!"#$

Disability & Rehabilitation, 2012; 34(23): 1943–1949© 2012 Informa UK, Ltd.ISSN 0963-8288 print/ISSN 1464-5165 onlineDOI: 10.3109/09638288.2012.665131

Purpose: The purpose of this review of the scientific literature was to investigate the incidence and prevalence of hemiplegia with lower limb amputation, and to identify outcomes following the dual disability of hemiplegia and amputation. Methods: Electronic searching of the literature identified major studies examining the e!ects of hemiplegia on rehabilitation following amputation. Data were extracted and levels of evidence assigned for each subtopic area. Results: The summary conclusions are Level 4 evidence. The prevalence of amputation and hemiplegia is 8–18% and amputation and hemiplegia occur most often in the same leg. Once individuals with hemiplegia and lower limb amputation are selected for prosthetic rehabilitation, rate of successful functional ambulation is greater than 58%. In general there is a lower rate of prosthetic success and independence with hemiplegia than without. Predictive factors associated with success include less severe hemiplegia, laterality of hemiplegia (ipsilateral and right side), transtibial level of amputation and absence of impaired mental function. There is wide variation in length of hospital stay, but a specialty multidisciplinary team reduces length of stay. Conclusions: Patients with dual disability of hemiplegia and amputation generally benefit from a prosthetic rehabilitation program. Further study on predictive factors for outcome would be beneficial.

Keywords: Ambulation, amputation, hemiplegia, outcome, prosthetic fitting, stroke

Introduction

%e majority of persons with lower limb amputation are older adults with dysvascular disease, commonly with the combination of diabetes mellitus and peripheral atherosclerotic disease [1]. Typically these patients have many risk factors that are shared with other vascular diseases including poorly controlled diabetes, smoking, hypertension and hypercholesterolemia. %erefore, many people with dysvascular amputation of the lower limb have signi&cant cardiac disease, peripheral arterial disease and cerebrovascular disease (stroke) leading to additional impairments other than that caused by the amputation alone. Similarly, a stroke may precede an amputation for the same reasons.

REVIEW

Comorbidities in amputation: a systematic review of hemiplegia and lower limb amputation

Jacqueline S. Hebert1,2, Michael W. C. Payne3,6, Dalton L. Wolfe3,4,5, A. Barry Deathe3,6 & Michael Devlin7,8

1Division of Physical Medicine and Rehabilitation, University of Alberta, Edmonton, Canada, 2Adult Amputee Program, Glenrose Rehabilitation Hospital, Edmonton, Canada, 3Department of Physical Medicine and Rehabilitation, University of Western Ontario, London, Canada, 4Aging, Rehabilitation and Geriatric Care Program, Lawson Health Research Institute, London, Canada, 5Bachelor of Health Sciences Program, University of Western Ontario, London, Canada, 6Southwestern Ontario Regional Amputee Program, Parkwood Hospital, SJHC, London, Canada, 7Division of Physiatry, Department of Medicine, University of Toronto, Toronto, Canada, and 8Amputee Rehabilitation Services, West Park Healthcare Centre, Toronto, Canada

Correspondence: Dr. Jacqueline S. Hebert, Glenrose Rehabilitation Hospital, 10230 111 Avenue, Edmonton, Alberta T5G 0B7, Canada. Tel: (780) 735-8218. Fax: (780) 735-6018. Email: Jacqueline.hebert@albertahealthservices.ca

%e prevalence of hemiplegia with lower limb ampu-tation ranges from 8 to 18%, most frequently a'ecting the same leg.%e majority of patients attain successful functional levels of ambulation with prosthetic rehabilitation, although lower rates than nonhemiplegic patients.Predictive factors associated with greater success include less severe hemiplegia, ipsilateral hemiple-gia, transtibial level of amputation and absence of impaired mental function.

Implications for Rehabilitation

(Accepted February 2012)

Disa

bil R

ehab

il D

ownl

oade

d fro

m in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of A

lber

ta o

n 02

/27/

13Fo

r per

sona

l use

onl

y.

!"## J. S. Hebert et al.

Disability & Rehabilitation

Both stroke and lower extremity amputation have a sig-ni&cant impact on many aspects of a patient’s life including activities of daily living (ADL), body image and sense of self, psychosocial health, vocational and avocational activities and mobility. Rehabilitation of individuals with a single diagno-sis of either hemiplegia secondary to stroke or major lower extremity amputation is directed at addressing the impair-ments to restore function. Few studies have examined the speci&c dual-diagnosis combination of both hemiplegia and amputation together from an epidemiological, rehabilitation or outcomes perspective. When a person has had a stroke or other cerebral insults resulting in hemiplegia the prosthetic rehabilitation training is made more di(cult since, in patients with hemiplegia, the ability to relearn ambulation is already greatly diminished [2].

At face value, one might accept that the combination of hemiplegia and amputation would produce worse outcomes, but we are unaware of the magnitude of the problem or the extent of disability among those with both diagnoses, and how one diagnosis may impact the other.

%e purpose of this review of the scienti&c literature was to investigate the incidence and prevalence of hemiplegia with lower limb amputation, and to identify the range of prosthetic and mobility outcomes following the dual dis-ability of hemiplegia and amputation. Understanding the burden of this dual disability may help with rehabilitation planning and prognosticating with respect to outcome for these patients.

%e speci&c objectives for this review were

i. to identify studies that evaluate the incidence and/or prevalence of hemiplegia with lower limb amputation;

ii. to determine the e'ect of hemiplegia on prosthetic &tting and functional ambulation;

iii. to describe the predictive factors of functional mobility and related outcomes in persons with the dual disability of hemiplegia and lower limb amputation; and

iv. to describe the impact of dual disability on hospital length of stay.

Methods

We reviewed studies examining the e'ects of hemiplegia on rehabilitation for major lower extremity amputations. Electronic searching of several bibliographic databases (i.e. Embase, Medline, PsychInfo, CINAHL, Cochrane) from 1980 to May 2011 using the search terms “lower limb amputa-tion, hemiplegia, stroke, cerebrovascular accident (CVA), complications, comorbidity, mobility and amputation reha-bilitation” and subsequent hand searching of reference lists from relevant articles identi&ed 43 articles addressing the presence of hemiplegia and lower limb amputation. Twelve of these articles were selected for inclusion and these studies included patients with various levels of amputation, etiolo-gies of amputation and severity of impairments. Studies were excluded if (i) they had a heterogeneous population which did not allow identi&cation of the speci&c impact of amputation and stroke (for example several diagnostic categories included

in the study population); (ii) they were case reports (n < 3) not allowing generalization of results; (iii) they reported only on upper extremity amputations, as mobility and lower limb amputation was the focus of this review; and (iv) if stroke or amputation was not a major focus.

For each of the selected articles, data were extracted and tabulated for clinically relevant issues or interventions (sub-topic areas) that were reported in at least three studies. Levels of evidence were assigned to the sum total of the evidence for each subtopic area using a modi&ed approach to that outlined by the Oxford Centre for Evidence-based Medicine [3] and employed in a similar evidence-based review of spinal cord injury rehabilitation practice [4]. In this approach, individual papers were assigned a level of evidence based on their study design (Table I) and an overall level of evidence was judged based on the highest level within the speci&c group of articles under consideration (see summary conclusions for each sec-tion). If at least 75% of the papers within a particular subtopic area did not show agreement in their &ndings, the level of evidence was deemed to be con)icting, although the articles with the highest quality of evidence were given more weight in this consideration. Individual study quality was assessed using the Downs and Black scale [5] for nonexperimental trials involving cohort, case control, case series, prepost and cross-sectional study designs; as there were no Level 1 or Level 2 trials found, the Physiotherapy Evidence Database (PEDro) scale [6] for experimental trials (randomized or nonrandomized controlled trials) was not appropriate in this review.

Results

%e results of this review are presented in four sections. %e &rst focuses on examining the reported incidence and/or prevalence of hemiplegia in persons with lower limb ampu-tation. %e second examines the e'ects of hemiplegia on prosthetic &tting and ambulation, while the third focuses on the predictive factors of ambulation and other functional out-comes (such as independence in ADL) in persons with the dual disability of hemiplegia and amputation. %e &nal sec-tion describes the impact of dual disability on length of hos-pital stay (LOS). In all cases, levels of evidence were assigned as Level 4 as ten of eleven studies meeting selection criteria involved case series study designs. In addition there was 1 prospective cohort study design [7] but due to methodology it was assigned as Level 4 evidence.

Prevalence%e prevalence of hemiplegia at the time of rehabilitation in patients with lower limb amputation ranged from 8% to 18% (Table II). In the current review, the prevalence rate was determined only for those studies where the baseline population of individuals undergoing lower limb amputation was provided [7–12]. No studies provided details regarding neither the incidence of amputation among those with hemiplegia nor the rate of hemiplegia occurring in those with amputation; therefore, conclusions on incidence could not be made.

Disa

bil R

ehab

il D

ownl

oade

d fro

m in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of A

lber

ta o

n 02

/27/

13Fo

r per

sona

l use

onl

y.

Hemiplegia and lower limb amputation !"#*

© Informa UK, Ltd.

Table I. Descriptions of study designs and levels of evidence (modi$ed from Strauss et al. 2005) [3].Level Research design DescriptionLevel 1 Randomized controlled

trial (RCT)Randomized controlled trial, PEDro score + 6. Includes within subjects comparison with randomized conditions and cross-over designs

Level 2 RCT Randomized controlled trial, PEDro score <6.Prospective controlled trial Prospective controlled trial (not randomized)Cohort Prospective longitudinal study using at least two similar groups with one exposed to a particular condition.

Level 3 Case control A retrospective study comparing conditions, including historical controlsLevel 4 Prepost A prospective trial with a baseline measure, intervention, and a post-test using a single group of subjects.

Post-test A prospective post-test with two or more groups – intervention, then post-test (no pretest or baseline measurement) using a single group of subjects.

Case series A retrospective study usually collecting variables from a chart review.Poor quality cohort and case series

A study that failed to adequately achieve any of the following: clearly de$ne comparison groups; measure exposure or outcome in the same (preferably blinded) way;

identify or control for known confounders; conduct su%ciently long or complete follow-up of study participants.

Level 5 Observational Study using cross-sectional analysis to interpret relations.Clinical consensus Expert opinion without explicit critical appraisal, or based on physiology, biomechanics or “$rst

principles.”Case report Prepost or case series involving one subject

PEDro, Physiotherapy Evidence Database.

Table II. Rate of hemiplegia and association of clinical features between side of hemiplegia and amputation.

Study Prevalence of dual disabilityLateralitya (ipsilateral vs contralateral)

Sequence ( hemiplegia then amputation)

Interval between hemiple-gia and amputation

Siriwardena 1991 [11]; UK; D&B = 18; Case Series

46/543 (8.5%) N/R N/R N/R

Altner 1987 [17]; USA; D&B = 9; Case Series

52/52b (14-year period) N/R 36/52 (69.2%) Mean 62.5 months

O’Connell 1989 [13]; USA; D&B = 12; Case Series

46/46b (7 of the 46 patients were simultaneous) (6-year period)

26 vs 13/39 (67% vs 33%) 29/39 (74.4%) N/R

Chiu 2000 [9]; Taiwan; D&B = 12; Case Series

23/264 (9%) 16 vs 7/23 (70% vs 30%) 15/23 (65.2%) Mean 4.9 months (range: 1 month to 27 years)

Neumann 1998 [7]; UK; D&B = 11; Cohort

19/194 (9.8%) over 12-month period

N/R N/R N/R

Garrison 1986 [14]; USA; D&B = 9; Case Series

22/ 22b (14-year period) 21 vs 1/22 (95% vs 5%) 22/22 (100%) Mean 43 months; range: 3–414 months (mean 32 months for diabetes mellitus patients – 55 months for non-diabetes mellitus)

Kullman 1987 [16]; Hungary; D&B = 8; Case Series

49/ 49b (2 cohorts) 21 vs 28/49 (43% vs 57%) 24/49 (49%) >1 year in 67% (33/49)

Badwey 1988 [2]; Italy; D&B = 8; Case Series

47/47b (all patients had hemiplegia $rst)

40 vs 7/47 (85% vs 15%) 47/47 N/R

Brunelli 2006 [8]; Italy; D&B = 19; Case Series

56/402 transfemoral and hemiplegia (14%)

25 vs 20/45 (56% vs 44%) 19/45 (42%) Median , 30 months 51% (23/45)>30 months 49% (22/45)

Varghese 1978 [15]; USA; D&B = 3; Case Series

30/30b 19 vs 10/29 (66% vs 34%) 15/29 (52%) N/R

Finch 1980 [10]; UK; D&B = 13; Case Series

14/133 (10.5%) N/R N/R N/R

Prvu-Bettger 2009 [12]; USA; D&B = 17; Cohort

864/4720 (18.3%) N/R 864/864 (100%) N/R

N/R, Not reported.aLaterality: Side of amputation relative to hemiplegia.bNo baseline population reported to estimate the frequency of hemiplegia.

Disa

bil R

ehab

il D

ownl

oade

d fro

m in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of A

lber

ta o

n 02

/27/

13Fo

r per

sona

l use

onl

y.

!"#- J. S. Hebert et al.

Disability & Rehabilitation

Overall, six studies [2,8,9,13–15] indicated a strong rela-tionship between the side of amputation and the side a'ected by hemiplegia (described as laterality: ipsilateral or contralat-eral). Ipsilateral side (amputation on same side of hemiplegia) was predominant, with a high frequency rate ranging from 56% to 95%. However, one study [16] reported the opposite relationship with a frequency rate of 57% for amputation on the contralateral side from the hemiplegic side.

%ere was an inconsistent relationship regarding the sequence of disability. %e studies conducted by Garrison and Badwey selected a cohort of patients with hemiplegia and followed them for onset of amputation and therefore do not assist with answering the question of which disability is more likely to occur &rst [2,14]. Of the studies that examined sequence, three studies reported a majority of subjects with onset of hemiplegia prior to amputation [9,13,17]. However, Kullman and Varghese found an almost even chance of either disability occurring &rst [15,16], while Brunelli reported the opposite relationship, with amputation occurring &rst in the majority (58%) of the subjects (see Table II) [8]. In these stud-ies, selection bias was not reported to be an issue, since the occasion of dual disability was searched from general reha-bilitation records, although di'erences between institutional admission criteria might still result in selection bias.

%e timing between onsets of each disability varied widely among the studies. Only Garrison, who studied patients with prior stroke currently undergoing rehabilitation for amputa-tion, reported a signi&cantly shorter duration between onset of hemiplegia to amputation for those with diabetes mellitus versus those without by nearly 2 years (see Table II) [14].

Prevalence – summary evidence statements%ere is Level 4 evidence that the frequency of concomi-tant lower limb amputation and hemiplegia is 8–18% although the relative onset of each condition varies widely, with con)icting evidence that hemiplegia tends to precede amputation.%ere is Level 4 evidence that amputation and hemiplegia occur most o.en in the same leg with a frequency rang-ing from 56–95%.

Prosthetic fitting and ambulation%e current review identi&ed eight studies reporting pros-thetic &tting and nine studies recording levels of ambulation in patients with dual disability (Table III).

Determining the rate of prosthetic &tting and ambula-tion outcome a.er dual disability is confounded by selection bias toward only those patients actually referred for pros-thetic &tting or rehabilitation. Neumann examined amputee patients referred for prosthetic rehabilitation versus the group of amputees not referred, and found more amputees with hemiplegia (19.3%) in the nonreferral group compared with their referral group (10.8%) [7]. Since those patients with dual disability seemed to be less likely to be referred to rehabilita-tion, it follows that this group was deemed to be less likely to bene&t from prosthetic &tting [7].

Few studies were found that clearly reported the rate of successful prosthetic &tting in this population. For most of

the studies the percentage of patients &t with a prosthesis was implied through the reporting of ambulation levels. We therefore attempted to estimate the rate of prosthetic &t from the total number of patients who were reported as having an ambulatory level with a prosthesis and/or with assistive devices (Table III), over the total number of patients reported as having dual disability as the baseline (Table II). Using this method to determine the rate of &tting allowed a comparison between di'erent studies.

Our analysis indicated anywhere from 10 to 100% of patients that were referred for rehabilitation were &t with a prosthesis, with &ve of seven studies for which this could be determined showing greater than 58% &t rate. Of the patients who were &t with a prosthesis, most of the studies demonstrated that patients with dual disability had reasonable ambulatory outcome following their prosthetic &tting, with nonambulatory rates of only 20–30% and the remainder being reported as indoor or community ambulators. %is would seem to indicate that most patients with dual disability who are referred for rehabilitation could bene&t from a prosthetic rehabilitation program (Table III).

%e studies by Neumann and Chiu also suggested that there is a reasonable likelihood for hemiplegic amputees referred for prosthetic rehabilitation and successfully &tted with a prosthesis to maintain use of the prosthesis over the &rst 12 months [7,9].

Prosthetic fitting and ambulation – summary evidence statements

%ere is Level 4 evidence that once individuals with hemiplegia and lower limb amputation are selected for prosthetic rehabilitation the &t rates and potential for achieving some level of functional ambulation are good, with most studies showing greater than a 58% success rate.

Predictive factors%e presence of hemiplegia was found to be a negative predic-tive factor overall for ambulation status [11] (see Table III) and also for prosthetic use following amputation as reported by Pohjolainen and Alaranta [18].

%e time interval between the onset of the two disabilities and the sequence of disability has also been shown to play an important role in predicting rehabilitation outcome. %ere appears to be an advantage in rehabilitating from one condi-tion &rst rather than coping with both severe conditions at the same time; i.e. a longer time period between each subsequent disability may predict better outcome [16]. Varghese stated that the amputees who had been &tted with their prosthesis prior to having a stroke had better ambulatory function [15]. O’Connell also demonstrated that patients who were ambulat-ing prior to the second disability were more likely to achieve independent mobility than those who failed to ambulate a.er the &rst disability [13].

%e most predominant clinical factors in)uencing reha-bilitation outcomes (mobility and ADL independence) in the hemiplegic amputees were: severity of hemiparesis, laterality of hemiparesis and transtibial level of amputation [8,9,13,15].

Disa

bil R

ehab

il D

ownl

oade

d fro

m in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of A

lber

ta o

n 02

/27/

13Fo

r per

sona

l use

onl

y.

Hemiplegia and lower limb amputation !"#/

© Informa UK, Ltd.

Tabl

e II

I. Pr

osth

etic

$tti

ng, l

evel

of a

mbu

latio

n an

d pr

edic

tive

fact

ors i

n pa

tient

s with

am

puta

tion

and

hem

iple

gia.

Stud

y

Initi

al sa

mpl

e re

ferr

ed fo

r re

habi

litat

ion

% $

t with

a

pros

thes

is

Leve

l of a

mbu

latio

n in

thos

e $t

with

a p

rost

hesis

Follo

w-u

p

Pred

ictiv

e fa

ctor

sIn

depe

nden

t or

com

mun

ityLi

mite

d

(indo

or)

Non

ambu

lato

rPo

sitiv

eN

egat

ive

Altn

er 1

986

[17]

; USA

5258

% (3

0/52

)27

% (8

/30)

53%

(16/

30)

20%

(6/3

0)N

/RN

/RN

/RO

’Con

nell

1989

[13]

; U

SA46

37%

(17/

46)

18%

(3/1

7)53

% (9

/17)

29%

(5/1

7)2–

120

mon

ths

– M

ilder

hem

iple

gia

– In

cont

inen

ce–

Tran

stib

ial a

mpu

tatio

nPr

ior a

mbu

latio

nC

hiu

2000

[9];

Taiw

an23

100%

(23/

23)

43.5

% (1

0/23

)21

.7%

(5/2

3)34

.8%

(8/2

3)At

disc

harg

e or

th

e la

st cl

inic

al

visit

– M

ild m

otor

invo

lvem

ent,

– Im

paire

d m

enta

l sta

tus

– Tr

anst

ibia

l am

puta

tion,

– A

mpu

tatio

n pe

rfor

med

bef

ore

CVA

,–

Age

<60

yea

rs (N

S)N

eum

an 1

998

[7];

UK

2382

.6%

(19/

23)

89.5

% (1

7/19

)aN

/R6–

12 m

onth

sN

RN

/RKu

llman

198

7 [1

6];

Hun

gary

4959

%59

%b (

29/4

9)35

%c (

17/4

9)N

/R–

Righ

t ips

ilate

ral

– Ci

rcul

atio

n–

Tran

stib

ial a

mpu

tatio

n le

vel

– C

hron

ic il

lnes

s–

Loco

mot

ion

– Ps

ychi

atric

diso

rder

s–

Sens

ory

(blin

dnes

s)Br

unel

li 20

06 [8

]; Ita

ly45

100%

(45/

45)

100%

d (45

/45)

N/R

At d

ischa

rge

– Ip

silat

eral

impa

irmen

tN

/R–

Mild

hem

iple

gia

Varg

hese

197

8 [1

5];

USA

3010

% (3

/29)

34%

(10/

29)

55%

(16/

29)

N/R

N/R

– A

mpu

tatio

n $r

st th

en

hem

iple

gia

– A

ge

– Ri

ght I

psila

tera

l–

Tran

stib

ial a

mpu

tatio

nSi

riwar

dena

199

1 [1

1];

UK

N/R

N/R

WA

I 0.6

wor

se in

thos

e w

ith h

emip

legi

a ve

rsus

no

hem

iple

gia

12 m

onth

sN

RN

/R

N/R

, Not

repo

rted

; WA

I, W

alki

ng A

bilit

y In

dex.

a Mob

ility

usin

g a

pros

thes

is w

as d

e$ne

d as

>30

m (9

), 11

–30

m (2

) and

<10

m (6

). &

e to

tal %

com

bine

s the

se th

ree

cate

gorie

s of d

istan

ces.

b Ind

epen

dent

wal

king

(3) o

r with

ass

istiv

e de

vice

(26)

.c W

heel

chai

r bou

nd (1

4) o

r non

ambu

lato

ry (3

).d M

obili

ty u

sing

a pr

osth

esis:

2/4

5 in

depe

nden

t wal

king

, 43/

45 w

ith a

ssist

ive

devi

ce.

Disa

bil R

ehab

il D

ownl

oade

d fro

m in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of A

lber

ta o

n 02

/27/

13Fo

r per

sona

l use

onl

y.

!"#0 J. S. Hebert et al.

Disability & Rehabilitation

%ese studies demonstrated that patients with a mild hemi-paresis, ipsilateral disability and a history of ambulation prior to their disability were statistically more likely to regain the ability to ambulate independently or attain independence in ADLs 8,13. Furthermore, Varghese concluded that the patients with right hemiplegia and right limb amputation attained a higher functional status than those who had le. hemiplegia and le. limb amputation [15]. Kullman also found better outcomes in those with right-sided hemiplegia [16]. Other predictive factors found in these studies were age [9], mental status (including alertness, orientation and learn-ing ability) [9] psychological and psychosocial status [16] and other comorbidities (Table III).

Inconsistent reporting of functional outcomes relating to ADL independence a.er dual disability limits conclusions. In O’Connell’s study, independence in ADLs was reported in 39% of patients which is consistent with the number reported as being &t with a prosthesis (at baseline all patients had dual disability) [13]. Lower numbers were reported by Varghese with only 26% (10/39) of those with hemiplegia and amputa-tion independent in ADLs (Table IV) [15]. Other studies had insu(cient data reported to determine rates of functional independence.

Predictive factors – summary evidence statements%ere is Level 4 evidence that the e'ect of the dual dis-ability of hemiplegia and lower limb amputation results in a lower rate of prosthetic success and independence in ADLs.

Predictive factors associated with greater success include less severe hemiplegia, laterality of hemiplegia (ipsilat-eral and right side), transtibial level of amputation and absence of impaired mental function.

Length of hospital stayLength of hospital stay (LOS) was reported in four studies, however the reporting of LOS within this population was inconsistent (Table V) [8,13,15,16].

Vargehese noted that the LOS of patients with dual disabil-ity ranged from 4 months to 1 year [15], whereas O’Connell reported an initial rehabilitation LOS of approximately 33 days [13]. O’Connell also noted that the majority of patients had either ongoing outpatient rehabilitation or a subsequent inpatient stay (Table IV) [13].

Brunelli found that functional outcomes such as indepen-dence in ADLs and ambulation did not di'er when comparing the length of stay between patients whose length of stay was less than 60 days and those more than 60 days [8]. Kullman reported that hemiplegic amputees treated by a specialized multidisciplinary team had a shorter LOS than those in an earlier cohort without a specialized program (average 1.7 months vs 5.2 months, respectively) [16].

Length of hospital stay – summary evidence statements%ere is con)icting Level 4 evidence about the LOS for patients with dual disability (hemiplegia and lower limb amputation) with wide variation in LOS ranging from 1 month to 1 year with uncertainty over whether better outcomes were obtained with longer LOS.

Table IV. Functional outcomes in patients with dual disability.Study Independence ADLs Functional independence ICF – Participationa

O’Connell 1989 [13]; USA 18/46 (39%) – 6/8 (75%) outpatient therapy – 2/46 (4.3%) returned to work– 10/17 (59%) inpatient therapy

Neuman 1998 [7]; UK N/R – BI = 15.5 at admission – Restricted lifestyle 3 months prior to amputation (FAI = 11, median)

– BI = 17.1 at discharge – Improved lifestyle a'er receiving a prosthesis (FAI = 11.5 at 12 months)

Kullman 1987 [16]; Hungary

N/R N/R – 14/30 participating and satisfactory psychosocial status;– 16/30 reported poor QOL

Brunelli 2006 [8]; Italy N/R – BI better for mild hemiplegia and younger age at discharge

N/R

Varghese 1978 [15]; USA 10/29 fully indepen-dent

N/R N/R

ADL, activities of daily living; BI, Barthel Index; FAI, Frenchay Activity Index; N/R, not reported; QOL, quality of life.aICF – Participation: Outcomes associated with participation under the International Classi$cation of Functioning, Disability and Health.

Table V. Length of hospital stay for patients with dual disability.Studies Sample size Length of stay Di(erences in functional outcomesO’Connell 1989 [13]; USA

46/46 Mean 33 days (Rehabilitation LOS) N/R

Kullman 1987 [16]; Hungary

49/49 49.6 days treated by multidisciplinary team (mean 1.7 months) N/R124.4 days not treated by multidisciplinary team (mean 5.2 months)

Brunelli 2006 [8]; Italy

45/45 Median 60 days No di(erences were found between patients who stayed <60 days or >60 days

Varghese 1978 [15]; USA

30/30 Range: 4 months to 1 year N/R

LOS, length of hospital stay; N/R, Not reported.

Disa

bil R

ehab

il D

ownl

oade

d fro

m in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of A

lber

ta o

n 02

/27/

13Fo

r per

sona

l use

onl

y.

Hemiplegia and lower limb amputation !"#"

© Informa UK, Ltd.

%ere is Level 4 evidence associated with one study that introduction of a specialty multidisciplinary team resulted in a reduced LOS from 5.2 months to 1.7 months.

Conclusions

Patients with dual disability have a high enough prevalence in the rehabilitation population that most rehabilitation centers will see and be treating these patients on a regular basis. Patients with dual disability have been shown to bene&t from a prosthetic rehabilitation program, although the presence of hemiplegia does result in a lower rate of success in ambulation and independence. Sequence of onset, laterality of disability, severity of hemiplegia, mental function and level of amputation were the most predominant factors in)uencing ambulation outcome in the rehabilitation of patients with dual disability. Having the amputation &rst, and being an ambulator prior to onset of ipsilateral stroke, appears to have the best outcome in dual disability.

%ere exists a de&nite selection bias with centre based studies in that the probable exclusion of a subset of patients that are never referred for rehabilitation a.er their second disability leads to the likelihood of an excessively positive outlook on potential outcome. Future studies should focus on surveys of patients in acute care both on stroke wards and on amputation units to determine true incidence and outcome for the broader population with dual disability.

Further study on predictive factors for outcome following prosthetic rehabilitation would also be useful; particularly to compare those populations that are not referred for rehabili-tation versus those that are, as well as to compare the group that is successful with prosthetic use for ambulation versus not successful with respect to independence and ambulation. Finally, more consistent reporting of length of hospital stay in those with dual disability compared with amputation alone may shed light on the additional impact and resource implica-tions of dual disability on prosthetic rehabilitation programs.

Declaration of Interest: %e authors declare no con)icts of interest. %e authors alone are responsible for the content and writing of the paper.

References 1. Dillingham TR, Pezzin LE, MacKenzie EJ. Limb amputation and limb

de$ciency: epidemiology and recent trends in the United States. South Med J 2002;95:875–883.

2. Badwey TM, Rice JC, Kerstein MD. Amputation as a consequence of stroke. J Cardiovasc Surg (Torino) 1988;29:563–566.

3. Strauss SE, Richardson WS, Glaszious P, Haynes RB. Evidence-based medicine: how to practice and teach EBM. 3rd ed. Edinburgh: Churchill Livingstone; 2005.

4. Eng JJ, Teasell RW, Miller WC, Wolfe DL, Townson AF, Aubut J et al. Spinal cord injury rehabilitation evidence: Method of the SCIRE sys-tematic review. Top Spinal Cord Inj Rehabil 2007;13:1–10.

5. Downs SH, Black N. &e feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 1998;52:377–384.

6. Sherrington C, Herbert RD, Maher CG, Moseley AM. PEDro. A data-base of randomized trials and systematic reviews in physiotherapy. Man &er 2000;5:223–226.

7. Neumann VC, Cotter DH, Geddes JM, Waxman R. &e in)uence of prior stroke on the prosthetic rehabilitation of lower limb amputees. Prosthet Orthot Int 1998;22:102–106.

8. Brunelli S, Averna T, Porcacchia P, Paolucci S, Di Meo F, Traballesi M. Functional status and factors in)uencing the rehabilitation outcome of people a(ected by above-knee amputation and hemiparesis. Arch Phys Med Rehabil 2006;87:995–1000.

9. Chiu CC, Chen CE, Wang TG, Lin MC, Lien IN. In)uencing factors and ambulation outcome in patients with dual disabilities of hemiple-gia and amputation. Arch Phys Med Rehabil 2000;81:14–17.

10. Finch DR, Macdougal M, Tibbs DJ, Morris PJ. Amputation for vas-cular disease: the experience of a peripheral vascular unit. Br J Surg 1980;67:233–237.

11. Siriwardena GJ, Bertrand PV. Factors in)uencing rehabilitation of arte-riosclerotic lower limb amputees. J Rehabil Res Dev 1991;28:35–44.

12. Prvu-Bettger JA, Bates BE, Bidelspach DE, Stineman MG. Short- and long-term prognosis among veterans with neurological disorders and subsequent lower-extremity amputation. Neuroepidemiology 2009;32:4–10.

13. OConnell PG, Gnatz S. Hemiplegia and amputation: rehabilitation in the dual disability. Arch Phys Med Rehabil 1989;70:451–454.

14. Garrison JH, Shankara B, Mueller MJ. Stroke hemiplegia and subse-quent lower extremity amputation: which side is at risk? Arch Phys Med Rehabil 1986;67:187–189.

15. Varghese G, Hinterbuchner C, Mondall P, Sakuma J. Rehabilitation outcome of patients with dual disability of hemiplegia and amputa-tion. Arch Phys Med Rehabil 1978;59:121–123.

16. Kullmann L. Evaluation of disability and of results of rehabilitation with the use of the Barthel index and Russek’s classi$cation. Int Disabil Stud 1987;9:68–71.

17. Altner PC, Rockley P, Kirby K. Hemiplegia and lower extremity ampu-tation: double disability. Arch Phys Med Rehabil 1987;68:378–379.

18. Pohjolainen T, Alaranta H. Predictive factors of functional ability a'er lower-limb amputation. Ann Chir Gynaecol 1991;80:36–39.

Disa

bil R

ehab

il D

ownl

oade

d fro

m in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of A

lber

ta o

n 02

/27/

13Fo

r per

sona

l use

onl

y.