Journal of Neurophysiology

Chemical Response Pattern of Different Classes of C-Nociceptors to Pruritogens and Algogens

M. Schmelz, R. Schmidt, C. Weidner, Marita Hilliges, H. E. Torebjörk, H. O. Handwerker


Vasoneuroactive substances were applied through intradermal microdialysis membranes and characterized as itch- or pain-inducing in psychophysical experiments. Histamine always provoked itching and rarely pain, capsaicin always pain but never itching. Prostaglandin E2 (PGE2) led preferentially to moderate itching. Serotonin, acetylcholine, and bradykinin induced pain more often than itching. Subsequently the same substances were used in microneurography experiments to characterize the sensitivity profile of human cutaneous C-nociceptors. The responses of 89 mechanoresponsive (CMH, polymodal nociceptors), 52 mechanoinsensitive, histamine-negative (CMiHis−), and 24 mechanoinsensitive, histamine-positive (CMiHis+) units were compared. CMiHis+ units were most responsive to histamine and to PGE2 and less to serotonin, ACh, bradykinin, and capsaicin. CMH units (polymodal nociceptors) and CMiHis−units showed significantly weaker responses to histamine, PGE2, and acetylcholine. Capsaicin and bradykinin responses were not significantly different in the two classes of mechano-insensitive units. We conclude that CMiHis+ units are “selective,” but not “specific” for pruritogenic substances and that the pruritic potency of a mediator increases with its ability to activate CMiHis+ units but decreases with activation of CMH and CMiHis− units.


The discovery of a specialized subgroup of primary C-nociceptors driven by histamine application has shed new light on peripheral itch mechanisms (Schmelz et al. 1997b). Activity in these units mirrored the subjective itch sensations, and therefore these fibers were named “itch” units. However, mediators other than histamine are known to provoke itch, albeit with lower potency: prostaglandins of the E group (Hägermark and Strandberg 1977; Woodward et al. 1995), serotonin (Hägermark 1992;Weisshaar et al. 1997), and acetylcholine (Rukwied and Heyer 1999; Vogelgsang et al. 1995). Under certain conditions bradykinin (Hägermark 1974; Rajakulasingam et al. 1991) and even capsaicin (Green and Shaffer 1993) may induce itch on topical application.

Here we have characterized pruritogenicity of these substances and studied the different C-fiber classes' sensitivity profile to these agents with different pruritogenic potency. The hypothesis was that the sensitivity of the applied substances, in histamine-responsive C-units (suggested “itch” fibers), should reflect the pruritogenic potency of these agents.



None of the subjects showed signs of neurological or dermatological disease. Recordings were obtained from 132 subjects (86 male, 46 female, age 20–35 yr) in the microneurography laboratories at Uppsala and Erlangen. Microdialysis experiments were performed in 55 subjects (32 male, 23 female, age 22–36 yr).

The subjects were financially compensated for the time spent in the experiment. They were instructed that they could withdraw from the experiment at any time and this would not affect the financial compensation. All subjects gave their informed consent according to the Declaration of Helsinki and the study was approved by the local ethics committees in Erlangen and Uppsala.

Microdialysis experiments

As injection-related pain could inhibit the ensuing itch sensation (Atanassoff et al. 1999) we used atraumatical delivery of the mediators via intradermal microdialysis catheters to assess their pruritogenicity. Microdialysis fibers (0.4 mm diam., cutoff 3,000 kDa, DermalDialysis, Erlangen, Germany) were inserted intracutaneously at a length of 1.5 cm in the volar forearm using a 25-G cannula as previously described (Schmelz et al. 1997a). No local anesthesia was used. The fibers were perfused with Ringer solution (Ringerlösung Fresenius) by a microdialysis pump (Pump 22, Harvard Apparatus) at a constant flow rate of 4 μl/min via a Tygon tubing (Novodirect). After a baseline period of 60 min, stimulation was performed with either histamine (5 × 10−6 M), serotonin (10−4M), bradykinin (10−4 M), PgE2 (10−4 M), acetylcholine (1%), or capsaicin (0.1%) for another 30 min (capsaicin, 5 min). The subjects noticed the switch of syringes but were unaware of their content. Concentrations of mediators were based on assessment of their dose–effect relation for the induction of pain and itch as previously reported (Lischetzki et al. 2001;Neisius et al. 2002). Pain and itch ratings were separately assessed on a numerical scale from 0 to 10. A value of “0” indicated no sensation and “10” indicated the maximum sensation the subject could imagine. For itch ratings a sensation provoking the need to scratch should be rated as “3”. Maximum pain and itch ratings during application of the mediators were used for statistical analysis.


Methods of microneurography employed in this study were described in detail elsewhere (Schmelz et al. 1994;Schmidt et al. 1995). Microelectrodes were inserted at the level of the fibular head into the superficial branch of the peroneal nerve. When a stable recording position in a nerve fascicle was obtained, innervation territories of single C-units were searched by transcutaneous electrical stimuli to avoid a bias toward mechanically receptive units. When C-fiber responses were encountered, a pair of needle electrodes, 0.2 mm diam., was inserted for intracutaneous electrical stimulation of that site, tips 5 mm apart. C-fiber responses were obtained to iterative constant current stimulation (0.2 ms, 10–120 V, 4-s interstimulus interval) delivered by the stimulus insulation unit of a Grass S88 stimulator.

“Marking” of activated C-units

When responses of one or several C-fibers to the intracutaneous electrical stimulation were recorded, the “marking” technique was used for characterizing the unit(s). This technique is based on the slowing of conduction velocity in a C-fiber when it is activated by an additional stimulus (Torebjörk 1974;Torebjörk and Hallin 1974). Pronounced slowing after repetitive firing is characteristic for nociceptive C-fibers (Gee et al. 1996; Serra et al. 1999;Thalhammer et al. 1994). The amount of the delay is strongly correlated to the number of additional spikes (Schmelz et al. 1995). In this study the marking technique was used for semiquantitative assessment of the chemical responses.

The C-units were functionally tested in the following order.


At first it was tested by maneuvers known to greatly increase the skin sympathetic sudomotor and vasoconstrictor outflow in conscious man (Hagbarth et al. 1970; Hallin and Torebjörk 1974; Torebjörk and Hallin 1970) whether the C-units were possibly sympathetic efferents. For this purpose sympathetic reflexes were provoked by loud unexpected noises or by inciting the subject to laugh or to perform a deep inspiration. The efficiency of these maneuvers was controlled by recording background activity of sympathetic burst discharges. C-units that showed latency increases related to sympathetic reflexes were classified as sympathetic fibers (Hallin and Torebjörk 1974;Schmelz et al. 1998).


The receptive skin fields of C-units were poked with a 750-mN von Frey nylon filament (Stoelting Co, Chicago, IL) to determine whether they were responsive to mechanical stimulation. If that was the case, this stimulus was also used for mapping the extension of their receptive fields (Schmelz et al. 1994; Schmidt et al. 1997). In addition, thinner calibrated von Frey filaments were used for determining mechanical thresholds. Heat stimuli were delivered by radiation from a halogen bulb focused to the receptive skin area and feedback controlled from a thermocouple attached to the skin (Beck et al. 1974). Mechanoresponsive units were tested inside their mechanoreceptive field and mechanoinsensitive units inside their electroreceptive field (Schmelz et al. 1994). For testing heat responsiveness, the skin temperature was increased by 0.25°C/s, from an adapting temperature of 32°C. Heating was stopped by the subject before the pain tolerance limit was reached (48–50°C). The cutoff temperature was 50°C.

Depending on the occurrence or absence of latency shifts related to these mechanical stimuli, afferent C-units were classified as in our previous studies into mechanoresponsive and mechanoinsensitive fibers (Weidner et al. 1999, 2000). Activity-dependent slowing of conduction velocity was assessed as absolute increase of response latency after 70 activations as described previously (Weidner et al. 1999).


For searching and identifying mechanoinsensitive units, transcutaneous electrical stimuli (0.2 ms, 30–50 mA) were delivered from a pointed surface electrode gently pressed to the skin. This surface electrode was also used for assessment of the innervation territory (electroreceptive field) of mechanoinsensitive units.


A Digitimer DS7 constant current stimulator was used for measuring the electrical threshold of the units to current application through a standardized probe. For this purpose, single pulses, 0.2 ms in duration, were delivered through a specially designed, hand-held applicator containing a round cotton disc, 5 mm diam., soaked in saline (Magerl et al. 1990). A large (5 × 10 cm) metal plate attached to the skin on the lower leg served as reference electrode.


The technique of iontophoretic application of histamine has been used by our group in several studies (Magerl et al. 1990). Histamine dihydrochloride (1%) was dissolved in a gel of 2.5% methylcellulose in distilled water. The 50-μl cavity of a 5-mm-diam. acrylic applicator was filled with this jelly. Current of 1 mA was delivered for 20 s from a silver-silverchloride electrode in this applicator to a large reference electrode applied to the skin distally and outside the territory of the peroneal nerve. According to the number of markings, the units were classified as histamine responsive (>40 markings) or histamine insensitive (<20 markings). Sustained histamine responses were restricted to mechanoinsensitive units, which resulted in three categories of C-fibers: 1) mechanoresponsive (“polymodal nociceptors,” all histamine negative), 2) mechanoinsensitive histamine-positive (“itch fibers”), and 3) mechanoinsensitive histamine-negative units. Responsiveness to ACh, PGE2, serotonin, bradykinin, and capsaicin was assessed and compared between these three fiber classes. Saline (0.9%) injections were used as vehicle controls.


The inflammatory mediators PgE2, serotonin, and bradykinin were dissolved in saline solution at concentration of 10-5M and kept frozen at −20°C until use. Capsaicin solutions (0.1%) were prepared in saline containing Tween 80 (LaMotte et al. 1991). For the injections 28-G needles and 0.3-ml syringes (Becton Dickinson) were used. Injections (20 μl) were given in the previously mapped innervation territory. Injection sites in one innervation territory were spaced by ≥1.5 cm. Pain sensations induced by the insertion of the needle and by the injection were assessed psychophysically. Subjects were asked to rate the sensation numerically on a scale from 0 to 10, in which 0 equaled “no pain” and 10 equaled “unbearable pain.” Maximum pain sensation induced by the injection was used for statistical analysis. Mechanical and heat thresholds at the injection sites were determined before and 5 min after the injection.

ACh (10% wt/vol) was dissolved in sterile water for each experimental day and kept refrigerated until use. ACh was applied by ionotophoresis using the same applicators described above for histamine, but for a longer duration (1 mA for 60 s = 60 mC).

Saline containing 0.1% capsaicin in Tween 80 (≤20 μl) (LaMotte et al. 1991) (i.e., a maximal dose of 20 μg) was injected intracutaneously inside the electrically mapped innervation territory. The injection was performed slowly (10 s) to allow the subject to rate continuously the increase of pain magnitude on a 10-point scale (0 = no pain, 10 = unbearable pain). Injection was stopped immediately when the subject reported a pain intensity of “5” to prevent losing the unit due to unintentional pain-provoked movements of the subject. Capsaicin was delivered from a 100-μl Hamilton syringe (Hamilton Bonaduz AG, Basel, Switzerland) with 1 μl partitions through a sterile nylon microfilter with 0.22-μm pores (Micron Separations Inc.), using a 27-G needle.

Data acquisition and analysis

C-unit responses to intracutaneous electrical stimulation were recorded on-line by a PC computer via an interface card (DAP, Microstar) using the SPIKE/SPIDI software package (Forster and Handwerker 1990). A suitable time segment of the recording following each electrical stimulus pulse was displayed and subsequent traces were written from top to bottom on the computer screen for on-line assessment of latency shifts of the activated C-units. Digital matched filtering was implemented to facilitate the tracking of the latency shifts (Hansson et al. 1998). In addition, the recordings were stored on hard disk for off-line analysis.

The amount of activation of C-units was documented by parameters derived from the marking response: the intensity and duration of the response was assessed by analyzing the number of traces in which the conduction delay of the respective unit was abruptly increased (activation periods, see Fig. 1). The period of injection or iontophoresis itself was excluded from this analysis, because activation by the needle or the current could not be differentiated from responses to the chemicals.

Fig. 1.

Maximum sensory ratings to intracutaneous application of histamine (n = 9), prostaglandin E2(PgE2; n = 9), serotonin (5HT;n = 11), acetylcholine (ACh; n = 11), bradykinin (Bk; n = 9), and capsaicin (Cap;n = 6) are shown for each subject. Itch and pain intensity were rated separately for each application. Ratings of each subject are shown interconnected with itch intensity represented by a filled circle (upper scale) and pain intensity by an open triangle (lower scale). Only for histamine itch ratings were significantly higher as compared with the other mediators. PgE2induced significantly less pain as compared with ACh, Bk and Capsaicin. Histamine induced significantly more itch than pain, whereas Bk and Cap provoked significantly more pain than itch (* P < 0.05, ** P < 0.01; Kruskal Wallis ANOVA, Mann WhitneyU test as post hoc test).


Intensity ratings and number of activation periods following chemical stimulation were evaluated by Kruskal–Wallis ANOVA and Mann–Whitney U tests as post hoc tests. Differences in conduction velocity and electrical threshold were calculated by ANOVA and Scheffé post hoc tests.




Application of various agents via intradermal microdialysis fibers provoked different intensities of itch and pain sensations. Maximum itch and pain ratings for each substance are shown in Fig. 1.


Histamine had the highest pruritic potency, significantly higher than each of the other mediators (P < 0.05; Kruskal Wallis ANOVA, Mann Whitney U test as post hoc test). The difference in itch intensity induced by PgE2, serotonin, and acetylcholine did not reach statistical significance (P> 0.2, Mann Whitney U test). Capsaicin did not induce itch sensations at all. Regarding the average itch responses one may put the “itching potency” of the tested substances in the following rank order: histamine ≫ PgE2 = serotonin = acetylcholine > bradykinin > capsaicin.


Capsaicin provoked the highest pain ratings, significantly higher than each of the other mediators (P < 0.05; Kruskal Wallis ANOVA, Mann Whitney U test as post hoc test). PgE2 induced pain ratings did not significantly differ from those provoked by histamine and serotonin but they were lower as compared with acetylcholine, bradykinin and capsaicin (P < 0.05, Mann Whitney U test) (see Fig.1). The tentative pain inducing potency of the tested substances on the basis of the average ratings is: capsaicin > acetylcholine = bradykinin > serotonin > histamine. Generally itch ratings diminished with increasing pain ratings and only 7 subjects reported a combination of itch and pain sensation (see Fig. 1).


Postinjection pain after bradykinin (2.5, 1–5; median, quartiles) and serotonin (2, 1–4) was significantly higher compared with saline (1.5, 1–2) (P < 0.05; Kruskal Wallis ANOVA, Mann WhitneyU test as post hoc test). Pain ratings after PgE2 injection (2, 1–3) did not significantly differ from that after saline and was significantly lower compared with bradykinin (P < 0.05; Kruskal Wallis ANOVA, Mann Whitney U test as post hoc test). Capsaicin injection provoked intense pain sensation (5, 4–6; median, quartiles) that by far exceeded the ratings for the other mediators.

Sample of C-nociceptors

89 mechano-responsive C-units (CMH), 52 mechano-insensitive histamine-negative (CMiHis-) and 24 mechano-insensitive histamine-positive C-units (CMiHis+) were studied. Conduction velocity of the CMH units was significantly higher as compared with the mechano-insensitive fibers (P < 0.001; ANOVA, Scheffé post hoc test). Within the mechano-insensitive group CMiHis+ had significantly lower conduction velocities than CMiHis- units (P = 0.01; ANOVA, Scheffé post hoc test) (Fig.2 A). Transcutaneous electrical thresholds and the activity dependent decrease in conduction velocity (“slowing”), did not differ significantly between the histamine positive and negative CMi units, whereas both parameters clearly differentiated between mechano-insensitive and mechano-responsive units (Fig. 2B,C) (Weidner et al. 1999).

Fig. 2.

Conduction velocity, electrical threshold and activity dependent slowing is depicted for polymodal nociceptors (CMH), mechano-insensitive histamine-positive (CMiHis+) and mechano-insensitive histamine-negative nociceptors (CMiHis-).

Responses of C-fibers


CMH and CMiHis- units were largely unresponsive to PgE2 injection, or showed only spurious activation. In contrast, most of CMiHis+ units were clearly activated and showed prolonged responses. A specimen of such a response is shown in Fig. 3.

Fig. 3.

Specimen of three chemical responses of a single mechano-insensitive nociceptor. The unit was activated by bradykinin (BK; A) and prostaglandin E2 (PgE2) injections (B) and histamine iontophoresis (C) in three different sites inside its innervation territory. Time of injection and iontophoresis is marked by open boxes. Responses to electrical stimulation at a frequency of 1/4Hz are shown in subsequent traces from top to bottom. Spikes of the unit under study are shown as filled boxes. All original action potential shapes of the unit are plotted (insets) for each panel. Bradykinin, PgE2 and histamine activated the unit with increasing effectivity as indicated by the irregular shifts of response latency (“marking”).

In the whole population of CMiHis+ units (as in the unit shown in Fig. 3) the median number of activation periods following PgE2 injection was lower as compared with histamine iontophoresis (P < 0.001, Mann WhitneyU test). PgE2 responses in CMiHis+ units were significantly larger as compared with bradykinin and capsaicin (P < 0.05, Kruskal Wallis ANOVA and Mann Whitney U test as post hoc test), but did not differ significantly from those induced by acetyl choline and serotonin. Saline injection did not activate any mechano-insensitive C unit (n = 7). Most of the mechano-sensitive nociceptors were activated during the injection, but only three out of fifteen units were activated for 1 or two traces after end of the injection (not shown).

In CMiHis+ units also activation by acetyl choline iontophoresis was significantly stronger than in CMH units (P < 0.05; Kruskal Wallis ANOVA, Mann WhitneyU test as post hoc test) and in histamine insensitive CMiHis- units (P < 0.05). No significant differences between the fiber classes were observed following injection of serotonin (Fig.4).

Fig. 4.

Number of activation periods induced by histamine, prostaglandin E2 (PgE2), serotonin, acetylcholine, bradykinin and capsaicin are shown for polymodal nociceptors (CMH), mechano-insensitive histamine-positive (CMiHis+) and mechano-insensitive histamine-negative nociceptors (CMiHis-). Note, that only histamine-positive nociceptors present lasting activation on PgE2 injection.

Bradykinin and capsaicin responses

Activation of nociceptors by bradykinin lasted shorter compared with activation by the above mediators. No significant differences between the nociceptor classes were observed. Injection of capsaicin activated virtually all nociceptors. The response pattern differed strikingly between the mechano-responsive and –insensitive group, with intense, but short-lasting activation in all the mechano-responsive units. Mechano-insensitive nociceptors showed longer lasting responses. A tendency for more activation periods were observed following injection in the histamine-negative subpopulation, but this difference was not significant (P = 0.2; Kruskal Wallis ANOVA, Mann Whitney U test as post hoc test) (Fig. 4).


Mechanical thresholds to probing with von Frey hairs were unchanged in CMH units following injections of PgE2 (30, 18–47 mM vs. 30, 14–32 mN) (median, quartiles; before vs. after injection), serotonin (30, 30–32 mM vs. 30, 12–32 mN), bradykinin (30, 14–65 mM vs. 30, 14–65 mN) and saline (30, 30–49 mM vs. 30, 12–46 mN). No originally mechano-insensitive fiber was sensitized to mechanical stimuli following PgE2.

Heat thresholds did not change significantly following injections of saline and serotonin. Bradykinin slightly lowered heat thresholds in CMH units from 41.0 ± 3.0°C to 40.5 ± 2.9°C (mean ± SD; n = 31), in CMiHis+ from 48.6 ± 3.5°C to 47.7 ± 1.4°C (n = 4), and from 48.1 ± 3.8°C to 48.0 ± 3.0°C (n = 7) in CMiHis- units. None of these differences was statistically significant. Two previously heat-insensitive, mechano-insensitive units were sensitized to heat by bradykinin (heat thresholds 48 and 50°C).

PgE2 injection slightly decreased heat thresholds in mechano-responsive units from 40.8 ± 3.1°C to 40.2 ± 2.4°C (mean ± SD; n = 28; P = 0.25,t-test). The drop in heat thresholds was by far more pronounced in CMiHis+ units, namely from 47.0 ± 1.2°C to 42.0 ± 1.6°C (n = 4,P < 0.01) and in histamine-negative mechano-insensitive units from 47.7 ± 2.6°C to 45.4 ± 2.3°C (n = 7, P < 0.05). Four initially heat insensitive mechano-insensitive units were sensitized to heat following PgE2 injection (heat thresholds 45, 47, 49 and 49°C) (Fig. 5).

Fig. 5.

Effects of prostaglandin E2 injection on heat threshold is shown for polymodal nociceptors (CMH), mechano-insensitive histamine-positive (CMiHis+) and mechano-insensitive histamine-negative nociceptors (CMiHis-). Heat thresholds of each unit before (control) and after injection (PgE2) are linked. For the 4 heat unresponsive fibers a dashed link is used.


The present results shed new light on the peripheral mechanisms of itch and pain sensations in healthy human skin. Mechano-insensitive nociceptors with a sustained response to histamine presented a graded response to pruritogenic substances such as prostanglandin E2 and serotonin, but were also excited by the algogenic substances bradykinin and capsaicin. In contrast, mechano-responsive C-fibers and histamine-insensitive CMi units were insensitive to PgE2 or showed spurious responses. They were, however, readily activated by algogenic substances.


Bradykinin did not sensitize C-nociceptors to mechanical stimuli, as could be expected from psychophysical investigations: intracutaneous injections of bradykinin did not induce mechanical hyperalgesia (Manning et al. 1991). However, in this in vivo investigation heat pain thresholds dropped by about 1°C in the subjects, which correlates well to animal data (Beck and Handwerker 1974). Interestingly, bradykinin-induced sensitization to heat is much more prominent under in vitro conditions (Jeftinija 1994; Khan et al. 1992;Koltzenburg et al. 1992; Liang et al. 2001; Mizumura et al. 1992; Rueff and Dray 1993), possibly because of lack of degrading enzymes.

Prostaglandin E2 induced clear heat sensitization, but no drop in mechanical thresholds or sensitization of mechano-insensitive nociceptors. These results are compatible with some in vitro data in rat skin (Lang et al. 1990). Recently, mechanical sensitization following PgE2 injection has been described in rat skin in vivo (Chen et al. 1999). The drop in mechanical threshold above saline control observed in this study was about 15%. As we used a series of calibrated v.Frey hairs with steps of more than 30% such a modest difference could not be detected in our study.

Responses of different classes of C-nociceptors and psychophysics

In our study chemical responsiveness was tested in various spots inside the innervation territories of nociceptors. Although there is a fairly pronounced heterogeneity of heat thresholds in mechano-insensitive units (Schmidt et al. 2002), chemical responsiveness was fairly homogenous; however, as there could be also spots inside the innervation territory which are insensitive to histamine, a negative histamine provocation does not prove that a unit is indeed histamine-insensitive in its entire receptive field thereby blurring our classification. However, our results show that only histamine-positive mechano-insensitive “itch” units showed strong and sustained responses to PgE2, whereas the other nociceptor classes were more or less unresponsive. Interestingly, also conventional mechano-heat sensitive nociceptors from rodents have been found largely unresponsive to PgE2(Lang et al. 1990; Mizumura et al. 1987;Rueff and Dray 1993). Activation of nociceptors by PgE2 has been reported only from nociceptive joint afferents in the cat (Schaible and Schmidt 1988;Schepelmann et al. 1992). The excitatory effect of PgE2 on histamine-positive mechano-insensitive nociceptors is therefore a novel finding, though it fits well with psychophysical observations that PgE2 applied by intradermal injection (Hägermark and Strandberg 1977) and by topic application in the conjunctivae (Woodward et al. 1995) causes itch. Taking into account the histamine sensitivity of these units, indirect activation via histamine released from mast cells has to be considered. Intradermal injection of PgE2 has been reported to induce only marginal whealing (Juhlin and Michaelsson 1969;Kingston and Greaves 1985), however, it provoked a small, albeit significant protein extravasation in other studies (Sabroe et al. 1997; Sciberras et al. 1987). Recently, dermal application of PgE2 via microdialysis has been combined with measurement of local protein extravasation and local blood flow (Neisius et al. 2002). In this study PgE2 did not increase protein extravasation, even at a concentration of 10-4M, but provoked a weak itch sensation and a pronounced vasodilation. In contrast, histamine provokes protein extravasation at lower concentrations as compared with the induction of itch (Lischetzki et al. 2001). Thus rather than being mediated by histamine release, the pruritic effect of PgE2 is most probably due to direct excitation of the CMihis+ units.

A direct comparison between psychophysics and nociceptor discharge is problematic as mediators were administered via different routes, i.e., iontophoresis and injection for electrophysiology and microdialysis for psychophysics. Microdialysis was chosen because it avoids the interaction of injection pain with the ensuing itch sensation. However, application of mediators via microdialysis catheters in the setting of microneurography is not feasible because of the time consuming procedure. Despite these limitations it appears reasonable to compare the response patterns of different nociceptor classes for a given mode of application.

Interaction of pain and itch

Although the responses in “itch units” reflected the pruritic potency of the mediators tested in this study, the strong activation of these units by capsaicin and bradykinin seems to contradict a specific role of these units in itch, since both substances are mainly algogenic and not pruritogenic. The ambiguity of “itch unit” discharge to pruritics and algogens may be solved by the central inhibition of itch by pain: it is common knowledge that scratching relieves itching. Thus it can be assumed that activity in mechano-sensitive nociceptors suppresses itch. To date there are many reports on itch suppression exerted by painful stimuli. These stimuli include electrical stimulation (Nilsson et al. 1997) or treatment with capsaicin (Brull et al. 1999). Recently, also the opposite effect, i.e., increasing of itch sensation by pain reduction has been clearly shown (Atanassoff et al. 1999). On a spinal level, opioids inhibit pain processing and thereby may provoke itch (Schmelz 2001). This mechanism is probably the basis for the antipruritic action of opioid antagonist like naloxone or naltrexone (Odou et al. 2001; Wolfhagen et al. 1997). However, it has to be stated that currently available data on central interactions of pain and itch are mainly based on psychophysics and therefore have to treated with caution.

The inhibition of itch by painful stimuli has to be taken into consideration when activity in “itch” units is correlated to the pruritic potency of the tested mediator (Schmelz 2001). Prostaglandin E2 exclusively excited “itch” nociceptors in our study, whereas acetylcholine activated a considerable number of nonitch nociceptors. Thus the pruritic effect of PgE2 can be explained by the activation of “itch” units and simultaneous the absence of activity in itch suppressing nociceptors. Conversely, the activation of “itch” units by acetylcholine does not provoke itch because simultaneously activated nonitch nociceptors suppress the itch and the perceived sensation is pain. Accordingly, capsaicin that readily activates itch and nonitch units, provokes strong pain and no itch sensation. Although our data support this concept, experimental proof for it can only be obtained in recordings from second order neurons. Noteworthy, acetylcholine provokes itch instead of pain in patients suffering from atopic dermatitis (Groene et al. 2001; Vogelgsang et al. 1995) indicating, that pain induced inhibition of itch might be compromised in these patients.

Activation of histamine-positive chemonociceptors by PgE2 is inline with the pruritogenic effects of prostaglandins. As PgE2 is only a weak pruritic it is not clear, whether positive PgE2 responses can be regarded characteristic for “itch-units” and thus could be useful to separate between pain and itch fibers. However, the histamine-positive fibers might not be classified as “itch specific”, because they are also excited by pure algogens and thus might rather be termed “itch selective” (McMahon and Koltzenburg 1992). Further support for the “specificity”, or rather “selectivity theory” comes from second order neurons in the cat that have recently been recorded. These neurons cannot be excited by mechanical stimulation, but are activated by histamine iontophoresis with a similar time course as compared with the primary afferents (Andrew and Craig 2001). Interestingly, conduction velocity in the primary afferents innervating these neurons was particularly low. In addition, their thalamic projection differed significantly from the conventional nociceptor specific and wide dynamic range neurons. From these results it appears that finally there is strong evidence to suggest that a subpopulation of nociceptors exist, which is responsible for the induction of itch and conveys their information in an itch specific central pathway.

In summary, the pruritic potency of inflammatory mediators is characterized by their ability to activate histamine positive mechano-insensitive C-nociceptors. However, concomitant activation of mechano-sensitive and mechano-insensitive histamine negative nociceptors will decrease the itch. Therefore the itch sensation is based on both, activity in the “itch-pathway” and absence of activity in the “pain-pathway”.


This work was supported by a Max Planck Award to H. E. Torebjörk, by the Deutsche Forschungsgemeinschaft, SFB 353, by the Swedish Medical Council, Project 5206 and a grant to R. Schmidt from the Swedish Foundation for Brain Research.


  • Address for reprint requests: M. Schmelz, Dept. Anesthesiology and Intensive Care Medicine, Faculty of Clinical Medicine Mannheim, University of Heidelberg, Theodor-Kutzer Ufer 1–3, 61087 Mannheim, Germany (E-mail:martin.schmelz{at}


View Abstract