Speech–Language Pathologist-Led Fiberoptic Endoscopic Evaluation of Swallowing: Functional Outcomes for Patients after Stroke Louise Bax, MSc,* Mary McFarlane, MSc,†‡ Emma Green, MSc,* and Anna Miles, PhD*
Background: Dysphagia is a common complication after stroke and is associated with the development of pneumonia. Early detection of dysphagia and specifically aspiration is, therefore, critical in the prevention of pneumonia. Fiberoptic endoscopic evaluation of swallowing (FEES) is a safe bedside instrumental tool for detecting dysphagia and aspiration and, therefore, has the potential to inform dysphagia management. This study investigated the clinical utility of a speech–language pathologist-led FEES service on functional outcomes for patients after acute stroke. Methods: A retrospective file audit was carried out on 220 patients before FEES was introduced and on 220 patients after the implementation of a speech–language pathologist-led FEES service. The primary outcome measure was incidence of pneumonia, and secondary outcome measures included mortality, diet on discharge, discharge destination, duration nil-by-mouth, incidence of nonoral feeding, and length of stay. Results: There was a significant increase in instrumental assessment use in the group that had access to FEES (P ,.001). There was a significant reduction of pneumonia rates in the group that had access to FEES (P 5.037). Patients were also significantly more likely to leave hospital on standard diets (P 5.004) but had longer periods of nonoral feeding (P 5 .013) and increased length of hospitalization (P , .001). Conclusion: When used selectively, FEES services have potential for improving functional outcomes for patients after stroke. Key Words: Fibreoptic endoscopic evaluation of swallowing—dysphagia—aspiration—pneumonia. Ó 2013 by National Stroke Association
Introduction Dysphagia (difficulties swallowing) after stroke is well established. Up to 64% of patients are likely to have dysphagia in the acute stages.1,2 Dysphagia is associated with increased risk of malnutrition, dehydration,
From the *Department of Speech Science, The University of Auckland, Auckland, New Zealand; †Waitemata District Health Board, Westlake, New Zealand; and ‡Department of Speech and Language Therapy, Northwick Park Hospital, Harrow, UK. Received August 5, 2013; accepted September 30, 2013. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article. Address correspondence to Anna Miles, PhD, Speech Science, University of Auckland, Private Bag 92019, Auckland 1142, New Zealand. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2013 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2013.09.031
pneumonia, poor functional outcomes, mortality, and reduced quality of life.3-7 After stroke, up to one third of patients will develop pneumonia and pneumonia remains the leading cause of death.5,8,9 Patients who aspirate are 10 times more likely to develop pneumonia compared with those who have normal swallowing on videofluoroscopic study of swallowing (VFSS; P , .0001) and with those who silently aspirate (aspirate without a protective cough response) 13 times more likely to develop pneumonia (P , .0001).10 In addition to the health implications associated with developing pneumonia, there are significantly increased health care costs.7,11 In view of these severe consequences, there is a critical need for accessible, reliable assessment tools for the identification of dysphagia. Currently, the only widely recognized methods for identifying aspiration are VFSS and fiberoptic endoscopic evaluation of swallowing (FEES). FEES has proven reliability for identifying the presence of aspiration compared with VFSS.12-18 FEES
Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2013: pp 1-6
1
L. BAX ET AL.
2
is safe, can be conducted at bedside, and concurs significantly less costs than VFSS.16,19 Finally, with the benefit of no radiation exposure, it can be repeated multiple times if clinically indicated.17,18,20 As a result, the use of FEES has received increasing support in acute dysphagia assessment, specifically in the acute stroke population.21 The aim of this study was to examine the clinical utility of a speech–language pathologist (SLP)led FEES service in an inpatient stroke unit.
Methods Study Design This study received appropriate regional ethics approval. Four hundred forty patients with a diagnosis of stroke admitted to an inpatient stroke unit in an urban hospital and referred to SLP for dysphagia assessment were included in the study. A retrospective file audit was carried out on 220 patients before FEES was introduced (pre-FEES group) and 220 patients after the implementation of an SLP-led FEES service (FEES group). Each audit group spanned 12 months of hospital admissions with a 6-month gap between audits while the FEES service was introduced. Twenty files were double audited to check reliability with strong agreement (100% agreement, K 5 1.0).
Referral to Instrumental Assessment In the pre-FEES group, the stroke unit had access to onsite VFSS in a radiology department. In the FEES group, the stroke unit had access to a ward-based FEES service and VFSS, with SLPs qualified to perform FEES independently. The instigation of the FEES service was prompted by the recruitment of an SLP experienced in FEES and the purchase of mobile FEES equipment. There was no addition in staffing. FEES were performed at the patient’s bedside using a 3.2-mm-diameter flexible video rhinolaryngoscope (ENF, V2; Olympus Corporation, Tokyo, Japan), an integrated light source and video processor (Olympus, OTV-SI; Olympus Corporation), and an LCD Monitor (Olympus OEV203; Olympus Corporation). No topical anesthesia was used. The endoscope was passed through the nose and positioned in the pharynx. Patients were administered varying quantities and consistencies of food and fluid colored with green food dye (apple sauce, milk, banana, cheese sandwich, continuous drinking of milk from a straw, and thickened fluids if indicated). The protocol was modified as required for patient safety. The criteria for referral to instrumental swallowing assessment did not differ between groups, with referrals made by the treating SLP and medical team based on individual patients’ clinical presentation. Cough reflex testing had also been introduced during the 2 recruitment periods. Cough reflex testing is a test of airway sensitivity and is considered to indicate patients
22
at risk of silent aspiration. In isolation, cough reflex testing has not been successful in reducing pneumonia rates, but it is assumed that it supports appropriate, timely referrals for instrumental assessment.23
Outcome Measures The primary outcome measure was pneumonia, using the criteria described by Mann et al,24 where 3 or more of the following variables constitute a diagnosis: fever (.38 C), productive cough with purulent sputum, abnormal respiratory examination (tachypnea [.22/ min], tachycardia, inspiratory crackles, bronchial breathing), abnormal chest radiograph, arterial hypoxemia (PO2 ,70 mm Hg), and isolation of a relevant pathogen (positive gram stain and culture). Other outcome measures included mortality, diet on discharge, discharge destination, length of stay (LOS), days kept nil-bymouth (NBM), and nonoral feeding. Diet was classified in terms of texture as per the Australian Dietitians’ Guidelines (standard, soft, minced–moist, puree, or NBM).25 Number and type of instrumental assessment (FEES or VFSS) and days until instrumental assessment were recorded. Definitions of variables are provided in Table 1.
Data Analysis Data were analyzed using SPSS Version 20 (SPSS, Chicago, IL). Chi-square analyses were used for categorical variables, and independent samples t tests were used for continuous variables. Multiple logistic regressions were applied to evaluate the efficacy of the FEES service adjusted for confounding variables based on bivariate analyses (gender, age, comorbidities, stroke severity, stroke site, instrument assessment, LOS, days nil-by-mouth, and days nonoral feeding) and the 2-way interactions including FEES service and LOS, FEES service and nonoral feeding, and FEES service and instrumental assessment. First the full model with all confounding factors was fit, and backward selection was used to select the main effect model. The two-way interactions were then added to the main effect model one by one for the final model. Multiple logistic regressions were then similarly applied to evaluate the development of pneumonia adjusted for confounding variables (age, LOS, gender, stroke type, stroke severity, comorbidities, experimental group, instrumental assessment, nonoral diet). A sample size of 219 participants in each group was calculated to achieve 80% power at a statistical level of .05.
Results There were no significant differences in baseline characteristics between the groups (Table 2).
SLP-LED FEES SERVICE: FUNCTIONAL OUTCOMES AFTER STROKE
3
Table 1. Definition of variables Variable
Definition
Pneumonia Mortality Days NBM Nonoral feeding Days nonoral Comorbidities Instrumental assessment Days instrumental Stroke severity LOS
Patient developed pneumonia as defined by Mann criteria at some point in their hospital admission. Patient passed away in hospital. Number of days before a patient either began an oral diet or began being fed nonorally. Patient was fed via NG or PEG tube at some point during their admission. Number of days in total a patient was nonorally fed during their hospital admission. Patient had pre-existing respiratory comorbidities before their hospital admission (eg, asthma, bronchiectasis). Patient received an instrumental assessment of swallowing (FEES or VFSS) during their hospital admission. Number of days a patient waited to have an instrumental assessment. Stroke severity as rated according to NIHSS. Number of days of hospital admission.
Abbreviations: FEES, fiberoptic endoscopic evaluation of swallowing; LOS, length of stay; NIHSS, National Institutes of Health Stroke Scale; NG, nasogastric tube; NBM, nil-by-mouth; PEG, percutaneous endoscopic gastrostomy; VFSS, videofluoroscopic study of swallowing.
Pneumonia A significant association was found between experimental group and pneumonia (pre-FEES pneumonia rate Table 2. Demographic data of pre-FEES and FEES groups
Age Mean (SD) Gender Male Ethnicity NZ European NZ Maori Pacific Island Asian Other Hemisphere Left Right Bilateral Type Ischemic Hemorrhagic Oxford classification LACI TACI POCI PACI NIHSS stroke Severity score, mean (SD) Comorbidities Yes
Pre-FEES (%)
FEES (%)
P value
n 5 220 76.5 (12.8) n 5 220 95 (43.1) n 5 220 150 (68.2) 9 (4.1) 12 (5.45) 12 (5.45) 37 (16.8) n 5 215 106 (49.3) 102 (47.4) 7 (3.3) n 5 217 190 (87.6) 27 (12.4) n 5 166 52 (31.3) 43 (25.9) 11 (6.7) 60 (36.1) n 5 182 8.53 (6.05)
n 5 220 76.3 (13.3) n 5 220 93 (42.2) n 5 218 151 (69.3) 6 (2.7) 10 (4.6) 12 (5.5) 39 (17.9) n 5 213 106 (49.8) 102 (47.9) 5 (2.3) n 5 214 183 (85.5) 31 (14.5) n 5 181 63 (34.8) 48 (26.5) 12 (6.6) 58 (32.1) n 5 203 8.14 (5.97)
.891
n 5 210 27 (12.9)
n 5 217 31 (14.3)
.923 .941
12%, FEES pneumonia rate 7%, proportion difference .5; 95% confidence interval [CI] .026, .05; unadjusted: c2 (1) 5 3.79, P 5 .037, adjusted: P , .001; odds ratio [OR] 5.5; 95% CI 6.27, 33.76). In the pre-FEES group, instrumental assessment was significantly associated with developing pneumonia (c2 5 4.5, P 5 .033; OR 5.7; 95% CI 1.15, 28.35). Conversely, in the FEES group, instrumental assessment was significantly associated with not developing pneumonia (c2 5 4.977, P 5 .026; OR 3.5; 95% CI 1.166, 10.751). Nonoral feeding was significantly associated with not developing pneumonia in both groups (pre-FEES; c2 5 5.12, P 5 .024; OR 2.9; 95% CI 1.15, 7.35, FEES; c2 5 6.6, P 5 .010; OR 6.88; 95% CI 1.58, 29.88).
Functional Outcomes
.898
.574
.868
.520
.675
Abbreviations: FEES, fiberoptic endoscopic evaluation of swallowing; NIHSS, National Institutes of Health Stroke Scale; LACI, lacunar infarct; TACI, total anterior circulation infarct; POCI, posterior circulation infarct; PACI, partial anterior circulation infarct.
There was no significant difference between groups in the duration of NBM status (pre-FEES group 1.23 days, FEES group 1.41 days, P 5 .195). There was no significant difference between groups in the number of patients nonorally fed (FEES group 15.9%, pre-FEES group 11.4%, P 5.105). In the FEES group, patients who were nonorally fed received nonoral feeding for significantly longer than patients who were nonorally fed in the pre-FEES group (pre-FEES group 8%, FEES group 14%, P 5 .013). If a patient left hospital on an oral diet, they were statistically more likely to leave hospital on standard rather than a modified diet (categorized as soft, minced–moist, and puree) in the FEES group compared with the pre-FEES group (pre-FEES group 51%, FEES group 66%, c2 5 8.24, P 5 .005). Patients in the pre-FEES group had significantly shorter LOS in the hospital admission (mean difference 5 2.55; 95% CI 1.23-3.85, P ,.001). There was no statistically significant difference between groups on discharge destination (P 5.459) or mortality (P 5.314).
Instrumental Assessment Use Patients in the FEES group were significantly more likely to receive an instrumental assessment (FEES or
L. BAX ET AL.
4
VFSS) than the pre-FEES group (pre-FEES group 6%, FEES group 38%, P , .001; Table 3). In the FEES group, 35 patients (16%) received multiple instrumental assessments during their admission. In the pre-FEES group, only 1 patient received a repeat VFSS. Patients in the FEES group waited significantly fewer days between admission and instrumental assessment taking place (pre-FEES group mean 5 10.5, FEES group mean 5 2.3, P 5 .008; Table 3).
Discussion Pneumonia rates were lower for the group with access to the SLP-led FEES service. The incidence of pneumonia in both groups is well below recent national data from a multisite randomized clinical trial of 311 patients with dysphagia after stroke (27% pneumonia rate).23 Furthermore, the 7% pneumonia rate in the FEES group is below the national rate of 10% reported in the general stroke Table 3. Comparison of clinical decision making and functional outcomes between groups Pre-FEES (%) Instrumental n 5 220 assessment Yes 14 (6.4) Instrumental n 5 14 assessment Days until; 10.5 (9.8) mean (SD) Nonoral feeding n 5 220 PEG or NG 25 (11.4) Nil-by-mouth n 5 215 d; mean (SD) 1.23 (1.4) Days nonoral feeding n 5 25 NG; mean (SD) 7.6 (6.5) Discharge diet n 5 186 Standard 95 (48.5) Soft 50 (25.5) Minced–moist 18 (9.2) Puree 23 (11.7) Length of stay n 5 220 Mean (SD) 17.34 (15.2) Discharge destination n 5 220 Home 96 (43.6) Rehabilitation 43 (19.5) Rest home 14 (6.4) Private hospital 41 (18.6) Death 23 (10.5) Other 3 (1.4) Mortality n 5 220 Yes 23 (10.5)
FEES (%)
P value
n 5 220
,.001
84 (38.2) n 5 84
.008
2.3 (4.9) n 5 220 .211 35 (15.9) n 5 208 .195 1.41 (1.5) n 5 34 .013 13.5 (11.1) n 5 192 .005 126 (61.8) 31 (15.2) 16 (7.8) 19 (9.3) n 5 220 ,.001 23.67 (20.2) n 5 220 .459 102 (46.4) 36 (16.4) 10 (4.5) 54 (24.5) 16 (7.3) 2 (.9) n 5 220 .314 16 (7.3)
Abbreviations: FEES, Fiberoptic endoscopic evaluation of swallowing; NG, nasogastric tube; PEG, percutaneous endoscopic gastrostomy.
26
population. Both specialized stroke units and nurse dysphagia screens are known to lead to improved outcomes for patients after stroke, and these were both present at the research site.27 However, these services were already present during the pre-FEES group data collection. Although patients after the implementation of the FEES service were significantly less likely to develop pneumonia if they received an instrumental assessment, the opposite was true for the pre-FEES group. Miles et al23 recently described similar findings. They report that no patients in their control group received an instrumental assessment until they developed pneumonia. This ‘‘wait-and-see’’ approach to dysphagia management is concerning and suggests that SLPs are hesitant to refer patients for instrumental assessment perhaps because of lack of accessibility. Ward-based FEES may be an efficient, fast means of instrumental assessment, therefore, avoiding patients developing pneumonia while they wait for VFSS and minimizing the inconvenience of waiting for a radiology procedure. The visibility and collaborative nature of FEES are also thought to lead to a better team understanding of dysphagia and, perhaps, in turn stricter adherence of SLP recommendations.28 The introduction of cough reflex testing in conjunction with the FEES service may have allowed more informed decisions regarding referral criteria for instrumental assessment as those who fail a cough reflex test are more likely to be referred for VFSS or FEES. Cough reflex testing has been shown to increase instrumental assessment rates in the stroke population.23 Combining a ward-based FEES service with a comprehensive clinical swallowing evaluation including cough reflex testing may avoid overuse of instrumental assessment and ensure the right patients receive the additional investigation. The FEES group, where pneumonia rates were lower, had significantly longer hospitalization. In contrast, similarly designed international studies have associated the development of pneumonia with longer hospitalization.8,11 It could be hypothesised that the introduction of the FEES service has provided SLPs with greater input into the rehabilitation process for patients and subsequently led to patients remaining in hospital for swallowing rehabilitation before being discharged. This theory is supported by the higher percentage of patients in the FEES group leaving hospital on standard diets. This has positive repercussions in terms of quality of life. Diet restrictions have been associated with reduced quality of life.29 Modified foods lose nutritional value through the process of modification, and therefore, diet modification leads to a risk of malnutrition and dehydration.29 Discharging with diet modifications may necessitate further community follow-up and, therefore, concur additional and, perhaps, unnecessary health care costs. Nonoral feeding was significantly associated with not developing pneumonia, although the effect size was much larger in the FEES group. Yet, previous research
SLP-LED FEES SERVICE: FUNCTIONAL OUTCOMES AFTER STROKE
has suggested a higher incidence of pneumonia in patients who are nonorally fed, with 1 study reporting that 41% of tube-fed patients after stroke developed pneumonia.30 It is possible that the FEES service allowed clinicians to be more assertive and confident in making earlier, objectively informed management decisions regarding nonoral feeding options.
Limitations The retrospective nature of the research means that data collection is reliant on access to, and accurate, documentation by health professionals. Similarly, the auditors are required to be both accurate and consistent in their interpretation. Both documentation and interpretation are susceptible to human error. Additionally, the between-group study design over 2 time periods means that confounding variables such as changes to hospital protocol or personnel cannot be accounted for.
Conclusions Introduction of an SLP-led FEES service was associated with a reduction in pneumonia rates in an inpatient stroke unit. Patients were discharged from hospital on less restrictive diets but following longer hospitalization and longer periods of nonoral feeding. Guidelines surrounding selection criteria of candidates for FEES may help to streamline the use of FEES and make such services more efficient. This would provide further strength to potential benefits of FEES in reducing pneumonia rates and improving functional outcomes for patients after stroke. Acknowledgment: The authors would like to acknowledge Waitemata DHB for their support in this study and for allowing researchers access their hospital databases.
References 1. Daniels SK, Foundas AL. Lesion localization in acute stroke patients with risk of aspiration. J Neuroimag 1999;9:91-98. 2. Hamdy S, Aziz Q, Rothwell JC, et al. Recovery of swallowing after dysphagic stroke relates to functional reorganization in the intact motor cortex. Gastroenterology 1998;115:1104-1112. 3. Martino R, Foley N, Bhogal S, et al. Dysphagia after stroke: incidence, diagnosis, and pulmonary complications. Stroke 2005;36:2756-2763. 4. Sellars C, Bowie L, Bagg J, et al. Risk factors for chest infection in acute stroke: a prospective cohort study. Stroke 2007;38:2284-2291. 5. Henon H, Godefroy O, Leys D, et al. Early predictors of death and disability after acute cerebral ischaemic event. Stroke 1995;26:392-398. 6. Reynolds PS, Gilbert L, Good DC, et al. Pneumonia in dysphagic stroke patients: effect on outcomes and identification of high risk patients. Neurorehabil Neural Repair 1998;12:15-21.
5
7. Katzan IL, Dawson NV, Thomas CL, et al. The cost of pneumonia after stroke. Neurology 2007;68:1938-1943. 8. Finlayson O, Kapral M, Hall R, et al. Risk factors, inpatient care, and outcomes of pneumonia after ischemic stroke. Neurology 2011;77:1338-1345. 9. Katzan IL, Cebul RD, Husak SH, et al. The effect of pneumonia on mortality among patients hospitalized for acute stroke. Neurology 2003;60:620-625. 10. Pikus L, Levine MS, Yang YX, et al. Videofluoroscopic studies of swallowing dysfunction and the relative risk of pneumonia. Am J Roentgenol 2003;180: 1613-1616. 11. Wilson RD. Mortality and cost of pneumonia after stroke for different risk groups. J Stroke Cerebrovasc Dis 2012; 21:61-67. 12. Colodny N. Inter-judge and intra-judge reliabilities in fiberoptic endoscopic evaluation of swallowing (FEES) using the penetration-aspiration scale: a replication study. Dysphagia 2002;17:308-315. 13. Langmore SE, Schatz MA, Olsen N. Endoscopic and videofluoroscopic evaluations of swallowing and aspiration. Ann Otol Rhinol Laryngol 1991;100:678-681. 14. Perie S, Laccourreye L, Flahault A, et al. Role of videoendoscopy in assessment of pharyngeal function in oro-pharyngeal dysphagia: comparison with videofluoroscopy and manometry. Laryngoscope 1998;108:1712-1716. 15. Wu C-H, Hsaio T-Y, Chen J-C, et al. Evaluation of swallowing safety with fiberoptic endoscope: comparison with videofluoroscopic technique. Laryngoscope 1997; 107:396-401. 16. Aviv JE. Prospective, randomized outcome study of endoscopy versus modified barium swallow in patients with dysphagia. Laryngoscope 2000;110:563-574. 17. Rao N, Bray SL, Chaudhuri G, et al. Gold-standard? Analysis of the videofluoroscopic and fiberoptic endoscopic swallow examinations. J Appl Res 2003; 3:89-96. 18. Tohara H, Nakane A, Murata S, et al. Inter- and intra-rater reliability in fiberoptic endoscopic evaluation of swallowing. J Oral Rehabil 2010;37:884-891. 19. Ajemian MS, Nirmul GB, Anderson MT, et al. Routine fiberoptic endoscopic evaluation of swallowing following prolonged intubation. Arch Surg 2001;136:434-437. 20. Ramsey DJC, Smithard DG, Kalra L. Early assessments of dysphagia and aspiration risk in acute stroke patients. Stroke 2003;34:1252-1257. 21. Leder SB, Espinosa JF. Aspiration risk after acute stroke: comparison of clinical examination and fiberoptic endoscopic evaluation of swallow. Dysphagia 2002; 17:214-218. 22. Miles A, Moore S, McFarlane M, et al. Comparison of cough reflex test against instrumental assessment of aspiration. Physiol Behav 2013;118:25-31. 23. Miles A, McLauchlan H, Zeng I, et al. Cough reflex testing in dysphagia following stroke: a randomized clinical trial. J Clin Med Res 2013;5:222-233. 24. Mann G, Hankey GJ, Cameron D. Swallowing function after stroke: prognosis and prognostic factors at 6 months. Stroke 1999;30:744-748. 25. Atherton M, Bellis-Smith N, Cichero J, et al. Texture modified foods and thickened fluids as used for individuals with dysphagia: Australian standardised labels and definitions. Nutr Diet, 2007; 64: S53-S76. 26. Child N, Fink J, Jones S, et al. New Zealand national acute stroke services audit: acute stroke care delivery in New Zealand. N Z Med J 2012;125:44-51.
6 27. Stroke Unit Trial Collaboration. Organised inpatient (stroke unit) care for stroke patients. Cochrane Database Syst Rev 2007;4:1-72. 28. Green E, McFarlane M, Miles A, et al. Staff perceptions following the introduction of a fibreoptic endoscopic evaluation of swallowing service for patients following stroke. Speech, Language and Hearing. In Revision.
L. BAX ET AL. 29. Keller H, Chambers L, Niezgoda H, et al. Issues associated with the use of modified texture foods. J Nutr, Health, Ageing 2012;16:195-200. 30. Langdon PC, Lee AH, Binns CW. High incidence of respiratory infection in nil-by-mouth tube-fed acute ischemic stroke patients. Neuroepidemiology 2009; 32:107-113.
Background: Dysphagia is a common complication after stroke and is associated with the development of pneumonia. Early detection of dysphagia and specifically aspiration is, therefore, critical in the prevention of pneumonia. Fiberoptic endoscopic evaluation of swallowing (FEES) is a safe bedside instrumental tool for detecting dysphagia and aspiration and, therefore, has the potential to inform dysphagia management. This study investigated the clinical utility of a speech–language pathologist-led FEES service on functional outcomes for patients after acute stroke. Methods: A retrospective file audit was carried out on 220 patients before FEES was introduced and on 220 patients after the implementation of a speech–language pathologist-led FEES service. The primary outcome measure was incidence of pneumonia, and secondary outcome measures included mortality, diet on discharge, discharge destination, duration nil-by-mouth, incidence of nonoral feeding, and length of stay. Results: There was a significant increase in instrumental assessment use in the group that had access to FEES (P ,.001). There was a significant reduction of pneumonia rates in the group that had access to FEES (P 5.037). Patients were also significantly more likely to leave hospital on standard diets (P 5.004) but had longer periods of nonoral feeding (P 5 .013) and increased length of hospitalization (P , .001). Conclusion: When used selectively, FEES services have potential for improving functional outcomes for patients after stroke. Key Words: Fibreoptic endoscopic evaluation of swallowing—dysphagia—aspiration—pneumonia. Ó 2013 by National Stroke Association
Introduction Dysphagia (difficulties swallowing) after stroke is well established. Up to 64% of patients are likely to have dysphagia in the acute stages.1,2 Dysphagia is associated with increased risk of malnutrition, dehydration,
From the *Department of Speech Science, The University of Auckland, Auckland, New Zealand; †Waitemata District Health Board, Westlake, New Zealand; and ‡Department of Speech and Language Therapy, Northwick Park Hospital, Harrow, UK. Received August 5, 2013; accepted September 30, 2013. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article. Address correspondence to Anna Miles, PhD, Speech Science, University of Auckland, Private Bag 92019, Auckland 1142, New Zealand. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2013 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2013.09.031
pneumonia, poor functional outcomes, mortality, and reduced quality of life.3-7 After stroke, up to one third of patients will develop pneumonia and pneumonia remains the leading cause of death.5,8,9 Patients who aspirate are 10 times more likely to develop pneumonia compared with those who have normal swallowing on videofluoroscopic study of swallowing (VFSS; P , .0001) and with those who silently aspirate (aspirate without a protective cough response) 13 times more likely to develop pneumonia (P , .0001).10 In addition to the health implications associated with developing pneumonia, there are significantly increased health care costs.7,11 In view of these severe consequences, there is a critical need for accessible, reliable assessment tools for the identification of dysphagia. Currently, the only widely recognized methods for identifying aspiration are VFSS and fiberoptic endoscopic evaluation of swallowing (FEES). FEES has proven reliability for identifying the presence of aspiration compared with VFSS.12-18 FEES
Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2013: pp 1-6
1
L. BAX ET AL.
2
is safe, can be conducted at bedside, and concurs significantly less costs than VFSS.16,19 Finally, with the benefit of no radiation exposure, it can be repeated multiple times if clinically indicated.17,18,20 As a result, the use of FEES has received increasing support in acute dysphagia assessment, specifically in the acute stroke population.21 The aim of this study was to examine the clinical utility of a speech–language pathologist (SLP)led FEES service in an inpatient stroke unit.
Methods Study Design This study received appropriate regional ethics approval. Four hundred forty patients with a diagnosis of stroke admitted to an inpatient stroke unit in an urban hospital and referred to SLP for dysphagia assessment were included in the study. A retrospective file audit was carried out on 220 patients before FEES was introduced (pre-FEES group) and 220 patients after the implementation of an SLP-led FEES service (FEES group). Each audit group spanned 12 months of hospital admissions with a 6-month gap between audits while the FEES service was introduced. Twenty files were double audited to check reliability with strong agreement (100% agreement, K 5 1.0).
Referral to Instrumental Assessment In the pre-FEES group, the stroke unit had access to onsite VFSS in a radiology department. In the FEES group, the stroke unit had access to a ward-based FEES service and VFSS, with SLPs qualified to perform FEES independently. The instigation of the FEES service was prompted by the recruitment of an SLP experienced in FEES and the purchase of mobile FEES equipment. There was no addition in staffing. FEES were performed at the patient’s bedside using a 3.2-mm-diameter flexible video rhinolaryngoscope (ENF, V2; Olympus Corporation, Tokyo, Japan), an integrated light source and video processor (Olympus, OTV-SI; Olympus Corporation), and an LCD Monitor (Olympus OEV203; Olympus Corporation). No topical anesthesia was used. The endoscope was passed through the nose and positioned in the pharynx. Patients were administered varying quantities and consistencies of food and fluid colored with green food dye (apple sauce, milk, banana, cheese sandwich, continuous drinking of milk from a straw, and thickened fluids if indicated). The protocol was modified as required for patient safety. The criteria for referral to instrumental swallowing assessment did not differ between groups, with referrals made by the treating SLP and medical team based on individual patients’ clinical presentation. Cough reflex testing had also been introduced during the 2 recruitment periods. Cough reflex testing is a test of airway sensitivity and is considered to indicate patients
22
at risk of silent aspiration. In isolation, cough reflex testing has not been successful in reducing pneumonia rates, but it is assumed that it supports appropriate, timely referrals for instrumental assessment.23
Outcome Measures The primary outcome measure was pneumonia, using the criteria described by Mann et al,24 where 3 or more of the following variables constitute a diagnosis: fever (.38 C), productive cough with purulent sputum, abnormal respiratory examination (tachypnea [.22/ min], tachycardia, inspiratory crackles, bronchial breathing), abnormal chest radiograph, arterial hypoxemia (PO2 ,70 mm Hg), and isolation of a relevant pathogen (positive gram stain and culture). Other outcome measures included mortality, diet on discharge, discharge destination, length of stay (LOS), days kept nil-bymouth (NBM), and nonoral feeding. Diet was classified in terms of texture as per the Australian Dietitians’ Guidelines (standard, soft, minced–moist, puree, or NBM).25 Number and type of instrumental assessment (FEES or VFSS) and days until instrumental assessment were recorded. Definitions of variables are provided in Table 1.
Data Analysis Data were analyzed using SPSS Version 20 (SPSS, Chicago, IL). Chi-square analyses were used for categorical variables, and independent samples t tests were used for continuous variables. Multiple logistic regressions were applied to evaluate the efficacy of the FEES service adjusted for confounding variables based on bivariate analyses (gender, age, comorbidities, stroke severity, stroke site, instrument assessment, LOS, days nil-by-mouth, and days nonoral feeding) and the 2-way interactions including FEES service and LOS, FEES service and nonoral feeding, and FEES service and instrumental assessment. First the full model with all confounding factors was fit, and backward selection was used to select the main effect model. The two-way interactions were then added to the main effect model one by one for the final model. Multiple logistic regressions were then similarly applied to evaluate the development of pneumonia adjusted for confounding variables (age, LOS, gender, stroke type, stroke severity, comorbidities, experimental group, instrumental assessment, nonoral diet). A sample size of 219 participants in each group was calculated to achieve 80% power at a statistical level of .05.
Results There were no significant differences in baseline characteristics between the groups (Table 2).
SLP-LED FEES SERVICE: FUNCTIONAL OUTCOMES AFTER STROKE
3
Table 1. Definition of variables Variable
Definition
Pneumonia Mortality Days NBM Nonoral feeding Days nonoral Comorbidities Instrumental assessment Days instrumental Stroke severity LOS
Patient developed pneumonia as defined by Mann criteria at some point in their hospital admission. Patient passed away in hospital. Number of days before a patient either began an oral diet or began being fed nonorally. Patient was fed via NG or PEG tube at some point during their admission. Number of days in total a patient was nonorally fed during their hospital admission. Patient had pre-existing respiratory comorbidities before their hospital admission (eg, asthma, bronchiectasis). Patient received an instrumental assessment of swallowing (FEES or VFSS) during their hospital admission. Number of days a patient waited to have an instrumental assessment. Stroke severity as rated according to NIHSS. Number of days of hospital admission.
Abbreviations: FEES, fiberoptic endoscopic evaluation of swallowing; LOS, length of stay; NIHSS, National Institutes of Health Stroke Scale; NG, nasogastric tube; NBM, nil-by-mouth; PEG, percutaneous endoscopic gastrostomy; VFSS, videofluoroscopic study of swallowing.
Pneumonia A significant association was found between experimental group and pneumonia (pre-FEES pneumonia rate Table 2. Demographic data of pre-FEES and FEES groups
Age Mean (SD) Gender Male Ethnicity NZ European NZ Maori Pacific Island Asian Other Hemisphere Left Right Bilateral Type Ischemic Hemorrhagic Oxford classification LACI TACI POCI PACI NIHSS stroke Severity score, mean (SD) Comorbidities Yes
Pre-FEES (%)
FEES (%)
P value
n 5 220 76.5 (12.8) n 5 220 95 (43.1) n 5 220 150 (68.2) 9 (4.1) 12 (5.45) 12 (5.45) 37 (16.8) n 5 215 106 (49.3) 102 (47.4) 7 (3.3) n 5 217 190 (87.6) 27 (12.4) n 5 166 52 (31.3) 43 (25.9) 11 (6.7) 60 (36.1) n 5 182 8.53 (6.05)
n 5 220 76.3 (13.3) n 5 220 93 (42.2) n 5 218 151 (69.3) 6 (2.7) 10 (4.6) 12 (5.5) 39 (17.9) n 5 213 106 (49.8) 102 (47.9) 5 (2.3) n 5 214 183 (85.5) 31 (14.5) n 5 181 63 (34.8) 48 (26.5) 12 (6.6) 58 (32.1) n 5 203 8.14 (5.97)
.891
n 5 210 27 (12.9)
n 5 217 31 (14.3)
.923 .941
12%, FEES pneumonia rate 7%, proportion difference .5; 95% confidence interval [CI] .026, .05; unadjusted: c2 (1) 5 3.79, P 5 .037, adjusted: P , .001; odds ratio [OR] 5.5; 95% CI 6.27, 33.76). In the pre-FEES group, instrumental assessment was significantly associated with developing pneumonia (c2 5 4.5, P 5 .033; OR 5.7; 95% CI 1.15, 28.35). Conversely, in the FEES group, instrumental assessment was significantly associated with not developing pneumonia (c2 5 4.977, P 5 .026; OR 3.5; 95% CI 1.166, 10.751). Nonoral feeding was significantly associated with not developing pneumonia in both groups (pre-FEES; c2 5 5.12, P 5 .024; OR 2.9; 95% CI 1.15, 7.35, FEES; c2 5 6.6, P 5 .010; OR 6.88; 95% CI 1.58, 29.88).
Functional Outcomes
.898
.574
.868
.520
.675
Abbreviations: FEES, fiberoptic endoscopic evaluation of swallowing; NIHSS, National Institutes of Health Stroke Scale; LACI, lacunar infarct; TACI, total anterior circulation infarct; POCI, posterior circulation infarct; PACI, partial anterior circulation infarct.
There was no significant difference between groups in the duration of NBM status (pre-FEES group 1.23 days, FEES group 1.41 days, P 5 .195). There was no significant difference between groups in the number of patients nonorally fed (FEES group 15.9%, pre-FEES group 11.4%, P 5.105). In the FEES group, patients who were nonorally fed received nonoral feeding for significantly longer than patients who were nonorally fed in the pre-FEES group (pre-FEES group 8%, FEES group 14%, P 5 .013). If a patient left hospital on an oral diet, they were statistically more likely to leave hospital on standard rather than a modified diet (categorized as soft, minced–moist, and puree) in the FEES group compared with the pre-FEES group (pre-FEES group 51%, FEES group 66%, c2 5 8.24, P 5 .005). Patients in the pre-FEES group had significantly shorter LOS in the hospital admission (mean difference 5 2.55; 95% CI 1.23-3.85, P ,.001). There was no statistically significant difference between groups on discharge destination (P 5.459) or mortality (P 5.314).
Instrumental Assessment Use Patients in the FEES group were significantly more likely to receive an instrumental assessment (FEES or
L. BAX ET AL.
4
VFSS) than the pre-FEES group (pre-FEES group 6%, FEES group 38%, P , .001; Table 3). In the FEES group, 35 patients (16%) received multiple instrumental assessments during their admission. In the pre-FEES group, only 1 patient received a repeat VFSS. Patients in the FEES group waited significantly fewer days between admission and instrumental assessment taking place (pre-FEES group mean 5 10.5, FEES group mean 5 2.3, P 5 .008; Table 3).
Discussion Pneumonia rates were lower for the group with access to the SLP-led FEES service. The incidence of pneumonia in both groups is well below recent national data from a multisite randomized clinical trial of 311 patients with dysphagia after stroke (27% pneumonia rate).23 Furthermore, the 7% pneumonia rate in the FEES group is below the national rate of 10% reported in the general stroke Table 3. Comparison of clinical decision making and functional outcomes between groups Pre-FEES (%) Instrumental n 5 220 assessment Yes 14 (6.4) Instrumental n 5 14 assessment Days until; 10.5 (9.8) mean (SD) Nonoral feeding n 5 220 PEG or NG 25 (11.4) Nil-by-mouth n 5 215 d; mean (SD) 1.23 (1.4) Days nonoral feeding n 5 25 NG; mean (SD) 7.6 (6.5) Discharge diet n 5 186 Standard 95 (48.5) Soft 50 (25.5) Minced–moist 18 (9.2) Puree 23 (11.7) Length of stay n 5 220 Mean (SD) 17.34 (15.2) Discharge destination n 5 220 Home 96 (43.6) Rehabilitation 43 (19.5) Rest home 14 (6.4) Private hospital 41 (18.6) Death 23 (10.5) Other 3 (1.4) Mortality n 5 220 Yes 23 (10.5)
FEES (%)
P value
n 5 220
,.001
84 (38.2) n 5 84
.008
2.3 (4.9) n 5 220 .211 35 (15.9) n 5 208 .195 1.41 (1.5) n 5 34 .013 13.5 (11.1) n 5 192 .005 126 (61.8) 31 (15.2) 16 (7.8) 19 (9.3) n 5 220 ,.001 23.67 (20.2) n 5 220 .459 102 (46.4) 36 (16.4) 10 (4.5) 54 (24.5) 16 (7.3) 2 (.9) n 5 220 .314 16 (7.3)
Abbreviations: FEES, Fiberoptic endoscopic evaluation of swallowing; NG, nasogastric tube; PEG, percutaneous endoscopic gastrostomy.
26
population. Both specialized stroke units and nurse dysphagia screens are known to lead to improved outcomes for patients after stroke, and these were both present at the research site.27 However, these services were already present during the pre-FEES group data collection. Although patients after the implementation of the FEES service were significantly less likely to develop pneumonia if they received an instrumental assessment, the opposite was true for the pre-FEES group. Miles et al23 recently described similar findings. They report that no patients in their control group received an instrumental assessment until they developed pneumonia. This ‘‘wait-and-see’’ approach to dysphagia management is concerning and suggests that SLPs are hesitant to refer patients for instrumental assessment perhaps because of lack of accessibility. Ward-based FEES may be an efficient, fast means of instrumental assessment, therefore, avoiding patients developing pneumonia while they wait for VFSS and minimizing the inconvenience of waiting for a radiology procedure. The visibility and collaborative nature of FEES are also thought to lead to a better team understanding of dysphagia and, perhaps, in turn stricter adherence of SLP recommendations.28 The introduction of cough reflex testing in conjunction with the FEES service may have allowed more informed decisions regarding referral criteria for instrumental assessment as those who fail a cough reflex test are more likely to be referred for VFSS or FEES. Cough reflex testing has been shown to increase instrumental assessment rates in the stroke population.23 Combining a ward-based FEES service with a comprehensive clinical swallowing evaluation including cough reflex testing may avoid overuse of instrumental assessment and ensure the right patients receive the additional investigation. The FEES group, where pneumonia rates were lower, had significantly longer hospitalization. In contrast, similarly designed international studies have associated the development of pneumonia with longer hospitalization.8,11 It could be hypothesised that the introduction of the FEES service has provided SLPs with greater input into the rehabilitation process for patients and subsequently led to patients remaining in hospital for swallowing rehabilitation before being discharged. This theory is supported by the higher percentage of patients in the FEES group leaving hospital on standard diets. This has positive repercussions in terms of quality of life. Diet restrictions have been associated with reduced quality of life.29 Modified foods lose nutritional value through the process of modification, and therefore, diet modification leads to a risk of malnutrition and dehydration.29 Discharging with diet modifications may necessitate further community follow-up and, therefore, concur additional and, perhaps, unnecessary health care costs. Nonoral feeding was significantly associated with not developing pneumonia, although the effect size was much larger in the FEES group. Yet, previous research
SLP-LED FEES SERVICE: FUNCTIONAL OUTCOMES AFTER STROKE
has suggested a higher incidence of pneumonia in patients who are nonorally fed, with 1 study reporting that 41% of tube-fed patients after stroke developed pneumonia.30 It is possible that the FEES service allowed clinicians to be more assertive and confident in making earlier, objectively informed management decisions regarding nonoral feeding options.
Limitations The retrospective nature of the research means that data collection is reliant on access to, and accurate, documentation by health professionals. Similarly, the auditors are required to be both accurate and consistent in their interpretation. Both documentation and interpretation are susceptible to human error. Additionally, the between-group study design over 2 time periods means that confounding variables such as changes to hospital protocol or personnel cannot be accounted for.
Conclusions Introduction of an SLP-led FEES service was associated with a reduction in pneumonia rates in an inpatient stroke unit. Patients were discharged from hospital on less restrictive diets but following longer hospitalization and longer periods of nonoral feeding. Guidelines surrounding selection criteria of candidates for FEES may help to streamline the use of FEES and make such services more efficient. This would provide further strength to potential benefits of FEES in reducing pneumonia rates and improving functional outcomes for patients after stroke. Acknowledgment: The authors would like to acknowledge Waitemata DHB for their support in this study and for allowing researchers access their hospital databases.
References 1. Daniels SK, Foundas AL. Lesion localization in acute stroke patients with risk of aspiration. J Neuroimag 1999;9:91-98. 2. Hamdy S, Aziz Q, Rothwell JC, et al. Recovery of swallowing after dysphagic stroke relates to functional reorganization in the intact motor cortex. Gastroenterology 1998;115:1104-1112. 3. Martino R, Foley N, Bhogal S, et al. Dysphagia after stroke: incidence, diagnosis, and pulmonary complications. Stroke 2005;36:2756-2763. 4. Sellars C, Bowie L, Bagg J, et al. Risk factors for chest infection in acute stroke: a prospective cohort study. Stroke 2007;38:2284-2291. 5. Henon H, Godefroy O, Leys D, et al. Early predictors of death and disability after acute cerebral ischaemic event. Stroke 1995;26:392-398. 6. Reynolds PS, Gilbert L, Good DC, et al. Pneumonia in dysphagic stroke patients: effect on outcomes and identification of high risk patients. Neurorehabil Neural Repair 1998;12:15-21.
5
7. Katzan IL, Dawson NV, Thomas CL, et al. The cost of pneumonia after stroke. Neurology 2007;68:1938-1943. 8. Finlayson O, Kapral M, Hall R, et al. Risk factors, inpatient care, and outcomes of pneumonia after ischemic stroke. Neurology 2011;77:1338-1345. 9. Katzan IL, Cebul RD, Husak SH, et al. The effect of pneumonia on mortality among patients hospitalized for acute stroke. Neurology 2003;60:620-625. 10. Pikus L, Levine MS, Yang YX, et al. Videofluoroscopic studies of swallowing dysfunction and the relative risk of pneumonia. Am J Roentgenol 2003;180: 1613-1616. 11. Wilson RD. Mortality and cost of pneumonia after stroke for different risk groups. J Stroke Cerebrovasc Dis 2012; 21:61-67. 12. Colodny N. Inter-judge and intra-judge reliabilities in fiberoptic endoscopic evaluation of swallowing (FEES) using the penetration-aspiration scale: a replication study. Dysphagia 2002;17:308-315. 13. Langmore SE, Schatz MA, Olsen N. Endoscopic and videofluoroscopic evaluations of swallowing and aspiration. Ann Otol Rhinol Laryngol 1991;100:678-681. 14. Perie S, Laccourreye L, Flahault A, et al. Role of videoendoscopy in assessment of pharyngeal function in oro-pharyngeal dysphagia: comparison with videofluoroscopy and manometry. Laryngoscope 1998;108:1712-1716. 15. Wu C-H, Hsaio T-Y, Chen J-C, et al. Evaluation of swallowing safety with fiberoptic endoscope: comparison with videofluoroscopic technique. Laryngoscope 1997; 107:396-401. 16. Aviv JE. Prospective, randomized outcome study of endoscopy versus modified barium swallow in patients with dysphagia. Laryngoscope 2000;110:563-574. 17. Rao N, Bray SL, Chaudhuri G, et al. Gold-standard? Analysis of the videofluoroscopic and fiberoptic endoscopic swallow examinations. J Appl Res 2003; 3:89-96. 18. Tohara H, Nakane A, Murata S, et al. Inter- and intra-rater reliability in fiberoptic endoscopic evaluation of swallowing. J Oral Rehabil 2010;37:884-891. 19. Ajemian MS, Nirmul GB, Anderson MT, et al. Routine fiberoptic endoscopic evaluation of swallowing following prolonged intubation. Arch Surg 2001;136:434-437. 20. Ramsey DJC, Smithard DG, Kalra L. Early assessments of dysphagia and aspiration risk in acute stroke patients. Stroke 2003;34:1252-1257. 21. Leder SB, Espinosa JF. Aspiration risk after acute stroke: comparison of clinical examination and fiberoptic endoscopic evaluation of swallow. Dysphagia 2002; 17:214-218. 22. Miles A, Moore S, McFarlane M, et al. Comparison of cough reflex test against instrumental assessment of aspiration. Physiol Behav 2013;118:25-31. 23. Miles A, McLauchlan H, Zeng I, et al. Cough reflex testing in dysphagia following stroke: a randomized clinical trial. J Clin Med Res 2013;5:222-233. 24. Mann G, Hankey GJ, Cameron D. Swallowing function after stroke: prognosis and prognostic factors at 6 months. Stroke 1999;30:744-748. 25. Atherton M, Bellis-Smith N, Cichero J, et al. Texture modified foods and thickened fluids as used for individuals with dysphagia: Australian standardised labels and definitions. Nutr Diet, 2007; 64: S53-S76. 26. Child N, Fink J, Jones S, et al. New Zealand national acute stroke services audit: acute stroke care delivery in New Zealand. N Z Med J 2012;125:44-51.
6 27. Stroke Unit Trial Collaboration. Organised inpatient (stroke unit) care for stroke patients. Cochrane Database Syst Rev 2007;4:1-72. 28. Green E, McFarlane M, Miles A, et al. Staff perceptions following the introduction of a fibreoptic endoscopic evaluation of swallowing service for patients following stroke. Speech, Language and Hearing. In Revision.
L. BAX ET AL. 29. Keller H, Chambers L, Niezgoda H, et al. Issues associated with the use of modified texture foods. J Nutr, Health, Ageing 2012;16:195-200. 30. Langdon PC, Lee AH, Binns CW. High incidence of respiratory infection in nil-by-mouth tube-fed acute ischemic stroke patients. Neuroepidemiology 2009; 32:107-113.
