This appeal requires us to review evidentiary rulings made during a jury trial of a statute of limitations issue in an asbestos case. That trial was held pursuant to a mandate of this court reversing earlier summary judgments for the defendants on limitations grounds and remanding for further proceedings to resolve disputes of fact material to the limitations issue. Cowgill v. Raymark Industries, Inc., 780 F.2d 324 (3rd Cir.1985) (Cowgill I ). Much of the excluded evidence in question was offered in support of a theory that was raised for the first time on remand and that was inconsistent with the facts found by the trial court in the initial proceedings and left undisturbed on appeal. The remaining evidence at issue was excluded under Fed.R.Evid. 401 and 403. Because we conclude that the plaintiff was barred from litigating her new theory and because we find no fault with the trial court's disposition of the remaining issues, we will affirm.

From 1926 until his retirement in 1961, George Cowgill worked as a pipefitter at Mobil Oil Company's Paulsboro, New Jersey refinery. In 1979, Mr. Cowgill was contacted by Mobil and asked to return to the Paulsboro refinery for a physical examination, a request he honored on November 16, 1979. Cowgill returned for further tests on November 29 and then again on November 30. He was examined on the latter two dates by Dr. James A. Barnshaw, a Mobil physician. Dr. Barnshaw diagnosed Cowgill as having "evidence of pleural thickening indicative of asbestos exposure." App. at 742. Whether and to what extent Barnshaw communicated this finding to Cowgill is disputed.

On April 30, 1980, Mr. Cowgill returned once again to the Paulsboro refinery, this time to participate in a group discussion about asbestos-related ailments. Dr. John McNally, who led the discussion, had reviewed Mr. Cowgill's file and concurred with Dr. Barnshaw's diagnosis of pleural thickening. Dr. McNally's records suggest that he may have relayed his conclusion to Mr. Cowgill, although this matter is in dispute. The parties do agree that Dr. Barnshaw wrote to Mr. Cowgill on November 20, 1981 advising him that Mobil's studies showed "minimal pleural thickness, most likely secondary to asbestos exposure" and "no other asbestos related findings." App. at 883.

On September 30, 1982, Mr. Cowgill was diagnosed as having lung cancer. He died on March 3, 1983.

Mrs. Cowgill filed this suit on August 26, 1983 in the United States District Court for the Eastern District of Pennsylvania, raising claims under New Jersey's Wrongful Death and Survival Acts. Her complaint alleged in part:

8(a) During the aforementioned years of employment [with Mobil], decedent was caused to be exposed to raw asbestos fiber and/or finished asbestos products of defendants which exposure caused him to contract asbestos related diseases and/or injury to his body systems, lungs, respiratory system, heart and other organs. As a result of said exposure, the said decedent was caused to suffer ...:

asbestosis, carcinoma of the lung

* * *

* * *

(b) It is averred that each said asbestos related disease was diagnosed on:

asbestosis--November 20, 1981

carcinoma of the lung--September 30, 1982

(c) It is averred that plaintiff's decedent knew of each such diagnosis on or about each of the above respective dates.

In their answers, the defendants, Raymark Industries and numerous other producers and distributors of asbestos, asserted that Mrs. Cowgill's claim was barred by a two year Pennsylvania statute of limitations. 42 Pa.Cons.Stat.Ann. Sec. 5524(2) (Purdon Supp.1985). They correctly pointed out that under Pennsylvania law all asbestos-related injuries constitute a single cause of action and the limitations period is triggered with respect to that cause of action when the first of a series of asbestos-related injuries is or reasonably should have been discovered. Ross v. Johns-Manville Corp., 766 F.2d 823 (3d Cir.1985) and cases cited therein.

After completion of discovery, the parties filed a Joint Pretrial Order in which the positions of the parties with respect to the date upon which Mr. Cowgill first learned he had asbestosis were described as follows:

[Plaintiff]:

George Cowgill knew he had an asbestos-related disease or illness in November of 1981 when Dr. Barnshaw informed him for the first time that he was suffering from asbestosis. George Cowgill had no reason to suspect that he suffered from an asbestos-related condition before that time.

[Defendant]:

George Cowgill was informed that he had an asbestos-related health condition on or about November 30, 1979 by Dr. Barnshaw.

Thereafter the defendants brought on motions for summary judgment based on their limitations defenses. Mrs. Cowgill resisted these motions on the grounds (1) that New Jersey law governed the limitations period and the point at which her cause of action accrued, and (2) that although Mr. Cowgill had an asbestos-related injury more than two years before the filing of suit, he neither knew nor should have known about that injury until November of 1981. The court, however, concluded that Pennsylvania law governed and that "it [was] factually indisputable that a reasonably diligent person [in Mr. Cowgill's position] would have attained knowledge of his injury and the cause thereof by November 30, 1979." 780 F.2d at 329. Accordingly, summary judgment was granted for the defendants.

A first timely appeal followed (Cowgill I ). This court first affirmed the district court's conclusion that Pennsylvania law governed the limitations issues. Turning to those issues, we accepted "as true Mr. Cowgill's assertions in his December 1982 deposition that ... no one informed him of his asbestosis until November 1981." 780 F.2d at 329. We further noted that Mrs. Cowgill did "not deny that, as of that date, decedent was aware of his asbestos-related injury." Id. at 326. We found conflicting evidence, however, on whether a reasonably diligent individual should have learned of his asbestos-related injury and its cause more than two years before the filing of the complaint. Id. at 330-1. We therefore remanded the case with explicit instructions regarding the issue to be determined in connection with defendant's limitations defense:

We conclude that issues of material fact remain in dispute and that, on this record, viewed in the light most favorable to appellant, it cannot be said as a matter of law that decedent should have known of his asbestos-related injury more than two years prior to the filing of appellant's suit. The grant of summary judgment for defendants must therefore be vacated and the matter remanded in order that a jury may determine when Mr. Cowgill knew or reasonably should have known of his injury and its cause.

Id.

Thus from the filing of the complaint through the first appeal, Mrs. Cowgill acknowledged that her husband had an asbestos-related injury more than two years prior to the institution of suit but insisted that he neither learned nor should have learned of it before November of 1981. As a result, her legal position with respect to limitations was that her cause of action accrued in November of 1981. At no time did she suggest that no injury occurred until a later date or that her cause of action did not accrue until a later date.

On remand, the district court severed the limitations issue for a separate jury trial. During the pre-trial proceedings on that issue, Mrs. Cowgill asserted an alternative theory for the first time. She advised the court that she intended to proceed on the theory that the pleural thickening could not be attributed to asbestos exposure and that the first time Mr. Cowgill knew he had an asbestos-related injury was when he was diagnosed as having cancer in March of 1983. App. at 62-63. Mrs. Cowgill proposed to call several witnesses in support of this contention including Dr. Barnshaw and Dr. Weiss, a medical expert engaged by Mobil to study Mr. Cowgill's file and x-rays.1 The trial judge excluded the testimony of these witnesses, however, concluding that the issue at the limitations trial should be limited to the issue remanded to the district court by this Court.

A seven-day trial followed at which Mrs. Cowgill was limited to the introduction of evidence tending to show that her husband, despite the exercise of reasonable diligence, knew nothing of his pleural thickening until November 20, 1981. Mrs. Cowgill failed to persuade the jury on this issue. In response to a special interrogatory, the jury determined that Mr. Cowgill had received information no later than April 30, 1980 "sufficient to establish that he knew or should have known that he had an asbestos-related injury." App. at 642. Based on this determination, the district court concluded that Mrs. Cowgill's cause of action accrued more than two years prior to the filing of suit and entered judgment for the defendants.

On this second appeal, Cowgill II, Mrs. Cowgill contends that the jury should also have been asked to determine whether Mr. Cowgill in fact had an asbestos-related injury at any time prior to August 27, 1981, two years before the filing of the complaint. She insists that if the jury had been able to hear the excluded testimony, they might well have determined that there was no asbestos-related injury, and, accordingly, no cause of action until some point within two years of the initiation of suit. Mrs. Cowgill also maintains that the district court erred in excluding certain other evidence relevant to what Mr. Cowgill knew or should have known about the condition of his lungs prior to August 27, 1981. We will examine each issue in turn.

When the district court entered its original judgment in favor of the defendants on limitations grounds, Mrs. Cowgill became collaterally estopped to contend that her husband had no asbestos-related injury prior to August of 1981. As she acknowledges in the briefing on this appeal, a necessary predicate of the court's conclusion that the limitation period had run prior to the filing of the complaint was a factual finding that Mr. Cowgill had an asbestos-related injury prior to that time. As soon as that legal conclusion was embodied in a final judgment, Mrs. Cowgill was bound by that finding at least in the absence of extraordinary circumstances rendering that result a manifest injustice. See, e.g., Huron Holding Corp. v. Lincoln Mine Operating Co., 312 U.S. 183, 61 S. Ct. 513, 85 L. Ed. 725 (1941); Restatement (Second) of Judgments Sec. 13, comments (f) and (g) (1982). The question thus becomes whether anything that happened in Cowgill I or the proceedings following remand lifted the constraint against Mrs. Cowgill pressing her new theory. We conclude that the constraint remains.

In Cowgill I, this court accepted the undisputed fact that Mr. Cowgill had pleural thickening more than two years before suit was filed and Mrs. Cowgill's allegation that this injury was asbestos-related. We also affirmed the district court's legal conclusion that Mrs. Cowgill's cause of action was barred if her husband knew or should have known of that injury more than two years prior to suit. Indeed, with one exception, we affirmed all aspects of the findings and conclusions supporting the district court's judgment and we remanded for a determination of a single narrow issue--when Mr. Cowgill knew or should have known of his injury and its cause.

When a court of appeals reverses a judgment and remands for further consideration of a particular issue, leaving other determinations of the trial court intact, the unreversed determinations of the trial court normally continue to work an estoppel. 1B J. Moore, J. Lucas & T. Currier, Moore's Federal Practice p 30.416, p. 517 (3d ed. 1984). When the estoppel is operative in proceedings in the same case on remand, courts frequently speak in terms of the law of the mandate or the law of the case rather than collateral estoppel but the underlying principle is the same. Todd & Co., Inc. v. S.E.C., 637 F.2d 154 (3d Cir.1980) (when an appellate court affirms in part and reverses in part, all issues necessarily disposed of in the affirmance become law of the case even though the case is remanded for proceedings on other issues). "The judicial system's interest in finality and in efficient administration" dictates that, absent extraordinary circumstances, litigants should not be permitted to relitigate issues that they have already had a fair opportunity to contest. Id. at 156.2

Mrs. Cowgill had a full and fair opportunity to litigate the issue of whether her husband had an asbestos-related injury prior to November of 1981 and she chose to concede that he did, contesting only whether he knew or should have known of its existence. Having made that election and pursued her theory of limitations through an appeal, considerations of finality and efficiency mandate that she be barred from relitigating the limitations issue on a different theory factually inconsistent with the one she elected to pursue. We note, for example, that if Mrs. Cowgill, in response to the defendants' limitations defense, had contended from the outset that no asbestosis injury existed prior to August of 1981, it is probable that there would have been no Cowgill I. If she had pointed at the outset to admissible evidence tending to show an absence of injury prior to that date, as she attempted to do after remand, the district court presumably would have denied summary judgment on the ground that there was a material dispute of fact regarding the existence of an injury. Having occasioned an appeal which on her present theory of the case was unnecessary, she is not entitled to ignore the consequences of those proceedings.

The situation that Mrs. Cowgill has created resembles in many ways the situation that confronted the Court of Appeals for the Ninth Circuit in Waggoner v. Dallaire, 767 F.2d 589 (9th Cir.1985), cert. denied, 475 U.S. 1064, 106 S. Ct. 1374, 89 L. Ed. 2d 601 (1986). In that case, the trustees of a multi-employer pension trust sued, in 1977, an employer that had ceased contributing to the trust fund in July of 1970. During the original district court proceedings, the employer stipulated that its president had falsely told the trustees prior to the cessation of contributions in 1970 that it had ceased being an active employer. After a bench trial, the court entered judgment for the defendant employer, basing that judgment upon a number of grounds including a finding that the trustees' claim was barred by a four year statute of limitations. On appeal, the Court of Appeals reversed and remanded for further proceedings, holding in part that the false report of the employer's becoming inactive had tolled the statute of limitations. On remand, the district court permitted the employer to withdraw its stipulation regarding the false report and the case was retried. Once again the district court entered judgment for the employer in part based on the conclusion that the trustees' claim was barred by limitations. The Court of Appeals again reversed, holding:

The law of the case controls unless the first decision is clearly erroneous and would result in manifest injustice, there has been an intervening change in the law, or the evidence on remand is substantially different. Planned Parenthood, 718 F.2d 938 at 949 (9th Cir.1983).

Dallaire [the employer] argues that the substantially different evidence exception applies here.... Withdrawal of the stipulation, Dallaire argues, made the evidence on remand substantially different and justified the district court's departure from the law of the case. Dallaire's argument overlooks the fact that the withdrawal of the stipulation was itself error. The stipulation and our first decision on its legal effect were part of the law of this case. The trial court thus could not permit Dallaire to withdraw the stipulation unless Dallaire presented evidence to the court showing that withdrawal of the stipulation was necessary to prevent manifest injustice. Nowhere in Dallaire's argument to this court did he indicate that he presented any such evidence. Thus we must conclude that the district court erred in permitting withdrawal of the stipulation. See McGregor Boulevard Church of Christ v. Walling, 428 F.2d 401, 404 (5th Cir.1970).

On remand, the trial court should only have considered " 'matters left open by the mandate of this court.' " Moore, 682 F.2d at 834 (quoting Quern v. Jordan, 440 U.S. 332, 347 n. 18, 99 S. Ct. 1139, 1148 n. 18, 59 L. Ed. 2d 358 (1979))....

767 F.2d at 593.

Here, as in Waggoner, Mrs. Cowgill tendered no evidence to the district court during the remand proceedings suggesting that release from her prior concession was necessary to prevent a manifest injustice. During the initial district court proceedings she had a full opportunity to conduct discovery and she became aware of most of the evidence that she relied upon in the remand proceedings to support her new theory. With respect to the remainder of that evidence she has suggested no reason why it or comparable evidence3 could not have been obtained by her before the original summary judgment was entered. Thus, this is not a case in which "newly discovered" evidence became available after the original decision was made. Similarly, this is not a situation in which the law changed materially after the initial decision. Nor does it come within any other category of case we have recognized as exceptions to the doctrines of law of the case and collateral estoppel. See Hayman Cash Register Co. v. Sarokin, 669 F.2d 162 (3d Cir.1982); Schultz v. Onan Corp., 737 F.2d 339, 345 (3d Cir.1984). In short, we perceive no reason why Mrs. Cowgill, after having had a full and fair opportunity to litigate the limitations issues at the trial level and through an appeal in this court, should be entitled to start afresh in the district court with a new theory fundamentally at odds with her prior position.4

Mrs. Cowgill's remaining arguments challenge the exclusion of evidence arguably relevant to the issue of whether her husband knew or should have known about his asbestos-related injury prior to August of 1981.

Dr. Barnshaw testified that he had no present recollection of his contacts with Mr. Cowgill. He indicated that his testimony regarding those contacts was based upon his records and his "general practice." App. at 777-779.5 With respect to the specific issue of what he told Mr. Cowgill on November 30, 1979, Mr. Cowgill's "Employee Medical Records" contained the following notation over Dr. Barnshaw's initials:

Consult. Discussed whether he has evidence of asbestosis--Told that he has mainly evidence of COPD [chronic obstructive pulmonary disease] probably 2o [secondary] to smoking--Does show evidence of pleural thickening indicative of asbestosis exposure.

App. at 742, 893. Not surprisingly, Dr. Barnshaw's testimony about what he told Mr. Cowgill rested primarily on this record notation.

Dr. McNally indicated that his testimony was based on his review of medical records regarding Mr. Cowgill. With respect to the notification issue, he relied upon his own notation in the records concerning a conference with Mr. Cowgill on April 30, 1980 when Mr. Cowgill was "informed again of his findings of COPD and pleural asbestosis." App. at 389, 894. On cross-examination, Mrs. Cowgill's counsel inquired of Dr. McNally, "In fact, at Mobil Oil, wasn't there a policy between 1979 and 1981 to not inform employees of what their findings were, although the forms indicated that you had told them?" App. 471-472. Dr. McNally denied that such a policy existed.

It was in this context that Mrs. Cowgill tendered and the trial judge excluded "rebuttal" testimony of four Mobil employees who were prepared to testify that they were not informed by Mobil of asbestos-related injuries. Nothing in the tender indicated that any of the four had medical records stating that they had been told of asbestos-related disease. Nor were any of them represented to have a medical situation similar to that of Mr. Cowgill. They were described only as Mobil employees having some form of asbestos-related injury. In this context, we find no fault with the ruling of the trial court.

To the extent Mrs. Cowgill sought to rely on a policy of concealment as affirmative evidence tending to make it more likely that her husband received no notice until 1981, the evidence was properly excluded because nothing in the tender indicated that the witnesses were qualified to testify as to Mobil's policy or habitual practice and because, as the trial court ruled, this was not rebuttal evidence.

To the extent Mrs. Cowgill sought to rely upon these four witnesses to undermine the credibility of the Mobil physicians, her argument is of more substance but we cannot say that the trial court abused its discretion in excluding the proffered testimony. As Mrs. Cowgill stresses, Dr. Barnshaw affirmatively relied on his "general practice" to support his testimony about what happened during his contact with Mr. Cowgill, and testimony of specific instances of concealment would have tended to refute any suggestion that there was a general practice of disclosure. As the trial judge noted, however, Dr. Barnshaw's testimony on the specific issue of what he told Mr. Cowgill on November 30, 1979 was based on his specific record notation as to what he said to Mr. Cowgill. Refutation of whether there was a "general practice" of disclosure would be of relatively little assistance to the jury with respect to this particular issue. On the other hand, as the trial court also observed, admission of the testimony of the plaintiffs four witnesses would have entailed "mini-trials" of each of their cases and of numerous other witnesses whom the defendants indicated they would call to testify that they had been told by the Mobil doctors of their asbestos-related injuries. Accordingly, the trial judge concluded that the value of the proffered testimony was outweighed by the diversions that pursuing these collateral issues would occasion. He did not abuse his considerable discretion under Fed.R.Evid. 403 in doing so. See, e.g., United States v. Long, 574 F.2d 761, 767 (3d Cir.), cert. denied, 439 U.S. 985, 99 S. Ct. 577, 58 L. Ed. 2d 657 (1978).

Mrs. Cowgill's final objection arises from the trial court's treatment of another facet of Dr. Barnshaw's testimony. Dr. Barnshaw's testimony was presented in the form of a videotaped deposition taken on October 16, 1986, seven years after his examination and diagnoses of George Cowgill. During this deposition, Mrs. Cowgill's counsel suggested to Dr. Barnshaw that the findings he made in November 1979 were "soft," and, at his prompting, Dr. Barnshaw agreed with the characterization. Counsel then elicited from Barnshaw an attempt to distinguish "soft" findings from "hard" findings, as relates to Cowgill's pleural thickening condition:

Q. And how do you--just so our terms are clear, a soft finding is a finding that is capable of more than one interpretation; isn't that right?

A. I think soft finding in this situation relates to a finding such as minimal pleural thickening in which one person may read it as being present, another may read it as not being present, and it's hard to know what it truly represents.

Q. And your testimony today is you would attempt to give to Mr. Cowgill as complete a picture as your understanding of the findings as possible; is that right?A. As I have said before, I would in dealing with minimal pleural thickening, point out that it's hard to know whether it truly is there or not. If it is there, and if it's real, it could be related to asbestos exposure as a very early change.

App. at 836-837.

It is to the trial court's exclusion of this testimony that Mrs. Cowgill objects, insisting that the testimony was critical to her case in light of several Pennsylvania cases suggesting that a tentative diagnosis, one unlikely to induce an individual of ordinary diligence to pursue his or her claim, does not commence the limitations period. See e.g. Trieshock v. Owens Corning Fiberglass Co., 354 Pa.Super. 263, 511 A.2d 863 (1986); Cathcart v. Keene Indus. Insulation, 324 Pa.Super. 123, 471 A.2d 493 (1984).6 Mrs. Cowgill insists that the tendered testimony would have suggested that Dr. Barnshaw's original diagnosis was equivocal.

The relevant issue, however, was what Dr. Barnshaw told Mr. Cowgill on or about November 30, 1979. Although Dr. Barnshaw's thoughts about the reliability of his original diagnosis seven years after the diagnosis was made may not be devoid of probative value in the context of that issue, we cannot fault the trial court for concluding that Dr. Barnshaw's comments were more likely to confuse or mislead the jury than to elucidate their analysis, a danger heightened by the focus on inherently opaque terms such as "soft findings" and "hard findings." Given the trial court's broad discretion under Rule 403 to balance probative value against the possibility that proffered evidence will prove confusing, we conclude that the trial court's exclusion of this portion of Dr. Barnshaw's testimony did not constitute error. See, e.g., United States v. Long, 574 F.2d 761, 767 (3d Cir.), cert. denied, 439 U.S. 985, 99 S. Ct. 577, 58 L. Ed. 2d 657 (1978) ("If judicial self-restraint is ever desirable, it is when a Rule 403 analysis of a trial court is reviewed by an appellate tribunal."); United States v. Robinson, 560 F.2d 507, 514 (2d Cir.1977), cert. denied, 435 U.S. 905, 98 S. Ct. 1451, 55 L. Ed. 2d 496 (1978).

The judgment of the district court will be affirmed.