Physician Prescribing Patterns of Oral Corticosteroids for Musculoskeletal Injuries

Methods

To obtain a description of the corticosteroid-prescribing patterns of sports medicine physicians, a two-page questionnaire was included in the registration packets of a national sports medicine meeting. Permission for distribution of the questionnaire was obtained from the sponsoring organization. This conference was attended by family physicians, internists, pediatricians, emergency medicine physicians, and physiatrists with interest and expertise in the practice of sports medicine. None of the attendants was an orthopedic surgeon. Personal identity of the surveyed physicians was not solicited. Limited demographic data were collected to assess prescription patterns of physicians in different regions, specialties, and experience levels. No data were collected from nonresponders. Completion of the survey was entirely voluntary.

Physician oral corticosteroid prescribing frequency for musculoskeletal injuries was collected using open-ended questions based on average prescriptions written per month. Survey participants were also asked to examine a list of both acute and chronic sports medicine injuries and indicate whether they would consider prescribing oral corticosteroids for these injuries. An open-ended response was also included for other injuries that might not have been included on the questionnaire. Physicians who prescribe oral corticosteroids were asked to indicate on a 6-point Likert scale the relative importance of various factors in their decision of whether to use oral corticosteroids. A 0 score was assigned to an unimportant factor and a 5 was assigned to a very important factor. Physicians who did not prescribe oral corticosteroids were asked to rate the relative importance of a variety of factors that they used to make their decision against corticosteroids on a similar Likert scale. The mean scores for each factor were calculated and ranked to determine their relative importance in the physicians’ decision-making process. A descriptive analysis was undertaken of the results.

Results

One hundred ninety-five questionnaires were included in registration packets given at the start of the conference. Ninety-nine were completed for a 51% return rate. Fifty-eight (58.6%) of the physicians reported ever prescribing oral corticosteroids for musculoskeletal injuries (Table 1). Physicians who prescribed corticosteroids for injuries averaged 6.6 prescriptions per month. Prednisone was the corticosteroid used by 82% of physicians. The average prescription length was 7 days. One half (51.7%) of the physicians tapered the dose. The most common starting dose (mode) was 60 mg.

There was no consensus among corticosteroid-prescribing physicians on indications for use. Approximately one third of physicians reported prescribing corticosteroids for acute conditions, one third for chronic conditions, and one third for both acute and chronic conditions (Table 2).

Those who received additional fellowship training in sports medicine or those holding a certificate of added qualifications in sports medicine appeared to be more likely to prescribe corticosteroids for nondiscogenic musculoskeletal conditions than did those who did not have additional training. There was a trend toward increased oral corticosteroid use with more practice experience (Table 3).

Among physicians who used corticosteroids to treat musculoskeletal injury, severity of injury was the primary factor in decision to prescribe. Among physicians who do not use oral corticosteroids, lack of medical literature supporting efficacy was the primary reason given for not prescribing (Table 4).

Discussion

There is limited research into the use of oral corticosteroids for nondiscogenic back pain. In three randomized, double-blind, placebo controlled studies, patients with carpal tunnel syndrome showed temporary benefit from oral corticosteroids.10–12

There are no human studies evaluating the use of oral corticosteroids in acute injuries, such as contusion, strains, or sprains. Some animal studies, however, have evaluated systemic steroids (intramuscular injection) for contusions.13 ,14 The steroids were given in dosages higher than those typically given to humans and were continued for longer durations. Histologic studies of injured tissue showed a decrease in initial inflammation but eventual decreased tensile strength of the affected tissue with the use of corticosteroids compared with controls. Direct injection of corticosteroids into a muscle defect after a strain is sometimes used in elite athletes and has been retrospectively reviewed.15 Results appeared promising; however, there was no control group and outcome measures were subjective.

Some authors advocate use of oral corticosteroids for adhesive capsulitis, but this recommendation is not based on any studies.16 ,17 The use of corticosteroids for chronic tendonitis is suspect, because this condition is degenerative rather than inflammatory.2 ,18

Corticosteroids have been prescribed extensively for a multitude of conditions, and their side effects are well documented. Serious side effects are most likely with prolonged or high-dose use. There is little documentation of incidence of side effects or risk with short-term (less than 1 week) and medium-dose (about 1 mg/kg) use of corticosteroids. The most common side effects with short-term, medium-dose steroids are changes in appetite and sleep pattern.19

Physicians that did not prescribe oral corticosteroids for musculoskeletal conditions might have been more likely not to complete and return the survey. Even so, it is important to note that many sports medicine physicians are prescribing oral corticosteroids, and there is no consensus on indications for their use. The demographic profile of the physicians surveyed was fairly narrow. Meeting attendees were primary care physicians with expertise in sports medicine. Prescribing patterns of orthopedic surgeons or physicians who do not practice much sports medicine might vary.

Conclusion

The use of oral corticosteroids as treatment of sports injuries is common. There is scant literature to support steroid use for most sports injuries. Animal studies that have examined the effect of corticosteroids on healing are not strictly applicable to human injury models. Injuries, dosages, length of treatment, and routes of administration might not be typical of the human sports injury. Because of the widespread use of oral corticosteroids and the anecdotal support they receive from clinicians, more research should be conducted on the efficacy of corticosteroids in different types of injuries. Serious side effects are possible but unlikely given typical dosing patterns.