ABSTRACT:
OBJECTIVE The aim of this study was to determine the prevalence of neuropathic pain and
characterize the quality of life of patients with osteoarthritis who
consulted a pain clinic in Southwestern Colombia.
METHODS A cross-sectional study was conducted via telephone survey. Participants ≥18
years of age with a diagnosis of osteoarthritis were included. The LANSS
questionnaire was used to evaluate symptoms and signs of neuropathic pain,
and the Short Form-8 was used to evaluate quality of life.
RESULTS Response rate was 54.1% (46/85). The male-to-female ratio was 5:1, with an
average age of 72 ±10 years. Most participants (91.3%) had severe pain. The
prevalence of neuropathic pain was 28.3% (95%CI = 15.99-43.46), and the
prevalence of neuropathic pain amongst women was 84.6% (95%CI = 54.55-98.01
). Dysesthesias and paroxysmal pain were present in 92.3% of individuals
with neuropathic pain. Regarding quality of life, limitations in physical
activity were the most significant, as 63% of individuals reported such
limitations.
CONCLUSION Neuropathic pain was found to be prevalent and had a negative impact on
physical function, highlighting the need for therapeutic strategies targeted
to specific neuropathic pain pathways in patients with osteoarthritis.
Keywords: chronic pain; neuralgia; quality of life; osteoarthritis.
RESUMO:
OBJETIVO O objetivo deste estudo foi determinar a prevalência de dor neuropática e
caracterizar a qualidade de vida de pacientes com osteoartrite que
consultaram um ambulatório de dor no sudoeste da Colômbia.
MÉTODOS Este foi um estudo transversal realizado por meio de entrevista telefônica.
Foram incluídos participantes ≥18 anos de idade com diagnóstico de
osteoartrite. O questionário Leeds Assessment of Neuropathic
Symptoms and Signs (LANSS) foi utilizado para avaliação dos
sintomas e sinais de dor neuropática e o Short Form-8
analisou a qualidade de vida.
RESULTADOS A taxa de resposta foi de 54,1% (46/85). A razão homem:mulher foi de 5:1, com
média de idade de 72 ±10 anos. A maioria dos participantes (91,3%)
apresentava dor intensa. A prevalência de dor neuropática foi de 28,3%
(intervalo de confiança [IC] de 95% = 15,99-43,46) e a prevalência de dor
neuropática entre mulheres foi de 84,6% (IC 95% = 54,55-98,01). Disestesias
e dor paroxística foram relatadas por 92,3% dos indivíduos com dor
neuropática. Em relação à qualidade de vida, as limitações na prática de
atividade física foram as mais significativas e relatadas por 63% dos
indivíduos.
CONCLUSÃO A dor neuropática foi prevalente e tinha impacto negativo na função física.
Isso destaca a necessidade de estratégias terapêuticas direcionadas a vias
específicas da dor neuropática em pacientes com osteoartrite.
Palavras-chave: dor crónica; neuralgia; qualidade de vida; osteoartrite.
FIGURES
| Citation: Zuluaga ME, Urbano I, Tovar-Sanchez MA, Baena C, Pacichana-Quinayaz SG, Bonilla-Escobar FJ. Prevalence of Neuropathic Pain in Patients with Osteoarthritis. 58(6):e924. doi:10.1055/S-0043-1776986 |
| Note: Work developed at the Department of Physical Medicine and Rehabilitation, Universidad del Valle, Cali, Colombia. |
| Financial Support: The present study was not funded. |
|
Conflict of Interests: The authors declare that they have no conflicts of interest. |
| Received: December 26 2022; Accepted: February 07 2023 |
INTRODUCTION
Osteoarthritis (OA) is the most common form of arthritis as it affects 3.3% to 3.6% of the global population.1 Among older adults, OA is one of the leading causes of deaths and disability worldwide, generating high medical expenses in the middle-aged and elderly populations.2 It is estimated that in the United States, Canada, UK, France and Australia, OA costs account for between 1% and 2.5% of the gross domestic product.3 OA common symptoms are joint pain, stiffness and swelling. Histological features include inflammation, cartilage damage and osteophyte formation resulting from repair attempts.4 Approximately 80% of people over the age of 65 have radiographic evidence of OA, with the hip and knee joints being the most affected.5
From many years ago the link between joint disease and peripheral neuropathy has been well established.6 Although historically, the pain associated with OA had been considered exclusively nociceptive, more recently, there is increasing evidence supporting the idea that has a neuropathic component that can co-exist.7,8 The International Association for the Study of Pain (IASP) states that neuropathic pain (NP) definition must include a central or peripheral lesion of the somatosensory system.9 The exact mechanism of OA-related peripheral neuropathy remains largely unknown.10
The prevalence of NP in the general population is estimated to be between 6.9% and 10%.11 In individuals with knee or hip OA, NP prevalence is around 23%.8 NP has significant implications for quality of life (QOL), including sleep disturbances, anxiety, and depression.12 Given the significant morbidity and mortality of OA there is a need for research addressing the relationship between NP and OA.
Therefore, combined with the paucity of evidence about OA in low- and middle-income countries, the primary aim of this study was to determine the prevalence of NP in patients with OA who consulted the pain clinic in a referral center in Southwestern Colombia. A secondary aim was to explore the QOL of these patients.
METHODS
Study Design
A cross-sectional study was conducted through a telephone interview. The survey used the Spanish versions of the Leeds questionnaire for the Evaluation of Symptoms and Signs of Neuropathic Pain (LANSS)13,14 and the Short Form-8 questionnaire to assess the QOL (SF-8)™.15 The study was approved by the ethics committees of the Universidad del Valle and the Hospital Universitario del Valle E.S.E (code 016-017).
Setting
Study participants were recruited from the outpatient pain clinic of the University Hospital of Valle (HUV), a tertiary care center in Southwestern Colombia. It has approximately 500 beds and serves as the main referral center covering a network of around 22195 km².
Study Population
The study population was identified by searching the registry of patients who consulted the pain clinic in the Department of Physical Medicine and Rehabilitation at HUV. Electronic medical records were reviewed and patients were retrospectively screened for OA according to the International Classification of Disease v.10 (ICD-10) classification with codes M15 to M19.16 Participants ≥18 years of age and with a positive diagnosis of primary or secondary OA were included.
Individuals were contacted between March and April 2018. Verbal informed consent was obtained from all participants prior to the interview. Individuals with altered mental or cognitive status, hearing impairment, previous diagnosis of NP of other causes, fibromyalgia, treatment for more than eight weeks with antineuralgic drugs at therapeutic doses,17 active cancer and autoimmune diseases were excluded.
Variables
Sociodemographic and clinical characteristics, including age, sex, highest academic degree, employment status and location of the worst pain, were collected. Pain characteristics and QOL were collected using the LANSS and the SF-8.15
The LANSS consists of two sections. The first section has four questions that explore essential symptoms of NP grouped into dysesthesias (pins and needles, punches), autonomic changes (if the skin looks mottled or red), pain provoked (if the skin is abnormally sensitive to the touch) and paroxysmal pain (if it has electric shocks, jumping or shocking). The second section performs a physical self-evaluation to look for the presence of positive signs of NP, allodynia, and a negative sign, hypoesthesia. A score greater than or equal to 12 is indicative of NP.14
The SF-815 questionnaire is an abbreviated version of the SF-36 consisting of 8 sub-scales. It evaluates the physical and mental aspects of QOL by asking participants to report the impact that a medical condition has had on various aspect of their life over the past four weeks, and it has been validated in Spanish.15
Data Collection
Telephone interviews were carried out over a period of five weeks. Before starting and to avoid cold calling, patients were asked about their health status and the date of their last and next appointment which is a regular activity in the clinic. Two of the principal investigators trained in applying the survey made the phone calls. Responses were recorded on printed questionnaires, and the information was subsequently entered into an Excel database for analysis.
If a patient could not carry out the physical exam or answer the questionnaire independently, such as in the case of older patients having difficulty understanding the questions, a family member was allowed to assist the participant.
Data Analysis
Data analysis was performed using Stata 16 (Stata Corp., TX, US)®. Initially an exploratory analysis of the data was carried out. This was done in order to identify missing data, typing errors, and inconsistent values and then corrected using the medical record or the paper registries.
Univariate descriptive analysis was performed in which measures of central tendency (mean, median) and dispersion (standard deviation, percentiles, maximum and minimum values) were calculated for continuous variables. Categorical variables were described as relative frequencies and percentages. For bivariate analysis, two groups were compared: patients with NP and without NP, using sociodemographic and clinical variables. Two-tail hypothesis tests were conducted based on the type of variable under comparisons; for categorical variables, chi2 or the Fisher's exact test was used and for continuous variables, t-test or the Wilcoxon test was used, as appropriate. Normality was assessed using the Shapiro-Wilk test and equality of variances was assessed using the variance ratio test. Significance was set at a p-value <0.05.
RESULTS
A total of 126 patients with a positive diagnosis of OA were screened from the registry. Of these, 41 were excluded. Eighty-five patients were listed to be interviewed by telephone; the investigators made an average of five call attempts to each of the participants. Thirty-nine patients were discarded: 32 because the telephone number was invalid, inactive, or there was no response; five had died; and two did not consent to participate. Thus, a total of 46 patients with a positive diagnosis of OA were surveyed (54.12% response rate) (►Fig. 1).
Sociodemographic and clinical characteristics of the participants are summarized in ►Table 1. Most participants were female (82.6%), and the mean age was 72 ±10 years. There were no significant differences in age by sex (p = 0.35) or the presence of NP (p = 0.71 ).
| Characteristic | All (n=46) | Neuropathic Pain | p Value | |
|---|---|---|---|---|
| No (n = 31) | Yes (n = 15) | |||
| Age, mean ± SD | 72 ±10 years | 72.68±10.52 | 71.47 ±10.24 | 0.71+ |
| Sex: Female, n (%) | 38 (82.6) | 26 (83.87) | 12(80) | 1.0† |
| Education, n (%) | 0.14† | |||
| < Primary | 25 (54.35) | 18(58.06) | 7 (46.67) | |
| Primary | 13 (28.26) | 6(19.35) | 7 (46.67) | |
| > Primary | 8(17.39) | 7 (22.58) | 1 (6.67) | |
| Marital status, n (%) | 0.92† | |||
| Married | 19(41.30) | 12(38.71) | 7 (46.67) | |
| Single/Divorced | 17(36.96) | 12(38.71) | 5 (33.33) | |
| Widowed | 10(21.74) | 7 (22.58) | 3(20) | |
| Employment status: Retired, n (%) | 39 (84.78) | 27(87.10) | 12(80) | 0.67† |
| Live in the city: Cali, n (%) | 30 (65.22) | 21 (67.74) | 9(60) | 0.61‡ |
| Severe pain: Yes, n (%) | 42(91.3) | 27(87.1) | 15(100) | 0.29† |
| Pain visual analogue scale, median ± IQR | 9(8-10) | 9(8-10) | 10(8-10) | 0.36* |
Abbreviatin: IQR, Interquartile range; LANSS, SD, Standard deviation.
+T-test. †Fisher's exact test. ‡ Chi2. * Wilcoxon test
►Table 1 show sociodemographic and presence of neuropathic pain of the participants
Regarding the intensity of pain measured with the Visual Analogue Scale (VAS), 91.3% (n = 42/46) had severe pain, corresponding to a score between 7 and 10. In patients with a LANSS score of ≥12,46.1% rated the intensity of the pain as a 10. Patients perceived the worst pain in the hips (38.5%) and the knee (30.8%) (►Fig. 2).
Overall, the prevalence of NP was 28.3% (n = 13, 95% CI = 16-43.5). Among women, the prevalence was 84.6% (n = ll, 95%CI = 54.5-98). In the segregated analysis of the variables from the LANSS questionnaire, paroxysmal pain was the most prevalent symptom (►Table 2). In participants with NP, the two most frequent types of pain were dysesthesia and paroxysmal pain (►Fig. 3).
| Questionnaire item | Frequency (n=46) | Percentage (%) |
|---|---|---|
| Essential Symptoms | ||
| Dysesthesias | 24 | 52.17 |
| Autonomic changes | 6 | 13.04 |
| Provoked pain | 23 | 50 |
| Paroxysmal pain | 27 | 58.7 |
| Physical Self-evaluation | ||
| Allodynia | 14 | 30.43 |
| Hypoesthesia | 19 | 41.3 |
►Table 2 describes LANSS questionnaire items and frequency among participants.
Regarding the assessment of QOL (►Table 3), Physical Health was mostly affected by General Health, Role Physical and Bodily Pain sub-scales. 60.9% of participants perceived their General Health to be "fair,", 63% reported "quite a lot" or "complete" limitations in Role Physical and more than half (60.8%) of the participants reported "severe" or "very severe" in Bodily Pain sub-scales. Mental Health was affected by Social Functioning and Mental Health sub-scales and somehow preserved in Role Emotional sub-scale where most participants reported "not at all" or "slight" emotional problems (52.1%).
| Sub-scale | Item | n | % |
|---|---|---|---|
| General health | Excellent | 1 | 2,17 |
| Very Good | 4 | 8,7 | |
| Good | 4 | 8,7 | |
| Fair | 28 | 60,87 | |
| Poor | 5 | 10,87 | |
| Very Poor | 4 | 8,7 | |
| Physical activities limitation (role-physical) | Not at all | 3 | 6,52 |
| Very little | 3 | 6,52 | |
| Somewhat | 11 | 23,91 | |
| Quite a lot | 17 | 36,96 | |
| Could not do physical activities | 12 | 26,09 | |
| Daily work difficulty (physical functioning) | Not at all | 9 | 19,57 |
| Very little | 6 | 13,04 | |
| Somewhat | 9 | 19,57 | |
| Quite a lot | 14 | 30,43 | |
| Could not do daily work | 8 | 17,39 | |
| Bodily pain | None | 2 | 4,35 |
| Very mild | 2 | 4,35 | |
| Mild | 2 | 4,35 | |
| Moderate | 12 | 26,09 | |
| Severe | 19 | 41,3 | |
| Very severe | 9 | 19,56 | |
| Vitality | Very much | 3 | 6,52 |
| Quite a lot | 6 | 13,04 | |
| Some | 16 | 34,78 | |
| A little | 14 | 30,43 | |
| None | 7 | 15,22 | |
| Limitation of usual social activities (social functioning) | Not at all | 13 | 28,26 |
| Very little | 4 | 8,7 | |
| Somewhat | 6 | 13,04 | |
| Quite a lot | 16 | 34,78 | |
| Could not do social activities | 7 | 15,22 | |
| Emotional problems (role-emotional) | Not at all | 13 | 28,26 |
| Slightly | 11 | 23,91 | |
| Moderately | 6 | 13,04 | |
| Quite a lot | 10 | 21,74 | |
| Extremely | 6 | 13,04 | |
| Daily activities limitation because of personal or emotional problems (mental health) | Not at all | 5 | 10,87 |
| Very little | 7 | 15,22 | |
| Somewhat | 14 | 30,43 | |
| Quite a lot | 14 | 30,43 | |
| Could not do daily activities | 6 | 13,04 |
Note: ►Table 3 shows eight sub-scales of the quality-of-life questionnaire with the frequency and percentage on each item evaluated.
DISCUSSION
In our study the overall NP prevalence was 28.3%, which can be contrasted with another study that found NP prevalence of 37%; however that study investigated individuals with knee pain exclusively.18 A systematic review estimated the prevalence of NP amongst individuals with OA to be 23%; however, the review included knee or hip OA.8 There is a lack of research investigating the presence of NP specifically amongst individuals with OA, and worldwide, there are few studies that characterize the prevalence of NP in those suffering from chronic pain of different etiologies.
Our study found that NP was more prevalent among females, which is in agreement with prior reports.11,19,20,21 The reason for this sex difference is not clear. However, it may be related to the hormonal deficiencies that develop in women around menopause, making them susceptible to OA and, therefore, to NP.19
Most of the participants (91.3%) reported severe pain. We found that in patients with OA, the hip and knee were reported as areas of the worst pain. NP among those with hip pain was overrepresented, since the prevalence was 38.5%, while the prevalence of NP in other areas of OA pain was 28.3%. Our findings show a high frequency of hip pain, contrasting reports from Latin America, that estimate 31.2% prevalence of knee OA, and only 1.3% of had hip OA.22 The high rates of hip OA in our study may be because most of the participants were female (82.6%) and past menopause, as it has been found that hormonal changes may be a risk factor for hip OA.20,21
In assessment of QOL, social functioning sub-scale half of the participants reported either "quite a lot" or "complete" limitations. As the average age of patients was 72 years old, these limitations in social functioning may be attributed to factors such as the narrowing social networks and changes in social roles that can occur with increasing age.23 At the same time, most of the participants reported "slight" or "no" emotional problems (52.1%), which could reflect the increased emotional stability that may also come with increasing age.23 However, 43% reported "quite a lot" or "complete" limitations in their daily activities due to personal or emotional reasons (mental health). As both social and daily physical functioning contribute to healthy aging,24 there is a need to further investigate the effect that OA and NP on social and emotional factors in the elderly.
As previously mentioned, OA pain can be nociceptive and neuropathic.10 There have been three important factors identified in the origin of OA pain. First is the increase of cytokines and interleukins in the synovial fluid locally in the joint. Second, general factors such as biomechanical and biochemical alterations due to obesity or diabetes. Third, neuroplastic changes such as peripheral sensitization due to the overgrowth of nerve fibers in articular cartilage that is not typically reversed,25 this mechanism could be the most significant with regard to the onset of NP in OA, as the NP definition must include a structure of the somatosensory system altered.8,9,10
Currently, in Colombia, the predominant treatment for OA is paracetamol and non-steroidal anti-inflammatory agents.22 However, as out study emphasize the prevalence of NP in individuals with OA, our results also highlight the need for more targeted interventions to address OA and NP. Pharmacological options such as Duloxetine already approved for knee OA as a chronic pain condition in other guidelines5 and Lipid lysophosphatidic acid (LPA) receptor blockade26 proposed as a pharmacological method to inhibit joint nerve damage should be the focus of further studies in OA and NP.27 Furthermore, non-pharmacological approaches include treatment tailored to the cognitive processes involved in the phenomenology of pain, such as pain-related catastrophizing, which has proven to be effective in the multidisciplinary management of pain.28
Limitations
The LANSS questionnaire validated for telephone application has a sensitivity and specificity of 52% and 78%, respectively. These are relatively low values; however, this is the only relevant questionnaire validated in Spanish for telephone use. Further research is required to validate and evaluate the reliability of other similar tools.
Another limitation is the small sample size which provides relevant information about response rate of participants enrolled in this type of research, information that has not been described in Colombia to the best of our knowledge. We attempted to locate participants with multiple phone calls trying to include as many subjects as possible. The power that we reached based on the sample size (n = 46), a null proportion of 37%,19 an alpha of 5%, and the prevalence that we described of 28.3%, was only 21%. As this is the first study describing OA and NP in Colombia and Latin America, the information may be of use of future researchers to use a different data collection method and to calculate a sample size based on our results.
Strengths
To the best of our knowledge, this is the first study in Latin America to assess both OA and NP. There is a lack of research about this topic around the globe. We used a validated tool to assess the outcome variable (NP). Interviews were carried out by resident physicians thus providing a high-quality data collection given the challenges of a survey that includes self-examination.
CONCLUSION
In our study, we found that the prevalence of NP in patients with OA was close to 30%. It has been increasingly understood that the pain associated with OA is not purely somatic. Recent evidence shows that there are alterations of the somatosensory system in the arthritic joint. Due to the prevalence of NP in individuals in our study, there is a need for further research investigating the physiological mechanisms behind NP in patients with OA. Understanding differences in the mechanisms of pain allows strategies targeted to specific pain pathways and enables healthcare providers to better predict and understand the patient response to treatments. Therefore, there is a need for further interdisciplinary studies that characterizes NP in individuals with OA.
ACKNOWLEDGMENTS
The authors thank Sohaib Haseeb, Annora Kumar and Dino Ventolini Zuluaga for their contributions in editing this document.
