We identified all available English language publications regarding the relationship between goiter and the esophagus, covering the period from 1 January 1975 to 1 July 2018. The search terms used in the scientific databases PubMed and EMBASE were variations of the terms for goiter, esophagus, swallowing, and dysphagia, with MESH terms being Goiter, Esophagus, Deglutition, and Deglutition disorders.

Initially, we screened publication titles and then selected abstracts of relevance. Each identified paper was screened in accordance with the inclusion criteria being: data obtained in humans; age ≥18 years; individuals should undergo thyroidectomy or be offered observation alone; any study should include a minimum of 15 patients. The reference list of each paper was screened for missed publications. The data extracted included number of patients, gender, study design, blinding, inclusion, and exclusion criteria, summary of results, and the presence of major confounders as well as limitations and biases. Biases were assessed for the individual studies using the “Cochrane Risk of Bias tool” (23).

From PubMed and EMBASE 3,040 relevant titles were included, supplemented by 20 articles obtained from the reference lists of included papers (Figure 1). We excluded 3,005 papers after review of titles and abstracts, leaving 55 for full text evaluation. Twenty-eight full text papers failed to meet the inclusion criteria, resulting in 27 papers for the final analyses.

Applying the “Cochrane Risk of Bias tool” across studies (23), we found a considerable risk of bias as many studies failed to describe inclusion procedures, include control groups, randomize for treatment, or blind observers or participants (Table 1). As a consequence, we made no attempt to perform a meta-analysis. Most of the included studies (Tables 2, 3) were small, with 60 or fewer individuals in 14 of the 27 studies. Only one study applied randomization (15) and three studies used some blinding of participants, personnel or the evaluated outcomes (25, 30, 31). There were five observational studies, 17 studies investigating the impact of thyroidectomy, and five retrospective studies with two studies investigating thyroidectomy (Tables 2, 3). As evident from the heterogenous studies, the level of evidense range from low to moderate (Table 4).

Affection of the esophagus was a common finding among patients with goiter (Table 2). Four studies evaluated, by chest X-ray, CT, or MRI, the prevalence of esophageal compression and/or deviation in patients undergoing thyroidectomy (8, 16, 24, 25). A study of 198 patients found esophageal deviation in 14% of individuals, and esophageal compression in even fewer (8%) (16). A higher prevalence of esophageal compression was found in another study of only 23 patients, with six patients (26%) having esophageal compression (8). A third study found 24 of 26 patients (95%) with substernal goiter to have esophageal midline deviation (24). All the above studies had shortcomings, such as unclear definitions of either esophageal deviation and/or compression.

A more recent study (25), with well-defined definitions of both esophageal deviation and compression, evaluated 65 well-characterized goiter patients undergoing neck MRI before and 6 months after thyroidectomy. In that study, the median esophageal deviation decreased significantly from 4 mm (range: 0–23 mm) to 3 mm (range: 0–10 mm). More importantly, the median smallest cross-sectional area of the esophagus (SCAE) increased from 95 mm² (range: 47–147 mm²) at baseline to 137 mm² (range: 72–286 mm²) after surgery, corresponding to a median increase of 34% (range, −17 to 253%) in SCAE (25). The anterior-posterior depth of the esophagus did not change significantly following surgery, while the median-lateral width of the esophagus achieved a small but significant increase from a median of 15 mm (range: 10–21 mm) to 17 mm (range: 12–24) after surgery (Table 2). The esophagus had a small midline deviation, and this persisted after surgery. More importantly, there was an increase in the cross-sectional area of the esophagus following surgery, and a change from a more round to a more ellipsoid shape. The preoperative volumes of the thyroid and the SCAEs were significantly inversely correlated, with reduction in SCAE of 0.35 mm² for every 10% increase in goiter volume. In addition, the SCAE increased by 1% for every 10% increase in the weight of the removed thyroid tissue (25). This implies that patients with the initially largest goiters also experienced the most pronounced improvements in SCAE, following thyroid surgery.

In relation to the impact of individual nodules on esophageal anatomy, one study found globus sensation to be present in 48% of patients. Nodules >3 cm in diameter, and/or located anteriorly and horizontally to the trachea were more often present in patients with globus sensation than in patients with smaller nodules or nodules located laterally to the trachea (26).

The swallowing function had been investigated by various objective methods (Table 2). Using scintigraphic methods for assessment, an increase in the esophageal transit time by 2–7 s were demonstrated in patients with a large goiter (>7 cm long or >3 cm in depth, evaluated by ultrasound), as compared to smaller goiters (27). Another study, also using nuclear medicine techniques, found 39% of goiter patients to have increased esophageal transit time (28).

A study of 36 patients identified a decrease in the upper esophageal sphincter (UES) pressure, using water manometry, in the early period after total thyroidectomy (13). However, these findings were normalized 18–24 months after surgery in a subsequent study of the same patient cohort (29). Another study of 33 patients, found normal esophageal motility, both before and 6 month after surgery (30). Only the UES basal pressure increased significantly from 71 to 88 mmHg 6 month after surgery, the latter value being within the normal limits (30). In our study, using the Chicago classification version 3.0 for assessing motility disturbances (42), only one of the 33 patients with a benign goiter (3%) had a minor peristaltic disorder preoperatively, defined as >50% weak or failed coordinated esophageal muscle contractions during 10 swallowing events (30). Two patients (6%) fulfilled these criteria 6 months after surgery. Importantly, as patient characteristics as well as the manometric examination techniques (i.e., water manometry and high resolution esophageal manometry), differed between the studies, a direct, comparison across studies is not meaningful. Nevertheless, it seems that decrease in the UES pressure potentially may be present in the early period after surgery as the goiter is removed. This issue is of no greater concern beyond 6 months after surgery.

A few studies have investigated the hyoid bone movement, epiglottic tilting, and laryngeal mobility in goiter patients (31–35). A study of 54 patients demonstrated a preoperative association between abnormal laryngeal mobility and the prevalence of dysphagia in patients with goiter (31). In the immediate weeks after surgery the laryngeal mobility was reduced, with recovery at 6 months postoperatively (32). Another study found the hyoid bone movement to be unchanged after surgery in an investigation of 34 patients (33). In a study with 54 patients divided into two groups using neither randomization nor blinding, trans-axillary thyroidectomy was associated with a lower degree of neck muscle adhesion and a greater hyoid bone movement, as compared with conventional thyroidectomy (34). The prevalence of reflux laryngitis was also found lower in patients with cervical compared to substernal goiter, at 21 vs. 42%, respectively (35).

Three studies, using validated swallowing specific questionnaires, have investigated the patients' perception of swallowing before and after thyroidectomy (Table 3) (12, 17, 18). By using the “swallowing related quality of life questionnaire,” eight (12) and nine (17) of 11 swallowing related domains were affected in studies of 116 and 224 patients with goiter, respectively, all of whom were referred for thyroidectomy. Using the validated Glasgow-Edinburgh throat scale, only two of 12 items improved at 3 months after surgery, while six items had improved after 12 months (18).

Nine studies, the largest including more than 100 patients, used the non-validated swallowing impairment score or modified versions hereof (11, 13, 15, 22, 29, 34, 36–38). Prior to thyroidectomy, the prevalence of dysphagia varied between 48 and 78% (11, 13, 36). At 1 week, 1 month, and 3 months after thyroidectomy, three studies found increased prevalence (i.e., worsening) of swallowing symptoms compared to the preoperative level (11, 22, 37), while a single study of 53 patients found a decreased level of dysphagia (15). However, at 6 months or later after surgery, patients had fewer symptoms than at baseline (11, 22, 29), although a subgroup had persistent swallowing complaints (38). Studies using other examination techniques, such as qualitative analyses, supported these findings (39–41). The impact on swallowing symptoms seemed uninfluenced by the surgical technique or whether hemi- or total thyroidectomy was performed (15, 17, 18, 22).

Two recent studies, carried out by our group on the same patient cohort, assessed the impact of thyroidectomy and esophageal dysfunction by use of the well-validated disease-specific ThyPRO questionnaire (25, 30). Goiter patients had a significantly affected mean Goiter Symptom score of 40 points at baseline (maximum of 100 points equals severe symptoms), which improved significantly to a mean of 10 points postoperatively (25) (minimum of 0 points equals no symptoms) (Table 2). All 11 items of the Goiter Symptom scale improved 6 months after surgery. The items related to swallowing difficulties, such as “discomfort swallowing,” underwent a significant score reduction from a mean of 1.8 points (maximum of 4 points) to 0.3 points, postoperatively. Similarly, the mean score for “difficulty swallowing” decreased significantly from 1.2 to 0.2 points after surgery, while “Globus sensation” decreased from 2.3 to 0.7 points.

Future studies should compare ¹³¹I therapy and ultrasound-guided thermal ablation techniques with various surgical techniques for goiter ablation, as for the effect on the esophagus. Such studies should be performed using validated swallowing and quality-of-life-related questionnaires, with measurements of the esophageal anatomy and function as the potential primary endpoints. Follow-up should be a minimum of 6 months, since improvement of swallowing dysfunction at an earlier time point cannot be expected, at least not after surgical goiter ablation.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.