- Top of page
Aims: Anorexia nervosa (AN) is subdivided into the restricting type (AN-R) and the binge-eating/purging type (AN-BP), but differences in cerebral blood flow between patients with these types of AN and healthy controls have not been investigated.
Methods: The present study was designed to elucidate any such differences using resting single photon emission computed tomography (SPECT) studies to compare the differences in cerebral perfusion among both types of AN and a healthy control group. Resting regional cerebral blood flow was assessed using SPECT with technetium-99m hexamethylpropyleneamine oxime in 13 female AN-R patients, 13 female AN-BP patients, and 10 healthy women as controls with 3-D stereotactic surface projections.
Results: The analytic program of the SPECT images showed bilateral decreased perfusion of the subcallosal gyrus (SCG), midbrain and posterior cingulate gyrus (PCG) in both AN-R and AN-BP patients, as compared with the controls. There were no clear differences between the AN-R and AN-BP groups. There were no significant differences in cerebral blood flow between patients with AN-R and AN-BP.
Conclusions: Abnormalities of the neuronal circuits containing the SCG, midbrain and PCG are possibly relevant to trait-related AN.
ANOREXIA NERVOSA (AN) is characterized by refusal to maintain bodyweight at or above a minimally normal weight for age and height; intense fear of gaining weight or becoming fat, even though underweight; disturbance in the way in which one's bodyweight or shape is experienced; undue influence of bodyweight or shape on self-evaluation, or denial of the seriousness of the current low bodyweight; and amenorrhea in postmenarcheal women. AN is further classified into two subtypes, that is, the restricting subtype (AN-R), in which the person has not regularly engaged in binge-eating or purging behavior, and the binge eating/purging subtype (AN-BP), in which the person has regularly engaged in binge-eating or purging behavior.1
Thus, there are some common or different clinical symptoms between AN-R and AN-BP patients. Several biological studies, such as neuroendocrine study,2–4 an event-related potentials study on P300,5,6 and a genetic study,7–9 have investigated the differences between AN-R and AN-BP. The understanding of the biophysiological differences between AN-R and AN-BP remains, however, unclear.
Neuroimaging studies have also been conducted to investigate brain functions in AN. Functional neuroimaging studies are among the most useful of methods to clarify the psychophysiological aspects of AN. To date, resting single photon emission computed tomography (SPECT) studies using voxel-based analysis have been performed.10,11 They have found a reduction of the regional cerebral blood flow (rCBF) in the brain cortex of AN patients before weight gain. To the best of our knowledge, only one resting SPECT study has been conducted to compare AN-R with AN-BP patients and control subjects,12 which demonstrated bilateral reduction of the rCBF in the anterior cingulate gyrus (ACG) in AN-R, but not in AN-BP patients.
Brain structural studies using computed tomography (CT) and magnetic resonance imaging (MRI) have generally found evidence of brain atrophy, so-called pseudo atrophy, in AN.13 Brain atrophy often affects the interpretation of SPECT images and causes measurement errors. Recently, however, analysis based on the development of 3-D stereotactic surface projections (3D-SSP) has been less influenced by brain atrophy,14,15 and accurate measurements of the rCBF in the brain cortex can be conducted even in patients with brain atrophy. Therefore, we used 3D-SSP for the analysis of SPECT images.
The primary aim of the present study was to investigate the differences in rCBF between AN patients and control subjects, and then to investigate the differences in rCBF between AN-R and AN-BP patients, by analysis of SPECT images using 3D-SSP.
- Top of page
In the present study, decreased perfusion in the bilateral SCG, PCG and midbrain was found in the AN patients (both AN-R and AN-BP) as compared with controls, but there was no difference between the two AN subtypes.
In a recent study, increases of rCBF were shown in the PCG following treatment of AN patients.10 Another study showed decreases of rCBF in the frontal lobe, including the PCG, in AN-R patients before treatment compared with controls.11 A previous PET study also showed decreases in regional cerebral metabolic rate of glucose in the PCG in AN patients compared with controls.26 The present findings that rCBF in the PCG decreased in AN patients compared with controls was in agreement with those studies. Several bodies of evidence have shown that the senses of taste and smell, and different foods and flavors activated specific regions such as the PCG, SCG, ACG, orbitofrontal cortex and amygdala, and these regions might be implicated in the modulation of the reward value of a sensory stimulus such as the taste of food.27–30 The decreased perfusion in the PCG might therefore be partially related to the pathophysiology of AN.
In contrast, there have been no studies on the changed rCBF in the SCG in AN subjects. PET studies of the 5-HT2A receptor showed an abnormality in SCG in AN patients, and this abnormality was shown in subjects who had recovered from both the restricting type recovered-AN (R-AN) (R-AN-R) and binge eating/purging type R-AN (R-AN-BP).31,32 In eating disorder patients the amygdala was shown to be activated by unpleasant stimuli, such as the patients' own body images and high-calorie foods.33,34 Deactivation of the SCG might result in failure of the function of the amygdala in reacting to unpleasant stimuli, because the SCG has interconnections with the amygdala.35 The SCG might be a region that plays a very important role in the psychopathology of AN.
Previous resting SPECT studies found decreased perfusion in the ACG in AN patients.12,19 In the present study, decreases of rCBF were observed in the SCG but not the ACG. Moreover, only one study has demonstrated that a bilateral decrease of the rCBF in the ACG is observed only in AN-R and not AN-BP patients in resting SPECT.12 Although the function of the ACG and SCG in AN is still unknown, the ACG (Broadmann area 24) and SCG (Broadmann area 25) play a common role involved in emotional response and are activated by sensory stimuli such as the taste of food.29,36 Thus, decreased rCBF in the SCG might be compensatory for changes in the ACG. Further studies are needed to address this issue. Another plausible explanation might be the differences in the analytic methods. We used 3D-SSP for analyzing the SPECT images, whereas others used statistical parametric mapping (SPM). However, further study using both methods would be needed to draw a definitive conclusion. As the third possibility, the absence of any significant differences in the duration of illness and the onset age between AN-R and AN-BP patients might explain the differences between the present results and the results of other studies. Further functional neuroimaging studies in AN subjects with detailed clinical background data will be essential to investigate the differences between AN-R and AN-BP subjects.
The present resting SPECT study found no significant differences in rCBF/activity between AN-R and AN-BP patients. The brain alternations were shared by the AN-R and the AN-BP patients. In other words, the present results might represent the common abnormalities between patients suffering from AN-R and AN-BP. In the clinical setting, AN-BP patients usually start the same way as AN-R patients, and frequently swing between the restricting type and bulimic/purging type of the disease.37 Eddy et al. stated that AN-R represents a phase in the course of AN rather than a distinct subtype.38
The present study also had the following limitations: although the diagnosis in the subjects was established by a minimum of two skilled psychiatrists, a structured clinical interview was not used in all of the subjects, nor was the severity of the AN or deficit in the physical status assessed in the subjects. Because of the limitation of the analytic program, 3D-SSP was unable to analyze the correlation between rCBF and clinical factors such as BMI, age, duration current episode of illness, onset age and duration of the current episode. It has been reported that changes in rCBF have been associated with interoceptive awareness in the recovery process from AN.10 In the present study we did not apply psychological assessments such as the eating disorder inventory. Some of the patient participants were taking mainly antidepressants. Numerous studies have demonstrated that administration of antidepressants did not decrease the cerebral blood flow.39,40 The tracer used in the study was HMPAO. The tracer has some quantification-related disadvantages compared with other tracers such as N-isopropyl-p-[123I] iodoamphetamine and 99mTc-ethyl-cistainate dimer. Last, we did not measure the extent of any of the subjects' brain atrophy using CT or MRI.
In conclusion, analysis of the resting SPECT images found no significant differences between patients with AN-R and AN-BP. Abnormalities of the neuronal circuits containing the SCG, midbrain and PCG are possibly relevant to the trait-related AN.