How to Treat a Tongue-tie: An Evidence-based Algorithm of Care (2024)

PMCID: PMC7859174

Rohil Shekher, MD,¹ Lawrence Lin, MD,¹ Rosaline Zhang, MD,¹ Ian C. Hoppe, MD,¹ Jesse A. Taylor, MD,¹ Scott P. Bartlett, MD,¹ and Jordan W. Swanson, MD¹

Author information Article notes Copyright and License information PMC Disclaimer

Methods:

A meta-analysis was performed to identify the extent of the benefit from frenotomy in breastfeeding measures, degree of tongue-tie, and maternal pain during feeding in randomized controlled trials. A structured literature review analyzed the optimal type and timing of repair. An algorithm was developed to incorporate this evidence into a management pathway.

Results:

Among 424 studies reviewed, 5 randomized controlled trials met inclusion criteria for meta-analysis. Frenotomy significantly improved the degree of tongue-tie, with a 4.5-point decrease in Hazelbaker Assessment Tool for Lingual Frenulum Function score compared with a decrease of 0 in those who did not undergo frenotomy (P < 0.00001). This was associated with improved self-reported breastfeeding (relative risk [RR] = 3.48, P < 0.00001) and decreased pain (Short-Form McGill Pain Questionnaire, P < 0.00001); however, Breastfeeding Self-Efficacy–Short Form and Latch, Audible Swallowing, Type of Nipple, Comfort, Hold scores did not significantly improve. Multiple studies demonstrated significant improvements following frenuloplasty when compared with frenotomy but demonstrated mixed results as to the effect of timing of tongue-tie division.

Conclusions:

Frenotomy is associated with breastfeeding improvements that vary individually but trend toward significance collectively during a critical time in infant development. Among patients with a severe Hazelbaker Assessment Tool for Lingual Frenulum Function score or difficulty breastfeeding, we conclude that simple frenotomy without anesthetic is generally indicated in infancy and frenuloplasty under general anesthesia for older children.

Table 1.

Criteria for Considering Studies for This Systematic Review and Meta-Analysis

Study Characteristics

All of the included studies measured outcomes following division by frenotomy or nondivision of ankyloglossia in at least 1 of the 3 following domains: breastfeeding, degree of tongue-tie, and maternal pain. Validated measures of breastfeeding efficacy utilized by the authors of included studies were the Latch, Audible Swallowing, Type of Nipple, Comfort, Hold (LATCH) score,^20,21 the Infant Breastfeeding Assessment Tool (IBFAT),²² and the Breastfeeding Self-Efficacy Score–Short Form (BSES-SF)²³ (Table ​(Table2).2). The validated measure of degree of tongue-tie, utilized by Buryk et al,¹⁷ was the Hazelbaker Assessment Tool for Lingual Frenulum Function (HATLFF).²⁴ Measures of maternal pain during feeding included a 1–10 scale, the Visual Analog Scale, and the validated Short-Form McGill Pain Questionnaire.^25,26

An overview of the findings of each included study is outlined in Figure ​Figure2.2. The studies included for meta-analysis involved a consistent surgical procedure. No anesthetic is used, and the baby is wrapped securely in a towel. An assistant holds the infant’s shoulders back with their hands, and uses their wrist to fix the infants head in place. The tongue is then held out of the way by the provider performing the division by direct pressure on the tongue with the aid of a grooved director, and the lingual frenulum is divided using sharp, blunt-ended scissors.

Breastfeeding Outcomes

Dollberg et al¹⁴ and Emond et al¹⁶ found no difference in LATCH scores following frenotomy or nondivision. Pooling of the data also showed no statistically significant difference between groups (P = 1.0) (Table ​(Table3).3). Furthermore, there was no difference in the change from baseline of LATCH scores between study groups when study data were combined (P = 1.0).

Following frenotomy or nondivision, IBFAT scores following frenotomy or nondivision from Buryk et al¹⁷ and Emond et al¹⁶ favor frenotomy (P < 0.00001; 95% confidence interval [CI], 0.83–1.30) (Fig. ​(Fig.3A).3A). In a similar fashion, there was a statistically significant benefit to frenotomy (P < 0.00001; 95% CI, 2.62–2.74) when examining the change from baseline of IBFAT scores following frenotomy or nondivision. Emond et al¹⁶ found no difference in BSES-SF scores following frenotomy or nondivision (P = 0.53), but the frenotomy group experienced a greater increase in BSES-SF scores when compared with the nondivision group (P = 0.002).

Meta-analysis showed that patients who underwent frenotomy had better absolute scores across validated breastfeeding efficacy scoring systems following the intervention than patients who did not undergo division (P = 0.04; 95% CI, 0.01–0.41), and patients who underwent frenotomy had a greater improvement across validated breastfeeding efficacy scores when compared with patients who did not undergo division (P = 0.003; 95% CI, 0.10–0.52) (Fig. ​(Fig.44).

There was a subjective improvement in breastfeeding, as reported by the mother, in studies by Berry et al¹⁵ and Hogan et al,¹⁸ following frenotomy when compared with nondivision (P < 0.00001; Relative Probability of Feeding Improvement 95% CI, 2.18–5.56) (Table ​(Table4).4). Berry et al¹⁵ found no dichotomous change in breastfeeding efficacy as documented by an objective observer between patients who underwent frenotomy or nondivision. However, meta-analysis shows that frenotomy is favored when pooling dichotomous qualitative assessment of breastfeeding when measured by the mother and an objective observer (P < 0.00001, Relative Probability of Feeding Improvement 95% CI, 1.75–3.58) (Fig. ​(Fig.33B).

Degree of Tongue-tie

Emond et al¹⁶ found that patients who underwent frenotomy had significant improvement in HATLFF score (mean = 4.5) compared with those who did not (mean = 0; P < 0.0001) and higher HATLFF scores following frenotomy (mean = 13.5) compared with nondivision (mean = 8; P < 0.0001).

Maternal Pain during Breastfeeding

The pooling of pain scores during breastfeeding by Berry et al,¹⁵ Dollberg et al,¹⁴ and Emond et al¹⁶ showed no significant difference following frenotomy or nondivision (P = 0.46; 95% CI, −0.98 to 0.44) (Table ​(Table5;5; Fig. ​Fig.5).5). Buryk et al¹⁷ demonstrated lower mean Short-Form McGill Pain Questionnaire scores in mothers of infants who underwent frenotomy (mean = 4.9) when compared with those who underwent nondivision (mean = 13.5; P < 0.00001) when imputed SDs were used for statistical analysis. In overall meta-analysis, frenotomy resulted in a significant decrease in maternal pain during breastfeeding across multiple validated scales (P = 0.004; 95% CI, −0.70 to −0.13) (Fig. ​(Fig.66).

Timing

Wakhanrittee et al²⁸ found no difference in success of exclusive breastfeeding at 3 months between frenotomy done before and after 24 hours of life nor a difference in the median age of infants at the time of frenulotomy who successfully exclusively breastfed at 3 months of age compared with those who did not exclusively breastfeed at 3 months. Steehler et al²⁹ concluded similarly in a retrospective study which utilized a prospective telephone survey. They found that there was no difference in the total length of breastfeeding between groups that underwent frenotomy within the first week of life or after the first week of life, but there was a subjective benefit, as reported by the mother, in the infant’s ability to feed if frenotomy was performed within the first week of life. In a small study by Praborini et al,³⁰ it was found that infants who underwent frenotomy before 8 days of age gained significantly more weight than those infants who underwent frenotomy after 8 days. A retrospective study by Sharma and Jayaraj³¹ found improved breastfeeding unrelated to whether division was before 30 days of age or after.³¹

Technique

Tongue-tie, when identified in neonates, is typically divided as an in-office procedure, with the use of topical or no anesthesia. Most studies outline a methodology that includes holding the child’s shoulders back and using blunt-ended scissors to divide the lingual frenulum.

In a randomized trial of patients older than 3 years of age with constrained articulation, 4-flap z-frenuloplasty was compared with the traditional horizontal-to-vertical frenuloplasty³² and demonstrated greater absolute increases in frenulum length, tongue protrusion, and speech articulation measured by 2 speech pathologists when >10 months from the operative tongue-tie release. Z-frenuloplasty demonstrated better outcomes at 3 months following tongue-tie release than simple frenotomy in areas of parental satisfaction, tongue mobility and elongation, and speech articulation in a randomized trial.¹⁹ Choi et al³³ found that z-frenuloplasty with partial myotomy of the genioglossus resulted in subjective speech improvement, and Klockars and Pitkäranta³⁴ found that frenuloplasty required no reoperations, compared with the frequency of reoperation in patients following simple frenotomy. Frenulectomy was rarely described, and no data suggested its superiority over frenotomy or frenuloplasty.

Use of Anesthetic

A double-blinded randomized study by Shavit et al³⁵ compared the use of topical 2% tetracaine or 20% benzocaine before frenotomy and found no difference between the 2 anesthetics in the first 30 days of life; furthermore, the study concluded that both anesthetics incompletely controlled infant pain. Another double-blinded study showed no difference between the use of topical benzocaine and no anesthesia during the procedure in infants 0–3 days of age.³⁶ When the mean age of patients in all of the studies discussed in this review are plotted, the use of topical or no anesthetic is generally utilized in approximately the first month of life (Fig. ​(Fig.7),7), with tongue-tie division under no anesthetic being performed up to the 115th day of life (Table ​(Table6).6). Operative general anesthesia was utilized in patients from the 5th to the 3614th day (9.9 years) of life, with the mean ages ranging from 960 to 2065 days (or 32 months to 5.65 years).

Intervention versus Nonintervention (Grade: B)

It is clear that the use of a scoring system in which functional considerations are taken into account is beneficial, as anatomic considerations alone do not correlate as strongly with breastfeeding difficulty.^37,38 In the current literature, there is a lack of uniformity in the grading of ankyloglossia. For this reason, we propose the use of the HATLFF scoring system in assessment of ankyloglossia, which has been shown to correlate with breastfeeding difficulty and success of intervention, and more specifically, the use of the “function impaired” HATLFF score as an indication for division.^39–41 Some infants cannot be classified by the HATLFF scoring system,^40,42 so infants who are experiencing breastfeeding difficulty should also undergo division, as our meta-analysis has shown that division is associated with breastfeeding improvement.

Age at Time of Intervention (Grade: D)

There is a lack of consistent empirical evidence to guide the timing of intervention in infants with tongue-tie, with available studies evaluating different metrics at varied time points. To facilitate breastfeeding, frenum release must necessarily be performed as early as possible in infancy. One study suggested infants over 4 months of age become more strong and aware, and division should be performed under general anesthesia.³⁹ This is an area requiring further study, but it is clear that there is an improvement in breastfeeding following division; therefore, ankyloglossia should be addressed as early as possible in affected infants.

Procedure for Division (Grade: B or C)

Randomized controlled trials have shown that z-frenuloplasty and 4-flap frenuloplasty are superior to simple frenotomy across multiple metrics but must be performed under general anesthesia. The benefit of this procedure must be weighed against the potential dangers that general anesthesia can pose to infants. No identified evidence supports the superiority of frenulectomy over either frenotomy or frenuloplasty. Because of these considerations, we only recommend performing z-frenuloplasty or 4-flap frenuloplasty under general anesthesia, which is more safely performed in childhood rather than in infancy.

Synthesizing this evidence about type and timing of procedure, we propose a treatment algorithm that identifies 2 ideal opportunities for addressing symptomatic ankyloglossia (Fig. ​(Fig.8).8). The first is in infancy, when frenotomy of a thin band of tissue can be performed with local anesthesia in the clinic setting and, in many cases, facilitate improved breastfeeding. The second is in childhood, when more extensive frenuloplasty can be performed more safely under general anesthesia. The evidence suggests either of these options is likely superior to a procedure under general anesthesia in infancy or a simple frenotomy without general anesthesia in childhood. However, we advocate for this algorithm being but one factor in a conversation between a physician and a patient’s family, with treatment tailored to the clinical situation.

This study has several notable limitations. There were few articles that met inclusion criteria, mainly due to the lack of randomized control trials on this topic. Further, there was significant variability among outcome measures used across studies, leading to incomparability. Future studies may benefit from using standardized outcome measures in measuring the degree of tongue-tie and outcome measures, so that strong conclusions can be drawn as to the indications and effectiveness of tongue-tie division on pediatric patients.

2. International Affiliation of Tongue-Tie Professionals (IATP). About tongue-tie.Accessed December 20, 2020. Available at http:www.tonguetieprofessionals.org/about-tongue-tie/.

6. Horta BL, Victora CS.Short-term effects of breastfeeding: a systematic review on the benefits of breastfeeding on diarrhea and pneumonia mortality.Available at http://apps.who.int/iris/bitstream/10665/95585/1/9789241506120_eng.pdf?ua=1. Accessed February 26, 2018.

7. Horta BL, Victora CS.Long-term effects of breastfeeding: a systematic review.Available at http://apps.who.int/iris/bitstream/10665/79198/1/9789241505307_eng.pdf?ua=1. Accessed February 26, 2018.