Timing of Admission to the Surgical Intensive Care Unit is Associated with in-Hospital Mortality


 Background: Knowledge about the relationship between timing of admission to the intensive care unit (ICU) and mortality among surgical patients admitted for acute care is limited.Objective: We aimed to investigate whether admission to the surgical intensive care unit (SICU) during after-hours (all the times when intensivists were not staffed in the ICU) was associated with in-hospital mortality.Methods: This retrospective cohort study was conducted at a tertiary academic hospital, in which we analyzed data of 571 patients admitted to the SICU whose complete medical records were available. The work-hours were defined as 07:00-19:00 from Monday to Friday during which intensivists were staffed in the ICU. The after-hours were defined as all other times during which intensivists were not staffed in the SICU. The primary outcome measurement was in-hospital mortality according to the time of SICU admission.Results: In all, 333 and 238 patients were admitted to the SICU during work-hours and after-hours, respectively. Unplanned admissions (33.3% vs. 47.1%, p<0.001), Acute Physiology and Chronic Health Evaluation II score ≥25 (11.1% vs. 23.9%, p<0.001), ventilator support (17.4% vs. 34.0%, p<0.001), and use of inotropics (33.3% vs. 50.0%, p<0.001) were significantly higher in the after-hours group than in the work-hours group. In the multivariate analyses of the association between the time of SICU admission and in-hospital mortality, the timing of SICU admission was an independent factor for in-hospital mortality (OR=2.526; 95% CI=1.010-6.320, p=0.048). Conclusions: In this study, we found that admission to the SICU during after-hours was associated with increased in-hospital mortality.


Introduction
Adequate treatment in the rst few hours after the initial insult is an important factor in uencing treatment outcomes in critically ill patients. [1,2] Patients can develop a critical condition at any time; hence, critical care needs to be available at all times. Previous studies reported that higher sta ng of intensivists in the intensive care unit (ICU) leads to a signi cant reduction in ICU and in-hospital mortality, and length of stay. [3][4][5][6][7][8] These improved outcomes were observed not only in the medical ICU, but also in the surgical intensive care unit (SICU). [9] However, due to the limited resources and high cost, it might not be always possible to staff the ICUs with intensivists at night and on weekends.
Most previous studies focused on patient outcomes related to intensivist sta ng during after-hours in the medical ICUs. [10][11][12][13][14] The outcomes of admission during after-hours in the SICU hours remains unclear.
In this study, we aimed to evaluate whether SICU admission during after-hours (all the times when intensivists were not staffed in the ICU) was associated with higher in-hospital mortality compared with admission during work-hours.

Methods And Materials
Study setting and patient enrollment This retrospective study was conducted between March 2018 and February 2019 in Severance Hospital, Yonsei University Health System. Severance Hospital Yonsei University Health System is a high-volume, teaching, and referral hospital in Seoul, Korea. It has a 143-bedded ICU spread across 11 areas. This study was performed on surgical patients in one SICU and one mixed ICU in 11 areas. Severance Hospital follows a high-intensity sta ng and a semi-closed ICU model. One intensivist with 3 years' experience and one second-year clinical fellow preparing for intensivist certi cation are staffed at the SICU from 7 am until 7 pm from Monday to Friday. A senior resident is in charge during after-hours. Real-time consultation during after-hours, by the work-hours intensivist is available according to the patient status.
Patients who were admitted to the SICU for any reasons were eligible for this study. Adult patients with trauma or post-transplant, and pediatric patients under 18 years of age were excluded. Of the 586 patients admitted to the SICU during the study period, 571 patients with complete medical records were nally enrolled.

Data collection
All demographic data including age, sex, Acute Physiologic and Chronic Health Evaluation (APACHE) II score, admission type, cause of ICU admission, number of ICU days, ventilator support required, and use of inotropic drugs were reviewed and analyzed. The primary outcome measurement was in-hospital mortality according to the time of SICU admission, while the secondary outcomes were ICU readmission within 48 hours, duration of ventilator support in days, length of ICU stay, length of hospital stay, and inhospital mortality.

De nitions
The work-hours were de ned to be from 7 am until 7 pm from Monday to Friday, during which intensivists were staffed in the SICU. The after-hours were de ned as all other times during which intensivists were not staffed in the SICU. Planned admission was de ned as SICU admission for postoperative monitoring or care after elective surgery, while all other admissions were de ned as unplanned admissions.

Statistical analysis
Between-group differences were evaluated using the chi-square test or Fisher's exact test. Continuous variables were compared using the Student's t-test. The Cox proportional hazard model was used for multivariate analysis. All statistical tests were two-sided, and a p-value < 0.05 was considered statistically signi cant. All statistical tests were performed using SPSS version 25.0 (IBM, SPSS Inc., Chicago, IL).

Ethics statement
The study was approved by the Institutional Review Board of Severance Hospital, Yonsei, University Health System (approval number 4-2019-0412). The requirement for informed consent was not necessary because of the retrospective nature of the study.

Results
In all, 333 (58.3%) and 238 (41.7%) patients were admitted to the SICU during work-hours and after-hours, respectively. The patients in the work-hours group were older than those in the after-hours group (mean age: 68.55 ± 12.52 vs. 64.06 ± 13.84, p < 0.001). The number of patients with an APACHE II score of ≥ 25 was two times higher in the after-hours group than in the work-hours group (23.9% vs. 11.1%, p < 0.001). Unplanned admissions were also one-and-a half times higher during after-hours than during work-hours (47.1% vs. 33.3%, p < 0.001). Compared to the work-hours group, the proportion of patients who required ventilator support was two times higher in the after-hours group (17.4% vs. 34.0%, p < 0.001). Further, the proportion of patients who were administered inotropics was also almost two times higher in the afterhours group (50.0% vs. 33.3%, p < 0.001; Table 1).  Table 2). Multivariate analyses showed that the time of SICU admission was an independent risk factor for in-hospital mortality (OR = 2.526; 95% CI = 1.010-6.320, p = 0.048). Unplanned admission (OR = 21.607; 95% CI = 2.789-167.407, p = 0.003) and ventilator support (OR = 52.269; 95% CI = 6.920-394.808, p < 0.001) were also independent predictors for in-hospital mortality with clinical signi cance (Table 3).

Discussion
Previous research showed that intensivist sta ng in the ICU leads to a signi cant reduction in ICU and inhospital mortality, and length of stay. [3][4][5][6][7][8][9] However, the Society of Critical Care Medicine guidelines do not recommend 24/7 intensivist sta ng in the ICU in high-intensity ICU models [5] based on only one randomized cohort study performed in the medical ICU. [15] To date, there is lack of evidence about whether admission in the SICU during after-hours is safe.
In this study, we investigated whether SICU admission during after-hours is associated with in-hospital mortality compared with admission during work-hours among patients who underwent surgical treatment. We found that admission to the SICU during after-hours increased the in-hospital mortality (OR = 2.526; 95% CI = 1.010-6.320, p = 0.048; Table 3). Moreover, unplanned admissions, APACHE II score ≥ 25, need for ventilator support, and administration of inotropic drugs were also markedly higher in the after-hours group than in the work-hours group. These ndings are important for the allocation and distribution of ICU resources, including intensivists.
Two systematic reviews reported that while patients admitted to an ICU over the weekend are at a higher risk of death, admission to the ICU during night hours did not appear to be associated with increased mortality. [16,17] The authors inferred that lower level of sta ng and intensity of care provided by many hospitals over the weekend might account for this nding. [5,16,17] In our study, "after-hours" was de ned as the absence of the intensivists, and the results showed increased mortality in patients admitted to the SICU during after-hours. Although there was no change in the level of nurse sta ng and intensity of care, and real-time consultation was provided by work-hour intensivists, absence of the intensivist might affect resolute decisions required during critical illness.
One recent study found that the out-of-o ce hours' ICU admission after elective surgery was associated with signi cantly higher rates of postoperative complications, increased hospital length-of-stay, and increased inter-hospital transfers. [18] In our study, the subgroup analysis of patients admitted for planned surgery showed a higher median length of hospitalization in the after-hours group than in the work-hours group. This result might suggest that the cases with higher risk are referred to the SICU outside regular work-hours. Therefore, surgeons and anesthesiologists should plan surgeries of patients with higher risk towards the beginning of the surgical schedule for the day.
Despite the retrospective design and short period of the study, these ndings might re ect the current status of most ICUs in Korea, considering that 82.3% of all ICUs are either in university teaching hospitals or in university-a liated hospitals. [19] One of the main strengths of this study is that we speci cally targeted surgical patients in conducting the research. Recently, O et al [20] reported a retrospective observational study of 10,708 postoperative patients showing an increased risk of 30-day mortality in the group without intensivist coverage. Compared to the previous studies performed in mixed or medical ICUs, [12,14,[21][22][23][24][25][26][27][28] we only focused on postoperative SICU admissions. In our study, we included patients with unplanned admission including those who underwent emergency operations, those who developed postoperative sepsis, bleeding or other medical complications and we also included patients who had an elective postoperative status. The enrolled patients in this study represented the general population of patients who required intensive care postoperatively, and the results were consistent with those of previous studies. [18,20,29] Additionally, we performed subgroup analyses according to the causes of SICU admission to adjust for negative effects of unplanned admission on mortality.

Conclusion
Admission to the SICU during after-hours was associated with increased in-hospital mortality. Our ndings have the potential to provide a guide to the hospitals for better patient safety and effective allocation of ICU resources.
Abbreviations ICU: Intensive care unit; SICU: Surgical intensive care unit; APACHE: Acute Physiologic and Chronic Health Evaluation