INTRODUCTION — Diabetic autonomic neuropathy (DAN) is a common and debilitating form of neuropathy. DAN may be detected in the majority of patients with diabetes with neurophysiologic testing but is classified as subclinical or clinical depending upon the presence or absence of symptoms [1]. A wide spectrum of manifestations affecting many different organ systems can occur, including the cardiovascular, gastrointestinal, genitourinary, pupillary, sudomotor, and neuroendocrine systems (table 1).
GLYCEMIC CONTROL AND RISK FACTOR REDUCTION — The therapy of DAN can be difficult. It is therefore desirable to prevent this complication or, once established, to slow disease progression.
Poor glucose control and vascular risk factors appear to be associated with the development of diabetic neuropathy [2]. This observation is supported by the results of the EURODIAB study, which found that the incidence of neuropathy was associated with poor glucose control, elevated triglyceride levels, elevated body mass index, smoking, and hypertension [3]. However, the effects of these risk factors on DAN progression are less clear. (See "Screening for diabetic polyneuropathy", section on 'Risk factors'.)
Results from a large prospective observational study suggest that the incidence of DAN is declining in type 1 diabetes, potentially reflecting improvements in the management of risk factors [4]. Furthermore, results from randomized trials suggest that intensive therapy reduces the onset and progression of autonomic neuropathy:
●In the Diabetes Control and Complications Trial (DCCT), intensive therapy with insulin in subjects with type 1 diabetes was found to reduce cardiovascular autonomic neuropathy incidence by 53 percent [5], and the benefit of prior intensive therapy was found to persist for up to 14 years in these subjects [6].
●In individuals with type 2 diabetes, the potential efficacy of intensive combined therapy in patients with type 2 diabetes and microalbuminuria was examined in the Steno type 2 trial [7]. In this prospective open-label trial, 160 patients were randomly assigned to standard or multifactorial intensive therapy. The intensive regimen consisted of behavioral therapy (including advice concerning diet, exercise, and smoking cessation) and pharmacologic intervention (consisting of the administration of multiple agents to attain several aggressive therapeutic goals) (table 2). DAN was present at baseline in 28 percent. At the end of the trial (mean follow-up of 7.8 years), intensive therapy reduced both microvascular and macrovascular disease [7]. In addition, intensive therapy reduced the rate of progression to DAN (30 versus 54 percent, relative risk [RR] 0.37, 95% CI 0.18-0.79). This benefit was sustained at a mean of 13.3 years (the trial intervention period plus an additional 5.5 years of observational follow-up) with a RR of 0.53 (95% CI 0.34-0.81) [8]. By contrast, there was no slowing of progression of peripheral neuropathy. The details of the protocol and overall results of this study are discussed elsewhere. (See "Overview of general medical care in nonpregnant adults with diabetes mellitus", section on 'Multifactorial risk factor reduction'.)
In some cases, extremely rapid improvements in glycemic control in patients with chronic hyperglycemia may be associated with the development of microvascular complications of diabetes, including autonomic and peripheral neuropathy [9]. Described as treatment-induced neuropathy of diabetes (TIND), the severity of neuropathy was linked to the magnitude of change in the glycated hemoglobin A1C over three months. The risk of neuropathic complications was approximately 10 percent if the hemoglobin A1C decreased by more than 3 percentage points in three months and exceeded 50 percent if the hemoglobin A1C decreased by 5 or more points in three months. These findings suggest that, in individuals with chronic hyperglycemia, the achievement of a target hemoglobin A1C (as suggested by the DCCT and Steno type 2 trials) should be achieved gradually and not exceed a 3-point change in hemoglobin A1C in three months. TIND is more common in those with type 1 diabetes, most likely because of the rapid change in glycemic control that can occur with the use of insulin. Based on the limited data available, chronic hyperglycemia of a year or more may be necessary for an abrupt decrease in average glucose levels to result in neuropathy development [10].
Not all experts are convinced that overly rapid glycemic control or reduction in hemoglobin A1C can cause neuropathy. The data supporting it come from uncontrolled and retrospective studies. The prevalence of this problem is unknown, as the most recent data on the subject are from a specialty tertiary referral center that may not reflect the general population. Although no evidence of the TIND phenomenon was observed in data from prospective controlled trials such as the DCCT, there was no measurement of small fiber neuropathy in such trials. Interestingly, "early worsening retinopathy" was noted within the DCCT trials, a phenomenon also seen in individuals with TIND [11-14]. The parallel development of retinopathy and neuropathy in acute glycemic control does suggest an inflammatory microvascular process. An alternative explanation is that the worsening of neuropathy is an epiphenomenon associated with but not caused by overly rapid glycemic control. To date, the data have only suggested an association, not causation. Thus, prospective, controlled studies would be helpful to resolve the relationship between glycemic control and the development of neuropathy.
CARDIOVASCULAR AUTONOMIC NEUROPATHY — Cardiovascular autonomic neuropathy (frequently abbreviated as CAN in the literature) is defined as the impairment of autonomic control of the cardiovascular system [15]. It is associated with several abnormalities in autonomic cardiovascular function. Subclinically, the disease is defined by cardiovascular reflex testing, which may have prognostic implications. Clinically, the impairment in autonomic function is associated with resting tachycardia, exercise intolerance, orthostatic hypotension, syncope, intraoperative cardiovascular instability, silent myocardial infarction and ischemia, and increased mortality [16].
The prevalence of cardiovascular autonomic neuropathy varies considerably, which reflects the tests used, diagnostic criteria, and the population studied. The Screen-Detected Diabetes in Primary Care (ADDITION) study reported an annual incidence of cardiovascular autonomic neuropathy of 1.8 percent in individuals with well-controlled diabetes [17]. This incidence is lower than in prior reports, which may reflect improvements in early detection and risk factor reduction.
Tachycardia and exercise intolerance — The earliest clinical manifestation of cardiac autonomic neuropathy may be a resting tachycardia. The increased resting heart rate is due to unopposed cardiac sympathetic nerve activity. As the autonomic neuropathy progresses, the heart rate gradually slows and in advanced cases will manifest as a fixed heart rate [18].
Exercise intolerance is usually due to impaired augmentation of cardiac output resulting from inadequate sympathetic modulation. Persistent sinus tachycardia can occur and there may be no variation in heart rate during activities that normally increase heart rate variability, such as deep breathing and the Valsalva maneuver.
Considerable interest has centered upon the role of abnormal myocardial electrical activity in arrhythmogenesis including, for example, QT prolongation and altered ventricular repolarization. In the EURODIAB Type 1 Complications Study, the prevalence of QT prolongation was 16 percent overall (11 percent in men, 21 percent in women) [19].
Cardiac denervation can occur in diabetic patients with advanced autonomic neuropathy. It is characterized by a fixed heart rate, in the range of 80 to 90 beats per minute, and is associated with painless myocardial infarction and sudden death [20].
Orthostatic (postural) hypotension — Orthostatic hypotension is defined as a fall in blood pressure of ≥20 mmHg systolic or ≥10 mmHg diastolic within three minutes of moving from a supine position to an upright tilt table test or standing position [21]. Orthostatic hypotension results from a combination of central and peripheral cardiovascular sympathetic denervation. It reflects failure of vasoconstriction in both the splanchnic and peripheral vascular beds. In its most severe disabling form, orthostatic hypotension can cause significant drops in blood pressure resulting in syncope. Retrospective data suggest that the presence of orthostatic hypotension is associated with microvascular and macrovascular complications of diabetes [22]. The presence of orthostatic hypotension in diabetes is associated with a significant increase in 10-year mortality [23,24]. Fortunately, this severity of orthostasis is rare, with the majority of patients having milder symptoms that are amenable to therapeutic interventions.
Orthostasis with DAN may have the following hemodynamic characteristics:
●A loss of the diurnal variation in blood pressure, with supine hypertension occurring at night [25]. This may also occur to a lesser degree in diabetic patients without neuropathy. In addition, ambulatory 24-hour blood pressure monitoring has detected hypertension in over 50 percent of subjects with type 2 diabetes and cardiovascular autonomic neuropathy, despite normal office blood pressure measurements [26].
●It is possible that day-to-day variability of symptoms may be exacerbated by insulin therapy, which has provoked hypotension in both diabetic [27] and nondiabetic [28] patients with autonomic failure, although the clinical relevance of this to most patients with diabetes is not known.
●Postprandial hypotension, whereby supine and standing systolic blood pressures may fall profoundly after meals [29]. The mechanism of postprandial hypotension in DAN is unclear. Both inadequate sympathetic compensation to meal-induced pooling of blood in the splanchnic circulation and vasodilatory gut peptides may contribute [30,31]. A hematocrit should also be checked, since anemia can result from erythropoietin deficiency secondary to renal denervation, which may exacerbate orthostatic symptoms [32]. (See "Mechanisms, causes, and evaluation of orthostatic hypotension".)
Postural tachycardia — A posture-induced tachycardia without a fall in blood pressure can result in significant postural symptoms of lightheadedness, dizziness, and presyncope. Although postural tachycardia syndrome (POTS) is a heterogeneous disorder, a diagnosis of POTS should not be made in individuals with diabetes; a more appropriate diagnosis would be DAN. Postural tachycardia is typically seen in patients with diabetes who have a high resting heart rate due to vagal cardiac neuropathy with an unopposed cardiac sympathetic nerve activity. This is very different from a diagnosis of POTS. (See "Postural tachycardia syndrome".)
The distinction between POTS and postural tachycardia related to DAN is a critical point for clinicians to be aware of, because many patients with diabetes are misdiagnosed with POTS. In such cases, the importance of DAN and its relationship to glycemic control are often dismissed, and the risks of morbidity and mortality associated with DAN are not typically considered.
Increased mortality — At least two meta-analyses of diabetic patients have found that cardiovascular autonomic neuropathy is associated with an increased risk of mortality [33,34]. In one meta-analysis, the mortality of autonomic neuropathy-free subjects over 5.5 years was approximately 5 percent, but this increased to 27 percent with the onset of abnormal cardiovascular reflex tests [33]. In a subsequent meta-analysis, the magnitude of the association was stronger for studies that required more than one abnormality of cardiovascular function to define cardiovascular autonomic neuropathy [34]. In the ACCORD trial, over a mean follow-up of 3.5 years, subjects with cardiovascular autonomic neuropathy at baseline had overall mortality and cardiovascular mortality rates that were approximately 1.6 to 2.1 and 1.9 to 3.0 times higher, respectively, compared with those who did not have cardiovascular autonomic neuropathy, even after adjusting for baseline cardiovascular risk factors [35].
Longitudinal studies of patients with DAN have typically found mortality rates over five years ranging between 16 and 53 percent (mean of approximately 30 percent) [36-40]. Although the majority of deaths result from associated macrovascular and microvascular disease, cardiorespiratory arrest secondary to autonomic denervation has been implicated in some diabetic patients [40,41]. In the EURODIAB Prospective Complications Study of over 2700 subjects with type 1 diabetes, an annual mortality rate of 5 per 1000 person-years was reported, with peripheral neuropathy (standardized hazard ratio [SHR] 1.88, 95% CI 1.06-3.35) and autonomic neuropathy (SHR 2.40, 95% CI 1.32-4.36) being the most important risk markers for mortality [42].
There are several potential mechanisms by which DAN may increase mortality rates. First, numerous studies have reported diminished perception of cardiac ischemia in individuals with diabetes [43-45]. Second, there is reduced cardiovascular response to physiologic stressors such as surgery or infection [46,47]. Third, there is an association between DAN and cardiac arrhythmias due to alterations in QT interval [48]. Fourth, alterations in the balance between sympathetic and parasympathetic systems have also been considered to be pro-arrhythmogenic [49]. Finally, denervated myocardial tissue that has focal areas of reinnervation may be at increased risk for arrhythmia [50].
Despite the evidence that DAN is associated with increased cardiac morbidity and mortality, sudden cardiac death in patients with DAN may have a stronger relationship with atherosclerotic heart disease and nephropathy than with DAN itself. In the Rochester Diabetic Neuropathy Study (RDNS), a prospective, longitudinal, population-based study, 21 cases of sudden cardiac death occurred in 462 patients with diabetes (151 with type 1) who were followed for over 15 years [51]. All 21 with sudden cardiac death had evidence of preceding severe atherosclerosis and myocardial damage. After adjusting for electrocardiogram (ECG) abnormalities (including evolving or previous Q wave changes indicative of myocardial infarction, left bundle branch block, or pacing) and stage of nephropathy, autonomic dysfunction was not significantly associated with sudden cardiac death.
Conversely, in a prospective observational study that followed patients with type 1 diabetes who had nephropathy (n = 197) or no nephropathy (n = 191) for 10 years, the presence of cardiovascular autonomic neuropathy (as measured by decreased heart rate variability) in the patients with nephropathy was an independent risk factor for cardiovascular morbidity and mortality [52]. Long-term follow-up data from the DCCT and Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study also found that individuals with type 1 diabetes diagnosed with cardiovascular autonomic neuropathy had a higher incidence of cardiovascular events over a 20-year period [53].
Studies reporting the highest mortality rates have typically enrolled symptomatic patients with abnormalities of both the sympathetic and parasympathetic divisions of the autonomic nervous system. Postural hypotension appears to predict a poorer prognosis [39].
Adverse cardiovascular events and silent ischemia — The presence of cardiovascular autonomic neuropathy is associated with adverse cardiac outcomes and with silent ischemia, as illustrated by the following studies:
●In the DIAD study, 522 patients with type 2 diabetes and no known or suspected coronary artery disease at baseline had adenosine stress radionuclide myocardial perfusion imaging studies, and silent ischemia was detected in 113 (22 percent) [54].
●A study of 120 patients with type 1 or type 2 diabetes and no history of myocardial infarction or angina but at least two additional cardiovascular risk factors followed for an average of 4.5 years found that a major cardiac event was significantly more common in patients with cardiovascular autonomic neuropathy than in those without cardiovascular autonomic neuropathy (24 versus 7 percent) [55].
●A meta-analysis of 12 cross-sectional studies found that patients with cardiovascular autonomic neuropathy had a significantly higher frequency of silent myocardial ischemia than patients without cardiovascular autonomic neuropathy (pooled prevalence rate ratio 1.96, 95% CI 1.53-2.51) [1].
The mechanisms of silent myocardial ischemia are complex and not completely understood. This issue is discussed separately. (See "Silent myocardial ischemia: Epidemiology, diagnosis, treatment, and prognosis", section on 'Pathophysiology'.)
Adverse renal and cerebrovascular outcomes — The presence of cardiovascular autonomic neuropathy may be associated with adverse renal and cerebrovascular outcomes, as suggested by the following studies:
●In a prospective study of subjects with type 1 diabetes who were followed for over 14 years, cardiovascular autonomic neuropathy and abnormal orthostatic diastolic blood pressure were associated with the subsequent development of renal complications [56].
●In patients with type 2 diabetes enrolled in the longitudinal Appropriate Blood Pressure Control in Diabetes (ABCD) trial, cardiovascular autonomic neuropathy was an independent risk factor for the occurrence of stroke [57].
●In a study including 1523 subjects with diabetes followed for 16 years, both a higher resting heart rate and a lower heart rate variability were associated with an increased risk of developing end-stage kidney disease [58].
Sleep apnea — There is evidence that sleep apnea and respiratory arrest may contribute to the increased mortality complicating DAN [59].
●Obstructive and/or central sleep apnea have been reported in 26 to 50 percent of subjects with type 1 diabetes and DAN [60,61].
●The number of oxygen desaturation episodes during sleep correlate with DAN severity [62].
●In an observational study of adults with a body mass index (BMI) ≤26 kg/m2 who had DAN and postural hypotension, the frequency of obstructive sleep apnea-hypopnea was >30 percent [63].
●In patients with obesity (mean BMI 35 kg/m2), DAN was associated with an increased risk of obstructive sleep apnea compared with other patients with obesity (with or without diabetes) [64].
Diagnostic testing — Subclinical autonomic neuropathy is often found in association with distal symmetric polyneuropathy and may only be detected by using cardiovascular reflex tests (such as the heart rate response to deep breathing and a Valsalva maneuver) or tests of peripheral sympathetic cholinergic function (such as quantitative sudomotor axon reflex testing or thermoregulatory sweat testing). Conventional measures of autonomic function use indirect methods relying on cardiovascular reflexes, which can detect early abnormalities in parasympathetic integrity but are relatively insensitive to sympathetic adrenergic deficits [65].
The selection of tests to evaluate autonomic function (table 3) has become reasonably standardized and should include measures of parasympathetic, sympathetic adrenergic, and sympathetic cholinergic function [66]. There is no evidence that any one test has diagnostic superiority [15,67,68], although it is rare that a single test would be administered.
Some experts believe, based on data from a meta-analysis, that a confident diagnosis of cardiovascular autonomic neuropathy requires abnormalities in more than one test, which evaluate different limbs of the reflex arc [1,34]. This notion underlies proposed criteria for the diagnosis and staging of cardiovascular autonomic neuropathy [15], which identify possible or early cardiovascular autonomic neuropathy on the basis of one abnormal cardiovagal test [69], definite or confirmed cardiovascular autonomic neuropathy on the basis of two abnormal cardiovagal tests [70], and severe or advanced cardiovascular autonomic neuropathy by the additional presence of orthostatic hypotension [71], which usually occurs late in the course of diabetes [67].
Direct assessment of cardiac sympathetic integrity is possible with radiolabeled analogues of norepinephrine, which are actively taken up by the sympathetic nerve terminals of the heart. The use of either 123-I-metaiodobenzylguanidine (MIBG, iobenguane I-123) or 11-C-hydroxyephedrine scintigraphy permits noninvasive assessment of the pattern of sympathetic innervation of the heart [72-75]. These tests have limited clinical utility because they are expensive and not widely available. They have, however, provided useful information because of their greater sensitivity to detect more subtle degrees of DAN than is possible by cardiovascular reflex testing:
●Among patients with type 1 diabetes and no evidence of DAN on cardiovascular reflex testing, abnormalities in cardiac innervation by cardiac MIBG scanning have been noted in some small studies [72,74].
●Diabetic patients with confirmed autonomic neuropathy have more pronounced abnormalities, but the MIBG scans do not differentiate between minimal or severe DAN on reflex testing [74,75].
●11-C-hydroxyephedrine undergoes highly specific uptake and retention in sympathetic nerve terminals [76,77] and, in comparison with MIBG, the images are less affected by non-neuronal uptake [78] and tissue attenuation [79]. This characteristic facilitates the quantitative regional characterization of sympathetic neuronal dysfunction and loss [80].
Treatment — Treatment is aimed primarily at prevention of disease progression. Medications that could cause or worsen neuropathy should be discontinued if possible (table 4). Modifiable risk factors include hyperglycemia, smoking, hypertension, and hyperlipidemia [3,8,81]. There is evidence that individuals who target these modifiable risk factors have minimal disease progression over a three-year period of time [82]. A cautionary note comes from the results of the ACCORD trial, in which intensive glycemic control was linked to increased mortality in type 2 diabetes [83]. The design of the ACCORD trial limited the ability to determine whether the differences in glycemia between the treatment groups or the different profile of medications utilized to achieve the glycemic levels was responsible for the excess mortality.
Additional interventions should include a program to increase overall cardiovascular fitness and weight control, since there is evidence that an exercise program can improve surrogate measures of both early and more advanced cardiovascular autonomic neuropathy [84,85]. However, a cardiac stress study is suggested before this program is initiated. In individuals with type 2 diabetes, there is a positive effect of weight loss on autonomic function, although it is unclear if this translates into a clinically meaningful autonomic outcome [86]. However, the clinical benefits of weight loss across all functional outcomes in type 2 diabetes support this clinical recommendation.
Beyond prevention of disease progression, patients with symptoms associated with DAN may require additional therapeutic intervention. In those with sympathetic adrenergic dysfunction and associated postural lightheadedness, the initial evaluation should include a detailed review and removal of any medications that may worsen orthostatic hypotension (eg, antihypertensive medications, antidepressant medications) [87]. Fluid intake should be increased and salt intake liberalized.
Pharmacologic therapy has included:
●Increasing the plasma volume with the mineralocorticoid fludrocortisone (0.1 to 0.4 mg/day) and a high-salt diet. This regimen may be helpful in more severe cases but can cause hypertension or peripheral edema. Higher doses of fludrocortisone increase the risk of side effects without evidence of additional improvement in symptoms, so are not recommended. (See "Treatment of orthostatic and postprandial hypotension", section on 'Fludrocortisone'.)
●Midodrine, an alpha-adrenoreceptor agonist, has been used to raise blood pressure and reduce symptoms of orthostatic hypotension [88,89]. Midodrine may cause severe supine hypertension, so patients should not use midodrine within six hours of bedtime.
●Droxidopa, an orally administered norepinephrine prodrug, has received regulatory approval in the United States and several Asian countries for treatment of symptoms associated with neurogenic orthostatic hypotension [90]. There is little published experience with the use of droxidopa in individuals with DAN and orthostatic hypotension.
●Octreotide (50 mcg three times daily, subcutaneously), a somatostatin analogue, may be helpful in diabetic patients with refractory and symptomatic postural or postprandial hypotension [29,31]. It acutely raises the blood pressure only in patients with autonomic failure but tends to be poorly tolerated because it can exacerbate bowel dysfunction and cause fluctuations in glycemic control.
●Treatment of anemia (including the judicious use of erythropoietin), if present, may also be helpful [91].
Nonpharmacologic therapy includes (see "Treatment of orthostatic and postprandial hypotension", section on 'Nonpharmacologic measures'):
●Making changes in posture slowly, ie, standing slowly in "stages."
●Elevation of the head of the bed by 10 to 20 degrees.
●Tensing the legs by crossing them while actively standing on both legs. In one report of patients with autonomic neuropathy, this procedure raised the cardiac output by 16 percent and the systemic blood pressure by 13 percent [92]. It can therefore minimize postural symptoms.
●Performing dorsiflexion of the feet or handgrip exercise before standing.
Measures aimed at increasing peripheral vascular tone (such as body stockings and gravity suits) are often tried. They may, however, prove ineffective since blood pooling probably occurs in the large splanchnic vascular bed [92]. Furthermore, in individuals with diabetic peripheral neuropathy, pressure sores can occur with compression stockings and should be avoided.
Supine hypertension (usually at night) is characteristic of autonomic neuropathy and may reflect a sympathetic/parasympathetic imbalance. Occasional patients have severe supine hypertension combined with orthostatic hypotension. Their management is challenging, and the approach is primarily guided by indirect evidence, clinical experience, and expert consensus [87]. Because of safety issues, treatment of the hypotension with volume expansion (eg, fludrocortisone) and sympathetic stimulation (eg, midodrine) usually takes precedence. However, this approach may worsen their supine hypertension. In order to minimize daytime hypotension, short-acting antihypertensive agents such as captopril, or hydralazine can be administered before bed, but patients need to be cautious about getting out of bed at night [87]. (See "Treatment of orthostatic and postprandial hypotension", section on 'Supine hypertension'.)
PERIPHERAL SUDOMOTOR AND VASOMOTOR NEUROPATHY — The peripheral sympathetic cholinergic system regulates sweat function and thermoregulation. In a length-dependent peripheral neuropathy there is loss of sweat function in a stocking and glove distribution. Clinically, this typically manifests as a compensatory proximal hyperhidrosis [93]. Progressive loss of sudomotor function can result in thermoregulatory impairment and hyperthermia. Peripheral autonomic nerve dysfunction may manifest as changes in the texture of the skin, itching, edema, venous prominence, callus formation, loss of nails, and sweating abnormalities of the feet [15]. The association between peripheral autonomic denervation and the resultant effects on the peripheral vasculature was recognized as early as 1941, when it was noted that diabetic patients with neuropathy had similar peripheral vasomotor reflexes as nondiabetic patients after sympathectomy [94]. This loss of sympathetic vascular innervation results in high peripheral blood flow through arteriovenous shunts and abnormal local reflex vascular control [16].
Peripheral autonomic neuropathy may be a contributing factor for the development of foot ulceration [95,96].
Peripheral autonomic neuropathy is thought to contribute to several other abnormalities:
●Symptoms such as aching, pulsation, tightness, cramping, dry skin, and pruritus.
●Peripheral edema, which is often associated with both foot ulceration and poor wound healing.
●The development of Charcot arthropathy (neuroarthropathy). In this condition, fractures can occur either spontaneously or with minimal stress and are followed by progressive bone disorganization with an increased risk of secondary ulceration (picture 1). (See "Diabetic neuroarthropathy".)
Diagnostic testing — Quantitative sudomotor axon reflex testing (QSART) can be used for the detection of early peripheral sympathetic denervation [97]. Galvanic skin response, also referred to as the sympathetic skin response, offers a surrogate measure of sympathetic innervation in the hands and feet [98]. This sympathetic skin response, however, is not a reliable measure of sudomotor function and is frequently absent in the older population.
Skin biopsy assessment has been increasingly utilized as a measure of small fiber neuropathy [99]. In addition, the assessment of sweat gland nerve fiber density has been studied as a marker of sudomotor neuropathy [100,101]. Despite growing acceptance as a useful marker of small fiber neuropathy, the clinical role of skin intraepidermal nerve fiber assessment for the quantitation of DAN remains unclear [102].
Measurement of vascular responses in the foot is an alternative method to detect peripheral sympathetic denervation. Thermal-induced vasoconstriction (rather than the normal vasodilation) reflects vascular denervation and is present only in those patients with both autonomic and somatic neuropathy [103]. Impairment of local axon reflex dilatation is thought to reflect depletion of local vasoactive neuropeptides. The role of alteration in the skin blood flow regulation in the development of foot ulceration is being evaluated. Although peripheral autonomic neuropathy correlates poorly with motor nerve dysfunction, motor nerve conduction velocity is decreased in patients with other evidence of small fiber damage, particularly loss of thermal sensation.
Treatment — Therapy of peripheral autonomic neuropathy mainly centers on careful foot care to prevent foot infection and ulceration. (See "Evaluation of the diabetic foot".)
Clinicians should avoid the temptation to treat a proximal compensatory hyperhidrosis because treatment increases the risk of hyperthermia. Lifestyle modifications such as cooling garments and avoidance of warm environments can improve symptoms.
GASTROINTESTINAL AUTONOMIC NEUROPATHY — The major gastrointestinal manifestations of DAN include gastroesophageal reflux disease (GERD), gastroparesis, and chronic diarrhea. This section will briefly review these problems, and they are discussed in greater detail elsewhere. (See "Diabetic autonomic neuropathy of the gastrointestinal tract".)
GERD may be caused by autonomic neuropathy with decreased lower esophageal sphincter (LES) pressure, increased number of transient LES relaxations due to hyperglycemia, impaired clearance function of the tubular esophagus, or delayed gastric emptying. The most common symptoms of GERD are heartburn (pyrosis) and regurgitation. Other extraesophageal manifestations of GERD include bronchospasm, laryngitis, and chronic cough. Dysphagia for liquids and/or solids is rarely seen in diabetes mellitus. (See "Diabetic autonomic neuropathy of the gastrointestinal tract", section on 'Gastroesophageal reflux disease'.)
Symptoms of gastroparesis include nausea, vomiting, early satiety, bloating, and/or upper abdominal pain. Patients with diabetic gastroparesis may also present with symptoms that are not directly related to the gastroparesis but are due to complications of poor glycemic control. The diagnosis of diabetic gastroparesis is established by evidence of delayed gastric emptying on scintigraphy in the absence of an obstructing structural lesion in the stomach or small intestine by endoscopy or imaging. Unlike other symptoms of autonomic neuropathy, gastroparesis may show some improvement with stable glycemic control [104]. Primary treatment of diabetic gastroparesis includes improved glycemic control, dietary modification, and administration of antiemetic and prokinetic agents in symptomatic patients. (See "Diabetic autonomic neuropathy of the gastrointestinal tract", section on 'Gastroparesis'.)
Diarrhea, and rarely steatorrhea, can occur in diabetics, particularly those with advanced disease. The pathogenesis of diabetic diarrhea is unclear, but multiple underlying mechanisms may be involved (table 5). The diarrhea is watery and painless, occurs at night, and may be associated with fecal incontinence. Bouts of diarrhea can be episodic with intermittent, normal bowel habits or even alternating with periods of constipation. Management of diabetic diarrhea includes general measures such as hydration and correction of electrolyte and nutrient deficiency, symptomatic treatment of diarrhea with antidiarrheals, and treatment of the underlying cause (eg, small intestinal bacterial overgrowth). (See "Diabetic autonomic neuropathy of the gastrointestinal tract", section on 'Diabetic diarrhea'.)
GENITOURINARY AUTONOMIC NEUROPATHY — Diabetic genitourinary autonomic neuropathy is responsible for several syndromes including bladder dysfunction, retrograde ejaculation, erectile dysfunction, and dyspareunia (due to decreased vaginal lubrication) [105]. Symptoms may be present in up to 50 percent of individuals with diabetes [106]. Erectile dysfunction is associated with the development of cardiovascular disease [107]. Diabetic bladder dysfunction initially presents as a decrease in the ability to sense a full bladder, secondary to loss of autonomic afferent innervation [108]. This abnormality leads to infrequent urination, while involvement of efferent nerves of the bladder results in incomplete emptying. These abnormalities can result in recurrent urinary tract infections and overflow incontinence with dribbling and poor urinary stream. Urinary incontinence is more common in type 1 diabetes. Lower urinary tract symptoms were present in approximately 20 percent of males with type 1 diabetes at the 10-year follow-up time point after completion of the Diabetes Control and Complications Trial (DCCT) [109]. In the same cohort, 38 percent of females reported urinary incontinence [110].
Sexual dysfunction is more common in females with diabetes compared with those without diabetes [111,112]. However, the evidence for diabetes as an independent risk factor for sexual dysfunction in females is equivocal; depression and anxiety appear to be more important predictors of sexual dysfunction [111,113]. (See "Overview of sexual dysfunction in females: Epidemiology, risk factors, and evaluation", section on 'Endocrine disorders'.)
In men, retrograde ejaculation reflects loss of coordinated internal urethral sphincter closure with external urethral sphincter relaxation during ejaculation. It may be manifest as cloudy urine postcoitally due to the presence of sperm. Impotence secondary to autonomic neuropathy usually occurs with other systemic manifestations of either somatic or autonomic neuropathy.
Treatment — Diabetic bladder dysfunction should initially be evaluated by a post-void residual, and further evaluated by complete urodynamic testing if necessary. The initial treatment consists of removal of medications that impair detrusor activity (anticholinergic agents, tricyclic antidepressants, and calcium channel antagonists) or agents that increase urethral sphincter tone (alpha-1 adrenoreceptor agonists). Treatment initially consists of a strict voluntary urination schedule, frequently coupled with the Crede maneuver. More advanced cases require intermittent catheterization.
The multiple potential etiologies of erectile impotence make it difficult to treat each patient with a standard protocol. An attempt should be made to establish the diagnosis. Potential causes include alcohol consumption, endocrine dysfunction, depression, functional (psychogenic) disorder, vascular disease, medications (iatrogenic), and autonomic neuropathy. (See "Evaluation of male sexual dysfunction".)
Phosphodiesterase-5 (PDE-5) inhibitors are first-line therapy for erectile dysfunction in those with diabetes. PDE-5 inhibitors are contraindicated in patients being treated with nitrates for heart disease. (See "Treatment of male sexual dysfunction".)
The management of female sexual dysfunction should be tailored to the specific sexual concerns and to underlying physical, psychological, and relationship factors. (See "Overview of sexual dysfunction in females: Management".)
OTHER MANIFESTATIONS — A number of other symptoms may occur with DAN, including:
●Pupillary abnormalities may result in failure in dark adaptation and difficulties in night driving.
●Alterations in neuroendocrine responses attributed to DAN include a reduction in glucagon and epinephrine secretion in response to hypoglycemia, thereby increasing the likelihood of hypoglycemic episodes, also referred to as hypoglycemia-associated autonomic failure [114-116]. In subjects with DAN, decreased counterregulatory catecholamine responses may increase the risk for severe hypoglycemia [117]. (See "Physiologic response to hypoglycemia in healthy individuals and patients with diabetes mellitus", section on 'Hypoglycemia-associated autonomic failure'.)
SCREENING — The American Diabetes Association recommends screening for DAN at the time of diagnosis of type 2 diabetes and five years after the diagnosis of type 1 diabetes [118].
●Screening should include a history and physical examination for signs of autonomic dysfunction
●Tests of heart rate variability may be indicated, including expiration-to-inspiration ratio, response to the Valsalva maneuver, and response to standing (table 3)
●If initial screening is negative, a history and physical examination for signs of autonomic dysfunction should be repeated annually
The identification of subclinical DAN is useful for risk stratification and for the individualization of glycemic, lipid, and blood pressure targets.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Neuropathy".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Nerve damage caused by diabetes (The Basics)")
●Beyond the Basics topics (see "Patient education: Diabetic neuropathy (Beyond the Basics)")
SUMMARY
●Classification – Diabetic autonomic neuropathy (DAN) is classified as subclinical or clinical depending upon the presence or absence of symptoms. A wide spectrum of manifestations can affect many different organs, including the cardiovascular, gastrointestinal, genitourinary, pupillary, sudomotor, and neuroendocrine systems (table 1).
•Cardiovascular autonomic neuropathy – Cardiovascular autonomic neuropathy is defined as the impairment of autonomic control of the cardiovascular system. The prevalence of cardiovascular autonomic neuropathy varies considerably, which reflects the tests used, diagnostic criteria, and the population studied. Subclinically and clinically, the severity of the disease is defined by cardiovascular autonomic reflex testing to assess parasympathetic and sympathetic adrenergic function. Clinically, the impairment in autonomic function is associated with resting tachycardia, exercise intolerance, orthostatic hypotension, intraoperative cardiovascular instability, silent myocardial infarction and ischemia, and increased mortality. (See 'Cardiovascular autonomic neuropathy' above.)
•Peripheral autonomic neuropathy – Peripheral autonomic nerve dysfunction may be manifest as changes in the texture of the skin, itching, edema, venous prominence, callus formation, loss of nails, and sweating abnormalities of the feet. Peripheral autonomic neuropathy may be a contributing factor for the development of foot ulceration, and may contribute to several other abnormalities such as aching, pulsation, tightness, cramping, dry skin and pruritus, peripheral edema, and the development of Charcot arthropathy (neuroarthropathy). (See 'Peripheral sudomotor and vasomotor neuropathy' above.)
•Gastrointestinal autonomic neuropathy – Gastrointestinal autonomic neuropathy can result in disorders of esophageal motility, gastric emptying (gastroparesis), and intestinal function. (See 'Gastrointestinal autonomic neuropathy' above.)
•Genitourinary autonomic neuropathy – Diabetic genitourinary autonomic neuropathy is responsible for several syndromes including bladder dysfunction, retrograde ejaculation, erectile dysfunction, and dyspareunia. (See 'Genitourinary autonomic neuropathy' above.)
●Recommendations for screening – The American Diabetes Association recommends screening for DAN at the time of diagnosis of type 2 diabetes and five years after the diagnosis of type 1 diabetes. (See 'Screening' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Martin Stevens, MD, who contributed to an earlier version of this topic review.
