Complications associated with surgery for RG
The presence of RG is historically known to carry a higher risk of finding a malignancy and of the development of postoperative complications (hematoma, transient/permanent hypoparathyroidism, and RLN dysfunction) compared to cervical goiters81. The most recent evidence has further confirmed these findings: for example, in a cohort of 1500 RLNs at risk, substernal extension and a thyroid volume >100 mL were the only significant factors for postoperative transient or permanent VC paralysis82. In the IONM era, a study extracting 42341 operations from the UK Thyroid database found that RG was an independent factor (OR= 1.36, 95%CI 1.05- 1.77) that predicted the risk of RN injury, along with revision surgery and patient age.83 A recent anatomical study has revealed that iatrogenic lesions of the RNL after RGs’ resections are more frequent on the right side in comparison to the left one because of the known anatomical and embryological differences in the development of the RLN on either side of the neck84.
Regarding postoperative hematoma formation, male sex, the use of preoperative anticoagulant therapy or postoperative subcutaneous heparin, and the presence of RG (38 cases) were all significant predictors in a group of 6900 thyroidectomies85. Instead, RG was not a significant risk factor in another series of 5900 operations, including a total of 148 RGs (2.5%)86. Both series, unfortunately, did not provide a definition of RG nor how the extension was determined. On the contrary, another analysis of the aforementioned UK registry (n= 53838 entries) revealed in multivariable analysis that male sex, increasing age, redo surgery, RG (whenever having an “extension to the thoracic inlet or below”), and total thyroidectomy were all correlated with an increased risk of reoperation for bleeding; interestingly, surgeon monthly thyroidectomy rate correlated with a decreased risk87.
The presence of RG is linked with a major risk of developing clinical symptoms of hypocalcemia (OR 10.26)88. A recent study published by Chen et al. provides evidence that there are predictive factors for this adverse event. This meta-analysis includes 23 studies and identifies twelve significant risk factors for postoperative hypocalcemia. In particular, hypoparathyroidism (OR 5.58), total thyroidectomy (OR 3.59), hypomagnesemia (OR 2.85), and preoperative vitamin D deficiency (OR 2.32), were those most associated with hypocalcemia89. In another observational study, RG and its extension beyond the carina showed a statistically significant higher risk for transient (less than one year) hypocalcemia (relative risk = 1.76) after total thyroidectomy90.
Van Slycke and colleagues reported that the higher risk of postoperative complication is influenced by the surgical approach chosen: out of 95 thyroidectomies performed for RGs, eighty patients (84%) were operated by cervicotomy and 15 (16%) by cervicosternotomy. The latter group had a higher risk of temporary recurrent laryngeal nerve palsy (21%) compared to cervicotomy (4%) and standard thyroidectomy (3%); also, the risk of temporary hypocalcemia after cervicosternotomy was higher than with transcervical approaches 62. In addition, another study highlights how postoperative complications associated with RG surgical removal are low in the hands of experienced, high-volume thyroid surgeons, regardless of transcervical or transthoracic approach91. Instead, the presence of RG was statistically significant for being an independent predictor of any complication (OR 2.1), when controlling for age, BMI, gender, and race91.
Tracheomalacia remains the most dreadful complication in the case of large RGs, despite its incidence seems to be very low92. In a series of 40 patients subjected to thyroidectomy with sternotomy, tracheomalacia was reported in 3 cases, in absence of any patient- or thyroid-related factor significantly associated with its development: one patient required tracheal resection with anastomosis, and two patients required tracheostomy92. In another paper, 17 cases of tracheomalacia out of 106 cases were reported, probably caused by long-standing tracheal compression from RG: this study suggested that prolonged intubation was sufficient to resolve airways collapse without tracheal reconstruction in all cases93. Zuo et al. reported how tracheomalacia could be successfully managed intraoperatively by suspending the trachea to the overlying skin with Prolene stitches, despite the fact that they didn’t exhibit either the exact number of patients affected or the long-term outcomes94. Despite all the aforementioned complications, RG was not considered a contraindication for outpatient surgery: in a recent national survey conducted by the American Association of Endocrine Surgeons, the diagnosis of RG had no or minimal effect on same-day discharge95.
The main issue in the analysis of the complications of surgery for RGs remains the non-standard definitions and characterizations. For instance, postoperative vocal cord dysfunction may be diagnosed by using a mirror52, flexible laryngoscopy57, or a mixture of both82. The timing of what constitutes a “permanent” hypocalcemia changes (after 657, versus after 12 months53); the same is true for hematoma/seroma that should be considered only when needing revision surgery according to some but not all the authors. To overcome these limitations, the use of the Clavien-Dindo system may be a solution: proposed in 1992 and revised in 2004, it appears reliable as a compelling tool for quality assessment in surgery worldwide. This classification system was reported as simple, reproducible, logical, useful, and comprehensive96 and it is beginning to be implemented also in thyroid surgery97.