William S. Harmsen

William S. Harmsen

Auto-generated profile, based on publicly available information

William S. Harmsen

Department of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, Minnesota

Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 February 2016 in Journal of Bone and Joint Surgery
Verified
Journal Article
Closed access
EWEric R. Wagner
Eric R. Wagner
AKAtul F. Kamath
Atul F. Kamath
KFKristin M. Fruth
Kristin M. Fruth
WHWilliam S. Harmsen
William S. Harmsen
DBDaniel J. Berry
Daniel J. Berry
Implant survival according to BMI as a continuous variable. Methods: Using prospectively collected data from our institutional total joint registry, we analyzed 21,361 consecutive hips (17,774 patients) treated with primary total hip arthroplasty between 1985 and 2012 at a single institution. The average BMI at the time of surgery was 28.7 kg/m2 (range, 15 to 69 kg/m2). Estimates of revision surgery and common complications associated with BMI were analyzed using the Kaplan-Meier method of assessing survivorship, with associations of outcomes assessed using a Cox model. Results: Utilizing smoothing spline parameterization, we found that reoperation (p < 0.001) and implant revision or removal rates (p = 0.002) increased with increasing BMI. Increasing BMI was associated with increased rates of early hip dislocation (p = 0.02), wound infection, and, most strikingly, deep periprosthetic infection (a hazard ratio of 1.09 per unit of BMI >25 kg/m2; p < 0.001). However, we found no association between increasing BMI and any revision for mechanical failure of the implant or between increasing BMI and revision for aseptic implant loosening. There was an inverse correlation between increasing BMI and risk of revision for bearing wear. Conclusions: The rates of reoperation, implant revision or removal, and common complications after total hip arthroplasty were strongly associated with BMI. Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence....

21 December 2016 in Journal of Bone and Joint Surgery
Verified
Journal Article
Closed access
EWEric R. Wagner
Eric R. Wagner
AKAtul F. Kamath
Atul F. Kamath
KFKristin Fruth
Kristin Fruth
WHWilliam S. Harmsen
William S. Harmsen
DBDaniel J. Berry
Daniel J. Berry
Background: High body mass index (BMI) is associated with increased rates of complications after total knee arthroplasty. To date, to our knowledge, studies have examined risk as a dichotomous variable using specific BMI thresholds. The purpose of this investigation was to quantify implant survival and the risk of common complications after total knee arthroplasty using BMI as a continuous variable. Methods: Using prospectively collected data from our institutional total joint registry, we analyzed 22,289 consecutive knees, in 16,136 patients, treated with primary total knee arthroplasty from 1985 to 2012. The mean BMI of these patients at the time of the surgical procedure was 31.3 kg/m2 (range, 11 to 69 kg/m2). The Kaplan-Meier survival method was used to estimate survivorship, reoperations, and common complications, with associations of outcomes assessed using a Cox regression model. Results: Utilizing smoothing spline parameterization, we found that reoperation (p < 0.001) and implant revision or removal rates (p < 0.001) increased with increasing BMI after total knee arthroplasty. Increasing BMI also was associated with increased rates of wound infection (hazard ratio [HR], 1.07; p < 0.001) and deep infection (HR, 1.08; p < 0.001) per unit of BMI over 35 kg/m2. A BMI of 35 to 40 kg/m2 was associated with a higher rate of implant revision for aseptic loosening (p < 0.001) and for polyethylene wear (p < 0.001) compared with a BMI of 18 to 24.99 kg/m2. There was no correlation between BMI and risk of venous thromboembolism, tibiofemoral instability, or need for knee manipulation. Conclusions: The rates of reoperation, implant revision or removal, and many common complications after total knee arthroplasty were strongly associated with BMI. Level of Evidence: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.

7 June 2017 in Journal of Bone and Joint Surgery
Verified
Journal Article
Closed access
EWEric R. Wagner
Eric R. Wagner
Matthew T. HoudekMatthew T. Houdek
Matthew T. Houdek
CSCathy Schleck
Cathy Schleck
WHWilliam S. Harmsen
William S. Harmsen
JSJoaquin Sanchez-Sotelo
Joaquin Sanchez-Sotelo
BEBassem T. Elhassan
Bassem T. Elhassan
Background: Although obesity is associated with increased complication rates after lower-extremity arthroplasty, there is a relative paucity of studies examining the effect of body mass index (BMI) on shoulder arthroplasty. The purpose of this investigation was to evaluate the effect of BMI on implant survival and the rate of complications after shoulder arthroplasty. Methods: Using an institutional total joint registry, 4,567 consecutive shoulder arthroplasty cases from 1970 to 2013 were studied. The mean BMI was 29.7 kg/m2 (range, 14 to 66 kg/m2), with 1,622 patients (36%) with a BMI of 30 to 40 kg/m2 and 297 patients (7%) with a BMI of >40 kg/m2. There were 2,493 female patients (55%). BMI was dichotomized after examination of the smoothing spline curve. The associations of factors and complications were assessed using Cox proportional hazard regression analysis. Results: Increasing BMI was associated with an increased risk of a revision surgical procedure, reoperation, revision for mechanical failure, and superficial infection, and it was negatively associated with risk of a periprosthetic fracture. The risk of a revision surgical procedure increased in a linear fashion with increasing BMI (hazard ratio [HR], 1.05, or a 5% increased risk per 1 unit of BMI; p = 0.03). Increased BMI was also associated with an increased risk of revision for mechanical failure (HR, 1.05; p = 0.004). In a multivariate model, the association of BMI and risk of a revision for any reason, revision for mechanical failure, and reoperation maintained significance (p ≤ 0.02). The most marked association between increasing BMI and any complication in shoulder arthroplasty was its association with superficial wound infection (HR, 1.09; p = 0.03). Conclusions: Increasing BMI is strongly associated with increased rates of revision surgical procedures and postoperative complications after shoulder arthroplasty. It is important to consider these findings when counseling patients, estimating risks, and estimating complication risks in policy decisions. Level of Evidence: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.

1 February 2015 in Journal of Bone and Joint Surgery
Verified
Journal Article
Closed access
PDF file
PRPhilipp von Roth
Philipp von Roth
MAMatthew P. Abdel
Matthew P. Abdel
WHW. Scott Harmsen
W. Scott Harmsen
DBDaniel J. Berry
Daniel J. Berry
Uncemented jumbo cups are commonly used for acetabular revision because they are technically straightforward to implant and provide good intermediate-term results. Understanding long-term survival is particularly important because this method is common and because jumbo cups do not provide notable bone stock restoration. The purpose of the present study was to determine the twenty-year results of jumbo cup use during revision total hip arthroplasty. In the original publication, we reported on eighty-nine patients who underwent revision with an uncemented jumbo cup with a single design (Harris-Galante) prior to 1993. The Harris Hip Score (HHS), radiographic results, and Kaplan-Meier survivorship curves were evaluated. Mean follow-up was twenty years. The mean postoperative HHS was 71 compared with 56 preoperatively (p = 0.001). A total of five jumbo cups were revised for aseptic loosening; one, for infection; and one, for recurrent dislocation. Eight liners were revised with retention of the metal acetabular component: six during femoral component revision, one for wear, and one for recurrent dislocations. Twenty-year survivorship was 88% free from aseptic loosening of the metal acetabular component, 85% free from aseptic loosening or radiographic evidence of definite loosening of the metal acetabular component, and 83% free from revision of the metal acetabular component for any reason. The twenty-year results of revision with uncemented jumbo acetabular components demonstrated acceptable clinical outcomes and radiographic stability. These results justify the use of jumbo cups as a common method of acetabular revision. Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.

1 September 2017 in Journal of Trauma and Acute Care Surgery
Verified
Journal Article
Closed access
MPMyung S Park
Myung S Park
GSGrant M. Spears
Grant M. Spears
KBKent R. Bailey
Kent R. Bailey
AXAiling Xue
Ailing Xue
MFMichael J. Ferrara
Michael J. Ferrara
JHJohn A. Heit
John A. Heit
BACKGROUND Reliable biomarkers predictive of venous thromboembolism (VTE) after acute trauma are uncertain. The objective of the study was to identify risk factors for symptomatic VTE after trauma, including individual plasma coagulome characteristics as reflected by thrombin generation. METHODS In a prospective, case-cohort study, trauma patients were enrolled over the 4.5-year period, 2011 to 2015. Blood was collected by venipuncture into 3.2% trisodium citrate at 0, 6, 12, 24, and 72 hours after injury and at hospital discharge. Platelet poor plasma was stored at −80 °C until analysis. Thrombin generation, as determined by the calibrated automated thrombogram (CAT) using 5 pM tissue factor (TF)/4 μM phospholipid (PS), was reported as peak height (nM thrombin) and time to peak height (ttPeak [minutes]). Data are presented as median [IQR] or hazard ratio with 95% CI. RESULTS Among 453 trauma patients (injury severity score = 13.0 [6.0, 22.0], hospital length of stay = 4.0 [2.0, 10.0] days, age = 49 [28, 64] years, 71% male, 96% with blunt mechanism, mortality 3.2%), 83 developed symptomatic VTE within 92 days after injury (35 [42%] after hospital discharge). In a weighted, multivariate Cox model that included clinical and CAT characteristics available within 24 hours of admission, increased patient age (1.35 [1.19,1.52] per 10 years, p < 0.0001), body mass index ≥30 kg/m2 (4.45 [2.13,9.31], p < 0.0001), any surgery requiring general anesthesia (2.53 [1.53,4.19], p = 0.0003) and first available ttPeak (1.67 [1.29, 2.15], p < 0.00001) were independent predictors of incident symptomatic VTE within 92 days after trauma (C-statistic = 0.799). CONCLUSION The individual’s plasma coagulome (as reflected by thrombin generation) is an independent predictor of VTE after trauma. Clinical characteristics and ttPeak can be used to stratify acute trauma patients into high and low risk for VTE. LEVEL OF EVIDENCE Prognostic, level III.

22 October 2024 in Circulation
Verified
Journal Article
Closed access
GOGustavo S Oderich
Gustavo S Oderich
YHYing Huang
Ying Huang
WHWilliam S. Harmsen
William S. Harmsen
ETEmanuel R. Tenorio
Emanuel R. Tenorio
ASAndres Schanzer
Andres Schanzer
SZSara L. Zettervall
Sara L. Zettervall
BACKGROUND: Fenestrated-branched endovascular aortic repair (FB-EVAR) has been used as a minimally invasive alternative to open surgical repair to treat patients with thoracoabdominal aortic aneurysms (TAAAs). The aim of this study was to evaluate aortic-related mortality (ARM) and aortic aneurysm rupture after FB-EVAR of TAAAs. METHODS: Patients enrolled in 8 prospective, nonrandomized, physician-sponsored investigational device exemption studies between 2005 and 2020 who underwent elective FB-EVAR of asymptomatic intact TAAAs were analyzed. Primary end points were ARM, defined as any early mortality (30 days or in hospital) or late mortality from aortic rupture, dissection, organ or limb malperfusion attributable to aortic disease, complications of reinterventions, or aortic rupture. Secondary end points were early major adverse events, TAAA life-altering events (defined as death, permanent spinal cord injury, permanent dialysis, or stroke), all-cause mortality, and secondary interventions. RESULTS: A total of 1109 patients were analyzed; 589 (53.1%) had extent I–III and 520 (46.9%) had extent IV TAAAs. Median age was 73.4 years (interquartile range, 68.1–78.3 years); 368 (33.2%) were women. Early mortality was 2.7% (n=30); congestive heart failure was associated with early mortality (odds ratio, 3.30 [95% CI, 1.22–8.02]; P =0.01). Incidence of early aortic rupture was 0.4% (n=4). Incidence of early major adverse events and TAAA life-altering events was 20.4% (n=226) and 7.7% (n=85), respectively. There were 30 late ARMs; 5-year cumulative incidence was 3.8% (95% CI, 2.6%–5.4%); older age and extent I–III TAAAs were independently associated with late ARM (each P <0.05). Fourteen late aortic ruptures occurred; 5-year cumulative incidence was 2.7% (95% CI, 1.2%–4.3%); extent I–III TAAAs were associated with late aortic rupture (hazard ratio, 5.85 [95% CI, 1.31–26.2]; P =0.02). Five-year all-cause mortality was 45.7% (95% CI, 41.7%–49.4%). Five-year cumulative incidence of secondary intervention was 40.3% (95% CI, 35.8%–44.5%). CONCLUSIONS: ARM and aortic rupture are uncommon after elective FB-EVAR of asymptomatic intact TAAAs. Half of the ARMs occurred early, and most of the late deaths were not aortic related. Late all-cause mortality rate and the need for secondary interventions were 46% and 40%, respectively, 5 years after FB-EVAR. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifiers: NCT02089607, NCT02050113, NCT02266719, NCT02323581, NCT00583817, NCT01654133, NCT00483249, NCT02043691, and NCT01874197.