RESEARCH

Predictors of prolonged mechanical ventilation after coronary artery bypass grafting among Filipino adults with coronary artery disease

SPMC J Health Care Serv. 2016;2(1):9. ark: http://n2t.net/ark:/76951/jhcs322urh

Joseph Jasper S Acosta,1 Jessie F Orcasitas1,2,3,4,5,6,7,8,9


1Department of Internal Medicine, Southern Philippines Medical Center, Bajada, Davao City, Philippines;
2Department of Internal Medicine, Brokenshire Memorial Hospital, Brokenshire Heights, Madapo, Davao City, Philippines;
3Pulmonary Section, Metro Davao Medical and Research Center, Bajada, Davao City, Philippines;
4Department of Internal Medicine, Davao Doctors Hospital, Quirino Avenue, Davao City, Philippines;
5Department of Internal Medicine, San Pedro Hospital of Davao City Inc., C Guzman St, Davao City, Philippines;
6Department of Internal Medicine, Davao Adventist Hospital, Bangkal, Davao City, Philippines;
7Department of Internal Medicine, Ricardo Limso Medical Center, Ilustre St, Davao City, Philippines;
8Department of Internal Medicine, Malta Medical Center, Toril, Davao City, Philippines;
9Community Health & Development Cooperative Hospital, Anda Riverside, Davao City, Philippines


Correspondence Joseph Jasper S Acosta, jj.acosta86@gmail.com
Received 16 September 2016
Accepted 21 December 2016
Cite as Acosta JJ, Orcasitas J. Predictors of prolonged mechanical ventilation after coronary artery bypass grafting among Filipino adults with coronary arterial disease. SPMC J Health Care Serv. 2016;2(1):9. http://n2t.net/ark:/76951/jhcs322urh


Abstract

Background. Identifying risk factors for prolonged mechanical ventilation (PMV) can improve postoperative outcomes of patients undergoing coronary artery bypass grafting (CABG).

Objective. To determine the occurrence rate and predictors of PMV among patients who underwent CABG.

Design. Retrospective cohort study.

Setting. Southern Philippines Medical Center Heart Institute, Davao City, Philippines.

Participants. 213 patients with coronary artery disease (CAD) who underwent CABG.

Main outcome measures. PMV occurrence rate; odds ratios (95% CI) of PMV for selected clinical characteristics.

Main results. There were 167 (78.4%) males and 46 (21.6%) females in this study. The patients had a mean age of 60.2 ± 9.68 years and a mean BMI of 25.8 ± 5.65 kg/m2. Post-CABG, PMV occurred in 18.87% of the patients. Univariate odds ratios of PMV were significantly high for renal dysfunction (OR=2.75; 95% CI 1.34-5.66), New York Heart Association functional class IV (7.53; 3.07-18.46), angina grade IV (4.52; 1.69-12.07), left ventricular ejection fraction <50% (2.80; 1.23-6.38), cardiogenic shock (12.14; 2.26-65.11), intraoperative IABP insertion (3.17; 1.46-6.88), postoperative acute kidney injury (AKI) (6.72; 2.99-15.10), postoperative hemodialysis (4.84; 2.21-10.60), postoperative neurological complications (13.04; 4.21-40.39), postoperative arrhythmia (2.59; 1.19-5.63), pulmonary complications (3.50;1.67-7.34), and other complications (3.44;1.22-9.68). On multiple regression analysis, AKI after CABG significantly increased the odds ratio of PMV (11.82;1.03-135.35).

Conclusion. PMV after CABG occurred in 18.87% of the patients in our study and was associated with poor preoperative cardiac and renal conditions, intraoperative IABP insertion, and postoperative complications. The development of AKI after CABG independently increased the odds ratio of PMV.


Keywords. early extubation, acute kidney injury, EuroSCORE II, Society of Thoracic Surgeons Adult Cardiac Surgery Risk


Introduction

Prolonged mechanical ventilation (PMV) after coronary artery bypass grafting (CABG) increases postoperative morbidity, mortality, and the cost of hospitalization.1 Post-cardiac surgery patients who are reintubated following extubation are likewise prone to more complications and have a higher mortality rate.2 3 In recent years, early extubation (EE)—or extubation within 8 hours of arrival at the postoperative care unit—has gained popularity because the practice has been shown to improve cardiac performance, reduce respiratory complications, allow early mobilization and feeding, increase patient autonomy and comfort,4 and reduce the workload of medical and nursing staff.5 EE, however, may not apply to all patients such as those who are at high risk for postoperative complications.4 5

Identifying patients at high risk of PMV can help physicians optimize health care to improve the outcomes of patients undergoing CABG. However, the key factors associated with early PMV and EE are poorly understood.6 To start with, PMV has been defined differently from study to study, with some setting the cutoff at 12 hours or less after surgery,2 6 7 and then others at 24 hours,8 9 10 11 48 hours,3 12 13 or even 72 hours14 15 postoperatively. Studies with PMV cutoffs lower than 24 hours usually report only a few predictors for PMV.6 7 In a study that defined PMV as mechanical ventilation for more than 12 hours after CABG, redo surgery, cardiopulmonary bypass (CPB) time of more than 91 minutes, intraoperative transfusion of more than 4 units of red blood cells, and left ventricular ejection fraction (LVEF) of ≥30% all increased the odds of having PMV.2 Another study, which set the PMV cutoff at 72 hours post-CABG, reported that advanced age, renal dialysis, peripheral vascular disease, hypertension, advanced stage of heart failure, elevated body mass index (BMI), reduced forced expiratory volume at 1 second (FEV1), and prolonged CPB are all associated with PMV.15 Age >70 years, diabetes, and the use of an intra-aortic balloon pump (IABP) have been identified as significant predictors of failure of EE among patients on off-pump CABG.16 Another study suggested that reducing the CPB time and keeping blood glucose levels low during CPB can help avoid delayed extubation.10

The predictors of PMV identified in these studies were varied, and sometimes conflicting. One study was able to show that the predictors of PMV may even change within the same institution, owing to the changes in patient demographics, and surgical and anesthetic techniques over time.3 This suggests that predictors are context-specific and that the application of previous findings on predictors of PMV cannot be fully extended to any other patient groups. We thought of doing this study since, as far as we know, there were no previous attempts to explore the risk factors for PMV among Filipino patients undergoing CABG. In this study, we aimed to determine the rate of occurrence of PMV after CABG among patients with CAD and to identify the preoperative, intraoperative and postoperative factors that are associated with PMV.

Methodology

Study design and setting
We employed a retrospective cohort study design in reviewing and analyzing the medical records of patients who underwent CABG at the Southern Philippines Medical Center Heart Institute (SPMC-HI) in Davao City, Philippines. Cardiac interventions and open heart surgical procedures have been offered by SPMC-HI since its opening in 2007. The institute caters to an average of 428 surgical procedures per year, around 15% of which are CABG.

Participants
All adult patients aged 19 to 75 years old with coronary artery disease (CAD) who underwent CABG at the SPMC-HI from 2007 to 2015 were eligible for inclusion in this study. We excluded patients who had previous percutaneous cardiac intervention or cardiac surgery.

Data collection
We collected sociodemographic data, which included age, sex, weight, and height. We also gathered clinical data such as smoking history, history of myocardial infarction within 3 months pre-CABG, comorbidities (hypertension, diabetes and chronic obstructive pulmonary disease [COPD]), Canadian Cardiovascular Society Angina Grading Scale (CCS AGS), New York Heart Association functional (NYHA FC) classification of heart failure, number of diseased vessels, and LVEF from the last 2D echocardiography prior to the CABG procedure.

We calculated BMI from the weight and height of each patient. The glomerular filtration rate (GFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula 17 available online. We also determined the chronic kidney disease (CKD) stage of the patient, albeit based on the calculated GFR alone and without considering urine findings or structural abnormalities. For each patient, we either noted the EuroSCORE (European System of Cardiac Operative Risk Evaluation) II18 and the online Society of Thoracic Surgeons Adult Cardiac Surgery Risk (STS ACSR) v2.8119 score as recorded in the patient’s chart, or calculated one or both scores using their respective calculators available online. Both EuroSCORE II and STS ACSR were based on models developed from pooled patients in Europe and America, respectively, and have been repeatedly validated in several studies.20 21 22 23 Both scores are commonly used to calculate the risks of morbidity and mortality from cardiac procedures and are presently used to screen patients for CABG under the Z-package program of the Philippine Health Insurance Corporation.24

We also collected data on preoperative, intraoperative and postoperative clinical characteristics of patients in relation to the CABG procedure. The preoperative variables we noted included demographic characteristics, comorbidities, severity of CAD, preoperative intubation status, presence of cardiogenic shock (with inotropic or intra-aortic balloon pump support), and the urgency of the surgery (elective or emergency). The intraoperative variables we collected were cardiopulmonary bypass duration (CBD) time, aortic cross-clamp (ACC) time, whether CABG was done off-bypass, number of units of blood products transfused, the use of Cell Saver® (autologous blood transfusion), the use of IABP, and the use of inotropic agents. We also gathered postoperative variables such as hematocrit level, PaO2/FiO2 ratio, and the occurrence of postoperative complications, namely, acute kidney injury (AKI; as defined by the Kidney Disease: Improving Global Outcomes [KDIGO]25—increase in creatinine by more than 0.3 mg/dl [26.5 μmol/L] from baseline within 48 hours or increase to 1.5 times from baseline within seven days), neurological sequelae, new onset arrhythmia and/or myocardial infarction, acute respiratory distress syndrome, pulmonary complication (nosocomial pneumonia, pleural effusion, atelectasis, pulmonary congestion, pulmonary edema), and other non-pulmonary complications (upper gastrointestinal bleed, decubitus ulcer, device related infection, surgical site infection, ischemic hepatitis, postoperative myasthenia gravis).

The main outcome measures for this study were the rate of occurrence and predictors of PMV. Since it has been the prevailing practice among cardiologists and pulmonologists in our institution to extubate patients the day—within 24 hours—after performing CABG, we operationally defined PMV as mechanical ventilation beyond 24 hours after the end of the CABG procedure. Patients who were immediately extubated (within 24 hours postoperatively) but who required reintubation within 72 hours of weaning from the mechanical ventilator were also classified as belonging to the PMV group. Patients who were successfully extubated within 24 hours from end of the CABG procedure comprised the EE group.We also looked at other outcomes, including mortality, length of hospital stay, and length of intensive care unit (ICU) stay.

The varying value of the sample size n in the Results section of this report reflects some missing data from patient records. Of the 213 patients included in this study, one patient died within 24 hours postoperatively, so only 212 could be assessed for the outcome of either EE or PMV. Two additional patients were transferred to another institution within 24 hours postoperatively, but after extubation, so only 210 could be assessed for mortality after 24 hours. Data on duration of stay were either missing from or inaccurate in some patient records, so only 209 could be assessed for length of ICU stay, and only 207 could be assessed for length of hospital stay.

Statistical analysis
We summarized continuous variables as means and standard deviations, and categorical variables as frequencies and percentages. We then used student t-test to compare continuous variables and chi-square test or Fisher’s exact test (for variables with frequency of <5) to compare categorical variables. To calculate the odds ratios (95% confidence intervals) of having PMV for each predetermined preoperative, intraoperative or postoperative variable, we performed univariate logistic regression. We dichotomized non-binary variables prior to performing logistic regression. For continuous variables, we determined the following cutoff points based either on the mean or on a clinically significant value: age >60 years; abnormal BMI (either less than 18.5 or more than 25); GFR <60 mL/min/1.73 m2 (renal dysfunction); LVEF <50%; EuroSCORE II mortality >5.0%; STS ACSR mortality >5.0%; CBP duration >180 min; ACC time >150 min; transfusion of packed red blood cells (PRBC) and/or whole blood (WB) >4 units; transfusion of total blood products >10 units; use of >2 kinds of inotropic agents; hematocrit level <0.38 for males or <0.34 for females; and PaO2/FiO2 ratio <200. We also dichotomized categorical variables with more than two categories as follows: NYHA FC IV (yes/no); CCS AGS IV (yes/no); and 3-vessel CAD (yes/no). For factors that significantly increased or decreased the odds ratios of having PMV by univariate logistic regression, we performed multivariable logistic regression to determine independent predictors of the outcome. For all statistical tests, the level of significance was set at <5%. All statistical tests were performed using Epi Info 7.1.4.0.

Results

A total of 213 patient charts were included in the analysis for this study. The summary of the sociodemographic and clinical profiles of patients are shown in Table 1. The mean age of the patients was 60.20 ± 9.68 years old, and the mean BMI was 25.80 ± 5.65 kg/m2. Among the patients 167/213 (78.40%) were males and 46/213 (21.60%) were females. Half of the patients (108/213, 50.70%) were smokers. Hypertension was the most common comorbidity, with 154/213 (72.30%) of the patients having it. More than half of the patients (123/213; 57.75%) had diabetes mellitus, while 17/213 (7.98%) had COPD.

The mean GFR was 71.48 ± 26.31 mL/min/1.73 m2; 69/206 (33.50%) patients had renal dysfunction (GFR <60 ml/min/1.73 m2), and 6/213 (2.82%) were on hemodialysis prior to the CABG procedure. The most frequent CKD stage among the patients was stage 2 (82/206; 39.81%).

In terms of cardiovascular status, the most frequent CCS AGS category among the patients was class II (41/123; 33.33%), while the most frequent NYHA FC was II (80/199; 40.20%). There were 73/211 (34.60%) patients who had a history of myocardial infarction within 3 months prior to CABG. The mean LVEF was 50.33 ± 13.54%, and 59/140 (42.14%) of the patients had LVEF less than 50%. Most of the patients (172/212; 81.13%) had 3-vessel CAD. Prior to CABG, 8/213 (3.76%) patients had cardiogenic shock, 4/213 (1.88%) were intubated for an indication that occurred preoperatively, and 2/213 (0.94%) had to undergo emergency CABG.

Table 1    Demographic and preoperative clinical characteristics of patients

Characteristics n* Value
Mean age ± SD, years 213 60.20 ± 9.68
Sex, frequency (%) 213
   Male 167 (78.40)
   Female 46 (21.60)
Mean BMI ± SD,kg/m2 212 25.80 ± 5.65
Smoker, frequency (%) 213 108 (50.70)
Hypertension, frequency (%) 213 154 (57.75)
Diabetes mellitus,frequency (%) 213 123 (57.75)
COPD, frequency (%) 213 17 (7.98)
Mean GFR ± SD, mL/min/1.73m2 206 71.48 ± 26.31
CKD stage†, frequency (%) 206
   1 58 (28.16)
   2 82 (39.81)
   3A 29 (14.08)
   3B 25 (12.14)
   4 6 (2.91)
   5 6 (2.91)
Renal dysfunction‡, frequency (%) 206 69 (33.50)
On hemodialysis, frequency (%) 213 6 (2.82)
CAD severity
   CCS AGS, frequency (%) 213
     I 34 (27.64)
     II 41 (33.33)
     III 12 (9.76)
     IV 36 (29.27)
   NYHA FC, frequency (%) 199
     I 59 (29.65)
     II 80 (40.20)
     III 35 (17.59)
     IV 25 (12.56)
   Recent Myocardial Infarctions, frequency (%) 211 73 (34.60)
   LVEF 140
     Mean ± SD, % 140 50.33 ± 13.54
     <50%, frequency (%) 140 59 (42.14)
   Number of diseased coronary vessels, frequency (%) 212
     1 12 (6.13)
     2 27 (12.74)
     3 172 (81.13)
Preoperative condition
   Cardiogenic shock‖, frequency (%) 213 8 (3.76)
   Intubated preoperatively, frequency (%) 213 4 (1.88)
   Urgency of procedure, frequency (%) 213
     Elective 211 (99.06)
     Emergency 2 (0.94)
BMI=body mass index; CAD=coronary artery disease; CCS AGS=Canadian Cardiovascular Society Angina Grading Scale; CKD=chronic kidney disease; COPD=chronic obstructive pulmonary disease; GFR=glomerularfiltration rate; IABP=intra-aortic balloon pump; LVEF=left ventricular ejection fraction; NYHA FC=New York Heart Association Functional Class.
* Value of n varies because of missing data.
† CKD stage solely based on calculated GFR.
‡ Patients with GFR <60mL/min/1.73m2
§ Within 3 months.
‖ Inotropes or IABP.

Of 212 patients, 172 (81.13%) had EE (within 24 hours postoperatively), while 40 (18.87%) had PMV (after 24 hours postoperatively). Table 2 shows the comparison of outcomes between patients who had EE and those with PMV. The PMV group had significantly higher mortality rate after 24 hours (10/39; 25.64% versus 3/171; 1.75%); <0.001), longer mean length of ICU stay (222.48 ± 292.02 hours versus 113.24 ± 69.58 hours; p<0.001), and longer mean length of hospital stay (30.08 ± 411.24 days versus 17.95 ± 9.35 days; p<0.001) compared to patients in the EE group.

Table 2    Comparison of outcomes between patients who had early extubation (EE) and patients with prolonged mechanical ventilation (PMV)

Outcomes n* EE n* PMV p-value
Mortality after 24 hours, frequency (%) 171 3 (1.75%) 39 10 (25.64%) <0.001†
Mean length of ICU stay ± SD, hours
   Overall 170 113.24 ± 69.58 39 222.48 ± 292.02 <0.001†
   Survivors‡ 167 113.24 ± 69.88 29 165.16 ± 119.37 0.0013†
Mean length of hospital stay ± SD, days
   Overall 170 17.95 ± 9.35 37 30.08 ± 411.24 <0.001†
   Survivors‡ 167 17.89 ± 9.33 27 31.30 ± 376.52 <0.001†
ICU=intensive care unit.
* Value of n varies because of missing data.
† Significant at p<0.05.
‡ Among those who survived.

Table 3 shows the comparison of demographic and preoperative clinical characteristics between patients who had EE and patients with PMV. The preoperative GFR of patients with PMV (57.68 ± 28.46 mL/min/1.73 m2) was significantly lower compared to that of patients who had EE (74.77 ± 24.83 mL/min/1.73 m2; p=0.0003). There were more patients with early-stage CKD in the EE group (53/167; 31.74% with CKD stage 1 and 67/167; 40.12% with CKD stage 2) than in the PMV group (5/38; 13.51% with CKD stage 1 and 15/38; 39.47% CKD stage 2). The PMV group had significantly higher proportions of patients with renal dysfunction (20/38; 52.63% versus 48/167; 28.74%; p=0.0047) and patients on hemodialysis (3/40; 7.50% versus 3/172; 1.74%; p=0.0480) compared to the EE group.

The most frequent angina CCS AGS classifications were class II among patients who had EE (37/101; 36.63%) and class IV among patients with with PMV (12; 57.14%; p=0.0158). Likewise, the most frequent NYHA FC were class II among patients who had EE (67/159; 42.14%) and class IV among patients with with PMV (14/39; 35.90%; p<0.0001). The PMV group had significantly higher proportions of patients with LVEF less than 50% (19/31; 61.29%), in cardiogenic shock (5/40; 12.50%), patients intubated preoperatively (4/39; 10.00%) and patients requiring emergency CABG (2/40; 5.00%) compared to the EE group. Patients with PMV group had significantly higher mean risk of mortality by EuroSCORE II (5.09 &pluamn; 4.97%) and mean mean risk of mortality by STS ACSR (3.85 ± 4.54%) than those who had EE. The rest of the demographic and preoperative clinical characteristics were comparable in between the two groups.

Table 3    Comparison of preoperative characteristics between patients who had early extubation (EE) and patients with prolonged mechanical ventilation (PMV)

Characteristics n* EE n* PMV p-value
Mean age ± SD, years 172 60.04 ± 9.63 40 60.51 ± 10.01 0.7506
Sex, frequency (%) 172 40 0.8913
   Male 135 (78.49) 31 (77.50)
   Female 37 (21.51) 9 (22.50)
BMI 172 40
   Mean ± SD, kg/m2 26.08 ± 6.00 24.58 ± 3.58 0.1307
   Obese†, frequency (%) 104 (60.47) 21 (52.50) 0.3563
   Malnourished patients‡, frequency (%) 7 (4.07) 3 (72.50) 0.3567
Hypertension, frequency (%) 172 124 (72.09) 40 29 (72.50) 0.9587
Diabetes, frequency (%) 172 102 (59.30) 40 20 (50.00) 0.2836
COPD, frequency (%) 172 12 (6.98) 40 4 (10.00) 0.5144
Mean GFR ± SD, mL/min/1.73m2 167 74.77 ± 27.83 38 57.68 ± 28.46 0.0003§
CKD stage‖, frequency (%) 167 38 0.0213§
     1 53 (31.74) 5 (13.51)
     2 67 (40.12) 15 (39.47)
     3A 23 (13.77) 5 (13.16)
     3B 18 (10.78) 7 (18.24)
     4 3 (1.80) 3 (7.89)
     5 3 (1.80) 3 (7.89)
Renal dysfunction¶, frequency (%) 167 48 (28.74) 38 20 (52.63) 0.0047§
On hemodialysis, frequency (%) 172 3 (1.74) 40 3 (7.50) 0.0480§
Smoker, frequency (%) 172 86 (50.00) 40 21 (52.50) 0.7758
CCS AGS, frequency (%) 101 21 0.0158§
   I 31 (30.96) 3 (14.29)
   II 37 (36.63) 4 (19.05)
   III 10 (9.90) 2 (9.52)
   IV 23 (22.77) 12 (57.14)
NYHA FC, frequency (%) 159 39 <0.0001§
   I 51 (32.08) 8 (20.51)
   II 67 (42.14) 12 (30.77)
   III 30 (18.87) 5 (12.82)
   IV 11 (6.92) 14 (35.90)
Recent myocardial infarction**, frequency (%) 170 54 (31.76) 40 18 (45.00) 0.1452
Mean LVEF ± SD, % 109 51.72 ± 12.80 31 46.16 ± 15.04 0.0516
LVEF <50%, frequency (%) 109 40 (36.70) 31 19 (61.29) 0.0144§
Number of diseased vessels, frequency (%) 171 40 0.7940
   1 10 (5.85) 3 (7.50)
   2 23 (13.45) 33 (82.50)
   3 138 (80.70) 33 (82.50)
Preoperative condition, frequency (%)
   Cardiogenic shock†† 172 2 (1.16) 40 5 (12.50) 0.0003§‡‡
   Intubated preoperatively 172 0 (0.00) 40 4 (10.00) <0.0001§‡‡
Urgency of procedure, frequency (%) 172 40 0.0032§‡‡
   Emergency 0 (0.00) 2 (5.00)
   Elective 172 (100.00) 38 (95.00)
Mean % risk of mortality by EuroSCORE II ± SD 92 2.16 ± 2.17 26 5.06 ± 4.97 <0.0001§
Mean % risk of mortality by STS ACSR ± SD 102 1.74 ± 3.04 26 3.85 ± 4.54 0.0055§
BMI=body mass index; CCS AGS=Canadian Cardiovascular Society Angina Grading Scale; CKD=chronic kidney disease; COPD=chronic obstructive pulmonary disease; EuroSCORE II=European System for Cardiac Operative Risk Evaluation II; GFR=glomerular filtration rate; IABP=intraaortic balloon pump; LVEF=left ventricular ejection fraction; NYHA FC=New York Heart Association Functional Class; RF=renal failure; STS ACSR=Society of Thoracic Surgeons Adult Cardiac Surgery Risk.
* Value of n varies because of missing data.
† BMI >25 kg/m2.
‡ BMI <18.5 kg/m2.
§ Significant at p<0.05.
‖ CKD stage solely based on calculated GFR
¶ GFR <60 mL/min/1.73m2.
** Within 3 months.
†† Inotropes or IABP.
‡‡ Fisher’s exact test.

Comparison of the intraoperative characteristics between patients who had EE and those with PMV are presented in Table 4. The mean CPB time of patients with PMV (200.23 ± 51.93 minutes) was significantly higher compared to that of patients who had EE (180.96 ± 41.87 minutes; p=0.0147). Likewise, the mean ACC time of patients with PMV (162.79 ± 46.16mins) was significantly higher than that of patients who had EE (149.54 ± 34.81 minutes; p=0.0476). The PMV group (14/40; 35.00%) had a significantly higher proportion of patients who had intraoperative insertion of IABP, than those in EE group (25/172; 14.53%; p=0.0026). Similarly, the proportion of patients who were given more than two kinds of inotropes intraoperatively in the PMV group (6/40; 15.00%) was significantly higher than that in the EE group (10/172; 5.85%; p=0.0476). The two groups were comparable in terms of the rest of the intraoperative characteristics.

Table 4    Comparison of intraoperative characteristics between patients who had early extubation (EE) and patients with prolonged mechanical ventilation (PMV)

Characteristics n* EE n* PMV p-value
CPB duration 163 38
   Mean ± SD, minutes 180.96 ± 41.87 200.23 ± 51.93 0.0147†
   Proportion of CPB >180mins 78 (47.56) 25 (64.10) 0.0633
Mean ACC time ± SD, minutes 161 149.54 ± 34.81 39 162.79 ± 46.16 0.0476†
OPCAB, frequency (%) 171 8 (4.65) 39 0 (0.00) 0.3571‡
Mean number of units of blood products ± SD
   PRBC 172 2.13 ± 1.88 40 2.55 ± 1.97 0.2193
   WB 172 0.70 ± 1.23 40 0.63 ± 1.21 0.7164
   PC 172 2.48 ± 2.45 40 2.28 ± 2.63 0.6440
   FFP 172 1.75 ± 1.73 40 1.73 ± 2.15 0.9376
   Total 172 9.97 ± 6.89 40 10.55 ± 7.80 0.6413
Use of Cell Saver®, frequency (%) 172 58 (33.72) 40 13 (32.50%) 0.8828
Intraoperative insertion od IABP§, frequency (%) 172 25 (14.53) 40 14 (35.00) 0.0026†
Use of <2 kinds of inotropes, frequency (%) 172 10 (5.85) 40 6 (15.00) 0.0476†
Other surgical procedures 172 6 (3.49) 40 1 (2.50) 0.6072‡
ACC=anodal closure contraction; CPB=cardiopulmonary bypass; FFP=fresh frozen plasma; IABP=intra-aortic balloon pump; OPCAB=off-pump coronary artery bypass; PC=platelet concentrate; PRBC=packed red blood cells; WB=whole blood.
* Value of n varies because of missing data.
† Significant at p<0.05.
‡ Fisher’s exact test.
§ Excluding those with IABP preoperativley.

Table 5 shows the comparison of the postoperative characteristics of patients who had EE and patients with PMV. Patients with PMV had significantly higher frequencies of the following postoperative complications: AKI (25/35; 71.43% versus 45/166; 27.11%; p<0.0001), hemodialysis (16/37; 43.24% versus 23/169; 13.61%; p<0.0001), neurologic sequelae (12/40; 30.00% versus 5/172; 2.91%; p<0.0001), arrhythmia (13/40; 32.50% versus 27/172; 15.70%; p=0.0144), other pulmonary complications (17/40; 42.50% versus 30/172; 17.44%; p=0.0006), and other non-pulmonary complications (7/39; 17.50% versus 10/171; 5.81%; p=0.0142) compared to patients who had EE. Other postoperative characteristics between the two groups were comparable.

Table 5    Comparison of postoperative characteristics between patients who had early extubation (EE) and patients with prolonged mechanical ventilation (PMV)

Characteristics n* EE n* PMV p-value
Mean hematocrit level ± SD 170/td> 0.3474 ± 0.0498 38 0.3308 ± 0.0444 0.4555
PaO2/FiO2
   Mean ± SD 169 272.01 ± 119.50 39 277.23 ± 215.33 0.8363
PaO2/FiO2 <200, frequency (%) 169 45 (27.11) 39 13 (35.90) 0.2157
Acute kidney injury† 166 45 (27.11) 35 25 (71.43) <0.0001‡
New indication for hemodialysis† 169 23 (13.61) 37 16 (43.24) <0.0001‡
Neurologic sequelae 172 5 (2.91) 40 12 (30.00) <0.0001‡
Arrhythmia 172 27 (15.70) 40 13 (32.50) 0.0144‡
New myocardial infarction‖ 172 2 (1.16) 40 2 (5.00) 0.1621§
Acute respiratory distress syndrome 172 10 (5.81) 40 5 (12.50) 0.1374
Other pulmonary complications¶ 172 30 (17.44) 40 17 (42.50) 0.0006‡
other non-pulmonary complications†† 171 10 (5.81) 39 7 (17.50) 0.0142‡
FiO2=fraction of inspired oxygen; PaO2=partial pressure of oxygen in arterial blood.
* Value of n varies because of missing data.
† Excluding those on hemodialysis preoperatively.
‡ Significant at p<0.05.
§ Fisher’s exact test.
‖ Excluding pre-existent myocardial infarction.
¶ Including nosocomial pneumonia, pleural effusion, atelectasis, pulmonary congestion, pulmonary edema
†† Including upper gastrointestinal bleed, decubitus ulcer, device related infection, surgical site infection, ischemic hepatitis, postoperative myasthenia gravis.

Presented in table 6 are the univariate odds ratios of PMV for preoperative, intraoperative and postoperative characteristics. Among the preoperative variables, renal dysfunction (OR=2.75; 95% CI 1.34 to 5.66; p=0.0058), NYHA FC IV (OR=7.53; 95% CI 3.07 to 18.46; p=<0.0001), CCS AGS IV (OR=4.52; 95% CI 1.69 to 12.07; p=0.0026), LVEF <50% (OR=2.80; 95% CI 1.23 to 6.38; p=0.0141), cardiogenic shock (OR=12.14; 95% CI 2.26 to 65.11; p=0.0036), EuroSCORE II mortality >5.0% (OR=5.56; 95% CI 1.88 to 16.46; p=0.0020), and STS ACSR mortality >5.0% (OR=7.34; 95% CI 1.90 to 28.42; p=0.0039) all significantly increased the odds ratio of having PMV. Only intraoperative insertion of IABP (OR=3.17; 95% CI 1.46 to 6.88; p=0.0036) among the intraoperative variables increased the odds ratio of PMV. Postoperative variables that significantly increased the odds of PMV include postoperative AKI (OR =6.72, 95% CI 2.99 to 15.10; p=<0.0001), a new postoperative indication for hemodialysis (OR =4.84, 95% CI 2.21 to 10.60; p=<0.001), postoperative neurological complications (OR =13.04, 95% CI 4.21 to 40.39; p=0.0001), postoperative arrhythmia (OR =2.59, 95% CI 1.19 to 5.63; p=0.0168), other pulmonary complications (i.e., nosocomial pneumonia, pleural effusion, atelectasis, pulmonary congestion, or pulmonary edema) (OR=3.50, 95% CI 1.67 to 7.34; p=0.0009), and other non-pulmonary complications (i.e., surgical site infection, upper gastrointestinal bleed, decubitus ulcer, device related infection, surgical site infection, ischemic hepatitis, or postoperative myasthenia gravis) (OR =3.44, 95% CI 1.22 to 9.68; p=0.0195).

Table 6    Univariate odds ratios (95% CI) of having prolonged mechanical ventilation (PMV) for selected preoperative, intraoperative, and postoperative characteristics of patients who underwent coronary artery bypass grafting (CABG)

Characteristics Odds ratio (95% CI) p-value
Preoperative variables
   Demographic
     Age <60 years 1.12 (0.56 to 2.24) 0.7403
     Male 0.94 (0.41 to 2.16) 0.8899
     Abnormal BMI* 0.81 (0.40 to 1.64) 0.5606
     Hypertension 1.02 (0.47 to 1.64) 0.9587
     Diabetes 0.69 (0.34 to 1.37) 0.2851
     Smoking 1.11 (0.56 to 2.20) 0.7758
     COPD 1.48 (0.45 to 4.86) 0.5162
     Renal dysfunction 2.75 (1.34 to 5.66) 0.0058†
     ESRD on hemodialysis 4.57 (0.89 to 23.53) 0.0694
   CAD severity
     NYHA FC IV 7.53 (3.07 to 18.46) <0.0001†
     CCS AGS IV 4.52 (1.69 to 12.07) 0.0026†
     Recent MI 1.76 (0.87 to 3.55) 0.1150
     LVEF <50% 2.80 (1.23 to 6.38) 0.0141†
     3 vessel disease 1.13 (0.46 to 2.77) 0.7940
   Preoperative condition
     Cardiogenic shock 12.14 (2.26 to 65.11) 0.0036†
     EuroSCORE II mortality >5.0% 5.56 (1.88 to 16.46) 0.0020†
     STS ACSR mortality >5.0% 7.34 (1.90 to 28.42) 0.0039†
Intraoperative variablesz
   CPB duration >180 mins 1.97 (0.96 to 4.05) 0.0662
   ACC time >150 mins 1.75 (0.85 to 3.57) 0.1271
   PRBC+WB >4 1.05 (0.44 to 2.50) 0.9053
   Total blood >10 0.77 (0.39 to 1.55) 0.4704
   Use of Cell Saver® 0.95 (0.45 to 1.97) 0.8831
   Intraoperative insertion of IABP 3.17 (1.46 to 6.88) 0.0036†
   Use of >2 kinds of intropes 2.86 (0.97 to 8.40) 0.0561
   Other procedures 0.71 (0.08 to 6.06) 0.7540
Postoperative variables
   Low hematocrit 1.01 (0.43 to 2.39) 0.9865
   Low paO2:FiO2 ratio 1.59 (0.76 to 3.33) 0.2179
   Acute kidney injury 6.72 (2.99 to 15.10) <0.0001†
   New indication for hemodialysis 4.84 (2.21 to 10.60) <0.0001†
   Neurological complications 13.04 (4.21 to 40.39) 0.0001†
   Arrythmia 2.59 (1.19 to 5.63) 0.0168†
   New myocardial infarction 4.47 (0.61 to 32.77) 0.1403
   other pulmonary complications‡ 3.50 (1.67 to 7.34) 0.0009†
   Other non-pulmonary complications§ 3.44 (1.22 to 9.68) 0.0195†
ACC=anodal closure contraction; BMI=body mass index; CAD=coronary artery disease; CCS AGS=Canadian Cardiovascular Society Angina Grading Scale; COPD=Chronic Obstructive Pulmonary Disease; CPB=cardiopulmonary bypass; ESRD=end stage renal disease; EuroSCORE II=European System for Cardiac Operative Risk Evaluation II; IABP=intra-aortic balloon pump; LVEF=left ventricular ejection fraction; NYHA FC=New York Heart Association Functional Class; PRBC=packed red blood cells; STS ACSR=Society of Thoracic Surgeons Adult Cardiac Surgery Risk; WB=whole blood.
* BMI <18.5 and >25
† Significant at p<0.05.
‡ Including nosocomial pneumonia, pleural effusion, atelectasis, pulmonary congestion, pulmonary edema.
§ Including upper gastrointestinal bleed, decubitus ulcer, device related infection, surgical site infection, ischemic hepatitis, postoperative myasthenia gravis.

All significant variables in the univariate logistic regression analyses were entered into a multiple regression model. The multivariable odds ratio (95% CI) of PMV are presented in Table 7. In this regression model, only AKI independently increased the odds ratio of having PMV (adjusted OR=11.82; 95% CI (1.03 to 135.35; p=0.0470).

Table 7    Multivariable odds ratio (95% CI) of having prolonged mechanical ventilation (PMV)

Characteristics Adjusted odds ratio (95% CI) p-value
Renal dysfunction 1.16 (0.13 to 10.15) 0.8913
NYHA FC IV 1.64 (0.10 to 28.15) 0.7342
CCS Angina IV 7.74 (0.49 to 121.28) 0.1449
LVEF <50 0.52 (0.07 to 3.96) 0.5256
Preoperative cardiogenic shock 10.79 (0.13 to 913.76) 0.2937
Intraoperative insertion of IABP 8.43 (0.97 to 72.96) 0.0530
Acute kidney injury 11.82 (1.03 to 135.35) 0.0470*
New indication for hemodialysis 2.75 (0.27 to 27.68) 0.3909
Neurological complications 2.98 (0.17 to 52.24) 0.4549
Arrhythmia 2.32 (0.32 to 17.16) 0.4069
Other pulmonary complications 5.27 (0.5923 to 46.9745) 0.1361
Other non-pulmonary complications 1.94 (0.1232 to 30.4484) 0.6382
CCS AGS=Canadian Cardiovascular Society Angina Grading Scale; IABP=intra-aortic balloon pump; LVEF=left ventricular ejection fraction; NYHA FC=New York Heart Association Functional Class.
* Significant at p<0.05.


Discussion

Key results
The frequency of PMV after CABG among patients with CAD in this study was 18.87%. Preoperative patient characteristics that increase the odds ratio of having PMV include renal dysfunction, cardiogenic shock, NYHA FC IV, CCS AGS class IV, LVEF <50%, greater than 5% mortality risk by EuroSCORE II, and greater than 5% mortality risk by STS ACSR. The intraoperative use of IABP and the postoperative development of AKI, neurological complications, arrhythmia, other pulmonary and non-pulmonary complications, as well as a new postoperative indication for hemodialysis, all significantly increase the odds ratio of having PMV. Multivariable regression analysis showed that postoperative AKI is an independent predictor of PMV.

Strengths and limitations
This is the first systematic and comprehensive description of the demographic and clinical profiles of patients with CAD who underwent CABG have been reported in our heart institute since its opening in 2007. To the best of our knowledge, this is also the first time that predictors of PMV have been explored among Filipino patients belonging to this subpopulation.

We identified some limitations that are inherent to the retrospective design of this study. Some important data on potential predictors and outcomes, which could possibly influence our results, were not reflected in the patient records that we reviewed. Moreover, we were not able to accurately account for potential variations in medical and surgical technology, surgical technique and experience, as well as prevailing health care approach to patients, throughout the 8-year span covered by this study. It is possible that these variations have significant impact on the outcomes that we measured in this study.

Interpretation
The demographic and clinical profiles of patients in our study were similar to those in previous studies done among similar patients in other older heart institutions in the Philippines (Table 8). The previous studies reported that the average ages of Filipino patients with CAD who underwent CABG were within the range of 59-62 years,26 27 28 and that the mean BMI was 24.8 kg/m2.27 Like patients in our study, majority of the patients in previous studies (75-85%) were males26 27 28 and with hypertension (69-72%).26 27 The mean LVEF of patients in our study was comparable to that of patients in another study.27 Our study however recorded higher rates of smoking, diabetes and COPD among patients. There were also more patients who presented with 3-vessel disease and who were admitted for elective surgery in our study. The observed differences in the rates of comorbidities probably reflect the increasing trends of these medical conditions over the years.29 30 This could also mean that more and more patients with comorbidities undergo CABG as a therapeutic procedure. However, this could be a form of selection bias, and further studies are needed to explore these differences.

Table 8    Comparison of demographic and clinical profiles of patients with coronary artery disease (CAD) who underwent coronary artery bypass grafting (CABG) in different heart institutions in the Philippines

Characteristics This study 2015
n=213
Enriquez, et al. 200826
n=225
Bastan, et al. 200727
n=296
Viela, et al. 200528
n=298
Number of years covered by the study 8 1 7 1
Mean age ± SD, years 60.20 ± 9.68 59.76 ± 9.18 59.7 ± 9.69 61.6 ± 9.15
BMI, kg/m2 25.80 ± 5.65 24.8 ± 3.58
Male sex, % 78.40 85.3 79 75.48
Hypertension, % 72.30 71.6 69.6
Diabetics, % 57.75 41.3 38.8
Smokers, % 50.70 33.3 38.2
COPD, % 7.98 4.7
Previous history of myocardial infarction, % 34.60 40.5 40.93
LVEF
   Mean ± SD 50.33 ± 13.54 53.4 ± 21.73
   EF<40%, % 10.4
EF<50%, % 30.71
3-vessel disease, % 81.13 59.5
Elective CABG, % 99.06 68.5 64.76
BMI=body mass index; COPD=chronic obstructive pulmonary disease; EF=ejection fraction; LVEF=left ventricular ejection fraction.

Compared to the PMV rates reported in other studies, which range from 2.4 to 10.4,3 5 8 9 12 14 the rate in our study (18.87%) appears to be the highest. The definition of PMV varied among the previous reports, with time cutoffs ranging from 12 hours to 72 hours postoperatively. 2 3 4 5 6 7 8 9 10 11 48 hours,3 12 13 14 15 16 Studies with >24 hours cutoff for PMV will tend to exclude more patients from being classified as having PMV and report lower PMV rates. On the other hand, a cutoff of 12 hours would further increase the PMV rate in our study because of our practice of extubating patients the day after the CABG procedure. The relatively high rate of PMV in our study could be due to the inclusion of more patients with worse cardiac, renal, and pulmonary preoperative conditions, higher operative mortality risks, worse intraoperative hemodynamic conditions, longer operative durations, and postoperative complications—clinical factors that proved to be associated with PMV.

In this study, PMV is associated with poor preoperative cardiac and renal status, and postoperative complications. Low cardiac output syndrome (LCOS), which is a consequence of myocardial dysfunction, is a common complication of CABG that requires intraoperative placement of IABP and inotropes.31 32 LCOS is associated with poorer outcomes and increased incidence of pulmonary complications, myocardial infarction and renal failure.3 The risk factors for LCOS are similar to the ones that we identified as associated with PMV, including low LVEF,32 33 renal failure,32 and emergency nature of the CABG procedure,33 probably suggesting a common pathophysiology for the two conditions. Higher operative risks scores (EuroSCORE II and STS ACSR) are associated with PMV because these scores are calculated based on preoperative variables that influence PMV, including cardiopulmonary and renal parameters.

The incidence of AKI after CABG may range from 3.6% to as high as 30%, depending on the definition used,34 35 36 and around 1% of the patients undergoing the procedure develop new indications for hemodialysis.37 Following the KDIGO definition of AKI—increase in creatinine by 0.3 mg/dl [26.5 μmol/L] from baseline or 1.5 times the baseline—26 we found that 34.83% of the patients in our study developed AKI post-CABG, and that 20.58% of patients who underwent CABG had new indications for dialysis postoperatively.

AKI can develop from several conditions, including intraoperative hypotension, postoperative cardiac complications that compromise kidney perfusion, hemolysis, atheroemboli, and contrast media exposure.37 Our findings revealed that AKI after surgery was the only independent risk factor for PMV among patients in our study. This is the first time that postoperative AKI is reported to be associated with PMV. AKI will ultimately manifest as low urine output and volume overload in the lungs and heart, exacerbating cardiac dysfunction and pulmonary congestion, which will require extended mechanical ventilation support.

In prolonged CABG procedures both CPB and ACC durations are increased. Unlike other open heart surgeries, those that require CBP has been known to decrease pulmonary function and increase the risk for postoperative pulmonary complications.38 This dysfunction, otherwise known as the pump lung or the post pump syndrome, is due to an acute pulmonary inflammatory response that can arise from ischemia-reperfusion injury, endotoxemia, operative trauma, pre-existing left ventricular dysfunction, non-pulsatile blood flow, or contact of blood components with the artificial surface of the bypass circuit.37 Many studies have looked into CPB time as a risk factor for PMV.2 3 8 9 13 Two studies were able to establish CPB as independent predictor of PMV.2 9 In one study, a CBP duration of more than 91 minutes significantly increased the odds ratio of having PMV of more than 12 hours to 1.39.2 The other reported that a CBP duration of more than 120 mins significantly increased the odds ratio of having PMV of more than 24 hours to 9.6.9 ACC time, on the other hand, refers to the duration that the aorta is clamped during bypass. Aorta clamping increases brain circulation but compromises the blood supply to the lower extremities and, more importantly, to the kidneys. Prolonged ACC because of its strong association with postoperative AKI12 may indirectly result in PMV.

Generalizability
CABG is an important surgical procedure in the treatment of CAD. Postoperative complications related to a patient’s preoperative medical condition and to the complexity of the CABG procedure—especially AKI—can predispose the patient to PMV. PMV, in turn, has been associated with prolonged hospital stay and mortality. Reducing the risk for AKI can reduce the incidence of PMV and subsequently reduce duration of hospital stay and mortality post-CABG. Every attempt should be made in order to avoid events or interventions that predispose patients to postoperative AKI such as exposure to contrast media, intraoperative hypotension, and cardiac complications that result in hypoperfusion of the kidneys.

We hope to be able to develop pre-emptive strategies for reducing the occurrence of PMV based on the findings of this research. To account for differences or changes in patients’ demographic and clinical characteristics, the growing experience of health care services, as well as improvements in surgical and anesthetic techniques over time,3 other institutions can come up with their own prediction models for PMV after CABG and regularly reassess or update their findings.

Conclusion

PMV occurred in 18.87% of the patients with CAD who had CABG in this study. Preoperative renal dysfunction, cardiogenic shock, NYHA FC IV, CCS AGS class IV, LVEF <50%, EuroSCORE II of >5% mortality risk, and STS ACSR score of >5% mortality risk increased the odds ratio of having PMV. The intraoperative insertion of IABP, as well as the occurrence of postoperative complications, namely—AKI, neurological complications, arrhythmia, other pulmonary and non-pulmonary complications, and a new postoperative indication for hemodialysis all significantly increased the odds ratio of having PMV. On multivariable regression analysis, postoperative AKI turned out to be an independent predictor of PMV.

In essence

It is important to identify patients at risk of prolonged mechanical ventilation (PMV) after coronary artery bypass grafting (CABG).


In this study, clinical factors associated with PMV include poor preoperative cardiac, renal and pulmonary status, long operative procedure and development of postoperative complications. Postoperative AKI independently increased the odds ratio of having PMV.


Coming up with local prediction models for PMV and regularly reassessing these models can help improve the outcomes of patients with CAD who undergo CABG.


Acknowledgments

We would like to express our sincerest thanks to Alvin S Concha for his invaluable mentorship in this research, and to the editors of the Southern Philippines Medical Center (SPMC) Journal of Health Care Services for contributing significantly to the editing process of this article. We would also like to thank the consultants of the Department of Internal Medicine in SPMC for their important inputs to this report, and the staff of the Medical Records Office of SPMC Heart Institute for allowing us access to the patient records that we used in this study.


Ethics approval

This study was reviewed and approved by the Department of Health XI Cluster Ethics Review Committee (DOHXI CERC reference P15101201).


Reporting guideline used

STROBE Checklist (http://www.strobe-statement.org/fileadmin/Strobe/uploads/checklists/STROBE_checklist_v4_combined.pdf)


Article source

Submitted


Peer review

External


Funding

Supported by personal funds of the authors


Competing interests

None declared


Access and license

This is an Open Access article licensed under the Creative Commons Attribution-NonCommercial 4.0 International License, which allows others to share and adapt the work, provided that derivative works bear appropriate citation to this original work and are not used for commercial purposes. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc/4.0/


References

1. Cislaghi F, Condemi AM, Corona A. Predictors of prolonged mechanical ventilation in a cohort of 5123 cardiac surgical patients. Eur J Anaesthesiol. 2009 May;26(5):396-403.


2. Cislaghi F, Condemi AM, Corona A. Predictors of prolonged mechanical ventilation in a cohort of 3,269 CABG patients. Minerva Anestesiol. 2007 Dec;73(12):615-21.


3. Knapik P, Ciesla D, Borowik D, Czempik P, Knapik T. Prolonged ventilation post cardiac surgery - tips and pitfalls of the prediction game. Journal of Cardiothoracic Surgery. 2011;6:158.


4. Higgins TL. Pro: early endotracheal extubation is preferable to late extubation in patients following coronary artery surgery. J Cardiothorac Vasc Anesth. 1992 Aug;6(4):488-93.


5. Serrano N, García C, Villegas J, Huidobro S, Henry CC, et al. Prolonged intubation rates after coronary artery bypass surgery and ICU risk stratification score. Chest. 2005 Aug;128(2):595-601.


6. Bansal S, Thai HM, Hsu CH, Sai-Sudhakar CB, Goldman S, et al. Fast track extubation post coronary artery bypass graft: a retrospective review of predictors of clinical outcomes. World Journal of Cardiovascular Surgery. 2013; 3(2):81-86.


7. Shahbazi S, Kazerooni M. Predictive factors for delayed extubation in the intensive care unit after coronary artery bypass grafting; a southern Iranian experience. Iran J Med Sci. 2012 Dec; 37(4):238-241.


8. Légaré JF, Hirsch GM, Buth KJ, MacDougall C, Sullivan JA. Preoperative prediction of prolonged mechanical ventilation following coronary artery bypass grafting. Eur J Cardiothorac Surg. 2001 Nov;20(5):930-6.


9. Natarajan K, Patil S, Lesley N, Ninan B. Predictors of prolonged mechanical ventilation after on-pump coronary artery bypass grafting. Ann Card Anaesth. 2006 Jan;9(1):31-6.


10. Suematsu Y, Sato H, Ohtsuka T, Kotsuka Y, Araki S, et al. Predictive risk factors for delayed extubation in patients undergoing coronary artery bypass grafting. Heart Vessels. 2000;15(5):214-20.


11. Gumus F, Polat A, Yektas A, Totoz T, Bagci M, et al. Prolonged mechanical ventilation after CABG: risk factor analysis. J Cardiothorac Vasc Anesth. 2015 Feb;29(1):52-58.


12. Piotto RF, Ferreira FB, Colósimo FC, Silva GS, Sousa AG, et al. Independent predictors of prolonged mechanical ventilation after coronary artery bypass surgery. Rev Bras Cir Cardiovasc. 2012 Dec;27(4):520-8.


13. Totonchi Z, Baazm F, Chitsazan M, Seifi S, Chitsazan M. Predictors of prolonged mechanical ventilation after open heart surgery. J Cardiovasc Thorac Res. 2014; 6(4):211-216.


14. Branca P, McGaw P, Light R. Factors associated with prolonged mechanical ventilation following coronary artery bypass surgery. Chest. 2001 Feb;119(2):537-46.


15. Saleh HZ, Shaw M, Al-Rawi O, Yates, Pullan DM, et al. Outcomes and predictors of prolonged ventilation in patients undergoing elective coronary surgery. Interact Cardiovasc Thorac Surg. 2012 Jul;15(1):51-56.


16. Cumpeeravut P, Visudharom K, Jotisakulratana V, Pitiguagool V, Banyatpiyaphod S, et al. Off-pump coronary artery bypass surgery: evaluation of extubation time and predictors of failed early extubation. J Med Assoc Thai. 2003 May;86 Suppl 1:S28-35.


17. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009 May 5;150(9):604-612.


18. Riskcalc.sts.org [Internet]. Online STS Adult Cardiac Surgery Risk Calculator. Available from: http://riskcalc.sts.org/. Accessed December 9, 2016.


19. EuroSCORE.org [Internet]. United Kingdom: European System for Cardiac Operative Risk Evaluation. Available from: http://www.euroscore.org/calc.html. Accessed December 9, 2016.


20. Riskcalc.sts.org [Internet]. Online STS Adult Cardiac Surgery Risk Calculator. Available from: http://riskcalc.sts.org/. Accessed December 9, 2016.


21. Poullis M, Pullan M, Chalmers J, Mediratta N. The validity of the original EuroSCORE and EuroSCORE II in patients over the age of seventy. Interact CardioVasc Thorac Surg. 2015;20(2):172-177.


22. Chalmers J, Pullan M, Fabri B, McShane J, Shaw M, et al. Validation of EuroSCORE II in a modern cohort of patients undergoing cardiac surgery. Eur J Cardiothorac Surg. 2013;43(4):688-694.


23. Kennedy JL, LaPar DJ, Kern JA, Kron IL, Bergin JD, et al. Does the Society of Thoracic Surgeons risk score accurately predict operative mortality for patients with pulmonary hypertension? J Thorac Cardiovasc Surg. 2013 Sep;146(3):631-637.


24. Kirmani B, Mazhar K, Fabri B, Pullan DM. Comparison of the EuroSCORE II and Society of Thoracic Surgeons 2008 risk tools. Eur J Cardiothorac Surg. 2013 Dec;44(6):999-1005.


25. Kidney Disease Improving Global Outcomes (KDIGO). Kidney international supplements: KDIGO clinical practice guideline for acute kidney injury. Official journal of the International Society of Nephrology. 2012 Mar;2(1).


26. Enriquez E, Vilela G. Can presence of toxic granulation predict subsequent congestive heart failure among post coronary artery bypass graft patients? Phil Heart Center J. 2008 Jan-Jun; 14(1):8-13.


27. Bastan M, Amil WA, Sy R, Silva G, Chungunco CN, et al. Clinico-demographic profile and in-hospital outcome of coronary arterial bypass graft patients at the Philippine General Hospital: A seven year review. Phil. Heart Assoc 36th Annual Convention Proceedings. 2007.


28. Vilela G, Milo JC. Estimating the risk of in-hospital mortality after CABG surgery at Philippine Heart Center: A comparison of two prediction models. Phil Heart Center J. 2005 Jan-Dec; 11:6-12.


29. Boyle JP, Honeycutt AA, Narayan KM, Hoerger TJ, Geiss LS, et al. Projection of diabetes burden through 2050: impact of changing demography and disease prevalence in the U.S. Diabetes Care. 2001 Nov;24(11):1936-40.


30. World Health Statistics 2008 [Internet]. COPD predicted to be third leading cause of death in 2030. 2008. Available from: http://www.who.int/gard/news_events/World_Health_Statistics_2008/en/. Accessed December 29, 2015.


31. Massé L, Antonacci M. Low cardiac output syndrome: identification and management. Crit Care Nurs Clin North Am. 2005 Dec;17(4):375-383.


32. Maganti M, Badiwala M, Sheikh A, Scully H, Feindel C, et al. Predictors of low cardiac output syndrome after isolated mitral valve surgery. J Thorac Cardiovasc Surg. 2010 Oct;140(4):790-796.


33. Sá MP, Nogueira JR, Ferraz PE, Figueiredo OJ, Cavalcante WC, et al. Risk factors for low cardiac output syndrome after coronary artery bypass grafting surgery. Rev Bras Cir Cardiovasc. 2012 Apr-Jun;27(2):217-23.


34. Corwin HL, Sprague SM, DeLaria GA, Norusis MJ. Acute renal failure associated with cardiac operations. A case-control study. J Thorac Cardiovasc Surg. 1989 Dec;98(6):1107-12.


35. Palevsky PM. Epidemiology of acute renal failure: the tip of the Iceberg. American Society of Nephrology. 2006; 1(1):6-7.


36. Loef BG, Epema AH, Smilde TD, Henning RH, Ebels T, et al. Immediate postoperative renal function deterioration in cardiac surgical patients predicts in-hospital mortality and long-term survival. J Am Soc Nephrol. 2005 Jan;16(1):195-200.


37. Rosner MH, Okusa MD. Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol. 2006 Jan;1(1):19-32.


38. Wynne R, Botti M. Postoperative pulmonary dysfunction in adults after cardiac surgery with cardiopulmonary bypass: clinical significance and implications for practice. Am J Crit Care. 2004 Sep;13(5):384-93.



Copyright © 2016 Acosta JJS, et al.

     

Published
December 29, 2016

Issue
Volume 2 Issue 1 (2016)

Section
Research