# Community-Acquired Pneumonia: Risk Stratification and Antibiotic Stewardship

## Overview

Community-acquired pneumonia (CAP) is the leading infectious cause of hospitalization and death in the United States. Risk stratification guides site-of-care decisions (outpatient, ward, or ICU), empiric antibiotic selection is based on the setting of care and risk factors for resistant organisms, and antibiotic stewardship focuses on appropriate duration, de-escalation, and avoidance of unnecessary broad-spectrum coverage.

## Diagnosis

### Clinical Features

Patients present with cough (productive or non-productive), fever, dyspnea, and pleuritic chest pain. Examination findings include tachypnea, tachycardia, hypoxemia, crackles, bronchial breath sounds, egophony, and dullness to percussion. Elderly or immunocompromised patients may present atypically with confusion, falls, or absence of fever.

### Diagnostic Workup

A chest X-ray is required to confirm the diagnosis, demonstrating a new infiltrate in the appropriate clinical context. CXR may be negative early (within 12 hours) or in severely dehydrated patients, and CT chest is more sensitive but not routinely needed. Blood cultures are not routine for all CAP but should be obtained for severe CAP, ICU admission, or when empiric MRSA/Pseudomonas coverage is being considered. Sputum culture and Gram stain are most useful for severe CAP or suspected resistant organisms. Urinary antigen testing for Legionella (in severe CAP, travel history, or treatment failure) and Pneumococcus (in severe CAP) adds diagnostic value. Respiratory viral panels during viral season (influenza, RSV, COVID-19) may allow antiviral therapy and avoidance of unnecessary antibiotics. Procalcitonin may help distinguish bacterial from viral pneumonia and is more useful for guiding antibiotic duration than initiation.

### Microbiology

Streptococcus pneumoniae remains the most common pathogen, though it is less frequently identified with modern diagnostics. Atypical organisms include Mycoplasma pneumoniae, Chlamydophila pneumoniae, and Legionella pneumophila. Viral causes (influenza, SARS-CoV-2, RSV) are increasingly recognized in adults. Staphylococcus aureus including MRSA is seen post-influenza, with cavitary disease, and in injection drug use. Gram-negatives including Pseudomonas are associated with structural lung disease, prior hospitalizations, and immunosuppression.

<image>Chest X-ray patterns in community-acquired pneumonia showing lobar consolidation (typical bacterial), diffuse bilateral infiltrates (atypical/viral), and cavitary lesion (S. aureus or anaerobes)</image>

## Risk Stratification

### PSI (Pneumonia Severity Index) / PORT Score

This 20-variable score incorporates age, comorbidities, physical exam findings, and laboratory values. Classes I through III indicate low risk (consider outpatient treatment), Class IV indicates moderate risk (short hospitalization or close outpatient follow-up), and Class V indicates high risk requiring hospitalization. It is better validated but may underestimate severity in young patients without comorbidities.

### CURB-65

| CURB-65 Criterion | Threshold |
|-------------------|-----------|
| Confusion | New mental confusion |
| Urea (BUN) | >7 mmol/L (>19 mg/dL) |
| Respiratory rate | ≥30 breaths/min |
| Blood pressure | SBP <90 or DBP ≤60 mmHg |
| Age | ≥65 years |

| CURB-65 Score | Disposition |
|--------------|-----------|
| 0-1 | Outpatient treatment |
| 2 | Consider short hospitalization |
| 3-5 | Hospitalization; ICU for score ≥4 |

This simpler tool scores Confusion, Urea above 7 mmol/L (BUN above 19), Respiratory rate of 30 or greater, Blood pressure (systolic below 90 or diastolic 60 or below), and age 65 or older. A score of 0-1 supports outpatient treatment, 2 warrants consideration of short hospitalization, and 3-5 requires hospitalization with ICU consideration for scores of 4 or higher.

### ATS/IDSA Criteria for Severe CAP (ICU Admission)

The major criteria (either one indicates severe CAP) are septic shock requiring vasopressors or need for mechanical ventilation. Minor criteria (3 or more indicate severe CAP) include respiratory rate of 30 or greater, PaO2/FiO2 of 250 or below, multilobar infiltrates, confusion, uremia (BUN 20 or above), leukopenia (WBC below 4,000), thrombocytopenia, hypothermia, and hypotension requiring aggressive fluids.

## Empiric Antibiotic Therapy (ATS/IDSA 2019)

### Outpatient — No Comorbidities

Options include amoxicillin 1 g three times daily, doxycycline 100 mg twice daily, or a macrolide (azithromycin) if local pneumococcal resistance is below 25%.

### Outpatient — With Comorbidities

For patients with chronic heart, lung, liver, or renal disease, diabetes, alcoholism, malignancy, or asplenia, a respiratory fluoroquinolone (levofloxacin 750 mg daily or moxifloxacin 400 mg daily) is appropriate, or alternatively amoxicillin-clavulanate 2 g twice daily plus a macrolide or doxycycline.

### Inpatient — Non-ICU

A beta-lactam (ceftriaxone 1-2 g IV daily, ampicillin-sulbactam 3 g IV every 6 hours, or cefotaxime) plus a macrolide (azithromycin), or respiratory fluoroquinolone monotherapy (levofloxacin 750 mg IV/PO daily).

### Inpatient — ICU

A beta-lactam (ceftriaxone or ampicillin-sulbactam) plus macrolide (azithromycin), or a beta-lactam plus respiratory fluoroquinolone. For penicillin allergy, a respiratory fluoroquinolone plus aztreonam is used.

### When to Add MRSA Coverage

Risk factors include prior MRSA isolation, recent hospitalization, severe necrotizing or cavitary pneumonia, and post-influenza pneumonia. Vancomycin or linezolid is added. If cultures are negative for MRSA at 48-72 hours and the patient is improving, de-escalation is appropriate. A nasal MRSA screen has high negative predictive value for MRSA pneumonia.

### When to Add Pseudomonas Coverage

Risk factors include structural lung disease (bronchiectasis, COPD with frequent exacerbations), prior Pseudomonas isolation, recent IV antibiotics, and immunosuppression. An anti-pseudomonal beta-lactam (piperacillin-tazobactam, cefepime, meropenem) is used, with de-escalation based on cultures.

<image>Empiric antibiotic selection for community-acquired pneumonia by treatment setting (outpatient, inpatient non-ICU, ICU) and risk factors for MRSA and Pseudomonas coverage</image>

## Corticosteroids in CAP

Growing evidence supports corticosteroids in severe CAP. The CAPE COD trial showed that dexamethasone 6 mg IV for 4 days reduced 28-day mortality in severe CAP requiring oxygen. Corticosteroids are not routinely recommended for non-severe CAP, and the benefit is most clear when CRP exceeds 150 mg/L. Corticosteroids are already standard in COVID-19 pneumonia requiring supplemental oxygen based on the RECOVERY trial.

## Antibiotic Duration

The standard duration is 5 days for most uncomplicated, non-severe CAP. Extension to 7 days is appropriate for severe CAP, slow clinical response, or complications (empyema, abscess). Criteria for stopping antibiotics include clinical stability (afebrile for 48 hours or more, improving symptoms, tolerating oral intake, and no more than one sign of instability). Procalcitonin-guided duration, stopping antibiotics when procalcitonin decreases by 80% or more from peak or drops below 0.25 mcg/L, reduces antibiotic duration without harm.

## Complications

Parapneumonic effusion or empyema requires thoracentesis if the effusion is significant. A pH below 7.2, positive Gram stain or culture, or loculated effusion indicates the need for chest tube drainage. Lung abscess requires prolonged antibiotics (4-6 weeks) with anaerobic coverage and drainage if large or refractory. Post-obstructive pneumonia should raise concern for underlying malignancy, especially in smokers or with non-resolving pneumonia.

## Follow-Up

Follow-up chest X-ray at 6 to 8 weeks is recommended for patients at risk of underlying malignancy (age over 50, smoking history). Clinical improvement is expected within 48 to 72 hours; if absent, broadening coverage, considering complications, and reassessing the diagnosis are warranted. Switching from IV to oral antibiotics when clinically stable and able to tolerate oral intake (early switch) is safe and encouraged.

<image>Procalcitonin-guided antibiotic duration algorithm showing initiation thresholds, monitoring intervals, and stopping criteria for community-acquired pneumonia</image>

## Clinical Pearls

A 5-day antibiotic course is sufficient for most CAP, and extending antibiotics "just in case" is low-value care that promotes resistance. Nasal MRSA PCR has greater than 95% negative predictive value for MRSA pneumonia and should be used to de-escalate vancomycin or linezolid early. Blood cultures should not be routinely obtained for low-risk CAP since they are rarely positive and infrequently change management. Fluoroquinolones should be reserved for patients with comorbidities or true beta-lactam allergies given their significant side effects (tendinopathy, QT prolongation, C. difficile, aortic dissection). Legionella should be specifically considered in severe CAP, especially with hyponatremia, GI symptoms, and failure to respond to beta-lactam monotherapy. Early mobilization and oral stepdown are key to reducing hospital length of stay.

## References
- Metlay JP, et al. Diagnosis and Treatment of Adults with CAP: ATS/IDSA 2019 Guideline. *Am J Respir Crit Care Med*. 2019;200:e45-e67.
- Dequin PF, et al. Hydrocortisone in Severe CAP (CAPE COD). *N Engl J Med*. 2023;388:1931-1941.
- Lim WS, et al. Defining CAP Severity on Presentation to Hospital: An International Derivation and Validation Study (CURB-65). *Thorax*. 2003;58:377-382.
- Schuetz P, et al. Effect of Procalcitonin-Guided Antibiotic Treatment on Mortality. *Lancet Infect Dis*. 2018;18:95-107.
- Fine MJ, et al. A Prediction Rule to Identify Low-Risk Patients with CAP (PSI/PORT). *N Engl J Med*. 1997;336:243-250.