Professional Documents
Culture Documents
Hypoglycemia
Perception of Neurogenic (Predominantly
Cholinergic) Rather Than Neuroglycopenic
Symptoms
DWIGHT A. TOWLER, CAROLYN E. HAVLIN, SUZANNE CRAFT, AND PHILIP CRYER
From the Division of Endocrinology, Diabetes, and Metabolism of the Department of Medicine; and the General Clinical Research Center; and the
Diabetes Research and Training Center, Washington University School of
Medicine, St. Louis, Missouri.
Address correspondence and reprint requests to Dr. Philip E. Cryer,
Division of Endocrinology, Diabetes, and Metabolism (Box 8127), Washington
University School of Medicine, 660 South Euclid Avenue, St. Louis, MO 63110
Received for publication 12 May 1993 and accepted in revised form 26 July
1993.
IDDM, insulin-dependent diabetes mellitus; RIA, radioimmunoassay; EPI,
epinephrine; NE, norepinephrine; NEFA, nonesterified fatty acid; HR, heart
rate; BP, blood pressure; sBP, systolic blood pressure; dBP, diastolic blood
pressure; ANOVA, analysis of variance; BMI, body mass index; (S-OHB,
p-hydroxybutyrate.
atrogenic hypoglycemia is a major problem for patients with IDDM because of intermittent hyperinsulinemia that results from imperfect insulin replacement
regimens and compromised physiological defenses
against hyperinsulinemia (1). In many patients, this problem is compounded by the syndrome of hypoglycemia
unawareness. In this syndrome, the patient experiences
partial or complete loss of the warning symptoms of
developing hypoglycemia that previously prompted its
recognition and led to food ingestion, which prevented
hypoglycemia's progression to severe hypoglycemia ( 1 7).
Conventionally, symptoms of hypoglycemia are divided into those directly attributable to glucose deprivation from the CNS (neuroglycopenic symptoms) and
those attributable to the autonomic nervous system activation triggered by hypoglycemia (neurogenic or autonomic symptoms) (8). Although some symptoms are
readily placed in one category or the other, others are not
(9-11). Clinical hypoglycemia unawareness in IDDM is
often assumed to be the result of the loss of neurogenic
symptoms of hypoglycemia (2,4), but this has not been
documented (8). Indeed, patients recall neuroglycopenic
as well as neurogenic symptoms after an episode of
hypoglycemia (10,11). Therefore, it is conceivable that
1791
AWARENESS OF HYPOGLYCEMIA
Atropine
(l.Omg)
cholinergic antagonist atropine sulfate (Elkins-Sinn,
Cherry Hill, NJ) injected intravenously at 0 and 60 min.
SEGMENTS:
AI
-BThe doses of these antagonists and the experimental
-60
-30
0
30
60
design are shown in Fig. 1.
TIME (minute.)
Observations (arterialized venous blood samples, BPs,
FIG. 1. Experimental design.
HRs, and symptom scores) were made at 15-min intervals (-30 through 120 min). Plasma glucose concentrations were, of course, measured more frequently.
patients use both neuroglycopenic and neurogenic Cognitive function tests were performed between 45 and
symptoms to recognize developing hypoglycemia. Thus, 60 min and 105 and 120 min.
whether awareness of hypoglycemia is normally based Analytical methods. Plasma glucose concentrations
on the perception of neurogenic symptoms, neuroglyco- were determined with a glucose oxidase method on an
penic symptoms, or both is not known. Likewise, whether analyzer (Beckman, Fullerton, CA). Plasma insulin (13),
clinical hypoglycemia unawareness in IDDM is the result C-peptide (13), glucagon (14), pancreatic polypeptide
of loss of neurogenic symptoms, neuroglycopenic symp- (15), growth hormone (16), and cortisol (17) concentratoms, or both is unknown.
tions were determined with RIA. Plasma EPI and NE
To determine which of an array of reported symptoms concentrations were assessed with a single isotope
are reproducible for hypoglycemia and which are neuro- derivative (radioenzymatic) method (18). Enzymatic
genic as opposed to neuroglycopenic in origin and to methods were used to measure serum NEFA (19) and
test the hypothesis that awareness of hypoglycemia is blood p-OHB (20), lactate (21), and alanine (22) concenthe result of perception of neurogenic rather than neuro- trations. BPs and HRs were recorded automatically (Dyglycopenic symptoms, we quantitated these variables namap, Critikon, Tampa, FL).
during euglycemia, hypoglycemia alone, hypoglycemia
Subjects scored each symptom from 0 (none) to 6
with combined a- and p-adrenergic blockade, and hy(severe). The symptoms included 16 that were thought to
poglycemia with panautonomic (adrenergic and muscabe indicative of hypoglycemia, i.e., difficulty thinking/
rinic cholinergic) blockade in normal, young adults.
confused, warm, shaky/tremulous, nausea, tired/drowsy,
hungry, weak, sweaty, headache, heart pounding, diffiRESEARCH DESIGN AND METHODS
culty speaking, nervous/anxious, dizzy, faint, tingling,
We studied 10 normal humans (7 men, 3 women), 22-29
yr of age with BMIs ranging from 18.7 to 30.7 kg/m2. and blurred vision; and three not thought to be indicative
Subjects gave their written consent to participate in this of hypoglycemia, i.e., cold, happy, sad. "Blood sugar
study, which was approved by the Washington University low" was used as a measure of awareness of hypoglySchool of Medicine Human Studies Committee and con- cemia.
Cognitive function tests included measures of global
ducted at the Washington University General Clinical
Research Center. The subjects were given no instruc- cognitive function (line orientation), attention (vigilance
tions about the symptoms of hypoglycemia before the and serial addition), and memory (immediate and delayed paragraph recall), as described previously (23).
study. However, they all had medical backgrounds.
Experimental design. As illustrated in Fig. 1, each Statistical analysis. To assess the impact of the various
subject was studied (after an overnight fast) on four antagonist drugs used, the mean incremental change in
occasions, using a hyperinsulinemic euglycemic and each parameter from segment A (mean of -30, -15, and
hypoglycemic clamp technique (12), in random se- 0 min values) to segment B (mean of 30, 45, and 60 min
quence. On all occasions, regular human insulin (Humu- values) (Fig. 1) under all four study conditions was tested
lin, Lilly, Indianapolis, IN) was infused intravenously in a with a repeated measures ANOVA. To assess the impact
dose of 12.0 pmol kg" 1 min ~1 (2.0 mU kg" 1 of hypoglycemia, the mean incremental change in each
min ~1) and plasma glucose concentrations were parameter from segment B to segment C (mean of 90,
clamped at 5.0 mM (90 mg/dl) with variable infusions of 105, and 120 min values) under all four study conditions
20% glucose before initiation of observations (at - 3 0 also was tested with a repeated measures ANOVA. For
min). On one occasion, euglycemia (-5.0 mM) was those parameters with significant overall treatment efmaintained through 120 min. On the other three occafects, differences among individual conditions were
sions, euglycemia was maintained through 60 min; then
tested with the Newman-Keuls multiple-range test. Data
glucose levels were lowered and clamped at 2.5 mM
are presented as means SE.
1. EUGLYCEMIA
5.0 mmol / L-
1792
100
5.0
4.0
80
GLUCOSE
o
1
3.0
60
m-- EU
= HYPO
40
o= HYPO+ADB
2.0
a = HYPO + PAB
-30
30
60
90
120
GLUCAGON
o
I 0.25
20
30
75
150
_j
100
50
30
60
90
^
a
60
90
120
PANCREATIC
POLYPEPTIDE
300
50
25
200
$rt
-30
120
30
90
120
TIME (min)
FIG. 2. Mean SE plasma glucose, Insulin, C-peptlde, glucagon, and pancreatic polypeptlde concentrations during hyperinsuiinemic (12.0
pmol kg~ 1 mirr 1 ) euglycemlc (-5.0 mM) clamps {EU) (-30 to 120 min) and euglycemic then hypoglycemic (-2.5 mM) clamps (60-120 mln)
alone {HYPO), with combined oc and p-adrenergic blockade {HYPO + ADB, phentolamlne and propranolol 0-120 min) and with panautonomic
blockade {HYPO + PAB, phentolamine and propranolol 0-120 mln plus atropine at 0 and 60 min).
RESULTS
Hormonal, metabolic, and hemodynamic changes.
Plasma glucose targets were achieved (Fig. 2). Plasma
insulin concentrations were comparable under all study
conditions. Plasma C-peptide levels were unaffected
by the antagonists; they fell during hypoglycemia
(P = 0.003) but were unaffected by adrenergic or panautonomic blockade (Fig. 2). Plasma glucagon and pancreatic polypeptide concentrations were unaffected by the
antagonists during euglycemia. The glucagon responses
to hypoglycemia (P = 0.020) were unaffected by adrenergic or panautonomic blockade (Fig. 2). The pancreatic
polypeptide responses to hypoglycemia (P< 0.0001)
were unaffected by adrenergic blockade, but were completely prevented by panautonomic blockade (Fig. 2).
Plasma EPI concentrations were unaffected, but plasma
NE levels were increased (P< 0.0001) by the antagonists during euglycemia. Both the EPI (P< 0.0001) and
NE (P< 0.0001) responses to hypoglycemia were increased by both adrenergic and panautonomic blockade
(Fig. 3). Plasma growth hormone and cortisol concentrations and their responses to hypoglycemia (both
P < 0.0001) were unaffected by adrenergic or panautonomic blockade (Fig. 3).
Serum NEFA and blood lactate, p-OHB and alanine
concentrations were unaffected by the antagonists during euglycemia. The NEFA (P = 0.002) and lactate
(P< 0.0001) responses to hypoglycemia were prevented by both adrenergic and panautonomic blockade
(Fig- 4).
HRs were unaffected by adrenergic blockade but were
increased (P < 0.0001) by panautonomic blockade (Fig.
5) during euglycemia. sBPs decreased (P = 0.031) more
prominently during adrenergic compared with panautonomic blockade during euglycemia; dBPs were unaffected (Fig. 5). Mean BP (data not shown) also
decreased ( P = 0.003). The HR responses to hypoglycemia (P< 0.0001) were prevented by both adrenergic
and panautonomic blockade (Fig. 5). Apparent differences in sBPs during hypoglycemia (Fig. 5) were not
statistically significant.
Awareness of hypoglycemia. No significant effects of
the antagonists (phentolamine and propranolol with or
without atropine) on the scores for "blood sugar low"
were noted during euglycemia (segment A vs. segment
B) (data not shown). However, significant (P< 0.001)
treatment effects were observed during hypoglycemia.
No change was observed in the mean SE score from
euglycemia to euglycemia, but the score increased significantly (2.0 0.4) from euglycemia (segment B) to
hypoglycemia (segment C) (Fig. 6). The score during
hypoglycemia was not significantly reduced by adrenergic blockade (1.6 0.5), but it was reduced significantly
1793
AWARENESS OF HYPOGLYCEMIA
1
1
EPI15.0 . NEPHRINE
- 4000
12.0
- 3000
90
20.0
2000
NOREPINEPHRINE
1500
i
i
r?n ; EU
10.0
5.0
0
- 2000 a
= HYPO
o = HYPO + ADB
D= HYPO+PAB
-30
30
//
^
- 1000
60
90
500
3.0
120
-30
30
60
90
120
80
GROWTH
HORMONE
3000
1000 a
I 6.0
60
2000
40
1000
20
-30
30
60
90
120
TIME (min)
FIG. 3. Mean SE plasma EPI, NE, growth hormone, and cortlsol concentrations during hyperlnsullnemic euglycemic clamps (EU) and
euglycemic then hypoglycemic clamps alone (HYPO), with combined a- and p-adrenergic blockade (HYPO + ADB) and with panautonomic
blockade (HYPO + PAB).
1794
200
200
NEFA
3200-
- 3200
150
I 100
E23= E U
= HYPO
50
o = HYPO + ADB
o = HYPO + PAB
-30
30
60
90
120
-30
400 -
30
60
90
- 200
200
_,
200 -
-30
30
60
120 '
90
" -30
30
TIME (min)
FIG. 4. Mean SE serum NEFA and blood Iactate, aianine and p-OHB concentrations during hyperinsulinemic euglycemic clamps (EU) and
euglycemic then hypoglycemic clamps alone (HYPO), with combined a- and p-adrenergic blockade (HYPO + ADB) and with panautonomic
blockade (HYPO + PAB).
DISCUSSION
These data indicate that awareness of hypoglycemia
normally is largely, if not exclusively, the result of perception of neurogenic rather than neuroglycopenic symptoms; this perception is mediated predominantly by
muscarinic cholinergic rather than adrenergic mechanisms. When asked to score the likelihood that their blood
glucose was low, subjects were clearly able to distinguish euglycemia (-5.0 mM) from hypoglycemia (~2.5
mM). However, subjects were significantly and substantially less able to distinguish euglycemia from hypoglycemia during panautonomic blockade (with phento-
30
30
60
90
120
TIME (min)
FIG. 5. Mean SE HRs, sBPs, and dBPs during hyperinsulinemic
euglycemic clamps (EU) and euglycemic then hypoglycemic clamps
alone (HYPO), with combined a- and p-adrenergic blockade
(HYPO + ADB) and with panautonomic blockade (HYPO + PAB).
1795
AWARENESS OF HYPOGLYCEMIA
SWEATY (p<0.00l)
SHAKY/TREMULOUS
* t
I
8*
to
HEART POUNDING
SYMPTOMS
(p< 0.001)
i
ADB
PAB
0
BLOOD SUGAR LOW
|p< 0.001)
NERVOUS/ANXIOUS
0
EU
ADB
PAB
NEUROGLYCOPENIC SYMPTOMS
DIFF. SPEAKING
WEAK (P-0.011)
(p-0.310)
Iii
DIFF. THINKING/
CONFUSED
BLURRED VISION
DIZZY (p=0.130)
|p = 0.161)
(p 0.004)
ADB
-HYPO
PAB
AOB
-HYPO
PAB
EU
ADB
PAB
-HYPO
1796
HAPPY (p0.697)
EU
AOB PAB
-HYPO
EU
ADB PAB
-HYPO
0ADB
PAB
-HYPO
HYPO
HYPO
WARM (P<0.00l)
< 2
(p= 0.002)
(p=0.071)
TINGLING (P-0.009)
EU
HEADACHE
NONSYMPTOMS
NEUROGENIC
HUNGRY lp<0.00i)
V'<<-\
NAUSEA (p = 0.056)
El
1.
COLD (p=0.185)
| p < 0.001)
TABLE 1
Cognitive function test scores during euglycemia, hypoglycemia alone, hypoglycemia with adrenergic blockade and
hypoglycemia with panautomatic blockade
Nominal plasma glucose (mM)
Test
Line orientation (errors)
Vigilance (errors)
Serial addition (errors)
Immediate recall (bits)
Delayed recall (bits)
Overall
P
value
Saline
5.0
NS
0.3
0
<0.01
<0.001 0.7
<0.04 14.0
<0.001 13.4
5.0
0.2
Adrenergic
blockade
Saline
0.3
0.3
0.3 0.5
1.4 15.4
1.4 14.8
5.0
2.5
0.2 0.3
0.2 0.2
0.2 1.0
1.4 12.6
1.2 11.4
0.2 0.3
0.1 0.7
0.3 0.9
1.3 13.6
1.3 11.8
5.0
2.5
Panautomatic
blockade
5.0
0.3 0.2 0.2
0.5 0.1 0 . 1 *
0.6 0.4 0 . 2 t
1.4 14.0 1.6
1.4 12.0 1.6*
2.5
0.3
2.1
3.6
12.2
9.0
0.2
0.6
0.6
2.0
1.7
ACKNOWLEDGMENTS
1797
AWARENESS OF HYPOGLYCEMIA
REFERENCES
1. Cryer PE: latrogenic hypoglycemia in IDDM: Consequences, risk
factors and prevention. In Diabetes Annual. Vol. 7. Marshall S, Home
PD, Alberti KGMM, Krall L, Eds. Amsterdam, Elsevier. 1993, p.
317-31
2. Pramming S, Thorsteinsson B, Bendtson I, Binder C: Symptomatic
hypoglycemia in 411 type I diabetic patients. Diabetic Med 8:21722, 1991
3. Ryder REJ, Owens DR, Hayes TM, Ghatei M, Bloom SR: Unawareness of hypoglycemia and inadequate glucose counterregulation:
no causal relationship with diabetic autonomic neuropathy. Br Med
J 301:783-87, 1990
4. Heller SR, Herbert M, Macdonald IA, Tattersall RB: Influences of
sympathetic nervous system on hypoglycemic warning symptoms.
Lancet ii:35963, 1987
5. Hepburn DA, Patrick AW, Eadington DW, Ewing DJ, Frier BM:
Unawareness of hypoglycaemia in insulin-treated diabetic patients:
prevalence and relationship to autonomic neuropathy. Diabetic Med
7:711-17, 1990
6. Gerich JE, Korytokowski M, Mokan M, Veneman T, Mitrakou A:
Hypoglycemia unawareness. Endocrinol Rev 12:356-71, 1991
7. Cryer PE: latrogenic hypoglycemia as a cause of hypoglycemiaassociated autonomic failure in IDDM: a vicious cycle. Diabetes
41:255-60, 1992
8. Cryer PE, Binder C, Bolli GB, Cherrington AD, Gale EAM, Gerich
JE, Sherwin RS: Hypoglycemia in IDDM. Diabetes 38:1193-99,
1989
9. Heller SR, Macdonald IA: Physiological disturbances in hypoglycemia: Effect on subjective awareness. Clin Sci 81:1-9, 1991
10. Hepburn DA, Deary JJ, Frier BM, Patrick AM, Quinn JD, Fisher BM:
Symptoms of acute insulin-induced hypoglycemia in humans with
and without IDDM: factor analysis approach. Diabetes Care 14:
949-57, 1991
11. Hepburn DA, Deary JJ, Frier BM: Classification of symptoms of
hypoglycemia in insulin-treated diabetic patients using factor analysis: relationship to hypoglycaemia unawareness. Diabetic Med
9:70-75, 1992
12. Schwartz NS, Clutter WE, Shah SD, Cryer PE: The glycemic thresholds for activation of glucose counterregulatory systems are higher
than the threshold for symptoms. J Clin Invest 79:777-81, 1987
13. Kuzuya H, Blix PM, Horwitz DL, Steiner DF, Rubenstein AH: Determination of free and total insulin and C-peptide in insulin-treated
diabetics. Diabetes 26:22-29, 1977
14. Ensinck JW: Immunoassays for glucagon. In Glucagon: Handbook
of Experimental Pharmacology. Vol. 66. New York, Verlag, 1983
15. Gingerich RL, Lacy PE, Chance RE, Johnson MG: Regional pancre-
1798
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.