British Journal of Dermatology (1975) 93J 53.

Secondary syphilis: a clinico-pathological review E.ABELL, R.MARKS* AND E.WILSON JONES St John's Hospital for Diseases of the Skin, Lisle Street, Leicester Square, London WC2H 7BJ Accepted for publication 21 October 1974

SUMMARY

The histological appearatices foutid in biopsies from fifty-seven patients with secondary syphilis have been correlated with the clinical morphology of the eruptions. Considerable variation of histological pattern was encountered, and the frequency with which some of the classically described changes were found to be absent or inconspicuous is stressed. Of particular interest were the findings that, in nearly one-quarter of the biopsies, plasma cell infiltration was either absent or very sparse, and that vascular damage was seen in less than half Where present, the vessel changes were almost entirely confined to swelling of the endothelial cells. Proliferation of the endothelial cells was most uncommon. The epidermis was very frequently involved in the inflammatory process. Exocytosis, spongiosis, parakeratosis, and acanthosis were the most frequent changes. No consistent histological difference between papular and papulo-squamous lesions could be found but macular lesions demonstrated more superficial and less intense dermal infiltration as well as less severe epidermal involvement. In late secondary lesions, the infiltrate became granulomatous, but in other respects the duration of the exanthem could not be correlated with the pathology. The differential diagnosis from pityriasis lichenoides and other inflammatory dermatoses is discussed and the value of histopathology in the diagnosis of secondary syphilis is emphasized.

An increasing incidence of early infectious syphilis has been recognized in many parts of the world over the last 20 years (Willcox, 1972). In Great Britain, despitea decline in the number of new cases in recent years (Report of the Chief Medical Officer, 1970) it continues to be prevalent, and particularly amongst male homosexuals (Waugh, 1972). Over these last two decades, the number of cases of secondary syphilis presenting at St John's Hospital has doubled, and the number of biopsies examined has increased by the same proportion. This accumulated clinical and histological material forms the basis of this report. The histopathology of secondary syphilis has usually been described in relationship to the clinical morphology of the eruption (Lever, 1967; Montogomery, 1967; Johnson, 1972; Grosshans, 1973). It was our impression from previous experience, that this relationship was imprecise. In order to * Present Address: Welsh National School of Medicine, University Hospital of Wales, Heath Park, Cardiff CF4 4XN.

53

54

EAbell, R.Marks and E.Wilson Jones

examine this clinico-his to logical relationship, sections were examined and recorded in the absence of the clinical details and correlated subsequently. MATERIALS AND METHODS

Biopsy specimens of fifty-seven serologically confirmed cases of secondary syphilis have been seen at St John's since 1952. Six patients had more than one biopsy performed, making a total of sixty-four biopsies. With the exception of six specimens all biopsies had been taken from the trunk or limbs. The remaining specimens consisted of two palmar, one plantar, two scalp and one facial lesion. The patients' ages ranged from 17-59 years, with more than half being in their third decade. The male to female ratio was 3 7 : 1 . The rashes had been present for between i week and i year. In one case the eruption had recurred over a 3-year period. As ascertained from the clincal records and photographs, twenty-nine of the lesions were classified as papules, twenty as papulo-squamous, and thirteen as macular lesions. In two patients the clinical data were insufficient for adequate classification. Haematoxylin and eosin sections were examined in all cases and in fourteen the following additional stains were used: Periodic add Schiff, orcein elastic, Gomori's reticulin, Perle's blood pigment and Masson-Fontana melanin. The various pathological changes were recorded on a standard proforma, conforming to the headings TABLE I. Histological features of sixty-four biopsies of secondary syphilis and the sixty-two with known morphology All lesions

No. of biopsies Surface change Hyperkeratosis Parakeratosis Acanthosis Spongiosis Keratinocyte necrosis Exocytosis Ci) all (ii) mononuclear (iii) mixed microabscess "] spongiform > pustulation J Papillary oedema Papillary RBC extravasation Vascular thickening Inflammatory cell infiltration (i) sparse (ii) moderate (iii) dense Plasma cells absent or very sparse

Number

(%)

Macules

Papules

Papulosquamuos

64 6

IOO

13 — —

29

20

I

14

I

II

8

4

23

13

3

II

5

12

9

3 7

3 5

19

19 41 31 625 30

54 36

84 55

7 6

28 18

18

i8

28

I

10

7

13

20



8

5

34

53

2

19

13

17 27

23 42

I

9

5

5

10

12

2



18 9

26 20

40

64 6 23

9 36

33

52

4 5 4

15

23

3

9

II

II

9 II

3

CHmco-pathologtcal review of secondary syphilis

55

of Table i. Only the presence or absence of each feature was recorded together with a qualifying descriptive comment. The dermal inflammatory infiltrate was scored by an arbitrary three plus system corresponding to sparse, moderate, or dense. The infiltrating cells were also characterized morphologically. At least two observers scored the lesions independently and the recorded data represented an amalgam of these observations. In eleven of the cases, the correct diagnosis was not suspected initially, and in nine, syphilis was only considered as a possibility in the differential diagnosis. In these twenty cases the clinical diagnoses most commonly made were in order of frequency—pityriasis lichenoides, psoriasis, eczema, insect bites, sarcoid, leishmaniasis and lymphoma. Hisiopathology

(a) The epidermis (Table i). The degree to which the epidermis was involved varied considerably. Nearly one-third (20 specimens) showed mild to moderate acanthosis and in two of these biopsies striking elongation of the rete ridges gave a psoriasiform outline to the epidermal contour (Fig. i).

FIGURE I. Papulo-squamous lesion (4 weeks) showing psoriasiform epidermal hypcrplasia and spongiform pustulation. The papillae are markedly oedematous and the infiltrate is predominantly disposed around dilated vessels. (H & E x ioo.)

In the remainder, the epidermis appeared either unchanged or attenuated. This thimiing of the epidermis was typically seen in lesions in which intense papillary oedema was also present (Fig. 2). Epidermal oedema of mild to severe extent characterized nearly two thirds (40) of the specimens but its intensity did not correlate directly with the intensity of the dermal reaction below. Exocytosis (defined as the presence of inflammatory cells in the epidermis) was the most constant feature (53) and was of two distinct types. In thirty-five, this infiltrate was composed of small round mononuclear cells only, occurring in the epidermis difl"usely or in small foci. In eighteen sections, both polymorphonuclear and small round cells had invaded the epidermis and in thirteen of these, aggregations had occurred producing collections of cells in the stratum corneum resembling Munro abscesses (Fig. 3) and less frequently spongiform pustulation (Fig. i).

E.Abell, RMarks and E.Wilson Jones

FIGURE 2. Papulo-sQuamous lesion (6 weeks). An irregularly acanthotic; epidermis is thinned over oedematous papillae. Note the pronounced plasma cell infiltration of the dermis.(H & E x ioo.)

FIGURE 3. Papule (2 weeks). Munro abscess collections of inflammatory cells in an oedematous epidermis. Note the heavy dermal infiltrate obliterating the epidermo-dermal interface. (H & E X 150.)

Hyperkeratosis, confined to twelve papular and papulo-squamous lesions, was mild and focal in ten and prominent in only two. Twenty-six sections demonstrated degrees of parakeratosis varying from mild and patchy to a continuous parakeratotic scale, well seen in some papulo-squamous lesions. Mild epidermal thickening and parakeratosis were seen only occasionally in macules, although more than half (seven) showed some mild inflammatory cell permeation of the epidermis. Occasionally (19), single or small groups of epidermal cells were encountered undergoing hyaline necrosis (Fig. 4). In four specimens partial erosion of the upper epidermis with crusting was noticed, but no example of complete ulceration of the Lues Maligna type was seen. (b) The dermis (Table i). The inflammatory infiltrate was always most intense in the papillary dermis but generally the infiltrate also extended perivascularly into the reticulat dermis. It did not tightly sleeve the vessels as in a figutate erythema, but was loosely dispersed to involve the surrounding tissue. In only four biopsies, all from macules, was the infiltrate limited to the upper dermis (Fig. 5).

Clinico-pathological review of secondary syphilis

57

FIGURE 4. Papule (i week). Focal epidermal parakeratosis and spongiosis with a group of necrotic keratinocytes (hyaline bodies). (H & E x 100.)

FIGURE 5. Macule (8 weeks). A loose lymphocytic infiltrate surrounds the superficial dilated vessels. Plasma cells are prominent in this case. (H & E x 60.)

Almost all papular and papulo-squamous lesions (47 of 49) demonstrated a moderate or dense inflammatory cell infiltrate which in many sections obliterated the overall perivascular pattern (Fig. 3), producing a lichenoid appearance in the upper dermis. In predominantly granulomatous lesions confluent areas of infiltrate also extended into the deeper dermis. In the larger biopsies, this perivascular arrangement could be seen to extend deeply around a single or a small number of deep dermal vessels. The follicles and sweat glands were very frequently sleeved by inflammatory cells and in eight biopsies the focal point of the reaction was centered around the pilosebaceous units giving the appear-

^ R.Marks and E.Wilson Jones

FIGURE 6. FolHcular papule (8 weeks). Note the destructive follicular inflammatory reaction with some sleeving of adjacent vessels. (H & E x 40.) ances of a foUiculitis (Fig. 6). Although the follicular epithelitim showed exocytosis and parakeratotic

hortiy plugging, destruction of the follicle was seen in only one biopsy. In two specimens, the infiltrate occupied the perineural space, but in none were the nerves themselves infiltrated. For descriptive purposes the infiltration was classified as being (a) predominantly lymphocytic (3); (b) lympho-histiocytic (43); (c) predominantly histiocytic (9); and (d) predominantly granulomatous (9). Plasma cell infiltration occurred in varying degree with each category. The density of this plasma cell infiltrate could not be correlated with either the stage of the disease or the types of other cells present. Large pale histiocytes, aggregations of epithelioid cells and multinuclear giant cells were seen in two lesions of 9 weeks duration and a granulomatous infiltrate characterized all lesions of 16 weeks duration or more (Fig. 7) with a single exception. In this instance antibiotic treatment had caused partial resolution before a specific diagnosis was reached. In two specimens focal granulomas were found in lesions of 4 weeks duration only. Severe papillary oedema accompanied the infiltrate in thirty-four biopsies and red cell extravasation also in seventeen cases. Vasodilatation was very common, but significant vessel damage was, by contrast, infrequent. In twenty-four biopsies noticeable swellingof the endothelial cells was apparent but usually not prominent. Endothehal proliferation was seen in only three cases, and in two of these the eruption had been present for 26 and 52 weeks respectively. In six patients multiple biopsies had been performed. The histological patterns differed only in those two cases in which different morphological lesions had been sampled. Little additional information was obtained from the histochemical stains. Some papules showed loss of elastic tissue. DISCUSSION

The old adage that syphilis is the great mimic is perhaps just as true for the histologist as it is for the

Clinico-pathological review of secondary syphilis

59

FIGURE 7. Papulo-squamous lesion (20 weeks) A diffuse granulomatous infiltrate occupies most of the dermis with cells extending into the epidermis as well. (H & E x 60.)

clinician. This study has detailed both the wide variety of histological change that occurs in secondary syphilis and its value in the diagnosis of the unsuspected case. Indeed on occasions it has been possible to alert the clinician to the possibility of unsuspected syphihs because of the histological appearances. Certainly many biopsies conform to the classically described features of secondary syphilis. Prominent among these are the presence of plasma cells in a perivascular infiltrate extending deeply in the skin, accompanied by evidence of endothelial damage to small blood vessels. However, these features cannot be relied upon, and in particular minor degrees of endothelial damage are difficult to assess and are a common feature in many other inflammatory dermatoses. Even the presence of plasma cells is not invariable, but in their absence it may still be possible to suspect syphilis by careful consideration of the total histological picture. In our material, plasma cells were absent or inconspicuous in nearly 25" „ of the biopsies. Alteration of small blood vessels was usually confined to endothehal swelling and prominent endothelial proliferation was very uncommon. In only one specimen the presence in the lower dermis of swollen tortuous arterioles showing obvious endarteritis was an exceptional finding. Excluding vasodilatation, the vessels were unaffected in more than half the biopsy specimens examined.

6o

E.Abell, R.Marks and E.Wilson Jones

In our material the epidermis was found to be involved in more than three-quarters of all the biopsies. While the severity and extent of this involvement varied considerably, we do not believe that the frequency with which this tissue is affected has been sufficiently emphasized in the past (Lever, 1967; Montgomery 1967). No clear histological differences between the papular and papulo-squamous lesions were detected. However, Table i shows that the incidence of epidermal involvement in macular lesions was less. There was also less dense inflammatory infiltration in the macular lesions, and it tended to be more superficially orientated. In many macular lesions, the histology was often not distinctive, particularly if plasma cells were few in number and biopsy may be of little practical help in the diagnosis of some of these cases. Differential diagnosis

In sections demonstrating a perivascular infiltrate as well as epidermal involvement, eczema, psoriasis, lichenoid reactions, erythema multiforme, and drug eruptions must be considered in the differential diagnosis of secondary syphilis. However, the presence of infiltrate at much deeper levels in the dermis in syphilis should normally rule out the essentially superficial reactions found in eczema, psoriasis, and the Uchenoid conditions. The characteristic histological changes of erythema multiforme are focused upon the epidermo-dermal interface, where oedema and epidermal cell vacuolation lead to necrosis and subepidermal vesicle and bulla formation (Ackerman, Penneys & Clark, 1971}. The eosinophilia frequently accompanying drug eruptions is a feature not seen in secondary syphilis. The presence of a heavy upper dermal infiltrate with invasion of mononuclear cells into the epidermis may simulate a premycosis or mycosis fungoides. In mycosis fungoides, however, large hyperchromatic mononuclear cells with crenated nuclear outline may be seen especially among these cells colonizing the epidermis. In syphilis, particularly in papular lesions, the infiltrate tends to take on a histiocytic or granuiomatous quality which is usually lacking in mycosis fungoides. Large hyperchromatic mononuclear cells were seen in only one biopsy, in which they represented an extremely small proportion of the lympho-histiocytic infiltrate which also contained abundant plasma cells. Pityriasis lichenoides is frequently the most difficult condition to exclude both clinically and histologically. Not infrequently, papular lesions were encountered showing focal epidermal changes of oedema, exocytosis, parakeratosis and keratinocyte necrosis overlying oedematous inflamed papillae with capillary haemorrhage. The syphilitic reaction pattern may closely resemble the changes of pit)'riasis lichenoides especially where plasma cells are absent. This pityriasis lichenoides-like reaction requires particular emphasis as this was also the most frequent chnical mis-diagnosis. However, deep perivascular infiltration is usually seen in pityriasis lichenoides only in association with acute necrotic lesions. The formation of granulomatous foci in the dermis was the sole feature which appeared to correlate with the duration of the eruption. In eruptions of 2-4 months' duration granulomatous foci were common, and in older untreated lesions they were consistently present. While no attempt has been made to classify each individual reaction pattern, the report by Jeerapaet & Ackerman (1973) also produced similar conclusions. They emphasized the epidermal involvement, particularly the exocytosis, spongiform pustulation, and keratinocyte necrosis. They also found a similar proportion of cases without plasma cell infiltration but quoted a much higher proportion of biopsies showing changes in the blood vessels. Johnson's review (1972) of forty-seven cases of secondary syphilis and the description by Pinkus & Mehregan (1969) both described the epidermal involvement and the relative paucity of plasma cells. Neither, however, gave the order of frequency with which these situations may arise.

Clinico-pathological review of secondary syphilis

6i

The characteristic features of secondary syphiUs seen in this study were those of a deeply extending perivascular inflammatory infiltrate containing a variable proportion of plasma cells. This infiltrate was mainly lymphohistiocytic but became frankly granulomatous in the older lesions. The dermal blood vessels were usually dilated but thickened endothelial cells were present in less than half of the specimens. Papillary oedema and erythrocyte extravasation sometimes accompanied these changes. The epidermis was very frequently involved, particularly in the papular lesions. Exocytosis, sometimes in the form of Munro abscess or spongiform pustulation, was the commonest reaction and was almost always associated with epidermal oedema. Parakeratosis, acanthosis and keratinocyte necrosis were less commonly present. ACKNOWLEDGMENTS

We wish to acknowledge the help of Mr E. Wheeler and Mr S.Robertson and their staff in preparation of the sections and photomicrographs for this work. REFERENCES AcKERMAN, A.B., PENNEYS, N . S . & CLARK, W.H. (i971) Erythema multiforme exudativtim, distinctive pathological process. British Journal of Dermaiology^ 84, 554. Annual Report of the Chief Medical Officer to the Department of Health and Social Security for the year (1970) H.M.S.O., London, 1972. GROSSHANS, E . C1973) Histopathologie der cutanen Syphilisiformen und ubrigen Spirochatosen. In: Spesielle Pathologische Anatomie (Ed. by U.W, Schynder), Vol. 7, p. 114. Springer, Berlin. JEERAPAET, P . & ACKERMAN, A.B. (1973) Histologic patterns of secondary syphilis. Archives oj Dermatology, 107, 373JOHNSON, W . C . in: GRAHAM, J.H., JOHNSON, W . C . & HELWIG, E . B . (1972) Dermal Pathology, p. 373. Harper and

Row, New York. LEVER, W . F . (1967) Histopathology of the Skin, 4th edn, p. 322. J.B. Lippincott Co., Philadelphia. MONTGOMERY, H . (1967) Dermaiopaihology, Vol. i, p. 41S. Harper and Row, New York. PiNKUS, H. & MEHREGAN (1969) A Guide to Dermatohisiopathology, p. 232. Appleton-Century-Crofts, New York. WAUGH, M . A . (1972) Studies on the recent epidemiology of early syphilis in west London. British Journal of Venereal Disease, 48, 534. WiLLCOX, R.R. (1972) A world wide view of venereal diseases. British Journal of Venereal Disease, 48, 163.

Secondary syphilis: a clinico-pathological review.

The histological appearances found in biopsies from fifty-seven patients with secondary syphilis have been correlated with the clinical morphology of ...
2MB Sizes 0 Downloads 0 Views