OUP user menu

Acute infections in children are accompanied by oxidative modification of LDL and decrease of HDL cholesterol, and are followed by thickening of carotid intima–media

Petru Liuba, Jerker Persson, Jukka Luoma, Seppo Ylä-Herttuala, Erkki Pesonen
DOI: http://dx.doi.org/10.1016/S0195-668X(02)00750-9 515-521 First published online: 2 March 2003


Background Atherosclerosis begins early in life. Infections might contribute to the pathogenesis of atherosclerosis. In this study, we investigated whether acute infections in children could alter the carotid wall morphology and the lipid profile.

Methods Mean carotid intima–media thickness (IMT) was measured by high-resolution ultrasound in 28 hospitalised children (mean age: 5±2 years), who fulfilled the diagnostic criteria of acute infections (body temperature, >38°C; C-reactive protein, >15mg/ml, and clinical), and in 20 age- and gender-matched controls. Antibodies against oxidised low-density lipoprotein (anti-oxLDL antibodies), as well as total and high-density lipoprotein cholesterol (HDL-C) were analysed in all children. The infection group was investigated both during the acute illness and 3 months after clinical recovery (post-infection).

Results During the acute illness, the infection group had elevated anti-oxLDL antibodies and decreased HDL-C, as compared to those obtained at 3 months and in controls Math. These changes in the infection group were followed, at 3 months, by thickening of carotid intima–media. Those who received antibiotics during their acute illness had less carotid thickening than those who were not treated with antibiotics Math.

Conclusion Acute infections in children seem to be accompanied by enhanced oxidative modification of LDL and by decrease in HDL-C. These lipid changes may be followed by thickening of carotid artery intima–media. These findings suggest that, in childhood, acute infections could be associated with increased risk of atherosclerosis, and warrant further studies on this topic.

  • Carotid intima–media thickness
  • Oxidised low-density lipoprotein
  • High-density lipoprotein
  • Acute infection

1 Introduction

Atherosclerosis, the major cause of coronary heart disease, begins during early childhood.1 The time-course of atherosclerosis development is not linear. It has been hypothesised that various acute damaging stimuli promote acute vascular injury, which resolves partially only after the cessation of these stimuli.2 Subsequent exposure to additional stimuli would cause further vascular damage. The summed effect of acute vascular damages and incomplete healings could ultimately lead to the development of atherosclerotic plaque.

Although the putative role of infections in the atherosclerosis development is still controversial, negative data being also presented,3,4 accumulative evidence suggests that infections might fit into the aforementioned scenario of atherosclerosis development.5–7 Humans experience, from early childhood, repeated acute infections. Chronic infections commonly undergo, within the sameindividual, repeated relapses accompanied by acute inflammatory and autoimmune reactions. These processes appear to be key-factors in the pathogenesis of atherosclerosis.8 Of note, organisms capable of generating life-long infections with repeated reactivation, such as Chlamydia pneumoniae, Helicobacter pylori, herpes viruses, hepatitis, have been most consistently associated with atherosclerosis development.9–11

Supporting this hypothesis, previous post-mortem studies have shown significant intimal thickenings of coronary arteries in children with signs of infection at their death.12 More recently, studies on young experimental animals have shown intimal thickenings in relation to certain types of acute infections, and provided clues for a possible progression of these changes toward an atherosclerotic lesion.13,14

As yet, there are no in vivo clinical studies questioning possible damaging effects of acute infections on the arterial wall. In the present study, we investigated whether clinically manifest acute infections in children could pose such effects, and we used carotid intima–media thickness (IMT) as a marker of arterial damage. IMT of carotid arteries has been previously validated as a reliable surrogate for tracking early atherosclerotic changes owing to various risk factors for atherosclerosis.15 The method, employing high-resolution ultrasound, is non-invasive and highly reproducible. We also investigated whether acute infections could enhance the oxidative modification of low-density lipoprotein (LDL) by measuring the levels of antibodies against oxidised low-density lipoprotein (anti-oxLDL antibodies). Oxidatively modified LDL has important atherogenic properties, and has been epidemiologically associated with early atherosclerosis.16 Total and high-density lipoprotein cholesterol (HDL-C) were also measured.

2 Methods

2.1 Study material

Twenty-eight children (mean age: 5±2 years), admitted to our hospital with clinically manifest acute infections, such as acute respiratory infections, urinary tract infections, gastroenteritis or other types of infections, were enrolled. Antibiotic treatment (cephalosporins or penicillin) was given to 17 children, depending on their clinical status and C-reactive protein levels. In addition, 20 healthy age and gender-matched controls were selected. Exclusion criteria were coronary heart disease, diabetes, hypertension, smoking and hypercholesterolemia, also if present among first-degree relatives. Written consent was obtained from all children older than 7 years, or from their guardians in case of those younger than 7 years. The study was approved by the Lund University's Ethics Committee for human research.

In the infection group, carotid ultrasound investigation was performed at two different time-points: during the acute infectious illness (body temperature, >38°C and C-reactive protein, >15mg/ml), before or within the first 24h of initiation of medical treatment, and 3 months after recovery from the infectious illness (post-infection). All controls were investigated once. Blood was taken from both the infection groups, during the acute illness and after recovery, and the control group. The samples were then centrifuged and stored at −70°C. The sera were analysed for anti-oxLDL antibodies, while the plasma samples were used for HDL-C and total cholesterol analyses. Data concerning blood pressure, weight, gender, and other clinical (body temperature, antibiotic treatment, when applicable) and laboratory (C-reactive protein, serum glucose) characteristics were also obtained.

2.2 Carotid ultrasound protocol

A high-resolution ultrasound system equipped with a 15-MHz imaging probe (Acuson Sequoia) was used. During each investigation, ECG was recorded continuously and displayed on the ultrasound system's monitor. The imaging protocol has been described in detail elsewhere.17 Briefly, longitudinal scans in bi-dimensional mode of the 1-cm-distal end of the left common carotid artery were imaged so that the lumen–intima and intima–media interfaces were distinguishable. All images corresponded to the R-wave on ECG. Up to four scans were obtained from each child, and were recorded on a videotape for off-line analysis of IMT. All ultrasound scans were taken by the same sonographer (P.L.). The mean carotid IMT was measured from each scan using a computer-assisted image analyser system, previously developed in collaboration between the Wallenberg Laboratory for Cardiovascular Research and the Chalmers University of Technology, Gothenburg, Sweden.17 Mean IMT obtained from all scans from the same subject were averaged, and the resulted mean IMT was used for statistical analyses. All the computer-assisted analyses of carotid IMT were performed by experienced B-mode image readers (Gerd Östling and Birgitta Frid), who were unaware of the clinical characteristics of the studied groups.

The method, using the same image analyser system, has been previously shown to be highly reproducible.17 In our pilot study on 15 children (two sonographers: P.L. and Gerd Östling), the inter-observer and intra-observer variability were 5.7 Math and 6.4% Math, respectively.

2.3 Anti-oxLDL antibodies, HDL-C and total cholesterol

A modified ELISA was used to determine anti-oxLDL antibodies. The technique has been described in detail previously.18 The data are expressed as the ratio of binding to oxLDL to binding to native LDL (oxLDL:native LDL).

All blood samples analysed for HDL-C and total cholesterol were collected in tubes containingethylenediamine tetra-acetic acid. Plasma HDL-C and total cholesterol levels were analysed by enzymatic methods using a Hitachi Modular-P system (Roche/Hitachi 912, Roche Diagnostics). All data are expressed in millimoles per liter.

2.4 Sample size calculations

A sample size of 28 was sufficient in order to detect a difference of interest of 0.05mm (approximately 10% of the baseline IMT) with a power of 86% using a paired t-test with a 0.05 two-sided significance level.

2.5 Statistical analysis

Two-way analysis of variance was used to calculate the statistical differences in IMT, anti-oxLDL antibodies, HDL-C, total cholesterol and their ratio (total choltesterol:HDL-C) in the infection group between the time-points (acute infection and post-infection), and between the infection and the control groups. In the infection group, Student's t-test was used to assess the statistical difference in carotid IMT between the antibiotic-treated and the non-treated sub-groups. Carotid IMT was calculated as a difference between carotid IMT at 3 months and carotid IMT during the acute illness. All data are expressed as means±SEM. Significance levels were set at Math. All analyses were madewith the statistical software StatView for Windows (version 5.0, SAS Institute Inc.).

3 Results

3.1 Carotid IMT

During the acute illness, the infection group had similar carotid IMT to those obtained in the control group (0.41±0.02 versus 0.43±0.02mm, respectively; Math; Fig. 1). Instead, 3 months after clinical recovery, the infection group had significantly greater carotid IMT than during the acute illness (0.48±0.02 versus 0.41±0.02mm, respectively; Math), and than the control group (0.48±0.02 versus 0.43±0.02mm, respectively; Math; Fig. 1). The thickening of carotid intima–media in the infection group was less in those who were treated with antibiotics during their acute illness than those who did not receive antibiotics (0.055±0.02 versus 0.112±0.015mm, respectively; Math; Fig. 2).

Fig. 1

Carotid IMT in the infection group during the acute illness (acute inf) and 3 months after recovery (post-inf), and in control subjects. Data are expressed as means±SEM; Embedded Image, and Embedded Image versus acute illness and controls, respectively.

Fig. 2

IMT in the infection group during the acute illness (acute inf) and at 3 months (post-inf) in relation to antibiotic treatment (left panel). Thickening of carotid intima–media in the antibiotic-treated and antibiotic-non-treated subgroups (right panel). Data are expressed as mean±SEM; Embedded Image.

3.2 Anti-oxLDL antibodies

The anti-oxLDL antibody data are presented in Fig. 3. During the acute illness, the levels of anti-oxLDL antibodies in the infection group were higher than in the control group (2.27±0.22 versus 1.55±0.17, respectively; Math). Although the level of anti-oxLDL antibodies in these children at 3 months was still higher than in controls (1.8±0.13 versus 1.55±0.17, respectively), the difference was not significant Math.

Fig. 3

Anti-oxLDL antibodies in the infection group during the acute illness (acute inf) and at 3 months (post-inf), and in controls. The antibodies against oxLDL are expressed as the ratio of binding to oxLDL to binding to native LDL. Data are expressed as mean±SEM; Embedded Image.

3.3 HDL-C and total cholesterol

The data pertaining to HDL-C, and total cholesterol:HDL-C ratio are illustrated in Fig. 4. The plasma levels of HDL-C were significantly decreased in the infection group during the acute illness (0.68±0.07mmol/l) compared to those at 3 months (1.23±0.11mmol/l, Math) and controls (1.35±0.06mmol/l, Math). Similar differences were observed with regard to total cholesterol, i.e. 3.2±0.15mmol/l during the acute illness versus 4.18±0.22 and 4.14±0.23mmol/l at 3 months and in controls, respectively, Math. The ratio between total cholesterol and HDL-C was significantly higher in the infection group during the acute illness (6.1±0.7) than at 3 months (3.6±0.4, Math) and than controls (3.1±0.2, Math). No significant differences in these variables were observed between the infection group at 3 months and the control group.

Fig. 4

HDL-C (Panel A), and total cholesterol (TC):HDL-C ratio (Panel B) in the infection group during the acute illness (acute inf) and at 3 months (post-inf), and in controls. Data are expressed as mean±SEM; Embedded Image, and Embedded Image.

There was no significant difference in body weight, blood pressure or serum glucose levels between the infection group, either during the acute illness or convalescence, and the controls.

4 Discussion

The findings indicate that clinically manifest acute infections in children are accompanied by enhanced oxidative modification of LDL and by decrease in plasma HDL-C. The lipid changes may be followed by thickening of carotid intima–media. Of note, antibiotic treatment during the acute infectious illness was associated with less carotid thickening.

Previous studies have demonstrated that the ultrasound-detected thickness of carotidintima–media correlates well with the anatomical thickness of carotid intima.19 In the presence of conventional risk factors for atherosclerosis, thickening of carotid intima–media may be observed by ultrasound in children.15,20,21 There are well-documented similarities in terms of extent and distribution of intimal thickening between carotid and coronary arteries.22

Seroepidemiological studies on adults have shown that several chronic infections are associated with accelerated atherosclerosis in the carotid arteries.9–11 Chronic infections may be accompanied by damage of endothelial cells, hypercoagulability, altered lipid profile andenhanced auto-immunity, all of which might serve as mechanisms in the pathogenesis of atherosclerosis.23 Although these data provide support to a possible contribution of infection to the development of atherosclerosis, it remains unclear whether these vascular changes follow a linear progression as a result of continuous injury triggered by chronic infections or whether they result from periodic spurts of vascular growth and incomplete healing due to repeated episodes of acute reinfections or reactivation of a chronic infection. The latter alternative is in keeping with the natural course of atherosclerosis development as being suggested by data from autopsy24 and serial angiographic25 studies, and is supported by our finding that acute infections could promote vascular changes.

We and other researchers have earlier demonstrated that repeated administration of infectious organisms could lead to progressive endothelial damage and atherosclerotic-like intimal thickening in experimental animals.26,27 These findings favour the idea that residual vascular inflammatory changes may persist following an acute infectiousepisode, and could be further aggravated by a new infectious stimulus. By extrapolating the data from animal studies in clinical setting, one could speculate that subsequent acute infections might have additive effects on a previously damaged and incompletely healed vascular wall and, thus, the process may further evolve toward a mature atherosclerotic lesion.

One attractive hypothesis that might explain the occurrence of intimal thickening in relation to acute infections involves the participation ofarterial endothelium. Arterial endothelial injury following acute infection has been demonstrated in vivo and in vitro in different animal models.26,28 The infectious components, the systemic inflammatory response, e.g. inflammatory cytokines, C-reactive protein, and changes in lipid and homocysteine metabolisms accompanying infections, might all concur to vascular endothelial damage which, in turn, initiates a cascade of humoral- and cell-mediated intravascular processes that could ultimately lead to thickening of vessel's intima.7,29

While a transitory decrease in HDL-C during acute infections has been previously detected in both pediatric and adult population,30,31 being also documented in our study, the finding of enhanced oxidative modification of LDL in children with acute infection is novel. The oxidatively modified LDL could lead per se to intimal thickening even after a short-time exposure.32 This thickening involves both inflammatory and proliferative intimal changes, illustrating the atherogenic properties of oxidised LDL. Their detrimental effects might be exacerbated by lessening in the protective vascular mechanisms due to decreased HDL-C.19 The oxidative modification of LDL in vivo is possibly part of a complex, but non-specific host response toinfection.33

In the present study, antibiotic treatment during the acute illness was associated with less IMT. Although beneficial effects of antibiotics on intimal thickening and atherosclerosis development in relation to particular types of infections have been observed in experimental studies and in a few clinical trials, other trials have failed to demonstrate this association.34 Besides their antimicrobial properties, the anti-inflammatory and antioxidant properties might also account for the beneficial effects of antibiotics. Alternatively, the less thickening in relation to antibiotic treatment might be rather due to the type of infection, i.e. bacteria versus viruses, since most of the antibiotic-treated children were likely to have more bacterial infections than the ones in non-treated group.

In conclusion, clinically manifest acute infections in children are accompanied by transitory pro-atherogenic lipid changes, and are followed by thickening of carotid intima–media. These findings suggest that acute infections might pose a risk of atherosclerosis in childhood, and are concordant with the epidemiologically suggested concept that infections might contribute to the development of atherosclerotic coronary heart disease. The findings warrant further studies on larger pediatric samples with acute infections, including larger control samples in order to avoid spurious significances, and follow-up studies in order to further elucidate their putative linkage to atherosclerosis. Such studies are imperative since the prevention of atherosclerotic coronary heart disease is best accomplished by combating its risk factors early in life.


We are indebted to Ms Gerd Östling and Ms Birgitta Frid, laboratory technicians, for their valuable assistance in computer-assisted analyses of carotid IMT, and to Laura Darcy, registered nurse, for assistance provided throughout the study. We also thank our colleagues and nurses from the Department of Infectious Diseases, Division of Paediatrics, Lund, for their cooperation in recruiting the children with infections.


View Abstract