- Email: [email protected]
Development of a clinical instrument to measure heel pad indentation
Clinical Biomechanics 15 (2000) 298±300
www.elsevier.com/locate/clinbiomech
Brief Report
Development of a clinical instrument to measure heel pad indentation Keith Rome a,*, Paul Webb b a
University of Teesside, Rehabilitation Research Unit, Middlesbrough General Hospital, TS1 3BA Cleveland, UK b Regional Medical Physics Department, North Tees Hospital, Stockton-on-Tees, Cleveland, UK Received 10 February 1999; accepted 25 May 1999
Abstract Objective. To provide an accurate, reliable, non-invasive, portable instrument to measure heel pad indentation in a clinical setting. Design. A novel instrument was applied to assess heel pad indentation. Background. For the lack of quantitative biomechanical tools for in vivo assessment, palpation is used to evaluate heel pad stiness subjectively in everyday clinical practice. Furthermore, previous studies have evaluated heel pad stiness using non-portable instrumentation. Methods. Cylindrical probe was attached to an electronic force gauge unit that passed through a Perspex plate for placement on heel pad. Displacement of the plate was connected to a linear variable dierential transformer. A laptop PC allowed for portability and storage of data. An exponential curve described the force±displacement data. Ten healthy subjects (mean 21.7; SD, 1.7 years) were assessed on ten separate occasions. The procedure was standardised for subject position and placement. Results. Accuracy and reliability of each device component was established. An intraclass correlation (2,1) of 0.90 and 0.88 demonstrated curve coecients b1 and b0 respectively. A paired t-test demonstrated no signi®cant dierence between the left and right foot coecients. Conclusion. The results demonstrated that each system component could be measured accurately and reliable. Reproducible results were obtained over separate occasions. The study has described a method to analyse the force±displacement curve. Relevance The development of a hand-held device may help the clinician to assess heel pad stiness in the clinical setting. Heel pad stiness may be associated in the development of plantar heel pain in athletes. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Heel pad; Indentation; Stiness; Soft tissue mechanical properties
1. Introduction For the lack of quantitative biomechanical tools for in vivo assessment, palpation is still widely used to evaluate heel pad stiness subjectively in everyday clinical practice. A number of biomechanical instrumentation have been advocated to assess the mechanical behaviour of the heel pad [1±3]. Morag et al. [1] reported using an ultrasound-based device to measure the in vivo characteristics of soft tissue under the calcaneal tuberosity. Hsu et al. [2] reported on a linear-array ultrasound transducer ®xed in a speci®cally designed jig to measure the stiness of the heel pad in young and elderly subjects without heel problems. Robbins et al. [3] described the *
Corresponding author. E-mail address: [email protected] (K. Rome).
use of a hand-held, spring-loaded spherical indentation system to measure the deformation of the heel pad, and reported good system repeatability and accuracy. The authors did not report on the reproducibility of the procedure or the time taken between readings for soft tissue recovery. The aim of this study was, therefore, to provide an accurate, reliable, non-invasive, easy-to-use, portable clinical instrument to measure the force on a metal probe, and its displacement into the plantar surface of the heel pad for clinical use.
2. Methodology An electronic force gauge unit (Salter, Weigh-Tronix, Washington, UK), incorporated a four-strain gauge load cell giving compression output. Attached to the
0268-0033/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 8 - 0 0 3 3 ( 9 9 ) 0 0 0 4 6 - 7
K. Rome, P. Webb / Clinical Biomechanics 15 (2000) 298±300
load cell was a stainless steel, ¯at-tipped, cylindrical actuator of 6-mm diameter. The actuator passed through a hole in the front face of the testing surface, made from 3 mm transparent Perspex, to allow a good view of the placement of the probe on the plantar surface of the heel pad. The Perspex plate was displaced as the probe was applied (Fig. 1). To measure the displacement of the Perspex plate, it was in turn connected to the actuator of a linear variable dierential transformer (LVDT). To operate the transducer, it was necessary to drive the primary with a sine wave, and monitor the output from the secondary that contain the positional information in amplitude and phase. Conversion to DC voltage allows varying linearly with position, and full thick ®lm circuits were available to perform these functions, by the use of an oscillator and a demodulator. These circuits were employed in the ®nal design of the instrument. As the instrument was to be battery powered, with a supply of 9 V, an excitation voltage of 5 vrms was chosen for the oscillator. Checks that the amplitude and frequency did not change as battery voltage deceased, hence introducing error. Measurement error did not prove to be a problem due to a battery voltage of 5 V, at which point a low battery indicator warns the user to change the batteries. A unity gain-inverting ampli®er with a variable oset was used to create a true zero, and the output from this was buered to produce a 0±1.5 V output representing 0±15 mm displacement. The voltage was fed to a peak hold circuit and panel meter, so that maximum displacement could be read directly on the instrument. The electronics were housed in a separate plastic enclosure, and cables passed the signals from the LVDT and the force gauge to it. The LVDT was mounted on the back of the force gauge. The two-analogue voltages from the force gauge and LVDT buer were passed to a two-channel analogue to digital converter in the form of a data acquisition module with WinDaq software (Metrum Information Storage Ltd, Wokingham, UK). Post-processing allowed X±Y plotting of data and export of data
Fig. 1.
299
onto spreadsheets for further analysis. A laptop PC was used for portability and for storage of data. Ten healthy subjects (eight females and two males) were chosen from a student population aged between 18 and 28 years (mean 21.7, SD: 1.71) to observe the variation on ten separate occasions. The inclusion criteria were no history of major trauma to the ankle or subtalar joints, and no history of plantar heel pain in the last 12 months. Twenty measurements on both feet were taken for each subject. A 20-min recovery time between repetitions was observed. The sampling rate was recorded at 240 samples/s for a maximum of 10 s. The procedure was standardised for subject position and placement of the test instrument. Each subject was in a reclined and fully supported supine position on a padded examination couch. The operator placed the actuator at right angle to the heel pad and indented into the plantar surface of the foot while maintaining contact between the plantar skin and the LVDT actuator. Intraclass correlations (2,1) [4] were used to assess to the reliability of the instrument over time. 3. Results A number of dierent curve-®tting procedures were undertaken, and a typical force±displacement curve was well described by a exponential function of the form using a non-linear regression equation: F b0eb1d , where d was the displacement (mm), F the force (N), e the constant, b0 and b1 were the curve coecients. To assess the LVDT, twenty consecutive measurements with a calibration block of a known distance of 8.50 mm produced a mean reading of 8.52, SD: 0.047 mm. A further twenty measurements were taken (1 h duration) and the results indicated a signi®cant correlation between the two occasions (R2 0.98, P < 0.05). The electronic force gauge device was calibrated against another recently checked and certi®ed electronic force gauge from the same manufacturers, ranging from 0± 100 N. The results indicated a signi®cant correlation (R2 0.99, P < 0.05) between the two devices. These results indicate that modi®cation of the force gauge does not alter the characteristics of the basic machine. The results demonstrated a low systematic error occurred across the range values (0±0.3%). The load-cell was further calibrated over its operating range (0±100 N) using a standard set of weights, and demonstrated a signi®cant correlation (R2 0.98, P < 0.01) between the indentation system and the standard weights. Heel pad indentation results demonstrated an ICC (2.1) [4] of 0.90 for b1 and 0.88 for b0 coecient over ten repeated measures. Twenty repetitions on one subject demonstrated a coecient of variation (SD divided by the mean) of 6.2%. A paired t-test demonstrated no signi®cant dierence between the left and right foot
300
K. Rome, P. Webb / Clinical Biomechanics 15 (2000) 298±300
coecients (b0: t 0.224, p 0.828; b1: t 0.236, p 0.819). 4. Discussion In this study, the results have demonstrated that a portable, hand-held device is an accurate and reliable tool for measuring heel pad indentation. Previous work has reported that the use of hand-held probes to be unreliable and inaccurate [5]. Main errors have included poor alignment; speed and angle of loading and inappropriate instrument application. To overcome the problems of poor alignment construction of Perspex plate encompassed the actuator. This allowed sucient view of the centre area of the plantar surface of the calcaneum. To reduce the problems with the angle of the probe held in a horizontal position, consecutive measurements was repeated on the same point. The results demonstrated high reproducibility (CV 6.2%), suggesting that speed and pressure angle did not critically aect the obtained data. Motorised step-motors have been reported to overcome the problems of alignment and operator error. However, inherent errors such as motor slippage and continuation of the steeping motor before motor reversal threshold is attained, and may also limit the use of the equipment being used in a clinical setting. To demonstrate the feasibility of using a hand-held probe to collect indentation data on human subjects the procedure was standardised for subject position and placement. The study has described a
method to analyse the force±displacement curve. The use of a device may enable the clinician to assess patients in a number of clinical settings in a quick and simple manner. Currently, research is being undertaken on evaluating the clinical signi®cance of the exponential curve, and on heel pad indentation data in the athletic population with diagnosed plantar heel pain. Future research may demonstrate that those patients suering from heel pain and other reduced shock-absorbing conditions may have tissue stiness alterations due to chronic overloading or pathological processess within the tissue.
References [1] Morag E, Lemmon DR, Cavannah PR. An in-vivo method to study the properties of the human heel pad. Presented at the American Society of Biomechanics, Clemson University, SC, September, 1997. [2] Hsu TC, Wang LW, Tsai WC, Kuo JK, Tang FT. Comparison of the mechanical properties of the heel pad between young and elderly adults. Arch. Phys. Med. Rehabil. 1998;79:1101±4. [3] Robbins SE, Gouw GJ, Hama AM. Running-related injury prevention through innate impact-moderating behaviour. Med. Sci. Sports. Exerc. 1989;21:130±9. [4] Rankin G, Stokes M. Reliability of assessment tools in rehabilitation : An illustration of appropriate statistical analyses. Clin. Rehab. 1998;12:187±99. [5] Kawchuk G, Herzog W. A new technique of tissue stiness assessment: Its reliability, accuracy and comparison with an existing model. J. Manipul. Physiol. Ther. 1996;19:13±8.
www.elsevier.com/locate/clinbiomech
Brief Report
Development of a clinical instrument to measure heel pad indentation Keith Rome a,*, Paul Webb b a
University of Teesside, Rehabilitation Research Unit, Middlesbrough General Hospital, TS1 3BA Cleveland, UK b Regional Medical Physics Department, North Tees Hospital, Stockton-on-Tees, Cleveland, UK Received 10 February 1999; accepted 25 May 1999
Abstract Objective. To provide an accurate, reliable, non-invasive, portable instrument to measure heel pad indentation in a clinical setting. Design. A novel instrument was applied to assess heel pad indentation. Background. For the lack of quantitative biomechanical tools for in vivo assessment, palpation is used to evaluate heel pad stiness subjectively in everyday clinical practice. Furthermore, previous studies have evaluated heel pad stiness using non-portable instrumentation. Methods. Cylindrical probe was attached to an electronic force gauge unit that passed through a Perspex plate for placement on heel pad. Displacement of the plate was connected to a linear variable dierential transformer. A laptop PC allowed for portability and storage of data. An exponential curve described the force±displacement data. Ten healthy subjects (mean 21.7; SD, 1.7 years) were assessed on ten separate occasions. The procedure was standardised for subject position and placement. Results. Accuracy and reliability of each device component was established. An intraclass correlation (2,1) of 0.90 and 0.88 demonstrated curve coecients b1 and b0 respectively. A paired t-test demonstrated no signi®cant dierence between the left and right foot coecients. Conclusion. The results demonstrated that each system component could be measured accurately and reliable. Reproducible results were obtained over separate occasions. The study has described a method to analyse the force±displacement curve. Relevance The development of a hand-held device may help the clinician to assess heel pad stiness in the clinical setting. Heel pad stiness may be associated in the development of plantar heel pain in athletes. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Heel pad; Indentation; Stiness; Soft tissue mechanical properties
1. Introduction For the lack of quantitative biomechanical tools for in vivo assessment, palpation is still widely used to evaluate heel pad stiness subjectively in everyday clinical practice. A number of biomechanical instrumentation have been advocated to assess the mechanical behaviour of the heel pad [1±3]. Morag et al. [1] reported using an ultrasound-based device to measure the in vivo characteristics of soft tissue under the calcaneal tuberosity. Hsu et al. [2] reported on a linear-array ultrasound transducer ®xed in a speci®cally designed jig to measure the stiness of the heel pad in young and elderly subjects without heel problems. Robbins et al. [3] described the *
Corresponding author. E-mail address: [email protected] (K. Rome).
use of a hand-held, spring-loaded spherical indentation system to measure the deformation of the heel pad, and reported good system repeatability and accuracy. The authors did not report on the reproducibility of the procedure or the time taken between readings for soft tissue recovery. The aim of this study was, therefore, to provide an accurate, reliable, non-invasive, easy-to-use, portable clinical instrument to measure the force on a metal probe, and its displacement into the plantar surface of the heel pad for clinical use.
2. Methodology An electronic force gauge unit (Salter, Weigh-Tronix, Washington, UK), incorporated a four-strain gauge load cell giving compression output. Attached to the
0268-0033/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 8 - 0 0 3 3 ( 9 9 ) 0 0 0 4 6 - 7
K. Rome, P. Webb / Clinical Biomechanics 15 (2000) 298±300
load cell was a stainless steel, ¯at-tipped, cylindrical actuator of 6-mm diameter. The actuator passed through a hole in the front face of the testing surface, made from 3 mm transparent Perspex, to allow a good view of the placement of the probe on the plantar surface of the heel pad. The Perspex plate was displaced as the probe was applied (Fig. 1). To measure the displacement of the Perspex plate, it was in turn connected to the actuator of a linear variable dierential transformer (LVDT). To operate the transducer, it was necessary to drive the primary with a sine wave, and monitor the output from the secondary that contain the positional information in amplitude and phase. Conversion to DC voltage allows varying linearly with position, and full thick ®lm circuits were available to perform these functions, by the use of an oscillator and a demodulator. These circuits were employed in the ®nal design of the instrument. As the instrument was to be battery powered, with a supply of 9 V, an excitation voltage of 5 vrms was chosen for the oscillator. Checks that the amplitude and frequency did not change as battery voltage deceased, hence introducing error. Measurement error did not prove to be a problem due to a battery voltage of 5 V, at which point a low battery indicator warns the user to change the batteries. A unity gain-inverting ampli®er with a variable oset was used to create a true zero, and the output from this was buered to produce a 0±1.5 V output representing 0±15 mm displacement. The voltage was fed to a peak hold circuit and panel meter, so that maximum displacement could be read directly on the instrument. The electronics were housed in a separate plastic enclosure, and cables passed the signals from the LVDT and the force gauge to it. The LVDT was mounted on the back of the force gauge. The two-analogue voltages from the force gauge and LVDT buer were passed to a two-channel analogue to digital converter in the form of a data acquisition module with WinDaq software (Metrum Information Storage Ltd, Wokingham, UK). Post-processing allowed X±Y plotting of data and export of data
Fig. 1.
299
onto spreadsheets for further analysis. A laptop PC was used for portability and for storage of data. Ten healthy subjects (eight females and two males) were chosen from a student population aged between 18 and 28 years (mean 21.7, SD: 1.71) to observe the variation on ten separate occasions. The inclusion criteria were no history of major trauma to the ankle or subtalar joints, and no history of plantar heel pain in the last 12 months. Twenty measurements on both feet were taken for each subject. A 20-min recovery time between repetitions was observed. The sampling rate was recorded at 240 samples/s for a maximum of 10 s. The procedure was standardised for subject position and placement of the test instrument. Each subject was in a reclined and fully supported supine position on a padded examination couch. The operator placed the actuator at right angle to the heel pad and indented into the plantar surface of the foot while maintaining contact between the plantar skin and the LVDT actuator. Intraclass correlations (2,1) [4] were used to assess to the reliability of the instrument over time. 3. Results A number of dierent curve-®tting procedures were undertaken, and a typical force±displacement curve was well described by a exponential function of the form using a non-linear regression equation: F b0eb1d , where d was the displacement (mm), F the force (N), e the constant, b0 and b1 were the curve coecients. To assess the LVDT, twenty consecutive measurements with a calibration block of a known distance of 8.50 mm produced a mean reading of 8.52, SD: 0.047 mm. A further twenty measurements were taken (1 h duration) and the results indicated a signi®cant correlation between the two occasions (R2 0.98, P < 0.05). The electronic force gauge device was calibrated against another recently checked and certi®ed electronic force gauge from the same manufacturers, ranging from 0± 100 N. The results indicated a signi®cant correlation (R2 0.99, P < 0.05) between the two devices. These results indicate that modi®cation of the force gauge does not alter the characteristics of the basic machine. The results demonstrated a low systematic error occurred across the range values (0±0.3%). The load-cell was further calibrated over its operating range (0±100 N) using a standard set of weights, and demonstrated a signi®cant correlation (R2 0.98, P < 0.01) between the indentation system and the standard weights. Heel pad indentation results demonstrated an ICC (2.1) [4] of 0.90 for b1 and 0.88 for b0 coecient over ten repeated measures. Twenty repetitions on one subject demonstrated a coecient of variation (SD divided by the mean) of 6.2%. A paired t-test demonstrated no signi®cant dierence between the left and right foot
300
K. Rome, P. Webb / Clinical Biomechanics 15 (2000) 298±300
coecients (b0: t 0.224, p 0.828; b1: t 0.236, p 0.819). 4. Discussion In this study, the results have demonstrated that a portable, hand-held device is an accurate and reliable tool for measuring heel pad indentation. Previous work has reported that the use of hand-held probes to be unreliable and inaccurate [5]. Main errors have included poor alignment; speed and angle of loading and inappropriate instrument application. To overcome the problems of poor alignment construction of Perspex plate encompassed the actuator. This allowed sucient view of the centre area of the plantar surface of the calcaneum. To reduce the problems with the angle of the probe held in a horizontal position, consecutive measurements was repeated on the same point. The results demonstrated high reproducibility (CV 6.2%), suggesting that speed and pressure angle did not critically aect the obtained data. Motorised step-motors have been reported to overcome the problems of alignment and operator error. However, inherent errors such as motor slippage and continuation of the steeping motor before motor reversal threshold is attained, and may also limit the use of the equipment being used in a clinical setting. To demonstrate the feasibility of using a hand-held probe to collect indentation data on human subjects the procedure was standardised for subject position and placement. The study has described a
method to analyse the force±displacement curve. The use of a device may enable the clinician to assess patients in a number of clinical settings in a quick and simple manner. Currently, research is being undertaken on evaluating the clinical signi®cance of the exponential curve, and on heel pad indentation data in the athletic population with diagnosed plantar heel pain. Future research may demonstrate that those patients suering from heel pain and other reduced shock-absorbing conditions may have tissue stiness alterations due to chronic overloading or pathological processess within the tissue.
References [1] Morag E, Lemmon DR, Cavannah PR. An in-vivo method to study the properties of the human heel pad. Presented at the American Society of Biomechanics, Clemson University, SC, September, 1997. [2] Hsu TC, Wang LW, Tsai WC, Kuo JK, Tang FT. Comparison of the mechanical properties of the heel pad between young and elderly adults. Arch. Phys. Med. Rehabil. 1998;79:1101±4. [3] Robbins SE, Gouw GJ, Hama AM. Running-related injury prevention through innate impact-moderating behaviour. Med. Sci. Sports. Exerc. 1989;21:130±9. [4] Rankin G, Stokes M. Reliability of assessment tools in rehabilitation : An illustration of appropriate statistical analyses. Clin. Rehab. 1998;12:187±99. [5] Kawchuk G, Herzog W. A new technique of tissue stiness assessment: Its reliability, accuracy and comparison with an existing model. J. Manipul. Physiol. Ther. 1996;19:13±8.