Global Journal of Anesthesiology Decreases Lidocaine Infusion Decreases Pain Scores in a Fibromyalgia Pain Population with Signi ﬁ cant Differential Pain Relief Secondary to Smoking Status

Study Objective: Our hypothesis is that systemic lidocaine can signi ﬁ cantly reduce chronic pain associated with ﬁ bromyalgia. A secondary goal of the study is to determine if other patient demographics can affect the response to IV lidocaine. Design: Retrospective chart review Patients: 55 patients being treated for ﬁ bromyalgia by lidocaine infusion Interventions: Intravenous infusion of lidocaine 5mg/kg over 1 hour Measurements: Two-tailed paired t-tests ( α =0.05) were performed to test for signi ﬁ cant change from pre- to post-infusion in BPI, PI, and VAS scores. Smoking status and race classi ﬁ cation were also identi ﬁ ed. Main Results: Mean BPI score decreased from 83.2±16.5 to 73.7±21.6 after infusion of 4 to 5 mg/ kg of lidocaine in 500 ml 5% dextrose (p-value<0.0001). Mean VAS decreased from 7.6±1.6 to 5.8±2.2 (p<0.0001). The post-infusion BPI was signi ﬁ cantly lower in non-smokers 68.5 ±19.6 vs. smokers 80.3 ±17.2 (p=0.028) and Caucasians 69.3 ±20.3 vs. African-Americans 83.6±21.4 (p=0.027). Conclusions: A retrospective analysis of 55 ﬁ bromyalgia patients who received IV lidocaine shows a statistically signi ﬁ cant decrease in pain scores. A subgroup analysis shows that baseline pain scores are higher in smokers, and response to IV lidocaine is attenuated in smokers compared to non-smokers. Case


Introduction
Fibromyalgia is a complex chronic pain syndrome that represents the extreme spectrum of musculoskeletal pain in the general population. It is defi ned as widespread pain lasting for more than three months, along with the presence of tenderness, discomfort, or pain in 11 or more of 18 tender points in the body [1]. In addition, fi bromyalgia is often associated with other constitutional symptoms such as disturbed sleep, emotional distress, and pronounced fatigue. Like many other clinical syndromes, fi bromyalgia has no single specifi c feature but represents a symptom complex of self-reported or elicited fi ndings [2].
Fibromyalgia is a relatively common disorder that is present in about 2% of the population with higher rates among women when compared to men, 3.4% and 0.5%, respectively.
The prevalence of fi bromyalgia increases with age. The highest frequency is between 60 and 79 years (greater than 7.0% in women). Characteristic features of fi bromyalgia including altered pain threshold levels and symptoms complexes are similar in community and pain clinic populations, but the overall severity and functional disability are more severe in the pain clinic population [3]. In fact, the chronic pain associated with fi bromyalgia is signifi cantly associated with a decrease in health-related quality of life. Effective management of fi bromyalgia symptoms, especially the chronic pain component, is associated with an increase in quality of life [4].
Despite new insights into the pathogenesis and treatment of fi bromyalgia, it remains a condition that is diffi cult to treat.  [5]. Evidence that suggests systemic lidocaine can induce a signifi cant and selective reduction of several components of pain. The observed preferential anti-hyperalgesic and antiallodynic effects of this drug suggest a selective central action on the mechanisms underlying these [6].
As such, the goal of this study is to corroborate and strengthen earlier fi ndings in the literature regarding reductions in pain, as well as identify patient characteristics that may affect pain levels or response to treatment. Our hypothesis is that systemic lidocaine can signifi cantly reduce the chronic pain associated with fi bromyalgia. A secondary goal of the study is to determine if other patient demographics can affect the response to IV lidocaine.
The pharmacologic treatment of fi bromyalgia often is comprised of a multi-faceted approach. Classes of drugs commonly prescribed include antidepressants -primarily duloxetine, milnacipran, tricyclic antidepressants, and antiepileptics, mainly pregabalin. Other drug classes commonly prescribed include opioid analgesics, muscle relaxants and sleep agents [7]. Often, complementary treatments such as psychological therapy, cognitive behavioral therapy [5], and mind-body techniques including hypnosis and biofeedback [8], are used to help with symptom management. However, regardless of these therapeutic options, many patients still have signifi cant chronic pain complaints associated with fi bromyalgia.
Certain behavioral factors can affect how people perceive and experience pain. For example, smoking status is independently associated with chronic pain, with smoking associated with musculoskeletal pain despite adjustment for confounding factors [9]. Though there are no clear associations between pain threshold levels and smoking status [10], habitual cigarette smoking correlates with lower physical and psychological indices of chronic pain. Many chronic pain patients frequently mitigate their pain by smoking. 57% of patients who smoke report having a need to smoke when they are in pain [11].
However, when smoking and non-smoking back pain patients were compared, smokers showed signifi cantly higher levels of emotional distress, tended to remain inactive, and relied on medication more often than the nonsmoking patients [11].
Additionally, several studies have found an association between race and pain tolerance levels. Studies have shown that Caucasians typically have higher pain thresholds, African Americans have an intermediate level, and Asian Americans have the lowest pain threshold level [12,13]. The genetic and cultural differences of pain perception among various races presents a challenge, as a patient's genetic background, cultural views, and support structure must be accounted for in the development and treatment of evolving chronic pain. As such, there remains a need for effective treatments for patients with fi bromyalgia that also account for the potential impact of factors such as race.
Local anesthetics have been used in medical treatments for many years. Traditionally local anesthetics, such as lidocaine, have been used as an anesthetic for local and regional anesthesia as well as an antiarrhythmic agent. The variability in the effects of local anesthetics is attributable to differential blockade of various cellular systems at different concentrations. Lidocaine blocks sodium channels to inhibit neuronal transmission of pain signals but also has systemic effects in addition to inhibiting neuronal conduction [14]. Intravenous (IV) lidocaine has been utilized in various areas of medicine owing to its analgesic, anti-hyperalgesic, and anti-infl ammatory properties [15]. In the perioperative setting, there is data supporting the use of IV lidocaine to reduce post-operative opioid consumption, early return of bowel function, and decreased the length of hospital stay [16][17][18]. In response to emerging literature, there have been increased efforts to elucidate fully the role of intravenous lidocaine, especially in treating chronic pain states that have failed conservative therapy.
Recently, there has been renewed interest in IV lidocaine infusions as a therapy in the treatment of chronic pain states.
It has been used as a therapy for a wide range of chronic pain syndromes, such as fi bromyalgia and complex regional pain syndrome (CRPS) [19,20]. IV lidocaine has also proved effi cacious in the treatment of a variety of neuropathic pain conditions, which include post-stroke pain, peripheral neuropathy, diabetic neuropathy, and CRPS type 1 and 2 [21].
In a study on CRPS patients, signifi cant reductions in multiple pain parameters were demonstrated, including dynamic and static mechano-allodynia, deep muscle pain, joint pain, and thermal allodynia [21]. Also, in a fi bromyalgia patient population, subsequent IV lidocaine infusions over a fi ve day period, demonstrated signifi cant improvements in quality of life and visual analog scale (VAS) scores [19]. In addition, IV lidocaine (5 mg/kg body weight) has been shown to be effective in relieving the signs and symptoms of painful diabetic neuropathy in the short term [22,23]. Patients with peripheral nervous system (PNS) injury reported substantially more pain relief than those with central nervous system (CNS) injury or with the pain of unknown etiology [24]. This suggests that the pathophysiology of chronic pain due to PNS injury is different from that due to CNS injury and idiopathic pain and that pain due to PNS damage may be blocked by local anesthetic agents [24].
Unfortunately, the exact mechanisms involved in the improvement of fi bromyalgia pain through lidocaine infusion have not been clearly defi ned. Intravenous lidocaine affects both peripheral and central voltage-gated sodium channels with more affi nity for open channels. Studies in rats suggest the effect on peripheral voltage gated sodium channels is the major contributor to reducing the pain of skin incision. In contrast, it appears that intravenous lidocaine reduces neuropathic pain via reduction of central hyperexcitability [14]. Lidocaine also has actions separate from effects on central and peripheral voltage gated Na+ channels. Additional lidocaine mechanisms of action include increase in spinal acetylcholine, inhibition of spinal glycine receptors, release of endogenous opioids, activation of NMDA receptors, and decrease of infl ammation through the release of ATP and activation of potassium channels [15].

Methods
After Duke University Institutional Review Board (IRB) approval, a retrospective chart review was performed on 55 patients being actively treated for fi bromyalgia under a waiver of consent conditions. The demographic information for these subjects is included in table 1. In this group of fi bromyalgia patients, most of the individuals had signifi cant pain that was refractory to other more conservative treatments. All patients underwent lidocaine infusion using 5mg/kg of lidocaine in D5W infused intravenously over 1 hour. The maximum dose used regardless of weight was 500mg. Patient information retrieved from the chart review included: sex, race, weight, pain duration, duration of relief after lidocaine infusion, smoking status, and individual pain scores as measured by brief pain inventory (BPI), visual analog scale (VAS), and pain interference scores (PIS).
As seen in fi gure 1, smoking status is known in only 32 of our 55 patients. Because of the importance of smoking as a potential effect, we proceeded with the analysis in that subset.
Although results involving smoking status are based only on that subset, we have no reason to think it is not representative of the sample, nor to suspect any bias in data collection.
As a normal distribution of the population could not be assured, Wilcoxon signed rank tests were also performed on each paired measure with =0.05. Then, separate repeatedmeasures ANOVA's were performed to test the effects of smoking status, sex, or race on overall BPI, PI, and VAS scores, as well as on pre-post change in these scores, with =0.05.
Finally, a two-tailed t-tests (=0.05) were performed to compare post infusion BPI, PI, and VAS scores based on smoking status and race groupings, as both were found to have signifi cant differences on the F-test among groups.

Results
As seen in fi gure 2, mean BPI score was found to be 83.2±16.5 before and 73.7±21.6 after infusion of 4 to 5 mg/kg of lidocaine in 500 ml 5% dextrose. The mean improvement in BPI score was signifi cant both in the t-test and Wilcoxon test (both p<0.0001).

Analysis
Statistical analysis was able to reject the null hypothesis of equal means between pre-infusion and post-infusion measurements of all three pain scales used. When paired with the signifi cance of the one-tailed analysis, the results show a statistically signifi cant reduction in pain after an infusion of 4 to 5 mg/kg of lidocaine in 500 ml 5% dextrose. These results were confi rmed by a Wilcoxon signed rank test. In all cases, the null hypothesis could be rejected with p<0.0001. All three pain scales showed an equal and comparable result, suggesting that all three are reliable measurements of pain levels in patients both before and after interventional procedures.
The analysis of repeated measures matched pairs, taking into consideration a second variable, provided important information about the effi cacy of treatment based on different patient characteristics. When using smoking status as a second variable, a signifi cant difference was found between treatment results in non-smoking vs. smoking groups. For every pain scale used in our analysis, the patients that identifi ed as smokers had a higher baseline pain score and a less robust decrease in pain score following intravenous lidocaine infusion when compared to non-smokers. The overall mean of VAS for smokers was found to be 7.4, while for non-smokers it was calculated as 6.8. When only post-infusion VAS scores were compared between smokers and non-smokers, statistically signifi cant differences were found between both groups using all three scales. In all cases, smokers had a higher level of post-infusion pain and less improvement in pain scores after lidocaine infusion than non-smokers receiving a comparable treatment.
Our conclusions about smoking effects are limited by the fact that 23 of 55 patients were unable to be assigned a smoking status, as a retrospective chart review was unrevealing. However, we have no reason to suspect any bias in data collection, or to think this subset was not representative. Even in our subset of 32 patients whose smoking status was known, the effect of smoking was large enough to achieve signifi cance.
A similar difference was found between patients grouped by race. Caucasians had a greater amount of pain relief than African-Americans in both post-infusion BPI and PIS scores. No such difference was found on post-infusion VAS. At the same time, grouped matched pair analysis showed statistically signifi cant differences between racial groups for all three scales. There was not suffi cient diversity in the study population to analyze race groups other than Caucasians and African Americans. There were too few males in the study to support reliable conclusions about gender differences in treatment.

Discussion
The analysis of a cohort of 55 fi bromyalgia patients treated with IV Lidocaine shows a statistically signifi cant decrease in pain as measured by mean Brief Pain Inventory (BPI) score, Pain Intensity Score (PIS), and Visual Analogue Score (VAS). Our results confi rm what other studies have demonstrated in the past: intravenous lidocaine infusions are effi cacious in treating patients with fi bromyalgia [5].
However, further analyzing our data for patient's smoking status in the subanalysis of matched pairs, we found that tobacco smokers had higher levels of baseline pain as measured by BPI, VAS, and PIS, along with a lower absolute reduction in pain scores after IV lidocaine infusions, as compared to nonsmokers. This higher baseline pain score in smokers, as well as a lower absolute reduction in pain scores after lidocaine infusion compared to non-smokers, were noteworthy. However, the association between smoking and pain is not straightforward. Tobacco abuse and its association with pain have been documented by numerous studies in the past.
Tobacco use has been shown to be proportionately greater in populations with chronic pain than in populations without chronic pain [26]. Direct comparison of any chronic pain in smokers as compared to the chronic pain in non-smokers is fraught with confounders, including comorbid medical, psychological, and behavioral conditions that are prevalent in this population. In essence, when focusing on comorbid pain and smoking, the question is whether a patient's pain affects tobacco use, or that tobacco use affects and modulates a patient's response to pain.
Nicotine has been shown to have direct pain-inhibitory effects in numerous animal studies, when administered subcutaneously, as measured by tail fl ick and hot plate models [27][28][29]. Also, nicotine has been demonstrated to provide the same analgesic effects via indirect administration in animals, as in tobacco smoke [30]. Nicotine exerts its complex physiologic effects by interacting with the nicotinic acetylcholine receptor (nAChR), which is a transmembrane complex composed of fi ve proteins that mediate transmembrane passage of Na + , Ca 2+ , and K + ions [31].
These ubiquitous receptors are located in both the peripheral and central nervous system. Nicotinic acetylcholine receptors composed of alpha 4, beta two subunits have been shown to be present in the dorsal horn of the spinal cord, and the thalamus, likely modulating the transmission of pain in the central nervous system [32,33]. A supraspinal component of nicotine's pain modulating effect is likely due to a complex interplay with the descending endogenous opioid system. In rats, as measured by the tail-fl ick test, the opioid antagonist, naltrexone, and a centrally acting nicotinic receptor antagonist, mecamylamine, has been shown to block the analgesic effects of nicotine [34].
In human studies, nicotine has shown to have an acute analgesic effect in experimental models [35]. Much like the well-studied animal models, nicotine likely modulates the perception of pain at the dorsal horn in humans and modulates pain processing via central mechanisms such as descending inhibitory opioid pathways.
However, epidemiologic and prospective cohort studies have shown a positive association with smoking and chronic pain conditions [36,37]. This change is likely due to receptor desensitization and tolerance due to chronic exposure of nicotine, blunting the acute effects of nicotine on the body.
Smokers who have tobacco withheld from them have shorter pain latency to heat pain and reduced tolerance to electrical pain stimulation [38,39]. Smokers have been demonstrated to have higher levels of Substance P in the CSF, than non-smokers [40]. Smokers have lower beta-endorphin levels as compared to non-smokers [41], while patients with fi bromyalgia have lower beta-endorphin levels than controls [42].
Studies have assessed the relationship between smoking and fi bromyalgia, as in our study patient population. Several studies have implicated smoking with an increase in widespread musculoskeletal pain compared to non-smokers [36,43]. In a cohort of American women, active smokers have been shown to have more chronic musculoskeletal pain than former smokers or non-smokers [44].  [49]. The race was merely associated with stereotyped behavior and emotional characteristics [49]. However, as we progressed towards the twenty-fi rst century, our inquest into elucidating racial differences in pain has expanded. Nevertheless, our examination of racial differences in pain is tempered by various contemporary social factors including prevalent socioeconomic disparities. Underserved populations have less access to health care, and persistent medical problems such as pain may be undertreated and cause greater psychological and physical distress [50]. An analysis of a cohort of African American and Caucasian women with fi bromyalgia showed that African American women had more widespread pain as compared to Caucasian women, who were noted to have more tender points and pain intensity at these points. They also noted greater depressive symptoms in African American cohort compared to the Caucasian cohort [51].

Complex genetic differences in pain modulation in African
Americans versus Caucasians may also contribute to clinical disparities. African Americans have been shown to have more pain associated with chronic medical conditions than Caucasians [49]. Experimental laboratory studies have shown that African Americans have reduced pain tolerance for various modalities of pain, including extremes of temperature and pressure pain [52,53]. Altered pain regulatory mechanisms could account for clinical differences in subjective pain relief.
African-Americans showed less blood pressure response and change in norepinephrine levels after both rest and stress conditions compared to Caucasians [53].
The limitations of our study should be noted. Our study was retrospective in nature, relying on documentation of multiple providers. Our sample size was only 55 patients who received treatment at one tertiary care referral site. Although the sample size was suffi cient to reject the null hypothesis, it is inadequate to make generalizations to entire populations. Our study only contained three male patients, consistent with the known gender ratio of fi bromyalgia, preventing any meaningful analysis of gender effect.
Although we were not able to obtain the smoking status of 23 of the 55 patients, the impact of tobacco on outcome still had adequate power, and the results are statistically signifi cant.

Conclusions
A retrospective analysis of a cohort of 55 fi bromyalgia patients who received IV lidocaine shows a statistically signifi cant decrease in pain scores. A subgroup analysis shows that baseline pain scores are higher in smokers, and response to IV lidocaine is attenuated in smokers compared to nonsmokers. Moreover, race also seems to infl uence the response to IV lidocaine. Therefore, it is reasonable to stress smoking cessation in this population, and further research is needed to understand better the effect of tobacco use and race in the management of fi bromyalgia pain.