Radiotherapy and Oncology 42 (1997) 99–106
Wound healing following radiation therapy: a review Martha K. Tibbs Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, 100 Blossom Street, Cox Building – Lower Level, Boston, MA 02114, USA Received 6 February 1996; revised version received 14 November 1996
Abstract Radiation therapy may interrupt normal wound healing mechanisms. Changes in vasculature, effects on fibroblasts, and varying levels of regulatory growth factors result in the potential for altered wound healing whether radiation is given before or after surgery. Surgical factors, such as incision size, as well as radiation parameters, including dose and fractionation, are important considerations in developing overall treatment plans. Experience suggests that certain practical measures may diminish the risk of morbidity, and investigations are ongoing. 1997 Elsevier Science Ireland Ltd. Keywords: Radiotherapy; Wound healing; Skin; wounds; Post-operative complications; Acute radiation effects; Late radiation effects
2. Normal wound healing
The process of wound healing allows the body to repair injured tissue and restore skin integrity . It requires the coordinated integration of cellular, physiologic, and biochemical events, and is therefore dynamic in nature. Poor wound healing may cause significant morbidity in the post-operative setting [1,30], and a number of factors may compromise a patient’s ability to heal such a wound. Excessive trauma to tissues during surgery, preexisting vascular disease, or improper surgical flap design may result in inadequate vascularization of involved tissues. Extreme nutritional deficit and corticosteroid therapy may contribute [15,37]. Prior radiation and concomitant chemotherapy [1,30], even as components of a carefully planned treatment regimen, may interrupt the normal wound healing mechanisms thereby increasing the likelihood of post-operative wound infections which in turn predispose the wound to a higher risk of dehiscence and failure . The process of wounding forces normal cells to proliferate, and in combination with radiation therapy may result in expression of otherwise late radiation effects in an acute setting. For example, expression of damage to cutaneous fibroblasts produced by therapeutic irradiation may be hastened by wounding .
In a healthy patient, a surgical wound requires at least 2 years to completely heal by primary intention, that is, for the surgical scar to attain its maximal strength. Even at that time, when measured by stressing it to failure, its maximum tensile strength is only 80% of normal uninjured tissue . In 1929, Edward Howes described normal wound healing and provided a basis for its subsequent study by correlating specific histologic stages with the wound’s gain in tensile strength . The normal wound healing response is depicted in Fig. 1. It represents a continuum of events that overlap in both time and activity with one stage merging into the next [15,30]. As depicted in Fig. 2, there are three stages of normal surgical wound healing. The first is the inflammatory phase which dominates the first 3 days of the wound healing process. The second stage is the proliferation phase which occurs between day three and week four following injury or surgical incision. The final stage is the remodeling or maturation phase which begins at approximately 4 weeks after wounding and continues for 2 years thereafter . Development of an infection or wound dehiscence interrupts the normal sequence of events, and this interruption may interfere with the delivery of subsequent ther-
0167-8140/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0167-8140 (96 )0 1880-4
M.K. Tibbs / Radiotherapy and Oncology 42 (1997) 99–106
Fig. 1. Normal wound healing requires multiple events, many of which occur simultaneously or overlap in time. (From Mast, B.A. The skin. In: Wound Healing: Biochemical and Clinical Aspects. Editors: I.K. Cohen, R.F. Diegelmann and W.J. Lindblad. W.B. Saunders, Philadelphia, 1992, p. 347. With permission.)
apy. In the case of a patient with cancer, where timeintensity relationships are important in the delivery of treatment and directly related to the chance of cure, wound healing plays a major role. 2.1. Inflammatory phase The inflammatory phase is a period of active cellular migration . It accomplishes both hemostasis and for-
Fig. 2. Three stages of normal wound healing. It requires 2 years to attain maximal strength which is about 80% of the strength of normal skin. (From Springfield, 1993 . With permission.)
mation of a temporary matrix on which fibroblasts and monocytes migrate . This stage begins as soon as a surgical wound is closed and lasts for 3 to 4 days. It is the stage most sensitive to the effects of chemotherapy and radiation therapy . At the time of wounding, the bleeding capillaries release blood products into the wound and initiate the coagulation cascade through the exposure of existing collagen bundles to blood-borne elements. Platelets assist in clotting and also release cytokines such as platelet derived growth factor (PDGF) and transforming growth factor beta (TGFbeta) [11,30]. The vascular bed becomes dilated increasing the influx of macrophages, monocytes, neutrophils, and lymphocytes. These cells reduce the number of bacteria in the wound and enhance the process of neovascularization through autocrine and paracrine processes [29,34]. Growth factors, or cytokines, mediate the process by stimulating mitogenesis and angiogenesis, and activating local fibroblasts that increase their collagen production [12,30]. In addition, epithelialization begins within the first 24 h following wounding with basal cells migrating over the incised dermis . Critically involved in the healing process are a number of cytokines [9,10]. Peptide growth factors initiate tissue repair by their chemotactic influence on macrophages and their varied influences on fibroblasts . TGF-beta, basic fibroblast growth factor (bFGF), and PDGF are all chemo-
M.K. Tibbs / Radiotherapy and Oncology 42 (1997) 99–106
tactic for fibroblasts; both PDGF and TGF-beta induce fibroblast proliferation; and TGF-beta causes fibroblast differentiation. In addition, both epidermal growth factor (EGF) and growth hormone (GH) enhance protein synthesis [19,37]. During the first few days, it is the fibrin clot that binds together the wound edges but results in a tensile strength of less than 5% of normal tissue . The monocyte-macrophage system that is responsible for the majority of events of the inflammatory phase provides the transition toward the next stage . 2.2. Proliferative phase The proliferative stage begins on day two or three following wounding, even before the completion of the inflammatory stage . It continues for at least 3 weeks, and the major event is the accelerated production of collagen, which is enhanced in the presence of PDGF and TGF-beta . Oxygen is a required substrate in collagen synthesis because it is involved in the hydroxylation of lysine and proline necessary for the release of collagen from the cells . Individual collagen fibrils with limited crosslinks then mature into fibers with many crosslinks between the lysyl and hydroxylysyl residues [5,30]. The production of collagen must be a continuous process because constant collagen breakdown is also occurring. Re-establishment of the epidermis is completed during this phase as the basal keratinocytes from the wound edge continue to migrate over the wound. The ‘free edge effect’ hypothesis suggests that the absence of adjacent cells at a wound margin stimulates the migration and proliferation of the keratinocytes. They cross either upon the temporary matrix laid by the products of the clotting cascade or over the basement membrane . bFGF has been shown to be a mitogen for keratinocytes contributing to an enhanced rate of epithelialization [12,17]. Following influx of endothelial cells and fibroblasts into the wound, bFGF, also a potent angiogenic molecule , peaks in concentration within full thickness wounds at approximately day thirteen . Capillary formation continues, fibroblasts proliferate, and mast cells and macrophages remain activated . Angiogenesis is further stimulated by local hypoxic conditions which may be present . Myofibroblasts also appear within the wound and produce contractile proteins containing actin and myosin. Either mechanical stress acting on the myofibroblasts or movement of the fibroblasts within the matrix, or possibly both in combination, stimulate contraction of the wound and reduce the distance that must be bridged by forming collagen . In addition, the presence of growth hormone may stimulate enhanced collagen deposition by fibroblasts . Production of collagen peaks at day seven after wound closure, but it continues at its peak rate for 2 to 4 weeks. The tensile strength of the wound reaches approximately 30% of normal at the end of week four .
2.3. Maturation phase The maturation stage begins around week three and lasts for 2 years. Collagen continues to mature with additional crosslink formation , and its fibrils align in the direction of greatest tension on the wound . Keratinocytes differentiate from basaloid morphology to their normal phenotype and attach to one another forming hemi-desmosomes . During this final phase, there is a reduction in the number of fibroblasts and macrophages, an increase in collagen content, and progressive gains in wound tensile strength as the wound regains most but not all of the strength of normal tissue. By the conclusion of this phase and after 2 years, the tensile strength has returned to approximately 80% of normal tissue.
3. Effect of radiation on wound healing Ionizing radiation produces both acute and delayed effects on skin and subcutaneous tissues  which have profound implications for surgical wound healing. Radiation induces acute degenerative changes in basement membranes and increases vascular permeability. Changes in vasculature may include stasis and occlusion, edema of vascular walls, and thrombosis, and radiation may reduce neovascularization [13,35]. Eventually, progressive loss of vessels and fibrous tissue replacement may occur. Fibroblasts may be permanently altered by radiation. They do not produce sufficient collagen to keep up with the demands of the wound, or the collagen that is produced does not mature quickly enough meet these demands of the acute phases of wound healing. Late radiation effects attributable to fibroblast injury include atrophy, contraction, and fibrosis [6,14,16,34]. The wound bursting strength (WBS) is a direct measure of the force required to break apart a healing linear incision. Although it is simplistic, it correlates with the collagen content of the wound, and more importantly, it accurately reflects the clinical setting. Bernstein et al.  have shown that wound bursting strength following 18 Gy was reduced to 62% of normal as measured one week following irradiation. Two weeks after irradiation it was 54% of normal. Therefore, inhibition of healing in irradiated wounds in this series persisted 2 weeks after wounding as compared with controls. Another parameter, collagen gene expression, is an indirect measure of wound strength and may be useful to provide insight into the particular mechanisms by which radiation impairs healing. In irradiated tissue, a reduction in synthesis of the alpha 1 chain of type I collagen was found 7 days after wounding and was significantly different when compared to controls in the same series. However, there was no difference at 14 days after wounding. The reduction in WBS, therefore, continued even after the collagen gene expression recovered, perhaps from
M.K. Tibbs / Radiotherapy and Oncology 42 (1997) 99–106
increased degradation of collagen, alterations in post-transcriptional collagen formation, or a defect in crosslinking . As mentioned previously, intimately involved in the healing process and shown to be important in repair following irradiation is the regulatory protein TGF-beta . It is found in fibroblasts, endothelial cells, lymphocytes, macrophages, and platelets, and its important functions include the induction of tissue matrix production and stimulation of chemotactic migration of monocytes and fibroblasts. While endogenous TGF-beta has a variety of stimulatory and inhibitory effects both via cascade mechanisms and in combination with bFGF depending upon relative proportions of the two substances , exogenous use of TGF-beta appears to improve healing in irradiation-impaired wounds [4,5,11,15]. Using a guinea pig model, the effect of TGF-beta on healing was evaluated in irradiation-impaired wounds . Wounds treated with small doses of TGF-beta (1–5 mg) demonstrated improved wound healing compared with controls as measured by an increase in WBS 7 days following wounding. The improved strength was accompanied by an increase in collagen type I gene expression in resident fibroblasts. Interestingly, higher doses of TGFbeta (20 mg) resulted instead in exaggerated impairment of wound healing as compared with controls. This finding emphasized the close attention necessary to avoid undesirable effects if exogenous TGF-beta is to find clinical application in promoting wound repair. TGF-beta has also been used as an investigative tool in studying the mechanism of radiation impairment of healing. Cromack et al.  have shown that TGF-beta-1 effectively accelerates soft tissue repair in both normal and delayed healing models. Rats were randomized to receive either total body irradiation (TBI), surface irradiation, or sham irradiation 2 days prior to wounding, with TGF-beta-1 applied at the time of wounding. TBI markedly reduces circulating monocyte numbers but spares the skin tissue, and thereby depresses the acute inflammatory phase of repair. Surface irradiation, in contrast, impairs surface healing but spares bone marrow elements, and thereby inhibits the collagen synthesis phase of repair. As measured on day seven and day twelve following wounding, TGF-beta-1 significantly improved WBS in TBI treated rats, demonstrating its ability to promote collagen synthesis in the absence of monocytes and macrophages. In contrast TGF-beta-1 did not reverse the healing deficit induced by surface irradiation and characterized by impaired function of dermal fibroblasts. These results imply that TGF-beta-1 to some degree eliminates the macrophage requirement for normal tissue repair. However, application of this growth factor does not completely reverse the deficit in repair induced by TBI, suggesting that TBI impairs another component of the healing process as well. Another polypeptide growth factor, PDGF has also been
studied in experimentally irradiated wounds . PDGF is released by macrophages, fibroblasts, and from platelets at sites of injury. It possesses potent chemotactic influences for monocytes, neutrophils, and fibroblasts, and it promotes fibroblast synthesis of fibronectin and collagenase which are significant extracellular components of tissue repair . Mustoe et al.  demonstrated that PDGF appears to recruit viable fibroblasts to the wound site in irradiated rats. The animals received either TBI at 8 Gy or surface irradiation at 25 Gy 2 days prior to performing parallel dorsal skin surgical incisions. A single dose of PDGF was topically applied to one incision. In the surface irradiated rats, treatment with PDGF recruited macrophages into the wounds and partially reversed healing impairment as noted at days seven and twelve. The influx of macrophages was associated with a subsequent increase in fibroblast influx and was found to stimulate TGF which promotes collagen synthesis. The wounds were observed to be 50% stronger than their paired controls as measured by WBS. In the TBI rats with depressed bone marrow cells, wound macrophages were reduced and the PDGF treatment was ineffective. Taken together with the previously described experiments, these results both emphasize the prominent responsibility of cytokines at multiple levels in the wound healing cascade and suggest that intervention may be possible at various points in the pathway. 3.1. Irradiation delivered prior to surgery Except for very low doses, pre-operative irradiation affects wound healing whether given in single or multiple fractions . Gorodetsky  found a clear dose dependency, with reductions in wound strength as radiation doses were escalated. The threshold at which this response was observed after a single fraction occurred at approximately 8 Gy, and a linear dose response continued with doses up to approximately 20 Gy. Injury to local components of the wound healing mechanism persists and results in diminished proliferative capabilities. Irradiation predominantly affects fibroblast function, and as previously mentioned, collagen production by fibroblasts does not meet the demands of the healing wound. Damage to the fibroblast’s ability to multiply, damage to the ability of fibroblasts to produce collagen and encourage collagen maturation, or a reduction in the fibroblast’s response to stimulatory effects of local growth factors have all been suggested as explanations. This effect seems to be permanent and may even be transmitted to daughter fibroblasts. In addition, irradiation damages keratinocytes and melanocytes leading to altered skin pigmentation and irregularities of the dermis [14,30]. Patients who have received their pre-operative radiation within the weeks to few months prior to surgery experience more wound healing difficulties as compared to patients who received similar doses of radiation more
M.K. Tibbs / Radiotherapy and Oncology 42 (1997) 99–106
Fig. 3. Effects of pre-operative irradiation on wound healing in rats. (From Moss’ Radiation Oncology, 1994 . With permission.)
than 6 months before undergoing surgery. Irradiated tissue demonstrates reduced wound tensile strength in the initial three weeks following the radiation, after which it becomes equivalent to wounds in unirradiated tissues. For the rat model, this difference is depicted in Fig. 3. Irradiated wounds eventually attain the same tensile strength as unirradiated wounds, but initially, they are weaker [14,30]. Though less pronounced, or perhaps less studied, a wound produced in tissue that received radiation months or even years before also exhibits an altered wound healing capacity. The skin and subcutaneous tissues are atrophic and less pliable than normal . Reduced vascularity and tissue hypoxia impair the normal response to wounding and predispose the wound to bacterial invasion. Some investigators estimate the complication rate at 30% in women undergoing breast biopsies in a previous radiation field . Another issue in considering the effect of radiation on wound healing is the surgery itself. More extensive surgery requires that the body heal a larger and more extensive wound. Complications from combining radiation and laparoscopy would naturally occur less than from the combination of radiation and laparotomy. Another example is intraoperative radiotherapy (IORT) which delivers high doses of radiation in a single fraction at the time of surgery, thereby maximally narrowing the interval between irradiation and wounding. Without careful planning then, IORT used on tissues involved in closure of a wound would be expected to have a larger potential for complications. While laboratory studies provide insight into the differences in wound healing between normal and irradiated tissues, clinical data to support these findings are rather
limited. Often wound complications are included within the larger categories of ‘post-operative morbidity’ or ‘moderate/severe post-operative complications’. When hidden within such broad groups, radiation parameters that may have considerable influence on a wound including fractionation, total dose, and timing become difficult to clarify and to study. The Stockholm Rectal Cancer Study Group evaluated low dose pre-operative irradiation in comparison to surgery alone for patients with resectable adenocarcinoma of the rectum . With a radiation dose of 25 Gy in 5 fractions, no significant difference was observed in the postoperative wound complication rate between the two groups. The pre-operative radiation group had a 4% rate of wound dehiscence and a 14% rate of wound infection, whereas the surgery alone group had a 1% dehiscence rate and a 9% infection rate. Mendenhall et al.  also evaluated pre-operative radiation for rectal cancer using regimens that varied in different treatment eras and included 40–50 Gy in 23–28 fractions, 35 Gy in 20 fractions, and 30 Gy in 10 fractions. The complication rate was also comparable for these different schedules. However, it should be noted that complications in patients occurring after discharge from the hospital or more than 30 days following surgery were often not reported. Because these series were not designed to address the effect of fractionation as it pertains to wound healing following pre-operative radiation, specific conclusions require caution. Metson et al.  retrospectively reviewed the wound healing difficulties in 61 patients receiving radiation either once daily (QD) at 1.8 Gy per fraction or twice daily (BID) at 1.6 Gy per fraction on a pre-operative basis to similar total doses for squamous carcinoma of the head and neck region. Of twenty-four patients receiving BID treatments, one suffered major flap loss, five developed wound infections, and five experienced wound separation or minor flap loss. Of 37 patients receiving conventional QD treatments, two suffered major flap loss, one developed a wound infection, and four experienced wound separation or minor flap loss. These differences did not reach statistical significance and therefore suggest that the interval between radiation treatments, within the usual clinical spectrum, may be a less important parameter. Budjko et al.  retrospectively reviewed 202 consecutive patients with soft tissue tumors who were treated with pre-operative radiation and surgical resection. In 89% primary closure of the wound was accomplished without skin grafts or special closure techniques. Overall, the incidence of wound healing morbidity was 37%: 24% wound dehiscence, 6% wound infection, 3.5% seroma, 3% skin graft breakdown, and 0.5% hematoma. Thirty-three of these patients, however, required 43 secondary surgical procedures to facilitate wound healing. In a multivariate analysis, risk factors associated with a delay in wound healing included location of the tumor, increasing age, and patho-
M.K. Tibbs / Radiotherapy and Oncology 42 (1997) 99–106
logic grade of tumor. High rates of retarded wound healing were found in patients with sarcomas of the knee, foot, and buttock regions. Arnold et al. , in considering the related topic of plastic reconstruction, demonstrated that muscle or musculocutaneous flaps elevated from normal tissue are superior to those elevated from irradiated tissue, even if the radiation was given years previously. There was a 32% complication rate for the transposition of an irradiated muscle, and for the transposition of a non-irradiated muscle, it was 19%. The complication rate for a muscle transposed into an irradiated field was 25%. This compares with a complication rate of 11.6% in a series of over 500 patients operated on by the same surgeon over the same time period who required muscle transposition but who had never received radiation treatment. The authors recommend that surgery on irradiated tissue be carefully and thoroughly planned to avoid undermining flaps, contributing to tissue ischemia, and causing hematomas and seromas. Doyle  examined the effects of fractionated radiation on neovascularization in the rat model. His work suggests that inhibition of newly initiated angiogenic responses may partially account for abnormalities in wound healing associated with radiation. Inhibited revascularization was noted at a cumulative dose threshold of 9 Gy when treatment was delivered within 48 h of implantation of polytetrafluoroethylene alloplasts adjacent to superficial epigastric vascular pedicles. Inhibition of alloplast neovascularization occurred to a much greater extent when radiation followed alloplast implantation compared with radiation given before implantation. 3.2. Irradiation delivered following surgery The initial 24 to 48 h after wounding are the most critical with respect to successful healing. Therapeutic irradiation administered 3 or 4 weeks after wound production is less likely to produce complications because the proliferative stage is complete, and delays that occur then have less clinical impact . Arbeit et al.  retrospectively analyzed the wound complication rate following surgical resection with or without peri-operative adjuvant brachytherapy in the treatment of superficial truncal and extremity sarcomas. The majority of patients received 45 Gy. Ninety-eight percent were afterloaded by the eight post-operative day and half of these on post-operative days 2 to 4. There were significant wound healing problems in 52% of the brachytherapy patients as compared with 26% of the patients who did not receive brachytherapy treatment. In this series peri-operative radiation also prolonged the time to complete healing. Ormsby  compared wound healing in patients treated for soft tissue sarcoma with surgery and then randomized to brachytherapy or no brachytherapy in a follow up of the Arbeit analysis. Afterloading of the radioactive
sources was delayed until postoperative days five to eight, and again 45 Gy was delivered. In the Ormsby study, no real difference was found in the wound complication rate between the brachytherapy and non-brachytherapy groups. However, a significant difference was found between the radiated group in this series and the irradiated group in the previous Arbeit study. Wound complications approached parallel control patients that had not received radiation when brachytherapy catheters were afterloaded on the fifth to eighth post-operative day. It may be postulated, then, that a certain critical amount of collagen formation and crosslinking had occurred by this time. These data suggest that even a short delay in postoperative radiation decreases the potential for complications. Mantravadi et al.  compared complications from pre-operative and post-operative radiotherapy in 152 patients with head and neck cancers. Issues of wound healing are often considered when combined modality therapy is contemplated, and radiation may be given either before or after surgery. In this series, post-operative radiation was given after a 3 to 4 week interval following surgery. Fifteen of 92 patients who had received pre-operative radiation experienced wound dehiscence, whereas none of 60 patients who had received post-operative radiation had wound dehiscence. In the context of the previous studies, it appears that the interval between surgery and postoperative radiation critically impacts on the ability of tissue to heal a wound. The optimal interval, however, has not been precisely clarified. In wounds with existing difficulties healing, those with post-operative infections, those that have dehisced, or those in patients with poor nutritional status, the delivery of radiation will further hinder healing. In pressing instances, irradiation may be given to patients with an open wound. Depending upon its size, this may be accomplished without complication, however, the wound may have difficulty healing by secondary intention and require skin grafting or plastic closure . Perhaps exogenous growth factors will find application and prove useful in such a setting . 3.3. Recommendations to avoid wound complications Much of the data that is available groups all morbidity under the severity of reaction rather than with reference to wound complications. As the literature lacks data to guide specific management, treatment is often empiric. Certain practical measures have proven useful in avoiding complications. At the time of surgery, it is extremely important to handle irradiated tissues with care. Whenever possible, the fascia and underlying muscle should be left attached to the skin and subcutaneous tissue as they are reflected from the tumor, and all potential dead space should be filled . Further, wounds should be drained and closed without tension . Seromas and hematomas are likely to
M.K. Tibbs / Radiotherapy and Oncology 42 (1997) 99–106
increase the incidence of late wound breakdown and subsequent infection and to delay the administration of radiation therapy. They should be evacuated. Some oncologic surgeons suggest that sutures should remain in place for a relatively long period of time, perhaps 1 month, but should preferrably be removed before radiation begins . In addition, some authors recommend that patients with lower extremity tumors should stay in bed for an extra few days, with rehabilitative activities beginning slowly . If a complication develops within the first 2 weeks after surgery, it is often better to return the patient early to the operating room to correct the problem which may involve bringing in a vascularized muscle, rather than waiting and hoping the wound will heal on its own. If, however, the patient develops a late but minor wound healing problem, observation is the treatment of choice. If no complication arises within the first 3 to 4 weeks after wound production, healing seems to proceed in a reasonably normal fashion [30,32]. Attempts to avoid difficulties in wound healing have led some investigators to use preoperative or post-operative hyperbaric oxygen therapy . At present, this is usually given in a protocol setting and involves 90-min sessions at 2.4 atmospheres. From the radiation oncologist’s perspective, treatment fields should be planned to minimize the irradiation of uninvolved tissue. Strips of skin and subcutaneous tissue, particularly in the extremities, should be spared from irradiation to allow for lymphatic drainage. Radiation fields should be planned to avoid possible donor areas of cutaneous or myocutaneous flaps. The effect of fractionation has not been systematically evaluated in the clinical setting. From first principles, larger doses would be expected to impair healing to a greater extent. The optimal interval between pre-operative radiation and surgery or between surgery and post-operative radiation is not clearly established. In all likelihood, the tumor site, tumor size, and nearby normal tissues all have impact. Theoretical considerations and the data described previously would suggest that surgery should not be performed sooner than 3 to 4 weeks after full dose radiation  and that post-operative radiation should be delivered no sooner than 3 to 4 weeks following surgery. Although commonly recognized in clinical practice, these suggestions are not founded on specific or randomized data. Much of the information available regarding the influence of radiation on wound healing is found in surgical, dermatologic, and reconstructive surgical literature. Radiation data is often limited to analysis of overall complications and in retrospect. It is incumbent on radiation oncologists to incorporate these considerations in trials where pre-operative or post-operative radiation is employed. References  Alekhteyar, K.M., Leung, D.H., Brennan, M.F. and Harrison, L.B.
The effect of combined external beam radiotherapy and brachytherapy on local control and wound complications in patients with high-grade soft tissue sarcomas of the extremity with positive microscopic margin. Int. J. Radiat. Oncol. Biol. Phys. 36: 321– 324, 1996. Arbeit, J.M., Hilaris, B.S. and Brennan, M.F. Wound complications in the multimodality treatment of extremity and superficial truncal sarcomas. J. Clin. Oncol. 5: 480–488, 1987. Arnold, P.G., Lovich. S.F. and Pairolero. P.C. Muscle Flaps in irradiated wounds: an account of 100 consecutive cases. Plast. Reconstr. Surg. 93: 324–329, 1994. Bernstein, E.F., Harisiadis, L., Salomon, G., Norton, J., Sollberg, S., Uitto, J., Glatstein, E., Glass, J., Talbot, T,. Russo, A. et al. Transforming growth factor beta improves healing of radiation-impaired wounds. J. Invest. Dermatol. 97: 430–434, 1991. Bernstein, E.F., Salomon, G.D., Harisiadis, L., Talbot, T., Harrington, F., Russo, A. and Uitto, J. Collagen gene expression and wound strength in normal and radiation-impaired wounds. J. Dermatol. Surg. Oncol. 19: 564–570, 1993. Bernstein, E.F., Sullivan, F.J., Mitchell, J.B. et al. Biology of chronic radiation effect on tissues and wound healing. Clin. Plast. Surg. 20: 435–453, 1993. Brem, H., Shing, Y., Watanabe, H., Tamvakopoulis, G., Marikovsky, M., Vogt, P.M., Klagsbrun, M., Grosfeld, J., Driksson, E. and Folkman, J. Temporal expression of basic fibroblast growth factor during wound healing. Surg. Forum 43: 664, 1992. Bujko, K., Suit, H.D., Springfield, D.S. and Convey, K. Wound healing after pre-operative radiation for sarcoma of soft tissues. Surg. Gynecol. Obstet. 176: 124–134, 1993. Buntrock, P., Jentzsch, K.D. and Heder, G. Stimulation of wound healing, using brain extract with fibroblast growth factor (FGF) activity. I. Quantitative and biochemical studies into formation of granulation tissue. Exp. Pathol. 21: 46, 1982. Buntrock, P., Jentzsch, K.D. and Heder, G. Stimulation of wound healing, using brain extract with fibroblast growth factor (FGF) activity. II. Histological and morphometric examination of cells and capillaries. Exp. Pathol. 21: 62, 1982. Cromack, D.T., Porras-Reyes, B., Purdy, J.A., Pierce, G.F. and Mustoe, T.A. Acceleration of tissue repair by transforming growth factor beta 1: identification of in vivo mechanism of action with radiotherapy-induced specific healing deficits. Surgery 113: 36–42, 1993. Davidson, J.M. and Broadley, K.N. Manipulation of the woundhealing process with basic fibroblast growth factor. Ann. NY Acad. Sci. 638: 306, 1991. Doyle, J.W., Li, Y., Salloum, A., FitzGerald, T.J. and Walton, R.L. The effects of radiation on neovascularization in a rat model. Plast. Reconstr. Surg. 98: 129, 1996. Drake, D.B. and Oishi, S.N. Wound healing considerations in chemotherapy and radiation therapy. Clin. Plast. Surg. 22: 31–37, 1995. Ehrlichman, R.J., Seckel, B.R., Bryan, D.J. and Moschella, C.J. Common complications of wound healing. Surg. Clin. North Am. 71: 1323–1351, 1991. Gorodetsky, R., Mou, X., Fisher, D.R. et al. Radiation effect in mouse skin: dose fractionation and wound healing. Int. J. Radiat. Oncol. Biol. Phys. 18: 1077–1081, 1990. Hebda, P.A., Klingbeil, C.K., Abraham, J.A. and Fiddes, J.C. Basic fibroblast growth factor stimulation of epidermal wound healing in pigs. J. Invest. Dermatol. 95: 626, 1990. Jorgensen, P.H. and Andreassen, T.T. Influence of biosynthetic human growth hormone on biomechanical properties of rat skin incisional wounds. Acta Chirurg. Scand. 154: 623, 1988. Kingsnorth, A.N. and Slavin, J. Peptide growth factors and wound healing. Br. J. Surg. 78: 1286, 1991. Mantravadi, R.V.P. Skolnik, E.M. and Applebaum, E.L. Complica-
M.K. Tibbs / Radiotherapy and Oncology 42 (1997) 99–106 tions of postoperative and preoperative radiation therapy in head and neck cancers. Arch. Otolaryngol. 107: 690–693, 1981. Mendenhall, W.M., Bland, K.I., Copeland, E.M. et al. Does preoperative radiation therapy enhance the probability of local control and survival in high-risk distal rectal cancer? Ann. Surg. 215: 696– 706, 1992. Metson, R., Freehling, D.J. and Wang, C.C. Surgical complications following twice-a-day versus once-a-day radiation therapy. Laryngoscope 98: 30–34, 1988. Cox, J.D. ed. Moss’ Radiation Oncology: Rationale, Technique, Results, 7th edn, Mosby, St. Louis, 1994. Modified from Nickson, J.J. et al., Roentgen rays and wound healing. II. Fractional irradiation: an experimental study. Surgery 34: 859–862, 1953. Mustoe, T.A. and Porras-Reyes, B.H. Modulation of wound healing response in chronic irradiated tissues. Clin. Plast. Surg. 20: 465– 471, 1993. Mustoe, T.A., Purdy, J., Gramates, P. et al. Reversal of impaired wound healing in irradiated rats by platelet-derived growth factorBB. Am. J. Surg. 158: 345–350, 1989. Ormsby, M.V., Hilaris, B.S., Nori, D. and Brennan, M.F. Wound complications of adjuvant radiation therapy in patients with softtissue sarcomas. Ann. Surg. 210: 93–99, 1989. Pierce, G.F., Tarpley, J.E., Yanagihara, D., Mustoe, T.A., Fox, G.M., and Thomason, A. Platelet-derived growth factor (BB homodimer), transforming growth factor-beta 1, and basic fibroblast growth factor in dermal wound healing: neovessel and matrix formation and cessation of repair. Am. J. Pathol. 140: 1375, 1992. Rudolph, R., Vande Berge, J. and Ehrlich, H.P. Wound contracture and scar contracture. In: Wound Healing: Biochemical and Clinical Aspects, Chapter 6, pp. 96–114. Editors: I.K. Cohen, R.F. Diegelmann and W.J. Lindblad. W.B. Saunders: Philadelphia, 1992.
 Shamberger, R. Effect of chemotherapy and radiotherapy on wound healing: experimental studies. Recent Results Cancer Res. 98: 17–34, 1985.  Springfield, D.S. Surgical wound healing. In: Multidisciplinary Treatment of Soft Tissue Sarcomas, Chapter 5, pp. 81–98. Editors: J. Verweij, H.M. Pinedo and H.D. Suit. Cancer Treat. Res. 67: 81– 98, 1993.  Stockholm Rectal Cancer Study Group. Pre-operative short-term radiation therapy in operable rectal carcinoma: a prospective randomized trial. Cancer 66: 49–55, 1990.  Suit, H.D. and Spiro, I.J. Role of radiation in the management of adult patients with sarcoma of soft tissue. Semin. Surg. Oncol. 10: 347–356, 1994.  Tibbles, P.M. and Edelsberg, J.S. Hyperbaric oxygen therapy. N. Engl. J. Med. 334: 1642–1648, 1996.  Tokarek, R., Bernstein, E.F., Sullivan, F., Uitto, J. and Mitchell, J.B. Effect of therapeutic radiation on wound healing. Clin. Dermatol. 12: 57–70, 1994.  van den Brenk, H.A.S., Sharpington, C., Orton, C. and Stone, M. Effects of X-radiation on growth and function of the repair blastema (granulation tissue). II. Measurements of angiogenesis in the Selye pouch in the rat. Int. J. Radiat. Biol. 25: 277, 1974.  Wiernik, G., Patterson, T.J.S. and Berry, R.J. The effect of fractionated dose-patterns of X-radiation on the survival of experimental skin flaps in the pig. Br. J. Radiol. 47: 343, 1974.  Zaizen, Y., Ford, E.G., Costin, G. and Atkinson, J.B. The effect of perioperative exogenous growth hormone on wound bursting strength in normal and malnourished rats. J. Pediatr. Surg. 25: 70, 1990.