HSC Section 6 Nov2016 Green Book

Otolaryngology–Head and Neck Surgery 153(3)

cases 1,5 and another demonstrating an LEMG-confirmed neuropathy in 100% of clinically diagnosed pareses. Dursun et al demonstrated that a thorough neurolaryngeal examina- tion in the hands of an experienced laryngologist can diag- nose superior laryngeal nerve paresis in 98% of cases with characteristic examination findings. 16 Merati et al demon- strated that 92% of patients with clinical vocal fold motion impairment had a neuropathy. 17 Furthermore, the ability to obtain reliable and accurate LEMG results in the larynx requires a significant degree of experience and expertise, specifically in the interpretation of laryngeal data, and it is heavily dependent on accurate and consistent needle place- ment, which limits its usefulness and availability in many centers. Last, it is an invasive test, not without attendant morbidity, and it does not often address the etiology of the paresis. Thus, it would not obviate the need for further ima- ging and laboratory testing in confirmed cases. A possible source of selection bias lies in the fact that 27% of paresis cases did not undergo CT. These cases likely had a clear etiology on history and examination, such as preceding surgical injury or known cervicothoracic malignancy. While this may raise the diagnostic yield of CT by excluding cases with a higher likelihood of a negative finding, they are not true idiopathic cases, and CT in this context would not likely provide additional diagnostic infor- mation, which is the key clinical question. Last, a possible source of misclassification bias lies in the fact that the clinical diagnosis of vocal fold paresis has evolved over the 10-year sample period. That is, we have had a higher index of suspicion and perhaps a lower thresh- old for the diagnosis of paresis in recent years, and patients with more subtle findings of paresis may have been excluded in earlier years of the study. However, a sensitivity analysis was performed comparing the diagnostic yields of the first 5 years of the study (2.2%) and the second 5 years (5.1%), and it revealed no statistically significant difference between these 2 periods (2.9%, P = .34). In the light of the above findings, our current investiga- tion of IUVFP involves a tailored approach. Patients who have additional localizing symptoms and examination find- ings are investigated in a targeted manner with imaging and appropriate blood tests. In true clinically idiopathic cases, we now offer the patient an informed choice between repeat videostroboscopic evaluation and initial CT, having dis- cussed the diagnostic and incidental yields of imaging. If there is a convincing history of a significant preceding upper respiratory infection and definitive coincident onset of laryngologic symptoms, we are more likely to advise repeat observation over up-front CT, as the underlying etiol- ogy is more likely a viral neuritis. Reliability of follow-up is also an important consideration. On serial examination, if a paresis evolves into paralysis or bilateral findings, our suspicion of an underlying nefar- ious lesion is heightened, and we then perform CT and other investigations as appropriate. Future prospective mul- ticenter studies validating clinical diagnosis of paresis against laryngeal EMG and CT are warranted.

fine-needle aspiration for thyroid nodules. In all but 1 case, these lesions were benign, and no direct complications from the aspiration procedure were recorded. Two patients under- went surgery for benign, otherwise asymptomatic lesions. In both cases, there were no complications from surgery. Only 1 of the 48 incidental cases was an incidental finding clearly beneficial to the patient. In this case, the patient had a ser- endipitous incidental finding of a thyroid malignancy, which was successfully treated with total thyroidectomy. Although it is not possible to definitively state how many of these lesions may have become clinically significant at a later date, none of the lesions became symptomatic or required further action over the period of follow-up. And while it can be argued that the incidental discovery of an early thyroid malignancy or a small descending thoracic aortic aneurysm is beneficial for the patient and may lead to improved survival, it cannot be argued that this justifies rou- tine CT in the evaluation of IUVFP. One must be clear that, in this context, CT investigation is a diagnostic test, trying to answer the question of etiology of paresis, and is not a screening investigation. While not a focus of this study, the underlying etiologies of vocal fold paresis further inform the finding of a low diag- nostic yield. Compared with paresis, a larger proportion of paralysis cases is due to neoplastic causes, and a smaller pro- portion is idiopathic. 8,15 In the 4 published studies (including this one), there is an increasing proportion of idiopathic cases and a much smaller proportion of cases due to neoplasia. In our study, the proportion of idiopathic cases that remained idiopathic was 88%, and the proportion of cases due to neo- plasia was 2.9%. Thus, there is already a low likelihood that CT will uncover an underlying malignancy. This is the largest published study to evaluate the role of CT imaging in the investigation of unilateral vocal fold par- esis. It has an adequate sample size for the desired precision and has a rigorous study method. Only 1 other study addressed the diagnostic yield of CT in the evaluation of IUVFP. Other strengths of this study include the long-term follow-up of patients with incidental findings and the use of 2 laryngologists from a single institution. Given the controversy in the literature in making a clinical diagnosis of paresis, having only 2 asses- sors may limit generalizability, but this singular clinical defini- tion of paresis maximizes internal validity. A possible weakness of this study is the lack of routine LEMG in the confirmation of a diagnosis of paresis, possi- bly leading to misclassification bias. Certain authors sug- gested that LEMG is an essential diagnostic tool in the evaluation of paresis and espoused its use in every case, 3,4 emphasizing that there is a marked discrepancy between clinical observations of paresis and LEMG findings, with discordance in 25% to 40% of cases. 2,4 This discrepancy, however, is not in the absence or presence of neuropathy in the larynx but on the paretic side and the nerve involved. In fact, there is excellent concordance between a clinical diag- nosis of paresis and LEMG findings of the presence of a neuropathy, with 1 study demonstrating LEMG confirma- tion of a clinically diagnosed mild paresis in 86.4% of

15