Antibiotic Gumbo

In my bwain

“There’s a little chef… in your bwain?” – Colette, Ratatouille

We’re half-way through Medical Microbiology and finished with most medically-important bacterial and fungal species; within another few weeks, I’ll also be much more well-versed in viral and parasitic diseases and the recommended treatments. Leading up to this Friday’s exam, though, I am definitely feeling both excited and challenged by the complicated nature of treating infectious diseases, particularly in the context of drug resistance and adverse effects. It seems appropriate that our introduction to infectious diseases should come so early during the second year: here, we can readily see the threat of lethal infections, the pressing need for rapid identification of the disease-causing agents, the profound benefits of proper and timely treatment, and the disastrous consequences of picking the wrong treatment.

I sometimes wonder how computers will come to aid medicine, and as time goes on, I realize just how complex a field it can be. Although a proponent of technological advancements, I would agree with our Chief of Medicine: the human brain is the only processor capable of properly diagnosing and treating diseases. Computerized technology can aid rapid access to information in a constantly growing and changing field, and it can help incorporate safety measures (e.g. standardized and easy-to-read electronic medical records, drug interaction warnings, when properly implemented). However, the thought process is a uniquely human ability, and will be even when the Cylons take over.


The pathogenesis of infectious diseases (and others) and their treatment involve much more than matching a drug to a collection of words describing symptoms. Symptoms themselves involve multiple dimensions. Take for example, the seemingly ambiguous symptom of pain: Where is the pain? How would you describe the pain? How bad is it, on a scale of 1-10? (or 1-100?) When did the pain start? How long did the pain last? How frequently have you felt this pain? Have you felt this pain previously during a separate occasion? If so, when? Tell me how it started. Does anything exacerbate or relieve the pain? Is the pain getting worse or getting better? Has the pain moved anywhere else? Is there anything else bothering you? (Anything that might be related to the pain?)

[These questions follow my system for evaluating a patient’s chief complaint. I modified the FARCOLDER system taught to us by our Chief of Medicine with a new order and a cute-overload-worthy acronym: “Little Cotton Tail Picks Peter Rabbit Always”

L – Location
C – Character
T – Timing (Duration, Onset, Frequency)
P – Precipitating Factors (Exacerbating, Relieving)
P – Progression
R – Radiation
A – Associated Symptoms]

Now, take some common symptoms of an infection: fever, malaise, chills, myalgia (muscle pain). What does this tell you? Not much, with respect to developing a differential diagnosis. One thing that many people forget (including doctors) is that the best, cheapest, most accessible information you can use to diagnose and treat disease comes directly from the patient. The key, then, is finding out how to get the pertinent information by asking the right questions. Tests can be very precise, but they are not useful if you don’t first come up with a proper system of thinking about a case: you need an accurate differential diagnosis first. In other words, even if you have a shotgun, you need to first pick the right direction to shoot in. Computer systems can be very good at providing exhaustive lists, but this is useless for sick and tired patients. Unlike computer systems, the human mind is flexible and immediately adaptive: the direction of the discussion can rapidly change, and questions can be precisely targeted and adjusted on a real-time basis. So let’s say this patient has a cough. What now? Is it the flu? Is it a cold? Is it pneumonia? If it’s pneumonia, a life-threatening disease, is it caused by Streptococcus pneumoniae, Legionella pneumophila, Staphylococcus aureus, or Mycoplasma pneumoniae? Or could it be Haemophilus influenzae, Bordetella pertussis, or Leptospira interrogans? Perhaps it’s not actually a lower respiratory infection (the lungs), but rather, a cough caused by an irritated and swollen throat: pharyngitis (upper respiratory infection). Should we then consider Streptococcus pyogenes (Strep throat), Corynebacterium diptheriae, or Neisseria gonorrhoeae? And then there’s meningitis, bacteremia, and genitourinary infections, all of which have many potential disease-causing agents.

But don’t forget you need to treat the patient, too! Let’s say it’s a bacterial infection, since that’s what I know. Does your patient have allergies to penicillin (as do 10% of all patients hospitalized with infections)? If so, all the pencillins (penicillin G, penicillin V, nafcillin, methicillin, oxacillin, cloxacillin, dicloxacillin, ampicillin, amoxicillin, carbenicillin, ticarcillin, mezlocillin, piperacillin) are out, and so are the cephalosporins (cephalexin, cefadroxil, cefazolin, cefaclor, cefamandole, cefotetan, cefoxitin, cefuroxime, cefuroxime axetil, ceftizoxime, cefotazime, ceftriaxone, cefdinir, ceftazidime, cefixime, cefoperazone, cefepime, etc.) and carbapenems which have cross-allergies (severe, anaphylactic reactions) with the penicillins in between 1-10% of cases. Lots of docs are using fluoroquinolones (ciprofloxacin, levofloxacin, etc.) these days for many bacterial infections, but hell will break loose if you prescribe these drugs, sulfonamides, tetracyclines, or metronidazole for a pregnant woman, or for some of them, babies and young kids. Linezolid, Synercid, and daptomycin are great if you have a MRSA (methicillin-resistant Staphylococcus aureus) infection, but you have to have good reason to believe that MRSA is your culprit because these are the last secret weapons against multi-drug resistant bugs – don’t use them with non-multi-drug-resistant bacteria when there are other things that can kill them! Then you have to think about how the drugs are absorbed: will they be absorbed well through the GI tract, or do you need to administer them intravenously, intrathecally, or in some other fashion? Your meningitis patient is not going to be happy (i.e. dead) if you treat his unidentified, bacterial meningitis with your broad-spectrum doxycycline (a tetracycline drug): it doesn’t reach the area of infection, so you should go with ceftriaxone (a cephalosporin drug) instead. But don’t forget to watch out for potentially life-threatening Clostridium difficile infections that arise from the actions of cephalosporin drugs (C. difficile lives in the colon, and cephalosporins kill many of the other normal bacteria, leaving C. difficile to go wild). Oh yeah, and antibiotics also cost money, so you have to see whether the patient will be able to pay for the treatment. And you have to adjust maintenance doses for patients with renal insufficiency, and account for drug interactions with other common drugs like warfarin, alcohol, ketoconazole, or even something as simple as grapefruit juice (diet, anyone?).

And this is me, as a second-year medical student with an incomplete knowledge of medicine, trying to grasp the complexity of treating infectious diseases when there are undoubtedly dozens of other possibilities and many other factors I haven’t listed here or considered.

One complaint that I’ve often heard about drug interaction databases with EMRs is that every drug you prescribe has contraindications with the other medications a given patient is taking. Then the question becomes, what will the side effects be? How serious will they be? Is it simply a matter of adjusting the doses to prevent adverse effects? At this time, many computer systems are made by programmers with negligible medical knowledge that design systems that are incompatible with the way physicians and nurses work (most often because the systems are too rigid, and are “stupid” and “wrong” in their adaptive methods). On the other hand, there are few immediate incentives for physicians and nurses to work on computerized tools to aid the process of diagnosis and treatment: they have too much on their hands already, immediate problems to solve, and no readily visible financial benefits. Given the lack of a viable alternative, I’m amazed how little respect many Americans have for their physicians and nurses who do much more than “push pills.” Or perhaps we just don’t know what complexity goes into “pushing pills.”

“We’re best friends just like… amoxicillin and clavulanic acid.” – From My Musical, Scrubs

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: