Preeclampsia (PE) is a pregnancy complication affecting 2–8% of pregnancies worldwide, defined by new-onset hypertension (SBP ≥140 or DBP ≥90 mmHg) after 20 weeks gestation combined with proteinuria (≥300 mg/24h) and/or end-organ damage. It is a leading cause of maternal and perinatal morbidity and mortality globally.
This QSP model integrates:
- Placental dysfunction and inadequate spiral artery remodeling
- Angiogenic imbalance: sFlt-1↑, PlGF↓, sEng↑
- Endothelial dysfunction: NO↓, ET-1↑, ROS↑
- Multiorgan involvement: cardiovascular, renal, coagulation, neurological, hepatic
- Drug PK/PD: Aspirin (prophylaxis), Labetalol, Nifedipine, MgSO4
Figure: Mechanistic map of preeclampsia pathophysiology and pharmacology. Click image to open interactive SVG. Generated with Graphviz (
neatolayout).15 subgraph clusters covering: Maternal Risk Factors · Placental Dysfunction · Angiogenic Imbalance · Endothelial Dysfunction · Cardiovascular/BP · Renal · Coagulation/HELLP · Neurological · Hepatic · Complement · Fetal · Aspirin PK/PD · Labetalol PK/PD · Nifedipine PK/PD · Magnesium Sulfate PK/PD · Clinical Endpoints
| # | Compartment | Description |
|---|---|---|
| 1 | DEPOT_ASP |
Aspirin gut absorption depot |
| 2 | ASPIRIN |
Aspirin central plasma (mg) |
| 3 | SALICYLATE |
Salicylate (hydrolysis product) |
| 4 | COX1_INH |
COX-1 inhibition state (0–1, irreversible) |
| 5 | DEPOT_LAB |
Labetalol gut depot |
| 6 | LABETALOL |
Labetalol central plasma |
| 7 | DEPOT_NIF |
Nifedipine gut depot |
| 8 | NIFEDIPINE |
Nifedipine central plasma |
| 9 | MG_PLASMA |
Magnesium plasma pool (mmol) |
| 10 | SFLT1 |
sFlt-1 plasma (pg/mL) |
| 11 | PLGF |
PlGF plasma (pg/mL) |
| 12 | SENG |
Soluble endoglin (ng/mL) |
| 13 | NO_EA |
Nitric oxide bioavailability (relative) |
| 14 | ET1 |
Endothelin-1 (pg/mL) |
| 15 | ROS |
Reactive oxygen species index |
| 16 | SBP |
Systolic blood pressure (mmHg) |
| 17 | DBP |
Diastolic blood pressure (mmHg) |
| 18 | GFR_C |
Glomerular filtration rate (mL/min/1.73m²) |
| 19 | PROTEINURIA |
Proteinuria (mg/24h) |
| 20 | PLATELET |
Platelet count (×10³/µL) |
| +2 | LDH_MK, SEIZURE_RISK |
HELLP/neurological markers |
| Scenario | Treatment | Rationale |
|---|---|---|
| 1 | No treatment | Natural PE progression |
| 2 | Aspirin 75 mg/day (12 wk→) | ASPRE trial: 62% reduction in early PE |
| 3 | Labetalol 200 mg BID (24 wk→) | CHIPS trial: α1+β blocker for BP control |
| 4 | Nifedipine 30 mg MR/day (24 wk→) | L-type Ca²⁺ blocker, CHIPS trial |
| 5 | MgSO4 4 g load + 1 g/h (30 wk→) | Magpie trial: 58% ↓ eclampsia |
| 6 | Combination (Aspirin + Labetalol + MgSO4) | Optimal multidrug approach |
| Biomarker | Normal | PE Threshold | Severe PE |
|---|---|---|---|
| SBP (mmHg) | <140 | ≥140 | ≥160 |
| DBP (mmHg) | <90 | ≥90 | ≥110 |
| sFlt-1/PlGF ratio | <38 | 38–85 | >85 |
| Proteinuria (mg/24h) | <300 | ≥300 | ≥5000 |
| Platelets (×10³/µL) | >150 | 100–150 | <100 (HELLP) |
| LDH (IU/L) | <600 | 600–800 | >800 (HELLP) |
| Mg²⁺ plasma (mmol/L) | 0.7–1.0 | 1.7–3.5 (therapeutic) | >3.5 (toxic) |
| File | Description |
|---|---|
pe_qsp_model.dot |
Graphviz source (15 clusters, 150+ nodes) |
pe_qsp_model.svg |
Scalable vector mechanistic map |
pe_qsp_model.png |
PNG thumbnail (150 dpi) |
pe_mrgsolve_model.R |
mrgsolve ODE model + 6 scenario simulations |
pe_shiny_app.R |
8-tab Shiny interactive dashboard |
pe_references.md |
60 curated references (PubMed links) |
README.md |
This file |
# Render mechanistic map
neato -Tsvg pe_qsp_model.dot -o pe_qsp_model.svg
neato -Tpng -Gdpi=150 pe_qsp_model.dot -o pe_qsp_model.png# Run mrgsolve simulation
install.packages(c("mrgsolve","dplyr","ggplot2","tidyr","patchwork"))
source("pe_mrgsolve_model.R")
# Launch Shiny dashboard
install.packages(c("shiny","shinydashboard","DT","gridExtra"))
shiny::runApp("pe_shiny_app.R")- Patient Profile — Parameters, clinical status at GA 36wk, threshold table
- Drug PK — Aspirin, Labetalol, Nifedipine, MgSO4 plasma concentrations over gestation
- Angiogenic Balance — sFlt-1, PlGF, sEng, sFlt-1/PlGF ratio trajectories
- Cardio-Renal — BP (SBP/DBP), endothelial markers, GFR, proteinuria
- HELLP & Neuro — Platelet count, LDH, seizure risk, Mg safety window
- Scenario Comparison — All 6 treatment arms side-by-side
- Biomarker Panel — Heatmap + data table across gestation
- About — Model description, calibration sources, thresholds
| Trial | Finding | Model Parameter |
|---|---|---|
| ASPRE (Rolnik 2017, Lancet) | Aspirin 62% ↓ early PE | F_Asp, kBP_Asp |
| Maynard 2003 (J Clin Invest) | sFlt-1 overexpression → PE phenotype | kprod_sFlt1, k_sFlt1_rise |
| Verlohren 2010 (AJOG) | sFlt-1/PlGF ratio cutoff 38 | Threshold encoding |
| CHIPS 2015 (NEJM) | Labetalol/Nifedipine BP control | Emax_Lab, Emax_Nif |
| Magpie 2002 (Lancet) | MgSO4 58% ↓ eclampsia | Emax_NMDA, seizure ODE |
Placental Hypoxia / Insufficient Trophoblast Invasion
↓
HIF-1α → ↑sFlt-1 secretion ──→ Sequestration of VEGF & PlGF
↓ ↓
↑sEng ──────────────────→ ↓ VEGFR2 signaling
↓
Endothelial Dysfunction (↓NO, ↑ET-1, ↑ROS)
↓
┌─────────────────┴──────────────────┐
↓ ↓
↑SVR → Hypertension Glomerular endotheliosis
↑TXA2 → Platelet aggregation → Podocyte injury
↓ ↓
HELLP Syndrome Proteinuria
Eclampsia (if MgSO4 ↓) ↓GFR → ↑Creatinine
Generated by Claude Code Routine (CCR) — QSP Disease Model Library | 2026-06-25
