Keiko Oka, M.D., Department of Dermatology Koishikawa Tokyo Hospital, 4-45-16 Ohtsuka, Bunkyo-ku, Tokyo 112-0012, Japan. Email: firstname.lastname@example.org
Eruptive pseudoangiomatosis is a skin eruption characterized by millet-sized erythema with an anemic halo appearing on exposed body areas. Insect bites, particularly mosquito bites, have been reported as one of the causes of eruptive pseudoangiomatosis. We experienced two cases of eruptive pseudoangiomatosis and the eruption was seen on the face and upper extremities of two women aged 48 and 77 years old. The two cases consented to be experimentally bitten by Culex pipiens mosquitoes and Aedes albopictus to determine if eruptive pseudoangiomatosis could be experimentally elicited by these mosquitoes. Our results showed that several minutes after a C. pipiens mosquito bite, an erythematous spot appeared on the bite site, followed by the formation of an anemic halo surrounding the erythema in 30 min; a successful reproduction of eruptive pseudoangiomatosis. The erythema lasted for more than a week and was not accompanied by any pruritus. With A. albopictus, we were able to reproduce a milder eruptive pseudoangiomatosis eruption: in case 1, a smaller erythematous spot with an ill-defined halo which disappeared within 1 week; and in case 2, an immediate response consisting of a wheal and erythema but not eruptive pseudoangiomatosis. We demonstrated that eruptive pseudoangiomatosis was the response manifested in individuals who normally did not demonstrate any immediate or delayed reaction to insect bites; and the typical eruptive pseudoangiomatosis eruption was elicited by C. pipiens mosquito bites. However, the mechanism resulting in the manifestation of eruptive pseudoangiomatosis is still unknown.
Eruptive pseudoangiomatosis (EP)1–8 is an entity defined by the following clinical features: (i) multiple millet-sized asymptomatic erythematous spots with an anemic halo distributed on exposed skin; (ii) the anemic halo may not appear with the spots on the face; (iii) the spots do not coalesce, and in the case of affected adults,3–8 (iv) systemic symptoms are absent; and (v) the eruption lasts for more than 1 week. This disease entity was first reported by Higuchi9 in 1943 as erythema punctatum, and thereafter a series of reports from Japan were published.10–13 As Ban et al.12 indicated, EP and erythema punctatum are similar disease entities. The more commonly-used term EP shall be used in this report.
The etiology of EP is unknown. Ohara et al.10 and Ban et al.12 reported that mosquito bites, especially by Culex pipiens mosquitoes were the cause of EP, and Restano et al.6 reported on EP after mosquito bites and flea bites, so insect bites are one of the causes of EP. However, Cherry,1 who first described the pediatric cases, suggested a viral etiology which was supported by several other authors,2–5,7 while another report suggested drug eruption11 as the cause.
The mosquito bite reaction is a cutaneous response to the saliva secreted by the mosquito just before it draws blood from the human skin.14 The mosquito bite reaction consists of an immediate skin reaction which is characterized by pruritus that is felt immediately after the bite, followed by a wheal and/or erythema appearing within several minutes; and a delayed skin reaction characterized by a pruritic papule and/or erythema which begins several hours after the bite and peaks around 24–48 h after the bite. Mosquito bite reactions differ from one individual to another, and the difference is due to the intensity of the immediate and delayed reactions in each individual. Mellanby15 and Heilesen16 demonstrated that repeated exposure to mosquito bites altered the bite reactions in humans from stage I to stage V (Table 1). If EP is a mosquito bite reaction, it should correspond to a stage V reaction. In a previous study,17 we performed experimental bite tests on the forearm of 162 subjects using Aedes albopictus, a mosquito indigenous to Japan, and found that 10 individuals belonged to Stage V. However, none of them demonstrated the symptoms of EP. Previous reports suggested that C. pipiens pallens was the mosquito species responsible for EP,10,12 but we suspected that A. albopictus was equally capable of eliciting EP. In order to confirm our suspicions, we performed experimental bite tests on the two clinical cases and five stage V volunteer subjects who had participated in our previous study by exposing them to A. albopictus. This time, the subjects were exposed to multiple bites and their response was observed for more than 1 week.
Table 1. Stage definition: mosquito bite reactions in an individual change from stages I–V depending on frequency of exposure15–17
Immediate reaction: the average diameter of the wheal and erythema 30 min later. Negative reaction (−): wheal <4 mm and erythema <4 mm. Positive (+): wheal >5 mm and/or erythema >5 mm. Delayed reaction: the average diameter of the papule and erythema 24 h later. Negative: erythema <4 mm and papule <3 mm. Positive: erythema >5 mm and/or papule >4 mm.
Case 1 was a 48-year-old Japanese woman. On the morning of 15 August 2004 (mid-summer in Japan), she woke up to find numerous millet-sized erythematous spots distributed on her upper limbs and lower eyelids. The lesions were asymptomatic, but she noticed that the spots on her upper limbs had an anemic halo surrounding them. The lesions persisted, which prompted her to consult our dermatology outpatient clinic 3 days later. The woman admitted that she had spent the night at her parents’ home in Kisarazu, a town in the countryside of Japan. She had worn a pair of short-sleeved pajamas to bed but did not apply any insect repellant to her skin before going to bed. The patient admitted that, of late, she had hardly noticed any mosquito bites on herself although she used to receive bites often in the past. We continued our clinical observation on her and found that the lesions lasted for more than 1 week.
Case 2 was a 77-year-old Japanese woman. On 24 July 2007, she noticed that numerous erythematous spots surrounded by an anemic halo had appeared on her arms (Fig. 1) and on her cheeks. The lesions were asymptomatic. Apparently she was not in the habit of applying insect repellant. She consulted our outpatient clinic on the fourth day, and a skin biopsy was performed on one of the spots on her left upper arm. Histopathology showed capillary dilatation and a lymphocytic perivascular infiltration in the upper dermis with no significant changes to the epidermis (Fig. 2). The infiltrate cells were proportionately CD4 and CD8 cells. The findings were consistent with non-specific inflammatory reaction. The eruption took a natural course of approximately 3 weeks to resolve.
Mosquito bite experiment
Cases 1 and 2, and five volunteer subjects who had participated in our previous study (two women aged 46 and 61 and three men aged 64, 74 and 81 years) who were classified as stage V to A. albopictus bites, and five new healthy subjects as control (two women aged 27 and 56, and three men aged 39, 40 and 62 years) participated in the experimental mosquito bite tests. The experiment was approved by the Ethics Committee of the Koishikawa Tokyo Hospital, and written informed consent was first obtained from all participants. The mosquitoes used were C. pipiens mosquitoes (C. pipiens pallens, C. pipiens quinquefasciatus) and A. albopictus which had been bred successively at the National Institute of Infectious Diseases (NIID). Only female mosquitoes 3 days and older were used. The mosquitoes were free of any virus or pathogens. Although C. pipiens quinquefasciatus is only naturally distributed in the southern parts of Japan like Okinawa and Amami and not in the Kanto region where the experiment was conducted, we included C. pipiens quinquefasciatus as one of the mosquitoes in the bite tests because C. pipiens mosquitoes are morphologically very similar; and while C. pipiens pallens only feed during the night, C. pipiens quinquefasciatus that were bred in NIID would feed even during the day.
In the single bite experiment, one female mosquito in a tube was allowed to bite and feed on the forearm of the subject and observations were made immediately after the mosquito had removed itself from the skin. In the multiple bite experiment, the subject would insert a forearm into a cage containing 50 female mosquitoes and exposed the forearm to bites for 5 min. Observations of bite reactions would begin immediately after the subject had withdrawn the forearm from the cage. Each subject was confirmed to have had at least 20 bites or more.
Immediate and delayed reactions on the bite sites were observed and evaluated according to the criteria in Table 1.17 The sites were observed for up to 1 week.
For positive control, a prick test was performed on all subjects using 1% histamine diluted in distilled water. A wheal between 5 and 8 mm was confirmed after 15 min.
Results of multiple bite experiment with C. pipiens mosquitoes
Several minutes after case 1 was exposed to C. pipiens quinquefasciatus, an erythematous reaction with an ill-defined halo appeared on the spots bitten by the mosquitoes. The reaction lasted several days, gradually subsided and completely disappeared 7 days later. This could be considered a successful reproduction of EP, minor type. The results of all mosquito bite tests are summarized in Table 2.
Table 2. Results of mosquito bite tests
Volunteers, five subjects who were classified as stage V during a previous bite test using A. albopictus. Control, five normal healthy subjects. *Minor form.
Culex pipiens mosquitoes
Five subjects in stage IV
One subject in stage III, two subjects in stage IV, two subjects not tested
We designed and performed a further test whereby case 1 was exposed to numerous C. pipiens pallens, a species known to bite and feed at night. Several minutes after the subject was exposed to C. pipiens pallens on the right forearm, erythematous spots started to appear. The spots displayed an anemic halo which gradually became more distinct. When the spots were evaluated for immediate reaction 30 min later, the erythematous spots on the forearm were 3 mm in size and surrounded by 6-mm halos. They remained unchanged when they were evaluated 24 h later (Fig. 3) and persisted relatively unchanged for more than 1 week. EP was successfully reproduced in case 1 after exposure to C. pipiens pallens.
Case 2 was exposed to multiple bites by C. pipiens quinquefasciatus, and the subject responded immediately with 2-mm sized erythematous spots surrounded by 7-mm halos. The spots remained visible when observed 24 h, 5 and 14 days later. Again, EP was successfully reproduced in case 2 after exposure to C. pipiens quinquefasciatus.
When the new five healthy volunteers were given separate test bites with two species of C. pipiens mosquitoes, we discovered that all of them showed stage IV responses to the two species of mosquitoes and there was no significant variation in the reaction to the two species.
Results of single bite experiment with A. albopictus
In case 1, an erythema approximately 2 mm in size appeared, gradually subsided and disappeared in 5 days. This could be considered a successful reproduction of EP minor type. Case 2 exhibited a positive immediate reaction (26-mm erythema with 7-mm wheal) but delayed reaction was negative (2-mm erythema).
Results of multiple bite experiment with A. albopictus on five volunteers
Five stage V subjects who had participated in our previous study reacted to multiple bites with responses similar to a previous single bite test: 30 min after exposure, they exhibited 2–3-mm erythematous spots and the spots remained relatively unchanged 24 h later (Fig. 4). The lesions were not accompanied by any wheals or pruritus. There were no distinct halos and the lesions disappeared after 1 week. The responses were considered successful reproductions of EP, minor type.
In this study, we found that EP patients and stage V individuals who normally show no response to mosquito bites would manifest symptoms of EP several minutes after being bitten by C. pipiens pallens or A. albopictus, and the response would require several days to disappear. We thought that none of the volunteers who were given test bites by A. albopictus in our previous study17 manifested EP, but careful observation during this study revealed that stage V volunteers demonstrated a minor type of EP. A. albopictus bites elicited millet-sized erythematous spots which were smaller and had less well-defined halos than those by C. pipiens mosquitoes. The spots also took a shorter time to disappear. In comparison, C. pipiens mosquitoes elicited typical EP lesions. It was not uncommon for the same individual to demonstrate different stage reactions to different mosquito species. However, the difference in the bite responses of stage V individuals to A. albopictus and C. pipiens mosquitoes is unclear.
How do we interpret the mechanism of mosquito bite responses? There were arguments in some reports that EP was associated with mosquito or flea bites. Some authors6,10 considered EP as a unique response to mosquito or flea bites, while others12 strongly argued that it was due to viruses transmitted by insects. However, because the mosquitoes used in our bite experiments were pathogen-free, and the eruption only appeared on the spot bitten by the mosquito, the reaction can only be considered the action of mosquito saliva and not a result of viral transmission.
It has been established that the human response to mosquito bites progresses from stage I to stage V with repeated exposure.15,16 Based on the assumption that the frequency of exposure to mosquito bite increases with age, the authors previously performed an experimental study on 162 volunteers ranging in age from 1–68 years and were able to classify the responses of these individuals to A. albopictus mosquito bites into five different stages. This study confirmed that the mosquito bite response of an individual progresses from stage II to stage V with increasing age.17 Using immunological tests to measure the levels of mosquito saliva-specific immunoglobulin E antibodies and lymphocyte stimulation tests (LST), the authors further demonstrated that the immediate reaction which appeared soon after a mosquito bite was an immediate-type allergy response and the delayed reaction was a delayed-type allergy response.18 The authors discovered that individuals who did not demonstrate any immediate or delayed responses were not in the unsensitized stage but rather in the desensitized stage, that is, stage V. Many individuals who were classified as stage V were either elderly subjects or subjects who admitted to have been frequently exposed to mosquitoes in the past.
Eruptive pseudoangiomatosis was more frequently seen among less active individuals such as the elderly or hospitalized patients, and in mass outbreaks.4,5 Many case reports were seen from the Mediterranean area,3 Far East Asia,7–13 that is, from areas where C. pipiens mosquitoes and other insects were said to inhabit naturally.19 It is therefore very likely that many of these EP cases were actually bite responses to C. pipiens mosquitoes and other insects.
The following is our explanation of why C. pipiens mosquitoes were not easily recognized as the cause of EP. A mosquito may bite on several skin sites of the host before drawing blood; and each time the mosquito bites, even though no blood is drawn, a very small amount of mosquito saliva is injected into the skin. A. albopictus usually bite during the day, and therefore are more easily noticed, whereas C. pipiens mosquitoes hide indoors and only actively bite and draw blood during nighttime. It is likely that because a stage V individual does not experience any pruritus when bitten, the individual only discovers that he/she has been bitten when EP appears on the exposed areas of the body. Even though stage V was defined as the stage without any immediate or delayed response, it does not mean that these individuals do not display any visible skin reaction at all. There was usually a small asymptomatic erythematous spot at the bite site, and the spot usually took several days to disappear. However, these reactions were not consistent with typical allergic bite reactions. Whether this response was due to the incomplete desensitization of the individual or a pharmacological response to mosquito saliva is unclear.
Mosquito saliva hardly contains any histamine.20 Saliva consists of components which act on platelet aggregation inhibitors; some of these components are species-specific and others, like apyrase, are common to all species. These substances may have been possible allergens that could elicit the immediate-type allergic response.21 The extent of desensitization may also vary from one mosquito species to another. It is also possible that EP may have been caused by the difference in pharmacological effects due to a variation in the proportion of these salivary components in the different species of mosquitoes. This is an area for future studies.