Article Text
PDF
Abstract
Introduction Cardiac surgery is frequently associated with vasoplegia and vasopressor treatment. Both may be associated with postoperative complications and prolonged length of stay. The most frequently used vasopressor is norepinephrine. However, in a pilot, double-blind, randomised controlled trial (RCT) in cardiac surgery patients, angiotensin II was effective in maintaining blood pressure and was associated with a shorter duration of hospital stay than norepinephrine. Furthermore, hyperreninaemic patients were more sensitive to angiotensin II. These findings support the need for a larger RCT to determine whether angiotensin II is superior to norepinephrine as a first-line treatment for low blood pressure after cardiac surgery.
Methods and analysis We will conduct a double-blind RCT comparing an infusion of either angiotensin II or norepinephrine intraoperatively and for up to 48 hours after the start of surgery. We will randomly allocate 400 cardiac surgery patients at multiple centres in two countries to either an equipotent angiotensin II or norepinephrine infusion, titrated to a mean arterial pressure of 70–80 mm Hg. The primary outcome will be length of hospital stay. Secondary outcomes will include a composite of renal, cardiovascular and neurological events. A subgroup analysis of patients with elevated baseline renin levels will be undertaken.
Ethics and dissemination Ethical approval has been granted by the Alfred Human Research Ethics Committee on 14 July 2023 (HREC/97814/Alfred-2023). Results will be published on completion of the trial.
Trial registration number Australian and New Zealand Clinical Trials Registry: ACTRN12623000848606.
- Cardiac surgery
- ANAESTHETICS
- Adult intensive & critical care
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The planned study is a multicentre double-blind, randomised trial, which will minimise bias and confounding.
The intervention begins early, after induction of anaesthesia, therefore avoiding a period of ‘non-study’ treatment for hypotension.
This trial is supported by the findings of a pilot double-blind randomised controlled trial at more than one centre.
Some patients may require only low doses or short duration of exposure to the study drug.
Due to the low mortality and low major morbidity of cardiac surgery patients, this study is not powered to identify differences in these patient-centred clinical outcomes.
Introduction
Cardiovascular disease is a major cause of morbidity and mortality, and each year more than 2 million people worldwide require cardiac surgery.1 Medical complications frequently occur in the immediate postoperative period and are a source of considerable cost, both in terms of patient morbidity and healthcare expenditure.2
Vasodilation occurs in most patients having cardiac surgery with varying severity. In this setting, vasopressor drugs are used to restore vasomotor tone and blood pressure in up to 80% of patients (unpublished data: Australian and New Zealand Society of Cardiothoracic Surgeons database). Severe vasodilation—termed ‘vasoplegia’ or ‘vasoplegic shock’—occurs in up to 20% of patients.3 Importantly, both vasoplegia and its treatment with vasopressor drugs are associated with serious complications, such as acute kidney injury (AKI), atrial fibrillation (AF) and stroke.4–6
The standard first-line vasopressor used in cardiac surgery is norepinephrine.7 However, recently, angiotensin II has been reported to be effective in restoring blood pressure in patients with vasoplegic shock resistant to standard vasopressors.8 Post-hoc analyses have reported improved renal and respiratory outcomes.9 10 Moreover, ACE inhibitors are frequently used in this group of patients; their use, combined with endothelial stress and the use of cardiopulmonary bypass which reduces the availability of ACE, suggests a high risk of angiotensin II deficiency. This implies that angiotensin II may be a particularly effective drug to treat vasoplegia in this group of patients.
Our recently completed pilot study, a double-blind, randomised controlled feasibility trial in 60 patients, compared angiotensin II to norepinephrine to determine whether a larger trial in cardiac surgery was feasible and justified.11 The trial showed a high degree of feasibility and preliminary evidence of benefit for angiotensin II. Moreover, we found that early postoperative renin increase was suppressed by angiotensin II compared with norepinephrine. Finally, we observed that in a subgroup of patients with higher baseline renin levels, higher doses of norepinephrine, but not angiotensin II, were needed to treat vasoplegia.12 These findings are consistent with other small studies.13
Given these preliminary findings suggesting a possible reduction in length of stay in the angiotensin II group, we hypothesise that angiotensin II infusion, when used to maintain a target perioperative mean arterial pressure (MAP) >70 mm Hg, will result in a shorter length of hospital stay compared with norepinephrine. We further hypothesise that this difference will be associated with and occur as a result of a reduction in the overall number of complications.
Methods and analysis
Participants
The trial will be carried out in hospitals across Australia and New Zealand. Patients ≥18 years undergoing cardiac surgery with cardiopulmonary bypass and an elevated risk of AKI will be eligible to participate.14 Increased risk of AKI will be defined as the presence of at least one of the following criteria: haemoglobin <130 g/L, creatinine >100 µmol/L, age >70 years, New York Heart Association (NYHA) class 4 and body mass index (BMI) >30 kg/m2; therefore, patients will need to meet at least one of these criteria to be included in the study.
Exclusion criteria are patients who underwent major aortic surgery, transplant surgery, pulmonary thromboendarterectomy, ventricular assist device placement, patients already receiving inotropic/vasopressor support, dialysis-dependent, prior renal transplant, pre-existing significant hypertension (persistent systolic blood pressure >180 mm Hg), severe pulmonary hypertension (estimated pulmonary systolic arterial pressure> 70 mm Hg, mean pulmonary artery pressure >40 mm Hg) with right ventricular systolic dysfunction and pregnancy or breastfeeding. Participant recruitment started in January 2024. As of 11 February 2025, 120 patients had been enrolled. We expect recruitment to be completed by the end of 2025.
Patient and public involvement
Patients and the public were not involved in the preparation of this protocol.
Intervention
The study drug will be administered by infusion through a central venous catheter after induction of anaesthesia, but prior to the institution of cardiopulmonary bypass, in response to a MAP <70 mm Hg. A predefined study drug administration protocol will then be applied, targeting a MAP of 70–80 mm Hg (figure 1). The protocol includes an algorithm for assessment and treatment of non-vasoplegic causes of hypotension or hypertension, such that all important physiological requirements are met (ie, adequate cardiac output and fluid status). A pragmatic approach using clinician discretion is to be used to determine that adequate cardiac output and fluid administration have been achieved. Once this assessment has taken place, the study drug can be started at 1 mL/hour and titrated as required (figure 1). All other aspects of care including the use of fluid resuscitation and open label vasopressor therapy (metaraminol/phenylephrine/norepinephrine/vasopressin/other) are at the discretion of the clinician. If the maximum rate of 20 mL/hour is reached and provides insufficient vasopressor effect, then an additional open label vasopressor is recommended (including open label norepinephrine). In our pilot and feasibility study, which used a similar design, open label norepinephrine use was required in 8/60 (13%) patients.11 The study drug infusion is continued in the intensive care unit (ICU) until it is no longer required to maintain MAP of 70–80 mm Hg, for a maximum of 48 hours (total) or until ICU discharge—whichever occurs first. The infusion can be paused at any point for control of blood pressure due to bleeding or for other surgical concerns. Cardiopulmonary bypass management is ‘goal-directed’ to achieve an oxygen delivery of 280 mL/min/m2,7 or to deliver an appropriate arterial blood flow indexed to patient size according to local perfusion policy (usually >2.2 L/min/m2). Haemoglobin concentration will be maintained ≥70 g/L.
Flow chart describing titration of drug to effect. MAP, mean arterial pressure.
To assist blinding of clinicians, infusion of the drug at the same rate in mL/hour will give a similar vasopressor effect (ie, the study drugs will be diluted to a concentration that results in similar vasopressor potency at equal infusion rates). Angiotensin II 2.5 mg will be added to 250 mL normal saline giving a concentration of 10 µg/mL; norepinephrine 15 mg will be added to 250 mL normal saline giving a concentration of 60 µg/mL. This ratio of concentrations is based on our feasibility study and prior work.11 15 16 The study drug will be administered as shown in table 1, with corresponding weight-based infusion rate ranges for patient weights of 50–100 kg. Based on feedback from our pilot trial, infusion rates are not primarily based on weight to increase the ease of protocol administration and reduce the potential for drug error. Nonetheless, clinicians can ascertain the possible weight-based dose and treat patients accordingly. The table will be used as a guide; infusion rates between the rates in the table (eg, 10 mL/hour) will be allowed at clinician discretion. Study drug use data will be collected during the study.
Range of allowable infusion rates, with corresponding dose ranges for 50–100 kg patients
All patients will have blood sampling for renin and aldosterone levels prior to surgery. A subset of 60 patients will have blood sampling prior to the start of the surgery, at the end of surgery and at 12–16 hours after ICU arrival.
Outcomes
Primary outcome
Length of hospital stay—this will be defined as the duration from anaesthesia start until hospital discharge.
Secondary outcomes
We will assess both a composite of key outcomes and individual secondary outcomes as described below.
A composite binary outcome of any one of the following complications within 7 days of surgery: any stage of AKI according to the Kidney Disease Improving Global Outcomes (KDIGO) criteria, new onset AF/atrial flutter, prolonged ICU length of stay (>48 hours) and delirium (as defined below).
Renal complications:
AKI according to KDIGO criteria—defined as the difference between preoperative creatinine and highest postoperative creatinine within the first 7 days after surgery.
Major adverse kidney events defined as death, renal replacement therapy or persistent loss of estimated glomerular filtration rate at discharge >50%.
Use of renal replacement therapy.
KDIGO stage 2 or 3 AKI.
Duration of KDIGO stage 2 or 3 AKI.
Cardiovascular complications:
Requirement for mechanical cardiovascular support postoperatively.
New onset AF/atrial flutter within 7 days of surgery.
Venous thromboembolism.
Other vasopressor use, other inotrope use,
Neurological complications:
New permanent stroke.
Delirium (use of psychotropic drugs for delirium control or bedside screening test (Confusion Assessment Method ICU or other validated method) positive for delirium).
Other:
Length of ICU stay.
Duration of mechanical ventilation.
Hospital mortality.
Return to the operating theatre.
Readmission (to hospital or ICU).
Any infectious complication.
Planned subgroup analyses
We will carry out planned subgroup analyses in two groups—patients with a high baseline renin and patients with additional biochemical samples available for analysis—as listed below:
We will compare a composite binary outcome of AKI (any stage by KDIGO criteria), new onset AF/atrial flutter, prolonged ICU length of stay (>48 hours) and delirium in a subset of patients with a baseline renin level >150 microunits/mL (approximately one-third of patients in pilot data, approximately equivalent to >90 pg/mL).11 Our hypothesis for this subgroup analysis is that patients given angiotensin II with high baseline renin levels will have a reduction in the composite outcome compared with norepinephrine.
In a further subset of 60 patients, we will explore temporal changes in the renin-angiotensin-aldosterone system (RAAS). Our main outcome of this analysis will be to confirm findings from a previous study suggesting renin levels are suppressed by angiotensin II relative to norepinephrine infusion. Secondary biological analyses will include temporal changes in the components of the RAAS, including:
ACE 1 and 2 levels and activity.
Angiotensin I and II levels.
Angiotensin 1–7.
Other markers related to RAAS physiology.
Primary outcome sensitivity analysis
It is possible that early deaths in one group could bias the length of stay in that group. To overcome this, we will compare days alive and at home within 30 days after cardiac surgery to allow for the competing risk of mortality (such that a death at any point in the first 30 days after surgery gives a score of zero). This outcome comprises both length of stay and mortality.
Recruitment, sample size and timeline
Patients will be identified from preoperative clinic lists or from inpatient surgical lists. We will seek informed consent from eligible patients prior to surgery. A participant timeline is shown in table 2.
Participant timeline
Our feasibility study showed a mean length of hospital stay of 9.9 (SD 7) days vs 7.1 (SD 1.9) days in the norepinephrine relative to the angiotensin II groups. This represented a 2.8-day mean difference. To allow for a more conservative estimate of effect, we will use a difference of 1.8 days between the two groups, with the same SD. Based on an independent groups student's t-test with equal allocation between treatment arms and assuming homogeneity of variance between arms and a power of 90%, the study will require a sample size of 346 patients (173 per treatment arm). To compensate for data loss and the possible effect of skewed data, we will randomise 400 patients.
Sequence generation and concealment
Patients will be allocated randomly to treatment arms in a 1:1 ratio. Randomisation will be stratified by site in randomly sized blocks of 2, 4 and 6. A random allocation sequence will be generated by computer and randomisation will occur using the REDCap database system.
The allocated study drug will be reconstituted by an unblinded member of staff or pharmacy not involved in the clinical care of the patient or research data collection. The study drug will be provided to the clinician and marked as ‘PORTHOS Study Drug’. Clinicians involved in patient care, the patient and researchers collecting or analysing outcome data will be blinded to group allocation. Given the short half-life (min) of the drug, it is unlikely unblinding will be necessary, as stopping the drug will result in rapid resolution of any drug-related effects.
Safety and monitoring
Deidentified data will be collected and stored on the REDCap database. Data will be collected contemporaneously. All patients will have their medical records reviewed and will be contacted 30 days after surgery to assess trial outcomes, safety endpoints and other adverse events.
An independent data and safety monitoring committee (DSMC) has been established. A DSMC charter is available on request. Interim analyses will be completed after 25% and 50% of patients receive the study drug (n=100 and n=200). Stopping rules for efficacy or harm will be according to the O’Brien-Fleming principles.
The CRF has been designed to capture safety endpoints following surgery routinely to enable the DSMC to review data in a meaningful way, by independent experts, with blinded separation of treatment groups enabling useful comparisons for the whole dataset. This then will enable the generation of a concise and relevant safety report for dissemination and review by all participating regulatory bodies at all participating centres.17 Safety endpoints and adverse events will be collected up to 30 days after surgery. Sites will directly report adverse events that are not captured as part of the CRF (ie, not trial or safety endpoints) and occur within 30 days. Trial conduct may be audited where concerns arise.
Statistical analysis plan
All data will be analysed on an intention-to-treat basis, defined as all assigned participants who undergo cardiac surgery except those for whom consent was withdrawn, without imputation of missing data unless specified.
Continuous variables will be assessed for normality and log transformed where appropriate. Baseline characteristics and processes of care data will be compared using χ2 tests for equal proportion, Student’s t-tests for normally distributed data and Wilcoxon rank sum tests otherwise, with results reported as counts with percentages, mean±SD or median (IQR), respectively.
The primary outcome will be analysed using a mixed-effects linear model with treatment group as a fixed effect and centre as a random intercept to account for the clustering effect of patients within hospitals. If a log transformation is applied, the results will be presented using geometric means and ratio of geometric means along with 95% CIs. A sensitivity analysis for the primary outcome of days alive at home within 30 days will be performed to account for the competing risk of mortality.
Binary outcomes (hospital mortality, return to theatre and complications) will be presented as proportions of participants in each treatment arm with the event. Generalised linear mixed-effects models with a binomial distribution and a log link will be used to estimate relative risks (95% CI), binomial regression with an identity link function to derive risk differences (95% CI) and logistic regression to derive ORs (95% CI) where binary model convergence is problematic.
Continuous secondary outcomes (ICU length of stay and duration of mechanical ventilation) will be analysed using mixed-effects linear models with treatment group as a fixed effect and sites treated as random effects. Non-parametric outcomes will be analysed using quantile regression with results reported as a difference of medians (95% CI).
In addition to the analyses described above, supportive analyses will be performed in the per protocol population excluding patients with major protocol deviations. All analyses will be performed using SAS V.9.4 (SAS Institute, Cary, North Carolina, USA) and a two-sided p value of 0.05 will be used to indicate statistical significance.
Ethics and dissemination
This study will be performed in accordance with ICH GCP (in accordance with the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use Good Clinical Practice Guidelines) notes for Guidance on Good Clinical Practice (ICH E6). Research ethics approval has been granted from the Alfred Hospital Research Ethics Committee (HREC/97814/Alfred-2023) on 14 July 2023 and the relevant ethics committee of each participating hospital. Informed consent will be obtained from trial participants by investigators and research staff in the department of anaesthesiology and perioperative medicine at participating sites. Data will be deidentified as described above. All data will be stored securely and indefinitely. The coordinating PI (Principal Investigator) will manage access to the final trial dataset. Any changes to the protocol will be communicated to relevant parties via the coordinating PI, steering committee and/or delegates thereof. Trial registration with the Australian and New Zealand Clinical Trials Registry occurred prior to enrolment of the first patient.
The dataset may be provided to other named investigators on request after trial completion. La Jolla Pharmaceutical Company (USA) will be granted access to the deidentified dataset after publication of the study. Results will be disseminated by publication and presentation. Publications will be written independently by the authors of this manuscript. Additional authors for subsequent manuscripts will be included according to ICMJE (International Committee of Medical Journal Editors) guidelines. Results will be disseminated to participants on completion of the trial.
The Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines were used in preparation of this protocol.18
Ethics statements
Patient consent for publication
Footnotes
X @LachieMiles
Contributors Protocol design and manuscript draft: TGC, LFM, DF, DP, SFM and RB. Statistical analysis plan: EP. Guarantor: TC.
Funding This study was supported by a grant from the Australian and New Zealand College of Anaesthetists (24/025). It also received support as an investigator-initiated research study from La Jolla Pharmaceutical Company, an affiliate of Innoviva Specialty Therapeutics, Inc.
Competing interests RB—received paid services as a member of a medical advisory board to La Jolla Pharmaceutical Company, Viatris and Paion AG. EP—none, TC—none, LM—none, DF—none, DP—none and SM—none.
Patient and public involvement Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.