366

Changes in Hemoglobin Concentration

and Oxygen Saturation in Human
Gingiva With Decreasing
Inflammation*
Takashi Hanioka, Satoshi Shizukuishi, and Akira Tsunemitsu

The purpose of this study was to determine if functional changes in the human
gingival vasculature were reversible following the resolution of gingival inflammation.
Ten patients with 40 inflamed gingival sites were evaluated before and 2 weeks after the
completion of treatment. We determined the hemoglobin concentration and the oxygen
saturation of hemoglobin at each site by tissue reflectance spectrophotometry. With the
use of treatment including motivation, oral hygiene instruction, and scaling, clinical
parameters such as the gingival and plaque indices, the Periotron score, and the probing
depth were altered toward a healthier state. With the resolution of gingival inflammation,
the increased hemoglobin concentration and decreased oxygen saturation in the inflamed
gingiva were restored to normal levels. These findings suggest that reversible changes
in the localhemoglobin concentration and oxygen saturation are associated with decreas-
ing gingival inflammation in human subjects. J Periodontol 1991;62:366-369.
Key Words: Gingivitis/therapy; hemoglobin concentration; oxygen saturation.

Since gingival health can be reestablished following me-
chanical toothcleaning, clinical change in gingival inflam-
mation is reversible in humans1'2 However, several studies
of the gingival vasculature have suggested that changes in
the blood vessel morphology associated with gingival in-
flammation appear to be irreversible.5,6 Although the mor-
phologic alterations occurring in gingival inflammation may
be related to functional changes in the gingival microcir-
culation,7 it is not certain whether the corresponding changes
in microvascular function are reversible with the resolution
of gingival inflammation in humans.
Tissue reflectance spectrophotometry (TRS) has been used
as a noninvasive measure of microvascular function to es-
timate the hemoglobin (Hb) concentration and Hb oxygen
saturation.8 Our previous works in this series have shown
that TRS can continuously measure the gingival Hb con-
centration and Hb oxygen saturation in situ.912 Further-
more, we have found an increase in Hb concentration and
a decrease in Hb oxygen saturation with increasing inflam-
mation in dog10 and human11 gingiva, suggesting that the
increase in blood supply may not be enough to meet the
oxygen demand in inflamed gingiva. The present study was
undertaken to determine whether the changes in Hb con-
centration and Hb oxygen saturation in inflamed human
gingiva were reversed with decreasing gingival inflammation.
MATERIALS AND METHODS
Subjects
Ten medically healthy patients aged 22 to 62 years (4 males,
6 females) were selected from among the patients at the
Clinic of Preventive Dentistry at Osaka University Dental
Hospital. All the subjects had all their anterior teeth and
presented with gingival inflammation and no or slight al-
veolar bone loss on the radiographs in these regions. In-
formed consent was obtained from each subject prior to the
study. The patients received motivation, oral hygiene in-
struction, and scaling every 2 weeks for about 3 months.
Clinical Evaluation
Before treatment and 2 weeks after the completion of treat-
ment, the gingival reflectance spectra were measured using
a tissue reflectance spectrophotometer TS-200,tn and the
periodontal status was assessed using the plaque index (P1I),13
the gingival index (GI),14 the measurement of gingival crev-
icular fluid flow with the Periotron,* and the probing depth.

'Department of Preventive Dentistry, Osaka University Faculty of Den- tSumitomo Electric Industries, Osaka, Japan.
tistry, Suita, Osaka 565, Japan. tHarco Medical Electronic Devices, Inc., Tustin, ca.
 Volume 62
Number 6 HANIOKA, SHIZUKUISHI, TSUNEMITSU 367

Reflectance spectrophotometry was always performed be- Table 1. Frequency Distribution of Gingival Index Scores and
fore the clinical assessment, as injury to the blood vessels Plaque Index Scores at 40 Sites Before and After Treatment
of the crevicular plexus by paper strips and the periodontal Gingival Index* Plaque Index*
probe may affect the spectral data. All measurements were 0 12 0 12 3
taken from 4 sites in the labial portion of the papillary Before Treatment 0 13 27 0 11 24 5
gingiva between the incisor and lateral incisor in both the After Treatment 27 13 0 25 15 0 0
maxilla and mandible. 'Significant change following treatment: < 0.001.
The gingival Hb concentration and Hb oxygen saturation
were estimated by TRS, as previously described.10'11 In
Table 2. Changes in the Periotron Score and Probing Depth With
brief, a flexible coaxial fiber bundle for illuminating and Treatment
reflected light was used as the light guide. The optical probe
(2.0 mm diameter) was placed in gentle contact and per- Probing Depth
Periotron Score (mm)
pendicular to the gingival surface to minimize the surface Before Treatment 39.8±27.9* 2.9±0.8
mirror reflection and to obtain diffusely reflected light from After Treatment 9.9 ±8.9 1.9 ±0.5
the gingival tissue. As any pressure on the tissue causes
MeanChanget -29.8±27.0 -0.8±0.6
expulsion of blood from the capillaries, an automatic re- Significant Level < 0.001 < 0.001

cording system with a touch detector at the tip of the optical
probe was used. The spectrophotometer was equipped with
a microcomputer and a linear array silicon photo-diode as
the detector and an iodine lamp (100 W, 12 V) was used
as the light source. The reflectance spectrum in the region
from 450 nm to 700 nm was measured over 40 to 60 msec
within 1.0 to 1.5 mm from the gingival surface.12 The spec-
trophotometer recorded the differences in absorption at each
wavelength (X¡) between a standard reference (halon white
plate) and the gingiva as follows: Relative absorbance (EX¡)
=
log(Ir(standard)/Ir(tiSsue)) where Ir(standard) and Ir(tissue) were
the intensity of diffusely reflected light from standard and
the tissue, respectively. The Hb concentration and Hb oxy-
gen saturation in gingival tissue were calculated from the
following equations: Hb concentration(AE) 569_615 =
'Mean value ± standard deviation for 40 sites.
tThe mean changes indicate the mean difference between measurements
before and after treatment, and negative values show a net decrease from
the initial measurement.
oglobin (oxyHb) concentration ( ) was derived as Hb
concentration Hb oxygen saturation/100, and the deox-
yhemoglobin (deoxyHb) concentration ( ) was calculated
by subtracting the oxyHb concentration from the Hb con-
centration. The mean values of these TRS indices were
calculated from 5 reflectance spectra measured at each site.
Statistical Analysis
Since the variability of the TRS parameters was relatively

2 657_680 xA.
-

large in treated and untreated gingiva within each subject,

Hb oxygen saturation (%) ß /( 569_615 2 657.680
—

=
the mean values for each site were tested for significance.
- ) 100 where A was obtained as 577_586
-

The chi-squared test of independence was employed to

9/17 569.586. In these equations, the absorbances at 569 compare the distributions of GI and P1I scores before and
—

and 586 nm were the isobestic points of oxyhemoglobin
and deoxyhemoglobin, and the absorbance at 577 nm was
1 peak of oxyhemoglobin. Absorbance at 615 nm was the
reference point, i.e., the lowest gingival tissue absorbance
in the region from 450 to 700 nm. The scattering effect in
gingival tissue between 569 and 615 nm was cancelled by
after the treatment. The mean values of the Periotron score,
probing depth, and TRS parameters before and after treat-
ment were compared using student's i-test for matched pairs.
Sites of observation were divided into two groups according
to the initial GI score and were further selected by indicat-
ing GI 0 after treatment; the means of the posttreatment
=

TRS parameters for these sites were compared using the 2-

2 657.680. The constant a(0.72) was calculated from
the scattering coefficient at 615 nm for erythrocytes with
all oxygenated and all deoxygenated Hb, and the constant
ß(3.66) was from the scattering coefficient for erythrocytes
with all oxygenated Hb. The Hb concentration reflects the
actual gingival tissue blood volume, since there was a good
correlation between the tissue Hb concentration measured
by transmittance spectrophotometry in vitro and the Hb
concentration estimated by TRS (r= 0.930).10 The rela-
tionship between the Hb oxygen saturation and the tissue
oxygen partial pressure in vivo was a sigmoid curve, re-
flecting the tissue Hb oxygen dissociation curve, and sug-
gesting that the Hb oxygen saturation depends on tissue
oxygen extraction or oxygen consumption.10 The oxyhem-
sample r-test.
RESULTS
Table 1 shows the distribution of GI and P1I scores before
and after treatment. These scores both showed a significant
change following treatment (P < 0.001). As shown in Ta-
ble 2, significant reductions in the Periotron score and prob-
ing depth were also observed after treatment. These results
indicate that mild to moderate gingival inflammation were
altered toward a healthier state by the treatment used.
Figure 1 shows representative gingival spectra before and
after treatment. Each spectrum has 2 absorption maxima at
542 and 577 nm, but some differences between the spectral
 J Periodontol
368 CHANGES IN HEMOGLOBIN AND OXYGEN LEVELS OF HUMAN GINGIVA June 1991

0.217 ± 0.016 and 71.6 ± 4.1%, respectively (means ±

standard deviation of 14 subjects).11 For the inflamed gin-
giva before treatment, both parameters showed significant
differences (P < 0.01), but the levels obtained after treat-
ment were not significantly different to those for clinically
healthy gingiva (data not shown). Thus, both parameters
seemed to return to normal with the resolution of gingival
inflammation following treatment. The percent change of
the deoxyHb concentration following treatment was the
greatest among the TRS parameters. We have previously
I _ _ _ _ _ shown that the increase in the deoxyHb concentration was
450 500 550 600 650 700
much more marked than the change in oxyHb concentration
WAVE LENGTH (nm) in inflamed dog10 and human11 gingiva. Thus, the deoxyHb
concentration appears to reflect sensitively changes in the
Figure 1. Representative gingival spectra before (A) and after (B) treatment.
inflammatory status of the gingiva.
Baab and Oberg15 compared the blood flow in treated
patterns can be observed. In particular, the posttreatment and untreated gingivitis in dogs using laser Doppler flow-
577.6]15 value was somewhat lower than that before treat-
metry. The blood flow rate measured quantitatively by this
ment. The TRS parameters before and after treatment are technique reflects both the volume of blood cells and their
summarized in Table 3. The mean Hb, the deoxyHb, and velocity. They found slight increases in blood flow in in-
oxyHb concentrations decreased with treatment ( 0.042, flamed gingiva, but could not detect any significant de-
-0.032, and -0.011, respectively), while the mean Hb
—

crease in blood flow following the resolution of inflammation

oxygen saturation increased 7.1%. These net changes with
treatment were statistically significant (P < 0.01 for oxyHb
concentration and < 0.001 for the other parameters). The
percent change of deoxyHb concentration with treatment
was greater than those of the Hb concentration, Hb oxygen
saturation, and oxyHb concentration. Table 4 shows dif-
ferences in the TRS parameters after treatment between 2
groups which were formed according to the initial GI score
and were further selected by GI 0 after treatment. The
=
mean Hb oxygen saturation in the sites showing GI 2 =
before treatment and GI 0 after treatment was signifi-
=
cantly lower (P < 0.05) than that in the sites with GI 1
before the treatment and GI 0 after the treatment. The
=
mean Hb and deoxyHb concentrations in the sites where
the initial GI was 2 were relatively high compared with
those in the sites where Gl was 1, but there were no sig-
nificant differences between these TRS parameters.
DISCUSSION
Our previous study showed that the Hb concentration and
Hb oxygen saturation in clinically healthy gingiva were
that occurred with treatment. Since their dogs had very mild
gingivitis, the changes in blood flow with treatment might
have been too small to detect with a laser Doppler flow-
meter. Hock and Kim16 measured blood flow in the healed
and inflamed periodontal tissues of dogs using radio-labeled
microspheres. Microspheres become entrapped in afferent
vessels and provide a quantitative estimate of the velocity
of blood flow into a region. They demonstrated that changes
in gingival blood flow were reversible when gingival in-
flammation was resolved. Although the components of vas-
cular function estimated by the TRS method are different
=
from those shown by radio-labeled microspheres, reversible
changes in TRS parameters were also found with decreasing
gingival inflammation in our study. The decrease of Hb
concentration may mainly reflect a decrease in tissue blood
volume, suggesting a decrease in the number of functioning
vascular units. Since the local blood supply to gingival
tissue depends on both the tissue blood volume and the
velocity of blood flow, this finding may indicate a decrease
in blood supply. The increase in Hb oxygen saturation with
treatment may be indicative of a change from hypoxia to

Table 3. Changes in TRS Parameters With Treatment

Hb HB Oxygen DeoxyHb OxyHb

Concentration Saturation(%) Concentration Concentration
Before Treatment 0.264±0.046* 63.6±7.0 0.096±0.031 0.166±0.025
After Treatment 0.222±0.033 70.8±7.2 0.066±0.021 0.156±0.022
MeanChanget -0.042±0.047 7.1 ±6.2 -0.032±0.028 -0.011±0.025
Significant Level < 0.001 < 0.001 < 0.001 <0.01
Percent Change 15.9 11.2 37.7 6.6

*Mean value ± standard deviation for 40 sites.

tThe mean changes indicate the mean difference between measurements before and after treatment, and negative
values show a net decrease from the baseline.
Volume 62
Number 6
Table 4: TRS Parameters After Treatment in Gingival
According toGl Score
Gl Score
Initial =
2;
Final =
0 Final
TRS Parameter (N 14)
=
Hb Concentration 0.223 ±0.027*
Hb Oxygen Saturation(%) 68.3±6.1t
DeoxyHb Concentration 0.071 ±0.017
OxyFIb Concentration 0.152±0.019
*Mean value ± Standard deviation.
tSignificant difference between the 2 groups: < 0.05.
normoxia in the treated gingiva. Thus, even if the local
gingival blood supply decreases after treatment, it may still
be adequate to meet the tissue oxygen demand.
In order to examine the extent of recovery in TRS pa-
rameters with decreasing gingival inflammation, posttreat-
ment parameters were compared in 2 groups of sites which
were divided according to the initial GI score and were
further selected as having a final GI of zero. We found that
the Hb oxygen saturation of moderately inflamed gingiva
did not change more than that of slightly inflamed gingiva.
Our previous study demonstrated that the Hb, deoxyHb,
and oxyHb concentration increased rapidly during the first
7 days after ligation in experimental Periodontitis of dogs,
while the Hb oxygen saturation decreased gradually over 7
weeks.10 Thus, these findings suggest that the Hb oxygen
saturation in moderately inflamed gingiva may not respond
quickly to increases or decreases in inflammation.
These findings suggest that functional changes in the gin-
gival vasculature are reversible following the resolution of
gingival inflammation. Also, TRS was shown to be a useful
method for noninvasive and continuous monitoring of the
gingival vascular changes that accompany inflammation. In
this study, the subjects did not have advanced periodontal
lesions, so further studies with an increased number of pa-
tients are required to examine the changes in Hb concen-
tration and Hb oxygen saturation in different degrees of
gingival inflammation following periodontal treatment.
REFERENCES
1. Löe H, Theilade E, Jensen SB. Experimental gingivitis
Periodontol 1965; 36:177-187.
HANIOKA, SHIZUKUISHI, TSUNEMITSU 369
Sites Divided 2. Axelsson P, Lindhe J. Effect of controlled oral hygiene procedures
on caries and periodontal disease. / Clin Periodontol 1978; 5:133-
151.
3. Lindhe J, Schroeder HE, Page RC, Munzel-Pedrazzoli S, Hugoson
Initial =
1; A. Clinical and stereologic analysis of the course of early gingivitis
=
0 in dogs. J Periodont Res 1974; 9:314-330.
(N 13) 4. Saxe SR, Greene JC, Bohannan H, Vermillion JR. Oral debris, cal-
0.218 ±0.041 culus and periodontal disease in the beagle dog. Periodontics 1967;
74.4 ± 6.7t 5:217-225.
0.057 ±0.022 5. Hock JM. Vascular morphology in noninflamed healed gingiva of
0.161 ±0.025
dogs. J Clin Periodontol 1979; 6:37-44.
6. Söderholm G, Egelberg J. Morphological changes in gingival blood
vessels during developing gingivitis in dogs. / Periodont Res 1973;
8:16-20.
7. Hock JM, Nuki K, Schlenker R, Hawks A. Clearance rate of Xenon-
133 in non-inflamed and inflamed gingiva of dogs. Archs Oral Biol
1980; 25:445^149.
8. Sato N, Matsumura T, Shichiri M, Kamada T, Abe H, Hagihara B.
Hemoperfusion, rate of oxygen consumption and redox levels of mi-
tochondrial cytochrome c ( + c) in liver in situ of anesthetized rat
measured by réflectance spectrophotometry. Biochem Biophys Acta
1981; 634:1-10.
9. Hanioka T, Shizukuishi S, Tsunemitsu A, Joh S, Sugiyama K. Basic
study on gingival hemodynamics and oxygen sufficiency estimated
by tissue reflectance spectrophotometry (in Japanese). / Dent Health
1987; 37:352-360.
10. Hanioka T, Shizukuishi S, Tsunemitsu A, Joh S, Sugiyama K, Sato
N. Haemoglobin concentration and oxygen saturation in dog gingiva
with experimentally induced Periodontitis. Archs Oral Biol 1989;
34:657-663.
11. Hanioka T, Shizukuishi S, Tsunemitsu A. Hemoglobin concentration
and oxygen saturation of clinically healthy and inflamed gingiva in
human subjects. / Periodont Res 1990; 25:93-98.
12. Hanioka T, Amano A, Tamagawa H, Shizukuishi S. Evaluation of
the depth sensitivity of tissue reflectance spectrophotometry in gin-
gival measurement. / Osaka Univ Dent Sch 1990; 30:1-7.
13. Silness J, Löe H. Periodontal disease in pregnancy. II. Correlation
between oral hygiene and periodontal condition. Acta Odontol Scand
1964; 22:121-135.
14. Löe H, Silness J. Periodontal disease in pregnancy. I. Correlation
between oral hygiene and periodontal condition. Acta Odontol Scand
1963; 21:533-551.
15. Baab DA, Oberg PÂ. Laser Doppler measurement of gingival blood
flow in dogs with increasing and decreasing inflammation. Archs Oral
Biol 1987; 32:551-555.
16. Hock JM, Kim S. Blood flow in healed and inflamed periodontal
tissue of dogs. J Periodont Res 1987; 22:1-5.
Send reprint requests to: Dr. Satoshi Shizukuishi, Department of Pre-
ventive Dentistry, Osaka University Faculty of Dentistry, 1-8, Yama-
in man. / daoka, Suita, Osaka 565, Japan.
Accepted for publication January 8, 1991.