Simultaneous Videofluoroscopic Swallow Study and Modified EvansBlue Dye Procedure: An Evaluation of Blue Dye Visualization inCases of Known Aspiration

Susan L. Brady, MS, CCC-SLP,1 Cynthia D. Hildner, MS, CCC-SLP,1 and Brad F. Hutchins, MA, CCC-SLP2

1Marianjoy Rehabilitation Hospital, Wheaton, Illinois; and2Rehabilitation Foundation, Inc., Wheaton, Illinois, USA

Abstract. The reliability of the modified Evans blue dye(MEBD) test for the detection of aspirated materials inpatients with tracheostomy has been questioned. The vid-eofluoroscopic swallow study (VFSS) has been the stan-dard procedure used to detect aspiration, but there areknown risks and the VFSS is not always an availableevaluation option for aspiration detection. The purposeof the present study was to investigate the visualizationof blue tracheal secretions in cases of known aspirationas documented by the VFSS. Twenty consecutive simul-taneous MEBD study and VFSS were completed on pa-tients with tracheostomies at an acute rehabilitation hos-pital. Overall, the MEBD showed a 50% false-negativeerror rate. The MEBD identified aspiration in 100% ofpatients who aspirated more than trace amounts butfailed to identify aspiration of trace amounts (0%).

Key words: Aspiration — Evans blue dye — Trache-ostomy — Videofluoroscopy — Dysphagia — Degluti-tion — Deglutition disorders.

The patient with a tracheostomy tube may experiencedysphagia. The physiologic factors that may contributeto the dysphagia in the patient with a tracheostomy in-clude reduced laryngeal elevation, reduced pharyngealsensation, reduced cough response, and disuse atrophy oflaryngeal musculature [1–4]. These factors compromiseglottal protection and reduce swallowing efficiency, thusincreasing the risk of aspiration in the patient with atracheostomy [5]. A tracheostomy tube with a cuff may

be placed to prevent aspiration of secretions, aspirationof food, and gastric contents [1,6,7]. However, a patientmay still aspirate some food with a fully inflated cuff dueto severe dysphagia or a tracheostomy site placed toohigh [6]. Aspiration is defined as the entry of material(food or liquid) into the larynx below the true vocalcords. Aspiration can lead to serious medical complica-tions such as pneumonia, airway obstruction, and evendeath. The identification and management of aspirationin the patient with a tracheostomy is important and oftenposes difficult and challenging clinical decisions [2].

Cameron et al. [8] described the Evans blue dyetest in which four drops of a 1% solution of Evans bluedye was placed on the tongues of patients with a trache-ostomy to document the incidence of aspiration of theirsaliva. The test was considered positive for aspirationwhen blue dye was suctioned through the tracheostomytube. Cameron et al. [8] found that aspiration was de-tected in 69% of the patients who had a tracheostomy.The modified Evans blue dye (MEBD) procedure isbased on the original Evans blue dye test [8,9]. In theMEBD, the patient is given food and liquids that havebeen dyed blue. The test is considered positive for aspi-ration when the blue food or liquid is suctioned throughthe tracheostomy [1,9,10]. Adding blue dye (e.g., meth-ylene blue, Evans blue dye, blue food coloring) is alsocommonly used to detect aspiration of enteral feedings[3].

The MEBD has become a standard clinical tool inthe evaluation of patients with tracheostomy and sus-pected dysphagia because it offers the advantages ofeconomy, simplicity, and availability. Thompson-Henryand Braddock [9] questioned the reliability of this pro-cedure in their article and concluded that, even thoughaspiration can be determined with the MEBD, more care-ful validation of the procedure is required. In that study,

Correspondence to:Susan Brady, M.S., Office of Professional Devel-opment, Marianjoy Rehabilitation Hospital, 26 W 171 Roosevelt Road,Wheaton, IL 60187, USA

Dysphagia 14:146–149 (1999)

© Springer-Verlag New York Inc. 1999

the investigators completed a retrospective review of fivepatients who underwent both the MEBD and a video-fluoroscopic swallowing study (VFSS). They reportedthat in all five cases the MEBD failed to detect aspirationin tracheotomized patients. Criticism of that study in-cludes nonspecified subject selection, the interval timebetween the MEBD procedure and the VFSS, and thelack of documentation of severity of aspiration observedduring the VFSS [10,11]. Thompson-Henry and Brad-dock [9] also reported that there was no available infor-mation regarding the efficacy of the MEBD procedure orthe probability of false-negative readings.

Another method for detecting aspiration is theVFSS, which is the fluoroscopic image of the swallowrecorded on videotape. This type of examination allowsfor the evaluation of the movement patterns of the oralcavity, pharynx, and esophagus during the swallow [12].The VFSS has long been considered the “gold standard”in evaluating patients with swallowing disorders[1,12,13]. This tool is not always used because it is un-available or patient medical complications prevent or de-lay its administration. In addition, for patients in which aMEBD does not detect aspiration, the VFSS may not beordered.

The purpose of the present study was to evaluatethe visualization of blue tracheal secretions followingaspiration as documented by the VFSS during simulta-neous MEBD/VFSS procedures. At issue is whether theMEBD procedure is sensitive in detecting aspiration inpatients with a tracheostomy.

There were two null hypotheses: (a) the MEBDfalse-negative error rate for patients who aspirate in traceamounts of the bolus during a swallow examination isless than 0.20, and (b) the MEBD false-negative errorrate for patients who aspirate more than a trace amount ofthe bolus during the swallow examination is less than0.05.

Materials and Methods

Subjects

Twenty individuals with tracheostomy participated in this study. Allwere patients in an acute inpatient rehabilitation or subacute rehabili-tation program, who had known or suspected dysphagia and were re-ferred for a VFSS. Age range for the subjects was 13–81 years, and themean age was 59 years, with a median age of 64 years. Four subjectswere female and 16 were male. Eight patients had a diagnosis ofneuromuscular weakness secondary to respiratory distress and/or arrest,five had a diagnosis of cerebral vascular accident, four had a diagnosisof traumatic brain injury, and three had a diagnosis of spinal cordinjury. All subjects possessed the cognitive abilities to perform theVFSS and MEBD procedures and the ability to accept food into themouth. The types of tracheostomy tubes, including the fenestration,cuff, and occlusion status during the procedure, were documented. The

subjects comprised all patients admitted to the facility during a 4-monthperiod with the exception of two individuals, one who declined toparticipate in the study and one who decannulated himself during theVFSS procedure.

Procedures

The simultaneous VFSS/MEBD studies were conducted by a physicianand a speech-language pathologist. The speech-language pathologists(SLPs) who assisted in the study had at least 1 year of dysphagiaexperience and had successfully completed the organization’s VFSScompetency training program. The physicians were board certified inphysical medicine and rehabilitation and were credentialed for VFSS.These SLPs and physicians were trained in following the research studyprotocol.

The subjects underwent the simultaneous VFSS/MEBD proce-dure. Each VFSS was recorded on videotape. Three to five drops ofblue dye was added to each 4 oz. of the barium-and-food mixture toensure adequate visualization of the blue dye. For the purpose of thestudy, consistencies evaluated were pureed solids, thick nectar liquids,and thin liquids. Bolus sizes were measured with a syringe and givenin 1, 2, 3, 4, 5, and 10 cc amounts. If judged by the attending physicianand/or SLP that a bolus consistency or size would be unsafe, it wasdeferred and documented on the data collection form. A respiratorytherapist suctioned tracheal secretions after each incidence of aspirationas determined by the VFSS and after the VFSS was completed. Thesuctioning catheter was inserted into the tracheostomy tube betweentwo and four times with all subjects who demonstrated aspiration. Thespecimen was examined for blue discoloration against a white back-ground under full room lighting. During the VFSS, inflated tracheos-tomy cuffs were deflated if medical clearance had been given to do so.In two subjects, the cuff remain inflated during the study procedure butwas deflated for suctioning.

Data Analyses

Two SLPs, each experienced in conducting VFSS evaluations, inde-pendently reviewed the VFSS videotapes to rate the presence and theamount of aspiration. The raters were blinded to the results of thesimultaneous MEBD/VFSS. The amount of aspiration was defined aseither trace or greater than trace. Trace was defined, for the purpose ofthe present study, as a very thin line of barium that was presumed to beless than 10% of a bolus. The amount of aspiration was estimated fromthe lateral two-dimensional fluoroscopic image. Each study was ana-lyzed as a whole rather than rating each swallow for the amount ofaspiration. Interrater reliability calculations of the presence and amountof aspiration were completed between the study clinician and the twoindependent raters by using the kappa correlation coefficient.

Results

Twenty simultaneous MEBD/VFSS studies were com-pleted. Aspiration was present in 40% (8 of 20) of thestudies as documented by the VFSS. Five patients dem-onstrated silent aspiration, one patient was a “quiet” as-pirator (characterized by a delayed, reduced intensity,nonproductive cough), and two patients demonstratedimmediate nonproductive coughing. Laryngeal penetra-tion without subsequent aspiration was present in 15% (3

S.L. Brady et al.: Evaluation of Blue Dye Visualization 147

of 20) of the studies. Suctioning failed to detect the pres-ence of blue tracheal secretions in four of the eight pa-tients who aspirated. Thus, the MEBD showed a 50%false-negative error rate in aspiration detection whencompared with the simultaneous VFSS.

The total amount of aspiration as documented bythe VFSS was classified as trace or greater than traceaspiration. Of the VFSS that had documented aspiration,50% (4 of 8) demonstrated greater than trace aspirationand 50% (4 of 8) documented trace aspiration. Of thestudies that showed greater than trace aspiration, 100%demonstrated blue tracheal secretions with suctioning.None (0%) of the studies that showed trace aspirationpresented with blue tracheal secretions with suctioning.None of the subjects who demonstrated laryngeal pen-etration without subsequent aspiration presented withblue tracheal secretions with suctioning.

Interrater reliability was calculated between thethree raters regarding the presence and amount of aspi-ration documented on the VFSS. The kappa correlationcoefficient was calculated between raters 1 and 2, raters2 and 3, and raters 1 and 3. The measure of agreement onamount of aspiration between raters 1 and 2 was 1.0 (p4 0.05). The measure of agreement between raters 2 and3 and between raters 1 and 3 was 0.750 (p 4 0.028). Theagreement between the two latter pairs of raters, althoughrelatively low, reflected agreement on seven of the eightstudies. The interrater reliability between the three pairsof raters for the presence of aspiration was 1.0, reflectingperfect agreement.

Aspiration was present with bolus sizes between1 and 4 cc and with thick liquid barium, thin liquidbarium, and puree-consistency barium. Of the eight stud-ies where aspiration was documented, seven of the eightpatients had a nonfenestrated tracheostomy tube and fourpatients had a cuff, with two inflated during the study.Types of tracheostomy tubes included three Bivona,three Shiley, and two Portex. Occlusion status includedthree studies with a capped tracheostomy tube, threestudies with the Passy-Muir speaking valve, and twostudies with the cuff inflated while the patient was on aventilator.

Discussion

Although the number of subjects in this study was small,the results indicate that the MEBD was not sensitiveenough for the detection of aspiration in trace amounts.However, for patients who aspirated greater than traceamounts, MEBD was sensitive for determining the pres-ence of aspiration.

The amount of aspiration as documented on theVFSS appeared to be the contributing factor as to wheth-

er the blue tracheal secretions were visible with suction-ing. A recognized limitation of this study is the lackof consensus in the field of dysphagia regarding the abil-ity to judge the quantity of the aspirated portion of thebolus. Although this lack of consensus is present, clini-cal practice often involves estimating the aspirated ma-terial as being trace or greater than trace [14–16]. Thepresence of a cuff, the fenestration status, and the con-sistency of the material aspirated did not appear to affectthe ability to detect the presence of blue tracheal secre-tions.

A previous study has questioned the reliability ofthe MEBD by comparing the results of MEBD withthose of VFSS completed at different times [9]. Conduct-ing the MEBD/VFSS simultaneously allowed the exam-iner to evaluate how effective the visualization of bluetracheal secretion was in cases of known aspiration asdocumented by the VFSS. Also, by conducting theMEBD/VFSS simultaneously, the same swallow wasevaluated at one time rather relying on inferences fromtwo separate events.

The usefulness for determining appropriate rec-ommendations regarding swallowing with the MEBDmust be approached with caution. The present resultssuggest that the MEBD could be best viewed as a screen-ing tool for determining the presence of greater thantrace aspiration. Because trace aspiration was not de-tected with the MEBD in the present study, a VFSS maybe indicated after a negative MEBD to rule out traceaspiration.

Future studies might include investigating thedifferences in detection rates of trace aspiration betweenand within subjects and replicating the study with a largenumber of subjects.

Acknowledgment.This work was supported by the Dr. Ralph and Mar-ian Falk Medical Research Foundation.

References

1. Logemann J:Evaluation and Treatment of Swallowing Disor-ders, 2nd ed. Austin: Pro-ed, 1998

2. Dettelbach M, Gross R, Mahlmann J, Eibling D: Effect of thePassy-Muir valve on aspiration in patients with tracheostomy.Head Neck 17:297–302, 1995

3. Potts R, Zaroukian M, Guerrero P, Baker C: Comparison of bluedye visualization and glucose oxidase test strip method for de-tecting pulmonary aspiration of enteral feedings in intubatedadults.Chest 103:117–121, 1993

4. Tolep K, Leonard G, Criner G: Swallowing dysfunction in pa-tients receiving prolonged mechanical ventilation.Chest109:167–172, 1996

5. Bone D, Davis J, Zuidema G, Cameron J: Aspiration pneumo-nia. Ann Thorac Surg 18:30–36, 1974

148 S.L. Brady et al.: Evaluation of Blue Dye Visualization

6. Bach J, Alba A: Tracheostomy ventilation: a study of efficacywith deflated cuffs and cuffless tubes.Chest 19:679–683, 1990

7. Elpern E, Scott M, Petro L, Ries M: Pulmonary aspiration inmechanically ventilated patients with tracheostomies.Chest105:563–566, 1994

8. Cameron J, Reynolds J, Zuidema G: Aspiration in patients withtracheotomies.Surg Gynecol Obstet 136:68–70, 1973

9. Thompson-Henry S, Braddock B: The modified Evan’s blue dyeprocedure fails to detect aspiration in the tracheotomized patient:five case reports.Dysphagia 10:172–174, 1995

10. Tippett D, Siebens A: Reconsidering the value of the modifiedEvan’s blue dye test: a comment on Thompson-Henry and Brad-dock (1995).Dysphagia 11:78–79, 1996

11. Leder S: Comment on Thompson-Henry and Braddock: themodified Evan’s blue dye procedure fails to detect aspiration in

tracheostomized patient: five case reports.Dysphagia 11:80–81, 1996

12. Logemann J:Evaluation and Treatment of Swallowing Disor-ders. San Diego: College-Hill Press, 1983

13. Gelfand D, Richter J:Dysphagia: Diagnosis and Treatment.New York: Igaku-Shoin, 1989

14. Arvedson J, Rogers B, Buck G, Smart P, Msall M: Silent aspi-ration prominent in children with dysphagia.Int J Pediatr Oto-rhinolaryngol 28:173–181, 1994

15. Logemann J, Pauloski B, Rademaker A, Colangelo L: Super-supraglottic swallow in irradiated head and neck cancer patients.Head Neck 19:535–540, 1997

16. Newton H, Newton C, Pearl D, Davidson T: Swallowing assess-ment in primary brain tumor patients with dysphagia.Neurology44:1927–1932, 1994

S.L. Brady et al.: Evaluation of Blue Dye Visualization 149