DipTher MSc EBPT
Director of Pharmacy Services/Clinical Pharmacist
Hillel Yaffe Medical Center Hadera/Teacher Practitioner
School of Pharmacy
Ben Gurion University of the Negev
Evidence-based antibiotic prescribing has been the cornerstone of cost-effective antibiotic utilisation in hospitals. In order to support this process it has been suggested that implementing and developing guidelines can be extremely useful and beneficial, not only in terms of local cost containment but also for maximising clinical effect while minimising the risk of treatment-induced adverse events and overcoming the emerging and serious problem of resistance. This article aims to provide the reader with practical tools that should be considered when seeking cost-effective antibiotic prescribing.
The following points should be taken into account when choosing antibiotics for guidelines:(1-9)
- Antimicrobial activity.
- Pharmacokinetic/pharmacodynamic (PK/PD) factors.
- Local and national microbial resistance patterns.
- Safety and side-effects.
- Local utilisation and prescribing patterns (experience and institutional politics).
- Evidence-based guidelines.
When looking at antimicrobial activity two factors should be considered: in-vivo and in-vitro activities. As a rule of thumb it would be useful to look at internationally accepted guidelines such as the Sanford Guide to Antimicrobial Therapy or other similar sources of information; local or national resistance patterns should also be considered. Thus, wherever possible, culture and sensitivity data should dictate the particular type of antibiotic.
The most important PK factor that contributes towards cost-effective use of antibiotics is the option for an early switch between intravenous (IV) and oral dosage forms for those classes with high bioavailability. Another important factor is achievable serum or target site drug concentration levels. Accordingly, for tissues with poor blood supply or infections within an organ, there is a need to select the right route of administration that will achieve concentrations above minimal inhibitory concentration (MIC), usually IV.
Furthermore, the route of elimination is also important when selecting an antibiotic, especially in urinary tract infections, where renal elimination is crucial in order to achieve high concentration levels in the bladder or urine, while antibiotics that are eliminated hepatically should be avoided (eg, moxifloxacin).
One should also consider antimicrobial activity in relation to drug concentration levels, such as with gentamicin, which allows for once-daily administration due to a postantibiotic effect. Other issues are the volume of distribution, as with highly hydrophilic drugs, especially in cases of dialysis, cirrhosis and congestive heart failure, disease states that require additional antibiotic dosing.
Finally, it is important not to cut corners and to use optimal doses that will achieve optimal MIC as well as full therapeutic effect (ie, concentration-dependent susceptibility). Some antibiotics are considered to have concentration-dependent killing. Consequently, the ratio of peak serum or tissue concentration over the MIC90 should be used, and the higher this ratio the better. For others, time-dependent killing or T>MIC should be considered, allowing for continuous infusion (eg, with macrolides, quinolones and beta-lactams), especially against respiratory pathogens.
Changes in bacterial DNA are due to chromosomal changes or are plasmid-mediated. The former are more stable and less common and usually limited to one species, while the latter are more common and may spread from species to species, expressed or repressed. Some antibiotics are more likely to induce resistance and, if so, this usually happens within two years. Furthermore, it should be remembered that resistance is not always related to the volume of use. The most important factor to consider while trying to control resistance is agent-specific formulary restriction, not related to the volume or duration of use, or antibiotic class per se. It should aim to avoid antibiotics with high resistance potential, such as third-generation cephalosporins, especially ceftazidime. Another technique that has been used and found to be ineffective is antibiotic rotation between different wards.
Some antibiotics are more likely to cause side-effects than others. These side-effects not only may cause unnecessary suffering to the patients but also may increase length of stay, laboratory testing and staff time. Among these side-effects are:
- Drug fever, especially with beta-lactams and sulpha-containing antibiotics.
- Drug rash.
- Cytopenias, especially with beta-lactams, vancomycin, TMP-SMX (trimethoprim/sulphamethoxazole), linezolid and chloramphenicol.ï¿½
- Gastrointestinal side-effects (with special consideration to Clostridium difficile and its consequences).
Other miscellaneous side-effects that should be considered are: nephrotoxicity with aminoglycosides; hepatotoxicity; pancreatitis with HIV medications; photosensitivity with tetracyclines; ototoxicity with aminoglycosides and QT prolongation with macrolides and quinolones.
Both direct and indirect costs should be considered when looking at the total antibiotic cost. Direct costs include acquisition costs. Indirect cost considerations include route and frequency of administration: aim for 12 or 24 hours dosing intervals and try to avoid combination therapy, as it may exceed acquisition cost. As a general rule, combination therapy should be reserved for preventing resistance, increased antimicrobial spectrum and potential synergy, which is tested in vitro. Other factors that may increase costs are: therapeutic drug monitoring and other laboratory testing; potential side-effects; PK considerations such as previously discussed IV/oral switch; and therapeutic failure due to a stepwise approach not using the right drug for the right disease.
In order to avoid pitfalls when developing antibiotic policy, it is important to take into consideration local utilisation patterns and reach as wide a consensus as possible; this, however, does not replace evidence-based medicine. Nevertheless, it would be correct to consult all staff involved before reaching a final decision. Thus, when aiming to change clinical practice, the process should be stepwise and cyclic. Crucial elements should be defined, based on evidence and consensus, tested in practice and adapted to local needs. The guidelines should be simple and compatible with routines as well as attractive and based on a credible source. The next stage should be identifying obstacles to change; these could be either personal and related to clinicians or organisational. The change process should also be considered as a whole to include dissemination, implementation and continuation. In these circumstances, a plan containing combined strategies should be developed. Furthermore, intermediate and long-term targets must be defined within a confined schedule. Obviously, there is a need to carry out the plan and evaluate the process. Strategies that have been proven effective in implementing and disseminating formulary instructions and clinical guidelines are: audit and feedback; sending reminders; providing costing information; interactive workshops as part of a continuing education plan; educational outreach visits (academic detailing); and the utilisation of local opinion ï¿½leaders.
Finally, this process should be accompanied by a systematic and accurate documentation, both in the patient file and centrally (ie, in the pharmacy) using a computer-based database. We believe that clinical pharmacists possess the clinical knowledge and communication skills required in order to be involved in the development and implementation of antibiotic policies and guidelines and, as such, that they can be used in an effective manner to target and promote rational and cost-effective antibiotic prescribing.
- Cunha BA. P&T 2003;28(6):396. Available from: http://www.ptcommunity.com/ptjournal/fulltext/28/6/PTJ2806396.pdf
- Cunha BA. P&T 2003;28(7):468-70. Available from: http://www.ptcommunity.com/ptjournal/fulltext/28/7/PTJ2807468.pdf
- Cunha BA. P&T 2003;28(8):524-7. http://www.ptcommunity.com/ptjournal/fulltext/28/8/PTJ2808524.pdf
- Cunha BA. P&T 2003;28(9):594-6. Available from: http://www.ptcommunity.com/ptjournal/fulltext/28/9/PTJ2809594.pdf
- Cunha BA. Principles of antibiotic formulary selection for P&T committees, part 5: the cost of antimicrobial therapy. P&T 2003;28(10):662-5. Available from: http://www.ptcommunity.com/ptjournal/fulltext/28/10/PTJ2810662.pdf
- Grol R. Beliefs and evidence in ï¿½changing ï¿½clinical practice. BMJ 1997;315:418-21.
- Belongia EA, Schwartz B. Strategies for promoting judicious use of ï¿½antibiotics by doctors and patients. BMJ 1998; 317:668-71.
- Anon. Antibiotic resistance and prescribing practice. MerRec BRIEFING 2003;21:1-8. Available from: http://www.npc.co.uk/MeReC_Briefings/2002/briefings_no_21.pdf
- Schwartzberg E, et al. Evaluation of the impact of a clinical pharmacist on the quality and cost of pharmacotherapy in a general surgical ï¿½department, the Hillel-Yaffe Medical Center. Harefuah 2003;142(1):5-9, 80.