Pulmonology

Value of integrated pulmonary index to predict exacerbation of chronic obstructive pulmonary Disease’s severity

a b s t r a c t

Purpose: In this study, we aimed to examine the correlation between current prognostic scores and the Integrated Pulmonary Index in patients admitted to the emergency department (ED) with exacerbation of chronic ob- structive pulmonary disease (COPD), and the diagnostic value of using the IPI in combination with other scores in determining patients who can be discharged safely.

Methods: This study was conducted as a multicenter and prospective observational study between August 2021 and June 2022. Patients diagnosed with COPD exacerbation (eCOPD) at the ED were included in the study and they were grouped according to the Global Initiative for Chronic obstructive lung disease (GOLD) classification. The CURB-65 (Confusion, Urea, Respiratory rate, Blood pressure, and age older than 65 years), BAP-65 (Blood urea nitrogen, Altered mental status, Pulse rate, and age older than 65 years), and DECAF (Dyspnea, Eosinopenia, Consolidation, Academia, and atrial Fibrillation) scores and IPI values of the patients were recorded. The correla- tion between the IPI and the other scores and its diagnostic value in detecting mild eCOPD were examined. The diagnostic value of CURB-IPI, a new score created by the combination of CURB-65 and IPI, in mild eCOPD was examined.

Results: The study was carried out with 110 patients (49 female and 61 male), mean age of 67 (min/max: 40/97). The IPI and CURB-65 had better predictive value in detecting mild exacerbations than DECAF and BAP-65 scores [Area under curves (AUC) were 0.893, 0.795, 0.735, 0.541 respectively]. The CURB-IPI score, on the other hand, had the best predictive value for detecting mild exacerbations (AUC 0.909).

Conclusion: We found that the IPI has good predictive value in the detection of mild COPD exacerbations, and its predictive value increases when used in combination with CURB-65. We think that the CURB-IPI score can be a guide when deciding whether patients with exacerbation of COPD can be discharged.

(C) 2023

  1. Introduction

An exacerbation of chronic obstructive pulmonary disease (eCOPD) is defined as an acute worsening of respiratory symptoms that results in additional therapy [1]. Many patients who experience eCOPD often require assessment in an emergency department (ED) and need hospi- talization. Prognostic severity scores enable ED physicians to identify eCOPD patients who can be discharged safely and who are at higher risk of an adverse evolution and are thus more likely to need intensive treatment [2]. However, because of the complicated nature of eCOPD, using only one biomarker or clinical parameter for prognosis evaluation is insufficient [3]. Therefore, some complex Clinical scores are used in clinical practice for risk classification of eCOPD.

* Corresponding author.

E-mail address: [email protected] (F.N. Karaarslan).

The CURB-65 (confusion, urea, respiratory rate, blood pressure, and age older than 65 years), BAP-65 (blood urea nitrogen, altered mental status, pulse rate, and age older than 65 years), and most recently the DECAF (dyspnea, eosinopenia, consolidation on chest X-ray, acidemia, and atrial fibrillation) scores are among these complex Clinical scores [4]. The integrated pulmonary index (IPI) is a mathematical algorithm that combines end tidal CO2 (PetCO2), respiration rate (RR), Arterial oxygen saturation (SpO2), and pulse rate (PR) into a single value between 1 and 10 and simplifies the interpretation of these parameters to assess a patient’s respiratory status [5]. Recent studies have demonstrated that the IPI is a good predictor for evaluating the ad- equacy of Ventilation and oxygenation in a noninvasive, dynamic way [6,7]. Because the severity of the disease is directly correlated with the respiratory status of the patient, IPI, which enables interpretation of re- spiratory status in one single value, may be helpful in predicting the clinical severity of eCOPD.

https://doi.org/10.1016/j.ajem.2023.05.043

0735-6757/(C) 2023

This study therefore aimed to evaluate the value of the IPI in patients with eCOPD to predict the severity of exacerbations. The secondary aim was to analyze the correlation between the IPI and the aforementioned prognostic scores in this patient group.

  1. Methods
    1. Study design

This was a prospective study that was conducted between August 1, 2021 and June 1, 2022 in the EDs of 2 training and research hospitals, 1 of which is a specialized respiratory center, with the approval of the local ethics committee.

    1. Subjects

Patients admitted to the ED with an Initial diagnosis of eCOPD were consecutively enrolled. The COPD diagnosis in the enrolled patients was confirmed via hospital records of the results of spirometry tests, which were previously performed in outpatient clinics. Exacerbation was diag- nosed if there was an acute deterioration in Respiratory functions and a need for additional therapy based on the clinical history and examina- tion findings of patients with COPD [1]. The participants were informed about the goals of the study and signed an informed consent form. All information was kept confidential.

Those who refused to give consent or who had eCOPD complicated by pneumothorax, pulmonary embolism, or Lung cancer were excluded. The other exclusion criterion was additional illness as a final diagnosis, such as asthma, bronchiectasis, or interstitial lung disease; acute coro- nary syndrome; acute decompensated heart failure; end-stage renal disease; or terminal malignancy (Fig. 1).

Image of Fig. 1

Fig. 1. Flow chart.

    1. Data collection

Data concerning patient demographics, medical and medication his- tory, and vital signs on arrival were recorded. Clinical information, vital signs, and electrocardiogram (ECG), laboratory, and radiographic data at admission were used to compile the CURB-65, BAP-65, and DECAF scores.

For the CURB-65 score, 1 point each was given for the presence of confusion, urea > 7 mmol/L, respiratory rate ?30/min, low systolic (<90 mmHg) or diastolic (?60 mmHg) blood pressure, and age

?65 years [8]. For the BAP-65 score, 1 point each was given for the pres- ence of a blood urea level > 8.9 mmol/L, altered mental status, and a pulse rate ?109 beats per minute. BAP score/class I was given when none of the criteria were met and age was <65 years, BAP score/class

II was given when none of the criteria were met but age was

?65 years, and BAP score/class III, IV, and V were given when the BAP score was 1, 2, or 3, respectively, regardless of age [9]. For the DECAF score, 1 point each was given for the presence of dyspnea (too breath- less to leave the house but independent in washing or dressing, or 2 points for being too breathless to be independent in washing or dress- ing), eosinopenia (eosinophil count <0.1 x 109/L), consolidation on chest X-ray, acidemia (pH < 7.30), and atrial fibrillation on ECG [10]. Higher scores acquired using all of these scoring systems correlate with increased risk of adverse evolution and in-hospital mortality. An online program called MDCalc(R) was used to calculate the scores.

The IPI was measured at the time of admission and before any treat- ment was administered. A Microstream(TM) bedside capnography moni- toring device was used for IPI measurement. The monitor’s Pulse oximeter probe was attached to the subject’s left index finger, and a nasal cannula (Oridion FilterLine(R)) was used for the capnography (PetCO2) sampling.

    1. Definitions of outcome measures

We identified 3 severity groups according to the recommendations of the Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines [1].

The mild group was the no respiratory failure group. Patients with an RR of 20-30 breaths per minute, no use of accessory respiratory mus- cles, no changes in mental status, hypoxemia improvement with sup- plemental oxygen (28-35% inspired oxygen [FiO2]) given via venturi mask, and no increase in PaCO2 were included in this group.

The moderate group was the nonlife-threatening acute respiratory failure group. Patients with an RR ?30 breaths per minute, use of acces- sory respiratory muscles, no change in mental status, hypoxemia im- provement with supplemental oxygen (25-30% FiO2) via Venturi mask, and hypercarbia (i.e., increased partial pressure of carbon dioxide [PaCO2] compared with the baseline or PaCO2 = 50-60 mmHg) were included in this group.

The severe group was the life-threatening acute respiratory failure group. Patients with an RR ?30 breaths per minute, use of accessory re- spiratory muscles, acute change in mental status, no improvement in hypoxemia with supplemental oxygen via Venturi mask or requiring FiO2 > 40%, hypercarbia (i.e., increased PaCO2 compared with the baseline or PaCO2 > 60 mmHg), or the presence of acidosis (i.e., pH

?7.25) were included in this group.

Although there may be hospital-specific differences in management and patient-based approaches, in general, according to the recommen- dations of the GOLD guide, members of the mild group can be treated as outpatients, whereas hospitalization is required for the moderate group, and intensive care follow-up is required for the severe group. For this reason, to provide the gold standard, sensitivity and specificity calculations were made based on the patient groups according to the GOLD guideline, not the patients’ ED outcomes (e.g., outpatient versus inpatient).

Image of Fig. 2Table 1

Demographic and clinical variables of the patients.

Age [median (IQR25-75)]

67 (61-73)

ECG Rhythm [N, (%)]

Female Sex [N, (%)]

49 (44.5)

Sinus Rhythm

92 (83.6)

Comorbidities [N, (%)]

Atrial Fibrillation

18 (16.4)

    1. Study protocol

Fig. 2. Improvement in CURB-IPI score.

Systemic Hypertension 66 (60)

Diabetes Mellitus 40 (36.4) ED treatments [N, (%)]

Coronary Artery Disease 31 (28.2) Nasal O2 105 (95.5)

Congestive Heart Failure 16 (14.5) O2 with mask 15 (13.6)

Chronic Kidney Disease 10 (9.1) Inhaler Bronchodilator 108 (98.2)

Hypothyroidism 10 (9.1) Intravenous Steroid 95 (86.4)

Patients admitted to the ED with respiratory distress, diagnosed with eCOPD, and who met the above-mentioned criteria were included in the study. The IPI was measured at the time of admission and before any treatment was administered. During the ED process, CURB-65, BAP65, and DECAF scores were calculated and, according to the severity of the disease, outcome groups were determined. The correlations be- tween the IPI and disease severity and prognostic severity scores were investigated.

    1. Improvement in CURB-IPI score

Based on the Receiver Operating Characteristic (ROC) analyses of the IPI for the detection of eCOPD severity, a new score was created using IPI and CURB-65 scores. An appropriate threshold for the IPI was deter- mined by conducting ROC analyses separately for the mild and severe eCOPD groups. According to this threshold value, if the IPI was >4, no points were added to the CURB-65 score of the patient. If the IPI was 3 or 4, 1 point was added to the CURB-65 score, and if the IPI was 1 or 2, 2 points were added. Thereby, a new score called CURB-IPI was created (Fig. 2).

    1. Sample size

The sample size of the study was determined using Dr. Hajian-Tilakis table for sample size estimation for diagnostic and prognostic studies [11]. Because the IPI is more important, especially for recognizing pa- tients who will not require hospital admission while under ED condi- tions, it was decided to take high sensitivity and a strong negative likelihood ratio (NLR) as expected values. Accordingly, given at least 90% sensitivity, at least 60% specificity, and an estimated NLR value of

0.17 (an upper limit of 0.30 in the 95% confidence interval) of the IPI

for determining which eCOPD patients will require hospital admission, it was decided that at least 110 patients should be included.

    1. Statistical analyses

The data were analyzed using Statistical Package for the Social Sci- ences (SPSS) for Windows 25 (Chicago, Illinois, USA). After analysis of the normality of distribution with the Shapiro-Wilk test and histo- grams, the data were defined as mean +- standard deviation if distrib- uted normally and as median and interquartile range if not distributed normally. The ordinal data were defined as frequencies. Interactions be- tween the groups were analyzed via Student’s t-test and the Mann- Whitney U, Kruskal-Wallis, and chi-square tests. For all calculations, p < 0.05 was considered statistically significant. Correlations between the IPI and other prognostic scores were analyzed with Spearman’s cor- relation test. Furthermore, ROC analysis was used to determine the di- agnostic values of the scores. Differences in the area under the curve (AUC) obtained by ROC analysis were compared using DeLong’s test.

  1. Results

One hundred and ten patients were included the study. The median age of the participants was 61 years (Interquartile Range [IQR] 25-75; 61-73). Forty-nine (44.5%) of the patients were female. According to the criteria outlined above, 43 (39.1%) of the patients were considered

Oral Steroid 2 (1.8)

Routine Treatments [N, (%)] NIMV 23 (20.9)

Nebulizer 69 (62.7) IMV 1 (0.9)

Oxygen Machine 59 (53.6)

CPAP-BiPAP device 36 (32.7)

ECG: Electrocardiography, ED: Emergency department, O2:Oxygen, CPAP: continuous positive airway pressure, BiPAP: Bilevel positive airway pressure, NIMV: Non invasive mechanical ventilation, IMV: Invasive mechanical ventilation.

to have mild, 43 (39.1%) moderate, and 24 (21.8%) severe exacerbation. The most common comorbid condition across all patient groups was hy- pertension, followed by diabetes mellitus and coronary artery disease. The patients’ demographic variables, comorbidities, ongoing outpatient treatments, and ED treatments during exacerbation are summarized in Table 1.

The correlations between the IPI and CURB-65, DECAF, BAP-65, and exacerbation severity were analyzed. There was a moderate negative correlation between the IPI and CURB-65 (p < 0.001, r – 0.406) and DECAF (p < 0.001, r – 0.379), whereas there was no correlation be- tween the IPI and BAP-65 (p 0.440). There was a strong negative corre- lation between the IPI and the GOLD classification of exacerbation severity (p < 0.001, r – 0.688). Data about the correlations are given in Table 2.

For evaluation of the diagnostic accuracy of the IPI in patients with mild eCOPD, ROC analysis was performed. The AUC was found to be 0.893. The most appropriate threshold of the IPI, with 97.67% sensitivity and 76.12% specificity, was determined as 4. The diagnostic accuracy of CURB-65, BAP-65, and DECAF for determining mild exacerbations was also analyzed. Of these scores, the highest diagnostic accuracy was ob- tained using CURB-65, with an AUC of 0.795. The diagnostic value of the IPI and other scores is given in Table 3. The AUC of 0.795 for CURB-65 increased to 0.909 for the improved CURB-IPI score, and there was a significant difference between ROC curves (p < 0.001, DeLong’s test) (Fig. 3).

The sensitivity of the CURB-IPI value of 3 for mild eCOPD was 94%. When higher sensitivity was targeted, with a CURB-IPI value of 2 as the threshold, sensitivity reached 100%, but specificity decreased to 21%.

  1. Discussion

The main result of the present study, in which the correlations be- tween the IPI and the severity of eCOPD and other prognostic clinical scores were analyzed, is that, of all the scores, the IPI had the highest di- agnostic accuracy in identifying mild eCOPD, and it was followed by the CURB-65 score. Our findings are consistent with previous studies that have shown the predictive value of clinical scores such as CURB-65, DECAF, and BAP-65 for eCOPD severity. However, to the best of our knowledge, there have been no previously published comprehensive studies regarding the use of the IPI in patients with COPD exacerbation in the ED. Our study fills this gap and demonstrates that IPI has a higher predictive value for COPD exacerbation severity than CURB-65, DECAF, and BAP-65 scores alone. We also showed that a new score, called CURB-IPI, which was created by combining CURB-65 and IPI scores, could determine mild COPD exacerbations with 94.03% sensitivity and 67.44% specificity. We also found that the CURB-IPI score had signifi- cantly higher predictive value than CURB-65 alone.

Table 2

Correlation between the IPI and the other scores.

P value (r coefficient)

CURB-65

BAP-65

DECAF

eCOPD severity?

IPI

eCOPD severity?

<0.001 (-0.406)

<0.001 (0.564)

0.440 (-0.075)

0.193 (0.126)

<0.001 (-0.379)

<0.001 (0.478)

<0.001 (-0.688)

* According to the GOLD classification. Statistically significant results are shown in bold characters.

Table 3

Diagnostic parameters of the IPI and other clinical scores for detection of mild eCOPD.

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

Cut-off Point

AUC

P value

95% CI

CURB-65

70.15

74.42

81.03

61.54

3

0.795

<0,001

0.681-0.867

BAP-65

77.61

30.95

64.2

46.43

2

0.541

0,503

0.423-0.658

DECAF

88.06

42.86

74.68

65.22

2

0.735

<0,001

0.635-0.834

IPI

97.67

76.12

72.41

98.08

4

0.893

<0,001

0.831-0.954

CURB-IPI

94.03

67.44

81.82

87.88

3

0.909

<0,001

0.854-0.963

In their retrospective study of 134 patients with eCOPD, Akhter et al. found that the AUC of the CURB-65 score was 0.78 for determining in- hospital mortality. The predictive value of CURB-65 for mortality was compared with that of Acute Physiology and Chronic Health Evaluation II and Sequential Organ Failure Assessment scores in the same study, and CURB-65 was more successful [12]. There are stud- ies comparing CURB-65, BAP-65, and DECAF scores in the literature. Parras et al. compared the predictive value of these 3 scores for mortal- ity in 164 patients with COPD exacerbation and showed that CURB-65 had the best predictive value. The AUCs of CURB-65, DECAF, and BAP- 65 were 0.860, 0.848, and 0.740, respectively [13]. Similarly, CURB-65 was found to have a higher predictive value for eCOPD severity than the BAP-65 and DECAF scores, although the outcome measured in our study was not mortality. However, there is an ongoing debate in the lit- erature about whICH Score is the most appropriate.

The IPI, which is new in clinical practice, is a mathematical algorithm that combines different Vital parameters into a single value and helps the clinician more easily interpret the patient’s respiratory status. The

Image of Fig. 3

Fig. 3. ROC curves of CURB-65 and CURB-IPI for mild eCOPD.

validation study of the IPI algorithm, performed with 523 patients, showed that for determination of respiratory problems, it has 83-100% sensitivity and 74-96% specificity [5]. Although IPI offers the possibility of integrated evaluation of 4 different parameters, it is not specific enough to guide treatment directly. Therefore, according to the current literature, its role is limited to that of an early warning sign of critical illness.

In their study, which included patients followed up during invasive mechanical ventilation for >24 h, Kaur et al. demonstrated that a de- crease in the IPI during the hour after extubation was an independent predictor of extubation failure [14]. Turan et al. studied the IPI in pa- tients under sedation during Endoscopic procedures, and in 5 of the 25 patients included, respiratory failure developed. The IPI was low in all the patients with respiratory failure, and in 3 of those 5 patients, de- spite low IPI values, oxygen saturation was normal. Based on these data, it was concluded that the IPI gave an earlier warning than SpO2 monitoring [15]. In another study, conducted in a pediatric ED, low IPI values were related to increased frequency of application of vital treat- ments and hospitalizations in the ED [16].

In our study, IPI had a higher predictive value for eCOPD severity than CURB-65, DECAF, and BAP-65 scores alone. As a simple mea- surement reflecting Respiratory physiology, IPI can detect mild exac- erbations with high accuracy. However, given that IPI alone may not provide sufficient information about other systemic effects in pa- tients, we examined its combination with CURB-65, which has the best predictive value among the other scores. With this combination, we aimed to incorporate the role of clinical parameters such as con- sciousness level, age, and blood urea level into the evaluation of ex- acerbation severity. Our analysis shows that this new score, which we call CURB-IPI, has a significantly better predictive value than CURB-65 (p < 0.001, DeLong’s test). Although it is controversial to state that CURB-65 is the most appropriate score for evaluating eCOPD, in the light of our data, we think that CURB-IPI can be easily applied in eCOPD patients in EDs and can be a reliable score with high predictive value for mild eCOPD.

In conclusion, although further studies are needed on the subject, we

think that the CURB-IPI score can be used to make decisions about safe discharge and to guide the identification of patients with mild eCOPD. Moreover, due to its ease of use and high predictive value, it may be used in triage in crowded EDs to safely identify low-risk patients.

  1. Limitations

The ED outcomes of the patients (i.e., discharge, inpatient clinic admission, or intensive care unit admission) were not evaluated be- cause these outcomes were directly related to local factors such as hospital capacity and service design. One of our study centers was a

hospital specializing in respiratory illnesses, and in the ED of this center, patients may stay a long time, and management of patients in the ED could be conducted as if it were an inpatient clinic. This prevented us from obtaining generalizable results for all hospitals. Therefore, we grouped the patients assuming that those with mild exacerbation–who did not need mechanical ventilation, had a PaO2 > 60 mmHg in room air (or under continuous oxygen therapy), had a respiratory rate < 30/min, and responded to the initial treatment–could potentially be discharged safely.

Another limitation was the evaluation of the IPI just before treat- ment and the lack of repeated IPI measurement after treatment. Because the IPI is a combination of vital parameters, it may change quickly after initial treatment of the exacerbation. Therefore, evaluation of the IPI after initial ED treatment of eCOPD might be more valuable.

In addition, despite the fact that we calculated the sample size before starting our study, when evaluating the generalizability of the study, the number of included cases should be considered.

Finally, because 1 study center was a respiratory illness hospital, most of the patients had severe COPD. This may have affected our results.

  1. Conclusion

For detection of mild eCOPD, the IPI had the best diagnostic accuracy among all the studied scores. It was followed by the CURB-65 score. The IPI was combined with CURB-65 to create the CURB-IPI score, and this study showed that with 94.03% sensitivity and 97.44% specificity, the CURB-IPI score had higher diagnostic accuracy than CURB-65 for detect- ing mild exacerbations.

We think that the CURB-IPI score can be a guide in the ED manage- ment of patients experiencing eCOPD due to its ease of application and its strong predictive value in identifying patient groups that are likely to be discharged due to their response to ED treatment. Moreover, the IPI may be used alone in triage in crowded EDs and help clinicians interpret the respiratory status of patients within minutes of admission.

Ethics committee approval

The study was conducted with the approval of Kecioren Training and Research Hospital Clinical Studies Ethical Committee with file number 2012-KAEK-15/2350 at 14.09.2021. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the institution’s human research committee.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Financial disclosure

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

CRediT authorship contribution statement

Fatma Nur Karaarslan: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Zeynep Saral Ozturk: Data curation. Gulsah Cikrikci Isik: Conceptualization, Formal analysis,

Investigation, Methodology, Writing – review & editing. Yunsur Cevik: Methodology, Investigation, Conceptualization.

Declaration of Competing Interest

The authors declare that there are no conflicts of interest.

Acknowledgements

None.

References

  1. Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diag- nosis, management, and prevention of chronic obstructive pulmonary disease 2019 report. URL: https://goldcopd.org/wp-content/uploads/2018/11/GOLD-2019- POCKET-GUIDE-FINAL_WMS.pdf; 2019. [Accessed January 24, 2019].
  2. Quintana JM, Esteban C, Unzurrunzaga A, Garcia-Gutierrez S, Gonzalez N, Barrio I, et al. Predictive score for mortality in patients with COPD exacerbations attending hospital emergency departments. BMC Med. 2014;12:66. https://doi.org/10.1186/ 1741-7015-12-66.
  3. Dogan NO, Varol Y, Kokturk N, Aksay E, Alpaydin AO, Corbacioglu SK, et al. 2021 guideline for the management of COPD exacerbations: emergency medicine Associ- ation of Turkey (EMAT) / Turkish thoracic society (TTS) clinical practice guideline task force. Turk J Emerg Med. 2021;21(4):137-76. https://doi.org/10.4103/2452- 2473.329630.
  4. Shafuddin E, Chang CL, Hancox RJ. Comparing severity scores in exacerbations of chronic obstructive pulmonary disease. Clin Respir J. 2018;12(12):2668-75. https://doi.org/10.1111/crj.12973.
  5. Ronen M, Weissbrod R, Overdyk FJ, Ajizian S. Smart respiratory monitoring: clinical development and validation of the IPITM (integrated pulmonary index) algorithm. J Clin Monit Comput. 2017;31(2):435-42. https://doi.org/ 10.1007/s10877-016-9851-7.
  6. Kaur R, Vines DL, Liu L, Balk RA. Role of integrated pulmonary index in identifying Extubation failure. Respir Care. 2017;62(12):1550-6. https://doi.org/10.4187/ respcare.05434.
  7. Freundlich RE, Walco JP, Mueller DM, Wanderer JP, Rothman BS, Shotwell MS, et al. Prospective randomised trial of the integrated pulmonary indexTM in low-acuity inpatients. Br J Anaesth. 2018;121(6):1375-7. https://doi.org/10. 1016/j.bja.2018.08.019.
  8. Lim WS, van der Eerden MM, Laing R, Boersma WG, Karalus N, Town GI, et al. Defin- ing community acquired pneumonia severity on presentation to hospital: an inter- national derivation and validation study. Thorax. 2003;58(5):377-82. https://doi. org/10.1136/thorax.58.5.377.
  9. Tabak YP, Sun X, Johannes RS, Gupta V, Shorr AF. Mortality and need for mechanical ventilation in acute exacerbations of chronic obstructive pulmonary disease: devel- opment and validation of a simple risk score. Arch Intern Med. 2009;169(17): 1595-602. https://doi.org/10.1001/archinternmed.2009.270.
  10. Steer J, Gibson J, Bourke SC. The DECAF score: predicting hospital mortality in exac- erbations of chronic obstructive pulmonary disease. Thorax. 2012;67(11):970-6. https://doi.org/10.1136/thoraxjnl-2012-202103.
  11. Hajian-Tilaki K. Sample size estimation in diagnostic test studies of bioMedical Informatics. J Biomed Inform. 2014;48:193-204. https://doi.org/10.1016/j.jbi. 2014.02.013.
  12. Akhter S, Warraich UA, Ghazal S, Rizvi N. Assessment and comparison of APACHE II (acute physiology and chronic health evaluation), SOFA (sequential organ failure as- sessment) score and CURB 65 (confusion; urea; respiratory rate; blood pressure), for prediction of inpatient mortality in acute exacerbation of chronic obstructive pulmo- nary disease. J Pak Med Assoc. 2019;69(2):211-5.
  13. Parras M, Bautista C, Chica G, Lopez M, Evaluation of DECAF. CURB-65 and BAP-65 scales as predictor of mortality risk in acute exacerbation of COPD in a retrospective cohort. Eur Respir J. 2017;50:PA931. https://doi.org/10.1183/1393003.congress- 2017.PA931.
  14. Kaur R, Vines DL, Patel AD, Lugo-Robles R, Balk RA. Early identification of Extubation failure using integrated pulmonary index and High-risk factors. Respir Care. 2021;66 (10):1542-8. https://doi.org/10.4187/respcare.08656.
  15. Turan G, Ayanoglu Tas B, Demiroluk O, Yildirim Ar A, Karadogan F, Bulut NE, et al. Integrated pulmonary index in endoscopic procedures on sedation. Bosphorus Medical Journal. 2015;2(3):85-8.
  16. Gurlu R, Tolu Kendir O, Baspinar O, Erkek N. Can non-invasive capnography and integrated pulmonary index contribute to patient monitoring in the pediatric emergency department? Klin Padiatr. 2022;234(1):26-32. English. https://doi. org/10.1055/a-1546-1473.