- Email: [email protected]
HIV infection in children—impact upon ENT doctors
International Journal of Pediatric Otorhinolaryngology (2003) 67S1, S85—S90
HIV infection in children–—impact upon ENT doctors Simon Hoare* Paediatric Infectious Diseases Unit, Newcastle General and Queen Elizabeth Hospitals, Tyne & Wear, 26A Heddon Banks, Heddon-on-the-Wall, Northumberland NE15 0BU, UK
KEYWORDS HIV; AIDS; ENT; Child(-hood, -ren)
Summary The global epidemic of HIV infection remains appalling. By 2001, there were an estimated 1.4 million HIV-infected children, with 4.5 million deaths. In the UK, paediatric cases are clustered around population centres where there are high concentrations of infected immigrant adults, and to a lesser extent, areas where IV drug abuse is common. The highest incidence remains in London and the southeast. With the national redistribution of immigrant and refugee families, any doctor in any speciality may expect to be involved with children who are HIV positive, or have clinical AIDS. The majority of children are infected vertically, i.e. infection of the infant from an infected mother in the pre-, peri-, or post-natal periods. Rates of transmission vary from 15—20% in the developed countries. Children with HIV infection may have their primary presentation to ENT doctors, who should have appropriate thresholds for suspecting the diagnosis. The most common presenting features include persistent generalised lymphadenopathy, hepatosplenomegaly, chronic/recurrent diarrhoea, poor growth, and fever. Fifteen to twenty percent of untreated children will present with an AIDS-defining illness by 12 months, typically with Pneumocystis pneumonia at ∼3—4 months of age. Seventy percent of perinatally infected children will exhibit some signs or symptoms by 12 months Without treatment, the median age to progression to AIDS is ∼6 years, and 25—30% will have died by this age. The median age of death is ∼9 years. Children may also present with repeated/unusual ear infections, sinus disease (inc. mastoiditis), tonsillitis, orbital/peri-orbital cellulitis, oral candidiasis, and dental infections. Infections with streptococcus pneumoniae and group A streptococcus are common, and often progress to severe systemic infection with an appreciable mortality. Infections may be due to unusual pathogens such as Pseudomonas, ‘typical’ and atypical Mycobacteria, Candida, Aspergillus, etc. Fungal infections of the sinuses (inc. Aspergillus and Rhizopus spp.) may be particularly devastating, with rapid spread to involve bone and the central nervous system. Another classical presentation, which may present to ENT doctors, is that of bilateral parotid enlargement, especially in children who are ‘slow progressors’, many of whom also have Lymphoid Interstitial Pneumonitis (LIP). A major attitudinal change has occurred due to advances in 3 main areas: (i) the multidisciplinary management of the infected mother (inc. counselling, antenatal screening, elective caesarean section, advising against breast feeding, etc.), (ii) the prevention of vertical transmission, using anti-retroviral therapy to the infected mother during pregnancy, and to the
∗ Tel:
+44-1661-852762. E-mail address: [email protected] (S. Hoare). 0165-5876/$ — see front matter © 2003 British Association for Paediatric Otorhinolaryngology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2003.08.036
S86
S. Hoare potentially infected infant in the first weeks of life, and (iii) major advances due to the advent of highly active anti-retroviral treatment. With effective use of these measures, transmission rates may be reduced to <2%. None of the measures though, affect a cure, and it will still be many years before the development of effective vaccines. ENT doctors may be referred children already known to be HIV-positive. Knowing how to talk to infected children (and their parents) is full of potential pitfalls, and requires careful forethought. Many infection-control policies have required considerable rethinking due to the AIDS epidemic. This has especially been the case with respect to needle-stick injuries, post-exposure prophylaxis, sterilization and re-use of equipment, and safe approaches to surgery. © 2003 British Association for Paediatric Otorhinolaryngology. Published by Elsevier Ireland Ltd. All rights reserved.
The impact of the HIV epidemic has ramifications for all of us. This short review approaches the topic in two parts: (1) an overview of HIV infection in children in the UK and abroad, and (2) the special relevance of HIV infection to ENT doctors. To begin, it is useful to look back over the progress of this infection over the last 20 or so years. It was in the early 1980s when a number of pharmacists in the USA noted increasing requests for intravenous co-trimoxazole that the first alarm bells started to ring. This was due to hitherto unanticipated numbers of cases of Pneumocystis pneumonia, particularly found in members of the homosexual community. These patients had an unusual immune deficiency, not previously recognised. By 1982, the acquired immune deficiency syndrome (AIDS) was defined for the first time and within a year of this, the human immuno-deficiency virus (HIV) was identified as its cause. By mid-1983 the presence of a heterosexual AIDS epidemic in Africa was revealed, the appalling impact of which was not to be appreciated by many for a significant number of years afterwards. By early 1985, Rock Hudson was the first of many public figures to disclose that he had AIDS, and later that year at least one case of HIV/AIDS had been reported from each region of the world. By 1987, the first therapy for AIDS (Zidovudine) was approved for use in the USA, and by 1994 the first treatment regimes to reduce mother-to-child transmission were introduced. By 1996, highly active anti-retroviral therapy (HAART) was first postulated, and by the late 1990s the very earliest vaccine trials were underway. AIDS has been defined in different ways over the years. It is important to make the distinction between HIV infection, during which many patients will be asymptomatic, versus clinical AIDS when infection is accompanied by signs and symptoms of an ‘AIDS-indicator disease’. A frequently used definition is ‘‘an illness characterised by one or more indicator diseases . . . with/without laboratory evidence of HIV infection (in the absence of another cause of immuno-deficiency)’’. The Centre
for Disease Control (CDC) has also used an extension of its definition to include all persons who are severely immuno-compromised (CD4 count of less than 200 × 106 l−1 ), irrespective of indicator disease. With special relevance to children, there are a large number of AIDS-defining conditions, some of which would indicate a presumptive diagnosis while others would lead to a definitive one. These presuppose the laboratory evidence of HIV infection. Of particular relevance to children is the development of recurrent and/or multiple bacterial infection in a child less than 13 years of age, primary cerebral lymphoma at any age, non-Hodgkins lymphoma, disseminated mycobacterial infection, and the HIV wasting syndrome/failure to thrive, which are all recognised AIDS-defining conditions. The route of transmission of the virus is well known to most clinicians. It includes sexual intercourse (whether anal or vaginal), contaminated needles (whether via intravenous drug usage, needle stick injuries or other injections), mother-to-child (vertical) transmission which may occur in utero (at birth or postnatally), and via organ and tissue donation (including blood products). For the sake of this title, it is mother-to-child transmission, and to a lesser extent needle stick injuries, that are of the most relevance. It is important to review the natural history of infection in adults and older children, which is different to that in infants. Classically, with early infection there is an increase in viral load with subsequent stabilisation. Although the viral load may continue in these stable numbers for many years, this is not insignificant infection and the number of virus particles is significantly high throughout. At the time of this initial transmission the CD4 count may drop transiently and then recover. This initial infection may be accompanied by fairly non-specific signs and symptoms, which may not come to medical carers attention at all. There is then a long, relatively asymptomatic period measured in years. This is then followed by the development of symptomatic AIDS, the hallmarks of which are an
HIV infection in children–—impact upon ENT doctors
S87
increasing viral load accompanied by immunodeficiency related to a fall in CD4 lymphocyte count. This, in a very simplistic way, is the natural history of the infection. In infants and children, however, this time scale is not seen and there is a rapid progression from birth to symptomatology in untreated patients. Thus, 15—20% of children develop AIDS-defining illnesses by the age of 12 months if they have been vertically infected. It is common for such children to present with Pneumocystis carinii pneumonia at about 3—4 months. About 70% overall will be symptomatic by the time they are 1 year old, and the median progression to clinical AIDS is approximately 6 years. Twenty-five to thirty percent of untreated children, especially in a Third World setting, may have died by this age. Over the course of this epidemic, though, it has become obvious that there are some children who represent slow versus rapid progression of the disease. The underlying reasons for this remain unclear, and certainly rapid progression seems to be a more major problem in the developing countries of the world. The global impact of this epidemic is appalling. By the end of the year 2000, an estimated 34.7 million adults and 1.4 million children were living with HIV/AIDS. It was estimated that, in that year alone, there were going to be 5.3 million new infections, representing about 16,000 per day. About 95% of all cases occur in the developing countries, and of these the major brunt of the epidemic is in sub-Saharan Africa. By the end of 1999, there were an estimated 13.2 million AIDS orphans, 2 million of which were in Southern Africa alone. The epidemic is anticipated to lead to an average expected reduction in life expectancy of approximately 17 years. The estimated impact on under-5 mortality rates has been assessed based on models with or without HIV infection and AIDS. Using Botswana as an example, there are approximately 30 deaths per thousand live births in the under-5 age group due to illnesses excluding AIDS, but this rises to greater than 150 deaths per thousand live births if AIDS is taken into account. Turning to Europe there are wide geographical variations in cases with approximately 90 per million cases in Spain and Portugal, falling to just over 10 per million in the UK and Holland. The UK HIV data base has recorded just over 51,000 reports of HIV-infected individuals from 1982 to 2002; 59% of whom have the HIV infection alone and 37% of whom have progressed to full-blown AIDS. Of the latter, 12,452 (665) have died. In the year 2000 in the UK, for the first time heterosexual cases had overtaken those in the homosexual population. About 70% of cases of AIDS and HIV infection occur and are seen in the Thames re-
gion. There is very encouraging data showing that the advent of HAART has led to increased life expectancy, but this does of course increase the pool of those who are infected and potentially infectious to others. Therefore, although great inroads have been made in terms of anti-retroviral treatment, primary and secondary preventative measures remain a formidable and continuing challenge. Comparing rates of HIV-infected individuals from the year of the first reported UK diagnosis, there have been significant changes. Comparing 1993—2001, rates occurring in the mother-to-infant group rose from 66 to 74 with a total of 861 cases overall. Rates amongst intravenous drug users had fallen from 204 to 104, but rather worryingly UK-acquired cases with no evidence of high-risk partners had risen from 63 to 165. Rates of those practising sex between men had fallen slightly from 1497 to 1415. In terms of the regional distribution of cases over this time period there was a total of 46,787 in England. Over 31,000 of these cases were in London: over the rest of the country the totals over this period range from 1624 (South West) to 2742 in the North West. Although paediatric data is limited, there are interesting trends with the introduction of HAART and active measures to prevent vertical transmission. Thus, from 1994 onwards there has been an increase in new diagnoses of HIV infection. However, over that same time scale the number of new reports of progression to AIDS has actually fallen as have the deaths related to AIDS. These figures may represent the fact that people are being diagnosed more efficiently, but also that progression and mortality is being influenced by such early diagnosis and subsequent aggressive management. This pattern is mirrored in that though the number of children born to HIV-infected mothers is rising, progression to HIV infection, AIDS and mortality is falling. Any attempt to influence progression of infection to AIDS is strengthened by detecting those mothers who are HIV-infected already. Historically in England the pick-up rate for HIV testing antenatally was poor, however, with the introduction of routine antenatal screening the total proportion of infection diagnosed has steadily risen from just over 40—70% or more. Many of these patients represent those who hitherto would have been undiagnosed and, therefore, preventative measures would not have been in place. This pattern is well demonstrated for inner London, but similar patterns has been shown for outer London, the rest of England and Wales and for Scotland. Partly linked with the timing of routine antenatal testing, children born in the UK are only recently showing a decrease in the number of cases per year in comparison to countries such as
S88 Spain, France and Italy. However, all of these European countries are showing significant decreases in cases with the advent of preventative treatment measures. In untreated mothers who are not actively managed, infection is transmitted to about 15—20% of their babies. The baseline rate is thought to double if the baby is also breastfed. In non-breastfed infants about 70% of the transmission occurs around the time of delivery. There are factors that increase the chances of transmission. These include a high viral load in the mother with a low CD4 count, clinical AIDS and prematurity. Also, the method of delivery has been felt to be important in transmission and most babies in the UK are now delivered by elective caesarean section. Alternatives to breastfeeding are also recommended in the UK, but obviously could not be applied to many parts of the world. Anti-retroviral medicines given to the mother and then to the baby, both before and after delivery, have played an important part in the reduction of the transmission rate. Implementing these three measures (not breastfeeding, anti-retroviral therapy for mother and baby and managed delivery) can mean a reduction in transmission rates to 2% or less, representing a phenomenally successful intervention. Highly active anti-retroviral treatment has become the accepted standard in most of the developed countries. There are various combinations of nucleoside and non-nucleoside reverse transcriptase inhibitors that are commonly used for children. By and large, the protease inhibitors tend to be avoided as first line therapy because of attendant side effects; therefore, most regimes are PI-sparing at the moment. There are often compliance and formulation problems with paediatric patients and there is a growing background of metabolic disturbances and lipodystrophy. Despite these problems, it is interesting to note that about 25% of the estimated 10,000 children in the United States who are HIV-infected and have progressed to clinical AIDS, have now survived to the teenage years. This then is the background of HIV/AIDS globally, and in the UK in particular. We now turn to the special relevance of this infection to doctors who specialise in oto-rhinolaryngology. In a study from Great Ormond Street Hospital of 66 paediatric patients over an 8-year period, 91% of these already-diagnosed children had ENT problems. The commonest were cervical lymphadenopathy, oesophageal candidiasis and recurrent otitis media. There were also some specific presentations, such as parotid swelling. In a study of 40 Brazilian children, ENT manifestations reached up to 100% of infected individuals and the majority had
S. Hoare their first manifestation of such ENT problems before HIV infection was diagnosed. They also noted a high occurrence of Herpes simplex gingivostomatitis and parotid hypertrophy. These two studies showed, therefore, that children who are infected with HIV suffer from common ENT illnesses, as well as more specific presentations. Again it is emphasised, the ENT condition may often be the initial presentation of HIV/AIDS. The pathogens involved include both common and uncommon organisms. Pneumococcal, Staphylococcal and Pseudomonal infections are especially important, and the latter in particular may run a very rapid clinical course. In the ill, HIV-infected child, empirical antibiotic treatment certainly needs to cover these organisms. There is often a requirement for broad-spectrum treatment and, especially in the case of Pseudomonas, dual synergistic therapy is often recommended. Other common infections include those with Haemophilus influenzae (including type B, since immunisation may not give appropriate protection in these children), the group A streptococcus, gram negative organisms, mycobacterial species, (including tuberculosis and atypical mycobacteria) anaerobes and fungi. Of the latter, Candida, Aspergillis and Rhizopus species are especially important pathogens with a high degree of morbidity and mortality. Viral infections are common, and pathogens include influenza and parainfluenza, RSV, adenovirus, CMV, EBV and HHV6, 7 and 8. The clinical course of these infections may be unusual or prolonged due to the underlying degree of immunity. In essence, the lower the CD4 count the greater the likelihood of overwhelming infection, or prolonged clinical course of infection with unusual organisms. Otitis and hearing loss are common in infected children. In one study of 10 children with symptomatic AIDS, otitis media was universal and 8 of the 10 also had persistent middle ear effusions. The aetiology of these effusions remains unclear and may be linked with humoral immunodeficiency, with Eustachian tube dysfunction, or possibly with lymphoid hyperplasia obstructing normal drainage. Six of seven children examined in this study also had conductive deafness, ranging between 10 and 65 dB losses. Sinusitis is frequent as an isolated finding. However, approximately 20% of episodes of serious bacterial sepsis may be associated with an underlying sinusitis. Aspirates most frequently show Streptococcus pneumonia, group G streptococci, Staphylococcus aureus and Staphylococcus epidermidis. Fungal infection of the sinuses is especially important. Aspergillis and Mucor infections, in particular, can cause rampaging infection with rapid extra
HIV infection in children–—impact upon ENT doctors
S89
sinus spread including that across tissue planes with extension into the central nervous system. In this scenario, the outlook is grim indeed. Allergic rhinitis is extremely common and markedly over represented in this population. It usually responds reasonably well to standard treatment and probably represents a degree of immune dysregulation. Bilateral parotid swelling is a classic feature of HIV/AIDS in children. It is seen to a much smaller degree in adults. It is associated with lymphoid interstitial pneumonitis, and is often seen in those children who are termed longer term or slow progressers. Clubbing is a frequent association. Such slow progressers may take several years to become obvious, and have CD4 counts and viral loads that alter over several years rather than months. Eventually, though, their CD4 counts decay and allow more classical clinical pictures of AIDS to appear. The parotid swelling associated with these children may be due to the presence of lymphoid tissue in the parotid gland that hypertrophies. Suppurative paratitis is much less common and usually unilateral. In terms of oral and throat disease, a wide range of problems have been associated with AIDS in the past including hairy oral leukoplakia, dental abscesses and caries, gingivitis, candidiasis and ulceration due to a variety of factors (including bacterial, herpetic and apthous ulcers). Nutritional problems are common in this group of children, remembering that failure to thrive and the HIV wasting syndrome are common presenting features. Kaposi’s sarcoma is much less common in the paediatric age range, as is Hodgkin’s disease and non-Hodgkin’s lymphoma. In terms of throat problems, bacterial and viral infection with group A streptococcus, staphylococcus, anaerobes and EBV, etc. are all common presenting clinically as tonsillitis and quinsy. In a very recent paper looking at laryngeal obstruction in 38 South African children, HIV infection was present in half of the patients admitted with this disorder to a paediatric intensive care unit. Obstruction related to lymphoid hyperplasia and orolaryngoesophageal candidiasis was over represented in the HIV-infected group. These children, however, had very similar requirements for supportive management, for the length of intubation and for ventilatory pressures used. The consensus of the authors was that even in countries with a very high HIV burden, children presenting with laryngeal obstruction do as well as their non-infected counterparts and certainly should be offered full supportive management. Another aspect of the interface between ENT doctors and children with HIV infection is the risk of HIV transmission to medical carers. In an anonymous HIV survey study of pa-
tients in Glasgow from 1992 to 1997, prevalence rates were applied to factors influencing risk. The study used the probability of percutaneous injury, the number of injuries sustained and whether post-exposure prophylaxis was taken. On the basis of three percutaneous injuries per year, specialists in renal and urological surgery who did not take any kind of post-exposure prophylaxis might expect a transmission rate of 1:2,000,000. For doctors working in ENT, though, this rate was estimated to be 1:200,000. Giving post-exposure prophylaxis to ENT doctors, though, would alter the transmission rate to 1:1,000,000. It is emphasised that ENT doctors who may be working with children who are HIV-infected need to practise excellent preventative techniques, but also be aware of their local unit’s policies and availability of medicines, guidance, advice, etc. with respect to post-exposure prophylaxis. ENT neoplasia is relatively uncommon in children who are HIV-infected. Kaposi sarcoma, lymphomas and particular B cell lymphomas are seen though. Previously, the development of neoplasia was associated with advanced disease and profound CD4-penia. It will be interesting to see if rates for neoplastic disorders alter with HAART. There is some early evidence that rates for Kaposi’s sarcoma may be influenced by aggressive anti-retroviral treatment, but the situation is less clear for other types of tumours. Finally, it is important to emphasise the psychodynamics of a consultation between parents and children who may be infected, and ENT doctors. This is unlikely for the majority of doctors in your speciality to be a frequent occurrence, but as we have discussed could happen any time, any place, anywhere. Several papers published in ENT journals have specifically addressed this issue, emphasising yet again that the otorhinolaryngologist may be the first physician to evaluate the patient with undiagnosed HIV infection. It is therefore important for ENT doctors to consider the diagnosis and to be aware of the clinical presentations. Many children will already be known to other specialities and good inter-disciplinary communication is essential. Some authors have emphasised that communication should also provide information and support for physicians who are ‘‘uncomfortable with the challenges of this epidemic’’. This infection and this epidemic are not going to go away in the near future. However, there is evidence of increasing medium- to long-term survival in the paediatric age group in those countries that can afford the preventative and treatment options outlined previously. It will be very important to see how the emergence of viral resistance and
S90 new medicines available for treatment affect this situation. It is already interesting that there are increasing numbers of women who are known to be HIV-infected who are now choosing to get pregnant because of an appreciable increase in confidence in anti-retroviral treatment. It will also be interesting if there are increasing numbers of non-infective sequelae in long-term survivors, such as neoplastic problems. For the majority of the world though, HIV remains an unpreventable, untreated and untreatable infection. In conclusion, the HIV epidemic emerged as a ‘‘bolt from the blue’’. In the latter 1970s, it would have been a brave physician to have predicted that within a decade the world would be swept by a previously unknown infectious disease, which essentially took the whole medical fraternity by surprise. More than 20 years after these events, the global pandemic of HIV remains appalling, remembering that more than 90% of AIDS occurs in the developing countries of the world. This epidemic is now leading to a situation where entire nations and governments fall under its influence, and there is an almost evolutionary change in population patterns that will have ramifications for many years. In terms of global control, there are few glimmers of hope. There is little in the way of medical advances that
S. Hoare are likely to change the picture for the foreseeable future in the developing countries, and much of the efforts towards control must still rest on prevention rather than cure or treatment. In the UK infected children are still mainly concentrated in the south east of the country, and especially in London. However, this is a situation that is dynamic and likely to alter with the redistribution of immigrant and refugee families nationally. In the UK, there is a sense of cautious optimism in terms of holding disease progression at bay with HAART, and a major sense of optimism in terms of prevention of mother-to-child transmission. At any stage, though, it is entirely feasible for HIV-infected children to present to ENT doctors before or after the diagnosis has been made. Such doctors must therefore ensure that they are able to suspect the infection clinically, to know the ENT manifestations of infection and disease, and to maintain an appropriate core of knowledge and communication relevant to HIV infection. Whilst there are many emotional and psychological upsets associated with those children with severe disease progression, there are many rewards and pleasures in dealing with the children and families who have this infection. It is up to all medical carers to put themselves in the position to recognise and manage such patients.
HIV infection in children–—impact upon ENT doctors Simon Hoare* Paediatric Infectious Diseases Unit, Newcastle General and Queen Elizabeth Hospitals, Tyne & Wear, 26A Heddon Banks, Heddon-on-the-Wall, Northumberland NE15 0BU, UK
KEYWORDS HIV; AIDS; ENT; Child(-hood, -ren)
Summary The global epidemic of HIV infection remains appalling. By 2001, there were an estimated 1.4 million HIV-infected children, with 4.5 million deaths. In the UK, paediatric cases are clustered around population centres where there are high concentrations of infected immigrant adults, and to a lesser extent, areas where IV drug abuse is common. The highest incidence remains in London and the southeast. With the national redistribution of immigrant and refugee families, any doctor in any speciality may expect to be involved with children who are HIV positive, or have clinical AIDS. The majority of children are infected vertically, i.e. infection of the infant from an infected mother in the pre-, peri-, or post-natal periods. Rates of transmission vary from 15—20% in the developed countries. Children with HIV infection may have their primary presentation to ENT doctors, who should have appropriate thresholds for suspecting the diagnosis. The most common presenting features include persistent generalised lymphadenopathy, hepatosplenomegaly, chronic/recurrent diarrhoea, poor growth, and fever. Fifteen to twenty percent of untreated children will present with an AIDS-defining illness by 12 months, typically with Pneumocystis pneumonia at ∼3—4 months of age. Seventy percent of perinatally infected children will exhibit some signs or symptoms by 12 months Without treatment, the median age to progression to AIDS is ∼6 years, and 25—30% will have died by this age. The median age of death is ∼9 years. Children may also present with repeated/unusual ear infections, sinus disease (inc. mastoiditis), tonsillitis, orbital/peri-orbital cellulitis, oral candidiasis, and dental infections. Infections with streptococcus pneumoniae and group A streptococcus are common, and often progress to severe systemic infection with an appreciable mortality. Infections may be due to unusual pathogens such as Pseudomonas, ‘typical’ and atypical Mycobacteria, Candida, Aspergillus, etc. Fungal infections of the sinuses (inc. Aspergillus and Rhizopus spp.) may be particularly devastating, with rapid spread to involve bone and the central nervous system. Another classical presentation, which may present to ENT doctors, is that of bilateral parotid enlargement, especially in children who are ‘slow progressors’, many of whom also have Lymphoid Interstitial Pneumonitis (LIP). A major attitudinal change has occurred due to advances in 3 main areas: (i) the multidisciplinary management of the infected mother (inc. counselling, antenatal screening, elective caesarean section, advising against breast feeding, etc.), (ii) the prevention of vertical transmission, using anti-retroviral therapy to the infected mother during pregnancy, and to the
∗ Tel:
+44-1661-852762. E-mail address: [email protected] (S. Hoare). 0165-5876/$ — see front matter © 2003 British Association for Paediatric Otorhinolaryngology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2003.08.036
S86
S. Hoare potentially infected infant in the first weeks of life, and (iii) major advances due to the advent of highly active anti-retroviral treatment. With effective use of these measures, transmission rates may be reduced to <2%. None of the measures though, affect a cure, and it will still be many years before the development of effective vaccines. ENT doctors may be referred children already known to be HIV-positive. Knowing how to talk to infected children (and their parents) is full of potential pitfalls, and requires careful forethought. Many infection-control policies have required considerable rethinking due to the AIDS epidemic. This has especially been the case with respect to needle-stick injuries, post-exposure prophylaxis, sterilization and re-use of equipment, and safe approaches to surgery. © 2003 British Association for Paediatric Otorhinolaryngology. Published by Elsevier Ireland Ltd. All rights reserved.
The impact of the HIV epidemic has ramifications for all of us. This short review approaches the topic in two parts: (1) an overview of HIV infection in children in the UK and abroad, and (2) the special relevance of HIV infection to ENT doctors. To begin, it is useful to look back over the progress of this infection over the last 20 or so years. It was in the early 1980s when a number of pharmacists in the USA noted increasing requests for intravenous co-trimoxazole that the first alarm bells started to ring. This was due to hitherto unanticipated numbers of cases of Pneumocystis pneumonia, particularly found in members of the homosexual community. These patients had an unusual immune deficiency, not previously recognised. By 1982, the acquired immune deficiency syndrome (AIDS) was defined for the first time and within a year of this, the human immuno-deficiency virus (HIV) was identified as its cause. By mid-1983 the presence of a heterosexual AIDS epidemic in Africa was revealed, the appalling impact of which was not to be appreciated by many for a significant number of years afterwards. By early 1985, Rock Hudson was the first of many public figures to disclose that he had AIDS, and later that year at least one case of HIV/AIDS had been reported from each region of the world. By 1987, the first therapy for AIDS (Zidovudine) was approved for use in the USA, and by 1994 the first treatment regimes to reduce mother-to-child transmission were introduced. By 1996, highly active anti-retroviral therapy (HAART) was first postulated, and by the late 1990s the very earliest vaccine trials were underway. AIDS has been defined in different ways over the years. It is important to make the distinction between HIV infection, during which many patients will be asymptomatic, versus clinical AIDS when infection is accompanied by signs and symptoms of an ‘AIDS-indicator disease’. A frequently used definition is ‘‘an illness characterised by one or more indicator diseases . . . with/without laboratory evidence of HIV infection (in the absence of another cause of immuno-deficiency)’’. The Centre
for Disease Control (CDC) has also used an extension of its definition to include all persons who are severely immuno-compromised (CD4 count of less than 200 × 106 l−1 ), irrespective of indicator disease. With special relevance to children, there are a large number of AIDS-defining conditions, some of which would indicate a presumptive diagnosis while others would lead to a definitive one. These presuppose the laboratory evidence of HIV infection. Of particular relevance to children is the development of recurrent and/or multiple bacterial infection in a child less than 13 years of age, primary cerebral lymphoma at any age, non-Hodgkins lymphoma, disseminated mycobacterial infection, and the HIV wasting syndrome/failure to thrive, which are all recognised AIDS-defining conditions. The route of transmission of the virus is well known to most clinicians. It includes sexual intercourse (whether anal or vaginal), contaminated needles (whether via intravenous drug usage, needle stick injuries or other injections), mother-to-child (vertical) transmission which may occur in utero (at birth or postnatally), and via organ and tissue donation (including blood products). For the sake of this title, it is mother-to-child transmission, and to a lesser extent needle stick injuries, that are of the most relevance. It is important to review the natural history of infection in adults and older children, which is different to that in infants. Classically, with early infection there is an increase in viral load with subsequent stabilisation. Although the viral load may continue in these stable numbers for many years, this is not insignificant infection and the number of virus particles is significantly high throughout. At the time of this initial transmission the CD4 count may drop transiently and then recover. This initial infection may be accompanied by fairly non-specific signs and symptoms, which may not come to medical carers attention at all. There is then a long, relatively asymptomatic period measured in years. This is then followed by the development of symptomatic AIDS, the hallmarks of which are an
HIV infection in children–—impact upon ENT doctors
S87
increasing viral load accompanied by immunodeficiency related to a fall in CD4 lymphocyte count. This, in a very simplistic way, is the natural history of the infection. In infants and children, however, this time scale is not seen and there is a rapid progression from birth to symptomatology in untreated patients. Thus, 15—20% of children develop AIDS-defining illnesses by the age of 12 months if they have been vertically infected. It is common for such children to present with Pneumocystis carinii pneumonia at about 3—4 months. About 70% overall will be symptomatic by the time they are 1 year old, and the median progression to clinical AIDS is approximately 6 years. Twenty-five to thirty percent of untreated children, especially in a Third World setting, may have died by this age. Over the course of this epidemic, though, it has become obvious that there are some children who represent slow versus rapid progression of the disease. The underlying reasons for this remain unclear, and certainly rapid progression seems to be a more major problem in the developing countries of the world. The global impact of this epidemic is appalling. By the end of the year 2000, an estimated 34.7 million adults and 1.4 million children were living with HIV/AIDS. It was estimated that, in that year alone, there were going to be 5.3 million new infections, representing about 16,000 per day. About 95% of all cases occur in the developing countries, and of these the major brunt of the epidemic is in sub-Saharan Africa. By the end of 1999, there were an estimated 13.2 million AIDS orphans, 2 million of which were in Southern Africa alone. The epidemic is anticipated to lead to an average expected reduction in life expectancy of approximately 17 years. The estimated impact on under-5 mortality rates has been assessed based on models with or without HIV infection and AIDS. Using Botswana as an example, there are approximately 30 deaths per thousand live births in the under-5 age group due to illnesses excluding AIDS, but this rises to greater than 150 deaths per thousand live births if AIDS is taken into account. Turning to Europe there are wide geographical variations in cases with approximately 90 per million cases in Spain and Portugal, falling to just over 10 per million in the UK and Holland. The UK HIV data base has recorded just over 51,000 reports of HIV-infected individuals from 1982 to 2002; 59% of whom have the HIV infection alone and 37% of whom have progressed to full-blown AIDS. Of the latter, 12,452 (665) have died. In the year 2000 in the UK, for the first time heterosexual cases had overtaken those in the homosexual population. About 70% of cases of AIDS and HIV infection occur and are seen in the Thames re-
gion. There is very encouraging data showing that the advent of HAART has led to increased life expectancy, but this does of course increase the pool of those who are infected and potentially infectious to others. Therefore, although great inroads have been made in terms of anti-retroviral treatment, primary and secondary preventative measures remain a formidable and continuing challenge. Comparing rates of HIV-infected individuals from the year of the first reported UK diagnosis, there have been significant changes. Comparing 1993—2001, rates occurring in the mother-to-infant group rose from 66 to 74 with a total of 861 cases overall. Rates amongst intravenous drug users had fallen from 204 to 104, but rather worryingly UK-acquired cases with no evidence of high-risk partners had risen from 63 to 165. Rates of those practising sex between men had fallen slightly from 1497 to 1415. In terms of the regional distribution of cases over this time period there was a total of 46,787 in England. Over 31,000 of these cases were in London: over the rest of the country the totals over this period range from 1624 (South West) to 2742 in the North West. Although paediatric data is limited, there are interesting trends with the introduction of HAART and active measures to prevent vertical transmission. Thus, from 1994 onwards there has been an increase in new diagnoses of HIV infection. However, over that same time scale the number of new reports of progression to AIDS has actually fallen as have the deaths related to AIDS. These figures may represent the fact that people are being diagnosed more efficiently, but also that progression and mortality is being influenced by such early diagnosis and subsequent aggressive management. This pattern is mirrored in that though the number of children born to HIV-infected mothers is rising, progression to HIV infection, AIDS and mortality is falling. Any attempt to influence progression of infection to AIDS is strengthened by detecting those mothers who are HIV-infected already. Historically in England the pick-up rate for HIV testing antenatally was poor, however, with the introduction of routine antenatal screening the total proportion of infection diagnosed has steadily risen from just over 40—70% or more. Many of these patients represent those who hitherto would have been undiagnosed and, therefore, preventative measures would not have been in place. This pattern is well demonstrated for inner London, but similar patterns has been shown for outer London, the rest of England and Wales and for Scotland. Partly linked with the timing of routine antenatal testing, children born in the UK are only recently showing a decrease in the number of cases per year in comparison to countries such as
S88 Spain, France and Italy. However, all of these European countries are showing significant decreases in cases with the advent of preventative treatment measures. In untreated mothers who are not actively managed, infection is transmitted to about 15—20% of their babies. The baseline rate is thought to double if the baby is also breastfed. In non-breastfed infants about 70% of the transmission occurs around the time of delivery. There are factors that increase the chances of transmission. These include a high viral load in the mother with a low CD4 count, clinical AIDS and prematurity. Also, the method of delivery has been felt to be important in transmission and most babies in the UK are now delivered by elective caesarean section. Alternatives to breastfeeding are also recommended in the UK, but obviously could not be applied to many parts of the world. Anti-retroviral medicines given to the mother and then to the baby, both before and after delivery, have played an important part in the reduction of the transmission rate. Implementing these three measures (not breastfeeding, anti-retroviral therapy for mother and baby and managed delivery) can mean a reduction in transmission rates to 2% or less, representing a phenomenally successful intervention. Highly active anti-retroviral treatment has become the accepted standard in most of the developed countries. There are various combinations of nucleoside and non-nucleoside reverse transcriptase inhibitors that are commonly used for children. By and large, the protease inhibitors tend to be avoided as first line therapy because of attendant side effects; therefore, most regimes are PI-sparing at the moment. There are often compliance and formulation problems with paediatric patients and there is a growing background of metabolic disturbances and lipodystrophy. Despite these problems, it is interesting to note that about 25% of the estimated 10,000 children in the United States who are HIV-infected and have progressed to clinical AIDS, have now survived to the teenage years. This then is the background of HIV/AIDS globally, and in the UK in particular. We now turn to the special relevance of this infection to doctors who specialise in oto-rhinolaryngology. In a study from Great Ormond Street Hospital of 66 paediatric patients over an 8-year period, 91% of these already-diagnosed children had ENT problems. The commonest were cervical lymphadenopathy, oesophageal candidiasis and recurrent otitis media. There were also some specific presentations, such as parotid swelling. In a study of 40 Brazilian children, ENT manifestations reached up to 100% of infected individuals and the majority had
S. Hoare their first manifestation of such ENT problems before HIV infection was diagnosed. They also noted a high occurrence of Herpes simplex gingivostomatitis and parotid hypertrophy. These two studies showed, therefore, that children who are infected with HIV suffer from common ENT illnesses, as well as more specific presentations. Again it is emphasised, the ENT condition may often be the initial presentation of HIV/AIDS. The pathogens involved include both common and uncommon organisms. Pneumococcal, Staphylococcal and Pseudomonal infections are especially important, and the latter in particular may run a very rapid clinical course. In the ill, HIV-infected child, empirical antibiotic treatment certainly needs to cover these organisms. There is often a requirement for broad-spectrum treatment and, especially in the case of Pseudomonas, dual synergistic therapy is often recommended. Other common infections include those with Haemophilus influenzae (including type B, since immunisation may not give appropriate protection in these children), the group A streptococcus, gram negative organisms, mycobacterial species, (including tuberculosis and atypical mycobacteria) anaerobes and fungi. Of the latter, Candida, Aspergillis and Rhizopus species are especially important pathogens with a high degree of morbidity and mortality. Viral infections are common, and pathogens include influenza and parainfluenza, RSV, adenovirus, CMV, EBV and HHV6, 7 and 8. The clinical course of these infections may be unusual or prolonged due to the underlying degree of immunity. In essence, the lower the CD4 count the greater the likelihood of overwhelming infection, or prolonged clinical course of infection with unusual organisms. Otitis and hearing loss are common in infected children. In one study of 10 children with symptomatic AIDS, otitis media was universal and 8 of the 10 also had persistent middle ear effusions. The aetiology of these effusions remains unclear and may be linked with humoral immunodeficiency, with Eustachian tube dysfunction, or possibly with lymphoid hyperplasia obstructing normal drainage. Six of seven children examined in this study also had conductive deafness, ranging between 10 and 65 dB losses. Sinusitis is frequent as an isolated finding. However, approximately 20% of episodes of serious bacterial sepsis may be associated with an underlying sinusitis. Aspirates most frequently show Streptococcus pneumonia, group G streptococci, Staphylococcus aureus and Staphylococcus epidermidis. Fungal infection of the sinuses is especially important. Aspergillis and Mucor infections, in particular, can cause rampaging infection with rapid extra
HIV infection in children–—impact upon ENT doctors
S89
sinus spread including that across tissue planes with extension into the central nervous system. In this scenario, the outlook is grim indeed. Allergic rhinitis is extremely common and markedly over represented in this population. It usually responds reasonably well to standard treatment and probably represents a degree of immune dysregulation. Bilateral parotid swelling is a classic feature of HIV/AIDS in children. It is seen to a much smaller degree in adults. It is associated with lymphoid interstitial pneumonitis, and is often seen in those children who are termed longer term or slow progressers. Clubbing is a frequent association. Such slow progressers may take several years to become obvious, and have CD4 counts and viral loads that alter over several years rather than months. Eventually, though, their CD4 counts decay and allow more classical clinical pictures of AIDS to appear. The parotid swelling associated with these children may be due to the presence of lymphoid tissue in the parotid gland that hypertrophies. Suppurative paratitis is much less common and usually unilateral. In terms of oral and throat disease, a wide range of problems have been associated with AIDS in the past including hairy oral leukoplakia, dental abscesses and caries, gingivitis, candidiasis and ulceration due to a variety of factors (including bacterial, herpetic and apthous ulcers). Nutritional problems are common in this group of children, remembering that failure to thrive and the HIV wasting syndrome are common presenting features. Kaposi’s sarcoma is much less common in the paediatric age range, as is Hodgkin’s disease and non-Hodgkin’s lymphoma. In terms of throat problems, bacterial and viral infection with group A streptococcus, staphylococcus, anaerobes and EBV, etc. are all common presenting clinically as tonsillitis and quinsy. In a very recent paper looking at laryngeal obstruction in 38 South African children, HIV infection was present in half of the patients admitted with this disorder to a paediatric intensive care unit. Obstruction related to lymphoid hyperplasia and orolaryngoesophageal candidiasis was over represented in the HIV-infected group. These children, however, had very similar requirements for supportive management, for the length of intubation and for ventilatory pressures used. The consensus of the authors was that even in countries with a very high HIV burden, children presenting with laryngeal obstruction do as well as their non-infected counterparts and certainly should be offered full supportive management. Another aspect of the interface between ENT doctors and children with HIV infection is the risk of HIV transmission to medical carers. In an anonymous HIV survey study of pa-
tients in Glasgow from 1992 to 1997, prevalence rates were applied to factors influencing risk. The study used the probability of percutaneous injury, the number of injuries sustained and whether post-exposure prophylaxis was taken. On the basis of three percutaneous injuries per year, specialists in renal and urological surgery who did not take any kind of post-exposure prophylaxis might expect a transmission rate of 1:2,000,000. For doctors working in ENT, though, this rate was estimated to be 1:200,000. Giving post-exposure prophylaxis to ENT doctors, though, would alter the transmission rate to 1:1,000,000. It is emphasised that ENT doctors who may be working with children who are HIV-infected need to practise excellent preventative techniques, but also be aware of their local unit’s policies and availability of medicines, guidance, advice, etc. with respect to post-exposure prophylaxis. ENT neoplasia is relatively uncommon in children who are HIV-infected. Kaposi sarcoma, lymphomas and particular B cell lymphomas are seen though. Previously, the development of neoplasia was associated with advanced disease and profound CD4-penia. It will be interesting to see if rates for neoplastic disorders alter with HAART. There is some early evidence that rates for Kaposi’s sarcoma may be influenced by aggressive anti-retroviral treatment, but the situation is less clear for other types of tumours. Finally, it is important to emphasise the psychodynamics of a consultation between parents and children who may be infected, and ENT doctors. This is unlikely for the majority of doctors in your speciality to be a frequent occurrence, but as we have discussed could happen any time, any place, anywhere. Several papers published in ENT journals have specifically addressed this issue, emphasising yet again that the otorhinolaryngologist may be the first physician to evaluate the patient with undiagnosed HIV infection. It is therefore important for ENT doctors to consider the diagnosis and to be aware of the clinical presentations. Many children will already be known to other specialities and good inter-disciplinary communication is essential. Some authors have emphasised that communication should also provide information and support for physicians who are ‘‘uncomfortable with the challenges of this epidemic’’. This infection and this epidemic are not going to go away in the near future. However, there is evidence of increasing medium- to long-term survival in the paediatric age group in those countries that can afford the preventative and treatment options outlined previously. It will be very important to see how the emergence of viral resistance and
S90 new medicines available for treatment affect this situation. It is already interesting that there are increasing numbers of women who are known to be HIV-infected who are now choosing to get pregnant because of an appreciable increase in confidence in anti-retroviral treatment. It will also be interesting if there are increasing numbers of non-infective sequelae in long-term survivors, such as neoplastic problems. For the majority of the world though, HIV remains an unpreventable, untreated and untreatable infection. In conclusion, the HIV epidemic emerged as a ‘‘bolt from the blue’’. In the latter 1970s, it would have been a brave physician to have predicted that within a decade the world would be swept by a previously unknown infectious disease, which essentially took the whole medical fraternity by surprise. More than 20 years after these events, the global pandemic of HIV remains appalling, remembering that more than 90% of AIDS occurs in the developing countries of the world. This epidemic is now leading to a situation where entire nations and governments fall under its influence, and there is an almost evolutionary change in population patterns that will have ramifications for many years. In terms of global control, there are few glimmers of hope. There is little in the way of medical advances that
S. Hoare are likely to change the picture for the foreseeable future in the developing countries, and much of the efforts towards control must still rest on prevention rather than cure or treatment. In the UK infected children are still mainly concentrated in the south east of the country, and especially in London. However, this is a situation that is dynamic and likely to alter with the redistribution of immigrant and refugee families nationally. In the UK, there is a sense of cautious optimism in terms of holding disease progression at bay with HAART, and a major sense of optimism in terms of prevention of mother-to-child transmission. At any stage, though, it is entirely feasible for HIV-infected children to present to ENT doctors before or after the diagnosis has been made. Such doctors must therefore ensure that they are able to suspect the infection clinically, to know the ENT manifestations of infection and disease, and to maintain an appropriate core of knowledge and communication relevant to HIV infection. Whilst there are many emotional and psychological upsets associated with those children with severe disease progression, there are many rewards and pleasures in dealing with the children and families who have this infection. It is up to all medical carers to put themselves in the position to recognise and manage such patients.