International Journal of Neuroscience, 2015; 125(6): 402–408 Copyright © 2014 Informa Healthcare USA, Inc. ISSN: 0020-7454 print / 1543-5245 online DOI: 10.3109/00207454.2014.943370
ORIGINAL ARTICLE
Delayed admission to ICU does not increase the mortality of patients post neurosurgery Jian-Cang Zhou,1 Kong-Han Pan,1 Xin Huang,2 Wen-Qiao Yu,1 and Hong-Chen Zhao1 1
Department of Critical Care Medicine, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China; 2 Department of Neurosurgery, 1st Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China Increasing shortage of intensive care resources is a worldwide problem. While routine postoperative admission to the intensive care unit (ICU) of patients undergoing neurosurgery is a long established practice for many hospitals. Therefore, some neurosurgical patients have to be cared in post anesthesia care unit (PACU) before ICU admission during high ICU occupancy. The aim of this study was to compare the outcome of neurosurgical patients immediately admitted to the ICU post operation with those who were required to wait for ICU bed in PACU and managed by anesthesiologists before ICU admission. All adult neurosurgical patients admitted to our ICU between January 2010 and July 2013 were retrospectively analyzed. Recorded data included demographic data, surgical categories, end time of operation, operation hours, postoperative complication, hospital/ICU length of stay and cost, Glasgow coma score (GCS) on ICU discharge and ICU mortality. A total of 989 neurosurgical patients were evaluated. Nine hundred thirty-seven (94.7%) patients were immediately admitted and 52 (5.3%) patients had delayed ICU admission. Median PACU waiting hours was 4.3 h (interquartile range: 2.0–10.2 h). Delayed ICU admission post neurosurgery was highly associated with the end time of operation (p = 0.019) and high ICU occupancy (p < 0.0001). Average GCS on ICU discharge was higher in immediately admitted group (13.0 ± 3.5 vs. 11.4 ± 4.5, p = 0.012). However, delayed admission to ICU post neurosurgery was not associated with prolonged ICU length of stay, increased ICU mortality, increased postoperative complication and hospital/ICU cost (all p > 0.05). Thus, an algorithm for appropriate disposition of neurosurgical patients is warranted so as to balance the quality of care and control of scarce intensive resources. KEYWORDS: neurosurgical patients, postoperative care, intensive care unit, survival
Introduction Shortage of intensive care resources is a worldwide problem that causing some critically ill patients have to be managed outside ICU by physicians not specifically trained for critical medicine. It has been associated with increased mortality and delays in treatments for emergency patients boarded in emergency department and managed by the emergency department providers [1], and sudden deteriorating hospitalized patients cared by
Received 5 May 2014; revised 6 July 2014; accepted 7 July 2014 Correspondence: Jian-Cang Zhou, Department of Critical Care Medicine, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, No. 3 Qingchun Road East, Hangzhou 310016, Zhejiang, China. Tel: +86 1386 7413 145, Fax: +86 8604 4817. E-mail: [email protected]
402
the ward physicians [2,3]. Moreover, unavailability of intensive care beds is one of the main reasons caused the postponement or cancellation of elective operations [4]. Routine postoperative admission to the intensive care unit (ICU) of patients undergoing neurosurgery is a long established practice. This was believed to allow earlier detection of serious postoperative complications, thereby facilitating prompt intervention and optimizing recovery [5]. Despite increasing controversies with respect to this practice, routine ICU admission after neurosurgery is still common practice in many hospitals. Thus, under high ICU occupancy days, some neurosurgical patients have to wait for ICU beds in post anesthesia care unit (PACU), which are commonly staffed with anesthesiologists rather than specialized intensive care physicians [6]. Nevertheless, PACU is designed for routine recovery of not critical surgical patients, and thus its staffs were not trained in critical care and may not
Delayed ICU admission poses no risk for neurosurgery
as experienced in caring for critical patients. Moreover, the PACU staffs will be distracted by the responsibility to care for other routine postoperative recovery patients. Therefore, it seems that neurosurgical patients who are waiting for ICU beds in PACU tend to be less closely monitored or treated less aggressively given the staff competence and staffing model in PACU. Thus, we hypothesized that the care quality for above-mentioned patients may be impaired and these patients may associated with increased mortality, yet few empirical data are available about this. Hence, the aim of this study was to compare the outcome of neurosurgical patients immediately admitted to the ICU post operation with those who were required to wait for ICU bed in PACU before ICU admission.
Materials and methods Subject The study was a retrospective review of all post neurosurgical patients admitted to our ICU from January 1, 2010 to July 31, 2013, to examine whether waiting in PACU and managed by anesthesiologist during high ICU occupancy for some neurosurgical patients was associated with worse outcomes. Since ours hospital only served for adult patients, all patients who underwent either elective or emergency neurosurgery and admitted to ICU post operation were included in the analysis. The institutional review board of Sir Run Run Shaw hospital approved this retrospective study. Our hospital is a tertiary and teaching hospital in east China. The ICU is a closed unit staffed by full time intensive care physicians while the PACU was staffed by anesthesiologists. Routine admission to ICU post neurosurgery is a common practice in our hospital. Patients were immediately admitted should there were ICU beds available. However, for patients underwent emergency operations at high ICU occupancy or who had booked an ICU bed preoperatively but ICU bed was not available yet on the time the operation terminated, they had to wait in PACU before ICU admission. Under this circumstance, patients were managed by the PACU nurses and anesthesiologists albeit ICU consultation was routine part of the treatments. On some occasions, an attempt to awake and extubation was performed for very few stable patients who may be transferred to ward directly from PACU or operating theatre given the significant shortage of ICU beds. The decision whether to wake each individual at the end of surgery or PACU was mainly based on the clinical assessment made by the anesthesiologists managing the patient, including his own experience, the timing of the surgical termination, and the scheduled disposition. C
2014 Informa Healthcare USA, Inc.
403
Data collection The ICU admission database was queried to identify all the ICU admissions post neurosurgery. The daily ICU occupancy was calculated as bed days used divided by bed days available at midnight once per day. For every included patient, the following data were extracted: (1) demographic characteristics (age, sex); (2) type of operations; (3) The American Society of Anesthesiologists (ASA) level; (4) operation hours and end time of the operation; (5) whether they had boarded in PACU for ICU beds and their waiting hours; (6) Acute Physiology and Chronic Health Evaluation (APACHE) II score on ICU admission; (7) daily ICU occupancy; (8) ICU and hospital length of stay; (9) ventilation hours; (10) ICU/hospital cost; (11) postoperative complications, such as readmission to ICU with the same hospitalization, postoperative bleeding, unplanned return to operating theatre; (12) Glasgow coma score (GCS) on ICU discharge and (13) ICU outcome. The operations were categorized into “neuro-oncological”, “neurovascular”, “head trauma”, and “others”. End time of operation was divided into 12:00–16:00 (most elective operations were expected to be finished in this time range), 16:00–20:00, 20:00–24:00, 00:00–04:00, 04:00–08:00 and 08:00–12:00. For the immediate admission group, PACU waiting time was counted as zero. While for those who had boarded in PACU before ICU admission, the waiting hours were divided into 0–2 h, and more than 2 h, respectively, because unplanned PACU stay of longer than 2 h is an important clinical indicator for anesthesia. Some very critically ill patients preferred to discharge from ICU and die at home imminently was categorized as “died in ICU” in the outcome analysis.
Outcome measures The primary outcome of the study was the ICU length of stay and mortality. Secondary outcomes included ICU/hospital cost, hospital length of stay, ventilation hours, readmission to ICU with the same hospitalization, postoperative bleeding, unplanned return to operating theatre and GCS on ICU discharge.
Statistics Normal distribution of continuous variable was test with the Kolmogorov–Smirnov test. Descriptive data were reported as either mean ± SD, median (interquartile range, IQR) or number and percentage. Patients were grouped into “Immediate ICU admission” group and “Boarded in PACU before admission” group. Also, patients were categorized into “elective”, “emergency” operation and “head trauma” groups in the analysis. With respect to the differences between the groups,
404
J.-C. Zhou et al.
categorical variables were compared using chi-square analysis or Fisher’s exact test. Continuous variables were compared using Independent Sample T test for normally distributed data and Mann–Whitney U test for non-normally distributed data. Multiple comparison analysis for trend test was performed using log-linear analysis for categorical data between different surgical categories, end time of operation and ICU occupancy groups. Statistical analysis was performed, using SPSS 16.0 (Chicago, Ill, USA). Statistical significance was defined as a p value 0.05). When it comes to the postoperative complications, there was no significant difference of ICU readmission, postoperative bleeding and unplanned reoperation rates between the two groups (Table 2, all p > 0.05). Meanwhile, there was no significant difference of ventilation hours, hospital length of stay, ICU and hospital cost between the two groups (Table 2, all p > 0.05). Nevertheless, the mean GCS on ICU discharge was higher in immediately admitted group compared to those had ever waited in PACU before ICU admission (13.0 ± 3.5 vs. 11.4 ± 4.5, p = 0.012). Moreover, more patients had a GCS ≥ 13 on ICU discharge in the immediately ICU admission group (p = 0.028). After stratification of elective or emergency operation and subgroup of head trauma, there were also no significant differences between immediately and delayed ICU admission groups with respect to the ICU mortality and median ICU length of stay (Table 3, all p > 0.05).
Discussion The study demonstrated that neurosurgical patients whose operations ended within “08:00–12:00” or high ICU occupancy were more likely to be boarded in PACU before ICU admission. Although average GCS on ICU discharge was higher in immediately admitted group compared to those had ever waited in PACU before ICU admission (13.0 ± 3.5 vs. 11.4 ± 4.5, p = 0.012), delayed admission to ICU was not associated with prolonged ICU length of stay, increased ICU mortality, increased postoperative complication, and hospital/ICU cost for neurosurgical patients underwent elective or emergency operations (all p > 0.05). Routine postoperative admission to the ICU has long been considered a necessity in the treatment of patients following neurosurgery. However, the costs and benefits of this practice have become increasingly scrutinized in this age, with neurosurgical and other International Journal of Neuroscience
Delayed ICU admission poses no risk for neurosurgery
405
Figure 1. Number of neurosurgical patients boarded in PACU for ICU beds grouped by ICU occupancy (p < 0.0001 for trend).
critically ill patients competing for scarce intensive care resources [7]. Overall incidence of postoperative intracranial hematoma ranges between 0.8% and 2.2% in patients after elective brain surgery [8]. Among that, most developed within the first hours after surgery, for what observation could be performed in a postoperaTable 2.
tive care setting like PACU [9]. Given patients undergoing elective craniotomy have been observed to make up as much as 27% of the general and 46% of the neurosurgical ICU admissions [7], it is debatable whether these patients really benefit from routine postoperatively ICU admission. If not, these intensive care resources
Outcomes for neurosurgical patients categorized by whether waiting in PACU before ICU admission.
Outcomes Mortality (%)a ICU readmissionb Postoperative bleeding Unplanned reoperation GCS on ICU dischargea Mean GCS
ORIGINAL ARTICLE
Delayed admission to ICU does not increase the mortality of patients post neurosurgery Jian-Cang Zhou,1 Kong-Han Pan,1 Xin Huang,2 Wen-Qiao Yu,1 and Hong-Chen Zhao1 1
Department of Critical Care Medicine, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China; 2 Department of Neurosurgery, 1st Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China Increasing shortage of intensive care resources is a worldwide problem. While routine postoperative admission to the intensive care unit (ICU) of patients undergoing neurosurgery is a long established practice for many hospitals. Therefore, some neurosurgical patients have to be cared in post anesthesia care unit (PACU) before ICU admission during high ICU occupancy. The aim of this study was to compare the outcome of neurosurgical patients immediately admitted to the ICU post operation with those who were required to wait for ICU bed in PACU and managed by anesthesiologists before ICU admission. All adult neurosurgical patients admitted to our ICU between January 2010 and July 2013 were retrospectively analyzed. Recorded data included demographic data, surgical categories, end time of operation, operation hours, postoperative complication, hospital/ICU length of stay and cost, Glasgow coma score (GCS) on ICU discharge and ICU mortality. A total of 989 neurosurgical patients were evaluated. Nine hundred thirty-seven (94.7%) patients were immediately admitted and 52 (5.3%) patients had delayed ICU admission. Median PACU waiting hours was 4.3 h (interquartile range: 2.0–10.2 h). Delayed ICU admission post neurosurgery was highly associated with the end time of operation (p = 0.019) and high ICU occupancy (p < 0.0001). Average GCS on ICU discharge was higher in immediately admitted group (13.0 ± 3.5 vs. 11.4 ± 4.5, p = 0.012). However, delayed admission to ICU post neurosurgery was not associated with prolonged ICU length of stay, increased ICU mortality, increased postoperative complication and hospital/ICU cost (all p > 0.05). Thus, an algorithm for appropriate disposition of neurosurgical patients is warranted so as to balance the quality of care and control of scarce intensive resources. KEYWORDS: neurosurgical patients, postoperative care, intensive care unit, survival
Introduction Shortage of intensive care resources is a worldwide problem that causing some critically ill patients have to be managed outside ICU by physicians not specifically trained for critical medicine. It has been associated with increased mortality and delays in treatments for emergency patients boarded in emergency department and managed by the emergency department providers [1], and sudden deteriorating hospitalized patients cared by
Received 5 May 2014; revised 6 July 2014; accepted 7 July 2014 Correspondence: Jian-Cang Zhou, Department of Critical Care Medicine, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, No. 3 Qingchun Road East, Hangzhou 310016, Zhejiang, China. Tel: +86 1386 7413 145, Fax: +86 8604 4817. E-mail: [email protected]
402
the ward physicians [2,3]. Moreover, unavailability of intensive care beds is one of the main reasons caused the postponement or cancellation of elective operations [4]. Routine postoperative admission to the intensive care unit (ICU) of patients undergoing neurosurgery is a long established practice. This was believed to allow earlier detection of serious postoperative complications, thereby facilitating prompt intervention and optimizing recovery [5]. Despite increasing controversies with respect to this practice, routine ICU admission after neurosurgery is still common practice in many hospitals. Thus, under high ICU occupancy days, some neurosurgical patients have to wait for ICU beds in post anesthesia care unit (PACU), which are commonly staffed with anesthesiologists rather than specialized intensive care physicians [6]. Nevertheless, PACU is designed for routine recovery of not critical surgical patients, and thus its staffs were not trained in critical care and may not
Delayed ICU admission poses no risk for neurosurgery
as experienced in caring for critical patients. Moreover, the PACU staffs will be distracted by the responsibility to care for other routine postoperative recovery patients. Therefore, it seems that neurosurgical patients who are waiting for ICU beds in PACU tend to be less closely monitored or treated less aggressively given the staff competence and staffing model in PACU. Thus, we hypothesized that the care quality for above-mentioned patients may be impaired and these patients may associated with increased mortality, yet few empirical data are available about this. Hence, the aim of this study was to compare the outcome of neurosurgical patients immediately admitted to the ICU post operation with those who were required to wait for ICU bed in PACU before ICU admission.
Materials and methods Subject The study was a retrospective review of all post neurosurgical patients admitted to our ICU from January 1, 2010 to July 31, 2013, to examine whether waiting in PACU and managed by anesthesiologist during high ICU occupancy for some neurosurgical patients was associated with worse outcomes. Since ours hospital only served for adult patients, all patients who underwent either elective or emergency neurosurgery and admitted to ICU post operation were included in the analysis. The institutional review board of Sir Run Run Shaw hospital approved this retrospective study. Our hospital is a tertiary and teaching hospital in east China. The ICU is a closed unit staffed by full time intensive care physicians while the PACU was staffed by anesthesiologists. Routine admission to ICU post neurosurgery is a common practice in our hospital. Patients were immediately admitted should there were ICU beds available. However, for patients underwent emergency operations at high ICU occupancy or who had booked an ICU bed preoperatively but ICU bed was not available yet on the time the operation terminated, they had to wait in PACU before ICU admission. Under this circumstance, patients were managed by the PACU nurses and anesthesiologists albeit ICU consultation was routine part of the treatments. On some occasions, an attempt to awake and extubation was performed for very few stable patients who may be transferred to ward directly from PACU or operating theatre given the significant shortage of ICU beds. The decision whether to wake each individual at the end of surgery or PACU was mainly based on the clinical assessment made by the anesthesiologists managing the patient, including his own experience, the timing of the surgical termination, and the scheduled disposition. C
2014 Informa Healthcare USA, Inc.
403
Data collection The ICU admission database was queried to identify all the ICU admissions post neurosurgery. The daily ICU occupancy was calculated as bed days used divided by bed days available at midnight once per day. For every included patient, the following data were extracted: (1) demographic characteristics (age, sex); (2) type of operations; (3) The American Society of Anesthesiologists (ASA) level; (4) operation hours and end time of the operation; (5) whether they had boarded in PACU for ICU beds and their waiting hours; (6) Acute Physiology and Chronic Health Evaluation (APACHE) II score on ICU admission; (7) daily ICU occupancy; (8) ICU and hospital length of stay; (9) ventilation hours; (10) ICU/hospital cost; (11) postoperative complications, such as readmission to ICU with the same hospitalization, postoperative bleeding, unplanned return to operating theatre; (12) Glasgow coma score (GCS) on ICU discharge and (13) ICU outcome. The operations were categorized into “neuro-oncological”, “neurovascular”, “head trauma”, and “others”. End time of operation was divided into 12:00–16:00 (most elective operations were expected to be finished in this time range), 16:00–20:00, 20:00–24:00, 00:00–04:00, 04:00–08:00 and 08:00–12:00. For the immediate admission group, PACU waiting time was counted as zero. While for those who had boarded in PACU before ICU admission, the waiting hours were divided into 0–2 h, and more than 2 h, respectively, because unplanned PACU stay of longer than 2 h is an important clinical indicator for anesthesia. Some very critically ill patients preferred to discharge from ICU and die at home imminently was categorized as “died in ICU” in the outcome analysis.
Outcome measures The primary outcome of the study was the ICU length of stay and mortality. Secondary outcomes included ICU/hospital cost, hospital length of stay, ventilation hours, readmission to ICU with the same hospitalization, postoperative bleeding, unplanned return to operating theatre and GCS on ICU discharge.
Statistics Normal distribution of continuous variable was test with the Kolmogorov–Smirnov test. Descriptive data were reported as either mean ± SD, median (interquartile range, IQR) or number and percentage. Patients were grouped into “Immediate ICU admission” group and “Boarded in PACU before admission” group. Also, patients were categorized into “elective”, “emergency” operation and “head trauma” groups in the analysis. With respect to the differences between the groups,
404
J.-C. Zhou et al.
categorical variables were compared using chi-square analysis or Fisher’s exact test. Continuous variables were compared using Independent Sample T test for normally distributed data and Mann–Whitney U test for non-normally distributed data. Multiple comparison analysis for trend test was performed using log-linear analysis for categorical data between different surgical categories, end time of operation and ICU occupancy groups. Statistical analysis was performed, using SPSS 16.0 (Chicago, Ill, USA). Statistical significance was defined as a p value 0.05). When it comes to the postoperative complications, there was no significant difference of ICU readmission, postoperative bleeding and unplanned reoperation rates between the two groups (Table 2, all p > 0.05). Meanwhile, there was no significant difference of ventilation hours, hospital length of stay, ICU and hospital cost between the two groups (Table 2, all p > 0.05). Nevertheless, the mean GCS on ICU discharge was higher in immediately admitted group compared to those had ever waited in PACU before ICU admission (13.0 ± 3.5 vs. 11.4 ± 4.5, p = 0.012). Moreover, more patients had a GCS ≥ 13 on ICU discharge in the immediately ICU admission group (p = 0.028). After stratification of elective or emergency operation and subgroup of head trauma, there were also no significant differences between immediately and delayed ICU admission groups with respect to the ICU mortality and median ICU length of stay (Table 3, all p > 0.05).
Discussion The study demonstrated that neurosurgical patients whose operations ended within “08:00–12:00” or high ICU occupancy were more likely to be boarded in PACU before ICU admission. Although average GCS on ICU discharge was higher in immediately admitted group compared to those had ever waited in PACU before ICU admission (13.0 ± 3.5 vs. 11.4 ± 4.5, p = 0.012), delayed admission to ICU was not associated with prolonged ICU length of stay, increased ICU mortality, increased postoperative complication, and hospital/ICU cost for neurosurgical patients underwent elective or emergency operations (all p > 0.05). Routine postoperative admission to the ICU has long been considered a necessity in the treatment of patients following neurosurgery. However, the costs and benefits of this practice have become increasingly scrutinized in this age, with neurosurgical and other International Journal of Neuroscience
Delayed ICU admission poses no risk for neurosurgery
405
Figure 1. Number of neurosurgical patients boarded in PACU for ICU beds grouped by ICU occupancy (p < 0.0001 for trend).
critically ill patients competing for scarce intensive care resources [7]. Overall incidence of postoperative intracranial hematoma ranges between 0.8% and 2.2% in patients after elective brain surgery [8]. Among that, most developed within the first hours after surgery, for what observation could be performed in a postoperaTable 2.
tive care setting like PACU [9]. Given patients undergoing elective craniotomy have been observed to make up as much as 27% of the general and 46% of the neurosurgical ICU admissions [7], it is debatable whether these patients really benefit from routine postoperatively ICU admission. If not, these intensive care resources
Outcomes for neurosurgical patients categorized by whether waiting in PACU before ICU admission.
Outcomes Mortality (%)a ICU readmissionb Postoperative bleeding Unplanned reoperation GCS on ICU dischargea Mean GCS
