- Email: [email protected]
Duration of patient immobilization in the ED
Duration of Patient Immobilization in the ED E. BROOKE LERNER, MS, EMT-P AND RONALD MOSCATI, MD In this article we seek to determine the duration of immobilization in patients presenting to the emergency department (ED). We conducted a 10-week prospective study of a convenience sample of patients transported to a level one trauma center immobilized with a backboard and cervical collar. Total backboard time (TBT) was measured from the time the ambulance left the scene to the time the patient was removed from the backboard, while total ED backboard time (TEDBT) was measured from the time of arrival at the ED to the time of backboard removal. There were 138 patients entered in the study. Insufficient data excluded 36 patients from further analysis. TBT was available for 92 patients and averaged 63.63 (__.45.87)minutes. Dividing patients into those who were removed from the backboard prior to radiographs (n = 85), the TBT average was 53.9 minutes (_+30.1), whereas the average for those who had radiographs prior to removal from the backboard (n = 7) was 181.3 minutes (_+41.6). There were 102 patients for whom TEDBT was available and averaged 46.36 (__.44.88)minutes. Dividing patients into those who were removed from the backboard prior to radiographs (n = 95), the TEDBT average was 37.6 minutes (-+29.6), whereas the average for those who had radiographs prior to removal from the backboard (n = 7) was 165.3 minutes (+_49.7).Patients are left on backboards for significant periods of time even when no radiographsare taken prior to backboardremoval. (Am J Emerg Med 2000;18:28-30. Copyright © 2000 by W.B. Saunders Company) Immobilizing patients with backboards and cervical collars is a common procedure utilized by out-of-hospital emergency care providers for all patients felt to have a mechanism of injury consistent with a potential vertebral i n j u r y Y These patients are fully immobilized prior to transport to the emergency department (ED), regardless of presenting signs and symptoms. 2-7 Recent studies have suggested that this is a potential source of iatrogenic patient pain. 8-12Several of these studies have shown that immobilizing noninjured, healthy patients for 30 to 80 minutes on a backboard causes most subjects to have pain and discomfort. 9-12 Furthermore, a study by Cordel113 shows that patients spend an average of 77 minutes on backboards in the ED before being cleared by any means, with the vast majority found to have no spinal injury. This study, to our knowledge, has only been published in abstract form and the full methods are not available. Further, Cordell does not provide the methods used to remove the patients from the From the State University of New York at Buffalo, School of Medicine and Biomedical Sciences, Department of Emergency Medicine, Buffalo, NY. Manuscript received October 16, 1998, returned November 5, 1998; revision received December 17, 1998, accepted March 18, 1999. Presented at the ACEP Research Forum, San Francisco, CA, October 18, 1997. Address reprint requests to E. Brooke Lerner, Department of Emergency Medicine, Erie, County Medical Center, 462 Grider St, Buffalo, NY 14215. E-mail: Lerner@ acsu.buffalo.edu Key Words: Immobilization, emergency medicine, backboarding, emergency medical services. Copyright © 2000 by W.B. Saunders Company 0735-6757/00/1801-0006510.00/0
28
backboard. Whereas those with spinal injuries clearly need to be protected, large numbers of patients are being subjected to iatrogenic pain when left on backboards for extended periods. These studies all advocate prompt removal o f patients from backboards by utilizing clinical criteria rather than waiting for a radiographic study. It is thought that removing patients from backboards after clinical evaluation may reduce the time patients spend on backboards in addition t6 avoiding unnecessary radiation exposure and increased health care cost. It is, however, unclear how long patients brought to the ED on backboards remain on backboards while in the ED. Our study was designed to quantify the amount of time patients spend on backboards in a clinical ED setting.
MATERIALS AND METHODS A prospective, observational study was conducted over a 10week period on a convenience sample of patients transported by ambulance to the regional adult level one trauma center. The trauma center is a public, urban, tertiary care facility with 400 inpatient beds and approximately 40,000 ED visits annually. ED patients are seen primarily by emergency medicine residents under the supervision of board-certified emergency medicine attendings. Typically, the attending physician or the third-year resident are the only staff members who are allowed to clear patients from backboards. There is no standard protocol for clinical clearance prior to radiographs; however, there is a standard practice among our attendings. No attempts were made to hide the study from our staff; however, it was not common knowledge that the study was taking place. To be included in the study, patients had to have been brought from the scene of injury by ambulance and to have been immobilized on a backboard in the field. In our system, any patient with a mechanism of injury that could possibly result in an injury to the vertebral column is fully immobilized. Patients were not enrolled if they had major multiple trauma or if they were transferred to our facility from another receiving facility. Data collectors were in the ED from 9 a.m. until 11 p.m. Monday through Saturday, and 2 p.m. until 7 p.m. on Sunday. They were instructed to enroll all patients meeting the inclusion criteria who presented during their data collection shift. Time data on patients who were still on backboards at the conclusion of their data collection shift may not have been completed. The data collectors reviewed patients' prehospital care record for the time the ambulance called en route to the hospital, which was used as an approximation of the time the patient was placed on the backboard. They also reviewed the nursing note to determine the time the triage nurse recorded that the patient had arrived at the ED. The time the patient was removed from the backboard, as well as whether the backboard was removed prior to taking radiographs, were directly observed and recorded by the data collectors. The recorded times were then used to calculate patients' total backboard time (TBT) by subtracting the time the ambulance left
LERNER AND MOSCATI I i ED PATIENT IMMOBILIZATION DURATION
the scene of injury from the time the patient was removed from the backboard in the ED. Total ED backboard time (TEDBT) was calculated by subtracting the time the patient arrived on location at the ED from the time the patient was removed from the backboard. Means with standard deviations and ranges were calculated for TBT and TEDBT for all patients as well as the subgroups removed from the backboard prior to radiographs or removed from backboard after radiographs. This study had institutional review board approval.
RESULTS One-hundred and thirty-eight patients were entered into the study. The average age of patients was 37.70 (+19.12) years and the sample was 69% male. Thirty-six patients were excluded from further analysis because insufficient data were collected. Twenty-nine were excluded because the time they were removed from the backboard was not recorded. Two were excluded because it was not recorded whether they were removed from the backboard before or after obtaining radiographs. An additional 5 were excluded because neither the time en route to the hospital nor the time of arrival at the ED were recorded. Of the remaining 102 subjects, 92 included all 3 time data points, whereas 10 included only the time of ED arrival and time of being cleared from the backboard. As a result, these patients were only included in the TEDBT analysis. The TBT was available for 92 patients, and the overall average TBT was 63.63 (_+45.87) minutes. Eighty-five patients were removed from the backboard prior to being radiographed in an average of 53.9 (+30.1, range 10-201) minutes, and 7 were removed from the backboard after being radiographed in an average of 181.3 ( + 41.6, range 133-239) minutes. Comparing the TBT for those patients who were cleared clinically to those who were cleared by radiograph showed a statistically significant difference between the two groups (P < .000001; 95% CI -151.58--- × -< -103.12). There were 102 patients for whom TEDBT was available, and the overall average TEDBT was 46.36 (+44.88) minutes. Ninety-five patients were removed from the backboard before being radiographed in an average of 37.6 (+29.6, range 4-196) minutes, and 7 were removed from the backboard after being radiographed in an average of 165.3 (-+49.7, range 98-231) minutes. Comparing the TEDBT for those patients who were cleared clinically to those who were cleared by radiograph showed a statistical significant difference between the 2 groups (P < .00000t; 95% CI -151.90--< × ~ -103.48). Comparison of the en route time with triage time showed that the overall mean transport time was 15.8 minutes (_+ 11.4). Comparing those patients who were cleared clinically (15.6 minutes -+ 11.3) to those who were cleared by radiograph (14.6 minutes + 8.1) showed there was no significant difference in their prehospital times (P < .72; CI 95% - 4 . 2 3 -< × -< 6.06). Of the 102 patients included in the study, 2 were determined to have spine fractures. One patient was found to have a T4 fracture, and one patient was found to have an L2 fracture. Of those 2 patients, 1 was radiographed prior to removal from the backboard. Both of these patients were
29
stable and did not receive surgery. There were no patients with cervical spine injuries in the study.
DISCUSSION Our results indicate that patients spend long periods of time immobilized on backboards. Removing patients from backboards before radiographs resulted in a decrease in the average amount of time patients spend immobilized on a backboard. However, the group of patients removed from the backboard before radiographs still spent an average of 53.9 minutes immobilized, which meets or exceeds the amount of time reported by most of the previous studies of iatrogenic pain in healthy volunteers. 9-12 The 16 minutes of transport time is below the amount of time shown by previous studies to cause pain, which suggests that immediate removal from the backboard at the "ED door" may reduce the incidence of iatrogenic pain. However, none of the iatrogenic pain studies has looked specifically at the effects of laying on a backboard for less then 30 minutes. Our study did not collect data on whether any of these patients had pain before or after immobilization. However, applying the conclusions of the iatrogenic pain studies, it is possible that many of our patients would have developed pain in the ED as a result of being immobilized that they did not have immediately following their injury. Iatrogenic pain could have resulted in increased time and cost because it might necessitate obtaining radiographs in patients who may not have had spine pain from the injury itself. Other studies have shown an increased incidence and severity of iatrogenic pain with increasing immobilization time. 8,12 This may affect patients who are perceived to be less seriously injured at triage and axe less likely to undergo as immediate an evaluation as those believed to have more severe injuries, particularly when the ED is busy. The immobilized patient without complaints would be immobilized longer waiting to be evaluated. Based upon our observed times and the time required to produce iatrogenic pain in the volunteer studies, this group of initially pain-free patients would be likely to develop pain from immobilization. These patients may in turn fail to meet the clinical criteria for spinal clearance, then necessitating radiographic clearance and still longer immobilization. These patients, in particular, suffer from the universal spinal immobilization concept. There are 3 approaches that could be taken to reduce iatrogenic pain in this group. The first would be to convince emergency physicians that these patients must be evaluated promptly upon arrival to the ED, because delay is painful. However, one must consider the potential disruption to other patient care by implementing such a policy. Secondly, more comfortable immobilization boards or techniques could be developed. Support in the natural contours of the spine that would not compromise immobilization might greatly reduce the incidence of iatrogenic pain. Several studies have shown that the use of padded materials during spinal immobilization reduces patient discomfort while maintaining adequate support, although none of these methods completely protected pa-
30
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 18, Number 1 • January 2000
tients from iatrogenic pain. 11,12,14 Lastly, criteria might be developed to permit clinical evaluation in the prehospital setting, thereby avoiding immobilization in those patients who do not need it. This last approach would be difficult to develop and implement nationally, but has been done successfully on smaller scales elsewhere. The large number of patients initially enrolled in the study who needed to be excluded because of insufficient data limits this study. While some of the patients had inaccessible times of immobilization or ED arrival, the majority of the excluded patients lacked recorded time of removal from the board. Many reasons could account for this, but at least some were due to patients being enrolled but not being removed from the backboard before the end of the data collection shift. It would have been interesting to have recorded data at all 3 times on the presence of spine pain in this patient population. This could have been used to verify the studies on iatrogenic pain conducted with healthy volunteers. It would also have provided data to correlate the length of time immobilized with the incidence of pain. Lastly, it would have provided additional information on the clinical sameness or differences between the clinical and radiographic groups. Our results show lengthy immobilization times in the ED even in those patients removed from the backboard prior to radiographs. Given previous reports of immobilization causing discomfort, even in healthy volunteers, an aggressive approach to clinically clear patients on ED arrival and shortening the time to obtain radiographs would seem to be in the patients' best interest.
REFERENCES 1. Mahoney BD: Spinal Injuries,/n Tintinalli JE, Krome RL, Ruiz E (eds): Emergency Medicine A Comprehensive Study Guide. New York, McGraw-Hill, Inc, 1992, p 922 2. Augustine JJ: Spinal Trauma, In Campbell JE (ed): Basic Trauma Life Support for Paramedics and Advanced EMS Providers, ed 3. Upper Saddle River, NJ, Brady, 1998, p 153 3. Caroline NL: Emergency Care in the Streets, ed 4. Boston, MA, Little, Brown, 1991, pp 320-321 4. Bledsoe BE, Porter RS, Shade BR: Paramedic Emergency Care ed 1. Englewood Cliffs, NJ, Brady, 1991, pp 424-425 5. Butman AM, Vomacka RW: Part 1, spine immobilization. Emergency 1991 ;23:48-51 6. Hockberger RS, Kirshenbaum K, Doris PE: Spinal Trauma, In Rosen P, Barkin RM, Braen GR, et al (eds): Emergency Medicine Concepts and Clinical Practice. St. Louis, Mosby Year Book, 1992, pp 406-409 7. Smith M, Bourn S, Larmon B: Ties that bind: Immobilizing the injured spine. J Emerg Med Serv 1989;14:28-35 8. Ausband S, March J, Hightower D, et al: Change in physical examination findings caused by use of spinal immobilization over time. Ann Emerg Med 1996;27:123 (abstr) 9. Lerner EB, Billittier AJ, Moscati R: Effects of neutral positioning with and without padding on spinal immobilization. Prehosp Emerg Care 1998;2:112-116 10. Chan D, Goldberg TR, Tascome A, et al: The effect of spinal immobilization on healthy volunteers. Ann Emerg Med 1994;23: 48-51 11. Walton R, DeSalvo JF, Ernst AA, et al: Padded versus unpadded spine board for cervical spine immobilization. Acad Emerg Med 1995;2:725-728 12. Cordell WH, Hollingsworth JC, Olinger ML, et al: Pain and tissue interface pressures during spine-board immobilization. Ann Emerg Med 1995;26:31-36 13. Cordell WH, Ayers JA, Olinger ML, et al: Duration of immobilization and pain experienced on rigid spine boards. Ann Emerg Med 1996;27:147-148 (abstr) 14. Hauswald M, Johnson DR, Stockhoff C: Superiority of a vacuum splint device over a traditional spine board for immobilization. Acad Emerg Med 1994;1 :A54 (abstr)
28
backboard. Whereas those with spinal injuries clearly need to be protected, large numbers of patients are being subjected to iatrogenic pain when left on backboards for extended periods. These studies all advocate prompt removal o f patients from backboards by utilizing clinical criteria rather than waiting for a radiographic study. It is thought that removing patients from backboards after clinical evaluation may reduce the time patients spend on backboards in addition t6 avoiding unnecessary radiation exposure and increased health care cost. It is, however, unclear how long patients brought to the ED on backboards remain on backboards while in the ED. Our study was designed to quantify the amount of time patients spend on backboards in a clinical ED setting.
MATERIALS AND METHODS A prospective, observational study was conducted over a 10week period on a convenience sample of patients transported by ambulance to the regional adult level one trauma center. The trauma center is a public, urban, tertiary care facility with 400 inpatient beds and approximately 40,000 ED visits annually. ED patients are seen primarily by emergency medicine residents under the supervision of board-certified emergency medicine attendings. Typically, the attending physician or the third-year resident are the only staff members who are allowed to clear patients from backboards. There is no standard protocol for clinical clearance prior to radiographs; however, there is a standard practice among our attendings. No attempts were made to hide the study from our staff; however, it was not common knowledge that the study was taking place. To be included in the study, patients had to have been brought from the scene of injury by ambulance and to have been immobilized on a backboard in the field. In our system, any patient with a mechanism of injury that could possibly result in an injury to the vertebral column is fully immobilized. Patients were not enrolled if they had major multiple trauma or if they were transferred to our facility from another receiving facility. Data collectors were in the ED from 9 a.m. until 11 p.m. Monday through Saturday, and 2 p.m. until 7 p.m. on Sunday. They were instructed to enroll all patients meeting the inclusion criteria who presented during their data collection shift. Time data on patients who were still on backboards at the conclusion of their data collection shift may not have been completed. The data collectors reviewed patients' prehospital care record for the time the ambulance called en route to the hospital, which was used as an approximation of the time the patient was placed on the backboard. They also reviewed the nursing note to determine the time the triage nurse recorded that the patient had arrived at the ED. The time the patient was removed from the backboard, as well as whether the backboard was removed prior to taking radiographs, were directly observed and recorded by the data collectors. The recorded times were then used to calculate patients' total backboard time (TBT) by subtracting the time the ambulance left
LERNER AND MOSCATI I i ED PATIENT IMMOBILIZATION DURATION
the scene of injury from the time the patient was removed from the backboard in the ED. Total ED backboard time (TEDBT) was calculated by subtracting the time the patient arrived on location at the ED from the time the patient was removed from the backboard. Means with standard deviations and ranges were calculated for TBT and TEDBT for all patients as well as the subgroups removed from the backboard prior to radiographs or removed from backboard after radiographs. This study had institutional review board approval.
RESULTS One-hundred and thirty-eight patients were entered into the study. The average age of patients was 37.70 (+19.12) years and the sample was 69% male. Thirty-six patients were excluded from further analysis because insufficient data were collected. Twenty-nine were excluded because the time they were removed from the backboard was not recorded. Two were excluded because it was not recorded whether they were removed from the backboard before or after obtaining radiographs. An additional 5 were excluded because neither the time en route to the hospital nor the time of arrival at the ED were recorded. Of the remaining 102 subjects, 92 included all 3 time data points, whereas 10 included only the time of ED arrival and time of being cleared from the backboard. As a result, these patients were only included in the TEDBT analysis. The TBT was available for 92 patients, and the overall average TBT was 63.63 (_+45.87) minutes. Eighty-five patients were removed from the backboard prior to being radiographed in an average of 53.9 (+30.1, range 10-201) minutes, and 7 were removed from the backboard after being radiographed in an average of 181.3 ( + 41.6, range 133-239) minutes. Comparing the TBT for those patients who were cleared clinically to those who were cleared by radiograph showed a statistically significant difference between the two groups (P < .000001; 95% CI -151.58--- × -< -103.12). There were 102 patients for whom TEDBT was available, and the overall average TEDBT was 46.36 (+44.88) minutes. Ninety-five patients were removed from the backboard before being radiographed in an average of 37.6 (+29.6, range 4-196) minutes, and 7 were removed from the backboard after being radiographed in an average of 165.3 (-+49.7, range 98-231) minutes. Comparing the TEDBT for those patients who were cleared clinically to those who were cleared by radiograph showed a statistical significant difference between the 2 groups (P < .00000t; 95% CI -151.90--< × ~ -103.48). Comparison of the en route time with triage time showed that the overall mean transport time was 15.8 minutes (_+ 11.4). Comparing those patients who were cleared clinically (15.6 minutes -+ 11.3) to those who were cleared by radiograph (14.6 minutes + 8.1) showed there was no significant difference in their prehospital times (P < .72; CI 95% - 4 . 2 3 -< × -< 6.06). Of the 102 patients included in the study, 2 were determined to have spine fractures. One patient was found to have a T4 fracture, and one patient was found to have an L2 fracture. Of those 2 patients, 1 was radiographed prior to removal from the backboard. Both of these patients were
29
stable and did not receive surgery. There were no patients with cervical spine injuries in the study.
DISCUSSION Our results indicate that patients spend long periods of time immobilized on backboards. Removing patients from backboards before radiographs resulted in a decrease in the average amount of time patients spend immobilized on a backboard. However, the group of patients removed from the backboard before radiographs still spent an average of 53.9 minutes immobilized, which meets or exceeds the amount of time reported by most of the previous studies of iatrogenic pain in healthy volunteers. 9-12 The 16 minutes of transport time is below the amount of time shown by previous studies to cause pain, which suggests that immediate removal from the backboard at the "ED door" may reduce the incidence of iatrogenic pain. However, none of the iatrogenic pain studies has looked specifically at the effects of laying on a backboard for less then 30 minutes. Our study did not collect data on whether any of these patients had pain before or after immobilization. However, applying the conclusions of the iatrogenic pain studies, it is possible that many of our patients would have developed pain in the ED as a result of being immobilized that they did not have immediately following their injury. Iatrogenic pain could have resulted in increased time and cost because it might necessitate obtaining radiographs in patients who may not have had spine pain from the injury itself. Other studies have shown an increased incidence and severity of iatrogenic pain with increasing immobilization time. 8,12 This may affect patients who are perceived to be less seriously injured at triage and axe less likely to undergo as immediate an evaluation as those believed to have more severe injuries, particularly when the ED is busy. The immobilized patient without complaints would be immobilized longer waiting to be evaluated. Based upon our observed times and the time required to produce iatrogenic pain in the volunteer studies, this group of initially pain-free patients would be likely to develop pain from immobilization. These patients may in turn fail to meet the clinical criteria for spinal clearance, then necessitating radiographic clearance and still longer immobilization. These patients, in particular, suffer from the universal spinal immobilization concept. There are 3 approaches that could be taken to reduce iatrogenic pain in this group. The first would be to convince emergency physicians that these patients must be evaluated promptly upon arrival to the ED, because delay is painful. However, one must consider the potential disruption to other patient care by implementing such a policy. Secondly, more comfortable immobilization boards or techniques could be developed. Support in the natural contours of the spine that would not compromise immobilization might greatly reduce the incidence of iatrogenic pain. Several studies have shown that the use of padded materials during spinal immobilization reduces patient discomfort while maintaining adequate support, although none of these methods completely protected pa-
30
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 18, Number 1 • January 2000
tients from iatrogenic pain. 11,12,14 Lastly, criteria might be developed to permit clinical evaluation in the prehospital setting, thereby avoiding immobilization in those patients who do not need it. This last approach would be difficult to develop and implement nationally, but has been done successfully on smaller scales elsewhere. The large number of patients initially enrolled in the study who needed to be excluded because of insufficient data limits this study. While some of the patients had inaccessible times of immobilization or ED arrival, the majority of the excluded patients lacked recorded time of removal from the board. Many reasons could account for this, but at least some were due to patients being enrolled but not being removed from the backboard before the end of the data collection shift. It would have been interesting to have recorded data at all 3 times on the presence of spine pain in this patient population. This could have been used to verify the studies on iatrogenic pain conducted with healthy volunteers. It would also have provided data to correlate the length of time immobilized with the incidence of pain. Lastly, it would have provided additional information on the clinical sameness or differences between the clinical and radiographic groups. Our results show lengthy immobilization times in the ED even in those patients removed from the backboard prior to radiographs. Given previous reports of immobilization causing discomfort, even in healthy volunteers, an aggressive approach to clinically clear patients on ED arrival and shortening the time to obtain radiographs would seem to be in the patients' best interest.
REFERENCES 1. Mahoney BD: Spinal Injuries,/n Tintinalli JE, Krome RL, Ruiz E (eds): Emergency Medicine A Comprehensive Study Guide. New York, McGraw-Hill, Inc, 1992, p 922 2. Augustine JJ: Spinal Trauma, In Campbell JE (ed): Basic Trauma Life Support for Paramedics and Advanced EMS Providers, ed 3. Upper Saddle River, NJ, Brady, 1998, p 153 3. Caroline NL: Emergency Care in the Streets, ed 4. Boston, MA, Little, Brown, 1991, pp 320-321 4. Bledsoe BE, Porter RS, Shade BR: Paramedic Emergency Care ed 1. Englewood Cliffs, NJ, Brady, 1991, pp 424-425 5. Butman AM, Vomacka RW: Part 1, spine immobilization. Emergency 1991 ;23:48-51 6. Hockberger RS, Kirshenbaum K, Doris PE: Spinal Trauma, In Rosen P, Barkin RM, Braen GR, et al (eds): Emergency Medicine Concepts and Clinical Practice. St. Louis, Mosby Year Book, 1992, pp 406-409 7. Smith M, Bourn S, Larmon B: Ties that bind: Immobilizing the injured spine. J Emerg Med Serv 1989;14:28-35 8. Ausband S, March J, Hightower D, et al: Change in physical examination findings caused by use of spinal immobilization over time. Ann Emerg Med 1996;27:123 (abstr) 9. Lerner EB, Billittier AJ, Moscati R: Effects of neutral positioning with and without padding on spinal immobilization. Prehosp Emerg Care 1998;2:112-116 10. Chan D, Goldberg TR, Tascome A, et al: The effect of spinal immobilization on healthy volunteers. Ann Emerg Med 1994;23: 48-51 11. Walton R, DeSalvo JF, Ernst AA, et al: Padded versus unpadded spine board for cervical spine immobilization. Acad Emerg Med 1995;2:725-728 12. Cordell WH, Hollingsworth JC, Olinger ML, et al: Pain and tissue interface pressures during spine-board immobilization. Ann Emerg Med 1995;26:31-36 13. Cordell WH, Ayers JA, Olinger ML, et al: Duration of immobilization and pain experienced on rigid spine boards. Ann Emerg Med 1996;27:147-148 (abstr) 14. Hauswald M, Johnson DR, Stockhoff C: Superiority of a vacuum splint device over a traditional spine board for immobilization. Acad Emerg Med 1994;1 :A54 (abstr)