CORRESPONDENCE * All letters must be typed with double spacing and signed by all authors. * No letter should be more than 400 words. * For letters on scientific subjects we normally reserve our correspondence columns for those relating to issues discussed recently (within six weeks) in the BMJ. * We do not routinely acknowledge letters. Please send a stamped addressed envelope ifyou would like an acknowledgment.
* Because we receive manv more letters than we can publish we may shorten those we do print, particularly when we receive several on the same subject.
Abolishing the prison medical service SIR,-The increasing suicide rate and level of mental illness in prisons make a parliamentary move to abolish the prison medical service and give its responsibilities to the NHS' fully understandable. Unfortunately, though, by itself such action would be most unlikely to improve the situation; the root cause is that prison is too often seen as a solution to society's unpleasant and difficult problems, whiWh other agencies cannot or will not deal with, and the system is thus overburdened. This is by no means the first time that it has been suggested that the health service should take over prison medicine, but the proposal has not been acted on seemingly because, firstly, it would cut across the borders of ministerial responsibility and, secondly, the NHS is understandably somewhat reluctant. It is in the large, busy remand prisons that most problems occur. These establishments can receive up to 100 or more new prisoners each night; the pressure so generated gives little opportunity for each inmate to make any consistent relationship with individual members of professional or other staff. Such prisons are labour intensive and expensive to run, and so the good housekeeping necessary means that manning levels must be reasonably controlled. That in turn means that to maintain good order inmates must be locked up for considerable periods. For disruptive, unstable people this can result in outbursts of violence towards themselves or others. Given the restricted ability to manipulate the environment in the short term, psychiatric options are therefore limited. Nevertheless, prison doctors and their staff, though in short supply for many years, have learnt to work within this restraint, placing those found to be mentally ill and disturbed in the prison hospitals for closer observation; these patients are then transferred to outside hospitals whenever possible. Greater understanding is therefore required of the difficulties inherent in running the prison system as a whole before its medical services are tampered with in isolation. D O TOPP
Scavnes Hill, Haywards Heath, Sussex RH 17 7NG I Smith R. Further call to abolish the prison medical service. BMJ 1991;302:1359. (8 June.)
Drug use and prisoners SIR,-As a part time prison medical officer, I was interested in Dr Pierre Van Damme and colleagues' finding of increasing drug use in Belgian prisons.
118
Their findings are similar to the findings of Dr Anthony Maden and colleagues in British prisons.2 One of the difficulties of estimating the prevalence of drug addiction by questioning or from questionnaires such as that used by Dr Maden and colleagues is that addiction is subjective. Unless valid consent is obtained for objective testing (such as testing of urine for drug residues) then in my view the extent of intravenous drug use before reception, or within prisons, will remain a matter of speculation. In view of the risk of transmission of HIV by illicit drug use in prisons it might be sensible to perform a pilot objective study in a local prison. P LUSH
HM Prison, Gloucester GL 1 2JN 1 Van Damme P, Clauwers M, Van Hal G, Peeters R. Drug use and prisoners. BMJ 1991;302:1464. (15 June.) 2 Maden A, Swinton M, Gunn J. Drug dependence in prisoners.
BMJ 1991;302:880. (13 April.)
Drugs, HIV, and prisons SIR,-The editorial by Drs Michael Farrell and John Strang on drugs, HIV, and prisons is a thoughtful contribution on an important subject,' but I cannot let pass the statement that prisons are a "gaping hole" in the "widening net of services" for injecting drug users. It is fair to say that provision in England and Wales-I cannot speak for Scotland-is less well developed in some prisons than others. But a void there is not, as their reference to work at Holloway prison acknowledges. It is also true that detoxification with methadone has not been practised everywhere, but that has reflected differences of view within the profession, not lack of knowledge or care on the part of prison doctors. The recent guidelines on drug users in prison for medical officers, to which Drs Farrell and Strang refer, encourage medical officers to develop an approach whereby such a programme is always considered. The authors damn with faint praise when they say that the guidelines are a start but then add that "prison medical officers need also to liaise with local specialists and ensure that they and prison officers receive training in managing drug use." In fact, medical officers are encouraged to liaise with local specialists, and they and other health care staff are receiving training. The charge of segregation of prisoners identified as being infected with HIV is misleading. No prisoner is segregated in the sense of having no association with other prisoners. In a recent survey of such prisoners that we commissioned, of 30 prisoners who replied to the question "With which prisoners are you currently able to associate?" 13 stated "all," 16 replied "selected prisoners only," and one replied "only those with HIV or subject to
viral infectivity restrictions." The viral infectivity restrictions system is to be reviewed, and in revised guidelines about HIV which I hope to issue to the service in the next few weeks I shall be emphasising again the policy of "normal" location of prisoners who are well. Finally, Drs Farrell and Strang comment about the prevalence of HIV infection and its relevance to policy. They mention figures suggesting that 24 HIV positive prisoners at Bristol prison identified themselves to prison officer counsellors whereas only two had been reported by the medical officer. The Prison Reform Trust deduced from these figures that the number of prisoners with HIV infection in England and Wales could be 12 times greater than the number officially recorded. That is not an inference that can be properly drawn from the Bristol figures. They are not comparable with the official figures because they are based on oral statements by prisoners unsubstantiated by results of tests. There may also have been some multiple counting because counsellors do not share information with each other. That said, we have always acknowledged that the confirmed cases reported by medical officers may be a small proportion of actual cases. As long ago as 1987 my predecessor estimated that the actual number at that time could be between 250 and 500 in a population of 50000. What is important is that policy and practice are based on the assumption that the prevalence of HIV infection is higher in the prison population than the general community. Our strategy for preventing HIV infection, in which education and counselling are key elements, reflects that assumption. More "client friendly" drug treatment programmes, desirable in themselves, will have an important part to play in furthering the strategy. ROSEMARY J WOOL
D)irector, Prison Medical Service, London SWIP 4LN I Farrell M, Strang J. Drugs, HIV, and prisons. B.&17 1991;302: 1477-8. 22 June.
Zidovudine after occupational exposure to HIV SIR,-Professor D J Jeffries proposes that prophylaxis with zidovudine should be available for laboratory and health care workers within one hour after they have been exposed to HIV.' Detailed counselling will be needed before prophylaxis is started because proof of efficacy is lacking and serious long term toxicity (particularly carcinogenicity) cannot be excluded. For laboratory workers propagating HIV or cells that might be infected with the virus counselling before any exposure occurs will remove the pressure of time from the counselling process and facilitate
BMJ VOLUME 303
13 JULY 1991
rapid initiation of prophylaxis after exposure in those who decide to receive it. Rapid delivery of prophylaxis will be much more difficult to achieve for health care workers. Professor Jeffries makes no detailed suggestions of how to do this. Options that are perhaps impracticable include counselling all health care staff before any exposure occurs and completing counselling and starting prophylaxis within the first hour after exposure. Two more feasible alternatives are available: accepting that the delay consequent on counselling and on getting the drug means that prophylaxis will not be started within one hour after exposure; and selecting for counselling before any exposure health workers who are considered to be at increased risk of exposure, and counselling after exposure those workers considered to be at lower risk, thus delaying prophylaxis for them. The second alternative is likely to be divisive (who is at increased risk of exposure?) and is difficult to justify as there is no proof that prophylaxis must be started within one hour (as opposed to, say, six hours). Until practical guidelines for starting prophylaxis with zidovudine in health care workers emerge implementation of Professor Jeffries's proposal will be delayed. DAVID J MORRIS
North Manchester Regional Virus Laboratory, Booth Hall Children's Hospital, Manchester M9 2AA 1 Jeffries DJ. Zidovudine after occupational exposure to HIV. BA!] 1991;302:1349-51. (8 June.)
SIR,-In his editorial Professor Don Jeffries reviews the conditions under which treatment with zidovudine can be considered after exposure to HIV through needlestick injuries. ' The BMA recommends that a prophylactic course of zidovudine should be considered if staff accidentally inoculate themselves with blood from a patient found to be positive for HIV.' Treatment with zidovudine is still experimental, and various prophylactic dose regimens are suggested.' Although a further failure of the drug to protect against seroconversion has recently been reported,4 its use for injured staff offers psychological reassurance and it is the only treatment currently available. Mangione et al recently reported that only 30% of needlestick injuries experienced by medical house staff at San Francisco General Hospital were reported, although 25% of junior staff were exposed to HIV infected blood.' They calculated that one in 1000 interns would become infected with HIV in the course of their work and that the fatality rate among interns could then be four times higher than that among police officers in
California. Underreporting of needlestick injuries was recognised as early as 1983, and continued failure to report exposures to blood infected with HIV leads to inaccurate estimation of the prevalence of exposure and subsequent underestimation of the overall occupational risk of acquiring HIV infection. Health care workers who fail to report forfeit the opportunity for early evaluation or treatment for exposure to hepatitis B virus and possible treatment with zidovudine. In a recent survey I found that 51 of 62 British health care workers indicated that they would report a needlestick injury; eight stated that they would probably report such an injury, depending on criteria such as its depth, whether the patient's blood had been transferred, and whether the patient was known to be positive for HIV or hepatitis B virus; and only four said that they would not report a needlestick injury.6 Interestingly, however, although 38% had suffered a needlestick injury in the past year, only 32 of 64 respondents had received a full course of hepatitis B vaccine and the remainder would therefore require immediate assessment in the event of exposure to hepatitis B virus.
BMJ
VOLUME
303
13
JULY
1991
It is surprising that after a decade of experience of HIV infection and AIDS in San Francisco American junior medical staff are still "too busy" or unaware of the health risks to report needlestick injuries and that American hospitals still need education programmes on infection control that supply information about the risk of seroconversion, provide psychological evaluation, and reinforce the importance of reporting all sharps injuries.5 In the United Kingdom the BMA's code of practice on sharps is available as a self contained training programme for managers to use with all health care staff.2 Until every hospital has instituted effective training and reporting procedures the actual hazards to staff and requirements for treatment after exposure will remain uncertain. DAVID R MORGAN Professional, Scientific, and International Affairs Division, BMA, London WC I H 9JP 1 Jeffries DJ. Zidovudine after occupational exposure to HIV. BM3r 1991;302:1349-50. (8 June.) 2 BMA. A code of practice for the safe use and disposal of sharps. London: BMA, 1990. 3 Singh VR, Shands JW, Jr. Risk of exposure of surgical personnel to patient's blood. N Englj Med 1991;324:57-8. 4 Durand E, Le Jeunne C, Hugues F-C. Failure of prophylactic zidovudine after suicidal self-inoculation of HIV-infected blood. N Englj Med 1991;324:1062. 5 Mangione CM, Gerberding JL, Cummings SR. Occupational exposure to HIV: frequency and rates of under reporting of percutaneous and mucocutaneous exposures by medical house staff. N EnglJ Med 1991;90:85-90. 6 Morgan DR. Preventing needlestick in'juries. BMJ 1991;302:1147. (11 May.)
SIR,-In his editorial Professor D J Jeffries suggests that hospitals should be able to give zidovudine within an hour after occupational exposure to HIV; he also makes it clear that to do this and give adequate counselling, especially at night, is difficult.' In this hospital, where the issue was considered over a year ago, it was decided that the consultant microbiologist on call was best placed to deal with the problem at night. A supply of zidovudine was made available in the pharmacy together with a counselling sheet, which sets out the evidence supporting the use of the drug and how to contact relevant staff. Our information made clear that there was no reason why the drug should not be started and then stopped after further discussion. We believe that it is important to get this across to staff as it alleviates the stress associated with such exposure and permits further review with the
advantage of reflection. HELEN HOLZEL ANTHONY HAINES Hospital for Sick Children, Great Ormond Street, London WC1N 3JH I Jeffries DJ. Zidovudine after occupational exposure to HIV.
BMJ 1991;302:1349-51. (8 June.)
Same day testing for HIV SIR,-We agree with Dr S B Squire and colleagues that the anxiety that people experience while waiting for the results of an HIV test is enormous. Any attempt to reduce this waiting time is probably of considerable benefit to patients' wellbeing, and Dr Squire and colleagues' efforts are admirable. We are concerned, however, about the separation of the HIV antibody testing from screening for sexually transmitted diseases. To screen for sexually transmitted diseases "if concerns about other sexually transmitted diseases were raised" is, we believe, unsatisfactory. Most HIV infection is sexually transmitted. Compared with the incidence of other sexually transmitted diseases, HIV infection is rare, and we suspect that with few exceptions
most patients requesting the tests have probably been at risk of other sexually transmitted diseases. It is possible to miss asymptomatic infections, particularly in women. Undiagnosed chlamydial infection mostly occurs in women and neonates. Though detecting a small proportion of people who are positive for HIV antibody, Dr Squire and colleagues are missing a large proportion of people with other sexually transmitted diseases with serious consequences. We analysed 159 consecutive people attending our clinic in April and May 1991 who requested an HIV antibody test, either for reassurance or because of their past sexual behaviour. Altogether 142 were asymptomatic, of whom (26%) had an associated sexually transmitted disease. There were 90 asymptomatic men, of whom three had Chlamydia trachomatis infection, one gonorrhoea, eight non-specific urethritis, and seven genital warts. One patient had both non-specific urethritis and warts. Of the 52 asymptomatic women, four had C trachomatis infection (one with genital warts as well) and 12 had vaginal candidiasis. Serological tests for syphilis gave negative results in all patients. HIV antibody tests and tests for hepatitis B gave negative results in all the women. Four homosexual men were positive for HIV antibody, and one for hepatitis B. Although the number of patients analysed is small, the results compare well with those reported in 1987.2 A higher prevalence (47%) of sexually transmitted diseases in asymptomatic people attending a clinic seeking HIV antibody testing has been reported.' As HIV infection is predominantly sexually transmitted and the risk of heterosexual transmission is increasing we believe that screening for other sexually transmitted diseases in patients seeking HIV antibody testing is, with a few exceptions, obligatory. S SIVAPALAN V HARINDRA R BASU ROY
Genitourinary Medicine, Royal Victoria Hospital, Bournemouth BH7 6JF 1 Squire SB, Elford J, Bor R, Tilsed G, Salt H, Bagdades EK, et al. Open access clinic providing HIV-1 antibody results on day of testing: the first twelve months. BMJ 1991;302:1383-6. (8 June.) 2 Coelho D, Roy RB. Incidences of STDs in worried well women requesting confidential HIV antibody tests. Hospital Doctor 1987;31: 14. 3 Mohanty KC. Sexually transmitted diseases among patients seeking HIV antibody tests for AIDS. International3journal of STD and AIDS 1990;1:207-8.
SIR,-Dr S B Squire and colleagues report the success of a clinic providing same day reporting of results of tests for HIV-I antibody.' They found that no serum specimen initially reactive in an enzyme linked immunosorbent assay for HIV-I antibody originated from a patient who was truly seronegative. Moreover, all serum specimens that gave initially equivocal results were from seronegative subjects. Our observations are rather different. We detected false positive results (albeit rarely) in two enzyme linked immunosorbent assays (Dupont and Wellcome) when screening specimens of serum collected from patients with a low prevalence of HIV-I antibody (those attending walk in self referral clinics for HIV antibody testing) or a very low prevalence of the antibody (those attending for in vitro fertilisation).2 The specificity of these assays increased when we used antigens derived from recombinant serum rather than antigens derived from infected cells, but it was still not absolute. Since our report we have begun to screen serum specimens from these clinics with a combined HIV-I and HIV-II antibody assay based on synthetic peptide (Biokit) and still find occasional false positive results (two reactive and two equivocal results among 411 truly negative specimens).
119
* Because we receive manv more letters than we can publish we may shorten those we do print, particularly when we receive several on the same subject.
Abolishing the prison medical service SIR,-The increasing suicide rate and level of mental illness in prisons make a parliamentary move to abolish the prison medical service and give its responsibilities to the NHS' fully understandable. Unfortunately, though, by itself such action would be most unlikely to improve the situation; the root cause is that prison is too often seen as a solution to society's unpleasant and difficult problems, whiWh other agencies cannot or will not deal with, and the system is thus overburdened. This is by no means the first time that it has been suggested that the health service should take over prison medicine, but the proposal has not been acted on seemingly because, firstly, it would cut across the borders of ministerial responsibility and, secondly, the NHS is understandably somewhat reluctant. It is in the large, busy remand prisons that most problems occur. These establishments can receive up to 100 or more new prisoners each night; the pressure so generated gives little opportunity for each inmate to make any consistent relationship with individual members of professional or other staff. Such prisons are labour intensive and expensive to run, and so the good housekeeping necessary means that manning levels must be reasonably controlled. That in turn means that to maintain good order inmates must be locked up for considerable periods. For disruptive, unstable people this can result in outbursts of violence towards themselves or others. Given the restricted ability to manipulate the environment in the short term, psychiatric options are therefore limited. Nevertheless, prison doctors and their staff, though in short supply for many years, have learnt to work within this restraint, placing those found to be mentally ill and disturbed in the prison hospitals for closer observation; these patients are then transferred to outside hospitals whenever possible. Greater understanding is therefore required of the difficulties inherent in running the prison system as a whole before its medical services are tampered with in isolation. D O TOPP
Scavnes Hill, Haywards Heath, Sussex RH 17 7NG I Smith R. Further call to abolish the prison medical service. BMJ 1991;302:1359. (8 June.)
Drug use and prisoners SIR,-As a part time prison medical officer, I was interested in Dr Pierre Van Damme and colleagues' finding of increasing drug use in Belgian prisons.
118
Their findings are similar to the findings of Dr Anthony Maden and colleagues in British prisons.2 One of the difficulties of estimating the prevalence of drug addiction by questioning or from questionnaires such as that used by Dr Maden and colleagues is that addiction is subjective. Unless valid consent is obtained for objective testing (such as testing of urine for drug residues) then in my view the extent of intravenous drug use before reception, or within prisons, will remain a matter of speculation. In view of the risk of transmission of HIV by illicit drug use in prisons it might be sensible to perform a pilot objective study in a local prison. P LUSH
HM Prison, Gloucester GL 1 2JN 1 Van Damme P, Clauwers M, Van Hal G, Peeters R. Drug use and prisoners. BMJ 1991;302:1464. (15 June.) 2 Maden A, Swinton M, Gunn J. Drug dependence in prisoners.
BMJ 1991;302:880. (13 April.)
Drugs, HIV, and prisons SIR,-The editorial by Drs Michael Farrell and John Strang on drugs, HIV, and prisons is a thoughtful contribution on an important subject,' but I cannot let pass the statement that prisons are a "gaping hole" in the "widening net of services" for injecting drug users. It is fair to say that provision in England and Wales-I cannot speak for Scotland-is less well developed in some prisons than others. But a void there is not, as their reference to work at Holloway prison acknowledges. It is also true that detoxification with methadone has not been practised everywhere, but that has reflected differences of view within the profession, not lack of knowledge or care on the part of prison doctors. The recent guidelines on drug users in prison for medical officers, to which Drs Farrell and Strang refer, encourage medical officers to develop an approach whereby such a programme is always considered. The authors damn with faint praise when they say that the guidelines are a start but then add that "prison medical officers need also to liaise with local specialists and ensure that they and prison officers receive training in managing drug use." In fact, medical officers are encouraged to liaise with local specialists, and they and other health care staff are receiving training. The charge of segregation of prisoners identified as being infected with HIV is misleading. No prisoner is segregated in the sense of having no association with other prisoners. In a recent survey of such prisoners that we commissioned, of 30 prisoners who replied to the question "With which prisoners are you currently able to associate?" 13 stated "all," 16 replied "selected prisoners only," and one replied "only those with HIV or subject to
viral infectivity restrictions." The viral infectivity restrictions system is to be reviewed, and in revised guidelines about HIV which I hope to issue to the service in the next few weeks I shall be emphasising again the policy of "normal" location of prisoners who are well. Finally, Drs Farrell and Strang comment about the prevalence of HIV infection and its relevance to policy. They mention figures suggesting that 24 HIV positive prisoners at Bristol prison identified themselves to prison officer counsellors whereas only two had been reported by the medical officer. The Prison Reform Trust deduced from these figures that the number of prisoners with HIV infection in England and Wales could be 12 times greater than the number officially recorded. That is not an inference that can be properly drawn from the Bristol figures. They are not comparable with the official figures because they are based on oral statements by prisoners unsubstantiated by results of tests. There may also have been some multiple counting because counsellors do not share information with each other. That said, we have always acknowledged that the confirmed cases reported by medical officers may be a small proportion of actual cases. As long ago as 1987 my predecessor estimated that the actual number at that time could be between 250 and 500 in a population of 50000. What is important is that policy and practice are based on the assumption that the prevalence of HIV infection is higher in the prison population than the general community. Our strategy for preventing HIV infection, in which education and counselling are key elements, reflects that assumption. More "client friendly" drug treatment programmes, desirable in themselves, will have an important part to play in furthering the strategy. ROSEMARY J WOOL
D)irector, Prison Medical Service, London SWIP 4LN I Farrell M, Strang J. Drugs, HIV, and prisons. B.&17 1991;302: 1477-8. 22 June.
Zidovudine after occupational exposure to HIV SIR,-Professor D J Jeffries proposes that prophylaxis with zidovudine should be available for laboratory and health care workers within one hour after they have been exposed to HIV.' Detailed counselling will be needed before prophylaxis is started because proof of efficacy is lacking and serious long term toxicity (particularly carcinogenicity) cannot be excluded. For laboratory workers propagating HIV or cells that might be infected with the virus counselling before any exposure occurs will remove the pressure of time from the counselling process and facilitate
BMJ VOLUME 303
13 JULY 1991
rapid initiation of prophylaxis after exposure in those who decide to receive it. Rapid delivery of prophylaxis will be much more difficult to achieve for health care workers. Professor Jeffries makes no detailed suggestions of how to do this. Options that are perhaps impracticable include counselling all health care staff before any exposure occurs and completing counselling and starting prophylaxis within the first hour after exposure. Two more feasible alternatives are available: accepting that the delay consequent on counselling and on getting the drug means that prophylaxis will not be started within one hour after exposure; and selecting for counselling before any exposure health workers who are considered to be at increased risk of exposure, and counselling after exposure those workers considered to be at lower risk, thus delaying prophylaxis for them. The second alternative is likely to be divisive (who is at increased risk of exposure?) and is difficult to justify as there is no proof that prophylaxis must be started within one hour (as opposed to, say, six hours). Until practical guidelines for starting prophylaxis with zidovudine in health care workers emerge implementation of Professor Jeffries's proposal will be delayed. DAVID J MORRIS
North Manchester Regional Virus Laboratory, Booth Hall Children's Hospital, Manchester M9 2AA 1 Jeffries DJ. Zidovudine after occupational exposure to HIV. BA!] 1991;302:1349-51. (8 June.)
SIR,-In his editorial Professor Don Jeffries reviews the conditions under which treatment with zidovudine can be considered after exposure to HIV through needlestick injuries. ' The BMA recommends that a prophylactic course of zidovudine should be considered if staff accidentally inoculate themselves with blood from a patient found to be positive for HIV.' Treatment with zidovudine is still experimental, and various prophylactic dose regimens are suggested.' Although a further failure of the drug to protect against seroconversion has recently been reported,4 its use for injured staff offers psychological reassurance and it is the only treatment currently available. Mangione et al recently reported that only 30% of needlestick injuries experienced by medical house staff at San Francisco General Hospital were reported, although 25% of junior staff were exposed to HIV infected blood.' They calculated that one in 1000 interns would become infected with HIV in the course of their work and that the fatality rate among interns could then be four times higher than that among police officers in
California. Underreporting of needlestick injuries was recognised as early as 1983, and continued failure to report exposures to blood infected with HIV leads to inaccurate estimation of the prevalence of exposure and subsequent underestimation of the overall occupational risk of acquiring HIV infection. Health care workers who fail to report forfeit the opportunity for early evaluation or treatment for exposure to hepatitis B virus and possible treatment with zidovudine. In a recent survey I found that 51 of 62 British health care workers indicated that they would report a needlestick injury; eight stated that they would probably report such an injury, depending on criteria such as its depth, whether the patient's blood had been transferred, and whether the patient was known to be positive for HIV or hepatitis B virus; and only four said that they would not report a needlestick injury.6 Interestingly, however, although 38% had suffered a needlestick injury in the past year, only 32 of 64 respondents had received a full course of hepatitis B vaccine and the remainder would therefore require immediate assessment in the event of exposure to hepatitis B virus.
BMJ
VOLUME
303
13
JULY
1991
It is surprising that after a decade of experience of HIV infection and AIDS in San Francisco American junior medical staff are still "too busy" or unaware of the health risks to report needlestick injuries and that American hospitals still need education programmes on infection control that supply information about the risk of seroconversion, provide psychological evaluation, and reinforce the importance of reporting all sharps injuries.5 In the United Kingdom the BMA's code of practice on sharps is available as a self contained training programme for managers to use with all health care staff.2 Until every hospital has instituted effective training and reporting procedures the actual hazards to staff and requirements for treatment after exposure will remain uncertain. DAVID R MORGAN Professional, Scientific, and International Affairs Division, BMA, London WC I H 9JP 1 Jeffries DJ. Zidovudine after occupational exposure to HIV. BM3r 1991;302:1349-50. (8 June.) 2 BMA. A code of practice for the safe use and disposal of sharps. London: BMA, 1990. 3 Singh VR, Shands JW, Jr. Risk of exposure of surgical personnel to patient's blood. N Englj Med 1991;324:57-8. 4 Durand E, Le Jeunne C, Hugues F-C. Failure of prophylactic zidovudine after suicidal self-inoculation of HIV-infected blood. N Englj Med 1991;324:1062. 5 Mangione CM, Gerberding JL, Cummings SR. Occupational exposure to HIV: frequency and rates of under reporting of percutaneous and mucocutaneous exposures by medical house staff. N EnglJ Med 1991;90:85-90. 6 Morgan DR. Preventing needlestick in'juries. BMJ 1991;302:1147. (11 May.)
SIR,-In his editorial Professor D J Jeffries suggests that hospitals should be able to give zidovudine within an hour after occupational exposure to HIV; he also makes it clear that to do this and give adequate counselling, especially at night, is difficult.' In this hospital, where the issue was considered over a year ago, it was decided that the consultant microbiologist on call was best placed to deal with the problem at night. A supply of zidovudine was made available in the pharmacy together with a counselling sheet, which sets out the evidence supporting the use of the drug and how to contact relevant staff. Our information made clear that there was no reason why the drug should not be started and then stopped after further discussion. We believe that it is important to get this across to staff as it alleviates the stress associated with such exposure and permits further review with the
advantage of reflection. HELEN HOLZEL ANTHONY HAINES Hospital for Sick Children, Great Ormond Street, London WC1N 3JH I Jeffries DJ. Zidovudine after occupational exposure to HIV.
BMJ 1991;302:1349-51. (8 June.)
Same day testing for HIV SIR,-We agree with Dr S B Squire and colleagues that the anxiety that people experience while waiting for the results of an HIV test is enormous. Any attempt to reduce this waiting time is probably of considerable benefit to patients' wellbeing, and Dr Squire and colleagues' efforts are admirable. We are concerned, however, about the separation of the HIV antibody testing from screening for sexually transmitted diseases. To screen for sexually transmitted diseases "if concerns about other sexually transmitted diseases were raised" is, we believe, unsatisfactory. Most HIV infection is sexually transmitted. Compared with the incidence of other sexually transmitted diseases, HIV infection is rare, and we suspect that with few exceptions
most patients requesting the tests have probably been at risk of other sexually transmitted diseases. It is possible to miss asymptomatic infections, particularly in women. Undiagnosed chlamydial infection mostly occurs in women and neonates. Though detecting a small proportion of people who are positive for HIV antibody, Dr Squire and colleagues are missing a large proportion of people with other sexually transmitted diseases with serious consequences. We analysed 159 consecutive people attending our clinic in April and May 1991 who requested an HIV antibody test, either for reassurance or because of their past sexual behaviour. Altogether 142 were asymptomatic, of whom (26%) had an associated sexually transmitted disease. There were 90 asymptomatic men, of whom three had Chlamydia trachomatis infection, one gonorrhoea, eight non-specific urethritis, and seven genital warts. One patient had both non-specific urethritis and warts. Of the 52 asymptomatic women, four had C trachomatis infection (one with genital warts as well) and 12 had vaginal candidiasis. Serological tests for syphilis gave negative results in all patients. HIV antibody tests and tests for hepatitis B gave negative results in all the women. Four homosexual men were positive for HIV antibody, and one for hepatitis B. Although the number of patients analysed is small, the results compare well with those reported in 1987.2 A higher prevalence (47%) of sexually transmitted diseases in asymptomatic people attending a clinic seeking HIV antibody testing has been reported.' As HIV infection is predominantly sexually transmitted and the risk of heterosexual transmission is increasing we believe that screening for other sexually transmitted diseases in patients seeking HIV antibody testing is, with a few exceptions, obligatory. S SIVAPALAN V HARINDRA R BASU ROY
Genitourinary Medicine, Royal Victoria Hospital, Bournemouth BH7 6JF 1 Squire SB, Elford J, Bor R, Tilsed G, Salt H, Bagdades EK, et al. Open access clinic providing HIV-1 antibody results on day of testing: the first twelve months. BMJ 1991;302:1383-6. (8 June.) 2 Coelho D, Roy RB. Incidences of STDs in worried well women requesting confidential HIV antibody tests. Hospital Doctor 1987;31: 14. 3 Mohanty KC. Sexually transmitted diseases among patients seeking HIV antibody tests for AIDS. International3journal of STD and AIDS 1990;1:207-8.
SIR,-Dr S B Squire and colleagues report the success of a clinic providing same day reporting of results of tests for HIV-I antibody.' They found that no serum specimen initially reactive in an enzyme linked immunosorbent assay for HIV-I antibody originated from a patient who was truly seronegative. Moreover, all serum specimens that gave initially equivocal results were from seronegative subjects. Our observations are rather different. We detected false positive results (albeit rarely) in two enzyme linked immunosorbent assays (Dupont and Wellcome) when screening specimens of serum collected from patients with a low prevalence of HIV-I antibody (those attending walk in self referral clinics for HIV antibody testing) or a very low prevalence of the antibody (those attending for in vitro fertilisation).2 The specificity of these assays increased when we used antigens derived from recombinant serum rather than antigens derived from infected cells, but it was still not absolute. Since our report we have begun to screen serum specimens from these clinics with a combined HIV-I and HIV-II antibody assay based on synthetic peptide (Biokit) and still find occasional false positive results (two reactive and two equivocal results among 411 truly negative specimens).
119
