In a jolt to the pharmaceutical industry last June, the federal government filed the first criminal case based on Food and Drug Administration regulations for reporting adverse drug reactions.

Earlier this month, the industry received a second jolt when SmithKline Beckman Corp. changed its plea from innocent to guilty to all 34 counts. Three of its physician-executives, who also had pleaded innocent, entered no-contest pleas to 14 counts.

Now there is what Judge Edward N. Cahn called the "possibility" of a third jolt: jailing the doctors for up a year on each count, for a maximum of 14 years. He held out that possibility on Dec. 12, after they had switched their pleas at a U.S. District Court hearing in Allentown, Pa.

But, in an unusual action, he told them that if he decides that jailing is an option, he will let them revive their innocent pleas and stand trial.

The detailed basis for Cahn's deliberations became public for the first time when the prosecution put into the hearing record a basically undisputed summary called "Elements of the Offense and Factual Basis for the Pleas."

The drug at issue is Selacryn (ticrynafen). Physicians prescribed it for high blood pressure during an eight-month period that ended in January 1980. George Washington University Medical Center scientists have linked it to 25 fatal and 315 nonfatal liver injuries in an estimated 300,000 to 350,000 U.S. users.

The judge's deliberations are rooted in the choices that the Food, Drug, and Cosmetic Act gives the government: to allege a felony, which requires proof beyond a reasonable doubt of intent to violate the law and carries stiff penalties (up to three years in prison and a $10,000 fine), or to allege a misdemeanor, which does not require proof of either specific or general intent and limits incarceration to one year and the fine to $1,000.

Behind the scenes in the SmithKline case, government attorneys hotly disputed which way to go, sources disclosed. Prosecutors convened an investigative grand jury, but did not seek felony indictments. Finally, in a criminal information filed in June, they alleged 34 misdemeanors. One reason for this decision reportedly was to make guilty or no-contest pleas more likely and close a case that already had been in the mill for 3 1/2 years. As part of the package, the government agreed to make no sentencing recommendations.

Consequently, even though the defendants are explicitly not charged with having acted intentionally, the practical situation is simply: If the judge concludes from the record that the physicians had intended to violate the law, or even that they had been grossly negligent, the possibility that they will go to jail -- and for long, rather than short, terms -- will be sharply increased.

The 2-hour-and-20-minute hearing made it apparent that the lines between intent and negligence are blurred, making it possible for opposing counsel to interpret the facts as differently as two people may view a painting: one sees a sunrise; the other, a sunset.

What defense attorney Donald J. Goldberg saw was this: The government had searched for "evil" in a 3 1/2-year investigation encompassing 600,000 documents, but had not found it because SmithKline is a company that is run by "very decent people," that "tries to do right" and that is "dedicated to good medicine." Never convicted of a crime, it is one of the nation's most profitable corporations and one of Philadelphia's largest employers.

In the Selacryn case, "human error" tripped up SmithKline, causing key data on adverse reactions to be "dropped through the cracks" rather than be reported to the FDA, Goldberg told the judge. Indeed, "The government has been frank to say that it has found no evidence of criminal intent," he said.

But what Assistant U.S. Attorney Peter J. Smith saw was vastly different. The prosecutor portrayed the evidence this way:

It was sufficient to prove that, over a period of several months, the physicians repeatedly had neglected to exercise their "authority and opportunity to prevent" the violations for which the law made them strictly liable (and it was also sufficient to induce them to enter pleas that, as Cahn pointed out, are equivalent in law to guilty). At the same time, however, the evidence admittedly was insufficient for the government to try to prove that the doctors had acted intentionally.

What does the evidence before Cahn really show? Intent? Sustained negligence? Inadvertence? A blur? A mix?

In 1973, a French pharmaceutical firm, Albert Rolland S.A. (Anphar), licensed SmithKline to sell ticrynafen in seven countries and agreed to alert it to adverse reactions. Anphar sells the drug (as Diflurex) in France, but with a strong warning about liver reactions in the labeling.

SmithKline began to study the drug's safety and effectiveness in preparation for seeking FDA marketing approval in 1977.

The NDA said that eight test patients in two particular groups had experienced liver disorders "most likely due to viral hepatitis," although Selacryn "could not be entirely ruled out."

Later, in November 1978, the same point was made to an agency advisory panel by Dr. Thomas G. Davis, vice president and medical director of the SmithKline & French Laboratories division.

The FDA approved the NDA on May 2, 1979, and SmithKline began at once to promote, distribute and sell Selacryn with labeling citing the liver problem "in a few patients," but saying "no causal relationship has been established."

The three physician-defendants, who had no connection with the NDA, become involved with Selacyryn after the FDA approved it. They were in SK&F's U.S. Pharmaceutical Products unit, which was responsible mainly for medical and scientific monitoring of all SmithKline prescription drugs, for accuracy and completeness of labeling, advertising and information, and for collecting, evaluating and reporting side effects to the FDA.

The doctors at the time: Vice President and Medical Director Philip J. Tannenbaum, now 55; Group Director for Medical Affairs Ralph M. Myerson, 66, and Associate Director Theodore Selby, 58. Tannenbaum supervised Myerson, and he, in turn, supervised Selby. Selby had direct responsibilities for Selacryn, including physician inquiries about adverse reactions and for originating reports of them to the FDA.

On March 20, 1979, Anphar wrote -- in French -- a lengthy memo on 13 liver (hepatic) disorders in ticrynafen users, including five "probably related to" the drug, and sent it to Philadelphia. SmithKline received it in late April, had it roughly translated on May 2 (the day FDA approved Selacryn) and had a full translation on June 1. A copy of the full translation went to Tannenbaum and Selby.

On June 22, 1979, Anphar representatives told SmithKline at a meeting in Philadelphia of liver disorders in 20 Diflurex users, including six who had experienced "positive rechallenge." In a rechallenge, a user who had a reaction that subsided when he went off the drug takes it anew -- to see if the reaction recurs. Scientists rate a rechallenge as very strong evidence of a cause-effect link between a drug and an adverse reaction. The meeting minutes, which discussed the rechallenges and the underlying clinical data, went to Tannenbaum, Myerson and Selby.

Two other SmithKline doctors who were at the session called the rechallenges "definitely drug-related" in a memo that went to the defendants on July 17. But, predicting hepatic problems would not occur "with any significant frequency," they said "it would seem judicious to closely monitor this now that 'Selacryn' is available to the general population."

During the next five months, SmithKline received numerous reports from doctors in private practice about patients with possible adverse reactions to the drug.

They included:

* Dr. Mervyn Lakin of Phoenix, who, on Aug 15., phoned Myerson to say he had been stricken by hepatitis after taking Selacryn for six weeks. Myerson filled out a "Side-Effect Report Cover Sheet." On Sept. 7, after Lakin had supplied more information, Selby classified the hepatitis in a new SERC as "unrelated" to Selacryn. Four days later, Lakin phoned with news of a positive rechallenge -- in himself. On Sept. 12, Selby listed the cause as "indeterminate."

Later, Selby and other SmithKline officials and Anphar discussed eight rechallenges -- Lakin's and the seven by then known to Anphar. On Oct. 2, Selby termed Lakin's hepatitis "probably related" to Selacryn and told the doctor never to take the drug again. Neither he nor his codefendants gave such advice in phone calls and letters to numerous other physicians whose patients had developed liver problems.

* Dr. Stephen De Gray, of Bluefield, W. Va., phoned Myerson on Sept. 18 to say that a patient who had been on Selacryn for two weeks had developed jaundice and was in a hepatic coma. Eight days later, Myerson wrote a memo. Eight days after that, on Oct. 4, Selby filled out a SERC in which he called the cause of the coma "indeterminate" and sent it to his codefendants.

* On Oct. 1, SmithKline learned that a patient of Dr. Milo Farnham of Independence, Mo., had developed jaundice while on Selacryn and that the symptoms had reappeared in a rechallenge. On Oct. 10, Selby phoned Farnham and wrote a memo. On Oct. 23, Selby wrote a SERC rating the rechallenge -- the ninth to be reported -- "possibly related" to the drug.

* Dr. Caroline Riley of the Yale School of Medicine in New Haven, Conn., told Selby Nov. 30 of a Selacryn patient who had developed jaundice and lapsed into a coma.

* Dr. Harold Berenzweig of Fort Worth on Dec. 5 told Selby of a Selacryn patient who had suffered liver disorders. On Dec. 14, he reported that the patient had died.

Agency rules require reporting of "unexpected" adverse reactions within 15 working days after receipt. That all nine rechallenges were unexpected -- because none had been listed in the NDA -- is undisputed. That SmithKline reported none as unexpected also is undisputed.

FDA regulations also require quarterly "full reports" of information that was not part of the application, including "unpublished reports of clinical experience." About 30 days before the Aug. 2 and Nov. 2 Selacryn quarterly reports were due, SmithKline sought to assure their completeness by circulating detailed instructions in an "alerting memo" that went to the three defendants. But it did not mention the June 1 French liver data in either quarterly report, not reporting it to the FDA until 6 1/2 months after getting the full translation.

As recommended by Selby, SmithKline disclosed Lakin's rechallenge not in an expedited unexpected-reaction report, but in the quarterly report of Nov. 2 -- 50 days after Lakin had reported it. It was grouped with 11 other reports, and an FDA reviewer didn't spot it for nearly a month.

Early in 1980, George Washington University's Dr. Hyman J. Zimmerman, an authority on liver diseases, phoned the FDA with an ominous message: The hepatic adverse reaction suffered by a Selacryn user in a rechallenge was irreversible. At the time, the agency knew of 52 cases of liver damage, including 30 cases of jaundice and five deaths.

The FDA then summoned SmithKline to a meeting on Jan. 15, 1980. SmithKline told the agency for the first time of the eight other rechallenges, the West Virginia and Connecticut hepatic comas, and the Texas death. SmithKline argued for relabeling, but the FDA told it to stop selling Selacryn, and it did so. Nine days later, the Public Citizen Health Research Group, founded by Ralph Nader, requested prosecution.

The company pleaded guilty to the 14 counts to which the three physicians pleaded no contest and to 20 additional counts, each alleging interstate shipment of Selacryn with improper labeling. Dr. Davis the SK&F vice president, is also charged in the 20 counts; his plea is innocent.