hypoglycemia - researchgate | find and share research · hypoglycemia philip e. cryer, md joseph n....

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/15274023

Hypoglycemia

Article in Diabetes Care · August 1994

DOI: 10.2337/diacare.17.7.734 · Source: PubMed

CITATIONS

130READS

2,695

3 authors, including:

Philip E Cryer

Washington University in St. Louis

407 PUBLICATIONS 32,216 CITATIONS

SEE PROFILE

Harry Shamoon

Albert Einstein College of Medicine

129 PUBLICATIONS 9,234 CITATIONS

SEE PROFILE

All content following this page was uploaded by Philip E Cryer on 03 February 2016.

The user has requested enhancement of the downloaded file.

https://www.researchgate.net/publication/15274023_Hypoglycemia?enrichId=rgreq-a7755baf1cd953c261922b417491b6e1-XXX&enrichSource=Y292ZXJQYWdlOzE1Mjc0MDIzO0FTOjMyNTA0OTY3MzM3MTY0OUAxNDU0NTA5Mjg5MjE1&el=1_x_2&_esc=publicationCoverPdf

https://www.researchgate.net/publication/15274023_Hypoglycemia?enrichId=rgreq-a7755baf1cd953c261922b417491b6e1-XXX&enrichSource=Y292ZXJQYWdlOzE1Mjc0MDIzO0FTOjMyNTA0OTY3MzM3MTY0OUAxNDU0NTA5Mjg5MjE1&el=1_x_3&_esc=publicationCoverPdf

https://www.researchgate.net/?enrichId=rgreq-a7755baf1cd953c261922b417491b6e1-XXX&enrichSource=Y292ZXJQYWdlOzE1Mjc0MDIzO0FTOjMyNTA0OTY3MzM3MTY0OUAxNDU0NTA5Mjg5MjE1&el=1_x_1&_esc=publicationCoverPdf

https://www.researchgate.net/profile/Philip-Cryer?enrichId=rgreq-a7755baf1cd953c261922b417491b6e1-XXX&enrichSource=Y292ZXJQYWdlOzE1Mjc0MDIzO0FTOjMyNTA0OTY3MzM3MTY0OUAxNDU0NTA5Mjg5MjE1&el=1_x_4&_esc=publicationCoverPdf


https://www.researchgate.net/institution/Washington_University_in_St_Louis?enrichId=rgreq-a7755baf1cd953c261922b417491b6e1-XXX&enrichSource=Y292ZXJQYWdlOzE1Mjc0MDIzO0FTOjMyNTA0OTY3MzM3MTY0OUAxNDU0NTA5Mjg5MjE1&el=1_x_6&_esc=publicationCoverPdf


https://www.researchgate.net/profile/Harry-Shamoon?enrichId=rgreq-a7755baf1cd953c261922b417491b6e1-XXX&enrichSource=Y292ZXJQYWdlOzE1Mjc0MDIzO0FTOjMyNTA0OTY3MzM3MTY0OUAxNDU0NTA5Mjg5MjE1&el=1_x_4&_esc=publicationCoverPdf


https://www.researchgate.net/institution/Albert_Einstein_College_of_Medicine?enrichId=rgreq-a7755baf1cd953c261922b417491b6e1-XXX&enrichSource=Y292ZXJQYWdlOzE1Mjc0MDIzO0FTOjMyNTA0OTY3MzM3MTY0OUAxNDU0NTA5Mjg5MjE1&el=1_x_6&_esc=publicationCoverPdf



T E C H N I C A L R E V I E W

HypoglycemiaPHILIP E. CRYER, MDJOSEPH N. FISHER, MDHARRY SHAMOON, MD

I atrogenic hypoglycemia causes recur-rent physical and recurrent or evenpersistent psychosocial morbidity, and

some mortality, in patients with insulin-dependent diabetes mellitus (IDDM), andin some patients with non-insulin-depen-dent diabetes mellitus (NIDDM) (1,2).There is now compelling evidence, fromthe Diabetes Control and ComplicationsTrial (DCCT), that metabolic control de-lays the development and progression ofretinopathy, nephropathy and neuropa-thy in IDDM, albeit at the expense of anincreased frequency of treatment-in-duced hypoglycemia (3). These findingswill almost assuredly provide further im-petus to patients and health care provid-ers to attempt to maintain plasma glucoselevels as close to the nondiabetic range aspossible. If so, hypoglycemia will becomean even more common problem for pa-tients with diabetes in the near future.

Clinical hypoglycemia, largely inIDDM, is the focus of this review. We firstsummarize selected background informa-tion. Then, the clinical issues are dis-cussed in more detail. Our intent is todefine the current body of knowledge andto point out relevant areas where knowl-edge is lacking. Obviously, the latter aresubstantial since hypoglycemia is a major

unsolved problem for the diabetes com-munity. However, our intent is to be se-lective rather than comprehensive. Theinterested reader is referred to a recentlypublished book on diabetes and hypogly-cemia (1) for further details and for otherpoints of view.

LITERATURE REVIEW ANDANALYSIS

BACKGROUND

CNS effects of hypoglycemiaThe brain is dependent on a continuoussupply of glucose from the circulation (4).The brain depends almost exclusively onglucose for its energy production underphysiological conditions. (A frequentlycited exception is prolonged fasting,when ketone bodies can provide as muchas two-thirds of brain energy metabolism[5], but this is hardly a physiological con-dition.) Indeed, glucose oxidation nor-mally accounts for almost all of the oxy-gen consumed by the brain (6), and thebrain respiratory quotient approaches 1.0(4).

The brain cannot synthesize glu-

From the Washington University School of Medicine (P.E.C.), St. Louis, Missouri; University ofTennessee College of Medicine (J.N.F.), Memphis, Tennessee; and Albert Einstein College ofMedicine (H.S.), Bronx, New York.

Address correspondence to Philip E. Cryer, MD, Division of Endocrinology, Diabetes andMetabolism, Washington University School of Medicine (Box 8127), 660 South Euclid Ave., St.Louis, MO 63110.

IDDM, insulin-dependent diabetes mellitus; NIDDM, non-insulin-dependent diabetes melli-tus; DCCT, Diabetes Control and Complications Trial; CNS, central nervous system; IQ, intel-ligence quotient.

cose and it can store only a few minutes'supply as glycogen (4). Therefore, it mustcontinuously derive its predominant met-abolic fuel from the circulation. Further-more, the brain cannot quickly increaseits extraction of glucose. Normally, therate of carrier-mediated (GLUT1) facili-tated glucose transport across the blood-brain barrier down a concentration gradi-ent exceeds the rate of brain glucosemetabolism. Thus, transport is not rate-limiting. However, if the plasma glucoseconcentration falls below a critical level(or if brain glucose metabolism increasessubstantially) glucose transport fromblood to brain becomes rate-limiting tobrain glucose metabolism and, thus,brain function and survival (4).

Hypoglycemia, sensed in thebrain itself (7) and in peripheral struc-tures such as the liver (8), triggers a seriesof central nervous system (CNS) medi-ated changes (9-11). These include, butare not limited to, changes in hormonesecretion, symptoms, cognitive dysfunc-tion, coma and, ultimately, death. Thestepped hypoglycemic clamp techniquehas been used to determine the glycemicthresholds for (i.e., the glucose concen-tration required to trigger) several of theseresponses to hypoglycemia (9,10). Meanarterialized venous glycemic thresholdsin nondiabetic humans are shown in Fig.1. Decrements in plasma glucose withinthe physiological range decrease insulinsecretion (9,11). Glucose decrements justbelow the physiological range increasethe secretion of glucose counterregula-tory hormones (9,10). Further glucosedecrements elicit symptoms of hypogly-cemia (9,10), while even further decre-ments cause cognitive dysfunction (10).As will become apparent later, however,these glycemic thresholds are dynamicrather than static.

Given the survival value of main-tenance of the plasma glucose concentra-tion, it is hardly surprising that physio-logical mechanisms that very effectivelyprevent or correct hypoglycemia (11)have evolved. Indeed, hypoglycemia is a

734 DIABETES CARE, VOLUME 17, NUMBER 7, JULY 1994

Technical Review

NORMAL GLUCOSE COUNTERREGULATION

IGLUCOSE

INSULIN (83±3mg/dL)

• I GLUCAGON (68±2mg/dL)

— - • 4EPINEPHRINE (69±2mg/dL)

(^ Brain Glucose Uptake, -67mg/dL)

• GROWTH HORMONE (66±2mg/dL)

4 CORTISOL (58±3mg/dL)

(Symptoms, ~54mg/dL;| Cognition, ~49mg/dL)

•*» GLUCOSE AUTOREGULATION (<50. >30mg/dL)

GLUCOSE

OTHER HORMONES, NEUROTRANSMITTERS,

OTHER SUBSTRATES

Figure 1—Schematic representation of the physiology of glucose counterregulation. Mean (±SE)

arterialized venous glycemic thresholds for the various responses/changes are also shown (9-11). From

CryerQl).

distinctly uncommon clinical event ex-cept in persons who use drugs, such asinsulin or sulfonylureas, that lower theplasma glucose concentration (12). Asdiscussed later, however, it is more thanjust the use of these glucose-loweringdrugs that explains the occurrence of hy-poglycemia in patients with diabetes.

Extra-CNS effects of hypoglycemiaAlthough the effects of hypoglycemia onthe brain per se are potentially most dev-astating, hypoglycemia elicits an array ofextra-CNS effects. The vast majority of therecognized responses are, however, CNSmediated. Some are clearly adaptive,some are seemingly maladaptive, andmany are teleologically obscure. They in-clude changes in hormone secretion in-cluding those relevant to the preventionand correction of hypoglycemia, summa-rized later, and the autonomic dischargethat results in hemodynamic changes andneurogenic (autonomic) symptoms.

The changes in hormone secre-

tion during hypoglycemia include decre-ments in insulin and increments in gluca-gon, epinephrine, growth hormone andcortisol (9-12). The physiological rele-vance of these is summarized shortly.Other hormones—corticotropin-releas-ing hormone, adrenocorticotropin, pro-lactin, vasopressin, oxytocin, pancreaticpolypeptide, renin, aldosterone, atrial na-triuretic hormone, gastrin, and parathy-roid hormone, among others—and sev-eral neuropeptides have been reported tobe released during hypoglycemia (13).

Both the sympathochromaffin(adrenomedullary and sympathetic neu-ral) and parasympathetic components ofthe autonomic nervous system are acti-vated during hypoglycemia (14). This un-derlies the hemodynamic changes (15)and the neurogenic (autonomic) symp-toms (16). The typical hemodynamic pat-tern includes an increase in heart ratewith widening of the pulse pressure (in-crements in systolic and decrements indiastolic blood pressure) with no change

in mean blood pressure. Increments incardiac output are the result of incre-ments in stroke volume as well as heartrate. Increased muscle sympathetic nerveactivity (14) presumably underlies net va-sodilatation, but the sympathetic re-sponse is differentiated. Splanchnic bloodflow has been reported to be increased(17) or unchanged (18) while renal bloodflow is reduced (15). Hypoglycemia alsoproduces lymphocytosis, probably epi-nephrine mediated, and neutrophilia, atleast in part cortisol mediated (19).

Physiology of glucosecounterregulationThe physiological mechanisms that nor-mally prevent or correct hypoglycemiahave been reviewed (11) and discussed indetail (20) recently and are summarizedin Fig. 1.

The principles of glucose counter-regulation are three. First, the preventionor correction of hypoglycemia is the re-sult of both dissipation of insulin and ac-tivation of glucose counterregulatory(glucose-raising) systems. Second,whereas insulin is the dominant glucose-lowering factor, there are redundant glu-cose counterregulatory factors. There aremultiple glucose-raising factors that col-lectively constitute a fail-safe system thatprevents or minimizes failure of the entiresystem despite failure of one, or perhapsmore, of its components. Third, there is ahierarchy among the glucoregulatory fac-tors. There is a ranked series of counter-regulatory factors, some more critical tothe effectiveness of the fail-safe systemthan others, that act in concert with dec-rements in insulin to prevent or correcthypoglycemia.

The first defense against fallingplasma glucose concentrations is de-creased insulin secretion; this occurs withglucose decrements within the physiolog-ical range, normally at a glycemic thresh-old of - 8 3 mg/dl (4.6 mM) (9,11). How-ever, biological glucose recovery fromhypoglycemia can occur despite an ap-proximately twofold peripheral hyperin-sulinemia and in the absence of portal hy-

DIABETES CARE, VOLUME 17, NUMBER 7, JULY 1994 735

Technical Review

Table 1—Range of frequencies (%) of individual symptoms of hypoglycemia reported ineight series of patients with insulin-treated diabetes mellitus

SweatingTremblingWeaknessVisual disturbanceHungerPounding heartDifficulty with speakingTingling around mouth

47-8432-7828-7124-6039-49

8-627-41

10-39

DizzinessHeadacheAnxietyNauseaDifficulty concentratingTirednessDrowsinessConfusion

11-4124-3610-445-20

31-7538-4616-3313-53

From Hepburn (31).

poinsulinemia (21). Thus, additional(glucose counterregulatory) factors mustbe involved. Critical glucose counter-regulatory systems are normally activatedat a glycemic threshold of ~68 mg/dl (3.8mM), well above the thresholds for symp-toms of hypoglycemia (~54 mg/dl [3.0mM]) and those for cognitive dysfunction(-49 mg/dl [2.7 mM]) (9,10). Among theglucose counterregulatory factors, gluca-gon plays a primary role. Epinephrine isnot normally critical, but it becomes crit-ical to glucose counterregulation whenglucagon is deficient. Because hypoglyce-mia develops or progresses when the se-cretion of both glucagon and epinephrineis deficient and insulin is present (22-29), these three hormones stand high inthe hierarchy of redundant glucoregula-tory factors. Other factors, such as growthhormone and cortisol in defense againstprolonged hypoglycemia, are known tobe involved. Furthermore, glucose auto-regulation (hepatic glucose production asan inverse function of ambient glucoselevels independent of hormonal and neu-ral glucoregulatory factors) may be oper-ative during severe hypoglycemia, andfatty acid elevations appear to mediate, inpart, the effects of epinephrine (11).Other hormones, neurotransmitters, andother metabolic substrate effects may alsobe shown to be involved. Nonetheless, allof these latter factors must stand low inthe hierarchy of redundant glucose coun-terregulatory factors since, despite theiractions, hypoglycemia develops orprogresses when both glucagon and epi-nephrine are deficient and insulin ispresent.

HYPOGLYCEMIA IN IDDM

ClassificationFor descriptive purposes treatment-in-duced hypoglycemia can be divided intothree categories: 1) asymptomatic (bio-chemical) hypoglycemia, 2) mild-moder-ate symptomatic hypoglycemia, and 3)severe hypoglycemia. Detection ofasymptomatic hypoglycemia requiresmeasurement of the plasma/blood glu-cose concentration. The patient is able torecognize and self-treat mild-moderatesymptomatic hypoglycemia. Severe hy-poglycemia, as defined in the DCCT pro-tocol, is temporarily disabling and, there-fore, requires the assistance of anotherperson (30).

While clinically and pedagogi-cally useful, this classification is arbitrary.Typically, the plasma glucose level fallsthrough asymptomatic and mild-moder-ate symptomatic phases before reachinglevels that produce severe clinical hypo-glycemia. As discussed later, however, theabsence of symptoms, or the failure of thepatient to recognize symptoms or to inter-pret symptoms as indicative of develop-ing hypoglycemia, is a major determinantof the frequency of severe hypoglycemia(30).

Clinical manifestationsBecause treatment-induced hypoglyce-mia is most often self-diagnosed, there is acritical reliance on symptoms. Spouses,other family members, friends, or co-workers, as well as medical personnel,also use signs to recognize hypoglycemia.

Symptoms. Patients with diabetes reportan array of symptoms during hypoglyce-mic episodes. Many of these, compiled byHepburn (31) from eight published series(32-39), are listed in Table 1. The varietyof reported symptoms and particularlythe wide range of their frequencies in thedifferent reports are noticeable. The latterlikely represents differences in patientpopulations, ascertainment methods, andsymptom assessment techniques. The dif-ficulty in providing a short list of reliablesymptoms is compounded by the fact thatsymptoms of hypoglycemia are often id-iosyncratic (40-42). In a given patient acertain symptom or symptom complex isoften reliable for that patient but not foranother patient. Thus each individual pa-tient must learn to recognize the individ-ual symptoms that most reliably suggestdeveloping hypoglycemia to him or her.

Although it has been suggestedthat hypoglycemic symptom patternsmay differ in patients with NIDDMtreated with oral agents (33), this remainsto be established. It appears that the hy-poglycemic symptom patterns of insulin-treated NIDDM and IDDM are the same(43).

It is conventional to divide thesymptoms of hypoglycemia into two cat-egories (16,44). J) Neuroglycopenicsymptoms are those that result directlyfrom brain glucose deprivation. Behav-ioral and cognitive changes are examples,as are coma and seizure. 2) Neurogenic(or autonomic) symptoms are those thatresult from perception of physiologicalchanges caused by the autonomic (adre-nomedullary, sympathetic neural andparasympathetic neural) discharge trig-gered by hypoglycemia. It could be rea-soned that this division is misleading be-cause the autonomic discharge isultimately the result of neuroglycopenia.Nonetheless, the division is clinically use-ful as discussed later. Parenthetically, it isimprecise to refer to neurogenic symp-toms as a group as "adrenergic." As dis-cussed later, some are mediated by cat-echolamines (adrenergic), but others aremediated by acetylcholine (cholinergic).


Technical Review

Table 2—Neurogenic and neuroglycopenic symptoms of hypoglycemia

Neurogenic Neuroglycopenic

Shaky/tremulousHeart poundingNervous/anxious

SweatyHungryTinglingBlood sugar low (awareness of hypoglycemia)

Difficulty thinking/confusedTired/drowsyWeakWarmDifficulty speakingIncoordinationOdd behavior(Coma, seizure, death)

Table is derived largely from Towler et al. (16).

It is conceivable that some might be me-diated by peptides (peptidergic). There-fore, the term neurogenic, or autonomicas some prefer (31), is more appropriate(44).

While many symptoms are readilycharacterized as neurogenic or neurogly-copenic, others have been more difficultto classify, and there has been some de-bate about these (31). Three approachesto such classification have been taken. 1)A physiological approach based on thephysiological mechanism (neurogenic ornon-neurogenic) thought to underlie agiven symptom (45). 2) Factor analysis toidentify statistical clusters of symptomscoupled with the physiological approachto deduce the mechanism of a given clus-ter (46-48). 3) A pharmacological ap-proach not unlike that used, in part, todefine physiological mechanisms in thepast but in this instance using adrenergicand cholinergic antagonists to attempt toblock both awareness of and individualsymptoms of hypoglycemia (16). Withthese approaches a consensus, albeit notquite complete, is emerging (16).

Major neurogenic and neurogly-copenic symptoms of hypoglycemia arelisted in Table 2. All three methods indi-cate that the symptoms of "shaky/tremu-lous," "heart pounding," and "sweaty" areneurogenic (16,45,48). Factor analysisand the pharmacological method indicatethat "hungry" is a neurogenic symptom(18,48). The pharmacological method in-dicates that "nervous/anxious," not as-

sessed with the other methods, is a neu-rogenic symptom (16). The symptom of"tingling" was found to be neurogenicwith the latter method (16); it did not fitin either category by factor analysis (47).As deduced from the pharmacologicalmethod and from factor analysis, "diffi-culty thinking/confused" and "tired/drowsy" are neuroglycopenic symptoms.Variously classified as neurogenic (46) orunclassified (47) by factor analysis, thesymptom of "weak" was found to be neu-roglycopenic with the pharmacologicalmethod (16) as was the symptom of"warm." The latter was classified as neu-rogenic in one factor analysis report (46)but not included in the others (47,48).The symptom "difficulty speaking" wasfound to be neuroglycopenic by factoranalysis (48); it was not a significantsymptom of hypoglycemia in the phar-macological study (16). The symptoms of"incoordination" and "odd behavior"were classified as neuroglycopenic by fac-tor analysis (48); these were not assessedin the pharmacological study (16). Fi-nally, neither "headache" nor "nausea"were found to be significant symptoms ofhypoglycemia in the pharmacologicalstudy (16), and these did not fall into ei-ther the neurogenic or neuroglycopeniccategories on factor analysis (48).

The neurogenic symptoms"shaky/tremulous," "heart pounding,"and "nervous/anxious" are reduced sub-stantially by adrenergic blockade (withthe nonselective a-adrenergic antagonist

phentolamine plus the nonselective j3-ad-renergic antagonist propranolol) duringhypoglycemia (16). Thus, these are ad-renergic neurogenic symptoms of hypo-glycemia. This conclusion is entirely con-sistent with earlier data (45) indicatingthat patients with cervical spinal cordtransections, and thus no sympathochro-maffin system outflow from the CNS, donot exhibit these symptoms (49-51).Since bilaterally adrenalectomized(52,53) and splanchniectomized (54) in-dividuals do not report palpitations dur-ing hypoglycemia, these are probablynormally mediated by epinephrine se-creted from the adrenal medullae. On theother hand, since such patients do reporttremor during hypoglycemia (53,54), thelatter is probably mediated by norepi-nephrine released from sympatheticnerves as well as by circulating epineph-rine (45,55).

In contrast, the neurogenic symp-toms of "sweaty," "hungry," and "tin-gling" are not reduced by adrenergicblockade during hypoglycemia but are re-duced substantially by panautonomicblockade that includes the muscariniccholinergic antagonist atropine (16).Thus, these are predominantly cholin-ergic neurogenic symptoms of hypoglyce-mia. The conclusion that these are neuro-genic symptoms is entirely consistentwith earlier data (45) indicating that pa-tients with cervical spinal cord transec-tions (49-51) and those who had under-gone bilateral adrenalectomy (52,53) orsplanchniectomy (54) do not report thesesymptoms during hypoglycemia. Theconclusion that "sweaty" is a cholinergicsymptom was expected since the dia-phoretic response to hypoglycemia is me-diated by cholinergic sympathetic nerves(45,55).

Awareness of hypoglycemia islargely, perhaps exclusively, the result ofthe perception of neurogenic symptoms(16). This finding is relevant to the clini-cal syndrome of hypoglycemia unaware-ness as discussed later. Interestingly,awareness is largely the result of the per-ception of cholinergic, rather than adren-


Technical Review

ergic, symptoms. Towler et al. (16) foundthat subjects' scoring of the symptom"blood sugar low" during hypoglycemiawas not reduced significantly by adrener-gic blockade but was reduced substan-tially (~70%) by panautonomic blockade(phentolamine and propranolol plus at-ropine).Signs. In contrast to the patient, who re-lies on neurogenic cues, behavioral orcognitive changes, which are manifesta-tions of neuroglycopenia, most oftenprompt recognition of hypoglycemia byobservers of the patient. This is done in-formally by a family member, friend, orco-worker, and by a more formal mentalstatus examination by medical personnel.Diaphoresis and pallor are common find-ings. Tachycardia and a widened pulsepressure are supportive but are often toosubtle to provide strong clues to the diag-nosis.

DiagnosisHypoglycemia is best documented byWhipple's triad: symptoms compatiblewith hypoglycemia, a low plasma glucoseconcentration, and relief of symptoms af-ter the plasma glucose concentrations israised (56). While devices for self-moni-toring of blood glucose are not precise,particularly at low plasma glucose con-centrations (57), they are invaluable inthe management of IDDM outside of ahospital environment since they permitestimation of the glucose level and thusdocumentation of hypoglycemia (as wellas euglycemia or hyperglycemia). Becauseof the imprecision of these devices andbecause the glycemic thresholds forsymptoms may lie at higher plasma glu-cose concentrations in patients withpoorly controlled IDDM (58,59) as dis-cussed later, it is not appropriate to use asingle glucose value to define hypoglyce-mia. If suggestive symptoms are presentthe glucose level should be estimated, ifpossible. If the glucose level is not ele-vated it is reasonable to treat (e.g. for thepatient to ingest carbohydrate) and see ifthe symptoms resolve. If it is not practicalto estimate or measure the plasma glucose

concentration, it is appropriate to treat onthe basis of symptoms alone since the riskof untreated hypoglycemia outweighs therisk of transient hyperglycemia.

While rigid definitions of asymp-tomatic (biochemical) hypoglycemia areuseful for epidemiological studies, moreflexible definitions should be used in themanagement of IDDM. In general, a glu-cose level of 50 mg/dl (2.8 mM) or less istoo low and should prompt treatment.However, in a given patient a level of 60mg/dl (3.3 mM), 70 mg/dl (3.9 mM), oreven higher might well prompt a manage-ment decision (e.g. food ingestion, defer-ral of exercise, change in insulin dose).

TreatmentAn episode of treatment-induced hypo-glycemia requires urgent treatment andshort-term measures to prevent recur-rence of hypoglycemia followed by con-sideration of the probable cause of the ep-isode and any adjustments that should bemade to reduce the likelihood of subse-quent episodes. Because hypoglycemia isso common and usually readily treated inIDDM, it is sometimes taken too lightly byhealth professionals, particularly physi-cians (60).

Most hypoglycemic episodes canbe treated effectively by ingestion of glu-cose per se or carbohydrate-containingjuices or soft drinks, candy, crackers, or ameal (61-64). The glycemic response isbetter correlated with the glucose thanthe carbohydrate content of the feeding(61). Recent data (64) support the earlierrecommendation (61) that a 20-g dose ofglucose (0.3 g/kg in children) is advisable.In a model of insulin-induced hypoglyce-mia in patients with IDDM (64), illus-trated in Fig. 2, 10 g of oral glucose raisedplasma glucose levels from 60 mg/dl (3.3mM) to only 97 mg/dl (5.4 mM) over 30min; the levels started to fall after 60 minand reached placebo levels in <2 h.Twenty grams of oral glucose raisedplasma glucose levels from 58 mg/dl (3.2mM) to 122 mg/dl (6.8 mM) over 45 min,with a greater response at 15 min; again,the levels started to fall after 60 min and

13.0

2.0

11.0

10.0

9.0

8.0

7.0

6.0

5.0

4.0

3.0

Ai•/ *

INSULIN 1

GLUCOSE

^ ^IDDM

220

200

180

160

140 "I

120

100

80

60

0 1 2 3 4 5 6 7 8 4 0

TIME (hours)

Figure 2—Mean (±SE) plasma glucose con-centrations, in a model oj insulin-induced hypo-glycemia in patients with IDDM, in response to 10g (O) and 20 g (•) oj oral glucose, 1.0 mg (A) ojsubcutaneous glucagon, or placebo (stippledarea). From Wiethop and Cryer (64).

approached placebo levels in ~2 h.Clearly, oral glucose is an effective buttemporary measure; persistent hypogly-cemia generally requires subsequent in-gestion of a more substantial meal to pre-vent recurrent hypoglycemia.

When the patient is unable, or un-willing because of neuroglycopenia, totake an oral treatment, glucagon, admin-istered subcutaneously or intramuscu-larly generally by a spouse, family mem-ber, friend or co-worker, is the treatmentof choice outside of the hospital setting(60,64-72). Glucagon stimulates hepaticglucose production. It increases both gly-cogenolysis and gluconeogenesis and isless effective in patients with glycogen de-pletion. The standard dose is 1.0 mg (15jULg/kg in children), but that can producesubstantial hyperglycemia (see Fig. 2). Aswith oral glucose, the glycemic responseis transient; glucose levels begin to fall af-ter —1.5 h (64). Lower doses would pro-duce less hyperglycemia but would be ex-pected to shorten that interval. Thus, thistoo is a temporary measure generally re-quiring subsequent ingestion of food toprevent recurrent hypoglycemia. Vomit-ing sometimes follows treatment of hypo-glycemia with glucagon (60,67).


Technical Review

Glucagon can also be used in theemergency treatment of hypoglycemia byparamedical personnel, although the lat-ter can usually administer glucose intra-venously (73). Although the hormonecan be used in the emergency room set-ting (67), intravenous glucose is perhapsmore appropriate (60). Intranasal gluca-gon, with a time course of action compa-rable to injected glucagon, is a promisingpractical advance in the out-of-hospitaltreatment of severe hypoglycemia (74-79).

Intravenous glucose is the stan-dard treatment of severe hypoglycemia inthe emergency room setting (60) and, in-creasingly, by paramedical personnel out-side of the hospital (73). The usual dose is25 g. Aside from occasional difficulty ob-taining intravenous access, particularly inan uncooperative neuroglycopenic pa-tient, and instances of phlebitis, there islittle to criticize about this rapidly effec-tive therapy. The response, of course, istransient, and subsequent glucose infu-sion, feeding, or both are required.

Experimental approaches to thetreatment of hypoglycemia, based on thepathophysiology of glucose counterregu-lation in IDDM (2), include oral adminis-tration of the amino acid alanine and oraland parenteral administration of the j82-adrenergic agonist terbutaline (64,80).The glycemic response to alanine islargely attributable to stimulation of en-dogenous glucagon secretion; the re-sponse to terbutaline is attributable toepinephrine-like direct j82-adrenergic ac-tions as well as stimulation of endogenousnorepinephrine release (80). The theoret-ical advantage of these agents is that, un-like glucose or glucagon, they producesustained glycemic responses. Therefore,they might be useful in the treatment ofmild hypoglycemia or in the preventionof hypoglycemia when food intake is notanticipated over the following severalhours (e.g. overnight). However, the clin-ical utility of these agents remains to beestablished.

Obviously, as mentioned earlierand discussed later, strategies that reduce

the frequency of iatrogenic hypoglycemiaare preferable to the treatment of hypo-glycemia.

FrequencyHypoglycemia is a fact of life for peoplewith IDDM. The frequency of severe hy-poglycemia can be determined with someprecision. That of mild-moderate symp-tomatic hypoglycemia can be estimated.The frequency of asymptomatic (bio-chemical) hypoglycemia can only be ap-proximated.Asymptomatic hypoglycemia. Usingseven-point multiple sampling during theday, Thorsteinsson et al. (81) docu-mented the fact that the frequency of hy-poglycemia is inversely related to the me-dian plasma glucose concentrationregardless of the glucose level used to de-fine biochemical hypoglycemia. At a eu-glycemic median glucose level of 90mg/dl (5.0 mM), their data predict glu-cose levels <54 mg/dl (3.0 mM) -10% ofthe time. This prediction is supported bythe data of Arias et al. (82). Using contin-uous blood glucose monitoring through-out a 24-h period, glucose levels <50mg/dl (2.8 mM) were found in 9 of 10patients treated to a mean glucose level of100 mg/dl (5.6 mM). These were not iso-lated low values. The mean duration ofglucose levels <50 mg/dl was —2.5 h, thelongest 7 h. Notably, the patients wereaware of only 6 of the 23 episodes (26%).

Asymptomatic hypoglycemia isparticularly common during the night(29,83-86). Gale and Tattersall (86)found 22 of 39 patients (56%) with IDDMto have blood glucose levels <36 mg/dl(2.0 mM) during overnight sampling.These episodes lasted >3 h in 17 (77%)of the affected patients. Only 8 (36%) ofthe 22 patients reported symptoms. Thus,more than one-third of the patients stud-ied had asymptomatic nocturnal hypo-glycemia of substantial proportions.

In general, plasma glucose con-centrations during the night are directlyrelated to the bedtime (as well as themorning) glucose levels (84,85,87-90).For example, in a study of 135 pediatric

patients with IDDM, Shalwitz et al. (90)found that the 10:00 P.M. glucose levelwas 100 mg/dl (5.6 mM) or less on 48%of nights with subsequent 2:00 A.M. glu-cose levels of 60 mg/dl (3.3 mM) or less.The risk of hypoglycemia at 2:00 A.M. wasabout one in four when the 10:00 P.M.glucose level was 100 mg/dl or less. Usinghourly sampling during the night, Pram-ming et al. (84) found even higher rates ofnocturnal hypoglycemia in a study of 58adults with IDDM. They calculated that ifthe bedtime glucose level was <108mg/dl (6.0 mM) the risk of nocturnal hy-poglycemia was 80%.

Clearly, asymptomatic iatrogenichypoglycemia is extraordinarily commonin patients with IDDM.Mild-moderate symptomatic hypogly-cemia. The best estimates of the fre-quency of mild-moderate symptomatichypoglycemia, i.e. symptomatic hypogly-cemia not requiring the assistance of an-other person, are based on data from theSteno Memorial Hospital in Denmark.Pramming et al. (91) studied 411 insulin-treated patients and found that theysuffered an average of 1.8 episodes ofsymptomatic hypoglycemia per week.Extrapolated over a lifetime of IDDM,these data suggest that a given patient suf-fers several thousand episodes of symp-tomatic hypoglycemia. While these pa-tients may have been treated aggressively,at a hospital dedicated to the manage-ment of diabetes, they were not, as agroup, being treated intensively by cur-rent criteria. Approximately 75% of thepatients were using a "split/mixed" insu-lin regimen (intermediate and rapid-acting insulin twice daily). Thus, the fre-quency of mild-moderate symptomatichypoglycemia might well be even higherin intensively treated IDDM.Severe hypoglycemia. Reported fre-quencies of severe treatment-induced hy-poglycemia range from 4.5 to 44% of pa-tients per year and from 5 to 140 episodesper 100 patient-years (92). The reasonsfor the widely discrepant results in thesereports (68,91,93-109) are not entirelyclear. One suspects factors such as differ-


Technical Review

= STANDARD (n = 132)

SEVERE30

g 20COh-ULU

CO

lo 10

26.00.54

9.8017

0 »—»13 38

0.6

0.4

•EXPERIMENTAL (n = 146)

COMA30 r -i 0.6

0.2

Oa>

laD

CO

zLU

2 0

COt—ULU—>CO

10

I — I " Jo23 79

19.90.34

6.1 0.12

0L I—II8 29

0.4 £AoCO

0.2

ZLU

L _ M Jo16 49

NUMBER OF SUBJECTS OR EVENTS

Figure 3—Proportion of patients affected and event rates for severe hypoglycemia and hypoglycemiccoma in the feasibility phase of the Diabetes Control and Complications Trial (DCCT). Data from Ref.100.

ent definitions of severe hypoglycemia(one study [99] required loss of con-sciousness), incomplete ascertainment ofsevere hypoglycemia (in 1 study [91] only6% of severe episodes were treated in ahospital emergency room), and wide vari-ance in the degree of metabolic controlachieved. In any event, this heterogeneityprecludes meaningful meta-analysis ofthese published data.

The Diabetes Control and Com-plications Trial (DCCT) provided infor-mative data on the frequency of severehypoglycemia in IDDM (3,30,100). TheDCCT was a large, prospective, random-ized trial designed to compare the effectsof conventional and intensive therapieson the development and progression ofthe early vascular and neurological com-plications of IDDM. In the 1-year feasibil-ity phase of the DCCT (96) 10% of 132conventionally treated patients suffered atleast one episode of severe hypoglycemia;26% of 146 intensively treated sufferedsevere hypoglycemia (Fig. 3); 6 and 20%,respectively, suffered hypoglycemic

coma. There were 17 and 54 episodes ofsevere hypoglycemia (and 12 and 34 ep-isodes of hypoglycemic coma) per 100patient-years in the conventionally andintensively treated groups respectively. Inthe full scale trial (3)—1,441 patients fol-lowed for a mean of 6.5 years with abroadened definition of severe hypogly-cemia that included episodes requiringassistance even if treatment was only oralcarbohydrate—the event rates wereslightly higher, 19 and 62 episodes of se-vere hypoglycemia per 100 patient-yearsin the conventionally and intensivelytreated groups respectively. Thus on av-erage a patient treated intensively to amean plasma glucose level of 155 mg/dl(8.6 mM) can be projected to suffer anepisode of severe, temporarily disablinghypoglycemia, often with coma or sei-zure, once every 1.6 years. A patienttreated conventionally to a mean plasmaglucose level of 231 mg/dl (12.8 mM) canbe projected to suffer such an episodeonce every 5 years. These should beviewed as minimum estimates since,

based on the experience in the feasibilityphase (100), patients with a history of re-current severe hypoglycemia were ex-cluded from the DCCT. As discussedlater, such a history is a major risk factorfor subsequent severe hypoglycemia (30).Obviously, these projections obscure thefact that there is heterogeneity among pa-tients. Some suffer repeated episodes ofsevere hypoglycemia, others suffer none.Nonetheless, the data document a highfrequency of severe iatrogenic hypoglyce-mia in patients with IDDM, especiallythose attempting to keep plasma glucoselevels as close to the nondiabetic range aspossible.

Aggressive treatment of IDDM isparticularly important before and duringpregnancy. However, intensive therapyduring pregnancy results in a frequencyof severe hypoglycemia comparable to, oreven greater than, that in the nonpreg-nant patient with IDDM (110,111). In-deed, evidence from studies in rats sug-gests that pregnancy per se impairsphysiological defenses against developinghypoglycemia (112).

Impact/OutcomePhysical morbidity. The physical mor-bidity of an episode of hypoglycemia islargely neurological. It ranges from an ar-ray of unpleasant symptoms (Table 2) toimpairment of cognitive function, behav-ioral changes, seizure, and coma. Revers-ible focal neurological deficits (113) anddecerebrate posturing (114) can occur.Generally, these symptoms and signsclear promptly after the plasma glucoseconcentration is raised. Sometimes recov-ery is delayed because of cerebral edema(60). Rarely, there are permanent gener-alized or focal neurological deficits (115).In one such patient a temporal lobe ab-normality, perhaps reflecting hypoglyce-mic brain damage, was found on mag-netic resonance imaging (115). Both thedepth and duration of hypoglycemia areprobably determinants of neurologicaldamage. In monkeys, 5-6 h at a bloodglucose of <20 mg/dl (1.1 mM) were re-


Technical Review

quired for the regular production of neu-rological damage (116).

A major unresolved issue is theextent to which recurrent iatrogenic hy-poglycemia produces permanent impair-ment of brain function over time in pa-tients with IDDM (117,118). There areseveral reports suggesting that childrenwith early-onset IDDM are at risk for de-velopment of cognitive impairment (119—123). In some reports this impairmentwas associated with previous recurrenthypoglycemia (121,123). Even if one as-sumes a cause and effect relationship be-tween recurrent iatrogenic hypoglycemiaand cognitive impairment in young chil-dren, whose brains are still developing, itwould be inappropriate to extrapolatethat to adults.

There are also several reports sug-gesting a relationship between recurrenthypoglycemia and cognitive impairmentin adults with diabetes (124-129). Lan-gan et al. (127), in a study of 100 adultswith IDDM using a retrospective measureof premorbid cognition, found a relation-ship between recurrent hypoglycemiaand lower intelligence quotients (IQ). Arelationship between recurrent hypogly-cemia and lower verbal IQ was also foundwhen this group of patients was con-trasted with a nondiabetic control group(128). On the other hand, despite an in-creased frequency of severe hypoglyce-mia in the intensively treated patients inthe 5-year Stockholm study (130), no dif-ferences in cognitive function betweenthis group and the conventionally treatedgroup were detected. Apparently, thatwas also the case in the DCCT (3), al-though the detailed data from the latterhave not yet been published. Althoughthe statistical power of the Stockholmstudy can be questioned, the DCCT datashould have sufficient power to detect ameaningful effect.

Psychosocial morbidity. Treatment-induced hypoglycemia can produce re-current or even persistent psychologicalmorbidity in IDDM (2,91,131-133).That includes both fear of developing hy-poglycemia and guilt about that fear (2)

and can be a major but unrecognized im-pediment to achieving glycemic control(132). Pramming et al. (91) found theirpatients to be as concerned about the de-velopment of an episode of severe hypo-glycemia as they were about the develop-ment of blindness or renal failure.Wredling et al. (133) found patients withrecurrent severe hypoglycemia to havehigher levels of anxiety and lower levels ofhappiness on psychological testing. Onepatient put it poignantly when she wrote,"Being fearful of hypoglycemia, is itwrong to be afraid? If so, I am" (2). Notonly was she afraid of hypoglycemia, butshe also had been made to feel guiltyabout that rational fear.

Patients with IDDM are barredfrom many forms of employment, and hy-poglycemia is said to be a major concernof prospective employers (134). Drivingperformance is demonstrably impairedduring hypoglycemia (135). Motor vehi-cle accidents attributable to iatrogenic hy-poglycemia have been documented(3,136-138). Interestingly, however,most studies do not indicate higher acci-dent rates in people with diabetes (137-139). A small age-adjusted increase has,however, been reported (140). The man-agement required to prevent hypoglyce-mia as well as hypoglycemia itself can in-trude on other routine activities such asexercise, sports, and social events (134).At best, an episode of hypoglycemia is anuisance and a distraction. It can also beembarrassing, and even lead to ostracismor be mistaken for disorderly or unlawfulbehavior.

Mortality. Unless the event is witnessedand hypoglycemia documented beforedeath, establishing treatment-inducedhypoglycemia as a cause of death is ex-traordinarily difficult (141). Thus, thetrue iatrogenic mortality rate from hypo-glycemia in IDDM is not known. Esti-mates from large retrospective seriesrange from 2 to 13% of deaths of patientswith IDDM (142-147); 4% (143,145) is acommonly quoted figure. Higher (148,149) and lower (150) rates have been re-ported in smaller studies. In an analysis of

35 deaths of patients using continuoussubcutaneous insulin infusion, Teutsch etal. (151) concluded that hypoglycemiawas the probable cause of death in 3;however, the relationship seemed clear inonly 1 patient. On the other hand, severalothers died without a clear explanation.The latter problem was highlighted by aninvestigation of 50 deaths of patients withIDDM by Tattersall and Gill (152). Twopatients had hypoglycemic brain damage,but 22 patients, most of whom retired tobed in apparent good health, were founddead in bed the following morning. Therelationship of this phenomenon to hypo-glycemia, if any, is unknown.

Regardless of the precise fre-quency, it is clear that some patients withIDDM die from hypoglycemia. There is aniatrogenic mortality rate.

Pathophysiology and clinical riskfactorsConventional risk factors. Conven-tional risk factors for treatment-inducedhypoglycemia in IDDM are based on thepremise that absolute or relative insulinexcess is the sole determinant of risk. Be-cause of the imperfections of current in-sulin replacement regimens, absolute orrelative insulin excess must occur fromtime to time in patients with IDDM. Itoccurs, for example, when:

1. Insulin doses are excessive, ill-timed, or of the wrong type.

2. The influx of exogenous glucose isdecreased, as following a missedmeal or snack or during an over-night fast.

3. Insulin-independent glucose utili-zation is increased (with or withoutincreased insulin absorption), asduring exercise.

4. Endogenous glucose production isdecreased, as following alcohol in-gestion.

5. Sensitivity to insulin is increased, asduring effective intensive therapy orafter exercise or in patients with hy-popituitarism or primary adreno-cortical insufficiency.


Technical Review

6. Insulin clearance is decreased, aswith progressive renal insufficiency.

However, the extensive experience of theDCCT indicates clearly that these conven-tional risk factors explain only a minorityof episodes of severe iatrogenic hypogly-cemia in IDDM (30).

The DCCT investigators analyzed714 episodes of severe hypoglycemia in216 patients with IDDM (30). Fifty-fivepercent of the episodes occurred duringsleep. Warning symptoms were absent in36% of the episodes that occurred whilethe patients were awake. Although 75%of the affected patients were in the inten-sive therapy group, the proportions of ep-isodes during sleep and without warningsymptoms were similar in the two groups.To identify risk factors, the DCCT inves-tigators first compared the frequency ofseveral conventional risk factors on thehypoglycemia day with that on a ran-domly selected nonhypoglycemia day. Ofthe factors examined, only a missed mealwas more frequent (P < 0.05) on the hy-poglycemia day. Using a time-dependentproportional hazards statistical model,they then identified five significant riskfactors (relative risk, with 95% confi-dence intervals, shown):

1. Previous severe hypoglycemia 2.5(1.7-3.9).

2. Duration of IDDM 9-12 years 1.7(1.1-2.9).

3. Recent hemoglobin Alc 1% lower1.4(1.2-1.8).

4. Baseline hemoglobin Alc 1% higher1.2(1.0-1.4).

5. Baseline insulin dose 0.1 units/kghigher 1.1 (1.0-1.2).

No other variables added to the risk. Theidentified factors explained only 8.5% ofthe variance in episodes of severe hypo-glycemia. The conventional risk factorswere noticeably absent.

Clearly, we must look beyond theconventional risk factors if we are to un-derstand the pathogenesis of most epi-sodes of severe treatment-induced hypo-glycemia in IDDM. The finding that a

history of prior severe hypoglycemia wasthe most powerful predictor of subse-quent severe hypoglycemia in the DCCT(30) suggests that there are unique fea-tures of patients at high risk. These fea-tures likely include those that compro-mise physiological and behavioraldefenses against hyperinsulinemia anddeveloping hypoglycemia (2). Thus, it ap-pears that iatrogenic hypoglycemia is theresult of the interplay of absolute or rela-tive insulin excess and compromised glu-cose counterregulation in IDDM (2).Syndromes of compromised glucosecounterregulation.Defective glucose counterregulation. Theglucagon secretory response to hypogly-cemia becomes deficient in the first fewyears of IDDM (153,154). This is a selec-tive defect; glucagon responses to otherstimuli are largely, if not entirely, intact.Therefore, it cannot be attributed to astructural abnormality of the a-cells perse and must represent a signaling abnor-mality. The mechanism of this defect isnot known, but it is tightly linked to ab-solute insulin deficiency (155). The defi-cient glucagon response appears to be ab-solute; even substantial hypoglycemiadoes not elicit a glucagon response (156).Because glucagon is normally the primaryglucose counterregulatory hormone, asdiscussed earlier, and a deficient gluca-gon response to falling plasma glucoseconcentrations is the rule in patients withIDDM, glucose counterregulation is al-tered in patients with established IDDM.Yet it appears to be adequate in the firstfew years of IDDM, perhaps because epi-nephrine compensates, at least in part.

A reduced epinephrine secretoryresponse to hypoglycemia developssomewhat later in the course of IDDM(154,156). This, too, is selective; epi-nephrine secretory responses to otherstimuli are intact (157). Therefore, the re-duced epinephrine response cannot be at-tributed to a structural lesion of the adre-nal medullae or the sympathochromaifinreflex arc that mediates all epinephrinesecretory responses, and must represent aregulatory abnormality. In contrast to the

reduced glucagon response, the reducedepinephrine response to a given degree ofhypoglycemia represents, at least in part,a shift to a higher glycemic threshold.Epinephrine responses can be elicited butlower plasma glucose concentrations arerequired (156). Thus, after 5-10 years ofIDDM most patients have no glucagon re-sponses to hypoglycemia and reducedepinephrine (as well as pancreaticpolypeptide) responses to a given degreeof hypoglycemia (64, 156).

The semantics of the glycemicthreshold concept warrant comment. Therationale for the terms used in this reviewand elsewhere (156) is as follows. If alower plasma glucose concentration is re-quired to elicit a given response, the gly-cemic threshold for that response is ele-vated, or higher, since a more intensehypoglycemic stimulus is required toelicit that response. Conversely, if a givenresponse occurs at a higher plasma glu-cose level the threshold for that responseis reduced, or lower, since a less intensehypoglycemic stimulus elicits the re-sponse.

In the setting of absent glucagonresponses, the development of deficientepinephrine responses is a critical eventin the pathophysiology of glucose coun-terregulation in IDDM. Patients withcombined deficiencies of their glucagonand epinephrine responses to fallingplasma glucose concentrations have de-fective glucose counterregulation (2,158-160). When compared with thosewith deficient glucagon but intact epi-nephrine responses, they have beenshown, in prospective studies, to be at25-fold or greater increased risk for se-vere iatrogenic hypoglycemia, at leastduring intensive therapy (158,159), as il-lustrated in Fig. 4.Hypoglycemia unawareness. As many as50% of patients with very longstanding(>30 years) IDDM (91) and an estimated25% of patients overall (161) have theclinical syndrome of hypoglycemia un-awareness (162-167). Affected patientsno longer have the warning neurogenicsymptoms that previously allowed them


Technical Review

100

80

uUJ

CO

3

60

40

20

0

2.46

89

8•1

0.10CD2

2.5

2.0

1.5 3

= ADEQUATE

COUNTERREGULATION

1.0 Z.to

0.5 5DEFECTIVE

COUNTERREGULATION

( n = 9 )

8 2 46

NUMBER OF SUBJECTS OR EVENTS

Figure 4—Proportion of patients affected and event rates for severe hypoglycemia during intensivetherapy of patients with IDDM and adequate or defective glucose counterregulation. Data from Whiteetal. (158).

to recognize developing hypoglycemiaand act (e.g. eat) to prevent its progres-sion to severe hypoglycemia. Therefore,the first symptoms are those of neurogly-copenia, and it is often too late for thepatients to treat themselves. Patients withhypoglycemia unawareness have elevatedglycemic thresholds (lower plasma glu-cose concentrations required) for auto-nomic, including epinephrine, as well assymptomatic responses to hypoglycemia(165,167). Patients with histories of hy-poglycemia unawareness have beenshown, in a prospective study, to be atabout sevenfold increased risk for severeiatrogenic hypoglycemia (168), as illus-trated in Fig. 5.

Elevated glycemic thresholds during inten-sive therapy. The first clue that glycemicthresholds for autonomic and symptom-atic responses to hypoglycemia are dy-namic rather than static was the clinicalimpression that intensively treated pa-tients with IDDM often tolerate lowplasma glucose levels without symptoms.This impression has been amply sup-ported by objective data. During inten-sive therapy of IDDM that effectively low-ers overall plasma glucose levels,glycemic thresholds for autonomic re-sponses and symptoms are elevated

(169,170). These responses can be elic-ited, but lower plasma glucose concentra-tions are required. Conversely, glycemicthresholds are reduced in patients withpoorly controlled IDDM (171). The

100

E 80

to

uUJ

2to

60

40

20

2.83

66

260.48

mechanism(s) of these shifts in glycemicthresholds is not known. However, pro-longed (days) hypoglycemia has beenshown to increase fractional extraction ofglucose into the brain in rats (172). Re-cent data indicate that 56 h of hypoglyce-mia of ~50 mg/dl (2.8 mM) betweenmeals is associated with increased glucoseextraction into the human brain at a givenlevel of hypoglycemia (173).

It is well established, from thelarge prospective experience of the DCCT(3,100), that effective intensive therapyresults in a threefold increase in the inci-dence of severe treatment-induced hypo-glycemia. However, the extent to whichelevated glycemic thresholds for auto-nomic and symptomatic responses to hy-poglycemia contribute to this is not clear.A major as yet unresolved issue is whetherglycemic thresholds for cognitive dys-function during hypoglycemia are also el-evated (lower plasma glucose levels re-quired) during intensive therapy. Data onboth sides of this issue are available. If theglycemic thresholds for cognitive dys-

2.5

2.0 s

1.5

1.0

0.5

3CO

ICO

LU

= AWARE(n=31)

= UNAWARE(n-29)

8 19 15 82

NUMBER OF SUBJECTS OR EVENTSFigure 5—Proportion of patients affected and event rates for severe hypoglycemia in patients withIDDM and normal or reduced awareness of hypoglycemia. Data from Gold et al. (168).


Technical Review

function are not elevated, as suggested bysome data (174-176), it would be reason-able to suggest that elevated glycemicthresholds for autonomic and symptom-atic responses are a major risk factor foriatrogenic hypoglycemia, since neurogly-copenia might precede neurogenic symp-toms. If the glycemic thresholds for cog-nitive dysfunction are also elevated, assuggested by other data (166,177,178),elevated thresholds for autonomic andsymptomatic responses might be less det-rimental.Elevated glycemic thresholds following recenthypoglycemia. Recent antecedent hypo-glycemia reduces autonomic and symp-tomatic responses to hypoglycemia innondiabetic humans (179-181) and pa-tients with IDDM (156,182,183). For ex-ample, Heller and Cryer (179) found thata single <2 h episode of afternoon hypo-glycemia reduced autonomic (includingepinephrine and pancreatic polypeptide)and symptomatic (both neurogenic andneuroglycopenic) responses to hypogly-cemia the following morning in nondia-betic subjects. Similarly, Dagogo-Jack etal. (156) found such an episode of after-noon hypoglycemia to cause both ele-vated glycemic thresholds for autonomic(including epinephrine and pancreaticpolypeptide) and symptomatic (bothneurogenic and neuroglycopenic) re-sponses and impaired physiological de-fense against hyperinsulinemia the fol-lowing morning in patients with IDDM.The potential clinical impact of thesefindings is discussed later (see Hypoglyce-mia-associated autonomic failure). This ef-fect may explain the observation of Gulanet al. (184) that epinephrine responses tohypoglycemia were reduced after 3months of intensive therapy of IDDMwith subcutaneous compared with intra-venous insulin since there were fewer hy-poglycemic episodes with the latter.Elevated glycemic thresholds during ft-ad-renergic blockade. By blocking the j32-adrenergic actions of epinephrine, /3-ad-renergic antagonists impair glucoserecovery from experimental hypoglyce-mia in patients with IDDM (185). This

class of drugs also elevates the glycemicthreshold (lower plasma glucose levels re-quired) for symptoms in IDDM (186).Since both the counterregulatory effectsof epinephrine and awareness of a givenlevel of hypoglycemia are reduced, it isreasonable to suspect that administrationof a j3-adrenergic antagonist might in-crease the frequency of iatrogenic hypo-glycemia in patients with IDDM. Compel-ling clinical support for this expectation islacking, but this issue has not been exam-ined critically in the setting of intensivetherapy of IDDM.Hypoglycemia-associated autonomic failure.As discussed in detail elsewhere (187),the syndromes of defective glucose coun-terregulation, hypoglycemia unaware-ness, and elevated glycemic thresholdsduring intensive therapy (perhaps a resultof recent antecedent hypoglycemia) seg-regate together clinically, are associatedwith a high frequency of severe treat-ment-induced hypoglycemia, and shareseveral pathophysiological features. Thelatter include elevated glycemic thresh-olds for autonomic responses to hypogly-cemia: adrenomedullary (epinephrine),parasympathetic (pancreatic polypep-tide), and sympathetic (neurogenicsymptoms) responses to a given level ofhypoglycemia are reduced (156,187). Inthe setting of absent glucagon responses,reduced epinephrine responses compro-mise physiological defense against devel-oping hypoglycemia. Reduced symptom-atic responses compromise awareness ofdeveloping hypoglycemia and thus theappropriate behavioral response. Thisline of reasoning led one of us (P.E.C.) topropose the concept of hypoglycemia-associated autonomic failure, of whichthese three clinical syndromes are exam-ples (187).

The pathogenesis of hypoglyce-mia-associated autonomic failure is notknown, need not necessarily be the samein all of the clinical syndromes of compro-mised glucose counterregulation, andcould be multifactorial even in a givensyndrome. Indeed, at a clinical level it isprobably best to consider each of the

component syndromes summarized ear-lier as separate, albeit often overlapping,entities, given current lack of a clear un-derstanding of the pathogenesis of eachsyndrome. However, at an investigativelevel some insight has been gained. Thestatus of this incomplete puzzle is sum-marized in the following paragraphs.

Recent antecedent iatrogenic hy-poglycemia may well be one factor in thepathogenesis of hypoglycemia-associatedautonomic failure. The hypothesis, illus-trated in Fig. 6, is that recent antecedentiatrogenic hypoglycemia is a major causeof hypoglycemia-associated autonomicfailure in IDDM and hypoglycemia-asso-ciated autonomic failure, by reducingboth physiological defense against andsymptoms of developing hypoglycemia,results in recurrent severe hypoglycemiathus creating a vicious cycle (187). Asmentioned earlier, recent data supporttwo key elements of this hypothesis: Re-cent antecedent hypoglycemia elevatesglycemic thresholds for autonomic andsymptomatic responses to subsequent hy-poglycemia and impairs physiological de-fense against hyperinsulinemia in pa-tients with IDDM (156). Furthermore,the phenomenon has been shown to bespecific for the stimulus of hypoglycemia;autonomic responses to standing, exer-cise, and a mixed meal are unaltered byrecent antecedent hypoglycemia (188).The phenomenon is not simply the resultof prior activation of the system sinceprior sympathochromafnn system activa-tion (by exercise 90 min earlier) does notreduce the response of that system to sub-sequent hypoglycemia (188). Finally, hy-poglycemia-associated autonomic failureis distinct from classical diabetic auto-nomic neuropathy (156,187).

Many of these phenomena havealso been demonstrated in a study of non-diabetic and diabetic BB/Wor rats (189).It was shown that 1) antecedent hypogly-cemia causes reduced counterregulatoryhormone (glucagon and epinephrine) re-sponses to subsequent hypoglycemia inrats, as it does in humans (156, 179—183); 2) the reduced responses induced


Technical Review

No Glucagon Responses to • Glucose Levels

• Epinephrine

Responses -

Imperfect Insulin

Replacement

-•Episodes of Hypoglycemia-*-

\HYPOGLYCEMIA-ASSOCIATED AUTONOMIC FAILURE

••Symptomatic Responses ( I Awareness)

• • Autonomic (Including Epinephrine) Responses '

Figure 6—Schematic diagram ojthe concept ofhypoglycemia-associated autonomic failure in IDDM.From Cryer (187).

by antecedent hypoglycemia are specificfor the stimulus of hypoglycemia in rats,as they are in humans (188); and 3) dia-betes per se, in the absence of antecedenthypoglycemia, causes reduced counter-regulatory responses to hypoglycemia inrats, as it does in humans (156).

To the extent that recent anteced-ent hypoglycemia is a causal factor in thepathogenesis of hypoglycemia unaware-ness, defective glucose counterregulation,or both, relatively short-term scrupulousavoidance of iatrogenic hypoglycemiashould reduce the frequency of subse-quent iatrogenic hypoglycemia (187). In-deed, a similar phenomenon—elevatedglycemic thresholds for counterregula-tory hormone responses, symptoms anddeterioration of cognitive function—occurs in nondiabetic patients with an in-sulinoma and is completely reversed fol-lowing removal of the tumor (190).Furthermore, Fanelli et al. (191) have re-ported that raising glycemic goals cou-pled with vigorous attempts to avoid hy-poglycemia in intensively treated patientswith relatively short-duration IDDM re-sulted in increased symptomatic, epi-nephrine, and perhaps glucagon re-

sponses to a given level of hypoglycemia 2weeks and 3 months later. The frequencyof hypoglycemic episodes was also re-duced. Glycemic control was compro-mised, but not greatly. These data suggestthat, at least in some patients, the syn-drome of hypoglycemia unawareness isreversible (191,192). If so, it follows thatrelatively short periods of scrupulousavoidance of iatrogenic hypoglycemiamight reduce the overall frequency of iat-rogenic hypoglycemia substantially(192). However, this plausible extrapola-tion remains to be systematically ex-plored.

Dagogo-Jack et al. (193) con-firmed the finding of Fanelli et al. (191)that awareness of hypoglycemia can berestored by scrupulous avoidance of iat-rogenic hypoglycemia in patients selectedinitially for hypoglycemia unawareness.However, they found no change in the(reduced) glucagon, epinephrine, andpancreatic polypeptide responses to hy-poglycemia during reversal of unaware-ness. Thus, it appears that while the syn-drome of hypoglycemia unawareness isreversible, that of defective glucose coun-terregulation may not be reversible (192).

It has also been suggested thattreatment with human compared withanimal insulin might cause hypoglycemiaunawareness in patients with IDDM(194-197). As reviewed elsewhere (198),there are some additional reports seem-ingly consistent with that suggestion(199-201). However, the bulk of the ev-idence does not support it (202-210).Notably, Colagiuri et al. (210) performeda prospective, randomized, double-blind,cross-over comparison of treatment withhuman and porcine insulin in 50 patientswho had previously reported reducedawareness of hypoglycemia after transferfrom porcine to human insulin. Theyfound no differences in the frequency ofhypoglycemia, or in that of hypoglycemiawith reduced or absent awareness, duringtreatment with human or porcine insulin.Thus, it does not appear that treatmentwith human compared with animal insu-lin produces clinically important hypo-glycemia unawareness.

In summary, although insulin ex-cess of sufficient magnitude can be ex-pected to produce hypoglycemia underany condition, the integrity of the glucosecounterregulatory systems, includingtheir effects to both defend against andwarn of developing hypoglycemia, deter-mines whether or not therapeutic hyper-insulinemia results in iatrogenic hypogly-cemia. Since the risk of treatment-induced hypoglycemia is the result of theinterplay of insulin excess and compro-mised glucose counterregulation, it fol-lows that both of these should be consid-ered in attempts to minimize thefrequency of hypoglycemia in IDDM.

HYPOGLYCEMIA IN NIDDMHypoglycemia also occurs in patientswith NIDDM treated with a sulfonylureaor with insulin. Most of the issues justdiscussed in relation to IDDM also applyto NIDDM. However, there are some dif-ferences. These are pointed out in theparagraphs that follow.

While it has been said that olderpatients with NIDDM have fewer neuro-genic symptoms of hypoglycemia (211),


Technical Review

Hepburn et al. (43) found the symptompatterns to be similar in NIDDM andIDDM patients.

Iatrogenic hypoglycemia appearsto be a less frequent problem among pa-tients with NIDDM than those withIDDM (211,212). In one study 20% ofsulfonylurea-treated NIDDM patients re-ported at least one episode of symptom-atic hypoglycemia in the preceding 6months, and 6% reported monthly epi-sodes (213). Rates of sulfonylurea-in-duced severe hypoglycemia of 1.9-2.5per 100 patient-years in NIDDM(212,214,215) contrast with those of19-62 per 100 patient-years in IDDM inthe DCCT (3). However, these figuresmay be misleading. In a preliminary re-port from a prospective study in whichpatients whose NIDDM was not con-trolled with diet were randomly assignedto insulin or sulfonylurea therapy andwhich included the glycemic goal of afasting plasma glucose concentration of<144 mg/dl (8.0 mM), the frequencies ofsymptomatic hypoglycemia over 3 yearswere 35% in those treated with insulin,29% in those treated with glyburide, and12% in those treated with chlorpropam-ide (216). The frequencies of severe hy-poglycemia were 6.6%, 3.7%, and 2.0%respectively. These data suggest that thereis a higher risk of severe hypoglycemiawith aggressive insulin treatment ofNIDDM, and that the frequency of severehypoglycemia in aggressively treatedNIDDM may be higher than previouslythought. Nonetheless, the rate of about1-2% per year (216) is considerablylower than that of —25% per year in theintensively treated IDDM patients in theDCCT (100). However, Hepburn et al.(43) found the frequency of severe hypo-glycemia to be similar in insulin-treatedNIDDM and IDDM matched for durationof insulin therapy. Furthermore, with re-spect to the impact of iatrogenic hypogly-cemia, it should be recalled that the num-ber of patients with NIDDM, and thus atrisk, is many fold greater than the numberof those with IDDM.

Taken together, the data raise the

possibility that while iatrogenic hypogly-cemia is a less frequent problem inNIDDM as a whole, it approaches that ofIDDM in those patients who reach the in-sulin-deficient end of the spectrum ofNIDDM. This possibility warrants furtherinvestigation.

In addition to short-term morbid-ity, it has been estimated that sulfonyl-urea-induced severe hypoglycemia re-sults in permanent neurological deficitsin ~ 5 % of survivors (217) and has a mo-rality rate of -10% (214,215,217-222).

While it is reasonable to considermany of the risk factors for iatrogenic hy-poglycemia in IDDM, discussed earlier, tobe relevant to NIDDM, these have beenlittle studied in NIDDM. With respect tothe pathophysiology of glucose counter-regulation in NIDDM there has been con-siderable controversy, much of it basedon methodological considerations, asnicely reviewed by Heller (223). In astudy of predominantly insulin requiringNIDDM, Bolli et al. (224) found a signif-icantly reduced, but certainly not absent,glucagon secretory response to hypogly-cemia produced by subcutaneous insulininjection. Growth hormone and cortisolsecretory responses were also slightly re-duced, the epinephrine response was notdifferent from control values, and the nor-epinephrine response was slightly in-creased. Although the burst of glucoseproduction that normally occurs earlyduring hypoglycemia was reduced in thepatients with NIDDM, glucose recoveryfrom hypoglycemia was only slightly re-duced because of limited glucose utiliza-tion. The net result was only a moderateprolongation of hypoglycemia. Thus, al-though subtle abnormalities are demon-strable, glucose counterregulation doesnot appear to be markedly defective inpatients with NIDDM. It is, of course,conceivable that studies of patients at theinsulin-deficient extreme of the spectrumof NIDDM might disclose more substan-tive defects in glucose counterregulation.

Additional risk factors for iatro-genic hypoglycemia in NIDDM includeadvancing age, poor nutrition, drug inter-

actions with sulfonylureas, and hepatic orrenal disease leading to altered metabo-lism and excretion of the drugs (211).Among the sulfonylureas, chlorpropam-ide and glyburide are associated with ahigher frequency of hypoglycemia(215,225).

Finally, the treatment of sulfonyl-urea-induced hypoglycemia in NIDDMdiffers from that of severe insulin-inducedhypoglycemia in some important ways.Because sulfonylurea-induced hypogly-cemia can be prolonged and can recur(226,227), hospitalization, with intrave-nous glucose infusion following initialtreatment with glucose injection, isstrongly recommended. Glucagon proba-bly should be avoided because it can stim-ulate endogenous insulin secretion inNIDDM (228). Drugs that inhibit insulinsecretion such as diazoxide (229,230) oroctreotide (231,232) might be consid-ered, in conjunction with glucose infu-sion as necessary.

CONCLUSIONS ANDCLINICAL PRACTICERECOMMENDATIONS

THE PROBLEMIatrogenic hypoglycemia is a major unre-solved problem for many patients withIDDM and some patients with NIDDM. Itis, in fact, the limiting factor in the man-agement of diabetes mellitus (2). Withdocumentation of the fact that effectiveglycemic control makes a difference withrespect to long-term complications (3), itis likely that hypoglycemia will becomean even more common problem in thefuture. From this it follows that 1) thisproblem should be acknowledged andaddressed by all concerned—patients,health care professionals, and the diabe-tes community as a whole—and 2) re-search relevant to hypoglycemia, rangingfrom the study of basic mechanisms tothat of applied clinical strategies, shouldbe encouraged.


Technical Review

RELATIVE RISKThe management of diabetes mellitus isempirical in the sense that one ideally at-tempts to hold plasma glucose concentra-tions as close to the nondiabetic range aspossible as long as one can do so safelyin a given patient. Given the currentlylimited clinical strategies to consistentlyminimize the frequency of iatrogenic hy-poglycemia without compromising glyce-mic control to a greater or lesser extent,judgments based on relative risk must bemade. Some episodes of asymptomatic ormild-moderate symptomatic hypoglyce-mia are a fact of life, at least in IDDM.However, recurrent episodes of severe,temporarily disabling hypoglycemia arenot acceptable. The very real short-termrisks outweigh the likely long-term bene-fits. The ultimate responsibility for guid-ing patients in this judgment lies with thephysician.

MANAGEMENT TECHNIQUES

GeneralThe issue of treatment-induced hypogly-cemia, along with other aspects of diabe-tes care, should be raised in all patientcontacts. The patient's views and con-cerns about hypoglycemia should besought explicitly by the physician orother health care provider involved.

Therapeutic goalsIn the authors' opinion, in many patientsapplication of the principles of intensivetherapy coupled with prudent and indi-vidualized glycemic targets can minimizethe frequency of treatment-induced hy-poglycemia without compromising glyce-mic control completely. It should be re-called that the DCCT data indicate adirect relationship between glycemic con-trol and long-term complications (as wellas an inverse relationship with hypogly-cemia) (3). Any improvement in glycemiccontrol would be expected to reduce therisk of long-term complications. Indeed,although there is not a clear threshold, therelationship is curvilinear with the great-est complications risk reduction by mod-

erate compared with poor glycemic con-trol. Thus, the risk of treatment-inducedhypoglycemia should not be used as anexcuse for not attempting to achieve thebest control possible in a given patient at agiven point in that patient's life. Clearly,however, glycemic targets must be indi-vidualized.

Principles of therapyApplication of the principles of intensivetherapy, albeit with individualized andgenerally higher glycemic targets, is justas important in a patient suffering recur-rent hypoglycemia as it is in a patient ableto achieve near euglycemia safely. Theseprinciples include extensive education,self-monitoring of blood glucose, andprofessional support.

Patients with diabetes, particu-larly those with IDDM or severe, insulin-requiring NIDDM, must manage their di-abetes effectively themselves if they are toachieve glycemic control. Health profes-sionals provide guidance and support,but self-management is the key to suc-cess. Therefore, the patient must be edu-cated about the nuances as well as thebasics of diabetes and become an expertin the management of his or her diabetes.With respect to hypoglycemia, they mustbe taught how to recognize developinghypoglycemia and to treat it effectively.They must keep glucose containing treat-ments with them at all times, at home, atwork, at all other activities, and in be-tween. The last of these warrants empha-sis; whether the distance is long or shortglucose must be available to the patientwhile traveling including, particularly, inthe automobile when the patient is driv-ing. Self-monitoring of blood glucose isparticularly important for patients withhypoglycemia unawareness. They mustmeasure their glucose level frequently,and without fail before performing a crit-ical task such as driving. Spouses, familymembers, friends, and coworkers shouldalso be educated about the recognition ofhypoglycemia and its treatment. The lat-ter includes the use of parenteral gluca-

gon when the patient is unable to takeglucose orally.

Therapeutic regimensAs mentioned earlier, all current insulinreplacement regimens are imperfect com-pared with insulin secretion from normalpancreatic j3-cells. In a patient sufferingfrom recurrent hypoglycemia, use of amore flexible insulin replacement regi-men such as a "basal-bolus" approach (in-termediate- or long-acting insulin pluspreprandial regular insulin) or continu-ous subcutaneous insulin infusion shouldbe considered (233). The goal here is rel-ative glycemic stability, at least initially ata somewhat higher than optimal glucoselevel, rather than near euglycemia. Again,the key issue is the selection of prudentglycemic goals for a given patient.

Risk factor reductionGiven that the risk of treatment-inducedhypoglycemia, at least in IDDM, is deter-mined by the interplay of absolute or rel-ative insulin excess and compromisedglucose counterregulation, it is reason-able to consider each aspect of both com-ponents to attempt to reduce risk factorsfor hypoglycemia (Table 3).

With respect to the conventionalrisk factor category, absolute or relativeinsulin excess: can the insulin regimen(doses, timing, type) be optimized? Par-ticularly if a fixed insulin regimen is used,is the meal plan, including snacks, appro-priate to the regimen and to the patient'slifestyle, preferences, and cultural back-ground? Does the patient understand andfollow it? Does the patient prepare for ex-ercise (food ingestion, decreased insulin,or both), avoid exercise at times of peakinsulin action, and cover unanticipatedexercise wisely? Are there relevant druginteractions, including alcohol? Is there areason, such as improved glycemic con-trol, weight loss, or physical training, tosuspect increased sensitivity to insulinand dictate regimen adjustments? Is therea reason, such as progressive renal insuf-ficiency, to suspect decreased clearance ofadministered insulin? In addition, the un-


Technical Review

Table 3—Risk factor reduction in the patients with recurrent hypoglycemia

Absolute or relative insulin excessCompromised glucose

counterregulation

Insulin doses, timing, and typeMeals and snacksExerciseDrug interactions, including alcoholSensitivity to insulinInsulin clearance (renal disease)

History of severe hypoglycemiaDefective glucose counterregulationHypoglycemia unawarenessEffective intensive therapyRecent antecedent hypoglycemia(/3-Adrenergic antagonists)

common causes of hypoglycemia inde-pendent of diabetes and its treatment (12)should be considered.

With respect to compromisedglucose counterregulation, recall that ahistory of severe hypoglycemia is a well-established predictor of subsequent se-vere hypoglycemia (30) and is probablyindicative of the presence of defective glu-cose counterregulation (158,159). Whileit is possible to test for the latter with aninsulin infusion test (158,159), that isprobably neither cost-effective nor clini-cally worthwhile. A positive test mightlead to a more cautious approach to ther-apy, but management would still be em-pirical and there is no known therapy fordefective glucose counterregulation. Onthe other hand, the history should beprobed for evidence of partial or completehypoglycemia unawareness and its rela-tion to effective glycemic control/recentantecedent hypoglycemia. Hypoglycemiaunawareness has management implica-tions, including the critical need for fre-quent self-monitoring of blood glucose asmentioned earlier. Furthermore, recentdata suggest that a relatively short period(e.g., 2 weeks) of scrupulous avoidance ofhypoglycemia might reverse this syn-drome (190-193) as discussed earlier. Atleast the clinician should point out thateven "mild" episodes of hypoglycemiamay have a long-term impact on the de-velopment of compromised glucosecounterregulation. Finally, althoughcompelling clinical data indicating a det-rimental effect are lacking, it would seemprudent to avoid j3-adrenergic antago-

nists on theoretical grounds if an alterna-tive drug is available. If not, a relativelyselective /^-adrenergic antagonist isprobably preferable to a nonselective an-tagonist.

SUGGESTIONS FOR FUTURERESEARCH— As mentioned earlier,given the magnitude of the problem, re-search relevant to hypoglycemia, rangingfrom the study of basic mechanisms toapplied clinical strategies, should be en-couraged. While a great deal has beenlearned about IDDM, we need to knowmore about the impact of NIDDM on de-fenses against developing hypoglycemia.With respect to IDDM, as well as NIDDM,critical issues include the following. 1)Does the glycemic threshold for cognitivedysfunction shift with those for auto-nomic and symptomatic responses in re-lation to antecedent glycemia? 2) What isthe mechanism(s) of these shifts in glyce-mic thresholds? 3) Do recurrent episodesof iatrogenic hypoglycemia cause perma-nent cognitive impairment? 4) What clin-ical strategies will minimize iatrogenichypoglycemia without compromisingglycemic control? Fundamentally, weneed to learn to replace insulin in a muchmore physiological fashion or to prevent,correct, or compensate for compromisedglucose counterregulation if we are toachieve euglycemia safely in the majorityof patients with diabetes mellitus.

References

1. Frier BM, Fisher BM (Eds.): Hypoglyce-mia and Diabetes. London, Edward Ar-nold, 1993

2. Cryer PE: Iatrogenic hypoglycemia inIDDM: Consequences, risk factors andprevention. In Diabetes Annual. Vol. 7.Marshall SM, Home PD, Alberti KGMM,Krall LP, Eds. Amsterdam, Elsevier,1993,p.317-331

3. The Diabetes Control and Complica-tions Trial Research Group: The effect ofintensive treatment of diabetes on thedevelopment and progression of long-term complications in insulin-depen-dent diabetes mellitus. N EnglJ Med 329:977-986, 1993

4. McCall AL: Effects of glucose depriva-tion on glucose metabolism in the cen-tral nervous system. In Hypoglycemia andDiabetes. Frier BM, Fisher BM, Eds. Lon-don, Edward Arnold, 1993, p. 56-71

5. Owen OE, Morgan AP, Kemp HG, Sulli-van JM, Herrera MG, Cahill GF Jr: Brainmetabolism during fasting. J Clin Invest46:1589-1595, 1967

6. Sokoloff L: Circulation and energy me-tabolism of the brain. In Basic Neuro-chemistry. Siegel G, Agranoff B, AlbersRW, Molinoff P, Eds. New York, Raven,1989,p.565-590

7. Biggers DW, Myers SR, Neal D, StinsonR, Cooper NB, Jaspan JB, Williams PE,Cherrington AD, Frizzell RT: Role ofbrain in counterregulation of insulin in-duced hypoglycemia in dogs. Diabetes37:7-16, 1989

8. Donovan CM, Halter JB, Bergman RN:Importance of hepatic glucoreceptors insympathoadrenal response to hypogly-cemia. Diabetes 40:155-158, 1991

9. Schwartz NS, Clutter WE, Shah SD,Cryer PE: The glycemic thresholds foractivation of glucose counterregulatorysystems are higher than the threshold forsymptoms. J Clin Invest 79:777-781,1987

10. Mitrakou A, Ryan C, Venemen T, EvronW, Jensen T, Cryer P, Gerich J: Hierar-chy of glycemic thresholds for activationof counterregulatory hormone secre-tion, initiation of symptoms and onset ofcerebral dysfunction in normal humans.


Technical Review

AmJ Physiol 260:E67-E74, 199111. Cryer PE: Glucose counterregulation:

The prevention and correction of hypo-glycemia in humans. Am J Physiol 264:E149-E155,1993

12. Cryer PE: Glucose homeostasis and hy-poglycemia. In Williams Textbook oj En-docrinology. 8th ed. Wilson JD, FosterDW, Eds. Philadelphia, Saunders, 1992,p. 1223-1253

13. Thompson CJ, Baylis PH: Endocrinechanges during insulin-induced hypo-glycemia. In Hypoglycemia and Diabetes.

Frier BM, Fisher BM, Eds. London, Ed-ward Arnold, 1993,p. 116-131

14. Berne C, Fagius J: Stimulation of the au-tonomic nervous system. In Hypoglyce-mia and Diabetes. Frier BM, Fisher BM,Eds. London, Edward Arnold, 1993, p.104-115

15. Fisher BM, Frier BM: Haemodynamic re-sponses and functional changes in majororgans. In Hypoglycemia and Diabetes.Frier BM, Fisher BM, Eds. London, Ed-ward Arnold, 1993 p. 144-155

16. Towler DT, Havlin CE, Craft S, CryerPE: Mechanism of awareness of hypogly-cemia: Perception of neurogenic (pre-dominantly cholinergic) rather thanneuroglycopenic symptoms. Diabetes42:1791-1798,1993

17. Beam AG, Billing BH, Sherlock S: Theresponse of the liver to insulin in normalsubjects and in diabetes mellitus: He-patic vein catheterisation studies. ClinSci 11:151-164, 1952

18. Hilsted J, Bonde-Petersen F, NorgaardM-B, Greniman M, Christensen NJ, Par-ving H-H, Suzuki M: Haemodynamicchanges in insulin-induced hypoglyce-mia in normal man. Diabetologia 26:328-332, 1984

19. Fisher BM, Frier BM: Peripheral blood,haemostasis and haemorheology. In Hy-poglycemia and Diabetes. Frier BM, Fisher

BM, Eds. London, Edward Arnold,1993,p.156-164

20. Cryer PE: Glucose counterregulation:The physiological mechanisms that pre-vent or correct hypoglycaemia. In Hypo-glycemia and Diabetes. Frier BM, Fisher

BM, Eds. London, Edward Arnold,1993,p. 34-55

21. Heller SR, Cryer PE: Hypoinsulinemia is

not critical to glucose recovery from hy-poglycemia in humans. Am] Physiol 261:E41-E48, 1991

22. Gerich J, Davis J, Lorenzi M, Rizza R,Bohannon N, Karam J, Lewis S, KaplanS, Shultz T, Cryer PE: Hormonal mech-anisms of recovery from insulin inducedhypoglycemia in man. AmJ Physiol 236:E380-E385, 1979

23. Rizza RA, Cryer PE, Gerich JE: Role ofglucagon, epinephrine and growth hor-mone in human glucose counterregula-tion. J Clin Invest 64:62-71, 1979

24. Kleinbaum J, Shamoon H: Impairedcounterregulation of hypoglycemia ininsulin-dependent diabetes mellitus. Di-abetes 32:493-498, 1983

25. Tse TF, Clutter WE, Shah SD, Cryer PE:Mechanisms of postprandial glucosecounterregulation in man: Physiologicroles of glucagon and epinephrine vis-a-vis insulin in the prevention of hypogly-cemia late after glucose ingestion. J ClinInvest 72:278-286, 1983

26. Rosen SG, Clutter WE, Berk MA, ShahSD, Cryer PE: Epinephrine supports thepostabsorptive plasma glucose concen-tration, and prevents hypoglycemia,when glucagon secretion is deficient inman. J Clin Invest 73:405-411, 1984

27. Boyle PJ, Shah SD, Cryer PE: Insulin,glucagon and catecholamines in the pre-vention of hypoglycemia during fastingin humans. Am J Physiol 256:E651-E661,1989

28. Hirsch IB, Marker JC, Smith LJ, SpinaRJ, Parvin CA, Holloszy JO, Cryer PE:Insulin and glucagon in the preventionof hypoglycemia during exercise in hu-mans. Am J Physiol 260:E695-E704,1991

29. Marker JC, Hirsch IB, Smith LJ, ParvinCA, Holloszy JO, Cryer PE: Cat-echolamines and the prevention of hy-poglycemia during exercise in humans.AmJ Physiol 260:E705-E712, 1991

30. The Diabetes Control and Complica-tions Trial Research Group: Epidemiol-ogy of severe hypoglycemia in the Dia-betes Control and Complications Trial.AmJ Med 90:450-459, 1991

31. Hepburn DA: Symptoms of hypoglycae-mia. In Hypoglycemia and Diabetes. Frier

BM, Fisher BM, Eds. London, Edward

Arnold, 1993, p. 93-10332. Goldgewicht C, Slama G, Papoz L, Tcho-

broutsky G: Hypoglycaemic reactions in172 type I (insulin dependent) diabeticpatients. Diabetologia 24:95-99, 1983

33. Mutch WJ, Dingwall-Fordyce I: Is it ahypo? Knowledge of the symptoms ofhypoglycaemia in elderly diabetic pa-tients. Diabetic Med 2:54-56, 1985

34. Aman J, Karlsson 1, Wranne L: Symp-tomatic hypoglycaemia in childhood di-abetes: a population based questionairestudy. Diabetic Med 6:257-261, 1989

35. Hepburn DA, Eadington DW, PatrickAW, Colledge NR, Frier BM: Symptom-atic awareness of hypoglycaemia: Does itchange on transfer from animal to hu-man insulin? Diabetic Med 6:586-590,1989

36. McFarlane PI, Walters M, Stutchfield P,Smith CS: A prospective study of symp-tomatic hypoglycaemia in childhood di-abetes. Diabetic Med 6:627-630, 1989

37. Ward CM, Stewart AW, Cutficld RG:Hypoglycaemia in insulin dependent di-abetic patients attending an outpatients'clinic. NZ Med J 103:339-341, 1990

38. Muhlhauser I, Heinemann L, Fritsche E,Von Lennep K, Berger M: Hypoglycemicsymptoms and frequency of severe hy-poglycemia in patients treated with hu-man and animal insulin preparations.Diabetes Care 14:745-749, 1991

39. Pramming S, Thorsteinsson B, BendtsonI, Binder C: Symptomatic hypoglycemiain 411 type I diabetic patients. DiabeticMed 8:217-222, 1991

40. Pennebaker JW, Cox DJ, Gonder-Fred-erick L, Wunch MG, Evans WS, Pohl SL:Physical symptoms related to blood glu-cose in insulin-dependent diabetics.Psychosomatic Med 43:489-500, 1981

41. Cox DJ, Gonder-Frederick L, Pohl S,Pennebaker JW: Reliability of symptom-blood glucose relationships among insu-lin-dependent adult diabetics. Psychoso-matic Med 45:357-360, 1983

42. Cox DJ, Clarke WL, Gonder-FrederickL, Pohl S, Hoover C, Snyder A, Zimbel-man L, Carter WR, Bobbitt S, Penne-baker J: Accuracy of perceiving bloodglucose in IDDM. Diabetes Care 8:529-536,1985

43. Hepburn DA, MacLeod KM, Pell ACH,


Technical Review

Scougal 1J, Frier BH: Frequency andsymptoms of hypoglycaemia experi-enced by patients with type 2 diabetestreated with insulin. Diabetic Med 10:231-237, 1993

44. Cryer PE, Binder C, Bolli GB, Cher-rington AD, Gale EAM, Gerich JE, Sher-win RS: Hypoglycemia in IDDM. Diabe-tes 38:1193-1199, 1989

45. Heller SR, Macdonald IA: Physiologicaldisturbances in hypoglycaemia: Effecton subjective awareness. ClinSci 81:1-9,1991

46. Hepburn DA, Deary IJ, Frier BM, PatrickAW, Quinn JD, Fisher BM: Symptoms ofacute insulin-induced hypoglycaemia inhumans with and without IDDM: Factoranalysis approach. Diabetes Care 14:949-957, 1991

47. Hepburn DA, Deary IJ, Frier BM: Classi-fication of symptoms of hypoglycaemiain insulin-treated diabetic patients usingfactor analysis: relationship to hypogly-caemia unawareness. Diabetic Med 9:70-75,1992

48. Deary IJ, Hepburn DA, MacLeod KM,Frier BM: Partitioning the symptoms ofhypoglycaemia using multi-sample con-firmatory factor analysis. Diabetologia36:771-777, 1993

49. Corrall RJM, Frier BM, McClemont EJW,Taylor SJ, Christie NE: Recovery mech-anisms from acute hypoglycaemia incomplete tetraplegia. Paraplegia 17:314—318,1979

50. Mathias CJ, Frankel JL, Turner RC,Christensen NJ: Physiological responsesto insulin hypoglycemia in spinal man.Paraplegia 17:319-326, 1979

51. Frier BM, Corrall RJM, Ratcliffe JG,Ashby JP, McClemont EJW: Autonomicneurol control mechanisms of substrateand hormonal responses to acute hypo-glycaemia in man. Clin Endocrinol 14:425-433, 1981

52. Ginsburg J, Paton A: Effects of insulinafter adrenalectomy. Lancet i:491-494,1956

53. Altorfer RM, Ziegler WH, Froesch ER:Insulin hypoglycaemia in normal andadrenalectomized subjects: comparisonof metabolic parameters and endocrinecounterregulation. Acta Endocrinol 98:413-419, 1981

54. French EB, Kilpatrick R: The role ofadrenaline in hypoglycaemic reactionsin man. ClinSci 14:639-651, 1955

55. Macdonald IA, Maggs DG: Cutaneousblood flow, sweating, tremor and tem-perature regulation in hypoglycaemia.In Hypoglycemia and Diabetes. Frier BM,

Fisher BM, Eds. London, Edward Ar-nold, 1993,p. 132-143

56. Whipple AO: The surgical therapy of hy-perinsulinism. J Int Chir 3:237-276,1938

57. Havlin CE, Parvin CA, Cryer PE: Bloodglucose monitoring devices provide onlyestimates of glucose levels. Clin Diabetes9:92-93, 1991

58. Boyle PJ, Schwartz NS, Shah SD, ClutterWE, Cryer PE: Plasma glucose concen-trations at the onset of hypoglycemicsymptoms in patients with poorly con-trolled diabetes and nondiabetics. NEnglJMed 318:1487-1492, 1988

59. Amiel SA, Sherwin RS, Simonson DC,Tamborlane WV: Effect of intensive in-sulin therapy on glycemic thresholds forcounterregulatory hormone release. Di-abetes 37:901-907, 1988

60. MacCuish AC: Treatment of hypogly-caemia. In Hypoglycemia and Diabetes.Frier BM, Fisher BM, Eds. London, Ed-ward Arnold, 1993,p. 212-221

61. Brodows RG, Williams C, Amatruda JM:Treatment of insulin reactions in diabet-ics. JAMA 252:3378-3381, 1984

62. Slama G, Traynard P-Y, Desplanque N,Pudar H, Dhunputh I, Letanoux M,Bornet FRJ, Tchobroutsky G: The searchfor an optimized treatment of hypogly-cemia. Arch Intern Med 150:589-593,1990

63. Moller J, Laker MF, Gillespie SM,Ovesen PG, Abildgaard N, Tian R, Jor-gensen JOL, Moller N: Lack of effects ofhypoglycemia on glucose absorption inhealthy men. Diabetes Care 15:1264-1266.1992

64. Wiethop BV, Cryer PE: Alanine and ter-butaline in the treatment of hypoglyce-mia in IDDM. Diabetes Care 16:1131-1136.1993

65. Elrick H, Witten TA, Arai Y: Glucagontreatment of insulin reactions. N Engl JMed 258:476-480, 1958

66. MacCuish AC, Muviro JF, Duncan LJP:

Treatment of hypoglycaemic coma withglucagon, intravenous dextrose, andmannitol infusion in a hundred diabet-ics. Lancet ii:946-949, 1970

67. Collier A, Steedman DJ, Patrick AW,Nimmo GR, Matthews DM, MaclntyreCCA, Little K, Clarke BF: Comparison ofintravenous glucagon and dextrose intreatment of severe hypoglycemia in anaccident and emergency department.Diabetes Care 10:712-715, 1987

68. Mulhauser I, Berger M, Sonnenberg G,Koch J, Jorgens V, Schernthaner G,Scholz V: Incidence and management ofsevere hypoglycaemia in 434 adults withinsulin-dependent diabetes mellitus. Di-abetes Care 8:268-273, 1985

69. Aman J, Wranne L: Hypoglycaemia inchildhood diabetes. I. Clinical signs andhormonal counterregulation. Acta Paedi-atrScand 77:542-547, 1988

70. Aman J, Wranne L: Hypoglycaemia inchildhood diabetes. II. Effect of subcuta-neous or intramuscular injection of dif-ferent doses of glucagon. Acta PaediatrScand 77:548-553, 1988

71. Namba M, Hanafusa T, Kono N, Tarui S,the GL-G Hypoglycemia Study Group:Clinical evaluation of biosynthetic glu-cagon treatment for recovery from hypo-glycemia developed in diabetic patients.Diabetes Res Clin Pract 19:133-138,1993

72. Hvidberg AM, Jorgensen S, Hilsted J:The effect of genetically engineered glu-cagon on glucose recovery after hypo-glycaemia in man. Br J Clin Pharmacol

34:547-550, 199273. Weston C, Stephens M: Hypoglycaemic

attacks treated by ambulance personnelwith extended training. Br Med J 300:908-909, 1990

74. Freychet L, Rizkalla SW, Desplanque N,Basdevant A, Zirinis P, Tchobroutsky G,Slama G: Effect of intranasal glucagon onblood glucose levels in healthy subjectsand hypoglycaemic patients with insulindependent diabetes. Lancet i: 1364—1366,1988

75. Pontiroli AE, Calderara A, Pajetta E, Al-beretto M, Pozza G: Intranasal glucagonas a remedy for hypoglycemia. DiabetesCare 12:604-608, 1989

76. Pontiroli AE, Pozza G: Intranasal admin-


Technical Review

istration of peptide hormones: Factorsaffecting transmucosal absorption. Dia-betic Med 7:770-774, 1990

77. Slama G, Alamowitch C, Desplanque N,Letanoux M, Zirinis P: A new non-inva-sive method for treating insulin-reac-tion: Intranasal lyophyllized glucagon.Diabetologia 33:671-674, 1990

78. Slama G, Reach G, Cahane M, Quetin C,Villanove-Robin F: Intranasal glucagonin the treatment of hypoglycaemic at-tacks in children: Experience at a sum-mer camp (Letter). Diabetologia 35:398,1992

79. Rosenfalck AM, Bendtson I, Jorgensen S,Binder C: Nasal glucagon in the treat-ment of hypoglycaemia in type 1 (insu-lin-dependent) diabetic patients. Diabe-tes Res Clin Pract 17:43-50, 1992

80. Wiethop BV, Cryer PE: Glycemic actionsof alanine and terbutaline in 1DDM. Di-abetes Care 16:1124-1130, 1993

81. Thorsteinsson B, Pramming S, LauritzenT, Binder C: Frequency of daytime bio-chemical hypoglycemia in insulin-treated diabetics: relationship to dailymedian blood glucose concentration.Diabetic Med 3:147-151, 1986

82. Arias P, Kerner W, Zier H, Navacues 1,Pfeiffer E: Incidence of hypoglycemic ep-isodes in diabetic patients under contin-uous subcutaneous insulin infusion andintensified conventional insulin treat-ment: Assessment by means of continu-ous blood glucose monitoring. DiabetesCare 8:134-140, 1985

83. Dornan TL, Peckar CO, Mayon-WhiteVA, Knight AH, Moore RA, HockadayDR, Bron AJ, Turner RC: Unsuspectedhypoglycaemia, hemoglobin A, and di-abetic control. QJMed 197:31-38,1981

84. Pramming S, Thorsteinsson B, Bendtson1, Ronn B, Binder C: Nocturnal hypogly-caemia in patients receiving conven-tional treatment with insulin. Br Med J291:376-379, 1985

85. Whincup G, Milner RDG: Predictionand management of nocturnal hypogly-caemia in diabetes. Arch Dis Child 62:333-337,1987

86. Gale EAM, Tattersall RB: Unrecognizednocturnal hypoglycaemia in insulin-treateddiabetes, lancet i: 1049-1052, 1979

87. Havlin CE, Cryer PE: Nocturnal hypo-

glycemia does not commonly result inmajor morning hyperglycemia in pa-tients with diabetes mellitus. DiabetesCare 10:141-147, 1987

88. Schiffrin A, Suissa S: Predicting noctur-nal hypoglycemia in patients with type 1diabetes treated with continuous insulininfusion. Am] Med 82:1127-1132,1987

89. Lerman IG, Wolfsdorf JI: Relationship ofnocturnal hypoglycemia to daytime gly-cemia in IDDM. Diabetes Care 11:636-642,1988

90. Shalwitz RA, Farkas-Hirsch R, WhiteNH, Santiago JV: Prevalence and conse-quences of nocturnal hypoglycemiaamong conventionally treated childrenwith diabetes mellitus. J Pediatr 116:685-689, 1990

91. Pramming S, Thorsteinsson B, BendtsonI, Binder C: Symptomatic hypoglycae-mia in 411 type I diabetic patients. Dia-betic Med 8:217-222, 1991

92. Tattersall RB: Frequency and causes ofhypoglycaemia. In Hypoglycemia and Di-abetes. Frier BM, Fisher BM, Eds. Lon-don, Edward Arnold, 1993, p. 176-189

93. Goldstein DE, England JD, Hess R, Raw-lings SS, Walker B: A prospective studyof symptomatic hypoglycemia in youngdiabetic patients. Diabetes Care 4:601-605,1981

94. Potter J, Clarke P, Gale EAM, Dave SH,Tattersall RB: Insulin induced hypogly-caemia in an accident and emergencydepartment: The tip of an iceberg? BrMed] 285:1180-1182, 1982

95. Basdevant A, Castigliola D, Lanoe JL,Goldewicht C, Triomphe A, Metz F, De-nys H, Eschwege E, Fardeau M, Tcho-broutsky G: The risk of diabetic control:A comparison of hospital versus generalpractice supervision. Diabetologia 22:309-314, 1982

96. Goldewicht C, Slama G, Papoz L, Tcho-broutsky G: Hypoglycaemic reactions in172 type 1 (insulin-dependent) diabeticpatients. Diabetologia 24:95-99, 1983

9 7. Casparie AF, Elving LD: Severe hypogly-cemia in diabetic patients: Frequency,causes, prevention. Diabetes Care8:141-145,1985

98. Moses RG, Hubert PA, Lewis-Driver DJ:Severe hypoglycaemia: A one year pro-spective survey in Wollongong. Med J

Aust 142:294-296, 198599. Muhlhauser I, Bruckner 1, Berger M,

Cheta D, Jorgens V, Ionescu-TirgovisteC, Sholz V, Mineu 1: Evaluation of anintensified insulin treatment and teach-ing programme as routine managementof type 1 (insulin-dependent) diabetes.Diabetologia 30:681-690, 1987

100. The Diabetes Control and Complica-tions Trial Research Group: DiabetesControl and Complications Trial(DCCT): Results of feasibility study. Di-abetes Care 10:1-19, 1987

101. Nilsson A, Tideholm B, Kalen J, Katz-man P: Incidence of severe hypoglyce-mia and its causes in insulin-treated di-abetes. Ada Med Scand 224:257-262,1988

102. Clausen-Sjobom N, Adamson U, LinsP-E: The prevalence of impaired glucosecounter-regulation during an insulin in-fusion test in insulin-treated diabetic pa-tients prone to severe hypoglycaemia.Diabetologia 32:818-825, 1989

103. Bergada I, Suissa S, Dufrenesne J, Schif-frin A: Severe hypoglycemia in IDDMchildren. Diabetes Care 12:239-244,1989

104. Daneman D, Frank M, Perlman K: Se-vere hypoglycemia in children with in-sulin dependent diabetes mellitus: Fre-quency and predisposing factors. JPediatr 115:681-685, 1989

105. AmanJ, Karlsson I, Wranne L: Symp-tomatic hypoglycaemia in childhood di-abetes: A population-based questionairestudy. Diabetic Med 6:257-261, 1989

106. Ward CM, Stewart AW, Cutfield RG:Hypoglycaemia in insulin diabetic pa-tients attending an outpatient clinic. N ZMed J 103:339-341, 1990

107. Bjork E, Palmer M, Schvarcz E, Berne C:Incidence if severe hypoglycaemia in anunselected population of patients withinsulin-treated diabetes melitus, withspecial reference to autonomic neuropa-thy. Diabetes Nutr Metab 4:303-309,1990

108. Reichard P, Rosenqvist U, Britz A: Inten-sified conventional insulin treatmentand neuropsychological inpairment. BrMcdJ 303:1439-1442, 1991

109. MacLeod KM, Hepburn DA, Frier BM:Frequency and morbidity of severe hy-


Technical Review

poglycaemia in insulin-treated diabeticpatients. Diabetic Med 10:238-245,1993

110. Rayburn W, Piehl E, Jacober S, SchorkA, Ploughman L: Severe hypoglycemiaduring pregnancy: Its frequency andpredisposing factors in diabetic women.MJ Gynaecol Obstet 24:263-268, 1986

111. Kimmerlie R, Heinemann L, Delecki A,Berger M: Severe hypoglycemia inci-dence and predisposing factors in 85pregnancies in type 1 diabetic women.Diabetes Care 15:1034-1037, 1992

112. Rossi G, Lapaczewski P, Diamond MP,Jacob RJ, Shulman G, Sherwin RS: In-hibitory effect of pregnancy on counter-regulatory hormone responses to hypo-glycemia in the awake rat. Diabetes 42:440-445, 1993

113. Wayne EA, Dean HJ, Booth F, Tenen-bein M: Focal neurologic deficits associ-ated with hypoglycemia in children withdiabetes. J Pediatr 117:575-577, 1990

114. Seibert DG: Reversible decerebrate pos-turing seondary to hypoglycemia. Am]Med 78:1036-1037, 1985

115. Chalmers J, Risk MTA, Kean DM, GrantR, Ashworth B, Campbell IW: Severeamnesia after hypoglycemia. DiabetesCare 14:922-925, 1991

116. Kahn KJ, Myers RE: Insulin-induced hy-poglycaemia in the non-human primate.I. Clinical consequences. In Brain Hy-poxia, Clinics in Developmental Medicine,

London, Heineman, 1971, p. 185-193117. Ryan CM: Neurobehavioral complica-

tions of type I diabetes. Diabetes Care 11:86-93, 1988

118. Gold AE, Deary 1J, Frier BM: Recurrentsevere hypoglycaemia and cognitivefunction in type 1 diabetes. Diabetic Med10:503-508, 1993

119. Ryan C, Vega A, Drash A: Cognitive def-icits in adolescents who developed dia-betes early in life. Pediatrics 75:921-927,1985

120. Holmes CS, Richman LC: Cognitive pro-files of children with insulin dependentdiabetes. J Dev Behav Pediatr 6:323-326,1985

121. RovetJF, Ehrlich RM, Hoppe M: Intel-lectual deficits associated with early on-set of insulin-dependent diabetes melli-tus in children. Diabetes Care 10:510—515,1987

122. RovetJF, Ehrlich RM, Hoppe M: Specificintellectual deficits in children withearly onset diabetes mellitus. Child Dev59:226-234, 1988

123. Golden MP, Ingersoll GM, Brack CJ,Russell BA, Wright JC, Huberty TJ: Lon-gitudinal relationship of asymptomatichypoglycemia to cognitive function inIDDM. Diabetes Care 12:89-93, 1989

124. Bale RN: Brain damage in diabetes mel-litus. BrJ Psychiatry 122:337-341,1973

125. SkenazyJA, Bigler ED: Neuropsycholog-ical findings in diabetes mellitus. J ClinPsychol 40:246-250, 1984

126. Wredling R, Levander S, Adamson U,Lins P-E: Permanent neuropsychologi-cal impairment after recurrent episodesof severe hypoglycaemia in man. Diabe-tologia 33:152-157, 1990

127. Langan SJ, Deary IJ, Hepburn DA, FrierBM: Cumulative cognitive impairmentfollowing recurrent severe hypoglyce-mia in adult patients with insulin-treated diabetes mellitus. Diabetologia34:337-344, 1991

128. Deary IJ, Crawford JR, Hepburn DA,Langan SJ, Blackmore LM, Frier BM: Se-vere hypoglycemia and intelligence inadult patients with insulin-treated dia-betes. Diabetes 42:341-344, 1993

129. Deary IJ, Langan SJ, Graham KS, Hep-burn DA, Frier BM: Recurrent severe hy-poglycaemia, intelligence and speed ofinformation processing. Intelligence 16:337-359, 1992

130. Reichard P, Britz A, Rosenqvist U: Inten-sified conventional insulin treatmentand neuropsychological impairment. BrMed] 303:1439-1432, 1991

131. Cox DJ, Irvine A, Gonder-Frederick L,Nowacek G, Butterfield J: Fear of hypo-glycemia: Quantification, validation andutilization. Diabetes Care 10:617-621,1987

132. Cox DJ, Gonder-Frederick L, Antoun B,Clarke WE, Cryer PE: Psychobehavioralmetaoblic paramenters of severe hypo-glycemic episodes. Diabetes Care 13:458-459, 1990

133. Wredling RAM, Theorell PGT, Roll HM,Lins PES, Adamson UKC: Psychosocialstate of patients with IDDM prone to re-current episodes of severe hypoglyce-mia. Diabetes Care 15:518-520, 1992

134. Gill G: Socioeconomic problems of hy-poglycemia. In Hypoglycemia and Diabe-tes. Frier BM, Fisher BM, Eds. London,Edward Arnold, 1993, p. 362-370

135. Cox DJ, Gonder-Frederick L, ClarkeWL: Driving decrements in type 1 dia-betes during moderate hypoglycemia.Diabetes 42:239-243, 1993

136. Frier BM, Matthews DM, Steel JM, Dun-can LJP: Driving and insulin-dependentdiabetes. Lancet i: 1232-1234, 1980

137. Eadington DW, Frier BM: Type I diabe-tes and driving experience: an eight yearcohort study. Diabetic Med 6:137-141,1989

138. Stevens AB, Roberts M, McKane R, At-kinson AB, Bell PM, Hayes JR: Motor ve-hicle driving among diabetics taking in-sulin and non-diabetics. Br Med J 299:591-595, 1989

139. Songer TJ, LaPorte RE, Dorman JS, Or-chard TJ, Cruickshanks KJ, Becker DJ,Drash AL: Motor vehicle accidents andIDDM. Diabetes Care 11:701-707,1988

140. Hansotia P, Broste SK: The effect of epi-lepsy or diabetes mellitus on the risk ofautomobile accidents. N EnglJ Med 324:22-26, 1991

141. Tattersall RB, Gale EAM: Mortality. InHypoglycemia and Diabetes. Frier BM,

Fisher BM, Eds. London, Edward Ar-nold, 1993,p.190-198

142 Paz-Guevara AT, Hsu T-H, White P: Ju-venile diabetes mellitus after forty years.Diabetes 24:559-565, 1975

143. Deckert T, Poulsen JE, Larsen M: Prog-nosis of diabetics with diabetes beforethe age of thirty-one. I. Survival, cause ofdeaths and complications. Diabetologia14:363-370, 1978

144. Nabarro JDN, Mustaffa BE, Morris DV,Walport MJ, Kurtz AB: Insulin deficientdiabetes: Contrasts with other endocrinedeficiencies. Diabetologia 16:5-12,1979

145. Tunbridge WMG: Factors contributingto deaths of diabetics under 50 years ofage. Lancet ii:569-572, 1981

146. Connell FA, Louden JM: Diabetes mor-tality in persons under 45 years of age.Am] Pub Health 73:1174-1177, 1983

147. Borch-Johnsen K, Nisan H, HendriksenE, Kreiner S, Sailing N, Deckert T, Ne-rup J: The natural history of insulin-dependent diabetes mellitus in Den-


Technical Review

mark: Long term survival with andwithout late diabetic complications. Di-abetic Med 4:201-210, 1987

148. Lestradet H, Papoz L, Hellouin de Meni-bus CL, Levasseur F, Besse J, Billaud L,Battisteli F, Trie Ph, Lestradet F: Long-term study of mortality and vascularcomplications in juvenile onset (type 1)diabetes. Diabetes 30:175-179, 1981

149. Joner J, Patrick S: The mortality of chil-dren with type 1 (insulin-dependent) di-abetes mellitus in Norway, 1973-1988.Diabetologia 34:29-32, 1991

150. Scibilia J, Finegold D, DormanJ, BeckerD, Drash A: Why do children with dia-betes die? Acta Endocrinol 279 (Suppl.):325-333,1986

151. Teutsch SM, Herman WH, Dwyer DM,Lane MJ: Mortality among diabetic pa-tients using continuous subcutaneousinsulin infusions pumps. N Engl ] Med310:361-368, 1984

152. Tattersall RB, Gill GV: Unexplaineddeaths of type I diabetic patients. Dia-betic Med 8:49-58, 1991

153. GerichJ, Langlois M, Noacco C, KaramJ, Forsham P: Lack of glucagon responseto hypoglycemia in diabetes: Evidencefor an intrinsic pancreatic alpha-cell de-fect. Science 182:171-173, 1973

154. Bolli G, DeFeo P, Compagnucci P,Cartechini MG, Angeletti G, Santeusa-nio F, Brunetti P, Gerich JE: Abnormalglucose counterregulation in insulin-dependent diabetes mellitus: Interactionof anti-insulin antibodies and impairedglucagon and epinephrine secretion. Di-abetes 32:134-141, 1983

155. Fukuda M, Tanaka A, Tahara Y, IkegamiH, Yamamoto Y, Kumahara Y, Shima K:Correlation between minimal secretorycapacity of pancreatic /3-cells and stabil-ity of diabetic control. Diabetes 37:81-88,1988

156. Dagogo-Jack SE, Craft S, Cryer PE: Hy-poglycemia-associated autonomic fail-ure in insulin dependent diabetes melli-tus. J Clin Invest 91:819-828, 1993

157. Hirsch BR, Shamoon H: Defective epi-nephrine and growth hormone re-sponses in type 1 diabetes are stimulusspecific. Diabetes 36:20-26, 1987

158. White NH, Skor DA, Cryer PE, Bier DM,Levandoski L, Santiago JV: Identifica-

tion of type 1 diabetic patients at in-creased risk for hypoglycemia during in-tensive therapy. N Engl J Med 308:485-

491.1983159. Bolli GB, DeFeo P, DeCosmo S, Periello

G, Ventura MM, Massi-Benedetti M,Santeusanio F, Gerich JE, Brunetti P: Areliable and reproducible test for ade-quate glucose counterregulation in typeI diabetes mellitus. Diabetes 33:732-

737.1984160. Sjobom NC, Adamson U, Lins PE: The

prevalence of impaired glucose counter-regulation during an insulin infusiontest in insulin-treated patients prone tosevere hypoglycaemia. Diabetologia 32:818-825, 1989

161. Gerich JE, MokanM, VenemanT, Koryt-kowski M, Mitrakou A: Hypoglycemiaunawareness. Endocrine Rev 12:356-371,1991

162. Heller SR, Herbert M, Macdonald IA,Tattersall RB: Influence of sympatheticnervous system on hypoglycemic warn-ing symptoms. Lancet ii:359—363, 1987

163. Hepburn DA, Patrick AW, EadingtonDW, Ewing DJ, Frier BM: Unawarenessof hypoglycaemia in insulin-treated dia-betic patients: Prevalence and relation-ship to autonomic neuropathy. DiabeticMed 7:711-717, 1990

164. Ryder REJ, Owens DR, Hayes TM,Ghatei M, Bloom SR: Unawareness ofhypoglycaemia and inadequate glucosecounterregulation: No causal relation-ship with diabetic autonomic neuropa-thy. Br Med] 301:783-787, 1990

165. Grimaldi A, Bosquet F, Davidoff P, DigyJP, Sachon C, Landault C, Thervet F,Zoghbi F, Legrand JC: Unawareness ofhypoglycemia by insulin-dependent di-abetics. Horm Metab Res 22:90-95,1990

166 Clarke WL, Gonder-Frederick LA, Rich-ards FE, Cryer PE: Multifactorial originof hypoglycemic symptom awareness ininsulin dependent diabetes mellitus. Di-abetes 40:680-685, 1991

167. Hepburn DA, Patrick AW, Brash HM,Thomson I, Frier BM: Hypoglycaemiaunawareness in type I diabetes: A lowerplasma glucose is required to stimulatesympathoadrenal activation. DiabeticMed 8:934-945, 1991

168. Gold AE, MacLeod KM, Frier BM: Fre-

quency of severe hypoglycemia in pa-tients with type I (insulin dependent) di-abetes with impaired awareness ofhypoglycemia. Diabetes Care. In press

169. Amiel SA, Tamborlane WV, SimonsonDC, Sherwin RS: Defective glucosecounterregulation after strict control ofinsulin-dependent diabetes mellitus. NEnglJ Med 316:1376-1383, 1987

170. Amiel SA, Sherwin RS, Simonson DC,Tamborlane WV: Effect of intensive in-sulin therapy on glycemic thresholds forcounterregulatory hormone release. Di-abetes 37:901-907, 1988

171. Boyle PJ, Schwartz NS, Shah SD, ClutterWE, Cryer PE: Plasma glucose concen-trations at the onset of hypoglycemicsymptoms in patients with poorly con-trolled diabetes and in nondiabetics. NEnglJ Med 318:1487-1492, 1988

172. McCall AL, Fixman LB, Fleming N,Tornheim K, Chick W, Ruderrnan NB:Chronic hypoglycemia increases brainglucose transport. Am ] Physiol 251:E442-E447,1986

173. Boyle PJ, Nagy RJ, O'Connor AM, Kem-pers SF, Yeo RA, Quails C: Adaptation inbrain glucose uptake following recurrenthypoglycemia. Proc Natl Acad Set USA. Inpress

174. Amiel SA, Pottinger RC, Archibald HR,Chusney G, Cunnah DTF, Prior PF, GaleEAM: Effect of antecedent glucose con-trol on cerebral function during hypogly-cemia. Diabetes Care 14:109-118,1991

175. Widom B, Simonson D: Glycemic con-trol and neuropsychologic function dur-ing hypoglycemia in patients with insu-lin dependent diabetes mellitus. AnnIntern Med 112:904-912, 1990

176. Maran A, Lomas J, Macdonald I, AmielS: Lack of protection of cerebral functionin well controlled diabetic patients withhypoglycemia unawareness (Abstract).Diabetes 42:17A, 1993

177. Ziegler D, Hiibinger A, Muhlen H, GriesFA: Effects of previous glycemic controlon the onset and magnitude of cognitivedysfunction during hypoglycaemia intype 1 (insulin dependent) diabetic pa-tients. Diabetologia 35:828-834, 1992

178. Jones TW, McCarthy G, TamborlaneWV, Rosessler E, Sherwin RS: Resistanceto neuroglycopenia: an adaptive re-


Technical Review

sponse during intensive insulin treat-ment of diabetes (Abstract). Diabetes 40:557A,1991

179. Heller SR, Cryer PE: Reduced neuroen-docrine and symptomatic responses tosubsequent hypoglycemia after one epi-sode of hypoglycemia in nondiabetichumans. Diabetes 40:223-226, 1991

180. Davis M, Shamoon H: Counterregula-tory adaptation to recurrent hypoglyce-mia in noraml humans. J Clin EndocrinolMctab 73:995-1001, 1991

181. Widom B, Simonson DC: Intermittenthypoglycemia impairs glucose counter-regulation. Diabetes 41:1597-1602, 1992

182. Davis MR, Mellman M, Shamoon H:Further defects in counterregulatory re-sponses induced by recurrent hypogly-cemia in type I diabetes. Diabetes 41:1335-1340, 1992

183. Lingenfelser T, Renn W, SommerwerckU, Jung MF, Buettner UW, Zaiser-Kas-chel H, Kaschel R, Eggstein M, JakoberB: Compromised hormonal counter-regulation, symptom awareness, andneurophysiological function after recur-rent short-term episodes for insulin-induced hypoglycemia in IDDM pa-tients. Diabetes 42:610-618, 1993

184. Gulan M, Perlman R, Sole M, AlbisserAM, Zinman B: Counterregulatory hor-mone responses preserved after long-termintravenous insulin infusion compared tocontinuous subcutaneous insulin infu-sion. Diabetes 37:526-531, 1988

185. Popp DA, Tse TF, Shah SD, Clutter WE,Cryer PE: Oral propranolol and meto-prolol both impair glucose recoveryfrom insulin induced hypoglycemia ininsulin dependent diabetes mellitus. Di-abetes Care 7:243-247, 1984

186. Hirsch IB, Boyle PJ, Craft S, Cryer PE:Higher glycemic thresholds for symp-toms during j3-adrenergic blockade inIDDM. Diabetes 40:1177-1186, 1991

187. Cryer PE: Iatrogenic hypoglycemia as acause of hypoglycemia-associated auto-nomic failure in IDDM: A vicious cycle.Diabetes 41:255-260, 1992

188. Rattarasarn C, Dagogo-Jack SE, Zach-wieja JJ, Cryer PE: Hypoglycemia-in-duced autonomic failure in IDDM is spe-cific for the stimulus of hypoglycemiaand is not attributable to prior auto-

nomic activation per se. Diabetes 43:809-818, 1994

189. Powell AM, Sherwin RS, Shulman GI:Impaired hormonal responses to hypo-glycemia in spontaneously diabetic andrecurrently hypoglycemic rats. J Clin In-vest 92:2667-2674, 1993

190. Mitrakou A, Fanelli C, Veneman T, Per-riello G, Calderone S, Platanisiotis D,Rambotti A, Raptis S, Brunetti P, CryerP, Gerich J, Bolli G: Reversibility of hy-poglycemia unawareness. N Engl J Med329:834-839, 1993

191. Fanelli CG, Epifano L, Rambotti AM,Pampanelli S, DiVincenzo A, ModarelliF, Lepore M, Annibale B, Ciofetta M,Bottini P, Porcellati F, Scionti L, Santeu-sanio F, Brunetti P, Bolli GB: Meticulousprevention of hypoglycemia normalizesthe glycemic thresholds and magnitudeof most neuroendocrine responses to,symptoms of and cognitive functionduring hypoglycemia in intensivelytreated patients with short-term IDDM.Diabetes 42:1683-1689, 1993

192. Cryer PE: Hypoglycemia begets hypo-glycemia in IDDM. Diabetes 42:1691-1693,1993

193. Dagogo-Jack SE, Rattarasarn C, CryerPE: Dissociation of symptomatic andneuroendocrine responses to hypogly-cemia in IDDM with hypoglycemia un-awareness and awareness and during re-versal of unawareness by avoidance ofiatrogenic hypoglycemia (Abstract). Di-abetes 43:141A, 1994

194. Teuscher A, Berger WG: Hypoglycaemiaunawareness in diabetics transferredfrom beef/porcine insulin to human in-sulin. Lancet ii:382-385, 1987

195. Berger W, Keller U, Honegger B, JaeggiG: Warning symptoms of hypoglycae-mia during treatment with human andporcine insulin in diabetes mellitus.Lancet i: 1041-1044, 1989

196. Egger M, Smith GD, Teuscher AU, Teus-cher A: Influence of human insulin onsymptoms and awareness of hypogly-caemia: a randomized double blindcrossover trial. Br Med J 303:622-626,1991

197. Egger M, Smith GD, Imhoof H, Teus-cher A: Risk of severe hypoglycaemia ininsulin treated diabetic patients trans-

ferred to human insulin: a case controlstudy. BrMedJ 303:617-621, 1991

198. Cryer PE: Hypoglycemia unawareness inIDDM. Diabetes Care 16 (Suppl. 3):40-47, 1993

199. Heine RJ, van der Heyden EAP, van derVeen EA: Responses to human and por-cine insulin in healthy subjects. Lancetii:946-949, 1989

200. Kern W, Lieb K, Kerner W, BornJ, FehmHL: Differential eifects of human andpork insulin-induced hypoglycemia onneuronal functions in humans. Diabetes39:1091-1098, 1990

201. Daneman D, Zinman B: Syndrome ofhypoglycemia unawareness whenchanging insulin species. Diabetes Care14:145-146, 1991

202. Sjobom NC, Lins P-E, Adamson U, The-odorsson E: A comparative study of thehormonal responses to insulin-inducedhypoglycaemia using semisynthetic hu-man insulin and pork insulin in patientswith type I diabetes mellitus. DiabeticMed 7:775-779, 1990

203. Bendtson I, Binder C: Counterregula-tory hormonal response to insulin-in-duced hypoglycaemia in insulin-depen-dent diabetic patients: a comparison ofequimolar amounts of porcine and semi-synthetic human insulin. J Intern Med229:293-296, 1991

204. Jones TW, Caprio S, Diamond MP, Hal-larman L, Boulware SD, Sherwin RS,Tamborlane WV: Does insulin speciesmodify counterregulatory response tohypoglycemia? Diabetes Care 14:728-731, 1991

205. Patrick AW, Bodmer CW, Tieszen KL,White MC, Williams G: Human insulinand awareness of acute hypoglycaemicsymptoms in insulin-dependent diabe-tes. Lancet 338:528-532, 1991

206. Maran A, ChildsJ, Hill C, Macdonald IA,Amiel SA: Human insulin has no effecton glucose counterregulation to hypo-glycaemia compared to pork insulin innondiabetic controls (Abstract). DiabeticMed 7:5A, 1990

207. Mulhauser I, Heinemann L, Fritsche E,von Lennep K, Berger M: Hypoglycemicsymptoms and frequency of severe hy-poglycemia in patients treated with hu-man and animal insulin preparations.


Technical Review

Diabetes Care 14:745-749, 1991208. Hepburn DA, Eadington DW, Patrick

AW, Colledge NR, Frier BM: Symptom-atic awareness of hypoglycaemia: does itchange on transfer from animal to hu-man insulin? Diabetic Med 6:585-690,1989

209. Orchard TJ, Maser RE, Becker D, Dor-man JS, Drash AL: Human insulin useand hypoglycaemia: insights from thePittsburgh epidemiology of diabetesstudy. Diabetic Med 8:469-474, 1991

210. Colagiuri S, Miller JJ, Petocz P: Doubleblind crossover comparison of humanand porcine insulins in patients report-ing lack of hypoglycaemia awareness.Lancet 339:1432-1435, 1992

211. Campbell 1W: Hypoglycaemia and type2 diabetes: sulfonylureas. In Hypoglyce-mia and Diabetes. Frier BM, Fisher BM,Eds. London, Edward Arnold, 1993, p.387-392

212. Clarke BF, Campbell IW: Long-termcomparative trial of glibenclamide andchlorpropamide in diet-failed maturityonset diabetes. Lancet i:246-248, 1974

213. Jennings AM, Wilson RM, Ward JD:Symptomatic hypoglycaemia in N1DDMpatients treated with oral hypoglycaemicagents. Diabetes Care 12:203-208,1989

214. Berger W: Incidence of severe side ef-fects during therapy with sulfonylureasand biguanides. Horm Metab Res 15(Suppl.):lll-115, 1985

215. Campbell IW: Metformin and sulfonyl-ureas: The comparative risk. Horm

Metab Res 15 (Suppl.):105-lll, 1985216. Fox C, Cull CA, Holman RR: Three year

response to randomly allocated therapywith diet, sulfonylurea or insulin in1592 type 2 diabetic patients (Abstract).Diabetic Med 8 (Suppl. 1):8A, 1991

217. Seltzer HS: Severe drug-induced hypo-glycemia: A review. Compr Ther 5:21-29,1979

218. GerichJE: Oral hypoglycemic agents. NEnglJ Med 34:1231-1245, 1989

219. Bailey CJ, Flatt PR, Marks V: Drugs in-ducing hypoglycaemia. Pharmacol Ther42:361-384, 1989

220. MelanderA, Bitzen P-O, Faber O, GroopL: Sulphonylurea antidiabetic drugs.Drugs 37:58-72, 1989

221. Seltzer HS: Drug-induced hypoglyce-mia: A review of 1418 cases. EndocrinolMetab Clin North Am 18:163-183, 1989

222. Campbell IW: Sulphonylureas and met-formin: Efficacy and inadequacy. In NewAutidiabetic Drugs. Bailey CI, Flatt PR,

Eds. Nishimura, Japan, Smith-Gordon,1990,p.33-51

223. Heller SR: Hypoglycaemia and type 2 di-abetes: Insulin therapy. In Hypoglycemiaand Diabetes. Frier BM, Fisher BM, Eds.London, Edward Arnold, 1993, p. 393-400

224. Bolli GB, Tsalikian E, Haymond MW,Cryer PE, Gerich JE: Defective glucosecounterregulation after subcutaneousinsulin in noninsulin dependent diabe-tes mellitus. J Clin Invest 73:1532-1541,1984

225. Ferner RE: Oral hypoglycemic agents.Med Clin North Am 72:1323-1355,1988

226. Gale EAM: Hypoglycaemia. Clin Endo-crinol Metab 9:461-475, 1980

227. Marks V: Drug induced hypoglycaemia.In Hypoglycaemia. 2nd ed. Marks V,Rose FC, Eds. Oxford, UK, BlackwellScientific, 1981, p. 357-386

228. Marri G, Cozzolino G, Palumbo R: Glu-cagon in sulphonylurea hypoglycaemia.Lancet i:303-304, 1968

229. Johnson SF, Schade DS, Peake GT:Chlorpropamide-induced hypoglyce-mia: Successful treatment with diazox-ide. Am] Med 63:799-804, 1977

230. Palatnick W, Meaterall RC, Tenebein M:Clinical spectrum of sulfonylurea over-dose and experience with diazoxidetherapy. Arch Intern Med 151:1859-1862,1991

231. Boyle PJ, Justice K, Krentz AJ, Nagy R,Schade DS: Octreotide reverses hyperin-sulinemia and prevents hypoglycemiainduced by sulfonylurea overdoses. JClin Endocrinol Metab 76:752-756,1993

232. Krentz AJ, Boyle PJ, Justice KM, WrightAD, Schade DS: Successful treatment ofsevere refractory sulfonylurea-inducedhypoglycemia with octreotide. DiabetesCare 16:184-186, 1993

233. Hirsch IB, Farkas-Hirsch R, Cryer PE:Continuous subcutaneous insulin infu-sion for the treatment of diabetic pa-tients with hypoglycemia unawareness.Diabetes Metab Nutr 4:41-43, 1991


View

pub

lication stats

View

pub

lication stats

https://www.researchgate.net/publication/15274023

hypoglycemia - researchgate | find and share research · hypoglycemia philip e. cryer, md joseph n....

Documents