RADIATION-INDUCED XEROSTOMIA I N CANCER

PATIENTS
Effect on Sa1ivai.y and Sei-iLm Electrolytes
SAM~JE L DDS, iMD, LEE R. BROWN,PHD, SANDRA
DREIZEN, HANDLER,
RS,
A N D BARNETM. LEVY,DDS

Saliva and s e r u m electrolyte concentrations were m o n i t o r e d in 30 patients

given a course of xerostomia-producing cancer radiotherapy. The mean flow r a t e
of stimulated w h o l e saliva decreased 83.3% during a 6-week t r e a t m e n t period.
The striking r e d u c t i o n in saliva o u t p u t was accompanied by significant increases
in saliva Na+, C1-, Cat+, Mg" and Prot.- concentrations and by a decrease i n
saliva HCO; content. The xerostomic saliva w a s more concentrated and had
a greater salinity t h a n t h e p r e t r e a t m e n t saliva in each instance. I n contrast, none
of t h e s e r u m electrolytes m e a s u r e d was significantly altered by t h e s u b t o t a l
salivary shutdown.
Cancer 38:273-278, 1976.

H IJMAN SALIVARY GLANDS IN THE PATH OF

tumor-suppressing ionizing radiation
undergo progressive glandular atrophy, fi-
brosis, and reduction in output beginning
shortly after the initial exposiire and intensi-
fying thereafter.5 Impaired production of sa-
liva strips the teeth of an important natural
defense against dental caries. T h e onset of
xerostomia, from whatever cause, is followed
invariably by the development of rampant
dental caries unless stringent measures are
taken to protect the Cross sectional
and longitiitlirial studies of tlie influence of
radiation-induced xerostomia on the human
oral microbiota reveal that the reduction in
salivation is paralleled by a shift to a highly
cariogenic microflora at the expense of non-
cariogenic organisms.lJJ Saliva protects the
From the University of Texas Dental Science I n -
stitute, Houston, Texas 77025.
Supported by Contract No. NIH-NIDR-71-2377
from the National Institute of Dental Research and
by a grant from the Kimberly-Clark Foundation, Inc.
Address for reprints: Samuel Dreizen, DDS, MD,
University of Texas Dental Science Institute, Houston,
T X 77025.
T h e authors are grateful to Drs. Thomas E. Daly
and Joe 1%.Drane, University of Texas M. D. Ander-
son Hospital, for patient selection, to Dr. Dennis
Johnston, M. D. Anderson Hospital, for statistical
consultation, to Dr. H. M. Stiles of the National
Institute of Dental Research, for administrative guid-
ance and to Jacqueline Blackburn and Patricia Carter,
for technical assistance.
Received for publication July 9, 1975.
273
teeth against caries by acting as a determinant
of the oral microflora a s well as ;I cleansing,
diluting, mineralizing, and buffering agent.
Normally, the salivary glands of man do not
play a decisive part in regulating water and
electrolyte balance since most of the salivary
constituents are reabsorbed by the alimentary
tract.13 T h e extent to wliicli radiation-iti-
duced histologic changes in the salivary
glands5 change tlie electrolyte composition of
saliva has not been documented. T o deter-
mine whether a radiation-instigated drastic
reduction in saliva output will materially
alter saliva and serum electrolyte levels, a
study of these parameters has been made in
patients given radiotherapy for head or neck
cii ricer.
k f A T E R I A I S AND METHoI)S
T h e study series was Lomposed of 30 pa-
tients with radiosensitive malignancies of tlie
lientl and/or neck treated at the LJniversity of
Texas M. D. Anderson Hospital arid T u m o r
Institute in Houston. Subject selection was
confined to patients who had not ieceived pre-
vious cancer treatment, had a natural denti-
tion in fair or good condition, were srlieduled
to receive a minimum of 5,000 r,itls of ioniL-
ing radiation at a level ok 1,000 rads a week
and whose radiation fields included coverage
of all major ralivary glands. T h e m e m age
ok the 21 men and nine women meeting these
274 1(3‘76 VOl. 38

- PH
CANCER

--- FLOW RATE (ml /min)

........ BUFFER CAPACITY
(ml 0.IN LACTIC ACID)
7.5

7.0
FIG. 1. Effect of iadiation-in-
2 .c ’ 6.5 duced xerostoinia on the mean
pH, flow rate, and buffer ca-
pacity of stimulated saliva from
30 patients with head or ncck
1 .S cancer
\
\
1.c \
\

0.s **.. ‘\,,

\

............. ................-.
-------__
:
.>q
*..
...

C I 1

RAOlATl ON RADIATION (WEEKS)

criteria was 50.5 years with ii range of from 17
to 76 years. Nine had carcinoma of the
tongue; eight, carcinoma of the nasopharynx;
five, carcinoma of the tonsil; three, carcinoma
of the floor of tlie mouth; two, carcinoma of
tlie oropharynx; two, carcinoma of tlie soft
palate; arid one malignant lymphoma of tlie
upper neck.
Samples of stimulated whole saliva gen-
el-atetl by the cliewing of sterile rubber bands
were obtained from each patient twice in tlie
week before radiation antl once weekly during
radiation. A venous blood specimen was
tlrawri a t each sampling period. T h e salivas
were collected between 8:30 antl 10:30 a.m. in
graduated g l x s tubes containing a layer of
par;iffiii oil. Collection periods were timed
precisely with a stop watch for calculation of
flow rates. Saliva in amounts adequate for
andysis were obtained from a11 30 patients in
radiation week 1; from 29, in weeks 2 arid 3;
from 28, in weeks 4 and 5; ant1 from 20 in
week 6.
T h e pH of each saliva sample was nieasurecl
with ;I Corning Model 12 Research pH Meter.
Saliva Ijuffcr capacity W;IS tlcterniinecl in I-nil
aliquots by tlie method ot Dreiren et al.“
Each saliva arid serum specimen was analyzed
quantitatively for carbon dioxide, sodium,
potassium, calcium, magnesium, inorganic
pliosphoi us, chloride, and total protein. Car-
bon dioxide was assessed in a Natelson
MotoriLed Microgasometer Model 650 by the
Natelson teclinic.’() Sodium and potassiuin
were measured by flame photometry in a
Coleman Model 21 Flame Photometer a n d
calcium aiid magnesium by atomic absorption
in ii Unicam SP 1900 Rtoniic Absorption
Spec trophotometer. Chloride and total pro-
tein were determined in a Beckman/Spinco
Ultramicro Analytical System by the ultra-
inicroada~~tations of the Schales and Schalesll
and Kingsleyi procedures, respectively. Inor-
ganic phosphorus W‘IS quantified by the micro-
colorimeti ic method of Taussky and S11orr.l~
All T ‘iliics were con\ er tetl into milliequiva-
lents per liter and means were calculated for
each sampling point. l h e data were examined
stati4cally by analysis of covariance1’ and by
the “1” test lor significant differences between
the preradiation ine~iiis and the means at
completion of i atliotherapy.
No. 1 SALIVARY IN RADIATIONXEROSTOMIA Dreizen et al.
ELECTROLYTES - 275
TABLE1. Analysis of Covariance: Magnitude and Direction of Significant Slope Changes in Salivary Parameters
During the DeveloDment of Xerostomia
Slope change
Salivary Overall* - magnitude Mean F
component mean f S D & directiont square: Ratio P
Flow rate 0.635 + ,586 - . 151 25.207 73.53 .003
Na+ 61.41 f 20.07 6.65 40797. 101.3 ,002
Ca++ 2.20 =k 1.06 ,218 44.278 39.64 008
Mg++ 0.811 f ,487 ,121 13.137 55.34 ,006
CI- 40.45 f 16.53 3.51 12999, 47.56 .006
Proteiii- 0.770 f 0.436 0.084 7,4530 39.30 ,008
* Overall mean from preradiation to 6000 rads (6 weeks).
t Magnitude and direction: negative value indicates a decreasing slope.
f Mean square = sum of squares since degrees of freedom equaled 1 in all instances.

OI~SERVATIONS was a slight decrease in pH from a preradia-

tion mean of 7.01 to a mean of 6.83 after 6
T h e mean pH, flow rate, and buffer ca- weeks of radiation. T h e average flow rate de-
pacity of stimulated whole saliva at each clined appreciably from 1.32 ml/min to 0.22
sampling period are graphed in Fig. 1. There ml/min in the same period. T h e 83.3% reduc-

TABLE
2. ComDarison of Saliva and Serum Values before and after Radiation
Saliva
Preradiation Postradiation Mean “t”
Component mean SD mean SD diff. value P
PH 7.01 0.37 6.83 0.45 0.18 1.15 > .05
flow rate* 1.32 1.33 0.22 0.24 1.10 3.64 <.001
buffer capacity’ 0.63 0.31 0.35 0.13 0.28 1.76 > .05
Na+ mEq/L 38.42 14.01 78.27 27.47 39.85 7.66 <.001
I<+ mEq/L 21.40 3.92 21.96 8.72 0.56 0.38 >.05
Ca++ m E q / L 1.51 1.07 2.80 1.42 1.29 3.72 < .001
Mg++ mEq/L 0.37 0.29 0.99 0.57 0.62 5.56 < ,001
Ci- mEq/L 24.68 18.43 45.03 17.63 20.35 4.09 < ,001
HPOa= mEq/L 7.32 2.40 6.15 2.62 1.17 1.66 > .05
HC0,- mEq/L 19.80 8.71 7.95 3.33 11.85 3.25 < .01
1’rot.- mEq/L 0.48 0.27 1.01 0.63 0.53 5.15 < ,001

Serum
Prerndiation Postradiation Mean “t”
Component mean SI) mean SD diff. value 1’
Na+ mEq/L 136.42 12.13 137.82 12.25 1.40 0.40 > .05
K+ mEq/L 4.75 0.95 4.29 0.82 0.46 1.67 > .05
Ca++ mEq/L 5.05 0.37 5.02 0.52 0.03 0.62 > .05
Mg++ InEq/L 1.81 0.24 1.73 0.17 0.08 1.32 > .05
C1- mEq/’L 97.58 12.23 95.24 11.26 2.34 0.59 >.05
HPO4= mEq/L 2.27 0.64 2.43 0.65 0.16 0.87 > .05
HCO,-mEq/L 27.04 1.95 28.16 1.85 1.12 0.43 > .05
Prot.- mEq/L 17.08 1.72 16.94 1.49 0.14 0.30 > .05
SD = Standard deviation. “t” = Significance ratio. p = Probability level.
* Millimeter per minute.
t Milliliters 0.1N lactic acid.
276 July 1976
CANCER Vol. 38

90

h
70
0.
Z 8
/

4 50 8
/ FIG. 2. Effect of radiation-in-

- 8
I duced xerostomia on the mean
saliva sodium, potassium, cal-
30 cium, and magnesium levels of
30 patients with head or neck

< 10
c(
cancer.

w 4
€
.......... ...............
2
=1 I 1 I I I 1

0
PRE 1 2 3 4 5 6
RADIATION RADIATION (WEEKS)

tion in flow rate was accompanied by a 44.4%
reduction in buffering capacity. Although the
reduction in flow rate was statistically impres-
sive, those for pH and buffer capacity were
not statistically significant (Tables 1 and 2).
T h e weekly mean saliva electrolyte concen-
trations are depicted in Figs. 2 and 3. Statis-
tically significant changes in slope magnitude
and direction during the development of
xerostomia were evidenced by Na+, Cat+,Mg”,
C1- and Prot.- (Table 1). As shown in Table 2,
the mean increases of 39.85 mEq/L in Na’,
20.35 mEq/L in C1-, 1.29 rnEq/L in Ca++,0.62
mEq/L in Mg”, and 0.53 mEq/L in Prot.- be-
tween the initial and final measurements were
all statistically meaningful ( p < .001). T h e
mean decrease of 11.85 mEq/L in HC0,- over
the same period €ailed to reach the same level
of significance (p < .Ol).
T h e xerostomia-related changes in the sali-
vary electrolyte levels occurred independent
of concurrent alterations in serum electrolyte
content. T h e r e were no significant changes in
either slope magnitude and direction during
radiotherapy or between the pretreatment and
posttreatment means for any of the serum
electrolytes tested (Table 2 ) .
DISCUSSION
T h e buffer capacity of saliva stems largely
from bicarbonates with some contributions
from phosphates and proteins.15 I n 1878,
Heidenhain6 noted that the salt content of
saliva rises to a plateau as the flow rate in-
creases in response to stimulation. This rela-
tionship, termed Heidenhain’s law by Langley
and Fletcher,s applies in general to hypotonic
salivas. I n man, Heidenhain’s law holds €or
total salt, sodium, chloride, and bicarbonate
and accounts for the greater neutralizing
power of stimulated saliva over that of resting
No. 1 SALIVARYELECTROLYTES XEROSTOMIA Dreizen et al.
IN RADIATION - 277
----- CI -
--- HC03'
......... HPO4=
60 - PROTEIN-

50

FK. 3. ERect of iadiation-in- 30

tlucetl xerostomia on the mean
saliva chloride, bicarbonate, 20
---- -----
.-
%\
phosphale, and protein levels --\-

of 30 patients with head or neck

cancer. -J
\
10
..............."............. "......
'\

U 0
Lu
€ 2.0
1.c
I I I I 1 I 1

PRE 1 2 3 4 5 6
RADIATION RADIATION (WEEKS)

saliva in the same indi~idual.4~13 stomic saliva a more concentrated secretion

Saliva formation is a two stage process.
Primary saliva with a tonicity similar to that
of plasma is elaborated by the acini and modi-
fied on passage through the ducts by partial
reabsorption of sodium and chloride and
secretion of bicarbonate and potassium.'3 T h e
bicarbonate secreted by the ducts is derived
from metabolic carbon dioxide converted into
carbonic acid by carbonic anhydrase.10 T h e
large xerostomia-related increases in saliva
sodium and chloride and decrease in
saliva bicarbonate found in this study reflect
radiation damage to both the acinar and duct
systems. Each of the other saliva electrolytes
tested increased slightly in concentration dur-
ing radiation to contribute to an overall in-
crease in total salt content. T h e changes in
the salt levels and the pronounced decrease
in water content combined to make xero-
than normal saliva.
Human salivary glands participate in the
fluid and electrolyte economy of the body
by producing from 1000 to 1500 ml of saliva
per day, most of which is recycled by absorp-
tion from the gut. T h e radiation-induced re-
duction of approximately 80% in saliva out-
p u t in this study did not materially alter any
of the electrolyte levels of the blood. These
findings indicate that the mechanisms re-
sponsible for maintaining the acid-base bal-
ance of the blood are sufficient to offset any
xerostomia-associated disturbances i n electro-
lyte metabolism.
T h e marked decrease in saliva production
that began during radiotherapy continued
to intensify after treatment and reached a
mean decline of 93.4% by 3 months postradia-
tion after which dental caries started t o be-
come clinically evident.
278 J u l y 197G
CANCER
REFERENCES
1. Brown, L. R., Dreizen, S., Handler, S., and John.
ston, D. A , : T h e effect of radiation-induced xeroatomia
on the human oral microflora. J . Dent. lies. 54:740-750,
1975.
2. del Regato, ,I. A.: Dental lesions observed after
roentgen therapy in cancer of the buccal cavity,
pharynx and larynx. A m . J. Roentgenol. 42:404410,
1939.
3 . Dreizen, S., Mann, A. W., Cline, J. K., and Spies,
T . D.: T h e buffer capacity of saliva as a measure of
dcntal caries activity. J. Dent. Re$. 25:213-222, 1946.
4. Dreizcn, S., Reed, A. I., Nicdermeier, W . , antl
Spies, T . D.: Sodium and potassium as constituents of
human salivary buffers. .I. Dent. Res. 32:497-503, 1953.
5. Frank, R. M., Herdly, J,, and Philippe, E.: Ac-
quired dental defects antl salivary gland lesions after
irradiation for carcinoma. J. A m . Dent. Asaoc. 70:868-
883, 1965.
6. Heidenhain, R.: Ueber secretorische und trop-
ische Drusennerven. Arch. GeJ. PhyAiol. 17:l-67, 1878.
7 . Kingsley, G. R.: T h e determination of serum
total protein, albumin and globulin by the biuret
reaction. J. Biol. Chern. 131:lS'i-200, 1939.
8. Langley, J . N., and Flectcher, H. M.: On the
secretion of saliva, chiefly on the secretion of salts in
it. Philos. Trans. R o y . Soc. L.ondon (Ser. B.) 180:109-
154, 1889.
VOl. 38
9. Llory, II., Dammron, A,, atid Frank, R. M.: Les
modifications de la flore buccale aerobie aprts radio-
therapie bucco-pharyngee. Arch. Oral Biol. 16:GlT-
630, 1971.
10. Natelson. S.: Routine use of ultramicromcthods
in the clinical laboratory. .I. Clin. Pathol. 21:1153-
1172, 1951.
11. Srhale5, O., and Schales, S. S.: A simple and
accurate method for the deterniinatiori of chloritle
in biological fluids. J. Biol. Chein. 140:879-884, 1941.
12. Scheffb, H.: T h e Analysis of Variance. New
York, John Wiley and Sons Inc., 1959; pp. 178-188.
13. Schneyer, L. H., Young, J. A,, antl Schneyer, C. h.:
Salivary secretion of electrolytes. PhysioE. Rev. 52:7 2 s
777, 1972.
14. Taussky, H. H., and Shorr, E.: A microcolori-
metric method for the determination of inorganic
phosphorus. J . Biol. Chern. 202:675-685, 1953.
15. Wah Leung, S.: A demonstration of the itn-
portance of bicarbonatc as a salivary buffer. J. Denl.
Res. 30:403-414, 1951.
16. Yoshimura, H.: Secretory mechanisms of saliva
and nervous control of its ionic composition. I n
Secretory Mechanisms of Salivary Glands. I.. H .
Schneyer and C. .A. Schneyer, Eds. New York, Aca-
demic Press, 1967, pp. 56-74.