Smart NOA Web Simulator — sources registry

SIMULATION — fictional patients — educational use only. Not a medical device. Not medical advice. Every dose, range, ceiling and timing in data/drugs.js is an illustrative educational value transcribed from the cited source. None is an order or a recommendation for any real patient. Labels differ by manufacturer and are revised often; real doses are individualized by the treating clinician.

Compiled 2026-11-05. Machine-readable copy: data/drugs.js (SOURCES).

How entries were checked

DailyMed, Drugs@FDA, PubMed, doi.org and label mirrors such as drugs.com could not be opened directly from the build environment (direct page fetches failed DNS resolution). Every entry was therefore checked through web-search records of FDA-hosted label PDFs, manufacturer prescribing-information pages, label mirrors (medlibrary.org, prescriberpoint.com, ndclist.com, medthority.com, drugcentral.org) and publisher or indexing records. The verifiedVia line of each entry says exactly what was seen. Anything not seen is marked "(identifier unverified)" or left out.

Entries S1–S22 reuse citations and verification text from breast-reconstruction-anesthesia-protocol/DOSING_SOURCES.md, renumbered:

S23–S45 are new (S40–S43 added later on 2026-11-05 for epinephrine, the Eleveld propofol model and sepsis fluid context; S44–S45 added in the 2026-11 audit for perioperative anaphylaxis — see AUDIT_2026-11.md). S38 and S39 are non-label sources used only for nitrous oxide (see below). S35 is the repository's own config.yaml. It is cited only so the simulator matches the Python controller's software ceilings and lockouts. It is not clinical evidence.

Data conventions

• Defaults: where the label names a usual, average or initial dose, that is the default. Otherwise the default is the lower bound of the stated range. Where a source gives only an upper limit (Foo 2021 lidocaine, Schwenk 2018 ketamine), min is 0 and the default is the limit. The ketamine maintenance default (0.2 mg/kg/h) is the Python simulation ceiling from S35, not a guideline value.
• Weight scalars: scalarBasis: 'guideline' for lidocaine (IBW, Foo 2021). 'label' for sugammadex (actual body weight, Bridion label). 'convention' (TBW) for propofol, etomidate, ketamine, fentanyl, rocuronium, succinylcholine, neostigmine and dexmedetomidine: their labels give mg/kg (or mcg/kg) doses, but this session found no label statement saying which weight to use.
• Ceilings: basis: 'label maximum' / 'guideline maximum' only where the source states a maximum: neostigmine, phenylephrine, ephedrine, ketorolac, acetaminophen, lidocaine and ketamine. 'study protocol value' (nitrous oxide 70%, from ENIGMA-II) marks a trial concentration, not a label maximum. 'label range upper bound' (propofol 200 mcg/kg/min, dexmedetomidine 1 mcg/kg/h, sevoflurane 3%, desflurane 8.5%, isoflurane 3.5% (2.5% maintenance bound plus the label's extra 1.0% with oxygen alone), sugammadex 16 mg/kg) is a simulator hard stop at the top of a labelled range that the label does not call a maximum.
• Timing: numbers come only from label statements. Where the label gives a range, the midpoint is used and the range is kept in timing.ranges. Where no statement was verified, the numbers are null and basis is 'not stated in label'. The engine then falls back to its own illustrative constants. This applies to dexmedetomidine, sevoflurane, desflurane, isoflurane, lidocaine, sugammadex, glycopyrrolate, dexamethasone, ondansetron, phenylephrine, ephedrine, ketorolac and acetaminophen.
• Severity versus label wording: contraindication messages quote or closely paraphrase the label/guideline text. Where the simulator's severity differs from the label, the message says so:
  • Dexmedetomidine with heart block is a LOCK, mirroring the controller. The label only says "use with caution".
  • Ketamine with uncontrolled hypertension is a LOCK (2026-11 audit), matching the label CONTRAINDICATION ("patients for whom a significant elevation of blood pressure would constitute a serious hazard"); applying it to the simulator's "uncontrolled hypertension" condition is a simulator mapping. (SPEC §4 originally listed this as WARN.)
  • Lidocaine with heart block (2nd/3rd degree, no pacemaker) or WPW is a LOCK (2026-11 audit), matching the IV lidocaine label CONTRAINDICATIONS (Stokes-Adams syndrome, WPW, severe sinoatrial/atrioventricular/intraventricular block). (SPEC §4 originally listed heart block as WARN.)
  • Ketorolac "prophylactic analgesic before any major surgery" (CONTRAINDICATIONS) and "peri-operative use should be avoided ... when hemostasis is critical" (WARNINGS) are runtime warnings keyed to the surgery timeline (before incision / during surgery) and shown at plan time for premed/induction/sedation and maintenance steps: the simulator cannot know whether surgery is "major" or whether haemostasis is critical. Bleeding-risk and peptic-ulcer/GI-bleed conditions are LOCKs.
  • Phenylephrine 200 mcg: the 2019 Biorphen label says "not to exceed a total dosage of 200 mcg"; some later copies say "additional boluses up to 200 mcg". The simulator applies the conservative reading: all boluses in the session together ≤ 200 mcg (scope: 'cumulative boluses'); the 200 mcg/min infusion limit is separate.
  • Ketorolac below eGFR 30: the label says "advanced renal impairment", and the numeric cut-off comes from S35.
  • Succinylcholine with pre-existing hyperkalaemia is a LOCK by simulator policy. The label's exact wording on this point was not retrieved.
  • Etomidate with sepsis and ketamine with psychosis history are policy warnings. The quoted label text does not name these conditions.
• when grammar: contraindications take a single token: condition:<id>, allergy:<id>, medication:<id>, egfr<N, age>=N or flag:<id>. adjust[].when may join several tokens with | (OR) and may also use asa>=N and weight<N. tests/drugs.test.mjs enforces this.

Registry

S1

Foo I, Macfarlane AJR, Srivastava D, et al. The use of intravenous lidocaine for postoperative pain and recovery: international consensus statement on efficacy and safety. Anaesthesia. 2021;76(2):238-250. https://read.qxmd.com/read/33141959
Identifiers: doi:10.1111/anae.15270; PMID 33141959

• Supports: Lidocaine loading dose ≤1.5 mg/kg over about 10 min; infusion ≤1.5 mg/kg/h; dosing on ideal body weight; hourly ceiling 120 mg/h for any patient; not for patients under 40 kg; not given concurrently with other local-anaesthetic interventions (nerve blocks, wound infiltration) or while their effect is still active; about 4 h separation between a block and IV lidocaine. IV lidocaine for analgesia is not an FDA-labelled indication.
• Used by: lidocaine
• Verified via: Copied from DOSING_SOURCES.md S1: WebSearch record of the abstract and full-text excerpts (Nottingham and Glasgow repository copies; read.qxmd.com/read/33141959), 2026-11-05. (date 2026-11-05)

S2

Gan TJ, Belani KG, Bergese S, et al. Fourth consensus guidelines for the management of postoperative nausea and vomiting. Anesth Analg. 2020;131(2):411-448. https://pubmed.ncbi.nlm.nih.gov/32467512/
Identifiers: doi:10.1213/ANE.0000000000004833; PMID 32467512

• Supports: Dexamethasone 4–8 mg IV at induction; ondansetron 4 mg IV at end of surgery; rescue with a different drug class.
• Used by: dexamethasone, ondansetron
• Verified via: Copied from DOSING_SOURCES.md S7: WebSearch records of PubMed 32467512 and of the APSF podcast notes and the Meyer TSHP handout summarizing Table 4, 2026-11-05. Superseded in part by S3. (date 2026-11-05)

S3

Gan TJ, et al. Fifth consensus guidelines for the management of postoperative nausea and vomiting: executive summary. Anesth Analg. Published online November 14, 2025. https://journals.lww.com/anesthesia-analgesia/
Identifiers: doi:10.1213/ANE.0000000000007816; PMID unverified
Type: consensus guideline

• Supports: Two interventions for adults at risk of PONV; dexamethasone 4–8 mg IV at induction (8 mg may be optimal); ondansetron 4 mg IV at end of surgery; dexamethasone plus a 5-HT3 antagonist as the best-studied pair. Baseline risk reduction includes avoidance of nitrous oxide during maintenance anaesthesia (with opioid-sparing analgesia, propofol TIVA where appropriate, regional anaesthesia, hydration).
• Used by: nitrous_oxide, dexamethasone, ondansetron
• Verified via: Copied from DOSING_SOURCES.md S8: WebSearch records of journals.lww.com (executive summary landing page), the LWW-hosted full report (cdn-links.lww.com/permalink/aa/f/aa_1_1_2025_11_24_cd_1_sdc1.pdf), scholars.duke.edu and a NYSORA summary, 2026-11-05. Full text not opened. PMID (identifier unverified). The nitrous-oxide baseline-risk-reduction item was confirmed only in a secondary summary (WebSearch record, nysora.com / dovepress review), not in the guideline text; secondary summaries add that the benefit of avoidance is duration dependent (Peyton & Wu meta-analysis cited by the consensus: NNT 9 for >2 h, 128 for <1 h). (date 2026-11-05)

S4

Apfel CC, Läärä E, Koivuranta M, Greim CA, Roewer N. A simplified risk score for predicting postoperative nausea and vomiting: conclusions from cross-validations between two centers. Anesthesiology. 1999;91(3):693-700. https://pubmed.ncbi.nlm.nih.gov/10485781/
Identifiers: doi:10.1097/00000542-199909000-00022; PMID 10485781

• Supports: Four risk factors (female sex, non-smoking, history of PONV or motion sickness, postoperative opioids), one point each; estimated PONV incidence for 0–4 factors about 10%, 21%, 39%, 61%, 79%.
• Used by: engine/UI only
• Verified via: Copied from DOSING_SOURCES.md S9: citation and percentages as audited in v2.0 of that repository against the PubMed record; not re-searched in this session (2026-11-05). (date 2026-11-05)

S5

Thilen SR, Weigel WA, Todd MM, et al. 2023 American Society of Anesthesiologists practice guidelines for monitoring and antagonism of neuromuscular blockade. Anesthesiology. 2023;138(1):13-41. https://pubmed.ncbi.nlm.nih.gov/36520073/
Identifiers: doi:10.1097/ALN.0000000000004379; PMID 36520073

• Supports: Quantitative monitoring; confirm TOFR ≥0.9 before extubation; sugammadex over neostigmine for deep, moderate and shallow rocuronium/vecuronium block; neostigmine reasonable at minimal depth (TOFR 0.4 to <0.9).
• Used by: sugammadex, neostigmine
• Verified via: Copied from DOSING_SOURCES.md S10: WebSearch record of PubMed 36520073 and the APSF summary, 2026-11-05. (date 2026-11-05)

S6

Sugammadex injection (BRIDION). Merck Sharp & Dohme LLC. FDA-approved prescribing information, revised 11/2022. https://www.accessdata.fda.gov/drugsatfda_docs/label/2022/022225Orig1s012lbl.pdf
Identifiers: NDA 022225

• Supports: 2 mg/kg at reappearance of T2; 4 mg/kg at 1–2 post-tetanic counts with no TOF twitch; 16 mg/kg for immediate reversal about 3 min after rocuronium 1.2 mg/kg; dosing on actual body weight. Contraindication: known hypersensitivity to sugammadex or any of its components. DESCRIPTION: sugammadex sodium molecular weight 2178.01; each mL contains 100 mg sugammadex, equivalent to 108.8 mg sugammadex sodium (used for the 1:1 molar binding).
• Used by: sugammadex
• Verified via: Dosing copied from DOSING_SOURCES.md S11 (WebSearch record of the FDA-hosted 2022 label, the Merck PI and the Medsafe datasheet, 2026-11-05). Contraindication wording "Known hypersensitivity to sugammadex or any of its components" confirmed in a WebSearch record of the US label, 2026-11-05. Median recovery times were found only in non-US datasheets and are not used. 2026-11 audit: molecular weight 2178.01 and the 100 mg sugammadex = 108.8 mg sugammadex sodium equivalence confirmed in WebSearch records of medlibrary.org Bridion label copies (US label text), the Dutch and Singapore registry entries (108.8 mg/mL), PharmaCompass/NCATS (2178.0 g/mol, C72H104Na8O48S8). The 2/4/16 mg/kg depth wording is used for a runtime dose-vs-depth warning. (date 2026-11-05)

S7

Neostigmine methylsulfate injection (BLOXIVERZ). Exela Pharma Sciences. FDA-approved prescribing information, 2023 revision. https://www.accessdata.fda.gov/drugsatfda_docs/label/2023/204078Orig1s014lbl.pdf
Identifiers: NDA 204078

• Supports: 0.03–0.07 mg/kg IV (0.03 mg/kg for shorter half-life agents such as rocuronium); maximum total 0.07 mg/kg or 5 mg, whichever is less; anticholinergic before or with neostigmine; TOF0.9 generally within 10–20 min. Contraindications: hypersensitivity to neostigmine; peritonitis or mechanical obstruction of the intestinal or urinary tract.
• Used by: neostigmine
• Verified via: Dosing copied from DOSING_SOURCES.md S12 (WebSearch record of the FDA-hosted 2023 and 2015 labels and the Amneal PI, 2026-11-05). Contraindications and the "TOF0.9 within 10 to 20 minutes" statement confirmed in a WebSearch record of the FDA-hosted 2023 label highlights and drugs.com/medlibrary copies, 2026-11-05. (date 2026-11-05)

S8

Glycopyrrolate injection, USP. Multiple manufacturers (label copies indexed at medlibrary.org/lib/rx/meds/glycopyrrolate-54/). FDA-approved prescribing information (revision dates vary). https://medlibrary.org/lib/rx/meds/glycopyrrolate-54/
Identifiers: none (multiple ANDAs)

• Supports: With neostigmine reversal, 0.2 mg glycopyrrolate per 1 mg neostigmine. Intraoperative use: 0.1 mg IV, repeated as needed at 2–3 min intervals.
• Used by: glycopyrrolate
• Verified via: Copied from DOSING_SOURCES.md S13: WebSearch record of several label copies (medlibrary.org; ndclist.com), 2026-11-05. The 0.1 mg wording was confirmed in one manufacturer's label but not tied to its section heading. (date 2026-11-05)

S9

Dexmedetomidine hydrochloride injection (PRECEDEX). Hospira, Inc. (Pfizer). FDA-approved prescribing information, 2026 revision. https://www.accessdata.fda.gov/drugsatfda_docs/label/2026/021038s061lbl.pdf
Identifiers: NDA 021038

• Supports: Procedural sedation of non-intubated patients: loading 1 mcg/kg over 10 min (0.5 mcg/kg over 10 min for less invasive procedures); maintenance started at 0.6 mcg/kg/h, titrated 0.2–1 mcg/kg/h. ICU maintenance 0.2–0.7 mcg/kg/h. >65 y: loading 0.5 mcg/kg over 10 min, consider reducing maintenance. Hepatic impairment: consider reduction. Contraindications: none. Warning: use with caution in advanced heart block and/or severe ventricular dysfunction. Use as a general-anaesthesia adjunct is off-label.
• Used by: dexmedetomidine (including the sedation_minimal regimen added 2026-11 — same label procedural regimen; the label does not define sedation levels); MAC — dexmedetomidine template note
• Verified via: Dosing copied from DOSING_SOURCES.md S14 (WebSearch record of the FDA-hosted 2026 and 2013 labels and Pfizer labeling pages, 2026-11-05; procedural 0.6 and 0.2–1 mcg/kg/h wording confirmed in the 2013 text). Contraindications "None" and the heart-block caution confirmed in a WebSearch record of the DailyMed Precedex listing and pfizermedicalinformation.com/precedex/warnings, 2026-11-05. No onset time was found. Addition (2026-11-05): in the label's procedural-sedation studies the maintenance infusion was titrated 0.2–1 mcg/kg/h to a target sedation score (Ramsay ≥ 2) and rescue midazolam was permitted to achieve/maintain the target (WebSearch records of medlibrary.org Precedex label copies and pfizermedical.com/precedex/clinical-studies). No label sentence saying dexmedetomidine alone may not reach deep sedation was found, so the simulator does not make that claim. (date 2026-11-05)

S10

Schwenk ES, Viscusi ER, Buvanendran A, et al. Consensus guidelines on the use of intravenous ketamine infusions for acute pain management from the American Society of Regional Anesthesia and Pain Medicine, the American Academy of Pain Medicine, and the American Society of Anesthesiologists. Reg Anesth Pain Med. 2018;43(5):456-466. https://pmc.ncbi.nlm.nih.gov/articles/PMC6023582/
Identifiers: doi:10.1097/AAP.0000000000000806; PMID 29870457

• Supports: Subanaesthetic ketamine for acute pain: bolus ≤0.35 mg/kg; infusions up to about 1 mg/kg/h; contraindications for acute pain largely mirror those for chronic pain. Analgesic ketamine is outside the FDA-labelled anaesthetic indications.
• Used by: ketamine
• Verified via: Copied from DOSING_SOURCES.md S15: WebSearch record of PubMed 29870457, PMC6023582 and the ASRA guideline page, 2026-11-05. The 0.35 mg/kg and 1 mg/kg/h figures were confirmed in a secondary summary. The specific contraindication list was NOT retrieved in this session. (date 2026-11-05)

S11

Propofol injectable emulsion (DIPRIVAN). Fresenius Kabi USA. FDA-approved prescribing information, 2022 revision; and generic propofol injectable emulsion labels (medlibrary.org copies). https://www.accessdata.fda.gov/drugsatfda_docs/Label/2022/019627s069lbl.pdf
Identifiers: NDA 019627

• Supports: Pain on injection may be reduced by using the larger veins of the forearm or antecubital fossa and by prior injection of IV lidocaine (1 mL of a 1% solution); lidocaine is given before propofol or added immediately before use (not exceeding 20 mg lidocaine per 200 mg propofol). DESCRIPTION: isotonic, pH 7 to 8.5. Induction <55 y about 2–2.5 mg/kg; elderly/debilitated/ASA III–IV about 1–1.5 mg/kg. Maintenance 100–200 mcg/kg/min (<55 y); 50–100 mcg/kg/min (elderly/ASA III–IV). MAC sedation initiation: 100–150 mcg/kg/min for 3–5 min or slow injection about 0.5 mg/kg over 3–5 min; slow infusion/injection preferable to rapid bolus; maintenance 25–75 mcg/kg/min during the first 10–15 min, then 25–50 mcg/kg/min; no rapid bolus for MAC sedation in elderly/debilitated/ASA III–IV. Onset: hypnosis usually within 40 s from start of injection. Contraindication: known hypersensitivity to propofol, egg or soybean (current label).
• Used by: propofol; lidocaine pretreatment regimen (10 mg = 1 mL of 1%, arithmetic); propofol injection-pain teaching note
• Verified via: Induction/maintenance copied from DOSING_SOURCES.md S16 (WebSearch record of the FDA-hosted 2022 label and the medlibrary.org label copy, 2026-11-05). MAC-sedation rates, slow-injection 0.5 mg/kg and the elderly no-rapid-bolus wording confirmed in WebSearch records of medlibrary.org propofol and Diprivan label copies (medlibrary.org/lib/rx/meds/propofol/, /diprivan/page/2/), 2026-11-05. 40 s onset confirmed in a WebSearch record of an ndclist.com label copy and the FDA-hosted 2014 label (019627s062). Contraindication wording from a WebSearch record of the current label highlights; older labels word it as anaphylaxis to eggs/egg products/soybeans/soy products. Additions (2026-11-05): the injection-pain sentences and the "isotonic and has a pH of 7 to 8.5" DESCRIPTION wording confirmed in WebSearch records of medlibrary.org Diprivan label copies, drugs.com/pro/diprivan, guidelinecentral.com Diprivan listings and an FDA-archived Diprivan PI (webarchive.library.unt.edu). (date 2026-11-05)

S12

Rocuronium bromide injection. Generic manufacturer labels (e.g., DailyMed setid a54a79f7-5207-43d8-a139-7d57557caa50), Pfizer prescribing information, and FDA-hosted 2026 rocuronium label (NDA 217472). https://www.accessdata.fda.gov/drugsatfda_docs/nda/2026/217472Orig1s000Lbl.pdf
Identifiers: NDA 217472 (2026 label); others vary

• Supports: Intubating dose 0.6 mg/kg; rapid sequence 0.6–1.2 mg/kg; maintenance guided by response (older labels: 0.1, 0.15 or 0.2 mg/kg at 25% recovery of T1). 0.6 mg/kg: maximum block in most patients in <3 min; 31 (15–85) min clinical relaxation. Contraindication (older label): hypersensitivity (e.g., anaphylaxis) to rocuronium bromide or other neuromuscular blocking agents. DESCRIPTION: C32H53BrN2O4, molecular weight 609.70 (used for the 1:1 molar sugammadex binding).
• Used by: rocuronium
• Verified via: Dosing copied from DOSING_SOURCES.md S17 (WebSearch record of the Pfizer dosage page, the DailyMed listing and an FDA-hosted 2026 label, 2026-11-05). Timing ("less than 3 minutes", "31 (15-85) minutes") from a WebSearch record of the FDA-hosted 2026 label; contraindication wording from an older label copy (ndclist.com / labeling.pfizer.com), 2026-11-05. 2026-11 audit: molecular weight 609.70 confirmed in WebSearch records of the Pfizer rocuronium label description page (pfizermedicalinformation.com) and a DrugCentral label copy. (date 2026-11-05)

S13

Fentanyl citrate injection, USP. Hospira, Inc. (Pfizer) and Fresenius Kabi (August 2023 PI) prescribing information. https://pfizermedicalinformation.com/fentanyl-citrate/dosage-admin
Identifiers: none (multiple NDAs/ANDAs)

• Supports: Individualized dosing; reduce initial dose in elderly/debilitated. General anaesthesia low dose 2 mcg/kg; moderate dose 2–20 mcg/kg with 25–100 mcg increments for maintenance. IV onset almost immediate; analgesia 30–60 min after a single IV dose up to 100 mcg. PRECAUTIONS: the initial dose should be appropriately reduced in elderly and debilitated patients; the effect of the initial dose should be considered in determining incremental doses (no figure given).
• Used by: fentanyl
• Verified via: Low-dose tier copied from DOSING_SOURCES.md S18 (WebSearch record of the Pfizer dosage page and Fresenius/medlibrary copies, 2026-11-05). Moderate-dose "2 mcg/kg to 20 mcg/kg, 25 mcg to 100 mcg" confirmed in a WebSearch record of the Fresenius Kabi August 2023 PI and ndclist.com label copies; the maintenance sentence itself was cut off in the snippet. Onset/duration from an archived (2008–2009) DailyMed label copy (webarchive.library.unt.edu), 2026-11-05. 2026-11 audit: PRECAUTIONS wording confirmed in WebSearch records of an archived DailyMed fentanyl citrate label (webarchive.library.unt.edu) and a Patterson Dental label PDF. A 25–50 mcg reduced-dose figure seen on one Pfizer-hosted page was not tied to the induction context and is not used. (date 2026-11-05)

S14

Naloxone hydrochloride injection, USP. Generic manufacturer labels (copies indexed at medlibrary.org/lib/rx/meds/naloxone-hydrochloride-14/; labeling.pfizer.com). https://labeling.pfizer.com/ShowLabeling.aspx?id=4542
Identifiers: none (multiple ANDAs)

• Supports: Postoperative opioid depression: 0.1–0.2 mg IV at 2–3 min intervals to desired reversal; repeat doses may be needed at 1–2 h intervals. IV onset generally apparent within 2 min; duration depends on dose and route.
• Used by: naloxone
• Verified via: Dosing copied from DOSING_SOURCES.md S22 (WebSearch record of multiple label copies, 2026-11-05). Onset "generally apparent within two minutes" confirmed in WebSearch records of ndclist.com and labeling.pfizer.com label copies, 2026-11-05. (date 2026-11-05)

S15

Ketorolac tromethamine injection, USP. Hospira, Inc. (Pfizer) prescribing information. https://www.pfizermedical.com/ketorolac-inj/dosage-admin
Identifiers: none

• Supports: <65 y: 30 mg IV every 6 h, max 120 mg/day; ≥65 y, renally impaired or <50 kg: 15 mg every 6 h, max 60 mg/day; ≤5 days. Contraindicated in advanced renal impairment or risk for renal failure due to volume depletion; in patients with allergic reactions to aspirin or other NSAIDs; in patients currently receiving aspirin or NSAIDs. Also contraindicated: as a prophylactic analgesic before any major surgery; in patients with suspected or confirmed cerebrovascular bleeding, hemorrhagic diathesis, incomplete hemostasis and those at high risk of bleeding (inhibits platelet function); active peptic ulcer disease, recent gastrointestinal bleeding or perforation; in the setting of CABG surgery. WARNINGS (hemorrhage): peri-operative use should be avoided, and postoperative use undertaken with caution, when hemostasis is critical.
• Used by: ketorolac
• Verified via: Dosing copied from DOSING_SOURCES.md S23 (WebSearch record of the Pfizer dosage page and medlibrary.org copies, 2026-11-05). Contraindication wording confirmed in a WebSearch record of pfizermedicalinformation.com/ketorolac-inj/contradictions and the Hospira DailyMed listing, 2026-11-05. 2026-11 audit: WebSearch records of pfizermedicalinformation.com/ketorolac-inj/contradictions (cerebrovascular bleeding / hemorrhagic diathesis / incomplete hemostasis / high risk of bleeding, quoted), medlibrary.org and Searle ketorolac label copies ("CONTRAINDICATED as prophylactic analgesic before any major surgery"; peptic ulcer, GI bleeding/perforation; CABG; "peri-operative use ... should be avoided and postoperative use be undertaken with caution when hemostasis is critical"). The exact phrase "intraoperatively when hemostasis is critical" was seen only in the Canadian monograph, so the US WARNINGS wording is used, as a runtime warning. (date 2026-11-05)

S16

Acetaminophen injection (OFIRMEV). Mallinckrodt Hospital Products. FDA-approved prescribing information. https://www.accessdata.fda.gov/drugsatfda_docs/label/2010/022450lbl.pdf
Identifiers: NDA 022450

• Supports: Adults ≥50 kg: 1000 mg every 6 h or 650 mg every 4 h, max single dose 1000 mg, max 4000 mg/day; <50 kg: 15 mg/kg every 6 h or 12.5 mg/kg every 4 h, maximum single dose 15 mg/kg up to 750 mg, maximum 75 mg/kg per 24 h up to 3750 mg; daily maximum counts all routes. Contraindicated in severe hepatic impairment; caution in hepatic impairment.
• Used by: acetaminophen
• Verified via: Copied from DOSING_SOURCES.md S24: WebSearch record of the FDA-hosted 2010 label, the Mallinckrodt PI and drugs.com label copy, 2026-11-05. 2026-11 audit: the <50 kg absolute caps (15 mg/kg up to 750 mg per dose; 75 mg/kg up to 3750 mg per 24 h) confirmed in WebSearch records of the FDA-hosted 2010 label (accessdata 022450lbl.pdf), drugs.com/pro/ofirmev and RxList. (date 2026-11-05)

S17

Midazolam hydrochloride injection, USP. Multiple manufacturers (e.g., DailyMed setid e22611dd-1830-40af-9139-7cdb07f5a426). https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=e22611dd-1830-40af-9139-7cdb07f5a426
Identifiers: none (multiple ANDAs)

• Supports: Individualized; IV sedation in a healthy adult starts with as little as 1 mg and should not exceed 2.5 mg initial dose, titrated slowly; reduce in older, debilitated or chronically ill patients and with opioids/CNS depressants. Sedation achieved within 3–5 min after IV injection. Contraindicated in acute narrow-angle glaucoma. Boxed warning on respiratory depression. Patients 60 years or older, debilitated or chronically ill: initial IV sedation dose no more than 1.5 mg over no less than 2 minutes; wait an additional 2 or more minutes to evaluate; further titration about 1 mg per 2 minutes; total doses greater than 3.5 mg are not usually necessary. Healthy adults under 60: no more than 2.5 mg over at least 2 minutes.
• Used by: midazolam
• Verified via: Dosing copied from DOSING_SOURCES.md S27 (WebSearch record of a prescriberpoint.com label copy and the DailyMed listing, 2026-11-05; age ≥60 figure confirmed in the 2026-11 audit). Onset "within 3 to 5 minutes" and the narrow-angle glaucoma contraindication confirmed in WebSearch records of US label copies (DailyMed listings; medlibrary.org), 2026-11-05. 2026-11 audit: the ≥60 y / debilitated / chronically ill tier (≤1.5 mg over ≥2 min; wait ≥2 min; ~1 mg per 2 min; >3.5 mg not usually necessary) and the healthy-adult ≤2.5 mg over ≥2 min wording confirmed in a WebSearch record of pfizermedicalinformation.com/midazolam/dosage-admin (US label text). UK SmPCs differ (0.5–1 mg in older patients) and are not used. (date 2026-11-05)

S18

Phenylephrine hydrochloride injection (BIORPHEN). FDA-approved prescribing information, 2019. https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/212909s000lbl.pdf
Identifiers: NDA 212909

• Supports: Hypotension during anaesthesia: initial IV bolus 40–100 mcg; additional boluses every 1–2 min as needed, not to exceed a total dosage of 200 mcg (2019 label wording; some later label copies read "additional boluses up to 200 mcg" — the simulator applies the conservative total-dosage reading); infusion 10–35 mcg/min, not exceeding 200 mcg/min.
• Used by: phenylephrine
• Verified via: Copied from DOSING_SOURCES.md S28: WebSearch record of the FDA-hosted 2019 label and drugs.com label copy, 2026-11-05. 2026-11 audit: WebSearch records of the FDA-hosted 2019 label (212909s000lbl.pdf; "Additional boluses may be administered every 1-2 minutes as needed; not to exceed a total dosage of 200 mcg"), ndclist.com Eton label copies (same wording) and drugs.com/pro/biorphen ("additional boluses up to 200 mcg"). Per-bolus wording on the current US label could not be verified, so the 200 mcg limit is applied to all phenylephrine boluses in the session combined. (date 2026-11-05)

S19

Ephedrine sulfate injection (AKOVAZ). Exela Pharma Sciences. FDA-approved prescribing information. https://www.drugs.com/pro/akovaz.html
Identifiers: none

• Supports: Clinically important hypotension in the setting of anaesthesia: IV bolus 5–10 mg as needed, not to exceed 50 mg; tachyphylaxis possible.
• Used by: ephedrine
• Verified via: Copied from DOSING_SOURCES.md S29: WebSearch record of drugs.com and medlibrary.org label copies, 2026-11-05. (date 2026-11-05)

S20

Chung F, Yegneswaran B, Liao P, et al. STOP questionnaire: a tool to screen patients for obstructive sleep apnea. Anesthesiology. 2008;108(5):812-821. https://pubmed.ncbi.nlm.nih.gov/18431116/
Identifiers: PMID 18431116 (DOI not re-verified)

• Supports: STOP items and Bang items (BMI >35, age >50, neck >40 cm, male sex).
• Used by: engine/UI only
• Verified via: Copied from DOSING_SOURCES.md S32: WebSearch record of PubMed 18431116 and qxmd.com/r/18431116, 2026-11-05. (date 2026-11-05)

S21

Chung F, Abdullah HR, Liao P. STOP-Bang questionnaire: a practical approach to screen for obstructive sleep apnea. Chest. 2016;149(3):631-638. https://pubmed.ncbi.nlm.nih.gov/26378880/
Identifiers: PMID 26378880 (DOI not re-verified)

• Supports: STOP-Bang 0–2 low, 3–4 intermediate, 5–8 high risk of moderate-to-severe OSA.
• Used by: engine/UI only
• Verified via: Copied from DOSING_SOURCES.md S33: WebSearch record of PubMed 26378880 and the UW-hosted STOP-Bang appendix, 2026-11-05. (date 2026-11-05)

S22

American Society of Anesthesiologists. Standards for Basic Anesthetic Monitoring. Last amended October 15, 2025. https://www.asahq.org/standards-and-practice-parameters/standards-for-basic-anesthetic-monitoring
Identifiers: none

• Supports: BP and HR at least every 5 min; continuous ECG; pulse oximetry; end-tidal CO2 during GA; exhaled-CO2 monitoring during moderate or deep sedation.
• Used by: engine/UI only
• Verified via: Copied from DOSING_SOURCES.md S35: WebSearch record of the asahq.org standards page and the OpenAnesthesia summary, 2026-11-05. (date 2026-11-05)

S23

Etomidate injection, USP (AMIDATE). Hospira, Inc. (Pfizer). FDA-approved prescribing information. https://www.pfizermedical.com/amidate/dosage-admin
Identifiers: NDA 018227 (FDA-hosted label 2017: 018227s032lbl.pdf)

• Supports: Adults: 0.2–0.6 mg/kg, individualized; usual 0.3 mg/kg over 30–60 s. Onset usually within 1 min; duration dose dependent, usually 3–5 min at 0.3 mg/kg. Reduced plasma cortisol after 0.3 mg/kg persisting about 6–8 h, unresponsive to ACTH. Geriatric patients may require reduced doses. Contraindicated in patients who have shown hypersensitivity to it.
• Used by: etomidate
• Verified via: WebSearch records of pfizermedical.com/amidate (dosage-admin), prescriberpoint.com etomidate label copies, the FDA-hosted 2017 label (accessdata.fda.gov 018227s032lbl.pdf, link only) and the drugs.com Amidate copy (updated Jan 2026) for onset/duration wording, 2026-11-05. (date 2026-11-05)

S24

Succinylcholine chloride injection, USP (ANECTINE; QUELICIN). Sandoz / Hospira prescribing information (medlibrary.org and medthority.com label copies; DailyMed setid 1d4302db-a0d5-d0e8-e063-6294a90a22f8). https://medlibrary.org/lib/rx/meds/anectine/
Identifiers: none

• Supports: Average adult intubating dose 0.6 mg/kg IV (range 0.3–1.1 mg/kg). After IV doses in this range blockade develops in about 1 min, maximum about 2 min, recovery within 4–6 min. Contraindications (older-format labels): personal or familial history of malignant hyperthermia, skeletal muscle myopathies, known hypersensitivity; not to be used after the acute phase of injury following major burns, multiple trauma, extensive denervation of skeletal muscle, or upper motor neuron injury. Boxed warning: hyperkalemic rhabdomyolysis in pediatric patients. Hyperkalemia risk in burn/injury patients peaks 7–10 days after injury.
• Used by: succinylcholine
• Verified via: WebSearch records of medlibrary.org Anectine copies (0.3–1.1 mg/kg; onset/peak/recovery), medthority.com Anectine/Quelicin copies (0.6 mg/kg; boxed warning; burn hyperkalemia timing), drugs.com and DailyMed listings (contraindication wording), 2026-11-05. Contraindication lists differ between label versions (some newer labels move burns/denervation to Warnings). (date 2026-11-05)

S25

Sevoflurane, USP, volatile liquid for inhalation (ULTANE). AbbVie. FDA-approved prescribing information. https://medlibrary.org/lib/rx/meds/ultane-1/
Identifiers: none confirmed

• Supports: MAC in oxygen by age: 25 y 2.6%; 40 y 2.1%; 60 y 1.7%; 80 y 1.4%. Surgical levels of anaesthesia usually achieved with 0.5–3% with or without nitrous oxide. Contraindicated in known or suspected genetic susceptibility to malignant hyperthermia and known sensitivity to sevoflurane or other halogenated agents.
• Used by: sevoflurane
• Verified via: WebSearch records of prescriberpoint.com sevoflurane label copies (MAC table; contraindications), medlibrary.org Ultane copy and pdr.net (0.5–3% maintenance sentence), 2026-11-05. The current US label full text was not opened; the 0.5–3% sentence appeared under an odd heading in the medlibrary copy. (date 2026-11-05)

S26

Mapleson WW. Effect of age on MAC in humans: a meta-analysis. Br J Anaesth. 1996;76(2):179-185. https://academic.oup.com/bja/issue/76/2
Identifiers: doi:10.1093/bja/76.2.179; PMID unverified
Type: meta-analysis (age-adjusted MAC model)

• Supports: MAC(age) = MAC40 × 10^(−0.00269 × (age − 40)), same slope for all inhaled agents above age 1 y (about 6% decline per decade). MAC40: sevoflurane 1.80%, isoflurane 1.17%, desflurane 6.6%, nitrous oxide 104% (halothane 0.75%, enflurane 1.63%); 95% limits about ±7% (desflurane ±10%).
• Used by: sevoflurane, desflurane, isoflurane, nitrous_oxide
• Verified via: WebSearch records of the BJA 76(2) issue listing (citation, DOI), a Mapleson-authored abstract on scite.ai (MAC40 1.80% sevoflurane; b = −0.00269), and a StatPearls review quoting the equation, 2026-11-05. Isoflurane 1.17%, desflurane 6.6% and nitrous oxide 104% MAC40 values confirmed in the same scite.ai abstract record and repeated in secondary pages (meddic.jp; medicalalgorithms.com), 2026-11-05. Full text not opened. (date 2026-11-05)

S27

Ketamine hydrochloride injection (KETALAR). Par Pharmaceutical / Endo. FDA-approved prescribing information, 2022 revision (also 2026 revision 016812s057); and Fresenius Kabi generic ketamine label. https://www.accessdata.fda.gov/drugsatfda_docs/label/2022/016812s051lbl.pdf
Identifiers: NDA 016812

• Supports: IV induction 1–4.5 mg/kg over 60 s; average 2 mg/kg produces 5–10 min of surgical anaesthesia within 30 s; maintenance increments of one-half to the full induction dose. Contraindicated in patients for whom a significant elevation of blood pressure would constitute a serious hazard, and in known hypersensitivity. Postoperative confusional states may occur during recovery.
• Used by: ketamine
• Verified via: WebSearch records of the FDA-hosted 2022 and 2026 Ketalar labels (1–4.5 mg/kg; maintenance increments; contraindications; emergence wording) and the Fresenius Kabi generic label copy (2 mg/kg, 5–10 min, within 30 s), 2026-11-05. (date 2026-11-05)

S28

Lidocaine hydrochloride injection, USP (intravenous, antiarrhythmic). Fresenius Kabi and Hospira prescribing information; archived DailyMed lidocaine labels. https://labeling.pfizer.com/ShowLabeling.aspx?id=4497
Identifiers: none

• Supports: Contraindication (Fresenius Kabi): not to be used in Stokes-Adams syndrome, Wolff-Parkinson-White syndrome, or severe degrees of sinoatrial, atrioventricular or intraventricular block. Warning (older-format labels): patients with severe hepatic disease are at greater risk of developing toxic plasma concentrations.
• Used by: lidocaine
• Verified via: WebSearch records of guidelinecentral.com Fresenius lidocaine label, labeling.pfizer.com (Hospira caution wording), the drugs.com lidocaine monograph and archived DailyMed lidocaine labels (webarchive.library.unt.edu), 2026-11-05. No dose is taken from this label. (date 2026-11-05)

S29

Ondansetron injection, USP. Pfizer prescribing information (pfizermedicalinformation.com/ondansetron). https://www.pfizermedicalinformation.com/ondansetron/contradictions
Identifiers: none

• Supports: Contraindicated in known hypersensitivity (e.g., anaphylaxis) to the product or its components, and with concomitant apomorphine (profound hypotension and loss of consciousness reported).
• Used by: ondansetron
• Verified via: WebSearch record of pfizermedicalinformation.com/ondansetron/contradictions and Canadian monographs (pdf.hres.ca), 2026-11-05. Dose is taken from S3, not from this label. (date 2026-11-05)

S30

Dexamethasone sodium phosphate injection, USP. Multiple manufacturers (e.g., Fresenius Kabi; label copies at medthority.com and drugcentral.org). https://drugcentral.org/label/9dbb2b90-471a-4de8-b105-6c9ede7d8d7e/view
Identifiers: none

• Supports: Contraindications: systemic fungal infections; hypersensitivity to any component of this product (including sulfites in some products).
• Used by: dexamethasone
• Verified via: WebSearch records of drugcentral.org and medthority.com dexamethasone sodium phosphate label copies, 2026-11-05. Dose is taken from S2/S3, not from this label. (date 2026-11-05)

S31

American Society of Anesthesiologists. Statement on Continuum of Depth of Sedation: Definition of General Anesthesia and Levels of Sedation/Analgesia. Original approval October 13, 1999; last amended October 23, 2024. https://www.asahq.org/standards-and-practice-parameters/statement-on-continuum-of-depth-of-sedation-definition-of-general-anesthesia-and-levels-of-sedation-analgesia
Identifiers: none

• Supports: Minimal sedation: normal response to verbal stimulation. Moderate: purposeful response to verbal or light tactile stimulation. Deep: purposeful response after repeated or painful stimulation (reflex withdrawal is not purposeful). General anaesthesia: not arousable even by painful stimulation.
• Used by: engine/UI only
• Verified via: WebSearch record of the asahq.org statement page (amendment date) and reproductions of the statement text, 2026-11-05. (date 2026-11-05)

S32

Practice guidelines for moderate procedural sedation and analgesia 2018: a report by the American Society of Anesthesiologists Task Force on Moderate Procedural Sedation and Analgesia, et al. Anesthesiology. 2018;128(3):437-479. https://pubmed.ncbi.nlm.nih.gov/29334501/
Identifiers: doi:10.1097/ALN.0000000000002043; PMID 29334501

• Supports: Framework for moderate procedural sedation (monitoring, titration, personnel). No dose is taken from it in this library.
• Used by: engine/UI only
• Verified via: WebSearch record of PubMed 29334501 and qxmd.com/r/29334501 (citation, DOI), 2026-11-05. Its capnography recommendation text was not retrieved; capnography statements in the simulator are tagged to S22. (date 2026-11-05)

S33

Devine BJ. Gentamicin therapy. Drug Intell Clin Pharm. 1974;8:650-655. https://wikem.org/wiki/Ideal_body_weight_calculator
Identifiers: none confirmed

• Supports: Ideal body weight: male 50 kg, female 45.5 kg, + 2.3 kg per inch over 60 in.
• Used by: engine/UI only
• Verified via: WebSearch records of secondary reproductions (wikem.org; evigrade.com; calcfyi.com) giving the citation and formula, 2026-11-05. Original article not opened. (date 2026-11-05)

S34

Janmahasatian S, Duffull SB, Ash S, Ward LC, Byrne NM, Green B. Quantification of lean bodyweight. Clin Pharmacokinet. 2005;44(10):1051-1065. https://pmc.ncbi.nlm.nih.gov/articles/PMC2485235
Identifiers: doi:10.2165/00003088-200544100-00004

• Supports: Lean body weight: male 9270 × TBW / (6680 + 216 × BMI); female 9270 × TBW / (8780 + 244 × BMI).
• Used by: engine/UI only
• Verified via: WebSearch records of bibliographic copies (citation, DOI, pages) and a PMC article quoting the denominators 6680 + 216×BMI and 8780 + 244×BMI, 2026-11-05. The 9270 numerator was not displayed in the search record (identifier/constant unverified in this session); check the original before reuse. (date 2026-11-05)

S35

Smart NOA Controller. config.yaml (simulation_rate_ceilings; ketorolac egfr_minimum; dexmedetomidine cardiac_conditions). Project configuration in this repository. config.yaml
Identifiers: none

• Supports: Python simulation rate ceilings: dexmedetomidine 0.5 mcg/kg/h, ketamine 0.2 mg/kg/h, lidocaine 1.5 mg/kg/h; ketorolac lockout below eGFR 30; dexmedetomidine lockout for heart block conditions.
• Used by: ketamine, dexmedetomidine, lidocaine, ketorolac
• Verified via: Read directly from /home/user/smart-noa-controller/config.yaml, 2026-11-05. This is not clinical evidence; it is cited so the simulator mirrors the Python controller. (date 2026-11-05)

S36

Desflurane, USP, volatile liquid for inhalation (SUPRANE). Baxter Healthcare. FDA-approved prescribing information (drugs.com copy revised July 2026; medlibrary.org and ndclist.com copies; DailyMed desflurane setid 21ba0e18-3dbc-4feb-a3a4-f6470c00910a). https://medlibrary.org/lib/rx/meds/suprane/page/2/
Identifiers: none confirmed

• Supports: Surgical levels of anesthesia in adults may be maintained with 2.5–8.5% with or without nitrous oxide; required concentration is age dependent. Can induce malignant hyperthermia in patients with known or suspected susceptibility based on genetic factors or family history (MH susceptibility listed as a contraindication). Concentrations exceeding 1 MAC may increase heart rate; a rapid increase in concentration is associated with marked increase in pulse rate, mean arterial pressure and catecholamines. Not recommended for induction in paediatric patients because of moderate to severe upper airway adverse events (laryngospasm, coughing, breath holding, apnea, increased secretions, desaturation). Contraindicated in patients with known sensitivity to SUPRANE or to other halogenated agents, and with a history of moderate to severe hepatic dysfunction following anaesthesia with SUPRANE or other halogenated agents.
• Used by: desflurane
• Verified via: WebSearch records of drugs.com/pro/suprane (current label, MH and heart-rate wording), medlibrary.org Suprane copies (2.5–8.5% maintenance sentence, older-format version; rapid-increase wording) and the DailyMed desflurane listing, 2026-11-05. The adult mask-induction statement ("not recommended for mask induction of anesthesia in adults") was seen only in the Canadian monograph and is not used as US wording. No US MAC-by-age table was retrieved. 2026-11 audit: halogenated-agent sensitivity and hepatic-dysfunction contraindications confirmed in WebSearch records of drugs.com/pro/suprane, medlibrary.org Suprane copies and the medthority.com (Henry Schein) US label copy; the same items appear in the EU SmPC and Canadian monograph. (date 2026-11-05)

S37

Isoflurane, USP (FORANE). Baxter Healthcare. FDA-approved prescribing information (FDA-hosted August 2006 PI, archived: webarchive.library.unt.edu ... fda.gov/MEDWATCH/safety/2006/Aug_PIs/Forane_PI.pdf). https://webarchive.library.unt.edu/eot2008/20081028114252/http:/www.fda.gov/MEDWATCH/safety/2006/Aug_PIs/Forane_PI.pdf
Identifiers: none confirmed

• Supports: Induction: start at 0.5%; 1.5–3.0% usually produces surgical anesthesia in 7–10 min. Maintenance: 1.0–2.5% in oxygen/nitrous oxide; an additional 0.5–1.0% may be required with oxygen alone. Average MAC in 100% O2: 1.28% (age 26 ± 4), 1.15% (44 ± 7), 1.05% (64 ± 5); in 70% N2O: 0.56%, 0.50%, 0.37%. Contraindications: known sensitivity to isoflurane or other halogenated agents; known or suspected genetic susceptibility to malignant hyperthermia.
• Used by: isoflurane
• Verified via: WebSearch records of the archived FDA-hosted 2006 Forane PI, the Baxter Canada Forane monograph and an Irish SmPC (consistent figures), 2026-11-05. US-format contraindication wording from the search record; drugs.com isoflurane page text was garbled and not relied on. (date 2026-11-05)

Non-label sources for nitrous oxide (S38, S39)

These two sources are not FDA labels or society guidelines. The owner asked for nitrous oxide to be included. US nitrous oxide cylinder labels give no clinical concentration and no contraindications, only handling, fire and asphyxia warnings. So for N2O only, an RCT (S38) supplies the illustrative concentration and an animal physiology study (S39) supplies the closed-gas-space rules. Each entry is labelled with its type.

S38

Myles PS, Leslie K, Chan MTV, et al; ANZCA Trials Group for the ENIGMA-II investigators. The safety of addition of nitrous oxide to general anaesthesia in at-risk patients having major non-cardiac surgery (ENIGMA-II): a randomised, single-blind trial. Lancet. 2014;384(9952):1446-1454. https://clinicaltrials.gov/study/NCT00430989
Identifiers: doi:10.1016/S0140-6736(14)60893-X; PMID 25142708; NCT00430989
Type: RCT

• Supports: Intervention arm: nitrous oxide at an inspired concentration of 70% in 30% oxygen; control: N2O-free with inspired oxygen 30%. 7112 patients ≥45 y with or at risk of coronary disease; death and cardiovascular complications at 30 days 8% in each group (RR 0.96, 95% CI 0.83–1.12). Avoiding nitrous oxide reduced severe PONV.
• Used by: nitrous_oxide
• Verified via: WebSearch records of the ClinicalTrials.gov NCT00430989 record ("inspired concentration of 70% in 30% Oxygen"), the University of Geneva archive record (citation, DOI, PMID), ANZCA repository and qxmd.com/r/25142708, 2026-11-05. Full text not opened. Non-label source: used because no US nitrous oxide label states a clinical concentration. (date 2026-11-05)

S39

Eger EI 2nd, Saidman LJ. Hazards of nitrous oxide anesthesia in bowel obstruction and pneumothorax. Anesthesiology. 1965;26:61-66. https://pubmed.ncbi.nlm.nih.gov/14257336/
Identifiers: PMID 14257336
Type: physiology study

• Supports: Nitrous oxide diffuses into closed gas spaces faster than nitrogen leaves, expanding them: in dogs breathing 75% N2O, pneumothorax volume roughly doubled by 10 min and roughly tripled-to-quadrupled (about +300%) by 60 min, whereas oxygen did not change it; bowel gas also expanded. Authors concluded nitrous oxide is relatively contraindicated in intestinal obstruction or pneumothorax.
• Used by: nitrous_oxide
• Verified via: WebSearch records of PubMed 14257336 (citation; no abstract listed), the Scilit record (conclusion "relatively contraindicated in intestinal obstruction or pneumothorax"), BestBets review table (dog data, +100% at 10 min, ~+300% at 60 min) and a textbook figure reproduction (75% N2O), 2026-11-05. Full text not opened; volume figures are from secondary reproductions. Animal study (pneumothorax created by air injection). Non-label source: used because no US nitrous oxide label states contraindications. (date 2026-11-05)

S40

Epinephrine injection, USP 1 mg/mL (ADRENALIN). Par Pharmaceutical / Par Health USA. FDA-approved prescribing information (DailyMed setid 3b7a4364-668d-4eb2-a20c-04adc35aabe4; FDA-hosted label files 204200s004 (2016), 204200s009/204640s009 (2019), 204640Orig1s018 (2023)). https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=3b7a4364-668d-4eb2-a20c-04adc35aabe4
Identifiers: NDA 204200 / NDA 204640 (as shown in FDA-hosted label file names; not otherwise re-verified)

• Supports: Anaphylaxis, adults and children over 30 kg: 0.3–0.5 mg (0.3–0.5 mL) undiluted IM or SC into the anterolateral thigh, up to a maximum of 0.5 mg (0.5 mL) per injection, every 5–10 min as necessary; monitor clinically for reaction severity and cardiac effects; do not inject into buttock, digits, hands or feet. Hypotension associated with septic shock: dilute 1 mg in 1000 mL D5W (1 mcg/mL), large vein, 0.05–2 mcg/kg/min titrated to the desired MAP. Contraindications: None. Contains sodium bisulfite; "presence of sulfite in this product should not deter use for anaphylaxis". "Epinephrine may induce cardiac arrhythmias and myocardial ischemia in patients, especially patients suffering from coronary artery disease, or cardiomyopathy."
• Used by: epinephrine
• Verified via: WebSearch records (2026-11-05) of DailyMed Adrenalin listings (setids 3b7a4364-…, 7663c49d-…, b293ad8e-…), drugs.com/pro/adrenalin-injection, the FFF Enterprises-hosted Adrenalin PI PDF, guidelinecentral.com and FDA-hosted Adrenalin labels 204200s004 / 204200s009 / 204640s018. The weight boundary appeared as "over 30 kg" (children "30 kg or less"); wording may differ between versions. Full label text not opened (DailyMed and accessdata.fda.gov unreachable from the build environment). (date 2026-11-05)

S41

Epinephrine injection (generic 1 mg/mL vial; NDA 215425). FDA-approved prescribing information, 2025 (215425s000lbl.pdf); and other generic epinephrine injection labels (NDA 205029, 2021–2025 revisions; Pfizer/Hospira labeling). https://www.accessdata.fda.gov/drugsatfda_docs/label/2025/215425s000lbl.pdf
Identifiers: NDA 215425; NDA 205029 (as shown in FDA-hosted label file names)

• Supports: Septic-shock hypotension: 0.05–2 mcg/kg/min titrated to the desired MAP; adjust e.g. every 10–15 min in increments of 0.05–0.2 mcg/kg/min; wean incrementally after stabilisation. "The ideal body weight (IBW) should be used as the weight parameter for dosing epinephrine in adult patients with septic shock associated hypotension" (215425 label). Sulfite should not deter use for septic-shock hypotension.
• Used by: epinephrine
• Verified via: WebSearch records (2026-11-05) of FDA-hosted labels 215425s000 (IBW sentence quoted verbatim), 205029Orig1s009 and 205029Orig1s015, DailyMed generic listings and a Pfizer highlights page ("0.05 mcg/kg/min to 2 mcg/kg/min, titrated to achieve desired mean arterial pressure (2.2) • Wean gradually"). The exact label carrying each titration/weaning excerpt was not pinned down. Full text not opened. (date 2026-11-05)

S42

Eleveld DJ, Colin P, Absalom AR, Struys MMRF. Pharmacokinetic-pharmacodynamic model for propofol for broad application in anaesthesia and sedation. Br J Anaesth. 2018;120(5):942-959. Corrigendum: Br J Anaesth. 2018;121(2):519. https://www.bjanaesthesia.org/article/S0007-0912(18)30051-5/fulltext
Identifiers: doi:10.1016/j.bja.2018.01.018; PMID 29661412
Type: published population PK-PD model

• Supports: The propofol PK model (see "Propofol PK model" below).
• Used by: propofol (pkModel: 'eleveld')
• Verified via: see "Propofol PK model" below. 2026-11 audit: the simulator now switches the covariate on for opioids only; the earlier extension to IV lidocaine was a simulator choice and was removed. (date 2026-11-05)

S44

Association of Anaesthetists of Great Britain and Ireland. Suspected anaphylactic reactions associated with anaesthesia. Anaesthesia. 2009;64(2):199-211. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3082210/
Identifiers: doi:10.1111/j.1365-2044.2008.05733.x; PMC3082210 (PMID not re-verified)
Type: society guideline

• Supports: Adrenaline is the most effective drug in anaphylaxis and should be given as early as possible. Adult IV dose 50 µg (0.5 mL of 1:10 000); several doses may be required for severe hypotension or bronchospasm; if several doses are required, consider an IV adrenaline infusion.
• Used by: epinephrine rescue_iv (50 mcg IV bolus, anaesthetised monitored adult); anaphylaxis scenario teaching point.
• Verified via: WebSearch records (2026-11-05) of the PMC3082210 full-text record (dose sentence quoted), the University of Southampton eprint 72661 (citation, DOI), NHS Scotland Right Decisions copies, and the NAP6 report chapter on rcoa.ac.uk ("AAGBI guidelines recommend an initial IV dose of 50 mcg, repeated as necessary"). Full PDF not opened (WebFetch unavailable). (date 2026-11-05)

S45

Resuscitation Council UK. Peri-operative anaphylaxis: initial management (algorithm for anaesthetists and intensivists), with the Association of Anaesthetists. 2024 (file "2526 AAP RCUK periop anaphylaxis"). https://www.resus.org.uk/sites/default/files/2024-01/2526%20AAP%20RCUK%20periop%20anaphylaxis-8C.pdf
Identifiers: none
Type: society guideline (algorithm)

• Supports: Peri-operative anaphylaxis (anaesthetists/intensivists): initial IV adrenaline bolus — adults and children > 12 y 50 micrograms (0.5 mL of 1 mg/10 mL), titrated; start an adrenaline infusion if repeated boluses are needed; IV crystalloid. RCUK 2021: IV adrenaline only by experienced specialists in an appropriate setting.
• Used by: epinephrine rescue_iv; anaphylaxis scenario teaching point.
• Verified via: WebSearch records (2026-11-05) of the resus.org.uk algorithm PDF, the RCUK 2021 guideline/draft ("Adults 50 micrograms" titration; specialist-only IV adrenaline) and SALG / Perioperative Allergy Network summaries. Full PDF not opened. (date 2026-11-05)

S43

Evans L, Rhodes A, Alhazzani W, et al. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock 2021. Crit Care Med. 2021 (published simultaneously in Intensive Care Med. 2021). https://sccm.org/sepsisguidelines
Identifiers: none confirmed (volume, pages, DOI and PMID not re-verified)
Type: guideline

• Supports: "For adults with sepsis-induced hypoperfusion or septic shock, we suggest that at least 30 mL/kg of IV crystalloid fluid should be given within the first 3 hours of resuscitation" — weak recommendation, low quality of evidence (downgraded from strong in 2016).
• Used by: engine only (context text logged when a sepsis patient's cumulative crystalloid reaches 30 mL/kg). No dose, threshold or effect size in the simulator is derived from it.
• Verified via: WebSearch records (2026-11-05) of sccm.org/sepsisguidelines, an AMiner executive-summary record, emDocs and Red Cross summaries and a CME PDF reproducing the wording and grading. The guideline text itself was not opened. (date 2026-11-05)

2026-11 additions (S46–S51)

Added for the owner-approved additions (lidocaine pretreatment, IV fluids, antibiotic prophylaxis, technique templates). Direct page fetches (ashp.org, PMC) still failed DNS resolution, so every entry was checked by WebSearch records only.

S46

Bratzler DW, Dellinger EP, Olsen KM, et al. Clinical practice guidelines for antimicrobial prophylaxis in surgery. Am J Health Syst Pharm. 2013;70(3):195-283. (Joint ASHP/IDSA/SIS/SHEA guideline; also published in Surg Infect (Larchmt). 2013;14(1):73-156.) https://academic.oup.com/ajhp/article/70/3/195/5112717
Identifiers: doi:10.2146/ajhp120568; PMID 23327981
Type: society guideline

• Supports: Adult doses: cefazolin 2 g IV (3 g for patients weighing ≥120 kg), redosing interval 4 h; clindamycin 900 mg IV, redosing interval 6 h; vancomycin 15 mg/kg IV, routine redosing not applicable. First dose should begin within 60 minutes before surgical incision; vancomycin (and fluoroquinolones) should begin within 120 minutes before incision because of prolonged infusion times. Redosing is needed when the procedure exceeds about two half-lives of the agent; the interval is measured from the initiation of the preoperative dose. For patients with β-lactam allergy, clindamycin or vancomycin are the alternatives most frequently recommended; cephalosporins are avoided in patients with an IgE-mediated (e.g. anaphylaxis) or severe reaction to a β-lactam.
• Used by: cefazolin, clindamycin, vancomycin (ANTIBIOTICS), prophylaxis window and redose alarms (engine/prophylaxis.js), β-lactam allergy lock
• Verified via: Citation, PMID and doses copied from the protocol registry DOSING_SOURCES.md S6 (WebSearch records of the OU publication list and institutional summaries — UNMC prophylaxis/redosing tables, ASA quality measure using the 4-h cefazolin interval), 2026-11-05. Re-checked 2026-11-05 by WebSearch: DOI 10.2146/ajhp120568 and the academic.oup.com AJHP 70(3):195 landing record; CHOC 2017/2023 guidelines reproducing Table 1 (cefazolin 2 g, 3 g for ≥120 kg, redose 4 h; clindamycin 900 mg, redose 6 h); UNMC/Thompson Health protocols quoting "within 60 minutes before surgical incision … vancomycin within 120 minutes"; OpenAnesthesia and Infection Control & Hospital Epidemiology summaries (clindamycin and vancomycin as the most frequent β-lactam-allergy alternatives; avoid cephalosporins after IgE-mediated or severe reactions). The guideline full text and Table 1 were not opened (ashp.org, PMC not reachable); the vancomycin "N/A" redosing cell and the exact allergy sentence were seen only in secondary reproductions. The 60–120 min wording in the additions request was NOT found for clindamycin (guideline: within 60 min); see SOURCES.md. (date 2026-11-05)

S47

Lactated Ringer's Injection, USP. Baxter Healthcare Corporation. FDA-approved prescribing information, revised 10/2025 (accessdata.fda.gov label 019632s067, 019634s058, 020002s037); and Baxter container labeling. https://www.accessdata.fda.gov/drugsatfda_docs/label/2025/019632s067%2C019634s058%2C020002s037lbl.pdf
Identifiers: NDA 019632 / 019634 / 020002 (as shown in the FDA-hosted label file name)
Type: FDA label

• Supports: Indicated as a source of water and electrolytes or as an alkalinizing agent. Approximate composition (per litre): sodium 130, potassium 4, calcium 1.4 mmol (2.7 mEq), chloride 109, lactate 28 mmol; osmolarity about 272–273 mOsmol/L; pH about 6.5. Dosage: the recommended dosage and duration are based on the patient's age, weight, clinical condition and concomitant therapy (no fixed dose). Warnings include: potassium-containing solutions may increase the risk of hyperkalaemia; patients with severe hepatic impairment may have impaired lactate metabolism (monitor lactate and acid-base status); do not give simultaneously with ceftriaxone in neonates ≤ 28 days (sequentially with line flushing in older patients); fluid/solute overload.
• Used by: lactated Ringer's (FLUIDS): composition shown in the fluids panel, label cautions (hyperkalaemia, hepatic impairment); no dose
• Verified via: WebSearch records (2026-11-05) of the FDA-hosted 2025 Baxter label (dosage based on age, weight, clinical condition and concomitant therapy; severe hepatic impairment and lactate; ceftriaxone wording), ndclist.com Baxter label copies 0338-0117 / 0338-9600 (container composition "Na 130 K 4 Ca 1.4 Cl 109 LACTATE 28 mmol/L, 272 mOsmol/L, pH 6.5"; "Potassium-containing solutions, including Lactated Ringer's Injection, USP, may increase the risk of hyperkalemia"; indication sentence) and a Canadian Baxter monograph (not for use in hyperkalaemia; not for lactic acidosis). Full label not opened. 2.7 mEq/L calcium is the mmol→mEq conversion of 1.4 mmol/L (×2, valence), arithmetic. No dose is taken from this label: rates and volumes in the simulator are UI choices (simulator convention). (date 2026-11-05)

S48

0.9% Sodium Chloride Injection, USP (VIAFLEX / VIAFLO containers). Baxter Healthcare Corporation; and Pfizer/Hospira 0.9% sodium chloride prescribing information. https://labeling.pfizer.com/ShowLabeling.aspx?id=4612
Identifiers: none
Type: FDA label

• Supports: Source of water and electrolytes. Per litre: sodium 154 mEq, chloride 154 mEq; osmolarity 308 mOsmol/L (calc.), isotonic. Warnings: may cause fluid and/or solute overload; use with particular caution (or avoid) in patients with or at risk for hypernatraemia, hyperchloraemia or metabolic acidosis; monitor serum sodium and chloride, fluid status and acid-base balance. Post-marketing: hypernatraemia, hyperchloraemic metabolic acidosis and hyponatraemia reported.
• Used by: 0.9% sodium chloride (FLUIDS): composition and label cautions; no dose
• Verified via: WebSearch records (2026-11-05) of pfizermedicalinformation.com sodium chloride description (154 mEq sodium and chloride per 1000 mL; 308 mOsmol/L calc.; isotonic), the Baxter 2B1324X/VIAFLO package inserts (pim.integratedmedsys.com) and medthority.com Baxter label copies (fluid/solute overload; caution in hypernatraemia, hyperchloraemia, metabolic acidosis; monitoring of sodium, chloride and acid-base balance; post-marketing hyperchloraemic metabolic acidosis). Exact dosage-section wording was not displayed; no fixed dose is taken from this label (rates/volumes are simulator conventions). (date 2026-11-05)

S49

Beloeil H, Garot M, Lebuffe G, et al. Balanced opioid-free anesthesia with dexmedetomidine versus balanced anesthesia with remifentanil for major or intermediate noncardiac surgery: the Postoperative and Opioid-free Anesthesia (POFA) randomized clinical trial. Anesthesiology. 2021;134(4):541-551. https://www.iasp-pain.org/publications/pain-research-forum/papers-of-the-week/paper/163431-balanced-opioid-free-anesthesia-dexmedetomidine-versus-balanced-anesthesia/
Identifiers: doi:10.1097/ALN.0000000000003725; PMID 33630043
Type: randomized clinical trial

• Supports: Dexmedetomidine-based opioid-free balanced anaesthesia versus remifentanil-based balanced anaesthesia: more of the composite of postoperative opioid-related adverse events (driven by hypoxaemia), more delayed extubation and longer PACU stay, and the trial was stopped early after five cases of severe bradycardia in the dexmedetomidine group; postoperative nausea and vomiting was less frequent in the dexmedetomidine group. Used only as an evidence note on the opioid-free template; no dose is taken from it.
• Used by: GA — opioid-free template evidence note only
• Verified via: WebSearch records (2026-11-05) of the IASP Pain Research Forum abstract record (PMID 33630043; composite 78% vs 67%; hypoxaemia 72% vs 61%; early termination for severe bradycardia; delayed extubation and PACU stay), typeset.io and revistachilenadeanestesia.cl copies, the Anesthesiology podcast page and editorial record (Kharasch & Clark). The PONV direction (less with dexmedetomidine) was seen in search excerpts whose percentages were inconsistently labelled, so no PONV figure is shown. Full text not opened. (date 2026-11-05)

S50

Semler MW, Self WH, Wanderer JP, et al. Balanced crystalloids versus saline in critically ill adults. N Engl J Med. 2018;378(9):829-839. https://www.uab.edu/medicine/coere/images/nejmoa1711584.pdf
Identifiers: doi:10.1056/NEJMoa1711584 (PMID not re-verified)
Type: randomized clinical trial (cluster, multiple crossover)

• Supports: SMART: 15,802 critically ill adults in five ICUs; major adverse kidney events within 30 days in 14.3% with balanced crystalloids versus 15.4% with 0.9% saline (adjusted OR 0.90, 95% CI 0.82–0.99). Critically ill (ICU) population, not elective surgery; used only as an evidence note on the fluids panel.
• Used by: fluids evidence note only
• Verified via: WebSearch records (2026-11-05) of the NEJM PDF copy (uab.edu), criticalcarereviews.com, The Bottom Line and EMRA summaries (citation, DOI, 14.3% vs 15.4%, OR 0.90). PMID was not confirmed in a search record and is not shown. (date 2026-11-05)

S51

Jalota L, Kalira V, George E, et al. Prevention of pain on injection of propofol: systematic review and meta-analysis. BMJ. 2011;342:d1110. https://www.bmj.com/content/342/bmj.d1110.abstract
Identifiers: doi:10.1136/bmj.d1110; PMID 21406529
Type: systematic review and meta-analysis

• Supports: 177 randomised trials, 25,260 adults; pain with propofol alone in about 60%. Most effective: antecubital vein instead of a hand vein (RR 0.14, 95% CI 0.07–0.30) and lidocaine pretreatment with venous occlusion (RR 0.29, 0.22–0.38); lidocaine–propofol admixture (0.40) and lidocaine pretreatment alone (0.47, 0.40–0.56) also effective. Used only as an evidence note; the dose comes from the propofol label (S11).
• Used by: propofol injection-pain teaching note only
• Verified via: WebSearch records (2026-11-05) of the bmj.com abstract page (bmj.d1110), the read.qxmd.com record (PMID 21406529) and library catalogue records (citation, abstract figures). Author list beyond the first three not re-verified (et al.). (date 2026-11-05)

2026-11 timing review (S52–S54; additions to S9, S11, S12, S13)

Verified by WebSearch records on 2026-11-05 (page fetches failed DNS: medlibrary.org, drugs.com). Timing statements added to existing entries: S9 dexmedetomidine — "Initiate at one mcg/kg over 10 minutes, followed by a maintenance infusion initiated at 0.6 mcg/kg/hour" (DailyMed / labeling.pfizer.com records); S11 propofol — MAC initiation over 3–5 min, then maintenance rates "during the first 10 to 15 minutes of sedation maintenance" (no "followed immediately" wording found for MAC); S12 rocuronium — maintenance doses "administered at 25% recovery of control T1 (defined as 3 twitches of train-of-four)" (archived FDA-hosted Zemuron PI 2008; generic label); S13 fentanyl — "the maximal analgesic and respiratory depressant effect may not be noted for several minutes" (archived DailyMed label). The full text of every entry is in data/drugs.js (supports, verifiedVia).

S52

Avidan MS, Zhang L, Burnside BA, et al. Anesthesia awareness and the bispectral index. N Engl J Med. 2008;358(11):1097-1108. (B-Unaware trial.) https://pubmed.ncbi.nlm.nih.gov/18337600/
Identifiers: doi:10.1056/NEJMoa0707361; PMID 18337600
Type: randomized clinical trial

• Supports: BIS-guided protocol with a target of 40 to 60 versus an end-tidal anaesthetic-gas protocol in about 2000 high-risk patients; definite awareness in 2 patients in each group; routine BIS monitoring not supported. Used only for the 40–60 GA band of the simulated processed-EEG index (a protocol target, not evidence of benefit).
• Used by: engine/timing.js GA_DEPTH_BAND (pre-incision item, tile/chart band, live check)
• Verified via: WebSearch records of PubMed 18337600, the ovid.com NEJM listing, the WashU news release and the BAG-RECALL protocol paper. Full text not opened. (date 2026-11-05)

S53

Avidan MS, Jacobsohn E, Glick D, et al. Prevention of intraoperative awareness in a high-risk surgical population. N Engl J Med. 2011;365(7):591-600. (BAG-RECALL trial.) https://pmc.ncbi.nlm.nih.gov/articles/PMC3447261/
Identifiers: PMC3447261 (DOI and PMID not confirmed in a search record; not shown)
Type: randomized clinical trial

• Supports: 6041 high-risk patients; BIS protocol with an audible alert below 40 or above 60 versus an end-tidal anaesthetic-concentration protocol; definite awareness 7/2861 (0.24%) vs 2/2852 (0.07%); superiority of the BIS protocol not demonstrated. Used only for the 40–60 GA band.
• Used by: engine/timing.js GA_DEPTH_BAND
• Verified via: WebSearch records of PMC3447261, digitalcommons.wustl.edu, aminer.org and the BAG-RECALL protocol (BMC Anesthesiol 2009;9:8, PMID 19948045). (date 2026-11-05)

S54

World Health Organization. WHO Guidelines for Safe Surgery 2009: Safe Surgery Saves Lives; and WHO Surgical Safety Checklist with Implementation Manual (2009). Geneva: WHO; 2009. https://www.leapfroggroup.org/sites/default/files/Files/WHO_SafeSurgeryChecklistImplementationManual.pdf
Identifiers: none (ISBN not verified, not shown)
Type: guideline / checklist

• Supports: Sign In, Time Out (before skin incision) and Sign Out. At the Time Out the team confirms patient, site and procedure; the checklist asks "Has antibiotic prophylaxis been given within the last 60 minutes?" (Yes / Not applicable). Not comprehensive; local modification encouraged.
• Used by: engine/timing.js TIME_OUT_REQUIRED (pre-incision / pre-procedure item; only the clinician confirms it)
• Verified via: WebSearch records of the Implementation Manual copy (leapfroggroup.org), checklist reproductions (caho.in, hqontario.ca), a PMC table and Wikipedia summaries; who.int originals not opened. (date 2026-11-05)

2026-11 timing cross-check (S55–S56; addition to S32)

Every timed element was cross-checked against labels and guidelines on 2026-11-05; the full table (element, simulator value, what each source says, agreement, classification, action) and the clinician-review checklist are in TIMING_EVIDENCE.md. Page fetches were refused by the session's egress proxy (accessdata.fda.gov, pubmed.ncbi.nlm.nih.gov), so every statement comes from WebSearch result records; exact wording that was not seen is marked as such there. S32 gained the titration statement (allow sufficient time between doses to assess each dose's effect; summarised in search records, verbatim sentence not retrieved), used for TIMING_CONFIG.MAC_SEDATIVE_AFTER_MIDAZOLAM_SEC.

S55

Frerk C, Mitchell VS, McNarry AF, et al. Difficult Airway Society 2015 guidelines for management of unanticipated difficult intubation in adults. Br J Anaesth. 2015;115(6):827-848. https://pmc.ncbi.nlm.nih.gov/articles/PMC4650961/
Identifiers: PMID 26556848; PMCID PMC4650961 (DOI not seen in a search record; not shown)
Type: guideline

• Supports: All patients preoxygenated before induction with 100% oxygen, effective face-mask seal, until the end-tidal oxygen fraction is 0.87–0.9; head-up 20–25° with CPAP delays hypoxia in obese patients. An end point, not a duration. Superseded by DAS 2025 (PMID 41203471), whose preoxygenation figures were not confirmed.
• Used by: engine/timing.js PREOXYGENATION_END_POINT (documentation entry; end-tidal O2 is not modelled) and the PREOXYGENATION_SEC rationale
• Verified via: WebSearch records of PMC4650961, PubMed 26556848 and a UKZN PDF copy; pages not fetched. No 3-min duration found in the DAS 2015 text. (date 2026-11-05)

S56

2022 American Society of Anesthesiologists Practice Guidelines for Management of the Difficult Airway. Anesthesiology. 2022;136(1):31-81. https://pubmed.ncbi.nlm.nih.gov/34762729/
Identifiers: doi:10.1097/ALN.0000000000004002; PMID 34762729
Type: guideline

• Supports: Supplemental oxygen before initiating and throughout difficult-airway management, including extubation. No preoxygenation duration or end-tidal target found.
• Used by: engine/timing.js PREOXYGENATION_END_POINT
• Verified via: WebSearch records of PubMed 34762729, the APSF update article and secondary summaries; full text not opened; author list not shown. (date 2026-11-05)

Propofol PK model

Propofol now uses the Eleveld 2018 published population PK model (pkModel: 'eleveld', arterial form, engine/pk.js ELEVELD_2018 / eleveldPropofol()); pkModelFallback: 'calibrated' keeps the illustrative model available. Order tried: Eleveld 2018 → (Schnider 1998/1999, Marsh 1991 not needed). Eleveld was implemented because its full parameter set could be checked in more than two independent sources.

How it was verified. The publisher page, PMC, PubMed and doi.org could not be opened from the build environment, and the PK theta values were not displayed in any publisher or PubMed search record. The parameter set was therefore taken from four sources that agree to the printed precision:

1. CRAN package tci 0.2.1 (R. Jarrett), R/poppk_mods.R pkmod_eleveld_ppf (GitHub mirror cran/tci, cloned 2026-11-05).
2. PyPI python-anesthesia-simulator 1.0.0 (B. Aubouin-Pairault), pk_models.py (thetas at full precision, e.g. Θ1 6.2830780766822).
3. vitaldb/tivatci, tivatci/__init__.py.
4. stanpumpR (S. L. Shafer) papers/digests/PMID_29661412.md: a digest of the paper PDF quoting Table 2/3 values with 99% CIs, page numbers, the printed equations and the corrigendum Q2 value. It is itself an AI-assisted reading, so it is secondary.

Cross-check against publication records: the PD/effect-site values (ke0 0.146 arterial, 1.24 venous; Ce50 3.08; age term −0.00635; BIS0 93.0; γ 1.47/1.89) were seen in a WebSearch record of the BJA table. The model structure was also seen in search records: three compartments, covariates age, weight, height, sex and concomitant drugs, with an opioid effect lowering CL and V3 in an age-dependent way. Citation, DOI, PMID and corrigendum were seen in bvsalud/mardi4nfdi/BJA records. The owner should still compare the Θ values below with Table 2 of the paper and the corrigendum, which asks implementers to use the supplementary NONMEM control streams.

Equations (as printed, p948, and identical in sources 1–4): V1 = Θ1·f(WGT)/f(70), f(x) = x/(x+Θ12); V2 = Θ2·(WGT/70)·exp(Θ10·(AGE−35)); V3 = Θ3·(FFM/FFMref)·fop(Θ13); CL = (Θ4 male | Θ15 female)·(WGT/70)^0.75·fCLmat/fCLmat,ref·fop(Θ11); Q2 = Θ5·(V2/Θ2)^0.75·(1+Θ16·(1−fQ3mat)); Q3 = Θ6·(V3/Θ3)^0.75·fQ3mat/fQ3mat,ref; fop(x) = exp(x·AGE) with concomitant drugs, else 1; fCLmat = PMA^Θ9/(PMA^Θ9+Θ8^Θ9); fQ3mat = (AGE·52+40)/(AGE·52+40+Θ14); PMA (weeks) = AGE·52+40. FFM = Al-Sallami 2015: male (0.88 + 0.12/(1+(AGE/13.4)^−12.7))·9270·WGT/(6680+216·BMI); female (1.11 − 0.11/(1+(AGE/7.1)^−1.1))·9270·WGT/(8780+244·BMI).

Reference values reproduced in tests (tests/models.test.mjs, implementation verification, not validation):

• Reference individual, no concomitant drugs: V1 6.28, V2 25.5, V3 273 L; CL 1.79, Q2 1.83 (corrigendum), Q3 1.11 L/min; ke0 0.146 min⁻¹. Female CL 2.10 L/min.
• Same individual with concomitant opioid: CL 1.619 L/min, V3 ≈ 168.2 L, Q3 ≈ 0.771 L/min (as computed in the stanpumpR digest; this code gives V3 168.4 because it uses the 3-significant-figure Θ3 = 273).
• Derived, not a published table value: bolus time to peak Ce ≈ 2.9 min at the reference (also the digest's derived figure).

Simulator choices (not part of the published model): the arterial form is used. The concomitant-drug switch is set once, when the propofol model is first created: it is ON if an opioid is in the plan or already given (opioids only, as published and as in the open-source implementations; IV lidocaine no longer switches it — 2026-11 audit). The simulated depth index, sedation level, MAP and respiratory effects still come from the illustrative PD in engine/pd.js. They are driven by Ce normalised to the peak Ce of a 2 mg/kg bolus in the Eleveld reference individual. The Eleveld BIS model is not used, and the depth index is not BIS.

Cross-checking mattered: one implementation's Schnider function (tci 0.2.1 pkmod_schnider) uses lean body mass where Schnider's CL equation uses weight. This was noticed while comparing sources, and Schnider is not used here.

Epinephrine and IV fluids

• Epinephrine (S40, S41, S44, S45): rescue = anaphylaxis IM 0.3 mg (range 0.3–0.5 mg; label maximum 0.5 mg per injection, hard stop), modelled with an illustrative IM absorption constant that slows when MAP is very low. rescue_iv (2026-11 audit) = perioperative anaphylaxis in an anaesthetised, monitored adult with IV access: 50 micrograms IV, titrated/repeated, then an infusion if several doses are needed (UK guideline dose, S44/S45; not an FDA label dose; specialist use). Each IV bolus is stopped at 0.05 mg (routeBolusMax, simulator policy using the guideline titration step); a bolus without a route is IM. maintenance = septic-shock infusion 0.05–2 mcg/kg/min on ideal body weight. The 2 mcg/kg/min hard stop is the top of the labelled range, which the label does not call a maximum. Warnings: CAD, and HFrEF as a policy mapping of "cardiomyopathy". The label lists no contraindications.
• IV fluids (2026-11 additions). Types: lactated Ringer's (S47) and 0.9% sodium chloride (S48); composition and warnings come from the labels, which give no fixed dose. Actions: fluid_bolus:<250|500>[:<fluidId>] (no type = FLUID_MODEL.DEFAULT_FLUID, lactated Ringer's), fluid_maintenance:<fluidId>:<mL/h> (0 = stop; input limit 1000 mL/h) and fluid_maintenance_stop; plan steps kind: 'fluid' (maintenance, continues across phases until changed) and kind: 'fluid_bolus'. state.fluids = { crystalloidMl, infusingMl, boluses, byType: { [fluidId]: { bolusMl, maintenanceMl } }, maintenance }; summary().fluids.byType gives totals by type. The default maintenance rate (100 mL/h), the input limit and the bolus sizes are simulator conventions (UI choices). Both fluids use the same illustrative volume model (see Model assumptions); chloride, lactate, potassium and acid-base effects are not modelled. Label cautions are logged once when they match the patient (LR: hyperkalaemia, liver disease/severe hepatic impairment; 0.9% NaCl: sepsis → metabolic-acidosis caution, HFrEF → fluid-overload caution — both simulator mappings). SMART (S50) is shown as an evidence note only. S43 is still logged as context text for sepsis patients at 30 mL/kg.
• Antibiotic prophylaxis (2026-11 additions; S46, timing only). Cefazolin 2 g (3 g if ≥ 120 kg; adjust when: weight>=120), clindamycin 900 mg, vancomycin 15 mg/kg (total body weight — the guideline weight basis was not confirmed, simulator convention). Window: dose started within 60 min before incision (vancomycin 120 min). Redosing interval from the start of the last dose: cefazolin 4 h, clindamycin 6 h, vancomycin not applicable. Selection: cefazolin, unless locked by the new allergy beta_lactam_severe (IgE-mediated or severe β-lactam reaction) → clindamycin, then vancomycin. Alarms (MEDIUM, logged): "Surgery started without antibiotic prophylaxis in the recommended window" at incision when prophylaxis is indicated; "Prophylaxis redose due" during surgery. Guided mode shows a checklist warning before Start surgery that the clinician must acknowledge (acknowledge_checklist:antibiotic); the full demonstration pauses for it; neither ever acknowledges it. "Indicated" is a plan flag (plan editor default: ticked for GA, unticked for MAC — simulator convention).
• Lidocaine pretreatment before propofol (2026-11 additions; S11): lidocaine.regimens.pretreatment = 10 mg fixed (label: 1 mL of a 1% solution). It counts toward the cumulative lidocaine total (summary().lidocaine: total mg, mg/kg IBW, pretreatment mg). A pretreatment dose and an IV lidocaine analgesic dose (Foo loading bolus or infusion) within 4 h of each other give a LOW alarm and a log/plan warning (Foo 2021, S1: ~4 h separation between local-anaesthetic interventions and IV lidocaine — applied to pretreatment as a simulator mapping). Injection pain is not modelled numerically; the teaching note states the label pH (7–8.5) and that pain is generally attributed to aqueous-phase propofol (mechanism not sourced).

Palliative and end-of-life modules (S100–S120)

Added 2026-11-05 for the owner-approved palliative pain management and end-of-life care modules (data/palliative.js, engine/palliative.js, ui/palliative.js). IDs start at S100 so they cannot collide with S55–S99 (another work stream). Machine-readable copy: data/palliative.js PALLIATIVE_SOURCES, merged into SOURCES at the end of data/drugs.js so the Sources panel lists them.

How these were checked. Every entry was verified through WebSearch records on 2026-11-05. WebFetch could not open any page (DNS blocked for journals.sagepub.com, eprints.lancs.ac.uk, ncbi.nlm.nih.gov, annalsofoncology.org and cdc.gov from the build environment), so no full text, table or PDF was opened. Each Verified via line says which records were seen and what was not found. Independent human verification by the owner is pending.

Framing. Palliative sedation relieves refractory suffering and is distinct from euthanasia and medical aid in dying: intent (symptom relief), proportionality (the lowest level of sedation that relieves the symptom) and titration to comfort distinguish it. Laws and policies differ by jurisdiction. The modules make no claims about survival, length of life or efficacy. Hard exclusions: no content, doses or guidance for medical aid in dying or euthanasia, and nothing about hastening death; tests/palliative.test.mjs scans the palliative files for violations.

S100

Cherny NI, Radbruch L; Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care. Palliat Med. 2009;23(7):581-593. https://pubmed.ncbi.nlm.nih.gov/19858355/ Identifiers: doi:10.1177/0269216309107024; PMID 19858355

• Supports: Sedation is an important and necessary therapy for selected palliative care patients with otherwise refractory distress; a 10-point framework for institutional guidelines covering how to prepare, organise, initiate, titrate, monitor and evaluate sedation. Distinguishes abuse (sedation aimed at hastening death) and injudicious use (e.g. demand generated by the family rather than the patient) from appropriate use; lists the use of inappropriate medications to achieve sedation, such as opioids, as substandard practice. Closeness to death is a criterion only for continuous deep sedation (exact time-frame wording not verified).
• Used by: palliative-sedation framework text, ethics panel, documentation gate (refractory, interdisciplinary, prognosis), "opioids are not used to sedate"
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Citation, DOI and PMID: PubMed 19858355, McMaster and Bar-Ilan research-portal records, eapcnet.org project page. Abuse / injudicious-use categories and "inappropriate medications to achieve sedation, such as opioids" seen in search excerpts of the SAGE reader page. "Closeness to death only relevant for continuous deep sedation" seen in a review comparing guidelines; the framework's own time-frame sentence was NOT found and is not quoted. Its Appendix 2 dose table was not seen; doses are taken from the 2024 revision (S101).

S101

Surges SM, Brunsch H, Jaspers B, et al. Revised European Association for Palliative Care (EAPC) recommended framework on palliative sedation: an international Delphi study. Palliat Med. 2024;38(2):213-228. https://pubmed.ncbi.nlm.nih.gov/38297460/ Identifiers: doi:10.1177/02692163231220225; PMID 38297460

• Supports: Revision of the 2009 framework (literature update + Delphi, 2020–2022; 42 statements with high/very high consensus; endorsed by the EAPC board January 2023). Terminology: "suffering" (physical, psychological, existential) and "refractory"; the principle of proportionality is part of the definition of palliative sedation; patient autonomy emphasised; stepwise pharmacological approach (step 1 midazolam); hydration is a separate, independent decision; no particular period of remaining life expectancy is specified. Step 1 midazolam: light palliative sedation 2.5 mg SC or 1.25 mg IV; deep palliative sedation 5–10 mg SC or 2.5–5 mg IV; a bolus of half the starting bolus may be repeated after 20 min SC or 5 min IV if necessary; continuous infusion: use 1 mg/h (SC and IV) and then adjust as needed; at doses higher than 10 mg/h consider adding or changing medication. Opioids (and haloperidol) should not be used to sedate a patient.
• Used by: midazolam step-1 doses, repeat interval, infusion start and 10 mg/h stop; proportionality; hydration as a separate decision; ethics panel
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Citation, DOI, PMID, method and consensus figures: PubMed 38297460, Lancaster and VUB portal records, bvsalud mdl-38297460, palliativeprojects.eu. Midazolam step-1 values seen in two independent records: a search excerpt of the SAGE PDF ("Step 1 · Midazolam · Light palliative sedation 2.5 mg SC 1.25 mg IV Deep palliative sedation 5–10 mg SC 2.5–5 mg IV Bolus with half of the starting bolus dosage may be repeated after 20 min SC or 5 min IV if necessary") and the EAPC e-book appendix table (eapcnet.pagetiger.com, page 57: "Use 1 mg/hr (SC and IV) and then adjust as needed"; "In the case of doses higher than 10 mg/hr, consider adding or changing medication"). "Opioids and haloperidol should not be used to sedate a patient" seen in the EAPC e-book page 50 excerpt. The table excerpts were fragmentary; the full table was not opened. Second-line agents (levomepromazine, propofol) were seen only in summaries and are NOT simulated.

S102

American Medical Association. Code of Medical Ethics Opinion 5.6: Sedation to unconsciousness in end-of-life care. https://code-medical-ethics.ama-assn.org/ethics-opinions/sedation-unconsciousness-end-life-care Identifiers: none

• Supports: Sedation to unconsciousness must never be used to intentionally cause a patient's death. Physicians may offer palliative sedation to unconsciousness for refractory clinical symptoms (not for existential suffering, which should be addressed through social, psychological or spiritual support); restrict it to patients in the final stages of terminal illness; document the rationale; obtain informed consent (patient or surrogate); discuss the degree and length of sedation and future life-sustaining treatment; monitor care once sedation begins. AMA opinions are ethics guidance, not standards of clinical practice or rules of law.
• Used by: ethics panel (intent, documentation, consent, jurisdiction note); consent and jurisdiction gate items; continuous-deep-sedation prognosis rule
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). WebSearch records of the AMA Code ethics-opinion page, the AMA-hosted 5.6 PDF and background report, and the 2013 Opinion 2.201 page. A 2019 resolution asking CEJA to review 5.6 was seen; its outcome was not found.

S103

American Medical Association. Code of Medical Ethics Opinion 5.7: Physician-assisted suicide. (Opinion 5.8, Euthanasia, is a separate opinion; its wording was not verified and is not quoted.) https://code-medical-ethics.ama-assn.org/ethics-opinions/physician-assisted-suicide Identifiers: none

• Supports: Definition used only to distinguish practices: physician-assisted suicide occurs when a physician facilitates a patient's death by providing the necessary means and/or information to enable the patient to perform the life-ending act; the opinion states that relief of suffering is a central goal of medicine, particularly at the end of life. Cited here for the distinction only; the simulator contains no content about these practices beyond the definitions.
• Used by: ethics panel definitions only (distinction)
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). WebSearch record of the AMA-hosted 5.7 PDF (2025-11 file) and the ethics-opinion page. Only fragments of 5.8 were seen (a page marked DRAFT), so 5.8 is named but not quoted.

S104

Ariadne Labs. Serious Illness Conversation Guide (updated May 2023). © 2015–2023 Ariadne Labs: A Joint Center for Health Systems Innovation (Brigham and Women's Hospital and Harvard T.H. Chan School of Public Health) and Dana-Farber Cancer Institute. Licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. https://www.ariadnelabs.org/serious-illness-care/ Identifiers: none

• Supports: Structure of a goals-of-care conversation (set up; assess illness understanding and information preferences; share prognosis; explore goals, fears and worries, sources of strength, critical abilities, trade-offs and family involvement; close with a recommendation and documentation). The simulator PARAPHRASES the topics only; it does not reproduce the guide's wording.
• Used by: goals-of-care topics (paraphrased) in the documentation gate and ethics panel
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Licence notice (CC BY-NC-SA 4.0, © 2015–2023 Ariadne Labs, "Updated May 2023") seen in search records of guide copies hosted by Blue Shield of California, TTUHSC, Fraser Health and the Aotearoa adaptation; Ariadne Labs' own licence page was not opened. The ariadnelabs.org URL is the program landing page as cited by those copies; it was not opened. Topic list paraphrased from those records.

S105

Dowell D, Ragan KR, Jones CM, Baldwin GT, Chou R. CDC Clinical Practice Guideline for Prescribing Opioids for Pain — United States, 2022. MMWR Recomm Rep. 2022;71(No. RR-3):1-95. https://www.cdc.gov/mmwr/volumes/71/rr/rr7103a1.htm Identifiers: doi:10.15585/mmwr.rr7103a1; PMID 36327391

• Supports: Scope: outpatients ≥18 y with acute (<1 month), subacute (1–3 months) or chronic (>3 months) pain; the recommendations do NOT apply to pain related to sickle cell disease, cancer-related pain treatment, palliative care or end-of-life care. Recommendations are voluntary and not inflexible standards. Rec. 4: before increasing total dosage to ≥50 MME/day, pause and carefully reassess individual benefits and risks; increases beyond 50 MME/day are progressively more likely to yield diminishing returns relative to risks. The 2022 guideline removed the 2016 recommendation to avoid ≥90 MME/day (it still refers to the 50 and 90 MME levels). Rec. 6 (acute pain): no greater quantity than needed for the expected duration of pain severe enough to require opioids. Rec. 8: offer naloxone when risk factors are present (implementation library trigger: ≥50 MME/day, concurrent benzodiazepines, substance use disorder history, sleep-disordered breathing). Rec. 9: review PDMP data. MME conversion factors (oral): morphine 1, oxycodone 1.5, hydromorphone 5 (hydromorphone changed from 4 in 2016) — for estimating total daily exposure, not for converting between opioids.
• Used by: MME factors (risk display), CDC scope statement, 50 MME/day pause + acknowledgement, naloxone-offer / PDMP / acute-pain teaching notes
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Citation and DOI copied from DOSING_SOURCES.md S21 and re-checked (read.qxmd.com record 36327391). Scope exclusion wording seen consistently in NCPA, AMiner, Oregon Pain Management Commission and ASA summaries. Rec. 4 and Rec. 6 wording in the CDC Opioid CDS implementation guide (fhir.org recommendation-04-05) and epocrates practice points. 90 MME change: Texas DSHS 2016-vs-2022 comparison, Minnesota Medical Association news, Analysis Group commentary. Rec. 8 triggers: fhir.org Library-OpioidCDSREC08. MME factors 1/1.5/5: New Jersey DURB newsletter reproducing the 2022 table. The MMWR page itself was not opened.

S106

World Health Organization. WHO Guidelines for the Pharmacological and Radiotherapeutic Management of Cancer Pain in Adults and Adolescents. Geneva: WHO; 2018. https://www.ncbi.nlm.nih.gov/books/NBK537492/ Identifiers: ISBN 978-92-4-155039-0; NCBI Bookshelf NBK537492

• Supports: NSAIDs, paracetamol and opioids, alone or in combination, should generally be used at the initiation of pain management; adjuvant steroids may be given to achieve pain control when indicated; a bisphosphonate should be used to prevent and treat bone pain in bone metastases; no recommendation for or against antidepressants for neuropathic cancer pain or for radioisotopes. Used for adjuvant teaching notes only; no dose is taken from it.
• Used by: adjuvant teaching notes (no doses)
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). NCBI Bookshelf records NBK537492 / NBK537479 / NBK537486 and the WHO IRIS record (ISBN and recommendation excerpts). The guideline's opioid titration and rescue-dose wording was NOT found in the records and is not used.

S107

Fallon M, Giusti R, Aielli F, et al; ESMO Guidelines Committee. Management of cancer pain in adult patients: ESMO Clinical Practice Guidelines. Ann Oncol. 2018;29(Suppl 4):iv166-iv191. https://academic.oup.com/annonc/article/29/Supplement_4/iv166/5046945 Identifiers: none confirmed (DOI and PMID not seen in a search record; not shown)

• Supports: Laxatives must be routinely prescribed for both the prophylaxis and the management of opioid-induced constipation [I, A]. The breakthrough dose is usually about 10%–15% of the total daily dose. Average relative potency of oral to intravenous morphine is between 1:2 and 1:3 [II, A] (oral to subcutaneous: 1:2 to 1:3 [IV, C]). Chronic kidney disease stage 4–5 (eGFR <30 mL/min): fentanyl and buprenorphine (transdermal or IV) are named as the safest opioids; in mild renal impairment smaller doses and longer intervals. Opioid switching may be used to improve analgesia or reduce adverse effects.
• Used by: bowel-regimen reminder, oral:IV morphine ratio 2:1–3:1, eGFR <30 morphine warning, ESMO breakthrough 10–15% note
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Records of the academic.oup.com and annalsofoncology.org full-text pages, ResearchGate PDF listing and a Spanish course-site PDF copy (laxative [I, A]; "+10% to 15% of the total daily dose"; oral:IV morphine 1:2–1:3 [II, A]; conversion table rows), plus Medical Independent and PMC review summaries (fentanyl/buprenorphine safest at eGFR <30). Author list beyond the first three not re-verified.

S108

National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Adult Cancer Pain (Swarm RA, et al.; J Natl Compr Canc Netw; and Version 2.2025, J Natl Compr Canc Netw. 2025;23(7):e250032). https://pmc.ncbi.nlm.nih.gov/articles/PMC5915297/ Identifiers: PMC5915297 (PMC record of an NCCN Adult Cancer Pain version; the issue was not pinned down)

• Supports: The rescue (breakthrough) dose is usually equivalent to 10% to 20% of the total daily dose, given every hour as needed (oral). If moderate-to-severe pain is unchanged or increased after reassessment, increase the dose by 50% to 100%. Reassess efficacy and adverse effects about 60 min after an oral dose (15 min after IV). Opioid rotation may be considered when pain is inadequately managed despite adequate titration.
• Used by: breakthrough 10–20%, PRN intervals (60 min oral / 15 min IV), titration +50–100% suggestion
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). "The rescue dose is usually equivalent to 10% to 20% of the total daily dose given every hour as needed" seen in a search excerpt of the PMC5915297 record; 10–20% also in an NCCN education module (2018) and MD Anderson 2025 algorithm citing NCCN. 50–100% increase and the 60 min / 15 min reassessment seen in Cancer Therapy Advisor and Medscape summaries of NCCN — secondary. Version 2.2025 (jnccn.org e250032) excerpts mention rotation after adequate titration; its numeric titration table was not seen. The NCCN guideline itself requires registration and was not opened. The 25–50% tier often quoted for pain 4–6 was NOT found and is not used.

S109

MASCC, ASCO, AAHPM, HPNA and NICSO expert panel. Opioid conversion in adults with cancer: MASCC-ASCO-AAHPM-HPNA-NICSO guideline. Support Care Cancer. 2025;33(3):243. https://link.springer.com/article/10.1007/s00520-025-09286-z Identifiers: doi:10.1007/s00520-025-09286-z

• Supports: Consensus conversion ratios (summary table): oral morphine → oral oxycodone 1.5:1 (34/38 panellists); oral morphine → oral hydromorphone 5:1 (33/37); subcutaneous → intravenous (any opioid) 1:1 (33/36). The panel did not reach consensus on routinely reducing the converted dose for incomplete cross-tolerance (clinicians commonly still reduce by 25–50%, per a UPMC commentary — not a guideline value). Recommendations cover pre-conversion assessment, conversion and post-conversion monitoring.
• Used by: conversion ratios 1.5:1 and 5:1; no default cross-tolerance reduction
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Springer article landing record (published 3 March 2025), PubMed 41197973 (clinical-insights review), MASCC guideline page and the MASCC summary-table / slide PDFs (ratio rows and panel agreement counts seen in search excerpts), ASCO Post summary and the UPMC "The Tablet" vol. 5 no. 7 commentary (cross-tolerance statement and "Support Care Cancer. 2025 Mar 3;33(3):243"). Author names not verified (listed as the expert panel). Parenteral morphine and hydromorphone rows were not seen and are not used.

S110

Morphine sulfate oral solution, 10 mg/5 mL and 20 mg/5 mL. FDA-approved prescribing information (e.g. DailyMed setid 9405556c-20e0-4e47-b1b0-7872bcc6517e; Precision Dose NDC 68094-001 label). https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=9405556c-20e0-4e47-b1b0-7872bcc6517e Identifiers: none (multiple ANDAs)

• Supports: Adults (opioid-naïve, initial dosage): 10 mg to 20 mg every 4 hours as needed, at the lowest effective dose, titrated to response. The 100 mg/5 mL (20 mg/mL) concentrate is for opioid-tolerant patients only.
• Used by: oral morphine starting dose 10–20 mg every 4 h
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). DailyMed listings (setids 9405556c-…, d8598654-…, ca359246-…, 242c2512-…) and ndclist.com 68094-001 label copy ("10 to 20 mg every 4 hours as needed" for the 10 mg/5 mL and 20 mg/5 mL strengths; concentrate restricted to opioid-tolerant adults).

S111

Morphine sulfate tablets (immediate release). FDA-approved prescribing information (label copies indexed at medlibrary.org/lib/rx/meds/morphine-sulfate-52/). https://medlibrary.org/lib/rx/meds/morphine-sulfate-52/page/7/ Identifiers: none (multiple ANDAs)

• Supports: Renal impairment: morphine pharmacokinetics are altered in renal failure; start these patients with a lower than usual dosage and titrate slowly, monitoring for respiratory depression, sedation and hypotension. Geriatric patients: titrate the dosage slowly. (No numeric reduction is stated.)
• Used by: morphine renal/geriatric wording
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). medlibrary.org label copy excerpts ("Start these patients with a lower than usual dosage of morphine sulfate tablets and titrate slowly"; "Titrate the dosage … slowly in geriatric patients"). The tablet starting-dose range was seen only in drug-reference summaries and is not used.

S112

Morphine sulfate injection, USP (NexJect / Carpuject). Hospira, Inc. (Pfizer). FDA-approved prescribing information. https://pfizermedicalinformation.com/morphine/highlights Identifiers: none

• Supports: Direct intravenous injection: the usual starting dose in adults is 0.1 mg to 0.2 mg per kg every 4 hours as needed for pain management. (The 50 mg/mL vial is for continuous infusion after dilution only and is not simulated.)
• Used by: IV morphine starting dose 0.1–0.2 mg/kg every 4 h
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Pfizer medical-information dosage and highlights pages (quoted: "The usual starting dose in adults is 0.1 mg to 0.2 mg per kg every 4 hours as needed"). A current DailyMed morphine injection listing was seen without the dosing excerpt.

S113

Oxycodone hydrochloride tablets (ROXICODONE) and oxycodone hydrochloride oral solution. FDA-approved prescribing information (DailyMed setid 5a8157d3-ba20-4abf-b0d4-6ad141b31d63; label copies at medlibrary.org and ndclist.com). https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=5a8157d3-ba20-4abf-b0d4-6ad141b31d63 Identifiers: none (NDA/ANDA numbers not re-verified)

• Supports: Initial dosage for patients not receiving opioid analgesics: 5 mg to 15 mg every 4 to 6 hours as needed; titrate to response; lowest effective dose for the shortest duration. Hepatic impairment and renal impairment: initiate with a lower than usual dosage and titrate carefully (no fraction stated). Patients with renal impairment (creatinine clearance <60 mL/min) had higher plasma oxycodone concentrations.
• Used by: oral oxycodone starting dose 5–15 mg every 4–6 h; renal/hepatic wording; CrCl <60 warning
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Starting range copied from DOSING_SOURCES.md S20 (DailyMed listing + drugs.com reproductions) and re-seen in an oxycodone oral solution label excerpt. "Initiate therapy in these patients with a lower than usual dosage … and titrate carefully" (hepatic and renal) in medlibrary.org Roxicodone page excerpts; CrCl <60 statement in an oxycodone tablet label excerpt. The "one-third to one-half" reduction was NOT found and is not used.

S114

Hydromorphone hydrochloride tablets (DILAUDID; and generic hydromorphone hydrochloride tablets). FDA-approved prescribing information. https://www.drugs.com/dosage/dilaudid.html Identifiers: none

• Supports: Initiate DILAUDID tablets in a dosing range of 2 mg to 4 mg orally every 4 to 6 hours as needed (lowest effective dose). Renal and hepatic impairment: initiate with one-fourth to one-half of the usual starting dose depending on the degree of impairment (exposure about 2-fold higher in moderate and 4-fold higher in severe renal impairment).
• Used by: oral hydromorphone starting dose 2–4 mg every 4–6 h; one-fourth to one-half in renal/hepatic impairment
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). drugs.com DILAUDID dosage page (January 2024 record) for 2–4 mg every 4–6 h; generic hydromorphone tablet label copy at medlibrary.org for "one-fourth to one-half the usual … starting dose depending on the degree of impairment" and the 2-fold/4-fold exposure data; the archived 2006 DILAUDID PI (FDA MedWatch copy) for lower starting doses in severe renal impairment. The renal wording was not seen in a current DILAUDID-brand tablet label.

S115

Hydromorphone hydrochloride injection, USP. Hospira, Inc. (Pfizer) and Fresenius Kabi labels (labeling.pfizer.com ShowLabeling id 4472; DailyMed setid e034eb96-03e0-46e1-8b92-5e1f94555e7b). https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=e034eb96-03e0-46e1-8b92-5e1f94555e7b Identifiers: none

• Supports: Intravenous starting dose 0.2 mg to 1 mg every 2 to 3 hours as needed, given slowly over at least 2 to 3 minutes; reduce the initial dose in elderly or debilitated patients (may be lowered to 0.2 mg); start hepatic or renal impairment at one-fourth to one-half of the usual starting dose. High-potency formulation (HP) only for opioid-tolerant patients.
• Used by: IV hydromorphone starting dose 0.2–1 mg every 2–3 h; elderly and renal/hepatic wording
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Copied from DOSING_SOURCES.md S19 (Pfizer labeling pages and DailyMed listing) and re-seen in Hikma / Fresenius Kabi label excerpts (elderly "may be lowered to 0.2 mg"; renal "one-fourth to one-half"; HP for opioid-tolerant only). Subcutaneous/IM starting doses were not confirmed and are not simulated.

S116

Scottish Palliative Care Guidelines (NHS Scotland, Right Decisions). Life-threatening opioid-induced respiratory depression requiring naloxone (2026; replaces the earlier "Naloxone" guideline). https://www.rightdecisions.scot.nhs.uk/shared-content/palliative-care/palliative-care/opioids/life-threatening-opioid-induced-respiratory-depression-requiring-naloxone/ Identifiers: none

• Supports: Naloxone only where there is concern about life-threatening respiratory depression; the aim is to reverse it without impacting symptom control or causing withdrawal. For people receiving palliative care: 100 microgram bolus doses (from 400 micrograms/mL, diluted); IV: 100 micrograms as a slow bolus every 2 minutes until respiratory status is satisfactory (≥8 respirations/minute); IM/SC: 100 micrograms every 5 minutes. Titrate to respiratory rate, not conscious level. Reversal of buprenorphine may be incomplete; methadone may need an infusion (not simulated).
• Used by: naloxone 0.1 mg increments every ≥2 min titrated to respiratory rate; RR <8 alarm
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Right Decisions pages (current guideline and the medicine-information page) and the guideline schedule (older naloxone guideline removed July 2026). The draw-up volume differs between pages (2 mL vs 2.5 mL of the diluted solution); the simulator uses the dose in milligrams only. The SIGN November 2025 consultation draft gives the same 100 microgram IV regimen.

S117

Hui D, Bohlke K, Bao T, et al. Management of dyspnea in advanced cancer: ASCO guideline. J Clin Oncol. 2021;39(12):1389-1411. https://pubmed.ncbi.nlm.nih.gov/33617290/ Identifiers: doi:10.1200/JCO.20.03465; PMID 33617290

• Supports: Non-pharmacological measures first (e.g. airflow from a fan directed at the cheek, positioning). Supplemental oxygen for hypoxaemic patients; not recommended when SpO2 >90%. Systemic opioids should be offered when non-pharmacological measures are insufficient. Short-acting benzodiazepines for dyspnoea-related anxiety persisting despite opioids. Continuous palliative sedation should be offered for dyspnoea refractory to all standard and palliative options in patients with an expected life expectancy of days.
• Used by: dyspnoea panel (fan, oxygen only if hypoxaemic, opioids); continuous palliative sedation with life expectancy of days
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). PubMed 33617290, zendy.io (DOI 10.1200/jco.20.03465), ASCO Post and The Limbic summaries, medicaldialogues summary (oxygen and SpO2 >90% wording; continuous palliative sedation and "days"). Full text not opened; recommendation strengths not shown.

S118

Wee B, Hillier R. Interventions for noisy breathing in patients near to death. Cochrane Database Syst Rev. 2008;(1):CD005177 (assessed as up to date 2017). https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD005177.pub2/information Identifiers: doi:10.1002/14651858.CD005177.pub2

• Supports: Four studies included; no evidence that any intervention, pharmacological or non-pharmacological, was superior to placebo for noisy breathing (terminal respiratory secretions). Used as the evidence note for the secretions panel; no dose is simulated.
• Used by: secretions panel evidence note (no dose)
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). Cochrane Library information and abstract-PDF records (CD005177.pub2), cochrane.org plain-language summaries and the PEDro record. Anticholinergic doses seen only in secondary GP notebook pages citing NICE are NOT used.

S119

Bush SH, Lawlor PG, Ryan K, et al; ESMO Guidelines Committee. Delirium in adult cancer patients: ESMO Clinical Practice Guidelines. Ann Oncol. 2018;29(Suppl 4):iv143-iv165. https://eprints.whiterose.ac.uk/130653/ Identifiers: doi:10.1093/annonc/mdy147 (seen in a search summary only)

• Supports: Prompt diagnosis and treatment of precipitating causes may reverse many delirium episodes; non-pharmacological measures and psychological support first; antipsychotics and benzodiazepines should be minimised and reserved for failure of non-pharmacological intervention; delirium is common in advanced disease and in the last hours or days of life. No drug dose is simulated.
• Used by: delirium panel note; delirium-risk display note
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). White Rose eprint 130653, ESMO guideline page and university portal records (citation and quoted recommendations). Author list beyond the first two not re-verified.

S120

Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002;166(10):1338-1344. https://pesquisa.bvsalud.org/colombia/resource/es/mdl-12421743 Identifiers: PMID 12421743

• Supports: 10-level scale from +4 (combative) to −5 (unarousable); 0 = alert and calm; −1 to −5 are sedation levels. Validated in adult ICU patients; the simulator uses it as the depth scale for palliative sedation targets (a palliative adaptation exists but was not verified and is not used).
• Used by: RASS scale for sedation targets and the unintended-sedation alarm
• Verified via: Verified 2026-11-05 through WebSearch records only (direct page fetches were not possible from the build environment: DNS blocked for publisher, NCBI, CDC and DailyMed hosts). bvsalud MEDLINE record mdl-12421743 (citation, scale range, reliability), NIH CDE RASS form, Merck Manual table. Level descriptors vary slightly between reproductions; the simulator shows short descriptors only.

Palliative modules — not simulated, and why

Palliative modules — model assumptions (illustrative)

Every symptom score, RASS value and respiratory rate in these modules is illustrative output of engine/palliative.js (constants in MODEL, all author_estimate_unvalidated, listed in MODEL_CARD.md): one-compartment opioid kinetics in oral-morphine-equivalent units with first-order oral absorption and an effect site; potency for the effect model only from the sourced ratios (oral) and author estimates (IV morphine 2.5, IV hydromorphone 12.5 — not used for conversion); slower elimination with eGFR <60/<30 and hepatic impairment; sigmoid analgesia, sedation and respiratory-rate effects scaled to opioid tolerance; seeded pain episodes; illustrative midazolam, naloxone, fan/positioning/oxygen, secretions, nausea and constipation terms. No survival or mortality is modelled.

Advisory haemodynamic rescue (November 2026; no new sources)

The advisory (engine/rescue-config.js, engine/rescue.js) reuses existing registry entries only: phenylephrine rescue bolus 40–100 mcg, additional boluses every 1–2 minutes, ≤ 200 mcg in all [S18]; ephedrine 5–10 mg IV bolus as needed, ≤ 50 mg [S19]; perioperative-anaphylaxis epinephrine guidance [S44, S45] (named only; epinephrine is never suggested). Simulator conventions (no source): the agent-by-HR rule and its 60 /min threshold; the ephedrine and other-vasopressor reassessment pauses (120 s; the ephedrine label states no interval); choosing the regimen lower bound; suppression during sensor faults and the anaphylaxis scenario; the MAP trigger and high-MAP thresholds are the simulator's alarm thresholds (build specification), not clinical standards. Sedation titration in guided mode reuses S11 (propofol MAC-sedation range 25–75 mcg/kg/min), S17 (midazolam repeat ≥ 4 min) and S32 (allow time between doses; no number); its step size and 180-s reassessment pause are conventions.

Not included, and why

Model assumptions (illustrative)

• All drugs except propofol use pkModel: 'calibrated'. The engine fits illustrative constants to the timing values above, or uses its own fallbacks when those are null. Propofol PK (with ke0) is the published Eleveld 2018 model (S42). Apart from that and the Mapleson age-adjusted MAC equation (S26), no PK/PD parameter is a published model. Every PD effect size (sigmoid weights, EC50s, gammas), including propofol's, is an author estimate. sim.modelInfo().inventory lists every parameter with its provenance.
• Epinephrine PD (engine/pd.js PROFILES.epinephrine, author estimates): concentration is normalised to steady state at 0.05 mcg/kg/min (u = 1). At u ≈ 1: MAP +8%, HR +6%, bronchodilation 0.5. A second, higher-EC50 component (mapHigh/hrHigh 0.35, EC50 10, γ 2; 2026-11 audit) keeps the response rising at high rates instead of plateauing: simulated healthy 80 kg adult, 10 min infusion — 0.05 → MAP +6%, 0.2 → +23%, 0.5 → +40%, 1 → +54% (HR similar), monotonic. Bronchodilation counters the bronchospasm (× (1 − 0.8·B)) and anaphylaxis (× (1 − 0.7·B)) scenario intensities. PK: k10 0.3, k12 0.05, k21 0.01, ke0 1.0 min⁻¹. IM absorption ka 0.15 min⁻¹ (SC 0.07), multiplied by a perfusion factor (ROUTE_PERFUSION_ILLUSTRATIVE: 1 at MAP ≥ 80% of baseline, falling linearly to 0.1 at ≤ 40% of baseline). Under GA with anaphylaxis (MAP ≈ 28), 0.5 mg IM now takes ≈ 5 min to bring MAP near baseline (was ≈ 2–3 min); a 50 mcg IV bolus acts within a minute and fades over several minutes, so repeat boluses or an infusion are needed. All illustrative and not from the label.
• 2026-11 audit PD recalibrations (all author estimates, illustrative):
  • dexmedetomidine hypnotic weight 0.55 → 0.35: labelled procedural sedation (1 or 0.5 mcg/kg over 10 min, then 0.6–1 mcg/kg/h) gives a depth index of ≈ 70–73 ("moderate") in the healthy and elderly examples instead of 57–64 ("general"); bradycardia unchanged (HR 72 → 58; elderly on a beta-blocker 64 → 52).
  • midazolam hypnotic EC50 0.6 → 1.0: 1 mg in the healthy example gives minimal sedation (depth ≥ 80); 2.5 mg gives moderate (≈ 75).
  • nitrous oxide hypnosis weight PATIENT_MODS.N2O_HYPNOSIS_FRACTION 0.5: the N2O MAC fraction counts half toward hypnosis (summed MAC still drives immobility, analgesia and respiratory effects). 70% N2O alone now gives depth ≈ 77 ("moderate"), not 55.
  • ketamine airway tone PATIENT_MODS.KETAMINE_AIRWAY_FRACTION 0.3: only 30% of ketamine's hypnotic effect counts toward upper-airway obstruction (effects.Hairway). Ketamine 2 mg/kg in the obese OSA example: minimum RR 17 (propofol 2 mg/kg: apnoea).
• Sugammadex–rocuronium binding (2026-11 audit): 1:1 molar, instantaneous, in the central compartment, using the label molecular weights (rocuronium bromide 609.70; sugammadex sodium 2178.01, with 100 mg sugammadex = 108.8 mg sugammadex sodium; S12, S6). Free sugammadex has illustrative slow elimination (ILLUSTRATIVE_PK.sugammadex k10 0.006 min⁻¹). The rocuronium effect-site rate constant is multiplied by 1 + 9·x⁴/(x⁴ + 1.5⁴), x = free sugammadex : rocuronium-in-body molar ratio (STIMULUS.SUGAMMADEX_*; author estimates). Result (80 kg adult, illustrative): label-tier doses reach TOFR 0.9 in ≈ 1.5–2 min (2 mg/kg at T2: 92 s; 4 mg/kg 5 min after 0.6 mg/kg: 85 s; 16 mg/kg 3 min after 1.2 mg/kg: 112 s); under-dosing is slower or incomplete (2 mg/kg at TOF 0: 203 s; 4 mg/kg 3 min after 1.2 mg/kg: 284 s; 2 mg/kg 3 min after 1.2 mg/kg: TOFR 0.8 at 60 min). A dose below the label tier for the current depth (TOF count ≥ 2 → 2 mg/kg; TOF 0–1 → 4 mg/kg; TOF 0 within 3 min of rocuronium ≥ 1.2 mg/kg → 16 mg/kg; post-tetanic count not modelled) gives a warning in the log/feedback and a medium alarm.
• Capnography during apnoea (2026-11 audit): with no breaths and no ventilatory support there is no exhaled gas, so state.vitals.etco2 is null and the monitor shows "— (no breaths)". The internal PaCO2 keeps rising and drives physiology.
• IV fluid bolus (engine/sim.js FLUID_MODEL, author estimates): 250 or 500 mL crystalloid, default 500 mL. Bolus sizes are UI choices, not recommendations. Run-in is 50 mL/min. Retained intravascular volume R decays with τ = 40 min. Restoration r = 1 − exp(−R/750 mL), which gives diminishing returns. A volume-loss event (hemorrhage) has its intensity multiplied by (1 − 0.8·r). Cap: fluids offset at most 80% of the event, because ongoing bleeding needs surgical control. Direct MAP effect is +4%·r when euvolaemic and +10%·r with the hypovolemia or sepsis condition. With these constants, 500 mL raised MAP by about 12 mmHg in a hemorrhage case and about 2 mmHg in a euvolaemic patient (fictional 80 kg adult). 2026-11 additions: a maintenance infusion adds its volume directly to R at its rate (no run-in queue); lactated Ringer's and 0.9% sodium chloride use identical constants (no type-specific physiology).
• Awake room-air oxygenation (fix, 2026-11-05): FRC (obesity/OSA/elderly/COPD) now scales only the O2 store, which sets how fast SpO2 falls during apnoea. It no longer scales the steady-state alveolar balance. Before the fix, an awake obese patient with no drugs settled near SpO2 40% on room air. Regression tests check all example patients stay within 2% of baseline, and that the obese patient still desaturates faster during apnoea.
• Additive MAC of co-administered inhaled agents (LIBRARY_NOTES.inhaledMacAdditivity). The engine treats MAC fractions as additive. This is standard teaching. No acceptable source was verified for it in this session, so it is uncited.
• Isoflurane and desflurane also use the Mapleson equation (MAC40 1.17% and 6.6%, the same slope). The isoflurane label table (1.28/1.15/1.05% at mean ages 26/44/64 y) is stored for display.
• N2O and FiO2: the maximum FiO2 is 1 − FiN2O (70% N2O leaves at most 0.30). This is simple arithmetic, so no source is cited.
• N2O cannot work alone: its MAC40 is 104% (Mapleson), so it cannot provide anaesthesia alone at 1 atm and is modelled only as an adjunct.
• Depth-high (awareness-risk) alarm (final-check fixes, 2026-11; engine/timing.js DEPTH_HIGH_ALARM_*): HIGH when the simulated processed-EEG index stays above the GA band upper bound 60 (B-Unaware [S52] / BAG-RECALL [S53] protocol band, reused) for ≥ 60 s during GA with the airway secured; clears at ≤ 58. The 60-s sustain, the 58 hysteresis and the HIGH priority are simulator conventions (no source gives an alarm duration); the index itself is an illustrative model output.
• Depth-low advisory, text only (owner-approved, 2026-11; engine/depth-advisory.js, engine/timing.js DEPTH_LOW_ADVISORY_*): a card (not an alarm) when the simulated index stays below the GA band lower bound 40 (B-Unaware [S52] / BAG-RECALL [S53] protocol band, reused) for ≥ 120 s during GA maintenance after incision; ends at ≥ 42. The 120-s sustain and the 42 hysteresis are simulator conventions; no new source was added. It suggests no drug, dose or rate and cites no outcome evidence for deep anaesthesia.
• Maintenance-hypnotic classes (final-check fixes, engine/plan-safety.js MAINTENANCE_HYPNOTIC_CLASSES): only volatile agents and hypnotic infusions (propofol) count as a GA maintenance hypnotic for the lockout check; N2O, dexmedetomidine and ketamine do not. A teaching rule of the simulator, not a sourced statement.
• Sevoflurane carries two MAC references: Mapleson MAC40 = 1.80% and the US label table (2.1% at 40 y in O2). They differ. The engine's effect model uses Mapleson, and the UI may show both.

Verification and validation status

Terms follow the FDA computational-modelling credibility vocabulary (verification = the code does what the model says; validation = the model matches the real system). This simulator has not been validated.

• Verification (automated tests, cd simulator && npm test, 145 tests passing after the 2026-11 audit; browser smoke test npm run test:ui, 17 subtests):
  • drug-library schema and source integrity (drugs.test.mjs): ids, ranges, when grammar, that every referenced S-id exists, and required contraindication rules;
  • dosing arithmetic and ceilings (dosing.test.mjs);
  • patient derivations: BMI, IBW/LBW/ABW, Apfel, STOP-Bang, flags (patient.test.mjs);
  • parity with the Python Smart NOA controller (parity.test.mjs): reproduces the EXAMPLE_RESULTS.json GREEN/RED/HOLD decisions and the test_safety_hardening.py regressions (NaN/None fail-safe, direct-write bypass, ceilings, matching);
  • engine behaviour (sim.test.mjs, run against a fixture library and the real library): PK mass balance and steady state, calibrated fits to label timing, determinism per seed, scenarios, alarms, interlocks, full GA/MAC runs;
  • model extensions (models.test.mjs): the Eleveld 2018 implementation reproduces published reference-individual values (V1 6.28, V2 25.5, V3 273 L; CL 1.79/2.10, Q2 1.83, Q3 1.11 L/min; ke0 0.146) and the opioid-switch values; epinephrine regimens, route block and qualitative effects; fluid tracking and qualitative response; the parameter inventory (every parameter has a provenance and existing source ids); and the awake room-air SpO2 regression. Reproducing published numbers checks the implementation only. It does not show that simulated patients behave like real ones.
• Face validity: no expert (clinician) review is recorded.
• External validation: none. No simulated output has been compared with patient data, monitor recordings or published clinical time courses.
• Parts that cannot be validated with the sources available: all PD effect sizes (depth index, MAP/HR/respiratory effects, opioid synergy, NMB/TOF mapping); the calibrated PK constants (fitted only to label onset/duration statements); epinephrine PK/PD and IM absorption; the fluid-response model; physiology gains and time constants (stimulus, apnoea, CO2, noise); scenario perturbation magnitudes; patient-modifier multipliers; alarm and extubation thresholds (simulator choices); and the depth index itself, which is not BIS. sim.modelInfo().inventory labels each of these author_estimate_unvalidated.