This appeal requires us to determine whether there was sufficient evidence presented at the trial of this medical malpractice action to warrant submission to the jury. Ruling that plaintiffs' case lacked the necessary evidence to establish the negligent act as the proximate cause of the injuries sustained, the trial judge granted a directed verdict under Rule 50 in favor of the defendant hospital.1 Before reaching the question of the sufficiency of the evidence, it is necessary to trace the wife-plaintiff's2 medical history prior and subsequent to the alleged act of negligence.

Testimony at trial disclosed that plaintiff was a forty-nine year old woman, five feet three inches tall and weighing over two hundred pounds. Prior to her admission to the defendant hospital in June, 1961, she had been under medical care for the better part of the last twenty years and in that period had undergone some remarkable surgical experiences: a removal of a kidney in 1947, an adult tonsillectomy in 1949, three separate gastric or duodenal ulcer operations in 1953 or thereabouts, and a dilation and curettage (D&C) in 1960.

The hospitalization giving rise to the present action was for the purpose of a vaginal wall repair and hysterectomy. During this latter operation on June 6, the right uterine artery tore loose from a surgical clamp. This necessitated still another operation, a laparotomy, an incision into the abdominal wall and cavity, in order to reach the uterine artery. Post-operatively, there was a packing of the vagina to control bleeding, but this, in turn, caused a bowel obstruction which was relieved by the insertion of a rectal tube. In addition, an infection developed in the surgical wound after the June 6th operation, which infection was described by the patient's doctor as one 'expected under the circumstances.'3

On June 17 the abdominal incision opened, causing a part of the intestinal tract to protrude through the incision. This complication required further surgical intervention consisting of a left rectal incision to repair the defect in the abdominal wall. A second infection developed, this time in the incision of June 17. Again her physician ascribed its cause to the same factors which caused the similar infection which appeared after the first operation.

Aside from these incidents of post-operative infection, the patient was undergoing an uneventful recuperation when suddenly, on July 2, she began to hemorrhage from the vagina. Fearing a fatal exsanguination, the hospital authorities immediately rushed her to the operating room--not sparing the time to put her into a transfer stretcher. A person in street clothes assisted in this process, helping to move the bed along the corridors and into the operating room where a surgical team went into immediate action. Determining that it was critical to find the cause of the bleeding, the surgeons again entered the abdominal cavity, and 'after a number of hours of searching' they finally tied off the bleeding blood vessel which had caused the hemorrhage.4

A week after this operation a third infection appeared, this time in the incision of July 2. This infection required the usual attention of draining, dressing, cleansing, and removing necrotic tissue. Thus, from June 6 through July 2, plaintiff underwent three separate abdominal operations, each of which was followed by a post-operative infection of the surgical wound. It is the presence of the infection following the third operation which forms the basis of the action against the hospital.

It is the theory of the plaintiff that the defendant hospital was negligent in permitting a person in unsterile street clothes to help wheel her into the operating room5 and that this violation of hospital regulations caused the post-operative infection.

At trial, plaintiff called two medical experts who were asked what caused this particular infection. Dr. Helene Young testified that by the late date she examined the patient, it was impossible to determine the cause. The second physician, Dr. Bernard Siegel testified on direct examination that it could have been caused by 'a number of things. I think the loss of blood, the many operations she had, the blood in the abdominal cavity as a result of the ruptured blood vessel, the fact that a loop of intestines had to be removed which exposed some contents of the intestinal tract; all these things reduce the resistance of tissue to inflammation and infection and thereby reduce their healing powers.'

These were the only two medical witnesses asked an opinion as to the cause of the infection. Neither attributed the cause, whether probably or possibly, to the activity of a person in street clothes wheeling the plaintiff into the operating room on July 2. The hospital laboratory report was introduced, showing the presence of 'staphylococcus' in the infection, which report, according to Dr. Siegel, if 'read * * * correctly, * * * show(ed) that the amount of staphylococcus present was minimal'. There was no evidence produced at trial describing the cause or effect of this particular strain of organism.6

Despite this lack of probative medical testimony, the appellants contend that once it was established that a person in unsterile clothing was present in the operating room, followed one week later by an indication of some staphylococcus in the surgical wound, sufficient evidence was presented for the jury to infer the nexus between the event and the subsequent infection. The District Court disagreed; it held that under these circumstances some expert testimony was necessary to establish a relationship between the alleged cause and effect.7

We perceive the presence or absence of medical expert testimony to be subordinate to the fundamental question involved: was there a sufficiency of evidence of any kind to merit submission of the cause-effect relationship to the jury?

In a laudatory effort to narrow the issues before us, the parties attempt to turn this decision on whether state or federal standards are to be employed to test the sufficiency. The appellant contends that federal standards control, and that under these standards sufficient evidence was present for the jury to establish the necessary causal relationship. The appellee insists that under the Erie doctrine we must apply the law of the forum, Pennsylvania, and under the practice of that state the case did not qualify for submission to a jury.

In the absence of a definitive statement by the Supreme Court as to what standards should apply in a diversity case, the circuits have adopted varying positions.8 This disparity among the circuit decisions has been reflected in this court. The earlier cases uniformly held that the question of sufficiency of the evidence was controlled by state law. This was probably because the litigants generally assumed the application of state law in their presentations to the court. But where the question was raised, earlier decisions uniformly held state law to be controlling. Zentz v. Buchman, 103 F.2d 850 (3 Cir. 1938); Warlich v. Miller, 141 F.2d 168 (3 Cir. 1944). In Waldron v. Aetna Casualty & Surety Co., 141 F.2d 230 (3 Cir. 1944), the court was explicit in ruling that:

'While questions of evidence ordinarily relate to matter of procedure, the sufficiency of the evidence goes to the maintenance of the substantive right and is, therefore, to be tested by local law in cases where such law controls.' 141 F.2d at 234.

Similarly, in Bastian v. Baltimore & Ohio R. Co., 144 F.2d 120 (3 Cir. 1944), we affirmed the application of a state rule of evidence on the grounds that it was 'in reality decisional law applied by the courts of Pennsylvania which affects the substantive rights of the parties'.

Contemporaneously with the development of this state-law sufficiency test, however, was the parallel enunciation of the concept that federal law controlled the division of functions between judge and jury in diversity actions. Thus, in Ettelson v. Metropolitan Life Ins. Co., 137 F.2d 62, 65 (3 Cir. 1943), it was concluded that 'well considered federal decisions have held that the division of function between court and jury in a federal court is to be made by federal, not state law'.

Ostensibly, it appeared that a dichotomy had developed in the decisions. On one hand it was held that state law controlled the question of whether sufficient evidence had been presented to justify the submission of a diversity action to the jury, while on the other hand it was determined that federal law controlled the division of judge-jury functions.

In 1958 the validity of this latter proposition--that federal law governed the judge-jury relationship in diversity actions--was confirmed by the Supreme Court in Byrd v. Blue Ridge Rural Elec. Corp., Inc., 356 U.S. 525, 78 S. Ct. 893, 2 L. Ed. 2d 953. Conceding that the Erie rule generally required the application of state law to diversity suits, the Court recognized that there were exceptions to this concept:

'But there are affirmative countervailing considerations at work here. The federal system is an independent system for administering justice to litigants who properly invoke its jurisdiction. An essential characteristic of that system is the manner in which, in civil common-law actions, it distributes trial functions between judge and jury and, under the influence--if not the command--of the Seventh Amendment, assigns the decisions of disputed questions of fact to the jury.'

Byrd thus made it clear that state law could not contravene the federal rule that only the jury may resolve disputed questions of fact. But the Byrd decision did not resolve the parallel problem of what law controlled the question of the sufficiency of the evidence required to submit a case to the jury, since this has always been regarded as a question of law for the court and not the jury. This is amply demonstrated by a comment contained in the Byrd decision in which the Court expressly distinguished its prior decision in Stoner v. New York Life Ins. Co., 311 U.S. 464, 61 S. Ct. 336, 85 L. Ed. 284 (1940). To appreciate fully the thrust of the comment, it is necessary to review the facts in Stoner.

In Stoner, the plaintiff had recovered a jury verdict against his insurer on a disability claim. On appeal, the state appellate court affirmed the award, holding that sufficient evidence had been presented by the plaintiff to have the case submitted to the jury. Thereafter, the insurer brought suit in federal court to test the claim of continuing disability. On substantially the same evidence as was presented in the state court, the district court held that sufficient evidence of disability was presented. The Court of Appeals reversed, holding that the evidence was insufficient as a matter of law to establish disability.

Reversing the Court of Appeals, the Supreme Court held that the federal courts were bound by the state determination that there was sufficient evidence on which to find disability.

Concerning Stoner, the Court in Byrd declared:

'Stoner v. New York Life Ins. Co., 311 U.S. 464, 61 S. Ct. 336, 85 L. Ed. 284 (1940), is not contrary. It was there held that the federal court should follow the state rule defining the evidence sufficient to raise a jury question whether the state-created right was established.'9

Since Byrd, the Supreme Court has made no further pronouncements in this area. In both Dick v. New York Life Ins. Co., 359 U.S. 437, 79 S. Ct. 921, 3 L. Ed. 2d 935 (1959) and Mercer v. Theriot, 377 U.S. 152, 84 S. Ct. 1157, 12 L. Ed. 2d 206 (1964), resolution of the issue was expressly declined. In the wake of this reluctance, our decisions have displayed aberrant tendencies. In Lind v. Schenley Industries, Inc., 278 F.2d 79 (3 Cir. 1960), we reiterated that 'the principle of Erie R. Co. v. Tompkins does not determine the division of functions between court and jury.' But in Pritchard v. Liggett & Myers Tobacco Co., 295 F.2d 292 (3 Cir. 1961), where the plaintiff contended that the test of the sufficiency of the evidence was controlled by federal law, the question was left unanswered. However, in Sleek v. J. C. Penney, 324 F.2d 467 (3 Cir. 1963), it was stated that: 'we are still of the view that in a diversity case where the sufficiency of the evidence goes to the maintenance of the substantive right, as here, the law of the state must prevail.' 324 F.2d at 471. See also Berwanger v. Delaware, Lackawanna & Western R. Co., 290 F.2d 588, 589 (3 Cir. 1961).

A contrasting view has been presented more recently in Woods v. National Life and Accident Insurance Co., 347 F.2d 760 (3 Cir. 1965) and Walsh v. Miehle-Goss-Dexter, Inc., 378 F.2d 409 (3 Cir. 1967). In Walsh, the court found that the question of whether the evidence 'was such as to warrant either the direction of a verdict or entry of judgment notwithstanding the verdict * * * is one which must be decided under federal law'.

Against the backdrop of these precedents, we are reluctant to embrace the contentions of the present litigants that 'the matter has been clearly settled'. It is true that certain consistencies have developed. Primary among these is the postulate that the federal forum will enforce its own standards in determining the judge-jury functions. But as the litany of past precedents attests, that postulate has not foreclosed the question of whether state or federal law shall determine the specific requirements for the establishment of a given prima facie case.

Recognizing the uncertain status of the law on this point, we have undertaken an examination of both the federal and state standards on the evidence necessary to establish medical causation. This examination convinces us that insofar as the present case is concerned they are substantially similar. Accordingly, we defer the choice of action urged upon us by the parties, and turn instead to the substance of the appeal.

Because this is an appeal from a directed verdict for the defendant, we must examine the record in a light most favorable to the plaintiff, and review the specific evidence in the record and all inferences reasonably capable of being drawn therefrom. We must determine whether, as a matter of law, the record is critically deficient of that minimum quantum of evidence from which a jury might reasonably afford relief. Continental Ore Co. v. Union Carbide & Carbon Corp., 370 U.S. 690, 82 S. Ct. 1404, 8 L. Ed. 2d 777 (1962). Lebrecht v. Bethlehem Steel Corp., 402 F.2d 585 (2 Cir. 1968); Schad v. Twentieth Century-Fox Film Corp., 136 F.2d 991 (3 Cir. 1943); Rawleigh, Moses & Co. v. Kornberg, 210 F.2d 176 (8 Cir. 1954).

Whether the plaintiff has met the burden of proof in the sense of a 'risk of non-persuasion of the jury' is, as its name commands, a matter to be determined by the jury. But the preliminary question, whether the quantity of evidence is fit to be considered by the jury, is a question of law; it is today a function of the trial judge; and it always has been so, in England and in our federal and state courts.

We find it necessary to reiterate this rather obvious conclusion because we have detected an increasing tendency on the part of Pennsylvania litigants invoking our diversity jurisdiction to place undue reliance on the case of Smith v. Bell Telephone Company of Pennsylvania, 397 Pa. 134, 153 A.2d 477 (1959) for the proposition that once some evidence, even a mere scintilla, is adduced at trial capable of generating some hypothetical inference, them the matter qualifies for consideration by the jury. We recognize Smith to be a landmark case in the law of that state; it abolished a previous requirement that proof, in cases based on circumstantial evidence, be so conclusive as to exclude any other reasonable inference inconsistent therewith, and pronounced a new standard that if 'there be sufficient facts for the jury to say reasonably that the preponderance favors liability', it is no longer necessary that 'every fact or circumstance point unerringly to liability'.

We do not construe this case to impinge upon the traditional responsibility of the trial judge to decide, as a matter of law, whether a prima facie case has been established. Although increasing the aggregate of inferences permitted to be drawn by the jury in Pennsylvania courts, the Smith case nevertheless reiterated certain minimum requirements: (1) although the jury may draw inferences based on all the evidence and the jurors' own knowledge and experiences, it may not be permitted to reach its verdict on the basis of speculation and conjecture; (2) 'it is the duty of the plaintiff to produce substantial evidence * * * upon which logically (the jury's) conclusion may be based';10 (3) there must be 'sufficient facts for the jury to say reasonably that the preponderance favors liability'.11

Smith therefore dictates that the trial judge must determine as a minimum requirement of a prima facie case whether the plaintiff has produced 'substantial evidence'12 of 'sufficient facts' having the capacity of supporting a logical conclusion.

Based on a review of the record and the opinion of the court below, we must first conclude that a jury could well have found that the hospital violated its regulations in the treatment of Catherine Denneny in permitting invasion of the operating room by one in unsterile clothes. That violation could have, in turn, led the jury to conclude that the conduct of the hospital's staff amounted to 'negligence'. The critical question, however, was whether there was sufficient evidence for the jury to reach a logical conclusion that this particular act was the proximate cause of the post-operative infection suffered by the patient.13

The trial judge ruled that 'expert medical testimony was required in this case to establish the cause of the infection complained of'. That conclusion is substantially supported by both the decisions of the Pennsylvania courts, under which law this diversity suit arose, and the decisions of this court in similar situations.14 The present case is not within that particular class of cases where the injury complained of is so inextricably connected with the acts of negligence, in time or effect, as to make the use of expert medical testimony superfluous.

Such were the circumstances in Schultz v. City of Pittsburgh, 370 Pa. 271, 88 A.2d 74 (1952), where the plaintiff fell through an open manhole and injured his back. In holding that the use of expert medical testimony in such circumstances was not necessary, the Pennsylvania Supreme Court stated:

'Where 'there is such close connection between the accident and the injury as to satisfy a reasonable person as to the cause of the injury, the relation between the two is sufficiently shown."15

The same result was reached by the Supreme Court in Sentilles v. Inter-Caribbean Shipping Corp., 361 U.S. 107, 80 S. Ct. 173, 4 L. Ed. 2d 142 (1959), a case involving the application of federal standards of evidence under the Jones Act. There it was held that the failure of medical witnesses to testify on the issue of causation did not impugn 'the jury's power to draw the inference that the aggravation of petitioner's tubercular condition, evident so shortly after the accident, was in fact caused by that accident * * *.'

Those circumstances are not present in the case bofore us. It cannot reasonably be said that the onset of post-operative infection was so inextricably connected with the act of negligence alleged on July 2 as to make a conclusion of causation logical; that Catherine Denneny suffered post-operative infections following her operations of June 6 and June 17, both immediately preceding the July 2 experience, strongly militates against such finding. As this court stated in Dixon v. Pennsylvania R. Co., 378 F.2d 392 (3 Cir. 1967), under similar circumstances where the issue of disability was in issue; 'Here, the jury would, at best only be able to hazard a guess, rather than draw a logical inference.'

The heavy reliance placed by plaintiff on Restatement, Torts 2d 432(2),16 is similarly of no avail unless proper evidence is submitted permitting the logical inference that the force, actively operating because of the actor's negligence, was sufficient to bring about the specific harm to another. Stripped of its surplusage, all we have in this case is testimony that 'a bug' caused the infection which appeared in the post-operative wound a week after the July 2 surgery. There was no evidence as to what specific 'bug' caused the infection. Although a laboratory report revealed the presence of 'staph', the doctor called by plaintiff explained that this report 'if read correctly * * * showed that the amount of staphylococcus present was minimal'. There was no testimony of the effect of staphylococcus, its nature, quantity, or necrotic effect, if any.17 In short, the word 'staph' appeared on a microbiology report as being found in a smear taken from the surgical wound. The same report disclosed the presence of other substances. No attempt was made by the plaintiff to equate the presence of these substances with the onset of infection, other than the single question directed to Dr. Siegel--and his answer eliminated it as a factor. No attempt was made to establish that any of the organisms found in the laboratory specimen could have resulted from the momentary presence in the operating room of a person in street clothing.

Indeed, other than the introduction of a hospital record composed in the typically mesozoic scrawl of the medical profession, which was admitted without any explanation of the entries, neither the court nor the jury were provided with any testimony describing the characteristics or consequences of the 'bug'. When this evidence, or lack thereof, is considered in the context of the patient's history of two post-operative infections occurring prior to the alleged negligent act, its basic deficiency becomes obvious.18 We must conclude that there was not substantial evidence of sufficient facts from which the jury could have drawn logical conclusions based on their own knowledge and experience. On such evidence as was presented, the jury's verdict could only have been the product of speculation and conjecture; neither federal nor state courts will permit this.

The judgment of the District Court will be affirmed.