GUIDELINE 11. ASSOCIATION OF LEVEL OF GFR WITH NEUROPATHY
Neuropathy develops during the course of chronic kidney disease and may become symptomatic.
- Patients with chronic kidney disease should be periodically assessed for central and peripheral neurologic involvement by eliciting symptoms and signs during routine office visits or exams.
- Specialized laboratory testing for neuropathy in patients with chronic kidney disease is indicated only in the presence of symptoms.
Neuropathy is a common complication of patients with kidney failure.438-440 Neuropathy may be manifested as encephalopathy, peripheral polyneuropathy, autonomic dysfunction, sleep disorders, and, less commonly, peripheral mononeuropathy. Occurrence of neuropathy is related to the level of kidney function, but not the type of kidney disease. However, there are certain causes of chronic kidney disease that also affect the central and/or peripheral nervous system. These are amyloidosis, diabetes, systemic lupus erythematosus, polyarteritis nodosa, and hepatic failure.438,439 In addition, there are congenital disorders that affect both the kidneys and nervous system, such as Von Hippel Lindau disease, Wilson's disease, and Fabry's disease.438
The pathophysiology of uremic neuropathy is not well understood. Levels of urea, creatinine, PTH, "middle molecules," and others have been correlated with reduction of nerve conduction velocity (NCV) and peripheral manifestations of neuropathy.438,439 In advanced stages there is evidence of histopathological damage with axonal degeneration and secondary demyelination of peripheral nerves.438
Markers of Neuropathy
Uremic neuropathy may affect the central, peripheral, or autonomic nervous systems. Early uremic encephalopathy may present with fatigue, impaired memory, or concentration. With more advanced uremia delirium, visual hallucinations, disorientation, convulsions, and coma may develop.438 Generally, uremic polyneuropathy is a symmetrical, mixed sensory and motor polyneuropathy, with distal nerves more severely affected. Patients may complain of pruritus, burning, muscle irritability, cramps, or weakness.438,439 Autonomic function abnormalities include impaired heart rate and blood pressure variability in response to respiratory cycle, postural change, and valsalva.
Signs on examination include muscle atrophy, loss of deep tendon reflexes, poor attention span, impaired abstract thinking, abnormal or absent reflexes (in particular ankle jerk), and impaired sensation (vibratory, light touch pressure, and pain).438,439 Later signs include meningismus, myoclonus, and asterixis.438 Electroencephalo!-!graphy (EEG) has generalized slowing, and bilateral spike and wave complexes have been described in up to 14% of patients, even in the absence of evident clinical seizure activity.438 EEG measures of sleep also are disturbed in dialysis patients.441 CT scan or MRI is not helpful, though there may be cerebral atrophy.438,442-444 The most sensitive test for detection of asymptomatic peripheral neuropathy is slowed sensory NCV; although motor NCV is slowed, there is a wide intra-individual day-to-day variation, and these findings occur with more advanced kidney dysfunction.438,439,445
Strength of Evidence
Neuropathy develops during the course of chronic kidney disease (R). Neuropathy is present in up to 65% of patients at the initiation of dialysis438,439; thus, it must begin to develop during an earlier phase of kidney disease. Symptoms of peripheral neuropathy generally do not present unless the GFR is under 12 to 20 mL/min, or uremia has been present for at least 6 months.438,439 Encephalopathy may become evident with less prolonged impairment of kidney function and can be seen with acute decline in GFR, although the correlation of central nervous system manifestations with level of kidney function is poor.438 Autonomic neuropathy is present in 20% to 80% of patients with diabetic nephropathy,442,444 in 66% of patients with severely impaired kidney function (creatinine clearance <8 mL/min), and in 50% of patients on dialysis.443
Objective findings of peripheral neuropathy as evaluated by NCV studies are present in 15% to 85% of individuals with decreased GFR.446-449 Sensory NCV is decreased in over 90% of patients, whereas motor NCV is decreased in only 40%.445 Among patients on dialysis, objective evidence of neuropathy is present in 50% to 100%,440,446 and the prevalence appears to increase with duration of dialysis.440
Objective evidence of central nervous system (CNS) dysfunction is not uniformly evident. EEG has been described to be minimally abnormal in a "small percentage" of patients445 or as slowed in most patients,450 with the degree of slowing more pronounced with more advanced dysfunction. EEG findings have been reported to improve after initiation of dialysis or with transplant.450 Tests of cognitive function were abnormal in all patients and were more impaired with increasing creatinine. Transplant and dialysis patients had somewhat better, but not normal, scores.450
Treatment with dialysis improves the more severe symptoms and findings of CNS involvement and improves the symptoms of polyneuropathy; however, NCV remains abnormal in up to 60% to 80%.440 The symptoms and findings of peripheral neuropathy are dramatically improved by transplantation.438
Neuropathy is associated with worse outcomes in chronic kidney disease (R). No articles were found that specifically related the presence of neuropathy to other outcomes among patients with chronic kidney disease. However, it is self-evident that impaired cognition and sleep, dysesthesias, and impaired autonomic function would at least lead to reduced quality of life and inability to function normally. If the neuropathy leads to skin ulcers, then certainly this would result in objective morbidity and potentially mortality. Advanced encephalopathy may result in seizures, coma, and death.438
Objective findings of neuropathy can be detected before symptoms arise (C, R). Several of the articles reviewed note that the majority of patients who have abnormalities in tests of nervous system function are asymptomatic.445,447,448 However, abnormalities are more profound among patients with symptoms.447,448
Onset and severity of neuropathy is associated with the level of GFR; there is insufficient evidence to define a specific threshold level of GFR that is associated with an increased prevalence or severity of neuropathy (C). The articles reviewed varied greatly in the levels of kidney function assessed, as well as in the measure of kidney function used, as some used only serum creatinine levels and other used GFR or creatinine clearance. Most studies demonstrated a relationship between kidney function and the particular marker of neuropathy. However, several studies only compared the particular marker with the normal or reference standard for the test or compared grouped data on patients with kidney disease with controls or patients on dialysis/transplant without providing data at various levels of kidney function. Summaries of the studies reviewed are presented in Tables 99 and 100.
Nerve conduction velocity (NCV) is slower in patients with decreased GFR (Table 99) (C).
There were 6 studies relating NCV to level of kidney function. The studies had sample sizes ranging from 40 to 210 subjects, with 29 to 72 patients with decreased kidney function not yet on dialysis. All but one450 of the studies showed that NCV was decreased below normal levels among patients with decreased kidney function. In three of the studies, the correlation between kidney function level and NCV was significant; in the other two correlation was suggested but lacked statistical significance. A threshold level of kidney function for abnormal motor NCV was only mentioned or deducible from three studies. Below a GFR of 8 to 13446,448 or serum creatinine above 7 to 8 mg/dL,445 50% or more patients with decreased kidney function had abnormal NCV. The threshold level at which 50% or more of patients have abnormal sensory NCV velocity was evaluated in only two studies and noted to be approximately 8 to 20 mL/min.445,448
These data generally confirm that NCV is decreased among patients with decreased kidney function. The reviewed studies do suggest a correlation between level of GFR and NCV, but they do not offer sufficient information to convincingly demonstrate a threshold level of GFR at which NCV becomes abnormal.
Memory and cognition are impaired in patients with decreased GFR (Table 100) (C). Only one study was found that evaluated memory and cognition among patients with decreased kidney function prior to the availability of erythropoietin.450 In this study of 177 subjects, of whom 72 had decreased kidney function not yet on dialysis, several cognitive functions were assessed, including sustained attention, selective attention, speed of decision-making, short-term memory, and mental manipulation of symbols. Each of these test measures was significantly lower among patients with decreased kidney function, correlated with level of dysfunction, and was improved to varying degrees among patients on dialysis and to a greater degree among patients with a kidney transplant.
Autonomic function is impaired in patients with decreased GFR (Table 101) (C).
Only three studies were found that objectively evaluated autonomic function among patients with kidney disease. These studies had between 42 and 123 subjects and between 21 and 67 patients with decreased kidney function not yet on dialysis. Each of these studies noted that autonomic function was impaired in more than 50% of patients with chronic kidney disease; however, only one of them found an association between level of kidney function and measures of autonomic nerve function.
The results of these studies cannot be extrapolated with confidence to the general population of patients with chronic kidney disease, as two were limited to patients with diabetes442,444 and thus confounded by the neuropathy ascribable to diabetes, and the third only had patients with very decreased kidney function (GFR <8 mL/min) or on dialysis.443
Symptoms of neuropathy, including sleep disturbances, are increased in patients with decreased GFR (C). Symptoms or clinical signs of peripheral neuropathy were evaluated or mentioned in four of the six studies of peripheral neuropathy reviewed for this guideline.445-448,450 The method of ascertaining the presence of symptoms or clinical findings was mentioned in only one of these studies447 as a "detailed neurological examination was carried out to find. . .evidence of clinically manifest neuropathy." The prevalence of symptoms or clinical findings ranged from 0% to 52%. Individuals with clinical symptoms had a greater reduction in NCV as compared to those without such symptoms in 2 studies,446,448 whereas there was no significant correlation between NCV and symptoms in one of the studies.447 None of the studies commented on the correlation between symptoms and level of kidney function; however, from a single study it was estimated that patients with symptoms had a lower mean level of kidney function (GFR = 6 mL/min) than patients without symptoms (GFR = 16 mL/min).446
The reviewed studies do not offer sufficient information to convincingly delineate a progressive increase in prevalence of symptoms with decreasing GFR.
Several of the reviewed articles included patients who had started dialysis or received a kidney transplant; information on these patients was used for background information and comparison. More articles than were reviewed were found with the literature search, but were not exhaustively reviewed as preliminary review suggested the lack of or inability to extract the necessary information. This may have led to the omission of some articles that may have provided further information.
These guidelines are limited by the inability to provide a definitive quantitative or semi-quantitative assessment of the relationship between level of kidney function and markers of neuropathy. This is in part due to the dearth of studies, the use of different measures of kidney function, the limited presentation of methods, and the failure to present adequate correlation data. In particular, there was extremely limited information on cognitive function and symptoms of neuropathy.
Lastly, many of the studies involved only a limited number of patients with mildly to moderately decreased kidney function, and two of the studies were limited to diabetics, confounding the results with the presence of diabetic neuropathy.
The data reviewed here suggest that symptoms of neuropathy begin to occur at very low levels of GFR. The inconclusive evidence presented herein has the implications that:
The applications suggested above are based on review of the available literature presented herein and opinion based on others' reviews of the problem. Treatment and assessment recommendations are beyond the scope of this guideline.
The only implementation issue arising from this guideline is to provide education regarding the prevalence of neuropathy, and the need to elicit symptoms and signs of this condition during routine office visits.
Much of the available information regarding neuropathy is derived from studies of patients with kidney failure. More information on neuropathy among patients with chronic kidney disease with earlier stages of chronic kidney disease may provide other means to follow progression of chronic kidney disease. In addition, if neuropathy were to be more carefully described and noted to have a high prevalence in earlier stages of chronic kidney disease and a relationship to kidney function, treatments to delay its progression could be considered.
The relationship between subjective and objective measures of neuropathy, and levels of kidney function, should be more accurately characterized. In addition, the relationship between neuropathy and kidney function should be separately studied among patients with additional risks of neuropathy, such as diabetics and patients with amyloidosis.