Robert S. Namba, MD, Heather A. Prentice, PhD, MPH, Elizabeth W. Paxton, PhD, MA, Adrian D. Hinman, MD, and Matthew P. Kelly, MD
Journal of Bone and Joint Surgery, American Volume. 2020
https://doi.org/10.2106/JBJS.19.01440
Summary by Ahna J. Donahue, BA
Nearly 50% of total knee arthroplasty (TKA) procedures in North America are performed with antibiotic-loaded bone cement (ABC) [1]. Yet, the efficacy of this more expensive alternative [2] to regular bone cement in reducing the risk of infection in patients undergoing TKA remains debated. In particular, the effectiveness of ABC over regular bone cement in reducing the risk of infection among high-risk subgroups such as diabetes, body mass index (BMI) ≥ 35kg/m2, or American Society of Anesthesiologists (ASA) classification ≥ 3 remains unknown [3]. This study sought to investigate the effectiveness of commercially available ABC in reducing the risk of infection following TKA both overall and among high-risk subgroups.
The Kaiser Permanente Total Joint Replacement Registry (KP TJRR) was utilized to identify 87,018 TKAs from 2008 to 2016 of which 27,539 (31.6%) were performed with ABC. Among the high-risk subgroups of diabetes, BMI ≥ 35kg/m2 and ASA ≥ 3, there were 16,644 diabetic patients, of which 38.6% received ABC (n=6,431), 21,691 patients with high BMI (35.25% received ABC; n=7,631) and 31,557 patients with a high ASA status (34.0% received ABC; n=10,719).
At 3 years, 261 (0.95%) of ABC recipients experienced an infection compared to 488 (0.82%) in the regular cement group. In the propensity-score weighted Cox models, there was no evidence to support that ABC reduced the risk of infection. Similarly, there was no evidence that regular cement was noninferior to ABC. Furthermore, there was no association with the risk of aseptic revision or revision for aseptic loosening observed when using ABC compared with regular cement. Among patients with diabetes, at 3 years, 0.60% of ABC recipients experienced an infection compared to 1.08% in those that received regular cement. Notably, ABC was associated with a significantly lower risk of infection (p = 0.049) and there was insufficient evidence that regular cement was noninferior to ABC in this group. In the high BMI subgroup, no association was observed for infection risk and there was no evidence of non-inferiority at 3 years. Similarly, in the high ASA status subgroup, there was no evidence for a reduced risk of infection when using ABC and noninferiority could be not concluded at 3 years.
In conclusion, ABC was found to reduce the risk of infection among patients with diabetes. However, in the overall group, the use of ABC was not cost-effective and generally did not reduce the risk of infection among primary TKA patients. One notable limitation of the study was the use of a non-standard protocol for selecting ABC or regular cement, allowing for potential bias in surgeon selection of ABC for particular patients. In addition, the study was not able to evaluate the potential effects of having more than one high-risk factor, due to the small sample size of subgroups.
References:
- Gandhi R. Antibiotic Bone Cement and the Incidence of Deep Infection after Total Knee Arthroplasty. The Journal of arthroplasty. 2009;24(7):1015-1018. doi:10.1016/j.arth.2008.08.004
- Gandhi R. Antibiotic Bone Cement and the Incidence of Deep Infection after Total Knee Arthroplasty. The Journal of arthroplasty. 2009;24(7):1015-1018. doi:10.1016/j.arth.2008.08.004
- Gandhi R. Antibiotic Bone Cement and the Incidence of Deep Infection after Total Knee Arthroplasty. The Journal of arthroplasty. 2009;24(7):1015-1018. doi:10.1016/j.arth.2008.08.004