Osteopenic consequences of botulinum toxin injections in the masticatory muscles: a pilot study.

Patients with temporomandibular muscle and joint disorder (TMJD) increasingly seek and receive treatment for their pain with botulinum toxin (BoNTA; botulinum toxin A). Used intramuscularly in therapeutic doses, it produces localised paresis. Such paresis creates risk of reduced bone mineral density, or 'disuse osteopenia'. Animal studies have frequently used BoNTA as a model of paralysis to induce bone changes within short periods. Osteopenic effects can be enduring in animals but have yet to be studied in humans. This is the first study in humans to examine bone-related consequences of BoNTA injections in the masticatory muscles, comparing oral and maxillofacial radiologists' ratings of trabecular bone patterns in the condyles of patients with TMJD exposed to multiple masticatory muscle injection sessions with BoNTA to a sample of patients with TMJD unexposed to masticatory muscle injections with BoNTA. Cone-beam computed tomography (CBCT)-derived images of bilateral condyles were evaluated in seven patients with TMJD receiving 2+ recent BoNTA treatment sessions for facial pain and nine demographically matched patients with TMJD not receiving BoNTA treatment. Two oral and maxillofacial radiologists evaluated CBCT images for evidence of trabecular changes consistent with osteopenia. Both evaluators noted decreased density in all participants exposed to BoNTA and in none of the unexposed participants (P < 0.001). No other abnormalities associated with reduced loading were detected. These findings need replication in a larger sample and over a longer time period, to ensure safety of patients with TMJD receiving multiple BoNTA injections for their pain.

Cysts and cystic lesions of the mandible: clinical and radiologic-histopathologic review.

Many lesions that occur in the mandible have a cystlike radiographic appearance. These lesions are often difficult to differentiate on the basis of their radiographic features alone. Mandibular lesions may be odontogenic or nonodontogenic. Among odontogenic lesions without mineralization, ameloblastomas, odontogenic keratocysts, and dentigerous cysts can all appear as well-defined, unilocular, well-corticated, lucent lesions that are often associated with the crowns of impacted or unerupted teeth. Most radicular cysts appear as round or pear-shaped, unilocular, lucent lesions in the periapical region. Among odontogenic lesions with mineralization, complex odontomas contain multiple masses of dental tissue and compound odontomas contain multiple teeth or toothlike structures. Odontogenic myxomas are characterized by lytic osseous changes of varying size, which may be demarcated and expansile or exhibit ill-defined borders. Nonodontogenic lesions that mimic odontogenic lesions include benign fibro-osseous lesions (conventional or juvenile ossifying fibroma, focal or periapical cemento-osseous dysplasia, florid osseous dysplasia), traumatic bone cyst, lingual salivary gland inclusion defect, central giant cell granuloma, brown tumor of hyperparathyroidism, arteriovenous malformation, and mucoepidermoid carcinoma. The clinical and radiographic features of these mandibular lesions help establish a differential diagnosis, although microscopic tissue evaluation is generally necessary to accurately identify the lesion.

Imaging of the temporomandibular joint: a position paper of the American Academy of Oral and Maxillofacial Radiology.

Various imaging techniques for the temporomandibular joint are discussed with respect to uses, strengths, and limitations. An imaging protocol is outlined for evaluating patients with a wide variety of temporomandibular joint related signs and symptoms.

Dosimetry and cost of imaging osseointegrated implants with film-based and computed tomography.

Thermoluminescent dosimeters were used to measure radiation doses at craniofacial sites in a tissue-equivalent phantom during film-based multidirectional tomography with the Tomax Ultrascan (Incubation Industries, Ivyland, Pa.) and during computed tomography with the Elscint Excel 2400 (Elscint Corp., Tel Aviv, Israel). Mean absorbed doses for presurgical mandibular and maxillary canine and molar implant assessments were converted to equivalent doses, which were then multiplied by published weighting factors and summed to give effective doses. The computed tomography device consistently delivered higher doses than the Tomax Ultrascan to all anatomic locations; the differences were most pronounced when only one or two implant sites were evaluated. The reasons for the dose disparities are considered both anatomically and procedurally. A survey of examination cost revealed film-based multidirectional tomography to be less expensive than computed tomography.

Niobium filtration of conventional and high-frequency x-ray generator beams for intraoral radiography. Effects on absorbed doses, image density and contrast, and photon spectra.

We have studied the effects of niobium beam filtration on absorbed doses, on image density and contrast, and on photon spectra with conventional and high-frequency dental x-ray generators. Added niobium reduced entry and superficial absorbed doses in periapical radiography by 9% to 40% with film and digital image receptors, decreased the radiation necessary to produce a given image density on E-speed film and reduced image contrast on D- and E-speed films. As shown by increased half-value layers for aluminum, titanium, and copper and by pulse-height analyses of beam spectra, niobium increased average beam energy by 6% to 19%. Despite the benefits of adding niobium on patient dose reduction and on narrowing the beams' energy spectra, the beam can be overhardened. Adding niobium, therefore, strikes the best balance between radiation dose reduction and beam attenuation, with its risks of increased exposure times, motion blur, and diminished image contrast, when it is used at modest thicknesses (30 microns) and at lower kVp (70) settings.

Coordination of oral cavity and laryngeal movements during swallowing.

In this study, dynamic imaging was used to track the movements of oral cavity and laryngeal structures during swallowing in 10 normal adults subjects. The movements of tiny lead pellet markers attached to the lips, tongue, mandible, and soft palate, as well as anatomic landmarks on the hyoid bone, were measured in relation to a reference pellet affixed to the upper central incisors. Sagittal views of the oral cavity were obtained using standard videofluorography. Each subject produced 10 swallows of 12 ml of tap water followed by 5 swallows with a bite block placed between the molars. The recorded video images were input to a microcomputer where the x- and y-coordinates of the pellets were measured. Results of the analyses revealed considerable temporal overlap in the timing of oral cavity and laryngeal movements, widespread individual variability in coordination patterns and movement trajectories, and selective effects of the bite block. These data suggest the existence of individual adaptive strategies in the programming and control of swallowing movements.

Alteration of irradiated shuttle vector processing by exposure of human lymphoblast host cells to single or split gamma-ray doses.

The repair of damaged DNA by mammalian cells exposed to single or split doses of radiation was probed with shuttle vector pZ189. Human lymphoblast hosts who received a single 120 cGy dose 2 h before transfection with 2500 cGy-damaged pZ189 yielded a two-fold higher frequency of progeny plasmids with mutations in their supF-tRNA target genes than did unirradiated host cells. Delaying transfection for 12 h, however, reduced the mutation frequency by half versus unirradiated controls. Plasmid survival was also affected by the time between host cell irradiation and transfection. Splitting doses of 50-500 cGy into two equal fractions separated by 4 h lowered mutation frequency and increased plasmid survival compared with equivalent acute doses; increasing the interval between dose fractions to 8 h, however, lowered plasmid survival compared with acute doses. Sequence analyses of the target gene in mutant plasmids revealed increased multiple-base substitution mutations among progenies recovered from irradiated hosts, indicating enhanced excision repair. These findings support modulation of mammalian cell DNA repair by ionizing radiation, disclose the transient nature of the effect of radiation on DNA repair, and demonstrate a quantitative difference in the effectiveness of single and split doses.

Mutations caused by gamma-radiation-induced double-strand breaks in a shuttle plasmid replicated in human lymphoblasts.

The mutagenicity of open-circular DNA (containing base damage and single-strand breaks) and linear DNA (containing base damage, single-strand breaks, and one double-strand break) produced in vitro by gamma-irradiation of shuttle vector pZ189, was analysed after the plasmid's repair and replication in the human lymphoblast line, GM606. By comparing the survival, mutation frequency, and types of mutations in descendants from the two DNA forms, the effects of the double-strand break were determined. The percentage of viable plasmids from linear DNA was two-fold lower than that from open-circular DNA, 7.8 versus 14.0 (compared with unirradiated, control DNA). The mutation frequency in progenies of the open-circular plasmid was 4.2 +/- 1.7 x 10(-3), compared with 7.8 +/- 0.1 x 10(-3) in progenies of the linear DNA, again, nearly a two-fold difference. Approximately 59% of the mutations from the linear DNA were deletions and 34% were base substitutions. In contrast, only 13% of mutations from open-circular DNA were deletions, but 87% were base substitutions. All recoverable deletions were small, ranging from 1 to 205 base pairs, and the majority contained direct repeats at the deletion junctions, indicating non-homologous recombinations. Thus, mutations found among descendants from the linear and open-circular DNAs were qualitatively similar but quantitatively different. The data suggests that producing one double-strand break in DNA by ionizing radiation causes a two-fold increase in both lethality and mutation frequency.

Mutation spectrum in gamma-irradiated shuttle vector replicated in ataxia-telangiectasia lymphoblasts.

Cells from ataxia-telangiectasia (AT) patients are hypersensitive to the lethal effects of ionizing radiation. To assess radiation mutagenesis in these cells, the SV40-based shuttle vector, pZ189, was used to analyze gamma-ray-induced mutations following the plasmid's replication in AT lymphoblasts. Progenies from the AT line GM2783 exposed to 50 Gy showed a mutation frequency of 7.6 x 10(-3), 63-fold over background; surviving plasmids were 3.4% of control. Both values were essentially the same as those of irradiated plasmids replicated in a normal lymphoblast line, GM606. In addition, pZ189 exposed to 25 Gy of gamma radiation and replicated in another normal lymphoblast line and in cells of two additional AT lymphoblast lines showed similar mutation frequencies and percentages of surviving plasmids. Qualitative comparison of plasmid mutations from AT and normal cells showed no significant differences, indicating that the damaged DNA was repaired with similar fidelity in AT and normal cells. These studies suggest that there is no correlation between the enhanced sensitivity of AT cells to killing by ionizing radiation and gamma-radiation-induced mutagenesis of plasmid DNA processed in these cells.

Dependence of the mutation spectrum in a shuttle plasmid replicated in human lymphoblasts on dose of gamma radiation.

The frequencies and types of mutations induced in the target gene, supF-tRNA, of the shuttle vector pZ189 were analysed following the replication of the gamma-irradiated plasmid in the human lymphoblastoid cell line, GM606. The mutation frequency measured in progeny of unirradiated pZ189 was 1.02 x 10(-4), increasing to 17.5 x 10(-4) at 1000 cGy, and to 63.4 x 10(-4) at 5000 cGy, approximately 17- and 62-fold over background levels, respectively. Simultaneously, the number of plasmids capable of replicating in Escherichia coli decreased with increasing radiation dose to 4% of the control value at 5000 cGy. Electrophoresis of the irradiated DNA showed a correlation between increases in mutation frequency and decreases in plasmid survival, and the formation of open-circular and linear DNA. The majority of the spontaneous (69.8%) and induced mutations (85.7%) at 1000 and 79.4% at 5000 cGy) were base substitutions and were generally of similar types among all groups. However, changes at 2500 (12.7%) and 5000 cGy (13.2%) involving A:T base pairs were greater than those in unirradiated controls (3.4%) or those at 1000 cGy (2.0%). This increase in A:T base pair mutations could be a result of reduced repair fidelity when the DNA is extensively damaged by high doses of ionizing radiation.

7,12-Dimethylbenz[a]anthracene adduct formation with Syrian hamster cheek pouch epithelial DNA: in vitro studies in organ explant culture.

Studies examined the binding of radiolabeled 7,12-dimethylbenz[a]anthracene (DMBA) to epithelial DNA of hamster cheek pouch (HCP) maintained in organ explant culture. Adduct formation was studied as functions of [3H]DMBA dose, of the time after single [3H]DMBA applications, and of the route by which the DMBA was administered--either topically or in the culture media. Total DMBA-DNA adduct formation [total binding index (TBI)] was determined by DNA-bound 3H activity, and qualitative binding characteristics were further studied by high-pressure liquid chromatography. [3H]DMBA was applied either in the culture media at concentrations of 0.005-0.5 micrograms/ml or topically in mineral oil or ethanol in doses of 0.005-0.5 micrograms to each tissue fragment. Histopathologic changes in DMBA-treated HCP fragments included substantial aberrations in maturation of cornified and keratin layers and focal squamatization and dysplasia of the basal epithelium--considerable tissue necrosis was encountered in the high-DMBA-dose groups. Dose-response data were qualitatively similar among treatment types, with the greatest TBIs in topical ethanol groups and the lowest TBIs in culture medium groups. Kinetics of adduct formation and removal showed a rapid increase in TBIs to peak values at 24-72 hours followed by a biphasic decrease in TBIs, which leveled off at 7%-20% of peak values at 120-240 hours. Chromatographic analyses of selected samples at various times from all treatment groups showed three major peaks that are likely to be the same 1,2,3,4-tetrahydro-3,4-dihydroxy-1, 2-oxide-deoxyribonucleoside adducts observed in other rodent in vivo and cell culture systems. These results are consistent with those of other laboratories studying DMBA-DNA interactions and suggest that in vitro studies of DMBA-treated HCP explants are useful in studying the molecular nature of DMBA-DNA interactions in oral mucosal carcinogenesis.

DNA adduct formation by 7,12-dimethylbenz(a)anthracene in Syrian hamster cheek pouch epithelium in vivo.

Studies examined the in vivo binding of radiolabeled 7,12-dimethylbenz(a)anthracene (DMBA) to hamster cheek pouch epithelial DNA. Adduct formation was studied as functions of [3H]DMBA dose and of the time after single [3H]DMBA applications in mineral oil. Total DMBA-DNA adduct formation was determined by DNA-bound 3H activity, and qualitative binding characteristics were further studied by high-pressure liquid chromatography. Adduct formation 24 h after single [3H]DMBA applications rapidly increased from DMBA concentrations of 0.05-5.0 micrograms. While binding also increased from DMBA concentrations of 5.0-50.0 micrograms, the variability in adduct formation at 50.0 micrograms was considerable. Adduct formation following single 5.0-micrograms [3H]DMBA applications rose slowly to a peak value of 76 pmol DMBA/mg DNA at 36 h. This level decreased very slowly in a biphasic manner through 240 h, at which time the adduct levels were 23% of maximum. Adduct levels of 1.5 pmol/mg DNA were measured as late as 5 wk after a single 5.0-micrograms [3H]DMBA application. Chromatographic analyses of the 24-, 36-, and 96-, and 240-h samples showed three major peaks which are likely to be 1,2,3,4-tetrahydro-3,4-dihydroxy-1,2-oxide-deoxyribonucleoside adducts. While these analyses were limited by the small amounts of radioactivity which could be retrieved from [3H]DMBA-treated pouches, the study suggests that the DMBA-induced hamster cheek pouch carcinoma may be useful in some molecular in vivo studies of chemical-DNA interactions in carcinogenesis.

Low level X-radiation effects on carcinogenesis by 7,12-dimethylbenz(a)anthracene in Syrian hamster cheek pouch epithelium: acute vs fractionated radiation dose studies.

Studies examined the effects of acute and fractionated low to moderate level X-ray exposures on hamster cheek pouch carcinogenesis in vivo by 7,12-dimethylbenz(a)anthracene (DMBA). Animals were grouped by treatment as follows: acute doses of 0.85-3.40 Gy X rays; 17 once weekly doses of 0.01-0.20 Gy X rays (fractionated radiation); topical DMBA for 10 weeks; DMBA plus fractionated radiation starting together; DMBA plus acute radiation in Week 1 or 10 of DMBA treatments; and sham irradiation, DMBA vehicle, or anesthesia controls. After 44 weeks, hamsters were sacrificed, and their cheek pouches were excised, serially sectioned, and examined by light microscopy for histopathology. No histologic changes were observed in radiation-only hamsters. Carcinoma incidences in DMBA-only groups ranged from 45 to 60%. Carcinoma incidences were greater in groups receiving DMBA plus fractionated radiation than in groups receiving either acute radiation + DMBA or DMBA alone. Carcinoma incidences in acute radiation plus DMBA groups were lower than those in DMBA-only groups. These results suggest complex interactions between radiation and DMBA, perhaps with radiogenic cell killing being a principal factor in acute radiation + DMBA groups, and reciprocal additive or synergistic effects of radiation and DMBA on cancer induction and manifestation in fractionated radiation + DMBA groups.

A clinical comparison of X-ray films for detection of proximal surface caries.

This clinical study compares the diagnostic yield of Kodak Ektaspeed with the diagnostic yield of Kodak Ultraspeed using a split-mouth experimental design and proximal surface carious lesions as the test objects. Assuming that proximal surface carious lesions are distributed randomly in a given patient and in the population at large, then any difference in the diagnostic quality of either of the films should be disclosed as a difference in the frequency and distribution of lesions identified with either of the films. No statistical difference was observed, which suggests that the two films perform equally well for the diagnostic task of detecting proximal surface carious lesions. Considering that there is a 40% to 50% reduction in radiation exposure to the patient when the faster film is used, it appears that Kodak Ektaspeed should be the preferred film for the dental practitioner.

Single, split and fractionated dose X-radiation-induced malignant transformation in A31-11 mouse BALB/3T3 cells.

Studies were conducted to determine the effects of a single, split and fractionated doses (separated by 1-h intervals) of 100 rad of X-irradiation on the morphological transformation of BALB/c 3T3 clone A31-11 mouse fibroblasts grown in media containing calf serum. Both spontaneous and radiation-induced transformation levels were lower for these cells grown in the cell-serum containing media than previously reported for these cells grown in fetal calf serum containing media. In the studies reported here, cells were irradiated either as density-inhibited plateau phase cultures or as low density cultures at 10-14 h after being reseeded from confluent dishes. We observed that a 4-fraction 100-rad dose resulted in a reduced yield of transformants compared to a single dose of 100 rad when plateau phase cultures were utilized for the radiation exposures, but not in low density cultures in which the cells were allowed to proliferate during the radiation exposures, these results suggest that the growth phase of the cells can play a major role in determining the yield of transformants induced by fractionated doses of radiation. It is noteworthy that, for the other data obtained in these studies, in none of 12 different experimental points (involving 5 separate experiments) did a fractionated dose protocol result in a reduced yield of transformants when compared to a single dose protocol.

Low-level X-radiation effects on functional vascular changes in Syrian hamster cheek pouch epithelium during hydrocarbon carcinogenesis.

Effects of repeated low-level X radiation on functional microvascular changes in hamster cheek pouch epithelium during and following carcinogenesis by 7,12-dimethylbenz[a]anthracene (DMBA) were studied. Prior studies showed enhancement of such carcinogenesis by repeated 20 rad head and neck X-radiation exposures, and it was proposed that one possible mechanism was radiogenic alteration of the functional microvasculature in a manner which favored subsequent tumor development. Hamsters were treated with either radiation, DMBA, radiation + DMBA, or no treatment. Animals were sacrificed at 3-week intervals from 0 to 39 weeks after treatments began. Pouch vascular volume and permeability changes were studied by fractional distributions of radiotracers and were analyzed by a variety of statistical methods which explored the vascular parameters, treatment types, elapsed time, presence of the carcinogen, and histopathologic changes. All treatments resulted in significant changes in vascular volume with time, while only DMBA treatments alone resulted in significant changes in vascular permeability with time. Prior to the appearances of frank neoplasms, volumetric changes in DMBA only and radiation only groups were similar, while volume changes in DMBA + radiation groups increased slowly to a peak later than in other groups and then declined steadily to levels similar to the radiation only group. As in prior studies, there were significant vascular volume differences between DMBA and DMBA + radiation groups of tumor-bearing cheek pouches. DMBA maxima were significantly higher than those of DMBA + radiation. Radiation significantly affected DMBA-associated vascular volume and permeability changes during carcinogenesis. Several possible explanations for the relationship of these changes to the enhancement of DMBA carcinogenesis include: radiation blocking normal capillary proliferative and/or dilatory responses to inflammation secondary to neoplastic changes; radiation-induced focal increases in the pericapillary connective tissue histohematic barrier, stimulating angiogenesis but reducing nutrient diffusion; radiation exposures sensitizing vascular endothelium to subsequent angiogenic stimulation from premalignant tissues; DMBA vascular and epithelial effects partially or completely blocking radiation effects on epithelial and/or endothelial cells; and radiation damage to vessel walls partially or fully inhibiting normal physiologic mechanisms of repairing DMBA damage to the vessels.

Comparison of biodistribution of 3H, 32P and 99mTc labeled Gpp(NH)p in tumor bearing hamsters.

Biodistribution and tumor uptake studies were carried out with intravenously injected tracer doses of Gpp(NH)p labeled with 3H, 32P or 99mTc . Syrian golden hamsters with cheek pouch carcinomas, induced by repeated topical applications of DMBA, were used as a tumor model. The biodistributions of these three radionuclides were different, indicating significant molecular cleavage of this nucleotide analog. It was also apparent that this compound labeled with 99mTc may not be useful for tumor imaging due to low tumor-to-blood specific activity ratio. The cheek pouch carcinoma tumor model may be valuable for the evaluation of tumor localizing radiopharmaceuticals.

Possible mechanistic roles of anatomical and functional vascular changes in rat urinary bladder carcinogenesis induced by N-butyl-N-(4-hydroxybutyl)nitrosamine.

Anatomical and functional vascular changes during rat urinary bladder carcinogenesis were studied by scanning electron microscopy of vascular casts, transmission electron microscopy of bladder capillaries, and fractional distributions of 51Cr-erythrocytes, 125I-human serum albumin, and 86RbCl which were used to determine vascular volume, permeability, and perfusion. Histopathological changes and focal capillary changes in vascular casts were measured quantitatively by an image analyzer. Male Wistar rats received 0.05% N-butyl-N-(4-hydroxybutyl)nitrosamine (BBN) in their drinking water for 8 weeks and were then maintained on tap water without BBN for an additional 32 weeks. Simple hyperplasia was first seen at Week 2. The percentage of the area of hyperplastic epithelium increased to about 95% by Week 8 and then decreased to 4 to 6% at Weeks 20 or 40. Papillary or nodular hyperplasia was first seen at Week 6. The percentage of the area of papillary or nodular hyperplasia increased with time to 31.0% at Week 40. Papillary transitional-cell carcinomas were found from Week 20, increasing with time, and their incidence was 100% after Week 35. Vascular cast diameters of normal-looking capillaries were larger during than after BBN treatment. Type 3 vascular proliferations were found beneath papillary or nodular hyperplasia and cancer. Capillaries beneath simple hyperplasia and type 3 capillaries beneath capillary or nodular hyperplasia and cancers were fenestrated and dilated. Changes in vascular volume were independent of changes in permeability and perfusion. Best-fit curve analyses showed the maximum vascular volume at 8 weeks and minimum at 25 weeks, and the permeability maxima at 4 and 25 weeks with minima at 15 and 32 weeks. While 86Rb values correlated 125I values (r = 0.58), they were unstable in intermediate time periods. Changes of vascular volume were coincident initially with increased areas of dilated capillaries beneath simple hyperplasia and later with areas of type 3 capillary proliferation beneath papillary or nodular hyperplasia and cancer. Changes of vascular permeability were related to inflammation indices throughout the study. Increases in permeability were coincident with fenestrated capillaries beneath simple hyperplasia in early stages, and subsequently with fenestrated type 3 capillaries beneath papillary or nodular hyperplasia and cancer. BBN appears to cause alterations in vascular volume via induction of capillary dilation and also possibly by enhancing the responsiveness of host endothelium to angiogenic stimulation from neoplastic or preneoplastic tissues.

Anatomical and functional vascular changes in hamster cheek pouch during carcinogenesis induced by 7, 12-dimethylbenz(a)anthracene.

Anatomical and functional vascular changes during hamster cheek pouch carcinogenesis were studied by light microscopy; scanning electron microscopy of vascular casts; transmission electron microscopy of cheek pouch capillaries; and fractional distributions of 51Cr-erythrocytes, 125I-human serum albumin, and 86RbCI which were used to determine vascular volume, permeability, and perfusion. Histopathological changes and focal capillary changes in vascular casts were measured quantitatively by an image analyzer. Male Syrian hamsters received 0.5% 7,12-dimethylbenz[a]anthracene (DMBA) in mineral oil for 11 weeks and were sacrificed at periodic intervals from 2 to 20 weeks after initial treatment. Simple hyperplasia was first seen at Week 1. The area of hyperplastic epithelium, expressed as percentage, increased to about 60% by Week 8 and then decreased to 30% at Week 20. Dysplastic foci were first seen at Week 2. The percentage of the area of dysplasia increased with time to 41% at Week 20. Squamous cell carcinomas occurred from Week 10, increased with time, and were found in all animals at Week 20. Vascular cast diameters of normal-looking capillaries were larger during than after DMBA treatment. Type 3 vascular proliferations were found beneath dysplasia and cancer. Capillaries beneath simple hyperplasia and type 3 capillaries beneath dysplasia and cancers were dilated but not fenestrated. Changes in vascular volume were independent of changes in permeability and perfusion and also occurred in contralateral untreated pouches of treated animals. While 86Rb values initially correlated with 125I values, the 86Rb values were unstable in intermediate and later time periods. Changes of vascular volume were accompanied initially by the presence of DMBA and were coincident with increased areas of dilated capillaries beneath simple hyperplasia and later with areas of type 3 capillary proliferation beneath dysplasia and cancer. Changes of vascular permeability were related to inflammation indices throughout the study. DMBA may lastingly alter capillary endothelium in a manner which allows or aids in its subsequent dilatory and proliferative responses to angiogenic stimulation from malignant tumors, and possibly from premalignant or malignantly transformed cells.