International Journal of Antimicrobial Agents 17 (2001) 443 – 450 Commentary Classification of oral cephalosporins. A matter for debate J.D. Williams *, K.G. Naber, A. Bryskier, N. Høiby, I.M. Gould, P. Periti, H. Giamarellou, B. Rouveix Editorial Office, 31 St Ola’s Court, City Business Centre, 25 Lower Road, London SE16 2XB, UK Abstract There are many cephalosporins available and various ways of classifying them for clinical use. Oral cephalosporins probably need a classification of their own. This informal discussion was prompted by the appearance of the recommendations of an expert committee of the Paul Ehrlich Gesellschaft. The views of several other commentators are included. There is considerable individual variation in preference for different styles of classification depending on what the classification is for. © 2001 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved. www.ischemo.org 1. Introduction One feature of the development of antibiotics over the last 40 years has been the production of many new agents based on the original parent compound. These products change the properties (microbiology, pharma- cology, tolerance, etc) of the parent in some way, hopefully for the better. No group of antibiotics has experienced greater di- versity than the cephalosporins, of which many differ- ent varieties are used. One has to be a dedicated cephalosporin-watcher to keep track of these develop- ments. Most clinicians get to know two or three cephalosporins and they change to another only when a newer analogue provides them with a very specific advantage over their existing choice. Unfortunately it is sometimes not easy to compare one cephalosporin with another. Classification of cephalosporins (and other antibiotics) is useful in order to fit new compounds into a pre-existing slot. Several different classifications have been suggested for cephalosporins, none of which have gained univer- sal acceptance. The use of ‘generations’ is still the most widely used and the least helpful. We now seem to have reached the fourth generation and apart from giving us some idea as to the date of first marketing, it does not group together cephalosporins with similar microbio- logical or pharmacological properties. Unfortunately all classifications need to simplify the scientific factors and reach compromises on the key properties that are ex- hibited. This was apparent with cephalosporins from the very first generation. Cephalothin has a half-life of less than an hour, considerably shorter than cephaloridine. So, even though the microbiological fea- tures were very similar, the dosage interval had to be taken into consideration. After the loss of cephaloridine, cefazolin was introduced as a replace- ment, especially for Gram-positive infections. So ce- fazolin is often regarded as first generation and is used mainly against Staphylococcus aureus. One repercussion of this is the use of cephalothin discs as a surrogate for cefazolin in susceptibility testing. That may be satisfac- tory for staphylococci, but the activity of cefazolin against Escherichia coli is more like cefuroxime (second generation) than cephalothin. Cephalosporins are all different. Any classification system involves compro- mises and the compromises are multiplied as more and more cephalosporins appear. Oral cephalosporins are now increasing rapidly in number. How should we classify these compounds? How do they compare? Are there any bases in the classification of parenteral cephalosporins that would * Tel.: +44-20-73212944; fax: +44-20-73212124. E-mail address: jdw@ischemo.demon.co.uk (J.D. Williams). 0924-8579/01/$20 © 2001 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved. PII:S0924-8579(01)00309-0