Placental architectural characteristics following laser ablation within monochorionic twins complicated by twin–twin transfusion syndrome

Introduction: Twin–twin transfusion syndrome (TTTS) complicates approximately 10%–15% of all monochorionic twin pregnancies. The aim of this review was to evalu - ate the placental architectural characteristics within TTTS twins following laser and elucidate their impact on fetal outcomes and operative success. Material and Methods: Five databases were searched from inception to August 2023. Studies detailing post-delivery placental analysis within TTTS twins post-laser were included. Studies were categorized into two main groups: (1) residual anastomoses following laser and (2) abnormal cord insertion: either velamentous and/or marginal or proximate. The primary outcome was to determine the proportion of TTTS placen - tas with residual anastomoses and abnormal cord insertions post-laser. Secondary outcomes included assessing residual anastomoses on post-laser fetal outcomes and assessing the relationship between abnormal cord insertion and TTTS development. Study bias was critiqued using the


| INTRODUC TI ON
[4] A unidirectional imbalance of blood flow leads to characteristic changes in polyhydramnios in one twin (the recipient) and oligohydramnios in the other (the donor), which is pathognomic of TTTS.
The severity of TTTS has been classified using the Quintero staging, with stages 2 and above recommended to undergo laser surgery. 5,6toscopic laser is an intrauterine treatment modality that intends to coagulate the intertwin vascular anastomoses. 7,8The number of residual anastomoses post-laser should be kept to a minimum, to reduce the risk of recurrent TTTS and post-laser twin anemia polycythemia sequence (TAPS). 9Clinicians have aimed to achieve this by adapting the surgical technique from the selective to the Solomon method, ensuring to ablate the entire vascular equator from each placental edge.1][12][13][14][15] There have been no previous systematic reviews describing or quantifying these residual anastomoses, or investigating factors that may influence numbers of residual anastomoses.The impact that residual anastomoses have on fetal outcomes also requires further investigation.
Abnormal placental cord insertion sites are more common within twin pregnancies compared to singletons. 168][19][20] It is suspected that abnormal cord insertion may predispose to the development of TTTS within MC twins, although current published evidence is conflicting. 19,21,22There is a scarcity of available literature delineating the proportions of different types of abnormal cord insertion including how this influences the TTTS development for MC twins specifically requiring laser.
The aim of this systematic review and meta-analysis is to evaluate the placental architectural characteristics within TTTS MC twins following laser ablation and elucidate their impact on fetal outcomes and operative success.

| MATERIAL AND ME THODS
This review was prospectively registered with PROSPERO (CRD42023476875) and results have been reported in accordance with the preferred reporting of systematic reviews and metaanalysis (PRISMA) guidelines. 23MEDLINE, EMBASE, Cochrane, Web of Science, and PubMed databases were searched from inception to August 2023.The MEDLINE/EMBASE search is detailed in Table S1.

Conclusions:
To the best of our knowledge, this is the first review to conjointly explore outcomes of residual anastomoses and abnormal cord insertions within TTTS twins following laser.A large prospective study is necessitated to assess the relationship between abnormal cord insertion and residual anastomoses development post-laser.

Key message
The aim of this systematic review and meta-analysis was to evaluate the placental architectural characteristics within monochorionic twins complicated by twin-twin transfusion syndrome (TTTS) following laser ablation and elucidate their impact on fetal outcomes and operative success.
To the best of our knowledge, this is the first review to conjointly explore outcomes of residual anastomoses and abnormal cord insertions within this twin cohort.This review enables clinicians the ability to clearly inform patients of potential post-laser risks during pre-and post-operative counselling.
Rayyan software was acquired to manage title, abstract, and full manuscript screening, including subsequent data extraction.
All abstracts and full texts were reviewed by two authors (JH and NE) and conflicts were discussed with a third author (LG).Reference lists were additionally screened to identify additional gray literature citations eligible for inclusion.Studies specifically detailing postdelivery placental analysis within MC twins complicated by TTTS that underwent intrauterine laser ablation were included.Pathology placental analysis was required to be detailed within the study methodological section for inclusion to ensure true estimates of placental anatomical characteristics could be established.In the uncommon event that an interventional group included patients undergoing various interventions (laser ablation or amnio drainage), then studies were deemed eligible for inclusion if a minimum of 60% of the cohort had undergone laser ablation.This was to capture a larger cohort of included studies while ensuring the results following laser ablation were not excessively diluted by alternative interventions.
Later subgroup analysis within the results excluded these studies to determine their influence on results.Studies examining singleton, dichorionic, monoamniotic, and higher-order pregnancies were excluded.Included studies were categorized into two main groups: (1)   studies identifying the presence of residual vascular anastomoses and (2) studies describing abnormal cord insertion: either velamentous and/or marginal or proximate.Several included studies encompassed both categories.

| Data analysis
Outcomes analyzed were individualized based on each group of included studies and were predetermined by an initial scoping review and prior to data extraction.For studies examining residual vascular anastomoses, the primary outcome was to determine the proportion of placentas with residual anastomoses present following laser ablation.Further analysis determines the mean number of residual anastomoses, subtypes of residual anastomoses, and their diameters.Secondary outcomes included the impact of placental position during laser on the number of placentas with residual anastomoses.
Further analyses assessed the impact of residual anastomoses on single and/or double intrauterine fetal death (IUFD), neonatal death (NND), recurrent TTTS, and post-laser TAPS.
For studies examining abnormal cord insertion, the primary outcome was to determine the proportion of different types of abnormal cord insertion within the TTTS cohort undergoing laser ablation.
Secondary outcomes included the influence of abnormal cord insertion on the development of TTTS versus other MC cohorts and the impact of abnormal cord insertion on IUFD and NND.Outcomes were considered eligible for meta-analysis if three or more studies were suitable for inclusion in the analysis.This was intended to improve the reliability of the effect estimates, given the methodological heterogeneity within the available literature. 24edetermined subgroup analysis was individualized for each group.For studies examining number of placentas with residual anastomoses, this included: dual survivors and single/double IUFD, Quintero staging, and laser ablation technique.For studies examining abnormal cord insertion, this included: donor/recipient cord insertion and studies with or without the total cohort undergoing laser.
Review Manager 5.4 (RevMan) was used to meta-analyze pairwise data. 25Dichotomous outcome data were assessed with 2 × 2 tables constructed to calculate odds ratios (OR) and 95% confidence intervals (CI).Mantel-Haenszel method random-effects models were utilized for meta-analysis.Summary OR were configured with forest plots, allowing visual inspection and quantitative assessment for heterogeneity using I 2 . 26Zero-cell adjustments to 0.5 were used as required.Studies containing raw data individually reporting on abnormal cord insertion and residual anastomoses were combined to form a mean and standard deviation (SD).Continuous variables, such as residual anastomoses and vascular subtypes, reported with an overall median and range were translated to proxy mean and SD using methodological equations described by Hozo et al and Wan et al. 27,28 Calculated means and SD for relevant groups from each study were combined together for an overall mean and SD using the Cochrane method for combing means. 29oportional data and 95% CI were calculated using R version 4.3.0. 30Random effects models were used to meta-analyze estimates of the overall proportional data due to the high likelihood of statistical heterogeneity.Heterogeneity arising from methodological diversity between studies, particularly due to nuances in placental analysis techniques and reporting, as well as specific inclusion/ exclusion study, led to a variation in observed intervention effects between studies.The inverse variance method with logit transformation of the rate was used.Summary proportions and CI were configured with forest plots, including quantitative assessment for heterogeneity using I 2 .I 2 values below 25% were considered low, approximately 50% considered moderate, and above 75% considered high levels of heterogeneity. 26Subgroup proportional data analysis was configured using visual forest plots, using random effects testing to assess for subgroup statistical differences with p values.

| Risk of bias and trustworthiness
Risk of bias in observational studies was assessed by two reviewers (JH and NE) using the most relevant and recent critical appraisal checklists from the Joanna Briggs Institute. 31The Cochrane risk of bias tool (RoB2) was used to assess bias in interventional studies. 32y discrepancies with bias assessment were arbitrated with a third author (LG).
The overall proportions of each anastomosis subtype were de- and diameters for each study are presented in Table 1.

| Secondary outcomes
The OR for all presented outcomes from each included study are demonstrated in Tables 2 and 3 3).

| Subgroup analysis
Several studies only included placental analyses from dual survivors, due to the potential for placental maceration following single or double IUFD, which could impede assessment of residual anastomoses.For studies explicitly commenting on the inclusion of either dual survivors, or both survivors and IUFD (single or double), while also specifying the rates of residual anastomoses for each group, the proportional data were calculated.The proportion of placentas with residual anastomoses within dual survivors was 16% (95% CI, 0.08-0.31;I 2 = 89%, 11 studies) versus 36% (95% CI, 0.08-0.78;I 2 = 68%, 5 studies) for studies with single or double IUFD.Analysis of subgroup differences between survivors and IUFD was not significant (p = 0.30) (Figure 4).Subgroup difference analysis specifically for AV anastomoses could not be performed as included studies were too few.
Three studies detailed the number of residual anastomoses for each Quintero stage.Quintero stages 1 and 2 were combined and compared against combined stages 3 and 4 to enable meta-analysis.

| Abnormal cord insertion
Thirteen studies were eligible for inclusion in the analysis of abnormal cord insertion within TTTS MC twins undergoing laser ablation.

| Secondary outcome
The OR for all presented outcomes from each included study are demonstrated in  Results were non-significant (OR, 1.04 (95% CI, 0.84-1.29);I 2= 0%).
Studies comparing only velamentous or marginal cord insertion between these groups, results were additionally non-significant (OR, Note: Only the first author is given for each study.Filled color means data entry is not applicable.

F I G U R E 4
Individual study proportion rates for overall residual anastomoses following laser ablation within cases of intrauterine fetal demise (single or double) and dual survivors for twin-twin transfusion syndrome monochorionic twins.
the included studies.Additionally, there were insufficient data within the included studies to compare abnormal cord insertion directly against the proportion of residual vascular anastomoses following laser ablation.I 2 = 0%, three studies) respectively.Individual study results can be visualized in Table 4.
TTTS could not be performed across all cord insertion groups following the removal of studies with <100% of the cohort undergoing laser ablation.

| Risk of bias
The full risk of bias assessments for all included studies are detailed in Table S2, within Supporting Information.Twelve studies were graded as low risk and one a moderate risk of bias.The included RCT was low risk across all domains of the Cochrane RoB2 tool.The most common reasons for a reduction in score were not identifying confounding factors and loss of follow-up not correctly recorded.

| Additional placental outcomes
In addition to the above outcomes that have been discussed, four included studies additionally detailed post-natal placental sharing and placental weight data. 4,38,41,49 Winden et al. 49 additionally compared fetal-placental weight ratio between TTTS MC twins and TTTS MC twins complicated with sIUGR and found that sIUGR was not an associating factor (p = 0.83).
Wang et al. 4 additionally noted that TTTS undergoing laser is not an influential factor on placental territory discordance ratio when compared to non-TTTS MC twins (p = 0.236).

| DISCUSS ION
This review provides quantitative findings on post-laser placental characteristics for TTTS MC twins.We demonstrate that nearly a quarter of all lasered placentas contained residual anastomoses.
Residual anastomoses are significantly associated with the development of post-laser TAPS, recurrent TTTS, IUFD, and NND.Placental positioning and Quintero staging are non-significant.
Abnormal cord insertion was detected in approximately half of all TTTS MC twins that underwent laser.Abnormal cord insertion is more frequently seen within donor twins.Abnormal cord insertion is not significantly associated with the development of TTTS requiring laser.
Our review represents the first study to amalgamate data on residual anastomoses post-laser for MC twins complicated by TTTS, including associated fetal outcomes.Residual anastomoses have been widely categorized into three main types: AA, AV, and VV.AA and VV anastomoses are more commonly superficial and classically bidirectional, yielding a protective effect for potential MC complications. 16,57AV anastomoses are obligate unidirectional and typically more deeply located within the placental tissue. 57Consequently, adequate visualization during fetoscopy and complete ablation can be troublesome.As a result, residual anastomoses post-laser can occur, which when uncompensated by the bilateral protective nature of AA/VV anastomoses, can increase the risk of developing post-laser complications and adverse perinatal outcomes. 16,42,58,59 our review, approximately 24% of placentas contained residual anastomoses, of which 75% contained AV anastomoses.Such a large proportion of residual unidirectional anastomoses, in collaboration with results demonstrating a significant increase in post-laser MC complications, reveals that our findings align closely with prior pathophysiological explanations within the literature. 3,16,57,58Importantly, In recent years, the surgical technique for ablating anastomoses has adapted from the selective to Solomon method, aiming to coagulate the entire vascular equator and reduce numbers of residual anastomoses. 9,60Prior literature has demonstrated a significant reduction in certain post-laser MC complications and adverse perinatal outcomes with the Solomon technique. 7,61,62]43 These findings are consistent with our presented results.However, our findings provide a unique exploration into the relationship between twins specifically requiring laser and abnormal cord insertion.
Interestingly, the most recent Royal College of Obstetricians and Gynaecologists guideline for MC twins notes abnormal cord insertion is often associated with TTTS. 63[66][67] While abnormal cord insertion may not be a critical determinant for the evolution of TTTS, including twins requiring laser treatment, we are unable to determine their association with residual anastomoses.[70] This given that those fetuses that were demised are likely to have had residual anastomoses.However, our subgroup analyses comparing survivors versus fetal demise aimed to rectify this concern.
Given the nature of certain outcomes presented within small numbers within individual studies, there was potential for publication bias.
Consequently, this limits the reliability of certain forest plots examining adverse fetal outcomes (left side limited to 0 value for outcome rates and wide confidence intervals).Furthermore, many included studies originated from high-income countries, where laser ablation is routinely available within tertiary centers, with rigorous TTTS pathways, including an established pathology department for placental analysis.
Surgical success and subsequent fetal outcomes may, however, differ for middle-/lower-income countries whereby these services may not be as established due to lack of healthcare resources.This potentially introduces publication bias within this review.Consequently, the results presented here may not achieve full worldwide generalizability.
Low-/middle-income countries may need to conduct their own local prospective studies to facilitate patients with region-specific outcome information.
In addition, included studies demonstrated a variation in their reporting techniques for each group examined, whether that be presenting the raw data, the mean and SD, or the median and range.
Consequently, the values we present for the overall anastomoses mean, including vascular subtypes and diameters, may slightly differ from the true values if all raw data were provided for calculation.
However, despite these restrictions, this paper represents the most comprehensive and up-to-date review of placental characteristics for post-laser TTTS MC twins.

CO N FLI C T O F I NTE R E S T S TATE M E NT
None.
twin groups within the included papers.Comparison groups included MC twins uncomplicated by TTTS, MC twins complicated by TTTS not undergoing laser, and MC twins complicated with both TTTS and selective intrauterine growth restriction (sIUGR).Studies comparing abnormal cord insertion in MC twins with TTTS undergoing laser ablation versus MC twins uncomplicated by TTTS were the only group containing the required number of studies for meta-analysis.Four studies compared the rates of combined abnormal cord insertion (velamentous and marginal) within MC twins with TTTS undergoing laser ablation versus MC twins uncomplicated by TTTS.
three studies), respectively.Subgroup proportional analysis of abnormal cord insertion for studies with the entire cohort undergoing laser versus studies containing <100% of the cohort undergoing laser ablation was not statistically different for combined (p = 0.93), velamentous (p = 0.33), or marginal (p = 0.07), respectively (Figure6).Removal of studies with less than 100% of the cohort undergoing laser continued to demonstrate significantly more combined and F I G U R E 5 Individual study proportion rates for abnormal cord insertion (combined, velamentous, and marginal) within twin-twin transfusion monochorionic twins following laser ablation.velamentous cord insertions within the donor twin compared to the recipient twin (OR, 5.09 (95% CI, 2.78-9.32);I 2= 18%, three studies) and (OR, 5.23 [95% CI 3.29-8.31];I 2= 42%, four studies), respectively.Marginal cord insertion within the donor versus recipient twin could not be meta-analyzed.Meta-analysis of MC twins complicated by TTTS undergoing laser ablation versus MC twins uncomplicated by TA B L E 4 Rates of abnormal cord insertion for twin-twin transfusion (TTTS) twins undergoing laser ablation versus non-TTTS MC twins and rates of donor versus recipient TTTS twins undergoing laser.
review demonstrates original insight by quantifying the relationship between residual anastomoses and post-laser MC complications.This review enables clinicians the ability to clearly inform patients of potential post-laser risks during pre-and post-operative counseling, such that if a surgeon anticipates residual anastomoses are likely present post-laser, then quantifiable risks presented within this review can help facilitate discussions regarding possible fetal outcomes.This review presents a strong association with recurrent TTTS and post-laser TAPS when residual anastomoses following laser ablation are present.Clinicians may consider increased awareness and surveillance for these potential fetal complications during the post-operative period to ensure earlier detection and subsequent antenatal management.
can ultimately lead to difficulty in achieving complete intraoperative ablation of the anastomoses.Determining the association between abnormal cord insertion and residual anastomoses may allow clinicians to risk stratify women pre-operatively and employ patient-specific counseling advice for continued post-operative management.A large prospective multicenter study is required to address this question.Additionally, Bonanni et al have recently F I G U R E 6 Individual study proportion rates for abnormal cord insertion (combined, velamentous, and marginal) within studies containing all participants undergoing laser ablation and studies containing fewer than 100% participants undergoing laser ablation.demonstrateda significant association with abnormal cord insertion and adverse fetal outcomes.This highlights the importance of improving the detection of these abnormal placental anatomical features pre-operatively, enabling timely intervention and improved operative planning.70While it is important to demonstrate evidence regarding postlaser TTTS placental architecture, it is equally important to consider alternative strategies to detect these placental characteristics preoperatively.An emerging interest within fetal medicine is the use of pre-operative MRI.71 Torrents-Barrena et al.72 proposed the first TTTS simulator for clinical use, allowing clinicians to pre-operatively visualize umbilical cord insertions and placental anastomoses, allowing for detailed TTTS surgical planning.Lewi et al.73 and Luks et al.74  additionally evaluated the use of pre-operative MRI, demonstrating an 89% accuracy for predicting twin placental volume distribution and enabling virtual simulation of the intertwin membrane location and placenta, respectively.Consequently, a detailed pre-operative analysis of the intrauterine environment ensures clinicians are fully equipped for surgery, ensuring correct location of the umbilical cords and vascular equator, and maximizing complete intertwin anastomotic ablation.To the best of our knowledge, this is the first systematic review that has explored outcomes of residual anastomoses and abnormal cord insertion within post-laser TTTS twins.An initial scoping search, a thorough formal literature search, and a robust methodological strategy, including a deep assessment of outcomes, are the main strengths of this review.Studies were additionally validated by two authors including consultation with a third author if any conflicts occurred.This intended to improve the accuracy of included studies and the objectivity of the findings.Many included studies were retrospective, non-randomized, containing small sample sizes while demonstrating a wide heterogeneity in methodological quality regarding placental analysis and participant inclusion criteria.These represent the main weaknesses of this review.Many studies assessing residual anastomoses only included dual survivors, potentially increasing the risk of selection bias.In turn, this may have underestimated the risks associated with residual anastomoses.Additionally, many studies automatically excluded single-or dual-fetal demise due to the potential risk of placental maceration.This may introduce further selection and publication bias, Residual anastomoses were detected within approximately a quarter of all post-laser TTTS twins.Residual anastomoses are significantly associated with an increased rate of recurrent TTTS, post-laser TAPS, IUFD, and NND.Abnormal cord insertion was detected in almost half of all post-laser TTTS twins.Abnormal cord insertion is not an influencing factor for TTTS development within twins requiring laser ablation.A large prospective study is necessitated to assess the relationship between abnormal cord insertion and residual anastomoses post-laser.AUTH O R CO NTR I B UTI O N SJack Hamer conceived the idea for the review, performed the literature search, screened texts, performed the risk of bias assessment, performed the statistical analysis, and wrote the draft and final manuscript.Nashwa Eltaweel conceived the idea for the review, screened texts as second reviewer, performed the risk of bias assessment, reviewed/edited the draft manuscript, and approved/ appraised the final submission.Rebecca Man and Matilde Rogerson performed the statistical analysis for the review, reviewed/edited the draft manuscript, and approved/appraised the final submission.Victoria Hodgetts Morton, Katie Morris, and Tamas Marton reviewed and edited the draft manuscripts and approved/appraised the final submission.Leo Gurney conceived the idea for the review, acted third reviewer to resolve literature inclusion conflicts, and reviewed/edited the final manuscript.
Individual study proportion rates for each residual anastomosis subtype following laser ablation within twin-twin transfusion monochorionic twins.velamentous, marginal, and proximate cord insertions.One study was an RCT, one was a prospective case series, two studies were retrospective case series, two were prospective cohort studies, and seven were retrospective cohort studies.The characteristics are included in 4,19,20,34,35,38,40,43,[46][47][48][49][50]Abnormal cord insertions encompassedF I G U R E 3Eight studies specifically explored the proportion of combined velamentous and marginal cord insertion in either one or both TTTS MC twins that underwent laser.Overall combined abnormal cord insertion (velamentous and marginal) was reported at a rate of 49% ((95% CI, 0.39-0.59);I 2 = 89%) for one or both TTTS twins upon placental examination following laser.Overall velamentous insertion was

Table 4 .
The rates of abnormal cord insertion within TTTS MC twins undergoing laser ablation were compared to several other TA B L E 2 Studies examining residual anastomoses following laser ablation for twin-twin transfusion monochorionic twins on the impact of different fetal outcomes.Paper Events

out of total participants with residual anastomoses (n/N) Events out of total participants without residual anastomoses (n/N) OR effect estimate (95% CI) I 2
Rate of intrauterine fetal demise (single or dual) Note: Only the first author is given for each study.Filled color means data entry is not applicable.Abbreviations: CI, confidence interval; OR, odds ratio; TAPS, twin anemia polycythemia sequence; TTTS, twin-twin transfusion syndrome.

of residual anastomoses with anterior placental position out of total cases (n/N) Cases of residual anastomoses with posterior placental position out of total cases (n/N) OR effect estimate (95% CI) I 2
1.11 (95% CI, 0.88-1.39);I2=0%)(fourstudies) and (OR, 0.96 (95% CI, 0.77-1.19);I2=20%)(threestudies), respectively.Results for each individual study can be visualized in Table4.Rates of IUFD and NND with respect to abnormal cord insertion could not be meta-analyzed due to insufficient amounts of data within TA B L E 3 Studies examining placental position and Quintero staging for twin-twin transfusion monochorionic twins and the influence on residual anastomoses post-laser.Paper Cases

Paper Events out of total TTTS MC laser participants (n/N) Events out of total non-TTTS MC participants (n/N) OR effect estimate (95% CI) I 2
Rate of combined abnormal cord insertion (velamentous and marginal) 38alek et al.41and Van Winden et al.49commented on the mean placental donor shares of 45.3% (10-90) and 40.0%(20-70), respectively.Van Winden et al.49noted that 21.9% of TTTS MC laser twins had a donor placental share <30%, whereas Favre et al.38noted 35% of donors had a placental