Could a saliva sample reveal signs of heart failure without a blood draw? This study found that a less perfect form of graphene could detect the signal more reliably than the cleaner, more orderly version. When heart muscles are damaged, NT-proBNP is released into the bloodstream and can also be found in saliva.

That makes it a promising marker for heart failure testing outside a hospital laboratory. Saliva is attractive because it is fast and easy to collect, without complex and invasive preparation. But the amount of NT-proBNP in saliva can be lower than one picogram per milliliter, making detection difficult.

The two sensor materials cost about the same, but their properties and usefulness differ. CVD graphene is an almost ideal, intact crystal, while graphene oxide is a semiconductor that keeps an energy gap. CVD graphene is described as an almost ideal intact crystal with zero band gap, whereas graphene oxide is semiconducting and retains an energy gap.

The study compared a device made with CVD graphene, functionalized with an aptamer designed to recognize NT-proBNP, against one made with reduced graphene oxide. Both devices were tested across a wide range of NT-proBNP concentrations in a liquid and in diluted artificial saliva, a model used to examine interference from bodily fluids.

As NT-proBNP bound to the catcher, both devices shifted their electrical response in the same direction, although the size of the shift differed. The change was stronger in the reduced graphene oxide device. In the CVD graphene device, the Dirac-point signal was weak and quickly reached saturation as the NT-proBNP concentration increased further.

In reduced graphene oxide, binding affected both electrostatic doping and direct doping to trapped states in rGO, producing a more pronounced response. This comparison matters because the two sensor designs respond in the same direction as NT-proBNP rises, even though their signal sizes differ.

In saliva-like fluid, the electrical background itself shifts strongly, yet both still register the target. Both biosensors showed low sensitivity to cTnI, indicating good selectivity, but unexpectedly showed higher sensitivity to the proBNP peptide instead.

The study suggests that the way the catcher lies on the graphene surface may increase this unexpected response. It also warns that sensing properties may decline after long-term storage. The devices were tested for detecting NT-proBNP in diluted artificial saliva.

The tested levels matched the working range expected in real saliva from patients with heart failure. The goal is to make non-invasive monitoring of NT-proBNP from saliva possible, and the study tested the aptasensors in spiked artificial saliva.

The comparison shows that choosing a sensor material requires more than looking at the material alone. The marker and the way it changes the electrical signal also matter, because this small peptide can interact with its catcher in different ways and create large measurement errors.

For this target, the study prefers the reduced graphene oxide device because its larger energy gap and trapped states can increase sensitivity and reduce noise; the assay was also demonstrated in diluted artificial saliva. The key lesson is that the best sensing material depends on what it must detect.

For this heart-failure marker, the rougher graphene-based device was more sensitive, pointing toward simple, non-invasive testing from saliva.