Clinical Radiology
Volume 65, Issue 2 , Pages 150-154, February 2010

Chest radiographic features of lymphocytic interstitial pneumonitis in HIV-infected children

  • R.D. Pitcher

      Affiliations

    • Division of Paediatric Radiology, Red Cross War Memorial Children's Hospital, Department of Radiation Medicine, University of Cape Town, South Africa
    • Corresponding Author InformationGuarantor and correspondent: Tel.: +27 215316123; fax: +27 215313164.
    • ,
  • S.J. Beningfield

      Affiliations

    • Division of Radiology, New Groote Schuur Hospital and University of Cape Town, South Africa
    • ,
  • H.J. Zar

      Affiliations

    • Department of Paediatric Pulmonology, Red Cross War Memorial Children's Hospital, School of Child and Adolescent Health, University of Cape Town, South Africa

Received 12 May 2009; received in revised form 12 September 2009; accepted 5 October 2009.

Article Outline

Aim

To review the radiological features of biopsy-proven lymphocytic interstitial pneumonitis (LIP) in human immunodeficiency virus (HIV)-infected children and establish whether these are based on systematic radiological analysis, and to investigate whether more specific radiological diagnostic criteria can be developed.

Materials and methods

A Medline search of English-language articles on the radiological features of biopsy-proven LIP in HIV-infected children was conducted for the period 1982 to 2007 inclusive. Radiological findings were compared with the Centers for Disease Control and Prevention (CDC) criteria for a presumptive diagnosis of LIP.1

Results

Pulmonary pathology was recorded as “diffuse” and “bilateral” in 125 (97.6%) of 128 reported cases of LIP. Twenty-five different terms were used to describe the pulmonary parenchyma. In 96 (75%), the terminology was consistent with CDC diagnostic criteria. Radiological evolution was documented in 43 (33.5%). Persistent focal opacification superimposed on diffuse pulmonary nodularity was demonstrated in 10 (7.8%). The method of radiological evaluation was described in six (4.6%). In no instance was the terminology defined.

Conclusion

The radiological features of LIP have not been systematically analysed. However, CDC criteria remain reliable, allowing diagnosis of at least 75% of cases. The sensitivity of these criteria may be increased by including cases with persistent focal pulmonary opacification superimposed on diffuse nodularity. Longitudinal studies utilizing standardized radiographic analysis are needed to elucidate the natural history of LIP.

 

Back to Article Outline

Introduction 

Lymphocytic interstitial pneumonitis (LIP) is estimated to occur in 30–40% of human immunodeficiency virus (HIV)-infected children.2, 3, 4, 5 LIP was included as an AIDS-defining illness by the United States Centers for Disease Control and Prevention (CDC) in 1985,6 but was reclassified in 1994 as a moderately symptomatic indicator (Category B),7 in recognition of the fact that children with LIP have improved survival compared with other indicator diseases.5

LIP generally has an insidious clinical presentation in HIV-infected children older than 2 years of age. Typical manifestations are a non-productive cough and mild hypoxaemia together with generalized lymphadenopathy and finger clubbing. LIP may also be associated with painless bilateral parotid gland enlargement. Although LIP is not confined to HIV-infected children, it is uncommon in HIV-infected adults and rare in HIV-uninfected individuals. Definitive diagnosis of LIP requires lung biopsy. Diffuse lymphocytic infiltration of the pulmonary interstitium occurs, including involvement of the alveolar septa and bronchiolar submucosa.8, 9 However, appreciation of the characteristic clinical and radiological features of LIP in HIV-infected children resulted in the CDC formulating criteria for a presumptive diagnosis of LIP in 1987,1 obviating the need for lung biopsy. Radiological criteria are persistence of diffuse, symmetrical, reticulo-nodular, or nodular pulmonary opacification (Fig. 1), with or without mediastinal adenopathy, for at least 2 months, with neither an identifiable pathogen nor a response to antibiotic therapy. These criteria, which have remained unchanged, are principally aimed at differentiating LIP from infections, including miliary or bronchogenic tuberculosis, viral infections, such as varicella or cytomegalovirus (CMV), and fungal infections.

The recognition of LIP as a common pulmonary complication of paediatric HIV infection coincided with two seminal radiological publications. The first suggested a new, semi-quantitative reporting tool for systematic radiographic analysis of diffuse pulmonary pathology,10 utilizing analytical principles previously applied to industrial lung disease. Although developed for use in epidemiological studies of adult pulmonary pathology, the principles outlined can be helpfully applied to diffuse pulmonary disease at all ages. The second article proposed standard terminology for chest radiographic reporting.11 Used in conjunction, the two approaches outlined useful principles for systematic analysis of radiographic change in diffuse lung disease. They afforded both a precise description of radiographic features and a scoring system for severity, providing tools for both epidemiological as well as clinical research. The diffuse radiographic change characteristic of LIP 9 is ideally suited to such a systematic analysis.

Two standardized reporting systems have thus far been invoked for LIP.12, 13 Nodule dimension formed the basis of the classification used by Oldham et al.12: type 1, a fine reticular or reticulo-nodular pulmonary pattern, with nodules less than 3mm in diameter; type 2, a coarse reticulo-nodular pattern with nodules 3–5mm in diameter; type 3, either type 1 or 2, associated with areas of confluent pulmonary opacification. Norton et al.13 utilized the reporting format of the Prospective Study of Paediatric Pulmonary and Cardiovascular Complications of Vertically Transmitted Human Immunodeficiency Virus Infection (P2C2 HIV study).14 In this, forced-choice responses evaluated lung volumes, nodular and reticular pulmonary opacities, consolidation, cysts, pleural effusions, pneumothoraces, hilar adenopathy, heart size, and bony changes.

The aims of the present study were to review the reported radiological features of LIP in HIV-infected children and establish whether these were informed by systematic radiological analysis utilizing standardized terminology, and to assess whether current knowledge of the radiological features of LIP merits modification of the criteria for its presumptive diagnosis.

Back to Article Outline

Materials and methods 

A Medline search of articles in English was conducted for the period 1982 to 2007 inclusive, using the terms “LIP in HIV-infected children” or “LIP in children with AIDS/AIDS related complex (ARC)” and “radiographic features” or “pathological features” or “diagnosis.” Inclusion criteria were children with biopsy-proven LIP and documented chest radiographic findings. Reference lists were examined sequentially and hand searches conducted, with extraction of radiographic and histological data. Articles recording radiological features of biopsy-proven LIP in HIV-infected children 13 years or younger, or children of the same age with AIDS/ARC as defined by the CDC1, 6, 15 were included. Radiographic findings were stratified according to the type of study, year of publication, country of origin, authorship, patient demographics, radiological methods, and radiological terminology used, and then summarized and compared with the radiological criteria for a presumptive diagnosis of LIP in children, as defined by the CDC.1

Back to Article Outline

Results 

Twenty-five publications documented the radiological features of 128 biopsy-proven cases of LIP in children ranging from 7–108 months of age. Most were from North America (22 articles, 103 cases),2, 5, 12, 13, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33 with single publications from Western Europe (four cases),34 the United Kingdom (five cases)35 and Southern Africa (16 cases).36 Seventeen publications appeared in the first decade of the HIV pandemic (1982–1991) and eight in the second; the last two were in 2001. The largest series involved 16 patients36; five articles reported single cases. Fifteen reports (83 cases) had a radiologist as a co-author, while six studies (24 cases) had radiological features as the primary focus.

In 125 cases (97.6%), chest pathology was recorded as “diffuse” and “bilateral.” Twenty-five different descriptors were used to further characterize the pulmonary parenchymal changes. In 96 (75%), these were recorded as either “nodular” or “reticulo-nodular,” and thus, consistent with the CDC diagnosis of LIP (Table 1). In 22 (17.2%), the descriptor used was non-specific; examples included “interstitial infiltrates,” “interstitial pattern,” “interstitial densities,” and “increased interstitial markings.” Had the authors been more precise in their descriptions of this group, it is possible that some cases would have been more accurately characterized as “nodular” or “reticulo-nodular.”

Table 1. Summary of radiological descriptors (n=128)
DescriptorNumber%
Interstitial infiltrates2217.2
Nodular pattern6147.7
Reticulo-nodular pattern3527.3
Reticular pattern21.6
Ring shadows53.9
Focal32.3

In 10 (7.8%) focal areas of confluent pulmonary opacification were superimposed on a diffuse parenchymal nodular pattern, accounting for some asymmetry in the radiological appearance.2, 12, 25, 31, 36 In three (2.3%), pulmonary features were focal, with pulmonary “cysts”,32 “consolidation,”13 and “consolidation with effusion”36 recorded in one case each.

Eight reports involving 25 patients (19.5%) recorded mediastinal abnormalities attributable to adenopathy.2, 19, 23, 25, 27, 28, 29, 30 There was a single case of adenopathy causing significant narrowing of the main bronchi.27

Nodule size was recorded in two reports, involving 11 patients (8.5%).12, 27 Bradford et al.27 found that nodules were most easily seen in the bases and periphery of the lung, were first discerned when approximately 1mm in diameter and were generally not larger than 3mm in diameter.

Evolution of radiological features was recorded in seven reports involving 43 patients (33.5%) that were followed for 18–84 months.2, 5, 12, 27, 28, 31, 33 These reports showed that the condition may be radiographically occult in the earliest stages of lymphocytic infiltration of the interstitium.2 With progressive infiltration, radiographic changes evolve sequentially as (1) bronchial wall thickening;2 (2) a fine reticular pattern;36 (3) a reticulo-nodular pattern, with nodule dimensions increasing from 1–5mm in diameter;12, 27, 33 and (4) coalescence of nodules to form areas of confluent opacification.2, 12, 36 Bradford et al.27 documented a single case where an increase in nodule size was discernable on radiographs taken 1 month apart. In all other cases, radiographic change was documented over more than 2 months. Where interstitial infiltration was sufficient to compromise adjacent bronchi, bronchiectasis could evolve distally.31 Four patients (3.1%) developed bronchiectasis in follow-up periods of 36 to 84 months. Mediastinal lymphadenopathy was an inconstant feature, which could cause proximal bronchial narrowing.2, 27

With the exception of bronchiectasis, all features were potentially reversible, either in response to therapy or in the natural course of the disease.2 Rubinstein et al.28 showed that 10 of 15 patients (60%) on steroid therapy demonstrated a decrease in the size and number of nodules over 16 months. Lynch et al. 5 showed radiographic resolution in 13 of 20 patients (65%) on anti-retroviral therapy within 1–4 years, and that this was not due to decreasing immune function.

A systematic method of radiological evaluation was described in only two publications involving a total of six cases (4.6%).12, 13 One used the classification developed by Oldham et al.,12 while the second used the standardized reporting method of the P2C2 study as previously described.14 In no instance was radiological terminology defined.

Back to Article Outline

Discussion 

Globally, approximately 2.1 million children under the age of 15 years are infected with HIV; 90% in sub-Saharan Africa.37 Given the estimated 30–40% incidence of LIP in HIV-infected children,2, 3, 4, 5 there may be as many as half a million children with LIP in sub-Saharan Africa alone. Consistent with this, LIP has been shown to be the commonest cause of chronic lung disease in HIV-infected South African children.36 LIP is thus an important component of the burden of HIV disease in African children.

The radiographic features of a relatively small number of biopsy-proven cases of LIP in HIV-infected children have been recorded over the past 25 years. A broad spectrum of terms has been used to describe the radiological findings, but these terms have not been consistently applied; the terms used have not been clearly defined and systematic reporting has been used in only six children.12, 13

Notwithstanding these limitations, studies have enhanced understanding of the radiology of LIP and elucidated the spectrum of features. The presentation chest radiographic pattern has been described as diffuse, bilateral, and symmetrical in approximately 90% of published cases and the parenchymal pattern has been found to be “nodular” or “reticulo-nodular” in 75% of cases. Pulmonary changes have been diffuse but asymmetrical in approximately 8% and focal in less than 2% of cases. Consolidation, mediastinal adenopathy, and bronchiectasis are recognized associations. The natural history of the condition remains poorly understood. Rather than the stable process originally conceived, it is now known that the radiological features of LIP may be influenced by anti-retroviral therapy or immune function and evolve over time.

In the present study, 75% of all biopsy-proven cases of LIP would have been correctly diagnosed by invoking the CDC criteria for a presumptive diagnosis. This figure could have been higher if standardized terminology had been used to report the diffuse bilateral interstitial change seen in the 17.2% of cases where descriptors were considered non-specific. Therefore, the CDC radiological criteria for the diagnosis of LIP remain reliable. The sensitivity of these criteria may be increased by inclusion of cases with an area of persistent confluent opacification against a background of diffuse bilateral symmetrical pulmonary nodularity. In the present study there were 10 such cases. Inclusion of these would thus have increased the sensitivity for the diagnosis by 7.8%. However, prospective studies using systematic reporting and standardized terminology in children with a presumptive diagnosis of LIP are needed before changing the CDC criteria.5

For the purposes of recording the diffuse radiographic change in LIP, recommended descriptors include: “nodule” or “rounded opacity;” “line” or “linear opacity;” “nodular pattern,” “reticular pattern,” or “reticulo-nodular pattern”.11 By contrast, the term “interstitial” is discouraged on the grounds that it is an anatomical term; and radiographic descriptors should not imply anatomical location or the nature of disease.10 The use of the term “density” is not recommended because of potential confusion with the more specific “radiographic density,” a fundamental characteristic of the radiograph.11 While descriptors such as “infiltrate” and “shadow” have qualified support in a broader clinical setting, their use in the context of chronic lung disease is not recommended if more precise terminology can be invoked.11

In addition to standard and defined terminology, systematic reporting requires that each radiographic descriptor, such as “nodule” or “line”, be further characterized by its size, margin, location, and profusion; findings are ideally recorded as forced-choice options, thus affording semi-quantitative analysis.10 Such a reporting tool for general used in chronic paediatric chest pathology has been developed.14 A modified reporting tool is proposed for specific use in prospective studies of LIP. This combines the reporting methods previously used by Oldham et al.12 and Cleveland et al.,14 with some analytical features invoked by McLoud et al.10 and terminology recommended by the Fleischner Society.11

A limitation of this study is its inclusion of only English language articles. The study was further limited to HIV-infected children only, due to the unique epidemiological features of LIP in this patient population; only 10 paediatric cases of LIP were reported prior to the beginning of the HIV pandemic in 1982. In addition to the published studies included in the present study, a single, small, prospective study of 12 HIV-infected children with a presumptive diagnosis of LIP (not confirmed on biopsy) has been reported. This study used the principles of systematic reporting and documented disease progression.38 However, for reliability of diagnosis, only biopsy-proven cases of LIP were included in the present study.

In conclusion, the 1987 CDC criteria for a presumptive diagnosis of LIP remain valid, as 75–92% of all biopsy-proven cases of LIP in HIV-infected children would have been diagnosed using these. Longitudinal studies of children with a presumptive radiological diagnosis of LIP are required, using systematic radiographic analysis and standardized terminology, to better elucidate the natural history of the condition including the impact of anti-retroviral therapy.

Back to Article Outline

References 

  1. Centers for Disease Control . Classification system for human immunodeficiency virus (HIV) infection in children under 13 years of age. MMWR. 1987;36:225–236
  2. Marquis JR, Berman CZ, DiCarlo F, et al. Radiographic patterns of PLH/LIP in HIV positive children. Pediatr Radiol. 1993;23:328–330
  3. Marks MJ, Haney PJ, McDermott MP, et al. Thoracic disease in children with AIDS. RadioGraphics. 1996;16:1349–1362
  4. Khare MD, Sharland M. Pulmonary manifestations of paediatric HIV infection. Indian J Pediatr. 1999;66:895–904
  5. Lynch JL, Blickman JG, terMeulen DC, et al. Radiographic resolution of lymphocytic interstitial pneumonitis (LIP) in children with human immunodeficiency virus (HIV): not a sign of clinical deterioration. Pediatr Radiol. 2001;31:299–303
  6. Centers for Disease Control . Revision of the case definition of acquired immunodeficiency syndrome for national reporting — United States. MMWR. 1985;34:373–375
  7. Centers for Disease Control . CDC 1994 revised classification system for HIV-infection in children less than 13 years of age. MMWR. 1994;43:1–10
  8. Carrington CB, Liebow AA. Lymphocytic interstitial pneumonia. Am J Pathol. 1966;48:36
  9. Liebow AA, Carrington CB. Diffuse pulmonary lymphoreticular infiltrations associated with dysproteinaemia. Med Clin North Am. 1973;57:809–843
  10. McLoud TC, Carrington CB, Gaensler EA. Diffuse infiltrative lung disease: a new scheme for description. Radiology. 1983;149:353–363
  11. Tuddenham WJ. Glossary of terms for thoracic radiology: recommendations of the Nomenclature Committee of the Fleischner Society. AJR Am J Roentgenol. 1984;143:509–517
  12. Oldham SAA, Castillo M, Jacobson FL, et al. HIV-associated lymphocytic interstitial pneumonia: radiologic manifestations and pathologic correlation. Radiology. 1989;170:83–87
  13. Norton KI, Kattan M, Rao JS, et al. P(2)C(2) HIV Study Group. Chronic radiographic lung changes in children with vertically transmitted HIV-1 infection. AJR Am J Roentgenol. 2001;17:1553–1558
  14. Cleveland RH, Schluchter M, Wood BP, et al. Chest radiographic data acquisition and quality assurance in multicenter studies. Pediatr Radiol. 1997;27:880–887
  15. Centers for Disease Control . Update: acquired immunodeficiency syndrome (AIDS) — United States. MMWR. 1984;32:688–691
  16. Oleske J, Minnefor A, Cooper R, et al. Immune deficiency syndrome in children. JAMA. 1983;249:2345–2349
  17. Rubinstein A, Sicklick M, Gupta A, et al. Acquired immunodeficiency with reversed T4/T8 ratios in infants born to promiscuous and drug-addicted mothers. JAMA. 1983;249:2350–2356
  18. Scott GB, Buck BE, Leterman JG, et al. Acquired immunodeficiency syndrome in infants. N Engl J Med. 1984;310:76–81
  19. Andiman WA, Martin K, Rubinstein A, et al. Opportunistic lymphoproliferations associated with Epstein–Barr viral DNA in infants and children with AIDS. Lancet. 1985;2:1390–1393
  20. Fackler JC, Nagel JE, Adler WH, et al. Epstein–Barr virus infection in a child with acquired immunodeficiency syndrome. Am J Dis Child. 1985;139:1000–1004
  21. Goldman HS, Ziprkowski MN, Charytan M, et al. Lymphocytic interstitial pneumonitis in children with AIDS: a perfect radiographic–pathologic correlation. AJR Am J Roentgenol. 1985;145:868
  22. Pahwa S, Kaplan M, Fikrig S, et al. Spectrum of human T-cell lymphotropic virus type lll infection in children. JAMA. 1986;255:2299–2305
  23. Rubinstein A, Morecki R, Silverman B, et al. Pulmonary disease in children with acquired immune deficiency syndrome and AIDS-related complex. J Paediatr. 1986;108:498–503
  24. Amodio JB, Abramson S, Berdon WE, et al. Pediatric aids. Semin Roengenol. 1987;122:66–76
  25. Zimmerman BL, Haller JO, Price AP, et al. Children with AIDS — is pathologic diagnosis possible based on chest radiographs?. Pediatr Radiol. 1987;17:303–307
  26. Schiff RG, Kabat L, Kamani N. Gallium scanning in lymphoid interstitial pneumonitis of children with AIDS. J Nucl Med. 1987;28:1915–1919
  27. Bradford BF, Abdenour GE, Frank JL, et al. Usual and unusual radiologic manifestations of acquired immunodeficiency syndrome (AIDS) and human immunodeficiency virus (HIV) infection in children. Radiol Clin North Am. 1988;26:341–353
  28. Rubenstein A, Bernstein LJ, Charyton M, et al. Corticosteroid treatment for pulmonary lymphoid hyperplasia in children with the acquired immune deficiency syndrome. Pediatr Pulmonol. 1988;4:13–17
  29. Zuckier LS, Ongseng F, Goldfarb CR. Lymphocytic interstitial pneumonitis: a cause of pulmonary gallium-67 uptake in a child with acquired immunodeficiency syndrome. J Nucl Med. 1988;29:707–711
  30. Haney PJ, Yale-Loehr AJ, Nussbaum AR, et al. Imaging of infants and children with AIDS. AJR Am J Roentgenol. 1989;152:1033–1041
  31. Amorosa JK, Miller RW, Laraya-Cuasay L, et al. Bronchiectasis in children with lymphocytic interstitial pneumonia and acquired immune deficiency syndrome. Pediatr Radiol. 1992;22:603–607
  32. Schroeder SA, Beneck D, Dozar AJ. Spontaneous pneumothrax in children with AIDS. Chest. 1995;108:1173–1176
  33. Izraeli S, Mueller BU, Ling A, et al. Role of tissue diagnosis in pulmonary involvement in paediatric human immunodeficiency virus infection. Pediatr Infect Dis J. 1996;15:112–116
  34. Boccon-Gibod L, Sacre JP, Just J, et al. Lymphoid interstitial pneumonia in children with AIDS or AIDS-related complex (ARC). Pediatr Pathol. 1986;5:238–239
  35. Sharland M, Gibb M, Holland F. Respiratory morbidity from lymphocytic interstitial pneumonitis (LIP) in vertically acquired HIV infection. Arch Dis Child. 1997;76:334–336
  36. Jeena PM, Coovadia HM, Thula SA, et al. Persistent and chronic lung disease in HIV-1 infected and uninfected African children. AIDS. 1998;12:1185–1193
  37. UNAIDS . AIDS epidemic update. http://www.unaids.org2007;[accessed June 2008].
  38. Prosper M, Omene JA, Ledlie S, et al. Clinical significance of resolution of chest x-ray findings in HIV-infected children with lymphocytic interstitial pneumonitis. Pediatr Radiol. 1995;25:S243–S246

PII: S0009-9260(09)00361-4

doi:10.1016/j.crad.2009.10.004

Clinical Radiology
Volume 65, Issue 2 , Pages 150-154, February 2010

Article Tools

* Image Usage & Resolution