In the U.S. there are more than 150 million emergency department (ED) visits each year, and the number is expected to rise.
ED physicians face enormous pressure to quickly and correctly diagnose patients, who often present with ambiguous symptoms. Choosing the right diagnostic test is essential for optimizing patient outcomes, while also expediting care and keeping down costs.
During a November webinar sponsored by Beckman Coulter, Melissa Naiman, PhD, Medical Director for Sepsis and Host Response Diagnostics at Beckman Coulter, discussed with Nathan Shapiro, MD, Professor of Emergency Medicine and attending physician at Boston’s Beth Israel Deaconess Medical Center and Harvard Medical School, practical tips to help organizations choose from a growing array of diagnostic options and align them with ED workflows.
Four key takeaways were:
1. Host response tests support faster, more accurate diagnostic decisions and reduce sepsis risk. Diagnostic tests can help determine patients’ current infection status and provide a prognosis of how sick they are likely to become. Physicians often use two complementary test approaches to deliver the right diagnosis and prognosis: tests that identify the pathogen and tests that measure how the individual patient responds to the infection (host response).
“Host response is probably the biggest differentiator of [determining] how sick somebody is,” Dr. Shapiro said. Host response tests are especially helpful when a physician may have estimated a patient to be at no risk or low risk of infection but the test indicates otherwise, which would prompt a more in-depth patient assessment.
2. To evaluate the potential utility of emerging lab tests, organizations should consider likelihood ratios. Manufacturers often calculate likelihood ratios associated with test results in regulatory documentation. The positive likelihood ratio is the probability of a true positive result divided by the probability of a false positive result and the negative likelihood ratio is the probability of a false negative result over the probability of a true negative result.
As a rule of thumb, a positive likelihood ratio greater than 2 or a negative likelihood ratio less than 0.5 indicates the result should have some impact on a provider’s initial clinical impression. For FDA approval, likelihood ratios are reported in interpretation bands (such as low, moderate, and high risk). “You have to integrate these likelihood ratios with your probability of disease to try and put the whole picture together,” said Dr. Shapiro.
3. Recently cleared host response tests can be diagnostic or prognostic, single or multi-marker. In sepsis assessments, diagnostic tests help determine the type of infection, prognostic tests look at how sick a person is or is might become. Recently cleared prognostic tests for recognizing sepsis include single-marker tests and multi-marker signatures.
Three single-marker prognostic tests:
- Monocyte Distribution Width (MDW) measures the size and volume distribution of monocytes that occur during the host response to sepsis.
- IntelliSep measures white cell response to mechanical stress. During an infection, deformity changes can occur making it possible to identify patients who are more likely to be septic.
- Pancreatic Stone Protein (PSP) is primarily produced by the pancreas and immune cells and is an early indicator of sepsis.
Three multi-marker signatures:
- TriVerity Critical Illness uses 29 mRNAs and an algorithm to predict the risk of critical illness.
- SeptiCyteLAB uses a 4 mRNA signature and is intended for a sicker population of adults admitted to the ICU.
- Sepsis ImmunoScore is based on up to 22 inputs from the EHR, including a combination of demographic and vital sign data along with routine labs and a few novel sepsis markers. An AI algorithm uses this data to identify sepsis risk.
4. Ultimately, picking the right tests should be mapped to patient acuity and clinical scenarios. Regulatory clearance indicates the safety and effectiveness of a test, however their actual clinical utility depends on the needs of each organization, the patient population they serve and specific implementation aspects, including lab capacity. “Picking a test to implement is only the beginning of the journey,” Dr. Naiman said.
To register for upcoming webinars, click here.
