U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Bast RC Jr, Kufe DW, Pollock RE, et al., editors. Holland-Frei Cancer Medicine. 5th edition. Hamilton (ON): BC Decker; 2000.

Cover of Holland-Frei Cancer Medicine

Holland-Frei Cancer Medicine. 5th edition.

Show details

Chapter 17Trauma and Inflammation

, MD.

The normal wear and tear of life induces a multiplicity of traumas which are rarely noted or quickly forgotten until the time arises to make something out of them.

Stewart, 1946

The role of trauma in the causation of cancer is a subject fraught with gross exaggerations and contradictions. The literature abounds in points of view ranging from detailed descriptions of case reports of trauma followed by malignant neoplasms to others in which this relationship is minimized or flatly denied.1–8

This problem is further confounded when medicolegal implications are considered. Compensation claims for development of a tumor ascribed to trauma are not infrequent. Although they rarely have a factual basis, some have been settled in favor of the injured party. Few cases claiming a causal relationship between trauma and neoplasia fulfill criteria for causation postulated by Ewing9 or Warren.10 These conditions include (1) authenticity of trauma, (2) sufficient severity of the trauma, (3) reasonable evidence of prior integrity of the injured area, (4) tumor appearance at the site of trauma, and (5) time interval not too remote for reasonable association of trauma and tumor. Additional criteria to further authenticate a possible relationship are (1) trauma of such magnitude that reparative proliferation of cells occurs, and (2) tumor of a type that might reasonably develop as a result of the regeneration and repair of specific tissues damaged during injury.10

Unfortunately, the application of these criteria is not in itself a warranty of a scientific and objective approach to this problem. One of the main sources of disagreement between proponents and opponents of the role of trauma in the causation of cancer has been the length and significance of the time interval between the two events. This factor appears to be somewhat flexible when compared with the appearance time of other neoplasms brought about by known carcinogens. The exact sequence of events leading to the development of leukemia following irradiation is not known, but in the case of radiation-related leukemia in Nagasaki and Hiroshima after atomic bomb exposure, peak appearance was 7.2 to 9.4 years later, but with wide variation.11 A similar interval until the peak (8.5 years) has been demonstrated in the case of thyroid neoplasms in children following neck irradiation.12 The appearance of clear cell vaginal adenocarcinoma in young women whose mothers received diethyl stilbestrol therapy during gestation occurred 14 to 22 years after the chemical traumatic event.13

Mechanisms Of Possible Traumatic Causation

A wide variety of malignant neoplasms have been described in association with trauma. In many cases, the association appears purely coincidental, judging from the lack of scientific evidence. Mechanical trauma may soon alert the patient to the presence of a pre-existing neoplasm in the affected part. Accepting the definition of trauma as a mechanical force received by the body, followed by a local reaction characteristic of injury, we exclude from consideration in this chapter ionizing irradiation (see Chapter 14), chemical insults (see Chapter 12), and ultraviolet radiation (see Chapter 15) as different forms of injury discussed elsewhere.

Different forms of skin cancer have been known to result from mechanical injury.

Draining Sinuses

In 1828, Marjolin described the development of malignant neoplastic changes in an old skin ulcer, probably the site of a draining sinus.14 In 1931, Benedict described 12 cases of cancer occurring in draining osteomyelitic sinuses and collected 52 similar cases from the literature.15 He studied 2,400 cases of osteomyelitic sinuses and found a 0.5% incidence of malignant change. The draining sinus in which cancer occurred had been present for an average of 30 years, and the subsequent neoplastic lesions were invariably slow growing. A review of all reports of cases in which metastases had been reported found only four cases.16 Because of the rarity of spread, the authors postulated that some of the cases reported as cancer could have been instances of pseudoepitheliomatous hyperplasia of skin around the sinus.

Thermal Injury

Since the description by Dupuytren in 1839 of a patient treated for cancer arising in the scar of a burn caused by sulfuric acid, many cases have been described in which heat has been the initial insult that puratively triggered the development of cancer at the site of scarring from injury.17 Excellent reviews on this subject are available, concluding that the potential of a scar to undergo malignant neoplastic degeneration, and indeed the type of resulting epithelioma, are related to the extent of the surface area involved and the depth of the burn.18,19 The type of burn is related to the nature of the agent (flame, metal, tar), to its temperature, to the tissue’s capacity for heat absorption, and to the duration of contact.1 Although most of the reported malignant lesions that follow burns are squamous cell carcinomas, basal cell cancers can also occur, usually when the burn is superficial and when the thermal injury resulted from hot solids. On the basis of a study of 2,465 cases of skin cancer, 2% of all squamous cell carcinomas and 0.3% of all basal cell carcinomas originate on skin subjected to thermal injury.19,20

Kangri Burn Cancer

The Indian kangri is an earthenware bowl heated by charcoal and worn against the skin of the thighs and abdomen. Owing to the constant application of heat, the skin in these areas becomes dry, hyperkeratinized, and frequently shows chronic dermatitis. Scars resulting from previous kangri burns are frequent and prone to undergo malignant change. The average age at onset is 55 years and the average duration of life is 15 months from the reported onset of the cancer, which suggests that metastasis does occur. The gross lesion is variable in appearance but microscopically is consistently squamous cell carcinoma.21

Kairo Burn Cancer

The Kairo burn cancer in Japan relates to another system for the maintenance of body warmth, the use of a light metal box containing embers that fits snugly against the contour of the abdomen. It was generally worn under kimonos for a period of at least 3 hours at a time. The continued or prolonged use of this utensil produced erythematous burns or chronic dermatitis leading to malignant neoplastic change.19

Lung Cancer

A relationship between lung cancer and pulmonary scars was first noted by Friedrich and Rossle, and several reports have confirmed this association. Their association has been the subject of numerous reports.4,22– 26 The exact frequency is difficult to determine. Luders and Themel reported a frequency of 28% in their study of 2,032 autopsies, whereas others report only 14%.27,28 These findings led to the concept of “scar cancer” as a morphologic entity embracing any inflammatory or vascular pulmonary lesion which led to the formation of scar tissue followed by the development of carcinoma. In most instances, the tumor arises peripherally; histologically, it shows the characteristics of pulmonary adenocarcinoma or bronchiolar carcinoma.29 Restrictive criteria have been postulated to differentiate true pulmonary scars from the dense connective tissue often encountered in lung cancer.29 Some observers believe that the association is higher than generally accepted because of the difficulty of convincingly demonstrating the presence of pre-existing scar tissue when examining pulmonary lesions in surgical or in autopsy material.26

Cancer of the Esophagus

This type of cancer is often related to previous injury. Stricture of the esophagus following lye ingestion is complicated by squamous cell carcinoma in at least 5.2% of cases.30 Other causes of injury have also been implicated, such as strong alcohol and smoking, but thermal irritation has been cited in certain cultures as the most constant factor predisposing to esophageal cancer.31,32 In regions where tea is drunk close to the boiling point, as in the Caspian littoral, esophageal cancer is the most common neoplasm. Nutritional deficiencies in such peoples confound the etiologic role of thermal trauma.

Cancer of the Oral Cavity

The role of tobacco and alcohol in the etiology of oral cancer is so dominant that they diminish the relative contribution of local trauma from dental irritation.33

Moles and Malignant Melanoma

A significant number of malignant melanomas are related to pre-existing nevi.34 A small percentage of nevi undergo malignant changes, but the true nature of this transformation is not well known. The fact that most malignant melanomas occur on exposed surfaces of the body, and that the incidence of the disease is higher in the sunnier parts of some countries support the concept that ultraviolet injury plays a major part in their causation.35,36 These assertions are based on studies among Caucasians. Among more highly pigmented races, trauma is considered to play a greater role.37 Studies of malignant melanoma in Australia showed that the most frequent location is the skin of the back in men. No direct relation was found to the belt area of the trunk, however, where the incidence of constant trauma is obviously higher.35

Trauma and Bone Tumors

Single traumas as causal events for bone tumors have been repeatedly postulated, although never proved. The preponderance of bone tumors occur in the same young age group in which the incidence of trauma is especially high. Most reports deal with observations of single cases. Nearly all reports deal with such short intervals between trauma and tumor that their causal relationship is highly improbable. To the contrary, the recollection of a remote trauma, otherwise forgotten and undistinguished until the appearance of tumor, is unpersuasive. The lack of significant increase in the number of bone tumors following wars, which cause many thousands of injured, also speaks against such a relationship.36

Other Types of Cancers

Other types of cancers have occasionally been reported in association with trauma. For mammary carcinoma and carcinoma of the uterine cervix, the evidence that trauma has etiologic significance is not persuasive.5,37,38

In one of the most important epidemiologic studies relating trauma and tumor incidence, Inskip and colleagues studied, over a 15-year period, 228,055 Danish residents who had been hospitalized for head injury and observed for at least 5 years, an average of 8, and a maximum of 17 years for benign and malignant brain tumors. Because of the superb Danish identification techniques, a search of the national tumor registry was conducted for the same 15-year period plus 1 year, and the results compared with the incidence in the Danish population. Excluding the first year of follow-up, which probably was heavily weighted by prevalent cases present before the injury, the standardized incidence ratio was 1.0 for gliomas, and 1.2 for meningiomas (with confidence intervals [CI] of 0.8 to 1.7). Thus, no significant relationship to trauma was found for the two most common types of brain tumor. Hemangioma and hemangioblastoma (N=15) were more frequent than in the general population, for a standardized incidence ratio of 2.6 (CI 1.4–4.2). This correlation needs independent confirmation.39

A multi-region study in the United Kingdom identified 794 men aged 15 to 49 years diagnosed with testicular germ cell tumor in a 33-month period. An age-matched control subject was selected, and all were interviewed concerning testicular trauma at least 2 years prior to the tumor. The odds ratio was 2.00 (CI 1.54–2.61). Although this difference is significant, it depends on memory of preceding events, which may have been enhanced by the subsequent attention to the testis.40

Interactions of Trauma and Tissue Repair

There is no experimental evidence to substantiate the production of tumors by the direct action of trauma, but there are numerous studies indicating its significance as a “promoter.” The studies of Rous and co-workers showed that under certain circumstances, in rabbits, a trauma can precipitate the formation of a tumor when the cells have been conditioned with previous oncogenic tar treatment.41,42 If trauma is to be considered independently as a tumor-causative factor, however, this could only take place as an extremely rare event during the process of regeneration and repair. A normal regenerative process implies the restoration of lost tissues by structurally and often functionally similar cells. A variety of organ-specific wound hormones (cytokines) liberated from the site of injury exert a stimulating effect on homologous tissue.43 Such putative cytokine release might eliminate or decrease an inhibitory gene regulatory effect,44 allowing increased cell cycling and growth. On purely numerical grounds, an increased number of mutations might then ensue. It is conceivable that persistent or repetitive cell damage by trauma could trigger an excessive proliferative effect. This is quite different from a single traumatic event, however.

The possibility of direct mutation from chronic inflammation or repair is controversial.45 Although this is the presumed mechanism of epithelial carcinogenesis in burn cancers and long-draining sinuses (although many cytokines may be involved) the rarity of soft tissue sarcomas compared with the frequency of wounding, operative and blunt trauma to mesenchymal tissues casts doubt on traumatic mutation in these tissues as an oncogenic event.


Abbas J S, Beecham J E. Burn wound carcinoma: case report and review of the literature. Burns Incl Therm Inj. 1988;14:222. [PubMed: 3048565]
DeNayer P P, Delloye C, Malghem J. Bone injury and late giant-cell tumor occurrence: a possible relation. A case report. Orthopedics. 1987;10:1279. [PubMed: 3628112]
Langer F, Pritzker K P, Gross A E, Shapiro I L. Giant cell tumor associated with trauma. Clin Orthop. 1982;164:245. [PubMed: 7067295]
Madri J A, Carter D. Scar cancers of the lung: origin and significance. Hum Pathol. 1984;15:625. [PubMed: 6378757]
Mosinger M, Glaunes J P, Fiorentini H, Bandler H. Tumeurs et cancers post traumatiques. Ann Med Leg. 1961;41:472. [PubMed: 14476492]
Stoll H L, Crissey J T. Epithelioma from single trauma. NY State J Med. 1962;62:496. [PubMed: 13917418]
Troost D, Tulleken C A. Malignant glioma after bombshell injury. Clin Neuropathol. 1984;3:139. [PubMed: 6090046]
Voutilainen A, Teir H, Kivivouri A. Causal relationship between trauma and malignant tumors. Ann Chir Gynaecol Fenn. 1967;56(Suppl 152):1. [PubMed: 4228058]
Ewing J. Bulkley lecture: modern attitude toward traumatic cancer. Arch Pathol. 1935;10:690.
Warren S. Minimal criteria to prove causation of traumatic or occupational neoplasms. Ann Surg. 1943;117:585. [PMC free article: PMC1617604] [PubMed: 17858205]
Bizzozero O J, Johnson K G, Ciocco A. Radiation-related leukemia, Hiroshima and Nagasaki, 1946–1964. I. Distribution, incidence and appearance time. N Engl J Med. 1966;274:1095. [PubMed: 5932020]
Winship T, Rosvoll R V. Childhood thyroid carcinoma. Cancer. 1961;14:734. [PubMed: 13785889]
Herbst A L, Ulfelder H, Poskanzer D C. Adenocarcinoma of the vagina: association of maternal stilbestrol therapy with tumor appearance in young women. N Engl J Med. 1971;284:878. [PubMed: 5549830]
Marjolin JN. Ulcére. In: Dict de méd, 2nd ed. 1846. p. 22.
Benedict E B. Carcinoma in osteomyelitis. Surg Gynecol Obstet. 1931;53:1.
Bereston E S, Ney C. Squamous cell carcinoma arising in a chronic osteomyelitic sinus tract with metastasis. Arch Surg. 1941;43:257.
Dupuytren G. Lélcons Orales de Cliníque Chirurgicale, 2nd ed. Paris, France: 1839.
Lever WF. Histopathology of the skin, 4th ed. Philadelphia, PA: JB Lippincott; 1967. p. 63.
Treves N, Pack G T. The development of cancer in burn scars. Surg Gynecol Obstet. 1930;51:749.
Lifeso R M, Rooney R J, Shaker M. Post-traumatic squamous cell carcinoma. J Bone Joint Surg. 1990;72:12. [PubMed: 2295659]
Neve E F. Kangri Burn Cancer. Br J Med. 1923;21:1255. [PMC free article: PMC2317796] [PubMed: 20771404]
Balo J, Juhasz E, Temes J. Pulmonary infarcts and pulmonary carcinoma. Cancer. 1956;9:918. [PubMed: 13364875]
Carroll R. The significance of lung scars on primary lung cancer. J Pathol Bacteriol. 1962;83:293. [PubMed: 13876892]
Friedrich G. Periphere Lungenkrebse auf dem bodem Pleuraher Narben. Virchows Arch Pathol Anat. 1939;304:230.
Raeburn C, Spencer H. Lung scar cancers. Br J Tuberculosis. 1957;51:237. [PubMed: 13446412]
Yokoo H, Suckow E. Peripheral lung cancers arising in scars. Cancer. 1961;14:1205. [PubMed: 14009145]
Gelzer J. Uber die peripheren Lungenkrebse in Bereich von Lungernarben. Virchows Arch Pathol Anat 1956; :329. [PubMed: 13409749]
Luders C J, Themel K G. Die Narbenkrebse der Lungen als Beitrag zur Pathogenese des peripheren Lungencarcinoms. Virchows Arch Pathol Anat. 1954;325:499. [PubMed: 13226703]
Castleman B. Healed pulmonary infarcts. Arch Pathol. 1940;30:130.
Joske R A, Benedict E B. The role of benign esophageal obstruction in the development of carcinoma of the esophagus. Gastroenterology. 1959;36:749. [PubMed: 13653298]
Victoria C G, Munoz N, Day N E. Hot beverages and esophageal cancer in Southern Brazil: a case control study. Int J Cancer. 1987;39:710. [PubMed: 3583451]
Watson W L, Goodner J. Carcinoma of esophagus. Am J Surg. 1957;93:259. [PubMed: 13394803]
Wynder E J, Bross I J, Feldman R M. A study of the etiological factors in cancer of the mouth. Cancer. 1957;10:1300. [PubMed: 13489682]
McGovern JJ. Malignant melanoma: clinical and histological diagnosis. New York, NY: Wiley, 1976. p. 47–54.
Davis N C, Herrow J J, McLeod G R. Malignant melanoma in Queensland. Analysis of 400 skin lesions. Lancet. 1966;2:407.
McKie R M, Atchison T. Severe sunburn and subsequent risk of primary cutaneous malignant melanomas in Scotland. Br J Cancer. 1982;46:955. [PMC free article: PMC2011208] [PubMed: 7150488]
Hewer T F. Malignant melanoma in colored races: role of trauma in its causation. J Pathol Bacteriol. 1935;41:473.
Dietrich A. Krebs Nach Kriegsverletzungen. Z Krebsforsch. 1942;52:91.
Boyd J T, Doll R. A study of the etiology of carcinoma uteri. Br J Cancer. 1964;18:419. [PMC free article: PMC2071087] [PubMed: 14219535]
Stewart F W. Occupational and post-traumatic cancer. Bull NY Acad Med. 1947;22:145. [PMC free article: PMC1870885] [PubMed: 19312521]
Inskip P D, Millemkjaer L, Gridley G, Olsen J H. Incidence of intracranial tumors following hospitalization for head injuries. Cancer Causes Control. 1998;9:109–116. [PubMed: 9486470]
U K. Testicular Cancer Study Group. Social, behavioural and medical factors in the aetiology of testicular cancer: results from the UK study. Br J Cancer. 1994;70:513–520. [PMC free article: PMC2033353] [PubMed: 8080739]
Friedwald W F, Rous P. The pathogenesis of deferred cancer. J Exp Med. 1950;91:459. [PMC free article: PMC2135979] [PubMed: 15415503]
Rous P, Kidd J G. Conditional neoplasms and sub-threshold neoplastic states. J Exp Med. 1941;73:365. [PMC free article: PMC2135131] [PubMed: 19871084]
Tier H, Kiljunen A, Putkonen T. Existence of growth promoting factor in the skin of the white rat. Ann Chir Gynaecol Fenn 1951; :40. [PubMed: 14838385]
Bullough W S. Mitotic and functional homeostasis: a speculative review. Cancer Res. 1965;25:1683. [PubMed: 5323643]
Warren S. In: Anderson, WAD, editor. Pathology. St. Louis, MO: Mosby; 1961. p. 447.
© 2000, BC Decker Inc.
Bookshelf ID: NBK20784


  • PubReader
  • Print View
  • Cite this Page