The exposure of salivary glands to high doses of radiation during therapy for head and neck malignancies leads to secretory hypofunction and several complications, including xerostomia, dysphagia, rampant dental caries, oral infections, oesophagitis and gustatory dysfunction. Thus, the oral complications of radiation-induced salivary hypofunction can be particularly distressing for the surviving patient. Not surprisingly, there has been a continuous effort, for more than 80 years, to understand the specific mechanisms that underlie such gland damage.

Most laboratory studies of radiation damage to salivary glands have been on rodents. Frequently, such studies have used a single exposure, and have evaluated gland structure, or occasionally secretory functions, within 1 month. Recently, Funegard and colleagues studied the effects of fractional radiation doses (5×4 Gy to 5×8 Gy) on rat whole saliva production for up to 6 months. Alterations were dose- and time-dependent, but every dosage resulted in reduced salivation by 26 weeks post-radiation. Because whole saliva was collected in these studies, it is unclear to what extent the two major salivary glands (parotid, submandibular) were individually affected. Indeed, it is generally considered that the parotid glands are more radiation-sensitive than the submandibular glands in both rodents and humans.

Previously, we have reported studies of parotid and submandibular secretory function in rats up to 40 days after head and neck radiation. The present study is an extension of those earlier investigations. We have now evaluated individual parotid and submandibular flow rates, glandular weights, and body weights at 3, 6, 9 and 12 months after single doses of either 2.5, 5, 7.5, 10 or 15 Gy X-irradiation to determine the longer-term effects of radiation in this commonly used model of radiation-induced salivary dysfunction.

Male Wistar rats, 250–300 g, were purchased from Harlan–Sprague Dawley (Walkersville, MD). They were kept in polycarbonate cages in an alternating 12-hr light/dark cycle, and were maintained on laboratory chow and water ad libitum. Total body weight, parotid and submandibular gland weights and stimulated parotid and submandibular salivary flow rates (below) were determined at 3, 6, 9, and 12 months post-irradiation. The numbers of rats in the various experimental groups are indicated in the figures. Additionally, in groups of rats receiving 15 Gy and their non-irradiated controls, we measured the weights of several other organs/tissues (liver, heart, spleen, epididymal fat) after 6 months, and collected blood to determine several routine laboratory serological and haematological measures.

Rats were anaesthetized and irradiated as described previously. All animals were given a single acute exposure of 2.5, 5, 7.5, 10 or 15 Gy delivered by two opposing 250-kV therapeutic X-ray tubes (Philips Medical Systems, Inc.) operated at 235 kV, 15 mA, filtration of 0.25 mm Cu and 0.55 Al, HVL 0.9 mm Cu, with an output of 1.26 Gy/min at 54 cm. Control animals were anaesthetized but were not exposed to radiation. All irradiation was done between 8:00 a.m. and 12:00 p.m.

Saliva was collected as previously reported. Animals were weighed, anaesthetized with an intraperitoneal injection of pentobarbital (50 mg/kg), and a tracheotomy was performed. Their main excretory parotid duct was identified through an extraoral incision and cannulated with polyethylene tubing (PE10; Clay Adams, Parsippany, NJ). The orifices of the main excretory ducts of the submandibular glands were identified intraorally and similarly cannulated. Salivary secretion was stimulated with pilocarpine HCl (5 mg/kg; Sigma). As soon as flow was noted, within 3–5 min, saliva was collected for 30 min into preweighted tubes kept on ice. The volume of saliva collected was determined gravimetrically assuming a sp. gr.=1.0; flow rate was expressed as the volume (in ? l) secreted in 30 min.

Data were expressed as means±SEM. Data were initially evaluated by ANOVA. Significant differences were resolved with Dunnett’s method or the Bonferroni test for comparison of multiple groups against a control. Differences between treatment groups were assessed with Dunn’s procedure for pairwise multiple comparisons. Statistical significance was defined by an -level of <0.05 for the pairwise multiple comparisons.

All irradiated and control rats that survived the first 2 weeks of the study then lived up to 3 months. Most early deaths were in the first few hours after treatment, probably as a complication of anaesthesia. At later time-points only a few, sporadic deaths (less than 8% of the total population) occurred in each dosage group except for the highest (15 Gy). All rats receiving 15 Gy died before the 12-month time-point was reached. These animals had a marked reduction in weight gain throughout their lifespan, although they had an apparently normal dentition and intact oral mucosa.

Fig. 1 shows body wts of irradiated animals at 3, 6, 9, and 12 months after exposure. At 3 months, they exhibited a radiation dose-dependent reduction in body wt; those which had received ?5 Gy showed statistically significantly less increase in body wt. The 15-Gy group on average weighted approx. 25% less than the non-irradiated controls. At later time-points only animals that had received ?7.5 Gy exhibited a consistent reduction in body wt compared to controls. By 9 months the weight of the 15 Gy group was approx. 50% less than that of the control. All animals in this group were dead by 12 months, while those that had received 10 Gy had an approx. 40% lower body wt at this time.

Fig. 1. Mean (±SEM) total body wt at (A) 3 months, (B) 6 months, (C) 9 months, and (D) 12 months after X-irradiation to the head and neck with the indicated doses. Weights significantly less than the non-irradiated controls (0-Gy group) are designated **(p<0.01). The number of animals in each group is shown in the bar.

Fig. 2 shows data on the weights of the parotid and submandibular glands at various times after radiation. Both glands showed radiation dose-dependent reductions in weight over the 12-month period. By 3 months, the average weight of the parotid gland in the 15-Gy group had fallen to about one-third that of control animals, and remained at this low level until death. By 12 months, each group of surviving irradiated rats (2.5–10 Gy) showed a statistically significant reduction in parotid-gland weight (from approx. 25% to 60%). Similarly, submandibular glands showed decreased weight 3 months post-irradiation. At this time the weight of glands in the 15-Gy group was approx. 40% less than that of the control. Further gland weight loss occurred and by 9 months the submandibular gland weighed about one-third of the control. By 12 months, additional evidence of reduced gland weight was seen.

Fig. 2. Mean (±SEM) salivary gland weights at (A) 3 months, (B) 6 months, (C) 9 months, and (D) 12 months after X-irradiation to the head and neck with the indicated doses. Weights significantly less than the non-irradiated controls (0-Gy group) are designated **(p<0.01). The number of individual gland measurements for each group is shown in the bar.

At the 6 month time-point, several other organs were removed from rats exposed to 15 Gy and weighed. As shown in Table 1, all organs weighed considerably less than those from non-irradiated controls. Liver, heart, spleen and epididymal fat were reduced by approx. 60%, 45%, 45% and 80%, respectively. This compares to an approx. 70% reduction in parotid weight and an approx. 40% reduction in submandibular weight at this time. Laboratory analysis of blood samples taken at 6 months revealed no differences between non-irradiated and irradiated (15 Gy) rats in the following measures: complete blood count, total protein, amylase, alkaline phosphatase, glucose, uric acid and creatinine. Conversely, we observed significant reductions in the irradiated rats compared to non-irradiated controls in triglyceride concentrations (mean values, n=6; 8.0 vs 15.6 mg/dl), white blood cells (5.6 vs 8.8×10³/ml), and lymphocytes (4.5 vs 8.3×10³/ml).