Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Autonomic Nervous System Testing A utonomic nervous system testing is used to document diabetic autonomic failure. Diabetic autonomic failure is defined as impaired function of the peripheral autonomic nervous system and can be divided into two categories: 1) autonomic neuropathy, in which there is a structural lesion of the peripheral autonomic neuron and 2) functional autonomic failure, in which no known structural lesion occurs. In addition to the classic neurotransmitters (acetylcholine and norepinephrine), newer neurotransmitters and neuromodulators such as substance P, neuropeptide K, calcitonin gene-related peptide, and nitric oxide are also involved. Also, the adrenal medullae are an integral part of the autonomic nervous system. With these caveats, the following definition is proposed, wherein autonomic abnormalities are classified as structural or functional with further subdivisions based on the overt or subclinical nature of the disorder, the specific subdivisions of the autonomic nervous system, and the specific organ systems involved. AUTONOMIC NEUROPATHY IN DIABETES — In the setting of diabetes mellitus without other causes of autonomic neuropathy there is a structural lesion that comprises a diffuse disorder of small nerve fibers of the cholinergic, adrenergic, and peptidergic nervous systems. This may be further divided into: 1) subclinical—that which is diagnosed only by tests and 2) clinical—that which presents with symptoms or signs. FUNCTIONAL AUTONOMIC FAILURE IN DIABETES— Reduced autonomic responses not attributable to classic diabetic autonomic neuropathy also occur and cause clinically important problems. An example is hypoglycemiaassociated autonomic failure (HAAF; 1,2). HAAF is a recently described functional disorder without a known structural lesion characterized by selectively reduced adrenomedullary (epinephrine) and parasympathetic (pancreatic polypeptide) responses to a given degree of hypoglycemia. It is associated with the interrelated clinical syndromes of defective glucose counterregulation, hypoglycemia unawareness, and elevated glycemic thresholds for symptoms of and autonomic responses to hypoglycemia during effective intensive therapy, and with a high frequency of iatrogenic hypoglycemia. A further example of functional autonomic failure is the impairment of gastric emptying with severe hyperglycemia (3). TESTS OF AUTONOMIC FUNCTION — The autonomic nervous system is usually tested by evaluating reflex arcs. A reflex arc involves a stimulus, a receptor, an afferent nerve, central processing, an efferent nerve and an endorgan response. In addition to the reflex arc, there are several synapses involved throughout the pathway and different neurotransmitters at each synaptic cleft. It is important that, where possible, the confounding variables, standardization of stimulus, and normal end-organ func- ADDRESS CORRESPONDENCE AND REPRINT REQUESTS TO DR. RICHARD KAHN, AMERICAN DIABETES ASSOCIATION, 1660 DUKE STREET, ALEXANDRIA, VA 22314. THIS MATERIAL IS BEING PUBLISHED SIMULTANEOUSLY IN NEUROLOGY AND MUSCLE AND NERVE. DIABETES CARE, VOLUME 15, SUPPLEMENT 3 , AUGUST 1992 tion be established before one elicits the reflex arc to test the autonomic nervous system. Many organs are dually innervated. Innervation of parasympathetic and sympathetic pathways often work as a check and balance system. Therefore, where possible, the autonomic nervous system needs to be evaluated recognizing that the result may reflect a decrease in one pathway or an increase in another. An ideal test should be simple, noninvasive, easy for the operator and subjects, reproducible, sensitive, relevant to known physiological functions, suitable for longitudinal evaluation, and specific. The confounding variables affecting the test should be fairly well delineated. Some of the current autonomic tests fulfill nearly all of these obligations. Standardization of testing To reduce great variability in assessing the autonomic nervous system, it is important to standardize the test where possible. It is known that eating, drinking coffee, smoking, volume status, upright posture, medicines, and exercise may affect the cardiovascular autonomic nervous system and, presumably, other autonomic nervous organ systems. Therefore, in an ideal situation, studies should be performed with the patient having had no acute illness for the preceding 48 h; unaccustomed vigorous exercise for 24 h, anticholinergic drugs (including antidepressants), antihistamines and over-the-counter cough and cold medications, 9 -a-fluorohydrocortisone, diuretics, sympathomimetic and parasympathomimetic medications, and aspirin for 18 h; alcohol or hypoglycemic episodes for 12 h; or food, caffeine, or tobacco products for 8 h. Moreover, the studies should be performed in the morning in a quiet relaxed atmosphere. The patient should have been taught and practiced the procedure, and at the time of the study should not be wearing compressive clothing or Jobst stockings, should have the blood glucose stabilized 1095 Autonomic nervous system testing Table 1—Some validated tests of specific subdivisions of the autonomic nervous system SYMPATHETIC d. Blood pressure/body temperature. These affect the response in predictable manners based on the heart-rate response. e. Heart rate. In normal individuals, heart rate increases and R-R variation decreases in a predictable manner with aging (7). In diabetic individuals, changes in heart rate are more complicated. With increasing duration of diabetes there is initially an increase followed by a slowing and finally a fixed heart rate (8). R-R variation, in contrast, decreases early and rapidly after the diagnosis of diabetes has been established (9). f. Position. Both standing and sitting significantly reduce R-R variation, therefore tests must be done in the lying position. BLOOD PRESSURE RESPONSE TO STANDING OR TILT. NOREPINEPHRINE RESPONSE TO STANDING. DARK-ADAPTED PUPIL SIZE AFTER PARASYMPATHET1C BLOCKADE QUANTITATIVE SUDOMOTOR AXON REFLEX TEST PARASYMPATHETIC R-R INTERVAL VARIATION WITH DEEP BREATHING PANCREATIC POLYPEPTIDE RESPONSE TO CLAMPED HYPOGLYCEMIA ADRENOMEDULLARY EPINEPHRINE RESPONSE TO CLAMPED HYPOGLYCEMIA (with insulin, if necessary), and should not be emotionally upset. To further define neuropathy, tests have been designed to focus on the specific subdivision affected (Table 1). neuropathy or stress, whereas a large decrease in R-R variation can result from either parasympathetic neuropathy or stress (f$-adrenergic stimulation). SPECIFIC SYSTEM INVOLVEMENT IN AUTONOMIC NEUROPATHY— R-R variation is influenced by many physiological factors: a. Respiratory rate. There is a decrease in R-R variation with increasing respiration rate. The greatest R-R variation occurs at a respiratory rate of 5 breaths/ min. Thus, it is not only important to standardize the respiration rate, but the rate should be 5 breaths/min to optimize the results. b. Age. R-R variation decreases with age and allowance must be made for this. c. Weight. Parasympathetic activity is reduced in obese individuals. The functional changes associated with autonomic failure are given in Table 2 and the tests for autonomic failure in Table 3. Cardiovascular abnormalities Testing for cardiovascular autonomic neuropathy involves a series of measurements. 1. Resting heart rate. 2. Beat-to-beat heart-rate variation. R-R variation is the magnitude of the sinus arrythmia and is measured by one of several methods (4): 1) standard deviation, 2) mean circular resultant, 3) maximal minus the minimal heart rate, 4) expiration-inspiration ratio, 5) Holter monitoring, and 6) power spectral analysis. R-R variation was considered to be exclusively under the control of the parasympathetic nervous system. However, subsequent studies demonstrate that both P-adrenergic stimulation (isoproterenol) (5) and 0-adrenergic blockade (propranolol) (6) decrease R-R variation. Both parasympathetic cholinergic blockade (atropine), and (3-adrenergic stimulation can nearly totally abolish R-R variation (5). Thus, a small decrease in R-R variation results from sympathetic 1096 Confounding variables in R-R autonomic testing (4): a. General: this includes eating, drinking coffee, and smoking. b. Sodium salicylate raises the R-R variation. c. Dehydration can alter the response. d. Coronary artery disease: it has been shown that patients with Table 2—Functional changes associated with autonomic failure SYSTEMS INVOLVED CARDIOVASCULAR MANIFESTATIONS RESTING TACHYCARDIA, IMPAIRED EXERCISE-INDUCED CARDIOVASCULAR RESPONSES, CARDIAC DENERVATION, ORTHOSTATIC HYPOTENSION, HEAT INTOLERANCE, IMPAIRED VASODILATAT1ON, IMPAIRED VENOARTER1OLAR REFLEX (DEPENDENT EDEMA) EYE DECREASED DIAMETER OF DARK-ADAPTED PUPIL (DARK-ADAPTED MIOSIS) GASTROINTESTINAL ESOPHAGEAL ENTEROPATHY, GALLBLADDER ATONY, IMPAIRED COLONIC MOTIL1TY (DIARRHEA, CONSTIPATION), ANORECTAL SPHINCTER DYSFUNCTION (INCONTINENCE) GENITOURINARY NEUROGENIC VESICAL DYSFUNCTION (DECREASED BLADDER SENSITIVITY/ INCONTINENCE/RETENTION), SEXUAL DYSFUNCTION, (MALE: PENILE ERECTILE FAILURE AND RETROGRADE EJACULATION; FEMALE: DEFECTIVE LUBRICATION) SUDOMOTOR ANHIDROSIS/HYPERHIDROSIS (HEAT INTOLERANCE), GUSTATORY SWEATING ENDOCRINE HYPOGLYCEMIA-ASSOCIATED AUTONOMIC FAILURE DIABETES CARE, VOLUME 15, SUPPLEMENT 3, AUGUST 1992 Autonomic nervous system testing Table 3—Tests for autonomic failure and their suitability for cross-sectional and longitudinal studies in individual and groups POPULATION KEY TEST QUANTITATIVE LONGITUDINAL REFERENCES STANDARDIZATION STUDIES STUDIES INDIVIDUAL GROUP 4 10 YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES R - R VARIATION VALSALVA RATIO POSTURAL BLOOD PRESSURE TEST 13 14 Q - T INTERVAL DARK-ADAPTED PUPIL SIZE AFTER PARASYMPATHETIC BLOCKADE 13 No YES YES YES YES 19,21 YES No No YES No 20 16 17 22 25,26 23 24 1,2 YES YES No YES No YES YES YES YES YES No No YES YES YES YES SEMI YES YES YES YES YES YES YES YES YES SEMI YES YES YES YES YES YES YES YES SOLID PHASE GASTRIC MOTILITY CYSTOMETROGRAM WITH BETHANECHOL SUPERSENSITIVITY TEST LATENCY OF SPINAL REFLEX-EVOKED POTENTIALS NOCTURNAL PENILE TUMESCENCE MONITORING INTRACAVERNOSAL INJECTION OF VASODILATORS THERMOREGULATED SWEAT TEST QUANTITATIVE SUDOMOTOR AXON REFLEX TEST SKIN POTENTIALS SWEAT IMPRINT CLAMPED HYPOGLYCEMIA EPINEPHRINE YES YES YES YES YES PANCREATIC POLYPEPTIDE YES YES YES YES YES SYMPTOMS SCORE YES YES YES No YES YES No YES YES YES 33 INSULIN INFUSION TEST inferior wall myocardial infarctions experience a bradycardiahypotensive syndrome, whereas patients with anterior wall myocardial infarctions experience tachycardia-hypertensive syndromes. Both can affect the R-R variation. Valsalva maneuver. The subject blows into the mouthpiece of a manometer to 40 mmHg for 15 s with continuous EKG monitoring before, during, and after the procedure. Healthy subjects develop tachycardia and peripheral vasoconstriction during strain, and an overshoot in blood pressure and bradycardia on release. The Valsalva ratio is the longest R-R/ shortest R-R. The Valsalva maneuver has also been well evaluated and studied (10). It encompasses a complex reflex arc involving both sympathetic and parasympa- DIABETES CARE, VOLUME 15, SUPPLEMENT 3, AUGUST thetic pathways to the heart, sympathetic pathways to the vascular tree, and baroreceptors in the chest and lungs. 4. Blood pressure response to standing. The blood pressure response is measured with the patient lying supine at rest and again after 1 and 5 min standing. The absolute fall is arbitrary but fall in systolic blood pressure greater than 20 mmHg accompanied by symptoms of orthostasis is taken to be evidence of sympathetic failure. Photoplethysmographic beat-tobeat blood pressure recording may have significant advantages and can be used to evaluate the components of the Valsalva maneuver (11). Its role in clinical studies awaits more extensive experience. 5. QTc interval. Examination of the EKG may reveal a prolonged QTc 1992 interval in patients with cardiac autonomic neuropathy (12,13). Eye Dark-adapted pupil size after total parasympathetic blockade may be a useful reproducible and valuable tool for evaluating autonomic nervous system function, and is related to poor dark vision (14). Motor disturbances of gastrointestinal tract 1. Esophageal and gallbladder enteropathy. These are usually discovered accidentally while reviewing upper gastrointestinal studies. 2. Gastroparesis. In evaluating a patient with suspected diabetic gastroparesis, the level of glycemic control should be assessed. Careful history of medications including 1097 Autonomic nervous system testing Table 3—Continued DEGREE OF TEST R - R VARIATION SENSITIVITY SPECI- REPRODUCIBIUTY (1-4) FICITY (1-4) 4 YES 3 How TO REPORT E - I RATIO, MEAN CIRCU- Pvs. CERTAINTY NP (1-4) COMMENTS 4 NP 4 CONFOUNDING VARIABLES ARE EASE 1-4 LAR RESULTANT WELL ESTABLISHED VALSALVA RATIO 3 YES 3 RATIO 4 P 4 CONFOUNDING VARIABLES ARE POSTURAL BLOOD PRESSURE 2 YES 2 CHANGE IN BLOOD PRES- 4 NP 4 INTRAVASCULAR VOLUME DE- Q - T INTERVAL 2 YES QTc 4 YES MM 4 2-3 P P 4 4 ISCHEMIA DEPENDENT DARK-ADAPTED PUPIL SIZE 4 4 AFFECTED BY RUBEIOSIS 1-2 No 1 MlN 3 NP 2 SOMEWHAT INVASIVE, SPECIAL UNKNOWN YES 3 YES/NO 1 NP 3 INVASIVE 3 YES UNKNOWN MS 1 UNKN 2 SIGNIFICANCE OF RESULTS IS 3 No 3 RIGIDITY (RELATIVE 2 EQUIP. 3 N O T AN ANS WELL ESTABLISHED TEST PENDENT SURE AFTER PARASYMPATHET1C BLOCKADE SOLID PHASE GASTRIC M O TIUTY CYSTOMETROGRAM WITH FACILITIES, RADIATION BETHANECHOL SUPERSENS1TIVITYTEST LATENCY OF SPINAL REFLEX- EVOKED POTENTIALS NOCTURNAL PENILE TUMES- UNCLEAR, INVASIVE DEPEN. UNITS), NUMBER OF CENCE MONITORING TEST, USED IN COMBINATION WITH OTHER ERECTILE EPISODES, TEST FOR DIAGNOSIS OF MINUTES OF DURATION EXCLUSION, RATINGS DONE ON SLEEP LAB EVALUATION INTRACAVERNOSAL INJECTION N/A No 2 NP YES/NO (ALTHOUGH COULD BE MADE OF VASODILATORS 3 WITH REPEATED QUANTITATIVE) TESTING THERMOREGULATED SWEAT 3 YES UNKNOWN 3 YES 3 2 2-3 YES 1-2 YES 3 TEST, USED IN COMBINATION WITH OTHER TEST FOR DIAGNOSIS OF EXCLUSION SURFACE AREA 2 NP 4 VERY CUMBERSOME AND ML/CM 2 3 NP 4 NEEDS SPECIAL EQUIPMENT YES/NO OR MV 3 2 NP NP 3 4 HABITUATES MESSY TEST QUANTITATIVE SUDOMOTOR N O T AN ANS AXON REFLEX TEST SKIN POTENTIALS SWEAT IMPRINT DENSITY OR DIAMETER NEEDS SPECIAL EQUIPMENT DISTRIBUTION CLAMPED HYPOGLYCEMIA RELATES TO INSULIN-TREATED DIABETES ONLY EPINEPHRINE UNKNOWN UNKNOWN 4 GLYCEMIC THRESHOLD 1 NP 3 1 NP 3 1 NP 3 2 NP 3 MM PANCREATIC POLYPEPTIDE UNKNOWN UNKNOWN 4 GLYCEMIC THRESHOLD MM SYMPTOMS SCORE UNKNOWN UNKNOWN 2 GLYCEMIC THRESHOLD MM INSULIN INFUSION TEST 3 3 4 YES/NO ASSESSES ABILITY TO DEFEND AGAINST HYPOGLYCEMIA 1, Low; 4, High; P, Parametric; NP, Nonparametric. 1098 DIABETES CARE, VOLUME 15, SUPPLEMENT 3, AUGUST 1992 Autonomic nervous system testing Abnormal NPT Normal Hormonal Profile Intracavemosal injection of vasodilators (X 3 with high doses) No erection Erection I I Vasculature probably competent but cannot rule out endothelial injury Suspicion of neuropathy Vascular problems ± neuropathy Figure 1—Diagnosis of diabetic impotence. ganglionic blocking agents and psychotropic drugs should be obtained. In addition, gastroduodenoscopy should be performed to exclude pyloric or other mechanical obstruction. After optimization of glycemic control, isotope scintigraphy to measure solid-phase gastric emptying times may be indicated (3). 3. Constipation. The extent of the evaluation in a diabetic patient complaining of constipation depends on the severity of the constipation and associated signs. All patients should have a digital examination to evaluate rectal sphincter tone. Other causes of constipation should be excluded. 4. Diarrhea. The diagnosis of diabetic diarrhea is established by excluding other causes of diarrhea and by confirming the presence of autonomic neuropathy (15). 5. Fecal incontinence. Anorectal sphincter function is evaluated by anorectal manometry, which quantitates maximal basal sphincter pressure and the rectoanal inhibitory reflex (inflation of a balloon in the rectum causes a reflex relaxation of the internal anal sphincter). Continence for solids and liq- DIABETES CARE, VOLUME 15, SUPPLEMENT 3, AUGUST uids is directly assessed by simulating the presence of stools with a solid sphere or rectally infused saline. Genitourinary tract disturbances 1. Autonomic neuropathy of the penis in diabetes. It is not presently possible to measure the status of the autonomic nerves of the penis. Because of this, diabetic penile neuropathy is a diagnosis that is reached after the exclusion of other causes of erectile dysfunction. The presence of impotence can be determined by history and nocturnal penile tumescence monitoring of sleep-related erections (16). Patients can then be tested for their response to intracavemosal vasodilators (papaverine, phentolamine, prostaglandin E : ) (17). If the patient has an abnormal nocturnal penile tumescence test and a normal hormonal profile, but responds with a full erection to the vasodilators, significant vascular disease is usually ruled out, suggesting that neuropathy is the predominant factor. However, because the vasodilators directly relax the smooth muscle, it is not possible to exclude vascular endothelial dys- 1992 function as a contributing factor leading to impotence. If the patient does not respond to several (3-5) intracavemosal injections, vascular dysfunction (arterial or venoocclusive disease), with or without associated neuropathy, is likely. The suggested sequence for the diagnosis of autonomic neuropathy of the penis is given in Fig. 1. The suspicion of diabetic penile neuropathy cannot be confirmed with any specific measurement of penile autonomic nerve function. Reflex-evoked potential studies measure function in sensory (bladder visceral) autonomic and sensory and motor somatic nerves but not in motor autonomic penile nerves. Therefore, abnormal latencies in these reflex-evoked potential studies support but do not confirm a diagnosis of autonomic neuropathy of the penile nerves (18). 2. Female sexual dysfunction. Sexual dysfunction in the female secondary to autonomic neuropathy has not been well defined. Anorgasmia may or may not be a feature. However, difficulty with vaginal lubrication may occur in diabetic women with autonomic neuropathy. Female sexual dysfunction using vaginal plethysmography to measure lubrication and vaginal flushing has not been well established. It may be valuable in the future, but much in the way of standardization needs to be accomplished before it can be recommended as a routine test. 3. Autonomic neuropathy of the bladder in diabetes. To arrive at the diagnosis of diabetic autonomic neuropathy of the bladder, the sensory and motor function of the bladder and need to be tested. Testing requires catheterization of the bladder, sophisticated urodynamic equipment, and an experienced professional with training 1099 Autonomic nervous system testing in the field of neurourology. A simple, noninvasive, and specific test to screen for the presence of bladder autonomic neuropathy is not available. A. Testing of bladder sensitivity, i. The integrity of bladder proprioception can be tested by progressive filling of the bladder with water or gas (CO2) to record when the patient first becomes aware of a desire to void (urge) and when bladder filling becomes painful (bladder capacity) (19). ii. Perception of visceral pain can be evaluated by determining the electrical perception threshold (19). A catheter with s t i m u l a t i n g electrodes is placed in the vesicourethral junction; the intensity of the stimulus is progressively increased until the patient first senses it. Elevated electrical perception thresholds are considered diagnostic of sensory loss (19). iii. Temperature sensation can also be evaluated by the introduction of warm or cold fluid through the catheter (19). iv. Another way of determining the status of bladder sensory nerves is by measuring the latency of spinal reflexevoked potentials (20). Electrical s t i m u l a t i o n through an indwelling catheter in the urethra or vesicourethral junction elicits a contractile response of the anal sphincter. This response can be recorded with a bipolar needle placed in the sphincter. The latency between the stimulus and the response can be measured to assess the integrity 1100 of this reflex pathway, in which the afferent limb comprises sensory autonomic fibers traveling in the pelvic nerves and the efferent limb, motor somatic nerve fibers in the pudendal nerve. It is possible that the abnormality in the reflex pathway is due to an alteration in the efferent rather than in the afferent limb. To differentiate an autonomic sensory from a somatic motor alteration, the bulbocavernosus reflex can be tested. In this reflex both limbs are formed by somatic fibers. If the latency of this reflex is normal, it can be concluded that the increase in the reflex latency after visceral stimulation is likely due to neuropathy of the sensory autonomic fibers. In addition, electromyography of the anal sphincter after voluntary contraction will assist in determining the presence or absence of neuropathy in this somatic motor pathway. The advantage of this method over those described in i, ii, and iii is that is does not rely on the patient reporting a sensation but rather on an electrophysiological measurement. The disadvantage of measuring the latency of spinal reflex-evoked potentials is that there is limited experience with this method and it requires sophisticated electrophysiological equipment. Although this method is quantitative and can be standardized, its reproducibility and specificity are not known. The invasive and complicated nature of this test makes it a poor candi- date for longitudinal studies. B. Testing of bladder motor function. The ability of the bladder to empty urine not only depends on the ability of the detrusor to contract but also on a nonobstructed bladder outlet. Therefore, when evaluating the motor function of the bladder, it is necessary to rule out bladder outlet obstruction. This is particularly important in older men in whom bladder outlet obstruction is prevalent, i. Detrusor motor function can be studied by performing a cystometrogram (19,20). This test studies the detrusor reflex function. The normal bladder, when filled with fluid or gas, responds with a reflex contraction of the detrusor. This contraction can be voluntarily suppressed. A bladder that does not contract with filling is considered an areflexic bladder. This can be due to an inability to suppress the psychogenic central influence controlling this reflex, to neuropathy of sensor autonomic nerves, and/or to neuropathy of motor autonomic fibers responsible for contraction of the detrusor muscle. Therefore, the finding of bladder areflexia alone is not sufficient for the diagnosis of autonomic motor neuropathy of the bladder. The bethanechol supersensitivity test is used to determine the presence of efferent denervation of the bladder (21). This test is based on Canon's law of denervation supersensitivity DIABETES CARE, VOLUME 15, SUPPLEMENT 3, AUGUST 1992 Autonomic nervous system testing of an organ to its neurotransmitter. Bethanechol (5 mg), a muscarinic agonist, is given subcutaneously and the cystometrogram is repeated. After the bladder has been filled with 100 ml of water or gas, if the intravesical pressure is >20 cm H2O above the intravesical pressure before the administration of bethanechol, the test is considered positive. If motor autonomic neuropathy is present, the test will be positive. Patients with psychogenic suppression of the detrusor reflex or sensory autonomic neuropathy in the absence of motor neuropathy would have a negative test. The sensitivity of the test is not known. It is considered to be specific for autonomic motor neuropathy, but due to its invasive nature, it is probably not useful in longitudinal studies. ii. Finally, measurement of volume of residual urine after urination may be used, in the absence of bladder outflow obstruction, as an indicator of bladder motor function. After the patient has urinated, the volume of residual urine can be determined non-invasively with an ultrasound bladder scanner or, invasively, by catheterization. Sudomotor sympathetic function Sudomotor function may be evaluated with the thermoregulatory sweat test, quantitative sudomotor axon reflex test, skin potentials, or sweat imprint quantitation. The thermoregulatory sweat test is a sensitive test of sweat distribution (22). The subject is dusted with an indi- DIABKTKS CARE, VOLUME 15, SUPPLEMENT 3 , AUGUST cator powder that turns purple when moist. This qualitative test can be rendered semiquantitative by charting the percentage of anterior body surface that is anhidrotic. The skin potential can be recorded with standard EMG equipment from the palm and sole (23). The stimulus is an electric shock, an inspiratory gasp, or other stimuli that activate type II and III mechanoreceptor afferents. The skin potential is readily evoked but habituates. The silastic skin imprint is obtained after the application of silastic material to stimulated skin (24). The sweat droplet indents the imprint and the count and diameter distribution can be determined. The usual stimulus is pilocarpine administered by iontophoresis. In the quantitative sudomotor axon reflex test the stimulus consists of the iontophoresis of acetylcholine via the stimulus compartment of a multicompartmental sweat cell (25,26). Postganglionic sympathetic nerve terminals are activated and the nerve impulse travels retrogradely, reaches a branch point, then travels orthogradely to activate a second population of sweat glands. The sweat response from this second population of sweat glands is recorded by a sudorometer. The stimulus compartment surrounds the central recording compartment, separated by an air gap and two ridges (to block diffusion). This test evaluates the integrity of the distal postganglionic sympathetic sudomotor axon. Four recording sites (distal forearm and 3 lower extremity sites) are used. The test has high sensitivity, a coefficient of variation of 20%, and, when used in conjunction with the thermoregulatory sweat test, defines the pre- or postganglionic site of the lesion. The test requires specialized equipment, trained technicians, and 20-30 min to complete. Endocrine tests for functional autonomic failure in diabetes HAAF appears to be distinct from classic diabetic autonomic neuropathy (27). 1992 First, the two disorders can occur independently. Second, deficient autonomic responses are specific for the stimulus of hypoglycemia in HAAF, whereas reduced sympathetic and parasympathetic responses to multiple stimuli characterize autonomic neuropathy. Third, whereas reduced epinephrine responses to hypoglycemia are a characteristic feature of HAAF, epinephrine responses are reduced little if at all in insulindependent diabetes mellitus (IDDM) with autonomic neuropathy compared with IDDM without autonomic neuropathy. Fourth, diabetic autonomic neuropathy, in sharp contrast to HAAF, is not a risk factor for iatrogenic hypoglycemia in IDDM (28,29). The pathogenesis of HAAF is not known; it probably is multifactorial and may be related to recent antecedent hypoglycemia (30-32). Tests for functional autonomic failure in IDDM are complex for the patient as well as the investigator and labor intensive. Ideally, both glycemic thresholds for symptoms and counterregulatory activation and glycemic defense against mild to moderate hyperinsulinemia should be determined in patients with IDDM and appropriate control subjects. Defense against hyperinsulincmic hypoglycemia is probably the more essential, at least for clinical purposes, because it has been shown in prospective studies (33,34) to have clinical predictive power, i.e., to identify patients at markedly increased risk (> 25-fold) for severe iatrogenic hypoglycemia. Glycemic thresholds can be defined with the hyperinsulinemic stepped hypoglycemic clamp technique (1,2,35). The stepped hypoglycemic clamp technique can be highly standardized (1,2). However, the test is complex. Precise quantitation of the glycemic thresholds requires both a euglycemic control clamp and a stepped hypoglycemic clamp in each individual studied. Arterialized blood samples are also required. The reproducibility of the stepped hypoglycemia clamp technique has not been assessed systematically. However, the 1101 Autonomic nervous system testing fact that identical glycemic thresholds for symptoms and glucagon and epinephrine release were calculated from data generated with the technique in two different laboratories (1,2) indicates that the test is probably reproducible. The technique is probably more suitable for populations rather than individuals, i.e., it might not identify all patients with minimally altered glycemic thresholds. The results of the stepped hypoglycemia clamp test should be reported as absolute glycemic thresholds (glucose concentration), with those of the insulin infusion test as negative (adequate glucose counterregulation) or positive (defective glucose counterregulation). For the stepped hypoglycemic clamp technique, comparison of each response (symptom scores, hormone concentrations) of each individual patient with the 95% confidence interval of the corresponding response for a nondiabetic control group is preferable (1,2), although comparisons of group means can be used. For the latter, either parametric or nonparametric tests will be appropriate to a given data set. Glycemic defense against hyperinsulinemia can be defined with an insulin infusion test (33). The insulin infusion test can be standardized with a relatively low insulin infusion dose (0.67 mU • kg" 1 • min • - 1 ) and a specific plasma glucose end point (<2 mM = defective glucose counterregulation). In addition, a neuroglycopenic end point must be included for ethical reasons. Inclusion of neuroglycopenia (difficulty thinking, blurred vision, dizziness, fatigue and faintness) as an additional end point increases the predictive power of the test (33). Although the test yields only a positive or negative conclusion—defective or adequate glucose counterregulation—it has >90% clinical predictive power for development of hypoglycemia with intensive insulin treatment. The test is labor intensive, requiring a minimum of two people (typically a physician and nurse); a third person is needed if cognitive function is assessed. It 1102 also requires accurate glucose measurements (e.g., with a Beckman or YSI analyzer not with a portable glucose monitor) at the bedside and the analytical capacity to perform the hormone measurements. Cognitive assessment generally requires a collaborating psychologist (2,36). The insulin infusion test has been shown to be highly reproducible (33,34). With three replicate tests at 3- to 4-wk intervals, coefficients of variation for glucose and key counterregulatory hormone (glucagon, epinephrine) responses were <8% (34). One can estimate sensitivity and specificity to be ~90% from the data of White et al. (33). But, these are minimum estimates. In the data of Bolli et al. (34), none of the patients with a negative test (nadir >2.2 mM) subsequently suffered severe clinical hypoglycemia, whereas all of those with a positive test (nadir <2.2 mM) suffered severe hypoglycemia during subsequent intensive therapy. SUMMARY AND RECOMMENDATIONS— Several organ systems can be monitored by reliable longitudinal testing. These include: 1. Cardiovascular system a. R-R variation b. Valsalva maneuver c. Postural blood pressure testing 2. Sudomotor a. Postganglionic function with QSART 3. Eye a. Dark-adapted pupil size after total parasympathetic blockade. Specific research projects may include other tests but these tests do not lend themselves easily for large, multicenter type of studies. The test listed above are noninvasive, quantitative, and are sufficiently standardized to allow longitudinal assessment in diabetic patients. Furthermore, these tests allow evaluation of at least three different organ systems. HAAF, an example of functional autonomic failure in diabetes, is best assessed by evaluation of the glycemic thresholds for epinephrine and pancreatic polypeptide secretion and hypoglycemic symptoms using a stepped hypoglycemic clamp protocol. The status of physiological defense against hyperinsulinemic hypoglycemia can be assessed with the insulin infusion test. References 1. Schwartz NS, Clutter WE, Shan SD, Cryer PE: The glycemic thresholds for activation of glucose counterregulatory systems are higher than the threshold for symptoms. J Clin Invest 79:777-81, 1987 2. Mitrakou A, Ryan C, Veneman T, Mokan M, Jennssen T, Kiss 1, DurrantJ, Cryer P, Gerich J: Hierarchy of glycemic thresholds for counterregulatory hormone secretion, symptoms, and cerebral dysfunction. Am] Physiol 260:E67-74, 1991 3. Achem-Karam SR, Funakoshi A, Vinik Al: Plasma motilin and migrating motor complex in diabetic gastroparesis. Gastroenterology 88:492-94, 1985 4. Genovely H, Pfeifer MA: RR-Variation: the autonomic test of choice in diabetes. Diabetes Metab Rev 4:255-71, 1988 5. Pfeifer MA, Cook D, BrodskyJ, Tice D, Reenan A, Swedine S, Halter JB, Porte Jr D: Quantitative evaluation of cardiac parasympathetic activity in normal and diabetic man. Diabetes 31:339-45, 1982 6. Weinberg CR, Pfeifer MA: Development of a predictive model for symptomatic neuropathy in diabetes. Diabetes 35: 873-80, 1986 7. Pfeifer MA, Halter JB, Weinber CR, Cook D, Best J, Reenan A, Porte D Jr: Differential changes of autonomic nervous system function with age in man. Am J Med 75:249, 1983 8. Ewing DJ, Campbell IW, Clarke BF: Heart rate changes in diabetes mellitus. Lancet 1:183, 1981 9. Pfeifer MA, Weinberg CR, Cook D, Reenan A, Halter JB, EnsinckJ, Porte Jr D: Autonomic neural dysfunction in recently diagnosed diabetic subjects. Diabetes Care 7:447-53, 1984 10. Rothschild AH, Weinberg C, Cook D, Halter J, Porte D, Pfeifer MA: Sensitivity of R-R variation and the Valsalva ratio in the assessment of cardiovascular diabetic DIABETES CARE, VOLUME 15, SUPPLEMENT 3, AUGUST 1992 Autonomic nervous system testing 11. 12. 13. 14. 15. autonomic neuropathy. Diabetes Care 40: 735-41, 1987 Sandroin P, Benarroch EE, Low PA: Pharmacologic dissection of the components of the valsalva maneuver in adrenergic failure. J Appl Physiol 71:1563-67, 1991 KahnJK, Sisson JC, Vinik Al: QT Interval prolongation and sudden cardiac death in diabetic autonomic neuropathy. J Clin Endocrinol Metab 64:751-54, 1987 Gonin JM, Kadrofske MM, Schmaltz S, Vinik Al: Corrected Q-T interval prolongation as diagnostic tool for assessment of cardiac autonomic neuropathy in diabetes mellitus. Diabetes Care 13:68-71, 1990 Pfeifer MA, Code DL, BrodskyJ: Quantitative evaluation of sympathetic and parasympathetic control of iris function. Diabetes Care 5:518-28, 1982 Barnett JL, Vinik Al: Gastointestinal disturbances in diabetes. In Therapy for Diabetes Mellitus and Related Disorders. Leb- ovitz AE, Ed. Alexandria, VA, Am. Diabetes Assoc, 1991, p. 279-87 16. Karacan 1, Salis PJ, Ware JC, Derveut B, Williams RL, Scott FB, Attia SL, Beutler LE: Nocturnal penile tumescence and diagnosis in diabetic impotence. AmJ Psychiatry 135:191-97, 1978 17. Krane RJ, Goldstein I, Saenz de Tejada I: Impotence. N EnglJ Ued 321:1648-59, 1989 18. Sarica Y, Karacan I: Bulbocavernosus reflex to somatic and visceral nerve stimulation in normal subjects and in diabetics with erectile impotence. J Urol 138:5558, 1987 DIABETES CARE, VOLUME 15, SUPPLEMENT 3 , AUGUST 19. Fridmodt-Moller C: Diabetic cystopathy. Incidence of severe hypoglycemia in an Danish Med Bull 25:49-59, 1978 unselected population of patients with insulin-treated diabetes mellitus with special 20. Bradley WE: Diagnosis of urinary bladreference to autonomic neuropathy. Diabeder dysfunction in diabetes mellitus. Ann tes Nutr Metab 4:303-309, 1990 Intern Med 92:323-26, 1980 21. Lapides J: Denervation supersensitivity 30. Heller SR, Cryer PE: Reduced neuroendocrine and symptomatic responses to as a test for neurogenic bladder. Surg subsequent hypoglycemia after one epiGynecol Obstet 114:241-49, 1962 sode of hypoglycemia in nondiabetic hu22. Fealey RD, Low PA, Thomas JE: Themomans. Diabetes 40:223-26, 1991 regulatory sweating abnormalities in diabetes mellitus. Mayo Clin Proc 64:617- 31. Widom B, Simonson DC: Effect of inter28, 1989 mittent hypoglycaemia on counterregu23. Shahani BT, HalperinJJ, Boulu P, Cohen latory hormone secretion (Abstract). DiJ: Sympathetic skin response: a method abetologia 33:A484, 1990 of assessing unmyelinated axon dysfunc- 32. Davis M, Shamoon H: Adaptive countertion in peripheral neuropathies. J Neural regulatory responses to hypogycemia in Neurosurg Psychiat 47:536-42, 1984 nondiabetic humans (Abstract). Diabetes 24. Kennedy WR, Sakuta M, Sutherland D, 38:5A, 1989 Goetz FC: Quantitation of the sweating 33. White NH, Skor DA, Cryer PE, Bier DM, deficit in diabetes mellitus. Ann Neurol Levandoski L, Santiago JV: Identification 15:482-88, 1984 of type 1 diabetic patients at increased 25. Low PA, Zimmerman BR, Dyck PJ: Comrisk for hypoglycemia during intensive parison of distal sympathetic with vagal therapy. N EnglJ Med 308:485-91, 1983 function in diabetic neuropathy. Muscle 34. Bolli GB, Defeo P, DeCosmo S, Perriello G, Nerve 9:592-96, 1986 Ventura MM, Massi-Benedetti M, Santeu26. Low PA, Caskey PE, Tuc RR, Fealey RD, sanio F, Gerich JE, Brunetti P: A reliable Dyck PJ: Quantitive sudomotor axon reand reproducible test for adequate glucose flex test in normal and neuropathic subcounterregulation in type I diabetes mellijects. Ann Neurol 14:573-80, 1983 tus. Diabetes 33:732-37, 1984 27. Cryer PE: Iatrogenic hypoglycemia as a 35. Boyle PJ, Schwartz NS, Shah SD, Clutter cause of hypoglycemia-associated autoWE, Cryer PE: Plasma glucose concennomic failure in IDDM: a vicious cycle. trations at the onset of hypoglycemic Diabetes 41:255-60, 1992 symptoms in patients with poorly con28. Diabetes Control and Complications trolled diabetes and in nondiabetics. N Trial Research Group: Epidemiology of EnglJ Med 318:1487-92, 1988 severe hypoglycemia in the diabetes con- 36. Hirsch IB, Boyle PJ, Draft S, Cryer PE: trol and complications trial. Am J Med Higher glycemic thresholds for symp90:450-59, 1991 toms during beta-adrenergic blockade in 29. Bjork E, Palmer M, Schvarcz E, Berne C: IDDM. Diabetes 40:1177-86, 1991 1992 1103