By Robert Langreth

Bloomberg News

NEW YORK — The safety of robots made by Intuitive Surgical Inc. is being probed by U.S. regulators, raising questions about the prospects of one of the hottest technologies in health care.



The Food and Drug Administration asked surgeons at key hospitals to list the complications they may have seen with the machines, which cost about $1.5 million each and were used last year in almost 500,000 procedures. The doctors were also surveyed on which surgeries the robots might be most and least suited for, and asked to discuss their training, according to copies of the survey obtained by Bloomberg News.



The answers may sway debate on whether robotic surgeries promoted as being less invasive are worth the extra cost. The findings may also determine the outlook for Intuitive. The da Vinci surgical system and related products generated most of the Sunnyvale, California-based company’s $2.2 billion revenue in 2012, and helped boost market value 70 percent over three years to about $23 billion, according to data compiled by Bloomberg.



The surveys were sent to hospitals that belong to a product safety network overseen by the FDA. What the agency is trying to determine is whether a rise seen in incident reports sent to the agency are “a true reflection of problems” with the robots, or the result of other issues, said Synim Rivers, an agency spokeswoman, in an e-mail. “It is difficult to know why the reports have increased,” she said.



Incident reports are sent to the agency by patients, medical professionals and companies. They “can contain incomplete, inaccurate, duplicative and unverified information,” Rivers said.



Intuitive fell 11 percent Thursday to close at $509.89. The stock recovered Friday, rising 8.5 percent to close at $553.40 in New York when at least four analysts reiterated their recommendations of the shares. Jeremy Feffer, a Cantor Fitzgerald analyst, raised his rating to buy from hold, saying “we see limited risk of this investigation materializing into significant punitive measures against the company.”



The adverse event reports sent to the FDA outline serious complications, including those involving gynecological surgeries that make up about half of the procedures involving Intuitive’s robotic system. They include damage to ureters and bowels as well as instances when instruments broke off from the robot and fell into patients. Reports also cited burns to various organs or vessels, most repaired during the surgery itself.



Many of the reports were filed by Intuitive Surgical and state that no robot malfunction was found or that the problem came from user error. Several originated from patients, who aren’t identified.



The robots are “extremely safe,” said Myriam Curet, Intuitive’s chief medical adviser, in a telephone interview. “There is a lot of redundancy in the system to make it as safe as possible.” The percentage of deaths and injuries involving the company’s da Vinci robotic system, “is extraordinarily small,” she added, and “hasn’t grown” over time.



While the number of adverse-event reports to the FDA is small compared with the total number of procedures, the agency’s reporting system has in the past served as an early warning system on concerns about medical-device safety.



“One of the risks is that the FDA could recall parts of the product or certain instruments, or somehow put restrictions on certain surgeries,” said Michael Matson, an analyst with Mizuho Securities USA in New York. “But I don’t think that’s very likely,”



A rise in adverse events isn’t necessarily alarming because the number of surgeries done with the robot has been growing rapidly, said Matson, who is based in New York and has a buy recommendation on Intuitive.



“The bigger concern would be that the patients would get scared,” he said. “Part of what’s driven this market is people seeking out robotic surgery; hospitals market it and the patients seem to think it’s better.”



Intuitive’s stock is likely to be under pressure until the company can prove the safety worries aren’t a significant issue, Matson said. In the meantime, “there’s going to be an overhang,” he said.



In robotic surgery, the physician sits at a video-game style console several feet from the patient, peering into a high-definition display while using foot pedals and hand controls to maneuver mechanical arms equipped with surgical tools. A camera provides a three dimensional view of the work being done inside the patient.



Martin Makary, a surgeon at Johns Hopkins Hospital in Baltimore who has researched the marketing of the systems, said safety should now become part of the debate.



“Patients need to know the truth about robotic surgery,” Makary said in a telephone interview. “The Achilles-heel feature” of the robot is its lack of tactile feedback, which can spur “inadvertent injuries if added caution is not taken.”



Among surgeons, there has been an intensifying debate about whether the benefits of robotic surgery justify the increased costs. A report published in the Journal of the American Medical Association on Feb. 20 found that surgery to remove the uterus with the Intuitive robot costs thousands of dollars more without reducing complications, compared with standard less-invasive surgery.



Examining data from 264,758 women who had a hysterectomy for benign conditions at 441 U.S hospitals, the study found that robotic operations cost hospitals $2,189 more per procedure than performing the same surgery without the robot.



A version of the letter sent to doctors by the FDA in January said the survey was being done because “a recent review of medical device report data is showing an increase in patient adverse reports.” A revision of the document requested by Intuitive and sent again on Jan. 22 deletes that wording, according to the company.



The first version was a draft that was sent out inadvertently, the FDA’s Rivers said. The agency only realized this after the company contacted the regulators, she said.



“We do not ask companies to weigh in” on this type of survey, she said.



The first version of the letter “did not accurately characterize what assumptions can or should be made from the number of reports in the FDA database,” Rivers said.