HSC Section 6 Nov2016 Green Book

Vojko Djukic, et al

Stroboscopy in Detection of Laryngeal Dysplasia

TABLE 3. Stroboscopic Signs for Patients With Severe Dysplasia Before Treatment and After 12 Months of Follow-Up or Before Retreatment

Severe Dysplasia

Stroboscopic Signs

Before Treatment (%)

After 12 Months (%)

Sig.

Glottic occlusion

1.000

Sufficient Insufficient

25/33 (75.8) 8/33 (24.2)

24/33 (72.7) 9/33 (27.3)

Phase symmetry

—

Symmetrical Asymmetrical

0/33 (0)

19/33 (57.6) 14/33 (42.4)

33/33 (100.0)

Periodicity

0.000*

Regular Irregular

1/33 (3.0)

19/33 (57.6) 14/33 (42.4)

32/33 (97.0)

Amplitude

0.000*

Normal

1/33 (3)

19/33 (57.6) 14/33 (42.4)

Decreased Increased

29/33 (87.9)

3/33 (9.1)

0/33 (0)

Mucosal wave

0.000*

Normal with 30–50% lateral travel

1/33 (3) 0/33 (0)

19/33 (57.6)

Increased with lateral travel greater than 50% Decreased with lateral travel less than 30%

0/33 (0)

32/33 (97)

14/33 (42.4)

Nonvibratory segment

0.013*

Presence Absence

18/33 (54.5) 15/33 (45.5)

8/33 (24.2) 25/33 (75.8)

Abbreviation: Sig., statistical significance. * P < 0.05.

moderate dysplasia, and 36.4% of the patients with severe dys- plasia. Type II cordectomy was performed in 35.9% of the pa- tients with mild dysplasia, 34.6% of the patients with moderate dysplasia, and 57.6% of the patients with severe dysplasia. Type III cordectomy was performed in only two (6%) patients with severe dysplasia. Recurrence of the disease occurred after 8 months in six patients, after 9 months in five patients, after 10 months in four patients, after 11 months in eight patients, and after 12 months in five patients. Most patients with recurrence were in a group with moderate dysplasia ( Table 5 ). Disease progression was noted in 10 patients with recurrence. Invasive carcinoma developed in four patients: one from group with moderate dys- plasia and three from group with severe dysplasia. Multivariate regression analysis was performed to determine how factors such as cordectomy type and stroboscopic signs (glottic occlusion, phase symmetry, periodicity, amplitude of vibrations, mucosal wave, and nonvibrating segment) corre-

lated with the histopathologic verification of different types of dysplasia ( Table 6 ). Some factors, such as the type of cordec- tomy and the existence of nonvibrating segment, were set apart from others and were proven to be significantly different in var- ious levels of dysplasia ( P < 0.05). Logistic regression was used to determine whether the recur- rence and progression of the disease could be anticipated by cordectomy type and stroboscopic signs (glottic occlusion, phase symmetry, periodicity, amplitude of vibrations, mucosal wave, and nonvibrating segment) ( Table 7 ). Of all the analyzed factors, only the amplitude of vocal fold vibrations in group with mild dysplasia ( P ¼ 0.03) was statistically significant for recurrent disease. In this group, some factors indicated a higher risk of recurrence but not statistically significant enough. Pa- tients with asymmetry in vocal fold vibrations and irregular vo- cal fold vibrations, with abnormal amplitude of vocal fold vibrations, and the existing nonvibrating segment were at higher risk of recurrence. None of the factors was statistically

TABLE 4. Treatment Modalities for the Different Grades of Dysplasia Cordectomy Types Mild Dysplasia (%)

Moderate Dysplasia (%)

Severe Dysplasia (%)

All Patients (%)

I

34/53 (64.1) 19/53 (35.9)

17/26 (65.4) 9/26 (34.6)

12/33 (36.4) 19/33 (57.6)

63/112 (56.2) 47/112 (42.0)

II

III

0/53 (0)

0/26 (0)

2/33 (6.0)

2/112 (1.8)

33