Cathy Adelman-Tremblay brought this diversity action against Jewel Companies, Inc. ("Jewel") and Pacific World Corp. ("Pacific World") to recover for injuries suffered when she applied artificial fingernails from a kit assembled by Pacific World and sold by Osco Drug Store, owned and operated by Jewel.1 She alleged strict product liability and breach of implied warranty2 against Jewel, and strict product liability, negligence and res ipsa loquitur against Pacific World. Plaintiff also asserted a claim for punitive damages against Pacific World. The district court granted summary judgment in favor of both defendants on all of plaintiff's claims. We affirm.

Summary judgment is appropriate if there are no genuine issues of material fact and the defendants are entitled to a judgment as a matter of law. See Fed.R.Civ.P. 56(c). In reviewing the grant of summary judgment, we read the record--including the pleadings, depositions and affidavits--in the light most favorable to the plaintiff, the nonmoving party. Miller v. A.H. Robins Co., Inc., 766 F.2d 1102, 1104 (7th Cir.1985). We recognize that tort actions generally are not disposed of by summary judgment because they typically involve a myriad of factual issues. Gracyalny v. Westinghouse Elec. Corp., 723 F.2d 1311, 1316 (7th Cir.1983); 10A C. Wright, A. Miller & M. Kane, Federal Practice and Procedure Sec. 2727 (1983). But where the facts are undisputed, as they appear to be in this case, and the plaintiff is precluded from recovery as a matter of law, a grant of summary judgment in favor of the defendants is appropriate. Id. Sec. 2729, at 197. With these principles in mind, we describe the background and events of this case.

The plaintiff purchased a "Nailene Nail Kit" at an Osco Drug Store in Wisconsin. The kit included artificial fingernails, a nail sander file and a tube of cyanoacrylate liquid glue (also known as "Crazy Glue" and "Super Glue"). After attaching the artificial nails to her own fingernails with the cyanoacrylate glue, the plaintiff experienced pain and permanently lost all of her natural fingernails. The three physicians who treated the plaintiff in the succeeding five months characterized her reaction as "allergic contact dermatitis." The plaintiff testified that she is allergic to many substances, including codeine, penicillin, aspirin, cats, dogs, trees, grass, nickel, pollen, bees, dust and certain fragrance products.

Pacific World has sold more than one million Nailene kits, all containing the cyanoacrylate glue, since it began marketing them in 1983. The injury to the plaintiff is the only adverse reaction known to Pacific World, and neither Pacific World nor Jewel has ever received a complaint about the Nailene kit.

The defendants moved for summary judgment, asserting that neither a manufacturer nor a seller are liable when an unusually susceptible consumer suffers a rare allergic reaction to a product not previously known to cause such a reaction. The plaintiff responded to the motion by filing an affidavit of Dr. James D. Hogan, a dermatologist to whom the plaintiff was referred by her treating physicians. The plaintiff planned to use Dr. Hogan as her medical expert at trial. Dr. Hogan stated that the plaintiff's injuries were caused by a toxic, not an allergic, reaction to the glue. On the basis of this affidavit, the plaintiff opposed summary judgment by arguing that a question of fact existed with respect to whether her injury was caused by a toxic or an allergic reaction. A toxic reaction would allegedly show that the product was defective. She also argued that resolution of this factual issue was material to the question whether the defendants' failure to warn about the product's dangerous side effect constituted negligence.

After briefing was completed on the defendants' summary judgment motion, Pacific World deposed Dr. Hogan and submitted an affidavit summarizing his testimony. The affidavit indicated that Dr. Hogan had changed his mind as to the nature of the plaintiff's reaction to the glue. Based on the results of a patch test and on his consultation with Dr. Alex Fischer, a dermatologist who is an authority on contact dermatitis, Dr. Hogan concluded that the plaintiff suffered an extremely rare allergic reaction previously unknown to him and unreported in the medical literature. Based on new evidence, Dr. Hogan no longer believed that the plaintiff suffered a toxic reaction. Because he had recently learned of only two other similar occurrences, Dr. Hogan indicated that he was unaware of any reason for giving a warning as to the use of the glue found in the nail kit.3

The district court issued a memorandum order indicating that, in light of Dr. Hogan's testimony, summary judgment was appropriate, but withheld granting the motion to allow the plaintiff to address the deposition of Dr. Hogan.

At oral argument on the summary judgment motion, the plaintiff filed a supplemental affidavit of Dr. Hogan, purporting to clarify his deposition testimony. In his affidavit, Dr. Hogan stated that the patch test showed that the plaintiff was allergic to the glue "at this point in time," but that it did not prove that her initial response to the glue was an allergic reaction. He speculated that the allergic reaction to the patch test may have been the result of the plaintiff's previous exposure to the glue, but that the "initial insult mechanism" causing plaintiff's injury could not be determined. Dr. Hogan also concluded that the patch test does not establish whether the plaintiff is unusually susceptible to the glue or whether the glue contains an ingredient potentially harmful to the general population. It was his opinion, however, that it was unlikely that the plaintiff was an unusually susceptible person, that a defect in the glue posed a risk of harm to the ordinary consumer and that, therefore, the product's label should warn of the possibility of an allergic or a toxic reaction. The plaintiff also submitted an article from the October 1987 issue of Dermatology Times that reported four case studies and recommended patch testing for sensitivity to cyanoacrylate glue.

The district court granted the defendants' summary judgment motion. The court stated that "a major part of Dr. Hogan's [second] affidavit flatly contradicted his deposition testimony." The court noted, for example, that Dr. Hogan's latest version of the cause of the plaintiff's injury--that she suffered from either a toxic or an allergic reaction--implied that the mechanism of the plaintiff's injury was unknowable and thus contradicted his deposition testimony that the plaintiff had suffered an allergic reaction. The court also noted that Dr. Hogan's opinion that ordinary consumers are endangered by the product contradicted his deposition testimony that the plaintiff's allergic reaction was a rare occurrence. Because there was no newly discovered evidence and the plaintiff had not lacked access to material facts--two compelling reasons for which courts will allow plaintiffs to contradict prior sworn testimony to avoid summary judgment--the district court held that Dr. Hogan's supplemental affidavit could not be used to create a factual issue. The district court concluded that because the plaintiff suffered an unusual allergic reaction to the glue, she would be unable to prove a case against defendants on any theory of product liability. Thus, summary judgment in favor of Jewel and Pacific World was appropriate. Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S. Ct. 2548, 2552-53, 91 L. Ed. 2d 265 (1986). We affirm.

On appeal, the plaintiff argues that Dr. Hogan's second affidavit precludes summary judgment by creating an issue of material fact as to whether she suffered a toxic or an allergic reaction to the glue. The plaintiff claims that Dr. Hogan's second affidavit--in which, contrary to his deposition testimony, he indicated that the plaintiff's reaction to the glue may have been toxic rather than allergic--was based on newly discovered evidence and was intended to clarify, not contradict, the "broad, general statements made at his deposition." Brief of Plaintiff-Appellant at 13. Thus, according to the plaintiff, the district court erred in concluding that this affidavit could not be used to create an issue of fact.

In this circuit, a party may avoid summary judgment by submitting an affidavit that conflicts with its earlier deposition testimony in only a limited number of circumstances. Babrocky v. Jewel Food Co., 773 F.2d 857, 861 (7th Cir.1985). For example, a subsequent affidavit may be allowed to clarify ambiguous or confusing deposition testimony. Id. In this case, however, Dr. Hogan's testimony was a model of clarity. See Appendix to this opinion. He essentially testified that he had not changed his mind about the nature of the plaintiff's injury; that, based on the results of a patch test and on his consultation with Dr. Fischer, he believed that the plaintiff suffered a rare allergic reaction rather than a toxic reaction to the glue. Dr. Hogan directly contradicted this testimony in his supplemental affidavit in which he stated that the patch test does not indicate whether the plaintiff's reaction upon initial exposure to the glue was allergic or toxic.

A contradictory supplemental affidavit is also permissible if it is based on newly discovered evidence. The plaintiff points to a journal article as newly discovered evidence that supports Dr. Hogan's position in his second affidavit. The article warns that the use of artificial nail products may cause contact dermatitis based on four case studies of adverse reactions to such products. Dermatology Times at 1, 10 (October 1987). The article does not, however, appear to describe any toxic reactions. Thus, it fails to provide support for Dr. Hogan's final conclusion that the plaintiff's initial reaction to the glue may have been toxic.

The plaintiff's submission of the affidavit is little more than a desperate attempt to resuscitate her claim that her reaction was toxic and that therefore the glue was defective and Pacific World was negligent in failing to warn. Because her opposition to summary judgment was based solely on the issue of toxicity versus allergic reaction to the glue, her desire to preserve the issue is understandable.

The rule against creating "sham" issues by submitting affidavits that contradict prior depositions thus far has been applied only to parties. See Diliberti v. United States, 817 F.2d 1259, 1263 (7th Cir.1987); Babrocky, 773 F.2d at 861; Miller, 766 F.2d at 1104. We can think of no reason, however, not to apply this rule to the present case involving the testimony and affidavit of the plaintiff's sole expert witness. The purpose of summary judgment motions--"to weed out unfounded claims, specious denials, and sham defenses," Babrocky, 773 F.2d at 861--is served by a rule that prevents a party from creating issues of credibility by allowing one of its witnesses to contradict his own prior testimony. Id. (quoting Camfield Tires, Inc. v. Michelin Tire Corp., 719 F.2d 1361, 1366 (8th Cir.1983)).

We must now determine whether the district court correctly concluded as a matter of law that the plaintiff may not recover for her allergic reaction to the glue. Wisconsin has not to our knowledge addressed the issue whether a consumer can recover from a seller or a manufacturer for injuries sustained in using a product where those injuries result from unusual allergic reactions. Jurisdictions that have examined such cases, however, refuse to impose liability on sellers or manufacturers for allergic or otherwise idiosyncratic reactions to products.

Wisconsin, like most jurisdictions, recognizes a cause of action in negligence for a manufacturer's failure to warn of a danger associated with a product. Kozlowski v. John E. Smith's Sons Co., 87 Wis. 2d 882, 898-99, 275 N.W.2d 915, 922-23 (1979). There is, however, in the majority of jurisdictions no duty to warn of the possibility of a rare and unusual allergic reaction. See W. Keeton, Prosser and Keeton on the Law of Torts Sec. 96, at 687 (5th ed. 1984);4 72 C.J.S. Products Liability Sec. 26, at 42 (Supp.1975) ("There is generally no duty to warn of harm from an unusual allergic reaction from use by a miniscule percentage of users, where the risk is so rare that the injury is considered to result from a personal idiosyncracy of the consumer.") (footnotes omitted); Annotation, Unusual Susceptibility to Injury, 26 A.L.R. 2d 963 (1952). Another way of stating this rule is that the plaintiff's own idiosyncracy or allergy is the proximate cause of his injury, not the failure to warn. Presbrey v. Gillette Co., 105 Ill.App.3d 1082, 1091, 61 Ill. Dec. 816, 823, 435 N.E.2d 513, 520 (1982). Liability should be imposed on a manufacturer only if it has actual knowledge that its product might cause harm to an identifiable class of sensitive users or if, based on the state of scientific knowledge, it is chargeable with such knowledge. Howard v. Avon Prods., Inc., 155 Colo. 444, 454, 395 P.2d 1007, 1012 (1964).

The plaintiff has offered no reason why Wisconsin would not follow these general principles with respect to liability of manufacturers for failure to warn of a product's danger to an unusually susceptible consumer. In the circumstances of this case, we find no reason to impose upon Pacific World a duty to warn its consumers of a potential allergic reaction to cyanoacrylate glue. Pacific World sold one million nail kits and the plaintiff's was the only reported adverse reaction. The medical literature, at the time Pacific World assembled and marketed the kits, indicated that although the glue's fumes were irritating, allergic reactions were "considered virtually impossible." A. Fischer, Contact Dermatitis 559 (1986). The plaintiff is now able to point to a few other cases of allergic contact dermatitis caused either by cyanoacrylate glue or by other artificial nail products. These all came to light shortly after plaintiff suffered her injury. The plaintiff has not come close to showing that she is a member of an identifiable class of consumers who are sensitive to cyanoacrylate glue. Nor has she pointed to scientific knowledge in existence at the time of her injury with which Pacific World should be chargeable.5

Under Wisconsin law, both Jewel and Pacific World could be held strictly liable if the cyanoacrylate glue were defective or unreasonably dangerous to the consumer. Sumnicht v. Toyota Motor Sales, Inc., 121 Wis. 2d 338, 351-52, 360 N.W.2d 2, 7-8 (1984); Dippel v. Sciano, 37 Wis. 2d 443, 459, 155 N.W.2d 55, 63 (1967); Restatement (Second) of Torts Sec. 402A (1965). Although the plaintiff has clearly suffered an injury, she has failed to allege a defect. Theoretically, at least, if she suffered a toxic reaction to the glue, the presence of a toxin or poison in the glue might render it defective or unreasonably dangerous. See id. Sec. 402A comment h (defective condition may arise "from harmful ingredients, not characteristic of the product itself ... [and] from foreign objects contained in the product"). But a consumer who suffers an allergic reaction to a product without any identifiable defect, such as the reaction of the plaintiff to the glue, may not invoke the doctrine of strict liability to recover from a manufacturer or a seller. Annotation, Products Liability: Strict Liability in Tort Where Injury Results from Allergenic (Side-Effect) Reaction to Product, 53 A.L.R. 3d 298, Sec. 3 (1973) ("[A] product, faultlessly manufactured and containing no impurities, is not rendered defective per se, within meaning of the doctrine of strict liability in tort, by the mere fact that it causes injury to certain individuals who, because of hypersensitivity or other peculiarity of makeup, suffer an allergenic or idiosyncratic reaction when exposed thereto.") (footnotes omitted); accord Helene Curtis Indus., Inc. v. Pruitt, 385 F.2d 841, 851 (5th Cir.1967) (under Texas law, to recover under strict liability for injury from hair bleach, plaintiff must show that bleach mixture is defective and that scalp is not hypersensitive), cert. denied, 391 U.S. 913, 88 S. Ct. 1806, 20 L. Ed. 2d 652 (1968); Thomas v. Gillette Co., 230 So. 2d 870, 873 (La.App.) (manufacturer of hair relaxant not strictly liable where plaintiff showed only her apparently unusual reaction but no defect in product), cert. denied, 255 La. 809, 233 So. 2d 249 (1970); Thomas v. Amway Corp., 488 A.2d 716, 722 (R.I.1985) (in strict liability action plaintiff must prove a defect and that defect caused her injury). In the present case, although the plaintiff has not alleged any defect in the cyanoacrylate glue, she argues that the defendants should be held strictly liable for failure to warn of the possibility of an adverse allergic reaction to the glue.

Wisconsin will impose strict liability upon a seller or a manufacturer when failure to warn of a danger inherent in the product's use renders the product unreasonably dangerous. See Schuh v. Fox River Tractor Co., 63 Wis. 2d 728, 738-39, 218 N.W.2d 279, 284 (1974) (describing duty to warn in a strict liability case); see Restatement (Second) of Torts Sec. 402A comment j ("In order to prevent the product from being unreasonably dangerous, the seller may be required to give directions or warning, on the container, as to its use.").6 Wisconsin has not indicated whether strict liability for failure to warn applies in cases involving unusually allergic consumers. Most jurisdictions, however, will not generally hold manufacturers or sellers strictly liable for failure to warn of the possibility of a rare allergic reaction. See, e.g., Oakes v. E.I. Du Pont de Nemours & Co., 272 Cal. App. 2d 645, 651, 77 Cal. Rptr. 709, 713 (1969) (manufacturer of herbicide not strictly liable for failure to warn of "unknown or unknowable allergies, sensitivities and idiosyncracies"); Booker v. Revlon Realistic Professional Prods., Inc., 433 So. 2d 407, 410 (La.App.1983) ("Unusual or rare idiosyncratic sensitivity on plaintiff's part would not provide a basis for recovery or even a requirement of a warning from the manufacturer on the product."); Thomas, 488 A.2d at 722 (soap manufacturer not strictly liable for failure to warn because plaintiff's rash not reasonably foreseeable; thus soap not defective); cf. Presbrey, 105 Ill.App.3d at 1092, 61 Ill. Dec. at 823, 435 N.E.2d at 520 ("Regardless of plaintiff's theory [of liability], the manufacturer [of antiperspirant] owes no prior duty to warn of a risk that is remotely possible to the unknown few in the population.").

Again, the plaintiff has not attempted to persuade us that Wisconsin would not follow the general rule that plaintiffs may not hold manufacturers and sellers strictly liable for injuries resulting from idiosyncratic or rare allergic reactions. And, in a case not unlike the one before us, a district court refused to hold a shampoo manufacturer strictly liable under Wisconsin law for one plaintiff's allergic response (a severe eczema attack). The court reasoned that where "there is no danger at all to the ordinary consumer" the manufacturer ordinarily will not be held strictly liable. Mountain v. Proctor & Gamble Co., 312 F. Supp. 534, 536 (E.D.Wis.1970).7

We need not adopt the reasoning of the Mountain decision--that a manufacturer and a seller are strictly liable only to the "ordinary consumer"--to relieve Pacific World and Jewel of strict liability in this case. Even if Wisconsin were to follow other jurisdictions that appear to treat plaintiffs more favorably, there would be no basis upon which the defendants in the present case could be held strictly liable for the plaintiff's injury. The plaintiff has alleged no facts indicating that she is part of "a substantial number of the population" who are allergic to cyanoacrylate glue. Restatement (Second) of Torts Sec. 402A comment j. Nor has she shown that defendants knew or had reason to know that any users of the glue might suffer allergic contact dermatitis. See Oakes, 272 Cal. App. 2d at 650, 77 Cal.Rptr. at 713 (under strict liability, manufacturer must warn of dangers of which it "knows or should know"); Presbrey, 105 Ill.App.3d at 1091, 61 Ill.Dec. at 823, 435 N.E.2d at 520 ("The unusual susceptibility of the consumer is generally recognized as a complete defense where the manufacturer did not know and had no reason to know that a very few users of his product might be injured."); Thomas, 488 A.2d at 722 (under strict liability, there is duty to warn only of "reasonably foreseeable" dangers).

Based on the undisputed facts of this case, we agree with the district court that the defendants cannot be held strictly liable for the plaintiff's injury. Pacific World has marketed one million nail kits, and the plaintiff's injury is the only adverse reaction reported to it. At the time Pacific World marketed the nail kits, cyanoacrylate glue was not known to cause allergic contact dermatitis.8 Thus, the plaintiff has failed to show that any defect in the glue caused the loss of her fingernails or to offer any reason for imposing on either defendant a duty to warn of her unusual allergic reaction.9

The reason most often quoted to support the general rule denying recovery to unusually allergic users of products was articulated by the Supreme Court of Utah:

Every substance, including food which is daily consumed by the public, occasionally becomes anathema to him peculiarly allergic to it. To require insurability against such an unforeseeable happenstance would weaken the structure of common sense, as well as present an unreasonable burden on the channels of trade.

Bennett v. Pilot Prods. Co., 120 Utah 474, 478, 235 P.2d 525, 527 (1951). We find this reasoning persuasive and believe that the Supreme Court of Wisconsin would also. See Dippel, 37 Wis. 2d at 459-60, 155 N.W.2d at 63 ("Strict liability does not make the manufacturer or seller an insurer nor does it impose absolute liability.").

Accordingly, summary judgment in favor of Jewel and Pacific World is

AFFIRMED.

At his deposition, Dr. Hogan answered the following questions:

Q And the purpose of the patch tests is to see if there is an allergic reaction?

A Yes.

Q To the glue?

A Yes, right.

....

A To the best of my--it's mostly from our model after reviewing and talking to Cathy, how did this happen, you know, what was the time sequence of it, the reaction, it was unclear to me how this could be since she had never been previously exposed to cyanoacrylate glue, how this could be a contact dermatitis, No. 1, and No. 2, that it had occurred within seven to eight hours after initial application, and that's--

Q Too short a time?

A Too short a time. So that seems to fly in the face of conventional wisdom based on the necessity for exposure and then perhaps re-exposure for an allergic contact phenomenon to develop.

Q So based upon that understanding of the history, you would exclude an allergic reaction and therefore adopt the--

A Toxic reaction, yes. And furthermore, that would be shored up by the literature that had been placed by Alex Fischer in his textbook in which he had stated in there that allergic contact dermatitis reactions to cyanoacrylate glue had not been reported. There were no cases on record of this.

Q This Dr. Fischer reported that in his literature?

A That is in his textbook.

Q All right, and that would be--you would rely upon that textbook as authoritative?

A That is a Northern United States Reference Source for contact dermatitis.

....

Q Okay, you performed a patch test, you're satisfied from the way that test was administered that the glue that you used in that patch test was the same as the glue which Cathy Tremblay used in January of 1986?

A Uh huh.

Q That she in fact is allergic to that glue?

A Yes.

Q Have you ever heard of an allergic reaction to this type of substance before?

A No.

Q And then I take it you've never treated a patient with an allergic reaction to this type of substance?

A No.

Q And in addition to having had no prior knowledge of this type of allergic reaction, you also discussed what you called an accelerated--

A Allergic contact dermatitis, yes.

Q Which means that not only is there an allergic reaction, but it happens more rapidly?

A Uh huh.

Q Or the onset of symptoms is more rapid than what one would normally expect with some type of contact dermatitis?

A With an ordinary type of allergic contact dermatitis, yes.

Q Then this is a new phenomenon?

A It appears to be a new phenomenon, yes.

Q And you have discussed this with the North American authority on contact dermatitis, Dr. Fischer, and he confirms--

A He is also--

Q Go ahead.

A He also was as surprised at this finding as I am.

....

Q As we sit here today, what is your opinion as to the mechanism of injury in Cathy Tremblay?

A Well, I think if we have to be consistent in our belief that patch testing is at all of any relevance, the mechanism has to be allergic contact dermatitis.

Q Heretofore unknown by you or to your knowledge other members of the medical profession with respect to this particular substance?

A That's right, uh huh.

Q In Exhibit No. 7 Dr. Baumgaertner stated on page two in the first full sentence, "In either event, this is apparently an extremely rare occurrence and one that the manufacturer did not anticipate and would therefore not be liable." I'm not going to ask you if you agree with that statement in its entirety. Do you agree with that statement, "this is apparently an extremely rare occurrence"?

A Yes.

....

Q With respect to Cathy Tremblay, this is the first case of such an occurrence that you're aware of?

A Right.

Q And it's the first case that you're aware that's ever occurred within the medical literature?

A Well, yes, except for those two cases that I was made aware of on Friday.

Q Sure.

A So this would be the third case, yes.

....

Q You're not aware of any prior knowledge or reason for anybody to have given a warning with respect to the use of this type of glue relative to an allergic reaction or sensitivity, are you?

A No.

....

Q Well, your opinion is that the cause of the problem was an allergic reaction to this glue, correct?

A My opinion had been that she, yes, that she had put the nail products on and she had developed a nail problem, and my operating assumption was that this is a toxic reaction.

Q All right, you have since determined that to be incorrect?

A Yes, I have changed my mind.

Q All right, it's allergic?

A Based on new evidence, it's allergic, yes, yes.

Q All right. And you, based upon your knowledge in your area of specialty within the medical profession and your research in this case, you didn't do a patch test for some seven months after you originally saw Miss Tremblay because of your opinion an allergic contact dermatitis was not indicated?

A Allergic dermatitis had not been reported.

Q Had not previously been reported?

A Yes.

Appendix of Plaintiff-Appellant at 15-21.