The characteristics and pre-hospital management of blunt trauma patients with suspected spinal column injuries: a retrospective observational study

Patients who have suffered blunt trauma resulting in spinal column injuries, such as spinal fractures or dislocations, are at risk of developing iatrogenic spinal cord injury (SCI) due to physical movement or manipulation [1‐5].

SCI is defined as a traumatic injury to the spinal cord that results in loss of motor and/or sensory functions [6]. In a European cohort (n = 250,584) of severely injured trauma patients (excluding penetrating injuries), 13.2 % of immobilised patients had vertebral column injuries and 1.8 % sustained a SCI [7, 8]. It has been postulated that spinal immobilisation by emergency medical services (EMS) is required for all patients with suspected vertebral column injuries to prevent SCI after blunt trauma [1].

According to the 8th edition of the Advanced Trauma Life Support (ATLS) guidelines, spinal immobilisation should be maintained by a rigid cervical collar in combination with head blocks strapped to a spine board [9]. In the Netherlands, the EMS spinal immobilisation guidelines have been adjusted in 2002 and 2006 in accordance with the Pre Hospital Trauma Life Support (PHTLS) guidelines [10, 11]. According to the guidelines of 2006, full spinal immobilisation is only indicated in patients who have sustained blunt trauma and show one or more of the following symptoms: neck/back pain or tenderness, altered level of consciousness, neurological deficits and evidence of intoxication or painful distracting injuries. Departure from the guidelines is allowed, however, in case of neck muscle spasms, increased pain, increase of neurological deficits, signs of increased intracranial pressure (ICP) or combativeness/resistance of the patient [11]. In these situations, the EMS staff can opt for a rigid collar only or head blocks with spine board only.

Despite the assumed beneficial effect of the method of spinal immobilisation advocated by the ATLS guidelines, there is growing evidence that immobilisation is associated with severe adverse effects including serious respiratory problems [11‐13], increased ICP [14‐16], delirium [17], iatrogenic pain or discomfort [18, 19] and possible deterioration of SCI [2, 20‐23]. Furthermore, spinal immobilisation can cause a delay in transportation time to the hospital, which can negatively influence outcomes in patients with SCI [24].

A limitation of these studies is that they were mainly hospital-based and lacked a full pre-hospital description of, for example, patient characteristics, immobilisation techniques and adverse effects of spinal immobilisation. There is a paucity of data on the characteristics of patients who received pre-hospital spinal immobilisation and on the adverse effects of immobilisation that may occur during this phase. Furthermore, it is unknown whether EMS staff follows the 2006 spinal immobilisation guidelines with regard to applied techniques when taking care of patients with spinal column injuries. Our study had three main research goals. First, we aimed to determine the pre-hospital characteristics of blunt trauma patients with suspected spinal injuries that were immobilised by EMS staff. Second, we assessed the reasons for spinal immobilisation and the choices made by EMS staff regarding spinal immobilisation techniques. Third, we researched the occurrence of possible adverse effects of immobilisation during the pre-hospital phase.

A total of 1082 patients received spinal immobilisation in the region Gelderland-Zuid between January 2008 and December 2012. A total of 654 (60.4 %) patients were transferred to a Level 1 trauma centre and 428 (39.6 %) patients to a Level 2 trauma centre.

To our knowledge, this is the first study that gives a comprehensive overview of the characteristics and pre-hospital management of blunt trauma patients with suspected spinal column injuries.

Our first objective was to determine the pre-hospital characteristics of this category of patients. The most important findings were that the EMS staff suspected spinal column injuries in 402 (37.2 %) patients and that 62 (5.7 %) patients had symptoms of SCI. The EMS staff did not report the NRS in 771 (71.3 %) patients. Painful distracting injuries were found in 146 (13.5 %) patients. The standard method of immobilisation was not used in 171 (15.8 %) patients. Finally, in 22 (2 %) cases adverse effects were reported by the EMS. These included pain due to spinal immobilisation (n = 10, 0.9 %,); shortness of breath (n = 3, 0.3 %); combativeness or anxiety (n = 6, 0.6 %); and worsening of pain when supine (n = 1, 0.1 %).

Our second objective was to assess the reasons for spinal immobilisation and the choices made by EMS staff regarding spinal immobilisation techniques The reason for spinal immobilisation was clear to the EMS staff in most cases (n = 1059, 97.9 %). There was a significant mechanism of injury (Fig. 1, box 1) combined with specific signs and symptoms (Fig. 1, box 2). We found that consensus on the implementation of the criterion ‘distracting injury’ was lacking among EMS staff. A total of 146 (13.5 %) of the patients who did not have spinal tenderness after blunt trauma were immobilised because of a painful injury. There is, however, a difference between a painful injury and a distracting injury. While a distracting injury is a criterion for spinal immobilisation according to Fig. 1, box 2, this criterion was interpreted differently by the various EMS staff. A distracting injury is defined by PHTLS as follows: long bone fractures, visceral injury requiring surgical consultation, large laceration, degloving or crush injury, large burns and any other injury producing acute functional impairment (Table 7) [10]. Although the most painful injuries may be considered as ‘distracting’, we could not demonstrate an association between high NRS scores at arrival of the EMS staff and distracting injuries as a reason for spinal immobilisation. The mean NRS score in the group of patients that were immobilised based on the criterion ‘painful distracting injuries’ was 3.2 (n = 52) with a mode of 0.

Previous studies by Hefferman et al. [28] and Domeier et al. [29] researched whether the definition of a distracting injury could be narrowed. Hefferman et al. focused on the c-spine and showed that patients with an upper torso injury, in cases of a non-tender cervical spine, might have sustained a cervical spine injury. Domeier et al. redefined the term ‘distracting injury’ as: ‘a suspected extremity fracture proximal to the wrist or ankle’.

There is increasing evidence that a distracting injury, as currently defined, does not affect the sensitivity of the physical examination. Konstantinides et al. [30] concluded that only the upper chest injuries may be significant enough to decrease the sensitivity of the physical examination of the cervical spine in alert and non-intoxicated patients blunt trauma patients. Furthermore, Dahlquist et al. [31] showed, that femur fractures should not be considered as distracting injuries for cervical spine assessment. Clinical examination is a sensitive screening method for thoracolumbar spine clearance in patients with distracting injuries [32].

Therefore, further research and clarification of the criterion ‘distracting injury’ or a narrowing of the definition (following Hefferman et al., Domeier et al.) based on pain scores combined with injury assessment is warranted. We believe this will ultimately lead to a reduction of adverse effects of spinal immobilisation and potentially to a decrease of the number of patients, that is unnecessarily exposed to X-rays.

In 23 (2.1 %) patients spinal immobilisation was based on the mechanism of injury criteria (Fig. 1, box 1). Departure from the standard method of spinal immobilisation was found in 171 (15.8 %) cases. A high CPI was not a reason for removal of the rigid collar, despite the fact that 75 (6.9 %) of our cohort had signs of increased ICP. No statistically significant differences could be found for the application of a rigid collar between groups of patients with or without signs of high ICP. The recommendation to remove the rigid collar in case of increased ICP was made by the PHTLS in 2002 and accepted the NPA in 2006. The reason that EMS staff did not depart from the standard method of spinal immobilisation in cases of high ICP may be explained by the non-explicit naming of the removal of the rigid collar in the NPA. A lack of awareness of this guideline can cause further increase in ICP [14]. The arguments used by EMS staff to depart from the standard method of spinal immobilisation were in accordance with the Dutch NPA protocol. One exception concerned a patient who was immobilised by spine board only because of complaints of lower back pain. The patient did not complain of pain in the cervical region. We found that the documentation of reasons for departure from the standard method was inadequate. It remains unclear, for instance, why 139 (12.8 %) patients were not immobilised based on the standard method. Departure from the standard method is permissible if the EMS staff can substantiate this departure. Patients and other healthcare providers have reason to expect that pre-hospital treatment follows current protocols or guidelines. It is therefore vital that the reasons for departure from the standard method are adequately documented by EMS staff. Adequate documentation leads to improved transparency of pre-hospital care and is necessary when the EMS staff or organisation has to account for its decisions [33].

The third objective of our study was to research the occurrence of possible adverse effects during the pre-hospital phase. Previous research has found adverse effects of spinal immobilisation during this phase. In their study on the effects of spinal immobilisation on healthy volunteers, Kwan et al. [18] found that 55 % of healthy volunteers complained of moderate to severe pain within 30 min after spinal immobilisation. In addition, Bauer and Kowalski [26] found that certain devices used for spinal immobilisation had restrictive effects on the pulmonary function in the healthy, non-smoking man. Since the average on-scene and transportation times exceeded 30 min, we expected to find adverse effects of spinal immobilisation. Our findings did not correspond with previous research, however, adverse effects were found in only 22 (2 %) cases (pain due to spinal immobilisation (n = 10, 0.9 %,); shortness of breath (n = 3, 0.3 %); combativeness or anxiety (n = 6, 0.6 %); and worsening of pain when supine (n = 1, 0.1 %). A possible explanation for this discrepancy could be that the pre-hospital time is too short for the occurrence of the described adverse effects. Another explanation could be that pre-hospital data were incompletely documented, which was frequently observed in our study. The chosen design (retrospective EPR study) may also underestimate the true rate of adverse effects [34]. Recently, there is a lot of debate and research ongoing in the call for alternative spinal motion restriction. There are many side effects known of spinal immobilisation, and researchers recently came up with evidence that cervical spine movement was up to four times as high during extrication by EMS as by controlled self-extrication [35]. To reduce the adverse effects caused by spinal immobilisation, the National Association of EMS physicians (NAEMSP) and the American College of Surgeons Committee on Trauma (ACS-COT) have published a position paper in 2013. In this paper they describe that the utilization of backboards for spinal immobilization during transport should be judicious and not be used at all times. A professional consideration should take into account the benefits as well as the risks [36].

Finally, we found a high number of patients with symptoms (determined by EMS staff) of SCI in our study compared with a large European cohort study (5.7 versus 1.8 %, respectively) [37]. Nevertheless, it is difficult to draw comparisons, for while EMS staff can detect symptoms of SCI such as muscle weakness, paralysis or altered sensation, numbness, tingling or loss of sensation in hands, fingers, feet or toes [38], it lacks a validated instrument to uniformly detect symptoms of SCI. The Emergency Department (ED) uses standards of the American Spinal Injury Association (ASIA-ICLOS) to assess damage to the spinal cord of patients. These standards are not applicable in the pre-hospital setting for practical reasons. Furthermore, the EPR of the EMS does not state that documenting symptoms of SCI is mandatory. This means the number of symptoms of SCI in the pre-hospital setting might be over- or underestimated in our study. For future research it is necessary to develop a validated and uniform instrument to measure symptoms of SCI in the pre-hospital setting. This enables clinicians to monitor symptoms of SCI over time and to see whether the patient shows signs of deterioration or improvement. This knowledge can contribute to the evaluation of the effectiveness of the current (pre) hospital spinal immobilisation guidelines and improve the quality of care.

This study has some limitations. Data were obtained from only one of the 25 EMS organisations in the Netherlands and may not be representative for spinal immobilisation care in other regions than Gelderland-Zuid. To overcome this limitation, we used a large sample size and we included both rural and urban areas. Despite the regional differences in training and education, all 25 EMS organisations follow a national protocol and their staff is certified by an independent national organisation (Dutch Ambulance Institute).

Another limitation is that we were not able to demonstrate the appropriate use of and adherence to the spinal immobilisation protocol as outlined in Fig. 1 because we did not include patients after blunt trauma who did not receive spinal immobilisation. Furthermore the final outcome of spinal immobilised patients is unknown. Only suspected cases are described and we were not able to compare the patients with suspected spinal column injury with patients in which actual vertebral column/cord injury was diagnosed.

A final limitation concerns the fact that data on a number of variables were incorrectly recorded or missing in the EPR. Adequate documentation by ambulance staff is not only vital for the provision of good care, but also important for trauma research [33, 39, 40]. The lack of adequate EMS documentation has been previously reported [40]. We tried to minimise information bias by analysing the coded data in combination with free text fields. Based on the results of our study, we recommend a revision of the EMS documentation protocol in order to improve the collection of data for research purposes.

This retrospective observational study described the characteristics and pre-hospital management of patients who received spinal immobilisation by EMS staff. Suspicion of spinal column injuries was documented in 37.2 % of the patients and in 5.7 % of the patients EMS staff reported symptoms of SCI. The combination of a rigid collar, spine board with straps and head blocks (full spinal immobilisation) was used in 84 % of the patients. The remaining 16 % received an alternative or incomplete form of spinal immobilisation. Although one-third of the patients showed signs of head injury and 7 % of these patients had signs of increased ICP, we could not demonstrate the removal of the rigid collar in cases of increased ICP (as advised by PHTLS).

Evident complications of spinal immobilisation were described in less than 2 % of the patients by EMS staff. A reduction in the number of patients who require spinal immobilisation can be achieved by clarifying the term ‘distracting injury’. Pain scores were under-recorded by the EMS staff in this cohort.

Finally, attention should be given to improving the process of data registration to make data more reliable. Good documentation is fundamental to scientific research. Therefore, management and staff of the ambulance services should be encouraged to improve their digital records to contribute to future pre-hospital research.