A review of Immediate and partial neural dysfunction after thyroid and parathyroid surgery: Need for recognition, laryngeal exam, and early treatment

Practice guideline articles seem like they should carry a lot of weight given the number of authors and their collective fame and recognition. Yet one of the pitfalls of writing by consensus is publishing something that is effectively the lowest common denominator. Since the article may be frequently quoted, let’s take a closer look at:

Dhillon VK, Randolph GW, Stack BC Jr, et al. Immediate and partial neural dysfunction after thyroid and parathyroid surgery: Need for recognition, laryngeal exam, and early treatment. Head & Neck. 2020;1–16. https://doi.org/10.1002/hed.26472

The authors introduce a new term, “immediate VFP” (VFP=vocal fold paralysis). Already we have a problem here because none of these words - immediate vocal fold paralysis - are physiologic descriptions of what actually may go wrong during surgery. 

  • Consider the word immediate. If you are a patient, whether an injury is immediate or delayed it not really relevant, it is still an injury. However, most injuries probably are immediate and occur during surgery. Ok, correct terminology, but helpful? The injured patient primarily cares about why their voice is not functioning like it did before surgery. They want to know what the injury is and will it get better. Ideally (and all too often in retrospect) the patient actually wants to know before surgery what are the potential pitfalls of surgery - what can go wrong with the voice, how often does it occur, how severe can it be and will it recover.
  • Consider the next term vocal fold. The thyroid surgeon never injures the vocal folds during thyroid surgery. It's possible the anesthesiologist might traumatize the vocal folds during intubation but that is not the essence of this guideline. The surgeon does not get near or injure the vocal folds.
  • Consider the final term paralysis. Paralysis means complete injury to a nerve, yet this may be the minority condition. At least if you consider paralysis relative to the number of partial nerve injuries (paresis). Paresis is the most common condition after a surgical nerve injury and is not well represented by this definition, “immediate vocal fold paralysis.” The authors add the term, “partial neural dysfunction” but we already have a more succinct word, paresis.

Surgical injury occurs to a nerve which supplies the muscles that move the vocal fold. So already, the semantic foundations of the paper might be built on sand. 

Just say, thyroid surgery has a risk of nerve injury. We already have three descriptive terms which well describe how a nerve is injured (Neuropraxia, physiologic block of nerve conduction within an axon; Axonotmesis - anatomical interruption of the axon with no or only partial interruption of the connective tissue framework; Neurotmesis - complete anatomical disruption of the both the axon and all of the surrounding connective tissue (rupture of the nerve)). These nerve injuries lead to reduction or complete cessation of contraction for the supplied muscle(s). 

I suspect that surgeons primarily think about neurotmesis when they are discussing the potential for nerve injury with patients before thyroid surgery. And actually severing the nerve is probably reasonably uncommon after all the training most surgeons receive. So a surgeon may feel comfortable saying, "I almost never injury the nerve," meaning they can't really recall ever cutting one of the laryngeal nerves.

However, traumatizing the nerve without cutting it may lead to axonotmesis and this is why many surgeons attempt to rely on intraoperative laryngeal EMG, trying to detect when they are pulling too hard on the recurrent laryngeal nerve and potentially internally injuring nerve conduction.

But I suspect the most common surgical injury is some devascularization of the nerve, leading to immediate or delayed axonotmesis or neuropraxia. 

Perhaps as a surgeon you think these “minor” nerve injuries aren’t worth worrying about. I assure you that patients can detect these injuries and many of them seek care elsewhere after you tell them nothing is wrong, leading you to think you hardly ever injure the nerves because you don't cut them, because there is a high rate of recovery and compensation and because patients quit bothering you.

Consensus statements

Let’s consider the statements utilized for the consensus discussion in the paper.

Statement #1 first part:

“Laryngeal dysfunction is complex and involves more than just motion impairment.”

“Laryngeal dysfunction” is complex because there are 4 nerves1 and 9 muscles2 involved, but “nerve injury” is straightforward in the case of motor nerves. Simply, an injured nerve results in reduced or no contraction of the muscle it innervates. So rather than thinking about the larynx as a whole not functioning, the astute examiner will consider what function each muscle of the larynx performs when it contracts. There is much less complexity when each muscle is considered on its own. 

For example, a lateral cricoarytenoid muscle only moves one vocal process medially during phonation. So an astute examiner will focus in turn on only one vocal process, during phonation, to determine if it is moving medially. The examiner isolates the function of each muscle in their mind and concentrates on whether it moves at the appropriate time. Then the examiner moves on to another function. This is why performing a video recording of laryngeal function is so important. Various parts of the larynx must be viewed on their own while other parts are ignored. Consequently a neurolaryngeal exam benefits from viewing motion multiple times. That is much easier to do from a recording on a computer then to perform it live during an exam. In fact, I would say an accurate neurolaryngeal exam is nearly impossible to perform with a mirror examination of the larynx.

Statement #1 second part:

“A comprehensive evaluation in the immediate postoperative setting is important if there is concern for any neural dysfunction.”

This is not sufficient. If a surgeon wants to know if they are injuring the nerves during thyroid surgery, a comprehensive neurolaryngeal evaluation must be performed both before and after surgery. It is the change in the exam which is the primary method for determining when and what nerve injury secondary to surgery. Without a complete preoperative neurolaryngeal examination, who knows what changed during surgery. 

Even the word “important” is insufficient here. More than “important,” examination is the only way to determine if a surgeon has caused an injury and what the injury is. If you examine everyone before and after surgery, you will find many more injuries. You may not want to find them, but not noticing an injury does not mean that it has not occurred.

Recommended statement #1:

Laryngeal nerve injury is relatively straightforward. A comprehensive neurolaryngeal examination before and after an intervention (in this case thyroid surgery) will document and reveal the incidence and degree of injury from surgery.

Statement #2:

Early identification of injury leads to improved patient outcomes. This is a huge hurdle to overcome. Given how often the laryngeal nerves spontaneously recover, very large sample sizes are required to determine if interventions alter outcomes. This is very similar to studying facial nerve injuries and very many proposed treatments for improving recovery after facial nerve injuries have failed to stand up to the test of time.

Recommended statement #2:

It is appropriate to perform a neurolaryngeal exam before and after any intervention which might impair the nerve supply of the larynx. If there are positive findings in the immediate postoperative period, neurolaryngeal exams should be repeated for up to a year until the injury resolves or stabilizes. Even a year is a short period of time if a surgeon really wants to know what the long-term rate of injury is for laryngeal nerve function. Recovery with neural dysfunction may continue to evolve for up to 20 years after an injury. I have seen increasing dyskinesis develop 20 years after thyroid surgical nerve injury, Based on the evolution of the patient’s symptoms, laryngospasm (inappropriate timing and degree of lateral cricoarytenoid muscle contraction) worsens over many years.

Statement #3 first part:

Wait for two months before you check out patient complaints. Have an “Index of suspicion.”

Waiting and indexing are poor triggers for evaluation. The surgeon relying on “index of suspicion” before examining laryngeal muscles is an ostrich. By putting patient complaints off, you are making the patient feel that they are pestering you with their complaints. By two months, they quit bothering you, but the problem injury may still be there, whether you look or not.

Recommended statement #3:

Evaluating pre-and post intervention vocal parameters (including maximum phonation time, vocal range and volume) is one of the strongest verifications of nerve injury. If a patient loses vocal volume or vocal range, there has been a nerve injury until proven otherwise on a detailed endoscopic examination. 

In addition, after surgery, if you looked at the onset of a patient complaint, you would learn what the injury was and if you looked again later, you would learn which injuries resolve, which injuries evolve and which ones are permanent.

Statement #4:

If you think there is an injury, do an “objective voice evaluation through validated questionnaires.”

Validated questionnaires are not objective voice evaluations. They are measures of patient feelings.

Recommended statement #4:

If a surgeon wants to learn how often they injure a nerve, they will document vocal capabilities (volume, range and duration), both before and after any intervention near the laryngeal nerve. If vocal capabilities are not equivalent at this interval, an injury has occurred. 

Go ahead and administer questionnaires if you like them, but they don't really tell you anything about neural function or injury.

Statement #5:

Stroboscopy is important.

Ok. Maybe, though it depends on how stroboscopy is performed and whether it is performed across a range of vocal pitch and vocal volume. The larynx is excellent at compensation for an injury. If stroboscopy is performed at a single pitch and the endoscope is far away from the larynx, most injuries will be missed.

Recommended statement #5:

Endoscopy and stroboscopy, using a neurolaryngeal approach, before and after an intervention can determine which muscles were injured, which recovered appropriately and which ones recovered inappropriately. Want details on a complete neurolaryngeal examination? Download a pdf on Neurolaryngeal examination.


Summary Statement #1: immediate paralysis

There's not much here to support adding the term “immediate” to our description of neural injury that occurs during surgery. 

It is that a big jump in this section to assume that using this new term, immediate, will lead to better patient outcomes after nerve injury if injection augmentation procedures are performed. Certainly that's a topic which should undergo research. But there is a strong possibility of selection bias in some articles.

On the rationale for “Early Identification,” the authors recommend, “a high clinical suspicion for unilateral VFP is important if the patient complains of a weak voice or has an audible breathy quality to the voice.” This triggering approach means that only the most severe injuries even graduate to the point of an examination.

Summary Statement #2: Preoperative laryngeal exams may be helpful.

The word, “may” is a very weak link here. Preoperative exams combined with postoperative exams are the only potential way to find even most of the injuries that occur. Not looking before surgery does not mean there wasn’t an injury during surgery. It is very difficult to go back and time and wish you had an exam that wasn’t performed.

Summary Statement #3: Partial neural dysfunction

It is easy to be concise with this statement. Partial neural dysfunction after thyroid surgery is widely under diagnosed. 

Let's figure out how we can improve this. Patients considering thyroid surgery deserve a more accurate understanding of the risk of nerve injury, especially considering the number of thyroid surgeries performed for what are often ultimately benign conditions. Just ask some of the singers I have seen who had thyroid removal for fear of a cancer that in hindsight had only benign tissue removed and their vocal injury never recovered completely. Was the vocal injury worth removal of benign tissue?

My summary

Nerve injury means a muscle does not contract partially (paresis) or at all (paralysis). Appropriate recovery means the muscle contracts at the appropriate time. Inappropriate recovery means the muscle contracts at an inappropriate time. When opposing muscles contract, one at an appropriate time and one at an inappropriate time, some degree of synkinesis results. When an unopposed muscle contracts at an inappropriate time, dyskinesis results. 

My concern with these expert practice guidelines is that they will continue to perpetuate the sense that neural injury is uncommon during thyroid surgery. Neural injury is underdiagnosed and will likely remain that way if neurolaryngeal examinations are only performed for an index of suspicion. 


1. Nerves: Left and right superior and left and right recurrent laryngeal nerves = 4 nerves.

2. Muscles: paired thyroarytenoid, lateral cricoarytenoid, posterior cricoarytenoid, cricothyroid equals 8 muscle and the interarytenoid may functionally act as a single muscle = 9 muscles.