The plaintiff has recovered a judgment for damages in an action for malpractice against the appellant, Dr. Rosenow. He claimed he sustained an injury to his left arm as the result of the negligent administration of a needle in the withdrawal of blood for laboratory examination. The appellant argues that the judgment cannot be supported by the evidence, that, in effect, the plaintiff proceeded on the mistaken theory of res ipsa loquitur. For reasons stated beyond, we reverse the judgment and grant a new trial.
The plaintiff suffered from a bladder ailment and consulted defendant Dr. Hans Goldmann for treatment. Dr. Goldmann referred him to the appellant for blood tests. The appellant withdrew blood without incident. About a month later, Dr. Goldmann again requested the plaintiff to see the appellant for additional tests. According to the plaintiff’s testimony, on this occasion he experienced “terrific pain” in his arm when the appellant inserted the needle; and his fingers shook uncontrollably. The appellant removed the needle and the pain and shaking subsided. The appellant then said that he could not draw blood from the vein on the inside of the arm and he took blood from the outside area of the plaintiff’s arm.
The plaintiff testified that during the following week the pain in his arm grew more severe and he returned to the appellant for alleviation. The latter examined the arm and told him to see Dr. Goldmann, who would inject vitamin B-12. Dr. Goldmann began a series of three injections, but the pain increased, so that Dr. Goldmann called the .appellant, who made an appointment for the plaintiff to see defendant Dr. Jack London, a neurosurgeon. Dr. London informed the plaintiff that a nerve had been injured and that further vitamin B-12 injections were necessary.
*242The plaintiff thereafter received 29 injections over a period of about three months. He said his pain was considerable during this time and that presently he felt occasional pain and numbness in his fingers.
The plaintiff’s case included the testimony of Dr. Karlan, to whom the plaintiff had gone for examination. Dr. Karlan testified that neurological tests disclosed some loss of sensation in the back of the plaintiff’s hand, which Dr. Karlan attributed to neuritis of the radial nerve that “ could be an inflammation or an injury of the radial nerve.” It was his opinion that the neuritis was causally related to the second occasion of the taking of a blood specimen by the appellant.
The plaintiff also called Dr. Wigderson, who had examined the plaintiff at the request of the appellant. Dr. Wigderson testified that his diagnosis was that there existed sensory evidence of radial nerve involvement, based on the plaintiff’s subjective response.
The plaintiff rested after reading from the depositions before trial of Dr. Goldmann and Dr. London. Dr. Goldmann testified as to the method of withdrawing blood by needle from a vein (venipuncture). Dr. London testified that the radial nerve i-is located about one inch deeper than the vein found in the inside of the elbow and a half inch away from it. He said that venipuncture usually poses no danger to any nerve.
The appellant rested on the plaintiff’s case and moved to dismiss the complaint.1 The Trial Term reserved decision; that motion and the appellant’s motion to set aside the verdict were thereafter denied.
The appellant’s chief argument is that the plaintiff failed to establish by expert testimony that the appellant had not performed the venipuncture in accordance with proper and accepted medical standards. If the appellant be correct in this proposition, then the judgment must be reversed, because the record contains no such testimony. Whether the proposition is correct perforce directs our inquiry .to the doctrine of res ipsa loquitur in the field of medical malpractice.
Res ipsa loquitur, as a rule regulating proof, is no more than the permissible drawing of an inference from circumstantial evidence (2 Harper and James, Torts, § 19.11). One of the conditions underlying its invocation requires that the casualty be of a character which does not usually occur, save *243through negligence (Foltis, Inc. v. City of New York, 287 N. Y. 108, 115; McCormick, Evidence, § 309, pp. 643-644). That condition necessarily implies that the circumstances of the occurrence can he reasonably appraised by the common knowledge or experience of a jury to reach the conclusion that negligence is an attending factor of the occurrence.
In a direct sense, therefore, the doctrine is merely a reflection of the principle that a jury of laymen ordinarily do not need expert testimony on the question of negligence, itself a function of the conduct of reasonable men (Meiselman v. Crown Heights Hosp., 285 N. Y. 389, 396; cf. Lince v. Monson, 363 Mich. 135). Hence, as the area of common knowledge grows, so does the necessity for expert testimony to aid the jury diminish.
Yet, the science of medicine is recondite; and, in the ordinary case of malpractice, a jury is not equipped to determine the issue of negligence simply as an inference from the circumstances (Benson v. Dean, 232 N. Y. 52, 56; Charlton v. Montefiore Hosp., 45 Misc 2d 153; see ann. 81 ALR 2d 597, 82 ALR 2d 1262). The information obtained from an expert witness supplies the jury with the means by which it can evaluate the conduct of the party charged with malpractice (cf. 7 Wigmore, Evidence [3d ed.], § 1917, pp. 3-4).
There are several kinds of particular acts of malpractice which are generally recognized to lie within a jury’s competence: (1) where the injury occurred outside the site of an operation (cf. Charlton v. Montefiore Hosp., supra; Ayers v. Parry, 192 F. 2d 181, cert. den. 343 U. S. 980); (2) where a foreign object is left within the body of a patient after an operation (cf. Benson v. Dean, supra; Blackburn v. Baker, 227 App. Div. 588); (3) where the injury occurred while the patient was under anesthesia (cf Matlick v. Long Is. Jewish Hosp., 25 A D 2d 538; Wolfe v. Feldman, 158 Misc. 656); and (4) where the nature of the acts alleged, such as a beating of the patient by a psychiatrist, itself bespeaks improper treatment (cf. Hammer v. Rosen, 7 N Y 2d 376; Agnew v. City of Los Angeles, 82 Cal. App. 2d 616). In all of these cases it was considered that enough was shown to cast a duty on the treating physician to explain his conduct.
The case before us does not fall within any of these categories. That in itself does not mean that the doctrine of res ipsa loquitur does not apply, because the common law constantly refashions its rules to deal with changing circumstances. But it does require us to meet the question whether *244an injury to a nerve following the use of a needle for the withdrawal of blood ordinarily connotes negligence.
In general, the courts which have dealt with this subject have not allowed an injured party recovery merely on the proof of an untoward result from the administration of a needle by a doctor (see, e.g., Toy v. Rickert, 53 N. J. Super. 27; Ayers v. Parry, 192 F. 2d 181, cert. den. 343 U. S. 980, supra; Bettigole v. Diener, 210 Md. 537; Hall v. United States, 136 F. Supp. 187, affd. 234 F. 2d 811; Hornbeck v. Homeopathic Hosp. Assn, of Del., 197 A. 2d 461 [Del.]; Talbot v. Dr. W. H. Groves’ Latter-Day Saints Hosp., 21 Utah 2d 73; Graham v. St. Luke’s Hosp., 46 Ill. App. 2d 147; Honeywell v. Rogers, 251 F. Supp. 841). The reasoning sustaining this result is that absent expert testimony a jury could not be expected to possess the requisite knowledge and experience to decide whether the injury incurred would have happened without negligence on the part of the physician.
There is authority to the contrary (e.g., Wolfsmith v. Marsh, 51 Cal. 2d 832; Bauer v. Otis, 133 Cal. App. 2d 439; Horner v. Northern Pacific Beneficial Assn. Hosp., 62 Wn. 2d 351; Herbert v. Travelers Ind. Co., 193 So. 2d 330 [La.]). The theory of these cases is that needle injections are so commonplace that it is well known that injuries do not occur unless they are unskillfully administered or the substance injected is harmful, so that it is fair that a duty of explanation should be imposed on the physician.
In appraising the effect of adopting a rule of res ipsa loquitur in a particular case two factors, not affecting the strict logic of the rule but truly its policy, enter into the consideration. First, it is not always easy for a plaintiff in a malpractice action to secure expert medical testimony (cf. McDermott v. Manhattan Eye, Ear & Throat Hosp., 15 N Y 2d 20, 27-28). The increasing number of malpractice suits suggests, however, that the difficulty may not now be as great as it was in the past. Second, eminent scholars in the field have remarked that the plaintiff permitted to prove a prima facie case by res ipsa loquitur rarely loses (2 Harper and James, Torts, § 19.11, p. 1099, supra).
On balance, then, we think that the doctrine of res ipsa loquitur should not be applied to this case. It is not enough that the use of a needle is not in most instances followed by serious harm; in addition, it must be found that among all the possible causes of the harm those which point to the physician’s negligence are so probable, under the circumstances, and those *245which would exonerate him so improbable, that accordingly the patient need not specify the one which actually led to the harm (Abbott v. Page Airways, 23 N Y 2d 502; 58 Am. Jur. 2d, Negligence, § 479). That is to say, application of the doctrine in a malpractice action must rest on a finding that an inference exonerating the physician is improbable as a matter of fact. To allow the plaintiff’s use of the doctrine to make out a prima facie case here requires us to say that a jury could, without expert testimony, conclude that radial nerve involvement would not result from a properly administered injection. We are not prepared to make that determination.
The plaintiff claims that, apart from res ipsa loquitur, a prima facie case was established. He argues that the jury could have reasonably inferred that, since blood was drawn on the first occasion without any negative result, the injury on the second occasion came about because of improper procedure. The only fact upon which that inference could be made is that the appellant “ stabbed ” the plaintiff five times on the second occasion. We are unable to say, however, that this fact is indicative of poor medical practice. We think that expert testimony is needed before a jury could reach such a conclusion.
The plaintiff also contends that the testimony of Dr. London that venipuncture usually does not pose any danger of injury to nerves, because of anatomical structure, and that if the veins were not prominent it would have been necessary for the appellant to palpate the plaintiff’s arm (a process which the plaintiff testified the appellant did not apply) was sufficient to prove a prima facie case. But this testimony does not rule out a danger to nerves by venipuncture even if accepted medical procedure were pursued; nor was there any evidence that the plaintiff’s veins were so obscure that palpation was required.
A new trial is appropriate so that the plaintiff may be given the opportunity to produce expert testimony. On the new trial the plaintiff should not read from the depositions from Dr. Goldmann and Dr. London, since they are no longer defendants in the action. Moreover, we note with disapproval that the plaintiff’s attorney admitted during the course of the trial that these two doctors had been joined as parties defendant solely for the purpose of obtaining their testimony as experts. Indeed, after reading portions of their depositions, the plaintiff’s attorney offered to discontinue the action as against them.
A plaintiff may of course call a defendant in a malpractice case to testify as an expert witness, to establish the plaintiff’s case. But, in so doing, as the Court of Appeals has said, “It *246is, of course, assumed that a plaintiff, in naming a doctor as a defendant, has done so in good faith, on the basis of his relationship with the case and not as a device or subterfuge in order to afford the plaintiff an opportunity to call him as an expert witness ” (McDermott v. Manhattan Eye, Ear & Throat Hosp., 15 N Y 2d 20, 30, n. 5, supra).
The judgment accordingly should be reversed insofar as appealed, on the law and the facts, and a. new trial granted as between the plaintiff and defendant Dr. Rosenow, with costs to abide the event and with appropriate severance of the action.
Martuscello, Latham, Shapiro and Brennan, JJ., concur.
Judgment reversed insofar as appealed from, on the law and the facts, and new trial granted as between plaintiff and said defendant, with costs to abide the event and with appropriate severance of the action.
. The Trial Term at that time granted motions to dismiss the complaint as to Drs. Goldmann and London.