You are here

Study Questions Accuracy of Finger-Prick Blood Testing

Theranos takes another hit

Scientists at Rice University have found that results from a single drop of blood are highly variable, and that as many as six to nine drops must be combined to achieve consistent results. The study, which appeared online in the American Journal of Clinical Pathology, examined the variation between blood drops drawn from a single finger prick. The results suggest that health care professionals must take care to avoid skewed results as they design new protocols and technologies that rely on finger-prick blood.

“We began looking at this after we got some surprising results from our controls in an earlier study,” said lead investigator Dr. Rebecca Richards-Kortum. “Students in my lab are developing novel, low-cost platforms for anemia, platelet, and white blood cell testing in low-resource settings, and one of my students, Meaghan Bond, noticed there was wide variation in some of the benchmark tests that she was performing on hospital-grade blood analyzers.”

The benchmark controls are used to gauge the accuracy of test results from the new technology under study, so the variations among the control data were a sign that something was amiss. What wasn’t immediately clear was whether the readings resulted from a problem with the current experiments or actual variations in the amount of hemoglobin, platelets, and white blood cells (WBCs) in the different drops of blood.

Richards-Kortum and Bond designed a simple protocol to test whether there was actual variation, and if so, how much. They drew six successive 20-mL droplets of blood from 11 donors. As an additional test to determine whether the minimum droplet size might also affect the results, they drew 10 successive 10-mL droplets from seven additional donors.

All of droplets were drawn from the same finger prick, and the researchers followed best practices in obtaining the droplets: the first drop was wiped away to remove contamination from disinfectants, and the finger was not squeezed or “milked,” which can lead to inaccurate results. For experimental controls, the researchers used venipuncture, the standard of care in most hospitals, to draw tubes of blood from an arm vein.

Each 20-mL droplet was analyzed with a hospital-grade blood analyzer for hemoglobin concentration, total WBC count, platelet count, and three-part WBC differential, a test that measures the ratio of different types of WBCs, including lymphocytes and granulocytes. Each 10-mL droplet was tested for hemoglobin concentration with a point-of-care blood analyzer used in many clinics and blood centers.

Richards-Kortum and Bond found that hemoglobin content, platelet count, and WBC count each varied significantly from drop to drop.

“Some of the differences were surprising,” Bond said. “For example, in some donors, the hemoglobin concentration changed by more than 2 grams per deciliter in the span of two successive drops of blood.”

Richards-Kortum and Bond found that averaging the results of the droplet tests could produce results that were on par with venous blood tests, but tests on six to nine drops of blood were needed to achieve consistent results.

The new finding is bad news for blood-testing companies that rely on finger-prick testing for diagnostics, such as California-based Theranos, according to an article posted on the BioSpace website.

In 2015, Theranos was the target of allegations over the efficacy of its finger-prick testing, and the Wall Street Journal ran a series of articles questioning the reliability and accuracy of the company’s testing methods. In response, Theranos said it stood by its technology and questioned the Journal’s use of “inexperienced and disgruntled former employees.” Moreover, some of the company’s laboratory practices have been questioned by both the FDA and the Centers for Medicare and Medicaid Services, according to BioSpace.

Sources: Rice University; February 17, 2016; and BioSpace; February 24, 2016.

Recent Headlines

Possible First Treatment Option for Rare Autoimmune Disease of the CNS
New Hematologic Biomarker Test Provides New Approach to Sepsis Triage and Diagnosis
Antibiotics, Statins, and Glucocorticoids All Show Promise
More Than 32% of Patients Responded in Trial
Current, Sole Therapy Not Always Successful/Suitable
Over 1.5 Million Americans Likely to have Wet AMD by 2020 
Potential For Use Against E. Coli, TB, Resistant Bacteria