Case report of nipple shield trauma associated with breastfeeding an infant with high intra-oral vacuum

A 35 year old primiparous woman had persistent and severe nipple pain from the onset
of breastfeeding. The medical, family and psychosocial history was unremarkable with
no breast or nipple trauma, surgery or piercings, and the woman expressed motivation
to breastfeed.

The female infant was born at 37.1 weeks gestation after induction for fetal growth
restriction. The birth weight, 2600 g, was on the 10
^th
centile and subsequent weights tracked along the 5
^th
centile of the World Health Organization weight-for-age girls percentile chart 5]. The infant remained healthy and developmentally appropriate.

Due to attachment difficulties, hand expressed colostrum was syringe fed to the infant
until secretory activation on the fourth postnatal day when a nipple shield was introduced
and breastfeeding commenced. Nipple pain was experienced at the first breastfeed,
and continued during and randomly between breastfeeds. The pain was described as “pinching,
tight and raw pain” and superficial nipple trauma occurred regularly. Mild blanching
and erythema of the nipples were infrequently observed and did not match the typical
clinical picture of nipple vasospasm. Some breastfeeds were less painful than others,
and regular paracetamol (acetaminophen) reduced the pain marginally.

In the second postnatal month severe burning and shooting pains were experienced in
the right breast during and after feeds. Nipple swabs were taken for microbiological
culture and concurrent prescribed courses of oral flucloxacillin and fluconazole were
completed. The breast pain abated within a few days of taking the medications, but
the nipple pain continued.

Nipple shields (24 mm, Medela AG, Baar, Switzerland) were used regularly as they modified
the pain, although health professionals had repeatedly discouraged this due to concerns
about an association between nipple shield use and reduced breastmilk supply. Use
of an electric breast pump (Swing, Medela AG, Baar, Switzerland) on the default (average
?53 mmHg) or slightly higher vacuum setting felt “tender rather than painful…really
low, low level pain…nothing like breastfeeding.”

During the first three postnatal months several consultations were undertaken with
an obstetrician, a community child health nurse, international board certified lactation
consultants, and family physicians. Bacterial infection and candidiasis were considered
likely diagnoses and so nipple swabs were sent for microscopy and culture on three
occasions. However no pathogenic growth was detected and neither courses of oral flucloxicillin
nor oral fluconazole had an effect on nipple pain. (Details of the drug doses and
course durations are not available). Ultrasound examination was conducted to exclude
nipple and breast pathology, and no abnormalities were detected. Wearing of breast
warmers (Promix HB, Solma, Sweden) in cold weather provided some comfort between feeds.

At 3 months the mother and infant attended the practice of the first author in a final
attempt to identify the cause of persistent, severe nipple pain. The infant had been
exclusively breastfed from birth and was feeding every 3 to 4 h during the day with
an overnight inter-feed interval of 6 h. Examinations of the infant’s mouth, maternal
breasts and nipples were unremarkable. Test weighing was performed before and after
a clinically assessed breastfeed. Positioning and attachment at the breast appeared
satisfactory, and the mother was in obvious pain; pale, sweating and barely able to
talk throughout the breastfeed. Immediately post feed the nipple was not compressed
and there was mild areolar edema radiating 1 cm from the nipple base. The infant fed
from both breasts transferring 136 mL in 10 min.

Feeding with a nipple shield was reported to be less painful although it intensified
pain in the nipple tip with a sensation of “…the nipple being drawn out through the
end of the shield.” Further questioning revealed the formation of blisters on the
nipple tips that correlated with each of the nipple shield holes.

The mother was invited to complete measurement of her 24 h milk production and to
attend a research facility for measurement of the infant’s intra-oral vacuum. The
24 h breastmilk production was measured by test-weighing the infant before and after
each feeding from each breast on an electronic scale (BabyWeighâ„¢, Medela Inc, McHenry
IL, USA, resolution 2 g, accuracy?±?0.034 %) for a period of 24 h plus one breastfeeding.
A corrected 24 h production for each breast then was determined with measurements
of breastfeed amounts and milk production expressed in grams that is considered to
be nearly equivalent to mL. Normal milk production was confirmed (Table 1) 6].

Table 1. Breastfeeding and 24 h breast milk production profile for an exclusively breastfeeding
dyad with high intra-oral vacuum in the infant

Intra-oral vacuum was measured during breastfeeding using a small silicone tube (Supplemental
Nursing System, Medela AG, Baar, Switzerland) filled with sterile water with one end
positioned alongside the mother’s nipple and the other end attached via a silicone tube (650 mm, 4 mm) and a 3-way tap to a pressure transducer (SP854, Memscap,
Bernin, France) with disposable clip-on dome (MLA844, AD Instruments, Castle Hill,
Australia). High intra-oral vacuum was measured both during active sucking and when
pausing on the breast, both with and without use of the 24 mm Medela nipple shield,
(Table 2, Fig. 1). During direct breastfeeding average peak, baseline and pausing vacuums were 43,
107 and 307 % higher than reference values 4]. Nipple shield use appeared to normalize baseline and pausing vacuums, however peak
vacuums were 207 % higher, and blisters were observed on the nipple tips immediately
after feeding (Fig. 2). Sub-mental ultrasound observation of tongue movements during breastfeeding demonstrated
compression of the nipple base when the tongue depressed during milk flow.

Table 2. Intra-oral vacuum measurements during direct breastfeeding (breast) and breastfeeding
with a nipple shield. Values are reported as mean?±?standard deviation (range)

Fig. 1. Intra-oral vacuum traces for infant with high vacuum during (a) direct breastfeeding and (b) breastfeeding with a nipple shield, and (c) infant with expected range of intra-oral vacuum

Fig. 2. Blisters resulting from nipple shield use while breastfeeding an infant with high
intra-oral vacuum

The following management strategies were offered to limit or alter exposure to high-intra-oral
vacuum during breastfeeding.

1. Remove infant from the breast when non-nutritive sucking begins towards the end
of a breastfeed.

2. Replace some breastfeeds with bottle feeds of expressed breastmilk.

3. Trial alternative breastfeeding positions (such as lying down) to vary the location
of nipple base pressure.

4. Trial a larger nipple shield to reduce the intra-oral space and therefore reduce
vacuum as per Boyle’s law.

Follow up at 4 months revealed that the mother replaced some breastfeeds with feeding
of expressed milk to rest her nipples, and when convenient she breastfed lying down
as it varied the location of pain at the nipple base. The mother reported that she
felt better able to cope once she was provided with a definitive reason for the pain
and some options for managing it. The nipple pain had reduced with use of a 28 mm
Mamivac conical nipple shield (KaWeKo, Ditzingen, Germany). However while more comfortable
than a 24 mm nipple shield, the nipple skin was still being drawn through the holes
of the shield. The family was not available for a follow up measurement of the infant’s
intra-oral vacuum when using the larger nipple shield. They gradually weaned from
the nipple shield during the fifth month and by 6 months breastfeeding had become
completely comfortable and enjoyable. The mother later reported that she achieved
her goal of breastfeeding for 12 months.

Discussion

While the distinct pattern of nipple trauma suggested an intra-oral vacuum anomaly,
the ability to measure intra-oral vacuum during breastfeeding was key in identifying
the cause of this mother’s extreme and continuing nipple pain. Early diagnosis and
management of breastfeeding problems is important to prevent early weaning. However
in the absence of a clinical screening tool for intra-oral vacuum, a differential
diagnosis of high intra-oral vacuum may not be considered. It is likely that a proportion
of these cases wean due to unexplained painful breastfeeding. Current recommendations
for management are restricted to reducing the frequency and/or duration of breastfeeds
to limit exposure to high intra-oral vacuum.

While evidence for the effect of high intra-oral vacuum on nipple pain is emerging
4], its etiology is not well understood. As peak intra-oral vacuum during breastfeeding
coincides with downward movement of the posterior tongue and soft palate 3] it is possible that anomalous tongue or palatal movements contribute. The growing
infant’s oral cavity enlarges vertically so that the tongue, which fills the oral
cavity of a newborn, occupies a lower proportion of the oral space in older infants
7]. It is not known whether the changes in oral anatomy impact intra-oral vacuum over
time and may explain gradual resolution of symptoms of high intra-oral vacuum in the
reported case. Further studies that include simultaneous intra-oral vacuum measurements
and ultrasound imaging in infants of different ages will further our knowledge of
this phenomenon.

Differing vacuum patterns were observed between direct breastfeeding and nipple shield
use, with the latter associated with higher peak and normalized baseline and pausing
vacuums. While it is not known whether nipple shield use impacts intra-oral vacuum
for all infants, altered baseline vacuum may explain in part how nipple shield use
facilitates sustained attachment in preterm infants 8].