- Email: [email protected]
Elevated serum inflammatory markers in post-poliomyelitis syndrome
Journal of the Neurological Sciences 271 (2008) 80 – 86 www.elsevier.com/locate/jns
Elevated serum inflammatory markers in post-poliomyelitis syndrome Christopher B. Fordyce a , Donald Gagne a , Farzaneh Jalili a , Sudabeh Alatab a , Douglas L. Arnold a , Deborah Da Costa c , Stefan Sawoszczuk a , Caroline Bodner a , Stan Shapiro b,d , Jean-Paul Collet b,d,e , Ann Robinson d , Jean-Pierre Le Cruguel d , Yves Lapierre a , Amit Bar-Or a,⁎, Daria A. Trojan a,⁎ a
Department of Neurology and Neurosurgery, Montreal Neurological Institute and Hospital, Montreal, Quebec, Canada Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, Quebec, Canada Division of Clinical Epidemiology, Department of Medicine, McGill University Health Centre, McGill University, Montreal, Quebec, Canada d Randomized Clinical Trial Unit, Jewish General Hospital, McGill University, Montreal, Quebec, Canada e Department of Pediatrics, University of British Columbia, Vancouver, British Columbia, Canada b
c
Received 17 December 2007; received in revised form 7 February 2008; accepted 26 March 2008 Available online 12 May 2008
Abstract Objectives: To determine (i) whether serum inflammatory markers TNFα, IL-1β. IL-6, and leptin are increased in post-poliomyelitis syndrome (PPS) compared to healthy controls; and (ii) whether an association exists between elevated inflammatory markers and clinical parameters in PPS. The cause of PPS is unknown, but abnormal inflammatory responses have been implicated in several small studies. Methods: Serum inflammatory markers were measured (by Luminex) in 51 PPS patients and 26 normal controls. Clinical parameters assessed included disease duration, muscle strength (Medical Research Council sumscore), fatigue (Fatigue Severity Scale and Multidimensional Fatigue Inventory), and pain (visual analog scale scores). Results: In PPS, TNFα levels, as well as IL-6 and leptin were significantly increased compared to controls (Wilcoxon rank-sum test, p = 0.03 for TNFα, p = 0.03 for IL-6, p = 0.01 for leptin). The elevated TNFα levels in PPS were associated with increased pain due to illness (Spearman correlation coefficient r = 0.36, 95% C.I. 0.09 to 0.57) and specifically, with muscle pain (r = 0.38, 95% C.I. 0.11 to 0.59). There were no correlations between inflammatory markers in PPS and joint pain, muscle strength, fatigue, or disease duration. Conclusions: Serum TNFα, IL-6 and leptin levels are abnormally increased in PPS patients. Elevated TNFα levels appear to be specifically associated with increased muscle pain. © 2008 Elsevier B.V. All rights reserved. Keywords: Post-poliomyelitis syndrome; Cytokines; Leptin; Pain; Weakness; Fatigue
1. Introduction Post-poliomyelitis syndrome (PPS) is a common neurological disorder [1,2]. Muscle or joint pain occurs frequently, and is associated with reduced quality of life [3]. Limited ⁎ Corresponding authors. Montreal Neurological Institute and Hospital, 3801 University St., Montreal, Quebec, Canada H3A 2B4. Tel.: +1 514 398 8911, +1 514 398 8531; fax: +1 514 398 7371. E-mail addresses: [email protected] (A. Bar-Or), [email protected] (D.A. Trojan). 0022-510X/$ - see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2008.03.015
insight into PPS pathophysiology, may explain the lack of an approved pharmacological treatment [2]. Several small studies have suggested abnormal inflammatory responses in PPS pathophysiology [4–8]. Establishing such an association may set the stage for immune intervention trials, which is attractive given the growing cadre of immune modulating agents. Unlike multiple sclerosis (MS) [9], the evidence for inflammatory responses in PPS pathogenesis is less clear. Initial studies reported perivascular and parenchymal infiltrates and active gliosis in the spinal cord [4], oligoclonal
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
bands in the cerebrospinal fluid (CSF) [5], and signs of inflammation in affected muscles [6]. More recently, increased CSF mRNA levels of proinflammatory cytokines, similar to MS patients, were found in PPS patients [7]. Intravenous immunoglobulin produced a decrease in the increased mRNA levels of proinflammatory cytokines in the CSF [8]. This is significant since such cytokines can maintain an inflammatory cascade and continuous neuronal dysfunction [10]. In MS, the pro-inflammatory cytokines TNFα, IL-1β and IL-6 are abnormally produced in the inflamed CNS and serum levels may be elevated [9,11–14]. Expression of TNFα, IL-1β, and IL-6 is often co-regulated. For example, insulin and lipopolysacharide produce an increase in these cytokines [15,16]. Leptin, a hormone which regulates body weight, is also considered to be a pro-inflammatory cytokine belonging to the same family of long-chain helical cytokines as IL-6 [17,18]. To our knowledge, serum levels of inflammatory markers such as TNFα, IL-1β, IL-6, and leptin have not been assessed in PPS. The main aim of this study was to assess serum pro-inflammatory markers (TNFα, IL-1β, IL-6, and leptin) in PPS patients, compared to healthy controls. A secondary aim was to evaluate the association of elevated inflammatory markers with clinical parameters in PPS, including muscle strength, fatigue, and pain. 2. Methods 2.1. Study design and patient population Patients with PPS were recruited from a well-established Post-Polio Clinic at the Montreal Neurological Institute and Hospital, based on formal diagnostic criteria described below. A cohort of 31 normal controls without neurological disease were recruited from local volunteers (n = 29) and genetically unrelated family members of patients (n = 2). Since PPS patients were greater than 40 years of age, normal controls who were ≥ 40 years of age (n = 26) were included as a comparison group. Recruitment of study subjects occurred from October, 2002 to July, 2006. All study subjects provided their informed consent for participation in the study which was approved by the Montreal Neurological Institute and Hospital Research Ethics Board. PPS patient inclusion criteria were based on those proposed by the March of Dimes [19]: (i) a prior history of paralytic polio with evidence of motor neuron loss, as confirmed by history, neurological exam, and/or evidence of denervation on electromyography; (ii) a period of partial or complete functional recovery after the acute paralytic polio, followed by an interval (usually ≥ 15 years) of neurological stability; (iii) the new development of rapid onset or gradually progressive and persistent new muscle weakness and/or abnormal muscle fatiguability (decreased endurance), with or without generalized fatigue, muscle atrophy, or muscle and joint pain; (iv) the persistence of such new symptoms for at least 1 year; and (v) the exclusion of other
81
neurological, medical, and orthopedic problems as causes of these new symptoms. PPS patient exclusion criteria were the same as those reported in a recent study of fatigue in MS [20]. Fifty-four ambulatory PPS patients were screened for the study. Of these patients, three were excluded; one due to a forced vital capacity (FVC) of b 60%, one chose not to participate after the details of study procedures were explained, and one had extremely high values (off scale) for all cytokines, likely due to contamination of the serum sample. The final study population included 51 PPS patients. 2.2. Data collection and clinical outcome measures PPS patients participating in the study came for two study appointments. The first visit was a screening visit with completion of medical history (including disease duration), physical examination (performed by the same physician for most patients), assessment of BMI (weight in kg/(height in m)2), completion of screening blood tests, spirometry, and the Centers for Epidemiological Studies Depression Scale (CES-D) [21]. No clinically significant abnormalities were found on detailed blood work including complete blood count, erythrocyte sedimentation rate or C-reactive protein, electrolytes, glucose, blood urea nitrogen, creatinine, calcium, phosphorus, magnesium, total protein, albumin, total bilirubin, AST, ALT, LDH, alkaline phosphatase, creatine kinase, thyroid stimulating hormone, and T4. For PPS patients, we recorded the length of time since the acute polio infection, as well as PPS duration. Patients who met inclusion/exclusion criteria came for a second study visit, which generally occurred within 3 months of the screening visit. During the second visit, an interval medical history was elicited, and subjects completed the Fatigue Severity Scale (FSS; [22]) the Multidimensional Fatigue Inventory (MFI; [23]), and pain visual analog scales. Whole blood was collected from each subject at approximately the same time of day in a BD Vacutainer with no additive (BD BioScience, Mississauga, ON, Canada) and delivered immediately to the Neuroimmunology research laboratory for same day processing. Normal controls underwent the identical study procedures as patient cohorts, with the exception of physical and neurological examination, screening blood tests, and spirometry. Muscle strength was assessed by the examining physician using the Medical Research Council (MRC) sumscore. This measure is based on the well established 0 to 5 MRC scale for 12 muscle groups (bilateral arm abduction, forearm flexion, wrist extension, leg flexion, knee extension, and foot dorsal flexion). The final score calculated for all muscle groups tested could range from 0 to 60 (normal strength) [24]. Fatigue was assessed by two self-administered questionnaires, the FSS and the MFI. The FSS was developed to assess fatigue in neurologic and medical disorders, and is
82
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
reliable in PPS [25]. A score from 1 to 7 is obtained, with a higher score indicating more fatigue [22]. The MFI is a newer scale that was developed to assess the different components of fatigue. This measure can assess fatigue on five subscales, with a score of 4 to 20 on each, a higher score indicating more fatigue. The five subscales are General Fatigue, Physical Fatigue, Reduced Activity, Reduced Motivation, and Mental Fatigue [23]. The MFI has good psychometric properties, and has been validated in both healthy individuals and in certain chronic disease populations [26]. In this study, we focused on General Fatigue, Physical Fatigue, and Mental Fatigue because all three types of fatigue are reported in the PPS population [2]. Pain due to illness (PPS), and muscle and joint pain were all assessed with pain visual analog scale (VAS) scores, using a 0 to 10 scale, with a higher score indicating greater pain. 2.3. Serum processing and multiplex bead immunoassay Following phlebotomy, blood was allowed to clot, and then centrifuged at 310 ×g for 12 min. The serum was transferred to a new collection tube, centrifuged again, and aliquotted into cryovials for storage at − 80 °C. On the day of the assay, serum samples were thawed and transferred to 96-well plates for batched analysis to avoid inter-assay variability. Each sample was simultaneously analyzed by Luminex for concentrations of TNF-α, IL-1β, IL-6 and leptin according to the manufacturer's instructions (BioSource, Camarillo, CA, USA). Briefly, 100 µL of standard or test serum sample was combined with the beads in a 1:1 ratio. After 2 h of incubation at room temperature, the plates were washed and 100 µL of biotinylated detector antibody was added to each well. After 1 more hour of incubation, the plates were washed and 100 µL of streptavidin-RPE was added to each well followed by another 30 min of incubation. The beads were then resuspended in 100 µL wash solution and analyzed. The reading was performed in a 96-well plate format and analyzed with a Luminex LX100 instrument (Luminex, Austin, TX, USA), using StarStation 2.0 software (Applied Cytometry, Sacremento, CA, USA). A minimum of 50 beads per region were analyzed. A five-parameter logistic curve was applied to each standard curve and cytokine concentrations were interpolated from these curves.
matory marker values, the data was still not normally distributed, and therefore non-parametric statistical methods were used. The non-parametric Wilcoxon rank-sum test (two-sided) was utilized for comparison of inflammatory markers between cohorts, and unpaired t-tests and Chisquared tests were used for comparison of other clinical parameters between patients and controls. Spearman correlation coefficients and their 95% confidence intervals were calculated to assess the association between inflammatory markers and clinical parameters in PPS patients, and to evaluate the association between leptin and BMI. We did not correct for multiple comparisons in our planned primary analysis of the results presented in the Tables due to the exploratory nature of this study [27]. We performed a secondary analysis, in which we accounted for multiple comparisons by correcting p-values from a one sided Wilcoxon rank-sum test (patients would not be expected to have lower levels of inflammatory markers than controls) using the method described by Benjamini and Hochberg. This technique controls strongly for the false discovery rate, and is usually more powerful than other adjustments such as Bonferonni [28]. 3. Results 3.1. Characteristics of study subjects PPS patients were similar to their healthy control cohort in terms of age (PPS patients were 59.5 ± 7.3 years and controls were 58.0 ± 11 years of age) and gender (19/51 (37%) of PPS patients were male and 10/26 (38%) of controls were male). For PPS patients, average age at acute polio was 4.3 ± 4.1 years, average time since acute polio was 55.2 ± 7.2 years, and average PPS duration was 12.9 ± 6.9 years. 49/52 (94%) of PPS patients had new weakness since recovery from acute polio, while 47/52 (90%) reported new muscle fatigue in addition to weakness. 3.2. Association of leptin with body mass index (BMI) To validate our leptin measurements, we evaluated the association of leptin with BMI in PPS patients. We found that elevated leptin was correlated with increased BMI: Spearman correlation coefficient r = 0.57 (95% confidence interval 0.34 to 0.73, p b 0.0001).
2.4. Statistical analysis
3.3. Inflammatory markers in patients and controls
The sample concentrations of inflammatory markers were transformed to natural log scores given the non-normal distribution of data and to minimize the effect of outliers. When inflammatory marker levels were less than detectable (under curve), a level was imputed by adding a small random number to half the minimum observed value before the log transformation (final value was less than the minimum observed value). Despite log transformation of the inflam-
The four inflammatory markers assessed in our study were compared between patients and normal controls in Table 1 and Fig. 1. In the primary analyses, TNFα (p = 0.03), IL-6 (p = 0.03), and leptin (p = 0.01) were significantly increased in PPS patients compared to controls, with a trend to increased levels of IL-1β (p = 0.07). Similar results were obtained in our secondary analysis when adjusting for multiple comparisons. TNFα (p = 0.03), IL-1β (p = 0.03), IL-6 (p = 0.02), and leptin
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86 Table 1 Comparison of serum inflammatory markers between PPS patients and normal controls Inflammatory
PPS
Controls
Marker
(n = 51)
(n = 26)
Log TNFα (pg/ml) Log IL-1β (pg/ml) Log IL-6 (pg/ml) Log Leptin (pg/ml)
2.10 ± 1.39 2.27 ± 2.69 4.98 ± 0.64 8.05 ± 1.01
1.08 ± 1.85 1.22 ± 2.54 4.67 ± 0.52 7.47 ± 0.77
PValue 0.03 0.07 0.03 0.01
Legend: Natural log transformed values of inflammatory markers are presented as means ± standard deviations. P-values are for a two-sided Wilcoxon rank-sum test. PPS = post-poliomyelitis syndrome.
(p = 0.03) were significantly increased in PPS patients compared to controls. The mean difference between PPS patients and controls for TNFα and IL-1β was approximately 1 on the natural log scale, indicating that both cytokine levels were more than double (by a factor of 2.718) in patients compared to
83
controls. The mean difference between PPS patients and controls for IL-6 was 0.31, or was greater in PPS patients compared to controls by a factor of 1.36, and for leptin was 0.58, or greater in PPS patients than controls by a factor of 1.79. 3.4. Association of inflammatory markers with clinical parameters in PPS patients We assessed the correlation of the four inflammatory markers with clinical parameters in PPS. Elevated TNFα was associated with increased overall pain due to illness (r = 0.36, 95% C.I. 0.09 to 0.57), and specifically with muscle pain (r = 0.38, 95% C.I. 0.11 to 0.59) (Fig. 2). There was a suggestion of an association between elevated TNFα and reduced muscle strength (r = − 0.19, 95% C.I. − 0.44 to 0.09). A suggestion of an association between elevated leptin and increased muscle pain (r = 0.26, 95% C.I. − 0.02 to 0.49), and overall pain due to illness (r = 0.23, 95% C.I. − 0.05 to 0.47)
Fig. 1. Plots of means and standard deviations of serum inflammatory markers in post-poliomyelitis syndrome patients compared to controls. PPS = postpoliomyelitis syndrome.
84
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
Fig. 2. Scatterplots of pain due to illness and muscle pain plotted against TNFα. Regression lines and Spearman correlation coefficients are presented. ⁎ P-values have not been corrected for multiple comparisons.
was also seen. There was no association between inflammatory markers and age, disease duration, or fatigue. 4. Discussion We demonstrate, for the first time, the presence of increased levels of circulating pro-inflammatory markers in the serum of patients with PPS. Specifically, PPS patients had significantly increased levels of TNFα, IL-6 and leptin when compared to normal controls of similar age and gender. We also found a suggestion of increased levels of IL-1β in PPS patients. In a secondary analysis, we found that elevated TNFα levels were associated with increased muscle pain. Our finding of increased levels of serum TNFα, IL-6, and leptin in PPS patients, decades after their acute disease, warrants consideration of a role for inflammation in the ongoing PPS disease process. In our PPS population, 94% had new weakness since resolution of acute polio, while 90% reported new muscle fatigue in addition to weakness, with a mean duration of PPS of 12.9 years, suggesting an ongoing disease state. Until recently, PPS has generally not been considered an immune-mediated disorder [2]. In contrast, TNFα and IL-6 have been implicated in MS neuroinflammation [9], and more recently, in other neurological conditions including amyotrophic lateral sclerosis [13] and Alzheimer's disease [11]. It is well accepted that these cytokines may contribute to the activation of microglia and macrophages, which in turn, could damage neurons through the release of reactive oxygen and nitrogen species, and of glutamate, among others [13,29,30]. TNFα, IL-1β, and IL-6 are also known to be directly toxic to neurons [31]. PPS patients exhibited multiple cytokine elevations. This may reflect a more global peripheral immune activation in PPS, and if appropriate longitudinal studies are performed, may represent a reasonable target for peripherally administered immune modulating agents. To our knowledge, our study is the first to implicate abnormal levels of circulating leptin in PPS. Leptin, originally found to be associated with satiety, can upregulate TNFα and IL-6, as well as IL-10 production by peripheral blood monocytes during the active phase of MS [18].
Therefore, increased leptin levels in PPS patients may reflect the presence of active ongoing inflammation. It is not known whether potential sites of ongoing inflammation in PPS may involve the CNS or the peripheral compartment, or whether simple peripheral measures of inflammation might serve as useful markers of inflammatory disease activity. Evidence of chronic inflammation is supported by other groups. One study has demonstrated, increased TNFα and IFN-γ mRNA transcripts in mononuclear cells from the CSF of 13 PPS patients. Of note, such abnormalities were not detected in the circulation of patients [7]. Administration of intravenous immunoglobulin decreased the abnormal cytokine mRNA expression in 16 PPS patients [8]. Our study extends these prior observations in a larger cohort and with detection of abnormal elevations in the translated protein of TNFα as well as IL-6 and leptin, in the circulation of PPS patients. In the case of TNFα, the blood-brain barrier is permeable to this molecule in both the injured and non-injured state [32,33]. Future longitudinal studies could potentially demonstrate the use of serum cytokines as biomarkers to assess target organ inflammation or to predict certain clinical parameters, such as pain due to PPS. The question of how inflammation may contribute to PPS, decades after acute poliovirus infection, and its relationship with ongoing symptoms in patients, is raised. The presence of poliovirus genomic sequences within the CSF has been reported in PPS patients [10,34]. While not active or having a direct ability to cause PPS symptoms due to restrained RNA synthesis and lack of evidence for direct toxicity, these poliovirus genomic particles have the potential to induce the production of cytokines, such as TNFα, and thereby contribute to gradually worsening chronic inflammation [35]. Other groups have not found evidence for poliovirus persistence [36,37]. This has led to other hypotheses for persistently elevated cytokines to include a poliovirus-induced autoimmune response directed against as of yet unknown neuronal or non-neuronal autoantigens, or an immune response that is completely secondary to CNS damage [10]. After demonstrating elevated levels of serum TNFα, IL-6, and leptin in PPS patients, we then performed a secondary
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
analysis to explore potential relationships between these cytokines and clinical parameters in PPS. Both muscle and joint pain occur in most PPS patients, and are associated with a significant reduction in quality of life [3]. Joint pain is likely caused by joint and soft tissue abnormalities, related to chronic abnormal use from weakness secondary to acute polio. The origin of muscular pain, is unclear, but has not been associated with greater weakness at acute polio, current muscle strength, and increased serum creatine kinase (CK) levels [3]. Recent studies have increasingly implicated proinflammatory cytokines in pathways of pain perception [38], including both peripheral and central neuropathy [39]. Ours is the first study to report an association in PPS patients between elevated TNFα levels and increased pain, specifically that of muscular origin. BMI may be confounder in this relationship because heavier patients may be more likely to have pain due to overuse. However, our finding is consistent with experimental evidence in rats, where TNFα has direct effects on nociception, and injured nerve fibers are sensitized to the excitatory effects of TNFα [40]. In a mouse model of chronic nerve injury, inhibition of TNFα receptors resulted in a reduction in thermal analgesia and mechanical allodynia [41]. Our findings are supported by a recent multi-center randomized controlled trial which reported that in a secondary analysis, intravenous immunoglobulin decreased pain in PPS patients. This suggests a link between pain and inflammation, however inflammatory markers were not assessed [42]. Our study is cross-sectional, and we do not wish to overstate potential causal relationships or links between inflammatory markers and PPS symptoms. We also acknowledge that circulating cytokines may be subject to circadian variation that could differ across groups, and therefore longitudinal studies with repeat serum sampling may provide further insights into such biological variability. Nonetheless, our PPS cohort represents one of the largest ones studied to date, and we were able to detect several abnormalities in inflammatory markers in the circulation of patients. We also found a specific association between TNFα and muscle pain. Together, our results suggest that inflammation may play a role in PPS. These results prompt further investigation into the importance of inflammatory mediators in the pathophysiology and symptomatology of PPS. Such follow-up studies are particularly relevant in view of the lack of specific treatments for this debilitating condition and the continued endemic existence of poliovirus in some areas of the world [43], at a time that a growing range of novel immune modulating therapies is under active development. Acknowledgements The authors acknowledge the help of Dr. Diane Diorio, Montreal Neurological Hospital, in patient recruitment for the study. We are grateful to Boli Fan from Dr. Bar-Or's research laboratory for her help in processing samples for the study. We are indebted to the patients and controls in this
85
study who donated their time to participate. The study was supported by the Multiple Sclerosis Society of Canada and the Montreal Neurological Institute. References [1] Jubelt B, Cashman NR. Neurological manifestations of the post-polio syndrome. Crit Rev Neurobiol 1987;3:199–220. [2] Trojan DA, Cashman NR. Post-poliomyelitis syndrome. Muscle Nerve 2005;31:6–19. [3] Vasiliadis HM, Collet JP, Shapiro S, Venturini A, Trojan DA. Predictive factors and correlates for pain in postpoliomyelitis syndrome patients. Arch Phys Med Rehabil 2002;83:1109–15. [4] Pezeshkpour GH, Dalakas MC. Long-term changes in the spinal cords of patients with old poliomyelitis. Signs of continuous disease activity. Arch Neurol 1988;45:505–8. [5] Dalakas MC, Elder G, Hallett M, Ravits J, Baker M, Papadopoulos N, et al. A long-term follow-up study of patients with post-poliomyelitis neuromuscular symptoms. N Engl J Med 1986;314:959–63. [6] Dalakas MC. Morphologic changes in the muscles of patients with postpoliomyelitis neuromuscular symptoms. Neurology 1988;38: 99–104. [7] Gonzalez H, Khademi M, Andersson M, Wallstrom E, Borg K, Olsson T. Prior poliomyelitis-evidence of cytokine production in the central nervous system. J Neurol Sci 2002;205:9–13. [8] Gonzalez H, Khademi M, Andersson M, Piehl F, Wallstrom E, Borg K, et al. Prior poliomyelitis-IVIg treatment reduces proinflammatory cytokine production. J Neuroimmunol 2004;150:139–44. [9] A Bar-Or. Immunology of multiple sclerosis. Neurol Clin 2005; 23:149–175, vii. [10] Dalakas MC. Pro-inflammatory cytokines and motor neuron dysfunction: is there a connection in post-polio syndrome? J Neurol Sci 2002;205:5–8. [11] Wyss-Coray T. Inflammation in Alzheimer disease: driving force, bystander or beneficial response? Nat Med 2006;12:1005–15. [12] Orr CF, Rowe DB, Mizuno Y, Mori H, Halliday GM. A possible role for humoral immunity in the pathogenesis of Parkinson's disease. Brain 2005;128:2665–74. [13] Wu DC, Re DB, Nagai M, Ischiropoulos H, Przedborski S. The inflammatory NADPH oxidase enzyme modulates motor neuron degeneration in amyotrophic lateral sclerosis mice. Proc Natl Acad Sci U S A 2006;103:12132–7. [14] Martino G, Furlan R, Brambilla E, Bergami A, Ruffini F, Gironi M, et al. Cytokines and immunity in multiple sclerosis: the dual signal hypothesis. J Neuroimmunol 2000;109:3–9. [15] Soop M, Duxbury H, Agwunobi AO, Gibson JM, Hopkins SJ, Childs C, et al. Euglycemic hyperinsulinemia augments the cytokine and endocrine responses to endotoxin in humans. Am J Physiol Endocrinol Metab 2002;282:E1276–85. [16] Watson GS, Craft S. Insulin resistance, inflammation, and cognition in Alzheimer's Disease: lessons for multiple sclerosis. J Neurol Sci 2006;245:21–33. [17] La Cava A, Matarese G. The weight of leptin in immunity. Nat Rev Immunol 2004;4:371–9. [18] Frisullo G, Angelucci F, Mirabella M, Caggiula M, Patanella K, Nociti V, et al. Leptin enhances the release of cytokines by peripheral blood mononuclear cells from relapsing multiple sclerosis patients. J Clin Immunol 2004;24:287–93. [19] March of Dimes. Post-polio syndrome: identifying best practices in diagnosis and care. White Plains: March of Dimes Birth Defects Foundation; 2001 (www.modimes.org). [20] Trojan DA, Arnold D, Collet JP, Shapiro S, Bar-Or A, Robinson A, et al. Fatigue in multiple sclerosis: association with disease-related, behavioural and psychosocial factors. Mult Scler 2007;13:985–95. [21] Radloff LS. A self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385–401.
86
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
[22] Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD. The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol 1989;46:1121–3. [23] Smets EM, Garssen B, Bonke B, De Haes JC. The Multidimensional Fatigue Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res 1995;39:315–25. [24] Kleyweg RP, van der Meche FG, Schmitz PI. Interobserver agreement in the assessment of muscle strength and functional abilities in Guillain–Barre syndrome. Muscle Nerve 1991;14:1103–9. [25] Finch LE, Venturini A, Mayo NE, Trojan DA. Effort-limited treadmill walk test: reliability and validity in subjects with postpolio syndrome. Am J Phys Med Rehabil 2004;83:613–23. [26] Schneider RA. Reliability and validity of the Multidimensional Fatigue Inventory (MFI-20) and the Rhoten Fatigue Scale among rural cancer outpatients. Cancer Nurs 1998;21:370–3. [27] Bender R, Lange S. Adjusting for multiple testing—when and how? J Clin Epidemiol 2001;54:343–9. [28] Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 1995;57:289–300. [29] Smith T, Groom A, Zhu B, Turski L. Autoimmune encephalomyelitis ameliorated by AMPA antagonists. Nat Med 2000;6:62–6. [30] Fordyce CB, Jagasia R, Zhu X, Schlichter LC. Microglia Kv1.3 channels contribute to their ability to kill neurons. J Neurosci 2005;25:7139–49. [31] Kim YS, Joh TH. Microglia, major player in the brain inflammation: their roles in the pathogenesis of Parkinson's disease. Exp Mol Med 2006;38:333–47. [32] Pan W, Banks WA, Kastin AJ. Permeability of the blood-brain and blood-spinal cord barriers to interferons. J Neuroimmunol 1997;76: 105–11. [33] Pan W, Banks WA, Kastin AJ. Blood-brain barrier permeability to ebiratide and TNF in acute spinal cord injury. Exp Neurol 1997;146: 367–73.
[34] Leon-Monzon ME, Dalakas MC. Detection of poliovirus antibodies and poliovirus genome in patients with the post-polio syndrome. Ann N Y Acad Sci 1995;753:208–18. [35] Colbere-Garapin F, Duncan G, Pavio N, Pelletier I, Petit I. An approach to understanding the mechanisms of poliovirus persistence in infected cells of neural or non-neural origin. Clin Diagn Virol 1998;9: 107–13. [36] Melchers W, de Visser M, Jongen P, van Loon A, Nibbeling R, Oostvogel P, et al. The postpolio syndrome: no evidence for poliovirus persistence. Ann Neurol 1992;32:728–32. [37] Salazar-Grueso EF, Grimaldi LM, Roos RP, Variakojis R, Jubelt B, Cashman NR. Isoelectric focusing studies of serum and cerebrospinal fluid in patients with antecedent poliomyelitis. Ann Neurol 1989;26: 709–13. [38] Sommer C, Kress M. Recent findings on how proinflammatory cytokines cause pain: peripheral mechanisms in inflammatory and neuropathic hyperalgesia. Neurosci Lett 2004;361:184–7. [39] Marchand F, Perretti M, McMahon SB. Role of the immune system in chronic pain. Nat Rev Neurosci 2005;6:521–32. [40] Schafers M, Lee DH, Brors D, Yaksh TL, Sorkin LS. Increased sensitivity of injured and adjacent uninjured rat primary sensory neurons to exogenous tumor necrosis factor-alpha after spinal nerve ligation. J Neurosci 2003;23:3028–38. [41] Sommer C, Schmidt C, George A. Hyperalgesia in experimental neuropathy is dependent on the TNF receptor 1. Exp Neurol 1998;151: 138–42. [42] Gonzalez H, Sunnerhagen KS, Sjoberg I, Kaponides G, Olsson T, Borg K. Intravenous immunoglobulin for post-polio syndrome: a randomised controlled trial. Lancet Neurol 2006;5:493–500. [43] Howard RS. Poliomyelitis and the postpolio syndrome. BMJ 2005;330:1314–8.
Elevated serum inflammatory markers in post-poliomyelitis syndrome Christopher B. Fordyce a , Donald Gagne a , Farzaneh Jalili a , Sudabeh Alatab a , Douglas L. Arnold a , Deborah Da Costa c , Stefan Sawoszczuk a , Caroline Bodner a , Stan Shapiro b,d , Jean-Paul Collet b,d,e , Ann Robinson d , Jean-Pierre Le Cruguel d , Yves Lapierre a , Amit Bar-Or a,⁎, Daria A. Trojan a,⁎ a
Department of Neurology and Neurosurgery, Montreal Neurological Institute and Hospital, Montreal, Quebec, Canada Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, Quebec, Canada Division of Clinical Epidemiology, Department of Medicine, McGill University Health Centre, McGill University, Montreal, Quebec, Canada d Randomized Clinical Trial Unit, Jewish General Hospital, McGill University, Montreal, Quebec, Canada e Department of Pediatrics, University of British Columbia, Vancouver, British Columbia, Canada b
c
Received 17 December 2007; received in revised form 7 February 2008; accepted 26 March 2008 Available online 12 May 2008
Abstract Objectives: To determine (i) whether serum inflammatory markers TNFα, IL-1β. IL-6, and leptin are increased in post-poliomyelitis syndrome (PPS) compared to healthy controls; and (ii) whether an association exists between elevated inflammatory markers and clinical parameters in PPS. The cause of PPS is unknown, but abnormal inflammatory responses have been implicated in several small studies. Methods: Serum inflammatory markers were measured (by Luminex) in 51 PPS patients and 26 normal controls. Clinical parameters assessed included disease duration, muscle strength (Medical Research Council sumscore), fatigue (Fatigue Severity Scale and Multidimensional Fatigue Inventory), and pain (visual analog scale scores). Results: In PPS, TNFα levels, as well as IL-6 and leptin were significantly increased compared to controls (Wilcoxon rank-sum test, p = 0.03 for TNFα, p = 0.03 for IL-6, p = 0.01 for leptin). The elevated TNFα levels in PPS were associated with increased pain due to illness (Spearman correlation coefficient r = 0.36, 95% C.I. 0.09 to 0.57) and specifically, with muscle pain (r = 0.38, 95% C.I. 0.11 to 0.59). There were no correlations between inflammatory markers in PPS and joint pain, muscle strength, fatigue, or disease duration. Conclusions: Serum TNFα, IL-6 and leptin levels are abnormally increased in PPS patients. Elevated TNFα levels appear to be specifically associated with increased muscle pain. © 2008 Elsevier B.V. All rights reserved. Keywords: Post-poliomyelitis syndrome; Cytokines; Leptin; Pain; Weakness; Fatigue
1. Introduction Post-poliomyelitis syndrome (PPS) is a common neurological disorder [1,2]. Muscle or joint pain occurs frequently, and is associated with reduced quality of life [3]. Limited ⁎ Corresponding authors. Montreal Neurological Institute and Hospital, 3801 University St., Montreal, Quebec, Canada H3A 2B4. Tel.: +1 514 398 8911, +1 514 398 8531; fax: +1 514 398 7371. E-mail addresses: [email protected] (A. Bar-Or), [email protected] (D.A. Trojan). 0022-510X/$ - see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2008.03.015
insight into PPS pathophysiology, may explain the lack of an approved pharmacological treatment [2]. Several small studies have suggested abnormal inflammatory responses in PPS pathophysiology [4–8]. Establishing such an association may set the stage for immune intervention trials, which is attractive given the growing cadre of immune modulating agents. Unlike multiple sclerosis (MS) [9], the evidence for inflammatory responses in PPS pathogenesis is less clear. Initial studies reported perivascular and parenchymal infiltrates and active gliosis in the spinal cord [4], oligoclonal
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
bands in the cerebrospinal fluid (CSF) [5], and signs of inflammation in affected muscles [6]. More recently, increased CSF mRNA levels of proinflammatory cytokines, similar to MS patients, were found in PPS patients [7]. Intravenous immunoglobulin produced a decrease in the increased mRNA levels of proinflammatory cytokines in the CSF [8]. This is significant since such cytokines can maintain an inflammatory cascade and continuous neuronal dysfunction [10]. In MS, the pro-inflammatory cytokines TNFα, IL-1β and IL-6 are abnormally produced in the inflamed CNS and serum levels may be elevated [9,11–14]. Expression of TNFα, IL-1β, and IL-6 is often co-regulated. For example, insulin and lipopolysacharide produce an increase in these cytokines [15,16]. Leptin, a hormone which regulates body weight, is also considered to be a pro-inflammatory cytokine belonging to the same family of long-chain helical cytokines as IL-6 [17,18]. To our knowledge, serum levels of inflammatory markers such as TNFα, IL-1β, IL-6, and leptin have not been assessed in PPS. The main aim of this study was to assess serum pro-inflammatory markers (TNFα, IL-1β, IL-6, and leptin) in PPS patients, compared to healthy controls. A secondary aim was to evaluate the association of elevated inflammatory markers with clinical parameters in PPS, including muscle strength, fatigue, and pain. 2. Methods 2.1. Study design and patient population Patients with PPS were recruited from a well-established Post-Polio Clinic at the Montreal Neurological Institute and Hospital, based on formal diagnostic criteria described below. A cohort of 31 normal controls without neurological disease were recruited from local volunteers (n = 29) and genetically unrelated family members of patients (n = 2). Since PPS patients were greater than 40 years of age, normal controls who were ≥ 40 years of age (n = 26) were included as a comparison group. Recruitment of study subjects occurred from October, 2002 to July, 2006. All study subjects provided their informed consent for participation in the study which was approved by the Montreal Neurological Institute and Hospital Research Ethics Board. PPS patient inclusion criteria were based on those proposed by the March of Dimes [19]: (i) a prior history of paralytic polio with evidence of motor neuron loss, as confirmed by history, neurological exam, and/or evidence of denervation on electromyography; (ii) a period of partial or complete functional recovery after the acute paralytic polio, followed by an interval (usually ≥ 15 years) of neurological stability; (iii) the new development of rapid onset or gradually progressive and persistent new muscle weakness and/or abnormal muscle fatiguability (decreased endurance), with or without generalized fatigue, muscle atrophy, or muscle and joint pain; (iv) the persistence of such new symptoms for at least 1 year; and (v) the exclusion of other
81
neurological, medical, and orthopedic problems as causes of these new symptoms. PPS patient exclusion criteria were the same as those reported in a recent study of fatigue in MS [20]. Fifty-four ambulatory PPS patients were screened for the study. Of these patients, three were excluded; one due to a forced vital capacity (FVC) of b 60%, one chose not to participate after the details of study procedures were explained, and one had extremely high values (off scale) for all cytokines, likely due to contamination of the serum sample. The final study population included 51 PPS patients. 2.2. Data collection and clinical outcome measures PPS patients participating in the study came for two study appointments. The first visit was a screening visit with completion of medical history (including disease duration), physical examination (performed by the same physician for most patients), assessment of BMI (weight in kg/(height in m)2), completion of screening blood tests, spirometry, and the Centers for Epidemiological Studies Depression Scale (CES-D) [21]. No clinically significant abnormalities were found on detailed blood work including complete blood count, erythrocyte sedimentation rate or C-reactive protein, electrolytes, glucose, blood urea nitrogen, creatinine, calcium, phosphorus, magnesium, total protein, albumin, total bilirubin, AST, ALT, LDH, alkaline phosphatase, creatine kinase, thyroid stimulating hormone, and T4. For PPS patients, we recorded the length of time since the acute polio infection, as well as PPS duration. Patients who met inclusion/exclusion criteria came for a second study visit, which generally occurred within 3 months of the screening visit. During the second visit, an interval medical history was elicited, and subjects completed the Fatigue Severity Scale (FSS; [22]) the Multidimensional Fatigue Inventory (MFI; [23]), and pain visual analog scales. Whole blood was collected from each subject at approximately the same time of day in a BD Vacutainer with no additive (BD BioScience, Mississauga, ON, Canada) and delivered immediately to the Neuroimmunology research laboratory for same day processing. Normal controls underwent the identical study procedures as patient cohorts, with the exception of physical and neurological examination, screening blood tests, and spirometry. Muscle strength was assessed by the examining physician using the Medical Research Council (MRC) sumscore. This measure is based on the well established 0 to 5 MRC scale for 12 muscle groups (bilateral arm abduction, forearm flexion, wrist extension, leg flexion, knee extension, and foot dorsal flexion). The final score calculated for all muscle groups tested could range from 0 to 60 (normal strength) [24]. Fatigue was assessed by two self-administered questionnaires, the FSS and the MFI. The FSS was developed to assess fatigue in neurologic and medical disorders, and is
82
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
reliable in PPS [25]. A score from 1 to 7 is obtained, with a higher score indicating more fatigue [22]. The MFI is a newer scale that was developed to assess the different components of fatigue. This measure can assess fatigue on five subscales, with a score of 4 to 20 on each, a higher score indicating more fatigue. The five subscales are General Fatigue, Physical Fatigue, Reduced Activity, Reduced Motivation, and Mental Fatigue [23]. The MFI has good psychometric properties, and has been validated in both healthy individuals and in certain chronic disease populations [26]. In this study, we focused on General Fatigue, Physical Fatigue, and Mental Fatigue because all three types of fatigue are reported in the PPS population [2]. Pain due to illness (PPS), and muscle and joint pain were all assessed with pain visual analog scale (VAS) scores, using a 0 to 10 scale, with a higher score indicating greater pain. 2.3. Serum processing and multiplex bead immunoassay Following phlebotomy, blood was allowed to clot, and then centrifuged at 310 ×g for 12 min. The serum was transferred to a new collection tube, centrifuged again, and aliquotted into cryovials for storage at − 80 °C. On the day of the assay, serum samples were thawed and transferred to 96-well plates for batched analysis to avoid inter-assay variability. Each sample was simultaneously analyzed by Luminex for concentrations of TNF-α, IL-1β, IL-6 and leptin according to the manufacturer's instructions (BioSource, Camarillo, CA, USA). Briefly, 100 µL of standard or test serum sample was combined with the beads in a 1:1 ratio. After 2 h of incubation at room temperature, the plates were washed and 100 µL of biotinylated detector antibody was added to each well. After 1 more hour of incubation, the plates were washed and 100 µL of streptavidin-RPE was added to each well followed by another 30 min of incubation. The beads were then resuspended in 100 µL wash solution and analyzed. The reading was performed in a 96-well plate format and analyzed with a Luminex LX100 instrument (Luminex, Austin, TX, USA), using StarStation 2.0 software (Applied Cytometry, Sacremento, CA, USA). A minimum of 50 beads per region were analyzed. A five-parameter logistic curve was applied to each standard curve and cytokine concentrations were interpolated from these curves.
matory marker values, the data was still not normally distributed, and therefore non-parametric statistical methods were used. The non-parametric Wilcoxon rank-sum test (two-sided) was utilized for comparison of inflammatory markers between cohorts, and unpaired t-tests and Chisquared tests were used for comparison of other clinical parameters between patients and controls. Spearman correlation coefficients and their 95% confidence intervals were calculated to assess the association between inflammatory markers and clinical parameters in PPS patients, and to evaluate the association between leptin and BMI. We did not correct for multiple comparisons in our planned primary analysis of the results presented in the Tables due to the exploratory nature of this study [27]. We performed a secondary analysis, in which we accounted for multiple comparisons by correcting p-values from a one sided Wilcoxon rank-sum test (patients would not be expected to have lower levels of inflammatory markers than controls) using the method described by Benjamini and Hochberg. This technique controls strongly for the false discovery rate, and is usually more powerful than other adjustments such as Bonferonni [28]. 3. Results 3.1. Characteristics of study subjects PPS patients were similar to their healthy control cohort in terms of age (PPS patients were 59.5 ± 7.3 years and controls were 58.0 ± 11 years of age) and gender (19/51 (37%) of PPS patients were male and 10/26 (38%) of controls were male). For PPS patients, average age at acute polio was 4.3 ± 4.1 years, average time since acute polio was 55.2 ± 7.2 years, and average PPS duration was 12.9 ± 6.9 years. 49/52 (94%) of PPS patients had new weakness since recovery from acute polio, while 47/52 (90%) reported new muscle fatigue in addition to weakness. 3.2. Association of leptin with body mass index (BMI) To validate our leptin measurements, we evaluated the association of leptin with BMI in PPS patients. We found that elevated leptin was correlated with increased BMI: Spearman correlation coefficient r = 0.57 (95% confidence interval 0.34 to 0.73, p b 0.0001).
2.4. Statistical analysis
3.3. Inflammatory markers in patients and controls
The sample concentrations of inflammatory markers were transformed to natural log scores given the non-normal distribution of data and to minimize the effect of outliers. When inflammatory marker levels were less than detectable (under curve), a level was imputed by adding a small random number to half the minimum observed value before the log transformation (final value was less than the minimum observed value). Despite log transformation of the inflam-
The four inflammatory markers assessed in our study were compared between patients and normal controls in Table 1 and Fig. 1. In the primary analyses, TNFα (p = 0.03), IL-6 (p = 0.03), and leptin (p = 0.01) were significantly increased in PPS patients compared to controls, with a trend to increased levels of IL-1β (p = 0.07). Similar results were obtained in our secondary analysis when adjusting for multiple comparisons. TNFα (p = 0.03), IL-1β (p = 0.03), IL-6 (p = 0.02), and leptin
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86 Table 1 Comparison of serum inflammatory markers between PPS patients and normal controls Inflammatory
PPS
Controls
Marker
(n = 51)
(n = 26)
Log TNFα (pg/ml) Log IL-1β (pg/ml) Log IL-6 (pg/ml) Log Leptin (pg/ml)
2.10 ± 1.39 2.27 ± 2.69 4.98 ± 0.64 8.05 ± 1.01
1.08 ± 1.85 1.22 ± 2.54 4.67 ± 0.52 7.47 ± 0.77
PValue 0.03 0.07 0.03 0.01
Legend: Natural log transformed values of inflammatory markers are presented as means ± standard deviations. P-values are for a two-sided Wilcoxon rank-sum test. PPS = post-poliomyelitis syndrome.
(p = 0.03) were significantly increased in PPS patients compared to controls. The mean difference between PPS patients and controls for TNFα and IL-1β was approximately 1 on the natural log scale, indicating that both cytokine levels were more than double (by a factor of 2.718) in patients compared to
83
controls. The mean difference between PPS patients and controls for IL-6 was 0.31, or was greater in PPS patients compared to controls by a factor of 1.36, and for leptin was 0.58, or greater in PPS patients than controls by a factor of 1.79. 3.4. Association of inflammatory markers with clinical parameters in PPS patients We assessed the correlation of the four inflammatory markers with clinical parameters in PPS. Elevated TNFα was associated with increased overall pain due to illness (r = 0.36, 95% C.I. 0.09 to 0.57), and specifically with muscle pain (r = 0.38, 95% C.I. 0.11 to 0.59) (Fig. 2). There was a suggestion of an association between elevated TNFα and reduced muscle strength (r = − 0.19, 95% C.I. − 0.44 to 0.09). A suggestion of an association between elevated leptin and increased muscle pain (r = 0.26, 95% C.I. − 0.02 to 0.49), and overall pain due to illness (r = 0.23, 95% C.I. − 0.05 to 0.47)
Fig. 1. Plots of means and standard deviations of serum inflammatory markers in post-poliomyelitis syndrome patients compared to controls. PPS = postpoliomyelitis syndrome.
84
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
Fig. 2. Scatterplots of pain due to illness and muscle pain plotted against TNFα. Regression lines and Spearman correlation coefficients are presented. ⁎ P-values have not been corrected for multiple comparisons.
was also seen. There was no association between inflammatory markers and age, disease duration, or fatigue. 4. Discussion We demonstrate, for the first time, the presence of increased levels of circulating pro-inflammatory markers in the serum of patients with PPS. Specifically, PPS patients had significantly increased levels of TNFα, IL-6 and leptin when compared to normal controls of similar age and gender. We also found a suggestion of increased levels of IL-1β in PPS patients. In a secondary analysis, we found that elevated TNFα levels were associated with increased muscle pain. Our finding of increased levels of serum TNFα, IL-6, and leptin in PPS patients, decades after their acute disease, warrants consideration of a role for inflammation in the ongoing PPS disease process. In our PPS population, 94% had new weakness since resolution of acute polio, while 90% reported new muscle fatigue in addition to weakness, with a mean duration of PPS of 12.9 years, suggesting an ongoing disease state. Until recently, PPS has generally not been considered an immune-mediated disorder [2]. In contrast, TNFα and IL-6 have been implicated in MS neuroinflammation [9], and more recently, in other neurological conditions including amyotrophic lateral sclerosis [13] and Alzheimer's disease [11]. It is well accepted that these cytokines may contribute to the activation of microglia and macrophages, which in turn, could damage neurons through the release of reactive oxygen and nitrogen species, and of glutamate, among others [13,29,30]. TNFα, IL-1β, and IL-6 are also known to be directly toxic to neurons [31]. PPS patients exhibited multiple cytokine elevations. This may reflect a more global peripheral immune activation in PPS, and if appropriate longitudinal studies are performed, may represent a reasonable target for peripherally administered immune modulating agents. To our knowledge, our study is the first to implicate abnormal levels of circulating leptin in PPS. Leptin, originally found to be associated with satiety, can upregulate TNFα and IL-6, as well as IL-10 production by peripheral blood monocytes during the active phase of MS [18].
Therefore, increased leptin levels in PPS patients may reflect the presence of active ongoing inflammation. It is not known whether potential sites of ongoing inflammation in PPS may involve the CNS or the peripheral compartment, or whether simple peripheral measures of inflammation might serve as useful markers of inflammatory disease activity. Evidence of chronic inflammation is supported by other groups. One study has demonstrated, increased TNFα and IFN-γ mRNA transcripts in mononuclear cells from the CSF of 13 PPS patients. Of note, such abnormalities were not detected in the circulation of patients [7]. Administration of intravenous immunoglobulin decreased the abnormal cytokine mRNA expression in 16 PPS patients [8]. Our study extends these prior observations in a larger cohort and with detection of abnormal elevations in the translated protein of TNFα as well as IL-6 and leptin, in the circulation of PPS patients. In the case of TNFα, the blood-brain barrier is permeable to this molecule in both the injured and non-injured state [32,33]. Future longitudinal studies could potentially demonstrate the use of serum cytokines as biomarkers to assess target organ inflammation or to predict certain clinical parameters, such as pain due to PPS. The question of how inflammation may contribute to PPS, decades after acute poliovirus infection, and its relationship with ongoing symptoms in patients, is raised. The presence of poliovirus genomic sequences within the CSF has been reported in PPS patients [10,34]. While not active or having a direct ability to cause PPS symptoms due to restrained RNA synthesis and lack of evidence for direct toxicity, these poliovirus genomic particles have the potential to induce the production of cytokines, such as TNFα, and thereby contribute to gradually worsening chronic inflammation [35]. Other groups have not found evidence for poliovirus persistence [36,37]. This has led to other hypotheses for persistently elevated cytokines to include a poliovirus-induced autoimmune response directed against as of yet unknown neuronal or non-neuronal autoantigens, or an immune response that is completely secondary to CNS damage [10]. After demonstrating elevated levels of serum TNFα, IL-6, and leptin in PPS patients, we then performed a secondary
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
analysis to explore potential relationships between these cytokines and clinical parameters in PPS. Both muscle and joint pain occur in most PPS patients, and are associated with a significant reduction in quality of life [3]. Joint pain is likely caused by joint and soft tissue abnormalities, related to chronic abnormal use from weakness secondary to acute polio. The origin of muscular pain, is unclear, but has not been associated with greater weakness at acute polio, current muscle strength, and increased serum creatine kinase (CK) levels [3]. Recent studies have increasingly implicated proinflammatory cytokines in pathways of pain perception [38], including both peripheral and central neuropathy [39]. Ours is the first study to report an association in PPS patients between elevated TNFα levels and increased pain, specifically that of muscular origin. BMI may be confounder in this relationship because heavier patients may be more likely to have pain due to overuse. However, our finding is consistent with experimental evidence in rats, where TNFα has direct effects on nociception, and injured nerve fibers are sensitized to the excitatory effects of TNFα [40]. In a mouse model of chronic nerve injury, inhibition of TNFα receptors resulted in a reduction in thermal analgesia and mechanical allodynia [41]. Our findings are supported by a recent multi-center randomized controlled trial which reported that in a secondary analysis, intravenous immunoglobulin decreased pain in PPS patients. This suggests a link between pain and inflammation, however inflammatory markers were not assessed [42]. Our study is cross-sectional, and we do not wish to overstate potential causal relationships or links between inflammatory markers and PPS symptoms. We also acknowledge that circulating cytokines may be subject to circadian variation that could differ across groups, and therefore longitudinal studies with repeat serum sampling may provide further insights into such biological variability. Nonetheless, our PPS cohort represents one of the largest ones studied to date, and we were able to detect several abnormalities in inflammatory markers in the circulation of patients. We also found a specific association between TNFα and muscle pain. Together, our results suggest that inflammation may play a role in PPS. These results prompt further investigation into the importance of inflammatory mediators in the pathophysiology and symptomatology of PPS. Such follow-up studies are particularly relevant in view of the lack of specific treatments for this debilitating condition and the continued endemic existence of poliovirus in some areas of the world [43], at a time that a growing range of novel immune modulating therapies is under active development. Acknowledgements The authors acknowledge the help of Dr. Diane Diorio, Montreal Neurological Hospital, in patient recruitment for the study. We are grateful to Boli Fan from Dr. Bar-Or's research laboratory for her help in processing samples for the study. We are indebted to the patients and controls in this
85
study who donated their time to participate. The study was supported by the Multiple Sclerosis Society of Canada and the Montreal Neurological Institute. References [1] Jubelt B, Cashman NR. Neurological manifestations of the post-polio syndrome. Crit Rev Neurobiol 1987;3:199–220. [2] Trojan DA, Cashman NR. Post-poliomyelitis syndrome. Muscle Nerve 2005;31:6–19. [3] Vasiliadis HM, Collet JP, Shapiro S, Venturini A, Trojan DA. Predictive factors and correlates for pain in postpoliomyelitis syndrome patients. Arch Phys Med Rehabil 2002;83:1109–15. [4] Pezeshkpour GH, Dalakas MC. Long-term changes in the spinal cords of patients with old poliomyelitis. Signs of continuous disease activity. Arch Neurol 1988;45:505–8. [5] Dalakas MC, Elder G, Hallett M, Ravits J, Baker M, Papadopoulos N, et al. A long-term follow-up study of patients with post-poliomyelitis neuromuscular symptoms. N Engl J Med 1986;314:959–63. [6] Dalakas MC. Morphologic changes in the muscles of patients with postpoliomyelitis neuromuscular symptoms. Neurology 1988;38: 99–104. [7] Gonzalez H, Khademi M, Andersson M, Wallstrom E, Borg K, Olsson T. Prior poliomyelitis-evidence of cytokine production in the central nervous system. J Neurol Sci 2002;205:9–13. [8] Gonzalez H, Khademi M, Andersson M, Piehl F, Wallstrom E, Borg K, et al. Prior poliomyelitis-IVIg treatment reduces proinflammatory cytokine production. J Neuroimmunol 2004;150:139–44. [9] A Bar-Or. Immunology of multiple sclerosis. Neurol Clin 2005; 23:149–175, vii. [10] Dalakas MC. Pro-inflammatory cytokines and motor neuron dysfunction: is there a connection in post-polio syndrome? J Neurol Sci 2002;205:5–8. [11] Wyss-Coray T. Inflammation in Alzheimer disease: driving force, bystander or beneficial response? Nat Med 2006;12:1005–15. [12] Orr CF, Rowe DB, Mizuno Y, Mori H, Halliday GM. A possible role for humoral immunity in the pathogenesis of Parkinson's disease. Brain 2005;128:2665–74. [13] Wu DC, Re DB, Nagai M, Ischiropoulos H, Przedborski S. The inflammatory NADPH oxidase enzyme modulates motor neuron degeneration in amyotrophic lateral sclerosis mice. Proc Natl Acad Sci U S A 2006;103:12132–7. [14] Martino G, Furlan R, Brambilla E, Bergami A, Ruffini F, Gironi M, et al. Cytokines and immunity in multiple sclerosis: the dual signal hypothesis. J Neuroimmunol 2000;109:3–9. [15] Soop M, Duxbury H, Agwunobi AO, Gibson JM, Hopkins SJ, Childs C, et al. Euglycemic hyperinsulinemia augments the cytokine and endocrine responses to endotoxin in humans. Am J Physiol Endocrinol Metab 2002;282:E1276–85. [16] Watson GS, Craft S. Insulin resistance, inflammation, and cognition in Alzheimer's Disease: lessons for multiple sclerosis. J Neurol Sci 2006;245:21–33. [17] La Cava A, Matarese G. The weight of leptin in immunity. Nat Rev Immunol 2004;4:371–9. [18] Frisullo G, Angelucci F, Mirabella M, Caggiula M, Patanella K, Nociti V, et al. Leptin enhances the release of cytokines by peripheral blood mononuclear cells from relapsing multiple sclerosis patients. J Clin Immunol 2004;24:287–93. [19] March of Dimes. Post-polio syndrome: identifying best practices in diagnosis and care. White Plains: March of Dimes Birth Defects Foundation; 2001 (www.modimes.org). [20] Trojan DA, Arnold D, Collet JP, Shapiro S, Bar-Or A, Robinson A, et al. Fatigue in multiple sclerosis: association with disease-related, behavioural and psychosocial factors. Mult Scler 2007;13:985–95. [21] Radloff LS. A self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385–401.
86
C.B. Fordyce et al. / Journal of the Neurological Sciences 271 (2008) 80–86
[22] Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD. The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol 1989;46:1121–3. [23] Smets EM, Garssen B, Bonke B, De Haes JC. The Multidimensional Fatigue Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res 1995;39:315–25. [24] Kleyweg RP, van der Meche FG, Schmitz PI. Interobserver agreement in the assessment of muscle strength and functional abilities in Guillain–Barre syndrome. Muscle Nerve 1991;14:1103–9. [25] Finch LE, Venturini A, Mayo NE, Trojan DA. Effort-limited treadmill walk test: reliability and validity in subjects with postpolio syndrome. Am J Phys Med Rehabil 2004;83:613–23. [26] Schneider RA. Reliability and validity of the Multidimensional Fatigue Inventory (MFI-20) and the Rhoten Fatigue Scale among rural cancer outpatients. Cancer Nurs 1998;21:370–3. [27] Bender R, Lange S. Adjusting for multiple testing—when and how? J Clin Epidemiol 2001;54:343–9. [28] Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 1995;57:289–300. [29] Smith T, Groom A, Zhu B, Turski L. Autoimmune encephalomyelitis ameliorated by AMPA antagonists. Nat Med 2000;6:62–6. [30] Fordyce CB, Jagasia R, Zhu X, Schlichter LC. Microglia Kv1.3 channels contribute to their ability to kill neurons. J Neurosci 2005;25:7139–49. [31] Kim YS, Joh TH. Microglia, major player in the brain inflammation: their roles in the pathogenesis of Parkinson's disease. Exp Mol Med 2006;38:333–47. [32] Pan W, Banks WA, Kastin AJ. Permeability of the blood-brain and blood-spinal cord barriers to interferons. J Neuroimmunol 1997;76: 105–11. [33] Pan W, Banks WA, Kastin AJ. Blood-brain barrier permeability to ebiratide and TNF in acute spinal cord injury. Exp Neurol 1997;146: 367–73.
[34] Leon-Monzon ME, Dalakas MC. Detection of poliovirus antibodies and poliovirus genome in patients with the post-polio syndrome. Ann N Y Acad Sci 1995;753:208–18. [35] Colbere-Garapin F, Duncan G, Pavio N, Pelletier I, Petit I. An approach to understanding the mechanisms of poliovirus persistence in infected cells of neural or non-neural origin. Clin Diagn Virol 1998;9: 107–13. [36] Melchers W, de Visser M, Jongen P, van Loon A, Nibbeling R, Oostvogel P, et al. The postpolio syndrome: no evidence for poliovirus persistence. Ann Neurol 1992;32:728–32. [37] Salazar-Grueso EF, Grimaldi LM, Roos RP, Variakojis R, Jubelt B, Cashman NR. Isoelectric focusing studies of serum and cerebrospinal fluid in patients with antecedent poliomyelitis. Ann Neurol 1989;26: 709–13. [38] Sommer C, Kress M. Recent findings on how proinflammatory cytokines cause pain: peripheral mechanisms in inflammatory and neuropathic hyperalgesia. Neurosci Lett 2004;361:184–7. [39] Marchand F, Perretti M, McMahon SB. Role of the immune system in chronic pain. Nat Rev Neurosci 2005;6:521–32. [40] Schafers M, Lee DH, Brors D, Yaksh TL, Sorkin LS. Increased sensitivity of injured and adjacent uninjured rat primary sensory neurons to exogenous tumor necrosis factor-alpha after spinal nerve ligation. J Neurosci 2003;23:3028–38. [41] Sommer C, Schmidt C, George A. Hyperalgesia in experimental neuropathy is dependent on the TNF receptor 1. Exp Neurol 1998;151: 138–42. [42] Gonzalez H, Sunnerhagen KS, Sjoberg I, Kaponides G, Olsson T, Borg K. Intravenous immunoglobulin for post-polio syndrome: a randomised controlled trial. Lancet Neurol 2006;5:493–500. [43] Howard RS. Poliomyelitis and the postpolio syndrome. BMJ 2005;330:1314–8.