undescended testicle removal does not extend life

[malernee]

408 The Veterinary Record, March 30, 2002
Papers & Articles
Decision analysis tree for deciding whether
to remove an undescended testis from a
young dog
M. A. J. Peters, F. J. van Sluijs
A decision analysis tree was constructed to estimate the life span which might be expected for a dog
presented with cryptorchidism at one year of age if it underwent a preventive orchidectomy or if it did not.
The tree was constructed by using risk factors associated with cryptorchidism and data on the prevalence of
complications after surgery from the literature. The expected life span without an orchidectomy was not
significantly different from the expected life span after an orchidectomy.
THE reported prevalence of cryptorchidism in dogs ranges
from 0·8 per cent (Priester and others 1970) to 9·7 per cent (Reif
and Brodey 1969) depending on the population studied, and
the condition is an important risk factor for the development
of testicular tumours. The prevalence of Sertoli cell tumours is
26 times higher and the prevalence of seminomas is 15 times
higher in undescended testes than in scrotal testes (Hayes and
others 1985). The prevalence of Leydig cell tumours in undescended
testes is similar to (Hayes and Pendergrass 1976) or
lower than that in scrotal testes (Reif and Brodey 1969, Lipowitz
and others 1973,Nielsen and Lein 1974,Nieto and others 1989).
Testicular tumours may induce a feminising syndrome
consisting of gynaecomastia, atrophy of the contralateral
testis, a pendulous prepuce, attractiveness to other male dogs,
bilateral symmetrical alopecia and, in severe cases, bone marrow
depression. The bone marrow depression may be sufficiently
severe to cause death, and death may also be caused by
metastases (Nielsen and Lein 1974, Theilen and Madewell
1987, Nielsen and Kennedy 1990).
An orchidectomy eliminates the risk of tumour-associated
morbidity and mortality, but introduces the risk of morbidity
and mortality due to the surgical procedure; these risks
have to be balanced before a decision can be made. Thirteen
years ago Bosch and others (1989) concluded that it was better
to do nothing, on the basis of a decision analysis in which
the life spans of the dogs and the mortality and morbidity of
preventive orchidectomy were estimated. The availability of
new data on the life span of dogs (Eichelberg and Seine 1996,
Bonnett and others 1997, Patronek and others 1997) and the
mortality and morbidity of surgical procedures in dogs
(Clarke and Hall 1990, Brunnberg and others 1997) provides
the opportunity to make a new decision analysis.
MATERIALS AND METHODS
Decision analysis
A two-legged decision tree was constructed as described by
Weinstein and Fineberg (1980),with preventive orchidectomy
and doing nothing as choices (Fig 1). The utility of each
choice, the expected utility value (EUV), was expressed in
terms of expected survival time. For the preventive orchidectomy
the following endpoints were used: death due to anaesthetic
complications,wound dehiscence, seroma/haematoma,
wound infection and uncomplicated recovery. For doing
nothing the endpoints were: death due to metastases, death
due to bone marrow depression, feminisation without bone
marrow depression, alopecia, and no complications due to a
Veterinary Record (2002)
150, 408-411
M. A. J. Peters, PhD,
F. J. van Sluijs, DVM, PhD,
DiplECVS, Department of
Clinical Sciences of
Companion Animals,
Faculty of Veterinary
Medicine, Utrecht
University, PO Box 80.154,
3508 TD Utrecht,
The Netherlands
Correspondance to
Dr van Sluijs
The Veterinary Record, March 30, 2002 409
testicular neoplasm. The expected survival time for the endpoints
was calculated separately for Sertoli cell tumours, seminomas
and Leydig cell tumours. Mixed tumours were
subdivided according to the types of tumour present.
Probabilities
Probabilities (Table 1) were derived from the literature.Mixed
tumours were not counted as a separate category, but their
components were considered as pure tumours and added to
the respective groups.
The probabilities of particular testicular tumours arising
(0·034, 0·024 and 0·005 for Sertoli cell tumours, seminomas
and Leydig cell tumours, respectively) were taken from a
retrospective study of 2912 cryptorchid dogs (Hayes and
others 1985).
The probability of death due to anaesthetic complications
was derived from a recent survey among 150 veterinary surgeons
from 53 small animal practices involving 20,814 anaesthetised
dogs. In this study, the mortality in the American
Society of Anesthesiologists categories 1 and 2 (healthy, or
with slight disease causing no incapacity) was 0·001 (Clarke
and Hall 1990).
The probabilities of wound dehiscence and seroma/
haematoma (0·023 and 0·05, respectively) were taken from a
retrospective study of 8724 dogs by Brunnberg and others
(1997); the probability of wound infection (0·032) was calculated
by averaging the results of this study and of a study of
1992 dogs by Vasseur and others (1988).
The probabilities of death due to metastases were calculated
for each tumour (0·05, 0·06 and 0·02 for Sertoli cell tumours,
seminomas and Leydig cell tumours, respectively) by averaging
the results of 10 retrospective studies (Scully and Coffin
1952, Brodey and Martin 1958, Cotchin 1960, Dow 1962,
Gembardt 1971, Lipowitz and others 1973, Kasbohm 1974,
Nielsen and Lein 1974,Nieto and others 1989,Reifinger 1998).
The probabilities of bone marrow depression (0·097, 0·001
and 0·007 for Sertoli cell tumours, seminomas and Leydig cell
tumours, respectively) were calculated by averaging the results
of four retrospective studies (Kasbohm and Saar 1975, Sherding
and others 1981,Morgan 1982, Post and Kilborn 1987).
The probabilities of feminisation without bone marrow
depression (0·603, 0·008 and 0·043 for Sertoli cell tumours,
seminomas and Leydig cell tumours, respectively) were calculated
by averaging the results from four studies in cryptorchid
dogs (Brodey and Martin 1958,Reif and Brodey 1969,
Weaver 1983, Nieto and others 1989).
The probabilities of alopecia without feminisation (0·22,
0·02 and 0·1 in Sertoli cell tumours, seminomas and Leydig cell
tumours, respectively) were calculated by averaging the results
of 13 retrospective studies (Scully and Coffin 1952,Brodey and
Martin 1958, Dow 1962, Reif and Brodey 1969, Gembardt
1971, Lipowitz and others 1973, Damodaran and others 1974,
Kasbohm 1974,Nielsen and Lein 1974, Steele and others 1977,
Weaver 1983, Post and Kilborn 1987,Nieto and others 1989).
Expected utility values
The average life span of a dog was estimated to be 10 years, on
the basis of a study by Eichelberg and Seine (1996). The average
ages at which Sertoli cell tumours and seminomas occur
Papers & Articles
Expected utility values Years Probabilities %
Average life span 10·00 Mortality due to anaesthesia 0·11
Age at the time of decision 1·00 Dehiscence 2·30
Age for SCT 8·20 Seroma/haematoma 5·00
Age for SEM 8·50 Infection 3·15
Age for LCT 10·00 SCT
Reduction for feminisation 0·50 Metastases 5·00
Reduction for alopecia 0·10 Bone marrow depression 9·72
Reduction for dehiscence 0·30 Feminisation 60·28
Reduction for seroma/haematoma 0·20 Alopecia 22·00
Reduction for infection 0·10 SEM
Metastases 6·00
Bone marrow depression 0·14
Alopecia 2·00
LCT
Metastases 2·00
Bone marrow depression 0·69
Feminisation 4·31
Alopecia 10·00
TABLE 1: Expected utility values in years and percentage probabilities of Sertoli cell tumours
(SCT), seminomas (SEM) and Leydig cell tumours (LCT) and their complications in cryptorchid
dogs. The average life span, the ages at which SCT, SEM and LCT occur and their probabilities
were derived from the literature (see text). Reductions in life span due to alopecia,
dehiscence, seroma and infection have been estimated
FIG 1: Decision tree for preventive orchidectomy in a cryptorchid dog. The tree has two branches: ‘operate’ and ‘wait’. The branch
‘wait’ has four possibilities: ‘Sertoli cell tumour’, ‘seminoma’, ‘Leydig cell tumour’ and ‘no tumour’. The calculation of the expected
utility values (EUV) (expected survival time in years when the decision is made) of each branch starts on the right and proceeds to
the left. The EUV of the possibilities ‘operate’, ‘Sertoli cell tumour’, ‘seminoma’ and ‘Leydig cell tumour’ are calculated by summing
the products of the EUV and the probability of each possible outcome. For the branch ‘operate’, this yields the EUV of this branch. The
EUV of the branch ‘wait’ is calculated by summing the products of the EUV and the probability of each possibility. The probabilities
of ‘Sertoli cell tumour’, ‘seminoma’ and ‘Leydig cell tumour’ are derived from the literature. The probability of the remaining
possibility ‘no tumour’ is 1 minus the sum of the other probabilities in the same branch. Pr Probability,  Decision,  Chance
410 The Veterinary Record, March 30, 2002
in cryptorchid dogs (8·2 years and 8·5 years, respectively) were
derived from a study by Hayes and Pendergrass (1976). The
average age at which Leydig cell tumours occur has been
reported to be 11·2 years (Nielsen and Lein 1974), beyond the
estimated life span, which was therefore used instead. The
expected survival time (Table 1) was corrected for morbidity
by subtracting time from the life span, the time subtracted
being estimated on the basis of the severity of the complication:
wound dehiscence –0·3 years, seroma/haematoma –0·2 years,
wound infection –0·1 year, feminisation –0·5 years, and alopecia
–0·1 year. The expected survival time was calculated as the
life span expected from the age at which the decision was made
which, for the purposes of this study, was taken to be one year.
RESULTS
The EUV for doing nothing was 8·98 years, compared with
8·97 years for preventive orchidectomy (Fig 1).
DISCUSSION
This decision analysis tree for a young cryptorchid dog
showed that the risk of anaesthetic or surgical complications
was similar to the risk of morbidity or mortality due to a testicular
tumour. This result is consistent with the conclusions
of Bosch and others (1989) although the general consensus
(Dunn and others 1968,Wallace and Cox 1980, Romagnoli
1991) has been that a preventive orchidectomy is beneficial.
Like other models, this analysis includes a number of assumptions
and simplifications. The data concerning the prevalence
of testicular tumours was derived from a large population of
cryptorchid dogs (Hayes and others 1985), but studies in
smaller groups that included relatively large numbers of old
dogs indicated that the prevalence of testicular tumours was
much higher (Mosier 1989). These data were not used
because in the present study the decision whether to operate
was made at the age of one year.
The average age at which Leydig cell tumours have been
reported is 11·2 years (Nielsen and Lein 1974), 1·2 years more
than the estimated life span used in this study.As a result, the
probability of morbidity and mortality associated with these
tumours may have been overestimated. The probabilities of
alopecia, metastases and feminisation may also have been
overestimated. Mixed tumours were added to each tumour
type present. As a result, alopecia, feminisation and metastases
were associated with all the types of tumour present
whereas they may be caused by only one component.
Feminisation and metastases were considered to occur independently
of each other, but theoretically they may affect the
same dog. These potential overestimations were accepted
because they were likely to be small and were unlikely to
change the outcome of the analysis.
The prevalence of feminisation is higher in cryptorchid dogs
than in dogs with scrotal testes (Reif and Brodey 1969, Nieto
and others 1989). Only data from cryptorchid dogs were used
to calculate the probability of feminisation in dogs with the different
types of testicular tumour. The prevalence of alopecia
has not been reported to differ between cryptorchid dogs and
dogs with scrotal testes. The probabilities for these complications
were derived from studies in which cryptorchidism was
either absent or not reported as a separate condition.
Bone marrow depression is generally attributed to the
excessive secretion of oestrogens by a tumour and is associated
with feminisation (Kasbohm 1974, Sherding and others
1981, Morgan 1982, Suess and others 1992). It was assumed
that these conditions do not occur separately, although one
study reported that two of 10 dogs with a testicular tumour
and bone marrow depression had no overt signs of feminisation
(Morgan 1982); however, additional tumours (adrenocortical
carcinoma and bronchiolar carcinoma) were found
in one of the dogs which might have contributed to the bone
marrow depression.
In this analysis, a possible breed predisposition for testicular
tumours was not taken into account. It is generally
believed that there is no breed predisposition for the development
of testicular neoplasms or abnormalities of the spermatogenic
epithelium (Nielsen and Kennedy 1990), except for
seminomas which have been reported to occur three times
more often in the boxer than in other breeds (Howard and
Nielsen 1965). If the probability of seminoma had been multiplied
by three, the EUV of both options remained the same.
There are no data on the prevalence of complications after
preventive orchidectomy in large groups of male dogs.
Instead, data on the complication rates after elective ovariohysterectomy
in bitches were used (Vasseur and others 1988,
Brunnberg and others 1997). Both procedures can be characterised
as a caudal midline coeliotomy in a healthy animal
and are technically similar.
Probably the greatest simplification is the estimate of the
average life span of dogs. It is well known that small dogs live
longer than large dogs (Bronson 1982, Eichelberg and Seine
1996, Li and others 1996, Patronek and others 1997). If the
expected life span exceeded 10·4 years, the analysis would no
longer favour doing nothing. Preventive orchidectomy may
therefore be considered in dogs with an expected life span
over 10·4 years, but the difference between the two options
will remain small.
Neoplastic changes in undescended testes may be complicated
by testicular torsion (Pearson and Kelly 1975).This complication
may introduce an additional risk factor, but the
number of reported cases is too small for its probability and
the associated EUV to be estimated reliably. For this reason,
testicular torsion was not considered in the decision analysis.
A decision tree is only a tool to help the clinician to estimate
the risks of a treatment. Other important factors leading
to a decision include the behavioural changes induced by
castration and the increased risk of obesity. Dog owners
should be informed about these side effects.Another consideration
is that cryptorchidism in dogs is considered to be an
inherited trait (Romagnoli 1991), and castration may therefore
be advisable to prevent breeding from affected animals.
Because the outcome of the decision tree indicated that
orchidectomy would not make a significant difference to the
dog’s life expectancy, it would seem advisable to monitor a
cryptorchid dog frequently for development of a testicular
tumour and to operate only when one is suspected.
References
BONNETT, B. N., EGENVALL, A., OLSON, P. & HEDHAMMAR, A. (1997)
Mortality in insured Swedish dogs: rates and causes of death in various
breeds. Veterinary Record 141, 40-44
BOSCH, A. G., VAN SLUIJS, F. J. & VAN NES, J. J. (1989) Medical decision
analysis in veterinary practice. An introduction with reference to the problem:
to operate or not in cryptorchism in dogs. Tijdschrift Diergeneeskunde
114, 369-375
BRODEY, R. S. & MARTIN, J. E. (1958) Sertoli cell neoplasms in the dog.
Journal of the American Veterinary Medical Association 133, 249-257
BRONSON,R. T. (1982) Variation in age at death of dogs of different sexes and
breeds. American Journal of Veterinary Research 43, 2057-2059
BRUNNBERG, L., RICHTER, K., KOHN, B. & HERMANNS, W. (1997)
Complications in wound healing in dogs and cats – a retrospective analysis
(1990-1994). Kleintierpraxis 42, 535-553
CLARKE, K.W. & HALL, L.W. (1990) A survey of anaesthesia in small animal
practice: AVA/BSAVA report. Journal of the Association of Veterinary
Anaesthetists 17, 4-10
COTCHIN, E. (1960) Testicular neoplasms in dogs. Journal of Comparative
Pathology 70, 232-252
DAMODARAN, S., RAMACHANDRAN, P. V. & PARTHASARATHY, K. R.
(1974) Testicular neoplasms in canines. Cheiron Tamil Nadu Journal of
Veterinary Science and Animal Husbandry 3, 107-111
Papers & Articles
The Veterinary Record, March 30, 2002 411
(c) British Veterinary Association. All rights reserved
DOW, C. (1962) Testicular tumours in the dog. Journal of Comparative
Pathology 72, 247-265
DUNN, M. L., FOSTER,W. J. & GODDARD, K.M. (1968) Cryptorchidism in
dogs: a clinical survey. Journal of the American Animal Hospital Association
4, 180-182
EICHELBERG, H. & SEINE, R. (1996) Life expectancy and cause of death in
dogs. I. The situation in mixed breeds and various dog breeds. Berliner
Münchener Tierärztliche Wochenschrift 109, 292-303
GEMBARDT,C.H. (1971) Beitrag zur Histologie hormonell aktiver und inaktiver
Hodenneoplasien des Hundes. Berlin, Freien Universität Berlin
HAYES, H. M. & PENDERGRASS, T. W. (1976) Canine testicular tumors:
epidemiological features of 410 dogs. International Journal of Cancer 18, 482-
487
HAYES, H. M., WILSON, G. P., PENDERGRASS, T.W. & COX, V. S. (1985)
Canine cryptorchidism and subsequent testicular neoplasia: case-control
study with epidemiologic update. Teratology 32, 51-56
HOWARD, E. B. & NIELSEN, S. W. (1965) Neoplasia in the boxer dog.
American Journal of Veterinary Research 26, 1121-1131
KASBOHM, C. (1974) Hodenneoplasien des Hundes aus klinischer Sicht.
Tierärtzliche Praxis 2, 435-444
KASBOHM, C. & SAAR, C. (1975) Östrogenbedingte Knochenmarkschäden
bei Hunden mit Hodenneoplasien. Tierärtzliche Praxis 3, 225-229
LI, Y., DEEB, B., PENDERGRASS,W. & WOLF, N. (1996) Cellular proliferative
capacity and life span in small and large dogs. Journals of Gerontology A,
Biological Sciences and Medical Sciences 51, B403-B408
LIPOWITZ, A. J., SCHWARTZ, A., WILSON, G. P. & EBERT, J. W. (1973)
Testicular neoplasms and concomitant clinical changes in the dog. Journal
of the American Veterinary Medical Association 163, 1364-1368
MORGAN, R. V. (1982) Blood dyscrasias associated with testicular tumours
in the dog. Journal of the American Animal Hospital Association 18, 970-975
MOSIER, J. E. (1989) Effect of aging on body systems of the dog. Veterinary
Clinics of North America: Small Animal Practice 19, 1-12
NIELSEN, S.W. & KENNEDY, P. C. (1990) Tumours of the genital system. In
Tumours in Domestic Animals. Ed J. E. Moulton. Berkeley, University of
California Press. pp 479-517
NIELSEN, S.W. & LEIN, D. H. (1974) Tumours of the testis. Bulletin of the
World Health Organization 50, 71-78
NIETO, J. M., PIZARRO, M., BALAGUER, L. M. & ROMANO, J. (1989) Canine
testicular tumours in descended and cryptorchid testes.Deutsche Tierärztliche
Wochenschrift 96, 186-189
PATRONEK, G. J., WATERS, D. J. & GLICKMAN, L. T. (1997) Comparative
longevity of pet dogs and humans: implications for gerontology research.
Journals ofGerontology A, Biological Sciences and Medical Sciences 52, B171-B178
PEARSON, H. & KELLY, D. F. (1975) Testicular torsion in the dog: a review of
13 cases. Veterinary Record 97, 200-204
POST,K.& KILBORN, S. H. (1987) Sertoli cell tumor: a medical records search
and literary review. Canadian Veterinary Journal 28, 427-431
PRIESTER, W. A., GLASS, A. G. & WAGGONER, N. S. (1970) Congenital
defects in domesticated animals: general considerations.American Journal of
Veterinary Research 31, 1871-1879
REIF, J. S. & BRODEY, R. S. (1969) The relationship between cryptorchidism
and canine testicular neoplasia. Journal of the American Veterinary Medical
Association 155, 2005-2010
REIFINGER, N. (1998) Statistiche Untersuchungen zum Vorkommen von
Hodenturmoren bei Haussaugetieren. Zentralblatt für Veterinärmedizin A 35,
63-72
ROMAGNOLI, S. (1991) Canine cryptorchidism. Veterinary Clinics of North
America: Small Animal Practice 21, 533-544
SCULLY, R. E. & COFFIN,D. L. (1952) Canine testicular tumors. Cancer 5, 592-
605
SHERDING, R. G., WILSON, G. P. & KOCIBA, G. J. (1981) Bone marrow
hypoplasia in eight dogs with Sertoli cell tumor. Journal of the American
Veterinary Medical Association 178, 497-501
STEELE,W. B.,MCNEIL, P.,MUNRO, C.D.,RENTON, J. P. & DOUGLAS,T. A.
(1977) Testicular tumours in dogs. Veterinary Record 101, 142
SUESS, R. P. Jr, BARR, S. C., SACRE, B. J. & FRENCH, T.W. (1992) Bone marrow
hypoplasia in a feminized dog with an interstitial cell tumor. Journal of
the American Veterinary Medical Association 200, 1346-1348
THEILEN, G. H. & MADEWELL, B. R. (1987) Tumours of the male genital
system. In Veterinary Cancer Medicine. 2nd edn. Eds G. H. Theilen, B. R.
Madewell. Philadelphia, Lea & Febiger. pp 591-600
VASSEUR, P. B., LEVY, J.,DOWD, E. & ELIOT, J. (1988) Surgical wound infection
rates in dogs and cats: data from a teaching hospital. Veterinary Surgery
17, 60-64
WALLACE, L. J. & COX, V. S. (1980) Canine cryptorchidism. In Current
Veterinary Therapy VII. Small Animal Practice. Ed R.W. Kirk. Philadelphia,
W. B. Saunders. pp 1244-1246
WEAVER, A.D. (1983) Survey with follow-up of 67 dogs with testicular Sertoli
cell tumours. Veterinary Record 113, 105-107
WEINSTEIN, M. C. & FINEBERG, H. V. (1980) Clinical Decision Analysis.
Philadelphia,W. B. Saunders
Papers & Articles

[malernee]

study below found 0 Leydig tumors in retained testicals and 22 in desended


1: Dtsch Tierarztl Wochenschr. 1989 Apr;96(4):186-9. Related Articles, Links


Canine testicular tumors in descended and cryptorchid testes.

Nieto JM, Pizarro M, Balaguer LM, Romano J.

A comparative study of canine testicular tumors in descended and cryptorchid testes was performed in order to associate pathologic features with clinical signs. Observation on 80 canine testicular tumors revealed the following distribution: in undescended testes, 4 tumors were classified as Seminomas (mean age: 8.8 +/- 3 years) and 14 as Sertoli cell tumors (8.7 +/- 1.7 years); in descended testes, 21 tumors were classified as seminomas (8.8 +/- 3 years), 13 as Sertoli cell tumors (9.8 +/- 1.8 years), 22 as Leydig cell tumors (11.5 +/- 2 years), 5 multiple primary tumors and 1 as an immunoblastic lymphoma. Histological features were studied and correlated with other clinical parameters: feminization manifestations, prostatic disease, perineal hernia and perianal gland tumor.

PMID: 2565801 [PubMed - indexed for MEDLINE]

[malernee]

some cancers and other problems are seen more frequently in castrated dogs.

http://evidencebasedvet.com/forum/viewtopic.php?t=15

[malernee]

The Canadian veterinary journal. La revue vétérinaire canadienne
Volume 47 | Issue 8 (August 2006)
Extratesticular interstitial and Sertoli cell tumors in previously neutered dogs and cats: a report of 17 cases
Can Vet J. August 2006;47(8):763-6.
Angela L Doxsee, Julie A Yager, Susan J Best, Robert A Foster
Department of Pathobiology, Ontario Veterinary College, University of Guelph, Guelph, Ontario.

Abstract
Primary neoplasms derived from testicular tissue and in an extratesticular location are extremely rare. Clinical and surgical information was collected and verified from 15 different submitting practices for 12 dogs and 5 cats that spontaneously developed neoplasms of testicular origin after castration. Eleven dogs had Sertoli cell tumors in an extratesticular location. One dog and all 5 cats had an extratesticular interstitial cell tumor. Six animals (1 dog, 5 cats) had developed secondary sexual characteristics that reversed after removal of the tumor. All had a palpable mass in the scrotum or at the site of the original prescrotal incision. No animals died of neoplasia-related disease and no metastases were identified. Several possibilities, including the presence of embryological ectopic tissue or the presence of testicular tissue transplanted during castration, are considered as causal.