Cognitive functional therapy intervention including biofeedback for LBP during cycling; a single case study

Introduction

Cycling is one of the most popular recreational sports throughout the world. The incidence of low back pain (LBP) among cyclists appears to range from 32-60%.^1-3 Most LBP lacks a specific diagnosis, and has been described as non-specific chronic low back pain (NS-CLBP).^4,5 The NS-CLBP population is a very heterogeneous group requiring sub-classification.⁵ One of the proposed subgroups in NS-CLBP is the Flexion Pattern (FP) which is the most common pattern4 within the ‘maladaptive motor control impairment’ subgroup.⁴ It is suggested that a well selected subgroup of cyclists with NS-CLBP present with inherent maladaptive motor control (FP) at the lower lumbar spine resulting in a more flexed lumbo-pelvic posture during cycling that is related to a significant increase in pain.^6-8
For the general LBP population several researchers have suggested that enhancing the lumbar lordosis can help to reduce the incidence of LBP.^9,10 Therefore, for a subgroup of cyclists where the LBP is related to a maladaptive flexed position on the bicycle,^2,11-14 trying to alter the adopted lumbo-pelvic position could be a relevant rehabilitation intervention. Testing this hypothesis can be classified in two domains.
Firstly, the low back position can be improved by addressing the non-personal modifiable factors, namely the geometric bike related variables. These includes factors such as saddle angle,² type of saddle,¹⁵ saddle height,¹¹ pedal unit position,¹⁶ type of bike² and reach.¹¹ Making changes to these variables have been proposed as a method to reduce LBP. Two studies examined the influence of such alterations. Bressel and Larson¹⁵ demonstrated that some saddles could increase anterior pelvic tilt and hypothesized this could reduce LBP during cycling. They concluded further investigation into saddle alteration and LBP are required. Further, Salai et al.² revealed that appropriate adjustment of saddle angle inclination caused a major reduction (72%) in the incidence and magnitude of LBP experienced during cycling.
Secondly, the low back position can be improved by addressing personal modifiable factors, namely regaining postural control through active repositioning of the symptomatic lower lumbar spine,¹⁷ without modifying the settings of the bike.

A multi-dimensional Cognitive Functional Therapy (CFT) intervention could be an appropriate and useful rehabilitation/prevention strategy for LBP in cyclists.¹⁸ The general aim of the intervention would be to alter the personal modifiable factors in order to reduce the LBP experienced during cycling, taking into account the complex multi-factorial nature of LBP in sports.¹⁹ This CFT should be specifically directed to regain postural control over the symptomatic lumbo-pelvic region and to facilitate a less end range flexed cycling posture.
While in clinical practice this postural control retraining has been shown to be effective in the management of NS-CLBP disorders,^20,21 its relationship to LBP during cycling has not been investigated rigorously.
Biofeedback (BFB) might be a useful tool to increase subjects’ awareness of their lumbo-pelvic posture during cycling, helping them to avoid provocative end-range postures and reduce the risk of recurrent and chronic LBP. There is evidence of reduced proprioceptive awareness in NS-CLBP²² and further evidence that postural feedback may be a useful adjunct to conventional management of NS-CLBP.²³
Despite the potential that regaining postural control could be a useful rehabilitation intervention for LBP during cycling, until now no cycling field studies have investigated this issue. Nor are there documented cases of the immediate effect of a rehabilitative approach on reducing LBP in cycling. Therefore, the aim of this case study was to evaluate the effect of a specific CFT intervention including BFB on lumbo-pelvic posture to influence both the postural control at the symptomatic lumbar region and the associated levels of LBP during cycling.

Case

A 19 year old male (mass 72.2kg, height 186cm, Body Mass Index: 20.9) was independently evaluated and subclassified based on O’Sullivan’s classification system⁴ by two physiotherapists.

The subject presented with bilateral LBP located at the lower lumbar spine (L4-5 region). The LBP was described as a dull and diffuse ache. Cycling and prolonged sitting aggravated his LBP. Walking and stretching the lower back into extension during cycling relieved his pain. He had a 5 year history of LBP. His average cycling pain was 5/10 (numerical pain rating scale: NPRS) and average pain during activities of daily living (ADL) was 2/10(NPRS). He had a low level of disability (Revised-oswestry disability index (R-ODI) = 6%) and moderate level of kinesiophobia (Tampa Scale for Kinesiophobia (TSK) = 38/68). He was otherwise healthy and took no medication.

Subject had a slouched usual sitting posture with a posterior rotated pelvis. Self-correcting his posture resulted in upper lumbar and thoracic extension. A loss of lumbo-pelvic lordosis and anterior pelvic rotation and increase in thoracic extension was also seen during different functional tasks (forward bending, squatting and sit-to-stand). The L4-5 segment was found hyper-mobile into flexion during passive motion testing and LBP was reproduced during palpation of the L4-5 segment.
Subject cycled with his race bike 5 days/week and an average distance of 400 km/week. He was a competitive cyclist who already cycled for 6 years, but because of the LBP he recently decided to stop competitive cycling. He had no specific structural spinal pathology, no spinal surgery and no neurological symptoms. Based on this musculoskeletal screening the subject was subclassified (by the two physiotherapists) as having a ‘Flexion Pattern LBP disorder’ that was considered directly attributable to the activities of cycling and sitting.
Subject’s personal race bike had a medial cut-out saddle and a 1° anterior tilted saddle angle (measured with a long arm goniometer (Gymna, Belgium).

Intervention – Cognitive Functional Therapy (CFT)

In order to improve postural control and reduce LBP at the lumbo-pelvic region during cycling, the subject underwent a specific CFT intervention including BFB. This consisted of several steps.
Firstly, the underlying mechanism behind the patient’s LBP was clearly explained by an experienced physiotherapist during an educational session on LBP. This cognitive component was deemed essential to give the subject a clear understanding of the relationship between his (prolonged) excessively flexed sitting posture and LBP, and the development and further provocation of peripheral nociceptive pain generation leading to LBP. This session was given by an experienced musculoskeletal physiotherapist (WD) and contained the basic principles on spinal loading, the importance of neutral posture and the concept of neuromuscular control strategies (as a personal modifiable factor) and its influence on pain control.
Thereafter, the subject was taught to regain postural control over his symptomatic lumbo-pelvic region. In practice, this control can be achieved by instructing and learning the subject to rotate the pelvis anteriorly, which has a critical role in facilitating the lumbo-pelvic musculature^24,25 and preventing excessive flexion at the lumbar spine.
Finally, the subject was taught individual exercises aiming to control anterior tilting of the pelvis in different positions (sitting and in four-point kneeling). Subject was asked to practice on a daily basis and to integrate the motor control strategies during ADL and cycling. A sheet with clear instructions for exercises was provided.
To test the effectiveness of this CFT intervention, the subject had to perform two cycling tasks, which are explained below.

Cycling task 1 (no-CFT/BFB) was performed before the start of the CFT intervention. The subject performed a two hour outdoor cycling task on a flat parcourse with his personal race bike. He was instructed to cycle as usual and was guided by a heart rate monitor (Polar, Belgium) to maintain a heart rate between 60-70% of his age-predicted maximum heart rate.^6,26
Cycling task 2 (CFT/BFB) was performed one month after the start of the CFT intervention. The task was identical, except for the following two aspects. To facilitate a less end-range cycling posture, the subject was instructed to actively rotate his pelvis anteriorly as practiced during the CFT. Further, he was provided with external auditory and somato-sensory (vibration) BFB of his lumbo-pelvic posture during cycling. To avoid any potential diurnal effects27,28 the start time was the same (13.30pm). Non-personal modifiable factors were unchanged.

Outcome measures and statistics

Lumbo-pelvic posture was measured using the BodyGuardTM (Sels Instruments nv, Belgium) (http://www.sels-instruments.be), as described in detail elsewhere.^14,29 The subject was positioned on his personal race bike and the maximal lower lumbar lordosis still allowing to cycle was determined. Based on pilot testing and clinical observation, an individualised threshold was set at 62% flexion range of motion (Fl ROM), 20% below his average lower lumbar posture (82.2% Fl ROM) during the first cycling task. External auditory and somato-sensory BFB was provided when the subject exceeded this threshold by assuming excessive lumbo-pelvic flexion.

The NPRS was used to measure the level of LBP, as described in detail elsewhere.¹⁴ The subject’s level of pain was measured at the start, every 15 minutes during cycling and at 30 minutes, one, two and 24 hours after cycling. R-ODI³⁰ and TSK³¹,³² were re-evaluated after the second cycling task.
Comfort level of sitting was recorded using the Category Partitioning scale (CP-50), as described in detail elsewhere.³³ This comfort level of sitting was measured since it has been suggested that an anterior pelvic rotation during cycling could increase perineal pressure¹⁵, leading to sitting discomfort. Scores were filled in at the start and immediately after the two hours of cycling.
BodyGuard^TM data were downloaded to a personal computer, uploaded to Microsoft Excel and compressed from 20Hz to an average value for each minute and per ten minutes of cycling. Paired t-tests were used to determine differences in lumbo-pelvic posture and level of pain between the two cycling tasks. A one-way repeated measures ANOVA was used to determine if the lumbo-pelvic posture changed between the two conditions across the 12 intervals of ten minutes. All statistical analyses were performed using SPSS Version 16.0. The significance level was set at p<0.05.
The subject was contacted by telephone for further follow-up after 2½ months and asked to report on his LBP during cycling and ADL activities.

Clinical outcome

The CFT/BFB intervention significantly reduced (p<0.001) the mean(±SD)% of total lumbo-pelvic flexion from 82.2(±5.2)% during the first cycling task (no-CFT/BFB) to 56.6(±3.6)% during the second cycling task (CFT/BFB) (Figure 1). This was associated with significantly less pain reported during cycling (p=0.01) (Figure 2). During the initial cycling task (no-CFT/BFB), the level of pain gradually increased to 7/10 while the lumbo-pelvic position remained the same. During the second cycling task, when the cyclist adopted a less end-range lumbo-pelvic lordosis using BFB, the level of pain was significantly reduced as it remained at 0/10 up till 90 minutes and then increased to a maximum level of 2/10 for a short period of time during cycling, returning back to 0/10 at the end (Figure 2).

The level of sitting comfort during cycling was the same (5/50), on both cycling tasks, indicating very low pressure/slight discomfort.

The average pain during cycling (average pain over the last week) prior to the first cycling task had decreased from 5/10 to 2/10 over the last week before the second cycling task (one month later). Similarly, the average pain during ADL (average pain over the last week) decreased from 2/10 to 1.5/10 before the first and the second cycling task respectively. R-ODI and TSK scores improved from 6 to 2% and from 38 to 30 respectively after the CFT intervention.

Discussion

This single case study revealed that a cycling specific CFT intervention including BFB significantly decreased lumbo-pelvic flexion over the entire two hours of cycling. This was accompanied by a significant reduction in cycling related LBP. During the initial cycling task, the level of pain gradually increased over time without increase of lumbo-pelvic flexion. In contrast, after the CFT intervention and using the BFB, the cyclist was able to maintain a more towards neutral lower lumbar lordosis during cycling. The LBP developed during cycling was significantly delayed and reduced compared to the initial cycling task. This suggests that an inherent maladaptive motor control dysfunction (FP disorder), resulting in a more end-range flexed posture, can be a key factor in the development of LBP in this cyclist.
Consistent with these findings, similar cognitive and active rehabilitation strategies have been recommended in sports like cycling^11,12,34 and rowing.^35,36 Recently this type of intervention has been shown to be effective in reducing the incidence and level of LBP and disability in rowers.^19,36 For instance, Thorpe et al.³⁶ compared an experimental with a control group through the rowing season till 10 weeks post season. The experimental group received a LBP education session, a physical conditioning program and a specific individualised prescribed physiotherapy intervention (based on a thorough musculoskeletal screening). The control group only received a LBP education session and a physical conditioning program. The findings revealed that a specific individually prescribed physiotherapy intervention was associated with a reduction in the prevalence of LBP in adolescent female rowers across the rowing season. More recently, Perich et al.¹⁹ revealed that a multi-dimensional intervention program (during the season) consisting of a LBP education session, a screening based individualised specific exercise intervention, combined with off-water strength and conditioning sessions resulted in a decreased incidence of LBP and the levels of pain en disability in schoolgirl rowers at mid- and end-season compared with a control group. In clinical practice this CFT intervention has been shown to be effective in the management of CLBP disorders,18,20 and lifting tasks.³⁷ For instance, Fersum et al. compared a 12-week classification based CFT intervention with a Manual therapy and exercise intervention in a randomised controlled clinical trial with 12 month follow-up. The results support the efficacy of the CFT intervention.

While the CFT/BFB approach was highly effective, it is interesting to mention that initially the subject experienced a “stiff and fatigued feeling in the lower back” during cycling when adopting a more anteriorly tilted pelvis. This phenomenon could be related to muscle fatigue and reduced back muscle endurance in this cyclist, and has been described in other LBP populations.^35,38,39 In addition, upright sitting (like instructed during the CFT) is positively correlated with activating key muscles and stimulating back muscle endurance.^24,38,40,41 In contrast, posterior pelvic tilt is correlated with flexion-relaxation42 (myo-electrical silence in the back extensor muscles at the mid-to end-range of trunk flexion) and inversely correlated with back muscle endurance.³⁸ In this study lumbo-pelvic posture did not deteriorate over time during cycling but instead stayed relatively consistent. This suggests that an altered motor control pattern could be the primary driver for LBP during cycling, rather than simply reduced endurance. Since LBP subjects with a Flexion Pattern dysfunction presents with altered motor control patterns resulting with deficits in the spinal stabilizing muscles⁴³ and adopting a more ‘passive’ posture with relative inactivity of these muscles,⁴⁴ the motor control deficit itself may contribute to the reduced endurance.⁴² This may suggest that for these paraspinal muscles, controlling the lumbo-pelvic region, proprioceptive and endurance training may be an important factor in the rehabilitation of chronic LBP. Further research is necessary to test this hypothesis in cyclists.

In this case-study, non-personal modifiable factors were not altered. It is noteworthy that the case subject was cycling on a medial cut-out saddle and a 1° anterior tilted saddle angle. Both factors can facilitate an anterior pelvic tilt without increasing perineal discomfort.¹⁵ While tilting the pelvis anteriorly may increase perineal pressure,¹⁵ in this case study the sitting discomfort levels remained the same, maybe by the tilt and type of the saddle. Most interestingly, it was only after the CFT/BFB intervention that the cyclist was able to actively control the lumbo-pelvic region. This finding further supports the major role of motor control in cycling.
The reduction in maximum pain during cycling from 7/10 during the first cycling task to 2/10 during the second cycling task exceeds the minimal clinically important difference (MCID) (2/10) for the NPRS. Further, the cyclist reported feeling more comfortable on the bike and being able to subjectively produce more power with the lower limbs. Interestingly, follow-up after 2½ months (telephone contact) revealed the subject still could cycling without LBP and that he was able to resume competitive cycling after an absence of one year because of LBP. He also reported significant decrease in LBP in daily life (e.g. during prolonged sitting), and improvements for the TSK and the R-ODI. The score on the TSK improved to a value (30/68) below the cut-off score of 37/68,³¹ meaning that the cyclist had less fear of movement after the CFT intervention.

Limitations and recommendations for further research

To confirm these case-study based findings, a well-powered RCT in a similar NS-CLBP population, with adequate follow-up, is required. Further research is necessary to discern the relative contribution of changing personal and/or non-personal modifiable factors in reducing LBP in cyclists. It is not unlikely that a combination targeting both factors may help.
Finally, from this case study it is not possible to draw conclusions regarding the specific contribution of the BFB device in improving the lumbo-pelvic posture. During the CFT intervention period (without BFB) the subject’s average LBP during cycling (over the last week) had reduced from 5/10 (prior to cycling task 1) to 2/10 (prior to cycling task 2). Further research, comparing FP subgroups of cyclists receiving CFT with or without BFB is required to evaluate the specific contribution of the BFB during the CFT intervention.

Conclusion

This is the first cycling field study revealing that a specific CFT intervention including BFB on lumbo-pelvic posture could significantly change lumbo-pelvic posture and reduce LBP during cycling. The results of this case study suggest that this intervention could be an appropriate and useful rehabilitation strategy for LBP in cyclists. Further research using a larger sample size is warranted.