Injury, Int. J. Care Injured 42 (2011) 488–491

Inner-city gunshot wounds – 10 years on

Matt Davies a,*, Mary Kerins b, Edward Glucksman b

a Trauma Audit and Research Network, University of Manchester, Clinical Sciences Building, Salford Royal Hospital, Eccles Old Road, Salford M6 8HD, United Kingdomb Emergency Department, King’s College Hospital, Denmark Hill, London SE5 9RS, United Kingdom

A R T I C L E I N F O

Article history:

Accepted 30 September 2010

Keywords:

Penetrating injury

Gun

Wound

Trauma

Inner-city

A B S T R A C T

Introduction: In 1997, Injury published one of the first research papers to document the incidence and

characteristics of civilian gunshot wounds in a UK urban environment. Since then there has been concern

that firearm deaths and injuries have increased despite little published clinical evidence.

Methods: We carried out a retrospective survey ten years on from the initial study. All patients

presenting to the Emergency Department (ED) of King’s College Hospital with gunshot wounds from 1st

January 2003 to 31st December 2004 were identified. Information regarding incidence, patient and

injury characteristics and outcome was determined.

Results: 46 patients presented with firearm injuries. 44 were male and the average age was 24 years. The

majority were from minority ethnic groups. 38/46 presented out of hours and the police were

documented to be involved in 36 cases. All injuries were due to assault by low energy projectiles. Of the

32 patients admitted the mean length of stay was 12.4 days. The majority of injuries were to the

musculo-skeletal system. Six patients died from their injuries–5 from head/neck or chest injuries and 1

from intra-abdominal injury.

Discussion: There appears to be little increase in firearm injuries seen over this 10 year period at our

hospital and predominately young, black males continue to be the victims. Most present out of hours,

potentially placing considerable challenges on junior medical staff. Most wounds were to the musculo-

skeletal system perhaps reflecting the desire to maim rather than kill and the absence of high velocity

injuries may reflect the UK’s stringent gun control legislation. The importance of high quality clinical

audit is necessary to effectively plan training, service provision and violence prevention efforts.

Conclusion: Despite public, political and media concerns that deaths and injuries caused by firearms

have increased dramatically, this study finds little change in incidence or characteristics of those injured

and attending an urban ED over a ten year period.

� 2010 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Injury

journa l homepage: www.e lsevier .com/ locate / in jury

Introduction

In 1997, Injury published one of the very first research articlesdescribing the incidence, injury characteristics and epidemiologyof patients injured through gunshot wounds in the UnitedKingdom.14 This single-centre study of all such wounds admittedto a major London teaching hospital over a two year period (1993–1995) was undertaken as there was concern at that time thatfirearm violence in UK urban populations was increasing. In fact,the study showed that the number of patients involved was low, 42out of 147,012 (0.029%) total Emergency Department (ED)attendances but primarily involved young black males, eight ofwhom died.

A number of very high profile murders and serious injuries overthe past few years may give the impression that injuries and deaths

* Corresponding author. Tel.: +44 0161 206 4397.

E-mail address: drmattdavies@btinternet.com (M. Davies).

0020–1383/$ – see front matter � 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.injury.2010.09.041

sustained through gunshot wounds have increased dramatically.Home Office statistics demonstrate an increase in crimes recordedby police in England and Wales involving firearms that causedinjury (excluding air guns) from 597 in 1992 to 2367 in 2003/2004.15 Nearly two-thirds of all non-air weapon firearm offences inEngland and Wales occurred in just three police authorities:Metropolitan (Greater London), Greater Manchester and WestMidlands.15 Table 1 demonstrates a considerable increase in thenumber of firearm offences and injuries that were reported to theMetropolitan Police over a 20 year period.12

Since 1997, there has been little medical research publishedabout UK civilian firearm injuries. Such data is important as theprovision of comprehensive and accurate information about injuryhas been a key element in stimulating injury prevention initiativesand also helps plan service provision and has implications fortraining emergency and surgical doctors of the future. Other UKstudies have shown that the incidence of injuries seems to beincreasing, primarily affecting young males from minority ethnicgroups disproportionately, who present out of hours (OOH).13,2

Table 1Offences in which firearms were used and were reported to the Metropolitan police

(moderate injury was added in 2001).

1983 1993 2003

Fatal 6 19 32

Serious 32 33 236

Moderate 671

Minor 404 213 1473

Total 442 265 2412

[()TD$FIG]

Fig. 1. Anatomic locations of gunshot wounds.

M. Davies et al. / Injury, Int. J. Car

There is strong evidence of a link between firearm-related offendingand the trade in illegal drugs, gang culture and areas with higherthan average levels of deprivation and unemployment.8 Low energytransfer wounds from hand-guns and shot-guns (as opposed torifles) seem to predominate and many can be managed conserva-tively or on an out-patient basis.2 However, a large proportion offirearm injuries require a variety of surgical skills that can be achallenge to the most experienced surgeon and represent aconsiderable financial cost to the NHS.2

We carried out a retrospective survey, ten years on from thisoriginal study, to examine the incidence, injury characteristics andepidemiology of those who presented to a major London teachinghospital with firearm injuries. This study provides clinical evidencefrom an under-utilised data set about this mechanism of injury.

Ethical committee approval was not required for this study.

Methods

King’s College Hospital is a 950 bed teaching hospital serving aninner city population of 250,000. As well as providing generalhospital services for the London boroughs of Lambeth, Southwarkand Lewisham it is also a tertiary centre for Neurosurgery,Cardiothoracic and Maxillo-facial surgery. King’s is also a receivinghospital for the Helicopter Emergency Medical Service (HEMS).

All patients sustaining gunshot wounds (excluding air-guninjuries) that presented to the Emergency Department (ED)between January 1st 2003 and January 1st 2005 were identifiedthrough a retrospective review of the department’s traumadatabase, resuscitation department log-book and the hospitalpatient database using relevant ICD 10 codes (X72, X73, X93, X95,W32, W34, Y22, Y23 and Y24). The hospital’s business intelligenceunit was asked to provide total ED attendance figures for 1993/1994 and 2003/2004.

Patients’ ED cards and hospital notes were then analysed toverify eligibility and provide the following information:

Patient demographicsMode and timing of arrivalInjury intentPolice involvementAnatomical locations of wound/s and treatmentInitial observations (GCS, SBP and RR)Length of stay (LOS)ICU admissionSurvival to discharge

Each patient’s Revised Trauma Score1 (RTS) (based on first setof assessments in the ED) was calculated using the followingformula:

RTS ¼ 0:9368 GCSþ 0:7326 SBPþ 0:2908 RR ðGCS

¼ Glasgow Coma Score; SBP ¼ Systolic Blood Pressure; RR

¼ Respiratory RateÞ

Results

46 out of 201,680 (0.023%) patient attendances were recordedas being seen at King’s College Hospital ED with gunshot wounds.Of these, 44 were male and 2 female with a median age of 24 years(IQR 19–32). One patient was 15 years of age.

Twenty-three patients were documented to be of African-Caribbean origin, four were Black British, three Black ‘other’, five‘other’, four Caucasian, three ‘unknown’ and one Asian. The policewere involved in 36 cases whilst, in the other ten, policeinvolvement was not documented.

The majority of patients (38) presented ‘out of hours’ (OOH)(outside 08.00–18.00 Monday–Friday), with the highest numberseen Thursday–Saturday (27). Each month saw at least 1 gunshotwound with July (7), September (6) and January (6) recording thehighest number of visits and November the least (1). Most patientsarrived via the London Ambulance Service by land transfer (27),though 5 arrived by the Helicopter Emergency Medical Service(HEMS). Of the others, 6 arrived by private transport, 3 walked inand 2 used by public transport. In 3 cases the mode of arrival wasunknown.

All injuries were believed to be due to low energy projectiles. 32patients were shot once and 12 were hit by multiple bullets (2cases were not documented). Four patients were shot twice, 4three times, 1 five times and 1 six times. In 2 cases the number ofbullet wounds was documented as ‘multiple’ but not accuratelystated. In some of the cases the bullet passed through more thanone anatomical area (Fig. 1).

The mean RTS from data available on 45/46 patients was 6.87and all injuries were alleged assault. Six patients died. 8 patientswere discharged the same day and in 2 cases, disposal wasunknown. Of the 32 patients who were admitted, the average LOSwas 12.4 days. 6 patients in total received care at some point on theIntensive Care Unit (ICU).

Injuries

Fatalities

Six patients died from their injuries, four of them in the ED, onein theatre and one on the ICU after two days. Of these, two patientssuffered single gunshot wounds to the chest, one of which had anemergency thoracotomy at scene by HEMS whilst the other was inasystolic cardiac arrest at scene. A further two patients hadwounds to the chest but these were combined with wounds toother areas. One patient was shot six times and involved the chest,

e Injured 42 (2011) 488–491 489

M. Davies et al. / Injury, Int. J. Care Injured 42 (2011) 488–491490

neck and left shoulder the other patient was shot three timesinvolving the neck and chest. Both patients were asystolic at scene.Other than chest injuries, one patient died after being shot in theright lumbar region and again was asystolic at scene whilst theother fatality received a gunshot wound to the right temporalregion with a large exit wound to the left parietal area, consistentwith a catastrophic brain injury.

Head + neck

Eleven patients in total had head and neck injuries. Three died.Four patients required simple debridement of gunshot pellets fromthe face and scalp. One patient sustained deep lacerations to the lefttemporal region that required ligating the temporal artery, thoughthe skull was not breached. Another patient was shot above the leftsupra-orbital ridge, fracturing the temporal bone and lodging in thetemporal lobe. This required craniotomy and debridement. Maxillo-facial input was required with three patients – one with a bulletlodging in the right maxilla, a bullet entering the right retromolartrigone and exiting the left tongue base and a further patient with agunshot wound to the base of the left neck where the patientultimately required a median sternotomy for a left haemothorax andcervical oesophageal rupture.

Chest

Fifteen patients had gunshot wounds to the chest. Four died.Four patients sustained superficial wounds from gunshot pelletswith no damage to underlying structures. Two patients solelyrequired placement of chest drains, no attempt being made toremove the bullet. One patient required a median sternotomy for aleft-sided haemothorax and ruptured oesophagus. In the remain-ing four patients the bullet caused minimal damage to thoracicorgans but passed through the diaphragm into the abdomenrequiring exploratory laparotomies.

Abdomen + pelvis

Thirteen patients presented with injuries to the abdomen/pelvis, one of whom died. Two sustained superficial injuries fromgunshot pellets that required only debridement, though onesubsequently required plastic surgery. Two sustained injuries totheir genitals, requiring exploration and in one case removal of anobliterated testicle. Of the remaining eight patients, all under-went laparotomy, two underwent hemi-nephrectomy, fourunderwent right hemi-colectomy, one required repair of a liverlaceration and one a Hartmann’s procedure following tears to themid-ileal mesentery and damage to the left iliac artery.

Limbs

23 patients presented with gunshot wounds to the musculo-skeletal system. One died from additional injuries. Three weretreated non-operatively with antibiotics alone. Ten required localdebridement (2 under local anaesthetic) whilst five required formaloperative exploration and debridement. Two patients sustainedinjuries to the knee and required arthroscopy and wash-out of metalfragments but no fracture was detected in either case. One patientrequired an intra-medullary nail for a comminuted femoral mid-shaft fracture, and two patients underwent debridement ofcompound wrist fractures that were initially left open.

Discussion

Our study shows only a very small increase in the actualnumber of firearm injuries seen over a similar 2 year period 10years on (46 vs 42). However, as a proportion of all ED attendances,the number has fallen. This data is consistent with an analysis of allfirearm injuries on the UK Trauma Audit and Research Network(TARN) database between 1998 and 2008 that demonstrated

annual peaks and troughs in the number of people seriouslyinjured through the use of firearms in the UK, but no sustainedincrease over a ten year period.3

The demographic of young black males being primarilyinvolved remains the same and this is reflected in other UK andinternational studies.2,10,11,13,16 According to Metropolitan Policedata for the period between April 2001 and October 2005, 63% ofthe victims of murder and attempted murder involving firearms inLondon were black.8

The majority presented using land ambulance. All were assaultsby low energy projectiles and this perhaps reflects the UK’s tightgun control laws on high velocity weapons and the fact that self-inflicted injury by firearm has an extremely high case fatality rate9

and are unlikely to survive to be taken to hospital. Most presentedOOH reflecting ethnographic analysis of when violence occurs11

this also highlights the added pressure on emergency and surgicalservices when senior staff are perhaps not immediately available. Itis of note that 4 of the 8 deaths in this study occurred in the ED (allOOH). The 2007 National Confidential Enquiry into PatientOutcome and Death (NCEPOD), ‘Trauma: Who cares?’, recom-mended that a consultant must be team leader for the manage-ment of the severely injured patient,4 although one study found norelationship between outcome and time of admission.7

Of those that died (6/46), 5 died from injury to the head and neckor chest region with the exception of 1 who died from intra-abdominal injuries. This is consistent with the initial study where 8/42 died, all from head and neck or chest injuries.14 Most woundswere peripheral and musculo-skeletal in location, perhaps suggest-ing the use of firearms to intentionally maim rather than to kill. Thisis reflected in generally low RTS scores of the cohort and the fact thata large minority of patients (14/46) were discharged after initialtreatment in the ED. This is consistent with other UK studies.2

It was noted in the previous study that the police were notinvolved in a substantial minority of cases (5/42 with 9/42 notdocumented) and raised the question of under-reporting of guninjury and crime.14 In our study, in 10/46 cases police involvementwas not documented in the notes, again raising the same concerns.Since 2003 the General Medical Council recommended that allfirearm injuries are reported to the police5; and more recently itrequires doctors to inform the police or social services wheneverthey treat a patient who is a victim of gun or knife crime6;however, there is no evidence to confirm that this is happeningconsistently.

This advice regarding reporting may be acting as a deterrent tomany of the less seriously injured victims of firearm crime seekingmedical attention in the first place, leading to under-reporting ofthe true scale of the problem. Additionally, fear of reprisal is welldocumented and for those involved in gangs, it may bepreferential to carry out revenge shootings and take matters intotheir own hands rather than involve the criminal justice system.16

There are some potential limitations with this study. It is asingle centre study over a short period of time. The identification ofpatients who met the inclusion criteria was dependant on accurateclinical coding of injuries for retrieval of relevant notes. Also, not allfirearm injuries that occur in the hospital’s catchment area willautomatically be brought to this hospital. The surrounding area iscovered by multiple hospitals and some patients may betransported by HEMS to the Royal London Hospital. Therefore, itis not possible to extrapolate our results to give a completeindication of what is happening locally or throughout London.With the recent development of the London Trauma Networks andassociated improved trauma audit we will hopefully be betterplaced to provide accurate pan-London statistics in the future.

It is also important to note that fatalities caused by firearmsmay not present to the ED and may be placed immediately underthe jurisdiction of the forensic pathologist and transferred directly

M. Davies et al. / Injury, Int. J. Care Injured 42 (2011) 488–491 491

to the mortuary. Consequently, this may lead to under-reporting ofgun fatalities in this study.

Conclusion

Gun crime has become a pressing public and political issue.However, ten years on from the publication of one of the firstarticles addressing firearm injuries in London, we have shown noappreciable change in the incidence, epidemiology or injurycharacteristics of those involved.

Conflict of interest statement

All authors declare no competing interests.

Funding

MD’s salary was part funded by the National Institute of Health(NIH). They played no further part in the study and the viewsexpressed are those of the researchers, not the NIH.

Data

All authors, external and internal, had full access to all of thedata (including statistical reports and tables) in the study and cantake responsibility for the integrity of the data and the accuracy ofthe data analysis.

References

1. Champion HR, Sacco WJ, Copes WS, et al. A revision of the trauma score. JTrauma 1989;29(5):623–9.

2. Cowey A, Mitchell P, Gregory J, et al. A review of 187 gunshot wound admissionsto a teaching hospital over a 54-month period: training and service implica-tions. Ann R Coll Surg 2004;86(2):104–7.

3. Trauma Audit and Research Network, personal communication.4. Findlay G, Martin IC, Carter S, et al. Trauma: who cares?. National Confidential

Enquiry into Patient Outcome and Death; 2007.5. General Medical Council. Reporting gun shot wounds. Guidance for Doctors in

Accident and Emergency Departments. London: General Medical Council; 2003.6. General Medical Council. Confidentiality. London: General Medical Council;

2009.7. Guly HR, Leighton G, Woodford M, et al. The effect of working hours on outcome

from major trauma. Emerg Med J 2006;23:276–80.8. Hales G, Lewis C, Silverstone D. Gun crime: the market in and use of illegal

firearms. Home office research study no. 189; 2006.9. Haw C, Sutton L, Simkin S, et al. Suicide by gunshot in the United Kingdom: a

review of the literature. Med Sci Law 2004;44:295–310.10. Hemenway D. Private guns, public health. Ann Arbor: University of Michigan

Press; 2004.11. Hemenway D. The epidemiology of US firearm injuries. J Public Health Policy

2003;24:380–5.12. Metropolitan Police, personal communication.13. Persad IJ, Reddy R, Saunders M, Patel J. Gunshot injuries to the extremities:

experience of a U.K. trauma centre. Injury 2005;36(March (3)):407–11.14. Porteus MJ, Edwards SA, Groom AF. Inner-city gunshot wounds. Injury

1997;28(June–July (5–6)):385–7.15. Povey D (ed), Coleman K, Kaiza P, Roe S. Homicides, firearm offences and

intimate violence 2007/08 (supplementary volume 2 to crime in England andWales 2007/08). Home Office; 2009.

16. Squires P. ‘Gun crime’. A review of evidence and policy. Centre for Crime andJustice, King’s College London; 2008.