Pills That Kill:
A Tale of Two Drugs

The old joke – "The medical treatment was a success but, unfortunately, the patient died" –  is not at all funny:  it’s an all too frequent reality when it comes to some medications that doctors routinely prescribe.  Consider two popular drugs still being sold:  an anti-depressant that can attack your liver, and a diet pill that goes after your heart.  It’s becoming increasingly evident that, like two little piggies, some drugs are rushed off to market when they probably should have stayed home.

With the drugs Serzone and Meridia, the Food and Drug Administration has had to step in and take measures to protect the lives of the very patients whose welfare the medications are intended to improve and protect.

Serzone The anti-depressant, Serzone, is made and marketed by Bristol-Myers Squibb Co.  Its generic name is nefazodone hydrochloride.  Since approval by the FDA in 1994, it has been used by more than 7.2 million patients in the United States alone.  It works by blocking the body’s absorption of the mood-regulating hormones, serotonin and norepinephrine.  Although its structure and composition put Serzone in a class all its own, it does bear some similarity to Prozac, Paxil, Zoloft and other drugs collectively known as selective serotonin reuptake inhibitors (SSRIs).  Despite pre-marketing evidence that Serzone posed no danger to the liver, within five years of its arrival in pharmacies there were at least three reports of patients who suffered liver failure while taking the drug.  The cases were summarized in American Family Physician in September 1999:

• A 54-year-old woman, being treated for anxiety, developed jaundice six weeks after starting to take Serzone.  The liver failure persisted despite being taken off the medication.  After another six weeks her liver, kidneys and brain progressively degenerated until she had to undergo a liver transplant.  She died soon after the transplant.
• Three months after a 16-year-old girl started taking Serzone for depression, she developed nausea, vomiting, fatigue and jaundice.  The medication was withdrawn but her symptoms persisted.  Two weeks later she started becoming more and more confused, sleepy and disoriented, symptoms of the central nervous system’s response to a failing liver. She had a liver transplant and was doing well 10 months later.
• A 57-year-old woman who had taken the anti-depressant for six months was admitted to a hospital with a variety of  symptoms of liver deterioration.  Her liver was found to be severely inflamed.  She was taken off the medication and listed for a liver transplant.  Her condition improved enough under treatment, however, and she did not need the transplant.

The FDA now estimates that Serzone is linked to one liver failure that results in a transplant or death for every 250,000 to 300,000 patient-years of treatment.  (Each patient-year is equivalent to one patient taking the drug for one year, or two patients each taking it for six months, etc.) 

This rate is three to four times higher than normal, and the FDA says the drug-induced failure rate may be even higher due to under-reporting.  That’s because the FDA gets reports of deaths and adverse reactions from voluntary submissions by physicians and hospitals.  (Drug manufacturers, however, are required to make such reports, but they don’t always know of the adverse effects caused by their products and don’t always comply with the requirement when they do know.  See “Meridia,” below.)  Thus, the number of actual adverse reactions is typically thought to be 10 times higher than reported.  As drug companies like to point out, however, listing a death in the FDA's files means the drug is suspected, not that it is proven that it caused the death. 

The FDA’s Response

The FDA was alarmed enough to order Bristol-Myers Squibb to add a “black box” warning – a warning within a bold black frame – to its printed prescription information, that fine-print document included in every package of pills.  The warning reads:

“Cases of life-threatening hepatic (liver) failure have been reported in patients treated with Serzone.” 

The pharmaceutical company also inserted the following in its printed “Information for Patients”: 

“Patients should be informed that Serzone therapy has been associated with liver abnormalities ranging from increases of asymptomatic reversible serum transaminase (an enzyme in the blood) increases to cases of liver failure resulting in transplant and/or death.  At present there is no way to predict who is likely to develop liver failure.” Click here to view the FDA's letter to healthcare practitioners regarding the black box warning. (Requires Adobe Acrobat Reader.)

Not the Only One

Serzone is but one of more than 800 drugs that have been implicated in liver disease.  Indeed, medications account for more than 30 percent of all cases of acute liver failure in the United States and for as many as 20 percent of all jaundice cases admitted to the hospital – and the risk level increases with age.  Drug-induced liver failure can be indistinguishable from other types of liver disease, even though each drug does tend to leave its own signature – its typical pattern of injury. 

Doctors are being advised to err on the side of caution by monitoring liver functioning in all patients for whom they prescribe Serzone, and to take them off the drug _ and keep them off – at the first sign of trouble.  Because many more patients will develop transient symptoms rather than full blown jaundice, this means the drug probably may sometimes be withheld unnecessarily, but the precaution is prudent because of the inability to predict which patients are at risk of a severe reaction to the drug.  Dire consequences that can arise from that reaction as about 10 percent of all types of drug-induced liver inflammation end in death. 

Serzone’s Other ‘Problems’

The liver-damage warning was added to an already lengthy list of harmful side effects and other problems identified in pre-marketing trials of Serzone.  Worldwide, 3,496 patients participated in the trials, but some 16 percent of them had to drop out because of adverse reactions.  The most common of adverse reactions (compared to those on placebos) were drowsiness, dry mouth, nausea, dizziness, constipation, asthenia (weakness or loss of energy), lightheadedness, blurred vision, confusion and abnormal vision. 

Also in those pre-marketing studies, 5.1 percent of the patients on Serzone developed low blood pressure, compared to 2.5 percent on a placebo.  This could pose particular problems for anyone with a history of heart disease or stroke, all too frequent precursors of depression, the very condition that Serzone is intended to relieve. 

Further, when combined with other drugs, SSRIs that work like Serzone have produced reactions ranging from severely high temperatures to “extreme agitation progressing to delirium and coma.”  It is not unlikely that Serzone may have the same incompatibilities.

Meridia The suspect diet pill is Meridia, made by Abbott Laboratories.  It is sold in 70 countries and has been used by some 9 million people worldwide since it went on the market in early 1998.  It had been approved by the FDA the previous year despite studies that showed the drug slightly increased the blood pressure and heart rate of some patients and substantially increased them in others.  The FDA’s medical officer who reviewed the drug concluded that it “has an unsatisfactory risk-benefit ratio and therefore this Reviewer recommends non-approval of the original submission.”The agency’s advisory committee voted five-to-four against its approval. The FDA approved it anyway, in November 1997.

How It Works

Meridia’s generic name is sibutramine. Like both the anti-depressant Serzone and the diet drug, fenfluramine (remember the fen-phen combination recalled in 1997 for causing heart valve damage?), Meridia affects the body’s use of serotonin.  However, whereas fen-phen boosted production of serotonin to fool the brain into making patients feel their bellies were full, Meridia uses the opposite approach, slowing the body's absorption of the brain’s natural serotonin output to achieve the same self-deception.

Death by Diet Pill

Since February 1998, when it went on the market, Meridia has been linked with:

• 34 deaths worldwide, 29 in the United States, 19 of them from heart attacks or other cardiovascular causes;
• 10 cardiac deaths of persons 50 years old or younger, including three women under 30;
• 397 serious adverse reactions reported to the FDA, including
• 152 patients whose reactions were severe enough to require hospitalization, and
• 143 patients in whom arrhythmia (an abnormal heartbeat) was reported.

Abbott Laboratories disputes the validity of the numbers, arguing that it hasn’t yet been proven that the diet pill caused the effects.  Numbers aside, the adverse reactions to Meridia that have been reported to FDA include such serious concerns as:

• Stroke
• Heart attacks and other cardiovascular injuries
• Arrhythmia
• Blood clots near the eyes
• Increased blood pressure and heart rate
• Body aches; neck and chest spasms
• Nervousness, hyperactivity, and anxiety

Meridia’s effectiveness as a weight-loss aid is also being questioned.  In approving the drug, the FDA reported that in clinical trials, obese people taking Meridia for one year could expect to lose an average of only 6.5 pounds compared to those taking a placebo. The risk-benefit ratio question:  is losing 6.5 pounds worth the heightened risk of a fatal heart attack?

Abbott Under Fire

March 19, 2002:

The consumer advocacy group Public Citizen cited these statistics in asking the FDA to recall Meridia. 

March 21-April 3, 2002:

Two days later, the FDA dispatched a pair of investigators to Abbott Laboratories’ plant in Illinois for two weeks to look for clues to determine whether Meridia was responsible for the 32 deaths then reported.  (They’d already been there in January on other, similar inspection visits regarding other Abbott products.)  In the pharmaceutical giant’s database they found at least one death that had never been reported, plus "several records … (involving seven other deaths in which) the adverse drug information reported to FDA was either not accurate, not supported by source data, or was missing additional information found in the source data,” according to the inspection report. 

May 22, 2002:

Public Citizen asked the government to press criminal charges against Abbot for withholding crucial information about adverse events and eight patient deaths. 

FDA’s Wrist-Slapping Response

The FDA took action of sorts on July 19, 2002.  Its Chicago office sent Abbott Laboratories a sternly worded letter, then posted the warning on the FDA’s website, with names and other identifying information blanked out (“redacted”).  The letter is available at http://www.fda.gov/foi/warning_letters/g3420d.htm.  Some telling excerpts: 

“The necessary data… (required to be reported to the FDA) are: (1) an identifiable patient; (2) an identifiable reporter; (3) a suspect drug; and (4) an adverse event or fatal outcome. These elements were present in the source documents for [identity redacted] but that event was not reported by your subsidiary, Knoll.  Your explanation – that the caller mentioned this serious and unexpected adverse drug event (a death) was a ‘rumor’– does not affect the requirement to report this event. Since the necessary data elements were available, your firm was required to promptly report the serious, unexpected adverse event to FDA…. 

“In further violation… you submitted adverse drug experience information in reports to FDA which was inaccurate in some cases, or did not completely reflect information available in the source documents, such as in three reports of death associated with the use of Meridia…. 

“Your response… contained additional information that was not in the reports submitted to FDA, and presented explanations for your selection and presentation of data in these reports. These explanations do not adequately address either the omission of information or the inaccurate presentation of information in the reports submitted to FDA.”

The FDA letter itemized the three faulty case reports:

1.	Abbott reported that the effect resulted from a single dose on one day, but source documents contained no such information.
2.	Abbott reported no autopsy findings, even though source data stated that an "autopsy did not reveal anything other than left ventricular hypertrophy” (enlargement of the heart’s left ventricle).
3.	Although source documentation stated the patient was in her 30s and “was not known to have any type of heart disease,” the Abbott report to FDA said her age was unknown and said nothing about having found no history of heart disease.

The FDA letter to Abbott Laboratories continued:

“FDA expects drug manufacturers to establish procedures to ensure that their foreign affiliates and corporate units rapidly transmit (information) to expedite reporting of serious and unexpected adverse drug experiences to FDA…. 

“You should take prompt action to correct these deviations. Failure to promptly correct these deviations may result in regulatory action without further notice.  These actions include, but are not limited to, seizure or injunction…. 

“Although your firm has taken some corrective actions to address deficiencies found during the … March-April 2002 inspections of your firm, you have not addressed many of our concerns. Of the deficiencies you address in your letter of April 18, 2002, we find many of your responses inadequate….” 

(09/13/02)