Anserine bursitis is a common clinical problem,1 also labeled as pes anserine tendinitis or simply “medial knee pain,” most commonly diagnosed by history and examination, being characterized by tenderness on the upper-medial tibia at a point 3–5 cm distal to the medial knee joint line. There are no validated diagnostic criteria for anserine bursitis, however, and the validity of the clinical maneuver (palpation) used to establish the diagnosis has not been assessed. Because many of these patients have medial compartment osteoarthritis or tenderness along the medial collateral ligament, clinicians often refer to these patients as having “medial knee pain” or “anserine syndrome,” without a label that implies specific pathology.2,3 The mechanism of onset of this disorder is unclear, but diabetes, obesity, and osteoarthritis of the knee are risk factors.2,3 Although there is evidence that physiotherapy and local corticosteroid injections are effective,2,3 the author’s clinical experience indicates that recurrences are common and response to treatment is incomplete. Given the common recurrence of medial knee pain labeled as anserine bursitis and limited interventions available, long-term effective strategies are needed.
No trials of foot orthoses exist in clinical populations identified as having anserine bursitis. Cambron et al.4 measured the change in Oswestry Disability Index scores in subjects with nonspecific chronic low-back pain at the end of 6 weeks of treatment with foot orthoses as compared with no orthosis treatment. A number of these subjects had lower-limb pain, but they were not specifically examined for findings typically associated with anserine bursitis. An uncontrolled trial5 has shown reductions in lower-limb pain with orthosis use but again was not in a cohort of subjects with anserine bursitis. The purpose of this study was to determine the proportion of subjects reporting recovery from anserine bursitis using prescribed custom-made foot orthoses in addition to local corticosteroid injection versus corticosteroid injection alone.
For the purpose of this study, a case definition of anserine bursitis was established by the presence of four criteria: 1) the presence of pain elicited on palpation of the pes anserine insertion; 2) absence of pain elicitation on palpation at a site 3 cm away from this site or the medial joint line; 3) absence of clinically detectable knee effusion; and 4) no other clinically evident, obvious reason for knee pain. The case definition was designed to mimic clinical practice because these patients are seldom referred for diagnostic imaging but rather given empirical therapy and assessment of the response before considering further investigation. The palpation method was based on the author’s clinical practice of manual palpation to a pressure of approximately 4 kg with the pulp of the thumb. Osteoarthritis is not an exclusion to this case definition, so long as clinically evident effusion was not present (i.e., osteoarthritis as evident on radiograph is common in this clinical population, and it would be difficult to find cases of “pure” anserine bursitis in practice).
During a period of nearly 4 months in early 2009, the author was acting as a consultant in two nearby primary care clinics in Edmonton, Alberta, Canada. Both clinics served similar clinical populations near the inner city, mainly lower-socioeconomic and worker populations. Primary care physicians referred patients with musculoskeletal disorders to the author. At one clinic, there were the space and the administrative support to provide custom-made foot orthoses immediately at the time of the initial consultation visit. At the second clinic, these were not available for at least 3 months. The author had been routinely collecting data using the Oswestry Disability Index in these types of patients as well as data concerning current medication use, typically reassessing patients at 8 weeks after consultation. Thus, the circumstances between these two clinics and the available data in all patients seen in follow-up provided an ideal opportunity for a cohort-controlled trial of custom-made foot orthoses, maintaining consistency of treatment approaches to the problem of anserine bursitis. That is, the same consultant was involved at both patient care clinics, and in some cases, the same primary care physician was involved in care at both clinics. Data were collected by the author in both practices; approval was obtained from the College of Physicians and Surgeons of Alberta as part of an ethics approval for practice audit of the author’s various clinical populations.
Subjects were recruited from a sample of consecutive patients presenting to their family physician with pain in the region of the medial knee and who were further suspected by the family physician to have anserine bursitis, on the basis of location of pain and point tenderness. These subjects were routinely referred to the author for care. Prospective subjects were further assessed for inclusion and exclusion criteria at the time of initial interview. Subjects included for study met the following criteria: older than 17 years, able to read and write in English at the grade 8 level or higher, and meeting the case definition for anserine bursitis. Exclusion criteria included current use of custom-made foot orthoses, neurological disorder (including sciatica with objective neurological signs), cancer, spinal stenosis, spinal or lower-limb surgery, recent or complicated fracture, known inflammatory arthropathy, severe osteoarthritis of the lower-limb joints, prosthetic joints, amputation, or congenital lower-limb deformity. Mild to moderate lower-limb joint osteoarthritis was not an exclusion because this is a common condition in primary care, especially in older patients. Subjects in the control group were excluded if they obtained a foot orthosis during the study period. Patients were also excluded if they had pain above the midthigh level for any reason for more than 1 day per week (i.e., if they had another pain, it was relatively minor compared with the knee region pain). Thus, these cohorts represented anserine bursitis with or without other higher-level lower-limb or low-back pain. These cohorts reflect typical primary care patients in their presentation with anserine bursitis. Referring physicians were aware of these criteria to refer patients who were likely to meet the criteria on evaluation by the author.
INTERVENTIONS AND GROUPS
The author, after an appropriate history and physical examination and confirmation of a case definition of anserine bursitis, prescribed a local corticosteroid injection to take place within 1 week at a local radiology clinic, with 40 mg of Depo-Medrol injected into the region of the bursa, unilaterally or bilaterally, as the subject required. The injection protocol and procedure were determined by the radiology clinic. The author also provided advice to continue follow-up with the primary care physician for analgesics. The author and the primary care physicians at the clinic, as part of their usual practice, routinely avoided prescribing other modalities for anserine bursitis, such as acupuncture or massage therapy, but allowed patients to seek these out if they so desired. The primary care physicians also had a typical medication regimen that included either nonsteroidal anti-inflammatory drugs or acetaminophen products, although the basis for this prescription is unknown. They tended not to use narcotics or sedatives, although the number of subjects who received these is unknown. These patients already had appropriate investigations to rule out fractures, radiculopathies, and other nonbenign causes of knee region pain but did not have investigations for internal derangement unless there was a clinical indication.
After assessment, each subject completed the Oswestry Disability Index. The author then obtained foam impressions of each patient in the seated position, with the patient asked to relax his/her lower-limb and allow the examiner to place downward pressure on the knee along the axis of the tibia with the knee flexed to 90°. This created the foam impression. All subjects received a standardized orthosis, a Footmaxx® Premium Allsport orthosis (Footmaxx, Toronto, Ontario, Canada), composed of a semirigid module, vinyl reinforcement, polyurethane foam cushioning, and large-profile metatarsal pad aligned with the third metatarsal ray and placed distal to the semirigid module. No thickness alteration was made for weight or activity. The distal part of the semirigid module ended before the metatarsal heads, and the dimensions of the metatarsal pad were those of an isosceles triangle with a base approximately 5 cm (2 in) and a height of 6 cm (2.25 in). The thickness of the metatarsal pad was approximately 0.6 cm (0.25 in) at its maximum.
The subjects were provided lifts for correction of leg-length discrepancies if measured to be 1.5 cm or greater, and the lift was provided at a measure of half this discrepancy. None of the subjects received valgus or varus postings. The subjects received their orthoses within 1 week of the molding, and all received the same general instruction on usage and footwear. Each subject receiving orthoses received an instruction sheet provided by Footmaxx (Toronto, Ontario, Canada). In brief, the subjects were advised to wear the orthoses for 1–2 hrs on the first day and to increase use by 1–2 hrs daily. They were advised to wear runners or work boots as appropriate if possible, but in most cases, the subjects had chosen their own footwear, and no data are available on this.
For the purposes of creating a cohort for analysis, the patients who met the inclusion and exclusion criteria and who received the prescribed custom-made foot orthoses were deemed to be the orthoses group. When reassessed by the author at 8 weeks, the subjects who reported ongoing or recurrent symptoms were referred for repeat corticosteroid injection.
These subjects were taken from patients consecutively referred to the author at the nearby second primary care clinic. Again, the primary care physicians (some of whom had worked in both clinics) were asked to refer patients with suspected anserine bursitis. The author undertook the same clinic activities as stated above for the orthoses group including the completion of the Oswestry Disability Index and referral for a local corticosteroid injection. Because orthoses are not yet a standard or proven therapy in this clinical group and because these could be offered at a later date, no ethical concerns regarding standard of care were raised. When reassessed by the author at 8 weeks, the subjects who reported ongoing or recurrent symptoms were referred for repeat corticosteroid injection.
At baseline, the subjects completed the Oswestry Disability Index, a 10-item questionnaire that indicates the extent to which a person’s functional level is restricted by pain and that is responsive to improvements in pain with orthosis use.5 Scores range from 0 to 5 for each item, with higher scores indicating a greater degree of restriction by pain. The scores for each item are added and divided by the denominator of a possible total score of 50, then converted into a percentage. Thus, the scores range from 0% (no restriction due to pain) to 100% (severe restriction due to pain). In addition, at baseline, the medications the claimant was using, including the type and the number of different classes of medications, were recorded to determine whether the subject was using some form of prescribed analgesic and whether this was specifically for the knee pain. No data were gathered on compliance with orthosis use. No data were gathered on the dosage of medications or use of over-the-counter medications. No data were gathered regarding other treatment modalities sought by the subject. Only analgesic use was ultimately recorded, without specifying the exact classes of drugs or the number of drugs used (i.e., use of one analgesic or more constituted “analgesic use”).
At follow-up at 8 weeks, the subjects were asked whether they felt that they had recovered from their “knee region pain,” with recovery arbitrarily being defined as having pain or symptoms in this region for 1 day per week or less. All subjects who failed to report recovery at 8 weeks underwent a repeat corticosteroid injection. At this 8-week follow-up, the medications the claimant was using, including the type and the number of different classes of medications, were recorded to determine whether the subject was using some form of prescribed analgesic and whether this was specifically for the knee pain. The outcome of recovery and medication use were measured again at 4 months.
SAMPLE SIZE AND STATISTICAL METHODS
Sample size was based on the aforementioned study,5 wherein a significant change (p < 0.05) in Oswestry Disability Index score was noted in a cohort of 30 subjects, suggesting effectiveness of the orthoses. Thus, at least 30 in each group were sought. There are no other studies in anserine bursitis from which to consider sample size. Data were analyzed using the Statistical Package for the Social Sciences (SPSS) 11.0 (MacIntosh version, FootMaxx, Toronto, Canada). Descriptive statistics were calculated for the cohorts, as were baseline scores for the Oswestry Disability Index. Differences between the groups at baseline were analyzed by the Student t-test. Proportions of the subjects using prescription analgesics at baseline and the follow-up points, as well as proportions of the subjects reporting recovery at the follow-up points, were compared using the χ2 test with Yates correction.
In forming the orthoses group, 42 patients had been referred and 8 were excluded (2 had frequent—more than 1 day per week—pain above the midthigh, 1 did not read or write in English at a grade 8 level, 2 had severe knee osteoarthritis, 1 had known medial meniscal tear, 1 had a clinically detectable knee effusion, and 1 had already obtained orthoses). In forming the control group, 47 subjects were referred. Of these, 12 were excluded (5 had frequent—more than 1 day per week—pain above the midthigh, 1 had incomplete data, 1 did not read or write in English at a grade 8 level, 1 had recent back surgery, 2 had severe hip or knee osteoarthritis, and 2 had obtained orthoses elsewhere). Thus, there were 34 subjects in the orthoses group and 35 subjects in the control group at baseline. Of these, one subject was lost to follow-up in the orthoses group and two subjects were lost to follow-up in the control group at 8 weeks. At follow-up at 4 months, the final group numbers were 32 in each group because of additional losses.
The baseline characteristics of the subjects in the orthoses group and the control group are shown in Table 1. There were no significant differences between the groups. Compliance with referrals for injections was equal between the groups at baseline and 8 weeks. Considering all injection referrals (i.e., both the initial and subsequent referrals at 8 weeks), the control group was 84% compliant with injection referrals, and the orthoses group was 88% compliant with injection referrals. There were six persons with diabetes in the control group and five in the orthoses group. The mean ± SD body mass index (BMI) in the control group was 27.1 ± 2.1 and that in the orthoses group was 27.3 ± 2.8. None of the subjects required a lift in his/her orthosis.
At 8 weeks, the proportion of the subjects reporting recovery (pain in the medial knee region ≤1 day per week) was 55% in the control group and 76% in the orthoses group. The number of subjects who reported recovery at 4 months was markedly different between the groups, with only 41% reporting recovery in the control group and 88% reporting recovery in the orthoses group. The control group thus reported a high rate of recurrence, despite referral for repeat corticosteroid injections at the 8-week follow-up. At the 4-month follow-up, a smaller proportion of the orthoses group was using prescribed analgesics for pain (p < 0.05) (Table 2.) There was no statistically significant correlation between age, sex, BMI, or duration of pain and recovery.
This study shows that in a primary care setting, the addition of custom-made foot orthoses to the treatment of anserine bursitis seems to improve outcomes and reduces prescription analgesic use. It should be noted that there were no criteria set for determining whether a subject required custom-made foot orthoses. The orthoses were routinely offered and were all essentially of the same design. This raises the possibility either that arch disorder (collapse or pes cavus) is etiologic in anserine bursitis, and thus this is a group of individuals who have a high prevalence of a clinically relevant disorder, or simply that although orthoses have their clinical effect, this effect can be achieved in clinical practice without knowledge of the arch and foot biomechanics. The key to this study is that simply adding custom-made foot orthoses as a routine measure generates a much higher proportion of subjects who recover from anserine bursitis.
There are a number of limitations to this study. First, it was not a randomized, controlled study. There could be a number of factors that affected the observed outcomes, including subject characteristics not measured, producing a selection bias. The author, although involved in the care of the subjects, did not have contact with any of them in the interval between baseline and follow-up, and the measures are unlikely to have been influenced by the author. It is possible that the primary care physicians learned of the subjects’ use of orthoses, and this may have influenced how they treated the subjects, thus affecting outcomes. It is also possible that there was a selection bias due to practitioner style and treatments between the two groups because the subjects were from two clinics. At the same time, the author was involved in the clinical care in all subjects. Thus, treatment approaches were highly standardized and very similar in both clinics. Given the innocuous nature of orthoses, however, compared with medications, and the ease with which these may be provided, their use in clinical practice should be further considered both for patient convenience and possibly for cost-effectiveness, if these reduce the need for other therapies. Further studies will be required with larger subject numbers in a variety of clinical populations. In addition, economic data should be gathered in future studies to quantify cost-effectiveness.