It is always useful to have a patient’s own account of what a medical or neurologic syndrome is like. It is especially valuable to have a first-person account from such a gifted writer as New York Times reporter Trip Gabriel. In his companion article in this issue of the journal, Mr Gabriel (Gabriel, 2017) describes his experience with transient global amnesia, one of the most dramatic and mysterious syndromes in medicine. Among the many accounts of this syndrome, Mr Gabriel’s, brought on by stressful sailing races and starting to abate as he lay in a magnetic resonance imaging (MRI) scanner about 9 hours later, must qualify as one of the more frightening attacks.
Transient global amnesia is a sudden loss of short-term memory, beginning abruptly and, by definition, ending within a day. Patients typically remember who they are and they recognize family members, but they do not recognize where they are or why they are there, and they typically ask repetitively about “where” and “why.”
Transient global amnesia was first reported by Morris Bender in 1956. The same year, the French authors Guyotat and Courjon described a similar syndrome under the name “les ictus amnésiques” [“amnesic seizures”].
I have adapted these diagnostic criteria from those proposed by Hodges and Warlow (1990):
- The attack must be witnessed by a capable observer.
- Patients must have both retrograde and anterograde amnesia during the attack.
- Patients must stay fully conscious during the attack and must remember who they are.
- Patients must not have focal neurologic signs during the attack.
- Patients must not show evidence of experiencing epileptic seizures during the attack.
- Patients must not have had a recent head injury, an active seizure disorder, or any other known cause of amnesia.
- The attack must resolve within 24 hours.
Going by Mr Gabriel’s account, he met all seven criteria.
Transient global amnesia encompasses both retrograde and anterograde amnesia. At the start of an attack, the retrograde amnesia may extend back weeks, months, or years. This is why, when Mr Gabriel’s wife called him, he knew who he was and who she was, but he could not remember where they lived. During the period of anterograde amnesia, the patient cannot create new memories. Once the anterograde amnesia resolves, the period of the retrograde amnesia shrinks, usually to hours. From then on, the patient has a permanent gap in memory for everything that happened during the attack and perhaps a brief period preceding it.
Mr Gabriel’s total episode lasted about 23 hours. It probably began around 1 PM, when the second sailing race—and the rain—were starting. He began to regain awareness at around 10 PM, when he “awoke” in the MRI scanner, but only about 14 hours later had his anterograde memory returned fully so that he could be discharged from the hospital. Once his anterograde amnesia began to ease and he could start forming new memories again, his period of retrograde amnesia began to shorten. His permanent memory gap extends for 12 hours, from about 10 AM, when he got his haircut, until he became aware of being in the MRI machine at about 10 pm.
As Mr Gabriel reviews the literature, he discusses precipitating factors such as swimming, sexual intercourse, and stress. All of these triggers are cited by Fisher and Adams, who published a brief report on the syndrome in 1958 and a more complete discussion in 1964. In their words, the syndrome could be precipitated by bathing “in the cold waters of the North Atlantic.” They also mentioned the case of a middle-aged man who developed the syndrome while in the company of his mistress.
Mr Gabriel notes that the cause of transient global amnesia is still not understood. We know that it is uncommon. Incidence estimates have been in the range of 20 to 30 per 100,000 population over age 50 (Miller et al, 1987; Zorzon et al, 1995). We also know that the syndrome tends to affect middle-aged and elderly people, who collectively are likely to have stroke risk factors such as hypertension or heart disease. Unlike patients with more typical symptoms of a transient ischemic attack, however, people who have had transient global amnesia face no higher risk of stroke than do patients of similar age who have not had a transient ischemic attack (Hodges and Warlow, 1990; Miller et al, 1987; Shuping et al, 1980; Zorzon et al, 1995).
We know that the phenomenon of amnesia relates to the function of the hippocampi, as Mr Gabriel writes, along with the other structures comprising the Papez (medial limbic) circuit. In transient global amnesia, some factor presumably prevents the hippocampus on one or both sides from functioning for a period of hours, and then reverses to allow full recovery.
The nature of transient global amnesia has fascinated neurologists since the syndrome was first described. We have some understanding of the anterograde amnesia. Because the brain is not laying down or consolidating new short-term memories, patients cannot remember anything that happens from the start of an attack until they recover the ability to form new memories.
The retrograde amnesia is less fully explained. We know that patients lose memories that have not yet “consolidated,” and so we would expect patients to have a brief period of permanent retrograde amnesia for the time just before an attack begins. For example, a football player who suffers a concussion may at first recall how he was hit, but he quickly loses that memory (Yarnell and Lynch, 1970), and later on he will not remember the play that led to his injury.
What we still need to understand is why, during an episode of transient global amnesia, the retrograde amnesia stretches back for months or years, but when the patient starts to build new memories again, the period of loss shrinks. This “shrinking retrograde amnesia” is also seen in patients who have amnesia after traumatic brain injury.
Gordon and Marin (1979) described their detailed examination of a patient in the midst of an episode of transient global amnesia. She could explain complex concepts such as the rules of bridge and she could perform immediate memory tasks such as a digit span, but as soon as rehearsal was prevented, she forgot the digits within 20 to 30 seconds. Her retrograde memory loss included major political events of the previous few years. The authors commented that during the attack she had impaired retrieval of older memories, but those memories were not permanently lost.
Transient global amnesia has a number of “mimics” beyond stroke and the transient ischemic attacks that are the warning symptoms of a stroke. Other imitators include alcohol “blackouts,” effects of hypnotic drugs, partial complex seizures, migraine attacks, head trauma, and even brain tumors.
Three neuroscientists who took the sedative-hypnotic drug triazolam (Halcion®) during transatlantic flights to prevent jet lag lost all memory of the trip (Morris and Estes, 1987). Similar effects have been reported with such other hypnotic drugs as zolpidem (Ambien®). New York Times columnist Maureen Dowd (2006) wrote that after taking zolpidem she had no recollection of eating a jumbo pack of Oreo cookies. The comedian Shelley Berman entitled one of his most famous sketches “The Morning After the Night Before.” Badly hungover, he is speaking on the telephone with the host of the hard-drinking party where he had behaved badly the previous evening. He is surprised to hear that he had thrown the host's cat through a closed window...and then the host's mother.
Memory gaps caused by drugs and alcohol can usually be distinguished from transient global amnesia through the history. Gaps caused by the other imitators may be more challenging to differentiate.
Most patients with transient global amnesia are brought to medical attention because a friend or family member recognizes that they are having a problem. If they are seen by a physician who is unfamiliar with the syndrome, the diagnosis may be missed. If they are taken to a hospital, they will be evaluated for stroke (usually with MRI, a computed tomogram, and/or an electroencephalogram) and for the other mimics. For most patients, the workup is negative and the neurologist can rule out the other conditions and reach a diagnosis of transient global amnesia by exclusion. However, because occasional patients are found to have a stroke or another of the imitator conditions, all patients with sudden-onset amnesia should have an emergency evaluation.
MRI is valuable in the differential diagnosis of an episode of amnesia. In my own experience, most patients with transient global amnesia have normal MRI scans, as did Mr Gabriel. However, several MRI studies have reported lesions in the hippocampus (Alberici et al, 2008; Lee et al, 2009; Strupp et al, 1998; Yang et al, 2008), and one has described frontal lobe abnormalities (Ay et al, 1998). A study with positron emission tomography reported reduced glucose metabolism in the frontal lobe (Eustache et al, 1997).
Two of my patients illustrate the role that MRI can play in evaluating amnesia.
Recently, I treated a neurologist who suffered transient global amnesia triggered by a stressful event. This man was driving his car when it stalled in a busy intersection and he could not restart it. Adding to his stress was his realization that he had left his cell phone at his office. He was able to flag down another driver, who let him use her cell phone to call his wife. She drove him to his office. Shortly after he arrived there, the attack began. An MRI scan showed that he had a hippocampal lesion, but it was so tiny as to be unlikely to represent a stroke. He recovered uneventfully.
By contrast, I treated another patient (Kirshner, 2011) who suffered amnesia but did not recover within 24 hours. The MRI scan showed a somewhat larger infarction in the left hippocampus. This patient’s diagnosis was not transient global amnesia, but stroke.
Mr Gabriel’s neurologist told him that he was unlikely to suffer another attack. Only about 15% to 20% of patients have recurrences (Miller et al, 1987; Nausieda and Sherman, 1979; Shuping et al, 1980). I believe that patients with recurrent episodes tend to be people whose attacks are triggered by migraine or seizure. For some of my patients who have multiple attacks suggesting partial seizures, I have found that treating with antiepileptic drugs can prevent recurrences. The dramatic response by several patients suggests that the two conditions are related.
Like patients who have one episode, those who have recurrences recover fully, apart from their permanent memory gaps.
After many descriptions, case series, and studies, transient global amnesia remains a mystery. We truly do not understand the cause. A hypothesis about venous insufficiency has not been confirmed (Cejas et al, 2010).
This state of the field was nicely summarized by Altamura and Vernieri (2010) as “investigating the case of [transient global amnesia], we have definitely identified the suspects, but we still have to prove them guilty.” Perhaps close attention to the details of the attack, such as Trip Gabriel has provided, will give us more clues to the cause.
Alberici E, Pichiecchio AE, Caverasi E, et al. 2008. Transient global amnesia: hippocampal magnetic resonance imaging abnormalities. Funct Neurol. 23:149–152.
Altamura C, Vernieri F. 2010. Internal jugular vein valve incompetence in transient global amnesia: more circumstantial evidence or the proof solving the mystery? Stroke. 41:1–2.
Ay H, Furie KL, Yamada K. 1998. Diffusion-weighted MRI characterizes the ischemic lesion in transient global amnesia. Neurology. 51:901–903.
Bender MB. 1956. Syndrome of isolated episode of confusion with amnesia. J Hillside Hosp. 5:212–215.
Cejas C, Cisneros LF, Lagos R, et al. 2010. Internal jugular vein valve incompetence is highly prevalent in transient global amnesia. Stroke. 41:67–71.
Dowd M. 2006. Valley of the rolls. The New York Times
. March 18, 2006: The Opinion Pages. Available at: www.nytimes.com/2006/03/18/opinion/valley-of-the-rolls.html
. Accessed February 16, 2017.
Eustache F, Desgranges B, Petit-Taboue MC. 1997. Transient global amnesia: implicit/explicit memory dissociation and PET assessment of brain perfusion and oxygen metabolism in the acute stage. J Neurol Neurosurg Psychiatr. 63:357–367.
Fisher CM, Adams RD. 1958. Transient global amnesia. Trans Am Neurol Assoc. 83:143–146.
Fisher CM, Adams RD. 1964. Transient global amnesia. Acta Neurol Scand. 40(suppl 9):1–83.
Gabriel T. 2017. The day that went missing: a first-person account of transient global amnesia. Cogn Behav Neurol. 30:1–4.
Gordon B, Marin OSM. 1979. Transient global amnesia: an extensive case report. J Neurol Neurosurg Psychiat. 42:572–575.
Guyotat J, Courjon J. 1956. Les ictus amnésiques [Amnesic seizures]. J Med Lyon. 37:697–701.
Hodges JR, Warlow CP. 1990. The aetiology of transient global amnesia: a case-control study of 114 cases with prospective follow-up. Brain. 113:639–657.
Kirshner HS. 2011. Transient global amnesia: a brief review and update. Curr Neurol Neurosci Rep. 11:578–582.
Lee SY, Kim WJ, Suh SH, et al. 2009. Higher lesion detection by 3.0T MRI in patients with transient global amnesia. Yonsei Med J. 50:211–214.
Miller JW, Petersen RC, Metter EJ, et al. 1987. Transient global amnesia: clinical characteristics and prognosis. Neurology. 37:733–737.
Morris HH III, Estes ML. 1987. Traveler’s amnesia: transient global amnesia secondary to triazolam. JAMA. 258:945–946.
Nausieda PA, Sherman IC. 1979. Long-term prognosis in transient global amnesia. JAMA. 241:392–393.
Shuping JR, Rollinson RD, Toole JF. 1980. Transient global amnesia. Ann Neurol. 197:281–285.
Strupp M, Bruning R, Wu RHH, et al. 1998. Diffusion-weighted MRI in transient global amnesia: elevated signal intensity in the left mesial temporal lobe in 7 out of 10 patients. Ann Neurol. 43:164–170.
Yang Y, Kim S, Kim JH. 2008. Ischemic evidence of transient global amnesia: location of the lesion in the hippocampus
. J Clin Neurol. 4:59–66.
Yarnell P, Lynch S. 1970. Retrograde memory immediately after concussion. Lancet. 1:863–864.
Zorzon M, Antonutti L, Mase G, et al. 1995. Transient global amnesia and transient ischemic attack: natural history, vascular risk factors, and associated conditions. Stroke. 26:1536–1542.