Management of renin–angiotensin system inhibitors prior to major surgery: insights from the STOP-or-NOT trial

Summary

Strong recommendations on how to manage renin–angiotensin system inhibitors, including angiotensin-converting enzyme inhibitors and angiotensin receptor blockers, before surgery are lacking because of a lack of evidence, which is mostly limited to data from observational studies. The STOP-or-NOT trial was a large multicentre randomised trial designed to determine whether chronic renin–angiotensin system inhibitors should be continued or discontinued before major noncardiac surgery. As principal investigators of the STOP-or-NOT trial, we discuss the trial's results and how they contribute to the existing literature on management of renin–angiotensin system inhibitors before surgery.

The STOP-or-NOT trial¹ was a large multicentre randomised trial designed to address whether patients taking chronic renin–angiotensin system inhibitors (RASIs) should continue or discontinue these medications before major noncardiac surgery (Fig. 1).

Fig 1.

Visual abstract of the STOP-or-NOT trial. CI, confidence interval; IQR, interquartile range; OR, odds ratio.

Context and need for the STOP-or-NOT trial

More than 20% of patients undergoing surgical procedures have a medical history of hypertension, diabetes mellitus or heart failure^2,3 requiring chronic treatment with a renin–angiotensin system inhibitor (RASI), such as angiotensin-converting enzyme inhibitors (ACE-Is) or angiotensin receptor blockers (ARBs). Observational studies and small randomised trials have suggested that continuing RASIs before major noncardiac surgery increases the risk of intraoperative hypotension,4, 5, 6 which might impact postoperative complications such as acute kidney injury, myocardial injury (MI), or stroke.⁷ These studies are, however, prone to unmeasured confounders, and the lack of large-scale randomised trials has led to conflicting guidelines. The 2014 ESC/ESA guidelines on noncardiac surgery stated that transient discontinuation of RASIs before noncardiac surgery in hypertensive patients should be considered (class IIa recommendation, level of evidence: C).⁸ In contrast, the 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery stated that continuation of RASIs perioperatively was reasonable (class IIa recommendation, level of evidence: B).⁹ They also recommended that if RASI therapy is withheld, it should be restarted as soon as feasible (level of evidence: C). The 2021 European Society of Cardiology guidelines¹⁰ also underscored the lack of data, but recommended that withholding RASIs in patients with chronic hypertension could be considered to prevent intraoperative hypotension, while continuing them in patients with heart failure is acceptable.

Key findings and their implications

The STOP-or-NOT trial¹ enrolled 2222 patients treated with chronic RASIs (defined as being treated for 3 months or more) and scheduled for major noncardiac surgery across 40 centres in France, of whom 1107 were randomised to continue RASIs until the day of surgery and 1115 were randomised to discontinue RASIs 48 h before surgery (i.e. last dose 3 days before surgery). Adherence to the protocol was 96.3%; 88% of participants had a history of hypertension, 16% had coronary artery diseases, and 8% had diabetes mellitus requiring treatment with insulin. The primary endpoint was a composite of all-cause mortality and major postoperative complications within 28 days after surgery (Table 1), which aligns with the Standardised Endpoints in Perioperative Medicine (StEP) initiative.^11,20,21 Postoperative complications included acute myocardial infarction, arterial or venous thrombosis, stroke, acute pulmonary oedema, cardiogenic shock, acute severe hypertensive crisis, de novo cardiac arrhythmia requiring therapeutic intervention, sepsis or septic shock (defined according to the Sepsis-3 definition), respiratory complications (determined by the need for reintubation or noninvasive ventilation for respiratory failure), unplanned ICU admission or readmission, acute kidney injury (based on the serum creatinine item of the Kidney Disease Improving Global Outcome [KDIGO] criteria), or hyperkalaemia (serum K⁺ >5.5 mM requiring therapeutic intervention) or need for surgical reintervention. Participants were not blinded to the intervention, but an adjudication committee, blinded to the treatment arm, reviewed and validated all events of the primary outcome.

Table 1.

The STOP-or-NOT trial outcomes, rationales, and potential relationship with the intervention at the time of the trial design. RAAS, renin–angiotensin–aldosterone system; RASI, renin-angiotensin system inhibitor.

Components of the primary outcome at 28 days	Rationale	Strength of the potential relationship with the intervention
Death	-Standardised and patient-centred cardiovascular outcome11	++
Postoperative cardiovascular events	-Standardized and patient-centred cardiovascular outcome -Strategy of RASI management might impact the incidence of postoperative cardiovascular events -Activation of RAAS promotes the occurrence of cardiovascular events, including atrial fibrillation12, 13, 14	+++
Unplanned ICU admission	- Standardised and patient-centred cardiovascular outcome	+++
Acute kidney injury	-Hypotension can decrease renal perfusion -RAAS is a major contributor to glomerular filtration regulation -Activation of RAAS associated with kidney injury15	+++
Postoperative episodes of sepsis	-Standardised and patient-centred cardiovascular outcome -Inhibition of RAAS can increase the risk of hypotension and need for vasopressor if infection develops16 -RASIs associated with immune regulation and might interfere with the risk of developing sepsis17, 18, 19	+
Surgical complications	-Hypotension or use of vasopressors might lead to poor wound healing -Important patient-centred outcome	+
Postoperative respiratory complications	-Standardised and patient-centred cardiovascular outcome -RAAS activation might contribute to lung injury	++

The trial results showed no difference in the primary outcome between the continuation and discontinuation groups, with both experiencing a 22% rate of death or major postoperative complications. However, the trial also found that patients in the continuation group experienced a greater incidence of intraoperative hypotension, defined as episodes of hypotension requiring vasopressor administration during anaesthesia and surgery (54% compared with 41% in the discontinuation group). The duration of MAP <60 mm Hg was only slightly longer in the continuation group, with a mean absolute difference of 3.7 min (95% confidence interval [CI]: 1.4–6.0). No other difference in secondary outcomes was observed (postoperative organ failure, hospital and intensive care unit length of stay).

Limitations of the STOP-or-NOT trial

The trial was conducted in a single country, and the results might not be generalisable to other healthcare settings with different practices or patient populations.

Moreover, the trial was underpowered to detect differences in some rarer outcomes or in specific surgical procedures. Only 6% of patients had a history of heart failure, which aligns with the incidence reported in large observational studies of patients undergoing major surgery,^2,3 but the small sample size limits interpretation of the results in this specific subpopulation. Finally, N-terminal prohormone of brain natriuretic peptide (NT-proBNP) was not systematically measured, limiting interpretation of the results based on a risk stratification using the preoperative biomarker level. Likewise, postoperative high-sensitivity troponin T (hsTnT) was not systematically measured, with <50% of patients having troponin measurements, and it was not used to define postoperative MI.²² In the subgroup of patients with postoperative troponin measurements, no difference in troponin values was observed.

Clinical relevance

Despite an increased incidence of hypotension, the short duration of hypotension did not translate into a higher rate of postoperative complications. Hypotension appeared to have been appropriately detected and corrected by the anaesthesia teams using vasopressors.³ The greater intraoperative use of vasopressors over recent years might explain why RASI continuation was not associated with prolonged periods of hypotension compared with early reports when vasopressor use was less common, and fluids were unlikely to correct hypotension induced by vasodilatory medications.³ The impact of preventing hypotension on postoperative outcomes in randomised trials has been inconsistent, with most trials showing no difference.²³

The results of the STOP-or-NOT trial also align with other randomised trials investigating the strategy of RASI management before surgery. In the recent POISE-3 trial, where holding RASIs the day before surgery was part of a hypotension-avoidance strategy bundle, of 7490 patients enrolled, 71.7% were chronically treated with ACEIs or ARBs. The hypotension-avoidance strategy did not lead to a lower incidence of the primary outcome (a composite of vascular death and non-fatal MI after noncardiac surgery, stroke or cardiac arrest at 30 days after randomisation) compared with the hypertension-avoidance strategy. Of note, adherence to the protocol was suboptimal for a long-term antihypertensive medication strategy, ranging between 68.3% and 74.6%. In the POISE-3 trial, RASIs were held the night before surgery compared with 48 h before in the STOP-or-NOT trial in the hypotension strategy group (and until day 3 after surgery).

In the SPACE trial,²⁴ 236 patients aged ≥60 yr undergoing elective noncardiac surgery were randomly assigned to either discontinue before surgery (2–3 days before) or continue RASIs in six UK centres. The primary outcome, masked to investigators, clinicians and subjects, was MI (defined as an elevated plasma hsTnT: by ≥15 ng L⁻¹ within 48 h after surgery or ≥5 ng L⁻¹ when the preoperative hsTnT was ≥15 ng L⁻¹). Discontinuing RASIs before surgery did not reduce MI but was associated with an increased risk of clinically significant acute hypertension after surgery (defined as systolic blood pressure >180 mm Hg, diastolic blood pressure >100 mm Hg, or both from the time of randomisation until 48 h after surgery). In a post hoc analysis, patients with low preoperative brain natriuretic peptide and randomised to the discontinuation group were found to be at higher risk of MI after surgery and also did not have better outcomes in the hypotension avoidance group compared with the hypertension avoidance group.^25,26

Future directions

Overall, current evidence does not suggest an impact of continuation vs discontinuation of RASIs before major noncardiac surgery. A small pilot trial did not observe significant differences in postoperative outcomes among patients undergoing cardiac surgery.²⁷ Exploring the strategy in higher-risk procedures would be of value, but such a trial will be challenging. The Management of Angiotensin Inhibitors During the Perioperative Period (AIPOP) trial, which planned to enrol 3200 patients (mostly with low baseline cardiovascular risk) to evaluate the impact of preoperative RASI management on postoperative acute kidney injury, was terminated early after enrolling 342 patients because of a slow recruitment rate (NCT04506372). The results are not yet available. Future studies should focus on patients and procedures with the greatest risk.

A greater incidence of hypotension is likely among patients who continued RASIs before surgery, although the duration of this hypotension appears to be only slightly longer. The best strategy to prevent and treat those hypotensive episodes should be further explored, including the best choice of vasopressors (e.g. norepinephrine vs phenylephrine).²⁸ Finally, alternative strategies to decrease the elevated incidence of postoperative complications after major noncardiac surgery should be explored.

Conclusions

In the STOP-or-NOT trial, no differences were observed in the primary outcome of all-cause mortality and major postoperative complications within 28 days after surgery between the continuation and discontinuation groups, with both experiencing a 22% rate of death and major postoperative complications. Given that both continuation and discontinuation strategies resulted in similar postoperative outcomes, clinicians might have greater flexibility in managing RASI therapy based on individual patient factors and the specifics of the surgery, or patient preference, while being attentive to more hypotension and reacting accordingly.

Declarations of interest

ML is supported by grant R01 GM151494–01 and R01 DK139484-01 from the US National Institutes of Health (Bethesda, MD, USA), and received consulting fees from Alexion, Radiometer, Viatris and La Jolla. The STOP-or-NOT trial was funded by the French Ministry of Health (PHRC 2016).