523 So.2d 815 (1988)
Gordy SMITH, et al.
v.
STATE of Louisiana Through the DEPARTMENT OF HEALTH AND HUMAN RESOURCES ADMINISTRATION and Dr. Edward Staudinger.
No. 87-C-2418.
Supreme Court of Louisiana.
April 11, 1988.
Russell Tritico, Lake Charles, for applicants.
Frederick Alexius, Stephen D. Wheelis, Provosty, Sadler & deLaunay, Alexandria, for respondents.
CALOGERO, Justice.
This is a wrongful death and survival action brought by the surviving spouse and children of one Ellie Smith. Mrs. Smith was admitted to the emergency room at the Huey P. Long Charity Hospital in Rapides Parish on December 10, 1982, at which time she complained that she was feeling bad and experiencing shortness of breath. About four and one-half hours later, while Mrs. Smith was still under observation in the emergency room, she began exhibiting symptoms of cardiac arrest. An hour later, after unsuccessful rescusitation efforts (CPR), she was pronounced dead.
The defendants are the State, through the Department of Health and Human Resources, as operator of the hospital, and Dr. Edward Staudinger, the emergency room physician on duty on the night in question. The plaintiffs allege that Mrs. Smith's death was caused by the negligent failure of the emergency room staff to observe and treat the decedent properly, and to perform promptly the necessary tests which would have allowed them to diagnose Mrs. Smith's heart condition and render preventive treatment.
After a trial on the merits, the district court found that the defendants' treatment of the decedent was negligent, chiefly because of the failure of the hospital staff to administer timely an electrocardiogram (EKG) and the failure of the attending nurses and physician to monitor properly Mrs. Smith's condition during the period that she was in the emergency room prior to her death. However, the trial court entered judgment for the defendants upon finding that the plaintiffs did not prove that defendant's negligence caused decedent's heart attack and subsequent death. The court of appeal affirmed, agreeing with the trial court that the defendants were negligent, but also agreeing that there was no causal link between the negligence and the death. 517 So.2d 1072 (La. App. 3rd Cir.1987).
We granted the plaintiffs' application for a writ of review, 519 So.2d 107 (La.1987), in order to consider the court of appeal's ruling on causation in light of our decision in Hastings v. Baton Rouge General Hospital, 498 So.2d 713 (La.1986). Hastings held that in a wrongful death action involving allegations of medical malpractice, the plaintiff need not prove that the patient would have survived if the defendant had undertaken preventive or other methods of treatment. Instead, we held essentially that the plaintiff may establish a compensable claim through evidence which demonstrates that a defendant's malpractice resulted in the loss of a chance of survival of a patient who thereafter expired.
After reviewing the record in this case, we agree with the court of appeal and the trial court that the plaintiffs did not prove that negligence on the part of the defendants lessened Mrs. Smith's chances of survival or otherwise contributed to her death. While we adhere to the "loss of chance" standard described in Hastings, we find that the standard was not met here by plaintiffs, who presented little or no evidence that defendants caused the decedent to lose any chance of survival. Accordingly, we affirm the judgments of the courts below.
(I) FACTS
Effie Smith was admitted to the emergency room at the Huey P. Long Charity Hospital in Rapides Parish, Louisiana at approximately 7:30 P.M. on December 10, 1982. Accompanied by a son and daughter, Mrs. Smith complained to the admitting nurse that she had been feeling poorly for the prior week, and was experiencing difficulty breathing. She was 58 years old at that time, and was described on the emergency room admission form as an obese female in no acute distress. The nurse took Mrs. Smith's vital signs, which revealed a slightly elevated blood pressure of 180/110. All other vital signs were within normal limits.
Between 7:30 and 8:00 P.M., the patient was examined by the physician on duty at the emergency room, Dr. Edward Staudinger. Dr. Staudinger noted that Mrs. Smith was experiencing shortness of breath, and he also observed mild edema, or swelling, in her extremities. The doctor's initial diagnosis was that Mrs. Smith was suffering from mild congestive heart failure. He prescribed twenty milligrams of Lasix (a diuretic designed to reduce blood pressure), a chest X-ray and an electrocardiogram (EKG). These orders were recorded on the patient's chart at 8:05 P.M.
Anthony Gomez was the registered nurse on duty at the time that Mrs. Smith was admitted to the emergency room. Nurse Gomez testified that he administered the Lasix to Mrs. Smith at approximately 8:05 P.M. He then had the patient sent to the radiology lab (a short distance away, and within the hospital) for the chest X-ray. Dr. Staudinger had ordered the X-ray in order to rule out the possibility that the patient had fluid in her lungs, a condition associated with congestive heart failure. Although the precise time that the X-ray was taken is not clear from the record, Dr. Staudinger testified that he reviewed the X-ray results between 9:00 and 9:15 P.M. Those results were normal, and at that point the doctor felt that there was no need to hospitalize Mrs. Smith.
The patient's chart reflects that Nurse Gomez sent a request for an EKG to the hospital lab at 9:15 P.M., shortly after the X-ray had been completed. The testimony at trial further established that the EKG was not actually administered until approximately 11:30 P.M.
In the interim, Nurse Gomez's shift ended, and he was replaced by Nurse Everett Lofton. Nurse Lofton testified that while he was making his rounds between 11:30 and 11:45 P.M., he observed that Mrs. Smith was in distress. Upon closer examination, he determined that she had a faint pulse and breathing difficulties (apnea). Believing that the patient had gone into cardiac arrest, he immediately summoned Dr. Staudinger.
Dr. Staudinger testified that at the very moment that he received the distress call from Nurse Lofton, he was in the process of reviewing the results of Mrs. Smith's EKG. Those results had just been transmitted to him, the test having been administered only a few minutes earlier.[1] The EKG results showed that Mrs. Smith was experiencing atrial fibrillation, which is an irregular heart rhythm.
Upon receiving word from Nurse Lofton of the patient's condition (faint pulse and breathing difficulties), the doctor went to her bedside and, together with the nurse, administered CPR for about forty-five minutes. Mrs. Smith did not respond, and was pronounced dead at 12:37 A.M. on December 11, 1982. Dr. Staudinger listed the cause of death as cardiac arrest, secondary to arrhythmia.
It is uncertain whether Mrs. Smith's condition ever changed appreciably between the time that she was initially examined (7:30-8:00 P.M.) and that point in time just preceding the onset of the cardiac arrest discovered by Nurse Lofton at 11:45 P.M. There is no indication on her medical chart that her vital signs were ever re-checked during the period between her initial examination and the cardiac arrest. Dr. Staudinger testified that he received confirmation from Nurse Gomez between 9:00 and 10:00 P.M. that there had been no change in the patient's vital signs, an inquiry which he made in order to determine if the Lasix was having any effect. The doctor also testified that after Mrs. Smith died, he confirmed from Nurse Lofton that the patient's vital signs had been taken, and were essentially unchanged, at the time Lofton began his shift (around 11:15 P.M.). Nurses Gomez and Lofton had no specific recollection of the frequency with which they had monitored Mrs. Smith's vital signs during the period prior to the cardiac arrest, although both nurses stated that it would have been their customary practice to monitor the patient's condition on a regular basis. However, Mrs. Smith's daughter testified that she was at her mother's side throughout the evening and saw no one check her mother's vital signs. The more probable likelihood is that Mrs. Smith's condition was not monitored on a regular basis, as the trial court found.
There is no explanation in the record of why the EKG, ordered by the doctor at about 8:00 P.M., was not administered until approximately 11:30 P.M. Dr. Staudinger admitted that this delay was unusual. Nurse Gomez testified that he did not request an EKG from the lab until 9:15 P.M. because he waited to make the request until after the patient had returned from the X-ray lab. He testified that he sent the patient to the X-ray lab before having the EKG done because, given the preliminary diagnosis of congestive heart failure, a chest X-ray was of higher priority than an EKG.[2]
The EKG results showed that Mrs. Smith was experiencing atrial fibrillation, an irregular heart rhythm. The test results also showed that her heart rate had increased (from 90 at the time of admission) to 170. Dr. Staudinger testified that while atrial fibrillation is not necessarily a life threatening condition, EKG results alerting him concerning that condition normally would prompt him to institute treatment of a sort to which Mrs. Smith was not exposed during her four hour stay in the emergency room. He did not have time to react to the EKG results because the patient went into cardiac arrest at about the time that the results of the test were being printed and reviewed by him. The doctor also expressed the opinion that Mrs. Smith was not in the midst of cardiac arrest at the time the test was administered.
Dr. Staudinger stated that he had no way of knowing whether the EKG would have revealed an atrial fibrillation if it had been taken earlier in the evening. He testified that atrial fibrillation is a condition that can arise gradually or suddenly, with sudden development more probable. If the test had been administered between 9:00 and 10:00 P.M., he was of the view that "it may have shown something bad, it may have shown nothing at all." Dr. Eli Sorkow, who testified by deposition as an expert witness on behalf of the plaintiff, similarly testified that it is difficult to know whether Mrs. Smith's condition ever changed appreciably prior to the cardiac arrest. Dr. Sorkow indicated that any answer to that question would be pure speculation, and stated "I wasn't there and it would be hard for me to say."
Dr. Staudinger also testified that if the EKG had been taken earlier and had yielded the same results (atrial fibrillation), he would have (1) placed Mrs. Smith on a heart monitor; (2) placed her in the intensive care unit; (3) prescribed an I.V. for medication "if she needed it" (the type of medication to be used was not specified) and (4) sought a medical consultation from another doctor in order to evaluate the atrial fibrillation.
Asked if the implementation of these procedures would have given him a "reasonable chance" of saving Mrs. Smith's life, Dr. Staudinger answered "no." He went on to explain his view that regardless of whether the cardiac arrest occurred in the emergency room or in the intensive care unit, "I don't believe it would have made any difference." Asked if he would have had a better chance of prolonging the patient's life if the atrial fibrillation had been discovered earlier in the evening, he replied "It is possible that it may have ..., then it is also very possible that it wouldn't have...."
Plaintiff's expert, Dr. Sarkow, indicated in his testimony that if the patient's condition did worsen earlier in the evening, the emergency room staff possibly could have increased her chances of survival by administering medication. He did not specify the type of medication which could have been used, and he was not asked to explain how any other forms of preventive treatment, such as a cardiac monitor, could have increased the patient's chances of survival. He also was not asked what the probabilities were that such measures would have been successful.
(II) ANALYSIS
The court of appeal correctly stated a plaintiff's burden of proof in a malpractice action against a physician:
"In a medical malpractice action based on the negligence of a physician, the plaintiff bears the burden of establishing the doctor's deviation from the standard of care required by others in the same field. Plaintiff also must establish a causal relationship between the doctor's alleged negligence and the injury resulting therefrom. LSA-R.S. 9:2794; Banda v. Danburg, 469 So.2d 264 (La.App. 1 Cir.1985)." 517 So.2d at 1075.
In a medical malpractice action against a hospital, the plaintiff must prove, as in any negligence action, that the defendant owed the plaintiff a duty to protect against the risk involved, that the defendant breached that duty, that the plaintiff suffered an injury, and that the defendant's actions were a substantial cause in fact of the injury. See Sibley v. Board of Supervisors of Louisiana State University, 477 So.2d 1094, 1099 (La.1985); Belmon v. St. Frances Cabrini Hosp., 427 So.2d 541 (La.App. 3 Cir.1983); Biggs v. United States, 655 F.Supp. 1093, 1094 (W.D.La. 1987). "A hospital is bound to exercise the requisite amount of care toward a patient that the particular patient's condition may require," and to protect the patient from "external circumstances peculiarly within the hospital's control." Hunt v. Bogalusa Community Medical Center, 303 So.2d 745, 747 (La.1974). A determination of whether a hospital has breached those duties depends upon the facts and circumstances of each particular case. Id.
We find no reason to disagree with the conclusion of the lower courts that the lack of medical attention provided to Mrs. Smith after her initial examination upon admission to the emergency room and during the period prior to the cardiac arrest did in fact fall below the applicable standard of care.[3] While there was an absence of direct testimony on the question of the standard of care owed by emergency room physicians and nurses to a patient admitted with Mrs. Smith's complaints, the record gives rise to the reasonable inference that a person diagnosed with congestive heart failure should be more closely observed, monitored, and perhaps treated than was done here.
Thus we do not disturb the finding of the courts below that the defendants' treatment of Mrs. Smith was negligent in certain respects.
We turn now to the more difficult issue of causation, an issue which prompted our writ grant in this case. Causation need not always be a difficult question in a medical malpractice case, but it is made so by the particular facts of this case and by the fact that the plaintiffs presented virtually no evidence on the issue.
La. R.S. 9:2794(A)(3) requires the plaintiff to prove that as a "proximate result" of the defendant's failure to use the required degree of care, "the plaintiff suffered injuries that would not otherwise have been incurred." In a situation where the patient dies, we have held that the plaintiff does not have to shoulder the "unreasonable burden" of proving that the patient would have lived had proper treatment been given. Hastings v. Baton Rouge General Hospital, 498 So.2d 713, 721 (La.1986). Instead, the plaintiff must prove "only that there would have been a chance of survival," and that the patient was denied this chance of survival because of the defendant's negligence. Id. at 720.
In this particular case there are two separate questions pertinent to causation, and both must be answered in the affirmative in order for the plaintiff to prevail on the issue. The first question is: If the defendants had regularly monitored the patient's condition or administered the EKG sooner, would they have discovered at an earlier point in time that Mrs. Smith was in imminent danger of suffering a heart attack? This question is important because the plaintiffs do not allege that the initial course of treatment proposed by Dr. Staudingerthe Lasix, the X-ray and the EKGwas inadequate. Instead, they contend that because the defendants failed to promptly administer the EKG and did not monitor the patient's condition, they missed an opportunity to save her life. Thus, it must initially be determined whether the defendants could have made an earlier diagnosis of a dangerous condition that called for a different course of treatment. If so, the second question is whether the defendants' failure to undertake that course of treatment decreased the patient's chances of survival.
There is no evidence in the record which supports the conclusion that if the defendants had monitored Mrs. Smith's condition on a more regular basis, and administered the EKG at an earlier time, they would have discovered a change in her condition that would have triggered more active management. Dr. Staudinger testified that the onset of atrial fibrillation is more likely to be sudden than gradual. He stated that he simply did not know whether that condition would have been detected if the EKG had been taken earlier. While Dr. Staudinger's testimony on this issue has to be assessed in light of the fact that he is a defendant in the lawsuit, plaintiff's expert Dr. Sorkow was hardly more helpful to plaintiff on this issue. He stated that it was "possible" that a gradual change in Mrs. Smith's condition occurred, but he was quick to qualify his testimony and to stress that there was no way of knowing whether her condition changed gradually or suddenly.
Plaintiffs argue that the reason that the patient's condition prior to the cardiac arrest is unknown is directly attributable to the fact that the defendants failed to monitor her condition during that period. They further argue that the defendants should not prevail on this causation question when it is their own negligence which makes it impossible to know whether Mrs. Smith's condition in fact worsened at an earlier point in time.
This argument is not without some appeal. However, even if we were somehow able to conclude from the record that it is more probable than not that the decedent's condition did worsen at an earlier point in time, and that the change in her condition could and should have been promptly discovered by the defendants, we still must face the question of whether such a discovery would have made any difference in terms of the patient's chances of survival.
The court of appeal stated that "[t]here has been no expert medical testimony presented here which would suggest that decedent would have survived if diagnosed sooner." 517 So.2d at 1077 (emphasis added). However, this is not the test for causation under Hastings. The plaintiffs need only show that the decedent had a chance of survival which was denied to her as a result of the defendant's negligence. They do not have a burden of establishing that Mrs. Smith would have lived if she had received different treatment as the result of an earlier diagnosis.
Nevertheless, we believe that the result reached by the court of appeal, judgment in favor of defendants, is correct, for the plaintiffs failed to establish that the defendants' negligence denied the patient a chance of survival. There is simply no evidence in the record as to whether treatment undertaken earlier would have provided the decedent a chance of survival that she otherwise would not have had. Plaintiff's expert said that it was "possible" that measures could have been undertaken which would have increased the patient's chances of survival, but he did not specify what those measures were and he did not assess in any detail what the quantitative chances of survival would have been. Dr. Staudinger indicated that he would have adopted a different treatment plan if he had received at an earlier time the same EKG result he obtained at 11:45 P.M. However, there is no evidence which shows that these measures would have increased the chances that the patient would have avoided or survived a cardiac arrest.
Specifically, Dr. Staudinger testified that if he had discovered the atrial fibrillation at an earlier point in time, he would have (1) placed the patient on a heart monitor (2) obtained a medical consultation (3) possibly placed the patient on an unspecified type of medication and (4) moved the patient to intensive care. However, there was no testimony to the effect that any of these measures would have increased the patient's chances of survival (and if so, how, and by how much). The only witness queried to that end was Dr. Staudinger, who stated that the patient could have had a heart attack regardless of whether she was in the emergency room or the intensive care unit.
Plaintiffs seem to argue that we should assume that the defendants' failure to employ preventive treatment measures decreased the patient's chances of survival. It is true that one can logically assume that certain alternative methods of treatment, such as moving the patient to intensive care and placing her on a heart monitor, would have constituted an effort to decrease the risk that she would go into cardiac arrest. But would the risk of cardiac arrest actually have been decreased? This is a question which the record before us does not answer, and absent at least some evidence on the issue, we cannot simply assume that the defendant's failure to employ a more active course of treatment denied this heart patient a chance of survival.
A different result is appropriate in those cases where the plaintiff does in fact demonstrate, through expert testimony or otherwise, that the defendant's malpractice resulted in the loss of the patient's chances for survival. Hamil v. Bashline, 481 Pa. 256, 392 A.2d 1280 (1978) is an example of a case which involves somewhat similar facts and in which the plaintiff did present sufficient evidence on causation. Emergency room physicians failed to promptly treat a patient who was admitted complaining of severe chest pains, and later died of a heart attack. The plaintiff's expert witness testified that if the defendants had timely administered an EKG and employed other specific treatment measures, including bed rest, oxygen and pain relieving drugs, the decedent would have had a 75% chance of surviving the heart attack. A number of other common law jurisdictions have found that the plaintiff may establish causation by proving that the defendant's negligence destroyed the patient's chances of avoiding injury or death, even in instances where the patient's chances of survival were less than fifty percent under any circumstances. See Herskovits v. Group Health Cooperative of Puget Sound, 99 Wash.2d 609, 664 P.2d 474, 479-87 (1983) (extensive discussion of "loss of chance" doctrine by Pearson, J., concurring).
Even prior to Hastings, at least two Louisiana appellate court decisions upheld the right of medical malpractice plaintiffs to recover in instances where it was proven that the defendant's negligence destroyed a significant chance that the patient's death or injury could have been avoided. See, e.g., Hunter v. Office of Health Services, 385 So.2d 928, 932-33 (La.App. 2nd Cir.), writ denied 393 So.2d 737 (La.1980) (evidence established that defendant's failure to timely diagnose malignant melanoma caused cancer victim "to fall from the category of persons who would statistically have been expected to survive to the category in which there was almost no chance of survival."); Fairchild v. Brian, 354 So. 2d 675, 679-80 (La.App. 1st Cir.1977) (delay in discovering detached retina until it was too late to save plaintiff's eyesight was directly attributable to defendant's negligence).
The common denominator among these loss of chance cases in which the plaintiff has prevailedHamil, Hunter, Fairchild, and the numerous cases cited in Justice Pearson's scholarly concurring opinion in Herskovitsis that in each case there was sufficient evidence to support the proposition that the patient's chances of avoiding injury or death had been decreased due to the defendant's negligence.
Similarly, we found in Hastings that there was sufficient evidence for the jury to find that the defendants' failure to promptly perform a surgical procedure on a stabbing victim who was allowed to bleed to death without surgical intervention, destroyed the victim's chances of survival, and for that reason we reversed the trial court's directed verdict in favor of the defendant.
The medical malpractice plaintiff does not have the unreasonable burden of proving that the patient would have lived if the defendant had not been negligent. However, the plaintiff does have the burden of establishing by a preponderance of the evidence that the defendant's conduct denied the patient a chance of survival. Even if the destruction of less than a fifty percent chance of survival is compensable, the plaintiff must in the first instance prove by a preponderance of the evidence that such a chance existed and that it was lost as the result of the defendant's negligence.
In this case, there is an almost total absence from the record of any evidence tending to show that the defendants' actions or inactions denied Mrs. Smith a chance of survival. The only testimony favorable to the plaintiffs on the causation issue are general and unexplained statements by Dr. Staudinger and the plaintiff's own expert Dr. Sorkow that possibly the decedent's condition had worsened earlier in the evening and that possibly preventive measures which could have been taken upon discovery of that change in condition would have decreased the patient's chances of suffering cardiac arrest. However both of these witnesses also made clear that it was also very possible that Mrs. Smith's condition did not change earlier in the evening, and that even if her condition had changed, preventive measures would not have saved her life.
Causation is a question of fact. Miles v. Dolese Concrete Co., 518 So.2d 999, (La. 1988); Hastings, 498 So.2d at 720. Here the evidence on causation was conflicting at best. The court of appeal was of the view that the trial court's finding that the plaintiffs failed to establish causation was not clearly wrong or manifestly erroneous. And our independent review of the record convinces us that the plaintiffs did indeed fail to prove by a preponderance of the evidence that the defendants' negligence denied the decedent a chance of survival.
Finally, plaintiffs urge the applicability of the doctrine of res ipsa loquitur, and argue that once negligence was established, the burden shifted to the defendants to nullify the inference that it was their negligence which caused Mrs. Smith's death. Res ipsa loquitur applies when "circumstances suggest the defendant's negligence as the most plausible explanation for the injury." Gonzales v. Winn-Dixie Louisiana, Inc., 326 So.2d 486, 489 (La.1976). See also White v. McCool, 395 So.2d 774, 776-77 (La.1981).
In Hastings, we indicated that res ipsa loquitur was applicable to the facts before us in that case, although the majority opinion's reference to the doctrine was possibly dicta, unnecessary to a determination of the limited issue of whether the defendant was entitled to a directed verdict on causation grounds. See 498 So.2d at 723 (Hall, J., concurring). As earlier noted, Hastings involved a stab wound victim who bled to death after his injuries were not treated by surgical intervention. The majority opinion posited that the res ipsa loquitur doctrine was available to the plaintiffs because "[i]n the ordinary course of events, a man does not bleed to death in a hospital emergency room over a two and one-half hour period without some surgical intervention to save his life." 498 So.2d at 723.
In this case, however, it cannot be said that a heart attack is the type of incident which does not normally occur without negligenceeven if the heart attack occurs some four hours after a patient has been admitted to an emergency room. It is common knowledge that heart attacks can occur suddenly, and there is evidence in the record to that effect. This is not to say that in any case involving a heart attack, it cannot be demonstrated that the failure of the defendants to undertake certain preventive measures denied the patient a chance of survival. Obviously such proof is possible, as shown by Hamil v. Bashline, 392 A.2d at 1280. However, because it cannot be said that the only reasonable explanation of the heart attack was the defendants' negligence, causation cannot be presumed. Instead, the plaintiffs here retained the burden of proving that the defendants' negligence decreased the patient's chances of survival. They failed to carry that burden, for they did not introduce any testimony or other evidence to that effect.
DECREE
For the reasons stated herein, the judgments of the district court and the court of appeal are affirmed.
DIXON, C.J., dissents with reasons.
WATSON, J., dissents and assigns reasons.
DIXON, Chief Justice (dissenting).
Everybody knows Mrs. Smith's condition worsened during her stay in the hospital. While she was there she certainly lost any chance she had of survival.
WATSON, Justice, dissenting.
Failure to give an EKG, prescribed by the emergency room physician for over four hours, establishes a res ipsa loquitur inference of causal negligence on the part of the defendant hospital, which was not rebutted. See Hastings v. Baton Rouge General Hospital, 498 So.2d 713 (La.1986). Conceding that the hospital was negligent, the majority erroneously requires plaintiffs to shoulder the impossible burden of proving that Ellie Smith would not have died of cardiac arrest if promptly diagnosed and properly treated.
Therefore, I respectfully dissent.
NOTES
[1] The exact time that the EKG was administered is not shown in the record. The teletyped document containing the test results reflects that it was printed at 11:45 P.M. Dr. Staudinger accordingly estimated that the test was actually administered at 11:30 P.M., based on the assumption that it usually takes about ten minutes for the EKG to be conducted and about five more minutes before the results are printed on the teletype machine.
[2] The technician who conducted the EKG died prior to trial, and thus any information he might have had on the cause of the delay was unavailable to the trier of fact.
[3] On the other hand, the plaintiffs do not contend that Dr. Staudinger's initial examination of the patient was in any way inadequate, or that the course of treatment prescribed by him based upon his initial diagnosis was improper. Also, there is no contention much less evidence that the emergency measures implemented after decedent went into cardiac arrest were inadequate.