Application of thermography to the evaluation of the histamine skin test in man

Application of Thermography Histamine Skin Test in Man

to the Evaluation of the

TOSHIHIKO UEMATSU, YOSHIHARU TAKIGUCHI,

ATSUHIROMIZUNO, KEIZO SOGABE,

AND MITSUYOSMI NAKASHIMA

The degree of skin flare IO min after intradermal injection of histamine (0.062 t.r,gin 0.1 ml saline) was evatuated both visually and by thermography in 6 healthy male volunteers. lntradermally injected histamine increased the skin flare area in a dose-dependent manner. There was a good exponential correlation (r = 0.963) between the flare area evaluated visually and that evaluated by thermography, indicating that the degree of underestimation of the flare area by visual inspection increased with increasing flare size. This could be due to the fact that the rather irregular shape of a flare was rounded by the eyes. Suppressive effects of an antihistamine, terfenadine (20, 40, 60, and 120 mg, p.o.), on the flare caused by histamine (2 yg/O.l ml saline) were evaluated in a similar way in a single-blind, crossover controlled trial using 15 healthy male volunteers. The histamine skin test was evaluated both visually and by thermography before and 4 hr after administration of placebo or each of 4 doses of terfenadine. Terfenadine (20, 40, and 60 mg) suppressed the skin flare in a dose-dependent manner, but 60 and 120 mg had almost the same effect. In all cases the flare area evaluated visualty was smaller than that obtained by the thermography, although the degrees of reduction of flare area by the drug were similar by both methods when expressed as a percentage of placebo. In conclusion, thermography could serve as a tool for the objective evaluation of the histamine skin test.

Key Words:

Histamine

skin test; Thermography;

Antihistaminic

terfenadine;

Flare area

INTRODUCTION An intraderm~l injection of hjstamine in man causes a local wheal and a flare around it associated with heat, which is known as a triple response, i.e., localized redness, flare, and wheal (Lewis, 1927). The suppressive effects of a drug on this response are frequently used to evaluate an antihistamini~ action (Kerp and Tie, 1966;

Joubert

et al., 1969;

Baugh

and Clalvert,

1977;

Huether

et al., 1977;

Harvey

and Schocket, 1980; Summers et al., 1981; Meyrick Thomas et al., 1985). Although either the wheal area, or the flare area, or both have been used as indicators of

Fromthe Department of Pharmacology, Hamamatsu University School of Medicine, Hamamatsu 43131, Japan. Address reprint requests to: Dr. Tosh~hiko Uematsu, Department of Pharmacology, Hamamatsu University School of Medicine, Hamamatsu 431-31, Japan. Received September 11, 1986; revised and accepted February 23,?987.

Journal of Pharmacological 6 lb7

Else&r

Methods

Science Publishing

18,703-110

(19871

Co., Inc., 52 Vanderbilt

0160-54oz871$03.50 Avenue, New York, NY 10017

104

T. Uematsu et al.

antagonism of the effect of histamine, the wheal area tends to be preferred (Fowie et al., 1971). The primary reason for this is the inaccuracy of the determination of a flare area. Thermography, which turns the distribution of heat radiation given off by human tissues, e.g., skin, eye, heart in an open-chest surgery etc., into an isothermal graph, has been proved to be effective in the diagnosis of jnflammatory diseases such as rheumatoid arthritis (Bacon et al., 1976; Colavita et al., 1983). As the flare following the intradermal injection of histamine is caused by vasodilatation and is, therefore, associated with heat, it can be expected that the flare area can be objectively measured using the thermography. In the present study we have first compared the visual estimation of the flare area with that obtained by thermography and, then, attempted to evaluate the antihistaminic effect of terfenadine (Merrel DOW, Japan), an ~~-histamine receptor blocking agent as compared with placebo (Kinsotving et al., 1975).

MATERIALS AND METHODS The thermal video system (Nippon Avionics Co., Ltd.) was used throughout this study. This system can show the difference of surface temperature of an object as the difference in colors with a range of 16 colors. The maximum power of discrimination is O.l”C. Experimental

Design:

Experiment

I

Six healthy male volunteers, aged 19-22 years, participated after giving informed consent. Each subject received a series of intradermal injections of 0.1 ml saline containing histamine (histamine phosphate, Eli Lilly) on the flexor surface of either forearm every hour. The doses of histamine were 0 (saline alone), 0.06, 0.125, 0.25, 0.5,1, and 2 pg in this sequence, and two sequential injections were performed on the contralateral forearms. Prior to each injection, a control thermography of the injection area was recorded in a diskette using a disc unit (TV-4019). As we had determined in preliminary experiments that the flare area reached a maximum by 5 to 70 min and remained at a maximum tiIJ 12 to 20 min depending on the dose, we measured the flare area 10 min after the injection. First, the thermography of the same area as the control was recorded in the diskette, and, then, the flare area visually observed (hereafter we call it “‘visual flare”) was delineated with a ballpoint pen. The visual flare delineated was taken on a video recorder (VO-5800, Sony) combined with the thermography by the use of a video camera that has a mixing device (TVM-4023). Afterward this video was played back and analyzed by an image analyzing device (PIAS EM256, Japan PC Systems Co.1 connected to a personal computer (PC-9801F2, NEC). The area of a visual flare was thus measured. For the determination of the area of a flare by the thermography (hereafter we call it a “thermal flare”), the thermography was afso played back, and the colors of the control were subtracted from those obtained after injection. The area of the remainj~g portion was also measured by the above mentioned image analyzing device.

105

106

7. Uematsu et al.

Experimental

Design:

Experiment

II

Fifteen healthy male volunteers, aged 23-58 (mean 38.9) years, were accommodated in an air-conditioned room at a constant temperature (25°C) every Sunday for 5 weeks. They were allocated randomly to 5 groups, and each group (3 subjects) was administered either vehicle (syrup alone), referred to as placebo, or 20,40, SO, or 120 mg of terfenadine in syrup on an empty stomach at 8:00 AM using a latinsquare method. Histamine skin tests were performed before and I, 2, 4, 6, and 12 hr after terfenadine admj~istration. Each subject received an intradermal histamine (2 ~g/O.~ ml saline) injection in the same manner as in Experiment 4. Injections and visual evaluations of flare and wheal were done throughout the study by the same dermatologist, who was not involved in the thermographical study and was blind to the te~e~adine doses administered. Only before and 4 hr after the administration of terfenadine, when the effect of drug was considered to be maximal Wuether et al., 1977; ~akashima et al., 19861, was the thermal area recorded on a videotape. Prior to video recording, the wheat and flare were delineated with a ballpoJnt pen, and the outlines were transferred to a case record form with transparent adhesive tape. tn this experiment the contralateral forearm served as a control. Thermography was played back on a dispiay, and the colors of the corresponding portion of the contralateral forearm were subtracted from the thermography of the flare. The remaining portion was delineated on a transparency put on a display, and the outlined area was measured with a planimeter. The minimal discrimination power of temperature of the thermography was set at 0.2”C in the Experiment I and 0.3”C in the Experiment il. The data are expressed as means of SEM. Statistical evaluations were performed using Student’s paired t-test and Scheffe’s multiple comparison procedure, where appropriate. Statistical probabilities less than 0.05 were considered to be significant.

RESUl_TS Both the visual and thermal flare areas increased with increasing doses of histamine. As can be seen from Figure 1, the thermal areas were significantly larger than the visual ones at the higher doses of histamine (‘I and 2 &g/O.1 mJ saline). Figure 2 shows the correlation between visual and thermographic evaluatjons of the areas of all the flares in the six subjects. The correlation was exponentia4 (Y = 7.49fe0.mx - 1); r = 0.963,p -C 0.01).This means that the Jarger the flare is, the greater is the visual underestjmation as compared with the thermographic assessment. The shape of the thermal area tended to be rather irregular, whereas the shape of the visual one tended to be more rounded. Table 1 indicates the time course of the flare and wheal evaluated visually in the subjects of Experiment II when they were administered placebo. The wheal area was rather constant until 8 hr after the administration, whereas the visual flare decreased by 4 hr and fluctuated afte~ard. The ant~histam4n~c terfe~adine from 20 to 60 mg reduced both the visual and the thermal flares in a dose”de~endent manner, but 60 and 120 mg produced similar effects (Figure 3). The percentage of change

Thermography

I

0.06

Flare

*

-=-Thermal -=-Visual

0

and Histamine

/

0125

0.25

0.5

Dose of Histamine

1.0

I

2.0

(pg)

FIGURE 1. Dose-response relationship of the histamine-induced skin flare area (cm21 evaluated by eye (Cl) and thermography (m). Each point and bar represent mean f SEM of responses of six subjects at each dose of histamine. Each dose of histamine was injected intradermally on the flexor surface of a forearm every hour, and the visual and thermal flares were evaluated 10 min after (*p < 0.05).

30y=749(eQoQX-1)

l

r=0.963 T :

/

.

.

t . . . /l .

20-

6” .h. z E t loif ~;I .l l l ,’

* 0 / ti* l* l *

. . .

4%

.

t 10

O” Visual

flare

I 20 (cm*)

FIGURE 2. Relationship between the visual and thermal flares. The data of Figure 1 were put together and plotted; an exponential correlation was obtained (r = 0.963; p < 0.01).

107

T. Uematsu et al.

terfenadine

0

40 120 20 60

mg

*-= 3 HGURE 3. Effect of orally admioisfer~ terfenadi~e on the histamine-induced skin flare: (Af visual flare; (B) thermal flare. The upper panel indicates the area of the skin flare 4 hr after the administration of placebo or each dose of terfenadine. The data are expressed as percentages of the predose value of flare. The predose value means the value of flare before the admfnistration of either placebo or each dose of terfenadine (see Table I). Symbols, * and +, show significant differences as compared with the placebo (0 mg in the figure), using Scheffe’s multfple comparison proc~ure (p < 0.01) and 5tudeot’s paired f-test tp < 0.05), respectively. fn the lower panel, the decrease of the flare of each dose of the drug is expressed as a percentage of change from the value of flare following placebo admioistratioo.

in flare area from placebo was similar in visual and thermal determinations (Figure although the latter was always larger than the former (Figure 3, upper panel). Around the injected portion a histamine wheal was clearly visible as a small area of lower temperature, compared to the surrounding flare in only 3 out of 1.5 volunteers. This did not disturb the accurate measurement of flare area. One subject was excluded from the analysis, because the well-developed superficial veins interfered with the evaluation of thermal flare.

3, lower panel),

We have defined the flare thermographicalfy as the area of skin that has a temperature of 0.2~0.3% higher than the surrounding area. At the lower doses of intradermal histamine, the thermal flare was almost equal to the visual one and our thermal definition seemed to be appropriate (Figure 1). The thermal flare was, however, larger than the visual flare at the higher doses of histamine. As shown in Figure

Thermography

and Histamine Flare

2, this means that the larger the flare is, the more our eyes tend to underestimate

the flare. Indeed, the shape of flare in the thermography was rather irregular, while the delineated shapes of visual flares were rather round. Thus, thermography can serve as a tool for more objective evaluation of histamine flare. The thermal flare was reduced by terfenadine up to 60 mg in a dose-dependent manner, and the degree of reduction was parallel to the change in the visual flare when the percentage of reduction of flare at each dose from the value at placebo was compared in each individual (Figure 3, lower panel). However, there was a large difference between the two kinds of flare when the reduction from the value of each flare before the administration of placebo or terfenadine, which is referred to as predose value, was compared (Figure 3, upper panel). The visual flare was largely reduced at the higher doses of the drug (by about 80%), whereas the thermal flare still remained about 50%. Throughout this series of experiment, the same dermatologist, who was not involved in the thermographical experiment, evaluated the flare and wheal areas visually. As can be seen from Table 1, it seemed that this examiner tended to underestimate the flare area with time, because the wheal remained rather constant at least until 6 hr after the administration, although there is still a possibility that the reduction of the flare might be due to the “diurnal change” in responsiveness to intradermal histamine. In addition, within the range of the flare area following placebo (Table I), it can be assumed that a small change of visual flare may cause a larger change in thermal flare (Figure 1). These results may explain the rather large discrepancy between the visual and thermal flares in Experiment II and suggest again that the thermography serves as a tool for more objective evaluation of histamine flare. We consider two other possible explanations for this phenomenon. First, there may be some discrepancy between the redness caused by vasodilatation and the heat produced by the inflammatory reaction, because rheumatoid arthritis does not necessarily accompany a local redness of skin and can be detected by the thermography as a local area of higher temperature (Colavita et al., 1983). An antihistaminic may suppress not only thevasodilatation but also the inflammatory reactions that produce a local heat, and there may be differences between the actions of an Hi-antagonist, terfenadine, on these two events. It has been shown that the skin reaction to intradermal histamine is mediated by not only H1- but also by Hz-receptors (Harvey and Schocket, 1980; Meyrick Thomas et al., 19851, and the latter contributes to the cellular reaction in the case of inflammation (Canellin and Parsons, 1982). Second, the eyes may recognize the outline of a flare by the degree of the change in redness, and the antihistamine may alter this feature of the flare. Unfortunately, the distribution of temperature within the flare could not be determined in all cases, because in some cases the higher temperatures of the skin were outside of the range of thermography, although this did not have any influence on the measurement of thermal flare area. Anyway, further experiments should be conducted to clarify this point. Although it is possible to measure a skin flare objectively with thermography, there are some problems yet to be solved. The superficial veins, which, of course,

109

110

T. Uematfu

et al.

do not look red, irradiate heat and interfere with an accurate measurement of the thermal flare. We have chosen a portion of the forearm on which such veins were not apparent, In ~x~er~rne~t 44, huwever, we could not determine the thermal ffare in one subject who was very iean and had many we44 deve4o~ed superficial veins. We excluded this subject from the study. Addjtio~ally, some mechanical stimulation of the skin, for example, scratching or rubbing, may cause vas~di4atati~n of the skin that would influence the evaluation of a flare area. In Experiment II (Figure 3), the visual flare was delineated by a ballpoint pen, the outline was transferred with a transparent tape, and, then, the thermal flare was evamated. This procedure might have caused some mechanical stimu4ation of the skin that might have interfered with an accurate evaluation and resuited in a poorer ~orre4atj~n between the v4sua4 and thermal fiares as compared to Experiment 4. The tatter problem coufd be solved by determ~~at4o~ of therma flare prior to a visual inspection. The thermal method requires an expensive equipment and, therefore, may not be accessible to some researchers.

Bacon PA, CoBins Al, Ring Fj, Cosh +!A(‘F976) Thermography in the assessment of inflammatory arthritis. C/in Rheum Dis 2:51-65. Baugh R, CIalvert RT (1977) The effect of diphenhydramine alone and in combination with ethanat on histamine skin response and mental performance. Eur j Ci& ~~armac0112:201-204. CoIavita N, Orazi C, Fusco A (1383) Role of teletbermography and arthrography in rheumatoid wrist. Diagn Imaging 52:189-1Y6. Cook I, Sbuster S 11980) Histamine wheaf formation in man. Brj P~~f~~~o/ ij9:579-585. Fowle A%, Hughes DID, Knight GI f1971) The evaluation of histamine antagonists in man. Eurj Ciin Pbarnxml3:215-220. Ganellrn CR, Parsons ME 11982) P~~~~~c~l~~y t-k&mine Receptors. Bristol: WrighVPSG.

of

Harvey RP, Schocket MD (1950) The effect of f-f and H blockade on cutaneous histamine response in man. j Ailer~y Clin ~~~~~0~ 65:136-X%9. Huether Kj, Renftier G, Barraud N, Burke jr, Kochweser J f1977) lnhibito~ activity of t~rfe~~d~~e on hjstam~n~-induced skin wheals in man. Eur f CIin Pharmacol12:195-199.

Joubert L, Gaut Z, Abrams WB (1969) A new ztp preach of the use of histamine skin tests in the study of antihistamine drugs. C/in Pharrnaco/ Pher 10:250-2.57. Kerp L, Tie PN (1966) Pruefund des Antihistamins HS 992 am Histam~n-bow. “4~8~‘-~~th~rn und bei Patienten mit aliergishen Sofortr~aktio~en~ Med weli S1:27Y4-2798. Kinsoiv~ng CR, Munro NL, Carr AA 11975) Separation of the CNS and Hi receptor effects of antihistamine agents. ~harmaca~ogis~ 15:221. Lewis T (19227)fhe B/o& Vessels of the Ffuman Skin ano’ Their Respomes, London: Shaw & Sons, Ltd. Meyrick Thomas RH, Browne PD, Kirby JRT (1985) The influence of ranitidine, alone and in combination with clemastine, on h~stamine~m~d~ated cutaneous wheal and flare reactions in human skin. Br j C/in Pharmacd 20:377-382. Nakashima M, Uematsu I, Takigucbi, Y, Hashimoto, H, Harada S, fei K (19861 &f~n~copharmacologicaf study of terfe~ad~ne, a new antihistaminic agent: safety study, pharmacokine~~cs, and inhibitory effect on wheat and flare induced by histamine on nofmat volunteers. 1 CIin %er hiled 2:?063-‘IO80 (in Japanese with English abstract!.