J. clin. Path., 1970, 23, 668-675

Infected infarcts of the testis: A study of 18 cases

preceded by pyogenic epididymoorchitis

D. O'B. HOURIHANEFrom the Central Histopathology Laboratory of the Federated Dublin Hospitals School ofPathology,Trinity College, Dublin

SYNOPSIS Eighteen cases of infected infarcts of the testis are presented, and evidence is

put forward that these result from venous occlusions in the epididymis and cord. The venous

lesions probably result from thrombosis during the course of an attack of epididymoorchitis.Granulomatous orchitis was present in some part of half of the orchidectomy specimens,

and the clinical histories, bacteriological findings, and histological data all suggest that thisform of inflammation results from pyogenic infection of the testicle. What the factor is whichdetermines whether the inflammation is granulomatous or not is unknown.

Apart from torsion and occasional involvementin polyarteritis nodosa testicular infarction is rare,and the receipt of several specimens stimulatedinterest in the lesion.Between January 1964 and December 1969 we

have received 18 testes with infarcts (excludingcases of torsion) and this paper presents thefindings from their study.The ages of the patients ranged from 26 years

to 74 years, the majority being between 30 and65 years (Table I). Most of the patients had a

history of urinary tract disease, mainly of attacksof cystitis, often with haematuria. The common

history was of 'cystitis' for seven to 10 days, thenof the development of testicular pain and swellingusually diagnosed as epididymoorchitis (Tables Iand II). Conservative treatment for three to seven

weeks failed to lead to resolution and manydeveloped a fluctuant mass in the testis. In six

patients a scrotal sinus developed and in six othersits imminence dictated immediate orchidectomy(Tables I and LI).

In most of the patients E. coli was culturedfrom the urine, and in all cases in which culture

Received for publication 12 February 1970.

Case Age Clinical Haema- E. coli Cal- PreviousNo. (yr) Epididymo- turia turedfrom Surgery

orchitisUrine Testis

1 55 0' +2 43 z + 03 60 0 +4 48 - + + Hernial repair5 69 - Ps. pyo.O Transurethral

resection ofprostate +hernial repair

6 30 + + - +7 56 + + 0 +8 70 - - +9 30 + + + +10 36 + - Proteusl1 26 - + 0 012 67 - - 0 Hernial repair13 71 + + + +14 48 + - +15 42 + - +16 45 + + 017 57 + + 0 +18 74 + + + + Hernial repair

and retro-pubic prostat-ectomy

Table I Clinical summary of 18 cases

10 = no culture performed.

copyright. on M

arch 28, 2021 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.23.8.668 on 1 N

ovember 1970. D

ownloaded from

669 Infected infarcts of the testis: A study of 18 cases preceded by pyogenic epididymoorchitis

Case Testicular Scrotal Skin Histology ofNo. Swelling (no. of Involved Surviving Testis

days beforeoperation)

1 50 + F2 44 03 38 04 21 F5 ? + F6 31 + F7 51 - G8 26 - G9 35 + F10 40 + 011 ? + M

12 ? + F13 64 - M14 36 + F15 42 + M16 28 ± M17 42 + G18 ? + F

Table II Length of history, involvement of skin, andhistology ofsurviving testes in 18 patients

0 = no recognizable viable testis.F = fibrotic tubules.G = granulomatous tubules.M = both fibrotic and granulomatous tubules.

of the testicular tissue was performed E. coli wasgrown. Four patients had had a hernial repair(three ipsilateral and one contralateral) and twopatients had had a recent prostatectomy.

Pathology

The specimens were extraordinarily similar sothat the naked eye appearances could almost leadone to predict the history of the patient.There were two types. In the first and most

common (12 cases) the testis was of normal size orslightly enlarged. The cord was thickened by grey-white fibrous tissue in which were scatteredyellow-brown foci, and the cut surface of the testiswas more or less replaced by a yellow-greensuppurative mass in which tubules could still beidentified (Fig. 1). The surviving testis (if any) wasrepresented by a peripheral rim of pale brown softtissue. The tunica was usually grossly thickened,and skin was attached to it and communicatedwith the underlying abscess in seven cases (Figs.1 and 2).In those cases with generalized enlargement the

testis was replaced by rubbery, firm, creamy-yellow lobulated tissue, from which the infarctcould be clearly distinguished (Fig. 3). The cordwas thickened as in the first group with yellowfoci scattered in dense white fibrous tissue.

Histology

All sections were stained with haematoxylin and

Fig. 1 Sagittal slice of orchidectomy specimenshows necrosis of most of the testis. A thin rim ofsurviving parenchyma is seen at the bottom right-handcorner of the photograph.

The fibrous thickening of the cord is well shown,and attached scrotal skin is indistinctly visible at theleft-hand side of the specimen.

Case 11: histology of epididymis is shown inFigures 9 and 10.

i; i

0 2 1 212'2t3z214 5

Fig. 2 Horizontal slice of orchidectomy specimenshows a small sharply demarcated zone ofnecrosis in theanterior part of the testis. Note the gross thickening ofthe tunica. The viable testis showedfibroticseminiferous tubules.

Case 12: histology of border of necrotic zone isshown in Figure 4.

111I111jol'Will

copyright. on M

arch 28, 2021 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.23.8.668 on 1 N

ovember 1970. D

ownloaded from

670 D. O'B. Hourihane

-Ig. 3.

Fig. 4.

Fig. 3 Testis enlarged by firm, pale tissue, with adistinct zone ofnecrosis in the lowerpart ofthe specimen.Cord grossly thickened.

Case 13: history of testicular swelling for 64 days.Histology in Figs. 6 and 7 shows a mixture ofgranulomatous and fibrotic tubules.

Fig. 4 Area of necrosis of tubules at the top of thephotograph, with fibrotic, viable tubules at the foot.A thin haemorrhagic zone between the two areas runsdiagonally across the photograph. Specimen is shownin Figure 2.

Case 12: haematoxylin and eosin x 54.

Fig. 5 Outline of a seminiferous tubule surroundedby inflammatory infiltrate.

Case 17: history 42 days. H and E x 750.

Fig. 5.

copyright. on M

arch 28, 2021 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.23.8.668 on 1 N

ovember 1970. D

ownloaded from

Infected infarcts of the testis: A stuidy of 18 cases preceded by pyogenic epididymoorchitis

Fig. 6.

Fig. 8.

Fig. 7.

Fig. 6 Fibrotic seminiferous tubules, lined by a singlelayer of mononuclear cells. These cells usually con-tained lipid, and are probably modified Sertoli cells.Same case as in Figures 3 and 7.

Case 13: history 64 days. H and E x 180.

Fig. 7 Intratubular granulomata, with intenseinterstitial inflammation.

Case 13: history 64 days. H and E x 180.

Fig. 8 A single seminiferous tubule lies vertically inthephotograph, and is filled by large, pink, mononuclearcells. Multinucleated giant cells can be seen, andpolymorphs are identifiable within the upper and lowerends of the tubule.

Case 17: history 42 days. H and E x 180.

Fig. 9 Interstitial histiocytic and lymphocyticinfiltrate between epididymal tubules. Specimen isshown in Figure 1.

Case 11:HandE x 180.

Fig. 10 Intratubular exudate in epididymis. Noteperitubular histiocytes.

Case 11: Hand E x 300.

671copyright.

on March 28, 2021 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.23.8.668 on 1 Novem

ber 1970. Dow

nloaded from

eosin, and all suspected vascular lesions werestained with Sheridan's method for elastic tissue.Ziehl-Neelsen stains were performed on allgranulomatous specimens. Brown pigment wasstained with Perls's acid ferrocyanide method forhaemosiderin, and with a long Ziehl-Neelsenmethod for acid fastness. Frozen sections werestained for fat with oil red 0.There was again a general resemblance between

the cases. No normal tubules or spermatozoawere identified in any specimen. In the testes ofnormal size the central necrotic zone showed amixture of infarction and suppuration (Figs. 4and 5) with polymorphs infiltrating necrotictubules, the outlines of which were still detectable(Fig. 5). Occasional haematoidin crystals wereseen within the necrotic zones. A thin haemor-rhagic border was present separating necroticfrom viable tubules in some specimens (Fig. 4).The surviving peripheral area of testis showedfibrotic tubules lined by Sertoli cells with occa-sional intratubular granulomatous zones (Fig. 6).The larger testes were swollen by a granulo-

matous intratubular reaction (granulomatousorchitis, Figs. 7 and 8) with some fibrotic tubulesin peripheral areas, but their necrotic zones weresimilar to those of the smaller testes.

Inflammation of the epididymis was present inall cases. In many it was purely histiocytic whilein others polymorphs were scattered in the histio-cytic infiltrate and in the neighbouring fibroustissue. Exudate was always present both betweenand within tubules (Figs. 9 and 10), althoughtubules were difficult to identify in some speci-mens. The cord was commonly thickened byfibrous tissue and scattered foci of lymphocytesand histiocytes, while dilated lymphatics andsclerosed blood vessels were prominent.A feature of 17 of the specimens was the

presence of totally occluded veins in the epidi-dymis and cord (Fig. 11). Many of these containedhaemosiderin granules and were heavily in-filtrated with plasma cells and lymphocytes (Fig.12) while others were recanalized and showedimmature collagen between the new channels(Figs. 13 and 14). Endophlebitis (Fig. 15) andextreme narrowing of veins by intimal collagenwere common findings (Fig. 16).Frozen sections were available from nine

specimens, including sections of epididymis andtestis in each case. Lipid was demonstrated by oilred 0. Large aggregates of lipid-laden histiocyteswere present in the epididymis but the venousocclusions were uniformly negative for fat. Manyof the seminiferous tubules were lined by lipid-containing cells. In general, those lined by a singlelayer contained more lipid than those filled bygranuloma. Interstitial lipid was focally presentbut not prominent.

Foci of intracellular brown pigment were foundin the interstitium of the epididymis and in occa-sional seminiferous tubules; most of this pigment

672 D. O'B. Hourihane

Fig. 9.

Fig. 10.

copyright. on M

arch 28, 2021 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.23.8.668 on 1 N

ovember 1970. D

ownloaded from

673 Infected infarcts of the testis: A study of 11% cases

44

S. ~ ~ ij ~

q' '34 ~ Vt V

A ~ 1! *4.xp.2~~~~~~~~I

Aw.,-I-N *

preceded by pyogenic epididymoorchitis

Fig. 13 Multiple channels of recanalized vein.Case 17: history 42 days. Sheridan's elastic stain,

counterstained with van Gieson stain x 180.

Fig. 11 Totally occluded vein in spermatic cord.Heavy cellular infiltrate occupies lumen.

Case 7: history 51 days. H and E x 750.

w * qllD #~;;'-

~. %# . ... F .E

4"4*c /11 * /fAS,A> f-

A * 4

Fig. 12 Higher magnification of intravascular tissuein Figure 11. Largely lymphoid cells with somehistiocytes and intercellular connective tissue.

Case 7: history 51 days. H and E x 1,200.

Fig. 14 Four small occluded veins in spermatic cord.Interstitial and intraluminal lymphoid infiltrate.

Case 14: history 36 days. H and E x 180.

. . . . . 9%

copyright. on M

arch 28, 2021 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.23.8.668 on 1 N

ovember 1970. D

ownloaded from

674 D. O'B. Hourihane

Fig. 15 Lymphoid intimal infiltrate of vein.Case 17: history 42 days. H and E x 750.

stained as lipofuscin of ceroid type, but in theepididymis some stained as haemosiderin.The interstitium of the testes showed an intense

infiltrate of plasma cells, lymphocytes, and histio-cytes in those with a granulomatous response, butthe areas of fibrotic tubules showed a minimalinflammatory infiltrate. No interstitial Leydigcells were identified in any specimen. Giant cellswere exceedingly rare within granulomatoustubules, being seen in only two cases, but a centralcollection of polymorphs was commonly presentwithin the lumen of the granulomatous tubules(Fig. 8).No spermatozoa were seen in any specimen.

Despite inflammation of the epididymis in allcases, with destruction of some tubules andaggregates of polymorphs and lipid-laden histio-cytes, extravasated spermatozoa or sperm headswere not seen.

Discussion

These lesions bear close similarities between eachother and show a mixture of infarction and in-fection of the testes. The appearances are, inmany respects, reminiscent of renal papillarynecrosis in so far as in each tubules undergo

Fig. 16 Severe intimal fibrosis of vein.Case 17: history 42 days. H and E x 750.

necrosis and are infiltrated by polymorphs andyet tubular outlines remain.The clinical histories and presentation suggest a

banal infection of the bladder which subsequentlyspreads to the epididymis. The tunica breaksdown and allows the formation of a dischargingscrotal sinus. There is no fundamental differencebetween patients with granulomatous and non-granulomatous lesions, and in four instancestransitions between fibrotic and granulomatoustubules could be seen, supporting the suggestionthat granulomatous orchitis results from a chronicpyogenic infection (Lynch, Eakins, and Morrison,1968).The role of the venous occlusions is of great

interest; the fact that most were organized or inthe process of organization means that they wereabout four to six weeks old. They may haveoccurred as thrombi at the time of the initialepididymoorchitis, either as a result of suppura-tive involvement or of venous stagnation, and ledto an area of venous infarction which subsequent-ly became infected, or they may have been manyyears old, following previous infections or thetrauma of surgery (ipsilateral hernial repair inthree cases). There is little evidence in favour ofsuppurative destruction of vein walls as in onlyone case was there circumstantial support for thisin the form of scarring of the medial coat, and thepresence of inflammatory cells within most of the

copyright. on M

arch 28, 2021 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.23.8.668 on 1 N

ovember 1970. D

ownloaded from

Infected infarcts of the testis: A study of 18 cases preceded by pyogenic epididymoorchitis

venous occlusions suggests that they were notvery old, a conclusion supported by the absenceof organized pink-staining collagen with vanGieson stain. The presence of necrotic tubuleswith an inflammatory or haemorrhagic reactionat the edge of the zone of necrosis would alsosuggest that these lesions were weeks rather thanyears old.Thrombosis during the course of epididymo-

orchitis would seem to be the most likely explana-tion for these vascular lesions, as the interval ofthree to nine weeks between onset and orchi-dectomy would allow time for organization totake place. Although the turgid friable organ ofvenous infarction was not found it is reasonableto suggest that the engorgement with blood hadbeen cleared by cellular activity before theorchidectomy.

Ischaemia has been proposed as the cause ofgranulomatous orchitis (Dreyfuss, 1956), andmany published series include cases with largeareas of testicular necrosis similar to thosedescribed here (Spjut and Thorpe, 1954; Lynchet al, 1968). Since 1964 we have seen two cases ofgranulomatous orchitis in which necrotic areas

were not evident, but these have not shownvenous occlusions and it would appear thatvascular obstruction is related more to thedevelopment of testicular necrosis rather than tothe granulomatous inflammatory reaction.The grouping of granulomatous and non-

granulomatous testes allowed comparisons be-tween the two, and it became clear after histo-logical study that transitions between the twowere common and might be seen in a single testis.These findings support the idea of pyogenic in-fection rather than an immune mechanism as thecause of this unusual histological reaction.The history of previous repair of an ipsilateral

inguinal hernia is of interest, as the possibility ofvenous injury at this time must be considered.The data on the remaining patients may be inade-quate, but it would seem unwise to ascribeaetiological significance to the earlier surgery.

The absence of normal spermatogenesis in the

surviving tubules probably reflects a general sup-pression as a result of the local inflammation, andis responsible for the absence of extravasatedspermatozoa around the inflamed epididymaltubules. This, together with the absence of meta-plastic epithelium and the indistinctly granulo-matous nature of the epididymal inflammation.contrasts with the findings in patients withspermatic granuloma (Glassy and Mostofi, 1956).The presence of lipid and lipofuscin pigment of

ceroid type within cells in granulomatous orchitisand spermatic granuloma has led Phillips (1961)to postulate that spermatozoa have given rise toeach of these inflammatory reactions. Similarfindings were present in some of the cases pre-sented here, but the evidence for a pyogenicinfection seems very strong, and it would appearmore likely that the lipid and pigment haveresulted secondarily from necrotic cells.From this study it seems reasonable to conclude

that when an abscess develops in the testis of apatient with epididymoorchitis this is associatedwith multiple occlusions of veins within the epi-didymis and spermatic cord, and a causal rela-tionship probably exists between the abscess andthe vascular lesions. The abscess usually has thehistology of an infected infarct, and is generallysituated at the antihilar border of the gonad.

My thanks are due to Mr F. Murray for meticu-lous care with the photographs, and to Miss J.Fowler for secretarial assistance.

References

Dreyfuss, W. (1954). Acute granulomatous orchitis. J. Urol.(Baltimore), 7, 483-487.

Glassy, F. J., and Mostofi, F. K. (1956). Spermatic granuloma ofthe epididymis. Amer. J. clin. Path., 26,1303-1313.

Lynch, V. P., Eakins, D., and Morrison, E. (1968). Granulomatousorchitis. Brit. J. Urol., 40, 451-458.

Phillips, D. E. (1961). Lipid granuloma of the testis and epidi-dyrmis. Brit. J. Urol., 33, 448-452.

Spjut. H. J., and Thorpe, J. D. (1956). Granulomatous orchitis.Amer. J. clin. Path., 26,136-145.

675copyright.

on March 28, 2021 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.23.8.668 on 1 Novem

ber 1970. Dow

nloaded from