dissenting.
I respectfully dissent. I would affirm the summary final judgments because the undisputed chronology reveals that the actions are time-barred.
Viewing the record and all attendant inferences in the light most favorable to Bo-gorff, the party opposing the motions for summary judgment, Holl v. Talcott, 191 So.2d 40 (Fla.1966), the facts giving rise to these actions may be summarized as follows. Adam Bogorff was born on October 16, 1966. In April, 1970, Adam was diagnosed as having undifferentiated lymphob-lastic leukemia by Dr. Edward Morse of the University of Connecticut medical school in Hartford, Connecticut. Adam was treated with the drugs oncovin, mecap-topurine and prednisone in order to induce a remission. In June, 1970, Adam and his parents, Robert and Thelma, moved to Florida. Pursuant to the recommendation of Adam’s physicians in Connecticut, the Bogorffs placed Adam under the care of Dr. Kjell Koch, a pediatric hematologist at the University of Miami medical complex. The Bogorffs selected Dr. Paul Winick as *1229Adam’s pediatrician. Dr. Winick, who was not affiliated with the University of Miami, cared for Adam before, during, and after his treatment by Dr. Koch.
Because Adam’s Connecticut physicians had successfully induced a remission in his leukemia, Dr. Koch placed Adam on a regimen of drug therapy to maintain his remission. Dr. Koch continued Adam on the drugs previously administered in Connecticut with only one change. He substituted methotrexate, a drug manufactured by Lederle, for mercaptopurine. The metho-trexate was administered orally. In the spring of 1971, Thelma Bogorff approached Dr. Koch regarding the use of some type of prophylaxis to prevent cancer cells from invading Adam’s central nervous system. She specifically inquired into the possibility of treating Adam with cranial radiation in view of the positive results that a Dr. Aur at St. Jude Children’s Research Hospital in Memphis, Tennessee, had achieved in child leukemia patients. Dr. Koch refused to commit to such treatment until he had obtained the quarterly report on leukemia therapy from the National Cancer Institute in May, 1971. In June, 1971, Dr. Koch determined that prophylaxis consisting of intrathecal (intraspinal) injections of metho-trexate was appropriate. Dr. Koch administered two intrathecal injections of metho-trexate.1 In the summer of 1971, Dr. Koch referred Adam to Dr. Komanduri Charyu-lu, the chief radiologist at the University of Miami. Dr. Charyulu ordered prophylactic cranial and spinal radiation administered to Adam. During the radiotherapy, Adam experienced several adverse side effects that Dr. Koch had told the Bogorffs could result from the combined treatment of intrathecal methotrexate and radiation, e.g., loss of hair and appetite, headaches, and nausea. In September, 1971, the Bogorffs reported to Dr. Koch that Adam had slept for an entire day and had subsequently appeared lethargic. These symptoms abated in the fall of 1971 when the radiation ceased. Adam received a third intrathecal injection of methotrexate in January, 1972, again accompanied by radiation. The following month the Bogorffs perceived distinct, alarming changes in Adam’s condition, including lethargy, headaches, seizures, jerking movements, slurred speech, and impaired motor skills. Three months later, in April, 1972, Adam suffered convulsions and lapsed into a coma. Dr. Koch discontinued the intrathecal injections of methotrexate. Dr. Robert, F. Cullen, a pediatric neurologist at the University of Miami, was called in to assess Adam’s condition. Dr. Cullen attributed Adam’s decline to a subdural implant indicating the presence of leukemia cells in the brain. He prescribed additional radiation followed by chemotherapy which included the oral administration of metho-trexate. Although Dr. Cullen’s treatment succeeded in rousing Adam from his comatose condition, Adam never regained his former state of health and never recovered his speech or motor skills.
By July, 1972, Adam had become paralyzed and unresponsive. At this time, the Bogorffs consulted Dr. Gary Giesecke, a neurologist who was not affiliated with the University of Miami. Pursuant to the Bo-gorffs’ request that Adam’s physicians maintain communications, Dr. Giesecke conveyed his findings by way of a letter to Dr. Koch wherein he noted that the Bo-gorffs had provided all of Adam’s medical records to assist him in- evaluating Adam. Dr. Giesfecke agreed that Adam had suffered brain damage and offered three possibilities as to the type of damage: localized leukemic implant, multifocal leuko-en-cephalopathy, or subcortical demyelination. Dr. Giesecke concluded his letter with the following observation: “I think the parents do realize that this child is quite ill and the prognosis is grave. They, of course are looking for help from any source, and this *1230in itself has led to some confusion on their part. I encouraged them to continue with their care at the University of Miami and discouraged them from seeking information or help from outside sources, since I do not think it would be of any further significant benefit to this child.”
The Bogorffs returned to Dr. Koch, who had ascribed Adam’s brain damage to either leukemia cells invading the brain or a viral infection.' Robert Bogorff, in his capacity as a university1 librarian, had discovered an article in a British medical journal which suggested a possible connection between methotrexate therapy in leukemia patients and encephalopathy or brain damage.2 The article detailed “encephalopathy in seven patients who were receiving prolonged methotrexate therapy, in part by intrathecal injection.” The abstract theorized that the brain damagé did not conform to the type of brain damage caused by Viral encephalitis. The abstract further explained that because the brain damage improved somewhat with the administration of folic acid, “the condition appears to be due to methotrexate.” Finally, the abstract characterized the relationship between methotrexate and encephalopathy as “a condition not previously described.” Bo-gorff shared the- article with Dr. Koch and inquired as to its relevance to Adam. According to Bogorff, Dr. Koch read the article in his presence, dismissed it as inapplicable to Adam, and discarded it in the trash.3 Dr. Koch maintained his position that Adam’s brain damage was the consequence of either a virus or proliferating cancer cells. In an attempt to secure a more definitive .diagnosis of Adam’s condition, Dr. Koch contacted Dr. Aur at St. Jude Children’s Research Hospital. Dr. Aur was unable to provide any additional insight. After 1973, Adam received no treatment for his leukemia, which remained in remission; instead, Dr. Koch simply monitored Adam’s leukemia by taking blood samples. When Dr. Koch left Miami in 1978, he did not take Adam’s medical records with him.
Dr. Cullen left the University of Miami to become director of pediatric neurology at Variety Children’s Hospital in Miami, but he continued to treat Adam for his neurological damage. Although Adam initially reacted favorably to a course of weekly intravenous B-12 and folic acid, the treatment proved unsuccessful and was abandoned in 1973 or 1974. Despite Dr. Cullen’s program of various antiseizure medications, he was unable to alter Adam’s cognitive functions. Regrettably, as Dr. Cullen noted in his neurology consultation of May, 1975, Adam remained “at a 10% level when one takes 100% as being normal.” In 1975, either Dr. Winick or Dr. Cullen referred Adam to Shands Teaching Hospital in Gainesville, Florida, for a complete metabolic evaluation during which an EEG was performed. No additional knowledge was obtained by these studies. Neither did Dr. Cullen change his treatment of Adam as a result of the Shands consultation.
In 1977, Adam was evaluated by Dr. Paul Zee and Dr. Billmeier at St. Jude Children’s Research Hospital. Concerned by Adam’s stunted growth, the Bogorffs hoped that Dr. Zee, as chief of nutrition and metabolism, could provide recommendations for improving Adam’s stature. Dr. Billmeier conducted a physical examination of Adam which included a CAT scan. Again the physicians were unable to recommend any protocol for Adam’s improvement. At Thelma Bogorffs request, Dr. Zee communicated his findings to Dr. Win-ick in a letter dated July 18, 1977. Dr. Zee summarized his diagnosis: “Encephalopathy with irreversible anatomical changes possibly secondary to radiation and intra-*1231thecal methotrexate.” In a separate letter to Thelma Bogorff dated July 19, 1977, Dr. Zee commended her on her care of Adam and advised her that the results of all of Adam’s tests at St. Jude’s had been sent to Dr. Winick.
In 1979, the Bogorffs applied for supplemental social security on Adam’s behalf but submitted no supporting documentation. The application was rejected. Also in 1979, the Bogorffs consulted counsel regarding the existence of a medical malpractice or products liability claim for Adam’s injury. They were advised that no cause of action existed. In 1982, the Bogorffs reapplied to the Social Security Administration for financial aid for Adam. To support their second application, they obtained Adam’s complete medical records from Dr. Winick. For the first time, the Bogorffs read the correspondence among the various physicians who had treated or evaluated Adam. The letters acknowledged the possibility that methotrexate could have been a cause of Adam’s brain damage. The Bogorffs consulted another attorney to investigate the viability of a lawsuit.4
In December, 1982, the Bogorffs filed a complaint against Dr. Koch and the University of Miami for medical malpractice and against Lederle for negligence and products liability. The subsequently amended complaint alleged, in relevant part, that Dr. Koch and the University were negligent in their treatment of Adam as evidenced by the failure to recognize methotrexate as the cause of Adam’s brain damage and to take prompt action to reverse the damage; the failure to seek the assistance of neurologists in reversing the damage; the failure to make full and fair disclosure of the fact that prophylactic intrathecal methotrexate was an experimental procedure; and the failure to inform the Bogorffs that intra-thecal administration of methotrexate had not been approved by the Food and Drug Administration [FDA] at the time of Adam’s treatment. The amended complaint against Lederle alleged that Lederle was negligent (1) in not providing, through package inserts or other information, adequate warnings to physicians of the adverse side effects, reactions, and contraindications of methotrexate; (2) in not providing full disclosure of necessary information so as to allow the physician to determine the ratio of risk to benefit for the patient; (3) in not warning the medical community that intrathecal doses of methotrexate could lead to meningitis and central nervous system deterioration in children; and (4) in not following applicable federal rules and regulations for labelling, testing, and marketing methotrexate.
Dr. Koch, the University, and Lederle answered the complaint and raised as an affirmative defense the expiration of the relevant statute of limitations.5 They subsequently moved for summary final judgment based upon the expiration of the statute of limitations. In opposition to the motions for summary judgment, the Bo-gorffs argued that the statute of limitations had been tolled by Dr. Koch’s fraudulent concealment and misrepresentations as to the connection between Adam’s brain damage and methotrexate. The Bogorffs also contended that Lederle’s fraud and concealment tolled the statute of limitations. The Bogorffs specifically alleged that Lederle had intentionally misrepresented the adverse side effects of methotre-xate through its annotation in the 1970 Physician’s Desk Reference which stated that the FDA had approved methotrexate *1232for intrathecal use.6
The controversy at the hearings on the motions for summary judgment centered on whether the four-year statute of limitations had run prior to the commencement of the Bogorffs’ actions in December, 1982. In support of their motion in opposition to summary judgment, the Bogorffs submitted their affidavits and the letters from Adam’s medical records. According to their pleadings, only Dr. Zee’s 1977 letter to Dr. Winick afforded the Bogorffs notice of the irreversible nature of Adam’s injury and possible nexus to intrathecal methotre-xate. The Bogorffs contended that the statute of limitations did not commence to run until 1982 when they first read Dr. Zee’s crucial letter. Dr. Koch, the University, and Lederle argued that the Bogorffs had access to Adam’s medical records at all times and that Robert Bogorffs awareness of the 1972 journal article belied his position of ignorance for a decade thereafter. Following the hearings, the trial court entered summary final judgments in favor of Dr. Koch, the University, and Lederle.7
I would affirm the summary final judgments in favor of Dr. Koch and the University. The record is devoid of evidence that Dr. Koch had fraudulently concealed the cause of Adam’s injury. In the absence of such concealment, the statute of limitations is not tolled. Nardone v. Reynolds, 333 So.2d 25, 39 (Fla.1976), clearly establishes, that a physician has a duty to disclose only “known facts and not conjecture and speculation as to possibilities.” See Almengor v. Dade County, 359 So.2d 892, 894 (Fla. 3d DCA 1978). The Bogorffs’ claim, that Dr. Koch’s denial of the possibility that Adam’s brain damage was caused by the intrathecal administration of methotrexate constituted “concealment,” is unsupported by the record or the law. There is no showing that, in the exercise of reasonable medical diligence, Dr. Koch should have known that the methotrexate was the cause of Adam’s brain damage. Dr. Koch’s rejection of the medical journal article linking methotre-xate treatment of leukemia patients to encephalopathy and his alternative medical opinion that the damage was the consequence of either a virus or cancer cells invading the brain do not amount to “a fraudulent withholding of the facts, sufficient to toll the running of the statute.” Nardone, 333 So.2d at 39.
According to the medical consensus in the record, methotrexate was merely one possibility in a lengthy catalogue of suspected causes. A survey of the various medical opinions regarding Adam’s condition demonstrates the lack of any fraudulent concealment by Dr. Koch. Dr. Winick had opined in a 1973 letter to Dr. Koch: “Whether this whole business is secondary to Methotrexate is difficult to ascertain.” In a 1977 letter, Dr. Winick speculated: “He [Adam] [sjeems to have some type of encephalopathy, which is related to either his leukemia or more likely to radiation or perhaps related to a folic acid deficiency.” In his 1975 neurological evaluation of Adam, Dr. Cullen stated: “He seems to have some type of peculiar encephalopathy, either related to his leukemia, radiation, or perhaps related to a folic acid deficiency accompanying use of Methotrexate.” Dr. Zee had thought the encephalopathy was “possibly secondary to radiation and intra-thecal methotrexate.” The radiologist, Dr. Charyulu, posited “that there may be some distant or remote connection with Metho-trexate toxicity.” The inability of all the medical specialists to identify the precise cause of Adam’s brain damage is best exemplified by Dr. Winick’s terse observation to Dr. Koch in 1973: “[T]he only way I think we will be able to make a definitive diagnosis, is a post.” To conclude that fraudulent concealment exists where the uncontroverted evidence suggests only a tenuous connection between methotrexate *1233and Adam’s brain damage is to require of Adam’s treating physician twenty-twenty foresight.
“Admittedly the science of medicine is not an exact science.” Bourgeois v. Dade County, 99 So.2d 575, 577 (Fla.1956). At best, the evidence in this case suggests only that in Dr. Koch’s medical judgment methotrexate could be rejected as the cause of Adam’s brain damage although other experts might not have eliminated methotrexate from consideration. A divergence of expert views cannot form the basis for fraudulent concealment. Fraudulent concealment exists (1) where a physician (or his employee, agent, or servant) engages in conduct to prevent inquiry or elude investigation as to the cause of action or (2) fails to reveal to the plaintiff facts (not possibilities or conjecture) relating to the nature or cause of the plaintiff’s condition. Almengor, 359 So.2d at 894. Here, the record shows that the Bogorffs did not blindly accept Dr. Koch’s diagnosis of a virus or spreading cancer cells but instead promptly consulted at least two neurologists, Dr. Cullen and Dr. Giesecke, who were similarly unable to isolate the etiological agent. Significantly, Dr. Koch did not mislead the Bogorffs as to the severity of Adam’s injury or encourage them as to his prospect for recovery. See Swagel v. Goldman, 393 So.2d 65 (Fla. 3d DCA 1981) (where plaintiff was incontinent after operation performed by defendant but defendant assured him that condition was only temporary, genuine issue of material fact existed as to whether plaintiff discovered or should have discovered injury prior to the time when defendant finally told him that incontinence was permanent). Neither were the Bogorffs discouraged or otherwise hindered in obtaining Adam’s medical records from Dr. Winick, Dr. Koch, Dr. Cullen, or any of the consulting physicians. The Bogorffs were obviously cognizant of the important data located in Adam’s records as demonstrated by their furnishing the record for Dr. Giesecke’s scrutiny in 1972. The Bogorffs admitted in their depositions that they enjoyed open communications with all of Adam’s treating physicians, including Dr. Koch. The Bo-gorffs had specifically directed the various physicians to keep one another informed of their findings, thereby indicating their awareness that Adam’s file contained valuable material.
“Under Florida’s discovery standard, the cause of action does not accrue, for limitations purposes, until the injured party discovers or has a ‘duty to discover the act constituting an invasion of his legal rights’.” Celotex Corp. v. Meehan, 523 So.2d 141, 145 (Fla.1988) (quoting Creviston v. General Motors Corp., 225 So.2d 331, 334 (Fla.1969)). The trial court correctly found that, as a matter of law, the Bogorffs knew or should have known of the existence of their cause of action more than four years prior to the filing of their complaint in 1982, notwithstanding Dr. Koch’s opinion that methotrexate was not the cause of Adam’s injury. See Nardone (limitations period commences when plaintiffs have knowledge of change in physical condition even where they do not know of causal connection of defendant’s act); Carr v. Broward County, 505 So.2d 568 (Fla. 4th DCA 1987) (seven-year statute of repose barred medical malpractice action brought ten years after child’s birth where any injury to child from medical malpractice was completed fact on or before time of birth), aff'd, 541 So.2d 92 (Fla.1989); Carlton v. Ridings, 422 So.2d 1067 (Fla. 1st DCA 1982) (medical malpractice claim time-barred where injury resulting from alleged negligent administration of radiation was discovered within four years from date of injury but suit was not filed before expiration of four-year statute of limitations); Robinson v. Sparer, 365 So.2d 438 (Fla. 3d DCA 1978) (malpractice action barred by statute of limitations where plaintiff was aware of and claimed damage from treatment by physician more than two years prior to filing complaint). See also Moore v. Morris, 475 So.2d 666 (Fla.1985) (summary judgment precluded where material issue of fact existed as to when parents of brain-damaged child should have discovered cause of injury); Florida Patient’s Compensation Fund v. Sitomer, 524 So.2d 671 (Fla. 4th DCA 1988) (statute of limita*1234tions issue properly submitted to jury where there was substantial question of fact regarding extent and timing of notice to plaintiff of injury and its cause); Schafer v. Lehrer, 476 So.2d 781 (Fla. 4th DCA 1985) (knowledge of physical injury alone, without knowledge that injury resulted from negligent act, does not trigger limitations).
Even accepting the Bogorffs’ argument that Dr. Zee’s 1977 letter to Dr. Winick was the only document which could have alerted them to intrathecal methotrexate as the etiological agent, the Bogorffs had both actual and constructive notice of the letter prior to 1978, the latest date of discovery in order to sustain an action filed in 1982. Dr. Zee had informed Mrs. Bogorff in 1977 that his medical findings had been sent under separate cover to Dr. Winick, thereby giving her actual notice of the letter’s existence. Nardone teaches that absent fraudulent concealment knowledge of the contents of a patient’s medical records will be imputed. Thus, the Bogorffs clearly had constructive notice of the information in Adam’s medical file. Even permitting the Bogorffs the latitude of a 1977 discovery date with Dr. Zee’s letter triggering the running of the statute of limitations, their action was time-barred by 1982 when their complaint was filed. Because the action against the University of Miami was based upon a theory of vicarious liability, I would affirm summary judgment in favor of the university as well.
I also agree with the trial court that summary judgment for Lederle is required as a matter of law because the sequence of events culminating in Adam’s brain damage is undisputed. Adam was treated with intrathecal methotrexate in 1971 and January, 1972. The Bogorffs perceived a deterioration in Adam’s condition by the spring of 1972. According to the timetable, both the alleged negligent act (the intrathecal injection of methotrexate) and the alleged injury (Adam’s irreversible brain damage) occurred, at the latest, in 1972 or 1973. Section 95.11(4), Florida Statutes (1971), provides a four-year statute of limitations for products liability actions.8 The Bo-gorffs’ complaint against Lederle was filed on December 1, 1982, well beyond the four-year limit, and was unequivocally time-barred.
Steiner v. Ciba-Geigy Corp., 364 So.2d 47 (Fla. 3d DCA 1978), cert. denied, 373 So.2d 461 (Fla.1979), is remarkably similar to the facts of this case. In Steiner, this court affirmed a summary final judgment for a defendant-drug manufacturer on the ground that the action was barred by the four-year statute of limitations. The plaintiff had suffered a loss of vision allegedly due to his ingestion of a prescription drug manufactured by the defendant. Both the drug ingestion and the vision loss had occurred in 1972. The plaintiff filed his action against the drug manufacturer in 1977. This court affirmed the summary judgment for the drug manufacturer upon a finding that “the facts known to the plaintiff were sufficient as a matter of law to begin the running of the statute of limitations.” Id. at 52. This court further observed that the “statute of limitations will begin to run only when the ‘moment of trauma’ and the ‘moment of realization’ have both occurred. By ‘trauma,’ we simply mean the ill effect, damage or injury; and by ‘realization,’ we mean the ‘known or should have known’ element associated with the trauma.” Id. at 53. This case falls within the Steiner paradigm since both the “moment of trauma” and the “moment of realization” transpired in 1972. The Bogorffs were on notice at that time as demonstrated by their proffer of the journal article to Dr. Koch and their ensuing consultations with other specialists in 1972 in search of a more definitive diagnosis and possible remedy. See generally *1235Annotation, Statute of Limitations: When Cause of Action Arises on Action Against Manufacturer or Seller of Product Causing Injury or Death, 4 A.L.R.3d 821 (1965).
“[O]nce there is notice of the negligent act, the statute of limitations begins to run even if the full consequences are not known.” Byington v. A.H. Robins Co., Inc., 580 F.Supp. 1513, 1517 (S.D. Fla.1984). The Bogorffs’ argument, that they did not realize that Adam’s brain damage was permanent until they read Dr. Zee’s letter, is unavailing. By any meter, the Bogorffs knew or should have known of the irreversible nature of the injury in 1973 when Dr. Cullen abandoned all attempts at palliation. Byington is instructive on the question of when the Bogorffs knew or should have known of methotrexate as the possible catalyst. In Byington, the district court, applying Florida law, granted summary judgment in favor of the manufacturer of an intrauterine device on the ground that the action was barred by the four-year statute of limitations. Relying on Steiner, the court found that the plaintiff’s “persistent pelvic pain, coupled with her doctor’s verbalized suspicions and Defendant’s public disclosure of the possible dangers of the Daikon Shield in 1974, should have put her on notice that her cause of action had accrued.” Id. at 1515-16. The court reasoned that “[although she did not ‘know for a fact’ that the IUD was the cause of her discomfit, Mrs. Byington certainly had available to her adequate facts ... to point in a single direction in search for the responsible party o[r] parties.” Id. at 1517 (quoting Steiner, 364 So.2d at 52). Rejecting Byington’s contention that the manufacturer’s fraudulent concealment of the dangers of the Daikon Shield tolled the statute of limitations, the court observed that “the failure of a possibly responsible party to ascertain and publish the fact of its responsibility is hardly sufficient to constitute fraudulent concealment.” Id. at 1518. Under Byington, there exists no genuine issue of material fact that the Bo-gorffs’ cause of action accrued more than four years prior to the filing of the complaint.
The questions of whether Lederle fulfilled its duty to warn the medical community of the potential dangers of intrathecal use of methotrexate9 and whether Dr. Koch met his concomitant duty as learned intermediary to disclose to the Bogorffs the side effects of methotrexate10 are not *1236genuine issues of material fact sufficient to preclude summary judgment. While the record is less than clear regarding the adequacy of the warnings given by Lederle and the reasonableness of the disclosure made by Dr. Koch, these considerations are rendered moot by the statute of limitations.11 A cause of action is not deemed to accrue when the wrong has been done or the duty has been breached but when the resulting damage is sustained and is capable of ascertainment. See Steiner; Krug v. Sterling Drug, Inc., 416 S.W.2d 143, 148-50 (Mo.1967). “A statute of limitation runs from the date the cause of action arises; that is, the date on which the final element (ordinarily, damages, but it may also be knowledge or notice) essential to the existence of a cause of action occurs.” Carr, 505 So.2d at 570. There is no merit in, and no record support for, Bogorff’s contention that Lederle’s failure to warn and Dr. Koch’s failure to obtain informed consent constituted a continuing wrong against which the limitations period may only begin to run from the time of cessation of the wrong. Accordingly, I would affirm the summary final judgments in favor of Dr. Koch, the University, and Led-erle.
. Treatment of acute leukemia in the child involves two phases: induction of remission and maintenance of remission_ For maintenance, it has been found that the intramuscular or oral administration of methotrexate twice weekly is superior to daily oral dose of methotrexate or mercaptopurine or to courses of methotrexate alternated with courses of mercaptopurine.... Methotrexate, given in the spinal canal, produces improvement in 80 percent of patients with meningeal leukemia.
3B L. Frumer & M. Friedman, Products Liability § 50.01 [26] (Supp.1986).
. Kay, Encephalopathy in Acute Leukaemia Associated with Methotrexate Therapy, 47 Archives of Diseases in Childhood 344 (1972).
. Dr. Koch acknowledged in his deposition that he had summarily told the Bogorffs that the article did not pertain to Adam's case. Dr. Koch pointed out several significant differences between the seven patients chronicled in the article and .Adam, the most dramatic being the much largdr doses of methotrexate the seven patients had received. Dr. Koch further stated that he did not dispose of the article in the trash. i
. The Bogorffs subsequently filed a legal malpractice complaint against the attorney they had consulted in 1979. The basis of their claim is that the attorney failed to apprise them of the statute of limitations restrictions.
. The motions for summary judgment cited section 95.11(6), Florida Statutes (1973), or section 95.11(4)(b), Florida Statutes (1975), as the controlling statute of limitations. The Bogorffs contended that section 95.11(4), Florida Statutes (1971), was applicable. Although the trial court’s orders granting summary final judgment for Dr. Koch, the University, and Lederle do not cite to a specific statute, it is clear that the parties and the trial court were working with a four-year-limitation period. My analysis is predicated on the statute of limitations advocated by the Bogorffs, section 95.11(4), Florida Statutes (1971).
. The Physician’s Desk Reference, published annually, consists of monographs summarizing FDA-approved package inserts.
. The trial court had previously entered summary judgments in favor of Dr. Koch, the University, and Lederle but had vacated them, pursuant to Bogorffs motion, on the authority of Ash v. Stella, 457 So.2d 1377 (Fla.1984). The orders under review on these appeals were entered on the renewed motions for summary final judgment filed by Dr. Koch, the University, and Lederle.
. 95.11 Limitations upon actions other than for the recovery of real property. — Actions other than those for the recovery of real property can only be commenced as follows:
[[Image here]]
(4) WITHIN FOUR YEARS. — Any action for relief not specifically provided for in this chapter.
§ 95.11(4), Florida Statutes (1971).
. As the manufacturer of a prescription drug, Lederle had a duty to warn the medical community of potential side effects resulting from the use of its product. Davis v. Wyeth Laboratories, Inc., 399 F.2d 121 (9th Cir.1968); Felix v. Hoffmann-LaRoche, Inc., 513 So.2d 1319 (Fla. 3d DCA 1987), rev. granted, No. 71,633 (Fla. Mar. 18, 1988); Ricci v. Parke-Davis & Co., 491 So.2d 1182 (Fla. 4th DCA), rev. denied, 501 So.2d 1283 (Fla.1986); Buckner v. Allergan Pharmaceuticals, Inc., 400 So.2d 820 (Fla. 5th DCA), rev. denied, 407 So.2d 1102 (Fla.1981); McEwen v. Ortho Pharmaceutical Corp., 270 Or. 375, 528 P.2d 522 (1974); Terhune v. A.H. Robins Co., 90 Wash. 2d 9, 577 P.2d 975 (1978). See generally Comment, The Drug Manufacturer’s Duty to Warn — To Whom Does It Extend?, 13 Fla.St.U.L. Rev. 135 (1985); Annotation, Liability of Manufacturer or Seller for Injury or Death Allegedly Caused by Failure to Warn Regarding Danger in Use of Vaccine or Prescription Drug, 94 A.L. R.3d 748 (1979), and cases cited therein.
. Where a product is available only on prescription or through the services of a physician, the physician acts as a "learned intermediary” between the manufacturer or seller and the patient. It is his duty to inform himself of the qualities and characteristics of those products which he prescribes for or administers to or uses on his patients, and to exercise an independent judgment, taking into account his knowledge of the patient as well as the product. The patient is expected to and, it can be presumed, does place primary reliance upon that judgment.
Buckner, 400 So.2d at 823 (quoting Terhune, 577 P.2d at 978). A doctor’s duty is to disclose to the patient the necessary information to enable him to balance the risks against the benefits. Buckner, 400 So.2d at 823 (quoting ZeBarth v. Swedish Hosp. Medical Center, 81 Wash.2d 12, 499 P.2d 1, 11 (1972)). The relevant standard for assessing the extent of the information which may be required is "whether a reasonable medical practitioner in the community would make the pertinent disclosures under the same or similar circumstances." Ritz v. Florida Patient’s Compensation Fund, 436 So.2d 987, 991 (Fla. 5th DCA 1983), rev. denied, 450 So.2d 488 (Fla.1984); Thomas v. Berrios, 348 So.2d 905, 908 (Fla. 2d DCA 1977); Ditlow v. Kaplan, 181 So.2d 226, 228 (Fla. 3d DCA 1965).
. The record contains the package inserts for methotrexate as well as the FDA applications and supporting documents and PDR annotations for 1970, 1971, and 1972. However, the record contains no evidence of the state of expert knowledge regarding the connection between intrathecal use of methotrexate and brain damage during the relevant time frame of 1971 to 1972. Drug manufacturers are held to the level of knowledge of an expert in the field. The duty to warn of a risk does not arise until the manufacturer knows or should know of the risk. Reyes v. Wyeth Laboratories, Inc., 498 F.2d 1264 (5th Cir.), cert. denied, 419 U.S. 1096, 95 S.Ct. 687, 42 L.Ed.2d 688 (1974); Davis v. Wyeth Laboratories, Inc., 399 F.2d 121 (9th Cir.1968); Ortho Pharmaceutical Corp. v. Chapman, 180 Ind.App. 33, 388 N.E.2d 541 (1979); McEwen v. Ortho Pharmaceutical Corp., 270 Or. 375, 528 P.2d 522 (1974). "[T]here is no duty to warn of unknown or unforeseeable risks. Dates are thus vitally important, since the duty to warn depends on when the risk became apparent." Basko v. Sterling Drug, Inc., 416 F.2d 417, 426 (2d Cir.1969).