Inclusion criteria were: hyperactive, inattentive, and impulsive behaviors that were impairing in at least 2 settings and a Conners' Teacher Hyperactivity rating greater than 2 SD above age- and sex-specific means.^{22, 23} The DSM-IV diagnosis of ADHD was based on the Parent Diagnostic Interview for Children and Adolescents,²⁴ Conners' Teacher Rating Scales,^{22, 23} and the Teacher Report Form.²⁵ A clinical psychologist administered the Wechsler Intelligence Scale for Children-Revised²⁶ to 110 patients with ADHD and the Wechsler Intelligence Scale for Children-III²⁷ to 41 patients (1 was too young to be tested). Exclusion criteria were a full-scale IQ of less than 80, evidence of medical or neurological disorders on examination or by clinical history, Tourette disorder, or any other axis I psychiatric disorder requiring treatment with medication at study entry.

[Fred A. Baughman Jr., MD:No matter the multiplicity of measures, not one is an objective, physical measure; no one or several positive results on such wholly subjective tests or scales equates to, suggests, or implies, organic disease of the brain.  Recall that "biological" psychiatry has yet to offer us their first biological proof.]

A total of 56 unrelated healthy female (mean initial age, 10.0 years; range, 5.2-16.1) and 83 male (mean initial age, 10.9; range, 4.5-19.0) controls were recruited from the community via the National Institutes of Health Normal Volunteer Office and outreach to local schools. Screening included an initial telephone interview, parent and teacher rating scales,²⁵ in-person assessment including physical and neurological examinations including handedness,²⁸ and clinical history obtained by a child and adolescent psychiatrist (J.N.G.)

[Fred A. Baughman Jr., MD:Parent and teacher rating scales they would have us believe are tools with which to validate the presence of organic disease of the brain.  Neurological examinations were done and apparently ruled out all objective neurological abnormalities; things such as abnormal gait, paralysis of one side or the other, visual field defects, facial palsy, objective physical abnormalities, as compared to the so-called invisible disabilities of special education and of psychiatry, generally]

. Vocabulary and block design subtests from the Wechsler Intelligence Scale for Children-Revised (n = 80), Wechsler Intelligence Scale for Children-III (n = 23), Wechsler Abbreviated Scale of Intelligence²⁹ (n = 20), Wechsler Preschool and Primary Scale of Intelligence³⁰ (n = 10), and Wechsler Adult Intelligence Scale-Revised³¹ (n = 1) were obtained. Five controls were not tested but were within the healthy range by reported academic history. Approximately 4 candidates were screened for every 1 accepted,³² with the most common exclusions being positive family psychiatric history and possible psychiatric diagnosis based on teacher report.

This study was conducted at the Child Psychiatry Branch of the National Institute of Mental Health in Bethesda, Md, between 1991 and 2001. The institutional review board approved the research protocol, and written informed consent and assent to participate in a study of brain development were obtained from parents and children, respectively, at study entry and at each subsequent MRI examination.

[Fred A. Baughman Jr., MD:It would be informative to obtain and review their informed consent documents]

Healthy volunteers and patients not currently participating in treatment studies were paid to participate.

Behavioral Measures

Primary symptom severity measures were those that remained constant across the study decade using the Attention Problems Factors from the Child Behavior Checklist and Teacher Report Form²⁵ and the Clinical Global Impressions scale for Severity of Illness.³³ Medication status was obtained from parental history.

[Fred A. Baughman Jr., MD:Nothing less than a life-history of medication usage can suffice in a study such as this]

MRI Acquisition

All patients and controls were studied on the same 1.5-T General Electric Signa scanner (Milwaukee, Wis). T1-weighted images with contiguous 1.5-mm slices in the axial plane and 2.0-mm slices in the coronal plane were obtained using 3-dimensional spoiled gradient recalled echo in the steady state. Imaging parameters were echo time of 5 ms, repetition time of 24 ms, flip angle of 45°, acquisition matrix of 256 <<...OLE_Obj...>> 192, number of excitations equals 1, and 24 cm field of view. Head placement was standardized as previously described.¹⁶

Image Analysis

T2-weighted images were obtained for evaluation by a clinical neuroradiologist. All raters were blind to demographic characteristics. Quantification of MRI images was performed via a 3-part fully automated image analysis process that determines the volumes of gray and white matter compartments in frontal, temporal, parietal, and occipital lobes as well as basal ganglia and cerebellum with excellent test-retest reliability as described elsewhere in detail.^34-36

Visual inspection of each scan revealed that 544 of 594 total scans (92%) were processed successfully; 50 were excluded because of classification and segmentation errors due to motion. Failure rate was significantly higher ( <<...OLE_Obj...>> ²₁ with Yates correction = 4.08, P = .04) in 34 of 317 patients (11%) than in 16 of 277 controls (6%). All remaining scans from patients with ADHD (283 scans) were used. The comparison group was selected from a pool of healthy controls after excluding siblings in order not to violate the statistical assumption of independence. The remaining 139 potential controls (ie, no more than 1 per family within the age range of our patients) were selected by the data manager (L.S.C.) to best match each target patient for sex, age, and longitudinal intervals, prior to morphometric analyses.

[Fred A. Baughman Jr., MD:By "morphomentric" they mean anatomic configuration]

Whenever precise matching on all parameters was not possible, patients and controls were matched on average-age across their own scans. Because we were unable to match all patients and controls 1-to-1, we made every effort to maintain proportional scan-densities across the entire age-range of the 152 patients.

Statistical Analyses

Demographic and clinical measures were compared by 2-way analyses of variance (testing main effects of diagnoses and sex and their interaction) or 2-sample t tests for continuous measures, and with <<...OLE_Obj...>> ² or Fisher exact test for nominal measures. Analyses of variance of the 10 regional brain measures and 3 summary measures obtained at initial scan (n = 291 independent participants) were initially performed with diagnoses and sex as between-participant factors. Because we did not obtain full-scale IQ scores from controls, Wechsler vocabulary standard score was used, as it is the single best predictor of full-scale IQ.²⁷ To account for between-group differences in vocabulary, height, weight, handedness, and medication status, analyses of covariance were performed with these potential covariates. Nonsignificant covariates were deleted from the final models. Pearson correlations were computed for symptom severity measures and brain volumes in the patient sample.

To examine the influence of medications more closely we compared patients with ADHD who were never previously treated with psychotropic medications (unmedicated ADHD), medicated patients (medicated ADHD), and controls.

[Fred A. Baughman Jr., MD:3 groups: (1) ADHD-never on any medication; (2) ADHD-medicated ( they do not say, but should, exact doses and for how long), and (3) normal controls]

The unmedicated ADHD patients were significantly younger than the medicated ADHD and controls; thus, we confirmed findings in age-matched subgroups (n = 128). All pairwise comparisons were conducted with Bonferroni corrections.

[Fred A. Baughman Jr., MD:   The fact that the unmedicated patients (32% of all ADHD patients) were "significantly younger" means they were also smaller and possessed of smaller brain.  This we are told was handled with a particular statistical treatment]

[Fred A. Baughman Jr., MD:There were 152 with ADHD, 103 (68%) of whom were medicated and 49 (32%) not medicated.   Had the goal of their research been to demonstrate that ADHD is a disease, perhaps one characterized by brain atrophy, they would have made this the one and only aim of their research and would have constituted the largest possible group of (1) never-medicated ADHD patients, comparing them with a like number of (2) normal controls.  This would have eliminated the drugs as a physical variable.  Swanson, at the CC lamented that they had difficulty recruiting never-drugged ADHD subjects because all or most are so quickly medicated.  This excuse does not fly; thousands of new diagnoses are made in the US daily, of this their can be no doubt.  Such a study could quickly be constituted and would be the pure study, absent the drugging confounder, that is needed.  To determine  whether the drugs are causing the brain atrophy or not would require a pure study comparing (1) ADHD patients never drugged to (2) ADHD patients-drugged.   It has seemed to me that the NIMH and research biological psychiatry has, instead, avoided doing these studies, continuing instead to produce and publicize studies of drugged ADHD subjects showing brain atrophy, in turn, representing such brain atrophy  as due to ADHD; proof it exists, saying little or nothing to the  public of the almost certain cause, the encepalopathic drugs they are on.]

Finally, longitudinal analytic methods^{37, 38} were used to examine growth patterns of caudate, cerebellum, total cerebrum, and the white and gray components of the 4 major lobes. The initial full longitudinal growth model was expressed as a cubic:

Size = Intercept + <<...OLE_Obj...>> ₁ <<...OLE_Obj...>> (Age - Mean Age) + <<...OLE_Obj...>> ₂ <<...OLE_Obj...>> (Age - Mean Age)² + <<...OLE_Obj...>> ₃ <<...OLE_Obj...>> (Age - Mean Age)³ + <<...OLE_Obj...>>

The model parameters (intercept and <<...OLE_Obj...>> coefficients) were initially allowed to reflect interactions between sex and diagnostic group. To account for within-person correlations, intercepts were treated as normally distributed random effects that varied by individual, while <<...OLE_Obj...>> coefficients for age, age-squared, and age-cubed terms were modeled as fixed effects. The full cubic model was compared with simpler quadratic, linear, and constant models with interactions. Once the order of the model was established, testing was performed to determine whether an additive model could replace the interactions between sex and diagnostic group for the height and shape parameters of the curves. With respect to shape of the curves, there were neither significant sex differences nor sex by diagnosis interactions for any structure. Consequently, final models allowed for sex and diagnosis effects in the height parameters (intercept) of the curves and included only diagnostic differences in shape parameters.

[Fred A. Baughman Jr., MD:numberable statistical treatments]

Hypothesis tests and model selection were based on F statistics. We included data from individuals who had only a single scan (about 40% of both groups), because single scans provide additional information about between-participant variation and overall curve shape. These methods have been useful for combining cross-sectional and longitudinal anatomic MRI data.^39-41 Statistical power exceeded 80% at P = .05 for all brain measures. Minimally detectable adjusted differences ranged from 2.7% (caudate and cerebellum) to 5% for occipital gray matter, and averaged 3% for cortical volumes. Statistical analyses were performed using SPSS version 10.0 (SPSS Inc, Chicago, Ill), except for the mixed-model random regression analyses, which were performed with SAS version 8.02 (SAS Institute Inc, Cary, NC), and the power analyses, which were conducted with PASS 2000 (NCSS Statistical Software, Kaysville, Utah). Two-tailed significance levels were defined as P <<...OLE_Obj...>> .05.

RESULTS
<<...OLE_Obj...>>
Participants

Final study participants consisting of 152 children and adolescents with ADHD and 139 controls were each successfully scanned up to 4 times over a decade.

[Fred A. Baughman Jr., MD:In the studies I suggest above, several years of follow-up and one set of scans of all is all that would be required.]

As Table 1 <../fig_tab/joc20194_t1.html> shows, there were several group differences between male and female patients (females were younger, shorter, and weighed less), and between patients and controls. Patients were shorter and weighed less, had lower vocabulary standard scores, and a lower percentage of individuals were strongly right-handed (scoring 10 or more of 12 items). Sex and diagnosis did not interact significantly for any demographic measure. Female and male patients with ADHD were comparable on vocabulary, handedness, parent and teacher attention problem scores, and prevalence of learning disorders.⁴²

[Fred A. Baughman Jr., MD:None of the following are physical variables: vocabulary , parent and teacher attention problem scores, and prevalence of learning disorders]

.Physician's Clinical Global Impressions ratings reflected significantly greater severity in females, who also had a higher percentage of combined-type ADHD, mood disorder (history of major depression and/or dysthymia) and lower prevalence of conduct disorder and tic disorder not otherwise specified.

[Fred A. Baughman Jr., MD:Of these, only tic disorder is a physical variable; if present prior to any medication it would be called Tourette's syndrome, if present only after stimulant treatment it would be called "stimulant-induced tic syndrome/disease]

At the time of the first scan, 103 patients (68%) were being treated with psychostimulants.

The 49

[Fred A. Baughman Jr., MD:32% of all ADHD subjects]

patients with ADHD (22 females, 27 males) who were successfully scanned before ever being treated with psychotropic medications (unmedicated ADHD) were significantly younger than the medicated patients (medicated ADHD) and controls (Table 2 <../fig_tab/joc20194_t2.html>).

[Fred A. Baughman Jr., MD:Which, without wholly valid statistical adjustment would mean that the ADHD-unmedicated group, as a whole is younger than ADHD-medicated and controls, making the ADHD-unmedicated group smaller (in height and weight) and possessed, as well, of smaller brains.  They will argue, and posit, within their study that they have statistically treated for this]

.   Unmedicated patients with ADHD were rated as comparable in severity by parents, but as significantly less severely affected by physicians and teachers.

[Fred A. Baughman Jr., MD:As if ADHD severity as judged by physicians, teachers or parents equated to "brain disease" severity]

They also tended to score higher on the vocabulary IQ subtest, but not significantly (P = .06).

Sixty-one patients (40%) were scanned once, 61 (40%) twice, 20 (13%) 3 times, and 10 (7%) 4 times. Fifty-two controls (37%) were scanned once, 55 (40%) twice, 29 (21%) 3 times, and 3 (2%) 4 times. Mean ages at each scan did not differ significantly between diagnostic groups (at first scan, F_1,289 = 2.28, P = .13; at second scan, F_1,176 = 0.08, P = .78; at third scan, F_1,60 = 0.02, P = .89; at fourth scan, F_1,11 = 1.06, P = .32). Female participants (mean, 9.7 years [SD, 2.6]) were significantly younger than male participants (mean, 10.7 years [SD, 3.3]; P = .006), regardless of diagnosis. Mean intervals between scans did not differ significantly between diagnostic groups (mean for patients with ADHD, 2.6 years [1.1]; mean for controls, 2.4 years [1.0]; T₂₂₉ = 1.60; P = .11). 

[Fred A. Baughman Jr., MD:Again, had the goal of their research been to demonstrate whether or not, yes or no, ADHD is a disease characterized by brain atrophy, they would have made this the one and only aim of their research and would have constituted the largest possible group of (1) never-medicated ADHD patients, comparing them with a like number of (2) normal controls.  This would have eliminated the drugs as a physical variable.  Swanson, at the CC lamented that they had difficulty recruiting never-drugged ADHD subjects because all or most are so quickly medicated.  This excuse does not fly; thousands of new diagnoses are made in the US daily, of this their can be no doubt.  Such a study could quickly be constituted and would be the pure study, absent the drugging confounder, that is needed.  To determine  whether the drugs are causing the brain atrophy or not would require a pure study comparing (1) ADHD patients never drugged to (2) ADHD patients-drugged.   It has seemed to me that the NIMH and research biological psychiatry has, instead, avoided doing these studies, continuing instead to produce and publicize studies of drugged ADHD subjects showing brain atrophy, in turn, representing such brain atrophy  as due to ADHD; proof it exists, saying little or nothing to the  public of the almost certain cause, the encepalopathic drugs they are on.  Both of the studies I suggest above would run for 5 years with all subjects matched for sex, and age, have a single, technically adequate scan at the start of the study and again at the end.  Growth  of the brain continues, at a much slower rate than before, until 12-15 years, when the average adult weight of 1230 to 1275 g is females and 1350 to 1410 g in males is attained.  Here we see that brain size is very much a function of age, and that the brain of females is smaller, generally than males.]

Analyses of Initial Scans

Table 3 <../fig_tab/joc20194_t3.html> contains the unadjusted means (SDs) of the 291 initial cross-sectional scans by diagnosis as well as the means (SEs) adjusted for all significant covariates. Three summary measures were obtained for the cerebrum, defined by excluding cerebellum, brainstem, and cerebrospinal fluid.

As expected,^{43, 44} all measures were significantly smaller in female participants

[Fred A. Baughman Jr., MD:see above, female brain smaller than male, with this difference implying nothing about functional capacity]

(F_{1, 287} ranged from 10.65 for parietal gray matter to 98.61 for cerebellum; P<.001), but sex did not interact significantly with diagnosis for any brain anatomy measure. Accordingly, mean values for sex and corresponding statistics are not presented here (they can be found at <http://intramural.nimh.nih.gov/research/chp/index2.html>). A significant main effect of diagnosis was found between patients with ADHD and controls for all measures with small-to-medium effect sizes ranging from 0.30 to 0.46, which remained significant or were somewhat enhanced (eg, adjusted effect size for temporal white matter = 0.64) when adjusted for the significant covariates of vocabulary, height, or medication status. When we adjusted for the significant group differences in total cerebral volume, the only brain region that remained significantly smaller in ADHD was the cerebellum (d = .27; 95% confidence interval [CI], 0.03-0.50; F_1,287 = 4.97; P = .03).

Effects of Prior Drug Treatment

Table 4 <../fig_tab/joc20194_t4.html> displays the contrasts between 3 nonoverlapping groups consisting of 49

[Fred A. Baughman Jr., MD:32%]

unmedicated patients with ADHD, 103 medicated patients

[Fred A. Baughman Jr., MD:68%]

with ADHD, and 139 healthy controls. Unmedicated patients with ADHD did not differ significantly from medicated patients with ADHD on any gray matter measures, or in caudate or cerebellum. By contrast, unmedicated patients with ADHD had strikingly smaller white matter volumes (F_2,288 = 11.65) compared with controls (-10.7%, P<.001) and with medicated children with ADHD (-8.9%; P<.001; all pairwise comparisons Bonferroni corrected). Unmedicated patients with ADHD had smaller cerebellar volumes (-6.2%, P<.001), smaller temporal gray (-4.6%, P = .02), and smaller total cerebral volumes (-5.8%, P = .002) compared with controls.

[Fred A. Baughman Jr., MD:No conclusions regarding the functional significance of these findings can be drawn.  The authors, however, draw the conclusion that these and similar findings explain their ADHD behaviors; behaviors, Carey asserts, are normal behaviors, having no pathological significance]

Differences between unmedicated patients with ADHD and controls in frontal (-3.8%) and parietal gray matter (-4.1%) would also have been significant if not corrected for multiple comparisons. Medicated patients with ADHD did not differ significantly from controls on any white matter measure.

[Fred A. Baughman Jr., MD:An observation the authors take to mean that stimulant medications do no damage to the brain]

Robust differences from controls remained for all gray matter measures (ranging from -3.4% to -6.6%), caudate (-4.3%), cerebellum (-3.6%), and the summary measures of total cerebral volume (-3.3%) and total gray volume (-3.9%).

Because the unmedicated patients with ADHD were significantly younger than the other 2 subgroups, and white matter increases with increasing age throughout the age range,⁴⁵ we performed secondary analyses restricted to an age-matched subset of 128 participants (consisting of 24 unmedicated patients with ADHD, 50 medicated patients with ADHD, and 54 controls [61 females]). All measures remained essentially unchanged.

[Fred A. Baughman Jr., MD:Again the need for statistical "corrections" of the fact that the ADHD-unmedicated  group-32% of the total ADHD group were "significantly younger."]

Relationship to Clinical Measures

We examined correlations between the 10 regional measures and behavioral ratings. Within the patient group, smaller volumes were significantly correlated in the expected direction with greater symptom severity. Frontal and temporal gray matter, caudate, and cerebellar volumes were significantly and negatively correlated with physician's Clinical Global Impressions rating (n = 139, Pearson coefficients ranged between -0.16 for frontal gray and -0.26 for cerebellum, all P<.05). The same 4 regions were also significantly and negatively correlated with parent-rated child behavior checklist attention problems with Pearson coefficients between -0.16 and -0.22 (all P<.05). Correlations were largely unaffected when adjusted for age.

Wechsler vocabulary standard score was significantly and positively correlated with all anatomic volumes in patients with ADHD (n = 151; r ranged from 0.19-0.35; all P<.02), and in frontal and occipital gray and white matter and cerebellar volumes in controls (n = 134; r ranged from 0.18-0.24; all P<.02). Although the magnitude of the correlations was greater in patients than in controls, none of the coefficients differed significantly from each other, and all regional volumes correlated significantly with the vocabulary score when the 2 groups were combined (n = 285; eg, for total cerebral volume, r = 0.31; P<.001).

Analyses of Initial and Follow-up Scans

Sixty percent of all participants had at least 2 scans (n = 178), including 62 (21%), who had at least 3 scans and 13 (4%), who had 4 scans obtained at 2- to 3-year intervals. Data from all 544 resulting scans were used to derive longitudinal growth curves for patients and controls of both sexes. The age range for male participants extended between 4.6 and 19.0 years, while female participants ranged between 5.2 and 16.3 years, reflecting our initial focus on males with ADHD.²⁰

Predicted longitudinal growth curve parameters did not differ significantly between male and female participants except for the height of each curve (intercept) at the corresponding age midpoint, which were significantly higher for males for all measures, regardless of diagnosis (empirical P<.001, derived from F statistics confirmed with permutation tests with 1000 iterations). There were no significant interactions between sex and diagnosis for any developmental growth patterns (intercepts or curve parameters <<...OLE_Obj...>> _1-3). Figure 1 <../fig_tab/joc20194_f1.html> shows the predicted developmental growth curves along with 95% CIs for each group's average total cerebral volume. Developmental curves were significantly higher in controls than in patients with ADHD for total cerebral volume and for all other brain measures. Diagnostic differences in curve height remained significant after adjusting for vocabulary standard score (total cerebral volume, P = .002). There were no significant differences in curve shape between patients and controls, except for caudate. After adjustment for diagnostic differences in total cerebral volume, only caudate (P = .02) and cerebellum (P = .003) remained significantly smaller in patients with ADHD.

Figure 2 <../fig_tab/joc20194_f2.html> depicts unadjusted predicted growth curves for caudate nucleus and cerebellum. Caudate was the only region in which the developmental trajectories did not remain statistically parallel for patients and controls (adjusted, P = .05). These differences in shape represent a normalization of caudate volume for patients by midadolescence. By contrast, diagnostic differences in cerebellar curves continue throughout our age range (unadjusted, P<.001; adjusted, P = .003), with a nonsignificant tendency toward a greater difference in late adolescence (unadjusted, P = .10). The general absence of diagnostic differences in curve shapes indicates that developmental curves for patients with ADHD, although significantly lower, were essentially parallel to curves for healthy controls, with the exception of the caudate nucleus.

[Fred A. Baughman Jr., MD:To speak of diagnostic differences is misleading.  The brain/part of brain volume differences described herein are group differences.  There is no kind of brain scan that is of diagnostic value or that can ever diagnose a single case of ADHD or, for that matter any psychiatric condition.  The reason there is no objective, physical test in all of "biological psychiatry" is that not a single psychiatric condition, in 35 years of disease claims, has a proven, demonstrated, demonstrable physical abnormality...not a gene, not a chemical imbalance, nothing.  Nor does the present study ruin their pristine record]

COMMENT
<<...OLE_Obj...>>
Fully automated measures of brain cortical and subcortical volumes from the initial scans of 291 male and female patients show that the cerebrum as a whole and the cerebellum are smaller in children and adolescents with predominantly combined-type ADHD

[Fred A. Baughman Jr., MD:They have proved no such thing]

. Rather than reflecting a selective frontal-striatal effect, volumes were decreased to a comparable extent in all 4 lobes and were statistically more prominent only in the cerebellum. Our findings were not ascribable to differences in cognitive level, height, age, weight, or handedness and were not related to comorbid diagnoses (data not shown).

This is the first neuroimaging study to our knowledge to include a substantial number (n = 49) of previously unmedicated children and adolescents with ADHD.

[Fred A. Baughman Jr., MD:Why, with Nasrallah et al having suggested that stimulant medication was causing brain atrophy in 1986, has it taken research psychiatry all this time to address this question, and, at that not yet authoring or performing the appropriate study necessary to answer the two, separate, fundamental questions (1) Is ADHD an abnormality/disease-yes or no? and (2) does Ritalin/amphetamine treatment cause brain atrophy?]

We attempted to recruit children with equivalent severity of ADHD symptoms by using identical diagnostic and symptom severity criteria. Unmedicated patients with ADHD did not differ from medicated children with ADHD on parent-rated attention problems, but they had significantly lower teacher and physician ratings, and higher vocabulary standard scores.

[Fred A. Baughman Jr., MD:as if objective measures]

These differences should have minimized anatomic brain differences between unmedicated patients with ADHD and controls. In fact, findings were generally as striking for the unmedicated patients with ADHD as for those who were being treated with medications, and were more pronounced for white matter volumes.

[Fred A. Baughman Jr., MD:Here, in a single reach, they would have us believe they have validated, ADHD as a disease, and, at the same time exonerated Ritalin/amphetamine treatment of all harm to the brain, that so strongly suggested, time and again by the scanning research of 1986-1998 and the present]

Thus, our analyses show that decreased brain volumes in ADHD in both white and gray matter compartments are not due to drug treatment. Conversely, we have no evidence that stimulant drugs cause abnormal brain development.⁴⁶

[Fred A. Baughman Jr., MD:Yes there is something to treat-ADHD! No! there is nothing to fear from the stimulant drugs; leave no child untreated, take them from the home of negligent parents refuse to treat them as is already being done by the hundreds of thousands in the US today.  ]

Patients with ADHD had developmental trajectories for nearly all brain regions that paralleled growth curves for controls but on a lower track. The one exception, foreshadowed by an earlier cross-sectional study,¹³ was the caudate nucleus, for which differences between patients and controls became negligible by midadolescence. As the caudate nucleus reaches its maximum volume around 10 years, the potential relationship between normalization of caudate volume in ADHD and decreased ratings of hyperactivity/impulsivity in children with ADHD,⁴⁷ as well as in quantitative measures of movement in normative samples,⁴⁸ should be addressed in future studies.

Longitudinal follow-up of functional outcome is continuing; hence, we cannot report definitively on the relationship between continuing anatomic deviance or normalization vs outcome. Preliminarily, global functional outcome in 64 patients with ADHD (20 females) evaluated 4 years after initial scan does not suggest any significant relationships between continuing anatomic deviance and clinical follow-up status.

We did not find evidence of a primarily frontal abnormality in ADHD. Instead, we found the smallest diagnostic effect sizes in frontal lobes. However, these results cannot be interpreted as definitive evidence against the frontal-striatal hypothesis of ADHD pathogenesis, because our units of analysis, while highly reliable, were too large. These methods have been useful in detecting age-, sex-, and diagnosis-specific differences in growth curves,^{39-41, 49} and their application to ADHD was warranted. Alternate approaches, such as unbiased pixel-based analyses,⁵⁰ may be needed to detect more localized anatomic abnormalities in regions such as cingulate, orbitalfrontal, or dorsolateral prefrontal cortex in patients with ADHD.⁵¹ However, these methods may also require even larger or more closely matched contrast groups (eg, twin or sibling controls) given the mostly modest effect sizes and substantial between-subject variations in brain anatomy.⁵²

Limitations of this study include the use of referred samples for patients and highly screened controls that may not be optimally representative. We recruited female patients with ADHD who were comparable in severity with our previous samples of males,⁵³ but in so doing may have selected females who are atypical of most community and clinical samples. We lost significantly more scans from children with ADHD because of excessive motion, but again, this bias should have removed the most symptomatic patients.

In conclusion, ADHD is associated with about a 3% (adjusted; 4% unadjusted) decrease in volume throughout the brain. Intriguingly, this decrease is most marked in white matter of unmedicated patients. Furthermore, with the exception of caudate nucleus, longitudinal growth curves are roughly parallel, suggesting that the fundamental developmental processes active during late childhood and adolescence are essentially healthy in ADHD, and that neuropsychiatric symptoms appear to reflect fixed earlier neurobiological insults or abnormalities. Future studies should focus on younger patients being enrolled into controlled treatment studies while in preschool and on the development of improved quantitative measures of brain anatomy and of the component endophenotypes of ADHD.⁵⁴ Finally, despite the importance of these findings, anatomic MRI studies remain appropriate only for research, as they cannot yet contribute to the diagnostic assessment of ADHD.

[Fred A. Baughman Jr., MD:Here, in the final sentence is the  confession that MRI for all the millions of dollars of tax and Pharma money spent on MRI/ADHD research is good for nothing more than deceiving all who read about it an all who read the press releases and abstracts (always cleansed of all mention that the brain atrophy, shrinkage, shriveling that  might just be due to those long-known encephalotoxins: Ritalin and the amphetamines given us as "treatment" for the never-proved disease; ADHD.  Has Castellanos, et al, proven that ADHD is not a fraud?  They have not. ]

Fred A. Baughman Jr., MD:  I invite my readers to read on.  In the appended material I expand upon many of the studies Castellanos, et al refer to, i.e., that appear in their bibliography.

CT AND MRI SCANS IN ADHD VS. NORMAL

In their joint presentation to the National Institutes of Health, Consensus Conference on ADHD, November 16-18, 1998.  James Swanson, Ph.D., of the University of California at Irvine, and F. Xavier Castellanos of the National Institute of Mental Health (NIMH), among the most prominent of ADHD researchers, reviewed the Biological Bases of Attention Deficit Hyperactivity Disorder: Neuroanatomy, Genetic, and Pathophysiology.  Who would guess, having heard their title, that there is no biological basis for ADHD.

Swanson and Castellanos wrote [1]: "One of the most important current developments has been the convergence of findings from magnetic resonance imaging studies of brain anatomy (aMRI)."   What, I wonder, does 'convergence of findings' mean relative to proof?  They continue: "We will present a meta-analysis of studies from several independent laboratories that have reported ADHD/HKD (hyperkinetic disorder, a term used in the UK) abnormalities in two specific but still coarsely defined brain regions of the frontal lobes and basal ganglia.  For example, Filipek and colleagues [2] reported that a group of children with ADHD/HKD had brain volumes about 10 percent smaller than normal in anterior superior regions (posterior prefrontal, motor association, and mid-anterior cingulate, anterior inferior regions, and anterior basal ganglia), and Castellanos and colleagues [3] reported that right anterior frontal, caudate, and globus pallidus regions were about 10% smaller in an ADHD/HKD group than in a control group. 

"The convergence of findings within and across investigators has not emerged for functional imaging studies using positron emission tomography (PET) [4,5]  as it has for aMRI studies."  What the authors mean here, is that only anatomic MRI (aMRI) studies have shown abnormalities-atrophy--in ADHD subjects,  relative to normal controls. 

Swanson, presenting at the Consensus Conference for himself and Castellanos [1], summarized: " Recent investigations provide converging evidence that a refined phenotype of ADHD/HKD is characterized by reduced size in specific neuroanatomical regions of the frontal lobes and basal ganglia."  

Nor did Swanson leave any doubt that he was claiming that the brain atrophy he had described was part and parcel of ADHD/HKD (by whatever name)-it's long-sought biological basis (ADD having been conceptualized-invented, for the DSM-III in 1980) [6]    Saying these brain abnormalities were a component of the ADHD 'phenotype,' Swanson posited that it had genetic basis-an abnormal 'genotype.'  Speaking of  'phenotype' one speaks of the somatic or physical manifestation of all the genes-the genotype.  Saying one has one has an abnormal 'phenotype,' one implies an abnormal gene or genes-an abnormal 'genotype' as it's cause.

Baughman, in attendance [7], took the microphone and asked:   "Dr. Swanson, why did you not mention that virtually all of the ADHD subjects in the neuroimaging studies have been on chronic stimulant therapy and that this is the likely cause of their brain atrophy?" 

Audience:  "ooh, wow!" 

Swanson:  "Well, that's a hypothesis.  I don't know the exact numbers of how many were or were not on medication and as I indicated I understand that this is a critical issue and in fact I am planning a study to investigate that.  I haven't yet done it."

Fourteen years ago, in a study utilizing computerized, cranial, tomographic scans (CCT), Nasrallah [8] suggested that the Ritalin/stimulant therapy itself was the cause of the brain atrophy.  

What then, of the conclusion reached by Swanson and Castellanos, and virtually all other contemporary, ADHD/aMRI researchers-that the brain atrophy is the physical abnormality that makes ADHD a bona fide disease!   Unlike Nasrallah [8], they say little or, like Swanson, nothing, of the possibility that the stimulant medication has caused the brain atrophy-the stimulant medication that virtually all of the ADHD subjects were on. 

In that the only difference (variable) between the ADHD subjects and the normal controls, was the stimulant drugs that nearly all of the ADHD subjects were on, the only plausible conclusion is that the brain atrophy in ADHD subjects is due to the stimulants  they were on. 

Why then, does psychiatry tell the public, so insistently, that ADHD is a disease?  Why are they told Ritalin/stimulant therapy is safe, and effective.  Effective against what?      Why, given 14 years of evidence, is it been kept from them that Ritalin/stimulant therapy  causes the brain to atrophy, shrivel, shrink? 

Acknowledgments that ADHD subjects were on stimulants are sparing and are found only in the small print of  original reports.  Diabolically, reviews, abstracts and press releases speak only of brain atrophy as a function of the ADHD. 

Recall--it was in 1986, that Nasrallah [8], suggested that the brain atrophy in their 'hyperactive' subjects might be due to the stimulant therapy.  Had psychiatry really wanted to know, all they would have had to do was an MRI study comparing drug-free,  patients with normal controls.  Had they done such a study, and found brain atrophy, they would have credible evidence of a physical basis for ADHD.  But in all of this time--in the entire 14 years since the Nasrallah report, they have failed to do such a study.  They have continued doing MRI studies, comparing ADD/ADHD patients to normal controls, but invariably, the ADD/ADHD subjects have been on brain damaging stimulants. 

One wonders if they have actually done a study of drug-free subjects, somewhere along the way, found their brains to be normal, and elected not to publish the 'negative results.'  This, after all would leave ADD/ADHD, invalid  and the main drugs used to treat this 'non-disease,' the stimulants, a proven cause of brain atrophy.  What might the reason be for their continuing to perform the same flawed experiment, time after time, for 14 years.  Just as they have avoided performing the one appropriate study--that on drug-free ADD/ADHD subjects, they have carefully avoided including stimulant 'non-responders' in any of their ADD/ADHD subject groups.  Could it be that they know that long-term stimulant treatment/exposure is necessary to assure a finding of brain atrophy on a scan--a finding they intended all along to represent as due to ADHD?  What other reason could there be. The last thing they want to prove with their research is that the drugs are causing brain shriveling and that ADHD does not exist.  Why else would Castellanos tells us in his Reader Digest interview of January, 2000, just as he told us in 1996, just as Swanson told us at the Consensus Conference, in 1998 we should do MRI scans in drug-naïve ADHD subjects, and we plan to, but we just haven't  gotten around to it yet. 

REVIEW OF CT, MRI SCANS IN ADHD-ALL OF THEM ON STIMULANTS

Nasrallah et al [1986], in a computerized cranial tomography (CCT) study (earlier, not as precise technology, as MRI) of 24 hyperactive subjects, found brain atrophy in more than 50%, leading the authors to conclude: "cortical atrophy may be a long-term adverse effect of this treatment."

Hynd et al [1991], using MRI, compared ADHD children "judged to be favorable responders to Ritalin," and controls.  The corpus callosum, the large bundle of nerve fibers connecting the two cerebral hemispheres was found to be smaller in the ADHD group and was said to be due to ADHD.  No mention was made of the possibility, actually a probability that the atrophy of the corpus callosum might be due to Ritalin. 

Giedd, et al. [1994], using MRI, found the  corpus callosum to be smaller in hyperactive boys than in normal controls and attributed it to their ADHD.

Matochik, J.A. et.al, AMJ Psychiatry 1994;151:658.  Subjects with ADHD meeting DSM III (1980) criteria for ADHD , the Utah criteria  for Attention Deficit Disorder in adulthood, having a definite childhood history of ADD with hyperactivity but no history of any other major psychiatric disorders, were studied before and after a minimum of six weeks of treatment with Methylphenindate, or Deamphetamine.  Neither changed global or whole brain metabolism.  Having failed to demonstrate abnormalities, with or without stimulant therapy, the authors emphasized that both drugs brought about significant improvement of behavior and strongly indicated that Methylphenindate and Deamphetamine were effective agents in the treatment of adults with ADHD 

Castellanos [1994], wrote:  "Thirty-nine of the50 patients had been previously treated with psychostimulants, and all patients participated in a 12 week double-blind trial of methylphenidate, d-amphetamine, and placebo, which is described elsewhere."  "The normal pattern of slight but significantly greater right caudate volume across all ages was not seen in ADHD."  "Total brain volume was 5% smaller in the ADHD boys..."

The majority of these patients, too, were treated, either with methylphenidate (Ritalin) or d-amphetamine (Dexedrine).  Here we have another study in which stimulant exposure in the ADHD group is the only physical variable, and, the likely cause.  

Lyoo, et al, [1996], studied 45 males and 6 females (51 total) with ADHD, and 28 controls.  Those with ADHD had a significantly  larger posterior lateral ventricles

[Fred A. Baughman Jr., MD:when brain tissue atrophies, the ventricles, filled with spinal fluid, enlarge, taking up the space vacated by brain matter]

.  A number of those in the ADHD group had co-existent conduct disorder, a second, confounding diagnosis, also, never proved to be a disease.   The majority were also on stimulants. 

Castellanos et al [1996], took magnetic resonance images (scans) of the brains of 57 boys, 5-18 years of age, said to have attention deficit hyperactivity disorder (ADHD) and of 55 "healthy" age-matched controls.  Those in the ADHD group were found to have significantly smaller, atrophic brains (diminution of total cerebral  and cerebellar volume relative to the " healthy"controls), and the loss, at many sites, of right more than left asymmetry (where right-sided structures are normally larger than that on the left) leading The authors conclude that the differences found in the ADHD group were "consistent with (previously) hypothesized (but never proven) dysfunction of right-sided prefrontal (front, right half, of the brain)-striatal (right sided, deep-lying nuclei, or cell groups) systems in ADHD."  In a nutshell, Castellanos, et al, concluded that the abnormalities found were due to ADHD, exactly as previously hypothesized-tending to prove that ADHD, virtually always diagnosed by elementary school teachers, armed with a pencil-paper behavior rating scale, is, indeed, a "brain disease"-something frequently claimed, but never proven, to that date.  A further indication that Castellanos, et al, were convinced that they had proof in hand that ADHD to be a disease, was their reference, must unscientifically, in the text of their article, to the normal control group as "'healthy' matched controls," leaving no doubt that the authors considered the ADHD subjects to be "unhealthy," or "diseased." 

If the 1996 Castellanos article was the first "proof" that ADHD is an actual disease with confirmatory, characteristic abnormalities of the brain, what proof were Castellanos and his NIMH colleagues referring to when, in NIH Publication 94-3572 "Attention Deficit Hyperactivity Disorder-Decade of the Brain" (1994), with "Scientific information and review (was) provided  by NIMH staff members (Castellanos, included)..." in which they refer to ADHD as "the disease" (page 7).   It is my understanding of the ADHD scientific literature, although voluminous and weighty, that there was, to that time, no proof that ADHD was a disease or a medical syndrome (about the same thing as a disease) with a confirmatory physical or chemical abnormality (pathology).  Be that as it may-what about the current claim of Castellanos, et al, that the MRI abnormalities in the current study comprise the long awaited, frequently claimed, proof that ADHD is a "disease."  There is fine print within the article that bears scrutiny, forcing us to draw a very different conclusion.  Under "Subjects with ADHD," was written "Fifty-three of them (of the 57 with ADHD) had been previously treated with psychostimulants, and 56 participated in a 12-week, double-blind trial of methylphenindate, dextoamphetamine and placebo..." 

Under "Comment," running fully a page and a half, is found the brief acknowledgement: "Because almost all (93%) subjects with ADHD had been exposed to stimulants, we cannot be certain that our results are not drug related."  This did not, however, deter Castellanos, et al, from stating, at the end of their "Comments": In summary, we have found decreased volume of several candidate brain regions previously hypothesized to be involved in the pathophysiology of ADHD..."

In 1997, Filipek, et al,  undertook volumetric MRI analysis "To test by MRI...the a priori hypotheses that developmental anomalies exist in attention-deficit hyperactivity disorder in left caudate and right prefrontal frontal/ and or/ posterior parietal hemispheric regions in accord with neurochemical, neuronal circuitry and attentional framework hypotheses, and prior imaging studies." 

"All subjects with ADHD had been placed on medication for at least 6 months prior to the study and were felt to be responding favorably at the time of the MRI."   Five of the subjects had not previously responded to methylphenidate or dextroamphetamine but responded to non-stimulant medication.  Nonetheless, Filipek et al, conclude: "This study is the first to report localized hemispheric structural anomalies in ADHD..."

Berquin , et al. (1998) undertook an MRI study of the cerebellum in attention-deficit hyperactivity disorder.  In 46 boys with ADHD, vermal  (vermis or the cerebellum) volume was significantly less than in the 47 matched controls.  From 'methods' we read, 'The 46... boys with ADHD were recruited for a drug-treatment study and were included in a prior report,"  The DSM-III-R, 1987 was used for diagnosis herein.

Berquin et al commented on the association of cerebellar atrophy with alcohol and acknowledged they could not fully rule out fetal alcohol exposure.  Making mention of alcohol exposure as a possible contributor to cerebellar atrophy and acknowledging that all of their patients (number = 46) were recruited from the among the 57 subjects in the study of Castellanos, et al, 93% of whom had been on stimulant therapy, Berquin, et al made no mention, as did Castellanos, of the fact that "Because almost all (93%) subjects with ADHD had been exposed to stimulants, we cannot be certain that our results are not drug related."  Clearly the cerebellar atrophy described in this study could have been, and probably was, stimulant induced.  

1999, Mostofsky: Brain abnormality linked to ADHD, April 20, 99 (Reuters Health) -- Compared with other children of the same age, children with attention deficit hyperactivity disorder (ADHD) have a smaller brain volumes, particularly smaller amounts of gray matter in the right frontal area of their brains, "There is a lot of evidence that the brain's right hemisphere is dominant in attentional processes," said study author and neurologist Stewart Mostofsky, MD, of the Kennedy Krieger Institute and Johns Hopkins School of Medicine in Baltimore. "Abnormalities in the brain's right frontal structure and function may be contributing to the behavioral impairments associated with ADHD."  Along with less right frontal gray matter, there searchers also found that ADHD patients had smaller volumes of left frontal gray matter as well as right and left frontal white matter when compared to children without ADHD. The study included 12 boys diagnosed with ADHD and 14 boys without ADHD. All boys were between the ages of seven and 13.   Nothing was said of drug status in the press release, I have had no answer to my letter of inquiry to Mostofsky, and there is no evidence as of July, 2000, that this work has been published.

All ADD/H subjects taking methylphenidate for a least 6 months and were judged to be responders in all settings by authors RS or JB

What were the ages and dates of MRI scans of the 10 subjects.  Were they done at or prior to the 97 Filipek report.  Were they done then and another time, prior to this year 2000 report or just prior to this, the year 2000 report?  What were dates of MRI scans (one or more)  and of psychometric testing?  Were they at about the same times, as one would have hoped, or were they at different times, different months or years. 

"Since response to stimulant medication  [fb-exposure] was linked to structural differences in children with ADD/H in our previous study (Filipek, 1997) the non-responders [fb-those not responding to it and not kept on it, i.e., those with least exposure] were excluded from the present study to control for medication response."

What they posit with this is that (1) ADD/H (not responding to Ritalin) is  one disease with one particular brain abnormality, and that (2)  ADD/H responding is another disease altogether with it's own particular brain abnormality.  What they conclude with this study is that the MRI abnormalities reported herein are those due to (2).  They have avoided doing MRIs on a relatively normal, unexposed group-group (1).  They have not acknowledged that ADD/H - untreated are normal or that the cause of the MRI abnormalities might be due to MPH effect on the previously normal brains of ADD/H.  

ADHD (untreated) = normal = normal brain and body

 normal + MPH (known brain toxin) = brain atrophy due to MPH exposure

ADHD + MPG                                  =  brain atrophy due to MPH

ADHD (nonresponder and therefore not kept on it exposed = normal MRI, little, if any brain damage, i.e., no brain atrophy of MPH exposure. 

"Furthermore, brain structural differences in children with ADHD, predominantly inattentive type, have not been studied."  This type is from the 1994, DSM-IV construct, which was not employed in the diagnosis of these subjects.  The Filipek ('97) subjects were likely recruited post 80 and post 87 using DSM III and III-R criteria.   

From the abstract: 10 males with ADD/H (8-17 and 11 controls (9-18) with neuropsychologic evaluation and MRI scan.

MPH = methylphenidate

Results: 

"As had been reported previously by these authors, the children with ADD/H were found to  have reversed asymmetry of head of the caudate, smaller volume of the left caudate head; smaller volume of  white matter of the right frontal lobe." 

"There is emerging evidence that compromised brain morphology of selected regions (above) is related to behavioral measures of inhibition and attention."

However, Breggin (In Breggin, P.R. Brain Disabling Treatments in Psychiatry, p 172)  points out the likelihood that these patients, in a day treatment program, were on stimulants or multiple psychotropic drugs and that the authors failed to discuss the issue of drugs at all.  Breggin concludes: There is reason to suspect that this is another incriminating study in regard to brain damage from Ritalin.  That the authors don't even mention the drug history of the children-and that the journal editor didn't insist upon the details-is very telling about the willingness of modern psychiatry to deny the potential damage of its treatments.  The January, 2000, Reader's Digest confession of Castellanos that, to that date, there were no such studies in drug-naïve ADHD subjects, covers all studies in print, this one, and that of Semrud-Clikeman, et al,  (accepted for publication, 10/5/99) included.  Giedd, like Castellanos is from the NIMH, therefore the ADHD subjects in this study may be one-and-the-same as those used in other studies from the NIMH or NIMH sponsored research.  Tannock (1988) points out that neuroimaging studies of ADHD come primarily from 3 centers.  It appears to me in reading this literature not only that the same ADHD patients are the subjects in several different reports, but that the same scans of the same subjects may be reported in multiple publications. [fb-my suspicion is that they include 'Ritalin reponders,' knowing they  have been on long-term Ritalin, knowing that the Ritalin is the only physical variable, and the likely cause of the callosal atrophy, and then they have an abnormality to represent 'as due to the ADHD.'  Ritalin-nonresponders would not long stay on the Ritalin, would likely have normal looking brains on MRI scan, and for this reason are excluded.]

Fb-again all of  the human beings who are their research subjects, or nearly all, regardless of the psychiatric disorder have been on a drug or drugs.  The only real, proven, organic abnormalities they find, attribute to the psychiatric disorder, report, and hold press conferences on, are those due to the drugs they give to the  subjects, all of whom were physically normal until the drugs were begun.   What they do and represent as science, medicine and research is absolutely unthinkable-pure duplicity.   It would be fascinating indeed to take 1, 5 or 10 NIMH research subjects and track how many reports they prove to be participants in, per 5 or 10 years.  I find it hard to follow their methods when they refer to subjects described not in the report one is reading but in one published at some prior date, as we have here with Casey et al, making use of subjects from the Castellanos, 96 report.  Sure this may be difficult to assail in court but eminently  possible, an with persistence it is a house of cards that will all tumble down.  Consider the end of the day conclusions of all of their research.  Here from the recent neuroimaging review of Hendren, et al, J Am Acad Child Adol Psychiatry 2000, 39 (7):815-828, a review of the neuroimaging of