What this course is, and how to use it
This is the eleventh course in University Medicine: pharmacology 药理学 — what the body does to a drug (kinetics) and what a drug does to the body (dynamics). MED02 supplied the receptor chemistry, MED05 the physiology, MED08 the antimicrobial targets.
Scope limit, stated once: a study course, not a medical degree; every case is fictional, every number is an illustrative teaching value inside a labelled calculation, drug names are international nonproprietary names (INN), and nothing here is prescribing guidance or a regimen.
- Units 11.1–11.2 are the engine: pharmacokinetics and pharmacodynamics.
- Units 11.3–11.6 are systems: autonomic, cardiovascular/renal, antimicrobial/anticancer, CNS and analgesia.
- Unit 11.7 is safety: adverse reactions, interactions, error prevention.
The pharmacology habit: mechanism predicts effect, arithmetic predicts exposure — every curve below is read in both directions, and every half-life is something you recompute.
Pharmacokinetics
Why it matters. Kinetics turns a dose into a concentration — and concentration, not dose, is what the body sees. Half-life 半衰期, clearance and volume are the three levers.
The idea, with a picture. Volume of distribution 表观分布容积 $V_d$ (the apparent volume a drug "occupies"), clearance 清除率 $CL$ (volume cleared per time), linked by $t_{1/2} = 0.693 \cdot V_d / CL$ ✓; absorption with bioavailability 生物利用度 $F$ (fraction reaching circulation); first-pass metabolism 首过代谢 cuts oral $F$ ✓.
The derivation (why 4–5 half-lives reach steady state). Repeated dosing adds drug each interval while clearance removes a fixed fraction: the gap closes exponentially toward steady state 稳态 — after 3.3 half-lives ~90 % of steady state, after 5 half-lives ~97 % ✓. A loading dose 负荷剂量 fills the tank at once: $D_L = C_{target} \cdot V_d / F$ (illustrative calculation: target 10 mg/L, $V_d$ 40 L, $F$ 1.0 → 400 mg — a teaching value, not a prescription) ✓.
Worked example 1 — the three levers (illustrative values). $V_d$ 40 L, $CL$ 5 L/h: $t_{1/2} = 0.693 \times 40 / 5 = 5.5$ h ✓. Halve clearance (liver disease): $t_{1/2}$ doubles to 11.1 h — the same dose now accumulates; steady state is later and higher ✓.
Worked example 2 — clinical link: the first-pass toll. (Fictional teaching value) a drug with oral $F$ = 0.6: of a labelled 100 mg oral dose, 60 mg reaches the circulation — the liver extracts 40 % on the first pass ✓. Portal failure or a bypass raises $F$ and the same tablet behaves like a bigger one: kinetics, not pharmacology, changed the effect.
Common mistakes. Equating half-life with duration of effect (effect depends on concentration vs the drug's target range). Reading $V_d$ as a real volume (it is a proportionality constant — about 3 L means the drug stays in plasma, 500 L means tissue binding). Forgetting maintenance dosing aims at $CL \cdot C_{ss}$.
Summary. $t_{1/2} = 0.693\,V_d/CL$; steady state ≈ 90 % at 3.3 $t_{1/2}$, 97 % at 5; loading $C_{target}V_d/F$; oral $F$ = fraction past the first pass; clearance sets maintenance, volume sets the load.
Pharmacodynamics
Why it matters. Dynamics is the grammar of every drug chart: what a receptor does with a molecule — potency, efficacy, and the antagonists that undo both.
The idea, with a picture. The dose–response curve (log dose): position along the x-axis = potency 效价 (EC₅₀); ceiling = efficacy 效能 ($E_{max}$) ✓. Agonists reach $E_{max}$; partial agonists 部分激动剂 reach a lower ceiling (and displace full agonists); competitive antagonists shift the curve RIGHT in parallel; non-competitive lower $E_{max}$ ✓.
The derivation (why the therapeutic index is a ratio). Therapeutic index 治疗指数 $TI = TD_{50}/ED_{50}$ — the toxic dose over the effective dose, each read from its own cumulative curve (illustrative: $TD_{50}$ 100 mg, $ED_{50}$ 10 mg → TI 10) ✓. A drug with TI 2 asks for care; TI 100 forgives. Narrow-TI drugs (the ones needing level monitoring) are those whose toxic and effective curves overlap.
Worked example 1 — reading a curve pair. (Fictional curves) drug A: EC₅₀ 2 mg, $E_{max}$ 100 %; drug B: EC₅₀ 20 mg, $E_{max}$ 100 % — B is 10× less potent but equally efficacious; drug C: EC₅₀ 2 mg, $E_{max}$ 60 % — a partial agonist, potent but capped ✓. Three claims, three curve features.
Worked example 2 — clinical link: the antagonist fork. (Fictional) a competitive antagonist co-administered: the curve shifts right — the agonist's effect can be overcome by concentration (surmountable). A non-competitive antagonist cuts $E_{max}$ — no concentration wins ✓. The overdose logic follows: one is a tug-of-war, the other removes the rope.
Common mistakes. Calling a potent drug a strong one (potency is dose-position; efficacy is the ceiling). Expecting a partial agonist to behave like a weak full agonist in combination (it COMPETES — lowering the mixture's ceiling). Reading TI from a single curve (it needs two: effect and toxicity).
Summary. Potency = EC₅₀ position; efficacy = $E_{max}$; partial agonist = lower ceiling that displaces; competitive → parallel right shift (surmountable); non-competitive → lower ceiling; $TI = TD_{50}/ED_{50}$ (illustrative 100/10 = 10).
Autonomic pharmacology
Why it matters. The autonomic receptor map (MED09 §5) is the most-prescribed territory in medicine — every agonist and blocker is a prediction exercise, and the toxidromes 中毒综合征 are its emergencies.
The idea, with a picture. The code: α1 constricts vessels and dilates pupils; β1 drives the heart; β2 opens airways; M3 drives secretion, constricts the pupil and speeds the gut, while M2 slows the heart ✓. Agonists push the receptor; blockers remove its tone.
The derivation (why toxidromes are the map upside down). Cholinergic excess = everything M stands for: salivation, lacrimation, miosis, bronchorrhoea, bradycardia, gut cramps and diarrhoea ✓. Anticholinergic = M blocked: dry mouth, dilated fixed pupils, tachycardia, urinary retention, hot dry skin ✓. Adrenergic excess = α+β on: tachycardia, hypertension, mydriasis, sweating. The syndrome is the arithmetic of the map.
Worked example 1 — predict four drugs. (Fictional INN teaching set) a β1 blocker lowers heart rate and contractility (cardio-selective logic); an α1 blocker lowers blood pressure and can cause postural drop; a β2 agonist opens airways with little cardiac effect; an M blocker dries secretions and blurs vision ✓ — four predictions straight off the table.
Worked example 2 — clinical link: two overdoses, two maps. (Fictional) patient A: pinpoint pupils, sweating, cramps and diarrhoea — cholinergic; patient B: dry, dilated, fast — anticholinergic ✓. Each finding is one receptor row; the collection names the class before any test returns.
Common mistakes. Expecting β-blockade to lower blood pressure instantly (output falls first; the full antihypertensive effect takes weeks of remodelling). Forgetting muscarinic versus nicotinic (M effects dominate toxidromes; N is ganglia and the NMJ). Treating sweating as an adrenergic sign only (sweat glands are sympathetic but cholinergic, at muscarinic receptors — the MED09 exception, which is why the anticholinergic patient is hot and dry).
Summary. α1 constrict/dilate pupil · β1 heart · β2 airway · M3 secrete/constrict pupil/speed gut, M2 slow heart; agonist = push, blocker = silence; cholinergic = wet and slow; anticholinergic = dry, blind, fast; adrenergic = fast and tight; predictions come off the map, not memory.
Cardiovascular and renal drugs
Why it matters. Hypertension, heart failure and thrombosis are mechanism-rich: each drug class is a named pressure-regulating or clotting step turned down, and the nephron is the diuretic 利尿剂 map.
The idea, with a picture. Antihypertensives by mechanism: diuretics (volume), ACE inhibitors ACE抑制剂 (block angiotensin II formation; the bradykinin cough), angiotensin receptor blockers, calcium-channel blockers (vasodilation), β1 blockers (output and renin) ✓. Diuretics by nephron site: loop (Na⁺-K⁺-2Cl⁻, thick ascending, ~25 % of Na⁺), thiazide (Na⁺-Cl⁻, distal tubule), potassium-sparing (collecting duct) ✓.
The derivation (why anticoagulants monitor differently). Unfractionated heparin acts immediately and is monitored with the aPTT; warfarin 华法林 blocks vitamin-K clotting-factor synthesis — existing factors must decay first, so onset lags days and the INR measures the effect, not the dose ✓; direct oral anticoagulants 抗凝药 target one factor (Xa or thrombin) with predictable kinetics, mostly unmonitored. The monitoring test is a property of the mechanism.
Worked example 1 — site to electrolyte. (Fictional) loop diuretic at the thick ascending limb: powerful Na⁺ loss, K⁺ loss (downstream flow), and Ca²⁺ loss; a potassium-sparing agent at the collecting duct: mild, K⁺-retaining ✓. Each site's chemistry predicts its electrolyte fingerprint.
Worked example 2 — clinical link: the cough that isn't infection. (Fictional) on an ACE inhibitor, a dry persistent cough: bradykinin no longer degraded in that pathway ✓ — mechanism names the side effect, and the class switch (to an ARB) follows from the mechanism, not trial and error.
Common mistakes. Placing thiazides in the loop (they act distally — weaker, slower, Ca²⁺-retaining). Expecting warfarin to act on factors already made (it blocks CARBOXYLATION of new ones). Confusing anticoagulants (clots) with antiplatelets (platelet plugs — different pipeline, different monitoring).
Summary. Volume/ACE/ARB/CCB/β1 = the antihypertensive mechanisms; loop 25 %, thiazide distal, K-sparing collecting duct; heparin→aPTT now; warfarin→INR after factor decay; DOACs one factor, predictable; mechanism names the side effect.
Antimicrobial and anticancer drugs
Why it matters. The two attack maps share one logic — selective toxicity 选择性毒性 against cells that differ from ours — and the plan defers to MED08 §6 for the antibacterial targets.
The idea, with a picture. Antibacterial targets (MED08): wall (β-lactams), ribosome, nucleic-acid enzymes, folate ✓. Anticancer: cell-cycle-specific 细胞周期特异性 agents hit cycling cells — S-phase (antimetabolites: methotrexate, fluorouracil), M-phase (vinca alkaloids, taxanes) — while alkylators damage DNA regardless of cycle ✓.
The derivation (why selectivity differs between the fields). Antibacterial selectivity rides on structures human cells LACK (wall, different ribosome) — a wide margin. Anticancer selectivity rides on TIMING (cycling cells take the hit) — a narrow margin, because marrow, gut and hair cycle too ✓. The margin predicts the side-effect profile before any list is memorised.
Worked example 1 — the arithmetic of resistance (from MED08). β-lactamase cleaves the ring (enzyme); altered PBP2a is target reshaping; the $10^{-7}$–$10^{-9}$ per-cell mutation supply explains why combinations outlast single agents ✓ — the pharmacology of failure is arithmetic.
Worked example 2 — clinical link: schedule is mechanism. (Fictional teaching schedule) an S-phase antimetabolite given as a short bolus misses cells not in S at that moment — the same total dose fractionated over time meets more cells in their vulnerable phase; an M-phase vinca depends on cells arriving at mitosis ✓. Cycle specificity dictates scheduling logic.
Common mistakes. Expecting anticancer "selectivity" to spare normal cycling tissue (marrow and mucosa cycle — the predictable toxicities). Treating resistance as a drug failure only (it is evolution in a dish: pressure selects). Calling antimetabolites non-specific (they are S-phase SPECIFIC).
Summary. Antibacterial: wall/ribosome/DNA/folate (MED08). Anticancer: S-phase antimetabolites, M-phase vinca/taxanes, alkylators cycle-independent; selectivity by ABSENCE (bacteria) vs TIMING (cancer); schedule follows the cycle.
CNS drugs and analgesia
Why it matters. CNS pharmacology is receptor physiology named: opioids, sedatives and antidepressants each sit at a known synapse — and tolerance 耐受性 and dependence 依赖性 are the learning the synapse does.
The idea, with a picture. Opioids 阿片类 act at μ receptors: analgesia with respiratory depression, constipation, miosis; tolerance and dependence develop ✓. NSAIDs block cyclo-oxygenase (COX-1 housekeeping/COX-2 inflammatory): less prostaglandin → less pain and inflammation, but gastric and renal cost ✓. Benzodiazepines are positive allosteric modulators at GABA-A; SSRIs block serotonin reuptake (weeks to full effect — receptor adaptation, not the synapse alone).
The derivation (why tolerance is physiology, not weakness). Chronic receptor activation down-regulates and desensitises: the same concentration does less — larger doses restore effect (tolerance), and abrupt withdrawal unmasks the adapted state (dependence) ✓. This is why benzodiazepines are short-course and why opioids' effective dose climbs; the WHO analgesic ladder (1986, revised 1996) is the classic framework naming pain severity to drug class.
Worked example 1 — three pains, three sites. (Fictional) prostaglandin-mediated inflammation — a NSAID at the peripheral COX site; severe post-operative pain — an opioid at μ receptors (with the respiratory price); neuropathic burning — neither class alone; agents acting on neuronal excitability ✓ — matching the site to the drug, not the severity alone.
Worked example 2 — clinical link: the paracetamol lesson. (Fictional teaching case) an overdose of paracetamol saturates the safe conjugation pathways and uses up glutathione, so the toxic metabolite (NAPQI) binds liver proteins — the MED10 necrosis rulebook applies ✓. Mechanism names both the antidote logic (acetylcysteine restores glutathione) and the organ at risk.
Common mistakes. Equating analgesic classes by strength (they act at different SITES — combinations are additive by mechanism). Expecting SSRIs to work in days (reuptake blockade is immediate; clinical effect needs weeks of adaptation). Forgetting constipation is the opioid effect that barely tolerises.
Summary. Opioid μ: analgesia + respiratory depression + constipation, tolerance/dependence; NSAID COX-1/2: periphery, gastric/renal cost; benzodiazepine GABA-A PAM; SSRI reuptake, weeks; tolerance = down-regulation; paracetamol = glutathione and liver.
Adverse reactions, interactions and safe-medication principles
Why it matters. Every drug chart is an experiment with adverse drug reactions 药物不良反应, interactions and human error — the last unit is where the course's mechanisms become safety habits.
The idea, with a picture. Type A reactions: DOSE-RELATED, predictable from pharmacology (β-blocker bradycardia); type B: idiosyncratic, not dose-related (allergy, rare toxicity) ✓. Interactions: enzyme induction 酶诱导 (rifampicin, phenytoin → a co-medication's levels fall) versus enzyme inhibition 酶抑制 (clarithromycin, grapefruit at CYP3A4 → levels rise) ✓.
The derivation (why induction and inhibition move levels for days). Induction builds new enzyme — days to build, days to unwind; inhibition occupies existing enzyme — begins with the first co-dose ✓. So a new inhibitor raises a substrate's level within doses (warfarin effect climbs with it), while an inducer's effect arrives late and outlasts its stop date. Timing is the mechanism made visible.
Worked example 1 — one interaction, three questions. (Fictional) warfarin + a new CYP inhibitor: INR rises (more warfarin effect); warfarin + rifampicin: INR falls (induction clears it); either change is detectable by the INR's rhythm, which is why monitoring interval follows interaction risk ✓.
Worked example 2 — clinical link: the five rights and the report. (Fictional ward) a near-miss caught at the bedside: the five rights — right patient, drug, dose, route, time — are the behavioural backstop for every mechanism above; national reporting systems (the UK MHRA Yellow Card scheme; the US FDA MedWatch) exist because type B reactions are found only by population reporting ✓ — dated jurisdiction-bound statements, as the plan requires.
Common mistakes. Blaming type B reactions on dose (they are idiosyncratic — dose neither predicts nor prevents them). Forgetting the inducer's tail (its effect on enzyme outlasts the course). Treating interactions as pharmacological trivia — most are kinetic, and the INR/aPTT/level is where they surface.
Summary. Type A = dose-related, type B = idiosyncratic; induction → levels fall (slow on, slow off), inhibition → levels rise (fast on); five rights as the behavioural backstop; Yellow Card (UK) / MedWatch (US) for population detection of type B.