A New Way To Quickly Detect The No. 1 Patient Killer In Hospitals

A “sensor system in a bottle” could help hospitals identify sepsis infections in patients more quickly.

Patients arrive at hospitals needing treatment for all sorts of deadly ailments. Sometimes those ailments kill them. But sepsis, the No. 1 cause of death in hospitals, often plays a role.


Sepsis, a type of systemic inflammation, is most often caused by bacterial infection. It’s treatable with antibiotics, but treatment is most likely to work if the problem can be identified quickly. All too often, it isn’t.

A company called Specific Technologies thinks it has a new weapon in the fight against sepsis and other bacterial infections: a sensor system in a bottle.

1CSA PP patterned substrate (2)

Traditionally, labs use CO2 sensors to determine whether dangerous bacteria are present in a patient’s blood. These sensors are useful, but they can’t detect the species and strain of the bacteria growing–they can just sense that something is there. Further examining the bacteria using a traditional gram stain (gram positive and gram negative bacteria are treated differently) can take hours longer, because a lab needs to examine the results.

The SpecID system is more specific and speeds everything up, potentially giving patients a better shot at survival, according to CEO Paul Rhodes. The system, which is essentially just a bottle with a postage stamp-sized sensor attached, has shown a specificity of 95% in trials. It can get a gram ID in under 10 hours and can pick up an initial species ID in under 11 hours–no lab workers needed. Conventional CO2 sensors take 12 hours for a basic “yes” or “no” answer on whether dangerous bacteria is there.

“[SpecID] is a negligible cost to print. It can and will be embedded in every single blood bottle,” says Rhodes. He expects SpecID to be in clinical labs within nine months. Specific Technologies is already in conversation with existing manufacturers.

In addition to its uses on the hospital ward, SpecID has applications in bio-defense, where it could be used to quickly detect dangerous microorganisms like anthrax. Down the road, Rhodes believes that similar technology could sense antibiotic-resistant strains of tuberculosis in the developing world, where the disease runs rampant.


“We expect we’ll be doing it faster than existing methods for the same reason we’re faster in blood culture,” he says.


About the author

Ariel Schwartz is a Senior Editor at Co.Exist. She has contributed to SF Weekly, Popular Science, Inhabitat, Greenbiz, NBC Bay Area, GOOD Magazine and more.