PEDIATRICS Vol. 120 No. 1 July 2007, pp. e129-e137 (doi:10.1542/peds.2006-2759) ARTICLE Abnormal Thermoregulatory Responses in Adolescents With Chronic Fatigue Syndrome: Relation to Clinical Symptoms Vegard Bruun Wyller, MDa,b, Kristin Godang, BScc, Lars Mrrkrid, MD, PhDd, Jerome Philip Saul, MDe, Erik Thaulow, MD, PhDa and Lars Wallre, MD, PhDb a Departments of Pediatrics c Endocrinology d Medical Biochemistry, Rikshospitalet-Radiumhospitalet Medical Center, Oslo, Norway b Department of Physiology, University of Oslo, Oslo, Norway e Department of Pediatrics, Medical University of South Carolina, Charleston, South Carolina OBJECTIVES. Chronic fatigue syndrome is a common and disabling disease of unknown etiology. Accumulating evidence indicates dysfunction of the autonomic nervous system. To further explore the pathophysiology of chronic fatigue syndrome, we investigated thermoregulatory responses dependent on catecholaminergic effector systems in adolescent patients with chronic fatigue syndrome. PATIENTS AND METHODS. A consecutive sample of 15 patients with chronic fatigue syndrome aged 12 to 18 years and a volunteer sample of 57 healthy control subjects of equal gender and age distribution were included. Plasma catecholamines and metanephrines were measured before and after strong cooling of 1 hand. Acral skin blood flow, tympanic temperature, heart rate, and mean blood pressure were measured during moderate cooling of 1 hand. In addition, clinical symptoms indicative of thermoregulatory disturbances were recorded. RESULTS. Patients with chronic fatigue syndrome reported significantly more shivering, sweating, sudden change of skin color, and feeling unusually warm. At baseline, patients with chronic fatigue syndrome had higher levels of norepinephrine, heart rate, epinephrine, and tympanic temperature than control subjects. During cooling of 1 hand, acral skin blood flow was less reduced, vasoconstrictor events occurred at lower temperatures, and tympanic temperature decreased more in patients with chronic fatigue syndrome compared with control subjects. Catecholamines increased and metanephrines decreased similarly in the 2 groups. CONCLUSIONS. Adolescent patients with chronic fatigue syndrome have abnormal catecholaminergic-dependent thermoregulatory responses both at rest and during local skin cooling, supporting a hypothesis of sympathetic dysfunction and possibly explaining important clinical symptoms. Key Words: chronic fatigue • thermal regulation • adolescents • pathogenesis Abbreviations: CFS—chronic fatigue syndrome • AVA—arteriovenous anastomosis • CDC—Centers for Disease Control and Prevention • ASBF—acral skin blood flow • TT—tympanic temperature • HR—heart rate • MBP—mean blood pressure • CI—confidence interval Accepted Jan 25, 2007. DISCUSSION The most important findings of this study are that (1) patients with CFS report several symptoms that might indicate thermoregulatory disturbances; (2) at baseline, patients with CFS have higher levels of norepinephrine and epinephrine and higher TT than control subjects; and (3) during cooling of 1 hand, the neuroendocrine responses are similar in the 2 groups, but ASBF is less reduced among patients with CFS, whereas the baseline differences in TT disappear. Furthermore, the relevance of these findings is strengthened by the strikingly homogeneous responses within the CFS group, creating significant differences from control subjects despite the small number of subjects studied. Baseline Observations The finding in this study of increased norepinephrine levels in patients with CFS seems to be novel, whereas increased levels of epinephrine have been reported sporadically.27,28 A higher level of acute emotional stress among patients with CFS might explain the epinephrine differences; however, plasma levels of norepinephrine are less influenced by such mechanisms.7 Thus, the findings seem to indicate a more substantial alteration of physiology. A high level of norepinephrine in the antecubital vein plasma might suggest increased sympathetic nerve activity to forearm skin and skeletal muscle.29 Likewise, a high plasma level of epinephrine might be a result of increased sympathetic nerve activity to the adrenals. However, there are several alternative explanations. Generally, high levels of plasma catecholamines could result from either increased spillover or reduced removal, which, in turn, depends on both sympathetic nerve activity, the capacity of different reuptake and breakdown pathways, and local blood flow.7,29 Furthermore, a high norepinephrine concentration in forearm venous blood might simply reflect increased arterial levels, which, in turn, could be because of enhanced spillover in other parts of the body. The plasma levels of metanephrines are not good markers of activity in either the adrenal medulla or the sympathetic neurons and are only weakly correlated with the plasma levels of the respective catecholamines.7,29 Thus, similar levels of metanephrines among patients with CFS and control subjects do not rule out a state of catecholamine excess in the former. The finding of increased TT in patients with CFS is in agreement with previous reports of increased skin temperature in this population30 but contrasts with 2 other studies that did not find any deviations in core body temperature.31,32 However, these latter studies focused primarily on alterations in circadian temperature rhythms. In this study, the increased TT might be partially caused by high levels of epinephrine, which increase basal metabolic rate and heat production.7 In addition, a tendency toward shivering, as reported in our patients with CFS, might contribute. Increased levels of thyroid hormone are an alternative explanation that has not been specifically addressed in this study; however, overt thyroid hyperfunction was ruled out in the patient group during routine clinical investigations. The high resting HR found in this study fits well with other studies, documenting similar hemodynamic abnormalities both at rest and during orthostatic stress.3,33 An abnormal sympathetic predominance of cardiovascular regulation is one possible interpretation of these results, which is consistent with the reduced ASBF found in the CFS group, because skin AVAs are strongly controlled by sympathetic neural activity.22,34 Taken together, our baseline observations might indicate a general enhancement of sympathetic nerve activity to different regions and organs, including the forearm, the adrenals, and the heart. Patients' report of shivering, sweating, and paleness further suggest enhanced sympathetic outflow to skeletal muscles, sweat glands, and skin arterioles, respectively, a possibility that should be the subject of further research. Observations During and After Cooling Strong, rapid cooling of 1 hand normally promotes a general enhancement of sympathetic nerve activity, causing increased plasma levels of both norepinephrine and epinephrine.7 The stimuli for this response are not only reduced skin temperature in the immersed hand but also painful sensations. The similar increase in norepinephrine and epinephrine among control subjects and patients with CFS, as documented in this study, suggests that patients with CFS have preserved response abilities toward local cooling within the sympathetic nervous system. Thus, there are no indications of gross autonomic neuropathy, as have been proposed by others.35 Although not the focus of this report, the significant reduction of normetanephrine and metanephrine on cooling observed in both groups is an interesting finding, which seems to be unique to this report. Moderate, slow cooling of 1 hand normally causes a gradual reduction of ASBF down to a certain temperature level, at which ASBF suddenly ceases, presumably because of a coordinated closure of all of the AVAs23 (Fig 1). Although the precise mechanisms behind this phenomenon have not been fully elucidated, one possible explanation is synthesis of local signal substances that increase the postsynaptic sensitivity for norepinephrine or antagonize neuronal reuptake.36,37 Still, the closure is also dependent on central sympathetic outflow.23 We have provided evidence of abnormally increased core body temperature at baseline among patients with CFS. If the part of the effector system that controls the AVAs functions normally, an adequate response would be to elicit heat loss, that is, ensuring preserved blood flow in the AVAs as the temperature in the water bath falls. Consequently, our observation of prolonged preservation of ASBF left in patients with CFS on cooling, which has not been reported before, might be interpreted as a normal regulatory response to abnormally increased core temperature at baseline. Accordingly, TT did fall in the CFS group during the experiment, indicating successful heat loss and normalization of core temperature, whereas the normal ASBF responses in the control subjects seemed to maintain TT constant in that group. Alternatively, our results might indicate a defect in local vasoconstrictor mechanisms. Interestingly, recent experimental studies applying acetylcholine to small skin areas have found stronger vasodilative responses among patients with CFS than healthy control subjects, suggesting subtle alteration of the endothelial microvascular regulatory system.38,39 Other studies have documented a strong relation between CFS and the postural orthostatic tachycardia syndrome,40 a condition that seems to be characterized by reduced norepinephrine reuptake in the sympathetic synapse.41 Altogether, the previous studies and our own results underscore the need for research specifically addressing the complicated interaction of adrenergic, cholinergic, and other microvascular control mechanism in patients with CFS.