Thread by @KirollosRoman

Formulating and appraising your PR (problem representation) is critical to the diagnostic process, and intrinsically welcomes cognitive bias

Work this case with me #MedTwitter

ED calls for admission: “40yo M w pneumonia”

*fist pump* sweet, easy admission

1/

“Fever, cough, dyspnea x2d. WBC 20, triggered sepsis. Got cx, fluids, abx. CXR RLL infiltrate. Needs admission for sepsis 2/2 CAP”

With that PR, slam dunk pneumonia.

100% match for my CAP illness script, don’t even need my dyspnea schema (courtesy of @DxRxEdu @CPSolvers)

#CognitiveBias of premature closure
+
cognitive load of retaking a full hx on such an “easy admission” 12hrs into call
=
Cutting corners in the name of “efficiency”

So you confirm a preconceived HPI

Fever✅
Cough✅
Leukocytosis✅
CXR infiltrate✅

Community acquired PNA✅

2/

Something doesn’t look right though.

Afebrile
Ill but nontoxic
HR 120s sinus despite fluid resuscitation
Profoundly tachypneic but SORA

Hm…

3/

More thorough hypothesis-driven hx taking reveals:

- Palpitations for 2d p/t illness
- Sx onset awoke him from sleep w pleurisy & dyspnea
- Primary sx now is pleurisy
- Exertional lightheadedness since onset
“Cough” - triggered by pleurisy
“Fever” - subjective diaphoresis

4/

You now have 2 conflicting PRs:

1. Acute onset fever cough dyspnea w leukocytosis & focal infiltrate

2. Acute onset pleurisy, dyspnea, exertional presyncope w persistent sinus tach

PR #2 sells a very different ddx than #1

Now you process illness scripts for PNA v PTE

5/

Cognitive error #2 — incomplete illness scripts

Most of us have a script for PTE which emphasizes HR, hemoptysis, dyspnea, DVT or risk factors (malignancy, immobilization, surgery, travel, etc)

Crackles?🚫
Infiltrates?🚫
Fever? 🚫
⬆️WBC? 🚫

Certainly on CAP script though

6/

Refine your script.

Patients w PTE can have:

Pleurisy (66%)
Tachypnea (54%)
Cough (30-40%)
Tachycardia (24%)
Rales (18-26%)
Pulm infiltrates (17%)
Fever (3%)
Leukocytosis (🤷🏻‍♂️%)

Hm. Sounds an awful lot like “sepsis” and PNA.

In fact, DVT on exam — flip of a coin

7/

Ultimately, CTA showed segmental filling defect c/f PTE.

Learning points:

- Be cautious - “efficiency” is a breeding ground for cognitive error
- Be curious & confirm nuanced hx
- Refine your illness scripts
- Master your PR

“A problem well stated is a problem half-solved”

Lastly, and most emphatically -

Sepsis is not an end-point diagnosis (s/o @PrathitKulkarni), nor is it specific for infection. Be curious.

https://twitter.com/prathitkulkarni/status/1557131592971214850

#MedTwitter #ClinicalReasoning #DiagnosticReasoning #CognitiveBias #CognitiveError @CPSolvers @DxRxEdu @rabihmgeha @rav7ks @PrathitKulkarni @medrants