Mild hyponatremia, attention deficits and falls in elderly

Mild hyponatremia is associated with impaired cognition and increased falls in community-dwelling older persons

Roshan Gunathilake, Christopher Oldmeadow, Mark McEvoy, Brian Kelly, Kerry Inder, Peter Schofield, and John Attia

Department of General Medicine & Hunter Medical Research Institute, John Hunter Hospital, Newcastle, NSW, AustraliaSchool of Medicine and Public Health & Centre for Translational Neuroscience and Mental Health, University of Newcastle, NSW, Australia

J Am Geriatr Soc 2013 Oct; 61(10): 1838–1839

Introduction

• Mild hypoNa+ is usually asymptomatic.• Few previous studies suggest a probable

association between mild hypoNa+ and cognitive deficits leading to falls1-3.

• Included hospitalised patients with ensuing risk of Berkson’s bias and cognitive testing was limited to attention

Hypothesis

• Mild hyponatremia (serum Na+ >130 and ≤135 mmol/l) is associated with clinically subtle yet significant cognitive deficits, impaired balance, and increased propensity to fall among community-dwelling older persons

MethodsHunter Cohort Study4

• Community dwelling males and females aged 55 - 85 years residing in Newcastle

• Randomly recruited between Dec 2004 – Dec 2007 using a modified Dillman strategy

• Persons living in residential aged-care facility were excluded.

• Approved by HNE Human Research Ethics Committee and the NSW Population and Health Service Research Ethics Committee

Data collection

• 2 self-administered questionnaires– morbidity, diuretic use, previous falls, nutrition,

physical activity• clinical measurements

– heart rate, blood pressure, visual acuity, hearing, vibration, TUG, Functional Reach, pedometry

• Serum Na+: – Vista® V-Lyte® system (Siemens, Newark, DE),

normal range 136-144 mmol/l

Cognitive testing with ARCS5

• 34-minute cognitive battery that is administered via an audio device to unsupervised patients

• Five key domains: memory, verbal fluency, visuospatial, language, and attention/executive function

• Sum of five domains calibrated to a population mean of 100 and SD 15

• Previously validated in HCS cohort5

Falls risk assessment

• Surrogate markers–Functional reach6

–Timed Up and Go7

• Self-reported falls• Admission related to falls on health record

linkage

Statistical Analysis

Serum Na+ Fall

Agegender

Diuretic use

Heart failure

SSRIs

Cardiovas. dis

statins

Statistical analysis

• Multiple linear regression to assess the effect of serum Na+ on ARCS, FR & TUG

• Multiple logistic regression to model the effect of sodium on the probability of fall

• Significance testing: Partial sums of square F-tests, Wald statistics

Results: Baseline characteristics

Analyte n* distribution†‡

Age at first clinic 3282 60 66 73 (66.8±7.8)

Gender 2460 51% (1252)

Serum sodium (mmol/l) 2550 139 140 142 (140.6±2.8)

Diuretic use 2653 18% (480)

Functional reach (cm) 2895 26 30.5 35.0 (30.5±7.4)

TUG (sec) 2900 8.0 9.0 10.0 (9.5±2.9)

Any fall 3327 15% (500)

Total ARCS score 2423 89 99 108 (98±16)

Memory domain score 2423 74 85 93 (83±15)

Fluency domain score 2425 30 35 42 (35.9±9.1)

Visuospatial domain 2429 8 9 10 (8.3±2.3)

Language domain 2426 9 9 10 (9.1±1.1)

Attention domain score 2424 13 18 24 (18.8±8.6)

† a b c represent the lower quartile a, the median b, and the upper quartile c‡ x±s represents mean± 1 SD

Lower ARCS scores were seen in subjects with low serum Na+ content

  d.f. Partial SS MS F p

Na+ 2 2093 1047 4.360.0130

non-linear 1 1513 1513 6.300.0122

diuretic 1 148 148 0.610.4330

TOTAL 3 2502 834 3.470.0156

ERROR 1721 413468 240

Table 1: The model effect of changing serum sodium on cognition

Na 135 130 mmol/l ARCS↓4.67 (95% CI 1.56 – 7.79, F=6.3, p=0.012)

After adjusting for age, gender and diuretic use, hypoNa+ subjects were still more likely to have suffered a fall

 2 d.f. p

Na+ 8.16 2 0.0169non-linear 6.16 1 0.0131age 8.34 4 0.0798non-linear 1.24 3 0.7445gender 0.13 1 0.7194diuretic 0.48 1 0.4889TOTAL NON-LINEAR 7.67 4 0.1044TOTAL 21.15 8 0.0068

Table 2: Wald statistics for any fall

• 5 mmol/l drop in sodium from 135 to 130, increased the odds of falls by 32% (OR 1.32, 95% CI 1.04-1.64).

• A change in age of 13 years, from 60 to 73, increased the odds of falls by 27% (OR 1.27, 95% CI 1.01-1.59); i.e. a drop of 5mmol/l in sodium had about the same effect on falls as aging by 13 years.

Predicted probabilities for the logistic models of falls vs. serum Na+, adjusted for age, gender and diuretic use. An absolute risk of ≈25% for a female aged 66 years, on no diuretics, at a sodium level of 130mmol/l .

No statistically significant relationship between hyponatremia and either TUG or FR

Conclusions

• Even mild degrees of hyponatremia can be associated with significant cognitive deficits and increased risk of falls in otherwise asymptomatic, medically stable older adults.

Limitations

• Serial serum Na+ measurements were not available.

• Surrogate markers of falls risk were crude and may not have detected subtle abnormalities in posture and gait reported by previous authors3.

References

1. Gankam Kengne F, Andres C, Sattar L, et al. Mild hyponatremia and risk of fracture in the ambulatory elderly. QJM. 2008;101:583-8.

2. Renneboog B, Musch W, Vandemergel X, et al. Mild chronic hyponatremia is associated with falls, unsteadiness, and attention deficits. Am J Med. 2006; 119:71.e1-8

3. Decaux G. Is asymptomatic hyponatremia really asymptomatic? Am J Med. 2006; 119:S79-82

4. McEvoy M, Smith W, D'Este C, et al. Cohort profile: The Hunter Community Study. Int J Epidemiol. 2010; 39:1452-63

5. Schofield PW, Lee SJ, Lewin TJ, et al. The Audio Recorded Cognitive Screen (ARCS): a flexible hybrid cognitive test instrument. J Neurol Neurosurg Psychiatry. 2010; 81:602-7

6. Duncan PW, Weiner DK, Chandler J, et al. Functional reach: a new clinical measure of balance. J Gerontol. 1990;45: M192-7

7. Podsiadlo D, Richardson S. The timed "Up & Go": a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991; 39:142-8