Total Knee Arthroplasty in Patients 40 Years of Age and Younger With Osteoarthritis

Abstract

The results of 32 total knee arthroplasties performed for osteoarthritis in 32 patients who were 40 years of age or younger are reviewed. At a mean followup of 7.9 years (minimum, 5 years), the Knee Society knee scores increased from an average of 47 to 88 points, and the function scores increased from 45 to 70 points. Overall, Knee Society knee scores were considered good or excellent in 82% of patients (26 knees) and fair or poor in 18% (six knees). Postoperative function scores were good or excellent in only 40% (13 knees). The average postoperative flexion arc was 110 °. If patients involved in worker’s compensation cases are excluded from analysis, the results improved substantially, with range of motion averaging 113 °, and Knee Society knee scores and function scores averaging 92 points and 77 points, respectively. Excluding the five patients involved in workmen’s compensation cases, knee scores were good or excellent in 91% of patients (25 knees) and function scores were good or excellent in 50% of patients (14 knees). Three revisions were performed for aseptic failure; one additional patient has radiographic evidence of tibial loosening, representing an aseptic failure rate of 12.5% at 8 years. Although slightly higher than observed in older patients, this failure rate still may be considered acceptable for this population of patients with severely affected knees who are not considered candidates for nonarthroplasty surgery. Despite a slightly higher tendency for aseptic failures in this group of patients, cemented total knee arthroplasty may provide some patients younger than 40 years of age with severe debilitating and recalcitrant osteoarthrosis, an important option with reasonable mid-and long-term results.

Total knee arthroplasty has proven effective in the treatment of primary osteoarthrosis in elderly patients. In an attempt to avoid performing total knee arthroplasty in younger patients with painful, debilitating osteoarthrosis, various treatment options historically have been offered. ^{14,18,25,28,34} Nonsurgical interventions, such as nonsteroidal antiinflammatory medications, hyaluronic acid or corticosteroid injections, and glucosamine and chondroitin Sulfate supplementation, all have been used in young patients with mixed results. ^{1,2,7,9,14,15,19,24,29,33,34} Additionally, nonprosthetic surgical options such as arthroscopic lavage or debridement, periarticular osteotomy, and arthrodesis, have a role in a select group of patients although the results tend to deteriorate with time. ^3,18,25

Total knee arthroplasty has been shown to be effective in patients 55 years of age and younger, in those patients with rheumatoid arthritis, and in patients with osteoarthritis. ^{10,12, 13,16,26,30} The concern in the orthopaedic community of accelerated failure in young, active patients because of repetitive loading, or presumed difficulty in performing the inevitable revisions, has prompted many surgeons to arbitrarily reserve total knee arthroplasty for those patients approaching or older than 65 years of age. It is being more commonly observed that many young patients present with severe debilitating arthrosis, which may not be amenable to surgical interventions other than total knee arthroplasty. It is the purpose of the current study to specifically address the results of total knee arthroplasty, after a minimum of 5-year followup, in patients who are younger than 40 years of age who have osteoarthrosis.

MATERIALS AND METHODS

A letter was sent to members of the Knee Society, soliciting data on patients, 40 years of age or younger, who underwent total knee arthroplasty for osteoarthrosis and were followed up for at least 5 years. Complete data, excluding those young patients who underwent total knee arthroplasty for inflammatory arthritis or secondary to fracture, were available for 44 patients from five centers. In this collected series of patients, 12 patients with Porous Coated Anatomic arthroplasties (Howmedica, Rutherford NJ) were excluded from analysis in the study because failures of this prosthesis likely were not related to patient age per se, but more to flaws in material qualities and design of the implant that ultimately have lead to its removal from the consumer market. The implants in nine of those 12 patients failed because of polyethylene wear and component loosening. This leaves 32 knees for inclusion in this study.

The average patient age was 35 years (range, 22–40 years). There were 13 women. The average height of the patients was 69 inches (range, 62–78 inches). The average weight of the patients was 200 lb (range, 109–380 lb). The average body mass index ([703 × weight]/height²) was 29.5. This suggests that the average patient was not by definition considered to be obese. All patients had either primary osteoarthrosis or posttraumatic arthrosis secondary to ligament injury. Eleven patients were receiving worker’s compensation related to injuries that eventually lead to arthrosis. Seventy-one prior surgeries were performed in 26 patients (mean, 2.7 surgeries per patient; range, 0–17 surgeries).

Total knee arthroplasties were performed between 1982 and 1994. Implants included Series 7000 (Osteonics, Allendale, NJ, n = 3), Insall-Burstein II (Zimmer, Warsaw, IN, n = 4), Anatomic Gradiated Component (Biomet, Warsaw, IN, n = 13), Duracon (Howmedica, Rutherford, NJ, n = 3), Press-Fit condylar (Johnson & Johnson/DePuy, Warsaw, IN, n = 7), Maxim (Biomet, n = 1), and Techmedica (Sulzer Orthopaedics, Hampshire, United Kingdom, n = 1).

The tibial and femoral components were cemented in 27 knees; two patients had hybrid fixation (tibia cemented, femur uncemented); and three were uncemented. Patellar resurfacing was performed in 28 of the 32 patients.

RESULTS

Preoperative flexion contractures averaged 8° (range, 0°–30°). Flexion arc averaged 92° (range, 20°–125°). Preoperative Knee Society knee scores ¹⁷ averaged 47 points (range, 5–80 points). Preoperative function scores averaged 45 points (range, 5–70 points).

Followup averaged 7.9 years (range, 5–17 years). Postoperative flexion contractures decreased to an average of 1.5° (range, 0°–20°). Flexion arc increased to 110° (range, 80°–130°). Postoperative Knee Society knee scores averaged 88 points (range, 53–96 points) and function scores averaged 70 points (range, 35–100 points).

Three revisions had been performed for polyethylene wear and loosening, at a mean 8-year interval after component implantation. Prostheses revised for aseptic loosening included Anatomic Gradiated Component (n = 1, 2 years), Insall-Burstein II (n = 1, 7 years), and Osteonics (n = 1, 5 years). One implant has been revised for infection. An additional patient has evidence of complete lucency beneath the tibial component, but revision has not yet been performed. Two of the aseptic failures occurred with cemented components; two occurred with cementless components. Using revision for aseptic failure as an endpoint, implant survival in this patient population at 8 years is 90.6%. Using radiographic loosening as an endpoint, aseptic failures occurred in 12.5% implants at 8 years. Aseptic failures occurred in two of the three cementless tibial implants and only two of the 29 cemented tibial implants.

Incomplete radiographic loosening beneath the tibial tray was observed in one patient. One patient required two neurectomies for neurogenic pain. One patient underwent successful treatment of a supracondylar femur fracture, but required three manipulations under anesthesia for arthrofibrosis. Eleven manipulations under anesthesia were necessary in patients in the current series. Patellar realignment was necessary in one patient.

Overall, Knee Society knee scores were considered good or excellent in 82% of patients and fair or poor in 18%. Postoperative function scores were good or excellent in only 40%.

When patients involved in worker’s compensation cases are excluded from the analysis, the results improved substantially, with range of motion (ROM) averaging 113°, Knee Society knee score averaging 92 points, and function score averaging 77 points. Excluding patients involved in worker’s compensation cases, knee scores were good or excellent in 91% of patients and function scores were good or excellent in 50% of patients.

There did not seem to be an association between the likelihood of unsuccessful results and prior surgeries, particularly proximal tibial osteotomy or patellectomy. One patient, who had an infection develop postoperatively, had a prior anterior cruciate ligament reconstruction and Maquet tibial tubercle osteotomy through separate incisions.

DISCUSSION

Total knee arthroplasty has proven effective long-term for elderly patients with osteoarthrosis. ^27,31 Our highly active and athletically oriented society has promoted a population at risk for ligamentous and chondral injuries in young patients that ultimately may induce chondral degeneration in this young patient population. ^21,22 Young patients with debilitating osteoarthrosis present a different challenge than those patients with rheumatoid arthritis, attendant to their higher activity levels. Concerns about component wear and implant longevity in this young group of patients have led many orthopaedic surgeons to withhold arthroplasty and to recommend other less predictable surgical and nonsurgical alternatives for this group of patients. ⁴ Often times, when total knee arthroplasty is pursued, knees have multiple scars, arthrofibrosis, and malalignment that ultimately make arthroplasty technically more challenging and may compromise results.

Although several articles have addressed the issue of total knee arthroplasty for patients younger than 55 years, to the authors’ knowledge, no other study specifically has addressed the role of total knee arthroplasty in patients younger than 40 years of age. ^8,10,12,26 The current series has shown that the followup Knee Society scores and function in this young group of patients is inferior to those scores observed in the more elderly counterparts. Some of this may be related to unreasonable expectations of the patients and to the reality of inferior results of total knee arthroplasty often observed in knees with multiple scars or arthrofibrosis. ²⁰ It is important to consider that these patients present some of the most difficult challenges in knee surgery—“easy” cases of osteoarthrosis in patients younger than 40 years generally are treated with nonarthroplasty options, such as distal femoral or proximal tibial osteotomies. Therefore, the results presented in the current series of 32 knees should be considered successful. If one excludes the excessive rate of failures observed in the presence of cementless tibial components, the option of total knee arthroplasty becomes even more attractive in this group of patients. In the entire series of young, active patients, the results of arthroplasty met or exceeded the expectations of 83% of the patients; 17% of patients were disappointed.

The issue of poor results in patients receiving worker’s compensation has been elaborated by Mont et al. ²³ In a series of 42 patients who underwent total knee arthroplasty, the results at a mean followup of 80 months were significantly worse (p < 0.01) in those patients receiving compensation compared with those who were not receiving compensation. Despite similar objective parameters in ROM, stability, and radiographic alignment, subjective scores were significantly inferior. The results in the current series corroborate those findings.

Diduch et al ¹⁰ reviewed the results of 103 knees using one implant in patients 55 years of age or younger for osteoarthrosis or posttraumatic arthrosis. No distinction was made regarding age categories. At a mean followup of 8 years, the average Knee Society knee score was 94 points, and the average functional score was 89 points. The authors reported an overall anticipated 18-year survival rate of 94%, although when patellar revision or spacer exchange were included in failures, expected survivorship was reduced to 87% at 18 years. ¹⁰ Activity level, according to the scoring system of Tegner and Lysholm ³² were improved in all but two patients.

In a separate study of total knee arthroplasties in patients younger than 55 years of age, Ranawat et al ²⁶ analyzed 93 total knee arthroplasties in 62 patients. Seventy-six patients had rheumatoid arthritis and 17 had osteoarthritis. The authors did not subdivide the patients according to age or activity level. However, good or excellent results were observed in 94% of patients with osteoarthritis and in 99% of those with rheumatoid arthritis at a mean followup of 6.1 years. Ninety-six percent survivorship was observed at 10 years using an endpoint of clinical or radiologic failure.

Duffy et al ¹² reported the results of 74 consecutive total knee arthroplasties in 54 patients who were 55 years of age or younger. Only 18 patients had a diagnosis of osteoarthrosis or posttraumatic arthrosis with the remainder having inflammatory or infectious arthritis. The authors reported a final Knee Society knee score of 84 points, and functional score of 60 points at a minimum 10-year followup. Two revisions were necessary either because of ligamentous laxity or aseptic loosening. Implant survivorship at 10 years was estimated to be 99% and at 15 years implant survivorship was 95%. These data are comparable with data observed in the current series, although there was no distinction made in the results between those patients who had inflammatory arthritides or osteoarthrosis. Additionally in the series by Duffy et al, ¹² all implants were cemented.

Dalury et al ⁸ reviewed 103 total knee arthroplasties in patients younger than 45 years of age. Once again, the majority of patients had rheumatoid arthritis (87%), with only a small proportion having osteoarthrosis. No distinction was made between the results in patients with inflammatory arthritis and osteoarthritis. At a mean followup of 7.2 years, no revisions were necessary for aseptic failure, but three revisions were necessary for patella failures and one for infection.

Coyte et al ⁶ reviewed the hospitalization statistics for 1301 revision total knee arthroplasties performed in Ontario, Canada between 1984 and 1991. They found that rheumatoid arthritis was associated with the lowest rate of revision (p = 0.03) and that young age (55 or younger) was associated with the shortest implant survival time (p < 0.0001). Patients who were 55 years of age had the greatest risk for implant failure; patients who were 65 years of age had a relative 85% risk of implant failure compared with the 55-year-old patients; and the patients who were 75 years of age had a 60% risk of failure relative to the 55-year-old patients. However, the authors also reported that the overall survivorship was so high (92% to 96%) that these differences were quantitatively small in relationship to the entire population of patients undergoing total knee arthroplasty. ⁶

The current results suggest that total knee arthroplasty for patients younger than 40 years of age may provide pain relief and return to functional activity, but results are slightly inferior to those observed in the elderly population. However, these patients are the most severely affected, with gonarthrosis not amenable to realignment procedures. In the current series various implants had been used and failures seemed to be related to older implant designs with poor material properties or the use of cementless components. Multiple prior surgeries did not seem to affect the overall results. Prior proximal tibial osteotomy did not effect outcomes in this group of patients.

Unlike total hip arthroplasty for very young patients, cemented total knee arthroplasty may produce excellent mid-and long-term results in patients younger than 40 years of age with end-stage arthrosis, provided they are willing to modify their activity levels. ^5,11 However, nonoperative interventions should be used at all costs until they are no longer effective. Additionally, other nonarthroplasty surgical options can be considered. Arthroscopic debridement and lavage, however, have provided variable pain relief. ³ For those patients with considerable limb malalignment, and unicompartmental disease, and for those patients who wish to continue working in labor-intensive jobs or continue their athletic involvement periarticular osteotomy or fusion (rarely) should be offered and encouraged before total knee arthroplasty. Today, most patients will not consent to arthrodesis. Proximal tibial osteotomy or distal femoral osteotomy for patients with unicompartmental arthrosis with malalignment of the mechanical axis may provide 60% to 65% satisfactory short-term results; however, deterioration of results beyond 5 years is common. ^18,25 Unicompartmental arthroplasty in young active patients has been fraught with complications, including accelerated failures. ²⁸

Despite a slightly higher tendency for aseptic failures in this group of patients, cemented total knee arthroplasty may provide some patients younger than 40 years of age, who have severe debilitating and recalcitrant osteoarthrosis, an important option with reasonable mid-and long-term results. Appropriate counseling before surgery regarding expectations and the possibility of eventual failure is important. Enhancing implant longevity in this group of patients is predicted on appropriate surgical technique, using carefully selected implants of acceptable design and material properties, and limiting the activity levels of these young patients.

Acknowledgments

The authors thank David Hungerford, MD; Kenneth Krackow, MD; Roger Emerson, MD; William Head, MD; Merrill Ritter, MD; Philip Faris, MD; E. Michael Keating, MD; Victor Goldberg, MD; and Adolf Lombardi, MD, who submitted cases for use in this series.