SCS. Based on such data, a new strategy has been suggested: “pharmacologically enhanced spinal cord stimulation” using receptor active drugs to improve the therapeutic effect. The present article reviews the animal data on which clinical trials have been based Department of Neurosurgery, Karolinska University Hospital, Sweden Author for correspondence: [email protected]
1 †
10.2217/PMT.11.44 © 2011 Future Medicine Ltd
Pain Manage. (2011) 1(5), 441–449
ISSN 1758-1869
441
Review Lind & Linderoth and summarizes the clinical experience up to the present. Relevant data exist for intrathecal baclofen as an adjuvant to SCS, but trials with clonidine and adenosine have also been performed. Available basic studies indicate that other substances might also prove useful in future trials. The present data thus only announce the beginning of ‘drug-enhanced spinal stimulation’. Since the presentation of the gate-control theory of pain modulation by Melzack and Wall [1] and the pioneering work of Shealy [2,3] , spinal cord stimulation (SCS) has continuously evolved as a minimally invasive, reversible, and cost-efficient therapy for some forms of chronic pain, resistant to pharmacological treatment [4–6] . The annual number of implants at present, worldwide, is estimated to be between 30,000 and 40,000, including battery replacements. In many cases of neuropathic pain, however, a trial of SCS may end up unsuccessful. Up to 30–50% even of well-selected patients with chronic neuropathic pain may not experience pain alleviation during a short-term trial with SCS [7,8] . There are also patients who face reduction of stimulation effect after an initial, sometimes extensive, period with good pain relief from SCS. Technical stimulation alterations, such as reprogramming, trials with rest periods from stimulation, variation of pulse rate and width, operative relocation of the electrode lead or simply adding more leads are warranted in this situation, but may turn out ineffective as well. During the two last decades experimental studies of the underlying mechanisms of SCS, mainly in animals, have provided information suggestive of a possibility that the effect of SCS can be augmented pharmacologically, thus enabling help for such patients [9] . Experimental background Only sparse and inconclusive data from analyses of lumbar cerebrospinal fluid in conjunction with SCS have been presented. Opioid mechanisms seem probably not to be involved. SCS tends to increase the content of substance P (SP) in human cerebrospinal fluid and of SP and serotonin in the dorsal horn of cats [10,11] . It is, however, likely that these changes of SP are not related to the pain relieving effect. Experimental studies on SCS have continued in nerve lesioned rats demonstrating hypersensitivity to previously innocuous stimuli as a model of neuropathic pain. Using microdialysis in the dorsal horn in this context it has been shown that SCS increases the release of GABA and in parallel significantly reduces
442
Pain Manage. (2011) 1(5)
the release of glutamate and aspartate (excitatory amino acids) [12] . The SCS effect on GABA release, however, only occurred in animals that had also responded to SCS with significant suppression of hind paw hypersensitivity in previous behavioral experiments [13] . Furthermore GABA was observed to decrease in the spinal cord after peripheral nerve injury [14] , and plausibly SCS may act by restoring normal GABA levels in the dorsal horn. In behavioral experiments it has been shown that the suppression of hypersensitivity that SCS exerts could be inhibited by intrathecal injection of a GABA B antagonist, while a GABA A antagonist (bicuculline) was less effective [12,15] . If instead GABA or a GABA B agonist, baclofen, was administered intrathecally the effect of SCS was markedly enhanced [15] . It was even shown that rats not responding to SCS (i.e., their mechanical hypersensitivity was not attenuated) could be converted to responders with intrathecal administration of a low dose of baclofen, a dose that was by itself ineffective [15] . A similar potentiating effect has been found also for R-PIA (acting on the adenosine A1 receptor) [16,17] and later on for other pharmaceuticals already in clinical use – gabapentin and pregabalin [18] as well as for clonidine (acting via different mechanisms than those mentioned above – for further discussion see [19]) [19,20] . The use of antagonists to reduce the effect and agonists to enhance it indicates that the improvement is largely receptor specific. We consider the benefits of the combined therapy to be primarily potentiation, although additive components cannot be excluded. It is thus highly probable that GABAergic and adenosinergic mechanisms are involved in the beneficial effects of SCS [15,17] . Furthermore, the antinociceptive effect of clonidine is believed to be related to activation of the spinal cholinergic system. If so, the effect of SCS might act also via involvement of these mechanisms, a notion supported by recent studies in rats [20,21] . In a spinal microdialysis study the release of acetylcholine (Ach) in the dorsal horn was increased in SCS responding nerve-injured rats while in the nonresponders the
future science group
Pharmacologically enhanced spinal cord stimulation for pain: an evolving strategy Ach levels were unchanged by SCS. When Ach receptor (muscarinic and nicotinic) antagonists were administered intrathecally activation of the muscarinic M4 and M2 receptors was shown to be essential for the SCS effect. Also a recent immunohistochemical study seems to confirm the important role of the muscarinic M4 receptor in the response to SCS after peripheral nerve injury [22] . The effect of activation of muscarinic receptors in conjunction with SCS has also been studied. The nonspecific muscarinic receptor agonist oxotremorine was administered intrathecally via a chronic catheter in rats with signs of neuropathy subjected to SCS. Following the routine in the laboratory the animals were tested using von Frey filaments, cold spray and focused radiant heat, under awake, freely moving conditions. It could thus be shown that intrathecal oxotremorine dose-dependently suppressed tactile hypersensitivity. As expected withdrawal thresholds and withdrawal latencies were enhanced by SCS and also cold scores improved. The combination of SCS and a subeffective dose of oxotremorine intrathecal led to a substantial increase of the SCS effect even in rats that did not display a satisfactory effect with SCS alone [22] . As mentioned above, it has been shown in cats that SCS induces an increase in dorsal horn serotonin content [23] . Serotonin has, in turn, been shown to have analgesic properties in rats [24–26] . However, spinal serotonin has also been implicated in the development of persistent pain [27] . In a recent experiment in rats the content of serotonin was measured both in the contralateral and the ipsilateral dorsal horn of animals subject to nerve injury [28] . Animals previously responding behaviorally to SCS were compared with animals not responding as well as with controls. It appeared that SCS induced increased ipsilateral dorsal horn serotonin levels in responders only. It was also tested if a low, in itself ineffective, dose of serotonin could enhance the effect of SCS in nonresponding animals and in this experiment it also appeared possible to convert a nonresponding animal to a SCS responder, by the combined therapy [28] . Two anticonvulsant drugs, gabapentin and pregabalin, have also been tested in conjunction with SCS [18] . Both substances were shown to reduce signs of tactile hypersensitivity in rats with experimentally induced mononeuropathy, in a dose-dependent manner. Subeffective doses
future science group
Review
of either drug could turn a rat nonresponsive to SCS into a responder if administered together with the SCS. The effect may, at least in part, be mediated by a supplementary suppression of wide dynamic range neuron hyperexcitability in the dorsal horn, as shown by extracellular recordings from animals with similar lesions. Gabapentin and pregabalin are interesting drugs in this context as they are widely used and can be administered perorally. As some antidepressants are also used for treatment of neuropathic pain the Karolinska group has studied interactions between three antidepressants and SCS. In a fashion similar to the previously mentioned studies subeffective doses were given intrathecally concomitantly to SCS. Three drugs were tested: a tricyclic antidepressant (amitriptyline), a selective serotonin reuptake inhibitor (fluoxetine) and a selective serotonin/noradrenalin reuptake inhibitor (milnacipran). In consideration of the common anticholinergic side effects of amitriptyline and the fact that part of the effect of SCS relies on cholinergic mechanisms it was feared that amitryptiline even might counteract the effect of SCS. However, it turned out that both amitriptyline and milnacipran produced a significant enhancement of the suppressive effect on mechanical hypersensitivity by SCS alone. Fluoxetine, on the other hand, did not influence the effect of SCS, at least not in the doses used in the experiment [29] . There were no signs of negative interactions, not even with higher doses of amitriptyline. In a recent study [30] , ketamine (an NMDA receptor antagonist) was used in a similar manner to convert rats not responding to SCS therapy after a partial sciatic nerve lesion to responders using individually determined subeffective doses of the drug administered intrathecally. The aim here was to depress the glutamine-induced central sensitization partly with the drug and additionally with stimulation. In summary, the potential involvement of serotonergic, GABAergic and cholinergic mechanisms in the pain relieving effect of SCS has been studied extensively in animal experiments. It has been shown that a substantial part of this effect is mediated via GABA B, muscarinic M4 and by several serotonin (5‑HT) receptors [12,13,21,28,31] . Furthermore, in these animal models of neuropathy intrathecal administration of the GABA B receptor agonist baclofen [15] , the anticonvulsants gabapentin
www.futuremedicine.com
443
Review Lind & Linderoth and pregabalin [18] , the adenosine A1 receptor agonist R‑PIA [16] , the a2-adrenoceptor agonist clonidine [20] , the muscarinic receptor agonist oxotremorine [22] and 5‑HT [28] seem to significantly enhance the pain relieving effect of SCS as well as the antidepressant drugs amitriptyline (tricyclic antidepressant) and milnacipran (selective serotonin/noradrenalin reuptake inhibitor) [29,32] . In conclusion, SCS probably activates multiple transmitter systems, simultaneously and/or in sequence and a number of publications point to complex interactions within the different neuronal circuits that may relate to the SCS effect. The combination of SCS and an appropriate drug may enhance the pain relieving effect. Clinical studies Baclofen
Based on the results from animal studies, as described above, clinical studies were initiated. In the first of these studies baclofen – being already in use for spinal administration – was tested on patients with neuropathic pain responding poorly to SCS. It was shown that simultaneous administration of intrathecal baclofen in low doses could in some patients enhance or restore the effect of SCS [33] . Baclofen was injected intrathecally as boluses in doses ranging from 25 to 100 µg in 43 patients. In the first patients catheters were implanted, but in the majority of patients daily
fine needle lumbar punctures could be used with few problems. The patients repeatedly evaluated their pain and after 90 min a 30 min session of SCS was started. Figure 1 shows an example of a patient’s responses to increasing doses of intrathecal baclofen and SCS. Patients responding favorably to the combination of intrathecal drug and SCS were offered implantation of a pump. However, only 20–25% of the tested patients responded so well that a pump implantation was warranted. A few patients could achieve a satisfactory result by combining SCS and oral baclofen. Unfortunately they all complained of side effects and quit the trial after some few weeks. Contrary to the findings in animal studies it did not prove possible to convert a totally SCS nonresponding patient into a responder. This observation may reflect a confounding factor/ different mechanisms between the stimulusevoked pain-related behavior response in the animals and the reduction of continuous neuropathic pain components observed clinically. The general relevance of animal data for the clinical use of SCS is beyond the scope of this review. The interested reader is referred to a review by Vierck et al. [34] . A long-term follow-up of the patients treated with pumps for intrathecal baclofen administration was performed between 2005 and 2006. The follow-up study also comprised patients treated with baclofen intrathecally as the sole therapy for their neuropathic pain. An extensive search of medical records as well
VAS score
100 80 60
SCS 25 µg + SCS 50 µg + SCS
40 20 0
75 µg + SCS 0
0.5 SCS
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Time (h)
Figure 1. Time course of pain relief for an individual patient receiving different doses of baclofen intrathecally and spinal cord stimulation. The combination of SCS and increasing doses of baclofen markedly enhanced and prolonged the therapeutic effect. SCS: Spinal cord stimulation; VAS: Visual analog scale. Adapted with permission from [33].
444
Pain Manage. (2011) 1(5)
future science group
Pharmacologically enhanced spinal cord stimulation for pain: an evolving strategy
Review
**
100
**
90 80 NS
VAS score
70 60 50 40 30 20 10 0 Preimplant
Early follow-up
Late follow-up
Figure 2. Mean visual analog scale score (± standard deviation) after 73 months of follow-up (late follow-up). The ‘combined therapy’ produces fairly stable long-term results. **p
Pharmacologically enhanced spinal cord stimulation for pain: an evolving strategy
Practice Points
Göran Lind†1 & Bengt Linderoth1
Clinical trials have been performed on the effect of combining spinal cord stimulation (SCS) with intrathecal drug administration.
Patients with neuropathic pain syndromes with insufficient benefits from SCS alone may be helped by the combined therapy of SCS plus a drug.
The standard requirements for an optimal SCS response have earlier been a reduction of the visual analog scale by at least 50% and a poststimulatory duration of >45 min.
Lately, however, several authors have challenged these criteria, questioning whether or not a 30% reduction would be valuable and if continuous low intensity SCS (as used by many patients by themselves) can be regarded as a beneficial therapy.
If the clinician and the patient deem the outcome of SCS treatment suboptimal, trials with SCS plus a drug can be performed according to the algorithms given in this review.
At present this begins with intrathecal baclofen in a bolus trial with the doses 25–75 µg.
If this proves beneficial, implantation of a pump could be considered.
Otherwise further options are given in the algorithms.
As indicated by animal data further drugs for this type of combined therapy are just around the corner.
Thus, we are only witnessing the beginning of drug-enhanced spinal stimulation.
SUMMARY
Spinal cord stimulation (SCS) as treatment for chronic neuropathic pain has developed into an important therapeutic strategy. However, several studies indicate that as many as 30–50% of patients do not respond sufficiently to technically well-functioning SCS. Experimental studies have revealed some of the possible neuronal systems and transmitters involved in SCS. Based on such data, a new strategy has been suggested: “pharmacologically enhanced spinal cord stimulation” using receptor active drugs to improve the therapeutic effect. The present article reviews the animal data on which clinical trials have been based Department of Neurosurgery, Karolinska University Hospital, Sweden Author for correspondence: [email protected]
1 †
10.2217/PMT.11.44 © 2011 Future Medicine Ltd
Pain Manage. (2011) 1(5), 441–449
ISSN 1758-1869
441
Review Lind & Linderoth and summarizes the clinical experience up to the present. Relevant data exist for intrathecal baclofen as an adjuvant to SCS, but trials with clonidine and adenosine have also been performed. Available basic studies indicate that other substances might also prove useful in future trials. The present data thus only announce the beginning of ‘drug-enhanced spinal stimulation’. Since the presentation of the gate-control theory of pain modulation by Melzack and Wall [1] and the pioneering work of Shealy [2,3] , spinal cord stimulation (SCS) has continuously evolved as a minimally invasive, reversible, and cost-efficient therapy for some forms of chronic pain, resistant to pharmacological treatment [4–6] . The annual number of implants at present, worldwide, is estimated to be between 30,000 and 40,000, including battery replacements. In many cases of neuropathic pain, however, a trial of SCS may end up unsuccessful. Up to 30–50% even of well-selected patients with chronic neuropathic pain may not experience pain alleviation during a short-term trial with SCS [7,8] . There are also patients who face reduction of stimulation effect after an initial, sometimes extensive, period with good pain relief from SCS. Technical stimulation alterations, such as reprogramming, trials with rest periods from stimulation, variation of pulse rate and width, operative relocation of the electrode lead or simply adding more leads are warranted in this situation, but may turn out ineffective as well. During the two last decades experimental studies of the underlying mechanisms of SCS, mainly in animals, have provided information suggestive of a possibility that the effect of SCS can be augmented pharmacologically, thus enabling help for such patients [9] . Experimental background Only sparse and inconclusive data from analyses of lumbar cerebrospinal fluid in conjunction with SCS have been presented. Opioid mechanisms seem probably not to be involved. SCS tends to increase the content of substance P (SP) in human cerebrospinal fluid and of SP and serotonin in the dorsal horn of cats [10,11] . It is, however, likely that these changes of SP are not related to the pain relieving effect. Experimental studies on SCS have continued in nerve lesioned rats demonstrating hypersensitivity to previously innocuous stimuli as a model of neuropathic pain. Using microdialysis in the dorsal horn in this context it has been shown that SCS increases the release of GABA and in parallel significantly reduces
442
Pain Manage. (2011) 1(5)
the release of glutamate and aspartate (excitatory amino acids) [12] . The SCS effect on GABA release, however, only occurred in animals that had also responded to SCS with significant suppression of hind paw hypersensitivity in previous behavioral experiments [13] . Furthermore GABA was observed to decrease in the spinal cord after peripheral nerve injury [14] , and plausibly SCS may act by restoring normal GABA levels in the dorsal horn. In behavioral experiments it has been shown that the suppression of hypersensitivity that SCS exerts could be inhibited by intrathecal injection of a GABA B antagonist, while a GABA A antagonist (bicuculline) was less effective [12,15] . If instead GABA or a GABA B agonist, baclofen, was administered intrathecally the effect of SCS was markedly enhanced [15] . It was even shown that rats not responding to SCS (i.e., their mechanical hypersensitivity was not attenuated) could be converted to responders with intrathecal administration of a low dose of baclofen, a dose that was by itself ineffective [15] . A similar potentiating effect has been found also for R-PIA (acting on the adenosine A1 receptor) [16,17] and later on for other pharmaceuticals already in clinical use – gabapentin and pregabalin [18] as well as for clonidine (acting via different mechanisms than those mentioned above – for further discussion see [19]) [19,20] . The use of antagonists to reduce the effect and agonists to enhance it indicates that the improvement is largely receptor specific. We consider the benefits of the combined therapy to be primarily potentiation, although additive components cannot be excluded. It is thus highly probable that GABAergic and adenosinergic mechanisms are involved in the beneficial effects of SCS [15,17] . Furthermore, the antinociceptive effect of clonidine is believed to be related to activation of the spinal cholinergic system. If so, the effect of SCS might act also via involvement of these mechanisms, a notion supported by recent studies in rats [20,21] . In a spinal microdialysis study the release of acetylcholine (Ach) in the dorsal horn was increased in SCS responding nerve-injured rats while in the nonresponders the
future science group
Pharmacologically enhanced spinal cord stimulation for pain: an evolving strategy Ach levels were unchanged by SCS. When Ach receptor (muscarinic and nicotinic) antagonists were administered intrathecally activation of the muscarinic M4 and M2 receptors was shown to be essential for the SCS effect. Also a recent immunohistochemical study seems to confirm the important role of the muscarinic M4 receptor in the response to SCS after peripheral nerve injury [22] . The effect of activation of muscarinic receptors in conjunction with SCS has also been studied. The nonspecific muscarinic receptor agonist oxotremorine was administered intrathecally via a chronic catheter in rats with signs of neuropathy subjected to SCS. Following the routine in the laboratory the animals were tested using von Frey filaments, cold spray and focused radiant heat, under awake, freely moving conditions. It could thus be shown that intrathecal oxotremorine dose-dependently suppressed tactile hypersensitivity. As expected withdrawal thresholds and withdrawal latencies were enhanced by SCS and also cold scores improved. The combination of SCS and a subeffective dose of oxotremorine intrathecal led to a substantial increase of the SCS effect even in rats that did not display a satisfactory effect with SCS alone [22] . As mentioned above, it has been shown in cats that SCS induces an increase in dorsal horn serotonin content [23] . Serotonin has, in turn, been shown to have analgesic properties in rats [24–26] . However, spinal serotonin has also been implicated in the development of persistent pain [27] . In a recent experiment in rats the content of serotonin was measured both in the contralateral and the ipsilateral dorsal horn of animals subject to nerve injury [28] . Animals previously responding behaviorally to SCS were compared with animals not responding as well as with controls. It appeared that SCS induced increased ipsilateral dorsal horn serotonin levels in responders only. It was also tested if a low, in itself ineffective, dose of serotonin could enhance the effect of SCS in nonresponding animals and in this experiment it also appeared possible to convert a nonresponding animal to a SCS responder, by the combined therapy [28] . Two anticonvulsant drugs, gabapentin and pregabalin, have also been tested in conjunction with SCS [18] . Both substances were shown to reduce signs of tactile hypersensitivity in rats with experimentally induced mononeuropathy, in a dose-dependent manner. Subeffective doses
future science group
Review
of either drug could turn a rat nonresponsive to SCS into a responder if administered together with the SCS. The effect may, at least in part, be mediated by a supplementary suppression of wide dynamic range neuron hyperexcitability in the dorsal horn, as shown by extracellular recordings from animals with similar lesions. Gabapentin and pregabalin are interesting drugs in this context as they are widely used and can be administered perorally. As some antidepressants are also used for treatment of neuropathic pain the Karolinska group has studied interactions between three antidepressants and SCS. In a fashion similar to the previously mentioned studies subeffective doses were given intrathecally concomitantly to SCS. Three drugs were tested: a tricyclic antidepressant (amitriptyline), a selective serotonin reuptake inhibitor (fluoxetine) and a selective serotonin/noradrenalin reuptake inhibitor (milnacipran). In consideration of the common anticholinergic side effects of amitriptyline and the fact that part of the effect of SCS relies on cholinergic mechanisms it was feared that amitryptiline even might counteract the effect of SCS. However, it turned out that both amitriptyline and milnacipran produced a significant enhancement of the suppressive effect on mechanical hypersensitivity by SCS alone. Fluoxetine, on the other hand, did not influence the effect of SCS, at least not in the doses used in the experiment [29] . There were no signs of negative interactions, not even with higher doses of amitriptyline. In a recent study [30] , ketamine (an NMDA receptor antagonist) was used in a similar manner to convert rats not responding to SCS therapy after a partial sciatic nerve lesion to responders using individually determined subeffective doses of the drug administered intrathecally. The aim here was to depress the glutamine-induced central sensitization partly with the drug and additionally with stimulation. In summary, the potential involvement of serotonergic, GABAergic and cholinergic mechanisms in the pain relieving effect of SCS has been studied extensively in animal experiments. It has been shown that a substantial part of this effect is mediated via GABA B, muscarinic M4 and by several serotonin (5‑HT) receptors [12,13,21,28,31] . Furthermore, in these animal models of neuropathy intrathecal administration of the GABA B receptor agonist baclofen [15] , the anticonvulsants gabapentin
www.futuremedicine.com
443
Review Lind & Linderoth and pregabalin [18] , the adenosine A1 receptor agonist R‑PIA [16] , the a2-adrenoceptor agonist clonidine [20] , the muscarinic receptor agonist oxotremorine [22] and 5‑HT [28] seem to significantly enhance the pain relieving effect of SCS as well as the antidepressant drugs amitriptyline (tricyclic antidepressant) and milnacipran (selective serotonin/noradrenalin reuptake inhibitor) [29,32] . In conclusion, SCS probably activates multiple transmitter systems, simultaneously and/or in sequence and a number of publications point to complex interactions within the different neuronal circuits that may relate to the SCS effect. The combination of SCS and an appropriate drug may enhance the pain relieving effect. Clinical studies Baclofen
Based on the results from animal studies, as described above, clinical studies were initiated. In the first of these studies baclofen – being already in use for spinal administration – was tested on patients with neuropathic pain responding poorly to SCS. It was shown that simultaneous administration of intrathecal baclofen in low doses could in some patients enhance or restore the effect of SCS [33] . Baclofen was injected intrathecally as boluses in doses ranging from 25 to 100 µg in 43 patients. In the first patients catheters were implanted, but in the majority of patients daily
fine needle lumbar punctures could be used with few problems. The patients repeatedly evaluated their pain and after 90 min a 30 min session of SCS was started. Figure 1 shows an example of a patient’s responses to increasing doses of intrathecal baclofen and SCS. Patients responding favorably to the combination of intrathecal drug and SCS were offered implantation of a pump. However, only 20–25% of the tested patients responded so well that a pump implantation was warranted. A few patients could achieve a satisfactory result by combining SCS and oral baclofen. Unfortunately they all complained of side effects and quit the trial after some few weeks. Contrary to the findings in animal studies it did not prove possible to convert a totally SCS nonresponding patient into a responder. This observation may reflect a confounding factor/ different mechanisms between the stimulusevoked pain-related behavior response in the animals and the reduction of continuous neuropathic pain components observed clinically. The general relevance of animal data for the clinical use of SCS is beyond the scope of this review. The interested reader is referred to a review by Vierck et al. [34] . A long-term follow-up of the patients treated with pumps for intrathecal baclofen administration was performed between 2005 and 2006. The follow-up study also comprised patients treated with baclofen intrathecally as the sole therapy for their neuropathic pain. An extensive search of medical records as well
VAS score
100 80 60
SCS 25 µg + SCS 50 µg + SCS
40 20 0
75 µg + SCS 0
0.5 SCS
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Time (h)
Figure 1. Time course of pain relief for an individual patient receiving different doses of baclofen intrathecally and spinal cord stimulation. The combination of SCS and increasing doses of baclofen markedly enhanced and prolonged the therapeutic effect. SCS: Spinal cord stimulation; VAS: Visual analog scale. Adapted with permission from [33].
444
Pain Manage. (2011) 1(5)
future science group
Pharmacologically enhanced spinal cord stimulation for pain: an evolving strategy
Review
**
100
**
90 80 NS
VAS score
70 60 50 40 30 20 10 0 Preimplant
Early follow-up
Late follow-up
Figure 2. Mean visual analog scale score (± standard deviation) after 73 months of follow-up (late follow-up). The ‘combined therapy’ produces fairly stable long-term results. **p
