Frequency of recovery from anorexia nervosa of a cohort patients re-evaluated on a long-term basis following intensive care

Transcription

1 ORIG I NAL RESEARCH PAPER Frequency of recovery from anorexia nervosa of a cohort patients re-evaluated on a long-term basis following intensive care L. Foppiani*, L. Luise**, E. Rasore**, U. Menichini**, and M. Giusti* *DiSEM, Cattedra di Endocrinologia, and **Cattedra di Psichiatria, Università di Genova, Italy ABSTRACT. The need to treat the acute onset of anorexia nervosa in a specialty unit is wel known, but nevertheless, even clinically recovered patients show a high rate of relapse. The aim of our study was to re-evaluate various clinical (hemoglobin, transferrin, insulin-like growth factor I, TSH, PRL, gonadotropins and 17β-estradiol) and psychiatric [semistructured interview Eating Disorder Inventory (EDI) Test] parameters in 19 female anorexic patients hospitalized in our department from 1983 to 1993 (with a 9-year median prior to the present study) for the treat ment of anorexia nervosa, and to compare these results with those of a previous follow-up per formed on the same subjects in In the present evaluation, no significant variation was found among the clinical, nutritional and hormonal parameters when compared to those of the first follow-up in which all parameters had improved with the exception of the PRL levels, which were significantly low. Meanwhile, the percentage of patients with spontaneous menses increased significantly from 50% to 70%, while the number of patients on psychopharmacologi cal therapy decreased significantly throughout the study. Furthermore the percentage o patients with altered (severe or mild) EDI profiles decreased to 50%. This study emphasizes the positive prognostic role of hospitalization and intensive care in a cohort of anorexic patients. The present study, in addition to demonstrating both a general maintenance of body weigh acquired over the years, albeit in the lower normal range, and an increase in the percentage o patients with a regular menstrual cycle, also highlights the persistence of psychiatric abnormali ties in a large number of patients, even in those diagnosed as clinically recovered, thus sug gesting the need for long-term psychiatric care. (Eating Weight Disord. 3, 90-94, 1998). 1998, Editrice Kurtis Key words: Anorexia nervosa, follow-up, endocrine profile, psychiatric evaluation. Correspondence: Massimo Giusti, M.D., Dipartimento di Scienze Endocrine e Metaboliche (DiSEM), Cattedra di Endocrinologia, Università di Genova, Viale Benedetto XV, 6, Genova, Italy. INTRODUCTION Anorexia nervosa (AN) is a chronic disease whose diagnostic symptoms include a body weight of 15% below the ideal body weight (IBW), an intense fear of gaining weight or becoming fat, even if underweight, amenorrhea, and a disturbance in the way the patient perceives her body weight, size or shape (1). It is well known that AN is a multifactorial disorder, in which biological, endocrine, psychiatric and socio-cultural factors interplay (2). The correct management of the disease is still being debated. Although day-care programs are gaining in increased recognition (3), most patients with significant body weight loss having a Body Mass Index (BMI) less than 17 Kg/m 2 must be hospitalized in a specialty unit (4). Long-term followup studies, lasting 10 to 20 years, have shown that although clinical parameters improve in many patients, psychiatric and behaviora alterations continue to persist (5-8). The aim of this study was to re-evaluate nutritional hormonal and psychiatric parameters in a cohort of anorexic patients hospitalized in the Department of Endocrine and Metabolic Sciences (DISEM), during the past 15 years to compare the results with a follow-up per formed 5 years earlier, and to determine any possible relation between the nutritional and hormonal status and psychopatologica development. METHODS Subjects and study design From 1983 to 1998, 58 female patients (21.1±0.6 years, ±SD, range years) with 90

2 Follow-up of anorexia nervosa TABLE 1 Some clinical, nutritional and hormonal parameters of the 19 anorexic patients studied. Weight BMI IBW T3 T4 TSH E2 IGF1 PRL Transferrin Hb (Kg) (Kg/m 2 ) (%) (ng/ml) (ng/ml) (μul/ml) (pg/ml) (ng/ml) (pg/ml) (mg/dl) (g/dl) Admission *37.1± ± ±1.6(A) ^0.8±0.1 ^^^65.3± ± ± ±13.6(a) ±28.0(e) 12.0±0.2(g) Discharge **41.3± ± ±6.0(B) ±0.4(h) 1 st follow-up ***49.1± ± ±4.8(C) ^^1.0±0.1 ^^^^78.0±4 1.1± ± ±14.4(b) 9.0±1.7(c) 307.1±30.8(f) 13.0±0.2(i) 2 nd follow-up 48.2± ± ± ± ± ± ± ± ±0.8(d) 287.8± ±0.3 **p<0.01 vs.*, (B) p<0.01 vs. (A), p<0.05 vs., (f) p<0.05 vs. (e); ***p<0.01 vs.**, (C) p<0.01 vs. (B), (b) p<0.05 vs. (a), (h) p<0.01 vs. (g) & (i); p<0.01 vs., ^p<0.05 vs.^^, (d) p<0.05 vs. (c) p<0.01 vs., ^^^^p<0.05 vs.^^^ AN according to DSM IV (1) were hospitalized in the DISEM. At the time of admission, average body weight was 38.4±0.4 Kg, the underweight versus IBW ratio was %±2.0% and the known duration of disease was 7.9±0.5 years. Blood samples were collected and assays performed to determine nutritional (haemoglobin, transferrin) and hormonal [T3, T4, TSH, PRL, LH, FSH, 17β-oestradiol (E2), insulin-like growth factor I (IGF-I)] levels. During hospitalization, the patients received enteral ( kcal/day) plus parenteral (500 kcal/day) nutrition or only parenteral nutrition ( kcal/day). On discharge, a dietary program in agreement with the patient and trained dieticians was provided. This program aimed to ensure an initial minimal caloric intake of Kcal/day with a mean composition of 60% carbohydrates, 23% lipids and 17% proteins. Every dietary program was personalized to the patient needs and involved the progressive addition of new foods in agreement with the patient. Caloric integrators were also added if accepted. The patients were first followed up on a weekly basis until a progressive and TABLE 2 Percentage of patients with spontaneous menses, low gonadotropin levels and on-therapy at the different times of the study. Spontaneous Low gonadotropin Psycopharmacological menses levels therapy Admission 0% 87.5% 57.8% 1 st follow-up 52.8% 41.1% 26.3% 2 nd follow-up 71.4% 28.6% 15.8% X 2, p<0.001, p<0.01, p<0.05 persistent weight gain was recorded. Afterwards, patients were seen either on a monthly or on a bimonthly basis. At the first follow-up study, performed in 1993, no criteria for selection was used, and 30 anorexic patients belonging to the cohort, who had been admitted to our department 5-15 years earlier (with a 9-year median) and a hospitalization period of days (with a 51-day median), were contacted and 19 agreed to participate. The second follow-up was performed at the end of 1997 on the same patients (27.9±5.0 years, ±SD, range years) of the first follow-up. Included in both studies was the evaluation of nutritional and hormonal levels as at the time of admission, a semi-structured interview, and a specific test for eating disorders, The Eating Disorder Inventory Test (EDI) (9-11). All the patients gave their informed consent to take part in the study which was approved by the Local Ethics Committee. Endocrine and nutritional assays Due to considerable changes in the assay method (i.e., RIA to IRMA) and sensitivity, the gonadotropin levels during the followup periods of the study could not be compared. At the time of admission, LH and FSH were assayed and reported (12). During the first and second follow-up periods, LH, FSH, T3, T4, TSH, E2, PRL, IGF1, transferrin and hemoglobin were assayed and reported (10, 13). In patients with spontaneous menses, blood samples for E2 were collected during the early (from the 4 th to the 7 th day) follicular phase. Psychiatric evaluation and management During hospitalization and after discharge, psychopharmacological therapy included 91

3 L. Foppiani, L. Luise, E. Rasore, et al. mainly antidepressant drugs (serotoninergics and tricyclics) and benzodiazepines to a lesser extent. The psychotherapeutic approach involved cognitive behavioral therapy performed by a psychiatrist and aimed to help patients recognize the connection between their dysfunctional thoughts and maladjusted behavior, and also permitted the patient to evaluate critically own attitudes and beliefs about weight and eating. The semi-structured interview evaluated the various aspects of the patient s history, family and treatment of the disease as well as current conditions and addressed not only an objective diagnosis but also the patient s subjective perception of change. The EDI questionnaire is widely used to measure various aspects of psychopathology linked to eating disorders. The current version includes 11 subscales: 3 scales to evaluate behavior and attitude towards body weight, food-intake and body shape, and 8 to scales evaluate general psychological aspects (11). Statistics Data are expressed as mean±sem, if not specified otherwise. Evaluation of the hormonal and nutritional data was performed by analysis of variance (ANOVA) followed by the Bonferroni test and the paired Student s t-test. Evaluation of the percentage of patients on estro-progestin therapy, with low gonadotropin levels and EDI score was performed by χ 2. Statistical significance was assumed at p<0.05. RESULTS Nutritional and endocrine parameters Some clinical, nutritional and hormonal parameters of anorexic patients during the different phases of the study are shown in Table 1. Body weight and BMI increased significantly (p<0.01) from the time of admission to discharge (Table 1). On discharge, the delta (Δ) of weight gain was significantly higher (p<0.05, 16.9%±3.7%) in patients (n=11; first group) who received enteral and parenteral nutrition than in those (n=8; second group) who received only parenteral nutrition (6.3%±1.9%). At the first follow-up, body weight and BMI had significantly (p<0.01) increased since admission (Table 1) and at discharge no significant difference was found in the Δ of weight gain between the first (25.1%±11.0%) and the second group (15.3%±8.4%) of patients. Hemoglobin values at the time of discharge were significantly (p<0.01) lower than on admission and were significantly (p<0.01) higher at the first follow-up. Transferrin, T3 and IGF1 values were significantly (p<0.05) higher at the first follow-up than at the time of admission (Table 1). On the second evaluation, no significant difference in the Δ of weight gain from the first follow-up was found between the first (-3.9%±5.1%) and the second group (3.6%±4.7%) of patients nor was any significant variation found either in the nutritional or endocrine parameters (Table 1). In the first follow-up 2 out of 19 (10.5%) patients had increased PRL levels (>20 ng/ml), which had normalized by the second evaluation. The percentage of patients not treated with estro-progestins, who had spontaneous menses, increased significantly (χ 2 =17.55, p<0.001) from admission to the second follow-up (Table 2). This finding followed quite closely the decrease (χ 2 =11.9, p<0.01) in the percentage of patients not on therapy with estro-progestins who had gonadotropin levels below or near the lower reference value (Table 2). Psychiatric parameters Analysis of the semi-structured interview results showed a significant presence of psychiatric disease in the families of 9 out of 19 (47.4%) patients and eating disorders in 6 out of 19 (31.6%). Most patients (8 out of 19) sought first help from their family doctor, while 4 out of 19 turned to endocrinologists, 3 out of 19 to gynecologists, 2 out of 19 to neurologists, and 1 out of 19 to a dietician or psychiatrist. While 3 out of 19 (15.8%) patients claimed to be currently cured of AN, 10 out of 19 (52.6%) patients claimed to have had a significant improvement in both clinical and psychological conditions, 4 out of 19 (21.1%) believed their condition to be unchanged and 2 out of 19 (10.5%) to have a bulimic condition. In addition 5 out of 19 (26.3%) patients still complained of affective disorders such as anxiety and depression and 4 out of 19 (21.0%) were on psychother - apy. Only 2 out of 19 (10.5%) patients treated judged the therapy as negative. The percentage of patients on psychopharmacological therapy (χ 2 =8.21, p<0.05) decreased significantly from the time of admission to the second follow-up (Table 2). At the second follow-up, the diagnosis was typical anorexia 92

4 Follow-up of anorexia nervosa nervosa in 8 patients, bulimia nervosa in 2 patients, and in 9 patients atypical anorexia nervosa (i.e., persistence of an anorexic disturbance with a weight loss of less than 15%). EDI results showed severely altered profiles in 4 out of 19 (21.0%) patients, mildly altered profiles in 6 out of 19 (31.6%) patients and normal profiles in 9 out of 19 (47.4%) patients. The most marked items were interpersonal distrust, a drive for thinness, inadequacy and perfectionism. The percentage of patients with severely or mildly altered EDI profiles decreased, although not significantly (χ 2 =3.04, p=0.08), from 84.2% to 52.6% from the first follow-up to the current followup. Furthermore no correlation was found between the IGF1 and PRL levels and the EDI scores (data not shown). DISCUSSION AN is a chronic disease whose correct management involves different specialties; indeed long-term mortality of anorexic patients is significantly lower when treated in a specialized unit rather than in a general psychiatric unit (4). Patients with severe emaciation (BMI less than 17 Kg/m 2 ), must be hospitalized to treat severe medical complications (1). In a previous study involving a cohort of anorexic patients re-evaluated a few years after hospitalization, we found a significant improvement in both the nutritional and hormonal levels, even if significant weight gain occurred in most patients after discharge, only about 50% of the patients had spontaneous menses (10) and reached 70% in this study, at a weight which was considerably similar to the first followup, at present the reason for these results cannot be confirmed. However, it is well known that the hypothalamic-pituitarygonadal (HPG) axis needs a prolonged period of stable body weight in order to regain its function, if indeed it does (14). A second possible explanation is that the number of patients who were on psychopharmacological therapy (e.g. antidepressants and anxiolitic drugs) decreased significantly throughout the study and it is well known that these drugs can affect the HPG axis by increasing the PRL levels, which, in turn, impair GnRH release from the hypothalamus (15, 16). In the current follow-up, no patient had increased PRL levels and mean PRL levels had decreased significantly in both studies (even though they remained within the normal range). Support also exists for the hypothesis that neurohormonal factors could be responsible for the fact that 30% of the patients, who were not on therapy with estro-progestins, were still amenorrheic at the present follow-up. Corticotropin releasing hormone (CRH) could also play an important role; indeed, CRH is able to impair the release of GnRH from the hypothalamus through the oppiod system, and increased CRH levels have been found in the cerebrospinal fluid of anorexic patients (17). In the patients studied, T3 and transferrin levels increased significantly from the time of admission to the first follow-up, and remained nearly unchanged at the second follow-up. It is well known that in a condition of undernourishment, as a consequence of a hypometabolic state, low- T3 syndrome (18, 19) and reduced transferrin levels (20) are usually present, but increase with weight gain (19, 20) as in our patients. IGF1 is another important endocrine marker of the nutrition state, and it is usually reduced in AN (21). In the patients studied, reduced IGF1 levels at the time of admission followed variations in body weight quite closely, and increased significantly at the first follow-up and then decreased slightly. The percentage of altered (severe or mild) EDI profiles decreased, though not significantly, from the first to the second follow-up, with 20% of the patients still showing severely altered profiles. The factor most seriously affecting EDI results in these patients concerned social relationship, thereby suggesting the image of a different AN, is more at odds with the external world than with body shape. The fact that a significant number of patients have a normal EDI profile, irrespective of clinical alterations (although improved), might depend on a real reduction in symptoms. This perhaps may be attributed not so much to the treatment received by the patient, but to self deception, which probably enables the patient to adapt to their own environment fairly well. Our study shows that hospitalization with intensive care constitutes a positive prognostic factor for both physical and psychological conditions in AN. Furthermore, an improvement in the physical condition is essential if psychiatric treatment is to be effective. This statement is supported by the significant reduction in the percentage of patients on 93

5 L. Foppiani, L. Luise, E. Rasore, et al. psychopharmacological therapy throughout the study, following improvement in clinical and endocrine conditions. In conclusion, our study shows a general improvement in nutritional, hormonal and psychiatric parameters over the years in a cohort of anorexic patients previously hospitalized, as already reported (5-8). However, the persistence of psychiatric alterations in a significant percentage of patients, including those defined as clinically recovered, justifies continued long-term care due to the risk of relapse of the disease. REFERENCES 1. Garner D.M.: Pathogenesis of anorexia nervosa. Lancet, 341, , Herzog D.B., Copeland P.M.: Eating disorders. N. Engl. J. Med., 313, , Kaplan A.S.: Day hospital treatment for anorexia nervosa and bulimia nervosa. Eat. Dis. Rev., 2, 1-3, Beumont P.J.V., Russel J.D., Touyz S.W.: Treatment of anorexia nervosa. Lancet, 341, , Rosenvige J.H., Mouland S.O.: Outcome and prognosis of anorexia nervosa: a retrospective study of 41 subjects. Br. J. Psychiatry, 156, 92-97, Eckert E.D., Halmi K.A., Marchi P., Grove W., Crosby R.: Ten years follow-up of anorexia nervosa: clinical course and outcome. Psychol. Med., 25, , Herzog D.B., Nussbaum K.M., Marmor A.K.: Comorbidity and outcome in eating disorders. Psychiatr. Clin. North Am., 19, , Strober M., Freeman R., Morrel W.: The long-term course of severe anorexia nervosa in adolescents: survival analysis of recovery, relapse, and outcome predictors over years in a prospective study. Int. J. Eat. Disord., 22, , Rasore E., Gabrielli F., Traversa C., Uva V., Foppiani L., Giusti M.: On mental anorexia and its therapy: clinical considerations and catamnestic examinations during a period of psychiatric counselling. 2nd International Rome Symposium on Eating Disorders: Advancement in Diagnosis and Treatment of Anorexia Nervosa, Bulimia and Obesity, Rome January 1993, Proceedings Book, p Foppiani L., Uva V., Falivene M., Valenti S., Traversa C., Rasore E., Giusti M.: Anoressia nervosa: follow-up di parametri nutrizionali, ormonali e psichiatrici. Minerva Med., 86, , Garner D.M.: Eating Disorder Inventory-2: Professional manual, psychological assessment resources. Odessa FL, Giusti M., Torre R., Traverso L., Cavagnaro P., Attanasio R., Giordano G.: Endogenous opiod blockade and gonadotropin secretion: role of pulsatile luteinizing hormone-releasing hormone administration in anorexia nervosa and weight loss amenorrhea. Fertil. Steril., 49, , Giusti M., Foppiani L., Ponzani P., Cuttica C.M., Falivene M.R., Valenti S.: Hexarelin is a stronger GH-releasing peptide than GHRH in normal cycling women but not in anorexia nervosa. J. Endocrinol. Invest., 20, , Kotsuji F., Kubo M., Takeuchi Y., Tominaga T.: Alternate-day GnRH therapy for ovarian hypofunction induced by weight loss: treatment of six patients who remained amenorrhoeic after weight gain. Clin. Endocrinol. (Oxf), 39, , Bohnet H.G., Dahlen H.G., Wuttke W., Schneider H.P.: Hyperprolactinemic anovulatory syndrome. J. Clin. Endocrinol. Metab., 42, , Bouchard P., Lagoguey M., Brailly S., Schaison G.: Gonadotropin-releasing hormone pulsatile administration restores luteinizing hormone pulsatility and normal testosterone levels in males with hyperprolactinemia. J. Clin. Endocrinol. Metab., 60, , Hotta M., Shibasaki T., Masuda A., Imaki T., Demura H., Ling N., Shizume K.: The responses of plasma adrenocorticotropin and cortisol to corticotropin-releasing hormone (CRH) and cerebrospinal fluid immunoreactive CRH in anorexia nervosa. J. Clin. Endocrinol. Metab., 62, , Docter R., Krenning E.P., De Jong M., Hennerman G.: The sick euthyroid syndrome: changes in thyroid hormone serum parameters and hormone metabolism. Clin. Endocrinol., 39, , Kiyohara K., Tamai H., Takaichi Y., Nakagawa T., Kumagai L.F.: Decreased thyroidal triiodothyronine secretion in patients with anorexia nervosa: influence of weight recovery. Am. J. Clin. Nutr., 50, , Matsubara M., Koyanagawa Y., Odagaki E., Nakagawa K.: Plasma transferrin levels in abnormal endocrine states. II: The changes in various endocrine states. Horm. Metab. Res., 21, , Muller EE.: The control of somatotropic function. Physiol. Rev., 67, ,