Management of Paediatric Head Injuries in Sweden: A 5 National Cross-Sectional Survey

utilise protocols for use routines and provide adequate information to patients/guardians at discharge.

utilise dose-reduction protocols for CT, use written observation routines and provide 1 adequate information to patients/guardians at discharge. 2 3 Keywords 4 mTBI; TBI; children; guidelines; initial management; Sweden; descriptive 5 Background 6 7 Traumatic brain injury (TBI) is recognized as a common cause of death and disability among 8 children worldwide [1]. The incidence varies; a recent review on global epidemiology of TBI 9 in children and adolescents showed that 691/100 000 children per year were treated in 10 emergency departments (ED´s), 74/100 000 was admitted to hospital and median mortality 11 rate was 9/100 000 children and year [2]. Most patients (70-98%) are classified as having 12 minimal or mild traumatic brain injury (mTBI) primarily from their initial level of 13 consciousness (Glasgow Coma Scale, GCS) with or without specific risk factors [3][4][5][6]. Two 14 thirds of these patients will swiftly recover without suffering from any persisting sequelae 15 [7,8]. However, a minority will suffer life-threatening intracranial haemorrhages in need of 16 urgent attention and often rapid surgical intervention [9]. The task of effectively identifying 17 these patients is a clinical challenge. 18 19 Computed tomography (CT) is the gold standard used to detect intracranial complications 20 after TBI, such as intracranial haemorrhages. CT is readily available in the western world and 21 relatively inexpensive and simple to use. However, as CT utilizes ionizing radiation, a scan 22 and achieve a high level of compliance. A previous national survey study [14], which showed 1 considerable variations in management, allows analysis of changes in management over time . 2 Also, although the SNC16 has not been officially implemented, the guideline may be in 3 clinical use due to the lack of accepted alternative options. 4 The aim of this study is therefore to describe current management of these children in 5 Sweden. Secondary aims are to analyse differences in management over time, to assess the 6 implementation of the SNC16 and to analysis possible variations in care. 7 Methods 8 This is a descriptive cross-sectional study including all emergency departments at hospitals 9 with the possibility of in-hospital care. Data was collected using a web-based survey designed 10 in collaboration with members of the Scandinavian Neurotrauma Committee (SNC). The 11 survey was designed to answer the most crucial questions regarding the current management 12 of children with TBI in Sweden. The questions were structured into 5 different sections (table  13 1), exemplifying questions from each part. The questionnaire was completed using the 14 program esMaker (Entergate AB). A primary pilot version was sent to 6 hospitals and after 15 minor details adjusted the complete form was finished and sent. 16 17 • Are these patients cared for by specialists or non-specialists?

Section 3: Radiology
• What primary radiology modality is used?
• Access to anaesthesiologist and diagnostic radiology Section 4: In-hospital observation • What department are patients admitted to in need of in-hospital observation?
• What parameters are being monitored during hospitalization?

Section 5: Discharge and follow-up
• Are patients and guardians provided with discharge information?
• Does your hospital arrange follow up?

1
The survey was answered once per participating hospital. Initial contact was established by 2 phone to ensure that a suitable responder, able to provide valid information and an overall 3 view, was reached. When appropriate recipients had been identified, the questionnaire was 4 distributed by e-mail. Data was collected continuously from June 2020 to March 2021. 5 Reminders were sent by e-mail twice to non-responders until >90% response rate was 6 reached. 7 Data is summarized and presented using descriptive statistics. A cross-comparison was 8 performed between four categories depending on the size of the hospital; local, regional, 9 university and children´s hospital. Further statistical analysis was performed using Fisher's exact test to detect differences between groups, when indicated. A two-tailed p-value of 1 <0.05 was considered significant. Due to known small numbers in the last two groups, the 2 categories were a priori dichotomised to smaller hospitals (local and regional) and larger 3 hospitals (university and children´s hospitals). As answers for a certain question were not 4 always 100% complete, the total number of responses is given with each question. 5 Only the overall management of children with mild head injury was collected in this study. 6 Neither personal data nor individual patient records were obtained. An ethical advisory 7 opinion was granted by the Swedish Ethical Review Authority, Dnr 2020-02693. 8

9
Initial contact was established with 76 hospitals in Sweden; 5 hospitals did not manage acute 10 TBI at all. Of the remaining 71 hospitals, responses were returned from 66 hospitals (overall 11 response rate 93%). 56 of these hospitals managed paediatric patients with TBI (10 only 12 managing adult TBI) and form the base of this study. The size and type of the included 13 hospitals (n = 56) varied; 28 local hospitals, 19 regional hospitals, 4 dedicated children´s 14 hospitals and 5 university hospitals. Of the 5 non-responding hospitals, 2 were local hospitals, 15 2 were regional hospitals and one was a university hospital (figure 1). 16 The age span which was used by participating units to define the patients as paediatric 4 differed. Most common span at the non-paediatric hospitals were 0-18 years of age. In 5 contrast, children´s hospitals had a different definition, with the upper limit differing between 6 14-16 years of age. 7 Nine (16%) hospitals reported limited access to a neurosurgical unit, with a transfer time of > 8 2 hours to the closest neurosurgical service. 9 In total, 76% (42/55) of the respondents reported use of an established guideline for the 11 management of paediatric TBI at their hospital ( Table 2). When comparing smaller hospitals 12 (local and regional) to larger ones (university and children´s), there was no statistically 13 significant difference in presence of guidelines (p=0.18). The most commonly used guideline 14 was the SNC16 guideline (n=31, 55%). Following this, the most commonly described was a 15 local modification of pre-existing validated guidelines (such as PECARN) or local guidelines 1 based on local expert opinion.  Children with TBI were predominantly managed by non-specialists (Table 3). There was no 9 statistical difference in the presence of specialists between small (local and regional) and 10 large (university and children´s) hospitals (p=0.17). Apart from dedicated children's hospitals, 11 children with TBI were rarely treated by a doctor with a paediatric speciality (Table 4). Large 12 (university and children´s) hospitals had significantly higher presence of paediatric 13 specialities than small (local and regional) hospitals (p<0.001). 14 15 16 4 "assistant physician, dependent", "assistant physician, independent", "intern", "resident" and "specialist".

5
For each "category" (experience level) of physician the respondent rated on a 5-grade scale (always; often; 6 sometimes; rarely; never) how frequent this category manage children with head trauma at their emergency 7 department. A dichotomisation of the experience levels to "specialist" and "non-specialist level" (in which 8 categories: "assistant physician, dependent", "assistant physician, independent", "intern" and "resident" 9 were merged) was done for the analysis. To further simplify presentation, grade "always" and "often" was 10 merged (implying the "most common" experience level for clinicians manging children with TBI) and presented 11 for respective hospital size. This means that if grade "sometimes", "rarely" or "never" was chosen for a 12 category (experience level) of physician it won´t be presented in the

14
As shown in table 4, most of the patients are managed by physicians in non-paediatric 15 specialities (75%), of which general surgery represents 71% (n=40). In 34% (n=19) of the 16 hospitals, emergency medicine physicians often or always manage paediatric patients with 17 TBI. It was uncommon (5%) that initial assessment always or often was done by a doctor 18 specialising in neurology (Table 5). 19 20  15 hospitals (27%) reported that these patients are occasionally discharged by a nurse at 12 triage without any doctor assessment. This management was more common (p<0.001) in 13 university and children´s hospitals (n = 6/9) when compared to local and regional hospitals (n 14 = 9/47). 8 of these units had written guidelines concerning this procedure; 6 of these used the 15 programs. However, many respondents (23/56, 41%) were not aware of any dose-reduction 21 protocol, with only a few hospitals (2/56, 4%) reporting that a dose-reduction protocol was 22 not routinely used and 3 respondents (5%) did not answer the question at all. If needed during 23 the head CT scan, 50% of the respondents reported occasional use of sedation in some form. 24 All hospitals had access to anaesthesiologists irrespective of time or day. 25 In 75 % of the hospitals (42/56), it was possible to admit patients for in-hospital observation. 1 Most commonly (64%), these children were admitted to a general ward. In 46% (13/28) of 2 smaller, local hospitals, children could not be admitted in-house and needed to be transferred 3 to another hospital if admission was necessary, see Table 6 for details. In local hospitals, 50% 4 (7/14) of children were observed in a non-paediatric ward. In larger hospitals, most children 5 were admitted to paediatric wards (96%, 25/26). In one hospital, the Intensive Care Unit 6 (ICU) was used for observation.  Respondents were asked to report which scale(s) that was used at their hospital for assessment of level of 8  relatively more common at large hospitals (7/9, 78% vs 14/47, 30%) and usually either at a 23 paediatric outpatient clinic (52%) and/or in the primary care sector (48%). 46% (26/56) did 24 not provide or plan follow-up in children following TBI; this was more common in small hospitals (25/26). There was a significant difference in follow-up routines between small 1 (local and regional) and larger (university and children´s) hospitals (p=0.015).  This national cross-sectional survey aims to describe and analyse the current management of 2 children with TBI in Sweden. With a high response rate (>90%) we have been able to 3 efficiently collect data concerning different aspects of mTBI management. Most hospitals 4 (76%) use an established guideline to aid in management and we did not observe a difference 5 of guideline use between sizes of hospital. Most use the SNC16 guideline and the majority of 6 the remaining hospitals use a locally constructed guideline, either a modification of an 7 established guideline, such as PECARN, or one based on expert opinion. 8 9 In 27 % of the hospitals, in particular larger hospitals, nurses can discharge children with 10 mTBI without any assessment by a doctor. These patients reasonably represent the mildest of 11 injuries and are in 53% (8/15) of the hospitals discharged using a guideline, mainly the 12 SNC16. If children can reliably be assessed by a nurse and judged to be in the mildest risk 13 group of TBI, this type of management may be efficient. Further studies could evaluate this 14 issue. 15 16 Children with TBI were managed predominately by non-specialists and (outside of dedicated 17 children´s hospitals) non-paediatricians. These findings reinforce the need of a nationally 18 implemented and accepted guideline for these patients, as most children with TBI will be 19 managed by inexperienced doctors from varying specialities, especially in smaller hospitals. 20 The field of emergency medicine is relatively young in Sweden, but may be the primary 21 group to manage TBI patients in the future. 22 Sweden is not a densely populated country with some concentrations of inhabitants in larger 1 cities. Due to this fact, large university hospitals are generally located in areas where many 2 inhabitants reside. Large parts of Sweden are therefore some distance from these hospitals 3 which contain the neurosurgery departments. Indeed, 16% of hospitals reported having at 4 least 2 hours transfer time to the nearest neurosurgical unit. This aspect is important, as the 5 severe complications after mTBI, although uncommon [9,27], require immediate attention 6 and often neurosurgical expertise. Due to the potential risk of ionizing radiation in children, the aspect of dose-reduction 18 protocols is highly relevant. Although approximately half of hospitals stated that routine 19 reduction of radiation dose was used, many respondents could not reliably answer this 20 question. In addition to the radiology departments, the referring party must also be made 21 aware of the radiation issues related to CT scans [17]. In order to fully investigate this issue, a 22 survey directed at the radiological department of the hospitals is warranted. 23 Following TBI, children may be manged (with CT, in-hospital observation, or both) in order 24 to detect possible severe complications. In the absence of these, these children may be discharged. The criteria for discharge may vary between hospitals but usually include the 1 absence of worrying signs and symptoms. More than half of hospitals did not have specific 2 discharge criteria for these patients. Although evidence is lacking in this area, written 3 discharge criteria may facilitate management and promote equality in patient management. 4 Discharge from the hospital, be that after initial assessment in the ED or after in-hospital 5 observation, should be accompanied with information regarding the injury, what to expect 6 and when to seek health care. Pleasingly, most hospitals provided this with only 3 hospitals 7 (5%) stating that they do not provide such information. in the SNC16 guideline, including other aspects than the risk criteria, such as observation routines, in Sweden is gratifying as it implies introduction of a sound evidence-based 1 approach in the management of head injuries in children. As these guidelines are still to be 2 internally validated in the Scandinavian setting a certain amount of vigilance is still required, 3 although external validation has shown encouraging results [22]. 4

5
The strengths of this study lie in the high response rate and the on-line survey system which 6 increases response accuracy and minimises ambiguous answers. As this group did a similar 7 survey in 2006, most questions are similar which allows a comparison over time. This study 8 has several limitations. Despite our best efforts to ensure that the respondent was fully aware 9 of all aspects of TBI management at their respective hospital, inaccuracies may still have 10 occurred. Also, the questions response rate was not always 100% within the same hospital 11 which may also account for some errors. However, these issues were minimal and have little 12 effect of the overall results. Ethics approval and consent to participate 10 The study does not include individual patient data. Ethics approval by the Swedish Ethical 11 Review Authority, Dnr 2020-02693 12 Consent for publication 13 Not applicable. 14 Availability of data and materials 15 The datasets used and/or analysed during the current study are available from the 16 corresponding author on reasonable request. 17 Competing interests 1 None of the authors have any financial competing interests. JU is a member of the SNC 2 committee, a non-profit organisation independent from financial company support, who are 3 responsible for the SNC16 guidelines. 4