Ventilator-Associated Pneumonia — Part 3: Diagnosis & Antimicrobial Treatment

Clinical diagnostic criteria, CPIS scoring table, microbiologic sampling strategies (ETA, BAL, mini-BAL) with quantitative thresholds, biomarker guidance, empiric antibiotic selection stratified by MDR risk with complete dosing tables and renal adjustments, de-escalation principles, short-course duration evidence, inhaled antibiotics, and treatment failure evaluation.

guidelinesMar 2026guidelines

10. Diagnostic Approach to VAP

10.1 Clinical Suspicion Criteria

VAP should be suspected in any mechanically ventilated patient (intubated ≥ 48 hours) who develops a new or progressive pulmonary infiltrate on chest imaging PLUS at least two of the following clinical features:12

  • Fever: Temperature > 38.0 °C (or hypothermia < 36.0 °C)
  • Leukocytosis or leukopenia: WBC > 12,000 cells/μL or < 4,000 cells/μL
  • Purulent tracheobronchial secretions: New onset or change in character (increased volume, thickened, or change in color to yellow/green)

Additional supportive findings (not required for diagnosis):

  • Worsening gas exchange (declining PaO2/FiO2 ratio, increasing FiO2 or PEEP requirements)
  • Hemodynamic instability or new sepsis
  • Increasing minute ventilation or tachypnea

Diagnostic challenge: Each individual clinical criterion has poor specificity for VAP. Fever and leukocytosis are extremely common in ICU patients from non-infectious causes (drug fever, transfusion reactions, adrenal insufficiency, pancreatitis, DVT). Purulent secretions are present in up to 50% of intubated patients without pneumonia due to tracheobronchitis. New infiltrates may represent atelectasis, ARDS, pulmonary edema, hemorrhage, or organizing pneumonia. The combination of clinical criteria has a sensitivity of approximately 69% and specificity of only 75% for histopathologically confirmed VAP.1

10.2 Clinical Pulmonary Infection Score (CPIS)

The CPIS was developed to standardize the clinical assessment of VAP and improve diagnostic accuracy by combining clinical, radiographic, and microbiologic variables into a numerical score.3

CPIS Scoring Table

Parameter0 Points1 Point2 Points
Temperature (°C)≥ 36.5 and ≤ 38.4≥ 38.5 and ≤ 38.9≥ 39.0 or ≤ 36.0
Blood leukocytes (cells/μL)≥ 4,000 and ≤ 11,000< 4,000 or > 11,000< 4,000 or > 11,000 PLUS band forms ≥ 50%
Tracheal secretionsFewModerate, non-purulentAbundant AND purulent
Oxygenation: PaO2/FiO2 (mmHg)> 240 or ARDS≤ 240 and no evidence of ARDS
Chest radiographNo infiltrateDiffuse or patchy infiltrateLocalized infiltrate
Culture of tracheal aspirateNo or light growthModerate or heavy growth PLUS pathogenic organism consistent with Gram stain

Interpretation:

CPIS ScoreInterpretationAction
< 6VAP unlikelyConsider alternative diagnoses; withhold antibiotics if clinical suspicion is low; repeat assessment at 48–72 hours
≥ 6VAP likelyObtain lower respiratory tract cultures; initiate empiric antibiotics

Limitations of CPIS:

  • Sensitivity 65–77%, specificity 42–85% — performance varies substantially across studies
  • Subjective components (secretion assessment, radiograph interpretation) limit inter-rater reliability
  • Does not reliably distinguish VAP from ventilator-associated tracheobronchitis (VAT) or non-infectious conditions
  • The 2016 guidelines from the major American thoracic and infectious disease professional societies do not endorse CPIS as a sole diagnostic tool but acknowledge its potential role in serial assessment (CPIS at day 0 and day 3) to guide antibiotic de-escalation or discontinuation.1
  • Modified CPIS variants have been proposed but lack adequate validation for clinical decision-making

10.3 Microbiologic Sampling

The 2016 guidelines recommend obtaining lower respiratory tract samples for culture before initiating or changing antibiotics, though antibiotic initiation should not be delayed while awaiting sampling.12

Sampling Methods Comparison

MethodTechniqueQuantitative ThresholdSensitivitySpecificityAdvantagesDisadvantages
Endotracheal aspirate (ETA)Blind suctioning via ETT; quantitative culture of aspirate≥ 10^5 CFU/mL (semi-quantitative: moderate to heavy growth)75–90%40–70%Non-invasive; no bronchoscopy needed; can be performed by RT/nurse; rapidHigh rate of false positives due to colonization; lower specificity
Bronchoalveolar lavage (BAL)Bronchoscopic wedge in affected segment; 120–240 mL sterile saline instilled and retrieved≥ 10^4 CFU/mL73–93%45–100%Directs sampling to affected area; higher specificity; allows cytologyRequires bronchoscopy; brief desaturation/hemodynamic changes; operator dependent
Mini-BAL (blind BAL)Non-bronchoscopic; catheter advanced blindly through ETT until wedged≥ 10^4 CFU/mL63–100%66–96%Does not require bronchoscopy; faster and less costly; can be performed by trained RTBlind technique; cannot target specific segments; risk of sampling non-affected areas
Protected specimen brush (PSB)Bronchoscopic double-catheter brush; protected from upper airway contamination≥ 10^3 CFU/mL33–100%50–100%Reduces contamination from upper airwaySamples small area; lower sensitivity; rarely used in current practice

Key recommendation from the 2016 guidelines:

The guidelines suggest non-invasive sampling with semi-quantitative cultures (i.e., endotracheal aspirate with semi-quantitative reporting) as the preferred initial approach, citing equivalent clinical outcomes between invasive (BAL/PSB) and non-invasive (ETA) strategies in a landmark Canadian RCT (n = 740).14

However, the European guidelines express a preference for quantitative cultures from invasive sampling (BAL), particularly when available, to improve specificity and guide more precise antibiotic de-escalation.2

Practical approach:

  1. Obtain ETA (semi-quantitative or quantitative) as the first-line sampling method in all patients with suspected VAP
  2. Consider BAL or mini-BAL when:
    • ETA is non-diagnostic or discordant with clinical presentation
    • Immunocompromised patients (broader differential diagnosis)
    • Prior antibiotic failure (to identify resistant organisms or alternative diagnoses)
    • Need for cytologic analysis (e.g., eosinophilic pneumonia, DAH)
  3. All samples should be sent before new antibiotics are started, but treatment initiation should not be delayed

10.4 Blood Cultures and Urinary Antigens

TestIndicationYieldNotes
Blood cultures (2 sets from separate sites)All patients with suspected VAPPositive in 8–20% of VAP casesA positive blood culture with the same organism as respiratory cultures strongly supports the diagnosis; also identifies concurrent bacteremia from other sources
Urinary antigen — Legionella pneumophila serogroup 1Suspected Legionella (travel, water exposure, immunosuppression, community-acquired presentation)Sensitivity 70–90% for serogroup 1; does not detect other serogroups or Legionella spp.Remains positive for weeks after infection
Urinary antigen — Streptococcus pneumoniaeNot routinely recommended in VAP (more useful in CAP)Moderate sensitivityMay be positive from prior colonization; more relevant to community-acquired presentation

10.5 Biomarkers in VAP Diagnosis

No single biomarker is sufficiently accurate to confirm or exclude VAP in isolation. Biomarkers should be used as adjuncts to clinical judgment.56

BiomarkerRole in VAPSensitivitySpecificityClinical Application
Procalcitonin (PCT)Distinguishing bacterial infection from non-infectious inflammation; guiding antibiotic de-escalation/discontinuation67–78% (for diagnosis)56–83% (for diagnosis)Better suited for guiding antibiotic duration than for initial diagnosis; a PCT < 0.5 ng/mL or decline of ≥ 80% from peak supports antibiotic discontinuation
C-reactive protein (CRP)Non-specific marker of inflammation56–86%43–79%Limited by poor specificity; serial trending (decline by day 4 of treatment suggests adequate therapy) may be more useful than single values
Soluble triggering receptor expressed on myeloid cells-1 (sTREM-1)Investigated as a rapid BAL fluid biomarker for bacterial pneumonia75–98%73–90%Not widely available; studied primarily in BAL fluid; promising but not recommended for routine clinical use
BAL fluid cell count≥ 2–3% intracellular organisms on direct microscopy has moderate specificity for VAP37–100%89–100%Rapid (available within 1–2 hours of bronchoscopy); highly specific when positive; low sensitivity

11. Empiric Antibiotic Therapy

11.1 Risk Stratification for MDR Pathogens

The cornerstone of empiric antibiotic selection for VAP is risk stratification for multidrug-resistant (MDR) organisms. The 2016 guidelines identify specific risk factors that should guide empiric regimen breadth.1

Risk Factors for MDR VAP

Risk FactorRationale
Prior intravenous antibiotic use within 90 daysSelects for resistant organisms in the oropharynx and GI tract
Hospitalization ≥ 5 days before VAP onsetLonger exposure to nosocomial flora; higher colonization with MDR organisms
Septic shock at time of VAP diagnosisNeed for broader coverage given high mortality of inadequate therapy
Acute respiratory distress syndrome preceding VAPAssociated with prolonged ventilation and antibiotic exposure
Acute renal replacement therapy before VAP onsetMarker of critical illness severity; associated with MDR risk
Local MRSA prevalence > 10–20% among S. aureus ICU isolatesUnit-specific risk for MRSA VAP
Local MDR gram-negative prevalence > 10–20% among gram-negative ICU isolatesUnit-specific risk for resistant Pseudomonas, Acinetobacter, ESBL Enterobacterales

MDR Risk Stratification Decision Framework

Does the patient have ANY of the following?
  • IV antibiotics within 90 days
  • Hospitalized ≥ 5 days
  • Septic shock at VAP onset
  • ARDS preceding VAP
  • RRT before VAP onset

  ┌─── NO to ALL ───────────────────────────────────────┐
  │                                                      │
  │  LOW MDR RISK → Empiric monotherapy                  │
  │  (Section 11.2)                                      │
  │                                                      │
  └──────────────────────────────────────────────────────┘

  ┌─── YES to ANY ──────────────────────────────────────┐
  │                                                      │
  │  HIGH MDR RISK → Dual gram-negative coverage         │
  │  ± MRSA coverage (Section 11.3)                      │
  │                                                      │
  └──────────────────────────────────────────────────────┘

Additional consideration:
  • If unit MRSA prevalence > 10–20% OR patient has MRSA risk
    factors → ADD anti-MRSA agent
  • If unit MDR gram-negative prevalence > 10–20% → USE two
    anti-pseudomonal agents from DIFFERENT classes

11.2 Empiric Therapy — Low MDR Risk (Monotherapy)

For patients with no risk factors for MDR pathogens and in units where local MDR prevalence is ≤ 10–20%, empiric monotherapy with a single anti-pseudomonal agent is recommended:1

AgentStandard Adult DoseInfusion StrategyKey Notes
Piperacillin-tazobactam4.5 g IV every 6 hoursStandard (30 min) or extended (3–4 hours)Broad gram-negative and anaerobic coverage; well-tolerated
Cefepime2 g IV every 8 hoursStandard (30 min) or extended (3–4 hours)Excellent gram-negative coverage including Pseudomonas; no anaerobic activity
Levofloxacin750 mg IV every 24 hours60–90 min infusionAlternative for beta-lactam allergy; rising Pseudomonas resistance limits utility in some units
Imipenem-cilastatin500 mg IV every 6 hours30 min infusionReserve carbapenems when possible; broader than needed for low-risk patients
Meropenem1 g IV every 8 hoursStandard (30 min) or extended (3 hours)As above; prefer beta-lactam/beta-lactamase inhibitors or cephalosporins first-line

Renal Dose Adjustments — Monotherapy Agents

AgentCrCl 30–50 mL/minCrCl 10–29 mL/minCrCl < 10 mL/min or HDCRRT
Piperacillin-tazobactam3.375 g IV q6h2.25 g IV q6h2.25 g IV q8h (give dose after HD)3.375–4.5 g IV q6–8h (institution-specific)
Cefepime2 g IV q12h1 g IV q12h1 g IV q24h (give dose after HD)1–2 g IV q12h
Levofloxacin750 mg IV q48h750 mg IV q48h500 mg IV q48h (give dose after HD)750 mg IV q48h
Imipenem-cilastatin500 mg IV q6–8h250 mg IV q6h250 mg IV q12h250–500 mg IV q6–8h
Meropenem1 g IV q12h500 mg IV q12h500 mg IV q24h1 g IV q12h

11.3 Empiric Therapy — High MDR Risk (Combination Therapy)

For patients with one or more risk factors for MDR organisms, empiric therapy should include two anti-pseudomonal agents from different classes and, when indicated, an anti-MRSA agent.12

Gram-Negative Coverage — Choose TWO Agents from Different Classes

Class 1 — Anti-pseudomonal beta-lactam (choose one):

AgentStandard Adult DoseInfusion Strategy
Piperacillin-tazobactam4.5 g IV every 6 hoursExtended infusion (3–4 hours) preferred in critically ill patients
Cefepime2 g IV every 8 hoursExtended infusion (3–4 hours) preferred
Ceftazidime2 g IV every 8 hoursStandard (30 min) or extended infusion
Meropenem1–2 g IV every 8 hoursExtended infusion (3 hours) preferred; use 2 g for suspected Pseudomonas with elevated MICs
Imipenem-cilastatin500 mg IV every 6 hours30 min infusion
Aztreonam2 g IV every 8 hours30 min infusion; alternative for severe beta-lactam allergy (cross-reactivity is negligible)

Class 2 — Non-beta-lactam anti-pseudomonal agent (choose one):

AgentStandard Adult DoseKey Notes
Amikacin15–20 mg/kg IV every 24 hours (based on ABW)Target peak 56–64 mg/L, trough < 4 mg/L; monitor renal function and levels
Gentamicin5–7 mg/kg IV every 24 hours (based on IBW)Target peak 20–30 mg/L, trough < 1 mg/L; higher nephrotoxicity than amikacin
Tobramycin5–7 mg/kg IV every 24 hours (based on IBW)Target peak 20–30 mg/L, trough < 1 mg/L; preferred aminoglycoside for Pseudomonas in some institutions
Ciprofloxacin400 mg IV every 8 hoursAvoid if fluoroquinolone used within 90 days; rising resistance limits utility
Levofloxacin750 mg IV every 24 hoursAs above
Colistin (polymyxin E)LD: 300 mg CBA (9 MU) IV x 1; MD: 150 mg CBA (4.5 MU) IV q12hReserve for XDR gram-negatives; nephrotoxicity monitoring essential; see pathogen-specific section
Polymyxin BLD: 2.5 mg/kg IV x 1; MD: 1.25–1.5 mg/kg IV q12hLess nephrotoxic than colistin; no renal dose adjustment needed (not renally eliminated)

MRSA Coverage — Add When Indicated

Add an anti-MRSA agent when any of the following apply:1

  • Unit MRSA prevalence > 10–20% among S. aureus ICU isolates
  • Patient has known MRSA colonization (nasal or respiratory)
  • Prior MRSA infection
  • Prior IV antibiotics within 90 days (MRSA risk factor)
  • Gram stain of respiratory sample showing gram-positive cocci in clusters
AgentStandard Adult DoseKey Notes
VancomycinLD: 25–30 mg/kg IV x 1; MD: 15–20 mg/kg IV q8–12h (target AUC/MIC 400–600 using Bayesian software, or trough 15–20 mg/L if AUC monitoring unavailable)First-line anti-MRSA; monitor renal function; 2020 consensus recommends AUC-guided dosing over trough-based monitoring
Linezolid600 mg IV/PO every 12 hoursNo renal adjustment; monitor platelets (thrombocytopenia with courses > 14 days); avoid with serotonergic agents; good lung penetration

Renal Dose Adjustments — Combination Therapy Agents

AgentCrCl 30–50 mL/minCrCl 10–29 mL/minCrCl < 10 mL/min or HDCRRT
Ceftazidime1 g IV q12h1 g IV q24h1 g IV q48h (after HD)1–2 g IV q12h
Aztreonam1 g IV q8h500 mg – 1 g IV q12h500 mg IV q12h (after HD)2 g IV q12h
AmikacinExtend interval; monitor levelsExtend interval to q24–48h; levels essentialPost-HD dosing; redose based on levels10 mg/kg q24h; monitor levels
Gentamicin / TobramycinExtend interval; monitor levelsExtend interval to q24–48h; levels essentialPost-HD dosing; redose based on levels2–3 mg/kg q24–48h; monitor levels
Ciprofloxacin400 mg IV q12h200 mg IV q12h200 mg IV q12h400 mg IV q12h
Colistin (CBA)Reduce MD by 25–50%Reduce MD by 50%Consult pharmacist; supplement post-HDReduced dose; institution-specific
VancomycinAdjust per AUC monitoringAdjust per AUC monitoring; may require q24h dosingPost-HD redosing based on levels15–20 mg/kg q24–48h; monitor levels
LinezolidNo adjustmentNo adjustmentNo adjustment (removed by HD; give after HD)No adjustment

11.4 Extended Infusion Beta-Lactam Strategy

For critically ill patients with VAP (especially those with septic shock or infections caused by organisms with elevated MICs), extended or continuous infusion of beta-lactam antibiotics optimizes pharmacokinetic-pharmacodynamic (PK-PD) target attainment:7

Infusion StrategyMechanismEvidence
Standard infusion (30 min)Intermittent bolus dosingAdequate for organisms with low MICs; standard approach
Extended infusion (3–4 hours)Maintains drug concentration above MIC for ~60–70% of the dosing intervalImproved clinical cure and survival in meta-analyses of critically ill patients; recommended as first-line approach in severe infections
Continuous infusion (24-hour)Maintains constant drug concentration above MICMaximum PK-PD optimization; requires dedicated IV line; stability considerations

Recommended extended infusion dosing:

AgentExtended Infusion DoseInfusion DurationStability at Room Temperature
Piperacillin-tazobactam4.5 g IV q6hOver 3–4 hours24 hours
Cefepime2 g IV q8hOver 3–4 hours24 hours
Meropenem1–2 g IV q8hOver 3 hours4–8 hours (stability is a key limitation)
Ceftazidime2 g IV q8hOver 3–4 hours24 hours

12. Antibiotic De-escalation

12.1 Principles

De-escalation — narrowing the spectrum of empiric antibiotics based on culture results — is a cornerstone of antibiotic stewardship in VAP management.18

Timeline for de-escalation:

TimepointAction
0–6 hoursInitiate empiric broad-spectrum antibiotics after obtaining cultures
24–48 hoursReview preliminary Gram stain and culture results; reassess clinical trajectory
48–72 hoursFinal culture identification and susceptibility results available → actively de-escalate
Day 3–5Reassess clinical response; stop antibiotics if VAP is excluded (cultures negative + clinical improvement + alternative diagnosis identified)

12.2 De-escalation Approach

ScenarioAction
Cultures identify a specific pathogen with susceptibilitiesNarrow to the most targeted effective agent; discontinue combination therapy
MRSA not isolated and nasal MRSA screen negativeDiscontinue vancomycin/linezolid (negative nasal MRSA swab has > 95% NPV for MRSA VAP)
Cultures negative at 48–72 hoursIf patient is clinically improving and VAP diagnosis is in doubt, consider stopping antibiotics; if clinical concern remains, continue narrowed therapy
Pseudomonas identified and susceptibleDe-escalate to a single agent with activity; continue monotherapy (combination therapy for the full course is not recommended when the organism is susceptible)
ESBL-producing organismTypically requires carbapenem (meropenem or imipenem); de-escalate from broader combination to carbapenem monotherapy

12.3 MRSA Nasal Screening for De-escalation

The MRSA nasal swab (PCR or culture) has emerged as a valuable tool for guiding empiric MRSA therapy in VAP:9

Test ResultMRSA VAP NPVAction
Negative nasal MRSA screen95–99%Strongly consider discontinuing empiric vancomycin/linezolid
Positive nasal MRSA screenLow PPV (~15–30%)Does not confirm MRSA VAP; MRSA colonization is common; continue empiric MRSA coverage until respiratory cultures finalize

13. Duration of Antibiotic Therapy

13.1 Short-Course Therapy — 7 Days

The 2016 guidelines strongly recommend 7 days of antibiotic therapy for VAP, rather than longer courses (10–21 days), for most patients.1210

Evidence:

  • A landmark multicenter RCT (n = 401) compared 8 days vs 15 days of antibiotic therapy for VAP. The 8-day group had equivalent 28-day mortality (18.8% vs 17.2%) and clinical cure rates, with significantly fewer antibiotic-free days (mean 13.1 vs 8.7 days, p < 0.001). No difference in recurrence except for non-fermenting gram-negative bacilli (primarily Pseudomonas), where 8-day therapy was associated with a higher recurrence rate (40.6% vs 25.4%), though without impact on mortality.10
  • A subsequent meta-analysis confirmed that short-course therapy (7–8 days) is associated with more antibiotic-free days, reduced emergence of resistant organisms, and similar clinical outcomes compared with longer courses.11

13.2 Exceptions — When Longer Courses May Be Warranted

SituationSuggested DurationRationale
Non-fermenting gram-negatives (Pseudomonas, Acinetobacter, Stenotrophomonas)7–14 days (individualized)Higher recurrence with 7-day courses; consider extending if slow clinical response
MRSA pneumonia7–14 daysSome experts recommend 10–14 days based on limited data; clinical response should guide duration
Lung abscess or necrotizing pneumonia14–21 days or longerInadequate source control with shorter courses
Empyema14–21 days minimumRequires adequate drainage plus prolonged antibiotics
Immunocompromised patientsIndividualized (typically 14+ days)Impaired host defenses may necessitate longer treatment
Bacteremia from VAPMinimum 7 days from first negative blood cultureStandard approach for bacteremia

13.3 Procalcitonin-Guided Antibiotic Discontinuation

Procalcitonin (PCT) can be used as an adjunctive tool to guide antibiotic discontinuation in VAP:15

PCT CriterionRecommendation
PCT < 0.5 ng/mL or decline ≥ 80% from peakConsider antibiotic discontinuation if clinically improved
PCT persistently elevated or risingContinue or reassess antibiotics; consider treatment failure, inadequate source control, or secondary infection
Serial PCT measurementsObtain baseline at VAP diagnosis, then every 48–72 hours

Important: PCT-guided algorithms should complement — not replace — clinical judgment. PCT should not be used as a sole criterion for continuing antibiotics in an improving patient.


14. Inhaled Antibiotics as Adjunctive Therapy

14.1 Indications

Inhaled (aerosolized) antibiotics are considered as adjunctive therapy (in addition to IV antibiotics, not as a substitute) for VAP caused by MDR gram-negative organisms that are susceptible only to polymyxins or aminoglycosides, or when clinical response to IV therapy alone is inadequate.112

14.2 Available Inhaled Agents

AgentDose (Nebulized)FrequencyVehicleKey Notes
Tobramycin300 mgEvery 12 hoursPreservative-free solutionMost commonly used inhaled aminoglycoside for VAP
Amikacin400 mgEvery 12 hoursPreservative-free solutionAlternative to tobramycin
Colistimethate (colistin)75–150 mg CBA (2–4 MU)Every 8–12 hoursReconstitute in sterile water; use within 24 hours of reconstitutionFirst-line inhaled agent for XDR gram-negatives (Acinetobacter, Pseudomonas)

14.3 Administration Requirements

RequirementDetails
Nebulizer typeVibrating mesh nebulizer preferred (e.g., Aerogen Solo); jet nebulizers are less efficient; ultrasonic nebulizers should be avoided with some formulations
Ventilator settingsIncrease inspiratory time (if possible); consider reducing flow rate to optimize aerosol delivery; target I:E ratio of 1:1
Circuit placementPlace nebulizer in the inspiratory limb, 15–30 cm proximal to the Y-piece
HME removalRemove any heat-moisture exchanger during nebulization; replace with heated humidifier or remove during treatment
MonitoringMonitor for bronchospasm during administration; pre-treat with bronchodilator (albuterol 2.5 mg) 15 minutes before inhaled antibiotic
DurationContinue for the full course of VAP treatment (typically 7 days; may extend for MDR organisms)

15. Treatment Failure Evaluation

15.1 Definition

Treatment failure should be suspected if there is no clinical improvement or clinical deterioration by 72 hours after initiation of appropriate empiric therapy.12

15.2 Differential Diagnosis of Treatment Failure

CategoryConsiderations
Incorrect initial diagnosisAtelectasis, ARDS, pulmonary embolism, pulmonary hemorrhage, drug-induced lung injury, organizing pneumonia, cardiogenic pulmonary edema
Resistant or untreated organismMDR pathogen not covered by empiric regimen; fungal infection; viral pneumonia; Mycobacterium tuberculosis; Nocardia; Pneumocystis jirovecii
Inadequate antibiotic dosingSubtherapeutic drug levels (particularly in augmented renal clearance, obesity, or high-volume fluid resuscitation); consider extended infusion, higher doses, or therapeutic drug monitoring
Complication of VAPEmpyema, lung abscess, necrotizing pneumonia requiring drainage or prolonged therapy
SuperinfectionNew organism acquired during treatment (often more resistant than the initial pathogen)
Non-pulmonary infectionCLABSI, CAUTI, Clostridioides difficile, intra-abdominal infection, sinusitis
Drug feverAntibiotics themselves can cause persistent fever

15.3 Evaluation Steps

  1. Repeat lower respiratory tract cultures (consider BAL if ETA was initially used)
  2. Obtain CT chest (more sensitive than portable CXR for empyema, abscess, alternative diagnoses)
  3. Review antibiotic susceptibilities and assess adequacy of dosing
  4. Consider therapeutic drug monitoring (vancomycin AUC, aminoglycoside levels, beta-lactam levels if available)
  5. Evaluate for non-pulmonary sources of infection
  6. Consider broadening empiric coverage to include XDR gram-negatives and/or fungal pathogens
  7. Infectious disease consultation recommended for treatment failure


  1. Kalil AC, Metersky ML, Klompas M, et al. “Management of adults with hospital-acquired and ventilator-associated pneumonia: 2016 clinical practice guidelines by the Infectious Diseases Society of America and the American Thoracic Society.” Clin Infect Dis. 2016;63(5):e61-e111. ATS/IDSA. DOI: 10.1093/cid/ciw353 ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎

  2. Torres A, Niederman MS, Chastre J, et al. “International ERS/ESICM/ESCMID/ALAT guidelines for the management of hospital-acquired pneumonia and ventilator-associated pneumonia.” Eur Respir J. 2017;50(3):1700582. ERS/ESICM/ESCMID/ALAT. DOI: 10.1183/13993003.00582-2017 ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎

  3. Pugin J, Auckenthaler R, Mili N, Janssens JP, Lew PD, Suter PM. “Diagnosis of ventilator-associated pneumonia by bacteriologic analysis of bronchoscopic and nonbronchoscopic ‘blind’ bronchoalveolar lavage fluid.” Am Rev Respir Dis. 1991;143(5 Pt 1):1121-1129. DOI: 10.1164/ajrccm/143.5_Pt_1.1121 ↩︎

  4. Canadian Critical Care Trials Group. “A randomized trial of diagnostic techniques for ventilator-associated pneumonia.” N Engl J Med. 2006;355(25):2619-2630. DOI: 10.1056/NEJMoa052904 ↩︎

  5. Schuetz P, Wirz Y, Sager R, et al. “Procalcitonin to initiate or discontinue antibiotics in acute respiratory tract infections.” Cochrane Database Syst Rev. 2017;10:CD007498. DOI: 10.1002/14651858.CD007498.pub3 ↩︎ ↩︎

  6. Gibot S, Cravoisy A, Levy B, Bene MC, Faure G, Bollaert PE. “Soluble triggering receptor expressed on myeloid cells and the diagnosis of pneumonia.” N Engl J Med. 2004;350(5):451-458. DOI: 10.1056/NEJMoa032655 ↩︎

  7. Roberts JA, Abdul-Aziz MH, Davis JS, et al. “Continuous versus intermittent beta-lactam infusion in severe sepsis: a meta-analysis of individual patient data from randomized trials.” Am J Respir Crit Care Med. 2016;194(6):681-691. DOI: 10.1164/rccm.201601-0024OC ↩︎

  8. Leone M, Bechis C, Baumstarck K, et al. “De-escalation versus continuation of empirical antimicrobial treatment in severe sepsis: a multicenter non-blinded randomized noninferiority trial.” Intensive Care Med. 2014;40(10):1399-1408. DOI: 10.1007/s00134-014-3411-8 ↩︎

  9. Parente DM, Cunha CB, Engemann JJ, Hollenbeck BK, Timbrook TT. “The clinical utility of methicillin-resistant Staphylococcus aureus (MRSA) nasal screening to rule out MRSA pneumonia: a diagnostic meta-analysis with antimicrobial stewardship implications.” Clin Infect Dis. 2018;67(1):1-7. DOI: 10.1093/cid/ciy024 ↩︎

  10. Chastre J, Wolff M, Fagon JY, et al. “Comparison of 8 vs 15 days of antibiotic therapy for ventilator-associated pneumonia in adults: a randomized trial.” JAMA. 2003;290(19):2588-2598. PneumA Trial. DOI: 10.1001/jama.290.19.2588 ↩︎ ↩︎

  11. Pugh R, Grant C, Cooke RP, Dempsey G. “Short-course versus prolonged-course antibiotic therapy for hospital-acquired pneumonia in critically ill adults.” Cochrane Database Syst Rev. 2015;8:CD007577. DOI: 10.1002/14651858.CD007577.pub3 ↩︎

  12. Solé-Lleonart C, Rouby JJ, Blot S, et al. “Nebulization of antiinfective agents in invasively mechanically ventilated adults: a systematic review and meta-analysis.” Anesthesiology. 2017;126(5):890-908. DOI: 10.1097/ALN.0000000000001570 ↩︎