Stillbirth in Iceland: Causes and clinicopathological phenotypes

Abstract

Aims: to explore the stillbirth rate (SBR) and examine the etiology of stillbirth in Iceland from 1996-2021 as well as review the placental histopathology.

Methods: Clinical information was obtained from medical records of mothers who had stillbirths and their infants (n=395). Infants were divided into groups according to gestational age (GA) at diagnosis of stillbirth: early preterm: ≥22 but < 28 weeks (n=140), late preterm: ≥28 but <37 weeks (n=130) and term: ≥ 37 weeks (n=125). Autopsy records and gross description of placenta were reviewed and microscopic slides re-evaluated and classified according to the Amsterdam placental workshop consensus into four major patterns of placental injury. Primary and associated causes of death were assigned according to the Stockholm classification of stillbirth. The SBR, maternal and fetoplacental characteristics, placental histopathology as well as cause of death were compared between two 13-year periods (1996-2008 and 2009-2021) and between GA groups. Chi-square tests were applied and p-level of 0.05 considered statistically significant.

Results: The SBR decreased from 4.10 to 2.88/1000 births (p=0.009) between the former and latter period, but the reduction was limited to stillbirths diagnosed before term. Fewer stillbirths in the latter period were attributed to causes such as infection and placental abruption as well as unknown/unexplained. The most common primary causes of stillbirth were reduced circulation in the umbilical cord (25.6%) and placental insufficiency (25.2%); both increased with advancing GA. Despite no difference in proportion of small for gestational age infants between the two periods, placental weight <10th percentile for GA was significantly more common in the latter period than the earlier (30.3% vs 21.3%, p= 0.002), when a larger proportion of stillbirths was attributed to placental insufficiency (37.0 vs 17.0%, p=0.0002). The stillbirth rate at term was 1.18/1000 term infants with no decrease between the two periods. In stillbirth at term, placentas weighing under the 10th percentile were more common in the latter (43.5 vs 31.7%, p<0.05). Chronic villitis of unknown etiology (VUE) was significantly more often diagnosed in the latter period (40.3 vs,12.7%, p<0.01. Cord complications and placental insufficiency were the most common causes of stillbirth at term as 72% (91/125) of deaths were attributed to these causes, either alone or together. Compared to the earlier period, stillbirth was significantly more often attributed to placental insufficiency (with or without cord complications) in the latter compared to the earlier (56.5% vs. 22.2%, p<0.001). Placental histologic slides were reviewed for 96.4% stillborn singletons and classified into four major patterns of placental injury. Maternal vascular malperfusion (MVM) was diagnosed in 19.0%, fetal vascular malperfusion (FVM) in 31.6%, acute chorioamnionitis (ACA) in 32.2% and VUE in 15.9%. More than one pattern of placental injury was found in 7.7% and none of the major patterns in 27.9% of placentas. A similar proportion of MVM was seen in all GA groups, FVM was more common ≥28 weeks whereas ACA was most frequent <28 weeks and VUE most often diagnosed at term. A higher proportion of maternal vascular malperfusion was found in stillbirths with small for gestational age (SGA) infants than non-SGA (23.0 vs 6.1%) as well as in stillbirths with maternal hypertensive disorder of pregnancy than in stillbirths with a normotensive mother (23.9 vs 11.8%). The umbilical cord was classified as at risk in half (53.8%) of singleton stillbirths, the proportion increased with gestational age (28.9% <28 weeks, 56.0% ≥28 but <37 and 71.7% at term). Hypercoiled, excessively long and wrapped cords were most common. Term stillbirths with cord at risk often also had placental MVM and/or VUE.

Conclusions: The SBR decreased in the latter period due to a reduction in preterm stillbirth but the SBR at term was unchanged. Reduced circulation of the umbilical cord and placental insufficiency were the commonest causes of death, both increasing with gestational age. Stillbirth due to infection and placental abruption as well as unexplained stillbirths decreased during the study period whereas deaths attributed to placental insufficiency became more common, reflecting lack of reduction of stillbirth at term in the latter period. Stillbirths with VUE and FVM, as well as umbilical cord at risk, were more common with advancing GA and often without recognized risk factors. Understanding major patterns of placental injury and their correlation to clinical phenotypes of stillbirth is valuable when counselling after stillbirth and planning surveillance of a subsequent pregnancy.

Markmið: Að rannsaka tíðni andvanafæðinga á Íslandi á árabilinu 1996-2021 og orsakir þeirra, meðal annars með því að endurskoða meinafræði fylgnanna.

Aðferðir: Klínískum upplýsingum var safnað úr sjúkraskrám mæðra sem höfðu fætt andvana barn eða börn á rannsóknartímanum (n=385) og barna þeirra (n=395). Börnunum var skipt í eftirfarandi hópa eftir meðgöngulengd þegar dauðsfallið greindist: ≥22 en <28 vikur (n=140), ≥28 en <37 vikur (n=130) and ≥37 vikur (n=125). Farið var yfir niðurstöður krufningar og lýsingu fylgju, belgja og naflastrengs og auk þess voru öll smásjárgler endurskoðuð og flokkuð samkvæmt Amsterdam samkomulaginu um fylgjumeinafræði (e. Amsterdam placental workshop consensus). Ákvörðuð var aðal (primary) og meðverkandi (associated) orsök dauðsfallsins samkvæmt Stokkhólms flokkun á andvanafæðingum (e. Stockholm classification of stillbirth). Tíðni andvanafæðinga, upplýsingum um mæður, nýbura og fylgjur þ.á.m. fylgjumeinafræði svo og orsakir andvanafæðinganna voru borin saman milli tveggja 13 ára tímabila (1996-2008 and 2009-2021), svo og milli þriggja meðgöngulengdarhópa. Beitt var kí-kvandrat prófi (Chi-square tests) og p-gildi 0.05 var talið tölfræðilega marktækt.

Niðurstöður: Tíðni andvanafæðinga lækkaði úr 4,10 í 2,88/1000 fæðingar (p=0,009) milli fyrri og seinni helminga rannsóknartímans vegna færri andvanafæðinga fyrir 37 vikna meðgöngu. Á seinni helmingi rannsóknartímans voru færri andvanafæðingar vegna sýkinga, fylgjuloss eða óútskýrðra orsaka. Algengustu aðalorsakir andvanafæðinga voru flæðistruflun um naflastreng (25,6%) og fylgjuþurrð (25,2%) og þessar orsakir voru báðar algengari eftir því sem meðgangan var lengri. Þrátt fyrir að ekki væri munur milli tímabila á hlutfalli andvanafæddra barna sem voru létt miðað við meðgöngulengd (undir 10. percentili) var marktækt hærra hlutfall fylgna léttar miðað við meðgöngulengd á seinni helmingi rannsóknartímans (30,3 vs 21,3%, p=0,002), þegar marktækt stærra hlutfall andvanafæðinga var rakið til fylgjuþurrðar (37,0 vs 17,0%, p=0,0002). Tíðni andvanafæðinga fullburða barna var 1,18/1000 börn fædd eftir 37 vikna meðgöngu og breyttist ekki yfir rannsóknartímann. Hjá andvanafæddum fullburða börnum var marktækt hærra hlutfall fylgna létt miðað við meðgöngulengd á seinni helmingi rannsóknartímans (43,5% vs. 31,7%, p<0.05), þegar langvinn bólga í fylgjutotum af óþekktum uppruna (e. chronic villitis of unknown etiology, VUE) greindist einnig marktækt oftar (40,3% vs. 12,7%, p<0,01). Algengustu orsakir andvanafæðinga fullburða barna voru flæðistruflun um naflastreng og fylgjuþurrð en 72% dauðsfall var rakin til annarrar þessara orsaka eða beggja. Marktækt hærra hlutfall andvanafæðinga fullburða barna var rakið til fylgjuþurrðar (með eða án flæðistruflunar um naflastreng) á seinni hluta rannsóknartímans (56,5% vs 22,2%, p<0,001). Vefjasneiðar frá 96,4% af fylgjum andvanafæddra einbura voru endurskoðaðar og flokkaðar í fjóra meginhópa fylgjuskemmda (e. four major patterns of placental injury). Fylgjuskemmdin maternal vascular malperfusion (MVM) greindist í 19,0% af fylgjum, fetal vascular malperfusion (FVM) í 31,6%, acute chorioamnionitis (ACA) í 32,2% og VUE í 15,9%. Fleiri en einn meginhópur fylgjuskemmda fannst í 7,7% fylgna, oftast eftir fulla meðgöngu, og ekkert meginhópanna fjögurra í 27,9%. Svipað hlutfall af MVM fannst í fylgjunum óháð meðgöngulengd, FVM var algengara ≥28 vikur en ACA fannst oftast <28 vikur og VUE greindist oftast í fylgjum fullburða barna. Hærra hlutfall af einangruðu MVM fannst í fylgjum einbura með lága fæðingarþyngd en annarra andvana fæddra einbura (23,0 vs 6,1%), svo og hverra mæður höfðu verið með háþrýstingssjúkdóm á meðgöngu samanborið við fylgjur andvanafædda einbura hverra mæður höfðu ekki verið með háþrýsting (23,9 vs 11,8%). Naflastrengir flokkuðust í áhættu fyrir flæðistruflun hjá rúmlega helmingi (53,8%) andvanafæddra einbura og hlutfallið jókst með lengd meðgöngu (28,9% <28 viku, 56,0% ≥28 en <37 og 71,7% >37 vikur). Algengustu frábrigðin voru ofsnúnir, of langir og vafðir naflastrengir, sem oft fór saman. Fullburða andvana einburar með áhættunaflastrengi voru oft einnig með MVM og/eða VUE í fylgjunni.

Ályktanir: Tíðni andvanafæðinga lækkaði á síðari hluta rannsóknartímans vegna fækkunar á andvanafæðingum fyrirbura en tíðni andvanafæðinga fullburða barna breyttist ekki. Algengustu dánarorsakirnar voru flæðistruflun um naflastreng og fylgjuþurrð og báðar þessar orsakir voru algengari eftir því sem meðgangan var lengri. Hlutfall andvanafæðingum vegna fylgjuþurrðar jókst á seinni hluta tímans enda fækkaði andvanafæðingum fullburða barna ekki. Með lengri meðgöngu greindist oftar VUE eða FVM í fylgjum andvanafæddra barna svo og áhættunaflastrengir, oft eftir meðgöngu án þekktra áhættuþátta. Skilningur á meginhópum fylgjuskemmda og tengslum þeirra við klíníska áhættuþætti getur hjálpað við að útskýra hvers vegna barn dó fyrir fæðingu og við að leggja drög að eftirliti á næstu meðgöngu