|
|
| |||||||||||||||||||||||||||||||
Background Rickettsialpox is caused by Rickettsia akari, which is transmitted from rodents to humans by bloodsucking mites. The initial skin lesion forms an eschar and is followed by the development of fever, malaise, myalgia, and 5 to 40 maculopapules and papulovesicles. The disease, which responds to tetracycline, can be mistaken for chickenpox. The diagnosis has been based on an increase in serum antibody titers against R. akari over a period of three to eight weeks. We discuss a more rapid technique that uses direct immunofluorescence to identify R. akari in paraffin-embedded tissue, and we describe the histopathological findings of lesional skin.
Methods We studied 13 patients (age, 11 months to 58 years) who were seen at Lincoln Hospital in New York City from 1980 to 1989 and were suspected of having rickettsialpox. In nine patients serum samples were obtained during the acute and convalescent phases of the illness for indirect fluorescent-antibody testing. Punch-biopsy specimens of skin lesions were examined by microscopy and by direct fluorescent-antibody testing with an anti-R. rickettsii globulin conjugated with fluorescein isothiocyanate.
Results The diagnosis was confirmed in all 13 patients by indirect or direct fluorescent-antibody techniques. Direct fluorescent-antibody testing of eschars from seven patients was positive in five patients, but negative in two patients who had serologically confirmed rickettsialpox. In contrast, direct fluorescent-antibody testing of papulovesicles from nine patients was positive in only one patient. Histopathological analysis of the eschars revealed extensive necrosis and inflammation. In biopsy specimens of papulovesicles, dermal edema, subepidermal vesicles, and vascular changes were present.
Conclusions The combination of direct fluorescent-antibody testing of an eschar from the presumed site of inoculation and histopathological examination of papulovesicles for distinctive features represents an improved method of diagnosing rickettsialpox.
Over 800 cases of rickettsialpox have been reported, but the true incidence is probably higher6. More than half these cases were reported during the first three years after the initial description of the disease7. Rickettsialpox has been reported infrequently in recent years. We are aware of published reports of only six cases in the United States during the 1980s8,9. The disease is probably underrecognized as well as underreported.
To date, the diagnosis of rickettsialpox has been based on an increase in serum antibodies against R. akari during convalescence. There are drawbacks to this approach to the diagnosis of rickettsialpox. Since the patients have already recovered, they may not return during the convalescent phase to have another serum sample drawn. There is also a delay of several weeks between the onset of the illness and the increase in the antibody titer.
The use of direct immunofluorescence to identify another member of the spotted-fever group, R. rickettsii, in fixed, paraffin-embedded tissue has been described10. The technique employs an anti-R. rickettsii globulin labeled with fluorescein isothiocyanate that has a high degree of cross-reactivity with R. akari antigens11. It has been used experimentally to identify the causative agent of rickettsialpox in tissue embedded in paraffin blocks (Figure 1). The method has potential advantages over serologic testing. It requires only one visit by the patient and yields results within a few days.
|
We conducted a study of patients suspected of having rickettsialpox at Lincoln Hospital, Bronx, New York, from 1980 through 1989 to assess the clinical usefulness of direct immunofluorescence as a diagnostic technique and to determine the characteristic histopathological appearance of the skin lesions.
Methods
We studied 13 patients who were seen at Lincoln Hospital from 1980 through 1989 with clinical features consistent with rickettsialpox. Serum samples were obtained from nine patients during the acute and convalescent phases of the illness for indirect fluorescent-antibody testing against antigens of the spotted-fever group. Testing was performed at the Centers for Disease Control and Prevention (CDC) with a previously described technique13.
Eighteen biopsies of papulovesicles and eschars from 11 untreated patients were performed with a 4-mm punch. The specimens were stained with hematoxylin and eosin and Giemsa stain and assessed for epidermal and dermal necrosis; spongiosis, vacuolar alteration, dermal edema, and vesicle formation; the distribution, composition, and intensity of the inflammatory infiltrate; and vascular changes. Seventeen tissue samples embedded in paraffin blocks were available from 10 of the patients for direct fluorescent-antibody testing, which was performed at the CDC with an anti-R. rickettsii globulin labeled with fluorescein isothiocyanate10.
Results
Thirteen patients suspected of having rickettsialpox were seen at Lincoln Hospital from 1980 through 1989. The patients ranged in age from 11 months to 58 years (Table 1). The chief symptom at presentation was a papulovesicular rash in 12 patients; 1 patient initially came to the hospital screening clinic with symptoms including fever, headache, and sore throat. A rash subsequently developed, and he returned to the hospital and was referred to the dermatology department. All patients had an eschar and papulovesicles at presentation to our clinic (Figure 2 and Figure 3); the majority also had red maculopapules. At the initial examination, the papulovesicular eruption had been present for two to seven days. The lesions were on the face, trunk, and extremities and ranged in number from 5 to approximately 30. Only a few patients or their parents knew when the eschar had developed; most indicated that it had been present for four to seven days before the onset of the papulovesicular rash.
|
|
|
Laboratory studies pertinent to rickettsialpox were performed in three patients. Two had normal white-cell counts; the third had leukopenia (2800 cells per cubic millimeter) with a leftward shift (31 percent band forms). The two patients who underwent urinalysis had normal results.
Nearly all the patients were treated with tetracycline starting the day they came to our clinic. The fever resolved within 24 hours, and all had recovered without complications when seen at a follow-up visit nine days to four weeks after the onset of the papulovesicular eruption. Patient 4, an 11-month-old infant, was seen three days after the onset of papulovesicles and was treated with erythromycin (Figure 3). She was reported to be afebrile within 24 hours and had only a few red papules at follow-up six days later. Patient 13 presented five days after the onset of the papulovesicular rash, was afebrile, and received no treatment. Her skin lesions had resolved at follow-up four days later.
Antibody titers against R. akari were four or more times higher in the serum samples obtained during convalescence than in those obtained during the acute phase of the illness in all nine patients for whom serial samples were available (Table 1). Titers against R. rickettsii and R. conorii also increased between the acute and convalescent phases of the illness; the increase was typically less than that observed for R. akari.
Direct fluorescent-antibody testing with an anti-R. rickettsii conjugate was performed on paraffin-embedded tissue from 10 patients, 6 of whom had serologic confirmation of the diagnosis of rickettsialpox. Serum samples from the acute or convalescent phase of the illness were not available for the other four patients. Direct fluorescent-antibody testing was positive in six samples from five patients (Table 1). Five of the positive results were obtained in eschars. The one papulovesicle in which a positive result was obtained was from a patient whose eschar was also positive. Of the eschar-biopsy specimens from three patients with serologic confirmation of rickettsialpox, only one was positive by direct fluorescent-antibody testing. None of the papulovesicles from the six patients with serologically confirmed rickettsialpox were positive.
A number of histopathological changes were seen in the skin lesions. In eschars, there was epidermal and dermal necrosis, which included hair follicles and eccrine glands, as well as focal necrosis of the panniculus (Figure 4). An extensive lymphohistiocytic infiltrate was present around blood vessels and adnexa and in the septa and lobules of the panniculus. Neutrophils were associated with necrotic areas. A spectrum of changes was seen in blood vessels, including endothelial swelling, necrotic and thrombosed vessels in the superficial dermis, and lymphocytic vasculitis in the middle and deep portions of the dermis. Medium-caliber vessels deep in the dermis and in the subcutis had hyperplastic endothelia containing small deposits of fibrin and were infiltrated by lymphocytes and a few epithelioid histiocytes. Organisms were not identified in sections stained with Giemsa stain.
|
|
Rickettsialpox typically has an incubation period of 9 to 14 days8. The initial lesion of the disease is a papulovesicle that develops at the site of inoculation and subsequently forms an eschar8,14,15,16. Local lymph nodes draining the primary lesion are usually enlarged and tender. Approximately one week after the development of the initial lesion, fever begins abruptly and is usually high (temperature, 39 to 40 °C). The usual accompanying symptoms are headache, malaise, and myalgia, most prominently involving the back. About half of patients have drenching sweats and shaking chills. Less frequently there is photophobia, conjunctival injection, rhinorrhea, sore throat, cough, anorexia, generalized lymphadenopathy, and nausea or vomiting. Usually within three days of the onset of fever, and no later than seven days afterward, red macules, papules, and papulovesicles develop over the body. The lesions are most often sparse, numbering roughly 20 to 40 in previously published reports, and resolve in approximately one week. There may be an associated enanthem of the oral cavity16.
Laboratory findings in rickettsialpox include leukopenia in approximately 75 percent of cases. Although the differential count is usually normal, there is a relative lymphocytosis in some patients and a leftward shift in others. The erythrocyte sedimentation rate may be slightly elevated, and some patients have a transient febrile albuminuria14. The Weil-Felix test is negative in most patients, but positive agglutination reactions have been reported in some, usually at a low titer9,15,16.
Antibiotic therapy ameliorates the course of rickettsialpox8,16,17. Tetracycline at a dose of 250 mg orally every six hours for two to five days is the treatment of choice8. Even without therapy, all patients described to date recovered without complications. Rickettsialpox has yet to be reported in an immunocompromised patient.
The clinical presentations in our 13 patients were consistent with those described in prior reports, although none of our patients had an enanthem, generalized lymphadenopathy, or gastrointestinal symptoms. Two of our patients had chest pain, which is not a typical feature of the illness. We found that fever usually began no more than 24 hours before the development of the papulovesicular eruption, which consisted of as few as 5 lesions or as many as 30. With one exception, the patients only sought medical care after papulovesicles were fully developed. Although all our patients had eschars, the lesions were apparently ignored or went unnoticed.
In patients with rickettsialpox, serum antibodies against R. akari develop that can be detected by complement fixation or indirect fluorescent-antibody techniques. Antibody titers peak three to four weeks after the onset of systemic symptoms in untreated patients and up to six to eight weeks after symptoms develop in treated patients8. The diagnosis of rickettsialpox has been based on the assessment of the antibody titer in serum samples obtained during the acute and convalescent phases of the illness. A fourfold or greater increase in titer is considered diagnostic. It takes several weeks to establish the diagnosis by serologic testing.
Rickettsialpox was confirmed in all nine of our patients for whom serial serum samples were available. Titers against R. akari rose between the acute and convalescent phases of the illness, as did titers against R. rickettsii and R. conorii. The serologic cross-reactivity between R. akari and these other members of the spotted-fever group13,18 does not present a problem in confirming the diagnosis of rickettsialpox, since its clinical presentation and course are usually distinct from those of Rocky Mountain spotted fever and the Eastern Hemisphere rickettsioses. If there is any doubt, a serologic cross-absorption technique for the specific diagnosis of rickettsialpox is available8.
In hopes of identifying a more convenient and rapid means of diagnosing rickettsialpox, we performed direct fluorescent-antibody testing with anti-R. rickettsii globulin conjugated with fluorescein isothiocyanate on biopsy specimens from 10 of our patients. This test was positive in six biopsy specimens from five of the patients (Table 1). In all five patients, a positive result was obtained in an eschar. In one of the five patients, a papulovesicle was also positive. Given that the eschar is the presumed site of inoculation of R. akari, it seems likely that this lesion would harbor the highest concentration of the organism. The fact that R. akari was identified in only one of the papulovesicles suggests that there are few organisms in this lesion. Sampling error might also lead to negative results. Nevertheless, the finding of R. akari in at least one papulovesicle suggests that the organism is directly responsible for the papulovesicular eruption of rickettsialpox. The exanthem probably does not represent an immunologic response to the rickettsiae in the eschar.
Two patients in whom the diagnosis of rickettsialpox was confirmed by indirect fluorescent-antibody testing had negative results on direct fluorescent-antibody testing of eschars. Although the exact onset of these two eschars is unknown, they might have been old lesions from which R. akari antigen had already been cleared. An alternative explanation relates to the anti-R. rickettsii conjugate used for testing. Although conjugates against the individual members of the spotted-fever group are all cross-reactive, homologous reagents often stain tissue at higher titers than heterologous reagents11. Therefore, a tissue sample that stains weakly for an anti-R. akari conjugate might not react at all with a conjugate directed against R. rickettsii.
Overall, we found that direct fluorescent-antibody testing of paraffin-embedded tissue is a useful diagnostic technique. Had the test not been used, the diagnosis of rickettsialpox would have been unconfirmed in 4 of our 13 patients. The test requires only one visit by the patient and can yield results within a few days after biopsy. Eschars should be tested in preference to papulovesicles, since only one of the papulovesicles was positive in our patients.
There are published descriptions of the histopathological appearance of eight skin-biopsy specimens from patients with rickettsialpox8,12; six were maculopapules, and two were papulovesicles. Many of the findings described in prior reports were similar to those we observed. These included a vacuolar alteration, a mononuclear infiltrate with an occasional mixture of neutrophils and eosinophils, and blood vessels with prominent endothelial cells and fibrin thrombi. We could find no previous mention of lymphocytic vasculitis, which was prominent in many of our biopsy specimens.
The early literature on rickettsialpox describes intraepidermal formation of vesicles12. A more recent paper suggested that the vesicles are subepidermal8. We found that in well-developed papulovesicles there is frank separation at the dermal-epidermal junction and, thus, subepidermal formation of vesicles.
The histopathological appearance of rickettsialpox is distinctive. Eschars are characterized by extensive necrosis and inflammation involving superficial and deep portions of the dermis and, frequently, the panniculus. Papulovesicles are characterized by extensive superficial edema of the dermis and vesicle formation. Blood vessels display a continuum of changes from edematous walls with a lymphohistiocytic infiltrate to lymphocytic vasculitis with fibrin in lumina and walls. The histopathological appearance of papulovesicles is virtually diagnostic of the illness when found in combination with the typical clinical presentation.
Rickettsialpox has been recognized in Russia, South Africa, and Korea and in several parts of the United States,6 including New York, Massachusetts, Connecticut, Pennsylvania, Ohio, and Utah7,19. The disease is most likely to occur in crowded areas with mouse-infested housing20. In addition, it has been suggested that epidemics of rickettsialpox may only occur in an environment in which the murine host has become less attractive to L. sanguineus21. Humans may then become a secondary target for the mite and may thus be susceptible to inoculation with R. akari. Lymphocytic choriomeningitis, which has been found in mice during outbreaks of rickettsialpox, may be the catalyst for the occurrence of epidemics of the disease. This infection and its associated findings, such as changes in body temperature, could cause L. sanguineus to abandon its natural host.
Although there have been very few reports of rickettsialpox in recent years, conditions favoring outbreaks of the illness are still present in large cities. Some cases may go unreported. We suspect that some patients are misdiagnosed as having chickenpox, the main alternative in the differential diagnosis of the illness in the United States. Rickettsialpox can be distinguished clinically from chickenpox by the presence of an eschar, the failure of the lesions to appear in crops, and the relative sparsity of lesions. In addition, the mature lesion in rickettsialpox is a papulovesicle with a prominent papular component; in chickenpox, it is a vesicle. The typical histopathological appearance of rickettsialpox, as described in this paper, is quite different from that of chickenpox.
In summary, the diagnosis of rickettsialpox should be considered in any patient with fever and a papulovesicular eruption. Direct immunofluorescence testing of eschars and routine microscopy are useful diagnostic tools in the context of the characteristic clinical presentation of rickettsialpox.
Source Information
From the Departments of Dermatology and Pathology, New York Medical College and Lincoln Hospital Center, New York (E.M.K., W.K.S, H.L., J.L., K.S.); Dermatopathology Associates of New York, New Rochelle (W.K.S.); and the Centers for Disease Control and Prevention, Atlanta (C.R.).
Address reprint requests to Dr. Szaniawski at Dermatopathology Associates of New York, 91 Weyman Ave., New Rochelle, NY 10805.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
