Dermatopathology Human orf and milkers' nodule: A clinicopathologic study R. W. Groves, MRCP,a E. Wilson-Jones, FRCP, FRCPath,b and D. M. MacDonald, FRcpa London, England We report the clinical and histopathologic features of 17 patients with orf or milkers' nodule infection. The majority were male, 12 to 65 years of age, and gave a history of contact with farm animals. Most lesions affected the hands or arms, ranged in size from 1 to 3 em, and occurred on average 3 weeks after presumed exposure. On low-power examination the epidermis showed endophytic strandlike proliferations and the dermal papillae were distended by intense edema. There was massive capillary proliferation and dilation and a dense inflammatory infiltrate. High-power examination revealed epidermal viral cytopathic changes with inclusion bodies, clumping of keratohyalin, and cytoplasmic vacuolation that had a distinctive "spongifarm" appearance within follicular structures. We conclude that orf and milkers' nodule infection have distinctive histopathologic features, and, in contrast to some previous reports, viral changes may frequently be found. (J AM ACAD DERMATOL 1991;25:706-11.) Orf is a viral illness endemic in sheep and goats. The causative organism belongs to the parapox subgroup of poxviruses, is oval, measures 260 X 160 nm, l and has a characteristic spiral coat. It is morphologically indistinguishable from the organism that causes milkers' nodule,2 although the two can be differentiated by subtle differences in tissue culture. 3 Because of the clinical similarity of the two diseases, the collective termfarmyard-pox has been proposed. 4 The disease may be transmitted to human beings either by contact with infected animals, either alive or dead, or from contaminated objects; the organism is resistant to heat, cold, and drying and may persist for some time on fences or feeding troughs,5 from which infection can subsequently be acquired. In humans, orf and milkers' nodule infection are common, 6 self-limiting, and probably easily recognized by those working with sheep and cattle, in which infection is most likely. We suspect that this knowledge of the natural history of the disease means that many patients do not seek medical attention, and even fewer have biopsies. Although the pathologic characteristics of orf and milkers' nodules have been reviewed,5, 7 certain imFrom the Laboratory of Applied Dermatopathology, United Medical and Dental Schools, Guy's Campus, Guy's Hospital"; and the Department of Histopathology, The Institute of Dermatology, 5t Thomas' HospitaJ.b Reprint requests: D. M. MacDonald, Department of Dermatology, Guy's Hospital, St Thomas' St., London SEI 9RT, U,K.
16/1/30377
706
portant aspects (e.g., the presence or absence of inclusion bodies) remain controversia1. 8 To define accurately the histopathology and to resolve some of these controversies, we have undertaken a detailed histopathologic study of orf and milkers' nodule infection.
MATERIAL AND METHODS Histopathologic material. Dermatology departments throughout the United Kingdom were asked to submit specimens diagnosed as orf or milkers' nodules. Twentysix specimens were received, in 17 of which adequate tissue was available for study, and the diagnosis was considered unequivocal on the basis ofclinical data (history, examination, contact with sheep or cattle) or ultrastructural demonstration of viral particles. Hematoxylin-and-eosin sections were examined in all instances and in selected cases reticulin stains and phloxine-tartrazine and chloroacetate-esterase histochemical testing were also performed. Electron microscopy. In three cases electron microscopy (EM) had been performed by the referring physician and in each of these there was evidence of typical parapox viral particles. EM was not attempted on the paraffin-embedded material submitted to us.
RESULTS
Clinical details Seventeen patients were studied, of whom 10 were male. Their ages ranged from 12 to 65 years (mean 33 years). Ten were farm workers or shepherds, three were school children who lived on or had visited a farm, one was a mechanic who had knocked
Volume 25 Number 4 October 1991
Orfand milkers' nodule 707
Fig. 1. At scanning power a nodular lesion is seen with zoned appearance. Papillary dermal edema is sandwiched between parakeratotic and papillomatous epidermis above and intense dermal inflammatory cell infiltrate below.
his head while visiting a farm, and one patient prepared dog food from sheep heads. In 14 cases the patient had been in contact with sheep, in two cases with cattle, and in one case the infecting animal was unknown. Nine lesions occurred on the hands (five, left hand; two, right hand; two, side unknown), one on the elbow, and five on the head or neck. Lesions had been apparent between 2 and 6 weeks at time of biopsy. There was a slight seasonal increase in the winter months (December through February, six patients; March through May, one; June through August, four; September through November, two). Two patients had associated immunodeficiencyone secondary to lymphoma and cytotoxic drug therapy, the other from a congenital T cell deficiency. Both these patients had large lesions. One led to amputation of the affected finger. Details of the patient with lymphoma have been reported previously.9 Histopathologic findings Overall morphology. Lesions ranged in size from 6 to 27 mm, and at low magnification a nodular lesion was apparent. In many cases there was a "zoned" appearance consisting of papillary dermal edema sandwiched by a superficial parakeratotic crust with endophytic epidermal proliferation and an underlying dermal inflammatory cell infiltrate (Fig. 1). Epidermis. The most striking feature was epidermal proliferation, which in some cases was extreme. Strands of epidermal keratinocytes sometimes only
one or two cells thick plunged into the dermis, in cases with apparent formation of a reticular network. Generally a mild to moderate degree of acanthosis was present and in all cases parakeratosis was seen. In six cases epidermal erosion was evident. Examination at high power revealed in all cases viral cytopathic changes, which varied in degree from subtle to extreme. These were present in both early (8-day) and late (6-week) lesions; cytoplasmic vacuolation (Fig. 2), nuclear vacuolation, and cytoplasmic inclusion bodies were seen. The cytoplasmic inclusions tended to occur in the upper epidermis and were deeply eosinophilic homogeneous bodies, 3 to 5 ,urn, often with a surrounding pale halo. They were frequently clustered and generally were seen at sitesthat showed other evidenceofcytopathic change (Fig. 3). Sections stained with phloxine-tartrazine showed inclusion bodies well, with a pink-red reaction product but these were also visible on routine hematoxylin-eosin-stained sections. In some instances, cytoplasmic vacuolation was so extreme that epidermal degeneration and vesicle formation occurred. In places, often within follicular structures, the keratinocyte degeneration had a distinctive "spongiform" appearance with vacuolated cells developing wispy strands of eosinophilic cytoplasm, a prominent cell membrane and pyknotic nuclei (Fig. 4). In areas of otherwise normal epidermis, individual dyskeratotic cells were occasionally found and in the downward fronds of epidermis there were occasional typical mitoses. In one case occasional multinucleate keratinocytes were seen, but this was
Journal of the American Academy of Dermatology
708 Groves et al.
Fig. 2. Cytoplasmic vacuolation was frequent, sometimes leading to vesicle formation. Note also granular appearance of keratinocyte cytoplasm in this area.
not a general finding and nuclear inclusions were not seen. Frequently there was an element of spongiosis that sometimes contributed to vesicle formation. Keratohyalin in areas of viral cytopathic change often had an inky, clumped appearance. The often intense dermal infiltrate (discussed later) frequently involved the epidermis. This epidermal inflammatory cell infiltrate in some instances was seen as an interface dermatitis (Fig. 5, A), in others spreading·into the epidermis as a diffuse infiltrate and occasionally collecting as an abscess (Fig. 5, B). In general it qualitatively reflected the underlying dermal infiltrate, which comprised lymphocytes, polymorphs, or eosinophils. In instances the intraepidermal infiltrate particularly involved follicular structures and sometimes led to their destruction. In every respect the tissue from the patients who had been in contact with cows (presumed milkers' nodule infection) was indistinguishable from that taken from the patients with orf. Dermis. A dense dermal inflammatory cell infiltrate was present in all cases. In general, this was predominantly lymphohistiocytic, but in some instances there was a polymorph component, possibly related to secondary infection, sometimes with accompanying eosinophils. In addition, the chloroacetate-esterase reaction showed a significant mast cell population. As already noted, this in:filtrate frequently also involved the epidermis. Along with this cellular infiltrate there was frequently marked dermal edema that appeared to expand the dermal pa-
pillae both vertically and horizontally and may in part contribute to the overall papillomatous appearance. Perhaps the most remarkable feature of the dermis was the presence of massive capillary proliferation and dilation, sometimes seen as huge capillary lakes that gave the impression of an angiomatous lesion (Fig. 6). Like the viral changes already described, this vascular proliferation was seen in both early and late lesions. Reticulin staining showed an intact reticulin pattern around these vessels. In places, sheets of endothelial cells with no apparent lumen were seen. These appeared to be actively dividing with many mitoses. In some cases dermal hemorrhage and neutrophil nuclear dust were evident. This presumably occurred as a secondary degenerative phenomenon because there was no other evidence of vasculitis. Again, no difference in dermal histologic features could be found between the orf and milkers' nodule cases. DISCUSSION
The clinical findings of this study are consistent with previous reports. That the disease affected more men than women in our sample is not surprising because men are more likely to be in direct contact with farm animals. The disease can clearly affect persons within a wide age range and it seems that immunity to orf is not acquired by those in contact with sheep because many patients in this survey were middleaged or older. Immunity to orf can be produced ex-
Volume 25 Number 4 October 1991
Or! and milkers' nodule 709
Fig. 3. Inclusion bodies were often clustered, as here, and generally had a surrounding pale halo.
perimentallyiO but is not permanent, as re-infection frequently occurs. ii The role of the immune system in controlling orf is clearly important, however, because two patients in our survey were immunosuppressed. It is noteworthy that each of these cases was atypical; both had extremely large lesions and one led to amputation of the affected finger. It is possible that with increasing organ transplantation and consequent use of immunosuppressive therapies and with further cases of the acquired immunodeficiency syndrome, more atypical orf infection will occur and the immune status of patients with giant or atypical orf should be investigated. Our series confirms that direct contact with animals is not required for orf infection because one patient acquired the disease from a contaminated barn beam. With one exception, all lesions occurred on extremities, either hands (usually the left), arms, or the head. This reinforces the idea that the infection is generally contracted from the handling of infected animals or dead tissue and that the site of infection depends on local practices. In Western Scotland, lesions generally occur on the left hands of shepherds (Dr. R. S. C. Fergusson, Beauly, Inverness-shire, Scotland, personal communication, July 1989) as they tend to be bitten on the nondominant hand (which holds open the animal's mouth) while dosing stock. Seasonal variation in the incidence of orf has been previously noted 6 and was seen in our series. The reason for this is unclear, although one explanation (Dr. Fergusson) is that in the winter
Fig. 4. "Spongiform degeneration." This was seen particularly within follicular structures. Note wisps of cytoplasm that remain in vacuolated cells, thick cell membrane, and persistent pyknotic nuclei.
months the only available feed is gorse, which is prickly, cuts the sheeps' lips, and allows infection. In New Zealand, however, cases of orf directly parallel the number of lambs slaughtered per month. ll
710 Groves et al.
Journal of the American Academy of Dermatology
Fig. 5. Epidermal inflammatory infiltrate ranged from (A) interface dermatitis to (B) abscess formation.
Fig. 6. Subepidermal capillary proliferation and dilation is marked, and forming (A) lakes or (B) aggregations of thin-walled vessels. Note also in (A) intense papillary dermal edema and in (B) heavy lymphocYtic infiltrate.
We have reported the histopathologic features of 17 cases of orf or milkers' nodule selected on a clinical basis or the ultrastructural demonstration of parapox viral particles. These features were characteristic; taken together, the endophytic strandlike epidermal downgrowths, papillary dermal edema and marked capillary dilation and proliferation are unique and unlike anything seen in other cutaneous
viral lesions such as herpes simplex, herpes zoster, molluscum contagiosum, coxsackie or human papillomavirus infections. In all our cases there were features typical of virus infection, with inclusion bodies, and cytoplasmic and nuclear vacuolation. These features were variable in degree, ranging from subtle to extreme. These viral cytopathic changes set orf apart from potentially confusing differential diag-
Volume 25 Number 4 October 1991
noses such as pyogenic granuloma or inflammatory vascular neoplasms. Other authors have failed to mention7 or have been unable to detect8 evidence of cytoplasmic inclusion bodies in orf and it may be that these changes can be subtle. However, even in the less typical cases of our series, such changes were present. Although previous authors 12 have described the presence of intranuclear inclusions, we were unable to see these and ultrastructural studies have shown the virus to be cytoplasmic rather than nuclear. 13 Virions are mainly found in degenerating keratinocytes scattered freely within the cytoplasm as single particles or in small groups together with viroplasm 13 and it may be that these are the ultrastructural correlates of the cytoplasmic inclusions seen on light microscopy. In common with previous descriptions, we were unable to identify any histopathologic difference between orf and milkers' nodule. 14,15 Atypical presentation without marked viral changes, although uncommon in our experience, may potentially cause diagnostic confusion that could be helped by the use of other diagnostic techniques such as EM or immunohistochemistry. Although previous investigators have attempted the latter unsuccessfully with polyclonal antiserum against the orf virus, 16 we are currently investigating this further with monoclonal antibodies. Initial studies have been encouraging. 17 EM should be regarded as the most accurate in establishing the diagnosis of paravaccinia infections but those cases that come to biopsy are often atypical and tissue is therefore not obtained for EM study. In addition, EM is frequently unavailable in outlying areas, precisely where orf is most common. We believe that in most cases, however, the light microscopic appearance is sufficiently characteristic for a diagnosis to be made and that conventional histopathologic technique has an important diagnostic role. Orf may well be an example of epidermal cytokines that have a profound influence on, and largely create, the disease. The orf virus is only to be found in upper, degenerating keratinocytes 13 and is unlikely to be directly responsible for the dermal changes. The epidermis is capable of producing many cytokines, including interleukins 1, 3, 6, and 8, granulocyte-macrophage colony-stimulating factor, and tumor necrosis factor-a. I8 Many of these me-
Or! and milkers' nodule 711 diators are neutrophil and lymphocyte chemoattractants; others are known to be angiogenic and to have mitogenic effects on keratinocytes. It seems likely that virus infection causes release of one or more cytokines with production of the typical lesion. Orf, although generally a trivial and self-limiting process, is an important condition to be able to diagnose correctly because inappropriate treatment may cause long-term disability. We believe that a close and thorough assessment of the histopathology should enable a positive diagnosis to be made in the majority of cases. REFERENCES 1. Nagington J, Whittle CH. Human orf: isolation of the vi-
rus by tissue culture. Br Med J 1961;2:1324-7. 2. Nagington J, Tee GH, Smith JS. Milkers nodule virus infections in Dorset and their similarity to orf. Nature 1965;208:505-7. 3. Nagington J. The growth of paravaccinia viruses in tissue culture. Vet Rec 1968;80:477-82. 4. Shelley WB, Shelley ED. Farmyard pox: parapox virus infection in man. Br J DermatoI1983;108:'725-7, 5. Leavell VW, McNamara MJ, Muelling R, et at Orf: report of 19 human cases with clinical and pathological observations. JAMA 1968;203:657-64. 6. PHLS Communicable Disease Surveilance Centre. Orf paravacciriia infections, British Isles 1975-1981. Br Med J 1982;284:1958. 7. Johannessen JV, Krogh H-K, Solberg r, et al. Human orf. J Cutan Pathol 1975;2:265-83. 8. Austin C, Vuitch F, Freeman R. Milkers' nodule and orf: a clinicopathologic evaluation [Abstract). J Cutan Pathol 1987;14:348. 9. Savage J, Black MM. "Giant" orfof finger in a patient with a lymphoma. Proc R Soc Med 1972;65:766-8. 10. Blakemore F, Abdussalam M, Goldsmith WN. A case of orf: identification of the virus. Br J Dermatol 1948;60: 404-9. 11. Robinson AJ, Petersen GV. Orfvirus infection of workers in the meat industry. N Z Moo J 1983;96:81-5. 12. Leavell UW, McNamara MJ, Muelling RJ. Ecthyma contagiosum (orf). South Med J 1965;58:239-43. 13. Yeh H-P, Soltani K. Ultrastructural studies in human orf. Arch Dermatol 1974;109:390-2. 14. Davis CM, Musil G. Milkers' nodule. Arch Dermatol 1970;101 :305-11. . 15. Leavell VW, Phillips lA. Milkers' nodules: pathogenesis, tissue culture, electron microscopy and calf inoculation. Arch Dermatol 1975;11] :1301-11. 16. Sanchez RL, Hebert A, Lucia H, et at Orf: a case report with histologic, electron microscopic and immunoperoxidase studies. Arch Pathol Lab Moo 1985;109:166-70. 17. Groves R W, Wilson Jones E, MacDonald DM. Human orf: morphologic characteristics and immunohistochemical diagnosis. J Cutan Pathol 1989;16:305. 18. Kupper TS. Production of cytokines by epithelial tissues. Am J Dermatopathol1989;11:69-73.
16/1/30377
706
portant aspects (e.g., the presence or absence of inclusion bodies) remain controversia1. 8 To define accurately the histopathology and to resolve some of these controversies, we have undertaken a detailed histopathologic study of orf and milkers' nodule infection.
MATERIAL AND METHODS Histopathologic material. Dermatology departments throughout the United Kingdom were asked to submit specimens diagnosed as orf or milkers' nodules. Twentysix specimens were received, in 17 of which adequate tissue was available for study, and the diagnosis was considered unequivocal on the basis ofclinical data (history, examination, contact with sheep or cattle) or ultrastructural demonstration of viral particles. Hematoxylin-and-eosin sections were examined in all instances and in selected cases reticulin stains and phloxine-tartrazine and chloroacetate-esterase histochemical testing were also performed. Electron microscopy. In three cases electron microscopy (EM) had been performed by the referring physician and in each of these there was evidence of typical parapox viral particles. EM was not attempted on the paraffin-embedded material submitted to us.
RESULTS
Clinical details Seventeen patients were studied, of whom 10 were male. Their ages ranged from 12 to 65 years (mean 33 years). Ten were farm workers or shepherds, three were school children who lived on or had visited a farm, one was a mechanic who had knocked
Volume 25 Number 4 October 1991
Orfand milkers' nodule 707
Fig. 1. At scanning power a nodular lesion is seen with zoned appearance. Papillary dermal edema is sandwiched between parakeratotic and papillomatous epidermis above and intense dermal inflammatory cell infiltrate below.
his head while visiting a farm, and one patient prepared dog food from sheep heads. In 14 cases the patient had been in contact with sheep, in two cases with cattle, and in one case the infecting animal was unknown. Nine lesions occurred on the hands (five, left hand; two, right hand; two, side unknown), one on the elbow, and five on the head or neck. Lesions had been apparent between 2 and 6 weeks at time of biopsy. There was a slight seasonal increase in the winter months (December through February, six patients; March through May, one; June through August, four; September through November, two). Two patients had associated immunodeficiencyone secondary to lymphoma and cytotoxic drug therapy, the other from a congenital T cell deficiency. Both these patients had large lesions. One led to amputation of the affected finger. Details of the patient with lymphoma have been reported previously.9 Histopathologic findings Overall morphology. Lesions ranged in size from 6 to 27 mm, and at low magnification a nodular lesion was apparent. In many cases there was a "zoned" appearance consisting of papillary dermal edema sandwiched by a superficial parakeratotic crust with endophytic epidermal proliferation and an underlying dermal inflammatory cell infiltrate (Fig. 1). Epidermis. The most striking feature was epidermal proliferation, which in some cases was extreme. Strands of epidermal keratinocytes sometimes only
one or two cells thick plunged into the dermis, in cases with apparent formation of a reticular network. Generally a mild to moderate degree of acanthosis was present and in all cases parakeratosis was seen. In six cases epidermal erosion was evident. Examination at high power revealed in all cases viral cytopathic changes, which varied in degree from subtle to extreme. These were present in both early (8-day) and late (6-week) lesions; cytoplasmic vacuolation (Fig. 2), nuclear vacuolation, and cytoplasmic inclusion bodies were seen. The cytoplasmic inclusions tended to occur in the upper epidermis and were deeply eosinophilic homogeneous bodies, 3 to 5 ,urn, often with a surrounding pale halo. They were frequently clustered and generally were seen at sitesthat showed other evidenceofcytopathic change (Fig. 3). Sections stained with phloxine-tartrazine showed inclusion bodies well, with a pink-red reaction product but these were also visible on routine hematoxylin-eosin-stained sections. In some instances, cytoplasmic vacuolation was so extreme that epidermal degeneration and vesicle formation occurred. In places, often within follicular structures, the keratinocyte degeneration had a distinctive "spongiform" appearance with vacuolated cells developing wispy strands of eosinophilic cytoplasm, a prominent cell membrane and pyknotic nuclei (Fig. 4). In areas of otherwise normal epidermis, individual dyskeratotic cells were occasionally found and in the downward fronds of epidermis there were occasional typical mitoses. In one case occasional multinucleate keratinocytes were seen, but this was
Journal of the American Academy of Dermatology
708 Groves et al.
Fig. 2. Cytoplasmic vacuolation was frequent, sometimes leading to vesicle formation. Note also granular appearance of keratinocyte cytoplasm in this area.
not a general finding and nuclear inclusions were not seen. Frequently there was an element of spongiosis that sometimes contributed to vesicle formation. Keratohyalin in areas of viral cytopathic change often had an inky, clumped appearance. The often intense dermal infiltrate (discussed later) frequently involved the epidermis. This epidermal inflammatory cell infiltrate in some instances was seen as an interface dermatitis (Fig. 5, A), in others spreading·into the epidermis as a diffuse infiltrate and occasionally collecting as an abscess (Fig. 5, B). In general it qualitatively reflected the underlying dermal infiltrate, which comprised lymphocytes, polymorphs, or eosinophils. In instances the intraepidermal infiltrate particularly involved follicular structures and sometimes led to their destruction. In every respect the tissue from the patients who had been in contact with cows (presumed milkers' nodule infection) was indistinguishable from that taken from the patients with orf. Dermis. A dense dermal inflammatory cell infiltrate was present in all cases. In general, this was predominantly lymphohistiocytic, but in some instances there was a polymorph component, possibly related to secondary infection, sometimes with accompanying eosinophils. In addition, the chloroacetate-esterase reaction showed a significant mast cell population. As already noted, this in:filtrate frequently also involved the epidermis. Along with this cellular infiltrate there was frequently marked dermal edema that appeared to expand the dermal pa-
pillae both vertically and horizontally and may in part contribute to the overall papillomatous appearance. Perhaps the most remarkable feature of the dermis was the presence of massive capillary proliferation and dilation, sometimes seen as huge capillary lakes that gave the impression of an angiomatous lesion (Fig. 6). Like the viral changes already described, this vascular proliferation was seen in both early and late lesions. Reticulin staining showed an intact reticulin pattern around these vessels. In places, sheets of endothelial cells with no apparent lumen were seen. These appeared to be actively dividing with many mitoses. In some cases dermal hemorrhage and neutrophil nuclear dust were evident. This presumably occurred as a secondary degenerative phenomenon because there was no other evidence of vasculitis. Again, no difference in dermal histologic features could be found between the orf and milkers' nodule cases. DISCUSSION
The clinical findings of this study are consistent with previous reports. That the disease affected more men than women in our sample is not surprising because men are more likely to be in direct contact with farm animals. The disease can clearly affect persons within a wide age range and it seems that immunity to orf is not acquired by those in contact with sheep because many patients in this survey were middleaged or older. Immunity to orf can be produced ex-
Volume 25 Number 4 October 1991
Or! and milkers' nodule 709
Fig. 3. Inclusion bodies were often clustered, as here, and generally had a surrounding pale halo.
perimentallyiO but is not permanent, as re-infection frequently occurs. ii The role of the immune system in controlling orf is clearly important, however, because two patients in our survey were immunosuppressed. It is noteworthy that each of these cases was atypical; both had extremely large lesions and one led to amputation of the affected finger. It is possible that with increasing organ transplantation and consequent use of immunosuppressive therapies and with further cases of the acquired immunodeficiency syndrome, more atypical orf infection will occur and the immune status of patients with giant or atypical orf should be investigated. Our series confirms that direct contact with animals is not required for orf infection because one patient acquired the disease from a contaminated barn beam. With one exception, all lesions occurred on extremities, either hands (usually the left), arms, or the head. This reinforces the idea that the infection is generally contracted from the handling of infected animals or dead tissue and that the site of infection depends on local practices. In Western Scotland, lesions generally occur on the left hands of shepherds (Dr. R. S. C. Fergusson, Beauly, Inverness-shire, Scotland, personal communication, July 1989) as they tend to be bitten on the nondominant hand (which holds open the animal's mouth) while dosing stock. Seasonal variation in the incidence of orf has been previously noted 6 and was seen in our series. The reason for this is unclear, although one explanation (Dr. Fergusson) is that in the winter
Fig. 4. "Spongiform degeneration." This was seen particularly within follicular structures. Note wisps of cytoplasm that remain in vacuolated cells, thick cell membrane, and persistent pyknotic nuclei.
months the only available feed is gorse, which is prickly, cuts the sheeps' lips, and allows infection. In New Zealand, however, cases of orf directly parallel the number of lambs slaughtered per month. ll
710 Groves et al.
Journal of the American Academy of Dermatology
Fig. 5. Epidermal inflammatory infiltrate ranged from (A) interface dermatitis to (B) abscess formation.
Fig. 6. Subepidermal capillary proliferation and dilation is marked, and forming (A) lakes or (B) aggregations of thin-walled vessels. Note also in (A) intense papillary dermal edema and in (B) heavy lymphocYtic infiltrate.
We have reported the histopathologic features of 17 cases of orf or milkers' nodule selected on a clinical basis or the ultrastructural demonstration of parapox viral particles. These features were characteristic; taken together, the endophytic strandlike epidermal downgrowths, papillary dermal edema and marked capillary dilation and proliferation are unique and unlike anything seen in other cutaneous
viral lesions such as herpes simplex, herpes zoster, molluscum contagiosum, coxsackie or human papillomavirus infections. In all our cases there were features typical of virus infection, with inclusion bodies, and cytoplasmic and nuclear vacuolation. These features were variable in degree, ranging from subtle to extreme. These viral cytopathic changes set orf apart from potentially confusing differential diag-
Volume 25 Number 4 October 1991
noses such as pyogenic granuloma or inflammatory vascular neoplasms. Other authors have failed to mention7 or have been unable to detect8 evidence of cytoplasmic inclusion bodies in orf and it may be that these changes can be subtle. However, even in the less typical cases of our series, such changes were present. Although previous authors 12 have described the presence of intranuclear inclusions, we were unable to see these and ultrastructural studies have shown the virus to be cytoplasmic rather than nuclear. 13 Virions are mainly found in degenerating keratinocytes scattered freely within the cytoplasm as single particles or in small groups together with viroplasm 13 and it may be that these are the ultrastructural correlates of the cytoplasmic inclusions seen on light microscopy. In common with previous descriptions, we were unable to identify any histopathologic difference between orf and milkers' nodule. 14,15 Atypical presentation without marked viral changes, although uncommon in our experience, may potentially cause diagnostic confusion that could be helped by the use of other diagnostic techniques such as EM or immunohistochemistry. Although previous investigators have attempted the latter unsuccessfully with polyclonal antiserum against the orf virus, 16 we are currently investigating this further with monoclonal antibodies. Initial studies have been encouraging. 17 EM should be regarded as the most accurate in establishing the diagnosis of paravaccinia infections but those cases that come to biopsy are often atypical and tissue is therefore not obtained for EM study. In addition, EM is frequently unavailable in outlying areas, precisely where orf is most common. We believe that in most cases, however, the light microscopic appearance is sufficiently characteristic for a diagnosis to be made and that conventional histopathologic technique has an important diagnostic role. Orf may well be an example of epidermal cytokines that have a profound influence on, and largely create, the disease. The orf virus is only to be found in upper, degenerating keratinocytes 13 and is unlikely to be directly responsible for the dermal changes. The epidermis is capable of producing many cytokines, including interleukins 1, 3, 6, and 8, granulocyte-macrophage colony-stimulating factor, and tumor necrosis factor-a. I8 Many of these me-
Or! and milkers' nodule 711 diators are neutrophil and lymphocyte chemoattractants; others are known to be angiogenic and to have mitogenic effects on keratinocytes. It seems likely that virus infection causes release of one or more cytokines with production of the typical lesion. Orf, although generally a trivial and self-limiting process, is an important condition to be able to diagnose correctly because inappropriate treatment may cause long-term disability. We believe that a close and thorough assessment of the histopathology should enable a positive diagnosis to be made in the majority of cases. REFERENCES 1. Nagington J, Whittle CH. Human orf: isolation of the vi-
rus by tissue culture. Br Med J 1961;2:1324-7. 2. Nagington J, Tee GH, Smith JS. Milkers nodule virus infections in Dorset and their similarity to orf. Nature 1965;208:505-7. 3. Nagington J. The growth of paravaccinia viruses in tissue culture. Vet Rec 1968;80:477-82. 4. Shelley WB, Shelley ED. Farmyard pox: parapox virus infection in man. Br J DermatoI1983;108:'725-7, 5. Leavell VW, McNamara MJ, Muelling R, et at Orf: report of 19 human cases with clinical and pathological observations. JAMA 1968;203:657-64. 6. PHLS Communicable Disease Surveilance Centre. Orf paravacciriia infections, British Isles 1975-1981. Br Med J 1982;284:1958. 7. Johannessen JV, Krogh H-K, Solberg r, et al. Human orf. J Cutan Pathol 1975;2:265-83. 8. Austin C, Vuitch F, Freeman R. Milkers' nodule and orf: a clinicopathologic evaluation [Abstract). J Cutan Pathol 1987;14:348. 9. Savage J, Black MM. "Giant" orfof finger in a patient with a lymphoma. Proc R Soc Med 1972;65:766-8. 10. Blakemore F, Abdussalam M, Goldsmith WN. A case of orf: identification of the virus. Br J Dermatol 1948;60: 404-9. 11. Robinson AJ, Petersen GV. Orfvirus infection of workers in the meat industry. N Z Moo J 1983;96:81-5. 12. Leavell UW, McNamara MJ, Muelling RJ. Ecthyma contagiosum (orf). South Med J 1965;58:239-43. 13. Yeh H-P, Soltani K. Ultrastructural studies in human orf. Arch Dermatol 1974;109:390-2. 14. Davis CM, Musil G. Milkers' nodule. Arch Dermatol 1970;101 :305-11. . 15. Leavell VW, Phillips lA. Milkers' nodules: pathogenesis, tissue culture, electron microscopy and calf inoculation. Arch Dermatol 1975;11] :1301-11. 16. Sanchez RL, Hebert A, Lucia H, et at Orf: a case report with histologic, electron microscopic and immunoperoxidase studies. Arch Pathol Lab Moo 1985;109:166-70. 17. Groves R W, Wilson Jones E, MacDonald DM. Human orf: morphologic characteristics and immunohistochemical diagnosis. J Cutan Pathol 1989;16:305. 18. Kupper TS. Production of cytokines by epithelial tissues. Am J Dermatopathol1989;11:69-73.
