The U.S. Food and Drug Administration will allow medical devices for the legs coated with the drug paclitaxel to remain on the market, but with increased scrutiny and enhanced patient warnings after the discovery that the devices are associated with long-term higher risks of death.

An updated letter to doctors posted by the FDA on Wednesday says the agency is "working with" devicemakers to include warnings on product labels explaining what's known about the risk. While the agency will continue allowing commercial sales of the devices, patients treated in clinical trials will receive enhanced informed consent disclosures and will be subject to closer ongoing scrutiny by safety-monitoring boards.

"Because of the demonstrated short-term benefits of the devices, the limitations of the available data, and uncertainty regarding the long-term benefit-risk profile of paclitaxel-coated devices, the FDA believes clinical studies of these devices may continue and should collect long-term safety (including mortality) and effectiveness data," the FDA said in a letter Wednesday.

Stock analysts, company officials and researchers said the FDA's announcement was not surprising. Analysts with Needham & Co. said the announcement was likely to increase use and sales of the devices somewhat, but not to the levels that had been projected before the mortality risk was first reported.

Paclitaxel is applied to the outsides of drug-eluting stents and drug-coated medical balloons to cut down on the risk that a diseased femoropopliteal artery in the leg will become reclogged months after being mechanically opened with a device. In patients with severe peripheral artery disease (PAD), the devices are seen as an alternative to surgically attaching a bypass vessel around the blockage.

The FDA created a lucrative and fast-growing product category in 2012 when it approved the nation's first paclitaxel-coated medical device for PAD. Sales of such devices eclipsed $500 million industrywide last year, including Cook Medical's Zilver PTX stent and Medtronic's In.Pact Admiral drug-coated balloon. Boston Scientific's Eluvia stent was seen as a highly promising device approved late last year, and Abbott Laboratories is working with Minnesota's SurModics to bring its own drug-coated balloon to market.

Needham analysts said that although the news was positive for Medtronic and Boston Scientific, paclitaxel device sales account for less than 1% of total revenue at each firm. For SurModics, a smaller company in Eden Prairie, the letter was a "significant positive" that removed uncertainties over its drug-coated balloon device pipeline, which could be worth $92 million in milestone payments from Abbott alone.

Paclitaxel has long been used on coronary stents and in chemotherapy, without the late mortality concerns. The burgeoning field of using paclitaxel to treat PAD had been on a strong upswing until the December issue of the Journal of the American Heart Association brought the growth to a halt.

A report in that issue, which has since been downloaded more than 40,000 times, said an independent meta-analysis of two dozen industry-sponsored trials found that patients who got paclitaxel for PAD had a higher risk of death at five years than patients randomized to the same treatment without the drug.

None of the individual studies was large enough to detect higher death rates, and they were not designed with that purpose in mind. But by pooling data from multiple trials, independent researchers revealed a larger pattern that showed an unexplained higher risk of death in the treatment arms of the studies.

The FDA quickly undertook its own analysis, amid assurances from the industry that none of the individual studies showed the "late mortality signal" claimed in the paper. The FDA also found an enhanced risk of death at five years among a handful of randomized controlled trials.

Specifically, the FDA analysis found that among 1,090 patients randomized in three trials, the all-cause five-year mortality rate was 19.8% in the paclitaxel patients and 12.7% in patients treated with uncoated devices. That means the paclitaxel patients died within five years about 57% more often than people in the control group. (The original JAHA paper said the risk of death was 93% higher for paclitaxel groups, while a subsequent analysis by a physicians' group called VIVA said the increased mortality rate was only 38% greater.)

"The magnitude of the signal should be interpreted with caution because of multiple limitations in the available data, including wide confidence intervals due to a small sample size, pooling of studies of different paclitaxel-coated devices that were not intended to be combined, substantial amounts of missing study data, no clear evidence of a paclitaxel dose effect on mortality, and no identified pathophysiologic mechanism for the late deaths," the FDA wrote.

An FDA panel of experts met for two days in Maryland in June and found that although a mortality signal was present, there was not enough evidence to conclude that people with higher doses of the drug were at greater risk of death. Also, the panel was not able to pinpoint any specific cause for the effect.

In response to the panel's findings, the FDA on Wednesday issued an updated letter to doctors recommending that patients' long-term health preferences and their individual risks for vessel reclosure should be taken into account when deciding whether to use paclitaxel to treat PAD.

"The Panel determined, and the FDA concurs, that additional clinical study data are needed to fully evaluate the late mortality signal," the agency said.

Mark Pacyna, general manager of Medtronic's peripheral vascular business, said Wednesday that the company was encouraged that the FDA wants to maintain the device's status as a first-line treatment option, especially for patients at risk of vessel reclosure (known as "restenosis").

"Overall it's a positive step in the next phase of this process," Pacyna said. "It certainly doesn't end anything. But the guidance allows physicians and patients to really get into the benefit and risk of these treatment options and supports that treatment decision, and that was the recommendation from the FDA panel."