Steering surgery fluids by cardiac output did not cut complications

Managing the circulation during a long operation is one of anaesthesia's central tasks, and for years the appealing idea has been to do it by the numbers — track how much blood the heart is pumping and top up fluids or drugs to keep that figure where it should be. A large European trial has now tested that idea directly in the patients who have the most to lose, and the answer is deflating: steering care by each patient's own cardiac index did not make major abdominal surgery any safer than ordinary hands-on management [s1].

The trial, led from University Medical Center Hamburg-Eppendorf and published in JAMA, was a multicentre, single-blind, superiority randomised clinical trial run at 12 university medical centres in Austria, the Czech Republic, Denmark, Germany and Spain [s1]. It enrolled high-risk adults scheduled for elective major abdominal surgery between October 8, 2023, and October 25, 2025, with follow-up completed on January 15, 2026 [s1]. Of 1,128 patients enrolled, 1,111 were randomised and 1,095 entered the primary analysis; their median age was 67 years and 49.4% were female [s1].

What the trial tested

Cardiac index is the amount of blood the heart pumps each minute, scaled to body size — a bedside gauge of whether the circulation is keeping the organs supplied. The premise behind "goal-directed" haemodynamic therapy is that holding that supply steady through the stress of surgery should head off the organ injuries that follow when it falls. The trick is deciding what "steady" means, since a healthy resting value differs from person to person.

This trial personalised the target. Patients assigned to the intervention had their cardiac index measured while awake and at rest before the operation, and the anaesthetic team then used fluids and drugs to hold the index at that individual baseline throughout surgery [s1]. The comparison group received routine haemodynamic management, with the cardiac-index monitor still attached but its readings hidden from the team [s1]. That blinded-monitor design is the trial's strength: both groups were instrumented identically, so any difference in outcomes could be attributed to acting on the number rather than merely measuring it.

The primary outcome was a composite of serious events in the first seven days after surgery — acute kidney injury, acute myocardial injury, severe infectious complications, nonfatal cardiac arrest, or death [s1]. These are the common, costly ways that a technically successful operation can still go wrong.

What it found

It made no measurable difference. The composite outcome occurred in 243 of 549 patients (44.3%) in the individualised cardiac-index group and in 256 of 546 patients (46.9%) in the routine-management group — a relative risk of 0.94, with a 95.1% confidence interval from 0.83 to 1.08 and a P value of 0.38 [s1]. An interval that comfortably spans 1.0 means the result is statistically compatible with no benefit, and the P value is nowhere near the conventional threshold. The individual components of the composite — the kidney, heart, infection, arrest and death endpoints — showed no between-group differences either [s1].

Two features of the trial make that null result hard to wave away. The event rate was high — close to half of patients in both arms had at least one complication — so this was genuinely a high-risk population in which a real benefit, had one existed, would have had room to show up [s1]. And because the monitor was running in both groups, the finding is specifically that titrating care to a personalised cardiac-index target added nothing, not that cardiac-output monitoring is useless information.

How to read it

The honest summary is that a plausible, physiologically grounded strategy failed its test. That is a common and important kind of result in critical care, where interventions that make sense at the bedside repeatedly turn out not to change what happens to patients once they are tested in a randomised trial. A single composite endpoint in elective abdominal surgery does not settle every question — the strategy might behave differently in emergency surgery, in other operations, or against a different target — but within the question this trial actually asked, the protocol did not earn its complexity [s1].

For readers, the practical translation is narrow and worth stating plainly: this is evidence about how anaesthetists should manage fluids and circulatory drugs during an operation, not advice about whether to have surgery or which hospital to choose. It adds to a run of recent trials interrogating what intensive perioperative management does and does not deliver, alongside work on intravenous iron for anaemia before cardiac surgery and fluid strategy in septic shock.

What to watch

The trial's authors frame cardiac-index targeting as unproven rather than disproven for every setting, and the registry record (NCT05648279) shows the study is now complete [s1][s2]. The larger lesson is methodological: monitors that generate ever more detailed numbers do not automatically improve care, and each goal-directed protocol has to be judged by whether acting on its number changes outcomes — not by how sophisticated the measurement looks.

This article describes research and is not medical advice. Decisions about anaesthetic and surgical care are for treating clinicians.