Perioperative Cardiovascular Assessment

Definition

A structured, stepwise framework for evaluating and managing cardiovascular risk in adult patients (≥18 years) undergoing noncardiac surgery (NCS). The goal is to identify patients who warrant further evaluation or optimisation before surgery, avoid unnecessary testing in low-risk patients, and ensure cardiovascular comorbidities are appropriately managed throughout the perioperative period.

Source status. The 2026 AHA/ACC multisociety guideline is a reaffirmation, not a revision: every recommendation, table and figure is identical to the 2024 document, which remains the substantive source for this page. (sources/periop-aha-2026, rating: very high)

Key Concepts

Surgical Risk Classification

• Emergency (<2h): No preoperative evaluation feasible; manage cardiac conditions perioperatively
• Urgent (2–24h): Limited time for targeted risk reduction
• Time-sensitive (up to 3 months delay): Allows evaluation and optimisation
• Elective: Full preoperative evaluation can occur
• Low risk: Predicted MACE <1%; Elevated risk: MACE ≥1% (traditionally RCRI >1)
• Highest-risk procedures: suprainguinal vascular, thoracic, transplant, neurosurgery
• Intermediate-risk: general, genitourinary, orthopaedic
• Lowest-risk: endocrine, breast, gynaecology/obstetrics (sources/periop-aha-2024, rating: very high)

Stepwise Assessment Algorithm (Figure 1, AHA 2024)

1. Emergency surgery? → Proceed; manage cardiac conditions perioperatively
2. Active cardiac conditions? (ACS, decompensated HF, severe symptomatic AS, unstable arrhythmia) → Evaluate and treat first
3. Low-risk surgery? → Proceed without further cardiac testing
4. Assess functional capacity using DASI and calculate perioperative risk with RCRI or NSQIP
5. If elevated risk + poor/unknown functional capacity → consider biomarkers, ECG, selective non-invasive imaging
6. Shared decision-making; proceed with surgery ± GDMT optimisation (sources/periop-aha-2024, rating: very high)

Cardiovascular Risk Indices

• No single risk index is recommended over the others; combining the RCRI with coronary calcium from an existing CT may improve stratification. (sources/periop-aha-2024, rating: very high)

Functional Capacity Assessment

• Measured in METs; threshold of <4 METs = poor functional capacity
• DASI (Duke Activity Status Index): 12-item validated self-report tool (score range 0–58.2); DASI ≤34 predicts increased 30-day death or MI — COR 2a (B-NR)
• Subjective clinician assessment of METs is NOT associated with outcomes (METs trial, n=1401)
• DASI added to RCRI significantly improves predictive performance (BASEL-PMI: functional capacity <2 flights of stairs → 1.63× 30-day MACE)
• Poor functional capacity alone does not mandate preoperative stress testing (sources/periop-aha-2024, rating: very high)

Frailty

• COR 2a: Validated frailty assessment for all patients ≥65 years (or <64 with perceived frailty) undergoing elevated-risk NCS
• Frailty → 3.71× increased 30-day mortality and 2.39× increased 30-day complications (meta-analysis, 1.1 million patients)
• Validated tools: Fried Phenotype, Clinical Frailty Scale (best predictor in NCS), FRAIL Scale, Edmonton Frail Scale, SPPB
• Prehabilitation (exercise + nutrition) may improve outcomes in selected frail patients (sources/periop-aha-2024, rating: very high)

Preoperative Biomarkers

• BNP/NT-proBNP (COR 2a, B-NR): In patients with CVD, age ≥65, or age ≥45 with CVD symptoms undergoing elevated-risk NCS
  • Abnormal thresholds: BNP >92 ng/L; NT-proBNP ≥300 ng/L
  • NT-proBNP >100 pg/mL independently predicts all-cause mortality (VISION substudy)
  • Addition of NT-proBNP to traditional risk scores did not significantly improve prediction beyond scores + functional status in one cohort (n=3597)
• Cardiac troponin (COR 2b, B-NR): Same population as BNP; provides baseline for postoperative interpretation; no management change warranted on elevated preoperative cTn alone
• No studies demonstrate that acting on elevated preoperative biomarkers improves outcomes (sources/periop-aha-2024, rating: very high)

Preoperative 12-Lead ECG

• COR 2a: Patients with known coronary disease, arrhythmia, PAD, cerebrovascular disease, structural heart disease, or active CVD symptoms undergoing elevated-risk surgery
• COR 2a: New ECG abnormalities should prompt further evaluation — ST elevation, ST depression, T-wave inversion, LVH, significant pathological Q waves, Mobitz II or higher AV block, bundle branch block, QT prolongation, or AF. Incidental Q waves or bundle branch block must not trigger revascularisation.
• COR 2b: Asymptomatic patients undergoing elevated-risk surgery without known CVD
• COR 3 No Benefit: Routine ECG in asymptomatic patients undergoing low-risk procedures
• Actionable findings: Mobitz II+ AV block, new-onset AF, QT prolongation (inform anaesthetic/antibiotic choice), new bundle branch block (sources/periop-aha-2024, rating: very high)

Stress Testing

• COR 2b (B-NR): Only if poor/unknown functional capacity AND elevated validated risk AND suspicion of high-risk ischemia (left main/multivessel disease) undergoing elevated-risk surgery
• COR 3 No Benefit (B-R): Low-risk patients, adequate functional capacity, or low-risk surgery
• An abnormal stress test should NOT prompt angiography unless it shows high-risk features (LM disease or severe multivessel + reduced EF)
• CARP trial (n=510): Preoperative coronary revascularisation did not reduce 30-day MI or mortality vs medical therapy before vascular surgery (sources/periop-aha-2024, rating: very high)

Coronary CT Angiography (CCTA)

• COR 2b: Elevated risk + poor/unknown functional capacity only if high-risk coronary anatomy suspected; provides high negative predictive value
• COR 3 No Benefit: Routine CCTA; contraindicated in urgent/emergency surgery
• CCTA overestimates risk in patients who will NOT have MACE (5× more false-positive than false-negative)
• Coronary calcium score 0 within 2 years → proceed without additional cardiac testing (sources/periop-aha-2024, rating: very high)

Invasive Coronary Angiography

• COR 3 No Benefit (C-LD): Routine preoperative ICA not recommended regardless of surgical risk
• ICA reserved for patients with ACS, accelerating angina despite GDMT, newly diagnosed moderate-severe ischemia on stress testing, or indicators of obstructive left main disease (sources/periop-aha-2024, rating: very high)

Perioperative Blood Pressure Management

• Intraoperative (COR 1, B-NR): Maintain MAP ≥60–65 mmHg or SBP ≥90 mmHg
  • Harm threshold: MAP <65 or SBP <90 sustained ~15 minutes
  • POISE-3 (n=7490): No benefit from targeting MAP ≥80 vs MAP ≥60 mmHg
• Postoperative (COR 1, B-NR): Treat hypotension (MAP <60–65 or SBP <90) promptly
• Preoperative: May defer elective elevated-risk surgery if SBP ≥180 mmHg or DBP ≥110 mmHg plus ≥1 RCRI risk factor (COR 2b)
• Continue antihypertensives throughout perioperative period in most patients (COR 2a)
• Restart antihypertensive medications as soon as clinically reasonable postoperatively (COR 1) (sources/periop-aha-2024, rating: very high)

MINS and Postoperative Myocardial Infarction

• Definition: acute myocardial injury (rise and/or fall with at least one value above the sex-specific 99th percentile URL), regardless of symptoms, in the post-operative period; asymptomatic in 80–90%. MINS is the subset deemed ischaemic without unexpected physiological stress or a non-coronary cardiac cause, and carries an eight-fold higher adjusted 30-day mortality. Under the Fifth Universal Definition most post-operative events are secondary MI — supply–demand imbalance from another acute condition — and confirmation now requires obstructive CAD without acute coronary pathology, or a new regional wall motion abnormality, rather than a biomarker multiple. (sources/5thMI-AHA-2026, rating: very high)
• Quantified by the 2026 definition: with active troponin surveillance, PMI occurs in ~1 in 8 high-risk non-cardiac surgery patients and primary MI in 1 in 20 of those with post-operative injury; up to 40% of patients with CVD or risk factors have a baseline elevated troponin, making a pre-operative value essential; most events occur within 2 days and up to two-thirds are asymptomatic. (sources/5thMI-AHA-2026)
• Incidence: ~20% of patients undergoing NCS
• Prognosis: 30-day mortality ~10%; proportional to peak cTn concentration (17% in highest quartile vs 1% in lowest); 34% population attributable risk of 30-day postoperative mortality
• Elevated postop cTn without ischaemic feature → 3× 30-day mortality hazard; with ischaemic feature → 5× hazard (VISION, n=21,842)
• Surveillance (COR 2b, B-NR): cTn at 24h and 48h post-surgery in patients with known CVD, CVD symptoms, or age ≥65 with risk factors undergoing elevated-risk NCS; NOT in low-risk NCS (COR 3)
• Management:
  • COR 2a: Cardiology consultation and GDMT optimisation (early referral → reduced 30-day mortality)
  • COR 2b: Consider antithrombotic therapy — MANAGE trial: dabigatran 110 mg BID reduced major vascular events but increased minor/GI bleeding; high dropout rate
  • Algorithm: elevated postop troponin → evaluate for ischaemic symptoms/ECG changes → classify as MINS vs STEMI/NSTEMI vs nonischaemic injury
  • Nonischaemic causes to exclude: pulmonary embolism, sepsis, decompensated HF, stroke (sources/periop-aha-2024, rating: very high)

When injury is classified as infarction, perioperative MI occurs in 0.9–15% of patients and is predominantly type 2, from supply–demand mismatch. Perioperative STEMI carries 30–35% in-hospital mortality and emergent invasive angiography is strongly considered; NSTEMI is managed with guideline-directed medical therapy as for spontaneous MI, with angiography in selected patients according to the balance of bleeding and thrombotic risk. Both decisions require multidisciplinary input from surgeon, anaesthesiologist and cardiologist. (sources/periop-aha-2024, rating: very high)

Perioperative Pharmacotherapy Summary

Anticoagulation Management

Most patients with AF on a DOAC are managed by time-based interruption without bridging. (sources/periop-aha-2024, rating: very high)

Antiplatelet Therapy and Timing of Surgery After PCI

• Continue aspirin 75–100 mg wherever possible (COR 1 B-R).
• Continue DAPT (COR 1 B-NR) when time-sensitive NCS falls within 30 days of a bare-metal stent or within 3 months of a drug-eluting stent, unless bleeding outweighs stent-thrombosis risk.
• Intravenous antiplatelet bridging (COR 2b B-NR) may be considered in high thrombotic risk — within 6 months of a DES or 30 days of a BMS — when oral agents must be stopped.
• Offset intervals: aspirin 4 days; clopidogrel 5–7; prasugrel 7–10; ticagrelor 3–5. (sources/periop-aha-2024, rating: very high)
• Local handbook, for comparison: stop ticagrelor 3–5 days, clopidogrel 5 days and prasugrel ≥7 days pre-operatively; do not stop DAPT within the first month after PCI; defer elective surgery by 3–6 months, preferably 12, after recent MI or where ischaemic risk is high, and by at least 1 month, preferably 6, otherwise. (sources/Handbook-HA-2024, rating: high)

Valvular Disease, Congenital Heart Disease and Transplantation

(Sources: sources/periop-aha-2024, rating: very high)

• Local handbook thresholds for intervening before elective intermediate- or high-risk surgery: severe symptomatic AS — postpone and consider AVR or valvuloplasty, and for asymptomatic severe AS proceed if LVEF ≥50% and consider AVR below that; severe AR — operate first if symptomatic, or if LVESD >50 mm or resting LVEF <50%; moderate-to-severe MS — PTMC or surgery first if symptomatic or SPAP >50 mmHg; severe MR — intervene first if symptomatic, or if LVESD ≥40 mm or LVEF <60%. (sources/Handbook-HA-2024, rating: high)

Intraoperative Management

• Anaesthetic technique (volatile vs TIVA, neuraxial vs general): no meaningful difference in cardiovascular outcomes
• Thoracic epidural analgesia: reduces perioperative MI in major abdominal surgery (COR 2a)
• Tranexamic acid: reduces blood loss and transfusions in NCS with expected blood loss (COR 2a); caution in cardiovascular high-risk patients
• Iron therapy for iron-deficiency anaemia before elective NCS: COR 2a B-R — either oral or intravenous. IV iron is more effective in supportive text, but the recommendation covers both routes; restrictive transfusion threshold 8 g/dL in orthopaedic and cardiovascular patients.
• TEE/FoCUS: COR 2a for unexplained haemodynamic instability where expertise exists; routine intraoperative TEE is COR 3 No Benefit
• Temporary mechanical circulatory support: COR 2b C-LD — may be used pre-emptively or as rescue in acute severe haemodynamic instability undergoing urgent or emergency NCS
• PA catheter: COR 3 No Benefit, LOE A for routine use; COR 2b C-LD for selective use when haemodynamically significant conditions (decompensated HF, severe valvular disease, combined shock, PH) cannot be corrected (sources/periop-aha-2024, rating: very high)
• In obstructive HCM, avoid reducing blood volume, systemic vascular resistance or venous capacitance, and avoid catecholamines. The four Class 3 Harm factors and their substitutes are tabulated on concepts/HCM-Management. (sources/Handbook-HA-2024, rating: high)

Special Populations: Patients with CIEDs

• Periprocedural CIED management requires dedicated multidisciplinary planning separate from the general perioperative cardiovascular assessment framework. (sources/periop-cied-aha-2024, rating: high)
• Preprocedural screening should identify CIED type, manufacturer, pacing dependency, battery status, and anticipated EMI sources before any invasive procedure. (sources/periop-cied-aha-2024)
• Magnet application to an ICD does not affect pacing function — pacing-dependent ICD patients require device reprogramming, not just magnet placement. Reprogramming to AOO, VOO or DOO is the route when the diathermy site is above the umbilicus and the patient is pacing-dependent. (sources/periop-cied-aha-2024)
• Leadless pacemakers do not respond to a magnet, so the Micra cannot be switched to asynchronous pacing that way; pacing-dependent patients require reprogramming to VOO before procedures with EMI above the umbilicus. (sources/periop-cied-aha-2024)
• A deactivated ICD requires continuous cardiac monitoring with external defibrillation immediately available; a ring magnet suspends inappropriate therapy during the procedure. (sources/Handbook-HA-2024)
• If external defibrillation is needed, place the paddles more than 12 cm from the generator. (sources/Handbook-HA-2024)
• Re-enable ICD therapies before discharge from the monitored setting — patient deaths have been reported following failure to do so. (sources/periop-cied-aha-2024)
• See concepts/Periprocedural-CIED-Management for the full device-specific framework.

Special Populations: Pulmonary Hypertension

• Continue all PAH-targeted therapy without interruption (COR 1): abrupt discontinuation of an ERA, PDE5 inhibitor or prostacyclin causes acute haemodynamic decompensation and death — maintain therapy including on the day of surgery. (sources/periop-aha-2024, rating: very high)
• Perioperative risk: PH confers 43% increased odds of postoperative death, MI or stroke; Group 1 PAH specifically carries 2.5× MACE risk and 5× cardiogenic shock risk versus patients without PH. (sources/periop-aha-2024)
• Specialist centre and multidisciplinary team (COR 2a): severe PH undergoing elevated-risk noncardiac surgery should be managed where PAH expertise exists, with a PH cardiologist, anaesthesiologist and intensivist. (sources/periop-aha-2024)
• Invasive haemodynamic monitoring (COR 2a): recommended for elevated-risk surgery; avoid fluid overload and agents that increase RV afterload. (sources/periop-aha-2024)
• Inhaled pulmonary vasodilators (COR 2a): inhaled nitric oxide or epoprostenol should be available for perioperative pulmonary hypertensive crises; norepinephrine or vasopressin is preferred to maintain RV perfusion pressure. (sources/periop-aha-2024)
• See concepts/Pulmonary-Hypertension for the haemodynamic definitions and group classification behind these recommendations.

Contradictions / Open Questions

• The local handbook and AHA 2024 describe different algorithms and different risk bands, and nothing on the page says which to follow. The primary source is AHA 2024; the handbook carries a seven-step version derived from the 2022 ESC guideline, with a different ordering of biomarkers and functional testing, and a separate surgical-risk banding (<1%, 1–5%, >5%) against the AHA's <1% / ≥1% split. The handbook's guidance is carried inside the relevant sections above rather than as a separate algorithm, which shows where the two disagree but does not resolve it. (sources/Handbook-HA-2024, rating: high)
• Beta-blocker initiation before surgery is recommended by the handbook and prohibited by the guideline. AHA 2024 makes starting a beta-blocker on the day of surgery Class 3: Harm on POISE, while the handbook advises considering initiation before high-risk surgery where there are ≥2 clinical risk factors or known CAD. Both concern the same decision: POISE started the drug on the day of surgery, the handbook implies preloading, but neither document states how many days of pre-treatment would separate a safe strategy from a harmful one. (sources/periop-aha-2024, rating: very high; sources/Handbook-HA-2024, rating: high)
• Preoperative biomarker-guided management has no outcome data. No RCT has shown that acting on an elevated preoperative BNP or troponin improves outcomes, so the test is recommended without evidence that the result changes what is done.
• Intraoperative blood-pressure targets remain contested. POISE-3 found no benefit from targeting MAP ≥80 rather than ≥60 mmHg, while INPRESS suggested benefit from tighter control — conflicting evidence for the same decision.
• Perioperative management of sacubitril/valsartan is completely unstudied.
• GLP-1 agonist guidance rests on ASA consensus only, not on RCT data.
• The incidence of postoperative myocardial injury is quoted two ways that point in opposite directions. The older figure is ~20% of all patients undergoing noncardiac surgery; the Fifth Universal Definition quotes ~1 in 8 (12.5%) of high-risk patients under active troponin surveillance. A high-risk subgroup cannot plausibly have a lower incidence than the whole population, so the discrepancy reflects different surveillance protocols and definitions rather than a difference between populations, and neither figure is directly usable for consent or planning. (sources/periop-aha-2024, rating: very high; sources/5thMI-AHA-2026, rating: very high)
• Optimal MINS management remains undefined — the MANAGE trial had significant limitations, with high dropout and a post-hoc change to the primary outcome.
• The optimal surveillance strategy for postoperative AF after noncardiac surgery is not established.

Connections

• Related to concepts/Periprocedural-CIED-Management — specialized framework for pacemakers, ICDs, CRT, and ILR perioperative management
• Related to entities/Heart-Failure — perioperative HF management; SGLT2i discontinuation
• Related to entities/Atrial-Fibrillation — perioperative AF; new-onset POAF stroke risk
• Related to concepts/Pulmonary-Hypertension — continue PAH therapy; specialised centre referral
• Related to entities/HCM — avoid LVOT aggravating factors
• Related to concepts/Aortic-Stenosis — perioperative AS algorithm; TAVI before NCS
• Related to concepts/Mitral-Stenosis — perioperative MS management
• Related to concepts/TAVI — NCS timing after TAVI
• Related to concepts/DAPT-Strategies — PCI timing and perioperative antiplatelet management
• Related to entities/Chronic-Coronary-Disease — preoperative revascularisation thresholds

Sources

• sources/periop-aha-2024
• sources/periop-aha-2026
• sources/periop-cied-aha-2024
• sources/5thMI-AHA-2026
• sources/Handbook-HA-2024