HIV, Vascular and Aging Injuries in the Brain of Clinically Stable HIV-Infected Adults: A 1H MRS Study

Background Cardiovascular disease (CVD) and premature aging have been hypothesized as new risk factors for HIV associated neurocognitive disorders (HAND) in adults with virally-suppressed HIV infection. Moreover, their significance and relation to more classical HAND biomarkers remain unclear. Methods 92 HIV− infected (HIV+) adults stable on combined antiretroviral therapy (cART) and 30 age-comparable HIV-negative (HIV−) subjects underwent 1H Magnetic Resonance Spectroscopy (MRS) of the frontal white matter (targeting HIV, normal aging or CVD-related neurochemical injury), caudate nucleus (targeting HIV neurochemical injury), and posterior cingulate cortex (targeting normal/pathological aging, CVD-related neurochemical changes). All also underwent standard neuropsychological (NP) testing. CVD risk scores were calculated. HIV disease biomarkers were collected and cerebrospinal fluid (CSF) neuroinflammation biomarkers were obtained in 38 HIV+ individuals. Results Relative to HIV− individuals, HIV+ individuals presented mild MRS alterations: in the frontal white matter: lower N-Acetyl-Aspartate (NAA) (p<.04) and higher myo-inositol (mIo) (p<.04); in the caudate: lower NAA (p = .01); and in the posterior cingulate cortex: higher mIo (p<.008– also significant when Holm-Sidak corrected) and higher Choline/NAA (p<.04). Regression models showed that an HIV*age interaction was associated with lower frontal white matter NAA. CVD risk factors were associated with lower posterior cingulate cortex and caudate NAA in both groups. Past acute CVD events in the HIV+ group were associated with increased mIo in the posterior cingulate cortex. HIV duration was associated with lower caudate NAA; greater CNS cART penetration was associated with lower mIo in the posterior cingulate cortex and the degree of immune recovery on cART was associated with higher NAA in the frontal white matter. CSF neopterin was associated with higher mIo in the posterior cingulate cortex and frontal white matter. Conclusions In chronically HIV+ adults with long-term viral suppression, current CVD risk, past CVD and age are independent factors for neuronal injury and inflammation. This suggests a tripartite model of HIV, CVD and age likely driven by chronic inflammation.


Standard neurobehavioral evaluation
Prior to the assessment of neurocognitive functions, information regarding participants' medical, social and occupational history was obtained through a basic demographic inventory, administered via an interview format. Depressive complaints, cognitive complaints, and Independence in Activities of Daily Living (IADL) were assessed with standard self-report instruments (Beck Depression Inventory-II [1,2], Personal Assessment of Own Functioning Inventory [3] and IADL [4]). Antiretroviral medication adherence was also recorded using a questionnaire was adapted from [5] recommendations. The neurobehavioral examination also included a brief neuropsychiatric semi-structured interview (Electronic Mini International Neuropsychiatric Interview -eM.I.N.I version 5.2 [6]). On this examination, we found that 11% of the HIV+ participant had experienced a major depressive episode. As per design, no participants were found to have current substance abuse or dependence disorders. All participants underwent a standard neuropsychological evaluation to assess functions of attention/working memory; verbal learning and memory, verbal generativity; fine-motor coordination; mental flexibility/inhibition and speed of information processing (See Table S1 for individual tests). The session duration was approximately two hours.
The standard neuropsychological testing was conducted by research assistants/students who were neuro/psychology graduates (EF, TL and DM were trained in the administration of the testing protocol under the supervision of a senior neuropsychologist, LAC). All measures were administered and scored according to standard procedures. Neuropsychological scores were converted into demographically corrected T-scores and deficit scores to compute overall impairment rate and domain specific performance scores Carey et al., 2004). Using published normative data [7,8], raw scores were converted to demographically corrected T-scores in order to minimize the influence as appropriate of age, education, sex and ethnicity (Caucasians versus other). T-scores were then transformed into a Global Deficit Score (GDS, [9]) according to the following criteria: T-scores greater than 40 reflects no impairment (deficit score = 0), whereas a deficit score of 1 reflects mild impairment (T score = 39 to 35), deficit score of 2 reflects mild to moderate impairment (T score = 34 to 30), 3 reflects moderate impairment (T score = 29 to 25), 4 reflects moderate to severe impairment (T score = 24 to 20), and 5 reflects severe impairment (T score <20). Deficit scores on all tests were averaged to create the GDS. A GDS ≥ 0.5 a clinically validated and reliable cutoff [9] which indicates that, on average, an individual is at least mildly impaired in at least half the single neuropsychological measure in the battery (the current battery was composed of 16 neuropsychological measures). Digit Symbol Coding: Requires the participant to fill the blank spaces according to a key for 120 seconds [8,11] Colour-Word Interference Test: A speed-based test with four conditions that require the participant to name ink colours, read words and inhibit prepotent responses [13] VERBAL GENERATIVITY Controlled Oral Word Association Test: A measure of executive functioning which evaluates the spontaneous production of words within one minute [8,14] File S2 Localization of the voxel for brain spectra acquisition

Right Caudate Nucleus / Basal ganglia
Voxel was positioned as much as possible within the right caudate nucleus. When severe atrophy was present the voxel was positioned in the basal ganglia.

Rationale for voxel selection
The frontal white matter has been systematically studied in the cART era neuroHIV studies and abnormalities related to age and HIV infection have been found (see introduction). The caudate nucleus was selected rather than other basal ganglia nuclei based on a review by Paul et al. [15] which has identified that the caudate nuclei are a primary target of HIV infection. The posterior cingulate cortex has been systematically used in normal aging and pathological aging studies (see introduction).

Rationale for unsuppressed water signal (H 2 O) as the main reference
We used the unsuppressed water signal (H 2 O) as the main reference rather than Creatine. There is now cumulative evidence that Creatine is not constant across brain region including in nonpathological state such as normal aging for example. The use of H 2 O as the reference marker is now advocated as an alternative except when water variations are expected in certain clinical 7 conditions. To gain specific understanding on the use of H 2 O as the reference and the associated MRS protocols we recommend the textbook by Barker, et al. [16].

File S3
Tonometry examination procedure HIV+ participants underwent non-invasive tonometry to estimate arterial stiffness of the radial artery. Pulse pressure was used as research indicated that it is a better predictor of cardiac ischemic events than systolic, diastolic, or mean brachial pressures [17]. The evaluation was performed in duplicate by a single operator at the left radial artery using a highly sensitive transducer (SphygmoCor, AtCor Medical, Sydney). An augmentation index was calculated by dividing the difference between the second systolic peak and the diastolic pressure by the difference between the first systolic peak and the diastolic pressure and expressed as a percentage. Systolic blood pressure was also recorded at this time.

Extra Results section Neurocognitive and mood and functional characteristics in study groups
The HIV+ individuals as a group showed a significantly greater degree of overall neurocognitive impairment (Table S2). We found that 23% of HIV+ participants had clinically relevant (GDS≥0.5) level of neuropsychological (NP) deficits or HAND. Cognitive complaints were significantly more frequent in the HIV+ group than the HIV-group. While the HIV+ group also had a mild level of depressive complaints, the HIV+ individuals who reported clinically significant levels of depressive complaints (16%; BDI-II total score > 17) were no more likely to be NP impaired than those who did not (p>.27). Almost a quarter of the HIV+ cohort selfreported a decrease in independence in activities of daily living. Furthermore, those who reported clinically significant decline activities of daily living were more likely to have HAND as 1. IADL significant change: Independence in Activities of Daily Living: self-report of at least 2 or more decrease in the capacity to perform IADL. The HAND criteria definition follows the 43 nomenclature and uses the IADL cut-off as previously defined to discriminate between ANI and MND. Note that the level of impairment in controls according to this nomenclature is close to be expected 15% when at least five cognitive domains have been measured. Also note that the GDS: Global Deficit Score dichotomous score and the HAND criteria provides same level of impairment in the HIV+ group, but not in the HIV-group.
2. PAOFI: Personal Assessment of Own Functioning Inventory total score 3. BDI-II: Beck Depression Inventory-II total score. 4. BDI-II clinically significant: A cut off of >17 was used to define clinically significant levels of depression.
Note that there was no significant correlation between the CVD scores and the global neuropsychological performance.