Drug and Device Injuries

People injured by a drug or medical device often sue not only their medical caregivers and the hospitals where the devices were implanted, but also the drug or device's manufacturers. Smart move, but there may be other avenues for recovery that should be explored. There are many players in the process that brings a drug or device to the market, and it may prove valuable to question whether any of these had a role in causing the claimant's injury and whether they can be reached for recovery.

One such contributor to the process that leads to the marketing of a medical device ' a medical device testing service hired by the manufacturer ' is now fighting a suit against it for alleged fraud and negligence that may have led to the injuries suffered by a woman who underwent surgery for a back problem. Wawrzyneck v. Statprobe Inc., Slip Copy, 2007 WL 3146792 (E.D.Pa., 10/25/07). Although the testing service moved for summary judgment of the suit based on a number of theories, the case has been allowed to go forward. The reasoning of the U.S. District Court for the Eastern District of Pennsylvania in denying summary judgment in this case is instructive on the issues of a plaintiff's duty to timely discover which parties may be liable and whether federal preemption should apply.

A Surgery's Bad Outcome

Eileen Wawrzyneck underwent spinal surgery at Elkins Park Hospital in February 1999. Prior to her operation, her physician, Dr. Leonard Bruno, consulted with her and told her that he intended to apply ADCON-L ' a gel, classified as a medical device, that is used to cover the surgical site ' to prevent scar tissue from forming in her spinal canal. Wawrzyneck consented to this application, and the surgery was performed as planned. A problem developed that required a second surgery, following which the plaintiff suffered a severe decomposition of her spinal column. This complication necessitated two additional surgeries to her back.

In December 2000, Wawrzyneck and her husband brought suit for medical malpractice and for negligent administration of antibiotic treatment against Dr. Bruno, another doctor, and the hospital. They were unsuccessful in these claims. They settled with ADCON-L's manufacturer, Gliatech.

There remained one source of potential recovery untapped: the company that compiled the safety and efficacy test data on ADCON-L. On March 1, 2005, the Wawrzynecks sued that company, Statprobe. The court dismissed plaintiffs' breach of contract claim, and defendant moved for summary judgment on their claims of fraud and negligence.

Testing ADCON-L

The plaintiffs' claim alleges several irregularities in the manner in which ADCON-L was tested and in the way the test results were reported. This, they say, caused medical providers and consumers to believe the device was more useful than it actually is.

ADCON-L was classified as a Class III medical device under the Medical Device Amendments (MDA) to the Food, Drug, and Cosmetics Act. A Class III device is one that 'presents an unreasonable risk of injury or illness.' 21 U.S.C. ' 360. Before a Class III medical device can be marketed to the public, the manufacturer must provide the FDA with 'reasonable assurance' that the device is safe and effective. 21 U.S.C. ' 360c(a)(1)(C). This the manufacturer does through the FDA's pre-market approval (PMA) process, which requires manufacturers to 'submit detailed information regarding the safety and efficacy of their devices, which the FDA then reviews, spending an average of 1,200 hours on each submission.' Steele v. Depuy Orthopaedics Inc., 295 F.Supp.2d 439, 442 (K.N.J.2003). To show such a device is 'effective,' the manufacturer must prove through valid scientific evidence 'that in a significant portion of the population, the use of the device for its intended conditions of use, when accompanied by adequate directions for use in warning against unsafe use will provide clinically significant results.' 21 U.S.C. ' 360e(d)(2)(A).

Gliatech had to get FDA approval before it could market ADCON-L, so it contracted with Statprobe to have it provide a statistical report of the device's safety and efficacy in preventing the formation of scar tissue. Statprobe, a bio-statistical firm and contract research organization, advertised its services as the 'highest quality clinical management, data management, programming, biostatistics, medical writing and medical safety services to the pharmaceutical, biotechnology and medical device industries worldwide.'

In order to conduct the required testing, Statprobe was tasked by contract with programming the study, monitoring it, managing its data and providing a statistical analysis of the results. Statprobe also developed and maintained the study's randomization code and was charged with alerting Gliatech to any safety problems it observed and with reporting any adverse medical events or side effects, in accordance with federal regulations.

The ADCON-L study was to be a double-blinded evaluation of the development of scar tissue in patients six months after lower back surgery. Approximately half of the study's subjects received ADCON-L, while the other half received no special treatment at all. Statprobe developed a randomization code that assigned each patient to one group or the other. The company was supposed to maintain the key to the code until the study was completed, with any changes to protocol being subject to prior approval by the FDA.

Dr. Jeffrey Ross, a neuroradiologist, was the only doctor employed to read the Magnetic Resonance Imaging (MRI) results taken on the study's participants to see how well they had healed. He assigned a score to each patient, in pen, on that patient's 'scar score sheet.' These scar score sheets were sent to Statprobe for entrance into its data bank and statistical analysis.

As part of the PMA process, Gliatech was required to submit interim data from the study to the FDA in mid-1997 and make a presentation of the final data to an FDA Advisory Panel on Dec. 12, 1997. The interim data taken in the summer of 1997 showed preliminarily that ADCON-L was effective. However, a month prior to the December 1997 final data submission date, Statprobe reported to Gliatech that the data showed roughly the same scar formation rate for those patients using ADCON-L as for those not using it.

The Statprobe statistician in charge of the ADCON-L project, Dr. Guoquin Su, agreed that the ADCON-L test results showed no efficacy at reducing scar build-up, but neither he nor anyone else from Gliatech or Statprobe mentioned this at the December 1997 FDA Advisory Panel meeting. Statprobe contends that this was because the data was preliminary and likely to change. And change it did, but in a manner that posed a problem.

The FDA Advisory Panel suggested that Gliatech and Statprobe do an 'Intraobserver Reliability Study,' which would test the consistency of the scar score readings Dr. Ross had made. The method of testing was this: Dr. Ross reread the original study's MRIs in the presence of two observers, one a doctor and the other a Gliatech employee. Both these observers were aware during this process of which of the patients had used ADCON-L and which had not, although Dr. Ross still did not know. As Dr. Ross read the results out loud, Dr. McKinley wrote them down, in pencil. When the process was completed, Dr. Ross signed off on the results as written down. However, FDA regulations require test results to be written in pen, and a later FDA investigation of this case found numerous erasures on the test result sheets.

The new results produced in this manner showed a great increase in the efficacy of ADCON-L in reducing the development of scar tissue and only a 1% probability that the difference between the ADCON-L and control group patients was a matter of chance. (To demonstrate the efficacy of ADCON-L, the manufacturer had to show that there was less than a 5% chance that any favorable results from the use of ADCON-L were due to chance only, and not to use of the device.) This new data was entered into Statprobe's database, without any indication that these were re-read scores that were not compiled in 1997, before the December 1997 FDA Advisory Panel met. Despite the fact that the re-read data was recorded in pencil, was represented as having been read in 1997 instead of 1998, was read after the study was unblinded, and was not read in compliance with the study's guidelines, Statprobe used it to provide a final clinical study report that did not mention the fact that it was based on substituted data. On May 27, 1998, the FDA gave conditional approval to Gliatech to manufacture ADCON-L.

When the FDA later learned of the testing and reporting irregularities, Gliatech was prosecuted by the Department of Justice. The company pleaded guilty to charges that it failed to submit adverse event reports, failed to maintain accurate files and submitted false and misleading reports to the FDA. Neither the FDA nor the Justice Department took any legal action against Stratprobe.

Statprobe Argues Suit Is Time-Barred

Statprobe asked for summary judgment on the basis of Pennsylvania's statute of limitations, 42 Pa..C.S. ” 5524-5525, which requires personal injury claimants to commence action within two years of the date of injury. Plaintiffs here argued that both the 'discovery rule' and the doctrine of fraudulent concealment tolled the statute of limitations.

Pennsylvania's discovery rule states that the statute of limitations begins to run on the first date that the injured person possesses sufficient critical facts to put him on notice that a wrong has been committed and that he needs to investigate to determine whether he is entitled to recovery. A plaintiff invoking the discovery rule bears the burden of proving he was unable to discover sufficient facts to assert a claim within the limitations period despite exercising reasonable diligence. Dalrymple v. Brown, 549 Pa. 217 (Pa. 1997).

Statprobe offered evidence that the plaintiffs could have discovered their involvement in the testing of ADCON-L much earlier, and within the two-year limitations period, if they had investigated Gliatech's Securities Exchange Commission (SEC) filings and had read an article recounting Statprobes involvement in ADCON-L's testing in a leading Ohio newspaper, the Plain Dealer. The court rejected this stance, stating, 'Statprobe's arguments place a weighty responsibility on Plaintiffs to scour the nation's newspapers for information related to their suit, a responsibility not reflected in case law. Plaintiffs are residents of Pennsylvania, and while it is true that the Plain Dealer is the self-proclaimed 'largest newspaper' in Ohio and is available online, Plaintiffs are not required as part of the concept of reasonable diligence to troll the local media of the nation's cities for possible defendants with potential responsibility for their injuries. Moreover, even if Plaintiffs had found the Plain Dealer article and Gliatech's Form 8-K (SEC filing) by 2000, neither of these documents identifies Statprobe by name. Thus, Plaintiffs still would have needed to locate some other source with more information of greater specificity.'

In addition, the court found sufficient evidence that Statprobe had fraudulently concealed its role in conducting and reporting the test results to preclude summary judgment on a statute of limitations basis. Specifically, in a July 1997 letter from Mark Becker, Vice President of Statprobe, to Raymond Silkaitis, Ph.D., Vice President of Gliatech, Dr. Becker expressed concerns that Gliatech was asking Statprobe to take part in bad statistical methods. He therefore asked in that same letter that Gliatech not use Statprobe's name in any way to indicate support of ADCON-L before any regulatory agency. From this evidence, said the court, 'a jury could conclude here that Statprobe took steps to conceal its identity by preventing publication of its name in connection with Gliatech or ADCON-L.'

In next month's issue, we'll discuss other bases for the U.S. District Court for the Eastern District of Pennsylvania's denial of summary judgment.

Janice G. Inman is Editor-in-Chief of this newsletter.

People injured by a drug or medical device often sue not only their medical caregivers and the hospitals where the devices were implanted, but also the drug or device's manufacturers. Smart move, but there may be other avenues for recovery that should be explored. There are many players in the process that brings a drug or device to the market, and it may prove valuable to question whether any of these had a role in causing the claimant's injury and whether they can be reached for recovery.

One such contributor to the process that leads to the marketing of a medical device ' a medical device testing service hired by the manufacturer ' is now fighting a suit against it for alleged fraud and negligence that may have led to the injuries suffered by a woman who underwent surgery for a back problem. Wawrzyneck v. Statprobe Inc., Slip Copy, 2007 WL 3146792 (E.D.Pa., 10/25/07). Although the testing service moved for summary judgment of the suit based on a number of theories, the case has been allowed to go forward. The reasoning of the U.S. District Court for the Eastern District of Pennsylvania in denying summary judgment in this case is instructive on the issues of a plaintiff's duty to timely discover which parties may be liable and whether federal preemption should apply.

A Surgery's Bad Outcome

Eileen Wawrzyneck underwent spinal surgery at Elkins Park Hospital in February 1999. Prior to her operation, her physician, Dr. Leonard Bruno, consulted with her and told her that he intended to apply ADCON-L ' a gel, classified as a medical device, that is used to cover the surgical site ' to prevent scar tissue from forming in her spinal canal. Wawrzyneck consented to this application, and the surgery was performed as planned. A problem developed that required a second surgery, following which the plaintiff suffered a severe decomposition of her spinal column. This complication necessitated two additional surgeries to her back.

In December 2000, Wawrzyneck and her husband brought suit for medical malpractice and for negligent administration of antibiotic treatment against Dr. Bruno, another doctor, and the hospital. They were unsuccessful in these claims. They settled with ADCON-L's manufacturer, Gliatech.

There remained one source of potential recovery untapped: the company that compiled the safety and efficacy test data on ADCON-L. On March 1, 2005, the Wawrzynecks sued that company, Statprobe. The court dismissed plaintiffs' breach of contract claim, and defendant moved for summary judgment on their claims of fraud and negligence.

Testing ADCON-L

The plaintiffs' claim alleges several irregularities in the manner in which ADCON-L was tested and in the way the test results were reported. This, they say, caused medical providers and consumers to believe the device was more useful than it actually is.

ADCON-L was classified as a Class III medical device under the Medical Device Amendments (MDA) to the Food, Drug, and Cosmetics Act. A Class III device is one that 'presents an unreasonable risk of injury or illness.' 21 U.S.C. ' 360. Before a Class III medical device can be marketed to the public, the manufacturer must provide the FDA with 'reasonable assurance' that the device is safe and effective. 21 U.S.C. ' 360c(a)(1)(C). This the manufacturer does through the FDA's pre-market approval (PMA) process, which requires manufacturers to 'submit detailed information regarding the safety and efficacy of their devices, which the FDA then reviews, spending an average of 1,200 hours on each submission.' Steele v. Depuy Orthopaedics Inc ., 295 F.Supp.2d 439, 442 (K.N.J.2003). To show such a device is 'effective,' the manufacturer must prove through valid scientific evidence 'that in a significant portion of the population, the use of the device for its intended conditions of use, when accompanied by adequate directions for use in warning against unsafe use will provide clinically significant results.' 21 U.S.C. ' 360e(d)(2)(A).

Gliatech had to get FDA approval before it could market ADCON-L, so it contracted with Statprobe to have it provide a statistical report of the device's safety and efficacy in preventing the formation of scar tissue. Statprobe, a bio-statistical firm and contract research organization, advertised its services as the 'highest quality clinical management, data management, programming, biostatistics, medical writing and medical safety services to the pharmaceutical, biotechnology and medical device industries worldwide.'

In order to conduct the required testing, Statprobe was tasked by contract with programming the study, monitoring it, managing its data and providing a statistical analysis of the results. Statprobe also developed and maintained the study's randomization code and was charged with alerting Gliatech to any safety problems it observed and with reporting any adverse medical events or side effects, in accordance with federal regulations.

The ADCON-L study was to be a double-blinded evaluation of the development of scar tissue in patients six months after lower back surgery. Approximately half of the study's subjects received ADCON-L, while the other half received no special treatment at all. Statprobe developed a randomization code that assigned each patient to one group or the other. The company was supposed to maintain the key to the code until the study was completed, with any changes to protocol being subject to prior approval by the FDA.

Dr. Jeffrey Ross, a neuroradiologist, was the only doctor employed to read the Magnetic Resonance Imaging (MRI) results taken on the study's participants to see how well they had healed. He assigned a score to each patient, in pen, on that patient's 'scar score sheet.' These scar score sheets were sent to Statprobe for entrance into its data bank and statistical analysis.

As part of the PMA process, Gliatech was required to submit interim data from the study to the FDA in mid-1997 and make a presentation of the final data to an FDA Advisory Panel on Dec. 12, 1997. The interim data taken in the summer of 1997 showed preliminarily that ADCON-L was effective. However, a month prior to the December 1997 final data submission date, Statprobe reported to Gliatech that the data showed roughly the same scar formation rate for those patients using ADCON-L as for those not using it.

The Statprobe statistician in charge of the ADCON-L project, Dr. Guoquin Su, agreed that the ADCON-L test results showed no efficacy at reducing scar build-up, but neither he nor anyone else from Gliatech or Statprobe mentioned this at the December 1997 FDA Advisory Panel meeting. Statprobe contends that this was because the data was preliminary and likely to change. And change it did, but in a manner that posed a problem.

The FDA Advisory Panel suggested that Gliatech and Statprobe do an 'Intraobserver Reliability Study,' which would test the consistency of the scar score readings Dr. Ross had made. The method of testing was this: Dr. Ross reread the original study's MRIs in the presence of two observers, one a doctor and the other a Gliatech employee. Both these observers were aware during this process of which of the patients had used ADCON-L and which had not, although Dr. Ross still did not know. As Dr. Ross read the results out loud, Dr. McKinley wrote them down, in pencil. When the process was completed, Dr. Ross signed off on the results as written down. However, FDA regulations require test results to be written in pen, and a later FDA investigation of this case found numerous erasures on the test result sheets.

The new results produced in this manner showed a great increase in the efficacy of ADCON-L in reducing the development of scar tissue and only a 1% probability that the difference between the ADCON-L and control group patients was a matter of chance. (To demonstrate the efficacy of ADCON-L, the manufacturer had to show that there was less than a 5% chance that any favorable results from the use of ADCON-L were due to chance only, and not to use of the device.) This new data was entered into Statprobe's database, without any indication that these were re-read scores that were not compiled in 1997, before the December 1997 FDA Advisory Panel met. Despite the fact that the re-read data was recorded in pencil, was represented as having been read in 1997 instead of 1998, was read after the study was unblinded, and was not read in compliance with the study's guidelines, Statprobe used it to provide a final clinical study report that did not mention the fact that it was based on substituted data. On May 27, 1998, the FDA gave conditional approval to Gliatech to manufacture ADCON-L.

When the FDA later learned of the testing and reporting irregularities, Gliatech was prosecuted by the Department of Justice. The company pleaded guilty to charges that it failed to submit adverse event reports, failed to maintain accurate files and submitted false and misleading reports to the FDA. Neither the FDA nor the Justice Department took any legal action against Stratprobe.

Statprobe Argues Suit Is Time-Barred

Statprobe asked for summary judgment on the basis of Pennsylvania's statute of limitations, 42 Pa..C.S. ” 5524-5525, which requires personal injury claimants to commence action within two years of the date of injury. Plaintiffs here argued that both the 'discovery rule' and the doctrine of fraudulent concealment tolled the statute of limitations.

Pennsylvania's discovery rule states that the statute of limitations begins to run on the first date that the injured person possesses sufficient critical facts to put him on notice that a wrong has been committed and that he needs to investigate to determine whether he is entitled to recovery. A plaintiff invoking the discovery rule bears the burden of proving he was unable to discover sufficient facts to assert a claim within the limitations period despite exercising reasonable diligence. Dalrymple v. Brown , 549 Pa. 217 (Pa. 1997).

Statprobe offered evidence that the plaintiffs could have discovered their involvement in the testing of ADCON-L much earlier, and within the two-year limitations period, if they had investigated Gliatech's Securities Exchange Commission (SEC) filings and had read an article recounting Statprobes involvement in ADCON-L's testing in a leading Ohio newspaper, the Plain Dealer. The court rejected this stance, stating, 'Statprobe's arguments place a weighty responsibility on Plaintiffs to scour the nation's newspapers for information related to their suit, a responsibility not reflected in case law. Plaintiffs are residents of Pennsylvania, and while it is true that the Plain Dealer is the self-proclaimed 'largest newspaper' in Ohio and is available online, Plaintiffs are not required as part of the concept of reasonable diligence to troll the local media of the nation's cities for possible defendants with potential responsibility for their injuries. Moreover, even if Plaintiffs had found the Plain Dealer article and Gliatech's Form 8-K (SEC filing) by 2000, neither of these documents identifies Statprobe by name. Thus, Plaintiffs still would have needed to locate some other source with more information of greater specificity.'

In addition, the court found sufficient evidence that Statprobe had fraudulently concealed its role in conducting and reporting the test results to preclude summary judgment on a statute of limitations basis. Specifically, in a July 1997 letter from Mark Becker, Vice President of Statprobe, to Raymond Silkaitis, Ph.D., Vice President of Gliatech, Dr. Becker expressed concerns that Gliatech was asking Statprobe to take part in bad statistical methods. He therefore asked in that same letter that Gliatech not use Statprobe's name in any way to indicate support of ADCON-L before any regulatory agency. From this evidence, said the court, 'a jury could conclude here that Statprobe took steps to conceal its identity by preventing publication of its name in connection with Gliatech or ADCON-L.'

In next month's issue, we'll discuss other bases for the U.S. District Court for the Eastern District of Pennsylvania's denial of summary judgment.

Janice G. Inman is Editor-in-Chief of this newsletter.