Unplanned ICU transfer from post-anesthesia care unit following cerebral surgery: A retrospective study

Background Unplanned transfer to intensive care unit (ICU) lead to reduced trust of patients and their families in medical staff and challenge medical staff to allocate scarce ICU resources. This study aimed to explore the incidence and risk factors of unplanned transfer to ICU during emergence from general anesthesia after cerebral surgery, and to provide guidelines for preventing unplanned transfer from post-anesthesia care unit (PACU) to ICU following cerebral surgery. Methods This was a retrospective case-control study and included patients with unplanned transfer from PACU to ICU following cerebral surgery between January 2016 and December 2020. The control group comprised patients matched (2:1) for age ({+/-}5 years), sex, and operation date ({+/-}48 hours) as those in the case group. Stata14.0 was used for statistical analysis, and p <0.05 indicated statistical significance. Results A total of 11,807 patients following cerebral surgery operations were cared in PACU during the study period. Of the 11,807 operations, 81 unscheduled ICU transfer occurred (0.686%). Finally, 76 patients were included in the case group, and 152 in the control group. The following factors were identified as independent risk factors for unplanned ICU admission after neurosurgery: low mean blood oxygen (OR=1.57, 95%CI: 1.20-2.04), low mean albumin (OR=1.14, 95%CI: 1.03-1.25), slow mean heart rate (OR=1.04, 95%CI: 1.00-1.08), blood transfusion (OR=2.78, 95%CI: 1.02-7.58), emergency surgery (OR=3.08, 95%CI: 1.07-8.87), lung disease (OR=2.64, 95%CI: 1.06-6.60), and high mean blood glucose (OR=1.71, 95%CI: 1.21-2.41). Conclusion We identified independent risk factors for unplanned transfer from PACU to ICU after cerebral surgery based on electronic medical records. Early identification of patients who may undergo unplanned ICU transfer after cerebral surgery is important to provide guidance for accurately implementing a patient's level of care.


Author Disclosure Statements
All authors indicate that the study does not have any conflicts of interest. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted March 14, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted March 14, 2022. patients and their families in medical staff and challenge medical staff to allocate scarce 48 ICU resources. This study aimed to explore the incidence and risk factors of unplanned 49 transfer to ICU during emergence from general anesthesia after cerebral surgery, and to 50 provide guidelines for preventing unplanned transfer from post-anesthesia care unit 51 (PACU) to ICU following cerebral surgery.

52
Methods: This was a retrospective case-control study and included patients with 53 unplanned transfer from PACU to ICU following cerebral surgery between January 2016 54 and December 2020. The control group comprised patients matched (2:1) for age (±5 55 years), sex, and operation date (±48 hours) as those in the case group. Stata14.0 was used 56 for statistical analysis, and p <0.05 indicated statistical significance. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 14, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022

73
The post-anesthesia care unit (PACU) is where surgical patients under general anesthesia 74 receive monitoring and care immediately after surgery, and is an intermediate transitional 75 unit to provide care for patients [1,2] . Postoperative patients were admitted to the PACU 76 and closely cared for by the anesthetic nurse and anesthesiologist until they recovered 77 from the anesthesia effects, and then safely transferred to the ward by the anesthetic 78 nurse [ 3 ] . 79 Unplanned transfer of patients from the PACU to the ICU after surgery is one of the 80 common critical events in the PACU [ 4 , 5 ] . It is well-known that there is a shortage of ICU 81 beds worldwide [ 6 -8 ] . Patients without plans to transfer to the ICU before surgery, and their 82 transport time is often delayed compared with those who have reserved ICU beds [ 9 ] . 83 Limited ICU resources and delays in transfers are associated with increased 84 hospitalization costs and morbidity, which can negatively affect their treatment and 85 prognosis [ 1 0 ] . 86 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 14, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 The risk of complications in cerebral surgery is high, and patients are prone to 87 unpredictable changes after surgery [ 1 1 , 1 2 ] . In previous clinical practice, patients were 88 usually admitted to the ICU for 12-24 hours of observation and treatment after cerebral 89 surgery; most of them were transferred to the general ward the next day [13][14][15] . With the 90 recent advancement in medicine and surgery, many studies have shown that patients do 91 not need to be routinely transferred to the ICU after cerebral surgery; patients should be 92 strictly evaluated and then judged whether they need to be transferred to the ICU for 93 treatment and care, even for emergency cerebral surgery [16][17][18][19][20][21] . Identifying postoperative 94 patients requiring ICU admission is a challenging but necessary daily task. In the clinical 95 setting, some patients undergoing cerebral surgery require unplanned ICU transfer despite 96 thorough evaluation [22,23] . Therefore, it is vital to improve the positive rate of 97 identification, especially for those patients assessed as transferred to the ward, but 98 transferred to the ICU after surgery. 99 The identification of risk factors is critical to improving preoperative assessment. 100 Therefore, this study aimed to examine preoperative, surgical, and PACU data for 101 potential risk factors for unplanned ICU transfer, which has implications for improving 102 the management and postoperative resource allocation of those vulnerable patients.

105
This was an observational, retrospective, case-control study in which each case was 106 matched with two controls. Unplanned transfer to the ICU was defined as no plan to 107 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted March 14, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 6 transfer the patient to the ICU before starting anesthesia (i.e., the decision to transfer to 108 ICU was made during or after surgery). CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted March 14, 2022. ; https://doi.org/10. 1101/2022 For each patient with an unplanned transfer to ICU after cerebral surgery, two controls of 128 the same type of surgery were randomly selected from the electronic medical records. 129 Patients in the control group were matched (2:1) for sex, age (±5 years), and operation 130 date (±48 hours). Researchers reviewed the data from patients in the control group to 131 ensure the integrity of the included patients' medical records. 132 Data collection 133 All data extraction and data input were completed by members of the research team who 134 have been trained in scientific research. 135 The following demographic and disease history data were collected: age, sex, smoking 136 history, drinking history, and the presence of lung disease, hypertension, diabetes, 137 hyperlipidemia, and/or cardiovascular disease. Patients who had quit smoking and 138 drinking were still categorized into the smoking and drinking groups, respectively. Data 139 on laboratory indicators, such as platelet volume, hemoglobin, total protein, albumin, uric 140 acid, and blood glucose levels, were also collected. Laboratory indicators were measured 141 on an empty stomach the day before the operation. 142 Surgical and anesthetic data, including the type of surgery, operation duration, blood loss 143 volume, fluid replacement volume, urine volume, blood transfusion volume, heart rate, 144 percentage of oxygen saturation (SpO 2 ), removal of tracheal intubation, reintubation, and 145 PACU duration, were collected. Volume of blood loss, of fluid replacement, and of urine 146 were obtained over the entire operation period; heart rate and blood oxygen were 147 obtained immediately after leaving the operating room.
148 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) in the control group (Fig 1).

166
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) We compared the preoperative, intraoperative, and PACU characteristics of cases 168 receiving routine postoperative care, and those with unplanned ICU transfer. There were 169 a number of significant differences between the two patient populations (  initial 20 variables, 7 remained in the final multivariate model (Fig 3) (Table 3).  ICU unplanned after cerebral surgery were those who were more likely to have 202 pulmonary disease, lower mean albumin, higher mean blood glucose, more likely to 203 undergo emergency surgery, more likely to receive intraoperative blood transfusion, 204 higher mean heart rate, and lower mean oxygen saturation. To best of our knowledge, this 205 is the first study to investigate the risk factors for unplanned ICU transfer from the PACU 206 in patients undergoing cerebral surgery. This present studyserves a reference for our next 207 step in developing an electronic medical record system-based, large data-based 208 intraoperative prediction model for unplanned ICU transfer of surgical patients.

209
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted March 14, 2022. . 243 Exploring the risk factors of unplanned transfer to ICU is of great significance to improve 244 the awareness of both medical staff and patients and help them take preventive measures 245 for potential risk factors. In the future, we hope to preoperatively stratify patients based 246 on these factors to predict which patients will need a higher level of postoperative care. 247 By preventing or predicting the need for unplanned transfer to ICU after surgery, we can 248 better allocate ICU resources, reduce the morbidity and mortality of vulnerable patients, 249 and ultimately improve the overall safety of cerebral procedures.

250
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted March 14, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022