Scolaris Content Display Scolaris Content Display

Intervenciones para la prevención del SHEO en los ciclos de TRA: una revisión global de revisiones Cochrane

Collapse all Expand all

Resumen

Antecedentes

El síndrome de hiperestimulación ovárica (SHEO) en los ciclos de tecnología de reproducción asistida (TRA) es una enfermedad inducida por el tratamiento que tiene una prevalencia del 20% al 33% en su forma leve y del 3% al 8% en su forma moderada o grave. Estos números podrían ser incluso mayores en las pacientes con alto riesgo como las que presentan ovarios poliquísticos o una alta producción de ovocitos a partir de la recolección de óvulos.

Objetivos

El objetivo de este resumen es identificar y resumir todas la evidencia de las revisiones sistemáticas Cochrane sobre las intervenciones para la prevención o el tratamiento del SHEO moderado, grave y general en parejas con subfertilidad sometidas a ciclos de TRA.

Métodos

Fueron elegibles para la inclusión en este resumen las revisiones sistemáticas Cochrane publicadas que informaron SHEO moderado, grave o general como un resultado en los ciclos de TRA. También se identificaron los protocolos Cochrane entregados y los títulos registrados para su futura inclusión en el resumen. La evidencia está actualizadas hasta el 12 diciembre 2016. Las revisiones, protocolos y títulos se identificaron mediante la búsqueda en la base de datos de revisiones sistemáticas del Grupo Cochrane de Ginecología y Fertilidad (Cochrane Gynaecology and Fertility Group) y en Archie (el sistema de gestión de la información de Cochrane) realizada en julio de 2016 sobre la efectividad de intervenciones para los resultados de SHEO moderado, grave y general. Se realizó por duplicado la selección de revisiones sistemáticas, la extracción de datos y la evaluación de la calidad. Para evaluar la calidad de las revisiones incluidas se utilizó la herramienta AMSTAR (que evalúa la calidad metodológica de las revisiones sistemáticas) y se utilizaron los métodos GRADE para evaluar la calidad de la evidencia para cada resultado. Se resumieron las características de las revisiones incluidas en el texto y en las tablas adicionales.

Resultados principales

En este resumen se incluyeron 27 revisiones. En general las revisiones fueron de alta calidad según las calificaciones AMSTAR y los estudios incluidos proporcionaron evidencia que varió de calidad muy baja a alta. Diez revisiones no han sido actualizadas en los últimos tres años. Siete revisiones describieron intervenciones que proporcionaron un efecto beneficioso en la reducción de las tasas de SHEO y una revisión adicional se categorizó como "alentadora". De las intervenciones efectivas, todas excepto una no mostraron efectos perjudiciales sobre los resultados del embarazo.

Evidencia de al menos calidad moderada indica que los médicos deben considerar las siguientes intervenciones en los ciclos de TRA para reducir las tasas de SHEO.

• Tratamiento con metformina antes y durante un ciclo de TRA en las pacientes con SOPQ (evidencia de calidad moderada).

• Protocolo de antagonista de la hormona liberadora de gonadotropinas (GnRH) en los ciclos de TRA (evidencia de calidad moderada).

• Agonista de GnRH (GnRHa) desencadenante en los programas de donación de ovocitos o "freeze all" (criotransferencia) (evidencia de calidad moderada).

Evidencia de calidad baja o muy baja indica que los médicos deben considerar las siguientes intervenciones en los ciclos de TRA para reducir las tasas de SHEO.

• Citrato de clomifeno para la estimulación ovárica controlada en los ciclos de TRA (evidencia de calidad baja).

• Cabergolina alrededor del momento de administración de la gonadotropina coriónica humana (hCG) o la recolección de ovocitos en los ciclos de TRA (evidencia de calidad baja).

• Líquidos intravenosos (expansores del plasma) alrededor del momento de la administración de la hCG o la recolección de ovocitos en los ciclos de TRA (evidencia de calidad muy baja).

• Progesterona para el apoyo de la fase luteínica en los ciclos de TRA (evidencia de calidad baja).

• "Coasting" (suprimir las gonadotropinas), una intervención alentadora que necesita estudios de investigación adicionales para la reducción del SHEO.

Según este resumen, se debe concluir que actualmente la evidencia no es suficiente para apoyar la práctica generalizada de la crioconservación de embriones.

Conclusiones de los autores

Actualmente se han realizado 27 revisiones en The Cochrane Library que informan o intentan informar sobre el SHEO en los ciclos de TRA. Se identificaron cuatro protocolos de revisión, pero no se identificaron títulos nuevos registrados que potencialmente se puedan incluir en este resumen en el futuro. Este resumen proporciona la evidencia más actualizadas sobre la prevención del SHEO en los ciclos de TRA de todas las revisiones Cochrane publicadas actualmente sobre las TRA. Los médicos pueden utilizar la evidencia de este resumen para elegir el mejor régimen de tratamiento para las pacientes individuales, un régimen que no sólo disminuya las probabilidades de desarrollar SHEO, sino que no comprometa otros resultados como la tasa de embarazo o de nacidos vivos. Sin embargo, los resultados de la revisión están limitados por la falta de estudios primarios recientes o revisiones actualizadas. Además, los elaboradores de políticas pueden utilizar este resumen para desarrollar protocolos o guías locales y regionales y puede mostrar las brechas en el conocimiento para los estudios de investigación futuros.

Resumen en términos sencillos

Intervenciones para la prevención del síndrome de hiperestimulación ovárica en los ciclos de fecundación in vitro: una revisión global de revisiones Cochrane

Pregunta de la revisión global

Este resumen de revisiones Cochrane tiene como objetivo identificar y resumir la evidencia de las revisiones sistemáticas Cochrane sobre las intervenciones que podrían prevenir o tratar el síndrome de hiperestimulación ovárica (SHEO) moderado, grave y general en parejas con subfertilidad a las que se les realizan ciclos de tecnologías de reproducción asistida (TRA) (es decir, fecundación in vitro [FIV] e inyección intracitoplasmática de espermatozoides [ICSI]).

Antecedentes

El SHEO en los ciclos de TRA es un evento adverso como consecuencia de la estimulación ovárica para la FIV. Es causado por una respuesta ovárica muy intensa a la medicación hormonal y da lugar a ovarios aumentados de tamaño y a un cambio en el paso de los líquidos de los vasos sanguíneos a la cavidad abdominal que provoca, por ejemplo, timpanismo abdominal, alto riesgo de coágulos dentro de los vasos sanguíneos (trombosis) y reducción en la irrigación de sangre a órganos importantes como los riñones y el hígado. La forma leve del SHEO se observa en casi el 20% al 33% de los ciclos, mientras que la forma moderada o grave se encuentra en aproximadamente el 3% al 8% de los ciclos y puede dar lugar a una carga de enfermedad grave o incluso a mortalidad si no se trata. Por lo tanto, es importante identificar los regímenes de tratamiento y las intervenciones que pueden reducir la incidencia de SHEO.

Características de los estudios

Se encontraron 27 revisiones Cochrane de TRA de alta calidad que fue posible incluir en este resumen. Estas revisiones intentaron informar sobre el SHEO en los ciclos de FIV o ICSI. No se incluyeron revisiones de inseminación intrauterina ni de inducción de la ovulación. La evidencia está actualizadas hasta el 12 diciembre 2016.

Resultados clave

De las 27 revisiones incluidas en este resumen, diez no se habían actualizado en los tres últimos años.

Siete revisiones describieron intervenciones que proporcionaron un efecto beneficioso en la reducción de las tasas de SHEO y una revisión adicional se categorizó como "alentadora". De las intervenciones efectivas, todas excepto una no mostraron efectos perjudiciales sobre los resultados del embarazo. Evidencia de al menos calidad moderada indican que los médicos deben considerar las siguientes intervenciones en los ciclos de TRA para reducir las tasas de SHEO.

• Tratamiento con metformina antes y durante un ciclo de TRA en las pacientes con SOPQ (evidencia de calidad moderada).

• Protocolo de antagonista de la hormona liberadora de gonadotropinas (GnRH) en los ciclos de TRA (evidencia de calidad moderada).

• Agonista de GnRH (GnRHa) desencadenante en los programas de donación de ovocitos o "freeze all" (criotransferencia) (evidencia de calidad moderada).

Evidencia de calidad baja o muy baja indican que los médicos deben considerar las siguientes intervenciones en los ciclos de TRA para reducir las tasas de SHEO.

• Citrato de clomifeno para la estimulación ovárica controlada en los ciclos de TRA (evidencia de calidad baja).

• Cabergolina alrededor del momento de administración de la gonadotropina coriónica humana (hCG) o la recolección de ovocitos en los ciclos de TRA (evidencia de calidad baja).

• Líquidos intravenosos (expansores del plasma sanguíneo) alrededor del momento de la administración de la hCG o de la recolección de los óvulos en los ciclos de TRA (evidencia de calidad muy baja).

• Progesterona para el apoyo de la fase luteínica en los ciclos de TRA (evidencia de calidad baja).

Una intervención alentadora que necesita estudios de investigación adicionales es el "coasting" (suprimir las gonadotropinas) para la reducción del SHEO. Según este resumen, se debe concluir que actualmente la evidencia no es suficiente para apoyar la práctica generalizada de congelar todos los embriones y recolocarlos posteriormente cuando ha desaparecido el SHEO.

Los médicos pueden utilizar la evidencia de este resumen para elegir el mejor régimen de tratamiento de las pacientes individuales, un régimen que no sólo disminuya las probabilidades de desarrollar SHEO, sino que no comprometa los resultados del embarazo. Sin embargo, los resultados de este resumen están limitados por la falta de estudios primarios recientes o revisiones actualizadas. Además, los elaboradores de políticas pueden utilizar este resumen para desarrollar protocolos o guías locales y regionales y puede mostrar las brechas en el conocimiento para los estudios de investigación futuros.

Authors' conclusions

This overview provides the most up‐to‐date evidence on prevention of OHSS in ART cycles from all currently published Cochrane reviews on ART. Clinicians can use the evidence summarised in this overview to choose the best treatment regimen for individual patients: a regimen that not only reduces the chance of developing OHSS but does not compromise other outcomes such as pregnancy or live birth rate. Furthermore, policymakers can use this overview when developing local and regional protocols or guidelines, and investigators can use it to identify knowledge gaps for future research.

Implications for practice

Evidence of at least moderate quality shows that clinicians should consider the following interventions to reduce OHSS rates in ART cycles.

  • Metformin treatment before and during an ART cycle for women with PCOS (moderate‐quality evidence) (Tso 2014).

  • Dopamine agonists around the time of hCG administration or oocyte pickup in ART cycles (moderate‐quality evidence) (Tang 2016).

  • GnRH antagonist protocol in ART cycles (moderate‐quality evidence) (Al‐Inany 2016).

  • GnRHa trigger in donor oocyte or 'freeze‐all' programmes, as it reduces OHSS and leads to lower pregnancy rates when embryo transfer is performed in the same cycle (moderate‐quality evidence) (Youssef 2014).

All of the above mentioned interventions are preventive measures used to reduce OHSS rates.

Evidence of low or very low quality indicates that clinicians can consider the following interventions to reduce OHSS rates in ART cycles.

  • Clomiphene citrate for controlled ovarian stimulation in ART cycles (low‐quality evidence) (Gibreel 2012).

  • Intravenous fluids (plasma expanders) around the time of hCG administration or oocyte pickup in ART cycles (very low‐quality evidence) (Youssef 2016b).

  • Progesterone for luteal phase support in ART cycles (low‐quality evidence) (van der Linden 2015).

Among the interventions mentioned above, clomiphene for ovarian stimulation and progesterone for luteal support are preventive measures used to reduce OHSS rates; dopamine agonists and plasma expanders are considered treatments for women with OHSS.

On the basis of this overview, we must conclude that evidence is currently insufficient to support the widespread practice of embryo cryopreservation and coasting (withholding of gonadotrophins) for reduction of OHSS (D'Angelo 2007; D'Angelo 2011).

Implications for research

This overview clearly identifies ways in which current evidence on effectiveness of interventions for prevention of OHSS is lacking. First, it highlights the need for review authors to update existing ART reviews to decrease knowledge gaps on this topic. Second, it should motivate clinicians and researchers to generate larger RCTs of higher quality to perform comparisons of new and existing interventions intended to reduce the incidence of OHSS.

The following three interventions have been shown to reduce OHSS on the basis of low‐quality or very low‐quality evidence. The fourth intervention listed here has shown a promising effect on reduction of OHSS and should be prioritised for examination by researchers in new high‐quality RCTs.

  • Clomiphene citrate for controlled ovarian stimulation in ART cycles (low‐quality evidence) (Gibreel 2012).

  • Intravenous fluids (plasma expanders) around the time of hCG administration or oocyte pickup in ART cycles (very low‐quality evidence) (Youssef 2016b).

  • Progesterone for luteal phase support in ART cycles (low‐quality evidence) (van der Linden 2015).

  • Coasting (withholding gonadotrophins) before hCG triggering in ART cycles (very low‐quality evidence) (D'Angelo 2011).

The uptake of subgroups 'moderate' and 'severe' OHSS in outcome reporting would be very useful for future studies and reviews, as these categories are the most clinically significant subgroups of OHSS. Reporting these subgroups separately provides clinicians and policymakers with a far more balanced reflection of treatment risks than is provided by mere use of the outcome 'total OHSS', which also includes mild OHSS and might not be as clinically important. Furthermore, clinicians would benefit if new RCTs would distinguish early and late types of OHSS, as the time of development of OHSS could influence the choice of therapy.

Large, well‐conducted RCTs are urgently needed to support the current evidence base for interventions described in this overview, including dose‐finding studies and research to determine the optimal timing of interventions. These trials should first examine interventions that seem to be effective but for which only very low‐quality, low‐quality or moderate‐quality evidence is available (Table 4).

Background

Description of the condition

Ovarian hyperstimulation syndrome (OHSS) is a serious complication of controlled ovarian hyperstimulation cycles used in assisted reproductive technologies (ART). OHSS is clinically characterised by abdominal tenderness and swelling due to increased ovarian volume along with a sudden increase in vascular permeability, which results in a shift of fluid to the extravascular space. However, the exact pathophysiology of OHSS has not been completely elucidated. Cases of spontaneous OHSS have been reported and are suspected to be linked to follicle‐stimulating hormone (FSH) receptor gene mutations (Delbaere 2004). However, the development of OHSS during ART cycles is mainly an iatrogenic side effect of the high doses of gonadotropin used for ovarian stimulation, resulting in multi‐follicular growth. A key role is suspected for vascular endothelial growth factor (VEGF), which is produced by multiple follicles following ovarian stimulation (Agrawal 1999). Higher VEGF levels induce hyperpermeability of ovarian blood vessels, which leads to a fluid shift from the intravascular to the third space. Also, the administration of human chorionic gonadotrophin (hCG) as an ovulation trigger or luteal phase support in high‐risk women with extensive luteinisation and supraphysiological levels of oestradiol and progesterone in the presence of multiple corpora lutea can trigger OHSS (Delbaere 2005). Moreover, the extra hCG‐rise accompanying (multiple) pregnancy after ART can aggravate already existing OHSS or induce late‐onset OHSS.

Over the years, several criteria have been used to classify OHSS severity (Appendix 1; Aboulghar 2003; Golan 1989; Navot 1992; Schenker 1978). In general, when OHSS progresses to a moderate stage, women experience abdominal pain and nausea and vomiting, and ascites can be seen around the ovaries on vaginal ultrasonography. If the condition progresses to severe OHSS, extravascular fluid can be found in pleural and pericardial spaces, and several haemodynamic changes take place, such as intravascular volume depletion, haemoconcentration, hypoalbuminaemia and electrolyte imbalances. These changes can lead to severe morbidity associated with thromboembolic events (Stewart 1997), respiratory distress and liver or renal failure. If left untreated, OHSS demonstrates rapid progression, with potentially life‐threatening or lethal complications (Braat 2006).

The mild form of OHSS is common, is of less clinical importance and occurs in an estimated 20% to 33% of ART cycles. The more clinically relevant moderate and severe forms of OHSS occur in an estimated 3% to 8% of ART cycles (3% to 6% moderate and 0.5% to 5% severe forms) (Delvigne 2002; Golan 1989; Schenker 1994). These large differences in reported OHSS incidence occur mainly because most reports involve single‐centre data, use different definitions of OHSS, do not require that diagnosis must be ascertained by a formal classification system or must have adequate follow‐up and lack reporting of mild or moderate forms. A large European report on 2010 ART practice (Kupka 2014) provided OHSS data for 25 participating countries and revealed prevalence of 0.3% in 349,402 simulated ART cycles. However, this report lacked data for some countries with a high volume of ART cycles (e.g. France, Sweden, the Netherlands, UK) and for other countries reported extremely low rates of OHSS, possibly as the result of reporting bias. A large Swiss retrospective cohort study reported a decline in OHSS incidence from 3.6% to 1% from 2005 to 2009 (De Geyter 2015). Globally, the incidence of OHSS is declining; a steady decrease has been reported since its peak incidence in the 1990s, when the main goal of ART was to produce a high number of oocytes (Kol 2011), and the incidence of severe OHSS was considered to be around 0.2% to 1% (Abramov 1999). With the emergence of new treatment regimens, more judicious use of gonadotrophins, increased cycle monitoring and improved knowledge of OHSS risks, the incidence of this disorder fell gradually over subsequent decades.

Although it is relatively rare, OHSS in ART cycles is an iatrogenic disease, and women who are affected should be monitored carefully to avoid life‐threatening complications. Early recognition of risk factors for OHSS can help clinicians tailor treatment regimens. Women with a priori risk for development of OHSS are those with polycystic ovaries (PCOs) (with or without PCO syndrome (PCOS)) or a high antral follicle count (e.g. at a young age). During a controlled ovarian stimulation cycle, women can acquire 'high risk' status when they prove to have high oestradiol levels, excessive growth of follicles or a large number of retrieved oocytes. Besides the early OHSS type that develops during, or immediately after, ovarian stimulation, we can distinguish a late type, which appears after embryo implantation has been established. The presence of a multiple gestation can trigger or exacerbate this late type of OHSS (Delbaere 2005; Mathur 2000).

Description of interventions and how the interventions might work

Interventions that aim to reduce OHSS incidence can target diverse portions of stimulated ART cycles.

  • Selection or identification of 'high risk' populations for tailoring of stimulation regimens.

  • Prevention of recurrent OHSS by adjustment of the dose of gonadotrophins in the next cycle.

  • Prevention of large numbers of follicles by tailored ovarian stimulation for specific risk groups (e.g. use of different treatment regimens, use of adjuvant medication).

  • Prevention of a rise in VEGF levels (e.g. by prevention of development of large numbers of follicles, by targeting of VEGF receptors (e.g. by dopamine agonists)).

  • Dose reduction or withholding of hCG administration for ovulation trigger or luteal support.

  • Prevention of a rise in oestradiol by withholding of gonadotrophins (‘coasting’).

  • Prevention of a further rise in oestradiol and of ovulation triggering and pregnancy by cycle cancellation.

  • Prevention of intravascular volume depletion by administration of plasma‐expanding intravenous (IV) fluids.

  • Prevention of pregnancy by freezing of all embryos and transfer back during a subsequent cycle.

Moreover, trials of interventions within an ART cycle that are not specifically aimed at preventing OHSS may report on OHSS as an outcome. These interventions are of interest to this overview and might reveal new mechanisms for lowering risk of OHSS.

Why it is important to do this overview

OHSS is an iatrogenic disease with an estimated incidence of 3% to 6% in ART cycles of the clinically relevant moderate or severe form (Delvigne 2002). If left untreated, OHSS can lead to severe morbidity and can be life‐threatening. Multiple treatment options are available for prevention of OHSS in ART cycles; therefore, it is important to provide consumers, health professionals, policymakers and guideline developers with a summary of evidence on OHSS prevention obtained from the Cochrane Library. We will comment upon this evidence in light of the overall effectiveness of studied interventions in the separate reviews. By doing so, we will identify existing knowledge gaps or reporting flaws within the Cochrane systematic reviews published in the Cochrane Library on the topic of OHSS in ART cycles. This means that we can provide clear suggestions for future research.

Objectives

The objective of this overview is to identify and summarise all evidence from Cochrane systematic reviews on interventions for prevention or treatment of moderate, severe and overall OHSS in couples with subfertility who are undergoing ART cycles.

Methods

Criteria for considering reviews for inclusion

Types of reviews

For this overview of reviews, we included all published Cochrane systematic reviews of randomised controlled trials (RCTs) that examined:

  • interventions that aimed to prevent OHSS with reporting on the incidence of moderate, severe or overall OHSS as a primary outcome; and

  • other interventions in ART cycles with reporting on the incidence of moderate, severe or overall OHSS as a secondary outcome.

Moreover, we listed the protocols of reviews and title registrations on OHSS prevention in a table included in the overview. Thus we will be able to identify and add new reviews, once published, at the time of the next overview update. We excluded reviews on non‐ART cycles and reviews on ART cycles that did not report on OHSS as an outcome.

Types of participants

We included reviews that enrolled women who underwent fresh ART cycles, including those who acted as oocyte donors. We considered Cochrane systematic reviews that reported on ‘high risk’ subgroups (e.g. minimum number of follicles, minimum number of oocytes retrieved, minimum oestradiol level, women with PCOS) and those that reported on unselected populations. We excluded reviews of non‐ART cycles, such as ovulation induction or intrauterine insemination cycles.

Types of interventions

We considered for inclusion reviews on two types of interventions.

  • Interventions specifically aimed at prevention of OHSS for which OHSS was reported as a primary outcome.

  • Any interventions in ART cycles for which OHSS was reported as a secondary outcome.

Search methods for identification of reviews

We searched for reviews within the Cochrane Database of Systematic Reviews and Archie (the Cochrane information management system) for the following keywords: in vitro fertilisation (IVF), intracytoplasmic sperm injection (ICSI), ART, adverse events and OHSS (search dates 18/11/2015, 24/7/2016 and 12/12/2016). The overview 'Assisted reproductive technology: an overview of Cochrane Reviews' by Farquhar 2015 identified all current reviews on ART that reported on OHSS as a primary or secondary outcome, and we used this as complementary guidance.

Data collection and analysis

We based the methods used for data collection and analysis for this overview on Chapter 22 of the Cochrane Handbook for Systematic Reviews of Interventions (Becker 2011; Higgins 2011).

Primary outcomes

The primary outcome measure is the incidence of moderate, severe and overall OHSS per woman randomised.

The OHSS subgroups of moderate and severe are defined by the criteria set forth by Aboulghar 2003, Golan 1989, Navot 1992, Rabau 1967, Rizk 1999 and Schenker 1978, or by any other classification used in the included reviews (Appendix 1).

Secondary outcomes

Secondary outcomes studied were live birth rate, clinical pregnancy rate, miscarriage rate, multiple pregnancy rate and any reported adverse effects that derived from the interventions studied (as reported by separate reviews, e.g. side effects of medication, admission to the hospital).

Selection of reviews

Two overview authors independently selected reviews for inclusion according to the criteria stated. A third overview author acted as a referee and discussed disagreements that arose. We added the following to the overview for future overview updates: protocols of reviews and title registrations on prevention of OHSS submitted to the Cochrane Library, and reviews on ART interventions that will report on OHSS as a secondary outcome.

Data extraction and management

Two overview authors (SM and JB) performed data extraction using a Microsoft Excel spreadsheet. If data from the reviews were unclear or seemed to be missing, we contacted review authors for clarification, searched primary RCTs or contacted primary study authors for details. A third overview author (CF) acted as a referee and discussed discrepancies or disputes that arose.

We extracted and summarised the following data for the additional tables.

  • Population demographics: participant characteristics, definition of high‐risk groups when applicable.

  • Review characteristics: number of included trials, number of participants, date the review was assessed as up‐to‐date (date of search), interventions and comparisons, all primary and secondary outcomes and limitations of the review.

  • Timing of intervention: e.g. pretreatment selection of participants, pretreatment adjuvant therapy, stimulation phase, stimulation phase adjuvant treatment, ovulation trigger, embryo transfer phase, luteal support phase.

  • Statistical summary: summary effects from relevant comparisons on our primary outcome of moderate, severe or overall OHSS.

We used the same summary effect measures as were used in the original reviews, in most cases odds ratios.

Assessment of methodological quality of included reviews

Two overview authors independently assessed the quality of the evidence derived from included systematic reviews. We resolved discrepancies by discussion, and a third overview author acted as an arbiter.

Quality of evidence from primary studies in included reviews

Using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) method, we summarised the quality of the evidence from primary studies in the included reviews (Guyatt 2008; Schünemann 2013). We prepared 'Summary of findings' tables using GRADEpro Guideline Development Tool (GDT) software (GRADEpro GDT) for overview outcomes for each comparison by taking ratings from the original review, or we appraised the review ourselves if the review had not yet been assessed through the GRADE approach.

  • Risk of bias of included trials.

  • Directness of the evidence.

  • Precision of the evidence.

  • Heterogeneity.

  • Risk of publication bias.

We summarised the evidence for each of the selected clinical outcomes in a 'Summary of findings' table, to which we added the summary risk estimate and 95% confidence intervals. We allocated the quality of evidence for the clinical outcome with a score for strength of the evidence, ranging from 'high' to 'very low'.

Quality of included reviews

We assessed the methodological quality of included reviews using the AMSTAR (Assessing the Methodological Quality of Systematic Reviews) instrument (Shea 2007). This instrument evaluates methods used in systematic reviews and the degree to which reviews are biased by comparing them on the basis of distinct criteria. Ratings used in AMSTAR include 'yes' (clearly done), 'no' (clearly not done), 'cannot answer' and 'not applicable' (Appendix 2).

Data synthesis

We undertook a narrative description of the included trials. We included an ‘Overview of reviews’ table, which shows the characteristics of included reviews. Moreover, we displayed a summary of the quality of evidence within individual reviews that was based on GRADE judgements, and we provided an AMSTAR rating for each included review.

We summarised the main results of the included systematic reviews and the effect on OHSS rates of their individual comparisons using the following framework.

  • Effective interventions: indicates that the review found evidence of effectiveness for an intervention.

  • Promising interventions (more evidence needed): indicates that the review found some evidence of effectiveness for an intervention, but more evidence is needed.

  • Ineffective interventions: indicates that the review found evidence of lack of effectiveness for an intervention.

  • Probably ineffective interventions (more evidence needed): indicates that the review found evidence suggesting lack of effectiveness for an intervention, but more evidence is needed.

  • No conclusions possible due to lack of evidence: indicates that the review found insufficient evidence for review authors to comment on the effectiveness of an intervention.

The choice of category to be allocated reflects the conclusions stated by authors of the individual reviews and our judgement as overview authors. We resolved disagreements by discussion.

We based our approach to summarising the evidence on the framework used for the ART overview (Farquhar 2015).

Results

Upon screening the ART overview (Farquhar 2015), we identified a total of 20 reviews reporting on OHSS as an outcome. We subsequently screened full texts for remaining reviews in the ART overview reporting on OHSS or adverse events that could include OHSS as an outcome. By doing this, we identified an additional seven reviews (see flow diagram of included reviews, Figure 1). We excluded 33 reviews from the ART overview for not reporting on OHSS. From the Cochrane Gynaecology and Fertility (CGF) Database of registered titles, we identified no titles that were expected to report on OHSS in ART cycles as an outcome. From the CGF database of submitted protocols, we identified four protocols that potentially would report on OHSS in ART cycles as an outcome (Appendix 3). Most often, we excluded titles and protocols because they did not concern ART cycles or because they concerned laboratory interventions.


Flow diagram of included reviews.

Flow diagram of included reviews.

Description of included reviews

In total, we included 27 Cochrane systematic reviews that reported on OHSS (85,497 participants). See Table 1 for a summary of the characteristics of these reviews (review title and author, numbers of randomised controlled trials and participants included, interventions and comparisons, outcomes, main limitations of each review). Of the 27 included reviews, two were empty reviews (reviews with no included studies), with last search dates in 2009 and 2011, respectively (Siristatidis 2009; Yossry 2006). We deemed two reviews to be stable, meaning that searches would be repeated only when review authors became aware of newly published evidence (D'Angelo 2007; Yossry 2006).

Open in table viewer
Table 1. Review characteristics

Review ID

Number of included trials

Population

Definition of high risk for OHSS (where applicable)

Intervention

Comparison intervention/control

Primary outcomesa

Review limitations

ADA563

D'Angelo 2011

Coasting (withholding gonadotrophins) for preventing ovarian hyperstimulation syndrome

4 RCTs

340 women with PCOS

downregulated by GnRHa, undergoing superovulation in IVF or ICSI cycles

High risk: women with PCOS

Coasting when

oestradiol levels were > 2500 pg/mL or > 9000 pmol/L

Early unilateral follicular

aspiration

No coasting or other interventions

OHSSa

Live birtha

Clinical pregnancy

Number of oocytes retrieved

Multiple pregnancy

Miscarriage

Comparisons based on limited trial data

Live birth reported in only 1 trial

Trials lacked blinding, and half the trials lacked details on allocation concealment and incomplete outcome assessment

ADA561

D'Angelo 2007

Embryo freezing for preventing ovarian hyperstimulation syndrome

2 RCTs

151 women

downregulated by GnRHa, undergoing superovulation in IVF or ICSI cycles.

High risk: as defined by included studies

Cryopreservation

Fresh embryo transfer

Intravenous albumin

OHSSa

Clinical pregnancya

Live birth

Admissions

Evidence based on 2 trials, 1 for each comparison

Live birth reported in only 1 trial

Issues around methodological quality of both trials

TH1338

Tang 2016

Dopamine agonists for preventing ovarian hyperstimulation syndrome

16 RCTs

2091 women at high risk of developing OHSS undergoing ART

High risk: as defined by included studies

Cabergoline quinagolide, bromocriptine,

cabergoline + albumin,

cabergoline + HES

Placebo/no treatment/other treatment:

Albumin alone

HES

Coasting

Prednisolone

OHSSa

Live birtha

Clinical pregnancy

Adverse effects

Miscarriage

Multiple pregnancy

Allocation concealment and blinding not adequately reported.

One study used a co‐intervention of albumin IV and 1 of HES

Different regimens of cabergoline administration between included studies

Live birth rate reported in only 2 studies

Incomplete reporting of multiple pregnancy rate, adverse effects and miscarriage rate

PMA481

Youssef 2016b

Volume expanders for prevention of OHSS

9 RCTs

1660 (albumin) + 487 (HES) women

at high risk of developing OHSS undergoing ART cycles High risk: determined as number of follicles or oestradiol levels on day of hCG, as defined by included studies

Human albumin

Hydroxyethyl starch (HES)

Placebo/no treatment

OHSSa

Clinical pregnancy

Number of oocytes retrieved

Multiple pregnancy

Miscarriage

Live birth

No reporting of live birth rate

Limited by incomplete data reporting and lack of (details on) blinding

HA413

Youssef 2016

Recombinant vs urinary hCG for final oocyte maturation triggering in IVF and ICSI cycles

18 RCTs

2952 women undergoing ART

Recombinant hCG

Recombinant LH

Urinary hCG

OHSSa

Clinical pregnancy

Miscarriage

Oocytes retrieved

Tolerance

Live birth

Review authors combined ongoing pregnancy and live births together

Only 7 trials reported on live birth

Trials lacked details on allocation concealment, randomisation and blinding

MM1690

Youssef 2014

GnRHa vs hCG for oocyte triggering in antagonist‐assisted reproductive technology

17 RCTs

1847 women undergoing ART

GnRH agonist

hCG

OHSSa

Live birth ratea

Ongoing pregnancy

Clinical pregnancy

Multiple pregnancy

Miscarriage rate

Risk of bias in included studies. Limitations included premature termination, failure to clearly report methods and substantial heterogeneity

Adverse events such as multiple pregnancy rate were not well reported

AWP1710

Pouwer 2015

Long‐acting FSH vs daily FSH for women undergoing assisted reproduction

6 RCTs

3753 women with subfertility

Long‐acting FSH

Daily FSH

OHSSa

Live birth ratea

Ongoing pregnancy rate

Clinical pregnancy rate

Multiple pregnancy rate

Miscarriage rate

Adverse events

Satisfaction

Limited by risk of attrition bias in some primary studies and by serious imprecision

LDT120

Tso 2014

Metformin treatment before and during IVF or ICSI in women with PCOS

9 RCTs

816 women with PCOS

Metformin

Placebo

No treatment

OHSSa

Live birtha

Clinical pregnancya Miscarriage

Adverse events Number of oocytes

retrieved

Total dose FSH (IU) Number of days

gonadotrophin treatment

Cycle cancellation rate

Serum E2 level (nmol/L)

Half the trials were not blinded and lacked details on allocation concealment and randomisation

AM1335

Gibreel 2012

Clomiphene citrate in combination with gonadotropins for controlled ovarian stimulation in women undergoing IVF

14 RCTs,

12 for meta‐analysis

2536 (12 trials)

Subfertile women undergoing ART

Clomiphene citrate

± additional treatments

Alternative treatments for

COH

OHSSa

Live birth ratea

Miscarriage rate

Ectopic pregnancy

Foetal abnormality

Ongoing pregnancy rate

Cancellation rate

Live birth reported in only 5 trials

Most studies suffered from suboptimal methods and information on some outcomes was insufficient

TA1860

Allersma 2013

Natural cycle IVF for subfertile couples

5 RCTs

382 subfertile women and couples undertaking IVF treatment

Natural cycle IVF

Modified natural cycle IVF

COH IVF

OHSSa

Live birtha

Pregnancy

Ongoing pregnancy

Number of oocytes retrieved

Time to live birth

Number of cycles required to conceive

Cumulative pregnancy/live birth rate

Multiple pregnancy

Lack of embryos for cryopreservation

Cycle cancellation

Gestational abnormalities

Cancellation of treatment

Cost‐effectiveness

Few studies, live birth reported in only 1 very small trial

Inclusion criteria differed

MV263

van der Linden 2015

Luteal phase support for ART cycles

94 RCTs

26,198 women with any cause of subfertility undergoing ART

Progesterone

hCG

Placebo or no treatment

hCG

Progesterone + oestrogen Progesterone + GnRHa

Live birtha

Clinical pregnancy

Ongoing pregnancy

Miscarriage

OHSS

Multiple pregnancy

Poor reporting of study methods and imprecision due to small sample sizes

HA412

Al‐Inany 2016

Gonadotrophin‐releasing hormone antagonists for ART

73 RCTs

12,212 women undergoing ART

GnRH antagonist

Long‐course GnRHa

OHSSa

Live birtha

Ongoing pregnancy

Clinical pregnancy

Miscarriage

Cycle cancellation

Only 12 trials reported live birth

Trial methods limited by lack of blinding

Poor reporting of study methods for OHSS

AMY731

Yossry 2006

IVF vs tubal re‐anastomosis (sterilisation reversal) for subfertility after tubal sterilisation

No RCTs

NA

IVF

Tubal re‐anastomosis

Live birtha

Clinical pregnancy Multiple pregnancy

Other serious maternal morbidity, (incl OHSS)

Empty review with no

trials

No longer being updated

ZP672

Pandian 2015

IVF for unexplained subfertility

6 RCTs

733 couples with unexplained subfertility

IVF

Expectant management

Intrauterine insemination Intrauterine insemination +

ovarian stimulation

Clomiphene citrate

Live birtha

OHSS

Clinical pregnancy Multiple pregnancy

Some evidence was based

on a single trial

Limitations included imprecision and heterogeneity for some outcomes

LA541

Albuquerque 2013

Depot vs daily administration of GnRHa protocols for pituitary desensitisation in assisted reproduction cycles

16 RCTs,

12 for meta‐analysis

1811 couples with any cause of subfertility undergoing IVF with COH with hFSH, hMG or rFSH

Pituitary downregulation with depot administration of GnRHa

Daily administration of GnRHa

OHSSa

Live birtha

Clinical pregnancya

Miscarriage

Multiple pregnancy

S tudy quality unclear due to poor reporting. O nly four stu dies reported live birth an d only five described adequate methods for allocation concealment .

IOK973

van Wely 2011

Recombinant vs urinary gonadotrophin for ovarian stimulation in ART cycles

42 RCTs

9606 normogonadotrophic women undergoing fresh and/or frozen thawed IVF or ICSI cycles

Recombinant FSH

Urinary FSH

OHSSa

Live birtha

Clinical pregnancy

Miscarriage

Multiple pregnancy

Adverse effects

No difference reported in moderate/severe OHSS

WPM1780

Martins 2013

FSH replaced by low‐dose hCG in the late follicular phase vs continued FSH for ART

5 RCTs

351 women undergoing COH for ART

Low‐dose hCG instead of FSH in late follicular phase

Continued FSH in late follicular phase

OHSSa

Live birtha

Clinical pregnancy

Miscarriage

Small studies and low event rate

Total OHSS incidence reported

DHH752

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pretreatment for ovarian stimulation protocols for women undergoing ART

23 RCTs

2596 women of any age with subfertility regardless of cause, undergoing ART

Pretreatment with combined oral contraceptive pills

Pretreatment with progestogens

No pretreatment

Placebo

Progestogens

Oestrogens

Live birtha

OHSS

Clinical pregnancy

Miscarriage

Multiple pregnancy

Adverse effects

Only 3/23 studies reported on OHSS

2 of these 3 studies did not define how they diagnosed the condition

IOK972

Kwan 2014

Monitoring of stimulated cycles in assisted reproduction (IVF and ICSI)

6 RCTs

781 women undergoing COH in an IVF/ICSI cycle

Transvaginal ultrasonography + Oestradiol measurement

Transvaginal ultrasonography

Live birtha

OHSS

Clinical pregnancy

Miscarriage

Multiple pregnancy

Adverse effects

Only total OHSS reported, including mild OHSS

CMB1261

Boomsma 2012

Peri‐implantation glucocorticoid administration for ART cycles

14 RCTs

1879 subfertile patients undergoing IVF/ICSI, regardless of cause of infertility

Glucocorticoids in the peri‐implantation phase

No glucocorticoids in the peri‐implantation phase

Live birtha

Multiple pregnancya

OHSS

Clinical pregnancy Miscarriage

Adverse effects

Only 2 studies, pooled total OHSS

VJP951

Siristatidis 2016

Aspirin for IVF

13 RCTs

2653 women undergoing IVF/ICSI and their partners

Aspirin

No treatment

Placebo

Live birtha

OHSS

Clinical pregnancy Miscarriage

Multiple pregnancy

Adverse effects

Only 1 of 13 studies reported on OHSS and without exact numbers or explanation for numerators/denominators

CS1400

Siristatidis 2009

In vitro maturation in subfertile women with PCOS undergoing assisted reproduction

None

0 women with PCOS and subfertility

In vitro maturation + IVF/ICSI in women with PCOS

Conventional IVF/ICSI in women with PCOS

Live birtha

OHSS

Effectiveness

Clinical pregnancy Miscarriage

Adverse effects

Empty review

IRS911

Cheong 2013

Acupuncture and ART

20 RCTs

4544 women undergoing ART, any type of acupuncture at any time point before, after or during ART, intended to improve ART outcome

Acupuncture of men, women or both during COH

Acupuncture + ART

Acupuncture alone

No treatment

Placebo

Sham acupuncture

Acupuncture + ART

Live birtha

OHSS

Clinical pregnancy Miscarriage

Multiple pregnancy

Adverse effects

No trials reported on OHSS

MHM931

Mochtar 2007

Recombinant luteinising hormone (rLH) for COH in assisted reproductive cycles

14 RCTs

2612 subfertile ovulatory women undergoing IVF or ICSI

High risk: NA

Combination of rLH and rFSH for COH in IVF/ICSI followed by ET in GnRHa and GnRH antagonist protocols

rFSH alone for COH in IVF/ICSI followed by ET in GnRHa and GnRH antagonist protocols

OHSSa

Live birtha

Clinical pregnancy Miscarriage

Only 4/14 trials reported on OHSS

Pooled OHSS

No GRADE assessment in old version

KH291

Duffy 2010

Growth hormone for IVF

10 RCTs

440 women part of a subfertile couple undergoing IVF

Adjuvant growth hormone during conventional IVF

Conventional IVF

Live birtha

OHSS

Clinical pregnancy

Adverse effects

Only 4 of 10 RCTs reported on adverse events (which could include OHSS)

1 study actually mentioned OHSS (however, no cases); pooled OHSS

SD265

Siristatidis 2015

GnRHa protocols for pituitary suppression in assisted reproduction

37 RCTs

3872 women/couples with all types of infertility undergoing ART and using GnRHa for pituitary downregulation

Long protocol

Long luteal protocol

Short protocol

Dose continued

Dose continued after hCG administration

Pretreatment 2 weeks

Short protocol

Ultrashort protocol

Long follicular phase protocol

Ultrashort protocol

Dose stopped

Dose reduced

Dose discontinued after hCG administration

Pretreatment 3 weeks

Live birtha

OHSS

Clinical pregnancy Adverse effects

Only 2 of 37 included RCTs reported on OHSS for 2 of 9 compared regimens

JC1630

Showell 2013

Antioxidants for female subfertility

28 RCTs

3548 subfertile women referred to fertility clinic who might or might not undergo ART (IVF, ICSI or IUI)

Adjuvant antioxidants in females

No treatment

Placebo

Another antioxidant

Live birtha

Clinical pregnancy Miscarriage

Multiple pregnancy

Adverse effects (incl OHSS)

Only 3 studies reported: 1 no data and 2 no cases

aPrimary review outcome.

ART: artifical reproductive technology.

COH: controlled ovarian hyperstimulation.

ET: embryo transfer.

FSH: follicle‐stimulating hormone.

GnRHa: gonadotrophin‐releasing hormone agonist.

hCG: human chorionic gonadotrophin.

HES: hydroxyethyl starch.

hFSH: human follicle‐stimulating hormone.

hMG: human menopausal gonadotrophin.

ICSI: intracytoplasmic sperm injection.

IUI: intrauterine insemination.

IVF: in vitro fertilisation.

LH: luteinising hormone.

NA: not applicable.

OHSS: ovarian hyperstimulation syndrome.

PCOS: polycystic ovary syndrome.

RCT: randomised controlled trial.

rFSH: recombinant follicle‐stimulating hormone.

rLH: recombinant luteinising hormone.

Pandian 2015 (IVF for unexplained subfertility) compared IVF versus IUI (we did not formally consider IUI a treatment for inclusion in this overview). As all studies also included a IVF/ICSI comparator group, we decided that we should include this review in the overview. Also, Cheong 2013 ('Acupuncture and assisted reproductive technology') could theoretically include studies on IUI or ovulation induction; however, all current comparisons in this review involve acupuncture around the time of oocyte retrieval and/or embryo transfer, which means that the current version of the review reports only on IVF/ICSI treatments.

Reporting on OHSS

A total of 15 reviews reported on OHSS as a primary outcome, and 12 reported on OHSS as a secondary outcome. The number of included primary studies per review ranged from zero to 94. We also noted large variation in the number of included primary studies that actually reported data on OHSS, which ranged from zero to 32 studies. For example, in the review that included 94 primary studies, only one study actually reported on OHSS (van der Linden 2015).

Four reviews focused specifically on prevention of OHSS and compared the following interventions: coasting versus no/other treatment (D'Angelo 2011), embryo freezing versus fresh transfer or intravenous albumin plus fresh transfer (D'Angelo 2007), volume expanders versus placebo or no treatment (Youssef 2016b) and dopamine agonists versus placebo or no/other treatment (Tang 2016). These reviews included studies that identified high‐risk groups on the basis of oestradiol levels, a minimum number of follicles of a certain size, a minimum number of retrieved oocytes or a diagnosis of PCOS. Some primary studies excluded extremely high risk groups on the basis of oestradiol levels.

Three reviews (D'Angelo 2007; D'Angelo 2011; Tang 2016) reported separately on the subgroups 'moderate' and 'severe' OHSS, and two reported only on 'severe OHSS' (Al‐Inany 2016; Youssef 2016). The other 22 reviews were described as reporting 'total OHSS' with or without defining this as inclusion of mild, moderate or severe cases.

Timing of intervention

Timing of interventions in the included reviews differed (see Table 1) as follows.

We could not classify the Yossry 2006 and Pandian 2015 reviews according to this framework because they studied IVF versus other strategies.

Main limitations of the reviews

The major and most frequent limitations of included reviews were the mere reporting of 'total OHSS', as opposed to reporting separately on the more clinically relevant subgroups 'moderate' and 'severe'; failure to include any or inclusion of only a few studies per comparison; and a generally low proportion of primary studies reporting data on OHSS.

The 12 reviews that did report on OHSS as a secondary outcome often described lack of statistical power for the outcome 'OHSS' due to the low incidence of the condition in general and more specifically in populations not selected for risk of developing OHSS. For example, given a population size of 2000 women undergoing ART, as well as a 5% margin of error and a 95% confidence interval, the required sample size would be 323 women. In light of the fact that the incidence of moderate to severe OHSS in this population would be set at 5% (range from literature 3% to 8%), at least 6460 women should be included in the study for enough women to develop OHSS that data would show differences in OHSS rates. For most countries and settings, this inclusion number is not realistically attainable for any study.

Last search date of the reviews

Table 2 shows the last search date per review. Only 17 of the 27 included reviews conducted a literature search within the past three years (to 12 December 2016), and overview authors deemed an additional two reviews (D'Angelo 2007; Yossry 2006) with an older literature search to be stable. At our third search date (12 December 2016), we became aware of four reviews that were in the process of being updated (D'Angelo 2011; Duffy 2010; Gibreel 2012; Mochtar 2007). Progress of these updates at the date of the search varied widely, from just starting the literature search to completing the final editorial phase.

Open in table viewer
Table 2. Last search date assessment

Review no.

First review author

Review title

Date last assessed
up to date

< 3 years since last assessed up to date or deemed stable

ADA561

D'Angelo 2007

Embryo freezing for preventing OHSS

26/11/2010

Stable

ADA 563

D'Angelo 2011

Coasting (withholding of gonadotrophins) for preventing OHSS

19/07/2010

X

TH1338

Tang 2016

Dopamine agonists for preventing OHSS

15/08/2016

PMA481

Youssef 2016b

Volume expanders for prevention of OHSS

21/09/2016

HA413

Youssef 2016

Recombinant vs urinary hCG for final oocyte maturation triggering in IVF and ICSI cycles

23/04/2015

MM1690

Youssef 2014

GnRHa vs hCG for oocyte triggering in antagonist‐assisted reproductive technology

08/09/2014

X

LDT1201

Tso 2014

Metformin treatment before and during IVF or ICSI in women with PCOS

15/10/2014

AWP1710

Pouwer 2015

Long‐acting FSH vs daily FSH for women undergoing assisted reproduction

8/06/2015

AM1335

Gibreel 2012

Clomiphene citrate in combination with gonadotrophins for controlled ovarian stimulation in women undergoing IVF

23/03/2012

X

TA1860

Allersma 2013

Natural cycle IVF for subfertile couples

5/03/2013

MV263

van der Linden2015

Luteal phase support for ART cycles

25/11/2014

HA412

Al‐Inany 2016

Gonadotrophin‐releasing hormone antagonists for ART

28/04/2016

AMY731

Yossry 2006

IVF vs tubal re‐anastomosis (sterilisation reversal) for subfertility after tubal sterilisation

15/05/2009

Empty, stable

ZP672

Pandian 2015

IVF for unexplained subfertility

4/05/2015

LA541

Albuquerque 2013

Depot vs daily administration of GnRHa protocols for pituitary desensitisation in assisted reproduction
cycles

3/07/2012

IOK973

van Wely 2011

Recombinant vs urinary gonadotrophin for ovarian stimulation in ART cycles

20/10/2010

X

WPM1780

Martins 2013

FSH replaced by low‐dose hCG in late follicular phase vs continued FSH for ART

5/02/2013

DHH752

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pretreatment for ovarian stimulation protocols for women undergoing ART

16/11/2008

X

IOK972

Kwan 2014

Monitoring of stimulated cycles in assisted reproduction (IVF and ICSI)

30/05/2014

CMB1261

Boomsma 2012

Peri‐implantation glucocorticoid administration for ART cycles

20/09/2011

X

VJP951

Siristatidis 2016

Aspirin for IVF

9/05/2016

CS1400

Siristatidis 2009

In vitro maturation in subfertile women with PCOS undergoing assisted reproduction

17/02/2011

Empty

IRS911

Cheong 2013

Acupuncture and ART

22/07/2013

MHM931

Mochtar 2007

Recombinant luteinising hormone (rLH) for COH in assisted reproductive cycles

25/01/2007

X

KH291

Duffy 2010

Growth hormone for IVF

20/07/2009

X

SD265

Siristatidis 2015

GnRHa protocols for pituitary suppression in assisted reproduction

23/04/2015

JC1630

Showell 2013

Antioxidants for female subfertility

15/04/2014

ART: artifical reproductive technology.

COH: controlled ovarian hyperstimulation.

FSH: follicle‐stimulating hormone.

GnRHa: gonadotrophin‐releasing hormone agonist.

hCG: human chorionic gonadotrophin.

ICSI: intracytoplasmic sperm injection.

IUI: intrauterine insemination.

IVF: in vitro fertilisation.

OHSS: ovarian hyperstimulation syndrome.

PCOS: polycystic ovary syndrome.

rLH: recombinant luteinising hormone.

✔ under 3 years since last assessed as up to date

X over 3 years since last assessed as up to date

Statistical summary

Quality of evidence from primary studies in included reviews

The quality of the evidence reported by primary studies in the included reviews assessed by the GRADE approach ranged from very low to high for individual comparisons. The main reasons for downgrading of reviews for quality included inadequate reporting of allocation concealment and randomisation methods, lack of blinding and imprecision. Eleven of the 27 reviews included fewer than 10 primary studies.

Methodological quality of included reviews

Quality of systematic reviews

We rated the quality of the included reviews using the AMSTAR tool (Shea 2007) and listed the domains per review in Table 3.

Open in table viewer
Table 3. AMSTAR assessment per review

Review no.

First review author + year

Review title

AMSTAR criteria

Prespecified question and inclusion criteria

Duplicate study selection and data extraction

Comprehensive literature search

Grey literature included

Lists included and excluded studies

Describes characteristics of included studies

Study quality assessed

Studies combined using appropriate methods

Likelihood of publication bias considered/tested

Potential for conflict of interest addressed

ADA561

D'Angelo 2007

Embryo freezing for preventing ovarian hyperstimulation syndrome

ADA 563

D'Angelo 2011

Coasting (withholding gonadotrophins) for preventing ovarian hyperstimulation syndrome

TH1338

Tang 2016

Dopamine agonists for preventing ovarian hyperstimulation syndrome

PMA481

Youssef 2016b

Volume expanders for the prevention of ovarian hyperstimulation syndrome

HA413

Youssef 2016

Recombinant versus urinary human chorionic gonadotrophin for final oocyte maturation triggering in IVF and ICSI cycles

MM1690

Youssef 2014

Gonadotropin‐releasing hormone agonist versus hCG for oocyte triggering in antagonist‐assisted reproductive technology

LDT1201

Tso 2014

Metformin treatment before and during IVF or ICSI in women with polycystic ovary syndrome

AWP1710

Pouwer 2015

Long‐acting FSH versus daily FSH for women undergoing assisted reproduction

AM1335

Gibreel 2012

Clomiphene citrate in combination with gonadotrophins for controlled ovarian stimulation in women undergoing in vitro fertilisation

TA1860

Allersma 2013

Natural cycle IVF for subfertile couples

MV263

van der Linden 2015

Luteal phase support for ART cycles

HA412

Al‐Inany 2016

Gonadotrophin‐releasing hormone antagonists for assisted reproductive technology

AMY731

Yossry 2006

In vitro fertilisation versus tubal re‐anastomosis (sterilisation reversal) for subfertility after tubal sterilisation

NA

NA

NA

NA

ZP672

Pandian 2015

In vitro fertilisation for unexplained subfertility

LA541

Albuquerque 2013

Depot versus daily administration of gonadotrophin‐releasing hormone
agonist protocols for pituitary desensitisation in assisted reproduction
cycles

IOK973

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

WPM1780

Martins 2013

FSH replaced by low‐dose hCG in the late follicular phase versus continued FSH for assisted reproductive techniques

DHH752

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pretreatment for ovarian stimulation protocols for women undergoing assisted reproductive techniques

IOK972

Kwan 2014

Monitoring of stimulated cycles in assisted reproduction (IVF and ICSI)

CMB1261

Boomsma 2012

Peri‐implantation glucocorticoid administration for assisted reproductive technology cycles

VJP951

Siristatidis 2016

Aspirin for in vitro fertilisation

CS1400

Siristatidis 2009

In vitro maturation in subfertile women with polycystic ovarian syndrome undergoing assisted reproduction

NA

NA

NA

NA

IRS911

Cheong 2013

Acupuncture and assisted reproductive technology

MHM931

Mochtar 2007

Recombinant luteinising hormone (rLH) for controlled ovarian hyperstimulation in assisted reproductive cycles

X

KH291

Duffy 2010

Growth hormone for in vitro fertilisation

X

SD265

Siristatidis 2015

Gonadotrophin‐releasing hormone agonist protocols for pituitary suppression in assisted reproduction

JC1630

Showell 2013

Antioxidants for female subfertility

Search date: 24/07/2016.

ART: artifical reproductive technology.

FSH: follicle‐stimulating hormone.

hCG: human chorionic gonadotrophin.

ICSI: intracytoplasmic sperm injection.

IUI: intrauterine insemination.

IVF: in vitro fertilisation.

NA: not applicable.

rLH: recombinant luteinising hormone.

  • All reviews had prespecified their clinical question and inclusion criteria.

  • All reviews conducted study selection and data extraction in duplicate.

  • All reviews conducted a comprehensive literature search.

  • All reviews included searches of grey literature.

  • All reviews listed included and excluded studies.

  • All reviews described the characteristics of included studies.

  • All reviews assessed study quality.

  • All reviews combined studies using appropriate methods.

  • A total of 25/27 reviews addressed the risk of reporting bias by using a statistical test when appropriate.

  • All reviews addressed the potential for conflict of interest.

Effect of interventions

We categorised all included intervention reviews by effectiveness for reduction of OHSS and by effectiveness for the primary pregnancy outcome stated in the review. In total, with regard to reduction of OHSS rates, seven reviews showed a beneficial effect of the intervention on the incidence of OHSS, one was promising,13 were ineffective and six remained inconclusive. Of the effective interventions, one intervention did reduce OHSS rates but had a detrimental effect on pregnancy outcomes (Youssef 2014).

We listed effects of the interventions on the incidence of OHSS in the 'Summary of findings' table (Table 4). Most reviews did not report the incidence of OHSS as the (sole) primary outcome. This implies that the effectiveness of studied interventions can very well be different for the main primary outcome and for reduction of OHSS. Among the 25 non‐empty reviews, the effect of the intervention on OHSS rates was beneficial in eight reviews, and the intervention had no effect on OHSS rates in 14 reviews. One review reported that the control group had lower OHSS rates than the intervention group; however, the only primary study in this review reporting on OHSS did not provide exact numbers, so we could not calculate the effect size (Siristatidis 2016). For three reviews, we could not calculate effect size because they included insufficient primary studies reporting on OHSS (Cheong 2013; Duffy 2010; Showell 2013).

Open in table viewer
Table 4. Summary of findings for OHSS: per review and/or per intervention

Review title and

comparison intervention/control

Assumed risk with comparator

Corresponding risk

with intervention

Relative effect

(95% CI)

Number of participants

(studies)

Quality of the evidence

(GRADE)

Comments

D'Angelo 2007

Embryo freezing for preventing ovarian hyperstimulation syndrome

(Embryo freezing vs fresh transfer)

Overall OHSS: 60 per 1000

Overall OHSS: 125 per 1000

(62 to 240)

OR 1.12

(0.01 to 2.29)

125

(1 study)

Low

Imprecision, number of events < 300

Evidence based on a single open‐label study with insufficient methodological details provided

D'Angelo 2007

Embryo freezing for preventing ovarian hyperstimulation syndrome

(Embryo freezing vs intravenous albumin)

Moderate or

severe OHSS:

77 per 1000

Moderate or

severe OHSS:

308 per 1000

(41 to 824)

OR 5.33

(0.51 to 56.24)

26

(1 study)

Very low

Imprecision, number of events < 300

Evidence based on a single open‐label trial

D'Angelo 2011

Coasting (withholding gonadotrophins) for preventing ovarian hyperstimulation syndrome

Moderate or

severe OHSS:

265 per 1000

Moderate or

severe OHSS:

58 per 1000

(11 to 241)

OR 0.17

(0.03 to 0.88)

68

(1 study)

Very low

Imprecision, number of events < 300

Evidence based on a single conference abstract
Insufficient methodological details provided

Tang 2016

Dopamine agonists for preventing ovarian hyperstimulation syndrome

Moderate or severe OHSS: 286 per 1000

Moderate or severe OHSS: 97 per 1000

(71 to 135)

OR 0.27

(0.19 to 0.39)

2091

(16 studies))

Moderate

Imprecision, number of events < 300

Lack of details for allocation concealment and blinding, selective reporting

Youssef 2016b

Volume expanders for the prevention of ovarian hyperstimulation syndrome

(human albumin vs placebo/no treatment)

Moderate or severe OHSS:

122 per 1000

Moderate or severe OHSS: 85 per 1000 (61 to 177)

OR 0,67

(0.47 to 0.95)

1452

(7 studies)

Very low

Imprecision, number of events < 300

Lack of details on allocation concealment and selective reporting

Youssef 2016b

Volume expanders for the prevention of ovarian hyperstimulation syndrome

(HES vs placebo)

Moderate or

severe OHSS: 164 per 1000

Moderate or

severe OHSS: 50 per 1000

(23 to 104)

OR 0.27

(0.12 to 0.59

272

(2 studies)

Very low

Imprecision, number of events < 300

Lack of details on allocation concealment and selective reporting

Youssef 2016b

Volume expanders for the prevention of ovarian hyperstimulation syndrome

(mannitol vs placebo)

Moderate or

severe OHSS: 517 per 1000

Moderate or

severe OHSS: 289 per 1000

(191 to 407)

OR 0.38

(0.22 to 0.64)

226

(1 study)

Low

Imprecision, number of events < 300

Lack of details on allocation concealment and selective reporting

Youssef 2016

Recombinant versus urinary human chorionic gonadotrophin for final oocyte maturation triggering in IVF and ICSI cycles (r‐hCG vs u‐hCG)

Overall OHSS: 27 per 1000

Overall OHSS:40 per 1000

(15 to 102)

OR 0.39

(0.25 to 0.61)

374

(3 studies)

Moderate

Imprecision, number of events < 300

One of the trials lacked methodological details on randomisation, allocation concealment and blinding

Youssef 2016

Recombinant versus urinary human chorionic gonadotrophin for final oocyte maturation triggering in IVF and ICSI cycles (r‐LH vs u‐hCG)

Overall OHSS: 10 per 1000

Overall OHSS:

17 per 1000

(11 to 84)

OR 1.76

(0.37 to 8.45)

417

(3 studies)

Low

Imprecision, number of events < 300

One of the trials lacked adequate methodological details

Youssef 2014

Gonadotropin‐releasing hormone agonist versus hCG for oocyte triggering in antagonist‐assisted reproductive technology

Overall OHSS: 5 per 1000

Overall OHSS:

1 per 1000

(0 to 2)

OR 0.15

(0.05 to 0.47)

989

(9 studies)

Moderate

Imprecision, number of events < 300

All studies at high risk of bias in 1 or more domains

None clearly reported blinded outcome assessment

Tso 2014

Metformin treatment before and during IVF or ICSI in women with polycystic ovary syndrome.

Overall OHSS: 270 per 1000

Overall OHSS: 97 per 1000

(62 to 153)

OR 0.29

(0.18 to 0.49)

798

(8 studies)

Moderate

Imprecision, number of events < 300

Pouwer 2015

Long‐acting FSH versus daily FSH for women undergoing assisted reproduction

(low dose)

Overall OHSS: 47 per 1000

Overall OHSS: 57 per 1000

(26 to 125)

RR 1.22

(0.56 to 2.66)

645

(3 studies)

Moderate

Imprecision, number of events < 300

Pouwer 2015

Long‐acting FSH versus daily FSH for women undergoing assisted reproduction

(medium dose)

Overall OHSS: 63 per 1000

Overall OHSS: 60 per 1000 (45 to 85)

RR 0.96

(0.68 to 1.35)

3075

(5 studies)

Low

Imprecision, number of events < 300

Confidence intervals compatible with clinically meaningful benefit in either arm or with no effect, plus high risk of attrition bias in 2 studies

Pouwer 2015

Long‐acting FSH versus daily FSH for women undergoing assisted reproduction

(high dose)

Overall OHSS: 0 per 1000

Overall OHSS: 0 per 1000

(0 to 0)

RR 1.73

(0.09 to 32.75)

33

(1 study)

Very low

Imprecision, number of events < 300

High risk of attrition bias

Gibreel 2012

Clomiphene citrate in combination with gonadotropins for controlled ovarian stimulation in women undergoing in vitro fertilisation

(clomiphene + gonadotropins vs gonadotropins)

Overall OHSS: 50 per 1000

Overall OHSS:12 per 1000

(5 to 27)

OR 0.23

(0.1 to 0.52)

1559

(5 studies)

Low

Imprecision, number of events < 300

Very wide 95% confidence interval crossing the threshold points of appreciable benefit or harm, which is 25%

Allersma 2013

Natural cycle IVF for subfertile couples

(natural cycle vs conventional IVF)

Overall OHSS: 67 per 1000

Overall OHSS:

13 per 1000

(1 to 393)

OR 0.10

(0.01 to 4.06)

60

(1 study)

Very low

Imprecision, number of events < 300

Only 1 study reporting on OHSS

Allocation concealment method not reported

van der Linden 2015

Luteal phase support for ART cycles

(hCG versus placebo/no treatment)

Overall OHSS: 41 per 1000

Overall OHSS:

155 per 100

(76 to 292)

OR 4.28

(1.191 to 9.6)

387

(1 study)

Low

Imprecision, number of events < 300

Poor reporting of study methods

van der Linden 2015

Luteal phase support for ART cycles

(progesterone vs hCG regimens)

Overall OHSS: 126 per 1000

Overall OHSS: 72 per 1000

(31 to 162)

OR 0.54

(0.22 to 1.34)

615

(4 studies)

Low

Imprecision, number of events < 300

Poor reporting of study methods

van der Linden 2015

Luteal phase support for ART cycles

(progesterone + GnRH agonist)

Overall OHSS: 50 per 1000

Overall OHSS: 50 per 1000

(17 to 137)

OR 1.00

(0.33 to 3.01)

300

(1 study)

Very low

Imprecision, number of events < 300

Poor reporting of study methods

van der Linden 2015

Luteal phase support for ART cycles

(progesterone vs progesterone + oestrogens)

Overall OHSS: 39 per 1000

Overall OHSS: 22 per 1000

(8 to 62)

OR 0.56

(0.2 to 1.63)

461

(2 studies)

Low

Imprecision, number of events < 300

Poor reporting of study methods

Al‐Inany 2016

Gonadotrophin‐releasing hormone antagonists for assisted reproductive technology

(GnRH antagonist vs GnRH agonist)

Overall OHSS: 114 per 1000

Overall OHSS: 73 per 1000

(62 to 85)

OR 0.61
(0.51 to 0.72)

7944

(36 studies)

Moderate

Methodological limitations including poor allocation concealment and lack of blinding

Yossry 2006

In vitro fertilisation versus tubal reanastomosis (sterilisation reversal) for subfertility after tubal sterilisation

(IVF vs tubal reanastomosis)

NA

NA

NA

NA

NA

Empty review

Pandian 2015

In vitro fertilisation for unexplained subfertility

(IVF vs IUI + gonadotropins/clomiphene citrate)

Overall OHSS: 58 per 1000

Overall OHSS: 66 per 1000

(26 to 158)

OR 1.15 (0.43 to 3.06)

324

(2 studies)

Low

Imprecision, number of events < 300

Only 2 studies on OHSS reported

Albuquerque 2013

Depot versus daily administration of gonadotrophin releasing hormone agonist protocols for pituitary desensitization in assisted reproduction cycles

(depot vs daily gonadotropins)

Overall OHSS: 3 per 100

Overall OHSS: 2 per 100
(1 to 6)

OR 0.84
(0.29 to 2.42)

570

(5 studies)

Low

Most studies were classified as at unclear risk of bias for all domains

Imprecision, number of events < 300
Studies were insufficient to assess publication bias

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

(rFSH vs HMG/HMG‐HP)

Overall OHSS: 17 per 1000

Overall OHSS: 17 per 1000

(10 to 28)

OR 1.00

(0.58 to 1.71)

3197

(11 studies)

High

Imprecision, number of events < 300

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

(rFSH vs FSH‐P)

Overall OHSS: 28 per 1000

Overall OHSS:49 per 1000

(25 to 95)

OR 1.79

(0.89 to 3.62)

1490

(6 studies)

Higha

Imprecision, number of events < 300

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

(rFSH vs FSH‐HP)

Overall OHSS: 28 per 1000

Overall OHSS:

31 per 1000

(20 to 48)

OR 1.11

(0.70 vs 1.75)

3053

(14 studies)

Higha

Two additional trials excluded in sensitivity analyses because it was unclear if data were reported according to ITT analysis (those were included for "Overall OHSS")

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

(rec‐hCG vs u‐hCG)

Overall OHSS: 19 per 1000

Overall OHSS:

22 per 1000

(16 to 30)

OR 1.18 (0.86 vs 1.61)

7740

(32 studies)

Higha

Imprecision number of events < 300

Martins 2013

FSH replaced by low‐dose hCG in the late follicular phase versus continued FSH for assisted reproductive techniques

(low‐dose hCG vs FSH in late follicular phase)

Overall OHSS: 3 per 100

Overall OHSS: 1 per 100

(0 to 4)

OR 0.30 (0.06 to 1.59)

351

(5 studies)

Very low

Imprecision, number of events < 300

Inconsistency, high risk of bias

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pre‐ treatment for ovarian stimulation protocols for women undergoing assisted reproductive techniques

(OAC plus antagonist vs antagonist)

Overall OHSS: 17 per 1000

Overall OHSS: 25 per 1000

(5 to 133)

OR 1.5 (0.26 to 8.8)

234

(1 study)

Very low

Single study reporting on OHSS

Imprecision, number of events < 300

Wide confidence intervals that cross line of no effect

High risk of attrition bias

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pre‐ treatment for ovarian stimulation protocols for women undergoing assisted reproductive techniques

(OAC plus antagonist vs agonist)

Overall OHSS: 55 per 1000

Overall OHSS: 35 per 1000

(12 to 100)

OR 0.63

(0.21 to 1.92)

290 (2 studies)

Very low

Imprecision, number of events < 300

One study at high risk of attrition bias

Kwan 2014

Monitoring of stimulated cycles in assisted reproduction

(IVF and ICSI)

(transvaginal ultrasound + estradiol vs transvaginal ultrasound)

Overall OHSS: 36 per 1000

Overall OHSS: 36 per 1000

(18 to 75)

OR 1.03

(0.48 to 2.20)

781

(6 studies)

Low

Imprecision, number of events < 300 with wide confidence intervals

Methods of randomisation inadequately described in 3 of 6 trials, allocation concealment inadequately described in all 6 trials and blinding inadequately described in 5 of 6 trials

No definition of OHSS provided by authors of these 6 studies

Boomsma 2012

Peri‐implantation glucocorticoid administration for assisted reproductive technology cycles

(adjuvant glucocorticoids vs no glucocorticoids)

Overall OHSS: 194 per 1000

Overall OHSS: 159 per 1000

(64 to 392)

OR 0.82 (0.33 to 2.02)

151

(2 studies)

Low

Imprecision, number of events < 300

Siristatidis 2016

Aspirin for in vitro fertilisation

(aspirin vs no treatment/placebo)

NA

NA

NA

NA

NA

Only 1 study reported on OHSS; no exact numbers or explanation of numerators/denominators given

Siristatidis 2009

In vitro maturation in subfertile women with polycystic ovarian syndrome undergoing assisted reproduction

(IVM vs conventional IVF)

NA

NA

NA

NA

NA

Empty review

Cheong 2013

Acupuncture and assisted reproductive technology

(acupuncture vs no acupuncture/sham acupuncture)

NA

NA

NA

NA

NA

No studies reported on OHSS

Mochtar 2007

Recombinant luteinizing hormone (rLH) for controlled ovarian hyperstimulation in assisted reproductive cycles

(combined rLH + FSH vs FSH )

Overall OHSS: 20 per 1000

Overall OHSS:

27 per 1000

(12 to 59)

OR 1.34

(0.58 to 3.09)

986

(7 studies)

Low

Imprecision, number of events < 300

Some methodological details unclear

Duffy 2010

Growth hormone for in vitro fertilization

(growth hormone vs no treatment/placebo)

NA

NA

NA

NA

NA

Only 1 study reported on OHSS; however, no cases of OHSS were reported

Siristatidis 2015

Gonadotropin‐releasing hormone agonist protocols for pituitary suppression in assisted reproduction

(different protocols vs other protocol)

Overall OHSS: 20 per 1000

Overall OHSS: 27 per 1000

(12 to 59)

OR 1.34

(0.58 to 3.09)

986

(7 studies)

Low

Imprecision, number of events < 300

Some methodological details unclear

Showell 2013

Antioxidants for female subfertility

(antioxidants vs no treatment/placebo/other antioxidant)

NA

NA

NA

NA

NA

Although 3 studies reported on OHSS, no numbers were given, so effect size could not be calculated

aReview authors GRADED these outcomes as 'high quality'; however, the total event rate < 300 would justify downgrading for this to moderate‐quality evidence.

ART: artifical reproductive technology.

FSH: follicle‐stimulating hormone.

FSH‐HP: highly purified FSH.

hCG: human chorionic gonadotrophin.

HES: hydroxyethyl starch.

ICSI: intracytoplasmic sperm injection.

IUI: intrauterine insemination.

IVF: in vitro fertilisation.

IVM: in vitro maturation.

NA: not applicable.

OAC: oral anticoagulant.

OHSS: ovarian hyperstimulation syndrome.

OR: odds ratio.

rFSH: recombinant follicle‐stimulating hormone.

r‐hCG: recombinant human chorionic gonadotrophin.

rLH: recombinant luteinising hormone.

Total: any grade of OHSS.

u‐hCG: urinary human chorionic gonadotrophin.

We summarise here the effectiveness of interventions for both reduction of OHSS and pregnancy outcomes, ranked by timing of the intervention within an ART cycle.

Moreover, for interventions that had a beneficial effect on OHSS rate, Figure 2 shows the extent of this effect in relation to the effect on live birth rate (when reported) or clinical pregnancy.


Extent of effect of interventions on OHSS rate and live birth rate (when reported) or clinical pregnancy: OR and 95% CI.

Extent of effect of interventions on OHSS rate and live birth rate (when reported) or clinical pregnancy: OR and 95% CI.

Effective interventions for reduction of OHSS with no impact on nor improvement in pregnancy outcomes

Pretreatment adjuvant therapy

  • Metformin treatment before and during IVF or ICSI in women with PCOS: No conclusive evidence suggests that metformin treatment before or during ART cycles improved live birth rates (low‐quality evidence). However, use of this insulin‐sensitising agent increased clinical pregnancy rates and decreased the risk of OHSS (moderate‐quality evidence) (Tso 2014). Evidence showed a beneficial effect of the intervention on OHSS rates.

Pituitary downregulation phase

  • Gonadotrophin‐releasing hormone (GnRH) antagonists for ART: Use of antagonists compared with long GnRHa protocols was associated with a large reduction in OHSS, and no evidence suggested a difference in live birth rates (moderate‐quality evidence) (Al‐Inany 2016). Evidence showed a beneficial effect of the intervention on OHSS rates.

Stimulation phase

  • Clomiphene citrate for controlled ovarian stimulation in women undergoing IVF: This review suggested that regimens with clomiphene could be used in controlled ovarian stimulation for IVF treatment without a reduction in pregnancy rates. However, further evidence is required before these regimens can be recommended with confidence as alternatives to gonadotrophins alone in GnRH long or short protocols (low‐quality evidence) (Gibreel 2012). Evidence showed a beneficial effect of the intervention on OHSS rates.

Ovulation trigger phase

  • Dopamine agonists for preventing OHSS: Dopamine agonists appeared to reduce the risk of OHSS in high‐risk women, especially for moderate OHSS. Use of dopamine agonists did not appear to affect clinical pregnancy rates or miscarriage rates, nor did they increase the risk of other adverse events (moderate‐quality evidence) (Tang 2016). Evidence showed a beneficial effect of the intervention on moderate or severe OHSS rates.

  • Volume expanders for prevention of OHSS: The volume expanders hydroxyethyl starch and mannitol decreased the incidence of moderate or severe OHSS without affecting pregnancy rates (very low‐quality evidence) (Youssef 2016b). Evidence showed a beneficial effect of the intervention on moderate or severe OHSS rates.

Luteal support phase

  • Luteal support phase in ART cycles: This review concluded that progesterone appears to provide the best method of providing luteal phase support, as it is associated with higher rates of live birth or ongoing pregnancy than placebo, and lower rates of OHSS than hCG. Addition of one or more doses of GnRH agonists to progesterone was associated with higher live birth and ongoing pregnancy rates than progesterone alone. Overall, addition of other substances such as oestrogen or hCG did not seem to improve outcomes, and hCG was associated with higher risk of OHSS. The route of progesterone administration did not seem to matter (quality of evidence was low for most comparisons) (van der Linden 2015). Evidence showed a beneficial effect of the intervention on OHSS rates for the comparison hCG versus placebo/no treatment. For the other comparisons, no evidence showed an effect on OHSS rates.

Effective interventions for reduction of OHSS with negative impact on pregnancy outcomes

Ovulation trigger phase

  • GnRHa versus hCG for oocyte triggering in antagonist ART cycles: Evidence suggested a lower live birth rate, a reduced ongoing pregnancy rate and a higher miscarriage rate among women who received a GnRHa. However, OHSS rates were reduced with GnRHa triggering; therefore, clinicians should consider the tradeoff between benefits and harms (moderate‐quality evidence) (Youssef 2014). Evidence showed a beneficial effect of the intervention on OHSS rates.

  • Volume expanders for prevention of OHSS: Evidence suggested that human albumin decreased the incidence of moderate or severe OHSS. However, contrary to the (very low‐quality) evidence found with hydroxyethyl starch (HES) and mannitol, human albumin appeared to have a detrimental effect on pregnancy rates (moderate‐quality evidence) (Youssef 2016b).

Promising interventions for reduction of OHSS with no impact on or improvement in pregnancy outcomes (more evidence needed)

Ovulation trigger phase

  • Coasting (withholding gonadotrophins) for preventing OHSS: Evidence was insufficient to show benefit derived from coasting done to prevent OHSS compared with no coasting or other interventions (very low‐quality evidence) (D'Angelo 2011). Evidence showed a beneficial effect of the intervention on OHSS rates, but this was reported only in a single abstract on an RCT that provided insufficient methodological details.

Ineffective interventions for reduction of OHSS with no impact on or improvement in pregnancy outcomes

Pretreatment adjuvant therapy

  • Oral contraceptive pill (OCP), progestogen or oestrogen pretreatment for ovarian stimulation protocols for women undergoing ARTs: Evidence suggested improved pregnancy outcomes with progestogen pretreatment and poorer pregnancy outcomes with combined OCP pretreatment (Smulders 2010). No evidence showed an effect of the intervention on OHSS rates.

Pituitary downregulation phase

  • GnRHa protocols for pituitary suppression in ART cycles: The pregnancy rate was higher when GnRHa was used in a long protocol as compared with a short or ultra‐short protocol (low‐quality evidence) (Siristatidis 2015). No evidence showed an effect of the intervention on OHSS rates.

  • Depot versus daily administration of GnRHa protocols for pituitary desensitisation in assisted reproduction cycles: No evidence suggested a significant difference in live birth or pregnancy outcomes between depot and daily GnRHa use for pituitary downregulation in IVF cycles using the long protocol, but substantial differences could not be ruled out (moderate‐quality evidence) (Albuquerque 2013). No evidence showed an effect of the intervention on OHSS rates.

Stimulation phase

  • FSH replaced by low‐dose hCG in the late follicular phase versus FSH alone for ARTs: Review authors were very uncertain about effects on live birth, OHSS and miscarriage, but evidence suggested that this intervention did not reduce the chances of ongoing and clinical pregnancy, and that it was likely to result in retrieval of an equivalent number of oocytes with less FSH expended (very low‐quality evidence) (Martins 2013). No evidence showed an effect of the intervention on OHSS rate.

  • Recombinant versus urinary gonadotrophin for ovarian stimulation in ART cycles: It appeared that all available gonadotrophins were equally effective and safe. The choice of one or the other product would depend upon the availability of the product, the convenience of its use and associated costs. Any specific differences are likely to be too small to justify further research (high‐quality evidence) (van Wely 2011). No evidence showed an effect of the intervention on OHSS rates.

  • Long‐acting FSH versus daily FSH for women undergoing assisted reproduction: A medium dose (150 to 180 μg) of long‐acting FSH appeared to offer a safe treatment option that was as effective as daily FSH in women with unexplained subfertility. Evidence showed a reduced live birth rate among women receiving a low dose (60 to 120 μg) of long‐acting FSH compared with daily FSH (moderate‐quality evidence) (Pouwer 2015). No evidence showed an effect of the intervention on OHSS rates.

  • Natural cycle IVF for subfertile couples: No evidence showed a significant difference between natural cycle and standard IVF for outcomes including live birth, OHSS, clinical pregnancy and multiple pregnancy (very low‐quality evidence) (Allersma 2013). No evidence showed an effect of the intervention on OHSS rates.

  • Monitoring of stimulated cycles in assisted reproduction (IVF and ICSI): RCTs provided no evidence to support cycle monitoring by ultrasonography plus serum oestradiol as more efficacious than cycle monitoring by ultrasonography only for the outcomes of live birth and pregnancy. A large well‐designed RCT is needed (low‐quality evidence) (Kwan 2014). No evidence showed an effect of the intervention on OHSS rates.

Ovulation trigger phase

  • Recombinant versus urinary hCG for final oocyte maturation triggering in IVF and ICSI cycles: Review authors concluded that urinary hCG remains the best choice for final oocyte maturation triggering in IVF and ICSI treatment cycles owing to availability and cost and no difference in live birth rates (moderate‐quality evidence) (Youssef 2016). No evidence showed an effect of the intervention on OHSS rates.

Embryo transfer phase

  • Peri‐implantation glucocorticoid administration for ART cycles: Overall, no clear evidence suggests that administration of peri‐implantation glucocorticoids in ART cycles significantly improved clinical outcomes (low‐quality evidence) (Boomsma 2012). No evidence showed an effect of the intervention on OHSS rates.

  • Embryo freezing for prevention of OHSS: Evidence was insufficient to show benefit for routine cryopreservation and the relative merits of intravenous albumin versus cryopreservation (low‐quality evidence) (D'Angelo 2007). No evidence showed an effect of the intervention on OHSS rates.

Luteal support phase

  • Recombinant luteinising hormone (rLH) for controlled ovarian hyperstimulation in assisted reproductive cycles: No evidence suggested that coadministration of rLH and recombinant follicle‐stimulating hormone (rFSH) in GnRHa‐downregulated women resulted in more live births than were reported with controlled ovarian hyperstimulation (COH) with rFSH alone. Nevertheless, all pooled pregnancy estimates, although not significantly different, pointed towards a beneficial effect of cotreatment with rLH, in particular with respect to pregnancy loss (low‐quality evidence) (Mochtar 2007). No evidence showed an effect of the intervention on OHSS rates.

ART versus other interventions

  • In vitro fertilisation for unexplained subfertility: IVF may be more effective than IUI plus ovarian stimulation (low‐quality evidence) (Pandian 2015). No evidence showed an effect of the intervention on OHSS rates.

Possibly ineffective interventions for reduction of OHSS with no impact on or improvement in pregnancy outcomes (more evidence needed)

  • None were reported.

No conclusions possible on effectiveness for reduction of OHSS (lack of evidence)

For six reviews, review authors could provide no conclusions on the effects of interventions on OHSS rates.

Pretreatment adjuvant therapy

  • Aspirin for IVF: Evidence from adequately powered RCTs was insufficient for review authors to reach a conclusion (Siristatidis 2016). No evidence showed an effect of the intervention on OHSS rates. No numbers were provided, so we could not calculate effect size.

  • Antioxidants for female subfertility: Antioxidants were not associated with increased live birth or clinical pregnancy rates, although more evidence is needed (low‐quality evidence) (Showell 2013). No trials reported actual numbers of cases of OHSS.

  • Acupuncture and ART: No evidence suggested benefit for acupuncture in improving live birth or clinical pregnancy rates in assisted conception (low‐quality evidence) (Cheong 2013). No trials reported on OHSS.

Stimulation phase

  • Growth hormone for IVF: We could not calculate the effect of the intervention on OHSS rates. Use of growth hormone in poor responders was associated with significant improvement in live birth rates (moderate‐quality evidence) (Duffy 2010).

  • In vitro maturation in subfertile women with polycystic ovarian syndrome (PCOS) undergoing assisted reproduction: This is an empty review (Siristatidis 2009).

ART versus other interventions

  • IVF versus tubal reanastomosis (sterilisation reversal) for subfertility after tubal sterilisation: This is an empty review (Yossry 2006).

Discussion

Summary of main results

We found seven interventions that were effective in reducing the occurrence of ovarian hyperstimulation syndrome (OHSS) while not influencing or even improving pregnancy outcomes: metformin pretreatment in women with polycystic ovary syndrome (PCOS), use of a gonadotrophin‐releasing hormone (GnRH) antagonist protocol for pituitary suppression, use of clomiphene citrate for controlled ovarian stimulation, use of dopamine agonists and the volume expanders hydroxyethyl starch (HES) and mannitol around the time of oocyte triggering, coasting before oocyte triggering and use of progesterone for luteal phase support (Al‐Inany 2016; D'Angelo 2011; Gibreel 2012; Tang 2016; Tso 2014; van der Linden 2015; Youssef 2016b).

Two additional interventions ‐ gonadotrophin‐releasing agonist (GnRHa) versus human chorionic gonadotrophin (hCG) for oocyte triggering, and use of the volume expander human albumin around the time of ovulation triggering ‐ proved effective in reducing OHSS but negatively impacted pregnancy outcomes in autologous cycles (Youssef 2014; Youssef 2016b).

Concerning the GnRHa trigger, this detrimental effect on pregnancy outcomes was not found in oocyte donation cycles. This would make the use of GnRHa for ovulation triggering suitable for oocyte donation programmes, as it would largely eradicate the chance of OHSS in the donor, without negatively influencing pregnancy outcomes in the recipient. GnRHa could also be useful in preventing OHSS in "freeze‐all" programmes (i.e. embryo transfer is not performed in the fresh autologous cycle) ‐ a regimen that is currently the topic of numerous research projects and new randomised controlled trials (RCTs).

Overall completeness and applicability of evidence

This overview of Cochrane reviews is complete and up‐to‐date as of December 2016. Only four reviews specifically addressed prevention of OHSS; by reporting all reviews on assisted reproduction technology (ART) that include OHSS as a primary or secondary outcome, we aimed to provide the most up‐to‐date overview of strategies to prevent OHSS currently included within the Cochrane Library. In keeping with the nature of a Cochrane overview, this body of work does not cover non‐Cochrane reviews on OHSS. Moreover, alternative or emerging strategies for prevention of OHSS may not yet have been covered in a Cochrane review and therefore cannot be found in this overview, for example, use of calcium gluconate infusion or the dual GnRHa and hCG trigger. Once such strategies have been assessed in new reviews, we can and will update this overview accordingly.

Clinically relevant moderate OHSS and severe OHSS are still rare conditions; moreover, inclusion of 'high risk' groups was not based on a variety of criteria and was not a prerequisite at all for many of the included reviews. In combination with the fact that most studies included OHSS as a secondary outcome, this means that most primary studies lacked statistical power to report on OHSS. However, as OHSS is an undesirable outcome in ART, it is unlikely that the prevalence will change, and it is considered unethical by many institutional ethics review boards to randomise very high‐risk groups as a control in current and future studies. Unfortunately, this means that it will be difficult to nearly impossible for researchers to perform well‐powered studies that address these shortcomings.

A major limitation of many of the included reviews is reporting of 'total OHSS' only. A total of three reviews reported separately on the subgroups 'moderate' and 'severe' OHSS. Three additional reviews reported only 'severe' OHSS; the remaining 21 reviews described that they reported on total 'OHSS' and provided no further explanation (e.g. this could have included moderate + severe, mild + moderate + severe, severe only). As almost all hyperstimulation cycles for in vitro fertilisation (IVF)/intracytoplasmic sperm injection (ICSI) are accompanied by some form of discomfort and ovarian enlargement, the subgroup 'mild OHSS' does not seem to represent a very significant clinical outcome (for a description of subgroups, see Appendix 1). Reporting only on the total incidence of OHSS with no further specification of cases per subgroup biases the comparability of studies and actual reporting on clinically significant OHSS rates. For example, some studies could have found mainly, or only, mild cases of OHSS, whereas other studies might not have even included or assessed mild cases and would have reported only moderate and/or severe cases of OHSS. This would mean that the former study would overestimate the number of OHSS cases for which the latter finding is more precise. Also, a study that reports only on hospitalised cases might underestimate moderate cases, which we consider clinically significant too. Moreover, reviews reporting on interventions that aim to reduce the incidence of OHSS included only 'high risk' populations, whereas most of the other reviews did not select their population specifically for this criterion, which makes it difficult to compare the effectiveness of different interventions within this overview. This heterogeneous method of reporting fails to acknowledge the incidence of moderate and severe cases, resulting in an evidence gap for this important adverse outcome of ART cycles.

Currently, many ART clinics apply preventive strategies such as natural cycle/mild stimulation IVF or cryopreservation of all embryos (the 'freeze‐all' approach) that have not been proven to have a beneficial effect on OHSS on the basis of low‐quality evidence (Allersma 2013; D'Angelo 2007; D'Angelo 2011). However, theoretical considerations suggest that these strategies are probably effective, as they eliminate certain steps of an ART cycle (preventing multiple follicle growth and implantation, thereby preventing an hCG surge); additional RCTs are needed to provide a robust evidence base for these practices.

Quality of the evidence

All included systematic reviews were prepared according to Cochrane guidelines and were of high quality in most respects, although only 17 of 27 had conducted a literature search within the past three years. Moreover, the included primary studies might be significantly older than the review publication date, thus sometimes reflecting outdated clinical practice or stimulation regimens. This cannot be avoided in an overview, as we summarise available evidence from existing reviews, and sometimes few or no recent RCTs are available.

This also has an impact on the quality of the evidence reported by primary studies in the included reviews. Using GRADE methods, we rated this quality level as very low to high. The main reasons for downgrading the quality of evidence included bias in the primary studies (inadequate reporting of allocation concealment and randomisation methods, lack of blinding) and imprecision. Evidence was frequently restricted to that provided by only a few included trials per comparison. Because clinically relevant OHSS is still a relatively rare outcome in ART cycles, and given that the primary study size and the number of studies per comparison have been limited for most reviews, the event rate of reported OHSS will remain low, as mild OHSS frequently is not reported. This implies that the quality of the evidence should be downgraded by one level (according to GRADE rules for downgrading dichotomous outcomes by one level for imprecision), and that the event rate < 300 and the total cumulative sample size were lower than the calculated optimal information size (OIS) (Schünemann 2013). As a result, the quality of the evidence on effectiveness of interventions for the outcome of OHSS remains 'very low' or 'low' for most interventions and comparisons.

Potential biases in the overview process

We identified no biases during the overview process.

Agreements and disagreements with other studies or reviews

Over the years, as new evidence from RCTs continues to emerge, a steady stream of publications aims to provide a comprehensive overview on the pathophysiology, prevention and treatment of OHSS. For example, in 2016 alone, Guo 2016 and Kwik 2016 were published, and recently, the American Society for Reproductive Medicine provided practice guidelines (ASRM 2016). In addition to this, an abundance of reviews have examined particular interventions covered in this overview, such as use of dopamine agonists (Baumgarten 2013; Kalampokas 2013; Kasum 2015; Leitao 2014).

Such publications encompass, for example, regional or national clinical practice guidelines or an overview of the literature. However, most also include data derived from retrospective and population‐based longitudinal studies or non‐randomised controlled trials. To our knowledge, this overview is the first to include solely systematic reviews on ART conducted according to rigorous Cochrane standards, thus showing high methodological quality. Moreover, we included not only reviews of interventions directly aiming to prevent OHSS, but also reviews of other ART interventions reporting on OHSS as an adverse effect, thus presenting a more complete overview of the literature currently available in the Cochrane Library.

Flow diagram of included reviews.
Figures and Tables -
Figure 1

Flow diagram of included reviews.

Extent of effect of interventions on OHSS rate and live birth rate (when reported) or clinical pregnancy: OR and 95% CI.
Figures and Tables -
Figure 2

Extent of effect of interventions on OHSS rate and live birth rate (when reported) or clinical pregnancy: OR and 95% CI.

Table 1. Review characteristics

Review ID

Number of included trials

Population

Definition of high risk for OHSS (where applicable)

Intervention

Comparison intervention/control

Primary outcomesa

Review limitations

ADA563

D'Angelo 2011

Coasting (withholding gonadotrophins) for preventing ovarian hyperstimulation syndrome

4 RCTs

340 women with PCOS

downregulated by GnRHa, undergoing superovulation in IVF or ICSI cycles

High risk: women with PCOS

Coasting when

oestradiol levels were > 2500 pg/mL or > 9000 pmol/L

Early unilateral follicular

aspiration

No coasting or other interventions

OHSSa

Live birtha

Clinical pregnancy

Number of oocytes retrieved

Multiple pregnancy

Miscarriage

Comparisons based on limited trial data

Live birth reported in only 1 trial

Trials lacked blinding, and half the trials lacked details on allocation concealment and incomplete outcome assessment

ADA561

D'Angelo 2007

Embryo freezing for preventing ovarian hyperstimulation syndrome

2 RCTs

151 women

downregulated by GnRHa, undergoing superovulation in IVF or ICSI cycles.

High risk: as defined by included studies

Cryopreservation

Fresh embryo transfer

Intravenous albumin

OHSSa

Clinical pregnancya

Live birth

Admissions

Evidence based on 2 trials, 1 for each comparison

Live birth reported in only 1 trial

Issues around methodological quality of both trials

TH1338

Tang 2016

Dopamine agonists for preventing ovarian hyperstimulation syndrome

16 RCTs

2091 women at high risk of developing OHSS undergoing ART

High risk: as defined by included studies

Cabergoline quinagolide, bromocriptine,

cabergoline + albumin,

cabergoline + HES

Placebo/no treatment/other treatment:

Albumin alone

HES

Coasting

Prednisolone

OHSSa

Live birtha

Clinical pregnancy

Adverse effects

Miscarriage

Multiple pregnancy

Allocation concealment and blinding not adequately reported.

One study used a co‐intervention of albumin IV and 1 of HES

Different regimens of cabergoline administration between included studies

Live birth rate reported in only 2 studies

Incomplete reporting of multiple pregnancy rate, adverse effects and miscarriage rate

PMA481

Youssef 2016b

Volume expanders for prevention of OHSS

9 RCTs

1660 (albumin) + 487 (HES) women

at high risk of developing OHSS undergoing ART cycles High risk: determined as number of follicles or oestradiol levels on day of hCG, as defined by included studies

Human albumin

Hydroxyethyl starch (HES)

Placebo/no treatment

OHSSa

Clinical pregnancy

Number of oocytes retrieved

Multiple pregnancy

Miscarriage

Live birth

No reporting of live birth rate

Limited by incomplete data reporting and lack of (details on) blinding

HA413

Youssef 2016

Recombinant vs urinary hCG for final oocyte maturation triggering in IVF and ICSI cycles

18 RCTs

2952 women undergoing ART

Recombinant hCG

Recombinant LH

Urinary hCG

OHSSa

Clinical pregnancy

Miscarriage

Oocytes retrieved

Tolerance

Live birth

Review authors combined ongoing pregnancy and live births together

Only 7 trials reported on live birth

Trials lacked details on allocation concealment, randomisation and blinding

MM1690

Youssef 2014

GnRHa vs hCG for oocyte triggering in antagonist‐assisted reproductive technology

17 RCTs

1847 women undergoing ART

GnRH agonist

hCG

OHSSa

Live birth ratea

Ongoing pregnancy

Clinical pregnancy

Multiple pregnancy

Miscarriage rate

Risk of bias in included studies. Limitations included premature termination, failure to clearly report methods and substantial heterogeneity

Adverse events such as multiple pregnancy rate were not well reported

AWP1710

Pouwer 2015

Long‐acting FSH vs daily FSH for women undergoing assisted reproduction

6 RCTs

3753 women with subfertility

Long‐acting FSH

Daily FSH

OHSSa

Live birth ratea

Ongoing pregnancy rate

Clinical pregnancy rate

Multiple pregnancy rate

Miscarriage rate

Adverse events

Satisfaction

Limited by risk of attrition bias in some primary studies and by serious imprecision

LDT120

Tso 2014

Metformin treatment before and during IVF or ICSI in women with PCOS

9 RCTs

816 women with PCOS

Metformin

Placebo

No treatment

OHSSa

Live birtha

Clinical pregnancya Miscarriage

Adverse events Number of oocytes

retrieved

Total dose FSH (IU) Number of days

gonadotrophin treatment

Cycle cancellation rate

Serum E2 level (nmol/L)

Half the trials were not blinded and lacked details on allocation concealment and randomisation

AM1335

Gibreel 2012

Clomiphene citrate in combination with gonadotropins for controlled ovarian stimulation in women undergoing IVF

14 RCTs,

12 for meta‐analysis

2536 (12 trials)

Subfertile women undergoing ART

Clomiphene citrate

± additional treatments

Alternative treatments for

COH

OHSSa

Live birth ratea

Miscarriage rate

Ectopic pregnancy

Foetal abnormality

Ongoing pregnancy rate

Cancellation rate

Live birth reported in only 5 trials

Most studies suffered from suboptimal methods and information on some outcomes was insufficient

TA1860

Allersma 2013

Natural cycle IVF for subfertile couples

5 RCTs

382 subfertile women and couples undertaking IVF treatment

Natural cycle IVF

Modified natural cycle IVF

COH IVF

OHSSa

Live birtha

Pregnancy

Ongoing pregnancy

Number of oocytes retrieved

Time to live birth

Number of cycles required to conceive

Cumulative pregnancy/live birth rate

Multiple pregnancy

Lack of embryos for cryopreservation

Cycle cancellation

Gestational abnormalities

Cancellation of treatment

Cost‐effectiveness

Few studies, live birth reported in only 1 very small trial

Inclusion criteria differed

MV263

van der Linden 2015

Luteal phase support for ART cycles

94 RCTs

26,198 women with any cause of subfertility undergoing ART

Progesterone

hCG

Placebo or no treatment

hCG

Progesterone + oestrogen Progesterone + GnRHa

Live birtha

Clinical pregnancy

Ongoing pregnancy

Miscarriage

OHSS

Multiple pregnancy

Poor reporting of study methods and imprecision due to small sample sizes

HA412

Al‐Inany 2016

Gonadotrophin‐releasing hormone antagonists for ART

73 RCTs

12,212 women undergoing ART

GnRH antagonist

Long‐course GnRHa

OHSSa

Live birtha

Ongoing pregnancy

Clinical pregnancy

Miscarriage

Cycle cancellation

Only 12 trials reported live birth

Trial methods limited by lack of blinding

Poor reporting of study methods for OHSS

AMY731

Yossry 2006

IVF vs tubal re‐anastomosis (sterilisation reversal) for subfertility after tubal sterilisation

No RCTs

NA

IVF

Tubal re‐anastomosis

Live birtha

Clinical pregnancy Multiple pregnancy

Other serious maternal morbidity, (incl OHSS)

Empty review with no

trials

No longer being updated

ZP672

Pandian 2015

IVF for unexplained subfertility

6 RCTs

733 couples with unexplained subfertility

IVF

Expectant management

Intrauterine insemination Intrauterine insemination +

ovarian stimulation

Clomiphene citrate

Live birtha

OHSS

Clinical pregnancy Multiple pregnancy

Some evidence was based

on a single trial

Limitations included imprecision and heterogeneity for some outcomes

LA541

Albuquerque 2013

Depot vs daily administration of GnRHa protocols for pituitary desensitisation in assisted reproduction cycles

16 RCTs,

12 for meta‐analysis

1811 couples with any cause of subfertility undergoing IVF with COH with hFSH, hMG or rFSH

Pituitary downregulation with depot administration of GnRHa

Daily administration of GnRHa

OHSSa

Live birtha

Clinical pregnancya

Miscarriage

Multiple pregnancy

S tudy quality unclear due to poor reporting. O nly four stu dies reported live birth an d only five described adequate methods for allocation concealment .

IOK973

van Wely 2011

Recombinant vs urinary gonadotrophin for ovarian stimulation in ART cycles

42 RCTs

9606 normogonadotrophic women undergoing fresh and/or frozen thawed IVF or ICSI cycles

Recombinant FSH

Urinary FSH

OHSSa

Live birtha

Clinical pregnancy

Miscarriage

Multiple pregnancy

Adverse effects

No difference reported in moderate/severe OHSS

WPM1780

Martins 2013

FSH replaced by low‐dose hCG in the late follicular phase vs continued FSH for ART

5 RCTs

351 women undergoing COH for ART

Low‐dose hCG instead of FSH in late follicular phase

Continued FSH in late follicular phase

OHSSa

Live birtha

Clinical pregnancy

Miscarriage

Small studies and low event rate

Total OHSS incidence reported

DHH752

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pretreatment for ovarian stimulation protocols for women undergoing ART

23 RCTs

2596 women of any age with subfertility regardless of cause, undergoing ART

Pretreatment with combined oral contraceptive pills

Pretreatment with progestogens

No pretreatment

Placebo

Progestogens

Oestrogens

Live birtha

OHSS

Clinical pregnancy

Miscarriage

Multiple pregnancy

Adverse effects

Only 3/23 studies reported on OHSS

2 of these 3 studies did not define how they diagnosed the condition

IOK972

Kwan 2014

Monitoring of stimulated cycles in assisted reproduction (IVF and ICSI)

6 RCTs

781 women undergoing COH in an IVF/ICSI cycle

Transvaginal ultrasonography + Oestradiol measurement

Transvaginal ultrasonography

Live birtha

OHSS

Clinical pregnancy

Miscarriage

Multiple pregnancy

Adverse effects

Only total OHSS reported, including mild OHSS

CMB1261

Boomsma 2012

Peri‐implantation glucocorticoid administration for ART cycles

14 RCTs

1879 subfertile patients undergoing IVF/ICSI, regardless of cause of infertility

Glucocorticoids in the peri‐implantation phase

No glucocorticoids in the peri‐implantation phase

Live birtha

Multiple pregnancya

OHSS

Clinical pregnancy Miscarriage

Adverse effects

Only 2 studies, pooled total OHSS

VJP951

Siristatidis 2016

Aspirin for IVF

13 RCTs

2653 women undergoing IVF/ICSI and their partners

Aspirin

No treatment

Placebo

Live birtha

OHSS

Clinical pregnancy Miscarriage

Multiple pregnancy

Adverse effects

Only 1 of 13 studies reported on OHSS and without exact numbers or explanation for numerators/denominators

CS1400

Siristatidis 2009

In vitro maturation in subfertile women with PCOS undergoing assisted reproduction

None

0 women with PCOS and subfertility

In vitro maturation + IVF/ICSI in women with PCOS

Conventional IVF/ICSI in women with PCOS

Live birtha

OHSS

Effectiveness

Clinical pregnancy Miscarriage

Adverse effects

Empty review

IRS911

Cheong 2013

Acupuncture and ART

20 RCTs

4544 women undergoing ART, any type of acupuncture at any time point before, after or during ART, intended to improve ART outcome

Acupuncture of men, women or both during COH

Acupuncture + ART

Acupuncture alone

No treatment

Placebo

Sham acupuncture

Acupuncture + ART

Live birtha

OHSS

Clinical pregnancy Miscarriage

Multiple pregnancy

Adverse effects

No trials reported on OHSS

MHM931

Mochtar 2007

Recombinant luteinising hormone (rLH) for COH in assisted reproductive cycles

14 RCTs

2612 subfertile ovulatory women undergoing IVF or ICSI

High risk: NA

Combination of rLH and rFSH for COH in IVF/ICSI followed by ET in GnRHa and GnRH antagonist protocols

rFSH alone for COH in IVF/ICSI followed by ET in GnRHa and GnRH antagonist protocols

OHSSa

Live birtha

Clinical pregnancy Miscarriage

Only 4/14 trials reported on OHSS

Pooled OHSS

No GRADE assessment in old version

KH291

Duffy 2010

Growth hormone for IVF

10 RCTs

440 women part of a subfertile couple undergoing IVF

Adjuvant growth hormone during conventional IVF

Conventional IVF

Live birtha

OHSS

Clinical pregnancy

Adverse effects

Only 4 of 10 RCTs reported on adverse events (which could include OHSS)

1 study actually mentioned OHSS (however, no cases); pooled OHSS

SD265

Siristatidis 2015

GnRHa protocols for pituitary suppression in assisted reproduction

37 RCTs

3872 women/couples with all types of infertility undergoing ART and using GnRHa for pituitary downregulation

Long protocol

Long luteal protocol

Short protocol

Dose continued

Dose continued after hCG administration

Pretreatment 2 weeks

Short protocol

Ultrashort protocol

Long follicular phase protocol

Ultrashort protocol

Dose stopped

Dose reduced

Dose discontinued after hCG administration

Pretreatment 3 weeks

Live birtha

OHSS

Clinical pregnancy Adverse effects

Only 2 of 37 included RCTs reported on OHSS for 2 of 9 compared regimens

JC1630

Showell 2013

Antioxidants for female subfertility

28 RCTs

3548 subfertile women referred to fertility clinic who might or might not undergo ART (IVF, ICSI or IUI)

Adjuvant antioxidants in females

No treatment

Placebo

Another antioxidant

Live birtha

Clinical pregnancy Miscarriage

Multiple pregnancy

Adverse effects (incl OHSS)

Only 3 studies reported: 1 no data and 2 no cases

aPrimary review outcome.

ART: artifical reproductive technology.

COH: controlled ovarian hyperstimulation.

ET: embryo transfer.

FSH: follicle‐stimulating hormone.

GnRHa: gonadotrophin‐releasing hormone agonist.

hCG: human chorionic gonadotrophin.

HES: hydroxyethyl starch.

hFSH: human follicle‐stimulating hormone.

hMG: human menopausal gonadotrophin.

ICSI: intracytoplasmic sperm injection.

IUI: intrauterine insemination.

IVF: in vitro fertilisation.

LH: luteinising hormone.

NA: not applicable.

OHSS: ovarian hyperstimulation syndrome.

PCOS: polycystic ovary syndrome.

RCT: randomised controlled trial.

rFSH: recombinant follicle‐stimulating hormone.

rLH: recombinant luteinising hormone.

Figures and Tables -
Table 1. Review characteristics
Table 2. Last search date assessment

Review no.

First review author

Review title

Date last assessed
up to date

< 3 years since last assessed up to date or deemed stable

ADA561

D'Angelo 2007

Embryo freezing for preventing OHSS

26/11/2010

Stable

ADA 563

D'Angelo 2011

Coasting (withholding of gonadotrophins) for preventing OHSS

19/07/2010

X

TH1338

Tang 2016

Dopamine agonists for preventing OHSS

15/08/2016

PMA481

Youssef 2016b

Volume expanders for prevention of OHSS

21/09/2016

HA413

Youssef 2016

Recombinant vs urinary hCG for final oocyte maturation triggering in IVF and ICSI cycles

23/04/2015

MM1690

Youssef 2014

GnRHa vs hCG for oocyte triggering in antagonist‐assisted reproductive technology

08/09/2014

X

LDT1201

Tso 2014

Metformin treatment before and during IVF or ICSI in women with PCOS

15/10/2014

AWP1710

Pouwer 2015

Long‐acting FSH vs daily FSH for women undergoing assisted reproduction

8/06/2015

AM1335

Gibreel 2012

Clomiphene citrate in combination with gonadotrophins for controlled ovarian stimulation in women undergoing IVF

23/03/2012

X

TA1860

Allersma 2013

Natural cycle IVF for subfertile couples

5/03/2013

MV263

van der Linden2015

Luteal phase support for ART cycles

25/11/2014

HA412

Al‐Inany 2016

Gonadotrophin‐releasing hormone antagonists for ART

28/04/2016

AMY731

Yossry 2006

IVF vs tubal re‐anastomosis (sterilisation reversal) for subfertility after tubal sterilisation

15/05/2009

Empty, stable

ZP672

Pandian 2015

IVF for unexplained subfertility

4/05/2015

LA541

Albuquerque 2013

Depot vs daily administration of GnRHa protocols for pituitary desensitisation in assisted reproduction
cycles

3/07/2012

IOK973

van Wely 2011

Recombinant vs urinary gonadotrophin for ovarian stimulation in ART cycles

20/10/2010

X

WPM1780

Martins 2013

FSH replaced by low‐dose hCG in late follicular phase vs continued FSH for ART

5/02/2013

DHH752

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pretreatment for ovarian stimulation protocols for women undergoing ART

16/11/2008

X

IOK972

Kwan 2014

Monitoring of stimulated cycles in assisted reproduction (IVF and ICSI)

30/05/2014

CMB1261

Boomsma 2012

Peri‐implantation glucocorticoid administration for ART cycles

20/09/2011

X

VJP951

Siristatidis 2016

Aspirin for IVF

9/05/2016

CS1400

Siristatidis 2009

In vitro maturation in subfertile women with PCOS undergoing assisted reproduction

17/02/2011

Empty

IRS911

Cheong 2013

Acupuncture and ART

22/07/2013

MHM931

Mochtar 2007

Recombinant luteinising hormone (rLH) for COH in assisted reproductive cycles

25/01/2007

X

KH291

Duffy 2010

Growth hormone for IVF

20/07/2009

X

SD265

Siristatidis 2015

GnRHa protocols for pituitary suppression in assisted reproduction

23/04/2015

JC1630

Showell 2013

Antioxidants for female subfertility

15/04/2014

ART: artifical reproductive technology.

COH: controlled ovarian hyperstimulation.

FSH: follicle‐stimulating hormone.

GnRHa: gonadotrophin‐releasing hormone agonist.

hCG: human chorionic gonadotrophin.

ICSI: intracytoplasmic sperm injection.

IUI: intrauterine insemination.

IVF: in vitro fertilisation.

OHSS: ovarian hyperstimulation syndrome.

PCOS: polycystic ovary syndrome.

rLH: recombinant luteinising hormone.

✔ under 3 years since last assessed as up to date

X over 3 years since last assessed as up to date

Figures and Tables -
Table 2. Last search date assessment
Table 3. AMSTAR assessment per review

Review no.

First review author + year

Review title

AMSTAR criteria

Prespecified question and inclusion criteria

Duplicate study selection and data extraction

Comprehensive literature search

Grey literature included

Lists included and excluded studies

Describes characteristics of included studies

Study quality assessed

Studies combined using appropriate methods

Likelihood of publication bias considered/tested

Potential for conflict of interest addressed

ADA561

D'Angelo 2007

Embryo freezing for preventing ovarian hyperstimulation syndrome

ADA 563

D'Angelo 2011

Coasting (withholding gonadotrophins) for preventing ovarian hyperstimulation syndrome

TH1338

Tang 2016

Dopamine agonists for preventing ovarian hyperstimulation syndrome

PMA481

Youssef 2016b

Volume expanders for the prevention of ovarian hyperstimulation syndrome

HA413

Youssef 2016

Recombinant versus urinary human chorionic gonadotrophin for final oocyte maturation triggering in IVF and ICSI cycles

MM1690

Youssef 2014

Gonadotropin‐releasing hormone agonist versus hCG for oocyte triggering in antagonist‐assisted reproductive technology

LDT1201

Tso 2014

Metformin treatment before and during IVF or ICSI in women with polycystic ovary syndrome

AWP1710

Pouwer 2015

Long‐acting FSH versus daily FSH for women undergoing assisted reproduction

AM1335

Gibreel 2012

Clomiphene citrate in combination with gonadotrophins for controlled ovarian stimulation in women undergoing in vitro fertilisation

TA1860

Allersma 2013

Natural cycle IVF for subfertile couples

MV263

van der Linden 2015

Luteal phase support for ART cycles

HA412

Al‐Inany 2016

Gonadotrophin‐releasing hormone antagonists for assisted reproductive technology

AMY731

Yossry 2006

In vitro fertilisation versus tubal re‐anastomosis (sterilisation reversal) for subfertility after tubal sterilisation

NA

NA

NA

NA

ZP672

Pandian 2015

In vitro fertilisation for unexplained subfertility

LA541

Albuquerque 2013

Depot versus daily administration of gonadotrophin‐releasing hormone
agonist protocols for pituitary desensitisation in assisted reproduction
cycles

IOK973

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

WPM1780

Martins 2013

FSH replaced by low‐dose hCG in the late follicular phase versus continued FSH for assisted reproductive techniques

DHH752

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pretreatment for ovarian stimulation protocols for women undergoing assisted reproductive techniques

IOK972

Kwan 2014

Monitoring of stimulated cycles in assisted reproduction (IVF and ICSI)

CMB1261

Boomsma 2012

Peri‐implantation glucocorticoid administration for assisted reproductive technology cycles

VJP951

Siristatidis 2016

Aspirin for in vitro fertilisation

CS1400

Siristatidis 2009

In vitro maturation in subfertile women with polycystic ovarian syndrome undergoing assisted reproduction

NA

NA

NA

NA

IRS911

Cheong 2013

Acupuncture and assisted reproductive technology

MHM931

Mochtar 2007

Recombinant luteinising hormone (rLH) for controlled ovarian hyperstimulation in assisted reproductive cycles

X

KH291

Duffy 2010

Growth hormone for in vitro fertilisation

X

SD265

Siristatidis 2015

Gonadotrophin‐releasing hormone agonist protocols for pituitary suppression in assisted reproduction

JC1630

Showell 2013

Antioxidants for female subfertility

Search date: 24/07/2016.

ART: artifical reproductive technology.

FSH: follicle‐stimulating hormone.

hCG: human chorionic gonadotrophin.

ICSI: intracytoplasmic sperm injection.

IUI: intrauterine insemination.

IVF: in vitro fertilisation.

NA: not applicable.

rLH: recombinant luteinising hormone.

Figures and Tables -
Table 3. AMSTAR assessment per review
Table 4. Summary of findings for OHSS: per review and/or per intervention

Review title and

comparison intervention/control

Assumed risk with comparator

Corresponding risk

with intervention

Relative effect

(95% CI)

Number of participants

(studies)

Quality of the evidence

(GRADE)

Comments

D'Angelo 2007

Embryo freezing for preventing ovarian hyperstimulation syndrome

(Embryo freezing vs fresh transfer)

Overall OHSS: 60 per 1000

Overall OHSS: 125 per 1000

(62 to 240)

OR 1.12

(0.01 to 2.29)

125

(1 study)

Low

Imprecision, number of events < 300

Evidence based on a single open‐label study with insufficient methodological details provided

D'Angelo 2007

Embryo freezing for preventing ovarian hyperstimulation syndrome

(Embryo freezing vs intravenous albumin)

Moderate or

severe OHSS:

77 per 1000

Moderate or

severe OHSS:

308 per 1000

(41 to 824)

OR 5.33

(0.51 to 56.24)

26

(1 study)

Very low

Imprecision, number of events < 300

Evidence based on a single open‐label trial

D'Angelo 2011

Coasting (withholding gonadotrophins) for preventing ovarian hyperstimulation syndrome

Moderate or

severe OHSS:

265 per 1000

Moderate or

severe OHSS:

58 per 1000

(11 to 241)

OR 0.17

(0.03 to 0.88)

68

(1 study)

Very low

Imprecision, number of events < 300

Evidence based on a single conference abstract
Insufficient methodological details provided

Tang 2016

Dopamine agonists for preventing ovarian hyperstimulation syndrome

Moderate or severe OHSS: 286 per 1000

Moderate or severe OHSS: 97 per 1000

(71 to 135)

OR 0.27

(0.19 to 0.39)

2091

(16 studies))

Moderate

Imprecision, number of events < 300

Lack of details for allocation concealment and blinding, selective reporting

Youssef 2016b

Volume expanders for the prevention of ovarian hyperstimulation syndrome

(human albumin vs placebo/no treatment)

Moderate or severe OHSS:

122 per 1000

Moderate or severe OHSS: 85 per 1000 (61 to 177)

OR 0,67

(0.47 to 0.95)

1452

(7 studies)

Very low

Imprecision, number of events < 300

Lack of details on allocation concealment and selective reporting

Youssef 2016b

Volume expanders for the prevention of ovarian hyperstimulation syndrome

(HES vs placebo)

Moderate or

severe OHSS: 164 per 1000

Moderate or

severe OHSS: 50 per 1000

(23 to 104)

OR 0.27

(0.12 to 0.59

272

(2 studies)

Very low

Imprecision, number of events < 300

Lack of details on allocation concealment and selective reporting

Youssef 2016b

Volume expanders for the prevention of ovarian hyperstimulation syndrome

(mannitol vs placebo)

Moderate or

severe OHSS: 517 per 1000

Moderate or

severe OHSS: 289 per 1000

(191 to 407)

OR 0.38

(0.22 to 0.64)

226

(1 study)

Low

Imprecision, number of events < 300

Lack of details on allocation concealment and selective reporting

Youssef 2016

Recombinant versus urinary human chorionic gonadotrophin for final oocyte maturation triggering in IVF and ICSI cycles (r‐hCG vs u‐hCG)

Overall OHSS: 27 per 1000

Overall OHSS:40 per 1000

(15 to 102)

OR 0.39

(0.25 to 0.61)

374

(3 studies)

Moderate

Imprecision, number of events < 300

One of the trials lacked methodological details on randomisation, allocation concealment and blinding

Youssef 2016

Recombinant versus urinary human chorionic gonadotrophin for final oocyte maturation triggering in IVF and ICSI cycles (r‐LH vs u‐hCG)

Overall OHSS: 10 per 1000

Overall OHSS:

17 per 1000

(11 to 84)

OR 1.76

(0.37 to 8.45)

417

(3 studies)

Low

Imprecision, number of events < 300

One of the trials lacked adequate methodological details

Youssef 2014

Gonadotropin‐releasing hormone agonist versus hCG for oocyte triggering in antagonist‐assisted reproductive technology

Overall OHSS: 5 per 1000

Overall OHSS:

1 per 1000

(0 to 2)

OR 0.15

(0.05 to 0.47)

989

(9 studies)

Moderate

Imprecision, number of events < 300

All studies at high risk of bias in 1 or more domains

None clearly reported blinded outcome assessment

Tso 2014

Metformin treatment before and during IVF or ICSI in women with polycystic ovary syndrome.

Overall OHSS: 270 per 1000

Overall OHSS: 97 per 1000

(62 to 153)

OR 0.29

(0.18 to 0.49)

798

(8 studies)

Moderate

Imprecision, number of events < 300

Pouwer 2015

Long‐acting FSH versus daily FSH for women undergoing assisted reproduction

(low dose)

Overall OHSS: 47 per 1000

Overall OHSS: 57 per 1000

(26 to 125)

RR 1.22

(0.56 to 2.66)

645

(3 studies)

Moderate

Imprecision, number of events < 300

Pouwer 2015

Long‐acting FSH versus daily FSH for women undergoing assisted reproduction

(medium dose)

Overall OHSS: 63 per 1000

Overall OHSS: 60 per 1000 (45 to 85)

RR 0.96

(0.68 to 1.35)

3075

(5 studies)

Low

Imprecision, number of events < 300

Confidence intervals compatible with clinically meaningful benefit in either arm or with no effect, plus high risk of attrition bias in 2 studies

Pouwer 2015

Long‐acting FSH versus daily FSH for women undergoing assisted reproduction

(high dose)

Overall OHSS: 0 per 1000

Overall OHSS: 0 per 1000

(0 to 0)

RR 1.73

(0.09 to 32.75)

33

(1 study)

Very low

Imprecision, number of events < 300

High risk of attrition bias

Gibreel 2012

Clomiphene citrate in combination with gonadotropins for controlled ovarian stimulation in women undergoing in vitro fertilisation

(clomiphene + gonadotropins vs gonadotropins)

Overall OHSS: 50 per 1000

Overall OHSS:12 per 1000

(5 to 27)

OR 0.23

(0.1 to 0.52)

1559

(5 studies)

Low

Imprecision, number of events < 300

Very wide 95% confidence interval crossing the threshold points of appreciable benefit or harm, which is 25%

Allersma 2013

Natural cycle IVF for subfertile couples

(natural cycle vs conventional IVF)

Overall OHSS: 67 per 1000

Overall OHSS:

13 per 1000

(1 to 393)

OR 0.10

(0.01 to 4.06)

60

(1 study)

Very low

Imprecision, number of events < 300

Only 1 study reporting on OHSS

Allocation concealment method not reported

van der Linden 2015

Luteal phase support for ART cycles

(hCG versus placebo/no treatment)

Overall OHSS: 41 per 1000

Overall OHSS:

155 per 100

(76 to 292)

OR 4.28

(1.191 to 9.6)

387

(1 study)

Low

Imprecision, number of events < 300

Poor reporting of study methods

van der Linden 2015

Luteal phase support for ART cycles

(progesterone vs hCG regimens)

Overall OHSS: 126 per 1000

Overall OHSS: 72 per 1000

(31 to 162)

OR 0.54

(0.22 to 1.34)

615

(4 studies)

Low

Imprecision, number of events < 300

Poor reporting of study methods

van der Linden 2015

Luteal phase support for ART cycles

(progesterone + GnRH agonist)

Overall OHSS: 50 per 1000

Overall OHSS: 50 per 1000

(17 to 137)

OR 1.00

(0.33 to 3.01)

300

(1 study)

Very low

Imprecision, number of events < 300

Poor reporting of study methods

van der Linden 2015

Luteal phase support for ART cycles

(progesterone vs progesterone + oestrogens)

Overall OHSS: 39 per 1000

Overall OHSS: 22 per 1000

(8 to 62)

OR 0.56

(0.2 to 1.63)

461

(2 studies)

Low

Imprecision, number of events < 300

Poor reporting of study methods

Al‐Inany 2016

Gonadotrophin‐releasing hormone antagonists for assisted reproductive technology

(GnRH antagonist vs GnRH agonist)

Overall OHSS: 114 per 1000

Overall OHSS: 73 per 1000

(62 to 85)

OR 0.61
(0.51 to 0.72)

7944

(36 studies)

Moderate

Methodological limitations including poor allocation concealment and lack of blinding

Yossry 2006

In vitro fertilisation versus tubal reanastomosis (sterilisation reversal) for subfertility after tubal sterilisation

(IVF vs tubal reanastomosis)

NA

NA

NA

NA

NA

Empty review

Pandian 2015

In vitro fertilisation for unexplained subfertility

(IVF vs IUI + gonadotropins/clomiphene citrate)

Overall OHSS: 58 per 1000

Overall OHSS: 66 per 1000

(26 to 158)

OR 1.15 (0.43 to 3.06)

324

(2 studies)

Low

Imprecision, number of events < 300

Only 2 studies on OHSS reported

Albuquerque 2013

Depot versus daily administration of gonadotrophin releasing hormone agonist protocols for pituitary desensitization in assisted reproduction cycles

(depot vs daily gonadotropins)

Overall OHSS: 3 per 100

Overall OHSS: 2 per 100
(1 to 6)

OR 0.84
(0.29 to 2.42)

570

(5 studies)

Low

Most studies were classified as at unclear risk of bias for all domains

Imprecision, number of events < 300
Studies were insufficient to assess publication bias

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

(rFSH vs HMG/HMG‐HP)

Overall OHSS: 17 per 1000

Overall OHSS: 17 per 1000

(10 to 28)

OR 1.00

(0.58 to 1.71)

3197

(11 studies)

High

Imprecision, number of events < 300

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

(rFSH vs FSH‐P)

Overall OHSS: 28 per 1000

Overall OHSS:49 per 1000

(25 to 95)

OR 1.79

(0.89 to 3.62)

1490

(6 studies)

Higha

Imprecision, number of events < 300

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

(rFSH vs FSH‐HP)

Overall OHSS: 28 per 1000

Overall OHSS:

31 per 1000

(20 to 48)

OR 1.11

(0.70 vs 1.75)

3053

(14 studies)

Higha

Two additional trials excluded in sensitivity analyses because it was unclear if data were reported according to ITT analysis (those were included for "Overall OHSS")

van Wely 2011

Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles

(rec‐hCG vs u‐hCG)

Overall OHSS: 19 per 1000

Overall OHSS:

22 per 1000

(16 to 30)

OR 1.18 (0.86 vs 1.61)

7740

(32 studies)

Higha

Imprecision number of events < 300

Martins 2013

FSH replaced by low‐dose hCG in the late follicular phase versus continued FSH for assisted reproductive techniques

(low‐dose hCG vs FSH in late follicular phase)

Overall OHSS: 3 per 100

Overall OHSS: 1 per 100

(0 to 4)

OR 0.30 (0.06 to 1.59)

351

(5 studies)

Very low

Imprecision, number of events < 300

Inconsistency, high risk of bias

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pre‐ treatment for ovarian stimulation protocols for women undergoing assisted reproductive techniques

(OAC plus antagonist vs antagonist)

Overall OHSS: 17 per 1000

Overall OHSS: 25 per 1000

(5 to 133)

OR 1.5 (0.26 to 8.8)

234

(1 study)

Very low

Single study reporting on OHSS

Imprecision, number of events < 300

Wide confidence intervals that cross line of no effect

High risk of attrition bias

Smulders 2010

Oral contraceptive pill, progestogen or oestrogen pre‐ treatment for ovarian stimulation protocols for women undergoing assisted reproductive techniques

(OAC plus antagonist vs agonist)

Overall OHSS: 55 per 1000

Overall OHSS: 35 per 1000

(12 to 100)

OR 0.63

(0.21 to 1.92)

290 (2 studies)

Very low

Imprecision, number of events < 300

One study at high risk of attrition bias

Kwan 2014

Monitoring of stimulated cycles in assisted reproduction

(IVF and ICSI)

(transvaginal ultrasound + estradiol vs transvaginal ultrasound)

Overall OHSS: 36 per 1000

Overall OHSS: 36 per 1000

(18 to 75)

OR 1.03

(0.48 to 2.20)

781

(6 studies)

Low

Imprecision, number of events < 300 with wide confidence intervals

Methods of randomisation inadequately described in 3 of 6 trials, allocation concealment inadequately described in all 6 trials and blinding inadequately described in 5 of 6 trials

No definition of OHSS provided by authors of these 6 studies

Boomsma 2012

Peri‐implantation glucocorticoid administration for assisted reproductive technology cycles

(adjuvant glucocorticoids vs no glucocorticoids)

Overall OHSS: 194 per 1000

Overall OHSS: 159 per 1000

(64 to 392)

OR 0.82 (0.33 to 2.02)

151

(2 studies)

Low

Imprecision, number of events < 300

Siristatidis 2016

Aspirin for in vitro fertilisation

(aspirin vs no treatment/placebo)

NA

NA

NA

NA

NA

Only 1 study reported on OHSS; no exact numbers or explanation of numerators/denominators given

Siristatidis 2009

In vitro maturation in subfertile women with polycystic ovarian syndrome undergoing assisted reproduction

(IVM vs conventional IVF)

NA

NA

NA

NA

NA

Empty review

Cheong 2013

Acupuncture and assisted reproductive technology

(acupuncture vs no acupuncture/sham acupuncture)

NA

NA

NA

NA

NA

No studies reported on OHSS

Mochtar 2007

Recombinant luteinizing hormone (rLH) for controlled ovarian hyperstimulation in assisted reproductive cycles

(combined rLH + FSH vs FSH )

Overall OHSS: 20 per 1000

Overall OHSS:

27 per 1000

(12 to 59)

OR 1.34

(0.58 to 3.09)

986

(7 studies)

Low

Imprecision, number of events < 300

Some methodological details unclear

Duffy 2010

Growth hormone for in vitro fertilization

(growth hormone vs no treatment/placebo)

NA

NA

NA

NA

NA

Only 1 study reported on OHSS; however, no cases of OHSS were reported

Siristatidis 2015

Gonadotropin‐releasing hormone agonist protocols for pituitary suppression in assisted reproduction

(different protocols vs other protocol)

Overall OHSS: 20 per 1000

Overall OHSS: 27 per 1000

(12 to 59)

OR 1.34

(0.58 to 3.09)

986

(7 studies)

Low

Imprecision, number of events < 300

Some methodological details unclear

Showell 2013

Antioxidants for female subfertility

(antioxidants vs no treatment/placebo/other antioxidant)

NA

NA

NA

NA

NA

Although 3 studies reported on OHSS, no numbers were given, so effect size could not be calculated

aReview authors GRADED these outcomes as 'high quality'; however, the total event rate < 300 would justify downgrading for this to moderate‐quality evidence.

ART: artifical reproductive technology.

FSH: follicle‐stimulating hormone.

FSH‐HP: highly purified FSH.

hCG: human chorionic gonadotrophin.

HES: hydroxyethyl starch.

ICSI: intracytoplasmic sperm injection.

IUI: intrauterine insemination.

IVF: in vitro fertilisation.

IVM: in vitro maturation.

NA: not applicable.

OAC: oral anticoagulant.

OHSS: ovarian hyperstimulation syndrome.

OR: odds ratio.

rFSH: recombinant follicle‐stimulating hormone.

r‐hCG: recombinant human chorionic gonadotrophin.

rLH: recombinant luteinising hormone.

Total: any grade of OHSS.

u‐hCG: urinary human chorionic gonadotrophin.

Figures and Tables -
Table 4. Summary of findings for OHSS: per review and/or per intervention