Patricia W. Salmon, Guardian Ad Litem of Timothy B. Salmon, a Minor v. Parke, Davis and Company

BUTZNER, Circuit Judge: The guardian ad litem of Timothy B. Salmon, a child who developed aplastic anemia after being treated with Chloromycetin, appeals from a summary judgment dismissing his action against Parke, Davis & Company, the manufacturer of the drug. 1 The principal allegations of negligence are that Parke, Davis failed to warn physicians effectively of the drug’s hazards and that it overpromoted the drug. Since jurisdiction is based on diversity of citizenship, the law of North Carolina, where Timothy was treated, is applicable. The district court held that Parke, Davis’ warning was adequate as a matter of law and that the evidence was insufficient to show over-promotion. While this appeal was pending, the North Carolina Court of Appeals decided the same issues adversely to Parke, Davis in Whitley v. Cubberly, 24 N.C.App. 204, 210 S.E.2d 289 (1974). Unfortunately, the district court did not have the *1361 benefit of that opinion; on the contrary, Parke, Davis had cited as authority the erroneous grant of summary judgment by the state trial court that Whitley reversed. Though the decision of an intermediate court is not controlling in diversity cases, Whitley’s exposition of North Carolina law is persuasive. Guided largely by it, we reverse and remand this action for trial. In May 1968, when Timothy was two years old, a physician prescribed Chloromycetin for an injury to his palate, and again in December for bronchitis. He soon developed aplastic anemia, which results from the failure of bone marrow to produce blood cells. The disease is generally irreversible and apparently has left Timothy permanently and seriously disabled. Chloromycetin, Parke, Davis’ trade name for chloramphenicol, is a potent, broad-spectrum antibiotic. Properly administered, it is a valuable, life-saving drug that can effectively treat stubborn infections. But it can be injurious — even fatal — if its use is not carefully monitored. According to the Food and Drug Administration, its most common, serious toxic effect is the development of anemia. Parke, Davis’ package inserts have always mentioned the drug’s potential danger, and in 1961 the company revised its warnings, emphasizing precautions that should be observed. Nevertheless, the adequacy of its warnings and the charge of overpromotion continue to be the subject of litigation. In every instance, the question of Parke, Davis’ negligence, both before and after the 1961 revision, has been held by appellate courts to present a question for the jury. Stevens v. Parke, Davis & Co., 9 Cal.3d 51, 107 Cal.Rptr. 45, 507 P.2d 653 (1973); Incollingo v. Ewing, 444 Pa. 263, 282 A.2d 206 (1971); Whitley v. Cubberly, 24 N.C.App. 204, 210 S.E.2d 289 (1974); Love v. Wolf, 249 Cal.App.2d 822, 58 Cal.Rptr. 42 (1967); cf. Lake v. Konstantinu, 189 So.2d 171 (Fla.Dist.Ct.App.1966); but cf. Stottlemire v. Cawood, D.C., 213 F.Supp. 897 (1963). In 1968 Parke, Davis again modified its warning, but apparently this revision was not in effect at the time Timothy was treated. The district court ruled that the uncontradicted evidence showed that the warning involved in this case stated: “Warning — Blood dyscrasias may be associated with the use of chloramphenicol. It is essential that adequate blood studies be made. (See enclosed warnings and precautions.)” The package insert read in part as follows: “Warning — Serious and even fatal blood dyscrasias (aplastic anemia, hypoplastic anemia, thrombocytopenia, granulocytopenia) are known to occur after the administration of chloramphenicol. Blood dyscrasias have occurred after both short-term and prolonged therapy with this drug. Bearing in mind the possibility that such reactions may occur, chloramphenicol should be used only for serious infections caused by organisms that are susceptible to its antibacterial effects. “Chloramphenicol should not be used when other less potentially dangerous agents will be effective; or in the treatment of trivial infections such as colds, influenza, or viral infections of the throat; or as a prophylactic agent. “Precautions — It is essential that adequate blood studies be made during treatment with the drug. While blood studies may detect early peripheral blood changes, such as leukopenia or granulocytopenia, before they become irreversible, such studies cannot be relied on to detect bone marrow depression prior to development of aplastic anemia.” Since Timothy’s prescription was filled by a pharmacist, neither his parents nor the physician saw the package warning and insert. The same material, however, was available to the doctor in the Physicians Desk Reference, and she had previously seen either the pre-1961 package warning and insert or its revision that Parke, Davis enclosed with a sample of the drug. *1362 A manufacturer of an ethical drug must exercise reasonable care, commensurate with the risk, to warn physicians effectively of the drug’s inherent dangers. Stevens v. Parke, Davis & Co., 9 Cal.3d 51, 107 Cal.Rptr. 45, 507 P.2d 653, 661 (1973); Whitley v. Cubberly, 24 N.C.App. 204, 207, 210 S.E.2d 289, 292 (1974); cf. Stegall v. Catawba Oil Co., 260 N.C. 459, 133 S.E.2d 138, 141 (1963); Restatement (Second) of Torts § 388. Though the law does not impose strict liability on the manufacturer of a potentially dangerous drug, it does exact a high degree of care. And when a manufacturer does- not change a warning it knows is widely disregarded, a jury may infer that the warning is insufficient. Incollingo v. Ewing, 444 Pa. 263, 282 A.2d 206, 219, 222 (1971). In North Carolina, as elsewhere, compliance with federal laws and regulations concerning a drug, though pertinent, does not in itself absolve a manufacturer of liability. Whitley, 210 S.E.2d at 292; see generally 1 Frumer & Friedman, Products Liability § 807[1]. Furthermore, overpromotion of the drug may erode the effectiveness of otherwise adequate warnings. Stevens, 507 P.2d at 661; Whitley, 210 S.E.2d at 292. But, of course, a jury need not reach the question of overpromotion if it decides the warning was inadequate in the first place. At this stage of the proceedings, review is limited to determining whether the case should have been decided summarily or submitted to the jury. There is no controversy about the historical facts, but that does not end the inquiry. The critical question is whether a genuine issue exists as to the inferences or conclusions that may properly be drawn from the evidence. American Fidelity and Casualty Co. v. London and Edinburgh Insurance Co., 354 F.2d 214, 216 (4th Cir. 1965); Pierce v. Ford Motor Co., 190 F.2d 910, 915 (4th Cir. 1951). The inferences must be viewed in the light most favorable to the appellant, and the judgment must be reversed if “inferences contrary to those drawn by the trial court might be permissible.” United States v. Diebold, 369 U.S. 654, 82 S.Ct. 993, 8 L.Ed.2d 176 (1962). Applying these principles, we conclude that conflicting inferences can properly be drawn from the facts for the following reasons: f Parke, Davis’ warning states that blood studies are essential during treatment with Chloromycetin. The Food and Drug Administration urged the company to add this statement: “Because of the necessity of repeated blood studies during therapy, it is desirable that patients be hospitalized if possible.” Parke, Davis resisted this suggestion because hospitalization is so expensive. The district court found its opposition to be neither negligent nor irresponsible. Considering the risk of aplastic anemia and the recognized need for tests and observation, a jury, however, could reasonably find that an adequate warning should conform to the F.D.A. suggestion despite the additional cost. If Parke, Davis’ warning states that blood studies cannot be relied on to detect bone marrow depression before aplastic anemia develops. The appellant charges negligence because Parke, Davis did not explain what other tests are essential for prudent use of the drug. The district court ruled that “since all the actual risks are communicated to the physician, the failure of Parke, Davis to refer to the iron plasma or other tests cannot be considered negligent.” From the scant record before us, it appears that Parke, Davis knew, or should have known, better than the average physician what other tests were necessary. Consequently, a jury could view the warning as incomplete if it did not provide physicians all of the essential information about this important subject. H Parke, Davis’ warning also states the drug “should be used only for serious . . . infections . . . not . . . in the treatment of trivial infections such as colds, influenza, or viral infections of the throat. . . . ” In contrast, a medical journal quoted by the appellant states: *1363 “The safest approach is to give chloramphenicol only after the following questions have been seriously considered: 1. Is this drug necessary to save life or to greatly minimize a valid risk to life? 2. Is it necessary to preserve a vital organ or its function or to greatly minimize serious risk of their loss? 3. Will any other antibiotic or course of action do as well in this situation? From the above appraisal it should be clear that most if not all candidates for chloramphenicol will be hospitalized patients or out-patients requiring very close follow up.” 2 Comparing the company’s warning with the article, a jury could infer that Parke, Davis’ version was unreasonably diluted. H Another medical journal, citing cases of fatal aplastic anemia for which “it seemed clear that [chloromycetin] was responsible,” suggested restriction of the drug to use in “impelling circumstances, i. e., conditions for which no other antibiotic is currently effective.” It recommended that advertising against indiscriminate use of the drug could be made more forceful. The journal blamed physicians for promiscuous use of the drug and stated that the manufacturer should not be “taken to task.” 3 Although Parke, Davis changed its package warning after the article appeared, a jury could reasonably find that even then the company failed to warn physicians, particularly in its advertising, with emphasis that was commensurate with the risk. A jury could also conclude that the fault of misuse should be shared by both the physician and the manufacturer. II The limited evidence presented on the motion for summary judgment supports the conclusion that chloromycetin can cause aplastic anemia. Parke, Davis’ warning did not contain an unequivocal statement of this causal relationship but spoke instead in terms of aplastic anemia being “known to occur after administration of” the drug. Significantly, the doctor who treated Timothy stated that although she was aware it was suspected that chloromycetin “could” cause aplastic anemia, she could not say with absolute certainty that the drug “can cause” this disability. Again, the jury could reasonably infer that Parke, Davis’ warning lacked the emphasis that the danger demanded, and that the company should have cautioned physicians in no uncertain terms that, according to present knowledge, the drug can cause aplastic anemia. If The F.D.A. suggested, and Parke, Davis opposed, language that would tell physicians they “must” take certain precautions and “must not” incur needless risks. The district court ruled that Parke, Davis’ insistence on the use of “should” was reasonable. Again, we believe that a jury might properly draw a contrary inference from the facts, especially since physicians apparently continued to use the drug carelessly to treat minor infections. f'Timothy’s physician had received a calendar advertising chloromycetin, together with a sample package containing a warning about the drug. The district court held that as a matter of law this evidence was insufficient to show overpromotion. Though the evidence is slight, we cannot say that its probative value is wholly lacking. It is foreseeable that a calendar might remain on a physician’s desk as a constant reminder to prescribe a drug long after the sample and its warning had been removed. A jury could infer, therefore, that the absence of a warning on an advertisement for the use of a drug as potentially dangerous as chloromycetin was a form of overpromotion which nullified the effect of even a valid warning on the package. Cf. Stevens v. Parke, Davis & Co., 9 Cal.3d 51, 107 Cal.Rptr. 45, 507 P.2d 653 (1973); Incollingo v. Ewing, 444 Pa. 263, 282 A.2d 206 (1971); Whitley v. Cubberly, 24 N.C.App. 204, *1364 210 S.E.2d 289 (1974). The likelihood of overpromotion by advertisements that lack a warning is increased when a physician writes a prescription without having either the package or its insert at hand and the patient obtains the drug from a pharmacist. Indeed, an article in the Journal of the American Medical Association observed that Parke, Davis’ advertisements “could be made more forceful.” 4 Our catalogue of permissible inferences is intended neither to define nor to decide the issues in this case. It serves simply to illustrate our reasons for concluding that summary judgment is inappropriate and that, as in negligence actions generally, it is for the jury to say whether the defendant breached its duty to exercise reasonable care. Spaulding v. Ads-Anker Data Systems— Midwest, Inc., 498 F.2d 517 (4th Cir. 1974); Pierce v. Ford Motor Co., 190 F.2d 910, 915 (4th Cir. 1951). The judgment is reversed and the case is remanded for trial on the merits. Footnotes [1] Timothy’s doctor is joined as a defendant, charged among other things with needlessly prescribing Chloromycetin for minor infections and failing to conduct essential blood tests. The question of the doctor’s liability is not an issue on this appeal. [2] Hartman, Chloramphenicol and the Code of Hammurabi, American Practitioner, 497, 499 (August 1962). [3] W. Dameshek, Chloramphenicol — A New Warning, 174 Journal of the American Medical Association, 1853 (Dec. 3, 1960). [4] See text at n. 3 supra.