Gold-Standard?

Analysis of the Videofluoroscopic and Fiberoptic Endoscopic
Swallow Examinations

 

Noel Rao, MD*

Susan L. Brady, MS, CCC-SLP†

Gouri Chaudhuri, MD*

Joseph J. Donzelli, MD‡

Michele W. Wesling, MS, CCC-SLP†

 

*Rehabilitation Medicine Clinic,

†Department of Speech-Language Pathology

Marianjoy Rehabilitation Hospital

Wheaton, Illinois

‡Otolaryngology Head & Neck Surgery, LTD

Carol Stream, Illinois

 

Work was completed at the Marianjoy Rehabilitation Hospital, Wheaton, Illinois.

This work was supported by the Dr. Ralph and Marian Falk Medical Research Foundation.

This paper was accepted for a poster presentation at the Academy of Physical Medicine and Rehabilitation Annual Assembly, November 21-24, 2002, Orlando, FL.

 

KEY WORDS: swallow, videofluoroscopy, rehabilitation, endoscopy

ABSTRACT

Objective: The goal of this study was to determine sensitivity and specificity values for laryngeal penetration, tracheal aspiration, and pharyngeal residue for both the videofluoroscopic (VFSS) and fiberoptic endoscopic (FEES) swallowing examinations. Sensitivity and specificity values were calculated twice, first using the VFSS as the gold standard, then using the FEES as the gold standard. Percentage of agreement for laryngeal penetration, aspiration, pharyngeal residue, diet recommendations, and compensatory strategies were also calculated.

Study Design: Prospective, consecutive design, set in a freestanding rehabilitation hospital.

Participants: Eleven patients who underwent simultaneous VFSS and FEES.

Interventions: Not applicable.

Main Outcome Measures: Presence or absence of laryngeal penetration, tracheal aspiration, and/or pharyngeal residue.

Results: When the VFSS was used as the gold standard, sensitivity of the FEES for laryngeal penetration was 0.87, aspiration 0.96, and pharyngeal residue 0.68. Specificity of the FEES for laryngeal penetration was 1.0, aspiration 1.0, and pharyngeal residue 0.98. When the FEES was used as the gold standard, sensitivity of the VFSS for laryngeal penetration was 1.0, aspiration 1.0, and pharyngeal residue 0.96. Specificity of the VFSS for laryngeal penetration was 0.58, aspiration 0.63, and pharyngeal residue 0.78. Agreement for the presence or absence of pharyngeal residue was 84.38%, laryngeal penetration 89.58%, and tracheal aspiration 96.69%. Diet recommendations were in agreement 100%, and compensatory swallowing strategies were in agreement 82%.

Conclusions: The sensitivity values were higher when FEES was used as the gold standard, and the specificity values were higher when VFSS was used as the gold standard. The one exception is that the sensitivity values for aspiration, regardless of whether VFSS or FEES was used as the gold standard, were similar.

INTRODUCTION

Many authors have suggested the “gold” or “criterion” standard for the instrumental assessment of the swallow as the videofluoroscopic swallowing evaluation (VFSS).1–7 However, an evidence report by the Agency for Health Care Policy and Research (AHCPR) branch of Health and Human Services concluded that currently in the literature there was no evidence to support that one instrumental assessment of the swallow provided more useful information than another.8 They concluded that neither the VFSS nor the fiberoptic endoscopic examination of the swallow (FEES) could serve as a perfect “gold” standard for the detection of aspiration because both examinations yield false-positive and false-negative results. Furthermore, they reported that without a third, more reference standard, the VFSS and FEES could not be compared with each other. Although the controversy exits as to whether the VFSS or FEES should be considered the gold standard, they remain the two most common types of instrumental assessments of the swallow.5,9–12

Several authors have reported the advantages and disadvantages of both the VFSS and FEES.5,13–20 Additionally, previous research has also compared the two examinations.16,17,21,22 Some studies have compared the two examinations on patients with dysphagia when the examinations were completed at separate times.16,17,22 However, because individuals with dysphagia often show variable abilities to swallow, an inherent weakness of studies that are not conducted simultaneously is that they are actually comparing two different behaviors.10 Furthermore, previous researchers have conducted sensitivity and specificity analysis for the FEES using the VFSS as the “gold” or “criterion” standard.23 According to Langmore,23 this step was important because the FEES was considered the “new” examination; to establish its validity, it needed to be compared to the “old” examination of the VFSS. An additional acknowledged limitation of these studies is that when the old examination represents the “truth,” it is very difficult to show that the “new” procedure is as valid.23

The purpose of this study was to determine sensitivity and specificity values for laryngeal penetration, tracheal aspiration, and pharyngeal residue for both the VFSS and FEES. Because sensitivity and specificity values require a gold standard, the values were calculated twice, first using the VFSS as the gold standard then using the FEES as the gold standard. Calculating the sensitivity and specificity twice is important to establish the validity of each examination. The current literature shows no evidence-based indication of the “gold” or “criterion” standard for the instrumental assessment of the swallow.

METHOD

Subjects

Eleven consecutive patients underwent the VFSS and FEES procedures simultaneously. Criteria selection included subjects who had a suspected laryngeal or pharyngeal abnormality or dysphonia and required an evaluation of the oral, pharyngeal, and esophageal phases of swallowing.

Age range was from 29 to 77 years, with a mean age of 50 (standard deviation ± 18.10 years). Nine subjects were men and two were women. See Table 1 for additional information on subject demographics. All subjects possessed the cognitive abilities to perform the VFSS and FEES procedures and the ability to accept food into the mouth.

Procedures

The subjects underwent simultaneous VFSS and FEES. Equipment used during the procedures included the C-arm fluoroscopy system, the Olympus flexible endoscope, the Olympus CKL halogen light source, and the Elmo one chip camera. All of the examinations were recorded on videotape. The examiners were not blinded to the results of each examination. Each bolus presentation was evaluated for the presence of laryngeal penetration, aspiration, and pharyngeal residue. Bolus size and consistency were not controlled.

A physiatrist and speech-language pathologist conducted all the VFSS. The physiatrist performing the VFSS had successfully completed a VFSS training program administered by a radiologist and had received credentials to complete VFSS at this institution. An otolaryngologist and speech-language pathologist performed the FEES. For the FEES procedure, the flexible endoscope was passed transnasally to the hypopharynx to provide a full view of the larynx. The endoscope was initially placed in the high position to just above the tip of the epiglottis before and during the swallow and then advanced to the low position to just above the vocal folds after the swallow to evaluate for the presence of laryngeal penetration or aspiration.

Three to five drops of blue dye were added to each 4 ounces of the barium-and-food mixture to ensure adequate visualization of the bolus. The ratio of barium to food/liquid was controlled. A thin liquid barium mixture consisted of one third liquid polibar barium and two thirds water, nectar-thick liquid consisted of one half liquid polibar barium and one half nectar liquid, and extra-thick liquid consisted of 2 teaspoons of liquid polibar barium and 4 oz of applesauce. Each 4 oz of pureed and solid foods were mixed with 3 teaspoons of powder barium.

An independent experienced rater established interrater reliability for both VFSS and FEES. The rater was blinded to the results of the examination and independently rated each swallow for both the VFSS and FEES. The interrater agreement for the presence of laryngeal penetration, tracheal aspiration, and pharyngeal residue for both examinations as compared with the initial rating was at 90% or higher for each parameter.

DATA ANALYSIS

The sensitivity and specificity values were calculated. To determine the sensitivity and specificity values, you need to establish one measure as a gold standard. For the purpose of this investigation, the sensitivity and specificity values were calculated twice, first using the VFSS as the gold standard, then using the FEES as the gold standard. To further evaluate the results, 2 x 2 contingency tables were used to compare the results for laryngeal penetration, aspiration, and pharyngeal residue during the FESS/VFSS.To correct for chance agreement on the contingency tables, the kappa correlation and Fisher’s exact tests were also completed. Percentage of agreement between the two examinations for laryngeal penetration, aspiration, pharyngeal residue, diet recommendations, and compensatory strategies were calculated.

RESULTS

A total of 100 boluses were presented across the 11 subjects. Four boluses were excluded from the study, as they were unable to be evaluated by both the FEES and VFSS secondary to the subject’s motion of the examination field.

Sensitivity and specificity values for the FEES when the VFSS was used as the gold standard are represented in Table 2. The sensitivity and specificity values for the VFSS when the FEES was used as the gold standard are represented in Table 3. The sensitivity values were higher when the FEES was used as the gold standard, and the specificity values were higher when the VFSS was used as the gold standard. The one exception was that the sensitivity values for aspiration were similar, regardless of which examination was used as the gold standard.

The results of the contingency tables for laryngeal penetration, aspiration, and pharyngeal residue are summarized in Table 4. The correlations indicate a moderate association between the FEES and VFSS. The most similar agreement between the FEES and VFSS was with aspiration and the least agreement was with pharyngeal residue.

Percentage of agreement between the two examinations for laryngeal penetration, aspiration, and pharyngeal residue was calculated and is shown in Table 5. Laryngeal penetration was present on 25% (24/96) of the boluses. The FEES and VFSS were in agreement for the presence or absence of laryngeal penetration 89.58% (86 of 96) of the time. On the 10 boluses that were not in agreement, the FEES detected laryngeal penetration all 10 times, whereas the VFSS did not. Aspiration was present on 9.4% (9/96) of the boluses. Agreement between the FEES and VFSS for the presence or absence of aspiration was 96.69% (93/96). On the three boluses that were not in agreement, the FEES detected aspiration all three times whereas the VFSS did not. Pharyngeal residue was present on 69.79% (67/96) of the boluses. Agreement between the FEES and VFSS for the presence or absence of pharyngeal residue was 84.38% (81 of 96). On the 15 boluses that were not in agreement, the FEES detected pharyngeal residue 14 times when the VFSS did not and the VFSS detected pharyngeal residue 1 time when the VEES did not.

Seven of the subjects presented with some type of laryngeal or pharyngeal abnormality as identified by the FEES, and 3 subjects presented with an anatomic abnormality as identified by the VFSS (Table 6). Excessive oropharyngeal secretions that had accumulated in the hypopharynx were identified in 6 patients by the FEES procedure. Diet recommendations were in agreement 100% of the time, and compensatory swallowing safety strategies were in agreement 82% (9 of 11) of the time (Table 6).

DISCUSSION

The results of this study support previous research in showing that both FEES and VFSS are valuable procedures for evaluating dysphagia and have been shown to be successful at the diagnosis and management of dysphagia.5,15–20 In this current investigation, the sensitivity values for aspiration were similar, regardless of whether the VFSS or FEES measure was used as the gold standard. The sensitivity value (the true-positive rate), which is the test’s sensitivity to detecting a disorder when it is actually present, was higher when FEES was used as the gold standard for laryngeal penetration and pharyngeal residue. The specificity value is the true-negative rate and answers the question of whether the test is sensitive to the construct being measured or whether it picks up other constructs as well. The specificity values were higher for laryngeal penetration, aspiration, and pharyngeal residue when the VFSS was used as the gold standard.

This current study also provided support that the VFSS and FEES are equally effective, comparable, valid instrumental procedures for swallowing and both deserve to be considered the “gold” standard. A moderate association between the FEES and VFSS was demonstrated in this current study, with the most similar values shown for aspiration. The choice of which instrumental assessment should be used should be dictated by clinical indications, equipment availability, and clinical expertise of the evaluators. Furthermore, it is important to understand and recognize the strengths and weaknesses of each diagnostic procedure. From a clinical standpoint, the VFSS provides greater information during the oral phase of swallowing, and the FEES may be uncomfortable for young children and individuals with severe cognitive disorders who may also be agitated.13 The FEES, however, may be able to provide the examiner with additional information on anatomy and physiology of the pharynx and larynx that the VFSS would be unable to provide. We propose that the VFSS and FEES can be used to complement each other. The results of this study suggest that patients who are referred for an instrumental assessment of the swallow and the clinical indicators of a dysphonia, suspected laryngeal or pharyngeal abnormality, presence of a tracheotomy tube or difficulty managing secretions may be best evaluated using FEES. Additionally, patients referred for an instrumental assessment of dysphagia along with the clinical indicators of a suspected esophageal abnormality might be best evaluated using the VFSS.

An advantage of this investigation was that VFSS and FEES were conducted simultaneously. Previous investigators attempted to show the sensitivity and specificity values of the FEES, using the VFSS as the gold standard, with the examinations completed at separate times.16,22 In this current investigation, the examiners were allowed to evaluate the same swallow under both diagnostic tools at the same time, rather than relying on inference from two separate events. The results of this study suggest that the most effective means for evaluating the validity of both the VFSS and FEES is by conducting simultaneous examinations.

A recognized limitation of this study is that the dysphagia evaluation team introduced appropriate swallowing strategies to minimize the risk of laryngeal penetration, aspiration, and pharyngeal residue to the patients. If this safety protocol was not in place, theoretically more patients may have shown increased symptoms of dysphagia. Because this safely protocol was in place for all patients, however, each patient was presented with different bolus types and amounts. Direction for future research may include replicating this study with a larger sample size with the patients undergoing identical swallowing protocols for bolus type and amount.

It was interesting to note that the overall recommendations for diet level remained unchanged even though minor differences in laryngeal penetration, aspiration, and pharyngeal residue detection rate were present. However, for two subjects in this current study, the recommendations for swallowing safety strategies were different. One difference was based on the results of the VFSS and other was based on the results of the FEES. Additionally, the FEES was able to provide additional diagnostic information by identify laryngeal abnormalities in six of the patients. However, it was unable to detect esophageal abnormalities that the VFSS was able to show in two of the patients. In only one patient could both VFSS and FEES be used to identify the same abnormality. In this case, the abnormality was prevertebral swelling after an anterior cervical spinal fusion surgery.

CONCLUSIONs

The FEES and VFSS are both valuable procedures for evaluating dysphagia and show good agreement regarding laryngeal penetration, aspiration, pharyngeal residue, diet recommendations, and compensatory strategies. The sensitivity values are higher when the FEES is used as the gold standard and the specificity values are higher when the VFSS is used as the gold standard. The one exception is that the sensitivity values for aspiration were similar regardless of whether which measure was used as the gold standard. Because both examinations yield valuable information in the assessment and clinical management of dysphagia, their selection should be clinically mandated.

ACKNOWLEDGMENTS

The Dr. Ralph and Marian Falk Medical Research Trust funded this study. The authors would also like to acknowledge Barbara Kremer, PhD, who provided statistical support for this study.

REFERENCES

1. Robbins J, Coyle J, Rosenbek J, et al: Differentiation of normal and abnormal airway protection during swallowing using the penetration-aspiration scale. Dysphagia 14:228–232, 1999.

2. Lindbichler F, Raith J, Uggowitzer M, Wuttge-Hannig A: Functional imaging of the pharynx using electron beam tomography. Dysphagia 13:101–104, 1998.

3. Mari F, Matei M, Ceravolo MG, et al: Predictive value of clinical indices in detecting aspiration in patients with neurological disorders. J Neurol Neurosurg Psychiatry 63:456–460, 1997.

4. Collins MJ, Bakheit AM: Does pulse oximetry reliably detect aspiration in dysphagic stroke patients? Stroke 28:1773–1775, 1997.

5. Wu CH, Hsiao TY, Chen JC, et al: Evaluation of swallowing safety with fiberoptic endoscope: Comparison with videofluoroscopic technique. Laryngoscope 107:396–401, 1997.

6. Yang WT, Loveday EJ, Metreweli C, Sullivan PB: Ultrasound assessment of swallowing in malnourished disabled children. Br J Radiol 70:992–994, 1997.

7. Cordaro MA, Sonies BC: An image processing scheme to quantitatively extract and validate hyoid bone motion based on real-time ultrasound recordings of swallowing. IEEE Trans Biomed Eng 40:841–844, 1993.

8. Eisenberg JM, Kamerow DB: Diagnosis and Treatment of Swallowing Disorders (Dysphagia) in Acute-Care Stroke Patients. Evidence Report/Technology Assessment Number 8. Agency for Health Care Policy and Research U.S. Department of Health and Human Services AHCPR Publication No. 99-E024, Rockville, MD, 1999.

9. Donzelli J, Brady S, Wesling M, Craney M: Simultaneous modified Evans blue dye procedure and video nasal endoscopic evaluation of the swallow. Laryngoscope 111:1746–1750, 2001.

10. Brady S, Hildner C, Hutchins B: Simultaneous videofluoroscopic swallowing study and modified Evan’s blue dye procedure: an evaluation of blue dye in cases of known aspiration. Dysphagia 14:146–149, 1999.

11. Ekberg O: Radiologic evaluation of swallowing. In: Groher ME, ed: Dysphagia Diagnosis and Management, 2nd ed. Stoneham, MA: Butterworth-Heinemann; 1992:163–189.

12. Splaingard ML, Hutchins B, Sulton LD, Chaudhuri G: Aspiration in rehabilitation patients: Videofluoroscopic vs. bedside clinical assessment. Arch Phys Med Rehabil 69:637–640, 1986.

13. Logemann J: Instrumental techniques for the study of swallowing. In: Logemann J, ed: Evaluation and Treatment of Swallowing Disorders, 2nd ed. Austin, TX: Proed; 1998:53–70.

14. Gelfand D, Richter J. Dysphagia: diagnosis and treatment. New York: Igaku-Shoin; 1989.

15. Langmore SE, Schatz K, Olsen N: Fiberoptic endoscopic examination of swallowing safety: A new procedure. Dysphagia 2: 216–219, 1988.

16. Langmore SE, Schatz K, Olsen N: Endoscopic and videofluoroscopic evaluations of swallowing and aspiration. Ann Otol Rhinol Laryngol 100:678–681, 1991.

17. Bastian R: The videoendoscopic swallow study: An alternative and partner to the videofluoroscopic swallow study. Dysphagia 8:359–367, 1993.

18. Kidder T, Langmore S, Martin B: Indications and techniques of endoscopy in evaluation of cervical dysphagia: Comparison with radiographic techniques. Dysphagia 9:256–261, 1994.

19. Aviv JE, Kaplan ST, Thomson JE, et al: The safety of flexible endoscopic evaluation of swallowing with sensory testing (FEESST): An analysis of 500 consecutive evaluations. Dysphagia 15: 39–44, 2000.

20. Leder SB: Serial fiberoptic endoscopic swallowing evaluations in the management of patients with dysphagia. Arch Phys Med Rehabil 79:1264–1269, 1998.

21. Schima W, Denk D, Schober E: Videofluoroscopic and videoendoscopic swallowing studies: Complementary methods for assessment of dysphagia (abstract). Dysphagia 15:115, 2001.

22. Perie S, Laccourreye L, Flahault A, et al: Role of videoendoscopy in assessment of pharyngeal function in oropharyngeal dysphagia: comparison with videofluoroscopy and manometry. Laryngoscope 108: 1712–1716, 1998.

23.           Langmore S: FEES: Building a foundation for research in endoscopy. In: Perspectives on Swallowing and Swallowing Disorders (Dysphagia). Washington, DC: American Speech-Language-Hearing Association, Division 13, 11: 6–8.

 

 

Table 1. Subject Demographics

 

Subject                  Age (y)                   Gender                              Diagnosis                       Boluses (n)

1                               60                         Male                    Anoxic encephalopathy                  16

2                               77                         Male                    Anoxic encephalopathy                   7

3                               63                         Male                                   CVA                                 12

4                               17                       Female                      Closed head injury                       5

5                               29                         Male                        Closed head injury                       8

6                               29                         Male                              Dysphagia                             6

7                               49                         Male                              Dysphagia                            18

8                               65                         Male                                   CVA                                  8

9                               59                         Male                                  ACSS                                 7

10                             50                         Male                   Eaton-Lambert syndrome                 1

11                             51                       Female                                 CVA                                  8

 

 

Table 2. VFSS as the Gold Standard, Sensitivity and Specificity of the FEES

 

Bolus Flow                  Sensitivity        Specificity

 

 

Laryngeal penetration    0.87                  1.0

Aspiration                       0.96                  1.0

Pharyngeal residue        0.68                 0.98

 

Table 3. FEES as the Gold Standard, Sensitivity and Specificity of the VFSS

Bolus Flow                  Sensitivity        Specificity

 

Laryngeal penetration     1.0                  0.58

Aspiration                        1.0                  0.63

Pharyngeal residue        0.96                 0.78

 

Table 4. Results of Contingency Tables for Laryngeal Penetration, Aspiration, and Pharyngeal Residue

 

                                        Kappa          Fisher’s

FEES/VFSS                    Correlation    Exact Test

 

Laryngeal penetration       0.68          P = 0.000

Aspiration                          0.75          P = 0.000

Pharyngeal residue           0.67          P = 0.000

 

 

Table 5. Agreement of VFSS and FEES

 

                 Laryngeal                         Pharyngeal
Subject     Penetration     Aspiration     Residue

 

1                   15/16             14/16          14/16

2                     5/7                 7/7              4/7

3                    9/12              12/12           7/12

4                     4/5                 5/5              4/5

5                     8/8                 8/8              8/8

6                     6/6                 6/6              6/6

7                   16/18             18/18          15/18

8                     8/8                 8/8              7/8

9                     6/7                 6/7              7/7

10                   1/1                 1/1              1/1

11                   8/8                 8/8              8/8

 

 

Table 6. Abnormalities and Recommendations

 

                                                                                                                                         Agreement of

                          Abnormality                       Abnormality                   Agreement            Compensatory

Subject           Detected By VFSS               Detected By FEES             of Diet Level               Strategy

1                None                                   None                                            Yes                Yes

 

2                None                                   Left vocal cord paresis,              Yes                Yes

                                                             small posterior glottal chink                               

 

3                None                                   Right paralyzed vocal cord          Yes                No; addition of

                                                             to eliminate penetration                                      a chin tuck

                                                             as detected by the FEES

                                                                                                                                       

4                None                                   Swollen mucosa of                      Yes                Yes

                                                             pyriform with floppy arytenoid                          

 

5                None                                   Posterior glottal chink                   Yes                Yes

 

6                T6-T7                                  None                                            Yes                No; addition of

                  Esophageal                                                                                                    a liquid wash

                  narrowing                                                                                                      because of the

                                                                                                                                        narrowing of the

                                                                                                                                        esophagus per

                                                                                                                                        VFSS results.

 

7                Esophageal                         None                                            Yes                Yes

                  abnormality with

                  backflow of
                  bolus into the pharynx

 

8                None                                   Left paralyzed                             Yes                Yes

                                                             vocal cord

 

9                Prevertebral                        Prevertebral                                 Yes                Yes

                  swelling                              swelling                                                             

 

10              None                                   None                                            Yes                Yes

 

11              None                                   Bilateral vocal                              Yes                Yes

                                                             cord paralysis