Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

188 Official Journal of FIMS (International Federation of Sports Medicine)

ISMJ

International SportMed Journal

Original research article

Effects of saddle height on knee forces of recreational cyclists with

and without knee pain

1 2*Dr Rodrigo Rico Bini, PhD, 2Professor Patria A Hume, PhD

1Laboratório de Pesquisa do Exercício, Universidade Federal do Rio Grande do Sul, Porto Alegre,

Brazil

2Sport Performance Research Institute New Zealand, Auckland University of Technology, Auckland,

New Zealand

*Corresponding author. Address at the end of text.

Abstract

Background: Low saddle height has been empirically linked to increased compressive force at the patellofemoral joint and to the development of overuse knee pain. Research question: How changes in saddle height would affect patellofemoral and tibiofemoral forces of cyclists with and without knee pain. Type of study: Cross-sectional. Methods: Sixteen cyclists without knee pain and eight cyclists with knee pain performed four 2-minute submaximal cycling trials using their preferred saddle height,

two saddle heights (high and low) eliciting ±10 change in knee flexion angle, in relation to their

preferred saddle height, and a saddle height eliciting 25 knee flexion at the 6 o’clock crank position. Workload (3.04 ±0.78 W/kg) was consistent and pedalling cadence was visually controlled at 90 rpm for every trial. Right pedal forces and joint kinematics recorded during all trials enabled calculation of patellofemoral and tibiofemoral forces using a musculoskeletal model. Results: Changes in saddle height did not substantially affect patellofemoral and tibiofemoral forces when comparing cyclists with and without knee pain. Compared to the low saddle height there were larger anterior tibiofemoral forces at optimal (35% without pain, 51% pain) and high saddle heights (76% without pain, 92% pain). Conclusions: Bicycle saddle height can probably be set within a large range of knee motion (i.e., ~44-65° determined during dynamic cycling at the 3 o’clock position) to minimise possible detrimental

effects of large patellofemoral and tibiofemoral forces. Keywords: bicycle configuration, joint kinetics,

joint kinematics, knee flexion angle

*Dr Rodrigo Rico Bini, PhD

Rodrigo’s research focuses on cycling biomechanics with special attention to internal forces. He is affiliated to the International Society of Biomechanics and at the International Society of Biomechanics in Sport. He received the International Student Travel Award from the International Society of Biomechanics to conduct his PhD in New Zealand in 2009. He was also received the II Biofenac Brazilian Award, along with the co-authors, for the best research in Exercise Science.

Professor Patria Anne Hume, PhD

Patria’s research focuses on improving sport performance using sports biomechanics and sports anthropometry, and focuses on reducing sporting injuries by investigating injury mechanisms and injury prevention methods. Patria has published in collaboration with her colleagues over 70 refereed journal articles, 50 edited articles, 15 books and chapters, and 170 technical reports. She has supervised 5 PhD and 8 Masters students to completion and she currently is supervising 18 PhD students. Email: patria.hume@aut.ac.nz

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

189 Official Journal of FIMS (International Federation of Sports Medicine)

Introduction

Overuse injuries may be common in cycling, with up to 85% of cyclists seeking medical attention for overuse injuries per year

1, 2. The

knee joint may be at particular risk of overuse injury, with the annual prevalence reported to be 39% among professional road cyclists and up to 65% among recreational touring cyclists 1, 2

. The position of the body on the bicycle has been suggested to play an important role in the

rates of overuse injuries 3. Studies of cycling

kinematics indicate that lower saddle height leads to greater knee flexion angle in cycling

4-

6. Repetitive forces applied at a large knee joint

angle (i.e. greater knee flexion angle – see Figure 1) may result in soft tissue damage

3,

where knee flexion angle is defined as the angle between the tibia and the forward projection line of the femur.

Figure 1: Schematic illustration of 25 of change in the knee flexion angle from 60 (A) to 35 (B) that should theoretically decrease patellofemoral compressive force (FP). Arrows indicate quadriceps muscle force (FQ) and patellar tendon force (FPT). Patellofemoral (FP) and tibiofemoral anterior shear (FTA) and compressive (FTC) force components illustrated in relation to the tibial plateau (C).

Previous studies have given attention to the effects of saddle height on knee forces

7, 8.

Patellofemoral joint compressive force and compressive and anterior-posterior tibiofemoral forces were the focus of previous investigations because they may be linked to overload in patellofemoral cartilage, menisci

and anterior-posterior cruciate ligaments 9.

However, results from previous studies on knee forces are conflicting. Ericson and Nisell

7

found increases in patellofemoral compressive force using a 10% lower saddle height whilst Tamborindeguy and Bini

8 did not observe

significant changes when lowering or

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

190 Official Journal of FIMS (International Federation of Sports Medicine)

increasing 3% the saddle height. Ericson and Nisell

10 reported that saddle height (11% lower

or 7% higher) followed an inverse relationship with tibiofemoral compressive force and anterior tibiofemoral force; however, no effects of saddle height on tibiofemoral compressive force were found for two other studies with 3-6% increases/decreases in saddle height

8, 11.

Most studies to date have assessed a limited number (i.e. 6-10) of non-cyclists using leg length based methods to set the height of the saddle. Knee joint kinematics are not similar between subjects when leg length methods are used compared to knee flexion angle methods 12

. Therefore, there is need for measurements of knee joint forces using cyclists pedalling at different saddle heights set by a given knee flexion angle.

None of the previous studies provided evidence for knee joint forces of cyclists who reported pain and/or injury in their knees using different saddle heights. Bailey et al.

13

observed greater knee medial projection in the frontal plane for cyclists with anterior knee pain. Therefore it is important to assess cycling motion because cyclists with knee pain may have different knee joint flexion angles compared to healthy cyclists. That would result from either a contained knee motion to reduce pain or to differences in self-selected saddle height or both. If there is an inverse relationship between knee forces and saddle height, cyclists with knee pain would benefit

from using a higher saddle height compared to healthy cyclists. However, there is no experimental evidence to support this hypothesis. No previous studies have assessed knee joint forces in cyclists with exiting knee pain while pedalling using different saddle height. That information is important to ascertain on a range of saddle heights that could potentially reduce the risk of knee injuries.

Therefore the aim of this study was to compare the effects of different saddle heights on patellofemoral and tibiofemoral forces of recreational cyclists with and without knee pain.

Methods

Participants

Twenty-four cyclists without competitive experience in cycling participated in the study. Cyclists reporting pain in the right knee as a result of overuse (i.e. non-traumatic injury) and those who remained riding a bicycle for a minimum of 10 km each day three times per week were included in the study. Cyclists were interviewed to assess training and existing pain related to bicycle riding. They were separated into groups with and without existing knee pain or injury on the right knee. Cyclists with bilateral knee pain were not included in our study due to limitations of using right sagittal plane analyses. The characteristics of the cyclists are shown in Table 1.

Table 1: General characteristics of the participants with differences between cyclists with and without pain reported as percentage differences along with effect size magnitudes.

Groups Pain

(n = 8)

No pain

(n = 16)

Pain vs. No pain

Age (years) 40 (11) 43 (9) 6%; 0.3, S

Body mass (kg) 82 (13) 78 (18) 6%; 0.3, S

Height

(cm)

181 (5) 178 (7) 2%; 0.4, S

Time of cycling training (hours/week)

7 (3) 7 (3) 1%; 0.1, T

Training volume (km/week)

105 (60) 80 (60) 32%; 0.4, S

Abbreviations used for effect sizes of trivial (T), small (S), moderate (M) and large (L).

Prior to the study, the cyclists were informed about possible risks and signed a consent form approved by the Ethics Committee of Human Research at the institution where the study was conducted.

Of the eight cyclists in the knee pain group, three had non-specific knee pain, one had medial knee pain, one had lateral knee pain, two had had surgery, and one had arthritis in the patella. The authors opted for assessing only eight subjects with varying types of knee pain due to difficulties in recruiting cyclists with

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

191 Official Journal of FIMS (International Federation of Sports Medicine)

only right knee pain on sustained bicycle riding at the time of the study.

Data collection

Anthropometrics (height and body mass) were measured according to International Society for Advancement of Kineanthropometry protocols

14. Each cyclist’s bicycle’s vertical

and horizontal position of the handlebars was measured to set up the stationary cycle ergometer (Velotron, Racemate, USA) at their “preferred height” configuration. Saddle height was measured from the central portion of the top of the saddle to the pedal spindle, with the crank in line with the seat tube angle

15 in each

cyclists’ bicycle along with the horizontal position of the saddle to the bottom bracket. These measures were replicated in the cycle ergometer to simulate the configurations used for cyclists on their bicycles. Knee joint flexion angle was measured using a goniometer with

1 of resolution (Physio-Med Services, UK),

while the cyclists held the crank at the 6 o’clock crank position (lowest pedal position in the crank cycle) after pedalling at 100 W for two minutes to accommodate in the ergometer. Before measurements were taken, cyclists were asked to replicate statically hip, knee and ankle angles that they would have during cycling motion. The literature reports varying individual ankle motions in cyclists

16, 17. The

authors then opted to replicate the self-selected ankle position for each cyclist. Saddle height was recorded when the saddle was changed from the preferred position to high (-

10 of knee flexion with respect to the

preferred height), low (+10 of knee flexion with respect to the preferred height), and to the

theoretical optimal (25 of knee flexion - see Figure 2). The optimal saddle height was chosen because it was supposed to reduce the risk of knee injuries

18.

Figure 2: Knee flexion angle and pedal stick illustration along with saddle heights (insert A) and instrumented pedal force system (insert B).

Cyclists then performed 10 minutes of warm-up cycling at 100 W of workload and 90±2 rpm of pedalling cadence on the stationary Velotron cycle ergometer using their preferred saddle height and horizontal position. Workload was then increased to match 3.04 (±0.78) W/kg

(243 ±78 W or 161 ±4 J) and pedalling cadence was visually controlled at 90 (±2) rpm for two minutes. Data were recorded during the first 20s of the second minute for each saddle height trial. One minute of static rest was allowed between trials with different saddle

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

192 Official Journal of FIMS (International Federation of Sports Medicine)

heights. The order of trials using the high, low and optimal saddle heights were randomly defined for each cyclist.

Force applied to the right pedal and right lower limb kinematics was recorded for the first 20s of the second minute during all saddle height conditions. As landmarks for the hip, knee and ankle joint axes, reflective markers were placed on the right side of the cyclists at the anterior superior iliac spine, greater trochanter, lateral femoral condyle, lateral malleolus, anterior and posterior pedal stick (device attached to the pedal surface for affixation of reflective markers). One marker was attached to the sacrum to measure the position of the cyclists when they were evaluated at the different saddle heights. Two markers were taped to the bicycle frame and used as a reference for image calibration. A custom-made 2D right pedal dynamometer with clip-in system

19 and one high speed camera

positioned perpendicular to the motion plane (AVT PIKE F-032, Allied Vision Technologies GmbH, Germany) were synchronised by an external trigger. The pedal force system enabled normal and anterior-posterior force measurements using strain gauges with cyclists and triathletes using cycling shoes with Look

® Delta cleats (see Figure 2). Errors of

calibration of normal and anterior-posterior components were computed as average percentage differences in voltage due to calibration load in relation to the output voltage. As an example, for the normal force of the right pedal, the difference in voltage from 0 kg to 5 kg was 0.1547 V and the difference in voltage from 5 kg to 10 kg was 0.1544 V, resulting in 0.19% difference in voltage due to load application. Kinematics were recorded at 60 Hz using AVT ActiveCam viewer software (Allied Vision Technologies GmbH, Germany) and force data were recorded at 600 Hz per channel employing a 16-bit analogical to digital converter (PCI-MIO-16XE-50, National Instruments, USA) using a custom Matlab (Mathworks Inc, MA) data acquisition script.

Data analyses

Video files were digitised and automatic tracking of markers were conducted in DgeeMe software (Video4Coach, Denmark) for x-y coordinates over time. Kinematic data were smoothed with a digital second order zero lag band pass Butterworth filter with cut-off frequency optimized to reduce signal residual 20

. Kinematics of the hip, knee, and ankle joints during the pedalling movement were calculated from the smoothed x-y coordinate data. Anterior and posterior pedal stick markers

were used to quantify the pedal angle and to determine the pedal spindle virtual marker. Knee flexion angle was defined as the angle between the tibia and the forward projection line of the femur. Correction of the hip joint centre based on the average coordinate between the marker on the anterior superior iliac spine and the greater trochanter was completed

21. The average relative horizontal

position of the marker on the sacrum to the bottom dead centre was computed over time across ten pedal revolutions for the analysis of body position on the saddle at the four saddle heights.

Pedal force application was defined at the centre of the pedal spindle due to the constrained motion between the shoe-cleat and the pedal. Pedal angle in relation to the global coordinate system was calculated to convert the forces on the pedal reference system to forces in the global reference system by means of trigonometric procedures. The right lower limb was modelled as a three-segment rigid body system (thigh, shank and foot-pedal) with segment mass and centre of mass estimated according to De Leva

22.

Linear and angular velocities and accelerations of the segments and joints were computed from smoothed kinematic data by a three points derivative method

20. Conventional

inverse dynamics were conducted to calculate the net joint moments at the hip, knee and ankle

23 using adapted scripts of van den

Bogert and de Koning 24

. Quadriceps muscle and patellar tendon forces were computed from knee extensor joint moment (inverse dynamics) and patellar tendon moment arm and patellar tendon angle were taken from Herzog & Read

25. Patellofemoral compressive

force was computed as described by Bressel 26

including corrections for quadriceps-patellar tendon force ratio

27. Tibiofemoral compressive

and anterior-posterior forces were computed adapted from description of Thambyah et al.

28.

Peak patellofemoral force and tibiofemoral components (anterior and compressive) were calculated across ten consecutive crank revolutions. Knee forces and flexion angle were computed using a custom written program in Matlab (Mathworks Inc, MA) for ten consecutive crank revolutions to determine means for each cyclist. Knee flexion angles at 3 o’clock and 6 o’clock crank positions were determined from knee angle-crank angle series data after interpolation to 360 samples. Knee joint forces were normalised by workload level (in Joules) for each cyclist.

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

193 Official Journal of FIMS (International Federation of Sports Medicine)

Statistical analyses

Data were reported as means and standard deviations for each group (with and without pain) and saddle height (preferred, high, low and optimal) for the tibiofemoral (anterior and compressive) and patellofemoral (compressive) peak forces and for the knee flexion angles at 3 o’clock and 6 o’clock crank positions. Data normality distribution and sphericity were evaluated by the Shapiro-Wilk and Mauchly tests, respectively using SPSS for Windows 16.0 (SPSS, NY, USA). Data normality of tibiofemoral compressive force was corrected using a logarithm transform.

The effects of changes in saddle height and differences between cyclists with and without knee pain on the patellofemoral compressive force, anterior and compressive tibiofemoral force, and knee angles at 3 o’clock and 6 o’clock crank positions were evaluated using Cohen‘s effect sizes (ES) computed as differences in mean values standardised by the pooled standard deviation. Confidence limits were computed for p<0.05 using within group standard deviations. We understand that inferential statistical for a small sample size as usually observed in studies on cycling biomechanics (e.g. 8 cyclists in our pain group) would be prone to errors. Low power of tests would preclude the extrapolation of the results to a wider population using inferential statistics. Therefore, we preferred using a more comprehensive approach (effect sizes) setting a “threshold” of ES = 1.0 (large effects) for substantial changes. This is a more conservative approach than previously

described 29

, but it would secure that the overlapping in mean scores would be greater than 55% (see Cohen

30 pgs. 21-23). Effect

sizes were rated as trivial (<0.25), small (0.25-0.49), moderate (0.5-1.0), and large (>1.0)

31.

Results

Knee flexion angles at the 3 o’clock and 6 o’clock crank positions derived from cycling motion were largely different amongst the four saddle heights with lower saddle heights leading to greater knee flexion angles. There were no differences in saddle heights comparing cyclists with and without knee pain. There were no differences between cyclists with and without knee pain from changes in saddle height for patellofemoral and tibiofemoral compressive forces (see Tables 2 and 3).

Anterior tibiofemoral peak force was larger at high and optimal saddle heights compared to the low saddle height for cyclists with and without pain. Large changes in knee flexion angle occurred for the 3 o’clock and 6 o’clock crank positions for both groups when saddle height was modified. Changes in height of the saddle were 3-6% for cyclists without pain and

2-7% for cyclists with knee pain to elicit ±10 of change in knee flexion angle measured statically with the crank at the 6 o’clock position in this present study. No differences between cyclists with and without knee pain have been found for any of the variables (see Tables 2 and 3).

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199. Available at URL: http://www.ismj.com

194 Official Journal of FIMS (International Federation of Sports Medicine)

Table 2: Mean (SD) results and ±95% confidence interval of saddle height, knee flexion angle at 3 o’clock and 6 o’clock crank positions, peak patellofemoral compressive force, anterior and compressive tibiofemoral forces presented for four saddle heights (preferred, high, low and optimal) for cyclists with and without knee pain.

Pain (n = 8) No pain (n = 16)

Optimal High Preferred Low Optimal High Preferred Low

Saddle height

(cm)

89 (4.7);

±2.33

88 (6.8);

±2.04

85 (4.7);

±1.95

84 (3.8);

±2.21

89 (4.7);

±3.31

89 (4.1);

±4.75

87 (3.9);

±3.28

84 (4.5);

±2.64

Knee flexion angle

3 o’clock crank position ()

47 (6)

±2.61

50 (5);

±2.41

56 (7);

±1,73

59 (7);

±2.81

48 (5);

±5.32

48 (6);

±3.24

52 (3);

±4.82

58 (6);

±4.75

Knee flexion angle

6 o’clock crank position ()

30 (7);

±3.29

35 (8);

±2.86

41 (9);

±2.26

47 (6);

±2.82

32 (7);

±5.09

31 (6) ;

±5.44

40 (5);

±6.94

45 (6);

±4.44

Patellofemoral compressive force

(%Workload)

1616 (521);

±255

1639 (381);

±187

1713 (331);

±162

1954 (375);

±184

1812 (861);

±597

1736 (657);

±456

1864 (784);

±543

1944 (882);

±611

Tibiofemoral anterior force

(%Workload)

75 (23);

±11

69 (27);

±13

86 (26);

±13

49 (25);

±12

71 (21);

±14

66 (27);

±18

56 (31);

±22

35 (22);

±15

Tibiofemoral compressive force

(%Workload)

995 (200);

±98

1059 (311);

±152

1031 (203);

±100

1055 (167);

±82

1111 (443);

±307

1072 (364);

±253

1031 (203);

±264

1055 (167);

±282

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199. Available at URL: http://www.ismj.com

195 Official Journal of FIMS (International Federation of Sports Medicine)

Table 3: Differences between cyclists with and without pain, and differences between saddle heights within a group, are reported as mean difference percentages along with effect size magnitudes and confidence intervals.

O vs. H O vs. P O vs. L H vs. P H vs. L P vs. L Opt High Prefered Low

Saddle height

Pain 4%; 0.1 (±1.0), T

3%; 0.1 (±1.1), T

7%; 0.1 (±1.2), T

7%; 0.2 (±1.0), T

11%; 0.3 (±1.0), S

4%; 0.1 (±1.0), T

Pain vs. no

pain

11%; 0.1 (±0.8) , T

6%; 0.1 (±0.8), T

8%; 0.1 (±0.8), T

1%; 0.3 (±0.8), S No

pain 1%; 0.1 (±0.7), T

6%; 0.2 (±0.8), T

21%; 0.8 (±1.0), M

4%; 0.2 (±1.1), T

16%; 0.8 (±1.4), M

12%; 0.7 (±0.8), M

Knee flexion angle

3 o’clock crank position

Pain 6%; 0.5 (±1.0), M

20%; 1.4 (±1.1), L

26%; 1.9 (±1.1), L

13%; 1.1 (±1.1), L

16%; 1.6 (±1.1), L

5%; 0.4 (±1.0), S

Pain vs. no

pain

3%; 0.1 (±0.8), T

5%; 0.3 (±0.8), S

6%; 0.4 (±0.9), S

1%; 0.1 (±0.8), T No

pain 4%; 0.4 (±0.7), S

9%; 0.9 (±0.7), M

19%; 1.7 (±0.8), L

13%; 1.5 (±0.8), L

20%; 2.2 (±0.9), L

9%; 1.5 (±0.8), L

Knee flexion angle

6 o’clock crank position

Pain 1%; 0.1 (±1.0), T

24%; 1.3 (±1.0), L

40%; 2.0 (±1.3), L

17%; 0.7 (±1.0), M

26%; 1.7 (±1.1), L

13%; 0.8 (±1.0), M

Pain vs. no

pain

7%; 0.2 (±0.8), S

10%;0.3 (±0.8), S

1%; 0.1 (±0.8), T

6%; 0.3 (±0.8), S No

pain 1%; 0.1 (±0.7), T

25%; 1.4 (±0.8), L

40%; 2.1 (±0.9), L

25%; 1.6 (±0.8), L

30%; 2.3 (±0.9), L

11%; 0.9 (±0.7), M

Patellofemoral compressive force

Pain 4%; 0.1 (±1.0), T

3%; 0.1 (±1.0), T

7%; 0.1 (±1.0), T

7%; 0.2 (±1.0), T

11%; 0.3 (±1.0), S

4%; 0.1 (±1.0), T

Pain vs. no

pain

11%; 0.1 (±1.0), T

6%; 0.1 (±1.0), T

8%; 0.1 (±1.0), T

1%; 0.3 (±1.0), S No

pain 1%; 0.1 (±0.7), T

6%; 0.2 (±0.7), T

21%; 0.8 (±0.7), M

4%; 0.2 (±0.7), T

16%; 0.8 (±0.7), M

12%; 0.7 (±0.7), M

Tibiofemoral anterior force

Pain 7%; 0.2 (±1.0), T

21%; 0.6 (±1.0), M

51%; 1.7 (±1.1), L

15%; 0.3 (±1.0), S

92%; 1.3 (±1.0), L

62%; 0.8 (±1.0), M

Pain vs. no

pain

6%; 0.4 (±1.0), S

30%;0.5 (±1.0), M

22%; 0.4 (±1.0), S

41%; 0.6 (±1.0), M No

pain 14%; 0.4 (±0.7), S

9%; 0.3 (±0.7), S

35%; 1.1 (±0.8), L

20%; 0.7 (±0.7), M

76%; 1.4 (±0.8), L

40%; 0.7 (±0.7), M

Tibiofemoral compressive force

Pain 3%; 0.1 (±1.0), T

1%; 0.1 (±1.0), T

2%; 0.1 (±1.0), T

3%; 0.1 (±1.0), T

6%; 0.2 (±1.0), T

3%; 0.1 (±1.0), T

Pain vs. no

pain

10%; 0.3 (±1.0), S

1%; 0.1 (±1.0), T

7%; 0.2 (±1.0), S

7%; 0.1 (±1.0), T No

pain 6%; 0.2 (±0.7), T

4%; 0.2 (±0.7), T

6%; 0.3 (±0.7), S

6%; 0.1 (±0.7), T

1%; 0.1 (±0.7), T

2%; 0.1 (±0.7), T

Abbreviations used for comparisons are optimal saddle height (O), preferred saddle height (P), high saddle height (H), low saddle height (L) and effect sizes of trivial (T), small (S), moderate (M) and large (L).

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

196 Official Journal of FIMS (International Federation of Sports Medicine)

Discussion

In this study, patellofemoral compressive, anterior and compressive tibiofemoral force, along with the knee flexion angle at two positions of the crank (3 o’clock and 6 o’clock) in cyclists with and without knee pain using four different heights of the saddle (preferred, low, high and optimal) were compared. Comparisons regarding the advocated optimal

saddle height that resulted in 25 of knee flexion angle when the crank is at the 6 o’clock crank position were conducted. Changes in saddle height from the low to the high and optimal heights were large, resulting in changes in anterior tibiofemoral force and knee flexion angles. Knee flexion angles were also reduced in cyclists who preferred the optimal saddle height and increased towards lower saddle heights (no pain group only).

Lower saddle heights resulted in greater knee flexion angle in this study, which is in agreement with previous studies

4, 5, 8. Knee

flexion angle was similarly affected by changes in saddle height for cyclists with knee pain compared to pain-free cyclists. One reason for similarities between cyclists with and without knee pain in this study may have been that they were capable of covering similar mileage during weekly cycling. Therefore the level of pain during their commuting cycling was not substantial enough to limit their practice at a sub-maximal effort level. Another reason may be related to the similar saddle height configuration, which may result in similar knee forces. Saddle height and knee flexion angles were not different between cyclists with and without pain at their preferred saddle height which indicates that similar knee forces may have been affecting both groups.

Although large changes were observed in the knee flexion angle, only anterior tibiofemoral force was largely affected by changes in saddle height (smaller force for low height compared to high saddle height) for cyclists with and without knee pain. These findings indicate that changes in the knee flexion angle from various combinations of saddle heights were not fully translated into large changes in patellofemoral and tibiofemoral compressive force. Previous studies reported either an increase

7 or no change in patellofemoral

compressive force 8 when saddle height was

reduced by 10% and 3% respectively, for non-athletes without knee pain. It may be expected that patellofemoral compressive force is only largely affected with greater changes in saddle height than the ones conducted in this study.

Cyclists are advised to configure the saddle

height to elicit 25-30 of knee flexion when the crank is static at the 6 o’clock position

18. This

configuration has been empirically chosen with reports of optimal cycling efficiency

32.

However, no study to date has assessed if this configuration of saddle height would reduce the patellofemoral compressive force compared to other configurations used by cyclists. Cyclists may not be able to replicate statically (e.g. knee flexion angle at the 6 o’clock crank position) similar joint angles as taken dynamically

33. The cyclists without

knee pain in this study presented greater knee flexion angles dynamically compared to the

results measured statically (i.e. 25) at the 6 o’clock crank position at the optimal saddle

height (~32 ±7 - no pain group; 31 ±10 - pain group). Therefore a statically measured knee flexion angle at the 6 o’clock crank position should not be used to configure saddle height. Among potential reasons these authors can highlight are that at static poses, cyclists have no angular momentum to drive their cranks. Therefore muscles are mostly relaxed and joint angles are not affected by shorter muscles (e.g. triceps surae and hamstrings), which is observed during the cycling motion. Dynamic measures of joint angles should provide a more accurate assessment of the effects of saddle height configurations for body position on the bicycle. Preferably, knee flexion angle at the 3 o’clock crank position should be used because it is a position of crank revolution closer to peak knee forces

7. In terms of knee

joint flexion angle measured dynamically, the

cyclists in this study presented ~40 (±5)

(without pain) and ~41 (±9) (with pain) for the 6 o’clock crank position at their preferred saddle height, which indicates a lower saddle height than the advocated optimal

18. However,

patellofemoral compressive force was not greatly affected by changes in saddle height. Only trivial to small changes were observed for tibiofemoral compressive force across the different saddle heights. The use of dynamic measures of joint angles has implications for both future research and for clinical bicycle-fitting. However, the validity and reliability of common tools used to do this in the field remain to be established.

Anterior tibiofemoral force was greater at the high (76-92% change) and optimal (35-51% change) compared to the low height for cyclists with and without knee pain. One explanation for this result is that the anterior tibiofemoral force depends on the patellar

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

197 Official Journal of FIMS (International Federation of Sports Medicine)

moment arm, which is expected to increase at the high and optimal saddle heights, using Herzog and Read’s model

25. However, the

importance of the anterior tibiofemoral force has been neglected because low levels of strain on the anterior cruciate ligament were found with in vivo measures

34. No reports of

anterior cruciate ligament injuries have been found in cyclists

1. These may explain why

cyclists with and without knee pain have not been affected by changes in saddle height in terms of anterior tibiofemoral force. When cycling is used in rehabilitation programmes following anterior cruciate ligament injury, these authors recommend that high saddle heights be avoided. This is due to the large anterior tibiofemoral forces observed when saddle height was increased in these subjects.

Some limitations may have affected the results of the present study. The authors did not control forward-backward body position on the saddle when saddle height was changed across trials. However, they measured body position on the saddle during all trials by video tracking of the position of the sacrum in relation to the bicycle frame, and observed trivial differences due to saddle height changes. Modelling joint kinetics (e.g. knee forces) without information of muscle activation and muscle anatomy (e.g. muscle moment arm) may not be as accurate as forward dynamics simulations

9. However, most

studies have used only the kinetic-kinematic model

7, 8 which enables the comparison of this

study’s results to those from other studies of similar design. Only right lower limb knee forces were computed due to technical limitations. Therefore the knee pain group may have been assessed under a higher workload level for their right lower limb. A greater number of cyclists with knee pain should be assessed in future studies with similarities in terms of severity, duration and type of knee pain.

Within group variation in this study for knee joint forces (i.e. 19% for the preferred position of cyclists without knee pain) indicates that an effect on knee forces must be much greater than the one observed in this study (1-21% for this group). Previous studies found changes in knee forces only when the workload is largely changed (see Bini and Hume

35). Previous

studies indicated that changes of a similar magnitude from the ones these authors conducted resulted in compensations from individual muscles without changes in cycling efficiency

36. The take-home message is that

neuromuscular strategies are adapted at different saddle heights in order to sustain performance. That would apply to this study’s findings with knee joint forces, which indeed is the most important finding in this article. In other words, even with large changes in joint motion, knee forces remain unchanged because workload and pedalling cadence are controlled.

Conclusion

Effects of saddle height were not meaningfully different for patellofemoral and tibiofemoral forces when comparing cyclists with and without knee pain. Compared to the low saddle height, there were large differences in anterior tibiofemoral forces at optimal (35% without pain, 51% pain) and high saddle heights (76% without pain, 92% pain). Based on this study’s results, it is likely that changes in saddle height do not result in large changes in knee joint forces, and there are no large differences in knee joint forces between cyclists with and without knee pain. However, as moderate and small changes may have clinical significance, future studies with greater subject numbers are necessary.

Acknowledgements

The first author acknowledges Capes-Brazil for his PhD scholarship. The authors acknowledge AUT University for supporting this research. Thanks to the cyclists who participated in the study.

Address for correspondence:

Dr Rodrigo Bini, Laboratório de Pesquisa do Exercício, Escola de Educação Física, Universidade Federal do Rio Grande do Sul, Rua Felizardo 750, Bairro Jardim Botânico Cep: 90690-200, Brazil Email: bini.rodrigo@gmail.com

References

1. Dettori NJ, Norvell DC. Non-traumatic bicycle injuries: A review of the literature. Sports Med 2006; 36(1):7-18.

2. Clarsen B, Krosshaug T, Bahr R. Overuse injuries in professional road cyclists. Am J Sports Med 2010; 38(12):2494-501.

3. Callaghan MJ. Lower body problems and injury in cycling. Journal of Bodywork and Movement Therapies 2005; 9(3):226-36.

4. Nordeen-Snyder KS. The effect of bicycle seat height variation upon oxygen consumption and lower limb kinematics. Med Sci Sports Exerc 1977; 9(2):113-7.

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

198 Official Journal of FIMS (International Federation of Sports Medicine)

5. Sanderson DJ, Amoroso AT. The influence of seat height on the mechanical function of the triceps surae muscles during steady-rate cycling. J Electromy Kinesiol 2009; 19(6):e465-e71.

6. Bini RR, Tamborindeguy AC, Mota CB. Effects of saddle height, pedaling cadence, and workload on joint kinetics and kinematics during cycling. J Sport Rehab 2010; 19(3):301-14.

7. Ericson MO, Nisell R. Patellofemoral joint forces during ergometric cycling. Phys Therap. 1987; 67(9):1365-9.

8. Tamborindeguy AC, Bini RR. Does saddle height affect patellofemoral and tibiofemoral forces during bicycling for rehabilitation? Journal of Bodywork and Movement Therapies. 2011; 15 (2):186-91.

9. Neptune RR, Kautz SA. Knee joint loading in forward versus backward pedaling: Implications for rehabilitation strategies. Clin Biomech. 2000; 15(7):528-35.

10. Ericson MO, Nisell R. Tibiofemoral joint forces during ergometer cycling. Am J Sports Med. 1986; 14(4):285-90.

11. McCoy RW, Gregor RJ. The effects of varying seat position on knee loads during cycling. Med Sci Sports Exerc 1989;21(S(2)):S79.

12. Peveler WW, Bishop P, Smith J, et al. Comparing methods for setting saddle height in trained cyclists. J Exerc Physiol Online. 2005; 8(1):51-5.

13. Bailey MP, Maillardet FJ, Messenger N. Kinematics of cycling in relation to anterior knee pain and patellar tendinitis. J Sports Sci 2003; 21(8):649-57.

14. Marfell-Jones M, Olds T, Stewart A, et al. International standards for anthropometric assessment. Kineanthropometry ISfAo, Ed. Potchefstroom, South Africa: ISAK; 2006.

15. Bini RR, Hume PA, Croft JL. Effects of bicycle saddle height on knee injury risk and cycling performance. Sports Med 2011; 41(6):463-76.

16. Chapman A, Vicenzino B, Blanch P, et al. Do differences in muscle recruitment between novice and elite cyclists reflect different movement patterns or less skilled muscle recruitment? J Sci Med Sport. 2009; 12(1):31-4.

17. Carpes FP, Dagnese F, Bini RR, et al. Pedaling kinematics characteristics of competitive cyclists of different disciplines. Portuguese Journal of Sports Sciences 2006; 6(1):7-14.

18. Holmes JC, Pruitt AL, Whalen NJ. Lower extremity overuse in bicycling. Clin Sports Med 1994; 13(1):187-203.

19. Candotti CT, Ribeiro J, Soares DP, et al. Effective force and economy of triathletes and cyclists. Sports Biomech 2007; 6(1):31-43.

20. Winter DA. Biomechanics and motor control of human movement. 3rd ed. New Jersey: John Wiley & Sons; 2005.

21. Neptune RR, Hull ML. Accuracy assessment of methods for determining hip movement in seated cycling. J Biomech 1995; 28(4):423-37.

22. De Leva P. Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters. J Biomech 1996; 29(9):1223-30.

23. Redfield R, Hull ML. On the relation between joint moments and pedalling rates at constant power in bicycling. J Biomech 1986; 19(4):317-29.

24. van den Bogert AJ, de Koning JJ, Eds. On optimal filtering for inverse dynamics. IXth Biennial Conference of the Canadian Society for Biomechanics; 1996; Vancouver. Canadian Society for Biomechanics.

25. Herzog W, Read LJ. Lines of action and moment arms of the major force-carrying structures crossing the human knee joint. J Anat. 1993; 182(2):213-30.

26. Bressel E. The influence of ergometer pedaling direction on peak patellofemoral joint forces. Clin Biomech 2001; 16(5):431-7.

27. Sharma A, Leszko F, Komistek RD, et al. In vivo patellofemoral forces in high flexion total knee arthroplasty. J Biomech 2008; 41(3):642-8.

28. Thambyah A, Pereira BP, Wyss U. Estimation of bone-on-bone contact forces in the tibiofemoral joint during walking. Knee 2005; 12(5):383-8.

29. Hopkins WG, Marshall SW, Batterham AM, et al. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc 2009; 41(1):3-12.

30. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Lawrence Earlbaum Associates; 1988.

31. Rhea MR. Determining the magnitude of treatment effects in strength training research through the use of the effect size. J Strength Cond Res 2004; 18(4):918-20.

32. Peveler WW. Effects of saddle height on economy in cycling. J Strength Cond Res. 2008; 22(4):1355-9.

Saddle height effects on knee forces International SportMed Journal, Vol.15 No.2, June 2014, pp.188-199.

Available at URL: http://www.ismj.com

199 Official Journal of FIMS (International Federation of Sports Medicine)

33. Farrell KC, Reisinger KD, Tillman MD. Force and repetition in cycling: Possible implications for iliotibial band friction syndrome. Knee 2003; 10(1):103-9.

34. Fleming BC, Beynnon BD, Renstrom PA, et al. The strain behavior of the anterior cruciate ligament during bicycling. An in vivo study. Am J Sports Med. 1998; 26(1):109-18.

35. Bini R, Hume P. Effects of workload and pedalling cadence on knee forces in cyclists. Sports Biomech. in press.

36. Connick MJ, Li FX. The impact of altered task mechanics on timing and duration of eccentric bi-articular muscle contractions during cycling. J Electromy Kinesiol in press.