Risk udary i upadki.pdf

ORIGINAL PAPER/ARTYKU£ ORYGINALNY

Risk factors for falls in post-stroke patients treated in a neurorehabilitation ward

Czynniki ryzyka upadków u chorych po udarze mózgu rehabilitowanych

na oddziale rehabilitacji neurologicznej

Anna Czernuszenko

Neurorehabilitation Unit of the Second Department of Neurology, Institute of Psychiatry and Neurology, Warsaw

Neurologia i Neurochirurgia Polska 2007; 41, 1: 28–35

A b s t r a c t

S t r e s z c z e n i e

Background and purpose: Patients with neurological

disorders, including post-stroke patients, are at high risk

for falls. The purpose of the study was to estimate the

number and type of falls and fall-related injuries in patients

of a neurological rehabilitation ward and to identify risk

factors for falls.

Material and methods: 353 consecutive post-stroke

patients of the Neurorehabilitation Unit at the Institute of

Psychiatry and Neurology in Warsaw were enrolled into the

study and observed during the hospital stay. Falls occurring

in patients during their hospital stay were registered and an

analysis of fall-related data and the routinely obtained

medical data was performed.

Results: In the evaluated group, 35 (10%) patients suffered

50 falls, which corresponds to an incidence rate of 5.02 [95%

confidence interval (95% CI): 3.70-6.52] falls per 1000

patient-days. Most falls occurred in patients’ rooms (80%),

most of them taking place while transferring from one place

to another or while changing position. 2% of falls caused

severe injury. There was a higher percentage of patients with

unilateral neglect in the group of fallers (p=0.0007), who also

demonstrated poorer performance in activities of daily living

as expressed by the Rankin score (p <0.0002) and the

Barthel Index (p <0.0001), and greater neurological deficit

in the Scandinavian Stroke Scale (SSS) (p=0.002) at

admission as well as at discharge from hospital than

Wstêp i cel pracy: Pacjenci ze schorzeniami neurologiczny-

mi, a wœród nich chorzy po udarze, s¹ nara¿eni na czêste

upadki. Badanie mia³o na celu identyfikacjê czynników ry-

zyka upadków u pacjentów rehabilitowanych szpitalnie

po udarze mózgu, a tak¿e oszacowanie ich liczby, rodzaju

i konsekwencji.

Materia³ i metody: Do badania w³¹czono kolejnych 353 pa-

cjentów po udarze mózgu rehabilitowanych na Oddziale

Rehabilitacji Neurologicznej Instytutu Psychiatrii i Neuro-

logii. Rejestrowano upadki pacjentów podczas ich pobytu

na oddziale, nastêpnie analizowano dane dotycz¹ce upad-

ków oraz rutynowo pozyskiwane dane medyczne.

Wyniki: W badanej grupie odnotowano 50 upadków u 35

(10%) chorych. WskaŸnik upadków wynosi 5,02/1000 oso-

bodni [95% przedzia³ ufnoœci (95%CI): 3,70–6,52]. Wiêk-

szoœæ upadków (80%) nast¹pi³a w pokoju chorego. Do upad-

ków najczêœciej dochodzi³o podczas przesiadania i zmian po-

zycji. 2% upadków skutkowa³o powa¿nymi obra¿eniami cia³a.

Grupa upadaj¹cych charakteryzowa³a siê wiêkszym odset-

kiem osób z zespo³em pomijania stronnego (p=0,0007), ni¿-

szymi ocenami pocz¹tkowymi i koñcowymi w skalach Rankin

(p <0,0002), Barthel (p <0,0001) i w Skandynawskiej Skali

Udarowej (p=0,002). Ryzyko upadku najsilniej zwi¹zane by-

³o z punktacj¹ poni¿ej 15/20 w skali Barthel [ryzyko wzglêd-

ne (RR)=10,3 (95% CI: 2,8–50,7)], z punktacj¹ powy¿ej

3 w skali Rankina [RR=5,12 (95% CI: 2,41–14,8)], punkta-

Address for correspondence: Anna Czernuszenko, MD, Neurorehabilitation Unit of the Second Department of Neurology, Institute of Psychiatry

and Neurology, Warsaw, Sobieskiego 9, 02-957 Warszawa; e-mail: beres@mp.pl

Received: 11.05.2006; accepted: 18.11.2006

Neurologia i Neurochirurgia Polska 2007; 41, 1

Falls in post-stroke patients

non-fallers. The risk of a fall showed greatest association with

low Barthel Index at admission (<15/20; RR=10.3; 95% CI:

2.8-50.7), performance in the Rankin score of >3

(RR=5.12; 95% CI: 2,41-14,8), low SSS score on admission

(0-39; RR=3.40; 95% CI: 1.62-9.14), and the presence of

hemispatial neglect (RR=3.42; 95% CI: 1.23-6.53). The

multiple-fallers group did not differ significantly from the

group of single-fallers.

Conclusions: Among post-stroke patients, falls are a

complication especially affecting persons with a large

neurological deficit, severe disability and the neglect syndrome.

Key words: falls, rehabilitation, stroke.

cj¹ poni¿ej 39 w Skandynawskiej Skali Udarowej [RR=3,40

(95% CI: 1,62–9,14)] oraz z obecnoœci¹ zespo³u zaniedbywa-

nia [RR=3,42 (95% CI: 1,23–6,53)]. Pacjenci, którzy prze-

wrócili siê wiele razy nie ró¿nili siê znamiennie od tych, którzy

upadli raz.

Wnioski: W grupie chorych po udarze upadki dotycz¹

zw³aszcza osób z du¿ym deficytem neurologicznym, ciê¿k¹

niepe³nosprawnoœci¹ i zespo³em zaniedbywania.

S³owa kluczowe: upadki, rehabilitacja, udar mózgu.

Introduction

rehabilitation ward predicted further falls during the

post-hospital period [5]. Awareness of the risk

factors, mechanisms and causes of falls may help in

limiting their number and sequelae in patients who

have had a stroke, which may significantly improve

the effects of post-stroke rehabilitation and the quality

of life of patients and their care-givers.

The aim of the present study was to assess the

incidence, circumstances and sequelae of falls among

patients of the Neurorehabilitation Unit and to verify

the risk factors for falls.

Falls constitute an important and difficult problem

in the elderly population. They are the main cause of

injuries and fractures in this age group [1]. The

incidence of falls in the general population increases

with age, while the risk of serious sequelae of falls also

increases with age and with worsening of health. Apart

from that, falls are one of the most frequent

complications of hospitalizations in neurological,

geriatric and post-stroke rehabilitation wards [2-10].

In post-stroke patients, the risk of falls is twice as high

as that in the general population [11] and refers to

approximately 40-70% of patients during the first year

after stroke [5,12,13]. This group of patients is

concomitantly exposed to dynamic development of

osteoporosis, especially on the paretic side [14]. It is

estimated that in patients with hemiparesis, a 14%

reduction in bone mineral density occurs during the

first year following stroke [15], which exceeds by

approximately twenty-fold bone mass reduction

observed per year in healthy persons at the same age.

High risk of falls together with large bone mass loss in

this group of patients results in a four-fold increase in

the risk of femoral neck fractures [1].

Specific risk factors for falls have been defined for

the population of post-stroke patients. They include:

male gender, limited independence in activities of

daily living, urinary incontinence, balance

disturbances, presence of left-sided or bilateral motor

deficit, sensory disturbances, unilateral neglect, use of

hypotensive, diuretic, anti-depressant, psychotropic

or sedative agents, history of previous falls, and

cognitive function disturbances [11,16-21]. It was

also observed that occurrence of falls in a

Material and methods

The study was conducted in the Neurorehabilitation

Unit at the Second Department of Neurology at the

Institute of Psychiatry and Neurology in Warsaw. It is a

forty-bed ward for inpatients. Patients with acquired

lesions of the central nervous system, most often as a

result of stroke, are admitted to the ward, usually within

the first three months following the onset of the disease.

The study was conducted in a prospective manner.

All 353 patients with established diagnosis of stroke

based on clinical criteria and neuroimaging studies

who were admitted to the ward between June 1, 2002

and July 31, 2003 were enrolled into the study.

Duration of observation ranged from 1 to 74 days

(mean: 28.2±13.3 days) and was equal to the number

of days of patients’ hospitalisation. Characteristics of

the studied group are presented in Table 1.

Falls were recorded in a “Notebook of falls”

immediately after such an event by the nursing staff.

The notes included information on the time, site,

circumstances and sequelae of falls.

Neurologia i Neurochirurgia Polska 2007; 41, 1

Anna Czernuszenko

Table 1. Characteristics of the study population

the scales assessing independence in the activities of

daily living (ADL) at baseline and by the end of the

hospital stay – the Rankin scale, the Barthel index, and

the Scandinavian Stroke Scale (SSS).

The results were presented as mean values,

frequencies expressed as percent values, and

incidence rates expressed as the ratio of the number

of events to the number of patient-days during the

observation period. To compare frequencies between

the groups, the

N=353

gender

199

56.4%

154

43.6%

age [years] mean (SD)*

62 (14)

duration of stay [days] mean (SD)

28.2 (13.3)

time since disease onset [days] median (range)

41.5 (5-1592)

type of stroke

ischaemic

or the Fisher exact test was used

depending on the sample size. Differences in mean

values were assessed using the Mann-Whitney

U-test. The level of statistical significance was set at

p<0.05. For the significant variables, relative risks

(RR) and 95% confidence intervals (95% CI) were

calculated. Statistica 6.0 software was used for the

statistical analysis.

300

85%

haemorrhagic

14.2%

not identified

0.8%

side of limb paresis

right

169

48.1%

left

154

43.9%

both

5.7%

none

2.3%

Results

dysphasia

137

38.8%

unilateral neglect

10.8%

Thirty-five of the observed 353 patients (10%)

fell. Fifty falls were reported. The incidence rate of

falls is 5.02 per 1000 patient-days (95% CI 3.70-6.52),

which corresponds to 1 fall per 5 days in the

conditions of our ward.

For the first fall, the incidence ratio is 3.85 per 1000

patient-days (95% CI 2.57-512), and for subsequent

falls, 21.9 per 1000 patient-days (95% CI 12.3-35.8).

Falls occurred most frequently during the first

(42%) and the second (30%) week of hospitalisation.

Similarly, the incidence rate of falls was highest in the

first week of hospitalisation and tended to decrease in

subsequent weeks. Figure 1 shows the relationship

between the incidence rate of falls and the duration of

hospital stay; it illustrates a reduction in the risk of a

fall with longer hospital stay.

sensory deficit

18.1%

current medication

antidepressants

102

28.9%

diuretics

27.8%

hypotensives

253

71.7%

sedatives

12.2%

anticonvulsants

9.6%

neuroleptics

4.3%

antiparkinsonian agents

0.8%

Rankin 0** mean (SD)

3.3 (1.2)

Rankin 1*** mean (SD)

2.7 (1.2)

Barthel 0** mean (SD)

11.3 (5.9)

Barthel 1*** mean (SD)

14.5 (5.4)

SSS 0** mean (SD)

37.0 (11)

SSS 1*** mean (SD)

42.6 (10.5)

During the day, a marked increase in the number

of falls was observed between 11 a.m. and 12 p.m., and

a second, although much smaller, peak between 5 a.m.

and 6 a.m. Figure 2 illustrates the circadian

distribution of falls.

Twenty-six of the 35 (74.3%) patients who fell

during hospitalisation had a single fall; in 9 (24.7%)

persons, at least two falls (2-4) were observed. Multiple

falls in the same patients accounted for 48% of all events.

The majority of falls occurred in patients’ rooms

(80%), while they much less frequently occurred in the

bathroom (8.3%) or in the hallway (6.3%).

falls

10%

* SD – standard deviation

** 0 – score at admission

*** 1 – score at discharg e

SSS – Scandinavian Stroke Scale

Medical data were obtained from the

Neurorehabilitation Unit Hospital Discharge

Summaries stored in digital form. After termination of

the study, information on falls as well as data pertaining

to the following parameters were analysed: patients’

gender, age, duration of hospital stay, diagnosis, type of

neurological deficit, medications used, and scores on

Neurologia i Neurochirurgia Polska 2007; 41, 1

Falls in post-stroke patients

Patients fell most often during such activities as

transferring from one place to another (26%) and during

changing position from sitting to standing and vice versa

(18%). Falls occurred relatively frequently in patients

sitting in a wheelchair (16%). A few falls took place

during standing and walking. In contrast, as many as

14% of the recorded falls were classified as falls from the

supine position. No falls were reported outside the ward,

for example during exercises in the gymnastic hall.

In 21 cases (62.4%), no injuries were observed. In

10 cases (15.6%), mild injuries like contusion of soft

tissues, bruises, or smaller wounds not requiring

surgical care were observed. Only in one case (2%) did

a fall result in a more severe injury – an incised wound

requiring application of a suture.

Table 2 compares patient subgroups with and

without falls. The fallers did not differ significantly from

non-fallers with regard to gender, age, time from disease

onset, type of past stroke, side of the paresis, presence of

sensory deficits or duration of hospital stay. Similarly, no

significant differences were observed in the frequency of

use of medication proven to affect the risk of falls in post-

stroke patients and in the general population.

The group of fallers differed from non-fallers in

the following features: the fallers had poorer scores,

both at baseline and at study completion, in the scales

assessing independence in activities of daily living –

the Rankin scale and the Barthel index – and a greater

neurological deficit assessed in the Scandinavian

Stroke Scale. The risk of a fall showed strongest

association with a score below 15/20 points in the

Barthel index, a score above 3 in the Rankin scale, low

score (below 39 points) in the Scandinavian Stroke

Scale, and with the presence of the unilateral neglect

syndrome. Detailed results are listed in Table 3.

Significant correlations were found between falls and

scores in the Rankin scale (R=0.23), the Barthel

index (R=- 0.22), and the Scandinavian Stroke Scale

(R=- 0.17), as well as between falls and presence of

the neglect syndrome (R=0.19). No significant

differences were found between the subgroup of

patients who experienced a single fall and the

subgroup of multiple fallers.

1 2 3 4 5 6 7

weeks

Fig. 1. Relationship between the incidence rate of falls and the duration

of hospital stay

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

hours

Fig. 2. Circadian distribution of falls

process of rehabilitation [1,9,22]. Awareness of the

fact that the majority of injuries and fractures in this

group of patients are a consequence of falls, that

femoral neck fracture in post-stroke patients is

associated with one-year mortality of 30%, and that

only 25% of patients regain their pre-fracture agility

[1] requires that the problem of falls and their

prevention should be treated with due seriousness.

The dimension of the phenomenon in relation to

patients is illustrated by the percent of patients who fall

– 10% in the present study. The results of our study do

not differ significantly from those obtained by the

majority of authors [3,10,12,18,23,24], where the

number of fallers ranged between 8.7 and 14%;

however, they do differ from the results observed by

Nyberg (36% and 39% of fallers) [12]. Yet, in the

study by Nyberg [12], mean duration of patients’

observation (48 days) was markedly longer than that in

our study (28 days). In the studies with enrolment of

patients discharged home from hospital, an even

higher (up to 73%) percentage of fallers is observed

during a longer time period after the occurrence of

stroke – during a half-year observation period [5].

Discussion

Falls are a frequent phenomenon among patients

undergoing post-stroke rehabilitation. They are also a

source of further complications that can disturb the

Neurologia i Neurochirurgia Polska 2007; 41, 1

Anna Czernuszenko

Table 2. Comparison between the groups of non-fallers and fallers

NON-FALLERS

FALLERS

N=318

N=35

gender

177

55.7%

62.9%

0.52

141

44.3%

37.1%

age [years] mean (SD)*

61 (15)

66 (10)

0.09

duration of stay [days] mean (SD)

27.8 (13.3)

31.5 (13.5)

0.07

time since disease onset [days] median (range)

42.5 (5-1592)

36.5 (10-366)

0.4

type of stroke

ischaemic

271

85.2%

82.9%

0.7

haemorrhagic

14.2%

14.3%

0.82

not identified

0.6%

2.8%

0.69

side of paresis

right

156

49.1%

37.1%

0.23

left

135

42.5%

54.3%

0.26

both

5.3%

8.6%

0.85

none

2.5%

0.73

dysphasia

120

37.7%

52.4%

0.58

unilateral neglect

8.5%

28.6%

0.0007

sensory deficit

17.0%

28.6%

0.14

current medication

antidepressants

28.6%

31.4%

0.88

diuretics

28.6%

20.0%

0.38

hypotensives

228

71.7%

71.4%

0.87

sedatives

11.6%

17.1%

0.5

anticonvulsants

10.1%

5.7%

0.6

neuroleptics

4.1%

5.7%

0.99

antiparkinsonian agents

0.9%

0.0%

0.7

Rankin 0 mean (SD)

3.2 (1.17)

4.1 (0.73)

<0.0001

Rankin 1 mean (SD)

2.6 (1.18)

3.3 (0.91)

0.0002

Barthel 0 mean (SD)

11.8 (5.9)

7.5 (4.1)

0.00005

Barthel 1 mean (SD)

14.8 (5.6)

12.2 (4.1)

0.0005

SSS 0 mean (SD)

37.6 (11.1)

30.3 (10.4)

0.002

SSS 1 mean (SD)

43 (10.7)

38.7 (9.3)

0.02

2 or the Fisher exact test was used. dependently on sample size. The Mann-Whitney U test was used for

calculation of differences between mean values. P values are given to present differences between the groups. For abbreviations and marks - see Table 1.

As approximately every fourth faller falls at least

twice, and subsequent falls constitute approximately

30% of all events, the percentage of fallers does not

illustrate the dimension of the phenomenon. The

incidence rate of falls, additionally normalized with

regard to the duration of observation, seems to be a

better measure in this case. Values of this index

reported in stroke and post-stroke rehabilitation wards

by other authors are greater than those observed based

on our data and are 8.9/1000 patient-days (6.2/1000

patient-days for the first and 17.9/1000 patient-days for

subsequent falls) according to Tutuarima [18] and

15.9/1000 patient-days in a study by Nyberg [12]. The

incidence rate of falls in our unit is even lower than the

Neurologia i Neurochirurgia Polska 2007; 41, 1

For the comparison of percent values between the groups of fallers and non-fallers the χ

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: