Preoperative Weight Gain Before Knee Arthroplasty Does Not Increase Revision Risks

A retrospective investigation examining 12,990 patients who underwent elective, primary, unilateral total knee arthroplasty at an urban academic medical center has demonstrated that substantial weight gain between the date of surgical booking and the date of surgery does not increase the risk of postoperative revision surgery or short-term clinical complications. Published in the journal Arthroplasty by Anzar Sarfraz, Theodor Di Pauli von Treuheim, Garrett Ruff, Braden V. Saba, Farouk Khury, Ran Schwarzkopf, Joshua C. Rozell, and Vinay K. Aggarwal, the analysis evaluated patients treated between September 2015 and March 2024. The research team observed that 60.4 percent of surgical candidates gained body weight during the waiting period leading up to their operative date. While patients who experienced a clinically meaningful increase in body mass index of more than 5 percent faced longer initial hospitalizations and were less frequently discharged directly home, their rates of 90-day adverse events and overall revision procedures remained statistically indistinguishable from those who maintained or lost weight.

These clinical findings emerge against a backdrop of escalating surgical volumes and rising obesity rates nationwide. In the United States, procedural demand for total knee arthroplasty is projected to expand by 139 percent by the year 2040 and by 469 percent by 2060, an increase driven in substantial part by climbing obesity rates among surgical candidates, which grow by roughly 0.97 percent annually. Obesity represents a firmly established clinical risk factor for adverse postoperative events, including periprosthetic joint infection, postoperative coagulopathies, and premature implant revision. Published registry and clinical data indicate that while periprosthetic infection rates remain beneath 1 percent in non-obese individuals, they reach approximately 2 percent among obese patients and climb by an additional 3 to 4 percent among morbidly obese cohorts. Elevated body mass index is likewise known to prolong operative durations, lengthen hospital stays, increase discharge rates to inpatient rehabilitation centers, and inflate overall 90-day healthcare costs.

In response to these procedural risks, the American Association of Hip and Knee Surgeons has advised delaying elective joint replacement for patients whose body mass index exceeds 40 kilograms per square meter, concluding that procedural risks outweigh functional benefits at higher weights. However, observational research cited by the study authors highlights the difficulty patients face when attempting to comply with strict preoperative weight cutoffs. For example, an observational review by Wilson and colleagues demonstrated that only 19 percent of morbidly obese candidates succeeded in lowering their body mass index below the 40 threshold within two years, and merely 5 percent ultimately proceeded to surgery. Similarly, prospective data from Springer and colleagues revealed that only 6.9 percent of 289 deferred patients achieved their target weight threshold despite nearly a quarter attempting formal lifestyle modifications or bariatric surgery. Furthermore, research by Riddle and colleagues demonstrated that substantial preoperative weight loss frequently predicts rapid postoperative weight regain, showing that patients often experience weight instability around the time of surgery.

Despite widespread scrutiny of preoperative optimization and long-term postoperative trajectories, the specific window separating surgical booking from the operative date has historically remained underexplored in orthopedic literature. This scheduling window holds unique clinical significance because patients have already satisfied institutional inclusion criteria and secured an operative slot. Once surgical clearance is confirmed, behavioral incentives can shift, yet orthopedic teams routinely proceed under the clinical assumption that patient weight remains stable through the final preoperative interval. In contemporary practice, weight fluctuations across this period carry immediate practical consequences, particularly as growing numbers of primary procedures migrate toward ambulatory surgery centers. Because freestanding ambulatory centers apply strict body mass index cutoffs to determine patient eligibility, unexpected weight gains between initial surgical booking and the morning of surgery risk triggering last-minute venue cancellations, facility transfers, or logistical postponements.

To rigorously characterize these perioperative dynamics, the investigative team reviewed consecutive adult patients aged 18 and older undergoing unilateral elective total knee arthroplasty for end-stage osteoarthritis under institutional review board approval. The protocol excluded revision operations, bilateral procedures, non-elective surgeries, and cases lacking at least 90 days of documented postoperative follow-up. Institutional guidelines at the academic facility recommended that patients with a body mass index greater than 40 postpone scheduling to pursue weight loss, while maintaining an absolute ineligibility ceiling at 50 kilograms per square meter that permitted exceptions only under extenuating circumstances. Patient heights and weights were captured directly from electronic medical records using calibrated institutional scales operated by clinical staff, first at the preoperative visit establishing surgical indication and booking, and again on the morning of surgery. Weight change was expressed both as absolute difference and as percentage variation relative to the booking date.

The overall study cohort was stratified into three exposure groups according to percentage changes in body mass index: Group 1 comprised individuals who lost weight; Group 2 included those who gained up to 5 percent of their baseline body mass index; and Group 3 comprised those who experienced a gain exceeding 5 percent, a boundary selected in accordance with the United States Food and Drug Administration definition of clinically meaningful weight change. Across the unadjusted pre-match sample of 12,990 patients, 39.6 percent lost weight after scheduling, 41.2 percent demonstrated modest weight gains between zero and 5 percent, and 19.2 percent gained more than 5 percent. To resolve baseline demographic disparities, investigators implemented a 1:1:1 propensity-score matching protocol centered on Group 3. Matching balanced age, baseline body mass index at scheduling, sex, smoking history, American Society of Anesthesiologists physical status classification, diabetes history, and socioeconomic status estimated via ZIP-code median household income quartiles. Standardized mean differences across all matched baseline covariates fell below 0.1, indicating balanced groups.

Within the matched study cohort, the average delay spanning surgical booking to the operating room was 61 days across all participants, averaging 61.1 days in Group 1, 57.3 days in Group 2, and 65.3 days in Group 3. While this intergroup scheduling disparity attained statistical significance, the investigators emphasized that a single-week variation holds minimal clinical relevance in the context of typical two-month scheduling intervals. On the operative day, mean body mass index measurements were 30.2 in Group 1, 31.8 in Group 2, and 34.0 in Group 3. Evaluation of medical histories showed that glucagon-like peptide-1 receptor agonist prescriptions were documented in 4.8 percent of Group 1, 4.4 percent of Group 2, and 3.5 percent of Group 3, a distribution that did not differ significantly between groups. Similarly, previous bariatric surgery histories were evenly distributed across cohorts, and the duration of postoperative monitoring averaged 2.2 years across all study participants.

Analysis of acute perioperative outcomes revealed that significant weight gain directly affected inpatient recovery metrics. Length of stay was longest among Group 3 patients, averaging 50.3 hours compared with 48.3 hours in Group 1 and 48.0 hours in Group 2 among propensity-matched cohorts, a statistically significant extension. In parallel, the proportion of patients discharged directly home rather than transferred to specialized rehabilitation or nursing facilities was lowest in Group 3 at 88.7 percent, compared with 89.4 percent in Group 1 and 91.1 percent in Group 2. In unadjusted pre-matched analyses, these disparities were similarly evident, with Group 3 recording an average hospital stay of 50.3 hours versus 48.6 hours in Group 1 and 47.1 hours in Group 2, alongside a home discharge rate of 88.7 percent compared with 89.8 percent and 91.7 percent in the respective comparator cohorts.

Despite differences in initial hospitalization metrics, subsequent clinical outcomes across extended postoperative follow-up showed no statistical variation linked to preoperative weight gain. The investigators reported no significant differences among the three exposure groups regarding operative times, 90-day emergency department visits, or 90-day hospital readmission rates. Likewise, postoperative deep vein thrombosis and infection-related surgical interventions, specifically irrigation and debridement or two-stage septic revisions, occurred at comparable frequencies regardless of interval weight changes. Multivariable logistic regression modeling confirmed that percentage changes in body mass index between booking and surgery exerted no statistically detectable impact on all-cause revision or septic revision rates, indicating that modest weight fluctuations immediately preceding surgery do not dictate midterm implant survival.

Multivariable logistic regression of the overall cohort identified independent baseline determinants of procedural revisions. All-cause revision risks decreased with advancing patient age (odds ratio 0.95), Asian racial background (odds ratio 0.43), and longer intervals from surgical booking to surgery (odds ratio 0.98 per week). Conversely, revision risks increased significantly among Black patients (odds ratio 1.65) and individuals categorized as American Society of Anesthesiologists class IV (odds ratio 6.36). Septic revisions similarly decreased with age (odds ratio 0.95), but demonstrated an elevated association with glucagon-like peptide-1 receptor agonist therapy (odds ratio 2.81). However, the authors underscored that this pharmacotherapy finding was based on a small subpopulation of 601 treated patients and only eight infection events (1.3 percent) compared with 53 events among untreated patients (0.4 percent), severely limiting the statistical power of that secondary observation.

In linear regression models assessing drivers of weight fluctuation, higher baseline body mass index, male sex, residency in top-quartile income neighborhoods, and former smoking history correlated with modest weight loss prior to surgery. Patient age, race, and the duration of surgical booking delays did not predict weight changes across the broader cohort. When stratifying patients by initial weight categories, however, booking delay exerted a selective effect among overweight individuals with baseline body mass index values between 25 and 30. Within this specific subgroup, scheduling delays of one to four months produced significantly larger median weight gains (+0.40) than delays under one month (+0.17), with delays exceeding four months yielding median gains of +0.50. This pattern was entirely absent among patients with baseline body mass index values under 25 or over 30, suggesting that overweight individuals represent a subpopulation uniquely prone to weight gain during prolonged preoperative waits.

Subject of Research: Impact of preoperative weight fluctuations between surgical booking and knee arthroplasty on surgical outcomes

Article Title: Does weight gain from time of indication to date of surgery affect outcomes in total knee arthroplasty?

Article References: Sarfraz, A., Di Pauli von Treuheim, T., Ruff, G., Saba, B. V., Khury, F., Schwarzkopf, R., Rozell, J. C., & Aggarwal, V. K. (2026). Does weight gain from time of indication to date of surgery affect outcomes in total knee arthroplasty?. Arthroplasty, 8(1), Article 35. https://doi.org/10.1186/s42836-026-00379-6

Image Credits: AI Generated

DOI: 10.1186/s42836-026-00379-6

Keywords: total knee arthroplasty, body mass index, preoperative weight gain, surgical booking delay, postoperative complications, all-cause revision, septic revision, length of stay, discharge disposition, osteoarthritis, propensity score matching, periprosthetic joint infection, American Association of Hip and Knee Surgeons, GLP-1 receptor agonists, orthopedic surgery, hospital readmission, surgical outcomes