Prevalence and Outcome of Cytomegalovirus-associated Pneumonia ...

ORIGINAL STUDIES

Prevalence and Outcome of Cytomegalovirus-associated Pneumonia in Relation to Human Immunodeficiency Virus Infection Marco Zampoli, MB BCh, FCPaeds, Cert Paed Pulm,* Brenda Morrow, PhD,† Nei-Yaun Hsiao, MB BCh, MMed,‡ Andrew Whitelaw, MB BCh, MSc, FCPath,§ and Heather J. Zar, PhD§

Aim: To investigate the antemortem prevalence and outcome of cytomegalovirus (CMV)-associated pneumonia in African children. Methods: A total of 202 children (median age, 3.2 months; 124 human immunodeficiency virus 关HIV兴-infected, 62%; 87 severely malnourished, 43%) sequentially hospitalized for severe pneumonia were prospectively investigated. In addition to routine microbiologic investigations, respiratory tract secretions and blood were submitted for CMV culture and qualitative and quantitative CMV polymerase chain reaction. Results: CMV-associated pneumonia was common (28%, 47/169) and more prevalent in HIV-infected than uninfected children (36% vs. 15%; odds ratio 关OR兴, 3.0; 95% confidence interval, 1.3–7.4). CMV-associated pneumonia was more common than Pneumocystis pneumonia (27%) and other viralassociated pneumonia (19%) in HIV-infected children. In-hospital mortality was 25% (51/202) with increased mortality in HIV-infected compared with uninfected children (43/124 关35%兴 vs. 8/76 关11%兴; OR, 4.5; 1.9 –11.8). Increased mortality occurred in HIV-infected children with CMV-associated pneumonia (OR, 2.5; 1.04 – 6.5) but this association was not evident after adjusting for CD4 ⬍15% (adjusted OR, 1.78; 0.6 – 4.6). Conclusions: CMV-associated pneumonia is common and associated with a poor outcome in children with advanced HIV disease. Improved diagnostic testing and increased access to antiviral therapy might improve the outcome of HIV-infected children with CMV-associated pneumonia. Key Words: cytomegalovirus, pneumonia, HIV, children (Pediatr Infect Dis J 2011;30: 413– 417)

C

ytomegalovirus (CMV) and Pneumocystis jirovecii are leading causes of severe pneumonia in human immunodeficiency virus (HIV)-infected infants. Pneumocystis pneumonia (PCP) and CMV pneumonia are clinically indistinguishable; dual infection is common and associated with severe disease.1,2 Unlike PCP, little

Accepted for publication November 11, 2010. From the *Department of Paediatrics and Child Health, Red Cross War Memorial Children’s Hospital, University of Cape Town, Cape Town, South Africa; †Division of Paediatric Critical Care and Children’s Heart Disease, Red Cross War Memorial Children’s Hospital, University of Cape Town, Cape Town, South Africa; and Divisions of ‡Clinical Virology and §Clinical Microbiology, National Health Laboratory Services, University of Cape Town, Cape Town, South Africa. Supported by National Research Foundation, South Africa; and ASTRA-Zeneca Respiratory Award from the South African Thoracic Society. Address for correspondence: Marco Zampoli, MB BCh, FCPeds, Cert Paed Pulm, Department of Paediatrics and Child Health, Red Cross War Memorial Children’s Hospital, Klipfontein Rd, Rondebosch, 7700, South Africa. E-mail: [email protected]. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.pidj.com). Copyright © 2011 by Lippincott Williams & Wilkins ISSN: 0891-3668/11/3005-0413 DOI: 10.1097/INF.0b013e3182065197

is known about the epidemiology of CMV infection or CMV pneumonia in African children. Current knowledge of CMV pneumonia is derived mostly from necropsy studies in which the prevalence is reported to range from 22% to 78%.3– 6 Few studies have investigated the antemortem incidence of CMV infection in African children. Earlier studies from South Africa have reported a high prevalence (47% and 51%) of CMV infection in children with severe pneumonia but neither was able to distinguish CMV disease from infection.7,8 A small lung biopsy study from South Africa in HIV-infected children ventilated for pneumonia reported histologically confirmed CMV pneumonia in 72% of cases.9 CMV is ubiquitous, with primary CMV infection reported in 85% of African infants by 1 year of age.10 Distinguishing asymptomatic infection from CMV disease is important for several reasons. First, CMV viral replication is associated with accelerated HIV disease progression and increased risk of CMV end-organ disease in HIV-infected children.11,12 Second, CMV disease such as pneumonia can be severe, especially in immunocompromised children. Third, specific therapy is recommended for the prevention and treatment of CMV disease in immunocompromised hosts.13,14 Access to this treatment is limited in sub-Saharan Africa due to cost and availability. Definitive diagnosis of CMV pneumonia requires histologic confirmation on lung biopsy. However, this is invasive, requires specific expertise, and may not be safe in a hypoxic child. Without lung biopsy, accurate diagnosis of CMV pneumonia is difficult. Monitoring of CMV viraemia by qualitative and quantitative polymerase chain reaction (PCR) is widely used in solid organ and bone marrow transplant recipients for the diagnosis and treatment of CMV infection and disease. In this setting, a systemic CMV viral load (VL) greater than 4.0 log copies is considered indicative of active CMV infection and disease.15,16 However, there are few data on the clinical utility of these methods in the diagnosis of CMV infection or disease in African infants. Knowledge of the prevalence of CMV-associated pneumonia, and distinguishing CMV disease from asymptomatic infection is important for the development of diagnostic strategies that would enable targeted treatment of children with CMV disease. The aim of this study was to investigate the prevalence and outcome of CMV-associated pneumonia in children hospitalized with severe pneumonia.

MATERIALS AND METHODS Subjects A prospective study of children admitted for severe pneumonia was conducted at Red Cross War Memorial Children’s Hospital, a tertiary referral hospital in Cape Town, South Africa, from December 2006 to June 2008. Ethical approval was granted by the Ethics Committee of the Faculty of Health Sciences, University of Cape Town. Written informed consent was obtained from a parent or legal guardian.

The Pediatric Infectious Disease Journal • Volume 30, Number 5, May 2011

www.pidj.com |

413

Zampoli et al

The Pediatric Infectious Disease Journal • Volume 30, Number 5, May 2011

Inclusion Criteria This study was conducted within a study of children admitted with suspected PCP. Sequential children admitted with severe pneumonia defined by the following criteria were included: acute onset (⬍2 weeks) of respiratory illness requiring hospitalization and presence of tachypnea (respiratory rate ⬎60 breaths per minute (bpm) if less than 2 months, ⬎50 bpm if less than a year, ⬎40 bpm if older than a year), hypoxia (room air saturation ⬍90%), diffuse bilateral lung disease not associated with wheezing, and one of the following risk factors: HIV-infected or clinically suspected; or HIV-exposed; or malnutrition; or on immunosuppressive therapy; or the presence of immunodeficiency disease other than HIV. Children in whom consent could not be obtained were excluded. Maternal antenatal HIV status and demographic and clinical parameters were collected. Severe malnutrition was defined as weight-for-age less than 3 standard deviations of the norm. Children were prescribed first-line intravenous antibiotics (ampicillin, 50 mg/kg 6 hourly; gentamicin 5 mg/kg once daily) by the admitting clinician according to hospital protocol. In addition, cotrimoxazole (10 mg/kg loading dose followed by 5 mg/kg 6 hourly for 21 days) and adjuvant corticosteroid therapy (1 mg/kg prednisone tapered) was prescribed for suspected PCP. Intravenous ganciclovir (5 mg/kg twice daily) was initiated according to clinician’s discretion. Children diagnosed with CMV-associated pneumonia received 2 weeks of intravenous ganciclovir after which they were shifted to oral valganciclovir (15 mg/kg twice daily) to complete 6 weeks of therapy. Children diagnosed with HIV infection who were eligible for highly active antiretroviral therapy (HAART) were initiated on HAART according to national guidelines. Admission to the intensive care unit (ICU) was subject to the availability of ICU resources at the discretion of the attending specialist.

Investigations Blood Tests A full blood count, blood culture, and lactate dehydrogenase were performed at the time of admission. In children in whom HIV status was not known, an enzyme-linked immunosorbent assay (Architect HIV Ag/Ab Combo ELISA, Abbott Laboratories, Abbott Park, IL) or rapid HIV test (Determine HIV1/2, Abbott Laboratories, Abbott Park, IL) was performed in those older than 18 months, whereas for those younger whole-blood HIV deoxyribonucleic acid (DNA) PCR (Amplicor HIV-1 DNA test version 1.5, Roche Diagnostics, GmbH Mannheim, Germany) was performed. CD4 count (PanLeucogated method)17 was measured in HIV-infected children. HIV VL (NucliSens EasyQ, bioMerieux, Boxtel, Netherlands) was determined in children commencing HAART.

Microbiologic Investigations A nasopharyngeal aspirate (NPA) and either induced sputum (IS) in nonintubated patients, or nonbronchoscopic bronchoalveolar lavage (BAL) in intubated patients were collected in all children according to previously described techniques.18,19 P. jirovecii identification was performed using direct immunflourescence (Detect IF PC, Axis-Shield, United Kingdom) on NPA, IS, and BAL samples. In addition, silver methenamine stain was performed on BALs. All respiratory tract samples (NPA, IS, and BAL) were screened by rapid antigen detection for respiratory syncytial virus (RSV), adenovirus, and influenza A/B (murine FITC-conjugated anti–RSV or adenovirus monoclonal, Chemicon, Temecula, CA; Simulfluor Flu A/B, Chemicon, Temecula, CA). In addition, all respiratory samples were inoculated into “in-house”

414

| www.pidj.com

shell vial cultures (SVC). Induced sputa and BAL were submitted for detection of Mycobacterium tuberculosis by microscopy and mycobacterial culture (Becton Dickinson and Company, Sparks, MD). Postmortem lung biopsy was done when possible with parental consent. CMV Investigations All respiratory tract samples were inoculated in CMV SVC (Murine anti-CMV monoclonal, PerkinElmer, Boston, MA; and rabbit FITC-conjugated antimouse antibodies, Dako, Glostrup, Denmark). Screening for CMV viraemia was performed on all patients by qualitative whole-blood nested PCR (Super-Therm, JMR Holdings, Kent, United Kingdom). Serum CMV IgG was not measured in infants due to the high prevalence of maternal seropositivity in our population. Levels of CMV viraemia were determined by quantitative PCR (Artus RealArt LC CMV PCR, Qiagen Hilden, Germany) on all PCR positive samples. DNA extraction was done on 200 ␮L of whole blood (MagNA Pure LC DNA Isolation kit I, Roche Diagnostics, GmbH Mannheim, Germany). Nested CMV PCR targeting the immediate early gene was performed. The analytical sensitivity of the nested PCR was 50 copies/mL. Definitions The following definitions were adopted for the purposes of this study. Active CMV infection was defined as CMV detected in either blood or respiratory tract secretions by PCR or SVC, respectively. CMV-associated pneumonia was defined as a positive CMV SVC from a lower respiratory tract sample (IS or BAL) associated with detectable systemic CMV viraemia. High levels of systemic CMV viraemia were considered present if the CMV VL was ⱖ4.0 log copies/mL.

Statistical Analysis Data were analyzed using STATA (version 10.0, Statcorp, TX). Continuous variables were expressed as means ⫾ standard deviation for normally distributed variables or medians and interquartile ranges (IQR) for nonnormally distributed variables. The Mann-Whitney U test was used for comparing nonnormally distributed measures. Categorical variables were compared using ␹2 test. Odds ratios (OR) and 95% confidence intervals (CIs) were calculated where applicable. Multiple logistic regression analysis was performed including univariate variables with significant values of P ⱕ 0.05.

RESULTS Study Population A total of 202 children were enrolled. Ninety-two (45%) were males and 87 (43%) were severely malnourished. HIV status was determined in 200 children (Table 1). Most HIV-infected children (117, 93%) were less than 1 year of age and 71% (75/106) were severely immunosuppressed (CD4 ⬍25%). Of 100 HIVinfected children, 76 (76%) had a HIV VL ⱖ1,000,000 copies/mL (ⱖ6.0 log copies/mL). Four HIV-infected children were on HAART prior to hospitalization. Severe malnutrition was present in both HIV-infected and uninfected children (47% and 40%; OR, 1.3; 95% CI, 0.7–2.5). A history of infant feeding practice was obtained in 191 patients: 117 (61%) infants were breast-fed at any time.

Microbiologic Results

CMV (129/196, 66%), respiratory viruses (RSV, 27 关13%兴; adenovirus, 18 关9%兴; parainfluenza-3, 8 关4%兴), and PCP (43, 21%) were the leading pathogens identified in all samples. Cultureconfirmed tuberculosis (6, 3%) or bacteremic pneumonia (7, 3%) © 2011 Lippincott Williams & Wilkins

The Pediatric Infectious Disease Journal • Volume 30, Number 5, May 2011

TABLE 1.

CMV-associated Pneumonia

Cytomegalovirus (CMV) Infection in HIV-infected and -uninfected Children With Severe Pneumonia Variable

All Subjects* N ⫽ 200

HIV Infected N ⫽ 124

HIV Uninfected N ⫽ 76

Age in months (IQR) CMV SVC ⫹ IS CMV SVC ⫹ BAL CMV viraemia (blood PCR⫹) CMV VL ⱖ4 log copies/mL CMV-associated pneumonia PCP Respiratory virus Mortality

3.1 (2– 4.6) 34/105 (32) 16/97 (15) 124/196 (63) 70/194 (36) 47/169 (28) 43 (21) 52/191 (27) 51 (25)

3.4 (2.5– 4.4) 26 (38) 12 (27) 88 (73) 57 (47) 37 (36) 33 (27) 23 (19) 43 (35)

2.5 (1.3– 4.9) 8 (25) 4 (9) 36 (47) 13 (17) 10 (15) 10 (13) 29 (40) 8 (11)

OR (95% CI)

P

1.8 (0.6 –5.3) 3.8 (1.02–17.6) 2.6 (1.3–5.0) 4.1 (1.9 – 8.9) 3.0 (1.3–7.4) 2.4 (1.05–5.8) 0.4 (0.2– 0.7) 4.5 (1.9 –11.8)

0.003 0.209 0.024 0.002 ⬍0.001 0.004 0.025 0.002 ⬍0.001

All categorical variables expressed as n (%); continuous variables as medians and (IQR). *HIV status not determined in 2 individuals. OR indicates odds ratio; CI, confidence interval; IQR, interquartile range; ⫹, positive; SVC, shell vial culture; IS, induced sputum; BAL, bronchoalveolar lavage; PCR, polymerase chain reaction; VL, viral load; PCP, Pneumocystis pneumonia.

TABLE 2.

Clinical Features and Outcome of Children With and Without CMV-associated Pneumonia Variable

CMV-associated Pneumonia N ⫽ 47

Without CMV-associated Pneumonia N ⫽ 123

CMV VL (log copies/mL) CMV VL ⱖ4.0 (log copies/mL) Received ganciclovir NPA CMV SVC⫹ n ⫽ 164 PCP coinfection CD4 ⬍15%, n (%)* Mortality

4.5 (4.0 –5.2) 35 (74) 33 (70) 33 (72) 17 (36) 19/32 (59) 18 (38)

3.9 (2.8 – 4.6) 25 (20) 61 (50) 20 (17) 21 (17) 13/32 (41) 25 (20)

OR (95% CI)

P

11.4 (4.8 –27.4) 2.4 (1.1–5.3) 12.4 (5.2–30.1) 2.7 (1.1– 6.2) 1.93 (0.7–5.0) 2.4 (1.08 –5.4)

0.003 ⬍0.001 0.015 ⬍0.001 0.007 0.131 0.016

All categorical variables expressed as n, (%); continuous variables as medians and (IQR). *HIV-infected children. CMV indicates cytomegalovirus; OR, odds ratio; CI, confidence interval; VL, viral load; NPA, nasopharyngeal aspirate; SVC⫹, positive shell vial culture; PCP, Pneumocystis pneumonia.

were uncommon. Thirty-five (17%) children were coinfected with 2 or more pathogens. No etiological cause for pneumonia could be identified in 86 (43%) children.

Active CMV Infection A total of 129 (129/196, 66%) children had active CMV infection. Five children had CMV cultured from a NPA without associated CMV viraemia. CMV was identified in 2 of 5 postmortem examinations; both cases were associated with high CMV VLs (Table 1). Of the breast-fed children, 76% (87/114) were actively CMV-infected compared with 51% (37/72) of non-breastfed children (OR, 3.0; 95% CI, 1.5– 6.0; P ⬍ 0.001). CMV viraemia, high CMV VL, and CMV from BAL was more common in HIVinfected than uninfected children (Table 1).

CMV-associated Pneumonia CMV-associated pneumonia was diagnosed in 47/170 (28%) children. There were no differences in age, clinical presentation, and nutritional status in children with and without CMVassociated pneumonia (Table 2). Higher total white cells, neutrophils, lymphocytes, and lactate dehydrogenase values were observed in children with CMV-associated pneumonia than without CMV-associated pneumonia but the differences were small and of doubtful clinical relevance. More than half (57%) of CMV-associated pneumonia cases were coinfected with other pathogens. The most frequent coinfections were P. jirovecii (36%) or respiratory viruses (25%). CMV-associated pneumonia was diagnosed in 10 HIV-uninfected children of whom 4 were HIVexposed, 5 preterm infants, 4 RSV coinfected, 1 had a congenital T-cell deficiency, and 1 acute lymphoblastic leukemia. © 2011 Lippincott Williams & Wilkins

CMV was isolated on NPA in 33 (72%) children with CMV-associated pneumonia, compared with 20 (17%) children without CMV-associated pneumonia (OR, 12.4; 95% CI, 5.2– 30.1). NPA SVC was insensitive but fairly specific for predicting CMV-associated pneumonia (Table, Supplemental Digital Content 1, http://links.lww.com/INF/A702). In multiple logistic regression analysis, a history of ever breast-feeding (OR, 3.8; 95% CI, 1.4 –9.9), HIV infection (OR, 2.9; 95% CI, 1.1–7.7), and a positive CMV SVC on NPA (OR, 9.5; 95% CI, 3.9 –22.8) were independent predictors of CMV-associated pneumonia.

Course and Outcome A total of 109 children (54%) were admitted to the ICU, including 53 (43%) who were HIV-infected. Ganciclovir was prescribed to 110 (54%) children within a median 2 days (IQR, 0 – 4) of hospitalization; 25/37 (68%) HIV-infected and 8/10 (80%) HIV-uninfected children with CMV-associated pneumonia received ganciclovir. Antiretroviral therapy was initiated in 73 (61%) HIV-infected children at a median of 10 days (IQR, 7–14) of hospital admission. Forty-seven HIV-infected children were not initiated on HAART: 26 (55%) died before HAART could be prescribed and 21 (45%) were discharged from hospital and referred for further HIV care including HAART to appropriate services. Among children who survived hospital admission, the time of hospitalization was longer in HIVinfected children compared with uninfected children (median, 23 days vs. 15 days; P ⬍ 0.001). In-hospital mortality was 25% (51/202) with most deaths occurring in HIV-infected children (Table 1). Mortality in HIVwww.pidj.com |

415

The Pediatric Infectious Disease Journal • Volume 30, Number 5, May 2011

Zampoli et al

infected children was highest (55%) in children with CMV-associated pneumonia and high CMV viraemia (Table, Supplemental Digital Content 2, http://links.lww.com/INF/A703). HIV-infected children who died had a significantly higher median CMV VL than children who survived (5.23 log copies/mL vs. 4.29 copies/mL, P ⫽ 0.001). Mortality in HIV-infected children with CMV-associated pneumonia and high CMV viraemia treated with ganciclovir (9/21, 43%) was lower than similar untreated children (7/8, 88%). This difference was however not significant (OR, 0.1; 95% CI, 0.1–1.1; P ⫽ 0.03). Among HIV-infected children, CMV-associated pneumonia and CMV-associated pneumonia with high CMV viraemia were associated with increased mortality on univariate analysis. Increased mortality in CMV-associated pneumonia was, however, not evident after adjusting for CD4 ⬍15% (Table, Supplemental Digital Content 2, http://links.lww.com/INF/A703).

DISCUSSION This study found a high prevalence of CMV infection and CMV-associated pneumonia in young South African children hospitalized with severe pneumonia. These findings are supported by postmortem and lung biopsy studies which identify CMV as a leading cause of severe or fatal pneumonia, particularly in young infants.3– 6,9 The high prevalence of CMV-associated pneumonia in the HIV-uninfected control group may be accounted for by the study entry criteria that selected patients at risk for opportunistic infections. The high CMV infection rate in infants is consistent with other African studies. This is attributed to early vertical and horizontal acquisition of CMV infection, high CMV seroprevalence among African adults, and higher rates of vertical CMV transmission in HIV-infected women.10,14,20,21 Furthermore, HIVinfected children are more susceptible to congenital or postnatal CMV infection in their first year of life compared with HIVuninfected children.11,22,23 Distinguishing asymptomatic CMV infection from CMV pneumonia is difficult without lung biopsy and we thus opted to use the term CMV-associated pneumonia. Diagnostic criteria for HIV-associated CMV pneumonia in children are not well defined or validated. However, routine lung biopsy in children with hypoxic pneumonia is neither feasible nor safe and would thus be considered unethical in our setting which has free access to ganciclovir and antiretroviral therapy. Clinicians need to therefore rely on less invasive methods to diagnose CMV pneumonia. Internationally accepted criteria for the clinical diagnosis of CMV pneumonia for transplant recipients include signs and symptoms of pulmonary disease combined with the detection of CMV by any means in BAL fluid or lung tissue.24 Similar criteria may, however, not be suitable in the setting of HIV infection as the significance of isolating CMV in BAL fluid of HIV-infected adults with pneumonia is reported to lack clinical utility in diagnosing CMV pneumonia.25–28 However, CMV pneumonia in adults commonly results from reactivation of latent CMV infection, whereas in infants, it is more likely to follow primary CMV infection. This study investigated the novel use of quantitative CMV PCR in children with suspected CMV pneumonia as CMV viraemia is associated with increased risk of CMV disease in HIVinfected individuals, even in patients receiving HAART.29,30 Data from longitudinal cohort studies on asymptomatic HIV-infected African children report peak CMV VL values occurring soon after primary CMV infection and in advanced HIV disease.31,32 The mean peak CMV VL reported in these small studies is however less than 4.0 log copies/mL. Furthermore, a recent autopsy-based study in HIV-infected adults reported that a cut off of ⬎4.0 log copies/mL has a specificity and positive predictive value of 100%,

416

| www.pidj.com

respectively, in predicting CMV disease.33 In our study, high CMV VL was observed in 47% of all HIV-infected children and 78% of those with CMV-associated pneumonia. Quantitative PCR may therefore be a valuable tool in the diagnosis of CMV pneumonia in HIV-infected children but cut off values that accurately predict CMV disease must still be established. The high cost and limited availability of quantitative PCR are however drawbacks that precludes its routine use in resource-limited settings. Although insensitive, NPA CMV culture may be an additional tool to diagnose CMV-associated pneumonia in situations where lower respiratory tract secretions cannot be obtained. The in-hospital mortality is similar to that previously reported.4,8,34 Most deaths occurred in HIV-infected infants with advanced HIV disease. Consistent with findings in adult and pediatric studies, CMV-associated pneumonia was associated with increased mortality in HIV-infected children.1,2,35 Low CD4 count is, however, an important confounder of this association, as indicated by the lack of association with increased mortality after adjusting for CD4 ⬍15%. This indicates the susceptibility of children with advanced HIV disease to developing more severe opportunistic CMV disease. The impact of CMV-associated pneumonia on mortality is difficult to interpret in this study as almost 70% of children with CMV-associated pneumonia received ganciclovir. A trend towards increased mortality was observed in the subgroup of children with probable CMV pneumonia, who did not receive ganciclovir. Interpretation of this finding is, however, limited by small numbers and a wide spectrum of disease severity and presentation. For example, ganciclovir may not have been prescribed in time in children who presented late with severe disease and died within hours of hospital admission. Furthermore, this study was not designed to investigate the efficacy of ganciclovir in treating CMV-associated pneumonia and is inadequately powered to detect differences in outcome due to small numbers. Other factors may have adversely affected outcome and represent additional limitations of this study. Despite free access to HAART and ganciclovir, prescribing patterns in the hospital were inconsistent in HIV/CMV coinfected individuals. Second, access to ICU was not possible for all HIV-infected children, particularly those with advanced disease in whom limitations of care were applied. Third, all patients routinely received corticosteroids. Although adjuvant corticosteroids in the treatment of PCP are associated with improved outcome, controversy exists as to whether corticosteroids may adversely affect the course of CMV pneumonia.1,36 –39 In conclusion, the findings in our study suggest a role for empiric ganciclovir treatment in the management of HIV-infected children who present with severe pneumonia and advanced immunosuppression. Improved diagnostic testing and access to ganciclovir treatment might improve the outcome of HIV-infected children with CMV-associated pneumonia.

ACKNOWLEDGMENTS The authors thank Louise LeSai for her assistance with enrolling patients, and Rauf Sayed and Landon Myer for statistical support. REFERENCES 1. Williams AJ, Doung T, McNally LM, et al. Pneumocystis carinii and cytomegalovirus infection in children with vertically acquired HIV infection. AIDS. 2001;15:335–339. 2. Boonsarngsuk V, Sirilak S, Kiattboonsri S. Acute respiratory failure due to Pneumocystis pneumonia: outcome and prognostic factors. Int J Infect Dis. 2009;13:59 – 66. 3. Chintu C, Mudenda V, Lucas S, et al. Lung diseases at necropsy in African children dying from respiratory illnesses: a descriptive necropsy study. Lancet. 2002;360:985–990.

© 2011 Lippincott Williams & Wilkins

The Pediatric Infectious Disease Journal • Volume 30, Number 5, May 2011

4. Ruffini D, Madhi SA. The high burden of Pneumocystis carinii pneumonia in HIV-1-infected children hospitalized with severe pneumonia. AIDS. 2002;16:105–112. 5. Jeena PM, Coovadia HM, Chrystal V. Pneumocystis carinii and cytomegalovirus infections in severely ill, HIV-infected African infants. Ann Trop Paediatr. 1996;16:361–368. 6. Ansari N, Kombe AH, Kenyon TA, et al. Pathology and causes of death in a series of human immunodeficiency virus-positive and negative pediatric referral hospital admissions in Botswana. Pediatr Infect Dis J. 2003;22:43– 47. 7. Rabie H, de Boer A, van den Bos S, et al. Children with human immunodeficiency virus infection admitted to a paediatric intensive care unit in South Africa. J Trop Paediatr. 2007;53:270 –273. 8. McNally L, Jeena PM, Gajee K, et al. Effect of age, polymicrobial disease, and maternal HIV status on treatment response and cause of severe pneumonia in South African children: a prospective descriptive study. Lancet. 2007;369:1440 –1451. 9. Goussard P, Kling S, Gie RP, et al. CMV pneumonia in HIV-infected ventilated infants. Pediatr Pulmonol. 2010;45:650 – 655. 10. Kaye S, Miles D, Antoine P, et al. Virological and immunological correlates of mother-to-child transmission of cytomegalovirus in The Gambia. J Infect Dis. 2008;197:1307–1314. 11. Kovacs A, Schluchter M, Easley K, et al. Cytomegalovirus infection in HIV-1 disease progression in infants born to HIV-1 infected women. N Engl J Med. 1999;341:77– 84. 12. Nigro G, Krzysztofiak A, Gattinara GC, et al. Rapid progression of HIV disease in children with cytomegalovirus DNAemia. AIDS. 1996;10:1127– 1133. 13. Hodson EM, Craig JC, Strippoli GF, et al. Antiviral medications for preventing cytomegalovirus disease in solid organ transplant recipients. Cochrane Database Syst Rev. 2008;16:CD003774. 14. Mofenson LM, Oleske J, Serchuck L, et al. Treating opportunistic infections among HIV-exposed and infected children: recommendations from CDC, the National Institutes of Health, and the Infectious Diseases Society of America. Clin Infect Dis. 2005;40:S1–S84. 15. Drew WL. Laboratory diagnosis of cytomegalovirus infection and disease in immunocompromised patients. Curr Opin Infect Dis. 2007;20:408 – 411. 16. Razonable RR, Emery VC; 11th Annual Meeting of the International Herpes Management Forum. Management of CMV infection and disease in transplant patients. Herpes. 2004;11:77– 86. 17. Glencross D, Scott LE, Jani IV, et al. CD45-assisted PanLeucogating for accurate, cost-effective dual-platform CD⫹4 T-cell enumeration. Cytometry. 2002;50:67–77. 18. Zar HJ, Tannenbaum E, Apolles P, et al. Sputum induction for the diagnosis of pulmonary tuberculosis in infants and young children in an urban setting in South Africa. Arch Dis Child. 2000;82:305–308. 19. Koumbourlis AC, Kurland G. Non-bronchoscopic bronchoalveolar lavage in mechanically ventilated infants: technique, efficacy and applications. Pediatr Pulmonol. 1993;15:257–263. 20. Adjei AA, Armah HB, Gbagbo F, et al. Seroprevalence of HHV-8, CMV, and EBV among the general population in Ghana, West Africa. BMC Infect Dis. 2008;188:111. 21. Ghebrekidan H, Ruden U, Cox C, et al. Prevalence of herpes simplex virus types 1 and 2, cytomegalovirus, and varicella-zoster virus infections in Eritrea. J Clin Virol. 1999;12:53– 64.

© 2011 Lippincott Williams & Wilkins

CMV-associated Pneumonia

22. Chandwani S, Kaul A, Bebenroth D, et al. Cytomegalovirus infection in human immunodeficiency virus type1-infected children. Pediatr Infect Dis J. 1996;15: 310–314. 23. Doyle M, Atkins JT, Rivera-Matos IR. Congenital cytomegalovirus infection in infants infected with human immunodeficiency virus type 1. Pediatr Infect Dis J. 1996;15:1102–1106. 24. Llungman P, Griffiths P, Carlos P. Definitions of cytomegalovirus infection and disease in transplant recipients. Clin Infect Dis. 2002;34:1094 –1097. 25. Mann M, Shelhamer JH, Masur H, et al. Lack of clinical utility of bronchoalveolar lavage cultures for cytomegalovirus in HIV infection. Am J Respir Crit Care. 1997;155:1723–1728. 26. Tamm M, Traenkle P, Brilli B, et al. Pulmonary cytomegalovirus infection in immunocompromised patients. Chest. 2001;119:838 – 843. 27. Angelici E, Contini C, Sebastiani G, et al. Cytomegalovirus in bronchoalveolar lavage specimens from patients with AIDS: comparison with antigenaemia and viraemia. J Med Microbiol. 1996;45:149 –152. 28. Uberti-Foppa C, Lillo F, Terreni MR, et al. Cytomegalovirus pneumonia in AIDS patients: value of cytomegalovirus culture from BAL fluid and correlation with lung disease. Chest. 1998;113:919 –923. 29. Wohl DA, Zeng D, Stewart P, et al. Cytomegalovirus viraemia and end-organ disease among patients receiving potent antiretroviral therapies. J Aquir Immune Defic Syndr. 2005;38:538 –544. 30. Deayton JR, Sabin CA, Johnson MA, et al. Importance of cytomegalovirus viraemia in risk of disease and death in HIV-infected patients receiving highly active antiretroviral therapy. Lancet. 2004;363:2116 –2121. 31. Boriskin YS, Sharland M, Dalton R, et al. Viral loads in dual infection with HIV-1 and cytomegalovirus. Arch Dis Child. 1999;80:132–136. 32. Slyker JA, Lohman-Payne BL, John-Stewart GC, et al. Acute cytomegalovirus infection in Kenyan HIV-infected infants. AIDS. 2009;23:2173–2181. 33. Brantsaeter AB, Holberg-Petersen M, Jeansson S, et al. CMV quantitative PCR in the diagnosis of CMV disease in patients with HIV-infection: a retrospective autopsy based study. BMC Infect Dis. 2007;7:127.Available at: http://www.biomedcentral.com/1471–2334/7/. Accessed March 2, 2009. 34. Zar HJ, Hanslo D, Tannenbaum E, et al. Aetiology and outcome of pneumonia in human immunodeficiency virus-infected children hospitalized in South Africa. Acta Paediatr. 2001;90:119 –125. 35. Benfield TL, Helweg-Larsen J, Band D, et al. Prognostic markers of short-term mortality in AIDS-associated Pneumocystis carinii pneumonia. Chest. 2001;119:844 – 851. 36. Briel M, Bucher HC, Boscacci R, et al. Adjunctive corticosteroids for Pneumocystis jirovecii pneumonia in patients with HIV-infection. Cochrane Database Syst Rev. 2006;3:CD006150. 37. Terblanche AJ, Green RJ, Rheeder P, et al. Adjunctive corticosteroids treatment of clinical Pneumocystis jirovecii pneumonia in infants less than 18 months of age—a randomised controlled trial. S Afr Med J. 2008;98: 287–290. 38. Madhi SA, Cutland C, Ismail K, et al. Ineffectiveness of trimethoprimsulfamethoxazole prophylaxis and the importance of bacterial and viral co-infections in African children with Pneumocystis carinii pneumonia. Clin Infect Dis. 2002;35:1120 –1126. 39. Jensen AM, Lundgren JD, Benfield T, et al. Does cytomegalovirus predict poor prognosis in Pneumocystis carinii pneumonia treated with corticosteroids? A note for caution. Chest. 1995;108:411– 414.

www.pidj.com |

417