Journal of Investigative Surgery, Early Online, 1–9, 2014 C 2014 Informa Healthcare USA, Inc. Copyright ISSN: 0894-1939 print / 1521-0553 online DOI: 10.3109/08941939.2014.975876
ORIGINAL RESEARCH
Thyroid Surgery and the Usefulness of Intraoperative Neuromonitoring, a Single Center Study Sophie J. R. de Danschutter, MD, Jennifer M. J. Schreinemakers, MD, PhD, Leoni H. M. Smit, MD, PhD, Lijckle van der Laan, MD, PhD, Hans K. S. Nuytinck, MD, PhD
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
Department of Surgery, Amphia Hospital, Molengracht, Breda, the Netherlands
ABSTRACT Purpose/Aim of study: To compare the use of intraoperative neuromonitoring (IONM) versus visualization of the recurrent laryngeal nerve (RLN) alone in thyroid surgery with regard to incidence in postoperative RLN injury and operation time. Materials and Methods: This retrospective cohort study was performed in the Amphia Hospital, the Netherlands. All thyroid gland operations were collected from September 2009 to October 2012. For each case we recorded the patient characteristics, indication for surgery, intraoperative data, complications, results of pathological evaluation, and consultation of a ENT-surgeon. Research of current literature and statistical analysis was performed. Results: In total, 147 patients were included and classified into an IONM and non-IONM group. Both groups were similar in demographical aspects and indications for surgery. In total, we had 170 nerves at risk (NAR). In both groups, there were 85 (50%) NAR. Overall injury to the RLN was 6%. A statistical significant decrease of permanent RLN injuries was noticed in the IONM group compared to the non-IONM group (n = 0 vs n = 6; p = .044). In transient RLN injury, no difference was noticed (n = 2 vs n = 2). Operation time with or without IONM was not significantly different for hemithyroidectomies, neither for total thyroidectomies. Conclusion: IONM is a useful tool as an adjunct in thyroid surgery to prevent RLN injury. A statistical significant decrease in permanent RLN injury with the use of IONM was found, but it did not significantly decrease time of operation. Keywords: thyroid surgery; intraoperative neuromonitoring; recurrent laryngeal nerve injury
INTRODUCTION
To reduce the incidence of post-operative RLN palsy, next to intraoperative visualization of the RLN, intraoperative neuromonitoring (IONM) has increasingly been used to identify the RLN [8–10]. The main goals of IONM are identification of the RLN, aid in dissection, investigation of functional integrity of the RLN and lesion site identification [11]. Several studies have shown the benefit of neuromonitoring for nerve identification in thyroid surgery. Yet, it remains unclear if the monitoring actually decreases the incidence of RLN injury, probably due to the relatively low number of RLN injury in thyroid surgery, even without IONM, and small sample sizes of series [12–18]. The aim of our study was to compare the use of intraoperative nerve integrity monitoring (NIM) versus visualization of the RLN alone as regards to the incidence in postoperative RLN injury and operation time in thyroid surgery.
In many thyroid gland disorders, both benign and malignant, surgery is the treatment of choice [1]. Apart from bleeding and hypocalcemia, injury to the recurrent laryngeal nerve (RLN), is the most invalidating complication after thyroid surgery [2]. The RLN has a great variability in anatomic course [3–5]. Injury to the RLN can occur as a result of stretch, pressure, crush, electrocautery, suction trauma or ischemia during thyroid, or parathyroid surgery [6]. Unilateral RLN injury can lead to dysphonia, with a potential great impact on quality of life. Consequences of bilateral RLN injury can even be worse, leading to possible bilateral vocal cord paralysis and obstruction of the airway, mostly requiring a tracheostomy. Incidences in literature vary from 0% to 11% in permanent RLN injury and from 0% to 7.1% in transient RLN injury [7]. Received 9 April 2014; accepted 8 October 2014.
Address correspondence to Sophie J. R. de Danschutter MD, Department of Surgery, Amphia Hospital, Molengracht 21, 4818 CK Breda, the Netherlands. E-mail: [email protected]
1
2 S. J. R. de Danschutter et al.
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
MATERIALS AND METHODS This study presents all thyroidectomies, performed in the Amphia hospital Breda, from September 2009 to October 2012. We collected data on all thyroid gland operations in our department of general surgery to create a retrospective cohort study for evaluating the effect of IONM. We consulted the institutional review board (AMOA) of the Amphia hospital and they confirmed that no formal written waiver for the need of ethics approval was required, because of the retrospective design of the study. Also there was no written consent needed from the patients, because of the anonymization of patients and the retrospective design. In our hospital, thyroid gland surgery is performed by general surgeons specialized in thyroid surgery. R 3.0 Nerve MonSince the introduction of the NIM itoring System in our hospital in July 2010, surgeons started using IONM to lower the RLN injury rate. The NIM equipment was introduced after a training at prof. Dralle in Halle by one of the surgeons and a senior resident. All patients who underwent a thyroidectomy before this period where included in the non IONM group. After introduction the NIM equipment was used consistently. No selection criteria were used to use the NIM in this retrospective study. However, some recommend that IONM should be used routinely to obtain adequate experience, especially for difficult cases[19,20]. Exclusion criteria included: first, patients with postoperative hoarseness in whom no postoperative laryngoscopy was performed. Second, cases where IONM was used with a loss of signal (LOS) already before dissection, which resulted in an unreliable measurement. The third and last exclusion criteria was the performing of only an isthmusresection or resection of solely the nodule, in which the RLN was not at risk. The resulting case mix also includes patients with huge goiters, as well as patients with indication for reoperation and patients partly referred from other hospitals. We recorded patient characteristics, indication for surgery, intraoperative data (NIM use, stimulation of the vagal nerve and the RLN, macroscopic injury, LOS, operation time), complications, results of pathological evaluation and consultation of an Ear, Nose, Throat (ENT) surgeon. Pre- and postoperative vocal cord assessment was recorded. During all thyroid surgical procedures propofol and sevoflurane were used as anesthetic agents in combination with sufentanil as opioid. For intubation propofol 2–3 mg/kg was used in combination with the shortacting neuromuscular blocking agent rocuronium in a lower as usual induction dose of 0.3 mg/kg. To facilitate an easier intubation an intravenous lidocain bolus of 1 mg/kg was also administered. This combination was used to make sure that muscular activity returned to normal before using the neuromonitor. No antibiotics were given perioperatively. Maintenance of anesthesia was performed using incremental boluses
of 5 mcg sufentanil in combination with sevoflurane in an end-expiratory concentration of at least 1.6 %. . The R 3.0 Nerve Monitoring System was used, which NIM monitors the electromyographic (EMG) activity from muscles innervated by the RLN. It converts muscle activity from the musculus vocalis into acoustic and electromyographical signals. A special endotracheal intubation tube with electrodes monitoring the nerves in both vocal cords is used to perform IONM. Grounding electrodes are placed on the patient’s shoulder or sternum, for correct monitoring. During surgery the vagal nerve and the RLN are stimulated pre- and postdissection to confirm their integrity The superior laryngeal nerve was not stimulated. If the nerve function changes, the NIM monitoring system provides visual and audible warnings [21–24]. RLN injury was defined as hoarseness, which has been split up in two categories. The first was permanent RLN injury with a confirmed paralysis of the vocal cord on postoperative laryngoscopy, existing longer than one year [12,13,15,17,18]. The second was transient RLN injury defined as confirmed pareses of the vocal cord on postoperative laryngoscopy with a complete remission on follow up. All patients with a RLN injury had serial laryngoscopy follow up with a minimum period of one year. Statistical analysis was performed using SPSS version 20. Parameters were tested using the chi-square test and Fisher’s exact test in case of small expected proportion, for categorical variables. The level of significance was set at 0.05. For continuous variables that were normally distributed, means are shown and compared with the independent sample test (Student’s t test), otherwise the Mann-Whitney U and the median were used. The incidence of RLN injury was calculated on the number of nerves at risk (NAR). Subgroup analysis was not performed, due to small number of cases.
RESULTS A total of 163 thyroid surgeries were performed of which 147 patients were included in our study and 16 excluded. Reasons why patients were excluded are: during surgery a LOS from the IONM before dissection (7), patients with postoperative hoarseness in whom no postoperative laryngoscopy was performed (6), patients whom underwent an isthmusresection (2), and resection of solely a nodule (1). The population consisted of 21 (14%) men and 126 (86%) women with a mean age of 53 years, ranging from 25 to 88 years. In 71 (48%) of the cases, we used IOMN compared to 76 (52%) cases where only visual identification was used. Indications for surgery were thyroid nodules (63%), multinodular goiter (17%), re-operation for carcinoma to complete a total thyreoidectomy (11%), intrathoracal goiter (5%), follicular thyroid carcinoma (1%), and others (3%) which included two cases of Journal of Investigative Surgery
Thyroid Surgery and Intraoperative Neuromonitoring
3
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
TABLE 1 Demographical aspects of both groups
Age (years) Mean (+/− SD) Gender n (%) Male Female Indication n (%) Thyroid nodules Multinodular goiter Intrathoracal goiter Follicular thyroid carcinoma Re-operation for carcinoma Othera Operation n (%) Hemithyroidectomy Total thyroidectomy
IONM
Non-IONM
Overall
54 (±16)
52 (±13)
53 (±15)
13 (18%) 58 (82%)
8 (11%) 68 (89%)
21 (14%) 126(86%)
43 (61%) 12 (17%) 5 (7%) 1 (1%)
50 (66%) 13 (17%) 3 (4%) 0 (0%) 8 (10%) 2 (3%)
93 (63%) 25 (17%) 8 (5%) 1 (1%) 16 (11%) 4 (3%)
67 (88%) 9 (12%)
124 (84%) 23 (16%)
p-value .321 .239
.826
8 (11%) 2 (3%)
.256 57 (80%) 14 (20%)
a. therapy resistant Graves’disease, amiodarone induced thyrotoxicosis, primairy surgery for hyperparathyreoidectomy, IONM = intraoperative neuromonitoring, SD = standard deviation, n = number. Percentages are calculated between columns.
therapy resistant Graves’ disease, one amiodarone induced thyrotoxicosis and a surgery with primary intention for primary hyperparathyroidectomy, during operation the surgeon found also a thyroid nodule. Both groups were similar in demographical aspects because no statistical differences were seen in baseline characteristics (gender, age, indication). Also, no significance for use of IONM was seen among the different surgical thyroid gland procedures (Table 1). Most surgical procedures were performed for benign disorders such as nodular hyperplasia (25%), follicular adenoma (22%), multinodular goiter (14%), and dysplastic nodule (10%). In total, 16% of the patients were operated for malignancy, which included papillary and/or follicular thyroid carcinoma. Our overall complication rate was 26%. The three main complications are hypocalcemia (17%), hoarseness (4%), and postoperative hemorrhage (5%). Other complications included stridor (2%), one wound abscess and one perforation of the oesophagus. An overview of the RLN injuries is given in Table 2. To calculate the RLN injury, the injured nerves were related to the number of NAR. In total, there were 170
NAR in 147 patients of whom 23 had a total thyreoidectomy. Eighty-eight (52%) NAR were located on the right and 82 (48%) on the left side of the thyroid gland. In both groups, there were 85 (50%) NAR. The overall injury to the RLN is 6% (10 nerves). In all cases where the RLN was identified, there was no macroscopic injury to the nerve during the procedure. Transient RLN injuries showed no difference between the two groups (n = 2 with IONM vs. n = 2 without). But we noticed a statistical significant decrease of permanent RLN injuries in the IONM group compared to the nonIONM group (n = 0 vs. n = 6; p = p = .044). This group of permanent injuries includes two patients requiring a tracheostoma because of postoperative hoarseness and stridor caused by bilateral RLN injury. In both patients, this was after total thyroidectomy for one preoperatively known carcinoma and one suspicious for carcinoma. In the first case, one RLN was already compromised preoperatively. A reduced abduction of the left vocal cord was seen on preoperative laryngoscopy, therefore we considered this RLN at risk. A third patient underwent hemithyroidectomy for a thyroid nodule. After surgery this patient showed unilateral
TABLE 2 Overview of total recurrent laryngeal nerve (RLN) injuries RLN injury IONM
RLN injury
Total nerves at risk (NAR)
No injury Permanent injury Transient injury
n (%) n (%) n (%) n (%)
Yes
No
Total
p-value
83 (98%) 0 (0%) 2 (2%) 85 (100%)
77 (91%) 6 (7%) 2 (2%) 85 (100%)
160 (94%) 6 (4%) 4 (2%) 170 (100%)
> .05 .044 >.05 >.05
RLN = recurrent laryngeal nerve, IONM = intraoperative neuromonitoring, NAR = Nerves at risk, n = number. Percentages are calculated between columns. C
2014 Informa Healthcare USA, Inc.
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
4 S. J. R. de Danschutter et al. permanent RLN injury. The last patient underwent a total thyroidectomy for a very large multinodular goiter (255 gram) which resulted in a unilateral permanent RLN injury. All cases were confirmed by laryngoscopy postoperatively. An overview of these patients is made in Table 3. In the non-IONM group, 88% of the nerves were identified, in contrast to 98% in the IONM group. Due to some complicated cases surgeons were not always able to find the RLN. Three patients in the non-IONM group, where the nerve was not identified, had permanent RLN injury. This also emphasizes the importance of identifying the RLN. We also calculated the RLN data on diagnostic hemithyroidectomies for benign disease (n = 91, Table 4). A total of 43 (47%) were NAR in the IONM group, and 48 (53%) were NAR in the non-IONM group. The overall injury rate to the RLN in diagnostic hemithyroidectomies was 2% (2 nerves). No permanent injuries occurred. There was no difference between both groups for transient injuries (n = 1 in both groups, p = .937). Those patients whom had transient injuries were confirmed on direct laryngoscopy with a paresis of one vocal cord. With speech therapy, the paresis resolved. In the 71 patients where IONM was used, a LOS occurred in six patients (8%). The LOS was present in one patient during resection, without a good explanation. LOS occurred in five patients after resection, all of who had a visually intact RLN. The LOS was probably due to a technical error. In all these cases, there was no sign of RLN injury postoperatively. During our study period, preoperative laryngoscopy was not commonly used, only in 45% (71/157) of the cases. It was not limited to re-operative surgery. Towards the end of the study period more and more patients underwent preoperative laryngoscopy. Surgeons got aware of the importance of evaluating the functional status of the vocal cords. Nowadays, it is common practice to do a pre- and postoperative laryngoscopy in our clinic. Table 5 shows the number of procedures performed by each surgeon and the individual RLN injuries. The differences in RLN injury are influenced by the fact that surgeon number one performed all the difficult cases (reoperations, large goiters and operations for malignancies) in this cohort study. The mean duration of surgery in hemithyroidectomies was not statistically significant comparing the IONM and non-IONM group (83 vs. 77 min; p = .164). Nor was there a difference for total thyroidectomies (121 min for the IONM group vs. 124 min for the nonIONM group; p = .845).
DISCUSSION The international guidelines recommend to use IONM in thyroid surgery to prevent RLN injury[11]. In our retrospective cohort study the overall RLN injury rate
was 6% (transient 2% and permanent 4%), which is in accordance to current literature[7]. We found that the use of IONM significantly diminished the number of permanent RLN injuries to zero. It did not decrease the number of transient injuries. Permanent injuries were only seen in reoperations, operations for carcinoma and total thyroidectomies. In diagnostic hemithyroidectomies, no permanent RLN injuries were seen. The composition of the case mix, therefore, strongly affects the incidence of RLN injury. The incidence of RLN injury is relatively high in the pre-IONM period. However, after introduction of the NIM equipment it was low. In our opinion this justifies the use of an IONM, especially in high risk cases. Increased surgeon volume and experience with the NIM equipment as the study progressed will partly influence the lowering incidence of RLN injury. However, the surgeon who performed the majority of the procedures has extensive experience. The learning curve of each surgeon is necessary to lower the RLN injury rate [25–27]. An overview of the current available literature concerning nerve visualization during surgery alone compared to IONM is given in Table 6a and 6b. There was only one randomized controlled trial (RCT) showing that IONM significantly decreased the incidence for transient injury (p = .011) but not for permanent injury (p = .368) [12]. In the group of patients in whom IONM was used, there was a lower rate of RLN injury than in the conventional group, 2.7% compared to 5.0% (p = .007). An extensive meta-analysis from Higgings et al. reported an overall RLN injury of 3.52% for IONM versus 3.12% for nerve identification alone, which was not statistically different [13]. Other reports also have contradictory results on whether IONM is useful to prevent RLN injury. Yet most studies show no benefit for IONM use in thyroidectomies [14–18]. These results of all these studies will undoubtedly be influenced by case mix. Ours contains a substantial number of patients with an increased risk for RLN injury, partly referred from other hospitals. These included cases with malignancies and total thyroidectomies for large goiters. Additional risk factors for patients who underwent a thyroidectomy are the relatively low-volume hospital and low-volume surgeons in our hospital compared to international data. However, in the Netherlands, these patient numbers are normal patient volumes and not low volume numbers. Dralle et al. (2004) performed a risk stratification for permanent RLN injury. These risks include recurrent benign and malignant goiter (OR 4.7 and 6.7 respectively), primary surgery in thyroid malignancy (OR 2.0), lobectomy (OR 1.8), no nerve identification (OR 1.4), low-or medium-volume hospital (OR 1.3), and low-volume surgeons (OR 1.2) [14]. Others also found significant higher incidences of postoperative RLN injury in cases with thyroidectomy for malignancy as for benign diseases (p = .025), and those undergoing secondary thyroidectomy compared with primary Journal of Investigative Surgery
5
Perm
Perm
Perm
Perm
Trans
Trans
Trans
Trans
1.
2.
3.
4.
5.
6.
7.
8.
Multinodular goiter (125 gr)
Intrathoracal goiter (318 gr)
Thyroid nodule
Reoperation Multinodular goiter (115 gr) Large multinodular goiter (225 gr) Multinodular goiter with trachea deviation (73 gr) Thyroid nodule
PHPT, also thyroid nodule
Indication
Total thyroidectomy
Total thyroidectomy
Hemi thyroidectomy right
Hemi thyroidectomy left
Total thyroidectomy
Total thyroidectomy
Total thyroidectomy
Hemi thyroidectomy
Operation
Yes
Yes
Yes
Yes
No
No
No
Yes
Identification RLN
Yes
No
Yes
No
No
No
No
No
IONM
Multinodular goiter
Multinodular goiter
Multinodular goiter
Follicular thyroid carcinoma Dysplastic nodule
Multi nodular goiter
Papillary thyroid carcinoma Nodular hyper plasia
Pathology
Hoarseness
Light stridor
Hoarseness
Hoarseness
Stridor
Stridor, hemorrhage Hoarseness
Hoarseness
Complication
Laryngoscopy
Reduced mobility of the left vocal cord which resolved with speech therapy Reduced mobility of the right vocal cord which resolved with speech therapy Reduced mobility of the left vocal cord which resolved with speech therapy Reduced mobility of the right vocal cord which resolved with speech therapy
Paralysis both vocal cords
Paralysis right vocal cord
Paralysis both vocal cords
Paralysis right vocal cord
RLN = recurrent laryngeal nerve, IONM = intraoperative neuromonitoring, Perm = permanent, Trans = transient, PHPT = primary hyperparathyroidism.
RLN injury
Case
TABLE 3 Cases with permanent and transient RLN injury
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
6 S. J. R. de Danschutter et al. TABLE 4 Overview of recurrent laryngeal nerve (RLN) injuries in diagnostic hemithyreoidectomies for a benign disease RLN injury IONM
RLN injury
n (%) n (%) n (%) n (%)
No injury Permanent injury Transient injury
Total nerves at risk (NAR)
Yes
No
Total
p-value
42 (98%) 0 (0%) 1 (2 %) 43 (100%)
47 (98%) 0 (0%) 1 (2%) 48 (100%)
89 (98%) 0 (0%) 2 (2%) 91 (100%)
>.05 >.05 .937 >.05
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
RLN = recurrent laryngeal nerve, IONM = intraoperative neuromonitoring, NAR = Nerves at risk, n = number. Percentages are calculated between columns.
surgery (p = .017) [15]. In our series, the patients with permanent RLN injury had risk factors as re-operation, malignant disease and large multinodular goiter. We believe that careful exploration and anatomical knowledge should be the keystone to avoid RLN injury. But in the previous mentioned high risk cases, the IONM can be of particular assistance in identifying the RLN. In diagnostic hemithyroidectomies for benign disease, there was no additional effect of using the IONM. Hoarseness and in particular the ones with a transient course, are not always due to RLN injury, it can also be derived from intubation injury. There are data to suggest that laryngeal complications after thyroidectomy are primarily caused by injury to the vocal cord from intubation (31.3%) and in a lesser extent by injury to the RLN (6.6%) [28]. In our study, patients with hoarseness were excluded if they did not have a postoperative laryngoscopy and vocal cord paralysis could not be assessed. This may also lead to underestimation of RLN injuries. Yet it may be that most of them could have had transient injury due to intubation. A standard pre- and postoperative laryngeal examination is a valuable tool to examine the vocal folds and distinguish between hoarseness due to RLN injury or intubation trauma [29]. Some advantages of IONM are alerting the surgeon to the nerve location before visual identification, subsequently allowing a quicker nerve dissection and it allows for documenting nerve function before and after resection, which could be useful in litigations. On the other hand, using of IONM is associated with an increased time of setup, increased cost of equipment (5%–7% increase), and the potential for false security
TABLE 5 Overview of number of surgeries and RLN injury by each surgeon Surgeon 1 2 3 4 5
Number of operations
Permanent RLN injury
Transient RLN injury
94 23 20 13 7
5 0 1 0 0
3 0 0 1 0
when no warning signals are generated owing to an improperly functioning system [13,17,30]. In our IONM group, we had a LOS in 8%, which is equal to another study [6]. Chiang et al. (2008) searched for the possible mechanisms of nerve injury. They had a LOS in 9.2% of the cases. Causes included transection of the nerve, a constricting band, in advert clamping of the nerve and the main cause was overstretching at the region of Berry’s ligament. In our opinion, LOS was in many cases was due to the learning curve of using and trying to deal with the technical errors of the NIM device for both surgeons and operating staff (nurses, anesthesiologists, etc.). In contrast to what might be expected, we noticed that the operation time for hemithyroidectomies was longer in the IONM group (6 min) in comparison with the non-IONM group. Yet, for total thyroidectomies, the operation time was three minutes shorter in the IONM group. It is debatable if this difference is clinically relevant. Another study also showed that the mean operating time was longer (9 min) using IONM (p < 0.001) [12]. Limitations of our study are the retrospective design and the fact that we could not calculate the validity of IONM. Sensitivity and specificity, as well as positive and negative predictive value could not be measured, due to the fact that not all patients were completely assessed with pre- and postoperative laryngoscopy. It was not a standard procedure in the beginning of the study, nowadays it is. Postoperative ENT consultation was mostly done in case of a presence of hoarseness. This could lead to an underestimation of the true incidence of preoperative or postoperative vocal cord dysfunction. Another limitation is the possibility of selection bias, because there were no clear cut inclusion criteria. This is difficult to overcome, due to the retrospective design of the study. However, it seems unfeasible to perform a good RCT, because it needs a large population to reach adequate power. The only RCT available could show a significant decrease in transient RLN injuries with the use of IONM [12]. The pre IONM and the IONM patient groups were very well comparable in this single center study. To validate this observation, a prospective multicenter trial is needed in the near future. Journal of Investigative Surgery
7
RCT
Studydesign
Retrospective cohort study
8. Current series
4650
372
2483
671
85
85
120
499
501
116
5517
17.832
755 (2.72%)
19 (1.9%)
Trans
Neurosign 100 system, electrodes through cricothyroid ligament Neurosign 100 system, 17 (3.4%) electrodes through cricothyroid ligament Neurosign 100 system, 32 (1.3%) electrodes through cricothyroid ligament R Medtronic Xomed 14 (2.09%) NIM 2, endotracheal tube electrodes R Medtronic Xomed 4 (3.45%) NIM 2, endotracheal tube electrodes R NIM 3 Nerve 2 (2%) Monitoring System, endotracheal tube electrodes
Neurosign 100 system, electrodes through cricothyroid ligament Different kind of devices for IONM
Kind of IONM
5 (4.24%)
2 (2%)
1 (0.86%) 0 (0%)∗
11 (2.96%)
99 (2.1%)
9 (0.4%)∗
2 (0.3%)
20 (4.0%)
-
885 (2.44%)
38 (3.8%)
Trans
4 (0.8%)
143 (0.80%)∗
208 (0.75%)
8 (0.8%)
Perm
>.05 >.05
>.05
.94
.368
–
.26
.011
Trans Perm
p-value
6 (7%)∗
3 (2.54%)
1 (0.3%)
>.05
.89
.40
.044
.62
>.99
37 (0.8%)∗ .05 >.05
ORIGINAL RESEARCH
Thyroid Surgery and the Usefulness of Intraoperative Neuromonitoring, a Single Center Study Sophie J. R. de Danschutter, MD, Jennifer M. J. Schreinemakers, MD, PhD, Leoni H. M. Smit, MD, PhD, Lijckle van der Laan, MD, PhD, Hans K. S. Nuytinck, MD, PhD
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
Department of Surgery, Amphia Hospital, Molengracht, Breda, the Netherlands
ABSTRACT Purpose/Aim of study: To compare the use of intraoperative neuromonitoring (IONM) versus visualization of the recurrent laryngeal nerve (RLN) alone in thyroid surgery with regard to incidence in postoperative RLN injury and operation time. Materials and Methods: This retrospective cohort study was performed in the Amphia Hospital, the Netherlands. All thyroid gland operations were collected from September 2009 to October 2012. For each case we recorded the patient characteristics, indication for surgery, intraoperative data, complications, results of pathological evaluation, and consultation of a ENT-surgeon. Research of current literature and statistical analysis was performed. Results: In total, 147 patients were included and classified into an IONM and non-IONM group. Both groups were similar in demographical aspects and indications for surgery. In total, we had 170 nerves at risk (NAR). In both groups, there were 85 (50%) NAR. Overall injury to the RLN was 6%. A statistical significant decrease of permanent RLN injuries was noticed in the IONM group compared to the non-IONM group (n = 0 vs n = 6; p = .044). In transient RLN injury, no difference was noticed (n = 2 vs n = 2). Operation time with or without IONM was not significantly different for hemithyroidectomies, neither for total thyroidectomies. Conclusion: IONM is a useful tool as an adjunct in thyroid surgery to prevent RLN injury. A statistical significant decrease in permanent RLN injury with the use of IONM was found, but it did not significantly decrease time of operation. Keywords: thyroid surgery; intraoperative neuromonitoring; recurrent laryngeal nerve injury
INTRODUCTION
To reduce the incidence of post-operative RLN palsy, next to intraoperative visualization of the RLN, intraoperative neuromonitoring (IONM) has increasingly been used to identify the RLN [8–10]. The main goals of IONM are identification of the RLN, aid in dissection, investigation of functional integrity of the RLN and lesion site identification [11]. Several studies have shown the benefit of neuromonitoring for nerve identification in thyroid surgery. Yet, it remains unclear if the monitoring actually decreases the incidence of RLN injury, probably due to the relatively low number of RLN injury in thyroid surgery, even without IONM, and small sample sizes of series [12–18]. The aim of our study was to compare the use of intraoperative nerve integrity monitoring (NIM) versus visualization of the RLN alone as regards to the incidence in postoperative RLN injury and operation time in thyroid surgery.
In many thyroid gland disorders, both benign and malignant, surgery is the treatment of choice [1]. Apart from bleeding and hypocalcemia, injury to the recurrent laryngeal nerve (RLN), is the most invalidating complication after thyroid surgery [2]. The RLN has a great variability in anatomic course [3–5]. Injury to the RLN can occur as a result of stretch, pressure, crush, electrocautery, suction trauma or ischemia during thyroid, or parathyroid surgery [6]. Unilateral RLN injury can lead to dysphonia, with a potential great impact on quality of life. Consequences of bilateral RLN injury can even be worse, leading to possible bilateral vocal cord paralysis and obstruction of the airway, mostly requiring a tracheostomy. Incidences in literature vary from 0% to 11% in permanent RLN injury and from 0% to 7.1% in transient RLN injury [7]. Received 9 April 2014; accepted 8 October 2014.
Address correspondence to Sophie J. R. de Danschutter MD, Department of Surgery, Amphia Hospital, Molengracht 21, 4818 CK Breda, the Netherlands. E-mail: [email protected]
1
2 S. J. R. de Danschutter et al.
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
MATERIALS AND METHODS This study presents all thyroidectomies, performed in the Amphia hospital Breda, from September 2009 to October 2012. We collected data on all thyroid gland operations in our department of general surgery to create a retrospective cohort study for evaluating the effect of IONM. We consulted the institutional review board (AMOA) of the Amphia hospital and they confirmed that no formal written waiver for the need of ethics approval was required, because of the retrospective design of the study. Also there was no written consent needed from the patients, because of the anonymization of patients and the retrospective design. In our hospital, thyroid gland surgery is performed by general surgeons specialized in thyroid surgery. R 3.0 Nerve MonSince the introduction of the NIM itoring System in our hospital in July 2010, surgeons started using IONM to lower the RLN injury rate. The NIM equipment was introduced after a training at prof. Dralle in Halle by one of the surgeons and a senior resident. All patients who underwent a thyroidectomy before this period where included in the non IONM group. After introduction the NIM equipment was used consistently. No selection criteria were used to use the NIM in this retrospective study. However, some recommend that IONM should be used routinely to obtain adequate experience, especially for difficult cases[19,20]. Exclusion criteria included: first, patients with postoperative hoarseness in whom no postoperative laryngoscopy was performed. Second, cases where IONM was used with a loss of signal (LOS) already before dissection, which resulted in an unreliable measurement. The third and last exclusion criteria was the performing of only an isthmusresection or resection of solely the nodule, in which the RLN was not at risk. The resulting case mix also includes patients with huge goiters, as well as patients with indication for reoperation and patients partly referred from other hospitals. We recorded patient characteristics, indication for surgery, intraoperative data (NIM use, stimulation of the vagal nerve and the RLN, macroscopic injury, LOS, operation time), complications, results of pathological evaluation and consultation of an Ear, Nose, Throat (ENT) surgeon. Pre- and postoperative vocal cord assessment was recorded. During all thyroid surgical procedures propofol and sevoflurane were used as anesthetic agents in combination with sufentanil as opioid. For intubation propofol 2–3 mg/kg was used in combination with the shortacting neuromuscular blocking agent rocuronium in a lower as usual induction dose of 0.3 mg/kg. To facilitate an easier intubation an intravenous lidocain bolus of 1 mg/kg was also administered. This combination was used to make sure that muscular activity returned to normal before using the neuromonitor. No antibiotics were given perioperatively. Maintenance of anesthesia was performed using incremental boluses
of 5 mcg sufentanil in combination with sevoflurane in an end-expiratory concentration of at least 1.6 %. . The R 3.0 Nerve Monitoring System was used, which NIM monitors the electromyographic (EMG) activity from muscles innervated by the RLN. It converts muscle activity from the musculus vocalis into acoustic and electromyographical signals. A special endotracheal intubation tube with electrodes monitoring the nerves in both vocal cords is used to perform IONM. Grounding electrodes are placed on the patient’s shoulder or sternum, for correct monitoring. During surgery the vagal nerve and the RLN are stimulated pre- and postdissection to confirm their integrity The superior laryngeal nerve was not stimulated. If the nerve function changes, the NIM monitoring system provides visual and audible warnings [21–24]. RLN injury was defined as hoarseness, which has been split up in two categories. The first was permanent RLN injury with a confirmed paralysis of the vocal cord on postoperative laryngoscopy, existing longer than one year [12,13,15,17,18]. The second was transient RLN injury defined as confirmed pareses of the vocal cord on postoperative laryngoscopy with a complete remission on follow up. All patients with a RLN injury had serial laryngoscopy follow up with a minimum period of one year. Statistical analysis was performed using SPSS version 20. Parameters were tested using the chi-square test and Fisher’s exact test in case of small expected proportion, for categorical variables. The level of significance was set at 0.05. For continuous variables that were normally distributed, means are shown and compared with the independent sample test (Student’s t test), otherwise the Mann-Whitney U and the median were used. The incidence of RLN injury was calculated on the number of nerves at risk (NAR). Subgroup analysis was not performed, due to small number of cases.
RESULTS A total of 163 thyroid surgeries were performed of which 147 patients were included in our study and 16 excluded. Reasons why patients were excluded are: during surgery a LOS from the IONM before dissection (7), patients with postoperative hoarseness in whom no postoperative laryngoscopy was performed (6), patients whom underwent an isthmusresection (2), and resection of solely a nodule (1). The population consisted of 21 (14%) men and 126 (86%) women with a mean age of 53 years, ranging from 25 to 88 years. In 71 (48%) of the cases, we used IOMN compared to 76 (52%) cases where only visual identification was used. Indications for surgery were thyroid nodules (63%), multinodular goiter (17%), re-operation for carcinoma to complete a total thyreoidectomy (11%), intrathoracal goiter (5%), follicular thyroid carcinoma (1%), and others (3%) which included two cases of Journal of Investigative Surgery
Thyroid Surgery and Intraoperative Neuromonitoring
3
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
TABLE 1 Demographical aspects of both groups
Age (years) Mean (+/− SD) Gender n (%) Male Female Indication n (%) Thyroid nodules Multinodular goiter Intrathoracal goiter Follicular thyroid carcinoma Re-operation for carcinoma Othera Operation n (%) Hemithyroidectomy Total thyroidectomy
IONM
Non-IONM
Overall
54 (±16)
52 (±13)
53 (±15)
13 (18%) 58 (82%)
8 (11%) 68 (89%)
21 (14%) 126(86%)
43 (61%) 12 (17%) 5 (7%) 1 (1%)
50 (66%) 13 (17%) 3 (4%) 0 (0%) 8 (10%) 2 (3%)
93 (63%) 25 (17%) 8 (5%) 1 (1%) 16 (11%) 4 (3%)
67 (88%) 9 (12%)
124 (84%) 23 (16%)
p-value .321 .239
.826
8 (11%) 2 (3%)
.256 57 (80%) 14 (20%)
a. therapy resistant Graves’disease, amiodarone induced thyrotoxicosis, primairy surgery for hyperparathyreoidectomy, IONM = intraoperative neuromonitoring, SD = standard deviation, n = number. Percentages are calculated between columns.
therapy resistant Graves’ disease, one amiodarone induced thyrotoxicosis and a surgery with primary intention for primary hyperparathyroidectomy, during operation the surgeon found also a thyroid nodule. Both groups were similar in demographical aspects because no statistical differences were seen in baseline characteristics (gender, age, indication). Also, no significance for use of IONM was seen among the different surgical thyroid gland procedures (Table 1). Most surgical procedures were performed for benign disorders such as nodular hyperplasia (25%), follicular adenoma (22%), multinodular goiter (14%), and dysplastic nodule (10%). In total, 16% of the patients were operated for malignancy, which included papillary and/or follicular thyroid carcinoma. Our overall complication rate was 26%. The three main complications are hypocalcemia (17%), hoarseness (4%), and postoperative hemorrhage (5%). Other complications included stridor (2%), one wound abscess and one perforation of the oesophagus. An overview of the RLN injuries is given in Table 2. To calculate the RLN injury, the injured nerves were related to the number of NAR. In total, there were 170
NAR in 147 patients of whom 23 had a total thyreoidectomy. Eighty-eight (52%) NAR were located on the right and 82 (48%) on the left side of the thyroid gland. In both groups, there were 85 (50%) NAR. The overall injury to the RLN is 6% (10 nerves). In all cases where the RLN was identified, there was no macroscopic injury to the nerve during the procedure. Transient RLN injuries showed no difference between the two groups (n = 2 with IONM vs. n = 2 without). But we noticed a statistical significant decrease of permanent RLN injuries in the IONM group compared to the nonIONM group (n = 0 vs. n = 6; p = p = .044). This group of permanent injuries includes two patients requiring a tracheostoma because of postoperative hoarseness and stridor caused by bilateral RLN injury. In both patients, this was after total thyroidectomy for one preoperatively known carcinoma and one suspicious for carcinoma. In the first case, one RLN was already compromised preoperatively. A reduced abduction of the left vocal cord was seen on preoperative laryngoscopy, therefore we considered this RLN at risk. A third patient underwent hemithyroidectomy for a thyroid nodule. After surgery this patient showed unilateral
TABLE 2 Overview of total recurrent laryngeal nerve (RLN) injuries RLN injury IONM
RLN injury
Total nerves at risk (NAR)
No injury Permanent injury Transient injury
n (%) n (%) n (%) n (%)
Yes
No
Total
p-value
83 (98%) 0 (0%) 2 (2%) 85 (100%)
77 (91%) 6 (7%) 2 (2%) 85 (100%)
160 (94%) 6 (4%) 4 (2%) 170 (100%)
> .05 .044 >.05 >.05
RLN = recurrent laryngeal nerve, IONM = intraoperative neuromonitoring, NAR = Nerves at risk, n = number. Percentages are calculated between columns. C
2014 Informa Healthcare USA, Inc.
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
4 S. J. R. de Danschutter et al. permanent RLN injury. The last patient underwent a total thyroidectomy for a very large multinodular goiter (255 gram) which resulted in a unilateral permanent RLN injury. All cases were confirmed by laryngoscopy postoperatively. An overview of these patients is made in Table 3. In the non-IONM group, 88% of the nerves were identified, in contrast to 98% in the IONM group. Due to some complicated cases surgeons were not always able to find the RLN. Three patients in the non-IONM group, where the nerve was not identified, had permanent RLN injury. This also emphasizes the importance of identifying the RLN. We also calculated the RLN data on diagnostic hemithyroidectomies for benign disease (n = 91, Table 4). A total of 43 (47%) were NAR in the IONM group, and 48 (53%) were NAR in the non-IONM group. The overall injury rate to the RLN in diagnostic hemithyroidectomies was 2% (2 nerves). No permanent injuries occurred. There was no difference between both groups for transient injuries (n = 1 in both groups, p = .937). Those patients whom had transient injuries were confirmed on direct laryngoscopy with a paresis of one vocal cord. With speech therapy, the paresis resolved. In the 71 patients where IONM was used, a LOS occurred in six patients (8%). The LOS was present in one patient during resection, without a good explanation. LOS occurred in five patients after resection, all of who had a visually intact RLN. The LOS was probably due to a technical error. In all these cases, there was no sign of RLN injury postoperatively. During our study period, preoperative laryngoscopy was not commonly used, only in 45% (71/157) of the cases. It was not limited to re-operative surgery. Towards the end of the study period more and more patients underwent preoperative laryngoscopy. Surgeons got aware of the importance of evaluating the functional status of the vocal cords. Nowadays, it is common practice to do a pre- and postoperative laryngoscopy in our clinic. Table 5 shows the number of procedures performed by each surgeon and the individual RLN injuries. The differences in RLN injury are influenced by the fact that surgeon number one performed all the difficult cases (reoperations, large goiters and operations for malignancies) in this cohort study. The mean duration of surgery in hemithyroidectomies was not statistically significant comparing the IONM and non-IONM group (83 vs. 77 min; p = .164). Nor was there a difference for total thyroidectomies (121 min for the IONM group vs. 124 min for the nonIONM group; p = .845).
DISCUSSION The international guidelines recommend to use IONM in thyroid surgery to prevent RLN injury[11]. In our retrospective cohort study the overall RLN injury rate
was 6% (transient 2% and permanent 4%), which is in accordance to current literature[7]. We found that the use of IONM significantly diminished the number of permanent RLN injuries to zero. It did not decrease the number of transient injuries. Permanent injuries were only seen in reoperations, operations for carcinoma and total thyroidectomies. In diagnostic hemithyroidectomies, no permanent RLN injuries were seen. The composition of the case mix, therefore, strongly affects the incidence of RLN injury. The incidence of RLN injury is relatively high in the pre-IONM period. However, after introduction of the NIM equipment it was low. In our opinion this justifies the use of an IONM, especially in high risk cases. Increased surgeon volume and experience with the NIM equipment as the study progressed will partly influence the lowering incidence of RLN injury. However, the surgeon who performed the majority of the procedures has extensive experience. The learning curve of each surgeon is necessary to lower the RLN injury rate [25–27]. An overview of the current available literature concerning nerve visualization during surgery alone compared to IONM is given in Table 6a and 6b. There was only one randomized controlled trial (RCT) showing that IONM significantly decreased the incidence for transient injury (p = .011) but not for permanent injury (p = .368) [12]. In the group of patients in whom IONM was used, there was a lower rate of RLN injury than in the conventional group, 2.7% compared to 5.0% (p = .007). An extensive meta-analysis from Higgings et al. reported an overall RLN injury of 3.52% for IONM versus 3.12% for nerve identification alone, which was not statistically different [13]. Other reports also have contradictory results on whether IONM is useful to prevent RLN injury. Yet most studies show no benefit for IONM use in thyroidectomies [14–18]. These results of all these studies will undoubtedly be influenced by case mix. Ours contains a substantial number of patients with an increased risk for RLN injury, partly referred from other hospitals. These included cases with malignancies and total thyroidectomies for large goiters. Additional risk factors for patients who underwent a thyroidectomy are the relatively low-volume hospital and low-volume surgeons in our hospital compared to international data. However, in the Netherlands, these patient numbers are normal patient volumes and not low volume numbers. Dralle et al. (2004) performed a risk stratification for permanent RLN injury. These risks include recurrent benign and malignant goiter (OR 4.7 and 6.7 respectively), primary surgery in thyroid malignancy (OR 2.0), lobectomy (OR 1.8), no nerve identification (OR 1.4), low-or medium-volume hospital (OR 1.3), and low-volume surgeons (OR 1.2) [14]. Others also found significant higher incidences of postoperative RLN injury in cases with thyroidectomy for malignancy as for benign diseases (p = .025), and those undergoing secondary thyroidectomy compared with primary Journal of Investigative Surgery
5
Perm
Perm
Perm
Perm
Trans
Trans
Trans
Trans
1.
2.
3.
4.
5.
6.
7.
8.
Multinodular goiter (125 gr)
Intrathoracal goiter (318 gr)
Thyroid nodule
Reoperation Multinodular goiter (115 gr) Large multinodular goiter (225 gr) Multinodular goiter with trachea deviation (73 gr) Thyroid nodule
PHPT, also thyroid nodule
Indication
Total thyroidectomy
Total thyroidectomy
Hemi thyroidectomy right
Hemi thyroidectomy left
Total thyroidectomy
Total thyroidectomy
Total thyroidectomy
Hemi thyroidectomy
Operation
Yes
Yes
Yes
Yes
No
No
No
Yes
Identification RLN
Yes
No
Yes
No
No
No
No
No
IONM
Multinodular goiter
Multinodular goiter
Multinodular goiter
Follicular thyroid carcinoma Dysplastic nodule
Multi nodular goiter
Papillary thyroid carcinoma Nodular hyper plasia
Pathology
Hoarseness
Light stridor
Hoarseness
Hoarseness
Stridor
Stridor, hemorrhage Hoarseness
Hoarseness
Complication
Laryngoscopy
Reduced mobility of the left vocal cord which resolved with speech therapy Reduced mobility of the right vocal cord which resolved with speech therapy Reduced mobility of the left vocal cord which resolved with speech therapy Reduced mobility of the right vocal cord which resolved with speech therapy
Paralysis both vocal cords
Paralysis right vocal cord
Paralysis both vocal cords
Paralysis right vocal cord
RLN = recurrent laryngeal nerve, IONM = intraoperative neuromonitoring, Perm = permanent, Trans = transient, PHPT = primary hyperparathyroidism.
RLN injury
Case
TABLE 3 Cases with permanent and transient RLN injury
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
6 S. J. R. de Danschutter et al. TABLE 4 Overview of recurrent laryngeal nerve (RLN) injuries in diagnostic hemithyreoidectomies for a benign disease RLN injury IONM
RLN injury
n (%) n (%) n (%) n (%)
No injury Permanent injury Transient injury
Total nerves at risk (NAR)
Yes
No
Total
p-value
42 (98%) 0 (0%) 1 (2 %) 43 (100%)
47 (98%) 0 (0%) 1 (2%) 48 (100%)
89 (98%) 0 (0%) 2 (2%) 91 (100%)
>.05 >.05 .937 >.05
J Invest Surg Downloaded from informahealthcare.com by Selcuk Universitesi on 12/21/14 For personal use only.
RLN = recurrent laryngeal nerve, IONM = intraoperative neuromonitoring, NAR = Nerves at risk, n = number. Percentages are calculated between columns.
surgery (p = .017) [15]. In our series, the patients with permanent RLN injury had risk factors as re-operation, malignant disease and large multinodular goiter. We believe that careful exploration and anatomical knowledge should be the keystone to avoid RLN injury. But in the previous mentioned high risk cases, the IONM can be of particular assistance in identifying the RLN. In diagnostic hemithyroidectomies for benign disease, there was no additional effect of using the IONM. Hoarseness and in particular the ones with a transient course, are not always due to RLN injury, it can also be derived from intubation injury. There are data to suggest that laryngeal complications after thyroidectomy are primarily caused by injury to the vocal cord from intubation (31.3%) and in a lesser extent by injury to the RLN (6.6%) [28]. In our study, patients with hoarseness were excluded if they did not have a postoperative laryngoscopy and vocal cord paralysis could not be assessed. This may also lead to underestimation of RLN injuries. Yet it may be that most of them could have had transient injury due to intubation. A standard pre- and postoperative laryngeal examination is a valuable tool to examine the vocal folds and distinguish between hoarseness due to RLN injury or intubation trauma [29]. Some advantages of IONM are alerting the surgeon to the nerve location before visual identification, subsequently allowing a quicker nerve dissection and it allows for documenting nerve function before and after resection, which could be useful in litigations. On the other hand, using of IONM is associated with an increased time of setup, increased cost of equipment (5%–7% increase), and the potential for false security
TABLE 5 Overview of number of surgeries and RLN injury by each surgeon Surgeon 1 2 3 4 5
Number of operations
Permanent RLN injury
Transient RLN injury
94 23 20 13 7
5 0 1 0 0
3 0 0 1 0
when no warning signals are generated owing to an improperly functioning system [13,17,30]. In our IONM group, we had a LOS in 8%, which is equal to another study [6]. Chiang et al. (2008) searched for the possible mechanisms of nerve injury. They had a LOS in 9.2% of the cases. Causes included transection of the nerve, a constricting band, in advert clamping of the nerve and the main cause was overstretching at the region of Berry’s ligament. In our opinion, LOS was in many cases was due to the learning curve of using and trying to deal with the technical errors of the NIM device for both surgeons and operating staff (nurses, anesthesiologists, etc.). In contrast to what might be expected, we noticed that the operation time for hemithyroidectomies was longer in the IONM group (6 min) in comparison with the non-IONM group. Yet, for total thyroidectomies, the operation time was three minutes shorter in the IONM group. It is debatable if this difference is clinically relevant. Another study also showed that the mean operating time was longer (9 min) using IONM (p < 0.001) [12]. Limitations of our study are the retrospective design and the fact that we could not calculate the validity of IONM. Sensitivity and specificity, as well as positive and negative predictive value could not be measured, due to the fact that not all patients were completely assessed with pre- and postoperative laryngoscopy. It was not a standard procedure in the beginning of the study, nowadays it is. Postoperative ENT consultation was mostly done in case of a presence of hoarseness. This could lead to an underestimation of the true incidence of preoperative or postoperative vocal cord dysfunction. Another limitation is the possibility of selection bias, because there were no clear cut inclusion criteria. This is difficult to overcome, due to the retrospective design of the study. However, it seems unfeasible to perform a good RCT, because it needs a large population to reach adequate power. The only RCT available could show a significant decrease in transient RLN injuries with the use of IONM [12]. The pre IONM and the IONM patient groups were very well comparable in this single center study. To validate this observation, a prospective multicenter trial is needed in the near future. Journal of Investigative Surgery
7
RCT
Studydesign
Retrospective cohort study
8. Current series
4650
372
2483
671
85
85
120
499
501
116
5517
17.832
755 (2.72%)
19 (1.9%)
Trans
Neurosign 100 system, electrodes through cricothyroid ligament Neurosign 100 system, 17 (3.4%) electrodes through cricothyroid ligament Neurosign 100 system, 32 (1.3%) electrodes through cricothyroid ligament R Medtronic Xomed 14 (2.09%) NIM 2, endotracheal tube electrodes R Medtronic Xomed 4 (3.45%) NIM 2, endotracheal tube electrodes R NIM 3 Nerve 2 (2%) Monitoring System, endotracheal tube electrodes
Neurosign 100 system, electrodes through cricothyroid ligament Different kind of devices for IONM
Kind of IONM
5 (4.24%)
2 (2%)
1 (0.86%) 0 (0%)∗
11 (2.96%)
99 (2.1%)
9 (0.4%)∗
2 (0.3%)
20 (4.0%)
-
885 (2.44%)
38 (3.8%)
Trans
4 (0.8%)
143 (0.80%)∗
208 (0.75%)
8 (0.8%)
Perm
>.05 >.05
>.05
.94
.368
–
.26
.011
Trans Perm
p-value
6 (7%)∗
3 (2.54%)
1 (0.3%)
>.05
.89
.40
.044
.62
>.99
37 (0.8%)∗ .05 >.05
