Issue: June 2011
June 01, 2011
3 min read

ACL registry improves safety and quality through continuous surveillance of implant performance

Issue: June 2011
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

It is my pleasure to feature Tadashi T. Funahashi, MD, for this month’s 4 Questions interview. I have closely watched the good work he and his colleagues have done in the Kaiser system with developing their registries. In the past, we have featured results from their joint replacement registry. This time, I have asked Dr. Funahashi to specifically share their ACL data with us. What makes their registries so valuable is they are applicable to what most clinicians experience. Their registries have many surgeons in different settings and are using common variations of the surgeries performed around the United States. This ACL registry and their other registries have contributed to our orthopedic profession and will continue to do so in the future.

— Douglas W. Jackson, MD
Chief Medical Editor

Douglas W. Jackson, MD: How does your ACL registry function and what kind of information does it offer the clinician?

Tadashi T. Funahashi, MD: The goal of the Kaiser Permanente Registries is to improve safety and quality through continuous surveillance of implant performance, comparative effectiveness, and adverse outcomes or recalls. Data is collected using our integrated electronic health records, administrative data, paper forms at the time of surgery and patient outcome surveys. The intraoperative form collects information on the patient’s injury, such as time to surgery and injury event, associated injuries, surgical approach, concomitant procedures (such as treatment of meniscal or cartilage injuries), graft selection, and method and type of fixation.

Tadashi T. Funahashi, MD
Tadashi T. Funahashi

Validation of the procedure and supplemental information, such as body mass index and race, are collected from our electronic medical record. Our fully integrated care model then allows us to longitudinally track the patient for surgical complications, such as thromboembolic events, surgical site infections, any reoperations and revision surgery. For a subset of our patients, we also collect preoperative and postoperative self-reported outcomes.

Individual surgeons may request a confidential report of their surgical profile including patient demographics, case volumes, breakdown of their surgical interventions and recorded outcomes. More significantly, the compilation of the data by more than 200 contributing surgeons allows us to collect and analyze data on approximately 2,500 new cases per year, with more than 11,900 cases registered at the end of 2010.

Our annual reports provide summary information on the patient population, surgical techniques, including meniscal injury and treatment, graft types, fixation methods, readmissions, reoperations, complications and revisions. Some recent findings include a meniscal repair rate 28%, which is higher than what is generally reported in the literature, and graft type distribution of 30% patellar tendon, 29% hamstrings and 39% allograft. Now in our sixth year since inception, we are beginning to provide additional important analyses, such as risk factors associated with early surgical complications, reoperations and failures, as well as differential survival rates of various graft types.

Jackson: In your registry, what are the reasons patients need reoperations of ACL reconstructed knees?

Funahashi: Our 2009 annual report evaluated 7,089 primary ACL reconstructions. A total of 332 (4.7%) cases had undergone another operation on the index or contralateral knee. Ipsilateral reoperations were commonly meniscal (39.6%), for hardware removal (14.8%) and procedures for stiffness (13.6%). There were 84 (1.2%) graft revisions with the main reason cited as graft retear in 71 cases and graft failure without specific injury in four cases. For the contralateral knee, most underwent a new ligament procedure (73.2%).

Jackson: What have you observed in the overall rates of deep vein thrombosis (DVT) and pulmonary embolism (PE) in this population? What prophylactic measures do you currently use?

Funahashi: While we do not have explicit guidelines and do not track the use of prophylaxis for DVT and PE, most surgeons are not using chemoprophylaxis. Nevertheless, we track the incidence of symptomatic thromboembolic events as defined by evaluation and treatment within our system. In our 2009 annual report, we found our incidence of DVT to be 0.2% and PE to be <0.1% for primary ACL reconstructions.

Jackson: What are the strengths and weaknesses of any ACL registry?

Funahashi: The strengths of registry research include the ability to collect in situ information on a large number of patients, determine the demographic characterization of the index population, describe overall patterns of utilization (such as case rate, implant types and grafts) within this population, describe graft/implant survival, determine the incidence of rare events including complications, and locate implants in case of recalls.

Our registry is a reflection of ACL interventions within the community and has a wealth of information regarding various types of procedures performed by a large number of surgeons in different locations. This allows us to evaluate the effectiveness of a specific graft, implant or technique. The volume of registered cases also allows us to track rare events and study the associated risk factors, which cannot be done by other study designs.

Some limitations of registries include the need for continuous collection of large volumes of data that precludes collection of detailed information. Adding more data points to a registry can create confusion and additional work for the user, especially if the definitions of terms are not clearly described/accepted, which can lead to decreased participation in the registry. The observational nature of a registry cannot determine causation between exposures and outcomes, and are limited to the assessment of associations. Finally, the large number of locations, surgeons and cases entered in the registry creates the need for rigorous assessment of selection bias, as well as who is lost to follow-up and why, in our population.