1428

Journal of Clinical Endocrinology and Metabolism Copyright 0 1994 by The Endocrine Society

Vol. 79, No. 5 Printed in U.S.A.

Suspected Postprandial Hypoglycemia Is Associated with ,&Adrenergic Hypersensitivity and Emotional Distress* IVAN BERLIN, ANDRfi FRANCOIS CESSELIN,

GRIMALDI, ALAIN

AND

CATHERINE J. PUECH

Dkpartements de Pharmacologic Clinique (LB., A.J.P), H6pital Piti&Salp&ridre, 75013 Paris, France

LANDAULT,

Diabktologie

(A.G.), and Biochimie

(CL., F.C.),

ABSTRACT Suspected postprandial (reactive or idiopathic) hypoglycemia is characterized by predominantly adrenergic symptoms appearing after meals rich in carbohydrates and by their rare association with low blood glucose level (C2.77 mmol/L). We studied heart rate, blood pressure, plasma insulin, C-peptide, and catecholamine responses during a 5-h oral glucose tolerance test in eight patients with suspected postprandial hypoglycemia and eight age-, sex-, and body mass indexmatched healthy controls. We also evaluated p-adrenergic sensitivity by using the isoproterenol sensitivity test. Psychological profile was assessed by the Symptom Checklist (SCL-9OR) self-report symptom inventory. Patients with suspected postprandial hypoglycemia had higher fladrenergic sensitivity (defined as the dose of isoproterenol required to increase the resting heart rate by 25 beats/min) than controls (mean +SEM, 0.8 * 0.13 us. 1.86 f 0.25 pg isoproterenol; P = 0.002). After administration of glucose (75 g) blood glucose, plasma C-peptide, plasma epinephrine, and plasma norepinephrine responses were iden-

tical in the two groups, but plasma insulin was higher in the patients (group effect, P = 0.02; group by time interaction, P = 0.0001). Both heart rate and systolic blood pressure were significantly higher (but remained in the normal range) after glucose administration in patients with suspected postprandial hypoglycemia than in controls (group by time interactions, P = 0.004 and 0.0007, respectively). After glucose intake, seven patients had symptoms (palpitations, headache, tremor, generalized sweating, hunger, dizziness, sweating of the palms, flush, nausea, and fatigue), whereas in the control group, one subject reported flush and another palpitations, tremor, and hunger. Analysis of the SCL-9OR questionnaire revealed that patients had emotional distress and significantly higher anxiety, somatization, depression, and obsessive-compulsive scores than controls. We may conclude that patients with suspected postprandial hypoglycemia have normal glucose tolerance, increased fl-adrenergic sensitivity, and emotional distress. (J Clin Endocrinol Metab 79: 142% 1433,1994)

I

may result from hypoglycemia could not be confirmed (8,9). Therefore, it seemsthat the symptoms or signs compatible with hypoglycemia are consequencesof an altered response to ingestion of meals rich in carbohydrates and not to a chemical hypoglycemia, and psychological factors may have a considerablecontribution. As symptoms of patients with suspectedpostprandial hypoglycemia are mostly of the adrenergic type, we undertook this study to assess1) plasma catecholamine concentrations during an oral glucose tolerance test, and 2) peripheral adrenergic sensitivity to determine whether these patients have higher plasma catecholamine levels or increased adrenergic sensitivity compared to age-, sex-, and body mass index-matched controls. Secondarily, we evaluated blood glucose, plasma insulin, and plasma C-peptide responsesto the oral glucoseload and quantified the degree of emotional distressof patients and controls.

DIOPATHIC postabsorptive hypoglycemia (1) has long been considered as a nondisease (2) with nonhypoglycemia (3). These patients with suspectedpostprandial hypoglycemia (4) are characterized by symptoms or signs compatible with hypoglycemia 2-5 h after ingestion of meals rich in carbohydrates. Diagnosis of this condition (5) is mainly based on exclusion of insulinoma, intestinal surgery, or an endocrine disorder (adrenal insufficiency, early diabetes mellitus, or hypothyroidism). Several studies have shown that a low blood glucose level (CO.5 g/L or 2.77 mmol/L) is not a necessary condition, because postprandial symptoms suggestiveof hypoglycemia are rarely or not associatedwith low blood glucose concentration (4, 6), and chemical hypoglycemia occursasoften in patients with suspectedpostprandial hypoglycemia as in controls (1). Anthony et al. (7) observed that patients with this disorder had a Minnesota Multiphasic Personality Inventory score compatible with hypochondriacal and hysterical personality features. Half of the patients who consider themselves hypoglycemic has a current psychiatric disorder and hysterical personality traits (8). The view that psychological symptoms Received November 16, 1993. Revision received April 28, 1994. Accepted July 28, 1994. Address requests for reprints to: Dr. I. Berlin, DPpartement de Pharmacologie Clinique, Hhpital Piti&Salp&ri&e, 47 boulevard de l’Hspita1, 75013 Paris, France. * This work was supported by Grant P920707 from Assistance Publique, Hhpitaux de Paris.

Subjects

and Methods

Subjects Eight patients and eight age-, sex-, weight-, and height-matched controls were studied. All patients and control subjects gave their informed written consent. The protocol of the study was approved by the ethics committee of the Piti&Salp&riPre Hospital. The patients with suspected postprandial hypoglycemia had episodes appearing after meals and characterized by symptoms such as tremor, sweating, palpitations, hunger, fatigue, dizziness, headache. Liver disease, adrenal

1428

Downloaded from jcem.endojournals.org by on December 6, 2008

POSTPRANDIAL insufficiency, and thyroid, cardiac, or renal disease could be excluded. A 24-h fast was performed for each patient in the hospital and provoked no hypoglycemia. All patients had blood pressure below 140/90 mm Hg after 10 min of rest. None had diabetes, and none was treated by insulin, oral antidiabetic agents, P-blockers, anxiolytics, or antidepressants. Patients were referred to the out-patient clinic of the Department of Diabetology by their physician. Control subjects were recruited by advertisement. They had no acute or chronic disease. They had normal laboratory test results (routine 20-parameter biochemistry, hematology, and thyroid hormones). All patients and control subjects had normal 12-lead electrocardiogram results. None of the patients or controls was receiving a special diet or reported a weight change during the last 6 months. Both patients and controls were from the urban area of Paris.

HYPOGLYCEMIA

1429

13% for norepinephrine, respectively. The detection limit for epinephrine was 82 pmol/L, and that for norepinephrine was 0.12 nmol/L.

Statistical

analysis

Means were compared by the two-tailed t test for independent samples. Baseline values were tested by one-way analysis of variance. Analysis of variance was performed on variables with repeated measures to evaluate the effects of group and time and their interaction (betweengroup difference in the evolution of a variable as a function of time). Results are expressed as the mean f SEM. Differences were considered significant at P < 0.05.

Results

Study outline Subjects were instructed to eat at least 200 g carbohydrates for 3 days before the test day. All eventual medication (three patients took propranolol, one patient had valproic acid stopped 2 months before, and one patient was treated with propranolol and bromazepam) was stopped 2 weeks before the test days. On test days subjects arrived in the laboratory after an overnight fast of at least 12 h. A polyethylene cannula was inserted into a forearm vein. The contralateral arm was used for blood pressure and heart rate recordings. The first blood sample and blood pressure and heart rate recordings were taken at least 30 min after the insertion of the cannula. In 5 min the subjects absorbed 75 g glucose diluted in 200 mL tap water at time zero. Blood was drawn for blood glucose, plasma insulin, plasma C-peptide, plasma epinephrine, and ulasma noreuineohrine concentrations 15 min before and 0, 30, 60, 90, i20, 150, ld0, 240, and 300 min after glucose intake. Blood pressure and heart rate were recorded just after blood samples were drawn. The subjects remained in bed until 5 h after glucose ingestion. After the glucose tolerance test, subjects had a standardized lunch (containing 800 Cal; 16% protein, 35% lipids, and 49% carbohydrates). Two hours after lunch, an isoproterenol sensitivity test to evaluate padrenergic responsiveness was performed as described previously (10). This test consists of repeated bolus iv administration of increasing doses of isoproterenol hydrochloride. For each isoproterenol sensitivity test, a log dose-response curve was constructed from isoproterenol dose and heart rate increase, and the dose required to raise resting heart rate by 25 betas/mm [isoproterenol chronotropic dose (CD,,)] was determined by interpolation, This dose was considered a measure of isoproterenol sensitivity (11, 12). Subjective symptoms were recorded by free interviews during the 5 h of the oral glucose tolerance test and during the isoproterenol sensitivity test (lasting -1 h). Between lunch and the isoproterenol sensitivity test, subjects completed the Symptom Check List 90R (SCL-90R) autoquestionnaire (13). The SCL-90R is a 90-item self-report symptom inventory designed to primarily reflect the psychological symptom patterns of psychiatric and medical patients. Each item is rated on a 5-point scale of distress (O-4), ranging from “not at all” at one pole and “extremely” at the other. The responses are scored and interpreted in terms of 10 symptom dimensions (somatization, obsessive-compulsive, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, psychoticism, and additional items) and 3 global indices (global severity index, positive symptom distress index, and positive symptom total).

There was no difference in characteristicsbetween patients with suspected postprandial hypoglycemia and controls (Table 1). There was no difference between the two groups in baseline values of blood pressure and heart rate, blood glucose, plasma insulin, plasma C-peptide, and plasma catecholamine concentrations before the oral 5-h glucose tolerance test. After glucose ingestion, blood glucose increased in both groups similarly (Fig. 1). The mean (~SEM) peak blood glucose level was 8.59 + 0.5 mmol/L in the patient group and 8.57 -C 0.57 mmol/L in the control group, and the times to peak blood glucose were 64 f 9 and 60 + 14 min, respectively. Mean blood glucose nadir and time to blood glucose nadir were similar in the two groups [patients, 3.61 * 0.3 (range, 2.7-5.4); controls, 3.57 + 0.15 mmol/L (range, 3.1-4.3); patients, 225 f 28 min; controls, 202 f 55 min). In spite of the identical blood glucoselevels, plasmainsulin was significantly higher in patients with suspected postprandial hypoglycemia than in controls (group effect, P = 0.02; group by time interaction, P = O.OOOl), but plasma C-peptide concentrations showed no difference (group effect, P = 0.619; group by time interaction, P = 0.807; Fig. 1). There was no overall difference between patients and controls for plasma epinephrine or norepinephrine concentrations (Fig. 2); however, peak plasma norepinephrine was lower in the patients than in the controls (2.23 + 0.17 US.3.3 j, 0.03 nmol/L; P = 0.018). A significant time effect was found for plasma epinephrine (P = 0.0003), corresponding to the increasein plasma epinephrine at the end of the oral glucose tolerance test. The time to peak plasma epinephrine was 225 * 10 min in the patient group and 223 f 27 min in the control group, and it paralleled the blood glucose nadir. After glucose ingestion, the evolution of systolic blood pressureand heart rate in patients and controls was different as a function of time (group by time interaction for systolic blood pressure,P = 0.0007; for heart rate, P = 0.004; Fig. 3).

Blood glucose was determined by the glucose oxidase method. Plasma insulin was determined by commercial kit (BioMerieux, Marcy-L’Etoile, France). Intraand interassay coefficients of variation were 3% and 4.5%, respectively; the detection limit of the method was 14 pmol/L. Plasma C-peptide was determined using a Ciba-Coming kit (CergyPontoise, France). Intra- and interassay coefficients of variation were 5% and 6%, respectively; the detection limit of the method was 83 pmol/L. Plasma epinephrine and norepinephrine concentrations were determined by a radioenzymatic method (14), with CAT-A-KIT (TRK 995 Amersham International, Cardiff, UK). Intra- and interassay coefficients of variations were 9% and 11% for epinephrine and 8% and

TABLE hypoglycemia

1. Characteristics and healthy

of patients controls

Aae bd wt (kg) Ht (cm) Body mass index Sex (m/f) Values

(kg/m2)

aremeans,with

the SEM in

Downloaded from jcem.endojournals.org by on December 6, 2008

with

suspected

postprandial

Patients

Controls

36.2 (4.6) 61.4 (3.2) 160.6 (2.7) 23.7 iO.Sj 117

36.1 (4.6) 64.2 (3.3) 166.7 (2.6) 23.1 (l.Oj 117

parentheses.

BERLIN

1430

JCE8zM.1’394 Vol79.No5

ET AL.

“z

400

!z 2 w

300

Z

E

-30

0

30

60

90

200

120150180210240270300

01

’

’

’

’

-30

0

30

60

90

’

’

’

’

’

’

J

120150180210240270300

36OJ z\ z

3100

E a z n T fj a I ” I 2

2600

3.0 2100 1600

c-\

2.5

i

1100

0 E,

2

2.0

600

w -30 1100 1000 z-

900

a E 55

a00

z 3 2 z 2

600

?

300

a

200

0

30

60

90

i20i50i80210240270300

!

e I a

1.5

f

1.0

W Ei cc 0 z

0.5

700

0.0

500

-30

0

30

60

400

90 TIME

120150180210240270300 (minutes)

FIG. 2. Mean (~SEM) plasma epinephrine and norepinephrine concentrations during the oral glucose tolerance test in patients with suspected postprandial hypoglycemia (0) and healthy controls (0). -30

0

30

60

90 TIME

12015018021024027o3oo (minutes)

1. Blood glucose, plasma insulin, and plasma C-peptide during the 5-h glucose tolerance test (75 g glucose) in patients with suspected postprandial hypoglycemia (0) and in age-, sex-, and body mass indexmatched healthy controls (0). Values are the mean f SEM.

FIG.

Z

Table 2 shows symptoms reported after glucoseingestion. More patients with suspected postprandial hypoglycemia reported more symptoms than controls. Seven patients reported 12 symptoms (number of total reports, 25), whereas 2 controls reported 4 symptoms (number of total reports, 4). Isoproterenol sensitivity, expressedas the dose of isoproterenol required to raise the resting heart rate by 25 beats/

Downloaded from jcem.endojournals.org by on December 6, 2008

POSTPRANDIAL

HYPOGLYCEMIA TABLE 2. Symptoms (75 g glucose)

1431 reported

during

the 5-h glucose

Patients with suspected postprandial hypoglycemia Palpitations Tremor Headache Sweating Hunger Dizziness Sweating of the palms Pallor Flush Weakness Nausea Fatigue Shown I

I

I

t

-30

0

30

60

90

120150180210240270300

-30

0

30

60

90

120150180210240270300

50

45 -30

I

’

’

’

’

0

30

60

90

I

’

’

I

I

’

I

I

I

’

’

’

’

(minutes)

FIG. 3. Heart rate and systolic and diastolic blood pressures during the oral glucose tolerance test in patients with suspected postprandial hypoglycemia (0) and in healthy controls (0). Values are the mean f SEM.

of subjects

that

Healthy

test

controls 1 1 0 0 I 0 0 0 1 0 0 0

reported

each symptom.

min (CD,,), was significantly higher in patients with suspected postprandial hypoglycemia (0.8 + 0.13 pg) than in controls (1.86 f 0.25 pg; P = 0.002), whereas the resting heart rates were similar (75 f 4 VS. 74 + 1 beats/mm). During the isoproterenol sensitivity test, 6 control subjects and 8 patients reported palpitations. Patients also experienced flush (n = 2), dizziness (n = 2), anxiety, weakness, headache, and fatigue (n = 1 each). Analysis of the SCL-90R questionnaire showed that the global severity index (0.855 f 0.21 US. 0.222 + 0.06; P = 0.012), the positive symptom distress index (39.12 + 7.4 ZIS. 15.37 + 4.5; P = 0.016), and the positive symptom total (1.87 f 0.18 VS. 1.38 f 0.11; P = 0.039) were significantly higher in patients than in controls. Patients had significantly higher scores than controls for the dimensions of somatization, obsessive-compulsive, depression, and anxiety (Table 3). Discussion

120150180210240270300 TIME

are the number

3 3 4 3 2 2 2 2 1 1 1 1

tolerance

The present study showed that patients with suspected postprandial hypoglycemia have increased/I-adrenergic sensitivity and emotional distress, and, after an oral glucose load, more symptoms, higher plasma insulin concentration, higher heart rate, and higher systolic blood pressure, but similar blood glucose and plasma catecholamine concentrations compared to age-, sex-, and body massindex-matched controls. Ingestion of glucose usually provokes an initial positive excursion of blood glucose followed by a smaller negative excursion. Ten percent of healthy asymptomatic subjects have glucose nadirs of 0.47 g/L (2.6 mmol/L) or below (15). Thus, a blood glucose level lower than 0.5 g/L (2.77 mmol/ L) is not uncommon and does not lead to hypoglycemic symptoms. Several studieshave shown (1,15,16) that during an oral glucose tolerance test, patients with suspectedpostprandial hypoglycemia had similar blood glucoselevels and blood glucosenadirs as controls, and two prospective studies (4, 6) confirmed that these patients had no chemical hypoglycemia when they experienced symptoms. Patients with suspectedpostprandial hypoglycemia seem not to differ in plasma insulin or glucagon concentrations from controls (1, 16, 17); however, increasedinsulin sensitiv-

Downloaded from jcem.endojournals.org by on December 6, 2008

1432 TABLE

Shown

BERLIN 3. Results

of the Symptom

Check

List

(SCL-9OR)

ET AL.

questionnaire

Dimension

Patients with suspected postprandial hypoglycemia

Somatization Obsessive-compulsive Interpersonal sensitivity Depression Anxiety Hostility Phobic anxiety Paranoid ideation Psychoticism Additional items

15.25 (31.8) 10.75 (26.9) 7.37 (20.5) 13.75 (26.4) 11.5 (28.8) 3.37 (14) 3.25 (11.6) 4.37 (18.2) 1.37 (3.4) 5.87 (21)

are means

with

the percentage

of the possible

JCE & M. 1994 Vol79.No5

maximal

Healthy

controls

2.25 (4.7) 2.75 (2.5) 4.5 (12.5) 2.62 (5) 0.75 (1.9) 1.5 (6.3) 0.75 (2.7) 2.25 (9.4) 0.25 (0.6) 3.5 (12.5)

P value 0.001 0.021 0.397 0.025 0.0001 0.053 0.057 0.27 0.143 0.208

score in parentheses.

ity was observed in a subgroup of patients characterized by chemical hypoglycemia associated with clinical symptoms and normal plasmainsulin concentrations (17). In the present study, basal fasting plasma insulin levels were similar in patients and controls, but after a glucose load, patients had significantly higher plasma insulin, as has been found in patients with major depression (18), but similar C-peptide concentrations ascontrols. This finding should be interpreted with caution and necessitatesfurther investigations. The fact that plasma C-peptide concentrations were similar in the two groups suggeststhat the higher plasma insulin concentration observed in the patient group is due to a reduced insulin clearance associatedwith increased insulin resistance.Catecholaminesmay induce insulin resistanceby reducing insulin binding (19, 20) by P-adrenergic mechanisms, and physiological epinephrine levels exert a pronounced insulin antagonistic effect that is mediated by ,&receptor stimulation (21). Epinephrine stimulates glucose production, inhibits glucose clearance(22), and may secondarily lead to insulin resistance and hyperinsulinemia. Cortisol, another stresshormone (that was not measured in this study), may also induce insulin resistancein man (23). On the other hand, insulin increases sympathetic nervous system activity (24) and, even within the physiological range, elevates sympathetic neural outflow (25). Thus, not only could the increasedsympathetic response result in hyperinsulinemia, but, conversely, hyperinsulinemia could theoretically lead to increasedsympathetic responses. As patients with suspectedpostprandial hypoglycemia do not have lower blood glucose, but show more adrenergictype symptoms than controls (l), and plasma epinephrine and norepinephrine concentrations have been found to be identical in thesepatients and controls during an oral glucose load (16), the question arose of whether patients with this disorder had an increased responseto catecholamines. The present study showed that patients with suspectedpostprandial hypoglycemia had increased P-adrenergic sensitivity. This finding is compatible with the observation of higher heart rate and systolic blood pressure after an oral glucose load in these patients. This adrenergic hypersensitivity may explain the fact that patients with suspected postprandial hypoglycemia have more symptoms after glucose ingestion than controls; the physiological posthyperglycemic drop in blood glucoseresulted in a plasma epinephrine increasethat was similar in both groups, but leading to adrenergic (hy-

poglycemia-like) symptoms only in the group of patients. According to the results of the SCL-90R questionnaire, patients with suspectedpostprandial hypoglycemia had high levels of emotional distress and differed from controls in dimensions of depression, obsessive-compulsive, anxiety, and somatization. These findings agree with those of Anthony et al. (7) and Johnson et al. (9), which showed that patients with suspected postprandial hypoglycemia have a conversion pattern on Minnesota Multiphasic Personality Inventory, and their symptoms were not related to chemical hypoglycemia (9). The clinical symptoms of patients with suspectedpostprandial hypoglycemia resemblethose of patients with panic attacks. Panic disorder patients are distinguished from patients with generalized anxiety disorder by their mostly physical complaints, corresponding to increased autonomic nervous system activity: increasedblood pressure, tachycardia, palpitations, dizziness, sweating, and difficulty of catching his or her breath (26). Panic disorder patients have increased noradrenergic activity (27), and panic attack can be launched by isoproterenol(26, 28, 29). Furthermore, patients with panic disorder have increased sensitivity to isoproterenol(29). The results of the SCL-90R questionnaire suggest that patients may have more problems dealing with stressthan controls. It is, therefore, possiblethat under the sameexperimental conditions, patients perceived more stressthan controls. Thus, it cannot be determined whether the higher systolic blood pressureand heart late after glucose ingestion and the increased adrenergic sensitivity observed in the patient group result from a greater sensitivity to stresssituations or from an intrinsic characteristic of the patients. In conclusion, patients with suspected postprandial hypoglycemia have an increased /3-adrenergic sensitivity, due perhaps to an anxiety-depression state and responsible for the hypoglycemia-like symptoms. Further studiesare needed to assess,from a psychiatric point of view, this group of patients. Propranolol ameliorated symptoms of reactive hypoglycemia in postgastrectomy patients (30). Therapeutic trials should evaluate the effectiveness of P-blockers or psychotropic drugs, such as tricyclic antidepressants and monoamine oxidase inhibitors (which are effective treatments in panic disorders), in patients with suspected postprandial hypoglycemia.

Downloaded from jcem.endojournals.org by on December 6, 2008

POSTPRANDIAL

15.

Acknowledgment The authors assistance.

thank

to Nathalie

Brialy

for

HYPOGLYCEMIA

her

excellent

technical

16.

1433

Lev-Ran A, Anderson

RW. 1981 The diagnosis

hypoglycemia.

30:996-999.

Diabetes.

Hogan MJ, Service FJ, Sharbrough glucose tolerance test compared of reactive hypoglycemia. Mayo

of oral glucose tolerance tests and mixed meals in patients with apparent idiopathic postabsorptive hypoglycemia. Absence of hypoglycemia after meals. Diabetes. 30:465-470. 2. Yager J, Young RT. 1974 Non-hypoglycemia is an epidemic condition. N Engl J Med. 291:907-908. on non-hypoglyce3. Cahill GF, Soeldner JS. 1974 A non-editorial mia. N Engl J Med. 291:905-906. 4. Palardy J, Havrankova J, Lepage R, et al. 1989 Blood glucose measurements during symptomatic episodes in patients with suspected postprandial hypoglycemia. N Engl J Med. 321:1421-1425. 5. Betteridge DJ. 1987 Reactive hypoglycaemia. Br Med J. 295:286287. 6. Snorgaard 0, Binder C. 1990 Monitoring of blood glucose concentration in subjects with hypoglycaemic symptoms during everyday life. Br Med J. 300:16-18. 7. Anthony D, Dippe S, Hofeldt FD, Davis JW, Forsham PH. 1973 Personality disorder and reactive hypoglycemia. Diabetes. 22:664675. a. Ford CV, Bray GA, Swerdloff RS. 1976 A psychiatric study of patients with a diagnosis of hypoglycemia. Am J Psychiatry. 133:290-294. 9. Johnson DD, Dorr KE, Swenson WM, Service J. 1980 Reactive hypoglycemia. JAMA. 243:1151-1155. beta10. Berlin I, Grimaldi A, Bosquet F, Puech AJ. 1986 Decreased adrenergic sensitivity in insulin-dependent diabetic subjects. J Clin Endocrinol Metab. 63~262-265. 11. Cleaveland CR, Rangno RE, Shand DG. 1972 A standardized isoproterenol sensitivity test. The effects of sinus arrhythmia, atropine, and propranolol. Arch Intern Med. 130:47-52. 12. George CF, Conolly ME, Fenyvesi T, Briant R, Dollery CT. 1972 Intravenously administered isoproterenol sulfate dose-response curve in man. Arch Intern Med. 130:361-364. 13. Derogatis LR. 1977 SCL-90-R (revised) version manual I. Baltimore: John Hopkins University School of Medicine. single isotope radioen14. Peuler JD, Johnson GA. 1977 Simultaneous zymatic assay of plasma norepinephrine, epinephrine and dopamine. Life Sci 21:625-636.

FW, Gerich JE. 1983 Oral with a mixed meal in the diagnosis Clm Proc. 58:491-496.

17.

Tamburrano G, Leonetti F, Sbraccia P, Giaccari A, Locuratolo N, Lala A. 1989 Increased insulin sensitivity in patients with idiopathic

18.

Winokur

References 1. Charles MA, Hofeldt F, Shackelford A, et al. 1981 Comparison

of postprandial

reactive

hypoglycemia.

A, Maislin

J Clin Endocrinol

G, Phillips

resistance after oral glucose depression. Am J Psychiatry.

Metab.

69:885-890.

JL, Amsterdam JD. 1988 Insulin

tolerance testing in patients with major 145:325-330. 19. Arner P, Hellmer J, Ewerth S, Ostman J. 1984 Effect of glucose on beta-adrenergic induced downregulation of insulin receptor binding in human fat cells. Biochem Biophys Res Commun. 122:97-122. 20. Lonnroth P, Wesslau C, Stenstrom G, Tisell L-E, Smith U. 1985 Reduced insulin binding to human fat cells following beta-adrenergic stimulation-experimental evidence and studies in patients with pheochromocytoma. Diabetologia. 28:901-906. 21. Lager I, Attvall S. Eriksson BM, von Schenk H, Smith U. 1986 Studies on the insulin-antagonistic effect of catecholamines in normal man. Diabetologia. 29:409-416. 22. Rizza RA, Cryer PE, Haymond MW, Gerich JE. 1980 Adrenergic mechanisms for the effects of epinephrine on glucose production and clearance in man. J Clin Invest. 65:682-689. 23. Rizza RA, Mandarin0 LJ, Gerich JE. 1982 Cortisol-induced insulin resistance in man: unpaired suppression of glucose production, stimulation of glucose utilization due to a postreceptor defect on insulin action. J Clin Endocrinol Metab. 54:131-138. 24. Rowe JW, Young JB, Minaker KL, Stevens AL, Pallotta J, Landsberg L. 1981 Effect of insulin and glucose infusions on sympathetic nervous system activity in normal man. Diabetes. 30:219-225. 25. Anderson EA, Hoffman RP, Balon TW, Sinkev CA, Mark AL. 1991 Hyperinsulinemia produces both sympathetic neural activation and vasodilation in normal humans. J Clin Invest. 87:2246-2252. 26. Gorman JM, Liebowitz MR, Fyer AJ, Stein J. 1989 A neuroanatomical hypothesis for panic disorder. Am J Psychiatry. 146:148161. 27. Charney DS, Heninger GR. 1986 Abnormal regulation of noradrenergic function in panic disorders. Arch Gen Psychiatry. 43:10421054. 28. Balon R, Yeragani VK, Pohl R, Muench J, Berchou R. 1990 Somatic and osvcholodcal svmutoms durine; isooroterenol-induced panic 1 attacks’f’syc~atry Rks. ‘32:103-112.A 29. Pohl R. Yeraeani VK. Balon R. et al. 1988 Isooroterenol-induced panic attacks.“Biol Psychiatry. 24:891-902. ’ 30 Leichter SD, Permutt MA. 1975 Effect of adrenergic agents on postgastrectomy hypoglycemia. Diabetes. 24:1005-1010.

Downloaded from jcem.endojournals.org by on December 6, 2008