We affirm the judgment below directing a verdict in favor of all defendants on both counts of the plaintiffs’ complaint. We rely primarily on the findings and reasoning set out in District Judge Flannery’s opinion, reported as Haven v. Randolph, 342 F.Supp. 538 (D.D.C.1972).
However, in his dissent Judge Van Pelt raises a question to which we should respond specifically. He believes that the jury could have reasonably inferred from the evidence produced at trial, particularly the testimony of Dr. Randolph himself, that Dr. Randolph had violated the applicable standard of care in administering the Hypaque dye to Roy Haven. We think that the trial court acted properly when it precluded the jury’s consideration of such an inference by directing a verdict.
In Kosberg v. Washington Hospital Center, 129 U.S.App.D.C. 322, 394 F.2d 947 (1968), we stated: “A prima facie case of medical malpractice must normally consist of evidence which establishes the applicable standard of care, demonstrates that this standard has been violated, and develops a causal relationship between the violation and the harm complained of.” 129 U.S.App.D.C. at 324, 394 F.2d at 949. The trial court concluded in this case that the plaintiffs had failed to clear the first hurdle of establishing the standard o'' care. The court stated- “[PjlainDpfs offered no expert medie-; i testimony which might be probative of the standard of care except for the testimony of defendant physicians themselves.” 342 F.Supp. at 543. Judge Van Pelt, however, responds that “negligence of a doctor can be shown either by expert testimony indicating that the community standards were not followed by him or by proven facts or by inferences which by_ common knowledge may be drawn from the facts.” Dissenting opinion, infra, p. 1072 (emphasis supplied). Allowing a jury to make its own inferences from the proven facts may be permissible when a physician has committed a blunder so egregious that a layman is capable of comprehending its enormity. An example is the case of a surgeon who leaves a sponge in an incision after removal of a kidney. See Rodgers v. Lawson, 83 U.S.App.D.C. 281, 284-285, 170 F.2d 157, 160-161 (1948). But where “the question turns on the merits and performance of scientific treatment, the issue may not be resolved by the jury without the aid of expert testimony.” Brown v. Keaveney, 117 U. S.App.D.C. 117, 118, 326 F.2d 660, 661 (1963).
In this case, Dr. Randolph performed a highly sophisticated transfemoral aortogram, which is the procedure of “injecting radio-opaque media into the renal artery by means of a catheter inserted into the femoral artery through an incision in the thigh near the groin, and moving it upwards to the renal artery, in order to visualize by x-rays the kidneys and surrounding areas.” Brief for Appellee Randolph at 5 n. 1. Clearly, in order to establish a standard of care applicable to administration of an aortogram and to demonstrate that Dr. Randolph breached that standard, expert testimony was necessary. None was supplied by plaintiffs in this case.
Despite the failure of plaintiffs to establish by expert testimony a standard of care and its breach, Judge Van Pelt argues that a jury could reasonably have inferred negligence from certain segments of Dr. Randolph’s testimony. These segments do not expose any fault in Dr. Randolph’s physical technique in *1071administering the aortogram. They merely indicate that the doctor may have injected dosages that were hazardous to the patient, Roy Haven. In focusing on these isolated segments of Dr. Randolph’s testimony, we think our colleague does not adequately evaluate the doctor’s assertions that the hazards he created were necessary in order to treat Roy’s death-threatening condition properly. Evaluation of a doctor’s conduct in treating a patient requires examination not only of the risks inherent in the doctor’s method of treatment, but also of whether running those risks was justified as a prerequisite to correcting the patient’s condition.
The passages from Dr. Randolph’s testimony cited by Judge Van Pelt demonstrate at most that Dr. Randolph may have taken a substantial risk in administering the aortogram to Roy in the manner he did.1 True, Dr. Randolph admitted “that the adult dosage was 25 c.c.’s [and] that this dosage was proportioned down for children.” Dissenting opinion, infra, p. 1072. But he also testified that “drug scaling down becomes a matter of reason, a matter of judgment, a matter of experience.” Transcript at 739. Dr. Randolph’s reason, judgment, and experience told him that scaling down the dosage for Roy too drastically would have made the amounts injected “unworkable,” Transcript at 738,2 and thus ineffective for the purpose of revealing the location of the stenosis blocking Roy’s renal artery. The doctor further stated, “I could not proceed with surgery without demonstration of the artery and the possible blockage . . . . ” Transcript at 668. No one has contested the proposition that surgery to correct Roy’s stenosis was necessary to save his life. Therefore, Dr. Randolph’s testimony was that the risk inherent in injecting 9-10 ce.’s of Hypaque was necessary in order to correct Roy’s life-threatening condition.
The doctor also admitted that he administered a total of 19 cc.’s within a period of twenty minutes to Roy, and that such repetition of dosages should “be avoided whenever possible.” Transcript at 1133. However, he supplied sound medical reasons for the consecutive dosages:
Well, my best considered judgment at that time, based on some years of training in children’s surgery, based on the specific study of the problem of renal hypertension and aortography in children, told me that the best thing to do was to proceed with the second injection because this boy was under general anesthesia; we knew that any general anesthesia was a threat to him; the catheter was in place; there was already some hazard to the renal artery — excuse me — to his femoral artery, which is the site of the operation to put the catheter; and to try to double all of that in the face of the body of information, which was that no child had ever been parlyzed [sic] by this procedure before under the age of nine, seemed to me, in my judgment, a much better decision for the boy.
Transcript at 669. Only an expert could determine whether Dr. Randolph’s exercise of medical judgment in this case measured up to the standard of care for physicians in the community. This is manifestly not a case in which the jury should have been allowed to infer negligence from admissions that the treatment methods employed involved some hazard to the patient.
Affirmed.
. There is some doubt about the magnitude of the risk since Roy’s extreme reaction to the Hypaque injection was apparently the first recorded instance of such a reaction to aortography in a child under nine years of age. Transcript at 651, 1094 (testimony of Dr. Randolph).
. Dr. Randolph asserted that in his experience the 9-10 ce. dosage that he administered to Roy during each test represented the “upper limits of safe dosage.” Transcript at 1130 (emphasis supplied).