OPINION OF THE COURT
Sullivan, J.
On March 31, 1970, plaintiff, Michael Mortensen, then 10 years old, underwent an operative procedure at Arden Hill Hospital for the removal of a Baker’s cyst from the back of his left knee. Upon making his incision, however, and after retracting certain tendons and nerves the surgeon, defendant Dr. Rakov, discovered a hard, gritty “rather indistinct” mass which appeared to have penetrated the entire popliteal space. Dr. Rakov considered his finding unusual since he had never seen [152]*152such a tumor in that part of the anatomy. He resected a portion of the mass, left the rest intact and closed the incision. A pathologist diagnosed the tumor as juvenile fibromatosis, a locally active, nonmalignant type, nonmetastasizing, but one which tends to invade nearby structures.
Discharged from the hospital on April 4, plaintiff thereafter visited Dr. Rakov at his office on April 9 and again on April 16. On both occasions, Dr. Rakov conducted a brief physical examination and removed stitches. When plaintiff still complained of soreness and discomfort at a subsequent visit on May 15, and his own examination revealed an area of decreased sensation and increased muscle atrophy in the left leg, Dr. Rakov decided to refer plaintiff to Dr. Knapper, a surgeon specializing in gastric and mixed tumors on the staff of Memorial Hospital. Dr. Rakov made the referral primarily because, as he explained, he “did not want to be the one to tell the patient that he needed a leg amputated.” Dr. Rakov also was hopeful that Dr. Knapper “might have some idea” as to how amputation could be avoided.
Dr. Knapper first saw plaintiff on May 27 and found that “[t]he mass [had] rapidly recurred and enlarged to four or five times the size it had attained prior to initial excision.” Due to the nature of the tumor Dr. Knapper decided that he would attempt a wide excision, but, if the mass recurred, amputation would be necessary. During exploratory surgery, which took place on June 9, Dr. Knapper found that the tumor was surgically unresectible, since the major artery and vein circulating blood to the leg below the popliteal area were encompassed by the mass, which was attached to the periosteum, with the lateral and medial popliteal nerves adhering to it. In light of plaintiff’s age and size Dr. Knapper also ruled out the possibility of any bypass or grafting procedure.
On June 12, 1970, Dr. Knapper amputated plaintiff’s left leg at midthigh. After his discharge from the hospital on June 26, plaintiff underwent a lengthy rehabilitation program at the Rusk Institute. In 1979 he commenced this action alleging medical malpractice against Dr. Rakov and others for his pain and suffering and his severe and permanent disability. The action was either discontinued against the other defendants or dismissed at the close of plaintiff’s case. Only the action against Dr. Rakov is at issue.
At trial, plaintiff’s expert testified that Dr. Rakov, after having excised only a portion of the mass, departed from accepted medical practice by failing to reoperate once the tentative diagnosis of juvenile fibromatosis had been made. At that time, [153]*153according to plaintiff’s expert, the vital structures were still unaffected, so that had a wide excision of the tumor mass as well as the tissue around it been performed, plaintiff would have had an 80% chance of achieving a complete recovery. Instead, according to plaintiff’s expert, Dr. Rakov failed to provide any adequate follow-up care; nor did he afford plaintiff the opportunity of receiving medical care from a more qualified specialist. This delay permitted the tumor to grow to inoperable proportions and deprived plaintiff of a chance for a cure.
Dr. Rakov’s expert testified that even by March 31, 1970 it was no longer possible to excise the tumor fully since it was growing from the inside out, and was located in an area where major vessels, arteries and nerves lie. Accordingly, once the nature of the tumor was diagnosed, Dr. Rakov exercised proper medical judgment in performing a partial excision and in foregoing any plan for further treatment.
Admittedly, however, Dr. Rakov had not attempted to follow any of the arteries, veins or other internal structures to determine the extent of their involvement with the mass. It should also be noted that his operative report did not make any mention of the popliteal artery or vein or that the tumor was adhering to the periosteum.
Dr. Knapper, who was a witness for Dr. Rakov, testified that, while the March 31, 1970 surgery was acceptable and in accordance with proper medical practice, he could not express an opinion as to whether Dr. Rakov’s course of conduct following the operation was in accordance with accepted standards of practice. He nevertheless felt that “anyone who fully understands the nature and progressive abilities of fibromatosis would probably give him further follow-up or made arrangements with him.” A pathologist at Memorial Hospital who reviewed slides from the March 31 operation testified that the tumor was not a juvenile fibromatosis, but, in fact, a desmoid tumor, “arising from the deep structures of the human body and progressing * * * slowly [over] a duration of years.” It should be noted, however, that although a desmoid tumor might be slightly faster growing and more locally malignant than a fibromatosis the two diagnoses are essentially the same.
During the course of a precharge conference, plaintiff’s counsel asked the court to “charge in your own language the law with respect to the Kallenberg case”
“In considering the question of proximate cause in this case, the law does not require that the plaintiff show to a certainty that his leg would have been saved had the defendant excised the tumor in early April of 1970. If you conclude that there was a substantial possibility that amputation could have been avoided had that surgery been performed at that time, that is in early April, and that the defendant’s negligence in failing to perform that surgery deprived the plaintiff of that substantial possibility, then the defendant is liable to the plaintiff.
“If, however, you conclude that that substantial possibility did not exist, then plaintiff has not established that he suffered any injury as a result of the defendant’s negligence and he is therefore not entitled to recover.”
In its charge-in-chief, the court submitted two special questions to the jury: (1) “Was Dr. Rakov negligent in not attempting to remove the mass in plaintiff’s left leg in early April, 1970?” (2) “Was there a substantial possibility that amputation could have been avoided had surgery been performed in early April, 1970?”
After commencing deliberations, the jury requested that the wording of the first question be changed. The court agreed and amended the question to read: “Was Dr. Rakov negligent in not taking any action with respect to the mass in plaintiff’s left leg in early April, 1970?”
Thereafter, the jury resumed deliberations and sent another note which asked, “can we award damages on question 1 if question 2 is answered no?” At this point, plaintiff’s counsel once again urged that the Kallenberg rule be charged (Kallenberg v Beth Israel Hosp., 45 AD2d 177, affd 37 NY2d 719) so that, he explained, plaintiff could recover if any possibility existed that amputation could have been avoided. The court refused.
At this juncture, the jury sent another note requesting guidance as to “a new category to base damages on.” Over the objection of plaintiff’s counsel, who then requested that the special questions be withdrawn and a general verdict taken, the court advised the jury that in order to award damages it would have to answer both questions in the affirmative. The court further charged the jury that it was not in a position to give a specific percentage, but that substantial possibility meant a “significant” or “realistic” possibility. Since, however, the court [155]*155was unclear as to the meaning of the third note, it asked the jury to clarify its request, which the jury did by submitting the following note: “What we mean is that we believe there was not substantial possibility, but there was negligence on the doctor’s part. We feel the plaintiff did not have his right to a second opinion and the doctor did not know enough to make a judgment to wait. So, being that there might have been the slightest possibility of saving the leg because of Dr. Rakov’s action or lack of it, plaintiff did not get that opportunity to find out. Therefore, we wish to award damages based on that negligence and not the loss due to the amputation. Question: Can we? If so, we need guidance.”
The court thereupon advised the jury that it was not legally permissible to award damages on such a basis, and that if the answer to the second question was in the negative, plaintiff would not be entitled to recover. Plaintiff’s counsel excepted, urging once again that, under Kallenberg (supra), even the slightest possibility of saving the limb would be sufficient to support a verdict in plaintiff’s favor. The jury thereafter returned with a verdict of no liability, finding, unanimously, that Dr. Rakov was negligent in not taking any action with respect to the mass in early April, 1970 but, by a 5 to 1 vote, that a substantial possibility of saving the leg did not exist even had such action been taken. Plaintiff appeals, citing error in the court’s charge with respect to causation, and arguing that all that need be shown to establish proximate cause is that plaintiff was deprived of an opportunity for cure.
At issue here is the degree of certainty required to show causation in a medical malpractice action. Apparently, this issue has been the subject of some confusion. For instance, Kallenberg v Beth Israel Hosp. (45 AD2d 177, affd 37 NY2d 719, supra), upon which plaintiff relies, is improperly interpreted to mean that a deprivation of a 2% chance of survival caused by a defendant’s failure to afford proper medical treatment is evidence of causation sufficient to make out a prima facie case of medical malpractice. A careful reading of Kallenberg elicits the conclusion, however, that the causal connection between the defendant’s malpractice and Mrs. Kallenberg’s demise was established by a much more substantial degree of certainty. Upon her admission to the hospital a diagnosis was made that Mrs. Kallenberg “was in need of a specific medication [Naturetin] to reduce her blood pressure”. This court found, however, that “[f]or some unexplained reason, this drug was never administered to the decedent, although its administration was consid[156]*156ered necessary to prepare her for needed surgery” (supra, at p 178), and that the failure to administer it “was a departure from accepted medical practice and procedure” (supra, at p 179). Even one of the defendant physicians “conceded at the trial that control of Mrs. Kallenberg’s blood pressure by proper medication meant the difference between life and death and that Naturetin was a critical drug for her” (supra, at pp 178-179). Thus, this court held, a jury could find that had Mrs. Kallenberg been treated properly she might have, even after suffering a second hemorrhage, improved sufficiently “to undergo surgery and make a recovery” (supra, at p 180).
The Kallenberg court also noted that plaintiff’s expert testified that, notwithstanding this second hemorrhage, “ ‘properly treated * * * the patient still has [would have had] a 20, say 30, maybe 40% chance of survival’ if surgery had been undertaken; and that surgery could have been performed, if the proper drugs had been administered” (45 AD2d, at pp 179-180). Thus, this court was dealing with a record that spoke with some degree of certainty that, if administered, the drug would have permitted surgery, which, in turn, would have afforded the patient a 20 to 40% chance of recovery.
In addition, and this is probably the source of the misinterpretation of Kallenberg, the plaintiff’s expert “also testified that if the proper drugs had been administered, even without surgery, she had a 2% chance of survival” (45 AD2d, at p 180). Consequently, with the medication, an additional slight chance existed that the surgery itself would have been unnecessary. A jury could thus reasonably conclude that the defendants’ malpractice deprived the deceased of a substantial possibility of survival. In any event, Kallenberg did not involve the propriety of a jury charge on proximate cause, but only the issue of whether the expert testimony made out a prima facie case.
In Kimball v Scors (59 AD2d 984), upon which Trial Term relied, the plaintiffs, apparently interpreting Kallenberg as does plaintiff here, had specifically requested a charge that “ ‘[t]he jury need only decide whether or not Mr. Kimball could have had a chance to survive had the malpractice not taken place.’ ” The Third Department disagreed with such an interpretation, stating (pp 984-985): “In our view, Kallenberg v Beth Israel Hosp. (45 AD2d 177, affd 37 NY2d 719) does not stand for the position urged by plaintiffs, i.e., that a jury need only determine that defendants’ malpractice deprived a decedent of a chance of survival, regardless of how small that chance might be. Such a charge is implicit with danger in that it could reasonably be [157]*157construed by jurors as judicial restraint on their obligation to find that the malpractice proximately caused the death. The ultimate finding cannot be whether the deceased would have a certain percentage chance of recovery; rather, it must be whether there was a substantial possibility the decedent would have recovered but for the malpractice. If the proof is ambivalent as to the question of whether the deceased would have died regardless of the malpractice, a pure factual issue is raised, as here, and such an issue can only be resolved by a jury determination of whether the malpractice proximately deprived the deceased of that substantial possibility.”
In 1981, in Monahan v Weichert (82 AD2d 102), the Fourth Department, confronted with conflicting evidence of proximate cause — “one being the physician’s conduct and the other being the course of plaintiff’s illness” — and citing Kimball and Kallenberg (supra) held, “To establish a prima facie case plaintiff need not eliminate entirely all possibility that defendant’s conduct was not a cause, but only offer sufficient evidence from which reasonable men may conclude that it is more probable that the injury was caused by defendant than that it was not (Restatement, Torts 2d, § 432, subd [2]; § 433B, Comment 6)” (supra, at p 108).
We agree with the decision in Kimball (supra) that proximate cause is a legal concept which cannot be dissected and measured in terms of percentages. As the Court of Appeals has noted: “The concept of proximate cause, or more appropriately legal cause, has proven to be an elusive one, incapable of being precisely defined to cover all situations * * * This is, in part, because the concept stems from policy considerations that serve to place manageable limits upon the liability that flows from negligent conduct (e.g., Ventricelli v Kinney System Rent A Car, 45 NY2d 950, 952; Palsgraf v Long Is. R.R. Co., 248 NY 339, 352 [Andrews, J., dissenting]). Depending upon the nature of the case, a variety of factors may be relevant in assessing legal cause. Given the unique nature of the inquiry in each case, it is for the finder of fact to determine legal cause, once the court has been satisfied that a prima facie case has been established (see, e.g., Nallan v Helmsley-Spear, Inc., 50 NY2d 507, 520, 521; Sheehan v City of New York, 40 NY2d 496, 502, 503; Kingsland v Erie County Agric. Soc., 298 NY 409, 424, 427). To carry the burden of proving a prima facie case, the plaintiff must generally show that the defendant’s negligence was a substantial cause of the events which produced the injury (Nallan v Helmsley-Spear, Inc., supra, at p 520; Restatement, Torts 2d, § 431).” (Derdiarian v Felix Contr. Corp., 51 NY2d 308, 314-315.)
[158]*158In a medical malpractice action, as in any negligence action, the plaintiff has the burden of proving, by a preponderance of the evidence, that the defendant’s negligence, in this case the departure from good and accepted medical practice, proximately caused the injury claimed. (Kletnieks v Brookhaven Mem. Assn., 53 AD2d 169, 176.) If such negligence is a substantial factor in producing the injury, it is a proximate cause of the injury. (See Dunham v Village of Canisteo, 303 NY 498, 505-506.)
Of course, in requiring that a defendant’s negligent act or omission be a substantial factor in bringing about the plaintiff’s injury, the substantial factor need not be the only cause which produces the injury. (Dunham v Village of Canisteo, 303 NY, at p 504.) A plaintiff is not required to eliminate every other possible cause. (Rosenberg v Schwartz, 260 NY 162, 166.) That another possible cause concurs with a defendant’s negligent act or omission to produce an injury does not relieve a defendant from liability if the plaintiff “shows facts and conditions from which the negligence of the defendant and the causation of the accident by that negligence may be reasonably inferred.” (Ingersoll v Liberty Bank, 278 NY 1, 7.) However, “where an [injury] is one which might naturally occur from causes other than a defendant’s negligence the inference of his negligence is not fair and reasonable”. (Cole v Swagler, 308 NY 325, 331, citing Foltis, Inc. v City of New York, 287 NY 108, 117; Galbraith v Busch, 267 NY 230, 233-234; Tortora v State of New York, 269 NY 167.) The burden, of course, of proving causation always remains with the plaintiff. If conflicting inferences may be drawn, the choice of inference is for the jury. (Foltis, Inc. v City of New York, supra, at p 118.)
Contrary to plaintiff’s arguments, he is not entitled to recover on the basis of a finding that Dr. Rakov’s malpractice deprived him of the possibility, no matter how slight, of saving his leg. Such a finding cannot support a damage award since proof of a mere possibility of cure does not satisfy a prerequisite to liability, i.e., that Dr. Rakov’s malpractice was a legal cause — a substantial factor — in bringing about plaintiff’s injury, which is, after all, the loss of his leg. If the loss of the leg was from a cause other than his malpractice, then Dr. Rakov is not liable.
In our view, the appropriate inquiry as to proximate cause in the situation here presented is whether, given the condition which existed in plaintiff’s left leg in the latter part of March or early April, 1970, it is more probable than not that the loss of the limb was caused by Dr. Rakov’s negligence in not reoperating immediately or taking other appropriate action. (Monahan v [159]*159Weichert, 82 AD2d 108, supra; Restatement, Torts 2d, § 432, subd [2]; § 433B, Comment b.) That, however, was not the basis of Trial Term’s charge. Instead, the jury was asked to consider whether Dr. Rakov’s malpractice deprived plaintiff of a substantial possibility of saving the limb. Although this court has had occasion to criticize the “substantial possibility” phrase as “purely speculative” when used by an expert as the basis for projecting an opinion as to whether a party would, at some future time, develop an additional ailment (see Cohen v Lizza, 63 AD2d 557), we believe that in determining proximate cause it accurately conveys to the layman the requirement that to be actionable the defendant’s negligence must, more probably than not, bring about the plaintiff’s injury.
The concepts of “substantial factor in bringing about an injury” and “substantial possibility of avoiding the injury”, which reflect, from both perspectives, the issue confronting a jury in such cases, are virtually indistinguishable. That this jury understood that Dr. Rakov’s malpractice had to be a substantial factor in bringing about the loss of plaintiff’s leg — by depriving him of a substantial possibility of avoiding amputation — is clear, as evidenced by its last note. It found Dr. Rakov negligent in not taking any further action after the March 31,1970 partial incision but, in light of the existing pathology, was unable to find that his inaction was a proximate cause of the amputation.
The “substantial possibility” charge has been approved in at least one other jurisdiction, Pennsylvania. (See Hoeke v Mercy Hosp., 445 A2d 140, 143 [“If there was any substantial possibility of avoiding injuries and the defendant has destroyed that possibility, he is liable to the plaintiff”].) In a Federal case in the Fourth Circuit, on an appeal from the dismissal of a complaint, the court held that Virginia law affixes liability, “[i]f there was any substantial possibility of survival and the defendant has destroyed it” (Hicks v United States, 368 F2d 626, 632). Recently in New York, in O’Connell v Albany Med. Center Hosp. (101 AD2d 637, 638), the Third Department noted, in affirming a verdict in plaintiff’s favor, that the charge permitted a jury to find, based on the testimony of plaintiff’s expert, “that there was a substantial possibility that plaintiff [’s] * * * recovery would have been faster, less painful and less disabling but for the malpractice of defendant”.
Thus, Trial Term quite correctly refused to charge in accordance with plaintiff’s request that the deprivation of any possibility of avoiding amputation was sufficient to impose liability. [160]*160Moreover, we conclude that Trial Term’s charge accurately conveyed to the jury the requirement that Dr. Rakov’s malpractice, to be actionable, had to be a substantial factor in bringing about the loss of plaintiff’s leg.
Accordingly, the judgment of the Supreme Court, New York County (Gammerman, J.), entered June 18,1983, dismissing the complaint, should be affirmed, without costs or disbursements.
Murphy, P. J., Kupferman and Kassal, JJ., concur.
Judgment, Supreme Court, New York County, entered on June 18,1983, unanimously affirmed, without costs and without disbursements.
Counsel was referring to Kallenberg v Beth Israel Hosp. (45 AD2d 177, affd 37 NY2d 719).