Posted by: notdeaddinosaur | January 5, 2012

Antibiotics; Facts and Fictions

Antibiotics are wonderful drugs. Penicillin, in fact, was the original “wonder drug”. Eighty years later, though, they’re not quite as wondrous as they used to be, due in large part to the fact that too many patients (and doctors!) don’t understand how to use them correctly.

Here is what antibiotics do: they kill bacteria. (Actually, some of them just stop bacteria from growing, which ends up doing the same thing, since the body will then get rid of the non-growing bacteria on its own. But I digress.) That’s basically all. They have to be absorbed and eliminated, which means some of them can have adverse effects on the GI tract (going in) and the liver and kidneys (going out). Sometimes, like any other foreign substance, they can elicit allergic reactions. Other bad things associated with antibiotics are really the result of them doing what they’re supposed to do, even when we don’t want them to, ie, killing off “good” bacteria normally present in assorted areas of the body. There are a couple of other odd antibiotic toxicities to be aware of (deafness; tooth staining) but killing bacteria is pretty much all they do.

Different antibiotics work differently. Some damage cell walls; some interfere with bacterial protein synthesis; some damage the bacteria’s genetic material. There are many different kinds of bacteria. Some have thick cell walls, which help keep antibiotics out (except for the ones that destroy cell walls); others have different ways of combating antibiotics. The point is that not all antibiotics kill all bacteria. So in order for antibiotics to work properly, you have to:

  1. Know what kind of bacteria you’re trying to kill, and
  2. Use an antibiotic that will kill that bacteria.

How do we know what bacteria we’re trying to kill? Bacteria are very small. We can identify them, but only if we have a lot of them to work with.  So what we do is take a sample of the infected material and incubate it under whatever conditions it takes for the bacteria to grow, then examine the colonies to figure out what kind of bacteria was in the original sample. This is called a culture. Once we have a lot of the bacteria around, the other thing we can do is divide them up into several little groups and subject each group to a different antibiotic. This way we can actually see in the lab that a certain antibiotic kills the bacteria from the patient. This is known as bacterial sensitivity testing.

Therefore ideally, every infection would be treated by taking a culture, identifying the bacteria, and testing it against different antibiotics so the patient can be treated with exactly the right one. It’s a little more complicated, of course, because different antibiotics penetrated different parts of the body to different degrees. There are also various routes; some antibiotics can’t be given orally; others can be given any which way (intravenously, jabbed into a muscle to be absorbed into the bloodstream, even injected directly into the spinal fluid surrounding the brain). However it goes in, though, it has to be able to get to the bacteria, and then kill it.

What if the patient is too sick to wait until we’ve gone through all that rigamarole to begin treatment? In that case, we look carefully at the likely source of the infection (lung/pneumonia, gut/diverticulitis, brain/meningitis, etc), weigh other patient-specific factors (age; travel; recent hospitalization or institutionalization; is the patient’s immune system intact?) plus other random circumstances (time of year; geography; known bacteria causing other acute infections in the community) and come up with an educated guess about what bacteria is most likely to be causing the problem. We call this empiric therapy. Note that after beginning antibiotics empirically, it’s still important to obtain a culture in order to confirm that the guess was indeed correct, and that the chosen antibiotic is indeed capable of killing the observed bacteria, especially if the patient is very sick.

Empiric therapy is often used without obtaining a culture, at least in primary care. There’s nothing wrong with this in principle, but you still have to know what you’re doing: using antibiotics to kill bacteria.

What about “preventing” infections with antibiotics? No such thing. If there are no bacteria, then there’s nothing (good) for the antibiotic to do. The use of “prophylactic” or “preventive” antibiotics technically applies to situations where there may be some bacteria involved (as in surgery, or  a contaminated wound), and if there’s an appropriate antibiotic around to kill those first few, they won’t grow into a full-fledged infection. This is one of the most misunderstood aspects of antibiotic management. If I had a nickel for every patient who insisted on having an antibiotic “so my cold won’t go into pneumonia”, I’d be rich beyond the dreams of avarice. There are very specific guidelines for prophylactic antibiotics; not that they’re followed as well as they should be. Sometimes it seems like the favorite words of every ER doc and pediatrician are, “Just in case.”

So the correct way to treat bacterial infections when you can get a sample of infected material (urine, in the case of a bladder infection, for example) is to send the culture and begin treatment with an antibiotic known to kill bacteria commonly causing the infection. If the culture shows that the bacteria happens to be resistant to whatever antibiotic you chose, it should be stopped (since it’s not killing the bacteria that are there) and the patient switched to one of the antibiotics that do kill their particular bacteria, according to the sensitivity report. This is why the commonly heard sentiment, “Isn’t it dangerous to stop an antibiotics before finishing the course?” makes no sense at all. It’s important to take an antibiotic long enough to kill all the bacteria causing the infection. After that, it’s useless.

What about when you can’t get a sample of infected material? Or, more likely, when obtaining the sample is far more involved, invasive, or expensive than the condition warrants. That’s when you have to go with your best guess, taking into account as many factors about the patient and the disease (the who, where, what, and when) as you can. References like the Sanford guide are invaluable in these cases.

What if a patient doesn’t get better with antibiotic treatment? There are several possibilities.

Sometimes the doctor has chosen the wrong antibiotic (one that isn’t effective against the patient’s bacteria.) Sometimes the patient’s bacteria is resistant to the antibiotic (even if other versions of the same bacteria are susceptible to it.) Note that these two things are not the same. Sometimes the dose wasn’t high enough. Sometimes the antibiotic can’t get to the bacteria (say, if it’s destroyed in the stomach and should have been given by injection instead of by mouth.) Sometimes the patient never took the antibiotic. It may have been too expensive, or the pills were too big, or they were too scared of the potential side effects.

Most frequently, thought, the patient didn’t have a bacterial infection (often not an infection of any kind at all). Why is it so hard for people to understand that you need an accurate diagnosis if you’re going to expect any kind of treatment to make you better? Viral upper and lower respiratory infections (colds, sinus infections, bronchitis) are almost always caused by viruses, not bacteria. Green mucus is caused by myeloperoxidases in white blood cells, not bacteria. It means nothing! Unnecessary antibiotics represent billions of dollars of wasted healthcare money every year.

Aside from the money, though, what’s the big deal about unnecessary antibiotics? “Isn’t it better to be safe than sorry?” I hear patients say all the time, to my eternal discouragement. “What’s the harm?” they ask.

There’s plenty of harm to be had from antibiotics, even when used correctly. Aside from nausea, vomiting, diarrhea, and other assorted direct adverse effects, the major harms are from allergic reactions, from killing off the body’s normal bacteria, and the emergence of resistant bacteria (that is, bacteria that cannot be killed by antibiotics).

Up until a few years ago, the phrase “resistant bacteria” made the eyes of everyone who wasn’t a doctor glaze right over. Now all you have to say is MRSA. Methicillin (or multiply) resistant staph aureus, also incorrectly called the flesh-eating bacteria is all over the news. I admit it’s gotten a tiny bit easier to talk patients out of unnecessary antibiotics by saying, “This causes MRSA, you know.” Resistant bacteria are a real danger because eventually a bacteria will emerge that cannot be killed by anything. That’s scary.

True allergic reactions to antibiotics are relatively rare. Still, they can be catastrophic. You can die from an allergic reaction to an antibiotic. If you have a life-threatening bacterial infection, that may seem like a reasonable risk to take. When it’s not actually necessary, not so much.

Perceived “allergic reactions” are actually more of a headache than real ones. Hives, throat swelling, and dropping your blood pressure to the point of collapse are signs of potentially dangerous allergic reactions. “I don’t know. Something happened when I was a baby and my mother said I was allergic to penicillin,” is the usual story behind far too many antibiotic “allergies”. According to UpToDate, as many as 85-90% of patients stating they were allergic to penicillin did not show true allergy upon appropriate testing.

Then there’s the problem of “cross-reactivity” between penicillin and another class of antibiotics called cephalosporins. Although the figure thrown around from med school onward is that 10% of people who are allergic to penicillin will also be allergic to cephalosporins, actual research shows it to be significantly less. In fact, only 2% of patients confirmed penicillin allergic by skin testing will actually react to cephalosporins.

What happens in real life, though? Anyone who says they’re allergic to penicillin (90% chance they’re not) is also immediately removed from consideration for treatment with cephalosporins, which just happen to be the drug of choice for most small minor skin infections, even though there’s only about a 2% chance of a problem. Putting those numbers together means that someone who says they’re allergic to penicillin has only a 2 in 1000 chance of reacting to a cephalosporin. So instead of a cheap, effective antibiotic, most of them are given clindamycin, a drug notorious for causing antibiotic-associated colitis (by killing off good bacteria in the lower bowel and letting dangerous bacteria proliferate unchecked), which can be fatal.

To sum up:

  • Antibiotics kill bacteria.
  • If you don’t have a bacterial infection, you should not be taking antibiotics.
  • Appropriate antibiotic therapy consists of the right drug for the right reason at the right dose for the right time/duration, and no more.


  1. Great overview of antibiotics — basically touched on everything we discuss in our medical school.

  2. AMEN.

    Only thing to add is that TMP/SMX is a great end run around those ‘penicillin allergies’ and covers MRSA very well and and strep reasonably well for superficial skin infections/abscesses in the outpatient setting. In fact, Sanford recommends BactrimDS 2 PO BID for large or multiple abscesses not amenable to I&D(a great way to avoid antibiotics entirely) as first line. Doxy also covers both reasonably well and both have less C.Dif rates than clinda. YMMV.

  3. I am printing this and taking it to the idiot ER doc that went nearly postal on us because we refused to put our 9 month old on two different antibiotics *at the same time* after he had a febrile seizure. “Just in case” the red ear drums were a bacterial infection, and despite the family history of serious allergic reaction to several classes of antibiotics. (Yes, true reactions. Full body hives, for me it’s -cillians and the sulfas.) He had roseola by the way, so the antibiotics would have done nothing but make him even sicker with the side effects.

    If you take post requests, I would love for you to weigh in on the use of prophylactic antibiotics in laboring women who test positive for Group B Strep at 36 weeks gestation. I’ve run the statistics and have an opinion, but would love to hear your thoughts.

  4. I wish that every part of the health care system was as conscientious about this as you are. In the last 10 years I have had antibiotics prescribed for strep throat, UI, and prostatitis. During that same period of time, my wife, who has an autoimmune disease so that even minor infections become big deals, has been prescribed antibiotics for bronchitis and sinus infections on multiple occasions. In no case was a culture taken, but if the original prescription didn’t do the trick we were just given another prescription for a different drug. This is in a metropolitan area of over half a million people and we have each changed our primary care providers several times because of this and other issues, but they all seem to operate with the “treat ’em and street ’em” philosophy.

  5. Great overview. The following article, albeit from halfway around the world where access to care may be scarce, really drives the point home about antibiotic resistance

  6. Good article although you should be aware of the many natural antibiotics that are available and their health benefits. Natural antibiotics do not have any long term health risks or cause damage to your liver. Antibiotic resistance is become more widespread, the “better safe than sorry” crowd will eventually be sorry.

  7. Great overview of antibiotics – largely touched in everything we discuss in our very own health class.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: