Despite advances in the management of HIV infection with the introduction of combination antiretroviral therapy (cART), it is well known that HIV can directly infect the central nervous system (CNS) and, as a result neuropsychological impairments can be manifested. However, in literature there are contrasting results on which cognitive functions are mainly affected, especially when different HIV seropositive populations are considered. In this study, we seek to determine whether seropositivity is associated with a poor neuropsychological performance in patients infected postnatally, namely haemophilic patients. The results suggest that HIV infection is associated with deficits in attention, short term spatial memory, phonemic fluency, abstraction and visual recognition. Such results have important implications for day-to-day functioning, as the level of impairment detected may cause difficulties in completing common everyday tasks.
Although neuropsychological profiles vary amongst
HIV+ individuals, several estimates indicate that as many as
50% of HIV+ individuals display some degree of
neuropsychological impairment when impairment is derived
from comparisons with normative performance standards
Clifford, Franklin, Woods et al., 2010). Also, the
longitudinal observation of these patients has given
contrasting results: some studies have not shown any
decline in neurocognitive functions
In addition, some authors have argued that such
difficulties might be more related to the presence of
important covariates, such as CD4 nadir count
In our view, it is important to notice that
neuropsychological impairments have been mainly studied
in two HIV seropositive (henceforth, HIV+) populations:
- patients infected vertically or perinatally (e.g. through
breast milk), a condition that presents adjunctive cognitive
disorders linked to neuro-developmental growth changes
- patients who contracted the infection postnatally or in
adulthood, often comprises participants who presented other
types of confounding factors, such as cocaine and opiates
use, alcohol abuse
By contrast, a population of HIV patients generally
infected postnatally at an older age, and that usually do not
present the drug-user populations confounding factors,
consists of patients medically induced to HIV infection
through blood transfusions
The interest in neurocognitive dysfunctions in hemophilia
HIV+ patients is quite innovative in the literature. Findings
are often inconsistent and difficult to summarize
Although there are some data available in the context of
paediatric populations, studies are totally underrepresented
in the adult population with mixed and confusing results
The present study aimed to assess the presence and the extent of neuropsychological impairments in such a patients’ group.
Fifteen HIV+ male haemophiliacs (mean age 45+8.4) were administered neuropsychological tests. Patients were recruited through their treating physicians via the haemophilia and thrombosis outpatient clinics in three Italian centres. The inclusion criteria were as follow: diagnosis of haemophilia, diagnosis of HIV with CD4+ counts consistently > 200cells/mm3, treatment with cART, age > 18. Exclusion criteria were diagnosis of AIDS, serious mental illness with a certificated diagnosis (e.g., major depression, anxiety, bipolar disorder) or known central nervous system pathology, including progressive multifocal leukoencephalopathy, brain cancer, neurosyphilis, active cytomegalovirus infection, seizures/epilepsy, drug use. multiple sclerosis, stroke,
Tests were also administered to a control group consisting of fifteen adults (mean age 49,3+6.84), comparable for age and education (Mann–Whitney test: z=-1.41, p=.16; z=-.35, p=.77, respectively). Participants’ educational and professional characteristics are presented in Table 1.
Participants completed six neuropsychological tests from
the Italian Brief Neuropsychological Examination battery
- visual attention and cognitive processing speed (Trail Making Test A, in which participants have to connect a series of 25 number in ascending order as quickly as possible: Tombaugh, 2004);
- visual attention and executive functioning (Trail Making
Test B, in which participants have to alternate between
numbers and letters (1, A, 2, B, etc.) connecting them in
sequential order as quickly as possible:Tombaugh, 2004);
- digit span memory
- phonemic fluency
- visual recognition
Patients were met immediately after a routine medical appointment, in which –among other things- overall clinical status, including the presence of concomitant diseases, was evaluated. After signing the data protection form, they were presented with the neuropsychological tests; each of them dealt with the tests individually in a quiet room. According to each test requirements, the answers were recorded either by the patient or by the experimenter. Patients completed the tests in about twenty minutes. Controls were tested individually in a quiet room; they completed the test in about twenty minutes.
As only three test out of six have the corrected score that allows to correct raw scores for age and educational level, we decided to analyse the raw data for all the tests, also considered that age and education do not differs in the two groups.
raw data for each neuropsychological test, for patients and controls.
Results show that patients’ performance in each neuropsychological test is worse than the corresponding performance by healthy controls. More in detail, we found that patients performed significant worse than controls in Trail Making A (T-test: t= 3.53, p=.001), in Trail Making B (T-test: t= 4.88, p<.0001), in Digit Span (T-test: t=-4.02, p<.0001), in Phonemic Fluency (T-test: t= -3.19, p=.003), and in Rey Tangled Test (T-test: t=-5..836, p<.0001). Within-group comparisons. As the HIV literature reveals a correlation between cognitive disorders and patients’ sociodemographic factors such as education and job status, we performed some comparisons to figure out whether the same effect holds for our participants. We thus divided both groups into two into sub-groups according to the educational level. We considered participants with 5 to 8 education years as “low educational level”, and participants with 13 to 18 years as “high educational level”. Nine patients and 9 controls fell into the low educational level, and 6 patients and 6 controls into the high educational level. Low school level patients performed worse than the high education group only in Trial Making B Test (MannWhitney Test: z=-2.125, p=.036).
We then divided patients according to their occupational status: employed individuals on one side (n=9) versus unemployed or with disability pension individuals (n=6). We found that the unemployed/disability pension group has a worse performance than the employed group only in Trial Making A Test (Mann-Whitney Test: z=-2.239; p=.026).
As these results were inconsistent with most of the existing literature, in addition we performed an ANCOVA analysis, controlling for the effect of age, education and working status. The analysis controlled for such covariates revealed that patients performed worse than controls on Attention with regard to the Trail Making A Test (F=8.470, p=.005), and Trail Making Test B (F=5.811, p=.022). However, the predicted main effect of education was not significant (F=4.89, p = .066, ηp2 = .004), neither was the predicted main effect of working status (F=.595, p = .455, ηp2= .004), and nor the main effect of age (F= 1.226, p = .292, ηp2 = .004). Although the introduction of covariates reduced the distance between the two groups of participants, they did not eliminate the effect of HIV infection on neuropsychological impairment completely. Therefore the presence of these covariates (age, educational level and working status) did not explain group differences per se.
Although neuropsychological impairments are well described in some specific HIV populations, very few studies investigated the cohort of adults seropositive patients with haemophilia, especially in the cART era. Furthermore, the few existing studies on children and adolescents, and on adults before the widespread use of cARTs, often obtain inconsistent results.
Our pilot study, conducted on a small but well-controlled sample of HIV + hemophiliacs, reveals that such patients show signals of neuropsychological impairments when compared with the controls’ performances.
In particular, considering the whole group of HIV+
patients, we found an HIV detrimental effect on tests
requiring attention and rapid information processing (Trail
Making Test A, and Trail Making B for patients with low
educational level), consistently with the classic
conceptualization of HIV as a subcortical disease targeting
frontal-striatal circuits supporting these abilities
Patient’s results on these two attention tasks, on the visual recognition task (Rey Tangled Lines), and on phonemic fluency are also consistent with one of the two existing study on adult seropositive hemophiliacs, by Riedel and colleagues (1992), who found that patients are impaired in visual attention, visuoperceptual speed, and verbal memory and fluency. However, they also found that patients’ impairment was a linear relationship with the decreasing of immune functioning, whereas we found deficits also in patients with high immune functioning (as all our patients are).
Our patients’ impairment on Digit span memory is in line
with Riedel’s work, and with several previous findings
showing that deficits in memory, and in particular in short
term memory, are among the strongest cognitive complain
in HIV (non-haemophilic) patients
The HIV literature reveals a correlation between cognitive disorders and socio-demographic factors (job status and education); however, we did not find such interactions. Further studies will be needed to understand these results. Taken together, our results reveal that also HIV+ hemophiliacs are impaired in several cognitive functions, as it happens to HIV+ patients infected vertically, as well as to patients infected in adulthood. This result suggests that the cognitive impairments might be related predominantly to the direct effect of the HIV virus, and that they are relatively independent from the age of infection and from confounding factors (such as drug abuse). However, the study is limited by its design as a pilot study, and it suffers from methodological challenges, such as potential instability of the data related to the small sample size. Future studies, with larger samples, should be conducted to replicate these findings and test for potential interaction effects among variables of interest.
Furthermore, future studies are needed to compare these results with those obtained by a group of seronegative hemophiliacs, to investigate to what extent the combination of both diseases could affect the results obtained.
Another limitation of this study is the lack of data on MRI abnormalities and other neurological indicators which made it impossible to link the neuropsychological assessment results with some neurobiological correlates or other medical observations.
On the other hand, the current study is strengthened by selective inclusion criteria (such as the level of CD4 > 200, and the absence of a psychiatric disorder such as a diagnosis of depression/anxiety), that allows the evaluation of a precise and representative cohort. It considerably enriches the currently scarce literature on European samples of HIV+ haemophiliac adult patients in the cART era.
The level of neuropsychological impairment we detected have relevant implications for day-to-day patients’ functioning: for instance they may cause difficulties in completing common everyday tasks, such as maintaining adherence to complex medication regimens (both for haemophilia and HIV), as well as holding down a full-time job. Continued research into the mechanisms related to HIV seropositivity and neurocognitive dysfunction may provide targets for meaningful interventions.
The first author was supported from Novo Nordisk grant “Changing possibilities in Haemophilia”. The authors wish to thank prof. Mannuccio Mannucci for his support, and the huand the AICE (Italian Association of Haemophilia Centres) for allowing to conduct this study in three Haemophilia centres.