Defining the Relationship between Obesity and Total Joint Arthroplasty
University of Massachusetts Medical School, Department of Orthopedics, 55 Lake Ave. North, Worcester, MA 01655. E-mail: firstname.lastname@example.org
Objective: The purpose of this study was to examine the relationship between obesity and patient-administered outcome measures after total joint arthroplasty.
Research Methods and Procedures: A voluntary questionnaire-based registry contained 592 primary total hip arthroplasty patients and 1011 primary total knee arthroplasty patients with preoperative and 1-year data. Using logistic regression, the relationships between body mass index and the several outcome measures, including Short Form-36 and Western Ontario and McMaster Universities Osteoarthritis Index, were examined.
Results: There was no difference between obese and non-obese patients regarding satisfaction, decision to repeat surgery, and Δphysical component summary, Δmental component summary, and ΔWestern Ontario and McMaster Universities Osteoarthritis Index scores (p > 0.05 for all). Body mass index was associated with an increased risk of having difficulty descending or ascending stairs at 1 year (odds ratio, 1.2 to 1.3).
Discussion: Obese patients enjoy as much improvement and satisfaction as other patients from total joint arthroplasty.
Most orthopedic surgeons are hesitant to operate on obese patients. Orthopedic surgeons suspect that obese patients are at greater risk of having complications, component failure, and worse outcomes after total joint arthroplasty than other candidates (1) (2). Patient-administered outcome measures have become as important as conventional physician-administered outcomes in the evaluation of total joint arthroplasty. It is unclear how obese patients’ subjective evaluations of their outcomes compare with their non-obese peers. The purpose of this study was to examine the relationship between body mass index (BMI) and patientadministered outcome measures after total hip arthroplasty (THA) and total knee arthroplasty (TKA).
Research Methods and Procedures
The study of human subjects in The Hip and Knee Registry was approved by the Committee on Human Subjects at the University of Massachusetts Medical School. Verbal informed consent was obtained from study participants. This research conforms to the principles of biomedical research with human subjects set forth in the World Medical Association Declaration of Helsinki.
The Hip and Knee Registry is a questionnaire-based registry in which physicians are asked to complete preoperative and follow-up questionnaires regarding physical exam, radiology findings, and complications. Patients completed preoperative and follow-up questionnaires that included the Short Form-36 and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) surveys. Other questions assessed demographic status, including race, income, marital status, and education.
The Short Form-36 is a reliable, valid, and accepted standard measurement for outcomes of THA (3). It is a self-administered questionnaire with eight domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These eight scores are often combined into two meta-scores, a physical component summary (PCS) and a mental component summary (MCS). The scores range from 0 to 100, with 50 representing the mean of the general U.S. population with a SD of 10. A higher score represents better function.
The WOMAC is a valid and reliable questionnaire that was developed to evaluate osteoarthritis of the hip and knee (4) (5). The WOMAC has 24 questions and consists of three parts: a physical function score, stiffness score, and pain score. The results were normalized by multiplying the total raw score of 0 to 96 (poor to excellent) by 100/96. The resulting normalized scores ranged from 0 to 100. A higher score represents better function.
The database was queried for patients with primary THA or primary TKA for osteoarthritis. Only patients with preoperative height and weight and 1-year satisfaction data were selected. BMI was defined as weight in kilograms divided by square height in meters. Divisions of BMI were defined as follows: <25 kg/m2, healthy; 25 to 29.9 kg/m2, overweight; 30 to 34.9 kg/m2, class I obese; 35 to 39.9 kg/m2, class II obese; and >40 kg/m2, class III obese (6).
A satisfaction question was phrased as follows: “How satisfied are you with the results of your joint replacement?” Possible responses were “very satisfied,” “somewhat satisfied,” “neutral,” “somewhat dissatisfied,” and “very dissatisfied.” For statistical analysis, patients were combined into two groups: “very satisfied” and “other.” Patients were also asked: “If you could go back in time and make the same decision again, would you choose to have this surgery?” Possible answers were “yes” and “no.”
Complications were classified as either medical or orthopedic. Medical complications included infection, deep vein thrombosis, nerve palsy, pulmonary embolism, and local or systemic complications. Orthopedic complications included hematoma, hemorrhage, fracture, loosening, infected prosthesis, and component failure. Hip-specific complications were trochanteric problems, osteolysis, and dislocation. The knee-specific orthopedic complication was extensor mechanism problems.
Due to missing data, denominators for different variables varied. All Δ scores were calculated by finding the mean of the differences of preoperative and 1-year postoperative scores. ANOVA and Student's t tests were used to determine significant differences in the means between groups; p < 0.05 was considered significant. Logistic regression was used to determine odds ratios (OR) and confidence intervals (CI). Power calculations were performed. All statistical analyses were performed using SAS statistical software (SAS, Cary, NC).
Demographic information is summarized in Table 1.
Table 1. Demographic information
|THA|| || || || || || || || || |
|p value|| ||p > 0.05|| || ||p > 0.05|| || || || |
|TKA|| || || || || || || || || |
|p value|| ||p > 0.05|| || ||p > 0.05|| || || || |
For unadjusted primary THA, there was no difference between BMI groups regarding satisfaction at 1 year; preoperative MCS, postoperative MCS, or ΔMCS scores; and preoperative PCS and ΔPCS scores. BMI was negatively correlated with unadjusted postoperative PCS scores (p < 0.05) as well as with preoperative and postoperative WOMAC scores (p < 0.05 for both) but not with ΔWOMAC scores (see Tables 2 and 3).
Table 2. Short Form-36 and WOMAC scores for THA patients
| ||p value||0.1352||0.2035||0.2545||0.0734||0.0274||0.4822||0.0286||0.0494||0.5352|
Table 3. THA patients
In an unadjusted analysis of primary hips, age was negatively correlated with BMI (p < 0.001). There was no statistically significant increased risk of a medical complication as BMI increased, but for each increase in BMI group, there was an OR of 1.3 (1.0 to 1.8) for having an orthopedic complication. BMI correlated positively with difficulty ascending stairs, with an OR of 1.3 (CI: 1.1 to 1.5), as well as with difficulty descending stairs, with an OR of 1.3 (CI: 1.2 to 1.5). There was no difference between BMI groups in terms of repeating the same decision at 1 year. Independent of BMI, >96% of primary THA patients would have surgery again.
Power was calculated for 1-year scores of THA patients to determine β error. There was adequate power to detect a 10-point difference in WOMAC scores and a 7-point difference in PCS scores.
In a crude analysis of TKA patients, PCS scores, WOMAC scores, age, difficulty descending stairs, and difficulty ascending stairs were negatively correlated with BMI (see Tables 4 and 5). The OR was 1.2 (CI: 1.1 to 1.3) for difficulty descending stairs and 1.2 (CI: 1.1 to 1.4) for difficulty ascending stairs. No association between BMI and satisfaction with the results of surgery or decision at 1 year was observed. Nonlinear differences between BMI and preoperative and postoperative MCS scores were observed, with the lowest MCS scores associated with the group with a BMI of 35 to 40 kg/m2. No associations were observed between BMI and Δ scores for PCS, MCS, or WOMAC.
Table 4. Short Form-36 and WOMAC scores for TKA patients
| ||p value||0.0283||0.0409||0.9765||0.0001||0.0232||0.4216||0.0001||0.0022||0.0819|
Table 5. TKA patients
Power calculations were performed to determine β error for 1-year scores. For TKA patients, there was sufficient power to detect a four-point difference in mean WOMAC scores and a five-point difference in mean PCS scores.
This study demonstrates that obese total joint arthroplasty patients are just as satisfied with their outcome as their non-obese peers. This finding contradicts a study of Medicare TKA patients that showed that predictors of a lower level of patient satisfaction included a higher BMI (7). This difference in satisfaction rates may be a result of the larger sample size in this study or of the stratified sampling techniques of the Medicare study. Obese patients were also as likely to make the same decision at 1 year as other patients. Furthermore, there is no difference in Δ scores for PCS, MCS, or WOMAC for different BMI groups. Obese patients are happy with their surgery and have just as much improvement as other patients.
This is not to suggest that obesity does not have an effect on the outcome of THA or TKA. As BMI increased, the risk of having difficulty ascending and descending stairs at 1 year increased. In another study with 10-year follow-up, obese TKA patients, as defined by BMI, had worse stair-stepping ability than healthy patients (p < 0.05) (8). Also, as BMI increased, postoperative WOMAC and PCS scores decreased. Despite the increased risk of having difficulty with stairs or worse postoperative scores, obese patients were as satisfied with the outcomes of their surgical procedure as other patients.
In this study, no difference in complication rates between BMI groups for TKA was observed, but obese THA patients had a small increased relative risk of complications. The literature examining this issue falls prey to the problem of multiple definitions of obesity (1) (9) (10) (11). Obesity has been arbitrarily defined as 120% of the ideal body weight for height based on the 1983 Metropolitan Life Insurance Company tables and as a BMI of ≥40 kg/m2 or ≥32 kg/m2 in various studies. It will be impossible to evaluate the impact of obesity on THA and TKA outcomes until the orthopedic community agrees on a standardized definition of obesity. We recommend that the BMI divisions set forth by Willett et al. (6), regarded as the consensus of the medical community and the World Health Organization, be adopted by the orthopedic community as a standardized definition of obesity so that different research can be compared.
There are several potential drawbacks to this study. The patients included in this study represent ∼24% of the pool of preoperative data because patients were excluded if there was no follow-up at 1 year. The demographics of the population used in the study and the database as a whole (see Table 1) were very similar with respect to mean age, mean BMI, and male-to-female ratio; however, the possibility of selection bias is still present. Because of the voluntary nature of The Hip and Knee Registry, the physicians enrolled may not represent North American orthopedic surgeons as a whole. Patients in this study only had 1-year of follow-up.
Although the relationship between obesity and complications after total joint arthroplasty remains unresolved, this study suggests that obese patients have as much improvement and satisfaction after THA or TKA as other patients.
The Hip and Knee Registry is supported by an unrestricted educational grant from Aventis Pharmaceuticals, Inc.