Factors Associated with Intubation Time and ICU Stay After CABG

OBJECTIVE The aim of this study was to identify factors associated with intubation time and intensive care unit stay after coronary artery bypass grafting with cardiopulmonary bypass. METHODS This was a retrospective study, whose data collection was performed in the hospital charts of 160 patients over 18 years, who underwent surgery from September 2009 to July of 2013 in a hospital in the state of Espirito Santo, Brazil. RESULTS The mean age of the subjects was 61.44±8.93 years old and 68.8% were male. Subjects had a mean of 5.17±8.42 days of intensive care unit stay and mean intubation time of 10.99±8.41 hours. We observed statistically significant positive correlation between the following variables: patients' age and intubation time; patients' age and intensive care unit stay; intubation time and intensive care unit stay. CONCLUSION In conclusion, the study showed that older patients had longer intubation time and increased intensive care unit stay. Furthermore, patients with longer intubation time had increased intensive care unit stay.


INTRODUCTION
Patients referred for coronary artery bypass graft (CABG) frequently have one or more comorbidities associated with coronary heart disease, as hypertension, diabetes mellitus (DM), peripheral vascular disease, cerebrovascular disease, renal dysfunction (RD) and class II heart failure or above according to New York Heart Association (NYHA) [1] . Some of these comorbidities have been associated with prolonged mechanical ventilation (MV), and a higher risk of re-intubation in the postoperative period after CABG [2] .
A recent study reported that postoperative total MV time was an independent risk factor for re-intubation following CABG [3] . The occurrence of complications is also increased in older patients undergoing CABG [1,4] .
Prolonged MV after CABG is closely linked to increased morbidity and mortality of patients, and longer stays both in hospital and in intensive care unit (ICU) [2,5,6] , which generates more costs for the service [5,6] . It is important to study the predictors of prolonged MV and ICU stay in these patients because this knowledge may contribute to optimise the management of the most appropriate care for CABG patients.
The aim of this study was to identify factors associated with intubation time (IT) and ICU stay after CABG with cardiopulmonary bypass.

METHODS
This is a retrospective study composed of patients undergoing CABG, from September 2009 to July 2013 in a hospital in the state of Espírito Santo, Brazil.
Data collection was conducted from October 2013 to February 2014 on hospital charts of patients. Inclusion criteria were all patients who underwent isolated CABG with cardiopulmonary bypass (CPB) use from September 2009 to July 2013, over 18 years, of both genders. Patients who underwent concomitant CABG with valve surgery or other surgeries were not included in accordance with the inclusion criteria. We excluded patients who died and those whose charts did not have enough data for the research. Of the total of 172 charts, 160 were included in the sample. Three were excluded because the patient died in the first days after surgery and other nine did not contain sufficient data.
Patient charts were carefully reviewed and data collected were: chart record number, age, gender, admission date, discharge date, smoking history, preoperative hypertension, preoperative DM, preoperative dyslipidemia, preoperative acute myocardial infarction (AMI), date and time of surgery, number of coronary grafts, length of CPB, date and time of entry in the ICU, Categorical variables were expressed as percentages and continuous variables as mean ± standard deviation (SD). Initially, we used the Kolmogorov-Smirnov test to evaluate the normality of the variables. Due to the breakdown of normality hypothesis in all cases, Spearman correlation coefficient (non-parametric) was used to check the correlation between variables. For comparison between groups we used the non-parametric Mann-Whitney test.

RESULTS
Study population had a mean age of 61.44±8.93 years old and 68.8% (n=110) were male. The average time of CPB was 58.85±21.89 minutes, and the mean number of grafts per surgery were 2.79±0.88. The mean IT was 10.99±8.41 hours. The mean duration of ICU stay was 5.17±8.42 days. These intraoperative and postoperative characteristics of patients are detailed in Table 1.
We observed significant positive correlation between patients' age and IT; between patients' age and ICU stay; and between IT and ICU stay. Number of grafts and time of CPB did not have significant correlation with IT or ICU stay (Table 2).
There was no significant relationship between ICU stay and any of categorical variables gender, smoking, hypertension, DM, dyslipidemia and previous AMI (Table 3). Similarly, statistically significant relationships between these categorical variables and IT weren't observed ( Table 3). The prevalence of these categorical variables observed in the study population was: 68.8% had hypertension; 38.8% had DM (not specified whether type I or II); 38.1% had dyslipidemia, 39.4% were smokers, and 23.1% had previous AMI.
Regarding the physiotherapy care received by patients, on the PO1, only 3.1% of patients received physiotherapy care and each of them received only one attendance a day; on the PO1 16.3% of patients received 1 attendance a day and 2.5% received 2 attendances a day; on the PO2 only 5.6% of patients received 1 attendance a day. During PO3, PO4, PO5 e PO6 only 2.5% of patients received 1 attendance a day. Due to the low number of patients who received physiotherapy care, it was not possible to apply statistical tests to relate this variable with the study outcomes.

DISCUSSION
The average age of patients was similar to other studies in recent years [1,3,[5][6][7] , which is in line with the current trend of increasing age of patients undergoing CABG [1] .
The major findings of the present study were that patients' age had a significant positive correlation with TI and ICU stay. In addition, TI had a significant positive correlation with ICU stay.
Piotto et al. [8] also reported that age was a predictor of prolonged MV and increased hospital stay. Another study also identified age as an independent predictor of prolonged MV [9] . Due to the increasing age of patients undergoing myocardial revascularization in last decades [1] , it is important to minimize the risk of prolonged TI and ICU stay in this segment of the population, since prolonged intubation results in significant acute and midterm morbidity and mortality [2] .
Regarding the positive correlation between IT and ICU found in this study, Cislaghi et al. [5] reported that prolonged MV is associated with higher length of stay in ICU and total hospitalization time. Similarly, Akdur et al. [10] showed increase in ICU stay for patients who spent more than 24 hours on MV.
In our results, the number of grafts did not have significant correlation with IT or ICU stay, and we found a mean of 2.79±0.88 grafts. This number is smaller than the average number of grafts reported by some authors [1,2,11,12] , but our average of grafts was similar to the mean found in another study that also did not observe relationship between number of grafts and length of MV or ICO stay [10] .
Guizilini et al. [13] compared CABG postoperative period with and without CPB, and concluded that there are impairments of lung function in both; however, this is significantly higher in patients operated with CPB. Most patients undergoing CPB suffer pulmonary dysfunction [11,[13][14][15] . A recent study concluded that compared with CABG with CPB, CABG without CPB reduced postoperative respiratory and renal morbidity, and shortened ICU and hospital stay in high-risk patients [12] .
In the present study, the mean time of CPB was 58.85±21.89 minutes. The study by Romanini et al. [16] showed average time of CPD similar to this. Other studies presented higher time of  [5][6]12,17] , and a recent epidemiological study showed smaller mean time of CPD (46.7±22.9 min) for patients underwent CABG [7] .
Akdur et al. [10] evaluated the length of MV on CABG postoperative period and observed that MV was longer in patients undergoing longer duration of CPB. The average time of CPB observed in our study (58.85±21.89 minutes) was smaller than the average times of both groups in that study and perhaps this might explain why we did not find correlation between time of CPB and IT.
CPB time longer than 91 minutes was shown to be an independent predictor of prolonged MV [5] . A recent publication [6] concluded that postoperative prolonged MV was associated with longer perfusion times in CABG patients, and showed that every 1-minute increase over 82.5 minutes of CPB time increased risk of delayed extubation by 3.5%. Regarding the observed differences in our results when they are compared with the two studies mentioned above, it is important to note that the average CPB time observed in our study was very shorter than average time of those studies and very few patients in our study had more than 91 minutes of CPB time. Furthermore, in the second study [6] , the patient population was comprised of a heterogeneous group of patients who had concomitant procedures along with CABG, such as valve procedures, features that may also explain at least in part the differences found.
In several studies, smoking has not been related with increased IT in the postoperative of CABG [2,6,8,10] , which corroborates our results. However, a recent study accessed a total of 3730 patients undergoing isolated CABG and reported that the smoking group had significantly longer duration of MV [18] .
In this study, the prevalence of previous AMI was 23.1%. An epidemiological study conducted in a Brazilian institution showed a prevalence of 46.9% for previous among AMI 3010 patients undergoing CABG [7] .