| |
Neurological Impairment Persists Despite HAART: "adjunctive therapies needed"
|
| |
| |
"reversible NP deficits were observed in 35 (37.2%) of 94 patients, whereas persistent NP deficits were documented in 59 (62.8%) of 94 patients."
JAIDS Journal of Acquired Immune Deficiency Syndromes:
1 June 2007
from Jules: like bone disease 2 years ago inadequate attention is being paid to neurologic, brain and cognitive impairment issues. There are a number of possible therapeutic interventions in HIV that are NOT being researched, as far as I know, in the ACTG nor NIH studies. HAART has NOT resolved neurologic problems. With the aging HIV patient population this could emerge into a devastating problem as studies suggest increased risk for alzheimers, neurologically related disabilities, increased mortality, and demand for greater care and services. We need a proportional response to the level of problem we face. HCV coinfection may also add to the factors causing brain dysfunction and neurologic impairment in HIV. This past month the IDSA published the first HIV bone treatment guidelines, although the subject needs more attention it's a start, and 2 additional bone guidelines are expected soon.
"adjunctive therapies are urgently needed in patients with persistent NP deficits despite HAART."
"sensitive markers to guide ARV therapy in patients with HIV-related NCI are urgently needed. Various laboratory markers of neuronal injury such as plasma HIV DNA level, CSF HIV RNA level,13,24 monocyte chemoattractant protein-1, and tumor necrosis factor-α26 have been proposed, but they have not been validated in HAART-treated patients"
-- "pathologic changes in Alzheimer disease may be present decades before clinical diagnosis of dementia is made and that sensitive cognitive measures allow identification of people with preclinical Alzheimer disease"......The association of lower education with persistent deficits supports the concept that lower educational attainment may increase the clinical expression of HIV-D(dementia)"
-- "These findings indicate that patients with more severe memory impairment and executive dysfunction are more likely to remain neurocognitively impaired despite HAART"...."because of less effective use of potent but complex current ARV regimens"
--"The association between the severity of NCI before HAART initiation and the persistence of NP deficits despite HAART makes a strong argument for the initiation of HAART as soon as NCI is diagnosed. As such, diagnostic studies for early detection of NP dysfunction should be considered in routine clinical practice in untreated patients."
"The results of this study extend previous observations indicating that current HAART regimens are inadequate to treat HIV-related NCI.....On a Kaplan-Meier estimate, the overall 65-month probability of showing persistent NP deficits despite HAART was 53.0%....Although a reversal of NP deficits was observed in more than one third of subjects, nearly two thirds of patients showed persistent NP deficits despite more than 5 years of HAART. ours is among the first study to provide an estimate of the probability of remaining neurocognitively impaired despite long-term HAART in patients with HIV-related NCI.....our findings provide further evidence that current standard HAART regimens, as prescribed in routine clinical practice, are not adequate to treat HIV-related NCI. . Although HAART can reverse HIV-related NCI,8-11 several epidemiologic and observational studies have consistently shown that HIV-related NCI continues to be detected with an almost unchanged21 or even increased15,22,23 prevalence despite the widespread availability of HAART
...the criteria for combination ARV regimens for patients with HIV-related brain disease still rely almost exclusively on expert opinion. Thus, our study findings strongly support the concept that targeted ARV treatment strategies for patients with HIV-related NCI should be carefully evaluated in prospective clinical studies...
...clinical and laboratory markers routinely used to assess the efficacy of ARV treatment and disease progression are of no use in predicting the evolution of neurologic disease...
....Our findings are consistent with previous observations demonstrating that NP improvement with HAART is not related to age, gender, disease stage, CD4 cell count, or virologic response to HAART with achievement of undetectable plasma viral load...
....The lack of association between reversible NP deficits and virologic and immunologic indicators, although apparently counterintuitive, might indicate the presence of a nonreversible component of neuronal injury possibly related to neuronal death. An alternative explanation might be that in some patients, the neuronal damage continues despite the success of ARV therapy. Data indicating the presence of continuing intrathecal immunoactivation despite 2 years of effective HAART with an HIV RNA level less than the detection limit in plasma and CSF25 are consistent with this hypothesis...
....We found an association between lower education and persistent NP deficits....alzheimers and education are associated...."
Persistence of Neuropsychologic Deficits Despite Long-Term Highly Active Antiretroviral Therapy in Patients With HIV-Related Neurocognitive Impairment: Prevalence and Risk Factors
JAIDS Journal of Acquired Immune Deficiency Syndromes:
1 June 2007
Tozzi, Valerio MD; Balestra, Pietro PsyD; Bellagamba, Rita MD; Corpolongo, Angela MD; Salvatori, Maria Flora DSc; Visco-Comandini, Ubaldo PhD; Vlassi, Chrysoula MD; Giulianelli, Marinella PsyD; Galgani, Simonetta MD; Antinori, Andrea MD; Narciso, Pasquale MD
From the *National Institute for Infectious Diseases IRCCS Lazzaro Spallanzani, Rome, Italy; and Department of Neuroscience, San Camillo Hospital, Rome, Italy.
Abstract
Objective: Although highly active antiretroviral therapy (HAART) can reverse HIV-related neurocognitive impairment (NCI), neuropsychologic (NP) deficits may persist in a substantial proportion of patients despite antiretroviral treatment. We assessed the prevalence and predictors of persistent NP deficits despite long-term HAART in patients with HIV-related NCI.
Methods: A group of 94 patients with HIV-related NCI underwent 2 to 7 serial NP batteries, neurologic examination, and brain imaging studies. Patients received HAART for a mean of 63 (range: 6-127) months. According to NP assessment results, patients were considered to have reversible or persistent NP deficits. Kaplan-Meier analyses and Cox proportional hazards models were used to analyze time to first evidence of NP deficit reversion.
Results: Persistent NP deficits were observed in 59 (62.8%) patients. Age, gender, Centers for Disease Control and Prevention stage, risk category, CD4+ cell count, plasma viral load, and use of central nervous system-penetrating drugs were not associated with persistent NP deficits. By contrast, patients with persistent NP deficits were less educated and showed poorer baseline performances in NP measures exploring concentration and speed of mental processing, memory, and mental flexibility. In multivariable analyses, only the baseline severity of NCI, as measured by the composite NPZ8 global score (odds ratio = 3.07, 95% confidence interval: 1.54 to 6.08; P = 0.001) remained significantly associated with persistent NP deficits.
Conclusions: The severity of NCI at HAART initiation seems to be the strongest predictor of persistent NP deficits despite long-term HAART. Our data indicate that HAART should be initiated as soon as NCI is diagnosed to avoid potentially irreversible neurologic damage.
HIV-related neurocognitive impairment (NCI) is regarded as a direct manifestation of HIV-1 infection of the central nervous system (CNS). It is characterized by cognitive slowing, impairment of memory and attention, disturbances in processing speed and fine motor functions, and behavioral manifestations.1,2 HIV-related NCI is a debilitating disorder that negatively affects a patient's ability to perform activities of daily living,3 quality of life,4 medication adherence,5 and survival.6
Neuropsychologically, there is a characteristic impairment in psychomotor speed, memory, and sustained attention. Thus, standardized neuropsychologic (NP) test batteries have been routinely included as primary outcome criteria to monitor neurologic response to antiretroviral (ARV) treatment in all published trials.7,8 The optimal treatment for HIV-related NCI has not been established, but there is strong evidence that highly active antiretroviral therapy (HAART) can improve the cognitive dysfunction.9,10 Clinical observations are supported by the demonstrations of the efficacy of various combination regimens in suppressing cerebrospinal fluid (CSF) HIV RNA levels11,12 and by the correlation between CSF virologic suppression and improvements on NP testing.13
Despite the overall benefit for cognitive function, the long-term efficacy of HAART in treating HIV-related NCI remains to be determined. First of all, the prevalence of HIV-related NCI seems to be only marginally affected by the widespread use of potent ARV treatment.14,15 Second, although HAART can reverse NCI, the benefits vary substantially among individuals,13 and slowly progressive forms of neurologic and cognitive dysfunction have been described in patients receiving HAART.16 Finally, neuropathologic studies have consistently shown that pathologic findings consistent with the most severe form of NCI, HIV-associated dementia (HIV-D), can still be detected in 25% to 60% of autopsies.17,18
Although current HAART regimens are not adequate to control HIV-related NCI, the prevalence and risk factors for persistence of NP deficits despite long-term HAART are largely unknown.
For these reasons, we have conducted a prospective observational study to examine the clinical course of HIV-related NCI in patients treated with HAART. Our objectives were to describe the prevalence of and risk factors for persistent NP deficits despite long-term HAART in these patients.
DISCUSSION
In this study, we examined the prevalence of and risk factors for persistent NP deficits in a complete clinic population of patients with HIV-related NCI treated with HAART. We found that 62.8% of patients persistently showed abnormal results on serial NP testing. On a Kaplan-Meier estimate, the overall 65-month probability of showing persistent NP deficits despite HAART was 53.0%. We found an association of persistent NP deficits with lower education and with greater impairment at baseline in different cognitive abilities such as concentration and speed of mental processing, memory, and mental flexibility. On a Cox model, the baseline severity of the cognitive impairment, as measured by the composite NPZ8 global score, was the strongest predictor of persistent NP deficits despite HAART.
The results of this study extend previous observations indicating that current HAART regimens are inadequate to treat HIV-related NCI. Although a reversal of NP deficits was observed in more than one third of subjects, nearly two thirds of patients showed persistent NP deficits despite more than 5 years of HAART. To our knowledge, ours is among the first study to provide an estimate of the probability of remaining neurocognitively impaired despite long-term HAART in patients with HIV-related NCI. Moreover, our findings provide further evidence that current standard HAART regimens, as prescribed in routine clinical practice, are not adequate to treat HIV-related NCI. Although HAART can reverse HIV-related NCI,8-11 several epidemiologic and observational studies have consistently shown that HIV-related NCI continues to be detected with an almost unchanged21 or even increased15,22,23 prevalence despite the widespread availability of HAART. Unfortunately, only 1 randomized clinical trial comparing different ARV combinations in patients with HIV-related NCI has been published to date,8 and the criteria for combination ARV regimens for patients with HIV-related brain disease still rely almost exclusively on expert opinion. Thus, our study findings strongly support the concept that targeted ARV treatment strategies for patients with HIV-related NCI should be carefully evaluated in prospective clinical studies.
Analyzing factors associated with the persistence of NP deficits despite HAART, we found that CD4 cell count and plasma viral load at baseline and over time, virologic response to HAART, HIV disease stage, age, CDC stage, and risk category were not associated with persistent NP deficits. Thus, clinical and laboratory markers routinely used to assess the efficacy of ARV treatment and disease progression are of no use in predicting the evolution of neurologic disease. Our findings are consistent with previous observations demonstrating that NP improvement with HAART is not related to age, gender, disease stage, CD4 cell count, or virologic response to HAART with achievement of undetectable plasma viral load.13,24 The lack of association between reversible NP deficits and virologic and immunologic indicators, although apparently counterintuitive, might indicate the presence of a nonreversible component of neuronal injury possibly related to neuronal death. An alternative explanation might be that in some patients, the neuronal damage continues despite the success of ARV therapy. Data indicating the presence of continuing intrathecal immunoactivation despite 2 years of effective HAART with an HIV RNA level less than the detection limit in plasma and CSF25 are consistent with this hypothesis. Various laboratory markers of neuronal injury such as plasma HIV DNA level, CSF HIV RNA level,13,24 monocyte chemoattractant protein-1, and tumor necrosis factor-α26 have been proposed, but they have not been validated in HAART-treated patients. Thus, sensitive markers to guide ARV therapy in patients with HIV-related NCI are urgently needed.
Our study also failed to detect an association between the use of CNS-penetrating drugs and NP outcome. For the purpose of this analysis, we have assessed a patient's exposure to ARV drugs and the number of ARV drugs known from the literature27 to have good CNS penetration in combination ARV regimens. Nevertheless, it must be considered that varying definitions of CNS penetration have been used so far and that ARV drug levels in brain tissue are largely unknown. Moreover, because of toxicity or virologic failure, our study patients experienced a mean of 4.6 changes in their combination ARV regimen. Thus, it was not possible to investigate the association between any specific ARV drug combination and NP testing results. Therefore, our negative findings must be considered with caution.
We found an association between lower education and persistent NP deficits. Education has a powerful effect in reducing the severity of cognitive impairment in other dementing illness, such as Alzheimer disease. It has been shown that pathologic changes in Alzheimer disease may be present decades before clinical diagnosis of dementia is made and that sensitive cognitive measures allow identification of people with preclinical Alzheimer disease. Subjects with more education have a greater cognitive reserve that allows them to mask their cognitive deficits.28 The association of lower education with persistent deficits supports the concept that lower educational attainment may increase the clinical expression of HIV-D.
Patients with persistent NP deficits showed greater baseline impairment in different cognitive abilities such as concentration and speed of mental processing, memory, and mental flexibility. By contrast, although motor dysfunction is a characteristic finding of HIV-related NCI, impairment in NP domains exploring fine motor functions were not associated with the persistence of NP deficits. These findings indicate that patients with more severe memory impairment and executive dysfunction are more likely to remain neurocognitively impaired despite HAART. Patients with memory problems and attention deficits may show persistent NP deficits despite HAART because of less effective use of potent but complex current ARV regimens. The most likely explanation for our findings is that poor medication adherence could be the driving factor behind this relation. Recent observations showed an association between NCI and reduced adherence to HAART. A cross-sectional study by Hinkin and colleagues5 demonstrated that cognitively compromised patients are at increased risk for poor medication adherence, especially with a complex dosing regimen requiring 3 doses of ARV drugs per day. Moreover, among different neurocognitive domains, memory impairment seems to be most strongly associated with adherence failure.29 Our findings are consistent with the hypothesis that memory impairment and executive dysfunction could be associated with the persistence of NP deficits despite HAART because they have a negative impact on adherence.
Using a Cox model, the severity of NCI before HAART initiation, as measured by the composite NPZ8 global score, was most strongly associated with persistent NP deficits despite long-term HAART. The OR in the multivariable analysis indicates that for a point decrease in the baseline NPZ8 global score, there was a 3 times greater probability of remaining impaired despite HAART. The pattern of NCI, as assessed by the profile of cognitive domains judged as abnormal at baseline, was similar in patients with persistent or reversible NCI. The only significant difference between the 2 groups was the severity of the impairment at baseline. Our data indicate that patients with reversible NCI were much closer to normal performance at baseline. The association between the severity of NCI before HAART initiation and the persistence of NP deficits despite HAART makes a strong argument for the initiation of HAART as soon as NCI is diagnosed. As such, diagnostic studies for early detection of NP dysfunction should be considered in routine clinical practice in untreated patients.
Our study has several limitations. First of all, medication adherence was not assessed. Thus, our hypothesis on the potential mechanisms of the persistence of NP deficits despite HAART in patients with memory problems and attention impairment remains a matter of speculation. Second, CSF studies were not performed. Thus, the possibility of ongoing HIV replication within the CNS in patients with an undetectable plasma viral load and persistent NP deficits must be taken into account. Finally, these results are from a single center, and more studies are needed to support our results. Although in terms of demographics, access to treatment, and ARV drug availability, our center is similar to other Italian centers, our results may not be generalized to patients treated in other counties.
Our study may have implications for the optimal management of HIV-infected patients. First, our findings highlight the need for improving the efficacy of HAART in treating HIV-related brain disease. Second, it seems that neurocognitive functions should be monitored in untreated HIV-positive patients and that HAART should be started as soon as HIV-related NCI is diagnosed to avoid potentially irreversible neurologic damage. Finally, adjunctive therapies are urgently needed in patients with persistent NP deficits despite HAART.
RESULTS
Patient Population
A total of 309 consecutive patients with symptoms of suspected NCI or with symptomatic HIV infection or a CD4 count less than 200 cells/µL were screened. Patients with current or past opportunistic infections or tumors of the CNS (n = 42), those with non-HIV-related major neurologic or psychiatric disorders (n = 10), and those using illicit drugs or sedative hypnotics (n = 68) were excluded. Patients with normal NP evaluation results (n = 95) were also excluded. The remaining 94 NP-impaired patients represent this study group.
Baseline Patient Characteristics
The baseline characteristics of the 94 study patients are shown in Tables 1 and 2. Approximately one fifth of the patients were female, 40% reported previous intravenous drug use as an HIV transmission modality, and 45% were already diagnosed with AIDS at the time of enrollment. Sixty-two (66.0%) patients were ARV naive, whereas 32 (34.0%) had received ARV therapy for 56.4 months (mean, SD = 27.8 months). Twenty AIDS patients were ARV naive at baseline because they were late presenters. They were found to be HIV-positive, together with the diagnosis of an AIDS-defining condition (HIV-D [n = 13], Kaposi sarcoma [n = 5], and Candida esophagitis [n = 2]). ARV-naive and ARV-experienced patients did not differ in terms of age, education, gender, clinical stage, HCV-positive serology, nadir CD4 cell count, or peak plasma viral load. As expected, at enrollment, ARV-experienced patients had a higher CD4 cell count (391 vs. 186 cells/µL; P < 0.001), lower plasma viral load (3.04 vs. 4.67 log copies/mL; P < 0.001), and longer time since first documented positive HIV test result (98.7 vs. 56.9 months; P = 0.002). ARV-experienced patients were older and reported a higher prevalence of previous intravenous drug use as an HIV transmission modality, but these differences were not statistically significant (see Table 1). Finally, the NP profile did not differ between the 2 groups (see Table 2).
Prevalence of Persistent Neuropsychologic Deficits Despite HAART
The mean duration of the follow-up period was 63 months (mean, range: 6-124 months). Overall, reversible NP deficits were observed in 35 (37.2%) of 94 patients, whereas persistent NP deficits were documented in 59 (62.8%) of 94 patients. All patients with reversible NP deficits remained unimpaired during the subsequent visits. All patients were seen at our clinic at least 6 months before data collection for statistical analyses. By means of a Kaplan-Meier estimate, the overall 65-month probability of showing persistent NP deficits despite HAART was 53.0% (Fig. 1).
Demographic and Clinical Variables Associated With Persistent Neuropsychologic Deficits
Demographic, clinical, and HIV illness variables associated with the persistence of NP deficits are reported in Table 3. Age, risk category, and duration of follow-up did not differ between patients with persistent or reversible NP deficits. Female patients were more likely to have persistent NP deficits (23.7% vs. 8.6% reversible), but this difference was not significant. By contrast, patients with persistent NP deficits were less educated (9.9 vs. 12.5 years; P = 0.002) and had a higher prevalence of HCV-positive serology (55.9% vs. 28.6%; P = 0.008). Notably, Centers for Disease Control and Prevention (CDC) stage, baseline CD4 cell count, baseline viral load, virologic response to HAART, and changes in viral load and in CD4 cell count during HAART did not differ between the 2 groups. Similarly, by means of a log rank test, the estimated 65-month probabilities of showing persistent NP deficits in patients with ≥10 (n = 47) and with <10 (n = 47) years of education were 47% and 59%, respectively (P = 0.0356). By contrast, age (older than vs. younger than the median), gender, previous drug use as a risk category (yes vs. no), disease stage (AIDS vs. non-AIDS), calendar year at enrollment (1996-1999 vs. 2000-2004), HCV serology (positive vs. negative), CD4 cell count at baseline and follow-up (greater than vs. less than the median), plasma viral load at baseline (greater than vs. less than the median), virologic response to HAART (failure vs. suppression), and months of exposure to a neuroactive HAART scheme (greater than vs. less than the median) were not significant.
Antiretroviral Treatment
As the initial HAART scheme, 51 (54.3%) patients received a PI-based regimen, 38 (40.4%) patients received a nonnucleoside reverse transcriptase inhibitor (NNRTI)-based regimen, and 5 (5.3%) patients received a triple-nucleoside reverse transcriptase inhibitor (NRTI) regimen. During follow-up, patients experienced 4.6 (mean, range: 1-9) changes in their HAART regimen because of toxicity or virologic failure. To investigate the association between ARV drug exposure and persistent NP deficits, we have assessed patients' exposure to each ARV drug and patients' exposure to HAART schemes that included at least 2 CNS-penetrating drugs. For the purpose of this analysis, stavudine, zidovudine, abacavir, lamivudine, efavirenz, nevirapine, and indinavir were considered CNS-penetrating drugs. Although patients with persistent NP deficits had an inferior exposure to abacavir, nevirapine, and efavirenz, the differences did not reach statistical significance. Finally, the number of CNS-penetrating drugs in the HAART scheme did not differ between patients with persistent or reversible NP deficits. Similar results were obtained when lopinavir/ritonavir was considered as a CNS-penetrating drug (data not shown in tables).
Neuropsychologic Variables Associated With Persistent Neuropsychologic Deficits
The relations between NP test results and persistence of NP deficits are shown in Table 4. A number of baseline NP test z scores were significantly associated with the persistence of NP deficits. Patients with persistent NP deficits showed a poorer performance on the composite concentration and speed of mental processing z score (-1.37 vs. -0.81; P = 0.003) and on the composite memory z score (-1.96 vs. -1.39; P = 0.022). Analyzing specific NP measures, a correlation between 4 of 6 measures exploring concentration and speed of mental processing (Wechsler Adult Intelligence Scale-Revised [WAIS-R] Digit Span forward: -0.92 vs. 0.37, P = 0.031; WAIS-R Digit Span backward: -1.43 vs. -0.75, P = 0.008; Stroop Word and Color: -1.41 vs. -0.37, P = 0.010; and Corsi Cube test: -1.39 vs. -0.40, P < 0.001) and 1 of 3 measures exploring mental flexibility (Trail-Making B: -4.41 vs. -2.51; P = 0.018) and persistent impairment was found. Finally, patients with persistent NP deficits despite HAART showed less-pronounced improvement in mean composite NPZ4 (0.35 vs. 1.56; P = 0.016) and NPZ8 (0.09 vs. 1.08; P = 0.010) global scores between baseline and the first visit after initiation of HAART. Notably, the 2 groups did not differ in any of the NP measures exploring fine motor functions. Similarly, on the log rank test, a number of baseline NP test z scores and composite z scores were significantly associated with persistent NP deficits. The estimated 65-month probabilities of showing persistent NP deficits in patients with composite NPZ4 scores greater than and less than the median of -1.19 were 29% and 79%, respectively (P < 0.0001), and in patients with composite NPZ8 scores greater than and less than the median of -1.23, they were 30% and 87%, respectively (P < 0.0001). Moreover, the estimated probability of showing persistent NP deficits was significantly higher in patients with poorer performance (z scores) on the composite concentration and speed of mental processing test (79% vs. 34%; P < 0.0001), composite memory test (81% vs. 39%; P = 0.0013), WAIS-R Digit Span backward test (76% vs. 31%; P = 0.0001), Corsi Cube test (60% vs. 39%; P = 0.0014), Trail Making B test (81% vs. 37%; P = 0.0005), Trail Making A test (79% vs. 30%; P < 0.0001), Rey Auditory Verbal Learning test (78% vs. 45%; P = 0.0034), Rey Auditory Verbal Learning test after 15 minutes (88% vs. 39%; P = 0.0009), WAIS-R Digit Symbol test (69% vs. 39%; P = 0.0009), and Rey Complex Figure (delayed) test (88% vs. 25%; <0.0001) (data not shown in tables).
Finally, the pattern of NCI, as assessed by the profile of cognitive domains judged as abnormal at baseline, was similar in patients with persistent and reversible NCI.
Factors Associated With the Persistence of Neuropsychologic Deficits Despite HAART Using a Multivariable Cox Model
In a multivariable Cox model, after controlling for gender, years of education, HCV-positive serology, and CD4 cell count at last visit, among the NP test battery measures and composite scores, poor performance on the baseline composite NPZ8 global score was most strongly associated with the persistence of NP deficits despite HAART (odds ratio [OR] = 3.07, 95% confidence interval [CI]: 1.54 to 6.08; P = 0.001; Table 5). The OR in the multivariable analyses indicates that for a point decrease in the baseline NPZ8 score, there was a 3 times greater probability of remaining impaired despite long-term HAART.
METHODS
Study Design, Setting, and Patient Population
This was a cohort study of patients with HIV-related NCI. Between June 1996 and December 2004, HIV-infected patients followed at the National Institute for Infectious Diseases Lazzaro Spallanzani (Rome, Italy) were screened for the presence of HIV-related NCI. The study was conducted in the context of usual clinical care according to an internal protocol approved by the local Ethics Committee.
Patients were eligible if they had symptoms of suspected NCI or had symptomatic HIV infection or a CD4 count less than 200 cells/µL. Exclusion criteria were current or past opportunistic infections or tumors of the CNS, non-HIV-related major neurologic or psychiatric disorders, and current use of illicit drugs or sedative hypnotics. After the exclusion of potentially confounding clinical conditions that could affect cognition through brain magnetic resonance imaging (MRI) studies, neurologic examinations, and laboratory testing, patients were referred for NP evaluation. Patients with normal results on NP evaluation were excluded. All patients with impaired NP performance were enrolled into the study. The day of completing the NP examination was considered as the day of enrollment into the study. After enrollment, patients were prospectively evaluated, and they represent this study cohort.
Neuropsychologic Testing
A detailed NP assessment was performed through a battery of 15 standardized NP tests administered by a trained neuropsychologist (PB). Tests were selected to include a wide spectrum of different cognitive domains: mental flexibility, concentration and speed of mental processing, memory, visuospatial and constructional abilities, and fine motor functions. The score for each test was adjusted for age, gender, and years of education.19,20 The NP scores resulting from each test were transformed into z scores as described elsewhere.21 Moreover, the following composite z scores were obtained: global NPZ8 score, global NPZ4 score, memory z score, concentration and speed of mental processing z score, mental flexibility z score, and fine motor functioning z score. Each z score was adjusted so that negative values indicated below-average performances.
Neurologic Examination and Diagnostic Criteria for Neurocognitive Impairment
Patients were classified as having or not having NCI based on their NP performance relative to the normative data. NCI was defined as performing below 1 SD from the normative mean on at least 2 NP tests or 2 SDs below the mean on at least 1 test. The diagnosis of NCI also required the exclusion, by neurologic examination, laboratory measures, and brain imaging studies, of other conditions that could explain the finding. The American Academy of Neurology (AAN) criteria2 were used to determine whether the degree of impairment met the criteria for HIV-D or for NCI without dementia.
Serial Neuropsychologic Examination and Criteria for Reversible Neuropsychologic Deficits
NP assessments were scheduled every 12 months and were repeated 2.7 times (mean, range: 1-6 times) during a follow-up period of 63 months (mean, range: 6-124 months). NP testing was considered as the primary outcome criterion. The NP testing and neurologic examination were used to identify subjects with reversible or persistent NP deficits. Reversible NP deficits were defined as an NP test performance within the normal range relative to the normative data in a previously impaired patient. By contrast, persistent NP deficits were defined as a performance below 1 SD from the normative mean on at least 2 NP tests or a performance 2 SD below the mean on at least 1 test. The diagnosis of persistent NP deficits required the exclusion by neurologic examination, laboratory measures, and brain imaging studies of other conditions that could explain the finding. The main end point was NP deficit reversion.
Antiretroviral Treatment and Virologic Response to Therapy
HAART regimens were prescribed as best judged by the primary medical care provider. The quantification of HIV-1 RNA was performed between 1996 and 1999 using the NASBA kit (Organon Teknika, Boxtel, The Netherlands) with a lower limit of detection of 80 copies/mL and between 2000 and the present using the Versant HIV RNA version 3.0 test (Bayer Diagnostics, Emeryville, CA) with a lower limit of detection of 50 copies/mL. Response to HAART was defined as durable suppression if the patient achieved and maintained a plasma viral load less than the detection limit or as virologic failure if the patient failed to achieve an undetectable viral load or experienced a confirmed virologic rebound to more then 1000 copies/mL after having achieved an undetectable viral load. Patients had a mean of 4 (range: 3-5) plasma viral load determinations each year. Moreover, viral load measurements were obtained concomitant with the times of NP assessments. Serial CSF HIV RNA data were available for 3 patients only and were not included in the present analyses.
Data Analysis and Statistical Methods
We evaluated the probability of remaining neurocognitively impaired despite HAART. The association between persistent NP deficits and selected demographic (age, gender, and years of education), clinical (HIV disease stage, hepatitis C virus [HCV] serology, virologic response to HAART, ARV treatment, and exposure to CSF-penetrating drugs), laboratory (CD4 cell count at enrollment and during follow-up and plasma HIV RNA level at enrollment and during follow-up), and NP (NP test z scores at enrollment and follow-up) characteristics was assessed by means of the χ2 test, or, when appropriate, the Student t test.
The Kaplan-Meier method was used to estimate the probability of showing persistent NP deficits despite HAART. The time to the first follow-up visit at which the patient was classified as not having NCI was computed. The log rank test was used to assess statistically significant differences between groups. The association between time to first reversion and the baseline severity of NCI was assessed by splitting the population sample at the median value of selected NP test z scores. A multivariable analysis was performed using Cox proportional hazard models, in which factors found significantly associated with persistent NP deficits at univariate analysis were included. Repeatedly measured variables were entered as time-dependent covariates. In addition, other variables, such as age, gender, CD4 cell count, and HCV-positive serology, were included in the model as correction variables.
P values were accepted as statistically significant at a 2-tailed 0.05 level. All date were analyzed using STATA statistical analysis software (version 8; Stata Corporation, College Station, TX).
|
|
| |
| |
|
|
|
