- Email: [email protected]
Critical care of medical and surgical patients in the ED: Length of stay and initiation of intensive care procedures
Critical Care of Medical and Surgical Patients in the ED: Length of Stay and Initiation of Intensive Care Procedures JAMES SVENSON, MD, BART BESINGER, BA, J. STEPHAN STAPCZYNSKI, MD Little is known about the extent of critical care delivered to patients in the emergency department (ED) and its impact on ED lengths of stay or patient outcomes. The purpose of this study was to characterize the timing of care for critically ill patients, both medical and surgical, in the ED. The design was a retrospective review. The setting was a university teaching hospital. The subjects were ED patients subsequently admitted to a medical or surgical intensive care unit (ICU). The average length of stay in the ED was 367 minutes. Thirty percent of patients were boarded in the ED because of lack of beds in the ICU. Stabilization procedures were performed on 45 (27%) patients, on average 102 minutes after ED admission. Monitoring procedures were performed on 35 (21%), on average 170 minutes after ED admission. There were no significant differences in length of stay, use, and timing of critical procedures in medical and surgical patients. Critically it] patients represent a significant portion of ED patients and may remain in the ED for prolonged periods of time. One of the major contributors to these prolonged stays are lack of beds. Both resuscitative and monitoring procedures are often performed in the ED setting for all types of critical patients. The timing of these procedures indicates that they are performed when necessary for patient care regardless of ED or ICU setting. Thus, ICU care is often initiated and maintained in the ED setting. EDs must be staffed adequately with appropriately trained personnel to care for these patients. (Am J Emerg Med 1997;15:654-657. Copyright © 1997 by W.B. Saunders Company) Emergency departments (EDs) care for a wide spectrum of patient complaints, from primary care to critical illness. Critically ill patients make up a substantial part of some ED patient populations, up to 8% of all patients, and over 25% of those admitted. 1 Critically ill patients receive many therapies in the ED before being admitted to an intensive care unit (ICU). These patients may spend a substantial length of time in the ED before admission.l,2 Little is known about the extent of critical care delivered to these patients, the factors that lead to prolonged ED times, and their impact on patient outcome. Vaxon et al (1994) 2 studied the use of procedures and lengths of stay in EDs for critically ill medical patients in the ED of a large urban hospital. They found that there were substantial delays in admission of patients to ICUs, and many ICU procedures were performed in the ED. Whether performance of these procedures in the ED led to delays in
From the Department of Emergency Medicine, University of Kentucky, Lexington. Received May 29, 1996; accepted January 22, 1997. Address reprint requests to Dr Svenson, Section of Emergency Medicine, University of Wisconsin, C7/379 CSC, 600 Highland Ave, Madison, WI 53792. Key Words:Critical care, length of stay, emergency service use. Copyright © 1997 by W.B. Saunders Company 0735-6757/97/1507-001055.00/0 654
transfer to the ICU was unclear. Similar analyses have not been performed for trauma or nontrauma surgical patients. The purpose of this study was to further characterize the timing of care for critically ill patients, both medical and surgical, in the ED.
METHODS All patients who presented to the ED of the University of Kentucky Hospital during a 6-month period were eligible for inclusion in this study. The University of Kentucky is a 434-bed teaching hospital with approximately 36,000 ED visits per year. The ED has a computerized tracking system that records times of patient arrival, consultation, admission, bed availability after admission has been initiated, and discharge from the department. There are six intensive care units in the hospital: medical, trauma, surgical, neurosurgical, cardiac, and pediatric intensive care units. Patients admitted to the medical, trauma, surgical, and neurosurgical ICUs were identified by a computer search of the ED tracking system. All patients adlnitted from the ED to one of these ICU beds, were eligible for inclusion. After identification of patients, demographic information and information of timing of consultation, admission decision, and discharge from the ED were extracted from the database. The patient's medical chart was retrospectively reviewed, and the use and timing of critical care procedures performed both in the ED and within 24 hours after admission to the ICU were recorded. Critical care procedures were dichotomized into (1) stabilization procedures: intubation, chest tube insertion, insertion of a CVP line for volume resuscitation, emergency cardioversion, and insertion of a pacemaker, and (2) patient monitoring procedures: insertion of an arterial line, insertion of an intracerebral pressure (ICP) monitor, insertion of a central venous pressure (CVP) or Swan-Ganz line for monitoring purposes. Differences in lengths of stay and timing until procedures were compared using the Wilcoxon rank sum test. Fisher's Exact test was used to compare differences in the performance of critical care procedures between groups. Statistical significance was defined as a P value of .05 or less.
RESULTS During the period of study, there were 201 patients admitted to one of the ICUs. Detailed information was available on 169 (84%) of these patients. Information was not available in 32 patients because of inability to locate the chart or lack of documentation in the ED record. Forty-nine (29%) were medical patients, 56 (33%) were nontrauma surgical patients, and 64 (38%) were trauma-related surgical patients. The admission diagnoses for the groups are shown in Table 1. Overall, the average length of stay in the emergency department was 367 +- 278 minutes (35 to 1737 minutes). The average length of stay for medical, nontrauma
SVENSON ET AL • CRITICAL CARE IN THE ED
TABLE 1.
655
Diagnosis on Admission to the Intensive Care Unit Diagnosis
Medical (n = 49) Chest pain Pulmonary edema Seizure Arrhythmia Toxic ingestion/overdose Gastrointestinal bleeding Cerebrovascular accident Pneumonia Sepsis Diabetic ketoacidosie COPD exacerbation Alcoholic ketoacidosis Hypernatremia Jaundice Vena cava thrombosis Meningitis Nontrauma Surgical (n = 56) Intracerebral bleed Gastrointestinal bleeding Subarachnoid hematoma Subdural hematoma Brain tumor Perforated viscous Dissecting abdominal aortic Aneurysm Bowel infarction Surgical complications Pericarditis Small bowel obstruction Brain abscess Liver failure Trauma Surgical (n = 64) Multiple trauma Burn Closed head injury Gunshot wound Epidural hematoma Subdural hematoma Hanging Shock
No. 12 6 5 5 3 3 2 2 2 2 2 1 1 1 1 1 17 8 7 5 4 3 3 2 2 1 1 1 1 t 40 8 5 4 2 2 2 1
surgical, and trauma-related surgical patients was 362 _+ 300, 343 + 306, and 393 + 232 minutes, respectively. Only the difference in length of stay between traumatized and nontraumatized surgical patients was significant (P = .02). Fifty-two patients (30%) were boarded in the ED because of bed unavailability in the ICU. The average length of boarding time was 321 + 289 minutes. Even after a bed was available in the ICU, it took an average of 110 + 132 minutes for the patient to leave the ED. The length of stay for boarders in the ED averaged 600 + 412 (range, 187 to 1825) minutes. Consultation with an admitting service was obtained 66 + 115 minutes after patient amval. Stabilization procedures were performed on 45 (27%) ED patients. Of these, 9 patients received more than one stabilization procedure. Fifteen additional patients received a stabilization procedure in the ICU. None of these patients had received a stabilization procedure in the ED. The mean time from admission until a stabilization procedure was performed was 102 + 142 minutes in the El) and 348 _+ 301
minutes in the ICU. Thirty-six (21%) patients were already intubated on arrival in the ED and were ventilated throughout their stay. There were no significant differences between the number of patients on whom a stabilization procedure was performed and the type of service to which they were admitted. There was no significant difference in the time until a stabilization procedure was performed and the type of service. The average time until the performance of a stabilization procedure by type of procedure is given in Table 2. Cardioversion was performed on only 2 patients, but was performed within 7 minutes of arrival on each. Four patients had a stabilization procedure (CVP for venous access) performed after a bed was already available in the ICU. Three patients had a stabilization procedure performed (2, intubation, and 1, chest tube insertion) after the lack of bed availability was known. These three procedures were performed on average 210 minutes after becoming ED boarders. Monitoring procedures were performed on 35 (21%) ED patients. Of these, 9 received more than one monitoring procedure. Thirty-six additional patients received a monitoring procedure in the ICU. Of these, 7 had received a monitoring procedure in the ED. The mean time from admission until a monitoring procedure was 170 _+ 131 minutes in the Ell) and 668 + 484 minutes in the ICU. Surgical nontrauma patients were more likely to receive a monitoring procedure (ICP monitoring) than medical or surgical trauma patients (P = .01). There was no difference in the performance of monitoring procedures between surgical trauma and medical patients. There was no significant difference in the time until a monitoring procedure was performed and the type of service (medical, surgical nontrauma, surgical trauma). The average time until the performance of a monitoring procedure by type of procedure is given in Table 3. Four patients had a monitoring procedure (arterial catheterization) performed in the ED after a bed was available in the ICU. Three patients who had to be boarded in the ED had a monitoring procedure performed after it was known that they were to be boarded in the ED. For those patients intubated on arrival, the mean El) length of stay was 364 -+ 319 minutes, for those not inmbated on arrival, the ED length of stay was 378 _+ 283 minutes (P = .97). There was no significant difference in length of stay of those on whom a stabilization procedure was performed and those for whom no procedure was performed. Lengths of stay in the El) tended to be longer in TABLE 2. ICU
Stabilization Procedures Performed in the ED and the
ED
ICU
Procedure
n
Mean Time (rain)
n
Mean Time (rain)
Intubation Central venous catheterization for resuscitation Chest tube placement Cardioversion Pacemaker placement
17
81
6
410
26 11 2 0
94 104 6 NA
3 0 1 0
235 NA 941 NA
ABBREVrATION: NA, not applicable.
656
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 15, Number 7 • November 1997
TABLE 3. Monitoring Procedures Performed in the ED and the ICU ED
Procedure
n
ICU
Mean Time (min)
n
Mean Time (min)
178
27
235
180 132
16 2
202 108
Arterial catheterization 25 Central venous catheterization for monitoring 11 Intracerebral pressure monitoring 10
those on whom a monitoring procedure was performed (451 + 372 minutes versus 367 +-- 291 minutes), but the difference did not reach statistical significance (P = .09). Twenty-eight patients (17%) did not survive, 6 (12%) medical patients, 13 (23%) nontrauma surgical patients, and 9 (14%) trauma-related surgical patients (P = .26). There was no difference in the ED length of stay in survivors (374 _+ 279 minutes) versus nonsurvivors (332 -+ 272 minutes) (P = .25). Nonsurvivors were much more likely to have had a procedure (either monitoring or stabilization) performed in the ED (65% versus 31%, P = .001), or were more likely to have already been intubated on arrival and ventilated throughout their ED stay (50% versus 15%, P < .001). The distribution of procedures among survivors and nonsurvivors is given in Table 4. DISCUSSION
Our study has addressed some issues regarding the treatment of critically ill patients the ED. The total time spent in our ED was large, on average over 5 hours. A substantial number of patients were officially boarded in the ED because of lack of availability of beds in the ICU. Thirty percent of our patients were boarded for an average of 10 hours in the ED. Lack of ICU beds is a major contributor to the prolonged length of stay in the ED for critically ill patients .3 In this era of contracting budgets for medical care, we must still plan for adequate spaces for these patients. Lack of beds was not the only contributor to prolonged length of stays. Even after beds were available, it took an average of almost 2 hours to transfer the patient to the ICU. What factors led to these delays are unclear from these data.
TABLE4. Differences in Critical Care Procedures Performed in the ED Among Survivors and Nonsurvivors
Type of Procedure Intubated on arrival (ventilated in ED) Intubated in ED Chest tube insertion Cardioversion Central venous catheter for volume replacement Arterial catheterization ICP Central venous catheter for monitoring
Nonsurvivors Survivors P (n = 28) n (%) (n = 141) n (%) Value 14 7 3 1
(50) (25) (10) (4)
22 (16) 10 (7) 8 (5) 1 (1)
<.001 .01 .39 .30
8 (28) 9 (32) 7 (25)
18 (13) 16 (11 ) 3 (2)
.04 <.001 <.001
5 (18)
6 (4)
.01
These transfers were not delayed because of ED procedures being performed. We have examined the use of critical care procedures in the ED. LeTourneau et al (1980) 6 previously examined critical procedures in a stabilization room of an urban referral hospital, and reported crude totals for a number of procedures. Varon et al (1994) 2 did not distinguish between stabilization or resuscitative procedures, which should be routinely performed in the ED, and those used for patient monitoring, more appropriate to the ICU setting. These investigators suggested that prolonged ED care may have a negative impact on patient care because many monitoring procedures were performed shortly after ICU admission in their population. Our data do not support this. Given the timing of the procedures, we suggest that patients had monitoring and stabilization procedures performed when necessary for their care. Very few patients had procedures performed in the ED because of prolonged stays attributable to bed unavailability. The average time to performance of a stabilization or monitoring procedure after transfer to the ICU was prolonged, indicating that procedures were performed because of changes in patient status, rather than not being performed in the ED. The average time from ED admission until performance of a stabilization procedure was substantial (102 minutes). The factors that caused the delays in the performance of these stabilization or resuscitation procedures in the initial phase of patient evaluation are not clear from these data. Some patients may be given a trial of maximal noninterventional therapy before more aggressive intervention, and this may lead to the delays seen. Stress and burnout are a substantial problem to the specialty of emergency medicine. 4'5 Treatment and monitoring of critically ill patients in the ED may add to this problem. However, our data suggest that at our university hospital setting there is substantial help for the ED physician in caring for these patients. Consultation with an appropriate admitting service was obtained quickly after ED arrival and stabilization. These services contribute to the ongoing care of the patient and assume administrative responsibility after admission decisions are made. These decisions were made substantially sooner than the patient actually left the department, so that intensive care of these patients was coordinated with the admitting team. This study was performed in an urban teaching hospital, where resident support is plentiful. The length of time of critical care and stress to the ED physician in private hospitals may be different. This study also addresses the issue of differences in care between critically ill surgical and medical patients. In this population, there was virtually no difference in ED length of stays, number or stabilization and monitoring procedures performed, and the time to initiation of those procedures between surgical and medical patients. In addition, even for traumatized patients, there was no difference in the time until performance of a critical procedure. Thirty-one percent of the patients in our study were intubated (either before arrival or in the ED) and mechanically ventilated in the ED. This is less than the 52% in the study of Varon et al, but still represents a sizable portion of this critically ill population. This again argues for some
SVENSON ET AL • CRITICAL CARE IN THE ED
extended training in the application and monitoring of mechanical devices, at least in centers with a high number of critically ill patients.
CONCLUSION Critically ill patients represent a significant portion of ED patients and may remain in the ED for prolonged periods of time. Both resuscitative and monitoring procedures are often performed in the ED setting for all types of critical patients. The timing of these procedures indicates that they are performed when necessary for patient care regardless of ED or ICU setting. Thus, ICU care is often initiated and maintained in the ED setting. EDs must be staffed adequately with appropriately trained personnel to care for these patients.
657
REFERENCES 1. Fromm RE, Gibbs LR, McCallum WG, et al: Critical care in the emergency department: A time-based study. Crit Care Med 1993;21 : 970-976 2. Varon J, Fromm RE, Levine RL: Emergency department procedures and length of stay for critically ill medical patients. Ann Emerg Med 1994;23:546-549 3. Andrulis DP, Kellerman A, Hintz EA, et al: Emergency departments and crowding in United States teaching hospitals. Ann Emerg Med 1991;20:980-986 4. Gallery ME, Whitley TW, KIonis LK, et al: A study of occupational stress and depression among emergency physicians. Ann Emerg Med 1992;21:58-64 5. Doan-Wiggins L, Zun L, Cooper MA, et al: Practice satisfaction, occupational stress, and attrition of emergency physicians. Acad Emerg Med 1995;2:556-563 6. LeTourneau B, Blegen C, Clinton J: Critical care in an emergency department. Ann Emerg Med 1980;9:126-130
From the Department of Emergency Medicine, University of Kentucky, Lexington. Received May 29, 1996; accepted January 22, 1997. Address reprint requests to Dr Svenson, Section of Emergency Medicine, University of Wisconsin, C7/379 CSC, 600 Highland Ave, Madison, WI 53792. Key Words:Critical care, length of stay, emergency service use. Copyright © 1997 by W.B. Saunders Company 0735-6757/97/1507-001055.00/0 654
transfer to the ICU was unclear. Similar analyses have not been performed for trauma or nontrauma surgical patients. The purpose of this study was to further characterize the timing of care for critically ill patients, both medical and surgical, in the ED.
METHODS All patients who presented to the ED of the University of Kentucky Hospital during a 6-month period were eligible for inclusion in this study. The University of Kentucky is a 434-bed teaching hospital with approximately 36,000 ED visits per year. The ED has a computerized tracking system that records times of patient arrival, consultation, admission, bed availability after admission has been initiated, and discharge from the department. There are six intensive care units in the hospital: medical, trauma, surgical, neurosurgical, cardiac, and pediatric intensive care units. Patients admitted to the medical, trauma, surgical, and neurosurgical ICUs were identified by a computer search of the ED tracking system. All patients adlnitted from the ED to one of these ICU beds, were eligible for inclusion. After identification of patients, demographic information and information of timing of consultation, admission decision, and discharge from the ED were extracted from the database. The patient's medical chart was retrospectively reviewed, and the use and timing of critical care procedures performed both in the ED and within 24 hours after admission to the ICU were recorded. Critical care procedures were dichotomized into (1) stabilization procedures: intubation, chest tube insertion, insertion of a CVP line for volume resuscitation, emergency cardioversion, and insertion of a pacemaker, and (2) patient monitoring procedures: insertion of an arterial line, insertion of an intracerebral pressure (ICP) monitor, insertion of a central venous pressure (CVP) or Swan-Ganz line for monitoring purposes. Differences in lengths of stay and timing until procedures were compared using the Wilcoxon rank sum test. Fisher's Exact test was used to compare differences in the performance of critical care procedures between groups. Statistical significance was defined as a P value of .05 or less.
RESULTS During the period of study, there were 201 patients admitted to one of the ICUs. Detailed information was available on 169 (84%) of these patients. Information was not available in 32 patients because of inability to locate the chart or lack of documentation in the ED record. Forty-nine (29%) were medical patients, 56 (33%) were nontrauma surgical patients, and 64 (38%) were trauma-related surgical patients. The admission diagnoses for the groups are shown in Table 1. Overall, the average length of stay in the emergency department was 367 +- 278 minutes (35 to 1737 minutes). The average length of stay for medical, nontrauma
SVENSON ET AL • CRITICAL CARE IN THE ED
TABLE 1.
655
Diagnosis on Admission to the Intensive Care Unit Diagnosis
Medical (n = 49) Chest pain Pulmonary edema Seizure Arrhythmia Toxic ingestion/overdose Gastrointestinal bleeding Cerebrovascular accident Pneumonia Sepsis Diabetic ketoacidosie COPD exacerbation Alcoholic ketoacidosis Hypernatremia Jaundice Vena cava thrombosis Meningitis Nontrauma Surgical (n = 56) Intracerebral bleed Gastrointestinal bleeding Subarachnoid hematoma Subdural hematoma Brain tumor Perforated viscous Dissecting abdominal aortic Aneurysm Bowel infarction Surgical complications Pericarditis Small bowel obstruction Brain abscess Liver failure Trauma Surgical (n = 64) Multiple trauma Burn Closed head injury Gunshot wound Epidural hematoma Subdural hematoma Hanging Shock
No. 12 6 5 5 3 3 2 2 2 2 2 1 1 1 1 1 17 8 7 5 4 3 3 2 2 1 1 1 1 t 40 8 5 4 2 2 2 1
surgical, and trauma-related surgical patients was 362 _+ 300, 343 + 306, and 393 + 232 minutes, respectively. Only the difference in length of stay between traumatized and nontraumatized surgical patients was significant (P = .02). Fifty-two patients (30%) were boarded in the ED because of bed unavailability in the ICU. The average length of boarding time was 321 + 289 minutes. Even after a bed was available in the ICU, it took an average of 110 + 132 minutes for the patient to leave the ED. The length of stay for boarders in the ED averaged 600 + 412 (range, 187 to 1825) minutes. Consultation with an admitting service was obtained 66 + 115 minutes after patient amval. Stabilization procedures were performed on 45 (27%) ED patients. Of these, 9 patients received more than one stabilization procedure. Fifteen additional patients received a stabilization procedure in the ICU. None of these patients had received a stabilization procedure in the ED. The mean time from admission until a stabilization procedure was performed was 102 + 142 minutes in the El) and 348 _+ 301
minutes in the ICU. Thirty-six (21%) patients were already intubated on arrival in the ED and were ventilated throughout their stay. There were no significant differences between the number of patients on whom a stabilization procedure was performed and the type of service to which they were admitted. There was no significant difference in the time until a stabilization procedure was performed and the type of service. The average time until the performance of a stabilization procedure by type of procedure is given in Table 2. Cardioversion was performed on only 2 patients, but was performed within 7 minutes of arrival on each. Four patients had a stabilization procedure (CVP for venous access) performed after a bed was already available in the ICU. Three patients had a stabilization procedure performed (2, intubation, and 1, chest tube insertion) after the lack of bed availability was known. These three procedures were performed on average 210 minutes after becoming ED boarders. Monitoring procedures were performed on 35 (21%) ED patients. Of these, 9 received more than one monitoring procedure. Thirty-six additional patients received a monitoring procedure in the ICU. Of these, 7 had received a monitoring procedure in the ED. The mean time from admission until a monitoring procedure was 170 _+ 131 minutes in the Ell) and 668 + 484 minutes in the ICU. Surgical nontrauma patients were more likely to receive a monitoring procedure (ICP monitoring) than medical or surgical trauma patients (P = .01). There was no difference in the performance of monitoring procedures between surgical trauma and medical patients. There was no significant difference in the time until a monitoring procedure was performed and the type of service (medical, surgical nontrauma, surgical trauma). The average time until the performance of a monitoring procedure by type of procedure is given in Table 3. Four patients had a monitoring procedure (arterial catheterization) performed in the ED after a bed was available in the ICU. Three patients who had to be boarded in the ED had a monitoring procedure performed after it was known that they were to be boarded in the ED. For those patients intubated on arrival, the mean El) length of stay was 364 -+ 319 minutes, for those not inmbated on arrival, the ED length of stay was 378 _+ 283 minutes (P = .97). There was no significant difference in length of stay of those on whom a stabilization procedure was performed and those for whom no procedure was performed. Lengths of stay in the El) tended to be longer in TABLE 2. ICU
Stabilization Procedures Performed in the ED and the
ED
ICU
Procedure
n
Mean Time (rain)
n
Mean Time (rain)
Intubation Central venous catheterization for resuscitation Chest tube placement Cardioversion Pacemaker placement
17
81
6
410
26 11 2 0
94 104 6 NA
3 0 1 0
235 NA 941 NA
ABBREVrATION: NA, not applicable.
656
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 15, Number 7 • November 1997
TABLE 3. Monitoring Procedures Performed in the ED and the ICU ED
Procedure
n
ICU
Mean Time (min)
n
Mean Time (min)
178
27
235
180 132
16 2
202 108
Arterial catheterization 25 Central venous catheterization for monitoring 11 Intracerebral pressure monitoring 10
those on whom a monitoring procedure was performed (451 + 372 minutes versus 367 +-- 291 minutes), but the difference did not reach statistical significance (P = .09). Twenty-eight patients (17%) did not survive, 6 (12%) medical patients, 13 (23%) nontrauma surgical patients, and 9 (14%) trauma-related surgical patients (P = .26). There was no difference in the ED length of stay in survivors (374 _+ 279 minutes) versus nonsurvivors (332 -+ 272 minutes) (P = .25). Nonsurvivors were much more likely to have had a procedure (either monitoring or stabilization) performed in the ED (65% versus 31%, P = .001), or were more likely to have already been intubated on arrival and ventilated throughout their ED stay (50% versus 15%, P < .001). The distribution of procedures among survivors and nonsurvivors is given in Table 4. DISCUSSION
Our study has addressed some issues regarding the treatment of critically ill patients the ED. The total time spent in our ED was large, on average over 5 hours. A substantial number of patients were officially boarded in the ED because of lack of availability of beds in the ICU. Thirty percent of our patients were boarded for an average of 10 hours in the ED. Lack of ICU beds is a major contributor to the prolonged length of stay in the ED for critically ill patients .3 In this era of contracting budgets for medical care, we must still plan for adequate spaces for these patients. Lack of beds was not the only contributor to prolonged length of stays. Even after beds were available, it took an average of almost 2 hours to transfer the patient to the ICU. What factors led to these delays are unclear from these data.
TABLE4. Differences in Critical Care Procedures Performed in the ED Among Survivors and Nonsurvivors
Type of Procedure Intubated on arrival (ventilated in ED) Intubated in ED Chest tube insertion Cardioversion Central venous catheter for volume replacement Arterial catheterization ICP Central venous catheter for monitoring
Nonsurvivors Survivors P (n = 28) n (%) (n = 141) n (%) Value 14 7 3 1
(50) (25) (10) (4)
22 (16) 10 (7) 8 (5) 1 (1)
<.001 .01 .39 .30
8 (28) 9 (32) 7 (25)
18 (13) 16 (11 ) 3 (2)
.04 <.001 <.001
5 (18)
6 (4)
.01
These transfers were not delayed because of ED procedures being performed. We have examined the use of critical care procedures in the ED. LeTourneau et al (1980) 6 previously examined critical procedures in a stabilization room of an urban referral hospital, and reported crude totals for a number of procedures. Varon et al (1994) 2 did not distinguish between stabilization or resuscitative procedures, which should be routinely performed in the ED, and those used for patient monitoring, more appropriate to the ICU setting. These investigators suggested that prolonged ED care may have a negative impact on patient care because many monitoring procedures were performed shortly after ICU admission in their population. Our data do not support this. Given the timing of the procedures, we suggest that patients had monitoring and stabilization procedures performed when necessary for their care. Very few patients had procedures performed in the ED because of prolonged stays attributable to bed unavailability. The average time to performance of a stabilization or monitoring procedure after transfer to the ICU was prolonged, indicating that procedures were performed because of changes in patient status, rather than not being performed in the ED. The average time from ED admission until performance of a stabilization procedure was substantial (102 minutes). The factors that caused the delays in the performance of these stabilization or resuscitation procedures in the initial phase of patient evaluation are not clear from these data. Some patients may be given a trial of maximal noninterventional therapy before more aggressive intervention, and this may lead to the delays seen. Stress and burnout are a substantial problem to the specialty of emergency medicine. 4'5 Treatment and monitoring of critically ill patients in the ED may add to this problem. However, our data suggest that at our university hospital setting there is substantial help for the ED physician in caring for these patients. Consultation with an appropriate admitting service was obtained quickly after ED arrival and stabilization. These services contribute to the ongoing care of the patient and assume administrative responsibility after admission decisions are made. These decisions were made substantially sooner than the patient actually left the department, so that intensive care of these patients was coordinated with the admitting team. This study was performed in an urban teaching hospital, where resident support is plentiful. The length of time of critical care and stress to the ED physician in private hospitals may be different. This study also addresses the issue of differences in care between critically ill surgical and medical patients. In this population, there was virtually no difference in ED length of stays, number or stabilization and monitoring procedures performed, and the time to initiation of those procedures between surgical and medical patients. In addition, even for traumatized patients, there was no difference in the time until performance of a critical procedure. Thirty-one percent of the patients in our study were intubated (either before arrival or in the ED) and mechanically ventilated in the ED. This is less than the 52% in the study of Varon et al, but still represents a sizable portion of this critically ill population. This again argues for some
SVENSON ET AL • CRITICAL CARE IN THE ED
extended training in the application and monitoring of mechanical devices, at least in centers with a high number of critically ill patients.
CONCLUSION Critically ill patients represent a significant portion of ED patients and may remain in the ED for prolonged periods of time. Both resuscitative and monitoring procedures are often performed in the ED setting for all types of critical patients. The timing of these procedures indicates that they are performed when necessary for patient care regardless of ED or ICU setting. Thus, ICU care is often initiated and maintained in the ED setting. EDs must be staffed adequately with appropriately trained personnel to care for these patients.
657
REFERENCES 1. Fromm RE, Gibbs LR, McCallum WG, et al: Critical care in the emergency department: A time-based study. Crit Care Med 1993;21 : 970-976 2. Varon J, Fromm RE, Levine RL: Emergency department procedures and length of stay for critically ill medical patients. Ann Emerg Med 1994;23:546-549 3. Andrulis DP, Kellerman A, Hintz EA, et al: Emergency departments and crowding in United States teaching hospitals. Ann Emerg Med 1991;20:980-986 4. Gallery ME, Whitley TW, KIonis LK, et al: A study of occupational stress and depression among emergency physicians. Ann Emerg Med 1992;21:58-64 5. Doan-Wiggins L, Zun L, Cooper MA, et al: Practice satisfaction, occupational stress, and attrition of emergency physicians. Acad Emerg Med 1995;2:556-563 6. LeTourneau B, Blegen C, Clinton J: Critical care in an emergency department. Ann Emerg Med 1980;9:126-130