dissenting. I concur with Justice Shea that it was improper for the trial court to admit into evidence all the testimony regarding the physiological effects of cocaine for the reasons stated in his concurring opinion. Nevertheless, I dissent from the result *528because I believe that this testimony was prejudicial and, therefore, deprived the defendant of a fair trial.
Although the majority concedes that the trial court improperly admitted into evidence the “possibl[y] prejudicial” expert testimony that cocaine use can cause “instantaneous death,” somehow it concludes that this was harmless. We have long held that “ ‘improper evidence that may have a tendency to excite the passions, awaken the sympathy, or influence the judgment of the jury, cannot be considered as harmless.’ ” State v. Loughlin, 149 Conn. 21, 26, 175 A.2d 367 (1961), quoting St. Martin v. New York, N.H. & H. R. Co., 89 Conn. 405, 411, 94 A. 279 (1915).
In the present case, Charles Reading, supervisor for the state of Connecticut toxicology laboratory, testified that cocaine absorbed through the body’s mucous membranes “results in a very intense physiological reaction. One of those reactions is a stimulation of the pleasure centers of the central nervous system. Another reaction, which is particularly a problem, is the irritation of the myocardium, that is the actual tissue of the heart itself, which has been recorded in many cases to cause almost instantaneous death.”1 Unlike the situation in Lane v. Commonwealth, 223 Va. 713, 292 S.E.2d 358 (1982), where the Supreme Court of Virginia concluded that the erroneously admitted testimony regarding a drug’s ill effects did not warrant reversal, Reading did not limit his improper remarks *529to instances of excessive use.2 Rather, as the defendant aptly notes, Reading’s testimony portrayed any use of cocaine as a form of “Russian roulette,” thereby inflaming the passions of the jury by implanting in their minds the notion that the defendant was peddling death for profit.
In similar circumstances, other jurisdictions have found that evidence of a drug’s physiological effect prejudices the defendant to the extent that fairness requires a new trial. United States v. Anderson, 584 F.2d 849, 852 (6th Cir. 1978); Smith v. Commonwealth, 223 Va. 721, 724-25, 292 S.E.2d 362 (1982).
Furthermore, the majority presumably relies upon the following from State v. Brigandi, 186 Conn. 521, 543, 442 A.2d 927 (1982), with which I have no quarrel: “ ‘The general principle is that a mistrial should be granted only as a result of some occurrence upon the trial of such a character that it is apparent to the court that because of it a party cannot have a fair trial . . . and the whole proceedings are vitiated.’” (Citations omitted.) In the present case, however, the majority fails to offer any analysis explaining how it came to the conclusion that, even though the “instantaneous death” testimony had no probative value and had a tendency to prejudice the defendant, he was not deprived of a fair trial.
Moreover, the majority fails to set forth the standard of review it applied. Based on the particular circumstances in Brigandi, this court stated that trial courts *530have “wide discretion in passing on motions for mistrial.” Id. The abuse of discretion standard, however, would be inappropriate in the present case because the trial court’s denial of the motion for mistrial was predicated on its erroneous ruling that the testimony had probative value and was, therefore, admissible. Had the trial court correctly ruled that the testimony about “instantaneous death” was irrelevant and had it stricken from the evidence, then the trial court’s decision to deny the motion for a mistrial would be entitled to an abuse of discretion standard of review. Obviously, this deferential standard should not be used when an erroneous evidentiary ruling provided the basis for the trial court’s decision on the mistrial motion.
I would find that admission of all such testimony pertaining to the physiological effects of the drug to be prejudicial. Accordingly, I would set aside the conviction and remand for a new trial. I, therefore, respectfully dissent.
Clearly, Reading’s testimony had an impact upon the jury. Reading, whose expertise as the supervising toxicologist for the state’s toxicology laboratory added credence to his testimony, predicated his testimony on his education, i.e., a bachelor’s degree in chemistry from the University of Utah and a doctoral degree in forensic toxicology from the University of Maryland, his membership in the American Board of Forensic Toxicologists as a diplomat and past vice-president, and his twenty years of experience as a toxicologist.
The Supreme Court of Virginia in Lane v. Commonwealth, 223 Va. 713, 717-18, 292 S.E.2d 358 (1982), concluded that “[u]nlike the testimony we considered in Smith [v. Commonwealth, 223 Va. 721, 292 S.E.2d 362(1982)], McGarry’s [testimony] was not such as would excite the passions of the jury. While he testified that excessive use of the drug is dangerous, he was careful to state its normal effect is that produced by a sleeping pill. We view this testimony as being innocuous, and while it was erroneously admitted, it does not justify a reversal of Lane’s conviction.”