Question 27

Discuss the factors that may affect your choice of antimicrobial agent in a critically ill septic patient, giving examples where relevant.

[Click here to toggle visibility of the answers]

College Answer

Patient Factors

History of current acute illness
Allergies
Previous antibiotic exposure
Co-morbidities like immunocompetence, Diabetes.
Social history e.g. nursing home resident, alcohol/drug abuse, occupation, contact with birds/animals, travel

Organism

Sensitivity profile
Inducible beta-lactamase producers (e.g. ESCAPPM)
Tendency to develop resistance to antimicrobial during treatment course e.g. Pseudomonas aeruginosa
Intracellular (e.g. aminoglycosides poorly active against strictly intracellular bacteria e.g. Rickettsia, Chlamydia, Coxiella burnetti)

Site of infection

Organs with non-fenestrated capillaries (e.g. brain, prostate, anterior chamber of eye) – poor penetration of non lipid-soluble drugs
Biliary and urinary sepsis – select drugs with hepatic (e.g. ceftriaxone) and urinary excretion (cefotaxime) respectively
Lung – e.g. daptomycin inactivated by surfactant, vancomycin poor penetration

Organ dysfunction

Renal or hepatic dysfunction may result in decreased elimination and increased toxicity

Toxicity

Renal and ototoxicity of aminoglycosides
Renal toxicity of vancomycin
Neurotoxicity of imipenem

Drug interactions

Synergy – beta lactams and aminoglycosides
Pharmacodynamic interactions e.g. macrolides plus other agents causing prolongation of QT

Non anti-microbial effects of antimicrobial

Anti-inflammatory effect of macrolides – may underlie outcome benefit when combined with beta lactams for bacteraemic pneumococcal pneumonia
Inhibition of toxin synthesis in toxic-shock syndrome by clindamycin and linezolid

Hospital factors

Local microbiology/ecology
Ability of monitoring drug levels (TDM)
Presence of an ID physician / Antibiotic Stewardship team in the hospital and their policies

Route of administration

Certain routes of administration may be unreliable in critically ill patients and drugs which can only be administered by that route are less desirable e.g. inhaled zanamivir

Cost

Cost-effectiveness of the antibiotic

Bactericidal vs bacteriostatic

Theoretical benefit from bactericidal drugs. Controversial whether there is a clinical benefit

Additional comments:

Candidates were not expected to provide long lists of antimicrobial agents but to mention some examples where relevant. Overall, the question was poorly answered with superficial answers showing a lack of depth of understanding of the topic. Some wrote about dosing and dose adjustment but not about the choice of antimicrobial agent. Some candidates included key phrases e.g., “time dependent killing” without any demonstration of understanding of how that concept affected the choice of antibiotic.

Discussion

This vaguely worded broad question was answered poorly. This is not because the trainees were unaware of the influence of susceptibilities or drug interactions on antibiotic choice, but because they could not guess what exactly the examiners wanted from them.

The tabulated answer below was largely derived from an article by Surbhi et al (2011) from the Mayo Clinic Proceedings.

Factors which Influence the Choice of Antibiotic Therapy
Factors		Discussion and examples
Disease specifics	Travel history	Geography of endemic regions (eg. leptospirosis) Known ongoing outbreaks (eg. Ebola, H1N1, MERS)
	Occupational exposure	Abbatoir workers (Coxiella burnetii) Fisherman (Vibrio vulnificus) Cattle farmers (Brucella sp.)
	Recreational exposure	IV drug use (endocarditis) Pets or animal exposure (eg. psittacosis or toxoplasma) Bushwalking (eg. tick-borne disease) Alcoholism (prognostic importance in community-acquired pneumonia)
	Recent antimicrobial use	Was it the right antibiotic? i.e. was the course of antibiotics ineffective because of poor agent choice? Did it select for a specific group of organisms? Prophylaxis vs. endemic pathogens (eg, malaria)
	Empiric vs. definitive	Are we convinced of the diagnosis? Is there a need to cover broadly?
	Urgency and timing	Septic patient (every hour delay is associated with a 1% mortality increase)
	Reliability of cultures	Are we sure we cultured the correct pathogen? Is a polymicrobial infection possible (eg. diabetic foot)?
Host factors	Clearance	Decreased renal clearance (by renal failure) Increased renal clearance (by dialysis, or in pregnancy) Decreased hepatic clearance (eg. cirrhosis) Exotically altered clearance (eg. plasma exchange, haemoperfusion, adsorption on to ECMO circuit surfaces, and so forth).
	Age	Paediatric dosing needs to be adjusted to weitght Geriatric dosing needs to account for change in volume of distribution and clearance
	Genetic variation	Genetic differences in side effects from antibiotics Congenital idiosyncracies preventing the use of certain antibiotics (eg. G6PD deficiency resulting in haemolysis when exposed to dapsone or nitrofurantoin) Hepatic enzyme defects
	Pregnancy and lactation	Early pregnancy teratogenesis (eg. nitrofurantoin, chloramphenicol, sulfonamides) Late pregnancy teratogensis (eg. tetracyclines)
	Immunocomptence	Steroid use Post-splenectomy, unvaccinated (susceptible to encapsulated organisms) Chemotherapy Solid organ or bone marrow transplantation

	Allergies	Fatal hypersensitivity reaction vs. some sort of mild scaly rash with a little itching.
Organism factors	Susceptibility	ESCAPPM, MRO, etc Community prevalence of drug resistance Tendency to develop resistance during treatment
	Biology	Intracellular pathogen vs. extracellular Unusual life cycle (eg. helminthes, malaria) - need to kill the eggs or dormant cocoons or whatnot
	Source control	Success of therapy overall is largely determined by this
	Duration of therapy	Short course, eg. in urosepsis Long course, eg. osteomyelitis
	Assessment of response	To repeat the cultures, or not? Is there a point in monitoring serology?
Drug factors	Cost	Fluconazole: $57.99 AUD for 28 capsules (200mg) Anidulafungin: ~$ 300 AUD per single 200mg dose. Cost of monitoring the drug levels How much is a life worth? you amoral monsters, etc.
	Toxicity	Risk vs benefit Some drugs (eg. chloramphenicl) are uniformly "too toxic for use", as there are less toxic alternatives in almost every situation.
	Bioavailability	Convenience of oral dosing Certainty of IV dosing Altered absorption via GI tract in context of critical illness, shock states, low flow, what have you.
	Site penetration	Basic chemistry of the drug influences this aspect. Eg: Penetration to the CSF (lipophilicity) Exclusive distribution into the circulating volume, (hydrophilicity, or high serum protein binding) Weird organ preference (eg. the strange affinity of fluoroquinolones for the prostate) Exclusion of a drug from a specific organ (eg. the inactivation of daptomycin by lung surfactant)
	Bactericidal vs bacteriostatic	Some agents are bacteriostatic against one pathogen and bactericidial against another There may not be any in-vivo difference
	Synergistic combination	Need for multiple agent therapy (eg. in Pseudomonas) Unquestioned need for synergy (eg. cocktail for TB) Advantage from synergy (eg. ampicillin with gentamicin for enterococci) Need for broad-spectrum coverage (in which case you use multiple agents to start with, and then narrow the spectrum of cover) Need for polymicrobial coverage (eg. surgical triple therapy, or in context of bone marrow transplant) Need to prevent emergence of resistance (eg. the argument offered to defend the use of selective digestive tract decontamination; also, a genuine argument for the use of rifampicin and fusidic acid together)

References

LONGACRE, AB. "Factors influencing the choice of antibiotics in therapy." The New Orleans medical and surgical journal 103.4 (1950): 160-167.

Leekha, Surbhi, Christine L. Terrell, and Randall S. Edson. "General principles of antimicrobial therapy." Mayo Clinic Proceedings. Vol. 86. No. 2. Elsevier, 2011.

[Submit a comment or correction]