The trusted source for
healthcare information and
By David L. Freedman, MD, JD, FAAEM, Emergency Physician; Attorney, Miller, Canfield, Paddock & Stone PLC, Ann Arbor, MI
Headache is one of the most common symptoms in patients presenting to emergency departments (EDs), accounting for up to 2.5% of all ED visits each year. The vast majority of these patients are not seriously ill, and require only appropriate analgesia and follow-up instructions after their evaluation and treatment. However, as many as 5% of these patients will have serious neurological pathology, and as many as 1% of these patients presenting to the ED with headache will have subarachnoid hemorrhage (SAH).1,2
Approximately 30,000 people in the United States have a nontraumatic SAH each year. It is the identification of the relatively small percentage of headache patients who have serious pathology, in particular those who suffer from SAH, that is the subject of this article. Defense of a physician who has discharged from the ED a headache patient who soon thereafter suffers a catastrophic SAH is inherently difficult, if for no other reason than the potential damages suffered by the patient are huge. If the physician has not performed a complete work-up of the headache (computed tomography [CT] examination of the head and, if negative, a lumbar puncture [LP]), the defense will be that much tougher. To be sure, not all headache patients require a complete work-up. However, before sending a headache patient out of the ED without a CT scan and LP, it is prudent to assume that the patient promptly will collapse that very evening at home due to an SAH, and then to consider whether the medical record clearly supports the clinical judgment that it was reasonable to conclude that ruling out SAH was unnecessary.
Sawtelle v. Southside Hospital, et al.3
On July 14, 1985, Jacqueline Sawtelle, a 29-year-old employed mother of two, came to Southside Hospital’s ED with a history of severe headache associated with nausea and vomiting. She had been having these symptoms for two days prior to coming to the ED. There also was a question as to whether she was experiencing a complaint of neck stiffness and whether she may have had a syncopal episode (symptoms known to be associated with SAH). After evaluation in the ED, Ms. Sawtelle was discharged. It was defendant Southside Hospital’s contention that its physician had performed a general and neurological examination that had shown no abnormalities.
On July 18, Ms. Sawtelle returned to Southside Hospital with the same complaints and again was evaluated in the ED. In the interval between the two ED visits, Ms. Sawtelle had experienced an episode in which her right eye had closed involuntarily, but she did not tell this to the ED physician. She contended in her subsequent malpractice case that the emergency physician was negligent in failing to elicit this information as part of the physician’s history-taking. On July 18, Ms. Sawtelle again was discharged home from the ED.
On July 24, Ms. Sawtelle returned to the Southside ED yet again. At this time, her right eye was closed completely, and the examining neurologist elicited the history of the transient ptosis that had occurred prior to her July 18 ED visit. It was Ms. Sawtelle’s claim in the subsequent malpractice case that her eye complaint was an almost classic sign of either an aneurysm or a tumor compressing her right third cranial nerve. A CT scan performed at the time of this visit showed that Ms. Sawtelle had an aneurysm compressing her third cranial nerve.
The neurologist at Southside Hospital interpreted the CT scan as showing a ruptured aneurysm, and an angiogram performed the following day, July 25, showed an aneurysm in the right posterior communicating artery. At this point, a decision was made to transfer Ms. Sawtelle to a tertiary care center for sur-gery, and New York University (NYU) was contacted. NYU and its attending neurosurgeon, Dr. Eugene Flamm, were advised by Southside Hospital that Ms. Sawtelle had suffered an SAH, and NYU scheduled the transfer for the following day, July 26, 1985.
It was the plaintiff’s contention in her malpractice case that a Southside Hospital radiologist had read her CT and interpreted it as indicating that the aneurysm had not yet ruptured, but had negligently failed to advise Ms. Sawtelle’s treating physicians at Southside Hospital or NYU of this interpretation. Ms. Sawtelle further claimed that Southside Hospital’s failure to immediately communicate this information to the other physicians was negligent and a proximate cause of her injuries.
When Ms. Sawtelle was subsequently transferred to NYU, her medical records indicated that the aneurysm had not yet ruptured and NYU’s physicians were so advised. However, based upon the initial information that the aneurysm already had ruptured, Ms. Sawtelle had been scheduled for surgery on the following Monday — July 29. At that time (1985), the timing of surgery for a cerebral aneurysm was based on whether the aneurysm had ruptured. If the aneurysm had ruptured, surgery was not immediately performed because of the concern that cerebral vasospasm could result. If, on the other hand, the aneurysm had not yet ruptured, immediate surgery was indicated to attempt to prevent a rupture.
As Ms. Sawtelle was being anesthetized prior to her scheduled surgery on July 29, her aneurysm ruptured. It was Ms. Sawtelle’s claim that surgery should have been immediately performed at NYU and, had that been done, she would not have suffered the rupture of the aneurysm.
Ms. Sawtelle was left with significant neurological deficits. At the time of the trial, she required a wheelchair when leaving her home. She needed support to walk, either from another person or, for short distances, by holding onto furniture or counters. In addition, Ms. Sawtelle suffered speech problems that were attributed to the medications she needed as a result of her stroke.
A Suffolk County, NY, jury returned a verdict on July 9, 1999, of $8,468,400, which consisted of $3 million for past pain and suffering, $5 million for future pain and suffering, $234,200 for past lost earnings, and $234,200 for future lost earnings.
Commentary: The most common recurrent risk management theme in ED Legal Letter is that every return visit should be viewed as an opportunity to correct a possible previous error. To do this, it is absolutely necessary that you always take a fresh look at the patient. That is, the physician must not simply rely upon the evaluation and conclusions of the previous physician, nor allow the re-evaluation to be focused too narrowly because of the previous diagnosis. It is critical to never let a previous physician’s evaluation and diagnosis, no matter how good you think that physician is, to cause you to "put blinders on" and not consider possibilities that the previous physician could have failed to consider.
In this case, the jury concluded that there was no negligence associated with Ms. Sawtelle’s first visit to the ED on July 14. The jury found, however, that Southside Hospital was negligent at the time of the second visit (July 18) in that its ED physician had failed to elicit an adequate history from the patient and that this was a proximate cause of Ms. Sawtelle’s damages. All defendants were found to share liability for the nearly two-week delay in Ms. Sawtelle’s surgery.
Anatomy and Epidemiology
A very brief review of the anatomy and epidemiology of SAH is in order before proceeding. SAH is defined as bleeding into the subarachnoid space of the central nervous system. Excluding head trauma, ruptured intracerebral aneurysm is the most common cause of SAH. Generally, when the term "subarachnoid hemorrhage" is used without any mention of trauma, it is assumed that the discussion is of nontraumatic SAH. Most aneurysms that cause SAH arise from arteries located in the circle of Willis and its branches, most commonly arising where vessels bifurcate. Eighty-five percent of intracerebral aneurysms arise from the anterior circulation, with the remaining 15% arising from the posterior circulation.4
It is not uncommon for patients to have multiple intracerebral aneurysms. In fact, at the time of initial presentation, approximately 25% of patients are found to have multiple aneurysms.5 Most commonly, aneurysms form in areas of vessels that have deficient medial layers associated with thin or absent internal elastic lamina.6
Hemodynamic stress likely plays a role in the formation of aneurysms. As with any other aneurysm, once an intracerebral aneurysm forms, LaPlace’s Law governs its physics and provides that the tension on the wall of the aneurysm is a function of the radius of the aneurysm and the pressure gradient across its wall. Therefore, as the size of the aneurysm increases, and/or the arterial pressure increases, the tension on the wall of the aneurysm will increase. As a result, the larger an aneurysm is, and the higher the patient’s blood pressure is, the more likely it is that the aneurysm will rupture.
Intracerebral aneurysms are not uncommon, although their exact incidence is not known and estimates of their incidence vary based upon the particular epidemiological technique utilized to make the estimate. Autopsy studies have projected an incidence of approximately 6% of the population.7 Many of the aneurysms included in these studies were quite small. Cerebral angiographic studies, on the other hand, have estimated an incidence of 0.5%-1%.8 The actual incidence likely falls somewhere between these two estimates. The annual incidence of SAH in the United States is approximately 30,000.9 Many people have small aneurysms that never rupture and remain asymptomatic throughout their lives, and the overall risk of rupture is approximately 0.5%-2% per year.
In the July 20, 1998, issue of Emergency Medicine Reports, the author, Jonathan A. Edlow, MD, provides an excellent summary of the epidemiology of SAH:
SAH can occur at any age, but it is rare in children, and the mean age at the time of presentation is about 50 years. Overall, there is a female:male ratio of 3:2; however, below age 40, men are more often afflicted than women. All ethnic groups are affected. However, first-degree relatives (especially siblings) of patients who have had SAH have a fourfold increased risk compared to the general population of developing SAH. Patients with a prior SAH are six times more likely to have a second hemorrhage later in life.10
SAH accounts for 8% of strokes, but the incidence of SAH is more heavily weighted toward a younger population than are thrombotic strokes.
Patients with SAH present with a range of signs and symptoms depending, in part, on the location of their aneurysm, but probably more importantly on the severity of the bleeding. Typically, a patient presents complaining of an excruciatingly severe headache, classically described as the worst headache of his or her life. The headache may be associated with nausea and vomiting, diaphoresis, and pallor. The patient may develop meningeal irritation and complain of neck pain and/or stiffness. A wide variety of neurological signs and symptoms may be present, depending upon the location of the aneurysm.
As many as 60% of patients who suffer an SAH will have a warning bleed prior to the full-blown rupture of the aneurysm. These warning, or "sentinel bleeds," can occur anytime from one day to months prior to the actual rupture of the aneurysm (mean interval is 14 days). These headaches typically are sudden and severe and often resolve spontaneously. It generally is believed that they represent a very small leak of blood from the aneurysm. It is, of course, important to identify the cause of these headaches, if possible, as prompt neurosurgical intervention often will prevent a full-blown SAH. While the neurological examination usually is normal at the time of a sentinel bleed, the cerebrospinal fluid almost always will show blood.
Hyatt et vir v. Raggio.11 According to the plaintiff’s version of the facts, Janis Hyatt, after experiencing headaches earlier in the year, sustained a new onset of headaches on Sept. 14, 1992. According to Mr. Hyatt, on Sept. 23, his wife "experienced a change in the intensity of the headaches and that she walked from the bathroom in their home, grabbed her head, asked him to turn the lights out, turn off the television, and leave her alone."12 He did so, and she fell asleep.
According to Mr. Hyatt’s version of the facts, he called Dr. John Raggio, a neurosurgeon who previously had treated him for back problems, and reported that his wife "was having terrible, pounding head-aches, that they had intensified that day, and that she was nauseated."13 Mr. Hyatt made an appointment for his wife to see Dr. Raggio on Sept. 29. Dr. Raggio, on the other hand, testified that he had no recollection of that telephone call and that his records did not indicate that such a call had been received.
Mr. Hyatt testified that on Sept. 28, one day prior to his wife’s scheduled appointment with Dr. Raggio, he again telephoned Dr. Raggio and told him that his wife was "complaining of intense, throbbing pain in her head and that he was taking her to the emergency room."14 It was his testimony that Dr. Raggio told him that he would "telephone ahead and order X-rays."15 According to the court, "[t]he record indicates that Mrs. Hyatt was examined at the hospital and that Dr. Raggio ordered a CT scan with and without contrast, which was determined to be negative."16 It appears that there was no ED physician directly involved in this case, although it seems reasonable to assume that any hospital large enough to have a neurosurgeon on staff would be likely to have an ED staffed by physicians. In any case, the ED record of Mrs. Hyatt’s visit contained the following notation by the nurse:
Headache — onset x2 weeks. Became severe last Wednesday (5 days ago) c/o photophobia, sounds bothering. States movement [arrow up] pain. Denies blurred vision, tunnel vision. States pain feels like pounding to forehead & top of head. AAO [alert, awake, and oriented]. Speech clear. MAE +/= [moves all extremities strongly, equally]. No distress noted.17
After examination, Dr. Raggio diagnosed Mrs. Hyatt’s headache as being related to tension. He ordered an injection of meperidine and gave her a prescription for Limbitrol. He then discharged Mrs. Hyatt. Mrs. Hyatt failed to keep her earlier scheduled appointment for the following day, although it was not clear whether Mrs. Hyatt had been instructed by Dr. Raggio at the time of her discharge to keep this appointment or not. According to Mr. Hyatt, his wife improved over the next couple of weeks.
On the morning of Oct. 18, 1992, Mr. Hyatt awoke to "find the bed shaking and his wife in a tremor."18 She then "became ill in bed and, grabbing her head, [and] stated that she felt that her head was exploding."19 She shortly thereafter collapsed and stopped breathing. Mr. Hyatt testified that he performed CPR while his daughter called the paramedics. Mrs. Hyatt was taken to the ED where Dr. Raggio examined her, ordered an arteriogram, and then diagnosed a ruptured aneurysm.
As a result of the SAH, Mrs. Hyatt was left in a totally helpless and dependent state, unable to walk, talk, feed, or otherwise care for herself. Mr. Hyatt subsequently brought a malpractice action on behalf of himself and his wife alleging that Dr. Raggio had failed to properly diagnose or perform proper diagnostic testing prior to the rupture of the cerebral aneurysm.
Commentary: Before revealing the outcome of this case, let us consider the standard of care issues that were involved. First, there were the events of Sept. 23. In retrospect, we know that Mrs. Hyatt very likely was suffering from a sentinel bleed at that time and an LP, if not a CT scan, would very likely have provided a diagnosis. There was, however, as the court noted, no record establishing that Mr. Hyatt had telephoned Dr. Raggio on Sept. 23 and given him the history as he testified. Interestingly, if the call had actually taken place as testified to by Mr. Hyatt, Dr. Raggio benefited from his own poor medical record documentation. Because there was no record of the call, the jury could believe Mr. Hyatt, or conclude that no such conversation ever took place. Had there been proof that the conversation had taken place as testified to by Mr. Hyatt, it would have been reasonable for a jury to conclude that Dr. Raggio’s response to the call (an appointment six days later) breached the standard of care.
More important were the events of Sept. 28. There was a record of these events, and at least some facts could not be contested. Clearly, Mrs. Hyatt had a headache sufficient in severity for her to go to the ED, and there was no evidence that she had ever done so before. The ED record indicated this was a headache of two weeks’ duration that had worsened five days previously, and that the headache was associated with photophobia and hyperacusis. In addition, a CT scan had been ordered — but for what reason?
Plaintiffs provided expert testimony that the ED record documented sufficient history to make SAH a part of the differential diagnosis and that Dr. Raggio deviated from the standard of care in failing to perform an LP after the CT scan was negative. Plaintiff’s neurosurgery expert testified that "the only reason a CT scan would have been ordered, as it was by Dr. Raggio, was to rule out a subarachnoid hemorrhage, and a negative CT scan alone, will not rule out the possibility."20 It was plaintiff’s expert’s testimony that, when the CT scan was normal, an LP should have been performed. This opinion was supported by the testimony of plaintiff’s pathology expert and her treating physicians.
There is no disputing the fact that plaintiff’s experts were correct that when ruling out an SAH, a negative CT scan should be followed by an LP. Therefore, would it not be true that since Dr. Raggio did not perform a lumbar puncture following the negative CT scan, he must have breached the standard of care? Yes, but only if the CT scan had been ordered to rule out SAH. Interestingly, there was no mention in the court’s opinion as to what the CT order or requisition indicated was Dr. Raggio’s reason for ordering the test. Why did he order a CT scan for this patient who presented with symptoms that were, if not entirely classic, at least suggestive of SAH? Dr. Raggio testified, "the CT scan was ordered to rule out an expanding lesion or tumor."21 If that were the case, then it would not follow logically that an LP must have followed the negative CT scan. We are, of course, considering here only the narrow issue of whether an LP must follow a negative CT scan, not the more general issue of whether it was appropriate for Dr. Raggio not to rule out SAH.
So, how should this case have been decided? The jury apparently found Dr. Raggio’s differential, which did not include SAH, to have been reasonable and believed his testimony as to why he had ordered a CT scan. As a result, the jury returned a verdict in favor of Dr. Raggio. However, the trial court judge granted Plaintiff’s motion for JNOV (judgment non obstante veredicto — judgment notwithstanding verdict) and overturned the jury’s verdict, stating: "I think that the evidence is overwhelming as to liability in this case."22 On appeal, the Court of Appeal of Louisiana reversed the trial court and reinstated the jury verdict for the defendant. Finally, the dissenting judge on the appellate panel disagreed with the reversal, stating: "I believe that the facts point strongly and overwhelmingly in the plaintiff’s favor . . . ."23
This was clearly a very close case but, I believe, would have been an easy case if it could have been shown that Dr. Raggio ordered the CT scan for the purpose of ruling out SAH. If that could have been established, the plaintiff should have prevailed.
There are three key risk management lessons to be learned from this case. First, if SAH is on the differential and a CT scan is ordered for the purpose of ruling out SAH, a negative CT scan must be followed by an LP, unless the patient refuses after being properly informed of the risks and benefits of the procedure. Second, it is crucial to make sure that X-ray requisitions are completed properly. This is a task generally left to clerks and ignored by physicians. Unless told otherwise, clerks routinely may write, "rule out bleed" on all head CT requisitions originating from the ED. If in error, this absolutely could defeat the defense of an otherwise defensible case. If a CT is ordered for a reason other than "rule out bleed," it is crucial to be sure that the requisition properly indicates this (e.g., "rule out mass"). If the plaintiff in Hyatt had been able to produce a requisition or order that contradicted Dr. Raggio’s testimony as to the reason he ordered the CT scan, the result in the case might very well have been different. Not only would it have established that the reason for the CT scan was to rule out SAH (and therefore, an LP should have followed the negative CT scan), it likely would have seriously damaged Dr. Raggio’s credibility if he testified that he had not ordered the CT for that purpose.
Finally, this demonstrates an area where a computerized medical record system could prevent a possible disaster. If the computerized medical record system is set up such that any time a CT is ordered for "rule out bleed" or "rule out SAH," and the CT is negative, a prompt appears that prevents the patient’s discharge without an LP (e.g., when LP lab requisitions are printed, clicking on the discharge button prompts an LP reminder). As a result, cases in which an LP is not done after a negative CT to rule out SAH could be eliminated. This LP prompt would have to be manually overridden if, for example, the patient refused the procedure.
The CT scan is a very sensitive test for SAH. It is not infallible, however. In the first 12 hours after an SAH, CT’s sensitivity approaches 100%, but the sensitivity drops to 93%-95% by 24 hours.24 There is continued progressive loss of sensitivity as the interval between the SAH and the CT scan increases past one day.25 Other factors also can limit the sensitivity of CT scanning — blood with a hemoglobin of less than 10 g/dL may be poorly visualized and small amounts of blood can be missed. As a result, the vast majority (if not all) of patients who undergo a CT scan to rule out SAH should have an LP if the CT is negative.
CT Before LP
American Universal Insurance Co. v. The Medical Malpractice Joint Underwriting Association of Massachusetts.26 This case primarily involved a claim by American Universal Insurance Co. (AU), an "excess coverage" insurer (i.e., a company that provides insurance for losses exceeding the limits of a policyholder’s primary coverage), that the defendant primary insurer, The Medical Malpractice Joint Underwriting Association of Massachusetts (JUA), was liable for bad-faith insurance practices that damaged AU. The case is interesting for two reasons. First, the underlying medical malpractice case involved a patient with SAH. Second, the case report documents in remarkable detail (because of the bad-faith insurance practices issue) the defense assessments of the case as the case progressed from filing to eventual settlement. It is very interesting to follow the progression of these assessments and see their relationship to the ultimate settlement.
On Sept. 14, 1983, "shortly before 3 o’clock in the afternoon," 43-year-old June LoCicero was brought by ambulance to the ED at St. Joseph’s Hospital in Lowell, MA (St. Joseph’s). She was complaining of a severe headache accompanied by nausea and vomiting. Ms. LoCicero was seen in the ED by Dr. Charles Rappaport, the ED physician on duty at the time.
Dr. Rappaport performed a "brief physical and neurological examination of Ms. LoCicero." He determined on examination that she exhibited meningeal signs. His differential at that time consisted of meningitis and SAH, with SAH being more likely in his opinion. Further examination by Dr. Rappaport of Ms. LoCicero did not reveal any signs of increased intracranial pressure or any "focal signs."
At that time, St. Joseph’s did not have a CT scanner; when a hospital patient needed a CT scan, arrangements were made with nearby facilities. The nearest CT scanner to St. Joseph’s was at Merrimack Valley CT Scanning, a facility located in a professional office building across the street from the hospital and connected to the hospital by an elevated walkway over the street.
It was Dr. Rappaport’s position in the subsequent malpractice action that it had been his experience that if he did not get word immediately that the Merrimack Valley CT Scanning facility was available to perform an emergency scan, there would then be a significant delay in making the necessary arrangements for a CT scan. When that was the case, he would not wait before investigating for meningitis. That is, he would not wait for the CT scan, but would proceed with an LP.
When Dr. Rappaport was not immediately told that the CT scan could be performed promptly at the Merrimack Valley CT Scanning facility, he apparently assumed (based upon his past experiences) that there would be a significant delay in obtaining a CT scan and proceeded with a LP at "about 3 o’clock." The precise time that the CT scan was ordered was not reported, but it appears that it could not have been more than a matter of minutes between the time Dr. Rappaport ordered the scan and when he performed the LP. The LP showed that there was blood in the cerebrospinal fluid consistent with an SAH.
Shortly after the LP was done, Dr. Rappaport learned that the Merrimack Valley CT Scanning facility was ready and an emergency scan could be performed. Unfortunately, at "just about that same time, he was also notified by a nurse that Ms. LoCicero had begun seizing and was unconscious."26 Ms. LoCicero was then transported across the street to the Merrimack Valley CT Scanning facility for a CT scan, which was performed at about 3:30 p.m. The CT revealed a massive intracerebral, as opposed to subarachnoid, hemorrhage with a "massive hematoma or clot on the left side of her brain."28
Dr. Lowell Rosman, a neurosurgeon, performed emergency surgery to evacuate the clot. He was of the opinion that Ms. LoCicero had experienced two separate episodes of bleeding, one before she came to the hospital, and one after the lumbar puncture had been done. He did not believe that the LP had caused or precipitated the second episode of bleeding. Plaintiff’s expert, of course, had a quite different opinion.
In the subsequent malpractice case, the LoCiceros alleged that the hemorrhage in Mrs. LoCicero’s brain itself did relatively little neurological damage. They claimed that it was the lumbar puncture that resulted in the more serious damage because it resulted in herniation of her brainstem and the catastrophic neurological damage that she ultimately suffered. It was their claim that Dr. Rappaport breached the standard of care by not obtaining a CT scan before deciding to do the LP. Had he waited for the CT scan, plaintiffs claimed that an LP would not have been done, there would have been no brainstem herniation, and Ms. LoCicero’s neurological deficits would not have been as great as they were.
Defense of the case was based on contesting both the negligence and causation elements of plaintiff’s case. As to the negligence element, it was the defense’s position that Dr. Rappaport was not negligent in performing the LP before obtaining a CT scan because he reasonably thought that a CT scan was unavailable. As to the causation element, it was the defense’s position that the LP did not cause Ms. LoCicero’s injuries, which were, according to the defense, all the primary result of her intracerebral hemorrhage.
The defense had the case, which was filed in July 1985, reviewed by numerous experts. As to the negligence issue, the court in the later insurance litigation noted, "the experts were not entirely enthusiastic about the course that Dr. Rappaport had taken."29 On the other hand, the experts were concordant and more emphatic that it was likely that the damages suffered by Ms. LoCicero were due to the hemorrhage and not related to the LP.
The initial assessment of Dr. Rappaport’s defense counsel, after he had obtained several expert reviews, was that the case could be successfully defended. He expressed the following opinions to the insurance carrier: "the case was not one of liability; it should not be settled at this time; the chance of winning at trial was 50%, depending on plaintiff’s medical expert’; and that he estimated the possible verdict range to be $500,000-$750,000, due to lifetime of institutional care.’"30 (Editor’s note: Do you agree?)
About a year later, defense counsel again reported to the insurance carrier. This time, he expressed an opinion that the case was "probably not" a case of liability (a year earlier he had said that it was "not one of liability"). He continued to advise against settlement and, while he continued to estimate the chance of prevailing at trial at 50%, he now estimated the verdict range if the case were tried and lost at $1 million to $6 million, a considerable change from his earlier estimate. This estimated verdict range clearly implicated the "excess coverage" carrier, as the primary insurance was for $1 million per claim. The "excess coverage" had a limit of $5 million.
In August 1987, the plaintiffs proposed that the case be settled for $6 million, an amount that was, probably not coincidentally, the aggregate amount of Dr. Rappaport’s insurance coverage. (Editor’s note: Would you settle? If yes, how much would you offer?) Defense counsel then submitted a third summary assessment of the case to JUA, the primary insurer. In this report, defense counsel, for the first time, expressed an opinion that the case should be settled. He recommended that a settlement offer of $500,000 be made and estimated that there was a 20% chance (down from 50%) of prevailing at trial. His estimate of the probable verdict range was now $5 million to $6 million, essentially the high side of his earlier estimate. This report apparently reflected his opinion that there was danger of a considerable verdict, though not a verdict so high as to justify meeting the plaintiff’s settlement demand. At a conciliation conference in September 1987, the defendants offered $100,000 to settle the case.
In February 1988, a neurosurgeon board member of JUA, Dr. Ronald Birkenfeld, reviewed the case and expressed to defense counsel his opinion that Ms. LoCicero initially had suffered an SAH, which presented no reason for refraining from performing a lumbar puncture, and that after the puncture, but not because of it, she had a further intracerebral hemorrhage. He was of the opinion that if she had had a mass or clot in the left side of her brain before the lumbar puncture was done, as the plaintiffs were asserting, she would have exhibited certain clinical signs, which she did not. This physician would not, of course, be providing official expert testimony in the case. After receiving this opinion, defense counsel now wrote to the JUA claims representative advising, "this is a defensible case."31 He reported an "excellent chance that our experts will convince the jury, both on liability and causation so that the jury will return its verdict for the defendants."32
Understandably, the JUA claims representative was confused. She wrote to defense counsel that: "I am extremely confused and dismayed at the letter regarding the defense of this case in light of the fact that on Aug. 24, 1987, you found that the case had a 20% chance to win, and now you feel the case is defendable. I am wondering if the change in your position is a result of our discussion before Dr. Ronald Birkenfeld [the neurosurgeon board member], and why, after this discussion, you have changed your percent chance to win in light of the fact that we continue to have the same expert opinions that we have had all along."33
Just before trial, defense counsel sent a fourth report to JUA. In this report, he made the same estimates regarding the prospects of winning the case and the probable verdict as he made in his August 1987 report. He continued to recommend a settlement offer of $500,000, but recommended that the case be tried "unless plaintiff’s ideas and demand are drastically reduced," because the "demand is excessive considering defendant’s chances of a compromise jury verdict."34
At the suggestion of the defense counsel assigned to him by the JUA (which he, incidentally, had chosen), Dr. Rappaport retained separate counsel to represent his personal interests because of the potential of a verdict exceeding his aggregate insurance coverage. In December 1987, Dr. Rappaport’s personal attorney notified both insurance carriers (JUA and AU) that Dr. Rappaport "authorized" them to settle the case for the $6 million amount that represented the aggregate limits of his policies. In April 1988, the plaintiffs made a new settlement demand of $5.45 million.
Not having reached a pretrial settlement, the case went to trial in September 1988. During the trial, settlement discussions continued without success — JUA, the primary insurer, had increased its settlement offer from $100,000 to $300,000, but the plaintiff’s demand remained at $5.45 million.
The defense suffered an early setback when the judge, in response to a plaintiffs’ motion, ruled that Dr. Rosman, the neurosurgeon who had operated on Ms. LoCicero, could only testify as a "fact witness" (i.e., testify only to the facts surrounding his treatment of Ms. LoCicero and not give any opinions) and not as an expert witness. The judge ruled that, under Massachusetts law, absent consent from Ms. LoCicero (which she, of course, would not give), he could not offer expert opinion. This was a significant setback for the defense, as Dr. Rosman was prepared to support the defense case. In what was shaping up as a "battle of the experts," it would have been of very substantial benefit to have the treating physician, the most objective expert in the case, side with the defense.
As could have been predicted, the question of whether a CT scan was readily available at the Merrimack Valley CT Scanning facility at the time Dr. Rappaport performed the lumbar puncture on Ms. LoCicero was central in the case. The plaintiffs focused on "actual availability" — the Merrimack Valley CT Scanning facility was actually available to perform a CT scan at the time Dr. Rappaport performed the LP. The defense’s response was to focus on "perceived availability." According to the defense, the more relevant fact was that, at the time he elected to proceed with the LP, Dr. Rappaport reasonably perceived that a CT scan was not available.
As this issue developed, the attorney for AU, the excess coverage carrier, who was observing at the trial, was of the opinion that plaintiffs had scored heavily on the CT availability issue, and that odds of the defense prevailing in the case were slight: "A verdict for the plaintiff is not only likely, but must be expected" since the testimony that the CT scan was in fact available "destroyed Dr. Rappaport’s defense."34 At this point, AU, the excess coverage carrier, demanded that JUA, the primary insurer, tender its policy (i.e., pay it over to AU) and allow AU to negotiate a settlement with the plaintiff. It was AU’s clear assessment that a verdict for the plaintiff would be returned if the case reached the jury.
On Oct. 5, 1988 (over five years after the events of the case and over three years after the case was filed), the jury returned a verdict awarding Ms. LoCicero damages in the amount of $6,110,953 plus interest in the amount of $2,340,494.90, and awarded Mr. LoCicero $1 million plus interest in the amount of $383,000 on his consortium claim. In November 1988, a settlement was reached in which the plaintiffs settled all their claims against Dr. Rappaport in exchange for a settlement package valued at $6 million paid by the insurers. In return for this settlement, the plaintiffs’ received Dr. Rappaport’s waiver of his right to appeal the case (i.e., a certain $6 million, instead of approximately $10 million which would be subject to appeal and ability to collect the excess from Dr. Rappaport personally). As a result of the settlement, Dr. Rappaport ultimately did not pay any money personally. Following the settlement, AU, the excess coverage insurer, sued JUA, the primary insurer, for bad-faith insurance practices in refusing to tender its policy to AU to allow it to negotiate a pretrial settlement.
Commentary: Was it within the standard of care for Dr. Rappaport to have performed the LP when he did? Not according to this jury. Prompt diagnostic intervention, as a general rule, is preferred. In this case, however, the jury apparently concluded that Dr. Rappaport made a precipitous decision to proceed without the CT scan — it seems that Dr. Rappaport waited only minutes after ordering the CT scan before concluding, without being told one way or the other, that a CT scan would not be immediately available. While there was conflicting testimony as to whether the LP caused Ms. LoCicero’s injuries, the jury likely felt that Dr. Rappaport had clearly breached the standard of care in not waiting to ascertain the availability of a CT scan before performing the LP. Having reached this conclusion, the jury probably had little interest in the causation defense.
There is certainly no question that minutes may well count in treating meningitis. If, however, it was reasonable to suspect that Ms. LoCicero might have bacterial meningitis, Dr. Rappaport might have initiated intravenous broad spectrum antibiotic treatment with appropriate meningitis coverage and then waited for CT scan availability. Had he done so, Ms. LoCicero’s care would not have suffered, even if she had bacterial meningitis, and this $6 million case would never have been brought.
It is unusual that a case report provides such detail of the chronological development of a defense attorney’s evaluation of a case. The lesson from this case is that an initial defense assessment (50% chance of success with a potential verdict of $500K-$750K in the LoCicero case) and the ultimate verdict (approximately $10 million) or settlement ($6 million) may be quite different. While Dr. Rappaport did not end up paying anything personally in this case, he certainly might have. It is, therefore, critical that a malpractice defendant monitor the progress of his or her case and attempt to be as objective as possible to avoid, in particular, an overly optimistic view of the case by defense counsel. As bad as it might be to settle a case for the policy limit, it is worse to take the case to trial and be on the wrong end of a verdict in excess of the policy limit.
It is routine to include in all oral board review courses a patient with incomplete vital signs (the patient’s temperature being the missing vital sign in most cases). In fact, such a case commonly is presented during the actual examination. Making sure that you always have a complete set of vital signs can be a pass-fail issue during the oral board examination; in the ED, however, it can be a life-and-death issue. Every ED physician must be sure that every patient has a complete set of vital signs and that any abnormal vital sign is appropriately responded to. Obviously, if the patient has a fever, the focus of suspicion will shift toward meningitis, as opposed to SAH.
Pay particular attention to the blood pressure. Although it is not necessary for the blood pressure to be elevated for a patient to suffer an SAH, hypertension will increase the likelihood that a cerebral aneurysm will rupture. Be very hesitant to ascribe a hypertensive patient’s headache to his or her hypertension. Hypertension alone rarely causes a headache.
All headache patients need a "complete" physical examination. (See "Evaluation of Head Pain Patients in the ED," at the end of this article.) That does not mean every headache patient needs an admission-type history and physical, or even a comprehensive neurological examination. What it does mean is that every headache patient needs an examination that at least briefly addresses all systems. The focus naturally will be on the neurological examination, which need not be comprehensive in most cases, but must cover all the basic components of a neurological examination (mental status, cranial nerves, sensory, motor, and reflexes).
Many patients with SAH (but certainly not all) will present with meningeal signs. Retrospective review of cases where SAH was missed shows a high incidence of "acute cervical strain" or other similar diagnoses as the discharge diagnoses. It certainly is true that an acute cervical strain may result in cephalgia. The first consideration, however, when a patient complains of head and neck pain (particularly of sudden onset) associated with a complaint of neck stiffness and/or neck stiffness on examination, must be SAH.
It is always a good idea to include evaluation of gait in any neurological screening examination. I have a vivid recollection of a young woman who presented with the acute onset of headache who seemed to have a perfectly normal neurological examination (including a perfectly supple neck) until I asked her to walk. When walking, she consistently seemed to list to one side. Her CT scan showed a large SAH. Gait is an excellent final discharge screening test (not just with headache patients), and it is always a good idea to watch your ambulatory patients walk before you discharge them.
Early diagnosis of SAH is critical to catch the patient at the sentinel bleed stage, if possible, before full-blown rupture of the aneurysm has occurred. Patients who present early in the SAH process and are most likely to benefit from prompt treatment (i.e., those who have only suffered a minor sentinel bleed) will likely have the least serious signs and symptoms and will often lack the "classic" symptoms of full-blown SAH (sudden, worst headache). These patients are the ones most often misdiagnosed and must, therefore, be a focus of risk management attention.
This situation is somewhat analogous to another major vascular emergency which commonly presents to the ED: acute coronary syndrome. When dealing with the diagnosis of acute coronary syndrome, ED physicians sometimes are focused too narrowly on AMI. To be sure, it is critical to have a very high index of suspicion for AMI. However, it is critical to identify acute coronary syndrome patients at the unstable angina phase, if possible, when prognosis will be best. Similarly, patients with cerebral aneurysm must, when possible, be identified at the sentinel bleed stage, before a full-blown rupture has occurred.
SAH is one of the "must-never-miss" diagnoses in emergency medicine. Studies have shown, however, that the rate of delayed diagnosis of SAH may be as high as 51%.36 Both consequences to the patient and the potential damages in a subsequent malpractice lawsuit will be huge. Despite the fact that many thousands of patients are seen in EDs annually with the complaint of headache, and the vast majority of them are not suffering from a serious problem, all headache patients must be treated seriously. The most useful strategies to avoid missing an SAH are to: 1) not exclude SAH from the differential simply because the patient does not have the "classic" presentation; 2) never treat a chronic headache "frequent flyer" any differently than you would any other patient when considering the likelihood of SAH; and 3) always follow a negative CT scan obtained for the purpose of ruling out SAH with a lumbar puncture. Other risk management strategies for avoiding SAH malpractice exposure are listed in Table 1, below.
|Table 1: Risk Management Strategies for Headache|
|1.||Take a careful history and, if the patient is complaining of either a) his or her first episode of significant headache; or b) if the patient has a history of previous headaches, a headache that is in any way different than his or her usual headache, consider subanachnoid hemorrhage (SAH) and rule it out unless there is a convincing reason (that you document in the record) not to.|
|2.||If you do a CT scan to rule out SAH and it is negative, do a lumbar puncture (LP) unless the patient refuses after a proper informed consent).|
|3.||When doing CT scans and LPs to rule out SAH, expect the incidence of positive tests to be quite low. This is expected and necessary in dealing with a diagnosis that neither you nor the patient can afford to miss.|
|4.||Develop a system (computer or otherwise) that will not allow patients to be discharged after having a negative CT scan to rule out SAH until they have had an LP or the treating physician has documented the patient’s refusal or another appropriate reason not to do the test.|
|5.||Do not rule out SAH based simply upon the fact that the onset of the headache was not sudden. Many patients with SAH will not exhibit the "classic presentation."|
|6.||Make a diagnosis of "viral syndrome" or "tension headache" in a patient who has come to the ED because of head pain with care, and only after considering more serious causes (SAH, meningitis, etc.).|
|7.||Never attribute a hypertensive patient’s headache to his or her blood pressure alone without considering SAH. Hypertension alone, even severe hypertension, rarely causes headache, and headache associated with hyper- tensive encephalopthy should resolve promptly with lowering of the blood pressure. Remember that hyperten- sion is a risk factor for SAH.|
|8.||Never dismiss the possibility of SAH based solely upon a patient’s response to analgesia. Some headaches associated with SAH resolve with ketorolac.|
|9.||Be very careful in concluding that a bloody tap is because of a "traumatic tap" when you had significant pre-LP suspicion of SAH.|
|10.||Maintain an increased level of suspicion of SAH when headache is associated with cocaine abuse.|
|11.||If meningitis and SAH both are being considered in the differential, a CT ordinarily still should precede an LP. The patient should be treated promptly, however, with appropriate intravenous antibiotics prior to being sent for a CT scan.|
1. Dhopesh V, Anwar R, Herring C. A retrospective assessment of emergency department patients with complaints of headache. Headache 1979; 10:37-42.
2. Leicht M. Nontraumatic headache in the emergency department. Ann Emerg Med 1980; 9(8):404-409.
3. Executive Edition, The New York Jury Verdict Reporter. Web: execed.newsday.com/database/jury0907.htm (visited 4/3/01).
4. Schievink WI. Intracranial aneurysms (see comments). N Engl J Med 1997; 336:28-40.
10. Edlow, J. Diagnosis of subarachnoid hemorrhage: Avoiding pitfalls and expediting assessment. Emerg Med Rep 1998; 19:154-163.
11. 757 So.2d 773 (La.Ct.App. 2000).
12. Id. at 774.
13. Id. at 775.
20. Id. at 779.
22. Id. at 778.
23. Id. at 782.
24. Sames TA, et al. Sensitivity of new-generation computed tomography in subarachnoid hemorrhage. Acad Emerg Med 1996; 3:16-20.
25. Vermeulen M, van Gijn J. The diagnosis of subarachnoid hemorrhage. J Neurol Neurosurg Psychiatry 1990; 53:365-372.
26. 1993 Mass. Super. LEXIS 149.
27. Id. at *16.
28. Id. at *17.
29. Id. at *21.
30. Id. at *29.
31. Id. at *30.
33. Id. at *30-31.
34. Id. at *33.
35. Id. at *51-52.
36. Neil-Dwyer G, Lang D. Brain attack’ — aneurysmal subarachnoid hemorrhage: Death due to delayed diagnosis (see comments). J Royal Coll Physicians Lond 1997; 31:49-52.
By David L. Freedman, MD, JD, FAAEM, Emergency Physician; Attorney, Miller, Canfield, Paddock & Stone PLC, Ann Arbor, MI
In evaluating the many ED patients who complain of head pain, whether or not head pain is the chief complaint, it is essential to identify those patients who are at risk for the two principal, treatable emergency conditions associated with headache that must not be missed: subarachnoid hemorrhage (SAH) and meningitis. Our concern here is with SAH, although the strategy applies to meningitis as well (although if uncomplicated meningitis is the only consideration, a CT scan will generally not be necessary). The first and most important step in identifying those patients who may have SAH is to divide headache patients into three groups, based upon a focused, but careful history:
• Group 1: Patients without a history of head pain who are presenting with a first (or very unusual) episode of significant head pain;
• Group 2: Patients with a history of chronic head pain, but whose headache is somehow different from their usual head pain;
• Group 3: Patients with a history of chronic head pain who are complaining of another episode of their usual head pain.
Obviously, it is the first two of these groups with which ED physicians must be most careful.
Classification of patients into these three basic risk groups may be accomplished by asking each headache patient several basic questions:
1. Have you had headaches before, other than minor headaches that were promptly relieved by a couple of Tylenol, aspirin, or the like? If the answer is no, the patient must be placed in Group 1. If the answer is yes (the patient has a history of headaches that he or she considers to have been significant), proceed to the next question to decide whether the patient belongs in Group 2 or Group 3.
2. Have you had headaches as bad or worse than this one before? If the answer is no, the patient must be placed in Group 2. If the answer is yes (the patient previously has had headaches as bad or worse than the current headache), proceed to the next question. It is important not to lead the patient when taking this history.
3. Is there anything unusual or unique about this headache; any complaint (sensitivity to light, location of the pain, rapidity of onset, etc.) that is present with this headache that routinely is not present when you suffer a headache? If the answer is yes, further questions to determine the significance of the unique symptom will be necessary and, unless there is a convincing reason to do otherwise, the patient should be placed in Group 2. If the answer is no, follow up with one additional question: So this headache is exactly the same as headaches you have frequently had, and you are here simply to get something for pain as you have done in the past? If the answer is yes to the follow-up question, the patient may be placed in Group 3.
While a complete problem-focused history should, of course, ultimately be obtained from every patient, these questions generally are sufficient to place all headache patients in one of the three groups. It is important that this history be well-documented in the medical record.
Over-reliance on the so-called "classic" presentation of SAH (sudden onset of an excruciating, focal thunderclap headache) is undoubtedly the most common cause of the failure to appropriately diagnose SAH. If SAH is considered and appropriately ruled out only in patients who present with the "classic" presentation, the diagnosis will be missed with alarming frequency and disastrous results. One reason that SAH is a significant risk management problem is that the potential damages suffered by the patient can be so devastating. It probably is not a coincidence that an SAH diagnosis is missed more often than it should be because there is a tendency to associate it too closely with its "classic" presentation. SAH is similar, in this sense, with at least two other leading risk management problems in emergency medicine—acute myocardial infarction (AMI) and acute appendicitis. Whenever an emergency condition has a "classic" presentation, there is a tendency to over-rely on this "classic" presentation in making the diagnosis. As with AMI and acute appendicitis, not all patients (in fact, probably not even the majority) with SAH present with the "classic" presentation. Therefore, it is folly to use the "classic" presentation as the reference point for one’s index of suspicion for any of these illnesses.
Evaluation of Group 1 Patients
Group 1 will contain most of the patients who ultimately are found to suffer from SAH. These patients have never had significant headaches before and have presented with a headache sufficiently severe to require them to come to the ED. This is a very significant fact — while headache is a frequent ED complaint, most patients who suffer a headache do not go straight to the ED. Unless there is a compelling reason not to do so, all of these patients require a noncontrast CT scan of the head and, if the CT scan is negative, a lumbar puncture. While, in the rare case, it may be appropriate to proceed to a cerebral angiogram after a negative lumbar puncture, the literature suggests that if both the CT and the lumbar puncture are negative, SAH will have been adequately ruled out.
Evaluation of Group 2 Patients
The evaluation of Group 2 patients really is no different than the evaluation of Group 1 patients. Remember that there is nothing about a chronic history of headache (e.g., monthly migraine headaches) that makes a patient any less likely to suffer an SAH than a person with no previous history of headaches. The major mistake made once a headache patient describes a history of chronic, recurrent headaches is to discount the possible significance of the current episode of headache and assume that the present episode is simply a recurrence of the chronic problem. This is a common feature of many SAH malpractice cases. Each episode of headache should be considered on its own particular merits and appropriately classified as discussed above.
Evaluation of Group 3 Patients
Clearly, the patients in this group are very unlikely to have an SAH. If the patient has an entirely negative-focused examination, including an entirely normal focused neurological examination, the patient can generally be given appropriate analgesia and discharged without a work-up to rule out SAH. Obviously, history is not the only consideration, and any abnormalities on physical examination must be considered and might provide reason for a CT and lumbar puncture.
The History in Head Pain Patients
After obtaining the basic history (i.e., the answers to the questions discussed above that will allow a patient to be assigned to an appropriate risk group), the remainder of a problem-focused history should be obtained, no matter to which group the patient may have been assigned. It is always possible that further history will be obtained that will cause you to change your preliminary assessment of SAH risk and, for example, decide that a Group 3 patient requires a full SAH work-up. The same is true, as mentioned above, with respect to findings during physical examination of the patient. It is crucial not to conclude irrevocably, on the basis of preliminary information (e.g., a history of monthly migraines), that the patient cannot have an SAH.
A general history of the present illness, as well as relevant past medical history, social history, family history (patients with a history of SAH in a first-degree relative are at increased risk for SAH), and at least a focused review of systems should always be obtained. Patients should be asked, at a minimum, about:
It is critical to make sure that, as SAH is ruled out, meningitis also is ruled out. Your history and physical examination should be designed to cover both possibilities. Fortunately, the definitive test for SAH — lumbar puncture — is also the definitive test for meningitis. Finally, do not forget the other causes of acute head pain that require immediate treatment (e.g., acute angle-closure glaucoma).