Improved clinical outcomes following introduction of an attending intensivist for patients admitted to the cardiac surgical intensive care unit after valvular heart surgery: a single-center experience

Introduction

An open intensive care unit (ICU) model may be the most widely adopted model in the cardiac surgical ICU (CSICU) (1). Although cardiac surgeons are the physicians with the most responsibilities in this model, they cannot always perform prompt postoperative management of patients admitted to the CSICU because they spend most of their time in the operating room (OR). Conversely, attending intensivists are responsible for the critical care of patients in the closed ICU model. This model has the advantage that attending intensivists can immediately respond to patients because they almost always remain within the CSICU during their working hours. Although several studies have revealed that the closed ICU model is associated with better clinical outcomes compared with the open ICU model (2-7), few studies have demonstrated the superiority of the closed model in CSICUs (8-10). In 2013, a transition from an open ICU model with coverage by surgeons and trainees to a closed ICU model with attending intensivist coverage occurred at our institution. The intensivist was board-certified for both cardiothoracic surgery and critical care. The intensivist remained in the CSICU during working hours and immediately made autonomous decisions on most medical issues, but communicated with cardiac surgeons about important surgical decisions, such as re-exploration due to post-cardiotomy bleeding. We hypothesized that the presence of an attending intensivist would be associated with a reduced mortality rate and postoperative complications in patients admitted to the CSICU after valvular heart surgery. We also hypothesized that the patients managed by the attending intensivist would have better clinical outcomes if patients with an equivalent level of predicted mortality underwent a similar type of surgery by the same surgeon. This study aimed to investigate the differences in patient outcomes due to the presence of an attending intensivist using propensity score matching and statistical analysis of patient records. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-581/rc).

Methods

Study population and outcomes measured

We retrospectively reviewed the electronic medical records of adult patients admitted to the CSICU following valvular heart surgery between January 2007 and June 2017 (n=808). During this period, several staffing changes occurred in the Department of Cardiac Surgery. However, we included only patients operated on by two cardiac surgeons (n=744), who worked at our institution for the entire study period and had more than 10 years of surgical experience in 2007 (surgeons A and B). The strategy for valvular heart surgery did not change significantly during that period as they were already established cardiac surgeons. The patients were categorized into two groups according to the presence of an attending intensivist. In 2013, our institution transitioned from an open ICU model with trainee coverage to a closed ICU model with attending intensivist coverage. From January 2007 to December 2012, a resident or fellow primarily managed patients admitted to the CSICU following valvular heart surgery with backup from a cardiac surgeon (control group, n=337). From January 2013 to June 2017, one attending intensivist provided postoperative critical care to patients who underwent valvular heart surgery in the CSICU (intensivist group, n=407). The intensivist stayed in the CSICU and managed postoperative patients during the day (08:00–18:00, Monday to Friday, except on holidays). On-duty trainees (resident or fellow) provided coverage for hours excluding weekdays. In order to maintain the continuity of care, guidelines for patient management were decided after discussion with surgeons and shared with on-duty trainees through daily rounds. Even after work, the intensivist backed them up by providing 24 hours home-call consultations. The intensivist worked as a primary decision-maker for all aspects of patient care. The intensivist made his own decisions on most medical issues but communicated with cardiac surgeons about critical surgical decisions, such as re-exploration due to post-cardiotomy bleeding. All parameters included in the European System for Cardiac Operative Risk Evaluation (EuroSCORE) II were used to calculate the predicted mortality rate. We reviewed every type of valvular heart surgery [aortic valve replacement (AVR), aortic valvuloplasty, mitral valve replacement, mitral valvuloplasty, tricuspid valve replacement, or tricuspid valvuloplasty (TVP)], and open heart surgery performed simultaneously with valve surgery [maze procedure, coronary artery bypass grafting (CABG), or aortic surgery]. We investigated whether the presence of an attending intensivist in the CSICU caused a reduction in mortality and the incidence of postoperative complications. The primary outcome was the 30-day mortality rate. We also investigated the rate of postoperative complications by classifying them into the following five categories: (I) mechanical ventilation: duration of mechanical ventilation, rate of prolonged mechanical ventilation for more than 48 hours, rate of early extubation within 6 hours, reintubation within 72 hours after extubation, and ventilator-associated pneumonia (VAP). (II) ICU and hospital stay: length of ICU stay, frequency of prolonged ICU stay for more than 72 hours, length of hospital stay, and rate of prolonged hospital stay for more than 14 days. (III) Post-cardiotomy bleeding: rate of re-sternotomy for bleeding control, cardiac arrest due to cardiac tamponade, and re-exploration in the ICU. (IV) Post-cardiotomy extracorporeal membrane oxygenation (ECMO) support: initiation of veno-arterial (VA) ECMO in the OR or insertion at the ICU bedside, extracorporeal cardiopulmonary resuscitation (ECPR) cases, ECPR time (defined as the time from initiation of cardiac massage to termination of standard CPR due to the initiation of ECMO support), and the number of patients who were successfully weaned from ECMO. (V) Common postoperative complications included delirium, cardiac arrest, deep sternal wound infection, acute kidney injury (AKI, defined as an increase in plasma creatinine greater than 50% over preoperative values), and stroke. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital (IRB No. B-2111-721-101). Informed consent was waived because of the retrospective nature of the study and the analysis using anonymized clinical data.

Statistical analysis

Continuous and categorical variables are expressed as mean ± standard deviation and number (%), respectively. The unmatched groups were compared statistically using the Student’s t-test and χ² test for continuous and categorical variables, respectively. The Wilcoxon rank-sum test was used to assess differences in ECPR time between the two matched groups. The differences in the weaning rate from ECMO and percentage of patients who underwent aortic valve (AV) + mitral valve (MV) + maze operation between the two matched groups were compared using the Fisher’s exact test. A propensity score was calculated for group assignment to overcome differences in baseline characteristics between the two groups. One-to-one propensity score matching was performed. The following underlying characteristics were used as covariates to calculate the propensity score: age, sex, EuroSCORE II, operator, type of valvular heart surgery, cardiopulmonary bypass (CPB) time, and operative time. The P value obtained during the Hosmer-Lemeshow test and the c-statistic of the propensity score model were 0.56 and 0.69, respectively. These values reflect the calibration and discrimination powers of the propensity score model. To evaluate the balance among covariates, the differences in baseline characteristics between the two groups were calculated before and after propensity score matching. It was considered acceptable if covariates with standardized mean difference (SMD) were less than 0.15. The SMDs of all covariates are presented in Table 1. The McNemar’s and Wilcoxon signed-rank tests were performed to compare categorical and continuous variables between the two matched groups. The relationships between the presence of an attending intensivist and clinical outcomes are presented as odds ratios (ORs) and 95% confidence intervals (CIs). SPSS 25 (SPSS Inc., Chicago, IL, USA) software was used for the statistical analysis. Statistical significance was set at P<0.05.

Results

A total of 285 patients were extracted from each group using a propensity score matching analysis. We considered it acceptable because the SMD of all covariates after matching was less than 0.15. Some baseline characteristics of the two groups differed significantly before propensity score matching was performed. The average age of patients in the intensivist group was higher than that in the control group (63.6±14.5 vs. 60.8±13.8 years, P=0.007). More patients who were operated by surgeon A were included in the control group than in the intensivist group (65.0% vs. 50.1%, P<0.001). The proportion of patients who underwent AVR in the intensivist group was higher than that in the control group (51.6% vs. 44.2%, P=0.045). A higher percentage of patients in the control group underwent TVP compared to those in the intensivist group (28.2% vs. 18.7%, P=0.002). In terms of detailed types of surgery before matching, the proportion of patients who underwent aortic valve surgery in combination with CABG in the intensivist group was higher than that in the control group (5.4% vs. 2.1%, P=0.032). The percentage of patients who underwent combined mitral and tricuspid valve surgery in the control group was significantly higher than that in the intensivist group (11.0% vs. 5.2%, P=0.005). All variables were evenly distributed between the two matched groups after propensity score matching (Tables 1,2). Demographic and clinical characteristics, which were not used as covariates for matching, were also examined to confirm whether they were evenly distributed between the two matched groups (Table 3). Although most baseline characteristics were evenly distributed, the intensivist group contained more patients with chronic renal failure (CRF) (7.0% vs. 3.2%, P=0.036) and lower creatinine clearance levels compared with those of the control group (60.9±28.6 vs. 67.7±28.8 mL/min, P=0.005). The primary outcome was the 30-day mortality rate. The intensivist group exhibited a significantly lower 30-day mortality rate compared with that of the control group (2.1% vs. 6.7%, P=0.015), which was lower than the predicted mortality calculated using EuroSCORE II. Conversely, the 30-day mortality rate in the control group was similar to that predicted. We also investigated the rate of postoperative complications by classifying them into five categories as described above (Table 4). In the intensivist group, patients underwent mechanical ventilation for a significantly shorter duration compared to the control group (21.8±69.8 vs. 39.2±115.3 hours, P=0.021). A higher percentage of patients in the intensivist group were extubated within 6 hours compared to the control group (53.7% vs. 42.8%, P=0.015). The incidence of ventilator-associated pneumonia (1.4% vs. 4.9%, P=0.031), cardiac arrest due to cardiac tamponade associated with post-cardiotomy bleeding (0.4% vs. 3.9%, P=0.002), and acute kidney injury (2.8% vs. 7.7%, P=0.011) was significantly lower in the intensivist group compared to the control group. Clinical outcomes with meaningful results were selected and expressed as OR plots (Figure 1).

Figure 1 OR plot of clinical outcomes in trainee coverage and attending intensivist coverage. OR, odds ratio; CI, confidence interval; ICU, intensive care unit.

Discussion

Over the last three decades, the mean age of patients undergoing cardiac surgery and frequency of their comorbid conditions have increased continuously (11). As the preoperative risk of patients undergoing cardiac surgery increases, a growing number of patients develop complications after surgery, leading to longer ICU stays and increased mortality rates (12). Many studies have been conducted to reduce postoperative complications and improve ICU quality (4-6). Studies comparing open and closed ICU models have revealed that the clinical outcomes of closed ICU models are superior to those of open models (4-7,13). The Society of Critical Care Medicine also recommended a closed ICU model because it was associated with a shorter ICU stay and a reduced mortality rate (14). Many ICUs are gradually changing from open to closed systems in accordance with the outcomes of these studies and recommendations. CSICUs are commonly established as open models, in which cardiac surgeons are the physicians with the most responsibilities who cooperate with consultants. Some studies using closed CSICU models have shown that the presence of a full-time intensivist is associated with reduced complications after cardiac surgery (8-10,15). In our previous study, we revealed that critical care provided by the attending intensivist was associated with improvements in CSICU quality and reductions in postoperative complications in patients who underwent elective isolated CABG (16). The results of our previous study revealed that the 30-day mortality rate decreased in the intensivist group, but we could not find a statistically significant difference between the two matched groups (0.3% vs. 1.7%, P=0.219). These findings may be attributed to the significantly lower mortality rate generally observed in patients undergoing elective isolated CABG compared to other forms of cardiac surgeries. In contrast to patients undergoing elective isolated CABG, those undergoing valvular heart surgery are expected to have more comorbidities and relatively higher operative mortality. We expected to find that the intensivist’s critical care was associated with a reduced mortality rate in the group of patients undergoing valvular heart surgery. However, we had to consider confounding factors associated with the heterogeneity of the included patients. Additionally, we need to consider that postoperative outcomes can vary greatly depending on the type of surgery, the difficulty of operation, and the surgeon. To solve this problem, we applied propensity score matching techniques to match patients with an equivalent level of predicted mortality and who underwent similar surgeries by the same surgeon. We used the EuroSCORE II risk-scoring system to compare patients with similar levels of predicted mortality. After propensity score matching was performed, the mean EuroSCORE II of the two matched groups did not differ significantly. Although we have several cardiac surgeons at our institution, we only included patients operated on by two cardiac surgeons (surgeons A and B) who worked at our institution throughout the research period and had more than 10 years of surgical experience in 2007. Although the surgical or anesthetic policy for performing valvular heart surgery did not change significantly during the research period, we matched patients according to every type of valvular heart surgery, including co-operations such as the Maze procedure, CABG, and aortic surgery. We also used CPB and operative times as covariates for matching to reflect the difficulty of the operation. Through these processes, we expected that the differences between the two matched groups, except for the presence of an attending surgical intensivist, would be minimized.

The primary outcome was the 30-day mortality rate. In contrast to the results of our previous study, we found that the 30-day mortality rate in the intensivist group was significantly lower than that of the control group. We assumed that this finding would be related to a decreased incidence of postoperative complications and improved quality of care in CSICU. We investigated the related variables in both groups by dividing them into five categories: mechanical ventilation, ICU and hospital stay, post-cardiotomy bleeding, VA-ECMO support, and postoperative complications. The mean duration of mechanical ventilation was significantly shorter in the intensivist group than in the control group. The frequency of prolonged mechanical ventilation was also reduced in the intensivist group; however, this difference was not statistically significant. These findings were likely due to earlier extubation in the intensivist group (17). In fact, we found that significantly more patients were extubated within 6 hours in the intensivist group than in the control group. Extubation within 6 hours was considered a criterion for a successful fast-track ventilator weaning protocol (18). Since the introduction of attending intensivist, a protocol for fast-track ventilator weaning has been employed. Once the patient meets criteria (elective surgery, awake, hemodynamically stable, minimal bleeding), a spontaneous-breathing trial is initiated. The trial involves the utilization of pressure support ventilation, with a pressure support level of 8 cmH₂O and a positive end-expiratory pressure of 5 cmH₂O. Patients who tolerate the spontaneous-breathing trial for 30 minutes without tachypnea, hypoxia, or hemodynamic instability are subsequently extubated. The number of patients who required reintubation within 72 hours after extubation did not differ significantly between the two matched groups, even though the rate of early extubation was higher in the intensivist group. This may indicate that the decision to extubate in the intensivist group was more appropriate. Because VAP and delirium are associated with prolonged mechanical duration (19,20), we expected a reduced rate of both complications. As expected, the incidence of VAP was significantly lower in the intensivist group compared with that in the control group. Although we could not find statistical significance, the incidence of delirium in the intensivist group was lower than that in the control group. These results may be very meaningful because the rate of reintubation and the incidence of VAP are among the indicators related to ICU quality that are recommended by the European Society of Intensive Care Medicine (21). The two matched groups did not differ significantly in terms of ICU and hospital stays, in contrast to the results of our previous study (16). Prolonged ICU stay, especially that longer than 72 hours, in patients undergoing cardiac surgery is associated with higher hospital mortality and lower long-term survival rates (22). Although we found that the average length of ICU stay was shorter and the proportion of patients with prolonged ICU stay was lower in the intensivist group, the difference was not statistically significant. One possible explanation for these findings is that shortening the ICU stay only by early extubation in patients undergoing valvular heart surgery may be difficult because they had more comorbid conditions and a higher perioperative risk than those undergoing elective isolated CABG. In fact, when we compared the average EuroSCORE II of patients in the intensivist group between this study and our previous study, the average EuroSCORE II of patients who underwent valvular heart surgery was higher than that of patients who underwent elective isolated CABG (6.6 vs. 1.9, after matching). Post-cardiotomy bleeding was one factor that prolonged the length of ICU stay in patients undergoing valvular surgery, besides mechanical ventilation, in relation to their higher perioperative risk. Compared with the results of our previous study, the rate of re-sternotomy for bleeding control in patients undergoing valvular surgery was higher than that in those undergoing elective isolated CABG (5.3% vs. 1.0%, intensivist group, after matching). In the present study, although the rate of re-sternotomy for bleeding control did not differ significantly between the two matched groups, the rate of cardiac arrest due to cardiac tamponade was significantly lower in the intensivist group. The intensivist group also exhibited a lower rate of re-exploration in the ICU due to hemodynamic instability. These findings may be due to the early diagnosis of critical post-cardiotomy bleeding prior to the development of cardiac tamponade. Although the protocol for re-exploration was not different between the two groups (mediastinal bleeding exceeding 200 mL/h for 4 hours or suspected cardiac tamponade), a notable change became evident in the diagnostic approach for cardiac tamponade after the introduction of intensivist. While the diagnosis of cardiac tamponade primarily depends on its characteristic physiology rather than echocardiography, the latter could serve as a valuable tool in assessing the need for re-exploration. Because the attending intensivist in this study was a certified echocardiologist, he performed regular echocardiograms and communicated with cardiac surgeons regarding the test results. Echocardiographic examination was performed more frequently in patients who were likely to develop cardiac tamponade due to postoperative bleeding. If the test results suggested impending cardiac tamponade, the result was promptly reported to the cardiac surgeon, and the decision to transfer the patient to the OR for re-exploration was made immediately. We believe that the reduced number of cases of cardiac arrest or re-exploration at the ICU bedside due to cardiac tamponade may be associated with regular echocardiographic examinations by the attending intensivist. These results may also have been attained because the attending intensivist thoroughly understood cardiac surgery and could communicate clearly with the cardiac surgeons, since the attending intensivist was board-certified in both cardiothoracic surgery and critical care. The advantage of an attending intensivist could also be found in the group of patients who were supported with VA-ECMO postoperatively. Although the number of cases of VA-ECMO initiated in the OR did not differ between the two matched groups, the incidence of VA-ECMO inserted at the ICU bedside, including ECPR cases, was significantly lower in the intensivist group. These findings may also be due to immediate intervention by the attending intensivist in response to changes in the patient’s vital signs, which led to the achievement of hemodynamic stability (23). Because the attending intensivist, as a primary decision maker, always remained in the CSICU during the daytime, the decision on most medical issues was made quickly. As a result, most patients were stabilized without ECMO insertion. Even if ECMO support was required, it was possible to proceed quickly, as evidenced by the significantly shorter ECPR time in the intensivist group. We also found that the incidence of AKI in the intensivist group decreased significantly. In the same context as mentioned above, these results might be because the attending intensivist immediately corrected electrolyte and metabolic abnormalities with appropriate management including fluid replacement, although further research is needed to support this hypothesis. In addition, we believe that the introduction of attending intensivist would be helpful for the management of human resources at our department. While the intensivist worked, the residents and fellows could concentrate on their work in the OR, improving the quality of training. Because the attending intensivist was in charge of not only medical management but also family care in the CSICU, surgeons could be free from concerns about CSICU patients and were able to focus on their original duties, such as surgery and research. Perfusionists were also able to significantly reduce the burden of ECMO-related work, which led to improved work efficiency. However, these benefits will need to be verified with specific data through future studies.

Our study has several limitations. First, we included all patients who underwent valvular surgery regardless of the type of operation. Although we used the propensity score matching technique to overcome the heterogeneity of the included patient population, it would be difficult to minimize the differences between the two matched groups, except for the presence of an attending intensivist. However, in our previous study, despite including more homogeneous patient populations who underwent elective isolated CABG, we failed to find a significant difference in 30-day mortality rate between the two matched groups (16). Although it could be pointed out that the heterogeneity of the population might lead to problems when interpreting our results, we included all patients who were managed postoperatively by the attending intensivist and finally succeeded in determining whether the presence of the attending intensivist positively affected 30-day mortality rate in patients undergoing valvular heart surgery using the same method as that in our previous study. It should also be taken into account that this study involved a much more homogeneous group of patients than previous studies including patients who underwent all types of heart surgery, including CABG, valve surgery and transplantation (9,10). The second limitation is that it is difficult to clearly distinguish the specific impact of an intensivist’s presence from other time-dependent factors, because we divided patients into two groups based on the time period. These factors encompass various aspects, such as the progressive improvement of surgical skills, advancements in the surgical environment, development in medications and equipment as well as improvements in postoperative care standards and the competency of nursing staff. It is widely acknowledged that surgical outcomes tend to exhibit an overall improvement as time progresses, a trend consistently observed across various databases (24). To overcome this problem, we included patients operated on by only two established cardiac surgeons who already had more than 10 years of surgical experience in 2007. Nevertheless, the reduction in 30-day mortality rate could be interpreted not because of the presence of an attending intensivist, but because the surgeons’ surgical skill improved over time. However, when comparing the two groups of patients after matching, we found that the intensivist group had lower preoperative creatinine clearance and a higher incidence of chronic renal failure. It could be interpreted that patients in the intensivist group had a higher preoperative risk despite propensity score matching. Nevertheless, reduced incidence of AKI in the intensivist group could be a basis for stating that the presence of the attending intensivist reduced mortality in addition to the gradual improvement of the cardiac surgeon’s skill. In addition, although it is true that this propensity-matched retrospective before-and-after observational study has statistical weaknesses, it should be considered that the same analysis method was applied in previously published studies (8,9). Finally, only one intensivist was included in this study. This limitation is the same as that mentioned in our previous study. More meaningful results could have been obtained if a homogeneous group of intensivists had been included in this study. However, recruiting ICU staff who are board-certified in both cardiothoracic surgery and critical care is a difficult task due to the lack of intensivists. For the same reason, previous studies have evaluated the impact of a heterogeneous group of intensivists with different specialties (8,10). It would be interesting to investigate whether homogeneous groups of intensivists with the same specialty had a greater impact on the clinical outcomes of patients admitted to the CSICU.

Conclusions

We found that transitioning from an open ICU model with trainee coverage to a closed ICU model with attending intensivist coverage resulted in improved clinical outcomes. Our findings suggest that the presence of an attending intensivist board-certified in both cardiothoracic surgery and critical care would be associated with a reduction of postoperative complications and 30-day mortality rate in patients undergoing valvular heart surgery.

Acknowledgments

Funding: None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-581/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-581/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-581/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-23-581/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital (IRB No. B-2111-721-101). Informed consent was waived because of the retrospective nature of the study and the analysis using anonymized clinical data.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Arora RC, Chatterjee S, Shake JG, et al. Survey of Contemporary Cardiac Surgery Intensive Care Unit Models in the United States. Ann Thorac Surg 2020;109:702-10. [Crossref] [PubMed]
2. Carson SS, Stocking C, Podsadecki T, et al. Effects of organizational change in the medical intensive care unit of a teaching hospital: a comparison of 'open' and 'closed' formats. JAMA 1996;276:322-8.
3. Hackner D, Shufelt CL, Balfe DD, et al. Do faculty intensivists have better outcomes when caring for patients directly in a closed ICU versus consulting in an open ICU? Hosp Pract (1995) 2009;37:40-50. [Crossref] [PubMed]
4. Pronovost PJ, Jenckes MW, Dorman T, et al. Organizational characteristics of intensive care units related to outcomes of abdominal aortic surgery. JAMA 1999;281:1310-7. [Crossref] [PubMed]
5. Dimick JB, Pronovost PJ, Heitmiller RF, et al. Intensive care unit physician staffing is associated with decreased length of stay, hospital cost, and complications after esophageal resection. Crit Care Med 2001;29:753-8. [Crossref] [PubMed]
6. van der Sluis FJ, Slagt C, Liebman B, et al. The impact of open versus closed format ICU admission practices on the outcome of high risk surgical patients: a cohort analysis. BMC Surg 2011;11:18. [Crossref] [PubMed]
7. Treggiari MM, Martin DP, Yanez ND, et al. Effect of intensive care unit organizational model and structure on outcomes in patients with acute lung injury. Am J Respir Crit Care Med 2007;176:685-90. [Crossref] [PubMed]
8. Kumar K, Singal R, Manji RA, et al. The benefits of 24/7 in-house intensivist coverage for prolonged-stay cardiac surgery patients. J Thorac Cardiovasc Surg 2014;148:290-7.e6. [Crossref] [PubMed]
9. Benoit MA, Bagshaw SM, Norris CM, et al. Postoperative Complications and Outcomes Associated With a Transition to 24/7 Intensivist Management of Cardiac Surgery Patients. Crit Care Med 2017;45:993-1000. [Crossref] [PubMed]
10. Huard P, Kalavrouziotis D, Lipes J, et al. Does the full-time presence of an intensivist lead to better outcomes in the cardiac surgical intensive care unit? J Thorac Cardiovasc Surg 2020;159:1363-1375.e7. [Crossref] [PubMed]
11. Çelik M, Durko AP, Bekkers JA, et al. Outcomes of surgical aortic valve replacement over three decades. J Thorac Cardiovasc Surg 2022;164:1742-1751.e8. [Crossref] [PubMed]
12. Rahmanian PB, Kröner A, Langebartels G, et al. Impact of major non-cardiac complications on outcome following cardiac surgery procedures: logistic regression analysis in a very recent patient cohort. Interact Cardiovasc Thorac Surg 2013;17:319-26; discussion 326-7. [Crossref] [PubMed]
13. Multz AS, Chalfin DB, Samson IM, et al. A "closed" medical intensive care unit (MICU) improves resource utilization when compared with an "open" MICU. Am J Respir Crit Care Med 1998;157:1468-73. [Crossref] [PubMed]
14. Haupt MT, Bekes CE, Brilli RJ, et al. Guidelines on critical care services and personnel: Recommendations based on a system of categorization of three levels of care. Crit Care Med 2003;31:2677-83. [Crossref] [PubMed]
15. Kumar K, Zarychanski R, Bell DD, et al. Impact of 24-hour in-house intensivists on a dedicated cardiac surgery intensive care unit. Ann Thorac Surg 2009;88:1153-61. [Crossref] [PubMed]
16. Kim DJ, Sohn B, Kim H, et al. The Impact of an Attending Intensivist on the Clinical Outcomes of Patients Admitted to the Cardiac Surgical Intensive Care Unit after Coronary Artery Bypass Grafting. Korean J Thorac Cardiovasc Surg 2020;53:8-15. [Crossref] [PubMed]
17. Dollin Y, Elliott B, Markert R, et al. Intensivist -Driven Ventilator Management Shortens Duration of Mechanical Ventilation in Coronary Artery Bypass Graft Surgery Patients. J Intensive Care Med 2022;37:1648-53. [Crossref] [PubMed]
18. Probst S, Cech C, Haentschel D, et al. A specialized post anaesthetic care unit improves fast-track management in cardiac surgery: a prospective randomized trial. Crit Care. 2014;18:468. [Crossref] [PubMed]
19. Wu D, Wu C, Zhang S, et al. Risk Factors of Ventilator-Associated Pneumonia in Critically III Patients. Front Pharmacol 2019;10:482. [Crossref] [PubMed]
20. Park SY, Lee HB. Prevention and management of delirium in critically ill adult patients in the intensive care unit: a review based on the 2018 PADIS guidelines. Acute Crit Care 2019;34:117-25. [Crossref] [PubMed]
21. Rhodes A, Moreno RP, Azoulay E, et al. Prospectively defined indicators to improve the safety and quality of care for critically ill patients: a report from the Task Force on Safety and Quality of the European Society of Intensive Care Medicine (ESICM). Intensive Care Med 2012;38:598-605. [Crossref] [PubMed]
22. Mahesh B, Choong CK, Goldsmith K, et al. Prolonged stay in intensive care unit is a powerful predictor of adverse outcomes after cardiac operations. Ann Thorac Surg 2012;94:109-16. [Crossref] [PubMed]
23. Siepe M, Pfeiffer T, Gieringer A, et al. Increased systemic perfusion pressure during cardiopulmonary bypass is associated with less early postoperative cognitive dysfunction and delirium. Eur J Cardiothorac Surg 2011;40:200-7. [Crossref] [PubMed]
24. Bowdish ME, D'Agostino RS, Thourani VH, et al. STS Adult Cardiac Surgery Database: 2021 Update on Outcomes, Quality, and Research. Ann Thorac Surg 2021;111:1770-80. [Crossref] [PubMed]