Clinical and legal significance of fragmentation of bullets in relation to size of wounds: retrospective analysis

BMJ 1999; 319 doi: (Published 14 August 1999) Cite this as: BMJ 1999;319:403
  1. Robin Coupland, surgeon (rcoupland{at}
  1. Unit of the Chief Medical Officer, International Committee of the Red Cross, 19 avenue de la Paix, 1202 Geneva, Switzerland
  • Accepted 7 April 1999


Objective: To examine the relation between fragmentation of bullets and size of wounds clinically and in the context of the Hague Declaration of 1899.

Design: Retrospective analysis of prospectively collected data on hospital admissions.

Setting: Hospitals of the International Committee of the Red Cross.

Subjects: 5215 people wounded by bullets in armed conflicts (5933 wounds).

Main outcome measures: Grade of wound computed from the Red Cross wound classification and presence of bullet fragments on radiography.

Results: Of the 347 wounds with fragmentation of bullets, 251 (72%) were large wounds (grade 2 or 3)—that is, those with a clinically detectable cavity. Of the 5586 wounds without fragmentation of bullets, 2915 (52.1%) were large wounds. Only 7.9% (251/3166) of large wounds were associated with fragmentation of bullets.

Conclusions: Fragmentation of bullets is associated with large wounds, but most large wounds do not contain bullet fragments. In addition, bullet fragments may occur in wounds that are not defined as large. Fragmentation of bullets is neither a necessary nor sufficient cause of large wounds, and surgeons should not diagnose extensive tissue damage because of the presence of fragments on radiography. Such findings also do not necessarily represent the use of bullets which contravene the law of war. Future legislation should take into account not only the construction of bullets but also their potential to transfer energy to the human body.

Key messages

  • The use of certain bullets has been prohibited in war

  • Wounds from bullets are caused by transfer of kinetic energy from the bullet to the tissues

  • The relation between size of wound and fragmentation of bullets can be examined using the Red Cross wound classification system

  • Fragments of bullets seen on radiographs of wounds sustained in wars do not necessarily represent large wounds or the use of illegal bullets

  • Existing legislation on the construction of bullets should be supplemented by legislation on how much energy is transferred to tissues


The St Petersburg Declaration of 1868 banned the use of bullets that explode on impact with the human body, and this was the basis of the legal notion of “superfluous injury or unnecessary suffering.”1 In the 1890s concern mounted about the effects on human beings of other bullets, including British dumdum bullets. This concern opened a debate about, on the one hand, the effectiveness of bullets or their stopping power and, on the other hand, how much injury civilised nations should inflict on their enemies.25 As a result, in the Hague Declaration of 1899, the contracting parties agreed “to abstain from the use of bullets which expand or flatten easily in the human body, such as bullets with a hard envelope which does not entirely cover the core or is pierced with incisions.” (Such bullets now have various names such as expanding, soft-point, and hunting; they are best referred to collectively as semi-jacket bullets.) Since then, all military bullets have been covered by a full metal jacket, an example of successful international legislation.

Wounds are caused by transfer of kinetic energy from projectiles to tissues of the body: the greater the transfer of energy the larger the wounds. The nature of the bullets' jackets is only one of the factors determining how much energy is transferred; other factors are velocity, mass, stability in flight, and the length of the track in the body.6 7 Therefore, if bullets with full metal jackets cause large wounds their use in armed conflict may comply with the word of the law but not with the spirit of the law. In other words, legislation about the jacket of bullets may not prevent the effect that gave rise to the concern that prompted the legislation. The issue has been further confused by the different focus of the disciplines concerned. While surgeons have invoked bullets' velocity or kinetic energy as the cause of large wounds,8 9 lawyers and technicians have remained focused on the construction of bullets.

The physical basis of wound ballistics and the causation of large wounds were understood as early as 1908.10 However, many wound ballistic experiments have been performed this century with a view to establishing whether certain bullets comply with the Hague Declaration of 1899. The detection of fragments of bullets either on radiography or by shooting a given bullet into simulated soft tissue is assumed to show non-compliance with the treaty or the use of bullets that have not been designed for military use.1115 This assumption is based on another: that the process of fragmentation of bullets is a necessary and sufficient cause of large wounds. Questioning these assumptions and so questioning the clinical and legal significance of fragmentation of bullets requires critical examination of the occurrence of large wounds and the extent to which this is associated with the fragmentation of bullets.

I used the Red Cross wound classification to examine the relation between fragmentation of bullets as detected on clinical radiography and the size of wounds sustained in the field.


Red Cross wound classification

The Red Cross wound classification permits documentation of the effects of missiles and explosions on people.1619 It is an anatomical classification alone and does not include a physiological variable. In a clinical setting this classification has been used to document the incidence of bullet disruption in armed conflict14 and the categories of wounds caused to civilians by hand grenades20 and to establish the size of wounds inflicted by conventional weapons.1

The box shows the six factors that can be scored for any wound. The M score indicates whether bullets or bullet fragments are visible in a radiograph and is relevant to this study (figs 1 and 2). Wounds are graded from the scores for entry, exit, and cavity to denote size and so reflect energy transfer. The V and M scores do not influence the computing of the grade. The correlation between grade and energy transfer or between grade and type of weapon is not precise, but bullets from handguns usually inflict grade 1 wounds and transfer up to 500 J of energy.6 Bullets from shotguns fired at close range would invariably cause wounds of grade 3 and are associated with transfer of more than 1500 J of energy.6 Military rifles can inflict all grades of wound.7 15

Red Cross wound classification

Factors to be scored

E, entry—Estimate the maximum diameter of the entry wound in centimetres

X, exit—Estimate the maximum diameter of the exit wound in centimetres (X=0 if no exit)

C, cavity—Can the cavity of the wound take two fingers before surgery? This may be obvious before operation or be established only after skin incision. For chest or abdomen wounds it refers to the wound of the chest or abdominal wall

C=0 if cavity cannot take two fingers

C=1 if it can

F, fracture—Is there a fracture?

F0=no fracture

F1=simple fracture, hole, or insignificant comminution

F2=clinically significant comminution

V, vital structure—Are brain, viscera, or major vessels injured? Is there breach of dura, pleura, or peritoneum?

V=0 if they are not injured

V=1 if they are

M, metallic body—Are bullets or bullet fragments visible on radiography?

M=0 if there are no metallic bodies

M=1 if there is one metallic body

M=2 if there are two or more metallic bodies

Grading of wounds

Grade 1—Skin wounds of <10 cm (E+X<10) without a cavity (C0) and without a comminuted fracture (F0 or F1)

Grade 2—Skin wounds of <10 cm (E+X<10) with a cavity (C1) or comminuted fracture (F2)

Grade 3—Skin wounds of ≥10 cm (E+X≥10) with a cavity (C1) or comminuted fracture (F2)

Fig 1.Fig 1.
Fig 1.

Bullet wound of left thigh. Top: Entry wound is indicated by pen and measured <1 cm. Exit wound on lateral side of thigh is large and obvious. Bottom: Radiograph shows comminution of femur with one large and several small metallic bullet fragments. Wound is grade 2 (scores: E=1, X=8, C=1, F=2, V=0, M=2)

Wound database

The wound database of the International Committee of the Red Cross originates from a system of data collection which was established in the organisation's independent hospitals in 1991. Included in the information recorded for each of the 26 636 patients are the cause of injury and the Red Cross wound classification.

I analysed all records of patients with one or two bullet wounds scored by the Red Cross wound classification by grade of wound or wounds and the corresponding M scores. I combined wounds of grade 2 and 3 to form a category of large wounds to compare the proportion of wounds associated with fragmentation of bullets.


A total of 5215 patients had scores for their bullet wounds. Of these 718 had two scored wounds. Thus, there were a total of 5933 scored wounds from which the grade could be computed. The table shows the grade of wound by M score.

Grade of wound by M score for 5933 bullet wounds

View this table:

Of the 347 wounds associated with fragmentation of bullets (total of M2 wounds), 251 (72%) were large wounds (grade 2 or grade 3) of the 5586 wounds not associated with fragmentation of bullets (total of M0 plus M1 wounds), 2915 (52.1%) were large wounds. Fragmentation of bullets (M2 wounds only) was associated with 96 (3.5%) of the 2767 grade 1 wounds and with 251 (7.9%) of the 3166 large wounds. Fragmentation of bullets was associated with large wounds (Pearson's χ2= 53.3, P<0.0001).

Fig 2.Fig 2.
Fig 2.

Bullet wound of left arm. Top: Entry wound is just proximal to elbow joint and is indicated by forceps in surgeon's hand. Exit wound in forearm is large and obvious. Bottom: Radiograph shows comminution of forearm bones and no metallic bullet fragments. Wound is grade 3 (scores: E=1, X=11, C=1, F=2, V=0, M=0)


These results show that fragmentation of bullets and large wounds is associated with large wounds, but fragmentation does not necessarily cause large wounds. Previous assumptions about the significance of fragmentation of bullets are misguided; this conclusion is supported by the fact that 82.1% of the large wounds occurred without fragmentation of bullets.

Limitations of the study

This study has several limitations related to the conditions under which the data were collected. A plain radiograph of the wounded part is taken routinely in Red Cross hospitals. A few may not have had this investigation either because the wounds were so serious that the person was taken directly to the operating theatre or because the wound did not require surgical intervention; in such cases the M score would be recorded as zero.

There are no data pertaining to those people who died before reaching hospital. This can be offset against the fact that some Red Cross hospitals admit only those who are more severely wounded; the others find treatment elsewhere. The results are unlikely to be affected by either factor as most of the wounded patients have wounds to their arms or legs that whatever their severity, rarely prevent the person reaching hospital.

Possible observer error must be addressed. The proportion of M2 wounds seen in four Red Cross hospitals in a previous study is higher than reported here.14 In that study, data were collected by retrospective review of radiographs. The most likely cause for the lower proportions of M2 wounds in the centralised Red Cross database is that the many surgeons deployed in these hospitals who are responsible for the routine scoring of wounds are not aware that small and less noticeable specks of lead ejected from the base of a full metal jacket bullet 6 13 should also be scored as M2. Multiple pellets from a shotgun cartridge would also be scored as M2; however, in my experience such wounds are rarely seen in Red Cross hospitals.

A bullet may fragment on passing through the metalwork of a vehicle Whatever the wound or wounds sustained by the person inside the vehicle, a radiograph is likely to show multiple fragments and would correctly be given an M2 score. There is no way to ascertain the proportion of people on the database who were wounded in this way This could account for some of the M2 scores of the grade 1 wounds.

Although most bullet wounds in this study were most likely inflicted by a Kalashnikov AK47 rifle, it is unclear how the proportions of the grades of wounds or the proportion with M2 wounds would change with different weapons.


None of these limitations refute the conclusion that fragmentation of bullets is an unreliable indicator of large wounds. Disproving the assumptions about the association between bullet fragmentation and large wounds has important implications. In a clinical setting surgeons should not diagnose extensive tissue damage because of the presence of bullet fragments on radiography. Likewise, lawyers should not claim that fragmentation of bullets as seen on clinical radiographs or in wound ballistic studies implies contravention of the Hague Declaration of 1899.

The results of this study are explicable by an understanding of wound ballistics. As a bullet passes along its track in the body, it lacerates and damages tissues by doing work on them—that is, by transferring to the tissues the kinetic energy it is carrying. An equal and opposite amount of work is done on the bullet by the tissues. Where along the track this work is done is determined, in part, by the construction of the bullet. A bullet which expands or flattens easily in the human body—that is, a semi-jacket bullet—is likely to cause a large wound because it transfers most of its energy in the first 10 cm of its track6; fragments of the bullet are seen on radiography as a result. A military bullet may, after 10-20 cm of minimal energy transfer, turn sideways in its track, maximise energy transfer, and be compressed enough to cause it to deform or break6; again, fragments may be seen as a result (fig 1). Whatever the construction of a bullet, if enough work is being done in the interaction between the projectile and the tissues, the bullet deforms or breaks. However, enough work can be done to cause a large wound without the bullet deforming or breaking; this is particularly the case for bullets with full metal jackets (fig 2).

Legislation about bullets' construction should be supplemented by legislation about how much kinetic energy is transferred to a soft medium and where along the bullets' tracks this energy is transferred.21 In this way, new projectiles could not circumvent the spirit of the law.


  • Funding No additional funding.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.