PUBLISH


                IN THE UNITED STATES COURT OF APPEALS

                        FOR THE ELEVENTH CIRCUIT
                                                                   FILED
                                                            U.S. COURT OF APPEALS
                                                              ELEVENTH CIRCUIT
                                                                   08/18/99
                                    No. 99-8053                THOMAS K. KAHN
                                                                    CLERK
                        D.C. Docket No. 93-CV-2051-RLV

CATHERINE “KIP” ALLISON,

                                                        Plaintiff-Appellant,

      versus

McGHAN MEDICAL CORPORATION and
MINNESOTA MINING & MANUFACTURING
COMPANY (3M),
                                                        Defendants-Appellees.



                   Appeals from the United States District Court
                       for the Northern District of Georgia

                                (August 18, 1999)

Before COX, Circuit Judge, FAY, Senior Circuit Judge, and NANGLE*, Senior
District Judge.

NANGLE, Senior District Judge:

_______________
*Honorable John F. Nangle, Senior U.S. District Judge for the Eastern District of
Missouri, sitting by designation.
     Catherine “Kip” Allison sought recovery in district court for injuries allegedly
suffered from breast implants manufactured by McGhan Medical Corporation and

Minnesota Mining & Manufacturing Company (“3M/McGhan”). She asserted claims

in negligence, fraud/misrepresentation and strict liability/failure to warn. After

holding a three day Daubert hearing, the district court ruled inadmissible Allison’s

proffered expert testimony on causation. See Daubert v. Merrell Dow Pharm., Inc.,

(1993).       The district court granted summary judgment on the

fraud/misrepresentation claims for failure to plead with particularity and to establish

a prima facie case, and on the strict liability claim because Georgia’s statute of repose

had run. Because of Allison’s inability to establish liability without the experts, the

district court granted final summary judgment to 3M/McGhan on the remaining

negligence and failure to warn claims. After careful but deferential review, we

conclude that the district court’s Daubert rulings were correct. Because the court

properly excluded the expert testimony, we affirm its grant of summary judgment on

the negligence claims. We additionally affirm summary judgment on Allison’s other

claims.

                                   I. Background

      In December 1979, at age 21, Kip Allison decided to get cosmetic silicone

breast implants. She discussed her decision with her parents and a cousin who had

implants. She also discussed the various associated risks of the implant surgery with


                                           2
her plastic surgeon, Dr. Harvey Weiss, although the parties dispute the contents of this

discussion. Dr. Weiss implanted a double lumen design,1                         manufactured by

3M/McGhan. After complications from the initial surgery developed, the left implant

was replaced in December 1980 by a model also manufactured by 3M/McGhan.

(Appellant’s Br. at 3.)

       In 1986 Allison was diagnosed with Hashimoto’s thyroiditis (diffuse infiltration

of the thyroid gland with white blood cells, resulting in diffuse goiter).2 In 1987,

Allison was diagnosed with Type I diabetes mellitus.3 She also began experiencing

debilitating fatigue, joint, muscle and nerve pain. In 1992 Dr. Bruce Bode, an

endocrinologist, tested Allison for antinuclear antibodies and found that she had an

extremely high titer of 1:5120.4 Bode referred Allison to Dr. Sam Schatten, a




       1
        Double lumen implants have an inner silicone shell filled with silicone gel surrounded by
an outer silicone shell that is filled with saline at the time of surgery.
       2
        Stedman’s Medical Dictionary 1812 (26th ed. 1995).
       3
        Diabetes mellitus is a severe autoimmune disorder in which the body attacks the pancreas,
eventually destroying insulin producing capacity. Type I diabetes cannot be “cured,” but it can be
managed by controlling blood sugar levels through diet and insulin injections. Uncontrolled
diabetes can cause a host of serious, even fatal health problems. (Bode Dep. at 9-15.)
       4
         Antinuclear antibodies (ANA) show an affinity for cell nuclei and are found in the serum
of a high proportion of patients with systemic lupus erythematosus, rheumatoid arthritis, and certain
collagen diseases, in some of their healthy relatives, and in about 1% of normal individuals.
Stedman’s at 100. Normal titers are less than 1:40 and then progress geometrically, i.e., 1:80, 1:160,
1:320 ad infinitum.

                                                  3
rheumatologist, who diagnosed her with Sjogren’s syndrome5 and fibromyalgia.6 Dr.

Schatten reported to Dr. Bode that he did not believe that Allison’s breast implants

were a source of her medical problems. (Appellees’ Br. at 10-11.) Although Allison

does not contend that her diabetes, thyroiditis, or neuropathies were caused or

exacerbated by the implants, the parties dispute the degree of debilitation caused by

Allison’s diabetes, which 3M/McGhan alleges was severe. (Appellant’s Reply Br. at

20; Appellees’ Br. at 9-10, 14; see also Bode Dep. at 16, 22, 39-40 (describing

Allison’s diabetes as chronically poorly controlled, aggravated by bulimia, and

contributing to her chronic fatigue); Schatten Dep. vol. I at 33 (describing onset of

symptoms of daily temperature, fatigue, malaise and chills at age sixteen and arising

from diabetes.))

       Throughout 1992 Allison worsened, and although no one determined the cause

of her ailments, she decided to have her implants removed at Dr. Bode’s

recommendation. (Bode Dep. at 60.) Dr. Philip Beegle, a plastic surgeon, performed

the explantation surgery in February 1993. The pathology report stated that the outer

shell of one implant was collapsed and the other contained minimal saline. Neither



       5
         Sjogren’s causes dry eyes, dry mouth, with resulting temperature, fatigue, malaise, flu-like
feeling and occasional chills. (R. 10, Schatten Dep. vol. I at 33-34, II at 40-41).
       6
       “A disorder characterized by muscle pain, stiffness and easy fatigability. The cause is
unknown and an estimated three million are affected in the U.S.A.” The On-line Medical Dictionary
(1997-98), .

                                                 4
implant showed a loss of integrity of the inner lumen containing the silicone gel.7 The

implants were photographed and subsequently destroyed. After removal of the

implants, Allison’s non-diabetic symptoms improved. Her ANA levels steadily

declined to a level of 1:80 in October 1997. Allison reported dramatic reduction of

joint and muscle pain, and less fatigue. (Appellant’s Br. at 5; Appellees’ Br. at 8.) Dr.

Schatten again evaluated Allison in 1993, shortly after the implants were removed.

He did not change his former opinion, but advised Allison that she needed psychiatric

help. (Id. at 11.)

       Allison filed her complaint in September 1993 in the Northern District of

Georgia seeking compensation for injuries allegedly caused by defectively

manufactured breast implants. The case was transferred to the Northern District of

Alabama as part of In re Silicone Gel Breast Implants Products Liability Litigation,

(J.P.M.L. 1992), before the Honorable Sam Pointer, Jr. for pretrial

disposition in multidistrict proceedings. The case was later remanded to the Northern

District of Georgia.


       7
         Although the pathologist made this initial assessment on gross examination, Allison asserts
that the three pathologists who conducted microscopic examinations by reviewing slides of her
breast tissue found silicone in the tissue “meaning, by definition, silicone had escaped from the gel
lumen.” (Appellant’s Reply Br. at 14-15; see also R. 61, App. 42, Pathology Report; R. 37 Ex. A,
Rule 26(a)(2) Disclosure Defs.’ Expert Darryl Carter; R. 99, Shanklin Dep. vol. I at 122, 158-59.)
3M/McGhan’s own expert pathologist, Dr. Darryl Carter, noted that on one side, “there are vacuoles
with refractile non-birefringent material consistent with silicone but there is no associated
inflammation. This is consistent with gel bleed.” 

remand, Allison stated orally to the court that she was proceeding on

claims of strict liability, negligence, misrepresentation and fraud. To establish

causation in the negligence claim, Allison proposed testimony from three physicians,

Drs. Eric Gershwin, Douglas Shanklin and Sam Schatten. The district court, in

accordance with the Supreme Court’s mandate for federal judges to exercise

gatekeeping functions to determine the reliability and relevance of scientific evidence,

held a Daubert hearing. The court heard three days of evidence and argument from

both sides and waded through literally volumes of paper of the documentary record

on the science related to breast implants. After consideration of this evidence, the

court granted 3M/McGhan’s motions to exclude the expert causation witnesses on the

basis that their testimony lacked reliability and relevance under Daubert. The court

dismissed Allison’s claims for local injuries and strict liability due to Georgia’s ten

year statute of repose.8            The court additionally barred the claims for

fraud/misrepresentation and failure to warn. Because causation is an essential element

in the negligence claim, and Allison was unable to prove causation without the

experts, the court subsequently granted final summary judgment to 3M/McGhan on

all remaining claims. Allison appeals.

                                 II. Standards of Review


      8
       See discussion infra at note 17 regarding the district court’s treatment of local injuries.

                                                6
      We review the district court's grants of partial summary judgment and summary

judgment de novo, reviewing all facts and reasonable inferences in the light most

favorable to the nonmoving party, and applying the same standard as the district court.

Rodgers v. Singletary, 

, 1253 (11th Cir. 1998); Hale v. Tallapoosa

County, 

, 1581 (11th Cir. 1995). A grant of summary judgment is

appropriate “if the pleadings, depositions, answers to interrogatories, and admissions

on file, together with the affidavits, if any, show that there is no genuine issue as to

any material fact and that the moving party is entitled to a judgment as a matter of

law.” Fed. R. Civ. P. 56(c). “If the record presents factual issues, the court must not

decide them; it must deny the motion and proceed to trial.” Clemons v. Dougherty

County, Ga., 

, 1369 (11th Cir. 1982). A grant of summary judgment

may be upheld on any basis supported by the record. Fitzpatrick v. City of Atlanta,

, 1118 (11th Cir. 1993).

      The Court reviews rulings on the admissibility of expert testimony for abuse of

discretion. General Elec. Co. v. Joiner, 

, 138-39 (1997). This deferential

standard is not relaxed even though a ruling on the admissibility of expert evidence

may be outcome-determinative. 

at 142-43. “Cases arise where it is very

much a matter of discretion with the court whether to receive or exclude the evidence;

but the appellate court will not reverse in such a case, unless the ruling is manifestly


                                           7
erroneous.” 

Co. v. Edgar, 

(1878)); see also N.V.

Maatschappij Voor Industriele Waarden v. A.O. Smith Corp., 

, 418

(1978) (pointing out that Rule 702, although broadening “the range of admissible

expert testimony,” does not alter the “manifestly erroneous” standard of review). The

burden of laying the proper foundation for the admission of the expert testimony is on

the party offering the expert, and admissibility must be shown by a preponderance of

the evidence. 

n.10 (citing Bourjaily v. United States, 

, 175-76 (1987)).

                                   III. Analysis

A. Strict Liability/Failure to Warn

      The district court properly granted summary judgment on the strict liability

claims. The court found that the claims were barred under Georgia’s ten year statute

of repose which states:

      No action shall be commenced pursuant to this subsection with respect
      to an injury after ten years from the date of the first sale for use or
      consumption of the personal property causing or otherwise bringing
      about the injury.

O.C.G.A. § 51-1-11(b)(2). Subsection (c) provides an exception to the above:

      The limitation of paragraph (2) of subsection (b) of this Code section
      regarding bringing an action within ten years from the date of the first
      sale for use or consumption of personal property shall also apply to the
      commencement of an action claiming negligence of a manufacturer as
      the basis of liability, except an action seeking to recover from a

                                          8
      manufacturer for injuries or damages arising out of the negligence of
      such manufacturer in manufacturing products which cause a disease or
      birth defect, or arising out of conduct which manifests a willful, reckless,
      or wanton disregard for life or property. Nothing contained in this
      subsection shall relieve a manufacturer from the duty to warn of a danger
      arising from use of a product once that danger becomes known to the
      manufacturer.

      The district court found that the above language provides an exception to the

statute of repose for negligence actions claiming failure to warn and disease

causation, but does not create an exception for these theories under strict liability

claims. Allison argues that her claim in strict liability/failure to warn should be

allowed to stand because Georgia’s Supreme Court found that “failure to warn causes

of action [are] outside the ambit of the statue of repose . . . .” Chrysler Corp. v.

Batten, 

, 213 (Ga. 1994). Allison acknowledges, however, that the

claim at issue in Batten was negligence rather than strict liability, but argues that

“there is no good reason to believe the Court intended to differentiate between the two

types of claims.” (Appellant’s Br. at 56.)

      Allison’s argument is unpersuasive because the Batten court had no reason to

differentiate between negligence and strict liability actions when the action before it

was only in negligence. Allison has failed to cite any Georgia cases in which a claim

for strict liability/failure to warn was excepted from the statute of repose.

Furthermore, Allison ignores the clear, unambiguous language of the statute which


                                           9
exempts only negligence actions. While Allison may argue that it would have been

logical for the legislature to have extended the exception to strict liability actions as

well as negligence, it did not do so, and rewriting the statute is outside the purview of

this or the district court. Accordingly, the district court correctly granted summary

judgment barring Allison’s strict liability claims.

B. Fraud/Misrepresentation Claims

      The district court also correctly granted 3M/McGhan’s motion for summary

judgment on the fraud/misrepresentation claims. To maintain a cause of action for

deceit or misrepresentation in Georgia, the injured plaintiff must show that defendant

made a wilful or reckless misrepresentation of a material fact to induce another to act

and upon which the other acts.            O.C.G.A. § 51-6-2(a)-(b).          Where the

misrepresentation is wilfully made, privity is not necessary to give rise to the cause

of action. O.C.G.A. § 51-6-2; Robert & Co. Assoc. v. Rhodes-Haverty Partnership,

, 504 (Ga. 1983); see also, Florida Rock & Tank Lines, Inc. v. Moore,

, 837 (Ga. 1988) (delineating circumstances allowing fraud claims

when third parties have detrimentally relied). The district court found that Allison had

failed to prove two essential elements, reliance and false representation.

      Allison alleges that 3M/McGhan knew that its product lacked sufficient safety

and efficacy data and that it, as well as the industry as a whole, knew that the


                                           10
3M/McGhan double lumen implant had serious gel migrating problems. Additionally,

she alleges that 3M/McGhan engaged in attempts to conceal negative facts about its

products from the plastic surgery community, while at the same time stating in the

package inserts that breast implants are “safe.” (Appellant’s Br. at 52-54.) Allison

states that she was not informed of risks such as gel bleed, gel migration,

inflammatory response, and systemic disease, and that she would not have consented

to the surgery had she been informed of these risks. (Id. at 54-55.)

      The district court found, and we agree, that Allison’s allegations fail because

she is unable to show any reliance on the alleged misrepresentations of 3M/McGhan.

Allison conceded that she never had any contact with 3M/McGhan, but that she

detrimentally relied on the assurances of her implanting surgeon, Weiss, who was in

turn misled by 3M. The record does not support this allegation. Weiss gave

undisputed testimony that he did not rely on information from implant manufacturers

and their sales representatives. He stated that he maintained proficiency in breast

implant surgery by consulting medical journals, attending medical conferences, and

by conferring with other “well-qualified” colleagues. (Weiss Dep. at 7-8, 12-13.)

Thus, Allison cannot sustain a cause of action under a theory of third party reliance

as outlined in Florida 

, because she cannot show that

3M/McGhan fraudulently induced Weiss to act in some manner on which she relied.


                                         11
      Additionally, Allison cannot show direct detrimental reliance because she stated

she never saw or read a package insert before her surgery. (Allison Dep. at 208-09.)

The package insert itself does not describe the implants as “safe,” but states that

“[a]ugmentation mammoplasty is considered . . . to be one of the most satisfying

procedures . . .,” before describing the various risks, including capsular contracture

and leakage. (Appellees’ Br. at 67 (citing R. 61, App. 44.)) Because Allison was

unable to show any direct or indirect detrimental reliance on representations by

3M/McGhan, her fraud/misrepresentation claims must fail, as the district court

correctly found.

      The district court dismissed the fraud count because Allison failed to plead with

particularity as required by Federal Rule of Civil Procedure 9(b). The Eleventh

Circuit has held that “[w]here a more carefully drafted complaint might state a claim,

a plaintiff must be given at least one chance to amend the complaint before the district

court dismisses the action with prejudice.” Bank v. Pitt, 

, 1112 (11th

Cir. 1991). Allison, upon remand to the district court, relied on the pleading in the

“Amended Master Complaint” in the multidistrict proceedings which stated,

“Defendants made misrepresentations to plaintiff that induced her to act to her

detriment and are liable to her for their fraud.” 3M/McGhan and other defendants

filed a “Master Answer” that raised the insufficient particularity of the fraud


                                          12
allegations as an affirmative defense. The transferee court, noting the shortcomings

of the truncated fraud pleading, stated that “an amended . . . complaint . . . may be

required before a case is scheduled for trial,” but directed that “amendments prior to

that time should generally be avoided.” (Appellant’s Br. at 50 (citing R. 61, App.

37.)) At no time did Allison amend her complaint. She now complains that the

district court’s summary judgment is an abuse of discretion because she merely

followed the transferee court’s instructions, that 3M/McGhan never sought a more

definite statement, and that she should be given a renewed opportunity to plead under

the doctrine of Bank.

      The transferee court’s order stated that an amended complaint may be required

before a case is scheduled for trial. Certainly, when an opposing party has filed for

summary judgment based on a failure to plead with particularity three weeks before

the scheduled trial date, the time is ripe. Despite the instructions of the transferee

court, litigants cannot cast on the district court the burden of prodding them into filing

appropriate pleadings. 3M/McGhan clearly raised indefiniteness as an issue in its

affirmative defense in its Master Answer adopted by the court. (Id. at n.17.) Thus,

Allison’s claim that 3M/McGhan never sought a more definite statement is contrary

to the record.

      Finally, Bank allows the plaintiff to make a more definite statement “[w]here


                                           13
a more carefully drafted complaint might state a claim . . . 

.

Because Allison’s bare bones fraud allegation requires the same proof of detrimental

reliance as the misrepresentation claim, which she was unable to substantiate, the

fraud claim must fail as well, regardless of any added particularity. Consequently, we

find that the district court correctly granted summary judgment denying the

fraud/misrepresentation claims.

C. Daubert Motions

      The Court next examines the exclusion of Allison’s expert witnesses. Allison

submitted proposed testimony by three experts to prove causation. Dr. Eric Gershwin

is a board certified immunologist, Dr. Douglas Shanklin is a board certified

pathologist and Dr. Sam Schatten was Allison’s treating board certified

rheumatologist. Federal Rule 104(a) provides:

      Preliminary questions concerning the qualification of a person to be a
      witness, the existence of a privilege, or the admissibility of evidence
      shall be determined by the court, subject to the provisions of subdivision
      (b). In making its determination it is not bound by the rules of evidence
      except those with respect to privileges.

Under Rule 104(a) the parties submitted hundreds of scientific studies and journal

articles for the district court to examine. In addition, accompanying affidavits and

depositions were submitted and were before the court during the three day Daubert

hearing. We note at the outset, despite 3M/McGhan’s bandying about of terms such


                                         14
as “junk science” and “science for hire,” that the district court was careful to note the

impeccable qualifications of the experts it was reviewing and that the court had before

it sufficient record to adequately assess the Daubert issues. See City of Tuscaloosa

v. Harcros Chem., Inc., 

, 565 & n.21 (11th Cir. 1998) (finding an abuse

of discretion when the court fails to conduct a suitable inquiry into the relevant factors

to determine whether expert testimony should be admitted), reh’g denied, 

(11th Cir. 1999).

1. The Daubert Standard

      The district court properly set out the standard enunciated in Daubert and its

progeny, noting its interaction with other pertinent rules of evidence. Summarizing

the applicable rules of admissibility of scientific evidence, this Court in City of

, stated that scientific expert testimony is admissible when

      (1) the expert is qualified to testify competently regarding the matters he
      intends to address; (2) the methodology by which the expert reaches his
      conclusion is sufficiently reliable as determined by the sort of inquiry
      mandated in Daubert; and (3) the testimony assists the trier of fact,
      through the application of scientific, technical, or specialized expertise,
      to understand the evidence or to determine a fact in issue.

The Daubert analysis does not operate in a vacuum. Any proffer of scientific evidence

is also subject to other rules of evidence. 3M/McGhan therefore challenged Allison’s

expert testimony on the basis of Rules 401, 402, 403, 702 and 703.

             a. Rules 401, 402 and 403
                                           15
      Federal Rules 401 and 402 deal with the admissibility of relevant evidence.

Rule 402 allows the admission of all relevant evidence “except as otherwise provided

by the Constitution of the United States, by Act of Congress, by these rules, or by

other rules prescribed by the Supreme Court pursuant to statutory authority. Evidence

which is not relevant is not admissible.” Rule 401 defines “relevant evidence” as that

which has “any tendency to make the existence of any fact that is of consequence to

the determination of the action more probable or less probable than it would be

without the evidence.”

      3M/McGhan challenged the expert evidence on the basis of Rule 403 as well.

While the district court only mentions this rule as one of the bases that 3M/McGhan

offers for exclusion, it apparently indirectly entered the court’s consideration. Rule

403 states, “Although relevant, evidence may be excluded if its probative value is

substantially outweighed by the danger of unfair prejudice, confusion of the issues,

or misleading the jury, or by considerations of undue delay, waste of time, or needless

presentation of cumulative evidence.” The Supreme Court recognized in Daubert the

intricate role of Rule 403 in an expert testimony admissibility analysis when it noted

that expert testimony could be “both powerful and quite misleading because of the

difficulty in evaluating 

(quoting Weinstein, Rule 702 of the

Federal Rules of Evidence is Sound; It Should Not Be Amended, 138 F.R.D 631


                                          16
(1991)); see also Hall v. Baxter Healthcare Corp., 

, 1407 n.43 (D.

Or. 1996) (finding a Rule 403 analysis applicable but unnecessary in making its

decision to exclude testimony).

      Thus, while Rules 401 and 402 reflect the general policy of the Federal Rules

for liberal admission of evidence, Rule 403, working in conjunction with Rules 702

and 703, militates against this general policy by giving courts discretion to preclude

expert testimony unless it passes more stringent standards of reliability and relevance.

These stricter standards are necessary because of the potential impact on the jury of

expert testimony. While the district court did not expressly exclude any testimony on

the basis of Rule 403, we note that its consideration would only serve to buttress the

court’s ultimate exclusion of the proffered experts.

      b. Rule 702 and Daubert’s Requirements of Reliability and Relevance

      Federal Rule of Evidence 702 states, “If scientific, technical, or other

specialized knowledge will assist the trier of fact to understand the evidence or to

determine a fact in issue, a witness qualified as an expert by knowledge, skill,

experience, training, or education, may testify thereto in the form of an opinion or

otherwise.” The Supreme Court imposed a special gatekeeping role on the trial judge

in 

, to ensure that scientific evidence is both reliable and

relevant before being admitted as evidence. The Supreme Court later acknowledged


                                          17
the difficulty of this role when it stated, “Neither the difficulty of the task nor any

comparative lack of expertise can excuse         the   judge   from exercising        the

‘gatekeeper’ duties that the Federal Rules impose . . . .” 

(Breyer, J., concurring) (noting also that judges are not trained scientists). The Court

opined that scientific knowledge is far afield from the normal expertise of judges and

that they should proceed with caution lest they exceed their grasp. 

(Rehnquist, J., and Stevens, J., concurring in part and dissenting in part).

      While meticulous Daubert inquiries may bring judges under criticism for

donning white coats and making determinations that are outside their field of

expertise, the Supreme Court has obviously deemed this less objectionable than

dumping a barrage of questionable scientific evidence on a jury, who would likely be

even less equipped than the judge to make reliability and relevance determinations and

more likely than the judge to be awestruck by the expert’s mystique. Also, a judge

may enlist outside experts to assist in this sometimes very difficult decision. Using

independent court-appointed experts may serve to quell the pseudo-scientist criticism.



      Some judges, noting the general complexity of some expert evidence and in the

penultimate exercise of caution and conscience, have exercised their inherent authority

to use outside experts and have engaged in elaborate Daubert inquiries in an effort to


                                          18
sort out conflicting scientific opinions in a comprehensive search for reliability and

relevance. Judge Sam C. Pointer, the multidistrict transferee judge for federal breast

implant cases, commissioned the National Science Panel (“NSP”) under Rule 706 to

consider the scientific evidence on whether silicone breast implants cause systemic

disease. The NSP is considered a prototype investigative panel allowing the court to

escape the heated rhetoric of the courtroom and obtain a more dispassioned analytical

look at the scientific evidence with the assistance of neutral scientific experts.9 United

States District Judge Robert E. Jones in Hall v. Baxter Healthcare Corp., 

(D. Or. 1996), conducted an extensive Daubert hearing by using court-appointed

technical advisors under Rule 10410 to help evaluate the “reliability and relevance” of

the scientific evidence in seventy cases brought against implant makers.11 Because of

       9
         The NSP in its initial report of December 1998 found no definitive scientific link between
silicone breast implants and disease: “The most likely conclusion from these several analyses is that
there is no meaningful or consistent association between breast implants or silicone gel-filled
implants and any of the conditions studied.” Betty A. Diamond et al., Silicone Breast Implants in
Relation to Connective Tissue Diseases and Immunologic Dysfunction III-24 (1998) (R. 90, Ex. A).
Depositions of the NSP members are ongoing. Courts with remanded cases were not required to
await the final outcome of the panel findings, but could proceed to trial after conducting their own
Daubert hearings as the district court did in this case. MDL-926 Order No. 31 at 6.
       10
         Experts may be appointed under either FRE 104 or 706. FRE 706 requires the experts to
act as additional witnesses, and they are subject to depositions and testifying at trial. Judge Jones
appointed the experts under Rule 104 to keep them independent of ongoing proceedings. 

n.8.
       11
          Judge Jones’ ninety page opinion (including appendices) is remarkable in its depth of
inquiry and clear exposition of the issues. It is worth reflective study by serious Daubert students.
The scientists, selected from several scientific specialties, recommended that the testimony of the
plaintiffs’ experts be excluded from trial because the claim lacked scientific validity. Judge Jones,

                                                 19
the painstaking analyses which district courts undertake in making these admissibility

determinations, their efforts are well deserving of the deference that the Supreme

Court has accorded through the abuse of discretion standard enunciated in Joiner.

While some courts on occasion use evidentiary findings of other courts as precedent

when reviewing essentially the same factual issues in their Daubert decisions, the

district court here did not use other breast implant causation precedents in making its

decision. See 

n.19 (noting on remand findings of other

circuits denying admissibility of expert testimony on the issue of Bendectin causation

of limb reduction). Nor do we depend on the conclusions of other courts regarding

causation of systemic disease in breast implant recipients as precedent for affirming

the decision of the district court. We merely note in passing that other courts, after

thoroughly sifting through the scientific data, have come to the same decision, and

indeed have even excluded some of the same experts as the district court did here in

the execution of its gatekeeping role (an indication that the district court was not

operating on the outer fringe of its discretion).

       The gatekeeper role, however, is not intended to supplant the adversary system

or the role of the jury: “[v]igorous cross-examination, presentation of contrary


aligning himself with the opinions of his scientific advisors, agreed. While not required, he deferred
the effective date of his decision pending the report of the NSP. The cases were subsequently settled
before his opinion was officially adopted.

                                                 20
evidence, and careful instruction on the burden of proof are the traditional and

appropriate means of attacking shaky but admissible evidence.” 

. The judge’s role is to keep unreliable and irrelevant information from the jury

because of its inability to assist in factual determinations, its potential to create

confusion, and its lack of probative value.

      (1) The Reliability Prong

      The Daubert Court listed four noninclusive factors courts should consider in

determining reliability under Rule 702: (1) whether the theory or technique can be

tested; (2) whether it has been subjected to peer review; (3) whether the technique has

a high known or potential rate of error; and (4) whether the theory has attained general

acceptance within the scientific community. 

at 593-94. Daubert’s

flexible four-pronged analysis supplanted the longstanding “austere” Frye standard

which allowed the admission of expert testimony when it was generally accepted in

the relevant scientific community. Frye v. United States, 

, 1013 (D.C. Cir.

1923). While Allison argues that the thrust of the Rules and of the Eleventh Circuit

has been for liberal admissibility of evidence, she fails to appreciate the tempering

qualities of Rules 403, 702 and 703 under Daubert and the fact that this Circuit has

been twice overruled on Daubert decisions in precedent setting Supreme Court

decisions in Joiner and Kumho Tire, both of which imposed stricter admissibility


                                          21
standards than the Eleventh Circuit had deemed appropriate. Joiner, 

;

Kumho Tire Co. v. Carmichael,        U.S. , 

(1999).

      The district court, citing Daubert, properly recognized that the above four

factors are not exhaustive and stated that its primary focus would “be solely on

principles and methodology, not on the conclusions that they generate.” 

. Thus, the proponent of the testimony does not have the burden of proving

that it is scientifically correct, but that by a preponderance of the evidence, it is

reliable. In re Paoli R.R. Yard PCB Litig., 

, 744 (3rd Cir. 1994). The

Daubert court itself recognized that the factors it listed were a mere starting point for

a court’s analysis. Some other factors which this and other courts have considered in

the Daubert analysis are reliance on anecdotal evidence (as in case reports), temporal

proximity, and improper extrapolation (as in animal studies). Willert v. Ortho Pharm.

Corp., 

, 981-82 (D. Minn. 1998); National Bank of Commerce v.

Dow Chem. Co., 

, 1504-05 (E.D. Ark. 1996).



      (2) The Relevance Prong

      Under the second prong of Daubert, the relevance requirement, the court must

“ensure that the proposed expert testimony is ‘relevant to the task at hand,’ . . . i.e.,

that it logically advances a material aspect of the proposing party’s case.” Daubert,


                                           

(on remand). Thus, the evidence must have a valid scientific

connection to the disputed facts in the case. 

(holding

“scientific validity for one purpose is not necessarily scientific validity for other,

unrelated purposes. . . . Rule 702's ‘helpfulness’ standard requires a valid scientific

connection to the pertinent inquiry as a precondition to admissibility”). This

connection has been appropriately denominated as “fit.” 

703

      Finally, 3M/McGhan challenges Allison’s expert testimony on the basis of Rule

703 which states:

      The facts or data in the particular case upon which an expert bases an
      opinion or inference may be those perceived by or made known to the
      expert at or before the hearing. If of a type reasonably relied upon by
      experts in the particular field in forming opinions or inferences upon the
      subject, the facts or data need not be admissible in evidence.

The Daubert Court pointed out that Rule 703 applies solely to expert opinions based

on otherwise inadmissible hearsay. 

at 595. The Eleventh Circuit has

alluded to another distinction to which some courts adhere, that Rule 702 governs only

the scientist’s “major premise” (the “principle, procedure, or explanatory theory

derived by the inductive, scientific technique”), while Rule 703 addresses “the sources

the expert may consult in collecting the case-specific information to serve as the minor

premise.” Davis v. Southern Bell Tel. & Tel. Co., (S.D. Fla. Feb. 1, 1994) (citing


                                          23
Edward J. Imwinkelreid, The “Bases” of Expert Testimony: The Syllogistic Structure

of Scientific Testimony, 67 N.C.L. Rev. 1 (1988)). Both rules are implicated in

3M/McGhan’ motions.

      The district court properly outlined the above standard in its opinion, and

indeed, Allison does not dispute the court’s delineation of the standard. Rather,

Allison takes issue

with the district court’s application of the standard to her proffered experts. The Court

will therefore next examine whether the district court abused its discretion in applying

this standard to Allison’s three causation experts.

2. The testimony of Dr. Eric Gershwin

      Dr. Gershwin proposed to testify that “silicone is capable of causing systemic

harm in exposed women.” (R. 62, Ex. 4, Gershwin Decl. at 3.) More specifically, “(a)

silicone is not inert; (b) silicone can induce inflammation; (c) silicone can induce

granulomas; (d) silicone migrates; (e) silicone implants can induce autoantibody

production; (f) silicone is an adjuvant; (g) silicone can emulsify; and (h) agents which

induce chronic inflammation and granulomas are associated with systemic complaints

and disease.” (Id. at 6.) Gershwin’s ultimate conclusion was that in his opinion, “to

a reasonable degree of medical certainty, . . . silicone breast implants cause or

exacerbate systemic conditions in some women.” (Id. at 22-23.) Gershwin stated the


                                           24
bases of his opinion were his own research, clinical experience, peer reviewed

literature on silicone as an adjuvant, the Marilyn Lightfoote study, studies examining

silicone oil treatment of detached retinas, silicone related antibody studies, studies

showing biomarkers in women with implants, case reports, and epidemiological

(human statistical) studies.

      The district court found that Gershwin’s opinion was unreliable under Daubert

because of improper scientific methodology, stating that Gershwin’s theories had not

been tested, were not subject to peer review and were not generally accepted by the

scientific community. While we disagree with some of the district court’s statements

regarding peer review, we find it did not abuse its discretion in ruling Dr. Gershwins’s

testimony inadmissible.

      The district court found that the five animal studies Gershwin relied on in

addition to the Lightfoote study were inadequate to support the theory that silicone is

an adjuvant. The court noted in passing that the Lightfoote unpublished study was not

subjected to peer review. Allison establishes that Lightfoote’s oral presentation was

subject to the review of her peers. This fact, however, is no more helpful than peer

review in other forums. As Allison and the Supreme Court point out, “Publication

(which is but one element of peer review) is not a sine qua non of admissibility; it

does not necessarily correlate with reliability.” 

. Peer review


                                          25
is significant under Daubert because “scrutiny of the scientific community is a

component of ‘good science,’ in part because it increases the likelihood that

substantive flaws in methodology will be detected.” 

peer review alone was

dispositive, then the Frye standard of general acceptability in the scientific community

would have remained adequate. Consequently, a finding that Lightfoote’s animal

study was peer reviewed does not mean it constituted an adequate basis for

Gershwin’s opinion that silicone breast implants cause systemic disease.

      The court found Gershwin failed to explain the correlation of the results of

Lightfoote’s rat studies in which the rats were directly injected with silicone to

symptoms in a human patient where the inner lumen of the implants had remained

intact. Similarly, the court found that Gershwin failed to convincingly extrapolate data

from the human retinal studies to cases involving unruptured implants. The court

specifically noted extrapolation or “leap” problems with Gershwin’s collagen

antibody studies having a causal connection to systemic disease.

      Daubert decisions in other courts warn against leaping from an accepted

scientific premise to an unsupported one. Moore v. Ashland Chem. Inc., 

, 278-79 (5th Cir. 1998) (citing Wheat v. Pfizer, Inc., 

, 343 (5th

Cir.1994) (finding that physician could not show reactive airways dysfunction

syndrome (“RADS”) was caused in a patient when his exposure level to toluene was


                                          26
unknown)), cert. denied, 

(1999); Braun v. Lorillard, Inc., 

,

235 (7th Cir. 1996) (finding use of asbestos detection test for buildings improper for

detection in human tissue); 

(rejecting experts’ opinions

who relied on animal studies, chemical structure analyses, and epidemiological data

when experts failed to clearly demonstrate scientific methodology); Conde v. Velsicol

Chem. Corp., 

, 814 (6th Cir. 1994) (finding animal studies inadequate for

showing causation of disease in humans with chlordane exposure); Cavallo v. Star

Enter., 

, 769 (E.D. Va. 1995) (finding methodology of studies on

toxic effects of chemicals sound but misapplied to the case at hand), rev’d in part on

other grounds, 

(4th Cir. 1996).

      Allison complains that Dr. Gershwin did indeed explain the linkage between

the rat studies and Ms. Allison’s disease. (Appellant’s Br. at 40-43.) While the

district court noted the explanation, it was within its discretion to simply find it

inadequate. Allison reasons that the adjuvancy papers

      were not published in the animal toxicology literature and were certainly
      not intended to provide information on how to treat arthritic rats; they
      were published in peer reviewed scientific journals intended to be read
      by clinicians and others treating real people[;] thus their relevance is
      established by the very books and journals in which they appeared.

(Id. at 43.) We are fully confident that the district court understood that these studies

were not undertaken to treat silicone exposed rats. Publication in a peer reviewed


                                           27
medical journal for humans, however, does not alone establish the necessary link

required under Daubert. Cf. In re 

(explaining requirement for

proper extrapolation from animal studies to show relevance). Furthermore, Allison

does not explain why the results of these animal studies should trump more than

twenty controlled epidemiological studies of breast implants in humans which have

found no valid increased risk of autoimmune disease. See Conde, 

(finding fault with expert who neither testified to the collective view of his scientific

discipline nor explained the grounds for his differences, citing Turpin v. Merrell Dow

Pharm., Inc., 

, 1360 (6th Cir. 1992)).

      Allison complains that the district court improperly rejected the retinal studies

on the basis that no linkage was found with Allison’s unruptured implants. Allison

states that there was without dispute, silicone in Ms. Allison’s breast tissue. Even

assuming gel bleed, a finding that silicone oil emulsifies in the eye indicates that

silicone gel similarly emulsifies in breast tissue and causes systemic disease is still

quite a leap. As the Supreme Court pointed out in 

, “[I]t

was within the District Court’s discretion to conclude that the studies upon which the

experts relied were not sufficient, whether individually or in combination, to support

their conclusions” [of causation]. As in Joiner, the district court, after conducting a

thorough review of the medical evidence, did not abuse its discretion by finding that


                                           28
Dr. Gershwin failed to adequately establish the link between the animal, retinal, and

anti-collagen studies and Allison’s complaints of disease.

       The district court next took a detailed look at the four epidemiological studies

Gershwin offered to support his opinion but, in each case, the court found reasons why

these studies did not supply an adequate foundation for Gershwin’s causation opinion.

Briefly, it found the Kayler12 study unreliable because it was a re-analysis of other

studies that had found no statistical correlation between silicone implants and disease.

It found the Friis13 study irrelevant because it specifically scrutinized muscular

rheumatism, not systemic disease. Similarly, the Giltay14 study found correlations

between implants and increased risk of joint pain, a complaint which Allison did not

have. While the study did support Allison’s claim for burning eyes, the court noted

that the women participating in the study were aware of the hypothesis, a factor which

could have created bias, skewing the results and ultimately making its conclusions

suspect. The court found that the Hennekens15 study, which had the most significant


       12
       L.K. Kayler et al., Breast Implants Increase the Risk of Arthralgias: An Epidemiological
Meta-Analysis, 43 J. Investigative Med. 129 (1995).
       13
         S. Friis et al., Connective Tissue Disease and Other Rheumatic Conditions Following
Breast Implants in Denmark, 39 Annals Plastic Surgery 1 (1997).
       14
        Eric J. Giltay et al., Silicone Breast Prostheses and Rheumatic Symptoms: A Retrospective
Follow-Up Study, 53 Annals Rheumatic Diseases 194 (1994).
       15
        Charles H. Hennekens et al., Self-Reported Breast Implants and Connective-Tissue
Diseases in Female Health Professionals: A Retrospective Cohort Study, 275 JAMA 616 (1996).

                                               29
statistical correlation of silicone and increased ANA, had a relative risk of only 1:24,

a finding so significantly close to 1.0 that the court thought the study was not worth

serious consideration for proving causation.16

       The court found that Gershwin’s proposed four studies were in direct contrast

to over twenty other epidemiological studies which found no statistical correlation

between silicone breast implants and systemic disease, strong evidence that a

consensus exists in the general scientific community that no correlation exists.

Allison complains that the district court erred by looking at conclusions rather than

methodology and principles as Daubert directed. While weighing the relative findings

of the studies may seem to be a resurrection of the Frye standard (general acceptance

in the scientific community), courts have noted that Daubert’s suggested criteria to

examine whether the theory has attained general acceptance within the scientific

community, 

-94, does just that. Joiner made it clear that

although principles and methodology were the focus, the court was not precluded from


       16
          The threshold for concluding that an agent more likely than not caused a disease is 2.0. A
relative risk of 1.0 means that the agent has no causative effect on incidence. A relative risk of 2.0
thus implies a 50% likelihood that the agent caused the disease. Risks greater than 2.0 permit an
inference that the plaintiff’s disease was more likely than not caused by the agent. Federal Judicial
Center, Reference Manual on Scientific Evidence 168-69 (1994). Allison points out that the
relative correlation of cigarette smoking to heart disease is only 1.5, and yet “[n]o one would
seriously quarrel that smoking can cause heart disease.” (Pl.’s Reply Br. at 5 n.6.) We note,
however, that this risk is more that twice that found in the Hennekens study, and we do not think the
district court abused its discretion in finding a 1.24 risk minimal in terms of causation. Moreover,
showing association is far removed from proving causation.

                                                 30
looking at conclusions:

      [C]onclusions and methodology are not entirely distinct from one
      another. Trained experts commonly extrapolate from existing data. But
      nothing in either Daubert or the Federal Rules of Evidence requires a
      district court to admit opinion evidence which is connected to existing
      data only by the ipse dixit of the expert. A court may conclude that there
      is simply too great an analytical gap between the data and the opinion
      proffered.

; see also Lust v. Merrell Dow Pharm., Inc., 

, 598

(9th Cir. 1996) (noting that a court may properly scrutinize anomalous conclusions

and reject expert opinion if the expert fails to identify and defend the reasons why his

scientific methodologies yielded novel results). We find that the district court did not

abuse its discretion by considering that the proffered conclusions in studies with

questionable methodologies were out of sync with the conclusions in the

overwhelming majority of the epidemiological studies presented to the court. We

additionally note that the district court lists Allison’s four problematic epidemiological

studies as only one of many factors for ultimately rejecting Gershwin’s testimony.

      Allison states that because these studies were published in peer reviewed

journals, then ipso facto their methodology has been determined sound. But as

regarding the Lightfoote study, while peer review increases the

likelihood that substantive flaws in methodology will be detected, scrutiny by one’s

peers does not insure admissibility. Again, it is well established that “[p]ublication


                                           31
. . . is not a sine qua non of admissibility.” 

.

      Allison complains that the district court failed to consider Gershwin’s testimony

that silicone can induce chronic inflammation, and chronic inflammation is associated

with systemic disease. Allison states that Dr. Gershwin’s opinion in this area is basic

textbook medicine, is unassailable and was excluded sub silentio by the district court.

While the district court did not specifically single out Gershwin’s chronic

inflammation theory for comment, we do not conclude that it failed to consider the

testimony. The court’s global conclusions regarding Dr. Gershwin’s opinions, that

they were untested and that extrapolations from animal studies were inadequate, is

reasonable in light of Dr. Shanklin’s testimony that “pathologists generally, as well

as myself, are still learning the full implications of [chronic inflammatory problems]”

and that no one to his knowledge had made the “connection” in the peer reviewed

literature between silicone induced chronic inflammation and systemic disease.

(Shanklin Dep. vol. I at 68-71.)

      Allison also states that the district court should not have found Gershwin’s

reliance on case studies improper methodology.          While we acknowledge the

importance of anecdotal studies for raising questions and comparing clinicians’

findings, in the face of controlled, population-based epidemiological studies which

find otherwise, these case studies pale in comparison. See 

                                          32
(finding that “case reports and case studies are universally regarded as an insufficient

scientific basis for a conclusion regarding causation because case reports lack

controls”; hence, they do not supply scientific knowledge upon which an opinion can

be based under Daubert); Casey v. Ohio Med. Prods., 

, 1385 (N.D.

Cal. 1995) (finding case reports do not provide reliable scientific evidence of

causation). The district court did not abuse its discretion by discounting Dr.

Gershwin’s reliance on case reports in the face of the overwhelming contrary

epidemiological evidence presented.

       While the court stated that Gershwin’s studies had not been subjected to peer

review, this factor has bare mention in the court’s analysis of Gershwin’s testimony.

We find, in contrast to the district court, that many of Dr. Gershwin’s theories had been

subjected to peer review. Dr. Gershwin is a prolific scientific author and has published

numerous articles in peer reviewed journals, and he himself is a peer reviewer. (R. 33, Pl.’s

Submission of Expert Reports, Ex. 2.) However, the parties dispute to what extent his

premise that breast implants cause systemic disease has been subjected to the relevant

scientific community for review. (See Pl.’s Reply Br. at 20; Defs.’ Br. at 49-51.) “Under

the regime of Daubert . . . a district judge asked to admit scientific evidence must determine

whether the evidence is genuinely scientific, as distinct from being unscientific speculation

offered by a genuine scientist.” Rosen v. Ciba-Geigy Corp., 

, 318 (7th Cir.1996).

While we do not question the scientific expertise of Dr. Gershwin, we find the district court

                                             33
correctly excluded his testimony on the individual and collective bases noted above. Even

assuming that Dr. Gershwin’s work had been subjected to the most rigorous scrutiny

by the scientific community, this factor would not nullify the court’s findings of

unreliable foundation, inadequate extrapolation, the lack of human models and “fit.”

See 

(“‘Fit’ is not always obvious, and scientific validity for

one purpose is not necessarily scientific validity for other, unrelated purposes.”).

Consequently, we find that the district court did not abuse its discretion in excluding

Dr. Gershwin’s testimony.

3. Testimony of Dr. Douglas Shanklin

      Dr. Shanklin, a pathologist, proposed testimony regarding his conclusions after

microscopic examination of Allison’s tissue slides. (Pl.’s Resp. Defs.’ Mot. Exclude

Ev. at 197.) Dr. Shanklin would have testified that (1) a positive result to a “silicone

sensitivity test” (“SST”) is evidence that silicone causes disease; (2) chronic

inflammation around a breast implant and granulomas found in breast capsule tissue

trigger autoimmune process; (3) crystalline silica can be identified in breast tissue by

a light microscope; and (4) crystalline silica in breast capsule tissue leads to

autoimmune disease. The district court excluded Dr. Shanklin’s opinions stating they




                                           34
were based on unreliable methodologies.17

       Allison complains that the district court rejected Shanklin’s entire testimony on

the basis of its rejection of his “silicone sensitivity test,” which Shanklin did not

intend to testify about because Allison had never had the test. Nevertheless, Allison

defended the merits of the SST during the Daubert hearing (Daubert Hearing

Proceedings Tr. vol. V at 13-20), even though she denied its applicability in her

response to the Daubert motions. Because the SST was put in issue, the district court

did not abuse its discretion by considering its reliability when evaluating Dr.

Shanklin’s proffered testimony.18 But even if this test had not been a point of

       17
          While the Court might have entertained partial allowance of Shanklin’s testimony as
causation evidence of local injury, claims of local injury have been abandoned on this appeal. The
parties dispute the actual ruling of the district court on whether chronic inflammation as a local
injury was barred by the statute of limitations. Appellee’s Br. at 4; Appellant’s Reply Br. at 24. The
Court need not address the lower court’s ruling, however, because the matter is first mentioned in
passing by Allison in her reply brief. Issues that are not clearly outlined in an appellant’s initial
brief are deemed abandoned. Federal Sav. & Loan Ins. Corp. v. Haralson, 

, 373-74 n.3
(11th Cir. 1987); 9 Jeremy C. Moore et al., Moore’s Federal Practice ¶ 228.01 (2d ed. 1985).

       Allison’s entire argument is that chronic inflammation is a cause of systemic disease.
Oblique references to local injury in the initial brief are clearly inadequate to preserve the
issue. See In re Trans World Airlines, Inc., 

, 132-33 (3rd Cir. 1998) (finding
broad and slight references inadequate to meet the substantive function of Fed. R. App. P.
28 requiring issues to be squarely addressed). Furthermore, this Court typically takes a dim
view of second guessing the strategies of litigants, refusing to grant “relief they did not
request, pursuant to legal theories they did not outline, based on facts they did not relate.”
Adler v. Duval County Sch. Bd., 

, 1481 n.12 (11th Cir. 1997), reh’g en banc
denied, 

(11th Cir. 1997).
       18
         Dr. Shanklin developed the test himself in conjunction with another physician. A
positive SST is purportedly evidence of an immune response to silicone. The test’s reliability
has been called into serious question by other scientists. See V. Leroy Young, Testing the Test:

                                                 35
contention between the parties, we do not think it an abuse of discretion for the court

to consider the general merits of an experts’ work in the field in which he will be

offering testimony, even if he will not be speaking specifically to each point the court

scrutinizes. The district court pointed out that Dr. Shanklin’s silicone sensitivity test

was flawed because it used crystalline silica rather than silicone and that he had failed

to produce any studies supporting his theory that silicone in the body breaks down to

silica and then acts as an antigen. His silicone conversion hypothesis does not have

support in the scientific literature.

        Other evidence in the record which supports the district court’s wholesale

exclusion of Dr. Shanklin’s testimony is that certain aspects of Allison’s purported

condition had not been documented through available medical tests.                               Allison

complained of muscle pain or myalgia, yet this clinical state, defined by Allison’s

subjective complaints, was never verified by a muscle biopsy, so that a diagnosis of

myositis, or inflammation of the muscles, was unsubstantiated.19 (Shanklin Dep. vol.

I at 74.) Therefore, Dr. Shanklin’s comments regarding causation of Allison’s


An Analysis of the Reliability of the Silicone Sensitivity Test (SILS) in Detecting Immune-Mediated
Responses to Silicone Breast Implants, 97 Plastic & Reconstructive Surgery 681 (1996) (discussing
the general merits of the test; the test provided invalid results in six out of six individuals, indicating
that it could not distinguish between persons with silicone breast implants and persons who never
had silicone breast implants).
        19
         The theory is that breakdown products from the implant get into muscle tissue causing
direct chemical toxicity. (Shanklin Dep. at 117.)

                                                    36
myalgia have dubious weight.

        Dr. Shanklin also indicated that his theories regarding chronic inflammation and

systemic disease are in their infancy. In response to a question regarding whether

chronic inflammation leads to multiple myositis, he could not confirm that it was

established in the literature: “Many papers have small details which are not in the

abstract. But the process is set up. The process is known to work as a matter of

making the final connection, and that work is going on as we speak.” (Id. at 78.) As

, Dr. Shanklin testified that “pathologists generally, as well as myself, are

still learning the full implications of [chronic inflammatory problems],” and to Dr.

Shanklin’s knowledge, no one had made this connection in peer reviewed literature.

(Id. at 68-71.)

        Dr. Shanklin was also questioned about silicone antibodies, cytokines,20 and

direct chemical toxicity, the mechanisms Shanklin stated led to the development of

systemic disease in breast implanted women. (Id. at 118.) Dr. Shanklin emphatically

stated that the mere presence of antibodies does not mean that a person will get sick



        20
         Dr. Shanklin’s theory is that chronic inflammation sets the stage for systemic disease–the
affected cells release cytokines, which in turn lead to multiple myositis, a diagnosis that was not
confirmed in Allison. (Shanklin Dep. vol. I at 88-89). Cytokines themselves indicate an activation
of the immune mechanism. Because of their short half-life, they are very difficult to measure in
humans. Although cytokines have been measured in some breast implanted women, this testing
appears to be in its initial stages and therefore is not routinely performed at this time. (Id. at 91-93.)
Hence, cytokine measurements were not done on Allison.

                                                   37
and that Allison had not been tested for any specific antibodies to silicone. (Id. at 140,

113.) When asked if he was aware that “the general consensus of the relevant

scientific community has been that there is no antibody produced in response to

silicone,” Shanklin stated that “[t]he relevant community is just now beginning to

weigh in on the subject.” Furthermore, he stated that a group of scientists in Great

Britain that had rejected his findings regarding immune responses were “simply

wrong.”21 As to cytokines, he stated, “We are about to break into the dawning era of

medicine by cytokine analysis. You are asking for data which won’t be around for ten

to fifteen years.” (Id. at 96-97.) Regarding direct chemical toxicity, also referred to

as silicone toxicity or siliconosis, Shanklin noted that it was “another mechanism

which can be invoked about which little is known presently,”22 even though he stated

Allison had the disease. (Id. at 80.)

       While Dr. Shanklin’s theories may be proven in the future, and although he has



       21
          See Silicone Gel Breast Implants: The Report of the Independent Review Group (Jill
Rogers Ass’n ed, July 1998) (“Following a careful consideration of the histopathological material
provided by Professor Shanklin, the IRG did not agree with him that any of the changes seen
constituted evidence of an immune response. In particular, neither vasculitis (inflammation of the
blood vessels, indicative of immunological involvement) nor any other histological change
suggesting an immune response could be seen in the tissues examined.”). 

(Shanklin Dep.
vol. II, Ex. 11.)
       22
         Shanklin remarked that he did not know whether Allison has “a supplemental direct
chemical toxicity or not. Nobody has tested her for that, and very little is known from the research
angle yet. That question won’t be answered for another five to seven years.” (Shanklin Dep. vol.
I at 117-18).

                                                38
strong beliefs regarding silicone related pathology, we find that his testimony is based

more on personal opinion than on scientific knowledge.23 In light of Dr. Shanklin’s

own admissions, we find that the district court did not abuse its discretion in finding

that his testimony was unreliable,24 was not generally accepted by the scientific

community, and was unsupported by other studies. Allison simply does not prove the

reliability of Dr. Shanklin’s testimony by a preponderance of the evidence. Therefore,

exclusion of Dr. Shanklin’s testimony was not error.

4. Testimony of Dr. Sam Schatten

       Dr. Schatten proposed to testify that the implants exacerbated Allison’s

fibromyalgia and Sjogren’s syndrome, resulting in her fatigue. He would state that

“there is a small subset of women with breast implants who become afflicted with

rheumatological disease.” (Pl.’s Resp. Defs.’ Mot. Exclude Causation Test. at 206.)

Additionally, he would testify that “there was a general consensus in the medical

community with which he was familiar that breast implants caused disease in ‘a


       23
         While scientific testimony need not be known to a certainty, Daubert does require
that assertions be derived from “scientific knowledge.” “Scientific” means proper grounding
in the methods and procedures of science, or the “scientific method.” “Knowledge” is more
than subjective belief or unsupported speculation, but “applies to any body of known facts
or to any body of ideas from such facts or accepted as truths on good grounds.” 

.
       24
         As with Dr. Gershwin, the Court notes that Dr. Shanklin has published many articles in
peer reviewed journals, yet this alone does not substantiate the scientific validity of his premise that
Allison’s silicone implants caused systemic disease.

                                                  39
percent of patients who have systemic rheumatic diseases.’” (Id.) Schatten recanted

his former opinion, that he did not believe that Allison’s breast implants were a source

of her medical problems, which he formed when Allison consulted him shortly after

the explantation surgery.

       Schatten’s revised opinion was made five years after the explantation surgery,

on the basis of Allison’s medical record and laboratory results and without benefit of

a follow-up examination and after being approached by Allison’s attorney. On the

basis of information supplied by the attorney regarding the explantation surgery, the

lowered ANA lab reports, and reported improvements in Allison’s fibromyalgia and

fatigue, Dr. Schatten stated that there was a possibility/probability25 that “[Allison’s]

breast implants have played a role [in her symptomatic improvements], period.”

Schatten Dep. at 75. Allison contends that Schatten relied on information he gained

by treating her, case reports and peer reviewed literature, as well as the information

supplied by her lawyer. The district court rejected Schatten’s testimony on the basis

that Schatten’s testimony was (1) inadmissible under Georgia law because he could

testify only to a probability rather than a possibility of causation; (2) his testimony

failed a Daubert analysis because it was tainted by reliance on unreliable sources; and


       25
          In Schatten’s deposition, he made approximately nine statements regarding the mere
possibility of causation and exacerbation countered by one “probability” statement in response to
a leading question by Allison’s counsel.

                                               40
(3) his conclusions were made in preparation for litigation.

      Allison argues that the district court erred by excluding Dr. Schatten’s

testimony based upon its erroneous application of Georgia law at this stage of the

proceedings and its erroneous finding that the testimony would be inadmissible under

Georgia law. Because this action is based on diversity, Georgia substantive standards

of law must apply. Erie Railroad Co. v. Tompkins, 

(1938). Proffered

expert testimony must meet the legal as well as the substantive issues of the case. See

In re Breast Implant Litig., 

, 1226 (D. Colo. 1998) (factoring in

Colorado’s standard of proof of “reasonable probability” in requiring a relative risk

ratio above 2.0 for epidemiological causation); 

(on remand)

(considering substantive tort requirements under California law in assessment of

whether expert evidence met Daubert’s “fit” or helpfulness prong).        The rule in

Georgia is

      [i]n cases that involve issues of causation which, by the nature of the
      situation, can be resolved solely by expert medical evidence standing
      alone, . . . the evidence must naturally be based at least on reasonable
      probability. “It appears to be well settled that medical testimony as to
      the possibility of a causal relation between a given [negligent act] and
      the subsequent [injuries alleged to have been caused by the negligence]
      is not sufficient, standing alone, to establish such relation.”

Maurer v. Chyatte, 

, 545 (Ga. App. 1985) (citations omitted). Allison

correctly states that the standard of proof in a civil case is preponderance of the


                                          41
evidence and that “reasonable medical probability” is the functional equivalent of

preponderance of the evidence. O.C.G.A. § 24-1-1; 24-4-3. Also, a plaintiff may

present medical as well as non-medical evidence to show causation. Estate of

Patterson v. Fulton-Dekalb Hosp. Auth., 

, 236 (Ga. App. 1998). That

breast implants can and did cause systemic disease in Allison is not a natural inference

that a juror could make through human experience. McDaniel v. Employers Mut.

Liab. Ins. Co., 

, 804 (Ga. App. 1961). Thus, medical expert testimony

was essential to prove causation in this case. Compare Georgia Cas. & Sur. Co. v.

Jernigan, 

, 616-17 (Ga. App. 1983) (refusing to strike expert’s

“possibility” opinion in its entirety because it was offered in conjunction with other

non-expert evidence authorizing a finding of the requisite causal connection in the

case).

         Allison’s causation evidence, however, consisted of testimony by three medical

experts, two of which were already excluded in the court’s opinion. Therefore,

Schatten was more than a “piece of the puzzle.” As the sole remaining causation

expert, it was not error for the district court to consider Georgia’s rule requiring

statement of his opinion to a reasonable degree of medical or scientific certainty (or




                                           42
by a preponderance of the evidence).26 Failure to meet this burden means that the sole

causation expert’s opinion would not assist the trier of fact under Rule 702 because

his degree of certainty would not be sufficient to establish probable cause and would

thus be irrelevant.

       Although Allison additionally argues that the court erred by applying summary

judgment standards to exclude Dr. Schatten’s testimony, we find this argument

irrelevant considering the fact that no other causation experts remained.27 Dr.

Schatten’s possibility testimony is not only excludable under Georgia law, but also

falls short of the standards for proving medical causation under Daubert because of

its lack of “fit.” Consequently, the court did not abuse its discretion by excluding Dr.

Schatten’s testimony which was based on mere possibility of causation.

       Additionally, the court found other independent grounds for excluding Dr.

Schatten’s opinions. The district court determined that Dr. Schatten’s opinions were

based on unreliable methodology. Dr. Schatten admitted that he had never tested his

theory that implants exacerbate Sjogren’s syndrome and that the scientific literature


       26
         Again, because local claims have been 

n.17, the Court does not entertain
an outcome premised on the possibility of partial allowance of Shanklin’s testimony as it related to
claims of local injury. Nevertheless, Schatten’s testimony is excludable on two other grounds, as
demonstrated infra.
       27
         In Hall, 

n.28, the court pointed out the distinctions of an admissibility
inquiry under Rule 702 that considers a “more likely than not” state law standard and a sufficiency
determination for summary judgment.

                                                43
does not support his theory. (Schatten Dep. vol. I at 94-95; II at 103-04.) He also

admitted that his exacerbation theory had never been peer reviewed and that the

scientific literature, except for case reports, does not support a relationship of any kind

between breast implants and Sjogren’s syndrome or fibromyalgia. (Id. vol. I at 92; II

at 29, 34.) Because the untested theories of Allison’s experts are not generally

accepted in the scientific community, they obviously have a high potential rate of

error.

         Finally, the court did not abuse its discretion by finding that Dr. Schatten’s

opinion was prepared in preparation for litigation. While Allison argues that

Schatten’s revised opinion was based on his examination, diagnosis and treatment of

her supplemented by information developed since her last visit, it was within the

district court’s purview to determine that such an about face in Schatten’s opinion

occurred because the opinion was developed in preparation for litigation. Also, his

reversal in opinion occurred without benefit of a follow-up examination.

         The mere coincidence of temporality of the dropping ANAs and Allison’s

subjective reports of decreased fatigue after explantation are questionable bases for

Dr. Schatten to reverse his prior opinion that was grounded on two clinical visits with

his patient. See In re Breast Implant 

(citing a string of

cases finding that temporality is not evidence of causation); In re 

                                            44
(finding self reports of symptoms in patients preparing for litigation unreliable when

not verified by other means). While courts frequently find that expert testimony such

as Dr. Schatten was planning to render is acceptable, even though it is solicited for

litigation and well rewarded, we agree that the court did not abuse its discretion in

scrutinizing the reversal of Schatten’s former opinion and the grounds upon which it

was reversed. We therefore find that it was within the district court’s discretion to

have excluded Dr. Schatten’s testimony for any one of the three reasons it offered.

      Courts have found that an abuse of discretion occurs when under Daubert the

admissibility bar is too high. Ruiz-Troche v. Pepsi Cola., 

, 85 (1st Cir.

1998). Defendants naturally favor strict admissibility standards, while plaintiffs argue

for more liberal standards:

      Trial judges must exercise sound discretion as gatekeepers of expert
      testimony under Daubert. [Defendant], however, would elevate them to
      the role of St. Peter at the gates of heaven, performing a searching
      inquiry into the depth of an expert witness's soul--separating the saved
      from the damned. Such an inquiry would inexorably lead to evaluating
      witness credibility and weight of the evidence, the ageless role of the
      jury.

McCullock v. H.B. Fuller Co., 

, 1045 (2nd Cir. 1995). Striking the

appropriate balance may sometimes be a difficult task. The Daubert remand court

stated that the gatekeeping task was particularly daunting when the dispute, as in this

case, “concerns matters at the very cutting edge of scientific research, where fact


                                          45
meets theory and certainty dissolves into probability.” 

.

Despite the difficulty of this case, we find that the district court suitably exercised its

discretion in excluding these experts; it found multiple grounds for their exclusion,

and on each point its rationale was based on a careful evaluation of the record in light

of the Daubert standards working in conjunction with other Federal Rules of

Evidence. There was no error.

D. Summary Judgment on Remaining Negligence Claims

       Finally, the district court did not err by granting summary judgment on the

remaining negligence claims based on failure to warn and disease causation. As

, Georgia’s ten year statute of repose created an exception for these

claims. Because the court found Allison’s causation testimony inadmissible, and such

testimony was essential to maintaining the negligence claims, Allison was unable to

assert a prima facie case. Therefore, summary judgment was proper.28

       While other courts, such as Hall and In re Breast Implant Cases, 942 F. Supp

958 (E.&S.D.N.Y. 1996) postponed motions for summary judgment for their own

reasons, we see no need for postponement in this case. As Daubert recognized,

       [s]cientific conclusions are subject to perpetual revision. Law, on the
       other hand, must resolve disputes finally and quickly. The scientific


       28
       As 

in note 17, the Court will not consider whether it was error to grant final
summary judgment on potential local injury claims because Allison abandoned those claims.

                                                 46
      project is advanced by broad and wide-ranging consideration of a
      multitude of hypotheses, for those that are incorrect will eventually be
      shown to be so, and that in itself is an advance. Conjectures that are
      probably wrong are of little use, however, in the project of reaching a
      quick, final, and binding legal judgment--often of great
      consequence--about a particular set of events in the past. We recognize
      that, in practice, a gatekeeping role for the judge, no matter how flexible,
      inevitably on occasion will prevent the jury from learning of authentic
      insights and innovations. That, nevertheless, is the balance that is struck
      by Rules of Evidence designed not for the exhaustive search for cosmic
      understanding but for the particularized resolution of legal 

. While we acknowledge that the debate regarding systemic disease

and silicone products may be ongoing for years to come, we concur with the district

court that final summary judgment is appropriate at this time and with these experts.

                                   IV. Conclusion

      We AFFIRM the district court’s ruling for summary judgment for 3M/McGhan

on the strict liability/failure to warn and fraud/misrepresentation claims. After careful

but deferential review, we AFFIRM the district court’s Daubert rulings excluding

Allison’s causation experts, finding that Allison has failed to show that the decision

is manifestly erroneous. Because the court properly excluded the expert causation

testimony, we AFFIRM its grant of final summary judgment to 3M/McGhan on the

remaining negligence claims.




                                           47