Do You Need Antibiotics?

Follow on

One hundred fifty million prescriptions for antibiotics are written each year in the United States. By some estimates, one third of them are unnecessary. One of the most common diagnoses for which antibiotics are inappropriately prescribed is upper respiratory tract infections (URIs). The overwhelming majority of these infections are viral—infections for which we have no treatment that speeds resolution of symptoms (with one possible non-antibiotic exception, discussed below).

Why are antibiotics so commonly overused? First, because patients so often want and expect them. Bacterial infections seem to be something almost everyone considers himself to be able to diagnose accurately. Second, people don’t realize that distinguishing between bacterial and viral infections is important, not realizing that antibiotics don’t work against the latter. Third, patients (and doctors) think antibiotics are harmless. This is, unfortunately, sometimes not true (as with all medical interventions). As one of my infectious disease specialist colleagues often says, “The dirty little secret of infectious diseases is that antibiotics often make you feel lousy.” Fourth, unless your doctor happens to be a direct primary care doctor, doctors often don’t have enough time to explain why antibiotics aren’t necessary. Occasionally, antibiotics even cause disease (sometimes serious disease like clostridium difficile colitis, a malady I myself once had the displeasure of contracting). Antibiotic use also breeds antibiotic resistance. And though any one course of antibiotics isn’t itself likely to breed a bacterial superorganism, repeated exposure ratchets up the risk. Finally, we’re just beginning to understand the complex biomass of bacteria, the microbiome, that inhabit our large intestines (the number of such bacteria exceeds the number of cells in our bodies) and the beneficial functions they may perform for us. Some have theorized that changes in the microbiome may give rise, or at least contribute to, a number of diseases (perturbations in the microbiome may even contribute to obesity). Though it’s far too early to draw any conclusions about this, it’s not too early to remain disinclined to disturb the microbiome unless we expect a great benefit from doing so (i.e., the speedy resolution of a bacterial infection that would otherwise not resolve).

What clues do clinicians use to figure out when antibiotics are necessary in URIs? A number of things help us sort out the probability that a URI is bacterial rather than viral:

    1. The infection involves only one body cavity. Viruses love to leap around, often beginning in the sinuses only then to leap into the throat and ear and then down into the chest. Bacterial infections, with rare exceptions, start in one place and stay there.
    2. Both viruses and bacteria can cause fevers. In viral infections, the fever tends to be lower (100-101ºF), however, and last only the first few days of the infection (one exception is influenza where the fevers are higher, 101-104ºF, and last 3-5 days). Bacterial infections, in contrast, cause slightly higher fevers that persist.
    3. Both viral and bacterial URIs can produce colored mucus, which only represents white blood cells that have fought valiantly on your behalf and died. The color of mucus does nothing to differentiate between the two types of infections.
    4. In general, bacterial infections are more severe than viral infections. Though you may feel terrible when you have a viral infection, you rarely feel so bad you can’t get out of bed after the first few days. People with bacterial infections just look sicker than people with viral infections (one notable exception is strep throat, which can present as mildly as viral pharyngitis).
    5. Viruses improve on their own. Many patients get an idea in their minds about how long their infection should last, and when it lasts longer, decide they need to be seen by a doctor because “something might be seriously wrong,” as numerous patients have explained to me over the years. But the average length of time viral symptoms last is three weeks. The take home point, however, is that if you start to improve on your own, you almost certainly have a virus.

Sometimes, however, even with the help of these guidelines, it’s difficult to tell if a URI is bacterial or viral. Some viruses can mimic bacterial infections and some bacterial infections can mimic viral ones. An example of the former would be influenza. An example of the latter would be mycoplasma. Mycoplasma is the smallest living bacteria that infects humans and like viruses can infect more than one body cavity at a time (sinuses and throat and ears and chest). But it has a stereotypical set of symptoms and clinical course as well as a long incubation period (meaning that unlike most viruses that cause us to become symptomatic generally within 3-5 days of exposure, we develop symptoms from mycoplasma 1-2 weeks after coming in contact with someone who has it). But as testing for it is better at identifying past infections than current ones, the diagnosis is usually made on clinical grounds. So if someone complains of a sore throat that’s every bit as painful on day 7 as it was on day 1, I often treat presumptively for mycoplasma with antibiotics (it will get better without them, but if they aren’t given, it tends to recur).

Finally, though in the 21st century we still don’t have a cure for the common cold (a viral URI), some studies do show that adults who begin taking zinc gluconate lozenges within the first 24 hours of developing viral symptoms will often shorten the duration and intensity of those symptoms. So I tell people to keep a bag of them around the house and to start taking one every two to three hours while awake at the first sign of symptoms and to continue like that for 5 days. It causes mild nausea in some people, and makes just about everything—including water—taste terrible. But it seems to work. And prevents patients from asking for antibiotics when they really don’t need them.

Sign up to get notified when a new blog post has been published.