Culture media for human pre-implantation embryos in assisted reproductive technology cycles

Numerous media are commercially available for culture of human pre-implantation embryos in assisted reproductive technology (ART) cycles. It is unknown which culture medium leads to the best success rates after ART. This is the first update of a Cochrane review published in 2015.

To evaluate the benefits and harms of different human pre-implantation embryo culture media used for in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) cycles, including intracytoplasmic morphologically selected sperm injection (IMSI).

We searched the Cochrane Gynecology and Fertility Group's Specialized Trials Register, Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, PsycINFO, the National Research Register, the Medical Research Council's Clinical Trials Register, and the NHS Centre for Reviews and Dissemination databases on 28 August 2025. We also examined the reference lists of all known primary studies, review articles, citation of relevant publications, and abstracts of major scientific meetings, and we contacted experts in the field for additional studies.

We included all randomized controlled trials that randomized women, cumulus-oocytes complexes, oocytes, or embryos, and compared any two or more commercially available culture media for human pre-implantation embryos in IVF, ICSI, or IMSI protocols.

Critical outcomes: cumulative live birth, live birth, and ongoing pregnancy (confirmed by heartbeat on ultrasound). Important outcomes: clinical and multiple pregnancy, miscarriage, preterm birth, birth weight, and congenital malformation. These outcomes were analyzed in the review, but we did not prioritize them for inclusion in our summary of findings tables; therefore, they are not reported in the synthesis of results section below.

To assess the risk of bias in the studies, we used the Cochrane tool RoB 1.

Two review authors independently selected studies, assessed their risk of bias, and extracted their data. We sought additional information from the study authors where necessary. Due to heterogeneity, we presented individual study results and did not conduct meta-analysis. We calculated effect estimates as odds ratios (OR) with 95% confidence intervals (CI) using the Mantel-Haenszel fixed-effect model. We assessed the certainty of the evidence using the GRADE method.

We screened 5779 search records and included 26 studies published as full articles in this updated review. Thirteen studies randomized 3913 women, two randomized 361 cycles, and 12 randomized 30,542 cumulus oocyte complexes, oocytes, or embryos. One study randomized women and oocytes.

Of the 26 included studies, eight studies (3315 women) were suitable for analysis of our critical outcomes: cumulative live birth, live birth, and ongoing pregnancy. However, we were unable to pool data because no appropriately designed studies compared the same culture media. Only one study reported cumulative live birth rate and suggested that G5 likely increases live birth compared to HTF (human tubal fluid) (44.1% versus 37.9%; OR 1.29, 95% CI 0.98 to 1.70; 836 participants; moderate-certainty evidence). Six studies (3063 participants) had data for the analysis of live birth rates, providing only very low certainty evidence. It is unclear if there is a difference in live birth rates between different culture media: G5 (22.3%) versus HTF (18.1%) (OR 1.30, 95% CI 0.92 to 1.82; 836 participants; fresh embryo transfers only from first cycle); Sydney IVF cleavage (21.7%) versus G1 series 3 (22.0%) (OR 0.98, 95% CI 0.63 to 1.54; 449 participants); G2 (26.4%) versus BlastAssist M1/M2 (18.0%) (OR 1.64, 95% CI 0.80 to 3.37; 176 participants); GM501 (16.5%) versus Sydney IVF (18.8%) (OR 0.85, 95% CI 0.39 to 1.87; 170 participants); and Sydney IVF (34%) versus EmbryoGen/BlastGen (22%) (OR 1.83, 95% CI 0.75 to 4.44; 100 participants). We presented the results of the sixth study in separate groups: EmbryoAssist (18.4%) versus EmbryoAssist plus GM-CSF (granulocyte-macrophage colony-stimulating factor) (24.0%; low human serum albumin (HSA) concentration) (OR 0.71, 95% CI 0.50 to 1.02; 743 participants) and EmbryoAssist (23.8%) versus EmbryoAssist plus GM-CSF (26.1%; high HSA concentration) (OR 0.96, 95% CI 0.67 to 1.38; 589 participants). Five studies (2520 participants) reported (cumulative) ongoing pregnancy rates. We judged one study to provide moderate-certainty evidence. This study showed a cumulative ongoing pregnancy rate of 45.3% with G5 compared to 38.7% with HTF (OR 1.32, 95% CI 1.00 to 1.73; 836 participants). We judged the remaining four studies to provide very low certainty evidence. Three of them reported the following ongoing pregnancy rates: GM501 (21.2%) versus ISM1 (21.8%) (OR 0.96, 95% CI 0.46 to 1.99; 172 participants); Sydney IVF (36.0%) versus EmbryoGen/BlastGen (22.0%) (OR 1.99, 95% CI 0.82 to 4.83; 100 participants); and ECM/Multiblast (50.0%) versus Global (70.0%) (OR 0.43, 95% CI 0.17 to 1.07; 80 participants). We presented the results of the fourth study in separate groups: EmbryoAssist (21.3%) versus EmbryoAssist plus GM-CSF (30.0%; low HSA concentration) (OR 0.63, 95% CI 0.45 to 0.88; 743 participants) and EmbryoAssist (28.1%) versus EmbryoAssist plus GM-CSF (29.0%; high HSA concentration) (OR 0.96, 95% CI 0.67 to 1.38; 589 participants). In the eight studies contributing to our critical outcomes, we assessed most risk-of-bias domains as unclear. The evidence was limited by methodological problems and incomplete reporting. Outcomes were inconsistently measured, few studies used appropriate designs for live birth outcomes, and cumulative live birth was rarely assessed. Variation in culture conditions further limited comparability across studies.

Embryo culture media are important for embryonic development and subsequent success of in vitro fertilization cycles. Determining which is the most appropriate embryo culture medium has been challenging due to many factors involved in assessing efficiency. Numerous studies have been performed, but no two studies that use an appropriate design and assess clinical outcomes have compared the same culture media. One study suggested that G5 medium probably increases cumulative live birth compared with HTF medium; however, the overall evidence base remains limited and inconsistent. We conclude that there is insufficient evidence to support or refute the use of any specific culture medium for human embryo culture. Properly designed and executed randomized trials are necessary.

This Cochrane review had no dedicated funding.

Protocol (2009) DOI: 10.1002/14651858.CD007524 Original review (2015) DOI: 10.1002/14651858.CD007876.pub2.