"Kill characteristics" is a made-up sounding term which describes the effect of dosing intervals on the antimicrobial effectiveness of antibiotics. It relates mainly to the post-antibiotic effects of the specific drug. The best way to describe kill characteristics is with the use of this diagram, which had come up in Question 3.1 from the second paper of 2014, and Question 24.1 from the first paper of 2008.
The graph represents a concentration-time curve, and describes the relationship between antibiotic concentration and their "killing power", the capacity to wipe out a significant population of microorganisms. The question is based on Table 72.2 from Oh's Manual, "Pharmacodynamic properties of selected antibiotics". It can be found on page 739 of the new edition. The table, in turn, is based on a 2009 article by Roberts and Lipman. The article even has a graph of concentration over time, which resembles the graph in the question. It was published one year after this college paper. In it, an excellent section entitled "Kill characteristics of antibiotics" describes the below concepts very well. This article, in turn, draws on material from an even earlier article from 2003 by William A. Craig, which is even more detailed. Generally speaking, William A. Graig seems to be the guru of antibiotic pharmacodynamics, and has at least one other excellent paper relevant to this topic.
At a basic level, the activity of antibiotics is described by three statements, according to the most important factor in their pharmacokinetics:
How do we decide which is which? It is surprisingly unscientific, much in the same way the way we define what an infection is, or what sepsis is. Most studies that explore this question tend to choose some relatively arbitrary clinical endpoint, eg. the resolution of fevers, negative blood cultures, improved haemodynamics, etc. The investigators then look at the pattern of drug concentration over time, and try to determine which parameter in this graph was most closely associated with improvement. This can occasionally produce some confusion. For example, some studies (Morrissey, 1997) will describe fluoroquinolones as concentration-dependent killers, whereas others (Schentag et al, 1991) found that peak/MIC or AUC/MIC were better predictors of a good clinical outcome.
This refers to the time spent marinading in a concentration over MIC - the "C" on the college graph.
The antibiotics most affected by this are:
Examples of this include β-lactams, carbapenems, monobactams, linezolid, linocosamides like clindamycin, and macrolides like erythromycin.
The concentration of the antibiotic does not need to remain over MIC for 100% of the time in order to be effective. At least with cephalosporins, it has been demonstrated that if even 40-50% of the dosing interval is spent at above MIC, the killing efficacy of the drug approaches maximum.
Specifically, the antibiotics requiring the longest time above MIC were the cephalosporins.
β-lactams required less time over MIC, and carbapenems even less. The key isssue there is the rate of killing - carbapenems seem to kill the fastest, and much of the killing occurs early in the dosing interval (thus, only 30% or so of the dosing interval needs to be spent at over-MIC concentrations).
For these antibiotics, the required time spent over MIC increases if the bacteria are in a site to which antibiotic penetration is limited. The increased time requirement probably reflects the fact that antibiotic concentration-time curves are usually measured in serum, and the concentration achieved within the infected site is probably much lower. This has been demonstrated in the scenario of osteomyelitis.
Pharmacokinetics do occasionally interfere with the kill characteristics, making some drugs appear as if they perform concentration-dependent killing, whereas in fact they are purely time dependent. This is the case with azithromycin, which can have an extremely long dosing interval. This is a mirage: the persistent effect of azithromycin is extremely prolonged because of its sequestration in macrophages, i.e. it is not a post-antibiotic effect.
This refers to the highest concentration reached.
Concentration dependent killing is a property of antibiotics which disable some sort of crucial step in bacterial metabolism or protein synthesis. The higher the concentration reached, the more synthetic enzyme molecules are inhibited. Examples include aminoglycosides, metronidazole, daptomycin, and to a lesser extent fluoroquinolones.
As an example of pharmacokinetics, for aminoglycosides to obtain a clinical response of around 90%, the peak level needs to exceed the MIC by eight to ten times.
There is an interesting idiosyncratic feature of aminoglycoside therapy here. Aminoglycosides killing is initially related to passive ionic binding of the drug to the bacterial lipopolysaccharide coats, but later it becomes more reliant on active uptake of the drug into the bacterial cell. Being exposed to aminoglycosides causes bacteria to down-regulate this uptake, and thus the first exposure to the drug increases the subsequent MIC. That first dose better be a big one; you are relying on the high peak of concentration to carry out the bulk of the genocide. Thereafter, resistant survivors will respond poorly. It is thought that once-daily dosing of aminoglycosides allows enough time for this effect to disspiate between doses.
Time and concentration dependent killing
This refers to antibiotics which rely on both time and concentration to kill bacteria - it is a property of those drugs which inhibit steps in DNA synthesis or replication, or other bacterial components which are crucial to cellular division.
Time is important because the inhibited enzymes are most active during cell division, and one must spend some time immersed in the drug in order to catch a large enough proportion of the bacteria at the point of replication. Concentration is important because higher concentrations disable more of the target cell components.
Examples include fluoroquinolones, azithromycin, tetracyclines, glycopeptides like vancomycin and teicoplanin, tigecycline, quinupristin/dalfopristin. and to some extent linezolid.
For instance, fluoroquinolone concentrations in serum need to average about four times the MIC for each 24 hours to produce virtually 100% survival in a variety of experimental animal infections.