Peak or Trough?

We were recently consulted on a patient with a history of repeated admissions with pneumonia who had developed AKI following treatment with Tobramycin. It got me thinking about the mechanisms of aminoglycoside toxicity. The traditional teaching is that aminoglycosides are best administered in once daily doses in order to reduce toxicity but given the fact that this should lead to higher peak levels, it seems counterintuitive that this might prevent nephrotoxicity. However, this makes sense when you understand how aminoglycosides damage the kidneys.

Aminoglycosides are freely filtered at the glomerulus. 90% of the drug is excreted unchanged in the urine while the rest is reabsorbed in the proximal tubule. As Nate mentioned in a previous post, they bind to phospholipids in the cell membranes in the S1 and S2 segments of the proximal tubule. During ischemic episodes, there is some uptake in the S3 segment also. They are subsequently endocytosed via the transmembrane protein, megalin, and accumulate in the cytosol where they mediate their toxicity. The degree of renal damage associated with the various aminoglycosides is related to the ability of the drug to bind to these phospholipids. The more binding that occurs, the more nephrotoxicity. Gentamicin is the most toxic, followed by tobramycin, amikacin and streptomycin. The reason why once daily dosing is better is that this transport mechanism is saturatable and beyond a certain concentration, no more drug is taken up. Thus, peak levels correlate with bactericidal ability while trough levels correlate with renal (and oto-) toxicity.

Interestingly, this is similar to the mechanism of renal toxicity associated with imipenem. There have been efforts to develop a drug that could be co-administered with aminoglycosides that would inhibit their transport in the proximal tubule and thus prevent their nephrotoxic effects, but unfortunately none have been successful.

Gentamicin is made up of four components and when these are administered separately, it turns out that one of them, C2, does not appear to be absorbed as readily into the proximal tubular cells while still retaining the bactericidal ability of the whole drug. This may have potential for reducing nephrotocity in patients who are particularly at risk – i.e. pre-existing AKI or ischemia.