- Email: [email protected]
Neuroimaging provides relevant clinical information in patients with burn injuries
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
xxx
ScienceDirect journal homepage: www.elsevier.com/locate/burns
Neuroimaging provides relevant clinical information in patients with burn injuries
,
$ $$
Elizabeth Concannon a, * , Louise Fitzgerald a, Emma Canniff b, John Birrane a , Joseph Harbison c , Odhran Shelley a a
Department of Plastic and Reconstructive Surgery, St James’s Hospital, Dublin, Ireland Department of Radiology, St James’s Hospital, Dublin, Ireland c Department of Medicine for the Elderly, St James’s Hospital, Dublin, Ireland b
article info
abstract
Article history:
Introduction: Neurological assessment of patients with burn injuries may be complicated by a
Accepted 29 August 2019
variety of factors including artificial ventilation and sedation, cerebral hypoxia and
Available online xxx
intoxication. Medically unstable intubated patients present logistical challenges for radiological imaging. The role of neuroimaging as an adjunct to clinical assessment of
Keywords: Burn injury Neuroimaging Fazekas score Brain injury Resuscitation Cerebrovascular disease White matter disease Rehabilitation
burn injured patients has not yet been determined. Aim: This study aims to investigate the indications, findings and outcomes of neuroimaging studies performed for burn injured patients. Methods: A retrospective case series study of adult burn patients admitted over an 8 year period was completed in the National Burns Centre at St James’s Hospital, Dublin. Neuroimaging studies carried out for patients admitted during the study period were reviewed by a Consultant Radiologist and Consultant Stroke Physician. Outcomes included neuroimaging findings, prevalence of white matter disease (Fazekas scale), length of stay, discharge destination, predicted and observed mortality. Results: 1328 consecutive patients with burn injuries were admitted during the study period. 56 patients underwent neuroimaging studies with computerised tomography, magnetic resonance imaging or both. 46 out of 56 neuroimaged patients (82.1%) had significant radiological findings, including 14 patients (25%) with acute findings. There was a high prevalence of white matter disease (mean total Fazekas score: 3.59) and acute cerebral infarction (7 patients). Patients with radiological findings required additional in-patient rehabilitation and had increased length of stay (Median 47.0 days vs. 27.5 days, p < 0.027). Patients with resuscitation burns or associated inhalation injury were significantly more likely to undergo neuroimaging (p < 0.0001) and to have positive radiological findings. Predicted mortality was higher in patients with positive neuroimaging findings compared to patients with normal neuroimaging studies, although there was no significant difference in observed mortality between these two groups.
$
This is the largest study to date investigating the role of neuroimaging in civilian burn patients. Neuroimaging provides valuable clinical information, when indicated, detecting findings that can exert direct influence on burn patient care, influence outcomes and predict rehabilitation needs. Won prize for Best Poster Presentation at the annual British Burns Association meeting May 2019, Leeds. Presented at the Irish Association of Plastic Surgeons Meeting, Cork, Ireland, May 2019 (Oral). * Corresponding author at: National Burns Unit, St James’s Hospital, James Street, Dublin 8, Ireland. E-mail addresses: [email protected] (E. Concannon), [email protected] (L. Fitzgerald), [email protected] (E. Canniff), [email protected] (J. Birrane), [email protected] (J. Harbison), [email protected] (O. Shelley). https://doi.org/10.1016/j.burns.2019.08.018 0305-4179/© 2019 Elsevier Ltd and ISBI. All rights reserved. $$
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
2
burns xxx (2019) xxx
xxx
Conclusion: Neuroimaging is used appropriately in patients admitted with burns and provides valuable applicable clinical information when indicated. © 2019 Elsevier Ltd and ISBI. All rights reserved.
1.
Background
Each year approximately 13,000 burn injuries which require hospital attention occur in the UK with 10% of those hospitalised suffering a life threatening burn [1]. Across Europe, between 4 and 22% are admitted to intensive care from presentation [2]. In 2004, nearly 11 million people worldwide were burned severely enough to require medical attention [3]. Burns, therefore, represent the fourth most common mechanism of all injuries after road traffic accidents, falls and interpersonal violence. Fortunately, an ongoing global reduction in burn related mortality has been demonstrated due to advances in the knowledge of burn pathophysiology and multidisciplinary critical care provided by specialised burn centres [4]. Neuroimaging represents a radiological adjunct to the secondary survey, when indicated, in Advanced Trauma Life Support1 management of trauma patients. However the role of neuroimaging in assessment and management of burns has not been specifically assessed. Burns are among the leading causes of disability-adjusted life-years lost in low- and middle-income countries. The cost of rehabilitation, time off work and loss of earnings represent a substantial cost to the community following acute hospital treatment [5]. This is increased by the presence of concomitant intra cranial pathology. Intra cranial injury, secondary to an associated trauma or hypoxic brain injury as a result of burn injury has the potential to impair and delay rehabilitation, however no previous research has specifically addressed this. Several criteria have been developed to guide indications for neuroimaging following head injury including New Orleans Criteria, Canadian CT Head Rules or National Emergency X-Ray Utilization Study-II clinical criteria [5]. Head injury assessment guidelines from the National Institute for Health and Care Excellence (NICE) govern practice in the United Kingdom and Ireland [6]. Computerised tomography (CT) brain studies are generally indicated following head injury if criteria for risk of neurosurgical intervention are present. No specific radiological criteria are available to guide neuroimaging of patients following burn injury with or without head injury. Neurological assessment of patients with burn injury may be confounded by a multitude of factors. Cerebral hypoxia due to significant inhalation injury, carbon monoxide toxicity or pulmonary oedema may indirectly alter cognitive function. Early endotracheal intubation of burn patients may be a lifesaving manoeuvre but represents an obvious barrier to neurological assessment as is opiate analgesia, to which elderly patients may be particularly sensitive. Burn patients have high rates of psychiatric comorbidity, including dementia [7], alcohol and polysubstance dependence [8] compared to the general population, which may complicate neurological observation and obscure the diagnosis of acute pathology. Haemodynamic instability may also cloud clinical examination. Furthermore, mechanisms of burn injury and their
interpretation may vary. For example, an explosion could be mistakenly interpreted as a flash flame burn. Head injury could also be sustained during rapid extrication from a fire. Primary neurological events such as acute stroke or seizure activity may also incite burn injury and a detailed history of events is not always available. CT brain undertaken to assess neurological function in these scenarios aims to investigate the possibility of intracranial pathology. An expansion in the current head injury assessment guidelines [5,6] is warranted to account for the wide range of indications for neuroimaging that may be present following burn injury. There is a paucity of data pertaining to the utility of neuroimaging in burn patients. The results of one retrospective study of burn inpatients who underwent CT brain imaging [9] demonstrated very few positive radiological findings and concluded that caution should be exercised when deciding to transport intubated, potentially unstable critical care patients to radiology departments, given the inherent risk of adverse events [10]. This study aims to assess the indications for neuroimaging in burn patients and whether their findings and outcomes provided useful clinical information.
2.
Methods
A retrospective review of all burns inpatients over the 8 year study period, from January 2008 to January 2016 was performed in the National Burns Centre at St James Hospital, to which all major burns in Ireland are referred. This quaternary referral centre provides a local, regional and national burns service to a population catchment of 4.7 million. Data was sourced from the HIPE system (Hospital In-Patient Enquiry), medical charts, electronic patients records and burns admission data. Burns data is collected electronically and coded prospectively within the National Burns Unit on patient admission using these databases. Approval for the study was granted by local institutional review board. As this observational research was carried out retrospectively with subjects who had received emergency treatment, a waiver of regulatory requirements for obtaining informed consent applies. All patients who underwent neuroimaging during their admission for management of burns were included. Data collection included age, gender, length of stay, total body surface area sustained (TBSA), burn mechanism, inhalation injury, indication for neuroimaging and modality of neuroimaging performed, neuroimaging findings and patient outcomes, including mortality, neurosurgical intervention and discharge destination. Imaging studies were reviewed by a Consultant Radiologist and by a Consultant Stroke physician with a special interest in neurovascular pathology (co-author JH). There were no new or additional radiological findings identified retrospectively when compared to the initial Consultant Radiologist report,
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
3
xxx
Table 1 – Demographics of burn patients admitted between 2008 and 2016.
Number Mean age Mean TBSA Inhalation injury Formal fluid resuscitation Median LOS Observed mortality Mean revised Baux score
a b
All burn patients
Neuroimaged burn patients
Positive radiological findings
No radiological findings
1328 54.8 5.6% 147 (11.0%) 215 19.5 39 (2.9%) 63 (2% mortality)b
56 57.5 14.8% 17 (30.3%) 25 41.5 2 (3.6%) 77 (5 10% mortality)b
46 63.5 15.8% 17 25 47.0a 2 (3.6%) 80 (5 10% mortality)b
10 51.1 11.4% 0 0 27.5a 0 66 (2 3% mortality)b
TBSA: Total Body Surface Area burned. LOS: Length of stay. Significant difference in LOS with p < 0.027, Kruskal Wallis test. Predicted probability of death using logistic regression model of revised baux score [12].
apart from calculation of a score for severity of white matter disease. The Fazekas scale [11] is a scoring system used to quantify the amount of white matter disease, best identified on T2-weighted FLAIR MRI images. Fazekas grading ranges from 0 in normal white matter, to 3 in severe white matter disease. Periventricular white matter (0 3) and deep (0 3) white matter disease scores were added to generate a Total Fazekas score (0 6), utilised for reporting the results of this study. Predicted mortality was calculated using logistic regression model of the revised Baux score [12]. This model has been validated for mortality prediction following burn injury in Ireland and internationally [13]. A standard statistical software package (SPSS1 v25) was used for data analysis with p-value <0.05 denoting statistical significance.
3.
Results
There were 1328 consecutive patients admitted to the National Burns Unit during the 8 year study period, between January 2008 and January 2016 (Table 1). The overall cohort was comprised of 697 males and 631 females with a mean age of 54.8 (Range 17 94). The mean TBSA burned of the overall cohort was 5.6% (Range 0.5 98%). Inhalation injury was present in 147 patients (11% of overall cohort). Fifty-six burn patients (4.2% of overall cohort of 1328 patients) underwent neuroimaging with CT or MRI. The neuroimaged patient cohort was comprised of 31 males and 25 females with a mean age of 57.5 years (Range 20 86 years). The mean TBSA of neuroimaged patients was 14.8% (Range 0.5 96%). The predominant burn injury mechanism in the neuroimaged patient cohort was thermal injury from flame, contact or scald. Similar burn injury mechanisms were predominant in burn patients who did not undergo neuroimaging studies. Three patients out of 56 (5.4%) sustained electrical injuries and a further three patients had burns secondary to an explosion. Seventeen patients (30.4%) had an associated inhalation injury. In terms of neuroimaging modality performed, 36 patients had CT imaging only, 7 patients had MRI imaging only and 14 patients had dual modailty imaging with CT and MRI. All MRI studies were carried out to clarify positive new findings
initially detected on CT. In total, 77 imaging studies were performed with several patients undergoing repeat imaging studies to monitor evolution or resolution of pathology identified. The indication for, timing and outcome in those who had significant acute findings on neuroimaging is presented in Table 2. The primary indications for neuroimaging included: neurological findings on primary survey (20/56); suspicion for head injury due to mechanism of burn injury (8/56); suspicion of hypoxic brain injury at presentation (1/56), dropping GCS (4/56), acute confusion (8/56), seizures (5/56), new focal neurological signs (5/56) and where a patient was slow to wean off mechanical ventilation (5/56) without any other obvious cause. Indications for neuroimaging in patients with positive new findings may be sub-divided into two time checkpoints (Table 3); early indications within 48 h of admission and later indications arising greater than 48 h after admission. Out of 56 patients who underwent neuroimaging, 46 patients had positive findings identified (82.1% of neuroimaged cohort, 3.5% of overall cohort), subdivided into significant acute findings and chronic findings below. A total of 14 out of 56 neuroimaged patients (25%) had a significant acute findings on neuroimaging (Fig. 1). This represents approximately 1% (14/1328) of all inpatient admissions during the study period. Seven out of 56 neuroimaged patients (12.5%) had acute cerebral infarctions, representing 0.5% (7/1328) of the overall cohort of burn patient admissions. One of these patients had extensive acute infarction secondary to carotid artery dissection. Three patients had diffuse hypoxic injury, a further three had contusions with intracranial haemorrhage, one patient had a new diagnosis of Huntington’s chorea and another had a new diagnosis of primary brain tumour (high grade glioblastoma multiforme — treated with palliative radiotherapy, as medical comorbidity precluded surgical intervention.). All positive findings were discussed with local neurosurgical services for further management but no acute neurosurgical interventions were ultimately carried out. Burns requiring formal intravenous fluid resuscitation are defined as >15% TBSA or <10% TBSA in >65 year olds [10] in our centre. Patients with resuscitation burns (25/215 vs. 31/1113 patients, p < 0.0001, Chi-squared test) or associated inhalation injury (17/147 vs. 38/1181 patients, p < 0.0001, Chi-squared test) were significantly more likely to undergo neuroimaging
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
4
burns xxx (2019) xxx
xxx
Table 2 – Acute Neuroimaging findings. Indication
Finding
Early: Primary survey — Explosion, reduced GCS
Extensive acute infarction of left cerebral hemisphere from traumatic carotid dissection with vasogenic oedema. Diffuse hypoxic brain injury. Diffuse hypoxic brain injury with occipital cortical necrosis. Multiple contusions, extradural haematoma and temporo-occipital parietal fracture with acute cortical infarction. Acute cortical infraction, subdural and subarachnoid haemorrhage. Multiple contusions and intraparenchymal haemorrhage. Acute medullary infarction. Acute posterior cerebellar artery infarction. Primary brain tumour — High grade Glioblastoma Multiforme Acute occipital infarct, old thalamic infarct. Small acute cortical infarct of anterior parietal lobe. Hypoxic brain injury with severe white matter disease. Acute cortical parietal infarct. Huntington’s disease, focal atrophy of bilateral caudate nuclei.
Early: Primary survey — Head injury, reduced GCS Early: Primary survey — Electrocution, reduced GCS Early: Primary survey — Explosion, reduced GCS Early: Primary survey — Explosion, reduced GCS Early: Primary survey — Reduced GCS Early: Primary survey — Reduced GCS Early: Acute confusion Early: Acute confusion Late: Slow to wean off mechanical ventilation Late: Acute Confusion Late: Acute Confusion & hyponatremia Late: New unilateral facial weakness Late: Headache and dizziness Early: Indication arising <48 h of admission. Late: Indication arising >48 h from admission. GCS: Glasgow Coma Scale.
compared to those who did not (Table 3). Positive findings were detected in all neuroimaging studies carried out for patients with resuscitation burns (25/25) and in all patients neuroimaged patients with associated inhalation injury (17/17). Chronic findings were identified in 32 of the 56 neuroimaged patients (57.1%), representing 2.4% (32/1328) of the overall cohort of burn patient admissions. Of these, 11 patients out of 56 neuroimaged (19.6%) had evidence of old established cerebral infarctions. A further 12 patients (21.4%) had findings of chronic microvascular ischaemia and 6 others (10.7%) had cerebral atrophy. Three patients had features of previous traumatic brain injury including craniotomy (1), skull fracture (1) and gliosis (1). The mean Total Fazekas score (Fig. 2) of studies performed in all 56 patients was 3.59 (Range 0 = Normal, Severe white matter disease = 3 6), suggesting a high overall rate of white matter disease. This is likely a reflection of the high rate of chronic cerebrovascular ischaemia detected. The Total Fazekas score in patients with positive new findings on neuroimaging was lower than that of the overall cohort at 1.9 (Range 0 5), suggesting a lower prevalence of white matter disease among patients diagnosed with acute neuropathology. Only ten out of 56 neuroimaged patients (17.9%) had entirely normal imaging studies.
One half of all patients with a significant new finding on neuroimaging required additional inpatient rehabilitation patient after discharge from the National Burns Centre. Four patients out of 14 (28.5%) were transferred to other Acute Hospitals closer to the patients home address to continue general inpatient rehabilitation. Two patients required transfer to a National Rehabilitation Hospital for specialist neurological rehabilitation and one was discharged to a residential facility with long-term nursing care. Of the 32 patients who had chronic findings on neuroimaging, 18/32 (56.2%) required transfer to another acute Hospital for further inpatient rehabilitation, long-term nursing care placement or increased home support services. Of the 10 patients who had normal neuroimaging, only 1 patient required transfer to another hospital for further rehabilitation, while the remaining 9 were discharged directly home from the National Burns Unit, independent in their activities of daily living. The median length of stay (LOS) in the overall cohort of 1328 patients was 19.5 days (Range 3 162 days). Patients with positive findings on neuroimaging (median 47 day LOS, SD 57.1) had a significantly increased length of stay compared to
Table 3 – Neuroimaging findings in burn patients requiring formal fluid resuscitation and burn patients with inhalation injury.
Total no. Neuroimaged no. No. with positive findings (Acute and Chronic) Positive findings as % of Total no. Positive findings as % of Neuroimaged no. Acute findings (% of Neuroimaged no.) Chronic findings (% of Neuroimaged no.)
Non-Resus (<15% TBSA)
Fluid Resus (>15% TBSA)
No Inhalation injury
Inhalation injury
1113 31 21 1.8% 67.7% 4 (12.9%) 17 (54.8%)
215 25 25 11.6% 100% 10 (40%) 15 (60%)
1181 38 28 2.4% 73.6% 6 (15.8%) 22 (57.8%)
147 17 17 11.6% 100% 7 (41.2%) 10 (58.8%)
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
xxx
5
Fig. 1 – Acute radiological findings. a) 20 year old man with hypoxic ischaemic injury following burn from high voltage electrical injury. Symmetric MRI FLAIR hyperintensity in the cerebral cortex and in deep grey nuclei for example the head of the caudate (arrow) with effacement of the sulci related to gyral oedema (multiple arrows) signifying severe hypoxic ischaemic injury pattern. b) 68 year old man with acute infarction following 60% TBSA burn from gas explosion. MRI FLAIR showing extensive high signal abnormality in the corpus callosum (stars). Bilateral cerebral hemispheres (large arrows), demonstrate marked high signal abnormality with sulcal effacement (small arrows) consistent with extensive acute infarct involving both cerebral hemispheres from traumatic carotid artery dissection. c) 74 year old man with primary brain tumour diagnosed after 20% scald burn to back. MRI Flair demonstrating moderately hyperintense mass (multiple arrows) within the periphery of the right frontal lobe with striking hyperintense signal throughout the frontal lobe in keeping with oedema. Associated effacement of the right lateral ventricle and subfalcine herniation. d) 40 year old man diagnosed with Huntington’s Disease after 1% scald. Non contrast CT brain demonstrating atrophy of both caudate nuclei (arrows) with compensatory enlargement of the frontal horns of the lateral ventricles (stars), giving a box like appearance, characteristic of Huntington’s disease.
those with normal imaging (median 27.5 day LOS, SD 24.7) with p < 0.027 (Kruskall Wallis test). Predicted and observed mortality is outlined in Table 1. The neuroimaged patient cohort had a higher mean revised Baux
score compared to patients who did not undergo neuroimaging although observed mortality was not significantly different between these groups. We had no complications as a result of transporting patients for neuro imaging except for
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
6
burns xxx (2019) xxx
xxx
Fig. 2 – MRI Flair images demonstrating Fazekas scale application. a) Normal brain parenchyma. No periventricular or deep white matter hyperintense lesions. b) Periventricular hyperintense smooth halo (large arrow) with punctate deep white matter hyperintense lesions (small arrows). c) Irregular periventricular hyperintense signal extending into the deep white matter (small arrows) with the confluence of deep white matter lesions (large arrow).
cancellation of one MRI study due to patient agitation on transfer to the radiology department.
4.
Discussion
It is often difficult to assess the neurological status of burn injured patients due to a wide range of factors. This brings about challenges in defining the indications for neuroimaging following burn injury that are not encompassed by current guidance relating to neuroimaging following head trauma [6]. A table of additional indications for neuroimaging specific to burns patients is proposed in Table 4, informed by the results of the present study. A variety of indications for neuroimaging can arise at two general time checkpoints during burn unit admission, to which burn care teams should remain alert. Common early indications (within 48 h of admission) include primary survey, where indicated due to burn injury mechanism, history or examination findings such as a suspicion for hypoxic brain injury or primary inciting neurological event such as cerebrovascular accident. Later indications (greater
than 48 h after admission) may arise in the setting of deteriorating neurological function, focal neurological signs, acute confusion, new seizure activity and where a patient is slower than expected to wake on weaning from mechanical ventilation without another identifiable cause. Hypoxic brain injury is an indication for both early and repeat scanning before and after 48 h. Early baseline imaging usually demonstrates non-specific findings of diffuse cerebral oedema and repeat imaging is warranted to confirm or refute this diagnosis. Burn injury mechanisms associated with increased risk of positive neuroimaging findings include explosion or electrical injury, present in 4/14 of patients with positive neuroimaging findings, as outlined in Table 2. The rate of positive neuroimaging findings in burn patients is unclear. One previous study directly investigated the rate of positive neuroimaging findings in burns patients [9], whereby 40 CT brain images from a cohort of 7000 burn inpatients over a 10 year period were retrospectively reviewed by an interventional radiologist. Common indications for neuroimaging in this study included positive findings on primary survey and suspected hypoxic brain injury. Just two cases of acute
Table 4 – Expanded indications for neuroimaging in adults after burn injury. Early indications: Within 48 h of injury Suspicion of concurrent head injury/indications as per head injury guidelines [5,6] Suspicion of primary inciting neurological event (e.g.: Cerebrovascular Accident) Mechanism of injury (e.g.; Explosion, Electrical injury, Fall from height) Suspicion of hypoxic brain injury (Baseline imaging) Late indications: After 48 h from time of injury Deteriorating neurological function (e.g.: Drop in GCS, new focal neurological deficit, new onset seizure activity, new delirium/confusion) Slow to wake on weaning from mechanical ventilation, without other obvious cause Suspicion of hypoxic brain injury (Repeat imaging to confirm diagnosis) High risk cohort: Resuscitation burns (>15% TBSA in adults) and burns with inhalational injury are associated with high rates of positive neuroimaging findings, when indicated
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
intracranial haemorrhage were identified in this study, one of whom required neurosurgical intervention. All neuroimaging studies carried out for the indication of suspected hypoxic brain injury were normal. The authors of this study concluded that caution should be exercised in deciding to transport intubated intensive care patients to radiology departments, when the results are likely to be normal. Even with intrahospital transfer guidelines in place, studies have reported one third of intensive care patients experienced adverse events on transfer to the radiology department [10]. Life threatening events have been reported in 8% of transfers, such as loss of intravenous access for inotropic support or endotracheal tube dislodgement. The findings of the present study suggest that abnormalities on neuroimaging are frequently detected in burn patients, contrary to the aforementioned study. The comparatively higher rate of positive neuroimaging findings in our study may be related to advances in neuroimaging technology over time, our inclusion of MRI studies in addition to CT studies and our wider definition of ‘positive neuroimaging findings’ to include additional neuropathology findings other than those requiring neurosurgical intervention. Neuroimaging findings can exert direct influence on burn patient care and may predict outcome and rehabilitation needs. While only a small proportion of the overall cohort of burn patients (56/1328 or 4.2%) had a clinical indication for neuroimaging, one in four of these patients were found to have positive new finding on radiology. This high frequency of positive radiological findings suggests that neuroimaging may be underutilised in burn patient cohorts, further supporting the expansion of imaging indications outlined in Table 4. The most common new findings were acute cerebrovascular events followed by hypoxic brain injury and intracranial haemorrhage. Positive findings were identified in all neuroimaging studies carried out for patients with burns requiring formal fluid resuscitation (25/25) and inhalation injury (18/18), although this finding may be reflective of more severe, larger TBSA burn injury. Major burn injuries (>35% TBSA) have been shown to result in increased release of neurotoxic high-molecular weight lipoproteins that can lead to encephalopathy and peripheral neuropathy [14]. Cerebral dysfunction may result from inhalation injury or burn wound sepsis via processes including cerebral perfusion impairment, altered neurotransmission and excessive microglial activation. This is supported by changes in neuroimaging, such as cytotoxic vasogenic oedema, white matter disruption and cerebral atrophy. Furthermore, there were no adverse incidents associated with neuroimaging in this study. Therefore, physicians should have a low threshold for proceeding with neuroimaging in burn patients where there is an appropriate indication. Acute (n = 7) and chronic cerebral infarction (n = 11) was seen in 18 of 56 (32.1%) neuroimaged burn patients in this study cohort. Chronic microvascular ischaemia was identified in 12 patients (21.4%) and 6 patients (10.7%) demonstrated cerebral atrophy. This high prevalence of cerebrovascular disease patients may reflect acute primary neurological inciting events which indirectly led to burn injury or a chronic disease process such as vascular dementia that can also predispose elderly patients to burns through deleterious effects on situational awareness, cognition, gait and mobility [15].
xxx
7
Previous study has shown that people with dementia have higher age-standardised incidence rates for burn injury than those without (22.7 per 100,000 vs. 14.2 per 100,000 population) equating to an incidence rate ratio of 1.6 (95% CI 1.3 2.0, p < 0.0001) [16]. Mean LOS was more than double (24 vs. 12 days) for people with dementia in the aforementioned study [16]. This is in keeping with the results of our cohort showing increased LOS in those with positive acute and chronic findings on neuroimaging (47 vs. 27 days). In addition to prolonging length of stay, cerebrovascular disease has been shown (in another large study of 31,338 burn patients) to significantly predict in-hospital mortality (OR 1.6) [17]. White matter disease is another common form of cerebrovascular disease and represents a risk factor for poor neurological outcomes [18]. Clinical symptoms related to white matter injury include cognitive and emotional dysfunction, sensorimotor impairment, urinary incontinence and pain. White matter disease was prevalent in this cohort of burn patients as indicated by the mean total Fazekas score (mean total score of 3.59 in neuroimaged burn patients). This score was calculated on review of all imaging studies in this study by a Consultant physician with a special interest in neurovascular pathology (JH). Fazekas scores were validated separately by a Consultant Radiologist, with whom there was 100% interobserver agreement. The Fazekas scale is a scoring system used to quantify the amount of hyper intense lesions identified on neuroimaging studies. This scale assigns scores to periventricular white matter disease, relating to a combination demyelination and gliosis (non-ischaemic changes) and deep white matter changes attributed to chronic small vessel ischaemia [19]. The deep white matter component of the Fazekas score is useful in the assessment of patients with possible dementia. Fazekas Grade 0 represents normal white matter while Grade 1 (mild white matter hyperintensities) is considered normal in elderly populations. Fazekas Grades 2 (moderate white matter disease) and 3 (severe white matter disease) are considered pathologic and although they may be seen in normally functioning individuals, they correlate with a high risk for disability, stroke and dementia [19,20]. Previous study of 600 normally functioning elderly people demonstrated that the Fazekas score predicted disability within one year [20]. Long term follow-up studies show that severe white matter changes independently and strongly predict rapid global functional decline, long term care and rehabilitation needs [21]. White matter disease therefore carries implications for service planning within acute care burn units, although more information on the natural course of vascular lesions and white matter changes is needed to directly inform patient management in terms of secondary prevention and therapeutic measures. Features of white matter disease may be subtle and review at a neuroimaging multidisciplinary team meeting with a specialist in neuropathology may be beneficial. There were two incidental but clinically significant new findings in this study: primary brain tumour and Huntington’s Chorea. Incidental findings on neuroimaging have been reported to occur with a frequency of 1.7 2.7% for MRI brain studies carried out in the general population [22,23] and 1% for CT brain studies [24] carried out following head trauma. The most common incidental CT brain findings in normal
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
8
burns xxx (2019) xxx
populations include cisterna magna enlargement followed by brain tumour and arachnoid cyst. Incidental MRI brain findings are common in the general population and include subclinical vascular pathologic changes, cerebral aneurysm and benign primary tumour [23]. The incidence of these findings increases with age. Mortality in patients with major burns as a direct result of cerebral lesions is rare but has been reported in one case of rapidly progressive glioma (not present on initial admission neuroimaging studies) [24]. Burn patients with neurological symptoms and signs should be investigated thoroughly to evaluate for the presence of pathology which may be associated with reduced survival, particularly in the setting of immunocompromise. In a military setting, predominated by ballistic trauma, 10% of burn patients sustained concomitant traumatic brain injury [25]. Traumatic brain injury was found to independently increase the risk of many complications (length of ICU and hospital stay, shock, amputation, wound sepsis and thromboembolism) when sustained with burns, although it did not independently increase mortality. Traumatic brain injury was seen in 6 patients or 10.7% of patients neuroimaged or 0.5% of our overall study cohort (3 diffuse hypoxic injury, 3 contusions with intracranial haemorrhage). Predicted (but not observed) mortality was higher in the neuroimaged cohort in the present study, although this reflect the higher mean TBSA burn in the former cohort. There is a paucity of research in civilian populations investigating intra-cranial injury in burn injury in terms of its incidence or potential to increase complications such as prolonged rehabilitation, delayed weaning off ventilation or increase length of stay. The results of this study indicate positive neuroimaging findings in civilian burn patients are not infrequent and predict prolonged length of stay. Neuroimaging also has discriminatory value in predicting rehabilitation requirements and patient discharge planning with approximately half of patients with acute (7/14) and chronic (18/32) neuroimaging findings requiring transfer to another inpatient facility after discharge from the burns unit for further inpatient rehabilitation. This has implications for burn patient prognosis and rehabilitation service planning in burn units. Future prospective study of neuroimaged burns patients in our unit will aim to further develop the general indications, outlined in Table 4, into specific recommendations or thresholds for neuroimaging. In particular, further research is warranted to investigate the relationship between TBSA, inhalation injury and positive neuroimaging findings, particularly given the logistical hazards of radiology department transfer for artificially ventilated patients with large burns.
5.
Conclusion
This study demonstrates the value of neuroimaging, when appropriate, in selected burn injured patients. Positive acute and chronic radiological findings have the potential to impact patient care, rehabilitation and outcome. A cohort of burn patients with increased risk of intra-cranial pathology have been identified, specifically those with larger burns undergoing formal intravenous fluid resuscitation and in those with inhalation injuries.
xxx
A high prevalence of acute and chronic cerebrovascular disease was detected in burns patients who underwent neuroimaging. This may reflect pre-existing neurodegeneration, a known risk factor for burn injury through impairment of situational awareness, cognition and gait. Cerebrovascular disease may also predict poorer outcomes such as increased length of stay and mortality. The clinical significance of white matter disease prevalent in burn patients, as indicated by the Fazekas score, is not clearly understood and warrants further study. Expanded indications for neuroimaging for patients following burn injury may be warranted and could be incorporated into any trauma systems dealing with acute burn injuries. Further study in large volume burn centres may be worthwhile to determine the true incidence, pathogenesis and clinical significance of positive neuroimaging findings specific to burn injured patients.
Author contributions E. Concannon: Study design, data analysis, interpretation, writing and critical revision. L. Fitzgerald: Data collection and writing. J. Birrane: Data analysis. E. Canniff: Data collection. J. Harbison: Data collection, analysis, interpretation. O. Shelley: Study design, data interpretation and critical review.
Conflict of interest None.
Funding None. REFERENCES
[1] Stylianou N, Buchan I, Dunn KW. A review of the international Burn Injury Database (iBID) for England and Wales: descriptive analysis of burn injuries 2003 2011. BMJ Open 2015;5 (February):e006184. [2] Brusselaers N, Monstrey S, Vogelaers D, Hoste E, Blot S. Severe burn aetiology in Europe: a systematic review of the incidence, aetiology, morbidity and mortality. Crit Care 2010;14(October (5)):R188. [3] Peck MD. Epidemiology of burns throughout the world. Part I: distribution and risk factors. Burns 2011;37(November (7)):1087 100. [4] Smolle C, Cambiaso-Daniel J, Forbes AA, Wurzer P, Hundeshagen G, Branski LK, et al. Recent trends in burn epidemiology worldwide: a systematic review. Burns 2017;43 (March (2)):249 57. [5] Douglas DB, Muldermans JL, Wintermark M. Neuroimaging of brain trauma. Curr Opin Neurol 2018;31(August (4)):362 70. [6] National Institute for Clinical Excellence (NICE). Clinical guideline 176: Head injury assessment and early management. Available at: https://www.nice.org.uk/guidan ce/cg176. [Accessed 1 May 2019].
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
[7] Alden NE, Rabbitts A, Yurt RW. Burn injury in patients with dementia: an impetus for prevention. J Burn Care Res 2005;26:267 71. [8] Fauerbach JA, Lawrence J, Haythornthwaite J, Richter D, McGuire M, Schmidt C, et al. Pre-burn psychiatric history affects post-trauma morbidity. Psychosomatics 1997;38(July (4)):374 85. [9] Branhall RJ, Badran M, Shelley OP, Philip B, Dziewulski P. CT brain scans in burn patients — a worthwhile investigation? Burns 2009;35(September (S1)):53. [10] Ott LK, Hoffman LA, Hravnak M. Intrahospital transport to the radiology department: risk for adverse events, nursing surveillance, utilization of a MET and practice implications. J Radiol Nurs 2011;30(June (2)):49 52. [11] Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. Am J Roentgenol 1987;149(August (2)):351 6. [12] Osler T, Glance L, Hosmer D. Simplified estimates of the probability of death after burn injuries: extending and updating the baux score. J Trauma 2010;68(March (3)):690 7. [13] Woods JF, Quinlan CS, Shelley OP. Predicting mortality in severe burns-what is the score? Evaluation and comparison of 4 mortality prediction scores in an Irish population. Plast Reconstr Surg Glob Open 2016;4(February (1)):e606. [14] Obeid DA, Abdulaziz K, Alhujayri AK, Aldekhayel S. Burn induced neuroepithelial changes as a delayed cause of mortality in major burns: a case report and literature review. Int J Burn Trauma 2018;8(December (6)):145 8. [15] Esechie A, Bhardwaj A, Masel T, Raji M. Neurocognitive sequela of burn injury in the elderly. J Clin Neurosci 2019;59 (January):1 5. [16] Harvey L, Mitchell R, Brodaty H, Draper B, Close J. Dementia: a risk factor for burns in the elderly. Burns 2016;42(March (2)):282 90.
xxx
9
[17] Thombs BD, Singh VA, Halonen J, Diallo A, Milner SM. The effects of preexisting medical comorbidities on mortality and length of hospital stay in acute burn injury: evidence from a national sample of 31,338 adult patients. Ann Surg 2007;245 (April (4)):629 34. [18] Wang Y, Liu G, Hong D, Chen F, Ji X, Cao G. White matter injury in ischemic stroke. Prog Neurobiol 2016;141(June):45 60. [19] Kim KW, MacFall JR, Payne ME. Classification of white matter lesions on magnetic resonance imaging in elderly persons. Biol Psychiatry 2008;64(August (4)):273 80. [20] Inzitari D, Simoni M, Pracucci G, Poggesi A, Basile AM, Chabriat H, et al. Risk of rapid global functional decline in elderly patients with severe cerebral age-related white matter changes: the LADIS study.; LADIS Study Group. Arch Intern Med 2007;167(January (1)):81 8. [21] Inzitari D, Pracucci G, Poggesi A, Carlucci G, Barkhof F, Chabriat H, et al. Changes in white matter as determinant of global functional decline in older independant outpatients: three year follow-up of LADIS (leukoaraiosis and disability) study cohort. LADIS Study Group. BMJ 2009;339(July): b2477. [22] Bunnik EM, Vernooij MW. Incidental findings in population imaging revisited. Eur J Epidemiol 2016;31(January (1)):1 4. [23] Vernooij MW, Ikram MA, Tanghe HL, Vincent AJ, Hofman A, Krestin GP, et al. Incidental findings on brain MRI in the general population. N Engl J Med 2007;357(November (18)): 1821 8. [24] Eskandary H, Sabba M, Khajehpour F, Eskandari M. Incidental findings in brain computed tomography scans of 3000 head trauma patients. Surg Neurol 2005;63(June (6)): 550 3. [25] Graybill JC, Aden J, Rizzo JA, Clemens M, Janak J, White C, et al. Concomittant traumatic brain injury and burns in combat trauma. J Am Coll Surg 2016;223(October (4)):e217.
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
burns xxx (2019) xxx
xxx
ScienceDirect journal homepage: www.elsevier.com/locate/burns
Neuroimaging provides relevant clinical information in patients with burn injuries
,
$ $$
Elizabeth Concannon a, * , Louise Fitzgerald a, Emma Canniff b, John Birrane a , Joseph Harbison c , Odhran Shelley a a
Department of Plastic and Reconstructive Surgery, St James’s Hospital, Dublin, Ireland Department of Radiology, St James’s Hospital, Dublin, Ireland c Department of Medicine for the Elderly, St James’s Hospital, Dublin, Ireland b
article info
abstract
Article history:
Introduction: Neurological assessment of patients with burn injuries may be complicated by a
Accepted 29 August 2019
variety of factors including artificial ventilation and sedation, cerebral hypoxia and
Available online xxx
intoxication. Medically unstable intubated patients present logistical challenges for radiological imaging. The role of neuroimaging as an adjunct to clinical assessment of
Keywords: Burn injury Neuroimaging Fazekas score Brain injury Resuscitation Cerebrovascular disease White matter disease Rehabilitation
burn injured patients has not yet been determined. Aim: This study aims to investigate the indications, findings and outcomes of neuroimaging studies performed for burn injured patients. Methods: A retrospective case series study of adult burn patients admitted over an 8 year period was completed in the National Burns Centre at St James’s Hospital, Dublin. Neuroimaging studies carried out for patients admitted during the study period were reviewed by a Consultant Radiologist and Consultant Stroke Physician. Outcomes included neuroimaging findings, prevalence of white matter disease (Fazekas scale), length of stay, discharge destination, predicted and observed mortality. Results: 1328 consecutive patients with burn injuries were admitted during the study period. 56 patients underwent neuroimaging studies with computerised tomography, magnetic resonance imaging or both. 46 out of 56 neuroimaged patients (82.1%) had significant radiological findings, including 14 patients (25%) with acute findings. There was a high prevalence of white matter disease (mean total Fazekas score: 3.59) and acute cerebral infarction (7 patients). Patients with radiological findings required additional in-patient rehabilitation and had increased length of stay (Median 47.0 days vs. 27.5 days, p < 0.027). Patients with resuscitation burns or associated inhalation injury were significantly more likely to undergo neuroimaging (p < 0.0001) and to have positive radiological findings. Predicted mortality was higher in patients with positive neuroimaging findings compared to patients with normal neuroimaging studies, although there was no significant difference in observed mortality between these two groups.
$
This is the largest study to date investigating the role of neuroimaging in civilian burn patients. Neuroimaging provides valuable clinical information, when indicated, detecting findings that can exert direct influence on burn patient care, influence outcomes and predict rehabilitation needs. Won prize for Best Poster Presentation at the annual British Burns Association meeting May 2019, Leeds. Presented at the Irish Association of Plastic Surgeons Meeting, Cork, Ireland, May 2019 (Oral). * Corresponding author at: National Burns Unit, St James’s Hospital, James Street, Dublin 8, Ireland. E-mail addresses: [email protected] (E. Concannon), [email protected] (L. Fitzgerald), [email protected] (E. Canniff), [email protected] (J. Birrane), [email protected] (J. Harbison), [email protected] (O. Shelley). https://doi.org/10.1016/j.burns.2019.08.018 0305-4179/© 2019 Elsevier Ltd and ISBI. All rights reserved. $$
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
2
burns xxx (2019) xxx
xxx
Conclusion: Neuroimaging is used appropriately in patients admitted with burns and provides valuable applicable clinical information when indicated. © 2019 Elsevier Ltd and ISBI. All rights reserved.
1.
Background
Each year approximately 13,000 burn injuries which require hospital attention occur in the UK with 10% of those hospitalised suffering a life threatening burn [1]. Across Europe, between 4 and 22% are admitted to intensive care from presentation [2]. In 2004, nearly 11 million people worldwide were burned severely enough to require medical attention [3]. Burns, therefore, represent the fourth most common mechanism of all injuries after road traffic accidents, falls and interpersonal violence. Fortunately, an ongoing global reduction in burn related mortality has been demonstrated due to advances in the knowledge of burn pathophysiology and multidisciplinary critical care provided by specialised burn centres [4]. Neuroimaging represents a radiological adjunct to the secondary survey, when indicated, in Advanced Trauma Life Support1 management of trauma patients. However the role of neuroimaging in assessment and management of burns has not been specifically assessed. Burns are among the leading causes of disability-adjusted life-years lost in low- and middle-income countries. The cost of rehabilitation, time off work and loss of earnings represent a substantial cost to the community following acute hospital treatment [5]. This is increased by the presence of concomitant intra cranial pathology. Intra cranial injury, secondary to an associated trauma or hypoxic brain injury as a result of burn injury has the potential to impair and delay rehabilitation, however no previous research has specifically addressed this. Several criteria have been developed to guide indications for neuroimaging following head injury including New Orleans Criteria, Canadian CT Head Rules or National Emergency X-Ray Utilization Study-II clinical criteria [5]. Head injury assessment guidelines from the National Institute for Health and Care Excellence (NICE) govern practice in the United Kingdom and Ireland [6]. Computerised tomography (CT) brain studies are generally indicated following head injury if criteria for risk of neurosurgical intervention are present. No specific radiological criteria are available to guide neuroimaging of patients following burn injury with or without head injury. Neurological assessment of patients with burn injury may be confounded by a multitude of factors. Cerebral hypoxia due to significant inhalation injury, carbon monoxide toxicity or pulmonary oedema may indirectly alter cognitive function. Early endotracheal intubation of burn patients may be a lifesaving manoeuvre but represents an obvious barrier to neurological assessment as is opiate analgesia, to which elderly patients may be particularly sensitive. Burn patients have high rates of psychiatric comorbidity, including dementia [7], alcohol and polysubstance dependence [8] compared to the general population, which may complicate neurological observation and obscure the diagnosis of acute pathology. Haemodynamic instability may also cloud clinical examination. Furthermore, mechanisms of burn injury and their
interpretation may vary. For example, an explosion could be mistakenly interpreted as a flash flame burn. Head injury could also be sustained during rapid extrication from a fire. Primary neurological events such as acute stroke or seizure activity may also incite burn injury and a detailed history of events is not always available. CT brain undertaken to assess neurological function in these scenarios aims to investigate the possibility of intracranial pathology. An expansion in the current head injury assessment guidelines [5,6] is warranted to account for the wide range of indications for neuroimaging that may be present following burn injury. There is a paucity of data pertaining to the utility of neuroimaging in burn patients. The results of one retrospective study of burn inpatients who underwent CT brain imaging [9] demonstrated very few positive radiological findings and concluded that caution should be exercised when deciding to transport intubated, potentially unstable critical care patients to radiology departments, given the inherent risk of adverse events [10]. This study aims to assess the indications for neuroimaging in burn patients and whether their findings and outcomes provided useful clinical information.
2.
Methods
A retrospective review of all burns inpatients over the 8 year study period, from January 2008 to January 2016 was performed in the National Burns Centre at St James Hospital, to which all major burns in Ireland are referred. This quaternary referral centre provides a local, regional and national burns service to a population catchment of 4.7 million. Data was sourced from the HIPE system (Hospital In-Patient Enquiry), medical charts, electronic patients records and burns admission data. Burns data is collected electronically and coded prospectively within the National Burns Unit on patient admission using these databases. Approval for the study was granted by local institutional review board. As this observational research was carried out retrospectively with subjects who had received emergency treatment, a waiver of regulatory requirements for obtaining informed consent applies. All patients who underwent neuroimaging during their admission for management of burns were included. Data collection included age, gender, length of stay, total body surface area sustained (TBSA), burn mechanism, inhalation injury, indication for neuroimaging and modality of neuroimaging performed, neuroimaging findings and patient outcomes, including mortality, neurosurgical intervention and discharge destination. Imaging studies were reviewed by a Consultant Radiologist and by a Consultant Stroke physician with a special interest in neurovascular pathology (co-author JH). There were no new or additional radiological findings identified retrospectively when compared to the initial Consultant Radiologist report,
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
3
xxx
Table 1 – Demographics of burn patients admitted between 2008 and 2016.
Number Mean age Mean TBSA Inhalation injury Formal fluid resuscitation Median LOS Observed mortality Mean revised Baux score
a b
All burn patients
Neuroimaged burn patients
Positive radiological findings
No radiological findings
1328 54.8 5.6% 147 (11.0%) 215 19.5 39 (2.9%) 63 (2% mortality)b
56 57.5 14.8% 17 (30.3%) 25 41.5 2 (3.6%) 77 (5 10% mortality)b
46 63.5 15.8% 17 25 47.0a 2 (3.6%) 80 (5 10% mortality)b
10 51.1 11.4% 0 0 27.5a 0 66 (2 3% mortality)b
TBSA: Total Body Surface Area burned. LOS: Length of stay. Significant difference in LOS with p < 0.027, Kruskal Wallis test. Predicted probability of death using logistic regression model of revised baux score [12].
apart from calculation of a score for severity of white matter disease. The Fazekas scale [11] is a scoring system used to quantify the amount of white matter disease, best identified on T2-weighted FLAIR MRI images. Fazekas grading ranges from 0 in normal white matter, to 3 in severe white matter disease. Periventricular white matter (0 3) and deep (0 3) white matter disease scores were added to generate a Total Fazekas score (0 6), utilised for reporting the results of this study. Predicted mortality was calculated using logistic regression model of the revised Baux score [12]. This model has been validated for mortality prediction following burn injury in Ireland and internationally [13]. A standard statistical software package (SPSS1 v25) was used for data analysis with p-value <0.05 denoting statistical significance.
3.
Results
There were 1328 consecutive patients admitted to the National Burns Unit during the 8 year study period, between January 2008 and January 2016 (Table 1). The overall cohort was comprised of 697 males and 631 females with a mean age of 54.8 (Range 17 94). The mean TBSA burned of the overall cohort was 5.6% (Range 0.5 98%). Inhalation injury was present in 147 patients (11% of overall cohort). Fifty-six burn patients (4.2% of overall cohort of 1328 patients) underwent neuroimaging with CT or MRI. The neuroimaged patient cohort was comprised of 31 males and 25 females with a mean age of 57.5 years (Range 20 86 years). The mean TBSA of neuroimaged patients was 14.8% (Range 0.5 96%). The predominant burn injury mechanism in the neuroimaged patient cohort was thermal injury from flame, contact or scald. Similar burn injury mechanisms were predominant in burn patients who did not undergo neuroimaging studies. Three patients out of 56 (5.4%) sustained electrical injuries and a further three patients had burns secondary to an explosion. Seventeen patients (30.4%) had an associated inhalation injury. In terms of neuroimaging modality performed, 36 patients had CT imaging only, 7 patients had MRI imaging only and 14 patients had dual modailty imaging with CT and MRI. All MRI studies were carried out to clarify positive new findings
initially detected on CT. In total, 77 imaging studies were performed with several patients undergoing repeat imaging studies to monitor evolution or resolution of pathology identified. The indication for, timing and outcome in those who had significant acute findings on neuroimaging is presented in Table 2. The primary indications for neuroimaging included: neurological findings on primary survey (20/56); suspicion for head injury due to mechanism of burn injury (8/56); suspicion of hypoxic brain injury at presentation (1/56), dropping GCS (4/56), acute confusion (8/56), seizures (5/56), new focal neurological signs (5/56) and where a patient was slow to wean off mechanical ventilation (5/56) without any other obvious cause. Indications for neuroimaging in patients with positive new findings may be sub-divided into two time checkpoints (Table 3); early indications within 48 h of admission and later indications arising greater than 48 h after admission. Out of 56 patients who underwent neuroimaging, 46 patients had positive findings identified (82.1% of neuroimaged cohort, 3.5% of overall cohort), subdivided into significant acute findings and chronic findings below. A total of 14 out of 56 neuroimaged patients (25%) had a significant acute findings on neuroimaging (Fig. 1). This represents approximately 1% (14/1328) of all inpatient admissions during the study period. Seven out of 56 neuroimaged patients (12.5%) had acute cerebral infarctions, representing 0.5% (7/1328) of the overall cohort of burn patient admissions. One of these patients had extensive acute infarction secondary to carotid artery dissection. Three patients had diffuse hypoxic injury, a further three had contusions with intracranial haemorrhage, one patient had a new diagnosis of Huntington’s chorea and another had a new diagnosis of primary brain tumour (high grade glioblastoma multiforme — treated with palliative radiotherapy, as medical comorbidity precluded surgical intervention.). All positive findings were discussed with local neurosurgical services for further management but no acute neurosurgical interventions were ultimately carried out. Burns requiring formal intravenous fluid resuscitation are defined as >15% TBSA or <10% TBSA in >65 year olds [10] in our centre. Patients with resuscitation burns (25/215 vs. 31/1113 patients, p < 0.0001, Chi-squared test) or associated inhalation injury (17/147 vs. 38/1181 patients, p < 0.0001, Chi-squared test) were significantly more likely to undergo neuroimaging
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
4
burns xxx (2019) xxx
xxx
Table 2 – Acute Neuroimaging findings. Indication
Finding
Early: Primary survey — Explosion, reduced GCS
Extensive acute infarction of left cerebral hemisphere from traumatic carotid dissection with vasogenic oedema. Diffuse hypoxic brain injury. Diffuse hypoxic brain injury with occipital cortical necrosis. Multiple contusions, extradural haematoma and temporo-occipital parietal fracture with acute cortical infarction. Acute cortical infraction, subdural and subarachnoid haemorrhage. Multiple contusions and intraparenchymal haemorrhage. Acute medullary infarction. Acute posterior cerebellar artery infarction. Primary brain tumour — High grade Glioblastoma Multiforme Acute occipital infarct, old thalamic infarct. Small acute cortical infarct of anterior parietal lobe. Hypoxic brain injury with severe white matter disease. Acute cortical parietal infarct. Huntington’s disease, focal atrophy of bilateral caudate nuclei.
Early: Primary survey — Head injury, reduced GCS Early: Primary survey — Electrocution, reduced GCS Early: Primary survey — Explosion, reduced GCS Early: Primary survey — Explosion, reduced GCS Early: Primary survey — Reduced GCS Early: Primary survey — Reduced GCS Early: Acute confusion Early: Acute confusion Late: Slow to wean off mechanical ventilation Late: Acute Confusion Late: Acute Confusion & hyponatremia Late: New unilateral facial weakness Late: Headache and dizziness Early: Indication arising <48 h of admission. Late: Indication arising >48 h from admission. GCS: Glasgow Coma Scale.
compared to those who did not (Table 3). Positive findings were detected in all neuroimaging studies carried out for patients with resuscitation burns (25/25) and in all patients neuroimaged patients with associated inhalation injury (17/17). Chronic findings were identified in 32 of the 56 neuroimaged patients (57.1%), representing 2.4% (32/1328) of the overall cohort of burn patient admissions. Of these, 11 patients out of 56 neuroimaged (19.6%) had evidence of old established cerebral infarctions. A further 12 patients (21.4%) had findings of chronic microvascular ischaemia and 6 others (10.7%) had cerebral atrophy. Three patients had features of previous traumatic brain injury including craniotomy (1), skull fracture (1) and gliosis (1). The mean Total Fazekas score (Fig. 2) of studies performed in all 56 patients was 3.59 (Range 0 = Normal, Severe white matter disease = 3 6), suggesting a high overall rate of white matter disease. This is likely a reflection of the high rate of chronic cerebrovascular ischaemia detected. The Total Fazekas score in patients with positive new findings on neuroimaging was lower than that of the overall cohort at 1.9 (Range 0 5), suggesting a lower prevalence of white matter disease among patients diagnosed with acute neuropathology. Only ten out of 56 neuroimaged patients (17.9%) had entirely normal imaging studies.
One half of all patients with a significant new finding on neuroimaging required additional inpatient rehabilitation patient after discharge from the National Burns Centre. Four patients out of 14 (28.5%) were transferred to other Acute Hospitals closer to the patients home address to continue general inpatient rehabilitation. Two patients required transfer to a National Rehabilitation Hospital for specialist neurological rehabilitation and one was discharged to a residential facility with long-term nursing care. Of the 32 patients who had chronic findings on neuroimaging, 18/32 (56.2%) required transfer to another acute Hospital for further inpatient rehabilitation, long-term nursing care placement or increased home support services. Of the 10 patients who had normal neuroimaging, only 1 patient required transfer to another hospital for further rehabilitation, while the remaining 9 were discharged directly home from the National Burns Unit, independent in their activities of daily living. The median length of stay (LOS) in the overall cohort of 1328 patients was 19.5 days (Range 3 162 days). Patients with positive findings on neuroimaging (median 47 day LOS, SD 57.1) had a significantly increased length of stay compared to
Table 3 – Neuroimaging findings in burn patients requiring formal fluid resuscitation and burn patients with inhalation injury.
Total no. Neuroimaged no. No. with positive findings (Acute and Chronic) Positive findings as % of Total no. Positive findings as % of Neuroimaged no. Acute findings (% of Neuroimaged no.) Chronic findings (% of Neuroimaged no.)
Non-Resus (<15% TBSA)
Fluid Resus (>15% TBSA)
No Inhalation injury
Inhalation injury
1113 31 21 1.8% 67.7% 4 (12.9%) 17 (54.8%)
215 25 25 11.6% 100% 10 (40%) 15 (60%)
1181 38 28 2.4% 73.6% 6 (15.8%) 22 (57.8%)
147 17 17 11.6% 100% 7 (41.2%) 10 (58.8%)
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
xxx
5
Fig. 1 – Acute radiological findings. a) 20 year old man with hypoxic ischaemic injury following burn from high voltage electrical injury. Symmetric MRI FLAIR hyperintensity in the cerebral cortex and in deep grey nuclei for example the head of the caudate (arrow) with effacement of the sulci related to gyral oedema (multiple arrows) signifying severe hypoxic ischaemic injury pattern. b) 68 year old man with acute infarction following 60% TBSA burn from gas explosion. MRI FLAIR showing extensive high signal abnormality in the corpus callosum (stars). Bilateral cerebral hemispheres (large arrows), demonstrate marked high signal abnormality with sulcal effacement (small arrows) consistent with extensive acute infarct involving both cerebral hemispheres from traumatic carotid artery dissection. c) 74 year old man with primary brain tumour diagnosed after 20% scald burn to back. MRI Flair demonstrating moderately hyperintense mass (multiple arrows) within the periphery of the right frontal lobe with striking hyperintense signal throughout the frontal lobe in keeping with oedema. Associated effacement of the right lateral ventricle and subfalcine herniation. d) 40 year old man diagnosed with Huntington’s Disease after 1% scald. Non contrast CT brain demonstrating atrophy of both caudate nuclei (arrows) with compensatory enlargement of the frontal horns of the lateral ventricles (stars), giving a box like appearance, characteristic of Huntington’s disease.
those with normal imaging (median 27.5 day LOS, SD 24.7) with p < 0.027 (Kruskall Wallis test). Predicted and observed mortality is outlined in Table 1. The neuroimaged patient cohort had a higher mean revised Baux
score compared to patients who did not undergo neuroimaging although observed mortality was not significantly different between these groups. We had no complications as a result of transporting patients for neuro imaging except for
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
6
burns xxx (2019) xxx
xxx
Fig. 2 – MRI Flair images demonstrating Fazekas scale application. a) Normal brain parenchyma. No periventricular or deep white matter hyperintense lesions. b) Periventricular hyperintense smooth halo (large arrow) with punctate deep white matter hyperintense lesions (small arrows). c) Irregular periventricular hyperintense signal extending into the deep white matter (small arrows) with the confluence of deep white matter lesions (large arrow).
cancellation of one MRI study due to patient agitation on transfer to the radiology department.
4.
Discussion
It is often difficult to assess the neurological status of burn injured patients due to a wide range of factors. This brings about challenges in defining the indications for neuroimaging following burn injury that are not encompassed by current guidance relating to neuroimaging following head trauma [6]. A table of additional indications for neuroimaging specific to burns patients is proposed in Table 4, informed by the results of the present study. A variety of indications for neuroimaging can arise at two general time checkpoints during burn unit admission, to which burn care teams should remain alert. Common early indications (within 48 h of admission) include primary survey, where indicated due to burn injury mechanism, history or examination findings such as a suspicion for hypoxic brain injury or primary inciting neurological event such as cerebrovascular accident. Later indications (greater
than 48 h after admission) may arise in the setting of deteriorating neurological function, focal neurological signs, acute confusion, new seizure activity and where a patient is slower than expected to wake on weaning from mechanical ventilation without another identifiable cause. Hypoxic brain injury is an indication for both early and repeat scanning before and after 48 h. Early baseline imaging usually demonstrates non-specific findings of diffuse cerebral oedema and repeat imaging is warranted to confirm or refute this diagnosis. Burn injury mechanisms associated with increased risk of positive neuroimaging findings include explosion or electrical injury, present in 4/14 of patients with positive neuroimaging findings, as outlined in Table 2. The rate of positive neuroimaging findings in burn patients is unclear. One previous study directly investigated the rate of positive neuroimaging findings in burns patients [9], whereby 40 CT brain images from a cohort of 7000 burn inpatients over a 10 year period were retrospectively reviewed by an interventional radiologist. Common indications for neuroimaging in this study included positive findings on primary survey and suspected hypoxic brain injury. Just two cases of acute
Table 4 – Expanded indications for neuroimaging in adults after burn injury. Early indications: Within 48 h of injury Suspicion of concurrent head injury/indications as per head injury guidelines [5,6] Suspicion of primary inciting neurological event (e.g.: Cerebrovascular Accident) Mechanism of injury (e.g.; Explosion, Electrical injury, Fall from height) Suspicion of hypoxic brain injury (Baseline imaging) Late indications: After 48 h from time of injury Deteriorating neurological function (e.g.: Drop in GCS, new focal neurological deficit, new onset seizure activity, new delirium/confusion) Slow to wake on weaning from mechanical ventilation, without other obvious cause Suspicion of hypoxic brain injury (Repeat imaging to confirm diagnosis) High risk cohort: Resuscitation burns (>15% TBSA in adults) and burns with inhalational injury are associated with high rates of positive neuroimaging findings, when indicated
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
intracranial haemorrhage were identified in this study, one of whom required neurosurgical intervention. All neuroimaging studies carried out for the indication of suspected hypoxic brain injury were normal. The authors of this study concluded that caution should be exercised in deciding to transport intubated intensive care patients to radiology departments, when the results are likely to be normal. Even with intrahospital transfer guidelines in place, studies have reported one third of intensive care patients experienced adverse events on transfer to the radiology department [10]. Life threatening events have been reported in 8% of transfers, such as loss of intravenous access for inotropic support or endotracheal tube dislodgement. The findings of the present study suggest that abnormalities on neuroimaging are frequently detected in burn patients, contrary to the aforementioned study. The comparatively higher rate of positive neuroimaging findings in our study may be related to advances in neuroimaging technology over time, our inclusion of MRI studies in addition to CT studies and our wider definition of ‘positive neuroimaging findings’ to include additional neuropathology findings other than those requiring neurosurgical intervention. Neuroimaging findings can exert direct influence on burn patient care and may predict outcome and rehabilitation needs. While only a small proportion of the overall cohort of burn patients (56/1328 or 4.2%) had a clinical indication for neuroimaging, one in four of these patients were found to have positive new finding on radiology. This high frequency of positive radiological findings suggests that neuroimaging may be underutilised in burn patient cohorts, further supporting the expansion of imaging indications outlined in Table 4. The most common new findings were acute cerebrovascular events followed by hypoxic brain injury and intracranial haemorrhage. Positive findings were identified in all neuroimaging studies carried out for patients with burns requiring formal fluid resuscitation (25/25) and inhalation injury (18/18), although this finding may be reflective of more severe, larger TBSA burn injury. Major burn injuries (>35% TBSA) have been shown to result in increased release of neurotoxic high-molecular weight lipoproteins that can lead to encephalopathy and peripheral neuropathy [14]. Cerebral dysfunction may result from inhalation injury or burn wound sepsis via processes including cerebral perfusion impairment, altered neurotransmission and excessive microglial activation. This is supported by changes in neuroimaging, such as cytotoxic vasogenic oedema, white matter disruption and cerebral atrophy. Furthermore, there were no adverse incidents associated with neuroimaging in this study. Therefore, physicians should have a low threshold for proceeding with neuroimaging in burn patients where there is an appropriate indication. Acute (n = 7) and chronic cerebral infarction (n = 11) was seen in 18 of 56 (32.1%) neuroimaged burn patients in this study cohort. Chronic microvascular ischaemia was identified in 12 patients (21.4%) and 6 patients (10.7%) demonstrated cerebral atrophy. This high prevalence of cerebrovascular disease patients may reflect acute primary neurological inciting events which indirectly led to burn injury or a chronic disease process such as vascular dementia that can also predispose elderly patients to burns through deleterious effects on situational awareness, cognition, gait and mobility [15].
xxx
7
Previous study has shown that people with dementia have higher age-standardised incidence rates for burn injury than those without (22.7 per 100,000 vs. 14.2 per 100,000 population) equating to an incidence rate ratio of 1.6 (95% CI 1.3 2.0, p < 0.0001) [16]. Mean LOS was more than double (24 vs. 12 days) for people with dementia in the aforementioned study [16]. This is in keeping with the results of our cohort showing increased LOS in those with positive acute and chronic findings on neuroimaging (47 vs. 27 days). In addition to prolonging length of stay, cerebrovascular disease has been shown (in another large study of 31,338 burn patients) to significantly predict in-hospital mortality (OR 1.6) [17]. White matter disease is another common form of cerebrovascular disease and represents a risk factor for poor neurological outcomes [18]. Clinical symptoms related to white matter injury include cognitive and emotional dysfunction, sensorimotor impairment, urinary incontinence and pain. White matter disease was prevalent in this cohort of burn patients as indicated by the mean total Fazekas score (mean total score of 3.59 in neuroimaged burn patients). This score was calculated on review of all imaging studies in this study by a Consultant physician with a special interest in neurovascular pathology (JH). Fazekas scores were validated separately by a Consultant Radiologist, with whom there was 100% interobserver agreement. The Fazekas scale is a scoring system used to quantify the amount of hyper intense lesions identified on neuroimaging studies. This scale assigns scores to periventricular white matter disease, relating to a combination demyelination and gliosis (non-ischaemic changes) and deep white matter changes attributed to chronic small vessel ischaemia [19]. The deep white matter component of the Fazekas score is useful in the assessment of patients with possible dementia. Fazekas Grade 0 represents normal white matter while Grade 1 (mild white matter hyperintensities) is considered normal in elderly populations. Fazekas Grades 2 (moderate white matter disease) and 3 (severe white matter disease) are considered pathologic and although they may be seen in normally functioning individuals, they correlate with a high risk for disability, stroke and dementia [19,20]. Previous study of 600 normally functioning elderly people demonstrated that the Fazekas score predicted disability within one year [20]. Long term follow-up studies show that severe white matter changes independently and strongly predict rapid global functional decline, long term care and rehabilitation needs [21]. White matter disease therefore carries implications for service planning within acute care burn units, although more information on the natural course of vascular lesions and white matter changes is needed to directly inform patient management in terms of secondary prevention and therapeutic measures. Features of white matter disease may be subtle and review at a neuroimaging multidisciplinary team meeting with a specialist in neuropathology may be beneficial. There were two incidental but clinically significant new findings in this study: primary brain tumour and Huntington’s Chorea. Incidental findings on neuroimaging have been reported to occur with a frequency of 1.7 2.7% for MRI brain studies carried out in the general population [22,23] and 1% for CT brain studies [24] carried out following head trauma. The most common incidental CT brain findings in normal
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
8
burns xxx (2019) xxx
populations include cisterna magna enlargement followed by brain tumour and arachnoid cyst. Incidental MRI brain findings are common in the general population and include subclinical vascular pathologic changes, cerebral aneurysm and benign primary tumour [23]. The incidence of these findings increases with age. Mortality in patients with major burns as a direct result of cerebral lesions is rare but has been reported in one case of rapidly progressive glioma (not present on initial admission neuroimaging studies) [24]. Burn patients with neurological symptoms and signs should be investigated thoroughly to evaluate for the presence of pathology which may be associated with reduced survival, particularly in the setting of immunocompromise. In a military setting, predominated by ballistic trauma, 10% of burn patients sustained concomitant traumatic brain injury [25]. Traumatic brain injury was found to independently increase the risk of many complications (length of ICU and hospital stay, shock, amputation, wound sepsis and thromboembolism) when sustained with burns, although it did not independently increase mortality. Traumatic brain injury was seen in 6 patients or 10.7% of patients neuroimaged or 0.5% of our overall study cohort (3 diffuse hypoxic injury, 3 contusions with intracranial haemorrhage). Predicted (but not observed) mortality was higher in the neuroimaged cohort in the present study, although this reflect the higher mean TBSA burn in the former cohort. There is a paucity of research in civilian populations investigating intra-cranial injury in burn injury in terms of its incidence or potential to increase complications such as prolonged rehabilitation, delayed weaning off ventilation or increase length of stay. The results of this study indicate positive neuroimaging findings in civilian burn patients are not infrequent and predict prolonged length of stay. Neuroimaging also has discriminatory value in predicting rehabilitation requirements and patient discharge planning with approximately half of patients with acute (7/14) and chronic (18/32) neuroimaging findings requiring transfer to another inpatient facility after discharge from the burns unit for further inpatient rehabilitation. This has implications for burn patient prognosis and rehabilitation service planning in burn units. Future prospective study of neuroimaged burns patients in our unit will aim to further develop the general indications, outlined in Table 4, into specific recommendations or thresholds for neuroimaging. In particular, further research is warranted to investigate the relationship between TBSA, inhalation injury and positive neuroimaging findings, particularly given the logistical hazards of radiology department transfer for artificially ventilated patients with large burns.
5.
Conclusion
This study demonstrates the value of neuroimaging, when appropriate, in selected burn injured patients. Positive acute and chronic radiological findings have the potential to impact patient care, rehabilitation and outcome. A cohort of burn patients with increased risk of intra-cranial pathology have been identified, specifically those with larger burns undergoing formal intravenous fluid resuscitation and in those with inhalation injuries.
xxx
A high prevalence of acute and chronic cerebrovascular disease was detected in burns patients who underwent neuroimaging. This may reflect pre-existing neurodegeneration, a known risk factor for burn injury through impairment of situational awareness, cognition and gait. Cerebrovascular disease may also predict poorer outcomes such as increased length of stay and mortality. The clinical significance of white matter disease prevalent in burn patients, as indicated by the Fazekas score, is not clearly understood and warrants further study. Expanded indications for neuroimaging for patients following burn injury may be warranted and could be incorporated into any trauma systems dealing with acute burn injuries. Further study in large volume burn centres may be worthwhile to determine the true incidence, pathogenesis and clinical significance of positive neuroimaging findings specific to burn injured patients.
Author contributions E. Concannon: Study design, data analysis, interpretation, writing and critical revision. L. Fitzgerald: Data collection and writing. J. Birrane: Data analysis. E. Canniff: Data collection. J. Harbison: Data collection, analysis, interpretation. O. Shelley: Study design, data interpretation and critical review.
Conflict of interest None.
Funding None. REFERENCES
[1] Stylianou N, Buchan I, Dunn KW. A review of the international Burn Injury Database (iBID) for England and Wales: descriptive analysis of burn injuries 2003 2011. BMJ Open 2015;5 (February):e006184. [2] Brusselaers N, Monstrey S, Vogelaers D, Hoste E, Blot S. Severe burn aetiology in Europe: a systematic review of the incidence, aetiology, morbidity and mortality. Crit Care 2010;14(October (5)):R188. [3] Peck MD. Epidemiology of burns throughout the world. Part I: distribution and risk factors. Burns 2011;37(November (7)):1087 100. [4] Smolle C, Cambiaso-Daniel J, Forbes AA, Wurzer P, Hundeshagen G, Branski LK, et al. Recent trends in burn epidemiology worldwide: a systematic review. Burns 2017;43 (March (2)):249 57. [5] Douglas DB, Muldermans JL, Wintermark M. Neuroimaging of brain trauma. Curr Opin Neurol 2018;31(August (4)):362 70. [6] National Institute for Clinical Excellence (NICE). Clinical guideline 176: Head injury assessment and early management. Available at: https://www.nice.org.uk/guidan ce/cg176. [Accessed 1 May 2019].
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018
JBUR 5926 No. of Pages 9
burns xxx (2019) xxx
[7] Alden NE, Rabbitts A, Yurt RW. Burn injury in patients with dementia: an impetus for prevention. J Burn Care Res 2005;26:267 71. [8] Fauerbach JA, Lawrence J, Haythornthwaite J, Richter D, McGuire M, Schmidt C, et al. Pre-burn psychiatric history affects post-trauma morbidity. Psychosomatics 1997;38(July (4)):374 85. [9] Branhall RJ, Badran M, Shelley OP, Philip B, Dziewulski P. CT brain scans in burn patients — a worthwhile investigation? Burns 2009;35(September (S1)):53. [10] Ott LK, Hoffman LA, Hravnak M. Intrahospital transport to the radiology department: risk for adverse events, nursing surveillance, utilization of a MET and practice implications. J Radiol Nurs 2011;30(June (2)):49 52. [11] Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. Am J Roentgenol 1987;149(August (2)):351 6. [12] Osler T, Glance L, Hosmer D. Simplified estimates of the probability of death after burn injuries: extending and updating the baux score. J Trauma 2010;68(March (3)):690 7. [13] Woods JF, Quinlan CS, Shelley OP. Predicting mortality in severe burns-what is the score? Evaluation and comparison of 4 mortality prediction scores in an Irish population. Plast Reconstr Surg Glob Open 2016;4(February (1)):e606. [14] Obeid DA, Abdulaziz K, Alhujayri AK, Aldekhayel S. Burn induced neuroepithelial changes as a delayed cause of mortality in major burns: a case report and literature review. Int J Burn Trauma 2018;8(December (6)):145 8. [15] Esechie A, Bhardwaj A, Masel T, Raji M. Neurocognitive sequela of burn injury in the elderly. J Clin Neurosci 2019;59 (January):1 5. [16] Harvey L, Mitchell R, Brodaty H, Draper B, Close J. Dementia: a risk factor for burns in the elderly. Burns 2016;42(March (2)):282 90.
xxx
9
[17] Thombs BD, Singh VA, Halonen J, Diallo A, Milner SM. The effects of preexisting medical comorbidities on mortality and length of hospital stay in acute burn injury: evidence from a national sample of 31,338 adult patients. Ann Surg 2007;245 (April (4)):629 34. [18] Wang Y, Liu G, Hong D, Chen F, Ji X, Cao G. White matter injury in ischemic stroke. Prog Neurobiol 2016;141(June):45 60. [19] Kim KW, MacFall JR, Payne ME. Classification of white matter lesions on magnetic resonance imaging in elderly persons. Biol Psychiatry 2008;64(August (4)):273 80. [20] Inzitari D, Simoni M, Pracucci G, Poggesi A, Basile AM, Chabriat H, et al. Risk of rapid global functional decline in elderly patients with severe cerebral age-related white matter changes: the LADIS study.; LADIS Study Group. Arch Intern Med 2007;167(January (1)):81 8. [21] Inzitari D, Pracucci G, Poggesi A, Carlucci G, Barkhof F, Chabriat H, et al. Changes in white matter as determinant of global functional decline in older independant outpatients: three year follow-up of LADIS (leukoaraiosis and disability) study cohort. LADIS Study Group. BMJ 2009;339(July): b2477. [22] Bunnik EM, Vernooij MW. Incidental findings in population imaging revisited. Eur J Epidemiol 2016;31(January (1)):1 4. [23] Vernooij MW, Ikram MA, Tanghe HL, Vincent AJ, Hofman A, Krestin GP, et al. Incidental findings on brain MRI in the general population. N Engl J Med 2007;357(November (18)): 1821 8. [24] Eskandary H, Sabba M, Khajehpour F, Eskandari M. Incidental findings in brain computed tomography scans of 3000 head trauma patients. Surg Neurol 2005;63(June (6)): 550 3. [25] Graybill JC, Aden J, Rizzo JA, Clemens M, Janak J, White C, et al. Concomittant traumatic brain injury and burns in combat trauma. J Am Coll Surg 2016;223(October (4)):e217.
Please cite this article in press as: E. Concannon, et al., Neuroimaging provides relevant clinical information in patients with burn injuries, Burns (2019), https://doi.org/10.1016/j.burns.2019.08.018