Size and Shape of pumping flanges: What the Scientific Literature Really Says

Size and Shape of pumping flanges: between marketing and clinical reality, what does the Scientific Literature Really Say?

The choice of flange size, and more recently, its shape, has long been presented as a central determinant of pumping comfort and efficiency. Yet, the evidence base underpinning current clinical recommendations remains surprisingly limited. Three recent publications allow for an update: a pilot study comparing two sizing methods (Anders et al., 2024), a study on flange opening angle (Sakalidis et al., 2020), and a conference presentation from Donna Geddes' group at the University of Western Australia (Gridneva et al., 2025). This article offers a critical synthesis of these three works, with particular emphasis on their methodological limitations and conflicts of interest.

1. Tunnel Size: Two Studies, Two Difficult-to-Reconcile Conclusions

Anders et al. (2024/2025): Flange Size Matters

This intra-subject crossover study (n = 36) compared, among parents exclusively pumping, a "standard-fit" flange (determined by manufacturer instructions, generally ≥ 21 mm) to a "smaller-fit" flange selected using the Flange FITS™ method. The latter starts from the nipple tip diameter and tests sizes both above and below. The most frequently used sizes in the "smaller-fit" group were 15 and 17 mm, compared to 21 and 24 mm in the standard group.

Results showed that with smaller flanges, milk volume was significantly higher (mean difference of 15 g per session, p = 0.004) and comfort was markedly improved (difference of 1.2 points on a 5-point scale, p < 0.001). The authors conclude that sizing must be individualized and that a smaller size, based on nipple tip measurement, can be used without compromising either volume or comfort, challenging the common belief that a too-small flange compresses milk ducts and reduces flow.

Gridneva et al. (2025): Efficacy Assessment of the Breast Shield Size

This presentation from the University of Western Australia's Centre for Human Lactation Research builds on previous work by the Geddes-Hartmann group on the ultrasound anatomy of the breast, which showed that duct diameter increases significantly during milk ejection. It compared, on a much larger sample (n = 157, 212 sessions), efficacy and comfort depending on whether the flange was correctly fitted (≥ 21 mm), not fitted, fitted at ≤ 18 mm, or too small (< 4 mm of space).

Notable result: despite an unintentional doubling of the recommended tunnel width (8 mm on average above the measured diameter) in part of the sample, no significant difference was observed in milk volume extracted, percentage of available milk removed (PAMR), or extraction efficiency. Only the maximum comfortable vacuum was significantly lower in the group using flanges that were too small. The authors conclude that correct fitting remains essential, but that a flange larger than necessary is not necessarily harmful: a conclusion opposite to that of Anders et al.

Key takeaway: These two studies, published months apart by different teams, arrive at opposing clinical messages regarding the benefit of reducing flange size. This reflects less a scientific contradiction than a sign of the fragility of the evidence base: small samples, short protocols, different populations, and non-overlapping outcome measures (grams of milk over a week vs. PAMR over a single 15-minute session).

2. Shape: The Flange Opening Angle

Sakalidis et al. (2020), a randomized crossover non-inferiority trial (n = 49), compared a flange with a 105° opening angle (oval shape) to the standard 90° flange (round shape). Results showed the 105° flange was non-inferior and even superior to the 90° for breast drainage (+3.9 to 4.5 PAMR points) and expressed volume (+9 mL in 15 minutes), with significantly better perceived comfort. The authors propose an anatomical explanation: lactiferous ducts, which are superficial and close to the nipple base, dilate by more than 50% during ejection; a wider angle would limit their compression by the flange rim.

These results are statistically robust (rigorous protocol, intention-to-treat and per-protocol analysis, independent blinded statistician), but the clinical magnitude of the effect remains modest: is a 4% drainage difference or 9 mL over a 15-minute session perceptible or relevant for a mother pumping occasionally? The authors themselves acknowledge this difference may not be clinically significant for occasional use, presenting it instead as a factor to combine with others (double pumping, maximum comfortable vacuum) for long-term pumping-dependent mothers.

3. Bias and Limitations: The Common Thread Across the Three Studies

This is perhaps the most important element for a critical reading of this research field.

Funding and Industrial Conflicts of Interest

• Sakalidis et al. (2020): The study was entirely designed, funded, and conducted by Medela AG, which markets the 105° flange tested. Four of the seven authors are Medela employees; only the CTU Bern statistician was truly independent and blinded to allocation. The authors themselves acknowledge this as "industry research" and attempt to limit bias by relying on published, peer-reviewed methodologies.

• Gridneva et al. (2025): Funded by an unrestricted grant from Medela AG to UWA; Donna Geddes, the last author, declares having sat on Medela's scientific advisory committee. The protocol also uses a Medela device and pumping mode (M2-PEP, Symphony pump).

• Anders et al. (2024): Funded internally (University of North Carolina Greensboro) rather than industrially, which is a positive point compared to the other two studies. However, a direct conflict of interest exists: co-author Jeanette Mesite Frem is the designer of the Flange FITS™ method tested in the study and receives commercial remuneration through training for healthcare professionals. She is therefore both judge and party to the method evaluated. The manufacturer Maymom LLC (whose flanges were used in the study) also provided equipment free of charge.

  
  

After the trainings, it appears that lactation consultants purchase Maymom flange size kits; thus, the question is there: is this training a Maymom manufacturer training under the guise of an IBCLC, who also provides CERPS points ?

Methodological Weaknesses Specific to Each Study

• Anders et al.: Convenience sample recruited via social networks and private clinics (likely selection bias towards parents already sensitized to pumping issues); only one week per condition without a wash-out period; self-reported comfort and volume measures, with no possible blinding (participants know which size they are using); pump settings (vacuum, frequency) not controlled or recorded; unrepresentative sample (89% with higher education, 89% married).

• Gridneva et al.: This is a conference abstract (ABREAST 2024) published in Proceedings (MDPI), not a full peer-reviewed article at the same level as a traditional journal article; thus, available methodological details are limited. The "too small flange" subgroup includes only 10 participants (12 sessions), insufficient to detect modest differences. The design is not randomized: group allocation depends on each participant's nipple morphology, introducing uncontrolled anatomical confounding factors.

• Sakalidis et al.: Very short pumping duration (15 minutes, two sessions spaced 7 to 21 days apart); no data collected on medium or long-term effects on milk production or risk of mastitis/engorgement, a limitation the authors explicitly acknowledge should be addressed in future studies.

None of the three studies followed participants beyond a few days to a few weeks. Yet the most clinically relevant question for families pumping long-term (premature infants, exclusive pumping) is the cumulative effect over weeks or months: repeated nipple trauma, impact on medium-term milk production, risk of mastitis. This gap remains.

4. What We Can Take Away IN PRACTICE

The postulate that a flange that is too small would necessarily compress ducts and reduce milk flow is not unambiguously confirmed by recent literature; conversely, the idea that a larger flange is always preferable for comfort is also not independently demonstrated.

Individualized fitting, tested in real pumping situations rather than on static nipple measurement alone, emerges as a convergent principle across the three studies.

Shape (opening angle) is a distinct factor from size that deserves consideration, but the magnitude of its demonstrated effect remains modest and comes from a single study funded by the manufacturer of the tested product.

The entire available evidence base on this topic is funded, wholly or in part, by breast pump manufacturers or the designer of a commercial flange sizing method: a situation that, as often in this field, calls for independent studies reproducing these results before making firm clinical recommendations.

(Please note: one of the study's authors is on a world tour delivering trainings to IBCLCs, awarding CERPS points for presenting the flange adjustment concept. An ethical question arises: Who finances this tour?)

In the end, let us remember that different flange shapes and sizes exist according to manufacturers, and depending on breast anatomy and nipple size, changing flanges may improve pumping yields.

For example, Maymom offers the "pano" shape for large breasts, the "saucer" or "crater" shapes to reduce lack of seal, available in sizes ranging from 10 to 28 mm.

Maymom flanges

As in many health-related sectors, breast pump and flange manufacturers can contribute to creating or accentuating certain "problems" to justify the marketing of increasingly segmented solutions.

This marketing logic should not forget that, historically, most mothers pumped their milk without pain and with satisfactory yields using standard flanges.

Pumpin Pal offers silicone flanges with a bent angle, allowing pumping in a reclined position rather than leaning forward.

Lactek proposes silicone flanges that would mimic a baby's suck by compressing the breast and creating a wave motion (although we know this suck model is obsolete, see Dr Pam Douglas website) and would be suitable for elastic nipples.

Medela is following the trend and has just released an 18 mm flange size.

References:

• Anders LA, Frem JM, McCoy TP. Flange Size Matters: A Comparative Pilot Study of the Flange FITS™ Guide Versus Traditional Sizing Methods. J Hum Lact. 2025;41(1):54-64.

  
  

• Sakalidis VS, Ivarsson L, Haynes AG, et al. Breast shield design impacts milk removal dynamics during pumping: A randomized controlled non-inferiority trial. Acta Obstet Gynecol Scand. 2020;99(11):1561-1567.

  
  

• Gridneva Z, Warden AH, McEachran JL, Perrella SL, Lai CT, Geddes DT. Efficacy Assessment of the Breast Shield Size. Proceedings 2025;112:7.