 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Preoperative Irradiation for Prevention of Heterotopic Ossification following Total Hip Arthroplasty*  , ROCHESTER, NEW YORKInvestigation performed at Strong Memorial Hospital, University of Rochester, Rochester
Eighty-six hips in eighty-five patients who were considered to be at risk for heterotopic ossification following a total hip arthroplasty were prospectively randomized or assigned to one of two treatment groups that received a single 800-centigray dose of limited-field radiation either preoperatively (Group I) or postoperatively (Group II). The risk factors for postoperative heterotopic ossification included previous heterotopic ossification following an operation about the hip, hypertrophic osteoarthrosis or post-traumatic osteoarthrosis characterized by the presence of extensive osteophytes, radiographic evidence of diffuse idiopathic skeletal hyperostosis, and ankylosing spondylitis. The hips in Group I were irradiated within 6.1 hours before the operation and those in Group II, within 51.3 hours after the operation. Either extra-field ossification or heterotopic ossification was observed in forty-one (48 per cent) of the eighty-six hips, thereby confirming the high risk for the population in this study. After a minimum duration of follow-up of six months, thirty-seven (76 per cent) of the forty-nine hips that had been treated with preoperative irradiation exhibited no new heterotopic ossification and eleven, progression to grade-I or II ossification. The remaining hip in that group was in a woman who had Paget disease as well as previous grade-IV (ankylosing) heterotopic ossification about the ipsilateral hip; heterotopic ossification progressed from grade II on the radiographs made immediately after the index revision procedure to grade III at the most recent follow-up assessment. Of the thirty-seven hips that had been treated with postoperative irradiation, twenty-seven (73 per cent) exhibited no new heterotopic ossification and nine had progression from grade-0 to grade-I ossification. The remaining hip in that group was in a man who had Parkinson disease and previous grade-III ossification about the ipsilateral hip; heterotopic ossification progressed from grade III immediately postoperatively to grade IV at the time of the most recent evaluation. Extra-field ossification was identified in twelve (24 per cent) of the forty-nine hips that had been irradiated preoperatively compared with three (8 per cent) of the thirty-seven hips that had been irradiated postoperatively (p = 0.05). Extra-field ossification was not associated with clinical symptoms of bursitis of the greater trochanter in any hip. Three of the ten hips that had a revision operation subsequently had a non-union of the greater trochanter; all three had been treated with preoperative irradiation. The findings of the present study suggest that preoperative irradiation is effective for the prevention of heterotopic ossification following total hip arthroplasty and that it eliminates the discomfort and morbidity that are associated with conventional postoperative treatment. Furthermore, the efficacy of preoperative irradiation suggests that osteogenic precursor cells that are active in this process are derived from the local tissues within the operative field rather than from distant blood-borne cell lines.
Heterotopic ossification has been reported as an incidental finding on the radiographs of as many as 84 per cent of patients who have had a total hip arthroplasty7,14,27,32,36,39,52,54,61,62,65. Three to 10 per cent of these patients have had functional impairment, primarily in the form of a diminished range of motion or pain secondary to the inflammatory reaction that accompanies the process of heterotopic ossification21,23,33,39,41,42,46,54,57,61,62. In response to this compromised outcome following total hip replacement, both chemotherapy and radiotherapy have been investigated as possible measures to prevent heterotopic ossification8,17,20,25,37,44,49,53,58,59,63,64,66,68. Prophylactic irradiation has been the favored intervention at our institution because of the potentially adverse effect of systemic chemotherapeutic agents on the growth of bone into a prosthesis that was inserted without cement1,2,18,22,35,44,46-48,50. Previous studies have demonstrated the efficacy of fractionated doses of 2000 and then 1000 centigray in the prevention of heterotopic ossification, provided that the treatment was initiated within four days after the operation1,16,44. More recently, both laboratory and prospective clinical investigations have demonstrated the accuracy of delivery as well as the efficacy of a single dose of 800 centigray administered to a limited field about the hip and have not shown that such treatment has any adverse effect on the early fixation of a prosthesis that was inserted without cement16,35,48. In response to a 31 per cent prevalence of extra-field ossification (which occasionally caused symptomatic bursitis) in the region of the greater trochanter in a series of sixty-two hips that had been treated with limited-field irradiation, the rectangular treatment field was modified to an L-shaped configuration to include the trochanteric ridge at the origin of the vastus lateralis47,48. However, radiation therapy after total hip replacement still is not without problems. The discomfort of the patient during transport to the radiation-therapy suite, and the dislocation of the hip that occasionally occurs during transfer there, provided us with the incentive to investigate the efficacy of preoperative treatment. Kantorowitz et al.29 described the results of a study in which a pellet of decalcified rat-bone matrix was implanted bilaterally into the quadriceps muscles of 111 Long-Evans rats. Single-fraction doses of 300, 800, 1800, 2400, and 3000 centigray were administered to the recipient thigh muscles of cohorts forty-eight hours before, one hour before, and forty-eight hours after the implantation of the pellet. Heterotopic ossification was noted in 40 to 50 per cent of control limbs by the sixteenth postoperative day. While irradiation was associated with a significant (p < 0.03) reduction of heterotopic ossification at all three time-intervals, 800 centigray yielded nearly maximum prophylactic effect; 300 centigray was clearly less effective, and doses of more than 800 centigray added little benefit. Moreover, the efficacy of prophylaxis with radiation was comparable between the group of thirty-eight animals that had been irradiated one hour before the implantation of the pellet (mean area of heterotopic ossification, 5.3 per cent) and the group of thirty-seven animals irradiated forty-eight hours after it (mean area of heterotopic ossification, 7.1 per cent). However, the thirty-six animals that had been irradiated forty-eight hours before the implantation exhibited significantly more heterotopic ossification (mean area of heterotopic ossification, 12.6 per cent) than did those that had been irradiated one hour before the operation (p = 0.02). The purpose of the present prospective study was to compare the efficacy, for the prevention of heterotopic ossification, of a single 800-centigray dose of limited-field radiation administered immediately preoperatively with that of the same dose given within seventy-two hours postoperatively.
Between August 1990 and December 1992, eighty-six hips in eighty-five patients who were to be managed with either primary or revision total hip arthroplasty at the University of Rochester, Strong Memorial Hospital, were considered to be at risk for postoperative heterotopic ossification and were entered into a prospective study that had been approved by the Institutional Review Board (Table I). The patients were randomized or assigned to one of two treatment groups that were scheduled to receive a single dose of 800 centigray of radiation delivered to a limited L-shaped field (or a rectangular field if a revision arthroplasty was done) either preoperatively (Group I) or postoperatively (Group II).
The identification of patients who were at risk for heterotopic ossification was based on previously described criteria3,7,12,19,39,45,51,54,61,62. Patients known to be at greatest risk are those who have had previous heterotopic ossification following an operation on the hip. Accepted risk factors in patients who have not had such an operation include hypertrophic osteoarthrosis or post-traumatic osteoarthrosis characterized by extensive osteophytes, radiographic evidence of diffuse idiopathic skeletal hyperostosis, ankylosing spondylitis, and male gender. The eighty-six hips that were thought to be at high risk for heterotopic ossification were classified into four groups of increasing risk, as described previously44, for the purpose of subsequent analysis of the results of treatment. Sixty-three hips (73 per cent) were considered to be in risk-class I, meaning that the patient had not had heterotopic ossification following a previous operation about either hip. Twenty-three hips (27 per cent) were deemed to be in risk-class II, III, or IV; this meant that the patient was at increased risk because heterotopic ossification had occurred following a previous operation about the contralateral hip (risk-class II), about the ipsilateral hip (risk-class III), or about the ipsilateral hip with resultant ankylosis (risk-class IV). Group I included thirty-four risk-class-I, nine risk-class-II, five risk-class-III, and one risk-class-IV hips; Group II included twenty-nine risk-class-I, six risk-class-II, and two risk-class-III hips. The predominant risk factor for heterotopic ossification in both treatment groups was radiographic evidence of either hypertrophic osteoarthrosis or diffuse idiopathic skeletal hyperostosis. Previous heterotopic ossification was noted in fifteen (31 per cent) of the hips that were to be irradiated preoperatively and in only eight (22 per cent) of the hips that were to be irradiated postoperatively. The indication for the total hip arthroplasty was hypertrophic osteoarthrosis in seventy-six hips (forty-one from Group I and thirty-five from Group II) and revision of a failed total hip arthroplasty in ten hips (eight from Group I and two from Group II). The prosthesis was implanted without cement in all but one of the hips in each treatment group; in the remaining hip in each group, an acetabular component was inserted without cement and a femoral stem was inserted with cement. Men outnumbered women by a ratio of two to one but were apportioned comparably between the two treatment groups (thirty-one [63 per cent] of the forty-nine hips in Group I were in men, compared with twenty-three [62 per cent] of the thirty-seven hips in Group II). The mean age of the patients who were managed with irradiation preoperatively was sixty-three years (range, thirty-two to eighty-six years), compared with sixty years (range, thirty-four to seventy-nine years) for those who were so managed postoperatively. Informed consent was obtained preoperatively from all patients who were enrolled in the study. The patients agreed to discontinue the use of all non-steroidal anti-inflammatory medications, glucocorticoids, and diphosphonates for seven days before the scheduled total hip replacement and for six months postoperatively. Initially, preoperative prospective randomization to one of the two treatment groups was carried out according to a computer-generated scheme, with an equal likelihood of assignment to either of the groups. Then, a preliminary analysis of the results, performed eighteen months after the start of the trial, suggested comparable outcomes in the two groups. In an effort to expedite the attainment of statistically valid data, and in view of our large experience with postoperative treatment, we elected to assign the last fifteen patients who were entered into the study directly to the group that was to receive preoperative irradiation (Group I) in order to increase the statistical power of subgroup analysis within that cohort. Group I was scheduled to receive radiation therapy within four hours before the start of the total hip arthroplasty, and Group II, within seventy-two hours after the conclusion of the procedure. Forty-eight patients (forty-nine hips), who were scheduled to have forty-one primary and eight revision total hip replacements, were randomized or assigned to Group I. Thirty-seven patients (thirty-seven hips), who were scheduled to have thirty-five primary and two revision total hip arthroplasties, were randomized to Group II. All primary procedures were performed through an anterolateral (Watson-Jones) approach. An osteotomy of the greater trochanter was not done in association with routine primary total hip arthroplasty; however, to improve exposure and to facilitate the removal of cement from deep within the femoral canal, a chevron trochanteric osteotomy was performed for all patients who had a revision of a cemented femoral component. Pre-existing heterotopic bone, when present, was excised from the operative field only to the extent that was necessary for exposure and dissection of the abductor muscles during a revision procedure. The greater trochanter was reattached with a three-wire technique. Two intramedullary wires were passed distally through the lateral cortex of the femur and proximally through the abductors over the proximal edge of the trochanter. The third wire was placed circumferentially around the shaft and the greater trochanter through a hole in the lesser trochanter. In all operations, both primary and revision, the operative field was irrigated copiously with a pulsatile-lavage device to remove osseous debris from the soft tissues. All wounds were closed over deep suction drains. All patients received prophylaxis against thromboembolic disease with warfarin beginning on the day before the operation. The use of aspirin and other non-steroidal anti-inflammatory agents was specifically avoided for six months postoperatively34. The radiation therapy was similar for both groups, with the exception of the timing of treatment. All therapy was given at the same facility with megavoltage equipment (generally, eighteen-megavolt photons were used, although ten-megavolt photons and cobalt-60 units were employed for a few patients) and use of isocentric, or single-focus-center, technique at a source-to-access distance of 100 centimeters. All patients received a single 800-centigray fraction that was given in opposed anteroposterior fields with the dose prescribed to the central axis, mid-plane depth. Before the irradiation, all patients had simulated treatment for planning and marking of the treatment portal on the skin. Patients who were to receive the radiation therapy preoperatively generally had the simulated treatment either one or two days before the actual treatment if the operation was scheduled as the first procedure in the morning, or on the day of the actual treatment if the operation was scheduled for later in the day. Patients who were to receive the radiation therapy postoperatively had the simulated treatment immediately before the actual treatment. Radiographs were made before the actual treatment to confirm the orientation of the treatment portal as well as the placement of the lead block (if used). An obliquely oriented rectangular portal with a lateral extension that covered the proximal tip of the greater trochanter (resulting in an L-shaped configuration) was centered over the site of the arthroplasty in patients managed with a primary procedure (Fig. 1). A simple rectangular portal without a lateral extension was used for all patients managed with a revision procedure, so as to exclude the osteotomy site from the treatment field. Since all procedures were performed through either an anterolateral or a transtrochanteric approach, the incision in the skin was excluded from the treatment field in every hip. The treatment field included the area between the lesser trochanter and the ischial ramus and extended laterally to include the area between the greater trochanter and the wing of the ilium. The porous surfaces of implants that had been, or were going to be, inserted without cement and that were intended for ingrowth of bone were excluded carefully from the treatment field with use of a rectangular lead block.
Radiographic evidence of heterotopic ossification was graded according to a modification of the classification scheme described by Brooker et al.7. Grade 0 indicated that no heterotopic bone was visible, grade I was characterized by islands of bone within the soft tissues about the hip, grade II indicated that bone originated from the pelvis or the femur and that there was at least one centimeter between apposing surfaces, grade III was the same as grade II except that there was less than one centimeter of space between apposing surfaces, and grade IV indicated that ankylosis of bone was apparent on the anteroposterior radiograph. The area in which the ectopic bone had formed was quantified with use of previously described classes that have been shown to correlate closely with functional outcome44. The area about the hip was divided into thirds on the standard anteroposterior pelvic radiograph, and the amount of heterotopic bone was quantified as class A when 1 to 33 per cent of the area was involved, as class B when 34 to 66 per cent of the area was involved, and as class C when 67 to 100 per cent of the area was involved (Fig. 2). Use of this method allowed for a more complete assessment of both the extent and the pattern of heterotopic ossification48.
The group to which each hip had been randomized or assigned was not known at the time of the radiographic analysis, performed by one of us (V. D. P., Jr.). An absolute grade of heterotopic ossification was determined for each hip, on the basis of the most recent radiograph, at a minimum of six months after the index operative procedure. An end point of six months was selected because progression of heterotopic ossification rarely has been seen after three months and has not been demonstrated after six months, to our knowledge. In addition, a relative grade was established as the difference between the absolute grade and the grade that had been determined on the basis of the immediate postoperative radiograph. Residual heterotopic bone from a previous procedure was thereby acknowledged, and the relative grade therefore reflected the actual postoperative progression of heterotopic ossification. Postoperative radiographs also were analyzed to determine the status of healing of the trochanteric osteotomy after a revision operation as well as the presence of heterotopic ossification outside the designated treatment field (extra-field ossification). Patients who had radiographic evidence of marked (grade-III or IV) heterotopic ossification were examined specifically for the purpose of this study to determine any limitation of motion or functional impairment. For all other patients, clinical data were collected at routine annual follow-up examinations for the evaluation of pain as well as functional limitations in activities of daily living, such as care of the feet, perineal hygiene, ascending and descending stairs, getting in and out of an automobile, and use of assistive devices for walking24,30. Statistical analysis was performed with use of the Pearson chi-square test with Yates continuity correction. The Fisher exact test was used as appropriate for small occurrence subsets.
All eighty-six hips were available for review at a minimum of six months after the index operation. The mean duration of follow-up after the operation was thirty-five weeks (range, twenty-six to 121 weeks) for the forty-nine hips that were irradiated preoperatively (Group I) and forty-two weeks (range, twenty-six to ninety-one weeks) for those that were irradiated postoperatively (Group II). The mean duration of follow-up was longer in Group II because, as mentioned, the last fifteen patients who were entered into the study were assigned exclusively to Group I. Despite this variation, it should be noted that the minimum duration of follow-up was six months in both groups. None of the arthroplasties had failed and no reoperations had been necessary at the time of follow-up. One hip, which had been irradiated postoperatively, dislocated during the transfer of the patient to the treatment table in the Radiation Oncology Department on the first postoperative day. Closed manipulative reduction, with intravenous sedation, was successful, and no subsequent episodes of instability had occurred by the time of the most recent follow-up. No untoward systemic or local wound effects were attributable to the radiation. In Group I, radiation therapy was initiated at a mean of 3.4 hours (range, 1.3 to 6.1 hours) before the operation. Thirty-seven of the forty-nine hips in that group were irradiated within the intended four-hour window before the operation; the remaining twelve hips were irradiated within 6.1 hours before the operation because of a slight delay in the start of the operation. In Group II, radiation therapy was initiated at a mean of 29.8 hours (range, 21.6 to 51.3 hours) after the operation; thirty-two hips were irradiated on the first postoperative day and five, on the second. Thirty-seven (76 per cent) of the forty-nine hips in Group I exhibited no new heterotopic ossification (Fig. 3), and twelve (24 per cent) had progression to grade-I, II, or III ossification (Table II). Of the latter twelve hips, eleven had grade-I or II and one had grade-III ossification. The latter hip with grade-III ossification was in a woman who had previous grade-IV (ankylosing) heterotopic ossification about the hip (risk-class IV) as well as radiographic evidence of Paget disease of the pelvis. She subsequently had a total knee arthroplasty, which was complicated by extensive heterotopic ossification in the extensor mechanism with compromised function of both the hip and the knee. She had symptomatic aseptic loosening of a metal-on-metal Ring total hip prosthesis (Zimmer USA, Warsaw, Indiana). Radiographs that were made immediately after the revision of this device demonstrated residual grade-II bone, which progressed to grade-III bone at the time of the most recent follow-up. Limited-field prophylactic radiation was used for this patient because of the anticipated use, without cement, of an osseous ingrowth porous-coated prosthetic device.
Similarly, twenty-seven (73 per cent) of the thirty-seven hips in Group II did not demonstrate new heterotopic ossification and ten (27 per cent) did (p = 0.99). Of the latter ten hips, nine had only grade-I ossification and one had grade-IV ossification. The hip with grade-IV ossification was in a man who had previous grade-III heterotopic ossification about the ipsilateral hip (risk-class III) as well as Parkinson disease. The index revision total hip arthroplasty was performed because of progressive posterior instability secondary to malpositioning of the acetabular component as well as polyethylene wear. Heterotopic ossification progressed from grade III on immediate postoperative radiographs to grade IV on those made at the time of the most recent follow-up; the patient had moderate functional limitation when putting on shoes and socks. The progression of heterotopic ossification after the operation was similar in both treatment groups (Table III). In Group I, heterotopic ossification advanced one grade in eleven hips (22 per cent) and two grades in one hip (2 per cent). Of the eleven hips in which heterotopic ossification advanced one grade, seven were risk-class I, one was risk-class II, two were risk-class III, and one was risk-class IV. The risk-class-IV hip was in the previously mentioned patient who had Paget disease; heterotopic ossification in this hip progressed from grade II immediately postoperatively to grade III at the time of the most recent follow-up. The one hip in which heterotopic ossification advanced two grades (from grade 0 immediately postoperatively to grade II at the time of the most recent follow-up) was risk-class II, as previous grade-IV ossification had occurred about the contralateral hip. In Group II, heterotopic ossification advanced one grade in ten hips (27 per cent) (p = 0.99). Four of the hips were risk-class I, five were risk-class II, and one was risk-class III. The risk-class-III hip was in the previously mentioned patient who had Parkinson disease. Patients who had a history of heterotopic ossification about the hip (risk-class II, III, or IV) had a significantly greater likelihood of progression of heterotopic ossification compared with those who did not (risk-class I) (p = 0.01).
The classifications derived with use of the system of Brooker et al.7, modified to include the area of bone formation, corresponded to functional impairment. Ten hips in Group I and nine hips in Group II had progression to grade-I ossification that was class A (involving a maximum of one-third of the area about the hip) at the time of the most recent evaluation. One hip, from Group I, had progression to grade-II ossification, which also was class A. None of these hips were associated with pain or loss of mobility related to heterotopic ossification. Grade-III ossification occurred in one hip in Group I, and grade-IV ossification occurred in one hip in Group II; the ossification in both hips was class-B as it involved two-thirds of the area about the hip. Both of these hips were in patients who had some difficulty with caring for the feet and with dressing, primarily secondary to restricted rotational motion about the operatively treated hip. However, both hips had a sufficient range of flexion to permit a normal sitting posture. Extra-field ossification (the appearance of bone outside of the area included by the treatment portal) most frequently involved the trochanteric ridge at the origin of the vastus lateralis or the lateral acetabular margin superior to the prosthetic component along the outer iliac wing. Extra-field ossification occurred in fifteen (17 per cent) of the eighty-six hips, including twelve (24 per cent) of the forty-nine hips in Group I and three (8 per cent) of the thirty-seven hips in Group II (p = 0.05) (Table IV). Extra-field ossification was not associated with symptomatic bursitis of the greater trochanter in any patient. As with a history of previous heterotopic ossification about the hip, the occurrence of extra-field ossification identified a group of patients who had a diathesis for ectopic bone formation. The prevalence of previous heterotopic ossification was 31 per cent (fifteen of forty-nine hips) in Group I and 22 per cent (eight of thirty-seven hips) in Group II (p = 0.33), for an over-all prevalence of 27 per cent (twenty-three of eighty-six hips). Taken together, a history of previous heterotopic ossification, the occurrence of extra-field ossification, or new heterotopic ossification following the index procedure all serve to identify the diathesis for ectopic bone formation in the at-risk population. Over-all, twenty-eight (57 per cent) of forty-nine hips demonstrated such a diathesis in Group I, compared with thirteen (35 per cent) of thirty-seven hips in Group II (p = 0.05). Furthermore, seven (21 per cent) of the thirty-four risk-class-I hips in Group I demonstrated extra-field ossification, compared with one (3 per cent) of twenty-nine such hips in Group II (p = 0.06).
An osteotomy of the greater trochanter was performed in ten hips (nine from Group I and one from Group II). Nine of the osteotomies were performed in conjunction with the revision of a cemented femoral component and the tenth, in conjunction with a reconstruction for the treatment of congenital hip dysplasia with femoral shortening and trochanteric advancement. One additional hip, from Group II, had a revision of the acetabular component, without removal of the greater trochanter. A trochanteric non-union occurred in three of the ten hips that had had an osteotomy; all three hips were from Group I. The non-union was not associated with displacement in two of the hips (one that had been treated with a revision of a cemented femoral component and one that had had trochanteric advancement with femoral shortening), but it led to separation and proximal migration of the trochanter in the third (which had been treated with a revision of a cemented femoral component). Analysis of the radiographs that had been made at the time of the simulated treatment demonstrated that the osteotomy site had been included in the treatment field of the hip in which a non-union occurred after femoral shortening and advancement trochanteric osteotomy for reconstructive treatment of congenital dysplasia; in the other nine hips in which a trochanteric osteotomy had been performed (including the other two hips in which a non-union occurred), the osteotomy site had been excluded from the treatment field.
The present study was initiated largely because of the need to minimize uncomfortable and inconvenient transport to postoperative radiation therapy. The results suggest that preoperative radiation therapy is as effective as postoperative therapy and that it eliminates the logistical problems associated with conventional treatment in the acute postoperative period. In addition, from a basic-science perspective, the study was designed to investigate the anatomical origin and chronology of progenitor cells involved in the process of heterotopic ossification. The efficacy of preoperative irradiation suggests that these progenitor cells originate within the native tissues that are contained in the treatment field (which is centered on the operative site) rather than being recruited from a more distant source in the peripheral blood or the intramedullary cavity after the operation. This observation is consistent with the finding that heterotopic ossification occurs about the hip and elbow following a head or spinal cord injury in the absence of direct trauma and may be an important clue to our understanding of the general process of bone formation. While some non-steroidal anti-inflammatory medications may be equally effective in the prevention of heterotopic ossification, the gastrointestinal side-effects seen in approximately one-third of the patients and the issue of patient compliance after discharge contribute to suboptimum use, and occasionally to premature discontinuation, of the medication8,10,17,22,53,58,59,63,68. The anti-platelet effects of these medications further complicate their use with warfarin, currently the most commonly employed thromboprophylactic agent, in the postoperative setting34. Most importantly, the systemic inhibitory effects of anti-inflammatory medication on osseous formation and growth into prostheses inserted without cement are of both theoretical and practical concern38,69. The ability to localize these inhibitory effects on osteogenesis precisely make prophylaxis with radiation an especially attractive modality in the setting of total hip arthroplasty performed without cement35,38. Prophylaxis with radiation is not without disadvantages5,56. Perhaps most important, and as yet not completely resolved, is the oncogenic risk imposed by radiation therapy6,26,56. Kim et al.31 observed no bone or soft-tissue sarcomas after the administration of less than 3000 centigray over a three-week period, and Tucker et al.70 reported no radiation-induced sarcomas of bone in patients who had been followed for as long as twenty-five years after the administration of less than 1000 centigray for the treatment of childhood cancers. However, malignant neurotumors have been seen, after a latency period of fifteen to twenty-four years, following the administration of as little as 1000 centigray to the head in children55, and malignant hematological tumors have been associated with the administration of low doses (less than 1000 centigray) of radiation to active bone marrow4,11,28. The advanced age of most patients who are managed with total hip arthroplasty, and the filling of the proximal part of the medullary canal with a prosthetic stem, minimize the number of active marrow elements subject to direct irradiation. However, as the latency period for the induction of malignant disease following irradiation is fifteen to twenty-four years55, patients who have been managed with radiation therapy specifically for prophylaxis against heterotopic ossification have not been followed for long enough to answer this question definitively60. Irradiation specifically for prophylaxis against heterotopic ossification was first described12 in 1981, to our knowledge. Consequently, sufficient follow-up data will not be available for analysis until the next decade. More practical disadvantages associated with irradiation for prophylaxis against heterotopic ossification include the involvement of an additional patient-care service, the transport of the patient in the early postoperative period, and the discomfort of the patient during transport and during transfer from one table to another in the radiation-therapy suite. Indeed, during the two years of the present study, the only postoperative dislocation of a hip occurred during the transfer of a patient from the hospital bed to the treatment table for prophylactic irradiation. The patient had no subsequent difficulties with instability after this isolated incident, and there was no radiographic evidence of malpositioning of the component at the time of the most recent examination. Preoperative treatment, potentially even before the patient is admitted to the hospital, encourages outpatient planning and the simulation of radiation therapy, which is a more predictable, convenient, and efficient use of time and personnel in the radiation oncology department. The patient's understanding of the process is enhanced by preoperative education in a setting in which he or she is alert and undistracted by perioperative pain. Although simulation and treatment in the radiation oncology department, for which the hospital charge is $700 to $1000, is considerably more expensive than ten days of treatment with indomethacin, radiation is effective and is essentially never complicated by the gastrointestinal intolerance that is associated with non-steroidal agents. Non-steroidal agents also pose a costly, albeit small, risk of intestinal bleeding. The current study demonstrates the comparable efficacy a single 800-centigray dose of limited-field radiation delivered either within approximately six hours before or forty-eight hours after a total hip replacement. Although, as far as we know, this is the largest prospective study to date in which the use of any form of prophylactic irradiation for the prevention of heterotopic ossification has been investigated, the sample size precludes determination of whether this result is significant. A power study suggests that, for conventional standards of alpha equaling 0.05 and beta equaling 0.20, 215 hips would be required in each treatment group in order for a 10 per cent difference in the prevalence of heterotopic ossification to be considered significant. The establishment of the equivalency of two regimens, as is suggested by the present study, would require an even greater number of hips to satisfy statistical considerations15,40. If the rate of progression of heterotopic ossification in Group II of the present report (ten of thirty-seven hips) is pooled with that in a group of comparable patients described in a previous report (seven of thirty-four hips)48, the over-all rate of progression after the postoperative administration of a single 800-centigray dose of limited-field radiation becomes 24 per cent (seventeen of seventy-one hips), but significance still is not attained (p = 0.88). Practical constraints on the utilization of hospital services, the decreasing duration of hospitalization, and the emphasis on moving inpatient care to the outpatient setting all contributed to the impracticality of completing a prospective study of such a large size. However, the current data are clinically relevant in the context of previous studies and the known results of alternative methods of prophylaxis against heterotopic ossification1,8,32,37,44,48,59,66. The rate of progression of heterotopic ossification in Group I (24 per cent; twelve of forty-nine hips) compares favorably with that in Group II (27 per cent; ten of thirty-seven hips) as well as with that in two groups of patients, described in a previous report48, who were managed with either a single-fraction dose of 800 centigray (21 per cent; seven of thirty-four hips) or a conventionally fractionated dose of 1000 centigray (21 per cent; six of twenty-eight hips) delivered postoperatively to a limited field. Both the present and the previous study included high-risk populations in which a large proportion of subjects had had heterotopic ossification after a previous operation about the hip. In the current series, 31 per cent (fifteen) of the forty-nine hips in Group I and 22 per cent (eight) of the thirty-seven hips in Group II had previous heterotopic ossification, and nearly half of the patients in the series demonstrated a diathesis for ectopic bone formation of some kind. Indeed, Group I included significantly more hips in patients who exhibited a diathesis for heterotopic ossification than did Group II (twenty-eight [57 per cent] of forty-nine hips compared with thirteen [35 per cent] of thirty-seven hips; p = 0.05). Furthermore, the risk of progression of heterotopic ossification, albeit functionally inconsequential in most patients, was significantly greater when there had been previous heterotopic ossification (risk-classes II, III, and IV) than when there had not (risk-class I) (48 per cent [eleven of twenty-three hips] compared with 17 per cent [eleven of sixty-three hips]; p = 0.01). These observations strengthen the findings of the present study that support the efficacy of preoperative irradiation. They also validate the predictive value of the risk-classification system based on the previous occurrence of heterotopic ossification of increasing severity in both the contralateral and the ipsilateral hip. The prevalence of extra-field ossification in the present study (17 per cent; fifteen of eighty-six hips) compares favorably with that reported previously in a similar prospective investigation (31 per cent; nineteen of sixty-two hips)48. In that series, three of the nineteen hips that had extra-field ossification about the greater trochanter were treated with repeated local injections of steroids because of painful bursitis. The simple rectangular treatment portal that was used in that study was modified to the currently employed L-shaped portal in order to include the area about the greater trochanter, where most extra-field ossification occurs. It is likely that this modification contributed meaningfully to the reduction in the occurrence of extra-field ossification and the elimination of the associated problem of bursitis of the greater trochanter in the present study. It is also interesting to note that the prevalence of extra-field ossification in the current study was significantly greater in Group I (24 per cent; twelve of forty-nine hips) than in Group II (8 per cent; three of thirty-seven hips) (p = 0.05). It can be hypothesized that preoperative placement of the treatment portal may decrease the accuracy with which the greater trochanter is included in the field; however, this difference was not clinically apparent in the present series. Additional study of the relationships among extra-field ossification, the extended treatment portal, and the preoperative delivery of radiation therapy is necessary to clarify this phenomenon. Several authors have noted that irradiation for prophylaxis against heterotopic ossification is associated with an increased risk of non-union at the site of an osteotomy of the greater trochanter1,2,12,18,37,44,47,48. The greater trochanter failed to heal in approximately one-third of the hips in those reports, irrespective of whether radiation had been delivered to a full, expanded treatment field or to a limited field that excluded the osteotomy site. We previously hypothesized that poor bone stock in the proximal part of the femur in the setting of revision arthroplasty probably is responsible for this increased rate of trochanteric non-union48. In the current series, three of ten osteotomy sites did not heal, once again independent of the fact that the limited treatment field excluded the osteotomy site from direct irradiation. Nine of the ten trochanteric osteotomies were performed to facilitate the removal of a cemented femoral stem during a revision arthroplasty. Two of the three non-unions occurred in these revised hips; the third non-union was noted radiographically in a patient managed with femoral shortening and trochanteric advancement after prosthetic arthroplasty for degenerative osteoarthrosis secondary to congenital dysplasia. The primary factor in the observed rate of trochanteric non-union associated with radiation therapy5 still appears to be poor bone stock in the proximal part of the femur leading to non-rigid fixation of the trochanter in patients who have been managed with a revision procedure. Two hips in the current series demonstrated clinically meaningful heterotopic ossification after a previous operation; both subsequently were treated with a revision arthroplasty, and both demonstrated postoperative progression of heterotopic ossification by one grade despite prophylaxis with radiation. One of these hips was in the woman who had Paget disease of the pelvis, and the other was in the man who had Parkinson disease; both patients had considerably compromised function of the hip at the time of the follow-up evaluation. While it is difficult to determine an exact prevalence of heterotopic ossification in these two unique patient populations because of difficulty in determining the denominator for the groups in question, our experience is consistent with the observation that patients who have Paget disease of bone about the hip and pelvis and those who have Parkinson disease exhibit extensive heterotopic ossification after an operation about the hip. We now consider these patients to be at increased risk for heterotopic ossification because of a poorly understood systemic diathesis. The clinical demonstration of the efficacy of preoperative irradiation in the prevention of heterotopic ossification after total hip replacement will have a meaningful impact on current orthopaedic practice. The demonstrated efficacy of preoperative irradiation, combined with the desirability of transferring service to the outpatient setting, of increasing patient comfort and understanding, and of minimizing the risk of postoperative complications, have led to the adoption of this technique as our routine protocol for prophylaxis against heterotopic ossification about the hip. However, the exact timing of this treatment remains in question. In the current series, twelve (24 per cent) of the forty-nine hips in Group I were not irradiated within the intended four-hour interval before the operative procedure because of an unanticipated delay in the start of the operation. Nonetheless, all hips were retained in the study for data analysis, and, ironically, the prevalence of heterotopic ossification was more than three times greater in the group that had been irradiated within four hours before the operation (eleven of thirty-seven hips; 30 per cent) than that in the group irradiated between four and approximately six hours before the operation (one of twelve hips) (p = 0.25). While the small sample size precludes this relationship from being significant, the time at which pluripotential mesenchymal cells are most sensitive to prophylactic irradiation for the prevention of differentiation along osteogenic cell lines has yet to be determined and clearly is deserving of additional study9,13,43,67. However, one very important implied message from the apparent efficacy of preoperative radiation is that the osteogenic precursor cells active in the process of heterotopic ossification can be presumed to be derived from the local tissues within the operative field rather than from distant blood-borne cell lines. Therefore, in addition to improving the predictability of clinical outcome after total hip arthroplasty, the study of preoperative irradiation in the prevention of heterotopic ossification contributes to our understanding of the fundamental control mechanisms responsible for the regulation of the processes involved in the formation of bone.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |