LETTERS TO THE EDITOR

of cidofovir is negligible, since the majority (>80%) is recovered unchanged in the urine. The principal systemic toxicity (nephrotoxicity) can be avoided by topical application. This initial case report suggests that topical cidofovir may represent a valuable addition to the armamentarium of hard to treat condyloma. However, a careful evaluation of the dose and frequency of cidofovir application is warranted.

EDITOR,-Despite the high prevalence of condylomata acuminata, their treatment remains unsatisfactory for both patients and physicians. Epidemiological studies estimated the prevalence of genital warts between 1-31% with a peak occurring in young adults. 1 As a consequence, the economic burden of human papillomavirus (HPV) infection in the United States is estimated to exceed $8.5 billion per year. 2 Current treatments rely on the ablation of warts (cryotherapy, laser vaporisation, electrodissection, or trichloroacetic acid) or the interruption of cell division (podophylox, intralesional or systemic interferon, and 5-fluorouracil). Recently, imiquimod has been successfully used as a topical immune response modifier for the treatment of external anogenital warts. 3 However, there remains a substantial number of patients who fail to respond to traditional and newer drugs. We report on such a patient with recalcitrant condylomata acuminata on the glans and shaft of the penis who was successfully treated using the novel virustatic cidofovir as a 1.5% gel.
A 48 year old man with a 2 1 ⁄2 year history of condylomata acuminata had received laser treatment, podophylox, and imiquimod. The patient's history was remarkable for diabetes mellitus. He presented with numerous, flesh coloured, flat topped papules in a circular manner on the outer preputium and the glans, whereas some lesions in the coronary sulcus had a more verruciform appearance (fig 1). On histological analysis, the typical picture of acanthosis, papillomatosis, and numerous koilocytes was seen. Papillomavirus typing revealed HPV-43 by nested PCR using consensus primers. 4 Cidofovir was evaluated in the indicator patient at 1.5% cidofovir in a viscous gel (propylene glycol, parabene). Initially, the patient was treated on an outpatient basis with two applications of cidofovir gel per week to the respective lesions without any side eVects. Thereafter, the patient was instructed to apply the gel three times a week by self application. At week 6 the patient pre-sented with small erosions surrounded by a marked erythema on all treated sites (fig 1). The lesions were painful. Condylomata were still present in the coronary sulcus. At this point treatment was stopped and antiseptic treatment was given with betadine solution once daily. Seven weeks later (week 13) all lesions had completely healed (fig 1). Neither scarring nor dysaesthesia were noted. No recurrence has occurred since. Cidofovir, 1-[(S)-3-hydroxy-2-(phosphono-methoxy)propyl]cytosine, is a member of a new class of antiviral agents (phosphonylmethylether nucleotide analogues). 5 It shows potent in vitro activity against a broad spectrum of herpesviruses, including human cytomegalovirus (CMV), HSV-1 and HSV-2, and adenovirus. 6 Recent in vitro and in vivo studies have demonstrated activity against papillomavirus and poxvirus. 6 7 Cidofovir is a nucleotide analogue of deoxycytidine monophosphate (dCMP). Analogous to the metabolism of dCMP to dCTP, cidofovir is converted to the active cidofovir diphosphate that inhibits viral DNA polymerases. 8 The uptake of cidofovir into cells is slow, but the intracellular half life of the various metabolites is between 6 and 87 hours, thus allowing infrequent dosing. 8 Compared with the general mechanism of activation of ganciclovir, which requires phosphorylation by the virus encoded UL97 gene, cidofovir does not depend on viral infection for its phosphorylation and can therefore prime cells to an antiviral state (prophylaxis).
The metabolism of cidofovir is negligible, since the majority (>80%) is recovered unchanged in the urine. The principal systemic toxicity (nephrotoxicity) can be avoided by topical application.
This initial case report suggests that topical cidofovir may represent a valuable addition to the armamentarium of hard to treat condyloma. However, a careful evaluation of the dose and frequency of cidofovir application is warranted. Bladder carcinoma presenting to genitourinary medicine departments EDITOR,-Large numbers of patients are seen in departments of genitourinary medicine with symptoms suggesting infection or inflammation of the genitourinary tract. Although bladder neoplasms typically cause painless haematuria, in a subgroup of patients they cause other urinary symptoms that may produce diagnostic confusion. We identified five patients who were referred to the genitourinary medicine service, and who were found to have bladder carcinoma (see table  1). Four of the patients presented to the genitourinary medicine department at High Wycombe (5500 new attendances per annum) between 1991 and 1998; the fifth patient presented to the Oxford genitourinary medicine department (9000 new attendances per annum) in 1997. None of the patients had an occupational history that placed them at higher risk for bladder cancer. Men with bladder carcinoma typically present in later life (median age 69 years), but the condition may occur at younger ages. 1 A subgroup of patients develop frequency, urgency, and dysuria-symptoms usually associated with bladder infection. 2 Rarely, penile and perineal pain mimicking prostatitis may be a presenting feature, as in patients 3 and 4, who have been described in more detail elsewhere. 3 Non-specific urethritis (NSU) is diagnosed commonly in genitourinary medicine clinics in men of all ages. In this series, patient 2 was referred with presumed NSU, and patient 4 had attended previously with a diagnosis of NSU, 2 years before the bladder cancer was diagnosed (at that time there were 5-10 white cells/high power field (×1000) on a urethral smear, and a chlamydia ELISA test and cultures for Neisseria gonorrhoeae were negative; no haematuria was detected). Both patients were subsequently noted to have neoplastic infiltration in the bladder neck area and prostatic urethra.
In all five cases a degree of persistent microscopic haematuria was noted at presentation; in patient 4 this was never greater than a trace on dipstick testing. Patient 1 reported intermittent painless macroscopic haematuria at presentation; he was referred by his general practitioner with suspected Figure 1 Condylomata acuminata with some lesions in the coronary sulcus having a more verruciform appearance. genitourinary infection, rather than suspected neoplasia, because of his young age (26 years).
Bladder neoplasia is especially liable to cause irritative symptoms when represented by, or associated with, carcinoma in situ of the bladder urothelium. 1 2 Urine cytology may be useful in this subgroup, and was abnormal in all three of the five patients in whom it was requested. When this process involves the prostatic urethra, symptoms mimicking prostatitis may arise. Early diagnosis of bladder neoplasia is of prognostic importance; the presence of carcinoma in situ or prostatic involvement by bladder carcinoma are poor prognostic features for which radical surgery may be required. 1 4 These cases highlight the importance of careful follow up of patients presenting with persistent irritative-type bladder symptoms, especially in an older age group, when specific tests for genitourinary infection are negative, and where microscopic haematuria is a feature. Bladder carcinoma should be considered in this subgroup; urine cytology and referral for cystourethroscopy may be indicated. Although rare in younger adult males, bladder cancer should not be ruled out in men under the age of 45 years, and our experience strengthens the case for continuing with routine urine testing in genitourinary medicine clinics.

Atrial myxoma and HIV infection
EDITOR,-Atrial myxoma has not previously been reported in HIV infection. We describe a patient with advanced HIV disease who underwent surgery for this condition.
The patient was diagnosed with asymptomatic HIV infection in February 1987 when she was aged 50 years. Her CD4 count was 690 ×10 6 /l at this time. HIV infection was acquired through sexual intercourse with a bisexual male partner. In December 1990 the CD4 lymphocyte count had fallen to 190 ×10 6 /l and zidovudine monotherapy was started. This was continued until 1996 when she was prescribed a combination regimen. Co-trimoxazole was given for Pneumocystis carinii prophylaxis, but the patient deferred starting this until December 1992.
In February 1992 the patient was admitted to another hospital with an acute myocardial infarction which was successfully thrombolysed. Fasting lipids were within the normal range. There were no cardiac risk factors apart from smoking.
In September 1995 the patient experienced a syncopal episode. An echocardiogram revealed a mass in the left atrium consistent with a left atrial myxoma. A coronary angiogram showed normal coronary arteries. Surgical resection of the myxoma was recommended.
In December 1995 the patient's CD4 count was 64 ×10 6 /l, but apart from oral candidiasis there had been no HIV related problems since diagnosis. Two leading UK HIV physicians were asked if they considered surgery to be advisable. They estimated the patient's likely survival from HIV disease to be 1-4 years. The risks of major heart surgery had to be balanced against the likelihood of recurrent symptoms from the myxoma in the next 1-4 years. The patient and her physician agreed to proceed with surgery.
On 4 December 1995 the patient underwent surgical resection of a pedunculated left atrial mass. Histological examination confirmed a benign atrial myxoma. The procedure was uncomplicated and she was discharged from hospital 4 days later. Annual cardiac review including an echocardiogram has shown no evidence of recurrence up to the present time. She remains free from cardiovascular symptoms. Her HIV disease is managed with combination therapy that consists of stavudine, lamivudine, and efavirenz. Current CD4 count is 564 ×10 6 /l and viral load less than 50 copies/ml (Chiron bDNA v3.0) Atrial myxoma is a rare tumour that is considered to be benign although recurrence and metastases have been described. 1 The myocardial infarction suVered by our patient may have been an embolic manifestation of the myxoma, and the normal serum lipids and normal coronary angiogram almost 4 years later would support this.
In 1995 expert opinion provided a very guarded prognosis for someone with a CD4 count of 60 ×10 6 /l who had been exposed to a single antiretroviral agent, zidovudine. Today there would be less debate over the merits of such a surgical procedure in this scenario, and this case demonstrates the excellent outcome that can be achieved with major surgery despite profound immunosuppression. The proved benefits of HAART (highly active antiretroviral therapy) have made it unacceptable to deny major surgical interventions to individuals with HIV.

The association between receptive cunnilingus and bacterial vaginosis
EDITOR,-We are puzzled by the surprisingly little, if any, serious work done to explain the epidemiological enigma of high prevalence of bacterial vaginosis (BV) in lesbians, 1 and the oft observed, but as yet unconfirmed association between BV and receptive cunnilingus in women in general.
In a detailed study of 17 consecutive lesbians attending the department of genitourinary medicine at the Royal Sussex County Hospital in Brighton, bacterial vaginosis was found in six women (35%). Of nine lesbians who practised receptive cunnilingus in the previous 4 weeks, six (67%) had BV. By contrast, no BV was present in all eight women who did not practise oral sex (table 1).
In a parallel prospective study of 256 consecutive heterosexual female patients attending the same department, 55 (21%) were diagnosed as having BV. Of 111 women who practised receptive cunnilingus in the previous 4 weeks, 41 (37%) had BV. Of 145 women who did not have oral sex, only 14 (10%) had BV (table 1). In both groups there was strong association between BV and receptive cunnilingus (p<0.001).
The evidence associating bacterial vaginosis with oral sex is too strong to be ignored and repeatedly dismissed. The mouth is full of Gram positive and Gram negative organisms including Bacteroides oralis and, albeit in much smaller quantities, lactobacilli. These organisms are part of normal flora in the mouth, but are they normal to the vagina? Might the tiny amount of lactobacilli be enough to act as a phage which destroys the

Is partner notification in the public interest?
EDITOR,-This ethical debate 1 calls for comment. Why did the clinicians only suspect AIDS? Surely at the second attendance the diagnosis was clinically obvious. As well as continuing treatment of candidiasis and starting prophylaxis of Pneumocystis carinii pneumonia, was not treatment for AIDS indicated? For fear of court proceedings a specimen of blood untested or surplus to routine haematological tests could have been stored to confirm, if necessary, the clinical diagnosis. A perspicacious defence lawyer could make much of this in terms of doctor thoroughness, cautiousness, and thoughtfulness-on behalf of his client.
In terms of contact tracing the word "disclosure" occurs repeatedly. Surely the first thing an index case is told when his/her cooperation is sought is that under no circumstances will their name be divulged. The contacts, when attending, will be refused any information regarding who has named them and immediately assured that the same confidentiality will be maintained if their cooperation is called for in the contact tracing process.
Only when it becomes widely known in a clinic that such confidentiality is thoroughly pursued will counterproductive fears be eliminated. With understanding and cooperation it can be done.

Sexual partner reduction and HIV infection
EDITOR,-We recently conducted a national urban random sample survey of 1400 men of sexually active age in the Dominican Republic to measure possible change in sexual behaviour. This sexual behaviour change (SBC) survey was prompted by results from the 1996 demographic and health survey, which found that 84.8% of a national random sample of Dominican men claimed that they had changed their behaviour in some way because of their fear of, or concern about, AIDS. The proportion of respondents reporting behaviour change such as becoming monogamous or reducing their number of sexual partners was about triple the proportion reporting condom adoption. In our SBC survey, 79% of respondents claimed to have changed behaviour because of concern about AIDS. A majority (52.2%) said they had become monogamous or reduced their number of sexual partners. This was followed by condom adoption (14.6%), only having sexual relations with a person they know (13.9%); avoiding relations with "prostitutes" (9.0%); or becoming abstinent (1.6%

Features of AIDS and AIDS defining diseases during the highly active antiretroviral therapy (HAART) era, compared with the pre-HAART period: a case-control study
EDITOR,-To assess the features of AIDS defining illnesses during the HAART era versus those observed before the introduction of HAART, the characteristics of 72 consecutive patients, diagnosed in 1997-9, were compared with those of 144 subjects randomly selected from the 436 patients diagnosed from 1985 to 1995, in a case-control study. An impressive drop in AIDS diagnosis was seen shortly after the introduction of HAART, with only 38, 21, and 13 cases per ∼1000 patient years observed in 1997, 1998, and 1999 respectively, versus a mean frequency >60 cases per ∼1000 patient years, demonstrated during 1991-5. A tendency towards an increased incidence of female sex was shown in 1997-9 compared with 1985-95 (33.3% versus 27.1%), together with a rise of mean CD4+ lymphocyte count (86.8 (SD 99.4) versus 72.1 (93.7) cells ×10 6 /l), while an increase in the mean patient age was highly significant (39.8 (8.3) versus 34.6 (7.7) years; p<0.0001). When considering the exposure to HIV infection, drug abuse became significantly less important in the HAART era (p<0.05), while heterosexual transmission was notably increased (34.7% versus 13.2% of cases; p<0.0003). The distribution of AIDS defining disorders during the HAART era showed an tendency to a reduction in cytomegalovirosis, cryptococcosis, mycobacteriosis, cryptosporidiosis, and HIV encephalopathy, while a relative increase in pneumocystosis, oesophageal candidiasis, wasting syndrome, tuberculosis, and non-Hodgkin's lymphoma was found; neurotoxoplasmosis and Kaposi's sarcoma were stable (table 1). However, while pneumocystosis, Candida oesophagitis, neurotoxoplasmosis, and Kaposi's sarcoma represented the four most frequent AIDS related events in both study periods, cytomegalovirosis, HIV encephalopathy, cryptococcosis, and mycobacteriosis (which ranked fifth to eighth in frequency during the pre-HAART era), virtually disappeared after the introduction of HAART (28 versus four overall cases; p<0.007), together with cryptosporidiosis. Neoplasms and HIV related disorders (encephalopathy and wasting syndrome), showed a slightly increased frequency during the HAART era (16.8% and 9.2% during 1997-9, versus 13.2% and 7.9% respectively, during the pre-HAART period). A considerable trend to increased mean CD4+ count was found during the HAART era for all AIDS related illnesses considered, except neurotoxoplasmosis. However, this increase in CD4+ count was significant only for Candida oesophagitis (p<0.04), wasting syndrome (p<0.03), and tuberculosis (p<0.03), probably because of small patient samples. Only seven of the 72 patients who developed AIDS since 1997 (9.7%), were eVectively treated with HAART for more than 3 months before diagnosis; in the remaining 65 cases HIV infection was detected concurrently with an AIDS defining event in subjects who were unaware of their condition (40 cases), or refused HAART or carried out it with poor adherence (25 patients).
Although a sharp decline in the incidence of multiple AIDS defining events was demonstrated since the introduction of HAART, the distribution of primary AIDS associated diseases showed limited modifications. 1-3 An increased incidence of women, a higher patient age, a greater role for heterosexual transmission compared with injecting drug addiction, and a rise in CD4+ count were disclosed by us in the HAART era compared with the pre-HAART period. Appreciable modifications of the spectrum of AIDS associated illnesses were also observed during the HAART era (a drop of cytomegalovirosis, cryptococcosis, mycobacteriosis, cryptosporidiosis, and HIV encephalopathy, with a parallel increase in pneumocystosis, oesophageal candidiasis, wasting syndrome, tuberculosis, and non-Hodgkin's lymphoma), together with a considerable trend towards an increased mean CD4+ count at diagnosis, as previously noted. 2 5 Disorders which are directly or indirectly associated with HIV damage itself, AIDS related neoplasms, and opportunistic diseases occurring with a less profound immunodeficiency, show a substantially stable or even increasing incidence among newly diagnosed cases of AIDS. 1 2 4 However, opportunistic diseases related to a severe immunodeficiency are still frequent among AIDS defining events, since the majority of cases identified during the HAART era occur in patients who are not aware of their disease, or fail HAART. Only early detection and aggressive treatment of HIV infection may definitively improve the epidemiology of AIDS; a continued surveillance of AIDS related disorders remains critical for the implementation of therapeutic and prophylactic strategies. ROBERTO (Danish, Finnish, Flemish, French, German, Greek, Italian, Portuguese, Spanish), and the full text (without illustrations) can be found online on the website (http://www.med. ic.ac.uk/df/dfhm/europap/hustling/press.htm).
How do you begin to address the sexual health needs of commercial sex workers (CSWs)? Here you will find (most of) the answers. This immensely practical book is essential for those setting up an outreach service, or simply wishing to know more about commercial sex work. It is the outcome of a series of projects and workshops, written by workers providing services to CSWs throughout Europe, and draws from the lessons learnt by these pioneering workers and clients. It is written with great clarity and frankness. The A4 layout is bold, imaginative, and attractive, with illustrations of promotional literature. Its European inclusiveness means that sadly it cannot be specific regarding, for example, the law as it applies to commercial sex. It does, however, give the broad framework with which providers must acquaint themselves wherever they work. It takes us through the steps; sources of funding, the scope of the service, useful contacts, where to make contact with CSWs, and so on. Importantly, in the current climate there are sections on evaluation and monitoring of the service, the legal and political context of the work, and dealing with the media. It stresses the heterogeneous nature of commercial sex workers whether male, female, or transsex, and the spectrum of commercial sex venues. Peer educator programmes are covered in some detail.
There are fascinating pieces of practical advice-for example, cooperate with police, but don't be identified too closely with law enforcement. Advising police of your outreach vehicle's registration number may prevent you being stopped for kerb crawling! You can set up a flawless screening service and find only a few CSWs attend. The book reminds us middle class, health aware professionals that, for many, sexual health is not a priority. We are perplexed when faced with "indiVerence, hostility and self destructive behaviour"; that her next fix, a roof over her head, or the desire to have a baby might be more important to the CSW than the nebulous risk of HIV. Address some of these needs and you have the carrot to attract attention to and confidence in your service. The spin oV is that clients can then benefit from STD screening and safer sex advice. Simply providing toilets and somewhere safe to have a cup of tea may be enough for some.
I would have liked to see a further reading list, but this book fulfils its remit excellently.
The Audit Commission, established in 1983, reports on a 2 year study of the specialist Child and Adolescent Mental Health Services (CAMHS) as provided by local authorities and NHS trusts. Local information has been processed centrally to generate facts and figures and comparative data. The 13 000 bodies providing CAMHS spend £100 billion (sic) of public money annually in England and Wales. The Commission's team of seven have met with external advisers with a view to shaping of the audit, its comments, and guidance. The aim is to achieve economy with eYciency and eVectiveness. The report is in five chapters and five helpful appendices. It lists 71 references and has an index.
Under the heading "The changing context" it is revealed that one in five children and adolescents (alas, not defined for females and males) suVer from a wide range of mental health problems of variable degrees of severity from social ineptitude through psychological to severe psychiatric disorder. Strong links are noted with juvenile crime, alcohol and drug abuse, eating disorders, and of course self harm.
The key components of the CAMHS are viewed as four "tiers": (a) Those providing primary intervention, eg, GPs, health visitors, residential social workers, juvenile justice workers, school nurses, and teachers. (b) Professional providers of services, eg, clinical and educational psychologists, paediatricians, child psychiatric nurses in the community, and child psychiatrists. (c) High grade specialist services for severe, complex and persistent disorders, eg, child psychiatrists, community psychiatric nurses, psychotherapists, occupational therapists and art, music, and drama therapists. (d) Consists of hospital services especially unnamed "highly specialised outpatient teams". This clearly applies to accident and emergency departments, obstetric and gynaecology departments, and genitourinary medicine departments. These deal very adequately with self poisoning episodes, premarital abortions, and sexually acquired infection, but fail to see the underlying behaviour as but one manifestation of an ongoing complex of medicosocial pathology. Clearly, services for the care of our adolescents, unlike paediatrics and geriatrics, are seriously fractionated.
What follows should help the holistically minded hospital doctor to increase his awareness and skills and so make more regular and early use of referral routes and emergency cover arrangements provided by developing CAMHS.
It is clear that in many areas there is an urgent need to plan how best to meet unmet needs, including appropriate monitoring. The final chapter of this book purports to show how, with national support, highly active local coordination can establish and advance improvements. Recommendations are provided. There are opportunities for masterly leadership.
As the first specialty to be nationalised in the United Kingdom, genitourinary medicine has come a long way from the days of "pox doctoring" in "clap clinics". Has the time come for it to give a lead in the development of more appropriate and comprehensive services for adolescents?
For the long sighted and adventurous GU physician this book suggests how to begin. "Venereal diseases are like the fine arts-it is pointless to ask who invented them." (Voltaire, Dictionaire philosophique).

R S MORTON
Sexually transmitted diseases (STDs) now rank among the top ten diseases for which adults in developing countries seek health care. The economic burden of STDs on both developed and developing countries is enormous. Infection with conventional STDs is a risk factor for transmission of infection with HIV, and therefore for the development and spread of the AIDS It is imperative that laboratory services are available to guide the clinician to the correct diagnosis and treatment of these conditions, and to give an accurate epidemiological picture of their prevalence in a particular community in order to target relevant populations and ensure optimal and economic use of available resources. Yet, the availability of both funds and technology varies widely between diVerent settings.
This manual sets out to give comprehensive guidance on tests available and applicable to the level of expertise and funding available.
Nine chapters cover the major STDs, encompassing bacterial and viral infections, and under the umbrella of vaginitis in adults; trichomoniasis, candidiasis, and bacterial vaginosis. Each chapter begins with a brief description of the microbiology of the infective agent and the clinical spectrum of disease. The detail given is not consistent, being comprehensive for chancroid and granuloma inguinale, and surprisingly brief for HIV and chlamydia by way of contrast. Then follows a description of collection and transport requirements, and of techniques for diagnosis. The emphasis is on tests that are possible in a reasonably well equipped laboratory, but not one capable of reference facilities. Tests that are suitable for use in the field are highlighted. An evaluation of sensitivity and specificity is also given. Other tests available in central or reference laboratories are mentioned in brief, usually with supporting references.
Two annexes cover media, reagents and stains, and details of equipment required to diagnose each condition. A third annex is an interesting table of which tests should be available at "peripheral," "intermediate," and "central" laboratories.
Overall, this manual is to be welcomed as an educational and reference source for medical microbiologists, technologists, and clinicians. However, I would recommend that the authors "road test" the manual to discover omissions in technical detail that would prevent the sole use of the manual in the field.
IndiVerent colour reproduction detracts from the quality of the text-for example, blue reactions appearing as red in the figure.
For the next edition, a chapter on basic microscopical techniques and another on the general principles and interpretation of laboratory tests would provide useful introductions to an otherwise excellent publication. GEOFFREY  Each tutorial is self contained (which does lead to some duplication) and has self assessment questions-usually with click and drag matching of statements or true/false boxes. The information itself is well illustrated and contains animations and a video clip, together with further information/annotations in pop up boxes. At the end of each section there is a set of summary points, a reading list, and further activities such as internet sites.
There is a searchable picture index which allows you to search, view, and save sets of images for reference and lectures (although copyright does apply), and a glossary of terms.
Overall this is an excellent CD Rom providing good information, presented in an attractive and usable way, with a wealth of illustrations. I would strongly recommend it. This is an interesting CD Rom which, gives a very personal guide to issues surrounding HIV-covering the experience of the patient, carer and healthcare professionals.

SARAH EDWARDS
Four main sections cover the following areas: Living with HIV, Is HIV diVerent? Loss, grieving and bereavement, Supporting people aVected by HIV.
These areas are illustrated by short video clips and backed up by further information. Basic information is given about HIV treatment, the impact of diagnosis and of ill health, and other related topics. Unfortunately the information about drug treatment is already outdated and there is no search facility.
The strength of this CD Rom is the view it gives of the emotional responses to HIV and the strategies for coping with the infection from the viewpoint of those involved. The academic content is limited but it is worth a look for the patient perspectives.