Ir J Med Sci DOI 10.1007/s11845-015-1369-3

ORIGINAL ARTICLE

Characterisation of the patellar tendon reflex in cerebral palsy children using motion analysis Rory O’Sullivan1 • Damien Kiernan1 • Michael Walsh1 • Tim O’Brien1 Yahya Elhassan1

•

Received: 5 May 2015 / Accepted: 5 October 2015 Ó Royal Academy of Medicine in Ireland 2015

Abstract Background The patellar tendon reflex (PTR) is an important spinal reflex and an important diagnostic tool assessing neurological disturbances. Reflexes are conveniently assessed but quantifying the response can be subjective. Motion analysis is commonly used to assess gait kinematics in a variety of populations. It can be used to objectively assess the PTR with the advantage that standard technique and hammer can be used without the need for bulky apparatus or fixing the subject position. Aim To compare the PTR in 15 cerebral palsy (CP) children with age and height matched controls. Methods EMG reflex latency in the rectus femoris was assessed using a Noraxon 2400T unit. Knee movement latency, knee angular displacement and peak angular velocity were captured using the CODA mpx 30 system. Results EMG reflex latency was significantly reduced in CP compared to control limbs (13.11 versus 18.11 ms; p \ 0.01) confirming a ‘brisk’ response in this population. The kinematic data found that while knee angular displacement was significantly reduced in CP (12.82° versus 20.06°; p \ 0.01) there was no significant difference in movement latency or peak angular velocity compared to controls. Conclusions Subjective evaluation of the PTR relies mostly on change in knee angle. Using motion analysis we have confirmed a difference in this variable in CP compared to controls. We have also shown reduced reflex latency associated with a brisk reflex. Knee movement & Rory O’Sullivan [email protected] 1

The Gait Laboratory, Central Remedial Clinic, Vernon Avenue, Clontarf, Dublin 3, Ireland

latency and peak angular velocity did not differentiate CP from normal. Further examination of the knee angular response of the PTR is warranted in CP. Keywords Motion analysis
Cerebral palsy
Patellar tendon reflex
Electromyography

Introduction Spinal reflexes are an important diagnostic tool which aid in the diagnosis and localisation of neurological problems [1]. Deep tendon reflexes (DTR) such as the patellar tendon reflex (PTR) can be obtained in a normal person as an immediate muscle contraction when muscle tendon is tapped briskly [2]. Depressed and hyperactive DTRs suggest peripheral and central nervous system compromise, respectively [3] and hyperreflexia, is characteristic of many neurological diseases such as stroke, traumatic brain or spinal cord injury, and cerebral palsy [4]. Therefore, accurate assessment of the PTR and the objective differentiation of a normal response from an abnormal response is an important part of clinical assessment. A recent study in cerebral palsy has shown that tendon reflex amplitude decreased at 2 and 4 weeks following botulinum-toxin injection [5] suggesting that reflex assessment may have a role in planning and monitoring treatment of spasticity. Clinical reflex assessment is inexpensive and quick but evaluation can be subjective and qualitative both in terms of the applied stimulus and assessing the response [6]. This disadvantage has led to attempts at quantifying the stimulus or response. Studies aiming to quantify the reflex input involve automated tapping devices or instrumented reflex hammers [7–9]. However, some of these methods involve

123

Ir J Med Sci

fixing the subject in position and tendon taps from an instrumented hammer can be unpleasant [10]. More objective methods than visual assessment for the quantification of the PTR have been described. Surface electromyography (EMG) is the most commonly applied method [11, 12] which, while convenient, can be subject to difficulties in relation to electrode placement [13]. Other methods include attaching force transducers [14] or accelerometers [6] but again these require the subject to be fixed in position or have bulky experimental set-ups. Motion capture systems are commonly used to assess joint kinematics during gait. Motion analysis has been used to quantify the healthy PTR and has the advantage of being easy to use without restricting the movement as well as being valid and reliable [10, 15]. This study compares the patellar tendon reflex in cerebral palsy (CP) compared to healthy controls using a motion analysis system for the first time. The aims of the study are to first objectively describe the PTR in CP and second to objectively describe the difference compared to healthy controls to aid with clinical assessment of CP.

Methods Fifteen children with a diagnosis of CP were recruited as a presenting sample from those attending our centre for physiotherapy. The 15 participants consisted of 8 diplegic and 7 hemiplegic subjects. Participants were classified as level I, II, or III on the Gross Motor Function Classification System (GMFCS) [16]. The GMFCS distinguishes between five levels of motor function based on functional mobility. Children in level I of the GMFCS are able to walk indoors and outdoors without assistance and can perform gross motor skills such as running and jumping. Children in level III require a handheld mobility device when walking indoors and use wheeled mobility when travelling long distances. GMFCS levels IV and V are non-ambulant and so are not assessed in the gait laboratory. None of the CP participants had surgery or botulinumtoxin injection within the previous 6 months. Fifteen healthy controls were recruited from children of staff members. The groups were matched in terms of age, weight and height (Table 1). Limbs were examined

Table 1 Demographics of control and cerebral palsy participants

123

individually giving a total of 30 normal limbs versus 23 CP limbs. The study was approved by the local ethics committee. The study protocol was explained to parents/guardians via a study information leaflet and all procedures were verbally explained to participants. Informed written consent was obtained from all parents/guardians and informed assent from participants prior to testing. Each participant sat at the edge of the plinth with both legs hanging freely (Fig. 1). Movement data were captured using the CODA mpx 30 active marker system. Markers were attached to the greater trochanter, lateral knee joint line, and lateral malleolus. A further marker was attached to the hammer to measure strike acceleration. EMG data were recorded from the rectus femoris muscle using a Noraxon 2400T unit. Electrode placement and skin preparation were as per standard SENIAM guidelines [17]. The patellar tendon reflex was elicited three times on each leg by the same investigator using a standard Queens Square reflex hammer (0.113 kg). Hammer acceleration was recorded each time. As the mass of the hammer remained constant it was assumed that hammer acceleration directly correlated with strike force. The following variables were examined: 1.

2.

3.

Controls (n = 15) Mean (SD)

Reflex latency: the time between hammer strike and onset of rectus femoris activity (Fig. 2). Peak hammer acceleration was found to be at the first contact of the hammer with the patellar tendon so this point was used to identify hammer strike consistently among all subjects. Rectus femoris onset was defined from the EMG graph as the point where activity exceeded baseline activity by 2 standard deviations and this was confirmed by visual inspection of the EMG plot [18]. Movement latency: the time between hammer strike and first movement of the knee joint into extension (Fig. 2). CODA markers placed on the greater trochanter, lateral knee joint line, and lateral malleolus allowed a knee flexion/extension graph to be plotted. Using this graph, the first movement of the knee into extension was identified when the value moved 2 standard deviations from the resting knee angle and was confirmed by visual inspection of the graph. Amplitude of movement: the maximal extension of the knee in the first excursion.

CP Participants (n = 15) Mean (SD)

p value

Age (years)

11.73 (3.47)

13.15 (3.20)

0.13

Height (cm)

146.78 (29.33)

145.48 (19.60)

0.81

Weight (kg)

46.28 (20.83)

40.99 (12.65)

0.28

Gender

10 F:5 M

7 F:8 M

Ir J Med Sci

Results

Fig. 1 Experimental set-up showing participant with motion analysis markers on greater trochanter, knee and ankle with a further marker on the hammer. Surface EMG electrode on the rectus femoris

There was no significant difference in hammer strike acceleration between the control (41.74 ± 5.44 m/s2) and CP (39.54 ± 5.35 m/s2) groups (p = 0.15). Therefore, it was assumed that the force of reflex elicitation was similar. The measured response variables are summarised in Table 2. CP participants were found to have shorter reflex latency (13.11 versus 18.11 ms; U = 127, p \ 0.0001) and reduced amplitude of movement (12.82° versus 20.06°; U = 105, p \ 0.0001) compared to controls. There were no significant differences in movement latency (U = 283.5, p = 0.27) or peak angular velocity [p = 0.71, mean diff. -0.072, 95 % CI (-0.46, 0.32)] of the response between groups. There was no significant correlation between GMFCS level and reflex latency (q = 0.27, p = 0.21), movement latency (q = 0.12, p = 0.59), amplitude of movement (q \ 0.02, p = 0.99) or peak angular velocity (q = -0.15, p = 0.49) of the response. There was no significant correlation between MAS and reflex latency (q = -0.12, p = 0.59), movement latency (q = 0.13, p = 0.55), amplitude of movement (q = 0.27, p = 0.22) or peak angular velocity of the response (q = 0.23, p = 0.30).

Discussion

Fig. 2 EMG response (red line) and knee movement (blue line) after hammer strike (dashed line)

4.

Peak angular velocity: the maximum angular velocity of knee extension.

Spasticity of the lower limb was estimated using the Modified Ashworth Scale (MAS) from the quadriceps and graded from 1 to 4. Stata/1C 12.1 (Texas 77845 USA, 2011) statistical software was used for all statistical analysis. Data were tested for normality using the Shapiro–Wilk test. Groups were compared using unpaired Student’s t test when normally distributed and the Mann–Whitney U test was used otherwise. Spearman’s Rank Correlation Co-efficients were calculated to examine the influence of GMFCS level and Ashworth Score on the reflex response in the CP group.

In this study we used motion analysis to characterize PTR in CP children versus matched healthy controls for the first time. A previous study has compared the response in adult normal compared to spastic limbs using an accelerometer but none of these were CP [6]. The PTR would be expected to be ‘brisk’ in cerebral palsy. This is confirmed by the significantly shorter EMG latencies in CP compared to control. Our control latency (18.11 ms) compares to previously published values of 13.0–18.0 ms from age 3 to 18 years [9]. A longer latency of 21 ms was reported but that was in a taller adult population, and both age and height were shown to influence reflex latency [11]. However, there was no significant difference in movement latencies in CP compared to control. Shorter movement latency has been reported in spastic participants compared to control but the height of the six spastic participants was also significantly less than the height of the six controls (174.5 and 166.5 m, respectively) [6]. The PTR has been shown to have a significant correlation with height [11, 19] suggesting that control data should be height matched as was done in our study.

123

Ir J Med Sci Table 2 Reflex response in control and CP participants

Control (n = 30 limbs) Mean (SD)

CP (n = 23 limbs) Mean (SD)

p value

Reflex latency (ms)

18.11 (5.44)

13.11 (3.24)*

\0.0001a

Movement latency (ms)

43.96 (11.40)

41.64 (19.31)

0.27a

Amplitude of movement (°)

20.06 (4.62)

12.82 (6.83)*

\0.0001a

Peak angular velocity (°/s)

119.75 (30.37)

123.76 (51.00)

0.71b

* Statistically significant differences p \ 0.05 a

Mann–Whitney U test

b

Unpaired Student’s t test

Our reported movement latencies in control and CP participants (43.96 and 41.64 ms, respectively) were slightly longer than those previously reported in normal and spastic participants (30.8 and 28.9 ms) [6]. However, they measured the response using a tri-axial accelerometer and defined the hammer strike time as the peak strike force time rather than time of first strike as was done in this study. This highlights the importance of establishing centre-specific normative data and ensuring consistency in motion analysis set-up if using this technology to measure the PTR. CP limbs demonstrated significantly reduced amplitude of movement compared to controls (12.82° and 20.06°). This is a significant finding for clinical assessment of the reflex to differentiate CP from normal. A study on subjective evaluation of the normal PTR found that raters relied most on change in knee angle when assessing the response [1]. Our study confirms that there is a significant difference in this variable in CP compared to normal. The reduced amplitude of movement in CP is most likely related to spasticity in the lower limb. This was also associated with a delayed return to the resting position. However, there was no correlation between either GMFCS level (q \ 0.01) or MAS (q = 0.02) and the amplitude of movement suggesting that increased spasticity or level of involvement did not result in a further decrease in amplitude of movement. The study numbers were small and further work is recommended to examine how the degree of involvement affects PTR response in CP. It would have been interesting to examine the pendulum test [20] which may correlate better with PTR response and further work may be warranted to examine the relationship between these two tests. Recent work suggests that the pendulum test can be used to discriminate between typically developing children and children with CP, as well as between various degrees of spasticity, such as spastic hemiplegia and spastic diplegia [21]. The PTR in CP has been described as ‘brisk’ but despite this we found no significant difference in the peak angular velocity of the response. Again, this finding differs from a previous study [6] which found significantly higher peak

123

angular velocity in adult spastic limbs compared to controls but as noted previously the populations were not matched for height and the spastic participants were not CP. PTR response has also been shown to vary with tapping force [15] and to a lesser extent with age [11]. Our subjects and controls were age matched and while we did not directly measure tapping force, subsequent analysis found no significant difference in average hammer acceleration at the point of impact between groups. As the mass of the hammer was constant, it is therefore assumed that the tapping force was not significantly different between groups. This study reports the patellar tendon response in CP children compared to matched controls using motion analysis for the first time. EMG latency was consistent with previously reported values and confirmed a ‘brisk’ response in CP. However, the kinematic data found that there was no difference between CP and control limbs in terms of movement latency or peak angular velocity of movement. The amplitude of movement was significantly less in CP. It has previously been shown that subjective evaluation of the PTR relies mostly on change in knee angle and our study confirms a significant difference in this variable in CP compared to normal. This suggests that when clinically assessing the PTR the amplitude of the response should be visually assessed rather than seeking to assess the ‘velocity of the response. Compliance with ethical standards Conflict of interest of the authors.

No conflict of interest has been declared by any

Ethical standard Ethical approval was obtained from the local ethics committee and informed consent was obtained from all participants.

References 1. Dafkin C, Green A, Kerr S, Veliotes D, McKinon W (2013) The accuracy of subjective clinical assessments of the patellar reflex. Muscle Nerve 47(1):81–88 (PubMed PMID: 23169260. Epub 2012/11/22. eng.)

Ir J Med Sci 2. Walker HK (1990) Deep tendon reflexes. In: Walker HK, Hall WD, Hurst JW (eds) Clinical methods: the history, physical and laboratory examinations. Butterworth Publisher, Boston, pp 365–368 3. Marshall GL, Little JW (2002) Deep tendon reflexes: a study of quantitative methods. J Spinal Cord Med Summer 25(2):94–99 (PubMed PMID: 12137223. Epub 2002/07/26. eng.) 4. Lebiedowska MK, Sikdar S, Eranki A, Garmirian L (2011) Knee joint angular velocities and accelerations during the patellar tendon jerk. J Neurosci Methods 198(2):255–259 (PubMed PMID: 21549755. Epub 2011/05/10. eng.) 5. Jang DH, Sung IY, Kang YJ (2013) Usefulness of the tendon reflex for assessing spasticity after botulinum toxin-a injection in children with cerebral palsy. J Child Neurol 28(1):21–26 (PubMed PMID: 22832767. Epub 2012/07/27. eng.) 6. Mamizuka N, Sakane M, Kaneoka K, Hori N, Ochiai N (2007) Kinematic quantitation of the patellar tendon reflex using a triaxial accelerometer. J Biomech 40(9):2107–2111 (PubMed PMID: 17140585. Epub 2006/12/05. eng.) 7. Simons DG, Lamonte RJ (1971) Automated system for the measurement of reflex responses to patellar tendon tap in man. American journal of physical medicine 50(2):72–79 (PubMed PMID: 5580262. Epub 1971/04/01. eng.) 8. Van de Crommert HW, Faist M, Berger W, Duysens J (1996) Biceps femoris tendon jerk reflexes are enhanced at the end of the swing phase in humans. Brain Res 734(1–2):341–344 (PubMed PMID: 8896845. Epub 1996/09/23. eng.) 9. Pereon Y, Nguyen The Tich S, Fournier E, Genet R, Guiheneuc P (2004) Electrophysiological recording of deep tendon reflexes: normative data in children and in adults. Neurophysiol Clin Clin Neurophysiol 34(3–4):131–139 (PubMed PMID: 15501682. Epub 2004/10/27. eng.) 10. Tham LK, Abu Osman NA, Wan Abas WA, Lim KS (2013) The validity and reliability of motion analysis in patellar tendon reflex assessment. PLoS ONE 8(2):e55702 (PubMed PMID: 23409022. Pubmed Central PMCID: PMC3567131. Epub 2013/02/15. eng.) 11. Frijns CJ, Laman DM, van Duijn MA, van Duijn H (1997) Normal values of patellar and ankle tendon reflex latencies. Clin Neurol Neurosurg 99(1):31–36 (PubMed PMID: 9107465. Epub 1997/02/01. eng.) 12. Stam J, van Crevel H (1989) Measurement of tendon reflexes by surface electromyography in normal subjects. J Neurol

13.

14.

15.

16.

17.

18.

19.

20.

21.

236(4):231–237 (PubMed PMID: 2760635. Epub 1989/05/01. eng.) Stam J, van Crevel H (1990) Reliability of the clinical and electromyographic examination of tendon reflexes. J Neurol 237(7):427–431 (PubMed PMID: 2273412. Epub 1990/11/01. eng.) Lebiedowska MK, Fisk JR (2003) Quantitative evaluation of reflex and voluntary activity in children with spasticity. Archiv Phys Med Rehabil 84(6):828–837 (PubMed PMID: 12808534. Epub 2003/06/17. eng.) Tham LK, Osman NA, Lim KS, Pingguan-Murphy B, Abas WA, Zain NM (2011) Investigation to predict patellar tendon reflex using motion analysis technique. Med Eng Phys 33(4):407–410 (PubMed PMID: 21146440. Epub 2010/12/15. eng.) Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B (1997) Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 39(4):214–223 (PubMed PMID: 9183258. Epub 1997/04/01. eng.) SENIAM. Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles: Biomedical Health and Research Program (BIOMED II) of the European Union. http://www.seniam. org/ Schillings AM, Van Wezel BM, Mulder T, Duysens J (1999) Widespread short-latency stretch reflexes and their modulation during stumbling over obstacles. Brain Res 816(2):480–486 (PubMed PMID: 9878872. Epub 1999/01/08. eng.) Kuruoglu R, Oh SJ (1993) Quantitation of tendon reflexes in normal volunteers. Electromyogr Clin Neurophysiol 33(6):347–351 (PubMed PMID: 8223334. Epub 1993/09/01. eng.) Wartenberg R (1951) Pendulousness of the legs as a diagnostic test. Neurology 1(1):18–24 (PubMed PMID: 14785753. Epub 1951/01/01. eng.) Szopa A, Domagalska-Szopa M, Kidon Z, Syczewska M (2014) Quadriceps femoris spasticity in children with cerebral palsy: measurement with the pendulum test and relationship with gait abnormalities. Journal of neuroengineering and rehabilitation 11:166 (PubMed PMID: 25516151. Pubmed Central PMCID: PMC4277843. Epub 2014/12/18. eng.)

123