Sanders and Sanders  introduced the idea of cycling antibiotics on special care units to circumvent the emergence of resistance in bacterial pathogens. The salient features of such an antibiotic cycling program, which include infection control procedures, elimination or avoidance of certain antibiotics and also continuous education and monitoring, were pointed out. As I understand the intention, the ideas presented in the paper were advanced to stimulate discussion, as no controlled trials were quoted to prove the efficacy of the strategy. Indeed, it might be difficult to devise a suitable control group.
What can happen to an antibiotic cycling program? To describe the worst scenario: within a few months of the introduction of such a cycling program, the staff will be disillusioned because nobody will remember which cycled regimen is in force, there being so many other tasks that have to be undertaken and regulated in intensive care units. It can also be foreseen that within a short period of time (less than 2 years of cycling), strains of multidrug-resistant Pseudomonas aeruginosa or Enterobacter spp., producing inducible β-lactamases and/or extended-spectrum β-lactamases with a loss or alteration of outer-membrane proteins, will invalidate this very mechanistic cycling process.
In my personal experience, mere may be at least one other way to handle infections in patients on special care units to avoid or to postpone the emergence of resistance. Based upon results of microbiological investigations, it is possible to create not what is called in the paper an ‘empirical’ but rather a ‘calculated’ chemotherapy for different sites of infection, e.g. blood, lung, abdomen, wound, urinary tract.
The rules for such a calculated chemotherapy have to be elaborated co-operatively by clinicians, microbiologists and pharmacists, and have to take into account five different aspects: prevalence of pathogens at a given site of infection; susceptibility patterns of these pathogens; pharmacokinetic aspects, including dosing; the condition of the patient (e.g. immunosuppression, renal function); and costs. On the basis of these data, it is possible to create very different therapeutic regimens tailored to the special requirements—dictated by the local epidemiologic situation, the site of infection and the individual patient characteristics—which can be used in parallel, thus avoiding the impact of a single regimen over even a short period of time on selection of resistance. It is possible to use ureidopenicillins, second-generation cephalosporins, and in some cases third-generation cephalosporins, as well as aminoglycosides or potent fluoroquinolones such as ciprofloxacin. It is important to adapt the regimens to the local situation. For example, in many of the small or medium-sized hospitals in Germany, infections by methicillin-resistant Staphylococcus aureus (MRSA) are uncommon and emergence of resistance in MRSA to fluoroquinolones does not present so great a threat as it may do elsewhere.
During the time that the guidelines are in use, the results of culture must be monitored very carefully by the microbiologist on a continuous basis, to detect all possible resistance mechanisms in the pathogens isolated. This should be done in ‘real time’, to give early warning of the emergence of resistant strains. When this happens, it is time to change the guidelines of calculated chemotherapy. The analysis should show not only the susceptibilities of first isolates from individual patients but also those of subsequent isolates, to determine whether resistance may develop during treatment. A further point is the need for rapid microbiological results; for several years it has been possible in many cases to obtain full results within 24 h . There are even some data showing that rapid microbiological results may improve survival of patients . By introducing these methods, it is possible to shorten the period of calculated chemotherapy and to switch to individualized chemotherapy, eliminating carbapenems or third- and fourth-generation cephalosporins whenever possible. Rapid microbiology significantly decreases selective pressure by avoiding overuse of highly efficient antibiotics.
My personal opinion is that microbiologists should promote such varied therapeutic regimens in cooperation with clinicians and pharmacists. For future investigation I would propose to take patients on a regimen of calculated chemotherapy in combination with rapid microbiology as a control group to compare with a group of patients on wards which have adopted cycling therapy.