Histamine response and local cooling in the human skin: involvement of H1- and H2-receptors


Dr M. Grossmann, Institute of Clinical Pharmacology, Medical Faculty, University of Technology, Fiedlerstr. 27, 01307 Dresden, Germany.


Aims Histamine may contribute locally to cutaneous blood flow control under normal and pathologic conditions. The objective of this study was to observe the influence of skin temperature on histamine vasodilation, and the roles of H1-and H2-receptors using novel noninvasive methods.

Methods Eleven healthy subjects received, double-blind, single doses of the H1-receptor antagonist cetirizine (10 mg), cetirizine (10 mg) plus the H2-receptor antagonist cimetidine (400 mg), or placebo on separate occasions. Histamine was dosed cumulatively by iontophoresis to the forearm skin at 34° C and 14° C. Laser-Doppler flux (LDF) was measured at the same sites using customised probeholder/iontophoretic chambers with Peltier cooling elements. Finger mean arterial pressure (MAP) was measured and cutaneous vascular conductance calculated as LDF/MAP.

Results Histamine vasodilation was reduced in cold skin. Cetirizine shifted the histamine dose-response at both temperatures: statistically significantly at 14° C only. Combined H1- and H2-receptor antagonism shifted the response significantly at both temperatures.

Conclusions H1- and H2-receptors mediate histamine-induced skin vasodilation. The sensitivity of these receptors, particularly the H1- receptor, is attenuated at low skin temperature. Whether the reduced effect in cold skin represents specific receptor or postreceptor desensitization, or nonspecific attenuation of cutaneous vasodilation remains to be elucidated.


Histamine, a well-recognized mediator of inflammatory vasodilation, may play a role in the local control of the normal cutaneous circulation, and has been implicated in the pathogenesis of cold urticaria. The principal clinical features of this syndrome, which affects up to 0.1% of adults, are urticaria, angioedema, and occasionally symptoms of hypotension after cold exposure [1]. Cooling affects the vascular response of the normal skin to several types of stimuli [2[3]–4]. Knowledge about the interaction of cooling and histamine receptor subtype activation may aid in the understanding of normal cutaneous vascular control, cutaneous blood flow control under extreme environmental cold, and may shed further light on the pathogenesis of cold urticaria.

Histamine receptors are classified as H1, H2 and H3 [5]. The H3-receptor appears to be an autoregulator of histamine release [6, 7]. The histamine response in various vascular beds is mediated by both H1- and H2-receptors [8[9]–10]. Stimulation of H1-receptors evokes release of nitric oxide (NO) from vascular endothelium in human extracranial arteries and rat skin [11, 12]. Local cooling decreases the sensitivity to exogenous NO but increases NO release after acetylcholine challenge in rabbit ear and femoral arteries [13, 14]. Stimulation of H2-receptors causes vasodilation through direct relaxation of vascular smooth muscle in large human arteries [15]. Cooling increases the sensitivity of rabbit vascular H2-receptors in vitro [16]. Little is known about the effect of local cooling on receptor subtype specific histaminergic vasodilation in the human cutaneous microcirculation. In part this reflects the lack to date of noninvasive methods to evaluate these actions.

Histamine provocation tests, such as the prick technique, are traumatic to the skin structure, provoking the release of vasoactive compounds which may confound the interpretation of histaminergic responses [17, 18]. Iontophoresis (local electrical current-driven delivery of drugs into the dermis) is atraumatic, and permits investigation of the effects of selected, exogenously administered molecules [19, 20].

The aim of the present study was to observe the roles of H1-and H2-receptors in the response of the normal skin vasculature to histamine at normal and low temperatures, using a new method which permits (i) iontophoretic application of histamine hydrochloride, (ii) simultaneous and colocalized measurement of skin blood flow by laser-Doppler fluximetry, and (iii) discrete local cooling and warming at the same site.



Eleven healthy volunteers (four male and seven female), 22–26 years old took part in this randomised, double-blind, placebo-controlled cross-over study. Their weight ranged from 52 to 101 kg (65±14 kg). All were found to be healthy as determined by complete physical examination, electrocardiogram, and normal routine blood and urine tests. Subjects were excluded if they had any acute or chronic illness, if they had ever taken H1- or H2-receptor antagonists regularly, or if they had a history of allergies. No subject suffered from any type of urticaria or atopic skin reactions. Each subject gave informed written consent before taking part in the study, which was approved by the Ethics Committee of the University of Technology, Dresden. The study was conducted in agreement with the Declaration of Helsinki (Somerset West Amendment, 1996). All medication was prohibited 2 weeks prior to study days.


Cetirizine dihydrochloride (Zyrtec), 10 mg, a selective H1-receptor antagonist, was supplied by UCB GmbH, Kerpen, Germany. Cimetidine (Tagamet 400, SmithKline Beecham Pharma GmbH, Munich, Germany), 400 mg, a selective H2-receptor antagonist, was obtained from commercial sources. Drugs and placebo (glucose powder) were presented in identical opaque gelatine capsules. Standard therapeutic doses were selected. At these doses the therapeutic actions have clearly been established to relate predominantly to histamine H1-and H2-antagonism, respectively. Maximum cetirizine concentrations in healthy volunteers are achieved 1 h after oral administration, and the plasma half-life of the parent compound is 7.4 h [21]. Time to peak concentration for cimetidine is 45–90 min after oral administration. The plasma half-life is 2 h [22].

Direct current iontophoresis was performed with 10 mmol l−1 (1.1 mg ml−1 ) solutions of histamine hydrochloride (Synopharm, Barsbuttel, Germany) in nonconducting propylene glycol (1,2-propanediol), prepared by the local pharmacy.

Study design

All experiments were performed during the winter season (November to January) to minimize exposure to allergens [20]. All female subjects were taking contraceptive medication. To minimize the influence of hormonal status on the histamine response [23[24]–25], all tests were carried out between day 12 and 21 of contraceptive pill intake. All contraceptives had low amount of estradiol (0.03–0.05 mg) and comparable amounts of progestin (chlormadinone acetate 2 mg, cyproterone acetate 2 mg, norgestrel 0.125 mg, or desogestrel 0.15 mg).

Using a Latin Square Design, subjects received in double-blind fashion: (a) one placebo capsule and one capsule with 10 mg cetirizine (b) one capsule with 10 mg cetirizine and one capsule with 400 mg cimetidine, or (c) two placebo capsules in 150 ml water on 3 respective study days at 08.00 h after a light breakfast. An interval of at least 1 week was allowed for complete washout between study days. Subjects were instructed not to drink caffeine-containing beverages 10 h prior to the study. One subject had to be excluded from the analysis because he did not finish the study for personal reasons.

Skin blood flow measurements and iontophoresis

Measurements of the cutaneous laser-Doppler flux (LDF) and simultaneous transcutaneous iontophoresis of histamine were performed as previously described [26]. Briefly, customised aluminium probeholders were used to permit simultaneous administration of drugs by iontophoresis, measurement of LDF (PeriFlux 4001 and 4002, Perimed AB, Sweden), and control of the skin temperature. The beam of red laser light (780 nm) penetrates the skin to a maximum depth of 1.5 mm and reflects skin microcirculation, including nutritional capillaries and arteriovenous shunt flow [27]. The laser-Doppler fluximeter was calibrated with respect to a latex suspension (Perimed AB, Sweden). The units of LDF are arbitrary (PU, perfusion units).

Iontophoresis current was delivered by means of a battery-powered constant current source (PeriIont 382 Power Supply, Perimed AB, Sweden). Skin temperature was recorded by a copper/constantan thermocouple (Omega Engineering, Stamford, CT) between probeholder and skin and maintained constant at 34±0.1° C by servo control of heating elements in the probeholder block. Two probeholders were equipped with cooling elements (Peltier elements), which were connected to a power supply. Temperature on these probeholders was maintained constant at 14±1° C by manual control.

Mean arterial pressure (MAP) was measured continuously from the middle finger of the same hand (Finapres BP Monitor, Ohmeda, Englewood, CO). Cutaneous vascular conductance (CVC, PU/mmHg) was calculated as CVC=LDF/MAP.


Subjects entered the laboratory between 10.00 and 11.00 h after a light breakfast, between 2–3 h after intake of the study medication. They lay supine in a temperature-controlled (24±1° C) laboratory at least 30 min prior to data collection. The forearm rested on a rigid arm support and four probeholders (two with heating and two with cooling capabilities) holding multireceiver laser probes (PF 415–113, Perimed AB, Sweden) were attached to the forearm skin sites with double-stick disks (3 Medical Device Division, St. Paul, MN). After 10 min baseline sampling period for the two probeholders with normal temperature (34±0.1° C) dose-responses to histamine were constructed according to a schedule based on preliminary experiments and reflecting the times required to reach steady-state response after each dose (2 min between dose 1 and 2, and 5 min between subsequent doses). Six ascending ‘doses’ (100; 200; 400; 800; 1200; 1600 μA·s) of histamine were administered. At the end of each iontophoretic histamine application, subjects were asked to describe the local itching sensation on a scale from 0 (no itch) to 3 (severe itching).

From the beginning of the experiment, the two other probeholders were cooled down to 14±1° C by means of Peltier elements (approximately −1° C min−1 to avoid reflex vasodilation [4]). After the dose–response curves at the 34° C probeholders were completed, baseline LDF at cooled probeholders were measured and similar dose–response curves for histamine were constructed at these sites as described above. At the end of the experiment, Peltier elements were switched off and local skin temperature returned to normal.

Data acquisition and analysis

The data acquisition and temperature servo control system was based on an IBM-compatible personal computer and data acquisition hard- and software (LabVIEW for Windows, National Instruments, Austin, TX). To reduce spatial variation of LDF mean values measured from the two normal and from the two cooled cold skin sites were calculated.

Since no plateau of response was seen at the highest doses in some experiments, particularly in the experiments where local cooling was applied, reliable estimates of conventional dose–response parameters could not be obtained by nonlinear curve fitting. Two-way repeated measurement analysis of variance (RM anova ) was therefore performed with histamine dose and local temperature or systemic treatment, respectively, as main factors. When an interaction was found, individual paired differences were determined by pairwise multiple comparison (Fisher's Least Square Difference Method). Results are expressed as mean±s.e.mean unless otherwise stated. 95% confidence intervals for the differences between the means (CI) were also calculated. The effect of treatment on the level of itching reported after histamine challenge was analysed by Friedman Repeated Measures Analysis of Variance on Ranks. A value of P<0.05 was considered to be significant.


Histamine iontophoresis caused dose-dependent vasodilation in human forearm skin. Local cooling diminished the response to histamine significantly (RM anova, dose×temperature, P<0.01, Figure 1).

Figure 1.

Dose–response curves of cutaneous vascular conductance (CVC) after histamine iontophoresis at normal skin temperature (34° C, ○) and local cooling (14° C, ●). Error bars are s.e.mean. *P<0.01 for 14° C compared with 34° C.

Histamine iontophoresis at normal skin temperature (34° C )

After systemic treatment with cetirizine plus cimetidine the dose-response to histamine was significantly lower compared with placebo (P<0.03, Figure 2). Cetirizine alone shifted the dose-response rightwards, but this was not a statistically significant effect. The highest dose of histamine (1600 μA·s) induced significantly higher vasodilation after treatment with cetirizine plus cimetidine compared with placebo (82±18 PU mmHg−1 (CI 41, 123), P<0.001), suggesting a nonspecific effect at that dose. Subjects reported significantly less itching after treatment with cetirizine and cetirizine plus cimetidine, respectively (P<0.001 for each treatment vs placebo). There was no significant difference in itching between both active treatments.

Figure 2.

Dose–response curves of cutaneous vascular conductance (CVC) after histamine iontophoresis at normal skin temperature (34° C). Subjects have been treated with cetirizine (□), cetirizine plus cimetidine (▿) or placebo (●). Error bars are s.e.mean. *P<0.05 for cetirizine plus cimetidine vs placebo. Some error bars have been omitted for clarity.

Histamine iontophoresis at cold skin temperature (14° C )

Treatment with cetirizine alone reduced the vasodilatory response to histamine at 14° C significantly compared with placebo (P<0.05, Figure 3). The combination of cetirizine and cimetidine further decreased the vasodilation (P<0.05 compared with cetirizine, P<0.01 compared with placebo).

Figure 3.

Dose–response curves of cutaneous vascular conductance (CVC) after histamine iontophoresis and local cooling (14° C). Subjects have been treated with cetirizine (□), cetirizine plus cimetidine (▿) or placebo (●). Error bars are s.e.mean. *P<0.05 for cetirizine vs placebo; +P<0.05 for cetirizine plus cimetidine vs placebo. #P<0.05 for cetirizine vs cetirizine plus cimetidine. Some error bars have been omitted for clarity. (Note that the scaling of the CVC axis is different from Figure 1 and Figure 2).

No subject reported itch during iontophoresis on cold skin. After the local skin temperature returned to neutral at the end of the experiments (25.5±0.4° C), subjects reported significantly more itching at cold sites after placebo compared with cetirizine or cetirizine and cimetidine (P<0.001). The vasodilation significantly increased during recovery from cooling. This increase was significantly higher after either cetirizine (302±20 PU mmHg−1; P<0.03) and cetirizine plus cimetidine (282±24 PU mmHg−1; P<0.001) compared with placebo (214±12 PU mmHg−1; Table 1).

Table 1.  Maximum cutaneous vascular conductance (CVC) after histamine iontophoresis at 14° C and after return to neutral temperature at the end of the experiment. Values represents means±s.e.mean (95 percnt; CI of differences, compared with placebo; compared with cetirizine alone). *P<0.02 cetirizine vs placebo; #P<0.01 cetirizine vs cetirizine plus cimetidine. Thumbnail image of


The main findings of the present study were: (i) local cooling to 14° C substantially attenuates histaminergic vasodilation in normal subjects, (ii) in cold and normal skin both H1- and H2-activation contribute to histamine vasodilation and (iii) H1-receptorantagonism shifts the histamine dose-response in cold and normal skin, but the shift is only statistically significant at 14° C. These findings suggest a relatively greater involvement of the H1-receptor in the overall diminished response to histamine on cooling.

We cannot say whether the diminished response to histamine at 14° C represents a specific reduction in receptor sensitivity or postreceptor mechanisms, or a nonspecific attenuation of cutaneous dilation in cold conditions. It has been observed that the binding of some H1-receptor antagonists increases at low temperatures in vitro [28, 29], but no information is available for cetirizine in vivo. Adequate penetration of histamine into the skin can be assumed from the significant increase in skin blood flow compared with baseline after recovery from cooling. The maximum CVC was lower compared with the 34° C control sites, since the ambient temperature at these sites was lower than 34° C.

Direct effects of histamine on isolated vessels depends on the degree of preconstriction [30]. In endothelin-1-preconstricted rabbit ear arteries (a model for cutaneous blood vessels), relaxation to histamine after blockade of H1-receptors could be blocked by H2-receptor antagonists [16]. Local cooling increases noradrenaline release [31, 32] and enhances α-adrenoceptor sensitivity [33], which may reduce the vasodilating effect of histamine. Both H1- and H2-receptors contribute to skin vascular vasodilation after histamine iontophoresis. Treatment with the H1-receptor antagonist cetirizine alone was not sufficient to significantly antagonize the histamine induced vasodilation in contrast to the combination of cetirizine with the H2-receptor antagonist cimetidine. The significant reduction in itching after cetirizine which did not further decrease after the combination of both drugs suggests predominance of the H1-receptor in itching [8]. However, since this study did not have a cimetidine-only arm, a more significant contribution of the H2-receptor in a redundant dual-receptor mechanism may have been obscured. Cimetidine has a lower specificity than newer antagonists (such as ranitidine [34] and famotidine [35]) for the H2-receptor. It is improbable that the effect of cimetidine when combined with cetirizine simply represents additional H1-receptor blocking action, since the relative potency of cimetidine on H1-receptors is more than 20 times lower compared with H2 receptors [36].

It has been shown that 10 mg cetirizine, the dose used in the present study, is able to reduce 60–80% of wheal and 90% of flare after histamine prick test 2 h after application [37, 38]. Whether or not the combination of H1- and H2-receptor antagonists is useful in the treatment of cold induced urticaria is still controversial [39[40][41][42]–43]. Significant reduction in skin blood flow response to histamine iontophoresis after cetirizine treatment alone, without an H2-receptor antagonist, compared to placebo has been reported [19]. The reason for the discrepancy with our study appears to be mainly methodological. We constructed dose-response curves to very low histamine doses with the highest dose of 2500 μA s cm−2, whereas the single dose used by van Neste et al. [19] was 17 times higher (42500 μA s cm−2 ); this caused significant local oedema, shown by the reduction of LDF in the centre of histamine iontophoresis. The significantly greater response to the highest dose of histamine in our experiments after treatment with cetirizine plus cimetidine compared with placebo illustrates the importance of the oedema in interpretation of the vasodilation: Vasodilation in the centre of histamine iontophoresis appears to be hidden under the wheal at the two highest histamine doses in the placebo phase. Combined treatment with cetirizine and cimetidine reduced the oedema and blood flow increased also significantly at the highest dose. In addition, the significantly higher maximum CVC during treatment with either cetirizine and cetirizine plus cimetidine compared to placebo after the cooled skin returned to ambient temperature confirms, that the increase in capillary permeability, which creates the wheal, is caused by combined H1- and H2-receptor stimulation [44, 45]. The flare response, several millimetres away from the histamine exposure is transmitted via axon reflex [44, 46] and is believed to be mediated mainly by H1-receptors [8]. In our experiments, vasodilation in the centre of histamine iontophoresis was caused by direct histamine effects.

In summary, this study shows that both H1- and H2-receptors are involved in histamine induced vasodilation in human skin in vivo and that the sensitivity of these receptors, particularly the H1-receptor, is attenuated at low skin temperature. The role of histamine, and the relative place of H1- and H2-receptors in cold induced disease, such as cold urticaria, await clarification in individuals with this disorder.


This work was supported by the P. G. Unna-Stiftung, Düsseldorf. We thank Siegfried Gerber for his excellent work in manufacturing the customised laser-Doppler fluximetry probeholders.