Secondary Logo

Journal Logo

White matter structure alterations in HIV-1-infected men with sustained suppression of viraemia on treatment

Su, Tanja; Caan, Matthan W.A.; Wit, Ferdinand W.N.M.; Schouten, Judith; Geurtsen, Gert J.; Cole, James H.; Sharp, David J.; Vos, Frans M.; Prins, Maria; Portegies, Peter; Reiss, Peter; Majoie, Charles B. on behalf of the AGEhIV Cohort Study

doi: 10.1097/QAD.0000000000000945
Epidemiology and Social

Objective: Cognitive impairment is highly prevalent in HIV-1-infected (HIV+) patients, despite adequate suppression of viral replication by combination antiretroviral therapy (cART). Cerebral white matter structure alterations are often associated with cognitive impairment and have commonly been reported in the natural course of HIV infection. However, the existence of these alterations in adequately treated HIV+ patients remains unknown, as well as its possible association with cognitive impairment.

Design: We used diffusion tensor imaging (DTI) to investigate whether white matter structure alterations exist in HIV+ patients with sustained suppressed viral replication on cART, and if such alterations are related to HIV-associated cognitive deficits.

Methods: We compared 100 aviraemic HIV+ men on cART with 70 HIV-uninfected, otherwise comparable men. Clinical and neuropsychological assessments were performed. From DTI data, white matter fractional anisotropy and mean diffusion were calculated. Subsequently, tract-based spatial statistics (TBSS) was performed, with and without masking out white matter lesions.

Results: HIV+ patients showed diffuse white matter structure alterations as compared with HIV-uninfected controls, observed as widespread decreased fractional anisotropy and an increased mean diffusion. These white matter structure alterations were associated with the number of years spent with a CD4+ cell count below 500 cells/μl, but not with HIV-associated cognitive deficits.

Conclusion: Cerebral white matter structure alterations are found in middle-aged HIV+ men with sustained suppression of viraemia on cART, and may result from periods with immune deficiency when viral toxicity and host-inflammatory responses were at their peak. These white matter structure alterations were not associated with the observed subtle HIV-associated cognitive deficits.

Video abstract:

aDepartment of Radiology, Academic Medical Center

bDepartment of Global Health, Academic Medical Center, and Amsterdam Institute for Global Health and Development (AIGHD)

cDepartment of Internal Medicine, Division of Infectious Diseases, Center for Infection and Immunity Amsterdam (CINIMA)

dDepartment of Neurology, Academic Medical Center, Amsterdam, the Netherlands

eThe Computational, Cognitive, and Clinical Neuroimaging Laboratory, Department of Medicine, Imperial College London, London, United Kingdom

fQuantitative Imaging Group, Delft University of Technology, Delft

gCluster Infectious Diseases Research, Public Health Service of Amsterdam

hDepartment of Neurology, OnzeLieveVrouweGasthuis (OLVG Hospital)

iHIV Monitoring Foundation, Amsterdam, The Netherlands.

Correspondence to Matthan W.A. Caan, PhD, Academic Medical Center, Department of Radiology, Location Z0-180, PO Box 22660, 1100 DD, Amsterdam, The Netherlands. Tel: +31 205664791; fax: +31 205669119; e-mail:

Received 31 March, 2015

Revised 11 September, 2015

Accepted 13 October, 2015

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 Website (

Back to Top | Article Outline


Although treatment of HIV-1-infected (HIV+) patients with combination antiretroviral therapy (cART) prevents severe HIV-related complications [1], cognitive impairment is often seen even when the infection appears adequately suppressed [2]. The reason for this observation is unclear, but might be explained by cerebral white matter structure alterations. Cerebral white matter structure is particularly susceptible to direct cytopathic effects of viral replication and persistent immune activation and inflammation [3]. Although cART potently suppresses viral replication and restores immune function, increased proinflammatory cytokine expression may persist [3,4]. This is supported by elevated cerebrospinal fluid (CSF) markers of immune-activation in patients on prolonged effective cART [5]. Moreover, toxicity of some cART-regimen may affect white matter structure [6]. HIV+ patients are at increased risk of developing ageing-associated pathologies such as cardiovascular disease [6–9], which may be accentuated by certain prevalent lifestyle factors (e.g. smoking, alcohol and recreational drug use), possibly impacting the white matter structure indirectly. As HIV+ patients age, all these factors may interact affecting tissue structure and eventually leading to cognitive deterioration [10].

White matter structure alterations can be captured by diffusion tensor imaging (DTI), which measures molecular diffusion of water, allowing white matter structure to be studied noninvasively [11]. DTI provides information on the degree and localization of white matter damage and type of tissue disruption [12,13].

Studies reporting DTI results in HIV-infection have thus far been inconclusive, with some studies showing white matter structure alterations in HIV+ patients [14–36], where others did not [37–39]. Furthermore, white matter structure alterations have been found to be variably related to the duration of HIV-infection, CD4+ cell counts and cognitive impairment [21–37,39,40]. Within today's context of HIV-treatment, these studies show a high variability in detectable viral load. Also, not all HIV-patients were adequately treated on cART. Finally, included HIV-uninfected controls were not always matched in terms of lifestyle and comorbid disease. Many of these studies have also used a region of interest approach to examine white matter structure, which in the context of a diffuse pathological process is prone to miss abnormalities [24,31,32]. As it remains unclear to what extent white matter structure alterations exist in chronic adequately treated HIV infection, and whether these alterations are related to cognitive impairment, a comprehensive whole brain approach is warranted [26,28].

Due to the evolving nature of the HIV/AIDS epidemic, it is likely that previous findings from studies of younger groups with high levels of viral load do not generalize to the majority of the HIV-infected population in the highly active antiretroviral therapy (HAART) era. It is essential that experimental groups are as representative of the HIV population as possible, hence our aim was to study a large group of middle-aged HIV+ men with sustained suppressed viral load on cART and include an HIV-uninfected men control group that shared socio-demographic background and had similar life style and risk factors. Within this adequately treated and well defined cohort, we aimed to identify the degree of white matter structure alterations and associations with cognitive deficits.

Back to Top | Article Outline


Study design

Eligible participants from the main AGEhIV Cohort were consecutively invited to participate in a nested neuroimaging substudy. The AGEhIV Cohort Study is an ongoing study on prevalence, incidence and risk factors of ageing-associated comorbidities and organ dysfunction among HIV+ patients and highly comparable HIV-uninfected controls at least 45 years of age (i.e. same geographic region with similar socio-demographic and behavioural (risk) factors) [8]. Inclusion criteria specific to this neuroimaging substudy were as follows: male gender, and for the HIV+ patients sustained suppression of HIV viraemia (plasma HIV-RNA < 40 copies/ml) for at least 12 months. The presence of viral ‘blips’ (transient low-level viraemia) was not an exclusion criterion. Specific exclusion criteria were current or past significant neurological disorders (i.e. traumatic brain injury with loss of consciousness >30 min, stroke, seizure disorders, multiple sclerosis and dementia), central nervous system infections and tumours. Current significant psychiatric disorders, injecting drug use, daily use of noninjection illicit drugs (with the exception of cannabis), excessive alcohol consumption (>48 units of alcohol/week), insufficient command of the Dutch language, intellectual disability and MRI contraindications were also reasons for exclusion.

Back to Top | Article Outline

Standard protocol approval, registration and patient consent

The protocol of the AGEhIV Cohort Study (including the abovementioned neuroimaging study) was approved by the institutional review board of the Academic Medical Center (AMC) and has been registered at (identifier: NCT01466582). Written informed consent was obtained separately for the cohort and neuroimaging studies, from all participants.

Back to Top | Article Outline

Life style, comorbidities and risk factors and HIV/antiretroviral therapy-related factors

All patients completed an extensive standardized questionnaire concerning a wide range of information on demographics, medical characteristics and life style factors. In addition, all patients underwent standardized screening for age-associated comorbidity and organ dysfunction including a wide range of physical and functional medical measurements. Blood and urine samples were obtained for extensive laboratory testing and detailed information concerning HIV and ART history was obtained. Further details are provided in a previous publication [8]. To examine the effect of biologically possible confounders on changes in diffusion properties, confounders of the following categories were selected from the cohort database: intoxicants, comorbidities and risk factors, biomarkers of immune activation and HIV/ART-related factors (see Tables 1 and 2 for an overview).

Table 1

Table 1

Table 2

Table 2

Back to Top | Article Outline

Neuropsychological assessment

All patients underwent a detailed neuropsychological assessment, covering fluency, attention, processing speed, memory, executive function and fine motor function. Using normative standards, test scores were converted to age and education corrected scores. Multivariate normative comparison (MNC) was performed to detect cognitive impairment [41]. Our previous work demonstrated that the MNC method can detect cognitive impairment more reliably as compared with the Frascati criteria for HIV-associated neurocognitive disorders (HAND) [42].

Back to Top | Article Outline

MRI data acquisition

All patients underwent a MRI examination at the AMC and scanning was performed on a 3T Intera and continued on a 3T Ingenia system (Philips Healthcare, Best, the Netherlands) due to a scanner upgrade. This upgrade was statistically accounted for in all analyses and the distribution of patients according to HIV serostatus per scanner system is provided in Table 1. The diffusion weighted MRI scanning parameters were as follows: echo time/repetition time (TE)/(TR) = 92/7081–9665 ms; 55 to 64 continuous slices depending on head size; data matrix 112 × 112; voxel size = 2 × 2 × 2 mm3; diffusion weighting of b = 1000 s/mm2 along 64 directions and four averages with b = 0 s/mm2. For volumetric analyses a sagittal magnetization prepared rapid gradient echo (MPRAGE) scan was acquired (TE/TR = 3.1/6.6 ms; 270 × 270 mm2 field-of-view (FOV), 170 sagittal slices of 1.2 mm thickness, 1.1 × 1.1 mm2 in-plane resolution) and to define white matter lesions of presumed vascular origin a three-dimensional fluid attenuated inversion recovery (3D-FLAIR) scan was performed (TE/TR = 355/4800 ms; inversion time = 1650 ms; FOV 250 × 250 mm2; 321 sagittal slices of 0.56 mm thickness; 1.1 × 1.1 mm2 in-plane resolution).

Back to Top | Article Outline

Image processing

All data were anonymized prior to analysis. Preprocessing of DTI data was performed with software developed in-house (Matlab; Math Works, Natick, Massachusetts, USA, using the HPCN-UvA Neuroscience Gateway and using resources of the Dutch e-Science Grid [43]. Head motion and deformations induced by eddy currents were corrected for by an affine registration of the diffusion weighted images (DWI) to the nondiffusion weighted image. The gradient directions were corrected by the rotation component of the transformation [44]. Rician noise in the DWI was reduced by an adaptive noise filtering method [45]. Diffusion tensors were calculated using a nonlinear least squares estimation. Subsequently, fractional anisotropy and mean diffusion maps were computed for each patient. All patients-data were then aligned into a common space using the nonlinear registration tool FNIRT [46].

To reduce the risk of partial volume effects, we focused our analysis on the central parts of the white matter tracts, for which a population-based fractional anisotropy-map was created and skeletonized (fractional anisotropy was thresholded at 0.2) [47]. The resulting fractional anisotropy-based skeleton represents the centre of all major tracts in the population-based template. For each individual patient, its aligned fractional anisotropy and mean diffusion maps were projected onto this fractional anisotropy skeleton (Fig. 1) [47]. Subsequently, each fractional anisotropy and mean diffusion maps were averaged to a white matter summary statistic or used for tract-based spatial statistics (TBSS) in the functional software library (FSL) [47,48].

Fig. 1

Fig. 1

Anatomical images were used for grey matter, white matter and CSF segmentation by SPM8 [49]. The intracranial volume (ICV) was computed by summing the grey matter, white matter and CSF volumes. The relative total brain volume was defined as the ratio of total brain volume (summing the grey matter and white matter) over ICV.

In addition to white matter summary statistics for each individual DTI metric, normal appearing white matter (NAWM) DTI metrics were derived in which white matter lesion areas (hyperintense on 3D-FLAIR) were masked out (Fig. 1) [50]. This involved training a ‘random-forest’ classification algorithm on a manual annotation set of 20 individuals with varying lesion load, to detect the presence of lesions [51]. This was then applied in the current dataset to identify regions of white matter lesions in HIV+ patients and controls.

Back to Top | Article Outline

Statistical analysis

Group comparisons of patient characteristics between HIV+ patients and controls were performed (Tables 1 and 2).

Across all patients, the effect of HIV serostatus on white matter fractional anisotropy (WMFA) and white matter mean diffusion (WMMD) measures were examined by linear regression analyses, while adjusting for age, relative brain volume and scanner system. To examine possible accelerated aging effects in HIV+ patients, interaction effects between age and HIV serostatus on WMFA and WMMD measures were assessed.

To identify confounders and determinants (intoxicants, comorbidities and risk factors, biomarkers of immune activation and HIV/ART-related factors) of WMFA and WMMD measures, stepwise linear regression analyses were performed (P < 0.05 probability to enter and P > 0.1 probability to remove), while adjusting for age, brain volume and scanner system.

To examine associations of WMFA and WMMD measures with cognition, HIV+ patients were classified as either cognitively impaired or unimpaired by MNC. Group comparison on cognitive status was performed on the WMFA and WMMD measures, adjusted for age, brain volume and scanner system. The test statistic of the MNC method was Hotelling's T2. To create a continuous measure of cognitive function and prevent a bimodal distribution, each Hotelling's T2 statistic was subtracted from the lowest Hotelling's T2 statistic and multiplied by the direction of deviation (i.e. positive or negative deviation reflecting better or poorer cognition as compared with the control group). Associations of this transformed statistic with WMFA and WMMD, were examined by linear regression, adjusting for age, brain volume and scanner system.

To provide spatial information on significant findings of white matter DTI metrics, TBSS group comparison and correlational analyses were performed, while adjusting for age, brain volume and scanner system. Multiple comparisons were corrected for by using permutation tests. This was implemented using the Randomise software within FSL, employing the threshold-free cluster enhancement (TFCE), in which P-values <0.05 were considered significant. All analyses were subsequently repeated with NAWM DTI metrics.

MNC was performed using R statistical software [52], whereas remaining analyses were done in SPSS (version 20.0; IBM, Armonk, New York, USA).

Back to Top | Article Outline


Demographic and clinical characteristics

Participants were enrolled to the neuroimaging study between December 2011 and August 2013. Neuroimaging data were available from 100 HIV+ patients and 70 controls. An overview of the demographics, neuroimaging and HIV/ART-related factors are shown in Table 1, whereas Table 2 provides an overview of intoxicants, comorbidities and risk factors and biomarkers of immune activation. The HIV+ patients [median age: 54 (interquartile, IQR 49–61) years] were highly comparable to the controls [median age: 53 (IQR 49–59) years]. Both groups were also similar in their substance use behaviour, except for ecstasy use, which was more common in the controls (11 vs. 2%, P = 0.02). Controls also had higher plasma concentrations of glycated haemoglobin (HbA1c, 37 vs. 35, P = 0.01) and BMI (26 vs. 24, P = 0.002). HIV+ patients fulfilled more often the criteria for central obesity (i.e. waist/hip ratio>0.9) and had greater lifetime tobacco exposure as measured by pack-years of smoking (P = 0.03, P = 0.01). Levels of soluble CD14 were higher in HIV+ patients, whereas CD4+/CD8+ ratios were lower as compared with controls (P < 0.001, P < 0.001). No group differences were found for the remaining factors.

HIV+ patients had been treated with antiretroviral therapy for a median duration of 11.4 (IQR 4.9–14.9) years and showed substantial immune recovery. Their median nadir CD4+ cell count was 170 (IQR 60–248), with a current CD4+ cell count of 620 (IQR 475–787) cells/μl.

Back to Top | Article Outline

Group comparisons on white matter diffusion properties

Across the white matter there were significant differences between the HIV+ patients and controls. HIV+ patients showed significantly lower WMFA (P = 0.03) and significantly higher WMMD (P = 0.02). See Table 1 for further details. Interaction effects were assessed and no interaction effect of age and HIV serostatus was found for these white matter DTI-metrics (fractional anisotropy: P = 0.59, mean diffusion: P = 0.58).

Voxel-wise comparison by TBSS showed a widespread pattern of lower fractional anisotropy and higher mean diffusion in HIV+ patients compared with controls (Fig. 2). Patterns of reduced fractional anisotropy were seen in projection- and thalamic fibres (i.e. cortical spinal tract and anterior thalamic radiation), all major association fibres (i.e. superior longitudinal fasciculus, inferior longitudinal fasciculus, inferior-fronto-occipital fasciculus and uncinated fasciculus), limbic system fibres (i.e. cingulum) and callosal fibres (i.e. forceps minor and forceps major). Differences in mean diffusion were less pronounced and mainly localized in the left hemisphere, although lowering the statistical threshold also showed contralateral effects (data not shown).

Fig. 2

Fig. 2

Back to Top | Article Outline

Determinants and confounders of altered white matter diffusion properties

Joint analysis of HIV+ patients and controls showed that HIV serostatus, age, lower brain volume and the number of antihypertensive medications used were significantly associated with lower WMFA and higher WMMD (Table 3: Model 1). Remaining possible confounding variables examined (see Tables 1 and 2 for a complete overview) did not contribute sufficiently to the white matter DTI-metrics and were therefore not selected for the final model.

Table 3

Table 3

When restricting the analysis to the HIV+ group, the number of antihypertensive medications used remained significantly associated. Additionally, higher LDL-cholesterol and duration spent with CD4+ cell count below 500 cells/μl, were significantly associated with higher WMMD (Table 3: Model 2). Patients who had been treated with mono- or dual therapy with nucleoside reverse transcriptase inhibitors before the start of cART showed significantly higher WMMD (P = 0.04). However, this effect was confounded by duration spent with CD4+ cell count below 500 cells/μl and did not remain significant. No collinearity was found between age or volume and the determinants.

Voxel-wise correlation analyses by TBSS showed a diffuse pattern of significantly increased mean diffusion with longer duration spent with a CD4+ cell count below 500 cells/μl (Fig. 2). This relation was found in projection and thalamic fibres, all major association fibres, limbic system fibres and callosal fibres.

Back to Top | Article Outline

Altered white matter diffusion properties and its association with cognitive performance

Sixteen percentage of the HIV+ patients were classified as cognitively impaired by MNC (alpha was 5%, one-tailed, assuming a specificity of 95%, which was previously verified) [42]. Comparing cognitively impaired and cognitively unimpaired HIV+ patients, no significant differences in WMFA (P = 0.82) or WMMD (P = 0.91) were found. Overall poorer cognitive performance was not associated with WMFA (ß = 0.007, P = 0.82, η2 < 0.001) or WMMD (ß = –0.051, P = 0.61, η2 = 0.003).

Back to Top | Article Outline

Repeating the analyses on the normal appearing white matter

All findings on WMFA and WMMD alterations persisted after excluding white matter lesion areas from the analyses (including effects of HIV serostatus, the number of antihypertensive medications used, LDL-cholesterol and duration spent with CD4+ cell count below 500 cells/μl). See supplementary Results and Tables 1 and 2,

Back to Top | Article Outline


Key findings

In this study, cerebral white matter structure was assessed in 100 middle-aged HIV+ men with well suppressed viral load on cART and compared with 70 HIV-uninfected, but otherwise highly comparable, controls. We found significant white matter structure alterations in HIV+ patients, which consisted of lower fractional anisotropy and higher mean diffusion, as assessed by DTI. TBSS showed that these effects were widespread throughout the brain. Additionally, deleterious effects of hypertension, dyslipidaemia and duration of past immune deficiency were found. HIV-associated cognitive deficits were not found to be associated with white matter structure alterations.

Back to Top | Article Outline

Interpretation of findings

The diffuse pattern of altered diffusion properties found in HIV+ patients relative to highly comparable HIV-uninfected controls may indicate subtle but widespread white matter injury. Consistent findings of alterations in fractional anisotropy and mean diffusion have been reported previously in middle-aged and older HIV+ patients compared with healthy controls [14,16,19,22,23,26,27,29,30,33,34,36]. The origin of the subtlety of our findings compared with previous studies may be two-fold. First, all HIV-patients in our cohort were adequately treated on cART. Second, healthy controls were carefully matched on lifestyle and comorbid disease. Comparing to previous work, one might infer that improved treatment diminishes HIV-induced white matter structure alterations. The fact that two comparable studies which exclusively included aviraemic HIV+ patients did not report white matter structure alterations might be due to smaller sample sizes as compared with our study [35,39]. Furthermore, one of the studies also excluded all possible comorbidities [39] and therefore may have excluded HIV effects, as HIV itself has been reported to be independently associated with cardiovascular disease and many other comorbidities [10].

In addition to the effects of HIV serostatus and ageing, we found independent associations of hypertension and dyslipidemia with white matter structure alterations. Effects of hypertension were particularly consistent, which might be partly HIV-mediated, as increased cardiovascular risk has been frequently reported in HIV [10]. Evidence for possible accelerated CNS ageing could not be derived from this cross-sectional analysis. Follow-up measurements are currently underway and may provide more insight into a possible interaction effect of age and HIV serostatus on white matter structure alterations in adequately treated HIV+ patients.

The total duration of immune deficiency (i.e. the number of years spent with CD4+ cell count lower than 500 cells/μl) was also strongly associated with white matter structure alterations. This may reflect irreversible damage that has occurred during immune deficiency by both direct viral and host-derived proinflammatory factors. Previously reported persistent white matter injury in HIV+ patients with partial immune reconstitution, provides evidence that such damage could be permanent in nature [33]. Moreover, the use of cART and higher current CD4+ cell counts have been associated with higher fractional anisotropy values [40], suggesting that prevention of immune deficiency may avert irreversible white matter structure damage.

Among HIV+ patients in the cART-era, effects of cumulative exposure to immune deficiency are possibly insufficiently captured by the nadir CD4+ cell count [33–35,39,40]. Subtle white matter structure alterations may be better captured by a measure of cumulative exposure to immune deficiency, as used in this study. These findings provide additional support for the current HIV treatment guidelines that stress the importance of preventing immune deficiency and initiating antiretroviral therapy in all patients irrespective of CD4+ cell count [53].

Pretreatment with nucleoside-analogue reverse transcriptase inhibitors before the start of cART was associated with white matter structure injury, but this relationship seemed to be driven by the duration of past immune deficiency. This is compatible with HIV+ patients diagnosed in the pre cART era to have been more likely to have experienced more prolonged periods of advanced immune deficiency.

Although associations between systemic markers of immune activation and inflammation (i.e. sCD14 and sCD16) with white matter structure alterations were not observed, ongoing proinflammatory reactions within the CNS affecting white matter structure cannot be ruled out. Postmortem studies have reported persistent levels of elevated markers of microglia/macrophage activation in cART treated HIV cases [54], suggesting proinflammatory reactions may not be normalized in the context of cART and that the continued presence of neuro- and myelinotoxic cytokines may induce subtle white matter alterations, such as those observed in the current study.

Although several studies have previously reported white matter structure alterations to be related to HIV-associated cognitive impairment [24,27,28,31,34,37], others did not report such an relationship [26,30,35,39]. An association between white matter structure alterations and cognitive deficits was not observed in the current study. The magnitude of the effect of HIV-serostatus on cognition and white matter structure alterations in the current study was small; hence a possible relationship between the two would likely be subtle. Note that in absolute numbers, our group of 16 cognitively impaired patients is of comparable size to the CHARTER cohort, reporting 10 impaired patients [28]. The successful treatment of HIV+ patients and exclusion of otherwise confounding factors are in our opinion key to the subtlety of effects we report. If white matter structure is to be used as a biomarker to predict cognitive impairment and subsequent deterioration in ageing HIV+ patients, then larger, multicentre cohorts with hundreds of patients are needed. Also, sufficient time between follow-up measurements is required in future studies, as cognitive decline was not related to a significant increase of mean diffusion in HIV+ patients after one year of follow-up [27].

The presence of white matter lesions of presumed vascular origin is associated with findings of white matter structure alterations [55]. Greater microstructural alterations have been reported in HIV+ patients with white matter lesions, compared with HIV+ patients with no lesions [31]. Although some studies have carefully reviewed the analysed regions for white matter lesions [15,26], we studied the NAWM separately by patient-wise excluding lesion areas. All effects persisted in whole brain measures.

Back to Top | Article Outline


Although controls enrolled in this study had very similar demographics and lifestyle, HIV+ patients reported more lifetime tobacco exposure and controls more ecstasy use. Moreover, although HIV+ patients had lower BMI, they fulfilled more often the criteria for central obesity, and showed increased monocyte activation (i.e. higher levels of soluble CD14) and lower CD4+/CD8+ ratio. Such vascular risk factors and increased immune activation are known complications of HIV-infection or its treatment. However, none of these parameters were related to white matter structure alterations in the current study. About one-third of HIV+ patients and controls were scanned on a different scanning system due to a scanner replacement. We have adjusted for the scanner effects by factoring its effects in the statistical model. When analysing the subset of patients on a single machine, the effects within HIV+ patients are persistent, the relation between CD4+ cell count less than 500 cells/μl and mean diffusion remains significant (beta = 0.26, P = 0.005). When comparing groups, effects sizes are of equal sign but slightly smaller, and P-values thus higher (fractional anisotropy: beta = –0.010, P = 0.19; mean diffusion: beta = 0.11, P = 0.09).

Back to Top | Article Outline


In this 3T DTI study, middle-aged HIV+ men with suppressed viraemia on cART showed pronounced white matter structure alterations as compared with highly comparable HIV-uninfected controls. The association with duration of exposure to immune deficiency suggests irreversible damage from previous periods of immune deficiency when host-inflammatory and virus toxicity were at their peak. In addition, independent associations between vascular risk factors and white matter structure abnormalities were found. Longitudinal follow-up studies are needed to determine the progression and synergistic effects of these risk factors.

Back to Top | Article Outline


We thank Merel Burgering for her assistance in MRI scanning. We thank Sandra van der Berg and Raschel Snoeks for their help considering MRI contraindications. We thank Paul Groot for his support concerning data transport and storage. Above all, we gratefully acknowledge all study participants for their co-operation.

This work was supported by the Nuts-OHRA Foundation (grant no 1003–026), Amsterdam, The Netherlands, as well as by The Netherlands Organisation for Health Research and Development (ZonMW) together with AIDS Fonds (grant nos 300020007 and 2009063, respectively). Additional unrestricted scientific grants were received from Gilead Sciences, ViiV Healthcare, Janssen Pharmaceutica N.V., Bristol-Myers Squibb, Boehringer Ingelheim, and Merck & Co.

None of these funding bodies had a role in the design or conduct of the study, the analysis and interpretation of the results, or the decision to publish.

T.S.: Collected study data, performed statistical analysis and the literature search, and drafted the manuscript.

M.C.: Processed the MRI data, supervised MRI data analysis, contributed to the data interpretation, and contributed to the writing of the manuscript.

F.W.: Contributed to the study design, supervised statistical analysis, contributed to data interpretation, and critically reviewed and revised the manuscript.

J.S.: Contributed to data collection, data interpretation, and critically reviewed and revised the manuscript.

G.G.: Contributed to data interpretation and critically reviewed and revised the manuscript.

J.C.: Contributed to data interpretation and critically reviewed and revised the manuscript.

D.S.: Contributed to data interpretation and critically reviewed and revised the manuscript.

F.V.: Contributed to the study design, data interpretation, and critically reviewed and revised the manuscript.

M.P.: Contributed to the study design, data interpretation, and critically reviewed and revised the manuscript.

P.P.: Contributed to study design, data interpretation, and critically reviewed and revised the manuscript.

P.R. conceived the main cohort study and the sub-study, obtained study funding, contributed to both study designs, to data interpretation, and critically reviewed and revised the manuscript.

C.M.: Conceived the sub-study and obtained study funding, contributed to its design, data interpretation, and critically reviewed and revised the manuscript.

Study funding: This work was supported by the Nuts-OHRA Foundation (grant no 1003-026), Amsterdam, The Netherlands, as well as by The Netherlands Organisation for Health Research and Development (ZonMW) together with AIDS Fonds (grant nos 300020007 and 2009063, respectively). Additional unrestricted scientific grants were received from Gilead Sciences, ViiV Healthcare, Janssen Pharmaceutica N.V., Bristol-Myers Squibb, Boehringer Ingelheim, and Merck&Co.

None of these funding bodies had a role in the design or conduct of the study, the analysis and interpretation of the results, or the decision to publish.

Disclosures: T.S. has received travel grants from Boehringer Ingelheim.

M.C. has received travel grants from Boehringer Ingelheim.

F.W. has received travel grants from Gilead Sciences, ViiV Healthcare, Boehringer Ingelheim, Abbvie, and Bristol-Myers Squibb.

J.S. has received travel grants from Gilead Sciences, ViiV Healthcare, and Boehringer Ingelheim.

D.S. is funded by a National Institute of Health Research Professorship (NIHR-RP-011-048) and has received an investigator-led grant from Pfizer, unrelated to the current work.

P.P. has been an ad hoc advisor to or speaking at various events sponsored by ViiV Healthcare, Gilead Sciences, Abbvie and Bristol-Myers Squibb.

P.R. through his institution has received independent scientific grant support from Gilead Sciences, Janssen Pharmaceuticals Inc., Merck&Co, Bristol-Myers Squibb, Boehringer Ingelheim and ViiV Healthcare, and travel support through his institution from Gilead Sciences and Janssen Pharmaceuticals Inc. In addition he has served on a scientific advisory board for Gilead Sciences and serves on a data safety monitoring committee for Janssen Pharmaceutica N.V., for which his institution has received remuneration.

Back to Top | Article Outline

Conflicts of interest

C.M., G.G., J.C., F.V. and M.P. have no conflicts of interest.

Back to Top | Article Outline


1. Wada N, Jacobson LP, Cohen M, French A, Phair J, Munoz A. Cause-specific life expectancies after 35 years of age for human immunodeficiency syndrome-infected and human immunodeficiency syndrome-negative individuals followed simultaneously in long-term cohort studies, 1984-2008. Am J Epidemiol 2013; 177:116–125.
2. Schouten J, Cinque P, Gisslen M, Reiss P, Portegies P. HIV-1 infection and cognitive impairment in the cART era: a review. AIDS 2011; 25:561–575.
3. Anthony IC, Bell JE. The neuropathology of HIV/AIDS. Int Rev Psychiatry 2008; 20:15–24.
4. Anzinger JJ, Butterfield TR, Angelovich TA, Crowe SM, Palmer CS. Monocytes as regulators of inflammation and HIV-related comorbidities during cART. J Immunol Res 2014; 2014:569819.
5. Edén A, Price RW, Spudich S, Fuchs D, Hagberg L, Gisslén M. Immune activation of the central nervous system is still present after >4 years of effective highly active antiretroviral therapy. J Infect Dis 2007; 196:1779–1783.
6. Manji H, Jäger HR, Winston A. HIV, dementia and antiretroviral drugs: 30 years of an epidemic. J Neurol Neurosurg Psychiatry 2013; 84:1126–1137.
7. Hasse B, Ledergerber B, Furrer H, Battegay M, Hirschel B, Cavassini M, et al. Morbidity and aging in HIV-infected persons: the Swiss HIV cohort study. Clin Infect Dis 2011; 53:1130–1139.
8. Schouten J, Wit FW, Stolte IG, Kootstra N, van der Valk M, Geerlings SG, et al. Cross-sectional comparison of the prevalence of age-associated comorbidities and their risk factors between HIV-infected and uninfected individuals: the AGEhIV Cohort Study. Clin Infect Dis 2014. 1–29.
9. Zanni MV, Schouten J, Grinspoon SK, Reiss P. Risk of coronary heart disease in patients with HIV infection. Nat Rev Cardiol 2014; 11:728–741.
10. Deeks SG. HIV infection, inflammation, immunosenescence, and aging. Annu Rev Med 2011; 62:141–155.
11. Basser PJ, Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI. 1996. J Magn Reson 2011; 213:560–570.
12. Sun S-W, Liang H-F, Trinkaus K, Cross AH, Armstrong RC, Song S-K. Noninvasive detection of cuprizone induced axonal damage and demyelination in the mouse corpus callosum. Magn Reson Med 2006; 55:302–308.
13. Song S-K, Sun S-W, Ramsbottom MJ, Chang C, Russell J, Cross AH. Dysmyelination revealed through MRI as increased radial (but unchanged axial) diffusion of water. Neuroimage 2002; 17:1429–1436.
14. Filippi CG, Ulug a M, Ryan E, Ferrando SJ, van Gorp W. Diffusion tensor imaging of patients with HIV and normal-appearing white matter on MR images of the brain. AJNR Am J Neuroradiol 2001; 22:277–283.
15. Pomara N, Crandall DT, Choi SJ, Johnson G, Lim KO. White matter abnormalities in HIV-1 infection: a diffusion tensor imaging study. Psychiatry Res 2001; 106:15–24.
16. Pfefferbaum A, Rosenbloom MJ, Adalsteinsson E, Sullivan EV. Diffusion tensor imaging with quantitative fibre tracking in HIV infection and alcoholism comorbidity: synergistic white matter damage. Brain 2007; 130:48–64.
17. Hoare J, Fouche JP, Spottiswoode B, Joska JA, Schoeman R, Stein DJ, Carey PD. White matter correlates of apathy in HIV-positive subjects: a diffusion tensor imaging study. J Neuropsychiatry Clin Neurosci 2010; 22:313–320.
18. Hoare J, Jacqueline H, Westgarth-Taylor J, Jenny W, Fouche J-P, Jean-Paul F, et al. A diffusion tensor imaging and neuropsychological study of prospective memory impairment in South African HIV positive individuals. Metab Brain Dis 2012; 27:289–297.
19. Müller-Oehring EM, Schulte T, Rosenbloom MJ, Pfefferbaum A, Sullivan EV. Callosal degradation in HIV-1 infection predicts hierarchical perception: a DTI study. Neuropsychologia 2010; 48:1133–1143.
20. Leite SCB, Corrêa DG, Doring TM, Kubo TT, Netto TM, Ferracini R, et al. Diffusion tensor MRI evaluation of the corona radiata, cingulate gyri, and corpus callosum in HIV patients. J Magn Reson Imaging 2013; 38:1488–1493.
21. Ragin AB, Storey P, Cohen BA, Edelman RR, Epstein LG. Disease burden in HIV-associated cognitive impairment: a study of whole-brain imaging measures. Neurology 2004; 63:2293–2297.
22. Thurnher MM, Castillo M, Stadler A, Rieger A, Schmid B, Sundgren PC. Diffusion-tensor MR imaging of the brain in human immunodeficiency virus-positive patients. AJNR Am J Neuroradiol 2005; 26:2275–2281.
23. Wu Y, Storey P, Cohen BA, Epstein LG, Edelman RR, Ragin AB. Diffusion alterations in corpus callosum of patients with HIV. AJNR Am J Neuroradiol 2006; 27:656–660.
24. Chen Y, An H, Zhu H, Stone T, Smith JK, Hall C, et al. White matter abnormalities revealed by diffusion tensor imaging in nondemented and demented HIV+ patients. Neuroimage 2009; 47:1154–1162.
25. Pfefferbaum A, Rosenbloom MJ, Rohlfing T, Kemper CA, Deresinski S, Sullivan EV. Frontostriatal fiber bundle compromise in HIV infection without dementia. AIDS 2009; 23:1977–1985.
26. Stebbins GT, Smith CA, Bartt RE, Kessler HA, Adeyemi OM, Martin E, et al. HIV-associated alterations in normal-appearing white matter. J Acquir Immune Defic Syndr 2007; 46:564–573.
27. Chang L, Wong V, Nakama H, Watters M, Ramones D, Miller EN, et al. Greater than age-related changes in brain diffusion of HIV patients after 1 year. J Neuroimmune Pharmacol 2008; 3:265–274.
28. Gongvatana A, Schweinsburg BC, Taylor MJ, Theilmann RJ, Letendre SL, Alhassoon OM, et al. White matter tract injury and cognitive impairment in human immunodeficiency virus-infected individuals. J Neurovirol 2009; 15:187–195.
29. Du H, Wu Y, Ochs R, Edelman RR, Epstein LG, McArthur J, et al. A comparative evaluation of quantitative neuroimaging measurements of brain status in HIV infection. Psychiatry Res 2012; 203:95–99.
30. Nakamoto BK, Jahanshad N, McMurtray A, Kallianpur KJ, Chow DC, Valcour VG, et al. Cerebrovascular risk factors and brain microstructural abnormalities on diffusion tensor images in HIV-infected individuals. J Neurovirol 2012; 18:303–312.
31. Stubbe-Drger B, Deppe M, Mohammadi S, Keller SS, Kugel H, Gregor N, et al. Early microstructural white matter changes in patients with HIV: a diffusion tensor imaging study. BMC Neurol 2012; 12:23.
32. Tate DF, Conley J, Paul RH, Coop K, Zhang S, Zhou W, et al. Quantitative diffusion tensor imaging tractography metrics are associated with cognitive performance among HIV-infected patients. Brain Imaging Behav 2010; 4:68–79.
33. Zhu T, Zhong J, Hu R, Tivarus M, Ekholm S, Harezlak J, et al. Patterns of white matter injury in HIV infection after partial immune reconstitution: a DTI tract-based spatial statistics study. J Neurovirol 2013; 19:10–23.
34. Nir TM, Jahanshad N, Busovaca E, Wendelken L, Nicolas K, Thompson PM, et al. Mapping white matter integrity in elderly people with HIV. Hum Brain Mapp 2014; 35:975–992.
35. Wright PW, Heaps JM, Shimony JS, Thomas JB, Ances BM. The effects of HIV and combination antiretroviral therapy on white matter integrity. AIDS 2012; 26:1501–1508.
36. Wright PW, Vaida FF, Fernández RJ, Rutlin J, Price RW, Lee E, et al. Cerebral white matter integrity during primary HIV infection. AIDS 2015; 29:433–442.
37. Ragin AB, Wu Y, Storey P, Cohen BA, Edelman RR, Epstein LG. Diffusion tensor imaging of subcortical brain injury in patients infected with human immunodeficiency virus. J Neurovirol 2005; 11:292–298.
38. Schulte T, Müller-Oehring EM, Javitz H, Pfefferbaum A, Sullivan EV. Callosal compromise differentially affects conflict processing and attentional allocation in alcoholism, HIV, and their comorbidity. Brain Imaging Behav 2008; 2:27–38.
39. Towgood KJ, Pitkanen M, Kulasegaram R, Fradera A, Kumar A, Soni S, et al. Mapping the brain in younger and older asymptomatic HIV-1 men: frontal volume changes in the absence of other cortical or diffusion tensor abnormalities. Cortex 2012; 48:230–241.
40. Gongvatana A, Cohen RA, Correia S, Devlin KN, Miles J, Kang H, et al. Clinical contributors to cerebral white matter integrity in HIV-infected individuals. J Neurovirol 2011; 17:477–486.
41. Huizenga HM, Smeding H, Grasman RP, Schmand B. Multivariate normative comparisons. Neuropsychologia 2007; 45:2534–2542.
42. Su T, Schouten J, Geurtsen GJ, Wit FW, Stolte IG, Prins M, et al. Multivariate normative comparison, a novel method for more reliably detecting cognitive impairment in HIV infection. AIDS 2015; doi:101097/QAD0000000000000573.
43. Shahand S, Benabdelkader A, Jaghoori MM, al Mourabit M, Huguet J, Caan MWA, et al. A data-centric neuroscience gateway: design, implementation, and experiences. Concurr Comput Pract Exp 2015; 27:489–506.
44. Leemans A, Jones DK. The B-matrix must be rotated when correcting for subject motion in DTI data. Magn Reson Med 2009; 61:1336–1349.
45. Caan MW, Khedoe G, Poot D, den Dekker A, Olabarriaga S, Grimbergen K, et al. Adaptive noise filtering for accurate and precise diffusion estimation in fiber crossings. Med Image Comput Comput Assist Interv 2010; 13:167–174.
46. Andersson JLR, Jenkinson M, Smith S. Nonlinear registration aka spatial normalisation. FMRIB Technial Report, TR07JA2; 2007.
47. Smith SM, Jenkinson M, Johansen-Berg H, et al. Tract-based spatial statistics: voxelwise analysis of multisubject diffusion data. Neuroimage 2006; 31:1487–1505.
48. Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TEJ, Johansen-Berg H, et al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 2004; 23 (Suppl 1):S208–S219.
49. Ashburner J, Friston KJ. Unified segmentation. NeuroImage 2005; 26:839–851.
50. Steenwijk MD, Pouwels PJW, Daams M, van Dalen JW, Caan MW, Richard E, et al. Accurate white matter lesion segmentation by k nearest neighbor classification with tissue type priors (kNN-TTPs). NeuroImage Clin 2013; 3:462–469.
51. Breiman L. Random forests. Mach Learn 2001; 45:1–35.
52. R Development Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2011.
53. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Rockville, Maryland: Department of Health and Human Services; 2007.
54. Anthony IC, Ramage SN, Carnie FW, Simmonds P, Bell JE. Influence of HAART on HIV-related CNS disease and neuroinflammation. J Neuropathol Exp Neurol 2005; 64:529–536.
55. Lange RT, Shewchuk JR, Heran MKS, Rauscher A, Jarrett M, Brubacher JR, et al. To exclude or not to exclude: further examination of the influence of white matter hyperintensities in diffusion tensor imaging research. J Neurotrauma 2014; 31:198–205.
56. Schmand B, Lindeboom J, van Harskamp F. Dutch adult reading test. Lisse: Swets en Zeitlinger; 1992.
    57. Letendre SL, Ellis RJ, Ances BM, McCutchan JA. Neurologic complications of HIV disease and their treatment. Top HIV Med 2010; 18:45–55.
    58. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2014; 37 (Suppl 1):S81–90.
      59. Pickering TG, Hall JE, Appel LJ, Falkner BE, Graves JW, Hill MN, et al. Recommendations for blood pressure measurement in humans: an AHA scientific statement from the Council on High Blood Pressure Research Professional and Public Education Subcommittee. J Clin Hypertens (Greenwich) 2005; 7:102–109.
      60. Horváth IG, Németh A, Lenkey Z, Alessandri N, Tufano F, Kis P, et al. Invasive validation of a new oscillometric device (Arteriograph) for measuring augmentation index, central blood pressure and aortic pulse wave velocity. J Hypertens 2010; 28:2068–2075.
      61. Baulmann J, Schillings U, Rickert S, Uen S, Düsing R, Illyes M, et al. A new oscillometric method for assessment of arterial stiffness: comparison with tonometric and piezo-electronic methods. J Hypertens 2008; 26:523–528.
      62. WHO. Waist circumference and waist-hip ratio: report of a WHO expert consultation. Geneva, Switzerland: WHO; 2008.

        aging; antiretroviral therapy; cerebral white matter; diffusion tensor imaging (DTI); HIV-1-infection; HIV-associated neurocognitive disorders (HAND); neuropsychological assessment

        Supplemental Digital Content

        Back to Top | Article Outline
        Copyright © 2016 Wolters Kluwer Health, Inc.