Published in the January 2019 issue of Today’s Hospitalist
PEAK FLU SEASON is approaching, bringing back memories of last winter, which proved to be the deadliest season for influenza in decades. In all, nearly 1 million people were hospitalized in the U.S. last year for flu and 80,000 died. Those somber statistics show that influenza is a tough diagnosis to make, according to Jennifer Babik, MD, PhD, an infectious diseases physician and associate clinical professor at the University of California, San Francisco (UCSF).
During last fall’s management of the hospitalized patient conference at UCSF, Dr. Babik spelled out the key clinical features, diagnostic tests and antiviral options for treating flu in inpatients. “We know that influenza is often under-diagnosed in the hospital,” she told audience members, “and antivirals are underutilized.”
Common misconceptions
Here’s one reason flu can be so hard to diagnose: “We always think that fever, cough, or fever and cough together are the hallmarks of influenza,” said Dr. Babik. But data indicate that none of these findings is very specific, although their specificity does go up in patients over age 60.
“Influenza is often under-diagnosed in the hospital, and antivirals are underutilized.”
~ Jennifer Babik, MD, PhD University of California, San Francisco
However, in patients 60 and older, only 35% of those presenting with flu have a fever. Those “classic” symptoms are even less likely in immunocompromised patients who are also more likely to need to be hospitalized and intubated. In addition, they shed virus longer—a median of eight days vs. five for patients who are immunocompetent—and 15% can shed for even longer periods, including more than 30 days.
Another factor that makes a flu diagnosis difficult: Doctors assume that patients who have been vaccinated don’t have the flu.
“Don’t take that into consideration for either diagnosis or treatment,” Dr. Babik pointed out. “Vaccine effectiveness is really a 50-50 shot in any given year, based on the predominant subtype.” Against influenza B or H1N1, for instance, vaccine effectiveness runs around 60%. But effectiveness falls to only 33% for H3N2 when the vaccine is a good match—and even lower (23%) when it’s not.
Clinicians also tend to put too much faith in rapid antigen testing in the ED when those results come back negative. “They’re useful if they are positive,” she explained, “but they’re only between 50% and 70% sensitive, so they cannot be used to rule out influenza during flu season.” UCSF has actually stopped using rapid antigen testing because of false negatives and missed flu cases. Instead, molecular assays “are very, very good, with about 95% sensitivity and specificity,” and some assays can determine influenza subtypes. “They’re really the test of choice if they’re available.”
Who should you test in the hospital during flu season? “All inpatients with an influenza-like illness or pneumonia,” Dr. Babik said. “That’s going to be a lot of people.”
Further, the test is typically an upper tract sample with a nasal swab. But shedding can decrease after about five days, she noted. Studies looking at paired samples in ICU patients have found that up to 40% of patients with a positive lower tract sample can have a negative upper tract one, which she characterized as a “terrifying” number.
“When patients are critically ill,” Dr. Babik said, “collect upper and lower tract samples, and do not stop empiric therapy until the lower tract one is negative.”
Flu and pneumonia In hospitalized patients testing positive for flu, chest X-rays often reveal pneumonia. Clinicians then need to determine whether it’s a case of influenza pneumonia or a secondary bacterial pneumonia.
That’s also very hard to tell, said Dr. Babik, because patients’ presentation with either is indistinguishable in terms of symptoms, labs and X-rays. Primary influenza pneumonia, which occurs in about 40% of patients hospitalized with the flu, is a severe illness. Twenty percent of these patients present with sepsis, while 10% present with septic shock and 50% need to be admitted to the ICU. Mortality runs around 20%.
Chest X-rays in the majority of patients with primary influenza pneumonia show consolidations and bilateral infiltrates, Dr. Babik explained, “but findings are unilateral in 30% or 40%. I’ve had to retrain my brain after medical school, where I learned that viral illness was always going to be diffuse and bilateral. Influenza can be a unilateral finding on chest X-ray.”
As for CT findings, influenza and other viral infections show three patterns: predominant ground glass opacities (GGO), GGO with consolidations and GGO with centilobular nodules.
Secondary bacterial pneumonia—which was likely responsible for most of the deaths in the 1918 flu pandemic—is less common. According to Dr. Babik, it affects about 10% of patients hospitalized with the flu, but probably 20% or 30% of those who are critically ill or who die.
A role for procalcitonin
The classic finding of secondary bacterial pneumonia, which was described in 1918, is that patients see a period of improvement before symptoms recur four to seven days later. But “many of these patients present around day 5 and have had no period of improvement,” Dr. Babik said. “Just because that’s not part of their history shouldn’t dissuade you from thinking of bacterial pneumonia.”
What does she rely on to distinguish viral vs. bacterial infection? It’s very challenging, she noted, but sometimes procalcitonin can help. Research published in the July 15, 2017, issue of Clinical Infectious Diseases looked at procalcitonin as a marker to help determine the etiology of community-acquired pneumonia.
“The procalcitonin cutoff they used—0.25—had a 70%-90% sensitivity and a negative predictive value of 80%-90% discriminating between viral and bacterial infections” in immunocompetent patients, Dr. Babik said. “That’s a useful adjunct, but it shouldn’t replace your clinical suspicion. A huge lobar pneumonia is less likely to be influenza, and I don’t care what the procalcitonin is in that situation.”
Clinical judgment should also guide when to give antibiotics. “If a patient is severely ill, give empiric antibiotics while you wait for more data.” Procalcitonin can help you make that determination in patients with mild to moderate illness. In addition, “a negative chest X-ray or minimal changes on a chest X-ray can make you more confident that it’s influenza pneumonia.”
Antivirals
As for treatment, the M2 inhibitors—amantadine, rimantadine—”target only influenza A and have widespread resistance, so we don’t use them,” Dr. Babik said. The drugs of choice, which are the neuraminidase inhibitors (oseltamivir, zanamivir and peramivir), target both influenza A and B.
The go-to drug, oral oseltamivir, has to be renally dosed, and it can produce side effects like nausea and vomiting. Studies indicate that the earlier inpatients are treated with oseltamivir, the more of a decrease you may get in terms of length of stay, mortality and the need for SNF care. “But you want to treat all inpatients with influenza regardless of the time of symptom onset,” Dr. Babik said. She typically treats for five days, which is the standard duration of therapy.
Because zanamivir is inhaled, she explained, “Nurses need special training to administer it, so we don’t use it as much.” Also, the drug can’t be used in intubated patients or those with underlying respiratory disease
Peramivir is administered through an IV, and it also needs to be renally dosed, although “it doesn’t really have any contraindications and is pretty well tolerated.” Dr. Babik typically reserves peramivir for patients for whom she’s concerned about oseltamivir absorption. While it’s approved as a single dose, her hospital guidelines suggest using it for up to five days.
New this year
The buzz this season is about baloxavir, which received fast-track FDA approval in October. In phase 3 trial results published in the Sept. 6, 2018, New England Journal of Medicine, the drug was tested against both oseltamivir and placebo.
“The decrease of symptoms with baloxavir was the same as with oseltamavir, but baloxavir had a more rapid decline in viral load,” Dr. Babik said. While physicians have no data on the drug in hospitalized patients, “I think it may be of use in oseltamivir resistance because it has a different mechanism of action.”
Oseltamivir resistance is something doctors should consider in patients who remain severely ill even after 10 days of treatment or in those who are immunocompromised. (Send resistance testing through your department of public health or the CDC.) While resistance is rare—showing up in less than 2% of isolates over the past two years—you’ll need to consider using IV zanamavir or, possibly, baloxavir to treat. (Oseltamivir resistance has cross-resistance with peramivir.)
And one question that constantly crops up during flu season, particularly among health care workers: Should you use antiviral prophylaxis if you’ve been exposed to someone with active flu? The CDC has weighed in on that, Dr. Babik said, and recommends considering prophylaxis for high-risk adults: those who are 65 or older; those with chronic medical or neurologic conditions; those who are immunocompromised, pregnant or post-partum within two weeks; Native Americans and Alaska natives; obese patients with a BMI of 40 or more; and residents of chronic care facilities.
And one question that constantly crops up during flu season, particularly among health care workers: Should you use antiviral prophylaxis if you’ve been exposed to someone with active flu? The CDC has weighed in on that, Dr. Babik said, and recommends considering prophylaxis for high-risk adults: those who are 65 or older; those with chronic medical or neurologic conditions; those who are immunocompromised, pregnant or post-partum within two weeks; Native Americans and Alaska natives; obese patients with a BMI of 40 or more; and residents of chronic care facilities.
“An alternative,” said Dr. Babik, “is really close monitoring and early therapy.” Prophylaxis is oseltamivir or zanamivir at half the therapeutic dose.
Phyllis Maguire is Executive Editor of Today’s Hospitalist.
Other viruses making the rounds
WHILE INFLUENZA always grabs the headlines, Jennifer Babik, MD, PhD, an infectious diseases physician and associate clinical professor at the University of California, San Francisco (UCSF), explained at a recent UCSF conference that several other respiratory viral infections—particularly respiratory syncytial virus (RSV) and human metapneumovirus—also make the rounds during the winter. Adenovirus and rhinovirus, on the other hand, are year-round culprits, while parainfluenza 3 is more prominent in the spring and summer.
• RSV: While “you absolutely cannot hang your hat on this,” Dr. Babik said, patients with RSV often have more wheezing than patients with influenza. Most patients have upper respiratory infection, and mortality in the elderly can approach 10%. Treatment for most patients is supportive care.
RSV is a huge problem in immunocompromised patients. Among bone marrow transplant patients, for instance, “simple upper respiratory infection can progress to lower tract disease.” Mortality rates can run as high as 80%; solid organ transplant patients “do a bit better,” but mortality can still be 20%. In these patients, ribavirin or ribavirin plus IVIG can decrease disease progression and mortality.
• Parainfluenza: Parainfluenza 3 is most common in adults and can be severe in the immunocompromised. “No treatment is clearly effective, and there is no benefit of ribavirin in bone marrow transplant patients.
• Human metapneumovirus: As many as 70% of patients with this virus are asymptomatic, with treatment again being supportive when symptoms occur.
• Adenovirus: This virus can cause pneumonia in the immunocompromised, but rarely in immunocompetent patients. “The classical features we think about—pharyngitis, conjunctivitis, rash—may be absent,” Dr. Babik pointed out. Many respiratory multiplex PCRs aren’t as good for adenovirus as for other viruses because there are so many subtypes. “If you have a high suspicion, you want to send a serum PCR because that’s more sensitive.” In immunocompromised patients, you can consider cidofovir. But, Dr Babik cautioned, “that’s very renally toxic so we don’t often do that.”
• Rhinovirus: The cause behind the common cold, rhinovirus is often found with both community- and hospital-acquired pneumonia. “It probably can predispose you to bacterial infections like all the other respiratory viruses, but it’s not really clear that it’s a primary pathogen,” Dr. Babik said. When she has a patient with respiratory illness, particularly if it’s severe, “I’m very hesitant to pin it on rhinovirus alone.”
• Swine flu: This is another virus to consider for those in the Midwest. “There are usually between 0 and 20 cases a year,” Dr. Babik pointed out. “It’s limited to person-to-person spread, and treatment is actually similar to that of seasonal influenza.”
