dissenting.
Plaintiff appeals from the grant of defendants’ motion for summary judgment.
Appellant Holley contends that the trial court erred in that summary judgment cannot issue based solely upon opinion evidence, since the trier of fact is not bound by opinion testimony, citing Ginn v. Morgan, 225 Ga. 192 (167 SE2d 393) (1969), and Harrison v. Tuggle, 225 Ga. 211 (167 SE2d 395) (1969). He asserts that there is but one exception to this general principle: summary judgment may issue based on opinion evidence in professional malpractice cases where there is no counter-expert opinion evidence presented. He has too narrowly defined the exception, which is stated in Howard v. Walker, 242 Ga. 406 (249 SE2d 45) (1978): “[I]n those cases where the plaintiff must produce an expert’s opinion in order to prevail at trial, when *367the defendant produces an expert’s opinion in his favor on motion for summary judgment and the plaintiff fails to produce a contrary expert opinion in opposition to that motion, then there is no genuine issue to be tried by the jury and it is not error to grant summary judgment to the defendant.” Id. at 408. Although Howard v. Walker was a legal malpractice case, the language of the opinion does not restrict the applicability of the exception to only legal malpractice cases. Moreover, as defendants point out, the courts in applying the Howard v. Walker exception have not limited it to malpractice causes of action. It has been applied in types of cases where there was an issue that required expert medical opinion.
In Carter v. Landy, 163 Ga. App. 509 (295 SE2d 177) (1982), the grant of partial summary judgment for defendant based entirely on opinion evidence in a false imprisonment action was affirmed. The issue was whether a physician exercised reasonable medical care in effectuating a patient’s transfer to a hospital for evaluation. In Jordan v. United Ins. Co., 158 Ga. App. 520 (281 SE2d 286) (1981), the exception was applied to a contract action against an insured for refusal to compensate the plaintiff for loss of a leg. The court held: “[I]nasmuch as expert opinion testimony was essential to a determination of the causal factors in the gangrenous condition of plaintiff’s leg, the expert testimony of plaintiff’s attending physician was sufficient to support the grant of defendant’s motion for summary judgment.” Id. at 521. Additionally, in Tony v. Pollard, 248 Ga. 86 (281 SE2d 557), a case involving a will contest, the Supreme Court cited Howard v. Walker, supra, in affirming the grant of summary judgment based on the undisputed medical opinion testimony relating to the testamentary capacity of the testatrix and the voluntariness of the execution of the will. “Hence the rule that opinion evidence cannot be used to support the grant of summary judgment is no longer absolute.” Id. at 90.
The present case involves the following questions of fact: first, whether trauma to the head was suffered in the collision, and second, whether the collision caused or aggravated the cancerous tumor and symptoms of seizures and depression claimed. Assuming plaintiff did suffer the blow to the head, a fact the establishment of which does not require expert opinion, the second question involves a subject not within the general knowledge of a layman; the cause of aggravation of cancer and its symptoms, and the cause of depression even if not the cancer, must be established by expert medical opinion. See in this connection Jordan v. United Ins. Co., 158 Ga. App. 520 (281 SE2d 286) (1981); McDaniel v. Employers Mut. Liability Ins. Co., 104 Ga. App. 340 (121 SE2d 801) (1961); and Callaway Mills Co. v. Hurley, 100 Ga. App. 781 (112 SE2d 320) (1959).
Defendants offered four medical experts who agreed that it is *368most probable that the tumor was the cause of plaintiff’s seizures and depression, that the cause of an astrocytoma tumor is unknown, and that trauma to the head does not cause cancer. Medical expert Austin categorically concluded that she is aware of no medical evidence or studies which “document or support the notion that trauma activates, spreads, stimulates or accelerates the growth of astrocytoma brain tumors” and that “it is absolute conjecture and speculation to conclude that trauma accelerated the symptoms of Mr. Holley’s tumor.” Expert Epstein testified that, “I have never seen a case where trauma activated, induced, aggravated, spread, stimulated, or accelerated the growth of an astrocytoma brain tumor or caused an astrocytoma brain tumor to swell or inflame. Moreover, I am aware of no medical literature, medical evidence or medical studies which document or conclude that trauma affects astrocytoma brain tumors in any manner.” He then summarized: “The possibility that any trauma Mr. Holley may have experienced in the accident on October 23, 1980 caused the brain tumor or effected [sic] a pre-existing tumor in any way is extremely remote, as remote as a meteor striking my desk in the next five seconds.”
Holley counters with the depositions of his treating physicians Davis and Reitt. Both of these medical experts testified that they could positively conclude that the trauma did not cause the cancer and resulting seizures. As to the possibility that trauma had contributed to the onset of plaintiff’s symptoms, their testimony was conjectural at most, as it related specifically to Holley. Much of the testimony related to a general class of persons in somewhat the circumstances of Holley.
The following are relevant portions of Reitt’s testimony on deposition: “J suppose an accident or some sort of trauma could aggravate a brain tumor.” (Emphasis supplied.) . . . “Let’s say that the patient has a tumor or abnormality and is a set up for a seizure but has had no seizures. Subsequently . . . stressful situation . . . may have caused very small hemorrhaging . . . Then perhaps the pressure or the irritability of the surrounding nerve cells around this lesion would be more affected. ...” (Emphasis supplied.) He was not discussing Holley’s specific case but rather only a hypothetical class inherently made up of cases with varying fact patterns, including Holley’s. Reitt’s testimony continued, “Assuming the tumor was there at the time, . . . and the man went into some shear impact where he turned over, I think it is likely that the trauma aggravated a pre-existing condition to some degree ... I think it may have led to a change in the tumor characteristics by the trauma either with small microscopic or other hemorrhages or a shift in the condition or some way made it more likely than not that the surrounding normal cells somehow became more irritable.” Upon being questioned about the possibility *369that a blow to the head or some trauma, if any, precipitated the symptoms brought about by the seizure, he said, “It could have.” (Emphasis supplied.)
When asked about probabilities versus possibilities as to Holley, Reitt stated: “Well, I don’t know. Now, the only thing is one cannot dismiss the fact that he had an abnormal symptom develop three weeks after an automobile accident and had had no apparent trouble prior to that time. And I don’t think no matter what we do we can throw that away. It is common sense to therefore assume in some way the seizure threshold of his neurons was affected by the trauma, albeit it may have injured the tumor that may have been there, but in some way the trauma made him more disposed to have seizures at that time.” (Emphasis supplied.) The jury is not able to make such an assumption when the plaintiff’s burden is to prove causation by a preponderance of the evidence.
When asked if the head trauma may have accelerated the onset of Holley’s symptoms, he replied: “I think that’s fair. Yeah that’s basically what I’m saying may have happened. If, indeed, the tumor was there at that time. . . .” (Emphasis supplied.) But, when asked if he would go further and say that it was more likely than not that trauma would have accelerated the symptoms of the tumor, he stated: “I wouldn’t go that far. I don’t know. I honestly don’t.” (Emphasis supplied.)
When questioned about the possibility that the accident had no effect on Holley, Reitt testified, “Í think it’s possible . . . It is a possibility the accident had no effect on the tumor ... I would really think it probably did have some effect, simply because of the fact that you have three weeks and something changed in this boy.” (Emphasis supplied.) Thus, the possible/probable testimony is based on an assumption of causation due solely to the proximity in time of the collision and the “something changed.” It is akin to an application of the res ipsa loquitur doctrine.
Although Reitt’s testimony as to the probability versus possibility of a causal relationship waffles back and forth, as evidenced above, “[t]he entirety of deponent’s expert medical testimony demonstrates” the lack of probability. (Emphasis supplied.) See Kirby v. Spivey, 167 Ga. App. 751, 756 (307 SE2d 538) (1983).
Similarly, plaintiff’s expert Davis testified that the collision “may have aggravated the onset of his seizures,” and that the accident was one of the “multifactoral” causes of plaintiff’s depression. (Emphasis supplied.) Although Davis testified that “I think it’s probable [that Holley had post-traumatic neurosis aside from the tumor and medication for the tumor and surgery] and mixed with all those other factors, he also stated, “[Y]ou can’t say that ten percent of it [symptoms] is due to the tumor and 20 percent to this. I think it’s all mixed *370in together.”
Decided March 15, 1985 Rehearing denied March 29, 1985 Glenville Haldi, for appellant. Charles H. Ivy, David J. Dempsey, for appellees.Indicating the mere possibility rather than probability of a causal relationship is not sufficient to rebut the expert testimony of Austin and Epstein. McDaniel, supra, 104 Ga. App. at 344; Callaway Mills Co., supra, 100 Ga. App. at 784. The jury’s finding for plaintiff would be based on lay conjecture and speculation in the absence of any sufficiently positive medical opinion as to proximate cause. The threshold degree of evidence to establish the fact is not present here. The testimony was equivocal at best. Appellant in his brief in fact concedes that “[n]one of the medical testimony even attempts or presumes to establish what is the cause of cancer or its aggravation,” and that “[t]he only cohesiveness between these two medical mysteries is the fact that no one knows the answers to either one.”
Expert medical testimony as to mere possibility is only sufficient in those cases where the evidence is such that upon consideration of all other facts of the case “a natural inference through human experience would be raised to indicate” a causal relationship. Because such a “natural inference” cannot be drawn in the case at bar involving the cause of cancer symptoms, the matter must be resolved solely by expert medical evidence and must therefore be based at least on reasonable probability. See McDaniel, supra at 343-344.
Thus, since expert opinion was required, and that which defendants produced was of such a character and nature as to carry their burden of showing that no material issue existed which plaintiff failed to rebut or undermine, Howard v. Walker, supra, the trial court did not err in granting defendants’ motion for summary judgment. The defendants’ affirmative evidence and plaintiff’s weak counter, which failed to meet the legal standard for creating an issue of fact, established that plaintiff could not prove the reasonable probability of causation necessary to raise a material issue of fact. Cf. Jackson v. Gershon, 251 Ga. 577 (308 SE2d 164) (1983).
“If the damage incurred by the plaintiff is only the imaginary or possible result of a tortious act or if other and contingent circumstances preponderate in causing the injury, such damage is too remote to be the basis of recovery against the wrongdoer.” OCGA § 51-12-8.