Ten Years' Experience With Pediatr ic Gunshot Wounds

By Barbara Barlow, Maria Niemirska, and Rajinder P. Gandhi N e w York City, N e w York

�9 Gunshot wounds in children have become a significant source of morbidity and mortality in our community in the last 10 yr. One hundred eight children, 16 yr of age and younger, were admitted to the Pediatric Surgical Service for gunshot wounds during this period; only 1 child was admitted for a gunshot wound in the 10 yr preceding this review. Rapid resuscitation and triage of major injuries directly to the operating room achieved a 94% sur- vival, Review of the circumstances of injury revealed that 4 2 % of the gunshot wounds were inflicted by children and 40% were known to have been inten- tional. Only 20% of the patients had known drug involvement; in general this was involvement in drug selling, not in drug abuse. Social service intervention can offer significant benefit to these children, but ultimately gun control laws with strict enforcement are needed to stop this type of violence toward children.

INDEX WORDS: Gunshot wounds; penetrating trau- ma.

C I V I L I A N gunshot wounds occur in societies with laxity toward gun control. When hand

guns are easy to obtain and are kept in many homes, gunshot wounds in children appear as a source of morbidity and mortality.

The available medical and social information on 108 children admitted to the Pediatric Surgi- cal Service of Harlem Hospital for gunshot wounds over the last decade were reviewed. Social data were not available for the early cases since this was considered privileged information and was not included in the chart. Seventy-five complete social records were available for chil- dren with gunshot wounds and these were com- pared to the social records of children admitted during the same period for simple fractures. These 108 patients represent only 80% of the children in this age group admitted for gunshot wounds since 15- and 16-yr-old children were also admitted to the General Surgical Service. Gunshot wounds are a new source of morbidity and mortality in our community: only 1 child with a gunshot wound was admitted to our service in the 10 yr preceding this review. This phenomenon appears directly related to the ease of obtaining hand guns in our area as well as to the change in the adult drug offender laws in New York State.

PATIENT GROUP

The age and sex distribution of the 108 children admitted for gunshot wound is presented in Fig. 1. Eighty-three percent of the children were male; 82% were between 12 and 16 yr of age. Fifty-six percent of the gunshot wounds occurred between 4 p.m. and midnight; 30% occurred during the summer months.

INJURIES

Gunshot wound sites for the 108 children are listed in Table I. Extremity wounds were the most common injuries. The major injuries sustained by these patients are listed in Table 2. Orthopedic injuries were the most frequent followed by pulmonary and neurologic injuries.

MANAGEMENT

Rapid transportation to the hospital, effective resuscita- tion, and immediate surgical intervention, reduces the mor- tality and morbidity in gunshot wounds. The majority of our patients reached the hospital within 30 min of injury; 30% by police car, 29% by ambulance, 25% by taxi, 10% by private car, and 6% on foot. After initial evaluation and resuscitation major injuries were tr iaged directly to the operating room. Radiologic evaluation was undertaken only in stable patients. In our hospital the emergency room staff is backed up by a full surgical team with all subspecialists available for rapid surgical treatment. All patients were operated on by the resident staff under attending supervision. The average time for arrival to surgery for penetrating head wounds was 71 min, ranging from 35 rain to 1 hr 45 min; for abdominal and thoraco-abdominal wounds, 42 min, ranging from 5 rain to 2 hr 20 min; and extremity wounds, 73 min, ranging from 5 min to 3 hr 50 rain.

Penetrating head wounds were intubated in the emergency room and started on mannitol, decadron, and antibiotics. Craniotomy was performed to debride necrotic brain tissue, remove bullets and bullet fragments and close the dura.

Face and neck wounds were evaluated for major vessel, esophageal, and tracheal injury. Only cases with such injury were explored, Angiography was used to evaluate gunshot wounds passing close to major vessels when clinical examina- tion did not confirm injury. This conservative approach to

From the Harlem Hospital Center, College of Physicians & Surgeons of Columbia University, New York, N.Y.

Presented before the Thirteenth Annual Meeting of the American Pediatric Surgical Association, Phoenix, Arizona, May 29-June 1, 1982.

Address reprint requests to Barbara Barlow, M.D., Chief, Pediatric Surgery, Harlem Hospital Center, 506 Lenox Ave., Room KP-17103, New York, N.Y. 10037.

�9 1982 by Grune & Stratton, Inc. 0022-3468/82/1706-49037501.00/0

Journal of Pediatric Surgery, Vol. 17, No. 6 (December), 1982 927

928

P e d i a t r i c G u n s h o t W o u n d s

"~ I 9 0 # 3~ 18

~ 3o

~ 25 ~ ~o

I 2 5 4 .5 6 7 8 9 I 0 I1 12 13 14 t5, 16

Age in Years

Fig. 1. Age and sex distribution of 108 children with gunshot wounds.

penetrating wounds of the neck avoided unnecessary surgery without complication. Penetration of the hypopharynx with- out evidence of additional injury was treated with observation and antibiotics. Endoscopy was used to rule out esophageal injury. Tracheostomy was only required in one patient with a tracheal injury.

All gunshot wounds to the chest resolved on tube thoracos- tomy. Indications for surgery were uncontrolled hemorrhage and cardiac tamponade; the only patient with such an injury presented DOA and could not be resuscitated.

All abdominal and thoraco-abdominal gunshot wounds were explored. One patient had a negative laparotomy with only a retroperitoneal hematoma foUnd a t surgery. Liver injuries were treated with suture control of bleeding and drainage. Liver resection and common duct decompression were not required in any of our patients. All gastrointestinal perforations, stomach, duodenum, small intestine and colon, were closed primarily without diversion. None of these clo- sures leaked. Injuries to the body and tail of the pancreas were oversewn and sump drained. One gunshot wound that destroyed the head of the pancreas, the ampulla of Vater, and the distal common bile duct, was treated with a modified Whipple's procedure and drained with sump drains. The patient now has normal glucose tolerance and has gained his preoperative weight without requiring dietary restrictions. Gall bladder perforation was treated with cholecystectomy. All retroperitoneal hematomas were explored. Injuries to the vena cava were sutured. The one transected ureter was

Table 1. Gunshot Sites in 108 Children

Site Number of Cases

Head 11

Face/neck 17

Chest 16

Thoraco-abdominal 6

Abdomen 11

Back 5

Extremity 42

BARLOW, NIEMIRSKA, AND GANDHI

Table 2. Major Injuries in 108 Pediatric Gunshot Wounds

Neurologic 20

Brain 9

Spine 6

Brachial plexus 4

Peripheral nerve 1

Orthopedic 35

Bone 27

Joint 6

Tendon 2

Vascular 7

Aorta 1

Popliteal artery 1

Femoral artery 2

Jugular vein 1

Vena cava 2

Pulmonary 28

Pharynx 3

Trachea 1

Lung 18

Heart 1

Diaphragm 5

Gastrointestinal 15

Esophagus 1

Stomach 4

Duodenum 2

Small intestine 4

Colon 4

Biliary-pancreatic 14

Liver 8

Gall bladder 1

Common duct 1

Pancreas 3

Pancreatic duct 1

Urologic 3

Kidney 2

Ureter 1

treated with primary anastomosis over a stent. Renal paren- chymal injury was treated with suture for hemostasis and drainage.

Gunshot wounds of the spine were observed. Myelogram was performed for deteriorating clinical findings. No patient had a decompression laminectomy.

All major vessel injuries were treated by debridement and reanastomosis. Only one femoral artery injury required a graft to bridge the gap. All repairs were successful.

Gunshot wounds to bone were treated as compound frac- tures with exploration, debridement, and antibiotics. Internal fixation was used where required. Joint injuries were explored and debrided with closure of the joint capsule.

All gunshot wounds were admitted to the hospital even if the injury was felt to be superficial. Broad spectrum antibiot- ics, with tract debridement and removal of accessible bullets was the usual treatment since low velocity bullets do not sterilize the bullet tract resulting in bullet tract infections and abscesses around retained bullets. This is particularly com- mon when the bullet has gone through clothing or through the bowel.

PEDIATRIC GUNSHOT WOUNDS 929

Table 3. Average Length of Hospitalization for 108 Children With Gunshot Wounds

Average Average Hospital

Injury Site No. ICU Days Days

Head 11 14 24 ,Face/neck 17 3 15

Chest 16 2 10 Thoraco-abdominal 6 24 59 Abdomen 11 6 18

Back 5 0 41 Extremity 42 1 11

H O S P I T A L I Z A T I O N

Table 3 shows the average length of intensive care treat- ment and average length of total hospitalization for the different types of gunshot injuries. Extremity and chest wounds had the shortest hospitalization.

D E A T H , C O M P L I C A T I O N , S E Q U E L A E

Six percent of the children died of gunshot wound. Five patients with penetrating head wounds died: three were DOA and two were decerebrate on admission and later expired. One child with a gunshot wound to the chest was DOA. One child admitted for an abdominal gunshot wound with multi- ple organ injury and injury to the aorta, vena cava, and spine, arrested in the emergency room and died during surgery from irreversible hypotension secondary to exsanguination.

Complications and sequelae are listed in Table 4. The complication rate was 14%; 6% of the patients had major complications, 8% had permanent sequelae, and 7% had temporary deficits.

Removal of the chest tube at 3 days was associated in two patients with reaccumulation of hemothorax requiring chest tube reinsertion at 5 and 8 days post injury, although chest tube drainage had stopped for 24 hr prior to the initial removal. Two patients with thoraco-abdominal gunshot wounds through full s tomachs had gastric contents in the chest tube drainage. Both patients developed empyemas necessitating open drainage. Thoracotomy with cleansing of

the chest cavity at the initial operation is mandatory in such cases.

One patient developed a fecal fistula from the right colon in an area where contusion without perforation was seen at the initial operation. Although the fistula resolved without secondary operation, such contused areas should be oversewn at initial operation.

Bullet tract infections or abscesses around a retained bullet were the most frequent complications in spite of routine tract debridement and removal of accessible bullets. There were no wound infections, intraahdominal abscess, or osteomyelitis.

Both patients with intestinal obstruction presented after initial discharge, 3 and 4 mo postoperatively. One responded immediately to long tube drainage; the other was explored with lysis of adhesions for a segmental volvulus.

Two patients received intravenous nutrition: the patient with the modified Whipple's procedure and one of the patients who developed empyema.

Major sequelae were all neurologic: 4 children with para- plegia, 1 quadriplegic, and 1 child with a mild left hemipare- sis.

C I R C U M S T A N C E S OF S H O O T I N G

The circumstance of shooting is known for 88 of the children, Sixty percent were considered accidental; 40% were intentional (see Fig. 2). Sixty-five percent were shot by known assailants. Thirty-five percent were shot in street shoot-outs, by street assailants who were not seen or through open windows in their homes. There is no way to estimate how many children fired the gun in these 35% of cases, nor do we know that all these cases were accidental. Sixty-five percent of the known assailants were children. Fifty-three percent of the guns are known to have belonged to children, usually without parental knowledge. Only 8% of the guns belonged to parents or relatives; 6% of the wounds were inflicted by relatives.

The police intentionally shot 8% of the children during crimes or street altercations, although none of the children were armed. Two children were running from street shoot- outs in which they were not involved.

Six wounds were self-inflicted; only one was a suicide attempt, the rest were accidental.

Table4. Complications and Sequelaein 108 Children With Gunshot Wounds

Injury Sites Cases Complications Permanent Sequelae Temporary Deficits

11 1 Brain abscess 1 (L) Hemiparesis 1 (R) Hemiplegia 1 Wound seroma 1 Global aphasia

17 1 T-M joint adhesion 4 Hypoesthesias 1 Bullet abscess 1 Paresis of arm extensors

16 2 Secondary chest tubes 6 2 Empyemas

11 1 Fecal fistula 2 Intestinal obstructions

1 Thrombophlebitis

Head

Face/neck

Chest Thoraco-abdominal

Abdomen

Back

Extremity

Total %

1 Horner's 1 Quadr ip legia

1 Paraplegia

1 Paraplegia

5 2 Paraplegias 42 2 Bullet tract infections 1 Growth arrest of finger

1 Tendon entrapment 100% 14% 8% 7%

930 BARLOW, NIEMIRSKA, AND GANDHI

Accidental

Assailant in 88 Pediatric Gunshot Wounds

40% Intentional

Child 7a~/o Street Assailant t 1 0%

/ 35% Unknown Assadant

Fig. 2. Assa i lants in shooting of 88 pediatric patients with gunshot wounds.

In 61 of the 88 cases only the patient was shot. In 13 cases there were 33 additional adults and children shot, including 9 people dead at the shooting scene. Fourteen children were injured in street gun fights where there is no record of the number of individuals injured.

Hand guns produced the majority of these wounds; there was one air rifle wound and one shotgun wound.

SOCIAL EVALUATION

Complete social histories were available for 75 children with gunshot wounds. These histories were compared to the social histories of 75 children admitted during the same period for simple fractures.

The family composition was similar for the two groups. Thirty-seven percent of the gunshot and 39% of the fracture patients were from two-parent families; 54% of the gunshot and 53% of the fracture patients were from single-parent households, 9% of the gunshot and 8% of the fracture patients lived with relatives. None of the children were in foster care.

Children with gunshot wounds were more likely to have self-supporting families; 57% of the gunshot and 39% of the fracture patients came from self-supporting families; 41% of

Table 5. Fami ly Deaths in Pat ien ts W i t h F rac tu re and Gunshot Wounds

Family Deaths Family Deaths Fracture Patients Gunshot Patients

Relative (75) (75)

Mother 4 6

Father 4 10 Stepparent 0 3

Sibling 1 4

Violent death 1/9 15/23

the gunshot and 61% of the fracture patients' families were on welfare.

Our initial impression that children with gunshot wounds rarely came from the group of children who receive primary care at Harlem Hospital was supported by the review. Eighty percent of the gunshot patients were first seen at our hospital for this injury and only 40% are now our regular clinic patients. Forty-five percent of the fracture patients were ne~ to our hospital but 73% are now followed in our clinics.

Of particular interest was the unusual number of violent deaths which had occurred in the immediate family of the children admitted for gunshot wounds. This was not found in the families of children admitted for fractures (see Table 5).

There was no difference in the history of drug involvement among the parents of the gunshot and fracture patients. Three percent of the children admitted for fractures had drug involvement--both angel dust abusers. Twenty percent of the children admitted for gunshot wounds had known drug involvement, one heroin snorter, one heroin addicted, and the rest involved in street selling of drugs but not drug abuse.

GUNSHOT WOUNDS AND DRUG LAWS

In 1973 N e w Y o r k S t a t e c h a n g e d the d r u g laws so tha t an adu l t a r r e s t ed for d r u g possession

rece ived a m a n d a t o r y j a i l s en tence o f 1 yr to l i fe

for a smal l a m o u n t and 15 yr to l ife for over 1 oz

( 1 9 7 3 - 1 9 7 9 ) , now over 2 oz (1979 to p resen t ) .

T h e F a m i l y C o u r t has j u r i sd i c t i on over j uve -

niles 16 and y o u n g e r for acts t ha t wou ld be

j u d g e d c r imes i f c o m m i t t e d by adul ts . S o m e 13

to 16 yr olds a r e t r i ed as adul ts , bu t not for d r u g

possession, d r u g sel l ing, o r gun possession, wh ich

a re hand l ed by F a m i l y C o u r t . P u n i s h m e n t for

such offenses a r e p roba t ion or s h o r t - t e r m incar -

ce ra t ion in a j uven i l e home . ~

T h e s e laws offer one e x p l a n a t i o n for adu l t s

r ec ru i t i ng ch i ld ren f r o m 12 to 16 yr o f age to sell

d rugs on the s t reets . T h e s e ch i ld ren w o r k in

g roups , a re a r m e d wi th h a n d guns by the adu l t

d r u g dea le r , and t a u g h t t h e code of behav io r for

d r u g sel l ing. T h e y can e a r n f r o m $150 to $2 ,000

per w e e k d e p e n d i n g on sales and expe r i ence .

M o s t o f these ch i ld ren d r o p ou t o f school; 19% of

our pa t i en t s w i th gunsho t wounds were no longer

a t t e n d i n g school. T h e s e ch i ld ren a re in cons t an t

d a n g e r o f v io len t in ju ry and the i r gun possession

puts un invo lved ch i ld ren a t r isk as well .

T h e Socia l S e r v i c e D e p a r t m e n t and the Divi -

sion o f Ch i ld P s y c h i a t r y a t H a r l e m H o s p i t a l

work c losely wi th t he P e d i a t r i c S u r g i c a l S e r v i c e

in e v a l u a t i n g the ch i l d r en a d m i t t e d for gunsho t

wounds . W e m a k e eve ry a t t e m p t to ge t ch i l d r en

b a c k in school and suppor t t h e m in c h a n g i n g

PEDIATRIC GUNSHOT WOUNDS 931

their dangerous life style. Three children who did not receive such attention returned with second gunshot wounds and a fourth committed suicide 1 wk after discharge. We are not uniformly successful but we do have many former patients

back in school, attending technical schools, or colleges--heading toward productive lives.

REFERENCES l. Allen Clark esq. Attorney at Law. Personal communi-

cation

Discuss ion

,4. Haller (Baltimore): I think the bottom line is not just in the children's gunshot injuries but the necessity for all of us to participate in more active hand gun controls throughout the country. Only 18% of her patient group were less than 12 yr of age, so she is essentially talking about young teenagers. Another important point is that all of these patients were taken for initial resusci- tation and management directly to the operating room. If there is not a trauma center within the hospital, this is the best way to handle resuscita- tion and management because of the implica- tions of immediate operative intervention. I was concerned that the colon perforations were closed. If there are multiple injuries within the abdomen I would think that exteriorization would be a better approach to the management of colonic injuries. There is an increase in the number of children: has there been any change in your referral pattern into the Harlem Hospital? Possibly the identification of a regional trauma center might be bringing more of these children in. The second question: Is there any advantage in excluding some of the older teenagers from a Children's Center? We have felt that if we brought all the drug users and homocides into our children's center we would destroy the over- all environment for children and their families, and have therefore organized an adolescent unit which is a part of, but in many ways separate from, the children's center.

Anthony Shaw (Duarte, Calif.): Dr. Barlow has asked me to discuss this significant paper because I was her predecessor as Chief of Pediat- ric Surgery at Harlem Hospital. Dr. Barlow's paper illustrates how an experienced trauma team can approach a wide variety of traumatic lesions in a systematic way that not only mini- mizes morbidity and mortality of the injury itself, but also avoids the morbidity of unneces- sary or excessive diagnostic procedures. I have a

few questions. First of all, the greatest number of wounds were of the extremities. In my experience with gunshot wounds over the past 11 yr, I observed that most of the extremity wounds were accidentally self-inflicted, and I wonder whether the extremity wounds fell into that category in this series. Second, in Los Angeles, teenage gang members shoot rival gang members. I wonder whether a significant number of Dr. Barlow's injuries were gang related. Third, was there also an increase in the number of other violent inju- ries such as stabbings seen at Harlem Hospital in the past 10 yr? Finally, I wonder if Dr. Barlow can tell us how she persuades a child making $2,000 a week to go back to high school.

G. IV. Shaftan (New York City): I practice at the Kings County Hospital in Brooklyn. This notorious area has, in the past several years, been substituting guns for knives as the weapon of aggression and pacification. Like Dr. Barlow, we have seen an increasing number of youthful injuries, primarily drug related when they have been deliberate. The leader of the major cocaine distribution ring in Brooklyn was 15 yr old and has three admissions in a 4-wk period for gunshot wounds. He died 2 wk after his last hospital discharge. We have used the policy of selective conservatism both in abdominal gunshot wounds and torso gunshot wounds, evaluating the patients on the basis of clinical signs. Peritoneal lavage has been used when the clinical signs did not demand exploration and we have regretted this policy only once in 26 yr. I would ask Dr. Barlow what her feeling is about the use of clinical observation and/or tap and lavage in the assessment of the core injuries secondary to gunshot wounds and, in a similar vein, the utility of angiography in extremity wounds in deter- mining the necessity for operation in children.

B. Barlow (closing): Our referral area, Cen- tral Harlem, has not changed. The Pediatric

932 BARLOW, NIEMIRSKA, AND GANDHI

Service has an adolescent ward which is impor- tant in managing the older children. Those chil- dren with major injuries are more easily per- suaded to alter their life style and return to school than those with minor injury. We have admitted approximately 70 children with stab wounds during the last decade but we have not reviewed the preceding 10 yr for stab wound admissions, so I cannot say if all violent injuries to children have increased. None of the gunshot wounds occurred during gang fights and only a few involved children were gang members. As

expected, extremity wounds were usually acci- dental, whereas wounds to the head, chest, and abdomen were mainly intentional. Colon closure for low velocity bullet wounds has been success- ful in adults and children at our institution and at other institutions. Since gunshot wounds of the abdomen usually produce injuries that require surgical correction, we feel that immediate lapa- rotomy without observation is essential in reduc- ing the morbidity and mortality of this type of trauma.