This is a medical malpractice action brought in 1977 by Charlotte Shelnitz for personal injuries and by her husband, Hyman Shelnitz, for loss of consortium, arising out of an ambulatory myelogram performed by the defendant neurosurgeon, Alvin Greenberg, who at the time of the procedure was a member of the defendant Neurosurgical Associates of New Haven, P. C.
The plaintiff’s1 complaint alleged negligence and a failure by the defendant2 to obtain informed consent. *60After a trial to a jury, a verdict was returned in favor of the plaintiff in the amount of $750,000 for Charlotte Shelnitz and $50,000 for Hyman Shelnitz. The defendant’s motions to set aside the verdict and for judgment notwithstanding the verdict were denied and the court rendered judgment for the plaintiff. This appeal followed.
The defendant raises six issues on appeal: (1) whether the evidence was sufficient to support the jury verdict on the issue of causation; (2) whether the court erred in failing to give a Secondino charge; (3) whether the plaintiff produced expert testimony sufficient to support her claim of lack of informed consent; (4) whether several of the court’s evidentiary rulings resulted in harmful error; (5) whether the court erred in refusing to grant the defendant’s motions for a mistrial; and (6) whether the verdict was excessive. We find no error.
I
On the advice of her internist, the fifty-four year old plaintiff consulted the defendant on November 22, 1974, with respect to a back problem. After an examination, the defendant recommended conservative treatment. The plaintiff was not “making much progress” and called the defendant on January 3,1975, at which time she was told that he would perform an ambulatory myelogram at the Neuro-Diagnostic Center in New Haven on January 9, 1975. The plaintiff testified that during this telephone conversation the defendant did not tell her about “any of the risks” nor did he inform her that the myelogram could be performed in a hospital.
The testimony at trial established that a myelogram is an invasive diagnostic procedure whereby a contrast material, in this case Pantopaque, is injected into the patient’s spinal column and then fluoroscoped and highlighted by x-ray to determine whether there is any *61pathology of the lumbar vertebrae and other portions of the patient’s anatomy. See also 2 Schmidt, Attorneys’ Dictionary of Medicine, M-173; Stedman’s Medical Dictionary (24th Ed.) p. 917. The spinal canal consists of three membranes: the dura, the arachnoid in the middle, and the pia matter. It is a closed, pressurized system in which cerebrospinal fluid (CSF) circulates, cleansing and nourishing the spinal canal and the brain and the other elements of the central nervous system.
A physician, in order to perform a lumbar myelogram, uses a needle to puncture the dura and arachnoid membranes in order to inject the contrast material into the closed system. As a result of the puncture, CSF leaks out of the closed system of membranes into the tissue around the puncture of the dura until the puncture heals and reseals itself. Every patient loses a small amount of CSF in the course of the procedure. The leak of CSF lowers the pressure in the closed system, the central nervous system, and causes irritation to the brain, nerves and other elements of the central nervous system. The CSF leak and the attendant drop in pressure are thought to cause postmyelogram headaches.
When the plaintiff arrived at the defendant’s office at the Neuro-Diagnostic Center on January 9,1975, she was told by a radiology technician that she could possibly experience postmyelogram headaches of up to six weeks duration. The defendant performed the myelogram on the plaintiff and afterwards placed her on a stretcher “for about ten minutes or so.” She was then allowed to dress and was driven home by her husband. There were no facilities at the center for a patient to lie down for a period of time exceeding ten minutes because of the times scheduled for various patients.
A card given to patients, including the plaintiff, by the defendant’s office, contained directions to “stay in *62bed, for no less than eight hours following myelogram. Patient may get up to use bathroom only, meals should be served in bed. Patient should force fluids . . . .” (Emphasis in original.) The purpose of these procedures is to decrease the pressure on the puncture site and to decrease the frequency and severity of spinal headache.
The day after the myelogram, the plaintiff began to experience headaches and telephoned the defendant’s office. Several of the plaintiff’s calls to the defendant to inform him of her headaches were not returned. The defendant finally spoke with the plaintiff about her complaint and he prescribed Valium.
On February 5, 1975, the defendant admitted the plaintiff to Yale-New Haven Hospital with a diagnosis of spinal headache as a result of continued CSF leak, and, after bed rest, she was discharged on February 13, 1975. A few days later, the plaintiff again contacted the defendant and complained of headaches. He asked her to come to Yale-New Haven Hospital as an outpatient on February 22, 1975, to undergo a procedure, called a saline bolus injection, to be performed by an anesthesiologist in order to seal the presumed CSF leak. The procedure was considered unsuccessful and no writing evidencing this procedure could be found in either the defendant’s or Yale-New Haven’s records.
The plaintiff sought treatment for her condition in the Boston area, and, from 1975 through 1982, she was admitted nine times to two hospitals for diagnostic tests and treatments. She became a patient of Richard Tyler, a physician at Peter Bent Brigham Hospital (PBBH). She underwent two venous blood patch procedures in 1975 at PBBH in an attempt to close what was considered to be a leak. In March, 1978, a myelogram was performed at PBBH. A radioisotope dye injection in March, 1978, at PBBH was done and an evaluation by *63its department of nuclear medicine stated that there was “no evidence of CSF leakage.” In May, 1979, a cervical myelogram was performed. The plaintiff was again admitted to PBBH in November, 1979. Another myelogram was performed, and an evaluation by PBBH’s department of radiology indicated that there was a leak of the Pantopaque used in the myelogram from the subarachnoid space into the epidural space, although the “location of this leak [was] not ascertained.” The discharge diagnosis was “chronic postmyelogram headache.” In January, 1980, the plaintiff was again admitted to PBBH and a laminectomy was performed by Keasley Welch, a neurosurgeon, in an attempt to discover the cause of the plaintiff’s headaches. The operative note indicated that there was no “obvious CSF leak” but that a dimple was found in the area where the defendant had performed the original myelogram. The discharge diagnosis stated that there was a “persistent cerebral spinal leak.” In September, 1982, the plaintiff was admitted to PBBH and another myelogram, using metrizamide, was performed and there was “some extravasation of dye from the lumbar area.” The discharge summary of October 18, 1982, stated that because a spinal tap had revealed a normal CSF pressure reading it “implied] that the headaches are not low pressure headaches which had been the theory . . . .” An October, 1982 report by Tyler to Morton Bender, one of the plaintiff’s treating physicians in 1982, stated that dye escaping into the lumbar area indicated that the plaintiff’s positional headache has “many characteristics of a spinal fluid leak.”
In November, 1982, the Journal of Neurosurgery contained an article written by Doctors Harrington, Tyler and Welch of Boston on the plaintiff’s condition and subsequent hospitalizations, although her identity was *64not disclosed in the article. The defendant testified that he had read the article and recognized the patient as the plaintiff.
The plaintiff offered expert testimony from Herbert Rabiner, a radiologist, and Lawrence Kaplan, a neurologist and psychiatrist. Both doctors practice in New York state but testified that they were familiar with the standards of practice for myelograms in Connecticut in 1975.3 Neither of the plaintiffs experts had examined her prior to trial. Rabiner testified that “there is a direct causal relationship between the early precipitous ambulation of this patient and the prolonged severe post-myelographic or post-spinal headaches which she suffered as a result of this early ambulation.” He further testified that instructing the patient to ambulate ten minutes following the myelogram “was a deviation from accepted medical care at the time.”
Kaplan also testified that it was a departure from acceptable practice to allow a patient “to be up and around” only ten minutes after a lumbar myelogram and that this violation was the cause of the plaintiffs headaches. He testified that the fact that the patient had not been properly confined to bed rest after the procedure caused a delay in the healing of the puncture wound and that this delay is “going to cause a persistent spinal fluid leak .... [W]hen this persistent spinal fluid leak occurs, as it did in this case, it is related to the fact that healing was not properly permitted.”
The defendant offered expert medical testimony from Franklin Robinson, a neurosurgeon associated with the defendant, William Scoville, a neurosurgeon, James *65Collias, a neurosurgeon, William Allen, a neuroradiological specialist, and Robert Shapiro, a radiologist and author of a book on myelography. Robinson testified that there was no causal relation between a patient’s ambulation ten minutes after the procedure to go home and postmyelogram headaches. Robinson testified that it was good practice to advise patients, as he does, to assume the horizontal position following a myelogram, although not necessarily immediately following the procedure. Scoville, Collias and Allen testified that it was within the standards of care exercised by physicians performing myelograms in Connecticut in 1975 to allow patients to go home directly after the procedure. Shapiro testified that in Connecticut in 1975 a patient was permitted to go home following a myelogram “as soon as in the opinion of the physician . . . the patient was suited to go home.” The textbook authored by Shapiro, Myelography (3d Ed.), a portion of which the plaintiff offered into evidence and upon which she examined, states, however, that “the patient should be instructed to lie in the prone position without a pillow for twenty-four hours.”
II
The defendant’s first claim of error is that the plaintiff failed to establish the necessary causal relation between the defendant’s performance of the myelogram and the alleged injuries on the basis of reasonable medical probabilities. The defendant also claims that because of this failure, it was error for the trial court not to grant his motions for a directed verdict and for judgment notwithstanding the verdict. We find no error.
The defendant, in his brief, does not dispute the probability of a CSF leak following the myelogram as the cause of the plaintiff’s headaches immediately thereafter, but he does dispute whether there was sufficient *66evidence of this leak continuing for eight years and causing the plaintiff’s persistent headaches. The defendant claims that because the plaintiff’s treating physicians in the Boston area did not testify, the plaintiff’s Boston hospital records form the only factual basis for the plaintiff’s claim of a causal relation between the myelogram and the headaches.
The plaintiff claims, however, that her expert witnesses, specifically relying upon the hospital records and trial testimony, provided the expert opinion on the causal connection, including the belief that the plaintiff’s disability is permanent. “Causation may be proved by circumstantial evidence and expert testimony. Slepski v. Williams Ford, Inc., 170 Conn. 18, 22, 364 A.2d 175 (1975).” Pisel v. Stamford Hospital, 180 Conn. 314, 340-41, 430 A.2d 1 (1980). In a medical malpractice case, expert testimony is necessary to establish the standard of proper professional skill or care on the part of a physician, surgeon or other similar practitioner because “in most such cases a layman does not and cannot have the requisite knowledge as to whether the proper treatment was given, procedure followed, or care used.” Chubb v. Holmes, 111 Conn. 482, 486, 150 A. 516 (1930); see Ardoline v. Keegan, 140 Conn. 552, 556, 102 A.2d 352 (1954).
The expert opinion that seeks to establish the causal connection between the injury and the alleged negligence “must rest upon more than surmise or conjecture.” Boland v. Vanderbilt, 140 Conn. 520, 525, 102 A.2d 362 (1953). “In evaluating damages in a tort action, a trier is concerned with reasonable probabilities, not with possibilities.” (Citations omitted.) Healy v. White, 173 Conn. 438, 443, 378 A.2d 540 (1977); Sheiman v. Sheiman, 143 Conn. 222, 225, 121 A.2d 285 (1956).
*67“In reviewing the evidence . . . we must give it the most favorable construction in support of the verdict of which it is reasonably capable.” Healy v. White, supra, 442. “The causal relation between an injury and its later physical effects may be established by the direct opinion of a physician, by his deduction by the process of eliminating causes other than the traumatic agency, or by his opinion based upon a hypothetical question.” Boland v. Vanderbilt, supra. The medical experts called by the plaintiff in this case did not examine the plaintiff but prior to the trial they were able to review her medical records for hospitalizations subsequent to the myelogram performed by the defendant. The defendant claims that Rabiner, one of the plaintiffs expert witnesses, never saw any of the patient’s records after the hospitalization of March 19, 1975, and that he, therefore, did not have any basis for an opinion regarding causation subsequent to that hospitalization. We need not determine whether Rabiner’s testimony on causation could have provided the jury with evidence of a causal connection beyond mere speculation because the plaintiff’s other expert witness, Kaplan, provided the needed testimony.
Kaplan, although he did not personally examine the plaintiff, reviewed all of her hospital records before he testified. “A doctor may give an opinion on a medical issue without having examined or treated the patient.” Kaye v. Newhall, 360 Mass. 701, 703, 277 N.E.2d 697 (1972); see Barksdale Lumber Co v. McAnally, 262 Ark. 379, 388, 557 S.W.2d 868 (1977); Williams v. Dawidowicz, 209 Md. 77, 87-88, 120 A.2d 399 (1956). He testified that, in his opinion based on a reasonable degree of medical probability, the plaintiffs headaches, as recently as October, 1982, were causally related to the myelogram and postmyelogram care received by the plaintiff.4 He testified further that “even if the hole *68is plugged and there is no further leakage of spinal fluid demonstrable, the patient continues to have symptoms because of the prolonged duration of time that this remained a leaking hole.” Kaplan testified that, in his opinion, the plaintiff will have “permanent difficulty” as a result of the procedure and that the headaches will persist for the rest of her natural life. This testimony, if the jury believed it, was sufficient to establish the necessary causal relation as a fact and not as a matter of conjecture or speculation. In addition, the plaintiff testified that she had not experienced severe headaches before the myelogram but that since the procedure in January, 1975, she has “never lived a day . . . without a headache and having to go to bed.”
Although the defendant stresses in his brief that all four of his expert witnesses testified that the defendant’s myelogram procedure was in accordance with the established Connecticut standard of care in 1975, it is within the jury’s province to judge the credibility of the various witnesses. See Healy v. White, supra, 444. “As to the conflicting expert testimony, the jury is free to accept or reject each expert’s opinion in whole or in part.” Slepski v. Williams Ford, Inc., supra, 22-23. In view of all the evidence, the trial court was correct in declining to exercise its power to set aside the verdict or to render a judgment notwithstanding the verdict because there was sufficient evidence for the jury reasonably and legally to reach a verdict for the plaintiff.
*69III
The defendant next claims that the trial court erred in failing to give the Secondino charge requested by him.5 See Secondino v. New Haven Gas Co., 147 Conn. 672, 165 A.2d 598 (1960). We find no error on this claim. In his brief, the defendant argues that he requested this charge to raise a negative inference “from the plaintiffs failure to produce any of the physicians who treated her during eight years of hospital admissions for her alleged persistent headaches.” He maintains that Tyler, her chief treating physician for the alleged headaches from 1975 through 1982, was one of “the treating physicians in Boston” that the plaintiff would naturally have produced. He also includes Welch in that claim and he contends that testimony by the plaintiffs own expert6 patently meets the test in *70Nichols v. Coppola Motors, Inc., 178 Conn. 335, 342, 422 A.2d 260 (1979), for showing “availability” under Secondino. Citing Nichols and Grabowski v. Fruehauf Trailer Corporation, 2 Conn. App. 167, 477 A.2d 685 (1984),7 the defendant argues that the trial court erred in failing to grant his request for a Secondino charge on the ground that he failed to satisfy his burden of producing evidence sufficient to warrant the requested charge. Moreover, he argues that the evidence “from the uncalled medical specialists in Boston” would not have been cumulative or corroborative of testimony already on the record since none of the “specialist physicians” who actually treated the plaintiff for her alleged headaches testified.
The plaintiff, however, claims that the trial court properly refused the Secondino request as the defendant had not presented any evidence that Tyler had been available. Additionally, she refers to the colloquy between the court and defense counsel, when the court was taking the exceptions to its charge, where the trial court asked: “How do we know that all of these Boston doctors aren’t in China?”8 The plaintiff also argues *71that because Tyler is from Massachusetts, he is not subject to a subpoena from Connecticut courts. In response to the defendant’s Secondino claim, it is also argued that even a doctor who has been treating a patient has not thereby agreed to travel to a different state to testify, particularly in a medical malpractice case. It is also argued that any testimony by Tyler would have been totally cumulative because all his reports and records were in evidence and referred to throughout the trial.
We conclude that the trial court properly refused to grant the Secondino charge requested by the defendant for two independent reasons.9 First, the request to charge was itself defective. The object in filing a request to charge is to inform the trial court of a party’s claim of the applicability of a principle of law to the *72case. Pickens v. Miller, 119 Conn. 553, 555, 177 A. 573 (1935); see Kast v. Turley, 111 Conn. 253, 257-58, 149 A. 673 (1930); Practice Book § 318; Maltbie, Conn. App. Proc. § 109. Our rules provide that each request to charge shall contain “a single proposition of law clearly and concisely stated with the citation of authority upon which it is based, and the evidence to which the proposition would apply. ” (Emphasis added.) Practice Book § 318; see Batick v. Seymour, 186 Conn. 632, 643, 443 A.2d 471 (1982); Ziskin v. Confietto, 137 Conn. 629, 633, 79 A.2d 816 (1951). A proper request to charge cannot, therefore, under our practice merely be a statement of an abstract proposition of law; Practice Book § 318; see Ratti v. Berry & Sons, Inc., 98 Conn. 522, 524, 119 A. 894 (1923); State v. Green, 86 N.J. 281, 290, 430 A.2d 914 (1981); any more than a court may give a charge, which though legally correct, is one “given in the abstract . . . and [is] insufficient fully to apprise the jury of the identity of the witness who was the subject of the charge.” Shea v. Tousignant, 172 Conn. 54, 59, 372 A.2d 151 (1976) (error in giving Secondino charge because given in abstract, made no reference to the evidence and did not fully apprise the jury of the identity of the “particular missing witness”). Our examination of the transcript and the exhibits indicates that Tyler and Welch appear to be those two Boston physicians who were the treating physicians. There were other physicians in Boston who treated the plaintiff, albeit not on the level of Welch and particularly Tyler. The request to charge should have specifically referred to the Boston physician or physicians to whom the Secondino request was directed. Such a reference accompanied by the evidence to which the request was claimed to apply would constitute a proper request. It would then have been adequate to permit the court to give instructions “sufficient to furnish a practical guide to the jury as to how the stated law [was] to be applied *73to the evidence before them.” Vita v. McLaughlin, 158 Conn. 75, 77, 255 A.2d 848 (1969); see Shea v. Tousignant, supra, 60. In that regard, we also note that in taking the exceptions, the trial court specifically pointed out to defense counsel that the request did not refer to “any specific witness.” Despite that, no names of any such witnesses were stated to the court at that time.
The second reason that the trial court properly declined to charge on Secondino as requested was because the defendant did not sustain his burden of demonstrating the availability of the witness. See, e.g., State v. Alfonso, 195 Conn. 624, 631, 490 A.2d 75 (1985); Doran v. Wolk, 170 Conn. 226, 229, 365 A.2d 1190 (1976); New England Whalers Hockey Club v. Nair, 1 Conn. App. 680, 685, 474 A.2d 810 (1984). The Secondino rule is: “ ‘The failure of a party to produce a witness who is within his power to produce and who would naturally have been produced by him, permits the inference that the evidence of the witness would be unfavorable to the party’s cause.’ ” Secondino v. New Haven Gas Co., supra, 675, quoting Ezzo v. Geremiah, 107 Conn. 670, 677, 142 A. 461 (1928). The party against whose cause an unfavorable inference is claimed may, of course, offer evidence to explain the failure to produce the witness. “There are two requirements for the operation of the rule: The witness must be available, and he must be a witness whom the party would naturally produce.” Secondino v. New Haven Gas Co., supra.
There would appear to be no real dispute between the parties that Tyler would be a witness whom the plaintiff would naturally have produced under Secondino. We need not, however, reach that question because of the basis of our disposition of this claim. The defendant claims that Tyler was “available” under that rule and he points to the following language in Nichols v. *74Coppola Motors, Inc., supra, 342, to support his claim: “Availability may be shown or determined not only from mere physical presence or accessibility for service, but also from the relationship, usefulness or nature of the expected testimony and this means only that the witness is in such relationship with the party that it is likely that his presence could be procured.” Our statement in Nichols must be considered in the light of our discussion of “availability” in Secondino and its progeny. Whether a particular witness is “available” is one of two distinct requirements for the operation of the rule. Secondino v. New Haven Gas Co., supra, 675. The language of Nichols, as we now examine it, overlaps the two distinct requirements of the rule. Although this charge has been dubbed the Secondino charge, “it appears to have been first formally recognized in Ezzo v. Geremiah, supra.” State v. Daniels, 180 Conn. 101, 109, 429 A.2d 813 (1980). In Ezzo, and later in Secondino, the rule is expressly linked to the maxim that “ ‘all evidence is to be weighed according to the proof which it was in the power of one side to have produced, and in the power of the other side to have contradicted.’ ” Ezzo v. Geremiah, supra; Secondino v. New Haven Gas Co., supra; see State v. Alfonso, supra, 632; Turner v. Scanlon, 146 Conn. 149, 161, 148 A.2d 334 (1959).
While we have not said and do not say that to be available a witness must be subject to subpoena, it is apparent that a witness is “available” if the party against whom the negative inference of Secondino is claimed is able to procure the witness’ testimony at the trial. See State v. Bennett, 171 Conn. 47, 57, 368 A.2d 184 (1976). There is little question that availability is related to the naturalness of production requirement. The notion of the power and the ability to procure the witness, by process or otherwise, is critical to availability. Before a negative inference can be drawn from a *75party’s failure to produce a particular witness, it must be shown that the party was able to procure the witness’ physical presence in court. This capability of the naturalness of production of a witness has caused Wig-more to state that it “is unquestioned” that the person involved must be “within the power of the party to produce.” (Emphasis in original.) 2 Wigmore, Evidence (Chadboum Rev.) § 286.10 This is consonant with our view of the meaning of “available.” It accords with common sense, especially if it is kept in mind that wisdom counsels caution in the application of this inference. See McCormick, Evidence (3d Ed. 1984) § 272; 2 Wigmore, supra, § 285. It also is borne out by our implicit overruling in Queen v. Gagliola, 162 Conn. 164, 169, 292 A.2d 890 (1972), of “such cases as Cascella v. Jay James Camera Shop, Inc., 147 Conn. 337, 343, 160 A.2d 899 [1960], which placed the burden of proof of nonavailability on the party who would naturally have produced the witness.” Doran v. Wolk, supra, 229. Of course, “ ‘[w]hether an inference can be drawn from the failure to call witnesses necessarily depends, as with inferences generally, upon the posture of the particular case and the state of the evidence.’ ” Grady v. Collins Transportation Co., 341 Mass. 502, 506, 170 N.E.2d 725 (1960), quoting Commonwealth v. O’Rourke, 311 Mass. 213, 222, 40 N.E.2d 883 (1942). The Secondino charge, therefore, should not be given unless there is a sufficient foundation to do so. In a malpractice case, one court, in finding error in giving the negative inference charge, stated, in adopting Griffin v. California, 380 U.S. 609, 614, 85 S. Ct. 1229, 14 L. Ed. 2d 106 (1965): “ ‘What the jury may infer, given no help from the court is one thing. What it may infer when the court *76solemnizes the silence . . . into evidence . . . is quite another.’ ” Wild v. Roman, 91 N.J. Super. 410, 415, 220 A.2d 711 (1966).
The defendant did not sustain his burden of demonstrating the availability of Tyler.11 The statements of Kaplan that Tyler “is in charge of neurology” at Peter Bent Brigham Hospital (PBBH), which was part of the Massachusetts General/Harvard Medical School in Boston, fall far short of that burden. Moreover, the defendant does not claim that there is any other evidence nor do we find any other evidence on this point.
IV
The defendant claims that the trial court erred in denying his motion for a directed verdict and in denying his motions to set aside the verdict and for judgment notwithstanding the verdict on the issue of informed consent because the plaintiff failed to produce expert medical testimony that a spinal headache of long duration is a risk of the myelogram procedure. Because we uphold the judgment in favor of the plaintiff on the issue of actionable negligence and because the jury’s award of damages on both issues is not severable, we need not address the informed consent issue since the damages awarded may stand on the negligence issue alone.12
V
The defendant raises as a fourth claim of error several of the trial court’s evidentiary rulings. The defendant claims that a hypothetical question posed by the *77plaintiff to her expert witness, Kaplan, was defective because it assumed a fact not in evidence. The defendant objected, the court overruled the objection, and an exception was taken. The plaintiff claims that the defendant, on cross-examination of the expert, was allowed to prove the issue that he had objected to in the hypothetical-—whether on each of the plaintiffs hospitalizations over the last eight years “she’s there because of a post-spinal headache . . . .’’The defendant claims that the plaintiff’s hospital records do not bear out that assumption and points to a discharge summary of the plaintiff from PBBH in 1982 which indicates that the headaches may be of a “possible” vascular etiology.
In Floyd v. Fruit Industries, Inc., 144 Conn. 659, 666, 136 A.2d 918 (1957), we stated that the determination of the admissibility of a hypothetical question “calls for the exercise of a sound discretion as to whether the question, even though it does not contain all of the facts in evidence, presents the facts in such a manner that they bear a true and fair relationship to each other and to the whole evidence in the case . . . is not so worded as to be likely to mislead or confuse the jury, and is not so lacking in the essential facts as to be without value in the decision of the case.” (Citations omitted.) The court did not abuse its discretion in admitting the question.
The hypothetical, while not a model of clear speech, asked the expert to assume that the plaintiff was hospitalized because of her postspinal headache. What a patient reports as her symptoms upon admission to a hospital is not conclusive upon the treating physician as to the actual diagnosis. The plaintiff’s hospital records reflect, in the admission histories, a continued complaint of headache that began following the 1975 myelogram. We note that the plaintiff’s Yale-New Haven and Boston hospital records were in evidence *78as full exhibits prior to the hypothetical question involved and also that Kaplan had reviewed these records. The hypothetical question cannot be said to have distorted the facts or to have misled or confused the jury.
The defendant’s next claim of error in evidentiary rulings is in the court’s application of the “best evidence” rule to certain testimony of Robinson, one of the defendant’s expert witnesses. The expert witness was asked on redirect examination: “[I]n the period of time . . . from 10/1/73 through 9/30/74, approximately, sir, how many ambulatory myelograms were performed by your group during that period?” The plaintiff objected and claimed that the best evidence rule barred this testimony unless it was introduced through records kept by the group. The best evidence rule requires, where the contents of a document are to be proved, that the original document be offered into evidence unless such production is excused. See Brown v. Connecticut Light & Power Co., 145 Conn. 290, 296, 141 A.2d 634 (1958); Tait & LaPlante, Handbook of Connecticut Evidence § 10.7, p. 169; McCormick, Evidence §§ 229 through 233. The court sustained the objection and stated that “the best evidence would be the records of what the actual number of myelograms were, rather than the expression of his opinion at this point as to how many were done during that course of time.” The defendant took an exception to the ruling. We agree with the defendant that the question posed by him did not seek to prove the contents of a document. The defendant claims that the question instead sought to elicit information about Robinson’s experience in performing ambulatory myelograms. The trial court referred not only to the best evidence rule in sustaining the objection but also to the apparently inadequate foundation for Robinson’s testimony on this subject. The trial court stated that the period in ques*79tion was an approximately eleven or twelve month period ten years prior to trial and that there had been a “long lapse of time” since then. The trial court did not abuse its discretion in sustaining the objection as the question was based on an inadequate foundation. Alternatively, the defendant has not demonstrated harmful error even if the trial court’s ruling is considered to have been based solely on the best evidence rule. The burden rests upon the defendant to demonstrate the harmfulness of the error. State v. Januszewski, 182 Conn. 142, 174, 438 A.2d 679 (1980), cert. denied, 453 U.S. 922, 101 S. Ct. 3159, 69 L. Ed. 2d 1005 (1981). Because Robinson had previously testified that he had performed “thousands” of ambulatory myelographies “up through the month of January, 1975,” there was sufficient evidence before the jury of Robinson’s professional experience in this area so that the ruling cannot be said to have been harmful error.
VI
The defendant also claims error in the court’s denials of his motions for mistrial because the “cumulative harm” resulting from the actions of the plaintiff’s counsel and the remark of the plaintiff’s witness created “insurmountable prejudice” against him. The defendant specifically cites five incidents during the trial, for which each time a motion for mistrial was made, denied, and an exception taken. The first incident involved a remark by the plaintiff’s counsel concerning the formation of the codefendant professional corporation, about which he stated: “It goes to the motive of this doctor in trying to steer his patients into this corporation for billing purposes as opposed to a hospital.” The trial court, stating that the question of financial arrangements was totally irrelevant and immaterial to the case, instructed the jury to disregard the comment.
*80The second and third incidents claimed by the defendant to require a mistrial involved the manner in which the plaintiff’s counsel “browbeat” and “badgered” the defendant during his testimony. The fourth incident involved the plaintiff’s daughter, who when on redirect examination was asked if she felt “a sense of anger and hostility about what happened to [her] mother,” responded: “Anger? I’d like to kill that man [the defendant] right there.” The court instructed the jurors to disregard the last question and answer and instructed them that “the mere fact that there is an emotional response ... is not to have any significance to you as to the merits of this case.” The final incident raised by the defendant is the “browbeating” of Robinson during the plaintiff’s cross-examination. The trial court, after the particular incident at issue, again directed the jury to disregard the comments.
A mistrial, as a general principle, should only be granted when it is apparent to the trial court that because of an incident during the trial a party cannot have a fair trial. See State v. Fleming, 198 Conn. 255, 264, 502 A.2d 886 (1986); State v. Festo, 181 Conn. 254, 265, 435 A.2d 38 (1980); Izzo v. Crowley, 157 Conn. 561, 565, 254 A.2d 904 (1969); Ferino v. Palmer, 133 Conn. 463, 466, 52 A.2d 433 (1947). The trial court has wide discretion in ruling on motions for mistrial. See State v. Dolphin, 195 Conn. 444, 453, 488 A.2d 812, cert. denied, 474 U.S. 833, 106 S. Ct. 103, 88 L. Ed. 2d 84 (1985); State v. Festo, supra; State v. Piskorski, 177 Conn. 677, 719-20, 419 A.2d 866, cert. denied, 444 U.S. 935, 100 S. Ct. 283, 62 L. Ed. 2d 194 (1979). “Moreover, we have consistently recognized that a careful, cautionary instruction given by the trial court is a relevant factor to be considered in determining whether the denial of such a motion constituted an abuse of the trial court’s broad discretion. See State v. Ruiz, 171 Conn. 264, 274, 368 A.2d 222 [1976]; State v. Savage, *81[161 Conn. 445, 449, 290 A.2d 221 (1971)]; State v. Miller, 154 Conn. 622, 631, 228 A.2d 136 [1967].” State v. Piskorski, supra, 720.
In the present case, the trial extended over several weeks and included extensive direct and cross-examination of the various witnesses. We have carefully reviewed the lengthy transcript in this case and find no error in the trial court’s denials of the defendant’s motions for mistrial. In the context of the entire trial, the above incidents involving the defendant or Robinson were reflective of the intense questioning by counsel from both sides. While not entirely proper or commendable examination tactics were used by the plaintiff’s counsel, the trial court did not abuse its discretion in denying the motions. As to the incident involving the plaintiffs daughter, it was an isolated incident in the trial. The trial court heard counsel for each of the parties before ruling on the motion, including the plaintiff’s counsel’s statement that the response was not anticipated or elicited but that it was “spontaneous.” “In making our independent assessment we do so under the limitations of a printed record, without the benefit of instant replay or of the sights and sounds of a trial in action. The trial judge, on the other hand, is in a better position to sense the atmosphere of the trial and therefore can apprehend far better than we can the effect of certain remarks on the jury. Marko v. Stop & Shop, Inc., 169 Conn. 550, 559, 364 A.2d 217 (1975); Butler v. Steck, 146 Conn. 114, 119, 148 A.2d 246 (1959).” Pisel v. Stamford Hospital, 180 Conn. 314, 322, 430 A.2d 1 (1980); see Pietrorazio v. Santopietro, 185 Conn. 510, 515, 441 A.2d 163 (1981). The trial court did not abuse its discretion in denying the defendant’s motions for mistrial.
VII
The defendant’s final claim of error is that the jury’s award of $800,000 in damages was excessive. The trial *82court denied the defendant’s motion to set aside the verdict and for judgment notwithstanding the verdict and rendered judgment for the plaintiff for $800,000. In his brief, the defendant claims that because the headaches suffered by the plaintiff cannot be causally connected to the defendant’s performance of the myelogram beyond a period of a few months, the award is “patently outrageous.” The defendant, in oral argument to this court, conceded that this claim would fail if we were to affirm the judgment on the negligence issue and find that causation was established. Because of our ruling on the causation issue, and in light of the defendant’s concession on this issue, we need not discuss this claim further.
There is no error.
In this opinion the other justices concurred.
The references to the plaintiff encompass both Charlotte and Hyman Shelnitz, unless otherwise stated.
References to the defendant encompass both Alvin Greenberg, M.D., and Neurosurgical Associates of New Haven, P.C.,unless otherwise stated.
The plaintiffs expert witnesses testified that they were familiar with the standards in Connecticut for performing myelograms in 1975. We note that since the trial in this case we have expanded the geographical standard for expert witnesses testifying in medical malpractice cases to “include the entire nation.” Logan v. Greenwich Hospital Assn., 191 Conn. 282, 301, 465 A.2d 294 (1983).
The defendant claims that Kaplan’s testimony concerning the venous blood patches performed by a physician in Boston in March, 1975, con*68tradicts the opinion of the physician in the medical records. Kaplan’s opinion that the venous blood patches “were not successful” is not directly contradictory, as the medical record indicates that the treating physician considered the second blood patch successful as it “relieved one component of her headache.” The evidence of the treating physician’s opinion of the patches in the hospital record was before the jurors for their consideration and determination whether Kaplan’s testimony was contradictory.
The defendant’s request to charge was the following: “The failure of a party to produce a witness whom it was within his power to produce and who would have naturally been produced by him permits an inference that the evidence of the witness would have been unfavorable to the party’s cause. A witness who would naturally have been produced by a party is one who was known to the party and who, by reason of his relationship to the party or to the issues, or both, could reasonably have been expected to have peculiar or superior information which was material to the case and would have been produced had it been favorable.” Secondino v. New Haven Gas Co., 147 Conn. 672, 165 A.2d 598 (1960).
The defendant’s brief points to the following from the direct examination of Kaplan concerning Tyler:
“Q. ... Do you know Dr. Tyler?
“A. Oh yes; very well.
“Q. Who’s Dr. Tyler, sir?
“A. Dr. Tyler is in charge of neurology at Peter Bent Brigham.
* ** *
“Q. Is he considered one of the authorities in the field of neurology, sir?
“A. Yes.
“Q. Nationwide?
“A. For a young man, I would say he is, yes. He is young.
“Q. And, sir, do you know the institution with which he is connected, the medical institution?
* * *
“A. Peter Bent Brigham is part of the Massachusetts General/Harvard Medical School. It’s a private hospital.”
In Grabowski v. Fruehauf Trailer Corporation, 2 Conn. App. 167, 477 A.2d 685 (1984), which involved a motor vehicle accident in Meriden, the Appellate Court sustained a Secondino charge, concerning a Meriden physician, which incorporated language from Nichols. In that case, there was testimony from two other physicians that the physician involved, who was the plaintiff’s family physician, was practicing medicine in Meriden “at the time of trial.” Grabowski v. Fruehauf Trailer Corporation, supra, 171.
During the course of taking defense counsel’s exceptions to the charge, the following took place:
“The Court: Incidentally, Attorney Willis, on that Secondino charge that you requested made.
“Mr. Willis: Yes.
“The Court: Now, you didn’t refer in that to any specific witness who the plaintiff failed to produce. Now, generally don’t you have to put on evidence in the course of the trial to show that that particular witness is available and within the control?
“Mr. Willis: No, if your Honor please.
“The Court: How do we know that all of these Boston doctors aren’t in China?
*71“Mr. Willis: It’s up to the plaintiff to establish that they were not available, if it was a witness that they should have called under the Secondino case.
“The Court: I thought under Secondino the burden is on the side that wants the inference charged to put on evidence to show that that doctor is available and within the control of the opposing party to produce him.
“Mr. Willis: I disagree with your Honor’s interpretation.
“The Court: We have a difference of opinion as what the Secondino—
“Mr. Willis: Right.
“The Court: —inference is.
“Mr. Willis: Yes.”
The posture of the defendant’s Secondino claim is perplexing. His request to eharge identified no missing witness at all. The colloquy at the close of the trial court’s charge during which the court referred to the “Boston doctors” elicited no name identification of any missing witness or witnesses. The defendant’s brief, while referring to Tyler and Welch as the treating physicians, focuses almost entirely, if not entirely, on Tyler. No other Boston physicians are named in his brief on this issue although the plaintiff did see other physicians, according to the transcript, in Boston. The defendant only refers us to evidence concerning Tyler’s “availability.” Consequently, mindful that it is the appellant’s obligation to ensure that this court is provided with an adequate appellate record to support his claims of error; State v. One 1977 Buick Automobile, 196 Conn. 471, 480, 493 A.2d 874 (1985); DeMilo v. West Haven, 189 Conn. 671, 681, 458 A.2d 362 (1983); we review this claim as to Tyler. We add, however, that there would be no basis, were we to reach it, upon which to review this claim as to any other of the “Boston doctors.”
The Court of Appeals for the Second Circuit cast it in terms of “control” in Savard v. Marine Contracting, Inc., 471 F.2d 536, 542 (2d Cir. 1972), cert. denied sub nom. Savard v. Perini Corporation, 412 U.S. 943, 93 S. Ct. 2778, 37 L. Ed. 2d 404 (1973), when it said: “No inference can be drawn from failure to produce evidence not within the party’s control.”
This also applies to Welch, who presumably was one of the “Boston doctors” who was not identified in the request or the colloquy on the exceptions.
The jury responded affirmatively to each of the following two interrogatories that were submitted to it: “1.) Do you find for the plaintiffs on Counts 1 and 3 which relate to medical malpractice? 2.) Do you find for the plaintiffs on Counts 2 and 4 which relate to lack of informed consent?”