Intra-operative anaesthetic management of older patients undergoing liver surgery

Older patients represent a growing proportion of the general surgical caseload. This includes those undergoing liver resection, with figures rising faster than the rate of population ageing. The physiology of ageing leads to changes in all body systems which may render the provision of safe anaesthesia more challenging than in younger patients. Anaesthesia for liver surgery has specific principles, largely aimed at reducing venous bleeding from the liver, and those related to complex major surgery. This review explores the principles of anaesthesia for liver resection and describes how they may require modification in the older patient. The traditional approach of low central venous pressure anaesthesia in order to reduce bleeding may need to be altered in the presence of a cardiovascular system less able to tolerate hypotension and hypoperfusion. These changes in physiology should also lower the threshold for invasive monitoring. The provision of effective analgesia perioperatively should be tailored to minimise the surgical stress response and opiate use. Careful consideration of general principles of intra-operative care for older patients, such as positioning, drug dosing, avoidance of excessively deep anaesthesia, and maintenance of normothermia are also important given the prolonged, complex nature of liver surgery. This individualised approach, with careful attention to changes in physiology allows liver resections to be undertaken in older patients without increases in mortality.

Multiple large infected scrotal sebaceous cysts masking Fournier's gangrene in a 32-year-old man

Extensive large sebaceous cysts on the scotum are rare and present a problem only when infected or when cosmesis is deemed unacceptable by the patient. Fournier's gangrene is an infective condition with a high death rate. We describe a case of Fournier's gangrene in a patient masked by multiple large infected scrotal sebaceous cysts. A 32-year-old man with a history of alcohol dependency, cirrhosis and multiple scrotal sebaceous cysts presented with acute scrotal pain and erythema. Necrosis of the area became evident within 12 h of his admission and an emergency surgical debridement was performed. The wound was left open to heal via secondary intention over 4 weeks without complication. Fournier's gangrene is a rapidly progressive condition and early surgical debridement is crucial to achieve satisfactory outcomes. In this case, prompt intervention allowed a large scrotal defect to heal without the need for skin grafting.

Comparison of internal emitter radiobiology in animals and humans

Investigations of radionuclide metabolism and effects in various mammalian species revealed important similarities between animals and humans and between some animal species. These include skeletal deposition of radium and radiostrontium in bone volume; deposition on bone surfaces of plutonium and other actinides; liver deposition of actinides; induction of skeletal or liver malignancies by these radionuclides; induction of tooth and jaw abnormalities; mammary cancer induction by radium in humans and in the beagle; depression of circulating cells in blood; and induction of bone fractures. There are also inter-species differences that may not have been noted if multiple species (including humans) had not been studied. Some of these are more rapid excretion of radium in humans compared with most other mammals; induction by radium of eye melanomas in animals but not humans; rapid loss of deposited plutonium from liver in many species of mice and rats but not in humans and dog; substantial sex-related differences in skeletal plutonium retention and bone sarcoma induction in mice but not in humans or dog; and induction of head sinus carcinomas by 226Ra in humans but not the beagle. Leukemia and other related neoplasms were not induced in radionuclide-injected lifespan dogs in excess of the occurrence in control animals. Much of our current understanding of skeletal biology and radionuclide behavior in mammals was derived from this and related projects. The primary goal of the Utah experiment of estimating toxicities of bone-seeking radionuclides relative to radium has been accomplished. For 226Ra = 1.0, comparative toxicities (ratios) of a single injection for bone tumor induction in beagles were about 16 +/- 5 for monomeric 239Pu (32 +/- 10 for chronic exposure), 6 +/- 0.8 for 241Am, 8.5 +/- 2.3 for 228Th, 6 +/- 3 for 249Cf, 4 +/- 2 for 252Cf, 6 +/- 2 for 224Ra (16 +/- 5 for 50 weekly injections), 2 +/- 0.5 for 228Ra, and between 0.01 +/- 0.01 and 1.0 +/- 0.5 for 90Sr, depending on the dose-rate, with the lowest dose-rates approaching a ratio of zero. Corresponding ratios in mice for 226Ra = 1.0 were 16 +/- 4 for monomeric 239Pu, 5.4 +/- 2.0 for 224Ra (16 for 50 weekly injections), 4.9 +/- 1.4 for 241Am, 5.0 +/- 1.4 for 249Cf, 2.6 +/- 0.8 for 252Cf, 4.4 +/- 1.8 for 243,244Cm and about 1.0 for 90Sr at high doses, decreasing to near zero for low doses.

Soft tissue tumors induced by monomeric 239Pu

Individual records of soft tissue tumor occurrence (lifetime incidence) among 236 beagles injected with 239Pu citrate as young adults and 131 comparable control beagles given no radioactivity enabled us to analyze the possible effects on soft tissue tumor induction resulting from internal exposure to 239Pu. A significant trend was identified in the proportion of animals having malignant liver tumors with increasing radiation dose from 239Pu. There was also a significant difference in the relative numbers of both malignant liver tumors (3.2 expected, 22 observed) and benign liver tumors (18.1 expected, 66 observed). Malignant tumors of the mouth, pancreas, and skin were more frequent among controls than among the dogs given 239Pu as were all tumors (malignant plus benign) of the mouth, pancreas, testis, and vagina. For all other tumor sites or types, there was no significant difference for both malignant and all (malignant plus benign) tumors. Mammary tumor occurrence appeared not to be associated with 239Pu incorporation. We conclude that the only soft-tissue neoplasia induced by the intake of 239Pu directly into blood is probably a liver tumor.

Mammary tumor occurrence in beagles given 239Pu

Comparison of 120 young adult female beagles given 0.026 to 106 kBq 239Pu kg-1 by intravenous injection and 63 comparable female control beagles showed that there were no significant differences in the risk of mammary tumor appearance between the two groups. This was the case for benign tumors only, for malignant tumors only, and for both malignant and benign tumors considered together. For malignant tumors the observed number was 73 as compared with 69 expected; for benign tumors, there were 131 observed and 126 expected; for all tumors (separate analysis, not just the addition of malignant plus benign), there were 199 observed and 199 expected. Chi-square analysis indicated that the p values for all these comparisons were > 0.05. There were 45 controls (71.4%) with any tumor vs. 67 dogs (55.8%) given Pu (95% C. I. = 46.9% to 86.2%). No significant differences could be established (Kaplan-Meier analysis) between these two groups for survival age at diagnosis of the first mammary tumor, 11.75 +/- 0.30 y for dogs given Pu vs. 11.90 +/- 0.36 y for controls. We reported previously that differences in mammary cancer occurrence had been identified between this same group of control dogs and 57 female beagles given 226Ra as young adults. The present study appears to support the earlier conclusion that something other than alpha irradiation of the skeleton (both 226Ra and 239Pu deposit in bone) seems to affect the appearance of mammary cancers, since internally deposited 226Ra does appear to induce these malignancies, possibly from initial deposition in mammary tissue of the parent radionuclide or the subsequent concentration in sensitive tissue of its radioactive progeny, 22Rn or isotopes of polonium, lead, and bismuth, which are absent in the case of 239Pu.

Fission fragment relative biological effectiveness for liver tumor induction

The risk coefficients for liver tumors as derived from 60 beagles with body burdens of 249Cf or 252Cf were used to determine the relative biological effectiveness of fission fragments relative to alpha particles. For liver malignancies the relative biological effectiveness was calculated to be about 2 +/- 3. The estimate based on the combined benign and malignant liver tumors was about 7, with a proportionately larger standard deviation of about +/- 14. Although the confidence intervals were wide, it is possible that the relative biological effectiveness of fission fragments relative to alpha particles, using liver neoplasia as the endpoint, is greater than 1.0 as compared to a value only slightly above zero when bone cancer was the lesion of interest.

Soft tissue tumors among beagles injected with 90Sr, 228Ra, OR 228Th

The occurrence of soft-tissue tumors in beagles given 90Sr (88 dogs), 228Ra (76 dogs), or 228Th (81 dogs) as young adults and followed throughout their lifespans was compared with that of 133 control beagles given no radioactivity. For animals injected with 228Ra, tumors of the eye were more prominent (p < 0.05) than in the controls, and soft-tissue tumors of cavities in the head (excluding the brain, mouth, and eye) were more prominent in dogs given 90Sr than in the controls (p < 0.05). There was some indication that eye tumors in animals given about 0.56 kBq 228Th kg-1 were associated with their radionuclide exposure. For tumors at a few other locations, the relative occurrence was greater (p < 0.05) in the controls. These included malignant tumors of the testis and malignant plus benign tumors of the mammae and vagina in 228Th dogs; both malignant and malignant plus benign tumors of the mouth and testis, and malignant plus benign tumors of the mammae and vagina in 228Ra dogs; and malignant plus benign tumors of the mammae in 90Sr dogs (p > 0.05 by Odds Ratio Chi Square analysis but p < 0.05 by Fisher's Exact Test). Differences in relative occurrence between radioactive dogs and controls of all other tumor types that appeared in any of the animals (notably lymphosarcoma, lymph node tumors, leukemia, mast cell tumors, liver tumors, etc.) were not statistically significant (p > 0.05). Intercurrent mortality, mainly from bone cancer, was higher in the radioactive dogs than in the controls. Mean survival was reduced in the dogs given 90Sr, 228Ra, or 228Th (13.17 +/- 2.64 y in controls, 10.95 +/- 4.06 y in 90Sr dogs, 9.07 +/- 3.61 y in 228Ra dogs, and 9.20 +/- 4.15 y in 228Th dogs). Attenuated lifespans could account, at least in part, for the relative paucity of soft-tissue tumors not induced by radiation among the groups of dogs given radioactivity and occurring near the end of life for control animals.

Soft tissue tumors in beagles injected with 241Am citrate

The occurrence of soft tissue tumors has been studied in 117 beagles assigned to 8 dosage groups of between 2 and 26 animals each and injected with 0.07 to 104 kBq 241Am kg-1 as the citrate. In addition, 133 control beagles given no radioactivity were used as a comparison group. All 250 dogs were maintained under identical conditions and were observed for their entire lifespans. An important competing risk for the appearance of soft tissue tumors appeared to be the occurrence of skeletal malignancy, and at the highest injected activity (104 kBq kg-1), kidney and liver failure brought about the death of both of the two dogs in this group. Thyroid and liver were the only soft tissues that exhibited greater concentrations of 241Am than the skeleton. Liver tumors were associated with 241Am exposure (p < 0.001), but the thyroid tumor rate was not increased significantly in the irradiated animals (p > 0.10) as compared with the occurrence in controls. There was a greater relative occurrence of all vaginal tumors in control animals than in dogs given 241Am, a situation also found for all tumors of the pancreas, skin, testis, and mammary glands and for malignant ovarian tumors. All of these differences were statistically significant. The survival of animals given 0.07 to 0.59 kBq 241Am kg-1 could not be established (p > 0.10) as significantly different from controls, but the survival of all groups given 1.8 to 104 kBq kg-1 was decreased (p < 0.05). There was no indication in our studies of a positive association between relative exposure to 241Am and the occurrence of mammary tumors, mast cell sarcomas originating outside the liver, lymphosarcoma or tumors of marrow, including leukemia.

Relative effectiveness of 239Pu and some other internal emitters for bone cancer induction in beagles

The toxicity ratio (relative effectiveness per gray of average skeletal dose) has been estimated for bone cancer induction in beagles injected as young adults for a number of bone-seeking internal emitters. These experiments yielded calculated toxicity ratios (+/- SD) relative to 226Ra = 1.0 of 239Pu = 16 +/- 5 (single exposure to monomeric Pu) and 32 +/- 10 (continuous exposure from an extraskeletal deposit in the body), 224Ra = 16 +/- 5 (chronic exposure) and approximately 6 +/- 2 (single exposure), 228Th = 8.5 +/- 2.3, 241Am = 6 +/- 0.8, 228Ra = 2.0 +/- 0.5, 249Cf = 6 +/- 3, 252Cf = 4 +/- 2, 90Sr = 1.0 +/- 0.5 (for high doses) and 0.05 +/- 0.03 (for low doses) and 0.01 +/- 0.01 (for extremely low doses). Because no skeletal malignancies were observed among beagles given only 253Es, the toxicity ratio is undefined.

Occurrence of mammary tumors in beagles given radium-226

A total of 128 primary mammary tumors (66 of them malignant) occurred in 35 female beagles injected with 226Ra at eight dose levels ranging from 0.2 to 440 kBq/kg body mass as young adults, while a total of 156 mammary tumors (57 of them malignant) were seen in 46 female control beagles not given any radioactivity. Sixty-three of 65 control dogs and 59 of 61 dogs given 226Ra survived the minimum age for diagnosis of mammary tumors of 3.75 years. Based on the observed age-dependent tumor incidence rates in the controls and on the corresponding number of dog-years at risk, the total number of observed malignant tumors in the radium group was statistically greater than the number of expected malignant tumors (66 observed vs 34 expected, P < 0.005). There was no such difference for the benign tumors. Cox regression analysis indicated no increased risk for the first tumor occurrence in irradiated dogs. Cox regression analysis of the multivariate risk sets showed no significantly increased risk for the occurrence of benign tumors but a statistically higher risk of 1.66 with a confidence interval of 1.15-2.40 for the occurrence of malignant tumors. The increased risk was dependent on dose, but a dependence on the frequency of previous occurrence of mammary tumors could not be confirmed. Censoring ovariectomized dogs at time of surgery decreased the relative risks slightly but did not alter the significance. Exposure to diagnostic X rays with cumulative exposures below 0.2 Gy had no effect on tumor formation. It is unknown whether the increased risk for malignant mammary tumors was due to some initial deposition of radium in sensitive tissue, a possible irradiation of fatty mammary tissue from transient radon-->polonium deposition, or a general effect of the overall radium deposition on the immune system of the dogs that lowered their resistance to formation of mammary tumors. Results of this study are potentially useful in understanding risks of radium-induced breast cancers in humans.

Distribution of skeletal malignancies in beagles injected with 239Pu citrate

The distribution of skeletal malignancies among our beagles injected with 239Pu as young adults roughly seems to follow the distribution of skeletal mass and skeletal 239Pu. These findings are similar to those we reported previously for a group of dogs given 26Ra. Although there were differences in tumor distribution between the animals given 226Ra and those given 239Pu, most of them were not statistically significant; however, the radium dogs seemed to show a greater sensitivity to bone tumor origin in the tibia, while there may have been a tendency among the plutonium dogs toward increased relative sensitivity in the scapula, lumbar vertebrae, sacrum, and ribs. In contrast, the most common site for the formation of naturally-occurring bone malignancy in the dog is the distal radius. Perhaps there were too few tumors and too few dogs to establish statistical significance. A correlation between tumor location and at least two anatomical-physiological factors in the skeleton indicated that these two factors (site-specific bone turnover rate and percent of red marrow at the site, which is correlated with vascularity) may influence the appearance of malignancies both individually and in combination. Except for the femur, there appeared to be no difference between the relative distribution of skeletal malignancies of low-level (30 Bq-2 Bq kg-1 injected) and high-level (3-122 kBq kg-1) dogs. Distribution of bone tumors between the axial and appendicular skeleton was 50% vs. 50% for 239Pu (42 and 42), but it was 39% axial vs. 61% appendicular (22 and 35, respectively) for dogs given 226Ra. This difference was not significant (p > 0.2).

Eye tumors and other lesions among beagles given 90Sr or 226Ra

Analysis of eye tumors and other eye lesions among beagles given either 90Sr or 226Ra, and among control animals, indicated that intraocular tumors in excess of the rate for our control animals were not associated with radiation from incorporated 90Sr + 90Y. It is unequivocal that eye melanomas were produced by injected 226Ra. Intraocular neoplasia, hyperplasia, hyperpigmentation, and melanosis in the eye all occurred in our control beagles given no radioactivity; however, tumor experience as currently reported for different beagle colonies may not be directly comparable because of differing rates of discovery, nonuniform nomenclature, and varying criteria for classification of lesions with their discordant interpretation by different pathologists.

Occurrence of metastases in beagles with skeletal malignancies induced by internal irradiation

Metastases from malignant bone tumors often are responsible for the fatal effects of these cancers. Characteristics of primary skeletal malignancies in beagles injected with bone-seeking radionuclides were studied by Thurman (1971) and summarized by Thurman et al. (1971). There were 212 tumors in 186 of these dogs for which we subsequently received information on bone tumor metastases. Evaluation of bone and soft tissue slides from these animals allowed us to compare parameters reported previously with the occurrence of grossly apparent bone tumor metastases. Data included growth-rate of the primary tumor, volume of the primary tumor at death, sex of the animal, growth period of the primary tumor, degree of calcification of the primary tumor, skeletal location of the primary tumor, cumulative radiation dose to the skeleton, dose equivalent to the skeleton, and year of death. For most of the comparisons, no significant differences could be established between dogs with and without metastases. However, tumor volume at death appeared to be correlated with probability of metastasis (p < 0.05), with the larger tumors being associated with higher rates of metastasis. Comparisons of dogs with and without metastases as a function of tumor growth-rate did not, for the most part, yield significantly different results between groups. Exceptions were when only one tumor per dog was considered for animals with multiple primary tumors and when only the tumor with the longest doubling time was included for all dogs with multiple primary tumors (p < 0.02). This effect was a result of only two tumors with doubling times > 45 d. Both had been characterized by Thurman (1971) as among the tumors with the least uncertainty in calculated doubling times. Rates of metastasis in dogs with primary tumors in paired bones were significantly higher than corresponding values of dogs with primary tumors in unpaired bones. Metastases in dogs with primary tumors in the ribs appeared to be more pronounced and those in the thoracic vertebrae appeared to be less pronounced than for animals with primary tumors in other bones as compared with the average for the whole skeleton.

Soft-tissue tumors among beagles injected with 226Ra

A total of 409 primary soft-tissue tumors (189 malignant) occurred among 87 of 120 young adult beagles (72.5%) injected with 226Ra in eight dose levels ranging from 0.2-440 kBq kg-1 body mass, while a total of 565 primary soft-tissue tumors (208 of them malignant) were seen among 117 of 133 control beagles not given radioactivity (88%). Because the p-value for the difference in these two percentages was > 0.05, further comparisons were not made of all tumor locations or types taken together but only of the individual tumor locations or types. There was a clear excess of malignant tumors and all tumors (benign plus malignant) in the eye among dogs injected with radium (p < 0.05, p < 0.01, respectively), but the occurrence of all the other types of soft-tissue tumors was not greater in irradiated vs. control dogs (p > 0.05). This was also true for hematopoietic tumor types (including just one leukemia in a control and none in irradiated dogs) in which there was no difference between controls and dogs given radium. The following total tumors (benign plus malignant) occurred in control dogs but not in radium dogs: brain = 3, peritoneum = 1, and pituitary = 4. Malignant tumors other than leukemia appearing in control animals and not among radium dogs were brain = 2, lymph nodes = 1, adrenal = 3, uterus = 1, and pancreas = 5. Tumors that occurred in dogs given radium and not in controls were 3 mast cell sarcomas and 2 tumors of the thymus (1 malignant). Age at first tumor diagnosis for corresponding tumor types did not seem to differ (p > 0.10 or p > 0.05) between radium dogs and controls except for the eye (p < 0.05), with radium dogs being somewhat younger than controls at first diagnosis, at death, or at loss from the colony. Cox regression indicated differences between radium dogs and controls in risk of dying with specific tumors. The following tumors had p values of < 0.05 and risk ratios of > 2.2:eye, mouth (mostly melanomas), and thyroid for malignant tumors and for malignant and benign tumors together. When all sarcomas were considered as a group, there was no difference between controls and radium dogs but there was a difference for all carcinomas taken together, even when mammary tumors and eye tumors were excluded and when eye tumors alone were excluded.

Skeletal malignancies among beagles injected with 241Am

Seventy skeletal malignancies in 44 dogs were identified among 117 beagles injected as young adults with graded dosages of approximately 0.07 to 104 kBq 241Am kg-1 and maintained for lifetime observation. All of these tumors were osteosarcomas except four fibrosarcomas of bone and four chondrosarcomas of bone. Of these 117 animals, 114 survived beyond the minimum age (of 2.79 y) for radiation-induced bone cancer, and all are now dead. An expression was derived that described the dependence of percent occurrence of bone sarcoma on skeletal radiation dose of A = 0.76 + 30D, where A = percent of dogs with skeletal malignancy within any dosage group, D = average skeletal dose (< 3 Gy) at 1 y before death (average skeletal dose was calculated to the presumed start of tumor growth, which we have taken to be 1 y before death), and 0.76 represents the lifetime percent malignant bone tumor response among 132 suitable control dogs in our colony not given any radioactivity. All dosage groups with skeletal doses of > 3 Gy at 1 y before death exhibited close to 100% occurrence and appeared to be beyond the region of linearity. Therefore, they were excluded from the derivation of this expression. Similar analysis of corresponding data for beagles given 226Ra as young adults, excluding the two highest dosage groups in which the bone tumor response was approximately 100%, yielded the expression, A = 0.76 + 4.7D, (D < 20 Gy). A ratio of the coefficients in these two expressions indicates the effectiveness at low radiation doses for bone-cancer induction of 241Am relative to 226Ra, or (30 +/- 2.6)(4.7 +/- 0.47)-1 = 6 +/- 0.8. This compares to the relative effectiveness at low radiation doses that was obtained earlier for a 239Pu:226Ra toxicity ratio of about 16 +/- 5.

Bone cancer occurrence among beagles given 239Pu as young adults

The occurrence of skeletal malignancies has been documented among 234 young adult beagles given single intravenous injections of monomeric 239Pu citrate. Occurrence has also been documented among 132 comparable control group animals surviving the minimum latent time period of 2.79 y for radiation-induced bone cancer, who were maintained for lifespan observation. Injected amounts ranged from about 0.02-106 kBq kg-1 body mass with factors of 2 or 3 between dose levels. There were 84 radiographically apparent bone tumors in 76 plutonium-injected dogs and one tumor in a control group dog. Most of these were osteosarcomas except for seven chondrosarcomas, one liposarcoma, and one plasma cell myeloma of bone. The relationship between percent of dogs at any dose level with bone malignancy and average skeletal dose at the presumed time of tumor initiation of 1 y before death appeared to be linear below about 1.3 Gy average skeletal dose. The observed data can be approximated by the expression A = 0.76 + 75 D, where A = percent of dogs with bone cancer at any dose level, D = average skeletal dose in Gy (for doses up to 1.3 Gy) at tumor initiation, and 0.76 represents the percent tumor response in the control animals not given plutonium. Similar analysis of our corresponding data for beagles given 226Ra, excluding the two highest dose levels (approximately 100% occurrence), yielded the expression A = 0.76 + 4.7 D, where D = the average skeletal dose in Gy (for doses up to 20 Gy) at 1 y before death. The ratio of coefficients indicates the effectiveness for bone cancer induction of 239Pu relative to 226Ra, or [(75 +/- 22.5)(4.7 +/- 0.47)-1] = 16 +/- 5 for a single, brief intake of either nuclide into blood.

Fission fragment RBE for bone sarcoma induction

Thirty beagles and 277 mice were injected with 249Cf, and 30 beagles and 274 mice were injected with 252Cf. The skeletal dose (in Gy) from 252Cf was about half from fission fragments and half from alpha particles, whereas 249Cf emits alpha particles in 100% of its transformations. Bone sarcomas (mostly osteosarcomas) were the main radiation-induced cancer. The relative biological effectiveness (RBE) of fission fragment dose relative to alpha-particle dose for bone sarcoma induction was calculated from the ratio of 249Cf/252Cf doses at equal times to bone sarcoma in (a) beagles and (b) mice, and (c) from the ratio 252Cf/249Cf risk coefficients in mice. The average RBE +/- standard deviation of the three evaluations was 0.1 +/- 0.1. The very low RBE for bone sarcomas is supported by the data of A. L. Batchelor, T. J. Jenner, and L. M. Cobb [Phys. Med. Biol. 28, 475-483 (1983)] for lung cancer induction in rats and by that of A. L. Brooks, J. A. Mewhinney, and R. O. McClellan [Health Phys. 22, 701-706 (1972)] for producing chromosome aberrations in the liver cells of Chinese hamsters. The low effectiveness of fission fragments relative to alpha particles, per gray of absorbed dose, is ascribed primarily to the much larger number of cells traversed by the alpha particles. Consideration might be given to decreasing the quality factor of fission fragments by an order of magnitude below that for alpha particles.

Primary pulmonic tumors in beagles

The age-specific frequency of 11 primary pulmonic neoplasms in a closed Beagle colony is presented. The first tumor occurred at 5 years, and the frequency increased progressively in the older age classes. All of the tumors arose from sites distal to the principal bronchi. Although metastases appeared to occur relatively late, with respect to course of tumor development, they were noted in approximately a half of the dogs and most frequently involved the bronchial lymph nodes.

Mammary neoplasia in a closed beagle colony

The incidence rate of mammary neoplasia in a large colony of beagles and the relationship to internal skeletal and/or liver radiation, age, relatively late ovariectomy (4 years and older), endometritis, parity status, and adrenal weight were examined. Of these various factors, age was the only condition that was clearly correlated with changes in the mammary tumor incidence. The rate became significant at approximately eight years of age and increased progressively throughout the older age classes. Among the female dogs, the incidence of mammary cancer was higher than that of any other form of spontaneous malignancy.