Salivary flow rates measured during radiation therapy in head and neck cancer patients: a pilot study assessing salivary sediment formation.

STATEMENT OF PROBLEM

Xerostomia often occurs in patients being managed for head and neck cancer who receive radiation therapy. Although accurate salivary sampling can be therapeutically important to measure during radiation, sampling errors can occur because of salivary sediments. Determining the impact that salivary sediments have on measured salivary flow rates during radiation is important for management of patients.

PURPOSE

The purpose of this study was to assess the magnitude of error associated with the inclusion of nonsalivary components (sediment) in the calculation of whole stimulated saliva flow rates prior to and during radiation therapy (SS and SSR) in patients with head and neck cancer.

MATERIAL AND METHODS

Whole paraffin-stimulated saliva was collected in large-mouth centrifuge tubes from 20 patients with head and neck cancer prior to and during the third week of radiation therapy. Gravimetric methods were used to calculate the flow rates at g/5 min. After centrifugation, supernatant saliva was removed and the sediment was oven-dried to remove residual moisture. Sediment weight was subtracted from the original weight of saliva specimens and flow rates were recalculated. Means and standard deviations were determined and flow rate differences before (BC) and after (AC) sediment correction were evaluated statistically with the paired t test (alpha=.05). A nonparametric analysis of the flow rate data with the Wilcoxon matched-pairs signed-ranks test was also used to examine the magnitude and direction of the intrapair (BC-AC) differences (alpha=.05).

RESULTS

On average, salivary sediment contributed less than 1% of the total uncorrected weight of saliva prior to radiation therapy. In specimens collected during radiation therapy, sediment contributed an average of 14% of the total uncorrected weight and as high as 95.4% in 1 patient. Sediment percentages were 20% and higher in 4 patients. In the Wilcoxon analysis, 19 out of 20 paired BC and AC flow rates were higher in the BC group in the SS and SSR samples.

CONCLUSIONS

The error associated with the inclusion of salivary sediment in the calculation of saliva flow rates prior to radiation treatment was small, but statistically significant. The magnitude of the sediment effect was more pronounced in specimens taken during radiotherapy and was significant, as determined by the Wilcoxon test, but the mean paired differences were not significantly different according to the t test.