The effects of high frequency, low intensity ultrasound therapyon parotid flow ratesVidya Sankar, DMD, MHS,a Thomas Prihoda, PhD,b and Geza Terezhalmy, DDS, MA,c SanAntonio, TXTHE UNIVERSITY OF TEXAS HEALTH SCIENCE CENTER AT SAN ANTONIO

Objectives. Saliva plays a critical role in maintaining oral health and comfort. Development of a single devicecapable of simulating salivary flow with no taste or adverse side effects is desirable.Study design. This clinical investigation on 23 healthy volunteers with no dry mouth complaints evaluated theeffectiveness of ultrasonic therapy in mechanical stimulation of the parotid gland to increase saliva flow. Stimulatedand unstimulated parotid saliva was collected. Differences in mean salivary flow rates and affects of age and ethnicitywere measured.Results. The mean difference between stimulated and unstimulated saliva flow rates was 0.009 mL/min � SD �0.042, t (df � 22) � 1.27 with P � .31. There were no significant differences of age (P � .26), gender (P � .38), orrace/ethnicity (P � .58) for the difference of stimulated and unstimulated conditions.Conclusions. Results indicate that high-frequency low-intensity ultrasound therapy is not effective in stimulating

salivary flow rates. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;106:e16-e19)

Saliva plays a critical role in maintaining oral healthand comfort. Saliva has antibacterial, lubricant, remi-neralizing, digestive, soft tissue reparative, buffering,and cleansing properties. Therefore, decreased salivaryproduction or altered salivary composition may resultin numerous clinical conditions affecting oral health,comfort, and quality of life.1-3 Many systemic diseasescan affect salivary gland function, either directly orindirectly. Medication usage is by far the most commoncause of salivary gland hypofunction and xerosto-mia.4-6

Current methods used to relieve xerostomia in pa-tients with diminished salivary output who retain theability to produce saliva include modification of med-ication schedules so that the peak xerostomic effectoccurs when there is maximal natural stimuli for sali-vation; substitution of one medication with anotherwith less xerostomic side effects; mechanical stimu-lants such as gum and candies; systemic salivary stim-

aAssistant Professor, Department of Dental Diagnostic Science, TheUniversity of Texas Health Science Center at San Antonio.bAssociate Professor, Department of Pathology, The University ofTexas Health Science Center at San Antonio.cEndowed Professor in Clinical Dentistry, Department of DentalDiagnostic Science, The University of Texas Health Science Centerat San Antonio.Received for publication Mar 31, 2008; returned for revision Jun 20,2008; accepted for publication Jun 24, 2008.1079-2104/$ - see front matter© 2008 Mosby, Inc. All rights reserved.

doi:10.1016/j.tripleo.2008.06.013

e16

ulants such as parasympathomimetic agonists; and ar-tificial salivas and/or oral lubricants.

There are various disadvantages associated withmany of these agents. For example, the increased sug-ars in gums and candies, along with the decreasedsalivary flow further increase the incidence and rate ofcaries; side effects of the parasympathomimetic ago-nists such as the increased sweating, diarrhea, hyper-tension, hypotension, and tachycardia restrict the use ofthese products; and many patients object to the tasteand/or texture associated with many of the artificialsalivas on the market. Additionally, patients find thatthe oral lubricants do not last long enough to providelasting beneficial results. Therefore, the development ofa single device with no taste, objectionable textures, oradverse side effects would be desirable.

Other therapies for treatment of xerostomia haveincluded electrostimulation using intraoral devices,which have yielded moderate to substantial improve-ment7 and transcutaneous electric nerve stimulation(TENS)8 using an extraoral device. TENS was capableof increasing parotid saliva flow in 68% of healthyadult subjects. However, a substantial number of sub-jects experienced side effects including twitching offacial musculature and anesthesia of cutaneous areasadjacent to the placement of the device.8

The proposed therapy for the treatment of xerostomiain this study is a device that uses high-frequency,low-intensity ultrasound. This device transmits a low-intensity ultrasound signal to the parotid gland usinghigh-frequency acoustic pressure waves to mechani-

cally stimulate the acinar tissue of the parotid gland.

OOOOEVolume 106, Number 5 Sankar et al. e17

The device provides 30 mW/cm2 intensity. This ultra-sound level is low and comparable to diagnostic ultra-sound levels used in sonograms; about 1% to 5% of theintensity of that used in conventional therapeutic ultra-sound. Human studies of the device thus far havecentered on its use in improving bone healing afterfracture.9-11 We believe that this device may be usefulin stimulating salivary flow because of its ability tomechanically stimulate parotid tissue and because thebiophysical effects of ultrasound include mechanicaleffects on the cell membrane and pressure changes thataffect cellular membrane transport pathways.12 Ultra-sound also has the ability to modulate intracellularcalcium and potassium levels via the Na-K ATPase.13

Salivary gland secretion is hypothesized to be modu-lated by the Na-K ATPase controlling intracellular cal-cium concentrations which change the osmotic gradientfor NaCl resulting in transepithelial movement of waterwithin the salivary glands.14

The objective of this clinical investigation was toevaluate the effectiveness of ultrasonic therapy to me-chanically stimulate the parotid gland to increase pa-rotid saliva flow.

METHODSThis study was approved by the University of Texas

Health Science Center Investigational Review Board. Itwas a clinical investigation to determine if ultrasonictherapy is effective in stimulating parotid salivary flowperformed at a single center. Subjects were healthyvolunteers with no dry mouth complaints and not cur-rently taking medications that have dry mouth sideeffects. Patients with dry mouth complaints, taking anymedication known to cause xerostomia, currently usingprescription or over-the-counter medications to controldry mouth, or with a history of Sjogren’s syndrome orxerostomia secondary to radiation or chemotherapywere excluded.

The device itself is composed of 2 piezoelectrictransducers secured in a plastic sheath that is connectedwith wires to an external ultrasonic power supply thatruns on C-cell batteries (Fig. 1). The transducers wereattached via adhesive tape extraorally covering the rightor left parotid gland of each patient (Fig. 2). To enhancetransmission of the ultrasound waves through the con-nective tissue, a thin coating of coupling gel was ap-plied onto the surface of the transducer that contactedthe skin. Ultrasound is conveyed only through a con-duction medium; thus, the effect is localized to the areaof the gland itself.

Parotid saliva was collected using the followingwell-known and noninvasive procedures. All salivacollection was done between 8:00 and 11:00 AM.

Subjects were instructed to refrain from oral activi-

ties such as eating, drinking, tooth brushing or floss-ing, and cigarette smoking for 1.5 hours before thecollection. The subjects were seated in an uprightposition for at least 15 minutes. The ultrasonic de-vice was attached extraorally to the right or leftparotid gland. Carlsen-Crittenden saliva collectioncups were then placed. After 2 minutes, saliva wascollected for 1 minute (unstimulated salivary flow).

Fig. 1. The high-frequency, low-intensity ultrasound deviceshown here contains only 1 of the 2 piezoelectric transducerssecured in a plastic sheath, which is connected with wires toan external ultrasonic power supply that runs on C-cell bat-teries.

Fig. 2. An illustration of the device in place (attached ex-traorally) to the skin covering the parotid gland. The Carlsen-Crittenden saliva collection cups are in place intraorally withthe collection tubes extending outside of the oral cavity.

Subjects then receive ultrasound therapy at the test

OOOOEe18 Sankar et al. November 2008

site for a full 2 minutes before saliva was collectedfor 1 minute. Flow rates were expressed as mL perminute per gland.

Sample size was calculated to be 23 subjects, whichgave 0.80 power to detect a mean difference with2-tailed significance level of .05.15 The n � 23 pergroup is very conservative and allows comparisons ofsubgroups of interest, e.g., males versus females, or theyoung half compared with the older half with goodpower to detect differences in these groups.

Paired t tests were used to determine differencesbetween stimulated and unstimulated saliva flow. Chi-squared tests were used to detect differences with re-spect to age. Differences in due to race/ethnicity, gen-der, and age were analyzed with analysis of covariance(ANCOVA). Analysis of the residuals used to verifythe results were based on valid assumptions about thedistribution and homogeneity of variance of the data.

RESULTSBetween March 2007 and July 2007, 30 subjects

were enrolled in the study; 7 failed to meet inclusion/exclusion criteria. Twenty-three subjects completed theentire study; 15 males (age range 27 to 56 years) and 8females (age range 21 to 53 years). The baseline sali-vary flow rate (unstimulated) was 0.043 � 0.05 mL/min (SD). Mean stimulated flow rate was 0.052 �0.055 mL/min. The mean difference between stimu-lated and unstimulated saliva flow rates was 0.009 �0.042 mL/min, t (df � 22) � 1.27 with P � .31. Therewere no significant differences of age (P � .26), gender(P � .38), or race/ethnicity (P � .58) for the differenceof stimulated and unstimulated conditions (Fig. 3).

None of the subjects experienced any side effects ordiscomfort at all, including heat, tingling, or pain.

DISCUSSIONAlthough high-frequency, low-intensity ultrasound

therapy works through piezoelectric effects and has

Fig. 3. Parotid salivary flow rates. Bars indicate standarddeviation.

been used successfully to facilitate and expedite healing

of the soft tissues and bone, it failed to stimulate parotidsaliva flow rates in nonxerostomic individuals.

The ultrasonic device used in this study was chosenbecause ultrasound has been found to mechanicallystimulate the parotid gland. Ultrasound has biophysicaleffects that mechanically cause cell membrane andpressure changes that affect cellular membrane trans-port pathways, and ultrasound has the ability to mod-ulate intracellular calcium and potassium levels via theNa-K ATPase, which has been hypothesized to play arole in salivary gland secretion. This study failed toyield those intended results.

There may be several reasons why the device did notwork. For one, it is possible that the device was not inplace long enough to elicit a response, perhaps stimu-lation for greater than 2 minutes would have yieldeddifferent results. Second, the device was held in placewith adhesive along with a lubricant gel. It is possiblethat greater pressure may have been needed to betterdeliver the ultrasound therapy. We did not apply pres-sure to the device since mechanically pressing on theglands would have expressed saliva from the glandbiasing the results. Therefore it is possible that therewas too much of a gap between the device and the skinsurface, which caused the ultrasound waves to dissipaterather than stimulate the salivary glands.

Statistically, the standard deviation was rather large,however it was what we planned on while calculatingthe power analysis. The difference between stimulatedand unstimulated salivary flow rates was smaller thanwhat we planned to find in the study. Additional infor-mation that would have given more insight into theability of the patients to produce saliva at an expectedlevel of responsiveness would have been to use con-ventional methods to stimulate salivary flow such ascitric acid. This may be a consideration for futureinvestigations.

Given the results of the present study in subjectswithout xerostomia, it seems that in patients with di-minished salivary output but who retain the ability toproduce saliva, current methods used to relieve xero-stomia are still their only choices. These methods in-clude modification of medication schedules so that thepeak xerostomic effect occurs when there is maximalnatural stimuli for salivation; substitution of one med-ication with another medication associated with lessxerostomic side effects; mechanical stimulants such asgum and candies; systemic salivary stimulants such asparasympathomimetic agonists; and artificial salivasand oral lubricants.

In summary, the salivary glands of healthy volun-teers with no xerostomic complaints showed no re-sponse to high-frequency, low-intensity ultrasound

therapy.

OOOOEVolume 106, Number 5 Sankar et al. e19

We thank the Vincent A. and Millie M. SegretoEndowment for Oral Medicine for their funding of thisstudy and Robert T. Bock for donating the use of thedevice tested. We also thank Deirdre Douglas forassistance with the study.

REFERENCES1. Baum BJ, Bodner L, Fox PC, Izutsu KT, Pizzo PA, Wright WE.

Therapy-induced dysfunction of salivary glands: implications fororal health. Spec Care Dentist 1985;5(6):274-7.

2. Atkinson JC, Fox PC. Salivary gland dysfunction. Clin GeriatrMed 1992;8(3):499-511.

3. Sreebny LM, Schwartz SS. A reference guide to drugs and drymouth—2nd edition. Gerodontology 1997;14(1):33-47.

4. Scully C.Drug effects on salivary glands: dry mouth. Oral Dis.2003;9(4):165-76.

5. Wynn RL. The top 50 prescription medications dispensed inpharmacies in 2001. Gen Dent 2002;50(5):400-4., 406

6. Smith RG, Burtner AP. Oral side-effects of the most frequentlyprescribed drugs. Spec Care Dent 1994;14(3):96-102.

7. Weiss WW Jr, Brenman HS, Katz P, Bennett JA. Use of anelectronic stimulator for the treatment of dry mouth. J OralMaxillofac Surg 1986;44(11):845-50.

8. Hargitai IA, Sherman RG, Strother JM. The effects of electro-stimulation on parotid saliva flow: a pilot study. Oral Surg Oral

Med Oral Pathol Oral Radiol Endod 2005;99(3):316-20.

9. Heckman JD, Ryaby JP, McCabe J, Frey JJ, Kilcoyne RF.Acceleration of tibial fracture-healing by non-invasive, low-intensity pulsed ultrasound. J Bone Joint Surg Am 1994;76:26-34.

10. Xavier CAM, Duarte LR. Ultrasonic stimulation of bone callus:clinical application. Rev Brazil Orthop 1983;18:73-80.

11. Duarte LR. The stimulation of bone growth by ultrasound. ArchOrthop Trauma Surg 1983;101:153-9.

12. Dinno MA, Dyson M, Young SR, Mortimer AJ, Hart J, Crum LAThe significance of membrane changes in the safe and effectiveuse of therapeutic and diagnostic ultrasound. Phys Med Biol1989;34(11):1543-52.

13. Chen B, Huang J, Wang J, Huang L. Ultrasound effects on theantioxidative defense systems of Porphyridium cruentum. Col-loids Surf B Biointerfaces 2008;61(1):88-92. Epub 2007 Jul 26.

14. Turner RJ. Mechanisms of fluid secretion by salivary glands.Ann N Y Acad Sci 1993;694:24-35.

15. Machin D, Campbell MJ. Statistical tables for the design ofclinical trials. London: Blackwell Scientific, 1987.

Reprint requests:

Vidya Sankar, DMD, MHS7703 Floyd Curl DriveMail Code 7919San Antonio, TX 78229.

sankarv@uthscsa.edu