by Richard B. Krueger, Michael H. Vvechsler, and Meg S. Kaplan

This chapter aims to review the rationale and data supporting bilateral orchiectomy as a possible treatment modality for sex offenders. Although the idea of archiectomy, or surgical castration, evokes strongly negative emotional responses in many individuals, several reviews of biological treatments for sex offenders examining recidivism data from European undies have concluded that castration had very significant positive effects on recidivism (Bradford, 1985; Freund, 1980; Meyer & Cole, 1997). A recent meta-analysis by Losel and Schmucker (2005) reviewed 69 studies and found that surgical castration and hormonal medication showed larger effects on sexual recidivism than psychosocial intervention. Finally, a recent review of pharmacotherapy in the treatment of sex offenders suggested that it might not be surprising if a “comeback” for surgical castration would occur (Rosier & Witztum, 2000) because of its efficacy and the low rates of recidivism reported by outcome studies.

Orchiectomy was selected as the title for this chapter because it more properly reflects the nature of the therapeutic procedure that will be discussed in depth. Orchiectomy (Hensyl, 1990, p. 1096) comes from the Greek, orchis, or testis, and ektome, or excision, and is referred to, alternatively, as orchidectomy, orchectomy, or testectomy, and consists of the removal of one or both testes.

The term “castration” comes from the Latin stem castrate meaning to castrate, prune, expurgate, deprive of vigor (Simpson & Weiner, 1989, p. 959) and has a broader meaning which includes mutilation, which undoubtedly has contributed to its negative associations. Castration for control, domination, punishment, mutilation, social advancement, and political reasons has been practiced since antiquity and has been the subject of several recent reviews (Ayalon, 1999; Marmon, 1995; Ringrose, 2003; Scholz, 2001; Taylor, 2000; Tougher, 2002; Tsai, 1996; Winslade, Stone, Smith-Bell, & Webb, 1998). Although castration is a term which is referred to in much of the literature cited in this article, unless otherwise indicated, it should be understood that insofar as can be determined bilateral orchiectomy was performed, and not a more extensive procedure.

Some of the literature also refers to “medical castration” by which is meant the use of hormonal or gonadotropin-releasing hormone agonist (CNRH) therapy to reduce testosterone to castrate levels. Finally. some of the literature on the treatment of prostate cancer refers to androgen deprivation therapy, which has the aim of reducing testosterone to castrate levels. and which can be accomplished by surgical or medical castration.



The first therapeutic castration based on a psychiatric indication was performed in 1892 by August Ford in Zurich for an “imbecilic” man who was on the verge of autocastration because of neurologic pains in his testes (Sturup, 1972). Denmark pioneered legalizing this treatment in 1929, followed hy Germany (1933), Norway (1934). Finland (1935), Estonia (1937), Iceland (1938), Latvia (1938), and Sweden (1944) (Bradford, 1985; Heim & Hutsch, 1979; Le Marie. 1956). Heim and Bursch (1979) reported that Switzerland, the Netherlands, and Greenland practiced this treatment without all)’ legal mandate, sometimes quite extensively. For instance, Heim (1981) reported that the procedure was widely used in Switzerland and that in the Zurich region alone more than 10,000 patients had been castrated for various psychiatric reasons since 1910. Heim and Hutsch (1979) reported that in Germany between 1934 and 1944 at least 2800 sex offenders were subjected to compulsory castration and that subsequently in West Germany, between 1955 and 1977, 800 sex offenders were castrated. They also reported 1100 cases of castration in Denmark since 1929. This procedure was also used in the Netherlands (Sturup, 1968a).

Castration has not been used in most countries for the treatment of sex offenders for several decades (Giis & Gooren. 1996), although Germany was reported (Wille & Beier, 1989) to have been conducting five castrations annually since 1980 and the Czechoslovakian republic was reported, in a paper presented at a conference in 1995 to have conducted 84 such procedures Since 1976 (Gijs & Gooren, 1996).

United States

In the late 1800s castration for criminals in the United States was more a subject of discussion in the penal literature than a practice, but many cases were reported (Gugliotta, 1998). Dr. Sharp in 1899 in Indiana began experiments on convicts (Le Marie 1956). Although described as castration, his procedure, in fact, involved only vasectomy to prevent the loss of secretions from the testes and thus preserve the loss of the “elixir of life” to promote well-being and health. More than 450 vasectomies were performed: on incarcerated men, both with and withont explicit legal sanction (Gugliotta, 1998).

Elsewhere in the United States castration was practiced in California from 1937 to 1948. A study reporting on the outcome of 40 castrated males was recently published (Weinberger et aI., 2005), but other reports suggest that total numher of inmates subjected to this procedure was in the neighborhood of 400 (Linsky, 1989).

More recently in 1996, in the United States, California became the first of nine states passing contemporary legislation authorizing the use of either medical or surgical castration for certain sex offenders (Scott & Holmberg, 2003). Of these nine states, four permit the use of medical castration only (Georgia, Montana, Oregon, and Wisconsin); four allow either medical castration or voluntary surgical castration (California, Florida, Iowa, and Louisiana) and only one (Texas) provides voluntary surgical castration as the only treatment option (Scott & Holmberg, 2003). While medical or surgical castration is described as mandatory in five of these stales (Scott & Holmberg, 2003), issues of informed consent are often not addressed by these statutes. Moreover, we have been able to locate only one published description from one of these states, Oregon, on the use of these statutes (Maletzky, Tolan, & McFarlan, 2006). This group reported on a 5-year follow-up of men referred under Oregon House Bill 2500 enacted in 1999. for evaluation as to whether medical treatment with medroxyprogesterone acetate (MPA) was indicated to reduce their risk. Two hundred seventy five men were evaluated, and sexual recidivism was established. Of these, 79 men were advised to receive MPA and did, in fact. received it; none of this group recidivated sexually. One hundred forty-one were not referred for MPA treatment and 14% recidivated sexually; 55 were advised to have MPA treatment, but for a variety of reasons did not receive it, and 18% of this group recidivated sexually. This study lends support to the efficacy of antiandrogen treatment in reducing sexual recidivism. The lack of other published reports or information from the other eight states suggests that these laws are not being utilized.


Aside from sporadic case reports of orchiectomy in the medical literature (Alexander, 1993; Joseph. 1993; Silcock. 1993) (of a British pedophile) or news media (CBS News, 2004) (of a pedophile in Texas) the authors know of only six recent cases, three of these were performed in Canada, two on sex offenders (Bradford, 2006, personal communication) and another for the treatment of autoerotic asphyxia (Fedoroff. 2006, personal communication), three were performed in Texas on individuals offered surgical castration under the recent Texas statute (Winslade. 2008, personal communication) Thus, the use and study of this procedure have fallen off extensively.

Ethical Issues and Use of Castration for Other Conditions

Castration has also quite continuously been the topic of many ethical debates and reviews (Alexander. 1993; Bingley, 1993; Bund, 1997; Cook, 1993; Eastman, 1993; Finch, 1993; Freund, 1980; Caensbauer, 1973; Gandhi, Purandare, & Lock. 1993; Heim & Hursch, 1979; Hicks, 1993; Icenogle, 1994; Joseph, 1993; Klerman. 1975; Silcock, 1993; Tancredi & Weisstllb. 1986: Taylor. 1993). In the United States, opponents of castration legislation cite First Amendment (concerns involving the protection of a person’s freedom of speech (including sexual expression); violation of the Eighth Amendment’s ban on cruel and unusual punishment; and violation of the Fourteenth Amcndment’s guarantee of due process and equal protection (subjecting prisoners to castration without adequate process of protection) (Scott & Holmberg, 2003).

However, although the use of castration is controversial for sex offenders, it has achieved the status of an accepted surgical treatment for a variety of medical conditions. The Nobel Prize in physiology or medicine was awarded to Charles Huggins in 1966 for his investigations during the 1940s into the role of sex hormones on prostate cancer and for his use of bilateral orchiectomy to treat patients with metastatic prostate cancer. (Corbin & Thompson, 2003; Huggins & Hodges. 2002). Although hormonal therapy and GNRH agonists havc largely supplanted the use of castration for prostate cancer, it is still a common urological procedme. Unilateral orchiectomy constitutes standard of care treatment for testicular cancer.

Furthermore, voluntary castration for a variety of motivations, including a feeling of control over one’s sexual urges and appetite, a sense of calmness, and cosmetic reasons, has been described with 23 of 234 respondents to an Internet survey in 2002, indicating that they had actually undergone castration for some of these reasons (Wassersug, 2004). A treatise on the psychological and other advantages of castration was published recently (Cheney. 2004).


Rat and Canine Studies

Research studies in rats demonstrate that castration results in a loss of sex drive and an abolishment of mating behavior. and that this can be restored by testosterone (Davidson, Stefanick. & Sachan, 1978). Beach (1976) provided a review and rationale for crossspecies comparison of sexual behavior and described a sequential reduction in ejaculation, intromission, and mounting in males of a variety of species following castration. He also summarized experiments on rats and dogs suggesting that previous sexual experience was n0t related to the variation in the effects of castration on animals (Beach 1970).

Primate Studies

The effects of caslmtion in primates have been reviewed by Dixson (1998, pp. 392-397). Chemical or physical castration has been studied in six species. with quite attenuated sexual functioning being uniformly demonstrated. For example, castration of sexually experienced adult male rheslIs monkeys results in a decline in sexual behavior, reversible with testos! erone (Dixson, 1998. pp. 392-393). In the first 2 to 4 weeks frequencies of ejaculation .md intromission are reduced; mounting behavior declines more gradually. However, there is marked individual variability with 5 out of 10 male rheslls monkeys studied by Phoenix (Phoenix, 1980; Phoenix. Slob, & Goy, 1973) still capable of intromission and three with ejaculatory responses, 1 year after the operation. Loy (1971) observed a castrated male rhesus monkey still capable of mounts, intromissions, and an ejaculatory pause chanlcteristic of intact males 7 years after operation.

Taken together, these data suggest that orchiectomy can markedly reduce sexual behavior in rodents and primates, but that its effects can be readily reversed by exogenous testosterone and there is much individual variation, with some animals having sexual functioning extinguished and others continuing with it.


Case Studies by Kinsey

Kinsey et al. (1953, pp. 731-745) reviewed a large amount of mostly anecdotal evidence on the effects of castration, including some of the European castration literature, and noted that there was large variability in the effects of the procedure. They described one male from their own studies who was married and normally sexually active 30 years after castration. They also noted that at 50 years of age, 7% of males were impotent and sexually unresponsive, whether castrated or not, and concluded that although castration was generally associated with reduced sexual functioning, castrated males were still capable of being aroused by tactile or psychological stimuli, and that the data did not justify an opinion that the public may be protected from socially dangerous types of sex offenders by castration laws.


Treatment of testicular cancer provides some information on the psychological effects of castration. For patients with early stage testicular cancer, radical unilateral orchiectomy is virtually always the primary therapeutic intervention (Jones & Vasey, 2003). Van Basten et al. (1999) reported on 21 patients who had undergone unilateral orchiectomy for testicular,: cancer, studying sex hormone levels and performin’ visual erotic stimulation (VES) tests and caverllosa~ artery duplex ultrasonography. They found that after a diagnosis of testicular cancer, sexual dysfunction’ was considerable, but within I year after treatment: there was some improvement, and the results of the VES test suggested that sexual dysfunction was more psychological than organically based. Jonker-Pool et al. (2001) performed a meta-analysis of 36 studies of individuals treated for testicular cancer. Four of these studies reported 011 groups treated with unilat. eral orchiectomy followed by surveillance, with 25% of subjects reporting a loss of desire, 24% orgasmic dysfunction, 16% ejaculation dysfunction, and 11% a decrease in sexual activit}’. Unilateral orchiectomy has no effect On the endocrine system. The clinical experience of one of the authors (Wechsler, 2006, personal communication) is that patients recover quite well psychologically over time.


Studies of individuals treated with prostate cancer provide data that is more relevant because castration is bilateral. The classic form of androgen deprivation is surgical castration by bilateral orchiectomy, which is the most immediate method of reducing circulating testosterone by > 90% within 24 hours (Maatman, Gupta, & Montie, 1985; Miyamoto, Messing, & Chang, 2004). Another summary indicates that surgi. cal castration, estrogens and LHRH agonists equally decrease plasma testosterone to between 5% and 10% of its original value (Baltogiannis, Giannakopoulos, Charalabopoulos, & Sofikitis, 2004). However. the adrenal glands still produce testosterone, estimated at 5% to 10% of the total amount (Sanford et al.. 1977; Salllen, 2003; Young & Landsberg, 2001).

Side effects of this procedure are loss of libido and potency, hot flashes, osteoporosis, loss of muscle mass, fatigue, weight gain and anemia, and psychological symptoms associated with definitive castration (Miyamoto et aI., 2004; Schroder, 1997).

Patient Preference of Medical over Surgical Treatment for Prostate Cancer

While orchiectomy is a relatively simple procedure with minor risks (Loblaw et al., 2004), it has fallen out of favor because of its psychological impact and the existence of viable medical alternatives. Despite this, some authors have suggested that it is underused. Polasky et al. (2001) used questionnaires to assess quality of life issues, and compared men who had received surgical castration with men who had received medical castration with LHRH agonists. Men who chose LHRH agonist therapy reporled greater problems with their overall sexual functioning ihan orchiectomy patients. despite both groups having a similar pretreatment level of functioning. The group receiving medical castration perceived themselves as less likely to be free of cancer. The authors suggested that these differences might have been partially related to regular injections that served as a constant reminder to the presence of disease. They also suggested that preference for injections over surgery might have been because of fear of pennanent mutilation, the ability to discontinue injections, and/or the loss of a masculine self-image. Another study suggested that patients preferred this approach for reasons of convenience and cost (Chadwick, Gillatl, & Gingell. 1991).

It is clear, however, that when patients are given the choice of medical or surgical castration, most choose medical approaches first (Cassileth et al.. 1989).

Studies of the Effects of Castration on Sexual Functioning in Patients Treated for Prostate Cancer

Greenstein et al. (1995) reported on a study of 16 mcn who had undergone physical (10, with posttreatment maintenance with Autamide) or hormonal (6, treated with diethylstilbestrol) castration. The mean interval from castration was 21 months. All patients had penile plethysmography while provided with erotic visual stimulation. Four of these patients achieved functional erections; all four of these had been physically castrated. All patients reported good erections and strong libido before castration with libido being markedly decreased after castration. In all patients the onset of erectile dysfunction was noticed a few weeks after castration. No patient reported spontaneous erections following castration and none had attempted intercourse.

Rousseau et al. (l988) administered a questionnaire and reported on biological and sexuality changes in prostate cancer patients receiving fllltamide with either surgical or medical castration with an LHRH agonist. Serum concentrations of testosterone were decreased to approximately 5% of pretreatment values with no difference in levels being found between those receiving orchiectomy and those with medical castration. The only difference was that there was a 2 to 3 week delay before castrate levels of serum androgens were achieved with LHRH agonists. In fact, a “flare” reaction is usually observed with the onset of GNRH agonist therapy for prostatic cancer, with a surge of testosterone during the first week or so after a depot injection. and some urologists will block this with a 1-week course of bicalutamide or a similar agent. Citabner et al. (2006) in Goodman & Gilman’s Textbook the Pharmacological Basis of Therapeutics suggest a 2 to 4 week administration of androgen receptor blocking (ARB) agents (p. 1388) for this purpose. Other authorities using GNRH agonists for the control of sexual behavior have suggested that ARB agents may be used, or that, as an alternative to using ARB agents, patients or their caregivers may be advised of the risk of hypersexuality and appropriate protective measures could be taken (Krueger & Kaplan, 2001). A major reduction in their interest for sexual intercourse was noticed by 70% of subjects; however. 20% of patients reported continuing sexual activity and interest, although the time from initiation of treatment to the assessment of current functioning was not specified.

Cost-Benefit Analyses

Several cost-benefit studies have been performed on the use of different types of antiandrogen therapies for the treatment of prostate cancer (Hellerstedt & Pienta, 2003). A recent study evaluated the cost effectiveness of six androgen suppression strategies to treat advanced prostate cancer. (Bayoumi, Brown, & Garber, 2000). Most provided similar outcomes in both survival and quality of life estimates. The annual cost of monotherapy with a GNRH agonist was 54995 in 2000; the cost has now fallen to approximately $3000. Orchiectomy had a one-time cost of $3360. which now in New York City is an outpatient procedure costing between $1500 and $2000. Cost concerns of medical castration led to the suggestion of surgical castration as an alternative (Oefelein & Resnick, 2003).

Degree of Androgen Suppression in Prostate Cancer

Androgen deprivation therapy in the treOitment of prostate cancer has the aim of achieving serum testosterone levels as low as possible so as to minimize androgenic stimulation of prostate cancer cells. Serum testosterone concentrations that correspond to castration levels have been set at less than 50 ng/dL (1.7 nmol/L). given variability of values in various reference laboratories (Bubley et al.. 1999; Sharifi. Gulley, & Dahut. 2005). Most men. however, achieve levels below 20 ng/dL (0.7 nmollL) after orchiectomy and it has been suggested that castration levels be redefined to use this as a threshold (Sharifi. Gulley, & Dahut. 2005). Two studies suggested that (or individuals treated with GNRH agonists who did not achieve castrate values of less than 20 ng/dL surgical castration should be considered (Oefelein, Feng, Scolieri. Ricchiutti, & Resnick, 2000; Oefelein & Resnick, 2003). although it is generally held that surgical and medical castration result in equivalent suppression o( testosterone.


In Czcchoslovakia Zverina et al. (1991) examined 84 castrated “sexual delinquents” 1 to 15 years after castration. Eighteen percent of subjects were capable of occasional sexual intercourse. and 21% lived in a stable heterosexual relationship. One-half had occasional erections in the morning. More than a third masturbated occasionally. One quarter of the subjects had objections to the results of the castration, most frequently because sex was lacking in their life. The authors did not observe serious physical or mental consequences of castration in the examined men.

In Denmark Sturup (1968, 1968) reported on 900 cases, and found that “asexualization” was present in 97% of castrated individuals, with 90% of individuals being satisfied with the operation. Castration was seldom followed by obesity, but there was a gynecoid distribution of fat. Hot flashes and sweating were common; beard and body hair were diminished but not head hair. Some individuals reported an increase in head hair. It was noted that diminished vitality, energy. and initiative was not definitely seen, and several of the men were noted to have increased energy and to have built new careers. Although it was noted that serious psychological consequences were not observed, at least 5, or 1.8%, of the men committed suicide.

In Germany Heim (1981) authored one of the most detailed studies of the cffects of castration on sex offenders released from prison. He studied ,9 men who had consented to castration while imprisoned in West Germany. The mean age of the subjects at castration was 42.5 years. and the median time since release from prison was 4.3 years. Sexual desire and sexual arousability were perceived by the subjects as having been considerably impaired by castration. and the frequency of coitus, masturbation, and sexual thoughts were perceived as strongly reduced after castration. However. 11 of 35 castrates stated they were still able to engage in sexual intercourse, and rapists proved to be more sexually active after castration than homosexual or pedophiliacs. The results of this study led him to conclude that there was a strong effect on sexual behavior only if castration were performed on males between 46 and 59 years of age and that the sexual manifestations of castration varied considerably between individuals.

In Norway Heim and Hursch (1979) and Bremer (1958) reported on a group of 244 Norwegian castrates, 102 of which were sex offenders. A questionnaire interview with the castrate and/or information from those in contact with him and an analysiS of documentary sources was conducted. In 66% of 157 persons for whom the effect of castration on sexual function could be judged. all sexual interest, reactivity, and activity had disappeared. Of 103 cases, 72% reported the effect immediately or just after the operation and 28% reported that it took a few months to a year for their sexual urges to disappear. However, 34% of the total group reported that their sexual interest and reactivity persisted for more than a year. Problems were reported in 18% of subjects after an observation period of 6 to 15 years, with weight gain, development of an aged appearance, climacteric symptoms, weakness, and deterioration in general health. There was an operative mortality of 2%, which nowadays would be considered extremely high and surgically unacceptable.

In Switzerland Heim and Hursch (1979) described details of a study by Cornu (1973), consisting of 127 sex offenders from Switzerland who were evaluated by psychiatrists and who had lived as castrates for at least 5 years after release from prison, with a comparison group of 50 who had refused to have their testes removed. Follow-up was 5 to 35 years. Criminal records and files from the court, police, and psychiatric hospitals were examined and 68 castrates were interviewed and medically examined. Cornu estimated the recidivism rate as being 4.13% after castration, in comparison with 52% in group of men who had had castration recommended to them, but refused, 10 years earlier. Of the 68 castrates, 63% said their libido and potency had been extinguished quickly after castration; 26% said there was a gradual decline; and 10% said they were able to have scxual intercourse 8 to 20 years after they were castrated. On medical examination. 51% were extremely overweight; gynecomastia was present in 10%; hair on the body was reduced in 63%. Thirty four percent of the castrates were adjudged to have a “castrate face,” apparently because their skin had become softer, more pliable, and slacker. Sixty subjects had X-rays of their vertebral column and 82% of these had osteoporosis diagnosed. Forty-three percent mentioned bone pains. mainly in the vertebral column.

Of the 68 castrates 28% indicated that the)’ had not experienced any psychological disorders; 40% had the opinion that castration had favorably influenced their moods, specifying that they felt calmer, happier, and more active than before the operation. Thirty two percent said that they felt miserable. with some indicating that they felt depressed, irritable, isolated. or maimed. Twenty two percent thought their capacity for work had deteriorated after castration; 16% thought that it had improved; and 61% thought there had been no change. Three men committed suicide. Seventy one percent said they were content with the operation because their abnormal sex drive had vanished, confinement was prevented, their state of health had improved. or marriage was possible for the first time; 16% were ambivalent; and 13% felt effeminate and mutilated.


A review of the recidivism literature is summarized in the accompanying table (see PDF). By today’s standards for conducting prospective clinical trials. most of these studies suffer from significant methodological difficulties. The subject groups are heterogeneous, not well described, and not exclusively sex offenders. Some individuals appeared to have had schizophrenia or mental retardation as primary diagnoses. Assignment was not random and treatment was not blinded. Length of time at follow-up was not clear and outcome measures, whether rearrest or reconviction, were often not specified. There also are not for the most part good comparison or control groups. However, at the time these trials and studies were conducted, the methodology was current and acceptable. Furthermore, standards applicable to clinical trials are not necessarily appropriate for forensic populations. There are se\’ere limitations on the conduct of research studies on such populations. Randomized and treatment blinded designs in situations where an outcome variable would involve victimization of an adult or child face substantial ethical challenge from institutional reviews boards. And many recidivism studies today use survival analysis techniques. which do not require control groups. It is clear that these studies are one of the most important sources of information concerning the effects of androgen deprivation therapy on sexually aggressive behavior. They do provide the equivalent of a series of retrospective analyses of open clinical trials conducted over very long periods of time (in many instances greater in duration of follow-up by a factor of at least three, compared with contemporary clinical studies (Bradford. 2006. personal communication), with assessments of the effects of this treatment on the most critical behavioral variable. commission of another sexual offense.

Czechoslovakia produced 2 studies (Taus & Susicka, 1973; Zverina, Zimanova, & Bartova, 1991) with small numbers, 5 and 84. Few details are available. but there were only 3 offenders relapsed out of these two groups for a very low recidivism rate.

Denmark yielded several studies (Hansen, 1991; Hansen & Lykke-Olesen, 1997; Le Marie, 19%; Ortmanll, 1980; Sand, Dickmeiss. & SchwalbeHansen, 1964; Stump, 1968, 1972; Weinberger ct al.. 2005), many with poorly defined or 110 control groups. All report a low recidivism rate with the largest by Sand who reported on 900 patients castrated from 6 to 30 years earlier and fOllnd a relapse rate of 1.1%, compared with relapse rates in comparison groups of 9.7%, 16.8%. or ;0% depending on the study and group.

Germany has produced a number of detailed studies (Heim, 1981; Heim & Hurseh, 1979; Langeliiddeke, 1963. 1968; Wille & Beier, 1989) that again support castration’s positive effect on recidivism. The largest, by Langeluddeke, reporting on patients in the community who had been castrated from 6 weeks to 20 years earlier reported a recidivism rate of 2.3% compared with a rate of 39.1% in 685 patients who were not castrated and released into the community. II should be noted that many of those castrated were under the Nazi regimen and were castrated involuntarily; this included individuals who were homosexuals.

A study is reported to have been done in the Netherlands (Meyer & Cole, 1997; Sturup. 1968) but details of this are not available beyond aggregate data.

Bremer (1958) reported on a group of 244 individuals castrated in Norway, 28 of whom were women, concluding that the indications. which. in addition to sex offenses. included schizophrenia and epilepsy, were too broad but that the procedure offered effcctive medical therapy to a subgroup of sex offenders.

Information is available from Sweden (Bremer, 1958; Lidberg. 1968; Meyer & Cole. 1997; Sturup. 1968). Switzerland (Cornu, 1973; Heim & Hursch. 1979: Sturup, 1972), and the Netherlands (Meyer & Cole, 1997; Sturup, 1968) which also suggest a dramatic effect on recidivism.

Finally, in California Weinberger ct al. (2005) reported on the follow-up of 40 sex offenders castrated in San Diego County during the period 1937 to 1948, none of whom reoffended.

By way of summary it would appear that effects 011 sexual interest and functioning begin almost immediately in the majority of patients. but that some report the maintenance of sexual functioning for a long period of time, and that the effects of castration have great individual variability. However imperfect. this castration literature supports the assertion that castration has a profound effect on sexual functioning and on recidivism of sexual crime.


Castration is a safe procedure that can be performed under local anesthesia as an outpatient. Anesthesia is injected into the inguinal cord. Sedation is often used. With the patient in a supine position bilateral incisions are made in the scrotum and the testicular artery, vein. and cord are ligated and then the testes are removed. Patients tolerate the procedure well. The procedure takes approximately half an hour to perform. As a rule, prostheses are not used. Morbidity includes bleeding, infection, and pain. Mortality is virtually nonexistent nowadays.

Hot flashes are commonly experienced with castration in medical patients. They usually require no treatment. However, estrogen, progesterone, or cyproterone acetate, each of which has its own risks and side effects. may ameliorate symptoms (Spetz, Zelterlund, Varenhorst, & Hammar. 2003). Clonidine and antidepressants can also be tried. However, this has not been reported to be a symptom that has been treated in the various studies of GNRH agonists with male sex offenders (Briken, Nika, & Berner, 2001; Krueger & Kaplan, 2001; Kruger & Kaplan, 2006; RosIer & Witztum, 1998; Saleh, Niel, & Fishman, 2004; Schober et aI., 2005; Thihaut, Cordier, & Kuhn, 1993; 1996).

Osteoporosis is a significant side effect of patients with paraphilias treated with antiandrogen therapy (Crasswick & Bradford, 2003) and of men treated with medical or surgical castration for prostate cancer (Oefelein & Resnick, 2004). Oefelein and Resnick (2004) reported on the incidence and management of this side effect in patients treated for prostate cancer, indicating that this is a significant side effect and that vitamin D (800 IU), calcium supplementation (1200 mg/d), weight-bearing exercise, parenteral estrogen therapy, and bisphosphonate therapy can be used. For paraphiliacs or sex offenders treated with GNRH agonists. bisphosphonate therapy, such as alendronate, has been used in conjunction with calcium and vitamin D. This has, in the experience of one of the authors (Krueger. 2006, personal communication), resulted in no apparent decrease in bone density according to bone density scans performed at baseline and then yearly.

Another risk of androgen deprivation therapy is that castrated patients. if they develop prostate cancer, would have a much worse prognosis. Low serum free testosterone has been reported to be a marker for high grade prostate cancer (Hoffman, DeWolf, & Morgentaler, 2000) and a recent study comparing finasteride. an inhibitor of 5A-reducates that inhibits the conversion of testosterone to dihydrotestosterone, the primary androgen ill the prostate, with placebo, suggested that this prevented or delayed the appearance of prostate cancer, there was an increased risk of high-grade prostate cancer. There is also a suggestion that the initiation of androgen suppressive therapy in men over 65 may be associated with increased cardiac events (D’Amico et aI., 2007).


A comprehensive review of the legal and ethical issues inherent in orchiectomy is beyond the scope of this chapter. However. some discussion of these issues is provided. Although several states have passed legislation (Scott & Holmberg, 2003) mandating chemical or physical castration, as far as can be determined these laws have not been used. Orchiectomy is a surgical procedure and. as such, for its ethical application, requires informed consent. Basic elements of informed consent for a medical or surgical procedure are the capacity to understand the nature of the procedure and its risks and benefits, the provision of information, and that a decision be freely made. (Applebaum & Gutheil, 1991). Clearly, if an individual refuses to subject himself to such a procedure and/or to give his consent. then to proceed with castration would violate medical ethical guidelines and be considered unethical.

Some have pointed to the existence of a choice of prison or orchiectomy as being inherently coercive (Gaensbauer. 1973; Klerman, 1990). Indeed, within the United States criminal justice system, all of plea-bargaining could be said to be inherently coercive, requiring a choice of an admission of guilt or taking the risk of trial .md a substantially longer prison sentence, yet this is considered legal and ethical. However, one can reason that sex offenders start from a different moral baseline than individuals who have not committed a sexual crime, in that they have victimized other individuals, and this fact mitigates against the coercive elements inherent in a choice of prison or castration (either medical or surgical) (Shajnfeld, 2008; Wertheimer, 1987).

Additionally, the choice of castration may be offered to an individual, and he may refuse it, but it cannot be forced onto a person. Accordingly, we would hold that presentation of a choice to an individual convicted of a sex offense and in need of such treatment as an alternative to incarceration is medically and morally ethical. Indeed, a case was recently reported of an individual with pedophilia who, faced with 10 years of failure of cognitive behavioral therapy to stop him from repeatedly sexually acting oul when released into the community, successfully sued to receive chemical castration (Krueger & Kaplan, 2006) He was then able to live in the community while treated with GNRH agonists without other sexual victims for the subsequent 10 years that he was treated. We would agree with Klerman, who emphasizes that prisoners should be allowed this option (Klerman. 1975).


Orchiectomy has several advantages over medical castration. Cost is one evident factor with orchiectomy being undoubtedly less expensive than GNRH agonists, but not necessarily progesterone. However, GNRH agonists have a side effect profile which is substantially better than the more traditional androgen suppressing agents (estrogen, progesterone, cyproterone acetate, or flutamide) (Rosier & Witztum, 1998; RosIer & Witztum, 2000) and GNRH agonists would seem to be the most evident equivalent to surgical castration. Other authorities point out that there has been much greater experience with more traditional androgen suppressing agents with thousands of sex offenders in past or current treatment with such agents, and that such agents may cause less side effects related to androgen deprivation and are more capable of titration (Bradford, 2006. personal communication).

Many of the effects and side effects of orchiectomy and of GNRH agonist medication are related to the removal of testosterone, but there are some medication specific risks to the GNRH agonists as well as to the other antiandrogens, such as progesterone or cyproterone acetate that would not exist with orchiectomy. Any exogenous substance can cause an allergic or anaphylactic response, and such responses have been reported with GNRH agonists (Raj, Karadsheh, Guillot, Raj, & Kumar, 1996). Skin reactions have also been reported (Neely et aI., 1992) (Labarthe, Bayle-Lebey, & Bazex. 1997; Monasco, Pescovitz, & Blizzard, 1993) although careful examination and consideration of injection technique, or a switch to another of the GNRH agonists with a different compound and different vehicle could resolve this. More serious side effects. such as pure red cell aplasia associated with leuprolide (Maeda et al., 1998) have been reported. The occurrence of these serious side effects could be avoided with different pharmacological agents or surgical castration. Some patients may prefer surgical to medical castration. choosing this to avoid injections or for other personal reasons.

Clearly, the ohvious advantage of GNRH agonists is that they are reversible and they do not entail the psychologically adverse effects of castration. Additionally. GNRH agonists may be lowered in dosage or the time interval between injections may be extended, allowing for some titration to allow for an increase in testosterone and the facilitation of unproblematic sexual behavior, if it is determined that this is appropriate. although experience with this is limited, and current dosage packaging limits the flexibility of titration. Castration would not allow for this option. However, many individuals discussed in the aforementioned studies, although castrated, were able to retain some sexual functioning. and it is not certain that sexual functioning will completely disappear with surgical castration. Further, while “add-back” therapy with testosterone to castrated sex offenders is a possibility, one study (Hansen, 1991; Hansen & Lykke-Olesen, 1997) reported a relapse with this treatment. and it would be important to carefully consider and weigh the risks and benefits of such therapy before initiating it in this population.


It should be emphasized that castration by itself is not an exclusive or final treatment. Patients who are castrated can still function sexually, with erections, intromission and ejaculation, and perpetrators can victimize individuals without using a sexual organ. Rather, castration’s main effect is to decrease or abolish the intensity of sexual cravings so as to allow iI concomitant increase in volitional capacity to maintain self-control (Berlin, 2005). This implies that an individual has to be willing to acknowledge that his sexual behavior is a problem and to be motivated and to agree to make it a target of treatment. One has to form an alliance with the patient in this effort.

Additionally, continual monitoring of testosterone, monitoring for osteopenia or osteoporosis with bone density assessments, and periodic laboratory and physical examinations are all required to follow individuals treated with this modality. Treatment to prevent osteopenia and/or osteoporosis with bisphosphonoates, calcium and vitamin D, are required medically as well as other treatments, including psychotherapy and sex offender specific therapy. Other modalities, such as ongoing monitoring by parole, housing, social services, and vocational rehabilitation, are important as well. Indeed. it may be that these other elements may contribute as much as or more than medical or surgical castration to reducing recidivism, and future research should configure study designs so as to assess the effects of sex offender therapy and other elements of a treatment plan, as well as medical or surgical castration. on recidivism of sexual crimes.

The largest open trial involving the use of a chemically castrating agent, tryptoreline, (RosIer & Witztum, 1998) relied on, among other measures, the self-report measure of the Bancroft Sexual Interest and Activity Scale (Bancroft, Tennent. Lougas, & Cass, 1974; Tennent, Bancroft, & Cass, 1974) and testosterone levels, both of which demonstrated change (a reduction in testosterone and in sexual interest and activity, as measured by a Likert scale and by asking the number of ejaculations a subject had experienced in the week before the questionnaire) related to GNRH treatment. These scales could be lIsed to assess outcomc and monitor a patient.

However, a recent study (Schober et aI., 2005) reported on the use of leuprolide acetate to treat pedophiles, using testosterone levels, sexual interest preference by visual reaction time, penile tumescence. and polygraphs to assess the effects of leuprolidc acetate. Subjects were noted to be deceptive regarding increased pedophilic urges and masturbatory frequency and it would be important to incorporate some of these objective modalities into treatment assessment and monitoring. Given the possibility of surreptitiolls anabolic steroids (Harris. Phoenix, Hanson, & Thornton, 2003, p. 65), it would also he important to screen for these in an ongoing way. Some in the United Stales have suggesed that a urologist be involved in such treatment (Schober, Byrne, & Kuhn, 2006) and others that an internist or family doctor offers an initial medical assessment for clearance for such treatmcnt.

Fundamentally, it cannot be said that castration is a be all and end all measure, and once castrated, society can be done with a sex offender. The marked individual variation in the effects of castration, with some individuals able to retain some sexual functioning, with a capacity for erections and ejaculation, and a motivation for sexual activity, even years after being castrated, suggest that it is critical to assess the actual effects of castration on the patient on whom it is performed. To this end, self-report scales, plethysmography, polygraphy, laboratory and clinical assessment, and observation should be utilized ill an ongoing way.


Animal and human studies show that the main effect of orchiectomy is a substantial reduction in sexually motivated behavior, but this can have great individual variation. However, all studies on orchiectomy of sex offenders support its efficacy in reducing recidivism. Furthermore, it should be clear that orchiectomy is not a cure-all for sex offenders, nor is it something that can be utilized without a system of aftercare that would include extensive provisions for monitoring, treatment, and support.

Orchiectomy for the treatment of sex offenders has had a dark history, in many instances being performed on individuals against their will or who because of other reasons were unable to consent. Moreover, psychiatry has a history in the century past of utilizing treatments, such as frontal lobotomy or extremely high-dose neuroleptics, that were ultimately established as being extremely harmful and of limited or no therapeutic value. Use of orchiectomy to treat sex offenders in Europe has virtually disappeared and in the United States, although several states have passed laws mandating surgical (or medical) castration for a variety of sexual crimes, contemporary usage of surgical castration has also ceased.

However, orchiectomy’s usage in the past century also antedated the modern era of biological treatment with much greater rigor in the classification of psychiatric disorders and research design and it can be anticipated that the future will bring better designed studies to objectively assess the effects of medical, if not surgical, castration, as well as the effects of other treatment modalities.

In spite of current pharmacological practice in the treatment of sex offenders, which includes serotoninselective reuptake inhibitors and other androgen suppressing agents (see other chapters in this volume), ,IS well as, more recently, GNRH agonists, cost considerations as well as medical ones suggest that orchiectomy could be a desirable alternative. Given the great number of sexual crimes and victims and the great expense sex offenders create for society, almost nothing is spent on research involving the biological treatment of sex offenders. This chapter should make clear that such treatments, including orchiectomy, have great promise and should be the subject of further consideration and investigation.

View PDF of this publication with tables and complete references.