Elixirs, Diluents, and the Passage of the 1938 Federal Food, Drug and Cosmetic Act

The Development of Elixir Sulfanilamide

The Elixir Sulfanilamide tragedy occurred at an epochal time in the history of medicine and in the development of the pharmaceutical industry. In 1932, Gerhard Domagk made the important discovery that a chemical, prontosil, protected against certain types of bacterial infections in mice [4]. Subsequent investigations showed that prontosil was metabolized to the active agent para-aminobenzenesulfonamide, also known as sulfanilamide [5]. In January 1937, after careful animal and clinical studies, Long and Bliss [6] endorsed the use of sulfanilamide for streptococcal infections. Because sulfanilamide was initially synthesized in 1908 and was therefore not subject to patent restrictions, several pharmaceutical companies such as Squibb, Merck, Winthrop, Eli Lilly, and Parke-Davis rushed to develop this new “wonder” drug [7]. Each preparation received the endorsement of the American Medical Association (AMA) Council on Pharmacy and Chemistry, an organization that reviewed new drugs on a voluntary basis.

About this time, the S.E. Massengill Company of Bristol, Tennessee, founded by Samuel E. Massengill (1871-1946), also began to produce sulfanilamide. This relatively small company initially manufactured capsules and tablets, as did the other drug companies. Soon thereafter, Massengill's detail men recognized that a demand existed for a liquid preparation. A solution of sulfanilamide had not been previously available because of the lack of an appropriate vehicle. Harold Watkins, Massengill's chief chemist, was assigned the task of finding the right diluent [8].

After investigating several possible solvents, Watkins formulated a raspberry-tasting pink preparation consisting of 10% sulfanilamide, 72% diethylene glycol, 16% water, and small amounts of elixir flavor, raspberry extract, saccharin solution, amaranth, and caramel [8]. The concoction was called Elixir Sulfanilamide despite the lack of ethanol, an ingredient that was required for a preparation to receive the elixir designation. The Massengill “control laboratory” evaluated the drug for appearance, flavor, and flagrance; found it acceptable; and approved it for distribution. The toxicity of the individual ingredients or the finished product was never tested [8]. In later testimony, a Food and Drug Administration (FDA) agent described the Massengill Company's new drug development strategy as throwing “drugs together, and if they don't explode they are placed on sale” [9].

By early September 1937, 240 gallons of Elixir Sulfanilamide had been manufactured by the Massengill Company. A total of 1304 shipments were distributed across the United States [8]. The bottles were labeled “Elixir Sulfanilamide: Each fluid ounce represents sulfanilamide, 40 grains” [10] (Figure 1). The presence of diethylene glycol was not divulged. One of the major points of delivery of the drug was Tulsa, Oklahoma.

On 2 October 1937, an editorial appeared in JAMA warning against an overzealous embrace of sulfanilamide [11]. It stated that although “seldom has a new drug introduced in medical practice aroused the enthusiasm that has developed for sulfanilamide,” clinicians needed to recognize the increasing number of adverse reactions associated with its use. These reactions included dermatitis, photosensitivity, granulocytopenia, and hemolytic anemia. Unbeknownst to the journal's publisher, the first deaths from the use of sulfanilamide had already occurred.

Within days of the publication of the editorial, James Stephenson, the president of the Tulsa County Medical Society, was notified that six local patients had unexpectedly died from renal failure after ingesting Elixir Sulfanilamide. In a telegraph to the AMA, Dr. Stephenson requested the composition of the elixir. The AMA responded that they were unaware of any product from the Massengill Company and had never approved a liquid sulfanilamide preparation [10].

The AMA then sent a telegraph to Dr. Massengill, requesting the composition of the elixir. Massengill released this proprietary information but urged that it be kept strictly confidential. He hypothesized that the deaths may have been caused by mixing the elixir with other drugs. Massengill and Watkins reluctantly admitted, however, that toxicity tests had not been done. To show confidence in his product, Watkins self-administered small amounts of diethylene glycol and elixir. No adverse effects were noted [3].

Nonetheless, on 20 October 1937, the AMA received the following telegram from Dr. Massengill: “Please wire collect by Western Union suggestion for antidote and treatment following use of Elixir Sulfanilamide.” The AMA sent a terse reply: “Antidote for Elixir Sulfanilamide-Massengill not known. Treatment presumably symptomatic” [12]. By this time the Massengill Company had sent out 1100 telegrams requesting the return of the elixir.

Almost immediately after this evolving catastrophe was recognized, researchers began to search for the etiologic agent. Analysis of the contents of an elixir bottle used by one of the Tulsa victims showed that it contained 72% diethylene glycol, 10% sulfanilamide, 15% water, and small amounts of saccharin and coloring—the exact ingredients stated by the manufacturer [13]. Contamination with lead, bismuth, mercury, arsenic, or other impurities was not found. Although the list of adverse reactions to sulfanilamide tablets was still growing at the time, nephrotoxicity had not been reported. The investigation therefore focused on the diluent—diethylene glycol.

Diethylene Glycol

Diethylene glycol Figure 2 is a condensation product of ethylene glycol production. Although it was discovered in 1869, commercial production did not begin until 1928 [14]. It proved to be an excellent solvent and was used as a glycerine substitute, as a moistening agent, and in the production of resins and explosives. Its use in food products, however, was not permitted because of the paucity of scientific data proving it safe for oral administration [15].

Before the elixir disaster, only two studies had been done on the toxicity of diethylene glycol. In 1931, Von Oettingen and Jirouch [16] determined that the minimum lethal dose of diethylene glycol in mice was 5 mL/kg body weight of a 50% solution given subcutaneously. Histologic analysis showed marked hydropic degeneration of the kidney. In early 1937, Haag and Ambrose [17] reported that ingestion of a 3% diethylene glycol solution was rapidly fatal in a rat model.

Although the clinical effects of diethylene glycol exposure were still unknown at this time, deaths from ethylene glycol had first been reported in 1930 [18]. Moreover, in 1934, an article appeared about an industrial accident involving five factory workers who had been using the solvent diethylene dioxide Figure 2, an anhydride derivative of diethylene glycol (known under the trade name Dioxan) [19]. Each victim died after renal failure developed. Postmortem examinations showed extensive renal and hepatic injuries.

Enter the Food and Drug Administration

By mid-October 1937, news of the Tulsa deaths reached the offices of the FDA [8]. Walter Campbell, the chief of the FDA, immediately ordered a massive investigation and assigned almost all of the FDA's 239 inspectors and chemists to the case [9].

The legal grounds for action on the part of the FDA were limited because of the lack of strong regulatory guidelines. The 1906 Pure Food and Drugs Act was the only previously enacted federal legislation that sought to protect the public from drugs and food potentially threatening to human health. This landmark act was passed during the progressive era of President Theodore Roosevelt at the urging of Dr. Harvey W. Wiley, the chief chemist at the Department of Agriculture [20]. Since the early 1880s, Wiley had unrelentingly investigated the problems of impure and contaminated food. In 1902, he organized a group of volunteers known as the “poison squad,” who did self-experiments on food preservatives [21]. Although Wiley's “poison squad” began to stir public interest in reform, it was the investigative reports of the muckrakers that fueled the passage of the much-needed legislation. In his expose in Collier's about the patent (secret) medicine industry, Samuel Hopkins Adams revealed many unsavory practices, such as the adding of opiates to soothing syrups for infants. The 1906 publication of Upton Sinclair's novel The Jungle exposed the unhygienic practices of the meat-packing industry, an event that ignited the public's furor more than anything else [22].

The 1906 statute was primarily concerned with preventing the distribution of adulterated food. Drug regulations, which were written to protect the public against misbranded patent medications, were of secondary importance. False claims in drug advertisement were not addressed, and dangerous drugs were not restricted. The 1906 regulations were enforced by the Bureau of Chemistry in the Department of Agriculture. In 1930, this unit became known as the FDA [23]. Under the 1906 law, the agency was primarily a policing organization that responded to violations of food and drug regulations after they occurred. In cases of suspected adulteration, drugs were tested for strength and purity. The law did not address the testing of new drugs.

Consequently, because the Elixir Sulfanilamide lacked ethanol, the only law with which the Massengill Company failed to comply was the misbranding regulation. Were it not for this slight technicality, the FDA would have not had the authority to seize the drug [3]. Fortunately, because of the FDA's diligence, 228 of the 240 gallons of the elixir were retrieved before distribution. It is estimated that if all 240 gallons had been consumed, the number of deaths could have exceeded 4000 [3].

The Epidemic

Three hundred fifty-three patients received Elixir Sulfanilamide during a 4-week period in the fall of 1937. There were 105 deaths and 248 survivors, for a case fatality rate of 30% [2]. Thirty-four children and 71 adults died. The drug was administered on the direction of a physician in 100 of the 105 patients who died. Reasons for use included gonorrhea, sore throat, ear infections, soft-tissue infections, and syphilis. The directions stated, “Begin with 2 to 3 teaspoonfuls in water every four hours. Decrease in twenty-four to forty-eight hours to 1 or 2 teaspoonfuls and continue at this dose until recovery” [10]. The problem, of course, was that none of these 105 patients recovered. The mean fatal dose of the elixir was 53 mL for children and 99 mL for adults. This is equivalent to a mean fatal diethylene glycol dose of 38 g in children and 71 g in adults. Survival time from the first dose was 9.4 days (range, 2 to 22 days) [2].

The earliest clinical symptoms were nausea and vomiting. Additional deaths would probably have occurred if many survivors had not discontinued use of the medication because of the gastrointestinal disturbances. Later symptoms included manifestations of renal failure such as polyuria, anuria, flank pain, coma, and, in some patients, seizures. Laboratory data were not usually obtained [2, 24, 25].

Postmortem examinations were most remarkable for “hydropic tubular nephrosis” [26], a finding we would now call vacuolar nephropathy. Liver examination showed central degeneration. These observations were remarkably similar to the findings described in patients who had Dioxan poisoning [26]. Renal calcium oxalate deposition, a common feature of ethylene glycol poisoning, was not noted [27, 28].

Geiling and colleagues [29] did toxicity studies to determine the exact toxic ingredient and to observe whether similar pathologic findings could be reproduced in animal models. The investigators compared the effects of sulfanilamide, diethylene glycol, Elixir Sulfanilamide, and a synthetic elixir that contained the same ingredients and proportions found in the Massengill preparation. These studies clearly showed that diethylene glycol was the responsible toxin. Unadulterated sulfanilamide was exonerated.

Congressional Action

This terrible epidemic stirred passions throughout the nation. In November 1937, both Houses of the U.S. Congress passed resolutions requesting a full investigation into the tragedy. In the report that followed, lax drug regulations were implicated for allowing the epidemic to occur [8]. The report urged that the current legislation be strengthened to protect the public from the dangers of premature distribution of new drugs.

In fact, legislation to revise the 1906 regulations had been considered by Congress ever since the first inauguration of President Franklin Roosevelt in 1933 [3]. Although supported by the Assistant Secretary of Agriculture, Rexford Tugwell, the legislation stalled on Capitol Hill and had not passed by the fall of 1937. Mandatory safety testing was not part of the proposal. Even if the proposal had passed, the elixir tragedy would probably not have been averted.

Similar to the upsurge in public pressure on Congress that arose from the fear of contaminated meat 30 years earlier, congressional resolve to pass a strengthened Food and Drug Act was dramatically altered by the unforeseen elixir tragedy. On 1 December 1937, Senator Royal S. Copeland of New York, a physician and one-time health commissioner of New York City, introduced Senate Bill S. 3073, which for the first time required pharmaceutical manufacturers to show product safety before distribution [3]. Six months later, President Franklin Roosevelt signed the 1938 Federal Food, Drug and Cosmetic Act into law [30]. This bill was one of the last major domestic acts of the New Deal. It required that pharmaceutical companies submit a new drug application report to the FDA showing drug safety before the interstate shipment of any new pharmaceutical agent. The new statute also banned dangerous drugs (for example, radium water) and false and misleading labeling. Formula disclosures of all active ingredients were now required. Directions for use and warnings about possible misuse were also required unless the drug was sold by prescription. Penalties for infractions of any of these regulations were also stiffened [31].

One consequence of the regulations was the increasing practice of requiring prescriptions for the dispensing of certain drugs. This change in how patients obtained medication would greatly affect the physician–patient relationship. Previously, most medications (except for narcotics) were readily available over the counter; only about 25% were ordered by prescription [32]. The 1951 Humphrey-Dunham Drug Prescription Act would codify these trends, clearly separating drugs into those that required a physician prescription and those that could be sold over the counter [33].

Despite the vast improvements of the 1938 law, significant regulatory limitations remained. Proof of efficacy was not required. Animal testing was not standardized, and human trials were often poorly executed. In fact, the FDA did not review a new drug application until the drug manufacturer had completed its own tests. Drugs studied in premarketing clinical trials were exempted from review. Furthermore, if the FDA failed to consider the new drug application within 60 days, the drug was automatically approved.

Twenty-three years passed before another significant attempt was made to strengthen drug safety regulations. This time, Senator Estes Kefauver would spearhead the drive to further regulate the pharmaceutical industry [34]. Similar to the early Wiley and Tugwell efforts, Kefauver's bill also confronted significant opposition. Another drug tragedy was required, this time caused by thalidomide, for further regulatory reform to occur.

Fortunately, the 1938 regulations and the diligent efforts of the FDA official, Dr. Francis Kelsey, prevented the marketing of thalidomide in the United States [34]. Most of this tragedy was confined to Europe. Nonetheless, under the guise of conducting a new-drug investigation, the Merrell Pharmaceutical Company managed to supply more than 2 500 000 thalidomide tablets to 1270 physicians in the United States for investigational use. These were distributed to 20 000 patients, including 624 pregnant women [35]. Ten cases of thalidomide embryopathy resulted. Such events made it increasingly evident that stronger regulations were required for the investigational process of a new drug. The resulting 1962 New Drug Amendments called for the FDA to closely monitor all stages of new-drug development. The investigational new-drug process now required comprehensive animal testing before extensive human trials were started. Proof of efficacy and safety was mandated. Time constraints for the disposition of new drug applications were removed [36]. The transition of the FDA from an agency responding to events to an agency actively scrutinizing new-drug development was complete.

Aftermath and Recent Diethylene Glycol Mass Poisonings

Senator Royal Copeland, sponsor of the 1938 Act, died of exhaustion 4 days after it was signed into law. Dr. Massengill pleaded guilty to 112 counts of adulteration and misbranding and paid a fine of $26 100 [37], at the time the largest fine ever levied for violation of the 1906 law. The chemist Watkins committed suicide [9]. Gerhard Domagk was awarded the 1939 Nobel Prize for Medicine for his discovery of the antibacterial properties of prontosil.

In the JAMA editorial in which the first deaths from Elixir Sulfanilamide were reported, a warning about what could happen in the future was clearly enunciated: “It would indeed be a pity if this tragedy were to be repeated again and again as more and more new remedies are put forth” [15]. Over the next few years, the antimicrobial revolution spurred the development of many new drugs at a pace never before seen. Fortunately, the American public now had some regulatory protection to prevent a recurrence of a similar tragedy. Unfortunately, these regulations carried little weight outside the United States.

Although most European countries (including Germany and Great Britain—home to most of the thalidomide babies) finally adopted strong new-drug testing regulations after the thalidomide disaster, developing countries have lagged much further behind in this pursuit [38]. Practices that have adversely affected Third World countries include dispensing drugs without a prescription, mislabeling, drug dumping, and bribery. In recent years, the World Health Organization has attempted to improve and harmonize drug regulations in these nations. Nevertheless, continuing lax regulations in several of these countries has facilitated additional episodes of mass poisonings, including diethylene glycol mass poisonings reminiscent of the elixir tragedy.

In 1969, an epidemic of fatal renal failure occurred among seven children in Cape Town, South Africa [39]. As therapy for fever, sedatives formulated with diethylene glycol instead of propylene glycol were given to each child. Shortly after the children received the medication, vomiting, diarrhea, and dehydration developed. Anuria, acidotic breathing, hepatomegaly, and unresponsiveness ensued. The blood urea nitrogen level ranged from 132 to 240 mg/dL. The patients were treated with fluid hydration and correction of the acidosis. Two of the children received peritoneal dialysis. Postmortem examination showed extensive tubular necrosis of the kidney and centrilobular hydropic degeneration of the liver.

In another diethylene glycol-mediated catastrophe, 14 unexpected deaths from renal failure occurred at a hospital in Bombay, India, in 1986 [40]. The patients had been hospitalized for various conditions, including brain tumors, head injuries, glaucoma, cataracts, cerebral vascular accidents, and iridocyclitis. All affected patients were supposed to receive medicinal glycerine for its osmotic diuretic effect. Instead, they each received the less costly industrial glycerine, which contained 18.5% diethylene glycol. Twelve of 14 patients received hemodialysis, but this treatment was unsuccessful.

Most recently, at a Nigerian hospital in 1990, 40 children ranging in age from 6 months to 23 months died unexpectedly from renal failure [1]. They had all previously received acetaminophen syrup for the treatment of upper respiratory tract infections or malaria. An investigation showed that local chemists had substituted diethylene glycol for propylene glycol in the acetaminophen preparation.

Other Glycols, Other Mass Poisonings

Apart from the U.S. elixir disaster, diethylene glycol poisoning has not been a common toxicologic problem in the United States. It continues to be manufactured as an industrial solvent and is a constituent of sprinkler antifreeze solutions, lubricants, paints, and cosmetic creams [41]. Clinical experience with this agent is mostly limited to the previously discussed mass poisoning episodes. Recently, Cantarell and colleagues [42] reported a case series from Spain of five patients with extensive burn injuries who died from acute renal failure and acidosis after the topical application of a 1% silver sulfadiazine burn preparation containing diethylene glycol [42]. In another recent account from Europe, certain white wines were found to be contaminated by diethylene glycol that had been added by wine producers to improve the taste of the wine [43]. No human toxicity was observed.

Today, ethylene glycol poisoning is a much more common problem. Ethylene glycol is readily available as automobile radiator antifreeze fluid, often at concentrations of 95%. It is not uncommonly consumed by alcoholic persons searching for an ethanol substitute or by young children exploring the car or garage. Ingestion of as little as one swallow in a child, or of one to two ounces in an adult, may result in metabolic acidosis and renal failure from accumulation of toxic metabolites. In 1993, more than 3500 human exposures to ethylene glycol were reported to poison centers in the United States [44]. Moderate to major toxicity developed in 275 patients, and 2 patients died.

Propylene glycol, another condensation product of ethylene glycol production, is considerably less toxic than ethylene or diethylene glycol. It is used as a diluent in many injectable medications, including diazepam, digoxin, and phenytoin. When administered properly, propylene glycol is nontoxic. If the solution is infused too quickly (for example, in rapid intravenous loading of phenytoin), life-threatening cardiac manifestations may result, including hypotension, conduction disturbances, bradycardia, and asystole [45]. Polyethylene glycol is also considered to have low toxicity. The oral administration of multiple-liter amounts of polyethylene glycol solutions is often used for bowel cleansing before surgery or colonoscopy and as a decontamination technique for selective toxin ingestions.

Significant epidemics of mass poisoning have also occurred when other more toxic substances were substituted for ethanol. During Prohibition, an ethanolic extract of Jamaica ginger known as the Jake was a popular substitute for ethanol, particularly in the Southern and Midwestern United States. Jake was usually sold adulterated with castor oil. During the early 1930s, triorthocresylphosphate, a little-known neurotoxin, was substituted for the more expensive castor oil. The ingestion of the Jake-triorthocresylphosphate mixture resulted in paralysis (ginger jake paralysis) or gait impairment (jake walk), or both, in 50 000 people [46]. In another example of mass poisoning involving an ethanol surrogate, the substitution of methanol for ethanol in bootleg whiskey in Atlanta in 1951 resulted in 323 cases of methanol poisoning, including 41 deaths [47].

Although no similar iatrogenic mass poisonings involving prescription medications and resulting in as many deaths have occurred in the United States since the 1938 regulations were implemented, other medication tragedies continue to occur. In 1981, 16 deaths were reported among premature neonates after they were repeatedly exposed to an intravenous flush solution containing the antimicrobial preservative benzyl alcohol [48, 49]. The accumulation of large amounts of benzyl alcohol and its metabolite benzoic acid in premature neonates caused a “gasping syndrome” manifested by severe metabolic acidosis, encephalopathy, and respiratory depression with gasping. The problem resolved when the benzyl alcohol was removed from the flush solution. The eosinophilia-myalgia syndrome that recently developed in more than 1500 people after they received L-tryptophan is yet another example of a medication-induced mass poisoning—this time involving an over-the-counter product [50].

Conclusion

The legacy of the Elixir Sulfanilamide tragedy was the successful passage of the 1938 Federal Food, Drug, and Cosmetic Act—regulatory legislation that for the first time required drug manufacturers to show drug safety before public sale and distribution. As with other drug legislation passed during this century, increased consumer safety resulted from an emotional reaction to a public health catastrophe. Given all the pharmaceutical agents that have been produced in the intervening years, the public protection afforded by the 1938 regulations has probably prevented similar misadventures from occurring in the United States. The relationship between the pharmaceutical industry and the government and its effect on the practice of medicine were significantly changed by this event.