- Email: [email protected]
Epidemiology of penetrating injuries in the United Kingdom: A systematic review
Accepted Manuscript Epidemiology of penetrating injuries in the United Kingdom: A systematic review G. Whittaker, J. Norton, J. Densley, D. Bew PII:
S1743-9191(17)30271-6
DOI:
10.1016/j.ijsu.2017.03.051
Reference:
IJSU 3686
To appear in:
International Journal of Surgery
Received Date: 1 March 2017 Revised Date:
16 March 2017
Accepted Date: 17 March 2017
Please cite this article as: Whittaker G, Norton J, Densley J, Bew D, Epidemiology of penetrating injuries in the United Kingdom: A systematic review, International Journal of Surgery (2017), doi: 10.1016/ j.ijsu.2017.03.051. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Epidemiology of penetrating injuries in the United Kingdom: A systematic review
Affiliations:
RI PT
G. Whittaker1, J. Norton1, J. Densley2, D. Bew3
School of Medical Education, King’s College London, London, UK
2
School of Law Enforcement and Criminal Justice, Metropolitan State University, Brooklyn Park,
SC
1
3
M AN U
Minnesota, USA
Department of Trauma and Emergency Surgery, King’s College Hospital NHS Foundation Trust,
London, UK
Correspondence:
TE D
Duncan Bew
Consultant Trauma & Emergency Surgeon
EP
Department of Trauma and Emergency Surgery
King’s College Hospital NHS Foundation Trust, London, SE5 9RS, UK
AC C
Ph: +44 (0)20 3299 9000
Fax: +44 (0)20 3299 3445
Email: [email protected]
ACCEPTED MANUSCRIPT
Epidemiology of penetrating injuries in the United Kingdom: A systematic review
RI PT
ABSTRACT Introduction: Penetrating injuries account for a significant number of deaths in the United Kingdom (UK) annually. Numerous articles have examined the epidemiology of penetrating trauma in various
SC
areas of the UK. This article aimed to systematically review the current literature and evaluate the incidence and mortality of penetrating injury according to region in the UK.
M AN U
Methods: A systematic literature search was performed using MEDLINE® (1946 to June 2016), EMBASE® (1974 to June 2016), and PsycINFO® (1806 to June 2016) databases. The following keywords were used in combination with Boolean operators: “epidemiology”, “incidence”, “frequency”, “pattern”, “distribution”; “penetrating”; “injuries”, “injury”, “trauma”; “United
TE D
Kingdom”, “UK”, “England”, “Scotland”, “Wales”, “London”.
Results: Eleven relevant studies were identified across five regions of the UK. Study periods ranged from 3 months to 16 years and encompassed between 343 and 127,191 patients. Relative incidence
EP
within individual studies ranged from 0.3% (Midlands) to 21.0% (London) and mortality ranged from 0.5% (London) to 15.4% (Midlands). The majority of patients were young males.
AC C
Discussion: An extensive range of incidence and mortality rates were observed between studies in all regions. This was largely dependent on the study population under review. London was found to have the highest incidence of penetrating injuries, however these studies tended to focus on populations of trauma patients. The high proportion of male victims may reflect the risk of becoming involved in gangs and violence. Conclusions: Our ambiguous results indicate the need for further work directed towards the epidemiology of penetrating injuries within regional trauma networks.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Keywords: epidemiology; penetrating; injury; trauma; United Kingdom.
ACCEPTED MANUSCRIPT 1.1 INTRODUCTION Penetrating injuries (PI), defined as any injury that occurs when an object pierces the skin and
RI PT
enters a tissue of the body, constitute an estimated 3.1% of adult [1] and 10% of paediatric [2] trauma in the United Kingdom (UK). Penetrating trauma is a major contributing factor to premature
mortality and permanent disability; it can have dramatic physical and psychological consequences for the individual, particularly when the injury is sustained via assault [3]. There are significant economic
SC
and political implications; the average treatment cost per patient amounts to nearly £8,000 [4], which places emphasis on the need for primary prevention through legislation and other interventions.
M AN U
Indeed, penetrating trauma tends to be a consequence of interpersonal violence [5-9]. Several epidemiological studies have examined incidence and mortality rates of penetrating trauma in various areas of the UK [9-13]. The extant literature is limited, however, because frequently it is localised and utilises specific study populations that differ in nature of injury. To date, no
TE D
comparison has been made between studies to draw general conclusions. The current study aimed to systematically review the published literature and evaluate the incidence and mortality rates of adult
AC C
EP
penetrating trauma in relation to geographical region within the UK.
1.2 PATIENTS AND METHODS
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) method directed this systematic review. Comprehensive searches of the MEDLINE® (1946 to June 2016), EMBASE® (1974 to June 2016), and PsycINFO® (1806 to June 2016) databases were performed to identify studies of potential relevance. Keywords were combined using “OR”/”AND”
ACCEPTED MANUSCRIPT operators and included the following: “epidemiology”, “incidence”, “frequency”, “pattern”, “distribution”; “penetrating”; “injuries”, “injury”, “trauma”; “United Kingdom”, “UK”, “England”, “Scotland”, “Wales”, “London”. Results were screened for applicability, with further scrutiny of the abstract if uncertain, prior to removal of duplicates. The remaining articles were subject to exclusion
RI PT
criteria and only those that met all conditions were included in this review. Studies from various areas of the UK were included. National epidemiological studies were used to evaluate the overall incidence of adult penetrating trauma in the UK. Research conducted
SC
locally at NHS Foundation Trusts was also incorporated to provide an insight into the patterns and disparities between regions. This included articles which specifically reviewed the incidence of
concerning adult populations.
M AN U
penetrating trauma. The criteria permitted inclusion of complete articles and conference abstracts
Publications that reviewed populations from outside the UK were excluded from this review, including those concerning the military and Republic of Ireland. Studies which lacked adequate
TE D
supporting data or evaluated only the incidence of non-penetrating trauma were also excluded. Furthermore, studies that examined the incidence of uncommon traumatic injuries (i.e. splenic
AC C
EP
trauma) or paediatric populations were excluded.
1.3 RESULTS
The literature search generated 118 potentially relevant publications for appraisal. Of these studies, 84 were deemed irrelevant and subsequently excluded as they related to diagnosis or management of PI. Eight were rejected from this review as they evaluated populations from outside the UK and a further 15 were excluded as they studied paediatric populations or the incidence of
ACCEPTED MANUSCRIPT specific visceral injuries. This rendered a total of 11 papers for final inclusion after reference checking [Figure 1]. A summary of the epidemiological evidence is presented in [Table 1]. Evidence was gathered from four modalities of data source across five geographical regions of the UK: England, Wales &
RI PT
Northern Ireland (EWNI), Scotland, Midlands, South West, and London. The mean study period was 46 months (range: three months to 16 years). The number of included patients ranged from 343 to 127,191, though one article did not disclose study population figures. High heterogeneity was also
SC
observed between studies regarding the frequency of penetrating trauma cases; incidence ranged from 0.3% to 21.0%. However, the latter figure arose from an article which specifically evaluated trauma
M AN U
patients. Mortality rates were reported to lie between 0.5% and 15.4% for the five publications which examined deaths from PI.
In all regions, the vast majority of cases were male with the highest proportion in the SouthWest (91.9%) and the lowest in the Midlands (84.7%). Patient ethnicity was reported in only one
TE D
article where 50% were assigned to the category Black or Black British. The highest reported incidence of penetrating trauma in each region is as follows: London (21.0%), Scotland (8.7%), South West (7.3%), EWNI (7.1%), and Midlands (3.6%). A study
EP
conducted in the Midlands was also found to have the lowest incidence rate of 0.3%. The mortality rates followed a different pattern: Midlands (15.4%), London (13.2%), EWNI (8.9%), and Scotland
AC C
(8.2%). Mortality in the South West was not reported.
ACCEPTED MANUSCRIPT 1.4 DISCUSSION Several surprising results were discovered upon statistical analysis of the included studies.
RI PT
Significant variation was observed between studies regarding the relative incidence (0.3 - 21.0%) and mortality rate (0.5 - 15.4%) of PI. One explanation for this heterogeneity would be the inclusion of articles which specifically reviewed major trauma patient populations, which have a higher incidence
SC
and mortality rates in comparison to those that investigated the general population.
The overall proportion of males that comprised the PI population was found to be similar for
M AN U
each region (84.7% - 91.9%) reflecting the high proportion caused by violent assault [5-9]. It is interesting to note that 10/11 publications did not disclose the ethnicity of PI patients despite revealing that this information was collected in the methodology. The reasons for this are not clear, especially as ethnicity is likely to be an important factor in the aetiology of PI as demonstrated by disproportionate rates (50%) of Afro-Caribbean gunshot victims in South East London, which, in
TE D
turn, may reflect disproportionate patterns of engagement in violent, illicit networks and gangs [14]. An explanation for these omissions may be the potentially negative connotations of associating a particular ethnicity with violent crime. However, different regions of the UK have distinct
EP
ethnographic preponderance of both social and economic challenges, which are associated with
AC C
deprivation and criminal activity [15, 16].
1.4.1
England, Wales & Northern Ireland (EWNI) Three articles reviewed the incidence of penetrating trauma in EWNI using national statistics from the Trauma Audit & Research Network (TARN) database. PI was accountable for 3.7% of all major trauma cases between January 2000 and December 2005, with the majority (91%) of patients
ACCEPTED MANUSCRIPT being young males at a median age of 30 years [4]. A relatively recent single-year review of this database was also conducted in 2012; the authors reported the incidence of PI to have dropped slightly to 3.1% which represents approximately 1,150 cases per year [1]. Another study examined 1000 mortalities across the UK in 1989 to ascertain the incidence
RI PT
and pattern of PI [17]. They found that knife- and firearm-related injuries occurred in 71 (7.1%) of cases, which is almost double that of recent reviews. This may be attributable to the relatively high mortality rate observed in this study, with only 24% of patients surviving the journey to hospital prior
SC
to the advent of trauma networks and helicopter transport.
A major limitation of these studies, particularly older ones, is that the TARN database
M AN U
originally received data from only 50% of trauma hospitals in EWNI. This number has increased over time and now 100% of hospitals receiving major trauma submit data to TARN, but the implication is that the overall incidence of penetrating trauma in the UK could be higher than reported.
TE D
1.4.2
Scotland
EP
One regional epidemiological study of penetrating trauma has been performed in Scotland to compare outcomes between trauma patients residing in urban and rural areas. Overall, PI accounted
AC C
for 8.7% of emergency department admissions with a reported mortality rate of 8.2% [10]. The proportion of patients with penetrating trauma was over five times higher in urban areas (9.9%) when compared to rural areas (1.9%), most likely due to higher rates of street violence and road traffic accidents. There was, however, no difference in mortality rates between these groups despite the longer pre-hospital times experienced by those living in rural areas.
ACCEPTED MANUSCRIPT 1.4.3 Midlands Independent research has also been completed in some of the largest urban areas in central
RI PT
England. The frequency of PI in South Yorkshire was reviewed using ambulance data which spanned nine years [18]. The results showed an escalating trend with a total growth of 65% over the study period, which the authors proposed was due to increased knife and gun-related injury. However, the overall proportion of penetrating trauma cases remained almost identical from 1999 (0.30%) to 2008
SC
(0.33%). A similarly broad study was conducted across the Mersey and North Wales regions over 20 years ago; however the authors reported an equivalent proportion of PI compared to the present
M AN U
(3.6%) [19]. These conflicting conclusions regarding the evolving incidence of PI potentially reflect the dynamic changes seen in society in general.
Interestingly, patients who suffered PI in the Mersey & North Wales region were found to have the highest mortality rate of all included studies (15.4%). This may be explained by a
TE D
combination of two factors. Firstly, of all included studies this was one of the earliest conducted, meaning that patients were unable to benefit from the advances in diagnostic technology and treatment options that have since been developed. Additionally, this region is a rural area with
EP
potentially prolonged travel times to the closest hospital, which may have had a dramatic impact on
AC C
the survival of these patients.
1.4.4 South West
Only one paper of relevance to this region was found. The authors studied a population of 510 assault victims in Bristol and found a PI caused by a sharp weapon in 7.3% of cases [11]. Lacerations were also observed in 42% of patients. However, the conclusions of the study are limited by the
ACCEPTED MANUSCRIPT exclusion of other common penetrating aetiologies such as road traffic collisions. Furthermore, this article was published over 25 years ago and therefore may not be representative of the current
1.4.5 London
RI PT
incidence of PI. The mortality rate from penetrating trauma in this region also remains unknown.
SC
The epidemiology of PI in London has received the most attention to date, with four
publications in total examining different areas of the capital. The frequency of adult trauma patients
M AN U
presenting with PI was prospectively reviewed at St George’s Hospital (SGH), the Major Trauma Centre (MTC) designated to South West London [20]. Having received 343 patients over a three month period, a total of 39 (11%) had a penetrating mechanism of injury. Another single centre study at London’s North East MTC, the Royal London Hospital (RLH), found that PI constituted 21% of trauma admissions and 13% of mortality [12]. The majority of mortalities in this group (31/40) were
TE D
caused by knife wounds as opposed to firearms or road traffic accidents. There also appears to be a growing number of violent assaults in London. Recent statistics
EP
from the Metropolitan Police revealed that over 4,200 knife-related offences were committed between April and August 2015 [21]. This is a 9.7% increase from the previous year, with 876 occurring in
AC C
August alone. This trend coincides with a 16-year review of the London Helicopter Emergency Medical Service (HEMS) database, which showed an annual rise of 23.2% and 11% in stabbings and shootings respectively [13]. A total of 1,564 (9.7%) of the 16,068 major trauma patients had sustained a deliberate PI over the study period. A further study of deliberate penetrating trauma has been carried out in perhaps the most prolific region of London – the South East. The inner-city borough of Southwark is reported to have one of the highest offence rates in London, with 261 recent knife crimes over a five month span representing an annual increase of 23.7% [21]. A cross-sectional study performed at King’s College
ACCEPTED MANUSCRIPT Hospital (KCH), the MTC in Southwark responsible for the South East London region, established that 938 knife injuries presented to the emergency department in 2011 [9]. The authors estimated the annual incidence rate in the South East population to be 737 per 100,000 persons. They also found that PI consumed a quarter (24%) of the trauma team’s workload despite the fact this group of patients
RI PT
comprised a tiny fraction (0.77%) of the 127,191 attendances. Furthermore, knives were responsible for 94.1% of penetrating trauma cases and assaults accounted for 44% of these presentations, though a
M AN U
SC
suspiciously-high proportion of cases (49%) were attributable to ‘accidents’.
1.5
LIMITATIONS
Two major limitations were identified in this review. Firstly, several of the included studies
TE D
did not report important variables such as mortality rate. This may limit any conclusions that can be drawn regarding mortality rates of PI patients. Furthermore, one study conducted in the Midlands lacked data on sample size and absolute number of penetrating trauma cases, with only the proportion
EP
of PI patients disclosed.
The second and perhaps more influential limitation is the inclusion of studies which evaluated
AC C
major trauma patient populations. These tended to exhibit the highest incidence and mortality rates, which may have skewed results. Our inclusion criteria intended to encompass all published data regarding the frequency of PI, however this has yielded potentially unrepresentative results. The low number of regional studies evaluating PI in general populations indicates the need for further epidemiological research, particularly within trauma networks. This will allow for accurate comparisons to be made between regions.
ACCEPTED MANUSCRIPT 1.6 CONCLUSIONS Penetrating trauma is common and has many medical and emotional consequences alongside
RI PT
the legal, economic and political implications. A systematic review of the current literature has yielded some unexpected results regarding the geographical pattern, incidence, and demographics of PI in the UK. A variance of incidence and mortality rate was observed between studies in all regions, depending on the specific study population under review. This may reflect the inherent variability of
SC
injury severity and outcome in accordance with classification of trauma patients.
M AN U
London was found to have the highest incidence rate of any region. However, the majority of studies originating from London were concerned with a population of major trauma patients. Therefore, the epidemiological data regarding PI in London may be of limited value when evaluating incidence within the general population. Mersey & North Wales, categorised within the Midlands region, was found to have the highest mortality rate of any published article, with similar rates
TE D
observed in London.
A significant majority of patients were male in all included studies which may represent the increased risk of males becoming involved in gangs and violence, as evidenced by the high
EP
contribution of assault to PI. Patients also tended to be fairly young with the exception of one study
AC C
which analysed the TARN database.
In conclusion, ambiguous incidence and mortality rates within regions have resulted in
uncertainty despite analysis of all relevant literature. Furthermore, the low number of studies conducted in each region limits the validity of any conclusions made. Therefore, further original work needs to be completed on the incidence, geographical distribution, and demographics of penetrating trauma including analysis of ethnicity. This should be conducted in the context of regional trauma networks within the UK.
ACCEPTED MANUSCRIPT 1.7 ACKNOWLEDGEMENTS This research received no funding from any agency in the public, commercial, or not-for-profit
M AN U
REFERENECES
SC
1.8
RI PT
sectors.
AC C
EP
TE D
[1] Woodford M, Lawrence T, Bouamra O, Edwards A, Lecky F. Major trauma audit: Metrics for improving care. African Journal of Emergency Medicine. 2013 12//: 3:S10 [2] Sinha CK, Lander A. Trauma in children: abdomen and thorax. Surgery (Oxford). 2013 3//: 31:123-9 [3] Holbrook TL, Hoyt DB. The impact of major trauma: quality-of-life outcomes are worse in women than in men, independent of mechanism and injury severity. The Journal of trauma. 2004 Feb: 56:284-90 [4] Christensen MC, Nielsen TG, Ridley S, Lecky FE, Morris S. Outcomes and costs of penetrating trauma injury in England and Wales. Injury. 2008 9//: 39:1013-25 [5] Brady RR, O'Neill S, Berry O, Kerssens JJ, Yalamarthi S, Parks RW. Traumatic injury to the colon and rectum in Scotland: demographics and outcome. Colorectal disease : the official journal of the Association of Coloproctology of Great Britain and Ireland. 2012 Jan: 14:e16-22 [6] Tambyraja AL, Scollay JM, Beard D, Henry JM, Murie JA, Chalmers RT. Aortic trauma in Scotland--a population based study. European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery. 2006 Dec: 32:686-9 [7] Skipworth RJ, McBride OM, Kerssens JJ, Paterson-Brown S. Esophagogastric trauma in Scotland. World journal of surgery. 2012 Aug: 36:1779-84 [8] Scollay JM, Beard D, Smith R, McKeown D, Garden OJ, Parks R. Eleven years of liver trauma: the Scottish experience. World journal of surgery. 2005 Jun: 29:744-9 [9] Pallett JR, Sutherland E, Glucksman E, Tunnicliff M, Keep JW. A cross-sectional study of knife injuries at a London major trauma centre. Annals of the Royal College of Surgeons of England. 2014 Jan: 96:23-6 [10] McGuffie AC, Graham CA, Beard D, et al. Scottish urban versus rural trauma outcome study. The Journal of trauma. 2005 Sep: 59:632-8 [11] Shepherd JP, Shapland M, Pearce NX, Scully C. Pattern, severity and aetiology of injuries in victims of assault. Journal of the Royal Society of Medicine. 1990 Feb: 83:75-8 [12] Chalkley D, Cheung G, Walsh M, Tai N. Deaths from trauma in London--a single centre experience. Emergency medicine journal : EMJ. 2011 Apr: 28:305-9 [13] Crewdson K, Lockey D, Weaver A, Davies GE. Is the prevalence of deliberate penetrating trauma increasing in London? Experiences of an urban pre-hospital trauma service. Injury. 2009 May: 40:560-3
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[14] Papachristos AV, Hureau DM, Braga AA. The corner and the crew: the influence of geography and social networks on gang violence. American sociological review. 2013: 78:417-47 [15] Gill B. The English Indices of Deprivation 2015: Department for Communities and Local Government, Government of the United Kingdom; 2015 [16] Jivraj S, Khan O. Ethnicity and deprivation in England: How likely are ethnic minorities to live in deprived neighbourhoods?: ESRC Centre on Dynamics of Ethnicity; 2013 [17] Anderson ID, Woodford M, Irving MH. Preventability of death from penetrating injury in England and Wales. Injury. 1989 Mar: 20:69-71 [18] Gray JT, Walker A. “At the sharp end”: does ambulance dispatch data from south Yorkshire support the picture of increased weapon-related violence in the UK? Emergency Medicine Journal. 2009 October 1, 2009: 26:741-2 [19] Gorman DF, Teanby DN, Sinha MP, Wotherspoon J, Boot DA, Molokhia A. The epidemiology of major injuries in Mersey Region and North Wales. Injury. 1995 Jan: 26:51-4 [20] Roche CD, McDonald CM, Liu B, Razik A, Bishop T. Missed orthopaedic injuries in adult trauma patients at a major trauma centre. Trauma. 2015 April 1, 2015: 17:123-7 [21] Metropolitan Police Service: Monthly Knife Crime Summary. 2015 August 2015 [cited 24 December 2015]; Available from: http://www.met.police.uk/foi/pdfs/priorities_and_how_we_are_doing/corporate/monthly_knife_crime _summary_august2015.pdf
ACCEPTED MANUSCRIPT TABLE 1. Sources of epidemiological evidence for penetrating injuries in the United Kingdom (UK).
4
Christensen et al. Anderson et al.17 McGuffie et al.10 Gray & Walker18 Gorman et al.19 Shepherd et al.11 Pallett et al.9 Chalkley et al.12 Crewdson et al.13 20 Roche et al.
Study period
Region
Data Number of source patients
Number of Mortality PI patients (%) rate (%)
EWNI
Apr 2012 - Apr 2013
TARN
37,353
1,158 (3.1)
N/R
EWNI EWNI Scotland Midlands Midlands South West London (SE) London (NE) London London (SW)
Jan 2000 - Dec 2005 1989 2003 - 2005 Apr 1999 - Mar 2008 May 1989 - Apr 1990 Jan - Dec 1986 Jan - Dec 2011 Jun 2004 - May 2008 Jan 1991 - Dec 2006 Jul - Sept 2012
TARN TARN ED YAS ED ED ED ED HEMS ED
~37,000 1,000 4,636 N/R 1,088 510 1,27,191 4,243 16,068 343
1,365 (3.7) 71 (7.1) 405 (8.7) N/R (0.3) 39 (3.6) 37 (7.3) 938 (0.8) 892 (21.0) 1,564 (9.7) 39 (11.4)
8.3 N/A 8.2 N/R 15.4 N/R 0.5 13.2 N/R N/R
RI PT
Study 1 Woodford et al.
AC C
EP
TE D
M AN U
SC
PI, penetrating injury; EWNI, England, Wales & Northern Ireland; SW, South West; NE, North East; SE, South East; N/R, not recorded; N/A, not applicable; TARN, Trauma Audit & Research Network; ED, Emergency Department; YAS, Yorkshire Ambulance Service; HEMS, Helicopter Emergency Medical Service.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT HIGHLIGHTS
EP
TE D
M AN U
SC
RI PT
11 articles reviewed Highest incidence in London Highest mortality in Midlands
AC C
• • •
S1743-9191(17)30271-6
DOI:
10.1016/j.ijsu.2017.03.051
Reference:
IJSU 3686
To appear in:
International Journal of Surgery
Received Date: 1 March 2017 Revised Date:
16 March 2017
Accepted Date: 17 March 2017
Please cite this article as: Whittaker G, Norton J, Densley J, Bew D, Epidemiology of penetrating injuries in the United Kingdom: A systematic review, International Journal of Surgery (2017), doi: 10.1016/ j.ijsu.2017.03.051. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Epidemiology of penetrating injuries in the United Kingdom: A systematic review
Affiliations:
RI PT
G. Whittaker1, J. Norton1, J. Densley2, D. Bew3
School of Medical Education, King’s College London, London, UK
2
School of Law Enforcement and Criminal Justice, Metropolitan State University, Brooklyn Park,
SC
1
3
M AN U
Minnesota, USA
Department of Trauma and Emergency Surgery, King’s College Hospital NHS Foundation Trust,
London, UK
Correspondence:
TE D
Duncan Bew
Consultant Trauma & Emergency Surgeon
EP
Department of Trauma and Emergency Surgery
King’s College Hospital NHS Foundation Trust, London, SE5 9RS, UK
AC C
Ph: +44 (0)20 3299 9000
Fax: +44 (0)20 3299 3445
Email: [email protected]
ACCEPTED MANUSCRIPT
Epidemiology of penetrating injuries in the United Kingdom: A systematic review
RI PT
ABSTRACT Introduction: Penetrating injuries account for a significant number of deaths in the United Kingdom (UK) annually. Numerous articles have examined the epidemiology of penetrating trauma in various
SC
areas of the UK. This article aimed to systematically review the current literature and evaluate the incidence and mortality of penetrating injury according to region in the UK.
M AN U
Methods: A systematic literature search was performed using MEDLINE® (1946 to June 2016), EMBASE® (1974 to June 2016), and PsycINFO® (1806 to June 2016) databases. The following keywords were used in combination with Boolean operators: “epidemiology”, “incidence”, “frequency”, “pattern”, “distribution”; “penetrating”; “injuries”, “injury”, “trauma”; “United
TE D
Kingdom”, “UK”, “England”, “Scotland”, “Wales”, “London”.
Results: Eleven relevant studies were identified across five regions of the UK. Study periods ranged from 3 months to 16 years and encompassed between 343 and 127,191 patients. Relative incidence
EP
within individual studies ranged from 0.3% (Midlands) to 21.0% (London) and mortality ranged from 0.5% (London) to 15.4% (Midlands). The majority of patients were young males.
AC C
Discussion: An extensive range of incidence and mortality rates were observed between studies in all regions. This was largely dependent on the study population under review. London was found to have the highest incidence of penetrating injuries, however these studies tended to focus on populations of trauma patients. The high proportion of male victims may reflect the risk of becoming involved in gangs and violence. Conclusions: Our ambiguous results indicate the need for further work directed towards the epidemiology of penetrating injuries within regional trauma networks.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Keywords: epidemiology; penetrating; injury; trauma; United Kingdom.
ACCEPTED MANUSCRIPT 1.1 INTRODUCTION Penetrating injuries (PI), defined as any injury that occurs when an object pierces the skin and
RI PT
enters a tissue of the body, constitute an estimated 3.1% of adult [1] and 10% of paediatric [2] trauma in the United Kingdom (UK). Penetrating trauma is a major contributing factor to premature
mortality and permanent disability; it can have dramatic physical and psychological consequences for the individual, particularly when the injury is sustained via assault [3]. There are significant economic
SC
and political implications; the average treatment cost per patient amounts to nearly £8,000 [4], which places emphasis on the need for primary prevention through legislation and other interventions.
M AN U
Indeed, penetrating trauma tends to be a consequence of interpersonal violence [5-9]. Several epidemiological studies have examined incidence and mortality rates of penetrating trauma in various areas of the UK [9-13]. The extant literature is limited, however, because frequently it is localised and utilises specific study populations that differ in nature of injury. To date, no
TE D
comparison has been made between studies to draw general conclusions. The current study aimed to systematically review the published literature and evaluate the incidence and mortality rates of adult
AC C
EP
penetrating trauma in relation to geographical region within the UK.
1.2 PATIENTS AND METHODS
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) method directed this systematic review. Comprehensive searches of the MEDLINE® (1946 to June 2016), EMBASE® (1974 to June 2016), and PsycINFO® (1806 to June 2016) databases were performed to identify studies of potential relevance. Keywords were combined using “OR”/”AND”
ACCEPTED MANUSCRIPT operators and included the following: “epidemiology”, “incidence”, “frequency”, “pattern”, “distribution”; “penetrating”; “injuries”, “injury”, “trauma”; “United Kingdom”, “UK”, “England”, “Scotland”, “Wales”, “London”. Results were screened for applicability, with further scrutiny of the abstract if uncertain, prior to removal of duplicates. The remaining articles were subject to exclusion
RI PT
criteria and only those that met all conditions were included in this review. Studies from various areas of the UK were included. National epidemiological studies were used to evaluate the overall incidence of adult penetrating trauma in the UK. Research conducted
SC
locally at NHS Foundation Trusts was also incorporated to provide an insight into the patterns and disparities between regions. This included articles which specifically reviewed the incidence of
concerning adult populations.
M AN U
penetrating trauma. The criteria permitted inclusion of complete articles and conference abstracts
Publications that reviewed populations from outside the UK were excluded from this review, including those concerning the military and Republic of Ireland. Studies which lacked adequate
TE D
supporting data or evaluated only the incidence of non-penetrating trauma were also excluded. Furthermore, studies that examined the incidence of uncommon traumatic injuries (i.e. splenic
AC C
EP
trauma) or paediatric populations were excluded.
1.3 RESULTS
The literature search generated 118 potentially relevant publications for appraisal. Of these studies, 84 were deemed irrelevant and subsequently excluded as they related to diagnosis or management of PI. Eight were rejected from this review as they evaluated populations from outside the UK and a further 15 were excluded as they studied paediatric populations or the incidence of
ACCEPTED MANUSCRIPT specific visceral injuries. This rendered a total of 11 papers for final inclusion after reference checking [Figure 1]. A summary of the epidemiological evidence is presented in [Table 1]. Evidence was gathered from four modalities of data source across five geographical regions of the UK: England, Wales &
RI PT
Northern Ireland (EWNI), Scotland, Midlands, South West, and London. The mean study period was 46 months (range: three months to 16 years). The number of included patients ranged from 343 to 127,191, though one article did not disclose study population figures. High heterogeneity was also
SC
observed between studies regarding the frequency of penetrating trauma cases; incidence ranged from 0.3% to 21.0%. However, the latter figure arose from an article which specifically evaluated trauma
M AN U
patients. Mortality rates were reported to lie between 0.5% and 15.4% for the five publications which examined deaths from PI.
In all regions, the vast majority of cases were male with the highest proportion in the SouthWest (91.9%) and the lowest in the Midlands (84.7%). Patient ethnicity was reported in only one
TE D
article where 50% were assigned to the category Black or Black British. The highest reported incidence of penetrating trauma in each region is as follows: London (21.0%), Scotland (8.7%), South West (7.3%), EWNI (7.1%), and Midlands (3.6%). A study
EP
conducted in the Midlands was also found to have the lowest incidence rate of 0.3%. The mortality rates followed a different pattern: Midlands (15.4%), London (13.2%), EWNI (8.9%), and Scotland
AC C
(8.2%). Mortality in the South West was not reported.
ACCEPTED MANUSCRIPT 1.4 DISCUSSION Several surprising results were discovered upon statistical analysis of the included studies.
RI PT
Significant variation was observed between studies regarding the relative incidence (0.3 - 21.0%) and mortality rate (0.5 - 15.4%) of PI. One explanation for this heterogeneity would be the inclusion of articles which specifically reviewed major trauma patient populations, which have a higher incidence
SC
and mortality rates in comparison to those that investigated the general population.
The overall proportion of males that comprised the PI population was found to be similar for
M AN U
each region (84.7% - 91.9%) reflecting the high proportion caused by violent assault [5-9]. It is interesting to note that 10/11 publications did not disclose the ethnicity of PI patients despite revealing that this information was collected in the methodology. The reasons for this are not clear, especially as ethnicity is likely to be an important factor in the aetiology of PI as demonstrated by disproportionate rates (50%) of Afro-Caribbean gunshot victims in South East London, which, in
TE D
turn, may reflect disproportionate patterns of engagement in violent, illicit networks and gangs [14]. An explanation for these omissions may be the potentially negative connotations of associating a particular ethnicity with violent crime. However, different regions of the UK have distinct
EP
ethnographic preponderance of both social and economic challenges, which are associated with
AC C
deprivation and criminal activity [15, 16].
1.4.1
England, Wales & Northern Ireland (EWNI) Three articles reviewed the incidence of penetrating trauma in EWNI using national statistics from the Trauma Audit & Research Network (TARN) database. PI was accountable for 3.7% of all major trauma cases between January 2000 and December 2005, with the majority (91%) of patients
ACCEPTED MANUSCRIPT being young males at a median age of 30 years [4]. A relatively recent single-year review of this database was also conducted in 2012; the authors reported the incidence of PI to have dropped slightly to 3.1% which represents approximately 1,150 cases per year [1]. Another study examined 1000 mortalities across the UK in 1989 to ascertain the incidence
RI PT
and pattern of PI [17]. They found that knife- and firearm-related injuries occurred in 71 (7.1%) of cases, which is almost double that of recent reviews. This may be attributable to the relatively high mortality rate observed in this study, with only 24% of patients surviving the journey to hospital prior
SC
to the advent of trauma networks and helicopter transport.
A major limitation of these studies, particularly older ones, is that the TARN database
M AN U
originally received data from only 50% of trauma hospitals in EWNI. This number has increased over time and now 100% of hospitals receiving major trauma submit data to TARN, but the implication is that the overall incidence of penetrating trauma in the UK could be higher than reported.
TE D
1.4.2
Scotland
EP
One regional epidemiological study of penetrating trauma has been performed in Scotland to compare outcomes between trauma patients residing in urban and rural areas. Overall, PI accounted
AC C
for 8.7% of emergency department admissions with a reported mortality rate of 8.2% [10]. The proportion of patients with penetrating trauma was over five times higher in urban areas (9.9%) when compared to rural areas (1.9%), most likely due to higher rates of street violence and road traffic accidents. There was, however, no difference in mortality rates between these groups despite the longer pre-hospital times experienced by those living in rural areas.
ACCEPTED MANUSCRIPT 1.4.3 Midlands Independent research has also been completed in some of the largest urban areas in central
RI PT
England. The frequency of PI in South Yorkshire was reviewed using ambulance data which spanned nine years [18]. The results showed an escalating trend with a total growth of 65% over the study period, which the authors proposed was due to increased knife and gun-related injury. However, the overall proportion of penetrating trauma cases remained almost identical from 1999 (0.30%) to 2008
SC
(0.33%). A similarly broad study was conducted across the Mersey and North Wales regions over 20 years ago; however the authors reported an equivalent proportion of PI compared to the present
M AN U
(3.6%) [19]. These conflicting conclusions regarding the evolving incidence of PI potentially reflect the dynamic changes seen in society in general.
Interestingly, patients who suffered PI in the Mersey & North Wales region were found to have the highest mortality rate of all included studies (15.4%). This may be explained by a
TE D
combination of two factors. Firstly, of all included studies this was one of the earliest conducted, meaning that patients were unable to benefit from the advances in diagnostic technology and treatment options that have since been developed. Additionally, this region is a rural area with
EP
potentially prolonged travel times to the closest hospital, which may have had a dramatic impact on
AC C
the survival of these patients.
1.4.4 South West
Only one paper of relevance to this region was found. The authors studied a population of 510 assault victims in Bristol and found a PI caused by a sharp weapon in 7.3% of cases [11]. Lacerations were also observed in 42% of patients. However, the conclusions of the study are limited by the
ACCEPTED MANUSCRIPT exclusion of other common penetrating aetiologies such as road traffic collisions. Furthermore, this article was published over 25 years ago and therefore may not be representative of the current
1.4.5 London
RI PT
incidence of PI. The mortality rate from penetrating trauma in this region also remains unknown.
SC
The epidemiology of PI in London has received the most attention to date, with four
publications in total examining different areas of the capital. The frequency of adult trauma patients
M AN U
presenting with PI was prospectively reviewed at St George’s Hospital (SGH), the Major Trauma Centre (MTC) designated to South West London [20]. Having received 343 patients over a three month period, a total of 39 (11%) had a penetrating mechanism of injury. Another single centre study at London’s North East MTC, the Royal London Hospital (RLH), found that PI constituted 21% of trauma admissions and 13% of mortality [12]. The majority of mortalities in this group (31/40) were
TE D
caused by knife wounds as opposed to firearms or road traffic accidents. There also appears to be a growing number of violent assaults in London. Recent statistics
EP
from the Metropolitan Police revealed that over 4,200 knife-related offences were committed between April and August 2015 [21]. This is a 9.7% increase from the previous year, with 876 occurring in
AC C
August alone. This trend coincides with a 16-year review of the London Helicopter Emergency Medical Service (HEMS) database, which showed an annual rise of 23.2% and 11% in stabbings and shootings respectively [13]. A total of 1,564 (9.7%) of the 16,068 major trauma patients had sustained a deliberate PI over the study period. A further study of deliberate penetrating trauma has been carried out in perhaps the most prolific region of London – the South East. The inner-city borough of Southwark is reported to have one of the highest offence rates in London, with 261 recent knife crimes over a five month span representing an annual increase of 23.7% [21]. A cross-sectional study performed at King’s College
ACCEPTED MANUSCRIPT Hospital (KCH), the MTC in Southwark responsible for the South East London region, established that 938 knife injuries presented to the emergency department in 2011 [9]. The authors estimated the annual incidence rate in the South East population to be 737 per 100,000 persons. They also found that PI consumed a quarter (24%) of the trauma team’s workload despite the fact this group of patients
RI PT
comprised a tiny fraction (0.77%) of the 127,191 attendances. Furthermore, knives were responsible for 94.1% of penetrating trauma cases and assaults accounted for 44% of these presentations, though a
M AN U
SC
suspiciously-high proportion of cases (49%) were attributable to ‘accidents’.
1.5
LIMITATIONS
Two major limitations were identified in this review. Firstly, several of the included studies
TE D
did not report important variables such as mortality rate. This may limit any conclusions that can be drawn regarding mortality rates of PI patients. Furthermore, one study conducted in the Midlands lacked data on sample size and absolute number of penetrating trauma cases, with only the proportion
EP
of PI patients disclosed.
The second and perhaps more influential limitation is the inclusion of studies which evaluated
AC C
major trauma patient populations. These tended to exhibit the highest incidence and mortality rates, which may have skewed results. Our inclusion criteria intended to encompass all published data regarding the frequency of PI, however this has yielded potentially unrepresentative results. The low number of regional studies evaluating PI in general populations indicates the need for further epidemiological research, particularly within trauma networks. This will allow for accurate comparisons to be made between regions.
ACCEPTED MANUSCRIPT 1.6 CONCLUSIONS Penetrating trauma is common and has many medical and emotional consequences alongside
RI PT
the legal, economic and political implications. A systematic review of the current literature has yielded some unexpected results regarding the geographical pattern, incidence, and demographics of PI in the UK. A variance of incidence and mortality rate was observed between studies in all regions, depending on the specific study population under review. This may reflect the inherent variability of
SC
injury severity and outcome in accordance with classification of trauma patients.
M AN U
London was found to have the highest incidence rate of any region. However, the majority of studies originating from London were concerned with a population of major trauma patients. Therefore, the epidemiological data regarding PI in London may be of limited value when evaluating incidence within the general population. Mersey & North Wales, categorised within the Midlands region, was found to have the highest mortality rate of any published article, with similar rates
TE D
observed in London.
A significant majority of patients were male in all included studies which may represent the increased risk of males becoming involved in gangs and violence, as evidenced by the high
EP
contribution of assault to PI. Patients also tended to be fairly young with the exception of one study
AC C
which analysed the TARN database.
In conclusion, ambiguous incidence and mortality rates within regions have resulted in
uncertainty despite analysis of all relevant literature. Furthermore, the low number of studies conducted in each region limits the validity of any conclusions made. Therefore, further original work needs to be completed on the incidence, geographical distribution, and demographics of penetrating trauma including analysis of ethnicity. This should be conducted in the context of regional trauma networks within the UK.
ACCEPTED MANUSCRIPT 1.7 ACKNOWLEDGEMENTS This research received no funding from any agency in the public, commercial, or not-for-profit
M AN U
REFERENECES
SC
1.8
RI PT
sectors.
AC C
EP
TE D
[1] Woodford M, Lawrence T, Bouamra O, Edwards A, Lecky F. Major trauma audit: Metrics for improving care. African Journal of Emergency Medicine. 2013 12//: 3:S10 [2] Sinha CK, Lander A. Trauma in children: abdomen and thorax. Surgery (Oxford). 2013 3//: 31:123-9 [3] Holbrook TL, Hoyt DB. The impact of major trauma: quality-of-life outcomes are worse in women than in men, independent of mechanism and injury severity. The Journal of trauma. 2004 Feb: 56:284-90 [4] Christensen MC, Nielsen TG, Ridley S, Lecky FE, Morris S. Outcomes and costs of penetrating trauma injury in England and Wales. Injury. 2008 9//: 39:1013-25 [5] Brady RR, O'Neill S, Berry O, Kerssens JJ, Yalamarthi S, Parks RW. Traumatic injury to the colon and rectum in Scotland: demographics and outcome. Colorectal disease : the official journal of the Association of Coloproctology of Great Britain and Ireland. 2012 Jan: 14:e16-22 [6] Tambyraja AL, Scollay JM, Beard D, Henry JM, Murie JA, Chalmers RT. Aortic trauma in Scotland--a population based study. European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery. 2006 Dec: 32:686-9 [7] Skipworth RJ, McBride OM, Kerssens JJ, Paterson-Brown S. Esophagogastric trauma in Scotland. World journal of surgery. 2012 Aug: 36:1779-84 [8] Scollay JM, Beard D, Smith R, McKeown D, Garden OJ, Parks R. Eleven years of liver trauma: the Scottish experience. World journal of surgery. 2005 Jun: 29:744-9 [9] Pallett JR, Sutherland E, Glucksman E, Tunnicliff M, Keep JW. A cross-sectional study of knife injuries at a London major trauma centre. Annals of the Royal College of Surgeons of England. 2014 Jan: 96:23-6 [10] McGuffie AC, Graham CA, Beard D, et al. Scottish urban versus rural trauma outcome study. The Journal of trauma. 2005 Sep: 59:632-8 [11] Shepherd JP, Shapland M, Pearce NX, Scully C. Pattern, severity and aetiology of injuries in victims of assault. Journal of the Royal Society of Medicine. 1990 Feb: 83:75-8 [12] Chalkley D, Cheung G, Walsh M, Tai N. Deaths from trauma in London--a single centre experience. Emergency medicine journal : EMJ. 2011 Apr: 28:305-9 [13] Crewdson K, Lockey D, Weaver A, Davies GE. Is the prevalence of deliberate penetrating trauma increasing in London? Experiences of an urban pre-hospital trauma service. Injury. 2009 May: 40:560-3
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[14] Papachristos AV, Hureau DM, Braga AA. The corner and the crew: the influence of geography and social networks on gang violence. American sociological review. 2013: 78:417-47 [15] Gill B. The English Indices of Deprivation 2015: Department for Communities and Local Government, Government of the United Kingdom; 2015 [16] Jivraj S, Khan O. Ethnicity and deprivation in England: How likely are ethnic minorities to live in deprived neighbourhoods?: ESRC Centre on Dynamics of Ethnicity; 2013 [17] Anderson ID, Woodford M, Irving MH. Preventability of death from penetrating injury in England and Wales. Injury. 1989 Mar: 20:69-71 [18] Gray JT, Walker A. “At the sharp end”: does ambulance dispatch data from south Yorkshire support the picture of increased weapon-related violence in the UK? Emergency Medicine Journal. 2009 October 1, 2009: 26:741-2 [19] Gorman DF, Teanby DN, Sinha MP, Wotherspoon J, Boot DA, Molokhia A. The epidemiology of major injuries in Mersey Region and North Wales. Injury. 1995 Jan: 26:51-4 [20] Roche CD, McDonald CM, Liu B, Razik A, Bishop T. Missed orthopaedic injuries in adult trauma patients at a major trauma centre. Trauma. 2015 April 1, 2015: 17:123-7 [21] Metropolitan Police Service: Monthly Knife Crime Summary. 2015 August 2015 [cited 24 December 2015]; Available from: http://www.met.police.uk/foi/pdfs/priorities_and_how_we_are_doing/corporate/monthly_knife_crime _summary_august2015.pdf
ACCEPTED MANUSCRIPT TABLE 1. Sources of epidemiological evidence for penetrating injuries in the United Kingdom (UK).
4
Christensen et al. Anderson et al.17 McGuffie et al.10 Gray & Walker18 Gorman et al.19 Shepherd et al.11 Pallett et al.9 Chalkley et al.12 Crewdson et al.13 20 Roche et al.
Study period
Region
Data Number of source patients
Number of Mortality PI patients (%) rate (%)
EWNI
Apr 2012 - Apr 2013
TARN
37,353
1,158 (3.1)
N/R
EWNI EWNI Scotland Midlands Midlands South West London (SE) London (NE) London London (SW)
Jan 2000 - Dec 2005 1989 2003 - 2005 Apr 1999 - Mar 2008 May 1989 - Apr 1990 Jan - Dec 1986 Jan - Dec 2011 Jun 2004 - May 2008 Jan 1991 - Dec 2006 Jul - Sept 2012
TARN TARN ED YAS ED ED ED ED HEMS ED
~37,000 1,000 4,636 N/R 1,088 510 1,27,191 4,243 16,068 343
1,365 (3.7) 71 (7.1) 405 (8.7) N/R (0.3) 39 (3.6) 37 (7.3) 938 (0.8) 892 (21.0) 1,564 (9.7) 39 (11.4)
8.3 N/A 8.2 N/R 15.4 N/R 0.5 13.2 N/R N/R
RI PT
Study 1 Woodford et al.
AC C
EP
TE D
M AN U
SC
PI, penetrating injury; EWNI, England, Wales & Northern Ireland; SW, South West; NE, North East; SE, South East; N/R, not recorded; N/A, not applicable; TARN, Trauma Audit & Research Network; ED, Emergency Department; YAS, Yorkshire Ambulance Service; HEMS, Helicopter Emergency Medical Service.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT HIGHLIGHTS
EP
TE D
M AN U
SC
RI PT
11 articles reviewed Highest incidence in London Highest mortality in Midlands
AC C
• • •