Introduction

Obesity continues to be a leading healthcare problem in the United States [17, 34]. The prevalence of adult obesity continues to be high, exceeding 30%, with 32% of adult men and 35% of adult women considered obese [17]. Obesity is a known risk factor for the development and progression of knee osteoarthritis [6, 8–10, 29, 32, 42, 45]. The incidence of knee osteoarthritis continues to rise, perhaps because of the greater proportion of the aging US population combined with the high obesity rates [42].

TKA improves a patient’s health-related quality of life [13]. Although a recent study using contemporary TKA implants found overall patient satisfaction was lower than that in earlier reports, overall patient satisfaction was still 81% [7]. The predictors of patient dissatisfaction were expectations not being met, a low 1-year WOMAC, preoperative pain at rest, and postoperative complications. There was no difference between satisfied patients and dissatisfied/neutral patients in terms of body mass index (BMI) [7].

Studies focusing on the obese patient population show most patients show improvements in outcome scores and are satisfied with TKA [1, 2, 12, 18, 19, 24, 26, 30, 33, 37, 43, 44, 48, 49]. However, high body weight amplifies the magnitude of joint load per step and may adversely affect polyethylene wear performance and the rate of aseptic loosening. Increased stress at the bone-prosthesis interface may decrease survivorship of the TKA. On the other hand, these loading force effects may be countered by an obese patient’s sedentary lifestyle that decreases the loading cycles per year [46].

Although metal-backed tibial (MBT) components and modern all-polyethylene tibial (APT) components have comparable success [5], little research is available regarding use of the latter in obese patients. Several studies of cemented TKA in obese patients report increased radiolucent lines around the components [2, 12, 24] or increased osteolysis in obese patients [43], whereas others show no radiographic differences between obese and nonobese patients [18, 30, 48]. However, those studies [12, 24] used MBT, not APT, components. Previous reports of APT components have focused mainly on clinical and radiographic evaluation in elderly [11, 22, 27, 35, 36] or sedentary patients [11]. Whether APT components are associated with similar osteolysis and loosening in obese patients is unclear.

We therefore determined the (1) function, (2) occurrence of osteolysis, and (3) complications in a cohort of obese patients receiving APT components.

Patients and Methods

Between September 17, 1996, and December 19, 2002, we implanted a cemented APT component (PFC/Sigma® TKA; DePuy Orthopaedics Inc, Warsaw, IN) in 273 patients (378 knees). We made the decision to use an APT component intraoperatively based on the patient’s age and activity demands. We considered any patient older than 70 years or any patient who had a low functional demand regardless of age for an APT component; patient weight or BMI did not affect this decision. During that same time, we treated a total of 1435 knees with TKA using any type of component. Of the 273 patients who had an APT component, 90 (125 knees) were characterized as obese (BMI of more than 30 kg/m2 as calculated by the standard formula: patient’s weight in kilograms divided by the square of his or her height in meters [15]). Of those 90 patients (125 knees), 24 (33 knees) died and 13 (17 knees) were lost to followup before the minimum 7-year followup period. Therefore, the final study group consisted of 53 patients (75 knees). Forty were women (59 knees) and 13 were men (16 knees) with an average age of 72 years (range, 61–83 years) and an average BMI of 34 kg/m2 (range, 30–48 kg/m2). The underlying diagnoses were osteoarthritis (52 patients, 74 knees) and avascular necrosis (one patient, one knee). The minimum followup was 7 years (mean, 10.4 years; range, 7.8–14 years). No patients were recalled specifically for this study; all data were obtained from medical records and radiographs.

All surgeries were performed by the senior author (DFD). All patients received prophylactic antibiotics administered within 1 hour of surgical start time. The senior author used a midline incision along with a medial trivector approach [16], made bone cuts using the appropriate cutting jigs, and performed soft tissue balancing in a standard fashion. All knees were cemented tricompartmental knees. The APT design had a cruciform stem, and all procedures were cruciate retaining. The polyethylene thickness varied: 8 mm (37 knees), 10 mm (34 knees), and 12.5 mm (four knees). The tibial component was implanted, and extruded cement was removed. The femoral component was then cemented, and the patella was resurfaced with an all-polyethylene component (Fig. 1).

Fig. 1A–B
figure 1

(A) AP and (B) lateral radiographs show the APT component for TKA used in our obese patients.

Full size image

The postoperative rehabilitation protocol was identical for all patients. Patients were encouraged to be out of bed and to ambulate with assistance with weightbearing as tolerated starting the day of surgery. Postoperative physical therapy and occupational therapy included five 30-minute sessions beginning on Postoperative Day 1. Physical and occupational therapy in the hospital was supervised. All patients started with a walker and progressed to a cane as tolerated. ROM was active and active assisted. Continuous passive motion machines (optional) were available for all patients. No knee immobilizers or braces were used. Patients were discharged to home or an inpatient rehabilitation facility on Postoperative Day 3 and continued to have in-home, inpatient, or outpatient physical therapy.

We evaluated patients clinically and radiographically preoperatively and postoperatively at 6 weeks, 12 weeks, 1 year, and every other year thereafter. In the clinical evaluation, we used the Knee Society Score (KSS) [25] to evaluate pain and function and to generate an overall score. We obtained weightbearing AP, lateral, and Merchant views in all patients.

While the Knee Society Total Knee Arthroplasty Radiographic Evaluation and Scoring System is reportedly associated with low interobserver variability [4], two of us (DFD, TCK) independently evaluated all radiographs for radiolucent lines and osteolysis. We did not measure interobserver variability of the measurements. We recorded any progressive radiolucent line defined as expanding to neighboring zones or increasing in width from previous radiographs or both. Radiolucencies, if identified, were recorded according to the specific component sites. We used criteria of the Knee Society knee radiographic score [14] to assess lucencies and osteolysis.

Of 53 patients (75 knees), complete postoperative clinical KSSs were available for 46 patients (63 knees) and complete postoperative radiographic studies were available for 38 patients (51 knees) with a minimum 7-year followup. The remaining patients who were not considered lost to followup after 7 years but who were missing clinical or radiographic evaluations were still being followed in the same orthopaedic practice and were noted by others in the practice to have well-functioning TKAs without any problems. For the data analysis, we included patients who had complete clinical KSS and complete radiographic evaluation with a minimum 7-year followup.

Results

The mean KSS was 51.6 points (range, 8–99 points) preoperatively and 92.4 points (range, 45–100 points) at the latest followup (Table 1). Preoperative ROM was similar to postoperative ROM. The mean knee extension was 3.3° (range, 0°–19°) preoperatively and 0.7° (range, 0°–10°) at final followup. The mean knee flexion was similar preoperatively and at last followup: 119.4° (range, 82°–140°) versus 118.2° (range, 75°–135°), respectively.

Table 1 Mean preoperative and postoperative KSS and ROM
Full size table

We identified five tibial radiolucencies in four patients, all of which were less than 1 mm and nonprogressive; one femoral radiolucency in each of two patients; and one case of minimal, nonprogressive tibial osteolysis. There were no femoral or patellar zones of osteolysis. No knee was radiographically positioned more than ± 3° relative to the mechanical axis, no femoral component was placed in more than 2° of femoral flexion, and no tibial component was placed in more than 6° of posterior slope. None of our patients had patellar subluxation or dislocation.

One patient required a reoperation after a traumatic event. This patient complained of instability after a fall that resulted in injury to the medial collateral ligament 21 months after the index surgery. The knee was converted to a stabilized implant. There were no implant-related failures or any implants at risk of failure at latest followup.

Discussion

In the practice of any orthopaedic surgeon, obese individuals comprise a substantial proportion of patients with osteoarthritis. For most obese patients, as for any patient, TKA results in improved pain and function scores with high satisfaction rates [2, 12, 26, 30, 33, 37, 49]. MBT components and modern APT components both have been shown to be successful in the general population; however, the use of APT components specifically in obese patients has not been well studied. We therefore determined the (1) function, (2) occurrence of osteolysis, and (3) complications in a cohort of obese patients receiving APT components.

There are several limitations to our study. First, we had no control group: we did not compare the use of APT and MBT components in the same obese population. A randomized controlled trial would better study this concept. Second, 41% of the patients died or were lost to followup before the 7-year minimum followup. That large a number lost may have affected the findings. At last followup between 6 months and 9 years, none of those patients had any complications or required revision. Third, our indications to use an APT component based on patient age and activity introduce bias to the study: our older patients (average age, 72 years) may have been inherently less active, and the correspondingly fewer cycles on the polyethylene may have produced less wear than might have occurred in a younger, more active obese population. Fourth, our study represents a midterm, not long-term, analysis, and additional followup is necessary. However, despite these limitations, our data may serve as a basis for consideration of choosing an APT component for TKAs in obese patients.

Multiple publications [3, 11, 20–23, 27, 28, 31, 36, 39–41, 47] also document clinical and radiographic success with the use of an APT component (Table 2). Three similar retrospective studies report clinical success with the use of an APT component. Ranawat et al. [38] reported on 54 TKAs using APT components. All patients were 60 years old or younger with an average BMI of 29.2 kg/m2 (range, 20.2–48.7 kg/m2). All knees received posterior-stabilized implants. At an average followup of 5 years (range, 2–11 years), their KSSs improved to 95 points (range, 72–100 points) with only one tibial component loosening secondary to trauma. We found similar KSSs (mean, 92 points), but our population differs from that of Ranawat et al. [38] because (1) our patients were older (mean age, 72 years; range, 61–83 years), (2) all of our patients were considered obese (only some in the other study were obese), and (3) we used a cruciate-retaining prosthesis. Dalury et al. [11] also previously reported midterm clinical findings of APT components in 88 patients (120 TKAs) older than 70 years or who had comorbidities suggestive of a low activity level. Unfortunately, patient BMI was not captured in that study. The average postoperative KSS in that study was 93.7 points. In 2007, Gioe et al. [22] reported on 443 APT components (378 patients) implanted by 12 surgeons in a community registry. The mean age was 77 years (range, 58–94 years), but BMI was also unfortunately not captured in this study for comparison. Although the authors noted APT components performed extremely well, they did not report KSS.

Table 2 Summary of published data regarding APT components
Full size table

Ranawat et al. [38] reported no radiographic evidence of component loosening, progressive radiolucent lines, osteolysis, or malalignment. Dalury et al. [11] noted 32 of 120 knees with nonprogressive radiolucencies, four knees with progressive radiolucent lines requiring observation, and no knee with impending failure. That study did not reveal osteolysis in any knee. The study by Gioe et al. [22] did not specifically address radiolucencies or osteolysis but noted one knee required revision for aseptic loosening. We found no progressive radiolucent lines, one implant with minimal osteolysis, and no implant-related failures.

The obese population in the United States has increased markedly over the past several decades. At an average 10-year followup, obese patients had satisfactory clinical and radiographic findings with the use of an APT component. These findings support the continued use of an APT component even in the obese population.