L5 Infection and Immunity Antimicrobials

Overview

This lecture covers how antimicrobial therapy is chosen and monitored in practice. It moves from the factors that shape empiric (before culture results) treatment, through severity assessment and local resistance data, to antimicrobial spectrum, laboratory susceptibility testing, and the pharmacokinetics/pharmacodynamics (PK/PD) that determine dosing. A worked case ties these together, and the lecture closes on antimicrobial stewardship principles (oral switch, shorter treatment courses).

Principles of Empiric Antimicrobial Therapy

Empiric (initial) treatment is chosen before the causative organism is known, based on:

• Clinical diagnosis (syndrome): history (symptoms, travel/occupational/sexual/drug-use/animal exposure), examination (signs), investigations (haematology, biochemistry, microbiology, radiology), and knowledge of the common microbiological causes of that syndrome.
• Nature of the host: e.g. different pathogens by age (neonatal vs adult meningitis); immunosuppressed patients acquire additional pathogens such as Listeria (meningitis) and Pneumocystis jiroveci (pneumonia); neutropaenic patients have worse outcomes if empiric therapy is ineffective.
• Severity of the infection.
• Risk of resistance: previous organisms isolated, antibiotic exposure in the past 30 days, local susceptibility patterns.

Not every syndrome needs treatment: acute sinusitis lasting ≤10 days is managed with self-care, not antibiotics (NICE guideline), because antibiotics make little difference to duration and carry adverse effects (diarrhoea, nausea). A bacterial cause is more likely if symptoms last >10 days, discharge is discoloured/purulent, there is severe unilateral facial/dental pain, fever, or deterioration after an initial milder phase; a back-up (“just in case”) prescription can be used in these situations. Immediate antibiotics (and hospital referral) are needed if the patient is systemically very unwell or has signs of a serious complication (severe systemic infection, intraorbital/intracranial complications).

Different clinical syndromes have characteristically different causative organisms (etiology can vary by geography), e.g.:

• Uncomplicated cystitis: E. coli (70–95%), Staphylococcus saprophyticus (5–10%).
• Community-acquired pneumonia: Streptococcus pneumoniae (39%), Mycoplasma pneumoniae (~16%), Legionella spp. (11%), Haemophilus influenzae (11%).
• Skin/soft tissue infection: β-haemolytic streptococci (72%), Staphylococcus aureus (12%).

Assessing Severity and the Cost of Inadequate Therapy

Severity is staged using a progression (Bone et al. 1992; other measures such as qSOFA also exist):

1. SIRS (systemic inflammatory response syndrome): temperature >38°C or <36°C, heart rate >90, respiratory rate >20 (or PaCO2 <32), white cell count >12,000 or <4,000/µL, or >10% bands.
2. Sepsis = SIRS + confirmed or suspected infection.
3. Severe sepsis = sepsis + end-organ damage.
4. Septic shock = severe sepsis + hypotension.

Myth: there is time to start narrow therapy and escalate later if needed. Fact: inadequate initial antimicrobial therapy increases mortality, and later switching from inadequate to appropriate therapy does not reliably reduce that excess mortality — a delayed appropriate start counts as inadequate therapy. Across multiple ICU studies, inadequate initial therapy was associated with roughly double (or more) the mortality of adequate initial therapy.

Antimicrobial Spectrum and Local Susceptibility Patterns

Local resistance rates vary geographically and must guide empiric choice (e.g. MRSA rates: New Zealand 10.8%, Australia 18%, USA 45%) — hence the instruction to always follow local empiric treatment guidelines (in NZ, informed by the annual hospital antibiogram, e.g. via the MicroGuide app or district guidelines). [slide table of local gram-negative susceptibility percentages partially obscured by an overlay box, exact values for some antibiotic/organism combinations not fully legible]

β-lactam spectrum (broad pattern, not exam detail):

• Penicillin: narrow, Gram-positive only.
• Amoxicillin: Gram-positive + some E. coli.
• Flucloxacillin/cefazolin (1st-gen cephalosporin): good Gram-positive (MSSA, streptococci) + some Gram-negative; Enterococcus and Listeria are intrinsically resistant to all cephalosporins.
• Cefuroxime (2nd-gen): better Gram-negative activity.
• Ceftriaxone/cefotaxime/ceftazidime (3rd-gen): improved stability against Gram-negative β-lactamases, ± Pseudomonas coverage.
• Cefepime (4th-gen): covers AmpC-producing organisms and Pseudomonas.
• Amoxicillin-clavulanate: Gram-positive + Gram-negative + anaerobes.
• Piperacillin-tazobactam/ticarcillin-clavulanate: broad Gram-positive + Gram-negative + Pseudomonas + anaerobes.
• Meropenem/imipenem: broadest coverage, essentially all groups including anaerobes; ertapenem is similarly broad but does not cover Pseudomonas.
• The ESCAPPM group (Enterobacter, Serratia, Citrobacter freundii, Hafnia, Aeromonas, Proteus vulgaris, Providencia, Morganella morganii) can develop inducible resistance to cephalosporins if treated with cephalosporin monotherapy.

Non-β-lactam spectrum:

• Vancomycin, daptomycin, linezolid, erythromycin, rifampicin: Gram-positive only (vancomycin and linezolid cover MRSA).
• Colistin: Gram-negative only, including Pseudomonas/ESCAPPM.
• Aminoglycosides: broad Gram-negative.
• Trimethoprim: narrow Gram-negative; co-trimoxazole: Gram-positive + Gram-negative.
• Ciprofloxacin: broad Gram-negative + Pseudomonas; moxifloxacin: Gram-positive + Gram-negative.
• Tetracycline: Gram-positive + some Gram-negative/anaerobe activity; clindamycin: Gram-positive + anaerobes.
• Metronidazole: anaerobes only.

Targeted Therapy and Susceptibility Testing

Once a pathogen is identified, treatment should shift to the narrowest-spectrum effective agent (“targeted therapy”), guided by culture and susceptibility results and local guidelines (e.g. Te Whatu Ora review antibiotic guidelines).

Specimen collection principles:

• Collect specimens before starting antibiotics, except that antibiotic initiation must not be delayed in severely unwell patients (e.g. meningitis, severe sepsis/septic shock).
• Collect from the site of infection; use pus/tissue rather than swabs where feasible.
• Collect the right number of specimens, e.g. blood cultures: 2 sets (aerobic + anaerobic bottle each), 3 sets if endocarditis is suspected.

Susceptibility testing aims to predict in-vivo success or failure of a drug, using standardised methods and interpretive criteria (CLSI, EUCAST):

• Minimum inhibitory concentration (MIC): lowest antimicrobial concentration that prevents visible growth after 18–24 h incubation.
• Minimum bactericidal concentration (MBC): lowest concentration that kills >99.9% of cells (not used clinically).
• Methods: microbroth dilution, disc diffusion (zone of inhibition around a standard-dose disc on a confluent bacterial lawn), E-test (antibiotic gradient strip).
• Categorical reporting: Susceptible, standard dosing (S) — high likelihood of success at standard dose; Susceptible, increased exposure (I) — high likelihood of success only with increased dose/exposure or high drug concentration at the infection site (e.g. urine); Resistant (R) — high likelihood of failure even with increased exposure.
• Breakpoints and dosing regimens are set per “bug-drug” combination using the population distribution of MIC values (with/without known resistance mechanisms), pharmacokinetics, pharmacodynamics, and clinical correlation of MIC with treatment response. Example: benzylpenicillin vs S. aureus — epidemiological cut-off (ECOFF) 0.125 mg/L; wildtype (susceptible, non-β-lactamase-producing) isolates are ≤0.125 mg/L, while β-lactamase-producing resistant isolates form the majority of a much higher-MIC population.

Pharmacokinetics and Pharmacodynamics of Antimicrobials

Pharmacokinetics (PK, “what the body does to the drug”) covers absorption, distribution, metabolism and excretion:

• Absorption: oral drug undergoes first-pass hepatic metabolism before reaching blood; IV drug enters blood directly.
• Distribution: in blood, drug exists as free drug in equilibrium with protein-bound drug; only free drug can act at the site of infection or reach storage tissue, and only free drug is active against bacteria (protein binding is therefore clinically important).
• Metabolism occurs in the liver; excretion occurs via the kidney.
• Absorption/distribution/metabolism/excretion together determine peak concentration (Cmax) and half-life (t1/2). IV bolus gives the highest, earliest peak with steady decline; IV infusion peaks later and lower; oral peaks a little later still. These parameters are usually measured in healthy volunteers and may differ substantially in critically unwell patients.

Pharmacodynamics (PD, “what the drug does to the bug”) classifies antimicrobials by how their killing relates to concentration over time:

• Time-dependent (T>MIC): efficacy depends on the time the concentration stays above the MIC, e.g. β-lactams.
• AUC24/MIC: efficacy depends on total drug exposure (area under the curve) relative to MIC, e.g. vancomycin, fluoroquinolones.
• Concentration-dependent (Cmax): efficacy depends on peak concentration relative to MIC, e.g. aminoglycosides.
• Post-antibiotic effect (PAE): continued suppression of bacterial growth after the drug concentration falls below the MIC; seen with aminoglycosides but not with β-lactams.

Biofilms

Bacteria exist as free-swimming (planktonic) cells or as biofilms: planktonic bacteria progressively aggregate into small clusters, then a continuous structured layer, then a mature biofilm with mushroom/tower-like structures that release planktonic bacteria back into the environment. Within a mature biofilm there is a metabolic activity gradient: high activity at the outer surface, decreasing through intermediate and low activity, to dormant bacteria deep inside — dormant/low-activity organisms are relevant because many antibiotics act on actively dividing bacteria, so biofilms are harder to eradicate.

Case 1: Applying the Principles (worked case)

A 25-year-old woman presented with fever, dysuria and left flank pain for 2 days, having been discharged one month earlier from an Indian hospital (10 days in ICU) after a motorbike accident during 3 months of backpacking in India and Nepal.

Findings: T 38.2°C, BP 85/45 mmHg, HR 125 bpm, RR 25/min, suprapubic and left flank tenderness; WCC 16.5×10⁹/L (neutrophils 14×10⁹/L), CRP 350 mg/L, creatinine 150 µmol/L, urea 23.9 mmol/L; urine microscopy showed >1000 leukocytes, 230 erythrocytes, and bacteria.

• Applying the SIRS→sepsis→severe sepsis→septic shock scale to these values places her at/near severe sepsis (hypotension, tachycardia, tachypnoea, abnormal WCC, organ dysfunction indicated by raised creatinine/urea).
• Syndrome: pyelonephritis/upper urinary tract infection (fever, flank pain, dysuria, pyuria/haematuria on microscopy).
• Usual UTI pathogens are typically E. coli (70–95% uncomplicated) and other Enterobacterales, but her recent prolonged hospital admission in India raised the risk of an antibiotic-resistant isolate, e.g. an ESBL (extended-spectrum β-lactamase) producer, which would be resistant to amoxicillin, amoxicillin-clavulanate, cefuroxime and ceftriaxone despite retaining susceptibility to meropenem.
• Country-level resistance data showed E. coli resistance in India was very high (~85–90%) to fluoroquinolones and 3rd-generation cephalosporins, ~65–67% to aminoglycosides and amoxicillin-clavulanate, but much lower (~15%) to carbapenems, compared with low resistance across all these classes in New Zealand.
• Given this risk profile, she was empirically treated with a carbapenem.
• Cultures (urine and blood) grew Klebsiella pneumoniae. Susceptibility testing showed a multidrug-resistant isolate, resistant even to amikacin and to meropenem, and PCR confirmed it carried an NDM-1 carbapenemase. [slide title for this susceptibility-testing slide not clearly legible; transcribed from visible annotations and photo labels]
• Remaining treatment options were narrow: initially the isolate was treated with colistin; when found resistant also to meropenem, treatment options were colistin or tigecycline.

Antimicrobial Stewardship and Duration of Therapy

Antimicrobial stewardship means ensuring patients get the right antimicrobial, at the right time, only when needed.

Early switch from IV to oral therapy (“oral is the new IV”) should be considered once the patient is haemodynamically stable, has reasonable source control, has no concerns about intestinal absorption, and the organism is susceptible to an appropriate oral agent. This avoids the complications of IV therapy (phlebitis, extravasation, thrombosis, local/systemic infection) and its costs (longer inpatient stay, pain/inconvenience, higher expense).

There is also a move towards shorter antibiotic courses where trial evidence supports equivalent efficacy to traditional longer courses, e.g. (short vs long, days): community-acquired pneumonia 3–5 vs 7–10; nosocomial pneumonia ≤8 vs 10–15; pyelonephritis 5–7 vs 10–14; intra-abdominal infection 4 vs 10; COPD exacerbation ≤5 vs ≥7; acute bacterial sinusitis 5 vs 10; cellulitis 5–6 vs 10; chronic osteomyelitis 42 vs 84.

Self-test

1. List the four categories of factors that determine empiric antibiotic choice for a given syndrome.
2. Explain why acute sinusitis of ≤10 days’ duration is not treated with antibiotics, and list two features that make a bacterial cause more likely.
3. Describe the progression from SIRS to septic shock, including the defining criteria at each stage.
4. Explain the relationship between initial inadequate antimicrobial therapy and mortality in critically ill patients, and why “starting narrow and escalating later” is not a safe strategy.
5. Why are Enterococcus spp. and Listeria spp. never treated with a cephalosporin, regardless of generation?
6. Distinguish targeted therapy from empiric therapy, and state the specimen-collection rule that should not be broken even in this situation.
7. Define minimum inhibitory concentration (MIC) and distinguish it from minimum bactericidal concentration (MBC).
8. Distinguish the three categorical susceptibility results (S, I, R) in terms of likelihood of therapeutic success.
9. Distinguish time-dependent, AUC/MIC-dependent and concentration-dependent antimicrobial classes, giving an example drug class for each, and define post-antibiotic effect.
10. Why is protein binding pharmacologically important for antimicrobial activity?
11. Describe how the metabolic activity of bacteria varies through the depth of a mature biofilm, and explain why this matters for antibiotic efficacy.
12. For the woman in Case 1, list the clinical and historical features that indicated a high risk of a resistant organism, and explain why a carbapenem was chosen empirically.
13. What organism was eventually cultured in Case 1, what resistance mechanism was confirmed, and what treatment options remained once it was also resistant to meropenem?
14. List four criteria that should be met before switching a patient from IV to oral antibiotic therapy, and state two costs or complications of prolonged IV therapy that this avoids.

Answers

Reveal answers

1. Clinical diagnosis (syndrome, from history/examination/investigations and known causative organisms), nature of the host (age, immunosuppression), severity, and risk of resistance (prior organisms, recent antibiotic exposure, local susceptibility patterns).
2. Antibiotics make little difference to symptom duration or number improving in short-duration sinusitis and can cause adverse effects (diarrhoea, nausea); features favouring a bacterial cause include symptoms lasting >10 days, discoloured/purulent discharge, severe unilateral facial/dental pain, fever, or marked deterioration after an initial milder phase (any two).
3. SIRS: temperature >38°C or <36°C, HR >90, RR >20 (or PaCO2 <32), WCC >12,000 or <4,000/µL or >10% bands. Sepsis = SIRS + confirmed/suspected infection. Severe sepsis = sepsis + end-organ damage. Septic shock = severe sepsis + hypotension.
4. Inadequate initial therapy is associated with substantially higher mortality (roughly double or more across ICU studies); changing from inadequate to appropriate therapy later does not reliably reduce this excess mortality, and a delayed appropriate start is itself considered inadequate therapy, so effective empiric coverage must be chosen from the start.
5. Enterococcus spp. and Listeria spp. are intrinsically resistant to all cephalosporins, across every generation.
6. Empiric therapy is chosen before the causative organism is known (based on syndrome/host/severity/resistance risk); targeted therapy is chosen using culture and susceptibility results and should use the narrowest effective spectrum. Specimens should be collected before starting antibiotics, but this must not delay treatment in a severely unwell patient (e.g. meningitis, severe sepsis/septic shock).
7. MIC is the lowest antimicrobial concentration that prevents visible bacterial growth after 18-24 h incubation. MBC is the lowest concentration that kills >99.9% of cells; MBC is not used in clinical practice.
8. Susceptible, standard dosing (S): high likelihood of success at a standard dose. Susceptible, increased exposure (I): high likelihood of success only with increased dose/exposure or high concentration at the infection site. Resistant (R): high likelihood of failure even with increased exposure.
9. Time-dependent (T>MIC, e.g. β-lactams): efficacy depends on time above MIC. AUC24/MIC (e.g. vancomycin, fluoroquinolones): efficacy depends on total exposure relative to MIC. Concentration-dependent (Cmax, e.g. aminoglycosides): efficacy depends on peak concentration relative to MIC. Post-antibiotic effect: continued suppression of bacterial growth even after the drug concentration falls below the MIC; seen with aminoglycosides, not β-lactams.
10. Only free (unbound) drug can act at the site of infection and against the bacteria; protein-bound drug is not active, so the free fraction determines effective antibacterial exposure.
11. Activity is highest at the outer surface of the biofilm and decreases with depth through intermediate and low activity to dormant bacteria at the core; many antibiotics act preferentially on actively dividing bacteria, so the low-activity/dormant deep layer is harder to eradicate, making biofilm infections more resistant to treatment than planktonic infections.
12. Recent prolonged hospitalisation (including ICU) in India, a country with very high local resistance rates for E. coli to fluoroquinolones, 3rd-generation cephalosporins and aminoglycosides; a carbapenem was chosen because carbapenem resistance remained comparatively low (~15%) even in the Indian data, making it the most reliable empiric option against a possible ESBL-producing organism.
13. Klebsiella pneumoniae was cultured from urine and blood; it was confirmed to carry an NDM-1 carbapenemase by PCR and was multidrug-resistant including to amikacin and eventually meropenem. Once meropenem-resistant, remaining options were colistin or tigecycline.
14. Susceptibility to an appropriate oral antibiotic, haemodynamic stability, reasonable source control, and no concerns about intestinal absorption. IV therapy carries complications such as phlebitis, extravasation, thrombosis and local/systemic infection, and costs such as prolonged inpatient stay and greater expense (any two).