Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

An Analysis of the FDA MAUDE Database and the Search for Cobalt Toxicity in Class 3 Johnson & Johnson/DePuy Metal-on-Metal Hip Implants

Kavanagh, Kevin T., MD, MS*; Kraman, Steve S., MD; Kavanagh, Sean P., BS

doi: 10.1097/PTS.0000000000000534
Original Articles

Introduction This study was designed to determine whether systemic cobalt toxicity as an adverse event could be documented using the Food and Drug Administration's (FDA) Manufacturer and User Facility Device Experience (MAUDE) database for cobalt-chromium containing hip implant recipients. Class 3 Johnson & Johnson (J&J)/DePuy devices were chosen for analysis because of the large number of adverse event reports related to their Pinnacle and ASR XL Acetabular hip replacement systems. A secondary goal was to characterize the reporters who are populating the information in the MAUDE database and to evaluate the quality of the data and information submitted.

Methods Using FDA MAUDE downloadable data files, 83,528 adverse event medical device narrative reports were identified with the product code of KWA (Prosthesis, Hip, Semiconstrained [Metal Uncemented Acetabular Component]) for J&J/DePuy (KWA Text File). These are class 3 devices and devices with known high failure rates. The ASR XL Acetabular hip replacement Systems and Pinnacle fall into this category. This group of implants was chosen because implant failure is associated with elevated cobalt levels. Two additional subfiles were created from Free Text records in the narrative reports containing key words that indicated a chromium or cobalt (CrCo) device and another for key words indicating elevated levels of cobalt or toxicity. These files were then searched for symptoms of systemic cobalt toxicity with Microsoft Excel using key words pertaining to symptom categories of: cognitive/memory loss, tremor, neuropathy, depression, auditory symptoms, visual symptoms, cardiac, and thyroid disease. Reports are submitted to the FDA at the device component level. It was common for multiple reports to be submitted for a single device.

Results It was not possible to differentiate systemic versus local symptoms for adverse event reports in the neuropathy category. This category was not analyzed. The number of adverse event reports in the other categories ranged from 119 to 16 for the J&J/DePuy KWA Text File and 59 to 4 in the file of records having key words indicating potential elevation of CrCo or toxicity. Cardiac, visual, and auditory conditions were most frequently reported. With the possible exception of cardiomyopathy, the numbers of these reports are far below the prevalence expected in the general population of comparable age. The content of the MAUDE database records often contained little objective data. We found less than 4% of 14,714 records, which indicated cobalt elevation or toxicity contained units for quantitative measurement. There was also frequent use of nondescript, all-encompassing words and phrases such as “mental anguish,” found in 321 KWA Text File records. Manufacturers submitted more than 99% of the reports and the most common reporter occupation was attorney, found in 42.24% of the 83,550 J&J/DePuy KWA Reporter File records. Physician was the reporter's occupation in 20.48% of reports but seldom reported directly to the FDA.

Conclusions We were not able find in the FDA MAUDE database meaningful warning signs to support the contention that chromium-cobalt–containing Class 3 J&J and DePuy hip implants caused systemic neurological or thyroid symptoms in patients. The incidence of reported cardiomyopathy was rare but frequent enough to be cause of concern. The redaction of most patient data along with the nonstructured nature of data entry would be expected to hinder the identification of warning signs. Even identification of the type of device could not be consistently carried out. In addition, the FDA needs to implement a methodology to identify and group all reports from a single device implanted into a patient, so duplication of event counting would not occur. Of 83,550 J&J/DePuy KWA Reporter File records, we found only two physician reports sent directly to the FDA. Almost all reports are submitted by manufacturers and are most commonly authored by attorneys. A standard of care needs to be set for physicians to report medical device adverse events to the FDA.

From the *Health Watch USA, Somerset;

University of Kentucky College of Medicine, Health Watch USA, Lexington, Kentucky; and

Health Watch USA, Solon, Ohio.

Correspondence: Kevin T. Kavanagh, MD, MS, Health Watch USA, Somerset, KY (e-mail:

The first author, K.T.K., has had complications from bone morphogenic protein implanted with a medical device into his neck, he is Board Chairman of Health Watch USA, a patient advocacy organization whose members have been active in raising awareness of the risks of cobalt-chromium implants. He has also received partial conference attendance and meeting support from the U.S. Department of Health and Human Services, National Quality Forum, National Patient Safety Foundation (NPSF), The Leapfrog Group and Consumer Union. He has served on the Centers for Medicare and Medicaid Services' Technical Expert Panel for Hospital Acquired Conditions, and most recently on the Strategic Working Group for AHRQ for quality indicators and is an Associate Editor for the Journal of Patient Safety for which he receives an honorarium. He also has a first-degree relative who is involved with development of pharmaceuticals at a university. None of the authors have an affiliation with or conflicts of interest with Johnson & Johnson. The other authors disclose no conflict of interest.

Supplemental digital contents are available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved