If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
One year effectiveness of neuromuscular exercise compared with instruction in analgesic use on knee function in patients with early knee osteoarthritis: the EXERPHARMA randomized trial
Address correspondence and reprint requests to: A. Holsgaard-Larsen, Orthopaedic Research Unit, Department of Orthopaedics and Traumatology, Odense University Hospital, Department of Clinical Research, University of Southern Denmark, Odense, Denmark.
Orthopaedic Research Unit, Department of Orthopaedics and Traumatology, Odense University Hospital, Department of Clinical Research, University of Southern Denmark, Odense, Denmark
Research Unit for Musculoskeletal Function and Physiotherapy, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark
Department of Mechanical Engineering, Stanford University, Stanford, CA, USADepartment of Orthopaedic Surgery, Stanford University, Stanford, CA, USAVA Joint Preservation Center, Palo Alto, CA, USA
Research Unit for Musculoskeletal Function and Physiotherapy, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark
To test long-term effectiveness of neuromuscular exercise (NEMEX) with instructions in optimized pharmacological treatment (PHARMA) on activities of daily living (ADL) in patients with early knee osteoarthritis.
Design
12-months follow-up from a randomized controlled trial. Participants with mild-to-moderate medial tibiofemoral knee osteoarthritis were randomly allocated to 8 weeks NEMEX or PHARMA. The primary outcome measure was the ADL-subscale of the Knee Injury and Osteoarthritis Outcome Score (KOOS). Secondary outcome measures included the other four KOOS-subscales, the University of California Activity Score (UCLA) and the European Quality of Life-5 Dimensions.
Results
Ninety-three patients (57% women, 58 ± 8 years, body mass index 27 ± 4 kg/m2) were randomized to NEMEX (n = 47) or PHARMA group (n = 46) with data from 85% being available at 12-months follow-up. Good compliance was achieved for 49% of the participants in NEMEX (≥12 sessions) and 7% in PHARMA (half the daily dose of acetaminophen/NSAIDs ≥ 28 days). Within-group improvements in NEMEX were considered to be clinically relevant (≥10 points) for all KOOS-subscales, except Sport/Rec whereas, no between-groups difference in the primary outcome KOOS ADL (3.6 [−2.1 to 9.2]; P = 0.216) was observed. For KOOS Symptoms, a statistically significant difference of 7.6 points (2.6–12.7; P = 0.004) was observed in favor of NEMEX with 47% improving ≥10 points.
Conclusions
No difference in improvement in difficulty with ADL was observed. NEMEX improved knee symptoms to a greater extent with half of patients reporting clinically relevant improvements.
Neuromuscular exercise (NEMEX) has demonstrated significant short-term effects on patient-reported outcome measures (PROs) in patients with knee or hip osteoarthritis (OA)
Immediate efficacy of neuromuscular exercise in patients with severe osteoarthritis of the hip or knee: a secondary analysis from a randomized controlled trial.
. However, in patients with early knee OA, we previously reported no superior effect of NEMEX compared to information on the recommended use of analgesics and anti-inflammatory drugs (PHARMA) on knee joint load or PROs immediately following the intervention at 2-month (the EXERPHARMA trial)
The effect of instruction in analgesic use compared with neuromuscular exercise on knee-joint load in patients with knee osteoarthritis: a randomized, single-blind, controlled trial.
One-year positive effects of non-surgical interventions including NEMEX have been reported in randomized trials of patients with moderate to severe knee OA
The efficacy of 12 weeks non-surgical treatment for patients not eligible for total knee replacement: a randomized controlled trial with 1-year follow-up.
Effect of preoperative neuromuscular training (NEMEX-TJR) on functional outcome after total knee replacement: an assessor-blinded randomized controlled trial.
Supervised neuromuscular exercise prior to hip and knee replacement: 12-month clinical effect and cost-utility analysis alongside a randomised controlled trial.
(within-group Cohen's d ranging between 0.47 and 0.87). One-year improvements have also been shown when NEMEX is delivered large-scale in clinical practice
Good Life with osteoArthritis in Denmark (GLA: D): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide.
. However, no controlled studies on long-term follow-up exist on the effectiveness of NEMEX in early knee OA, a group mostly seen in primary care.
In this 12-month follow-up study we test the hypothesis that long-term effectiveness of NEMEX, which encourages participants to maintain the exercise program following completion, would be superior in patient-reported physical function compared with instructions in optimized use of analgesics and anti-inflammatory drug use (PHARMA) in patients with early knee OA.
Method
Study design
An extended 12-month follow-up of the EXERPHARMA randomized controlled, prospectively designed, single-center, single-blind, 2-arm (1:1) trial
The effect of instruction in analgesic use compared with neuromuscular exercise on knee-joint load in patients with knee osteoarthritis: a randomized, single-blind, controlled trial.
The effect on knee-joint load of instruction in analgesic use compared with neuromuscular exercise in patients with knee osteoarthritis: study protocol for a randomized, single-blind, controlled trial (the EXERPHARMA trial).
approved by the Regional Scientific Committees for Southern Denmark (identifier: S-20110153) compliant with the Helsinki Declaration following the CONSORT recommendations for non-pharmacological trials and registered at ClinicalTrials.gov (NCT01638962).
Participants
Men and women with a clinical diagnosis of knee OA aged 40–70 years were recruited via general practitioners in the municipalities of Odense and Middelfart, Denmark, and from advertisements in local clubs, libraries, print media, and Facebook. A full list of inclusion and exclusion criteria has been published
The effect of instruction in analgesic use compared with neuromuscular exercise on knee-joint load in patients with knee osteoarthritis: a randomized, single-blind, controlled trial.
. In summary, included subjects had a clinical diagnosis of knee OA, with or without radiographic changes, had no contraindication for exercise, NSAIDs, or X-rays, and had not had any leg surgery/trauma within the last 6 months.
Participants were assessed for eligibility by a general practitioner/project manager. Subsequently, informed consent was obtained and clinical assessments were performed.
Randomization and allocation concealment
As described before, eligible participants were randomly allocated with concealed allocation (only accessible by the biostatistician) i.e., after the participant had been tested at baseline, the envelope was opened and the participant was informed about the allocation, either 8-week exercise (NEMEX) or instruction in analgesic use (PHARMA)
The effect of instruction in analgesic use compared with neuromuscular exercise on knee-joint load in patients with knee osteoarthritis: a randomized, single-blind, controlled trial.
The effect on knee-joint load of instruction in analgesic use compared with neuromuscular exercise in patients with knee osteoarthritis: study protocol for a randomized, single-blind, controlled trial (the EXERPHARMA trial).
. In brief it consists of five parts: warming up (10 min of aerobic activity at “rather strenuous level”), functional, proprioceptive, endurance strengthening, and cooling down.
The training took place in groups (up to 10 participants) twice a week (each session 60 min) for 8 weeks, and was supervised by physiotherapists at local clinics having taken a 2-day course on NEMEX. To allow for progression, three to four levels of difficulty were given for each exercise
The effect on knee-joint load of instruction in analgesic use compared with neuromuscular exercise in patients with knee osteoarthritis: study protocol for a randomized, single-blind, controlled trial (the EXERPHARMA trial).
. No restrictions on home exercises or participation in additional exercise programs besides the NEMEX were provided. Pain up to 2 on the VAS-scale was considered “safe”; between 2 and 5 as “acceptable”; and pain >5 as “avoid”. Over-the-counter and prescribed pain-relieving drugs were allowed, but not recommended.
PHARMA
The PHARMA group received information (by video and a pamphlet) on how best to use acetaminophen and oral NSAIDs, in doses consistent with Danish guidelines and described in detail
The effect on knee-joint load of instruction in analgesic use compared with neuromuscular exercise in patients with knee osteoarthritis: study protocol for a randomized, single-blind, controlled trial (the EXERPHARMA trial).
. Participants were encouraged to take their medication according to their need and to adjust the medication according to their pain levels.
If pain relief from over-the-counter acetaminophen was not sufficient, the pamphlet informed participants to contact their GPs for additional prescription of NSAIDs.
Adherence to the therapy
As reported previously, only 49% of the NEMEX group fulfilled the pre-defined criteria for good compliance of ≥12 sessions, while 19% participated in 10–12 sessions and 32% participated in ≤9 sessions
The effect of instruction in analgesic use compared with neuromuscular exercise on knee-joint load in patients with knee osteoarthritis: a randomized, single-blind, controlled trial.
. Only 7% demonstrated good compliance to the PHARMA intervention defined as taking at least 2,000 mg/daily of acetaminophen or equivalent dose of NSAID for at least 28 days. As a consequence the a priori specified per protocol analysis was not performed
The effect of instruction in analgesic use compared with neuromuscular exercise on knee-joint load in patients with knee osteoarthritis: a randomized, single-blind, controlled trial.
The effect of instruction in analgesic use compared with neuromuscular exercise on knee-joint load in patients with knee osteoarthritis: a randomized, single-blind, controlled trial.
. The primary outcome of the present “sequel analysis”, solely based upon PROs, is the between-group difference in change scores from baseline to 12 months in self-reported difficulty with activities of daily living (ADL) as evaluated by the Knee injury and Osteoarthritis Outcome Score ranging from 0 (extreme problems) to 100 (no problems) (KOOS, www.koos.nu). Secondary outcome measures include the other four KOOS subscales and physical activity evaluated by the University of California at Los Angeles (UCLA) Activity Score
stating the priority of the outcomes to be tested and statistical methods to be used was made publically available before data was delivered to the trial statistician (RC) who was masked to group allocation. Power and sample size considerations for this 12-month follow-up study were evaluated a posteriori: if at least 74 participants completed the trial, the long-term trial would yield sufficient power (80.8%) to detect (P < 0.05) a between-group difference of 10 KOOS points with a SD of 15 (www.koos.nu). Continuous outcome measures were analyzed based on the ‘as observed population’ (respecting the original randomization) using repeated measures mixed linear models, including participants as a random effect, with fixed factors for group (2 levels) and time to follow-up (2 levels [months 2, and 12]) and the corresponding interaction, adjusted for baseline values. The proportion of patients who achieved a minimal clinical important improvement of ≥10 KOOS points (“responders”) was analyzed using Chi-squared test. A 95% CI excluding differences between groups of greater than 10 KOOS points would be interpreted as indicating the absence of a clinically meaningful difference, excluding the risk of performing a type-2 error
In total 72 individuals were initially assessed for eligibility by a general practitioner and 372 by e-mail or telephone interview between October 1st 2012 and May 27th 2015 (Appendix). At 12 months (finalized May 27th 2016), data from 85%, or 41 patients from the NEMEX and 38 patients from the PHARMA group, were available (Appendix). Baseline characteristics are given in Table I.
Table IBaseline characteristics of study participants
Baseline characteristics
PHARMA (n = 46) Mean (SD)
NEMEX (n = 47) Mean (SD)
Women, n (%)
25 (54.3)
29 (61.7)
Age (years)
58.3 (8.1)
57.9 (7.9)
Weight (kg)
80.2 (12.7)
79.1 (12.4)
Body Mass Index (kg/m2)
27.0 (2.9)
26.8 (3.3)
Study knee in right knee, n (%)
25 (54.3)
26 (55.3)
Radiographic knee OA severity of study knee (Kellgren–Lawrence)
Grade 0, n (%)
3 (6.5)
0 (0.0)
Grade 1, n (%)
16 (34.8)
26 (55.3)
Grade 2, n (%)
17 (37.0)
14 (29.8)
Grade 3, n (%)
10 (21.7)
7 (14.9)
Grade 4, n (%)
0 (0.0)
0 (0.0)
Social economic status
Working full-time or part time, n (%)
25 (54.3)
24 (51.1)
Unemployed, n (%)
1 (2.2)
2 (4.3)
Sick leave, n (%)
0 (0.0)
1 (2.1)
Pensioner, n (%)
20 (43.5)
20 (42.6)
Living alone, n (%)
5 (10.9)
9 (19.1)
College education or equivalent, n (%)
26 (56.5)
32 (68.1)
Outcome measures
KOOS subscale scores
ADL
68.4 (17.1)
68.2 (15.5)
Pain
60.1 (15.0)
61.6 (13.7)
Symptoms
66.6 (16.5)
66.1 (16.7)
Sport/Rec
42.6 (22.1)
35.3 (22.4)
QOL
45.6 (16.6)
45.2 (16.6)
UCLA
6.8 (2.1)
7.1 (1.9)
EQ Health state
75.9 (15.2)
73.6 (15.9)
Variables are expressed as the Mean (SD).
Abbreviations: PHARMA, information on pharmacological pain-relief; KOOS, Knee injury and Osteoarthritis Outcome Score ranging from 0 to 100 with high scores indicating fewer problems; EQ-Health state, EuroQol Five Dimensions Questionnaire Time Trade Off.
The within-group mean improvements (±SE) in KOOS ADL after 12 months were 11.4 (±2.0) and 7.9 (±2.0) points in the NEMEX and PHARMA groups, respectively, corresponding to the absence of a statistically and clinically meaningful between groups difference of 3.6 points (95% CI, −2.1 to 9.2; P = 0.216) (Table II), On the individual patient level, the number of responders (improving > 10 points) on KOOS-ADL was 22 patients (47%) in the NEMEX group and 13 (28%) in the PHARMA group; this difference was potentially in favor of NEMEX (NNT of 5.3 patients (Risk Difference (RD) = 0.19; P = 0.065)).
Table IIMean difference within-groups and difference between-groups at 2 and 12 months follow up
2 months follow-up (mean change from baseline)
12 months follow-up (mean change from baseline)
P-value
PHARMA (mean ± SE) (within-group)
NEMEX (mean ± SE) (within-group)
Between-group difference in change (mean [95% CI])
PHARMA (mean ± SE) (within-group)
NEMEX (mean ± SE) (within-group)
Between-group difference in change (mean [95% CI])
Abbreviations: PHARMA, information on pharmacological pain-relief; KOOS, Knee injury and Osteoarthritis Outcome Score ranging from 0 to 100 with high scores indicating fewer problems; EQ-Health state, EuroQol Five Dimensions Questionnaire Time Trade Off.
* Within-group difference in change score P < 0.05, ** Within-group difference in change score P < 0.001.
Despite generally larger and on average clinically relevant effects, in the KOOS subscales for the NEMEX group at 12 months, a statistically significant between-group difference was only observed for KOOS Symptoms, revealing a 7.6 points (2.6–12.7; P = 0.004) difference in favor of NEMEX (NNT of 5.3 patients (RD = 0.19; P = 0.065)). No statistically significant within- or between-group differences were seen for UCLA (0.1, 95% CI, −0.6 to 0.7; P = 0.852) and EQ-5D (2.6, 95% CI, −2.9 to 8.1; P = 0.347) (Table II).
Discussion
Summary
The present study was designed to evaluate potential long-term differences in patient reported ADL and other patient-reported outcomes from two common primary care interventions provided to patients with knee OA. Due to poor adherence, especially in the PHARMA group, and a similar intake of pain relievers in both groups, we effectively compared the addition of 8 weeks of NEMEX to ‘care as usual’. We found that the NEMEX group generally had larger and on average clinically relevant improvements at 12 months. However, the between-group difference at 12 months only reached statistical significance for improvement in symptoms from the knee. Thus, NEMEX could be the preferred choice for long-term relief of symptoms such as swelling, stiffness, catching and noise while avoiding the potential side-effects of the alternative treatment option analgesics and anti-inflammatory non-steroidal drugs.
Patient-reported outcomes
Patient-reported outcomes (PROs) continued to improve from 2 to 12 months. At 12 months, but not at 2 months, the observed within-group improvements compared to baseline in the NEMEX group were considered to be clinically relevant (≥10 points) for all KOOS subscales, except for Sport/Rec. The confidence interval for the between-group difference of 7.6 points (2.6–12.7; P = 0.004) on KOOS Symptoms include 10 which is considered to represent a minimal clinical important improvement (www.koos.nu). Together with the clinical interpretation that 47% improved more than 10 points corresponding to an NNT of 5.3 patients (RD = 0.19; P = 0.065) a clinically relevant improvement for KOOS symptoms in favor of NEMEX cannot be excluded.
Our finding of sustained improvement in PROs during a year is in line with two recent randomized trials investigating the long-term effects of 12 weeks NEMEX
The efficacy of 12 weeks non-surgical treatment for patients not eligible for total knee replacement: a randomized controlled trial with 1-year follow-up.
Exercise therapy versus arthroscopic partial meniscectomy for degenerative meniscal tear in middle aged patients: randomised controlled trial with two year follow-up.
. Both studies observed continued within-group improvements at 12 or 24 months follow-up for the NEMEX groups. Based upon the current trial, and previous 1–2-year follow-ups available
The efficacy of 12 weeks non-surgical treatment for patients not eligible for total knee replacement: a randomized controlled trial with 1-year follow-up.
Effect of preoperative neuromuscular training (NEMEX-TJR) on functional outcome after total knee replacement: an assessor-blinded randomized controlled trial.
Supervised neuromuscular exercise prior to hip and knee replacement: 12-month clinical effect and cost-utility analysis alongside a randomised controlled trial.
, it seems that long-term self-reported effects from 4 to 12 weeks of NEMEX can be achieved in individuals with mild to severe OA. We speculate that the continued improvement following intervention may be due to past participation in exercise as participants were encouraged to maintain the exercise program.
Study limitations
The low adherence with the respective treatments restricts the ability to draw firm conclusions from this study. However, we deliberately chose wide inclusion criteria and to perform the exercise therapy intervention in ordinary private Physical Therapy practices to reflect ‘real world scenario’ and thus, to increase the external validity and generalizability. This approach may however threaten the internal validity of the study. Indeed the adherence was low to moderate and the quality of delivery of the interventions was not monitored and guaranteed the way it can be in a laboratory setting. Our study also had other limitations. The follow-up rate at 12 months was 85%, which is considered acceptable in long-term evaluation of RCTs but may still pose threats to the validity
. Furthermore, PROs are prone to influence of unspecific effects (i.e., placebo). However placebo effects are greater for more invasive treatments and therefore expected to be greater for orally administered analgesics than for exercise. Also, placebo effects wear off with increased time to follow-up, which is opposite to what we found for the NEMEX group in this study. Finally, participation in physical activity and/or structured exercise therapy programs during the follow-up time was not monitored and as such we cannot determine whether the clinically relevant effects seen at 12 months but not at 2 months are caused by the interventions per see or potential behavioral changes.
Conclusion
In this trial, limited by poor adherence, we found no difference at 12 months in improvement in difficulty with ADL. However, NEMEX improved knee symptoms to a greater extent with half of patients reporting clinically relevant improvements. Thus, NEMEX could be the superior choice for long-term relief of symptoms such as swelling, stiffness and mechanical problems while avoiding the potential side-effects of analgesics and anti-inflammatory drugs.
Author contributors
All Authors: conceptualization and design of the study. AHL, BC: collection and assembly of data. AHL, BC, RC, ER: analysis of data. AHL, BC and ER: interpretation of data. AHL: first draft of manuscript. RC: statistical expertise. All authors: critical revision of manuscript for important intellectual content and approval of final version.
Conflict of interests
All authors declare that they have no conflict of interests.
Role of funding source
The authors certify that the grant sponsor has no involvement in study design, collection, analysis and interpretation of data, in the writing of the manuscript, and in the decision to submit the manuscript for publication.
Acknowledgements
This projected was funded by: the Region of Southern Denmark PhD fund, €67,000; the Region of Southern Denmark research fund, €67,000; The Danish Rheumatism Association, €42,000; The Danish Rheumatism Association Ryholts grant, €2,000; the University of Southern Denmark Scholarship, €55,000; The Association of Danish Physiotherapists, €5,000; Odense University Hospital free research funds, €44,000; and Family Hede Nielsens fund, €2,600. The Parker Institute, Bispebjerg and Frederiksberg Hospital (R. Christensen) is supported by a core grant from the Oak Foundation (OCAY-13-309).
Appendix 1. Flow diagram
Flow diagram of participants in the study. Due to low adherence to interventions the per protocol analysis was not performed.
Immediate efficacy of neuromuscular exercise in patients with severe osteoarthritis of the hip or knee: a secondary analysis from a randomized controlled trial.
The effect of instruction in analgesic use compared with neuromuscular exercise on knee-joint load in patients with knee osteoarthritis: a randomized, single-blind, controlled trial.
The efficacy of 12 weeks non-surgical treatment for patients not eligible for total knee replacement: a randomized controlled trial with 1-year follow-up.
Effect of preoperative neuromuscular training (NEMEX-TJR) on functional outcome after total knee replacement: an assessor-blinded randomized controlled trial.
Supervised neuromuscular exercise prior to hip and knee replacement: 12-month clinical effect and cost-utility analysis alongside a randomised controlled trial.
Good Life with osteoArthritis in Denmark (GLA: D): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide.
The effect on knee-joint load of instruction in analgesic use compared with neuromuscular exercise in patients with knee osteoarthritis: study protocol for a randomized, single-blind, controlled trial (the EXERPHARMA trial).
Exercise therapy versus arthroscopic partial meniscectomy for degenerative meniscal tear in middle aged patients: randomised controlled trial with two year follow-up.
31276-1&id=fx1.jpg){kind=link}