Assessment of gradient between partial pressure of arterial carbon dioxide and end-tidal carbon dioxide in acute respiratory distress syndrome

Context End-tidal carbon dioxide (EtCO2) is used as a noninvasive bedside test to assess the adequacy of ventilation and physiologic dead space in mechanically ventilated patients. The gradient difference between EtCO2 and partial pressure of arterial carbon dioxide (PaCO2) is directly related to an increase in the physiologic dead space. Aim The aim of this study was to evaluate the role of measuring the gradient between EtCO2 and PaCO2 in adults with acute respiratory distress (ARDS). Settings and design This was a prospective consecutive enrollment study. Patients and methods Overall, 51 cases were recruited after the diagnosis of ARDS was made according to the Berlin definition. Patients were mechanically ventilated as per the lung-protective protocol. Daily arterial blood gases were collected and for every sample, the EtCO2 value was collected electronically by capnography using an endotracheal tube for the first 5 days. Results Cases were classified into survivors and nonsurvivors: 26 cases were because of extrapulmonary causes and 25 cases were because of pulmonary causes. The mean value of the APACHE II score for all cases on admission was 21.6. The mean length of ICU stay was 12.7 days. For all study cases, PaO2/FiO2 was the lowest at day 1 and the highest at day 5. We found a significant negative correlation between PaO2/FiO2 and the gradient at days 2, 4, and day 5, and a significant positive correlation between the gradient on admission and the APACHE II score (r=0.4, P≤0.05). Nonsurvivors had a significantly higher gradient and lower EtCO2 and PaO2/FiO2 levels at all time intervals, whereas PaCO2 alone was found to be nonsignificant. Conclusion In ARDS, EtCO2 and gradient are reliable indicators of severity.


Introduction
According to Berlin guidelines [1], acute respiratory distress syndrome (ARDS) is classified as follows: For years, capnography has been used as a standard of care in the ICU to ensure correct placement and patency of the endotracheal tube. end-tidal carbon dioxide (EtCO 2 ) is also used in cardiopulmonary resuscitation to ensure that the proper technique is used. In healthy adults, EtCO 2 approaches the arterial carbon dioxide (PaCO 2 ); hence, capnography can provide quick real-time continuous monitoring of the levels of CO 2 (and hence ventilation) in patients. However, in diseases of the lungs causing ventilation perfusion mismatches, the relationship between EtCO 2 and PaCO 2 becomes altered and many studies have shown that EtCO 2 alone cannot be used as a surrogate for PaCO 2 when the lungs are diseased [2].
Physiologic dead space is the part of ventilation that is not involved in the gas exchange process and it includes the anatomical dead space (airways) and the alveolar dead space (alveoli not receiving perfusion). In ARDS patients on mechanical ventilation, the dead space increases as the tidal volume reaches non perfused lung segments. Overdistension of the normally perfused segments by high tidal volumes or high ventilation pressures (PEEP) can also increase dead space. It has also been shown that in ARDS microemboli and endothelial damage occurs, which may also lead to the dead space effect. In patients with ARDS, dead space plays a prognostic role and can be used to control ventilator parameters. The gradient between arterial and EtCO 2 [P(a−et)CO 2 ] widens with increased dead space [3].
In our study, we attempted to study the P(a−et)CO 2 in ARDS patients and whether this gradient correlates with disease severity and outcome.
Our aim is to evaluate the role of measuring the gradient between EtCO 2 and PaCO 2 in ARDS mechanically ventilated patients as a detector of disease severity.

Study design
This study is a consecutive prospective enrollment study that was carried out from January 2015 to January 2017 at the respiratory ICU.

Ethical approval and consent to participate
The ethical committee of our university approved the study and, therefore, our study was carried out in accordance with the ethical standards of Declaration of Helsinki, laid down in 1964, and its later amendments. All candidates in the research signed an informed consent after an explanation was provided of the details of the study and the possibility of publication. Consent was signed by the patients or the legal next of kin.

Consent for publication
Informed consent on the data that were collected and/ or published was explained and signed by each patient or the legal next of kin. All authors agreed to publish the data.

Patients
The study included 52 mechanically ventilated patients with ARDS. Patients were 18 years of age and older, recruited within 48 h of mechanical ventilation after the diagnosis of ARDS as per Berlin definition. We excluded patients who were hemodynamically unstable (mean arterial pressure of 64 mmHg despite the use of vasopressors), patients who previously had any chronic lung diseases leading to hypercapnia (PCO 2 >45), hypoxemia (PO 2 <5 mmHg), pulmonary hypertension (PAP>40 mmHg), polycythemia, morbid obesity (BMI>35), or home oxygen/ventilator dependence. Patients were enrolled between January 2016 and January 2017 in the ICU of our university hospitals.

Mechanical ventilation
The days; the shortest time of stay was for a case who died 3 days after admission. All descriptive data are shown in Table 1.
Nonsurvivors had a significantly higher gradient and lower EtCO 2 and P/F levels at all time intervals. PaCO 2 alone was not found to be significantly different between survivors and nonsurvivors. These findings are shown in Table 2 and Fig. 1.
We found a significant negative correlation between the P/F ratio and the arterio-EtCO 2 gap at days 2, 4, and day 5 as shown in Fig. 2. Figure 3 shows that a significant positive correlation was found between the gradient on admission and APACHE (r=0.4, P≤0.05).
ROC analysis of the EtCO 2 gradient and predicted mortality showed a cut-off value for the EtCO 2 gradient to predict mortality of 17.  (Fig. 4).

Discussion
Our main finding was that nonsurvivors had higher gradients and lower EtCO 2 at all study days compared with survivors and that the gradient and EtCO 2 were significantly related to P/F ratios (the higher the gradient, the lower the P/F); hence, the gradient increased as ARDS worsened, whereas PaCO 2 alone   did not change significantly in nonsurvivors. To our knowledge, very limited studies have been carried out to assess the gradient in adult patients with ARDS.

Limitations of our study
The main limitations of our study are that although we studied the P/F ratio, we did not stratify patients into those with mild, moderate, and severe ARDS; hence, we did not establish a relationship between P(a−et) CO 2 and the degree of ARDS as per the Berlin classification. Also, we did not study the level of PEEP and its effects on dead space and hence P (a−et)CO 2 .
These data are in agreement with a study carried out by Brinton and colleagues, who studied 35 patients with various degrees of ARDS and concluded that the gradient widens with increasing ARDS severity [4].

Figure 4
Receiver operating characteristic analysis of the EtCO 2 gradient and predicted mortality. A similar study by Yousuf and colleagues in 2016 established P(a−et)CO 2 in ARDS patients. The aim of this study was to detect a gradient in each grade of severity of ARDS; PaCO 2 , PaO 2 /FiO 2 , and EtCO 2 were measured. This study found that the gradient between PaCO 2 and EtCO 2 in ARDS worsened with increasing severity of ARDS, which indicates the value of gradient as an indicator of the severity of ARDS [5].
A study was carried out by Mehta and colleagues in neonatal and pediatric ICUs of a tertiary care children hospital. The study included 101 mechanically ventilated patients (35 children and 66 neonates). EtCO 2 and PaCO 2 were reported and data were analyzed. The researchers concluded that EtCO 2 is a good predictor of PaCO 2 , and the correlation between EtCO 2 and PaCO 2 was affected significantly by the P/ F ratio [6].In 2012, another study detected a positive correlation between EtCO 2 and PaCO 2 through all modes of mechanical ventilation and even while the candidates were on a T tube. Their final conclusion was that EtCO 2 is an accurate predictor of PaCO 2 in mechanically ventilated patients and the use of EtCO 2 may reduce the need for invasive monitoring and/or repeated arterial blood gases [7].
Although studies in this field are limited and although our study was also limited by the number of patients and the fact that the PEEP effect was not studied, most data of the studies carried out are in agreement.
More studies on a greater number of patients are required.

Conclusion
The gradient between EtCO 2 and PaCO 2 , and EtCO 2 plays a good role in predicting severity and prognosis in patients with ARDS and its determination is recommended as a noninvasive, reliable predictor in mechanically ventilated patients.

Financial support and sponsorship
Nil.

Conflicts of interest
There are no conflicts of interest.