and Greenblatt reported that adverse reactions to DPX in hospitalized patients were infrenquent and mild. The adverse reactions, although qualitatively similar, occurred less often than with codeine and other analgesics used in hospitalized patients. Standard tolerance studies in volunteers revealed no significant differences between DPX and placebo (Ref. 14). In contrast, Goodman and Gilman state that in doses equianalgesic to codeine it is likely that the incidence of side effects would be similar to those of codeine (Ref. 15).

Reports of deaths in connection with DPX use have frequently relied upon statistics received from the Drug Abuse Warning Network (DAWN) system. This system, from which data are cited in the HRG petition, is a large-scale data-collecting system, initiated in September of 1972 and operated for the Federal Government on contract by IMS America, Ambler, PA DAWN collects data from over twenty large metropolitan areas in the continental United States and tabulates them as the number of "mentions" of a drug after persons have been in contact with or treated by one or three types of facilities: emergency rooms in non-Federal short-term general hospitals (as defined by the American Hospital Association), offices of medical examiners or coroners, and crisis intervention centers. An "episode" is either a drug-related death or a drug-related visit to an emergency room, and a "mention" is the report of a drug associated with an episode. If three drugs were reported for one episode, for example, three drug mentions would be recorded. Certain analytical problems may arise because of faetors such as the lack of precision in reporting (e.g. the names of the drugs involved may be given to an emergency room in jargon that makes it impossible to assign the mention precisely to a particular drug or drugs) and the limitations in the system itself (e.g. the number and characteristics of the facilities reporting to the DAWN system have not remained constant). Despite these problems. DAWN data are regarded as useful in identifying trends or indicating the development of drug problems. Although the data are not measures of the absolute size of a drug problem, they illuminate aspects of the nature of such a problem, and are helpful in making comparisons among drugs. The DAWN data which follow include only mentions from emergency rooms and medical examiners or coroners, excluding crisis intervention center reports. Although for many analyses it is appropriate to limit the data for a given period to that reviewed from consistent reporters, that was not done in this case because of the importance of not omitting any useful information.

Table 2 compares DAWN data on coroners' reports of deaths (associated with DPX alone or in conjunction with other factors) with data on emergency room visits. Although there is a slight increase in deaths in 1977 compared with the previous 3 years, this difference is of questionable significance. In most instances, other substances (e.g. tranquilizers) are also implicated in the deaths.

TABLE 2.-CORONERS' REPORTS AND EMERGENCY ROOM VISITS IN WHICH PROPOXYPHENE (DPX) IS MENTIONED1

Comparisons on safety of DPX and other drugs are shown in Tables 3 and 4. Not only are total DAWN mentions (coroner and emergency room) for the drugs provided, but also comparisons indicating the ratios of DPX-associated deaths to prescriptions dispensed. The data indicate that DPX is the most frequently mentioned single drug on coroner's reports. However, the ratio of DPX-associated deaths (coroners' mentions) to dispensed prescriptions is lower than that for the barbiturates, ethchlorvynol, glutethimide, methaqualone, amitriptycline. doxepin, and pentazocine, as shown in Table 3. When comparisons are made according to drug groupings, as in Table 4, the propoxyphene ratio is considerably lower than that for three other drug groups ("barbiturates," "other sedative/hypnotics," and "antidepressants").

TABLE 3.-COMPARISON OF PROPOXYPHENE WITH OTHER DRUGS; ASSOCIATIONS WITH EMERGENCY ROOM (ER) MENTIONS AND CORONER MENTIONS (DEATHS), 19771

TABLE 4.-COMPARISON OF PROPOXYPHENE WITH OTHER DRUG GROUPINGS; ASSOCIATIONS WITH DEATHS, 1977 1

1 Source: DAWN and NPA data.

2 Phenobarbital and meprobamate were intentionally excluded since their predominant use, as anticonvulsant and "muscle relaxant," respectively, differs from other drugs in the same pharmacologic category.

The circumstances under which the DPX-related deaths occurred are a matter of special interest. Particularly relevant are considerations such as whether other drugs or alcohol were also involved, whether an overdose of DPX was taken, and to what extent the deaths were intentional. Although it is impossible to determine precisely the answers to such questions, some generalizations can be made from available data.

1. DPX is a common cause of drug-associated death. These cases involve both suicide and accidents, but a majority of the deaths appear to be intentional. Thus, a tabulation of the 72 DPX-related deaths reported in 1971-1975 by the San Francisco Coroner's Office indicates that 58 percent of them were suicides (this compares with 10 codeine-related deaths, of which 50 percent were suicides). Analysis of data available from different sources, as shown in Table 5, supports the hypothesis that a substantial proportion of DPX deaths are the result of use by those in younger age groups, for suicidal purposes or associated with abuse. Thus, 8-22 percent of the deaths are in the 10-19 age group, which accounts for only 7 percent of the prescriptions; 48-58 percent of the deaths are in the 20-39 age group with approximately 30 percent of the DPX prescriptions. Regardless of age considerations, however, it is apparent that DPX is one of the prescription drugs most frequently associated with suicide and accidental deaths, ranking behind only the barbiturates as a group in total number and behind only barbiturates, other sedative-hypnotics and antidepressants in deaths per million prescriptions dispensed (Table 4).

TABLE 5.-PROPOXYPHENE: REPORTED PRESCRIBING, EMERGENCY ROOM VISITS, AND ASSOCIATED DEATHS. BY AGE

5 Drug Enforcement Administration, Drug Abuse Warning Network, January 1975 to August 1978. This figure is for the age group 50 and over.

FDA spontaneous adverse reaction reporting program. (In 15 percent of the reports age was not reported.) 8 Reference 17.

2. A majority of the DPX-related deaths appear to have occurred when DPX was taken in conjunction with alcohol or other drugs. Thus, information from various sources, shown in table 6, indicates that in about 12-28 percent of the deaths, DPX alone was involved; in the others alcohol and/or other drugs were also present.

3. At present there is no clear evidence of deaths attributed to DPX products alone when taken in recommended doses and without alcohol or tranquilizers also being involved. There are, however, several "accidental" deaths that have occurred apparently as a result of the consumption of DPX in quantities only slightly in excess of recommended therapeutic dosage, usually combined with alcohol or tranquilizers or both. Dr. Larry Lewman, Multnomah County (Oregon) Medical Examiner, in testimony before the Senate Subcommittee cited previously in this notice, presented data in support of this possibility (Ref. 11). While reports such as this are very infrequent, given the wide availability of DPX, they raise concern that death of persons taking the drug at or near the recommended doses may be more common than is currently appreciated.

4. The mechanism of death in cases of DPX overdose is commonly attributed to respiratory depression, a typical action of narcotics. This theory is substantiated by a large number of case reports from a wide variety of sources. However, the possibility of a specific and primary cardiotoxic effect, independent of respiratory depression has been raised. The demonstration of dose-related progressive condition block appears clear in experimental animals, and in some patients with acutely toxic overdoses there are reported electrocardiographic (ECG) changes. This is not unexpected in view of the local anesthetic activity of both DPX and its primary metabolite norpropoxyphene. It has been postulated that, with chronic dosing, DPX accumulates to near toxic levels and adversely affects myocardial conduction, but this has not been the experience in heroin addicts on long-term, high-dose DPX napsylate maintenance. Moreover, when there are ECG changes in DPX overdoses and the CNS depressant effects are reversed by naloxone, the ECG changes rapidly revert to normal when respiration returns (or is mechanically supported) and acidosis is corrected. Therefore; the cardiac changes are most likely secondary to hypoxia rather than norpropoxyphene toxicity, which would take at least several hours to be reversible. Moreover, as shown in Table 5, only a small percentage of the deaths are in the over 60 age group which accounts for 35 percent of the reported prescribing. This population would be presumably more sensitive to any cardiovascular toxicity associated with DPX, but the paucity of deaths in this age group is notable. Cardiotoxicity at a therapeutic dose has not been observed.

5. DPX can produce psychological and physical dependence of the opiate type when taken for an extended period of time. It will substitute for other opiates in addicted persons, but only to a limited extent. Because of the abuse potential of DPX, it was placed in Schedule IV of the Controlled Substances Act. The

Health Research Group believes the restrictions of Schedule IV are not sufficient to protect the public from the dangers of DPX use and has proposed it be transferred to the most restricted control, Schedule II.

REFERENCES

The following items specifically cited in this notice, as well as a number of other items related to the DPX hearing, are on file and available for inspection in the office of the Hearing Clerk, at the address specified at the beginning of this notice.

1. l'etition from Health Research Group, Washington, D.C., to Secretary J. Califano, November 21, 1978.

2. National Academy of Sciences/National Research Council; Report of Panel on Drugs for the Relief of Pain on Darvon Compound, NDA 10–996.

3. Beaver, W. T., "Mild Analgesics, A Review of Their Clinical Pharmacology (Part II)," American Journal of Medical Science, 25:576-599, 1966.

4. Miller, R. R., A. Feingold and J. Paxinos, "Propoxyphene Hydrochloride," Journal of American Medicine, 213(6):996-1006, 1970.

5. Beaver, W. T., Memorandum to Henry E. Simmons, Director, Bureau of Drugs, Food and Drug Administration, May 18, 1971.

6. Moertel, C. G., D. L. Ahmann, W. F. Taylor, and N. Schwartau, "A Comparative Evaluation of Marketed Analgesic Drugs," The New England Journal of Medicine, 286:813-15, 1972.

7. Hopkinson, J. H., IV, et al., "Acetaminophen Versus Propoxyphene Hydrochloride for Relief of Pain in Episiotomy Patients," The Journal of Clinical Pharmacology, 113:251-263, 1973.

8. Gruber, C. M., Jr., "Codeine and Propoxyphene in Postepisiotomy Pain," The Journal of the American Medical Association, 237: 2734–35, 1977.

9. Sunshine, A. J. Slafta and C. Gruber, Jr., "A Comparative Analgesic Study of Propoxyphene, Fenoprogen, the Combination of Propoxyphene and Fenoprofen, Aspirin, and Placebo," The Journal of Clinical Pharamacology, 18:556–563, 1978.

10. Miller, R. R., "Propoxyphere, A Review," American Journal of Hospital Pharmacy, 34:413-423, 1977.

11. Propoxyphene Hearings (January 31, February 1 and 5, 1979) of the Monopoly and Anticompetitive Activities Subcommittee of the Select Committee on Small Business, U.S. Senate.

12. Moertel, C. G., D. L. Ahmann, W. F. Taylor, and N. Schwartau, "Relief of Pain by Oral Medications," The Journal of the American Medical Association, 229(1):55-59, 1974.

13. Bauer, R. O., A. Baptisti, Jr., and C. M. Gruber, Jr., “Evaluation of Propoxyphene Napsylate Compound in Postpartum Uterine Cramping," The Journal of Medicine, 5:317-328, 1974.

14. Miller, R. and D. J. Greenblatt, "Drug Effects in Hospitalized Patients: Experiences of the Boston Collaborative Drug Surveillance Program, 1966-1975," John Wiley & Sons, New York, pp. 162–164, 1975.

15. Jaffe, J. H. and W. R. Martin, "Narcotic Analgesic and Antagonists," in "The l'harmacological Basis of Therapeutics," 5th Ed., Edited by Goodman, L. S., and A. Gilman, MacMillan Publishing Co., Inc., New York, pp. 270-271, 1975.

16. Baselt, R. C. and J. A. Wright, J. E. Turner, and R. H. Cravey, "Propoxyphene and Norpropoxyphene Tissue Concentrations in Fatalities Associated with Propoxyhene Hydrochloride and Propoxyphene Napsylate," Archives of Toricology, 34:145-152, 1975.

17. Finkel, B. S., K. L. McCloskey, G. F. Kiplinger, and I. F. Bennett, "A National Assessment of Propoxyphene in Postmortem Medicolegal Investigation, 1972-1975," Journal of Forensic Sciences, 21(4):706–742, 1976.

18. Hine, C. H., J. A. Wright, D. J. Allison, B. G. Stephens, and A. Pasi, Analysis of Fatalities due to Acute Narcotism in a Major Urban Area (umpub.).

PUBLIC HEARINGS

The Food and Drug Administration announces that a public hearing will be held to obtain additional information and recommendations relevant to consideration of further regulatory actions on DPX-containing drug products. The