Diagnosis and Severity Assessment of COPD Using a Novel Fast-Response Capnometer and Interpretable Machine Learning.

INTRODUCTION: Spirometry is the gold standard for COPD diagnosis and severity determination, but is technique-dependent, nonspecific, and requires administration by a trained healthcare professional. There is a need for a fast, reliable, and precise alternative diagnostic test. This study's aim was to use interpretable machine learning to diagnose COPD and assess severity using 75-second carbon dioxide (CO2) breath records captured with TidalSense's N-TidalTM capnometer. METHOD: For COPD diagnosis, machine learning algorithms were trained and evaluated on 294 COPD (including GOLD stages 1-4) and 705 non-COPD participants. A logistic regression model was also trained to distinguish GOLD 1 from GOLD 4 COPD with the output probability used as an index of severity. RESULTS: The best diagnostic model achieved an AUROC of 0.890, sensitivity of 0.771, specificity of 0.850 and positive predictive value (PPV) of 0.834. Evaluating performance on all test capnograms that were confidently ruled in or out yielded PPV of 0.930 and NPV of 0.890. The severity determination model yielded an AUROC of 0.980, sensitivity of 0.958, specificity of 0.961 and PPV of 0.958 in distinguishing GOLD 1 from GOLD 4. Output probabilities from the severity determination model produced a correlation of 0.71 with percentage predicted FEV1. CONCLUSION: The N-TidalTM device could be used alongside interpretable machine learning as an accurate, point-of-care diagnostic test for COPD, particularly in primary care as a rapid rule-in or rule-out test. N-TidalTM also could be effective in monitoring disease progression, providing a possible alternative to spirometry for disease monitoring.

Assessment of sidestream end-tidal capnography in ventilated infants on the neonatal unit.

OBJECTIVES: Continuous monitoring of carbon dioxide (CO2 ) levels can be achieved by capnography. Our aims were to compare the performance of a sidestream capnograph with a low dead space and sampling rate to a mainstream device and evaluate whether its results correlated with arterial/capillary CO2 levels in infants with different respiratory disease severities. WORKING HYPOTHESES: End-tidal carbon dioxide (EtCO2 ) results by sidestream and mainstream capnography would correlate, but the divergence of EtCO2 and CO2 results would occur in more severe lung disease. STUDY DESIGN: Prospective cohort study. PATIENT-SUBJECT SELECTION: Fifty infants with a median (interquartile range) gestational age of 31.1 (27.1-37.4) weeks and birth weight of 1.37 (0.76-2.95) kg. METHODOLOGY: Concurrent measurements of EtCO2 in ventilated infants were made using a new Microstream sidestream device and a mainstream capnograph (gold standard). Results from both devices were compared with arterial or capillary CO2 levels. The ratio of dead space to tidal volume (Vd/Vt) was calculated to assess respiratory disease severity. RESULTS: The mean difference between the concurrent measurements of EtCO2 was -0.54 ± 0.67 kPa (95% agreement levels - 1.86 to 0.77 kPa), the correlation between the two was r = .85 (P < .001). Sidestream capnography results correlated better with partial pressure of CO2 (PCO2 ) levels in infants with less (Vd/Vt < 0.35; r2 = .66, P < .001) rather than more severe (Vd/Vt > 0.35; r2 = .33, P = .01) lung disease. CONCLUSIONS: The sidestream capnography performed similarly to the mainstream capnography. The poorer correlation of EtCO2 to PCO2 levels in infants with severe respiratory disease should highlight to clinicians increased ventilation-perfusion mismatch.

Clinical and experimental validation of a capnodynamic method for end-expiratory lung volume assessment.

INTRODUCTION: Lung protective ventilation can decrease post-operative pulmonary complications. The aim of this study was to evaluate a capnodynamic method estimating effective lung volume (ELV) as a proxy for end-expiratory lung volume in response to PEEP changes in patients, healthy subjects and a porcine model. METHODS: Agreement and trending ability for ELV in anaesthetized patients and agreement in awake subjects were evaluated using nitrogen multiple breath wash-out/in and plethysmography as a reference respectively. Agreement and trending ability were evaluated in pigs during PEEP elevations with inert gas wash-out as reference. RESULTS: In anaesthetized patients bias (95% limits of agreement [LoA]) and percentage error (PE) at PEEP 0 cm H2 O were 133 mL (-1049 to 1315) and 71%, at PEEP 5 cm H2 O 161 mL (-1291 to 1613 mL) and 66%. In healthy subjects: 21 mL (-755 to 796 mL) and 26%. In porcines, at PEEP 5-20 cm H2 O bias decreased from 223 mL to 136 mL LoA (34-412) to (-30 to 902) and PE 29%-49%. Trending abilities in anaesthetized patients and porcines were 100% concordant. CONCLUSION: The ELV-method showed low bias but high PE in anaesthetized patients. Agreement was good in awake subjects. In porcines, agreement was good at lower PEEP levels. Concordance related to PEEP changes reached 100% in all settings. This method may become a useful trending tool for monitoring lung function during mechanical ventilation, if findings are confirmed in other clinical contexts.

Investigating Capnography Innovation for Better Patient Monitoring in the Resource Limited Surgical Setting.

BACKGROUND: Access to basic anesthetic monitoring in the developing world is lacking, which contributes to the 100 times greater anesthesia-related mortality in low- and middle-income countries. We hypothesize that an environmental sensor with a lower sampling rate could provide some clinical utility by providing CO2 levels, respiratory rate, and support in detection of clinical abnormalities. MATERIALS AND METHODS: A bench-top lung simulation was created to replicate CO2 waveforms, and an environmental sensor was compared with industry-available technology. Sensor response time and respiratory rates were compared between devices. Additionally, an in silico model was created to replicate capnography pathology as waveforms would appear using the environmental sensor. RESULTS AND CONCLUSION: Breath simulations using the bench-top lung simulation produced similar results to industry standards with a degree of variability. Respiratory rates did not differ between the environmental sensor and all other devices tested. Finally, pathological waveforms created in silico carried a certain level of detail regarding ventilatory pathology, which could provide some clinical insight to an anesthesiologist. We believe our prototype is the first step toward making low-cost and portable capnography available in the resource-limited setting, and future efforts should focus on bridging the gap to safer anesthesia and surgery globally.

Using continuous quantitative capnography for emergency department procedural sedation: a systematic review and cost-effectiveness analysis.

End-tidal CO2 has been advocated to improve safety of emergency department (ED) procedural sedation by decreasing hypoxia and catastrophic outcomes. This study aimed to estimate the cost-effectiveness of routine use of continuous waveform quantitative end-tidal CO2 monitoring for ED procedural sedation in prevention of catastrophic events. Markov modeling was used to perform cost-effectiveness analysis to estimate societal costs per prevented catastrophic event (death or hypoxic brain injury) during routine ED procedural sedation. Estimates for efficacy of capnography and safety of sedation were derived from the literature. This model was then applied to all procedural sedations performed in US EDs with assumptions selected to maximize efficacy and minimize cost of implementation. Assuming that capnography decreases the catastrophic adverse event rate by 40.7% (proportional to efficacy in preventing hypoxia), routine use of capnography would decrease the 5-year estimated catastrophic event rate in all US EDs from 15.5 events to 9.2 events (difference 6.3 prevented events per 5 years). Over a 5-year period, implementing routine end-tidal CO2 monitoring would cost an estimated $2,830,326 per prevented catastrophic event, which translates into $114,007 per quality-adjusted life-year. Sensitivity analyses suggest that reasonable assumptions continue to estimate high costs of prevented catastrophic events. Continuous waveform quantitative end-tidal CO2 monitoring is a very costly strategy to prevent catastrophic complications of procedural sedation when applied routinely in ED procedural sedations.

Assessment of Bohr and Enghoff Dead Space Equations in Mechanically Ventilated Children.

BACKGROUND: Recent findings suggest that using alveolar PCO2 (PACO2 ) estimated by volumetric capnography in the Bohr equation instead of PaCO2 (Enghoff modification) could be appropriate for the calculation of physiological dead space to tidal volume ratio (VD/VT Bohr and VD/VT Enghoff, respectively). We aimed to describe the relationship between these 2 measurements in mechanically ventilated children and their significance in cases of ARDS. METHODS: From June 2013 to December 2013, mechanically ventilated children with various respiratory conditions were included in this study. Demographic data, medical history, and ventilatory parameters were recorded. Volumetric capnography indices (NM3 monitor) were obtained over a period of 5 min preceding a blood sample. Bohr's and Enghoff's dead space, S2 and S3 slopes, and the S2/S3 ratio were calculated breath-by-breath using dedicated software (FlowTool). This study was approved by Ste-Justine research ethics review board. RESULTS: Thirty-four subjects were analyzed. Mean VD/VT Bohr was 0.39 ± 0.12, and VD/VT Enghoff was 0.47 ± 0.13 (P = .02). The difference between VD/VT Bohr and VD/VT Enghoff was correlated with PaO2 /FIO2 and with S2/S3. In subjects without lung disease (PaO2 /FIO2 ≥ 300), mean VD/VT Bohr was 0.36 ± 0.11, and VD/VT Enghoff was 0.39 ± 0.11 (P = .056). Two children with status asthmaticus had a major difference between VD/VT Bohr and VD/VT Enghoff in the absence of a low PaO2 /FIO2 . CONCLUSIONS: This study suggests that VD/VT Bohr and VD/VT Enghoff are not different when there is no hypoxemia (PaO2 /FIO2 > 300) except in the case of status asthmaticus. In subjects with a low PaO2 /FIO2 , the method to measure VD/VT must be reported, and results cannot be easily compared if the measurement methods are not the same.

Assessment of dead-space ventilation in patients with acute respiratory distress syndrome: a prospective observational study.

BACKGROUND: Physiological dead space (VD/VT) represents the fraction of ventilation not participating in gas exchange. In patients with acute respiratory distress syndrome (ARDS), VD/VT has prognostic value and can be used to guide ventilator settings. However, VD/VT is rarely calculated in clinical practice, because its measurement is perceived as challenging. Recently, a novel technique to calculate partial pressure of carbon dioxide in alveolar air (PACO2) using volumetric capnography (VCap) was validated. The purpose of the present study was to evaluate how VCap and other available techniques to measure PACO2 and partial pressure of carbon dioxide in mixed expired air (PeCO2) affect calculated VD/VT. METHODS: In a prospective, observational study, 15 post-cardiac surgery patients and 15 patients with ARDS were included. PACO2 was measured using VCap to calculate Bohr dead space or substituted with partial pressure of carbon dioxide in arterial blood (PaCO2) to calculate the Enghoff modification. PeCO2 was measured in expired air using three techniques: Douglas bag (DBag), indirect calorimetry (InCal), and VCap. Subsequently, VD/VT was calculated using four methods: Enghoff-DBag, Enghoff-InCal, Enghoff-VCap, and Bohr-VCap. RESULTS: PaCO2 was higher than PACO2, particularly in patients with ARDS (post-cardiac surgery PACO2 = 4.3 ± 0.6 kPa vs. PaCO2 = 5.2 ± 0.5 kPa, P < 0.05; ARDS PACO2 = 3.9 ± 0.8 kPa vs. PaCO2 = 6.9 ± 1.7 kPa, P < 0.05). There was good agreement in PeCO2 calculated with DBag vs. VCap (post-cardiac surgery bias = 0.04 ± 0.19 kPa; ARDS bias = 0.03 ± 0.27 kPa) and relatively low agreement with DBag vs. InCal (post-cardiac surgery bias = -1.17 ± 0.50 kPa; ARDS mean bias = -0.15 ± 0.53 kPa). These differences strongly affected calculated VD/VT. For example, in patients with ARDS, VD/VTcalculated with Enghoff-InCal was much higher than Bohr-VCap (VD/VT Enghoff-InCal = 66 ± 10 % vs. VD/VT Bohr-VCap = 45 ± 7 %; P < 0.05). CONCLUSIONS: Different techniques to measure PACO2 and PeCO2 result in clinically relevant mean and individual differences in calculated VD/VT, particularly in patients with ARDS. Volumetric capnography is a promising technique to calculate true Bohr dead space. Our results demonstrate the challenges clinicians face in interpreting an apparently simple measurement such as VD/VT.

Capnographic Monitoring of Moderate Sedation During Low-Risk Screening Colonoscopy Does Not Improve Safety or Patient Satisfaction: A Prospective Cohort Study.

OBJECTIVES: Appropriate monitoring during sedation has been recognized as vital to patient safety in procedures outside of the operating room. Capnography can identify hypoventilation prior to hypoxemia; however, it is not clear whether the addition of capnography improves safety or is cost effective during routine colonoscopy, a high volume, low-risk procedure. Our aim was to evaluate the value of EtCO2 monitoring during colonoscopy with moderate sedation. METHODS: We conducted a prospective study of sedation safety and patient satisfaction before and after the introduction of EtCO2 monitoring during outpatient colonoscopy with midazolam and fentanyl using the validated PROcedural Sedation Assessment Survey (PROSAS). Complications of sedation and PROSAS scores were compared among colonoscopies with and without capnography. RESULTS: A total of 966 patients participated in our study, 465 in the pre-EtCO2 group and 501 in the EtCO2 group. On multivariate analysis, patients and nurses reported higher levels of procedural discomfort after adoption of capnography (1.71 vs. 1.00, P<0.001). No serious adverse events were seen, and minor sedation-related adverse events occurred with similar frequency in both groups (8.2% pre-EtCO2 vs. 11.2% EtCO2, P=0.115). The cost of implementing EtCO2 in our unit was $40,169.95 and added $11.68 per case. CONCLUSIONS: Colonoscopy with moderate sedation is a low-risk procedure, and the addition of EtCO2 did not improve safety or patient satisfaction but did increase cost. These data suggest that routine capnography in this setting may not be cost effective and that EtCO2 might be reserved for patients at higher risk of adverse events.

The ¹³C bicarbonate method: an inverse end product method for measuring CO2 production and energy expenditure.

We reconsider the principle of the (13)C bicarbonate (NaH(13)CO3) method ((13)C-BM) for the determination of the CO2 production to obtain an estimate of energy expenditure (EE). Its mathematical concept based on a three-compartmental model is related to the [(15)N]glycine end product method. The CO2 production calculated by the (13)C-BM, RaCO2((13)C) is compared to the result from the indirect calorimetry, RCO2(IC). In an interspecies comparison (dog, goat, horse, cattle, children, adult human; body mass ranging from 15 to 350 kg, resting and fasting conditions) we found an excellent correlation between the results of (13)C-BM and IC with RCO2(IC) = 0.703 × RaCO2((13)C), (R(2) = 0.99). The slope of this correlation corresponds to the fractional (13)C recovery (RF((13)C)) of (13)C in breath CO2 after administration of NaH(13)CO3. Significant increase in RF((13)C) was found in physically active dogs (0.95 ± 0.14; n = 5) vs. resting dogs (0.71 ± 0.10, n = 17; p = .015). The (13)C recovery in young bulls was greater in blood CO2 (0.81 ± 0.05) vs. breath CO2 (0.73 ± 0.05, n = 12, p < .001) and in ponies with oral (0.76 ± 0.03, n = 8) vs. intravenous administration of NaH(13)CO3 (0.69 ± 0.07; n = 8; p = .026). We suggest considering the (13)C-BM as a 'stand-alone' method to provide information on the total CO2 production as an index of EE.

A novel continuous capnodynamic method for cardiac output assessment in a porcine model of lung lavage.

BACKGROUND: We have evaluated a new method for continuous monitoring of effective pulmonary blood flow (COEPBF ), i.e. cardiac output (CO) minus intra-pulmonary shunt, during mechanical ventilation. The method has shown good trending ability during severe hemodynamic challenges in a porcine model with intact lungs. In this study, we further evaluate the COEPBF method in a model of lung lavage. METHODS: COEPBF was compared to a reference method for CO during hemodynamic and PEEP alterations, 5 and 12 cmH2 O, before and after repeated lung lavages in 10 anaesthetised pigs. Bland-Altman, four-quadrant and polar plot methodologies were used to determine agreement and trending ability. RESULTS: After lung lavage at PEEP 5 cmH2 O, the ratio of arterial oxygen partial pressure related to inspired fraction of oxygen significantly decreased. The mean difference (limits of agreement) between methods changed from 0.2 (-1.1 to 1.5) to -0.9 (-3.6 to 1.9) l/min and percentage error increased from 34% to 70%. Trending ability remained good according to the four-quadrant plot (concordance rate 94%), whereas mean angular bias increased from 4° to -16° when using the polar plot methodology. CONCLUSION: Both agreement and precision of COEPBF were impaired in relation to CO when the shunt fraction was increased after lavage at PEEP 5 cmH2 O. However, trending ability remained good as assessed by the four-quadrant plot, whereas the mean polar angle, calculated by the polar plot, was wide.

Comparison of a self-inflating bulb syringe and a colorimetric CO2 indicator with capnography and radiography to detect the misdirection of naso/orogastric tubes into the airway of critically ill adult patients.

OBJECTIVE: The objective of this study was to develop a mechanism of discovering misdirection into the airway of naso/orogastric (NG) tubes before they reach their full depth of placement in adults. METHODS: A prospective, observational study was performed in humans, evaluating both the self-inflating bulb syringe (SIBS) and a colorimetric CO2 detector. A prospective convenience sample of 257 NG tube placements was studied in 199 patients in medical ICUs of a tertiary care medical center. Findings were compared to a "standard" (ie, end tidal CO2 results of a capnograph and the results of a chest radiograph performed at the completion of the tube placement). RESULTS: On the first tube placement attempt in any patient, the SIBS had a sensitivity of 91.5% and a specificity of 87.0% in detecting nonesophageal placement, while the colorimetric device exhibited 99.4% sensitivity and 91.3% specificity. On subsequent insertions, the SIBS showed 95.7% sensitivity and 100% specificity, while the colorimetric device exhibited 97.8% sensitivity and 100% specificity. The colorimetric device was eight times more expensive than the SIBS. CONCLUSIONS: The SIBS and the colorimetric CO2 detector are very good at detecting NG tube malpositioning into the airway, although the colorimetric device is slightly more sensitive and specific. Neither method adds substantial time or difficulty to the insertion process. The colorimetric device is substantially more expensive. The decision as to which method to use may be based on local institutional factors, such as expense.

Point of care ultrasound for orotracheal tube placement assessment in out-of hospital setting.

AIM OF THE STUDY: The percentage of unrecognised orotracheal tube displacement in an out-of-hospital setting has been reported to be between 4.8% and 25%. The aim of our study was to assess the sensitivity and specificity of Point-of-Care-UltraSound (POCUS) for confirming the proper tube position after an urgent orotracheal intubation in an out-of-hospital setting and the time needed for POCUS. METHODS: Our single-centred prospective study included all patients who needed out-of-hospital orotracheal intubation. After the intubation, bilateral chest auscultation and assessment of bilateral lung sliding and diaphragm excursion within POCUS were done. Spectrographic quantitative capnography was used as the reference standard to confirm a proper tube position. RESULTS: We enrolled 124 patients. For auscultation, sensitivity and negative predicted value were 100%, specificity was 90% and positive predicted value 30% (95% confidence interval). Sensitivity, specificity, positive predicted value, and negative predicted value for POCUS alone and for a combination of auscultation and POCUS were 100% (95% confidence interval). In three patients, we detected endobronchial tube displacement with auscultation and POCUS. Capnography failed to detect displacement in all three cases. The median time needed for POCUS was 30s. CONCLUSION: Results of our study support POCUS as an accurate and reliable method for confirming the proper orotracheal tube placement in trachea and it is feasible for out-of-hospital setting implementation. POCUS also seems to be time saving method but to make definitive conclusion more studies should be done.

Novel continuous capnodynamic method for cardiac output assessment during mechanical ventilation.

BACKGROUND: It is important to be able to accurately monitor cardiac output (CO) during high-risk surgery and in critically ill patients. The invasiveness of the pulmonary artery catheter (PAC) limits its use, and therefore, new minimally invasive methods for CO monitoring are needed. A potential method is estimation of CO from endogenous carbon dioxide measurements, using a differentiated Fick's principle to determine effective pulmonary blood flow (EPBF). In this study, we aimed to validate a novel capnodynamic method (COEPBF) in a wide range of clinically relevant haemodynamic conditions. METHODS: COEPBF was studied in 10 pigs during changes in preload, afterload, CO increase, and bleeding. An ultrasonic flow probe around the pulmonary artery was used as reference method of CO determination. CO was also measured using a PAC thermodilution technique (COPAC). CO and other haemodynamic data were recorded before and during each intervention. Accuracy and precision and also the ability to track changes in CO were determined using Bland-Altman, four-quadrant plot and polar plot analysis. RESULTS: COEPBF and COPAC showed equally good agreement, with a tendency to overestimate CO (bias 0.2 and 0.3 litre min(-1), respectively). The overall percentage error was 47% for COEPBF and 49% for COPAC. The concordance for tracking CO changes was 97 and 95% for COEPBF and COPAC, respectively, with an exclusion zone of 15% and radial limits of ±30°. CONCLUSIONS: COEPBF showed reliable trending abilities, equivalent to COPAC. COEPBF and COPAC also showed low bias but high percentage errors. Further studies in animal models of lung injury and in high-risk surgery patients are warranted.

Bench test assessment of mainstream capnography during high frequency oscillatory ventilation.

To assess the feasibility, stability and predictability of pCO2 measurement (PETCO2) using a main stream capnograph in a high frequency oscillatory ventilation circuit. A commercially available capnograph was mounted into a high frequency oscillatory ventilator patient circuit, adjustable CO2 flow was introduced into an artificial lung and the output of the CO2 sensor assessed under varying ventilator settings. Influence of oxygen content, pressures, heat and moisture were recorded. A linear relationship between CO2 flow rate and PETCO2 was found. Varying ventilator settings influenced the measurements, but the results for PETCO2 remained within a range of 1.5 mmHg above or under then mean measurement value. Measurements remained stable despite humidification, heat, pressure amplitudes or mean airway pressure changes. From this bench test, we conclude it is feasible to measure PETCO2 using a main stream capnograph during high frequency oscillatory conditions, these measurements were stable during the experiment. Changes in CO2 production or output can be detected. The system may prove to be of clinical value, but further in vivo measurements are warranted.

Assessment and confirmation of tracheal intubation when capnography fails: a novel use for an USB camera.

A 62 year old male with a right pyriform fossa lesion extending to the right arytenoid and obscuring the glottic inlet was planned for laser assisted excision. Direct laryngoscopic assessment after topicalization of the airway, showed a Cormack Lehane grade 3 view. We report a case where, in the absence of a fiberscope, a novel inexpensive Universal Serial Bus camera was used to obtain an optimal laryngoscopic view. This provided direct visual confirmation of tracheal intubation with a Laser Flex tube, when capnography failed to show any trace. Capnography may not be reliable as a sole indicator of confirmation of correct endotracheal tube placement. Video laryngoscopy may provide additional confirmation of endotracheal intubation.

Disparity between mainstream and sidestream end-tidal carbon dioxide values and arterial carbon dioxide levels.

BACKGROUND: Measuring and monitoring end-tidal carbon dioxide (PETCO2) is an important aspect of caring for critically ill patients. The 2 methods used for PETCO2 measurement are the mainstream and sidestream methods. OBJECTIVE: To assess the agreement between PETCO2 measurements performed by mainstream and sidestream methods with the PaCO2 values. METHODS: This was a prospective observational study. A total of 114 subjects were enrolled in the study. PETCO2 measurements using mainstream and sidestream methods were performed simultaneously with the arterial blood sampling in subjects who were observed in the emergency department and required arterial blood gas analysis. Agreement between the PETCO2 measurements and the PaCO2 values obtained from arterial blood gas analysis were evaluated using the Bland-Altman method. RESULTS: Sixty subjects (52.6%) were female, and the mean age was 60.9 years (95% CI 58.3-63.6). The mean PaCO2 was 35.16 mm Hg (95% CI 33.81-36.51), the mainstream PETCO2 was 22.11 (95% CI 21.05-23.18), and the sidestream PETCO2 was 25.48 (95% CI 24.22-26.75). Bland-Altman analysis showed an average difference between mainstream PETCO2 and PaCO2 values of 13 mm Hg (95% limits of agreement -0.6 to 25.5) and moderate correlation (r = 0.55, P < .001). The average difference between the sidestream PETCO2 and PaCO2 values was 9.7 mm Hg (95% limits of agreement -5.4 to 24.7) and poor correlation (r = 0.41, P < .001). CONCLUSIONS: PETCO2 values obtained by mainstream and sidestream methods were found to be significantly lower than the PaCO2 values. There was essentially no agreement between the measurements obtained by 2 different methods and the PaCO2 values.

Combination of quantitative capnometry, N-terminal pro-brain natriuretic peptide, and clinical assessment in differentiating acute heart failure from pulmonary disease as cause of acute dyspnea in pre-hospital emergency setting: study of diagnostic accuracy.

AIM: To determine the diagnostic accuracy of the combination of quantitative capnometry (QC), N-terminal pro-brain natriuretic peptide (NT-proBNP), and clinical assessment in differentiating heart failure (HF)-related acute dyspnea from pulmonary-related acute dyspnea in a pre-hospital setting. METHODS: This prospective study was performed in the Center for Emergency Medicine Maribor, Slovenia, January 2005-June 2007. Two groups of patients with acute dyspnea apnea were compared: HF-related acute dyspnea group (n = 238) vs pulmonary-related acute dyspnea (asthma/COPD) group (n = 203). The primary outcome was the comparison of combination of QC, NT-proBNP, and clinical assessment vs NT-proBNP alone or NT-proBNP in combination with clinical assessment, in differentiating HF-related acute dyspnea from pulmonary-related acute dyspnea (asthma/COPD) in pre-hospital emergency setting, using the area under the receiver operating characteristic curve (AUROC). The secondary outcomes end points were identification of independent predictors for final diagnosis of acute dyspnea (caused by acute HF or pulmonary diseases), and determination of NT-proBNP levels, as well as capnometry, in the subgroup of patients with a previous history of HF and in the subgroup of patients with a previous history of pulmonary disease. RESULTS: In differentiating between cardiac and respiratory causes of acute dyspnea in pre-hospital emergency setting, NT-proBNP in combination with PetCO2 and clinical assessment (AUROC, 0.97; 95% confidence interval [CI], 0.90-0.99) was superior to combination of NT-proBNP and clinical assessment (AUROC, 0.94; 95% CI, 0.88-0.96; P = 0.006) or NT-proBNP alone (AUROC, 0.90; 95% CI, 0.85-0.94; P = 0.005). The values of NT-proBNP> or = 2000 pg/mL and PetCO2 < or = 4 kPa were strong independent predictors for acute HF. In the group of acute HF dyspneic patients, subgroup of patients with previous COPD/asthma had significantly higher PetCO2 (3.8 +/- 1.2 vs 5.8 +/- 1.3 kPa, P = 0.009). In the group of COPD/asthma dyspneic patients, NT-proBNP was significantly higher in the subgroup of patients with previous HF (1453.3 +/- 552.3 vs 741.5 +/- 435.5 pg/mL, P = 0.010). CONCLUSION: In differentiating between cardiac and respiratory causes of acute dyspnea in pre-hospital emergency setting, NT-proBNP in combination with capnometry and clinical assessment was superior to NT-proBNP alone or NT-proBNP in combination with clinical assessment.