Two hundred and eighty-five women who were scheduled for bunion surgery
were enrolled in the study. They had a mean age of forty-nine years (range,
eighteen to eighty-one years). Sixty-seven orthopaedic surgeons, twelve in
academic practice and fifty-five in private practice, who were members of the
American Orthopaedic Foot and Ankle Society participated in the present study.
The inclusion criteria for the study were a hallux valgus angle of
>20°, an age of eighteen years or greater, no inflammatory arthritis,
and pain that was unresponsive to conservative treatment with shoe-wear
modification. The study was approved by the institutional review board at the
principal investigator's institution.
Prior to surgery, the patients were asked by their physician to complete a
baseline AAOS Foot and Ankle Outcomes Data Collection Questionnaire. This
validated questionnaire includes the Short Form-36 (SF-36), which, when
scored, provides eight separate domain scores and two composite scores from 0
to 100, with 100 being a perfect score for the ten following items: physical
function, role-physical (the patient's perception of his or her physical
abilities), bodily pain, general health, vitality, social function,
role-emotional (the patient's perception of his or her emotional function),
mental health, physical composite score, and mental composite score.
Thirty-five additional questions with regard to the foot and ankle were
included, and five additional scales were
calculated5. The
physical health and pain scale is actually the mean of the three physical
health scales (physical function, role-physical, and bodily pain) from the
SF-36 questionnaire and is not derived from the additional questions. Five
questions yield the score for treatment expectations, one question concerns
satisfaction with symptoms, twenty questions relate to the global foot and
ankle scale, and nine questions concern shoe comfort. Satisfaction with
symptoms is rated on a scale from 1 to 5, while the other four scales are
scored from 0 to 100, with 100 indicating perfect function.
The responses to the questionnaire were analyzed with use of Passport
software, developed by the American Academy of Orthopaedic Surgeons, after
answers were manually entered into the database. As previous normative
databases for the general population have demonstrated significant differences
among age-groups, the data were stratified into age-groups consistent with
those reported by Ware et al. in the SF-36 normative
database3. The data
were imported into Microsoft Excel (Redmond, Washington), where the means and
standard deviations were calculated for each age-group under every category.
The p values correlating the clinical rating score of the general population
and that of the patients with a bunion were calculated, with use of the
unpaired t test function (SigmaStat 2.0; Jandel Scientific, San Rafael,
California). The mean scores for the general population on the global foot and
ankle scale and the shoe comfort
scale4 were likewise
correlated, with use of the unpaired t test, with the mean scores for the
patients with a bunion. No age-specific comparisons were made for the global
foot and ankle or shoe comfort scale as normative data were not available by
age-group.
Physicians who enrolled their patients in the study were given a data sheet
requesting preoperative radiographic angular data for the hallux valgus and
the intermetatarsal angle between the first and second metatarsals. The data
on the hallux valgus angle were then broken down into groups that had mild
deformity (hallux valgus of <30°), moderate deformity (30° to
40°), and severe deformity (>40°). Similarly, the data on the
preoperative intermetatarsal angle between the first and second metatarsals
were stratified into mild (<10°), moderate (10° to 15°), and
severe (>15°) groups. The mean (and standard deviation) for each of the
fifteen scores for each of these groups (hallux valgus and intermetatarsal
angles) was calculated and correlated with a one-way analysis of variance
test.
In order to evaluate the effect of age on the SF-36 scores for the general
population and for the patients with a bunion, a linear regression was
performed with use of Microsoft Excel to determine whether functional status
scores are influenced by increasing age. A trend line was added to the data,
and the slope and the correlation coefficient (R2) of the trend
line were calculated. A positive slope indicates that the score increased with
increasing age, whereas a negative slope indicates a decrease in score. The
correlation coefficient indicates how well the trend line fits the data
points, with an exact fit having a R2 value of 1.
The results for eight of the SF-36 scales (excluding the composite scores)
are presented in Figures 1,
2,
3,
4,
5,
6,
7,
8, with all p values presented
in Table I. The mean value for
the general female population for these SF-36 scores is presented in the same
graphs to allow comparison with the patients undergoing bunion surgery. Across
all age-groups, the scores for bodily pain were found to be significantly
worse for patients undergoing bunion surgery, with the scores ranging from 10
to 23 points lower than those for the respective general population cohort
(Fig. 6). The scores for
physical function and role-physical for the patients undergoing bunion surgery
were significantly lower for two age-groups in each scale (Figs.
7 and
8). Conversely, the general
health of patients undergoing bunion surgery was significantly better in those
over the age of thirty-five years compared with that of the general population
(Fig. 5).
Evaluation of the five scales that are specific to the lower extremity
revealed that the physical health and pain score decreased with age for
patients with a bunion (Fig.
9). Treatment expectations were somewhat variable and were noted
to be somewhat lower for patients in the twenty-five to thirty-four-year
age-group (Fig. 10). The
scores for satisfaction with the symptoms were consistently found to be low
(=2 of 5), which is not surprising as the patients were about to undergo
bunion surgery and had pain according to the inclusion criteria
(Fig. 11). The global foot and
ankle scores for the patients having bunion surgery were found to trend lower
for each advancing decade of age and ranged from 72 to 80, with a mean of 75.9
(Fig. 12), which is
significantly lower (p < 0.001) than the mean foot and ankle scale score of
93.2 in recently published normative data for the general
population4. This
difference would be expected as the global foot and ankle scale is designed
specifically for foot and ankle function. Similarly, the shoe comfort scale
was low, ranging from 21.0 in the eighteen to twenty-four-year age-group to
37.4 in the sixty-five-year and older age-group
(Fig. 13). The average shoe
comfort score was 34.3, which differed significantly from the average score of
73.9 for the general population (p < 0.001).
Stratification of the baseline outcome scores according to the severity of
the deformity, whether with use of the hallux valgus angle or the
intermetatarsal angle between the first and second metatarsals, revealed no
significant differences for any of the fifteen outcome scores (see
Appendix).
Slopes and correlation coefficients for the general population and for
patients undergoing bunion surgery demonstrated that, except for mental
health, the other seven scale scores for the SF-36 decrease with age, with the
largest decreases in physical function, role-physical, bodily pain, and
general health. Similar decreases in function were noted for patients
undergoing bunion surgery, except the slope of the decrease was not as
pronounced for bodily pain and was actually positive for general health
(Table II).
The present study describes the preoperative general health status (SF-36
scores) and lower extremity functional scores (AAOS lower extremity
questionnaire) for patients undergoing bunion
surgery5. Normative
data for the SF-36 from the general population of patients, which provided
baseline values, have previously been
published3. In the
present study, the baseline data for women undergoing bunion surgery were
reported and were compared with the data for sex and age-matched groups from
the SF-36 normative data. The SF-36 normative data were obtained from a
population in which 78% of the individuals were white, 16% were black, and 6%
were other minorities, and 42% of the individuals had greater than a
high-school
education3. In our
study, 88% of the patients were white, 6% were black, and 6% were other
minorities, and 74% had greater than a high-school education. It is unclear
whether these differences in ethnicity and education level had any impact on
the results.
Evaluation of the SF-36 scores for the general population and the patients
undergoing bunion surgery revealed many interesting findings. The slopes for
the data for the eight subscales of the SF-36 score (excluding the composite
scores) revealed a substantially negative trend with each increasing decade of
age, especially for general health, bodily pain, physical function, and
role-physical for the general population. A less consistent decrease in social
function, role-emotional, and vitality was noted with a mild improvement in
mental health. These slopes for age-related changes emphasize the importance
of matching patient populations for age when designing a study comparing
patients with population-based normative data.
The bodily pain score was noted to be significantly less (indicating
greater bodily pain) for the patients having bunion surgery in each of the
seven age-intervals than that for the general population. The magnitude of
difference in bodily pain scores from the normative populations ranged from 10
to 25 points, with the smaller differences noted in the oldest age-groups. A
trend toward decreased physical function and role-physical scores, with the
scores for two age-groups in each scale significantly lower than those for the
general population, was also noted. In general, the mental health scores also
increased slightly with age for the patients undergoing bunion surgery, which
was similar to the scores for the general population. For one age-group, the
patients having bunion surgery had a significantly better mental health score
than the general population. This finding is in contrast to some clinicians'
perception that many patients undergoing bunion surgery have underlying
psychological problems, which would be expected to decrease the mental health
score. In general, the role-emotional, social function, and vitality scores
tend to be comparable for the general population and the patients with
bunions. Conversely, the general health of patients undergoing bunion surgery
was better than that of the general population in the four oldest age-groups.
This finding could be a reflection of the better health, in general, among
patients with bunions, suggesting that they have higher functional
demands.
The AAOS lower extremity questionnaire has thirty-five questions focusing
on the lower
extremity5. Five
additional scores specific to the lower extremity are also generated. The
physical health and pain score is actually the mathematical average of the
role-physical, physical function, and bodily pain scores of the
SF-365. The
treatment expectation score is calculated from five questions regarding what
benefit the patients expect from the impending intervention. The questions
address whether the patient believes that he or she will have relief from the
symptoms, be able to perform everyday activities, sleep better, return to
work, and do recreational activities. In this study, a consistently high
treatment expectation score of approximately 80 or above suggests that the
patients had high expectations for a successful outcome after the surgery. The
score for satisfaction with the symptoms is generated from the response to a
single question regarding the level of dissatisfaction that the patient would
have if he or she had to spend the rest of his or her life with the current
symptoms. It was consistently =2 on a 5-point scale, which is logical as
patients about to undergo bunion surgery would be expected to be dissatisfied
with the current symptoms. The global foot and ankle scale is derived from
twenty questions regarding such things as walking ability, giving-way, and
pain. This score was found to be consistently between 70 and 80 for all
age-groups, indicating relatively poor function, compared with an average of
93 for the general patient population. Although age-specific normative data
for the global foot and ankle scale are not available, the mean global foot
and ankle score for the patients with bunions was significantly decreased. The
shoe comfort score is based upon nine questions, but four are not scored with
the algorithm for this
instrument5. The
algorithm was not available at the start of the study, and therefore these
four questions were included but should be deleted if the instrument is to be
used in the future. The five questions that are scored for the shoe comfort
scale include a yes or no answer to whether the patient can wear any dress
shoe, most dress shoes, walking or casual shoes, prescription shoes, or all
shoes. The markedly decreased scores for shoe comfort suggest that this is the
most sensitive scale for assessing disability or discomfort in patients with
bunions.
When evaluating all fifteen scales that are calculated from the AAOS lower
extremity questionnaire, the scores that revealed the greatest impact for the
patients with bunions compared with the general population were the bodily
pain, global foot and ankle, and shoe comfort scores. This finding suggests
that the questions used to derive these scales are the most sensitive for
evaluating bunion surgery. One weakness of the current questionnaire is that
it has thirty-six questions to generate the SF-36 scores and an additional
thirty-five questions for the AAOS lower extremity questionnaire. These
questionnaires require approximately ten to fifteen minutes to complete, on
the basis of the education level of the respondents. Shorter, more focused
questionnaires would be easier to administer and score.
The patients' flawed memories of the preoperative state have recently been
demonstrated by readministering the outcome questionnaires six months or more
following surgery2.
The present study describes a prospective database of data collected from
patients about to have bunion surgery at multiple centers across the United
States. The age-stratified results reveal different values for patients of
different ages, which may allow for a more accurate comparison in
retrospective studies on patients with bunions.
The baseline scores of the patients for each of the fifteen scales were
correlated with the severity of the bunion deformity with use of both the
hallux valgus angle and the intermetatarsal angle between the first and second
metatarsals. We hypothesized that, with increasing deformity, the patients
would have a decreased score on the bodily pain scale (i.e., more pain), shoe
comfort scale, and global foot and ankle scale. When the deformity was
stratified as mild, moderate, or severe, no significant difference was noted
among the varying severities of deformity for any of the fifteen scales, which
disproved our hypothesis. The results suggest that once even a mild bunion
deformity has developed, the magnitude of the clinical symptoms and the
disability are similar regardless of the magnitude of the deformity. As the
amount of disability associated with a less severe deformity is similar to
that associated with a more severe deformity, these findings suggest that
surgeons should be more confident in offering surgery to patients with a less
severe deformity after conservative treatment has failed.
In conclusion, the trends found in this study suggest that female patients
with a bunion are as healthy or healthier than the general population, but
they have significantly increased bodily pain. Additionally, they have
decreased scores on the global foot and ankle and shoe comfort scales. The
SF-36 bodily pain score and the AAOS global foot and ankle and shoe comfort
scores demonstrate the largest magnitude of deviation from the scores for the
general population. Regardless of the severity of the deformity as measured
radiographically, all patients had a similar degree of pain and disability as
a result of the bunion. The present study provides age-stratified baseline
data for comparison when evaluating patients undergoing bunion surgery.
Tables presenting the radiographic findings (hallux valgus scores and
intermetatarsal angle between the first and second metatarsals) correlated
with the outcome measures are available with the electronic versions of this
article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM). ?