10/04/2026
A new Journal of Human Lactation paper should give a lot of people pause.
This study compared newborns with tongue tie, newborns with upper lip tie, and newborns with no ties at all. It did not rely only on a quick look inside the mouth or on subjective impressions. It used ultrasound during breastfeeding, alongside LATCH scores and maternal ni**le pain scores, to examine what was actually happening.
Its findings were clear. There were no significant differences between the groups in LATCH scores or maternal ni**le pain. Ultrasound identified some differences in sucking dynamics, but those differences did not translate into worse breastfeeding scores or more pain across the groups. In plain terms, visible anatomical variation did not reliably predict poorer breastfeeding outcomes (Temizel et al., 2026).
This sits very comfortably beside what Dr Robyn Thompson has been arguing for years. Breastfeeding is unique. It is genetic. Every baby has a unique oral function.
The current conversation around tongue tie, lip tie, buccal tie, and so called tethered oral tissues has become increasingly simplistic. A visible frenulum is often treated as though it explains ni**le pain, unsettled behaviour, repeated feeding, reflux, colic, poor sleep, wind, milk transfer concerns, and a long list of other problems. But this is not what the evidence shows.
The American Academy of Pediatrics states that fewer than half of infants with physical findings consistent with ankyloglossia have breastfeeding difficulty. It also states that labial and buccal frenae are normal oral structures and that there is no evidence that surgical incision of buccal frenae improves breastfeeding. Upper lip ties remain uncertain in clinical relevance, and appearance alone is not a reliable basis for diagnosis or intervention (Thomas et al., 2024).
The evidence around upper lip tie is particularly weak. Prospective cohort research found no correlation between upper lip frenulum grade and breastfeeding success, maternal pain, or "poor latch". It also found no evidence that tongue tie and upper lip tie consistently occur together in the same infant (Shah et al., 2021).
So we need to be honest. The confidence with which some babies are being labelled does not match the quality of the evidence.
Women’s distress is real. Their pain is real. Their babies may be unsettled, difficult to soothe, or feeding in a way that feels exhausting and unsustainable. None of that should be dismissed. But it is one thing to recognise suffering. It is another to assign a cause too quickly.
This is where Dr Robyn Thompson’s work remains so important. Her research did not start with the assumption that a small fold of tissue must be the primary problem. It looked instead at what was happening during the first and early breastfeeds. In her 2016 study, ni**le trauma was significantly associated with facio mandibular asymmetry, ni**le malalignment and the cross cradle technique. Her conclusion was not that babies needed oral procedures. It was that commonly taught breastfeeding techniques appeared to interfere with intra oral function by restricting movement of the craniocervical spine and limiting the baby’s instinctive ability to draw the ni**le and breast tissue effectively without causing trauma (Thompson et al., 2016).
Her earlier review made the same broader argument. Breastfeeding has been increasingly medicalised and pathologised. Complicated and unnatural teaching techniques interfere with instinctive sensory and mammalian behaviours and contribute to pain, trauma, delay, and early cessation (Thompson et al., 2011).
This is the part of the conversation that still does not get enough attention.
Breastfeeding not determined by staring at a frenulum and assigning a grade. It is a dynamic physiological process that shouldn't be mathematical or medicalised.
When we ignore that physiology and instead focus narrowly on appearance, we risk missing the actual source of the problem.
The more recent Thompson method implementation study also strengthens this position. When a physiological method of breastfeeding support was implemented in hospital care, direct breastfeeding trends at discharge improved, and women who discharged hospital exclusively breastfeeding were more likely to continue exclusive breastfeeding at 3 months. The method centred cradle hold, mouth to ni**le alignment, baby led connection and seal, maternal fine tuning for symmetry, and a hands off midwifery role. It moved away from breast shaping, ni**le to nose, forceful manoeuvres, and restricting the baby’s head and neck (Allen et al., 2023).
Read beside the new Journal of Human Lactation paper, the picture becomes much clearer. Babies with tongue tie or upper lip tie did not automatically have worse breastfeeding outcomes. Some ultrasound differences were present, yes, but breastfeeding scores and pain scores were not significantly different. Even the authors acknowledged important limitations, including small sample size and short follow up (Temizel et al., 2026). However the builds on the knowledge we know to be true in clinical care.
So no, this paper does not prove that all frenula are irrelevant.
What it does show is that the appearance based model is far weaker than many people have been led to believe.
It also raises serious questions about the increasing readiness to move from diagnosis to surgical incision.
Some women do report short term improvement after the procedure. That cannot simply be ignored. But short term improvement is not the same thing as strong long term evidence. Pain in early breastfeeding often improves over time anyway. The Cochrane review found short term reduction in maternal ni**le pain, but no consistent evidence of improved infant breastfeeding outcomes. The evidence base was small and methodologically weak (O’Shea et al., 2017). The FROSTTIE trial then found no difference in breastfeeding outcomes at 3 months between babies who had the procedure and babies who received breastfeeding support without surgery.
Meanwhile, rates of diagnosis and incision continue to rise, while breastfeeding duration remains poor.
That should concern every clinician in this space.
It suggests we may be intervening more aggressively on infant oral tissue while failing to address the factors that are actually shaping breastfeeding outcomes in the early days: birth intervention, early separation, delay to the first breastfeed, forceful techniques, poor symmetry, maternal pain, and disruption of the baby’s instinctive oral function.
It is also important to say clearly that this is not, in the overwhelming majority of cases, an emergency. A newborn with a visible frenulum does not usually require a rushed procedure simply because someone can see tissue they believe is restrictive. If an older child later has a clearly demonstrated functional problem that genuinely requires surgical assessment, that is a different clinical situation. In that case, specialist review, proper anaesthesia, and postoperative pain relief are far more appropriate than treating the newborn period as a window for quick oral procedures based on uncertain criteria.
The real question is not whether a frenulum exists.
The real question is whether it is functionally significant, whether breastfeeding has been thoroughly assessed, whether oral function has been optimised, whether forceful and distorting techniques have been removed, and whether the mother and baby have been given the time and skilled support needed to establish breastfeeding physiologically.
Dr Robyn Thompson has been saying for years that if we improve symmetry, protect the cranio cervical spine, stop forcing babies onto the breast, and work with mammalian physiology rather than against it, outcomes improve.
The new Journal of Human Lactation paper does not weaken that argument.
It strengthens it.
