In reoperative thyroid surgery, there is higher rate of complications because adjacent tissue is scarred and anatomy is distorted. The scarring makes dissection more difficult, and bleeding further compromises the operation field. Distorted anatomy increases the injury rate of important adjacent structures . Even for an experienced surgeon, reoperative surgery is stressful.
RLN injury in reoperative surgery occurs primarily because of difficulty in identifying the RLN. Identifying the RLN lowers the incidence of permanent RLN injury from 5.2% to 1.2% . In addition, the RLN injury rate in reoperative surgery is around 1% to 12% in different studies [5–7]. IONM helps to identify and map the route of the RLN during reoperation, but whether it prevents RLN injury is still controversial [1, 9–11]. One study  showed that the rate of RLN injury was higher in a group of reoperative thyroidectomies done without IONM (19%) than in a group done with IONM (7.8%). Another study  reported 1.9% and 1.7% permanent RLN injury rates for reoperative thyroidectomies with and without IONM, respectively . However, no study presents the actual RLN injury rate during reoperative thyroid surgery with RLN re-exploration, nor does any show that IONM is beneficial for re-exploring the RLN.
Another study  includes 89 and 157 patients with and without IONM usage, respectively. And the incidence of transient nerve palsy rate is higher in the group with IONM usage (6.2 vs. 2.5%), but showed no statistically significance. The incidence of permanent injury rate is 0% in the group with IONM usage, and 0.6% in those without. In this study, <10% of the patient was diagnosed as thyroid malignancy and the operation is performed by eight surgeons with different experience. Besides, the patient group is selected according to the surgeon’s preference, which would cause huge selection bias. In our study, patients with IONM usage showed significant benefit in preventing RLN injury and the reason may be that we included more thyroid malignancy in our patient group. In our study, 42.9% of patients with IONM and 80% of patients without IONM are diagnosed with thyroid malignancy and this may increase the difficulty of reoperation due to tumor invasion. All the operations in our study are performed by a single high-volume surgeon and all patients received IONM usage after introduction of IONM, which would significantly reduce bias from patient selection and surgeon’s technique.
In our study, both groups of patients had a definite history of exploration and re-exploration of RLNs. The actual injury rate of each re-exploration was compared. For the group without IONM, the accidental injury rate was 20% (3 of 15 nerves), and for the group with IONM, it was a significantly (P <0.05) different 1.43% (1 of 70 nerves). The overall reoperation injury rate in our institute is 1.90% (4 of 210 cases). Although the benefit of using IONM for reoperative thyroid surgery may not be indisputable, we hypothesize that if re-exploration near RLNs is planned and done using IONM, significantly fewer accidental RLN injuries will occur. IONM will also give surgeons more confidence when re-exploring RLNs, as shown with the patients in this study.
Several potential pitfalls of IONM have been reported [15, 16], such as device malfunction, improper device setup, misuse of muscle relaxant, anatomic variation of the RLN, and shunt stimulus, which can cause misleading information. Device set-up problems are the most common cause of false IONM results. In the one case of accidental injury in this study, an RLN was transected because the monitor temporarily stopped functioning during surgery. To prevent this from occurring, a standard manual for device set-up and proper endotracheal tube insertion is necessary . However, the possibility that the IONM device will malfunction or require readjustment during the operation still exists. Intermittently stimulating the vagus nerve using a recently designed anchor electrode [17, 18] in addition to a conventional handheld bipolar stimulation electrode during surgery may alert the surgeon to device malfunction. In our first case of RLN injury, device malfunction was identified as the main reason of injury. Continuous IONM would gain a lot of benefit in this situation and prevent nerve injury .
There are other advantages of using IONM during thyroid surgery. For accidental transection or deliberate sacrifice of the RLN, IONM helps when searching for the transected end. Once the transected end is identified, better re-anastomosis of the RLN can be achieved. In addition, IONM helps identify the transected end of RLNs with very short ends and RLNs that are buried in the Berry’s ligament. This should allow a more precise nerve repair or even re-anastomosis with the ansa cervicalis. However, the actual benefits require more investigation . IONM revealed advantage in endoscopic surgery in animal study, but the benefit in human need further investigation .
For patients with bilateral disease or preoperative unilateral RLN paresis, IONM may guide the strategy of the thyroid operation. IONM can be used to monitor the RLN function because that function may still be impaired even though the RLN visually appears to be intact. In one study , 33 patients developed unilateral cord paralysis, but only five nerve injuries were recognized during surgery. In another , 40 patients developed cord paralysis with only three recognizable injuries. The possible mechanisms include forceps clamping, nerve stretching, electrothermal injury, ligature entrapment, and ischemia. A stretching injury may be the main reason of temporary paresis . If a temporary or permanent nerve dysfunction occurs during the operation, the surgical strategy may be changed for safety . For patients who already have unilateral vocal cord paresis, an operation on the contralateral side using IONM will help, not to prevent an RLN injury, but to detect RLN paresis, to guide postoperative care, or even to prompt the surgeon to do a protective tracheostomy after the thyroid surgery. One of our patients developed RLN paresis during the operation and we changed our strategy to a total laryngectomy with a tracheostomy.
This study has some limitations. The first is the lack of randomized control groups because we have used an IONM device for every patient given a reoperation since it was introduced. Therefore, we used as our control group patients operated on before the introduction of the IONM. It is almost impossible to do a prospective randomized control study for this topic for three reasons. First, there are not many such cases every year. Second, if we select cases from several surgeons, differences in technique will significantly bias the results. Third, it is unethical to put high-risk patients into a control group that does not allow the surgeon to use currently available technology to help identify the route of the nerve so that it will not be accidently damaged. Therefore, this study presents the usefulness of IONM during thyroid surgery done by a single high-volume surgeon. The second is that we did more reoperation after introduce of IONM since 2006 (15 cases in 5 years vs. 56 cases in 5 years). Before introduction of the device, we reoperated in a limited indication. And the percentage of thyroid cancer is higher in the group without IONM usage. After IONM usage, we must say that we gain more assurance and widen the indication of reoperation to recurrent benign goiter or Graves’ disease. In our experience, avoiding direct dissection over the RLNs is the best way to prevent injury. The IONM helps to identify or map the route of RLNs, but it does not always prevent the surgeon from transecting the nerve.