Clinical Commander

All dossiers
pulm.atypical-pneumonia.v1

Atypical Pneumonia (Mycoplasma / Chlamydophila / Legionella / psittacosis / Q fever)

pulmonologysubacuteacuteadultoutpatientacuteinpatient
Start encounter →Print handoutPRODUCTION

Atypical CAP = canonical PATHOGEN-DISCRIMINATION engine. §5.5.2 wired as data: epidemiologic exposure prior + extrapulmonary clues + targeted micro discriminate Mp vs C. pneumoniae vs Legionella vs psittacosis vs Q fever vs viral vs typical CAP vs TB vs eosinophilic/HP. Each pivot carries a discriminator + test characteristic + PMID across sibling_differentiation (6 blocks), severity_triggers, and DIFFERENTIAL/INITIAL_WORKUP phase logic. Cross-dossier routing via workups[].branches_to (atypical_pneumonia_pathogen_panel) + sibling_differentiation[].sibling_engine_id using real engine_ids: pulm.cap.core.v1, pulm.tuberculosis.v1, id.sepsis.core.v1, id.covid19.core.v1, id.influenza.core.v1, pulm.hypersensitivity_pneumonitis.v1 (6 engine_ids). Effect-size / test-characteristic numbers wired with PMIDs (≥5): Legionella urinary antigen pooled sens 0.79 / serogroup-1 0.86 / spec ~1.00 → very high LR+ (Kawasaki 34972680); Legionella PCR sens 83% spec 90% (Cristovam 28463665); FQ vs macrolide Legionella mortality OR 0.94 (0.71-1.25), 6.9% vs 7.4% (Jasper 32296816); MRMP prevalence >90% East-Asia / A2063G 81.7% / switch-failure 20.9% vs 0% / fever 6.5 vs 4.0 d (Zheng 42088748; Wang 39143259); atypical-coverage inpatient CAP all-cause mortality RR 1.13 (0.82-1.54) with Legionella clinical-success advantage (Shefet 16186469/15846713); multiplex PCR LoD 5-10 copies (Wagner 29397298); Mp ~25% extrapulmonary (Lofgren 34532621); CURB-65 AUROC 0.726 / PSI 0.879 (Iftikhar 40033304). 12 phases populated, none empty — incl. RED_FLAGS severe-Legionella, INITIAL_WORKUP targeted micro, DIFFERENTIAL pathogen panel, RISK_STRATIFICATION CURB-65/PSI, TREATMENT empiric→directed, BRANCHING/MONITORING resistance + extrapulmonary screen, FOLLOWUP public-health + chronic-Q-fever surveillance. Regimen axis atypical_pneumonia_empiric_to_directed: 5 steps (outpatient empiric → inpatient empiric → pathogen-directed Legionella → pathogen-directed zoonotic psittacosis/Q-fever → resistance/treatment-failure MRMP) + supportive non-stepwise. Severity × pregnancy (macrolide pathway, avoid doxy/FQ) × resistance-risk (MRMP→doxy/FQ) × comorbidity; renal/hepatic adjust; durations 3-5 d typical / 7-14 d Legionella / up to 21 d immunocompromised / 10-14 d zoonotic; escalation severe Legionella; de-escalation IV→PO + short course; corticosteroid only severe inpatient / refractory hyperinflammatory Mp (2025 ATS CAP). Special populations: immunocompromised (broaden differential, extend duration, early BAL), pregnancy (azithromycin — avoid doxycycline/FQ), pediatric pointer (Mp common school-age; MRMP East-Asia), elderly (severe Legionella risk), travel/exposure (psittacosis-birds, Q-fever-livestock/raw-dairy), outbreak/cluster (Legionella/psittacosis/Q-fever public-health reporting + environmental source investigation). RxCUI FIX: prior file used 7407 for "normal_saline" — RxNav (2026-05-16) resolves 7407 = NICOTINE; corrected to 9863 = sodium chloride. All regimen RxCUIs RxNav REST-confirmed 2026-05-16: azithromycin 18631, doxycycline 3640, levofloxacin 82122, moxifloxacin 139462, clarithromycin 21212, ceftriaxone 2193, oxygen 7806, prednisolone 8638, sodium chloride 9863. No hand-authored/unconfirmed CUIs. SCHEMA-GAP NOTES: (1) _types.ts has no first-class field for diagnostic LR / test-characteristic tables — encoded in severity_triggers, sibling_differentiation feature rows, phase purpose/advance_when, calculator guideline_basis, regimen rationale, and the .depth.md pathogen-discriminator LR table (authoritative payload); (2) RequiredCalculator.drives enum lacks "diagnostic_gate" — CURB-65 reuses disposition, qSOFA/SOFA reuse risk_stratification; (3) no pathogen-prior / epidemiologic-exposure field — captured in CONTEXT phase purpose + severity_triggers + .depth.md; (4) acuity widened to [subacute, acute] to reflect the true clinical frame (was [acute]); (5) no RegimenDrug field for resistance-mechanism (23S rRNA) — captured in trigger strings + rationale. PRODUCTION blockers unchanged per dispatch scope: (1) manifest stub kept (no manifest/package pointer change); (2) clinical-tools-registry import not added; (3) no engine-specific test file beyond shared dossier-contract.test.ts.

Entry points (5)

  • symptom
    Subacute dry cough, low-grade fever, headache, myalgia, sore throat — gradual onset (2025 ATS CAP; Mp/C.pneumoniae pattern)
    subacute_dry_cough_low_grade_fever
  • symptom
    Extrapulmonary features — rash/mucositis or hemolysis or neurologic (Mp ~25% extrapulmonary, Lofgren PMID 34532621); GI/encephalopathy/relative bradycardia (Legionella)
    extrapulmonary_features
  • imaging
    Patchy / interstitial infiltrate disproportionate to exam (2025 ATS CAP; PMID 40679934)
    patchy_interstitial_infiltrate
  • lab_abnormality
    Urinary Legionella antigen positive (serogroup-1 sens 0.86, spec ~1.00; Kawasaki PMID 34972680)
    urinary_legionella_antigen_positive
  • history
    Hot tub / cooling tower / cruise / hotel exposure or local Legionella cluster; bird exposure (psittacosis); livestock/parturient-animal exposure (Q fever)
    travel_water_exposure_outbreak

Required inputs (22)

  • agerequired
    demographic • used at CONTEXT
    Mp peaks school-age/young adults; Legionella severe disproportionately elderly/smokers; doxycycline avoided <8 yr — drives pathogen prior + drug selection (2025 ATS CAP PMID 40679934)
  • sbprequired
    vital • used at CONTEXT
    Severity / sepsis screen — hypotension is a severe-CAP minor criterion and Legionella severe-disease risk (2025 ATS CAP)
  • rrrequired
    vital • used at CONTEXT
    CURB-65 element + tachypnea severity (Lim Thorax 2003 PMID 12728155)
  • spo2required
    vital • used at CONTEXT
    Hypoxemia threshold for admission/ICU; disproportionate hypoxia favors atypical/viral (2025 ATS CAP)
  • temprequired
    vital • used at CONTEXT
    High fever ≥39 °C with relative bradycardia (Faget sign) raises Legionella suspicion
  • hr
    vital • used at CONTEXT
    Pulse-temperature dissociation (relative bradycardia) — Legionella, psittacosis, Q fever clue
  • confusion
    symptom • used at CONTEXT
    CURB-65 element + Legionella encephalopathy (disproportionate to pneumonia severity)
  • bird_or_livestock_exposure
    history • used at CONTEXT
    Psittacine/poultry exposure → Chlamydia psittaci; parturient livestock / unpasteurized dairy → Coxiella burnetii (Q fever) — changes empiric duration/agent
  • immunocompromiserequired
    history • used at CONTEXT
    Transplant/biologic/HIV/steroid → broader differential (PCP, Nocardia, Legionella longjbeachae), longer duration, lower threshold for BAL (2025 ATS CAP)
  • pregnancy
    history • used at CONTEXT
    Doxycycline and fluoroquinolone relatively contraindicated → macrolide pathway; psittacosis in pregnancy is severe (azithromycin)
  • sodium
    lab • used at INITIAL_WORKUP
    Hyponatremia (SIADH) is a classic Legionella discriminator vs typical CAP
  • creatininerequired
    lab • used at INITIAL_WORKUP
    eGFR for renal dosing of FQ/macrolide; renal failure is a Legionella severe-disease marker
  • lft
    lab • used at INITIAL_WORKUP
    Transaminitis common in Legionella, Q fever, psittacosis; baseline before doxy/macrolide/FQ
  • cpk
    lab • used at INITIAL_WORKUP
    Rhabdomyolysis association with Legionella (AKI risk)
  • crp_procalcitonin
    lab • used at INITIAL_WORKUP
    Markedly elevated CRP/PCT favors typical bacterial or Legionella over Mp/C.pneumoniae/viral; low PCT supports holding/narrowing (2025 ATS CAP)
  • urinary_legionella_antigen
    lab • used at INITIAL_WORKUP
    L. pneumophila serogroup-1: pooled sens 0.86 / spec ~1.00 → very high LR+ when positive but cannot exclude non-sg1 (Kawasaki PMID 34972680)
  • urinary_pneumococcal_antigen
    lab • used at INITIAL_WORKUP
    Positive supports typical pneumococcal disease and argues against pure-atypical pathway (2025 ATS CAP)
  • respiratory_multiplex_pcr
    lab • used at INITIAL_WORKUP
    Multiplex RT-PCR detects Mp / C.pneumoniae / Legionella / viruses (LoD 5-10 copies; Wagner PMID 29397298); also identifies 23S rRNA macrolide-resistance mutation where assayed
  • cold_agglutinins
    lab • used at BRANCHING_WORKUP
    Cold-agglutinin hemolysis supports Mycoplasma (insensitive ~50%, low specificity — supportive only, not a rule-in)
  • atypical_serology
    lab • used at BRANCHING_WORKUP
    Paired acute/convalescent serology for C. psittaci, Coxiella burnetii (phase II IgG/IgM), Mp where PCR unavailable — retrospective confirmation
  • cxrrequired
    imaging • used at INITIAL_WORKUP
    Patchy/interstitial vs lobar; multilobar disease is a Legionella severity + severe-CAP minor criterion (2025 ATS CAP)
  • current_medsrequired
    medication • used at CONTEXT
    QTc-prolonging meds before macrolide/FQ; statin + macrolide rhabdo; warfarin + macrolide/FQ INR interaction (2025 ATS CAP)

12-phase flow (12)

  1. 1FRAME
    Frame as atypical-CAP pathogen-discrimination pathway (Mp / C. pneumoniae / Legionella / C. psittaci / C. burnetii) when subacute pattern + interstitial CXR + exposure/extrapulmonary clues, OR a positive atypical test, in an adult (2025 ATS CAP PMID 40679934)
    advance: Atypical pattern or positive atypical test established; chronic/non-infective routed out
  2. 2ENTRY
    Trigger from subacute cough + low-grade fever, exposure history, extrapulmonary feature, or positive urinary antigen / multiplex PCR (2025 ATS CAP)
    inputs: age
    advance: Entry symptom / exposure / test documented
  3. 3CONTEXT
    Epidemiologic exposure history is the prior: water/travel/outbreak (Legionella), bird (psittacosis), parturient livestock/raw dairy (Q fever), school/family cluster + age (Mp), high-MRMP-prevalence region (East Asia); plus severity vitals, pulse-temp dissociation, comorbidity/immunocompromise, pregnancy, QTc/interactions (2025 ATS CAP; Yu)
    inputs: sbp, rr, spo2, temp, hr, confusion, bird_or_livestock_exposure, immunocompromise, pregnancy, current_meds
    advance: Exposure prior + severity inputs captured
  4. 4RED_FLAGS
    Severe Legionella: respiratory failure / septic shock / multilobar disease / encephalopathy / acute renal failure / rhabdomyolysis — escalate empirics + ICU; also screen sepsis (qSOFA) (2025 ATS CAP; route → id.sepsis.core.v1)
    inputs: sbp, spo2
    actions: calc.qsofa
    advance: No emergent escalation OR ICU/sepsis pathway started
  5. 5INITIAL_WORKUP
    Targeted microbiology: CXR; CBC; BMP (Na+, Cr); LFTs; CPK; CRP/PCT; urinary Legionella + pneumococcal antigen; respiratory multiplex PCR (Mp/C.pneumoniae/Legionella/viruses, LoD 5-10 copies — Wagner PMID 29397298); blood cultures if severe; sputum for Legionella culture on BCYE if hospitalised/severe. Interpret urinary Legionella antigen as serogroup-1 sens 0.86 / spec ~1.00 — positive = high LR+, negative does NOT exclude non-sg1 or non-pneumophila (Kawasaki PMID 34972680)
    inputs: cxr, creatinine, sodium, lft, cpk, crp_procalcitonin, urinary_legionella_antigen, urinary_pneumococcal_antigen, respiratory_multiplex_pcr
    actions: panel.cbc, panel.renal, panel.lft, panel.inflammation
    advance: Antigen + PCR + CXR + severity labs resulted or sent
  6. 6BRANCHING_WORKUP
    Pathogen-specific branches: cold agglutinins + haptoglobin/LDH if hemolysis (Mp, supportive only); paired serology for C. psittaci / Coxiella burnetii phase II / Mp; Legionella sputum culture on BCYE for genotyping (outbreak link); 23S rRNA mutation testing if Mp + macrolide failure (MRMP — Wang PMID 39143259); CT chest + bronchoscopy/BAL if poor response or immunocompromised
    inputs: cold_agglutinins, atypical_serology
    advance: Targeted tests obtained when triggered
  7. 7DIFFERENTIAL
    §5.5.2 pathogen discrimination AS DATA: Mp vs C. pneumoniae vs Legionella vs C. psittaci vs C. burnetii vs viral (influenza/SARS-CoV-2/RSV) vs typical bacterial CAP vs TB vs eosinophilic pneumonia / hypersensitivity pneumonitis — each pivot has discriminator + test characteristic (see sibling_differentiation + severity_triggers + workups.branches_to)
    actions: atypical_pneumonia_pathogen_panel
    advance: Working pathogen category assigned with discriminating evidence
  8. 8RISK_STRATIFICATION
    CURB-65 (Lim PMID 12728155) / PSI for disposition (PSI/CURB-65 outperform generic scores for in-hospital mortality — Iftikhar PMID 40033304); 2025 ATS severe-CAP criteria; SOFA/qSOFA in severe Legionella; flag Legionella severe-disease risk (multilobar, hyponatremia, AKI, immunocompromise, elderly)
    inputs: sbp, rr, confusion
    actions: calc.curb65, calc.sofa
    advance: Severity tier + Legionella-severe-risk assigned; disposition implied
  9. 9TREATMENT
    Empiric atypical-active therapy by severity: outpatient macrolide vs doxycycline vs respiratory FQ; inpatient β-lactam + macrolide OR respiratory FQ monotherapy; severe → β-lactam + macrolide/FQ + atypical coverage always (2025 ATS CAP). Pathogen-directed: Legionella → FQ or azithromycin 7-14 d (FQ vs macrolide no mortality difference, OR 0.94 — Jasper PMID 32296816); psittacosis/Q fever → doxycycline/tetracycline; resistance branch: macrolide-failure Mp → switch to doxycycline/tetracycline or FQ (MRMP). Add corticosteroids only in severe inpatient CAP (2025 ATS CAP)
    inputs: creatinine
    advance: Empiric/pathogen-directed regimen + duration + de-escalation plan documented
  10. 10DISPOSITION
    Outpatient if CURB-65 0-1 + SpO2 ≥92% + no immunocompromise/severe-Legionella features; ward if CURB-65 2 or hypoxia; ICU if Legionella with respiratory failure/shock/encephalopathy or ≥3 severe-CAP minor criteria (2025 ATS CAP)
    advance: Disposition decided
  11. 11MONITORING
    Clinical response at 48-72 h (defervescence slower in Legionella/psittacosis); resistance/treatment-failure workup if no improvement (consider MRMP, wrong pathogen, TB/fungal, empyema, malignancy, PE); LFTs, QTc, sodium normalisation; IV→PO when Halm-stable; extrapulmonary complication screen (Mp hemolysis/MIRM/encephalitis — Lofgren PMID 34532621; Legionella endocarditis/abscess; Q fever endocarditis)
    advance: Stability achieved or therapy escalated / diagnosis revised
  12. 12FOLLOWUP
    Public-health reporting for Legionella (notifiable; environmental source investigation if cluster) and psittacosis/Q fever (zoonotic notifiable); Q fever serology surveillance for chronic endocarditis in valvulopathy/immunocompromise; pneumococcal + influenza + COVID vaccination; smoking cessation; CXR follow-up at 6 wk if smoker / age ≥50
    advance: Reporting + prevention + chronic-complication surveillance complete