- Email: [email protected]
A data-logging system to monitor bone fracture movement continuously
Copyright
PII: S1350-4533(96)00068-9
ELSEVIER
Med. Eng. Phys. Vol. 19, No. 3, pp. 286290, 1997 0 1997 Elsevier Science Ltd for IPEM. Printed in Great Britain All rights reserved 13564533/97 $17.00 + 0.00
A data-logging system to monitor movement continuously C. I. Moorcroft, Staffordshire Received 6 March,
P. J. Ogrodnik, University accepted
P. B. M. Thomas
and North 18 October
Staffordshire
bone fracture
and S. Verborg Hospital,
Stoke-on-Trent,
UK
1996
ABSTRACT A compact system is presented,
which can continuously monitor the occurrence of fracture site movement in patients with tibia1 shaft fractures treated with an Orthojiti external jixator allowing limited axial movement. The system comprises a microswitch and data logger board, which are both attached to thejxator. Each time the switch is closed by the movement of thejixator, the data logger records it as an eoent. The number of events per half hour is stored to the board’s memory. The data logger can record the frequency of movement for at least a four-week period, at which time, the data can be transferred to a computer via a serial link. The system has been proven in a limited patient trial. The results highlight the variation in patient mobility. The signijicance of continuously monitoring the fracture site movement during the healing period is discussed. 0 1997 Elseoier Science Ltd fm IPEM.
Keywords:
External
fixator,
dynamized,
Med. Eng. Phys., 1997, Vol. 19, 286290,
movement,
In the treatment of unstable long bone fractures, it is usually necessary to support the broken bone with some kind of fixation system. One method available is external fixation. In this instance, sets of pins are screwed into the bone on either side of the fracture site and protrude out from the soft tissues. Fitted to each set of pins, and bridging the fracture gap is an external column. The external fixation system then supports part or all of the load placed on the injured bone whilst restraining the fracture ends to maintain the alignment of the bone fragments. Restoring the anatomical position and alignment of the bone is termed a ‘reduction of the fracture’. The biological response to a fresh fracture involves the formation of a callus. The production of a callus is the natural way that the body reacts to attempt to bridge the fracture gap and restore continuity to the bone. For callus-bridging to take place, movement must occur between the fracture ends soon after the fracture has occurred. It has been found that axial movement is beneficial to fracture-healing’.‘. Many fixators incorporate a mechanism to allow a limited amount of axial movement to stimulate the callus response. A fixator allowing movement within its design is usually termed ‘dynamized’. Many factors influence the production of callus. One factor is the amount of movement between to: C. I. Moorcroft
healing,
continuous
monitoring
April
1. INTRODUCTION
Correspondence
fracture
the fracture ends. It has not yet been determined whether there is an optimum frequency or amplitude of movement to promote callus formation. Animal studies have recorded improved healing by applying one large displacement per day. Goodship’ recorded that 500 cycles per day of a small amount of movement improved the healing of tibia1 osteotomies in sheep. Clinical trials in humans, in which between 500 and 900 cycles per day of axial movement were induced in tibia1 fractures, showed an improvement in healing2*4,5. Normal dynamization utilizes the patient’s mobility to induce movement at the fracture site. With the fixator dynamized to allow axial movement, the load placed on the injured limb during every day activity produces the movement. The problem with this method, when attempting to monitor the effects of dynamization, is that it has been impossible to determine the frequency of movement outside the clinical environment. The frequency of fracture fragment movement during the patient’s normal daily activities has not been examined over long periods. If the amount of movement could be monitored continuously, then the effects of differing patient mobility patterns on fracture healing could be studied. This paper presents a new data-logging system that can monitor the frequency of axial movement of unstable isolated tibia1 shaft fractures during normal activity of patients treated with a dynamized external fixator. The system is compact and does not require physical links with external equipment. In this way, it is possible to monitor
the patient We present
remote to the hospital and the clinic. the results of a limited patient trial.
2. OPERATING
PRINCIPLE
Figure 1 (a) illustrates the configuration of an axially dynamized external fixator with no load applied to the injured bone. Attached to one half of the sliding mechanism of the fixator is a transducer that is able to detect axial movement. In this instance, it is a microswitch. The switch trigger rests on the opposite side of the sliding mechanism and remains open while the fixator remains unloaded. When sufficient load is applied to the fixator, the sliding mechanism allows a small amount of axial displacement, limited by the cushion ring. il;igure 2 represents the hysteresis curve for loading and unloading the Dyna-ring at various rates. It is shown that the relative stiffness of the cushion increases with load. This relative movement between the two sides of the fixator causes the microswitch to close as illustrated in F&re I(b). The closure of the switch is recorded by a data
Logger Boar4
(b)
Adjustment Plate,
Load
Cushion
3. PATIENT
TRIAL
3.1. Methods
and materials
1
Compressed
Figure 1 illustration
logger board, which is also attached to the fixator. While systems are commercially available for remote data acquisition, none met the design requirements. In general, the units were too large and would encumber the patient. One prime concern was that the unit should be attached to the fixator and not clipped to the patient’s belt; this reduces the risk of removal of the unit, or its batteries and damage to long wires that would be needed for a device worn on the body. The datalogging unit was specially designed in association with the Medical Electronics Research Group of Staffordshire University. The actual design of the data logging unit is beyond the scope of this paper, but a synopsis of its function is as follows. The data-logging unit interrogates the microswitch every 0.001 s. This is done by passing a small microamp signal to the switch and checking the return signal. If the switch is open, logic 0, then the return signal is also 0. If the switch is closed, logic 1, then the return signal is also 1. The data-logger notes if the logic of the switch has changed from the previous interrogation. If the logic changes from 0 to 1, then this is termed an ‘event’ and is recorded. Every 30 min, the number of events is stored in the unit’s m’emory. The unit has 500 kB of onboard memory, sufficient for four weeks’ continuous logging. The use of a passive sensor and low-energy memory further means that battery life is beyond four weeks. Should battery failure occur, the memory holds data already recorded, and data are not lost. The device is inexpensive and compact. It will fit within a cylinder 18 mm in diameter 70 mm in length. The data-logger board can be interrogated, using associated software, by a PC serial link. The recorded information can be transferred to the PC for further analysis. The data are presented in two columns. The first column is a 30 min time reference number, and the second is the number of events in the 30 min period.
1 (a). An illustration of the system configuration. I(b). of the fixator comprersed, closing the microswitch
An
The system was bench- and soak-tested successfully before attachment to patients treated with an axially dynamized OrthofixB fixator. The patients fitted with the system were already included in an
287
Continuous
jactuacture movement
monitting:
C. I. Momcroft
et al.
existing trial protocol and had given informed consent to be included in this trial. The aim of the trial was to test the performance of the data-logging system and to record the frequency of fracture site movement continuously over long time periods. The system was attached to the fixators of patients treated at the North Staffordshire Hospital in Stoke-on-Trent where they were seen in a research clinic. The fixator was unlocked to allow axial movement. Figure 3 presents a photograph of a patient with the datalogging system attached to the fixator. The microswitch was positioned such that it closed when a substantial proportion of the patient’s weight was supported on the injured limb. Therefore, the logic of the switch changed from 0 (nonweight-bearing) to 1 (weight-bearing) when a threshold displacement was exceeded. The switch required 0.25 mm of relative movement to operate and release, i.e. a change from logic 0 to logic 1. If this amount of movement was not present, the switch remained in a fixed state. Therefore, the system could only detect movements of 0.25 mm or greater form the operating point, set when the fixator was dynamized. The logger board was attached to the fixator within a protective aluminium sealed housing. Its small size and positioning did not cause any further encumbrance to the patient, and it did not directly affect patient mobility. The patients were fitted with the device for three to six weeks. The maximum time between
exchanges of the logger board was four weeks. The used logger was then interrogated to extract the recorded information. Once interrogated, the logger memory was reset, and a new battery was fitted. The data-logger was then considered to be ‘clean’ for the next patient. To determine the level of axial movement occurring during loading, a linear potentiometric displacement transducer was secured to the bone pins in the research clinics. The transducer gave a d.c. voltage output that was directly proportional to the axial displacement of its core shaft. The signal was recorded by a laptop computer containing an analogue-to-digital converter board. 4. RESULTS The results from a patient are shown in Figure 4. The events recorded over the 28-day trial are low. This was due to the dependence of the microswitch on a fixed operating point rather than any fault with the logger board. The patient was assessed to determine the level of axial movement of the fixator. It was found that the fixator was only moving 0.19 mm during weight bearing at the hospital. This was below the minimum needed for correct operation of the microswitch. The few events that were recorded must have been due to load levels in excess of those recorded at the hospital. F@LW 5 presents the results collected over a 19day period from the first dynamization of the fixator of a second patient. The fixator was dynamized at seven days ost-fracture, and the patient was able to bear 71 %o of his full body weight. The microswitch was set to close at this load. The patient was assessed in the clinic seven days from the start of data collection and was able to bear 73% of his body weight on the injured limb. The negligible increase in weight bearing is reflected in Figure 5, with a continued low event count recorded. On the 14th day of assessment, the patient demonstrated that he could comfortably fully bear weight on the injured limb. It is shown in Figure 5 that, from the eighth day, there is a trend towards increased activity as time progresses. This is most likely to be due to the patient’s increased ability to take more weight on the injured leg and therefore increase his mobility. Figure 6 presents a typical daily movement profile for the same patient. It shows that, throughout a normal day, the frequency of movement can vary considerably, from zero events recorded for long periods during the night to long periods of sustained movement during the day. 5. DISCUSSION
Figure
288
3
Photograph
of the system
fitted
to a Patient’s
fixator
A compact system has been developed that can monitor continuously the frequency of fracture site movement in patients suffering tibia1 shaft fractures treated with a dynamized external fixator. The system can be attached immediately after the fixator has been applied and can remain in place permenently. The data-logging board can
18
250
/ ---
-,
289
Continuousfracturemm-t
monitoring:
C. I. Moormoft
et al.
record and store over four weeks of data before it has to be exchanged. It is feasible to modify design parameters in order to adapt the system to monitor other fixator designs. The system has been proven in a limited trial. The operation of the transducer (microswitch) detected movements of 0.25 mm and above. It is possible to modify the transducer to record other desired parameters. For example, a passive pressure switch could be incorporated to record known load levels. We have demonstrated that it will be possible to record the frequency of movement that occurs throughout the healing period of a tibia1 fracture. This has a number of important implications. Goodship’ concluded that a database of normal healing could be used effectively to identify problems at an early stage, allowing prompt correction or modification to the treatment regime. It will be possible to integrate the results form the datalogger into the assessment of the fracture healing. For example, patients showing a poor callus response may exhibit generally a particular event profile, and action can be taken to encourage mobility or modify treatment. As the fracture heals, the increasing stiffness, resulting from the formation of callus, will reduce the proportion of load applied to the fixator. Therefore, the number of events recorded by the data-logger will decline. As patient data are collected, trends in the profile of this decline may emerge. From previous knowledge, it could then be possible to predict when a fracture had healed. By extrapolating the decline in events per day, it should also be possible to estimate when a fracture will heal. Furthermore, if a decline in events is unusual or not detected, it may be an indication of complications such as delayed union. Analysis of the event patterns and healing rates of patients could yield an
290
optimum frequency and timing of movement for the shortest healing time. In this instance, a modified ‘intelligent’ logger could inform the patient, with a flashing LED for instance, when he has reached the optimum level of movement for promoting callus production. ACKNOWLEDGEMENTS The authors acknowledge the contribution by Dr S. Grainger and Dr D. Hitchings of the Medical Electronics Group of Staffordshire University, and the support of EPSPRC (grant 555984) and Department of Health under the LINK Medical Implants Programme. The statements are those of the authors and do not necessarily reflect those of the Department of Health. REFERENCES 1. Goodship, A. E., Kelly, D. J., Rigby, H. S., Watkind, P. E. and Kenwright, J. The effects of different regimes of axial micromovement on the healing of experimental tibia1 fractures. Orthopaedic Transactions, 1987, 11, 285. 2. Kenwright, J., Richardson, S. B., Cunningham, J. L., White, S. H., Goodship, A. E., Adams, M. A., Magnussen, P. A. and Newman, J. H. Axial movement and tibia1 fractures. Journal ofBone and Joint Surgery, 1991, 7.3-B, 654-659. 3. Lindholm, R. V., Lindholm, T. S., Toikkanan, S. and Leino, R., Effect of forced intrafragmental movement on the healing of tibia1 fractures in rats. Acta Orthopaedica Scandinavica, 1970, 40, 721-728. 4. Kenwright, J., Goodship, A. E., Kelly, D. J., Newman, J. H., Harris, J. D., Richadson, J. B., Evans, M., Spriggins, A. J., Burrough, S. J. and Rowley, D. I. Effects of controlled axial micromovement on healing of tibia1 fractures. The Lancet, 1986, 2, 1185-1187. 5. Kenwright, J., Controlled mechanical stimulation in the treatment of tibia1 fractures. Clinical Orthopaedics, 1988, 241, 36-47. 6. Goodship, A. E., The role of fixator frame stiffness in the control of fracture healing. An experimental study. Journal of Biomechanics, 1993, 26(9), 1027-1035.
PII: S1350-4533(96)00068-9
ELSEVIER
Med. Eng. Phys. Vol. 19, No. 3, pp. 286290, 1997 0 1997 Elsevier Science Ltd for IPEM. Printed in Great Britain All rights reserved 13564533/97 $17.00 + 0.00
A data-logging system to monitor movement continuously C. I. Moorcroft, Staffordshire Received 6 March,
P. J. Ogrodnik, University accepted
P. B. M. Thomas
and North 18 October
Staffordshire
bone fracture
and S. Verborg Hospital,
Stoke-on-Trent,
UK
1996
ABSTRACT A compact system is presented,
which can continuously monitor the occurrence of fracture site movement in patients with tibia1 shaft fractures treated with an Orthojiti external jixator allowing limited axial movement. The system comprises a microswitch and data logger board, which are both attached to thejxator. Each time the switch is closed by the movement of thejixator, the data logger records it as an eoent. The number of events per half hour is stored to the board’s memory. The data logger can record the frequency of movement for at least a four-week period, at which time, the data can be transferred to a computer via a serial link. The system has been proven in a limited patient trial. The results highlight the variation in patient mobility. The signijicance of continuously monitoring the fracture site movement during the healing period is discussed. 0 1997 Elseoier Science Ltd fm IPEM.
Keywords:
External
fixator,
dynamized,
Med. Eng. Phys., 1997, Vol. 19, 286290,
movement,
In the treatment of unstable long bone fractures, it is usually necessary to support the broken bone with some kind of fixation system. One method available is external fixation. In this instance, sets of pins are screwed into the bone on either side of the fracture site and protrude out from the soft tissues. Fitted to each set of pins, and bridging the fracture gap is an external column. The external fixation system then supports part or all of the load placed on the injured bone whilst restraining the fracture ends to maintain the alignment of the bone fragments. Restoring the anatomical position and alignment of the bone is termed a ‘reduction of the fracture’. The biological response to a fresh fracture involves the formation of a callus. The production of a callus is the natural way that the body reacts to attempt to bridge the fracture gap and restore continuity to the bone. For callus-bridging to take place, movement must occur between the fracture ends soon after the fracture has occurred. It has been found that axial movement is beneficial to fracture-healing’.‘. Many fixators incorporate a mechanism to allow a limited amount of axial movement to stimulate the callus response. A fixator allowing movement within its design is usually termed ‘dynamized’. Many factors influence the production of callus. One factor is the amount of movement between to: C. I. Moorcroft
healing,
continuous
monitoring
April
1. INTRODUCTION
Correspondence
fracture
the fracture ends. It has not yet been determined whether there is an optimum frequency or amplitude of movement to promote callus formation. Animal studies have recorded improved healing by applying one large displacement per day. Goodship’ recorded that 500 cycles per day of a small amount of movement improved the healing of tibia1 osteotomies in sheep. Clinical trials in humans, in which between 500 and 900 cycles per day of axial movement were induced in tibia1 fractures, showed an improvement in healing2*4,5. Normal dynamization utilizes the patient’s mobility to induce movement at the fracture site. With the fixator dynamized to allow axial movement, the load placed on the injured limb during every day activity produces the movement. The problem with this method, when attempting to monitor the effects of dynamization, is that it has been impossible to determine the frequency of movement outside the clinical environment. The frequency of fracture fragment movement during the patient’s normal daily activities has not been examined over long periods. If the amount of movement could be monitored continuously, then the effects of differing patient mobility patterns on fracture healing could be studied. This paper presents a new data-logging system that can monitor the frequency of axial movement of unstable isolated tibia1 shaft fractures during normal activity of patients treated with a dynamized external fixator. The system is compact and does not require physical links with external equipment. In this way, it is possible to monitor
the patient We present
remote to the hospital and the clinic. the results of a limited patient trial.
2. OPERATING
PRINCIPLE
Figure 1 (a) illustrates the configuration of an axially dynamized external fixator with no load applied to the injured bone. Attached to one half of the sliding mechanism of the fixator is a transducer that is able to detect axial movement. In this instance, it is a microswitch. The switch trigger rests on the opposite side of the sliding mechanism and remains open while the fixator remains unloaded. When sufficient load is applied to the fixator, the sliding mechanism allows a small amount of axial displacement, limited by the cushion ring. il;igure 2 represents the hysteresis curve for loading and unloading the Dyna-ring at various rates. It is shown that the relative stiffness of the cushion increases with load. This relative movement between the two sides of the fixator causes the microswitch to close as illustrated in F&re I(b). The closure of the switch is recorded by a data
Logger Boar4
(b)
Adjustment Plate,
Load
Cushion
3. PATIENT
TRIAL
3.1. Methods
and materials
1
Compressed
Figure 1 illustration
logger board, which is also attached to the fixator. While systems are commercially available for remote data acquisition, none met the design requirements. In general, the units were too large and would encumber the patient. One prime concern was that the unit should be attached to the fixator and not clipped to the patient’s belt; this reduces the risk of removal of the unit, or its batteries and damage to long wires that would be needed for a device worn on the body. The datalogging unit was specially designed in association with the Medical Electronics Research Group of Staffordshire University. The actual design of the data logging unit is beyond the scope of this paper, but a synopsis of its function is as follows. The data-logging unit interrogates the microswitch every 0.001 s. This is done by passing a small microamp signal to the switch and checking the return signal. If the switch is open, logic 0, then the return signal is also 0. If the switch is closed, logic 1, then the return signal is also 1. The data-logger notes if the logic of the switch has changed from the previous interrogation. If the logic changes from 0 to 1, then this is termed an ‘event’ and is recorded. Every 30 min, the number of events is stored in the unit’s m’emory. The unit has 500 kB of onboard memory, sufficient for four weeks’ continuous logging. The use of a passive sensor and low-energy memory further means that battery life is beyond four weeks. Should battery failure occur, the memory holds data already recorded, and data are not lost. The device is inexpensive and compact. It will fit within a cylinder 18 mm in diameter 70 mm in length. The data-logger board can be interrogated, using associated software, by a PC serial link. The recorded information can be transferred to the PC for further analysis. The data are presented in two columns. The first column is a 30 min time reference number, and the second is the number of events in the 30 min period.
1 (a). An illustration of the system configuration. I(b). of the fixator comprersed, closing the microswitch
An
The system was bench- and soak-tested successfully before attachment to patients treated with an axially dynamized OrthofixB fixator. The patients fitted with the system were already included in an
287
Continuous
jactuacture movement
monitting:
C. I. Momcroft
et al.
existing trial protocol and had given informed consent to be included in this trial. The aim of the trial was to test the performance of the data-logging system and to record the frequency of fracture site movement continuously over long time periods. The system was attached to the fixators of patients treated at the North Staffordshire Hospital in Stoke-on-Trent where they were seen in a research clinic. The fixator was unlocked to allow axial movement. Figure 3 presents a photograph of a patient with the datalogging system attached to the fixator. The microswitch was positioned such that it closed when a substantial proportion of the patient’s weight was supported on the injured limb. Therefore, the logic of the switch changed from 0 (nonweight-bearing) to 1 (weight-bearing) when a threshold displacement was exceeded. The switch required 0.25 mm of relative movement to operate and release, i.e. a change from logic 0 to logic 1. If this amount of movement was not present, the switch remained in a fixed state. Therefore, the system could only detect movements of 0.25 mm or greater form the operating point, set when the fixator was dynamized. The logger board was attached to the fixator within a protective aluminium sealed housing. Its small size and positioning did not cause any further encumbrance to the patient, and it did not directly affect patient mobility. The patients were fitted with the device for three to six weeks. The maximum time between
exchanges of the logger board was four weeks. The used logger was then interrogated to extract the recorded information. Once interrogated, the logger memory was reset, and a new battery was fitted. The data-logger was then considered to be ‘clean’ for the next patient. To determine the level of axial movement occurring during loading, a linear potentiometric displacement transducer was secured to the bone pins in the research clinics. The transducer gave a d.c. voltage output that was directly proportional to the axial displacement of its core shaft. The signal was recorded by a laptop computer containing an analogue-to-digital converter board. 4. RESULTS The results from a patient are shown in Figure 4. The events recorded over the 28-day trial are low. This was due to the dependence of the microswitch on a fixed operating point rather than any fault with the logger board. The patient was assessed to determine the level of axial movement of the fixator. It was found that the fixator was only moving 0.19 mm during weight bearing at the hospital. This was below the minimum needed for correct operation of the microswitch. The few events that were recorded must have been due to load levels in excess of those recorded at the hospital. F@LW 5 presents the results collected over a 19day period from the first dynamization of the fixator of a second patient. The fixator was dynamized at seven days ost-fracture, and the patient was able to bear 71 %o of his full body weight. The microswitch was set to close at this load. The patient was assessed in the clinic seven days from the start of data collection and was able to bear 73% of his body weight on the injured limb. The negligible increase in weight bearing is reflected in Figure 5, with a continued low event count recorded. On the 14th day of assessment, the patient demonstrated that he could comfortably fully bear weight on the injured limb. It is shown in Figure 5 that, from the eighth day, there is a trend towards increased activity as time progresses. This is most likely to be due to the patient’s increased ability to take more weight on the injured leg and therefore increase his mobility. Figure 6 presents a typical daily movement profile for the same patient. It shows that, throughout a normal day, the frequency of movement can vary considerably, from zero events recorded for long periods during the night to long periods of sustained movement during the day. 5. DISCUSSION
Figure
288
3
Photograph
of the system
fitted
to a Patient’s
fixator
A compact system has been developed that can monitor continuously the frequency of fracture site movement in patients suffering tibia1 shaft fractures treated with a dynamized external fixator. The system can be attached immediately after the fixator has been applied and can remain in place permenently. The data-logging board can
18
250
/ ---
-,
289
Continuousfracturemm-t
monitoring:
C. I. Moormoft
et al.
record and store over four weeks of data before it has to be exchanged. It is feasible to modify design parameters in order to adapt the system to monitor other fixator designs. The system has been proven in a limited trial. The operation of the transducer (microswitch) detected movements of 0.25 mm and above. It is possible to modify the transducer to record other desired parameters. For example, a passive pressure switch could be incorporated to record known load levels. We have demonstrated that it will be possible to record the frequency of movement that occurs throughout the healing period of a tibia1 fracture. This has a number of important implications. Goodship’ concluded that a database of normal healing could be used effectively to identify problems at an early stage, allowing prompt correction or modification to the treatment regime. It will be possible to integrate the results form the datalogger into the assessment of the fracture healing. For example, patients showing a poor callus response may exhibit generally a particular event profile, and action can be taken to encourage mobility or modify treatment. As the fracture heals, the increasing stiffness, resulting from the formation of callus, will reduce the proportion of load applied to the fixator. Therefore, the number of events recorded by the data-logger will decline. As patient data are collected, trends in the profile of this decline may emerge. From previous knowledge, it could then be possible to predict when a fracture had healed. By extrapolating the decline in events per day, it should also be possible to estimate when a fracture will heal. Furthermore, if a decline in events is unusual or not detected, it may be an indication of complications such as delayed union. Analysis of the event patterns and healing rates of patients could yield an
290
optimum frequency and timing of movement for the shortest healing time. In this instance, a modified ‘intelligent’ logger could inform the patient, with a flashing LED for instance, when he has reached the optimum level of movement for promoting callus production. ACKNOWLEDGEMENTS The authors acknowledge the contribution by Dr S. Grainger and Dr D. Hitchings of the Medical Electronics Group of Staffordshire University, and the support of EPSPRC (grant 555984) and Department of Health under the LINK Medical Implants Programme. The statements are those of the authors and do not necessarily reflect those of the Department of Health. REFERENCES 1. Goodship, A. E., Kelly, D. J., Rigby, H. S., Watkind, P. E. and Kenwright, J. The effects of different regimes of axial micromovement on the healing of experimental tibia1 fractures. Orthopaedic Transactions, 1987, 11, 285. 2. Kenwright, J., Richardson, S. B., Cunningham, J. L., White, S. H., Goodship, A. E., Adams, M. A., Magnussen, P. A. and Newman, J. H. Axial movement and tibia1 fractures. Journal ofBone and Joint Surgery, 1991, 7.3-B, 654-659. 3. Lindholm, R. V., Lindholm, T. S., Toikkanan, S. and Leino, R., Effect of forced intrafragmental movement on the healing of tibia1 fractures in rats. Acta Orthopaedica Scandinavica, 1970, 40, 721-728. 4. Kenwright, J., Goodship, A. E., Kelly, D. J., Newman, J. H., Harris, J. D., Richadson, J. B., Evans, M., Spriggins, A. J., Burrough, S. J. and Rowley, D. I. Effects of controlled axial micromovement on healing of tibia1 fractures. The Lancet, 1986, 2, 1185-1187. 5. Kenwright, J., Controlled mechanical stimulation in the treatment of tibia1 fractures. Clinical Orthopaedics, 1988, 241, 36-47. 6. Goodship, A. E., The role of fixator frame stiffness in the control of fracture healing. An experimental study. Journal of Biomechanics, 1993, 26(9), 1027-1035.