The relationship between symptom duration in lumbar disc herniation and the prevalence of hip and knee osteoarthritis: A cross-sectional study Authors: Dusan Ondirko, Peter Solar, Tomas Valousek, Petra Wesela, Natalia Jankanicova, Radim Jancalek Published: 2026 ABSTRACT AbstractObjectiveDegenerative spine disease commonly presents as lumbar disc herniation (LDH), a major cause of radiculopathy. Altered biomechanics may contribute to secondary joint pathology; however, the relationship between LDH-associated radiculopathy and hip or knee osteoarthritis (OA) remains unclear.Research questionThis study evaluated the association between symptoms due to LDH and the presence and laterality of hip and knee symptomatic radiographic OA.Material and methodsWe analyzed data from a prospectively designed institutional study involving patients undergoing microdiscectomy for LDH. Consequently, this study evaluated exclusively symptomatic radiographic OA. Preoperative assessment included pain intensity, disability and motor deficit. Based on predefined clinical criteria (joint pain, stiffness, or restricted range of motion), symptomatic patients underwent targeted orthopedic and radiographic assessment for hip and knee OA.ResultsA total of 1344 patients were included. Longer duration of LDH-associated symptoms was associated with a higher prevalence of hip and knee symptomatic radiographic OA. A significantly higher prevalence of hip (OR = 34.9, 95% CI (10.0; 121.9) and knee (OR = 6.5, 95% CI (2.0; 20.7) symptomatic radiographic OA was observed in patients with a radicular motor deficit lasting >24 months versus those with symptoms ≤1 month.Radiculopathy laterality corresponded with ipsilateral joint involvement.Discussion and conclusionThe study demonstrates that a significant association appears to exist between the duration of radicular symptoms and the prevalence of lower limb symptomatic radiographic OA. However, because this study population consists entirely of surgically treated individuals, these findings may limit generalizability to the broader, non-surgical population.Keywords: Lumbar disc herniation, Degenerative disc disease, Microdiscectomy, Low back pain, Osteoarthritis Introduction Degenerative disease of the spine is a multifactorial condition that typically manifests between the ages of 30 and 50, affecting all components of the spinal motion segment, including the intervertebral discs, vertebral bodies, facet joints, ligaments, and paraspinal muscles. Its etiology reflects a complex interplay between genetic predisposition and environmental factors (Modic and Ross, 2007). Among the earliest and most common manifestations is degenerative disc disease (DDD), which frequently presents as low back pain (LBP) (Ravindra et al., 2018). Progression of DDD may lead to disc herniation, resulting in radicular symptoms such as radiating pain, sensory disturbances, and muscle weakness in the distribution of one or more lumbosacral nerve roots (Amin et al., 2017). Spinal pathologies, such as lumbar disc herniation (LDH), can lead to secondary functional alterations that affect the biomechanics of the musculoskeletal system. Conversely, lower limb osteoarthritis (OA) itself is well known to alter gait patterns, restrict joint mobility, and modify spinopelvic alignment, which may subsequently contribute to or exacerbate spinal symptoms. Importantly, pre-existing hip or knee OA may independently drive or reinforce lumbar spinal pathology by shifting the biomechanical burden upward along the kinetic chain. This chronic, asymmetric mechanical stress may not only accelerate intervertebral disc degeneration but could also render subsequent LDH symptoms less responsive to conservative therapies, potentially leading to delayed clinical recovery, prolonged symptom duration, and ultimately, a higher rate of delayed surgical interventions. This suggests a complex, bidirectional biomechanical interdependency where spinal and joint pathologies co-exist and potentially reinforce one another (Redmond et al., 2015; Tateuchi et al., 2016). OA, another highly prevalent musculoskeletal disorder, is defined as a degenerative disease primarily affecting articular cartilage and associated joint structures (Abramoff and Caldera, 2020). In recent decades, numerous studies have explored the association between degenerative spinal disease and knee OA (Amarasinghe et al., 2023; Katsumi et al., 2023; Govil et al., 2022; Bindawas et al., 2015). Concurrent involvement of the hip and spine is frequently observed, particularly in older adults, and is often characterized by overlapping clinical symptoms. Based on these findings, the present study aimed to evaluate the association without implying direct or exclusive causality between radicular symptoms due to LDH and the presence of radiographic OA in the hip and knee joints. Furthermore, we investigated whether there is an association between the side of radicular symptoms and the laterality of hip or knee radiographic OA. We hypothesized that a significant biomechanical interdependency exists, wherein a prolonged duration of lumbar radicular symptoms and motor deficits is associated with an increased prevalence and greater radiographic severity of hip and knee symptomatic OA; furthermore, we hypothesized that unilateral radicular symptoms exhibit a side-specific, localized spatial association with the laterality of lower limb joint degeneration, potentially reflecting a mutual reinforcement of both spinal and joint pathologies. Materials and methods Study design This study was conducted within the framework of a prospectively designed institutional study primarily focused on the microbiological investigation of herniated disc tissues and was approved by the Institutional Review Board of St. Anne's University Hospital Brno (Approval No. 33V/2015). Patient enrollment, preoperative clinical protocols, and data collection were performed prospectively between January 2018 and July 2025. The primary outcome of this study was to evaluate the relationship between the duration of preoperative LBP, radicular pain, and motor deficits and the presence of hip and knee symptomatic radiographic OA. Further evaluations concerning the laterality of joint involvement and specific radiographic severity grades were considered exploratory and supplementary analyses. Patient selection Patients were included if they met the following criteria: radiologically confirmed LDH on magnetic resonance imaging or computed tomography, and persistent radicular symptoms refractory to at least 6 weeks of conservative treatment. In cases of severe radicular pain or moderate-to-severe motor deficits, patients underwent surgical intervention within hours or days of symptom onset. Patients were systematically screened during enrollment, and exclusion criteria included inconclusive imaging findings, preexisting rheumatologic or other systematic inflammatory disease, preexisting knee or hip surgery (either due to trauma or congenital pathology), the presence of a hip or knee implant or arthroplasty, spinal stenosis, or instability requiring stabilization. Written informed consent was obtained from all eligible participants prior to enrollment. Patient characteristics Demographic data included sex, age, body mass index (BMI), workload (light, heavy, sedentary, or retirement pension), and the involved intervertebral levels. Clinical data comprised the duration of LBP, radicular pain, and motor deficits, as well as the presence of hip and/or knee joint radiographic OA. For patients with a history of recurrent or intermittent symptoms, the duration was strictly defined as the time elapsed since the onset of the current, persistent episode, intermittent pain, or an acute exacerbation of pain that directly led to the surgical indication. Preoperative assessment of radicular pain, disability, and motor deficit After meeting the inclusion criteria and accepting participation in the study, patients completed the standardized questionnaires one day prior to the planned microdiscectomy. The intensity of LBP and referred leg pain was quantified by the Numeric Rating Scale (NRS). The overall disability was assessed with the Oswestry Disability Questionnaire (ODQ). The degree of disability was then converted into the Oswestry Disability Index (ODI), which is calculated as a percentage of the total score from all questions in the ODQ. Motor deficit was assessed and graded with the Medical Research Council (MRC) scale. The severity of the motor deficit was then divided into severe (MRC 0-2), mild (MRC 3-4), or absent (MRC 5). Presence of hip and knee radiographic osteoarthritis Patients with clinical suspicion of OA characterized by localized hip or knee joint symptoms including pain, restricted range of motion, stiffness, or joint swelling were indicated for radiographic assessment of the respective joints. Patients presenting with these symptoms or a documented history of OA were referred for an orthopedic evaluation and bilateral X-ray imaging. Bilateral radiographs were obtained routinely for all referred patients using a standardized institutional imaging protocol that remained consistent throughout the entire study period. Consequently, throughout the entire manuscript, the prevalence of hip and knee OA is strictly defined as and refers to the prevalence of symptomatic radiographic OA, as asymptomatic joint degeneration was not radiologically evaluated. The diagnosis and radiographic severity of hip or knee OA were then verified by a board-certified orthopedist blinded to both the laterality and duration of the radicular symptoms. Importantly, the blinded orthopedic assessments were systematically cross-referenced and verified to be in strict correlation with the standard, official radiological evaluations performed by institutional radiologists at the time of imaging. In cases of borderline radiographic findings, a consensus agreement between the orthopedist and the radiological report was utilized to establish the definitive Kellgren-Lawrence grade. Radiographic OA severity was categorized into mild (Grades I–II) and moderate-to-severe (Grades III–IV) for the purpose of statistical analysis. This categorization allowed for a comparison between early structural changes and advanced joint degeneration in relation to the duration of LDH symptoms. Statistical analysis Continuous variables were represented as medians and interquartile ranges (IQRs), and discrete parameters as counts (N) and corresponding percentages. Since the normality of continuous variables was not met in all groups, as indicated by the Shapiro-Wilk test, the Kruskal-Wallis test was employed, followed by a post-hoc Dunn's test with the Bonferroni correction. For categorical variables, the Chi-square test with Yates' continuity correction and subsequent post-hoc Chi-square test with Bonferroni correction was used. Logistic regression with adjustment for age, BMI, and workload was used to evaluate the relationship between radiographic OA of the hip and knee and the duration of lower back pain, radicular pain, and motor deficit. For evaluation of the relationship between radiographic OA severity represented by grades of the hip and knee and the duration of LBP, radicular pain, and neurological deficit, Firth's bias-reduced penalized-likelihood logistic regression, also adjustment for age, BMI, and workload, was employed. For post hoc comparison of groups, Bonferroni correction was used. The significance level was defined at p < 0.05. Statistical analysis was performed using R version 4.2.3 (R Core Team, 2021; R Foundation for Statistical Computing, Vienna, Austria). Results In our cohort study, we enrolled 1344 patients with LDH indicated for lumbar microdiscectomy (Fig. 1). We divided patients according to duration of preoperative leg pain, LBP, or motor deficit into four distinct, mutually exclusive groups: ≤1 month; 2-6 months; 7-24 months; and >24 months. Flowchart of patient enrollment, screening, and group allocation. Demographic data Baseline patient characteristics are shown in Table 1. The cohort was 56.8% male (n = 763) and 43.2% female (n = 581), with a median age of 48.0 years (IQR 38.0; 59.0) and median overall BMI of 27.2 kg/m2 (IQR, 24.5-30.4). Regarding work status, 19.8% patients (n = 266) had light workload, 37.4% (n = 502) heavy workload, 26.9% (n = 361) sedentary work, and 16.0% (n = 215) were retired. Table 1Basic clinical characteristics of the included patients.Table 1TotalMenWomenMen, N (%)1344 (100%)763 (56.8%)581 (43.2%)Age, y48.0 (38.0; 59.0)48.0 (38.0; 58.0)47.0 (38.0; 59.0)BMI, kg/m227.2 (24.5; 30.4)27.4 (24.7; 30.3)26.8 (24.0; 30.5)Workload, N (%)Light266 (19.8%)132 (17.3%)134 (23.1%)Heavy502 (37.4%)322 (42.2%)180 (31.0%)Sedentary361 (26.9%)190 (24.9%)171 (29.4%)Retirement pension215 (16.0%)119 (15.6%)96 (16.5%)LDH, N (%)L1/27 (0.5%)5 (0.6%)2 (0.3%)L2/342 (3.1%)23 (3.0%)19 (3.3%)L3/4108 (8.0%)73 (9.6%)35 (6.0%)L4/5576 (42.9%)322 (42.2%)254 (43.7%)L5/S1611 (45.5%)340 (44.6%)271 (46.6%)Coxarthrosis of the hip, N (%)114 (8.5%)61 (8.0%)53 (9.1%)Gonarthrosis of the knee, N (%)82 (6.1%)37 (4.8%)45 (7.7%) Basic clinical characteristics of the included patients. Duration of preoperative low back pain in relation to hip and knee symptomatic radiographic osteoarthritis Patients with preoperative LBP showed a higher incidence of hip symptomatic radiographic OA with increasing symptom duration. Hip symptomatic radiographic OA was more frequent in patients with LBP lasting 7-24 months (OR = 3.5, 95% CI (1.0; 12.3), p = 0.046) and >24 months (OR = 11.9, 95% CI (3.4; 41.5), p < 0.001) compared to those with LBP ≤1 month. Similarly, the prevalence of hip symptomatic radiographic OA was higher in patients with LBP >24 months versus 2-6 months (OR = 6.6, 95% CI (3.1; 13.9), p < 0.001) (Fig. 2; Table 6). Prevalence of hip radiographic osteoarthritis according to the durations of preoperative low back pain, radicular leg pain, and neurological deficit. The charts display percentages (%) representing the proportion of patients with versus without hip osteoarthritis within each cohort. The red segments of the bars indicate the proportion of patients with hip osteoarthritis, whereas blue segments indicate patients without hip osteoarthritis. Asterisks denote statistically significant differences between groups (*p < 0.05; **p < 0.01; ***p < 0.001). Table 6 Pairwise comparisons of osteoarthritis risk (hip and knee OA) based on the duration of preoperative symptoms and neurological deficits.Table 6Duration of Preoperative LBP in Relation to Hip OAOR (95% CI)7-24 months2-6 months≤1 month> 24 months3.4 (1.6; 7.1) p < 0.0016.6 (3.1; 13.9) p < 0.00111.9 (3.4; 41.5) p < 0.0017–24 months-2.0 (0.9; 4.1) p = 0.1023.5 (1.0; 12.3) p = 0.0462–6 months--1.8 (0.5; 6.3) p = 0.999Duration of Preoperative LBP in Relation to Knee OAOR (95% CI)7-24 months2-6 months≤1 month> 24 months3.2 (1.4; 7.3) p = 0.0014.6 (2.1; 10.3) p < 0.0015.7 (1.7; 19.8) p = 0.0017–24 months-1.5 (0.6; 3.3) p = 0.9991.8 (0.5; 6.4) p = 0.9992–6 months--1.2 (0.4; 4.3) p = 0.999Duration of Preoperative Radicular Pain in Relation to Hip OAOR (95% CI)7-24 months2-6 months≤1 month> 24 months7.6 (3.0; 19.1) p < 0.00111.0 (4.9; 24.7) p < 0.00127.4 (7.8; 96.2) p < 0.0017–24 months-1.5 (0.7; 3.0) p = 0.9993.6 (1.1; 12.1) p = 0.0322–6 months--2.5 (0.8; 7.7) p = 0.206Duration of Preoperative Radicular Pain in Relation to Knee OAOR (95% CI)7-24 months2-6 months≤1 month> 24 months2.9 (1.1; 7.9) p = 0.0323.9 (1.6; 9.4) p < 0.0015.0 (1.5; 16.3) p = 0.0027–24 months-1.3 (0.6; 3.0) p = 0.9991.7 (0.6; 5.3) p = 0.9992–6 months--1.3 (0.5; 3.6) p = 0.999Duration of Preoperative Motor Deficit in Relation to Hip OAOR (95% CI)7-24 months2-6 months≤1 month> 24 months9.4 (3.6; 24.6) p < 0.00112.7 (5.6; 28.4) p < 0.00134.9 (10.0; 121.9) p < 0.0017–24 months-1.3 (0.6; 2.9) p = 0.9993.7 (1.1; 12.7) p = 0.0312–6 months--2.8 (0.9; 8.6) p = 0.106Duration of Preoperative Motor Deficit in Relation to Knee OAOR (95% CI)7-24 months2-6 months≤1 month> 24 months6.1 (1.9; 19.3) p < 0.0013.9 (1.7; 9.1) p < 0.0016.5 (2.0; 20.7) p < 0.0017–24 months-0.6 (0.2; 1.7) p = 0.9991.1 (0.3; 3.8) p = 0.9992–6 months--1.7 (0.6; 4.6) p = 0.999 Pairwise comparisons of osteoarthritis risk (hip and knee OA) based on the duration of preoperative symptoms and neurological deficits. Comparable conclusions were observed for knee symptomatic radiographic OA, with longer LBP duration associated with higher prevalence (Fig. 3; Table 6). Prevalence of knee radiographic osteoarthritis according to the duration of preoperative low back pain, radicular leg pain, and neurological deficit. The charts display percentages (%) representing the proportion of patients with versus without knee osteoarthritis within each cohort. The red segments of the bars represent the proportion of patients with knee osteoarthritis, and the blue segments represent patients without knee osteoarthritis. Asterisks indicate statistically significant differences between groups (*p < 0.05; **p < 0.01; ***p < 0.001). Patients with varying LBP duration also differed significantly in age (p = 0.014), (p = 0.003), workload (p = 0.001), and pain intensity (p = 0.013). No statistically significant differences in distribution of patient sex were observed between the groups (Table 2, Table 3, Table 4). To account for potential confounding effects of age, BMI and workload, logistic regression adjustments were performed. Table 2Differences in age, gender and BMI between patients with different duration of preoperative low back pain, radicular leg pain and neurological deficit caused by lumbar disc herniation.Table 2Duration of preoperative low back painAgeMedianIQRp-value≤1 month47.039.0; 60.00.0142-6 months46.037.0; 56.07-24 months51.038.0; 61.0>24 months51.040.0; 58.0Duration of preoperative leg painAgeMedianIQRp-value≤1 month45.037.0; 57.50.0082-6 months47.038.0; 57.07-24 months53.039.0; 62.0>24 months52.042.0; 58.0Duration of preoperative neurological deficitAgeMedianIQRp-value≤1 month44.037.0; 56.250.0032-6 months47.038.0; 58.007-24 months53.039.0; 62.00>24 months52.042.0; 58.00Duration of preoperative low back painGenderMaleFemalep-value≤1 month122 (59.8%)82 (40.2%)0.8212-6 months344 (56.1%)269 (43.9%)7-24 months186 (56.5%)143 (43.5%)>24 months111 (56.1%)87 (43.9%)Duration of preoperative radicular leg painGenderMaleFemalep-value≤1 month154 (58.6%)109 (41.4%)0.7922-6 months415 (55.6%)332 (44.4%)7-24 months136 (57.9%)99 (42.1%)>24 months58 (58.6%)41 (41.4%)Duration of preoperative neurological deficitGenderMaleFemalep-value≤1 month171 (57.8%)125 (42.2%)0.8582-6 months415 (55.8%)329 (44.2%)7-24 months117 (57.6%)86 (42.4%)>24 months60 (59.4%)41 (40.6%)Duration of preoperative low back painBMIMedianIQRp-value≤1 month26.223.6; 29.30.0032-6 months27.124.7; 30.37-24 months27.824.9; 31.2>24 months27.224.4; 30.4Duration of preoperative radicular leg painBMIMedianIQRp-value≤1 month26.524.1; 29.20.0012-6 months27.124.4; 30.47-24 months27.825.2; 31.4>24 months27.324.6; 30.2Duration of preoperative neurological deficitBMIMedianIQRp-value≤1 month26.423.9; 29.20.0012-6 months27.324.5; 30.87-24 months27.525.2; 31.3>24 months27.224.4; 30.1 Differences in age, gender and BMI between patients with different duration of preoperative low back pain, radicular leg pain and neurological deficit caused by lumbar disc herniation. Table 3Type of workload in patients with different duration of preoperative low back pain, radicular leg pain and neurological deficit caused by lumbar disc herniation.Table 3Duration of preoperative low back painWorkloadLightHeavySedentaryRetirement pensionp-value≤1 month43 (21.1%)83 (40.7%)45 (22.1%)33 (16.2%)0.0012-6 months117 (19.1%)233 (38.0%)173 (28.2%)90 (14.7%)7-24 months50 (15.2%)130 (39.5%)81 (24.6%)68 (20.7%)>24 months56 (28.3%)56 (28.3%)62 (31.3%)24 (12.1%)Duration of preoperative radicular leg pain≤1 month53 (20.2%)108 (41.1%)63 (24.0%)39 (14.8%)0.0012-6 months139 (18.6%)282 (37.8%)208 (27.8%)118 (15.8%)7-24 months39 (16.6%)88 (37.4%)59 (25.1%)49 (20.9%)>24 months35 (35.4%)24 (24.2%)31 (31.3%)9 (9.1%)Duration of preoperative neurological deficit≤1 month61 (20.6%)120 (40.5%)75 (25.3%)40 (13.5%)0.0012-6 months137 (18.4%)282 (37.9%)205 (27.6%)120 (16.1%)7-24 months33 (16.3%)77 (37.9%)49 (24.1%)44 (21.7%)>24 months35 (34.7%)23 (22.8%)32 (31.7%)11 (10.9%) Type of workload in patients with different duration of preoperative low back pain, radicular leg pain and neurological deficit caused by lumbar disc herniation. Table 4Intensity of preoperative low back pain, disability, and radicular leg pain.Table 4Duration of preoperative low back painNRS backMedianIQRp-value≤1 month5.03.0; 6.00.0132-6 months5.04.0; 6.07-24 months5.04.0; 6.0>24 months5.04.0; 6.0Duration of preoperative low back painOswestry Disability IndexMedianIQRp-value≤1 month40.028.0; 58.00.6272-6 months44.032.0; 54.07-24 months44.032.0; 54.0>24 months46.032.5; 56.0Duration of preoperative radicular leg painNRS legsMedianIQRp-value≤1 month6.05.0; 7.00.0542-6 months6.04.0; 7.07-24 months6.05.0; 7.0>24 months5.04.0; 6.5 Intensity of preoperative low back pain, disability, and radicular leg pain. Duration of preoperative radicular pain in relation to hip and knee symptomatic radiographic osteoarthritis Similar to LBP, patients with longer duration of radicular pain showed higher prevalence of hip and knee radiographic OA. Hip radiographic OA was more common in patients with radicular pain lasting for 7-24 month compared to those with radicular pain ≤1 month (OR = 3.6, 95% CI (1.1; 12.1), p = 0.032) and the same for >24 months (OR = 27.4, 95% CI (7.8; 96.2), p < 0.001). Likewise, patients with radicular pain >24 months had higher hip radiographic OA prevalence than those with 7-24 months of pain (OR = 7.6, 95% CI (3.0; 19.1), p < 0.001) (Fig. 2; Table 6). Longer-duration radicular pain was also associated with higher prevalence of knee symptomatic radiographic OA (Fig. 3; Table 6). Patients with varying durations of preoperative radicular pain differed significantly in age (p = 0.008), BMI (p = 0.001), and workload (p = 0.001), but not in leg pain intensity and gender (Table 2, Table 3). Logistic regression was adjusted for age, BMI, and workload as potential confounders. Duration of preoperative motor deficit in relation to hip and knee symptomatic radiographic osteoarthritis Longer-duration motor deficit in patients with LDH was associated with a higher prevalence of hip and knee radiographic OA. Hip symptomatic radiographic OA was more frequent in patients with motor deficit 7-24 months compared to those with motor deficit ≤1 month (OR = 3.7, 95% CI (1.1; 12.7), p = 0.031) and the same for >24 months (OR = 34.9, 95% CI (10.0; 121.9), p < 0.001) compared to ≤1 month. Similarly, patients with motor deficit >24 months had higher hip symptomatic radiographic OA prevalence than those with motor deficit 2-6 months (OR = 12.7, 95% CI (5.6; 28.4), p < 0.001) and 7-24 months (OR = 9.4, 95% CI (3.6; 24.6), p < 0.001) (Fig. 2; Table 6). For knee symptomatic radiographic OA, prevalence was higher in patients with motor deficit >24 months compared to ≤1 month (OR = 6.5, 95% CI (2.0; 20.7), p < 0.001), 2-6 months (OR = 3.9, 95% CI (1.7; 9.1), p < 0.001) or 7-24 months (OR = 6.1, 95% CI (1.9; 19.3), p < 0.001) (Fig. 3; Table 6). Patients with different durations of preoperative neurological deficit also differed significantly in age (p = 0.003), BMI (p = 0.001), workload (p = 0.001) and severity of motor deficit (MRC scale; p < 0.001) (Table 2, Table 3, Table 5). Table 5Severity of motor deficit in patients with lumbar disc herniation-induced preoperative neurological deficit of varying duration.Table 5Duration of preoperative neurological deficitSeverity of motor deficitp-valueMRC 0-2MRC 3-4MRC 5≤1 month53 (17.9%)201 (67.9%)42 (14.2%)<0.0012-6 months33 (4.4%)404 (54.3%)307 (41.3%)7-24 months5 (2.5%)116 (57.1%)82 (40.4%)>24 months1 (1.0%)77 (76.2%)23 (22.8%) Severity of motor deficit in patients with lumbar disc herniation-induced preoperative neurological deficit of varying duration. Association between the side of unilateral hip and knee symptomatic radiographic osteoarthritis and radicular symptoms We observed a significant side-specific association between radicular symptoms and unilateral symptomatic radiographic OA of the hip and knee. This specific analysis was strictly restricted to patients presenting with both unilateral radicular symptoms and unilateral joint involvement; patients with bilateral symptoms or bilateral OA were excluded from this subset. Patients with LDH-associated unilateral radicular symptoms more frequently exhibited ipsilateral, rather than contralateral, hip and knee radiographic OA (p < 0.001 and p = 0.001, respectively) (Fig. 4). Association between unilateral hip and knee osteoarthritis and the side of radicular symptoms. The charts display percentages (%) representing the proportion of left versus right joint osteoarthritis within each cohort. Dark blue bars indicate osteoarthritis affecting the left lower limb, whereas light blue bars indicate osteoarthritis affecting the right lower limb. Subgroup denominators (n) are presented on the X-axis. Asterisks denote statistically significant differences between groups (**p < 0.01; ***p < 0.001). Degree of unilateral hip and knee symptomatic radiographic osteoarthritis according to the duration of radicular symptoms For the investigation of this hypothesis, only a subgroup of patients with unilateral hip symptomatic radiographic OA (n = 77) and unilateral knee symptomatic radiographic OA (n = 66) were included. Due to the small number of patients with a duration of LDH-related symptoms ≤1 month, this subgroup was merged with patients reporting symptoms lasting 2-6 months. While certain tendencies were observed, these findings must be interpreted with caution due to the limited size of the stratified subgroups, which resulted in very wide confidence intervals. Patients with an LBP duration exceeding 24 months exhibited more grades from worse degree category compared to the patients with an LBP duration ≤6 months (OR = 14.1; 95% CI (1.4; 142.0), p = 0.020) (Fig. 5). The group of patients with radicular leg pain lasting for more than 24 months exhibited a higher proportion of cases with OA of degrees III–IV than the group with pain lasting for ≤6 months (OR = 10.5, 95% CI (1.3; 86.1), p = 0.024) and the same for the group 7-24 months in comparison to ≤6 months (OR = 9.4, 95% CI (1.3; 68.0), p = 0.021) (Fig. 5). Similar results were found for the patients with an LDH-associated neurological deficit exceeding 24 months compared to the group ≤6 months (OR = 34.3, 95% CI (2.7; 442.4), p = 0.003) and for the group of the neurological deficit lasting for 7-24 months compared to the ≤6 months (OR = 14.7, 95% CI (1.3; 170.7), p = 0.027) (Fig. 5). No association was found between the duration of LDH-related symptoms and the severity of unilateral knee symptomatic radiographic OA. Association between the duration of preoperative low back pain, radicular leg pain, and neurological deficit and the severity grade of unilateral hip osteoarthritis. The charts display percentages (%) representing the proportion of mild (Grades I–II) versus moderate-to-severe (Grades III–IV) osteoarthritis within each cohort. Subgroup denominators (n) are presented on the X-axis. Asterisks indicate statistically significant differences between groups (*p < 0.05; **p < 0.01). Discussion In our cohort study of consecutive patients indicated for microdiscectomy, we found a significant association between the duration of preoperative symptoms related to LDH and the presence of symptomatic radiographic OA in the hip and knee joints. Specifically, a longer duration of LDH-associated symptoms was linked to a higher prevalence of hip and knee symptomatic radiographic OA among patients indicated for microdiscectomy. Furthermore, we observed a lateral association with the side of unilateral symptomatic radiographic OA in the hip or knee, aligning with the side of radicular symptoms. Understanding this association is clinically actionable as it highlights a potential clinical link to joint degeneration. If prolonged radiculopathy (exceeding 24 months) is linked to a 39% prevalence of hip symptomatic radiographic OA, as seen in our cohort, it underscores a substantial clinical overlap between these conditions. However, given the cross-sectional nature of our data, this association should not be interpreted as an indication that early surgical decompression would prevent joint degeneration; longitudinal studies are strictly required to investigate whether addressing spinal symptoms chronologically impacts secondary mechanical stress on weight-bearing joints. This clinical link appears robust regarding symptomatic radiographic OA prevalence; however, corresponding findings regarding the radiological severity grades are associated with very wide confidence intervals and must be strictly viewed as preliminary, exploratory tendencies rather than definitive conclusions, owing to the small, stratified subgroups. Importantly, the differences between groups regarding symptom durations in age, BMI and workload were statistically significant. To account for these differences, which are also clinically meaningful, our logistic regression models were adjusted for age, BMI and workload. However, patient sex was not included in the final regression models because the differences between the symptom duration groups regarding this variable were not statistically significant. The analysis confirmed that the duration of radicular symptoms was associated with a greater occurrence of symptomatic radiographic OA. This indicates that while aging and body mass remain foundational drivers of joint pathology, the observed symptomatic radiographic hip and knee OA may also contribute to the chronicity of the spinal condition. It is estimated that approximately 20-40% of patients with LDH require surgical intervention despite an initial trial of conservative management (Leiden – The Hague Spine Intervention Prognostic Study Group (SIPS) et al., 2011). Numerous high-quality studies support early surgical treatment, demonstrating superior postoperative outcomes compared to those of patients undergoing prolonged non-operative treatment (Atlas et al., 2005; Weinstein et al., 2008). Nevertheless, in routine clinical practice, surgical interventions are often delayed. Prolonged conservative treatment may have a negative impact on postoperative recovery and overall outcomes, particularly in patients with persistent or progressive symptoms (Peul et al., 2007). Further evidence of the systemic consequences of degenerative spinal pathology comes from a long-term Finnish population-based cohort study with a 26-year follow-up, which demonstrated that severe lumbar disc degeneration significantly increased the hazard ratio for total hip arthroplasty (Salo et al., 2025). Similarly, spinal degeneration frequently coexists with knee OA, and these patients have been shown to experience worse outcomes following knee arthroplasty (Betzler et al., 2022). One possible explanation involves alterations in the lumbo-pelvic sagittal alignment, particularly pelvic retroversion, which may contribute to the progression of knee OA in patients with degenerative spinal disease (Amarasinghe et al., 2023; Yasuda et al., 2020). Unlike general population studies where OA prevalence is primarily linked to age and systemic factors, our findings demonstrate a significant side-specific association between radiculopathy and joint degeneration. This suggests that unilateral neurological impairment and subsequent asymmetric gait patterns may play a role in focal joint stress, a factor often overlooked in generalized OA cohorts. It is well established that unilateral lower extremity pain is frequently characterized by a variety of compensatory movement patterns including altered joint kinematics, restricted range of motion, and reduced gait velocity, all aimed at minimizing discomfort in the affected area. These changes commonly co-occur with core instability, muscle weakness, radicular pain, and sensory deficits on the symptomatic side (Wang et al., 2022). In patients with LDH, lower extremity symptoms and subsequent compensatory mechanisms are hypothesized to potentially contribute to gait abnormalities, such as reduced ankle plantarflexion, limited hip extension, and decreased sagittal plane hip motion during both the stance and swing phases, as described in existing literature (Haddas et al., 2025). These biomechanical alterations could theoretically be further influenced by core instability, muscle weakness, radicular pain, and sensory deficits on the symptomatic side (Bonab et al., 2023). Pathological muscle activation patterns have also been described in previous literature concerning patients with LDH. For example, elevated surface electromyographic amplitudes in sagittal plane stabilizers, such as the rectus femoris, reflect impaired motor coordination (Ciesielska et al., 2015). These neuromuscular impairments may contribute to characteristic spatiotemporal gait changes, including reduced gait speed, cadence, and increased gait variability in patients with LDH (Natarajan et al., 2022). Although we did not evaluate these factors directly, such neuromuscular impairments represent speculative pathways that might contribute to characteristic spatiotemporal gait changes. In our study, long-term biomechanical stress is hypothesized as a potentially plausible, indirect driver of hip and knee joint degeneration. The finding that symptomatic radiographic changes were significantly more prevalent on the side of the radiculopathy demonstrates a strong spatial association, which is consistent with our hypothesis regarding asymmetric gait alterations; however, our cross-sectional design precludes us from determining whether unilateral radiculopathy precedes and accelerates focal joint wear, or if pre-existing asymmetric joint degeneration conversely exacerbates the chronicity of spinal symptoms. This underscores the systemic impact of chronic spinal pathology on the entire kinetic chain. The coexistence of hip and spinal pathology, mediated through alterations in sagittal pelvic alignment, is recognized as MacNab's hip-spine syndrome (Weinberg et al., 2017). Furthermore, the cross-sectional nature of our study requires a direct consideration of the possibility of reverse causation. While we hypothesized that chronic radiculopathy contributes to joint wear, it is highly plausible that pre-existing hip or knee OA independently drives or exacerbates lumbar spinal pathology. Advanced joint degeneration characteristically restricts lower limb mobility and alters sagittal pelvic tilt profiles during daily activities. This baseline joint dysfunction shifts the biomechanical burden upward along the kinetic chain, subjecting the lower lumbar motion segments to continuous, asymmetric mechanical stress. Such persistent spinopelvic instability may not only accelerate intervertebral disc degeneration but could also render subsequent LDH symptoms less responsive to conservative therapies. Consequently, patients with pre-existing lower limb OA may experience a higher likelihood of delayed clinical recovery and prolonged symptom duration, ultimately leading to a higher rate of surgical indications and their subsequent accumulation within our longest-duration cohort (>24 months). While certain cadaveric studies have demonstrated that lumbar spinal degeneration often predates hip joint changes, the cross-sectional nature of our clinical data allows us to confirm only a strong spatial and statistical association, rather than a chronological causal sequence (Bajwa et al., 2013). As a result, the primary source of disability in hip-spine syndrome often remains unclear, since pathologic changes in both regions may potentiate one another (Devin et al., 2012). The prevalence of hip or knee OA in patients with lumbar spinal degeneration has been reported to range from 0 to 54%, depending on the joint assessed and the definition of OA used (Young et al., 2022). Lumbar spinal degeneration is associated with loss of lumbar lordosis, commonly referred to as flatback, a progressive reduction in spinopelvic mobility, and a loss of compensatory posterior pelvic tilt during postural transitions (Buckland et al., 2020). Furthermore, variations in increased pelvic incidence (PI) and sacral slope (SS) with variable pelvic tilt (PT)/lumbar lordosis (LL) profiles affect how forces are transmitted through the hip and thus contribute to hip OA (Yoshimoto et al., 2005). Degenerative lumbar spine changes are often associated with LBP, which is recognized as a negative prognostic factor in patients with hip OA (Stupar et al., 2010). Supporting this suggestion, we found an association between longer duration of LBP and a higher prevalence of symptomatic radiographic OA in the knee and hip joints. Interestingly, hip arthroplasty has been shown to improve LBP in certain patients, further supporting the biomechanical interdependence between these regions (Ben-et al., 2007). Conversely, pre-existing lumbar pathology, including LDH, may negatively influence hip replacement outcomes, likely due to disrupted spinopelvic dynamics that alter hip joint behavior (Blizzard et al., 2016). Despite the promising nature of our findings, several limitations must be acknowledged. Primary among these and a critical factor influencing the interpretation of our findings is that hip and knee radiographs were not performed systematically across the entire cohort, but only in patients exhibiting clinical joint symptoms. Consequently, our findings represent strictly the prevalence of symptomatic radiographic OA, which inherently underestimates the true radiological burden by completely omitting asymptomatic joint degeneration. While the use of blinded orthopedic clinical assessments combined with radiographic grading provided a robust surrogate for identifying clinically relevant OA, this selection criteria introduces an unavoidable ascertainment bias that must be factored into the interpretation of the data, as the true background prevalence of radiographic joint changes remains unknown. Furthermore, while all patients reporting subjective joint symptoms were systematically referred for radiographic examination, our database only captured individuals with confirmed structural changes (ranging from mild to moderate-to-severe radiographic OA). Because the exact numbers of symptomatic patients who presented with completely negative radiographic findings were not recorded, we cannot provide the total denominator of all screened individuals. Additionally, subjective symptoms in this surgical cohort could have potentially originated from non-OA sources, such as referred spinal pain from advanced LDH. Next, the categorization of hip and knee symptomatic radiographic OA was necessarily simplified due to the limited number of patients in each subgroup, which restricted more detailed statistical analysis. Additionally, formal interobserver or intraobserver reliability statistics for the radiographic grading were not calculated. However, as noted above, our reliance on clinical examinations, the blind nature of the orthopedic assessment, and its systematic correlation with standard institutional radiological reports offered a highly practical, reproducible, and reliable approach to evaluating symptomatic joint pathology within this surgical cohort. We acknowledge that standardized patient-reported outcome measures (PROMs) specific to joint function were not collected. However, as noted above, our reliance on clinical examinations and radiographic grading offered a practical and reliable approach to evaluating symptomatic joint pathology within this surgical cohort. Moreover, it is important to note that our cohort represents a continuous flow of patients indicated for surgery at a tertiary neurosurgical center. The group with symptom duration ≤1 month primarily consisted of patients with acute, severe clinical presentations requiring urgent intervention, whereas the >24-month group represented patients with chronic, relapsing symptoms who underwent prolonged conservative management before opting for surgery. Crucially, because this study population consists entirely of surgically treated individuals from a specialized tertiary referral institution, these findings may limit generalizability to the broader, non-surgical LDH population. Patients managed in primary care or outpatient rehabilitative settings typically exhibit less severe clinical presentations and shorter symptom durations; therefore, the strong spatial and statistical associations identified in our surgical cohort should be interpreted within the context of advanced spinal pathology. Crucially, the assessment of symptom duration in this study is inherently vulnerable to recall bias, which represents a notable limitation. Patients with a long-standing history of chronic or relapsing-remitting radiculopathy often find it challenging to precisely differentiate between the onset of the current acute exacerbation and historical, self-limiting episodes of LBP or radicular pain. While this retrospective reporting of symptom onset may introduce some classification variability between our intermediate and chronic cohorts, all patients strictly met the same radiological criterion of significant LDH correlating with their objective clinical presentations at the time of surgical indication. Furthermore, we acknowledge that certain relevant confounding factors, including smoking status, presence of diabetes mellitus, baseline physical activity, genetics or direct joint overload were not recorded in our patient database. Since metabolic conditions like diabetes and lifestyle factors like smoking can independently affect microvascular joint health and cartilage degradation, the absence of these variables prevents us from ruling out some potential confounding. These limitations highlight the need for a prospective study incorporating systematic clinical and radiological evaluation of hip and knee OA severity in patients undergoing spinal surgery for degenerative spinal disease. Conclusion In conclusion, a longer duration of lumbar disc herniation-related symptoms including radicular leg pain, LBP, and motor deficit was associated with a higher prevalence of symptomatic radiographic hip and knee OA within this selected surgical cohort. However, because this study population consists entirely of individuals indicated for microdiscectomy at a tertiary referral center, these findings are restricted to advanced spinal pathology and cannot be directly generalized to the broader, non-surgical population. Crucially, given the cross-sectional nature of our data, no definitive chronological sequence or causal relationship can be inferred from these observed associations. Whether chronic spinal symptoms accelerate lower limb joint wear or, conversely, pre-existing joint degeneration contributes to prolonged and less responsive spinal symptoms remains an untested hypothesis that represents a critical avenue for future longitudinal investigation rather than an immediate clinical implication. Author contributions Solar and Ondirko contributed to conceptualization; Solar, Ondirko, and Valousek contributed to the methodology, conducted operations, and collected data; Wesela performed data analysis; Solar, Ondirko, Wesela, Valousek, and Jancalek contributed to writing and editing the manuscript; Jancalek supervised the study. All authors reviewed the manuscript and agreed to its final version. Disclosures The authors have no personal, financial, or institutional interest in any drugs, materials, or devices described in this article. Statement During the preparation of this work the authors used the Gemini 3 Flash AI model in order to refine the English grammar, enhance the stylistic fluency, and improve the overall clarity of the manuscript. After using this tool, the authors reviewed and edited the content as needed and takes full responsibility for the content of the published article. Funding The study was supported by the Ministry of Health of the Czech Republic in cooperation with the Czech Health Research Council under project No. NW25J-10-00060. Declaration of competing interest The authors have no personal, financial, or institutional interest in any drugs, materials, or devices described in this article. References