End-tidal carbon dioxide measurement in expiration air in the diagnosis of pulmonary emboly and assessment of the weight of the disease

Objective: Pulmonary embolism (PE) is still one of the leading causes of cardiovascular death worldwide. Due to the variability in the clinical symptoms of PE, difficulties may be experienced in the diagnosis of the disease. In some studies, it has been reported that alveolar dead space indices (PCO₂-ETCO₂ gradient) calculated by end-tidal carbondioxide (ETCO₂) level measurement with capnography are a practical and supportive method for confirming or excluding the diagnosis of PE. In our study, we aimed to evaluate the role of alveolar dead space indices calculated with ETCO₂ measured in expiratory air with a capnography device in the diagnosis of PE, the determination of the severity of the disease and the follow-up of the clinical course in patients with suspected symptoms and signs of PE. Material and Method: All consecutive patients older than 18 years of age who were evaluated in the emergency and in patient services of Ondokuz Mayıs University Medical Faculty Hospital between September 2022 and February 2023 with suspected clinical findings of PE such as dyspnea, tachycardia, hypotension, syncope, and chest pain and who signed with informed written consent was included in the study. Pregnant patients, patients with Type 2 respiratory failure, chronic obstructive pulmonary disease, and patients who did not give written consent were excluded from our study. In addition to routine tests (complete blood count, biochemistry, D-dimer, ECG, arterial blood gas, echocardiography, extremity deep vein Doppler ultrasonography, thorax CT angiography) performed in all patients with suspected PE according to Wells criteria and clinical evaluation, end-tidal carbondioxide measurement, ETCO₂ measurements were made in expiratory air with Nelcor Capnostream 35 Portable Respiratory Monitorand Smart CapnoLine Guardian adult carbondioxide nasal pressure cannula working with the sidestream method. Patients who were excluded from the diagnosis of PE (n=50) were determined as the control group. Patients with confirmed PE diagnosis (n=50) were divided into 3 groups as massive, submassive and nonmassive embolism. ETCO₂ measurements were performed on the patients diagnosed with PE at the time of diagnosis (on day 1) and on day 7, and the PCO₂-ETCO₂ gradientand alveolar dead space indices were calculated. Results: 100 patients with suspected PE were included in the study. CT angiography was performed in all of these patients and 50 patients were diagnosed with PE (Group 1), and filling defects compatible with PE were not observed in the other 50 patients (Group 2). Of the patients (Group 1 + Group 2), 45% were male (n=45) and 55% were female (n=55), and the mean age was 63±15 years. The most common presenting symptoms in all patients were dyspnea and tachycardia. The most common additional chronic diseases in Group 1 + Group 2 patients were malignancy, hypertension, diabetes mellitus, coronary artery disease, chronic renal failure and congestive heart failure. The most common predisposing condition in all patients was surgery and immobilization in the last 1 month. DVT was detected in 26 of the patients in Group 1 and Group 2. Twenty-four of the patients with DVT were in Group 1 and 2 patients were in Group 2. Eighteen of 26 patients with DVT had clinical signs of DVT. In the group with PE (Group 1); Filling defects were detected in the main pulmonary artery in 18 patients, in the lobar branch in 15 patients, in the segmental branch in 16 patients, and in the subsegmental branch in 1 patient. Mean PAP; It was 31±14 mmHg in the group with PE (Group 1) and 26±12 mmHg in the group without PE (Group 2). As initial treatment, LMWH was started in 44 of the patients in the PE group (Group 1), and r-tPA was started in 6 of them. In maintenance treatment, LMWH was started in 33 patients, vitamin K antagonists in 15 patients, and NOACs in 2 patients. In the group with PE (Group 1), the mean ETCO₂ on Day 1 was 22.78 ± 5.14 mmHg, and the mean ETCO₂ on Day 7 was 30.87 ± 8.76 mmHg (p<0.05). In the group with PE (Group 1), the mean PaCO₂ - ETCO₂ gradient at the time of diagnosis was 11.48 ± 6.95 mmHg. In the group with PE (Group 1), the mean AVDSf at the time of diagnosis was 310 ± 160 mL. In the group without PE (Group 2), the mean ETCO₂ on Day 1 was 31.82 ± 5.58 mmHg. In the group without PE (Group 2), the mean PaCO₂ - ETCO₂ gradient was found to be 6.5 ± 4.84 mmHg. In the non-PE group (Group 2), the mean AVDSf at the time of diagnosis was 160 ± 110 mL. There was a significant difference between Day 1 ETCO₂, PaCO₂ - ETCO₂ gradient and AVDSf in the group with and without PE (p<0.05). In the massive PE group, the mean ETCO₂ on Day 1 was 15.66 ± 3.21 mmHg, and the mean ETCO₂ on Day 7 was 33.00 ± 4.58 mmHg (p:0.003). In the submassive PE group, the mean ETCO₂ on Day 1 was 20.90 ± 4.17 mmHg, and the mean ETCO₂ on Day 7 was 27.68 ± 11.65 mmHg (p:0.036). In the nonmassive PE group, the mean ETCO₂ on Day 1 was 25.28 ± 4.67 mmHg, and the mean ETCO₂ on Day 7 was 33.54 ± 4.07 mmHg (p<0.05). The best cut-off point for ETCO2 measured to support or exclude PE in Group 1 + Group 2 patients was found to be 26.5 mmHg (78% sensitivity and 86% specificity). The positive predictive value for this value was 84.78%, while the negative predictive value was 79.63% (AUC: 0.881, p value<0.05). The combination of ETCO₂ and D-dimer and the combination of ETCO₂ and Wells score were found to be significant to exclude the disease with a specificity rate of 86%. There was no significant difference between the use of ETCO₂ alone and the use of ETCO₂ in combination with D-dimer or Wells score to exclude disease. The best cut-off point for the PaCO₂-ETCO₂ gradient to support or exclude disease was found to be 8.5 mmHg (66% sensitivity and 76% specificity). Negative predictive value was 74.47%, positive predictive value was 71.70% (AUC: 0.727 and p value<0.05). The best cut-off point for alveolar dead space fraction (AVDSf) to support or exclude disease was found to be 230 mL (74% sensitivity, 76% specificity). Negative predictive value was 74.47% and positive predictive value was 71.70% (AUC:0.778 p value<0.05). For the combination of ETCO₂ and AVDSf, the sensitivity was 80% and the specificity was 84% (p<0.05). When the ETCO₂ values of the patients in the PE group (Group 1) were examined before the treatment (day 1) and after the treatment (day 7), it was observed that there was a significant improvement in the ETCO₂ values of the patients on the 7th day after the treatment. When the pre-treatment and 7th day values were compared in both groups of patients with PE (Group 1) who received LMWH and rt-PA treatment, it was observed that there was an improvement in ETCO2 values with treatment in both groups (p<0.05). Conclusion: The results of our study show that ETCO2 measured by capnography can be used together with Wells scoring and D-dimer to support or exclude the diagnosis in patients with suspected PE. ETCO2 measurements, which are made with an inexpensive and practical method, can be used as an auxiliary method in the treatment and clinical course follow-up of patients diagnosed with PE and in the evaluation of the severity of the disease. We think that new prospective studies with larger patient numbers are needed to evaluate the use of capnography in PE. Keywords: pulmonary embolism, capnography, end-tidal carbondioxide (ETCO₂), alveolar dead space fraction (AVDSf)