NEW YORK — The man who would be known as Patient No. 1 emerged from routine open-heart surgery at Columbia University Medical Center in stable condition. Then he began to bleed uncontrollably. Surgeons rushed him back to the operating room to reopen his chest, but by the time they could stop the hemorrhaging, Patient No. 1 was barely breathing and in a coma.
On Aug. 15, 2000, shortly before he was discharged on his way to a nursing home, a physician wrote a terse final diagnosis in his chart: “Medical disaster.”
Patient No. 1, along with more than 200 other open-heart surgery patients, was part of a two-year medical study at Columbia that government regulators now say was carried out with ethical and regulatory mistakes and may have caused harm to some patients. The study was testing a commonly used intravenous surgical fluid that previous studies had shown could cause hemorrhaging at high doses. At least two patients in the study died shortly after receiving the fluid and more than two dozen others required transfusions, according to documents submitted to the federal government by the hospital and obtained by the Huffington Post Investigative Fund.
In the past decade, Columbia has conducted three separate internal reviews of the study. The reviews raised serious questions about the drug trial’s design, management and oversight. But they concluded that there was no evidence that the fluid caused deaths or other medical problems for the patients and that there was no need to provide the patients with additional information about the study.
Now federal regulators have decided not to accept that conclusion. They have taken the rare action of demanding that Columbia track down the patients and their families, and acknowledge that they never were informed about the “true nature” of the drug study, the risks they faced or the consequences of their participation.
New information shows that “at least some of the subjects appear to have suffered harms that were a function of the design and procedures of the study,” the federal Office of Human Research Protections wrote to the hospital in a June 8, 2009, letter obtained by the Investigative Fund.
Federal officials also demanded that Columbia turn over a newly completed internal analysis of how the patients fared in the study.
The issues raised by the Columbia study, which was indirectly funded by a pharmaceutical company, reflect the ongoing national debate over flaws in the system designed to protect people who participate in medical research. The federal oversight office has cited more than 40 hospitals and academic medical centers in the past two decades for falling short. The Columbia case stands out for the bitter controversy it has engendered for years inside the hospital, the courts and the federal government – reported here for the first time – and for the hospital’s failure to contact patients even after federal investigators recommended it do so in 2003.
The study, conducted between December 1999 and February 2001 in the famed heart surgery unit at what is now called New York-Presbyterian Hospital/Columbia University Medical Center, involved four blood expanders approved by the U.S. Food and Drug Administration. The fluids are generally administered by anesthesiologists and combat medics when patients or soldiers have lost significant quantities of blood.
Two of the blood expanders in the study contained a substance known as hetastarch, a clear fluid made of a starch and salt solution. Published studies dating back to 1981 showed that hetastarch can prevent blood from clotting properly, especially when used at higher doses. According to documents filed by the hospital in New York state court, one purpose of the Columbia trial was to test whether a new formulation of hetastarch, manufactured by Abbott Laboratories, was less likely to trigger serious bleeding at high doses than the other fluids. It was largely funded from a $150,000 unrestricted grant given by the drug company to the hospital and lead researcher, records show.
In the consent form used in the study, patients were told that they would receive one of four fluids approved by the FDA and routinely “used to replace blood and fluid lost during surgery.” The consent form stated that the researchers would extract a few tablespoons of blood from the patient to test a machine that monitors clotting. Patients were not told that they could be given high doses of the fluids or that they faced a risk of serious bleeding, according to a copy of the consent form obtained by the Investigative Fund.
Documents later filed in court show that about half of the 215 people who agreed to participate were given hetastarch, and some received up to three times the level recommended by the manufacturers. Some of the subjects were Spanish-speaking patients who lived in low-income neighborhoods near the hospital and were admitted through the emergency room, according to people who worked at the hospital at the time. The names of the patients and details about their cases have not been made public because of medical privacy rules.
Two hospital doctors raised concerns about the study with hospital authorities in 2000, triggering the internal Columbia reviews. The hospital decided in 2002 to discipline the study’s lead researcher because, Columbia alleged, he had not properly disclosed the nature of the drug study to the hospital or the patients and had failed to report promptly a “substantial number” of medical complications among the participants, according to court papers. The researcher, Elliott Bennett-Guerrero, an anesthesiologist, subsequently filed a lawsuit against the hospital and its officials that vigorously challenged their claims and decision. The lawsuit ended with a confidential settlement in 2003, court records show, and Bennett-Guerrero left Columbia for another hospital.
Columbia hospital officials declined requests for interviews and would not discuss the recent findings by federal regulators that some patients appear to have been harmed or the government’s demand that the hospital notify the study’s participants.
In a statement to the Investigative Fund, Columbia said its internal reviews had concluded that neither patients nor the hospital board that approves clinical trials “were adequately informed of the risks posed by one of the treatments in the study.” Nevertheless, the hospital said, its most recent review completed in 2008 — which included outside experts — analyzed patient records and concluded that the medical outcomes did not meet the definition of “harm.”
Columbia also said that as a result of its investigations it had made “substantial improvements” in its procedures for overseeing research on humans.
In the lawsuit he filed against Columbia in 2003, Bennett-Guerrero said that proper consent was obtained from all the patients in the study. He said there was no misrepresentation of the study’s design or purpose, that hospital officials had been fully informed and had approved every aspect. He contended that their actions against him were meant to hide weaknesses in their own hospital procedures.
Bennett-Guerrero, who joined Duke University Health System in 2003, declined a request for an interview. He said in e-mails: “It is hard to imagine that an unbiased expert in cardiac surgery clinical trials could conclude that subjects were harmed in this study, since with only 50 patients per group the study was not designed or powered to prove any differences in major complications including death.”
Bennett-Guerrero wrote that the study proposal and consent form “were approved by Columbia’s Institutional Review Board. The Columbia IRB sought comments from members of the Departments of Surgery, Anesthesiology, and Medicine and the IRB had before it the package inserts for each of the four FDA approved fluids, as well as the protocol and the consent form.”
He added: “Please understand that I am, and have throughout my entire professional career been, committed to patient safety and improving patient outcomes. Indeed, as a practicing anesthesiologist who takes care of high risk patients, my primary focus in the operating room is patient safety and reducing pain and suffering.”
An Unrestricted Grant
The Columbia study came at a time when Abbott Laboratories, the manufacturer of one of the blood expanders, was looking to boost its share of the business. The fluids were often needed during more than half a million cardiac surgeries each year and in the late 1990s the market for blood expanders containing hetastarch was growing due to a shortage of albumin, one of the older, more commonly-used products.
In 1999, Bennett-Guerrero, then 34, was recruited to serve as clinical director of Columbia’s division of cardiothoracic anesthesiology. Within two years, records show, he was simultaneously running 25 clinical trials. He received approximately $150,000 in the form of an unrestricted grant from Abbott Laboratories as reimbursement for the comparative study of blood expanders, according to his statements in his lawsuit against the hospital. Medical centers welcome such grants, since they typically can take a portion for overhead.
Several previous studies had shown that the original hetastarch product could sometimes trigger excessive bleeding during surgery. If Abbott’s new formulation of hetastarch – called Hextend — turned out to be safer in high doses, anesthesiologists might be persuaded to switch, even though Abbott’s price was about 40 percent higher.
Before the study could begin, it had to pass muster inside the hospital. Under federal regulations, every clinical trial must be approved by an institutional review board, or IRB — a panel of doctors, other medical professionals and at least one non-medical professional from outside the hospital — that is charged with protecting human test subjects and ensuring that they are fully informed of the potential risks. The board must also ensure that studies are properly designed.
According to court documents, half of the open-heart patients in the study were slated to receive one of the two hetastarch solutions. The other half would get either albumin or a salt solution.
The original proposal requested a waiver from the standard requirement of obtaining written patient consent, on the grounds that participation in the trial “will not increase the likelihood of patients requiring blood transfusions . . . [or] any additional discomfort or risk.”
The study proposal and request for a waiver was reviewed in 1999 by the IRB, whose approximately 12 members met once a month. But there was no expert in blood expanders on the board and no member examined the published studies about the risks of high levels of hetastarch, according to court documents.
The review board did insist that patients sign a written consent form. Columbia investigators later concluded that the consent form failed to inform patients of the risk of bleeding. They also found that the IRB was unaware of those risks, in part because the panel “failed to adequately use data provided” by the hospital’s departments of surgery and anesthesia, which also reviewed the proposed study. A surgeon on the review board told the investigators that the board’s members, 11 of whom had other full-time duties in the hospital, didn’t have enough time to probe. “When we do these reviews we are presented with the investigator’s stack of IRB stuff,” he said, according to the court papers. “Most of us barely get to read the birthday cards from our kids . . .”
‘A Very Common Deficiency’
In November 2000, two Columbia anesthesiologists – Marc Dickstein and Mark Heath– sought out the head of the institutional review board, Paul Papagni, a lawyer. They told Papagni that they had been in the operating room when a number of patients had hemorrhaged. They feared the study’s design virtually guaranteed that there would be more who would suffer hemorrhaging, according to Heath’s statements to hospital investigators, included in court filings.
Dickstein would later tell Columbia investigators that he and Heath assumed the study would be suspended and reviewed since they had alerted the board. In court documents, he said, “We were two reasonably senior members of the cardiac anesthesia team coming in saying patients are being harmed. . . . [we thought] anyone who actually would look at the literature [on blood expanders] . . . would come to the same conclusion.” But the IRB did not suspend the study.
Heath and Dickstein declined to comment for this article. Attempts to contact Papagni were unsuccessful.
Court records show that Bennett-Guerrero and his department head, Margaret Wood, disagreed with the assessments of Heath and Dickstein. Columbia investigators later suggested that the concerns raised by Heath and Dickstein may have been initially cast aside as rooted in professional rivalry with Bennett-Guerrero.
Five months after the study ended, Heath and Dickstein wrote to Gerald Fischbach, dean of Columbia’s medical school, with their concerns. According to court documents, Fischbach soon ordered Bennett-Guerrero to stop enrolling patients in studies, pending the results of an investigation.
Fischbach later removed Bennett-Guerrero as clinical director of the division of cardiothoracic anesthesiology, according to court records. The university took him off tenure track in 2002, barred him from conducting research, and told him in a letter that he could not publish the results of the blood expander study.
In September 2002 Columbia sent a letter about the matter to the federal Office of Human Research Protections, part of the Department of Health and Human Services. The letter, obtained through the Freedom of Information Act, affirmed the chief complaints that Dickstein and Heath had raised. Columbia alleged that Bennett-Guerrero “failed to convey” the purpose of the study to the review board and patients, had not informed the patients of the risks and did not appropriately report serious medical complications. The letter also faulted the manner in which fluids were prepared for the study (they were allegedly mixed in an unsanitary, blood-spattered room). Columbia also assured federal regulators in the letter that it was overhauling its review process.
But hospital officials stated that patients had not suffered harm as a result of participating in the study. Columbia’s investigation, said the letter, “failed to show a causal relationship” between the fluids and the two deaths. It also added that there was “no evidence of harm to any particular patient that could be attributed to the study.”
What the letter did not say was how Columbia investigators calculated harm. Columbia reviewed only 14 patient charts of the more than 200 in the study, according to court documents. The investigators looked for kidney damage, another potential side-effect of blood expanders, but at that point did not report on bleeding.
Fischbach, who retired as dean in 2006, declined requests for an interview.
Federal officials responded to Columbia in a letter dated January 2003. They suggested that the university “re-consent patients,” which meant finding them or their survivors and informing them that the study may have put them at greater risk than they had been told when they gave informed consent. But the federal regulators didn’t force the issue. Columbia decided, as it had the year before, that there was no need to tell the patients.
That same month, Bennett-Guerrero filed suit against Columbia and Fischbach in New York Supreme Court, claiming the university had harmed his reputation and wrongly stripped him of his ability to conduct and publish research. Bennett-Guerrero attacked the university’s findings about the study and denied that he acted improperly in any way. The settlement, reached in June 2003, is confidential but the court file remains public.
E-mails obtained through a public records request from the Office of Human Research Protections show that Heath and Dickstein continued to ask the government to re-examine the study and its outcomes. In July 2006, a compliance officer responded that the federal agency would not challenge Columbia again. Failing to disclose risks to subjects in completed research was “a very common deficiency,”the officer wrote in an e-mail, adding that the agency “is not inclined, at this time, to investigate this matter further.”
Then in early 2007, for reasons that are not apparent from any documents, Columbia initiated yet another internal examination of the blood expander study. It was completed in the fall of 2008 and the hospital contacted federal regulators and acknowledged some deficiencies in a letter dated March 31, 2009. In the letter, the hospital again concluded that “the conditions necessary for a finding of patient harm had not been met.”
This time, federal regulators balked. In its letter to Columbia on June 8, the oversight office wrote that “new information” provided by the hospital now showed that among study participants who received hetastarch there was “a statistically significant higher rate” of “negative clinical outcomes, including bleeding events (requiring use of transfusions) and decreases in renal [kidney] function. Beyond that, there was the trend toward increased need for re-operation.” The letter said that the analysis “supports the hypothesis” that patients who received hetastarch did worse than the others.
The federal office instructed the hospital to draft a letter explaining the study to its former patients. Regulators also asked Columbia to hand over a full accounting of what happened to the patients who agreed to be part of its study. The federal agency has issued such a directive only three other times since 2000. Although the office has no direct authority to enforce its demands, it can cut off federal research funding to institutions that fail to comply.
This week, neither Columbia nor the federal government would say when the patients and families will learn the whole story behind the heart operations they underwent 10 years ago.
Jeanne Lenzer is an independent medical investigative journalist and a frequent contributor to the BMJ (formerly the British Medical Journal). She can be reached at email@example.com. Shannon Brownlee is a senior research fellow at the New America Foundation and the author of Overtreated: Why Too Much Medicine is Making Us Sicker and Poorer. She can be reached at firstname.lastname@example.org.
Help support this work
Public Integrity doesn’t have paywalls and doesn’t accept advertising so that our investigative reporting can have the widest possible impact on addressing inequality in the U.S. Our work is possible thanks to support from people like you.