Orthopaedic surgeons at ten sites in the United States will participate in a registry to collect postoperative venous thromboembolism incidence in lower extremity joint arthroplasty patients using the mobile compression device. Patients will be included in the registry if they are eighteen years of age or older and undergoing primary unilateral hip arthroplasty (including hip resurfacing) or primary unilateral knee arthroplasty (including unicondylar knee arthroplasty) using only the mobile compression device with or without aspirin for venous thromboembolism prevention. Patients will be excluded if they are scheduled for a revision surgery, have a history of venous thromboembolism, a coagulation disorder, a solid tumor malignancy within the last five years, or had a major surgery in the three months prior to the joint surgery. An attempt to enroll every patient consecutively will be made at each institution and only those patients who did not meet the inclusion criteria or declined participation will not be included. Use of the ActiveCare+S.F.T® mobile compression device will begin intra-operatively and continue for a minimum of 10 days. The decision to use aspirin is at the discretion of each surgeon. All patients will be evaluated within three months postoperatively, documenting evidence of deep venous thrombosis or pulmonary emboli.
Registry Procedures
- Patients will complete a 1 page questionnaire regarding their usage of the mechanical compression device. The de-identified information from the questionnaire will be entered into a web based computer database by a research coordinator at each of the ten sites. The web based registry was created by a third party (Global Visions Technology, Inc.).
- Once the study is closed to enrollment and data entry is complete, the database manager will pull all the data from the web based registry and import into a statistical software program for analysis.
- A power analysis found that sample sizes of 1500 in the device group and any drug group would be sufficient to achieve power in excess of 90% to detect a non-inferiority margin difference between the venous thromboembolism proportions in the two groups of 1.0%. In these calculations, we considered drug group venous thromboembolism rates between 0.5% and 1.0% and the device group venous thromboembolism rate was taken to be the drug group rate +1.0% under the null hypothesis of inferiority. Power was computed for the case when the actual device venous thromboembolism rate was identical to the drug comparator rate. The test statistic used was the one-sided score test, with the significance level set at 0.025.
- SPSS version 13.0 (SPSS, Chicago, Illinois) and NCSS Version 7.1.21 (NCSS LLC, Kaysville, Utah) will be used for analysis of the registry data. Means will be calculated to describe continuous variables (age, height, weight) and frequencies will be calculated to describe categorical variables (surgery type, aspirin use, incidence of symptomatic venous thromboembolism). Upper bound 97.5% CIs will be calculated around the observed venous thromboembolism rate difference between the mobile compression device and each drug comparator.
Quality Factors
- Institutional Review Board (IRB)approval will be obtained at each site.
- Patient privacy will be protected per the Health Insurance Portability and Accountability Act (HIPAA). Patients will remain anonymous; all patient identifiers will be purged from the data at the originating center. Patient care will NOT be affected in any way by participation in the study.
- Accounts were set up for each coordinator with respective passwords for security. A data entry guide will be distributed to each coordinator, which will give instructions for entering each variable. Additionally, each coordinator will be required to view an instructional video online before they will be permitted to begin data entry into the registry.
- Validation criteria will be set up for each field so that data are entered in the correct format (eg, must enter four digits for the year, sex must be entered as "M" or "F", etc). Ranges will be constructed to minimize data entry errors (eg, "number of days the device was worn" will be restricted to a number between 0 and 15 only). If any required fields are left blank, a prompt will come up to ask the coordinator to fill in the missing data. The record will be marked as incomplete until all required fields are completed. Each of the 10 sites will only have access to view their own data.
- Periodic data checks will be performed by the database manager, who will query all fields and to keep track of how enrollment is going at each site. When the study is closed, the database will be locked by the database manager.