Author + information
- Received June 17, 1998
- Revision received April 9, 1999
- Accepted May 16, 1999
- Published online September 1, 1999.
- Neil E Bowles, PhD∗,‡,
- Debra L Kearney, MD‡,§,
- Jiyuan Ni, MD∗,‡,
- Antonio R Perez-Atayde, MD∥,
- Mark W Kline, MD†,‡,
- J.Timothy Bricker, MD, FACC∗,‡,
- Nancy A Ayres, MD, FACC∗,‡,
- Steven E Lipshultz, MD¶,2,
- William T Shearer, MD, PhD†,‡ and
- Jeffrey A Towbin, MD, FACC∗,‡,#,1,2,* ()
- ↵*Reprint requests and correspondence: Dr. Jeffrey A. Towbin, Department of Pediatrics (Cardiology), Room 333E, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030
The aim of this study was to investigate the frequency of viral nucleic acid detection in the myocardium of human immunodeficiency virus (HIV)-infected children to determine whether an association exists with the development of heart disease.
As improved medical interventions increase the life expectancy of HIV-infected patients, increased incidences of myocarditis and dilated cardiomyopathy (DCM) are becoming more apparent, even in patients without clinical symptoms.
Myocardial samples were obtained from the postmortem hearts of 32 HIV-infected children and from 32 age-matched controls consisting of patients with structural congenital heart disease and no myocardial inflammation and no cardiac or systemic viral infection. The hearts were examined histologically and analyzed for the presence of viral sequences by polymerase chain reaction (PCR) or reverse transcription-PCR.
Myocarditis was detected histologically in 11 of the 32 HIV-infected patients, and borderline myocarditis was diagnosed in another 13 cases. Infiltrates were confined to the epicardium in two additional hearts. Virus sequences were detected by PCR in 11 of these 26 cases (42.3%); adenovirus in 6, CMV in 3 and both adenovirus and CMV in 2. Two cases without infiltrates were also positive for adenovirus: one had congestive heart failure (CHF) and the other adenoviral pneumonia. No other viruses were detected by PCR, including HIV proviral DNA. All control samples were negative for all viruses tested.
These data suggest that the presence of viral nucleic acid in the myocardium is common in HIV-infected children, and may relate to the development of myocarditis, DCM or CHF and may contribute to the rapid progression of HIV disease.
One of the consequences of the development of improved therapies for the treatment of human immunodeficiency virus (HIV) infection and the acquired immunodeficiency syndrome (AIDS) and the associated longer survival of infected patients has been the emergence of diseases not directly associated with opportunistic infections. These have included cardiac symptoms including those characteristic of myocarditis or dilated cardiomyopathy (DCM) (1–6). The prevalence of myocarditis in HIV-infected patients has been reported to range from nearly nonexistent (7,8)to 100% (1,4)of patients. This wide variation is probably due to a range of factors, including interobserver variability in making the histologic diagnosis of myocarditis, sampling error and the paucity of cellular infiltrates in this immunocompromised population (7). However, it is clear that cardiac abnormalities play a significant role in the morbidity and mortality associated with an HIV-infected state. In one study, 25% of children who died had evidence of cardiomyopathy or succumbed to sudden death; of these, 83% had premorbid cardiomyopathy or arrhythmias (9).
A number of etiologic agents have been proposed to be responsible for the initiation of the pathologic processes leading to the development of myocarditis and DCM in HIV-infected patients. These have included infection of myocytes with HIV (10,11)or other agents, including fungi, parasites or other viruses capable of initiating these diseases (6,12,13), cardiotoxicity resulting from drugs, such as heroin (14), or pharmacologic agents commonly used by AIDS patients, such as 3′-azido-3′ deoxythymidine (AZT, zidovudine) (15,16).
In the general population, myocarditis and its sequela, DCM, are believed to be initiated by a number of different agents including viruses, parasitic organisms and drugs; viral myocarditis is believed to be most common. Over the past decade various studies have established a role for persistent enterovirus infection of the myocardium in the pathogenesis of myocarditis and DCM (17–19)by using molecular hybridization or polymerase chain reaction (PCR) methods. We have reported the prevalence of viral sequences, detected by PCR and reverse-transcription PCR (RT-PCR), within myocardial samples from pediatric patients with myocarditis or DCM but without known HIV infection (20–22). Over 300 samples have been studied; adenovirus was detected in 19%, enterovirus in 13%, herpes simplex virus (HSV) or cytomegalovirus (CMV) in ∼1% and parvovirus or Epstein-Barr virus (EBV) in less than 1%.
Investigations of the role of virus infection of the myocardium in HIV-infected patients who develop myocarditis or DCM have primarily focused on HIV and CMV as responsible agents. Herskowitz et al. (13)detected sequences, by in situ hybridization, within the myocytes of 5 of 33 samples from patients with congestive heart failure (CHF) but in none of eight samples from HIV-infected individuals without evidence of heart disease. However, in another study no HIV sequences were detected in myocardial biopsy samples (23). Herskowitz and colleagues (13)also reported that 16 of the 33 samples were positive for the CMV immediate early-2 (CMV IE-2) gene transcripts, including all 5 that were positive for HIV-specific sequences. In isolated pediatric patients, HIV infection of the myocardium has been reported (10,24), but the role of other viruses has not been investigated.
In the study reported here, we describe the use of PCR and RT-PCR to investigate the frequency of detection of virus nucleic acid sequences in the myocardium among a cohort of HIV-infected pediatric patients, some having clinical evidence of cardiomyopathy or CHF at the time of death.
Thirty-two HIV-infected children, aged 3 months to 12 years, formed the patient cohort for this study. All but one child had perinatally acquired infection. The eldest patient became infected following blood transfusions he received when diagnosed with acute lymphocytic leukemia at five years old. He had not received the cardiotoxic drug adriamycin as part of his chemotherapeutic regimen. Clinical records and echocardiograms, performed in 26 of the 32 patients (81%), were reviewed. Upon death, all patients had complete postmortem examination. Fourteen children were part of a National Heart, Lung and Blood Institute (NHLBI)-sponsored multicenter study known as the Pediatric Pulmonary and Cardiac Complications of Vertically Transmitted HIV infection study (P2C2HIV study). Autopsies on these patients were performed between 1991 and 1997 at the participating institutions, which included Texas Children’s Hospital, Boston Children’s Hospital and the Children’s Hospital of Los Angeles. As part of the protocol for postmortem examination, the myocardium was cultured for bacteria, fungi, acid-fast bacilli and viruses in 10 patients. The remaining 18 HIV-infected patients were autopsied at Texas Children’s Hospital from 1985 to 1997; myocardial cultures were performed on 10 of these children as well.
Samples of formalin-fixed right ventricular myocardium were collected during postmortem examination for molecular viral genome analysis using PCR. In addition, a minimum of four sections, two from each ventricle, were processed by routine histologic techniques for light microscopic examination. Hematoxylin and eosin stained slides were examined with specific attention to the presence of inflammatory cell infiltrates and organisms. The diagnosis of myocarditis was established according to the “Dallas criteria,” which requires the presence of both lymphocytic infiltrates and myocyte degeneration/necrosis (25). Infiltrates without myocyte degeneration or necrosis were termed “borderline myocarditis.” Because these children were generally small for their age, reflecting poor somatic growth, all recorded heart weights were normalized to body length rather than patient age.
Right ventricular myocardial samples from 32 age-matched, non HIV-infected patients with structural congenital heart disease, no histologic evidence of inflammatory cell infiltrates and no evidence of cardiac or systemic viral infection served as control tissues for PCR analysis.
Myocardial samples were placed in viral transport media, homogenized, and the supernatant was used to inoculate susceptible cell lines capable of supporting the replication of enteroviruses, adenovirus, CMV, HSV, varicella zoster, influenza, parainfluenza, respiratory syncytial virus (RSV), measles or mumps viruses. The tubes were incubated at 37°C, and maintenance media appropriate to each cell line was changed weekly. Cultures were maintained for 21 days and assessed 5 days each week for viral cytopathic effects.
RNA and DNA template preparation
Tissue samples were first homogenized in RNAzol using disposable RNase-free pestles (PGC Scientific, Gaithersburg, Maryland). Total RNA and genomic/viral DNA were isolated as previously described (20–22).
Adenovirus Type 5, coxsackievirus B4, CMV strain AD169, HSV Type 1, EBV, influenza A, parvovirus B19, RSV, parainfluenza virus and an HIV gag-polsubclone were used as positive viral controls after nucleic acid extraction, for the PCR analysis (20–22,26).
Reverse transcription and pcr
Primer pairs (Gibco-BRL, Rockville, Maryland) designed to amplify viral genomic sequences and the beta-actin gene (27)are shown in Table 1. For the detection of genomic nucleic acid of the RNA viruses complementary DNA (cDNA) was synthesized from 3 μl of extracted total nucleic acid in the presence of 20 units of an RNase inhibitor, RNasin (Promega). This mixture was heated to 95°C for 5 min, then snap-cooled on ice. After the addition of 5 pmol of the reverse transcription primer, 4 μl of 5 mmol/liter each dNTPs, 4 μl of 5× RT buffer (Gibco-BRL), 200 units of Moloney murine leukemia virus (MoMLV) reverse transcriptase (Gibco-BRL) and diethyl pyrocarbonate (DEPC)-treated dH2O to 20 μl, the samples were incubated at 37°C for 1 h. Next, 2 μl of this first-strand cDNA or 500 ng of control human genomic DNA was combined with 25 pmol of each primer pair, 5 μl of 10× PCR buffer (Gibco-BRL), 2.5 μl of 50 mmol/liter magnesium chloride and 5 μl of 2 mmol/liter dNTPs, in a final volume of 50 μl. TaqDNA polymerase, 2.5 units (Gibco-BRL), was added after an initial 5-min incubation at 95°C and then 35 rounds of amplification were performed using either a Stratagene Robocycler or an MJ Research Thermocycler. In the case of enterovirus amplifications the conditions were as follows: 95°C for 30 s, 57°C for 30 s and 72°C for 1 min, whereas for parainfluenza virus, influenza A virus and RSV the conditions were 94°C for 45 s, 60°C for 45 s and 72°C for 45 s.
For the beta-actin control, or for adenovirus, CMV, EBV, HSV, or parvovirus (i.e., DNA viruses) or HIV proviral DNA, 2 μl of extracted total DNA was combined with 1 μl of each of the appropriate primers (10 μmol/liter), 2 μl of 10× PCR buffer, 1 μl of 50 mmol/liter magnesium chloride and 2 μl of 2 mmol/liter dNTPs in final volume of 20 μl. TaqDNA polymerase (2.5 units) was added after an initial 5-min incubation at 95°C. For each virus, 35 rounds of amplification were performed using a Stratagene Robocycler, under the following conditions: 94°C for 45 s, 64°C for 45 s and 72°C for 45 s (for EBV and CMV the second annealing step was performed at 68°C and for parvovirus, 45°C).
Analysis of PCR products
Polymerase chain reaction products were analyzed by agarose gel electrophoresis containing 0.5 μg/ml of ethidium bromide (Sigma) and visualized by ultraviolet transillumination. In all cases, positive (purified viral nucleic acid) and negative (water or nucleic acid from a tissue sample known to be negative) control reactions were performed simultaneously with the test samples. All samples were analyzed without prior knowledge of clinical, histologic or culture data for each patient; all samples were required to give positive reactions with the beta-actin primers; and all viral PCR-positive samples were required to have duplicate results.
The adenovirus-positive samples were reamplified and the PCR products purified using a purification kit (Qiagen, Valencia, California) according to the manufacturer’s instructions, and resuspended in 30 μl of TE (10 mM Tris, 1 mM EDTA, pH 8). The DNA sequence was determined using an Applied Biosystems automatic DNA sequencer using primer 928 as the sequencing primer. Sequence comparison was performed by BLAST search of GenBank databases. Although the primer binding sites are highly conserved between serotypes, the region between them is divergent.
At the time of death, 31 of 32 HIV-infected patients were severely symptomatic with illnesses that fulfilled the definition for AIDS, clinical category “C,” according to the 1994 Center for Disease Control (CDC) revised classification for HIV infection in children (28). One moderately symptomatic child was category “B.” Twenty-six of the 29 patients for whom CD4 lymphocyte counts were available were severely immunosuppressed, and placed in immunologic category “3.” Accordingly, 26 patients were CDC class “C3” reflecting advanced disease with severe immunosuppression. Fifteen (47%) of the 32 patients had clinical manifestations of heart disease present during the terminal hospitalization, including 3 with chronic cardiomyopathy, 4 with nonspecific chronic cardiac disease that had previously required inotropes on more than one occasion and 6 with acute CHF. Three patients had cor pulmonale (Table 2).
One or more echocardiograms were performed in 26 patients; the time interval between the last study and death ranged from two days to 18 months (mean 4.03 months, median 1.5 months). The findings from the last echocardiogram are presented in Table 2. Six of 26 patients (23%) with echocardiographic studies had decreased left ventricular function as assessed by percent shortening fraction. Among 23 patients with available data, the left ventricle was dilated in 3 patients, 2 of whom had decreased function. In three additional patients with poor left ventricular function (Patients 5–7), including two with chronic cardiomyopathy, parameters of left ventricular size were not available from the final echocardiogram; however, the left ventricle had been dilated in previous studies. Four patients had mild concentric left ventricular hypertrophy, including one (Patient 16) with congenital aortic stenosis, and two (Patients 1 and 20) with right ventricular dysfunction and cor pulmonale. Left ventricular mass was specifically assessed in the echocardiograms of 13 patients, including the final study in 9 patients. Left ventricular mass was increased in the most recent echocardiogram with available data from 5 patients (Patients 1, 2, 5, 13 and 14). Seventeen patients (53%) had no clinical cardiac symptoms and 12 (46%) had a normal final echocardiogram.
Viruses were cultured from six (30%) of the 20 patients who had myocardial cultures performed. Cytomegalovirus was isolated from four patients; one each grew adenovirus and picornavirus (Table 2).
Cardiomegaly, defined as an increase in heart weight ≥2 SDs above that expected for patient height (Z score ≥2), was seen in 18 (56%) of the 32 patients. Histologic review of the myocardial sections revealed interstitial lymphocytic infiltrates in 24 patients (75%); most infiltrates were focal and very mild (Table 2). Eleven patients (34%) had active lymphocytic myocarditis. The infiltrates were borderline for myocarditis in the remaining 13 patients. Two additional patients had mild lymphocytic infiltrates confined to the epicardium. Organisms were identified histologically in the myocardium of six patients; three were CMV infected, one had Histoplasma capsulatum, one Cryptococcus neoformansand one Acanthamoeba.
Results of the PCR analysis are summarized in Table 2. In 13 (41%) of the 32 samples from HIV-infected patients, one or more virus types were detected. The most commonly detected virus was adenovirus (10 of 32, 31%; Fig. 1), and CMV was detected in five samples (16%), including two that were simultaneously positive for adenovirus. No enterovirus, HSV, EBV, parvovirus, parainfluenza, influenza A virus or RSV genomic sequences were detected. In addition, no HIV proviral DNA was detected in any sample. No viruses were detected in any of the control myocardial samples.
The DNA sequence analysis of the adenoviral amplification products followed by comparison with published sequences (as well as the type 5 positive control) indicated that adenovirus type 5 was detected in each case. It should be noted, however, that the DNA sequences of all adenovirus serotypes have not been reported.
Detection of viruses in the myocardium
With the advent of improved therapies for the treatment of opportunistic infections suffered by AIDS patients and the development of drugs to enhance the immune system or reduce HIV load, the survival of HIV-infected patients has improved (29). One consequence of this has been the development of diseases not directly associated with opportunistic infections such as dementia and cardiomyopathy (1–6). Previous studies have suggested that HIV and CMV may be involved in the development of heart disease in these patients (11,13). In non-HIV-infected patients, however, the adenoviruses (20–22), enteroviruses (17–19), and CMV (30)are the important viruses relevant in the development of myocarditis and DCM.
In the present study, myocardial samples from 32 pediatric patients with advanced HIV disease as well as 32 non-HIV-infected, age-matched control samples from children with structural congenital heart disease and without inflammatory heart disease, were analyzed by PCR or RT-PCR for the presence of viral genomic sequences (Table 2). In 13 (41%) of the 32 samples from HIV-infected children, we detected one or more virus types. The most common virus identified was adenovirus (10 of 32, 31%), followed by CMV (5 of 32, 16%). In addition, two patients (2 of 32, 6%) had CMV-inclusion bodies in the myocardium but were PCR negative (patients 8 and 29). Because these inclusion bodies stain positive for CMV antigens, it is likely that these bodies comprise either empty or incompletely assembled virus particles. No other viral sequences were detected in any sample, including HIV proviral DNA. It currently cannot be unequivocally stated that the detection of viral genomic sequences indicates infection of myocytes or infiltrating inflammatory cells, although it is likely.
The frequency of virus detection described here is similar to that seen in non-HIV-infected pediatric myocarditis or DCM patients. Furthermore, although an unidentified picornavirus was cultured from the heart of one patient with disseminated infection, no enteroviral sequences were detected by PCR, suggesting that this virus was from another picornaviral genus. This differs from that typically seen in non-HIV-infected myocarditis/DCM patients, where enterovirus is detected almost as frequently as is adenovirus. The failure to detect viruses in a proportion of the patients with heart disease could reflect several factors, including sampling error, the presence of another virus that was not investigated, an earlier-triggering viral infection that has been cleared or simply that the heart disease was unrelated to viral infection.
The DNA sequence analysis of the adenoviruses amplified from the HIV-infected patient samples only demonstrated adenovirus type 5. This is in contrast to our data from non-HIV-infected children with myocarditis or DCM, where type 2 adenovirus has predominantly been detected (unpublished data). This difference may reflect a different spectrum of adenoviral susceptibility between HIV-infected and noninfected children or a difference in viral pathogenesis in immunocompromised children. However, it does seem that the group C adenoviruses are most commonly identified in myocardial samples.
We analyzed myocardial samples for the presence of HIV proviral DNA sequences rather than for viral RNA because we believed that this would provide a more accurate relationship between virus detection and pathogenicity. The demonstration of HIV-specific RNA sequences by RT-PCR could simply result from free virus contamination of the sample or from a nonproductive infection of myocytes, as suggested by Herskowitz et al. (13). The presence of proviral sequences would have strongly suggested that myocyte infection had occurred. However, the failure to detect HIV DNA sequences in these samples suggests that HIV rarely infects myocytes resulting in viral DNA integration. Such infrequent infection of myocytes suggests that HIV does not play a direct role in the development of myocarditis or DCM in HIV-infected children, but does not exclude indirect effects.
It has been reported that CMV immediate early region (IE)-2 specific RNA could be detected in 11 of 17 (64%) cases of myocarditis and 5 of 16 (31%) cases of DCM without an inflammatory infiltrate in HIV-infected adult patients (13). In our pediatric patient group, the frequency of CMV detection was lower (16% by PCR, 22% by PCR, culture or histology), which could indicate a different spectrum of viral exposure between adults and children. Alternatively, the PCR primers used in the present study were designed to amplify a different region of the viral genome (the pp150 protein encoding region) and it is possible that this region is more prone to sequence variation. However, these data suggest that CMV infection of the myocardium is an important etiologic agent of acquired heart disease in HIV-infected patients.
Myocardial pathology associated with viral infection
Histologic evidence of active myocarditis was observed in 11 (34%) of the 32 HIV-infected patient myocardial samples, whereas infiltrates borderline for myocarditis were observed in another 13 cases. Adenovirus was detected in 4 of the 11 samples with myocarditis, in 3 samples with borderline infiltrates, in 1 patient with infiltrates confined to the epicardium and in 2 patients with no histologic evidence of inflammation. In the two patients with adenovirus but no inflammation, one was reported to have died from CHF and the other from adenoviral pneumonia. Of the 10 patients positive for adenovirus, 4 (40%) had clinical cardiac symptoms, 3 (Patients 2, 6 and 25) had CHF, 1 (Patient 6) had evidence of chronic heart dysfunction and 1 had terminal hypotension with radiographic evidence of cardiomegaly. Adenovirus was detected in three of the six patients (50%) with CHF; only one had myocardial infiltrates and these were confined to the epicardium. Among the three patients with dilated cardiomyopathy, one was positive for adenovirus. Seven of the 18 patients (39%) with postmortem cardiomegaly (heart weight Z score ≥2: Table 2) were PCR-positive for adenovirus.
Conversely, 70% of patients with PCR evidence of adenovirus had postmortem cardiomegaly. Two patients (Patients 2 and 25; Table 2) were reported to have adenoviral pneumonia at the time of death; both cases were positive for adenoviral DNA by PCR including one with disseminated infection and positive myocardial culture. It is quite possible that the adenovirus detected by PCR in this patient with disseminated infection was present only in the blood. Future studies using in situ hybridization and immunochemical staining will confirm that virus detected in this patient, as well as all of the other samples studied here, is present in myocardial cells.
It is interesting to note that 6 of 10 adenovirus-positive patients had other organisms identified in the heart. All six had myocardial inflammation; however, only one had clinical cardiac symptoms. This contrasts sharply with the findings in four patients in whom adenovirus was the sole myocardial isolate; all four were symptomatic and only two of four had myocardial infiltrates. The frequency of postmortem cardiomegaly was similar in both groups of patients. These clinical and pathologic features in patients with PCR evidence of adenovirus support a pathogenic role for this virus in the development of heart disease in HIV-infected pediatric patients.
Cytomegalovirus was detected in three myocarditis samples (2 by PCR) and in four samples (3 by PCR) with borderline lymphocytic infiltrates. Extracardiac, systemic infection with the virus was detected by culture or by histology in six of the seven patients. Two patients (28%) had clinical cardiac symptoms. One patient who was PCR positive for CMV (Patient 24) had terminal acute CHF and myocardial infiltrates borderline for myocarditis. The most recent echocardiogram performed 10 days before death was normal and there was no postmortem evidence of systemic infection. In the other patient (Patient 29), CMV inclusions were only identified in the heart histologically and were associated with borderline myocarditis and disseminated systemic CMV infection. Clinically, the heart was enlarged by chest X-ray and the patient was hypotensive. The last echocardiogram performed 12 months before death was normal.
Another patient who was CMV positive by myocardial culture and PCR was clinically asymptomatic but had myocarditis and mildly decreased left ventricular function assessed one week before death. Among the 18 cases with postmortem cardiomegaly, only two (11%) had evidence of CMV in the myocardium (Patients 28, 29); in one (Patient 28) adenovirus was also present.
In the control group, the medical regimens included anticongestive therapy or beta-adrenergic blocking agent therapy. The HIV-infected group was treated with either no additional therapy or AZT monotherapy in most cases. A small number of children received ddl or d4T added to the AZT monotherapy. No differences were seen regarding cardiac function between these therapeutic groups.
The relatively mild inflammatory infiltrates in most of the virus-positive samples could be the result of a number of different possibilities, including the fact that these HIV-infected patients were immunocompromised, precluding a significant cellular immune response against infected cells. Indeed, in 26 of 29 patients with CD4 lymphocyte counts available to permit CDC classification, class C3 reflected severe immunosuppression. In addition, we have observed in non-HIV-infected myocarditis patients that the level of inflammatory infiltration is less in adenovirus-infected samples than in, for example, enterovirus-infected patients (21).
In this cohort of 32 HIV-infected children, clinical cardiac dysfunction was seen in 15 patients (47%) and contributed to death in 9 patients (28%). Four of these nine were PCR positive for adenovirus; two others had evidence of CMV. These findings suggest that viral-mediated cardiac disease is likely to gain greater clinical significance as preventive measures and therapeutic interventions decrease the mortality of other opportunistic infections. We acknowledge the possibility that systemic disease and medical therapy might have contributed to the histopathologic abnormalities observed.
Finally, the frequency of adenoviral sequences within the heart (31%) and the apparent disproportionate association of this virus with cardiomegaly (70%) suggest that the role of adenovirus infection of the myocardium and the development of heart disease in the morbidity and mortality of HIV-infected children should be further investigated.
The Clinical Centers and Principal Investigators of the P2C2HIV Infection Study are Baylor College of Medicine/Texas Children’s Hospital, Houston, Texas, William T. Shearer; Children’s Hospital, Boston/Harvard Medical School, Boston, Massachusetts, Steven Lipshultz; Mount Sinai School of Medicine, New York, New York, Meyer Kattan; Presbyterian Hospital in the City of New York/Columbia University, New York, New York, Robert B. Mellins; and UCLA School of Medicine, Los Angeles, California, Samuel Kaplan. The Clinical Coordinating Center is the Cleveland Clinic Foundation, Cleveland, Ohio, Mark D. Schluchter, Principal Investigator. The work was performed in part in the Phoebe Willingham Muzzy Pediatric Molecular Cardiology Laboratory at Texas Children’s Hospital, Baylor College of Medicine, Houston, Texas. We thank the Child Health Research Center DNA Sequencing facility at Baylor College of Medicine for the DNA sequence analysis.
↵2 Drs. Lipshultz and Towbin were supported by an NIH grant, the Pediatric Cardiomyopathy Registry.
↵1 Dr. Towbin was supported by the Texas Children’s Hospital Foundation Chair in Pediatric Cardiac Research.
☆ This study was supported, in part, by P2C2HIV Infection Study Contracts N01-HR-96037, 96038, 96039, 96040, 96041, 96042 and 96043 from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland, Hannah H. Peavy, Project Officer; by NIH General Clinical Research Center Grants RR-00188, RR-02172, RR-0533, RR-00071, RR-00645, RR-0865, and RR-00043; and by NIH Grants AI 39131 and AI 36211 and the Immunology Research Fund of Texas Children’s Hospital.
- acquired immunodeficiency syndrome
- Center for Disease Control
- congestive heart failure
- dilated cardiomyopathy
- Epstein-Barr virus
- human immunodeficiency virus
- herpes simplex virus
- polymerase chain reaction
- respiratory syncytial virus
- reverse transcription
- Received June 17, 1998.
- Revision received April 9, 1999.
- Accepted May 16, 1999.
- American College of Cardiology
- Corallo S,
- Mutinelli R,
- Muroni M,
- et al.
- Kearney D.L,
- Popek E.J
- Grenier M.A,
- Karr S.S,
- Rakusan T.A,
- Martin G.R
- Herskowitz A,
- Wu T.C,
- Willoughby S.B,
- et al.
- ↵Salgo M, Liebman J. Cardiomyopathy or congestive heart failure (CHF) in patients on nucleoside anti-retrovirals including ddC. Bethesda (MD): U.S. Department of Health and Human Services, 1997.
- Bowles N.E,
- Rose M.L,
- Taylor P,
- et al.
- Kandolf R,
- Ameis D,
- Kirschner P,
- et al.
- Martin A.B,
- Webber S,
- Fricker F.J,
- et al.
- Aretz H.T,
- Billingham M.E,
- Edwards W.D,
- et al.
- Ou C.Y,
- Kwok S,
- Mitchell S.W,
- et al.
- Ng S.Y,
- Gunning P,
- Eddy R,
- et al.
- Schonian U,
- Crombach M,
- Maser S,
- Maisch B