Abstract
Purpose
Is the presence of cytoplasmic strings (CS) in human blastocysts associated with the probability of clinical pregnancy with fetal heart (CPFH) after transfer.
Methods
This case-control study involved 300 single blastocyst transfers. 150 of these resulted in a CPFH (cases) while 150 did not (controls). All embryos were cultured in Embryoscope+ and AI software (IVY) was used to select the blastocyst with the highest score from the cohort for transfer. An embryologist, blind to the transfer outcome, recorded the CS number, location, and duration of their activity.
Results
There was a significant difference in the number of blastocysts that contained CS, with 97.3% of women’s blastocysts resulting in +CPFH containing the CS compared to 88.7% of blastocysts in women who did not have a pregnancy (p = 0.007, OR; 4.67, CI 95% 1.5–14.2). CS appeared 2.4 h earlier in embryo development in the +CPFH group compared to their negative counterparts (p = 0.007). There was a significant difference in the average number of CS/blastocyst with a higher number being present in those that achieved a clinical pregnancy (mean: 6.2, SD 2.9) compared to those that did not (mean: 4.6, SD 3.0) (p ≤ 0.0001). There was a significant increase in the number of vesicles seen traveling along the CS with more seen in the blastocysts resulting in a +CPFH (mean: 4.3 SD 2.1) compared to those in the −CPFH group (mean: 3.1, SD 2.1).
Conclusion
This study has shown that the presence of cytoplasmic strings in human blastocysts is associated with the probability of clinical pregnancy with fetal heart.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10815-021-02213-1.
Keywords: Blastocyst, Clinical pregnancy, Fetal heart, Cytoplasmic strings, Time-lapse
Introduction
Single blastocyst transfer is predominately becoming standard practice in IVF clinics worldwide in order to reduce the well-documented maternal and neonatal risks associated with multiple pregnancies [1, 2]. Consequently, this has placed increasing pressure on the embryologist to select a single embryo from the patient’s cohort for transfer that will result in the highest implantation potential for that cycle.
Embryo selection methods traditionally have involved grading systems based on morphology [3–6]. These morphological assessments are performed at specified time points during the embryo’s development. However, due to the nature of the developing embryo, these assessments could differ considerably throughout the day if performed more frequently. The introduction of time-lapse technology into the laboratory’s daily practice has opened up additional means of embryo assessment.
In the past few years, the literature has been heavily focused on the morphokinetic parameters of embryo development to construct algorithms, or identify biomarkers that seek to predict the embryo’s implantation potential [2, 7–12]. However, this task has been more complex than originally expected, due to the highly dynamic nature of the preimplantation stages [13]. Time-lapse monitoring (TLM) provides an optimal culture environment [14], while allowing a detailed analysis of key developmental events by providing the embryologist with an array of uninterrupted images. These image sequences can be further analyzed by the embryologist or by sophisticated computer software using artificial intelligence (AI) [15].
These images have made it possible for the identification of cytoplasmic strings (CS) in the developing blastocyst. CS are long thin projections that connect the inner cell mass (ICM) cell to the trophectoderm (TE) cells [16] and range from 0.1 to 3.0μm in diameter [17]. Historically in the literature, these structures have been described as “string-like” [18], “tongue-like” [16, 19], or “trophoblast processes” [20], as well as “cytoplasmic bridges” [21, 22] and “filopodia” [16]. Long traversing CS have been identified in 40% of in vivo produced mouse blastocysts, and almost all in vitro cultured mouse blastocysts [16], this has been suggested to be due to suboptimal mouse culture conditions [16, 19]. By using time-lapse image analysis, the long traversing CS can now be readily identified and accurately categorized in human blastocysts.
CS in human blastocysts are currently of unknown function and the literature is scarce. Most reports are in animal blastocysts [16, 20, 23] with only a handful reported in human blastocysts [18, 19, 21, 24]. The reports that do exist in humans are mostly limited to observations of their existence. It has been hypothesized that they may function as an ultrastructure transport mechanism [16] of unknown importance. In terms of clinical significance, the available data is limited to only one study, suggesting that once the blastocyst has completely expanded, the presence of these structures has a negative impact on implantation [19], but with no detailed information as to how this conclusion was reached.
The aim of this study was for the first time to focus on the clinical significance of these structures and determine if the presence of CS in human blastocysts is associated with the probability of clinical pregnancy with fetal heart after transfer (CPFH).
Materials and methods
Study design
This was a two-center case-control study performed between February 2018 and November 2019. Approval for this study was obtained from the IVFAustralia Research and Development Committee (2017/104). The study involved 300 single blastocyst transfers from 300 women undergoing either IVF or ICSI utilizing fresh oocytes. One hundred fifty of these resulted in a CPFH (1 sac, 1 FH) (cases) while 150 had no CPFH (0 sac, 0 FH) (controls). The blastocysts were assessed for CS in the days following embryo transfer prior to transfer outcome being known. Once transfer outcome was known the assessed blastocyst was designated to either the +CPFH group or the −CPFH group. Blastocysts that resulted in a pregnancy but not in 1 sac, 1 FH, e.g., anembryonic pregnancies or pregnancies that were miscarried before FH could be seen, were excluded. Blastocysts were assessed for CS until 150 met the criteria of the +CPFH group.
All embryos had been cultured in Embryoscope+ post-insemination and until day 5. Fresh autologous and donor oocytes were included in the study. Blastocysts were not suitable for the study if they did not reach the full blastocyst stage (i.e., exceeding early blastulation) by day 5, 112–120 h post insemination (hpi). Time-lapse videos generated from the Embryoscope+ were run through the artificial intelligence software (IVY) in order to select the embryo with the highest probability of a CPFH for embryo transfer from the patient cohort [15]. Videos of the embryo transferred that were out of focus or had debris in the well were excluded from the study (n = 14) so that IVY could accurately analyze the images.
Ovarian stimulation, oocyte retrieval, and semen preparation
Ovarian stimulation, oocyte retrieval, and semen preparation were performed as described previously [25]. Patients were stimulated using gonadotrophin-releasing hormone (GnRH) and transvaginal ultrasounds along with serum estradiol level were used to monitor follicular growth. Final oocyte maturation was triggered with human chorionic gonadotropin (hCG) and standard ultrasound-guided oocyte retrieval was performed 36 h after the trigger injection. Semen was prepared using 40:80% Puresperm density gradients (Nidacon, Sweden) and isolated into a pellet. The pellet was re-suspended in GMOPS media (Vitrolife, A/S, Aarhus, Denmark) and held at room temperature until use.
Oocyte insemination and embryo culture
Oocytes assigned for IVF were cultured in a pre-equilibrated 4 well NUNC dish (ThermoFisher Scientific, USA) at 37°C, 5% O2, and 6% CO2 in a MINC bench top incubator (COOK Medical, Australia). Each well contained G-IVF with an Ovoil mineral oil overlay (Vitrolife, A/S, Aarhus, Denmark). IVF insemination occurred between 39 and 41 h post trigger injection. Oocyte denudation occurred in those oocytes assigned for ICSI at 39–40 h post trigger injection. This was performed in an EmCell (HD Scientific, Australia) using Hyase-10X (Vitrolife, A/S, Aarhus, Denmark) and mechanical pipetting with a stripper tip (CooperSurgical, USA). All metaphase II oocytes present at the time of injection 39–41 h post trigger injection were injected at × 200 magnification using standard ICSI procedures. The EmbryoSlide ® (Vitrolife, A/S, Aarhus, Denmark) was prepared with warmed equilibrated G-TL with an Ovoil overlay (5% O2 and 6% CO2 and at 37 °C). Following insemination (ICSI day 0 and IVF day 1), the zygotes were loaded into the EmbryoSlide®and into the EmbryoScope+ 37 °C, 5% O2, 6% CO2 balance N2 immediately at completion of ICSI, or following stripping of the oocyte-cumulus complex at IVF fertilization check.
Embryo assessment
For ICSI inseminated zygotes fertilization was assessed 16-19 h post-insemination through the use of images provided by the time-lapse incubator software. IVF inseminated zygotes were assessed using a stereomicroscope (Leica, Germany), at 16–19-h post-insemination, and zygotes were transferred to the EmbryoSlide® and placed into the EmbryoScope+. Embryo development was monitored using the image software EmbyroViewer® (EmbryoScopeTM, Vitrolife, A/S, Aarhus, Denmark). Embryo morphology was also recorded without removal from the EmbryoScope+ using Gardner criteria on day 5 of culture (110–115 h post-insemination) [5].
Time-lapse monitoring and artificial intelligence software
EmbryoScope+ image acquisition was set to every 10 min at 11 focal planes. Key events in embryo development were annotated using the software EmbryoViewer®. Once at day 5 of culture, videos generated from the EmbryoViewer® software for all of the patient’s embryos in the EmbryoSlide® were downloaded as a single video file. AI software (IVY, Harrison AI, Australia) was applied to the entire cohort to identify the blastocyst with the highest clinical probability of a fetal heart (CPFH) for transfer. IVY software is a deep learning model that processes the raw time-lapse images from the Embryoscope+. This is executed through interconnecting neurons that are many layers deep [15]. Prior to embryo transfer, a trained embryologist verified the blastocyst selected by IVY. This was to ensure that the blastocyst video was clear of debris, bubbles, and shadowing so that an accurate AI score could be attained. If this criterion was not met, then the blastocyst was excluded from the study. Software version 0.1.2 was used for all blastocysts assessed.
Cytoplasmic string assessment
Cytoplasmic string assessment was undertaken in the days following embryo transfer. A single assessor, blind to the outcome of the embryo being transferred, examined the presence of CS in each blastocyst with the highest CPFH probability as selected by the AI Software. Time-lapse videos were reviewed using the EmbyroViewer® Software. CS (Fig. 1) were identified in the blastocysts at the start of blastulation by visualizing the videos at a controlled speed, using a Griffin PowerMate Control Knob (Griffin Technology, USA). Images were played back and forth while utilizing the 11 focal planes until the CS were no longer identified in the developing blastocyst. The presence and absence of CS were examined for each transferred blastocyst. The time points and stage of embryo development in which the CS first appeared and the time their activity ceased were recorded. The number present and location of the CS within the developing blastocyst was also noted. The location of CS formation was based on the blastocyst polarity regions as defined by Gardner, 1997 [26]. Regions where CS activity occurred were classified as mural, polar-mural junction (p-mj), and both (mural and polar-mural junction). CS were further assessed to determine if they contained the presence and absence of vesicle-like bulges. The number of vesicles was recorded as a single CS that presented with the largest number of vesicles seen traveling along it, throughout the blastocyst’s entire development. These bulges do not appear to have a name in the literature; however, it can be seen in Fig. 2 a–e and Supplementary Video 1, the movement of these bulges along the CS. Vesicle direction was recorded in three categories: vesicles originating from the trophectoderm (TE) cells and traveling towards the inner cell mass (ICM), originating from ICM and traveling towards the TE cells, or bidirectional (back and forth trajectory between the ICM and TE cells). For each blastocyst containing the CS with vesicles, the number and direction of vesicle movement were recorded.
Fig. 1.
A good-quality expanded blastocyst at day 5 (104.8 hpi). A single cytoplasmic string (arrow) can be seen traversing the blastocoel cavity, extending between the ICM and TE cells
Fig. 2.
a–e An early day 5 blastocyst shown between 104.8 and 107 hpi. A single cytoplasmic string can be seen traversing the blastocoel cavity, where it connects the ICM and the TE cells. A single vesicle can be seen emerging from the TE cells (a). As the blastocyst develops and expands, the single vesicle moves along the cytoplasmic string and another vesicle emerges from the TE cells (b). These vesicles continue to move along the cytoplasmic string from the TE cells towards the ICM (c, d). The two vesicles are delivered to the ICM and absorbed, taking a total of 2.2 h to occur (e)
Statistical analysis
All continuous variables are presented as mean and range. Categorical variables are presented as proportions. Considering the observational, retrospective nature of this study, statistical adjustment for baseline differences identified through bivariate (p < 0.05) analyses was performed. This was achieved by constructing appropriate multivariable logit regression models using the generalized estimating equations framework. All statistical analyses were performed with STATA (v.14.2, StataCorp, USA) and statistical significance was set at p ≤ 0.05.
Results
A total of 300 individual patients contributed to the study (i.e., 300 cycles) from these 2365 oocytes were retrieved (mean: 7.9 oocytes, SD: 4.6). There were 150 women in the group that had a positive clinical pregnancy with a FH (+CPFH) (i.e., positive BHCG, 1 gestational sac, and 1 fetal heart seen on ultrasound) and 150 women comprised the control group (−CPFH) (i.e., negative BCHG, 0 gestational sac, and 0 FH). There was a significant difference in the live birth rate in +CPFH blastocysts containing CS 88.4% (129/146) compared to those blastocysts that did not contain CS 50% (2/4) (p = 0.229). Additionally, those blastocysts that had CS present but did not result in a live birth, 8.2% (12/146) resulted in miscarriage, 2.1% (3/146) in a missed abortion, 0.6% (1/146) were terminated due to trisomy 21 and 0.6% (1/146) resulted in neonatal death (21 weeks of gestation).
Baseline and embryological characteristics
When the baseline characteristics of the two groups were investigated, there were no significant differences detected in the patient age, AMH, type of infertility, FSH starting dose or the duration of stimulation, the stimulation type, and the type of infertility (Table 1). There was a significant difference in the indication for treatment with 47% of the women who had a −CPFH falling into the unknown indication category, compared to 33% of women in the +CPFH group (p = 0.037).
Table 1.
Baseline demographics and cycle characteristics for cycles of transferred blastocysts that resulted in a clinical pregnancy with a fetal heart, compared to the control group
| Parameter | +CPFH (n = 150) | −CPFH (n = 150) | p-value |
|---|---|---|---|
|
Female age (years) Mean (range) |
33.8 (23.74–48.27) |
34.7 (18.8–44.39) |
0.088 |
|
Male age (years) Mean (range) |
35.1 (23.08–55.91) |
36.1 (19.92–52.75) |
0.163 |
|
AMH (pmol/L) Mean (range) |
24.5 (0.1–160) |
23.7 (1–127.1) |
0.954 |
| Type of infertility (%) | Primary: 144 (96.0) | Primary: 143 (95.3) | 0.784 |
| Secondary: 6 (4.0) | Secondary: 7 (4.7) | ||
|
Starting FSH dose (IU) Mean (range) |
192.13 (75–525) |
218.6 (50–1575) |
0.155 |
| GnRH analog (%) | Antagonist: 142 (94.7) | Antagonist: 142 (94.7) | 0.797 |
| Agonist: 8 (5.3) | Agonist: 8 (5.3) | ||
|
Duration of stimulation (days) Mean (range) |
7.7 (3–34) |
7.5 (3–31) |
0.312 |
|
Total FSH dose (IU) Mean (range) |
1985.0 (450–7200) |
2184.3 (500–16500) |
0.240 |
| Indication for treatment | Male factor: 30 | Male factor: 17 | 0.037 |
| Female factor: 35 | Female factor: 28 | ||
| Combination: 14 | Combination: 8 | ||
| Unknown: 50 | Unknown: 71 | ||
| Social: 21 | Social: 26 |
When investigating the embryological characteristics (Table 2) there was a significant difference in the number of oocytes collected (p = 0.005), with those women achieving a +CPFH having a higher number of oocytes collected (mean: 8.7 oocytes, SD: 5.0) compared to those women who had a −CPFH (mean: 7.2 oocytes, SD: 4.1). There was also a significant difference in the number of oocytes injected between the two groups (p = 0.011), with women in the +CPFH group having a higher number of oocytes injected (mean: 4.6 oocytes injected, SD: 5.2) compared to those women who had a −CPFH (mean: 3.2 oocytes injected, SD: 4.1). There were no significant differences detected in the insemination method between the two groups. In addition, there were no significant differences between the insemination method used and the presence of CS. 91.9% (125/136) of the blastocysts that had been inseminated by IVF had CS present, while 93.9% (154/164) of the blastocysts that had been inseminated using the ICSI technique had CS present (p = 0.377). There were no significant differences detected in the number of oocytes injected between the two groups.
Table 2.
Embryological characteristics for cycles of transferred blastocysts that resulted in a clinical pregnancy with a fetal heart, compared to the control group
| Parameter | +CPFH (n = 150) | −CPFH (n = 150) | p-value |
|---|---|---|---|
| Insemination method |
IVF: 60 ICSI: 90 |
IVF: 76 ICSI: 74 |
0.634 |
|
Number of oocytes collected Mean (range) |
8.7 (1–27) |
7.2 (1–20) |
0.005 |
|
Number of oocytes injected Mean (range) |
4.6 (0–26) |
3.2 (0–16) |
0.011 |
| Blastocyst quality (%) | 0.0001 | ||
| Top: AA, BA | Top quality: 103 (68.7) | Top quality: 69 (46) | |
| Medium: BB, AB | Medium quality: 39 (26) | Medium quality: 49 (32.7) | |
| Poor: anything containing a C | Poor quality: 8 (5.3) | Poor quality: 32 (21.3) | |
| Stage of embryo transfer | BL: 17 | BL: 32 | 0.002 |
| XBL: 69 | XBL: 80 | ||
| HGBL: 64 | HGBL: 38 | ||
|
AI score Mean (range) |
40.9 (0–68.0) |
32.2 (0–64.9) |
< 0.0001 |
BL blastocyst, XBL expanding blastocyst, HGBL hatching blastocyst
There was a significant difference in the embryo stage of development at the time of transfer (p = 0.002), with a larger proportion being at the later stages of blastulation in the positive FH group compared to those in the −CPFH group. As expected, there was a significant difference in the AI scores of the two groups with women in the +CPFH group having a higher score (mean: 40.9, SD: 15.8) compared to their −CPFH counterparts (mean: 32.2, SD: 17.6) (p < 0.001). There was also a significant difference in blastocyst quality, with the +CPFH group including more top-quality blastocysts (68.7%) compared to those with a −CPFH (46%), (P < 0.001).
Morphokinetic parameters
When known key embryological developmental events (Supplementary Table 1) were examined there were significant differences in the timings detected at the PN appearance, tPNf, t2, t4, t6, t7, tSB, tB, and tEB stages. The embryos in the +CPFH group all developed faster than those in the −CPFH group. There were no significant differences detected at the 2PB, t3, t5, t8, tM, and tHB stages (Appendix Table 6) between the two groups.
Table 6.
Morphokinetic markers assessed for blastocysts that were transferred, which resulted in a clinical pregnancy with a fetal heart, compared to the control group
| Parameter | +CPFH (n = 150) |
−CPFH (n = 150) |
p-value |
|---|---|---|---|
| 2PB |
3.5 (1.5–7.6) |
3.6 (1.8–9.3) |
0.790 |
| PN appear |
17.3 (5.0–25.8) |
18.1 (7.8–30.2) |
0.005 |
| tPNF |
22.7 (16.8–29.0) |
23.6 (18.3–35.6) |
0.002 |
| t2 |
25.8 (19.3–35.8) |
26.8 (21.1–43.8) |
0.001 |
| t3 |
35.6 (24.7–45.5) |
36.4 (25.6–57.5) |
0.086 |
| t4 |
37.0 (28.8–51.1) |
38.2 (28.9–57.8) |
0.009 |
| t5 |
47.7 (33.0–60.2) |
48.8 (33.1–71.0) |
0.069 |
| t6 |
49.7 (33.0–63.0) |
51.3 (34.9–89.1) |
0.019 |
| t7 |
51.7 (40.2–74.2) |
53.6 (35.6–89.9) |
0.012 |
| t8 |
55.1 (40.9–78.9) |
56.9 (37.4–97.4) |
0.066 |
| tM |
84.6 (65.9–109.3) |
86.1 (63.1–106.4) |
0.095 |
| tSB |
95.5 (82.1–112.5) |
97.1 (83.1–112.1) |
0.013 |
| tB |
102.2 (88.2–116.3) |
104.7(91.7–116.7) | 0.001 |
| tEB |
106.7 (90.8–118.7) |
108.7 (97.5–123.4) |
0.004 |
| tHB |
109.9 (97.7–117.6) |
110.7 (100.9–117.8) |
0.377 |
Cytoplasmic string presence and activity
When looking at the CS presence (Table 3) and the number of blastocysts that contained CS, there was a significant difference with 97.3% of women’s blastocysts that +CPFH contained the CS compared to 88.7% of blastocysts in those women who did not have a pregnancy (p = 0.007, OR; 4.67, CI 95% 1.5-14.2).
Table 3.
Cytoplasmic string activity recorded for blastocysts that were transferred, which resulted in a clinical pregnancy with a fetal heart, compared to the control group
| Parameter | +CPFH (n = 150) Mean (range) |
−CPFH (n = 150) Mean (range) |
p-value |
|---|---|---|---|
| Number of blastocysts with or without CS | With CS: 146 | With CS: 133 |
0.007 OR; 4.67, (CI 95%1.5–14.2) |
| Without CS: 4 | Without CS: 17 | ||
| CS appearance time (h) |
101.1 (84.9–114.8) |
103.5 (88.9–114.8) |
0.001 |
| CS end time (h) |
113.5 (100.1–119.5) |
114.1 (100.3–124.8) |
0.175 |
| Duration of CS presence (h) |
12.4 (1.2–26.8) |
10.6 (0.1–25.6) |
0.006 |
| Average number of CS /blastocyst |
6.2 (0–15) |
4.6 (0–12) |
< 0.0001 |
| Stage of blastocysts growth when CS appear | EBL: 102 | EBL: 87 | 0.060 |
| BL: 41 | BL: 35 | ||
| XBL: 2 | XBL: 11 | ||
| HGBL: 1 | HGBL: 0 | ||
| Stage of blastocysts growth when CS cease | EBL: 1 | EBL: 4 | 0.041 |
| BL: 18 | BL: 22 | ||
| XBL: 71 | XBL: 75 | ||
| HGBL: 56 | HGBL: 32 | ||
| CS formation (region of blastocyst) | p-mj: 1 | p-mj: 3 | 0.188 |
| Mural: 58 | Mural: 41 | ||
| Mural + p-mj: 87 | Mural + p-mj: 89 | ||
| Average number of vesicles traveling on a single CS |
4.3 (0–12) |
3.1 (0–8) |
< 0.0001 |
| Direction vesicles traveled | TE-ICM: 100 | TE-ICM: 84 | 0.113 |
| ICM-TE: 6 | ICM-TE: 13 | ||
| Bidirectional: 38 | Bidirectional: 27 |
CS cytoplasmic strings, BL blastocyst, XBL expanding blastocyst, HGBL hatching blastocyst, TE trophectoderm, ICM inner cell mass, p-mj polar-mural junction
The CS appeared 2.4 h earlier in blastocyst development in the +CPFH group (mean: 101.1 hpi, SD: 5.8), compared to their negative counterparts (mean: 103.5 hpi, SD: 6.1) and this was a significant difference (p = 0.001). There was no significant difference detected in the time when CS activity ceased between the two groups. However, the duration of CS activity was significant with blastocysts in the +CPFH group having CS active for 1.8 h longer (mean: 12.4 h, SD: 5.3) compared to their negative counterparts (mean: 10.6 h, SD: 5.5) (p = 0.005).
There was a significant difference in the average number of CS per blastocyst with a higher number of CS being present in those that did achieve a clinical pregnancy (mean: 6.2, SD 2.9) compared to those which did not (mean: 4.6, SD 3.0) (p ≤ 0.0001).
There was no significant difference detected in CS formation between the two groups when the polarity region of the blastocyst was investigated. In both groups, the majority of CS were identified in both the mural and the polar-mural junction regions of the developing blastocyst. There was a significant increase in the number of vesicles seen traveling along the CS with more vesicles seen in those blastocysts in the +CPFH group (mean: 4.3 vesicles on a single CS/blastocyst SD 2.1) compared to those in the −CPFH group (mean: 3.1 vesicles on a single CS/blastocyst, SD 2.1). When the direction of the vesicle movement was analyzed, there were no significant differences detected between the two groups. However, within both groups, majority of the vesicles were seen traveling from the TE cells to the ICM.
Blastocyst quality and cytoplasmic string presence
There was no significant difference in live birth rate and blastocyst quality when the blastocysts containing cytoplasmic strings in the +CPFH group were investigated (Table 4). Within this group, 85.1% (80/94) of blastocysts were graded AA, 87.5% (7/8) graded BA, 100% (12/12) graded BB, 91.7% (22/24) graded AB, and a 100% live birth rate in those blastocysts graded with a C grade in either the ICM or TE (p = 0.470).
Table 4.
Live birth rate and blastocyst quality (Gardner criteria) of blastocysts containing cytoplasmic strings in the positive clinical pregnancy with a fetal heart group
| Blastocyst grade | Live birth rate in blastocysts with the presence of CS |
|---|---|
| AA | 85.1% (80/94) |
| BA | 87.5% (7/8) |
| BB | 100% (12/12) |
| AB | 91.7% (22/24) |
| Anything containing a C | 100% (6/6) |
| p-value | 0.470 |
Multivariable regression analysis
As there was a significant difference in the developmental stage of transfer between the two groups compared, a multivariable regression analysis was performed to adjust for this potential confounding factor (Table 5). This analysis showed that the presence of CS is still an independent predictive factor of CPFH with an odds ratio of 4.7 (p = 0.007, OR; 4.7, CI 95% 1.5–14.2) even when controlling for the developmental stage of the embryo. Furthermore, as all the hatching blastocysts in this study contained CS, a sensitivity analysis was performed by excluding them, and it was found that the presence of CS was still significantly associated with CPFH with an odds ratio of 3.4 (p = 0.034, OR; 3.4, CI 95% 1.1–10.6).
Table 5.
Multivariable regression analysis for the presence of cytoplasmic strings when adjusted for the developmental stage of the blastocyst at transfer
| Parameter | Odds ratio (95% CI) |
p-value |
|---|---|---|
| Presence of CS |
4.7 (1.5–14.2) |
0.007* |
|
Presence of CS (excluding HGBL) |
3.4 (1.1–10.6) |
0.034* |
CS cytoplasmic strings, HGBL hatching blastocyst
Discussion
This study for the first time has identified a positive association of the presence of CS in human blastocysts with the probability of clinical pregnancy with fetal heart. This finding was confirmed with a multivariable regression analysis, implying a stronger effect size when controlling for the potential confounding factor of the blastocyst development stage. Furthermore, it was identified that blastocysts that resulted in a +CPFH had a higher number of CS present. The CS were also present in those blastocysts that had a +CPFH for a longer duration throughout blastocyst development compared to those blastocysts that resulted in a −CPFH.
In addition to this, the +CPFH blastocyst group had a higher degree of vesicle activity compared to those that resulted in a −CPFH. They contained a higher quantity of vesicles that were seen traveling along the CS between the TE cells and ICM. This was also found to be positively associated with the establishment of a +CPFH. The observation of an increase in both the number of vesicles and the duration of the CS presence is likely to be attributed to the higher number of CS being present in the group of blastocysts that resulted in a +CPFH.
Within the current literature, there are limited findings available on the importance of CS in human blastocysts, with conclusions stating that their role and function are of unknown significance [18, 21]. However, this study is the first to indicate a possible role of both the CS and their vesicles in blastocysts transferred that result in a +CPFH.
The current study found that CS are identified in almost all in vitro–produced human blastocysts, which was similar to that reported in mouse blastocysts [16]. In addition, it was found that there was a stronger association of the presence of CS in blastocysts that produced a +CPFH. Contrary to this finding, one study reported no difference in pregnancy rates when blastocysts were transferred that contained CS compared to those that did not [22]. This finding may be attributed to the larger sample size in the current study or the different time-lapse systems being used in the two studies.
Another study hypothesized that their presence may have a negative effect on embryo development that may be caused by poor culture conditions [19]. However, this was hypothesized based on the author’s unpublished data. It is difficult for us to draw conclusions between the current study and this suggestion as the methodology and design of how this was attained is not available. Additionally, this study was performed several years earlier with media and culture conditions changing drastically since the introduction of commercially available time-lapse systems.
In the current study, we observed a general trend that the CS appear during the early stage of blastulation which is consistent with that reported in another study [19]. Additionally, all hatching blastocysts within the current study were seen to have the presence of CS during the time of their development. It could be argued that this is likely due to blastocoel cavity expansion, thus allowing for visualization without the background noise of the TE cells. However, when the blastocyst developmental stage was controlled for, the presence of CS was still significantly associated with a fetal heart.
One of the strengths of the current study is that CS assessment occurred by a single assessor, therefore limiting inter-observer variation and to ensure maximum consistency in agreement between observations. However, the argument of intra-observer variation may still exist and this area requires further exploration in future studies. Furthermore, the assessor was blind to the outcome of that transfer (whether it resulted in CPFH or not) and this was to avoid systematic bias in the assessment process. Finally, this is the largest study to date assessing CS utilizing currently available systems that allow for higher resolution dynamic images, which ensures proper identification and evaluation of the CS as well as the findings being applicable to current clinical practice.
The current study also has limitations that need to be discussed. The case-control design of the study was chosen to maximize the power of the study in assessing the primary outcome of interest, which was CPFH. However, this means that a potential cohort effect of the CS presence cannot be evaluated, as only blastocysts that were considered of the highest quality from the cohort that were eventually transferred were included. It still remains unknown whether there is an association between the CS and the vesicle numbers, the other blastocysts within the patient’s cohort, poor-quality blastocysts, or if other factors may be involved and this warrants further investigation.
Additionally, the Embryoscope+ allows 11 focal planes for embryo assessment thus the possibility exists of under-diagnosis of the CS. In contrast, standard microscopy has an infinite number of focal planes and it could be argued that the static images obtained via this method may allow for additional focal planes in order to assess the CS. However, the exposure of CS would be limited to the assessor depending on the time the assessment occurred. Our study identified the presence of CS not only at different development stages but also for different durations (ranging from 0.1 to 26.8 h). Therefore, static images selected at key developmental timings to assess CS would prove to be a challenge in order to achieve an accurate representation of their activity.
It is well known that current grading systems of blastocysts are highly subjective and are open to interpretation [27–29]. Although they are detailed and employ a range of morphological parameters, they still suffer from both inter- and intra-observer agreement even from experienced embryologists [28]. It is likely that the assessment of CS would also experience some degree of inter- and intra-observer agreement; however, this remains to be explored in further studies.
It could be argued that the CS are similar to other filaments reported in the literature. However, CS appear to be different from other filaments that arise earlier in the preimplantation stages of embryo development. Perivitelline threads [30] appear at the 2-cell stage and originate from the corona radiata, while the long traversing CS appear at early blastulation and originate from the ICM and TE cells. As these are separated by approximately 60 h in development, it appears these threads and CS are unrelated. Additionally, short cellular projections identified as “short filopodia” were identified protruding from the CS into the blastocoel cavity in mouse blastocysts [16]. These structures were not identified in the current study and similarly were also reported as absent by another study [22]. These findings may be attributed to the magnification limitations of the current time-lapse systems being used, or that they may not be present in human blastocysts.
In conclusion, the current study is the first to suggest that the presence of CS in human blastocysts is associated with the probability of clinical pregnancy with fetal heart. With modern time-lapse incubators allowing for the ultrastructure of the blastocyst to be assessed in multiple focal planes along with morphokinetic analysis, it is logical that new methods of assessing embryos can be expected. The findings of this study suggest that in addition to blastocyst quality and the blastocyst development stage, the presence of CS could be used as an aid as an ultrastructure marker in the identification of blastocysts with the highest potential to produce a clinical pregnancy. However, further studies are needed to confirm these findings and clarify the potential value of CS assessment in modern ART.
Supplementary information
(DOCX 13 kb)
A timelapse video exported from the Embryoscope+ showing the same day 5 blastocyst as seen in Fig. 2a–e. A single cytoplasmic string can be seen traversing the blastocoel cavity between 104.8 - 107 hpi. It connects the ICM and the TE cells. A single vesicle can be seen emerging from the TE cells and as the blastocyst develops and expands, the single vesicle moves along the cytoplasmic string. Another vesicle emerges from the TE cells and these vesicles continue to move along the cytoplasmic string from the TE cells to where they are delivered to the ICM and absorbed. This process takes a total of 2.2 h to occur. (AVI 859 kb)
Appendix
Author contribution
All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Jessica Eastick, Christos Venetis, Simon Cooke, and Michael Chapman. The first draft of the manuscript was written by Jessica Eastick and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Data availability
http://handle.unsw.edu.au/1959.4/resource/collection/resdatac_1105/1
Code availability
Not applicable
Declarations
Ethics approval
This is an observational study. The IVFAustralia Research Ethics Committee has confirmed that no ethical approval is required.
Consent to participate
Not applicable
Consent for publication
Not applicable
Conflict of interest
The authors declare no competing interests.
Footnotes
Key message
For the first time, this case-control study has shown that the presence of cytoplasmic strings in human blastocysts is associated with clinical pregnancy with fetal heart following embryo transfer.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Gardner DK, Balaban B. Assessment of human embryo development using morphological criteria in an era of time-lapse, algorithms and ‘OMICS’: is looking good still important? MHR: Basic Sci Reprod Med. 2016;22(10):704–718. doi: 10.1093/molehr/gaw057. [DOI] [PubMed] [Google Scholar]
- 2.Pribenszky C, Mátyás S, Kovács P, Losonczi E, Zádori J, Vajta G. Pregnancy achieved by transfer of a single blastocyst selected by time-lapse monitoring. Reprod BioMed Online. 2010;21(4):533–536. doi: 10.1016/j.rbmo.2010.04.015. [DOI] [PubMed] [Google Scholar]
- 3.Fisch JD, Rodriguez H, Ross R, Overby G, Sher G. The Graduated Embryo Score (GES) predicts blastocyst formation and pregnancy rate from cleavage-stage embryos*. Hum Reprod. 2001;16(9):1970–1975. doi: 10.1093/humrep/16.9.1970. [DOI] [PubMed] [Google Scholar]
- 4.Gabrielsen A, Andersen AG, Andersen AN, Petersen K, Lindenberg S, Ziebe S. Embryo morphology or cleavage stage: how to select the best embryos for transfer after in-vitro fertilization. Hum Reprod. 1997;12(7):1545–1549. doi: 10.1093/humrep/12.7.1545. [DOI] [PubMed] [Google Scholar]
- 5.Gardner DKS, William B. Culture and transfer of human blastocysts. Curr Opin Obstet Gynaecol. 1999;11(3):307–311. doi: 10.1097/00001703-199906000-00013. [DOI] [PubMed] [Google Scholar]
- 6.Steer CV, Mills CL, Tan SL, Campbell S, Edwards RG. SHORT: the cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and embryo transfer programme. Hum Reprod. 1992;7(1):117–119. doi: 10.1093/oxfordjournals.humrep.a137542. [DOI] [PubMed] [Google Scholar]
- 7.Ciray HN, Campbell A, Agerholm IE, Aguilar J, Chamayou S, Esbert M, Sayed S, Time-Lapse User Group Proposed guidelines on the nomenclature and annotation of dynamic human embryo monitoring by a time-lapse user group. Hum Reprod. 2014;29(12):2650–2660. doi: 10.1093/humrep/deu278. [DOI] [PubMed] [Google Scholar]
- 8.Herrero J, Meseguer M. Selection of high potential embryos using time-lapse imaging: the era of morphokinetics. Fertil Steril. 2013;99(4):1030–1034. doi: 10.1016/j.fertnstert.2013.01.089. [DOI] [PubMed] [Google Scholar]
- 9.Kirkegaard K, Kesmodel US, Hindkjær JJ, Ingerslev HJ. Time-lapse parameters as predictors of blastocyst development and pregnancy outcome in embryos from good prognosis patients: a prospective cohort study. Hum Reprod. 2013;28(10):2643–2651. doi: 10.1093/humrep/det300. [DOI] [PubMed] [Google Scholar]
- 10.Basile N, Vime P, Florensa M, Aparicio Ruiz B, Garcia Velasco JA, Remohi J, et al. The use of morphokinetics as a predictor of implantation: a multicentric study to define and validate an algorithm for embryo selection. Hum Reprod. 2015;30(2):276–283. doi: 10.1093/humrep/deu331. [DOI] [PubMed] [Google Scholar]
- 11.Petersen BM, Boel M, Montag M, Gardner DK. Development of a generally applicable morphokinetic algorithm capable of predicting the implantation potential of embryos transferred on Day 3. Hum Reprod. 2016;31:2231–2244. doi: 10.1093/humrep/dew188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chen AA, Tan L, Suraj V, Reijo Pera R, Shen S. Biomarkers identified with time-lapse imaging: discovery, validation, and practical application. Fertil Steril. 2013;99(4):1035–1043. doi: 10.1016/j.fertnstert.2013.01.143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Gardner DK, Meseguer M, Rubio C, Treff NR. Diagnosis of human preimplantation embryo viability. Hum Reprod Update. 2015;21(6):727–747. doi: 10.1093/humupd/dmu064. [DOI] [PubMed] [Google Scholar]
- 14.Meseguer M, Rubio I, Cruz M, Basile N, Marcos J, Requena A. Embryo incubation and selection in a time-lapse monitoring system improves pregnancy outcome compared with a standard incubator: a retrospective cohort study. Fertil Steril. 2012;98(6):1481–9.e10. doi: 10.1016/j.fertnstert.2012.08.016. [DOI] [PubMed] [Google Scholar]
- 15.Tran D, Cooke S, Illingworth PJ, Gardner DK. Deep learning as a predictive tool for fetal heart pregnancy following time-lapse incubation and blastocyst transfer. Hum Reprod. 2019;34:1011–1018. doi: 10.1093/humrep/dez064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Salas-Vidal E, Lomeli H. Imaging filopodia dynamics in the mouse blastocyst. Dev Biol. 2004;265(1):75–89. doi: 10.1016/j.ydbio.2003.09.012. [DOI] [PubMed] [Google Scholar]
- 17.Fleming TP, Warren PD, Chisholm JC, Johnson MH. Trophectodermal processes regulate the expression of totipotency within the inner cell mass of the mouse expanding blastocyst. J Embryol Exp Morpholog. 1984;84(1):63. [PubMed] [Google Scholar]
- 18.Alpha Scientists in Reproductive M, Embryology ESIG Istanbul consensus workshop on embryo assessment: proceedings of an expert meeting. Reprod BioMed Online. 2011;22(6):632–646. doi: 10.1016/j.rbmo.2011.02.001. [DOI] [PubMed] [Google Scholar]
- 19.Scott L. Oocyte and embryo polarity. Semin Reprod Med. 2000;18(2):171–183. doi: 10.1055/s-2000-12556. [DOI] [PubMed] [Google Scholar]
- 20.Ducibella T, Albertini DF, Anderson E, Biggers JD. The preimplantation mammalian embryo: characterization of intercellular junctions and their appearance during development. Dev Biol. 1975;45(2):231–250. doi: 10.1016/0012-1606(75)90063-9. [DOI] [PubMed] [Google Scholar]
- 21.Hardarson TVLL, Jones G. The blastocyst. Hum Reprod. 2012;27(S1):i72–i91. doi: 10.1093/humrep/des230. [DOI] [PubMed] [Google Scholar]
- 22.Ebner T, Sesli Ö, Kresic S, Enengl S, Stoiber B, Reiter E, Oppelt P, Mayer RB, Shebl O. Time-lapse imaging of cytoplasmic strings at the blastocyst stage suggests their association with spontaneous blastocoel collapse. Reprod BioMed Online. 2020;40(2):191–199. doi: 10.1016/j.rbmo.2019.11.004. [DOI] [PubMed] [Google Scholar]
- 23.Fierro-González JC, White MD, Silva JC, Plachta N. Cadherin-dependent filopodia control preimplantation embryo compaction. Nat Cell Biol. 2013;15:1424. doi: 10.1038/ncb2875. [DOI] [PubMed] [Google Scholar]
- 24.Ebner TMM, Sommergruber M, Gaiswinkler U, Shebl O, Jesacher K, Tewas G. Occurrence and developmental consequences of vacuoles throughout preimplantation development. Fertil Steril. 2005;83(6):0015–0282. doi: 10.1016/j.fertnstert.2005.02.009. [DOI] [PubMed] [Google Scholar]
- 25.Eastick J, Venetis C, Cooke S, Storr A, Susetio D, Chapman M. Is early embryo development as observed by time-lapse microscopy dependent on whether fresh or frozen sperm was used for ICSI? A cohort study. J Assist Reprod Genet. 2017;34(6):733–740. doi: 10.1007/s10815-017-0928-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gardner RL. The early blastocyst is bilaterally symmetrical and its axis of symmetry is aligned with the animal-vegetal axis of the zygote in the mouse. Development. 1997;124(2):289–301. doi: 10.1242/dev.124.2.289. [DOI] [PubMed] [Google Scholar]
- 27.Ahlström A, Westin C, Reismer E, Wikland M, Hardarson T. Trophectoderm morphology: an important parameter for predicting live birth after single blastocyst transfer. Hum Reprod. 2011;26(12):3289–3296. doi: 10.1093/humrep/der325. [DOI] [PubMed] [Google Scholar]
- 28.Storr A, Venetis CA, Cooke S, Kilani S, Ledger W. Inter-observer and intra-observer agreement between embryologists during selection of a single Day 5 embryo for transfer: a multicenter study. Hum Reprod. 2017;32(2):307–314. doi: 10.1093/humrep/dew330. [DOI] [PubMed] [Google Scholar]
- 29.Arce J-C, Ziebe S, Lundin K, Janssens R, Helmgaard L, Sørensen P. Interobserver agreement and intraobserver reproducibility of embryo quality assessments. Hum Reprod. 2006;21(8):2141–2148. doi: 10.1093/humrep/del106. [DOI] [PubMed] [Google Scholar]
- 30.Derrick R, Hickman C, Oliana O, Wilkinson T, Gwinnett D, Whyte LB, Carby A, Lavery S. Perivitelline threads associated with fragments in human cleavage stage embryos observed through time-lapse microscopy. Reprod BioMed Online. 2017;35(6):640–645. doi: 10.1016/j.rbmo.2017.08.026. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(DOCX 13 kb)
A timelapse video exported from the Embryoscope+ showing the same day 5 blastocyst as seen in Fig. 2a–e. A single cytoplasmic string can be seen traversing the blastocoel cavity between 104.8 - 107 hpi. It connects the ICM and the TE cells. A single vesicle can be seen emerging from the TE cells and as the blastocyst develops and expands, the single vesicle moves along the cytoplasmic string. Another vesicle emerges from the TE cells and these vesicles continue to move along the cytoplasmic string from the TE cells to where they are delivered to the ICM and absorbed. This process takes a total of 2.2 h to occur. (AVI 859 kb)
Data Availability Statement
http://handle.unsw.edu.au/1959.4/resource/collection/resdatac_1105/1
Not applicable