In the present study we evaluated the CSI-Pol on a group of chronic spinal subjects and a separate group of control subjects. The study was approved by the Ethical Board of the Warsaw Medical University (Poland) (Consent number: KB/66/2019 obtained 15/04/2019). All participants in this study agreed to participate and signed informed consent forms before the study. All methods were performed in accordance with the relevant guidelines and regulations.
Participants
We recruited 152 outpatients with chronic spinal pain (some with CLBP only, some with chronic neck pain only, and some with both painful areas, and all with more than 3 months of pain duration) from the Neurological Outpatient Clinic and the Rehabilitation Clinic from National Institute of Geriatrics Rheumatology and Rehabilitation Clinic, Poland. The inclusion criteria were: age between 20 and 80 and chronic spinal pain with more than 3 months of pain duration. We excluded patients with cancer in the brain or in the spine, and other neurological diseases which could cause primary neuropathic pain as polyneuropathies (any causes; patients with diabetes were excluded), no patients had history of heavy alcohol consumption; dementia, previous spine surgery, or recent trauma in anamnesis and poor Polish comprehension skills. Patients with cervical or lumbar herniations with clinical symptoms awaiting surgery, were also excluded. All of the 152 patients completed the CSI-Pol, but one patient was excluded due to incomplete data, leaving 151 patients for analysis, including 24 with chronic neck pain (CNP) and 73 with CLBP and 54 with both CNP and CLBP. The control group consisted of 30 healthy subjects, with no reported spinal pain conditions, who were recruited from medical staff and their families, and agreed complete the CSI-Pol.
Measures
All of the 151 chronic spinal pain patients completed a battery of patient-reported measures. The 30 healthy subjects completed the CSI-Pol only. Additional demographic data were also collected.
Demographic and clinical data
Pain intensity was measured with a Numeric Pain Rating Scale (NRS), from 0 (no pain) to 10 (the worst imaginable pain) over the last 4 weeks. Age and sex of each participant was recorded. Each subject was weighed and their body mass index (BMI) was calculated. Patients were asked in an interview if they were employed, if they used alcohol excessively (over 60 g of 100% alcohol for men and 40 g for women daily) and if they had sleep disturbances (in “yes” or “no” format) caused by pain that prevented them from falling asleep or staying asleep during the night.
Central sensitization inventory- polish version (CSI-pol)
The CSI-Pol can be found in Appendix A (Supplement A) and at https://www.pridedallas.com/questionnaires [30]. The CSI consists of two parts: Part A is a 25-item self-report questionnaire which assess health-related symptoms common to CS and Central Sensitivity Syndromes. Each item is rated on a 5-point Likert-type scale (0 = never and 4 = always), with total scores of 0–100 [26]. Part B is not scored. It is designed to determine if the subject has been diagnosed with other CS-related disorders, including restless leg syndrome, chronic fatigue syndrome, fibromyalgia, temporomandibular joint disorder, migraine or tension headaches, irritable bowel syndrome, multiple chemical sensitivities, neck injuries (including whiplash), anxiety or panic attacks, and depression [31].
Neck disability index (NDI)
The NDI measures perceived level of disability in subjects with neck pain [32]. The Polish version was used in the present study [33]. The test consists of 10 items concerning various daily activities and other domains and is scored from 0 to 100% disability, with higher scores indicating greater perceived disability [32].
Oswestry disability index (ODI)
The Oswestry Disability Index measures perceived level of disability in subjects with low back pain [34]. The Polish version was used in the present study [35]. The test consist of 10 items concerning various daily activities and other domains and is scored from 0 to 100% disability, with higher scores indicating greater perceived disability [32].
Clinical psychological diagnostic system. Depression symptoms measurement questionnaire (KPD)
The Depression Symptoms Measurement Questionnaire consists of 75 statements, to which the respondent responds on a 4-point scale. It is used in Poland to assess the symptoms of depression. It contains five problems: Cognitive Deficits and Energy Loss,; Suicidal Tendencies, Pessimism and Alienation; Guilt and Anxiety; Psychosomatic Symptoms and Loss of Interest; Self-regulation. Cognitive Deficits and Loss of Energy.
Scale measures cognitive difficulties such as attention, learning, memory, psychomotor speed, executive functions resulting from depressed mood. Mortality, Pessimism and Alienation Scale examines the subjectively experienced loss of meaning in life, measures the sense of alienation and social isolation. Guilt and Anxiety tension.
concerns feelings of guilt, anxiety, fear, sadness. The items of this scale measure an attitude of dwelling on one’s failures and difficulties. Psychosomatic Symptoms and Decline in Interest measures the subjective evaluation of one’s own health and psychophysical performance. The Self-Regulation scale measures the subject’s emotional and cognitive resources that protect against depression. Depression Symptoms Measurement Questionnaire also has a total score scale measuring overall level of depression which is a sum of scores obtained from individual scales. High scores on these scales indicate high levels of depressed mood symptoms.
The Berlin social support scale (BSSS)
The Berlin Social Support Scale is a battery of self-report questionnaires developed by Schulz and Schwarzer to measure perceived social support [36]. The Polish version of the BSSS was used in the present study [37].
Quantitative sensory testing
We assessed mechanical allodynia using a brush, thermal allodynia using ice in a glove and pinprick hyperalgesia using a wooden cocktail-stick.
Procedure
All 151 participants completed the CSI-Pol, then completed it again one week later to determine test-retest reliability. Using total CSI-Pol scores, we used the algorithm proposed by Nijs et al. 2015 to differentiate predominant neuropathic, nociceptive and CS pain for each subject [31]. According to their suggestions the neuropathic component of pain was suspected if it was determined to be neuroanatomically logical, with the eventually presence of allodynia, hyperalgesia, with pain characterization of burning, shooting, or pricking (mostly radicular pain). If a neuroanatomically illogical pattern of pain was seen (with or without presence of allodynia and hyperalgesia in these locations), with disproportionate experience of pain to the nature and extent of the injury or pathology (structural impairments which might cause nociceptive LBP) spinal pain with CS was suspected [25]. The diagnosis was made by experienced physicians (neurologists and physiatrists). We divided the patient sample into 5 severity levels groups, as has been recommended previously, to aid the clinical interpretation process, including subclinical = 0–29 points; mild = 30 to 39 points, moderate = 40–49 points; severe = 50–59 points and extreme from 60 to 100 points [22].
Statistics
Data were analyzed using the statistical package STATISTICA 12.0 (licensed by StatSoft PL, Cracow, Poland) and IBM SPSS Statistics, version 22.0 (IBM, Armonk, NY, USA). Qualitative variables were characterized by the number of important cases (n) and the percentage of the total (%). Categorical comparison of groups was made using two-way tables and chi square test or Fisher exact test. The normality of continuous variables was determined using Kolmogorov–Smirnov. Lilliefors and Shapiro–Wilk tests were used to assess the homogeneity of dispersion from normal distribution.
Variables that were normally distributed were presented as a mean and standard deviation and compared between groups by one-way analyses of variance (more than 3 groups) with post-hoc analysis using Neuman-Keulus test or by student t test (2 groups). A Brown–Forsythe test was used to evaluate the homogeneity of variance (significance < 0.05). Variables that were not normally distributed, or for which the criterion of homogeneity of variance was not completed, were presented as a median and interquartile range (IQR) and compared between groups with Kruskal–Wallis analysis of variance (ANOVAs) (more than 3 groups) with post-hoc testing using Mann–Whitney U-tests with Bonferroni correction of p values or by Mann–Whitney U-test (two-way variables). Effect sizes and power analyses were performed using the G*Power 3.1.9.7 tool. Categorical variables were compared between groups using chi2 test or Fisher exact tests. Spearman’s correlation coefficients were used to examine the associations between the Polish version of the Central Sensitization Inventory (CSI-Pol) scores and pain intensity (NRS), NDI and ODI. Correlations were evaluated using the Spearman rank correlation test. The correlation strength was presented according to the Guilford classification [38]:
-
Poor when the correlation coefficient r = 0.1 < │r│ ≤ 0.3
-
Moderate when 0.3 < │r│ ≤ 0.5
-
High (strong) when 0.5 < │r│ ≤ 0.7
-
Very high (very strong) when 0.7 < │r│ ≤ 0.9
-
Nearly full when 0.9 < │r│ < 1
-
Full when │r│ = 1
The criterion for significant differences was p < 0.05. Data were given as means (M) with standard deviations (SD). Construct validity and factor structure were determined through the use of questionnaire principal component analysis with Maximum Likelihood Extraction (MLE), with the requirements for extraction being the satisfaction of all three points: scree plot inflection point, Eigen value > 1.0 and accounting for > 10% of variance [39]. The recommended minimum ratio of five participants-per-item was satisfied.
Additionally, the original 4-factor model suggested by Mayer et al. was tested via confirmatory factor analysis (CFA) with ordinal data [17]. In the 4-factor model, Factor 1 originally was named “physical symptoms” and included items 1, 2, 5, 6, 8, 9, 12, 14, 17, 18, and 22; Factor 2 was named “emotional distress” and included items 3, 13, 15, 16, 23, and 24; Factor 3 was named “headache/jaw symptoms” and included items 4, 7, 10, 19, and 20; and Factor 4 was named “urological symptoms” and included items 11, 21, and 25.
Internal consistency of the scale items was determined from Cronbach’s α coefficients [40]. Reliability was determined by test-retest ICC. An error range of 0 ± 10% was allowed in determining the test–retest reliability. The standard error of measurement (SEM) was calculated using the formula: SEM = s√(1 − r), where s = the mean and standard deviation (SD) of Time 1 and Time 2; r = the reliability coefficient for the test and Pearson’s correlation coefficient between test and retest values. Thereafter, the Minimal Detectable Change 90 (MDC90) was calculated using the formula: MDC90 = SEM × √2 × 1.96.