Abstract
Background:
Since the complications associated with endometriosis during pregnancy are rare, there is not enough evidence to indicate that this disease has major detrimental effects on pregnancy outcome. Therefore, the present study was conducted with the aim of investigating pregnancy complications related to endometriosis in pregnant women.
Materials and methods
This case-control study was conducted on pregnant women referred to the Obstetrics and Gynecology Clinic at Amiralmomenin Hospital (Zabol, Iran) from 2021 to 2023 to evaluate obstetric complications in those with endometriosis. Fifty pregnant women with endometriosis were compared to fifty pregnant women without endometriosis in terms of pregnancy complications and outcomes. The significance level was considered less than 0.05.
Results
The mean age of the participants was 28.8 ± 5.3 years. The youngest and oldest women were 16 and 43 years old, respectively. 47% of all women had at least one pregnancy complication (54% in the endometriosis group and 40% in the normal group (P=0.161). Women with endometriosis had significantly hagher rates of abortion than the control group (P=0.046), but other complications related to pregnancy were not significantly different between the two groups.
Conclusion
In this study, abortion was more frequent among pregnant women with endometriosis than among controls. However, no statistically significant between-group differences were observed for the other assessed obstetric complications.
Keywords
Abortion, Complications, Endometriosis, Pregnancy, Spontaneous
Introduction
Endometriosis is a chronic gynecological condition defined by the presence of endometrial-like glands and stroma outside the uterine cavity. The disease predominantly affects women of reproductive age, with a reported incidence rate of 6-10% in this population . These ectopic lesions can be found on the peritoneum, ovaries, and other sites within the pelvis . Endometriosis is a major clinical concern, being identified as the underlying cause of infertility in 30-50% of affected women (1-4).
Clinical manifestations of endometriosis vary in women. Patients often present with intermenstrual bleeding, painful menstruation, painful intercourse (dyspareunia), painful defecation (dyschezia), and painful urination (dysuria). Pelvic pain may appear before menstruation. Nonetheless, endometriosis may be asymptomatic and only found as an incidental finding during infertility evaluations (5, 6).
In addition, endometriosis may be associated with changes in ovulation, an increase in inflammatory cells in the peritoneal fluid, ovarian endometrioma, and disruption of the normal endometrium. These changes involve the uterine environment and may endanger normal fetus development. It is possible that such disorders in the peri-implantation period may continue in all subsequent stages of pregnancy and lead to adverse outcomes for both mother and her fetus (7-10).
Many studies in recent years have shown a relationship between endometriosis and adverse maternal and fetal outcomes, such as premature birth, preeclampsia, placenta previa, and postpartum hemorrhage. However, the findings in the literature are contradictory. For instance, some studies report a significant association between endometriosis and such complications (11-14), while others find no such relationships. This inconsistency highlights the need for further researcha. Additionally, many studies include pregnant women who conceived via assisted reproduction, which may confound the relationship between endometriosis and adverse perinatal outcomes, due to the higher prevalence of endometriosis among women requiring fertility treatment (13, 15-18). Given the inconsistent findings in the literatures and the limited evidence from Iranian populations, this study aimed to provide comparative data on a broad range of obstetric complications among pregnant women with and without endometriosis.
Materials and methods
This comparative cohort study enrolled pregnant women referred to the Obstetrics and Gynecology Clinic at Amiralmomenin Hospital (Zabol, Iran) between 2021 and 2023. A consecutive sampling method based on the census approach was employed. Accordingly, 50 eligible women diagnosed with endometriosis were selected, and 50 women without endometriosis were included as a one-to-one matched control group. No predetermined effect size was considered for sample size calculation, which is acknowledged as a limitation of the study.
Inclusion criteria for the case croup: Pregnant women with a confirmed diagnosis of endometriosis who had been referred to the clinic for fertility evaluation and treatment prior to their current pregnancy. The diagnosis was established either through laparoscopic surgery with histological confirmation or via clinical assessment supported by imaging findings from transvaginal sonography (TVS) or magnetic resonance imaging (MRI). Patients with a histological diagnosis were classified according to the revised American Society for Reproductive Medicine (ASRM) staging system.
Exclusion criteria: Women who did not consent to participate in the study, or women who did not consent to imaging [TVS, trans abdominal sonography (TAS)], transrectal sonography (TRS), or MRI. Additionally, women without a confirmed diagnosis of endometriosis or those who had other confounding conditions affecting pregnancy outcomes were excluded.
Pregnant women with endometriosis who met the inclusion criteria formed the case group. Women with no history of endometriosis and TVS before pregnancy were included in the control group.
Pregnancy complications were monitored during routine follow-ups. Maternal underlying conditions, including hypertension (HTN) and gestational diabetes mellitus (GDM), were assessed during and after pregnancy. Complications such as abortion [first-trimester abortion (under 20 weeks of pregnancy)], preterm delivery, pre-eclampsia, ectopic pregnancy, placenta previa, fetal abnormalities, stillbirth, and intrauterine growth restriction (IUGR) were evaluated.
Data from both the endometriosis group and the control group were compared to identify differences in obstetric outcomes. The stages of endometriosis were considered in the analysis to evaluate their impacts on pregnancy complications.
Ethical approval
This work complies with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. All experimental protocols were approved by the Ethics Committee of the Zabol University of Medical Sciences (IR.ZBMU.REC.1401.047). An informed consent letter was obtained from all study participants or their parents or legal guardian(s). They were assured that their information would be kept confidential.
Statistical analysis
Data were analyzed using SPSSv.26 (IBM Corp., Armonk, NY, USA) software to describe them in the form of frequency, percentage, mean and standard deviation. The comparison of women with and without endometriosis in terms of various obstetric outcomes was performed using chi square test. The signonficance level was considered less than 0.05.
Results
In this study, 100 pregnant women (50 women with endometriosis and 50 women without endometriosis) were assessed for pregnancy complications. The mean age was 28.8 ± 5.3 years and the youngest and oldest were 16 and 43 years old, respectively. The mean age in the endmetriosis group was 29.2 ± 4.6 years and in the control group was 28.3 ± 6 years (P=0.402).
Overall, 47% of the patients had at least one pregnancy complication and 53% of them completed their pregnancy without any complications (54% in the endometriosis group and 40% in the normal group (P=0.161).
In Table 1, a summary of the data on age and underlying diseases of the pregnant women are presented.
Table 1.
|
| ||||
|---|---|---|---|---|
| Underlying diseases | Group | P value | ||
| Endometriosis | Control | |||
|
| ||||
| Age (Y) | 29.2 ± 4.6 | 28.3 ± 6.0 | 0.402 | |
| GDM | 0.160 | |||
| Yes | 2 (4) | 7 (14) | ||
| No | 48 (96) | 43 (86) | ||
| HTN | >0.990 | |||
| Yes | 1 (2) | 2 (4) | ||
| No | 49 (98) | 48 (96) | ||
|
|
Data are presented as mean ± SD or n (%). P value based on chi square test. GDM; Gestational diabetes mellitus and HTN; Hypertension.
In Table 2, we evaluated and compared of complications of pregnancy in the two groups.
Table 2.
|
| ||||
|---|---|---|---|---|
| Complications of pregnancy | Group | P value | ||
| Endometriosisn (%) | Controln (%) | |||
|
| ||||
| Obstetric complications | 0.161 | |||
| Yes | 27 (54) | 20 (40) | ||
| No | 23 (46) | 30 (60) | ||
| Ectopic pregnancy | 0.117 | |||
| Yes | 4 (8) | 0 (0) | ||
| No | 46 (92) | 50 (100) | ||
| Abortion | 0.046 | |||
| Yes | 8 (16) | 2 (4) | ||
| No | 84 (84) | 48 (96) | ||
| IUGR | 0.242 | |||
| Yes | 3 (6) | 0 (0) | ||
| No | 47 (94) | 50 (100) | ||
| Preterm delivery | >0.990 | |||
| Yes | 4 (8) | 3 (6) | ||
| No | 46 (92) | 47 (94) | ||
| Fetal abnormality | >0.990 | |||
| Yes | 1 (2) | 0 (0) | ||
| No | 49 (98) | 50 (100) | ||
| Preeclampsia | >0.990 | |||
| Yes | 2 (4) | 2 (4) | ||
| No | 48 (96) | 48 (96) | ||
| Placenta previa | >0.990 | |||
| Yes | 2 (4) | 2 (4) | ||
| No | 48 (96) | 48 (96) | ||
| Stillbirth | >0.990 | |||
| Yes | 1 (2) | 1 (2) | ||
| No | 49 (98) | 49 (98) | ||
|
|
Data are presented as n (%). IUGR; Intrauterine growth restriction. P value based on chi square test.
Discussion
In this study, we found that endometriosis was significantly associated with a higher frequency of abortion. However, no statistically significant differences were observed between the endometriosis and control groups for other pregnancy complications, including preterm delivery, pre-eclampsia, ectopic pregnancy, placenta previa, fetal abnormalities, stillbirth, and IUGR. Porpora et al. (19) evaluated 145 pregnant women with endometriosis and reported a higher risk of obstetric complications, such as abortion and preterm delivery, along with a higher frequency of cesarean section compared to the control group are in contrast to our results.
The discrepancy between the studies could be attributed to several factors. These may include differences in the ethnic background of the study populations, variations in the mean age of the participants, or differences in the clinical characteristics and severity staging of the enrolled endometriosis patients.
Saraswat et al.’s (20) study showed that women with endometriosis have a higher risk of early pregnancy complications such as abortion and ectopic pregnancy compared to those without endometriosis. In agreement with Saraswat’s study, we found that abortion was associated with endometriosis, as it had a higher frequency in women with endometriosis compared to the control group. However, based on our results, the occurrence of ectopic pregnancies didn not differ between women with or without endometriosis. It seems that abortion has a significant correlation with endometriosis; however, more investigation is required to confirm this data.
In general, in the context of pregnancy complications related to endometriosis, various findings have been reported in different studies, suggesting that pregnancyvcomplications may differ based on the type of endometriosis lesion. In fact, the presence of deep endometriosis lesions in pregnant women is underestimated, but such lesions may cause unexpected and severe complications during pregnancy, such as the risk of perforation of the surrounding tissue, especially when endometriosis involves the intestines. Also, bleeding during pregnancy caused by endometriosis tissue on the ileum or terminal colon has been reported (21, 22).
Berlac et al. (23), have also shown that women with endometriosis were at risk of several issues such as preeclampsia and placental complications in pregnancy and childbirth. We found that 4% of pregnant women with endometriosis had the risk of suffering from preeclampsia or placenta previa, and these rates did not differ from those in normal women. Therefore, the findings of our study was in contrast with Berlac et al.’s (23) study.
Harada et al. (24), evaluated pregnant women with endometriosis and mentioned that the risk of complications such as premature rupture of membranes, premature delivery, and placenta previa were significantly higher compared to women without endometriosis. The study by Glavind et al. (25) has also shown that women with endometriosis are at risk of pre-eclampsia, preterm delivery, and required cesarean section, regardless of the use of assisted reproductive techniques. In the present study, despite the fact that the frequency of complications such as ectopic pregnancy, IUGR, preterm delivery, and fetal abnormalities in pregnant women with endometriosis was higher than in the control group, but unlike the studies mentioned above, these differences were not statistically significant.
Our finding of no significant association for several obstetric complications is supported by other studies. For instance, a large population-based study by Stephansson et al. (26) found that, after adjustments, endometriosis was not an independent risk factor for adverse outcomes such as pre-eclampsia or stillbirth. This consistency suggests that the influence of endometriosis on specific pregnancy complications may be less pronounced than previously thought, and that the apparent risks could be confounded by other factors such as maternal age or the use of assisted reproductive technology.
Pregnancy complications associated with endometriosis may be explained by some pathogenic mechanisms, such as chronic inflammation associated with endometriosis, the presence of adhesions and their consequences, and invasion of the ectopic endometrium into the vascular wall (27-29). Nonetheless, the potential mechanisms underlying pregnancy complications in women with endometriosis are still largely unknown. However, evidence has shown that peritoneal fluid in women with endometriosis is at least partially due to pro-inflammatory changes, including increased levels of cytokines and angiogenic factors, which may be a possible explanation for the higher risk of preterm birth in these patients. A local increase in the peristaltic activity of the uterus may cause microtraumatization and implantation disorder, all of which can ultimately justify the risk of preeclampsia and premature delivery in pregnant women with endometriosis (30-32).
Conclusion
Abortion was more frequent among pregnant women with endometriosis than among controls, whereas no statistically significant between-group differences were observed for the other assessed obstetric complications. However, based on the results of the current study, no relationship is observed between other pregnancy complications and endometriosis. Therefore, although pregnant women with endometriosis may not have any issues, it is necessary to be aware of the increase in some pregnancy complications like abortion and IUGR.
Acknowledgments
There is no financial support in this study.
Conflict of interest
The authors declare no competing interests.
Author’s Contributions
M.K.; Conceptualization, Project administration, Investigation, and Writing the original draft. K.S.; Formal analysis, Methodology, and Writing the original draft. M.K., K.S., M.A.; Investigation and Writing the original draft. H.H., Kh.R.K., Sh.Ch., A.M.K.; Writing the original draft, Supervision, Review, and Editing the manuscript. All authors read and approved the final manuscript.
References
- 1.As-Sanie S, Mackenzie SC, Morrison L, Schrepf A, Zondervan KT, Horne AW, et al. Endometriosis: A Review. JAMA. 2025;334(1):64–78. doi: 10.1001/jama.2025.2975. doi: 101001/jama20252975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Egami N, Muta R, Anami A, Koga H. Impact of clinical practice guidelines for vacuum-assisted delivery on maternal and neonatal outcomes in Japan: a single-center observational study. J Obstet Gynaecol Res. 2021;47(1):167–173. doi: 10.1111/jog.14448. doi: 101111/jog14448. [DOI] [PubMed] [Google Scholar]
- 3.Mizrachi Y, Weissman A. Reply: Should women receive luteal sup port following natural cycle frozen embryo transfer?. A systematic review and meta-analysis.Critical assessment of a review and meta-analysis. Hum Reprod Update. 2021;27(4):799–800. doi: 10.1093/humupd/dmab021. doi: 101093/humupd/dmab021. [DOI] [PubMed] [Google Scholar]
- 4.Shafrir AL, Farland LV, Shah DK, Harris HR, Kvaskoff M, Zonder van K, et al. Risk for and consequences of endometriosis: a critical epidemiologic review. Best Pract Res Clin Obstet Gynaecol. 2018;51:1–15. doi: 10.1016/j.bpobgyn.2018.06.001. doi: 101016/jbpobgyn201806001. [DOI] [PubMed] [Google Scholar]
- 5.Chauhan S, More A, Chauhan V, Kathane A. Endometriosis: a re view of clinical diagnosis, treatment, and pathogenesis. Cureus. 2022;14(9):e28864–e28864. doi: 10.7759/cureus.28864. doi: 107759/cureus28864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Brosens I, Gargett CE, Guo SW, Puttemans P, Gordts S, Brosens JJ, et al. Origins and progression of adolescent endometriosis. Re prod Sci. 2016;23(10):1282–1288. doi: 10.1177/1933719116637919. doi: 101177/1933719116637919. [DOI] [PubMed] [Google Scholar]
- 7.Taylor HS, Kotlyar AM, Flores VA. Endometriosis is a chronic sys temic disease: clinical challenges and novel innovations. Lancet. 2021;397(10276):839–852. doi: 10.1016/S0140-6736(21)00389-5. doi: 101016/S0140-6736(21)00389-5. [DOI] [PubMed] [Google Scholar]
- 8.Kisovar A, Becker CM, Granne I, Southcombe JH. The role of CD8+ T cells in endometriosis: a systematic review. Front Immunol. 2023;14:1225639–1225639. doi: 10.3389/fimmu.2023.1225639. doi: 103389/fimmu20231225639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Imanaka S, Maruyama S, Kimura M, Nagayasu M, Kobayashi H. Towards an understanding of the molecular mechanisms of endo metriosis-associated symptoms.World Acad Sci J.2020; 2(4); 12.doi: 10. 3892/wasj. 2020 53. [Google Scholar]
- 10.Horne AW, Missmer SA. Pathophysiology, diagnosis, and manage ment of endometriosis. BMJ. 2022;379:e070750–e070750. doi: 10.1136/bmj-2022-070750. doi: 101136/ bmj-2022-070750. [DOI] [PubMed] [Google Scholar]
- 11.Peres LC, Risch H, Terry KL, Webb PM, Goodman MT, Wu AH, et al. Racial/ethnic differences in the epidemiology of ovarian cancer: a pooled analysis of 12 case-control studies. Int J Epidemiol. 2018;47(2):460–472. doi: 10.1093/ije/dyx252. doi: 101093/ije/dyx252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Nezhat C, Young S, Burns MK, Johns E, Wang P. Pregnancy com plications in patients with endometriosis. Fertil Steril. 2017;108(4):602–603. doi: 10.1016/j.fertnstert.2017.08.014. doi: 101016/jfertnstert201708014. [DOI] [PubMed] [Google Scholar]
- 13.Borisova AV, Konnon SRD, Tosto V, Gerli S, Radzinsky VE. Ob stetrical complications and outcome in patients with endometrio sis. J Matern Fetal Neonatal Med. 2022;35(14):2663–2677. doi: 10.1080/14767058.2020.1793326. doi: 101080/1476705820201793326. [DOI] [PubMed] [Google Scholar]
- 14.Opøien HK, Fedorcsak P, Omland AK, Abyholm T, Bjercke S, Ertzeid G, et al. In vitro fertilization is a successful treatment in endometriosis-associated infertility. Fertil Steril. 2012;97(4):912–912. doi: 10.1016/j.fertnstert.2012.01.112. [DOI] [PubMed] [Google Scholar]
- 918.doi: 10. 1016/j. fertnstert. 2012 01112. [Google Scholar]
- 15.Elsherbini M, Koga K, Maki E, Kumasawa K, Satake E, Taguchi A, et al. Impact of chronic exposure to endometriosis on perinatal outcomes: establishment of a mouse model. Biomedicines. 2022;10(10):2627–2627. doi: 10.3390/biomedicines10102627. doi: 103390/biomedicines10102627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Horton J, Sterrenburg M, Lane S, Maheshwari A, Li TC, Cheong Y. Reproductive, obstetric, and perinatal outcomes of women with adenomyosis and endometriosis: a systematic review and meta analysis. Hum Reprod Update. 2019;25(5):592–632. doi: 10.1093/humupd/dmz012. doi: 101093/ humupd/dmz012. [DOI] [PubMed] [Google Scholar]
- 17.Ono Y, Furumura K, Yoshino O, Ota H, Sasaki Y, Hidaka T, et al. Influence of laparoscopic surgery for endometriosis and its re currence on perinatal outcomes. Reprod Med Biol. 2022;21(1):e12456–e12456. doi: 10.1002/rmb2.12456. doi: 101002/rmb212456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang JQ, Zhang JM, Qian B. Adverse pregnancy outcomes for women with endometriosis: a systematic review and meta-analy sis. Ginekol Pol. 2024;95(9):668–676. doi: 10.5603/GP.a2021.0081. doi: 105603/GPa20210081. [DOI] [PubMed] [Google Scholar]
- 19.Porpora MG, Tomao F, Ticino A, Piacenti I, Scaramuzzino S, Si monetti S, et al. Endometriosis and pregnancy: a single institution experience. Int J Environ Res Public Health. 2020;17(2):401–401. doi: 10.3390/ijerph17020401. doi: 103390/ijerph17020401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Saraswat L, Ayansina DT, Cooper KG, Bhattacharya S, Miligkos D, Horne AW, et al. Pregnancy outcomes in women with endometrio sis: a national record linkage study. BJOG. 2017;124(3):444–452. doi: 10.1111/1471-0528.13920. doi: 101111/1471-052813920. [DOI] [PubMed] [Google Scholar]
- 21.Setúbal A, Sidiropoulou Z, Torgal M, Casal E, Lourenço C, Kon inckx P. Bowel complications of deep endometriosis during preg nancy or in vitro fertilization. Fertil Steril. 2014;101(2):442–446. doi: 10.1016/j.fertnstert.2013.11.001. doi: 101016/jfertnstert201311001. [DOI] [PubMed] [Google Scholar]
- 22.Exacoustos C, Lauriola I, Lazzeri L, De Felice G, Zupi E. Complica tions during pregnancy and delivery in women with untreated rec tovaginal deep infiltrating endometriosis. Fertil Steril. 2016;106(5):1129–1135. doi: 10.1016/j.fertnstert.2016.06.024. e1doi: 101016/jfertnstert201606024. [DOI] [PubMed] [Google Scholar]
- 23.Berlac JF, Hartwell D, Skovlund CW, Langhoff-Roos J, Lidegaard Ø. Endometriosis increases the risk of obstetrical and neonatal complications. Acta Obstet Gynecol Scand. 2017;96(6):751–760. doi: 10.1111/aogs.13111. doi: 101111/aogs13111. [DOI] [PubMed] [Google Scholar]
- 24.Harada T, Taniguchi F, Onishi K, Kurozawa Y, Hayashi K, Harada T, et al. Obstetrical complications in women with endometriosis: a cohort study in Japan. PLoS One. 2016;11(12):e0168476–e0168476. doi: 10.1371/journal.pone.0168476. doi: 101371/journalpone0168476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Glavind MT, Forman A, Arendt LH, Nielsen K, Henriksen TB. Endo metriosis and pregnancy complications: a Danish cohort study. Fer til Steril. 2017;107(1):160–166. doi: 10.1016/j.fertnstert.2016.09.020. doi: 101016/jfertnstert201609020. [DOI] [PubMed] [Google Scholar]
- 26.Stephansson O, Kieler H, Granath F, Falconer H. Endometriosis, assisted reproduction technology, and risk of adverse pregnancy outcome. Hum Reprod. 2009;24(9):2341–2347. doi: 10.1093/humrep/dep186. doi: 101093/hum rep/dep186. [DOI] [PubMed] [Google Scholar]
- 27.Manresa MC, Godson C, Taylor CT. Hypoxia-sensitive pathways in inflammation-driven fibrosis. Am J Physiol Regul Integr Comp Phys iol. 2014;307(12):R1369–R1380. doi: 10.1152/ajpregu.00349.2014. doi: 101152/ajpregu003492014. [DOI] [PubMed] [Google Scholar]
- 28.Erikson DW, Chen JC, Piltonen TT, Conti M, Irwin JC, Giudice LC. Inhibition of epidermal growth factor receptor restores decidualiza tion markers in stromal fibroblasts from women with endometriosis. J Endometr Pelvic Pain Disord. 2014;6(4):196–211. doi: 105301/ je5000198. [Google Scholar]
- 29.Leeners B, Damaso F, Ochsenbein-Kölble N, Farquhar C. The ef fect of pregnancy on endometriosis-facts or fiction? Hum Reprod Update. 2018;24(3):290–299. doi: 10.1093/humupd/dmy004. doi: 101093/humupd/dmy004. [DOI] [PubMed] [Google Scholar]
- 30.Navarro R, Poder L, Sun D, Jha P. Endometriosis in pregnancy. Abdom Radiol (NY) 2020;45(6):1741–1753. doi: 10.1007/s00261-020-02486-7. doi: 101007/s00261-020-02486-7. [DOI] [PubMed] [Google Scholar]
- 31.Leyendecker G, Bilgicyildirim A, Inacker M, Stalf T, Huppert P, Mall G, et al. Adenomyosis and endometriosis.Re-visiting their associa tion and further insights into the mechanisms of auto-traumatisa tion.An MRI study. Arch Gynecol Obstet. 2015;291(4):917–932. doi: 10.1007/s00404-014-3437-8. doi: 101007/s00404-014-3437-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Leyendecker G, Wildt L, Mall G. The pathophysiology of endome triosis and adenomyosis: tissue injury and repair. Arch Gynecol Ob stet. 2009;280(4):529–538. doi: 10.1007/s00404-009-1191-0. doi: 101007/s00404-009-1191-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.