- Email: [email protected]
Haemodialysis alters peripheral nerve morphology in end-stage kidney disease
Clinical Neurophysiology xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Haemodialysis alters peripheral nerve morphology in end-stage kidney disease Adeniyi A. Borire a, Ria Arnold b, Bruce A. Pussell a, Natalie C. Kwai a, Leo H. Visser c, Luca Padua d,e, Neil G. Simon f, Matthew C. Kiernan g, Arun V. Krishnan a,⇑ a
Prince of Wales Clinical School, University of New South Wales, Sydney, Australia School of Medical Sciences, University of New South Wales, Sydney, Australia St Elisabeth Ziekenhuis, Tilburg, The Netherlands d Department of Geriatrics, Neurosciences and Orthopaedics, Università Cattolica del Sacro Cuore, Rome, Italy e Don Gnocchi Foundation, Milan, Italy f St Vincent’s Clinical School, University of New South Wales, Darlinghurst, NSW, Australia g Brain and Mind Centre, University of Sydney, Sydney, Australia b c
a r t i c l e
i n f o
Article history: Accepted 12 September 2016 Available online xxxx Keywords: End stage kidney disease Neuropathy Nerve ultrasound Cross-sectional area Hypoechoic fraction Total neuropathy score
h i g h l i g h t s Patients with end-stage kidney disease had larger and more hypoechoic nerves compared to normal
controls. The degree of nerve enlargement correlated significantly with electrophysiological parameters and
clinical severity. There was a significant decrease in nerve cross-sectional area and hypoechoic fraction following a
single dialysis session.
a b s t r a c t Objective: We explored the nerve ultrasound (US) characteristics of 15 patients with end-stage kidney disease (ESKD) and correlated these findings with clinical severity and electrophysiological parameters of neuropathy. Methods: 15 ESKD patients on thrice-weekly high-flux haemodialysis and 15 healthy controls were enrolled. Sonographic and electrophysiologic studies were conducted before and after a single session of haemodialysis. Serial measurements of median nerve cross-sectional area (CSA) and hypoechoic fraction (HF) were performed at the same non-entrapment site in the mid-forearm. Neuropathy severity was quantified using the total neuropathy score (TNS). Results: 86.7% of the ESKD cohort had neuropathy (TNS > 1). ESKD patients had significantly higher baseline CSA (8.9 ± 1.2 mm2 vs 7.5 ± 1.0 mm2, p < 0.05) and HF (56.0 ± 1.0% vs 54.0 ± 1.1%, p < 0.05) compared with the control group. The CSA correlated significantly with TNS (r = 0.826; p < 0.0001) and other electrophysiological parameters. There was a reduction in both the CSA (8.3 ± 1.4 mm2; p < 0.01) and HF (55.0 ± 1.6%; p < 0.05) after a single session of HD. A significant relationship was also found between the change in CSA and change in serum K+ after dialysis (r = 0.782, p < 0.01). Conclusions: This study shows that peripheral nerves in ESKD patients are larger and more hypoechoic and that these morphological abnormalities may be reversed by dialysis. Significance: US may be useful as an early marker of neuropathy in ESKD. Ó 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction
⇑ Corresponding author.
Neuropathy is the most common neurological complication of end-stage kidney disease (ESKD) occurring in 60–90% of patients
E-mail address: [email protected] (A.V. Krishnan). http://dx.doi.org/10.1016/j.clinph.2016.09.010 1388-2457/Ó 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
2
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
(Krishnan and Kiernan, 2009; Hojs-Fabjan and Hojs, 2006; Laaksonen et al., 2002; Van den Neucker et al., 1998). The most prominent clinical features are paraesthesia, numbness, reduction in deep tendon reflexes, impaired vibration sense, muscle atrophy and weakness, which are indicative of damage to large myelinated nerve fibers. Typically, symptoms progress in a length-dependent fashion, with greater lower-limb than upper-limb involvement (Tilki et al., 2009; Krishnan et al., 2009; Krishnan and Kiernan, 2007). Nerve conduction studies (NCS) often reveal a generalized neuropathy which is predominantly axonal with reductions in sensory and motor amplitudes (Krishnan and Kiernan, 2007). Nerve excitability studies, which provide information on axonal ion channel and membrane potential, have been studied extensively in ESKD patients and have shown prominent changes consistent with axonal membrane depolarization (Kiernan et al., 2002, 2000). Further studies have shown that these excitability abnormalities are rapidly reversed after a single session of dialysis (Krishnan et al., 2005, 2006a, 2006b). This is in contrast to traditional NCS which demonstrate minimal change over a dialysis session (Laaksonen et al., 2002). Despite the wealth of information provided by these studies, there are no studies exploring the changes in nerve morphology that may occur following dialysis in patients with ESKD. A considerable number of studies have evaluated the diagnostic utility of nerve ultrasound (US) as a marker of nerve morphology in mononeuropathies (Borire et al., 2016a,b; Simon et al., 2015; Cartwright et al., 2012), and more recently in acquired and inherited polyneuropathies (Ebadi et al., 2015; Goedee et al., 2013; Gallardo et al., 2015; Grimm et al., 2014; Shen and Cartwright, 2016; Visser and Beekman, 2011; Zaidman et al., 2009). These studies have demonstrated changes in the nerve cross-sectional area, fascicular size and arrangement as well as echogenicity. More recently, a few studies have assessed the sonographic characteristics of metabolic neuropathies, focussing largely on diabetic patients (Arumugam et al., 2016; Watanabe et al., 2009, 2010; Liu et al., 2012; Wei et al., 2012; Riazi et al., 2012). In this prospective study, we examined changes in nerve morphology across a dialysis session in a cohort of ESKD patients. We also explored the correlation of these findings with clinical severity and electrophysiological parameters.
2. Methods The study was approved by the Human Research Ethics Committee of the Prince of Wales Hospital, Sydney. Written informed consent was obtained from all study participants. This was a blinded, prospective, cross-sectional study of ESKD patients on hemodialysis (HD), recruited from two outpatient dialysis centers in Sydney (Prince of Wales and War Memorial Hospitals). Inclusion criteria comprised of ESKD patients aged 18–85 years, who were able to give informed consent, and who were maintained on hemodialysis for at least 6 months. Patients with a clinical history and/or electrodiagnostic features of carpal tunnel syndrome or previous traumatic median nerve injury were excluded from the study. The causes of ESKD in these patients included focal segmental glomerulosclerosis (n = 1), diabetes (n = 3), polycystic kidney disease (n = 2), IgA nephropathy (n = 2), glomerulonephritis (n = 2), interstitial nephritis (n = 1), reflux nephropathy (n = 2), obstructive uropathy (n = 1) and hypertensive nephrosclerosis (n = 1). With the exception of diabetes, patients did not have a history of other illnesses known to cause neuropathy. In total, 15 patients were recruited, receiving thrice-weekly HD using a PolyfluxÒ 201H (surface area 2.1 m2) dialyzer with a GambroÒ 200S dialysis machine (Gambro, Hechingen, Germany). Each dialysis session lasted between 4 and 6 h. The HD machines dialyzed
against pure water and GambroÒ Select Bag AX250G dialysis concentrate containing sodium (Na+) 140 mmol/L, bicarbonate 34 mmol/L, potassium (K+) 2.0 mmol/L, calcium 1.5 mmol/L, magnesium 0.50 mmol/L and glucose 1.0 g/L. All patients were adequately dialyzed as verified by the urea reduction ratio (URR > 65%) and equilibrated Kt/V (>1.2) (Daugirdas et al., 1997; Eknoyan et al., 2002). Clinical assessment as well as US and electrophysiologic studies were conducted before the first dialysis session for the week following a 2-day dialysis free period. Changes in hydration across dialysis were measured with three standard parameters: interdialytic weight change and blood pressure; and volume of ultrafiltration. Post-dialysis ultrasound measurements were performed 30 min after completion of HD. Nerve US was performed by the same investigator (AB) for all subjects, while electrophysiological studies were performed by a different investigator (RA). 15 ageand gender-matched controls were selected from a list of volunteers who had previously expressed interest in undertaking research studies. 2.1. Clinical assessment of neuropathy All patients underwent comprehensive clinical neurological examination prior to dialysis. The presence and severity of neuropathy was assessed using a modified version of the total neuropathy score (TNS) (Cornblath et al., 1999), a validated clinical measure of neuropathy severity (Arnold et al., 2013a, 2013b, 2016). The TNS is a composite score comprising of the following 8 domains: (i) sensory and (ii) motor neuropathic symptoms, (iii) pin prick sensibility, (iv) vibration detection, (v) strength assessment, (vi) deep tendon reflexes and lower limb (vii) sensory and (viii) motor nerve conduction studies. Each domain is allocated a severity score from 0 (normal) to 4 (severely abnormal). The sum of the individual domains gives the TNS ranging from 0 (no neuropathy) to 32 (disabling neuropathy). Serum electrolytes, urea, creatinine, calcium, magnesium, phosphate, parathyroid hormone and b-2 microglobulin and eGFR (estimated by the Modification of Diet in Renal Disease (MDRD) formula) (Levey et al., 1999) were also collected before and after dialysis. Nerve conduction studies (NCS) were undertaken on the tibial, sural, and median nerves using a Medelec Synergy system (Oxford Instruments, Abingdon, United Kingdom), based on standardized protocols (Kimura, 2013). Throughout the study, skin temperature was maintained above 32 °C. The following NCS parameters were recorded; sural nerve sensory nerve action potential (SNAP) amplitude and conduction velocity; tibial nerve compound muscle action potential (CMAP) amplitude and median nerve CMAP amplitude. Our local laboratory reference values were used. 2.2. Ultrasonography All patients underwent sonographic examination using a MyLabOne system with a 10–18 MHz linear probe (Esaote, Italy). Serial measurements of median nerve cross-sectional area were performed before and after dialysis at a non-entrapment site in the non-fistula arm. ‘Musculoskeletal’ factory preset (acoustic power 100%, line density set at medium, dynamic range set at 14, persistence set at 1) was used throughout the study. All machine settings, such as depth, gain and focus were also kept constant during each measurement. All patients were evaluated while sitting comfortably, with the forearm fully supinated and supported by an arm-rest, the elbow flexed to 80–90° and fingers semi-extended. The median nerve was first identified in the carpal tunnel inlet at the level of the pisiform bone. The nerve was then traced proximally, as it runs
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
3
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
between the superficial (flexor digitorum superficialis) and deep (flexor pollicis longus and flexor digitorum profundus) flexor muscles of the forearm. The median nerve was scanned in both transverse and longitudinal planes. The median nerve cross-sectional area (CSA) was measured (in mm2) using the continuous trace method, by outlining the inner margin of the epineurium (Cartwright et al., 2012; Borire et al., 2016). For each measurement, 2 separate CSA values were obtained and the average was used. The measurement site was at the junction of the middle and distal third of the forearm. The site was carefully marked to ensure consistent recordings before and after dialysis. The US probe was kept perpendicular to the nerve in order to maintain reproducibility of results. To eliminate confounding structural factors, we excluded patients with CTS and ensured that sonographic measurements were performed in the mid-forearm, a non-entrapment site. We also avoided the fistula arm to prevent potential vascular phenomena like vascular steal mechanism and venous hypertension, conditions which may affect nerve size and function.
2.3. Echogenicity The hypoechoic fraction (HF) is a validated objective marker of nerve echogenicity. The hypoechoic fraction is the percentage of the total region of interest that is below a set threshold of echogenicity. A high hypoechoic fraction suggests increased intraneural edema and vice versa. To obtain the HF, we analyzed the sonographic images of all subjects using ImageJ, an open-access software (National Institutes of Health, Bethesda, Maryland, USA). After uploading the images, images were converted to 8bit. The region of interest (ROI) was carefully selected by outlining the inner margin of the epineurium (Simon et al., 2015) and the ‘MaxEntropy’ automatic thresholding method was applied to identify the hypoechoic area within the ROI. The hypoechoic fraction (HF) was then calculated by using the ‘Measure Area Fraction’ function (with ‘Limit to Threshold’ selected) (Lee et al., 2016; Kapur et al., 1985). This process was repeated twice for each image and the mean value was then used for analysis (Fig. 1).
2.4. Data analysis Statistical analyses were conducted using SPSS (version 23.0, SPSS Inc., Chicago, IL, USA). The Shapiro–Wilk and Kolmogorov– Smirnov tests were used to test for normality of all continuous variables. Comparison between the pre-dialysis and post-dialysis measurements was performed using paired-samples t-test. Mann Whitney U test was used to test differences between other nonnormally distributed groups. Pearson (r) and Spearman coefficients (rho) were used to determine correlation for parametric and nonparametric variables respectively. Statistical significance was defined as p < 0.05. 3. Results 3.1. Clinical findings The baseline characteristics of the study participants and controls are presented in Table 1. The two groups were matched for age, gender and handedness. The mean TNS in the patient group was 10.4 ± 9.1, consistent with neuropathy of moderate severity. The average number of years on hemodialysis was 5.2 years, while the average duration of each dialysis session was 4.9 h. Dialysis adequacy as measured by URR (0.84 ± 0.12) and eKt/V (2.21 ± 0.87) were above expected targets (URR > 0.65 and Kt/V > 1.2). There was a difference between the pre-dialysis and post-dialysis measurements of systolic blood pressure (140.1 ± 23.3 mmHg vs 126.1 ± 17.0 mmHg, p = 0.020) and body weight (78.7 ± 13.0 mmHg vs 76.0 ± 12.9 mmHg, p < 0.001). The average volume of the ultrafiltration was 3.2 L. The prevalence of neuropathy in this cohort (defined as TNS > 1) was 86.7%, consistent with previous studies (Krishnan and Kiernan, 2009; HojsFabjan and Hojs, 2006; Laaksonen et al., 2002). 3.2. Ultrasound findings Median nerve US values are shown in Fig. 2. The mean crosssectional area of the control group (7.5 ± 1.0 mm2) was comparable to values from earlier studies: 7.43 ± 1.00 mm2 (Kang et al., 2016); 7.5 ± 1.7 mm2 (Grimm et al., 2014) and 6.6 ± 1.6 mm2 (Kerasnoudis et al., 2013). There was a significant difference between the baseline CSA values of patients and healthy controls (8.9 ± 1.2 mm2 vs 7.5 ± 1.0 mm2, p < 0.05). Similarly, the pre-dialysis HF was significantly higher in the patient group (56.0 ± 1.0% vs 54.0 ± 1.1%, p < 0.05). The optimal cut-off value for abnormal CSA and HF was
Table 1 Characteristics of patients and controls at baseline.
Fig. 1. Upper panel – Cross-sectional images of a healthy control’s median nerve in the mid-forearm showing (A1) normal fascicular pattern and (A2) hypoechoic fraction following the application of ‘MaxEntropy’ automatic thresholding method. Hypoechoic area is shown in black. Lower panel – Cross-sectional images of an ESKD patient showing (B1) loss of fascicular pattern and (A2) increased hypoechoic fraction. Hypoechoic area is shown in black.
ESKD patients (n = 15)
Control group (n = 15)
Demographics Age Male Gender (n,%) Diabetes (n)* Dry weight (kg) Pre-dialysis weight (kg)
60.1 ± 14.4 12 (80) 3 76.7 ± 13.4 78.6 ± 13.0
59.2 ± 5.9 12 (80) 0 83.6 ± 11.8 N/A
Pre-Dialysis serum biomarkers Creatinine (lmol/L) Urea (mg/dL) Potassium (mmol/L) Sodium (mmol/L) Parathyroid hormone (pmol/L) b2 microglobulin (mg/L)
809.1 ± 336.8 23.7 ± 7.5 4.9 ± 0.9 138.5 ± 3.2 27.1 ± 22.2 28.7 ± 15.3
N/A N/A N/A N/A N/A N/A
Except where indicated otherwise, the data are presented as mean ± SD. ESKD = end stage kidney disease, n = number of patients, N/A = not applicable. * p < 0.05.
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
4
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
Fig. 2. Ultrasound parameters in ESKD patients (before and after hemodialysis) and in normal controls. CSA – cross-sectional area; HD – hemodialysis; NC – normal controls; HF – hypoechoic fraction, ⁄ significant difference.
derived from the control data, using the sum of the mean and two standard deviations. 60% and 47% of the ESKD patients had abnormal baseline CSA and HF respectively compared to 40% and 33% respectively post-dialysis. Both US parameters correlated significantly at baseline (r = 0.402, p < 0.05). Following a session of hemodialysis, there was a significant reduction in both CSA and HF measured from the same site. The CSA was reduced from 8.9 ± 1.2 mm2 to 8.3 ± 1.4 mm2 (p < 0.01), while the HF reduced from 56.0 ± 1.0% to 55.0 ± 1.6% (p < 0.05).
p = 0.103). A significant relationship was also found between the change in CSA (difference between pre- and post-dialysis values) and change in serum K+ (difference between pre- and postdialysis serum K+) after a single session of dialysis (r = 0.782,
3.3. Correlation between ultrasound parameters, clinical assessment and serum electrolytes Correlation analysis was undertaken to investigate the potential relationship between US parameters and clinical findings. We found a strong correlation between the median nerve CSA at baseline and the TNS (r = 0.826; p < 0.0001). Pre-dialysis median nerve CSA also strongly correlated with NCS parameters. Specifically, correlations were noted between the baseline median nerve CSA and the following nerve conduction values: median CMAP amplitude (r = 0.825, p < 0.01) and sural SNAP amplitude (r = 0.682, p < 0.05) (Fig. 3). There was no significant relationship between sural nerve conduction velocity and baseline CSA (r = 0.494,
Fig. 4. Correlations between changes in ultrasound parameters and volume of ultrafiltration. CSA – cross-sectional area; DCSA – difference between pre- and post-dialysis CSA values; HF – hypoechoic fraction; DHF – difference between preand post-dialysis HF values. No significant relationship exists between the change in CSA or HF and the volume of ultrafiltration after a single session of dialysis.
Fig. 3. Correlations between nerve size, neuropathy severity and electrophysiological parameters. TNS – total neuropathy score; CSA – cross-sectional area; SNAP – sensory nerve action potential; CMAP – compound muscle action potential. Increasing nerve size correlated strongly with increasing neuropathy severity (A) as well as with changes in (B) sural sensory amplitude (lV) and (C) median motor amplitude (mV).
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
p < 0.01). However, no correlation was found between the change in CSA and changes in hydration status as measured by weight change (r = 0.275, p = 0.341), blood pressure change (r = 0.301, p = 0.295) or volume of ultrafiltration (r = 0.282, p = 0.374) (Fig. 4). There was no correlation between the baseline HF and TNS or NCS parameters. Similarly, there was no correlation between the baseline CSA or HF and other clinical variables such as years on dialysis treatment, duration of dialysis session, and measures of dialysis adequacy (URR and eKt/V).
5. Discussion The primary objective of this study was to evaluate the presence of morphological abnormalities in peripheral nerves of patients with ESKD receiving HD. The study demonstrates that morphological abnormalities in peripheral nerves in ESKD are significantly correlated with clinical measures of neuropathy. Furthermore, measures of CSA and HF are significantly altered after completion of dialysis which highlight acute changes in nerve morphology induced by dialysis. Baseline assessment demonstrated enlargement of the median nerve in ESKD patients, with significant correlations between the median nerve CSA and nerve conduction measures, specifically sural sensory amplitude and median motor amplitude. These changes were accompanied by higher HF in the patient group compared to normal controls, possibly reflecting intraneural edema. Earlier studies have suggested that pathological nerves are more hypoechoic than normal nerves due to intraneural edema (Martinoli et al., 2007; Watanabe et al., 2010). The exact mechanisms leading to intraneural edema in polyneuropathies remain unclear, however some of the theories proposed in entrapment neuropathies include perineurial thickening, vascular proliferation and increased vascular permeability (Simon et al., 2015). The finding of median nerve enlargement at baseline is consistent with the results of an earlier US study of ESKD patients (Wei et al., 2012), which demonstrated sural nerve enlargement (1.86 ± 0.53 mm2 vs 1.38 ± 0.25 mm2). Earlier reports have shown that the degree of nerve enlargement seen in axonal neuropathies is smaller compared to that seen in chronic demyelinating and hereditary neuropathies, which is in line with the results this study (Goedee et al., 2013; Zaidman et al., 2009; Watanabe et al., 2010). Similarly, conduction velocity did not correlate with nerve size, which is the opposite of what is seen in demyelinating neuropathies. This lends further support to the notion that the processes leading to nerve enlargement in metabolic and demyelinating neuropathies are different. This study also revealed rapid changes in the degree of nerve enlargement following a single dialysis session. The finding that dialysis induces acute changes in nerve morphology is novel and provides insight into the pathophysiology of neuropathy in ESKD as well as illustrating the dynamic nature of nerve US as an investigational tool. The acute change in nerve size and echogenicity is unlikely to be due to chronic structural changes in the nerve. While studies of nerve pathology in ESKD have shown segmental demyelination and remyelination as well as axonal degeneration (Thomas et al., 1971), these chronic changes are unlikely to be affected by acute interventions like HD. It is more likely that the change in nerve morphology was induced by dynamic processes such as the movement of electrolytes and water, which occur during HD. Although it might be expected that changes in nerve morphology may be related to removal of fluid during dialysis, we found no correlation between the US findings and measures of fluid shifts across the dialysis procedure. It does however seem likely that the improvements in US findings are short-lived and
5
future studies are needed to elucidate the duration of this improvement prior to the next dialysis session. Investigation of contributing serum metabolites revealed a strong correlation between changes in nerve morphology and K+ concentration. Previous studies have demonstrated that axons of ESKD patients are in a depolarized state (Kiernan et al.; 2000, 2002; Arnold et al., 2013c) and a recent interventional study demonstrated that hyperkalaemia was likely to be the underlying cause of the altered nerve membrane potential (Arnold et al., 2014). It has been hypothesized that prolonged membrane depolarization may induce morphological changes (Baker, 2002) by causing swelling of the node of Ranvier and myelin sheath and expansion of the periaxonal space, reflecting the movement of water underneath and within the myelin sheath (Baker, 2002; Love and Cruz-Höfling, 1986; Love et al., 1986). The myelin sheath plays a role in buffering excess extracellular K+, by expressing abundant K+ and chloride (Cl ) channels. These channels allow the passage of K+ and Cl ions together in an electroneutral manner and lead to movement of water into the Schwann cell, which results in swelling (Love and Cruz-Höfling, 1986). A potential shortcoming of this study is the fact that only one nerve site was studied. Sonographic examination of multiple nerves or multiple sites of the median nerve may provide greater certainty regarding the conclusions of the study. Anthropometric measurements were also not obtained in this study, and this may have an association with CSA. NCS were not repeated after HD and while this may have provided further information on the relationship between morphological and functional changes, previous work has demonstrated no significant change in NCS parameters following a single HD session (Laaksonen et al., 2002). In conclusion, this present study has shown that peripheral nerves of ESKD patients are enlarged and more hypoechoic, and that the degree of enlargement is related to the clinical and electrophysiological severity of neuropathy in ESKD patients. The study has also demonstrated rapid changes in nerve morphology across a single dialysis session, thus providing more insight into the effects of external influences such as electrolytes and fluid shifts on peripheral nerve morphology. Future studies are needed to explore the potential utility of US as an early marker of neuropathy in chronic kidney disease and to assess whether the degree of nerve enlargement varies according to dialysis type. Such studies may provide information on the type of dialysis that is most appropriate in preventing the development of chronic nerve injury in ESKD. Conflict of interest None of the authors have potential conflicts of interest to be disclosed. References Arnold R, Pussell BA, Pianta TJ, Lin C, Kiernan M, Krishnan A. Association between calcineurin inhibitor treatment and peripheral nerve dysfunction in renal transplant recipients. Am J Transplant 2013a;13:2426–32. Arnold R, Pussell BA, Pianta TJ, Grinius V, Lin CS, Kiernan MC, et al. Effects of hemodiafiltration and high flux hemodialysis on nerve excitability in end-stage kidney disease. PLoS One 2013b;8:e59055. Arnold R, Kwai NC, Krishnan AV. Mechanisms of axonal dysfunction in diabetic and uraemic neuropathies. Clin Neurophysiol 2013c;124:2079–90. Arnold R, Pussell BA, Howells J, Grinius V, Kiernan MC, Lin CS, et al. Evidence for a causal relationship between hyperkalaemia and axonal dysfunction in endstage kidney disease. Clin Neurophysiol 2014;125:179–85. Arnold R, Pussell BA, Kiernan M, Krishnan A. Comparative study to evaluate the effects of peritoneal and hemodialysis on peripheral nerve function. Muscle Nerve 2016;54:58–64. Arumugam T, Razali SN, Vethakkan SR, Rozalli FI, Shahrizaila N. Relationship between ultrasonographic nerve morphology and severity of diabetic sensorimotor polyneuropathy. Eur J Neurol 2016;23:354–60.
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
6
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
Baker MD. Electrophysiology of mammalian Schwann cells. Prog Biophys Mol Biol 2002;78:83–103. Borire A, Hughes A, Lueck C, Colebatch J, Krishnan A. Sonographic differences in carpal tunnel syndrome with normal and abnormal nerve conduction studies. J Clin Neurosci 2016. http://dx.doi.org/10.1016/j.jocn.2016.05.024. Borire A, Visser LH, Padua L, Colebatch J, Huynh W, Simon N, et al. Utility of maximum perfusion intensity as an ultrasonographic marker of intraneural blood flow. Muscle Nerve 2016. http://dx.doi.org/10.1002/mus.25200. Cartwright MS, Hobson-Webb LD, Boon AJ, Alter KE, Hunt CH, Flores VH, et al. Evidence-based guideline: neuromuscular ultrasound for the diagnosis of carpal tunnel syndrome. Muscle Nerve 2012;46:287–93. Cornblath DR, Chaudhry V, Carter K, Lee D, Seysedadr M, Miernicki M, et al. Total neuropathy score: validation and reliability study. Neurology 1999;53:1660–4. Daugirdas JT, Depner TA, Gotch FA, Greene T, Keshaviah P, Levin NW, et al. Comparison of methods to predict equilibrated Kt/V in the HEMO Pilot Study. Kidney Int 1997;1395–1405. Ebadi H, Siddiqui H, Ebadi S, Ngo M, Breiner A, Bril V. Peripheral nerve ultrasound in small fiber polyneuropathy. Ultrasound Med Biol 2015;41:2820–6. Eknoyan G, Beck GJ, Cheung AK, Daugirdas JT, Greene T, Kusek JW, et alHemodialysis (HEMO) Study Group. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med 2002;347:2010–9. Gallardo E, Noto Y, Simon NG. Ultrasound in the diagnosis of peripheral neuropathy: structure meets function in the neuromuscular clinic. J Neurol Neurosurg Psychiatry 2015;86:1066–74. Goedee HS, Brekelmans GJ, van Asseldonk J, Beekman R, Mess WH, Visser LH. High resolution sonography in the evaluation of the peripheral nervous system in polyneuropathy – a review of the literature. Eur J Neurol 2013;20:1342–51. Grimm A, Heiling B, Schumacher U, Witte OW, Axer H. Ultrasound differentiation of axonal and demyelinating neuropathies. Muscle Nerve 2014;50:976–83. Hojs-Fabjan T, Hojs R. Polyneuropathy in hemodialysis patients: the most sensitive electrophysiological parameters and dialysis adequacy. Wien Klin Wochenschr 2006;118(Suppl. 2):29–34. Kang S, Kim SH, Yang SN, Yoon JS. Sonographic features of peripheral nerves at multiple sites in patients with diabetic polyneuropathy. J Diabetes Complications 2016;30:518–23. Kapur JN, Sahoo PK, Wong AKC. A new method for gray-level picture thresholding using the entropy of the histogram. Comput Vision Graph 1985;29:273–85. Kerasnoudis A, Pitarokoili K, Behrendt V, Gold R, Yoon MS. Cross sectional area reference values for sonography of peripheral nerves and brachial plexus. Clin Neurophysiol 2013;124:1881–8. Kiernan MC, Burke D, Andersen KV, Bostock H. Multiple measures of axonal excitability: a new approach in clinical testing. Muscle Nerve 2000;23:399–409. Kiernan MC, Walters RJ, Andersen KV, Taube D, Murray NM, Bostock H. Nerve excitability changes in chronic renal failure indicate membrane depolarization due to hyperkalaemia. Brain 2002;125:1366–78. Kimura J. Electrodiagnosis in diseases of nerve and muscle: principles and practice. 4th ed. New York: Oxford University Press; 2013. Krishnan AV, Kiernan MC. Uremic neuropathy: clinical features and new pathophysiological insights. Muscle Nerve 2007;35:273–90. Krishnan AV, Kiernan MC. Neurological complications of chronic kidney disease. Nat Rev Neurol 2009;5:542–51. Krishnan AV, Phoon RK, Pussell BA, Charlesworth JA, Bostock H, Kiernan MC. Altered motor nerve excitability in end-stage kidney disease. Brain 2005;128:2164–74. Krishnan AV, Phoon RK, Pussell BA, Charlesworth JA, Bostock H, Kiernan MC. Neuropathy, axonal Na+/K+ pump function and activity-dependent excitability changes in end-stage kidney disease. Clin Neurophysiol 2006a;117:992–9.
Krishnan AV, Phoon RK, Pussell BA, Charlesworth JA, Kiernan MC. Sensory nerve excitability and neuropathy in end stage kidney disease. J Neurol Neurosurg Psychiatry 2006b;77:548–51. Krishnan AV, Pussell BA, Kiernan MC. Neuromuscular disease in the dialysis patient: an update for the nephrologist. Semin Dial 2009;22:267–78. Laaksonen S, Metsarinne K, Voipio-Pulkki LM, Falck B. Neurophysiologic parameters and symptoms in chronic renal failure. Muscle Nerve 2002;25:884–90. Lee H, Brekelmans GJ, Visser LH. Quantitative assessment of nerve echogenicity as an additional tool for evaluation of common fibular neuropathy. Clin Neurophysiol 2016;127:874–9. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461–70. Liu F, Zhu J, Wei M, Bao Y, Hu B. Preliminary evaluation of the sural nerve using 22MHz ultrasound: a new approach for evaluation of diabetic cutaneous neuropathy. PLoS One 2012;7:e32730. Love S, Cruz-Höfling MA. Acute swelling of nodes of Ranvier caused by venoms which slow inactivation of sodium channels. Acta Neuropathol 1986;70:1–9. Love S, Cruz-Höfling MA, Duchen LW. Morphological abnormalities in myelinated nerve fibres caused by Leiurus, Centruroides and Phoneutria venoms and their prevention by tetrodotoxin. Q J Exp Physiol 1986;71:115–22. Martinoli C, Tagliafico A, Bianchi S, Bodner G, Padua L, Schenone A, et al. Peripheral nerve abnormalities. Ultrasound Clin 2007;2:655–67. Riazi S, Bril V, Perkins BA, Abbas S, Chan VW, Ngo M, et al. Can ultrasound of the tibial nerve detect diabetic peripheral neuropathy? A cross-sectional study. Diabetes Care 2012;35:2575–9. Shen J, Cartwright MS. Neuromuscular ultrasound in the assessment of polyneuropathies and motor neuron disease. J Clin Neurophysiol 2016;33:86–93. Simon NG, Ralph JW, Poncelet AN, Engstrom JW, Chin C, Kliot MA. Comparison of ultrasonographic and electrophysiologic ‘inching’ in ulnar neuropathy at the elbow. Clin Neurophysiol 2015;126:391–8. Thomas PK, Hollinrake K, Lascelles RG, O’Sullivan DJ, Baillod RA, Moorhead JF, et al. The polyneuropathy of chronic renal failure. Brain 1971;94:761–80. Tilki HE, Akpolat T, Coskun M, Stalberg E. Clinical and electrophysiologic findings in dialysis patients. J Electromyogr Kinesiol 2009;19:500–8. Van den Neucker K, Vanderstraeten G, Vanholder R. Peripheral motor and sensory nerve conduction studies in haemodialysis patients. A study of 54 patients. Electromyogr Clin Neurophysiol 1998;38:467–74. Visser LH, Beekman R. Ultrasound of peripheral nerves. In: Walker FO, Cartwright MS, editors. Neuromuscular ultrasound. Philadelphia: Elsevier Saunders; 2011. p. 30–1. Watanabe T, Ito H, Morita A, Uno Y, Nishimura T, Kawase H, et al. Sonographic evaluation of the median nerve in diabetic patients: comparison with nerve conduction studies. J Ultrasound Med 2009;28:727–34. Watanabe T, Ito H, Sekine A, Katano Y, Nishimura T, Kato Y, et al. Sonographic evaluation of the peripheral nerve in diabetic patients: the relationship between nerve conduction studies, echo intensity, and cross-sectional area. J Ultrasound Med 2010;29:697–708. Wei M, Zhu J, Wang N, Xue Q, Liu F, Hu B, et al. Preliminary study of sural nerve morphological changes in uremic hemodialysis patients using 22-MHz highfrequency ultrasound. Muscle Nerve 2012;46:937–42. Zaidman CM, Al-Lozi M, Pestronk A. Peripheral nerve size in normals and patients with polyneuropathy: an ultrasound study. Muscle Nerve 2009;40:960–6.
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Haemodialysis alters peripheral nerve morphology in end-stage kidney disease Adeniyi A. Borire a, Ria Arnold b, Bruce A. Pussell a, Natalie C. Kwai a, Leo H. Visser c, Luca Padua d,e, Neil G. Simon f, Matthew C. Kiernan g, Arun V. Krishnan a,⇑ a
Prince of Wales Clinical School, University of New South Wales, Sydney, Australia School of Medical Sciences, University of New South Wales, Sydney, Australia St Elisabeth Ziekenhuis, Tilburg, The Netherlands d Department of Geriatrics, Neurosciences and Orthopaedics, Università Cattolica del Sacro Cuore, Rome, Italy e Don Gnocchi Foundation, Milan, Italy f St Vincent’s Clinical School, University of New South Wales, Darlinghurst, NSW, Australia g Brain and Mind Centre, University of Sydney, Sydney, Australia b c
a r t i c l e
i n f o
Article history: Accepted 12 September 2016 Available online xxxx Keywords: End stage kidney disease Neuropathy Nerve ultrasound Cross-sectional area Hypoechoic fraction Total neuropathy score
h i g h l i g h t s Patients with end-stage kidney disease had larger and more hypoechoic nerves compared to normal
controls. The degree of nerve enlargement correlated significantly with electrophysiological parameters and
clinical severity. There was a significant decrease in nerve cross-sectional area and hypoechoic fraction following a
single dialysis session.
a b s t r a c t Objective: We explored the nerve ultrasound (US) characteristics of 15 patients with end-stage kidney disease (ESKD) and correlated these findings with clinical severity and electrophysiological parameters of neuropathy. Methods: 15 ESKD patients on thrice-weekly high-flux haemodialysis and 15 healthy controls were enrolled. Sonographic and electrophysiologic studies were conducted before and after a single session of haemodialysis. Serial measurements of median nerve cross-sectional area (CSA) and hypoechoic fraction (HF) were performed at the same non-entrapment site in the mid-forearm. Neuropathy severity was quantified using the total neuropathy score (TNS). Results: 86.7% of the ESKD cohort had neuropathy (TNS > 1). ESKD patients had significantly higher baseline CSA (8.9 ± 1.2 mm2 vs 7.5 ± 1.0 mm2, p < 0.05) and HF (56.0 ± 1.0% vs 54.0 ± 1.1%, p < 0.05) compared with the control group. The CSA correlated significantly with TNS (r = 0.826; p < 0.0001) and other electrophysiological parameters. There was a reduction in both the CSA (8.3 ± 1.4 mm2; p < 0.01) and HF (55.0 ± 1.6%; p < 0.05) after a single session of HD. A significant relationship was also found between the change in CSA and change in serum K+ after dialysis (r = 0.782, p < 0.01). Conclusions: This study shows that peripheral nerves in ESKD patients are larger and more hypoechoic and that these morphological abnormalities may be reversed by dialysis. Significance: US may be useful as an early marker of neuropathy in ESKD. Ó 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction
⇑ Corresponding author.
Neuropathy is the most common neurological complication of end-stage kidney disease (ESKD) occurring in 60–90% of patients
E-mail address: [email protected] (A.V. Krishnan). http://dx.doi.org/10.1016/j.clinph.2016.09.010 1388-2457/Ó 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
2
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
(Krishnan and Kiernan, 2009; Hojs-Fabjan and Hojs, 2006; Laaksonen et al., 2002; Van den Neucker et al., 1998). The most prominent clinical features are paraesthesia, numbness, reduction in deep tendon reflexes, impaired vibration sense, muscle atrophy and weakness, which are indicative of damage to large myelinated nerve fibers. Typically, symptoms progress in a length-dependent fashion, with greater lower-limb than upper-limb involvement (Tilki et al., 2009; Krishnan et al., 2009; Krishnan and Kiernan, 2007). Nerve conduction studies (NCS) often reveal a generalized neuropathy which is predominantly axonal with reductions in sensory and motor amplitudes (Krishnan and Kiernan, 2007). Nerve excitability studies, which provide information on axonal ion channel and membrane potential, have been studied extensively in ESKD patients and have shown prominent changes consistent with axonal membrane depolarization (Kiernan et al., 2002, 2000). Further studies have shown that these excitability abnormalities are rapidly reversed after a single session of dialysis (Krishnan et al., 2005, 2006a, 2006b). This is in contrast to traditional NCS which demonstrate minimal change over a dialysis session (Laaksonen et al., 2002). Despite the wealth of information provided by these studies, there are no studies exploring the changes in nerve morphology that may occur following dialysis in patients with ESKD. A considerable number of studies have evaluated the diagnostic utility of nerve ultrasound (US) as a marker of nerve morphology in mononeuropathies (Borire et al., 2016a,b; Simon et al., 2015; Cartwright et al., 2012), and more recently in acquired and inherited polyneuropathies (Ebadi et al., 2015; Goedee et al., 2013; Gallardo et al., 2015; Grimm et al., 2014; Shen and Cartwright, 2016; Visser and Beekman, 2011; Zaidman et al., 2009). These studies have demonstrated changes in the nerve cross-sectional area, fascicular size and arrangement as well as echogenicity. More recently, a few studies have assessed the sonographic characteristics of metabolic neuropathies, focussing largely on diabetic patients (Arumugam et al., 2016; Watanabe et al., 2009, 2010; Liu et al., 2012; Wei et al., 2012; Riazi et al., 2012). In this prospective study, we examined changes in nerve morphology across a dialysis session in a cohort of ESKD patients. We also explored the correlation of these findings with clinical severity and electrophysiological parameters.
2. Methods The study was approved by the Human Research Ethics Committee of the Prince of Wales Hospital, Sydney. Written informed consent was obtained from all study participants. This was a blinded, prospective, cross-sectional study of ESKD patients on hemodialysis (HD), recruited from two outpatient dialysis centers in Sydney (Prince of Wales and War Memorial Hospitals). Inclusion criteria comprised of ESKD patients aged 18–85 years, who were able to give informed consent, and who were maintained on hemodialysis for at least 6 months. Patients with a clinical history and/or electrodiagnostic features of carpal tunnel syndrome or previous traumatic median nerve injury were excluded from the study. The causes of ESKD in these patients included focal segmental glomerulosclerosis (n = 1), diabetes (n = 3), polycystic kidney disease (n = 2), IgA nephropathy (n = 2), glomerulonephritis (n = 2), interstitial nephritis (n = 1), reflux nephropathy (n = 2), obstructive uropathy (n = 1) and hypertensive nephrosclerosis (n = 1). With the exception of diabetes, patients did not have a history of other illnesses known to cause neuropathy. In total, 15 patients were recruited, receiving thrice-weekly HD using a PolyfluxÒ 201H (surface area 2.1 m2) dialyzer with a GambroÒ 200S dialysis machine (Gambro, Hechingen, Germany). Each dialysis session lasted between 4 and 6 h. The HD machines dialyzed
against pure water and GambroÒ Select Bag AX250G dialysis concentrate containing sodium (Na+) 140 mmol/L, bicarbonate 34 mmol/L, potassium (K+) 2.0 mmol/L, calcium 1.5 mmol/L, magnesium 0.50 mmol/L and glucose 1.0 g/L. All patients were adequately dialyzed as verified by the urea reduction ratio (URR > 65%) and equilibrated Kt/V (>1.2) (Daugirdas et al., 1997; Eknoyan et al., 2002). Clinical assessment as well as US and electrophysiologic studies were conducted before the first dialysis session for the week following a 2-day dialysis free period. Changes in hydration across dialysis were measured with three standard parameters: interdialytic weight change and blood pressure; and volume of ultrafiltration. Post-dialysis ultrasound measurements were performed 30 min after completion of HD. Nerve US was performed by the same investigator (AB) for all subjects, while electrophysiological studies were performed by a different investigator (RA). 15 ageand gender-matched controls were selected from a list of volunteers who had previously expressed interest in undertaking research studies. 2.1. Clinical assessment of neuropathy All patients underwent comprehensive clinical neurological examination prior to dialysis. The presence and severity of neuropathy was assessed using a modified version of the total neuropathy score (TNS) (Cornblath et al., 1999), a validated clinical measure of neuropathy severity (Arnold et al., 2013a, 2013b, 2016). The TNS is a composite score comprising of the following 8 domains: (i) sensory and (ii) motor neuropathic symptoms, (iii) pin prick sensibility, (iv) vibration detection, (v) strength assessment, (vi) deep tendon reflexes and lower limb (vii) sensory and (viii) motor nerve conduction studies. Each domain is allocated a severity score from 0 (normal) to 4 (severely abnormal). The sum of the individual domains gives the TNS ranging from 0 (no neuropathy) to 32 (disabling neuropathy). Serum electrolytes, urea, creatinine, calcium, magnesium, phosphate, parathyroid hormone and b-2 microglobulin and eGFR (estimated by the Modification of Diet in Renal Disease (MDRD) formula) (Levey et al., 1999) were also collected before and after dialysis. Nerve conduction studies (NCS) were undertaken on the tibial, sural, and median nerves using a Medelec Synergy system (Oxford Instruments, Abingdon, United Kingdom), based on standardized protocols (Kimura, 2013). Throughout the study, skin temperature was maintained above 32 °C. The following NCS parameters were recorded; sural nerve sensory nerve action potential (SNAP) amplitude and conduction velocity; tibial nerve compound muscle action potential (CMAP) amplitude and median nerve CMAP amplitude. Our local laboratory reference values were used. 2.2. Ultrasonography All patients underwent sonographic examination using a MyLabOne system with a 10–18 MHz linear probe (Esaote, Italy). Serial measurements of median nerve cross-sectional area were performed before and after dialysis at a non-entrapment site in the non-fistula arm. ‘Musculoskeletal’ factory preset (acoustic power 100%, line density set at medium, dynamic range set at 14, persistence set at 1) was used throughout the study. All machine settings, such as depth, gain and focus were also kept constant during each measurement. All patients were evaluated while sitting comfortably, with the forearm fully supinated and supported by an arm-rest, the elbow flexed to 80–90° and fingers semi-extended. The median nerve was first identified in the carpal tunnel inlet at the level of the pisiform bone. The nerve was then traced proximally, as it runs
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
3
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
between the superficial (flexor digitorum superficialis) and deep (flexor pollicis longus and flexor digitorum profundus) flexor muscles of the forearm. The median nerve was scanned in both transverse and longitudinal planes. The median nerve cross-sectional area (CSA) was measured (in mm2) using the continuous trace method, by outlining the inner margin of the epineurium (Cartwright et al., 2012; Borire et al., 2016). For each measurement, 2 separate CSA values were obtained and the average was used. The measurement site was at the junction of the middle and distal third of the forearm. The site was carefully marked to ensure consistent recordings before and after dialysis. The US probe was kept perpendicular to the nerve in order to maintain reproducibility of results. To eliminate confounding structural factors, we excluded patients with CTS and ensured that sonographic measurements were performed in the mid-forearm, a non-entrapment site. We also avoided the fistula arm to prevent potential vascular phenomena like vascular steal mechanism and venous hypertension, conditions which may affect nerve size and function.
2.3. Echogenicity The hypoechoic fraction (HF) is a validated objective marker of nerve echogenicity. The hypoechoic fraction is the percentage of the total region of interest that is below a set threshold of echogenicity. A high hypoechoic fraction suggests increased intraneural edema and vice versa. To obtain the HF, we analyzed the sonographic images of all subjects using ImageJ, an open-access software (National Institutes of Health, Bethesda, Maryland, USA). After uploading the images, images were converted to 8bit. The region of interest (ROI) was carefully selected by outlining the inner margin of the epineurium (Simon et al., 2015) and the ‘MaxEntropy’ automatic thresholding method was applied to identify the hypoechoic area within the ROI. The hypoechoic fraction (HF) was then calculated by using the ‘Measure Area Fraction’ function (with ‘Limit to Threshold’ selected) (Lee et al., 2016; Kapur et al., 1985). This process was repeated twice for each image and the mean value was then used for analysis (Fig. 1).
2.4. Data analysis Statistical analyses were conducted using SPSS (version 23.0, SPSS Inc., Chicago, IL, USA). The Shapiro–Wilk and Kolmogorov– Smirnov tests were used to test for normality of all continuous variables. Comparison between the pre-dialysis and post-dialysis measurements was performed using paired-samples t-test. Mann Whitney U test was used to test differences between other nonnormally distributed groups. Pearson (r) and Spearman coefficients (rho) were used to determine correlation for parametric and nonparametric variables respectively. Statistical significance was defined as p < 0.05. 3. Results 3.1. Clinical findings The baseline characteristics of the study participants and controls are presented in Table 1. The two groups were matched for age, gender and handedness. The mean TNS in the patient group was 10.4 ± 9.1, consistent with neuropathy of moderate severity. The average number of years on hemodialysis was 5.2 years, while the average duration of each dialysis session was 4.9 h. Dialysis adequacy as measured by URR (0.84 ± 0.12) and eKt/V (2.21 ± 0.87) were above expected targets (URR > 0.65 and Kt/V > 1.2). There was a difference between the pre-dialysis and post-dialysis measurements of systolic blood pressure (140.1 ± 23.3 mmHg vs 126.1 ± 17.0 mmHg, p = 0.020) and body weight (78.7 ± 13.0 mmHg vs 76.0 ± 12.9 mmHg, p < 0.001). The average volume of the ultrafiltration was 3.2 L. The prevalence of neuropathy in this cohort (defined as TNS > 1) was 86.7%, consistent with previous studies (Krishnan and Kiernan, 2009; HojsFabjan and Hojs, 2006; Laaksonen et al., 2002). 3.2. Ultrasound findings Median nerve US values are shown in Fig. 2. The mean crosssectional area of the control group (7.5 ± 1.0 mm2) was comparable to values from earlier studies: 7.43 ± 1.00 mm2 (Kang et al., 2016); 7.5 ± 1.7 mm2 (Grimm et al., 2014) and 6.6 ± 1.6 mm2 (Kerasnoudis et al., 2013). There was a significant difference between the baseline CSA values of patients and healthy controls (8.9 ± 1.2 mm2 vs 7.5 ± 1.0 mm2, p < 0.05). Similarly, the pre-dialysis HF was significantly higher in the patient group (56.0 ± 1.0% vs 54.0 ± 1.1%, p < 0.05). The optimal cut-off value for abnormal CSA and HF was
Table 1 Characteristics of patients and controls at baseline.
Fig. 1. Upper panel – Cross-sectional images of a healthy control’s median nerve in the mid-forearm showing (A1) normal fascicular pattern and (A2) hypoechoic fraction following the application of ‘MaxEntropy’ automatic thresholding method. Hypoechoic area is shown in black. Lower panel – Cross-sectional images of an ESKD patient showing (B1) loss of fascicular pattern and (A2) increased hypoechoic fraction. Hypoechoic area is shown in black.
ESKD patients (n = 15)
Control group (n = 15)
Demographics Age Male Gender (n,%) Diabetes (n)* Dry weight (kg) Pre-dialysis weight (kg)
60.1 ± 14.4 12 (80) 3 76.7 ± 13.4 78.6 ± 13.0
59.2 ± 5.9 12 (80) 0 83.6 ± 11.8 N/A
Pre-Dialysis serum biomarkers Creatinine (lmol/L) Urea (mg/dL) Potassium (mmol/L) Sodium (mmol/L) Parathyroid hormone (pmol/L) b2 microglobulin (mg/L)
809.1 ± 336.8 23.7 ± 7.5 4.9 ± 0.9 138.5 ± 3.2 27.1 ± 22.2 28.7 ± 15.3
N/A N/A N/A N/A N/A N/A
Except where indicated otherwise, the data are presented as mean ± SD. ESKD = end stage kidney disease, n = number of patients, N/A = not applicable. * p < 0.05.
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
4
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
Fig. 2. Ultrasound parameters in ESKD patients (before and after hemodialysis) and in normal controls. CSA – cross-sectional area; HD – hemodialysis; NC – normal controls; HF – hypoechoic fraction, ⁄ significant difference.
derived from the control data, using the sum of the mean and two standard deviations. 60% and 47% of the ESKD patients had abnormal baseline CSA and HF respectively compared to 40% and 33% respectively post-dialysis. Both US parameters correlated significantly at baseline (r = 0.402, p < 0.05). Following a session of hemodialysis, there was a significant reduction in both CSA and HF measured from the same site. The CSA was reduced from 8.9 ± 1.2 mm2 to 8.3 ± 1.4 mm2 (p < 0.01), while the HF reduced from 56.0 ± 1.0% to 55.0 ± 1.6% (p < 0.05).
p = 0.103). A significant relationship was also found between the change in CSA (difference between pre- and post-dialysis values) and change in serum K+ (difference between pre- and postdialysis serum K+) after a single session of dialysis (r = 0.782,
3.3. Correlation between ultrasound parameters, clinical assessment and serum electrolytes Correlation analysis was undertaken to investigate the potential relationship between US parameters and clinical findings. We found a strong correlation between the median nerve CSA at baseline and the TNS (r = 0.826; p < 0.0001). Pre-dialysis median nerve CSA also strongly correlated with NCS parameters. Specifically, correlations were noted between the baseline median nerve CSA and the following nerve conduction values: median CMAP amplitude (r = 0.825, p < 0.01) and sural SNAP amplitude (r = 0.682, p < 0.05) (Fig. 3). There was no significant relationship between sural nerve conduction velocity and baseline CSA (r = 0.494,
Fig. 4. Correlations between changes in ultrasound parameters and volume of ultrafiltration. CSA – cross-sectional area; DCSA – difference between pre- and post-dialysis CSA values; HF – hypoechoic fraction; DHF – difference between preand post-dialysis HF values. No significant relationship exists between the change in CSA or HF and the volume of ultrafiltration after a single session of dialysis.
Fig. 3. Correlations between nerve size, neuropathy severity and electrophysiological parameters. TNS – total neuropathy score; CSA – cross-sectional area; SNAP – sensory nerve action potential; CMAP – compound muscle action potential. Increasing nerve size correlated strongly with increasing neuropathy severity (A) as well as with changes in (B) sural sensory amplitude (lV) and (C) median motor amplitude (mV).
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
p < 0.01). However, no correlation was found between the change in CSA and changes in hydration status as measured by weight change (r = 0.275, p = 0.341), blood pressure change (r = 0.301, p = 0.295) or volume of ultrafiltration (r = 0.282, p = 0.374) (Fig. 4). There was no correlation between the baseline HF and TNS or NCS parameters. Similarly, there was no correlation between the baseline CSA or HF and other clinical variables such as years on dialysis treatment, duration of dialysis session, and measures of dialysis adequacy (URR and eKt/V).
5. Discussion The primary objective of this study was to evaluate the presence of morphological abnormalities in peripheral nerves of patients with ESKD receiving HD. The study demonstrates that morphological abnormalities in peripheral nerves in ESKD are significantly correlated with clinical measures of neuropathy. Furthermore, measures of CSA and HF are significantly altered after completion of dialysis which highlight acute changes in nerve morphology induced by dialysis. Baseline assessment demonstrated enlargement of the median nerve in ESKD patients, with significant correlations between the median nerve CSA and nerve conduction measures, specifically sural sensory amplitude and median motor amplitude. These changes were accompanied by higher HF in the patient group compared to normal controls, possibly reflecting intraneural edema. Earlier studies have suggested that pathological nerves are more hypoechoic than normal nerves due to intraneural edema (Martinoli et al., 2007; Watanabe et al., 2010). The exact mechanisms leading to intraneural edema in polyneuropathies remain unclear, however some of the theories proposed in entrapment neuropathies include perineurial thickening, vascular proliferation and increased vascular permeability (Simon et al., 2015). The finding of median nerve enlargement at baseline is consistent with the results of an earlier US study of ESKD patients (Wei et al., 2012), which demonstrated sural nerve enlargement (1.86 ± 0.53 mm2 vs 1.38 ± 0.25 mm2). Earlier reports have shown that the degree of nerve enlargement seen in axonal neuropathies is smaller compared to that seen in chronic demyelinating and hereditary neuropathies, which is in line with the results this study (Goedee et al., 2013; Zaidman et al., 2009; Watanabe et al., 2010). Similarly, conduction velocity did not correlate with nerve size, which is the opposite of what is seen in demyelinating neuropathies. This lends further support to the notion that the processes leading to nerve enlargement in metabolic and demyelinating neuropathies are different. This study also revealed rapid changes in the degree of nerve enlargement following a single dialysis session. The finding that dialysis induces acute changes in nerve morphology is novel and provides insight into the pathophysiology of neuropathy in ESKD as well as illustrating the dynamic nature of nerve US as an investigational tool. The acute change in nerve size and echogenicity is unlikely to be due to chronic structural changes in the nerve. While studies of nerve pathology in ESKD have shown segmental demyelination and remyelination as well as axonal degeneration (Thomas et al., 1971), these chronic changes are unlikely to be affected by acute interventions like HD. It is more likely that the change in nerve morphology was induced by dynamic processes such as the movement of electrolytes and water, which occur during HD. Although it might be expected that changes in nerve morphology may be related to removal of fluid during dialysis, we found no correlation between the US findings and measures of fluid shifts across the dialysis procedure. It does however seem likely that the improvements in US findings are short-lived and
5
future studies are needed to elucidate the duration of this improvement prior to the next dialysis session. Investigation of contributing serum metabolites revealed a strong correlation between changes in nerve morphology and K+ concentration. Previous studies have demonstrated that axons of ESKD patients are in a depolarized state (Kiernan et al.; 2000, 2002; Arnold et al., 2013c) and a recent interventional study demonstrated that hyperkalaemia was likely to be the underlying cause of the altered nerve membrane potential (Arnold et al., 2014). It has been hypothesized that prolonged membrane depolarization may induce morphological changes (Baker, 2002) by causing swelling of the node of Ranvier and myelin sheath and expansion of the periaxonal space, reflecting the movement of water underneath and within the myelin sheath (Baker, 2002; Love and Cruz-Höfling, 1986; Love et al., 1986). The myelin sheath plays a role in buffering excess extracellular K+, by expressing abundant K+ and chloride (Cl ) channels. These channels allow the passage of K+ and Cl ions together in an electroneutral manner and lead to movement of water into the Schwann cell, which results in swelling (Love and Cruz-Höfling, 1986). A potential shortcoming of this study is the fact that only one nerve site was studied. Sonographic examination of multiple nerves or multiple sites of the median nerve may provide greater certainty regarding the conclusions of the study. Anthropometric measurements were also not obtained in this study, and this may have an association with CSA. NCS were not repeated after HD and while this may have provided further information on the relationship between morphological and functional changes, previous work has demonstrated no significant change in NCS parameters following a single HD session (Laaksonen et al., 2002). In conclusion, this present study has shown that peripheral nerves of ESKD patients are enlarged and more hypoechoic, and that the degree of enlargement is related to the clinical and electrophysiological severity of neuropathy in ESKD patients. The study has also demonstrated rapid changes in nerve morphology across a single dialysis session, thus providing more insight into the effects of external influences such as electrolytes and fluid shifts on peripheral nerve morphology. Future studies are needed to explore the potential utility of US as an early marker of neuropathy in chronic kidney disease and to assess whether the degree of nerve enlargement varies according to dialysis type. Such studies may provide information on the type of dialysis that is most appropriate in preventing the development of chronic nerve injury in ESKD. Conflict of interest None of the authors have potential conflicts of interest to be disclosed. References Arnold R, Pussell BA, Pianta TJ, Lin C, Kiernan M, Krishnan A. Association between calcineurin inhibitor treatment and peripheral nerve dysfunction in renal transplant recipients. Am J Transplant 2013a;13:2426–32. Arnold R, Pussell BA, Pianta TJ, Grinius V, Lin CS, Kiernan MC, et al. Effects of hemodiafiltration and high flux hemodialysis on nerve excitability in end-stage kidney disease. PLoS One 2013b;8:e59055. Arnold R, Kwai NC, Krishnan AV. Mechanisms of axonal dysfunction in diabetic and uraemic neuropathies. Clin Neurophysiol 2013c;124:2079–90. Arnold R, Pussell BA, Howells J, Grinius V, Kiernan MC, Lin CS, et al. Evidence for a causal relationship between hyperkalaemia and axonal dysfunction in endstage kidney disease. Clin Neurophysiol 2014;125:179–85. Arnold R, Pussell BA, Kiernan M, Krishnan A. Comparative study to evaluate the effects of peritoneal and hemodialysis on peripheral nerve function. Muscle Nerve 2016;54:58–64. Arumugam T, Razali SN, Vethakkan SR, Rozalli FI, Shahrizaila N. Relationship between ultrasonographic nerve morphology and severity of diabetic sensorimotor polyneuropathy. Eur J Neurol 2016;23:354–60.
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010
6
A.A. Borire et al. / Clinical Neurophysiology xxx (2016) xxx–xxx
Baker MD. Electrophysiology of mammalian Schwann cells. Prog Biophys Mol Biol 2002;78:83–103. Borire A, Hughes A, Lueck C, Colebatch J, Krishnan A. Sonographic differences in carpal tunnel syndrome with normal and abnormal nerve conduction studies. J Clin Neurosci 2016. http://dx.doi.org/10.1016/j.jocn.2016.05.024. Borire A, Visser LH, Padua L, Colebatch J, Huynh W, Simon N, et al. Utility of maximum perfusion intensity as an ultrasonographic marker of intraneural blood flow. Muscle Nerve 2016. http://dx.doi.org/10.1002/mus.25200. Cartwright MS, Hobson-Webb LD, Boon AJ, Alter KE, Hunt CH, Flores VH, et al. Evidence-based guideline: neuromuscular ultrasound for the diagnosis of carpal tunnel syndrome. Muscle Nerve 2012;46:287–93. Cornblath DR, Chaudhry V, Carter K, Lee D, Seysedadr M, Miernicki M, et al. Total neuropathy score: validation and reliability study. Neurology 1999;53:1660–4. Daugirdas JT, Depner TA, Gotch FA, Greene T, Keshaviah P, Levin NW, et al. Comparison of methods to predict equilibrated Kt/V in the HEMO Pilot Study. Kidney Int 1997;1395–1405. Ebadi H, Siddiqui H, Ebadi S, Ngo M, Breiner A, Bril V. Peripheral nerve ultrasound in small fiber polyneuropathy. Ultrasound Med Biol 2015;41:2820–6. Eknoyan G, Beck GJ, Cheung AK, Daugirdas JT, Greene T, Kusek JW, et alHemodialysis (HEMO) Study Group. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med 2002;347:2010–9. Gallardo E, Noto Y, Simon NG. Ultrasound in the diagnosis of peripheral neuropathy: structure meets function in the neuromuscular clinic. J Neurol Neurosurg Psychiatry 2015;86:1066–74. Goedee HS, Brekelmans GJ, van Asseldonk J, Beekman R, Mess WH, Visser LH. High resolution sonography in the evaluation of the peripheral nervous system in polyneuropathy – a review of the literature. Eur J Neurol 2013;20:1342–51. Grimm A, Heiling B, Schumacher U, Witte OW, Axer H. Ultrasound differentiation of axonal and demyelinating neuropathies. Muscle Nerve 2014;50:976–83. Hojs-Fabjan T, Hojs R. Polyneuropathy in hemodialysis patients: the most sensitive electrophysiological parameters and dialysis adequacy. Wien Klin Wochenschr 2006;118(Suppl. 2):29–34. Kang S, Kim SH, Yang SN, Yoon JS. Sonographic features of peripheral nerves at multiple sites in patients with diabetic polyneuropathy. J Diabetes Complications 2016;30:518–23. Kapur JN, Sahoo PK, Wong AKC. A new method for gray-level picture thresholding using the entropy of the histogram. Comput Vision Graph 1985;29:273–85. Kerasnoudis A, Pitarokoili K, Behrendt V, Gold R, Yoon MS. Cross sectional area reference values for sonography of peripheral nerves and brachial plexus. Clin Neurophysiol 2013;124:1881–8. Kiernan MC, Burke D, Andersen KV, Bostock H. Multiple measures of axonal excitability: a new approach in clinical testing. Muscle Nerve 2000;23:399–409. Kiernan MC, Walters RJ, Andersen KV, Taube D, Murray NM, Bostock H. Nerve excitability changes in chronic renal failure indicate membrane depolarization due to hyperkalaemia. Brain 2002;125:1366–78. Kimura J. Electrodiagnosis in diseases of nerve and muscle: principles and practice. 4th ed. New York: Oxford University Press; 2013. Krishnan AV, Kiernan MC. Uremic neuropathy: clinical features and new pathophysiological insights. Muscle Nerve 2007;35:273–90. Krishnan AV, Kiernan MC. Neurological complications of chronic kidney disease. Nat Rev Neurol 2009;5:542–51. Krishnan AV, Phoon RK, Pussell BA, Charlesworth JA, Bostock H, Kiernan MC. Altered motor nerve excitability in end-stage kidney disease. Brain 2005;128:2164–74. Krishnan AV, Phoon RK, Pussell BA, Charlesworth JA, Bostock H, Kiernan MC. Neuropathy, axonal Na+/K+ pump function and activity-dependent excitability changes in end-stage kidney disease. Clin Neurophysiol 2006a;117:992–9.
Krishnan AV, Phoon RK, Pussell BA, Charlesworth JA, Kiernan MC. Sensory nerve excitability and neuropathy in end stage kidney disease. J Neurol Neurosurg Psychiatry 2006b;77:548–51. Krishnan AV, Pussell BA, Kiernan MC. Neuromuscular disease in the dialysis patient: an update for the nephrologist. Semin Dial 2009;22:267–78. Laaksonen S, Metsarinne K, Voipio-Pulkki LM, Falck B. Neurophysiologic parameters and symptoms in chronic renal failure. Muscle Nerve 2002;25:884–90. Lee H, Brekelmans GJ, Visser LH. Quantitative assessment of nerve echogenicity as an additional tool for evaluation of common fibular neuropathy. Clin Neurophysiol 2016;127:874–9. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461–70. Liu F, Zhu J, Wei M, Bao Y, Hu B. Preliminary evaluation of the sural nerve using 22MHz ultrasound: a new approach for evaluation of diabetic cutaneous neuropathy. PLoS One 2012;7:e32730. Love S, Cruz-Höfling MA. Acute swelling of nodes of Ranvier caused by venoms which slow inactivation of sodium channels. Acta Neuropathol 1986;70:1–9. Love S, Cruz-Höfling MA, Duchen LW. Morphological abnormalities in myelinated nerve fibres caused by Leiurus, Centruroides and Phoneutria venoms and their prevention by tetrodotoxin. Q J Exp Physiol 1986;71:115–22. Martinoli C, Tagliafico A, Bianchi S, Bodner G, Padua L, Schenone A, et al. Peripheral nerve abnormalities. Ultrasound Clin 2007;2:655–67. Riazi S, Bril V, Perkins BA, Abbas S, Chan VW, Ngo M, et al. Can ultrasound of the tibial nerve detect diabetic peripheral neuropathy? A cross-sectional study. Diabetes Care 2012;35:2575–9. Shen J, Cartwright MS. Neuromuscular ultrasound in the assessment of polyneuropathies and motor neuron disease. J Clin Neurophysiol 2016;33:86–93. Simon NG, Ralph JW, Poncelet AN, Engstrom JW, Chin C, Kliot MA. Comparison of ultrasonographic and electrophysiologic ‘inching’ in ulnar neuropathy at the elbow. Clin Neurophysiol 2015;126:391–8. Thomas PK, Hollinrake K, Lascelles RG, O’Sullivan DJ, Baillod RA, Moorhead JF, et al. The polyneuropathy of chronic renal failure. Brain 1971;94:761–80. Tilki HE, Akpolat T, Coskun M, Stalberg E. Clinical and electrophysiologic findings in dialysis patients. J Electromyogr Kinesiol 2009;19:500–8. Van den Neucker K, Vanderstraeten G, Vanholder R. Peripheral motor and sensory nerve conduction studies in haemodialysis patients. A study of 54 patients. Electromyogr Clin Neurophysiol 1998;38:467–74. Visser LH, Beekman R. Ultrasound of peripheral nerves. In: Walker FO, Cartwright MS, editors. Neuromuscular ultrasound. Philadelphia: Elsevier Saunders; 2011. p. 30–1. Watanabe T, Ito H, Morita A, Uno Y, Nishimura T, Kawase H, et al. Sonographic evaluation of the median nerve in diabetic patients: comparison with nerve conduction studies. J Ultrasound Med 2009;28:727–34. Watanabe T, Ito H, Sekine A, Katano Y, Nishimura T, Kato Y, et al. Sonographic evaluation of the peripheral nerve in diabetic patients: the relationship between nerve conduction studies, echo intensity, and cross-sectional area. J Ultrasound Med 2010;29:697–708. Wei M, Zhu J, Wang N, Xue Q, Liu F, Hu B, et al. Preliminary study of sural nerve morphological changes in uremic hemodialysis patients using 22-MHz highfrequency ultrasound. Muscle Nerve 2012;46:937–42. Zaidman CM, Al-Lozi M, Pestronk A. Peripheral nerve size in normals and patients with polyneuropathy: an ultrasound study. Muscle Nerve 2009;40:960–6.
Please cite this article in press as: Borire AA et al. Haemodialysis alters peripheral nerve morphology in end-stage kidney disease. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.09.010