Oxytocin is a phylogenetically ancient hypothalamic hormone with diverse influences on the brain and body (1,2). There is an ever increasing understanding of oxytocin as a hormone of importance for the well-being of the organism, both psychologically where it promotes tranquillity, attachment, concentration and memory, and physically where it promotes tissue healing and decreases pain, blood pressure and heart rate (3). Many of these effects can be characterised as anti-stress effects and some consider oxytocin an anti-stress hormone (3).
In clinical practice, the main foci of oxytocin research have been female reproduction and mother - infant bonding. Little is known of the importance of oxytocin in health and disease in children. It has been shown that children with recurrent abdominal pain of psychosomatic origin have low concentrations of plasma oxytocin in samples obtained after an overnight fast (4). It has been speculated that this decrease could be an expression of the chronic stress from which they are suffering. This idea is supported by other data showing a decrease in plasma cortisol in this group and a positive correlation between oxytocin and cortisol (4), since low plasma cortisol in plasma may be an indicator of chronic negative stress (5).
Oxytocin has both peripheral and central effects. Plasma oxytocin is a direct reflection of its effects peripherally. Since the nonapeptide oxytocin does not readily cross the brain barrier (1), oxytocin in plasma probably does not reflect to any substantial degree what is happening in the central nervous system.
This study sought to evaluate the plasma oxytocin concentration in children with recurrent abdominal pain of either psychosomatic origin or organic origin and to test the hypothesis that chronic negative stress is associated with decreased plasma oxytocin concentration. The specific aims of the study were: to determine whether children with recurrent abdominal pain of psychosomatic origin have lowered plasma oxytocin concentration; to measure plasma oxytocin in other groups of children with recurrent abdominal pain of non-psychosomatic origin; to study the concentration of plasma cortisol in a subgroup of children with psychosomatic abdominal pain and to study the relationship of cortisol and oxytocin in these children; to study the relationship between the recurrent pain, other associated symptoms and oxytocin; and to evaluate the plasma oxytocin in children with psychosomatic abdominal pain one year after consultaton.
MATERIALS AND METHODS
Classification of Recurrent Abdominal Pain
Children with recurrent abdominal pain can be classified according to the recommendations of von Baeyer and Walker (6), which are derived from the Apley criteria (7). This classification system allows for the identification of subgroups of children with recurrent abdominal pain such as those with stress-induced psychosomatic RAP (8), the major focus of this study. In international diagnostic systems such as Diagnostic and Statistical Manuel of Mental Disorders (DSM) IV, stress-induced RAP would be classified as a somatoform disorder and in the International Statistical Classification of Diseases (ICD)-10-CM as pain disorder associated with psychological factors. According to the Rome II criteria for gastrointestinal disorders, it would be diagnosed as functional recurrent abdominal pain (9). However, since recurrent abdominal pain caused by stress is not clearly defined in any of these systems, none of these classifications were used here.
Groups and Subgroups of Children with Oxytocin Measurements
Forty-eight children aged 6 to 15 years (mean, 9.6 years; 12 boys, 36 girls) attending our center for consultation about recurrent abdominal pain who fulfilled von Baeyer and Walker's classification were identified (6). Of these children, 32 had abdominal pain of psychosomatic origin (psychosomatic subgroup) and 7 had probable psychosomatic pain. Both subgroups were diagnosed according to published criteria (8). Nine of these 48 children had a non-psychosomatic diagnosis (non-psychosomatic subgroup), 5 with Helicobacter pylori gastritis, 2 with lactose intolerance, 1 with constipation, and 1 with pain of unknown origin. These 48 children were selected from a group of 100 consecutive cases of children with recurrent abdominal pain presented in another article (8). The first 60 of these 100 children were asked to participate in a hormone analysis in connection with having a test for H. pylori serology and all but 5 accepted. The results of H. pylori serologies have been published previously (10).
One year follow up of psychosomatic pain.
Twenty-five of the children with psychosomatic recurrent abdominal pain agreed to a second venipuncture one year after the first evaluation. These children and their parents were provided good care and general advice about the management of abdominal pain. All children received physiotherapy and half received psychological therapy.
A separate group of 15 children with inflammatory bowel disease (IBD) attending the Huddinge Child Hospital was selected for plasma evaluation, 4 with Crohn disease and 11 with ulcerative colitis (range, 6–19 years; mean, 12 years; 6 girls, 9 boys).
79 school children age 6–15 years (mean, 10.9 years; 49 girls, 30 boys) were enrolled by the school nurse.
Subgroups for plasma cortisol measurements.
Fasting morning plasma cortisol was measured in a subgroup of 23 of the patients with psychosomatic abdominal pain (16 girls, 7 boys; mean age, 9.5 years) and in a subgroup of 21 controls matched for age and gender with the psychosomatic pain patients (16 girls, 5 boys; mean age, 10.5 years).
A pain score was calculated for each child based on three parameters: intensity, measured with a Visual Analogue Scale (VAS), a valid instrument for children over six years of age (11); duration of attacks measured in minutes; and episode frequency per week. Each parameter was divided into three scores (Table 1) and the scores of each were added. Thus, a child could receive a maximum score of 9. Since muscular tender points associated with stress and pain is common in children with recurrent abdominal pain of psychosomatic origin (12,13), tender points measured according to published criteria (14), were counted. Muscles studied were the trapezoids, the subclavians, the lateral insertion of the greater pectorals, the proximal insertion of the brachioradialis, and the temporal muscles. In all there were 10 possible tender points.
After applying local anesthetic to the skin over the antecubital vein, 5 ml of blood was collected from fasting patients and controls in tubes containing 10 U/ml of heparin and 500 U/ml Trasylol. Blood was usually drawn between 8 a.m. and 9 a.m. and never later than 10 a.m. Samples were stored at −20°C until analysed. All children and parents gave informed consent.
Oxytocin was measured by radioimmunoassay after SEP-PAK extraction, using antibody EuroDiagnostica (Sweden) Oxytocin Antisera KA19, as described by Stock & Uvnäs-Moberg (1988) (15). The intra-and interassay coefficient of variation were 14% and 15%. Cortisol (free plus CBG-bound) was measured by AutoDELFIATM (Dissociation-Enhanced Lanthanid FluoroImmunoAssay, Wallac Oy, Finland).
Uni- and bivariate statistical analyses were performed using the Statistica computer program, version 6. Hormone levels are expressed as means +/− the 95% confidence interval. Differences between groups were determined using the Mann-Whitney U-test. The Wilcoxon matched paired test was used when comparing oxytocin concentrations at time zero and after one year. Spearman's correlation test was used to study correlation. Discriminant analysis was also done on Statistica computer program version 6.
The study was approved by the Ethics Committee of the Huddinge Hosptial (No 35/96).
The mean oxytocin concentration in the 48 patients referred for recurrent abdominal pain was 36.7 pmol/L (95% confidence interval 29.0–43.2). The subgroup of 32 children with psychosomatic abdominal pain had a mean oxytocin concentration of 30.5 pmol/L (95% confidence interval 24.6–36.5), which was significantly lower than the control mean of 45.0 pmol/L (95% confidence interval 41.6–48.4), (P < 0.0001). The 15 patients with inflammatory bowel disease had a mean plasma oxytocin concentration of 35.0 pmol/L (95% confidence interval 20.6–49.1), which was also significantly lower than controls (P < 0.001), but was not difference from the children with psychosomatic abdominal pain (P =.3) (Fig. 1 and Table 2).
The 26 children with psychosomatic abdominal pain analysed after a year had a mean plasma oxytocin of 38.7 pmol/L (95% confidence interval 30.6–46.7), which was significantly higher that the mean obtained one year previously (P < 0.05) (Table 2).
The subgroup of children with psychosomatic abdominal pain in whom cortisol levels were measured had a lower mean plasma cortisol concentration of 263 nmol/L (confidence interval 232–293) compared to the subgroup of controls whose mean cortisol was 313 nmol/L (confidence interval 264–361). The difference between these means was not significant (P = 0.08). The cortisol and oxtyocin levels correlated positively in children with psychosomatic abdominal pain (R = 0.45; P < 0.05), but not in the subgroup of control children (R = 0.15; P = 0.5) (Fig. 2).
In a discriminant analysis for the variables oxytocin and cortisol, oxytocin discriminated 52% of the patients with psychosomatic abdominal pain and 90% of the controls. Plasma cortisol added only 2.5% and 2.7%, respectively to the discriminatory ability.
No correlation was found between oxytocin level and pain score (R = 0.12; P = 0.49) nor the number of tender points (P = 0.5). After one year, no difference was found in the oxytocin concentration in the psychosomatic group between 8 children with a pain score of zero and 11 with a pain score >2 (P = 0.88). When the patients with psychosomatic pain were combined with the group of probable psychosomatic pain (n = 39), no difference in the oxytocin concentration was found between children with and without headache (P = 0.30) or those with and without tender points (P = 0.51). There was no difference in the oxytocin levels of 13 children with and 22 children without reported loss of appetite (P = 0.97), nor between 23 children with and 8 children without pain accompanying meals (P = 0.81).
This study shows that children with abdominal pain of psychosomatic origin have plasma oxytocin levels less than control, a finding in accordance with an earlier study (4). The significance of this finding is not clear and can be interpreted in several ways. It is thought but not proven that heredity (16) and early life experience are of importance (16,17). Can the decrease in oxytocin shown in this study be related to stress and pain? Oxytocin is important for attachment and in anti-stress reactions (3). In chronic negative stress, the organism reacts with withdrawal and detachment (18). Not much is known about how oxytocin is regulated in such a state. One possibility, supported by the findings in this study is that oxytocin in children may decrease in response to chronic stress and detachment. This explanation is supported by the observed decrease in cortisol and its correlation with oxytocin levels which may also be a response to chronic stress (5,19). The increase in oxytocin after a year may reflect improvement in psychosomatic stress. Another less likely possibility is that pain itself triggers the lowering of the oxytocin concentration. The low concentration of oxytocin could be an expression of a hereditary trait of infantile origin. Recurrent abdominal pain is known to cluster in families (17), but we do not know if a low concentration of oxytocin is genetic. The increase of oxytocin after one year supports the view that the low level is due to stress and that the low initial level is due to state, not trait.
The oxytocin levels in other groups of children with abdominal pain showed variable results. The children with inflammatory bowel disease, a condition which is often associated with abdominal pain, were low compared to controls and were similar to those of the children with psychosomatic pain. The subgroup of nine children with non-psychosomatic recurrent pain had normal oxytocin levels. These groups were small, so interpretations must be drawn with caution.
The reason for the low oxytocin levels in the patients with inflammatory bowel disease is not clear. No relationship is known between inflammation and oxytocin secretion. Pain more often increases than decreases oxytocin (20). We speculate that inflammatory bowel disease might be associated with stress similar and thus result in lower oxytocin levels.
The determination of plasma oxytocin may be of some help in evaluating recurrent abdominal pain syndromes. IBD can be excluded with specific diagnostic testing (21).
The cortisol concentration measured in a subgroup of children with psychosomatic abdominal pain were not statistically different from controls. Cortisol and oxytocin showed a positive correlation, and significantly so in the subgroup of psychosomatic children. Ten years ago, it was shown for the first time that plasma cortisol in morning samples was decreased in a group of children with psychosomatic abdominal pain. At that time, the conclusion that low cortisol was a reflection of negative stress was met with scepticism. However, data indicating that cortisol secretion decreases in negative chronic stress (19,22) are accumulating. Our understanding of the regulation of the CRF-ACTH-cortisol axis in stress is still unclear and contradictory (22). Theoretically, the lowering of cortisol can be explained by many mechanisms such as changes in sensitivity of the cortsiol receptor in the CNS (17), or changes in the amygdala governed regulation of CRF in hypothalamus (23), which better explains the correlation we noted between cortisol and oxytocin levels. The amygdala, the center of stress regulation and of hypothalamic hormone secretion, could well be the governing center controlling the output of both oxytocin and CRF-ACTH-cortisol.
No relationship was found between oxytocin concentration and pain score, muscular tender points, loss of appetite or headache. This is somewhat surprising as we predicted that the amount of pain and number of symptoms should reflect the amount of negative stress in children with psychosomatic pain.
There may be a relationship between recurrent abdominal pain of psychosomatic origin and plasma oxytocin concentration perhaps as a reaction to chronic negative stress. Measurement of plasma oxytocin may be of help in differential diagnosis of recurrent abdominal pain. Research on organic disorders characterized by abdominal pain are needed to clarify the relationship between pain, stress and oxytocin level.
The author wishes to thank Kerstin Uvnäs-Moberg and Tor Lindberg for fruitful discussions, Jan Ejderham for collecting the IBD blood samples, and Renée Andersson for doing the oxytocin analysis. The author also wishes to thank Psykosomatisk Klinik och Forskning (PKF), FRF and Förstamajkommittén for their financial support of this study. The language has been checked by translator Isaac Austin and the statistics by statistician Göran Granath.
1. Gimpl G, Fahrenholtz F. The Oxytocin receptor system: structure, function and regulation. Physiological Review 2001;81:629–83.
2. Van de Kar LD, Blair ML. Forebrain pathways mediating stress-induced hormones secretion. Frontiers in neuroendocrinology 1999;20;1–148.
3. Uvnäs-Moberg K. Antistress pattern induced by oxytocin. News Physiol Sci 1998;13:13–26.
4. Alfvén G, de la Torre B, Uvnäs-Moberg K. Depressed concentration of cortisol and oxytocin in children with recurrent abdominal pain of non-organic origin. Acta Paediatr 1994;83:1076–80.
5. Heim C, Ehlert U, Hellhammer DH. The potential role of hypo-cortisolism in the pathophysiology of stress-related bodily disorders. Psychoneuroendocrinology 2000;25;1–35.
6. Baeyer CL, Walker LS. Children with Recurrent Abdominal Pain: Issues in the Selection and Description of Research Participants. Development and Behavioral Pedatrics 1999;20:307–13.
7. Apley J. The child with abdominal pains. Oxford: Blackwell, 1975.
8. Alfvén G. One hundred cases of recurrent abdominal pain in children. Diagnostic procedures and a presentation of eight criteria for psychosomatic diagnosis. Acta Paediatr 2003;92:43–50.
9. Rasquin-Weber A, Hyman PE, Cucchiara S, et al. Childhood functional gastrointestinal disorders. Gut 1999;45(Suppl II):II60–8.
10. Cassall H, Alfvén G, Drapinski M, Bergström M, Dahlström KA. One week treatment with Omeprazol, Clarithromycin, and Metronidazole in children with Helicobacter pylori infection. J Pediatr Gastroenterol Nutr 1998;27:415–8.
11. McGrath P. Pain in Children. New York: The Guildford Press, 1990.
12. Alfvén G. Preliminary findings of increased muscle tension and tenderness and recurrent abdominal pain of non-organic origin. A clinical study. Acta Paediatr 1993;82:400–3.
13. Alfvén G. Psychosomatic pain in children. A psycho-muscular tension reaction?European Journal of Pain 1997;1:1–10.
14. Wolfe F, Smythe HA, Yunus MB, et al. The American College of Rheumatology 1990 criteria for classification of fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum 1995; 38:19–28.
15. Stock S, Uvnäs-Moberg K. Increased plasma levels of oxytocin in response to afferent electrical stimulation of the sciatic and vagal nerves and in response to touch and pinch in anaesthetized rats. Acta Physiol Scand 1988;132:29–34.
16. Henry PJ, Wang S. Effects of early stress on adult affiliative behavior. Psychoendocrinology 1998;23:863–75.
17. Champagne F, Meaney MJ. Like mother, like daughter: evidence for non-genomic transmission of parental behavior and stress responsitivity. Prog Brain Res 2001;133:287–302.
18. Alfvén G. Recurrent abdominal pain of non-organic origin in childhood. Clinical, muscular, epidemiological, endocrinological, and etiological aspects. [thesis]. Karolinska institute, 1993.
19. Yehuda R. Sensitization of the hypthalamic-pituitary-adrenal axis in posttraumatic stress. Ann N Y Acad Sci 1997:21;821:57–75.
20. Brown DC, Perkowski S. Oxytocin content of the cerebrospinal fluid of dogs and its relationship to pain induced by spinal cord compression. Vet Surg 1998;27:607–11.
21. Olafsdottir E, Aksnes I, Fluge G, Bertad A. Faecal calprotectin levels in infants with infantile colic, healthy infants, children with inflammatory bowel disease, children with recurrent abdominal pain and healthy children. Acta Paediatr 2002;91:45–50.
22. Ehlert U, Gaab J, Heinrichs M. Psychoneuroendocrinologal contributions to the etiology of depression, posttraumatic stress disorder, and stress-related bodily disorders: the role of the hypothalamus-pituitary-adrenal axis. Biological Psychology 2001;57:141–52.
23. Gray TS, Bingaman E. The amygdala: corticotropin-realsing factor, steroids, and stress. Critical Reviews in Neurobiology 1996; 10:155–68.
© 2004 Lippincott Williams & Wilkins, Inc.