Adequacy of oxygenation parameters in elderly patients undergoing mechanical ventilation

ABSTRACT Objective: To compare ideal PaO2 with PaO2 found, ideal PaO2/FiO2 of room air with the one found, and ideal FiO2 with FiO2 found in mechanically ventilated elderly patients. Methods: Cross-sectional study that evaluated elderly mechanically ventilated patients for at least 72 hours and who underwent three subsequent blood gas analyses. Results: The sample consisted of 48 elderly with mean age of 74.77±9.36 years. There was a significant difference between the ideal PaO2 and the one found (p<0.001), between FiO2 corrected and the offered one, and also between ideal PaO2/FiO2 of room air and the PaO2/FiO2 found (p<0,001). Conclusion: A significant increase was seen in PaO2 and FiO2 and in alterations of gas exchange by PaO2/FiO2 index than those found in normal parameters.


INTRODUCTION
According to Brazilian Ministry of Health, ageing population in Brazil is growing and currently achieved more than 19 million of individuals (1) . Ageing brings with it an increase in chronic-degenerative diseases rate and in some cases leads to the need of interventions that require admission to Intensive Care Units (ICU) (2) . In 2008, pneumonia was the second leading cause of admissions among ageing population; the hypertension was the first leading cause (3) . Pneumonia could lead to respiratory insufficiency, which requires invasive ventilation support (4) . Cohen et al. (5) reported that elderlies were benefited by treatment in ICU and by mechanical ventilation (MV).
Senility causes changes in several organic systems such as the respiratory system, therefore harming its function. These harms could be higher depending on environmental issues as smoking and chronic degenerative diseases (4,6,7) . Alterations could be divided into structurally and functionally. Enlargement of deadspace, rigidity of chest wall, decrease of mucociliary clearance, strength and muscle mass are considered examples of structural changes whereas reduction of complacency of chest wall, the increase of pulmonary complacency, changes in pulmonary capacity and volumes (reduced vital capacity and increased residual volumes) are example of functional changes (6,(8)(9)(10)(11)(12)(13) .
After define the use of MV the constant monitoring of respiratory function is fundamental to detect complications early, to enable the prognosis analysis, to evaluate response to treatment and to reduce complications related to long duration of MV, therefore enabling the weaning from ventilator as soon as possible (14) . Among ventilation monitoring in patients, the following could be used: pulse oximetry, gasometry (blood gas analysis), measurement of oxygenation and ideal fraction of inspirited oxygen (FiO 2 ), among others. The PaO 2 / FiO 2 ratio is used to determine patient's oxygenation index, which is also called pulmonary oxygenation capacity index. The PaO 2 /FiO 2 relations >301 indicate an adequate oxygenation between 201 and 300, acute lung injury (ALI); and relations <200 characterize a sign of acute respiratory distress syndrome (ARDS) (15)(16)(17) . In ICUs the measurement of pulmonary oxygenation capacity index is important in order to monitor each patient in the most accurate way.
Age is included in the calculation of ideal PaO 2. The measurement is carried out with patient in supine position, and it uses the following formula: PaO 2 = 109-(0.43 x age) (18) . For this reason, elderly people commonly present a lower PaO 2 . Mechanical ventilated patients receive higher amount of FiO 2 than room air (21%), which is called oxygenotherapy. The FiO 2 delivered could be corrected using the corrected FiO 2 formula=(delivered FiO 2 x ideal PaO 2 /PaO 2 found) (8) . When oxygen (O 2 ) is administered in high doses and for longer periods, it might generates deleterious effects such as absorption atelectasis, interstitial pulmonary edema, and changes in cell structure and function (19,20) . The use of high doses of O 2 , besides being potentially harmful for patients, increases costs in hospital stay because of O 2 supplement high cost.

OBJECTIVE
To analyze parameters of gas exchange expected and found in mechanically ventilated elderly patients considering ideal and found PaO 2 , ideal and found PaO 2 /FiO 2 index and corrected and delivered FiO 2 . We hypothesize that mechanically ventilated elderly patients are receiving FiO 2 beyond necessary, and as a consequence they are subject to deleterious effects of high doses of O 2 . Once a different relation is established, which is corrected by age, concerning the PaO 2 /FiO 2 relationship, patients would receive lower doses of FiO 2 . In addition, this establishment would enable few numbers of possible ALI/ARDS cases and complications due to O 2 toxicity.

METHODS
This was a quantitative, cross-sectional and descriptive study. Data were collected from patients' medical records admitted at the hospital where the study was conducted from January 1 st to May 31 st 2011.
Inclusion criteria were men and women aged 60 years or older admitted at the ICU independently of the diagnosis, and submmitted to MV for at least 72 hours and blood gas analysis during three consecutive days after MV.
We excluded those patients who died before 72 hours after invasive MV, not underwent blood gas analysis within the first 3 days after invasive MV and evolved with airway extubation before 72 hours of MV.
Patients' medical records were used to collect information. We instructed an assistant researcher graduated in physiotherapy to use a form created by the researchers of this study; this assistant was the responsible to collect the data. In order to create an epidemiological profile of patients, information were collected using a nursing notepad that included patients' name, age, gender and diagnosis. Medical records and the nursing notepad were analyzed in the ICU, and no documents were taken outside the unit.
During patients' hospitalization, in the morning, a nurse collected the sample using a BD 3mL or 5mL syringe (Becton, Dickinson and Company, New Jersey, USA) for blood gas analysis. We did not request to change the collection time because our study analyzed only medical records. A laboratory technician or the on-duty nurse collected the blood gas analysis. After that, samples were analyzed by a laboratory hired by the hospital. The laboratory informed that the blood gas analyzer (Gem ® Premier™ 3000, Instrumentation Laboratory, Hamburg, Germany) was automatically calibrated after the use. Medical records lack the information if collection was done through radial or femoral artery.
Data from 3 consecutive days since the first day that patient underwent MV were used. We observed blood gas analysis and adequate MV based on these data regarding the delivered FiO 2 , the PaO 2 considered ideal, and also the SpO 2 .
This study was approved by the Ethical and Research Committee of the Universidade Católica de Brasília (UCB), protocol n. 328/2010.
A sample descriptive analysis was conducted using means, standard deviation and frequencies. We used the paired Student's t-test to compare ideal and found PaO 2 , corrected and delivered FiO 2 , and relation between ideal and found PaO 2 /FiO 2 . A of CI95% and p≤0.05 were considered for the analysis.

RESULTS
A total of 335 patients were admitted to hospital ICU from January 1 st to May 31 st 2011, of them 177 were men. Patients' mean age was 62.83±18.56 years with maximum of 101 and minimum of 13 years old, being 61.50% inpatients aged 60 years or over. The mean hospital stay was 11.15±26.23 days with maximum of 222 and minimum of 1 day. In all, 82.93% of patients stayed in the ICU for 11 days. Of these inpatients 23.28% died, 5.37 were transferred and 71.34% were discharged.
Among patients admitted to ICU in the period, 50 fulfilled the study inclusion criteria. At the beginning, data normality was assessed using the Kolmogorov-Smirnov test and no deviations were seen. However, two outliers were found later, but they were excluded from the sample. Sample was composed by 48 patients with mean age of 74.77±9.36 years old with maximal of 101 and minimal of 60 years old; of them 25 were women. The mean time of hospital admission was 37.10±38.93 days with maximal of 184 days and minimal of 3 days. Of included patients 77.08% died, 20.83% were discharged and 2.08% were transferred. In moment of hospital admission 38% of patients were diagnosed with more than one disease. Main events are described in table 1.
Among events in the study, diseases with higher incidence are described in table 2.

DISCUSSION
Paiva et al. (21) analyzed the epidemiologic profile of patients admitted to ICU during 7 years. They observed that 47.37% of patients were 60 years or older. Rocha Hernández et al. (22) also reported that patients over 60 years old comprises the highest percentage of those admitted to ICU. Our results corroborate with these studies because 61.50% of patients admitted to ICU were 60 years or older.
In the present study, the most frequent disease was acute respiratory failure followed by pneumonia and stroke. In Paiva et al. (21) study, most prevalent disease in patients older than 60 years were acute myocardial infarction and acute respiratory failure. This latter high prevalence emphasizes the need for further studies on the adequate FiO 2 , once acute respiratory failure evolves quite inevitable to MV and, as a consequence, to the administration of supplemental O 2 , which when poorly administered can lead to toxic effects.
Based on patients' age their PaO 2 were proportionally higher than the ideal PaO 2 . This information is relevant because free radicals interaction have been described as an increase in "disease status" that include inflammation and ischemia (23) . Main consequences of such free radicals interaction are damage to cell membrane, enzyme inactivation and changes to the molecular genetic material (24) . Halliwell et al. (25) stated that O 2 is widely used in clinical medicine, however, this use must not blind the professional understanding concerning its toxic effects because present O 2 in inspired air slow damage effects that depend on patients' organism, age and nutritional status. On the other hand, in a study conducted in pigs by Aoki et al. (26) the deleterious effects of low doses of O 2 (40%) were more remarkable within the first 4 weeks of administration than by 8 weeks in which the damages were stagnated. In our study, measures of FiO 2 and PaO 2 were conducted for a short-term, which indicates that a careful attention to rigorous administration of this gas is necessary; this regulation could be done using the ideal PaO 2 .
Metnitz et al. (27) reported that patients diagnosed with ARDS have a highly compromised antioxidative system, both by administration of high FiO 2 and inflammatory status. In the present study, although the elderlies were not diagnosed with this syndrome, the population presented high doses of FiO 2 and, in several cases, inflammatory status due to underlying disease. Li et al. (28) observed that rats exposed to high doses of FiO 2 developed similar lung conditions to those found in patients with ARDS. Therefore, PaO 2 and FiO 2 control is fundamental to MV patients' recovery.
We observed that the PaO 2 /FiO 2 relation found lead to misinterpretation of gas exchange condition because patients presented a PaO 2 extremely above the ideal level and, as a consequence, the relation that results in this index was even lower. Lang et al. (29) observed high levels of free radical markers in patients who received high FiO 2 and who were at risk of developing an important inflammatory process and finally ARDS, which again raises the importance of a rigorous PaO 2 control to calculate PaO 2 /FiO 2 , an adequate indicator of this high mortality syndrome especially in ageing individuals who already have important changes in lungs due to ageing.
Other issue that emphasizes the need to control FiO 2 administrated to ageing people is the statement by Araújo Neto (20) that O 2 toxicity is age dependent placing ageing individuals at higher risk of complications.
To reduce what could be wrongly considered low PaO 2 /FiO 2 relations, we suggest to consider the ideal PaO 2 /FiO 2 calculation, i.e., the ideal gas exchange of that particular patient will be the ideal PaO 2 divided by room air FiO 2 . Such "ideal" values would be considered when delivered FiO 2 and found PaO 2 are adjusted. Therefore, it is important to emphasize that PaO 2 range in hemoglobin dissociation curve ranged from 60 to 100mmHg for an appropriate saturation. The ideal PaO 2 calculation for people aged until 100 years old refers to PaO 2 within the range (66mmHg) showing that to use this formula as a parameter is adequate. Therefore, it is advisable to calculate PaO 2 to adjust the FiO 2 .
Metnitz et al. (27) showed that nutrient replacement with antioxidative capability, which helps the organism to fight against free radicals, was not enough to compensate the needs of patients submitted to increased O 2 fraction. Del Maestro study (23) affirmed that in cases of ischemia there is the need to administrate O 2 adequately and give enough time for cellular mechanisms to protect themselves against free radicals that could adapt and neutralize actions of these aggressors. In this way, such actions could avoid damages to cells, which are not seen in clinical practice, but can be seen in our study by PaO 2 increase.
For this reason, we believe that increased PaO 2 is harmful both for the toxic effects caused by O 2 and for the false impression of poor gas exchange that most of the times guide MV therapy. Our study did not include correlated increased PaO 2 with mortality. Patients' severity was not analyzed in our study, however, this issue should be considered in future analyses.
There is the hypothesis that ageing patients could be affected by O 2 toxicity because high doses of O 2 were delivered.
The present study limitation was the absence of a precise measurement of time elapsed between the collection and the blood gas analysis. However, the service where the collection was done has the protocol of analyzing the material immediately upon receive in order to ensure the exam's accuracy. We did not observe clinical impacts by administration of high doses of O 2 . The scarcity of references in the literature on the subject was also a limitation because the studies found were outdated or used animal model. Considering the scarcity of researches in the field, further studies seem warranted.

CONCLUSION
In the study population no significant differences were seen between normal parameters and found PaO 2, as well as alterations of exchange according to PaO 2 /FiO 2 and FiO 2 delivered.