1.2.2.1

Swimming

Given that swimming is an activity in which the participant’s bodyweight is fully supported, this activity is frequently recommended to LBP patients after an FRP. Aquatic exercises have demonstrated efficacy . However, the impacts of the various strokes in an LBP patient population have hardly been studied. From a biomechanical point of view, only butterfly stroke (requiring a more intense effort, with hyperlordosis and activation of the erector spinae muscles) is likely to promote the recurrence of LBP or even the occurrence of spondylolysis . We did not find any reports of an exacerbation of back pain for strokes other than butterfly. A case-control study comparing high-level swimmers and leisure swimmers did not find any difference in terms of the prevalence of the pain – regardless of the type of stroke . A greater observed prevalence of disc degeneration (L5/S1) in high-level swimmers suggested the presence of a relationship with training volume, training intensity and the number of years of training. The intensity and duration of effort can be easily modified and the range of available strokes (backstroke, breaststroke, crawl, etc.) and the fact that the body’s weight is fully supported means that the biomechanical stimulations can be easily modulated.

There is no proof that leisure swimming after an FRP increases pain levels (grade C). The observation of lower biomechanical loads during aerobic activity suggests (but does not prove) the presence of a beneficial effect. Butterfly should only be performed under supervision (grade C).

1.2.2.2

Walking

In principal, walking is the easiest means of performing physical activity.

A three-dimensional analysis of gait at different cadences in the healthy subject (based on the inverse solution approach) shows that the loads placed on the lumbopelvic region while walking are often lower than those in many activities of daily living, regardless of the cadence (with the exception of pelvic rotation, which is accentuated during brisk walking). Low-amplitude arm swings (i.e. during slow gait) tend to increase the “static” load on the lumbopelvic region, which becomes cyclic and thus better tolerated when amplitude of the arm swing is greater (i.e. during brisk walking) .

Sculco et al. randomized 35 chronic LBP patients into two groups: an exercise group, with 45-minute sessions of aerobic walking or cycling at 60% of the age-predicted maximum heart rate (HR _max ) for four times a week for 10 weeks, and a control group in which the participants maintained their normal everyday levels of activity. The exercise group had significantly better mood and psychological status (relative to the controls) but there was no significant difference in perceived pain levels between the groups. Two years later, there was no inter-group difference in the frequency of medical consultations for pain. However, over the mid- to long-term, regular walking did not appear to be associated with new pain or the exacerbation of existing pain.

The results of a study by Taylor et al. showed that moderate levels of walking are beneficial in the routine management of acute LBP. The exercise consisted in walking on a treadmill for 10 minutes at a self-chosen speed and then (after a two-minute break) walking rapidly for 5 minutes. The researchers observed a decrease in pain immediately after the walking. Brisk walking did not lead to an apparent exacerbation of the symptoms. However, on the basis of a follow-up assessment at 6 weeks, it was not possible to say whether or not the decrease in pain was maintained over time.

In a literature review of just four sufficiently high-quality studies, Hendrick et al. concluded that there was enough evidence to state that walking was effective for chronic LBP. The low compression loads involved during walking and the cyclic nature of the movements, can improve disc nutrition the latter’s load-bearing capacity. However, aged-related disc degeneration may compromise this adaptation .

Nordic walking is especially popular in northern Europe and is performed by applying force to special walking poles with each stride (as in classic-style cross-country skiing). A prospective, single-blind, randomized study evaluated the effect of Nordic walking on function and pain levels over an 8-week period in a three groups of chronic LBP patients :

• •
  

  a group that performed unsupervised Nordic walking after a session with an instructor;

  
  
• •
  

  a supervised group with two sessions a week;

  
  
• •
  

  a control group given advice on staying active.

The participants were assessed at 8 weeks, 6 months and 12 months. The functional and pain scores in the supervised Nordic walking group did not differ significantly from those of the other two groups. However, the mean improvement was greater in the supervised group. No harmful effects of Nordic walking were apparent .

Conclusions on walking in the LBP patient are contradictory (grade C). None of the types of walking are associated with adverse effects (grade B). In view of the low loads involved, regular walking (with or without poles) can be recommended (grade C).

1.2.2.3

Cycling

Leisure cycling appears to be a favoured means of promoting aerobic training. Although many studies have focused on the relationship between sports cycling and LBP, very few have looked at the positive or negative impacts of leisure cycling as a physical activity in the LBP patient.

In biomechanical terms, the cyclist’s seated posture implies the maintenance of lumbar kyphosis. The greater the proportion of bodyweight placed on the arms, the lower the total mechanical load on the spine. However, to reduce wind resistance, a cyclist must decrease his/her frontal surface area by flexing the thoracolumbar spine and the hips. This position increases the intradiscal pressure and thus accentuates the risk of LBP related to repeated or excessive spinal loading. The contraction of the lumbar paravertebral muscles is proportional to the pedalling cadense and is low in aerodynamic positions. The abdominal muscles are relaxed in all positions and at all pedalling cadenses. This imbalance can cause back pain in people lacking adequate physical preparation . Adjustment of the handlebar height, the saddle angle and the saddle’s distance from the handlebars and position relative to the bottom bracket axle have an influence on posture and spinal loading ( Fig. 2 ). When the saddle is positioned in front of the bottom bracket axle and the tip of the saddle lowered , lumbopelvic posture is improved and back pain incidence likely diminished.

Cycling postures: 1: on a racing bike, the cyclist’s lumbar spine is in kyphosis; 2: on a mountain bike, the cyclist’s lumbar spine is closer to lordosis. Bottom right: the saddle in position normal then advanced further forward relative to the bottom bracket. I: hamstring; S: sacrum; L: lumbar spine.

In a cross-sectional study, back injuries were the fifth most prevalent injuries (30%) in leisure cyclists . In view of this high prevalence, Salai et al. demonstrated that the above-mentioned, simple technical adjustments could decrease the incidence and intensity of LBP in almost of 70% of the cyclists studied . Other than a demonstration of the beneficial effect of supervised cycling exercises , no studies have recorded a beneficial effect of leisure cycling in LBP patients.

Cycling may be beneficial for the LBP patient because it constitutes aerobic activity. The cycling position influences the spinal load and the requirement for technical adjustments (grade C). A “hybrid” or “town” bike should be recommended because the riding position is similar to that adopted on a mountain bike.

1.2.2.4

Tai chi

Although tai chi is not well known in the Western world, this non-contact martial art is very widespread in south-east Asia and China and can be employed as a physical activity in LBP. Only one study on this topic has been published.

In a cross-sectional study, Hall et al. divided 160 LBP patients (aged 18–70) into two groups: the interventional group ( n = 80), who received two 40-minute tai chi sessions a week for 8 weeks and then one session a week for two further weeks, and the control group ( n = 80), who continued to go about their everyday activities. In the tai chi group, the bothersomeness of back pain (measured on a 0–10 visual analogue scale) fell by 1.7 points and the pain’s intensity fell by 1.3 points. The self-reported disability (measured on the 0–24 Roland–Morris Disability Questionnaire scale) fell by 2.6 points.

Hence, tai chi had beneficial effects on persistent LBP, with a reduction in pain and the related disability (grade C). However, further research will need to confirm that practicing tai chi can be considered as a reliable, efficacious technique in chronic LBP.

1.2.2.5

Team sports

The incidence of the spinal injuries in team sports (soccer, rugby, handball, basketball, etc.) is low in (often young) professional players and is therefore less studied in amateur players.

Soccer is the most-played sport in France (with 1,641,283 licensed club players in 2011). Kicking is the moment of play that most involves the spine: spinal hyperextension during kick preparation and then sudden hyperflexion upon contact with the ball. Leisure football training (e.g. an hour twice a week) increases trunk activity and coordination. If the subject can react more rapidly to a sudden load, the spine is better stabilized and the risk of lumbar injury is lower (decreasing the trunk’s reaction time to load has a preventive effect on the occurrence of LBP) . In theory, soccer training has a favourable effect on LBP. However, intensive football practice in the adolescent is associated with a risk of LBP .

Rugby, handball and basketball are high-intensity sports in which physical contact between players is frequent. The movements are forceful, repeated and sudden. However, there are no studies on rugby, handball and basketball in a context of LBP.

Leisure football training may have a beneficial effect on lumbar coordination (grade C). Intensive football playing is not advisable and any resumption after an FRP should be progressive.

1.2.2.6

Tennis

In biomechanical terms, the tennis serve results from a combination of lumbar hyperextension, lateral bending of the trunk and shoulder rotation, followed by trunk flexion. Tennis players are prone to prolapsed discs (43%) . More generally, repeated microtrauma to the spine (repetitive rotational forces applied to the disc and coupled with hyperextension forces) increases the shear forces on the caudal lumbar discs and thus augments the risk of back problems . In fact, the quality of the technique, the intensity of the activity and the court’s surface appear to have decisive impacts on the occurrence of LBP . For example, a clay court has a soft, loose surface that limits load along the spinal axis, enables players to slide after a change in direction and thus decreases the shear forces on the vertebral discs.

It appears that LBP is a frequent problem in professional tennis players . A cross-sectional study has confirmed that the prevalence of back pain did not differ significantly when comparing 388 amateur tennis players and 245 controls . In contrast, there are no studies on LBP patients’ tolerance of playing tennis.

In view of the stresses described above, tennis appears to expose the player to additional risk of back injury (grade C). The resumption of tennis after an FRP and in the absence of supervision cannot be recommended. There is justification for not performing an overarm serve (at least temporarily) and for playing on a clay court (grade C).

1.2.2.7

Horse riding

Horse riding subjects the spine to compression forces and shear forces during both hyperextension and hyperflexion. The rider’s position on the saddle and the length of the stirrups influence the extent of lumbar lordosis . Decreasing the lumbar lordosis increases the stress on the muscles and ligamentous structures of the lumbar spine that work to maintain the body’s posture and balance. The proportion of the forces related to the horse’s movements that is transmitted to the rider varies as a function of the horse’s speed (walking, trotting or galloping) . Lumbar lordosis enables the rider to absorb the compression forces more easily. When the thighs are at an angle of 30–40° (relative to the horizontal), the lordosis lumbar is low. In this posture, the forces exerted on the discs and lumbar vertebrae exceed 65% of the rider’s bodyweight. If the stirrups are lengthened, the thighs can be at an angle of 45° and lumbar lordosis is maintained. Furthermore, this posture lowers the muscle and ligament tensions in the lumbar regions and hips. Use of longer stirrups also leads to a reduction in pelvic retroversion (which can be as high as 20° with short stirrups). Hence, so-called “western”, long-stirruped saddles are better suited to the maintenance of lumbar lordosis than short-stirruped English saddles ( Fig. 3 ).

Types of saddle: 1: Western; 2: English.

Horse riding is one of the most dangerous leisure activities, with an injury rate of 35.7 per 100,000 riders. Back injuries (mainly bruising or muscle strains) account for 19% of these injuries . Indeed, back pain is one of the most frequent symptoms in competitive and professional riders, in whom the incidence of spinal pain is very high (72.5%) . There are no data on leisure riding. In an open, non-controlled study, Hakanson et al. studied the effects of equine-assisted therapy (EAT, which does not solely include horse riding) on 24 patients with chronic, invalidating LBP . The patients had already undergone a course of physiotherapy. The EAT programme included one or two individual or group sessions per week for periods of between 6 weeks and 12 months. The patients rode under the supervision of a physiotherapist and a riding instructor. A questionnaire was used to evaluate the EAT.

Hakanson et al. concluded that EAT relieved pain, decreased muscular stress and improved self-confidence, well-being, balance, mobility and posture. However, the study’s many sources of bias (an open design, an ill-defined interventional programme and evaluation of the results with a non-validated questionnaire) prevent firm conclusions from being drawn. Leisure riding may have beneficial effects in LBP patients and can be recommended (grade C). Use of a “western”, long-stirruped saddle should be recommended for biomechanical reasons (grade C).

1.2.2.8

Judo and other martial arts

Few studies have addressed the relationship between back pain and martial arts defined broadly and none have looked at the impact of performing these sports on LBP. However, all the available data show that the prevalence of LBP in martial arts performers is lower than in the equivalent age classes of the general population and the higher the rank obtained and the greater the frequency of training, the lower the prevalence of LBP.

1.2.2.9

Gymnastics

Artistic and rhythmic gymnastics, aerobics, acrobatics and tumbling expose the lumbar vertebral column to repeated hyperextensions; the prevalence of LBP in gymnasts has been estimated at between 30% and 85%. Rhythmic gymnastics is the most harmful for the spine . There are no studies on performance of gymnastics by LBP sufferers.

1.2.2.10

Golf

In golf, it is primarily the swing phase ( Fig. 4 ) that creates considerable load on the vertebral column, with anteroposterior shearing and compression during axial rotation and lateral bending. The trunk’s speed of rotation is very high (up to 200°/second) . Axial rotation has already been identified as a risk factor for back pain/injury, whereas the peak compression load is equivalent to eight times the bodyweight . Both “standard” and “modern” swing techniques lead to the same type of stresses: lateral bending, combined with rapid, high-energy rotation .

Types of golf swing: modern swing where players are turning freely from their shoulders (left); old swing where players keep their head still (right).

The incidence of golf-related lumbar injuries is between 15% and 34% in amateur golfers and between 22% and 24% in professionals. Overall, the incidence of LBP is 25% to 36% in male golfers and 22% to 27% in female golfers . In a one-year, prospective study of 196 golfers , the lifetime incidence of LBP was 63%, with incidence rates of 28% in the previous month and 8% at the end of study. Burdorf et al. did not find any evidence of an association between back pain and the frequency or duration of golf playing, although golfers with a history of severe LBP had a greater likelihood of relapse. However, 80.5% of the golfers with LBP did not consider that golf increased their pain levels .

Studies of possible predictors of LBP have identified a harmful role of a sub-10-minute warm-up, carrying the golf bag, repeated or intensive movements (more than 200 drives/weeks and more than 4 rounds a week) and poor swing technique . Modifying the technique can have a beneficial effect on chronic LBP. In one study, 145 LBP patients participated in a multidisciplinary golf rehabilitation programme based on correct grip and how to perform a standard swing with a shorter backswing (a 45° reduction in amplitude); 83% of the participants changed their swing techniques and 98% were able to resume golf at their previous level . In fact, use of a short backswing did not decrease golfing performance (i.e. there was no decrease in the speed of the head of the club upon contact with the ball) and protected the golfer’s lumbar spine by reducing the risk of back injury .

Golf can be recommended as long as the player warms up for at least 10 minutes, does not carry the golf bag and improves his/her swing technique (grade C).

1.2.2.11

Running

Running is a sport that involves repeated impacts. At heel strike, the lumbar spine bears a compression load of between 2.7 and 5.7 times the bodyweight .

The lifetime prevalence for LBP in runners is 75% . Footwear quality (cushioning, energy return, etc.) has a protective effect by decreasing shock transmission to the spine . Progressive training and smooth efforts also have a positive role .

In a randomized, controlled study, Turner et al. evaluated running as the primary intervention in 96 patients (aged 20–65) having suffered from chronic LBP for least 6 months . The 12-month study featured a control group and three interventional groups (behavioural therapy with aerobic exercise [BE]; behavioural therapy only [B] and aerobic exercise only [E]). Over an 8-week period, the BE and E groups performed five 10- to 20-minutes sessions of rapid walking or slow jogging per week (at 60–70% of HR _max ) and received advice on warming-up and stretching. At 8 weeks, Turner et al. observed a significant improvement in the BE group (relative to the control group) in terms of pain intensity and physical functioning. At the end of the 12-month follow-up period, there were no significant differences between the three interventional groups in terms of self-assessed parameters (pain intensity, functioning, etc.) but there was still a significant improvement in all three groups relative to the control group. Moderate-intensity running does not accentuate LBP and even improves it significantly.

Intensive running appears be a risk factor for LBP (grade C). Regular, progressive training and the use of appropriate shoes and insoles that decrease shock transmission to the lumbar spine can be recommended for moderate running by LBP patients after an FRP (grade C).