157

Adapted Physical Activity Quarterly, 2011, 28, 157-172© 2011 Human Kinetics, Inc.

The authors are with the Biomechanics and Technical Aids Department, Hospital Nacional de Para-pléjicos, Toledo, Spain.

Relation Between Kinematic Analysis of Wheelchair Propulsion and Wheelchair

Functional Basketball Classification

Beatriz M. Crespo-Ruiz, Antonio J. Del Ama-Espinosa, and Ángel M. Gil-Agudo

National Hospital for Spinal Cord Injury, Spain

The objective was to conduct a methodological pilot study to analyze wheelchair propulsion upper limb kinematics in standard competitive play considering the functional classification of each athlete. Ten basketball players with a functional classification ranging from 1 to 4 were included in the study. Four camcorders (Kinescan-IBV) and a treadmill for wheelchairs were used. Temporal parameters were analyzed and the upper limb kinematics was obtained using ISB recom-mendations. The value of the temporal parameters such as push phase duration, the ratio of push phase/recovery phase, contact, and propulsion angle seems to reduce as the functional classification increases. A methodological protocol has been developed that allows the analysis of kinematic characteristics of wheelchair propulsion in basketball players taking into account their functional classification.

One of the most crucial issues in wheelchair sports is the functional clas-sification that allows the grouping of players with a similar level of functional capacity based on their ability of perform movements, which aims to eliminate competitive inequalities due to the greater or lesser severity of the impairment of different athletes. Wheelchair basketball is probably the most popular disabled sport. According to estimates of the IWBF (International Wheelchair Basketball Federation), approximately 30,000 people play worldwide. Wheelchair basketball has one of the best developed functional classification systems in disabled sports (IWBF, 2002). The current player classification system adopted by the IWBF was used for the first time in international competition in July 1984 at the VIIth World Paralympic Games in Aylesbury, UK (Weiss, & Curtis, 1985). Under this system, players are tested on their ability to play the game, not on their medical disability. The wheelchair basketball functional classification is based on players’ capacity in terms of playing skills—pushing, pivoting, shooting, rebounding, dribbling, passing, and catching (IWBF, 2002). Earlier studies have tried to validate the wheelchair basketball classification system using field performance analysis in elite athletes (Vanlandewijck, Spaepen, & Lysens, 1995; Vanlandewijck et al., 2003). Although

158 Crespo-Ruiz, Del Ama-Espinosa, and Gil-Agudo

the functional classification system is well established, it would be interesting to further explore the biomechanical analysis of player classes to support the clas-sification system with objective data on how playing skills are performed.

The initial purpose of most biomechanical studies in wheelchair sports was to prevent injuries related to propulsion (Boninger, Robertson, Wolff, & Cooper, 1996; Burham, Chan, Hazlett, Laskin, & Steadward, 1994). The objective of preventing sport injuries was later broadened to include optimizing athletic performance (Van-landewijck, Theisen, & Daly, 2001). The sports performance optimization question has been approached from the perspectives of ergonomics and skill proficiency. In earlier studies, biomechanical analysis techniques have been used to compare and examine the propulsion techniques of senior male, senior female, and junior male athletes and to determine the relationship between the kinematic variables and performance in an 800-m race (Goosey, Fowler, & Campbell, 1997). Kinematic patterns were later analyzed for a range of wheelchair propulsion speeds (6.0, 6.5, and 7.0 m/s) and the relation between wheelchair propulsion and pushing economy was examined (Goosey & Campbell, 1998). It was concluded that adaptations to speed changes occur initially by decreasing cycle time and increasing cycle rate and later by increasing elbow flexion (Goosey et al., 1997).

As indicated above, one of the questions that require quantitative information is classification systems. Biomechanical analysis techniques are among the most indicated to contribute scientific evidence to the validation of different methods of classification (Gil-Agudo, Del Ama-Espinosa, & Crespo-Ruiz, 2010b). Among the most significant examples are the studies by Chow et al. of the application of kinematic analysis techniques to the movement characteristics of wheelchair field events, such as shot-putting (Chow, Chae, & Crawford, 2000) and javelin throw, performed by wheelchair athletes of different functional classes (Chow, Kuenster, & Lim, 2003). In both studies, the different classes are described and characterized using kinematics data.

In wheelchair basketball, one earlier study tried to describe the characteris-tics of different wheelchair basketball classes in biomechanical terms and also examined physiologic differences (Vanlandewijck, Spaepen, & Lysens, 1994). The user-related parameter was the force applied to the pushrim at different speeds on a wheelchair ergometer. The authors found that the level of functional disability had little impact on the isometric and dynamic forces applied to the pushrim in class II and III athletes (ISMWSF wheelchair basketball classification; Vanlandewijck et al., 1994); however, it may be more appropriate to analyze the key movements on which classifiers focus when classifying players. The definition of these movements makes it relatively easy to introduce biomechanical analysis techniques. It has been reported the need for further study to determine the biomechanical characteristics of wheelchair basketball shooting so that comparisons could be made among classes and individuals (Malone, Nielsen, & Steadward, 2000).

Still, one interesting and novel use of biomechanics in wheelchair basketball is in the analysis of player classifications. Despite being generally accepted, the current wheelchair classification system is based on the observations of classifiers, which can be somewhat subjective. Objective knowledge of the characteristics of each of the classes is needed. In this field, biomechanical analysis may have much to offer (Gil-Agudo et al., 2010b). Little, if any, quantitative research has been completed to date on the mechanics of wheelchair basketball (Malone, Gervais, & Steadward,

Wheelchair Basketball Classification    159

2002). The available literature tends to be qualitative in nature, based on coaches’ opinions and subjective analyses; however, an understanding of the mechanism of the movements in wheelchair basketball is essential. A further distinction must be made regarding objective differences in the mechanics of movement in each player classification group (Malone et al., 2002).

Up to date, biomechanical characteristics of manual wheelchair propulsion in wheelchair basketball players have not been studied in depth. We undertook this pilot study to explore the characteristics of wheelchair propulsion kinematics among wheelchair basketball players with different levels of disability by kine-matic analysis of manual wheelchair propulsion. The objective of this study was to conduct a methodological pilot study to analyze wheelchair propulsion upper limb kinematics in standard competitive play considering the functional classification of each athlete.

Method

Participants

Ten male wheelchair basketball players (age 26.9 ± 6.6 years) participated in this study. All were classified by European IWBF classifiers and were considered elite players because they had participated in international events. According to current classification system (IWBF, 2002), the results of the sample analyzed were the fol-lowing: 3 players in class 1, 3 in class 2, 2 in class 3, and 2 in class 4 (Table 1). All participants signed an informed consent form before the study began. Declaration

Table 1 Characteristics of Study Subjects in a Kinematic Analysis of Player Classifications

ParticipantsFunctional

ClassAge

(Years) InjuryIndividual Speed

Registered (Km/h)1 1 33 SCI D4 Complete 112 1 39 SCI D5 Complete 8.73 1 33 SCI D5 Complete 94 2 19 SCI D10 Complete 12.25 2 29 SCI L1 Complete 16.16 2 24 SCI D9 Complete 11.17 3 25 Right hip

disarticulation 158 3 19 SCI D8 Incomplete 13.19 4 27 Double, above–knee,

lower–limb amputee 11.510 4 21 Congenital lower–

limb malformation 12.4

160 Crespo-Ruiz, Del Ama-Espinosa, and Gil-Agudo

of Helsinki guidelines were followed in every case. We certify that all applicable institutional and government regulations concerning the ethical use of human volunteers were followed during the course of our research.

Instrumentation

Kinematics.  The right upper limb was used for kinematic analysis. Reflective markers were positioned following ISB recommendations (Wu et al., 2005) to define local reference systems on the hand, forearm, and arm (Figure 1). Twenty-two reflective markers were positioned on the following bony landmarks: 3 on the trunk (C7 spinous process, right acromio-clavicular joint, and left acromio-clavicular joint), 4 on the arm, 2 on the forearm, 3 on the hand, 4 on the wheel-chair, and 2 clusters of three markers each on the upper arm and on the lower forearm (Figure 1; Gil-Agudo, Del Ama-Espinosa, Pérez-Rizo, Pérez-Nombela, & Rodríguez-Rodríguez, 2010a). An upper-limb model was defined following ISB recommendations for local reference systems on the hand, forearm, and arm. The axes of this reference system have been described in detail elsewhere (Gil-Agudo, Del Ama-Espinosa, Pérez-Rizo, Pérez-Nombela, & Crespo-Ruiz, 2009). Three-dimensional markers positions were collected at 50 Hz with four camcorders, and the video recordings were digitalized using adequate software (Kinescan–IBV, Instituto de Biomecánica de Valencia, Valencia, Spain). Previously, a reference system was calibrated that included the entire surface of the treadmill and sufficient

Figure 1 — Upper limb marker placement: 3 on the trunk (C7 spinous process, right acro-mioclavicular joint, and left acromioclavicular joint), 4 on the arm, 2 on the forearm, 3 on the hand, and 2 clusters of three markers each on the upper arm and on the lower forearm.

Wheelchair Basketball Classification    161

height to allow trunk markers to be recorded. Spatial marker coordinates were smoothed out using a mobile means procedure (Gil-Agudo et al., 2009).

Data  Compilation.  A treadmill of suitable dimensions for wheelchairs was used to simulate realistic propulsion conditions (Bonte Zwolle B.V., BO Sys-tems, GTR-2.50). Each athlete was analyzed in the usual standard competitive play for the player’s classification with regard to wheelchair configuration and strapping because these are the circumstances in which the player’s functional class is valid. Any change in wheelchair configuration or strapping may change this classification. The subject then was instructed to propel the wheelchair on the treadmill at a comfortable speed for a two-minute adaptation period. After this period, a 1-s static video capture was made to calibrate the marker model. Participants propelled their wheelchair during a 1 min-test on the treadmill with no ramp. The treadmill speed for each player was the mean reached previously in three 20-m sprints on the playing court. We conducted propulsion trials on the treadmill with a safety system. A spotter at the front of the treadmill controlled the safety tether.

Data Simplification.  Data were collected in the middle 20-s interval to avoid the effect of acceleration and braking. Five cycles (a cycle begins when the hand contact the handrim and lasts until it returns to contact again, so the movement includes both push and recovery phase) were selected from the 20-s data recording and each cycle was analyzed separately to obtain the mean value of the 5 cycles. The cycles then were normalized from 0% to 100% because the time spent in each cycle varied across individuals and across cycles. Five events in the cycle have been identified for systematizing the description of the different articular events (Davis & Growney, 1998). Video images were taken into account to identify the start/finish of the push phase. Different propulsion events and angles were identified using a marker on the right wheel axis and a marker on the metacarpo-phalangeal joint of the third right finger (Davis & Growney, 1998). The angles that characterize the propulsion cycle (contact angle, release angle, and propul-sion angle) are shown in Figure 2. All necessary equations and calculations were processed with Matlab (The Mathworks Inc, Natick, MA, USA).

The temporal variables and upper limb kinematics were analyzed focusing on the maximum and minimum values of each articular movement and its range of motion (ROM).

Statistical Analysis

For each variable, descriptive analyses (mean ± SD) were computed for each class. Analyses were made with SPSS 12.0 (SPSS Inc., Chicago, IL, USA). The value of each variable is presented in numerical format in both the tables and graphs. Due to the small group size, only descriptive analyses have been made.

ResultsThe individual speed registered is presented in Table 1. The descriptive results of the temporal variables are listed in Table 2. The kinematic variables for each class are listed for the shoulder, elbow and wrist (Table 3). Results were not compared between classes due to the small number of participants in each class. A visual

162

Figure 2 — Description of biomechanical wheelchair propulsion variables: contact angle, release angle and propulsion angle.

Table 2 Biomechanical Temporal Parameters of Wheelchair Propulsion

Temporal Parameter Class 1 Class 2 Class 3 Class 4Speed (Km/h) 10.97 ± 1.97 11.30 ± 2.33 12.10 ± 1.05 13.43 ± 2.46Cadence (strokes/s) 1.60 ± 0.34 1.69 ± 0.30 1.90 ± 0.42 1.80 ± 0.85Duration of push phase (s) 0.21 ± 0.03 0.15 ± 0.05 0.14 ± 0.02 0.09 ± 0.02Duration of recovery phase (s) 0.39 ± 0.12 0.37 ± 0.12 0.41 ± 0.10 0.54 ± 0.29Push phase/Recovery phase 0.59 ± 0.23 0.40 ± 0.08 0.34 ± 0.04 0.19 ± 0.06Contact angle (°) 119.12 ± 2.71 107.27 ± 14.09 101.72 ± 0.93 81.40 ± 6.76Release angle (°) 19.96 ± 15.28 13.95 ± 4.17 7.84 ± 4.05 20.16 ± 14.23Propulsion angle (°) 99.16 ± 14.35 93.32 ± 11.53 93.88 ± 3.12 61.24 ± 7.47

Note. Statistical Data Group Variables (Mean ± SD)

    163

Table 3 Upper Limb Kinematics

Kinematic Variables Class 1 Class 2 Class 3 Class 4Shoulder: Maximum elevation –26.88 ± 18.49 –26.03 ± 3.04 –25.91 ± 15.63 –26.49 ± 15.63Shoulder: Minimum elevation –56.13 ± 6.73 –67.73 ± 3.88 –63.09 ± 10.10 –69.07 ± 4.77Shoulder: ROM elevation 29.25 ± 23.14 41.70 ± 5.27 37.18 ± 25.73 42.58 ± 10.86Shoulder: Maximum internal rotation 40.60 ± 8.61 22.37 ± 28.54 24.53 ± 34.62 63.34 ± 25.86Shoulder: Minimum internal rotation –65.41 ± 23.95 –69.20 ± 16.52 –90.68 ± 29.48 –27.86 ± 35.60Shoulder: ROM internal rotation 106.01 ± 29.46 91.57 ± 40.30 115.22 ± 5.13 91.19 ± 9.74Shoulder: Maximum plane of elevation 51.01 ± 13.62 60.37 ± 5.22 60.20 ± 3.38 13.37 ± 65.68Shoulder: Minimum plane of elevation –46.08 ± 11.60 –23.93 ± 29.12 –24.52 ± 12.49 –52.36 ± 52.55Shoulder: ROM plane of elevation 97.10 ± 24.44 84.30 ± 28.11 84.72 ± 9.11 65.74 ± 13.13Elbow: Maximum flexion 95.34 ± 12.90 100.52 ± 2.79 98.52 ± 0.89 94.20 ± 4.59Elbow: Minimum flexion 30.04 ± 6.43 21.87 ± 8.07 19.40 ± 9.20 21.74 ± 8.23Elbow: ROM flexoextension 65.30 ± 18.28 78.65 ± 6.97 79.12 ± 10.09 72.46 ± 12.83Elbow: maximum pronation 141.68 ± 15.47 132.64 ± 6.14 164.53 ± 22.36 165.02 ± 45.27Elbow: minimum pronation 103.79 ± 8.21 92.14 ± 2.33 106.61 ± 2.5 107.11 ± 27.36Elbow: ROM pronation 37.89 ± 9.25 40.49 ± 8.18 57.92 ± 24.86 57.92 ± 17.91Wrist: maximum ulnar deviation 20.16 ± 0.91 26.19 ± 3.17 22.41 ± 1.01 29.42 ± 7.74Wrist: maximum radial deviation 23.08 ± 4.98 16.22 ± 4.04 0.94 ± 5.91 –7.64 ± 11.94Wrist: ROM ulnar-radial 4.64 ± 2.01 9.98 ± 3.65 21.46 ± 4.9 37.06 ± 19.69Wrist: maximum flexion 29.77 ± 13.3 12.31 ± 14.92 5.44 ± 12.91 –0.18 ± 9.86Wrist: maximum extension 17.55 ± 9.36 34.06 ± 6.61 32.62 ± 4.23 25.98 ± 6.97Wrist: ROM flexo-extension 18.51 ± 14.53 21.75 ± 19.5 27.18 ± 17.14 26.16 ± 2.9

Note. Group Descriptive statistics (Mean + / - Standard Deviation)

164 Crespo-Ruiz, Del Ama-Espinosa, and Gil-Agudo

inspection of Figure 3 suggests that temporal parameters such as push phase dura-tion, the ratio of push phase/recovery phase, and contact and propulsion angle appears to be reduced while the class increases (Figure 3). For the kinematic parameters the same tendency for ROM values of the shoulder elevation plane was observed and the wrist radial-ulnar deviation ROM also seems to increase with functional classification value (Figure 4–6).

DiscussionThis study was a pilot study to assess the suitability of using biomechanical analysis techniques to compare and define functional classifications in wheelchair basketball.

The decrease in several temporal parameters with functional classification was observed, such as push phase duration, the ratio of push phase/recovery phase, and contact and propulsion angle. These observations could be related to players with a low lesion level; the hand is required to be in contact less time with the pushrim for propulsion, probably due to a stronger impulse. In regard to the upper limb kinematic values, the tendency for an increase or decrease has been detected in several parameters such as ROM values of the shoulder elevation plane and the wrist radial-ulnar deviation ROM. These data were inspected to assess the relationship with functional classes. Furthermore, given the importance of trunk movement in the functional classification of wheelchair basketball, it will be necessary to include in future studies a biomechanical model that includes trunk parameters in wheelchair propulsion.

There were different values of propulsion velocity and wheelchair configura-tion parameters among players, which conditioned the respective starting posi-tion and execution of players. Both points probably had an impact on the results. However, our aim was to work with the current functional classification system so each player was examined in his respective standard competitive play. The stan-dard competitive play is defined by the player’s functional capacity and physical interaction with the wheelchair, which is conditioned by the wheelchair design and strapping. Any change in wheelchair configuration or strapping could change the player’s assigned class. If wheelchair conditions would be standardized, we would have been analyzing the players themselves rather than their functional class, which is the product of player-wheelchair interaction. The same reasons are applicable to velocity. According to the current classification system, the player is asked to execute a sprint at maximum speed on the playing floor, because each player’s speed depends on his or her intrinsic functional capacity. The player is classified during propulsion at his or her maximum speed. Therefore, we applied the same criterion and treadmill speed was determined by calculating the mean of three previous sprints on the playing floor.

We are aware that speed is a variable that modifies kinematic findings, but if all players would perform at same propulsion velocity, functional differences would not have revealed.

Interest in the biomechanical analysis of manual wheelchair propulsion has increased as earlier studies have reported an increasingly older population of people with spinal cord injury (SCI) and a high incidence of upper limb pathology (Gell-man, Sie, & Waters, 1988; Sie, Waters, Adkins, & Gellman, 1992; Silfverskold & Waters, 1991). Biomechanical analysis of wheelchair propulsion yields pertinent

    165

Figure 3 — Kinematics parameters and wheelchair basketball player classes (Media ± SD).

166

Figure 4 — Wrist spatial-temporal parameters and wheelchair basketball player classes (Media ± SD).

    167

Figure 5 — Elbow spatial-temporal parameters and wheelchair basketball player classes (Media ± SD).

168

Figure 6 — Shoulder spatial-temporal parameters and wheelchair basketball player classes (Media ± SD).

Wheelchair Basketball Classification    169

information for identifying the factors that predispose to upper limb injuries (Gil-Agudo et al., 2010a). The objective of our work was to initiate a methodological pilot study to analyze wheelchair propulsion upper limb kinematics in standard competitive play considering the functional classification of each athlete. The study of relevant movements during wheelchair propulsion in athletes could be related to the prevention of upper limb injuries.

Wheelchair propulsion in wheelchair basketball players has been studied ear-lier (Coutts, 1990). The differences of wheelchair propulsion kinematics between basketball and track athletes on a wheelchair ergometer were compared. During the first 3 pushes, basketball players had a higher push rim velocity throughout effort but a higher wheelchair velocity only at the end of the first push (Coutts, 1990). Another key point in wheelchair basketball classification is trunk movement. Curtis et al. (1995) described a methodology for the kinematic analysis of trunk mobility in wheelchair users. This study provided strong evidence that a chest strap increases functional reach in the sagittal plane, but functional reach does not appear to be enhanced by a thigh strap in subjects with high or low thoracic paraplegia (Curtis, Kindlin, Reich, & White, 1995). It could be related to strapping in wheelchair basketball players.

In relation to biomechanical wheelchair basketball analysis, it may be more appropriate to analyze the key movements as shooting on which classifiers focus when classifying players and reproduce the standard competitive play (wheelchair configuration and strapping; Nunome, Doyo, Sakurai, Ikegmai, & Yabe, 2002).

Comparisons between studies are often difficult because of different testing procedures, units of measurement, equipment employed, and characteristics of the sample studied (Finley, Rasch, Keyser, & Rodgers, 2004). In this study, propulsion analysis was carried out using a wheelchair placed on a treadmill, which other authors have characterized as the ideal mechanical situation (Richter, Rodriguez, Woods, & Axelson, 2007). Other investigators have used dynamometers (DiGiovine, Cooper, & Boninger, 2001; Kulig et al., 1998; Newsam et al., 1999) or ergometers (Niesing, Eijskoot, Kranse, den Ouden, & Storm, 1990).

Methodological limitations can be raised to the current study because each player used his own adapted wheelchair, so wheelchair parameters were not standardized. Although the four cameras were electronically synchronized, errors may have resulted from the limited resolution of the video images. Given the pilot characteristic of this work, we consider that all the variables analyzed have the potential to give relevant information. Moreover, some additional variables should and will be considered: variables related with the trunk movement.

One controversial issue may derive from functional classifications being defined in terms of playing actions and being made during actual games, whereas biomechanical analysis is conducted in the experimental laboratory setting, and we are focusing our analysis uniquely on linear wheelchair propulsion. We think that it would be interesting to produce a better classification system and to con-tinue this study with a larger sample. This study represents just an example about how to introduce biomechanical techniques to assess functional classification in wheelchair basketball. Numerous skills are not studied (e.g., shooting, dribbling, rebound, and manual propulsion). It would be necessary to develop a new set up to analyze those skills.

One of our aims was to introduce objective assessment in the current classifica-tion system but in some way we maintained this subjectivity in our data analysis

170 Crespo-Ruiz, Del Ama-Espinosa, and Gil-Agudo

because our analysis procedure is essentially an attempt to find an association between a number of variables and the functional classification level produced by (subjective) human observers. This study was a pilot experiment to assess the suit-ability of using biomechanical analysis techniques to compare and define functional classifications in wheelchair basketball but we are agreeing that is necessary to defined more specific key movements.

ConclusionClassification systems are a key element in sports for athletes with disabilities, especially wheelchair basketball. Biomechanical analysis can help to define objec-tively propulsion characteristics for each wheelchair basketball functional class. The present study is a pilot experience to describe a methodology to analyze upper limb kinematic characteristics of wheelchair propulsion in standard competitive play (speed and wheelchair configuration) and wheelchair basketball classification. The data seems to be related when temporal parameters are inspected. This may suggest that more emphasis should be placed on biomechanical analysis to assess functional classification. More quantitative data are needed for the development of a database on wheelchair propulsion characteristics and other key points of clas-sification (dribbling, braking, rebounding, etc.), especially in the play situation.

Acknowledgment

This work was part of a project financed by the Consejería de Sanidad de Castilla-La Mancha, which does not have any commercial interest in the results of this investigation (Ref: 06006-00).

ReferencesBoninger, M.L., Robertson, R.N., Wolff, M., & Cooper, R.A. (1996). Upper limb nerve

entrapments in elite wheelchair racers. American Journal of Physical Medicine & Rehabilitation, 75, 170–176.

Burham, R.S., Chan, M., Hazlett, C., Laskin, J., & Steadward, R.D. (1994). Acute median nerve dysfunction from wheelchair propulsion: The development of a model and study of the effect of hand protection. Archives of Physical Medicine and Rehabilitation, 75, 513-518.

Chow, J.W., Chae, W.S., & Crawford, M.J. (2000). Kinematic analysis of shot-putting performed by wheelchair athletes of different medical classes. Journal of Sports Sci-ences, 18, 321–330.

Chow, J.W., Kuenster, A.F., & Lim, Y. (2003). Kinematic analysis of javelin throw perfor-mance wheelchair athletes of different functional classes. Journal of Sports, Science, and Medicine, 2, 36–46.

Coutts, K.D. (1990). Kinematics of sport wheelchair propulsion. Journal of Rehabilitation Research and Development, 27, 21–26.

Curtis, K.A., Kindlin, C.M., Reich, K.M., & White, D.E. (1995). Functional reach in wheel-chair users: The effects of trunk and lower extremity stabilization. Archives of Physical Medicine and Rehabilitation, 76, 360–367.

Davis, L.A., & Growney, E.S. (1998). Three-dimensional kinematics of the shoulder complex during wheelchair propulsion: A technical report. Journal of Rehabilitation Research and Development, 35, 61–72.

Wheelchair Basketball Classification    171

DiGiovine, C.P., Cooper, R.A., & Boninger, M.L. (2001). Dynamic calibration of a wheel-chair dynamometer. Journal of Rehabilitation Research and Development, 38, 41–55.

Finley, M., Rasch, E., Keyser, R., & Rodgers, M. (2004). The biomechanics of wheelchair propulsion in individuals with and without upper-limb impairment. Journal of Reha-bilitation Research and Development, 41, 395–402.

Gellman, H., Sie, I., & Waters, R.L. (1988). Late complications of the weight-bearing upper-extremity in the paraplegic patient. Clinical Orthopaedics and Related Research, 233, 132–135.

Gil-Agudo, A., Del Ama-Espinosa, A., Pérez-Rizo, E., Pérez-Nombela, S., & Rodríguez-Rodríguez, L.P. (2010a). Upper limb joint kinetics during manual wheelchair propulsion in patients with levels of spinal cord injury. Journal of Biomechanics, 43, 2508–2515.

Gil-Agudo, A., Del Ama-Espinosa, A.J., & Crespo-Ruiz, B. (2010b). Wheelchair basketball quantification. In Saunders (Ed.), Quality of life in physical medicine and rehabilita-tion clinics of North America (pp. 141-156). Nueva York (Estados Unidos): Elservier.

Gil-Agudo, A., Del Ama-Espinosa, A.J., Pérez-Rizo, E., Pérez-Nombela, S., & Crespo-Ruiz, B. (2009). Shoulder joint kinetics during wheelchair propulsion a treadmill at two dif-ferent speeds in spinal cord injury patients. Spinal Cord, 48, 290–296.

Goosey, V.L., & Campbell, I.G. (1998). Pushing economy and propulsion technique of wheelchair racers at three speeds. Adapted Physical Activity Quarterly, 15, 36–50.

Goosey, V.L., Fowler, N.E., & Campbell, I.G. (1997). A kinematic analysis of wheelchair propulsion techniques in senior male, senior female, and junior male athletes. Adapted Physical Activity Quarterly, 14, 156–165.

IWBF. (2002). Player classification system wheelchair basketball. Internacional Wheelchair Basketball Federation. IWBF web site. Avalaible online at: www.iwbf.org/classifica-tion/.

Kulig, K., Rao, S.S., Mulroy, S.J., Newsam, C.J., Gronley, J.K., Bontrager, E.L., et al. (1998). Shoulder joint kinetics during the push phase of wheelchair propulsion. Clini-cal Orthopaedics, 354, 132–143.

Malone, L.A., Gervais, P.L., & Steadward, R.D. (2002). Shooting mechanics related to player classification and free throw success in wheelchair basketball. Journal of Rehabilitation Research and Development, 39, 701–709.

Malone, L.A., Nielsen, A.B., & Steadward, R.D. (2000). Expanding the dichotomous outcome in wheelchair basketball shooting of elite male players. Adapted Physical Activity Quarterly, 17, 437-449.

Newsam, C.J., Rao, S.S., Mulroy, S.J., Gronley, J.K., Bontrager, E.L., & Perry, J. (1999). Three dimensional upper extremity motion during manual wheelchair propulsion in men with different levels of spinal cord injury. Gait & Posture, 10, 223–232.

Niesing, R., Eijskoot, F., Kranse, R., den Ouden, A.H., & Storm, J. (1990). Computer-controlled wheelchair ergometer. Medical & Biological Engineering & Computing, 28, 329–338.

Nunome, H., Doyo, W., Sakurai, S., Ikegmai, Y., & Yabe, K. (2002). A kinematic study of the upper-limb motion of wheelchair basketball shooting in tetraplegic adults. Journal of Rehabilitation Research and Development, 39, 63–71.

Richter, W.M., Rodriguez, R., Woods, K.R., & Axelson, P.W. (2007). Stroke pattern and handrim biomechanics for level and uphill wheelchair propulsion at self-selected speeds. Archives of Physical Medicine and Rehabilitation, 88, 81–87.

Sie, I.H., Waters, R.L., Adkins, R.H., & Gellman, H. (1992). Upper extremity pain in the postrehabilitation spinal cord injury patient. Archives of Physical Medicine and Reha-bilitation, 73, 44–48.

Silfverskold, J., & Waters, R.L. (1991). Shoulder pain and functional disability in spinal cord injury patients. Clinical Orthopaedics and Related Research, 272, 141–145.

172 Crespo-Ruiz, Del Ama-Espinosa, and Gil-Agudo

Vanlandewijck, Y., Theisen, D., & Daly, D. (2001). Wheelchair propulsion biomechanics. Implications for wheelchair sports. Sports Medicine (Auckland, N.Z.), 31, 339–367.

Vanlandewijck, Y.C., Evaggelinou, C., Daly, D.D., Van Houte, S., Verellen, J., Aspeslagh, V., et al. (2003). Proportionality in wheelchair basketball classification. Adapted Physical Activity Quarterly, 20, 369–380.

Vanlandewijck, Y.C., Spaepen, A.J., & Lysens, R.J. (1994). Wheelchair propulsion: functional ability dependent factors in wheelchair basketball players. Scandinavian Journal of Rehabilitation Medicine, 26, 37–48.

Vanlandewijck, Y.C., Spaepen, A.J., & Lysens, R.J. (1995). Relationship between the level of physical impairment and sports performance in elite wheelchair basketball players. Adapted Physical Activity Quarterly, 12, 139–150.

Weiss, M., & Curtis, K. (1985). Controversies in medical classification of wheelchair athletes. In C. Sherrill (Ed.), Sport and disabled athletes. Champaign, IL: Human Kinetics.

Wu, G., van der Helm, F.C., Veeger, H.E., Makhsous, M., Van, R.P., Anglin, C., et al. (2005). ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion–Part II: Shoulder, elbow, wrist and hand. Journal of Biomechanics, 38, 981–992.