Since 1976, all medical devices must be classified and approved by the U.S. Food and Drug Administration (FDA) before they can be legally marketed to the public. It is the fond belief of many Americans that FDA approval is a ringing endorsement of a product's safety, a guarantee that the device has been tested and re-tested by professionals who are presumably trained to put the device through all kinds of rigor to triple-check its safety and effectiveness before allowing it anywhere near our fragile human bodies.

It is, as we say, a fond belief, but not an accurate one. It seems that many medical devices - including the recently recalled DePuy ASR XL Acetabular System - are given an FDA approval and subsequently placed inside people's bodies without ever going through a single test at the FDA.

The FDA department responsible for the regulation and review of medical devices is the Center for Devices and Radiological Health (CDRH). CDRH divides medical devices into three classes: life-sustaining, life-supporting or implantable devices are designated as Class III.

Hip replacements, including the ASR XL system, qualify as a Class III device and are considered high risk to the patient's health.

Class III devices can be approved by the FDA through one of two processes. The first, the premarket approval process (PMA) is more or less what we think of when we imagine our device being tested by the FDA: an extensive review including rigorous clinical trials that show it to be safe for use in humans. The PMA is quite detailed, lengthy, and expensive, usually costing the manufacturing company upwards of $250,000 and taking as much as two years to complete.

A quarter of a million is a large chunk of change - but thankfully for companies with a firm eye on their bottom line, there's always the second FDA approval process.

The premarket notification approval process, also known as the 510(k) approval process, is a sort of shortcut to approval that lowers the price point considerably - and also, conveniently, doesn't involve any pesky testing that might reveal a flaw in the product that would then have to be revamped and retested for another quarter-million.

The 510(k) process basically pre-approves any product that is deemed "substantially equivalent" to another product already legally approved for sale. It costs less than $5,000, requires no clinical trials, and only takes an average of 3-6 months to complete. It's a good shortcut for products that are essentially identical to their forebears. After all, it hardly makes sense to waste the FDA's time when they could be helping get new, innovative, life-saving products to market. But what's to stop companies from claiming their products are "substantially equivalent" to other FDA-approved devices even if they're not remotely the same?

What, indeed

The ASR XL Acetabular hip replacement system got its seal of approval from the FDA using the 510(k) process, claiming that its product was substantially equivalent to other hip replacement devices already on the market. The FDA official in charge of approving the device would have gone through a checklist that included the following questions (simplified; the actual document is fairly dense):

• Does the new product treat or solve the same problems as previously approved products?
• Does the new product have the same design, materials, and energy sources as previously approved products?
• Could any new characteristics or combinations of characteristics affect safety or effectiveness?

The first question is easy: there are dozens of hip replacement devices on the market, all designed to help people walk even after their bum hip throws them out of commission. The second question gives us a bit of a hiccup: the ASR XL Acetabular hip replacement system had two parts, a ball and socket

The ball was equivalent to an approved product; the socket was equivalent to an entirely different approved product; and the materials were equivalent to still a third approved product.

Thankfully for DePuy, the FDA had thoughtfully included a caveat that allowed for such an eventuality. Which brings us to question three: could any new characteristics affect safety?

In DePuy's 501(k) application, it declared that the "subject device does not raise any new issues of safety or effectiveness." A brave statement, though one wonders how the company confirmed such a diagnosis without going through any clinical trials. Perhaps if it had, the problem of metallosis would have become evident before the hips were implanted into approximately 93,000 people.  

With such prerequisites for FDA approval, it is perhaps not surprising that only 10% of 510(k) applications are referred back for supporting human, clinical or animal testing.

Which is to say that 90% of claims require no clinical testing whatsoever before being sold to patients.

That's not to say that there is no testing at all. If a mechanical test is available, 510(k) approval process requires that the new product show that it yields similar results to a previously approved product. The DuPuy ASR hip replacement system took a few spins on a robot that went through simulated movements. In a legal suit brought against DePuy for a different hip implant last year, the mechanical test did not include the full range of motion of which a human being is capable.

If we assume that a new product truly is "substantially equivalent" to previous products and if the new product has the same safety and effectiveness-regardless of any new materials or technologies-then the 510(k) system might work just fine. But that's presuming the device's manufacturer put the product through enough testing to determine independently that the device could withstand the same circumstances as other products.

What's more, products are allowed to be "substantially equivalent" to products that are themselves "substantially equivalent" to other products, which are in turn "substantially equivalent" to still other products. It's a dangerous game of telephone, particularly when you recall that it is permissible to combine two or more products and claim your new device is "substantially equivalent" to all of them.

The DePuy ASR system was approved because it claimed to be substantially equivalent to threeproducts which themselves claimed to be substantially equivalent to other products. None of the products in the chain was ever subjected to clinical trials before being brought to market.

Many people note that the 501(k) process amounts to an enormous loophole, through which many companies bring under-tested products to market. With better testing, the DePuy hip recall might have been avoided altogether.

The DePuy ASR XL Acetabular system was approved through this FDA loophole, despite the fact that it had a materially different design than its predecessors. But it was designed to treat the same problem, so it must have been all right. It was made of a metal ball and a metal socket - but other hip implants had successful metal-on-metal designs, so it must have been all right. And it was designed to allow the bone to grow directly onto the implant - but still other hip implants had used this tactic successfully, so it must have been all right.

Except that, when put all together, it just wasn't.

On August 4, 2010, the FDA issued two reports that recommend sweeping changes to the 510(k) process and procedures. We can only hope they include actually testing the products to which the FDA affixes its seal of approval. 

Recommended Readings:

Hip Implant Risks May Include Premature Failure

How the ASR XL Acetabular System is Constructed

DePuy Asks Hip Implant Patients to Sign Away Their Legal Rights