Pre-Surgical Risk Stratification Pathway
One pass over the risk before theatre, resolved through the ACC/AHA steps · v1.3- Enter the patient data once. The tool fills the derived fields: BMI, creatinine clearance, the STOP-BANG demographics, the Caprini age bands, the RCRI renal and surgical criteria.
- Fields marked ● are calculated from your entries above.
- A blank optional field marks that assessment as not done, rather than silently scoring it as normal.
- You get eight domains as one verdict: RCRI, functional capacity, ASA, STOP-BANG, Caprini, ARISCAT pulmonary risk, frailty and the surgical risk grade.
- Plus the actions each specialty needs, and what this page advises settling before an elective list.
- Children and adolescents. Every score here is the adult one, and the tool accepts an age of 18 to 110.
- The clearance itself. It stratifies risk; the decision belongs to the anaesthesiologist and the surgical team who have seen the patient.
- Preoperative BNP or NT-proBNP, and postoperative troponin surveillance for myocardial injury after non-cardiac surgery. Both have a defined role in the 2024 guideline and are read there.
- The anaesthetic technique and the intraoperative plan, and postoperative care beyond the specialty actions it lists.
1. Surgical Urgency & Active Cardiac Conditions
2. Demographics & Vitals
3. Planned Procedure & Surgical Risk Grade
4. Functional Capacity (METs Assessment)
5. Revised Cardiac Risk Index (Lee / RCRI)
6. ASA Physical Status Classification
7. Airway Assessment
8. STOP-BANG Questionnaire (OSA Screen)
9. Caprini VTE Risk Assessment (2005 Model)
10. Comorbidities & Critical Medications
Perioperative Risk, Domain by Domain
1. ACC/AHA Stepwise Algorithm (Cardiology)
Five questions in order. The first affirmative answer ends the algorithm.
| Step | Question | If yes |
|---|---|---|
| 1 | Is surgery emergent? | Proceed to OT. No further testing changes management |
| 2 | Active cardiac condition: ACS, decompensated HF, significant arrhythmia, severe valvular disease? | Delay surgery and obtain cardiology consultation |
| 3 | Low-risk surgery, MACE <1%? | Proceed without further cardiac testing |
| 4 | Adequate functional capacity, ≥4 METs, without symptoms? | Proceed without further testing |
| 5 | Poor or unknown functional capacity: use RCRI | RCRI 0-1, proceed. RCRI ≥2, consider pharmacological stress testing ONLY if it will change management |
2. The 4-MET Question (Physician's Bedside Tool)
1 MET = 3.5 mL O2/kg/min. Surgery imposes 4-6 METs of stress on a patient who has no way of declining it, so a patient who cannot reach 4 METs in ordinary life has no reserve left for the theatre. That is why this one bedside question predicts better than most laboratory tests.
- Ask: "Can you climb one flight of stairs without stopping?" or "Can you walk 500 metres on flat ground without breathlessness?" If yes, functional capacity is likely ≥4 METs.
- 2024 refinement. Subjective clinician estimation of METs performs poorly. Where the answer is borderline or unclear, the 2024 ACC/AHA guideline prefers a structured questionnaire: a Duke Activity Status Index (DASI) score above 34 identifies patients at low risk.
- Self-reported inability to climb two flights of stairs is independently associated with postoperative complications.
3. RCRI (Lee Index, 1999), with the 2017 MACE Re-estimates
The critical threshold is 2: RCRI ≥2 with poor functional capacity triggers the decision for non-invasive stress testing. Two sets of percentages exist for the same score, and the tool prints the newer one.
| RCRI | MACE, Duceppe 2017 (printed by this tool) | MACE, Lee's own 1999 derivation cohort |
|---|---|---|
| 0 | 3.9% | 0.4% |
| 1 | 6.0% | 0.9% |
| 2 | 10.1% | 6.6% |
| 3 or more | 15%+ | 11% |
The four percentages this tool prints are the re-estimates published by Duceppe and colleagues in the Canadian Cardiovascular Society guideline of 2017, reference 18 below.
Indian Context: South Asian patients may carry higher baseline cardiovascular risk due to the "thin-fat" Indian phenotype (visceral adiposity with normal BMI), higher insulin resistance, and earlier-onset CAD. A lower threshold for cardiac evaluation in Indian patients with metabolic syndrome, even where the RCRI reads reassuringly, is the practice this page follows. It was attributed here to Yaddanapudi et al.; that citation could not be traced on 28 August 2026 (see reference 17), so the recommendation is offered as practice rather than as a sourced statement until it is.
4. STOP-BANG and Non-Obese OSA
- Bands: 0-2 = Low; 3-4 = Intermediate; 5-8 = High risk for OSA. Score ≥3 mandates postoperative continuous pulse oximetry and caution with opioid analgesia.
- Non-obese patients. STOP-BANG was validated in predominantly obese Western populations. In non-obese or underweight elderly patients, common in India, a score of 3-4 driven solely by age + gender + hypertension has lower specificity for obstructive sleep apnoea. Consider central sleep apnoea or upper-airway anatomical factors rather than classic obesity-driven OSA. Clinical correlation is essential.
5. Caprini VTE and Indian Context
Surgery activates all three arms of Virchow's triad: endothelial injury, stasis, and hypercoagulability. VTE is the most preventable cause of in-hospital death.
- Which Caprini. This tool uses the 2005 model as adopted by the ACCP (Gould et al., Chest 2012): malignancy scores 2 points, and the highest-potency events (stroke, elective arthroplasty, hip or leg fracture, multiple trauma, spinal cord injury, each within 1 month for the acute events) score 5 points each. A patient listed for elective hip or knee replacement therefore starts at 5 points and is high risk before any other factor is counted.
- Indian data. The ASI guidelines recommend pharmacological thromboprophylaxis for Caprini ≥3. Venous thromboembolism was long held to be less common in Indians. That belief rests largely on unscreened series, and under-prophylaxis remains a real problem on Indian surgical wards. The comparable-incidence claim was attributed here to Agarwala et al.; that citation could not be traced on 28 August 2026 (see reference 18), so it is stated as the direction the evidence points rather than as a sourced figure until it is.
6. ARISCAT: Postoperative Pulmonary Risk
Postoperative pulmonary complications are as frequent as cardiac ones and carry comparable mortality, yet most preoperative pathways score only the heart. ARISCAT (Canet et al., Anesthesiology 2010) predicts the composite of respiratory failure, infection, effusion, atelectasis, pneumothorax and bronchospasm from seven routine items, every one of them already on the chart.
| Item | Points |
|---|---|
| Age 51-80 | +3 |
| Age >80 | +16 |
| Preoperative SpO2 91-95% | +8 |
| Preoperative SpO2 ≤90% | +24 |
| Respiratory infection in the last month | +17 |
| Preoperative Hb ≤10 g/dL | +11 |
| Upper abdominal incision | +15 |
| Intrathoracic incision | +24 |
| Duration 2-3 h | +16 |
| Duration >3 h | +23 |
| Emergency surgery | +8 |
- Bands: <26 low risk (1.6% complications), 26-44 intermediate (13.3%), ≥45 high (42.1%).
- Intermediate and high scores justify preoperative chest physiotherapy training, incentive spirometry, smoking cessation, treating any active infection, lung-protective intraoperative ventilation, and a lower threshold for postoperative monitored care.
7. Frailty and Postoperative Delirium (Age 65+)
Frailty, not age, is the risk factor. The 2024 ACC/AHA guideline recommends preoperative frailty screening in adults of 65 and older; a validated instrument such as the Clinical Frailty Scale (Rockwood) takes under a minute.
- CFS 5 or more predicts mortality, complications, delirium, institutional discharge and prolonged stay better than chronological age does. It should prompt Comprehensive Geriatric Assessment, prehabilitation, and an explicit goals-of-care conversation before major elective surgery.
- Delirium prevention is mostly non-pharmacological: correct hearing and vision (bring the spectacles and hearing aid to the ward), reorientation cues, early mobilisation, night-time sleep protection, adequate analgesia while avoiding benzodiazepines, anticholinergics and pethidine, and daily screening with a brief instrument such as the 4AT.
- Frail patients also warrant scrutiny of every anticholinergic and sedative on the drug chart.
8. Medication Reconciliation (2024 Guidance)
| Drug | Perioperative instruction |
|---|---|
| Beta-blockers | CONTINUE if already taken. Never start on the day of surgery (POISE) |
| Statins, thyroid replacement, antiepileptics, inhaled bronchodilators | CONTINUE |
| ACEi/ARBs | HOLD on the morning of surgery to limit refractory intraoperative hypotension, especially before neuraxial anaesthesia. If blood pressure control is needed while they are held, a substitution such as amlodipine is a reasonable bridge. Continuation is reasonable in heart failure with reduced ejection fraction after discussion with the anaesthesiologist |
| Metformin | HOLD on the day of surgery only. Hold 48 hours around iodinated contrast or if eGFR is significantly reduced |
| Sulfonylureas and other oral hypoglycaemics | HOLD on the morning of surgery (hypoglycaemia while fasting) |
| SGLT2 inhibitors | HOLD 3 days (ertugliflozin 4 days) before surgery, for euglycaemic DKA risk |
| GLP-1 receptor agonists (2024 multisociety guidance) | Most patients on a stable dose without gastrointestinal symptoms may CONTINUE with a 24-hour liquid diet before the procedure. Three groups are not: the patient still escalating the dose, the patient on the maximum dose, and the patient with nausea, vomiting, bloating or reflux, all at elevated risk of retained gastric contents. In those, hold daily agents on the day of the procedure and weekly agents for one week. The alternative is to manage the patient as a full stomach: rapid sequence induction, and gastric ultrasound where available |
| Aspirin (POISE-2) | In patients without a coronary stent, routine perioperative aspirin does not reduce cardiac events and increases major bleeding: discontinue primary-prevention aspirin 4-5 days before surgery. CONTINUE perioperatively in patients with stents wherever the surgical bleeding risk allows |
| Prednisolone ≥5 mg daily for ≥3 weeks | Possible adrenal suppression: continue the usual dose and give hydrocortisone 50-100 mg IV at induction, scaled to the magnitude of surgery |
- PCI to surgery timing (2024 ACC/AHA): balloon angioplasty, delay elective surgery ≥14 days. BMS, ≥30 days. DES, ≥6 months for elective surgery (12 months after ACS or complex PCI); time-sensitive surgery may proceed from 3 months after discussion with interventional cardiology.
- Interrupting a P2Y12 inhibitor before surgery: clopidogrel 5 days, ticagrelor 3 days, prasugrel 7 days, continuing aspirin. Premature DAPT discontinuation causes stent thrombosis with >30% mortality.
Neuraxial anaesthesia and anticoagulation. The risk is epidural haematoma, and the DOACs and the P2Y12 inhibitors need longer holds before a neuraxial block than before surgery itself.
| Drug | Interval before a neuraxial block |
|---|---|
| Warfarin | INR <1.5 |
| LMWH | Last dose ≥12 hrs (prophylactic) or ≥24 hrs (therapeutic) |
| Apixaban, rivaroxaban | 72 hrs |
| Dabigatran | 3-5 days graded by CrCl (≥80: 3 days; 50-79: 4 days; 30-49: 5 days; below 30: avoid neuraxial) |
| Unfractionated heparin | Prophylactic 5000 units SC, 4-6 hrs with a normal aPTT; therapeutic IV, 4-6 hrs with a normal aPTT |
| Clopidogrel, ticagrelor (ASRA) | 5-7 days |
| Prasugrel (ASRA) | 7-10 days |
| Aspirin | Alone, does not preclude a neuraxial block |
9. Prehabilitation Checklist
| Target | What it buys |
|---|---|
| Smoking cessation ≥4 weeks preop | Reduces pulmonary complications by 30-40% |
| Glycaemic control | HbA1c >8% = 2-3x higher SSI. Target perioperative glucose 140-180 mg/dL (NICE-SUGAR) |
| Anaemia correction | Hb <10 g/dL independently increases 30-day mortality. IV iron infusion 2-4 weeks preop can restore Hb. Particularly relevant in India where nutritional anaemia is highly prevalent |
10. What This Tool Does Not Print
- A preoperative biomarker result. BNP or NT-proBNP has a defined role in the 2024 guideline. In an elevated-risk patient with poor or unknown functional capacity, a BNP < 92 ng/L or NT-proBNP < 300 ng/L supports proceeding without further cardiac testing. An elevated value should prompt a team discussion and consideration of postoperative troponin surveillance. The tool does not score it.
- Postoperative troponin surveillance for myocardial injury after noncardiac surgery (MINS), which may be considered in elevated-risk patients even without symptoms.
- A sourced Indian figure for either the metabolic syndrome threshold or the venous thromboembolism incidence claim. Both citations were untraceable on 28 August 2026 and both statements are offered as practice rather than as sourced figures.
- A revascularisation recommendation. The CARP principle at the head of this page governs: test only where the result would change perioperative management.
References
- Thompson A, Fleischmann KE, Smilowitz NR, et al. 2024 AHA/ACC/ACS/ASNC/HRS/SCA/SCCT/SCMR/SVM Guideline for Perioperative Cardiovascular Management for Noncardiac Surgery. Circulation. 2024;150(19):e351-e442. [Current edition; supersedes Fleisher 2014]
- Fleisher LA, et al. 2014 ACC/AHA Guideline on Perioperative Cardiovascular Evaluation. Circulation. 2014;130(24):e199-e267. [Superseded September 2024; source of the stepwise algorithm reproduced above]
- Lee TH, et al. Derivation and Validation of a Simple Index for Cardiac Risk. Circulation. 1999;100(10):1043-1049.
- Chung F, et al. STOP-Bang Questionnaire. Chest. 2016;149(3):631-638.
- Caprini JA. Thrombosis Risk Assessment. Disease-a-Month. 2005;51(2-3):70-78.
- Gould MK, et al. Prevention of VTE in Nonorthopedic Surgical Patients: ACCP Evidence-Based Clinical Practice Guidelines (9th ed). Chest. 2012;141(2 Suppl):e227S-e277S.
- Canet J, Gallart L, Gomar C, et al. Prediction of Postoperative Pulmonary Complications in a Population-Based Surgical Cohort (ARISCAT). Anesthesiology. 2010;113(6):1338-1350.
- Devereaux PJ, et al. Aspirin in Patients Undergoing Noncardiac Surgery (POISE-2). N Engl J Med. 2014;370(16):1494-1503.
- Devereaux PJ, et al. Effects of Extended-Release Metoprolol Succinate in Patients Undergoing Non-Cardiac Surgery (POISE). Lancet. 2008;371(9627):1839-1847.
- Kindel TL, Wang AY, Wadhwa A, et al. Multisociety Clinical Practice Guidance for the Safe Use of Glucagon-like Peptide-1 Receptor Agonists in the Perioperative Period. Clin Gastroenterol Hepatol. 2025;23(12):2083-2085.
- Rockwood K, et al. A Global Clinical Measure of Fitness and Frailty in Elderly People (Clinical Frailty Scale). CMAJ. 2005;173(5):489-495.
- McFalls EO, et al. CARP Trial. N Engl J Med. 2004;351(27):2795-2804.
- Halvorsen S, Mehilli J, Cassese S, et al. 2022 ESC Guidelines on cardiovascular assessment and management of patients undergoing non-cardiac surgery. Eur Heart J. 2022;43(39):3826-3924. [Supersedes Kristensen 2014]
- Kristensen SD, et al. 2014 ESC/ESA Guidelines on Non-Cardiac Surgery. Eur Heart J. 2014;35(35):2383-2431. [Superseded 2022]
- NICE-SUGAR Study Investigators. N Engl J Med. 2009;360(13):1283-1297.
- Yaddanapudi LN. Perioperative Cardiovascular Evaluation for Non-Cardiac Surgery. Indian J Anaesth. 2015;59(2):106-113. [Could not be verified, 28 August 2026.]
- Agarwala S, et al. VTE Prophylaxis in Surgical Patients: Practice Survey from India. Indian J Surg. 2018;80(6):575-581. [Could not be verified, 28 August 2026.]
- Duceppe E, et al. Canadian CCS Guidelines on Perioperative Cardiac Risk. Can J Cardiol. 2017;33(1):17-32.
How to Cite This Tool
DOIhttps://doi.org/10.5281/zenodo.22401626
AMA Style:Umakanth S. Pre-Surgical Risk Stratification Pathway. Version 1.3. MEDiscuss Clinical Decision Support System. Published 2026. Accessed . https://mediscuss.org/cdss/pre-surgical-clearance. doi:10.5281/zenodo.22401626
Vancouver Style:Umakanth S. Pre-Surgical Risk Stratification Pathway [Internet]. Version 1.3. MEDiscuss.org; 2026 [cited ]. Available from: https://mediscuss.org/cdss/pre-surgical-clearance. doi:10.5281/zenodo.22401626
Save this calculation
Whatever you type here stays on this device and is not sent anywhere. The server receives only a scrambled code made from it, so nobody with access to the server can tell which patient a saved calculation belongs to. Enter the same nickname the next time to see that patient's earlier values. What is stored
