Abstract
Objective
This study aims to explore the associations between clinical characteristics, including medical history, and the frequency, severity, and type of pain events (PEs) over a two-year period in patients with knee osteoarthritis (OA).
Design
A post-hoc analysis was conducted on data from two double-blinded, randomized, placebo-controlled, multicenter phase-III trials assessing the efficacy and safety of salmon calcitonin in patients with painful and radiographic knee OA. 2206 patients aged 51–80 years were included. Adverse events (AEs) were recorded and categorized into PEs and non-PEs. Logistic regression models were used to assess associations, adjusting for confounders such as age, sex, and body mass index (BMI).
Results
Of the 8183 AEs recorded, 27.6 % were classified as PEs. Higher BMI was associated with lower odds of reporting PEs (OR: 0.94, p = 0.02), while male patients reported fewer PEs than females (OR: 0.83, p = 0.04). Insomnia and depression were linked to increased odds of PEs. In the PE population, higher BMI was associated with increased odds of moderate and severe PEs. KL grade 3 was linked to lower odds of moderate and severe PEs.
Conclusions
The study identifies female sex and certain medical histories as significant predictors of PEs in knee OA patients. A higher BMI was associated with lower odds of reporting PEs but for those who reported PEs, higher BMI was associated with more severe PEs. These results should be interpreted with caution due to methodological limitations, however the findings indicate risk factors for pain events and suggest further research into these associations.
Keywords: Osteoarthritis, Knee, Pain, Obesity, Randomized controlled trials
1. Introduction
Osteoarthritis (OA) is the most prevalent chronic joint disease with knee OA causing the greatest burden to the population in terms of pain, stiffness and disability [1]. Currently, no medical treatments are approved for modification of the disease despite extensive efforts in clinical development [2,3]. Due to the limited treatment options, there has been a shift in focus towards identifying modifiable risk factors to help alleviate the disease burden [4,5]. Understanding the role that clinical characteristics and medical history (MH) play in knee OA and the risk of experiencing pain events is therefore important and may contribute with new insight into the design, conduct or content of existing and new interventions [6,7].
Pain is the predominant symptom in OA, with the majority of patients experiencing joint pain on a daily basis [8]. Due to the nature of OA, patients with knee OA are over a period of time likely to experience a worsening of their existing joint pain. Hence, events of worsening of existing joint pain is one of the most frequently reported adverse events (AEs) in a typical knee OA population, as reported in large randomized controlled trials. In addition to the existing knee OA pain, various other types of pain are commonly observed in knee OA patients [[9], [10], [11]].
There is substantial variability in how patients report Pain Events (PEs). The type of PEs can range from headaches, pain in specific joints or body locations, to unspecific pain or pain with potential neurogenic or neuropathic origins [[9], [10], [11]].
Despite this diversity in pain reporting, we currently have limited understanding of the factors associated with incidence and frequency of pain beyond the joint-specific pain most frequently collected as the main efficacy outcomes in OA-trials.
Previous studies have demonstrated that the risk of knee OA pain progression is correlated with increasing age and Body Mass Index (BMI) [6,7] and female sex [12,13], with the same trend in other types of pain, such as lower back pain [14]. Similarly, it has been established that patients with a medical history of especially psychiatric disorders, such as insomnia, have a higher risk of both knee OA pain progression and intensity [15] as well as PEs of other origin, such as lower back pain [16].
The objective of this study is to investigate associations between clinical characteristics including relevant medical history, and the frequency, severity and type of PEs reported during two large trials in a knee OA population.
2. Method
2.1. Study population
This is a post-hoc analysis of two large phase III trials of two, double-blinded, randomized, placebo-controlled, multicenter phase III clinical trials assessing the efficacy and safety of an oral formulation of 0.8 mg salmon calcitonin in patients with painful and radiographic knee OA (NCT00486434 (trial 1) and NCT00704847 (trial 2)) [17]. No significant differences were observed between participants receiving placebo and those receiving salmon calcitonin in terms of efficacy endpoints and reported AEs [17]. For the current analysis both participants receiving placebo and active treatments were included.
Each independent trial recruited patients aged 51–80 years with painful OA in the target knee and meeting the American College of Rheumatology (ACR) criteria for diagnosis of OA. On the Western Ontario and McMasters Osteoarthritis Index (WOMAC) subscales, patients were to score ≥150 mm (500 mm being the maximum score) for pain and/or ≥ 510 mm (1700 mm being the maximum score) for function. The radiographic inclusion criteria for the target knee included Kellgren-Lawrence (KL) grades 2 or 3, and a Joint Space Width (JSW) of ≥2.0 mm. An overview of the study population is shown in Fig. 1 and further details on the study population can be found in previously published reports [17].
Fig. 1.
Study Population and Pain Events.
Overview of Study Total Study population and distribution of Pain Events in the Pain Event Population, grouped by severity by most severe Pain Event per patient and subtype of Pain Events.
2.2. Study design
A total of 2206 patients were recruited at 19 sites in 11 countries, of which 1487 patients completed the trial. At screening, medical history and demographics were recorded by medical staff based on patient statements and medical records. X-rays were obtained during screening and derived parameters classified by a central reader. At the baseline visit, remaining clinical characteristics were collected, and BMI was derived by weight and height measurement. Patients were followed for two years with regular clinic visits. The trials were conducted in accordance with the Helsinki Declaration and ICH GCP, and were approved by all applicable Independent Review Boards, Ethics Committees, and regulatory bodies.
2.3. Adverse event reporting
Incidence of adverse events were assessed by investigators based on report by participants at study visits during the entire two-year study period, starting from the time of informed consent [17]. AEs are defined as any untoward medical occurrence in a patient administered a pharmaceutical product, regardless of causal relationship with this treatment. An AE can therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product, whether or not considered related to the medicinal product [18]. All AEs were coded to the most appropriate AE term using the Medical Dictionary for Regulatory Activities during study conduct, finalized before the study was locked and unblinded. All AEs analysed in the present report were categorized by their respective MedDRA Preferred Term and System Organ Class. Pain Event classification:
For the analysis of associations between clinical parameters and the occurrence of PEs, reported AEs were categorized into two groups: PEs and non-PEs. The categorization of all reported AEs was performed by the authors prior to data analysis. Reported AEs were categorized based on specific terms as described below, and medical judgment by one author (ABK) the proposed categorizations were reviewed (ARB). Any disagreements were resolved by discussion. PEs included AEs with “pain” or similar wording in the AE term, such as lower back pain, and conditions considered likely to be associated with pain such as bone fractures and headache. A full overview of reported terms for the AEs categorized as PEs are listed in Supplemental Listing 1.
2.4. Pain event severity
All PEs were graded by severity as mild, moderate, or severe. This assessment was performed by investigators upon reporting based on the Common Terminology Criteria for Adverse Events (CTCAE) with increasing severity levels reflecting a higher impact on daily activities [19].
2.5. Pain event subtype
The PEs were further categorized into subtypes according to the nature of the pain, distinguished by its duration (acute or chronic) and its origin (musculoskeletal (MSK) or non-MSK). PEs were classified as acute or chronic according to the reported duration, with conditions with a duration of less than three months being defined as acute, and those persisting for more than three months being defined as chronic. PEs were further identified as chronic if the adverse term included the word “chronic,” or if they were inherently chronic in nature, such as chronic migraine or hip arthrosis. The classification of the condition's origin as MSK or non-MSK was determined based on the PE description, where all conditions involving joints, bones, muscles or regional conditions, such as back pain, were classified as MSK. Non-MSK PEs were pain events originating from other sources, such as headaches or stomach pain. This sub-classification generated four groups based on patients having at least one of the PE subtypes: acute-MSK, acute non-MSK, chronic MSK and chronic non-MSK. Reported terms for the AEs categorized according to subtype are listed in Supplemental Listing 2.
2.6. Exposures
Exposures were defined as clinical characteristics and medical history.
2.7. Medical History
For the analysis, a selection of medical history (MH) terms was defined. The selected MH terms comprised of three categories, “Psychiatric”, “Musculoskeletal” and “Neurological”. Author 1 (ABK) manually selected a number of terms, based on their known or suspected correlation with pain and their relevant incidence. Author 2 (ARB) reviewed the selected terms and both authors agreed on a final list. A full list of selected Preferred Terms is included in Supplemental Listing 3. Based on the recorded MH terms, a binary classification was applied to indicate the presence or absence of the condition for all patients.
2.8. Sub populations
For the analysis, patients were stratified into populations based on the presence or absence of at least one PE, or characteristics of the PE reported as outlined below and in Fig. 1.
For the overall study population, this generated two populations 1) The total study population 2) Patients reporting at least one PE (The PE population.
For the PE population, the patients were further divided in subgroups based on the characteristics of the reported PEs:
Severity: patients were categorized in three groups (mild, moderate, severe) based on the severity of their most severe PE.
Subtype: patients were grouped by the presence of at least one PE of a specific subtype (acute or chronic and MSK or non-MSK). Patients were able to be represented in more that one subtype group, in case several PEs within different subtypes were reported.
2.9. Outcome evaluation
The occurrence of at least one PE during the trial was defined as the outcome. For those experiencing at least one PE, secondary outcomes were the occurrence of specific PEs by severity and subtype. Exposures were clinical characteristics and the presence of selected medical history terms.
2.10. Statistical analyses
The analyses were performed on the two populations.
2.10.1. Total study population
To assess the association between clinical characteristics and recorded Medical History at baseline, and the occurrence of PEs during the study period, logistic regression models were employed. These models were adjusted for potential confounders, including age, sex and BMI, BL joint space width (JSW), BL KL grade and BL WOMAC pain scores. The available data was analysed, and no imputations or actions were taken concerning missing data. Odds Ratios (ORs) were calculated to quantify these associations, reporting the OR of the occurrence of at least one PE.
2.10.2. Pain Event population
To assess the association between clinical characteristics at baseline, and the occurrence PEs of a certain severity or subtype during the study period, logistic regression models were employed. These models were adjusted for potential confounders, including age, sex and BMI, BL joint space width (JSW), BL KL grade and BL WOMAC pain scores. Odds Ratios were calculated to quantify these associations, reporting the OR of the occurrence of the severity of the most severe PE and the occurrence of at least one specific PE subtype.
For improved interpretability of the regression coefficients, certain OR-variables were transformed: For age, the OR was expressed in 5-year increments and for BMI in 3 kg/m2 increments, both reflecting what is considered a clinically significant change. WOMAC pain scores were expressed in 10-point increments to reflect a clinical important change. It has been demonstrated that improvements larger than approximately 7.5 out of 100 on the WOMAC pain scale may be considered clinically important [20].
Statistical analyses were performed using R software (version 4.4.2, 2024, The R Foundation, Vienna, Austria). A two-tailed P < 0.05 was considered statistically significant, and no corrections for multiplicity were performed.
3. Results
A total of 2206 participants were randomized in the trials, and during the two-year study period, 8183 AEs were recorded. Of these, 27.6 % (N = 2258) were classified as PEs. PEs were reported in 1225 patients, accounting for 55.5 % of the total study population. Fig. 1 illustrates the distribution of patients according to two populations: The total study population and the PE population including the severity and subtypes of PE in the PE population.
The clinical characteristics and prevalence of medical history of patients in the total study population comparing those with one or more PEs vs. no PE showed no major differences and are displayed in Table 1.
Table 1.
Clinical characteristics in Total Study Population.
| Unit | Patients w. no PEs N = 981 |
Patients w. ≥1 PEs N = 1225 |
|
|---|---|---|---|
| Age | Years, mean (SD) | 64.3 (6.9) | 64,6 (6.7) |
| Sex | Males, n (%) | 369 (37.6) | 407 (33.2) |
| BMI | Kg/m2, mean (SD) | 29.3 (5.0) | 28.8 (4.9) |
| JSW, TK | Mm, mean (SD) | 3.41 (0.99) | 3.42 (0.99) |
| KL grade, TK | Grade 2, n (%) | 816 (83.2) | 1025 (83.7) |
| Grade 3, n (%) | 165 (16.8) | 200 (16.3) | |
| WOMAC, TK | 0–100, mean (SD) | 47.9 (14.6) | 48.8 (14.8) |
| Number of AEs | Count, mean (SD) | 5.4 (2.2) | 4.8 (2.9) |
| Medical history, prevalence | |||
| Insomnia | n (%) | 15 (1.5) | 48 (3.9) |
| Depression | n (%) | 41 (4.2) | 72 (5.9) |
| Anxiety | n (%) | 12 (1.2) | 15 (1.2) |
| Migraine | n (%) | 24 (2.4) | 51 (4.2) |
| Headache | n (%) | 55 (5.6) | 95 (7.8) |
| Fracture | n (%) | 166 (16.9) | 230 (18.8) |
| Musculoskeletal pain | n (%) | 20 (2.0) | 49 (4.0) |
| Back pain | n (%) | 129 (13.1) | 204 (16.7) |
| Sciatica | n (%) | 12 (1.2) | 23 (1.9) |
| Arthroplasty | n (%) | 16 (1.6) | 23 (1.9) |
Abbreviations: TK: Target Knee, BMI: Body Mass Index. SD: Standard Deviation. KL: Kellgren-Lawrence. JSW: Joint Space Width. WOMAC: Western Ontario and McMaster Universities Arthritis Index.
In the PE population, the median number of PEs was 1 (Min 1; Max 8), and the majority of the PE were graded moderate. The PEs predominantly comprised acute pain conditions accounting for 81.0 % (N = 1829) of all PEs. In terms of origin, 54.3 % were MSK (N = 1032) and 45.7 % were non-MSK (N = 1032).
3.1. Total study population
3.1.1. Associations of clinical characteristics with incidence of PEs
In the total study population, male patients were found to report fewer PEs compared to females as they had a 17 % decreased odds (OR: 0.83, 95 % CI: 0.70, 0.99, p = 0.04) of reporting at least one PEs compared to none, as shown in Fig. 2.
Fig. 2.
Odds ratios for the presence of at least one Pain Event vs. none.
Odds ratio for experiencing at least one Pain Event during the study period based on clinical characteristics. Logistic regression analysis adjusted for shown covariates. KL-grade and WOMAC score are referring to the target knee.
A higher BMI was linked to a lower likelihood of reporting PEs, with each 3-point increase in BMI resulting in a 6 %-points lower odds (OR: 0.94, 95 % CI: 0.89, 0.99, p = 0.02) of reporting one or more PEs.
No further associations with PE reporting were observed; Age, target knee (TK) WOMAC pain scores and TK KL scores showed no increased or decreased odds of reporting one or more PE compared to none.
3.1.2. Associations of medical history with incidence of PEs
As shown in Table 2 the presence of several conditions in medical history were associated with higher odds of experiencing PEs, especially certain psychiatric disorders and musculoskeletal disorders. Patients with insomnia had a 13 % increased odds of having at least one PE compared to none (OR: 1.13, 95 % CI: 1.05, 1.38, p = 0.003), while those with depression exhibited a 10 % increased odds (OR: 1.10, 95 % CI: [1.00, 1.38], p = 0.048). Anxiety, however, showed no associations with the odds of experiencing PEs (OR: 1.11, 95 % CI: [0.90, 1.60], p = 0.308).
Table 2.
Association between Medical History and incidence of at least one Pain Event in Total Study Population.
| Medical History Terms | OR [95 % CI] for having at least one PE vs zero PEs | P-value |
|---|---|---|
| Psychiatric: | ||
| Insomnia | 1.13 [1.05; 1.38] | 0.003 |
| Depression | 1.10 [1.00; 1.38] | 0.048 |
| Anxiety | 1.11 [0.90; 1.60] | 0.308 |
| Neurological: | ||
| Migraine | 1.10 [0.98; 1.58] | 0.071 |
| Headache | 1.05 [0.97; 1.58] | 0.179 |
| Musculoskeletal: | ||
| Fracture | 1.03 [0.98; 1.59] | 0.291 |
| Musculoskeletal pain | 1.10 [1.02; 1.30] | 0.013 |
| Back pain | 1.06 [1.00; 1.12] | 0.037 |
| Sciatica | 1.08 [0.91; 1.60] | 0.283 |
| Arthroplasty | 1.03 [0.88; 1.61] | 0.695 |
Within the musculoskeletal category, individuals with a medical history of musculoskeletal pain at baseline had a 10 % increased odds of having one or more PEs compared to none (OR: 1.10, 95 % CI: [1.02, 1.30], p = 0.013), and pre-study presence of back pain showed a 6 % increased odds of reported PE during the study (OR: 1.06, 95 % CI: [1.00, 1.12], p = 0.037). Prior fracture, arthroplasty and sciatica showed no impact on the odds of reporting one or more PEs compared to none. Within the neurological category, there were no significant association between migraine and headache and the odds of experiencing PEs, however there was a tendency towards patients with migraine having an increased odds of encountering PEs (OR: 1.10, 95 % CI: [0.98, 1.58], p = 0.071).
3.2. Pain Event Population
3.2.1. Associations of clinical characteristics with severity of PEs
As shown in Table 3, within the group of patients reporting at least one PE, increasing BMI was associated with risk of PE by severity. Higher BMI was incrementally associated with higher odds of moderate and severe PE compared to mild. Each 3-point of BMI increase was associated with 35 % higher odds (OR: 1.35, 95 % CI: 1.22, 1.49, p < 0.001) of experiencing at least one moderate PE, and with 63 % higher odds (OR: 1.63, 95 % CI: 1.41, 1.90, p < 0.001) of experiencing at least one severe PE.
Table 3.
Association between clinical characteristics and severity of the most severe Pain Event in Pain Event Population.
| OR [95 % CI] for most severe PE being mild, moderate or severe |
|||||
|---|---|---|---|---|---|
| Mild N = 307 | Moderate N = 776 | P-value | Severe N = 142 | P-value | |
| Age, pr 5 years | Ref | 0.96 [0.86; 1.06] | 0.42 | 1.05 [0.92; 1.19] | 0.45 |
| Sex, male | Ref | 1.39 [1.03; 1.88] | 0.03 | 0.88 [0.61; 1.27] | 0.49 |
| BMI, pr 3 kg/m2 | Ref | 1.35 [1.22; 1.49] | <0.001 | 1.63 [1.41; 1.90] | <0.001 |
| JSW, TK, pr mm | Ref | 0.89 [0.77; 1.03] | 0.11 | 0.86 [0.68; 1.09] | 0.20 |
| KL3, vs KL2, TK | Ref | 0.58 [0.41; 0.82] | 0.002 | 0.24 [0.11; 0.49] | <0.001 |
| WOMAC, TK pr 10 points | Ref | 0.99 [0.90; 1.08] | 0.75 | 0.99 [0.86; 1.15] | 0.91 |
OR: Odds Ratio, PE: Pain Event, CI: Confidence Interval, TK: Target Knee, BMI: Body Mass Index. SD: Standard Deviation. KL: Kellgren-Lawrence. JSW: Joint Space Width. WOMAC: Western Ontario and McMaster Universities Arthritis Index.
Patients with a KL grade of 3 versus 2 were less likely to report moderate or severe PEs, as they had a 42 % decreased odds of having a moderate PE (OR: 0.38, 95 % CI: 0.41, 0.82, p = 0.002) and a 76 % decreased odds of having a severe PE (OR: 0.34, 95 % CI: 0.11, 0.49, p < 0.001) when compared to mild PE.
Male patients were more prone to experience moderate PEs compared to mild with a 39 % increased odds (OR: 1.39, 95 % CI: 1.03, 1.88, p = 0.03), but there were no association between sex and the risk of reporting severe PEs.
Additional clinical variables including age, target knee joint-space width or WOMAC pain at baseline were not associated with an increased or decreased risk of reporting moderate or severe PEs as compared to mild.
3.2.2. Associations of clinical characteristics with subtype of PEs
Patients with a KL grade of 3 versus 2 had a 30 % decreased odds of having an acute, non-MSK PE (OR: 1.30, 95 % CI: 0.49, 0.97, p = 0.04) when compared to having different types of PEs, as shown in Table 4. There were no further associations with KL grade and the risk of experiencing PE of other subtypes.
Table 4.
Association between clinical characteristics and incidence of subtypes of Pain Events in Pain Event Population.
| OR [95 % CI] for the incidence of PEs being acute or chronic, musculoskeletal (MSK) or non-MSK |
||||||||
|---|---|---|---|---|---|---|---|---|
| ≥1 acute MSK PEs vs none N = 870 | P-value | ≥1 chronic MSK PEs vs none N = 356 | P-value | ≥1 acute non-MSK PEs vs none N = 959 | P-value | ≥1 chronic non-MSK PEs vs none N = 73 | P-value | |
| Age, pr 5 years | 0.96 [0.91, 1.01] | 0.11 | 0.98 [0.94, 1.04] | 0.56 | 0.88 [0.81, 0.96] | <0.001 | 1.06 [0.89, 1.26] | 0.51 |
| Sex, male | 1.10 [0.94, 1.28] | 0.23 | 1.00 [0.86, 1.16] | 0.97 | 0.80 [0.63, 1.02] | 0.08 | 0.66 [0.36, 1.12] | 0.14 |
| BMI, pr 3 kg/m2 | 0.98 [0.94, 1.03] | 0.42 | 0.99 [0.95, 1.04] | 0.78 | 0.95 [0.89, 1.01] | 0.11 | 0.93 [0.80, 1.07] | 0.34 |
| JSW, TK, pr mm | 1.01 [0.94, 1.09] | 0.77 | 1.00 [0.93, 1.08] | 0.92 | 0.90 [0.80, 1.01] | 0.08 | 0.84 [0.64, 1.09] | 0.19 |
| KL3, vs KL2, TK | 0.90 [0.72, 1.10] | 0.31 | 1.10 [0.91, 1.33] | 0.33 | 0.70 [0.49, 0.97] | 0.04 | 1.13 [0.6, 2.03] | 0.69 |
| WOMAC, TK pr 10 points | 0.99 [0.94, 1.04] | 0.64 | 1.02 [0.98, 1.07] | 0.35 | 1.02 [0.94, 1.09] | 0.67 | 1.07 [0.92, 1.24] | 0.40 |
OR: Odds Ratio, PE: Pain Event, CI: Confidence Interval, MSK: musculoskeletal, TK: Target Knee, BMI: Body Mass Index. KL: Kellgren-Lawrence. JSW: Joint Space Width. WOMAC: Western Ontario and McMaster Universities Arthritis Index.
It was observed that older patients had decreased odds of experiencing acute non-MSK PE; with a 5-year increase in age patients had a 12 % decreased odds of having an acute, non-MSK PE (OR: 0.88, 95 % CI: 0.81, 0.96, p < 0.001) when compared to having different types of PEs. There were no additional correlations between age and the risk of encountering PE of other subtypes.
Additional clinical variables including male sex, BMI, TK JSW and TK WOMAC pain showed no associations with subtype of PEs.
4. Discussion
The findings from this post-hoc analysis provide important insights into the predictors of unspecific pain events in patients with knee OA and highlights the complex interplay between clinical characteristics including medical history and the occurrence, severity, and subtype of pain events.
One of the principal findings is the association between BMI and the risk of experiencing a higher severity of PEs among participants who are reporting one or more PEs. Specifically, in the PE population a higher BMI was linked to increased odds of reporting moderate and severe PEs compared to mild. The incremental relationship between obesity and pain complaints in general, both acute and chronic, has been underscored by a large volume of studies indicating that pain complaints become more prevalent with a higher BMI [[21], [22], [23], [24]]. It is well known in the literature, that obesity is a significant risk factor for OA-related pain, likely due to increased mechanical stress on joints and potential inflammatory pathways associated with adipose tissue [25,26]. Furthermore, a higher BMI has also been proved to be a risk factor for other common types of pain events such as lower back pain [27] or headaches [28]. The relationship between obesity and more severe pain events is multifaceted, involving mechanical, inflammatory, metabolic, and psychological components, with all the different factors contributing to pain events becoming more severe and in to a higher extent conflicting with patient's daily life.
A converse finding in this study was an inverse relationship between BMI and the overall incidence of pain events in the total study population, where higher BMI was associated with lower odds of reporting one or more PEs. While the specific causes of this observation are unknown, this paradoxical finding may suggest that individuals with higher BMI might underreport pain or that other factors, such as pain tolerance or reporting bias, could be at play and warrants further investigation. The seemingly conflicting finding may not be generalizable to the broader knee OA population, but suggest the presence of a currently undescribed, complex interplay between several factors.
A notable discovery in the total study population is the association between certain medical disorders and the increased odds of experiencing PEs. The link between psychiatric disorders, such as insomnia and depression, and increased pain perception is well-documented in the literature. The comorbidity between sleep disturbances, defined as a broad array of sleep-related outcomes including insomnia, can exacerbate pain sensitivity and reduce pain tolerance, potentially due to the impact of sleep deprivation on the central nervous system and its role in pain modulation [29,30]. Similarly, depression is known to influence pain perception through both psychological and physiological pathways, including alterations in neurotransmitter levels and increased inflammatory responses [31,32]. Interestingly, anxiety was not found to have a significant association with a higher pain reporting in this study, which in previous research has been associated with pain disorders in general and pain severity [33,34].
Our findings indicate an association between insomnia and the odds of experiencing PEs. However, it is important to consider the potential for a bilateral association between these variables. It is evident that poor sleep quality and insomnia are associated with increased knee OA pain [35]. Similarly poor sleep is common in patients with OA [36], and sleep typically worsens with increasing severity of OA [37]. Hence, insomnia and many of the selected exposures may share bilateral associations with the pain outcomes underlining the complex interaction between variables and aiding further investigation of the causal relationship.
The study also found that male patients reported fewer PEs compared to females. The underlying causes were not assessed in this report, but hypothetically, one possible explanation is the potential differences in central pain processing mechanisms, which is consistent with previous research indicating sex differences in pain perception and reporting [38,39]. This could be attributed to biological differences, such as hormonal influences, or sociocultural factors affecting pain expression, resulting in an overall higher pain levels in females [40,41].
The study suggests that in the PE population male patients had a higher risk of reporting a moderate PE, but no clear associations with the risk of reporting a severe PE. This warrants further research to investigate the sex differences in relation to severity of pain events. However, these findings may suggest a gender-based underreporting by male patients when it comes to severity, where male patients are more prone to understate their pain resulting in a less severe grading [42,43]. This aligns with the majority of clinical, basic human, and rodent literature reports that females are more sensitive to pain [44]. Clinical studies find females are more likely than males to report pain [45] and report higher pain intensity [43,46].
The current study shows an association between more severe radiographic knee OA and a lower severity of pain events in the PE population, which could be attributed to several factors. One possible explanation is the phenomenon of pain adaptation, where individuals with more advanced OA may develop coping mechanisms or adaptations that reduce the perception of pain [47]. Another factor could be the variability in individual pain thresholds and the subjective nature of pain perception, which can lead to differences in reported pain severity despite similar structural damage [48]. Lastly, the discordance between structural damage and clinical manifestations is well known in the literature, with a poor correlation between radiographic findings and symptom severity in knee OA [49]. However, the correlation between radiographic severity of OA and other pain events is a topic that not fully investigated, warranting further research. Another interesting aspect of this analysis is the population existing of patients with chronic pain due to knee OA, for which predictors of new-pain onset may be different from those in the general population.
The lack of significant associations between other clinical variables and PEs underscores the multifactorial nature of pain events and highlights the need for a more comprehensive approach to understanding pain in patients with knee OA. Furthermore, the associations between certain medical disorders and PEs underscore the importance of considering a patient's comprehensive medical history when assessing their risk for experiencing pain in general and in the context of knee OA.
4.1. Limitations
This study has several limitations. As a post-hoc analysis, the findings are hypothesis-generating, necessitating cautious interpretation of causal relationships between clinical variables and pain events. Multiple testing has been applied without statistical adjustment, which could cause an increase in false positives with the number of tests conducted.
The definition of exposures and categorization of PEs based on reported terms were defined and categorized by the authors which introduces a hypothetical risk of bias. Consequently, the PE categorization approach may not capture the full complexity of pain experiences and overlook certain events that, while categorized as non-PE in this analysis, were indeed painful, such as influenza with headache and pronounced muscular tenderness. However, considering the size of the dataset, the general robustness of the analysis is considered unaffected by potentially minor misclassifications due to human error.
Additionally, the study population, derived from the included clinical trials, may not fully represent the broader OA population, potentially limiting the generalizability of the findings. Data was collected from 11 countries and 19 centres resulting in data collection by various instruments, even despite efforts for uniforming the process, which could influence exposures and outcomes. For example, collected medical history term could vary due to different diagnosis criteria. Conversely, as a classical large OA RCT dataset, it may reflect very well on future interventional RCTs in OA as opposed to as an epidemiological measure of the background OA population.
Despite efforts to adjust for known confounders, the potential for residual confounding remains, and it may affect the interpretation of the results.
5. Conclusion
This study highlights the complex interplay between clinical characteristics, medical history and pain in a population with chronic pain due to knee OA. An important finding include that a higher BMI was associated with higher odds of having reported a moderate or severe PE, however it was associated with a lower odds of reporting one or more PEs as compared to no AEs or any non-PE.
Patients with a medical history of insomnia, depression, musculoskeletal pain and back pain had a higher odds of experiencing at least one PE compared to no PE.
Male sex and KL grade 3 was found to be associated with lower odds of reporting several PEs, and KL grade 3 was associated with lower odds of reporting moderate and severe PEs. KL grade 3 and higher age was further associated with a lower odds of reporting acute non-MSK PEs.
Further research is needed to substantiate the observed findings and to evaluate the underlying pathways that may mediate these associations.
Patient consent for publication
N/A.
Author’s contributions
Conception and design: ABK, ARB. Literature search: ABK. Analysis and interpretation of the data: ABK, JMB, ISA, ARB. Drafting of the article: ABK. Critical revision of the article for important intellectual content: ABK, JMB, ISA, MAK, ARB. Final approval of the article: ABK, JMB, ISA, MAK, ARB.
Data availability statement
The data is publicly available.
The patient and public involvement statement
Patients or members of the public were not involved in the design of the trial.
Ethics approval
The two trials ((NCT00486434 and NCT00704847) were approved by all applicable Independent Review Boards, Ethics Committees, and regulatory bodies.
Funding
No specific funding was received for this study. The work of JMB is partially funded by The Innovation Foundation, Denmark. All authors were compensated through their regular salaries, and the funding source had no role in the study design, data collection, analysis, interpretation, or writing of the manuscript.
Declaration of competing interest
The authors declare that they have no competing interests.
Acknowledgments and affiliations
We thank all the participants and study staff who participated in the clinical trials on which this paper was based.
Handling Editor: Professor H Madry
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ocarto.2026.100746.
Contributor Information
Andrea Bak Kaaber, Email: aka@nbcd.com.
Jakob Mejdahl Bentin, Email: jmb@sanos.com.
Ida Sofie Adrian, Email: iad@nbcd.com.
Morten Asser Karsdal, Email: mk@nordicbio.com.
Asger Reinstrup Bihlet, Email: abi@nbcd.com.
Appendix A. Supplementary data
The following are the supplementary data to this article:
References
- 1.Litwic A., Edwards M.H., Dennison E.M., Cooper C. Epidemiology and burden of osteoarthritis. Br. Med. Bull. 2013;105(1):185–199. doi: 10.1093/BMB/LDS038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Qvist P., Bay-Jensen A.C., Christiansen C., Dam E.B., Pastoureau P., Karsdal M.A. The disease modifying osteoarthritis drug (DMOAD): is it in the horizon? Pharmacol. Res. 2008;58:1043–6618. doi: 10.1016/j.phrs.2008.06.001. (Print)):1-7. [DOI] [PubMed] [Google Scholar]
- 3.Karsdal M.A.A., Michaelis M., Ladel C., et al. Disease-modifying treatments for osteoarthritis (DMOADs) of the knee and hip: lessons learned from failures and opportunities for the future. Osteoarthr. Cartil. 2016;24(12):2013–2021. doi: 10.1016/j.joca.2016.07.017. [DOI] [PubMed] [Google Scholar]
- 4.Georgiev T., Angelov A.K. Modifiable risk factors in knee osteoarthritis: treatment implications. Rheumatol. Int. 2019;39(7):1145–1157. doi: 10.1007/S00296-019-04290-Z. [DOI] [PubMed] [Google Scholar]
- 5.Grässel S., Muschter D. Recent advances in the treatment of osteoarthritis. F1000Res. 2020;9 doi: 10.12688/F1000RESEARCH.22115.1/. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Thijssen E., Van Caam A., Van Der Kraan P.M. Obesity and osteoarthritis, more than just wear and tear: pivotal roles for inflamed adipose tissue and dyslipidaemia in obesity-induced osteoarthritis. Rheumatology. 2015;54(4):588–600. doi: 10.1093/RHEUMATOLOGY/KEU464. [DOI] [PubMed] [Google Scholar]
- 7.Sampath S.J.P., Venkatesan V., Ghosh S., Kotikalapudi N. Obesity, metabolic syndrome, and osteoarthritis—an updated review. Curr. Obes. Rep. 2023 12:3. 2023;12(3):308–331. doi: 10.1007/S13679-023-00520-5. [DOI] [PubMed] [Google Scholar]
- 8.Katz J.N., Arant K.R., Loeser R.F. Diagnosis and treatment of hip and knee osteoarthritis: a review. JAMA. 2021;325(6):568–578. doi: 10.1001/JAMA.2020.22171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Le Graverand M.P.H., Clemmer R.S., Redifer P., et al. A 2-year randomised, double-blind, placebo-controlled, multicentre study of oral selective iNOS inhibitor, cindunistat (SD-6010), in patients with symptomatic osteoarthritis of the knee. Ann. Rheum. Dis. 2013;72(2):187–195. doi: 10.1136/annrheumdis-2012-202239. [DOI] [PubMed] [Google Scholar]
- 10.Hochberg M.C., Guermazi A., Guehring H., et al. Effect of intra-articular sprifermin vs placebo on femorotibial joint cartilage thickness in patients with osteoarthritis: the FORWARD randomized clinical trial. JAMA. 2019;322(14):1360–1370. doi: 10.1001/JAMA.2019.14735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Karsdal M.A., Byrjalsen I., Alexandersen P., et al. Treatment of symptomatic knee osteoarthritis with oral salmon calcitonin: results from two phase 3 trials. Osteoarthr. Cartil. 2015;23(4):532–543. doi: 10.1016/j.joca.2014.12.019. [DOI] [PubMed] [Google Scholar]
- 12.Segal N.A., Nilges J.M., Oo W.M. Sex differences in osteoarthritis prevalence, pain perception, physical function and therapeutics. Osteoarthr. Cartil. 2024;32(9):1045–1053. doi: 10.1016/j.joca.2024.04.002. [DOI] [PubMed] [Google Scholar]
- 13.Tschon M., Contartese D., Pagani S., Borsari V., Fini M. Gender and sex are key determinants in osteoarthritis not only confounding variables. A systematic review of clinical data. J. Clin. Med. 2021;10(14) doi: 10.3390/JCM10143178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Iwane N., Hashizume H., Murata S., et al. Long-term prognosis and risk factors for low back pain-related disorders in the general population: a 7-Year Follow-Up of the Wakayama spine study. Spine Surg Relat Res. 2025;9(6) doi: 10.22603/SSRR.2025-0122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Thorlund J.B., Skarpsno E.S., Vestergaard J.J., et al. Sleep problems and insomnia are common and associated with pain intensity, number of comorbidities and analgesic use in patients with knee and hip osteoarthritis: a cross-sectional study using data from the good life with osteoarthritis in Denmark (GLA:D…. Rheumatol. Int. 2025;45(5) doi: 10.1007/S00296-025-05878-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lu W.S., Wang Y.Q., An Y., et al. Modifiable risk factors that mediate the effect of insomnia on the risk of low back pain: a network mendelian randomization study. Hereditas. 2024;161(1) doi: 10.1186/S41065-024-00341-Z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Karsdal M.A.A., Byrjalsen I., Alexandersen P., et al. Treatment of symptomatic knee osteoarthritis with oral salmon calcitonin: results from two phase 3 trials. Osteoarthr. Cartil. 2015;23(4):532–543. doi: 10.1016/j.joca.2014.12.019. [DOI] [PubMed] [Google Scholar]
- 18.Cancer Institute N . 2017. Common Terminology Criteria for Adverse Events (CTCAE) Common Terminology Criteria for Adverse Events (CTCAE) v5.0.https://www.meddra.org/ Published online. [Google Scholar]
- 19.National Cancer Institute . 2017. Common Terminology Criteria for Adverse Events (CTCAE)https://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/ctcae_v5_quick_reference_5x7.pdf Version 5.0. November 27. [Google Scholar]
- 20.Angst F., Aeschlimann A., Stucki G. Smallest detectable and minimal clinically important differences of rehabilitation intervention with their implications for required sample sizes using WOMAC and SF-36 quality of life measurement instruments in patients with osteoarthritis of the lower extremities. Arthritis Rheum. 2001;45(4):384–391. doi: 10.1002/1529-0131(200108)45:4<384::AID-ART352>3.0.CO;2-0. [DOI] [PubMed] [Google Scholar]
- 21.Higgins D.M., Kerns R.D., Brandt C.A., et al. Persistent pain and comorbidity among operation enduring freedom/operation Iraqi freedom/Operation new dawn veterans. Pain Med. 2014;15(5):782–790. doi: 10.1111/PME.12388. [DOI] [PubMed] [Google Scholar]
- 22.Hitt H.C., McMillen R.C., Thornton-Neaves T., Koch K., Cosby A.G. Comorbidity of obesity and pain in a general population: results from the southern pain prevalence study. J. Pain. 2007;8(5):430–436. doi: 10.1016/J.JPAIN.2006.12.003. [DOI] [PubMed] [Google Scholar]
- 23.Stone A.A., Broderick J.E. Obesity and pain are associated in the United States. Obesity. 2012;20(7):1491–1495. doi: 10.1038/OBY.2011.397. [DOI] [PubMed] [Google Scholar]
- 24.Mccarthy L.H., Bigal M.E., Katz M., Derby C., Lipton R.B. Chronic pain and obesity in elderly people: results from the einstein aging study. J. Am. Geriatr. Soc. 2009;57(1):115–119. doi: 10.1111/J.1532-5415.2008.02089.X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Elbaz A., Debbi E.M., Segal G., et al. Sex and body mass index correlate with Western Ontario and McMaster universities osteoarthritis index and quality of life scores in knee osteoarthritis. Arch. Phys. Med. Rehabil. 2011;92(10):1618–1623. doi: 10.1016/j.apmr.2011.05.009. [DOI] [PubMed] [Google Scholar]
- 26.Silverwood V., Blagojevic-Bucknall M., Jinks C., Jordan J.L., Protheroe J., Jordan K.P. Current evidence on risk factors for knee osteoarthritis in older adults: a systematic review and meta-analysis. Osteoarthr. Cartil. 2015;23(4):507–515. doi: 10.1016/J.JOCA.2014.11.019. [DOI] [PubMed] [Google Scholar]
- 27.Su C.A., Kusin D.J., Li S.Q., Ahn U.M., Ahn N.U. The association between body mass index and the prevalence, severity, and frequency of low back pain: data from the osteoarthritis initiative. Spine. 2018;43(12):848–852. doi: 10.1097/BRS.0000000000002601. [DOI] [PubMed] [Google Scholar]
- 28.Chai N.C., Scher A.I., Moghekar A., Bond D.S., Peterlin B.L. Obesity and headache: part I--a systematic review of the epidemiology of obesity and headache. Headache. 2014;54(2):219–234. doi: 10.1111/HEAD.12296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Herrero Babiloni A., Brazeau D., Jodoin M., et al. The impact of sleep disturbances on endogenous pain modulation: a systematic review and meta-analysis. J. Pain. 2024;25(4):875–901. doi: 10.1016/j.jpain.2023.10.023. [DOI] [PubMed] [Google Scholar]
- 30.Afolalu E.F., Ramlee F., Tang N.K.Y. Effects of sleep changes on pain-related health outcomes in the general population: a systematic review of longitudinal studies with exploratory meta-analysis. Sleep Med. Rev. 2018;39:82–97. doi: 10.1016/j.smrv.2017.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.IsHak W.W., Wen R.Y., Naghdechi L., et al. Pain and depression: a systematic review. Harv. Rev. Psychiatr. 2018;26(6):352–363. doi: 10.1097/HRP.0000000000000198. [DOI] [PubMed] [Google Scholar]
- 32.Sheng J., Liu S., Wang Y., Cui R., Zhang X. The link between depression and chronic pain: neural mechanisms in the brain. Neural Plast. 2017;2017 doi: 10.1155/2017/9724371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gureje O., Von Korff M., Kola L., et al. The relation between multiple pains and mental disorders: results from the world mental health surveys. Pain. 2008;135(1–2):82–91. doi: 10.1016/J.PAIN.2007.05.005. [DOI] [PubMed] [Google Scholar]
- 34.De Heer E.W., Gerrits M.M.J.G., Beekman A.T.F., et al. The association of depression and anxiety with pain: a study from NESDA. PLoS One. 2014;9(10) doi: 10.1371/JOURNAL.PONE.0106907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Kirkham-Wilson F., Westbury L., Bevilacqua G., Laskou F., Fuggle N., Dennison E. Relationships between pain, physical activity and sleep quality among older adults with radiographic knee osteoarthritis: findings from the hertfordshire cohort study. Aging Clin. Exp. Res. 2025;37(1) doi: 10.1007/S40520-025-03238-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Atkinson H.H., Cesari M., Kritchevsky S.B., et al. Predictors of combined cognitive and physical decline. J. Am. Geriatr. Soc. 2005;53(7):1197–1202. doi: 10.1111/J.1532-5415.2005.53362.X;CTYPE:STRING:JOURNAL. [DOI] [PubMed] [Google Scholar]
- 37.Allen K.D., Renner J.B., DeVellis B., Helmick C.G., Jordan J.M. Osteoarthritis and sleep: the johnston county osteoarthritis project. J. Rheumatol. 2008;35(6):1102. https://pmc.ncbi.nlm.nih.gov/articles/PMC4800314/ [PMC free article] [PubMed] [Google Scholar]
- 38.Fang W.H., Huang G.S., Chang H.F., et al. Gender differences between WOMAC index scores, health-related quality of life and physical performance in an elderly Taiwanese population with knee osteoarthritis. BMJ Open. 2015;5(9) doi: 10.1136/BMJOPEN-2015-008542. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Unruh A.M. Gender variations in clinical pain experience. Pain. 1996;65(2–3):123–167. doi: 10.1016/0304-3959(95)00214-6. [DOI] [PubMed] [Google Scholar]
- 40.Fillingim R.B., King C.D., Ribeiro-Dasilva M.C., Rahim-Williams B., Riley J.L. Sex, gender, and pain: a review of recent clinical and experimental findings. J. Pain. 2009;10(5):447–485. doi: 10.1016/J.JPAIN.2008.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Morgan J., Magwood K., Smith J., Jenkins M.R., McGregor A.J., Quesnelle K.M. Sex and gender differences in pain perception and management in clinical settings. Life. 2024;17(1) doi: 10.1080/26895293.2024.2367421. [DOI] [Google Scholar]
- 42.Wise E.A., Price D.D., Myers C.D., Heft M.W., Robinson M.E. Gender role expectations of pain: relationship to experimental pain perception. Pain. 2002;96(3):335–342. doi: 10.1016/S0304-3959(01)00473-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sluka K.A., Berkley K.J., O'Connor M.I., et al. Neural and psychosocial contributions to sex differences in knee osteoarthritic pain. Biol. Sex Differ. 2012 3:1. 2012;3(1):26. doi: 10.1186/2042-6410-3-26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Mogil J.S. Qualitative sex differences in pain processing: emerging evidence of a biased literature. Nat. Rev. Neurosci. 2020;21(7):353–365. doi: 10.1038/S41583-020-0310-6. [DOI] [PubMed] [Google Scholar]
- 45.Ruau D., Liu L.Y., Clark J.D., Angst M.S., Butte A.J. Sex differences in reported pain across 11,000 patients captured in electronic medical records. J. Pain. 2012;13(3):228–234. doi: 10.1016/j.jpain.2011.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Fillingim R.B., King C.D., Ribeiro-Dasilva M.C., Rahim-Williams B., Riley J.L. Sex, gender, and pain: a review of recent clinical and experimental findings. J. Pain. 2009;10(5):447–485. doi: 10.1016/j.jpain.2008.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Mease P.J., Hanna S., Frakes E.P., Altman R.D. Pain mechanisms in osteoarthritis: understanding the role of central pain and current approaches to its treatment. J. Rheumatol. 2011;38(8):1546–1551. doi: 10.3899/JRHEUM.100759. [DOI] [PubMed] [Google Scholar]
- 48.Sethi V., Anand C., Della Pasqua O. Clinical assessment of osteoarthritis pain: contemporary scenario, challenges, and future perspectives. Pain Ther. 2024;13(3):391. doi: 10.1007/S40122-024-00592-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Bedson J., Croft P.R. The discordance between clinical and radiographic knee osteoarthritis: a systematic search and summary of the literature. BMC Muscoskelet. Disord. 2008;9 doi: 10.1186/1471-2474-9-116. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data is publicly available.