Appropriate Use of Antibiotics in the ICU

Prescribing antibiotics in the ICU always poses a dilemma to the treating physician

"Bacteria are becoming increasingly resistant and there aren't too many new antibiotics in the pipeline" - Dr Ashit Hegde Consultant, Medical Intensive Care & Head of Critical Care PD Hinduja Hospital Mumbai

Penicillin was discovered by Sir Alexander Fleming in 1928. Florey and his colleagues used penicillin for the first time on a policeman with orbital cellulitis. The policeman improved rapidly. Unfortunately, the penicillin stock was exhausted, the infection relapsed and the patient died 10 days later. This use of penicillin by Florey, opened up the golden era of antibiotics - an era which is about to come to an end. Bacteria are becoming increasingly resistant and there aren't too many new antibiotics in the pipeline. We, therefore, need to use antibiotics optimally in order to contain bacterial resistance.

Prescribing antibiotics in the ICU always poses a dilemma to the treating physician. On the one hand, he would like to avoid the emergence of multidrug-resistant micro-organisms. On the other, he has an immediate duty to save the life of his seriously-ill patient by making sure he gets the right antibiotic.

These are some of the questions a physician should ask himself before prescribing an antibiotic:

• Should an antibiotic be prescribed at all?
• Should I use antibiotics prophylactically?
• At what dose?
• For how long?
• What is the initial choice?
• What to do once cultures are available?

Should an Antibiotic be Prescribed at All?

Bacterial infections are over diagnosed in the ICU. Physicians must understand that:

Fever not always means infection: Many patients in the ICU have fever for a variety of reasons other than infections- drug fever, blood transfusions, pancreatitis and cerebro-vascular accidents. Physicians often mistake antibiotics for antipyretics. In an otherwise stable patient, non- infectious cause of fever should always be considered before rushing to prescribe or change antibiotics.

Leucocytosis not always means infection: Patients in the ICU quite often have leucocytosis even if they do not have infection. For instance, due to corticosteroids, seizures, cathecolamines and blood transfusions.

Colonisation not equal to infection: Most ICU patients have colonisation of their respiratory secretions, urinary catheters, central lines and bed sores. The bacteria are lying quiet without causing invasion and without provoking an inflammatory response. Whenever a positive culture is obtained from a non-sterile site ( tracheal secretions , central lines, urine, bed sores ) the physician should spend some time trying to analyse whether that positive culture represents true infection or merely colonisation ( which is often the case) .

Infection needs antibiotics, colonisation does not. Treating colonisation with antibiotics merely selects out even more resistant organisms. Differentiating colonisation from infection is therefore a very important aspect of ICU care.

Conversely there are many sick, immuno-compromised, elderly patients in the ICU who may not have fever and yet have infection. Fever is neither sensitive to nor specific for infection. One third of septic patients present with normal temperatures and 10 per cent are hypothermic.

Consider using antibiotics even in the absence of fever if the patient has some of the following :

• Hypothermia.
• Thrombocytopenia.
• Leucocytosis.
• Tachypnoea , tachycardia (with normal chest and heart ).
• Unexplained hypotension .
• Unexplained oliguria.
• Altered sensorium.
• Feed intolerance.
• Oedema.

Avoid Inappropriate Prophylactic Use of Antibiotics

Many ICU patients are prescribed 'prophylactic' antibiotics, merely because they have various indwelling lines or tubes. Administration of prophylactic antibiotics might delay the onset of infection by a few days, but when the infection does occur it is more likely to be caused by a resistant organism.

Physicians often confuse virulence and resistance. Patients with resistant infections have worse outcomes because they often do not get appropriate antibiotics in time and not because the resistant infections are necessarily more virulent. Remember that virulent infections are not always resistant and resistant infections are not always virulent. Also remember, community acquired infections may be virulent, but are not usually resistant. Nosocomial infections may not be virulent, but are usually resistan

At what dose should the antibiotic be given?

It has been shown that bacterial killing is a function of drug concentration and time of exposure. Some antibiotics are most effective when their peak concentration is high (aminoglycosides, metronidazole ). The dosing strategy for such 'concentration-dependant antibiotics' is to administer them in a high enough dose, just once a day.

Other antibiotics (B- lactams) exhibit 'time dependant killing.' The longer their levels are above a certain threshold, the more effective their bacterial killing. The dosing strategy for such antibiotics would be to dose them more frequently or probably even better, to administer them as continuous or prolonged infusions rather than as boluses.

For how long should the antibiotic be given?

The longer a patient is exposed to an antimicrobial, the greater the likelihood that colonisation with resistant organisms will occur.

Most antibiotics are being given for far too long. Start antibiotics immediately. Stop antimicrobial treatment, when infection is cured.

(If patient is afebrile and well for 48 hours). This usually means approximately seven days of therapy for most ICU infections.

Sometimes an antibiotic is prescribed because the physician suspected infection, but the subsequent course and investigations reveal an alternative cause for the patient's condition. The physician does not have to complete a course of antibiotics just because he started it. A physician can safely stop antibiotics when cultures are negative and infection is unlikely.

What should the initial empiric antibiotic be?

Initial empirical anti-infective therapy should include one or more drugs that have activity against the likely pathogens (bacterial or fungal) and that penetrate into the presumed source of the sepsis. Inappropriate treatment represents the use of antibiotics with poor or no in vitro activity against the microorganisms causing infection. Inappropriate antimicrobial treatment of serious infections has been shown to be an important determinant of hospital mortality. Changing treatment based on the subsequent culture results may not reduce the excess risk of hospital mortality associated with inappropriate initial treatment. It is therefore crucial to get it right the first time when treating serious ICU infections.

It is also very crucial that the first dose of antibiotic be administered as soon as possible (immediately after cultures are drawn).

In a study conducted by Kumar et al (Crit Care Med 2006;34:1589 ) 'In Patients with Septic Shock'- each hour of delay in antibiotics was associated with an average decrease in survival of 7.6 per cent. Time to appropriate antibiotic therapy was the single strongest predictor of outcome. The most important cause of improper initial therapy is failure to cover for resistant organisms.

The most important reason why the treating physician got it wrong the first time was that he did not suspect that the infection could be caused by resistant organisms and therefore failed to cover for these organisms. Physicians often confuse virulence and resistance. Patients with resistant infections have worse outcomes because they often do not get appropriate antibiotics in time and not because the resistant infections are necessarily more virulent.

Remember that virulent infections are not always resistant and resistant infections are not always virulent. Also remember, community acquired infections may be virulent, but are not usually resistant. Nosocomial infections may not be virulent, but are usually resistant.

Risk Factors

Patients with any of the risk factors (mentioned in the box), should be covered for resistant organisms in the initial empiric choice of antibiotics. Physicians should also be aware of the microbiologic flora prevalent in their practice environment. The flora may vary from unit to unit even within the same hospital and may vary over time within a particular unit. It is therefore important for each hospital to have an updated and accurate anti-biogram reflecting the bacterial pathogens and their antimicrobial susceptibility. Without such data, logical prescribing is going to be very difficult.

The physician should also be aware of the 'antibiotic gap' (which organisms are not covered) of each antibiotic. It is very likely that the next infection will be caused by an organism that is within the antibiotic gap of the antibiotic that the patient is already on. For instance, if the patient is already on a Carbapenem, the next infection is likely to be caused by MRSA and MRSE, Enterococcus faecium, Stenotrophomonas maltophilia, Burkholderia cepacia, fungus. These organism are not covered (are within the antibiotic gap ) by Carbapenems.

Antibiotic De-escalation

A major concern is that increased use of empiric broad antimicrobial agents will result in greater resistance. Antimicrobial de-escalation balances the twin objectives of providing appropriate initial treatment while limiting antimicrobial resistance. De-escalation means that the initial therapy (after sending cultures) is broad enough and high enough to treat the most likely pathogens . After the culture reports are obtained, the antibiotics are narrowed down (de-escalated) . Antibiotics are discontinued quickly once the patient recovers. Responsible antibiotic prescribing by physicians is the only way we can prevent the emergence of the post antibiotic era.

ashit_hegde@vsnl.com