How to Compare Mesh Studies in Breast Reconstruction (and Why 9% and 38% Can Both Be Right)

Put five published series of the same material side by side and the headline complication rates can range from under 10% to nearly 40%. That does not mean the material behaves unpredictably. It usually means the studies measured different things, in different patients, over different lengths of time. Comparing mesh studies is a skill, and most of the arguments about which matrix is “better” are really arguments about which numbers can legitimately be compared.

This article gives a working method for reading mesh and matrix studies in breast reconstruction, then applies it to the published bovine pericardium literature, the evidence base closest to ExaShape®, so the method can be seen at work on real numbers.

Seven questions before comparing any two numbers

Design. Randomised comparisons of support materials in breast reconstruction are few and small. Most of the evidence is retrospective, which means the material and the patient were chosen together by the surgeon. A retrospective series can describe what happened; it cannot, on its own, attribute it to the material.

Plane and staging. A matrix extending a muscle in a two-stage subpectoral reconstruction is doing a different job, under different loads, from a matrix forming the entire pocket in a direct-to-implant prepectoral one. The two Dutch and Swedish randomised trials of porcine dermis were subpectoral;^1,2 their results do not describe prepectoral practice.

Population. Flap thickness dominates. A series restricted to thin flaps will report more rippling than one restricted to thick ones, whatever the material.

Follow-up. Different complications appear on different clocks, which the next figure makes explicit.

Definitions. “Complication rate” can mean any event, only events needing treatment, or only those needing surgery; it can be counted per patient or per breast. Rippling is a complication in some papers and an aesthetic finding in others.

People. The Massachusetts General Hospital DTI series showed complications halving between the first year of a technique and later years.³ Single-surgeon series, early-adoption series and series whose authors consult for the manufacturer all deserve that context, and good papers state it.

Follow-up decides what a study can see

The method, applied: the bovine pericardium literature

Bovine pericardium has a small but growing clinical literature in breast reconstruction. Laid out with the seven questions, the apparent spread in results becomes readable.

Applied to these series, the questions explain the spread. The Danish series’ 38.2% counts every Clavien-Dindo event, including minor ones, and its authors drew the conclusion that careful selection and perfusion imaging reduced risk.⁴ The Spanish series included many patients outside conservative selection criteria and reported a low failure rate.⁵ The Italian thin-flap series selected for exactly the patients most likely to ripple and reported rippling accordingly.⁶ And the multicentre Italian pocket series studied direct-to-implant prepectoral reconstruction with a pre-shaped pocket over the longest prepectoral follow-up in the group.⁷

None of these is a comparison against another material, apart from the small Chinese study, which compared pericardium-assisted reconstruction with latissimus dorsi and two-stage reconstruction and found similar complication rates, BREAST-Q scores and hospital costs.⁸ The honest reading is that the bovine pericardium literature shows the material class is safe and usable in prepectoral reconstruction across several countries and several patient groups; it does not establish superiority over any other class, and it does not yet contain the long follow-up needed to describe late contracture with confidence.

Where head-to-head data exist

Direct comparisons between materials are rare, and the few randomised ones are small. In a Swedish within-patient randomised trial, each woman having bilateral reconstruction received a biological matrix (a bovine pericardium product) on one side and an absorbable synthetic mesh on the other. Across 48 breasts, seroma was markedly more frequent on the biological side, 38% against 3.8%, and implant loss was 8.5% against 2% without reaching significance.⁹ In an Austrian pilot trial of 48 patients, an acellular matrix and a titanised synthetic mesh had similar overall complication rates, but more implant losses occurred in the matrix group.¹⁰

Applying the questions to these studies matters as much as applying them to the pericardium series: both were small, both used specific products in specific planes, and neither was designed to settle the class question. They are discussed further in biologic versus synthetic mesh.

Configuration is a variable too

The last reason two studies of “the same material” disagree is that the material was not used the same way. A before-and-after series using a dermal matrix in prepectoral reconstruction reported a large fall in major complications after changing how the matrix was fixed, with no change in the product.¹¹ When studies are compared, the configuration of the material, wrap, partial cover or sling, fixed or unfixed, fenestrated or solid, deserves the same attention as its source tissue. It is also the argument for devices supplied pre-configured, so that the shape is designed rather than improvised at the table, which is the thinking behind the pre-shaped ExaShape® pockets.

The short version

Before comparing two mesh numbers, check the design, the plane, the staging, the population, the follow-up, the definitions and the people. Applied to the bovine pericardium literature, that method turns an apparently scattered set of results into a coherent picture: a material class used safely in several settings, with outcomes that track selection and technique, and with long-term data still to come.

References

1. Dikmans REG, Negenborn VL, Bouman MB, et al. Two-stage implant-based breast reconstruction compared with immediate one-stage implant-based breast reconstruction augmented with an acellular dermal matrix: an open-label, phase 4, multicentre, randomised, controlled trial. Lancet Oncol. 2017;18(2):251-258. doi:10.1016/S1470-2045(16)30668-4
2. Lohmander F, Lagergren J, Roy PG, et al. Implant based breast reconstruction with acellular dermal matrix: safety data from an open-label, multicenter, randomized, controlled trial in the setting of breast cancer treatment. Ann Surg. 2019;269(5):836-841. doi:10.1097/SLA.0000000000003054
3. Colwell AS, Damjanovic B, Zahedi B, Medford-Davis L, Hertl C, Austen WG. Retrospective review of 331 consecutive immediate single-stage implant reconstructions with acellular dermal matrix: indications, complications, trends, and costs. Plast Reconstr Surg. 2011;128(6):1170-1178. doi:10.1097/PRS.0b013e318230c2f6
4. Bentzon TT, Abu-Khalil M, Venzo A, Matzen DR, Damsgaard TE. Immediate prepectoral breast reconstruction using bovine pericardium matrix. Dan Med J. 2025;72(2). doi:10.61409/A08240522
5. Nogueira Sixto M, De Castro Parga G, Lamas González MJ, et al. The role of bovine pericardium matrix in immediate prepectoral breast reconstruction: insights from the largest published retrospective series. JPRAS Open. 2025;46:278-289. doi:10.1016/j.jpra.2025.09.005
6. Casella D, Rocco N, Luridiana G, et al. Enhancing breast reconstruction with bovine pericardium: a preliminary STEP (Surgical Techniques and Efficacy in Pericardium use) towards improved outcomes. J Clin Med. 2025;14(17):6296. doi:10.3390/jcm14176296
7. De Vita R, Mangialardi ML, Pozzi M, et al. A pericardium bovine matrix pocket in DTI prepectoral breast reconstruction. Clin Breast Cancer. 2024;24(7):e613-e621. doi:10.1016/j.clbc.2024.06.004. Disclosure: the first author is a consultant for ExaShape.
8. Wang S, He S, Zhang X, et al. Acellular bovine pericardium matrix in immediate breast reconstruction compared with conventional implant-based breast reconstruction. JPRAS Open. 2021;29:1-9. doi:10.1016/j.jpra.2021.03.005
9. Hansson E, Edvinsson AC, Elander A, Kölby L, Hallberg H. First-year complications after immediate breast reconstruction with a biological and a synthetic mesh in the same patient: a randomized controlled study. J Surg Oncol. 2021;123(1):80-88. doi:10.1002/jso.26227
10. Gschwantler-Kaulich D, Schrenk P, Bjelic-Radisic V, et al. Mesh versus acellular dermal matrix in immediate implant-based breast reconstruction: a prospective randomized trial. Eur J Surg Oncol. 2016;42(5):665-671. doi:10.1016/j.ejso.2016.02.007
11. Tettamanzi M, Arrica G, Ziani F, et al. Prepectoral breast reconstruction with prosthesis and acellular dermal matrix: a new technique of ADM implantation and fixation. Plast Reconstr Surg Glob Open. 2024;12(6):e5895. doi:10.1097/GOX.0000000000005895

This article is educational material for healthcare professionals. Series cited for the bovine pericardium material class used several different commercial products; results from one product should not be attributed to another. No clinical superiority is claimed for any product. ExaShape® is CE marked per EU MDR 2017/745 and is for professional use only.