There is insufficient evidence that most complementary treatments for in vitro fertilization improve fertility

The press release reflects the study fairly accurately, but it simplifies some important nuances and, above all, presents certain conclusions as more definitive than they are in the article, where they remain uncertain or model-dependent.

The press release gets the essentials right: the meta-analysis included 85 reliable randomized clinical trials (RCTs). Of 157 potentially eligible trials, it excluded 72 and concluded that most add-ons have no proven benefit or insufficient evidence. It is also correct in highlighting that there are three interventions with weak signs of benefit: endometrial scratching, which could be described as “a minor scraping of the uterine lining to try to promote implantation,” EmbryoGlue (an embryo culture medium containing a component that is supposed to increase embryo implantation), and PICSI (a microinjection of sperm selected using a special, more physiologically similar culture medium).

Where the article becomes somewhat more optimistic than the original paper is in its language regarding these three add-ons, because the original paper emphasizes that the effect of EmbryoGlue on live births was not robust under the random-effects model, and that the evidence for PICSI and scratching is of moderate or low certainty depending on the outcome. Furthermore, the article summarizes the limitations well but does not emphasize enough that the paper itself discusses heterogeneity, publication bias, imprecision, and risk of bias across various sets of studies.

The overall methodological quality is good for a review on this topic: a comprehensive search across multiple databases, double screening, duplicate data extraction, assessment using Cochrane RoB 1 and GRADE, and sensitivity and heterogeneity analyses. A key strength is that the authors did not merely count studies; rather, they excluded trials with serious reliability issues, which is particularly relevant in assisted reproduction.

Even so, the article itself makes clear that many trials were small, outdated, or at high risk of bias, and that for several outcomes, the certainty was low or very low. Thus, while the quality of the study is high, the quality of the available evidence for some interventions remains limited.

Overall, the main conclusions are supported by the presented data. The assertion that most add-ons show no clear benefit is consistent with the pooled results for acupuncture, corticosteroids, endometrial receptivity testing, intralipid therapy, PRP (platelet-rich plasma), and PGT-A (preimplantation genetic testing for aneuploidies) in their usual use as it relates to the outcome of live birth or cumulative outcomes.

However, statements regarding “possible benefit” should be read with caution. In the case of EmbryoGlue, for example, the positive effect was not consistent across statistical models, and the article explicitly acknowledges this. For PGT-A, the reduction in miscarriage does not clearly translate into more cumulative live births, which means that its actual clinical utility remains debatable.

The authors took several methodological confounding factors into account, primarily through subgroup comparisons, sensitivity analyses, and the GRADE approach, and they also attempted to minimize bias by excluding studies with reliability issues. However, in this type of review, “confounders” are not only clinical; there is also confounding due to study design: different populations, different indications, different IVF protocols, and outcomes measured at non-equivalent time points.

There are several significant limitations: the TRACT checklist has not been validated; some excluded studies might have been reliable; many trials were small or outdated; and some of the evidence comes from populations or practices that may not reflect current IVF practices. In the second study—the website trial—generalizability is also limited because all participants lived in Australia.

This study aligns well with the previous literature and, in part, refines it. The article itself notes that previous Cochrane reviews were outdated, included retracted or problematic studies, and that this meta-analysis incorporated 28 new trials. In other words, it does not radically contradict previous evidence but rather makes it more rigorous by filtering based on reliability.

The practical conclusion is quite consistent with the well-known trend in assisted reproduction: many add-ons are used more for marketing, habit, or biological plausibility than for proven clinical benefit. What is new here is the combination of a more rigorous review with a parallel trial on patient information, which reinforces the idea that the problem is not only therapeutic but also informational.

For clinical practice, the message is that most add-ons should be offered with great caution—if at all—and only with a clear explanation that the benefit is unproven or uncertain. This is especially relevant in private settings, where financial costs and information asymmetry can be high.

For patients, the study supports a more transparent shared-decision-making approach: distinguishing between standard interventions, experimental interventions, and interventions with weak evidence. For research, the priority should be large, pre-registered, and methodologically sound trials, as this field continues to be affected by reliability issues in primary studies.

The Spanish Fertility Society (SEF) has published two monographs specifically aimed at assisted reproduction professionals. One is titled “ADD-ONS” Adjuvant Treatments in Andrology and Assisted Reproduction Laboratories, prepared by the Clinical Embryology Interest Group, and the other is “ADD-ONS” Adjuvant Treatments in Assisted Reproduction, prepared by the Reproductive Endocrinology Interest Group. These guidelines contain a series of evidence-based recommendations reviewed and drafted by leading professionals in our country, which can serve as a reference for the add-ons included in the article and others. We are fortunate to be a country recognized worldwide as a leader in assisted reproduction.

Does the press release accurately reflect the study?

“Broadly speaking, yes. The press release correctly conveys the study’s main message: for most of the add-ons used in assisted reproduction, there is no solid evidence that they improve the likelihood of a live birth, and much of the available literature has significant methodological limitations. It also accurately reflects that some procedures, such as endometrial scratching, PICSI or EmbryoGlue, show signs of a possible benefit, albeit with varying degrees of uncertainty and also depending heavily on the patients in whom these add-ons are used.”

Is the study of good quality? Are the conclusions supported by robust data?

“Yes. This is a methodologically robust study that systematically reviews 85 randomised clinical trials on ten different add-ons. The authors applied very strict criteria for quality and reliability, excluding approximately half of the potentially eligible studies due to methodological problems or doubts about their reliability. Furthermore, they used internationally recognised tools such as GRADE to assess the quality of the evidence. However, the strength of the conclusions depends on each intervention. For some add-ons, the evidence is moderate, whilst for others it remains low or very low due to the small size of the available studies or the risk of bias”.

How does this study fit in with the existing evidence?

“The results are, in general, consistent with previous systematic reviews and Cochrane reviews. The main novelty is that the authors apply stricter criteria to assess the reliability of clinical trials and update the evidence by incorporating new studies published in recent years. Despite excluding numerous studies previously included in other reviews, the overall conclusions remain largely unchanged, which reinforces the robustness of the findings. This study also reflects a growing trend in evidence-based medicine: it is not enough simply to accumulate studies; it is essential to critically assess their methodological quality and reliability.”

Have the authors taken confounding factors into account? Are there any important limitations to bear in mind?

“Yes. The authors carried out heterogeneity analyses, risk-of-bias assessments and sensitivity analyses. Furthermore, they took into account factors such as the patients’ age, the presence of repeated implantation failure, and the use of placebo controls or sham procedures where possible. Among the main limitations, it is worth noting that many of the available trials are small, are at risk of bias, or were conducted years ago using IVF protocols that no longer fully reflect current clinical practice. Furthermore, for some interventions, the evidence remains scarce or highly uncertain. The authors themselves acknowledge that some excluded studies may contain valid data, even though they did not meet current standards of reliability.”

What are the implications for real-world practice?

“The main implication is that patients and reproductive medicine professionals should exercise caution before incorporating costly or invasive complementary treatments that have not clearly been shown to improve the likelihood of having a live newborn at home. The results should not be interpreted as marking the end of research in this field. Some add-ons show promising signs and could be beneficial in certain patient groups, but larger and more rigorous clinical trials are needed to determine precisely who might actually benefit from these interventions.

As a reproductive specialist, I would add an important caveat: the absence of solid evidence does not always equate to evidence of a lack of benefit. In many cases, it simply means that we do not yet have sufficiently robust studies to provide a definitive answer to the question.”

This study provides a comprehensive and methodologically rigorous assessment of the available evidence on ten of the most commonly used complementary treatments (add-ons) in in vitro fertilisation. For most of these treatments, there is currently no robust evidence that they improve reproductive outcomes, such as the probability of pregnancy or the live birth rate, whilst for some of them the potential benefits remain uncertain and require further, higher-quality studies.

One of the main strengths of this research is the approach adopted by the authors. They did not merely combine the results of the available trials, but carried out a critical assessment of the quality and reliability of the included studies, analysing the risk of bias and the certainty of the evidence using recognised tools. Furthermore, the exclusion of studies with significant reliability issues allows for a more conservative and likely more realistic estimate of the effect of these treatments.

The results are consistent with the evidence accumulated in recent years, which has shown that many complementary IVF treatments have been incorporated into clinical practice before sufficient evidence was available to demonstrate their efficacy and safety. Although some may have a role in specific situations or in certain subgroups of patients, their widespread use is not supported by high-quality evidence.

As the authors themselves point out, the study also has some limitations, including the possibility that certain trials excluded on the basis of reliability criteria may have contained valid information. Furthermore, for some adjuncts, the scarcity of available studies makes it difficult to draw definitive conclusions.

From a clinical and patient-care perspective, this study has important implications. In a field such as assisted reproduction, where patients often face a heavy emotional and financial burden, it is essential that any additional treatment is offered alongside clear and balanced information on its potential benefits, risks and the degree of uncertainty involved. The introduction of new complementary treatments should be guided by sound scientific evidence and, where such evidence is insufficient, should preferably be considered within well-designed research protocols.

This article presents a rigorous analysis of published trials that examine the effectiveness of complementary procedures used by assisted reproductive technology clinics, highlighting that a significant proportion of the studies had to be excluded due to doubts about their scientific basis. Among the remaining studies, it was found that most of the procedures have low scientific evidence supporting their effectiveness and, therefore, there is considerable uncertainty regarding their application in IVF. This type of study is essential for highlighting the lack of criteria and scientific basis supporting the use of the procedures described in the article and marketed by assisted reproduction clinics.

Reproduction is a highly regulated biological process whose cellular and molecular mechanisms are finely coordinated to ensure an essential evolutionary goal: the perpetuation of the species. All of its stages—from gamete formation and selection to fertilization, early embryonic development, and embryo implantation—depend on complex regulatory networks that ensure their proper development. However, despite the fundamental importance of these processes, our understanding of the cellular and molecular bases that govern them remains limited, and many of the mechanisms involved have yet to be identified or characterized.

The involvement of the private sector as the main driver of technological innovation in assisted reproduction has accelerated the introduction of new procedures in in vitro fertilization techniques. However, the adoption of these innovations is not always supported by robust scientific evidence and, in many cases, precedes the establishment of adequate regulatory mechanisms that would allow for a rigorous evaluation of their efficacy and safety.

Although the study focuses on a specific population (Australia), it is nonetheless significant and can be applied globally due to the socioeconomic characteristics of this country compared to other developed nations.

This study highlights the need for countries, through their health and pharmaceutical regulatory and oversight agencies, to strengthen the mechanisms for evaluating, monitoring, and authorizing new technologies applied to assisted reproduction. Furthermore, in my view, substantial investment in basic research on the reproductive process will provide well-verified new scientific evidence that can subsequently be applied to improve procedures, thereby preventing economic interests from taking precedence over the development of assisted reproductive technologies—a development that would have a major impact on the future of the population.