Predictors of delayed extubation post-coronary artery bypass graft and its relationship with length of stay and in-hospital mortality in Jordan

Ethical Considerations

This study was reviewed and approved by the Ethical and Research Committee at Alexandria University and by the selected hospitals before data collection (IRB00013620 [19/9/2025]). All patients who agreed to participate in the study signed an informed consent form, including permission to review their medical records. All participants received a detailed explanation of the study, its benefits, and risks. They were assured that their participation was voluntary and that they could withdraw at any time without consequences. All data were coded and kept in a password-protected computer, and aggregate data were used for this publication.

Design, Sample, and Setting

A non-experimental, prospective, observational design was used to meet the objectives of this study. A consecutive sampling technique was used. All patients who visited the cardiothoracic surgery outpatient clinics of four hospitals (one governmental, one teaching, and two private) in Alexandria, Egypt, were screened for the following eligibility criteria: (1) aged 18 years or older, (2) able to read and write Arabic, (3) non-emergency surgery, (4) no preoperative mechanical ventilation, (5) no combination with non-cardiac surgeries such as lobectomy. Patients were excluded from the study if they had clinical data loss.

To ensure that the sample size was sufficient to reach statistical significance and meet the objectives of this study, two steps were taken: (1) A logistic regression sample-size calculator was used to determine predictors of DE. The criteria assumed an estimated event rate (occurrence DE) of 20% [3], a two-tailed test with a power of 0.8, 16 independent variables, and an alpha of 0.05. Based on those assumptions, the required sample size was 232 patients. (2) An independent t-test was used to assess the difference in LoS between those who developed DE and those who did not. The assumptions for that test were a two-tailed test, type I error of 0.05, and power of 0.8. Based on those assumptions, the required sample was 128 patients. Therefore, we recruited 300 participants (Figure 1).

Figure 1.

Patient flow diagram.

Procedure

Well-trained research assistants holding master’s degrees in critical care approached all participants at the cardiothoracic surgery clinics who met the inclusion criteria, explained the study in detail, and answered participants’ questions. Before surgery (mean±standard deviation [SD], 20±14 hours), participants filled out a sociodemographic questionnaire that collected their sex, age, marital status, and working status. All other needed information was collected from the medical records after discharge: history of hypertension or diabetes mellitus, previous myocardial infarction, previous angina, smoking, ejection fraction (EF), body mass index (BMI; kg/m²), cardiopulmonary bypass (CPB) and aortic cross-clamp (ACC) time, transfusion of blood products, operative duration and use of intra-aortic balloon pump, pre- and postoperative hemoglobin levels, estimated glomerular filtration rate, postoperative hospital LoS, ICU LoS, and in-hospital mortality.

Study Outcomes

Patients were considered to have DE if extubation occurred more than 48 hours postoperatively or if the patient required reintubation during that period [3]. LoS was collected from medical records after discharge and recorded in days for both the ICU LoS and total hospital LoS. In-hospital mortality was also collected from medical records.

Data Analysis

The IBM SPSS version 26 (IBM Corp.) was used to analyze the data, and any p-value less than 0.05 was considered statistically significant. For aim (a), which assessed the incidence of DE following CABG surgery, numbers and percentages were used. For aim (b), which identified key predictors of DE, continuous variables were converted into categorical ones based on clinically relevant and practical thresholds: age ≥60 years, EF ≤ 50%, operation time ≥7 hours, CPB ≥120 minutes, and BMI ≥28 kg/m² [3]. A two-step analysis was conducted: (1) Point-biserial correlations were used to assess relationships between the independent variables and the dependent variable (DE); (2) Variables showing significant correlations were entered into a logistic regression model. For aim (c), which evaluated whether hospital and ICU LoS differed between the delayed and non-delayed groups, two independent-samples t-tests were performed—one for hospital LoS and one for ICU LoS. For aim (d), which assessed differences in the mortality rate between groups, Fisher’s exact test was used to provide a conservative interpretation because the total number of deaths was 6 (5 in the DE group and 1 in the non-DE group).

Demographics and Clinical Characteristics

In total, 300 patients were included in this study: 200 men and 100 women. The mean age was approximately 60 years, and most of the participants (240, 80%) were married. More than half of the patients were current smokers, had hypertension, or BMI ≥28 kg/m², and nearly three-quarters had previous angina (Table 1).

Table 1.

Demographic and clinical characteristics based on extubation category

Pre- and Postoperative Information

More than one-third of the sample had a CPB time ≥120 minutes. Ninety patients (30%) had an operation duration of ≥7 hours. Cryoprecipitate was the most common blood component transferred after the operation. Sixty percent of the participants were anemic before the operation, compared with 84% postoperatively. Two percent of the participants died during hospitalization (Table 1).

Compared with those who did not develop DE, patients who developed DE were older and had more comorbidities, including a history of previous myocardial infarction and diabetes. They were also more obese, had lower EF, and had longer surgery and ACC durations. They stayed at the hospital and ICU longer than those who did not develop DE. No other significant differences in demographic or clinical characteristics were observed between the groups (Table 1).

Predictors of DE Post-CABG Surgery

The correlations and logistic regression models in Tables 2 and 3 show that 5 predictors for the development of DE were found. Patients aged ≥60 years had a 1.71 times higher risk of developing DE than those aged less than 60 years. Patients with an operation time ≥7 hours were at 1.88 times higher risk for developing DE than those whose operation time was less than 7 hours. Patients with CPB time ≥120 minutes were at 2.25 times higher risk of developing DE than those with a CPB time less than 120 minutes. Patients with a BMI ≥28 kg/m² were at 2.33 times higher risk for developing DE than those with a BMI less than 28 kg/m². The only protective variable was EF: patients with EF ≥50% were 21% less likely to develop DE than those with EF less than 50%.

Table 2.

Significant correlations between DE and demographic and clinical variables

Table 3.

Independent predictors of delayed extubation by Binary logistic regression

Differences in ICU and Hospital LoS by DE Status Post-CABG Surgery

Participants who developed DE stayed in the ICU (mean±SD, 7.9±1.60 vs. 3.1± 1.2; P<0.001) and hospital (mean±SD, 22.6± 3.1 vs. 15.1± 2.9; P<0.001) longer than those who did not develop DE (Table 4).

Table 4.

The difference in the ICU and hospital LoS based on the DE group

Mortality Rate Difference by DE Status Post-CABG Surgery

Six patients (2%) in the total sample died during hospitalization; five of them were in the DE group, with only one patient in the non-delayed group. Fisher’s exact test was performed and indicated a significant difference (P=0.021) between the groups.

Incidence of DE Post-CABG Surgery

Overall, 75 participants (25%) developed DE after CABG surgery. In this study, we considered postoperative extubation to be delayed if mechanical ventilation lasted more than 48 hours or the patient was reintubated within the first 2 days after surgery [3]. The incidence of DE post-CABG reported in recent literature varies widely. For instance, Li et al. [3] found that the frequency of DE after cardiac surgery (including CABG) was 13.6%, whereas Gumus et al. [16] reported that only 5.6% of post-CABG patients required prolonged postoperative ventilation. Both of those studies defined DE as extubation occurring more than 48 hours after surgery. On the other hand, Kotfis et al. [7] reported an incidence of 3.7% for intubation time greater than 24 hours after CABG. Variation in participants’ demographic and clinical characteristics could contribute to the higher incidence of DE observed in this study, compared with previous studies [3,16]. Our results also demonstrate significant correlations between DE and participants’ demographic and clinical variables, which supports that explanation. Moreover, variations in perioperative management, including differences in anesthesia techniques, postoperative care protocols, and extubation criteria, can influence the timing of extubation.

Predictors of DE Post-CABG Surgery

DE post-CABG is influenced by several personal and intra-operative variables: age, BMI, EF, operation time, and CPB time. First, age should be considered when predicting the likelihood of DE in patients undergoing cardiac surgery [3,17]. Advanced age is associated with both frequent comorbidities and alterations in pulmonary function. According to Cho and Stout-Delgado [18], normal lung aging causes anatomical and functional alterations in the respiratory tract that lead to diminished lung function and therefore DE.

Second, high patient weight is associated with DE post-CABG surgery [19]. Specifically in our study, overweight patients (BMI ≥28 kg/m²) had a higher risk of DE than those with a BMI <28 kg/m². This finding supports the risk-prediction model developed by Li et al. [3] that identified 10 risk factors associated with DE after cardiac surgery, including BMI ≥28 kg/m². Along the same line, although some overweight patients can tolerate fast-track extubation, they have a higher incidence of early extubation failure and require a longer time for extubation than non-obese patients [20]. This association can also be explained by the comorbidities and body alterations associated with being overweight.

On the other hand, EF, which measures the heart’s pumping efficiency, acts as a protective factor against DE in patients undergoing CABG surgery. In particular, a better EF has been linked to a lower risk of prolonged intubation, which is consistent with the results reported by Rahimi et al. [21]. Evidently, low cardiac output leads to hemodynamic instability, the use of inotropic and vasoactive medications, and consequently failure to meet the weaning criteria for mechanical ventilation [17,21,22].

In terms of operative variables, our results are consistent with those of previous studies, which reported that prolonged operation time and CPB time are significant risk factors for DE in open heart surgeries, including CABG [23,24] and aortic dissection surgery [1]. The synthetic circuit of the extracorporeal circulation triggers an inflammatory response and causes ischemia–reperfusion injury, which can lead to pulmonary complications such as altered gas exchange, transient hypoxemia, or even acute respiratory distress syndrome [25]. Those pulmonary complications make it more difficult for the respiratory system to function independently, consequently delaying extubation after surgery. Although previous studies reported different cut points for CPB, we found that CPB time ≥120 minutes doubled the risk of developing DE post-CABG surgery.

Differences in ICU and Hospital LoS by DE Status Post-CABG Surgery

DE contributes to a higher incidence of postoperative complications, including the use of continuous renal replacement therapy, neurological complications [3], and delirium [26], leading to prolonged ICU and hospital LoS. Taylor et al. [15] summarized the evidence from four randomized trials that reported ICU LoS based on the extubation time after cardiac surgery. Extubation occurred in the operating room in two of those studies, within six hours of surgery in one, and within eight hours of surgery in one. They concluded that early extubation after cardiac surgery significantly reduces ICU LoS. Similarly, a retrospective analysis of 1,904 patients undergoing CABG revealed that patients extubated ≤12 hours after CABG had lower 30-day mortality and shorter ICU and hospital LoS than those intubated for >12 hours [7].

Differences in the Mortality Rate by DE Status Post-CABG Surgery

Although only 6 patients died during hospitalization in our study population, almost 84% of them were in the DE group. A large-scale study that examined the effects of immediate versus DE in 52,982 patients admitted to ICUs postoperatively in Austria found that day 2 extubation, as opposed to day 1 extubation, was an independent risk factor for in-hospital mortality. In general, postoperative DE affects the mortality rate [1]. A comparison of clinical outcomes specific to CABG surgeries in patients with and without DE showed that DE contributes to a higher incidence of postoperative complications, cardiac arrest, and death [3].

This study has limitations. Organizational and systemic variables were not measured in this study, which might have affected the results by failing to account for the possible effects of those variables on the DE predictors. Those variables should be incorporated in future studies.

The incidence of DE among patients post-CABG surgery was 25%. Diverse risk factors affect the likelihood of DE, among which some are modifiable and others are not. Working on the modifiable risk factors, such as lowering BMI, improving EF, and shortening the operation time, might decrease the incidence of DE and its associated morbidity, mortality, and long LoS.