Evaluation of gingival microcirculation by a laser-doppler flowmeter

j. Cranio-Max.-Fac.Surg. 17 (1989) j. Cranio-Max.-Fac. Surg. 17 (1989) 105-109 © Georg Thieme Verlag Stuttgart • New York

Summary

Up to now, only qualitative studies have been performed on gingival microcirculation whereas quantitative methods have been used in animal experiments, by the injection of microspheres for instance. The authors had the opportunity of using a LaserDoppler Flowmeter (LDF) which had already served to explore the cutaneous microvascular blood flow. At first, it was necessary to modify an occlusal splint for good immobilisation of the probe on the gingival mu-

Evaluation of Gingival Microcirculation by a Laser-Doppler Flowmeter Preliminary Results

cosa.

Franck Boutault, Henry Cadenat, Pierre Jean Hibert

Then they made a study on twenty healthy students with measurements of red cell velocity in superficial maxillary gingival mucosa. Each subject was tested twice at an interval of one-month. A statistical analysis shows very good reproducibility of the tests, especially for the pulsative velocity, despite a rather high variability between the subjects. The haemodynamic parameters (heart rate, systolic or diastolic blood pressure) do not seem to have a significant influence on the values obtained. The LDF is a very good instrument for the evaluation of gingival microvascular flow, as has been demonstrated for cutaneous microcirculation. A wide range of applications can be considered for this sensitive and harmless technique.

Service de Stomatologie et Chirurgie maxillo-faciale (Head: Prof. H. Cadenat, M.D., D.M.D.) Hopitai Rangueil, Toulouse, France

Submitted 29.2.88; accepted 5.7.88

Introduction

The study of the vascularisation of the face is surely a prerequisite which is indispensable to the success of many surgical procedures in our speciality. The anatomy of the vessels with a calibre greater than i mm is now well known at the level of the soft tissues, notably by the aid of the diaphanisation technique (Cadenat et al., 1974a and b). Nevertheless, the exact appreciation of the vascular flow in the capillary vessels (with a calibre less than I ram) is still difficult. This final phase of the blood circulation constitutes however the reason for its existence, because it is at the capillary level that the gaseous or nutritionnal exchanges indispensable to the tissue metabolism are produced. To this end, we had the opportunity to use a non-invasive device which measures the capillary flow, the Laser-Doppler Flowmeter which has up till now been used essentially to measure cutaneous microcirculation (Stern et al., 1977; Boccalon et al., 1984; Oberg et al., 1984; Zeghal et al., 1986). It was necessary in the first stage to elaborate a specific device which permits the use of a probe in the mouth, and then in a second stage to evaluate the reliability and reproducibility of this method on a series of 20 healthy subjects. Material and

Methods

Technical equipment The Periflux Laser Doppler Flowmeter (LDF) elaborated by the "Perimed-Stockholm" society (Fig. 1) was lent to us for this study by Prof. Boccalon who uses it in his Haemodynamic Exploration Department at the University Hospital of Toulouse-Rangueil. It is based on the Doppler effect and it analyses the variations in the wavelength of a monochromatic light after its reflection from the tissues of the zone to be examined. The technical features of the apparatus (Fig. 2), can be identified as: - a laser light source of helium-neon type, producing a coherent beam (wavelength = 632.8 nm) with power not exceeding 2 mW,

105

Key w o r d s Gingival microcirculation - Laser Doppler flowmetry - Blood flow evaluation

the two receptor-analysers charged with the evaluation of the modifications of the wavelength of the light beams reflected, - an electronic amplifier and recorder which produces a print-out graph, - soft optic fibres which link the probe (placed almost in contact with the mucosa) to the two receptors and to the light source. In the volume explored by the LDF which corresponds to a i mm diameter half-sphere, the red blood cells cause, by their movements, some modifications of the wavelength of reflected light. As they circulate in all directions, they produce a multitude of differences in frequencies, distributed according to a Gaussian curve which is centred on the initial frequency. The width of this curve is proportional to the average velocity of the blood cells in the volume explored. A signal treatment device allows the instantaneous registration on a graphic recorder of the evaluation of the average velocity in volts. This can be considered as directly proportional to the blood flow in the capillary vessels(Zeghal et al., 1986). -

Parameters Two parameters can be evaluated by the LDF as made at the skin level (Boccalon et al., 1984, Zeghal et al., 1986). On each of the recordings (Fig. 3) signal oscillations are

106

J. Cranio-Max.-Fac. Surg. 17 (1989)

F. Boutault et al.

er ht PhL~t~)t°r L-1 amplifier ~iPt~rOsa~

b probe

Fig.1

Fig.2

The Periflux LDF system.

I

siprocessor gnal

d°t;=°r I

f tissu

l

OUtpUt signal

]

Schematic diagram of the components of the LDF.

Fig.3 Example of recording for one anatomical site, Low velocity=3 cm/minute, high velocity = 2 cm/second.

5,8 6,4 25

point by provoking total ischaemia, contrary to what is done in the study of cutaneous vessels of the limbs by placing a tourniquet. We also noted the values of the pulse (P), systolic arterial pressure (SAP) and diastolic arterial pressure (DAP) at the time of the examination.

Adaptation

60~

Fig. 4 Dimensions of the probe support and the interocclusal splint.

noted, synchronous with heart beats as well as the proof of the short rapid recording and of which the amplitude corresponds to the Pulsatile Velocity (PV). There are also secondary oscillations (connected with the respiratory system), and even tertiary (connected with the variations of the vasomotor tone) of which we do not take account in o u r protocol. The value of PV is measured in millivohs by a simple arithmetical calculation. The other parameter studied is the average basal velocity (ABV). It corresponds to the average value of the signal during a three minute recording, in comparison with the zero-point. It must be stressed that for the exploration of the buccal mucosa, it is impossible to determine the zero-

For the study of cutaneous vascularisation, an adhesive tube is used which allows strict immobilisation of the probe in relation to the skin, at approximately 1 mm from it, throughout the whole period of the examination. Indeed, the least movement can involve interference with the recordings. In the light of the impossibility of using this device in the buccal cavity, we have conceived and designed an original probe support with the aid of the prosthetists of the Maxillofacial Surgery Department (Fig. 4). It comprises an acrylic splint cut out at the average dimension of the dental arch to which is sealed the support of the probe made from a stainless steel tube 25 mm long and 6 mm in diameter. In order to improve the stability of the set-up, a 2mmthick silicone rubber sheet was applied to the upper-side of the splint. In this way, the cusps of the upper teeth can lightly engage this non-slippery layer. Therefore, it eliminates all movement. On the other hand, the mandibular teeth can slip slightly on the lower surface of the plate without leading to disturbances in the measurements of gingival vascular flow of the upper alveolar arch. In this way, the effects of the involuntary micromovements of the mandible are eliminated.

Evaluation of Gingival Microcirculation

J. Cranio-Max.-Fac. Surg. 17 (1989)

I

~

\

107

~ \

/

i

1 \

I

RI

RPM

\ \

LI

LPM

Fig.7 Standardized locations of the probe: RPM = right premolar, RI = right incisor, LI = left incisor, LPM = left premolar.

Fig.5

The four support-plates with their probe-holders.

15 VP

Time

1

10 A

?-

VP 5

10

Time

0 15

Fig.6

Situation of the support-plate and the probe during the experiment.

Fig.8 Correlation diagram of the values of PV at RI (right incisor) between TimeO and Time 1 (one month later) - Each subject is represented by a triangle - The linear regression equation is: PV1 = 0.845 PVO+ 1.68 (r = 0.861 - p <0.001)

We have concentrated in this study on the maxilla, but it is obvious that we could use an analogous procedure for the mandibular gingival mucosa. It is sufficient just to reverse the plate. In fact, we have fabricated 4 support-plates (Fig. 5) in order to be able to explore 4 zones: - the attached gum between the central and lateral incisors on each side, - the attached gum between the first and second premolars on each side.

For each of them, two series of measurements were done at a i month interval, with identical protocols all the time: - subject supine, at rest for at least 5 minutes, - adaptation in a successive manner of the 4 supportplates and recording a graph for each of them, at slow speed for 3 minutes, followed by a short recording at a fast speed (Fig. 6). In total, 8 graphs were obtained for each subject, with evaluation of the blood velocity by PV and ABV at four standardized sites (Fig. 7): right and left incisors (RI and LI), right and left premolars (RPM and LPM), at time 0 and time 1 (1 month later). The aim of this study being the affirmation of the reproducibility of the measurements for the same subject at an interval of one month, we have achieved a statistical evaluation of the results. The method employed was polynomial

Protocol Twenty healthy volunteers (most of them medical students) participated in this study. Their ages ranged from 20 to 30 years with a mean of 28. The two sexes were equally distributed. The subjects all had a good dentition.

108

J. Cranio-Max.-Fac.Surg. 17 (1989)

regression by couples of values and comparison of the means for each couple (Student's test). The influence of the haemodynamic parameters was evaluated by polynomial regression as well as a nonparametric test (Wilcoxon test).

Results

Study of the reproducibility (Table 1) All the calculated values of correlation coefficients are statistically signifiant, site by site, between time 0 and time 1. As an example, Fig. 8 offers a graphic representation of the parameters obtained for the values of PV at the level of the right incisor (r = 0.861).

Comparison between the right and left side (Table 2) For each parameter (PV and ABV) we have calculated at time 0 and time 1 the correlation coefficients between the right and left sector, for the incisors as for the premolars. A mean comparison by a Student's test was done too. Only one result is unconcordant: PV-RPM/PV-LPM at time 0.

Comparison between the incisor and premolar zones (Table 3) For each parameter (PV and ABV) we have calculated the mean 'T' between the values "RI" and "LI" and the mean "PM" between the values "RPM" and "LPM". There is no statistically significant difference except for PV at time 0 (p < 0.05), a result which is confirmed by the Wilcoxon Ttest.

F. Boutault et al. Table1 Correlation coefficients for each site between Time 0 and Time1; * = p<0.05; ** = p<0.01; *** = p<0.001; (RPM = right premolar, RI = right incisor, LI = left incisor, LPM = left premolar). RPM

RI

LI

LPM

ABV0/ABVl

0.672

0.532

0.509

0.695

PVO/PVl

0.702

0.861

0.738

0.889

Table2 Correlation coefficients between right and left side; (1) = mean comparison not significant; (2) = mean comparison statistically significant.

RI/LI

For this last part, we have calculated the arithmetic mean of obtained values for each of the four sites of the probe at time 0 and looked for the correlations between the three haemodynamic parameters and the two parameters explored (PV and ABV). None of the values of "r" is statistically significant but some are quite high.

ABVl

PV0

PVl

0.677

0.326

0.461

0.763

*

**

**

RPM/LPM

(i)

(I)

(I)

(I)

0.533

01478

01585

01777

*

*

**k

***

(1)

(1)

(2)

(1)

Table3 Mean comparison between incisor and premolar sites; (I and PM) for PV and ABV (Student's test); NS = non significant, * = p<0.05.

Study of the influence of haemodynamic parameters (Table 4)

ABV0

I/PM

ABV0

ABV1

PV0

PVl

NS

NS

*

NS

Table4

Correlation coefficients between mean values of PV or ABV and the three haemodynamic parameters studied (r = 0.433 for p < 0.05). Pulse

SAP

DAP

0.196 0.430

0.313 0.151

0.170 0.398

Discussion

A great number of techniques have been developed up till now to study the microcirculation at the mucosa level. These methods have most of the time been first tested at the cutaneous level. The more precise use radioactive microspheres (Kaplan et al., 1973) or clearance of Xe133 after direct injection (Hock et al., 1980). But they are invasive methods reserved for experimental studies in animals and cannot be envisaged for clinical human studies. They were, therefore, used to validate other methods in animals, and particularly the LDF (Stern et al., 1977, 1979; Oberg et al., 1984). If the great interest of Doppler ultrasonic flowmetry is the study of vessels larger than 1 or 2 mm, it is obviously unadapted for the appreciation of the microcirculation. Nevertheless, this is just the level which is convenient to know and study. Two other techniques have been used in man for the study of intraoral vascularization: - capillaroscopy of the gingival margin (Forsslund 1959), relatively old and especially less precise; -electrical impedance plethysmography (Kinnen and Goldberg 1978) which requires the placement of elec-

AVB PV

trodes in direct contact with the mucosa. This must surely affect the results. It seems that the LDF was at first used in 1972 by Riva et al. who were ophthalmologists. In the last ten years, several studies have been published about the real interest of the LDF for cutaneous microvascular exploration (Boccalon et al., 1984, Bruce-Chwatt 1986). The good reproducibility of this method for studying PV and ABV has already been proved at the skin level (Stern et al., 1977; Zeghal et al., 1986). The LDF seemed to us a particularly reliable method in the exploration of gingival microcirculation. We indeed accept slight inconveniences such as: - the small size of the volume explored (1/2mm 3) which could lead to a modification in the results obtained in the case of minimum displacement of the probe in relation to the density of the vascular network of the gum,

Evaluation o f Gingival Microcirculation

- the difficulty in positioning the probe, with the necessity of elaborating a special device for application in the buccal cavity, - and as a consequence, the necessity of having a trained operator. On the other hand, the advantages are numerous: this method is not vulnerable, not invasive, comfortable for the patient, rather rapid, immediately interpretable and quantified. O u r statistical study associating binomial regression and mean comparison shows that the reproducibility is established for the gingival vascularisation too, despite quite great variability between the subjects. The most reliable measurement seems to be the pulsatile velocity (PV) and not the average basal velocity (ABV), contrary i.e. to that established for the skin by Zeghal et al. (1986). This is certainly directly linked to the impossibility of carrying out a zero-point at the level of the gum, which affects the ABV parameter to a certain degree of error. This is not the case for PV. This constancy is moreover reinforced by the results of Table 2 comparing the right and left sides. The correlation is also excellent, better for PV than for ABV. The fact that we note a significant mean difference between R P M and LPM for PV at time 0 must be interpreted with care: in multiplying the number of statistical tests, we surely increase the risk of seeing a paradoxical result. It is no longer obligatory that we have a capillary vascularisation absolutely symmetrical in a total number of quite weak subjects. An analog interpretation can be effected on the results of Table 3: we see a significant difference between the PV values of the incisors and those of the premolars. Moreover, it must be equally accepted that these two results, slightly surprising at first, were obtained for the first series of measurements (time 0) and not for the second. We can, therefore, envisage the eventuality of manipulative errors during the first measurements and a greater precision gradually acquired by the experimenter in the progress of the examination programme. Finally, there is no statistically significant correlation between the values obtained for PV or ABV and the haemodynamic parameters studied, in accordance with those which were found by Zeghal et al. (1986) at the skin level. Nonetheless, we can observe that the correlation coefficient between PV and the pulse is quite high, which indicates the influence of the cardiac frequency on this measure. A complementary study on a greater number of subjects will certainly be necessary. Conclusions

The LDF gives an immediate and reproducible quantitative evaluation of the gingival microcirculation, without disturbance at the level of the measurement. It offers to the maxillo-facial surgeon a wider field of application: - c l i n i c a l studies focussing on parodontal problems, stomatitis or localisation of general diseases in the buccal mncosa, - pharmacological studies centred on inflammation or ischaemia, etc. - evaluation of the eventual consequences on the vascularisation of maxillo-facial surgery, and especially of

J. Cranio-Max.-Fac. Surg. 17 (1989)

109

osteotomies (regarding the choice of incisions for instance). With a custom-made device designed for palatal mucosal examination, it should be possible to explore the microvascular blood flow before and after surgery, especially in cleft patients. Some other applications can certainly be found in our speciality. This preliminary study has therefore to be completed by numerous other studies to appreciate its real interest. Acknowledgements

The statistical studies were carried out on a VAX minicomputer in the Medical Computing Department of CHR Toulouse with the aid of Dr. Charlet. We also thank Dr. Gladstone for his contribution in translating the original french paper. References

Boccalon, H., M.C. Venerandi, A. Lozes, P. Pueh Introduction ?~la v~locim&rie Doppler au laser. Etude de la vascularisation cutan&. J. Mal. Vascul. 9 (1984) 11 Bruce-Chwatt, A.J.: Free flap monitoring using a microcomputer linked to a laser Doppler flowmeter. Brit. J. Hast. 8urg. 39 (1986) 229 Cadenat, H., IL Combelles, R. Barthelemy, M. Fabie: La vascularisation du maxillaire sup6rieur - Ses cons6quences en chirurgie orthop6dique. Rev. Stomatol. 75 (1974 a) 139 Cadenat, H., R. Combelles, R. Barthelemy, M. Fable, M. CIouet: Pent-on aborder les ost6otomies du maxillaire sup&ieur en r4alisant d'importants d~collements vestibulaires ou palatins? Rev. Stomatol. 75 (1974 b) 743 Forsslund, G.: The structure and function of the capillary system in the gingiva in man. Acta Odont. Scand. 17 (suppl. 26) (1959) 1 Hock, J., K. Nuki, R. Schlenker, A. Hawks: Clearance rates of Xenon-133 in non inflamed and inflamed gingiva of dogs. Arch. Oral Biol 25 (1980) 445 Kaplan, M.L., M.A. Davis, P. Goldhaber: Blood flow measurement in selected oral tissues in dog using radiolabelled microspheres and Rubidium 86. Arch. Oral Biol 25 (1973) 281 Kinnen, E., H. J. V. Goldberg: The application of electrical impedence plethysmography to the study of gingival circulation. J. Periodont. 49 (1978) 528 Oberg, P. A., T. Tenland, G. E. Nilsson: Laser Doppler Flowmetry: a non invasive and cutaneous method for blood flow evaluation in microvascular studies. Acta Med. Scand. 687 (1984) 17 Riva, C., B. Ross, G.B. Benedek: Laser Doppler measurements of blood flow in capillary tubes and retinal arteries. Investigative Ophthalmology 11 (1972) 936 Stern, M.D., D.L. Lappe, P. D. Bowen, J. E. Chimosky, G.A. Holloway, H.R. Keiser, R. L. Bowman: Continuous measurement of tissue blood flow by laser-doppler spectroscopy. Am. J. Physiol. 232 (1977) H441 Stern, M.D., P.D. Bowen, P. Betta, R.W. Osgood, R.L. Bowman, J.H. Stein: Measurement of renal cortical and medullary blood flow by laser-doppler spectroscopy in the rat. Am. J. Physiol. 236 (1979) F 80 Zeghal, K., P. Geslin, A. Mantel, G. Lagrue, F. Lhoste: La v41ocim6trie Laser-Doppler - nouvelle technique d'6valuation de la microcirculation. Etude de la reproductibilit6. Presse M6d. 15 (1986) 1997

Dr. F. Boutault Service de Stomatologie et Chirurgie maxillo-faciale Hopital Rangueil F-31054 Toulouse France