Clinical background
Headache is one of the most common reasons adults seek care in the emergency department. Within that large and heterogeneous group, subarachnoid hemorrhage (SAH) is uncommon but carries substantial morbidity and mortality when diagnosis is delayed. Many patients with aneurysmal SAH present with relatively preserved neurologic examinations, which makes selective use of neuroimaging and, when appropriate, cerebrospinal fluid evaluation a central part of safe emergency practice.
The Ottawa Subarachnoid Hemorrhage (SAH) Rule is a clinical decision aid developed and prospectively validated to help clinicians identify which patients with a specific headache presentation should undergo investigation to rule out SAH. It is not a substitute for bedside judgment, shared decision-making, or institutional pathways for “thunderclap” or high-risk headache, but it provides a structured checklist aligned with published derivation and validation cohorts.
Why a rule is needed
Non-contrast computed tomography (CT) of the brain is the usual first test when SAH is suspected, but its sensitivity depends heavily on time from symptom onset, scanner quality, and reader experience. Patients who present later may have false-negative CT results, which is why lumbar puncture or other strategies sometimes remain part of care pathways. At the same time, indiscriminate imaging and LP for every headache leads to crowding, cost, patient discomfort, and downstream evaluation of incidental findings.
A well-specified rule can help clinicians risk-stratify patients who share a narrow inclusion profile: neurologically normal adults with a new, severe headache that reached maximal intensity quickly. The Ottawa SAH Rule encodes six historical and examination features that, when any one is present, identify patients in whom investigation for SAH is appropriate within that framework.
Population to which the rule applies
The rule was studied in patients who were neurologically intact on examination (for example, without focal deficits and with a normal level of consciousness appropriate to the study definition), had a new acute non-traumatic headache, and whose headache reached maximal intensity within one hour of onset. Patients outside this envelope were excluded from the derivation and validation cohorts; applying “all variables negative” logic to defer testing in excluded populations is not supported by the published evidence.
Commonly cited exclusions from the rule’s intended use include:
- Recurrent headache patterns defined as three or more similar headaches over a period longer than six months.
- Headache attributed to acute head trauma.
- Presence of a ventriculoperitoneal shunt, known intracranial mass, or prior aneurysmal SAH.
- Headache that did not peak within one hour (the rule’s validation context assumes rapid time to maximal intensity).
When any of these situations apply, clinicians should follow broader emergency neurology and neuroimaging pathways rather than relying on the Ottawa SAH Rule alone to justify omitting investigation.
The six Ottawa SAH Rule variables
In an appropriate patient, investigation for SAH is indicated if any one of the following six variables is present. Age is evaluated as a continuous threshold; the other five are historical features or a targeted physical examination finding.
| Variable | What to assess |
|---|---|
| Age ≥ 40 years | Chronological age at presentation. This is a standalone risk modifier in the rule and does not require accompanying symptoms. |
| Neck pain or stiffness | Patient-reported neck symptoms associated with the headache presentation. This should be distinguished from chronic musculoskeletal complaints when possible by tying symptoms to the index event. |
| Witnessed loss of consciousness | Documented syncope or transient loss of consciousness observed by another person in association with the headache onset or early course. |
| Onset with exertion | Headache begins during or immediately after exertional activity such as heavy lifting, straining, valsalva, or sexual activity. The mechanism is thought to relate to transient rises in blood pressure and wall stress on intracranial vessels. |
| Thunderclap quality | Headache reaches maximal intensity immediately at onset (often described as “the worst headache of my life” with sudden onset). This is conceptually related to, but not identical with, the one-hour peak criterion used for study inclusion; thunderclap features remain a distinct positive variable within the rule. |
| Limited neck flexion on examination | Objective reduction of passive or active neck flexion compared with normal, interpreted in clinical context (fever, muscle spasm, cervical disease, and patient cooperation can mimic or obscure meningismus). |
Interpretation when variables are absent
In validation work among eligible patients, when none of the six variables were present, the rule exhibited very high sensitivity for identifying subarachnoid hemorrhage within the studied definitions and follow-up. That performance characteristic supports using “all negative” as one element of a safe disposition strategy only when the patient truly meets inclusion criteria, the history and examination are reliable, and institutional policy allows.
Sensitivity in research settings does not guarantee zero miss risk in every real-world encounter. Language barriers, analgesic effects, prior opioid use, psychiatric comorbidity, anemia affecting CT density, delayed presentation, and atypical histories can all alter pre-test probability and the reliability of historical elements. Negative rule variables therefore never remove the obligation to reassess if symptoms evolve or new findings appear.
How investigation typically proceeds when the rule is positive
When any variable is positive in an appropriate patient, clinicians generally proceed with urgent evaluation for SAH. Non-contrast CT brain imaging is the usual initial study. If CT is performed very early after symptom onset and is read as negative, some pathways incorporate shared decision-making about whether further testing is necessary, informed by time from onset, local protocols, and individual risk factors. For presentations beyond early windows, many guidelines emphasize additional testing because CT sensitivity falls as hours pass.
Lumbar puncture, CT angiography, and vascular imaging each have roles depending on pre-test probability, contraindications, availability, and suspicion for alternative diagnoses such as cervical artery dissection or cerebral venous thrombosis. The Ottawa SAH Rule does not specify which second-line test to choose; it addresses whether the patient’s presentation falls into the risk stratum that warrants SAH-directed investigation in the first place.
Performance characteristics in context
Prospectively validated decision rules for rare outcomes almost always trade specificity for sensitivity. In practice, a rule with very high sensitivity will label many patients as needing investigation, including those who ultimately have benign primary headaches, migraine, viral syndromes, or other non-SAH diagnoses. That pattern is expected: the purpose is to avoid missing a treatable vascular catastrophe, not to minimize imaging rates at the expense of safety.
Conversely, clinicians should not interpret “rule negative” as identical to “no serious disease.” Even within neurology, other emergent entities may present with sudden severe headache and normal early CT. The Ottawa SAH Rule targets SAH risk in a defined population; it does not comprehensively rule out all neurologic emergencies.
Documentation, communication, and follow-up
Clear documentation should capture time of onset, time to maximal intensity, associated symptoms (syncope, vomiting, neck symptoms), anticoagulant use, blood pressure, focused neurologic examination including neck flexion, and the rationale for testing or observation. When investigation is deferred after shared decision-making, record the discussion, specific return precautions (worsening headache, new focal deficits, seizures, fever with meningismus), and arrangements for primary or neurology follow-up.
Patients often benefit from plain-language explanation of why SAH is considered even when it is unlikely, and why CT may be normal while further testing is still recommended in selected time windows. Framing the rule as a standardized safety checklist can reduce perceived conflict between “low probability” and “we still need to rule out a bleed.”
Using this calculator
The companion calculator on this site walks through the same inclusion exclusions and six variables used in teaching versions of the Ottawa SAH Rule. Enter the patient’s age and mark historical and examination features as present or absent. The tool highlights when investigation is indicated by the rule, when the rule should not be applied, and offers general next-step prompts. It remains an educational aid only and must be interpreted together with the treating clinician’s assessment and local emergency neuroimaging pathways.