Introduction
The etiology of cervical disc herniation (CDH) is primarily multifactorial and associated with chronic degenerative changes.1 Given the chronic nature of the condition, patients often undergo prolonged conservative management, yet long-term prognostic data beyond one year for various therapeutic interventions remain scarce.
While our previous study2 demonstrated significant pain relief and functional improvement with integrative Korean medicine (IKM) at the one-year follow-up, evaluating longer-term durability is critical to monitor for late recurrence in CDH. As part of an ongoing 5-year prospective study, this research letter presents the interim three-year clinical outcomes to provide valuable longer-term evidence on the enduring effectiveness and safety of IKM.
Methods
This study is a three-year follow-up extension of a prospective, multicenter, observational trial. Between July 2019 and December 2021, 120 participants (aged 19–65) with MRI-confirmed cervical disc herniation and radiculopathy [baseline numeric rating scale (NRS) ≥6] were recruited across four Jaseng Hospital centers in South Korea.
The IKM interventions—consisting of herbal medicine, acupuncture, pharmacopuncture, and Chuna therapy—were administered according to clinical protocols identical to those detailed in our previous one-year report.2 Participants received treatment for approximately two months (average treatment visits: 15.35±4.45, approximately twice weekly), with long-term follow-up conducted via telephone or in-person visits, during which any additional healthcare utilization for cervical disc herniation was also recorded.
The primary outcome was the NRS for neck and arm pain. Secondary outcomes included the visual analogue scale (VAS), neck disability index (NDI), EuroQol-5-dimension (EQ-5D), patient global impression of change (PGIC), work productivity and activity impairment questionnaire (WPAI), and range of motion (ROM). Safety was assessed by monitoring adverse events (AEs). For statistical analyses, continuous variables were presented as mean and standard deviation, and categorical variables as frequency (%). Longitudinal changes from baseline were assessed using a linear mixed model (LMM) with time and baseline outcomes as fixed effects and a patient-specific random intercept, and an unstructured covariance matrix. The LMM handled missing data under the missing at random assumption, and sensitivity analysis was performed via multiple imputation for missing data. No formal correction for multiple comparisons was applied, as the Numeric Rating Scale (NRS) was pre-specified as the single primary outcome for confirmatory interpretation, while other parameters were treated as exploratory secondary outcomes. Analysis were conducted using SAS Version 9.1.3 statistical package (SAS Institute, Cary, NC, USA) or R software (Version 4.1.3; R Foundation for Statistical Computing, Vienna, Austria) with statistical significance set at p <0.05. All procedures were approved by the Institutional Review Board (JASENG 2019–04-009, JASENG 2019–04-010, JASENG 2019–04-011, and JASENG 2019–04-013) and registered at ClinicalTrials.gov (NCT03959098) and cris.nih.go.kr (KCT0004158).
Results
Of the 120 eligible participants, 115 completed the 2-month treatment protocol. Follow-up assessments were conducted for 114 participants at 1 year and 112 at 3 years. (Cumulative loss to follow-up: 3 withdrawal of consent, 3 loss of contact, and 2 other reasons; no adverse-event dropouts). The primary outcome, the NRS score, significantly decreased by 4.57 points (95% CI, 4.24 to 4.90) from baseline, reaching 2.68 (95% CI, 2.35 to 3.01) at the 3-year follow-up (P<0.001; Table 1). This reduction represents a sustained clinical improvement of >50% maintained since 8 weeks post-treatment (Figure 1).
Table 1.
Outcomes for the Study Period
| Value | Week 4 | Week 8 | Week 16 | Week 24 | 1 Yr | 3 Yr |
|---|---|---|---|---|---|---|
| NRS | 4.55 (4.22 to 4.87) |
3.29 (2.97 to 3.62) |
2.87 (2.54 to 3.19) |
2.91 (2.58 to 3.24) |
2.84 (2.51 to 3.17) |
2.68 (2.35 to 3.01) |
| Difference | 2.71 (2.38 to 3.03)*** |
3.96 (3.63 to 4.29)*** |
4.38 (4.06 to 4.71)*** |
4.34 (4.02 to 4.67)*** |
4.41 (4.08 to 4.75)*** |
4.57 (4.24 to 4.90)*** |
| VAS of Neck pain | 47.83 (44.77 to 50.89) |
32.25 (29.15 to 35.35) |
29.83 (26.19 to 33.47) |
29.24 (25.19 to 33.28) |
28.81 (24.29 to 33.32) |
23.66 (15.72 to 31.61) |
| Difference | 23.58 (20.52 to 26.64)*** |
39.16 (36.06 to 42.26)*** |
41.58 (37.94 to 45.22)*** |
42.17 (38.13 to 46.22)*** |
42.61 (38.09 to 47.12)*** |
47.75 (39.81 to 55.69)*** |
| VAS of Arm pain | 38.29 (34.88 to 41.70) |
24.66 (21.20 to 28.12) |
21.79 (17.76 to 25.82) |
21.71 (17.24 to 26.18) |
21.72 (16.75 to 26.69) |
18.14 (9.47 to 26.81) |
| Difference | 24.66 (21.25 to 28.06)*** |
38.28 (34.82 to 41.74)*** |
41.15 (37.12 to 45.18)*** |
41.23 (36.77 to 45.70)*** |
41.22 (36.25 to 46.19)*** |
44.80 (36.13 to 53.47)*** |
| EQ-VAS | 68.82 (66.15 to 71.48) |
71.80 (69.10 to 74.51) |
74.22 (71.56 to 76.88) |
75.04 (72.35 to 77.73) |
73.36 (70.64 to 76.08) |
73.81 (71.09 to 76.53) |
| Difference | −8.26 (−10.93 to −5.60)*** |
−11.25 (−13.95 to −8.54)*** |
−13.66 (−16.32 to −11.00)*** |
−14.49 (−17.18 to −11.80)*** |
−12.80 (−15.52 to −10.08)*** |
−13.25 (−15.97 to −10.53)*** |
| EQ-5D-5L | 0.79 (0.77 to 0.80) |
0.82 (0.80 to 0.84) |
0.82 (0.80 to 0.84) |
0.82 (0.80 to 0.83) |
0.82 (0.81 to 0.84) |
0.84 (0.82 to 0.85) |
| Difference | −0.09 (−0.11 to −0.07)*** |
−0.12 (−0.14 to −0.11)*** |
−0.12 (−0.14 to −0.11)*** |
−0.12 (−0.14 to −0.11)*** |
−0.13 (−0.14 to −0.11)*** |
−0.14 (−0.16 to −0.12)*** |
| NDI | 27.83 (26.02 to 29.65) |
20.44 (18.61 to 22.28) |
19.29 (17.48 to 21.10) |
19.46 (17.63 to 21.29) |
18.98 (17.13 to 20.82) |
17.11 (15.26 to 18.96) |
| Difference | 11.28 (9.46 to 13.09)*** |
18.67 (16.83 to 20.50)*** |
19.82 (18.01 to 21.63)*** |
19.65 (17.82 to 21.48)*** |
20.13 (18.29 to 21.98)*** |
22.00 (20.15 to 23.85)*** |
Notes: Differences represent the change from baseline, derived from the estimated marginal means (emmeans) of the LMM, and are displayed with 95% confidence intervals (CIs). ***: p<0.001.
Abbreviations: LMM, linear mixed model; NRS, numeric rating scale; VAS, visual-analogue scale; EQ-5D-5L, EuroQol 5-dimension 5-level; NDI, neck disability index.
Figure 1.
Linear mixed model graph for NRS change for study. Treatment-period: baseline-Wk8, Follow-up period: Wk8-3year. Values are estimated means with 95% confidence intervals. Patients (n) below the x-axis indicates the number of participants with an observed NRS score at each time point.
Abbreviations: Wk, Week; NRS, Numeric rating scale.
All secondary endpoints—including neck and arm pain VAS, disability (NDI), and quality of life (EQ-5D)—demonstrated significant longitudinal improvements consistent with the primary outcome (P<0.001 for all; Table 1). Cervical ROM, WPAI scores, and PGIC also showed significant early improvements that remained stable through year 3 (P<0.001). These clinical outcomes remained stable throughout the extended follow-up period, with no significant waning of clinical improvements observed between years 1 and 3. Concomitant healthcare utilization during the 3-year follow-up was minimal (eg, only 6 and 5 cases of injection therapy or conventional medication, respectively, with no surgical conversions), suggesting that long-term clinical improvements were sustained without significant reliance on rescue therapies. Sensitivity analysis using repeated-measures analysis of variance (ANOVA) with multiple imputation yielded results consistent with the primary analysis.
Safety monitoring revealed 118 AEs in 69 participants during the 1-year follow-up, 21 (17.8%) of which were potentially treatment-related and predominantly mild (eg, headache, diarrhea). While eight serious AEs (SAEs) occurred, they were largely unrelated to the intervention (eg, trauma). Two SAEs involving hospitalizations for exacerbated neck pain resolved following non-surgical interventions. Crucially, no additional treatment-related AEs or SAEs were reported during the extended 3-year follow-up, supporting a favorable long-term safety profile.
Discussion
This three-year follow-up demonstrates the long-term therapeutic durability and safety of IKM for patient with CDH. While our previous report confirmed one-year efficacy, current results indicate that significant improvements in pain, disability, and quality of life remained stable without waning effect through the third year. Notably, the mean reduction in NRS continued to substantially exceed both the minimal clinically important difference (MCID) of 2.2 and the minimal detectable change (MDC) of 4.1 established for the conservative management of patients with cervical radiculopathy.3 Crossing the MDC threshold suggests that the long-term pain relief is likely to reflect a meaningful variance distinct from simple measurement fluctuations. Furthermore, these outcomes appear favorable compared to other active non-surgical treatments; for instance, the long-term pain reduction observed here is within a highly comparable range to results reported for epidural steroid injections at one- and two-year marks, although direct comparisons warrant caution due to differences in study populations and designs.4 The sustained efficacy of IKM may be attributed to the synergistic, multi-target mechanisms of its integrated components. Specifically, acupuncture and Chuna therapy facilitate muscle relaxation and functional restoration while inducing neurotransmitter secretion and activating pain modulation pathways.5,6 Simultaneously, the herbal medicines and pharmacopuncture solutions used in this protocol exert potent anti-inflammatory, neuroprotective, and analgesic effects.7 Managing complex spinal disorders through this multi-modal approach addresses both localized inflammation and adjacent soft tissue dysfunction simultaneously, which likely contributes to preventing chronic relapse and maintaining the low rate of surgical conversion observed in our cohort.
The favorable safety profiles established during the first year was maintained throughout the three-year follow-up, with no additional treatment-related adverse events or delayed-onset complications reported. The lack of surgical conversion further suggests that IKM serves as a sustainable conservative alternative to invasive procedures. This study has some limitations. First, the single-arm observational design without a control group makes it difficult to distinguish IKM’s specific effects from the natural history of CDH, spontaneous recovery, placebo effects, or regression to the mean. Second, the potential for residual confounding and selection and recall biases cannot be entirely ruled out, which may limit the generalizability of our findings. Therefore, causality cannot be definitively established from these observational data alone. Nonetheless, the high three-year retention rate provides significant real-world evidence regarding the long-term sustainability and safety of IKM. In conclusion, IKM is associated with sustained pain relief, functional recovery, and long-term safety for CDH patients for up to three years, suggesting its role as a primary conservative option. Future well-designed randomized controlled trials with long-term follow-up are warranted to rigorously confirm these findings.
Data Sharing Statement
The deidentified participant datasets supporting the findings of this study are available from the corresponding author upon reasonable request.
Ethics Approval and Informed Consent
This study was approved by the Institutional Review Board of the Jaseng Hospital of Korean Medicine (JASENG 2019-04-009, JASENG 2019-04-010, JASENG 2019-04-011, and JASENG 2019-04-013). All participants provided written informed consent in accordance with the Declaration of Helsinki.
Disclosure
The authors report no conflicts of interest in this work.
References
- 1.Hammer C, Heller J, Kepler C. Epidemiology and pathophysiology of cervical disc herniation. Semin Spine Surg. 2016;28(2):64–4. doi: 10.1053/j.semss.2015.11.009 [DOI] [Google Scholar]
- 2.Lee SW, Ahn Y-J, Kim S. Effectiveness and Safety of Integrative Korean Medicine Treatment for Cervical Disc Herniation with Radiculopathy: one-year Follow-up Results of a Prospective Observational Study. Perspect Integr Med. 2026;5(1):17–27. doi: 10.56986/pim.2026.02.004 [DOI] [Google Scholar]
- 3.Young IA, Cleland JA, Michener LA, Brown C. Reliability, construct validity, and responsiveness of the neck disability index, patient-specific functional scale, and numeric pain rating scale in patients with cervical radiculopathy. Am J Phys Med Rehabil. 2010;89(10):831–839. doi: 10.1097/PHM.0b013e3181ec98e6 [DOI] [PubMed] [Google Scholar]
- 4.Goyal KK, Kristoff TJ, Sinopoli JT, et al. Benefit of pain and function after cervical transforaminal epidural steroid injections, an observational study. Intervent Pain Med. 2023;2(4):100299. doi: 10.1016/j.inpm.2023.100299 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chen T, Zhang WW, Chu Y-X, Wang Y-Q. Acupuncture for pain management: molecular mechanisms of action. Am J Chin Med. 2020;48(4):793–811. doi: 10.1142/S0192415X20500408 [DOI] [PubMed] [Google Scholar]
- 6.Lee J, Cho J-H, Kim K-W, et al. Chuna manual therapy vs usual care for patients with nonspecific chronic neck pain: a randomized clinical trial. JAMA Network Open. 2021;4(7):e2113757. doi: 10.1001/jamanetworkopen.2021.13757 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Park Y-G, Ha C-W, Han C-D, et al. A prospective, randomized, double-blind, multicenter comparative study on the safety and efficacy of Celecoxib and GCSB-5, dried extracts of six herbs, for the treatment of osteoarthritis of knee joint. J Ethnopharmacol. 2013;149(3):816–824. doi: 10.1016/j.jep.2013.08.008 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The deidentified participant datasets supporting the findings of this study are available from the corresponding author upon reasonable request.