SHORT REPORT of VCT in;{verting further HIV infections4. Cotrimoxazole prophylaxis, recommended by UNAIDS as part of a minimum package of care for people living with AIDS in Africa5, should be considered for HlV-seropositive patients, and this may provide individual benefit. Antiretroviral therapy may become accessibleto the population in the future. Health workers must take a lead in this difficult area, and can begin by ..breakingthe silence". Acknowledgements We thank the DepatTment for International Deyelopment (DFID), UK, the Nomegian Agency Jbr Development Cooperqtion (NORAD) and the Royal Dutch Tuberculosis Association (KNCV) for financial support as pqrt of their aid contribution to operational Research of the Malawi National ruberculosis control programme. The study receiyed the approval of the National Health Science Research Committee. This article has previously been published in Tropical Doctor in 2002.

Authors: Menco M Weismullerl, Sander Meijnenl, Niels JM Claessensl, Felix M Salaniponi MSc PhD2, Anthony D Harries MD FRCP2 1 Academic Medical Centre, University of Amsterdam, Amsterdam,The Netherlands 2 National Tuberculosis Control Programme, Ministry of Health, pOBox303j7, Capital City, Lilongwe 3, Malawi Address for Correspondence: Professor AD Harries, c/o British High Commission, pO Box 30042, Lilongwe 3, Malawi Fax: (265) 172 657; Email: [email protected] References 1. LINAIDS. Malawi - epidemiologicalfact sheeton HIV/AIDS md sexuallytrmsmitied infections.2000 Update.UNAIDS andWorld Health Organization. 2. Kwmjma JH Haries AD GausiF Nymgulu DS SalaniponifM. TB-HIV seroprevalence il patientswith tuberculosis in Malawi.MalawiMedicalJoumal2001: ll: 7- 10. 2. crant AD Sidibe K DomouaK et al. spectrumof diseasemong HlV-infected adults hospitalised in a respiratorymedicineunit in Abidjan,Coted'Ivoire.Int J TubercLuns D i s 1 9 9 82; : 9 2 6- 9 3 4 . 4. SweatM GreogorichS Smgiwa.G et al. Cost-effectiveness of volunttryHIV_1 counsellingandtestingin reducingsexualtrmsmissionof HIV-i in Kenyaandranzmia. L a n c e2t 0 0 0 3; 5 6 :I 1 3- l 2 l . 3. UNAIDS.ProvisionalWHO,{JNAIDSSecretriatrecommendations on the useof cotrimoxazole prophylaxisin adultsandchildrenliving with HIV/AIDS in Africa. UNAIDS / WHO 2OOO.

REIEWARTICLE

The impact of HIV infection on childhood pneumonia: comparison betweendevelopedand developing regions StephenM Graham Summary Respiratory disease is the commonest cause of morbidity and mortality in IllV-infected children. While the pattern of HlV-related pneumonia in African adults is well documented and is recognised as quite different from that which occurs among HlV-infected adults in high-income regions, less is known of the situation in children. Most children are infected by mother-to-child transmission and presentation of HlV-related pneumonia is often in infancy or early childhood, an age group in which conlirmation of the cause of pneumonia is difficult. However, aetiological data are important. Poor response of the infant with severe pneumonia to standard antibiotic (such as chloramphenicol) or of the older child with chionic pneumonia to anti-tuberculosis treatment are two very common clinical dilemmas that many Malawian health workers would recognise. This review aims to present the available data relevant to Malawi, contrast with experience from the developed world and to describe common HlV-related pneumonias such as PCP and LIP. Unlike for adults, the pattern of HfV-related pneumonia in Malawian children may not be so different in cause from that described for children in developed countries prior to the use of PCP prophylaxis and anti-retroviral therapies. The most important contrast is the higher prevalence and poorer outcome. Introduction Respiratory diseaseis a major causeof morbidity and mortality in HlV-infected children in developed and developing countries.'-'Prior to the HIV epidemic, there were already important differences between the regions that still exist. In developing countries, acute childhood pneumonia is more often due to bacteria, most commonly Streptococcus pneumoniae and Haemophilus influenzae,and more likely to be fatal.aThe prevalence is much higher of important risk factors for pneumonia morbidity and mortality such as fetal and early childhood malnutrition. In addition, the community prevalenceof smear-positive pulmonary tuberculosis (PTB) has increaseddramatically in HIV endemic regions and maternal illness and death is cofitmon. It is therefore a difficult environment for an HlV-infected child to negotiate and perhapsnot surprising that the majority of HIVinfected infants from a resource-poor region such as tropical Africa have died by 3 years of age.t6Only about 25Vosurvive up to 5 years compared to over 80Voin USA or western Europe.l Thus the majority of cases of HlV-related respiratory disease presentin infancy and early childhood. However, becausechildhood HIV infection is so common in some countries, the presentation of HlV-infected school-agedchildren often with chronic respiratory diseaseis not unusual. Causes of pneumonia in HlV-infected children living in highincome countries such as USA or UK are well documented.3'?'8 It may not be correct to assumethat the pattern of diseaseis similar in HlV-infected children in resource-poorAfrica where it is estimated that over 90Voof childhood HIV infection now Malawi Medical Jomal

\

REVIEW ARTICLE ., occurs.e'tvlajordifferences between the regions have been rec- antibiotic-resistantorganismsis more common.2oChildren with ognized in HlV-infected adults such as the relative incidence of advanced HIV diseaseare at greater risk of recurrent infection pulmonary tuberculosis (PTB) and Pneumocystis carinii pneu- and death.H.influenzae is also common and the studiesof South monia (PCP). Etiology of childhood respiratory diseaseis diffiAfrican children have found that StaphylococcusaLffeusis parcult to confirm in developing countries because diagnostic ticularly common in HlV-infected children with chronic lung options are very limited yet such data arc critical for effective disease.lllsSalmonella typhimurium was the second "o-rnon"J prevention and managementstrategies.Children with pneumo- blood isolate after pneumococcusfrom Malawian children with nia are mainly managedaccording to standardguidelineso,HIV acute severe pneumonia.l6 Other bacteria identified in HIVstatus is rarely known and therapeutic options such are limited infected African children with pneumonia include Klebsiella to a few cheapantibiotics and oxygen delivered via nasopharyn- pneumoniae, Escherichia coli and Pseudomonas aerugigeal catheter. nosa.t4't7't8 These Gram-negative bacteria are associated with malnutrition, as they were prior to the HIV epidemic. Clinical The pattern of respiratory disease in HlV-infected children managementalgorithms guide diagnosis and choice of antibiotin developed countries ic for most children with pneumonia.o'2'Althoughthe range of It is more relevant to compare the pattern of respiratory disease available antibiotics is limited, it would seem that these recomas it was in developedcountries prior to the availability of high- mendations are still appropriatefor HlV-infected children. ly active anti-retroviral therapy and the routine use of cotrimoxazole prophylaxis in HlV-exposed infantsTs,as this is PNEUMOCYSTIS CARINII PNEAMONIA still the scenariofor most HlV-infected African infants. The inci- Autopsy'3-" and clinicallu2' studieshave shown that PCP is comdence of bacterial pneumonia was far higher than in HIV-uninmon in HlV-infected African infants between 2 and 6 months of fected children and recurent bacterial diseaseis an AIDS-indiage and accountsfor one-third or more of the deathsin this age cator disease.TThe common bacterial pathogensare S.pneumo- group. Mixed infection with bacterial pneumonia or niae and H.influenzae. The commonest opportunistic infections cytomegalovirus (CMV) is common. PCP is usually the frst were PCP and lymphoid interstitial pneumonitis (LIP).&10Most presentationof HlV-related diseaseand is often clinically recogcasesof PCP were in infants, although it did occur among older rizable. PCP presents with very severe pneumonia, hypoxia is children, and PCP was associatedwith a poor survival.10 PCP is common and there is no responseto standardparenteral antibinow uncommon due to the routine use of cotrimoxazole pro- otics. Intravenous cotrimoxazole is usually not available so phylaxis in HfV-exposed infants.2 It may still present as first treatment is high-dose oral cotrimoxazole, steroids and continuevidence of HIV infection in infants born to immigrant mothers ing parenteral antibiotic and oxygen. Demand for oxygen therafrom HlV-endemic regions who have not received proper ante- py is very high and yet outcome is very poor.'6'"This can create natal care.3Cytomegalovirus (CMV) is also a cause of HIVa dilemma when oxygen supply is limited as more curable cases related pneumonia in infants, often in associationwith PCP.LIP such as very severebacterial pneumonia may be denied oxygen. occurs in an older age group and is associatedwith a "favorable" prognosis.'oLIP is often complicated by bacterial pneumonia VIRALPNEUMONIA and bronchiectasis."'" Tuberculosis is HlV-related but uncom- The proportion of pneumonia in hospitalized HlV-infected mon.t'8 South African children due to viruses was significantly lower than in HlV-uninfected children becausepneumonia due to other The pattern of respiratory disease in HlV-infected African causessuch as bacteria and PCP was more common.'3 However, children the estimated burden of disease of viral pneumonia in HIVCohort studies indicate that acute and chronic pneumonia are infected children was increasedfor RSY influenza virus, paraincommon causesof morbidity in HlVinfectedAfrican children.t6 fluenza and adenovirus, and mortality was significantly higher. However, these studies did not investigate for specific etiology. As for developed countries, respiratory syncytial virus (RSV) is Autopsy studieshave consistently shown that respiratory disease the commonest cause of viral pneumonia in developing counis a very corrrmon cause of death'3-lsbut tend to over-represent tries. In tropical regions it does not have a similar strong seacausesof pneumonia that do not respond to standardavailable sonal variation and this may be even less evident in HlV-infecttreatments or that present with end-stageHIV disease.Clinical ed children. The RSV pneumonia in HlV-infected African chilstudiesof infants and children hospitalized with severepneumo- dren is associatedwith less wheeze or typical features of bronnia in Malawi and South Africa found rates of HIV infection of chiolitis, is more commonly complicated by secondarybacterial 45-657o.t6'8 Case-fatality rate was increasedthree to six-fold in pneumonia and mortality is higher.'3Autopsy and clinical studthese children and this was mainly due to the poor outcome of ies have commonly identified CMV in HlV-infected infants with CMV pneumonia PCP. It is from South Africa that most etiologic data are emerg- pneumonia, often in associationwith PCP.I5'18 ing. Childhood HIV infection is now very common there and is difficult to recognize or confirm and therapy such as gancidiagnostic facilities more sophisticated than in other sub- clovir is not available. Measles is also more common in HIVSaharan counffies. However, similar studies still need to be infected children and more likelv to be fatal'zobut immunisation undertaken in tropical Africa where the pattern of diseasedoes is usually protective. differ, such as the importance of Salmonella as an invasive pathogen of children in malaria endemic regions.le LYMPHOID INTERSTITIAL PNEUMONITIS LIP is a common cause of chronic respiratory diseasein HIVBACTERIAL PNEUMONIA infected African children.''"'' Diagnosis is rarely confirmed as Streptococcuspneumoniae is consistently the cornmonestcause this requires lung biopsy but when this was done in HlV-infectof bacterial pneumonia and, similar to experience in the USA, ed South African children with chronic respiratory symptoms, the incidence is far higher than in HlV-uninfected children.l618 LIP was the commonest diagnosis.'sAs LIP was not a clinical The clinical presentationand treatment responsefor most cases problem prior to the HIV epidemic, many health workers have are similar to that in HlV-uninfected children but infection with little knowledge of the condition and LIP is often misdiagnosed Malawi Medical Joumal

21

REVIEWARTICLE as miliary or PTB.' Children with LIP usually present after 2 years of age and associatedclinical featuresinclude generalised lymphadenopathy,bilateral non-tenderparotid enlargement,digital clubbing and marked hepatomegaly.Secondarybacterial disease due to pneumococcus or Salmonella is common. Corticosteroids are useful in alleviating symptoms but prophylaxis or even treatment for PTB also needsto be considered. PULMONARY TUBERCULOSIS The association of TB and HIV in African children is still not entirely clear and is certainly not as strong as it is in adults.' The difficulty in defining the epidemiology in the conrext of HIV infection relates to difficulties in diagnosis in young children and to the overlap in clinical and epidemiological presentation. As mentioned, children with other forms of HlV-related lung diseasesuch as LIP are often registered and treated as smear-negativePTB.26Notwithstanding, there is now good evidence that the incidence of PTB is increased in HlV-infected African children, partly becauseof immune suppressionincreasing susceptibility to TB diseasebut also becauseHlV-infected children are at greater risk of TB infection due to high rates of PTB among their parents.rT'18'26.27 The presentationof PTB is similar to that in HlV-uninfected children but the tuberculin test is less useful and treatment responseis poorer especially in those with advancedHIV disease.'u'"Aside from LIP, other diagnoses to consider in the HlV-infected child with chronic respiratory disease who is not responding to anti-tuberculous therapy are bronchiectasisand pulmonary Kaposi's sarcoma.

Thble 1: Causesof HlV-related lung diseasein children AGEGROUP

MOSTCOMMOM

LESSCOMMON

INFANTS

Bacterial Pneumonia

ViralPneumonia(eg;CMV)

PCP

Tuberculosis

Bacterial Pneumonia

Viral pneumonia (eg; measles)

LIP

Pulmonary Kaposi's Sarcoma

Tuberculosis

Nocardiosis

CHILDREN

Candidiasis

Table 2: HlV-related clinical issuesthat affect incidence, aetiology and outcome of acute pneumonia Opportunistic pathogense.g.. Pneumocystis carinii pneumonia Uncommon pathogens e.g., Klebsiella pneumoniae, Salmonella Mixed infections Nosocomial infection Antibiotic resistance

Summary More etiological data are needed from resource-poor regions. However, unlike for adults, available evidencedoes suggestthat the pattern of HlV-related pneumonia in children from subSaharanAfrica is not markedly different from that described in HlV-infected children from developed countries in the earlier stagesof the epidemic. Important differences are the far greater size of the problem, the increasedfrequency of disease,the lack of diagnostic and therapeutic options and the high early mortality. These differencesiargely reflect the samerisk factors and the same problems of pneumonia managementihat existed prior to the HIV epidemic. In communities that are highly HIV endemic, the burden of HlV-related childhood pneumonia is enormous and affects the ability of already under-resourcedand over-burdened health servicesto care for all sick children, HlV-infected or not. There is great potential to reduce this burden by reducing rates of vertical HIV transmission and by interventions such as PCP prophylaxis and conjugate vaccinesagainstcommon bacteria. In practice, such measuresare still a long way off. Dr SM Graham, Reseach Felloq Wellcome Trust Researcb Laboratories, PO Box 30096, Blantyre. Fax:765-675774 Email: [email protected]

Underlying chronic lung disease Advanced immunosuppression Malnutrition Matemal illness or death

Table 3: Common clinical features of PCP in Malawian children Age: 2-6 months Afebrile or 1ow-gradefever Marked respiratorydistress Auscultation: Clear chest or diffuse rather than focal signs CXR: Hyperinflation or diffuse interstitial pattem Poor responseto standard antibiotic teatment for sevete pneumonia Severe and persistent hypoxia First presentation of HlV-related disease HIV seropositive

Malawi Medical Jomal

!li-

REVIEWARTICLE References 1. Graham SM, Coulter JBS, Gilks CF. Pulmonary diseasein HlV-infected Africm children. Int J Tuberc Lung Dis 2001l-5: 12-23. 2. Abrams EJ. Opportunistic infections and other clinical mmifestations of HIV disease in children. Pediatr Clin N Am 2000: 47: 79-108 3. Williams AJ, Duong T, McNally LM et al. Pneumocystis carinii pneumonia and cytomegalovirus infection in children with vertically acquired HIV infection. AIDS 2001;l5: 335-9 4. RasmussenZ, Pio A, Enarson P Case management of childhood pneumonia in developing countries: recent relevant reserch and cunent initiatives. Int J Tuberc Lung Dis 2000: 4:807-26 5. Spira R, Lepage P, Msellati P et al. Natural history of human immunodeficiency virus type 1 infection in children: a five-year prospective study in Rwanda. Pediatrics 1999; 104: e56 6. Taha TE, Graham SM, Kumwenda NI et al. Morbidity among HlV-infected and uninfected African children. Pediatrics 2000; i06: e77. 7. Dankner WM, Lindsey JC, Levin MJ and the Pediatric AIDS Clinical Trials Group. Correlates of opportunistic infections in children infected with the human

Table4: Clinical featuresof LIP Usualll older than 2 years of age \ot severeh'rvasted .\ssociatedtearures: Marked generalisedlymphadenopathy Hepatosplenomegaly Parotid enlargement Finger clubbing CXR: diffuse reticular pattern and bilateral adenopathy +/- local opacities or consolidation Differential diagnosis: miliary TB, pulmonary TB, bronchiectasis

Thble 5: Impact of HIV infection on.the usefulness of features used to diagnose PTB in children Diagnostic feature

Impact of HIV

Chronic symptoms

less specific

Smear-positive contact (if paren|

less specific

Malnutrition or failure to thdve

less specific

Positive tuberculin test

less sensitive

"Characteristic" CXR abnormalities

less specific

Satisfactory responseto TB fteatment

less sensitive

Table 6: Clinical differentiation of miliary TB from LIP in children Miliary TB Clinical

-:-. ..rrs spllatory syrn-

Persi"wasiing

'rever

-l+

+++

++

++

+++

Generalisedlymphadenopathy

+++

Parotid enlargement

++

Clubbing

+

Hepatomegaly

CXR Features:-

Diffuse micronodrrlar Diffuse reticular

++

Lymphadenopathy

++

Malawi Medical Jomal

imunodeficiency virus mmaged before highly active mtiretroviral therapy. Pediatr Infect Dis J 2001120: 40-8 8. Vm Dyke RB. Opportunistic infections in HlV-infected children. Seminr Pediatr Infect Dis 1995; 6: 10-6 9 . UNAIDS. AIDS epidemic update: December 2000. Geneva: LINAIDS,^VHO, 2000. 1 0 . Scott GB, Hutto C, Makuch RW et al. Survival in children with perinatally acquired immunodeficiency virus type I infection. N Engl J Med 1989 321: 1791-6. 1 1 . Shrland M, Gibb DM, Hollmd F Respiratory morbidity from lymphocytic interstitial pneumonitis (LIP) in vertically acquired HIV infection. Arch Dis Child 1997:.76: 334-6. 12. SheikhS,MadirajuK,SteinerP,RaoM.BronchiectasisinpediatricAIDS.Chestl99T; 112: 1702-7. 13. LucasSB.PeacockCS.HounnouAetal.DiseaseinchildreninfectedwithHlVin Abidjan, Cote d'Ivoire. BMJ 1996: 312: 335-8. 14. Ikeogu MO, Wolf B, Mathe S. Pulmonary manifestations in HIV seropositivity and malnutrition in Zimbabwe. Arch Dis Child 1997: 76: 1.24-8. 15. Jeena PM, Coovadia HM, Chrystal V. Pneumocystis crinii and cytomegalovirus infections in severely ill, HlV-infecred Africm infants. Ann Trop Paediatr 1997; 16:361-8. 16. Graham SM, Mtitimila EI, Kamanga HS, Walsh AL, Hart CA, Molyneux ME. The clinical presentation and outcome of Pneumocystis carinii pneumonia in Malawian children. Lancet 2000: 355: 369-73 17. Madhi SA, Petersen K, Madhi A, Khoosal M, Klugman KP. Increased disease burden md antibiotic resistanceof bacteria causing severe community-acquired lower respiratory tract infections in human immunodeficiency type l-infected children. Clin Infect Dis 2000; 31: 170-6 18. Zar H, Hanslo D, Tannebaum E et al. Aetiology and outcome of pneumonia in humm imunodeficiency virus-infected children hospitalized in South Africa. Acta Paediatr 2001;90:119-25 19. Graham SM, Molyneux EM, Walsh AL, Cheesbrough JS, Molyneux ME, Hart CA. Non-typhoidal Salmonella infections of children in tropical Africa. Pediatr Infect Dis J 2000; 19: ll89-96. 20. Madhi SA, PetersenK, Madhi A, WasasA, Klugman KP. Impact of human immunod eficiency virus type 1 on the disease spectrum of Streptococcuspneumoniae in South African children. Pediatr Infect Dis J 2000:, 19: 1141.-7. 21. Shann F. The management ofpneumonia in children in developing countries. Clin Infect Dis 1995; 21 (suppl 3): 5218-25. s K, Hussey G. Pneumocystis cainii pneumonia in immunodeficiency virus. Pediatr Infect Dis J 2000; 19: 603-7. K, Blackbum N, Klugman B, Simmk burden of 23. Madhi SA, Schoub respiratory viral associatedsevere lower respiratory tract inf'ections in children -with wirus type-l. I Pediatr 2000; 137: 78-84. humm ipruodeficienel' lrc, Quinn TC, Makowitz LE, Batter V Heyward WL. Measles m hospitalised African children with human immunodeficiency virus. Am J Dis Child 1988:165: 1271-2. 25. Jeena PM, Coovadia HM, Thula SA, Blythe D, Buckels NJ, Chetty R. persistent and chronic lung diseasein HIV-I infected and uninfected African chiidren. AIDS 1998: l2: I 185-93 26. Kiwanuka J, Graham SM, Coulter JBS et al. Diagnosis ofpulmonary tuberculosis il children in an HIV endemic uea in Malawi. Ann Trop paediah 2001;21: 5-14. 27. Mukadi YD, Wiktor SZ, Coulibaly I-M et al. Impact of HIV infection on the development, clinical presentation and outcome of tuberculosis among children in Abidjan, Cote d'Ivoire. AIDS 1997 11: l15l-8.

The impact of HIV infection on childhood pneumonia: comparison between developed and developing regions.

Respiratory disease is the commonest cause of morbidity and mortality in HIV-infected children. While the pattern of HIV-related pneumonia in African ...
6MB Sizes 0 Downloads 6 Views