The resurgence of interest in breast feeding has been accompanied by a lively debate about the significance of “tongue tie” or ankyloglossia. Symptoms attributed to tongue tie include nipple pain and trauma, difficulty in the baby attaching to the breast, frequent feeding, and uncoordinated sucking. These problems may result in the mother deciding to terminate breast feeding prematurely, slow weight gain for the baby, and even hypernatraemic dehydration. Speech defects have also been attributed to tongue tie. Strong views have been expressed by many eminent authors on the subject (box 1).

This paper reviews what is known about tongue movements and the significance and treatment of tongue tie. It is based on two literature reviews, one conducted on behalf of NICE^1,2 by one of us (MR) and updated by further searches of published and grey literature and conference abstracts. The publications reviewed for this paper are summarised in table 1.

As our review found little high quality objective evidence, we begin by making explicit the personal experience and bias with which we commenced the review. One of the authors (MR) felt that tongue tie is an important issue—she experienced pain for many weeks while breast feeding her first child, who exhibited features said to be typical of tongue tie, and has since discussed this issue widely with lactation specialists and women having similar problems. The other (DH) accepted that ankyloglossia occurs in dysmorphic infants¹⁰ and occasionally in otherwise normal babies,¹¹ but was sceptical about the high prevalence of the condition now being described by several authors.

ANATOMY AND PHYSIOLOGY

The tongue is a highly mobile organ made up of longitudinal, horizontal, vertical, and transverse intrinsic muscle bundles. The extrinsic muscles are the fan-like genioglossus which is inserted into the medial part of the tongue and the styloglossus and hyoglossus into the lateral portions. The sub-lingual frenulum is a fold of mucosa connecting the midline of the inferior surface of the tongue to the floor of the mouth. Tongue tie is the name given to the condition arising when the frenulum is unusually thick, tight, or short. There are many variations and differing degrees of severity (fig 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

 Six examples of babies diagnosed as having tongue tie, showing the variation in the thickness and insertion of the frenulum (reproduced with kind permission from Carolyn Westcott, Princess Anne Hospital, Southampton).

The movements of the tongue during infant feeding have been studied by cine-radiography and more recently by ultrasound.^12,13 Ultrasound reveals some similarities between the movements made by the baby when either breast or bottle feeding,¹⁴ but also some important differences.¹⁵ The tongue is projected further forward in breast feeding¹⁶ and the human nipple elongates during each suck in a way that an artificial teat cannot do.¹⁴ During feeding, the artificial teat, or the nipple together with some breast tissue, is held fully in the mouth with the tongue covering the lower gum ridge. The nipple is protected from damage and pain at the back of the baby’s mouth.¹⁶ The baby’s lower jaw is then elevated, compressing the artificial teat, or the breast immediately behind the nipple, while the front of the tongue moves up to aid the expression of milk. In breast feeding, this is by compression of the milk ducts under the areola. A wave of upward movement of the medial part of the tongue progresses backwards, and the expression of the milk is further facilitated by negative pressure generated by downward movement of the back of the tongue and the lower jaw and, in breast feeding, by the active expulsion of milk once the let down occurs.

In coordinated feeding, the sucking, swallowing and breathing movements follow in a 1:1:1 sequence. This can take several days to become established in healthy full term infants. In pre-term infants and in some term infants a variety of poorly coordinated feeding movement patterns are observed and sometimes persist.¹⁷ Antenatal ultrasound studies show that mouth and tongue movements are already well developed in association with intra-uterine yawning and crying.^18,19

HYPOTHESES

Review of the literature and expert opinion gave rise to the following hypotheses:

• Tongue tie is a definable condition

• Tongue tie affects 3–4% of infants

• The tight frenulum prevents the infant from getting the tongue over the lower lip and gum ridge and therefore can cause feeding problems, particularly affecting breast feeding, leading to pain for the mother and poor infant weight gain; it can also affect bottle feeding

• The impact of a tight frenulum varies between mother–baby dyads

• A tight frenulum can also cause problems in older children and adults, involving speech, dental hygiene, licking ice cream, and French kissing

• Division of the frenulum (frenulotomy) is a low risk effective treatment

• The condition is genetic

CASE DEFINITION

Can tongue tie be defined—and to what extent do individual observers agree on the diagnosis? The length of attachment of the frenulum varies widely. In some babies it extends to the tip of the tongue. There may be an indentation of the anterior edge, referred to as a heart shaped tongue. The appearance of the tongue is not sufficient on its own to make a diagnosis, as the thickness and elasticity of the frenulum also vary widely and affect the extent to which normal tongue movements are inhibited (see fig 1).

In four published studies of tongue tie in babies, the initial selection of possible cases from the whole newborn population was based only on appearance and was done by one individual (Ballard and colleagues⁵), or by one individual with confirmation of positive diagnoses by one other observer (Messner and colleagues³), or accomplished by providing staff with a selection of photos (Hogan and colleagues,⁴ Ricke and colleagues⁶). We have not found any formal data on observer agreement or variation in this process. The method of selection in the study by Masaiti and Kaempf²⁰ is unclear but cases were probably selected on the basis of breast feeding problems. Two authors (Messner,³ Ricke⁶) aimed to reduce bias by trying to avoid specific mention of tongue tie to mothers but acknowledge that this was difficult.

All authors agreed that function is more important than appearance and Hazelbaker designed an Assessment Tool for Lingual Frenulum Function (ATLFF).²¹ Ballard used this tool but did not examine inter-rater reliability. Ricke et al found that the inter-rater agreement using the ATLFF was only moderate and that many infants did not fit in any of the categories defined by Hazelbaker.

BIRTH PREVALENCE OF TONGUE TIE

There is agreement among authors that tongue tie is found in around 3–4% of babies, with the exception of Hogan et al who report 10.7%. All report a male preponderance with ratios varying from 1.5:1 to 2.6:1.

THE IMPACT OF TONGUE TIE

Maternal pain during feeding, sometimes accompanied by trauma, and difficulty in the baby taking the breast, are the main breast feeding problems attributed to tongue tie.^3–6 Attributing pain during breast feeding to tongue tie is not straightforward, however, since pain is a common problem that can result from several other causes, including attachment problems unconnected with tongue tie, and infection.^16,22–24 Ricke and colleagues⁶reported that more tongue tied infants than controls were bottle fed at one week but there was no significant difference at one month, though attrition meant that numbers were small. Messner et al³ found no difference in the rate of breast feeding between tongue tied infants and controls at 2 months but a significant difference in the numbers of mothers reporting problems with breast feeding.

Ramsay⁷ measured the distance from nipple tip to the junction of the hard and soft palate by sub-mental ultrasound. The distance decreased from 7.99 mm (±2.80) to 6.49 mm (±1.87) seven days after frenulotomy. This change, though statistically significant, is small and its practical significance is unknown. The tongue movements were said to become “more normal”.

VARIATION BETWEEN DYADS

Hogan and colleagues⁴ found that more than half of babies with tongue tie had no problems breast feeding but could not show any correlation between severity of tongue tie and feeding difficulty. This is perhaps surprising, but it may be that only a small shift in positioning on the breast is sufficient to eliminate pain and improve feeding. Ricke and colleagues⁶ reported that 80% of tongue tied infants were breast feeding successfully at one week. It is of interest that in two case series of older children presenting with speech difficulties and other problems attributed by the authors to tongue tie, 21/25 mothers (Messner and Lalakea⁸) and 80% of an unspecified number (Fernando⁹) who were asked about breast feeding reported no significant difficulties.

OTHER PROBLEMS

Several case series report a range of other problems in older children associated with ankyloglossia—speech defects, difficulty in licking the lips or in kissing, dribbling, etc. These are difficult to evaluate as the authors do not give details of the catchment population, referral patterns, or detailed criteria for inclusion in the series.

INTERVENTION

All authors agree that frenulotomy in the newborn is a low risk minor procedure, performed without anaesthetic. The presence of the deep lingual vein just lateral to the midline means that significant venous bleeding could occur if technique is not meticulous but we found no reports of serious adverse events. In older children the procedure needs an anaesthetic and sometimes a frenuloplasty, which carries some risk of scarring.

Ballard and colleagues⁵ reported a marked fall in maternal pain scores after the procedure. Hogan and colleagues⁴ randomised their cases to immediate or delayed intervention and found that frenulotomy was much more effective than advice from a lactation counsellor. They reported dramatic and rapid, often immediate, improvement after the procedure in most of their cases; improvement was noted in 95% of babies. The measurement of outcomes was not blinded. The precise criteria for improvement were not specified. Improvement was not always immediate, but this could be due to the need for sore nipples to heal or for the baby to re-learn optimal patterns of suckling.

In case series of older children and adults, some striking improvements were noted after surgery, but many of the children showed only gradual or modest improvement, particularly where the articulation of speech was concerned; these children often needed continuing speech therapy and this was attributed to the need to un-learn established patterns of articulation.⁹ The absence of any comparison or control cases makes these reports impossible to evaluate and we found no comparative studies or randomised trials addressing the role of tongue tie or frenulotomy in older children.

IS ANKYLOGLOSSIA INHERITED?

To define the inheritance of a condition, a robust case definition is needed, but tongue tie varies markedly in severity and is not an all-or-none condition. When an anomaly is identified in a newborn infant, the family searches its collective memory for other similar cases but, in the case of tongue tie, it would be impossible to assess the validity of that diagnosis in retrospect. None of the studies we reviewed considered these issues and none had gathered systematic family data across a number of families with and without the condition. Notwithstanding the comments made in several papers, no conclusions can currently be drawn about family history.

AN OVERVIEW

Tongue tie is at first glance a minor issue, but from the results of the only randomised controlled trial yet conducted, Hogan and colleagues⁴ suggest that at least 3% of newborns (57/1866) would benefit from frenulotomy and that this would increase the rate of continuing breast feeding. Most of the literature on tongue tie has been in connection with breast feeding; however, Hogan et al report that of the 57 babies in their study who benefited from frenulotomy, 40 were breast fed but 17 were artificially fed. If they are correct, this is a very common congenital anomaly that affects both breast fed and bottle fed babies, and up to 18 000 such procedures should be performed each year in the UK. It is therefore important to ask whether the evidence supports that rate of intervention and to scrutinise the evidence with a particularly critical eye.

There were a number of methodological problems with most of the studies we reviewed. These included:

• Inadequate assessment of inter-observer reliability of the initial diagnosis, the dynamic assessment of feeding and the maternal symptoms

• Ethical and practical difficulties in concealing the suspected diagnosis from the mother, thus potentially introducing a bias by raising the expectation of breast feeding problems and of improvement from intervention—this is, however, a common limitation in most studies of breast feeding problems

• Poorly defined outcome measures; it is particularly difficult to establish an objective assessment of improvement, when the primary outcome measure is reduction in maternal pain during breast feeding

• The dilemma of when to assess and intervene for tongue tie; if done very early, before breast feeding is established, as in the Ballard et al study, improvements may be wrongly attributed to the procedure (because suckling efficiency improves over the first few days and weeks¹³), but if done later (as in Hogan et al), many mothers may already have sore nipples or have given up breast feeding.

CONCLUSIONS

We began this review by stating our personal bias. While DH confesses to still being somewhat more sceptical than MR, we are in complete agreement on the following conclusions:

• Individual case histories suggest that some babies do have a tight frenulum (tongue tie) which can inhibit breast feeding and, in some cases, bottle feeding as well

• Although frenulotomy is a simple low risk procedure, it should be carried out only by those who have been trained in the procedure²⁵

• It can be justified only if it is likely to lead to significant improvement in the comfort and the continuation of breast feeding, or of other longer term problems for the child

• We do not know the true prevalence of significant tongue tie

• There is no evidence one way or the other about inheritance

• On current evidence, there is no justification for actively searching for tongue tie during routine examination, but when mothers are having difficulty in breast feeding this should be considered as one of several possible causes

• The diagnosis should rest primarily on observation and analysis of feeding difficulties rather than the static appearance of the tongue

• It may be wise to be particularly cautious in making this diagnosis in the first two or three days before lactation is established

• The problem is of sufficient interest and importance to merit further studies of both breast and bottle fed babies, in which more precise case definition, measures of inter-observer reliability of pre- and post-intervention assessment, and ultrasound imaging are likely to play a key role

• Given the evidence that breast feeding has many advantages for both mother and baby, funding should be sought for carefully planned definitive studies on the issue.