Case reasoning report · CDEMO01

Acute Dyspnea After Arthroplasty

A resident debrief on post-operative pulmonary embolism: stabilise before diagnosing, select D-dimer versus CTPA by pretest probability, anticoagulate with baseline labs, and stay inside the thrombolysis safety boundary.

ResidentEmergency DepartmentManagement & safetyTest stewardship

Edit case

What the cohort showed

Main pattern:Only 64% pinned the safety red flags before advancing to the workup.

What it means:Learners collected data but did not separate the safety-critical signal from the noise before acting.

Teach next:Run a focused timeline review and gate the workup on flagging the red flags first.

Expected reasoning pathway

What the case was designed to test, in order

Key clues

Pleuritic chest pain
Acute dyspnea
Unilateral calf swelling
Heart Ratered flag
Respiratory Ratered flag

Expected differential

Pulmonary embolismWorking diagnosis
Acute coronary syndromeCan't-miss
Community-acquired pneumoniaReasonable alternative
Aortic dissectionCan't-miss
Spontaneous pneumothoraxReasonable alternative

Expected tests

D-dimerDiagnostic workup
Baseline CBC, creatinine and coagulation panelPretreatment baseline
CT pulmonary angiography (CTPA)Diagnostic workup
12-lead ECG and continuous telemetry with repeat saturationsSafety monitoring

Expected management

Supplemental oxygen and continuous monitoring
Therapeutic anticoagulation — apixaban (DOAC)
Analgesia for pleuritic pain — paracetamol
Admit for observation and risk stratification

Completion funnel

Where learners dropped off

Case startedDifferential submittedPlan submittedCase completed

Assigned

—

Started

−2

Completed timeline

−3

Submitted DDx

−1

Submitted plan

−3

Completed case

−2

Signal selection — do learners know what matters?

Recall per pin intent; the distractor row is fixation risk

Pinned findingTagged pin

Pin intent	Opportunities	Pinned	Rate	Median time
Diagnostic clue	180	144	80%	45s
Safety red flag	180	115	64%	63s
Context factor	144	94	65%	65s
Management modifier	72	44	61%	74s
Distractorlower is better	72	30	42%	—

Test strategy & stewardship

Ordering as questions — expected vs ordered, and reveal rate for earned tests

Test orderedOrders submittedHidden tests revealed

Order intent	Expected	Ordered	Reveal rate
Diagnostic workupRight test, right time	72	74over	89%
Safety monitoringMonitoring gates	36	23under	63%
Pretreatment baselineReadiness before treatment	36	21under	58%

Diagnostic reasoning — did learners land it, near-miss, or wander?

Diagnosis rankedDifferential submittedResult viewed

Exact match47%×1
More specific (descendant)11%×0.7
Sibling (same parent)18%×0.4
Cousin (nearby)7%×0.2
No match17%×0

Tiers and multipliers from DDx scoring policy v1.0.0. ×n is the credit a match in that tier earns relative to an exact hit.

Can't-miss omissions

Aortic dissection39% omitted
Acute coronary syndrome28% omitted

Anchoring

73% updated after discriminator

27% kept their top diagnosis unchanged after a discriminating result.

Top wrong diagnoses

Community-acquired pneumonia18% · sibling
Spontaneous pneumothorax9% · cousin

Management & safety

Contraindicated actions, baseline readiness, and sequencing

Intervention selectedDrug orderedPlan submitted

Killer / contraindicated selections

Systemic thrombolysis (alteplase) in a haemodynamically stable patient17%high severity
High-dose NSAID alongside anticoagulation13%medium severity
Discharge home without anticoagulation9%high severity

41%

missed baseline labs before treatment

35%

sequenced management unsafely

Knowledge & evidence

Quiz performance by domain and reasoning objective

Question submittedAnswer selected

By domain

Test interpretation

74%n=36

Management

67%n=36

Patient safety

60%n=72

By objective kind

Reasoning gate

69%n=72

Platform gap assessment

62%n=36

Evidence application

63%n=36

Knowledge–deployment gap: learners answer the reasoning-gate items at 69% but their patient-safety behaviour scores 60% — they know the rule but don't yet deploy it under uncertainty.

Faculty debrief

A 10-minute debrief plan, grounded in the case teaching

1
Stabilise before you diagnose
2
Choose the test by pretest probability
3
Treat safely — and know the killer moves

Recommended for: Acute Dyspnea After Arthroplasty · drawn from the case teaching module.

Per-cohort pillar scores

How each assigned cohort performed on this case

Cohort	Diagnostic ability	Test stewardship	Management & safety	Knowledge & evidence
PGY-1 Internal Medicine	76	68	52	74
PGY-3 Internal Medicine	84	81	82	66
Med-2 Clerkship	64	41	48	70

Cells colored by status thresholds from DDx scoring policy v1.0.0 — success ≥ 70, warning ≥ 40.