# Sources — Bruce Lee

> **SYNTHETIC RECORD, NOT A REAL MEDICAL RECORD**

Accessed November 5, 2026. The canonical source metadata and claim links are in `case.json`.

## Historical evidence

1. Bruce Lee Enterprises. “Bruce Lee” (short and long biography). Supports that “Bruce Jun Fan Lee was born … on November 27, 1940 at the Jackson Street Hospital in San Francisco's Chinatown” while his parents were on tour with the Chinese opera; that “the English name, BRUCE, was given to the baby boy by a nurse”; that he returned to Hong Kong at three months and was raised there until 18; his study of wing chun under Yip Man from 13; his return to the United States in 1959, the University of Washington and his schools in Seattle, Oakland and Los Angeles; Jeet Kune Do, “the way of the intercepting fist”; his marriage in 1964 and the births of his children; his father's death in Hong Kong in 1965; the back injury of 1970, when “without warming up … Bruce picked up a 125-pound barbell and did a ‘good morning’ exercise severely injuring his back,” found to be “an injury to the fourth sacral nerve,” with six months in bed and “a chronic source of pain for the rest of his life”; his films for Golden Harvest; and, on July 20, 1973, the headache, the Equagesic tablet and the “nine-day coroner's inquest.” https://brucelee.com/bruce-lee
2. South China Morning Post. “Do we know what killed Bruce Lee? Yes, and no – and that's left ample room for speculation as to the real cause” (July 19, 2018). Supports the Queen Elizabeth Hospital report that his death “was due to acute cerebral oedema”; that “there was disagreement about the cause of the cerebral oedema between doctors at the official inquest”; that “traces of hash were found in Lee's stomach and intestine”; that Equagesic is “a painkiller formed of meprobamate and aspirin” and that “neither meprobamate nor aspirin … are known to cause a cerebral oedema”; that “a verdict of death by misadventure was recorded at the inquest”; and the May 10 collapse while “dubbing lines for Enter the Dragon at film company Golden Harvest's Hammer Hill Road studios,” when “he ate some hash, felt faint,” fainted, vomited and fainted again, and a neurosurgeon “diagnosed a cerebral oedema and gave him the drug Mannitol.” https://www.scmp.com/culture/film-tv/article/2155978/do-we-know-what-killed-bruce-lee-yes-and-no-and-thats-left-ample
3. Lui M. “What killed Bruce Lee? 4 leading theories medical and martial arts experts proposed after his shock death.” South China Morning Post (July 19, 2023). Cites the Post's reports of July 22 and September 19 and 21, 1973. Supports the cannabis traces, the May 10 diagnosis of cerebral oedema when he “collapsed at a film studio and was in critical condition,” the hypersensitivity explanation proposed by a London professor of forensic medicine and supported by the hospital pathologist, that “the air conditioning in the dubbing room had been turned off to avoid sound issues with the recording,” the removal of his armpit sweat glands “a few months earlier,” and the heat stroke and hyponatraemia theories. https://www.scmp.com/lifestyle/entertainment/article/3228184/what-killed-bruce-lee-4-leading-theories-medical-and-martial-arts-experts-proposed-after-his-shock
4. Villalvazo P, Fernandez-Prado R, Niño MDS, et al. Who killed Bruce Lee? The hyponatraemia hypothesis. *Clin Kidney J* 2022;15(12):2169-2176 (PMID 36381374; PMC9664576; open access, CC BY). Supports his birth name Lee Jun-fan; the May 10 events as told by his biographer, including that “his body began convulsing,” that he was treated at Baptist Hospital with mannitol, and that “a seizure was considered a potential cause … although as it was a sole episode, a diagnosis of epilepsy was not made”; the neurological evaluation of May 29 and 30, 1973, with an electroencephalogram that found no abnormality, after which “phenytoin was prescribed”; and the authors' hypothesis that “the cause of death was cerebral oedema due to hyponatraemia.” https://pubmed.ncbi.nlm.nih.gov/36381374/
5. Vancini RL, de Lira CAB, Andrade MS, Arida RM. Could epilepsy have been the cause of Bruce Lee's death?: “The athlete and myth of martial arts.” *Epilepsy Behav* 2020;111:107310 (PMID 32693379). Title and citation only. https://pubmed.ncbi.nlm.nih.gov/32693379/
6. HISTORY.com. “Actor and martial-arts expert Bruce Lee dies at age 32.” Supports that he died in Hong Kong on July 20, 1973, “from a brain edema possibly caused by a reaction to a prescription painkiller”; his birth in San Francisco while “his father, a Chinese opera star, was on tour in the United States”; some 20 films as a child actor; and burial in Seattle. https://www.history.com/this-day-in-history/july-20/bruce-lee-dies-at-age-32
7. Phoenix Television (ifeng.com). 李小龙逝世50周年 (July 21, 2023). Supports his Chinese name, 李振藩, given by his parents, and his stage name, 李小龙. https://news.ifeng.com/c/8RbWf9Zu3UO
8. Hong Kong Leisure and Cultural Services Department. Press release on the exhibition 「武‧藝‧人生─李小龍」 (June 26, 2013). Supports that he grew up and became famous in Hong Kong, “是香港的驕傲” (the pride of Hong Kong). https://www.tourism.gov.hk/resources/tc_chi/paperreport_doc/press/2013-06-26/Bruce_Lee_exhibition_tc.pdf
9. HISTMED Master Roster v3. Internal project source; not treated as independent corroboration.
10. Bruce Lee Foundation. “About Us.” Supports its mission, “to share the teachings and legacy of Bruce Lee through martial arts and mindfulness programs to empower youth mental wellness,” and its founding in 2002. https://bruceleefoundation.org/about

## Clinical guidance

1. Krumholz A, et al. Evidence-based guideline: Management of an unprovoked first seizure in adults (American Academy of Neurology and American Epilepsy Society). *Neurology* 2015;84(16):1705-1713 (PMID 25901057): recurrence risk greatest within two years (21% to 45%); risk factors (prior brain insult, epileptiform EEG, significant imaging abnormality, nocturnal seizure); immediate treatment likely reduces two-year recurrence but does not improve long-term remission; adverse events in 7% to 31%; an individualized decision with the patient. https://pubmed.ncbi.nlm.nih.gov/25901057/
2. Krumholz A, et al. Practice Parameter: evaluating an apparent unprovoked first seizure in adults. *Neurology* 2007;69(21):1996-2007 (PMID 18025394): EEG and brain imaging as part of the routine evaluation; laboratory tests such as electrolytes (particularly sodium) and toxicology as the circumstances indicate. https://pubmed.ncbi.nlm.nih.gov/18025394/
3. Fisher RS, et al. Operational classification of seizure types by the International League Against Epilepsy. *Epilepsia* 2017;58(4):522-530 (PMID 28276060). https://pubmed.ncbi.nlm.nih.gov/28276060/
4. Fisher RS, et al. ILAE official report: a practical clinical definition of epilepsy. *Epilepsia* 2014;55(4):475-482 (PMID 24730690). https://pubmed.ncbi.nlm.nih.gov/24730690/
5. Beghi E, et al. Recommendation for a definition of acute symptomatic seizure. *Epilepsia* 2010;51(4):671-675 (PMID 19732133): severe metabolic derangements, drug or alcohol intoxication and withdrawal, exposure to well-defined epileptogenic drugs, and acute brain insults. https://pubmed.ncbi.nlm.nih.gov/19732133/
6. Qaseem A, et al. Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain (American College of Physicians). *Ann Intern Med* 2017;166(7):514-530 (PMID 28192789). https://pubmed.ncbi.nlm.nih.gov/28192789/
7. Transport Department, Hong Kong. Supplementary Health Information Sheet for Driver Assessment: lists the diseases and disabilities in the First Schedule of the Road Traffic (Driving Licences) Regulations (Cap. 374B), including “epilepsy” and “liability to sudden attacks of disabling giddiness or fainting,” for which “the Commissioner for Transport shall refuse his application for driving licence or cancel his driving licence.” https://www.td.gov.hk/filemanager/en/content_4964/Supplementary%20Health%20Information%20Sheet%20for%20Driver%20Assessment.pdf
8. Driver and Vehicle Licensing Agency (UK). Neurological disorders: assessing fitness to drive. For a first unprovoked or isolated seizure, car drivers must not drive for six months, or twelve months if there is an underlying causative factor that may increase risk. Used as an example of common licensing practice, not as a rule that applies in Hong Kong. https://www.gov.uk/guidance/neurological-disorders-assessing-fitness-to-drive
9. Centers for Disease Control and Prevention. “First Aid for Seizures” (May 15, 2024). https://www.cdc.gov/epilepsy/first-aid-for-seizures/index.html
10. U.S. Preventive Services Task Force A and B recommendations, HIV (2019), hepatitis C (2020) and unhealthy drug use (2020); CDC hepatitis B screening and testing recommendations (2023; PMID 36893044) and the Adult Immunization Schedule by Age; Medicare Preventive Services reference (Aether Forge, 2026). https://www.uspreventiveservicestaskforce.org/uspstf/recommendation-topics/uspstf-a-and-b-recommendations; https://pubmed.ncbi.nlm.nih.gov/36893044/; https://www.cdc.gov/vaccines/hcp/imz-schedules/adult-age.html; https://patrickrutledge.github.io/preventative_services/

## Editorial resolutions

- **Sources.** The case relies on Bruce Lee Enterprises (his family's company), two South China Morning Post articles (one citing the Post's own reports from July and September 1973), the peer-reviewed Clinical Kidney Journal article, HISTORY.com, Phoenix Television, a Hong Kong government press release and the Bruce Lee Foundation. The New York Times archive, the 1973 inquest record and the Post's 1973 articles themselves were not read; no biography page was found on the Bruce Lee Foundation's site, so its About page is used only for his legacy. Britannica, Wikipedia and aggregator sites were not used. The Epilepsy & Behavior letter was read as a title only.
- **Recent-death rule.** He died in 1973, so the rule does not apply; only documented claims are used, and a sick visit is included.
- **Living people.** His wife and daughter are living or may be, and are not named anywhere in the record, including source locators; nothing is attributed to them. Others present on July 20, 1973, and the clinicians who treated him in 1973 are not named in claim statements; the London pathologist (deceased) is named only in a quoted source locator, and his parents only in the Chinese-name locator. His father appears in a claim and in the family history. No real person is given a modern statement, decision or health fact; the colleague who describes the seizure and his health care agent are synthetic and unnamed.
- **Name.** The display name is Bruce Lee. “Lee” alone is not an alias, because it appears in other records. The aliases are Lee Jun-fan (the romanization in the Clinical Kidney Journal; Bruce Lee Enterprises writes “Jun Fan” as a given name) and 李振藩 (Phoenix Television); neither appears in any other record. The stage name 李小龍 (Lee Siu-lung; “Lee Siu Loong” in Bruce Lee Enterprises) is recorded in a claim but not used as an alias. identity.name_parts is not used: the name splits simply into given name Bruce and family name Lee.
- **Cause of death.** The cause of the cerebral oedema that killed him is recorded as disputed. The inquest verdict (misadventure) and the hospital's finding (acute cerebral oedema) are D (CLM-0012, CLM-0013). The explanation favored at the inquest, hypersensitivity to Equagesic (CLM-0014), and the later proposals of hyponatraemia (CLM-0015), heat stroke (CLM-0016), epilepsy (CLM-0017) and cannabis (CLM-0018) are R with certainty “disputed”; none is adopted. Theories of assassination and the like are not recorded. Claims about anabolic steroids, diuretics and cocaine that the Clinical Kidney Journal article attributes to one biographer were not used: they rest on a single secondary account and are not needed for the episode.
- **The May 10, 1973 collapse.** That he collapsed twice while dubbing, vomited, was in critical condition and was treated for cerebral oedema with mannitol is D (CLM-0007). That he convulsed rests on his biographer's account, quoted in the Clinical Kidney Journal (CLM-0008, D “probable”), as does the unventilated room (CLM-0009). The Clinical Kidney Journal names Hong Kong Baptist Hospital; the Post articles do not name the hospital in the text read.
- **Index age and dates.** The index visit on May 14, 2026 maps to May 14, 1973, four days after the collapse, when he was 32 and in Hong Kong. The synthetic date of birth is 1993-11-27, keeping the true month and day, a shift of 53 years. The other visits (May 18, June 15, 22 and 29, and July 13, 2026) map to the same days of 1973, all before his death on July 20. No visit falls between May 19 and June 14, 1973, because he was examined in the United States on May 25 and on May 29 and 30 (CLM-0011); the date he returned to Hong Kong is not documented in the sources read, and nothing read places him away from Hong Kong after mid-June.
- **Setting.** The project facility (St. Posthumous Health Center) is modeled as a clinic in Kowloon with an affiliated epilepsy clinic, and the driving advice uses Hong Kong rules. The preventive visit and the guideline frameworks are the U.S. ones used throughout the dataset; this is a modeling choice.
- **Divergence from 1973.** The modern episode does not reproduce the 1973 care. The modern CT and MRI show no brain swelling, whereas cerebral oedema was diagnosed in 1973; the modern clinicians start no antiseizure medicine, whereas phenytoin was prescribed in 1973; the modern sodium values are normal, and the record does not test or endorse the hyponatraemia hypothesis. The modern fluid advice (drink to thirst, with salt in the heat) is general teaching, not a claim about his death.
- **Presenting problem and diagnosis.** The presenting problem is a convulsive collapse, seizure versus convulsive syncope (R55, verification provisional), resolved at the epilepsy clinic into a first unprovoked seizure (R56.9, recorded at the epilepsy clinic, placement both_with_link to the documented collapse). R56.9 is used because a single seizure is not epilepsy; no G40 code is used.
- **Unprovoked, not acute symptomatic.** Judgment call: cannabis the same morning, a week of short sleep, missed meals and heat are treated as triggers, not acute symptomatic causes, because sodium, glucose and calcium were normal, he was not intoxicated by his account, and cannabis is not an established epileptogenic drug (Beghi 2010). A reviewer could reasonably class the seizure as provoked; the treatment decision would be the same.
- **No antiseizure medicine.** Judgment call under the AAN/AES 2015 guideline: with a normal MRI and EEG, a daytime seizure and no prior brain insult, he chose no immediate medicine after the risks and benefits were explained, with levetiracetam or lamotrigine planned after a second unprovoked seizure.
- **Driving.** Hong Kong's driving licence regulations list epilepsy and liability to sudden fainting among the conditions for which a licence must be refused or cancelled; the Transport Department form read gives no seizure-free interval, and the regulations themselves were not read. The clinic advises at least six months without driving, citing the UK DVLA rule for a first unprovoked seizure as an example of common practice. This is a judgment call.
- **Cannabis.** His cannabis use on the day of the May 10 collapse and the cannabis found at autopsy are recorded from the cited accounts (sensitive content), without judgment. The modern episode records occasional use, counsels without judgment and diagnoses no cannabis use disorder (no F12 code), so no safety assessment is required; he chooses to stop for now.
- **Back pain and medicine choices.** The chronic low back pain (M54.50) is anchored to the documented 1970 injury (CLM-0005); its description of an injury to the fourth sacral nerve is as Bruce Lee Enterprises reports it. Acetaminophen as needed, exercise therapy and, at the sick visit, a seven-day course of naproxen were chosen because they neither lower the seizure threshold nor interact with antiseizure medicines; tramadol, bupropion, cyclobenzaprine, carisoprodol and meprobamate were avoided. The NSAID advice includes not forcing water, because NSAIDs impair water excretion.
- **Synthetic clinical picture.** Everything in the modern episode is synthetic: the symptoms, the witness account, the emergency department values (narrative), every laboratory value, the ECG, MRI and EEG findings, the pain scores, PHQ-2 and AUDIT-C, the screening results (HIV, hepatitis B and C, narrative), medicines and outcomes. Height (171 cm) and weights (57.6 to 59.4 kg) are synthetic; his biographer's report that he weighed 54 kg before his death is not used as a measurement. Reference ranges are typical adult male ranges.
- **Sick visit.** A flare of low back pain with muscle spasm after a training kick, chosen because it follows from his documented back injury and tests medicine choices after a seizure, without insinuating an undocumented diagnosis. The saddle sensation and bladder and bowel questions are asked because of the reported sacral nerve injury.
- **Preventive visit.** A periodic adult preventive visit at 32 (CPT 99395 and 99497 in plain text). Judgment calls: HIV, hepatitis C and hepatitis B screening (the last because he was raised in Hong Kong, where hepatitis B is more common), Tdap and the first dose of HepB-CpG that day, HPV vaccine declined by shared decision at 32, and influenza and COVID-19 vaccines deferred to the autumn. A health care agent from his household is recorded without a name.
- **CMID.** The CMID has no monograph for seizures or low back pain, so it is not cited; the focused visits use the AAN/AES, ILAE and ACP guidance instead.
- **Focused visits.** The three chronic-care follow-ups (the epilepsy clinic four days after intake, one month and two months) are documented as focused visits on the AAN/AES 2007 evaluation and 2015 management guidance, the ILAE 2017 classification and 2014 definition, and the ACP 2017 low back pain guideline. One deviation is recorded instead of an invented result: the creatine kinase and creatinine repeat due about a week after the first visit was missed because he was away for work and no visit was booked; at five weeks both were normal (168 U/L and 0.96 mg/dL), so no harm resulted, and the corrective action is booked laboratory visits with a reminder. Repeat MRI and EEG are recorded as not indicated, with reasons. Synthetic findings introduced: the normal MRI and EEG, the falling creatine kinase and the back flare.
- **Consultation.** The epilepsy referral and the consultation by Priyanka Abernathy-Leung, MD (a fictional consultant), at the epilepsy clinic four days after intake, are **S**. The consultant answers the five referral questions in line with what the record shows afterwards: an epileptic seizure rather than syncope, classed as unknown-onset motor (tonic-clonic); a normal MRI and EEG; a first unprovoked seizure that is not epilepsy; no antiseizure medicine, which he chose, with one to start after a second seizure; and no driving for six months, no heights or swimming alone, no cannabis, fluids to thirst with electrolytes, and avoidance of tramadol, bupropion, carisoprodol and meprobamate. The six-month review is a plan only.
