Medication, Prescribing, and Timing

There is an important variable that arises in pharmacology that is rarely addressed: after the choice of an antibiotic, how often a day should it be given, what time of day should it be given, and how do doctors compensate for the fact most patients stop antibiotics before the prescribed time period is up?

Sometimes we just don't know, sometimes we forget that patients stop almost all Rx's too early, and sometimes published clinical results are not followed because they are not in the PDR. One concrete example is malaria prophylaxis. Many of the drugs have to be continued for one to four weeks after the patient leaves the malarious area, and many patients forget about the fourth week. My practice is to have the pharmacist label the bottle: "Take.........until the bottle is empty", and then I emphasize to the patient the importance of following this advice.

If you prescribe antibiotics to be taken three times a day, you are fooling yourself, because most patients forget the midday dose, and only in the hospital can you depend on this schedule being followed (by the floor nurse, not the patient). Just ask your residents the next time you see them how well they complied the last time they were prescribed antibiotics. That is one reason I rather prescribe Augmentin, which can be overkill, rather than Amoxicillin, because the former is 2x/day, the the latter 3x/day. For the same reason I prefer Ceftin at b.i.d. to Keflex, especially when treating abscesses, where constant penetration at a bactericidal level is critical and, occasionally, once a day Levaquin to twice a day Cipro, even though the latter is available as a generic.

A classic study was done at NY Presbyterian's pediatric clinic about 40 years ago. Mothers were prescribed the classic bottle of liquid (yellow) penicillin suspension, with the instructions to give their child 1 tsp. four times a day, and warned about the strep throat/rheumatic fever connection. When the mothers returned with their bottles in 10 days, only 10% of them had empty bottles. Assuming that none emptied the bottles that AM out of embarrassment, this means that at least 90% of the most vitally interested persons did not follow the instructions. Add to this the belief of most patients that the less medicine they take the less sick they are helps us to understand why intelligent drug companies prescribe an antibiotic for longer than is necessary (e.g. 3 days of Zithromax is generally equipotent to 5 days, because of its persistence in the necessary tissues and cells) in order to be sure of observing a positive effect.

We all are aware that as the blood sugar climbs above 200, blood becomes more of a non-Newtonian fluid with an atypical cross-sectional velocity profile, and an enlarged viscosity-dependent boundary layer, so that Schlichting's classical theory does not apply. Two well-known results from this are (a) reduced diapedesis of white blood cells, thereby lessening resistance to containing infection, and (b) a non-linear increase in the pressure required to drive blood through the circulatory system, thereby increasing the amount of heart failure. But does the level of blood sugar affect the bactericidal effect of antibiotics? This question is not answered for most antibiotics, which is unfortunate, because diabetics have difficulty controlling the fasting and the post-prandial glucose level equally well, and that should therefore determine the optimum time of the day to deliver a once daily or twice daily antibiotic.

As far as I am aware, although some anti-cancer drugs are tested AM vs. PM to look for a diurnal variation effect, no antibiotic has been. We do know that for some antibiotics the peak level is important, for others the trough, and for still others the area under the dose-time curve, but it is difficult to translate these results into concepts the patient can understand. And the question of with food or on an empty stomach is often answered theoretically rather than clinically. For instance, does it really matter if Synthroid is taken before or after breakfast so long as the patient is consistent by taking the medicine at the same time each day, to reduce the variation of the blood free T4 level? Should a daily antibiotic be given in the AM, when the serum cortisol is at a peak, or in the PM, when the serum iron is at its peak? One could make a case for AM dosing, because one way the body apparently fights infection is to lower the serum iron, but I am unaware of any studies on this possible effect.

Another important point is to make sure certain drugs are dispensed only in tablet rather than capsule form. This is especially important when prescribing Doxycycline, since if a capsule with a pH of 12 gets stuck in the esophagus and dissolves there, the resulting chemical burn can take two to three weeks to heal. I also routinely give skiers the sunburn warning, because if snow blindness is of concern, then so is sunburn, especially on the lips.

The problem of generics also should be addressed (and I realize that I am drifting from my main topic). The generic and the brand name drug are usually compounded differently, with different binders and dyes to make up the tablet. I have seen several cases where the allergic reaction of the patient was due to the binders or the drug-carrying vehicle, and not to the parent antibiotic or drug, e.g. in a reaction to an IM steroid given to treat an acute allergic reaction.

We still do not know the optimum length of time to treat acute sinusitis or acute otitis media or acute cystitis or acute prostatitis. ENT physicians seem to add a Medrol Dospak to any infection they treat above the neck, and we internists do not, but I am not aware of any definitive comparative studies on this subject, nor does it appear in the Cochrane Report. We don't know if we should treat sub-acute Lyme disease for 14 or 21 days, but if you are concerned over malpractice, then 21 days always wins out.

And despite our encouragement NOT to treat viral infections, and the awareness of the public of the problem of drug resistance, I can tell you that if my wife gets a URI and she is not treated, the number of friends who tell her to see another doctor (not myself in either case) so she can get "put on an antibiotic and be properly treated" is astronomical.

Let me close with a related anecdotal story. When I was an intern in NYC, I admitted a male alcoholic with pneumonia one freezing December PM. We put him straightaway into the ICU because of trilobar pneumonia (in those days the admitting intern also covered his patients in the ICU with the help of the ICU resident, not the ward resident). The patient died within 12 hours of admission, and I filled out the death certificate: "Cause of Death...Acute Pneumonia". The supervising ward manager called me down to his office, telling me that in NYC if the death certificate did not specify either viral or bacterial pneumonia, then it became a ME case due to health law, and the body could not be released to the family for burial until an autopsy was performed to determine if a virus or a bacterium was responsible for the death. He asked me to insert an adjective for the family's sake, and I did. So much for the accuracy of death certificates---patients and their families can and should take priority over accurate facts under certain conditions.