Definitional Criteria Working Group 1: Toward an Updated Nosology for HIVassociated Neurocognitive Disorders Prepared by Igor Grant for Working Group 1 consisting of: Desiree Byrd, Mariana Cherner, David Clifford, Igor Grant, Jeymohan Joseph, Justin McArthur, Michael Nunn, Victor Valcour, Valerie Wojna Frascati, June 13, 2005
AAN 1991 Criteria: Probable HAD 1. a. Acquired abnormality in at least two cognitive (not motor) abilities b. Cognitive dysfunction causing impairment in work or ADLs 2. At least one of: a. Motor abnormality b. Behavioral abnormality (motivation, emotional control, social behavior) 3. Sufficient consciousness to assess cognitive abilities 4. Absence of other etiology 2
AAN 1991 Criteria: Probable HAD Subtypes HAD-both: Criteria 1 & 2 met, with both motor and behavioral Sx HAD-motor: Criterion 1 met & Criterion 2 met with only motor Sx HAD-behavioral: Criterion 1 met & Criterion 2 met with only behavioral Sx 3
AAN 1991 Criteria: Probable MCMD Must have EACH of the following: 1. Cognitive/motor/behavioral abnormalities. a. Hx of at least 2 of: impaired attention/concentration; mental slowing; impaired memory; slowed movements; incoordination; personality change/irritability/lability. b. Acquired cognitive/motor abnormality verified by clinical exam or NP testing (# abnormal areas not specified). 2. Abnormalities from #1 cause mild impairment in work or activities of daily living. 3. Does not meet criteria for HAD or HIVassociated myelopathy. 4. No evidence of other etiology. 4
Some issues that emerged regarding the AAN 1991 criteria Degree of NP impairment is underspecified in AAN HAD and MCMD criteria Number of areas showing objectively documented decline underspecified in AAN MCMD criteria Ability to classify HAD with "mild" ADL decline has some overlap with MCMD diagnosis 5
HNRC Criteria for HIV neurocognitive complications: Asymptomatic Neuropsychological Impairment (ANI) 1. Performance at least 1 standard deviation below demographically corrected norms in at least 2 different cognitive areas* 2. Impairment has been present at least one month. 3. The impairment cannot be explained by comorbid conditions (e.g., substance abuse, medications, developmental disorder) 4. The impairment does not occur solely as part of a delirium (e.g., due to CNS toxoplasmosis, lymphoma, CMV) *At least 5 of the following 8 ability areas must be assessed: attention/information processing; language; abstraction/executive; complex perceptual motor; learning; recall/forgetting; motor skills; sensory If the individual with suspected ANI also satisfies criteria for a major depressive episode or substance dependence, the diagnosis of ANI should be deferred to a subsequent examination conducted at a time when the major depression has remitted or at least 1 month has elapsed following termination of dependent-substance use. From: Grant, I. (2002). The neurocognitive complications of HIV infection. In Encyclopedia of the Human Brain (V.S. Ramachandran, Ed.) San Diego, Academic Press, 2, 475-489. 6
HNRC Criteria for HIV neurocognitive complications: HIV-1 Associated Mild Neurocognitive Disorder (MND) 1. Acquired mild-moderate impairment in cognitive function documented by performance at least 1.0 standard deviation below demographically corrected norms in at least 2 different cognitive areas 2. Cognitive impairment produces at least mild interference in daily functioning (report of reduced mental acuity, inefficiency at work, home, social activities) 3. Cognitive impairment has been present at least one month. 4. Cognitive impairment does not meet criteria for delirium or dementia. 5. There is no evidence of another preexisting cause for the MND. If the individual with suspected MND also satisfies criteria for a major depressive episode or substance dependence, the diagnosis of MND should be deferred to a subsequent examination conducted at a time when the major depression has remitted or at least 1 month has elapsed following termination of dependent-substance use. *At least 5 of the following 8 ability areas must be assessed: attention/information processing; language; abstraction/executive; complex perceptual motor; learning; recall/forgetting; motor skills; sensory 7
HNRC Criteria for HIV neurocognitive complications: HIV-1 Associated Dementia (HAD) 1. Acquired moderate-severe impairment in cognitive function documented by performance below 2 standard deviations below demographically corrected norms in at least 2 different cognitive areas* 2. Cognitive impairment produces marked interference in daily functioning (work, home, social activities) 3. Cognitive impairment has been present at least one month. 4. Cognitive impairment does not meet criteria for delirium or dementia. 5. There is no evidence of another preexisting cause. If the individual with suspected MND also satisfies criteria for a major depressive episode or substance dependence, the diagnosis of MND should be deferred to a subsequent examination conducted at a time when the major depression has remitted or at least 1 month has elapsed following termination of dependent-substance use. *At least 5 of the following 8 ability areas must be assessed: attention/information processing; language; abstraction/executive; complex perceptual motor; learning; recall/forgetting; motor skills; sensory 8
Predictive validity of AAN and HNRC criteria with autopsy confirmation of HIV encephalitis based on cases (n=39) in Cherner et al Neurology 2002 AAN91 HNRC HIVE #Cases HAD HAD HIVE+ 4 HAD MND HIVE+ 9 HAD NPI HIVE+ 1 Diagnostic agreement = 21/39 = 54% Agreement on normal/abnormal dx = 31/39=79% MCMD normal HIVE+ 1 MCMD normal HIVE- 2 normal MND HIVE+ 2 normal NPI HIVE+ 2 normal NPI HIVE- 1 normal normal HIVE+ 8 normal normal HIVE- 9 AAN dx corresponds to HIVE HNRC dx corresponds to HIVE 9
Diagnostic Accuracy of AAN and HNRC Nomeclatures vis-à-vis post mortem HIVE Positive Predictive Power: Number of cases with NP impairment who have HIVE Using AAN criteria: 15/17 = 88% Using HNRC criteria: 18/19 = 95% Misses: Number of cases deemed NP Normal who have HIVE Using AAN criteria: 12/22 = 54% Using HNRC criteria: 9/20 = 45% Sensitivity: Number of cases with HIVE who have NP impairment Using AAN criteria: 15/27 = 56% Using HNRC criteria: 18/27 = 67% Specificity: Number of cases deemed NP Normal who do not have HIVE Using AAN criteria: 10/12 = 83% Using HNRC criteria: 11/12 = 92% 10
Three scenarios of diagnostic disagreement 1) Normal by AAN criteria and impaired by HNRC criteria 3 cases: Have mild cognitive impairment but no ADL decline 1 case: Has mild cognitive impairment and mild ADL decline but no selfreported complaints/ Hx of cognitive, motor, or behavioral decline 2) MCMD by AAN criteria and normal by HNRC criteria 3 cases: Have normal cognitive functioning but motor or behavioral changes, along with mild ADL decline 3) HAD by AAN criteria and MND by HNRC criteria 6 cases: Have mild cognitive impairment 4 cases: Have mild to moderate cognitive impairment 11
Diagnosis of HIV neurocognitive disorders among women in Puerto Rico based on several available scales 100% 90% 11.6% 20.3% 23.2% 23.2% 80% 70% 60% 50% 40% 30% 20% 10% 29% 42% 17.4% 33.3% 46.4% 18.8% 53.6% 18.8% 20.3% 31.1% <8 2 & 3 HAD mild CI MC/MD asymptomatic equivocal normal 0% MSK HDS AAN M-AAN Scales for HIV Dementia Source: Valerie Wojna 12
What is the course of neurocognitive impairment over time? ************* Some data from HNRC, the Hawaii Aging with HIV Cohort, and NEAD 13
Stability of NP Impairment Longitudinal data from HNRC 534 HIV+ 141 HIVat least 3 visits (total number of visits = 3722) 14
Proportion NP Impaired at baseline by HIV Status (HIV+ = 541; HIV- = 141) % Impaired at baseline 60 50 40 30 20 10 14 27 44 52 0 HIV- CDC A CDC B CDC C 15
NP Course for HIV neurocognitive states N=534 60 50 40 % 30 47 stably normal n=249 stably impaired n=60 stably improved n=95 stably declined n=24 fluctuated n=102 20 18 19 10 11 4 0 All HIV+ 16
Hawaii data: Diagnostic Transitions from Baseline to Year 1 Hawaii Aging with HIV Cohort Year 1 Diagnosis as a % of Baseline Diagnosis 100% 0% 17.07% 21.95% 60.98% Normal (N=37) 2.94% 41.18% 38.24% 17.65% NP Abnormal (N=53) 17.86% 53.57% 17.86% 10.71% MC/MD (N=39) 67.86% 28.57% 3.57% HAD (N=30) Nearly 1/3 (53/159) of participants who have at least some NP impairment do not meet MCMD or HAD criteria Approximately 45% of these NP abnormal participants progress at one year Baseline Diagnosis Normal NP Abnormal MC/MD HAD Source: Victor Valcour 17
NEAD Data: Frequency of transitions among neurological states in NEAD cohort (subjects with at least 1 transition): 55% showed 1 transition Transition pattern Frequency % Normal MCMD 49 23 Normal HIV-D 10 4.6 MCMD Normal 45 21 MCMD HIV-D 39 18 HIV-D Normal 26 12 HIV-D MCMD 48 22 Source: Justin McArthur 18
Importance of demographic adjustments in interpreting neurocognitive results Data from New York and San Diego 19
New York data: Diagnoses based on AAN vs. Clinical Ratings: Ethnic Disparities AAN CR Age = 44.2 (7.6) Education = 12.2 (2.9) Median CD4 = 191.4 ** p<0.01 % Frequency 100 90 80 70 60 50 40 30 20 10 0 ** ** ** ** Normal HIV Impaired Normal HIV Impaired Ethnic Minorities Caucasians (n=216) (n=69) Source: Desireé Byrd 20
Results of Using Different Normative Data Sets to Identify NP Impairment in African American Subjects % Impaired 80 70 60 50 40 30 33-14% 71 44-27% 38 African American (Heaton et al 1991 norms) African American (AANP 1994 norms) White (Heaton et al 1991 norms) 20 10 0 19 15 Normal Controls (n=21) (n=114) HIV+ (n=89) (n=401) New Heaton et al 2004 norms are appropriate for use with African Americans Source: Bob Heaton 21
Proportion of healthy (HIV-) Spanish-speaking subjects classified as impaired using the HRB 2004 norms vs new Spanish language norms 40 30 31 English Norms Spanish Norms % Impaired 20 15 Expected base rate 10 0 Figure Learning Test n=363 Source: Cherner et a,, INS, Feb. 2005 22
Misclassification most salient at lower levels of education/literacy (Figure Learning Test) 100% 80% English norms Spanish Norms % Impaired 60% 40% 20% Expected base rate 0% 0 1-5 6-10 11-13 14+ Education 23
Review of Findings 1. Current AAN criteria have good sensitivity and specificity for predicting future HIVE diagnosis, but positive predictive power can be enhanced by considering asymptomatic neurocognitive impairment. 2. Approximately 20% of cases with documented neurocognitive impairment do not have sufficient everyday functioning change to meet current AAN criteria. 24
Review of Findings 3. A substantial proportion of cases with HIV have bidirectional changes in presence or level neurocognitive impairment, and a revised nosology needs to recognize this. 4. Presence and degree of neurocognitive impairment should constitute the fundamental criterion for establishing diagnosis. Other criteria, e.g., motor disorders, emotional or personality changes, should be considered ancillary or corroborative, or criteria for defining disorder subtypes. 5. Determination of neurocognitive impairment should be based on appropriately normed tests and consider additional possible confounds. 25
Recommendation Working Group 1 recommends revision of diagnostic criteria for HIV-associated neurocognitive disorders taking into account the observations above. Revised criteria should be field tested and refined accordingly before final recommendations are made. 26
Additional slides 27
AAN 1991 Criteria: Probable HAD Resulting Diagnoses - ADL decline Severity of ADL decline Mild: Conspicuous decline in work performance, daily living activities, social activities, and complicated tasks, but not completely dependent. Can perform self-care. Moderate: Can t work; needs assistance with daily living activities, self care, and walking. Can communicate basic needs. Severe: Unable to perform any activities of daily living or self-care without assistance; requires continual supervision; nearly or completely mute. 28
Baseline background characteristics by HIV status Mean (sd) or % HIV+ HIVn=141 p n=534 Age 37.1 (8.7) 35.6 (9.0) ns Education 13.5 (2.4) 14.2 (2.6).001 % women 15 26.003 % white 68 68 ns CD4 count 390.1 (261.0) 895.4 (303.4).0001 Nadir CD4 255.9 (233.7) - Log Plasma RNA 4.0 (1.2) - % HAART % other ARV 32 29 - - 29
Baseline background characteristics by HIV status and baseline NP impairment Mean (sd) or % Impaired (n=199) HIV+ p HIV- p Normal (n=331) Impaired (n=20) Normal (n=121) Age 38.3 (9.0) 36.3 (8.3).008 33.8 (9.33 35.9 (8.9) ns Educ 13.5 (2.4) 13.5 (2.4) ns 14.1 (2.2) 14.3 (2.7) ns % women 16 14 ns 25 26 ns % white 62 72.04 70 69 ns CD4 count 373.7 (265.9) 400.0 (257.9) ns 837.9 (243.4) 905.4 (312.6) ns Nadir CD4 224.5 (203.8) 276.2 (249.6).05 - - Log Plasma RNA 4.0 (1.3) 4.02 (1.1) ns - - % HAART % other ARV 39 30 27 29.002 - - - - 30
Baseline background characteristics of HIV+ Ss by NP course over all available visits (minimum 3) HIV+ p Mean (sd) or % Stably Normal Stably Impaired Improved Wobbler Declined (n=249) (n=60 ) (n=95) (n=102) (n=24) Age 36.3 (8.5) 41.5 (8.7) 37.6 (8.6) 35.8 (8.3) 36.2 (7.8).0004 Education 13.6 (2.2) 13.9 (2.9) 13.6 (2.2) 12.9 (2.3) 13.0 (3.7).08 % women 15 15 17 16 8 ns % white 75 67 61 61 67.06 CD4 count 422.1 (272.1) 391.6 (264.8) 363.1 (255.7) 341.8 (236.2) 373.4 (232.6).08 Nadir CD4 287.8 (263.1) 196.0 (201.7) 235.8 (203.9) 243.6 (199.4) 240.9 (227.051) ns Log Plasma RNA 3.9 (1.2) 3.7 (1.3) 4.2 (1.2) 4.2 (1.2) 4.4 (1.03) ns % HAART 26 43 36 31 42.06 % other ARV 28 26 35 31 25 31
Stability of NP status over time in 534 HIV+ and 141 HIV- Ss (minimum 3 visits) 80 75 70 60 58 50 % 40 30 25 42 HIV- HIV+ 20 10 0 NP Stable NP Change 32
Proportion NP Impaired Over Time by HIV Status 80 70 72 60 50 % 40 30 47 HIV- HIV+ 20 10 0 3 11 Stably Impaired Stably normal 33
NP Course by Baseline NP Status, HIV+ only 100 % 90 80 70 60 50 75 48 HIV+ Impaired HIV+ WNL 40 30 20 30 22 18 10 7 0 stably normal stably impaired stably improved stably declined wobbled course 34
Transitions in neurocognitive category between first and last visit in HIV+ HNRC participants (minimum 3 visits) 100% 1 0 3 6 12 4 80% 19 35 50 60% 40% 90 68 16 25 HAD MND NPI Normal 20% 45 12 0% normal n=332 NPI=114 MND=80 HAD=8 Baseline 13 35