ETCO2 for Predicting Pulmonary Embolism
ETCO2 measurement is NOT recommended as a standalone screening tool to exclude pulmonary embolism and has no role in current evidence-based diagnostic algorithms. Major international guidelines from the European Society of Cardiology and American College of Physicians do not include ETCO2 in their diagnostic pathways for PE 1.
Why ETCO2 Is Not Part of Standard PE Diagnosis
The established diagnostic approach for PE relies on validated clinical prediction rules (Wells score, Geneva score, PERC criteria), D-dimer testing, and imaging with CTPA or V/Q scanning 1. ETCO2 measurement is conspicuously absent from all Class I recommendations in the 2019 ESC Guidelines and the 2015 American College of Physicians Best Practice Advice 1.
The recommended diagnostic strategy is:
- Low-risk patients: Apply PERC criteria; if not all met, obtain D-dimer (age-adjusted cutoff: age × 10 ng/mL for patients >50 years) 1
- Intermediate-risk patients: Obtain D-dimer first, then CTPA if elevated 1
- High-risk patients: Proceed directly to CTPA or bedside echocardiography without D-dimer 1
Research Evidence on ETCO2: Limited and Inconsistent
While several small studies have explored ETCO2 for PE diagnosis, the results are inconsistent and insufficient to change practice:
Studies Suggesting Potential Utility:
- A 2010 study (n=298) found ETCO2 ≥36 mmHg had 96.6% negative predictive value, increasing to 97.6% when combined with Wells score <4 2
- A 2014 UK study (n=100) reported 100% sensitivity at ETCO2 <4.3 kPa (32.3 mmHg), but only 68% specificity 3
- A 2019 study (n=100) found ETCO2 >34 mmHg combined with clinical probability had 100% sensitivity and NPV 4
Studies Showing Poor Performance:
- A 2016 study (n=159) found ETCO2 had only 61.5% specificity with AUC of 0.751, concluding diagnostic value was "low" 5
- A 2020 study demonstrated variable cutoff points (28.5 mmHg) depending on clinical scoring system used 6
Critical Limitations of ETCO2 for PE
The fundamental problem is that ETCO2 alone cannot reliably exclude PE 4. Key limitations include:
- Poor specificity: Even in the most optimistic studies, specificity ranges from 53-73%, meaning high false-positive rates 2, 5
- Variable cutoff points: Different studies report optimal cutoffs ranging from 28.5 to 36 mmHg, indicating lack of standardization 4, 6, 2, 3
- Small sample sizes: All studies are single-center with <300 patients, insufficient for guideline-level recommendations 4, 6, 2, 5, 3
- No prospective validation: None of these studies have been validated in large multicenter trials 2
What You Should Use Instead
Follow the validated diagnostic algorithm recommended by ESC and ACP 1:
Assess pretest probability using Wells score, Geneva score, or clinical gestalt 1
For low-risk patients: Apply PERC criteria (all 8 must be met to avoid testing); if not all met, obtain high-sensitivity D-dimer 1
For intermediate-risk patients: Obtain D-dimer with age-adjusted cutoff (age × 10 ng/mL for patients >50 years); if elevated, proceed to CTPA 1
For high-risk patients with hemodynamic instability: Proceed directly to bedside echocardiography or emergency CTPA without D-dimer testing 1
D-dimer is NOT recommended in high clinical probability patients, as normal results don't safely exclude PE even with highly sensitive assays 1
Common Pitfalls to Avoid
- Do not use ETCO2 as a substitute for validated clinical prediction rules and D-dimer testing 1
- Do not measure D-dimer in hospitalized patients with multiple comorbidities (infection, inflammation, chronic kidney disease), as these cause persistently elevated levels 7
- In dialysis patients, skip D-dimer entirely and proceed directly to CTPA, as D-dimer is essentially non-diagnostic in this population 7
- Do not obtain CT venography as an adjunct to CTPA (Class III recommendation) 1
The evidence-based approach prioritizes validated clinical prediction rules, selective D-dimer testing with age-adjusted cutoffs, and appropriate imaging—not ETCO2 measurement 1.