Notice. Illustration and research prototype. Not clinically validated, not a medical device, not for patient care. Example data are fictional.
Live demonstration. The sliders and test cases work — every score recomputes in your browser. The data is generated, no patient is involved, and nothing here has been verified by a clinician.

pharmacology.yaml

Configuration · pharmacology.yaml · 316 lines · editable in place

PHARMACOLOGY — onset, duration, half-life, renal behaviour ============================================================================ Two problems this solves, both of which the other engines currently get wrong by omission.

1. ATTRIBUTION The trajectory engine treats every movement as if the patient were untreated. A MAP falling while noradrenaline is being titrated up means something entirely different from a MAP falling on no support. And a urine output "improving" twenty minutes after furosemide is the drug, not the patient.

So a value moving inside a drug's onset-to-peak window is EXPECTED, and a value failing to move outside that window is a finding.

2. ACCUMULATION In renal failure, half-life is the number that matters and it is not the one in the formulary. Vancomycin runs about 6 hours normally and over 100 in anuria. The AKI engine already flags "review dose" — this makes it say when the drug will actually clear.

All of it is a table and a formula. No model.

Sources: standard pharmacology references; furosemide half-life and onset from StatPearls / LITFL; vancomycin renal accumulation from Kidney International. Values are typical adult figures and must be checked against the unit's own formulary before use.

meta
version0.1.0
ownerUNASSIGNED — set to the responsible clinician or pharmacist
verified bynot set
note
Typical adult values. Critical illness alters volume of distribution,
protein binding and clearance in ways these single numbers do not capture.
Treat as a prompt to think, not a dosing authority.
renal model
reference crcl100
minimum crcl5
note
A simplification of the Dettli method. It gives the right order of
magnitude and the right direction; it is not a dosing calculation.
drugs
- namenoradrenaline
---- VASOACTIVE -------------------------------------------------------- Very short half-lives: the trajectory question is whether the patient is responding, not whether the drug will accumulate.
aliases[norepinephrine, levophed]
classvasopressor
onset min1
peak min2
duration min5
half life min2.5
renal fraction0.0
affects[map, sbp, dbp, heart_rate, shock_index]
expected direction{map: up, sbp: up, shock_index: down}
no response after min15
note
Acts within a minute or two. A MAP that has not responded within fifteen
minutes of an increase is a finding, not a waiting game.
- nameadrenaline
aliases[epinephrine]
classvasopressor
onset min1
peak min2
duration min10
half life min3
renal fraction0.0
affects[map, sbp, heart_rate, lactate]
expected direction{map: up, heart_rate: up, lactate: up}
note
Raises lactate through beta-2 stimulation without tissue hypoperfusion.
A rising lactate on adrenaline may be the drug, not worsening shock —
the single most misread trend in this table.
- namevasopressin
aliases[argipressin]
classvasopressor
onset min5
peak min15
duration min30
half life min20
renal fraction0.0
affects[map, sbp, urine_ml_kg_hr]
expected direction{map: up}
- namedobutamine
classinotrope
onset min2
peak min10
duration min10
half life min2
renal fraction0.0
affects[heart_rate, map, pulse_pressure]
expected direction{heart_rate: up, pulse_pressure: up}
- namefurosemide
---- DIURETIC ----------------------------------------------------------
aliases[frusemide, lasix]
classdiuretic
onset min5
peak min30
duration min360
half life min120
renal fraction0.85
affects[urine_ml_kg_hr, map]
expected direction{urine_ml_kg_hr: up}
no response after min120
note
IV onset 5 minutes, peak 30, effect 6 hours. Urine output rising within
that window is the drug. Half-life extends substantially in renal
impairment, and no diuretic response at two hours in AKI is itself
information.
- namepropofol
---- SEDATION AND ANALGESIA --------------------------------------------
classsedative
onset min1
peak min3
duration min10
half life min40
renal fraction0.0
affects[map, gcs, heart_rate]
expected direction{map: down, gcs: down}
note
Causes hypotension directly. A falling MAP after a propofol increase is
expected and should not read as deterioration.
- namemidazolam
classsedative
onset min2
peak min5
duration min120
half life min180
renal fraction0.5
affects[gcs, map, resp_rate]
expected direction{gcs: down}
accumulatestrue
note
Active metabolite accumulates in renal impairment. Prolonged sedation
after stopping is frequently this rather than a neurological event.
- namemorphine
classopioid
onset min5
peak min20
duration min240
half life min180
renal fraction0.6
affects[resp_rate, gcs, heart_rate]
expected direction{resp_rate: down, gcs: down}
accumulatestrue
note
Morphine-6-glucuronide accumulates in renal failure and is more potent
than the parent. Fentanyl is preferred where clearance is poor.
- namefentanyl
classopioid
onset min2
peak min5
duration min45
half life min200
renal fraction0.1
affects[resp_rate, gcs]
expected direction{resp_rate: down}
note
Context-sensitive half-time lengthens markedly with prolonged infusion —
the single figure understates it after days of use.
- namevancomycin
---- ANTIMICROBIAL -----------------------------------------------------
classantibiotic
onset min60
peak min120
duration min720
half life min360
renal fraction0.9
affects[]
accumulatestrue
monitorTrough level before the fourth dose
note
Half-life approximately 6 hours normally, over 100 hours in anuria.
This is the clearest case for computing adjusted half-life rather than
reading the formulary number.
- namepiperacillin-tazobactam
aliases[pip-tazo, tazocin]
classantibiotic
onset min30
peak min60
duration min480
half life min60
renal fraction0.7
affects[]
- namemeropenem
classantibiotic
onset min30
peak min60
duration min480
half life min60
renal fraction0.7
affects[]
- namegentamicin
classantibiotic
onset min30
peak min60
duration min480
half life min150
renal fraction0.95
affects[]
accumulatestrue
monitorTrough level; extend the interval rather than reduce the dose
- nameenoxaparin
---- ANTICOAGULANT -----------------------------------------------------
aliases[clexane, lovenox]
classanticoagulant
onset min180
peak min240
duration min720
half life min270
renal fraction0.8
affects[]
accumulatestrue
reversalProtamine, partial only — approximately 60% of anti-Xa activity
note
Accumulates below 30 mL/min creatinine clearance, and reversal is
incomplete. Unfractionated heparin is preferred where clearance is poor
or bleeding risk is high.
- nameheparin
aliases[unfractionated-heparin, ufh]
classanticoagulant
onset min1
peak min5
duration min120
half life min90
renal fraction0.1
affects[]
reversalProtamine, complete
- nameparacetamol
---- ANTIPYRETIC -------------------------------------------------------
aliases[acetaminophen]
classantipyretic
onset min30
peak min60
duration min300
half life min150
renal fraction0.05
affects[temperature, map]
expected direction{temperature: down}
note
A temperature trend within 4-6 hours of a dose describes the drug, not
the patient. IV paracetamol also causes transient hypotension.
- namehydrocortisone
---- OTHER -------------------------------------------------------------
classsteroid
onset min60
peak min240
duration min480
half life min100
renal fraction0.1
affects[map]
expected direction{map: up}
- nameinsulin
classhormone
onset min5
peak min30
duration min120
half life min10
renal fraction0.3
affects[]
accumulatestrue
noteRequirements fall in renal failure — insulin is cleared renally.
attribution
expected window multiplier1.0
A value moving inside the onset-to-peak window of a drug that acts on it is expected, and should be labelled rather than flagged.
flag non responsetrue
Beyond no_response_after_min with no movement in the expected direction, the absence itself is the finding.
clearance half lives5
Half-life multiples until a drug is effectively gone.
accumulation warning ratio3.0
adjusted half-life this many times normal