Bulimic adolescents benefit from massage therapy.

Field, Tiffany ; Schanberg, Saul ; Kuhn, Cynthia 等

Bulimia nervosa was originally thought to be a derivative of anorexia, but it is now recognized as a disorder of its own. A diagnosis of bulimia requires the following symptoms: (1) recurrent episodes of binge eating; (2) a feeling of lack of control over eating behavior during the binges; (3) regularly engaging in self-induced vomiting, use of laxatives or diuretics, strict dieting or fasting, or vigorous exercise in order to prevent weight gain; (4) an average of two or more binge-eating episodes a week for at least three months; and (5) persistent overconcern with body shape/weight (American Psychiatric Association, 1987).

The exact etiology of bulimia nervosa has not yet been determined, but the behaviors and symptoms presented by the majority of patients suggest a combination of psychological, social, and physiological factors. Depressed affect is so commonly seen that some believe bulimia is simply a type of affective disorder. According to Edelstein, Haskew, and Kramer (1989), 20-30% of patients with bulimia meet the diagnostic criteria for depression. Bulimic patients who vomit show lower urinary serotonin (Kaye, Ebert, & Gwirtsman, 1984), and elevated plasma norepinephrine (Robinson, Checkley, & Russell, 1985; Smythe, Bradshaw, & Vining, 1983).

Some have suggested that bulimics are difficult to medicate because they do not keep the medication in their systems long enough to absorb it. Nevertheless, significant decreases in bulimic and depressive symptoms have been demonstrated for tricyclic antidepressants, serotonergic agents, and MAO inhibitors. Investigators have found that some bulimic patients who are not depressed respond to antidepressant medication, and some who do suffer from depression may binge less while remaining depressed (Brotman, Herzog, & Woods, 1984; Walsh, Stewart, Roose, Gladis, & Glassman, 1984). Some of the relief experienced by bulimic patients may be due to lowered anxiety and suppressed appetite caused by tricyclic antidepressants rather than to the activity of the antidepressant itself (Pope & Hudson, 1986).

These treatment approaches, however, have not been sufficient on their own. To be successful, treatment must alleviate depressive symptoms and alter any neuroendocrinological abnormalities. Massage therapy has proven effective in these areas, namely reducing depression and cortisol (Field, Morrow, Valdeon, Larson, Kuhn, & Schanberg, 1992). For example, massage has been found to lower both self-reported and observed anxiety and depression as well as salivary cortisol levels in a sample of depressed adolescents (Field et al., 1992).

Given these positive findings, it was hypothesized that massage therapy would similarly be effective in decreasing depression, anxiety, and cortisol levels with a sample of eating-disorder patients. In addition, massage therapy was expected to reduce several other psychological and behavioral traits common in these patients.

Psychotherapy and pharmacotherapy have been somewhat effective, although the majority of patients have continuing eating problems (Garfinkel, Moldofsky, & Garner, 1977; Hsu, 1986). The present study sought to determine whether massage therapy is an effective adjunct.

METHOD

Sample

The subjects were 24 adolescent female bulimic inpatients at a residential treatment center. The inclusion criterion was a DSM-III-R diagnosis of bulimia nervosa: weight loss is usually present, amenorrhea is variable, vomiting/purging is normally present, as is fear of fatness (Hsu, 1986). The patients ranged in age from 16 to 21 and came from middle- to upper-SES (M = 2.2 on the Hollingshead Index) Hispanic (68%) and non-Hispanic White (32%) families. The patients were randomly assigned to a massage therapy or a standard treatment (control) group.

Procedure

Massage therapy. The massage therapy group received a massage 2 days a week for 5 weeks, for a total of 10 massages. The massages were administered by massage therapists. The massage therapy covered several parts of the body (which was fully clothed) and included 15 minutes in a supine position and 15 minutes in a prone position. It consisted of exerting traction upon the neck with the patient in a supine position, followed by smooth strokes across the forehead, jaw, and face, and depressing the shoulders. The therapist then exerted traction on each arm, followed by massage of the hand and smooth strokes over the length of the arm. The torso was gently rocked. The same movements used with the arms and hands were applied to legs and feet. In a prone position, the Achilles tendon was stretched and long strokes were made from the hip to the toes. Also included were lateral lumbar stretching, strokes from the back to the arms, trapezius and neck squeezing, friction alongside the spine, sacral traction, and long, soothing strokes from the head to the feet.

Standard treatment. All adolescents residing at the Renfrew Treatment Center are evaluated by a psychiatrist to assess their goals and identify a treatment plan. The adolescents then meet their primary clinician and begin participating in daily small-group, individual, family, and community therapy. In addition, they work with a registered dietician to become more knowledgeable about nutrition and the basic principles of physiology and metabolism. They also participate in nonverbal therapies, such as movement therapy. In total, residents attend 30 to 40 group therapy sessions per week.

Measures - Immediate Effects (Pre/Post Therapy Sessions)

The State-Trait Anxiety Inventory (STAI). The 20-item state scale of the STAI (Spielberger, Gorsuch, & Lushene, 1970) measures how the subject feels at that moment (e.g., "I feel: very nervous, nervous, or not nervous"). This scale was administered immediately prior to and again 30 minutes following the massage therapy on the first and last days of the study. Research has demonstrated that the STAI has adequate concurrent validity (Spielberger, 1972) and adequate internal consistency (r = .83; Spielberger et al., 1970). In addition, scores have been found to increase in response to stress and decrease under relaxing conditions (Spielberger et al., 1970).

Profile of Mood States Depression Scale (POMS-D). The POMS Depression Scale (McNair, Lorr, & Droppleman, 1971) consists of 19 adjectives, rated on a 5-point scale ranging from not at all to extremely, which reveal depressed mood "right now." The scale has adequate concurrent validity and good internal consistency (r = .95; McNair & Lorr, 1964) and is an adequate measure of intervention effectiveness (Pugatch, Haskell, & McNair, 1969). This scale was administered immediately prior to and again 30 minutes following the massage therapy on the first and last days of the study.

Behavior Observation Scale (BOS). Behavior Observation Scale (Field et al., 1992) ratings were made based on affect and anxiety observed during the 30 minutes prior to massage therapy and the 30-minute period after the session. Behaviors were coded at 30-second intervals by a coder who was blind to the hypotheses of the study and the group assignment of the subjects. The behaviors were rated on a 3-point scale and then summed and averaged. Intercoder reliability (10 sessions) was calculated using Cohen's Kappa to correct for chance agreement. BOS ratings have reliably discriminated depressed and adjustment-disorder children and adolescents before and after massage therapy (Field et al., 1992) and relaxation therapy (Platania-Solazzo, Field, Blank, Seligman, Kuhn, & Schanberg, 1992).

Salivary cortisol. Since cortisol levels in saliva reflect stress levels 20 minutes prior to their sampling, samples were collected just before massage therapy and ten minutes after the therapy. Salivary cortisol samples were obtained by having subjects place a cotton swab (dipped in lemonade-flavored crystals) along their gumline for approximately 30 seconds. The swab was then placed in a syringe and the plunger depressed to inject the saliva into a microcentrifuge tube, which was sealed, frozen, and sent to Duke University to be assayed for cortisol. Cortisol assays have proven to be an effective index of stress (Field et al., 1992; Ironson et al., 1996) and were included because stress levels are typically elevated in eating-disorder patients (Robinson et al., 1985; Smythe et al., 1983).

Measures - Longer-Term Effects (First Day/Last Day)

Eating Disorders Inventory (EDI). The 64-item EDI (Garner, Olmsted, & Polivy, 1983) consists of eight subscales measuring psychological and behavioral traits common in eating-disorder patients: drive for thinness, bulimia, body dissatisfaction, ineffectiveness, perfectionism, interpersonal distrust, interoceptive awareness, and maturity fears. Items are answered on a 6-point scale ranging from always to never. Criterion-related validity (patients correctly classified) has been established for this scale (92%). In addition, the scale has very good internal consistency (alpha = .90), and convergent/discriminant validity has been demonstrated for each of the subscales (Garner et al., 1983).

Center for Epidemiological Studies Depression Scale (CES-D). The 20-item CES-D (Radloff, 1977) assesses depressive symptomatology over the past week. Scale items represent the major symptoms of depression as identified by clinical judgment, frequency of use in other questionnaires, and factor analytic studies. Responses are made on a 4-point scale (rarely or none of the time, some or little of the time, a lot of the time, and most or all of the time). The CES-D has high internal consistency (alpha = .86) and test-retest reliability. Validity has been established through correlation with other self-report measures and clinical ratings of depression (Radloff & Locke, 1986; Radloff, 1991).

Urine samples. Urine samples were collected from the subjects prior to the end of the first and last days of therapy. An aliquot of each sample was frozen and sent to Duke University to be assayed for cortisol, serotonin (5-HIAA), creatinine, and catecholamines (norepinephrine, epinephrine, and dopamine). (See Kuhn, Schanberg, Field, Symanski, Zimmerman, Scafidi, & Roberts, 1991, for a description of the procedure.) Although bulimics show depleted serotonin (Kaye et al., 1984), they are noted to have elevated norepinephrine levels (Robinson, Checkley, & Russell, 1985; Smythe, Bradshaw, & Vining, 1983). In previous studies, massage therapy reduced norepinephrine levels in depressed adolescents and increased serotonin levels in stressed men (Ironson et al., 1996). Thus, decreased norepinephrine and cortisol and increased serotonin levels were expected following massage therapy.

RESULTS

Immediate Effects (Pre/Post Therapy Sessions)

Repeated-measures analyses of variance by group (massage/control) and post hoc Bonferroni t tests were conducted to assess the immediate effects (pre/post) of the massage therapy sessions on the first and last days of treatment. These analyses yielded interaction effects suggesting that on both days the massage group reported significantly lower anxiety and less depressed mood after therapy. On the first day (but not the last), the massage group showed significantly less stress and was observed to have more positive affect and less anxiety. There were no significant effects for the control (standard treatment) group (see Table 1).

Longer-Term Effects (First Day/Last Day)

Repeated-measures analyses of variance by group (massage/control) and post hoc Bonferroni t tests were conducted to assess the longerterm effects (first day/last day) of repeated massages. Significant interaction effects were found (see Table 2). The massage group had improved scores on the Eating Disorders Inventory subscales (drive for thinness, bulimia, body dissatisfaction, ineffectiveness, perfectionism, interpersonal distrust, interoceptive awareness, and maturity fears), reported less depression, and had higher dopamine and lower cortisol levels.

DISCUSSION

Both self-report and behavior observations revealed an immediate decrease in anxiety and depression in these bulimic females after massage [TABULAR DATA FOR TABLE 1 OMITTED] sessions. It was surprising, therefore, that the salivary cortisol levels decreased only slightly. However, the levels were already very low at baseline, probably because sampling occurred late in the afternoon when cortisol typically reaches its lowest level in the diurnal cycle, making it difficult to decrease further. Yet, the decrease in salivary cortisol from the first to last day and the significant decrease in urinary cortisol suggest a notable decrease in stress.

Table 2

Mean Scores (First Day/Last Day) for Massage Therapy Group (Control
Group in Parentheses)

Measure First Day Last Day p

Self-report measures

Eating Disorder Inventory 87.6 48.7 .001
 (81.2) (82.7) (ns)

Depression (CES-D) 36.2 27.2 .001
 (38.8) (30.9) (.05)

Biochemical measures

Norepinephrine 21.7 24.9 ns
 (32.3) (24.3) (ns)

Epinephrine 5.1 4.3 ns
 (8.3) (5.7) (ns)

Dopamine 146.5 210.8 .05
 (183.0) (150.1) (ns)

Urinary cortisol 222.4 152.2 .05
 (162.1) (140.4) (ns)

The convergence of measures, in this case the decreases in depression and cortisol, is consistent with the behavioral findings. Nonetheless, depression levels remained high, as might be expected, since bulimia is considered a depressive disorder (Edelstein et al., 1989). More intensive, longer-term massage therapy may be needed to decrease depression further.

The significance of the increased dopamine levels is less clear, although dopamine is often lower in depressed adults. Although serotonin rose, the increase was not significant, and norepinephrine showed a slight increase rather than a decrease.

Whether decreased depression contributed to improved attitudes regarding their eating disorder or improved attitudes contributed to less depression is also unclear. Nevertheless, massage therapy may have raised the patients' awareness of their bodies, which some have said is critical for resolving body perception dissonance (Vandereycken & Meerman, 1984).

Massage therapy attenuated several major problems associated with bulimia: anxiety, depression, neuroendocrinological abnormalities, and poor self-image. Although more extensive studies are needed, these results suggest that massage therapy is an effective adjunct to standard treatment.

REFERENCES

American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author.

Brotman, A. W., Herzog, D. B., & Woods, S. W. (1984). Antidepressant treatment of bulimia: The relationship between binging and depressive symptomatology. Journal of Clinical Psychiatry, 45, 7-9.

Edelstein, C. K., Haskew, P., & Kramer, J. P. (1989). Early cues to anorexia and bulimia. Patient Care, 23, 155-175.

Field, T., Morrow, C., Valdeon, C., Larson, S., Kuhn, C., & Schanberg, S. (1992). Massage reduces anxiety in child and adolescent psychiatric patients. Journal of the American Academy of Child and Adolescent Psychiatry, 31, 125-131.

Garfinkel, P. E., Moldofsky, H., & Garner, D. M. (1977). The outcome of anorexia nervosa, significance of clinical features, body image, and behavior modification. In R. A. Vigersky (Ed.), Anorexia nervosa (pp. 315-330). New York: Raven Press.

Garner, D. M., Olmsted, M. P., & Polivy, J. (1983). The Eating Disorders Inventory: A measure of cognitive-behavioral dimensions of anorexia nervosa and bulimia. In P. L. Darby, P. E. Garfinkel, D. M. Garner, & D. V. Coscina (Eds.), Anorexia nervosa: Recent developments in research (pp. 173-184). New York: Alan R. Liss.

Hsu, L. K. G. (1986). The treatment of anorexia nervosa. American Journal of Psychiatry, 443, 573-581.

Ironson, G., Field, T, Scafidi, F., Kumar, M., Patarca, R., Price, A., Goncalves, A., Hashimoto, M., Kumar, A, Burman, I., Tetenman, C, & Fletcher, M. A. (1996). Massage therapy is associated with enhancement of the immune system's cytotoxic capacity. International Journal of Neuroscience, 84, 205-218.

Kaye, W. H., Ebert, M. H., & Gwirtsman, H. E. (1984). Differences in brain serotoninergic metabolism between bulimic and nonbulimic patients with anorexia nervosa. American Journal of Psychiatry, 141, 1598-1601.

Kuhn, C., Schanberg, S., Field, T., Symanski, R., Zimmerman, E., Scafidi, F., & Roberts, J. (1991). Tactile/kinesthetic stimulation effects on sympathetic and adrenocortical function in preterm infants. Journal of Pediatrics, 119, 434-440.

McNair, D. M., & Lorr, M. (1964). An analysis of mood in neurotics. Journal of Abnormal Social Psychology, 69, 620-627.

McNair, D. M., Lorr, M., & Droppleman, L. F. (1971). POMS - Profile of Mood States. San Diego, CA: Educational and Industrial Testing Service.

Platania-Solazzo, A., Field, T., Blank, J., Seligman, F., Kuhn, C., & Schanberg, S. (1992). Relaxation therapy reduces anxiety in child/adolescent psychiatry patients. Actapaedopsychiatrica, 55, 115-120.

Pope, H. G., & Hudson, J. I. (1986). Antidepressant drug therapy for bulimia: Current status. Journal of Clinical Psychiatry, 47, 339-345.

Pugatch, D., Haskell, D., & McNair, D. M. (1969). Predictors and patterns of change associated with the course of time-limited psychotherapy. Mimeo report.

Radloff, L. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.

Radloff, L. (1991). The use of the Center for Epidemiologic Studies Depression Scale in adolescents and young adults. Journal of Youth and Adolescence, 20, 149-165.

Radloff, L. S., & Locke, B. Z. (1986). The Community Mental Health Assessment Survey and the CES-D Scale. In M. Weissman, J. Myers, & C. Ross (Eds.), Community surveys. New Brunswick, NJ: Rutgers University Press.

Robinson, P. H., Checkley, S. A., & Russell, G. F. M. (1985). Suppression of eating by fenfluramine in patients with bulimia nervosa. British Journal of Psychiatry, 146, 169-176.

Smythe, G. A., Bradshaw, J. E., & Vining, R. S. (1983). Hypothalamic monoaminergic control of stress induced adrenocorticotropin release in a rat. Endocrinology, 113, 1062-1071.

Spielberger, C. D. (1972). Anxiety as an emotional state. In C. D. Spielberger (Ed.), Anxiety: Current trends in theory and research. New York: Academic Press.

Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). The State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press.

Vandereycken, W., & Meerman, R. (1984). Anorexia nervosa: A clinician's guide to treatment. New York: Walter de Gruyter.

Walsh, B. T., Stewart, J. W., Roose, S. P., Gladis, M., & Glassman, A. H. (1984). Treatment of bulimia with phenelzine. Archives of General Psychiatry, 41, 1105-1109.

The authors would like to thank the adolescents who participated in this study, the massage therapists from Educating Hands and the Florida Institute of Massage Therapy, the staff at the Renfrew Treatment Center, and Lia Haley, Olga Quintino, and Julie Malphurs for helping with data collection. This research was supported by an NIMH Research Scientist Award (#MH00331) and an NIMH Research Grant (#MH46586) to Tiffany Field, and a grant from Johnson & Johnson to the Touch Research Institute.

Saul Schanberg, M.D., Ph.D., and Cynthia Kuhn, Ph.D., Duke University Medical School.

Tory Field, Karen Fierro, Tanja Henteleff, and Cynthia Mueller, Touch Research Institute, University of Miami School of Medicine.

Regina Yando, Ph.D., Harvard Medical School.

Seana Shaw, M.D., Department of Psychiatry, University of Miami School of Medicine.

Iris Burman, Educating Hands Institute.