Journal of Postgraduate Medicine
 Open access journal indexed with Index Medicus & EMBASE  
     Home | Subscribe | Feedback  

Year : 1989  |  Volume : 35  |  Issue : 1  |  Page : 43-8  

Role of pulsed electromagnetic fields in recalcitrant non-unions.

DF Delima, DD Tanna 

Correspondence Address:
D F Delima


Twenty-nine patients of recalcitrant nonunion of long bones were treated by pulsed electromagnetic fields in an attempt to bring about osteogenesis. The pulse used was rectangular, equal mark space wave in the astable, continuous mode operating at a frequency of 40 Hertz. The success rate was 82.5%. The result was not dependent on the age, sex, time of nonunion or the presence of infection. However, the results were uniformly poor when infection and fracture instability were coexistent in the same patient.

How to cite this article:
Delima D F, Tanna D D. Role of pulsed electromagnetic fields in recalcitrant non-unions. J Postgrad Med 1989;35:43-8

How to cite this URL:
Delima D F, Tanna D D. Role of pulsed electromagnetic fields in recalcitrant non-unions. J Postgrad Med [serial online] 1989 [cited 2022 Sep 24 ];35:43-8
Available from:

Full Text


Electrical methods are playing an increasing role in the therapeutic management of un-united fractures.[1],[7],[11] In fact the stimulation of osteogenesis by electrical methods is being considered as an alternative to bone grafting in delayed union and non-union and as a supplement to fresh autogenous cancellous grafting.[2] Electrical treatment options in non-union are either surgically invasive methods or surgically non-invasive methods. The latter mode uses pulsed electromagnetic fields as one such method. The pulsed electromagnetic fields are delivered by two opposite coils of wire mounted on the external surface of the cast or skin.[1],[3] As the electromagnetic waves pulse through the extremity, they induce weak, time varying currents in the tissue that are similar to those generated by bone in response to deformations These currents trigger calcification of gap tissue and result in bony union. Although there is experimental evidence to Drove pulse specificity[1],[3] -i.e. pulses designed to evoke specific biological responses, it appears that widely different signals produce similar effects arguing against any great specificity of signal.[4]

The aim of this study was to prove whether pulsed electromagnetic fields using a rectangular pulse wave, of equal mark space period, in the continuous mode at a frequency of 40 Hertz was capable of osteogenesis in 29 recalcitrant non-unions of long bones.



The pulse generator supplied a time varying current to a pair of coils in a Helmholtz aiding state. The driving voltage was between 12 volts in the upper extremity and 30 volts in the lower extremity. The wave form was rectangular in the astable, continuous equal mark space mode with a frequency of 40 Hertz. The pulse train was continuous.

The Helmholtz coils were copper wire wound on an air former. They were positioned facing each other at 180 degrees with the extremity or cast between them. The coils were wound in such a way that one 'pushes' and other 'pulls'. Care was also taken to see that the intercoil distance was equal to or less than the coil diameter, resulting in a Helmholtz aiding state and a central uniform magnetic field. The coil placement was done under radiographic control to prevent proximal, distal and rotational skew and also to prevent shadowing of bone in the presence of an implant.


All patients were stimulated for 16-18 hours a day. During this period, the patients were kept strictly non-weight bearing. Serial check radiographs were taken at 8 weeks and 3 months to see for any signs of osteogenesis. Calcification and haziness of the fracture gap with a decrease in the sclerosis of the fracture margins was an indication that calcification of the fibrocartilage was taking place. At this stage the patient was instructed to perform axial compression exercises and maintain the non-weight bearing status for a further 2-3 months. Electrical stimulation was continued during this period. When mature lamellar bone was seen bridging the fracture gap at atleast 2 sites the patient was allowed partial weight bearing.


The prestimulation operative details of the patients in this study were as shown in[Table 1].

Twenty nine patients comprised this study, of these 20 were males and 9 females. Eighteen patients were between 21 and 40 years, 10 patients between 41 and 60 years and 1 patient 65 year old.

Of the 29 patients, the non-union or delayed union followed operative intervention in 26 patients of which 20 were simple injuries and 6 compound injuries. Two patients had non-union following closed treatment for a simple and compound fracture respectively. One patient was stimulated following an osteoclastoma resection of the upper end of the tibia, reconstruction and autogenous cancellous bone grafting.

The time interval of the delayed/nonunion ranged from 0 to 3 months in 2 patients, 4 to 9 months in 7 patients, 9 to 24 months in 18 patients and greater than 24 months in 1 patient.

In this period of non-union/delayed union, 8 patients had 1 surgery prior to stimulation, 9 patients had 2 surgeries prior to stimulation, 6 patients had 3 surgeries, 2 patients had 4 surgeries and 1 patient had 6 surgeries prior to stimulation. This was over and above the primary fracture treatment offered when they presented with a fresh fracture. Three patients had no surgery between primary surgery and electrical stimulation.

There were 7 non-unions of the humerus, the prestimulation details of which were as follows-6 simple fractures and 1 compound fracture. All fractures were treated primarily either by intramedullary nail or compression plate with a cancellous graft [Table 1]. Just prior to stimulation the fixation was considered to be stable in 5 patients and unstable in 2 patients. There were 2 infected cases.

There were 15 non-unions of the tibia/fibula, the prestimulation details of which were as follows-10 simple fractures, 4 compound fractures and 1 patient of an osteoclastoma of the upper end of the tibia. All fractures were treated primarily either by an intramedullary nail or compression plate with cancellous autograft. Prior to stimulation the fractures were considered stable in 13 patients and unstable in 2 patients. Of the 15 patients in this group, 8 were infected fractures.

There were 6 non-union of the femur; 4 were simple fractures and 2 compound injuries. All patients were treated by an intramedullary nail and autogenous graft except for patient no. 24 who was supplemented with a fibular graft and autograft, patient no. 26 and 28 were not grafted at all and in patient no. 27 only a fibular graft was inserted with no internal fixation. There were 2 infected fractures, one of whose internal fixation was considered unstable.

There was only 1 patient of an non-union radius/ulna. Prior to stimulation, it presented as a compound fracture treated by compression plating without grafting. At the time of stimulation the fracture had been revised, stable but infected.


Of the 29 patients included in this study, only 1 patient (no. 3) was lost to follow up. Twenty two patients were followed upto a time interval 1-2 years, 5 patients for more than 2 years and 1 patient for less than a year.

We had 5 failures with an overall success rate of 82.14%. Of the 18 patients in the 21-40 years age group, there were 3 failures and a success rate of 77.77%. Of the 10 patients in the 41-60 year age group, 8 united with a success rate of 80%. The single patient above 60 years was considered a success.

There were 7 non-unions of the humerus, of which 1 was lost to follow up, 4 united and 2 patients (no. 5 and no. 6) were considered failures. There were 15 non-unions of the tibia/fibula, 13 of which united and 2 failed.

There were 6 non-unions of the femur, 5 united and 1 failure. The single case of non-union of the radius/ulna united.

There were 9 fractures where the time of delayed union was 9 months and less, all 9 fractures united, between 9 months and 2 years, 14 fractures united and 3 failed with one patient lost to follow up. Both the fractures with a non-union time of greater than 2 years failed.

There were 3 patients who had no surgical intervention between primary surgery and stimulation and all three united, 8 patients had 1 surgery and all 8 united, 8 patients had 2 surgeries and 5 united 2 failed and 1 was lost to follow up, 6 patients had 3 surgeries, 4 of which united and 2 failed, 1 patient had 4 surgeries which united and 1 patient who had 6 surgeries was also considered a success.

Of the 29 patients in this study, 7 patients had no fracture gap and all united, 9 patients had a gap of less than 1 cm and all united, 10 patients had a gap between 1 cm and less than 2 cm of which 7 united, 2 failed and 1 was lost to follow up. The 3 patients with a gap of more than 2 cm were considered failures.

Of the 29 patients in this study, we had 15 stable fractures prior to stimulation, 14 of which united and 1 was lost to follow up. One grossly unstable fracture was a failure.

Of the 9 infected fractures with a rigid fixation all 9 united. Of the 4 infected fractures with poor unstable fixation, all 4 failed. Of the 9 stable, infected fractures, 5 patients had no fracture gap, 1 patient less than 1 cm and 2 patients between 1 and less than 2 cm. All fractures united.

Of the 4 infected unstable fractures, there were 2 patients with a fracture gap of 1 cm and 2 patients with a fracture gap of 2 cm. All the four patients were considered failures.


The treatment of delayed and non-union of long bones using pulsed electromagnetic fields is a well known therapeutic method.[2],[3] Although most successful experiments seem to have an estimated internal voltage gradient of 0.5 to 1.5 mV per sq. cm in common,[6] the hypothesis of pulse specificity[1],[3] i.e. pulses designed to evoke specific biological response was not acceptable to us as several experiments with failure to obtain stimulation[9],[10] were performed under similar conditions.

The pulsed electromagnetic field stimulator developed by us generated a wave that was rectangular, equal mark/space, continuous and at a frequency of 40 Hertz. The driving voltage was either 40 volts for the lower extremity and 25 volts for the upper extremity. Taking the usual precautions of coil positioning and orientation, we feel that such a pulsed electromagnetic field is capable of osteogenesis as is evidenced by a success rate of 82.14%. This compares favourably to other reports of osteogenesis by pulse specificity.[1],[2],[3]

The result was not influenced by either the age or the sex of the patient. The failures did increase as the number of surgeries the patient was subjected to, prior to stimulation increased.[5] As regards to osteogenesis and the time of non-union, our results were good upto a non-union time of 2 years, beyond this time interval the results were bad. However, the poor result cannot be attributed solely to the time of nonunion as these fractures were also complicated by infection and fracture instability. These very same factors came into play when assessing osteogenesis and fracture non-union. We had good results in patients whose gap non-union was less than 2 cm.

The presence of infection in the presence of fracture stability did not seem to affect the role of pulsed electromagnetic fields in osteogenesis. However, we had poor results in patients whose fractures were unstable in the presence of infection and as such we do not recommend electrical pulsed electromagnetic stimulation in unstable infected fractures.

We do not routinely remove the implant prior to stimulation. However, in those cases where the fixation was poor with the fracture grossly unstable (case no. 4), the fixation did not serve any purpose at all. In such cases the implant was removed prior to surgery [Fig. 1]. We have had to resort to implant removal prior to electrical stimulation in two cases [Fig. 2].

We conclude that the pulsed electromagnetic fields of the type used by us are capable of bringing about osteogenesis. The result compares favourably with other known methods including surgical methods[2],[4] but has the advantage of being surgically noninvasive and an out-patient procedure. It should be given a consideration as a therapeutic aid in bringing about osteogenesis in patients with the proper indication.


(1) This project was carried out under a research grant from the Nair Golden Jubilee Research Foundation.

(2) The paper was read at:

-Western India Regional Orthopaedic Conference, 1985, Bombay.

XI All India Symposium cum workshop on Biomedical Engineering, Bhabha Atomic Research Center, December, 1985, Bombay.

(3) We remain grateful to the Dean, B.Y.L. Nair Hospital for use of hospital data.


1Bassett, C. A. L., Mitchell, S. N. and Gaston, S. R.: Pulsing electromagnetic field treatment in ununited fractures and failed arthrodeses. J. Amer. Med. Assoc., 247: 623-628, 1982.
2Bassett, C. A. L., Mitchell, S. N. and Schink, M. M.: Treatment of therapeutically resistant nonunions with bone grafts and pulsing electromagnetic fields. J. Bone & Joint Surg" 64-A: 1214-1220, 1982.
3Bassett, C. A. L., Mitchell, S. N. and Gaston, S. R.: Treatment of ununited tibial diaphyseal fractures with pulsing electromagnetic fields. J. Bone & Joint Surg., 63A: 511-523, 1981.
4Black, J.: Tissue response to exogenous electromagnetic signals. Orthop. Clin. N. Amer., 15: 15-31, 1984.
5Boyd, H. B., Lipinski, S. W. and Wiley J. H.: Observation on non-union of the shafts of the long bones, with a statistical analysis of 842 patients, J. Bane & Joint Surg., 43A: 159-168, 1961.
6Brighton, C. T.: Treatment of nonunion of the tibia with constant direct current. J. Trauma., 21: 189-195, 1981.
7Brighton, C. T., Black, J. and Friedenberg, Z. B., Esterhai, J. L., Day, L. J . and Connolly, J. F.: A multicenter study of the treatment of non-union with constant direct current. J. Bone & Joint Surg., 63A: 2-13, 1981.
8Cochran, G. V., Pawluk, R. J. and Bassett, C. A. L.: Electromechanical characteristics of bone under physiologic moisture conditions. Clin. Orthop., 58: 249-270, 1968.
9Harris, W. H., Moyen, B. J. L. and Thrasher, E. L., Davis, L. A., Cobden, R. H., Mackenzie, D. A. and Cywinski, J. K.: Differential response to electrical stimulation: A distinction between induced osteogenesis in intact tibiae and the effect on fresh fracture defects in radii. Clin. Orthop., 124: 31-40, 1977.
10Hassler, C. R., Rybicki, E. R. and Diegle, R. B. and Clark, R. C.: Studies of enhanced bone healing via electrical stimuli. Comparative effectiveness of various parameters. Clin. Orthop., 124: 9-19, 1977.
11Paterson, D. C., Lewis, G. N. and Cass, C. A.: Treatment of delayed union and nonunion with an implanted direct current stimulator. Clin. Orthop., 148: 117-128, 1980.

Saturday, September 24, 2022
 Site Map | Home | Contact Us | Feedback | Copyright  and disclaimer