統合分析報告性別手術後後悔率為1%，但納入研究追蹤時間短且流失嚴重

性別肯定手術後的後悔：盛行率的系統性回顧與統合分析

Authors: Valeria P. Bustos, Samyd S. Bustos, Andres Mascaro, Gabriel Del Corral, Antonio J. Forte, Pedro Ciudad, Esther A. Kim, Howard N. Langstein, Oscar J. Manrique
Journal: Plastic and Reconstructive Surgery Global Open
Year: 2021
Source: https://doi.org/10.1097/GOX.0000000000003477

Summary

This review combined 27 studies of nearly 8,000 people who had gender surgery and estimated that about 1% regretted it. The authors concluded regret is extremely rare and credited better screening of patients. The 1% figure is now widely quoted. But the authors themselves say the evidence underneath is weak: most studies were rated poor or fair, almost none used tested ways of measuring regret, and results varied a lot between studies. Critics point out that people were often followed for only one or two years, while regret typically shows up after about eight years. Many people were lost track of (36% in the largest study), and some numbers were copied wrongly. Most people studied were adults treated under older, stricter rules. The 1% is best seen as a minimum drawn from incomplete data, not a reliable regret rate for today's patients.

Key findings

回顧納入 27 項研究，涵蓋來自 14 個國家的 7,928 名手術患者；67% 接受男變女手術，33% 接受女變男手術，只有一名患者被記錄為非二元性別。
合併計算，約 1% 的患者對手術感到後悔（可能介於不到 1% 到 2% 之間），但各研究間的結果差異很大；總共有 77 人回報後悔。
後悔的比例在女變男手術和乳房切除後不到 1%，男變女手術後為 1%，建造陰道的手術後為 2%。
最常見的後悔原因是在新性別角色中難以被家人和社會接受，其次是手術或身體的結果不佳。
幾乎所有研究都使用自製的問題或訪談，多數被評為品質差或普通，追蹤時間從不到 1 年到 9 年不等。
有跡象顯示，後悔率低的研究更可能被發表；針對這點調整後，估計值從 1% 降至 0.5%。
2022 年的更正在作者重新執行回顧後修正了表格和圖表，已發表的信函（Expósito-Campos 和 D'Angelo 2021；Bewley 2022）則對其方法和結論提出質疑。

Abstract

Background: There is an unknown percentage of transgender and gender non-confirming individuals who undergo gender-affirmation surgeries (GAS) that experiences regret. Regret could lead to physical and mental morbidity and questions the appropriateness of these procedures in selected patients. The aim of this study was to evaluate the prevalence of regret in transgender individuals who underwent GAS and evaluate associated factors.

Methods: A systematic review of several databases was conducted. Random-effects meta-analysis, meta-regression, and subgroup and sensitivity analyses were performed.

Results: A total of 27 studies, pooling 7928 transgender patients who underwent any type of GAS, were included. The pooled prevalence of regret after GAS was 1% (95% CI <1%-2%). Overall, 33% underwent transmasculine procedures and 67% transfemenine procedures. The prevalence of regret among patients undergoing transmasculine and transfemenine surgeries was <1% (IC <1%-<1%) and 1% (CI <1%-2%), respectively. A total of 77 patients regretted having had GAS. Twenty-eight had minor and 34 had major regret based on Pfäfflin's regret classification. The majority had clear regret based on Kuiper and Cohen-Kettenis classification.

Conclusions: Based on this review, there is an extremely low prevalence of regret in transgender patients after GAS. We believe this study corroborates the improvements made in regard to selection criteria for GAS. However, there is high subjectivity in the assessment of regret and lack of standardized questionnaires, which highlight the importance of developing validated questionnaires in this population.

Conclusion

Based on this review, there is an extremely low prevalence of regret in transgender patients after GAS. We believe this study corroborates the improvements made in regard to selection criteria for GAS. However, there is high subjectivity in the assessment of regret and lack of standardized questionnaires, which highlight the importance of developing validated questionnaires in this population.

Methodological limitations

People not followed for long enough: Follow-up ranged from under 1 year to 9 years, and many studies followed people for only 1–2 years. Regret has been estimated to appear on average after 8 to 8.5 years. Cohn (Archives of Sexual Behavior, 2023) found no included study that kept track of over 85% of people for at least 8 years each.
Many people lost track of: The largest study (Wiepjes et al. 2018), about half of all people in the review, lost track of 36%. Even the five better-quality studies (174 people, 3% of the total) lost 28% to over 40%, including people who had died (Expósito-Campos & D'Angelo, PRS Global Open 2021).
Weak underlying studies: Most studies were rated "poor" or "fair" quality. Critics count 23 of 27 as at moderate-to-high risk of giving misleading results. None had a comparison group, and many just looked back at old medical files.
Regret measured in different, untested ways: Almost all studies made up their own questions, and each defined regret differently. The authors admit this, and add that patients may hide regret from the surgeons or clinics that treated them.
Number errors and a correction: Critics said the number of people counted from Wiepjes et al. was inflated from 2,627 to 4,863, which makes regret look rarer. The authors defended their figure, but a 2022 correction fixed the tables and figures after the review was re-run.
Questionable choice of studies: Letter writers said well-known relevant studies were missed. One included study asked about regret over the surgical technique chosen, not about having surgery at all.
No plan registered in advance, and missing details: Bewley (PRS Global Open 2022) found the authors did not publicly register their plan beforehand, left out standard reporting items, did not give each study's design or missing-data rates, and did not check whether weaker studies gave different answers.
Results varied a lot and may be skewed by publication: The studies disagreed with each other a great deal, and there were signs that low-regret studies were more likely to be published.
May not apply to today's patients: Most people studied were adults treated under older, stricter rules. Today's patients are more often teens when distress starts, more often born female and more often non-binary, and the claim that results apply to non-binary people rests on one patient.
Credit given without evidence: The authors say low regret shows that better patient screening works, but the data never tested this. Studies using different screening rules were not compared.
Narrow search: The search only covered English and Spanish papers up to May 2020 and found only 74 records in the databases.
Possible conflicts of interest: The authors said they had no financial interests. Bewley noted that several are plastic surgeons who perform these operations and questioned whether that statement holds.

detrans.ai

Summary

Key findings

Abstract

Conclusion