# ED Adventures v4 — clinical model and review notes

Prepared October 28, 2026. Educational fictional adult scenarios; no patient data.
This is a deterministic teaching model, not a validated physiologic or pharmacokinetic simulator, prescribing tool, or complete resuscitation protocol. A qualified emergency clinician should review the content and institutional dosing conventions before teaching use.

## How the simulation behaves

Orders are independent of narrative nodes. Multiple orders can be initiated at the same simulated time. Immediate actions abstract preparation and delivery to zero minutes; delayed orders have explicit due times. The clock advances only through the 1- or 5-minute controls. Events complete at the start of their due minute, before that minute's physiologic update. Learners can reassess, inspect results, and choose disposition at any point.

Monitoring reveals vital signs. Glucose is an intermittent measurement with a timestamp, not a continuously visible true value. Serial checks generate persistent results every five minutes during stabilization. Results do not disappear after unrelated orders. Browser-local saves are separate for each case and are versioned v4; no account, server storage, analytics, or external scripts are used. A saved encounter must be explicitly resumed. Clearing browser data removes it.

Vital signs are simplified trend variables. Respiratory rate responds to modeled oxygenation/perfusion; mental status responds to glucose in the glyburide case. Oxygen can modestly support saturation but cannot correct anaphylactic, obstructive, or hemorrhagic shock. IV access is required before modeled IV therapies. Crystalloid has an immediate finite effect and blood slows, but does not stop, ongoing hemorrhage. No real-world rate or outcome probability should be inferred from the model.

Untreated severe shock/hypoxemia ends in critical deterioration after a short illustrative period; full resuscitation is outside this model. VT is the exception: pulse loss opens a simplified CPR/defibrillation rescue pathway. ROSC is scripted for teaching, not guaranteed clinically. Do not interpret it as a complete arrest algorithm. The simulation stops at 120 minutes if no earlier endpoint occurs.

## Case-specific rationale and dosing

### Anaphylaxis — The Swelling Patient

The adult has ceftriaxone-associated shock, wheeze and hoarseness. IM epinephrine is first-line. The displayed adult example is 0.5 mg IM, using 0.5 mL of 1 mg/mL into the lateral thigh. This is the RCUK adult regimen; local US protocols may use 0.3–0.5 mg. Confirm the route, concentration and dose; do not confuse IM concentration with an IV resuscitation preparation. Repeat after five minutes for persistent ABC symptoms. The initial dose produces partial improvement in this particular case, so reassessment and a second dose are expected.

Crystalloid 500 mL IV is a starting adult bolus with reassessment; further fluid depends on response and comorbidity. Stop the trigger, position safely, monitor, and prepare expert airway rescue. Antihistamines are adjuncts for skin symptoms, not shock/airway treatment. The intentionally unsafe 1 mg IV bolus option demonstrates harm in a patient with a pulse; it is not a recommended alternative.

After two IM doses, persistent ABC compromise warrants expert-led refractory treatment; the infusion action requires two prior doses and specialist consultation. No infusion concentration/rate is taught here because it must follow a monitored local protocol. IM effects wane after a compressed teaching window and recurrence may occur. Initial shock and airway involvement justify monitored high-acuity handoff in this case; the simulation does not establish discharge readiness or complete risk-based observation.

Sources: [RCUK anaphylaxis guidance](https://www.resus.org.uk/library/additional-guidance/guidance-anaphylaxis/emergency-treatment-anaphylactic-reactions), [2025 special circumstances guidance](https://www.resus.org.uk/professional-library/2025-resuscitation-guidelines/special-circumstances-guidelines).

### Unstable wide-complex tachycardia — The Racing Heart

The rhythm is specifically regular monomorphic VT with a pulse and shock. Apply pads and verify synchronization markers. Prioritize synchronized cardioversion; sedation is appropriate if feasible without delay. The 2025 AHA electrical cardioversion algorithm lists 100 J for monomorphic VT. Follow the specific device and local protocol; verify/re-enable sync for repeat shocks. This energy must not be generalized to AF or polymorphic VT. The broader 2025 tachyarrhythmia algorithm emphasizes device-recommended energy and maximum setting when unknown.

Amiodarone 150 mg IV over 10 minutes is shown as a stable wide-QRS strategy; waiting for it in this unstable presentation permits deterioration. Pulseless VT requires high-quality CPR and unsynchronized defibrillation. Unsynchronized shock may also be necessary if synchronization cannot be achieved in a critically unstable patient, although this case assumes synchronization is available. That exception is not modeled as a separate pathway. Post-conversion ECG, cause evaluation and monitored care remain necessary. Post-ROSC care is summarized, not simulated in full.

Sources: [2025 AHA algorithms](https://cpr.heart.org/en/resuscitation-science/cpr-and-ecc-guidelines/algorithms), [electrical cardioversion PDF](https://www.heart.org/-/media/CPR-Files/CPR-Guidelines-Files/2025-Algorithms/Algorithm-ACLS-Electrical-Cardioversion-250514.pdf), [tachyarrhythmia with pulse PDF](https://www.heart.org/-/media/CPR-Files/CPR-Guidelines-Files/2025-Algorithms/Algorithm-ACLS-Tachycardia-250514.pdf).

### Tension pneumothorax — The Silent Left Chest

Severe respiratory compromise and hypotension justify immediate decompression before imaging. Technique/site/device must follow the clinician's training and local trauma protocol; the simulator does not teach a procedural insertion technique. Needle or finger decompression may be temporizing. Definitive pleural drainage is modeled separately and can be ordered directly. Temporary decompression starts to fail after five simulated minutes without drainage, a teaching device rather than a real patency prediction. Positive-pressure ventilation before decompression can worsen obstruction. After stabilization, evaluate drain function, lung expansion and associated injuries.

Source: [NICE NG39, chest trauma recommendations](https://www.nice.org.uk/guidance/ng39/chapter/Recommendations).

### Hypotensive MVC abdominal trauma — The Unstable MVC Patient

Suspected intraperitoneal hemorrhage requires parallel trauma activation, warmed blood-product resuscitation, bedside assessment and urgent source control. FAST returns positive. A near-normal early hemoglobin does not rule out major blood loss. CT transport consumes time and is unsafe while persistently unstable in this scenario. Blood supports physiology but does not stop bleeding; definitive handoff is to the operating room, even if BP has not normalized. Handoff does not falsely claim hemorrhage was controlled before surgery. Exact transfusion ratios, calcium replacement and TXA dosing are not simulated; use the institutional hemorrhage protocol, injury timing and specialist assessment. Warming/calcium surveillance is represented as a support action.

Sources: [NICE NG39](https://www.nice.org.uk/guidance/ng39/chapter/Recommendations), [EAST blunt abdominal trauma evaluation](https://www.east.org/education-resources/practice-management-guidelines/archived/blunt-abdominal-trauma-evaluation-of) (archived guideline; used only for the established unstable/FAST-positive operative principle).

### Persistent glyburide hypoglycemia — The Glucose That Will Not Stay Up

An older adult with poor intake and renal impairment has severe symptomatic hypoglycemia. Immediate IV glucose is needed because oral intake is initially unsafe. The adult example is dextrose 25 g, expressed as D10W 250 mL. Verify IV patency and concentration; titrate to local practice and reassess promptly. Other adult concentrations/volumes can be appropriate but are not modeled as selectable regimens.

The bolus corrects glucose transiently. Without octreotide the model drops glucose by a compressed teaching rate, causing recurrence in about 20 minutes; this is not a glyburide kinetic forecast. Octreotide 50 micrograms SC is one adult starting example. Subsequent dosing (for example every eight hours in the Queensland protocol), an IV regimen, duration and glucose-infusion titration require poison-service advice. The single SC dose is treated as effective only within this short encounter. It does not replace immediate glucose correction. A dextrose infusion may be needed but does not remove the insulin-driven mechanism. Serial bedside checks are modeled every five minutes during stabilization; real frequency changes with clinical state and treatment phase.

Hold glyburide, review renal function, and consult toxicology. Admit for recurrent/prolonged risk, continuing checks and observation after cessation of glucose/octreotide. Observation is often at least 12 hours after therapies stop and may need to be longer; local poison-service guidance and the clinical course govern. The model ends at admission and does not simulate that observation period.

Sources: [Queensland Poisons octreotide](https://www.poisonsinfo.health.qld.gov.au/for-health-professionals/antidote-stocking-recommendations/octreotide) (Australian adult regimen; adapt to local practice), [RCH oral hypoglycaemic poisoning](https://www.rch.org.au/clinicalguide/guideline_index/Oral_Hypoglycaemic_Poisoning/) (pediatric-focused guideline that also gives an adult octreotide example; not a general adult dosing source).

## Debrief rubric

Each case has observable clinical objectives, including early therapy, monitoring, reassessment and definitive/monitored disposition. Score = rounded percentage of objectives met minus 10 for each distinct modeled safety event, floored at zero. Repeating an action does not repeatedly earn points. First completion timestamps determine early-treatment objectives; persistent flags and counts support repeat-treatment checks. A recent clinical reassessment means within five simulated minutes of the endpoint. The rubric is intentionally case-specific and is not a validated assessment. A correct handoff may coexist with missed earlier objectives. Every home discharge in these short, high-risk encounters is treated as unsafe.

## Extension structure

Each case's metadata remains in cases/*.js. Clinical action definitions, time progression, consequences and rubric rules live in assets/simulation.js. The renderer in assets/ui.js does not contain the physiology. New cases require metadata plus clinical rules, actions, disposition criteria, source review, and positive/negative tests; adding a case title alone is insufficient. The six additional cases are now implemented in cases/expansion.js with per-case clinical hooks, actions, disposition criteria and tests; see the v5 expansion clinical notes.



# v5 expansion — six additional scenarios

Prepared October 28, 2026. These fictional cases extend v4/v4.1 to eleven cases. Existing clinical limitations still apply. All progression is deterministic and compressed for teaching. This is not bedside prescribing support or a validated performance assessment. A qualified emergency/pediatric clinician must review local dosing and escalation pathways before teaching use.

## Mixed overdose

70 kg adult with oxycodone, clonazepam and amitriptyline exposure. Ventilation, opioid reversal and sodium-channel toxicity are separate problems. Initial naloxone example is 0.1 mg IV, titrated to respiration with assisted ventilation available; severe apnea may need a larger starting dose under local protocol. Four selectable naloxone doses are a simulator limit, not a clinical maximum. Expert airway support represents ongoing support during transfer, not proof that intubation was necessary. Naloxone's modeled six-minute response window is compressed and is not its clinical duration. QRS narrows after bicarbonate in this scripted responder; serial gas/electrolyte/ECG surveillance remains essential. No bicarbonate recipe or infusion dosing is prescribed. Empiric flumazenil and unprotected oral charcoal trigger safety feedback. Admission still requires further co-ingestant testing and psychosocial assessment when appropriate.

Sources: AHA 2025 special circumstances (https://cpr.heart.org/en/resuscitation-science/cpr-and-ecc-guidelines/adult-and-pediatric-special-circumstances-of-resuscitation/); Queensland Poisons naloxone (https://www.poisonsinfo.health.qld.gov.au/for-health-professionals/antidote-stocking-recommendations/naloxone); RCH TCA poisoning (https://www.rch.org.au/clinicalguide/guideline_index/Tricyclic_Antidepressant_(TCA)_Poisoning/).

## Flash pulmonary edema

82 kg adult with severe hypertension and respiratory distress. This phenotype supports early NIV and carefully titrated nitrates after contraindication screening, alongside individualized diuresis and precipitant evaluation. A normal troponin or ECG is not used to dismiss ACS. The model offers expert airway rescue if noninvasive support fails. Numerical BP reduction and improvement are scripted, not treatment targets or a recommended rate of BP reduction. It does not model every contraindication, device setting or individualized dose. IV fluid loading worsens this particular overloaded patient; this must not be generalized to every heart-failure presentation.

Source: CAEP 2025 acute heart failure guidance (https://link.springer.com/article/10.1007/s43678-025-00961-y). The simulator is independently written, does not reproduce the checklist/figures and does not imply CAEP endorsement.

## Pediatric status epilepticus

5-year-old, measured 20 kg, with one adequate EMS midazolam dose already received. The selectable second total dose is 3 mg IV (0.15 mg/kg, maximum 10 mg/dose) over two minutes. The simulated child remains convulsing after that dose, prompting second-line therapy. Levetiracetam is 800 mg IV (40 mg/kg, maximum 3 g) over five minutes under the PCH regimen. Other accepted protocols use different loads; verify institutional practice. The weight is fixed, not an editable dosing calculator. Additional repeated benzodiazepines model respiratory harm. Convulsive cessation does not prove full neurological recovery; continue respiratory assessment, cause evaluation and possible EEG. Anesthetic infusion dosing and a complete refractory-status protocol are outside scope.

Sources: Perth Children's Hospital status epilepticus, endorsed July 2026 (https://pch.health.wa.gov.au/for-health-professionals/emergency-department-guidelines/status-epilepticus); NICE NG217 (https://www.nice.org.uk/guidance/ng217/chapter/7-Treating-status-epilepticus-repeated-or-cluster-seizures-and-prolonged-seizures).

## Intussusception

14-month-old, 10 kg, initially poorly perfused. 100 mL of 0.9% saline represents a 10 mL/kg aliquot with reassessment, capped at two modeled aliquots before specialist escalation. This cap is not a universal resuscitation maximum. Ultrasound confirms ileocolic intussusception. Enema reduction is available only after stabilization, imaging and specialist arrangement; candidacy is checked again at completion. Peritonitis emerges after a compressed 15-minute untreated interval; this is a teaching event, not a real onset prediction. Surgery can be the appropriate handoff if reduction is unsafe. The scripted successful reduction does not imply all reductions succeed or remove the need for observation.

Source: RCH intussusception, February 2026 (https://www.rch.org.au/clinicalguide/guideline_index/intussusception/).

## Hemolytic uremic syndrome

4-year-old, 16 kg, with established oliguric overload after bloody diarrhea. CBC, hemolysis studies and AKI findings support HUS; stool testing returns Shiga toxin positive. Other TMAs still require specialist consideration. This differs from early diarrheal hypovolemia, where appropriate rehydration may be protective. Cardiac calcium does not reduce potassium; shifting therapy has a delayed effect. High-risk insulin/glucose and calcium protocols are represented as supervised actions without a dose recipe; verify local weight-based orders and concentrations. Serial glucose is represented every five simulated minutes for close observation, not a universal clinical schedule. Definitive renal support is arranged, not magically completed. Handoff may be safe with persistent potassium elevation only when cardiac protection, temporary therapy and immediate tertiary renal care are organized. Model potassium trends are illustrative. Do not routinely treat STEC with antibiotics or antimotility agents; separate infections can require antibiotics.

Sources: CDC STEC clinical guidance (https://www.cdc.gov/ecoli/hcp/guidance/index.html); NHS GGC HUS guidance (https://www.clinicalguidelines.scot.nhs.uk/ggc-paediatric-guidelines/ggc-paediatric-guidelines/kidney-diseases/management-and-investigation-of-bloody-diarrhoea-and-haemolytic-uraemic-syndrome/); RCH hyperkalemia (https://www.rch.org.au/clinicalguide/guideline_index/Hyperkalaemia/).

## Testicular torsion with abdominal pain

15-year-old, 58 kg, already three hours into symptoms. A respectful, chaperoned examination reveals high-suspicion findings. Surgical activation and operative handoff should not be delayed for routine imaging. Doppler can be appropriate after surgical assessment when it does not delay definitive care. Manual detorsion is an expert temporizing maneuver; symptom relief cannot establish complete detorsion and does not replace exploration/fixation. Increasing ischemic-risk feedback after 15 simulated minutes and ending at 60 minutes are educational thresholds, not a viability clock. Near-normal vitals may persist despite organ-threatening pathology; this case intentionally does not require cardiovascular collapse to demonstrate harm.

Sources: RCH acute scrotal pain (https://www.rch.org.au/clinicalguide/guideline_index/Acute_scrotal_pain_or_swelling/); RACS torsion guidance (https://www.surgeons.org/about-racs/position-papers/acute-scrotal-pain-and-suspected-testicular-torsion-guidelines-2022).

## Automatic clock and saved encounters

All eleven cases start paused and support one simulated minute per 10, 30 or 60 real seconds. Tab hiding pauses; there is no catch-up. Orders remain independent and delayed results persist. Manual time is disabled during automatic progression. Terminal outcomes stop the clock. Resume restores clinical state but stays paused. Pediatric labels and weight-specific doses are shown per case, and the renderer does not apply the original adult BP normalization to these six models.

Clinical review is still required. Testing can confirm software behavior and arithmetic, not the clinical validity of a simplified physiological model.