Behind the Scenes in the Flu Lab

Flu has a lot in common with winter. Its severity varies from one year to the next, but the worst of both usually occurs in December, January, and February. The impact of each on an individual can range from mild to unpleasant to quite dangerous. And though there is really no way to know in advance how difficult either a flu season or a winter will be, experts try to make predictions to help people stay safe.

When it comes to influenza, this involves a worldwide collaborative effort among physicians and scientists—including Yale Medicine experts who work directly with lab specialists and epidemiologists at the Centers for Disease Control and Prevention (CDC) in studying flu-related developments in real time. We spoke recently with Marie Louise Landry, MD, a Yale Medicine virologist and infectious disease specialist and director of the Yale Clinical Virology Lab about what that means for patients—but, first, here’s a look at why this in-the-moment tracking is important.

Each February or March, the World Health Organization chooses the strains that the next flu vaccine is meant to protect against. The long lead time is necessary to make and distribute the vaccine by the following October.

But the circulating viruses don’t always align with the vaccine prediction. “We call this a mismatch,” explains Dr. Landry. “One thing we know about influenza is that though we try to predict what the virus is going to do next, it may change and outwit us.”

Ten years ago, this was especially true. In 2009, Americans were already braving a brutal flu season when a new, dangerous strain—never before identified in animals or people—emerged. That 2009 H1N1 strain—initially dubbed “swine flu” as it was believed to have originated in pigs—was declared a pandemic because it sickened so many people worldwide. In the U.S., there were more than 60 million cases of swine flu, causing more than 12,000 deaths—80% of which occurred in people younger than 65. In a typical flu season, 70 to 90% of deaths are in people older than 65. 

Differentiating the 2009 pandemic flu from other strains was critical, so that proper precautions to stop its spread could be taken. Yale was uniquely equipped for the task. As soon as a new flu strain was recognized, specialists at the Yale Clinical Virology Lab were able to test for that specific strain, something rapid tests offered in doctors’ offices and in most other hospitals could not do. That’s because Yale New Haven Hospital’s lab is one of only a few nationally equipped with specialized technology, including advanced lab techniques and instruments, provided by the CDC.   

Here is our conversation with Dr. Landry about this equipment, the impact it made in 2009, and why it matters now. 

Most doctors’ offices use what is called a rapid flu test. For many years, the only such test available was a simple one that detected the proteins or “antigens” of the virus particle. It typically required no equipment. Just a nasal or throat swab that was inserted into a liquid buffer and mixed well. Then, the liquid was applied to the test strip or cartridge. Ten to 15 minutes later, the result is read by eye and consists of a colored line on the strip.   

This was a big advance when it first became available, but the test requires about a million virus particles to generate a positive result. So, it is not what is called “sensitive.” Interestingly, children have more virus in their nasal passages than adults, so these rapid tests work better in them—in adults, there is a higher rate of a “false negative” result. 

In 2018, the Food and Drug Administration (FDA) began enforcing higher sensitivity requirements for flu tests. For some of the tests now, small instruments are used to read the results. But the old test is still used in many walk-in clinics and doctors’ offices due to simplicity, speed, and cost.  

The big change at hospitals in the past few years has been a transition to tests that detect the genetic materials of the influenza virus. The key to test sensitivity is that the viral genome is multiplied by a biochemical reaction a million- to a billion-fold before the detection step. As a result, you can have small amounts of virus in the sample, but the test itself amplifies it, so it is detectable. These types of tests are called “molecular assays,” and the first and most commonly used one is known as "polymerase chain reaction (PCR)."

Initially, molecular tests took one or two days to complete and required specialized expertise not available in most hospitals. In recent years, simpler, faster versions of molecular flu tests have become available, but they are much more expensive and require more expensive instruments and service contracts. In the hospital setting, though, where infection prevention (i.e., avoiding transmission to other patients) is critical, having the most sensitive test is key. 

Influenza is unique in its ability to change itself. Changes in the virus allow it to elude our immune defenses and be missed by diagnostic tests. As new strains appear, we need to be sure we can detect them. The CDC continuously monitors circulating strains. Every year we check in to see if there are any changes we need to make. During the year, if we see changes in how the test is performing, we let the CDC know. 

Having the CDC Influenza PCR in our laboratory at Yale was most useful in 2009 when the new pandemic flu appeared. We were able to immediately set up the new assay and were the only hospital in the state—and one of very few in the country, other than public health laboratories—that could identify the new pandemic strain of influenza virus. 

Yale New Haven Hospital has a separate clinical virology laboratory, which is uncommon in the hospital setting. Diagnostic virology embraced molecular amplification methods like PCR very early. Since viruses can be difficult, dangerous, or just very slow to grow in culture, we were highly motivated and developed the expertise to be able to offer PCR for all of the common viruses, including influenza, 15 to 20 years ago. When the CDC came out with an improved influenza assay in December 2008, we immediately adopted it. This was very timely, since a few months later the 2009 pandemic struck, and we were ready. 

In our emergency department, we do the rapid commercial influenza test for new admissions. If it is negative, we can retest with the CDC test. If a new strain appears, we will be able to identify it more quickly and can update our test based on CDC recommendations. 

[In 2009] it was important for people to know if they had the standard seasonal flu or the new pandemic strain because the new strain was more serious—especially for children and younger adults—it differed from seasonal flu in terms of sensitivity to antiviral therapy, and there was no vaccine. That meant we needed enhanced protection measures, including asking health care workers infected with the new flu strain to stay out of work longer. Knowing what you are dealing with is very useful in terms of patient management and treatment. 

We are diagnosing only a few flu cases per week in our laboratory. In a typical year, influenza activity can be expected to pick up substantially in mid to late December. 

While the vaccine is imperfect, it remains our best protection. The hope is even if it fails to completely protect us, it will reduce its severity and save lives. It’s like wearing a seat belt. It will not prevent all injuries or deaths, but we are better off with it than without it.