To evaluate the effect of body mass index (BMI) on short-term functional outcome and complications in primary total knee arthroplasty.

All patients undergoing primary total knee arthroplasty at a single institution between 2007 and 2013 were identified from a prospective arthroplasty database. 2180 patients were included in the study. Age, gender, BMI, pre- and post-operative functional scores [Western Ontario and McMaster University Arthritis Index (WOMAC) and SF-36], complications and revision rate were recorded. Patients were grouped according to the WHO BMI classification. The functional outcome of the normal weight cohort (BMI < 25) was compared to the overweight and obese (BMI ≥ 25) cohort. A separate sub-group analysis was performed comparing all five WHO BMI groups; Normal weight, overweight, class 1 obese, class 2 obese and class 3 obese.

With a mean age of 67.89 (28-92), 2180 primary total knee replacements were included. 64.36% (1403) were female. The mean BMI was 31.86 (18-52). Ninty-three percent of patients were either overweight or obese. Mean follow-up 19.33 mo (6-60 mo). There was no significant difference in pre or post-operative WOMAC score in the normal weight (BMI < 25) cohort compared to patients with a BMI ≥ 25 (P > 0.05). Sub-group analysis revealed significantly worse WOMAC scores in class 2 obese 30.80 compared to overweight 25.80 (P < 0.01) and class 1 obese 25.50 (P < 0.01). Similarly, there were significantly worse SF-36 scores in class 2 obese 58.16 compared to overweight 63.93 (P < 0.01) and class 1 obese 63.65 (P < 0.01) There were 32 (1.47%) superficial infections, 9 (0.41%) deep infections and 19 (0.87%) revisions overall with no complications or revisions in the normal weight cohort (BMI < 25).

Post-operative functional outcome was not influenced by BMI comparing normal weight individuals with BMI > 25. Patients should not be denied total knee arthroplasty based solely on weight alone.

Total knee arthroplasty (TKA) is an effective surgical treatment of osteoarthritis of the knee, with 700000 procedures performed in the United States annually with the demand for TKA projected to increase 673% by 2030[1]. Occurring in tandem with this increase in demand is the exponential increase in obesity in society. Currently in Ireland, 36% of the population are overweight and 14% obese and this is estimated to increase further in the future[2]. It is well established that obesity confers an increased risk for a number of medical conditions including ischaemic heart disease, diabetes and stroke[3]. It has also been shown that obesity increases the risk of development of osteoarthritis, particularly in the knee, which has potential implications for the demand for TKA in the future[4].

Obesity has a number of implications for surgery in general, but in particular for elective surgery such as TKA. Obesity is an independent risk factor for a number of perioperative complications including acute coronary syndrome, wound infection and urinary tract infection[5]. The outcome of obese patients that undergo TKA as compared to non-obese patients is of particular interest. Currently the evidence is unclear with some studies indicating that obese patients achieve inferior outcomes[6] with others showing equivalent functional outcome[7,8].

The aim of this study was to assess the effect of body mass index (BMI) on functional outcome in primary total knee arthroplasty.

Patients that underwent primary total replacement were identified from a prospectively collected joint registry at a single institution. Ethical approval was obtained for the establishment of the joint registry and for on going research. The joint registry is maintained by a full time clinical nurse specialist and all demographic and clinical information for each arthroplasty procedure performed at the institution is prospectively anonymously recorded. Two thousand one hundred and eighty patients were identified during the period 2007-2013. Demographic data including age and gender were collated for each patient. Body mass index (BMI) was calculated for each patient at pre-operative assessment using the standardised formula; weight in kilograms squared, divided by height in metres squared. Functional outcome scores, the Western Ontario and McMaster University Arthritis Index (WOMAC) and the Short Form 36 (SF-36) were collected pre-operatively and 6 mo post operatively. Complications including revision, superficial and deep infection, deep venous thrombosis (DVT) and pulmonary embolism (PE) were recorded prospectively in the postoperative period.

Patients were divided into two comparative groups for the purpose of the study, those with a normal BMI (less than or equal to 25) Group 1 and those who were overweight or obese (greater than 25) Group 2 according to the WHO BMI classification[9]. A separate sub-group analysis was performed comparing all five WHO BMI groups; Normal weight, overweight, class 1 obese, class 2 obese and class 3 obese.

The Primary outcomes assessed were pre-operative and six-month post-operative WOMAC and the SF-36 scores. The WOMAC score is a validated self-administered questionnaire that assesses the condition of patients with hip and knee arthritis. It has a scale of 0 to 100, with a higher score equalling more pain, stiffness and functional limitation[10]. The SF-36 is also a self-administered questionnaire that assesses quality of life. It has a scale of 0 to 100, with a higher score equating to a greater quality of life[11]. Secondary outcomes assessed included complications revision, superficial and deep infection, DVT and PE.

A total of 2180 primary total knee replacements were performed at the institution between 2007 and 2013. The mean age was 67.9 years (28-92), with 36% Male and 64% Female. The mean follow up was 19.3 mo, with a range of 6 mo to 5 years. The distribution according to BMI group is shown in Figure 1. The mean BMI was 31.9 (18-52) with 63% obese, 30% overweight and 7% normal weight. The two comparative study groups consisted of Group 1 (n = 162 patients) and Group 2 (n = 2018 patients).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Distribution of patients according to World Health Organization body mass index classification. BMI: Body mass index.

Overall the study revealed no significant difference in short-term post-operative functional outcome in patients with a normal BMI as compared to overweight or obese patients. Sub-group analysis found significantly lower functional outcome scores in class 2 obese patients (BMI 35-39.9) compared to both overweight and class 1 obese patient.

The study highlights that, the vast majority of patients now presenting to our institution for total knee replacement, 93% are either overweight or obese. This is significantly higher than the baseline levels in the general population, where 50% are either overweight or obese[2]. This finding is mirrored in previous studies, which also revealed a considerable proportion of patients undergoing total knee arthroplasty are now obese[12]. According to the latest National joint registry in the United Kingdom (NJR) figures (2013), the mean BMI of patients undergoing TKA in the United Kingdom is now 30.8 (Class 1 obese)[13]. This underlines the significant burden that this increase in BMI will place on orthopaedic services now and in the future.

The principle finding in this study of equivalent functional outcome comparing normal weight BMI < 25 individuals with BMI > 25 is in keeping with a recent study of 13673 primary total knee replacements by Baker et al[14] using NJR data. They found that the improvement of patient reported outcomes (PROMs) were similar irrespective of BMI. Similarly, Desmukh et al[15] revealed no correlation with BMI and functional outcome at 1 year. However, a consensus has yet to be reached in the literature, as there is also evidence that increasing BMI, particularly greater than 40 results in inferior clinical outcomes. Collins et al[16] reviewed 445 total knee replacements and found inferior clinical outcome scores in individuals with a BMI greater than 30 at 9 years follow up. Interestingly, although obese patients achieved lower outcome scores as compared to non-obese patients, they achieved significant absolute functional improvement and the authors concluded that they “found no reason to limit access to total knee replacement in obese patients”. While there was no difference in our main outcome measure, the sub-group analysis revealed significantly worse functional outcomes in the class 2 obese cohort compared to both the overweight and class 1 obese cohort. Interestingly the class 3 obese cohort did not demonstrate any significant difference in functional outcome scores. All cohorts achieved significant absolute improvements in functional outcome measures compared to preoperative values. The significance of our finding of inferior outcomes in the class 2 cohort is unclear. While we have included a relatively large cohort in this study (2180), perhaps larger numbers found in registry studies are necessary to define clear sub-group differences.

Despite no difference in functional outcome, the incidence of all complications was higher in the overweight and obese cohort as compared to the normal weight cohort. The evidence in the literature is clear in relation to the increased risk of perioperative complications with increasing BMI in TKA. The aetiology behind this is multifactorial. Wound healing and the development of both superficial and deep peri-prosthetic joint infections are significantly more common with increasing BMI. A recent meta-analysis by Kerkhoffs et al[17] revealed an odds ratio of 1.9 for all infection and 2.38 for deep infection in obese patients as compared to non-obese patients in an analysis of 15276 and 5061 patients respectively. Obese patients are also at a higher risk of thromboembolic disease post operatively[18]. It is imperative that patients are counselled in detail regarding the increased risk of perioperative complications with increasing BMI. While it would seem intuitive that patients should attempt to lose weight prior to surgery, some recent evidence suggests that obese patients that lose a significant proportion of bodyweight preoperatively, actually have a higher rate of surgical site infection compared to control[19]. Further research is needed in relation to perioperative weight management, however it raises interesting questions about the best way to manage this ever-expanding cohort of overweight and obese patients.

Due to the current demographics of our patient cohort, there were relatively few normal weight individuals presenting for surgery and therefore available for inclusion in the study. Similarly, larger studies using registry data may be necessary to elucidate clear sub-group differences. While the patient numbers were sufficient to statistically compare the functional outcome scores, there was an insufficient incidence of complications to draw any statistical conclusions in relation to complication differences. We acknowledge that early functional outcome may not correspond to long-term functional outcome and further research in this area is required.

In conclusion, overall there was no difference in early post-operative functional outcome comparing normal weight individuals with those of a BMI > 25 in a cohort of 2080 primary total knee replacements. Patients should be counselled regarding the potential increased risk of complications with increasing BMI, however they should not be denied TKA based solely on weight if medically fit to undergo the procedure.