Abstract
Introduction
Given the association of osteoarthritis with obesity, the typical patient requiring total knee arthroplasty (TKA) is often obese. Obesity has been shown to negatively influence outcomes following TKA, as it is associated with increased perioperative complications and poorer clinical and functional outcomes. Achieving proper limb alignment can be more difficult in the obese patient, potentially requiring a longer operation compared to non-obese patients. Patient specific instrumentation (PSI), a technique that utilizes MR- or CT-based customized guides for intraoperative cutting block placement, may offer a more efficient alternative to manual instruments for the obese patient. We hypothesize that the additional information provided by a preoperative MRI or CT may allow surgeons to achieve better alignment in less time compared to manual instrumentation. The purpose of this study was to assess whether PSI offers an improved operation length or limb alignment compared to manual instruments for nonmorbidly and morbidly obese patients.
Methods
In this retrospective cohort study, we evaluated 77 PSI TKA and 25 manual TKA performed in obese patients (BMI≥30) between February 2013 and May 2015. During this period, all patients underwent PSI TKA unless unable to undergo MR scanning. All cases were performed by a single experienced surgeon and utilized a single implant system (Zimmer Persona™). PSI cases were performed using the MR-based Zimmer Patient Specific Instrumentation system. Tourniquet times were recorded to determine length of operation. Long-standing radiographs were obtained preoperatively and 4-weeks postoperatively to evaluate limb alignment. Cases were subdivided by nonmorbid obesity (30≤BMI<40) and morbid obesity (BMI≥40) to assess the effect of increasing obesity on outcomes.
Results
PSI and manual cohorts were similar with regards to age, gender, and preoperative alignment. Tourniquet time was significantly shorter in the PSI group for nonmorbidly obese patients (PSI 49.8 minutes vs manual 58.3 minutes; p=0.005) (Figure 1). Postoperative mechanical axis was similar between groups for both nonmorbidly obese (PSI 1.8° vs manual 2.9°; p=0.338) and morbidly obese patients (PSI 4.0° vs manual 3.6°; p=0.922). Mechanical axis outliers (greater than 3° neutral), though nonsignificant, were fewer in the PSI group for nonmorbidly obese (PSI 21.8% vs manual 35.3%; p=0.318) and morbidly obese patients (PSI 46.1% vs manual 75.0%; p=0.362).
Discussion
We found that PSI significantly shortened operation length for nonmorbidly obese patients compared to manual instruments. Obesity is strongly associated with increased perioperative infection rates, as is prolonged operation length. The decreased operation length achieved with PSI in the nonmorbidly obese patient may as a consequence decrease infection rates, though further study is necessary. Though not statistically significant, PSI showed a trend toward decreasing overall mechanical axis outliers for both nonmorbidly obese and morbidly obese patients. The use of patient specific instrumentation compared to manual instruments has been controversial in the literature. However, patient specific instrumentation may be favorable in the obese patient, offering a shorter operation length and possibly improved alignment.
