Savina v. Sterling Drug, Inc. ( 1990 )

247 Kan. 105 (1990) 795 P.2d 915

RICHARD K. SAVINA, Appellant,
v.
STERLING DRUG, INC., GUST H. NELSON, M.D., and ST. JOSEPH MEDICAL CENTER, INC., Appellees.

No. 63,655

Supreme Court of Kansas.

Opinion filed July 13, 1990.

Richard D. Cordry, of Wichita, argued the cause, and Frances A. Hartman and Sharon A. Werner, of the same firm, were with him on the briefs for appellant.

Donald Patterson, of Fisher, Patterson, Sayler & Smith, of Topeka, argued the cause, and Steve R. Fabert, of the same firm, was with him on the brief for appellee Sterling Drug, Inc.

Hal D. Meltzer, of Turner & Boisseau, of Wichita, argued the cause, and W. John Badke II, of the same firm, was with him on the brief for appellee Gust H. Nelson, M.D.

Amy S. Lemley, of Foulston & Siefkin, of Wichita, argued the cause, and Jay F. Fowler, of the same firm, was with her on the brief for appellee St. Joseph Medical Center.

Dwight A. Corrin, of Corrin & Krysl, Chartered, of Wichita, was on the brief for amicus curiae Kansas Trial Lawyers Association.

The opinion of the court was delivered by

ALLEGRUCCI, J.:

This is a products liability action against Sterling Drug, Inc., (Sterling Drug) and a medical malpractice action against Gust H. Nelson, M.D., and St. Joseph Medical Center, Inc., in Wichita. Plaintiff alleges that the paralysis he experiences in the lower part of his body was due either to the dye metrizamide, used in his myelogram, or to the negligent performance of the myelogram procedure. Plaintiff appeals from the order of the district court granting summary judgment to each of the defendants.

The district court entered three separate journal entries, one as to each of the defendants. As to Sterling Drug, the court made 34 separate findings of uncontroverted material facts. As to Dr. Nelson, the facts set forth in his memorandum brief were adopted by the court, and summary judgment was granted to appellee St. Joseph Medical Center based "on the uncontroverted and undisputed facts presented by all the parties."

Since the trial court resolved this case by summary judgment, the facts need to be recited in detail. In October 1981, plaintiff first saw Dr. Schnelle and Dr. Miller, who practiced together, for pain in the lumbar region of his back. Plaintiff saw Dr. Schnelle again on April 11, 1983, and was admitted into St. Joseph Medical Center for pain in his lower back. After three days in traction, a myelogram was performed upon Dr. Schnelle's recommendation.

A myelogram is a procedure used to obtain a picture of an individual's spinal column. Although the picture can be done by fluoroscope or x-ray, soft tissues do not photograph unless a contrast medium is used. In April 1983, two radiopaque contrast agents were commonly used in myelograms: Pantopaque and Amipaque, the brand name for metrizamide. The contrast agent is inserted by needle into the thecal sac, which is the sheath of dura mater enclosing the spinal cord. The contrast medium is inserted into the spine below the spinal cord. In plaintiff's case, it was inserted between lumbar 2 and lumbar 3 (L-2 and L-3) discs. The puncture that occurs when the thecal sac is penetrated by the needle, and which remains after the needle is withdrawn, reduces the hydrostatic pressure surrounding the spinal column. If a higher-level obstruction exists above the area of the puncture, such as a hematoma cyst or thoracic-level herniated disc, the drop in hydrostatic pressure in the lower part of the thecal sac can increase pressure on the spinal cord at the higher obstruction, which can cause additional injury.

Pantopaque is an oil-based contrast medium that will not mix with spinal fluid. Pantopaque is neurotoxic and must be removed from the thecal sac after the myelogram is completed. Because removing all of the Pantopaque after the myelogram is mechanically impossible, the patient continues to risk a neurological problem, called arachnoiditis. Dr. Nelson, who performed the myelogram on plaintiff, preferred using Pantopaque in the cervical area but, otherwise, preferred metrizamide.

Metrizamide, which is the active ingredient of Amipaque, is a water-soluble contrast medium developed more recently than Pantopaque. Because it is water soluble, Amipaque is metabolized by the surrounding tissues and absorbed into the bloodstream after the procedure is completed, making manual removal unnecessary. Metrizamide is less dense and fills the nerve root sleeves better than Pantopaque, giving a nicer image aesthetically and diagnostically.

Both Pantopaque and Amipaque were used by physicians in Wichita. After plaintiff's myelogram, defendant St. Joseph Medical Center returned to using Pantopaque exclusively until a radiopaque contrast agent known as Omnipaque was introduced. Pantopaque is still available today and is used by some radiologists.

Plaintiff was not told that he risked developing either temporary or permanent paralysis of his lower extremities as a result of a myelogram using Amipaque or metrizamide. Dr. Nelson explained to plaintiff the procedure used in performing a myelogram and the probability of headaches, nausea, and vomiting afterwards. A patient consent was obtained by a student nurse at St. Joseph Medical Center. She stated that the consent was obtained the night before the myelogram, although the date and time on the form indicates it was signed after the procedure was completed.

Dr. Nelson, a board-certified radiologist, performed the myelogram on April 14, 1983. A myelogram tray was prepared by Mr. MacDonald, a radiology technician. He mixed 16 cc's of diluent (a diluting agent) with a full ampule of metrizamide. A syringe and needle is used to draw the diluent from a bottle with a rubber stopper and to then place it into an ampule of metrizamide. The two are mixed for delivery into the spinal column.

Dr. Nelson explained each step to plaintiff as he performed the myelogram. First a local anesthetic, Xylocaine, is injected into the patient's back. The metrizamide dye is then injected into the thecal sac, and the table is lowered to a 40° angle. Plaintiff felt the local anesthetic, needle, and dye going into his spine. He expressed discomfort and then fainted when the local anesthetic was injected. After waiting for the fainting to clear, Dr. Nelson injected the dye and noticed no immediate reaction. Dr. Nelson believed plaintiff had a herniated disc at lumbar 5 — sacral 1, pointing to this on the monitor during the myelogram procedure.

The myelogram procedure, which took about 20 minutes, was completed by 9:45 a.m. Plaintiff was placed in the hallway of the x-ray department to await transfer to his room. Approximately two to three minutes after being placed in the hall, plaintiff felt numbness, swelling, and pain from his waist to his toes. His feet felt like they were swelling and on fire until he was placed on pain medication. Plaintiff remained in the hallway approximately 30 minutes before being transported to his room.

Dr. Nelson learned of plaintiffs complaints when the nurse telephoned at approximately 11:30 a.m. Dr. Nelson had never heard of a patient complaining of burning and swelling in the lower legs after a myelogram procedure. Nor had Dr. Nelson ever heard of a case of permanent paralysis following a metrizamide myelogram. Plaintiff had pain in his legs and trouble using his legs and arms. Dr. Nelson instructed the nurse to call Dr. Schnelle. The two doctors met at approximately 1:30 p.m. and decided to consult a neurologist. Dr. Dilawer Abbas, a neurologist, saw plaintiff for the first time at St. Joseph Medical Center on April 14, 1983, at approximately 5:00 p.m. He found a decreased sensation at the L-4/L-5 level and no sensation below the L-3/L-4 level.

The evening of April 14, Dr. Schnelle, Dr. Hered, who was a neurosurgeon brought in for consultation, Dr. Abbas, and Dr. Nelson discussed plaintiffs condition. They decided to attempt a second myelogram to determine whether a mechanical obstruction existed higher in the spinal cord, such as a tumor or hematoma. Dr. Nelson placed a needle between the L-3/L-4 interspace and attempted to place a second needle at the L-2/L-3 interspace, but no fluids were returned in either needle. The contrast medium being used, Pantopaque, was not inserted because the thecal sac had collapsed due to the loss of fluid from the earlier myelogram. Collapse of the thecal sac is not an unusual occurrence after it has been punctured for a spinal tap. When the second myelogram failed, a computerized tomography (CT) scan was completed. The radiologist who read the CT scan did not find evidence of a hematoma or lesion at or above the level of the initial myelogram that would explain plaintiffs condition.

Currently, plaintiff has loss of motor control and sensation in his lower legs, can walk only with the aid of crutches and braces, has some bladder and bowel dysfunction, and has sexual dysfunction. All are considered permanent conditions. Three and a half years after the first myelogram by Dr. Nelson, a diagnostic device, Magnetic Resonance Imaging (MRI), revealed a thoracic-level herniated disc. MRI was not available in Wichita in 1983.

When the Amipaque was purchased, St. Joseph Medical Center had a contrast medium guide from Sterling Drug containing a reprint of the package insert and information available in the Physicians Desk Reference (PDR). This information gave detailed instructions, contraindications, warnings, risk of use, recommendations as to patient selection, and possible side effects then known. This was information concerning metrizamide contained in the PDR in 1982.

The record before the district court indicated that defendant Sterling Drug prepared drug experience reports (DER) during the development and marketing of metrizamide that indicated patients had suffered from temporary or transient paralysis after myelography with metrizamide. Plaintiffs three experts reviewed the DER prepared by Sterling Drug prior to April 1983, the package insert distributed with metrizamide, and plaintiff's medical records. Plaintiffs experts concluded that the DER compiled by Sterling Drug prior to April 1983 indicated that both transient and permanent paralysis could be caused by metrizamide. Therefore, in their opinion, the warnings contained in the package inserts were insufficient to adequately advise a physician using metrizamide of either the potential for an adverse reaction resulting in paralysis or the proper treatment to be employed in the event of paralysis.

One of plaintiff's experts, Dr. Berger, who is a neurologist and professor at Albert Einstein University, found at least 12 cases where patients suffered adverse reactions similar to plaintiff's after a myelogram using metrizamide. Because Sterling Drug did not check back with any doctor who reported an adverse reaction, it was impossible to ascertain whether the reported paralysis was in fact transient. According to Dr. Berger, Sterling Drug should have prepared a stronger, more explicit warning that severe transient or permanent paralyses were possible side effects of the use of metrizamide. Dr. Schaumburg, a neurologist and toxicologist as well as a professor at Albert Einstein University, agreed that Sterling Drug was not aggressive or emphatic enough in warning about the possible damage to nerve roots in using metrizamide. He indicated the package insert did not contain sufficient warning or discussion of appropriate subsequent treatment should a problem arise.

Dr. Heinz, who is a neuroradiologist and professor of radiology at Duke University, criticized the package inserts for referring to the complaints reported in the DER as "transitory." He noted that, for most physicians, the term "transitory" means a few seconds or minutes instead of three weeks as was reported in one DER. At the pretrial conference, plaintiff indicated that he would rely primarily upon the testimony of Dr. Heinz to establish the liability of the defendants. All parties used Dr. Heinz's deposition to support their respective positions regarding the claims raised in their motions for summary judgment and the response thereto.

At the deposition conducted on October 15, 1987, plaintiff's counsel asked Dr. Heinz if he had an opinion based upon a reasonable degree of medical probability of whether one would expect the results of a lumbar myelography with metrizamide as experienced by plaintiff to occur "but for the negligence of either the manufacturer of the drug or the negligence of the administrator of the drug, i.e., Dr. Nelson, or the negligence of the hospital through its personnel in mixing or preparing the drug for injection." Dr. Heinz indicated he did have an opinion. First, he found no question of negligence regarding the hospital. Second, he saw no direct evidence of negligence by Dr. Nelson. Third, he considered the drug itself. Knowledge about the drug dated from a 1947 paper which recognized that the concentration and duration of contrast material in contact with neighboring tissue are two factors indicating the constituency of neurotoxicity. Based upon cases where transient, persistent, or even chronic changes occurred in relation temporally to the placement of metrizamide, the doctor's opinion was that plaintiffs injury was probably caused by metrizamide and, perhaps, its persistence in his neural sac for a longer-than-average period of time. Elsewhere during his testimony, Dr. Heinz also noted that, when a therapeutic procedure such as opening the skull, the chest, or abdomen is performed, complications might be expected, but when the procedure is merely diagnostic, a significant complication is not anticipated. Dr. Heinz recognized that plaintiff had a more severe reaction than had been previously reported from the use of metrizamide, but neurologically his symptoms were quite similar, just more severe, than other patients who had experienced, but recovered from, some degree of disability following a myelogram using metrizamide.

Defendant Sterling Drug denied that metrizamide caused the injury plaintiff experienced following the myelogram. Plaintiff claimed that metrizamide caused his injuries and that Sterling Drug was aware that permanent paralysis was a potential adverse effect but did not discuss it in the package inserts.

Dr. Nelson denied he was negligent, but plaintiff pointed out that he used an 18-gauge needle to inject metrizamide instead of the 22-gauge needle recommended. Plaintiff does not offer evidence indicating that using a different size needle would cause an injury resulting in the symptoms plaintiff experienced. Nor can plaintiff identify any evidence to support his claim that Dr. Nelson was negligent, except for the fact that plaintiff was injured following the myelogram.

St. Joseph Medical Center also denied negligence by any of its staff and employees. Plaintiff claims that St. Joseph Medical Center's radiology technician, MacDonald, may have been negligent in the preparation of the metrizamide solution. Plaintiff notes that a phenomenon called "coring" can occur when a solution is contaminated during preparation by a part of the rubber stopper entering into the needle when the needle is placed in the rubber stopper to withdraw medication. If coring does occur, the foreign material may not be identified in a myelogram or a CT scan. Defendant Sterling Drug's expert, Dr. Sackett, a neuroradiologist, admitted that if coring occurred, an injection of the contaminated solution into the patient's spinal cord could cause injury. Here, MacDonald, following standard procedures, discarded the metrizamide and unused diluent from the first myelogram.

Dr. Nelson admitted that a patient undergoing a myelographic procedure has no control over performance of the procedure. The patient does not choose or control the equipment or the contrast medium.

Plaintiff's medical history gave no indication that he would suffer an unusual reaction to the use of metrizamide.

The package insert accompanying the Amipaque used for plaintiff's myelogram contained references to possible paralytic adverse reactions. The injuries suffered by plaintiff are commonly referred to as a "cauda equina" syndrome. The package insert referred to "persistent" cauda equina syndrome as a possible adverse reaction to use of the product but made no reference to a permanent reaction. The disabilities suffered by plaintiff are similar to those present in Guillain-Barre syndrome, which the package insert describes as a potential allergy or idiosyncrasy that can result in permanent impairment.

At the hearing on the summary judgment motion, Sterling Drug relied upon the discovery record to support its motion to determine the product to be a Comment k product, i.e., an "unavoidably unsafe" product, and thus within an exception to strict liability in tort. In lieu of the live testimony of Dr. Nelson, the parties stipulated as follows:

"1. A Myelogram as a diagnostic tool is beneficial and necessary for medical personnel to treat patients who have suspected spinal lesions.

"2. Any Myelogram itself is a form of photograph. It can be visualized by X-Ray, CT Scan or Fluoroscope.

"3. A contrast medium makes a Myelogram possible. Without a contrast medium, there can be no Myelogram.

"4. The alleged neurotoxic effects of Metrizamide were not avoidable by any means other than a decision not to use the product."

The parties had previously stipulated in the pretrial order as follows: "The Metrizamide manufactured by Sterling Drug, Inc. and used by Dr. Nelson in the myelogram procedure performed upon the plaintiff was not contaminated in any way and was in the form and content intended by its manufacture at the time of purchase by St. Joseph Medical Center."

Following a hearing on the motions for summary judgment on February 16, 1989, the next day the court sustained all of the defendants' motions for summary judgment. At that time, the court explained its decision from the bench as follows:

"I find there is no active negligence alleged or capable of proof based on the affidavit and deposition and so on that are on file, the statements, technical title that you file with your motions for summary judgment slips me right now. Based on those, I find that there are no issues of fact; and, of course, if it is controverted, it is controverted so it doesn't become a part of — so it is only the uncontroverted, undisputed facts that I am relying on and also with respect to the claim of res ipsa loquitur, I am saying that res ipsa loquitur could not be used in a malpractice case; but there is really no suggestion as to what the negligence was or what could have happened that would give rise to an inference of negligence.

"As I mentioned yesterday, the problem of the cascading cask of the traditional res ipsa loquitur case sort of goes by the wayside once you have discovery that allows you to inspect the premises; but still I am recognizing that in the proper case, res ipsa loquitur can be proper; but there still has to be some way that you are saying what went wrong; and based upon the entire file, I can't say what went wrong.

"I also find that the drug is a Comment k drug; and, of course, any time you are — based on the Funke case, any time you are dealing with the spinal column it is not so recognized that Mr. Savina went into the hospital with what he thought was to be a relatively minor procedure and came out paralyzed. Indeed a tragedy; but, nevertheless, I think the rulings I have to make require me to make the rulings I have made."

All issues raised in the appeal concern whether the trial court erred in granting summary judgment in favor of the appellees. In reviewing a motion for summary judgment that has been granted by the district court, this court must read the record in the light most favorable to the party who defended against the motion for summary judgment. Danes v. St. David's Episcopal Church, 242 Kan. 822, 830, 752 P.2d 653 (1988). In so doing, if "`the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law,'" summary judgment is proper. 242 Kan. at 830 (quoting Hollingsworth v. Fehrs Equip. Co., 240 Kan. 398, 400, 729 P.2d 1214 [1986]).

We first consider whether the trial court erred in finding that metrizamide qualified as a product described in Comment k of § 402A of the Restatement (Second) of Torts (1986). Plaintiff seeks to impose liability upon Sterling Drug under the doctrine of strict liability in tort. This doctrine had its genesis in a concurring opinion by Justice Roger Traynor of the California Supreme Court, where he suggested that a manufacturer should be absolutely liable if it placed a product in the market that it knew would be used without inspection and which had a defect that caused injury. Escola v. Coca Cola Bottling Co., 24 Cal. 2d 453, 461, 150 P.2d 436 (1944). The public policy considerations underlying the doctrine of strict liability are that the manufacturer can anticipate and guard against the recurrence of hazards, that the cost of injury, which may be overwhelming to an injured individual, can be distributed by the manufacturer among the consuming public, and that the marketing of defective products should be discouraged. Brown v. Superior Court, 44 Cal. 3d 1049, 1056, 245 Cal. Rptr. 412, 751 P.2d 470 (1988).

This court adopted the doctrine of strict liability in tort for the sale of a dangerously defective product as set out in the Restatement (Second) of Torts § 402A in Brooks v. Dietz, 218 Kan. 698, 702, 545 P.2d 1104 (1976).

Under the strict liability theory, a plaintiff is not required to establish misconduct by the maker or seller but, instead, is required to impugn the product. The plaintiff must show the product is in "a defective condition unreasonably dangerous," which means that it must be defective in a way that subjects persons or tangible property to an unreasonable risk of harm. Prosser and Keeton, Law of Torts § 99, p. 695 (5th ed. 1984). A product can be defective in one of the following three ways: (1) a flaw is present in the product at the time it is sold; (2) the producer or assembler of the product fails to adequately warn of a risk or hazard related to the way the product was designed; or (3) the product, although perfectly manufactured, contains a defect that makes it unsafe. Prosser, § 99, pp. 695-98.

In the drafting of the Restatement, a group of "unavoidably unsafe products" were to be excluded from the coverage of strict liability in § 402A. These products were described in Comment k, which provides:

"k. Unavoidably unsafe products. There are some products which, in the present state of human knowledge, are quite incapable of being made safe for their intended and ordinary use. These are especially common in the field of drugs. An outstanding example is the vaccine for the Pasteur treatment of rabies, which not uncommonly leads to very serious and damaging consequences when it is injected. Since the disease itself invariably leads to a dreadful death, both the marketing and the use of the vaccine are fully justified, notwithstanding the unavoidable high degree of risk which they involve. Such a product, properly prepared, and accompanied by proper directions and warning, is not defective, nor is it unreasonably dangerous. The same is true of many other drugs, vaccines, and the like, many of which for this very reason cannot legally be sold except to physicians, or under the prescription of a physician. It is also true in particular of many new or experimental drugs as to which, because of lack of time and opportunity for sufficient medical experience, there can be no assurance of safety, or perhaps even of purity of ingredients, but such experience as there is justifies the marketing and use of the drug notwithstanding a medically recognizable risk. The seller of such products, again with the qualification that they are properly prepared and marketed, and proper warning is given, where the situation calls for it, is not to be held to strict liability for unfortunate consequences attending their use, merely because he has undertaken to supply the public with an apparently useful and desirable product, attended with a known but apparently reasonable risk."

Comment k establishes an exception to the test for strict product liability as set forth in § 402A. While Comment g defines "defective condition" as "a condition not contemplated by the ultimate consumer, which will be unreasonably dangerous to him," Comment k defines a category of "unavoidably unsafe" products, which are not considered defective or unreasonably dangerous "when properly prepared, and accompanied by proper directions and warnings." In order to apply Comment k, the product must be properly prepared and must be accompanied by proper directions and warnings. The Comment cannot apply to products that contain a manufacturing flaw or an inadequate warning but, instead, applies only where the plaintiff alleges a design defect. Comment k is meant to shield a manufacturer from liability when the product cannot be designed more safely, not when the product was mismanufactured or was not accompanied by adequate warnings. Toner v. Lederle Laboratories, 112 Idaho 328, 732 P.2d 297 (1987).

The plaintiff contends that it is unclear if the trial court's conclusion that metrizamide is a Comment k drug was based upon the rationale that all prescription drugs are entitled to Comment k status, as defendants urged, or on a proper risk/benefit analysis. Plaintiff argues that, in either case, the trial court was in error. As to the former, plaintiff argues that Kansas does not recognize a blanket exception for prescription drugs and, as to the latter, there is no evidence to make the necessary analysis that metrizamide is a Comment k drug.

In Johnson v. American Cyanamid Co., 239 Kan. 279, 718 P.2d 1318 (1986), this court considered whether Orimune, which is a Sabin or live polio vaccine, was an "unavoidably unsafe product" that came within the protection of Comment k. This court concluded that the vaccine was an "apparently useful and desirable product, attended with a known but apparently reasonable risk." 239 Kan. at 286. In reaching its decision, the court considered the drug itself in determining whether it was an unavoidably unsafe product rather than giving this protection to all prescription drugs. Our decision indicates that the application of Comment k is made on a case-by-case determination.

This is the procedure followed by most courts that have discussed the issue. In Toner, the Idaho Supreme Court noted that, in order to find Comment k applicable, the seller had to establish that the product's risk was unavoidable under the present state of human knowledge and therefore incapable of being made safe. Toner, 112 Idaho at 336-37. Furthermore, a more feasible design, which accomplishes the product's purpose with a lesser risk, must not be available. 112 Idaho at 337. If such a product were available, then the risk would not be "unavoidable" or "apparently reasonable." Nor would the marketing and use of the product be fully justified if an alternative of lesser risk existed. In Toner, the court noted that the evaluation of a purported alternative design in the subject product should consider "the magnitude of the subject product's risk that the alternative avoids, the financial costs of the compared designs, the benefits of the compared designs, and the relative safety of the compared designs, including any new risk that the alternative would pose." 112 Idaho at 337.

In Brochu v. Ortho Pharmaceutical Corp., 642 F.2d 652 (1st Cir.1981), the court concluded that plaintiff stated a cause of action in strict liability when she established that the risk of the injury she suffered from a birth control pill she had taken was eliminated without excessive cost or loss of product efficiency by another birth control pill defendant marketed at the same time, which contained a much lower content of estrogen.

Many other courts have rejected an argument that a prescription drug is per se unavoidably unsafe under Comment k, instead requiring that the court determine whether the product qualifies as unavoidably unsafe under Comment k on a case-by-case basis. Hill v. Searle Laboratories, 884 F.2d 1064, 1069 (8th Cir.1989) (intrauterine device); Feldman v. Lederle Laboratories, 97 N.J. 429, 479 A.2d 374 (1984) (tetracycline drug); White v. Wyeth Laboratories, Inc., 40 Ohio St. 3d 390, 533 N.E.2d 748 (1988) (whooping cough vaccine); Castrignano v. E.R. Squibb & Sons, Inc., 546 A.2d 775 (R.I. 1988) (diethylstilbestrol [DES]).

In attempting to decide whether a given drug is unavoidably unsafe, Judge Kelly, in Graham v. Wyeth Laboratories, 666 F. Supp. 1483, 1496 (D. Kan. 1987), used the approach set forth in Kearl v. Lederle Laboratories, 172 Cal. App. 3d 812, 218 Cal. Rptr. 453 (1985), overruled in Brown v. Superior Court, 44 Cal. 3d 1049, 245 Cal. Rptr. 412, 751 P.2d 470 (1988). Although the California Supreme Court disagreed with the decision in Kearl and, instead, found that all prescription drugs should be entitled to Comment k protection, the approach taken by the court in Kearl in its efforts to determine whether a given drug should be considered unavoidably unsafe was relied upon by Judge Kelly in Graham and is helpful in applying Comment k on a case-by-case basis. The approach provides as follows:

"In our view, the decision as to whether a drug, vaccine, or any other product triggers unavoidably dangerous product exemption from strict liability design defect analysis poses a mixed question of law and fact and can be made only after evidence is first taken, out of the jury's presence, on the relevant factors to be considered. [Citations omitted.] A trial court should take evidence as to: (1) whether, when distributed, the product was intended to confer an exceptionally important benefit that made its availability highly desirable; (2) whether the then-existing risk posed by the product was both `substantial' and `unavoidable'; and (3) whether the interest in availability (again measured as of the time of distribution) outweighs the interest in promoting enhanced accountability through strict liability design defect review. In determining the first aspect of the second factor (i.e., whether the risk posed was `substantial'), a court should consider whether, at the time of distribution, the risk posed permanent or long-term disability (e.g., loss of body function, organs, or death) as opposed to mere temporary or insignificant inconvenience (e.g., skin rash, minor allergic reaction, etc.). In determining the second aspect of the second factor (i.e., whether the risk posed was `unavoidable'), a court should consider (i) whether the product was designated to minimize — to the extent scientifically knowable at the time it was distributed — the risk inherent in the product, and (ii) the availability — again, at the time of distribution — of any alternative product that would have as effectively accomplished the full intended purpose of the subject product." 172 Cal. App. 3d at 829-30.

The Idaho Supreme Court, in Toner, relied upon this approach in its attempt to determine whether the diphtheria, pertussis, and tetanus (DPT) vaccine being considered should be given Comment k status. 112 Idaho at 336-40. The court said:

"We do not believe comment k was intended to provide nor should it provide all ethical drugs with blanket immunity from strict liability design defect claims. The comment refers to `some' products which are unavoidably unsafe; the comment states such products are `especially common in the field of drugs;' the comment cites certain examples from that field deserving of its protection and notes that `[t]he same is true of many other drugs, vaccines, and the like ... [and] of many new or experimental drugs....' (Emphasis added.) Obviously, the comment does not apply to all drugs. Rather, the comment applies `when the situation calls for it,' which is when the product is unavoidably unsafe but is `an apparently useful and desirable product, attended with a known but apparently reasonable risk,' or with an unknown risk which was not yet reasonably discoverable at the time of marketing. It is equally obvious that not all drugs are so perfectly designed that they cannot be made more pure or more safe, or that there are not safer, suitable alternatives; nor do the benefits of all drugs necessarily outweigh their risks. Brochu [v. Ortho Pharmaceutical Corp.], 642 F.2d at 655; Singer [v. Sterling Drugs], 461 F.2d at 290-91; Flood v. Wyeth Laboratories, Inc., 183 Cal. App. 3d 1272, 228 Cal. Rptr. 700, 702-03 ([Div. 5] 1986); Kearl [v. Lederle Laboratories, Inc.], 218 Cal. Rptr. 463-64; Feldman [v. Lederle Laboratories], 479 A.2d at 380, 383; Schwartz, [Unavoidably Unsafe Products: Clarifying the Meaning and Policy Behind Comment K, 42] Wash. & Lee L. Rev. at 1141." Toner, 112 Idaho at 339.

We feel the analysis employed in Kearl as a framework for deciding whether metrizamide is an unavoidably unsafe product is sound.

Plaintiff contends that, by applying the above analysis to the present case, the marginal benefits of metrizamide do not outweigh its serious health risks. In Johnson, this court stated that the trial judge should have heard evidence on the issue of whether the polio vaccine in question was an unavoidably unsafe product. This court indicated the evidence should have been received outside the presence of the jury and a determination made before the case was submitted to the jury. 239 Kan. at 288.

In weighing the benefits and risks of metrizamide, this court must base its considerations upon information that was known at the time the product was distributed and was used on plaintiff, even though this information may have developed after metrizamide was originally introduced into the market. See Toner, 112 Idaho at 337-38.

Here, in finding metrizamide to be a Comment k drug, the trial court made no findings of fact to support its decision. The parties stipulated that the metrizamide "was not contaminated in any way and was in the form and content intended by its manufacture at the time of purchase by St. Joseph Medical Center." This stipulation established that the metrizamide used on plaintiff did not contain manufacturing flaws and was in the condition intended by Sterling Drug. Plaintiff presented no evidence that the drug contained a design defect. At the time this myelogram was performed, the only two contrast agents available in the Wichita area were Pantopaque and Amipaque. Because the oil-based Pantopaque does not mix with spinal fluid and must be removed, the mechanical impossibility of removing all of the Pantopaque leaves a continuing risk that the patient will develop arachnoiditis.

Dr. Nelson, who performed the myelogram here, had stopped using Amipaque during the pendency of the lawsuit, but expressed his desire to use it for the benefit of his patients. Plaintiff does not contradict evidence indicating that metrizamide was preferable to Pantopaque in most cases because it was less dense and filled the nerve root sleeves better, which resulted in a nicer image aesthetically and diagnostically, and because it was metabolized by surrounding tissues and absorbed into the bloodstream. Although Pantopaque was preferred by some radiologists for limited situations, the testimony of the experts established that Amipaque, containing metrizamide, was the preferred contrast agent at the time of plaintiffs myelogram. Later, preference for metrizamide was replaced by other water-soluble contrast agents.

Based upon the testimony by the experts, it is difficult to determine whether the risk posed by the use of metrizamide was substantial. Sterling Drug had received repeated reports of people having problems with transient paralysis. Apparently plaintiffs was the first case resulting in permanent paralysis. The evidence presented by the plaintiff concerning the risk addressed the issue of the adequacy of the warnings, not the risk posed by the product itself. Plaintiff appears to recognize that at least a slight risk of paralysis was unavoidable based upon the knowledge of the industry at the time. The parties stipulated that the only way in which paralysis could have been avoided here was to not use metrizamide. Sterling Drug recognizes and freely admits that it had received reports of paresis but insisted that it had no information indicating such disability would be permanent. Plaintiff argues that Sterling Drug's knowledge of one case of three-week paralysis should have suggested that a permanent disability was possible.

Sterling Drug did provide warnings to lessen the flow of the metrizamide toward the brain to minimize the possibility of seizures. Plaintiff has presented no evidence to show that metrizamide had a design defect that, if corrected, could have posed less of a risk. Although plaintiff argues that the contrast agent Pantopaque was available at the time, the testimony of all the radiologists indicates that, although Pantopaque may be utilized for some limited situations, the preferred contrast agent at the time of plaintiffs myelogram was Amipaque. Thus, Pantopaque was not an alternative product that would have as effectively accomplished the full and intended purpose of metrizamide. We reject the amicus argument that diagnostic drugs are not entitled to Comment k protection. The policy considerations underlying strict liability and Comment k would apply to a diagnostic drug as well as to a drug used for treatment.

The trial court here did not hear testimony but received over 30 volumes of depositions. It heard oral argument on the motions for summary judgment one afternoon and ruled the next day. The trial court did not discuss the record in finding that metrizamide was a Comment k drug. The depositions of plaintiffs expert witnesses, as well as those for the defendants, indicate that the metrizamide used here did not contain a design defect or a manufacturing flaw and that the benefits it offered out-weighed the risks its use posed in light of knowledge at the time of plaintiffs myelogram. We therefore cannot conclude that when we apply the risk/benefit analysis to the record in the present case, the trial court erred in finding metrizamide "qualifies as a product described in Comment k." Having made the determination that metrizamide was a drug described in Comment k, the court further concluded that, since the plaintiff stipulated that there was no manufacturing flaw or defect or no design defect asserted in the pretrial order, the only theory of liability remaining was the adequacy of the warnings.

In Johnson, we made a similar analysis as to the drug Orimune:

"Orimune, the Sabin-type vaccine, is an `unavoidably unsafe product' that is an `apparently useful and desirable product, attended with a known but apparently reasonable risk' as a matter of law. Public policy requires that the mere manufacture of the vaccine not be actionable on the ground of design defect. The trial judge should have heard the evidence on this issue outside the presence of the jury and made the determination thereon. There is no claim that the vaccine administered to plaintiffs child was improperly manufactured or that a defective product was delivered. The vaccine was properly prepared and marketed and was exactly what it was intended to be. As a matter of law there is no manufacturing or design defect in the product at issue herein.

....

"In determining warning issues, the test is reasonableness. To impose liability on a manufacturer, the plaintiff must show negligence on the part of the manufacturer. In Kearl v. Lederle Laboratories, 172 Cal. App. 3d 812, 218 Cal. Rptr. 453 (1985) (a case also involving Orimune but where the vaccine was administered through a clinic rather than a learned intermediary), the applicable rules were stated as follows:

"`As an initial matter we question the commonly assumed and often asserted proposition that in products liability cases failure to warn or inadequacy of a warning may be a basis for imposition of strict liability. A review of the cases discloses that the analysis called for in this situation is not based on strict liability, but negligence.

"`As noted above (ante, p. 822) the characteristic that distinguishes strict liability from negligence is proof of actual or constructive knowledge of risk: In a negligence action we focus on the defendant's conduct and require plaintiff to show defendant acted unreasonably in light of a known or constructively known risk. In strict liability actions, on the other hand, we focus not on the reasonableness of a defendant's conduct but on the product, and we either ignore the question of a manufacturer's actual or constructive knowledge of risk (as in a "consumer expectations" design defect case) or we in effect impute to the manufacturer defendant current scientific knowledge of the risk caused by his product (as in a risk/benefit design defect balancing case). (See, e.g., Comment, The Failure to Warn Defect: Strict Liability of the Prescription Drug Manufacturer in California (1983) 17 U.S.F.L. Rev. 743, 755.) But in all warning cases — even if the plaintiff or the court claims to analyze failure to warn or inadequacy of warning in the context of a strict products liability claim — the tests actually applied condition imposition of liability on the defendant's having actually or constructively known of the risk that triggers the warning. (E.g., Carmichael v. Reitz, supra, 17 Cal. App. 3d 958, 988; Christofferson v. Kaiser Foundation Hospitals, supra, 15 Cal. App. 3d 75, 79-80; Oakes v. E.I. Du Pont de Nemours & Co., Inc. (1969) 272 Cal. App. 2d 645, 650-651, 77 Cal. Rptr. 709; Toole v. Richardson-Merrell, Inc., supra, 251 Cal. App. 2d 689, 709-710; Dunn v. Lederle Laboratories, supra, 328 N.W.2d 576, 580; Woodill v. Parke Davis & Co., supra, 402 N.E.2d 194, 197-198, and cases cited; Petty v. United States, supra, 740 F.2d 1428, 1432; Reyes v. Wyeth Laboratories, supra, 498 F.2d 1264, 1274-1275; Basko v. Sterling Drug, Inc. supra, 416 F.2d 417, 426; Sterling Drug, Inc. v. Yarrow (8th Cir.1969) 408 F.2d 978, 992-993; Davis v. Wyeth Laboratories, supra, 399 F.2d 121, 129; Kidwell, The Duty to Warn: A Description of the Model of Decision (1975) 53 Tex. L. Rev. 1375, 1377-1378; Note, supra, 48 Fordham L. Rev. at pp. 745-750; Prosser & Keeton on Torts (5th ed. 1984) § 99, at p. 697; see also Comment, Strict Liability and the Tortious Failure to Warn (1984) 11 N. Ky. L. Rev. 409, 419-422 [criticizing but recognizing the present result].)

"`Just as liability for failure to warn of product risk is based on negligence, the adequacy of a warning is also judged under a reasonableness standard — even if the claim is made under the rubric of a strict products liability "defect." (Sterling Drug, Inc. v. Yarrow, supra, 408 F.2d 978, 992-993; Feldman v. Lederle Laboratories, supra, 479 A.2d 374, 386 ["negligence and strict liability in warning cases may be deemed to be functional equivalents"]; Dunn v. Lederle Laboratories, supra, 328 N.W.2d 576, 580; Franklin & Mais, supra, 65 Cal. L. Rev. at p. 762, and cases cited in fn. 33; Note, supra, 32 DePaul L. Rev. at p. 254, fn. 29; Kidwell, supra, 53 Tex. L. Rev. at pp. 1378-1379; Prosser & Keaton, supra, § 99, at p. 697; but see Noel, Products Defective Because of Inadequate Directions or Warnings (1969) 23 Sw. L.J. 256, 267, 272.)' 172 Cal. App. 3d at 831-33.

"Was the warning adequate as a matter of law? That is, was it a reasonable warning by a manufacturer to a learned intermediary? Was the manufacturer negligent in the warning supplied?" Johnson v. American Cyanamid Co., 239 Kan. 279, 286-87, 718 P.2d 1318 (1986).

Here, as in Johnson, the danger created by using the drug cannot be eliminated, but, due to its benefits, the drug can be marketed provided it is accompanied by "proper directions and warnings." The liability of the manufacturer of the drug is based upon negligence in failing to adequately warn.

We concluded in Johnson that, as a matter of law, the warnings were adequate and, therefore, it was error to have submitted the question of liability to the jury. However, in Johnson, we found that "[t]he warning obviously warns that in rare instances a person in close contact with a vaccinee may develop polio. That is exactly what happened to the plaintiff herein." 239 Kan. at 288. Here, what happened to the plaintiff is not exactly what is warned of in the package insert.

Here, plaintiff presented testimony from three doctors, who criticized the adequacy of information in the package inserts about DER that Sterling Drug received before plaintiffs myelogram. These three doctors testified by deposition that the warnings should have contained a more explicit statement of the severe transient or permanent paralysis that had been experienced by cases included in the DER. Drs. Berger, Schaumburg, and Heinz all testified that the warnings were inadequate. If their testimony creates a question of fact, then this court cannot conclude as a matter of law that reasonable minds could not differ in finding the warning adequate. See Johnson, 239 Kan. at 288-89.

A motion for summary judgment is to be sustained only where the record conclusively shows that no genuine issues of material fact exist and that the party moving for summary judgment is entitled to judgment as a matter of law. The trial court must resolve all facts and inferences which may reasonably be drawn from the evidence in favor of the party against whom the ruling is sought. Although the party opposing summary judgment is not required to prove its case, it has an affirmative duty to come forward with facts to support its claim. If factual issues exist, they must be material to preclude summary judgment. Bacon v. Mercy Hosp. of Ft. Scott, 243 Kan. 303, 306-07, 756 P.2d 416 (1988). On appeal, the court must apply the same rule and deny summary judgment where reasonable minds could differ about the conclusions to be drawn from the evidence. 243 Kan. at 306.

The decision of which contrast agent to use in this case was made by Dr. Nelson. Plaintiff had no choice in the product. This case involves an individualized medical decision and the learned intermediary rule applies. Therefore, if the warning provided to the doctor was adequate, Sterling Drug will have fulfilled its duty to warn.

Here, material questions of fact exist about the adequacy of the warnings to inform the physician of the potential risks that arise from the use of metrizamide. Sterling Drug admits that it was aware of several cases resulting in transient paralysis after the use of metrizamide in a myelogram. In the videotaped trial testimony of Dr. Heinz, the doctor discusses four DER that he believes indicated occurrences parallel to those symptoms experienced by plaintiff. Sterling Drug objects to the doctor's reliance upon one of the DER because the doctor did not list it specifically at the time of his discovery deposition. The injury described in the DER indicated damage to multiple nerve roots including injury to the right leg and lack of anal and bladder sphincter control. The symptoms in the patient improved, and Dr. Heinz agreed that the case would accurately fit the term "persistent paralytic result."

On cross-examination, Dr. Heinz admitted that the cases were different in many ways, but pointed to parallels that he believes should have alerted Sterling Drug to the possibility of permanent paralysis.

Sterling Drug established through its cross-examination that, at the time of plaintiffs myelogram, the cases reported in DER had not included one involving permanent paralysis. It argues that it had no obligation and, in fact, was prohibited by Food and Drug Administration (FDA) regulations from reporting unsubstantiated comments and concerns from doctors. It further argues that, even if the warnings had contained statements indicating cases of more extensive paralysis, the doctors would have used Amipaque anyway and not included this in the discussion with the patient about possible adverse reactions.

The language Dr. Heinz had particular problems with was the following sentence: "Rarely, persistent though transitory weakness in the leg or ocular muscles has been reported." Dr. Heinz admitted that he was not an expert on FDA regulations and in fact had never read them. His opinion that the warnings were inadequate was based upon his 30 years as a practicing radiologist, a teacher, and a scholar. If the warnings had included the omitted information, Dr. Heinz believed some doctors may have chosen to continue using Pantopaque. Although agreeing that the patient would probably not have been advised of the possibility of permanent paralysis, Dr. Heinz believed the drug manufacturer had a duty to fully advise the physician of the worst case scenario to enable him to react appropriately when symptoms of paralysis occurred.

Comments by the FDA stated that the purpose of prescription drug labeling is "to provide physicians with a clear and concise statement of the data and information necessary for the safe and effective use of the drug." Rules & Regulations of FDA 44 Fed. Reg. 37,435 comment 6 (1979). Prescription drug labeling is directed to health care professionals, not the ultimate consumer. 44 Fed. Reg. 37,438-39 comment 21. The agency intends the labeling to ensure that the medical community is provided a complete and accurate explanation of the drug, not to influence the civil tort liability of the physician. 44 Fed. Reg. 37,437 comment 13. Use of prescription drug labeling as evidence in malpractice litigation "is an unintended by-product of FDA's regulatory activities." 44 Fed. Reg. 37,435 comment 6.

The rules and regulations of the FDA define an adverse reaction as "an undesirable effect, reasonably associated with the use of the drug, that may occur as part of the pharmacological action of the drug or may be unpredictable in its occurrence." 21 CFR § 201.57(g) (1989). Adverse reactions are to be categorized and listed in decreasing order of frequency, except that significantly more severe reactions are to be listed before others. 21 CFR § 201.57(g)(2).

The section of labeling titled "[c]ontraindications" should describe "those situations in which the drug should not be used because the risk of use clearly outweighs any possible benefit." Only known hazards and not theoretical possibilities are to be listed. 21 CFR § 201.57(d).

The "[w]arnings" section of the labeling is intended to describe "serious adverse reactions and potential safety hazards, limitations in use imposed by them, and steps that should be taken if they occur." Labeling must be revised to include a warning as soon as reasonable evidence exists "of an association of a serious hazard with a drug; a causal relationship need not have been proved." 21 CFR § 201.57(e).

Sterling Drug classifies the information Dr. Heinz suggests should be included in the package insert as merely the personal opinion of individual physicians. The FDA rejected including "the impressions or beliefs of physicians, although they are honest and may prove to be valid" because they are not an appropriate ground upon which to base statements in prescription drug labeling. 44 Fed. Reg. 37,443 comment 46. But this comment was discussing statements about the effectiveness of a drug, which must be founded upon adequate and well-controlled clinical investigations. 21 CFR § 201.57(b)(2).

Sterling Drug argues that it cannot include information about unexplained, isolated adverse effects. The comments note that an adverse reaction "would not include unsubstantiated reactions, disease symptoms, or apparently undesirable effects coincidental to the use of a drug, because they would not be reasonably associated with the drug." Yet the commissioner continued by advising "that in determining whether an undesirable effect is reasonably associated with the use of a drug, analysis of available data[,] detailed study of individual cases, and new investigations with the drug may be necessary." 44 Fed. Reg. 37,453 comment 103. Sterling Drug argues that this comment prohibits listing all possible adverse reports. Yet the reactions plaintiffs experts believed should be listed would arguably be more than unsubstantiated reactions coincidental to use of metrizamide.

Sterling Drug argues that emphasizing one risk over another is prohibited. Adverse reaction information is to be categorized by organ system, severity, frequency, or toxicological mechanism. Any of these categories may be used. The adverse reactions within each category are to be listed in descending order of frequency, except that an adverse reaction "that is significantly more severe that the other reactions listed in a category ... shall be listed before those reactions regardless of its frequency." 21 CFR § 201.57(g)(2); 44 Fed. Reg. 37,455 comment 116.

Sterling Drug further argues that it cannot be liable because the Kansas Product Liability Act, K.S.A. 60-3301 et seq., presumes Amipaque to be safe. The product liability act applies to all product liability claims regardless of the substantive theory of recovery. Therefore, under K.S.A. 60-3302(c), the provisions of the Act apply to actions based on strict liability in tort as well as negligence, breach of express or implied warranty, and breach of or failure to discharge a duty to warn or instruct.

Sterling Drug argues that K.S.A. 60-3304(a) creates an irrebuttable presumption of safety when the design or performance of the product is in compliance with applicable state or federal regulations. K.S.A. 60-3304(a) provides:

"When the injury-causing aspect of the product was, at the time of manufacture, in compliance with legislative regulatory standards or administrative regulatory safety standards relating to design or performance, the product shall be deemed not defective by reason of design or performance, or, if the standard addressed warnings or instructions, the product shall be deemed not defective by reason of warnings or instructions, unless the claimant proves by a preponderance of the evidence that a reasonably prudent product seller could and would have taken additional precautions."

Sterling Drug argues that the first clause of this statute creates an irrebuttable presumption of safety when the design or performance of the product complies with applicable state or federal regulations. It argues that a rebuttable presumption arises only for the provisions of the statute concerning warnings or instructions. To rebut this presumption that the warnings or instructions are adequate, the statute requires plaintiff to show that "a reasonably prudent product seller could and would have taken additional precautions."

Pursuant to K.S.A. 60-3304(b), noncompliance with regulatory or administrative standards presents prima facie proof of defectiveness of a product subject to rebuttal by showing that non-compliance "was a reasonably prudent course of conduct under the circumstances." Sterling Drug argues that, when these two subsections are read together, they prohibit any attempt by a plaintiff to rebut the presumption created in subsection (a) by showing that a seller should have engaged in conduct that would have been contrary to administrative rules and regulations, as prohibited by subsection (b).

The problem with the analysis by Sterling Drug is that the regulations imposed by the FDA are minimal standards. A drug company is not prohibited from providing additional warnings and additional information that is not required by the FDA. Sterling Drug's analysis is further flawed by concluding that the last clause of subsection (a) of K.S.A. 60-3304, concerning the ability of a claimant to prove what a prudent product seller could and would have done, applies only to the provision relating to adequate warnings and instructions. In resolving this case, we do not need to decide whether this last clause should apply to the presumption concerning whether a product is deemed defective. The statute clearly provides that compliance with FDA rules and regulations creates only a rebuttable presumption that the warnings are adequate, which can be rebutted by a preponderance of the evidence that a reasonably prudent product seller could and would have taken additional precautions.

The testimony presented by plaintiffs experts, including Drs. Heinz, Schaumburg, and Berger, presents a question of fact concerning the adequacy of the package inserts with metrizamide. Although Dr. Schaumburg and Dr. Heinz noted that the information they would include was not necessarily required by the FDA, this is not defeated by a rebuttable presumption. Instead, plaintiff now has the burden of proving by a preponderance of the evidence that a reasonably prudent product seller would have included additional information in the package inserts.

The plaintiff next argues that the trial court erred in ruling the doctrine of res ipsa loquitur was unavailable against all the defendants. In rendering his decision in this case, the trial court stated, "[W]ith respect to the claim of res ipsa loquitur, I am saying that res ipsa loquitur could not be used in a malpractice case." In the next paragraph, the court stated, "I am recognizing that in the proper case, res ipsa loquitur can be proper; but there still has to be some way that you are saying what went wrong; and based upon the entire file, I can't say what went wrong."

Plaintiff argues that the trial court erred in stating that res ipsa loquitur was not available for use by a plaintiff in a medical malpractice case. In support of his argument, plaintiff notes that, in Voss v. Bridwell, 188 Kan. 643, 364 P.2d 955 (1961), this court found the doctrine of res ipsa loquitur to be applicable in a case alleging medical malpractice through negligence when a patient was rendered incompetent due to improper administration of general anesthesia during a mastoid operation, which was never performed. Plaintiff directs this court's attention to numerous other jurisdictions that have applied the doctrine of res ipsa loquitur in a variety of cases involving medical negligence claims.

The amicus curiae brief filed on behalf of the Kansas Trial Lawyers Association (KTLA) points out that, although a number of Kansas cases have declined to apply res ipsa loquitur to specific claims of medical malpractice, Kansas recognizes the availability of the doctrine for the right facts. At least two Kansas cases have applied the doctrine. In Voss v. Bridwell, the doctrine was applied in the action against the physician who inserted an endotracheal tube into the patient's esophagus instead of his larynx, which deprived the patient of oxygen during anesthesia and left him decerebrate and totally incapacitated from the administration of anesthesia. We recognized that established rules and rationale militate against the application of res ipsa loquitur in medical malpractice cases, stating the doctrine as follows:

"In medical malpractice cases where the action is not founded upon the mere failure of a physician or surgeon to secure the desired result, or upon the merits of a diagnosis or of scientific treatment, the doctrine of res ipsa loquitur has application to those situations in which the injury results from an unusual occurrence, not ordinarily found where the service performed followed the usual procedure of those skilled in that particular practice, provided a layman is able to say as a matter of common knowledge and observation, that the consequences of professional treatment were not such as would ordinarily have followed if due care had been exercised." 188 Kan. 643, Syl. ¶ 6.

The court also noted two general rules that particularly militate against application of res ipsa loquitur in medical malpractice cases. First, a physician or surgeon "is presumed to have carefully and skillfully treated or operated on his patient, and there is no presumption of negligence to be indulged from the fact of injury or adverse result by reason of such treatment or operation." Voss, 188 Kan. at 659. Second, to establish liability in a medical malpractice case, expert medical testimony must be presented as proof that a lack of due care was exercised or that approved procedure and methods were not followed, except under certain facts that are within the common knowledge of mankind and which may be testified about by anyone familiar with the facts. 188 Kan. at 659 (citing Riggs v. Gouldner, 150 Kan. 727, 728-29, 96 P.2d 694 [1939]). Application of res ipsa loquitur in medical malpractice cases must be considered on a case-by-case basis. 188 Kan. at 660.

In Tatro v. Lueken, 212 Kan. 606, 611, 512 P.2d 529 (1973), this court recognized that, in Kansas, res ipsa loquitur had been applied in only two medical malpractice cases, including Voss. In Tatro, the court stated that the doctrine of res ipsa loquitur applies in a medical malpractice case only where a layman is able to say as a matter of common knowledge and observation, or from the evidence can draw an inference, that consequences of professional treatment were not such as ordinarily would have been followed if due care had been exercised. 212 Kan. at 611.

The doctrine of res ipsa loquitur is available in an appropriate case to a plaintiff alleging medical malpractice based upon negligence. The applicability must be decided on a case-by-case basis. Therefore, we must determine whether the trial court erred in concluding that res ipsa loquitur was unavailable to the plaintiff under the facts of this case.

Three conditions must be met for the doctrine of res ipsa loquitur to apply. First, the thing or instrumentality causing the injury or damage must be within the exclusive control of the defendant. Second, the injury must be of the kind that ordinarily would not occur in the absence of someone's negligence. Third, the injury must not be due to the contributory negligence of plaintiff. Arterburn v. St. Joseph Hospital & Rehabilitation Center, 220 Kan. 57, 64, 551 P.2d 886 (1976). To meet the first condition, plaintiff must be able to show two things: (1) The specific thing or instrumentality which actually caused his injury or damage, and (2) that the thing or instrumentality which caused his injury or damage was within the exclusive control of the defendant. The doctrine does not apply where the thing or instrumentality which caused the injury or damage is unknown or cannot be shown. Arterburn, 220 Kan. at 65.

In Tatro, 212 Kan. 606, plaintiff underwent an abdominal hysterectomy and, a few days after the surgery, experienced a leakage of urine from the bladder into the vagina. Expert testimony concluded that this was caused by a vesicovaginal fistula, which could have been caused by (1) stitching the bladder and vagina together during surgery, (2) clamping the bladder and vagina together during surgery, or (3) an abscess which is a normal risk of the surgery. The first two causes would result from a deviation from the standard of care by the surgeon during the operation, but he denied both of them. It was impossible to determine which of the three actually caused the fistula. Under these circumstances, the court held that the submission of an instruction on res ipsa loquitur would be inappropriate. 212 Kan. at 615.

The plaintiff in Arterburn, 220 Kan. 57, alleged that his spleen was ruptured when a hospital bed malfunctioned. A jury trial resulted in a verdict for the hospital and the plaintiff appealed, challenging the correctness of an instruction on circumstantial evidence and the refusal to instruct on res ipsa loquitur. The expert medical testimony gave four possible causes for splenic rupture: (1) spontaneous rupture with no apparent explanation; (2) disease; (3) external trauma, which would include the malfunction of the hospital bed; and (4) surgical trauma. 220 Kan. at 59. This court concluded that the trial court did not err in ruling that the doctrine of res ipsa loquitur was not applicable because plaintiff failed to show that the malfunction of the hospital bed was the thing or instrumentality that caused his spleen to rupture. 220 Kan. at 67.

In making its ruling herein, the trial court referred to Funke v. Fieldman, 212 Kan. 524, 512 P.2d 539 (1973), which was a malpractice action against an anesthesiologist for injuries resulting from the administration of a spinal anesthetic. Plaintiff had previously been administered a caudal anesthesia for the birth of her second child. When she was advised that she needed a hysterectomy, she and her anesthesiologist agreed that a spinal anesthetic would be appropriate. When the doctor initiated the spinal tap at the L-2/L-3 interspace, plaintiff experienced terrible pain. After inserting a small amount of the anesthetic, the doctor withdrew the needle and reinserted it at the L-3/L-4 interspace. When the anesthesia disappeared, paralysis remained on the left side with total anesthesia to pain as well as temperature, and with reduced sensitivity to touch. 212 Kan. at 527-28.

In considering the appropriateness of res ipsa loquitur, the court noted that evidence in the record established a possibility of nerve damage such as appellant sustained could occur from a spinal injection. The court noted the longstanding general rule that expert medical testimony is ordinarily required to establish negligence or lack of skill on the part of a physician or surgeon in his medical diagnosis, his performance of surgical procedures, and his care and treatment of the patients. An exception to this rule exists when the results of the medical treatment are so patently bad that they are obvious to lay persons or within the common knowledge and experience of mankind generally. Funke, 212 Kan. at 530 (citing Collins v. Meeker, 198 Kan. 390, 394, 424 P.2d 488 [1967]). This court held that the record contained substantial competent evidence in the form of expert medical testimony to support the trial court's finding that appellee was not negligent in the performance of the spinal anesthetic injection. 212 Kan. at 530. In summarizing its decision in Funke, this court stated:

"In our opinion the administration of spinal anesthesia which results in permanent nerve damage to the patient is a procedure so complicated, considering the delicate anatomy of the human spine and the various possibilities of injury from the needle or anesthetic solution, as to lie beyond the realm of common knowledge and experience of laymen as to whether such result would not ordinarily occur in the absence of negligence." 212 Kan. at 538.

The trial court here cited to Funke when granting the motion for summary judgment. Apparently the trial court concluded that plaintiff did not present substantial competent evidence in the form of expert medical testimony to support his claim that his injury was caused by the negligence of Dr. Nelson, by the negligence of the radiologist, MacDonald, or other staff at St. Joseph Medical Center, or by the failure of Sterling Drug to more completely warn of the possible adverse effects of using metrizamide.

Concerning the failure to adequately warn, the court in Funke noted that the doctor administering the spinal anesthetic had a duty to obtain the informed consent of Mrs. Funke before its administration. Yet the physician's failure to fulfill his obligation did not establish liability to the patient. The court stated:

"An unrevealed risk that should have been made known must materialize, otherwise the omission, however unpardonable, is legally without consequence. Occurrence of the risk must be harmful to the patient, for negligence unrelated to injury is nonactionable. And, as in malpractice actions generally, there must be a causal relationship between the physician's failure to adequately divulge the risks and damage to the patient." 212 Kan. at 535.

The court concluded that a causal connection would exist only when disclosure of significant risks incidental to treatment would have resulted in a decision against it. In Funke, the patient did not establish that she would not have undergone the procedure if she had been adequately advised of the possible risks. 212 Kan. at 535-36.

Here, plaintiff suggested that his injury could have resulted from the improper placement of the needle by Dr. Nelson. Yet plaintiff presented no medical expert testimony that Dr. Nelson's placement of the needle was incorrect. No evidence was offered to establish that placement of the needle was erroneous. In fact, plaintiffs expert, Dr. Heinz, testified that he saw no conduct by Dr. Nelson that was outside the normal standard of competency expected by a doctor performing a myelogram.

Plaintiff also suggested that Dr. Nelson was negligent because he used an 18-gauge needle to inject the metrizamide instead of the 22-gauge needle recommended in the package insert. Again, plaintiff offered no evidence indicating that the use of a different size needle caused the injury suffered by plaintiff. Finally, plaintiff suggested that Dr. Nelson should have more completely advised plaintiff of the risks and possible consequences of a myelogram with metrizamide. Dr. Nelson indicated that he told plaintiff the information he tells all his patients regarding the performance of a myelogram with metrizamide. Dr. Nelson did not know if he would have provided a more complete discussion of the risks and possible consequences if the package insert had contained more complete information about the drug reaction reports that Sterling Drug had received from the use of metrizamide.

Plaintiff suggested that St. Joseph Medical Center should be liable for the negligence of its radiology technician, MacDonald, because he may have been negligent in the preparation of the metrizamide solution. Yet plaintiff presented no medical testimony to establish that plaintiff's injury was caused by negligent preparation of the metrizamide solution. Plaintiff also suggested that coring could have occurred during the preparation of the metrizamide solution when MacDonald punctured the rubber stopper with the needle and withdrew the medication. Once again, however, plaintiff presents no medical testimony to establish that coring occurred, but merely suggests that it was a possibility.

Finally, medical testimony indicates that the injuries suffered by plaintiff could have resulted from an allergic reaction of plaintiff to metrizamide or from a thoracic-level herniated disc that was revealed when an MRI procedure was performed on plaintiff three and one-half years after his injury.

Plaintiff does not have to show that the requisite control was within the exclusive control of one individual or a singular defendant. The doctrine of res ipsa loquitur can be applicable to multiple defendants. Bias v. Montgomery Elevator Co., 216 Kan. 341, 343, 532 P.2d 1053 (1975). Plaintiff does not have to eliminate all other possible causes of the accident. 216 Kan. at 344 (citing Prosser, Law of Torts § 39, p. 211 [4th ed. 1971]). Yet, plaintiff must produce sufficient evidence from which a reasonable person could say that, on the whole, it is more likely than not that the defendant was negligent. 216 Kan. at 344. Here, the evidence establishes that it is at least equally probable that the negligence of another or a condition of plaintiff was the cause of the injury. Because plaintiff failed to present evidence establishing that the acts of the defendants caused plaintiff's injury, the trial court properly granted summary judgment on the issue of res ipsa loquitur. See Bacon, 243 Kan. at 314.

The final issue raised by plaintiff is whether the trial court erred in granting summary judgment in favor of Dr. Nelson and the medical center. Both Dr. Nelson and the medical center argue that plaintiff needed to present expert testimony to avoid summary judgment on grounds that his claim of negligence against them could not be shown as a matter of law. The expert testimony presented by plaintiff did not indicate negligence by either Dr. Nelson or the medical center. Instead, plaintiff's expert, Dr. Heinz, testified that he saw no negligence or misconduct by these parties. The testimony presented by all of plaintiff's experts centered upon the inadequacy of the package inserts included with metrizamide.

At the hearing on the motions for summary judgment, plaintiff admitted that his claims against the doctor and hospital were based upon the doctrine of res ipsa loquitur. Summary judgment is seldom proper in negligence cases, but summary judgment motions are intended to isolate and dispose of factually unsupported claims or defenses. Bacon, 243 Kan. at 307. Negligence in a medical malpractice case is never presumed and cannot be inferred merely from the lack of success of an operation or procedure or from an adverse result from treatment. Tatro, 212 Kan. at 611. A plaintiff in a medical malpractice case bears the burden of showing not only the doctor's negligence, but also that this negligence caused the injury. See Funke, 212 Kan. at 535.

Unless the lack of reasonable care or the existence of proximate cause is apparent to an average layman based upon common knowledge or experience, a plaintiff must present expert testimony in a medical malpractice case to establish that the accepted standard of care was not met and to prove causation. Bacon, 243 Kan. at 307. Expert witnesses must confine their opinions to matters in issue which are certain or probable and should not testify to mere possibilities. Under the common knowledge exception, expert testimony is not needed "when the treatment or care of a patient has such bad results that lack of reasonable care would be obvious from the everyday knowledge of persons generally[;] then persons other than physicians may testify" to the facts needed to establish negligence. Crowley v. O'Neil, 4 Kan. App. 2d 491, Syl. ¶ 4, 609 P.2d 198, rev. denied 228 Kan. 806 (1980).

As the summary of the testimony presented in the discussion of the inapplicability of res ipsa loquitur noted, plaintiff presented no expert testimony indicating that the conduct of Dr. Nelson or the hospital was the certain or probable cause of the injury to plaintiff. Plaintiff presented testimony about possibilities. It was possible the doctor placed the needle improperly, used the wrong size needle, or did not provide more extensive warnings. But plaintiff presented no testimony, expert or otherwise, indicating that any of these events was the certain or probable cause of plaintiffs injury. Regarding the hospital, plaintiff presented testimony that it was possible the metrizamide was mixed improperly or that coring occurred during its preparation. Yet, again, plaintiff presented no testimony, expert or otherwise, indicating that this conduct was the certain or probable cause of plaintiff's injury.

This is not a case in which the treatment or care of the patient had such bad results that lack of reasonable care would be obvious from the everyday knowledge of persons generally. As this court suggested in Funke, the administration of a spinal tap is a complicated, delicate procedure that lies beyond the realm of common knowledge and experience of a layman. 212 Kan. at 538. This court has long recognized that a physician or surgeon is presumed to have carefully and skillfully treated or operated on his patient, and negligence cannot be presumed from the fact of injury or adverse result. Tatro, 212 Kan. at 611. Plaintiff was required to present expert testimony establishing a rational basis for concluding that it was more probable than not that defendants' negligence caused plaintiff's injury. Arterburn, 220 Kan. at 63. Because plaintiff failed to meet this burden, the trial court did not err in granting the motion for summary judgment as to Dr. Nelson and the hospital on plaintiffs claims of negligence.

We affirm the district court in granting summary judgment in favor of defendants Dr. Nelson and St. Joseph Medical Center. The judgment of the district court granting summary judgment in favor of defendant Sterling Drug is reversed, and the case is remanded for further proceedings.

McFARLAND, J., concurring and dissenting:

I disagree with the reversal of the summary judgment entered in favor of Sterling Drug. There had never been a single reported case of permanent paralysis associated with the drug. The warning adequately stated all known adverse reactions. The plaintiffs experts would require Sterling Drug, Inc., to speculate that a temporary (three week) paralysis in one patient could give rise to a permanent paralysis in another and include such speculation as a known risk associated with the use of the drug.

I believe the trial court correctly applied Comment k.