Background and Purpose. Improving functional performance in patients with chronic low back pain is of primary importance. The purpose of this study was to examine the effects of 2 proprioceptive neuromuscular facilitation (PNF) programs on trunk muscle endurance, flexibility, and functional performance in subjects with chronic low back pain (CLBP). Subjects. Eighty-six women (40.2±11.9 [X̅±SD] years of age) who had complaints of CLBP were randomly assigned to 3 groups: rhythmic stabilization training, combination of isotonic exercises, and control. Methods. Subjects trained with each program for 4 weeks with the aim of improving trunk stability and strength. Static and dynamic trunk muscle endurance and lumbar mobility were measured before, at the end of, and 4 and 8 weeks after training. Disability and back pain intensity also were measured with the Oswestry Index. Results. Multivariate analysis of variance indicated that both training groups demonstrated significant improvements in lumbar mobility (8.6%–24.1%), static and dynamic muscle endurance (23.6%–81%), and Oswestry Index (29.3%–31.8%) measurements. Discussion and Conclusion. Static and dynamic PNF programs may be appropriate for improving short-term trunk muscle endurance and trunk mobility in people with CLBP. [Kofotolis N, Kellis E. Effects of two 4-week proprioceptive neuromuscular facilitation programs on muscle endurance, flexibility, and functional performance in women with chronic low back pain. Phys Ther. 2006;86:1001–1012.]
- Back pain
- Muscle performance
- Muscle strength
- Proprioceptive neuromuscular facilitation
- Range of motion
Exercise is one of the most frequently used modalities in the rehabilitation of subjects with chronic low back pain (CLBP).1 The primary goals of physical exercise in the management of CLBP are to gain muscle strength (force-generating capacity), flexibility, and endurance, to restore injured tissues, and to contribute to the ability to sustain normal life activities, such as those at work.2
Exercise programs for managing CLBP differ with regard to duration, training frequency, intensity, and the mode of exercise. Previous studies1,3 have shown that isometric training can have positive effects on back pain. Passive extension4 or resistance5 training also has been shown to be beneficial. Furthermore, many different clinical trials using a wide variety of general dynamic fitness programs have produced positive clinical effects.3,6,7
In the chronic phases of low back pain, tailored exercise programs have been shown to have positive effects on physical impairments and limitations.1,6 However, the effects of particular types of exercise and the distinct effects of the particular components of exercise length and dosage on performance need to be determined.6,8 Some people with CLBP have benefited from long-term programs, whereas a positive short-term effect can be achieved with very intensive functional restoration programs.9,10,11 However, the particular effectiveness of short-term versus long-term programs is not clear. In our experience, patients very often ask for rapid restoration of functional ability, pain reduction, and enhancement of muscle function for financial reasons. In this respect, the application of short-term intensive exercise programs for CLBP treatment is worthwhile.
Some exercise programs (often called “general exercise programs”12) are designed to enhance trunk performance through the training of long trunk muscles (erector spinae and rectus abdominis), whose primary function is to generate movement. However, current research has shown that in most cases of CLBP, certain muscles of the back (musculi multifidi and transversus abdominis) that stabilize the spine are reflexively inhibited after injury.8,9 These muscles do not spontaneously recover even if patients are pain-free with a return to normal activity levels. Enhancement of function of such muscles may improve trunk muscle strength, endurance, and flexibility. Therefore, research examining the effects of exercise programs (often called “stabilization programs”) that aim to improve trunk stability and strength by training such muscles is worthwhile.
Neurophysiologic studies have linked pain development in the lumbar spine region of the vertebral column with disturbances in the mechanoreceptors and probably with impairment of the superior proprioception centers.10,11 Therefore, exercise programs that enhance proprioception may be beneficial for managing CLBP.
Evidence on the role of stabilization modalities in CLBP with respect to symptom recurrence led to controversial conclusions.12–15 In particular, some studies13,14 supported the use of stabilization exercise programs over general exercise programs for improving the cross-sectional area of the musculus multifidus, whereas other studies found the opposite results.12,15 In a recent study, Koumantakis et al12 found that an 8-week general exercise program reduced disability in the short term to a greater extent than did a stabilization-enhanced exercise approach in patients with recurrent nonspecific low back pain. It was suggested that stabilization exercises do not provide additional benefit in reducing pain symptoms and enhancing functional ability in patients with low back pain. However, the effects of stabilization exercises on muscle endurance and flexibility were unclear. Furthermore, the stabilization program was a combination of static and dynamic exercises performed together with general strength exercises. These facts do not allow clear conclusions to be drawn with regard to which type of exercise is most effective for managing CLBP.
Proprioceptive neuromuscular facilitation (PNF) exercises are designed to enhance the response of neuromuscular mechanisms by stimulating proprioceptors. The patterns of PNF exercises have a spiral, diagonal direction, and the performance of these patterns is in line with the topographic arrangement of the muscles being used.16 The performance of movements in PNF patterns may permit muscles to act in ways that are close to the actions and movements found in various sports.17–19 Therefore, these exercises may be better suited for performance enhancement than are conventional single-plane or single-direction weight-training programs. Furthermore, PNF techniques often have been used to improve the range of motion of a joint19–23 and endurance17,18 as well as performance in a vertical jump.17,18
There are different forms of PNF exercises. Two commonly used forms are rhythmic stabilization training (RST) and combination of isotonic exercises (COI). The RST technique uses isometric contraction of antagonistic patterns and results in co-contraction of the antagonists if the isometric contraction is not broken by the physical therapist. It is used mainly to manage conditions in which weakness is a primary factor and in which stabilization provides stimulation of the agonistic pattern.16 The COI technique is another form of PNF exercise used to evaluate and develop the ability to perform controlled purposeful movements. It involves the performance of alternating concentric, eccentric, and isometric contractions and is used to treat deficiencies in strength and range of motion.24 Although these forms of PNF exercises are used in physical therapist practice, their effects on CLBP are not clear.
To date, information on the effectiveness of dynamic and combined dynamic-static contraction exercises for trunk muscle stabilization and strength is lacking.14 Moreover, the influence of different types of programs on trunk muscle strength, endurance, and range of motion is unclear. If the aim of exercise programs for CLBP treatment is to improve trunk stabilization and flexibility, then examination of the effectiveness of PNF exercises for this purpose may yield useful results. Furthermore, studies on lower-extremity muscles have identified different muscle responses to concentric work and eccentric work, indicating that muscle adaptive responses vary in accordance with the specific training regimen used.25 Whether these differences apply to trunk musculature and PNF programs is not clear. The purpose of this study was to examine the effects of 2 commonly used modified PNF techniques, RST and COI, on range of motion and on dynamic and static trunk muscle endurance.
The experimental design of this study was a factorial design. Baseline muscle endurance, lumbar sagittal mobility, and history of low back pain were recorded approximately 1 week before the start of the program. Subjects then were randomly assigned to 1 of 3 groups: RST (n=28), COI (n=28), and control (C; n=30). The training groups were assigned to their respective programs for a period of 4 weeks. Endurance and mobility testing measurements were repeated at the end of the program and 4 and 8 weeks later.
One hundred eight women (40.2±11.9 [X̅±SD] years of age) with CLBP took part in the study. For the majority of the subjects, the duration of the episode of low back pain26 at the time of initial assessment was longer than 24 weeks. The subjects were selected from a larger sample of subjects with CLBP on the basis of satisfaction of any 1 of the 3 inclusion criteria: they had complaints of low back pain during or after activity, during or after sitting, and during walking on stairs. After being informed about the study, all subjects signed consent forms. None of the subjects received additional physical therapy interventions during the study period.
Physical Activity Questionnaire
The subjects completed a structured questionnaire during the initial assessment phase of the study.18 Physical activity at work and during leisure time was graded according to the frequency and intensity of exercise. The reliability and validity of data obtained with this method have been shown to be acceptable.27
With traditional techniques (Canadian Society for Exercise Physiology28), the participants’ height (in centimeters) and body mass (in kilograms) were used to calculate body mass index (in kilograms per square meter).
Selection of Outcome Measures
People with CLBP often demonstrate reduced muscle strength and endurance levels and altered flexibility accompanied by low-intensity pain levels and reduced functional ability.1,29–32 Therefore, the effectiveness of any exercise program could be tested against the occurrence of each (or some) of these symptoms. Our aim was to combine an array of qualitative measures related to pain and functional ability with quantitative measures related to trunk muscle endurance and flexibility. In this way, the effectiveness of each program could be examined not only on the basis of self-reported indices of pain but also on more quantifiable measurements, thereby providing a more complete profile of the effectiveness of training. The types of tests selected in this study were based on 2 factors: first, the tests should provide an adequate index of muscle performance (strength, endurance, and flexibility), and second, all subjects can perform the tests reliably and without difficulties. In particular, outcome measures included lumbar sagittal mobility (extension and flexion range of motion), static trunk extension and flexion endurance (in seconds), dynamic trunk extension and flexion endurance (number of repetitions), functional impairment, and low back pain symptom scale measures.
Lumbar sagittal mobility.
Lumbar sagittal mobility was measured by the flexicurve technique.33 Total lumbar extension (T12–S2) and total lumbar flexion were recorded to an accuracy of 1.1±0.6 (X̅±SD) degrees. Accuracy was determined in a pilot study in which movement by 3 subjects was recorded with a standard video camera (DCR-TRV460E,* sampling rate=30 Hz). The accuracy of the flexicurve scores was determined by comparing sagittal mobility measurements with data determined by digitization of digital video files.
Trunk flexion endurance.
The dynamic endurance of the abdominal muscles was measured with the curl-up test. Before the main test, there was a demonstration of the movement by a physical therapist, and then 3 to 5 familiarization efforts were performed. In particular, the subjects lay supine with the knees at an angle of 90 degrees and with arms straight at the sides of the body and pointing toward their knees. A Velcro† strap was used to secure their feet on the table. From this position, they performed as many consecutive curl-ups as possible at a rate of 25 per minute to a maximum of 25.34 The rate of exercise was monitored with a metronome, and special care was taken to maintain the predetermined range of motion and arm position throughout the test. If a reduction in the range of motion throughout the test was observed, then the test was terminated.
To measure the static endurance of the abdominal muscles, the subjects were instructed to curl up with straight arms pointing toward their knees until their iliac crests were raised from the table and to hold this posture for a maximum of 240 seconds.34 During the test, we visually checked the maintenance of performance. The test was terminated when the subject could not maintain the same position. The recorded time for the test was used for further analysis. Verbal instructions on correct positioning were provided only at the start of the test.
Trunk extension endurance.
Trunk extension endurance was measured by use of a modification of the Sorensen back extension test.35 The participants lay face down along the top platform of the steps used in the Canadian Physical Activity, Fitness, and Lifestyle Appraisal step test,28 with the trunk cantilevered from the top of the steps. A single mat was placed on top of the steps for the participants’ comfort. Velcro straps were used to stabilize the buttocks and the midthigh of the subjects, and their calves were held by the physical therapist. Their iliac crests were positioned at the edge of the top platform of the steps. The subjects then performed a minimum of 5 familiarization efforts statically and dynamically. In particular, for static endurance assessment, the participants maintained a horizontal position for as long as possible for a maximum of 240 seconds with no rotation or lateral shifting. The test was terminated when the upper torso dropped below the horizontal. The recorded time for the test was used for further analysis. For dynamic endurance assessment, the participants performed as many consecutive trunk extension repetitions as possible at a rate of 25 per minute to a maximum of 25.35 Special care was taken to monitor subject performance throughout the test so that the same technique was maintained throughout. If subjects could no longer perform a full repetition at the predetermined rate and range of motion, then the test was terminated. The number of repetitions performed was recorded.
Functional impairment and low back pain symptoms.
The degree of functional impairment was assessed by means of the Oswestry Low Back Pain Disability Questionnaire.36 This questionnaire is a 10-item scale; each item has 6 ranked detractors, scored from 0 to 5, yielding a maximum score of 50.36 The first section is a pain-related scale, and the other sections deal with various daily activities that are relevant to low back capability.
The intensity of the low back pain symptoms was assessed by means of the Borg Back Pain Intensity Scale (10 points).37 Subjects were required to rate their pain level from normal (0 points) to emergency (10 points). Values of between 1 and 3 indicate low-intensity low back pain, whereas pain scores of higher than 3 indicate high-intensity pain. Pain symptoms were monitored during each testing session as well as throughout the training period.
Administration of Tests
To avoid fatigue effects, the above tests were performed on 2 separate days (2 sessions before training and 2 sessions after training). The order of static, dynamic, and lumbar mobility tests was randomized across subjects and visiting sessions. The Oswestry Low Back Pain Disability Questionnaire and the Borg Back Pain Intensity Scale were administered on the day before and the day after the training program.
The reliability and the reproducibility of these muscle tests have been reported elsewhere.38,39 In a separate session, 10 subjects underwent the tests twice within a 1-week period. The intraclass correlation coefficients [ICC(3,1)] for lumbar extension and flexion mobility measurements were .94 and .93, respectively. Static trunk extension and flexion measurements demonstrated ICCs of .90 and .89, respectively. The ICCs for dynamic extension and flexion strength measurements were .86 and .88, respectively. The ICC for Borg Back Pain Intensity Scale measurements was .79.
The subjects were randomly assigned to 3 groups (Fig. 1): the RST group, the COI group, and the C group. The groups were homogeneous and showed nonsignificant differences in basic characteristics (Tab. 1). The C group was instructed to avoid structured exercise or activities other than those required for normal daily living throughout the study.
The 2 experimental groups (RST and COI) participated in 4-week programs that aimed to develop strength, endurance, and lumbar flexion-extension mobility as well as to improve overall body control. The progression of the groups was monitored by measuring lumbar sagittal mobility (as explained above) weekly for both groups.
The training frequency for both groups was 5 times per week. A typical weekly schedule included training sessions on Monday, Tuesday, Thursday, Friday, and Saturday, with no training sessions on Wednesday and Sunday. Both groups performed standardized warm-up exercises (stationary bicycling for 7–10 minutes and stretching exercises) and cool-down exercises as part of each training session. The training volume per session included 3 sets of 15 repetitions. The rest intervals between repetitions and sets were 30 seconds and 60 seconds, respectively. The speed of performance was monitored with a metronome. Exercise was performed with the subject in a seated position facing the physical therapist. Resistance was provided by the physical therapist by placement of the hands on the upper part of the chest (for trunk flexion) or the scapula-shoulder region (for trunk extension). All training sessions were controlled by the same physical therapist and had a total duration of 30 to 45 minutes. Given the maximal resistance provided each time by the physical therapist, intensity progression through the 4-week period was carried out according to PNF principles.
The RST program consisted of alternating (trunk flexion-extension) isometric contractions against resistance for 10 seconds, with no motion intended16 (Appendix 1). Subjects performed 3 sets of 15 repetitions at maximal resistance provided by the same physical therapist.16 Rest intervals of 30 seconds and 60 seconds were provided after the completion of 15 repetitions for each pattern and between sets, respectively.17,18
The COI program consisted of alternating concentric and eccentric contractions of agonists without relaxation (Appendix 2)16: resisted active concentric contraction for 5 seconds (trunk flexion), resisted eccentric contraction for 5 seconds (trunk flexion), and resisted maintained contraction for 5 seconds (trunk flexion-extension). Three sets of 15 repetitions at maximal resistance were performed. Rest intervals and warm-up and cool-down exercises were the same as those described above.
All tests were performed before, immediately after, 4 weeks after, and 8 weeks after training. Lumbar sagittal mobility, trunk muscle endurance, Oswestry questionnaire, and Borg scale data were checked for normality by use of the Kolmogorov-Smirnov test.
A multivariate analysis of variance (MANOVA) was applied to examine the differences among the 4 test measurements (before, immediately after, 4 weeks after, and 8 weeks after training) for the 3 groups (RST, COI, and C). The level of significance was set at P<.05.
Of the 108 subjects, a total of 86 subjects completed all training and testing measurements (Tab. 1).
The MANOVA indicated a significant interaction (group × time) effect on the dependent variables (P<.05). Univariate analysis of variance (ANOVA) designs then were applied to determine the group × time effect on each dependent variable.
Lumbar Sagittal Mobility
The univariate ANOVA results indicated a significant interaction (group × time) effect on lumbar range of motion both in flexion (Tab. 2) and in extension (Tab. 3). Compared with the pretesting measurements, the gains in range of motion ranged from 8.6% to 24.1%. Post hoc Tukey tests indicated that the RST and COI groups showed significantly increased lumbar flexion after the program (Tab. 2). In contrast, lumbar extension increased significantly only in the RST group (Tab. 3). No significant changes were observed in the C group.
Dynamic Trunk Muscle Endurance
The univariate ANOVA indicated a significant (P<.05) interaction effect on the dynamic trunk flexion and extension tests (Fig. 1). Post hoc Tukey tests indicated that both training groups showed significantly increased trunk flexion performance after training. The posttraining trunk extension scores showed increases of 39% (RST group) and 81% (COI group) over the pretraining scores. The corresponding values for trunk flexion scores were 43% and 50% for the RST and COI groups, respectively. The COI group showed significantly (P<.05) higher gains in trunk extension performance than did the RST group (Fig. 1).
Static Trunk Muscle Endurance
There were significant (P<.05) interaction effects on static trunk flexion and extension tests (Fig. 2). Post hoc Tukey tests indicated that both training groups showed significantly (P<.05) increased trunk flexion and extension performance after training. Furthermore, the COI group showed significantly (P<.05) higher gains in trunk muscle endurance than did the RST group, mainly during extension (Fig. 2). The posttraining trunk extension scores showed increases of 41.6% (RST group) and 69.5% (COI group) over the pretraining scores. The corresponding values for trunk flexion scores were 23.6% and 26.4% for the RST and COI groups, respectively.
Functional Impairment and Low Back Pain Symptoms
The Oswestry functional disability score was low (<18.5%) and had decreased significantly (P<.05) for the RST and COI groups 4 weeks after the end of the training program (Tab. 4). The posttraining scores showed increases of 29.3% (RST group) and 31.8% (COI group) over the pretraining scores.
The univariate ANOVA showed a significant interaction (P<.05) effect on the Borg Back Pain Intensity Scale measurements (Tab. 5). Post hoc comparisons indicated that although each exercise group (RST and COI) demonstrated a significant decrease in back pain measurements after training compared with pretesting measurements, there were no significant differences among the 3 groups for each testing session (Tab. 5).
The main findings of the present study are that 4 weeks of PNF exercises significantly increase the lumbar spinal range of motion and spinal muscle endurance in people with CLBP. To our knowledge, this is the first study to examine the use of PNF exercises for management of CLBP.
In the present study, lumbar flexion mobility increased after both RST and COI programs (Tab. 2). This finding is in agreement with previous findings3 indicating that it is possible to significantly increase range of motion and endurance in people with CLBP by use of a 4-week intensive PNF exercise program. The positive effects of the present training programs could be attributed to the nature of PNF exercises, which are designed primarily to maximize improvements in flexibility. Such exercises take advantage of the body’s inhibitory reflexes to improve muscle relaxation. This muscle relaxation allows a greater stretch magnitude during stretch training, which should result in superior gains in flexibility. These results provided further support of previous findings on the positive effects of PNF techniques on joint flexibility.19–23
Our results also indicated that at 4 weeks after the end of the COI program, there were lesser gains in extension range of motion than in flexion range of motion (Tab. 3). This result would seem to be related to the fact that there is generally less available lumbar range of motion in extension (Tab. 3) than in flexion (Tab. 2). These findings could be attributed to the fact that trunk extensor musculature works more statically and has a higher proportion of connective tissue than does trunk flexor musculature.40 For this reason, untrained people demonstrate stiffer trunk extensor musculature than trunk flexor musculature.41 Because of differences in stiffness, it is reasonable that the application of the same training program would result in greater increases in trunk lumbar flexion mobility than in extension mobility.
Trunk muscle endurance has been identified as a potential risk factor for the development of back pain.39 In the present study, significant gains in both dynamic (Fig. 1) and static (Fig. 2) muscle endurance were observed. This finding could be attributed to the fact that both exercise techniques involve muscle work at significant intensity levels that result in muscle strength and endurance improvements. Exercise intensity was progressively increased and adjusted to each subject’s performance; therefore, significant muscle system adaptations were observed at the end of each exercise program.42 In the present study, the RST and COI exercise programs were based on the performance of static and dynamic muscle actions, respectively. The principles of training theory would suggest that muscle adaptations are specific to the type of exercise applied.43 However, training adaptations may not always follow this principle, depending on the training status of the individual. For people with low levels of strength, like those in the present study, improvements throughout the force-velocity spectrum may be produced regardless of the training resistance or style used.44
Although muscle endurance measurements were significantly improved for both exercise groups, careful inspection of the results indicates that the COI group demonstrated greater lumbar mobility (Tabs. 2 and 3) and muscle endurance (Figs. 1 and 2) gains than did the RST group. This finding could be attributed to the dynamic nature of the COI exercises, which used all muscle action types (eccentric, concentric, and isometric) through a progressively increased range of motion, as opposed to the static nature of the RST exercises.22,45 These features also may explain the greater strength and sagittal mobility adaptations observed for the COI group than for the RST group. The significant improvements in sagittal mobility indicated that the RST form of PNF exercises is particularly useful for subjects who are unable or reluctant to perform dynamic exercises.
There were positive changes in the Oswestry Index for both exercise groups (Tab. 3). This finding is in agreement with the results of Manniche et al,46 who demonstrated that among patients with low back pain, intensive training had to continue for more than 2 months to achieve significant pain reduction, whereas a 3-week daily intensive training program was found to be equally efficient.6 The improvements in functional ability (as registered by the Oswestry Index) could be seen as a direct result of flexibility and endurance improvements, thereby providing further support for the effectiveness of PNF exercises for CLBP treatment.
Despite the improvements in muscle mobility, endurance, and functional ability, the back pain intensity scale measurements showed similar improvements in both the training groups and the C group (Tab. 4). This finding indicated that improvements in physical measures do not necessarily lead to pain reduction in subjects with low back pain, at least as registered with the Borg Back Pain Intensity Scale. This finding could be attributed to the moderate correlation between the performance of functional tests by subjects with CLBP and pain intensity ratings.47 Apart from physical factors, pain symptoms in people with CLBP may be affected by psychological or social factors,11,48 which may not be affected easily by a 4-week exercise program.
The exercise programs applied in the present study were short-term intensive programs. An important limitation of such programs is that any improvements may not be permanent. The results of the present study indicated that training improvements were evident even 8 weeks after the exercise programs ended. This finding is in agreement with previous data10 and suggests that improvement in trunk muscle endurance in people with CLBP during 4 weeks of progressive exercise is possible. However, our study did not monitor the long-term effects of the exercise programs; therefore, the long-term effects of PNF exercises on CLBP are not clear. Improvements in muscle strength and mobility may be reduced in the longer term, and further treatment may be required.49 Despite this situation, the percentages of increases after training reached 20% for lumbar mobility, 80% for endurance, and 31.8% for Oswestry Index scores; these data provide a basis for the application of COI and RST exercises for CLBP treatment. Of course, continuation or replication of such programs may produce even better results.
The results of the present study are applicable to people with neither acute nor very long lasting CLBP. This conclusion also is evident from the values for the muscle tests performed at the initial baseline measurement session compared with the values reported for subjects who were healthy in other studies34 as well as the mild reduction in functional ability indicated by the Oswestry Index. The effectiveness of PNF modalities in managing acute or more intense CLBP problems is unclear and deserves further investigation.
The application of 4-week RST and COI PNF programs increased the muscle endurance of people with CLBP by 23.6% to 81%. Back pain intensity and functional disability also decreased significantly. These results suggest that short-term programs with dynamic or static PNF exercises are particularly effective in improving trunk muscle endurance and mobility as well as in reducing back pain symptoms and improving functional performance in people with CLBP. Because the COI group showed greater improvements, the use of dynamic PNF exercises for the management of CLBP appears to be more effective. The resulting changes appeared to last for a period of 2 months after training. Further studies are needed to examine the long-term effects of PNF exercises.
Dr Kofotolis provided concept/idea/research design, data collection, subjects, and facilities/equipment. Dr Kellis provided data analysis, project management, and institutional liaisons. Both authors provided writing.
This study was approved by the Aristotle University of Thessaloniki ethics committee.
↵* Sony Corp, 6-7-35 Kitashinagawa, Shinagawa-ku, Tokyo 141-0001, Japan.
↵† Velcro USA Inc, 406 Brown Ave, Manchester, NH 03103.
- Received August 12, 2005.
- Accepted January 30, 2006.
- Physical Therapy