The Short-Term Effect of Integrated Complementary and Alternative Medicine Treatment in Inpatients Diagnosed with Lumbar Intervertebral Disc Herniation: A Prospective Observational Study

Abstract

Objectives:

This study aimed to investigate the short-term effect of hospital-based intensive nonsurgical treatment in lumbar intervertebral disc herniation (IDH) inpatients admitted to an integrated hospital that offers both complementary and alternative medicine (CAM) and conventional medicine treatment.

Design:

A prospective observational study.

Settings:

A private Korean medicine hospital inpatient setting in Korea.

Patients:

A total of 524 inpatients diagnosed with lumbar IDH admitted from June 1, 2012, to May 31, 2013.

Interventions:

The participants received treatment according to a CAM treatment protocol (herbal medicine, acupuncture, bee venom pharmacopuncture, and Chuna manipulation) and conventional medicine treatment as needed.

Outcome measures:

Numeric rating scale (NRS) of low back pain (LBP) and leg pain, Oswestry Disability Index (ODI), and patient global impression of change. The study also assessed whether improvement was obtained over minimal clinically important difference (MCID) in LBP or leg pain.

Results:

The average hospital stay was 24.4 ± 13.2 days. The majority of patients received CAM treatment and a few selected conventional medicine, such as pain killers (22.7%; 4.2 ± 3.0 administrations) or nerve blocks (14.1%; 1.4 ± 0.7 sessions). At discharge, the average reduction in NRS was 3.18 ± 2.29 (95% confidence interval [CI], 2.99–3.38) for LBP and 2.61 ± 2.60 (95% CI, 2.38–2.83) for leg pain the average reduction in ODI was 19.45 ± 19.53 (95% CI, 17.77–21.12). Two-hundred and seventy patients (51.5%) showed improvement over MCID in both NRS and ODI, 150 (28.6%) in either NRS or ODI, and 104 (19.8%) in neither.

Conclusions:

Integrated CAM treatment during hospitalization was effective for patients with lumbar IDH who had severe LBP and disability. However, these results must be investigated further to assess whether the effects surpass those seen with placebo and are cost-effective.

Introduction

Lumbar intervertebral disc herniation is the most common disorder causing low back pain (LBP) and radiating leg pain in working adults, with an annual incidence of 5/1000 in the general population.^1,2 The economic impact of LBP and radiating leg pain has been well documented in the Dutch study, and the amount of work loss and disability is larger than that seen with any other medical condition.³ LBP caused by such diseases as spondylosis and intervertebral disc disorders was the second highest factor for admission in persons aged 18–44 and the fourth highest in patients aged 45–64 years in the United States in 2005.⁴ The Healthcare Cost and Utilization Project Nationwide Inpatient Sample observed in 2011 that in populations aged 18–65 years, every 40 in 10,000 patients were admitted for intervertebral disc disorders, with an average hospital stay of 2.5–3.2 days.⁵

Korean medicine and conventional medicine coexist as individual treatment options within the South Korean medical system. According to data from the National Health Insurance Statistical Yearbook on conventional medicine in 2011, 189,804 patients received inpatient treatment for lumbar disc herniation during the preceding year and the average length of stay was 11.8 days.⁶ For Korean medicine, it was reported that 12,840 patients received inpatient care due to lumbar disc herniation during the same period for an average period of approximately 20 days, and this ranked third in reasons for hospitalization in Korean medicine hospitals, with unspecified lumbar pain and lumbar sprain coming first and second.⁶

In disorders related to lumbar disc herniation, the purpose of admission may be varied. Short stays will more likely be for surgery, whereas relatively long stays may be for surgical sequelae or for administration of nonsurgical treatment.⁵ Conservative treatment methods for lumbar disc herniation are diverse. However, there are conflicting opinions on whether the therapeutic effect of conservative treatment is better than the natural progress of disc herniation and whether conservative treatment is a safe and economic treatment choice.^7,8

Of the nonsurgical approaches, nonsurgical treatment using complementary and alternative medicine (CAM), such as acupuncture, pharmacopuncture, herbal medicine, and spinal manipulation, is popular in Korea. Jaseng Hospital of Korean Medicine treats more than 900,000 cases of spinal disorders a year as a spine specializing hospital of Korean medicine, designated as such by the Korean Ministry of Health and Welfare. Jaseng manages patients using an integrated system of Korean and conventional medicine wherein conventional medicine doctors use such diagnostic imaging tools as radiography and magnetic resonance imaging (MRI) and take on a supportive role to Korean medicine doctors (KMDs), providing emergency care as needed, while KMDs supervise the overall treatment regimen.^9,10 In this unique setting, patients unable to receive outpatient treatment because of reduced mobility due to LBP and radiating pain caused by lumbar disc herniation frequently opt for admission, with the aim of receiving intensive nonsurgical integrative treatment.

Various treatments are available to patients admitted for lumbar disc herniation. However, the number of studies on the therapeutic effect of inpatients treated with CAM for disc herniation is limited. Furthermore, the situation in Korea, where the medical system allows for an integrative approach focusing on CAM treatment in inpatient care, may seem strange. Thus, the aim of the current study is to propose a pragmatic research design on how integrative treatment is applied to patients diagnosed with lumbar disc herniation in real-life settings and to report the short term effects of an integrative treatment for inpatients focused on CAM.

Materials and Methods

Study design

This prospective observational study included inpatients presenting with LBP and radiating leg pain who were diagnosed with lumbar intervertebral disc herniation and admitted from June 1, 2012, to May 31, 2013, to Jaseng Hospital of Korean Medicine, which provides both conventional and Korean medical services. The last patient to be discharged during this period was released on June 23, 2013, and the data up to this point were retrieved and analyzed. The study was approved by the institutional review board of Jaseng Hospital of Korean Medicine, and written informed consent for participation in the study was obtained from participants at admission.

Participants

The inclusion criteria were hospitalization due to LBP and/or radiating leg pain, positive MRI findings of lumbar disc herniation, and diagnosis by KMDs that the causative factor for LBP and/or radiating pain was the herniated disc.

The exclusion criteria were main symptom something other than LBP or radiating leg pain, cause of LBP not related to spine or soft tissue (e.g., spinal tumors, pregnancy, urolithiasis, rheumatoid arthritis), and refusal to provide the information needed for clinical research.

KMDs evaluated whether the patients' present symptoms were caused by the herniated disc by conducting neurologic disability examinations (sensory, muscle strength, deep tendon reflex, and straight-leg raising tests) and checking the radiology specialists' MRI readings. KMDs visited and interviewed inpatients meeting the inclusion criteria in the inpatient ward on the day of admission and followed them up at discharge. In cases where one patient was admitted several times during the study period, only the first inpatient record was included for analysis.

Interventions

Although the research was planned so that the interventions applied in the study protocol consisted of the most frequently used treatment package contents administered to patients with disc herniation at this hospital, the study allowed for use of modalities other than those specified in the treatment protocol to all practitioners and patients participating in the study and used the electronic medical record system to record the types and frequency of divergent treatment use in a practical manner. If the patients needed such conventional treatments as analgesics, nerve block injections (lidocaine, steroids, adhesion barriers), or other treatments besides the CAM treatment stated in the treatment protocol, as assessed by KMDs, the patients were referred to conventional Western doctors for further treatment.

The CAM treatment protocol consisted of herbal medicine, acupuncture, pharmacopuncture, bee venom pharmacopuncture, and Chuna manipulation.

Herbal medicine

Herbal medicine was taken 3 times daily in dried powder (2 g) and water-base decoction form (120 mL) (Ostericum koreanum, Eucommia ulmoides, Acanthopanax sessiliflorus, Achyranthes japonica, Psoralea corylifolia, Saposhnikovia divaricata, Cibotium barometz, Lycium chinense, Boschniakia rossica, Cuscuta chinensis, Glycine max, and Atractylodes japonica). These herbs were selected from herbs prescribed in Korean medicine and Traditional Chinese Medicine literature¹¹ for LBP treatment. The prescription was further developed through clinical experience amassed at Jaseng Hospital of Korean Medicine.⁹ The anti-inflammation¹² and nerve¹³ and joint protection effects¹⁴ of the main herbal ingredients of the medicine (A. sessiliflorus Seem, A. japonica Nakai, Saposhnikovia divaricata Schischk, C. barometz J. Smith, G. max Merrill, and E. ulmoides Oliver) were reported as an herbal mixture named GCSB-5. The herbal medication received approval from the Korean Food and Drug Administration in January 2011 on the basis of research proving its therapeutic effects with less adverse digestive reactions in long-term administration. After the standard development process,^15

–18 it was categorized as an antipyretic, analgesic, anti-inflammatory prescription-only drug by the Korean Ministry of Health and Welfare and is currently available under the brand name Shinbaro (Greencross Pharmaceuticals).¹⁹

Acupuncture

Acupuncture treatment was administered 1–2 times daily in the lumbar sacral area using mainly Ah-shi points and acupuncture points related to LBP.

Pharmacopuncture

Select ingredients similar to those included in the oral herbal medicine (O. koreanum, E. ulmoides, A. sessiliflorus, A. japonica, P. corylifolia, S. divaricata, C. barometz, L. chinense, B. rossica, C. chinensis, G. max, and A. japonica) were freeze-dried into powder form after decoction, then diluted in normal saline and adjusted for acidity and pH to be used in injections.²⁰ The pharmacopuncture injections were injected once daily to the amount of 1 mL at Huato Jiaji (the erector spinae muscle area) and Ah-shi points in the lumbar area (CPL, 1 mL, 26-G × 1.5 syringe, Shinchang Medical Co., Korea).

Bee venom pharmacopuncture

Bee venom pharmacopuncture²¹ was applied after confirmation of a negative reaction to the hypersensitivity skin test. Diluted bee venom (saline:bee venom ratio, 10,000:1) was injected at 4–5 acupoints around the lumbar spine at the physician's discretion. Each acupuncture point was injected with approximately 0.2 mL to a total of 0.5–1 mL using disposable injection needles (CPL, 1 mL, 26-G × 1.5 syringe, Shinchang Medical Co., Korea).

Chuna manipulation

Chuna was administered 3–5 times a week. This is a Korean version of spinal manipulation that incorporates conventional spinal manipulation techniques^22,23 for mobilization involving high-velocity, low-amplitude thrusts to joints slightly beyond the passive range of motion and gentle force to joints within the passive range of movement.

Outcome measures

All patient examinations were conducted by previously trained KMDs. Patient characteristics (e.g., sex, age, occupation, smoking and drinking habits, and medical history), current pain levels, and neurologic examinations were assessed and conducted on the first day of admission. Follow-up examinations were administered at discharge.

Numeric rating scale

On a numeric rating scale (NRS), the patients were asked to quantify their current LBP and radiating leg pain, setting no pain as 0 and the most severe pain that they could possibly imagine as 10.^24,25

Oswestry disability index

The Oswestry Disability Index (ODI) is a survey assessing the patients' degree of functional disability in daily life, consisting of 10 questions with 6 choices each, corresponding to 0–5 points. The scores for each item are added, divided by 50, then multiplied by 100 to calculate the degree of disability. The Korean version of the ODI, the reliability and validity of which were verified through the research of Jeon et al., was used in this study.²⁶

Patient Global Impression of Change

The Patient Global Impression of Change instrument (PGIC) was administered at follow-up to evaluate patient treatment satisfaction and subjectively assess improvement. It is a 5-point scale: much improved, minimally improved, no change, minimally worse, and much worse.²⁴

Physical and Neurologic Examinations

The range of flexion and extension was assessed to objectively measure lumbar mobility. Straight-leg raising (SLR), sensory function, muscle strength, and deep tendon reflex tests were conducted for neurologic assessment. Range of motion measurement is reliable (r = 0.94) and valid (r = 0.97)²⁷ but not very responsive (effect size, 0.1–0.6).²⁸ SLR measurement is reliable (intraclass correlation coefficient = 0.95),²⁹ its sensitivity is 80% (72%–97%) and its specificity 40% (11%–66%),³⁰ but it is also not highly responsive (effect size = 0.2).

Safety assessments

Adverse events were carefully monitored throughout the hospital stay. Skin and general reactions, aggravation or changes in pain before and after acupuncture, pharmacopuncture, and Chuna manipulation treatment were observed. Especially because bee venom hypersensitivity reactions are well documented, whether antihistamine drugs were used for hypersensitivity reaction was rigorously checked.²¹ Regular blood tests (blood cell counts, liver and renal function tests, inflammatory activity indexes) were administered, and if KMDs and medical doctors (MDs) concurred that an adverse reaction was possible, the patients were rechecked during hospitalization. To assess possible drug-induced liver injury due to herbal and/or conventional medicine intake, liver and renal function tests were conducted. Liver injury was defined as (A) an increase of more than two times the upper limit of normal in alanine aminotransferase or direct bilirubin or (B) a combined increase in aspartate aminotransferase, alkaline phosphatase, and total bilirubin, provided one of them was above two times the upper limit of normal.³¹ All adverse events were recorded and reported.

Statistical analysis

Descriptive analyses were performed for all data using SPSS software for Windows, version 18.0 (IBM, Armonk, NY). The sample size was chosen on the basis of the number of patients admitted for lumbar disc herniation. NRS and ODI, the main outcome measurements, were statistically analyzed at follow-up to determine whether the improvement was larger than the minimal clinically important difference (MCID). According to Ostelo and de Vet,³² a clinically significant difference in LBP is a reduction of 3.5 or more in NRS in acute cases, 2.5 or more in chronic cases, and a reduction of 10 or greater in ODI regardless of chronicity. Therefore, the authors set a statistically significant reduction in LBP or radiculopathy as 3.5 or greater in NRS within 4 weeks of onset and 2.5 or greater in those over 4 weeks. In ODI scores, a reduction of 10 or greater was construed and analyzed as statistically significant regardless of onset. For continuous variables, such as NRS, ODI, SLR, lumbar flexion, and extension, the mean difference between follow-up and baseline was analyzed for significance with a 95% confidence interval.

To determine predictive factors for improvement over MCID in pain or functional recovery after inpatient treatment, the variables measured at baseline were assessed with univariate analysis. After univariate analysis, a multivariate logistic regression analysis was performed on variables with p < 0.10, adjusted for sex and age, using a backward approach to remove the nonsignificant variables with the highest p-value in a stepwise manner until all variables had a p-value <0.05.

Results

During the study period, 1394 patients were admitted; of these, 1178 had a chief complaint of LBP or leg pain. One thousand and one patients (85%) had lumbar MRI performed at Jaseng or other clinics or hospitals after presentation of the current pain episode. Of these patients, 739 patients' current pain was considered directly related to the lumbar intervertebral disc herniation (disc protrusion, extrusion, or migration) diagnosed through a combination of readings by radiology specialists and neurologic and physical examinations and interviews conducted by KMDs. The baseline and follow-up results of 524 patients were obtained (Fig. 1). The characteristics and results of the 524 participants are listed in Table 1.

FIG. 1.

Flow diagram of the study. KMD, Korean medicine doctor; MRI, magnetic resonance imaging.

Table 1.

Patient Demographic Characteristics and Clinical Features at Admission (n = 524)

Variable	Values
Age (yr)	42.1 ± 13.2
Men	49.2 (258)
Smoking status
No	63.5 (333)
Yes	19.5 (102)
Quit	17.0 (89)
Drinking
Yes	84.0 (440)
No	16.0 (84)
Body mass index (kg/m²)	23.8 ± 7.2
Previous back pain
None	65.6 (344)
Disc herniation	29.6 (155)
Other (e.g., fracture)	4.8 (25)
Comorbid illnesses,^a yes	14.7 (77)
Radiating pain
None	19.7 (103)
Unilateral	62.0 (325)
Bilateral	18.3 (96)
Radiating pain below knee	62.6 (328)
Muscle weakness^b	13.1 (68)
Sensory loss^c	12.0 (57)
Straight-leg raising test result <60°	33.4 (175)
Range of lumbar flexion ≤60°	56.7 (297)
Previous back surgery	9.4 (49)
Previous treatment (for current episode)
No. of treatments	2.4 ± 4.2
Nerve blocks	29.3 (149)
Pain killers	27.3 (143)
CAM	18.7 (98)
Duration of current episode (d)	289.9 ± 677.1
<1 mo	39.1 (205)
1–6 mo	30.9 (162)
>6 mo	30.0 (157)
Low back pain, NRS (0–10)	6.0 ± 2.4
Radiating leg pain, NRS (0–10)	5.2 ± 3.2
Oswestry disability index (0–100)	48.6 ± 20.7

Values are expressed as number (percentage) of patients or as mean ± standard deviation.

Any self-reported gastritis, tuberculosis poliomyelitis, cardiovascular diseases, or uterine myoma or hepatitis B carrier.

The data from 6 patients are missing.

The data from 57 patients are missing.

CAM, complementary and alternative medicine; NRS, numeric rating scale.

The average duration of hospital stay was 24.4 ± 13.2 days. Most of the patients were treated primarily with CAM, with focus on Korean medicine. For herbal medicine administration, 81.3% of patients were prescribed with the decoction form, 97.7% with the pill form suggested in the treatment protocol, and the rest with different types of herbal medicine at the KMD's discretion. About 29% of patients (n = 151) received conventional medicine with CAM. Of these patients, 22.7%p used analgesics (pain killers) an average 4.2 ± 3.0 times, and 14.1%p received an average 1.4 ± 0.7 nerve blocks (Table 2). The most frequently prescribed analgesics were nonsteroidal anti-inflammatory drugs, antidepressants, and muscle relaxants. Baseline pain and functional disability were more severe in patients who were administered conventional medical treatment in conjunction with CAM (Appendix Tables 1 and 2).

Table 2.

Treatment Methods Administered During Hospital Stay

Variable	Patients (%)	Mean ± standard deviation
Length of stay (d)		24.38 ± 13.21
CAM
Herbal medicine^a	99.4	151.7 ± 101.6
Protocol decoction	81.3
Protocol pills	97.7
Acupuncture	100.0	36.96 ± 21.35
Pharmacopuncture	98.7	23.16 ± 13.89
Bee venom pharmacopuncture	50.1	19.58 ± 13.45
Chuna (Korean manipulation)	83.8	13.50 ± 8.25
CAM + conventional treatment	28.8
Analgesics (pain killers)	22.7	4.22 ± 3.04
Nerve blocks^b	14.1	1.45 ± 0.69

Herbal medicine protocol: The same herbal medicine prescription was recommended for all patients before commencement of the trial. However, the patients could be given different herbal prescriptions, taking into account individual health conditions as assessed by the Korean medicine doctors.

Medications such as anesthetics and steroids were injected in close proximity to the nerve root area affected by the herniated disc.

Comparison of the NRS for LBP and radiating leg pain and the ODI scores between follow-up assessment at discharge and baseline scores at admission revealed statistically significant reductions in pain. Also regarding the main discomfort of LBP or radiating leg pain, a large proportion of patients showed reduction over MCID in pain levels (reduction in NRS of 3.5 or greater in acute pain with an onset of less than 4 weeks, 2.5 or greater in chronic cases) and ODI scores (reduction in ODI of 10 or greater) (Table 3). Two hundred and seventy patients (51.5%) showed reduction over MCID in both NRS and ODI scores, 150 (28.6%) showed reduction over MCID in either NRS or ODI, and 104 (19.8%) did not show reduction over MCID in NRS or ODI. A total of 502 patients answered the survey for PGIC, and among these patients, 89.0% checked “satisfied” or higher for level of satisfaction (Fig. 2).

FIG. 2.

Inpatient satisfaction rate and difference in numeric rating scale (NRS) and Oswestry Disability Index (ODI) compared with minimal clinically importance difference (MCID). ^aThe data from 22 patients are missing. The satisfaction rate of treatment was measured with a five-point scale at discharge. ^bThe number of patients with NRS and/or ODI scores greater than MCID for the main symptom of low back or radiating pain (MCID: for NRS, 3.5 [acute pain] or 2.5 [nonacute]; for ODI, 10).

Table 3.

Difference in NRS and Oswestry Disability Index Scores Between Admission and Discharge

Variable	Baseline (admission)	Follow-up (discharge)	Difference	95% CI
Low back pain, NRS	6.00 ± 2.40	2.82 ± 1.88	3.18 ± 2.29	2.99–3.38
Radiating leg pain, NRS	5.15 ± 3.15	2.54 ± 2.09	2.61 ± 2.60	2.38–2.83
ODI	48.60 ± 20.74	29.15 ± 15.00	19.45 ± 19.53	17.77–21.12
NRS; MCID^a			1.18 ± 2.12	1.00–1.37
ODI; MCID^b			9.45 ± 19.53	7.77–11.12

The difference in the NRS and MCID of NRS are indicated to evaluate the reduction in the main symptom of low back pain or radiating pain. In acute pain (duration ≤4 weeks), the MCID of NRS is set at 3.5; in nonacute cases (duration >4 weeks), MCID is set at 2.5).

The MCID of ODI was set at 10. Therefore, 10 was subtracted from the initial difference between admission and discharge ODI scores.

ODI, Oswestry Disability Index; MCID, minimal clinically importance difference.

Predictor variables at admission were assessed in patients who did not show improvement over MCID at discharge; abnormal range of lumbar motion and whether or not the patient was naive to Korean medicine treatment were found to have influenced the results. Improvement over MCID was not observed in patients with lumbar motion within normal range and those with experience with Korean medicine treatment (Table 4).

Table 4.

Factors Associated with Lack of Success (MCID Not Met) at Discharge

	OR (95% CI)
Variable	Univariate	Multivariate	p-Value
Age (continuous)	1.02 (1.00–1.04)	1.01 (0.99–1.03)	0.19
Sex, male (ref., female)	0.71 (0.46–1.09)	0.78 (0.49–1.22)	0.27
Smoking status
Quit (ref., nonsmoking)	0.87 (0.48–1.57)
Yes (ref., nonsmoking)	0.69 (0.38–1.24)
Drinking (ref., no)	1.32 (0.76–2.31)
Body mass index (continuous)	0.97 (0.91–1.03)
Previous back pain (ref., no pain)	0.77 (0.49–1.23)
Comorbid illnesses (ref., no comorbidity)^a	1.17 (0.65–2.11)
Radiating pain (ref., no pain)	1.44 (0.80–2.57)
Radiating pain below knee (ref., no pain)	1.10 (0.71–1.73)
Straight-leg raising test ≥60° (ref. <60°)	0.35 (0.20–0.60)
Range of lumbar flexion ≥60° (ref. <60°)	0.28 (0.17–0.46)	0.30 (0.18–0.51)	0.00
Surgery (ref., no surgery)	0.65 (0.28–1.49)
Previous treatment (for current episode)
Nerve blocks	0.83 (0.51–1.35)
Pain killers	1.84 (1.11–3.05)
CAM^b	1.31 (0.82–2.09)	1.90 (1.12–3.20)	0.02
Duration of current episode (days)
<1 mo
1–6 mo (ref. <1 mo)	1.81 (1.05–3.11)
>6 mo (ref. <1 mo)	2.16 (1.26–3.7)

In the multivariate analysis, variables that are statistically significant with a p-value of 0.1 or lower in each univariate analysis were adjusted for age and sex; by using a backward approach, nonsignificant variables were removed in a stepwise manner until all variables had a p-value <0.05.

Any self-reported gastritis, tuberculosis poliomyelitis, cardiovascular diseases, or uterine myoma or hepatitis B carrier.

OR, odds ratio; CI, confidence interval; ref, reference group.

Liver function tests were conducted in all inpatients for safety assessment; 209 of these patients were identified as requiring follow-up assessments by KMDs and were reassessed at discharge. Five patients were identified as having liver injury on blood screening at admission and 3 patients at discharge. Among the 3 patients confirmed as having liver injury on the blood test at discharge, 2 had liver injury at admission. The remaining patient was not diagnosed with liver injury at admission but was identified as having liver injury at discharge. However, in the post hoc analysis, this patient was verified as having normal liver function with no specific symptoms and continued intake of the same herbal medicine for 2 months. Therefore, this patient was not considered as having drug-induced liver injury. Eight patients were prescribed antihistamine drugs for allergic reactions at an average 2.3 ± 1.8 sessions of bee venom treatment. No other adverse reaction requiring additional treatment or emergency action was reported.

Discussion

Improvement in pain and lumbar function was investigated in patients with LBP and radiating leg pain who had disc herniation and received inpatient treatment focusing on CAM. The NRS for LBP and radiating leg pain and ODI all showed statistically significant reductions. The minimum scores indicating statistical significance and clinical significance may differ, and the NRS and ODI score results showed improvement above MCID in both chronic and acute cases. In addition, after taking into account that 420 patients (80.2%) improved over MCID in NRS or ODI, it would seem that this clinically significant effect influenced patient satisfaction rate, too, as 447 patients (89.0%) replied that they were “satisfied.”

The main strengths of this research are that it shows a model of integrative conventional medicine and Korean medicine treatment focused on CAM in actual clinical practice. In setting the standardized treatment protocol, the authors selected evidence-based treatment that is frequently used and has been tested in prior studies; in execution they selected a feasible design in which they recorded the rate of divergence in treatment modalities, allowing for tailoring to patients' individual differences. The satisfaction survey conducted at discharge reflected not only the treatment effect but also the increased medical cost incurred by the various treatments received in this study. Considering that the participants were not recruited through public advertisements but were actual patients not provided with any financial compensation, the fact that most patients' satisfaction rate was high is worthy of attention. Therefore, the results of this research serve as evidence^33
–35 that multidisciplinary treatment is better than individual treatment in patients diagnosed with lumbar disc herniation. The results also imply that this integrative treatment model is applicable in real clinical settings with consideration of medical cost.

The biggest limitation of this study is the problem inherent to a prospective observational study with no control group. It is not possible to draw a definite conclusion that the integrative CAM treatment model suggested in this study is more effective than the treatment that would be received in a control group (e.g., surgery, conventional nonsurgical treatment) or natural progress. Another major limitation is the heterogeneity of the patient group and interventions. The study includes participants with acute and chronic pain whose natural course is generally different; because the study did not exclude such conventional treatments as nerve blocks or analgesics, no definite conclusions indicating that the results are the effects of CAM treatment can be drawn. It would be more appropriate to view them as the combined effects of the integrated system of composite Korean and conventional medicine treatment. Patients with higher levels of pain or functional disability due to lumbar disc herniation at admission were more likely to receive conventional and CAM treatment together initially, and later only CAM treatment.

The average cost of multidisciplinary treatment used in this study is estimated at $100–$150/day, and while the effect of multidisciplinary treatment may be superior to other single treatments, future studies should assess whether multidisciplinary treatment is cost-effective.

Another limitation of this study is that it mainly evaluated pain and functional disability at discharge and did not rigorously assess for progressive neurologic deficit. It did not record motor weakness or sensory loss at discharge and only checked SLR for neurologic testing. In SLR results, 33.4% of patients (n = 175) had an SLR angle of <60°at admission, and 17.0% (n = 89) had limited SLR at discharge. It is possible that the reason why 8 patients responded that they were dissatisfied at discharge in PGIC may have been progressive neurologic deficit or lack of pain relief.

The adherence rate is relatively low considering the short follow-up period, with 524 (86%) of the 609 patients with lumbar disc herniation followed up at discharge. This was probably related to patient attitudes toward clinical research. Because this study did not offer the participants any direct benefits, the patients, already irritable and uncomfortable from pain, may have been more negatively predisposed toward the study. Many patients declined to participate in the study or did not answer certain sections of the survey. In consideration of the patients' attitudes and low adherence, it is possible that participants dissatisfied with inpatient treatment declined to respond to the follow-up. Finally, a major limitation is the lack of long-term follow-up after discharge monitoring to determine how long the treatment effects were maintained.

Few studies have assessed the treatment results of inpatients with a main symptom of LBP or lumbar disc herniation. This is probably partly related to the patients' health care system. According to the 2011 U.S. statistics on patients with a principal diagnosis of International Classification of Diseases, Ninth Revision, diagnostic code of 722.10 in the Healthcare Cost and Utilization Project database, lumbar disc displacement is associated with an average hospital stay of 2.6 days; admissions are generally related to surgical treatment rather than conservative treatment, with 86% of the total 133,677 patients receiving laminectomy, excision of intervertebral disc, or spinal fusion.⁵

It would seem that the shortage of studies on hospitalization in this patient group is due to the relative insignificance of comparing short-term hospital stays before and after surgery. One study on 150 inpatients reported that almost 80% of patients who underwent an average 2.2 ± 1.1 sessions of epidural glucocorticoid injections and bed rest showed complete or partial pain relief (19% and 61%, respectively) during an average hospital stay of 9.9 ± 4.3 days. Of these patients, 13% received surgical treatment during their hospital stay.³⁶ A few other studies evaluated rehabilitation treatment of patients with subacute or chronic LBP. In Germany, Wagner et al. compared the effects of 3 weeks of multidisciplinary rehabilitative intervention in 136 inpatients who had chronic LBP with a waiting list control group of 34 patients; they found that the treatment group significantly improved in pain intensity, activity, participation, and health-related quality of life compared with the control group.³⁷ Härkäpää et al. reported the results of a controlled study on inpatient and outpatient treatment in patients with LBP. A total 476 patients were randomly assigned to the inpatient, outpatient or control group; results showed that the inpatient and outpatient treatment groups improved in pain and disability over the control group, and the inpatient group reported slightly more pain than the outpatient group after treatment and no difference in disability.³⁸ Several other studies mainly described the treatment effect of various types of rehabilitative exercise or supplements to exercise.^39
–41

The integrative treatment model of Korean medicine hospitals in Korea is considerably different from the models of CAM treatment usually found in Western contemporaries. Although CAM treatment has been more widely used in the West in recent years,⁴² it would seem the concept of “complementary” in “complementary and alternative medicine” is emphasized more and that medical doctors are the main administrators of acupuncture and naturopathy as an adjunct to conventional treatment after receiving training or advice from CAM specialists. In many cases, these CAM specialists do not have full legal authorization to independently treat patients. However, the Korean medical system is a dual medical system granting equal practice rights and legal authority to KMDs and MDs.

Moreover, because patients in Korea do not have a family doctor system, they are required to initially choose between conventional and Korean medicine treatment in selecting primary medical care.⁴³ The participants of this study, which was set in a Korean medicine hospital, initially chose a Korean medicine treatment approach primarily for their herniated disc disorder.

Therefore, the integrative treatment model implemented in this study is almost the opposite of its contemporary Western counterparts, where CAM is supplementary to conventional medicine; consequently, the bulk of treatment consists of such CAM treatment as acupuncture, herbal medicine, Chuna manipulation, and bee venom pharmacopuncture, with a few patients referred to conventional medical doctors for minor treatment sessions as needed. Although 22.7% of the participants of this study were prescribed analgesics, the average number of administrations was 4.22 ± 3.04, which roughly correlates to 2 days (based on a twice-daily dosage) during an average hospital stay of 24.4 days; only 14.1% received nerve block therapy. After taking into account the fact that these patients selected hospital stay for their level of discomfort, the demand for conventional medicine appears fairly small in this patient group.

It seems a reasonable conjecture that the main purpose of choosing hospital stay for conservative treatment of lumbar disc herniation–related disorders would be pain reduction. The results of this study, which show pain and functional recovery over MCID in many patients without resorting to surgical means during their hospital stay, hold important implications for patients considering Korean medicine hospitals for conservative treatment. Several randomized controlled trials among patients with lumbar disc herniation allowed for crossover treatment, with patients randomly allocated to surgical or nonsurgical groups, which resulted in high levels of crossover in treatment; one study found that almost 40% of participants received treatment of the nonallocated group.^44

–47 A cohort study that observed nonrandomized groups (with patients actively selecting the treatment group from surgical and nonsurgical treatment groups) of surgical and nonsurgical treatment patients also reported similar results, with a relatively large percentage of patients (14%) crossing over to the counterpart treatment within 3 months.⁴⁵

There is a high possibility that this digression from the initial plan in many studies is due to change in patients' pain intensity. Regardless of whether the patient actively selected or was randomly allocated to the surgical group, if the pain is sufficiently reduced during the waiting period, the patient may select conservative treatment instead. Likewise, patients in the nonsurgical treatment group may reconsider surgery if pain levels become unbearable. Therefore, in real treatment settings—as opposed to a rigid study setting—the main selection criterion for surgical versus nonsurgical treatment would probably be swift pain reduction, meaning that this would also be of major importance and possibly the main purpose for admission in inpatients receiving nonsurgical treatment.

One possible explanation for the clinically significant improvement in patients with restricted lumbar flexion over those with normal lumbar flexion range of motion could be that patients with restricted range of motion generally have higher NRS and ODI scores, leaving more room for improvement. The difference related to CAM experience could have occurred because higher expectation levels of patients naive to CAM treatment were reflected in the results, as it is probable that the patients were already dissatisfied with previous treatments; this could lead to higher hopes and expectancy for the new treatment modality and possibly greater improvement over MCID than seen in patients with CAM experience. The fact that such factors as the duration period, presence or absence of leg pain, comorbidity, or history of surgery did not influence improvement over MCID is noteworthy and suggests that the integrative CAM treatment applied in this study may be applicable to diverse patient groups.

Conclusions

Intensive integrative inpatient treatment focused on CAM resulted in clinically significant improvement in patients with lumbar and radicular leg pain who were diagnosed with lumbar disc herniation, and it may continue to yield satisfactory results in clinically applicable patients. However, further investigations are needed to assess whether these effects surpass placebo and natural progress and are cost-effective.

Footnotes

Acknowledgments

Areum Choi and Eun Hya Chi at Jaseng Medical Foundation supported the collection and management of data, and Hyejin Kim gave helpful advice on the statistical design of the trial. The trial is registered at ClinicalTrial.gov (NCT02257723; date of registration: October 3, 2014).

Author Disclosure Statement

No competing financial interests exist.

Appendix Table 2.

Pharmacologic Classification and Main Chemical Components of Conventional Medicine Prescribed in Patients Who Received Both CAM and Conventional Treatment

Pharmacologic classification	Main chemical component (generic name)	Patients (n)
Antacid	Almagate	135
NSAID	Naproxen	92
Antidepressant	Amitriptyline hydrochloride	85
Muscle relaxant	Eperisone hydrochloride	67
Opioid	Acetaminophen	63
NSAID	Diclofenac sodium	43
Other		17

NSAID, nonsteroidal anti-inflammatory drug.

References

Cherkin

, Deyo

, Loeser

, et al. An international comparison of back surgery rates. Spine, 1994; 19:1201–1206.

Frymoyer

. Back pain and sciatica. N Engl J Med, 1988; 318:291–300.

van Tulder

, Koes

, Bouter

. A cost-of-illness study of back pain in the Netherlands. Pain, 1995; 62:233–240.

Merrill

, Elixhauser

. Hospitalization in the United States, 2002. Rockville, MD: Agency for Healthcare Research and Quality, 2005.

Healthcare Cost and Utilization Project Nationwide Inpatient Sample (NIS). Rockville, MD: Agency for Healthcare Research and Quality, 2011.

2011 National Health Insurance Statistical Yearbook. Seoul, Korea: National Health Insurance Service, 2013.

Cherkin

, Deyo

, Wheeler

, et al. Physician views about treating low back pain. The results of a national survey. Spine, 1995; 20:1–9; discussion 9–10.

Nachemson

. Newest knowledge of low back pain. A critical look. Clin Orthop Relat Res, 1992; 8–20.

Stevens

, Duarte

, Park

. Promising implications for integrative medicine for back pain: a profile of a Korean hospital. J Altern Complement Med, 2007; 13:481–484.

10.

Robinson

, Liu

. Oriental and traditional medicine—supporting the vision for integrated health. Eur J Integr Med, 2012; 4:e363–365.

11.

Lin

, Chen

. Advances on study of treatment of lumbar disk herniation by Chinese medicinal herbs. Zhongguo Zhong Yao Za Zhi, 2007; 32:186–191.

12.

Chung

, Lee

, Shin

, et al. Modulation of acute and chronic inflammatory processes by a traditional medicine preparation GCSB-5 both in vitro and in vivo animal models. J Ethnopharmacol, 2010; 130:450–459.

13.

Kim

, Yoon

, Lee

, et al. Protective effect of GCSB-5, an herbal preparation, against peripheral nerve injury in rats. J Ethnopharmacol, 2011; 136:297–304.

14.

Kim

, Park

, Kang

, et al. Effect of GCSB-5, a Herbal formulation, on monosodium iodoacetate-induced osteoarthritis in rats. Evid Based Complement Alternat Med, 2012; 2012:730907.

15.

Cha

, Lee

. HPLC analysis and screening of standard compound on cibotiirhizoma for standardization of GCSB-5 preparation. Kor J Pharmacogn, 2010; 41:48–53.

16.

Cha

, Lee

. HPLC analysis and screening of standard compound on saposhnikoviae radix for standardization of GCSB-5 preparation. Kor J Pharmacogn, 2009; 40:103–108.

17.

Lee

, Cha

. Quantitative analysis of glycine semen nigra and eucommiae cortex for standardization of GCSB-5 preparation. Kor J Pharmacogn, 2009; 40:18–24.

18.

Lee

, Cha

. Quantitative analysis of acanthopanacis cortex and achyranthis radix for standardization of GCSB-5 preparation. Kor J Pharmacogn, 2008; 39:316–323.

19.

Lee

, Kwon

, Lee

. SHINBARO, a new herbal medicine with multifunctional mechanism for joint disease: first therapeutic application for the treatment of osteoarthritis. Arch Pharm Res, 2011; 34:1773–1777.

20.

Kim

, Kang

. A descriptive statistical approach to the Korean pharmacopuncture therapy. J Acupunct Meridian Stud, 2010; 3:141–149.

21.

Lee

, Pittler

, Shin

, et al. Bee venom acupuncture for musculoskeletal pain: a review. J Pain, 2008; 9:289–297.

22.

Assendelft

, Morton

, Yu

, et al. Spinal manipulative therapy for low back pain. A meta-analysis of effectiveness relative to other therapies. Ann Intern Med, 2003; 138:871–881.

23.

Bronfort

, Haas

, Evans

, et al. Evidence-informed management of chronic low back pain with spinal manipulation and mobilization. Spine J, 2008; 8:213–225.

24.

Farrar

, Young

Jr , LaMoreaux

, et al. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain, 2001; 94:149–158.

25.

Turk

, Rudy

, Sorkin

. Neglected topics in chronic pain treatment outcome studies: determination of success. Pain, 1993; 53:3–16.

26.

Jeon

, Kim

, et al. Validation in the cross-cultural adaptation of the Korean version of the Oswestry Disability Index. J Korean Med Sci, 2006; 21:1092–1097.

27.

Saur

, Ensink

, Frese

, et al. Lumbar range of motion: reliability and validity of the inclinometer technique in the clinical measurement of trunk flexibility. Spine, 1996; 21:1332–1338.

28.

Hart

, Werneke

. Re: Pengel LHM, Refshauge KM, Maher CG. Responsiveness of pain, disability, and physical impairment outcomes in patients with low back pain. Spine, 2004; 29:2475–2476.

29.

Chow

, Adams

, Herbert

. Straight leg raise test high reliability is not a motor memory artefact. Aust J Physiother, 1994; 10:107–111.

30.

Rebain

, Baxter

, McDonough

. A systematic review of the passive straight leg raising test as a diagnostic aid for low back pain (1989 to 2000). Spine, 2002; 27:E388–395.

31.

Benichou

. Criteria of drug-induced liver disorders. Report of an international consensus meeting. J Hepatol, 1990; 11:272–276.

32.

Ostelo

, de Vet

. Clinically important outcomes in low back pain. Best Pract Res Clin Rheumatol, 2005; 19:593–607.

33.

Buchner

, Neubauer

, Zahlten-Hinguranage

, et al. The influence of the grade of chronicity on the outcome of multidisciplinary therapy for chronic low back pain. Spine, 2007; 32:3060–3066.

34.

Flor

, Fydrich

, Turk

. Efficacy of multidisciplinary pain treatment centers: a meta-analytic review. Pain, 1992; 49:221–230.

35.

van Tulder

, Ostelo

, Vlaeyen

, et al. Behavioral treatment for chronic low back pain: a systematic review within the framework of the Cochrane Back Review Group. Spine, 2001; 26:270–281.

36.

Berthelot

, Rodet

, Guillot

, et al. Is it possible to predict the efficacy at discharge of inhospital rheumatology department management of disk-related sciatica? A study in 150 patients. Rev Rhum Engl Ed, 1999; 66:207–213.

37.

Wagner

, Ehrenhofer

, Lackerbauer

, et al. Rehabilitation of non-specific low back pain. Results of a multidisciplinary in-patient program. Schmerz, 2007; 21:226, 228–33.

38.

Harkapaa

, Mellin

, Jarvikoski

, et al. A controlled study on the outcome of inpatient and outpatient treatment of low back pain. Part III. Long-term follow-up of pain, disability, and compliance. Scand J Rehabil Med, 1990; 22:181–188.

39.

Kool

, Bachmann

, Oesch

, et al. Function-centered rehabilitation increases work days in patients with nonacute nonspecific low back pain: 1-year results from a randomized controlled trial. Arch Phys Med Rehabil, 2007; 88:1089–1094.

40.

Keel

, Wittig

, Deutschmann

, et al. Effectiveness of in-patient rehabilitation for sub-chronic and chronic low back pain by an integrative group treatment program (Swiss Multicentre Study). Scand J Rehabil Med, 1998; 30:211–219.

41.

Frisch

, Widder

. Radicular low back pain: is there a relationship between the treatment regimen and outcome?. Psychiatr Prax, 2004; 31 Suppl 1:S149–151.

42.

MacLennan

, Wilson

, Taylor

. The escalating cost and prevalence of alternative medicine. Prev Med, 2002; 35:166–173.

43.

Chun

, Kim

, Lee

, et al. Republic of Korea: Health system review. Health Systems in Transition, 2009; 11:1–184.

44.

Weber

. Lumbar disc herniation. A controlled, prospective study with ten years of observation. Spine, 1983; 8:131–140.

45.

Atlas

, Keller

, Chang

, et al. Surgical and nonsurgical management of sciatica secondary to a lumbar disc herniation: five-year outcomes from the Maine Lumbar Spine Study. Spine, 2001; 26:1179–1187.

46.

Mazanec

, Okereke

. Interpreting the Spine Patient Outcomes Research Trial. Medical vs surgical treatment of lumbar disk herniation: implications for future trials. Cleve Clin J Med, 2007; 74:577–583.

47.

van den Hout

, Peul

, Koes

, et al. Prolonged conservative care versus early surgery in patients with sciatica from lumbar disc herniation: cost utility analysis alongside a randomised controlled trial. BMJ, 2008; 336:1351–1354.