Osteoid osteoma is a benign bone tumor that is usually found in children and young adults. It produces pain that characteristically responds to treatment with non-steroidal anti-inflammatory drugs. Patients who have an osteoid osteoma usually are managed by an orthopaedic surgeon, but they may initially be seen by a pediatrician, family practitioner, or rheumatologist.
The growth potential of the tumor is limited. There may be considerable inflammation surrounding the tumor, but the tumor itself (the nidus) rarely exceeds 1.5 centimeters in diameter. There is no malignant potential. Although there is some evidence that the lesion may regress spontaneously22, it is usually removed because patients are unwilling to tolerate the pain and wish to avoid the long-term use of non-steroidal anti-inflammatory medications. Prolonged presence of the tumor may lead to complications such as growth disturbance, scoliosis, and osteoarthrosis15,17.
Despite the small size of the lesion, the operative procedure for its removal can be extensive. The tumor may be difficult to identify at the time of the operation, and it may recur if it is not removed completely. The tumor has a propensity for the major weight-bearing bones, especially the femur. There is a risk of fracture at such sites if a substantial amount of cortical bone is removed, and therefore internal fixation and bone-grafting may be necessary. The location within a given bone is variable, and the operative approach can be difficult.
Improved methods for the precise localization of an osteoid osteoma with use of radioisotope scanning11,20,25 and computed tomography26-28,40,45 have made it possible to treat this lesion with more limited operations4,14,18,43, arthroscopic removal36, and percutaneous removal2,3,8,21,23,29,32,36,41,42.
The method of treating osteoid osteomas with thermocoagulation with use of a percutaneously placed electrode has been previously described by us37,38 and others7. A minimum amount of bone is removed during this procedure, and the patient can return to normal function almost immediately. The technique is one of the growing number of minimally invasive alternatives to conventional operative procedures. We studied the impact of this percutaneous method on the management of patients at our institution by comparing the length of the hospital stay, the rate of complications, and the clinical outcome with those associated with operative excision.
*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.
†Departments of Radiology (D. I. R.) and Orthopaedics (F. J. H., M. W. W., L. C. J., M. C. G., and H. J. M.), Massachusetts General Hospital, 32 Fruit Street, Boston, Massachusetts 02114. E-mail address for Dr. Rosenthal: rosenthal.daniel@mgh.harvard.edu.
All patients who had been managed at our institution between 1978 and 1995 because of an osteoid osteoma were identified through a search of operative and radiological databases, and the records were reviewed.
A patient was included in the study if definitive treatment of the osteoid osteoma had been carried out at our institution and if the patient had been followed for at least two years. The study included patients who had been managed for a primary lesion as well as those who had been managed for a recurrent lesion. Patients who had been managed initially for a recurrent lesion at our institution were considered separately from those who had had a recurrence after the index treatment. A patient was excluded if another diagnosis had been made by the pathologist on examination of the operative specimen. However, if the surgeon or the radiologist had believed that the lesion was an osteoid osteoma at the time of treatment, the patient was included, even if the pathologist had not been able to identify the nidus (the histological findings were negative). Procedures that had been done to treat complications were recorded separately. Patients who had had a spinal lesion were excluded, as we are reluctant to treat such tumors with thermocoagulation.
The age and gender of the patient, the location of the tumor, the date and type of the procedure, and the length of the hospital stay (in days) were recorded. The length of the hospital stay referred only to the time needed for treatment of the lesion; any additional hospitalizations that were needed for the treatment of complications were recorded separately. The histological findings were noted.
A questionnaire regarding the current clinical status was sent to the last known address of each patient to supplement the information found in the medical records. The patients were asked to report on any additional procedures that had been performed elsewhere as well as on symptoms and medications. The patients' responses were reviewed by three individuals, two of whom (F. J. H. and M. W. W.) were orthopaedic surgeons who had not participated in either the operative procedures or the percutaneous ablations. The third individual (D. I. R.) was a radiologist. When a response was ambiguous, the patient was contacted by telephone.
The treatment was classified, by consensus, as clinically successful only if the patient was free of pain and was taking no medications. The treatment was considered to have failed if a subsequent procedure had been performed to remove tumor or if the symptoms had not improved. If the symptoms were different in nature or less severe, or both, the status was considered to be improved.
The method of treatment with radiofrequency has been described elsewhere7,37,38. In brief, a conventional bone-biopsy needle is percutaneously introduced into the lesion under the guidance of computed tomography. A sample of tissue is obtained, and then a radiofrequency electrode with a five-millimeter exposed tip is introduced through the cannula of the biopsy needle (Figs. 1-A, 1-B, and 1-C). The electrode is connected to a commercially available radiofrequency generator (Radionics, Burlington, Massachusetts) that simultaneously monitors the temperature at the tip of the electrode. The power output is adjusted to gradually bring the temperature of the tip to 90 degrees Celsius. This temperature is maintained for six minutes, during which time a sphere of tissue approximately one centimeter in diameter is ablated through thermal conduction. The radiofrequency treatment is thus not specific for osteoid osteoma and will ablate any tissue within the small range of the electrode. Although the procedure usually is done on an outpatient basis, it is most often done with the use of general anesthesia; a spinal or local anesthetic sometimes is used if requested by an adult patient. The patient is allowed to bear weight on the involved extremity immediately after the procedure. No crutches or external supports are used, and pain medication usually is needed only for the first twenty-four hours after the procedure.
Statistical Methods
All of the analyses were performed with use of a commercially available statistical software package (SAS, Cary, North Carolina). The Fisher exact test (two-tailed) was used to compare the data on outcome for the patients who had had operative treatment and those who had had percutaneous treatment. Linear regression analysis was used to determine the effect of the date of treatment on the length of the hospital stay.
A total of 125 patients met the criteria for inclusion in the present study. Operative excision was performed in eighty-seven patients: sixty-eight who had a primary lesion and nineteen who had a recurrent lesion. Most of the patients in the latter group had the initial treatment elsewhere. Percutaneous ablation with radiofrequency was performed in thirty-eight patients: thirty-three who had a primary lesion and five who had a recurrent lesion. The operative procedures were performed by fifteen different surgeons, but most (seventy-one) were performed by only three surgeons, including two of us (M. C. G. and H. J. M.). All of the percutaneous procedures were performed by one radiologist (D. I. R.).
Treatment of Primary Lesions
Sixty-eight patients were managed with operative excision of a primary lesion. Forty-six of these patients were male and twenty-two were female; the average age was twenty-two years. The average length of the hospital stay was 4.7 days, and the average duration of follow-up was almost nine years.
Six (9 per cent) of the sixty-eight patients returned for an additional procedure. Five patients had a second operation (two also had a third operation), and one had percutaneous ablation. Three of these patients were male and three were female; the average age was twenty-three years. The average interval between the first and second procedures was one year (range, 0.4 to 1.4 years).
Thirty-three patients were managed with percutaneous ablation of a primary lesion. Twenty-six of these patients were male and seven were female; the average age was twenty-three years. The average length of the hospital stay was 0.18 day; four patients stayed in the hospital for one night because of symptoms related to anesthesia, and one patient stayed for two nights. The average duration of follow-up was 3.4 years.
Four (12 per cent) of the thirty-three patients returned for an additional procedure. Two patients were managed with a repeat percutaneous ablation and two, with an operative procedure. Three of these patients were male and one was female; the average age was nineteen years. The average interval between the first and second procedures was thirteen months (range, one to twenty-two months).
With the numbers available for study, we could detect no significant difference, with regard to the rate of recurrence, between the patients who had been managed with an operation and those who had been managed with percutaneous ablation (p = 0.725, Fisher exact test).
The diagnosis was confirmed by biopsy at the time of the operative treatment in fifty-two patients and at the time of the percutaneous ablation in twenty-five patients. When only these patients were considered, an additional procedure was necessary for four (8 per cent) of the patients who had been managed with an operation and for two (8 per cent) of the patients who had been managed with a percutaneous procedure. With the numbers available, we could detect no significant difference between these two groups with respect to the rate of recurrence (p = 0.577, Fisher exact test).
Treatment of Recurrent Lesions
Nineteen patients were managed with operative excision of a recurrent lesion. Fourteen of these patients were male and five were female; the average age was twenty-four years. The average length of the hospital stay was 5.1 days. Of the nineteen patients, two (both of whom were female) had a subsequent operation for recurrence.
Five patients were managed with percutaneous ablation of a recurrent lesion. Two of these patients were male and three were female; the average age was twenty-two years. None of these five patients had additional intervention because of subsequent recurrence.
Subsequent Recurrence
Overall, additional intervention (either operative treatment or percutaneous ablation) was necessary for eight (9 per cent) of the eighty-seven patients who had been initially managed with operative excision and for four (11 per cent) of the thirty-eight patients who had been initially managed with percutaneous ablation. With the numbers available for study, we could detect no significant difference between these two groups with regard to the overall rate of recurrence (p = 0.750, Fisher exact test). The results for the patients who had had treatment of a primary lesion were comparable with those for the patients who had had treatment of a recurrent lesion.
Complications
A total of five secondary procedures were performed for two patients because of complications after en bloc excision of the tumor. One patient had arthroscopy and subsequent repair of the posterior cruciate ligament, and the other had exploration of the sciatic nerve, transfer of the peroneal tendon, and subsequent revision of the tendon transfer.
Clinical Success
The questionnaire regarding the current clinical status was returned by fifty-three (42 per cent) of the 125 patients. The response rate was particularly low for the patients who had been managed operatively, probably because of the long time-interval since many of the operative procedures. None of the patients reported having had an additional procedure elsewhere. Therefore, we believe that our records represent the complete treatment experience of this group of patients.
Twenty-six patients who had had percutaneous treatment with radiofrequency returned the questionnaire. Twenty patients (77 per cent) reported that they were completely free of pain; eighteen of these patients were male and two were female. The remaining six patients, three of whom were male and three of whom were female, reported that the symptoms had not improved.
Twenty-seven patients who had had operative excision returned the questionnaire. Nineteen patients (70 per cent) reported that they were completely free of pain; fourteen of these patients were male and five were female. The treatment had failed in four patients (all of whom were female), and the symptoms had improved in four patients (two of whom were male and two of whom were female).
With the numbers available for study, we could detect no significant difference, with regard to the rate of clinical success, between operative treatment and percutaneous treatment with radiofrequency (p = 0.722, Fisher exact test).
At our institution, the minimally invasive procedure of percutaneous ablation with radiofrequency has progressively replaced operative excision as the preferred treatment for osteoid osteoma of the long bones or the pelvis (Fig. 2). This change in clinical practice is due to the fact that the percutaneous procedure requires fewer health-care resources and is associated with a shorter period of convalescence. However, the relative efficacy of the two approaches has not previously been evaluated, to our knowledge.
We could detect no significant difference between the two procedures with respect to the rate of recurrence. Although true recurrence after complete removal of the tumor30,44 and the existence of more than one nidus1,5,12,16,31 have been reported, it is generally believed that recurrence is most often due to incomplete removal of the tumor15,25,30,34. For this reason, symptoms usually occur early after treatment, but a late return of symptoms, even after many years, has been reported30,33.
Because most of the recent procedures were performed percutaneously, the average duration of follow-up for the patients who had been managed with radiofrequency was shorter than that for the patients who had been managed with operative excision. The time-interval between the initial procedure and the recurrence of symptoms was not known in the present study. However, all of the second procedures were performed within two years after the initial procedures, which indicates that the symptoms had recurred within that interval. We therefore believe that the minimum duration of follow-up of two years is adequate for comparison of the two groups. While the rate of recurrence probably would be slightly higher after a longer period of follow-up, it is improbable that the difference would be sizable.
We wished to understand the probability of a successful outcome of each procedure in the context of the clinical presentation of a suspected lesion. Therefore, we studied patients in whom the diagnosis had been confirmed histologically as well as those in whom it had not. Because the small size of an osteoid osteoma makes it difficult to identify or remove the nidus—resulting in a poor outcome for many patients—the inclusion of only patients in whom the diagnosis has been verified histologically may introduce an outcome bias favorable to treatment. As expected, the apparent rate of recurrence decreased when we evaluated only the patients in whom the diagnosis had been confirmed by biopsy at the time of the operation.
Furthermore, the inclusion of patients in whom the diagnosis has not been confirmed histologically is supported by studies that have suggested that the removal of the surrounding sclerotic bone (so-called unroofing) or other intralesional procedures may relieve symptoms in a substantial proportion of patients, even when the nidus has not been identified30,49. As an explanation of this observation, it has been suggested that perhaps the nidus was removed by the surgeon but was not recognized or sent for histological examination, that perhaps the nidus was present in the specimen but was overlooked by the pathologist, or that perhaps the specimen was damaged by handling so as to render the nidus unrecognizable39. Some lesions have appeared to heal spontaneously after operative treatment that did not include removal of the tumor9,39.
The rate of recurrence (or persistence) of the tumor and of tumor-related symptoms is unclear because most series in the literature have been relatively small, the duration of follow-up has varied, and patients who have not had histological confirmation of the lesion often have been excluded. In one series from the Mayo Clinic, there were no recurrences in ninety-five patients in whom the diagnosis had been confirmed histologically6. Jackson et al.19, in an extensive review of the literature, reported a rate of recurrence of 4.5 per cent (thirty-nine of 860) but noted that this rate was based on those reported in many different studies, with no uniformity or minimum standard for the duration of follow-up. Kohler et al.24 reported on a series by Bouyala in which the rate of recurrence was 10 per cent, but the number of patients was not provided. Norman30 reported recurrence in nine (12 per cent) of seventy-three patients. In that study, the nidus had not been identified at the time of the operation in seven of the nine patients who had a recurrence. Golding13 reported a recurrence of symptoms in five (25 per cent) of twenty patients. Freiberger et al.10 reported that five (6 per cent) of eighty patients needed more than one procedure. The rate of recurrence in our series thus falls within the range of rates that have been reported by others.
An important finding of our study is that the frequency of persistent symptoms was considerably higher than the rate of recurrence; six (23 per cent) of twenty-six patients who had had percutaneous ablation and eight (30 per cent) of twenty-seven who had had operative excision had persistent symptoms. Although some of the patients who had persistent or recurrent symptoms may have had undocumented recurrence of the tumor, others may have had pain related to the procedure.
One reason for the relative success of the percutaneous method is that the procedure is performed under the guidance of computed tomography, which ensures accurate placement of the electrode. Improved identification of the lesion preoperatively or intraoperatively has been shown to decrease the prevalence of recurrence after operative treatment25,28.
Our results demonstrate that percutaneous ablation with radiofrequency is essentially equivalent to operative excision for the treatment of both primary and recurrent osteoid osteoma of an extremity. Percutaneous ablation with radiofrequency is the preferred treatment for extraspinal osteoid osteoma because hospitalization is unnecessary, it has not been associated with complications, and it has been associated with a shorter period of recovery than has operative treatment.