Medical Hypotheses
(Abstract 330 words)
(Article 2062 words, 32 references, 2 figures)
Pre-eclampsia – the “uterine reinnervation” view.
MJ QUINN, MB ChB, MD, MRCOG, LLM.
The author declares no financial interests or conflicts of interest
associated with this article.
Corresponding author:
MJ Quinn, MD, LLM.
38 Harwood Road, Fulham, London, SW6 4PH.
Telephone:
E-mail:
0044-7710-288477
mjquinn001@btinternet.com
The “uterine reinnervation” view.
Summary
Difficult vaginal deliveries, gynaecological surgery, and, persistent straining during
defaecation injure uterine nerves. Cytokines released from injured, uterine nerves
cause regeneration of new nerves with altered structures and functions.
In
structural terms, these new nerves proliferate in chaotic and dysfunctional patterns
with abnormal, cross-sectional profiles. In functional terms they are particularly
sensitive to “stretch” or mechanosensory transduction. Release of neural cytokines
also causes hyperplasia of the walls of adjacent, denervated uterine arterioles that
may reduce uteroplacental blood flow during pregnancy.
In the “uterine reinnervation” view, “stretch” applied to injured uterine nerves
triggers uterorenal nerves to cause vasoconstriction in the renal cortex,
hypertension and proteinuria i.e. the key features of preeclampsia. There are two
intrauterine mechanisms that stretch injured, uterine nerves (a) in the placental bed,
(b) in the extraplacental myometrium, respectively. In “early-onset” preeclampsia
(<34 weeks), continuing increases in maternal plasma volume, increase blood flow
through denervated, and, narrowed uterine arterioles in the placental bed,
stretching injured perivascular nerves resulting in preeclampsia with a small-forgestational-age fetus.
In “late-onset” preeclampsia (>34 weeks), nulliparity,
multiple pregnancy, concealed abruption and polyhydramnios increase myometrial
tension and results in preeclampsia with an appropriate-for-gestational-age fetus.
Widespread activation of autonomic nerves results in multi-system features of these
syndromes. Changes in placental site and circulatory compliance may contribute to
different phenotypes of the preeclamptic syndromes in subsequent pregnancies.
The “uterine reinnervation” view offers an explanation of the common clinical
features of the preeclamptic syndromes through a single pathophysiological
mechanism, namely, prepregnancy injury to uterine nerves. Importantly, it offers
2
The “uterine reinnervation” view.
an explanation for resolution of the symptoms and signs of preeclampsia with
delivery of the fetus, the “early” and “late-onset” preeclamptic syndromes, and, the
established clinical associations of the condition including nulliparity, hydramnios,
multiple pregnancy, molar pregnancy, concealed abruption, etc. Establishing the
presence of injured nerves expressing mechanoreceptors in the uterus, and, neural
cytokines in thickened, uterine arterioles, will assist in developing this view.
However, myometrial hyperplasia during the second half of pregnancy separates
injured uterine nerves from injured uterine arterioles ensuring that the key
pathoanatomical relationship in preeclampsia will be difficult to demonstrate.
3
The “uterine reinnervation” view.
Preeclampsia is characterized by the onset of hypertension and proteinuria in the
second half of pregnancy (140/90 mm, >0.3g proteinuria per 24 hours). It affects
5-8% of women with increased incidences in teenagers, women over 35 years of
age, and, first-degree relatives (1). Clinical associations of preeclampsia include
nulliparity, multiple pregnancy, polyhydramnios, concealed abruption, hydatidiform
mole, pre-existing hypertension, diabetes mellitus, obesity, and, renal disease.
Severe preeclampsia (BP 160/110 mm, > 5g proteinuria per 24 hours) frequently
results in multi-system disease that manifests as thrombocytopenia, abnormal liver
function, pulmonary oedema, and seizures.
Recent clinical and histological observations suggest there are two, distinct, clinical
and placental phenotypes of the pre-eclamptic syndromes (2). Preeclampsia prior
to 34 weeks gestation (“early-onset”) demonstrates placental hypoplasia with many
placental vascular lesions that result in a small-for-gestational-age fetus. After 34
weeks gestation (“late-onset”) there are fewer placental lesions and an appropriatefor-gestational-age fetus. Evidence from retrospective, cohort studies suggests that
women with preeclampsia delivering small-for-gestational-age babies, suffer
increased risks of cardiovascular disease in later life (3). Different, though
overlapping, intrauterine mechanisms may account for the development of
preeclampsia, and, their remote consequences, in these two, distinctive clinical
syndromes.
4
The “uterine reinnervation” view.
Previous theories of preeclampsia
The “uterine reinnervation” view describes some of the diverse and varying
consequences of injuries to uterine nerves (Fig. 1). It brings together two previous
strands of research in the history of preeclampsia; the “mechanical distension”
theory and the “placental ischaemia” theory. Both take origin from the initial
observation of proteinuria in women with eclampsia by Dr Charles Lever (1843,
Guy’s Hospital, London) (4). The “mechanical distension” theory gathered
momentum in the second half of the nineteenth century when WH Dickinson
(1868, St Georges Hospital, London) proposed that proteinuria resulted from
compression of the renal veins and inferior vena cava by the pregnant uterus (5). In
1929, RH Paramore (Rugby Hospital) published an updated version of the
“mechanical distension” theory though it received less attention with the
emergence of the “placental ischemia” theory in the 1930’s (6).
The “placental ischemia” theory originated in 1914 when Dr James Young
(University College Hospital, London) found histological evidence of villous
overcrowding adjacent to areas of placental infarction on the fetal side of the
maternal-fetal interface. However, Young continued to attribute preeclampsia to
“toxins” released from these infarcts (7). Dr EW Page (1939, University of
California at San Francisco) formally set out the “placental ischaemia” hypothesis
that achieved histopathological endorsement with the observations of Drs. HG
Dixon and WB Robertson (8, 9). They described irregular narrowing of uterine
spiral arterioles in the placental bed on the maternal side of the maternal-fetal
interface that may be the source of the placental infarcts described by Dr Young
and earlier workers [7, Fig. 2a-d].
In recent years narrowing of uterine arterioles has been associated with different
patterns of reproductive loss including miscarriage, cornual ectopic pregnancy,
5
The “uterine reinnervation” view.
midtrimester loss, intrauterine growth retardation, preterm labour and placental
abruption (10). Originally, this lesion was held to arise from “two waves of
trophoblast invasion” though there are continuing doubts about the site and timing
of these histological events (11). The lesion also occurs in non-pregnant women
with endometriosis-associated infertility, and, in cornual, ectopic pregnancies at
eight weeks gestation (12, Fig. 2b-d). Narrowing of uterine spiral arterioles may,
therefore, be a non-specific finding indicative of an injury to uterine vasomotor
nerves, that is an antecedent of both preeclampsia and other forms of reproductive
loss.
The “uterine reinnervation” view has its modern origins in statements by Dr Otto
Spiegelberg (1878, Breslau) who said that eclampsia was caused by “severe renal
vasospasm, reflexly elicited by uterine distention, and that prolongation of the
vasoconstriction would result in irreversible renal disease” (13). In his scholarly
account (14), Dr Chesley records that both Gabelchoverus (1596) and Mauriceau
(1694) attributed preeclampsia to “irritation of the uterus” (15, 16). He further
notes that GJ Sophian revived this hypothesis when observing the experimental
basis for neural connections between the uterus and kidneys in rabbits (17). He
found that distending an intrauterine balloon caused reflex renal vasoconstriction,
and, that cutting the nerves between the uterus and kidneys abolished this reflex
(17). Recent Doppler studies demonstrating increases in the resistance of uterine
and renal circulations that vary directly with increasing blood pressure, provide
evidence for uterorenal connections in women with preeclampsia though further
studies describing their relationships with changes in the uterine artery waveform,
and, their changes at delivery might be helpful (18, 19). These abstracted
observations within the history of research into preeclampsia do not sit in any
coherent theoretical framework, however, they contribute important elements to
the “uterine reinnervation” view.
6
The “uterine reinnervation” view.
Injuries to uterine nerves
Uterine sympathetic nerves take origin from T10-12 and L1-2 whilst
parasympathetic fibres arise from S2-4. They traverse the hypogastric plexi before
entering the uterus with the uterine artery, and, through the uterosacral ligaments
(Fig. 1). Subserosal and endometrial-myometrial nerve plexi sparsely innervate the
myometrium before the nerves continue into the muscle layers of the Fallopian
tube.
Injuries to uterine nerves result from, difficult first labours, excessive traction to
the cervix and over-vigorous curettage during minor gynaecological procedures,
and, persistent straining during defaecation (Fig. 2b, 20, 21). Physical efforts
during defaecation affect 20-30% of Western populations, and, may affect different
pelvic organs depending on the shape of the pelvis, and, pattern of straining efforts
(22). The site, nature and extent of the neurological injury affects its histological
appearance and clinical consequences (20, 21, 23). Injuring uterine vasomotor
nerves causes regenerating aberrant nerves to sprout in close apposition to adjacent,
denervated blood vessels (Fig. 2a-d). These injuries contrast with chaotic
reinnervation following difficult vaginal deliveries, and, traction to the cervix, or,
over-vigorous curettage during minor gynaecological procedures (21).
Varying injuries to autonomic nerves have diverse consequences including changes
in visceral form and, function (20, 24). The structural effects in the uterus include
myometrial and endometrial hyperplasia that results in some forms of leiomyoma
and adenomyosis (21, 25). The functional effects include altered afferent function
where stretching injured, myometrial nerves generates autonomic signals between
the uterus and kidneys that result in renal vasoconstriction, hypertension and
proteinuria [26]. There is clear evidence from bladder studies that stretching
epithelial cells releases adenosine triphosphate (ATP) that acts on injured nerves to
7
The “uterine reinnervation” view.
initiate impulses in sensory pathways (27). Clinical evidence suggesting that
myometrial “stretch” is important in the aetiology of some forms of preeclampsia
includes increased rates of preeclampsia in nulliparous women, multiple pregnancy,
hydramnios, hydatidiform mole and concealed, antepartum haemorrhage. Daily
clinical experience confirms that removal of the “stretch” stimulus at delivery of
the fetus and placenta leads to prompt resolution of the clinical syndrome; a feature
of the condition that has not been satisfactorily explained by previous theories (1).
8
The “uterine reinnervation” view.
The “uterine reinnervation” view
The central feature of the “uterine reinnervation” view is that applying stretch to
regenerated, intrauterine nerves at different anatomical sites and different
gestational ages, generates neural signals in the uterorenal nerves that cause renal
vasoconstriction, hypertension and proteinuria. In the “early onset” syndrome
(before 34 weeks) it is also necessary to explain concurrent narrowing of uterine
arterioles in the placental bed that results in a small-for-gestational-age baby (SGA).
(a) Narrowing of uterine arterioles
Prepregnancy injuries to uterine vasomotor nerves have important consequences.
Firstly, release of neural cytokines at the time of the injury results in regeneration of
“new” nerves in close proximity to the arteriole. These nerves have different
anatomical profiles, and, different properties. Specifically, in anatomical terms
there are many more nerves after the injury, and, the cross-sections of these nerves
are abnormal (Fig. 2b-d). At the same time the flood of neural cytokines released
from the proximal stump of the nerve also act on the now-denervated, arteriolar
walls, causing irregular hyperplasia of the vessel walls with narrowing of their
lumens (Fig. 2b-d). Uteroplacental flow is reduced, and, may lead to a small-forgestational-age fetus in a subsequent pregnancy. In this account, narrowing of
uterine arterioles is simply a marker of injury to uterine vasomotor nerves; it is not
specific to any form of pregnancy loss.
(b) Intrauterine “stretch” receptors
In addition to the anatomical changes in the injured uterine nerves, there are
important changes in their functional properties. The key change is that these new,
regenerative nerves are particularly sensitive to “stretch”. There is clear evidence in
the bladder that injured nerves express purinergic mechanoreceptors e.g. P2X2,
P2X3, P2X7, that are sensitive to “stretch” and use adenosine triphosphate as a
9
The “uterine reinnervation” view.
neurotransmitter. Such receptors have yet to be demonstrated in the uterus in
hypertensive pregnancy though regenerating autonomic nerves may generically,
express these receptors in different forms depending on the nature of the injury
and the tissue of origin. In the “uterine reinnervation” view they form the primary
sensory mechanism that activate the uterorenal nerves to cause renal
vasoconstriction, hypertension and proteinuria.
(c) “Early” and “late” onset syndromes
There are two, distinct, mechanisms that stretch injured, uterine nerves. In “earlyonset” preeclampsia (before 34 weeks), continuing increases in maternal plasma
volume up to 34 weeks gestation, increase blood flow through denervated uterine
arterioles in the placental bed, and, stretch injured perivascular nerves resulting in
preeclampsia with a small-for-gestational-age fetus (Fig. 2d). In “late-onset”
preeclampsia (after 34 weeks), nulliparity, multiple pregnancy, concealed abruption
and polyhydramnios increase myometrial tension and results in preeclampsia with
an appropriate-for-gestational-age, fetus. In both circumstances, stretch
(mechanosensory transduction) results in neural signals from the uterus to the
kidney that lead to vasoconstriction in the renal cortex with hypertension and
proteinuria.
(d) Clinical variations of preeclampsia
There are many clinical variations of the preeclamptic syndromes that require
explanation within the framework of “uterine reinnervation”. Concomitant injuries
to uterorenal nerves may result in a small-for-gestational-age fetus without the renal
consequences of hypertension and proteinuria (Fig. 2d). In exceptional
circumstances, proliferating molar tissue in injured uterine arterioles may account
for severe preeclampsia in the first half of pregnancy, as well as concomitant
pulmonary metastases (28). Pre-existing hypertension and renal disease may reduce
renal thresholds for developing hypertension and proteinuria. Preterm labour and
10
The “uterine reinnervation” view.
placental abruption complicating preeclampsia, represent further diverse
consequences of denervation of the lower genital tract that may include
opportunist infection, and, injury to denervated, myometrial blood vessels (29).
Post-partum “eclampsia” on days 5-6 arises from temporary hypertension
associated with physiological readjustments of the maternal circulation, rather than
the consequences of intrauterine changes (30).
(e) Physical separation of nerve and blood vessels during pregnancy
Uterine nerves do not elongate during the myometrial hypertrophy of the second
half of pregnancy (30). This physiological process separates injured uterine nerves,
that do not stretch along their longitudinal axis, from denervated uterine blood
vessels that do respond to the endocrine environment during this phase of
pregnancy (31). Confirmatory evidence for this anatomical separation arises from
histology of narrowed uterine spiral arterioles that show diminishing evidence of
injured perivascular nerves during the first half of pregnancy (Fig. 2b-c), and, that
continues into the second half of pregnancy (Fig. 2d). Previous studies in the
placental bed in women with adverse pregnancy outcomes have shown no evidence
of nerves in a small number of uteri, as might be anticipated (32).
(f) Autonomic denervation in the uterus and other pelvic organs
Many clinical conditions result from the diverse and varying consequences of
injuries to autonomic nerves that include changes in form and function, sensitivity
to infection and different toxins, pain and sensitization of the central nervous
system (22, 23). In reproductive biology, the three primary uterine pathologies of
leiomyoma, adenomyosis and endometriosis result from injuries to uterine nerves
(20, 21, 24). Most forms of pelvic, uterine, vaginal and vulval pain are
consequences of injuries to uterine nerves (20, 21). Irritative bladder and bowel
symptoms also demonstrate aberrant reinnervation (23). Now, in this framework,
it seems likely that many of “great obstetric syndromes” including preeclampsia,
11
The “uterine reinnervation” view.
preterm labour, placental abruption, etc. may be consequences of autonomic
denervation (33). A woman’s reproductive health may depend on the integrity of
her pelvic autonomic nerves. Preventing injury to these nerves through common
procedures such as good diet and bowel habits, better gynaecological surgery, and,
improved first labours, is a feasible ambition for contemporary clinical care.
Moreover, many extrapelvic, Western diseases demonstrate unexplained
denervation or reinnervation that may also result from similar injuries (23).
Conclusion
The “uterine reinnervation” view provides a mechanism for many of the clinical
and histopathological features of the preeclamptic syndromes. Straining during
defaecation and minor gynaecological surgery prior to a first pregnancy, injure
uterine nerves that regenerate to express primitive stretch receptors. Applying
stretch to these injured uterine nerves at different anatomical sites and gestational
ages, stimulates uterorenal nerves to cause vasoconstriction in the renal cortex,
hypertension and proteinuria. Key evidence to support this view might include
demonstrating neural cytokines in the walls of injured blood vessels, and,
demonstrating mechanoreceptors on injured, uterine nerves.
For the first time, this view offers explanations for the key associations of
preeclampsia including nulliparity, hydramnios, multiple pregnancy, concealed
abruption, molar pregnancy, etc., the early and late-onset syndromes, and, the
resolution of hypertension and proteinuria with removal of intrauterine “stretch” at
the time of delivery. However, many clinicians are unfamiliar with the morphology
of the pelvic autonomic nerves, and, myometrial hypertrophy during the second
half of pregnancy separates injured uterine nerves from denervated uterine blood
vessels, concealing the key pathoanatomical relationship in preeclampsia.
12
The “uterine reinnervation” view.
References
(1)
Sibai B, Dekker G, Kupferminc M.
Pre-eclampsia.
Lancet. 2005; 365(9461):785-99.
(2)
Nelson DB, Ziadie MS, McIntyre DD, Rogers BB, Leveno KJ.
Placental pathology suggesting that preeclampsia is more than one disease.
Am J Obstet Gynecol 2014; 210:66.e1-7.
(3)
Barker DJP, Osmond C, Winter PD, Margetts B, Simmonds SJ.
Weight in infancy and death from ischaemic heart disease.
Lancet 1989; 2(8663):577-80.
(4)
Lever JCW.
Causes of puerperal convulsions with remarks.
Guys Hospital Reports 1843; 1 (2d series): 495-517.
(5)
Dickinson WH.
On the pathology and treatment of albuminuria.
London, Longmans Green, 1868, p115.
(6)
Paramore RH.
An introduction to the mechanistic conception of eclampsia.
Lancet 1928; 2:914.
(7)
Young J.
The Aetiology of Eclampsia and Albuminuria and their Relation to Accidental
Haemorrhage: (An Anatomical and Experimental Investigation.).
Proc R Soc Med. 1914;7(Obstet Gynaecol Sect):307-48.
(8)
Page EW.
The relation between hydatid moles, relative ischemia of the gravid uterus, and the
placental origin of eclampsia.
Am J Obstet Gynecol 1939; 37:291-3.
(9)
Dixon HG, Robertson WB.
A study of the vessels of the placental bed in normotensive and hypertensive women.
J Obstet Gynaecol Br Emp. 1958; 65(5):803-9.
(10) Brosens IA, Pijnenborg R, Vercryusse L, Romero R.
The “Great Obstetrical Syndromes” are associated with disorders of deep placentation.
Am J Obstet Gynecol 2011; 204(3):193-201.
(11) Pijnenborg R, Dixon G, Robertson WB, Brosens I.
Trophoblastic invasion of human decidua from 8 to 18 weeks of pregnancy.
Placenta. 1980; 1(1):3-19.
(12) Wu XQ, Chen HY, Wan L, Quinn MJ.
Abnormal innervation of uterine blood vessels in cornual ectopic pregnancy.
J Obstet Gynaecol 2014 (in press).
13
The “uterine reinnervation” view.
(13) Spiegelberg O.
Lehrbuch der Geburtsch fur Artze u. Studierende. Schauenburg.
Lahr. 1878.
(14) Chesley LC.
Hypertensive disorders in pregnancy.
Appleton-Century-Crofts, 1978.
(15) Gabelchoverus.
Artzneybuch, darninnen vast fur alle des menschlichen Leibs, anlingen und Gebrechen,
ausserlesene und bewehrte Artzneyen usw. Tubingen., Gruppenback, 1596.
(16) Mauriceau F.
Traite des Maladies des Femmes Grosses, et celles Qui Sone Accouchees Einseignant la
Bonne et Verritable Methode pour Bien Aider , etc. Book II, Chap 28, Paris, d’Houry,
1694.
(17) Sophian GJ.
Toxaemias of pregnancy.
London: Butterworth, 1953.
(18) Miyake H, Nakai A, Koshino T, Araki T.
Doppler velocimetry of maternal renal circulation in pregnancy-induced hypertension.
J Clin Ultrasound. 2001; 29(8):449-55.
(19) Nakai A, Miyake H, Oya A, Asakura H, Koshino T, Araki T.
Reproducibility of pulsed Doppler measurements of the maternal renal circulation in
normal pregnancies and those with pregnancy-induced hypertension.
Ultrasound Obstet Gynecol. 2002; 19(6):598-604.
(20) Quinn MJ.
Endometriosis; the consequence of uterine denervation-reinnervation.
Arch Obstet Gynaecol 2011; 284:1423-9.
(21) Atwal G, du Plessis D, Armstrong G, Slade R, Quinn M.
Uterine innervation after hysterectomy for chronic pelvic pain with, and without,
endometriosis.
Am J Obstet Gynecol. 2005;193(5):1650-5.
(22) Heaton KW, Cripps HA.
Straining at stool and laxative taking in an English population.
Dig Dis Sci 1993; 38(6):1004-8.
(23) Quinn M
Autonomic denervation and Western diseases.
Am J Med 2014; 127(1):3-4.
(24) Quinn M.
Uterine innervation in fibroids: a qualitative study.
J Obstet Gynaecol. 2007; 27(5):489-92.
14
The “uterine reinnervation” view.
(25) Quinn MJ.
Perivascular nerve fibre proliferation: the consequence of prolonged straining.
J Obstet Gynaecol. 2007; 27(2):185-8.
(26) Goldblatt H, Lynch J, Hanzel RF, Summerville WW.
Studies on experimental hypertension. I. The production of persistent elevation of
systolic blood pressure by means of renal ischemia.
J. Exp. Med. 1934; 59:347–379.
(27) Burnstock G.
Purinergic mechanosensory transduction and visceral pain.
Mol Pain. 2009; 30(5):69-81.
(28) Page EW.
On the pathogenesis of pre-eclampsia and eclampsia.
J Obstet Gynaecol Br Commonw. 1972 ;79(10):883-94.
(29) Goldenberg RL, Culhane JF, Iams JD, Romero R.
Epidemiology and causes of preterm birth.
Lancet. 2008; 371(9606):75-84.
(30) Walters BN, Thompson ME, Lee A, de Swiet M.
Blood pressure in the puerperium.
Clin Sci (Lond). 1986; 71(5):589-94.
(31) Wikland M, Lindblom B, Dahlström A, Haglid KG.
Structural and functional evidence for the denervation of human myometrium during
pregnancy.
Obstet Gynecol. 1984 ; 64(4):503-9.
(32) Khong TY, Tee JH, Kelly AJ.
Absence of innervation of the uteroplacental arteries in normal and abnormal human
pregnancies.
Gynecol Obstet Invest. 1997; 43(2):89-93.
(33) Brosens I, Pijnenborg R, Vercruysse L, Romero R.
The “great obstetrical syndromes” are associated with disorders of deep placentation
Am J Obstet Gynecol 2011; 204(3):193-201.
15
The “uterine reinnervation” view.
16