The impact of pulmonary functions on outcome of intubated patients with chronic obstructive pulmonary disease

Background: Assessment of lung mechanics and respiratory muscle determinants is considered to be useful for adjustment of ventilator settings to minimize dynamic hyperinflation in patients with chronic obstructive pulmonary disease (COPD). Objective: The aim of this study is to assess the impact of pulmonary functions on the outcome of mechanically ventilated COPD patients. Patients and methods: This study was carried out in the respiratory ICU, chest department, Assiut University Hospital. A total of 47 mechanically ventilated COPD patients were recruited into the study. The patients were divided into two groups on the basis of outcomes: successful group (36 patients) and failed group (11 patients). Analyses of different lung mechanics and respiratory muscle determinants were carried out for both groups. Results: There was no significant difference in age between both the groups. Arterial blood gases analysis showed a significantly decreased pH, PaO 2 , and SaO 2 in the failed group and a significantly increased PaCO 2 in the failed group after 1, 24, and 48 h of mechanical ventilation. Values of lung mechanics in the failed group, compared with the successful group, showed a significantly lower respiratory compliance (29.18 ± 1.47 vs. 45.92 ± 4.08 ml/cm H 2 O), a significantly higher respiratory resistance (24.73 ± 1.19 vs. 22.39 ± 1.10 cm H 2 O/L/S), and a significantly higher intrinsic positive end expiratory pressure (8.36 ± 0.67 vs. 7.58 ± 0.94 cm H 2 O). Values of respiratory muscle determinants in the failed group showed a significantly lower negative inspiratory force (−17.18 ± 2.82 vs. −26.44 ± 3.32 cm H 2 O), a significantly lower vital capacity (382.27 ± 102.75 vs. 810.00 ± 133.03 ml), and a significantly higher occlusion pressure P0.1 (6.09 ± 1.04 vs. 1.72 ± 0.66 cm H 2 O). Conclusion: Pulmonary functions in mechanically ventilated COPD patients including lung mechanics and respiratory muscle determinants had an important impact on outcome. A significant decrease in respiratory compliance, negative inspiratory force, and vital capacity with a significant increase in respiratory resistance, intrinsic positive end expiratory pressure, and occlusion pressure P0.1 was observed in the COPD group with failed extubation.


Introduction
Chronic obstructive pulmonary disease (COPD) has become a major epidemic with an increasing incidence worldwide, particularly in developing countries [1], and is now one of the leading causes of death and disability, with a prevalence of more than 10% in individuals above 40 years of age in most countries [2].
COPD is a disease characterized by poorly reversible airfl ow limitation of the airways that is usually progressive and associated with an abnormal infl ammatory response in the lung [3]. Th e abnormal infl ammatory response is usually triggered by smoking [4] or exposures to other environmental irritants [5] that interact with genetic factors [6], leading to both airway and systemic infl ammation, resulting in airway injury and lung damage.
COPD is a disease characterized by airway narrowing that produces a critical increase in airway resistance, especially during exhalation. Breathing through narrowed airways imposes an additional load on the respiratory muscles [7].
In the presence of high resistance to expiratory fl ows and short expiratory times, the respiratory system cannot return to its resting volume at the end of exhalation. As a result, a new resting state is established, such that there is a positive recoil pressure [(intrinsic positive end expiratory pressure (PEEPi)] at the end of expiration [8].
Th is state of air trapping or dynamic hyperinfl ation is common in patients with COPD. Initially, hyperinfl ation tends to keep the airways open, reduces airway resistance, increases elastic recoil, and tends to improve expiratory fl ow. However, hyperinfl ation exerts several deleterious eff ects. Th e positive pressure within regions of hyperinfl ated lung increases the mean intrathoracic pressure and causes the inspiratory The impact of pulmonary functions on outcome of intubated patients with chronic obstructive pulmonary disease Ahmed Metwally a , Khaled Hussein a , Ashraf Zein El-Abdeen a , Ahmed Hamed a , Azaa Ez-Eldeen b Background Assessment of lung mechanics and respiratory muscle determinants is considered to be useful for adjustment of ventilator settings to minimize dynamic hyperinfl ation in patients with chronic obstructive pulmonary disease (COPD).
Objective The aim of this study is to assess the impact of pulmonary functions on the outcome of mechanically ventilated COPD patients.

Patients and methods
This study was carried out in the respiratory ICU, chest department, Assiut University Hospital. A total of 47 mechanically ventilated COPD patients were recruited into the study. The patients were divided into two groups on the basis of outcomes: successful group (36 patients) and failed group (11 patients). Analyses of different lung mechanics and respiratory muscle determinants were carried out for both groups.

Results
There was no signifi cant difference in age between both the groups. Arterial blood gases analysis showed a signifi cantly decreased pH, PaO 2 , and SaO 2 in the failed group and a signifi cantly increased PaCO 2 in the failed group after 1, 24, and 48 h of mechanical ventilation. Values of lung mechanics in the failed group, compared with the successful group, showed a signifi cantly lower respiratory compliance (29. 18  muscles to operate at a higher than resting lung volume. Th us, dynamic hyperinfl ation places the respiratory muscles at a considerable mechanical disadvantage and further impairs respiratory muscle function [9].
PEEPi also imposes a huge inspiratory threshold load because a negative intrapleural pressure equal to the level of PEEPi has to be generated before inspiratory fl ow begins within alveoli [10].
In emphysema, although static compliance is high because of the overall eff ect of loss of elastic recoil, dynamic compliance is often low on account of the airways obstruction and because of loss of traction of elastic tissue on the airways [11]. Pulmonary hyperinfl ation along with an increase in the work of breathing that occurs in COPD are the main contributing factors to respiratory muscle dysfunction [12].
During mechanical ventilation, a study to minimize dynamic hyperinfl ation is the main goal to improve the weaning outcome. Assessment of lung mechanics and respiratory muscle determinants is considered to be useful for adjustment of ventilator settings to minimize dynamic hyperinfl ation in patients with COPD.

Patients and methods
Th is study was carried out in the respiratory intensive care unit, chest department, Assiut University Hospital. Th e study included mechanically ventilated COPD patients with respiratory failure who fulfi lled the following inclusion criteria.

Inclusion criteria
(1) Disturbed conscious level because of hypercapnia.

Exclusion criteria
(1) Age: younger than 20 years or older than 80 years.
(2) Pregnancy. All patients at admission to the respiratory intensive care unit were subjected to the following: Arterial blood gases Th ese were obtained by blood samples from the radial artery and analyzed using a blood gases analyzer (Rapid lab850; CHIRON/Diagnostics; critical care systems), and included pH, PaCO 2 , bicarbonate level, partial pressure of arterial oxygen (PaO 2 ), and arterial oxygen saturation (SaO 2 ).
During mechanical ventilation, patients were subjected to the following: (1) Continuous monitoring of heart rate, blood pressure (systolic and diastolic), respiratory rate, and arterial oxygen saturation using a pulse oximeter (SpO 2 ).

Mechanics of breathing
Mechanics of breathing were recorded after 24 and 48 h of mechanical ventilation.
(1) Respiratory compliance: Th is was assessed by pressing the inspiratory pause button of the ventilator. Th is is automatically calculated and displayed as the ratio between the exhaled tidal volume and the diff erence between plateau pressure and the unoccluded PEEP. (2) Respiratory resistance: Th is was assessed by pressing the inspiratory pause button of the ventilator. Resistance was automatically calculated and displayed by dividing the plateau pressure subtracted from peak pressure by fl ow. (3) Intrinsic positive end expiratory pressure: Th is is displayed automatically by pressing the expiratory pause button of the ventilator.

Respiratory muscle determinants
Respiratory muscle determinants were recorded after 48 h of mechanical ventilation.
(1) Negative inspiratory force (NIF): We measured NIF three times and the most negative result was taken. A negative force that was weaker than −20 cm H 2 O implies respiratory muscle weakness and diffi cult extubation (Fig. 1). (2) Vital capacity (VC): We measured VC three times and the average of the results was obtained. (3) Occlusion pressure (P0.1): We measured P0.1 fi ve times over a period of 60-90 s and the average of these measurements was obtained.

Medical treatment
Medical treatment included nebulized salbutamol and ipratropium bromide, which were administered through a piece connected to the ventilator circuit near the mouth. Intravenous hydrocortisone, 100 mg/12 h, was administered to all COPD patients until discharge from the ICU. Th eophylline was administered intravenously 6 mg/kg over 20-30 min, followed by a continuous infusion of 0.6 mg/kg/h. Antibiotics were administered as combination therapy with cefepime 1 g or ceftazidime 1 g/12 h plus levofl oxacin 500 mg/24 h or amikacin 500 mg/12 h as all our patients had bacterial infection.
According to the weaning outcome, we classifi ed patients into two groups: successful group and failed group.

Statistical analysis
Data were recorded to statistical package for social science, version 16. Data were described as mean ± SD and frequencies for nominal data. Th e correlation between continuous variables was determined using Spearman's correlation coeffi cient. P less than 0.05 was considered signifi cant. Figure 2 shows the outcome of mechanically ventilated COPD patients and indicated that 76.6% of patients had successful outcome, whereas 23.4% of patients had failed outcome. Table 1 shows the demographic and historical data of both groups of COPD patients and indicated a signifi cantly increased smoking index in the failed group. Also, there was a signifi cantly increased number of exacerbations last year, previous hospitalizations, and incidence of previous mechanical ventilation in the failed group. Table 2 shows the baseline clinical and gasometric data. A signifi cantly decreased pH was found in the failed group, with no signifi cant diff erence in the other baseline clinical and gasometric data in both groups of COPD.

Results
As shown in Table 3, from baseline laboratory data in both groups of COPD, it is clear that there was a signifi cantly increased TNF-α and high-sensitivity C-reactive protein in the failed group, whereas there was a signifi cantly decreased serum albumin in the failed group.    increase in the respiratory rate, minute ventilation, peak pressure, and plateau pressure in the failed group. However, there was a signifi cant (P < 0.01) decrease in the exhaled tidal volume and PaO 2 /FiO 2 in the failed group. Table 4 shows the follow-up of lung mechanics in both groups of COPD patients. A signifi cantly decreased respiratory compliance and a signifi cantly increased respiratory resistance and PEEPi were found in the failed group. Table 5 shows the respiratory muscle determinants in both groups of COPD patients. Signifi cantly decreased NIF and VC, and a signifi cantly increased P0.1 were found in the failed group.       Figure 5 shows that there was a significant positive correlation (r = 0.958, P = 0.000) between NIF and VC in both groups of COPD patients. Figure 6 shows that there was a signifi cant negative correlation (r = 0.970, P = 0.000) between NIF and duration of MV. Figure 7 shows that there was a signifi cant positive correlation (r = 0.441, P = 0.002) between peak airway pressure and respiratory resistance.

Discussion
Increased number of exacerbations in the last year, more frequent hospitalizations, and previous mechanical ventilation are associated with an increased risk of failed extubation [13]. Th is is consistent with our results.
Our results indicated a signifi cantly increased smoking index in the failed group; these results are in agreement with those of Bakr et al. [14], who reported that there was a signifi cant diff erence in the smoking index in both groups of COPD: 38 ± 33.5 packs/years in the successful group and 62.82 ± 29.68 packs/years in the failed group.
Follow-up of arterial blood gases was performed in both groups of COPD patients. Signifi cantly decreased pH, PaO 2 , and SaO 2 and a signifi cantly increased PaCO 2 were found in the failed group; these results are in agreement with those of Ucgun et al. [15], who reported that COPD patients with failed extubation were characterized by a signifi cantly decreased PaO 2 (52.8 ± 7.0 mmHg), a signifi cantly decreased pH (7.20 ± 0.97), and a signifi cantly increased PaCO 2 (75.8 ± 10.1 mmHg). Also, Bakr et al. [14] reported an increase in mortality with derangement of all arterial blood gases items in patients with COPD, with mean pH (7.24 ± 0.03 and 7.20 ± 0.01), mean PaCO 2 (54.8 ± 6.39 and 76.1 ± 11.1 mmHg), and mean PaO 2 (54.8 ± 7.1 and 50.2 ± 4.9 mmHg) for survivors and nonsurvivors, respectively; this was statistically signifi cant (P < 0.05).
Th ere was a statistically signifi cant decrease in PaO 2 / FiO 2 in the failed group; these results are in agreement with those of Park and Koh [16], who reported a signifi cantly decreased PaO 2 /FiO 2 in the failed group.  A signifi cant positive correlation between NIF and VC. NIF, negative inspiratory force; VC, vi tal capacity.

Fig. 5
A signifi cant negative correlation between NIF and duration of MV. NIF, negative inspi ratory force. A signifi cant positive correlation between peak pressure an d resistance.
Also, Madkour and Adly [17] reported that the mean PaO 2 /FiO 2 in COPD patients with failed extubation was signifi cantly lower than that in the group of patients with successful extubation.
Measurement of lung mechanics indicated a signifi cantly decreased respiratory compliance and signifi cantly increased respiratory resistance and PEEPi in the failed group. Th ese results are in agreement with those of Jubran and Tobin [18], who reported that COPD patients who failed the weaning trial of mechanical ventilation developed worsening of lung mechanics compared with patients who tolerated the weaning trial. Also, Lin et al. [19] reported that patients who had better underlying lung mechanics (higher respiratory compliance and lower PEEPi) had better chances of weaning from mechanical ventilation.
Nava et al. [20] reported that respiratory compliance measured in the fi rst 24 h from intubation enabled good diff erentiation between the patients who were weaned successfully and those who failed.
Our results indicated a signifi cant increase in resistance and PEEPi in the failed group. Th ese results are in agreement with those of Broseghini et al. [21], who reported that patients with acute exacerbation of COPD were characterized by high respiratory resistance and severe pulmonary hyperinfl ation, with 'intrinsic' PEEP (PEEPi) up to 22 cm H 2 O. Increased respiratory resistance and pulmonary hyperinfl ation were associated with increased incidence of failed extubation.
Increased respiratory resistance in patients with COPD can be explained by the fact that in advanced COPD, the alveolar attachments that normally keep the smaller airways open by radial traction are lost [22], increased airway collapsibility occurs because of destruction of the lung parenchyma, and there is loss of lung elastic recoil [23].
In respiratory muscle determinants, signifi cantly decreased NIF and VC, and a signifi cantly increased P0.1 were found in the failed group; these results are in agreement with those of Graham and Nausherwan [24], who reported that patients with obstructive pulmonary disorders had P0. Li et al. [25] documented that NIF among the group of COPD patients who failed extubation was −13 ± 3 cm H 2 O, whereas NIF in the group of COPD patients with successful extubation was −21 ± 4 cm H 2 O.
Th ere was a signifi cant positive correlation between NIF and VC; this result is in agreement with that of Prigent et al. [26], who reported that there was a positive correlation between NIF and VC. NIF can be used to evaluate patients with hypoventilation because of inspiratory muscle weakness, whereas VC is limited by weakness of both inspiratory and expiratory muscles and by compliance of the respiratory system.
Lawn et al. [27] reported that decreased VC less than 15 ml/kg and NIF less than −30 cm H 2 O may be predictors for invasive mechanical ventilation.

Conclusion
Pulmonary functions in mechanically ventilated COPD patients including lung mechanics and respiratory muscle determinants had an important impact on outcome. A signifi cant decrease in respiratory compliance, NIF, and VC, with a signifi cant increase in respiratory resistance, intrinsic PEEP, and occlusion pressure P0.1, was observed in the COPD group with failed extubation.