What your blood pressure reading misses

**SPEAKER_1** (0:02)
This is a podcast by The Straits Times.

**Joyce Teo** (0:08)
Hi, I'm Joyce Teo. Welcome to Health Check.
In this podcast, we will look at the importance of fat loss over weight loss, and why a normal blood pressure reading may not always tell you the full story.
If you walk into a clinic for a routine checkup, the doctor will typically verify your blood pressure using a standard arm calf to ensure it remains within a healthy range. And for millions of people worldwide living with hypertension, keeping blood pressure numbers in the normal zone is the ultimate goal. But what if that piece of mind is an illusion? Recent findings from the remodeled study at the National Heart Center Singapore suggest that the true danger to your heart might be completely invisible to standard clinical checks. The study was published as a letter in the journal MedCom in February 2026 Today, we sit down with the senior author of that study to find out more about it. He's Associate Professor Calvin Chin. Dr. Chin is the Senior Consultant in Cardiology at the National Heart Center Singapore and the Deputy Director of the National Heart Research Institute Singapore. Dr. Chin, welcome to the show.

**Calvin Chin** (1:14)
Well, thanks for having me.

**Joyce Teo** (1:16)
So I want to start with a statistic from your paper, right? You wrote that nearly 30% of health complications related to high blood pressure occur in patients whose BP appears perfectly normal on paper. You know, how is that possible?

**Calvin Chin** (1:31)
Blood pressure is only one of the many risk factors for heart disease. And obviously at this point, controlling blood pressure is still the right thing to do. But it's only one piece of the puzzle that we know.
Two patients with very similar blood pressure reading can have very different underlying biology. And that's what we are seeing in our study. That one patient can have relatively healthy metabolic profile. The other one can have ongoing inflammation, for instance, adverse heart remodeling or excess visceral fat that may not be apparent during a routine clinical visit. So our findings suggest that some of these patients will have higher risk despite having blood pressures that appear to be well controlled. And that just means that they have got other underlying biology that's driving the adverse outcomes.

**Joyce Teo** (2:18)
Okay, so it means the blood pressure numbers that we've been relying on are just kind of like scratching the surface. We don't really know.

**Calvin Chin** (2:25)
Yes, it gives an incomplete picture, but it is obviously a very important piece of the puzzle. But I think there are other things that are going on.

**Joyce Teo** (2:33)
Okay, so tell us a little bit about the remodels study.

**Calvin Chin** (2:36)
So the remodels study is actually a cohort that was started more than 10 years ago. And in consent of participants, we will collect their information from cardiac MRIs, ECG, blood biomarkers, or clinical information to study biological variation. And then these patients or participants are then followed up for health complications over time.
And the whole idea of setting up this remodel cohort is to help us move towards a more personalized risk assessment, so that we can identify some of these individuals earlier who may benefit from more intensive interventions.

**Joyce Teo** (3:12)
Okay, so these are the patients with hypertension?

**Calvin Chin** (3:14)
Yes, hypertension with some of them also has got concomitant, bad cholesterol, diabetes, fatty liver, obesity, for instance.

**Joyce Teo** (3:23)
Oh, okay, but in the study, you use machine learning to track them over time, right? So the AI, the machine doesn't know that whether they are healthy?

**Calvin Chin** (3:30)
Yes, the algorithm is agnostic to their baseline characteristics. But what we do is that we put all of this data, the blood imaging clinical data, such as the amount of fat in the body, heart function, the blood pressure, blood markers of heart injury, inflammation.
Then we use a data-driven clustering approach. Basically, what it does is it looks across all these different measurements to identify individuals with similar biological characteristics and then group them into different risk categories. We identify two risk categories in this study, a low risk and a high risk.

**Joyce Teo** (4:07)
Okay, so tell us what did the high risk group look like?

**Calvin Chin** (4:10)
Yeah, so very interestingly, we found that the high risk individuals on average was about 54 years old and they were about four years younger than the ones at low risk, contrary to what we believe, right? And these young high risk individuals also has got significantly greater amount of visceral fat and these are the type of fat that's hidden inside the body. And this was observed despite having similar amount of subcutaneous fat. And this is fat that is under the skin that you can see.