Evaluación de Tos
Evaluación de Tos
Evaluación de Tos
[Original]
Differences in the Peak Cough Flow among Stroke Patients With and Without
Dysphagia
1
Department of Rehabilitation, Kyushu Nutrition Welfare University, Japan. Kokuraminami-ku, Kitakyushu 800-0298,
Japan
2
Department of Rehabilitation Medicine, School of Medicine, University of Occupational and Environmental Health,
Japan. Yahatanishi-ku, Kitakyushu 807-8555, Japan
Abstract : Coughing is an important protective mechanism for keeping the airway clear, and adequate voluntary
coughing reduces the risk of aspiration in patients with deglutition disorders. The purpose of this study was to com-
pare the peak cough f low (PCF) of stroke patients with and without dysphagia and to identify the physical and re-
spiratory determinants of PCF. Using a spirometer, we measured and compared the PCFs of 10 stroke patients with
dysphagia (SPD), 20 stroke patients without dysphagia (SP) and 10 gender and age matched healthy controls (HC)
recruited by using a notice at a clinic and in newspapers. The PCF of the SPD (mean ± SD, 160.1 ± 68.7 l/min) was
significantly lower than that of the SP and HC (297.2 ± 114.2 l/min and 462.0 ± 84.4 l/min, respectively; one-way
ANOVA, Scheffe’s test, P < 0.05). The vital capacity (VC) and inspiratory reserve volume (IRV) of the SPD were
lower than those of the HC. Stepwise multivariate regression analysis revealed that IRV and ambulation function
(Functional Ambulation Categories, FAC) contributed 50% and 17% to the variance of PCF (P < 0.05), respectively.
It is suggested that respiratory function, especially IRV, is important for maintaining PCF in SPD.
Key words : deglutition disorders, dysphagia, stroke, cough, inspiratory reserve volume.
Corresponding Author: Yoshiko Kimura RPT, PhD, Department of Rehabilitation, Kyushu Nutrition Welfare University, 1-5-1, Kuzuharatakamatsu,
Kokuraminami-ku, Kitakyushu 800-0298, Japan, Tel: + 81-93-471-7912, Fax: +81-93-473-4161, E-mail: Kimura@krc.ac.jp
10 Y Kimura et al
of the predictive factors for aspiration pneumonia. As previously reported to be 98%, and it had the advantage
the first step in a study on the prevention of aspiration of being simple, safe and inexpensive [8]. The high
pneumonia in SPD by physical therapies, we measured sensitivity of these non-VFG tests means that they have
the PCF in stroke patients with and without dysphagia a low rate of false negatives, so we suspected dyspha-
to reveal whether SPD have a lower PCF than stroke gia if the results of any one of these tests was judged to
patients without dysphagia (SP) and healthy controls be abnormal, or if coughing occurred during swallow-
(HC). As our final goal is to investigate effective re- ing, if there was difficulty with eating, if there was wet
habilitative approaches to prevent aspiration in stroke hoarse dysphonia after deglutition, or if the subject had
patients, the second purpose of this study is to find any a past history of suffocation, we suspected dysphagia.
factors related to the volume of PCF, including respira- Finally, 10 stroke patients suspected to have dysphagia
tory functions. underwent VFG to confirm the dysphagia.
Healthy men were recruited as HC by using a notice
Patients and Methods at the clinic and in newspapers. The inclusion criteria
for a control were as follows: 1) had no medical his-
The study subjects consisted of 45 consecutive male tory of stroke or dysphagia, and the same as 2) ~ 7) of
stroke in-patients who were referred to the department the inclusion criteria for the stroke patients. Finally,
of rehabilitation for further rehabilitative treatments. 10 men were adopted as HC.
The inclusion criteria for this study were as follows: Approval for this study by the ethics committee.
1) the patient was diagnosed with cerebral hemorrhage
or infarction by magnetic resonance imaging and com- Degree of penetration aspiration scale of SPDs
puted tomography of the head; 2) was male; 3) was A physiatrist (M.T.) at the dysphagia clinic measured
55~86 years old; 4) had no history of cardiopulmo- the degree of penetration aspiration scale (P/A) from
nary diseases; 5) had no physical disabilities induced VFG. This scale is one of the most standard scales for
by musculo-skeletal disorders; 6) could understand the the degree of dysphagia (Table 1) [9].
examination; 7) was able to undergo spirometry with a
maximum effort, sitting on a chair. Table 1. Penetration aspiration scale
Fifteen patients were excluded from this study. Eight Score Contrast P/A
could not perform the trial of spirometry with a maxi- 1 Contrast does not enter the airway No penetration
mum effort, five had a history of ischemic heart disease
2 Contrast enters the airway; remains above Penetration
or suspected chronic obstructive pulmonary disease, the vocal folds
and two had physical disabilities due to a degenerative 3 Contrast remains above the vocal folds with Penetration
joint disorder. Finally, 30 patients met the inclusion visible residue
criteria. They were divided into two groups according 4 Contrast contacts vocal folds; no residue Penetration
to their clinical signs and symptoms, 10 SPD patients 5 Contrast contacts vocal folds; visible residue Penetration
and 20 SP patients, by a physiatrist (M.T.) at the dys-
6 Contrast passes glottis; no subglottic residue Aspiration
phagia clinic. We recruited three screening tests for
7 Contrast passes glottis; visible subglottic Aspiration
dysphagia and videoflurography (VFG), if necessary. residue despite response
These screening tests consisted of the Repetitive Salvia 8 Contrast passes glottis; visible subglottic Aspiration
Swallowing test (RSST: a patient is asked to swallow residue; absence of response
his saliva as many times as possible in 30 seconds), the P/A: penetration aspiration scale (Reproduced from ref. [9] with
water swallowing test (WST: a patient is asked to swal- permission of CHEST (American College of Chest Physicians))
low 3 ml of water poured on his tongue), and the pud-
ding swallowing test (PST: a patient is asked to swal- Measurement of pulmonary function and PCF
low 4 g of pudding placed on his tongue). The summed One of the authors (Y.K.) carefully explained the
scores of the PST and WST had a sensitivity of 90% maneuver of the spirometer (Autospiro AS-505, Mi-
and a specificity of 56%. The sensitivity of RSST was nato Medical Science, Osaka) and measured pulmo-
Peak Cough Flow among Stroke Patients With and Without Dysphagia 11
nary functions including tidal volume (TV), vital the correlation between PCF and other variables: age,
capacity (VC), inspiratory reserve volume (IRV), ex- days since onset, pulmonary functional parameters,
piratory reserve volume (ERV), and PCF to all the sub- (VC, TV, IRV, ERV), and Spearmanʼs rank correla-
jects. A subject was asked to sit on a chair, wearing a tion coefficients to determine the correlation between
face mask instead of a mouthpiece attached to the spi- PCF and ambulatory function. Stepwise multivariate
rometer, to prevent leakage from the mouth and nose. regression analysis was then applied to determine the
The subject was instructed to breathe out, breathe influence of the independent variables on the PCF.
in deeply, and then cough as hard as possible under The independent variables were selected based on the
shouts of encouragement. First, they tried a few times Pearsonʼs correlation coefficients and Spearmanʼs rank
to get accustomed to the measurement and then were correlation coefficients with significance. A statistical
asked to make their maximal efforts for three times. software package (SPSS 17.0J, SPSS Japan, Tokyo)
PCF was defined as the highest point of the flow vol- was used for these analyses, and their level of signifi-
ume curve obtained during a cough, and a maximum cance was set at P < 0.05.
value of three measurements was used for analysis.
Results
Severity of hemiplegia and gait disturbance
The severity of hemiplegia was assessed with The characteristics of the subjects are shown in
Brunnstromʼs recovery stage [10] for upper and lower Table 2. There were no significant differences in age,
extremities, and the mode and median values were ob- height, weight or body mass index among the SPD,
tained for the SPD and SP groups. The ambulatory SP and HC groups (one-way ANOVA, P > 0.05), and
function was assessed based on function ambulation there was no significant difference between the SPD
categories (FAC), and was coded as follows for analy- and SP groups in the number of days since the onset of
sis: 0 for “ambulator-independent” and “ambulator-in- stroke between the SPD and SP groups (Student t-test,
dependent, level surfaces only”, 1 for “supervison” and P > 0.05). Although there were no significant differ-
“ambulatory-dependent for physical assistance-level ences in the severity of hemiplegia (Mann-Whitney U
I”, 2 for “ambulator-dependent for physical assistance- test, P > 0.05), the gait disturbance of the SPD group
level II,” and 3 for “non-functional ambulatory” [11]. was significantly worse than that of the SP group
One of the authors (Y.K.) explained the purpose and (Mann-Whitney U test, P < 0.05).
details of this study to all the participants, and obtained Magnetic resonance imaging and computed tomog-
their written consent to participate in this study under raphy of the head showed that 17 patients had unilater-
approval of the institutional ethics committee. al hemispheric lesions, three had bilateral hemispheric
lesions, and 10 had lesions in the brain stem or cerebel-
Data analysis lum. 70% of the SPD and 30% of the SP group had
Data of the three groups were stored in a spread- lesions in bilateral hemispheres or a lesion in the brain
sheet; interval scales were expressed as means and stem or cerebellum and the difference in the rate of le-
standard deviation, and ordinal scales were expressed sions was significant (Fisherʼs exact test, P < 0.05).
as median and mode. Age, height, weight, body mass The PCF of the SPD group was significantly lower
index, and spirograms in the three groups were ana- than that of the SP and HC groups, and the PCF of the
lyzed with analysis of variance (ANOVA), followed SP group was also significantly lower than that of the
by post hoc Scheffeʼs tests, if the difference was sig- HC group (one-way ANOVA, Scheffeʼs test, P < 0.05,
nificant. The difference in number of days since the Table 3). The VC of the SPD and SP groups was signif-
onset of stroke were compared between SP and SPD icantly lower than that of the HC group, and IRV of the
groups by using a Student t-test, and the severity of SPD group was also significantly lower than that of the
hemiplegia and FAC were compared between the two HC group (one-way ANOVA, Scheffeʼs test, P < 0.05).
groups by using a Mann-Whitney U test. Pearsonʼs No significant differences in TV or ERV were found
correlation coefficients were obtained to determine among the three groups (one-way ANOVA, P > 0.05).
12 Y Kimura et al
The P/A score and PCF of the SPD group are shown 0.01), VC (0.73, P < 0.01), and age (r = -0.56, P <
in Table 4. Their PCF tended to be lower regardless of 0.01), and also had significant Spearmanʼs rank cor-
their P/A score (PCF range 37.8 - 293.4 l/min). The relation coefficients with FAC (Fig. 1, r = -0.72, P <
PCF score of only one patient (No. 9) was equal to the 0.01). According to a stepwise multivariate regression
average PCF of SP group. analysis, the IRV and FAC were found to be significant
The PCF of the stroke patients had significant Pear- independent variables, and these variables predicted
sonʼs correlation coefficients with IRV (r = 0.72, P < 67.0% of the variance in the PCF (Table 5).
A. C.
600 600
r = -0.56
500 p < 0.01 500
400 400
PCF l /min
PCF l /min
100 100
0 0
40 50 60 70 80 90 0 500 1000 1500 2000 2500
Age (yr) IRV(ml )
B. D.
600 600
PCF l /min
p < 0.01
300 300
200 200
100 100
0 0
0 2000 4000 6000 0 1 2 3
VC (ml ) FAC
Fig. 1. Correlation between age, VC, IRV, FAC and PCF of all stroke patients (SPD and SP). A. Correlation between PCF and
age, B. Correlation between PCF and VC, C. Correlation between PCF and IRV, D. Correlation between PCF and FAC, Age,
VC, IRV, and FAC are all correlated with PCF (P < 0.01). VC: vital capacity, IRV: inspiratory reserve volume, FAC: functional
ambulation categories, PCF: peak cough flow.
14 Y Kimura et al
Table 5. Multiple linear regression analysis of PCF pathway and described that central diaphragmatic im-
Variables R 2
P pairment may cause hypoxia and respiratory dysfunc-
IRV 0.50 0.0001 tion in acute stroke patients [16]. In the initial phase
FAC 0.17 0.0005 of stroke, a decrease in the expiratory capacity, rather
IRV: inspiratory reserve volume, FAC: functional ambulation than the inspiratory capacity, causes restrictive pulmo-
category, R2: adjusted R-squared, PCF: Dependent variable, IRV, nary dysfunction, and inspiratory capacity appears to
FAC: Independet variables decrease with time. Because the diaphragm muscles
are the main factor of ventilation, decreased hemidia-
phragmatic excursion and reduced rib cage expansion
Discussion due to contracture may cause inspiratory restriction
[17]. Trebbia et al. also reported that the maximal
In this study, we found that stroke patients had a inspiratory capacity contributed 44% of the variance
reduced capacity for voluntary cough compared with in the PCF of 155 neuromuscular patients, while the
age-matched healthy males, and, moreover, that the ERV contributed 13% of the variance [18]. They con-
SPD group had a still lower capacity for voluntary cluded that augmenting inspiration was crucial to the
cough than the SP group. Although there have been improvement of PCF, but that the role of the expiratory
only a few reports on the measurement of voluntary muscles could not be ignored. On the other hand, we
cough in stroke patients, Hammond et al reported that demonstrated that expiration was not an essential factor
objective measures of voluntary cough can identify for maintaining PCF in this study. This difference may
stroke patients who are at risk for aspiration. Patients come from the differences in the diseases of the sub-
with Parkinsonʼs disease also show a decreased ca- jects. For example, the subjects examined by Trebbia
pacity for voluntary cough and a swallowing distur- et al. consisted of patients with myotonic dystrophy,
bance from the early stages [12]. Pitts, et al described Duchenne muscular dystrophy, spinal cord injury, and
that the decreased ability to adequately clear material poliomyelitis, and the % predicted value of ERV was
from the airway with a voluntary cough may exacer- only 33 ± 29, whereas the ERV of our stroke patients
bate symptoms resulting from airway penetration in was as much as the HC group.
patients with Parkinsonʼs disease [13]. Ebihara, et al In the present study, the second strongest predictor
measured the PCF of patients with Parkinsonʼs disease, of PCF was the ambulatory function. Although the
and reported that it was 230 ± 74 l/min in the early SPD and SP groups had the same severity of hemiple-
stage and 186±60 l/min in the advanced stage [14]. gia, the ambulatory function of the SPD group was
The PCF of patients with advanced Parkinsonʼs dis- worse than that of the SP group. Possible reasons why
ease approximates that of our SPD group. the SPD group had lower ambulatory function may be
Stroke patients were reported to have a reduced VC reduced progress in training induced by dysphagia-re-
and forced expiratory volume in one second (FEV1.0), lated conditions, general fatigue due to low nutrition,
which may have been due to the affected neurologic low activity in daily life, or a chance relationship.
control of respiratory function and decreased lung We didnʼt include VC in the stepwise regression,
volume excursions due to deteriorated movement of because VC has a strong correlation with TV, IRV and
the chest wall caused by weakness or spasticity of the ERV, and it is a sum of these respiratory factors. If we
rib cage and abdominal muscles [15]. In the present had included VC in our stepwise multivariate regression
study, the IRV was a major factor for reduction of VC model, the result would have been that not IRV but VC
in stroke patients, because VC is composed of TV, ERV became the most significant independent variable.
and IRV, and there were no differences in TV and ERV Our study has some limitations because it was a pre-
between the SPD, SP and HC groups. The decrease liminary cross-sectional study of a small number of
in IRV explained approximately 50% of the variance male stroke patients. It will be necessary to perform a
in the PCF of the stroke patients. Khedr et al exam- prospective study on a larger number of stroke patients
ined the effect of stroke on the corticodiaphragmatic with or without dysphagia, including the measurement
Peak Cough Flow among Stroke Patients With and Without Dysphagia 15
of daily activity and other factors, to reveal whether in- videof luorography. Dysphagia 18: 126-134
spiratory training prevents aspiration pneumonia. 9 . Pitts T, Bolser D, Rosenbek JC, Troche M, Okun MS
& Sapienza C (2009): Impact of expiratory muscle
Conclusion strength training on voluntary cough and swallow
function in Parkinson disease. Chest 135: 1301-1308
We measured the peak cough flow (PCF) in stroke 10 . Brunnstrom S (1970): Movement therapy in hemiple-
patients with and without dysphagia (SPD and SP) to gia: A neurophysiological approach. Harper & Row,
reveal whether SPD have a lower PCF than SP and New York 192pp
healthy controls (HC). The factors relating with the 11 . Holden MK, Gill KM & Magliozzi MR (1986): Gait as-
volume of PCF, including respiratory functions, were sessment in the neurologically impaired patients: Stan-
then investigated to find effective rehabilitative ap- dards for outcome assessment. Phys Ther 66: 1530-
proaches to prevent aspiration in stroke patients. The 1539
results were that the PCF of the SPD group was lower 12 . Smith Hammond CA, Goldstein LB, Horner RD, Ying
than that of the SP and HC groups, and the decrease in J, Gray L, Gonzalez-Rothi L & Bose DC (2009): Pre-
the inspiratory reserve volume (IRV) was a major fac- dicting aspiration in patients with ischemic stroke:
tor associated with the low PCF. comparison of clinical signs and aerodynamic mea-
sures of voluntary cough. Chest 135: 769-777
References 13 . Pitts T, Bolser D, Rosenbek J, Troche M & Sapienza C
(2008): Voluntary cough production and swallow dys-
1 . Daniels SK, Brailey K, Priestly DH, Herrington LR, function in Parkinsonʼs disease. Dysphagia 23: 297-
Weisberg LA & Foundas AL (1998): Aspiration in pa- 301
tients with acute stroke. Arch Phys Med Rehabil 79: 14 . Ebihara S, Saito H, Kanda A, Nakajoh M, Takahashi H,
14-19 Arai H & Sasaki H (2003): Impaired efficacy of cough
2 . Gordon C, Hewer LR & Wader DT (1987): Dysphagia in Patients with Parkinson disease. Chest 124: 1009-
in acute stroke. Br Med J 295: 411-414 1015
3 . Horner J, Massey EW, Riski JE, Lathropdl & Chase 15 . Thompson EC & Murdoch BE (1995): Respiratory
KN (1988): Aspiration following stroke: clinical cor- function associated with dysarthria following upper
relates and outcome. Neurology 38: 1359-1362 motor neurone damage. Aust J Disord in Commu 23:
4 . Mann G, Hankey GJ & Cameron D (1999): Swallow- 61-87
ing function after stroke: Prognosis and prognostic fac- 16 . Khedr EM, Shinawy EI, Khedr T, Ali YA & Awad EM
tors at 6 months. Stroke 30: 744-748 (2000): Assessment of corticodiaphragmatic pathway
5 . Kidd D, Lawson J, Nesbitt R & MacHahon J (1993): and pulmonary function in acute ischemic stroke pa-
Aspiration in acute stroke: A clinical study with video- tients. Eur J Neurol 7: 323-330
f luoroscopy. Q J Med 86: 825-829 17 . Fugl-Meyer AR, Linderholm H & Wilson AF (1983):
6 . Horner J & Massey EW (1988): Silent aspiration fol- Restrictive ventilatory dysfunction in stroke: Its relation
lowing stroke. Neurology 38: 317-319 to locomotor function. Scand J Rehabil Med Suppl 9:
7 . Smith Hammond CA, Goldstein LB, Zajac DJ, Gray 118-124
L, Davenport PW & Bolser DC (2001): Assessment of 18 . Trebbia G, Lacombe M, Fermanian C, Falaize L,
aspiration risk in stroke patients with quantification of Lejaille M, Louis A, Devaux C, Raphael JC & Lofaso F
voluntary cough. Neurology 56: 502-506 (2005): Cough determinants in patients with neuromus-
8 . Tohara H, Saitoh E, Mays KA, Kuhlemeier K & Palmer cular disease. Resp Physiol Neurobiol 146: 291-300
JB (2003): Three tests for predicting aspiration without
16 Y Kimura et al
脳血管障害患者における嚥下障害の有無による最大咳流速の差
木村 美子1,高橋 真紀2,和田 太2,蜂須賀 研二2
1
九州栄養福祉大学 リハビリテーション学部 理学療法学科
2
産業医科大学 医学部 リハビリテーション医学講座
要 旨: 咳嗽は気道の浄化における重要な防御機構であり,十分に強い咳嗽は,嚥下障害を有する患者の誤
嚥リスクを抑制する. 今回の研究目的は,脳血管障害患者の嚥下障害の有無による最大咳流速(PCF)の違いを調
べることであり,さらにはこれらの患者における PCF の身体および呼吸関連因子を明確にすることである. 肺活
量計を用いて,嚥下障害を有する 10 名の脳血管障害患者(SPD),嚥下障害を有しない 20 名の脳血管障害患者
(SP)
と臨床や新聞で募集した性・年齢の一致する 10 名の健常コントロール群(HC)の PCF を測定し比較した.その結果,
SPD の PCF は 160.1 ± 68.7 l/min であり,SP の 297.2 ± 114.2 l/min,
HC の 462.0 ± 84.4 l/min と比べて有意に低下して
いた(one-way analysis of variance, ANOVA, Scheffeʼs test, P < 0.05).また,SPD の肺活量(vital capacity, VC)と予備吸
気量(inspiratory reserve volume, IRV)は,HC の VC や IRV より減少していた. 多変量回帰分析より,IRV と移動能力
(Functional Ambulation Categories, FAC)の PCF に対する寄与率は,それぞれ 50% と 17% であることが分かった.呼吸
機能特に IRV が,SPD の PCF 維持に重要であることが示唆された.
キーワード:嚥下障害,脳血管障害,咳嗽,予備吸気量.
J UOEH(産業医大誌)35
(1)
:9 - 16
(2013)