Unilateral vocal fold paralysis: can laryngoscopy predict recovery? A prospective study

Classifying Vocal Folds Fixation from Endoscopic Videos with Machine Learning

Vocal folds motility evaluation is paramount in both the assessment of functional deficits and in the accurate staging of neoplastic disease of the glottis. Diagnostic endoscopy, and in particular videoendoscopy, is nowadays the method through which the motility is estimated. The clinical diagnosis, however, relies on the examination of the videoendoscopic frames, which is a subjective and professional-dependent task. Hence, a more rigorous, objective, reliable, and repeatable method is needed. To support clinicians, this paper proposes a machine learning (ML) approach for vocal cords motility classification. From the endoscopic videos of 186 patients with both vocal cords preserved motility and fixation, a dataset of 558 images relative to the two classes was extracted. Successively, a number of features was retrieved from the images and used to train and test four well-grounded ML classifiers. From test results, the best performance was achieved using XGBoost, with precision = 0.82, recall = 0.82, F1 score = 0.82, and accuracy = 0.82. After comparing the most relevant ML models, we believe that this approach could provide precise and reliable support to clinical evaluation.Clinical Relevance- This research represents an important advancement in the state-of-the-art of computer-assisted otolaryngology, to develop an effective tool for motility assessment in the clinical practice.

Factors Predicting the Recovery of Unilateral Vocal Fold Paralysis After Thyroidectomy

BACKGROUND

We used voice analysis and clinicopathological factors to explore the prognosis of unilateral vocal fold paralysis after thyroid surgery.

METHODS

The medical records of 63 females who developed unilateral vocal fold paralysis after thyroidectomy were reviewed. All patients were divided into two groups: those who recovered from vocal fold paralysis and those who did not. We analyzed clinical parameters and voice analysis results in a search for correlations with recovery from paralysis.

RESULTS

Of the 63 patients, 37 (58%) recovered from paralysis. A small tumor size, incomplete paralysis, the absence of arytenoid tilting, no compensatory movement of the normal side, lower postoperative shimmer, a higher postoperative maximum phonation time (MPT), and lower postoperative subglottic pressure correlated significantly with recovery from vocal fold paralysis. Multivariate analysis confirmed that the absence of compensatory movement of the normal side on videostroboscopy was independently prognostic. A postoperative MPT of 6.86 appeared to be optimal for prediction of recovery. Most patients recovered within 6 months, but those with incomplete paralysis recovered about 3 months earlier. At the 12-month follow-up, the thyroidectomy-related voice questionnaire scores had returned to preoperative values in only 12 patients (19.0%); 51 patients (81.0%) did not fully recover.

CONCLUSION

Compensatory movement of the normal side evident on videostroboscopy was a poor prognostic factor. Voice analysis can be helpful in counseling vocal fold paralysis patients after thyroidectomy, and early intervention may be considered in patients who are expected to have a poor prognosis.

Paclitaxel inhibits post-traumatic recurrent laryngeal nerve regeneration into the posterior cricoarytenoid muscle in a canine model

OBJECTIVES/HYPOTHESIS

To investigate the efficacy of paclitaxel, a potent microtubule inhibitor with a more favorable therapeutic index as compared with vincristine, in preventing post-traumatic nerve regeneration of the recurrent laryngeal nerve into the posterior cricoarytenoid muscle in a canine laryngeal model.

STUDY DESIGN

Experimental animal study.

METHODS

Forty-nine canine hemilaryngeal specimens were divided into five experimental groups. Under general anesthesia, a tracheostomy, recurrent laryngeal nerve (RLN) transection and repair, and laryngeal adductory pressures (LAP) were measured pre-RLN injury. The approach to the posterior cricoarytenoid (PCA) muscle for neurotoxin injection was transoral or open transcervical, at 0 or 3 months. At 6 months, postinjury LAPs were measured and the animals were sacrificed at 6 months to allow for laryngeal harvesting and analysis.

RESULTS

Paclitaxel demonstrated increased mean laryngeal adductory pressures (70.6%) as compared with saline control (55.5%). The effect of paclitaxel was the same as observed with vincristine at 0 months and with a delayed injection at 3 months. There was no difference between transoral or open injection groups.

CONCLUSIONS

PCA muscle injection with paclitaxel resulted in improved strength of laryngeal adduction. This effect was similar to that of vincristine at both 0 and 3 months following nerve injury. A single intramuscular injection of paclitaxel was well tolerated. Additional human studies are needed to determine the degree of clinical benefit of this intervention.

LEVEL OF EVIDENCE

NA Laryngoscope, 127:651-655, 2017.

Compound Motor Action Potential Quantifies Recurrent Laryngeal Nerve Innervation in a Canine Model

Objective:

The compound motor action potential (CMAP) is the summated action potential from multiple muscle fibers activated by a single nerve impulse. The utility of laryngeal muscle CMAP for quantifying innervation following recurrent laryngeal nerve (RLN) injury was investigated.

Method:

In a series of 21 canine hemi-laryngeal preparations, RLNs were exposed and a stimulating electrode placed. Maximum CMAP amplitudes and area under the curve from the thyroarytenoid (TA) muscles were obtained at baseline and at 6 months following injury to the RLN. Injury mechanisms included crush, stretch, cautery, and complete transection with microsuture repair.

Results:

Prior to injury, baseline CMAP amplitudes and area under the curve were 15.81 mV and 15.49mVms, respectively. Six months following injury, CMAP amplitude and area under curve were 105.1% and 102.1% of baseline for stretch, 98.7% and 112.7% for crush, 93.3% and 114.3% for cautery. The CMAP amplitude and area under the curve in the transection/repair group had a 54.3% and 69.4% recovery, respectively, which were significantly different than baseline ( P < .01, P < .05). These values were correlated with vocal fold motion.

Conclusion:

The CMAP is a measure of vocal fold innervation. The technique could be further developed for clinical and experimental applications.