The human health threat posed by exposure to ionizing radiation continues to present itself on many fronts. In the post Cold War era, nonproliferation agreements have resulted in a new set of challenges that include the movement and dismantlement of nuclear weapons as well as the cleanup of radiological contamination at nuclear weapons production facilities and surrounding sites. A threat of nuclear terrorism now exists due to the possibility of theft from closed and/or ill-maintained Cold War facilities. In addition, the peacetime applications of radioactive materials continue to expand in the fields of energy production, medical diagnosis and treatment, and scientific and medical research. The continuing dangers are exemplified by a lethal exposure in 1994 in Estonia. One man was killed and several of his relatives were exposed to a stolen, discarded cesium source. Industrial accidents continue to occur, and severe overexposures may occur even during radiotherapy.

Findings concerning the complex nature of radiation injuries and of potential prevention, assessment, and treatment strategies were reported at the April 1993 Second Consensus Development Conference on the Treatment of Radiation Injuries. The conference, conducted in Bethesda, Maryland, was sponsored by the Armed Forces Radiobiology Research Institute and was supported in part by educational grants from Amgen, Inc., Thousand Oaks, California; Immunex Research and Development Co., Seattle, Washington; US Bioscience, Conshohocken, Pennsylvania; Monsanto/GD Searle Co., St. Louis, Missouri; Sandoz Cytokine Development Unit, East Hanover, New Jersey; and Nycomed/Bioreg Co., Oslo, Norway.

This volume reflects the conference proceedings in 38 papers from leading researchers in 18 countries. While a number of authors took the opportunity to update their contributions during an extensive review and coordination process, it is clear from the literature and results presented at recent meetings and symposia that, while significant insights have been gained into the regulation of hematopoietic stem cell proliferation and differentiation as well as into the pathogenesis of sepsis, little clinical documentation has been reported in the treatment of myelosuppression and in the prevention of the morbidity and mortality associated with infectious disease in the immunocompetent, let alone the myelosuppressed, host.

Therapeutic approaches to enhancing hematopoietic recovery from radiation or chemotherapy-induced myelosuppression have continued to focus on the evaluation of several lineage-specific cytokines for thrombopoiesis and the use of combination protocols or the single agent GM-CSF/IL-3 (granulocyte-macrophage colony-stimulating factor/interleukin 3) fusion protein PIXY321 to enhance dual lineage recovery of neutrophils and platelets. At the same time, the characterization of cytokine receptors, receptor antagonists, and antiendotoxin antibodies in conjunction with "blockade therapy" has focused renewed research efforts on endotoxin and the proinflammatory cytokines IL-1 and tumor necrosis factor (TNF) as key players in the pathophysiology of sepsis and septic shock.

Sepsis and septic shock have been the focus of an intense effort in preclinical research and clinical trials over the past decade. However, much to the dismay of the research and medical communities, new attempts at immunotherapy directed at endotoxin and inflammatory mediators and cytokines have not resulted in clear and reproducible beneficial effects in clinical trials.

Since 1982, over 15 multicenter clinical trials, using thousands of patients, have recorded no consistent reduction in short-term mortality from septic shock, which has consistently ranged from 28 to 40 percent.

Therefore, the recommendation of a new treatment modality for septic patients is precluded by the lack of any clear-cut beneficial effect from the many clinical trials in sepsis, trials using agents such as antibodies to core endotoxin and TNF as well as soluble receptors to TNF and the IL-1 receptor antagonist. It is apparent from the results of these trials, when considered in the light of increasing knowledge concerning the mechanisms and pathogenesis of sepsis, that combined treatment modality may be necessary to control this common, deadly, and complex disease. This approach is most likely to benefit the immunocompromised, irradiated patient for whom a beneficial result may require treatment designed to both augment host defense mechanisms and dampen the proinflammatory action of endotoxin and cytokines.

New generation hematopoietic growth factors and lineage-specific cytokines have moved into the various phases of cell biology examination, preclinical research, and clinical trials. Two of these new generation cytokines are represented by the flt3/flk2 ligand and c-mpl ligand, which is also known by the trade names megakaryocyte growth and development factor (MGDF) and thrombopoietin (Tpo). The flt3 ligand has shown promise as a potent synergistic hematopoietic growth factor for proliferation, expansion, and myeloid differentiation of primitive hematopoietic progenitor cells when used in combination with other multilineage and synergistic growth factors. At this writing, clinical trials have not been initiated.

The c-mpl ligand has shown potent effects on megakaryocytopoiesis and platelet production. It has lessened the degree and duration of thrombocytopenia in several animal models of radiation- or drug-induced thrombocytopenia. The data accumulated to date suggest that c-mpl ligand is a lineage-restricted, physiologic regulator of platelet production with important therapeutic benefits. It has recently entered clinical trials.

Several other hematopoietic growth factors are currently in various phases of clinical trials. These include the GM-CSF/IL-3 fusion protein PIXY321, IL-3, IL-6, IL-11, and the synthetic IL-3 receptor agonist, Synthokine. Phase I/II clinical studies in bone marrow transplantation and chemotherapy-induced myelosuppression have documented the ability of PIXY321 to promote platelet, neutrophil, and in some cases red cell recovery. Phase I clinical studies of IL-11 have demonstrated thrombopoietic activity in patients treated prior to chemotherapy. Phase II trials are in progress. IL-3, while documented as a multilineage hematopoietic growth factor with an important role in the viability, proliferation, and differentiation of hematopoietic progenitor cells, has proved equivocal in preclinical and clinical studies evaluating therapeutic efficacy in producing platelets and/or neutrophils consequent to radiation or chemotherapy-induced myelosuppression. IL-6 is also in the midst of clinical trial evaluation both as a single agent and in combination with G-CSF (Neupogen) or IL-3. The results suggest that IL-6 may accelerate hematopoietic recovery from chemotherapy-induced thrombocytopenia as well as that post autologous bone marrow transplantation, although transfusion requirements have not been reduced. The Synthokine however has shown preclinical therapeutic efficacy for the production of platelets while also significantly lessening the neutrophil nadir in a nonhuman primate model of radiation-induced myelosuppression. These results forecast the clinical potential of such an IL-3 alternative molecule.

Current directions in the treatment for severe radiation-induced marrow aplasia also include a renewed focus on transplantation of hematopoietic stem and/or progenitor cells. It is the source, activation state, function, and manipulation of the cells used for grafting that has generated the increased potential of transplantation in the severe radiation accident scenario. Cytokines have played a major role in this advance due to their ability to mobilize both pluripotent stem cells and committed progenitor cells from the marrow into the peripheral blood. The noted heterogeneity of the mobilized hematopoietic cells provides a marked advantage to this treatment modality in that the graft may be engineered to provide a high dose of committed progenitor cells that can accelerate the short-term engraftment while the content of stem cells can provide for long-term reconstitution. Recent clinical studies have emphasized the successful use of allogeneic peripheral blood/stem cell transplantation. In this regard the manipulation of the graft by CD34+ immunoadsorption can, while concentrating the hematopoietic stem and progenitor cells, also deplete the graft of alloreactive T-cells and thus reduce the potential for acute and chronic graft-versus-host disease (GVHD).

Selected cytokines have also shown utility in the ex vivo expansion of stem and progenitor cells regardless of source. A better understanding of the growth factors involved in expanding pluripotent stem cells versus committed progenitor cells will allow graft engineering and selective emphasis on short-term radioprotection versus long-term reconstitution. Such ex vivo expansion may be used as a primary graft or in support of primary engraftment, delayed engraftment, or secondary failure of the initial graft.

Additional advances in the transplantation of hematopoietic stem cells will undoubtedly come from the area of cord blood stem cell transplantation and/or ex vivo expansion of such cells. At this writing over 100 transplants have been performed with cord blood derived stem cells. The available data suggest that umbilical cord blood is an acceptable source of transplantable hematopoietic stem cells with low GVHD potential.

It is apparent from results to date that continued evaluation of alternative strategies is required to further reduce the obligate periods of neutropenia and thrombocytopenia consequent to radiation or drug-induced myelosuppression. The data in this volume continue to reflect the body of knowledge on which will be based further developments in radiation injury prevention, assessment, and treatment, particularly in the areas of sepsis and of the hematopoietic, gastrointestinal, and cutaneous syndromes.

It is recognized that as our ability to treat the hematopoietic syndrome improves, the gastrointestinal syndrome will emerge as the next acute dose-limiting organ system. Thus, consideration of current and potential treatment strategies for gastrointestinal radiation injuries is most appropriate. The dilemma facing the researcher and clinician lies in the multifactorial consequences and different etiology of radiation damage to the gastrointestinal system. Treatment strategies developed to correct for the dysfunction in motility and epithelial cell transport that contribute to the severe and debilitating diarrhea characteristic of this syndrome, will most likely differ from those directed at regenerating the gastrointestinal stem cells and epithelial mucosa. This approach may be somewhat similar to that taken with the hematopoietic system in that stem cells are responsible for repopulation of the gastrointestinal tract while the viability and integrity of mucosal epithelial cells are necessary for a functioning organ. Results from definitive clinical trials relative to evaluation of treatment modalities from above-mentioned radiation- and/or chemotherapy-induced gastrointestinal injury are not available.

It has become apparent that regulation of proliferation and differentiation of epithelial stem cells and the viability and function of mucosal cells is accomplished through a complex interaction between growth factors, the extracellular matrix, and the basal lamina. New pleiotropic cytokines such as IL-11 and IL-15, and a family of protease-resistant peptides designated trefoil factors may play important roles in the regulation of mucosal cell proliferation and integrity. Effective therapuetic protocols will depend upon deciphering the complex interactions between growth factors such as cytokines and trefoil peptides, matrix constituents, lamina propria, neural mediators, and luminal contents.

New ideas are forthcoming and should lead to clinical trials that will provide information on the efficacy of treatments for radiation-induced cutaneous injuries.

The editors gratefully acknowledge the contributions of numerous individuals to the success of the conference and the publication of this volume. The consensus panels were as follows.

Consensus summary on the treatment of radiation injuries

The radiation environment during a nuclear accident or disaster is likely to be uncontrolled and ill defined. This was apparent in the three most recent and largest accidents—the reactor explosion in Chernobyl, U.S.S.R. (1986); the internal and external exposure of victims to cesium-137 in Goiania, Brazil (1987); and the exposure of workers in a radiation sterilization facility in San Salvador, El Salvador (1989). The radiation may differ in quality, energy, and dose rate. The exposure may be nonuniform and heterogenous; it may involve only part of the body, depending on the location, position, and movement of the subject in relation to the radiation source and available shielding. A subject may also have been traumatized by burns and/or wounds.

Many individuals who have received total-body irradiation at doses that lie within the range that causes the hematopoietic syndrome will require intensive medical care. A patient with combined injuries—one who has been traumatized as well as irradiated—will require additional care for the associated trauma and consequent immunosuppression. The ill-defined and uncontrolled nature of radiation exposure and nuclear accidents usually forecasts a nonuniform exposure, with the variable dose distribution suggesting a possible sparing of some bone marrow and/or gastrointestinal stem cells. With this in mind, it is possible that recovery may occur even when radiation doses extend well into the lethal range of the hematopoietic and gastrointestinal syndromes.

Success with therapeutic regimens of replacement (platelets, fluids, and electrolytes) and substitution (antibiotics) supports the concept that infection and hemorrhage are the primary factors in the lethal consequences of radiation exposure in the hematopoietic syndrome. Successful treatment depends on the fact that damage to the stem cell system is reversible and that the surviving fraction of hematopoietic stem cells are capable of spontaneous regeneration.

These supportive and maintenance therapies were successfully used after the nuclear disaster in Chernobyl. Patients with clinical manifestations of the acute radiation syndrome were treated using isolation, antimicrobial decontamination of the intestine, systemic antibiotics, and transfusion of red blood cells and platelets.

Supportive procedures for empirical antiinfection treatment were highly effective. There were almost no deaths caused solely by infection in patients experiencing acute radiation sickness.

The immediate treatment of associated injuries is a primary determinant of survival. The degree of radiation-induced marrow aplasia (reversible or irreversible) may not be known for days because of the uncontrolled nature of the exposure and its potential to be nonuniform and heterogeneous. These facts make reliance on a physical dose estimate impossible and underscore the need for monitoring biological parameters to estimate the severity of injuries and the probability of survival. Reliable triage and good clinical care based on comprehensive biological data will ensure the best chance for casualty recovery should a critical number of stem cells survive the radiation exposure.

Data for uncomplicated human radiation exposures within the hematopoietic syndrome range are relatively limited. The evidence (exclusive of five recent accidents: Chernobyl, Goiania, San Salvador, Belarus and Israel) comes from five primary sources: (1) persons exposed to the nuclear weapons detonations in Hiroshima and Nagasaki (2), therapeutic total-body bilateral exposure to gamma radiation of persons with Ewing's sarcoma (3), two nuclear accidents (Vinca, Yugoslavia; and Oak Ridge, TN) involving nine people exposed to mixed neutron and gamma radiation (4), two radiation accidents (New Jersey) involving two people exposed to cobalt-60 radiation, and (5) bone marrow transplant recipients receiving total-body irradiation. Information collected from these exposures and recent radiation accidents as well as experimental data provide a basis for a consensus on the treatment of radiation injuries.

Consensus on a treatment protocol
The consensus presented here is based on the discussions of four panels (Marrow Aplasia, Sepsis, Gastrointestinal Injury, Cutaneous Radiation Syndrome) at the Second Consensus Development Conference as well as deliberations at the First Consensus Development Conference (1) and other background material. The references given at the end of this chapter, in general, do not duplicate the references provided by the other authors in this book.

Based on the severity of the radiation exposure, casualties can be classified into four treatment categories: mild, moderate, severe, and lethal. Although a consensus was not reached on the specific dose ranges for these categories, primarily because of the difficulty in converting an air dose to a meaningful tissue dose. While the goal of the categorization is to increase the likelihood of survival with appropriate triage, the most reliable guide for treatment is the change in levels of blood cells and cytogenetics.

Medical management of radiation injuries
The medical management of radiation and/or combined injuries can be divided into three stages: triage, emergency care, and definitive care. Quality care can be provided when there are few casualties and a well-equipped facility. The therapeutic modalities will vary according to current medical knowledge and experience. Precisely implemented modalities will depend on the number of casualties, available medical facilities, and resources. The panel concurred that the recommendations for the treatment of a few casualties may not apply to the treatment of mass casualties because of limited resources.

Triage
During the triage stage, the triage director categorizes victims by degree of injury from trauma, radiation, or their combination, and estimates chances for survival. Triage is extremely important in mass casualty situations that require allocation of limited medical resources to those who require treatment and have the best chances of survival. Because there is no immediate life-threatening hazard from a survivable dose of radiation alone, the associated injuries should be treated first.

An acute radiation syndrome flowchart for each patient should be maintained. The flowchart provides a continuity of information throughout prolonged therapy that may be delivered by various caregivers and in different facilities.

Clinical monitoring and laboratory testing
Estimation of the degree of radiation damage and exposure is difficult. Prodromal symptoms, which begin within hours of exposure, are characterized by gastrointestinal and neurovascular signs and symptoms, including nausea, vomiting, diarrhea, fatigue, weakness, fever, and headache. The gastrointestinal symptoms generally do not last longer than 24-48 hours after exposure. The time of onset, severity, and duration of these signs are dose dependent and dose-rate dependent and should be used in conjunction with early biological parameters, such as granulocyte and lymphocyte levels, to determine the presence and severity of the acute radiation syndrome. These signs and symptoms must be used initially to determine if the casualty is one that is related to radiation exposure.

The rate and degree of decrease in blood cells are dose dependent. Blood samples should be taken daily during the first 2 weeks. A useful rule of thumb: If lymphocytes have decreased by 50 percent and are less than 1 x 10exp9/L within 24-48 hours, the patient has received at least a moderate dose of radiation. In combined injuries, lymphocytes may be an unreliable indictor. Patients with severe burns and/or trauma to more than one system often develop lymphopenia. Associated injuries (trauma/burn) should be assessed by standard procedures, keeping in mind that the signs and symptoms of tissue injuries can mimic and obscure those caused by acute radiation effects.

Laboratory testing is primarily focused on protecting the patient and determining biological dosimetry. The following are necessary studies:

• Evaluation of patient and excreta (urine/stool) for residual radioactivity
• CBC/differential/platelet counts daily for minimum of 1 week
• HLA-subtyping before lymphocyte count falls
• CMV, HbsAg, HCVAb, HIV, VSV

When possible, laboratory testing should include the following:

• Immediate lymphocyte cytogenetics
• Bone marrow biopsy and aspirate (potentially from both pelvis and sternum). Cells/sections should be frozen for future analysis.
• T4, TSH assay
• Serum and plasma for frozen storage
• Storage of clothing for later dosimetry evaluation
• Dental biopsy for dosimetry testing
• Autopsy in the case of fatal accidents, photograph of relevant findings, and storage of frozen tissues. Investigational studies: Serial whole-body magnetic resonance imaging (MRI) to evaluate bone distribution of bone marrow irradiation and in vitro culture of bone marrow cells for hematopoietic progenitor cell growth unless hospitalization is mandated by trauma, burns, or other serious illness. Monitoring can be performed on an outpatient basis. Patients with potential exposure to radioactive iodine should receive cold iodine (Lugol's solution) to block thyroid uptake of radioisotope.

Effective triage relies on accurate analysis of early signs and symptoms, substantiated by biological parameters associated with radiation exposure. Treatment must be determined after assessment of the radiation injury and the indicated therapies for the type of organ damage. Conventional injuries that are benign by themselves can become lethal when combined with radiation exposure.

Emergency care
Following a radiation accident, injured persons should be stabilized (intravenous fluids, bandages, and ventilatory support provided) at the site before transportation to local treatment centers for further evaluation and treatment. Those with trauma and burns and without radiation injury (determined by their location at the time of the accident or by the lack of general prodromal symptoms such as nausea and vomiting) should be referred to specialized treatment centers according to their injuries. Patients who probably suffered from radiation injury (identified by signs and symptoms) should be referred to treatment centers that can evaluate and treat bone marrow failure. Radiation injuries, burns, or multiple traumas will be treated at these centers by a team. Treatment should be managed by representatives from hematology, infectious disease, surgery, and psychiatry services. Other centers should be alerted to the possible need for platelets and red cells, and they should verify that all blood products have been irradiated. Hospitals in the vicinity should be alerted to the possible need for supplies and blood products.

Therapeutic support for severely irradiated patient: hematopoietic syndrome
Conventional Therapy of Neutropenia and Infection: Prevention and Management of Infection. The quantitative relationship between the degree of neutropenia and increased risk of infectious complications was demonstrated by Bodey and colleagues (2) almost 30 years ago. Empiric antimicrobial therapy for the febrile, neutropenic patient was first proposed bySchimff and colleagues (3) more than 20 years ago. The consensus report by the Working Committee of the Infectious Disease Society of America (4) stated that antibiotic prophylaxis should only be considered in afebrile patients at the highest risk for infection; that is, profound neutropenia (< 100 cells/µL) with an expected duration of greater than 7 days. Although the degree of neutropenia (absolute neutrophil count, ANC < 100/µL) is the greatest risk factor for developing infection, other factors influence the treatment choice and outcome. Such factors include duration of neutropenia, bactericidal function of surviving neutrophils, alteration of physical defense barriers, the patient's endogenous microflora, and organisms endemic to the hospital and community. It has also become apparent that, as the duration of neutropenia increases, the risk of secondary infections such as invasive mycoses also increases. It is for these reasons that adjuvant therapies such as the cytokines GM-CSF (granulocyte-macrophage colony-stimulating factor) or G-CSF may prove invaluable in the treatment of the severely irradiated person.

Prevention of Infection. Initial care of medical casualties with moderate and severe radiation exposure should probably include early institution of measures to reduce pathogen acquisition, with emphasis on low-microbial-content food, acceptable water supplies, frequent hand washing (or wearing of gloves), and air filtration. Prophylactic use of selective gut decontamination with antibiotics that suppress aerobes but preserve anaerobes is recommended. These measures help control the alimentary canal source (mouth, esophagus, and intestines) of postinjury infections. Maintenance of gastric acidity (avoidance of antacids and H2 blockers) may prevent bacteria from colonizing and invading the gastric mucosa and may reduce the frequency of nosocomial pneumonia due to aspiration of these organisms. The use of sucralfate or prostaglandin analogues may prevent gastric hemorrhage without decreasing gastric activity. When possible, early oral feeding is preferred to intravenous feeding to maintain the immunologic and physiologic integrity of the gut. Surgical implantation of a subcutaneously tunneled central venous catheter should be considered to allow frequent venous access. Meticulous attention to proper care is necessary to reduce catheter-associated infections.

Management of Infection. The management of established or suspected infection (neutropenia and fever) in irradiated persons is similar to that used for other febrile neutropenic patients, such as solid tumor patients receiving myelotoxic chemotherapy. First, an empirical regimen of antibiotics should be selected, based on the pattern of bacterial susceptibility and nosocomial infections in the particular institution and the degree of neutropenia. Broad spectrum empiric therapy with high doses of one or more antibiotics, avoiding aminoglycosides whenever feasible due to associated toxicities, should be used. Therapy should be continued until the patient is afebrile for 24 hours and ANC is greater than or equal to 500/µL. Combination regimens often prove to be more effective than monotherapy. The potential for additivity or synergy should be present in the choice of antibiotics.

Modifications of this initial antibiotic regimen should include a thorough evaluation of the history, physical findings, laboratory data (including chest radiograph), and epidemiological information. Antifungal coverage with amphotericin B should be added, if indicated, for patients who remain persistently febrile for 7 days or more on antibiotic therapy in association with clinical evidence of infection, or if they have new fever on or after day 7 of treatment with antibiotics. If there is evidence of resistant gram-positive infection, vancomycin should be added. If diarrhea is present, stool cultures should be examined for Salmonella, Shigella, Camphlobacter and Yersinia. Oral and pharyngeal mucositis and esophagitis suggests Herpes simplex infection of candidiasis. Empiric acyclovir or antifungal therapy should be considered.

Surveillance cultures may be useful for monitoring acquisition of resistant bacteria during prophylaxis and emergence of fungi. A once or twice weekly sampling of surveillance cultures from natural orifices and skin folds (e.g., axillae, groin) would be reasonable, but should be modified based on the institutional patterns of nosocomial infections. A chest radiograph should be considered at initiation of empiric therapy. This may aid in definitive diagnosis of a new pulmonary infiltrate obtained during the course of neutropenia.

Several principles (5) have been stated that are generally applicable to the febrile neutropenic patient and provide a foundation upon which a specific initial regimen may be selected.

• Principle 1: The spectrum of infecting organisms and antimicrobial susceptibility patterns vary among institutions and over time.
• Principle 2: Life threatening, gram negative bacterial infections are universal among neutropenic patients, but the prevalence of life-threatening, gram-positive bacterial infections varies greatly among institutions.
• Principle 3: Current empiric antimicrobial regimens are highly effective for initial management of febrile, neutropenic episodes.

Overall recommendations (5)

1. A standardized plan for the management of febrile, neutropenic patients must be devised.
2. Empiric regimens must contain antibiotics broadly active against gram-negative bacteria, but antibiotics directed against gram-positive bacteria need be included only in institutions where these infections are prevalent.
3. No single antimicrobial regimen can be recommended above all others, considering the premise of principle 1 that pathogens and susceptibility vary with time
4. If infection is documented by cultures, the empiric regimen may require adjustment to provide appropriate coverage for the isolate. This should not narrow the antibiotic spectrum.
5. If the patient defervesces and remains afebrile, the initial regimen should be continued for a minimum of 7 days.

Growth Factors. Hematopoietic growth factors, such as G-CSF and GM-CSF, are potent stimulators of hematopoiesis. Preclinical studies in nonhuman primates, rodents, and canines have demonstrated significant reduction of the duration of neutropenia and time to recovery of neutrophils to preirradiated values. In large animal models of severe radiation-induced myelosuppression where clinical support in the form of antibiotics and fresh, irradiated platelets or whole blood is used concurrently with G-CSF or GM-CSF, a marked reduction in infectious complications translated to reduced morbidity and mortality. It is documented that the risk of infection and subsequent complications are directly related to depth and duration of neutropenia (2).

G-CSF or GM-CSF have been shown to reduce the incidence of febrile neutropenia in several major randomized trials in adults. The American Society of Clinical Oncology (ASCO) consensus group (6) and this consensus recommend primary administration of CSF for patients expected to receive significant levels of febrile neutropenia consequent to severe myelosuppression. In this case the CSFs should be administered prior to any occurrence of neutropenia or febrile neutropenia. Further efficacy of CSFs has been shown in the significantly shortened period of neutropenia and reduction in infectious complications in patients with severe myelosuppression with autologous bone marrow transplant.

The beneficial role of G-CSF or GM-CSF in reducing the duration and degree of neutropenia cannot be underestimated. It has been determined that the longer the duration of severe neutropenia the greater the risk of secondary infections, especially with invasive mycoses. An additional benefit of the CSFs is their ability to increase the functional capacity of the neutrophil and thereby contribute to the prevention of infection as an active part of cellular host defense.

Recommendations for use of G-CSF or GM-CSF for patients expected to experience severe levels of febrile neutropenia are as follows:

• Dose, Route
• G-CSF 2.5-5.0 µg/kg/day (100-200 µg/mexp2/day)
• GM-CSF 5.0-10.0 µg/kg/day (200-400 µg/mexp2/day)
• Both CSFs administered subcutaneously (s.c.) or intravenously (i.v.); s.c. recommended; QD protocol

Initiation and Duration. The appropriate timing of CSF administration relative to the radiation exposure as well as the duration and schedule of CSF are important concerns in trying to achieve maximum clinical response. ASCO recommends that G-CSF or GM-CSF be started within 24 to 72 hours subsequent to the exposure to provide the opportunity for maximum recovery. CSF administration should continue, with daily consecutive injections, through the desired effect of an ANC of 10 × 10exp9/L subsequent to the ANC nadir (6).

The delayed initiation of CSF therapy in the case of radiation accidents is a persistent problem. The recommendation is to initiate CSF therapy (as stated above) without significant delay.

Although reports are available that suggest the delay of CSF administration did not compromise recovery of neutrophils (7-9), others suggest that delay of CSF administration results in loss of therapeutic efficacy (10-14). The lack of a clear beneficial effect relative to CSF scheduling may be due, in part, to the variable degrees of myelosuppression used in the preclinical and clinical studies available. Schedule variation may be less well tolerated in the cases of severe myelosuppression.

Comparative Efficacy of CSFs. There is no data available from large-scale, prospective clinical trials evaluating comparative CSF efficacy. Physician preference and experience should dictate the choice of CSF.

Comparative Toxicity of CSFs. The predominant side effect noted with administration of G-CSF is medullary bone pain observed shortly after initiation of G-CSF treatment and again just before onset of neutrophil recovery from nadir. G-CSF may exacerbate preexisting inflammatory conditions. The most noted side effects with administration of GM-CSF are fever, nausea, fatigue, headache, bone pain, and myalgia. The recent ASCO consensus statement states that it is not clear whether side effects of G-CSF or GM-CSF differ markedly when conventional doses are administered (6).

It should be noted that there are two forms of GM-CSF, sargramostim, which is glycosylated, and molgramostim, which is not. The literature suggests that glycosylated GM-CSF has a longer serum half-life, greater neutrophil stimulating activity, less leukotriene production, and fewer side effects. Fluid retention, dyspnea, fever, and myalgias/bone pain may be less prominent with the yeast-produced formulation sargramostim (15,16).

Other considerations
Radiation Factors. Consideration must be given to the type of radiation exposure (internal or external), the presence of nonhematopoietic organ injury, and the radioisotope. Growth factor therapy may not be helpful and may be detrimental where there is prolonged internal irradiation by long-lived isotopes. Potential risks include acceleration of leukemogenesis and prevention of neutrophil migration into soft tissues. Early granulocyte recovery following growth factor therapy may indicate lower-than-estimated radiation dose exposure.

Data on growth factors administered to accidentally irradiated humans is limited; and in each case, administration occurred at 1 week or more postirradiation (17-20). In 11 cases the growth factor was administered about 3-5 weeks postexposure. In 13 cases, results have not yielded data that could be evaluated. Earlier administration of growth factors (1-3 days postexposure) may prove more beneficial in inducing granulocytosis.

Antibiotic-Induced Release of Endotoxin. Endotoxin, the lipopolysaccharide components of gram-negative bacterial cell membranes, are potent inducers of the proinflammatory cytokines, including tumor necrosis factor (TNF) and interleukin-1 (IL-1) that are known to play a pivotal role in the pathogenesis of sepsis. It has been shown in experimental models, in vitro and preclinical, as well as in clinical studies that endotoxin concentration in the blood and urine can increase following antibiotic treatment of gram-negative infections. It has been demonstrated that antibiotics (equivalent minimum inhibitory concentrations, MICs) differ in their capacity to cause endotoxin release from gram-negative bacteria. It has been suggested from preclinical studies that this effect may translate into increased morbidity and mortality. The clinical relevance of these differences between antibiotics in endotoxin-releasing potential is unknown.

A recent randomized, double-blind, clinical study (21) comparing imipenem (a low endotoxin-releasing antibiotic) with ceftazidime (a high endotoxin-releasing antibiotic) in the treatment of gram-negative urosepsis suggests that differences in endotoxin release may influence the speed of defervescence early after the start of treatment and may also affect the inflammatory response in gram-negative sepsis. Serum and urine cytokine levels increased after 4 hours of ceftazidime treatment; there were no increases in the imipenem-treated patients. Further preclinical and clinical research is required for a definitive recommendation. These considerations may be of value in developing an antibiotic protocol for severely irradiated patients.

Granulocyte Transfusions (GTX) from CSF-Stimulated Donors. Extensive literature (22-24) is available regarding the therapeutic and prophylactic efficacy of transfusing granulocytes (polymorphonuclear neutrophils, PMNs) into neutropenic individuals. Two major problemsþcell dose and alloimmunizationþhave limited the applicability and evaluation of granulocyte transfusions.

The first problem of cell dose may be overcome through the use of G-CSF to significantly increase the number of donor circulating PMNs over a duration that would allow multiple sampling (25,26). The use of HLA-compatible donors would possibly avoid the problem of alloimmunization. An additional benefit may be derived from the fact that G-CSF enhances the phagocytic and microbicidal activity of stimulated PMNs.

Besinger et al. (25) recently reported that G-CSF was safe to administer to normal individuals, significantly improved the quantity of granulocytes collected, and when transfused resulted in significant circulating levels of granulocytes in neutropenic patients.

GTX of G-CSF-stimulated PMNs could prove effective therapy for severely neutropenic patients with sepsis and who have failed to respond to appropriate antibiotic therapy. Patients with prolonged, severe neutropenia caused by continuing bone marrow failure as well as those with refractory infection may benefit.

Further preclinical and clinical research is required to elucidate the effects of G-CSF in normal donors, the quality of the stimulated and harvested PMNs, and the efficacy of G-CSF-stimulated GTX as treatment for infections in neutropenic hosts.

Selective Decontamination of the Digestive Tract. The approach of selective decontamination has not led to fewer febrile episodes or to a lower mortality in neutropenic patients (27). Based on an analysis of current risks and benefits, selective decontamination should be considered experimental and cannot be recommended as a routine preventive strategy (28). However, Schimpff (29) has concluded that the patient who is likely to benefit from an intensive, oral nonabsorbable antibiotic regimen is one who is likely to experience prolonged, profound granulocytopenia, especially when related to some degree of mucosal damage. It is in this context that we extend our recommendation for use of selective decontamination. Verhoef (27) states that the rationale for selective decontamination with trimethoprim-sulfamethoxazale (TMP-SMZ) or quinolones was that the elimination of potentially pathogenic aerobic gram-negative bacilli from the gastrointestinal tract would prevent colonization and subsequent infection. The noted efficacy of TMP-SMZ or quinolones in decreasing the frequency of gram-negative infections is clouded by the inability to attribute the efficacy to selective gut decontamination or increased tissue levels of the drugs.

New Adjuvant Therapies. Sepsis and its sequelae are the major causes of death in severely irradiated patients. This is also true in medical and intensive care units in the United States. In spite of an intense research effort toward identifying causative inflammatory mediators and therapeutic protocols for the treatment of this lethal condition, relatively few advances have occurred. In recent years, several candidate therapies, including immunomodulators, corticosteroids, antiendotoxin antibodies, and inflammatory cytokine antagonists, have been evaluated. While significant insights have been realized into the pathophysiology of sepsis, none of the clinical trials to date have achieved statistically significant reductions in mortality according to the prescribed criteria for efficacy.

Conventional Therapy of Thrombocytopenia: Platelet Support. The requirement for platelet support depends on the patient's condition. In irradiated patients with or without other major medical problems (infection, gastrointestinal problems, or trauma), the platelets should be maintained at greater than 20 × 10exp9/L. The need to maintain platelet counts at 20 × 10exp9/L has recently been questioned (30). Analysis of platelet counts versus hemorrhage suggests that 10 × 10exp9/L is adequate in the absence of any indication of accompanying frank hemorrhage. If surgery is needed, the platelet count should be greater than 75 × 10exp9/L. Transfusion of platelets remains the primary therapy to maintain adequate platelet counts. As general supportive measures, one should avoid the use of aspirin and NSAIDS.

Limited platelet support is likely to come from random donors. Should refractoriness develop, family members as well as HLA-compatible donors from the general population can be considered as platelet donors. The use of platelet products from which white blood cells have been removed is desirable to minimize both allosensitization and the risk of transmission of viral illnesses, such as cytomegalovirus. All blood products should receive 15-20 Gy of radiation before infusion to prevent graft-versus-host disease through infusion of mononuclear cells present in the products. If a transplant is to be performed, the use of platelets from related donors should be avoided.

Growth factor/cytokine therapy for thrombocytopenia
Use of thrombopoietic agents immediately after radiation injury is not currently recommended for the following reasons: (1) thrombocytopenia is not generally life-threatening in this situation, (2) evidence of benefit is not yet available from chemotherapy models, (3) extramedullary toxicities or compromise of CSF benefit might prove detrimental, and (4) clinical evaluation of native IL-3 has proven equivocal to date. However, further drug development may alter the accepted pattern of care, and the introduction of PIXY321, IL-6, IL-11, megakaryocyte growth and development factor/thrombopoietin (MGDF/Tpo) or the synthetic IL-3 receptor agonist Synthokine early in the postradiation accident setting may be appropriate in the near future.

Use of thrombopoietic agents (PIXY321, IL-6, IL-11, MGDF/Tpo, or Synthokine) should be considered in a patient who has neutrophil recovery but remains platelet transfusion dependent after accidental irradiation. In this situation, therapy should be initiated at a subcutaneous (s.c.) phase II dose recommended by the manufacturers and titrated based on response and patient tolerance. Initial treatment should probably be attempted for 2-3 weeks with a posttreatment observation period of 1-2 weeks.

Conventional therapy of anemia
Transfusion of peripheral red blood cells (PRBCs) remains the primary therapy to maintain hemoglobin above 8 gm/dl. PRBC transfusions should be irradiated, leukocyte-filtered (whenever possible), and from an unrelated donor if allogeneic transplantation is a consideration.

Risks of PRBC transfusion may include cytomegalovirus (CMV) transmission and alloimmunization.

Erythropoietin Therapy of Anemia. Use of Epo after radiation injury is not recommended even though it is likely to be safe. Anemia is not generally life-threatening in this situation; endogenous Epo levels are often elevated already after highly cytotoxic therapy; and evidence of benefit is not yet available from clinical chemotherapy models.

Bone marrow transplantation
The dilemma presented to the physician is whether or not a transplant is required to ensure the best chance of survival for the severely irradiated person. We know that total-body irradiation produces dose-dependent toxicity to the bone marrow. Increasing doses lead to more protracted pancytopenia and increasing risk of death from infection and/or hemorrhage. The dose of radiation to the bone marrow that would effectively prevent recovery within a period required for survival is unknown. We also know that this dose can be effectively increased with the use of conventional clinical support and cytokine therapy. Preclinical studies using a canine model of lethal whole-body irradiation have shown that therapeutic support (including fluids, platelets or whole blood, antibiotic regimens and administration of G-CSF) has induced survival in otherwise uniformly lethal doses of radiation. When experiments were performed over a range of lethal doses, clinical support alone resulted in a dose reduction factor (DRF) of 1.3 whereas the addition of G-CSF therapy increased the DRF to approximately 2.0. The radiation alone value for the LD50/30 for canines of 260 cGy was increased respectively to 340 cGy and 502 cGy (31).

A reasonable estimate for an LD50/60 in humans receiving clinical support is 600 cGy free-in-air for total-body uniform exposure. The use of broad spectrum antibiotics, blood product support, and hematopoietic growth factors may provide a human a reasonable chance of surviving a radiation dose of 800 cGy or greater. Considerations revolve around knowledge of dose uniformity, heterogeneity, dose rate, and total dose received in addition to the facts that clinical support and cytokine therapy combine to hasten recovery and increase survival in otherwise lethally irradiated animal models. Furthermore, the use of G-CSF or GM-CSF is recommended in high-risk patients such as those exposed to acute, high-dose radiation with or without associated trauma. The subset analysis of several clinical trials has suggested that patients with protracted neutropenia may benefit the most from the therapeutic use of CSFs as adjuncts to antibiotic therapy. The evidence-based review cited by the ASCO report (6) suggests that CSFs can successfully shorten the period of neutropenia and reduce infectious complications in patients undergoing high-dose cytotoxic therapy with autologous bone marrow transplantation (ABMT). The benefit of CSF use in patients undergoing allogeneic BMT (allo-BMT) is not clearly defined.

The major complications following ABMT are the same as those consequent to severe radiation exposure. Obligate periods of neutropenia and thrombocytopenia present the increased risk for infection and hemorrhage, the need for platelet and/or red blood cell (RBC) transfusions and delayed or incomplete reconstitution, and potential “other organ” damage. Allo-BMT also presents the risk of graft-versus-host disease (GVHD) and graft rejection. According to Miller et al., (6) “There has been no evident increase in complications of graft-versus-host disease, graft rejection, or relapse associated with CSF use in randomized studies of allo-BMT.” (32–36)

Physicians should consider allo-BMT under the following circumstances:

• A fully matched sibling donor is available.
• The patient has an absolute lymphocyte count (ALC) < 1000/µl.
• The radiation dose is unknown or is likely to be under 2,000 cGy (give patient benefit of doubt).
• There are no other injuries that preclude survival or preclude transplantation (e.g., severe burns).
• Irradiation is not ongoing from an internal source.

Following are recommended transplantation conditions:

• Marrow should be T-cell depleted to avoid GVHD (consider aliquot of unpurged marrow as backup in case of engraftment failure).
• The marrow is T-cell depleted; no specific age limit need apply.
• The transplantation can be performed promptly after irradiation.
• No immunosuppression is given to enhance the likelihood of marrow engraftment.

The timing of marrow grafting is crucial and presents a dilemma. Animal data suggest that the marrow should be infused within the first 3-5 days of radiation exposure. This coincides with the peak period of immunosuppression; and graft rejection, therefore, will be less likely. These findings stress the importance of developing reliable clinical and laboratory parameters to assess the degree of radiation damage to the marrow as quickly as possible and to determine which patients should be given marrow transplants. Waiting for a week or longer after the radiation exposure would require some form of immunosuppressive treatment to prepare the patient for a marrow graft. Such treatment may be less tolerated by the patient who is a radiation accident victim. The marrow transplant must be carried out at a treatment facility equipped for such procedures.

Timing of the marrow graft should also be viewed in the context of recommended cytokine therapy for the neutropenic patient. Assuming all indications point toward a high radiation exposure, it is recommended that cytokine therapy be initiated as soon as possible. In this case the patient would have been treated with G-CSF or GM-CSF prior to the decision to transplant. In addition, further drug development and results from ongoing clinical trials may alter the recommendations with the introduction of thrombopoietic agents such as IL-6, IL-11, PIXY-321, MGDF/Tpo, or Synthokine. The effects of prior cytokine administration on marrow engraftment is unknown.

CSF Administration. The same doses, routes, and schedules of CSF administration (G-CSF or GM-CSF) as mentioned previously for the neutropenic patient should be followed in the transplant setting.

Delayed Engraftment. The occurrence of delayed or inadequate primary marrow engraftment or secondary graft failure following autologous or allogeneic stem cell transplantation can be treated with CSF administration and/or peripheral blood progenitor cell transfusion. According to Miller et al. (6), “An attempt at CSF therapy is reasonable in view of the high infectious mortality rates in this population and the general desperation inherent in this situation.” Only limited data is available assessing the value of CSFs in stimulating recovery and improving survival of patients with poor engraftment after transplant (37-43).

Peripheral blood stem cell transplantation (PBSCT)
There is increasing evidence that PBSCT of cells mobilized by growth factors are capable of reliable, rapid, and durable autologous hematopoietic engraftment (44,45). An early assumption is that autologous mobilized (primed) PBSCT offered more rapid recovery of granulocytes and platelets than BMTs derived from normal, resting marrow. A recent comparison of engraftment from primed PBSCs to that derived from the primed marrow after the growth factor-induced mobilization suggests mobilized and primed PBSCs are comparable to primed BMSCs derived after the same GF stimulation (46).

These data derived from autologous PBSCT suggest that mobilized PBSCs might also shorten the duration of cytopenia in the setting of allogeneic transplantation. Cautious use of allogeneic PBSCT was based on the unknown toxicities from cytokine administration in donors and the increased risk of GVHD from the large number of T-cells infused. However, early results indicated that unmanipulated allo-PBSCT can be performed with cytokine-mobilized cells and provide rapid engraftment without an appreciable greater incidence of acute GVHD than expected with allo-BMT (47-52). Further evaluation is necessary to determine the effect of the large T-cell component of PBSCTs on chronic GVHD. Further use of graft engineering is required. A step in this direction has already been taken with the use of CD34+ immunoabsorbent columns for positive selection of CD34 cells for the allograft. Such positive selection can achieve a several-log depletion of T-cells, thereby achieving in one step both the enrichment of PBSCs and the depletion of T-cells (44,53,54). It is probable that the average yield from two apheresis procedures of mobilized CD34+ cells can be several-fold greater than an average single bone marrow harvest.

Cell Dose for PBSCT. The critical number of allo-CD34+ cells from growth factor mobilized blood to ensure a durable graft is unknown. An optimal autologous PBSCT CD34+ cell number is in the range of 2-5 × 10exp6 per kg bw (body weight). The estimate for allo-PBSCT CD34+ cells is in the range of 5-15 × 10exp6 per kg. This should be achievable with two apheresis.

Delayed Marrow Engraftment. Cryopreserved, mobilized PBSCs have been successful in the treatment of graft failure (55-57). It is reasonable that allo-PBSCs be harvested and cryopreserved in the event of allo-BMT failure to engraft or secondary graft failure. Efforts should also proceed to evaluate the engraftment potential of primed bone marrow derived from the GF-stimulated donor at the approximate time of PBSC harvest by apheresis.

Combination Allo-BMT Plus Allo-PBSCT. Further consideration must also be given to the combination transplant protocol of PBSC plus BM transplantation (58,59). Such a protocol would potentially take advantage of the dual engraftment properties available in PBSC grafts. The mobilized peripheral blood cells contain large quantities of committed progenitors in addition to hematopoietic stem cells. These committed progenitors would provide for an earlier, although unsustained, phase of engraftment. The more primitive stem cells contained in both the PB and BM graft would then provide for the later, durable, long-term reconstitution.

Cytokine Dose and Duration for Mobilization. The predominant use of G-CSF (Filgrastim) for mobilization of PBSCs is in the dose range of 2-16 µg/kg/day. The higher doses have been administered as a divided dose, s.c., twice a day for 5 or 6 consecutive days. The lower doses have been administered once a day, s.c., for the same duration. Leukapheresis can be performed for 2 consecutive days beginning on day 5 of G-CSF administration. The specific protocol and dose of G-CSF should be the choice of the transplant team. The choice of cytokine as a single agent or in combination with other GFs should also be the choice of the transplant team, based on experience and current knowledge of GF-induced mobilization efficiency. The cytokines GM-CSF, c-kit ligand, erythropoietin, and IL-3 have been used both alone or in combination to induce PBSC mobilization. It is probable that the flt-3 ligand and the Synthokine molecule will be evaluated for mobilization potential.

Therapeutic support for the severely irradiated patient: gastrointestinal syndrome
The effects of radiation on the gastrointestinal tract and the associated symptomatology can be categorized into four major phases that correspond to the elapsed time from exposure to manifestation. The phases are (1) prodromal phase, in which nausea, vomiting, and diarrhea occur minutes to hours after exposure; (2) subacute phase, in which diarrhea and vomiting occur hours to days after exposure; (3) acute phase, in which diarrhea, toxemia, and septicemia occur days to weeks after exposure; and (4) chronic phase, in which fibrosis, bleeding, and fistulas occur months to years after exposure. Of particular interest to this meeting were those symptoms associated with the prodromal, subacute, and acute phases of gastrointestinal injury. As discussed elsewhere in this volume, the mechanisms underlying these three phases are likely to be different. Thus, treatments directed against these effects will require at the least a bimodal approach.

Areas of consensus
Nausea and vomiting associated with the prodromal effects of radiation exposure can be prevented and/or ameliorated by the new generation of 5HT3 antagonists such as ondansetron and granisetron. The efficacy of these agents in controlling prodromal diarrhea is presently being established in a number of clinical trials. However, in cases in which radiation dose is unknown (e.g., in the case of accidental exposure), the early use of an antiemetic may be precluded since the time of onset and the duration of vomiting can be useful in dose estimation. An antiemetic can then be given once emesis has commenced. During the subacute and acute phases, fluids and electrolytes should be administered to prevent or correct dehydration. In addition, transfusions should be given as needed.

Treatment areas requiring research and development
Treatment of Radiation-Induced Diarrhea. A number of factors may contribute to radiation-induced diarrhea. Diarrhea associated with the prodromal and subacute phases of gastrointestinal injury is most likely related to neurohumoral factors affecting gastrointestinal motility and transport. Loss of the epithelial cell lining is not observed until later during the acute phase of gastrointestinal injury. As a result, treatment for post-irradiation diarrhea will require several different approaches.

For the early prodromal and subacute phases of diarrhea, agents directed against or counteracting the effects of neurohumoral factors on gastrointestinal cells should be considered. These include antidiarrheal/antisecretory agents such as anticholinergics, metamucil, amphogel, and loperamide. Loperamide may offer distinct advantages as this agent affects both intestinal cell transport and motility, each of which may contribute to diarrhea. Antisecretory agents, however, will be of limited effectiveness against the acute phase of gastrointestinal injury, during which the loss of epithelial cell lining has progressed to denudation of the intestine.

Presently, our level of understanding concerning the processes involved in stimulating proliferation and/or maintaining the intestinal cell lining following radiation exposure prevents the recommendation of specific therapies for the acute phase of gastrointestinal injury. Cytokines that have proven efficacy in promoting recovery of hematopoietic tissues may be beneficial to the intestine as well. However, sufficient data concerning the efficacy of these agents or gut-related growth factors and elemental diets in stimulating gastrointestinal regeneration are not yet available. Promising results have been provided by Dr. Wilmore and his associates, who have observed the benefits of glutamine administration (0.5 g/kg/day, swallowed if possible; complemented with i.v. administration of the remainder of the dose if necessary). However, confirmation of these results by multiple health care centers is required before specific recommendations can be made.

Decontamination and Antibiotics. The use of antibiotics should be considered for specific infections. Gut decontamination has been recommended following hematopoietic injury. However, the potential adverse effects of gut decontamination also require consideration. Because of the complex relationship between the gut and the immune system, gut decontamination may eliminate the potentially beneficial effects derived from the normal flora in stimulating immune defenses and preparing the immune system for subsequent infections. In the future, the capability to maintain the intestinal integrity following radiation exposure may reduce the present emphasis on gut decontamination.

The bactericidal effect of gastric acid on intestinal flora is well known. However, gastric acid also stimulates pancreatic and biliary secretions, both of which have adverse effects on gastrointestinal integrity post-irradiation. Thus, the need to maintain gut integrity may preempt the desire to stimulate normal bactericidal mechanisms by increasing gastric acid secretion. Indeed, suppression of gastric acid secretion may be preferred. This can be accomplished by the administration of Histamine2 antagonists (e.g., cimetidine) or H+,K+ ATPase inhibitors (e.g., omeprazole). However, the efficacy of these agents in irradiated patients is unproven.

Research Requirements. Considerable study of the gastrointestinal effects of radiation exposure must be undertaken before the development of treatments directed against these effects. The specific etiology of radiation-induced diarrhea needs to be determined. The interaction of the gastrointestinal and immune systems requires clarification. The role of growth factors and diet in the maintenance of normal gut integrity must also be determined. Only then can candidate agents directed against various phases of radiation-induced gastrointestinal injury be selected for additional study. Each of the factors involved in gastrointestinal injury must be analyzed separately to identify relevant factors and to determine useful approaches. This is not possible in patients in whom a variety of treatments are currently administered in an effort to promote survival. Thus, the formulation of treatments against radiation-induced gastrointestinal injury requires additional experiments in animal models in which various parameters of the experiment can be controlled and the influence of individual factors can be discerned.

Cutaneous radiation syndrome: diagnosis and treatment

Considerations
The cutaneous radiation syndrome is a complex pathological syndrome that follows a typical clinical course and is characterized by signs and symptoms that appear relatively early after exposure. The intensity and duration of the signs and symptoms are dose dependent and radiation quality related. The following phases ensue after a brief exposure to highly penetrating radiation (25-30 Gy). The prodromal phase is characterized by a topical and brief erythema. The time of onset following exposure and the magnitude of associated symptoms are useful in estimating the seriousness of the exposure and, therefore, have diagnostic and prognostic value.

In the acute phase, after a variable subclinical or latent period, a secondary erythema develops, followed by swelling caused by an increased vascular permeability and loss of fluids to the extravascular compartment. Moist desquamation is inaugurated by the rupture of vesicles or bullae and the debridement of dead skin; a superficial ulcer develops. Deep ulceration and necrosis are dose and radiation quality dependent. These phenomena result from an ischemic process swing that, due to the hyperplasia of endothelial cells, occludes the capillaries underlying the lesion. Additional clinical manifestations that may be present during this phase include epilation, skin dryness (destruction of sebaceous and sweat glands), telangiectasia, and pigmentary changes.

Beta lesions usually do not affect the entire dermis and underlying structures while gamma-induced lesions involve skin, subcutaneous tissue, muscle and, occasionally, bony structures.

The management of these lesions includes the aspiration of fluids, the debridement of devitalized tissues, the application of bacteriostatic agents coated in nonadherent dressings, and the use of nonsteroidal antiinflammatory drugs to relieve pain, reduce edema, and delay blistering for limited periods. Such agents should not impede healing or adversely affect the patient's overall condition. The fluid obtained from the vesicles should be analyzed for the presence of microorganisms as well as for the presence of cytokines or vasoactive substances.

The extent of these injuries is established by noninvasive procedures, such as magnetic resonance imaging, technetium scanning (angioscintillography), thermography, and ultrasound. For injuries affecting muscles, biopsies (histo and immunocytochemical studies) might be performed to detect inviable tissues and areas of poor blood supply.

Therapeutic procedures
The following options were discussed:

• Distal Extremity Injuries. Conservative treatment for superficial lesions and surgery (ulcerectomy, necrectomy, and amputation) for painful deep ulcerations and necrosis
• Truncal Lesions. Conservative treatment whenever low-penetrating radiations are concerned. Lesions caused by highly penetrating radiations should be treated with drugs intended to improve microcirculation (pentoxifylline-trental) whenever adequate perfusion at affected and surrounding tissues exists. In cases of profound ulcerations or necrosis (ischemic process), the lesion should be excised and the wound bed should be covered with a good quality, full-thickness skin graft.
• Delayed Cutaneous Lesions. These lesions are characterized by pigmentary changes, atrophy, increased vulnerability to trauma, thermal and pressure changes, subcutaneous sclerosis, and keratosis. They also are especially susceptible to reopening periodically due to inflammation of the vascular endothelium (vasculitis). The therapeutic approach is initially conservative (avoidance of trauma, rehabilitation, skin hydration). The use of drugs to reduce fibrosis (interferon-gamma, superoxide dismutase) should be considered. Surgery is indicated in cases of deep necrotic ulceration.

Potential areas of research and development
Diagnostic. Studies comparing magnetic resonance imaging with other noninvasive methods (thermography, angioscintillography); studies of the role of fibroblast, endothelial cell, and keratinocyte receptors and that of certain cytokines (IL-1, IL-6, TNF).

Therapeutic Possibilities. Evaluation of the importance of the topical use of steroids and the potential of the new NSAID (low gastric toxicity) in reducing inflammation and pain; evaluation of the use of epidermal and fibroblast growth factors to accelerate the healing process; evaluation of the use of drugs to improve microcirculation and to avoid the formation of microthrombi; and evaluation of the use of interferon-gamma and superoxide dismutase to reduce fibrosis.

Pathophysiology of the Radiation-Induced Cutaneous Syndrome. Explanation of the increased vascular permeability after irradiation; explanation of increased susceptibility of lesions to trauma; studies of the radiosensitivity of different cells, lineages of skin, and adjacent tissues; and explanation of the etiopathogenesis of obliterative endarteritis caused by radiation.

Definitive care of combined injury
Combined injury is defined as a concurrent trauma (mechanical or thermal) and radiation injury occurring during a period of time before recovery from any one of the injuries. Combined injury may afflict variable numbers of casualties. Because radiation injury is not immediately life threatening, initial care should address the associated conventional injuries, for example, thermal burns and wounds. The emergency medical procedures for ventilation, perfusion, and hemorrhage should be provided first, and then casualties should be stabilized. After stabilization, radioisotope decontamination should be performed before emergency surgery, definitive care, and treatment of radiation injuries. Collection of biological samples during the resuscitation stages will supplement the initial data collected during triage.

Ideally, definitive care should immediately follow resuscitation. However, in mass casualty situations with limited resources, the assessment during triage will indicate which individuals would benefit the most from immediate surgery and those who would benefit from observation and standard medical therapy. During the definitive care stage, the assistance of various medical specialists and consultants is desired, especially in managing difficult cases. In mass casualty situations, however, the primary caregiver (any trained or untrained individual) should be capable of providing basic therapies.

There may be many psychological problems accompanying radiation accidents. Professional help may be necessary to treat these problems.

Surgical procedures play a part in both emergency care and definitive care of patients with combined injuries. Recognizing the difficulties associated with the management of soft tissue wounds, serious consideration should be given to alternative ways to close the wound, for example, biological wound coverings and skin grafts. Surgical correction of life-threatening and other major injuries should be carried out as soon as possible (within 36-48 hours); elective procedures should be postponed until late in the convalescent period (45-60 days) following hematopoietic recovery.

Treatment of thermal burns should include early excision of potentially septic tissue and closure of the wounds, preferably by skin grafting. Radiation burns and thermal burns should be treated differently, especially when using surgery, which should be delayed in the case of radiation burns.

References

1. Browne, D., Weiss, J. F., MacVittie, T. J., et al. Eds. Treatment of Radiation Injuries New York: Plenum Press, 1990.
2. Bodey, G. P., Buckley, M., Sathe, Y. S., et al. Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Ann Intern Med 64:328-40, 1971.
3. Schimpff, S., Satterlee, W., Yound, V. M., et al. Empiric therapy with carvenicillin and gentamicin for febrile patients with cancer and granulocytopenia. N Eng J Med 284:1061-65, 1971.
4. Hughes, W. T., Armstrong, D., Bodey, G. P., et al. Guidelines for the use of antimicrobial agents in neutropenic patients with unexplained fever. J Infect Dis 161:381-96, 1990.
5. Shenep, J. L., Empiric antimicrobial treatment in febrile neutropenic cancer patients. Infections in Medicine, April 1992, p 39-47.
6. Miller, L., Anderson, J.R., Anderson, P.N., et al. American Society of Clinical Oncology recommendations for the use of hematopoietic colony-stimulating factors: Evidence-based, clinical practice guideline. J Clin Oncol 12L 2471-2508, 1994.
7. Meisenberg, B. R., Davis, T. A., Melaragno, A. J., et al. A comparison of therapeutic schedules for administering granulocyte colony-stimulating factor to nonhuman primates after high-dose chemotherapy. Blood 79:2267-2272, 1992.
8. Morstyn, G., Campbell, L., Lieschke, G. Treatment of chemotherapy-induced neutropenia by subcutaneously administered granulocyte colony-stimulating factor with optimization of dose and duration of therapy. J Clin Oncol 7:1554-1562, 1989.
9. Farese, A.M., Williams, D. E., Seiler, F. R., et al. Combination protocols of cytokine therapy with interleukin-3 and granulocyte-macrophage colony-stimulating factor in a primate model of radiation-induced marrow aplasia. Blood 82:3012-3018.
10. Schuening F., Storb, R., Goehle, S. Effect of recombinant human granulocyte colony-stimulating factor on hematopoiesis of normal dogs and on hematopoietic recovery after otherwise lethal total body irradiation. Blood 74:1308-1313, 1989.
11. Vadhan-Raj S., Broxmeyer, H. E., Hittelman, W. N., et al. Abrogating chemotherapy-induced myelosuppression by recombinant granulocyte-macrophage colony-stimulating factor in patients with sarcoma: Protection at the progenitor cell level. J Clin Oncol 10:1266-1277, 1992.
12. Neidhart, J. A., Mangalik A., Stidley, C. A., et al. Dosing regimen of granulocyte-macrophage colony-stimulating factor to support dose-intensive chemotherapy. J Clin Oncol 10:1460-1469, 1992.
13. Fukuda, M., Nakano, A., Kinoshita, A., et al. Optimal timing of G-CSF administration in patients receiving chemotherapy for non-small cell lung cancer (NSCLC) (abstr). Proc Am Soc Clin Oncol 12:447, 1993.
14. Crawford, J., Kreisman, H., Garewal, H., et al. A pharmacodynamic investigation of recombinant human granulocyte colony stimulating factor (r-metHuG-CSF) schedule variation in patients with small cell lung cancer (SCLC) given CAE chemotherapy (abstr). Proc Am Soc Clin Oncol 11:299, 1992.
15. Denzlinger, D., Tetzloff, W., Gerhartz, H. H., et al. Differential activation of the endogenous leukotriene biosynthesis by two different preparations of granulocyte-macrophage colony-stimulating factor in healthy volunteers. Blood 81:2007-2013, 1993.
16. Dorr, R. T. Clinical properties of yeast-derived versus Escherichia coli-derived granulocyte-macrophage colony-stimulating factor. Clin Ther 15:19-29, 1993.
17. Radiological Accident in Goiania, IAEA, STIPUB 815 Vienna, 1988, Unipub 4611-F, Assembly Dr., Lanham, MD.
18. Baranov, A.E., this volume.
19. Radiological Accident in Soreq, IAEA, STIPUB 925 Vienna, 1988, Unipub 4611-F, Assembly Dr. Lanham, MD.
20. Nagler, A., Naparstek, E., Drakos, P., et al., this volume.
21. Prins J. M. , van Agtmael, M. A., Kuijper, E. J., et al. Antibiotic-induced endotoxin release in patients with gram-negative urosepsis: A double blind study comparing Imipenem and Ceftazidime. J Infect Dis 172:886-91, 1995.
22. Schiffer, C. A. Granulocyte transfusions: An overlooked therapeutic modality. Transf Med Rev 4:2, 1990.
23. Strauss, R. G. Granulocyte transfusions. In Principles of Transfusion Medicine. E.C. Rossi, T.L. Simon, G.S. Moss, G.S. Eds. Williams and Wilkins, Baltimore, Maryland, 1991, p. 287.
24. Strauss, R.G. Therapeutic granulocyte Transfusions in 1993. Editorial. Blood 81:1675-78, 1993.
25. Besinger, W. I., et al. The effects of daily recombinant human granulocyte colony-stimulating factor administration on normal granulocyte donors undergoing leukapheresis. Blood 81:1883-88, 1993.
26. Caspar, C. B., Seger, R. A., Bruger, J., et al. Effective stimulation of donors for granulocyte transfusions with recombinant methionyl granulocyte colony stimulating factor. Blood 81:2866-2871, 1993.
27. Verhoef, J. Prevention of infections in the neutropenic patient. Clin Infect Dis 17:Suppl 2, S359-67, 1993.
28. Craven D. E. Use of selective decontamination of the digestive tract. Ann Intern Med 117:609-10, 1992.
29. Schimpff, S. C. Infection prevention in patients with cancer and granulocytopenia. In: Infections in the Abnormal Host M.H. Grieco Ed. New York Medical Books, New York. pp. 926-50, 1980.
30. Beutler, E. Platelet transfusions: The 20,000/uL trigger. Blood 81:1411-13, 1993.
31. MacVittie, T. J., Monroy, R., Vigneulle, R. M., et al. The relative biological effectiveness of mixed fission-neutron gamma radiation on the hematopoietic syndrome in the canine: Effect of therapy on survival. Radiat Res 128:529-536, 1991.
32. Masaoka, T., Moriyama, Y., Kato, S., et al. A randomized placebo-controlled study of KRN8601 (recombinant human granulocyte colony-stimulating factor) in patients receiving allogeneic bone marrow transplantation. Jpn J Med 3:233-239, 1990.
33. DeWitte, T., Gratwohl, A., Van Der Lely, N. Recombinant human granulocyte-macrophage colony stimulating factor accelerates neutrophil and monocyte recovery after allogeneic T-cell-depleted bone marrow transplantation. Blood 79:1359-1365, 1992.
34. Powles, R., Smith C., Milan S., et al. Human recombinant GM-CSF in allogeneic bone-marrow transplantation for leukaemia. Double-blind, placebo-controlled trial. Lancet 336:1417-1420, 1991.
35. Blaise, D., Vernant, J. P., Fiere, D., et al. A randomized, controlled, multicenter trial of recombinant human granulocyte colony stimulating factor (filgrastim) in patients treated by bone marrow transplantation (BMT) with total body irradiation (TBI) for acute lymphoblastic leukemia (ALL) or lymphoblastic lymphoma (LL). Blood 80:982a, 1992 (abstr; suppl 1).
36. Linch, D. C., Scarffe, H., Proctor, S., et al. Randomised vehicle-controlled dose-finding study of glycosylated recombinant
    human granulocyte colony-stimulating factor after marrow transplantation. Bone Marrow Transplant 11:307-311, 1993.
37. Weisdorf, D.J., Vefaillie, C. M., Davies, S. M., et al. Hematopoietic growth factors for graft failure after bone marrow transplantation: A randomized trial of granulocyte-macrophage colony-stimulating factor (GM-CSF) versus sequential GM-CSF plus granulocyte-CSF. Blood 85:3452-3456, 1995.
38. Nemunaitis, J., Singer, J.W., Buckner, D. C., et al. Use of recombinant human granulocyte-macrophage colony stimulating factor in graft failure after bone marrow transplantation. Blood 76:245-253, 1990.
39. Ippoliti, C., Przepiorka, D., Giralt, S., et al. Low dose non-glycosylated rhGM-CSF is effective for treatment of delayed hematopoietic recovery after autologous marrow or peripheral blood stem cell transplantation. Bone Marrow Transplant 11:55-59, 1993.
40. Vose, J. M., Bierman, P. J., Kessinger, A., et al. The use of recombinant human granulocyte-macrophage colony stimulating factor for the treatment of delayed engraftment following high dose therapy and autologous hematopoietic stem cell transplantation for lymphoid malignancies. Bone Marrow Transplant 7:139-143, 1991.
41. Sierra, J., Terol, M. J., Urbano-Ispizua, A., et al. Different response to recombinant human granulocyte-macrophage colony-stimulating factor to primary and secondary graft failure after bone marrow transplantation. Exp. Hematol 22:566, 1994.
42. Brandwein, J. M., Nayar, R., Baker, M. A., et al. GM-CSF therapy for delayed engraftment after autologous bone marrow transplantation. Exp Hematol 19:191-195, 1991.
43. Klingemann, H.G., Eaves, A. C., Barnett, M. J., et al. Recombinant GM-CSF in patients with poor graft function after bone marrow transplantation. Clin Invest Med 13:77-81, 1990.
44. Brugger, W., Henschler, R., Heimfeld, S., et al. Positively selected autologous blood CD34+ cells and unseparated peripheral
    blood progenitor cells mediate identical hematopoietic engraftment after high dose VP16, ifosfamide, carboplatin, and epirubin. Blood 84:1421-26, 1994.
45. Weaver, C.H., Bucknmer, C.D., Login, K., et al. Syngeneic transplantation with peripheral blood mononuclear cells collected after the administration of recombinant human granulocyte colony-stimulating factor. Blood 82:1981, 1993.
46. Janssen, W.E., Smilee, R.C., Perkins, J., et al. Platelet and neutrophil recovery following autotransplant with G-CSF primed marrow or G-CSF mobilized blood is not different. Exp Hematol 23:863, 422a, 1995.
47. Oblon, D.J., Paul, S., Oblon, M.B., et al. Allogeneic peripheral blood stem cell transplants: Preliminary results. Blood 84:94a, 1994
48. Laporte, J.P., Lesage, S., Douay, L., et al. Allogeneic transplantation with peripheral blood hematopoietic progenitors mobilized with G-CSF. Blood 84:90a, 1994.
49. Russell, J., Bowen, T., Brown, C., et al. Allogeneic blood cell transplants for high-risk hematologic malignancy- A comparison of engraftment and acute GVHD with BMT. Blood 84:337a, 1994.
50. Bensinger, W. I., Weaver, C.H., Appelbaum, F. R., et al. Transplantation of allogeneic peripheral blood stem cells mobilized by recombinant human granulocyte colony stimulating factor. Blood 85:1655-58, 1995.
51. Korbling, M., Prezepiozka, D., Huh, Y.O., et al. Allogeneic blood stem cell transplantation for refractory leukemia and lymphoma: Potential advantage of blood over marrow allografts. Blood 85:1659-65, 1995.
52. Schmitz, N., Dreger, P., Suttorp, M., et al. Primary transplantation of allogeneic peripheral blood progenitor cells mobilized by Filgrastim (granulocyte colony stimulating factor). Blood 85:1666-72, 1995.
53. Dreger, P., Viehmann, K., Steinmann, J., et al. G-CSF-mobilized peripheral blood progenitor cells for allogeneic transplantation: Comparison of T-Cell depletion strategies using different CD34+ selection systems campath-1. Exp Hematol 23:147-54, 1995.
54. Cottler-Fox, M., Cipolone, K., Yu, M., et al. Positive selection of CD34+ hematopoietic cells using an immunoaffinity column results in T-cell depletion equivalent to elutriation. Exp Hematol 23:230, 1995.
55. Arseniev, L., Tischler, H. J., Battmer, K., et al. Treatment of poor marrow graft function with allogeneic CD34+ cells immunoselected from G-CSF-mobilized peripheral blood progenitor cells of the marrow donor. Bone Marrow Transplant 14:791-797, 1994.
56. Molena, L., Chabannon, C., Virot, F., et al. Granulocyte colony stimulating factor-mobilized allogeneic peripheral blood stem cells for rescue of graft failure after allogeneic bone marrow transplantation in two patients with acute myeloblastic leukemia in first complete remission. Blood 85: 1678-79, 1995 (correspondence).
57. Dreger, P., Suttorp, M., Haferlach, T., et al. Allogeneic granulocyte colony-stimulating factor-mobilized peripheral blood progenitor cells for treatment of engraftment failure after bone marrow transplantation. Blood 81:1404, 1993 (letter).
58. Nemunaitis, J., Albo, V., Zeigler, Z.R., et al. Reduction of allogeneic transplant morbidity by combining peripheral blood and bone marrow progenitor cells. Leuk, Lymphoma 10:405-406, 1993.
59. Link, H., Arseniev, L., Bähre O., et al. Combined transplantation of allogeneic bone marror and CD34+ blood cells. Blood 86: 2500-2508.