Use of capnography for prediction of obstruction severity in non-intubated COPD and asthma patients

Pertzov, Barak; Ronen, Michal; Rosengarten, Dror; Shitenberg, Dorit; Heching, Moshe; Shostak, Yael; Kramer, Mordechai R.

doi:10.1186/s12931-021-01747-3

Cited by 10 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After adjusting for potential confounders, we found that the parameters that could distinguish between mild/moderate and severe cases were vascular disorders, HDL, plasma fibrinogen, fructosamine, standard bicarbonate concentration, pCO 2 , age, eosinophil count, lymphocyte ratio, and apolipoprotein A1. The factors that could distinguish between mild/moderate and very severe cases were vascular disorders, HDL, plasma fibrinogen, fructosamine, pO 2 , plasma carbon dioxide concentration, standard bicarbonate concentration, pCO 2 , age, sex, allergic diseases, eosinophil count, lymphocyte (14,17). For instance, Chen et al developed a model for predicting disease severity in patients hospitalized for COPD exacerbation and found that neutrophil count percentage and demographic parameters were associated with a higher risk of COPD exacerbation; the area under the receiver operating characteristic curve was 0.84 (14).…”

Section: Discussionmentioning

confidence: 99%

Development of a model for predicting the severity of chronic obstructive pulmonary disease

Chen²,

Qiu³

et al. 2022

Front. Med.

View full text Add to dashboard Cite

BackgroundSeveral models have been developed to predict the severity and prognosis of chronic obstructive pulmonary disease (COPD). This study aimed to identify potential predictors and construct a prediction model for COPD severity using biochemical and immunological parameters.MethodsA total of 6,274 patients with COPD were recruited between July 2010 and July 2018. COPD severity was classified into mild, moderate, severe, and very severe based on the Global Initiative for Chronic Obstructive Lung Disease guidelines. A multivariate logistic regression model was constructed to identify predictors of COPD severity. The predictive ability of the model was assessed by measuring sensitivity, specificity, accuracy, and concordance.ResultsOf 6,274 COPD patients, 2,644, 2,600, and 1,030 had mild/moderate, severe, and very severe disease, respectively. The factors that could distinguish between mild/moderate and severe cases were vascular disorders (OR: 1.44; P < 0.001), high-density lipoprotein (HDL) (OR: 1.83; P < 0.001), plasma fibrinogen (OR: 1.08; P = 0.002), fructosamine (OR: 1.12; P = 0.002), standard bicarbonate concentration (OR: 1.09; P < 0.001), partial pressure of carbon dioxide (OR: 1.09; P < 0.001), age (OR: 0.97; P < 0.001), eosinophil count (OR: 0.66; P = 0.042), lymphocyte ratio (OR: 0.97; P < 0.001), and apolipoprotein A1 (OR: 0.56; P = 0.003). The factors that could distinguish between mild/moderate and very severe cases were vascular disorders (OR: 1.59; P < 0.001), HDL (OR: 2.54; P < 0.001), plasma fibrinogen (OR: 1.10; P = 0.012), fructosamine (OR: 1.18; P = 0.001), partial pressure of oxygen (OR: 1.00; P = 0.007), plasma carbon dioxide concentration (OR: 1.01; P < 0.001), standard bicarbonate concentration (OR: 1.13; P < 0.001), partial pressure of carbon dioxide (OR: 1.16; P < 0.001), age (OR: 0.91; P < 0.001), sex (OR: 0.71; P = 0.010), allergic diseases (OR: 0.51; P = 0.009), eosinophil count (OR: 0.42; P = 0.014), lymphocyte ratio (OR: 0.93; P < 0.001), and apolipoprotein A1 (OR: 0.45; P = 0.005). The prediction model correctly predicted disease severity in 60.17% of patients, and kappa coefficient was 0.35 (95% CI: 0.33–0.37).ConclusionThis study developed a prediction model for COPD severity based on biochemical and immunological parameters, which should be validated in additional cohorts.

show abstract

Section: Discussionmentioning

confidence: 99%

Development of a model for predicting the severity of chronic obstructive pulmonary disease

Chen²,

Qiu³

et al. 2022

Front. Med.

View full text Add to dashboard Cite

show abstract

“…These features included the following: α, β, γ, and δ angles (Fig. 2 ) [ 11 – 13 ]; gradients and residuals derived from fitting curves to phases, such as the expiratory plateau [ 14 ]; absolute and short-term variability of pCO 2 [ 15 ]; curvature and other higher-order time-based features such as the ratio of the expiratory to inspiratory phase [ 12 , 13 , 16 ]; and area under the curve (AUC), which is commonly calculated in volumetric capnography [ 16 , 17 ]. These features have a basis in the respiratory physiology literature, and in many cases have been hypothesized to relate to clinical airway obstruction.…”

Section: Methodsmentioning

confidence: 99%

Machine diagnosis of chronic obstructive pulmonary disease using a novel fast-response capnometer

Talker¹,

Neville

Wiffen

et al. 2023

Respir Res

View full text Add to dashboard Cite

Background Although currently most widely used in mechanical ventilation and cardiopulmonary resuscitation, features of the carbon dioxide (CO2) waveform produced through capnometry have been shown to correlate with V/Q mismatch, dead space volume, type of breathing pattern, and small airway obstruction. This study applied feature engineering and machine learning techniques to capnography data collected by the N-Tidal™ device across four clinical studies to build a classifier that could distinguish CO2 recordings (capnograms) of patients with COPD from those without COPD. Methods Capnography data from four longitudinal observational studies (CBRS, GBRS, CBRS2 and ABRS) was analysed from 295 patients, generating a total of 88,186 capnograms. CO2 sensor data was processed using TidalSense’s regulated cloud platform, performing real-time geometric analysis on CO2 waveforms to generate 82 physiologic features per capnogram. These features were used to train machine learning classifiers to discriminate COPD from ‘non-COPD’ (a group that included healthy participants and those with other cardiorespiratory conditions); model performance was validated on independent test sets. Results The best machine learning model (XGBoost) performance provided a class-balanced AUROC of 0.985 ± 0.013, positive predictive value (PPV) of 0.914 ± 0.039 and sensitivity of 0.915 ± 0.066 for a diagnosis of COPD. The waveform features that are most important for driving classification are related to the alpha angle and expiratory plateau regions. These features correlated with spirometry readings, supporting their proposed properties as markers of COPD. Conclusion The N-Tidal™ device can be used to accurately diagnose COPD in near-real-time, lending support to future use in a clinical setting. Trial registration: Please see NCT03615365, NCT02814253, NCT04504838 and NCT03356288.

show abstract

“…As the number of breaths per capnogram varied, the median and standard deviation were calculated for the per-breath features in each capnogram. These features included the following: α, β, γ , and δ angles (Figure 2) [11, 12, 13]; gradients and residuals derived from fitting curves to phases, such as the expiratory plateau [14]; absolute and short-term variability of pCO 2 [15]; curvature and other higher-order time-based features such as the ratio of the expiratory to inspiratory phase [12, 13, 16]; and area under the curve (AUC), which is commonly calculated in volumetric capnography [16, 17]. These features have a basis in the respiratory physiology literature, and in many cases have been hypothesized to relate to clinical airway obstruction.…”

Section: Methodsmentioning

confidence: 99%

Machine Diagnosis of Chronic Obstructive Pulmonary Disease using a Novel Fast-Response Capnometer

Talker¹,

Neville

Wiffen

et al. 2023

Preprint

View full text Add to dashboard Cite

Background: Although currently most widely used in mechanical ventilation and cardiopulmonary resuscitation, features of the carbon dioxide waveform produced through capnometry have been shown to correlate with V/Q mismatch, dead space volume, type of breathing pattern, and small airway obstruction. This study applied feature engineering and machine learning techniques to capnography data collected by the N-TidalTMdevice across four clinical studies to build a classifier that could distinguish CO2recordings (capnograms) of patients with COPD from those without COPD. Methods: Capnography data from four longitudinal observational studies (CBRS, GBRS, CBRS2 and ABRS) was analysed from 295 patients, generating a total of 88,186 capnograms. CO2sensor data was processed using Cambridge Respiratory Innovations′ regulated cloud platform, performing realtime geometric analysis on CO2waveforms to generate 82 physiologic features per capnogram. These features were used to train machine learning classifiers to discriminate COPD from 'non-COPD' (a group that included healthy participants and those with other cardiorespiratory conditions); model performance was validated on independent test sets. Results: The best machine learning model (XGBoost) performance provided a class−balanced AUROC of 0.968 ± 0.017 and a positive predictive value (PPV) of 0.911 ± 0.028 for a diagnosis of COPD. The waveform features that are most important for driving classification are related to the alpha angle and expiratory plateau regions. These features correlated with spirometry readings, supporting their proposed properties as markers of COPD. Conclusion: The N-Tidal device can be used to accurately diagnose COPD in near-real-time, lending support to future use in a clinical setting.

show abstract

Use of capnography for prediction of obstruction severity in non-intubated COPD and asthma patients

Cited by 10 publications

References 22 publications

Development of a model for predicting the severity of chronic obstructive pulmonary disease

Development of a model for predicting the severity of chronic obstructive pulmonary disease

Machine diagnosis of chronic obstructive pulmonary disease using a novel fast-response capnometer

Machine Diagnosis of Chronic Obstructive Pulmonary Disease using a Novel Fast-Response Capnometer

Contact Info

Product

Resources

About