This new labeling would be required for all OTC drug products containing an IAAA active ingredient, whether marketed under an OTC drug monograph or an approved NDA. FDA bases this proposal on its reviews of the medical literature, data provided to FDA, and recommendations made by NDAC. FDA has tentatively concluded that new labeling for OTC IAAA drug products is necessary for the safe and effective use of these products by consumers.

II. Background

FDA believes that acetaminophen and NSAIDs, when labeled appropriately and used as directed, are safe and effective OTC drug products that benefit tens of millions of consumers every year. FDA believes that these products should continue to be accessible to consumers in the OTC setting.

A. Development of OTC IAAA Drug Product Warnings

The development of a monograph for OTC IAAA drug products began in 1977 with publication of an expert panel report and continued in 1988 with publication of the TFM. The development of labeling for OTC IAAA drug products is recorded in the following documents.

1. Warnings for Aspirin and Acetaminophen

In the Federal Register of July 8, 1977 (42 FR 35346), FDA published the report of the Advisory Review Panel on OTC Internal Analgesic, Antipyretic, and Antirheumatic Drug Products (the IAAA Panel) for OTC IAAA active ingredients: Acetaminophen, aspirin, carbaspirin calcium, choline salicylate, magnesium salicylate, and sodium salicylate. The recommendations included labeling and warnings for:

In the Federal Register of November 16, 1988 (53 FR 46204), FDA published a tentative monograph with the following warnings for:

2. Warnings in the Professional Labeling for Aspirin

In the Federal Register of October 23, 1998 (63 FR 56802), FDA published labeling for health professionals (not available in OTC drug product labeling) that provided for cardiovascular and rheumatologic indications. The labeling listed adverse reactions reported in the literature, e.g., hypotension (low blood pressure); tachycardia (rapid heart rate); dizziness; headache; dyspepsia (indigestion); bleeding, ulceration, and perforation of the gastrointestinal (GI) tract; nausea; and vomiting. FDA determined that consumers were not able to determine when they needed to take aspirin to prevent cardiovascular events, such as stroke, myocardial infarction (damage to the heart muscle), or other conditions. FDA did not
[[Page 77316]]
consider it possible to provide adequate directions and warnings to enable the layperson to make a reasonable selfdiagnosis of these cardiovascular and rheumatologic conditions.

3. Alcohol Warnings for Acetaminophen and NSAIDs

In the Federal Register of October 23, 1998 (63 FR 56789), FDA published a final regulation stating that any OTC drug product, labeled for adult use, containing acetaminophen, aspirin, carbaspirin calcium, choline salicylate, ibuprofen, ketoprofen, magnesium salicylate, naproxen sodium, and sodium salicylate must bear an alcohol warning statement in its labeling. Section 201.322 requires the following statements:

Alcohol Warning: If you consume 3 or more alcoholic drinks every day, ask your doctor whether you should take acetaminophen or other pain relievers/fever reducers. Acetaminophen may cause liver damage.

Alcohol Warning: If you consume 3 or more alcoholic drinks every day, ask your doctor whether you should take (name of active ingredient) or other pain relievers/fever reducers. (Name of active ingredient) may cause stomach bleeding.

Alcohol Warning: If you consume 3 or more alcoholic drinks every day, ask your doctor whether you should take (insert acetaminophen and one other IAAA active ingredientincluding, but not limited to aspirin, carbaspirin calcium, choline salicylate, magnesium salicylate, or sodium salicylate) or other pain relievers/fever reducers. Acetaminophen and (insert name of one other IAAA active ingredient including, but not limited to aspirin, carbaspirin calcium, choline salicylate, magnesium salicylate, or sodium salicylate) may cause liver damage and stomach bleeding.
4. Proposed Amendment to Include Ibuprofen as a Generally Recognized Safe and Effective OTC IAAA Active Ingredient

In the Federal Register of August 21, 2002 (67 FR 54139), FDA proposed to include ibuprofen in the monograph for OTC IAAA drug products with additional warnings:

Ask a doctor before use if you have:

FDA received several comments (Refs. 1 and 2) about the proposed warning for kidney disease and reopened the administrative record on June 4, 2003 (68 FR 33429), to allow for additional public comment. FDA continues to propose a warning about kidney disease for ibuprofen and other NSAIDs in this document. In a future issue of the Federal Register, we will publish our final decision about this warning and the proposed inclusion of ibuprofen in the monograph.

B. Completion of the OTC IAAA Drug Products FM

In the process of completing the FM for OTC IAAA drug products, FDA reviewed a variety of data regarding the safety of acetaminophen, aspirin, and other NSAIDs. FDA continued to receive serious adverse event reports associated with the use of these products during this review. These serious adverse events included unintentional acetaminophen hepatotoxicity and NSAIDrelated GI bleeding and renal toxicity. Although the occurrence of these events is rare, relative to the extensive use of the products, as described in the text that follows, FDA believes that labeling changes are necessary for the safe and effective use of these products and to reduce the associated morbidity.

1. Unintentional Acetaminophen Hepatotoxicity

Acetaminophen is widely available in numerous single ingredient and combination OTC drug products, and in many prescription drug products, as a pain reliever and/or fever reducer. OTC acetaminophen drug products, as currently labeled and used, have been reported to be associated with unintentional overdose that may lead to serious hepatotoxicity (Ref. 3). The IAAA Panel discussed overdoserelated hepatotoxicity (42 FR 35346 at 35413 to 35414), and FDA addressed it in the IAAA TFM (53 FR 46204 at 46213 to 46218). (See section II.A.1 of this document.)

2. Aspirin and Other NSAIDsGI Bleeding and Renal Toxicity

Aspirin and other NSAIDs are available OTC for the treatment of minor aches and pain, for the treatment of headaches, and for fever reduction. Per aspirin's professional labeling (not part of the OTC drug product labeling), aspirin may be used to reduce the risk of serious cardiovascular events when taken on a daily basis under the direction of a physician. Aspirin is also effective in treating a variety of rheumatologic diseases under the direction of a physician. The professional labeling also includes information about the potential risk of GI bleeding and renal toxicity associated with aspirin.

OTC nonaspirin salicylates include the NSAIDs ibuprofen, naproxen sodium, and ketoprofen. The product labels for these products are not required to contain warnings about GI bleeding and renal toxicity. These ingredients are, however, also available by prescription at strengths higher than in OTC products and the prescription product labeling contains warnings about these risks.

III. NDAC Meeting

At a September 19 and 20, 2002, meeting, NDAC considered products currently marketed with OTC IAAA ingredients, including acetaminophen, aspirin, carbaspirin calcium, choline salicylate, ibuprofen, ketoprofen, magnesium salicylate, naproxen sodium, and sodium salicylate. FDA expressed its belief that these products should remain available OTC given their overall effectiveness and safety, the benefit to consumers of having a pain reliever and fever reducer available OTC, and the use of these products by tens of millions of people weekly. FDA suggested that certain interventions could decrease the frequency and morbidity of these serious adverse events. NDAC members were asked to consider which additional interventions were necessary to reduce the occurrence of serious adverse events. The presentations made at the meeting, and NDAC's findings, are summarized in this document. More information about the September 2002 NDAC meeting is available on the Internet and in the Division of Dockets Management (see ADDRESSES). A. Data and Information Reviewed

FDA provided NDAC with the following data and information (Ref. 3):

NDAC also considered submissions and presentations from industry and individuals during the open public sessions (Refs. 4 and 5). B. Acetaminophen

On the first day of the meeting (September 19, 2002), NDAC considered safety issues related to the use of acetaminophen, unintentional overdose, and the potential for hepatotoxicity from both OTC and prescription acetaminophen products.

1. Points for Discussion

FDA asked NDAC to discuss possible factors that might contribute to unintentional overdose (Ref. 3) and provided the following points for consideration:

FDA asked NDAC what additional measures could be taken to better ensure that prescribers and other people are aware of the potential risks associated with exceeding the recommended dose of prescription or OTC drug products containing acetaminophen and with using multiple products containing acetaminophen. FDA suggested the following possible measures for OTC drug products:

For prescription products, FDA suggested:

FDA also asked NDAC if there are identifiable factors that might make some individuals more susceptible to hepatic toxicity (e.g., underlying liver disease, malnutrition, drug interactions, and alcohol users). If subpopulations at increased risk of acetaminopheninduced hepatotoxicity could be identified, FDA asked NDAC what reasonable measures could be taken to decrease their risk. FDA suggested some possible measures:

FDA asked NDAC whether additional studies are needed to evaluate these issues. FDA suggested a number of subjects for potential research:

As a leadin to the liver toxicity discussion, Dr. William Lee, of the University of Texas Southwestern Medical Center at Dallas, presented the results of acute liver failure (ALF) studies in the United States (Ref. 6). He estimated that between 1,000 and 2,000 ALF cases occur in the United States each year and are associated with high mortality. Dr. Lee conducted a retrospective analysis of 177 cases of ALF reported in the literature between 1986 and 1998. Of these, 20 percent were attributed to acetaminophen toxicity. To study ALF prospectively, Dr. Lee also formed a study group of 25 treatment centers in 1998. Details of the group's initial 308 cases are presented in table 1. Approximately 40 percent of the cases were due to acetaminophen toxicity, which was increased when compared to the rate of acetaminophen toxicity in the cohort from Dr. Lee's retrospective analysis.
Table 1. Study Group Series of ALF Cases (N = 308) ALF Etiology
Case Report Data All Other Acetaminophen Induced Drug (Not Acetaminphen) Indeterminate Causes P value (n=120) Induced (n=40) Cause (n=53) (n=95) Sex (% Female) 79 73 60 72 NS* Age (years) 36 41 38 43 0.02 Jaundice (days) 1 12 12 4 < 0.001 Coma (%) 50 43 47 47 NS Alanine aminotransferase (ALT) (International Units/ 4310 574 947 1060 < 0.001 Liter (IU/L))**
Bilirubin 4.3 20.2 24.5 12.6 < 0.001 [[Page 77318]]
Transplant (%) 6 53 51 36 < 0.001 Spontaneous survival (%) 68 25 17 33 < 0.001 Overall survival (%) 73 70 64 61 NS * Not significant ** ALT (normal range 035 IU/L)

Of the 120 acetaminophen toxicity cases identified in Dr. Lee's series, 12 were omitted due to concomitant patient issues that would have confounded the analysis. The remaining 108 cases were analyzed and showed that alcohol use was reported in 57 percent of the cases and alcohol abuse was reported in 19 percent of the cases. Individuals in 38 percent of the cases were taking both narcoticacetaminophen prescription products and OTC acetaminophen products at the same time, some for as long as 2 to 3 months. In 70 percent of the cases, patients ingested more than 4 grams (g) of acetaminophen per day (recommended maximum daily dose), and 32 percent of the cases reported ingestion of more than 10 g per day.

A comparison was conducted among the 108 cases of toxicity due to accidental (ingestion of drugs for pain relief, without suicidal intent) and suicidal (ingestion with admitted suicidal attempt) ingestion. The type of ingestion could not be determined in 5 cases, resulting in a comparison of 103 cases (table 2). More than half of the acetaminophen toxicity cases (57 percent) were accidental.
Table 2.Suicidal vs. Accidental Acetaminophen ALF Cases Accidental (n=59) Suicidal (n=44) pvalue Age 39 33 0.011 Acetaminophen total (g) 20 29 NS Antidepressant 36% 34% NS Alcohol (nonabuse use) 55% 61% NS Double use* 24% 5% 0.02 Narcotic/acetaminophen 54% 14% 0.001 ALT (IU/L) 3,616 5,929 < 0.001 Creatine 2.5 1.3 0.008 Survival 71% 75% NS * Use of more than one acetaminophen containing product.

The incidence of use of antidepressants and alcohol was nearly identical in the accidental and suicidal groups. The accidental cases included a larger percentage of subjects who doubledosed or used a narcotic/acetaminophen combination product. Survival rates were also similar. Lee concluded that acetaminophen toxicity accounted for about a third of all deaths from ALF in this case series and appears to be a growing problem in the United States.

FDA staff presented a safety analysis of hepatotoxicity associated with acetaminophen (Ref. 7). The cases were reported as ``intentional overdose'' and ``unintentional overdose.'' The reported doses were rarely within the recommended range. Four national databases were used to estimate the occurrence of these events:

1. National Hospital Ambulatory Care Survey: Emergency Department (ED) Componenta probability survey sampling of visits made to emergency departments and short stay hospitals in the United States.

2. National Electronic Injury Surveillance Systemcollects information on consumer productrelated injuries treated in emergency departments of 66 selected hospitals.

3. National Hospital Discharge Surveya probability survey sampling of patient discharge records from nonFederal, short stay hospitals in the United States.

4. Multiple Cause of Death Filesa data file that contains information from death certificates.

Acetaminophen overdose (unintentional and intentional) was associated with an annual average of over 56,000 emergency department visits (1993 to 1999) and more than 26,000 hospitalizations (1990 to 1999). Between 1996 and 1998, an annual average of 458 deaths was attributed, at least in part, to acetaminophen overdose. Unintentional acetaminophen overdose was associated with an annual average of over 13,000 emergency department visits (1993 to 1999), 2,189
hospitalizations (1990 to 1999), and 100 deaths (1996 to 1998). Each event in these tallies is independent from the others. No information about associated
[[Page 77319]]
hepatotoxicity was available for these cases. FDA reviewed the age distribution for acetaminophen overdoses. Medication use varies by age and different OTC drug products containing acetaminophen are available for different age groups. The age distribution of unintentional overdose cases varies among reporting databases and is shown in table 3. While emergency department visits are most prevalent among young people, this age group accounts for the lowest percentage of cases of mortality.
Table 3.Age Distribution of Unintentional Cases Age (years)
<17 1764 >65 Emergency department visit 74% 25% < 1% Hospitalization 23% 70% 7% Mortality 1% 75% 23%

Chronic liver disease has been postulated to be one of the factors that increases the risk of hepatotoxicity from acetaminophen. Using the multiple cause of death database, the presence of chronic liver disease among cases of unintentional and intentional overdose with mortality outcomes was examined (table 4).
Table 4.Percent of Liver Disease Reported Among Fatal Acetaminophen Overdose Cases, Mortality Data, 19961998 Liver Disease Reported Unintentional (N=235) Intentional (N=1,010) Chronic alcoholic 13% 1% Other chronic liver disease 48% 8%

These findings suggest that chronic liver disease, in the presence or absence of alcohol, may be a risk factor for developing or increasing severity of hepatotoxicity among people with unintentional overdose. However, this analysis has limitations. If the presence of alcohol or alcohol use was not mentioned on the death certificate, alcohol related liver disease may be misclassified as other chronic liver disease. In addition, suicide cases may be misclassified as unintentional overdose to protect privacy.

FDA also presented an analysis of cases of hepatotoxicity associated with acetaminophen from the published literature. A MEDLINE search identified all U.S. case series containing at least 10 cases that had been published in the previous 10 years (Ref. 7). Eight case series were identified, four of which were derived exclusively from review of hospital medical charts. In two series, cases were obtained from hospitals, published cases, the FDA adverse event reporting system, and poison control center databases. One case series was from a registry of cases reported by hepatologists and other practitioners. One case series was obtained exclusively from a consortium of liver transplant centers. The number of cases per series ranged from 47 to 73. Two case series were largely pediatric, and the remaining six case series consisted of largely adult populations. Six of the case series reported gender, and in all six there was a preponderance of females. Intentionality was reported in five of the series. Table 5 shows the acetaminophen dose reported among in the unintentional overdose groups. Table 5.Hepatotoxicity Series: Unintentional Toxicity Cases No. of Cases With Typical Case Series Reported Daily Doses (g/day) No. of Cases in Series Daily Dose of < =4g/day Johnston 1.320 53 9 Schiodt 230 21 3 Zimmerman ``<4''``>15''* 67 27 Whitcomb 3.525 21 None Broughan 15.9 (mean) 8 None * Dose was reported categorically.

Nine people in the Johnston case series and three people in the Schiodt case series ingested 4 g/day or less of acetaminophen. In the Zimmerman case series, 27 people used acetaminophen at the recommended dose, while 13 people used between 4.1 and 6 g/day. In the Whitcomb case series, 3 people used acetaminophen at, or slightly above, the recommended dose (i.e., 3.5 to 5 g/day in one case and 4 to 6 g/day in two cases). In the Broughan case study, none of the people took acetaminophen at the recommended dose.

[[Page 77320]]

Table 6 compares the number of deaths and serious outcomes for the unintentional and intentional groups. Intentionality could only be compared in the adult case series. Serious outcomes were defined as hepatic coma, acute liver failure, and liver transplant.

Table 6.Comparison of Unintentional and Intentional Toxicity Groups: Cases of Death or Serious Outcome Case Series Unintentional Intentional Johnston 17/53 NA* Schiodt 11/21 4/50 Zimmerman 13/67 NA* Whitcomb 5/21 NR** Broughan 2/8 0/40 *NA: Not applicable; **NR: Not reported

FDA also presented case data from the TESS of the American Association of Poison Control Centers (AAPCC). At that time, AAPCC had a repository of over 27 million human poison exposures reported by over 60 participating centers. These centers covered over 90 percent of the U.S. population. Examination of AAPCC's annual reports from 1995 to 1999 of cases listing acetaminophen as the primary (first) agent showed acetaminophen to be the leading cause of poisonings. In 1999, acetaminophenrelated calls represented 10 percent of all calls to AAPCC. There was a decrease in calls between 1995 (111,175) and 1999 (108,102). In 1999, nearly 50 percent of the poison victims associated with the calls received treatment in health care facilities. Two percent of these victims were reported to have developed major effects resulting from the poisoning, i.e., the signs or symptoms occurring as a result of acetaminophen exposure were lifethreatening or resulted in significant residual disability. Fifty percent of the calls involved children and adolescents (19 years of age or under). Of the acetaminophen related calls regarding children under 6 years of age (approximately 40,000 calls), 22 percent occurred in children who ingested adult formulations of acetaminophen.

In 1995, there were at least 76 acetaminophenrelated fatalities. By 1999, the number of acetaminophenrelated fatalities increased to 141. Of these, 92 (65 percent) were a result of suicidal intent, 43 (30 percent) were unintentional, and the dosing intent for 6 (4 percent) was undetermined. Among the 43 unintentional fatalities, 28 (65 percent) took one OTC drug product containing only acetaminophen; 4 (9 percent) took one prescription product containing acetaminophen, and 11 (26 percent) took more than one acetaminophen product simultaneously. These AAPCC data may underreport the actual number of acetaminophen toxicity cases, because serious cases that go directly to emergency departments, and chronic users of acetaminophen, are unlikely to generate poison control center contacts.

FDA staff reviewed spontaneous reports of hepatoxicity in FDA's adverse event reporting system (AERS). U.S. cases were identified that had been received by FDA between January 1998 and July 2001 and in which one or more acetaminophen containing products had been ingested. Of 633 reports, 43 were duplicates. Another 283 were excluded for various reasons, primarily to exclude cases in which there was apparent suicidal intent. A total of 307 cases were included in FDA's analysis (25 pediatric and 282 adult cases).

Pediatric cases (of children age 1 day to 8 years) consisted primarily of males (approximately 70 percent), although gender was not reported in each case. Fifteen of the 25 pediatric cases involved severe, lifethreatening liver injury. Of the 25 children, 10 died, 21 were hospitalized, and 2 required only treatment in an emergency department. The dose was estimated, based upon reported daily doses and weight, in 10 cases to be 106 to 375 milligrams/kilogram (mg/kg) per day. The recommended pediatric dose is 75 mg/kg/day (Ref. 7). Twenty two of the children (88 percent) took only 1 product containing acetaminophen and 3 children (12 percent) took 2 or more products containing acetaminophen. Sixteen of the cases (53 percent) reported ingestion of a single ingredient acetaminophen product (APAP), 12 cases (40 percent) reported ingestion of an ``unspecified APAP product'' and the remainder of the cases reported ingestion of combination products. Of the single ingredient products, concentrated drops containing acetaminophen 100 mg/milliliter (mL) were reportedly ingested in seven cases.

In 20 of the pediatric cases, 1 or more medication errors were reported. In three cases, the wrong product was used, i.e., the concentrated drops instead of the children's acetaminophen liquid formulation. In four cases, incorrect measuring devices were used, i.e., teaspoonfuls instead of dropperfuls. Five cases reported instances of misinterpretation of labeled dosing guidelines or misinterpretation of instructions provided by a health care provider.

Sixty percent of the 282 adult cases (15 to 85 years old) were female and 229 required hospitalization. A total of 169 adults experienced severe, lifethreatening liver injury; 124 of these patients died and 7 required a liver transplant. One hundred ninety nine (71 percent) adults reported using an acetaminophen product for a therapeutic indication, primarily analgesia. In 74 (26 percent) cases, the indication for use was unknown, and in 9 (3 percent) cases, abuse of a narcoticacetaminophen prescription product was reported. One hundred thirtyeight (38 percent) cases listed an unspecified acetaminophen product (unknown whether single ingredient or combination product and whether OTC or prescription), 122 (33 percent) cases involved the use of a narcoticacetaminophen prescription product, and 76 (21 percent) cases reported use of an OTC single ingredient acetaminophen product. Approximately 25 percent of all adult cases reported use of more than one acetaminophen product. When more than one acetaminophen product was reported, a narcoticacetaminophen prescription product in combination with an OTC product containing acetaminophen was used more often than any other
[[Page 77321]]
combination of acetaminophen products. These cases also used higher doses than people who took only one acetaminophencontaining product.

Dosing amounts were reported in 132 of the 282 adult cases. The mean and median daily dose were 6.5 and 5 g, respectively, but ranged from 650 mg to 30 g/day. Where the dosage strength was known, 500 mg acetaminophen was reported most often. If a dose range was reported in the case, the midpoint was used in the analysis. If the strength was unknown, a 500mg strength was assumed. Dosing in the 65 adults with severe liver injury from this group showed a mean and median daily dose of 7.1 and 6.23 g, respectively. Twentythree of the 65 cases with severe liver injury reported doses of less than 4 g/day. People who used more than one acetaminophen product reported taking higher doses than people who took a single product. Qualitative dosing information was provided for an additional 43 (15 percent) cases with terms such as ``excessive doses'' or ``recommended doses.'' Two out of three of these cases suggest that greater than recommended doses were used.

Alcohol use was reported in 116 of the adult cases and the content of the reports was highly variable. Alcohol use in these cases was defined by FDA as alcoholism or alcohol abuse in 64 cases; regular, daily, or moderate use in 23 cases; occasional use in 10 cases; previous use in 6 cases; and 13 cases did not provide a description. Eightysix (74 percent) of the 116 alcohol users developed severe liver injury. For those cases with acetaminophen dose information, the mean dose associated with toxicity was lower for alcohol users compared to nonalcohol users (table 7).
Table 7.Acetaminophen Dose and Alcohol Use Category of Liver Disease (Developed Post
Acetaminophen) Alcohol Users (Mean Dose) NonUsers (Mean Dose) All (N=132) 5.6 g (N=53) 6.9 g (N=79) Severe only (N=65) 6.0 g (N=38) 8.6 g (N=27)

A history of prior liver disease, or possible underlying liver disease, was reported in 70 cases. In at least 20 of these cases, the preexisting liver disease was reportedly due to alcohol. Twentythree people reported a history of, or possible, viral hepatitis. Among the 70 cases with preexisting liver disease, 49 percent (70 percent) developed severe liver injury. Table 8 shows the dose that was associated with liver injury for cases with and without preexisting liver disease. The first row includes all cases (all degrees of acute liver injury) that reported dosing information. The second row shows a dose comparison in people who experienced severe liver injury after acetaminophen exposure.
Table 8.Acetaminophen Dose and Liver Disease Category of Liver Injury Associated With Cases With Preexisting Liver Cases With No Preexisting Liver Acetaminophen Dosing Disease (Mean Dose) Disease (Mean Dose) All (N=132) 5.4 g (N=36) 6.8 g (N=96) Severe only (N=65) 5.7 g (N=23) 7.8 g (N=42)

Some additional factors may have contributed to the development of hepatotoxicity in these adults. Use of other medications that may have contributed to hepatotoxicity was reported in 93 cases, including 63 cases that involved products that are labeled with warnings about potential hepatotoxicity. A small number of reports also mentioned the existence of concomitant malnutrition or decreased oral intake.

FDA noted that there are limitations to interpreting the AERS data. Dosing information may be unreliable. Acetaminophen products are generally taken on an asneeded basis, so the actual dose ingested can be difficult to ascertain. There is no certainty that all of the adult cases included in this analysis were unintentional. Stigma may be associated with reporting suicide, so cases may be reported as unintentional when they were intentional overdoses. In addition, spontaneous reporting systems cannot provide certainty that acetaminophen was the cause of any of the reported adverse event. Furthermore, incidence rates cannot be determined, because the numerator or denominator descriptors for the entire population are not available. Overall, spontaneous reports may be subject to significant underreporting.

The AERS cases strongly suggest that particular circumstances were likely to have led to hepatotoxicity. Some examples of those circumstances follow:

FDA presented NDAC with several questions that remained unaddressed by FDA's review:

It is clear that unintentional acetaminophen doses are associated with a large number of emergency department and hospital admissions and are related to an estimated 100 deaths each year. Using a number of data sources, analyses have shown that circumstances leading to [[Page 77322]]
acetaminophen hepatotoxicity are multifactorial. FDA asked the committee to consider the contribution of each of the following in producing unintentional overdose toxicity:

Several drug manufacturers and other interested parties provided additional comment (Ref. 4):

1. The precise incidence of harmful, unintentional overuse cannot be accurately determined from the current databases. Fortyeight million American adults use products containing acetaminophen in any single week; thus, harm is rare and is caused by inadvertent overdose.

2. There are limitations to the AERS data set for assessing hepatic events. Patients consistently underestimate the dose taken, and suicide attempts are often not recorded in patients who are found unconscious or intoxicated. The AERS reports found to be definitely associated with acetaminophen involved substantial overdose in individuals with self abusive behaviors (e.g., alcohol abuse, bulimia). Causality cannot be ascertained using retrospective data, especially case reports, because the dose history is often inaccurate.

3. Formulations most commonly reported were OTC singleingredient and prescription combination acetaminophen products. OTC acetaminophen combination products were rarely reported.

4. Serious hepatotoxicity occurs following substantial overdose (a single dose of approximately 15 g or use of approximately 12 g for multiple days).

5. FDA focused on unintentional misuse. The manufacturer noted they had implemented labeling changes to minimize the inadvertent overuse of analgesics. The manufacturer recommended an organ specific overdose warning.

1. In overdose situations, in any given year, the number of deaths for acetaminophen reported by the AAPCC is approximately 20 times that for ibuprofen.

2 . Unintentional overdose of acetaminophen can put consumers in a lifethreatening situation due to the delayed onset of clinical symptoms of toxicity.

3. Advertising portrays acetaminophen as a totally safe ingredient. This portrayal may exacerbate use and contribute to the silent danger resulting from overdose.

1. The median and average doses were between 6 and 8 g/day. These values are above the recommended maximum daily dose (4 g/day), but below the 10 to 15 g dosage usually considered to be toxic. There were three cases of intentional overdose and three cases of unintentional overdose involving prescription acetaminophen products. OTC products were associated with 20 intentional and 21 unintentional overdoses. More patients in the unintentional overdose group used single ingredient acetaminophen (i.e., not a combination product). The primary reason reported for exceeding the maximum dose was to treat unrelieved pain. Many patients stated that they knew they were exceeding the recommended dose and did so because they thought it was a safe drug. Thirty percent of the patients used the drug over a period of greater than 7 days.

2. The unintentional overdose group experienced greater morbidity and mortality than the intentional overdose group. The peak acetaminophen levels in the intentional overdose group were much lower compared to the unintentional overdose group (27.8 versus 115.1 mg/L). The unintentional overdose group had much higher peak levels of Alanine aminotransferase (ALT) (5,193 versus 3,065 units/L), Aspartate aminotransferase (AST) (6,819 versus 2,742 units/L), International Normalized Ratio (INR) (4 versus 2.5), and total bilirubin (5.87 versus 1.87 mg/dL). Patient outcomes were generally worse in the unintentional overdose group, in which more patients failed to have resolution of the liver problems from the overdose (31 versus 4 percent). More patients were evaluated for transplants (11 versus 9), received transplants (2 versus 0), and died (3 versus 0) as a result of unintentional overdoses.

3. Compared to the intentional overdose group, the unintentional overdose group was more likely to have one or more of the following risk factors for acetaminophen toxicity: (1) Hepatic disease, (2) acute or chronic alcohol use, (3) drug abuse, or (4) concomitant disease. Ninetysix percent of cases in the unintentional overdose group had one or more of these risk factors, as compared to 70 percent in the intentional group. Acute and chronic alcohol use was present in 87 percent of unintentional overdose cases, as compared to 61 percent of the intentional overdose cases. Thus, the existence of risk factors may have an impact on toxicity in unintentional ingestions.

Several drug manufacturers and others submitted additional information for the committee to review (Ref. 5):

1. AERS serves as a signal generating system for rare, unexpected adverse events in marketed products. It cannot be used to determine event rates, dose ingested, or patient dosing intent.

2. FDA's review of the AERS data set was intended to exclude obvious suicide, usually associated with very large drug ingestion. Thus, the reported dosage (which could only be estimated in 48 percent of the reports in the data set) is skewed significantly toward labeled directions for use, so cases may falsely appear to be consistent with inadvertent misuse.

3. The selective data in FDA's AERS review cannot be used to determine an acetaminophen toxicity threshold associated with any patient condition (i.e., concomitant drug, alcohol history, or pre existing concomitant disease).

4. The quality of the 281 adult reports in AERS was evaluated by the manufacturer. The manufacturer concluded that 168 reports (24 percent) contained insufficient information to estimate the dose taken and 212 reports (88 percent) contained no liver pathology information. AST and ALT levels were not reported in 108 cases (38 percent). Only 61 reports (25 percent) had information about viral hepatitis testing and, of these, 29 reports were positive for hepatitis A, B, or C.

5. There are flaws in the derivation of FDA's theory that alcohol use, underlying/history of liver disease, and potentially the use of hepatotoxic concomitant medications, may increase susceptibility to acetaminophenassociated hepatotoxicity at unexpectedly low doses of acetaminophen. The manufacturer provided arguments that the existence of any of these factors in a case report may each inherently interfere, for various reasons, with establishing the correct assessment of a hepatotoxic dose of acetaminophen.

Based on an assessment of several databases, a sponsor calculated that the worst case scenario of deaths from acetaminophen overdose is estimated to be 213 deaths per year (Ref. 5, Tab A).

1. The majority of hepatotoxicity cases (65 percent of cases in the year 2000) involved use of one acetaminophencontaining analgesic product.

2. Acetaminophencontaining cough/cold medications were not a significant contributor to the total number of reports of acetaminophen associated hepatotoxicity (2 percent of cases in the year 2000).

3. Only 1 percent of the reported cases of hepatotoxicity in 2000 involved use of an OTC acetaminophencontaining cough/cold product concomitantly with other acetaminophencontaining product(s).

NDAC unanimously agreed that the evidence of risk associated with unintentional overdose of acetaminophen warrants FDA labeling changes, without awaiting the outcome of further studies. NDAC noted the following four major areas of concern:

1. Unintentional use of multiple acetaminophen containing products

2. Exceeding the recommended dose without recognizing the consequences

3. Improper dosing of infants

4. The unknown consequences of use in special populations, such as alcohol abusers.

NDAC recommended that the minimum requirements for change should include, for all products containing acetaminophen (including those available by prescription), the addition of distinctive labeling (highlighted or bold type) on the front panel or PDP to state that the products contain acetaminophen. FDA noted that the nonproprietary name of prescription drugs must appear in labeling in letters at least half the size of the brand name (see 21 CFR 201.10(g)(2)). NDAC recommended that a similar provision also be applied to OTC drug products containing acetaminophen, such as a standard to ensure prominence of important information. NDAC stated that consumers need to be informed that combining products containing acetaminophen can result in exceeding the recommended dose.

NDAC commented that there are insufficient data in the OTC setting on risk management, understanding consumer behavior, and the effectiveness of warnings on labels. This lack of data makes it difficult to determine which factors contribute to liver injury. Although these factors are not clearly understood, NDAC concluded that labeling revisions are needed to help minimize any risks.

NDAC did not propose exact language. It was believed that it was important that the message not refer to ``overdose,'' but rather to a statement such as ``do not take more'' or ``do not exceed the recommended dose.'' NDAC believed that the term ``overdose'' would not be understood to be pertinent to consumers whose intent was to use the product safely. One NDAC member stated the term ``exceed'' is not part of consumers' common vocabulary and proposed that it would be more useful to inform consumers of a specific allowable total dose (e.g., not to take more than a specified number of tablets in a given period).

NDAC reexamined the currently required alcohol warning for acetaminophen, which states: ``Alcohol Warning: If you consume 3 or more alcoholic drinks every day, ask your doctor whether you should take acetaminophen or other pain relievers/fever reducers. Acetaminophen may cause liver damage.'' NDAC inquired why ``three drinks'' were used in the alcohol warning. FDA responded that the number is from recommendations of the American Heart Association as to what constitutes excessive alcohol use. FDA stated that it recognized this may seem arbitrary and asked NDAC to provide further
recommendations. NDAC questioned whether doctors are wellinformed with proper information about the relationship between alcohol and acetaminophen use and whether educational efforts should also include educational efforts directed at health care professionals and consumers. NDAC was concerned about the lack of available data on which to base such advice, noting that there is a lack of information about how to determine the amount of alcohol that may be harmful to any individual. NDAC noted that reducing the risk of drug adverse events is the goal, but believed that more data are essential for them to make specific recommendations.

FDA asked NDAC to comment on whether the current maximum allowable daily dose of acetaminophen should be used by individuals consuming three or more drinks per day. One NDAC member agreed that was prudent to lower the dose, however, the majority of NDAC members believed that more information is needed before dose reductions could be implemented for this population. NDAC stated that, intuitively, a lower dose would decrease potential toxicity, but noted that there is a lack of information to support such labeling.

One NDAC member mentioned that although some evidence appeared to show an association of increased acetaminophen toxicity for patients with preexisting liver disease, this finding is contrary to hepatologists' experience with acetaminophen. Generally, acetaminophen is considered safe for use in patients with liver disease, including people awaiting liver transplantation. Most hepatologists recommend acetaminophen for such patients, but at reduced doses, such as 2 g maximum in a 24hour period. NDAC urged more studies, not only of risk factors, but of a plan to reduce risk.

NDAC stated that many physicians and pharmacists may not be aware of the risks of unintentional overdose. NDAC added that, along with consumer education, professional programs are important, because prescription products containing acetaminophen are widely used. Education of pharmacists would be needed to support the use of additional labeling information (stickon labels, etc.) attached to prescription containers. NDAC stated that auxiliary labeling is critical to conveying information that the prescription product contains acetaminophen.

C. Aspirin and Other NSAIDs

On the second day of the meeting (September 20, 2002), NDAC considered safety issues related to the use of aspirin and other OTC NSAIDs. The primary areas for discussion included the potential for GI bleeding and renal toxicity from using these drugs. The prescription labeling for NSAIDs and the professional labeling for aspirin have warnings for GI bleeding and possible renal toxicity. Aside from the alcohol warning required on all OTC NSAID drug products, current OTC labeling does not have warnings about damage to specific organs. 1. Points for Discussion

FDA asked NDAC to consider the relative risks for GI bleeding and renal toxicity associated with OTC doses of NSAIDs, including aspirin, and to consider the following issues:

GI bleeding

FDA staff described cases of GI bleeding (spontaneous reports from AERS received by FDA between 1998 and 2001) in individuals who used OTC NSAIDS (including aspirin) as an analgesic and/or antipyretic (Ref. 8). The review was limited to cases that mentioned ``OTC'' in the narrative of the report. Any cases that appeared to involve prescription NSAID products were excluded. A total of 279 cases of
[[Page 77325]]
GI bleeding were included: 82 for aspirin and 197 for nonaspirin NSAIDs (i.e., ibuprofen, ketoprofen, and naproxen). The mean age was 59 years (ranging from 1 to 99 years). There were 138 (49.5 percent) males, 119 (42.7 percent) females, and 22 cases (7.9 percent) in which gender was not reported.

Cases that specified the location in the GI tract of the bleed included: Stomach (63 cases), duodenum (35 cases), unspecified upper GI site (15 cases), esophagus (13 cases), and rectum/colon/small intestine (9 cases). For nonaspirin NSAIDs, the median time to onset was 7 days. Time to onset was defined as the time between each person's first use of the drug and the time that bleeding occurred. For aspirin, time to onset was about 30 days. For both aspirin and nonaspirin NSAIDs, there was a wide range in time to onset. FDA reviewed the cases for common risk factors for GI

FOR FURTHER INFORMATION CONTACT

Marina Chang, Center for Drug Evaluation and Research, Food and Drug Administration, 10903 New Hampshire Ave., Silver Spring, MD, 209930002, 3017962090.
[[Page 77315]]