Every surgery asks a patient to trust a future they cannot fully see, and one of its quieter risks is a stroke that may arrive in the weeks afterward. Researchers at Hackensack Meridian have built a calculator that reads the information already resting in a patient's medical record, such as age, past strokes, and narrowed blood vessels, and turns it into an estimate of that risk before the knife ever touches skin. Its promise is measured not only in statistical strength but in the hope that a warning given early can change what happens next, a hope that remains, for now, still waiting to be pr
New Web Calculator Aims to Flag Stroke Risk Before Surgery Using Existing Medical Record Data
Prediction alone does not show that a single stroke has been prevented.
So what does this calculator actually do?
It estimates a patient's chance of stroke within 30 days of a procedure, using things already in their chart: age, prior stroke, narrowed vessels, and the type of procedure.
Why build a new one? Aren't there already risk tools?
Yaghi says existing ones often exclude high-risk procedures and miss conditions like carotid stenosis and intracranial atherosclerosis. Those gaps are what this one targets.
Before we get excited, the 0.87 and 0.86 figures are the team's own reported numbers. They show the model separates higher from lower risk. They don't show that anyone has been helped.
What stood out among the predictors?
Prior stroke or TIA, with nearly sevenfold higher odds. Vascular, neurosurgical, and open cardiovascular procedures also raised risk.
Worth noting that's odds, not a patient's absolute chance. The interview doesn't give absolute rates, so I wouldn't read sevenfold as seven times likelier in plain terms.
How might it change what happens before surgery?
Possibly extra neurological evaluation, tighter control of vascular risk factors, an individualized plan, or sometimes a change in timing or type of procedure.
Those are possibilities Yaghi describes. Whether they actually lower stroke rates hasn't been tested yet. He says so himself.
And the rollout?
Hackensack Meridian plans to deploy it across its medical centers. The challenges are consistent data collection and not adding workload.
Also, it was built from three hospitals. Broader validation is still pending, so how it performs elsewhere is unknown.
O Pulso
- A tool built from nearly 445,000 procedures now offers surgeons a number that signals who may be most vulnerable to a stroke in the thirty days after an operation.
- Patients with a prior stroke or transient ischemic attack face nearly sevenfold higher odds, sharpening the urgency of knowing who is at risk.
- The calculator draws on existing electronic records, so it can slip into busy preoperative routines without collecting new data.
- Its accuracy scores held steady in an outside validation group, suggesting the signal is not just a quirk of one set of hospitals.
- Hackensack Meridian plans a system-wide rollout, yet the critical question of whether acting on these scores actually prevents strokes is still unanswered.
Every surgery asks a patient to trust a future they cannot fully see, and one of its quieter risks is a stroke that may arrive in the weeks afterward. Researchers at Hackensack Meridian have built a calculator that reads the information already resting in a patient's medical record, such as age, past strokes, and narrowed blood vessels, and turns it into an estimate of that risk before the knife ever touches skin. Its promise is measured not only in statistical strength but in the hope that a warning given early can change what happens next, a hope that remains, for now, still waiting to be proven.
A patient walking into the operating room rarely hears a word about the stroke that might follow in the weeks after. For those who do suffer one, the outcome can mean lasting disability, a longer hospital stay, or death, and around one in four adults worldwide will have a stroke at some point in life.
A new web-based calculator aims to bring that hidden risk into the light. Developed by a team led by Shadi Yaghi, a neurologist at Hackensack Meridian, it estimates the chance of a stroke within thirty days of a procedure using information already in a patient's electronic record. Yaghi says the idea came from the bedside, where neurologists see complications that might have been anticipated. Many existing tools miss certain high-risk operations and overlook conditions like carotid stenosis and intracranial atherosclerosis, so the team set out to build something practical that required no new data collection.
The model drew on nearly 445,000 procedures from three hospitals. A prior stroke or TIA stood out, carrying nearly sevenfold higher odds of a postprocedural stroke, and vascular, neurosurgical, and open cardiovascular operations also raised the risk. The calculator separated higher-risk from lower-risk patients well, with strong scores in both the development group and an external validation group.
The value lies in the window before surgery, when something can still change. A flagged patient might receive further neurological evaluation, tighter control of vascular risk factors, or a tailored perioperative plan, and in some cases the timing or type of procedure could shift. Yaghi is clear that the tool supports, rather than replaces, the judgment of doctors and patients.
Hackensack Meridian plans to roll it out across its medical centers, and Yaghi expects the integration to be manageable. The open questions are whether the results hold across more diverse settings and whether using the scores to guide care will truly reduce strokes. Those answers are still to come.
A patient heading into surgery carries a risk that rarely shows up in the conversation beforehand: a stroke in the weeks that follow. Around 1 in 4 adults worldwide will have a stroke at some point in their lives, and for people who have one after a procedure, the consequences can be permanent disability, a longer hospital stay, or death.
A new web-based calculator is meant to bring that risk into view before the operation. It estimates the chance of a stroke within 30 days of a procedure, using details already sitting in a patient's electronic medical record: age, a prior stroke, narrowed blood vessels. The work was published in the Journal of the American Heart Association in June 2026, and NeurologyLive spoke with one of its developers around World Stroke Day.
That developer is Shadi Yaghi, MD, FAHA, chair of neurology at Hackensack Meridian Neuroscience Institute at Jersey Shore University Medical Center and a professor of neurology at Hackensack Meridian School of Medicine. He says the idea grew out of what stroke neurologists see at the bedside: patients who suffer these complications and might have been identified in advance. In his account, existing risk calculators often leave out high-risk procedures and overlook important cerebrovascular conditions, including carotid stenosis and intracranial atherosclerosis. The team wanted something practical that gave individualized estimates without requiring new data collection.
The model was built from nearly 445,000 procedures across three hospitals. Among the strongest predictors of ischemic stroke were a prior stroke or transient ischemic attack, intracranial atherosclerosis, carotid artery stenosis, and the type and setting of the procedure. Patients with a history of stroke or TIA had nearly sevenfold higher odds of a postprocedural stroke. Vascular, neurosurgical, and open cardiovascular procedures were also linked to higher risk.
On performance, the calculator reached an area under the curve of 0.87 in the development cohort and 0.86 in an external validation cohort. Yaghi describes that as a strong ability to separate higher-risk patients from lower-risk ones.
The practical value, he argues, lies in the window before surgery, when something can still be done. A patient flagged as high risk might get additional neurological evaluation, better control of vascular risk factors, and an individualized perioperative plan. In some cases the result could affect the timing or type of the procedure. Yaghi stresses that the tool is not meant to replace clinical judgment, only to inform decisions made by physicians and patients together.
Hackensack Meridian Health plans to roll the calculator out across its medical centers. Because it draws on data already in electronic health records, Yaghi expects integration into preoperative workflows to be relatively straightforward. The main hurdles he anticipates are consistent data collection across hospitals and fitting the tool into practice without adding unnecessary work for clinicians.
Two questions remain open. Yaghi says further validation across diverse healthcare settings will be important, despite the model's strong results. And the team hopes to learn whether using the calculator to guide decisions actually reduces postprocedural strokes, which prediction alone does not show. His longer-term vision is for such tools to sit inside electronic health records and automatically flag high-risk patients.
Citações Notáveis
The calculator is not intended to replace clinical judgment but to help physicians and patients make more informed decisions.— Shadi Yaghi, MD, FAHA, Hackensack Meridian Neuroscience Institute
The team hopes to evaluate whether using the calculator to guide decisions can actually reduce postprocedural strokes.— Shadi Yaghi, MD, FAHA (paraphrased)