Introduction

Acute and chronic/occult clinical umbilical cord (UC) compromise (compression, true knot, prolapse) are common obstetric complications contributing to perinatal morbidity [1]. Less frequently, they can be life-threatening to the fetus [2]. In fact, UC complications may be the second most common cause of stillbirth after placental insufficiency, particularly around term [3–6]. However, most fetal deaths are caused by a variety of placental pathologies, including UC compromise [7, 8].

Pathomorphologic confirmation of clinical UC compromise on post-mortem or placental examination can be elusive, especially in cases of entanglements and prolapse [9]. In rare cases, a clinically unsuspected cord encirclement can be retrospectively diagnosed on perinatal autopsy and placental examination [10]. Gross UC abnormalities may also be associated with stillbirth, abnormal perinatal outcome, fetal growth restriction, nonreassuring fetal heart rate tracings (particularly atypical variable decelerations in the first stage of labor), meconium-stained amniotic fluid, increased rate of labor induction, and emergency cesarean section [11–18]. Most authors relate clinical risk factors and placental pathomorphological features only to selected UC abnormalities [19] such as abnormal UC length [20, 21] or diameter [4, 13], abnormal coiling [4, 12, 14, 16, 22–28], prolapse [29, 30], entanglement [21, 25], edema [31], abnormal insertion [3, 4, 32], true knots [33], stricture [34, 35], ulcer [10, 19], varix, aneurysms, and pseudocysts [3].

The impact of acute and chronic UC compromise or abnormal morphology (such as abnormal insertion or morphological lesions) on clinical outcome and other placental pathology is largely unknown or misunderstood. Although it is known that obstructive cord lesions are associated with fetal thrombotic vasculopathy (FTV) [18, 36], little is known about the results of comprehensive placental examination in cases of UC complications. This analysis assesses the usefulness and limitations of diagnosing placental villous sequelae associated with UC compromise and gross morphological UC abnormalities.

Material and methods

The author examined 5634 consecutive placentas from pregnancies with a gestational age at birth ≥21 weeks. These examinations were performed during the period 1994–2013 in four tertiary care centers: Department of Pathology, University of Cincinnati Medical Center, Cincinnati, Ohio; Department of Pathology, Sheffield Children’s NHS Trust, Sheffield, Great Britain; Department of Anatomical Pathology, Canterbury Health Laboratories, Christchurch, New Zealand; and Division of Pathology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio. Placentas were submitted for examination at the discretion of obstetricians for the following reasons: abnormal gross placental features at delivery, high-risk pregnancy, operative delivery, poor obstetric outcome, and poor condition of the neonate. Placental examinations were performed according to generally accepted criteria [1, 37, 38]. The definitions of placental lesions have been given previously [39, 40], particularly of those that are not yet in general use. Definitions of some UC lesions are listed in Table 1.

Table 1 Definitions of morphological lesions of the umbilical cord
Full size table

Four groups of placentas were compared retrospectively (Table 2):

  1. Group 1.

    —182 placentas from pregnancies with only clinical features of UC compromise

  2. Group 2.

    —1355 placentas with only morphological UC abnormalities (UC itself or its abnormal insertion) (Fig. 1)

  3. Group 3.

    —152 placentas with at least one phenotype from group 1 and one from group 2

  4. Group 4.

    —3945 remaining placentas with no clinical or pathological UC-related phenotypes (control group)

Table 2 Composition of four studied groups
Full size table
Fig. 1
figure 1

Morphological abnormalities of umbilical cord (group 2). A. Battledore placenta (marginal insertion). B. Velamentous insertion. C. Furcate insertion. D. Chordae (amniotic webs). E. Massive edema. F. Thin. G. Uncoiled H. Hypercoiled. I. Long. J. Absent. K. Periumbilical stricture. L. Torsion. M. Ulcers. N. Amniotic band. O. Hypoplastic artery. P. Fusion of umbilical arteries. Q. Remnant of second umbilical vein. R. Single umbilical artery. S. Varix of umbilical vein. T. Vessel unprotected by Wharton jelly. U. Aneurysm of umbilical artery. V. W. Hematoma. Amniotic cyst. X. Hemangioma. Y. Necrotic media of umbilical vein. Z. Candida funisitis. AA. Necrotizing subacute (barber pole) funisitis

Full size image

Differences among the groups were analyzed by ANOVA or χ 2 with 3 degrees of freedom and 0.0007353 Bonferroni adjustment for 68 clinical and placental phenotypes analyzed.

Results

Statistically significant differences existed in 31 phenotypes: 13 clinical (Table 3) and 18 placental (Table 4). The most common phenotypes in the first three groups of placentas were as follows:

  1. Group 1.

    —2 clinical phenotypes (oligohydramnios and abnormal fetal heart rate tracing) and 0 placental phenotypes

  2. Group 2.

    —4 clinical phenotypes (abnormal umbilical artery Dopplers, nonmacerated stillbirth, multiple pregnancy, and fetal growth restriction) and 2 placental phenotypes (clusters of maternal floor multinucleate giant cells and amnion nodosum)

  3. Group 3.

    —7 clinical phenotypes (gestational hypertension, polyhydramnios, induction of labor, cesarean section, macerated stillbirth, congenital malformations, and abnormal third stage of labor) and 16 placental phenotypes

Table 3 Statistically significant clinical differences among groups 1–4
Full size table
Table 4 Statistically significant placental differences among groups 1–4
Full size table

Therefore, the frequencies of the most common abnormalities were as follows: group 3 > group 2 > group 1. In the three groups, frequencies for clinical phenotypes were 7, 4, and 2 (respectively) and frequencies for histopathological placental phenotypes were 16, 2, and 0 (respectively).

Group 4 (the control group) contained the least frequent statistically significant variables as follows:

  • 10 of the 13 least frequent abnormal clinical variables (gestational hypertension, oligohydramnios, polyhydramnios, abnormal antepartum fetal heart rate tracing, abnormal Dopplers, induction of labor, cesarean section, nonmacerated stillbirth, and multiple pregnancy), and

  • 11 of the 18 least frequent placental variables (normoblastemia of fetal blood, deep (decidual) meconium penetration, post uterine chronic hypoxic placental injury, chorionic microcysts of the chorionic disk, excessive number of extravillous trophoblasts, choriodecidual hemosiderosis, luminal vascular abnormalities of chorionic villi, diffuse villous fibrosis, fetal vascular thrombi, clusters of avascular chorionic villi, periarterial stem edema, and dilatation of fetal veins (umbilical, chorionic disk, and stem villi)).

In the remaining 37 variables analyzed (clinical or placental), no statistically significant differences existed after Bonferroni correction (not included in Tables 3 or 4).

Discussion

Frequencies of abnormal clinical phenotypes occurred mostly in single digits in group 1, except for abnormal fetal heart rate, induction of labor, and cesarean section. Fetal heart rate abnormalities are common in cord complications, particularly in cases of prolonged pregnancy [44]. However, gestational age did not differ among the groups studied. Also, induction of labor was reported to be associated with cord prolapse [30]. Therefore, it is not only possible but likely that the abnormal results of fetal heart monitoring affected the rates of labor induction and cesarean section [15, 32]. However, the clinical signs of UC compromise were not associated with increased pathomorphological placental abnormalities in group 1. Other authors have also reported that clinical UC complications are unlikely to be associated with abnormal villous lesions [45]. Our previous analysis showed that the full-term placentas were more likely to suffer from acute in utero hypoxia manifesting clinically with fetal distress (abnormal fetal heart rate tracings and thick meconium, possibly resulting from unrecognized UC compromise) and deep meconium penetration [6].

By comparison, the coexistence of abnormal clinical and morphological placental phenotypes (group 3) was often statistically significantly associated with the following:

  • Placental histological signs of decreased fetal blood flow (stasis-induced FTV) (Fig. 2),

  • Placental hypoxic lesions and patterns, and features of shallow placental implantation (Fig. 3).

Pathophysiologically, fetal vascular obstructive lesions result from the Virchow’s triad: (1) stasis (UC compromise), (2) hypercoagulability (poorly controlled diabetes mellitus, autoimmune conditions, maternal or fetal thrombophilias), and (3) vascular damage within the fetal circulation of the placenta (inflammation, meconium toxicity, hemorrhagic endovasculitis) [46]. FTV is associated with potentially obstructive lesions of the cord and with oligohydramnios (the latter is not in our material) [47].

Our results thus strengthen the evidence that the coexistence of clinical and placental UC abnormalities may indeed play a part in perinatal morbidity and, particularly, in the rate of macerated stillbirth [5, 7, 48, 49]. Of these stillbirths, a proportion shows features of FTV, which are more likely to be seen in unexplained stillbirths [50]. Various UC abnormalities and placental clusters of sclerotic/hemosiderotic chorionic villi were most common in macerated stillbirths [40]. Also, several histological features of prolonged umbilical cord compromise [51] were encountered among the statistically significant pathomorphological findings in group 3. However, in addition to thrombotic lesions (Fig. 2), several hypoxic lesions (Fig. 3) were also common in this group. This finding agrees with the results of those authors who claim that mixed placental pathologies are frequently seen in third trimester stillbirths [11, 36, 52]. We have also found recently that placental hypoxic overlap lesions associated with clinical complications of pregnancy predispose to thrombotic lesions, some of which are most likely stasis-induced [53].

Fig. 2
figure 2

Statistically significant patterns and lesions related to decreased or absent fetal blood flow (objective magnifications). a Luminal vascular abnormalities of stem villi (×20). b Increased extracellular matrix of chorionic villi, stillbirth, 23 weeks of pregnancy (×20). c Dilatation of stem vein (×4). d Thrombi in a stem vessel (×20). e Cluster of avascular chorionic villi (×10). f Villous hemosiderosis in lobular distribution with calcified microthrombi (×20). G. Intimal cushion in stem vein (×10). h Stem perivascular edema and obliterative endarteritis (×20)

Full size image
Fig. 3
figure 3

Statistically significant hypoxic placental lesions and patterns (objective magnifications). a Normoblasts in fetal blood (×40). b Meconium macrophages in amniotic mesenchyme and decidua (arrows) (×20). c Laminar decidual necrosis of membranes (×10). d Diffuse post uterine pattern of chronic hypoxic placental injury (22 weeks) (×10). e Excessive amount of extravillous trophoblasts in chorionic disk (×4). f Microscopic chorionic pseudocyst of a cell island (×4). g Clusters of maternal floor multinucleate trophoblastic giant cells and decidual hemosiderosis (arrows) (×20). h Intervillous thrombus (dissecting microscopy ×2)

Full size image

Abnormal pathomorphological UC features without clinical signs of UC compromise (group 2) were associated at a moderate frequency with abnormal clinical and placental findings—except at the highest congenital malformation rate or in the presence of clusters of maternal floor multinucleate giant cells. Therefore, the macroscopic morphological features of the UC alone cannot reliably predict the presence or absence of histological placental lesions [54].

The problem of UC-related complications and associated placental pathology is also important because of its medicolegal aspect. Although infrequent, litigation may occur following the poor outcome associated with UC accidents [5]. Of 158 cases subjected to litigation because of cerebral palsy, 63 % showed clinical and/or pathologic evidence of UC obstruction [55]. In only a portion of these cases could the litigation have been averted by prenatal screening for certain important UC lesions [5, 12, 19, 56]. Litigation might also be avoided by screening for fetal heart rate abnormalities to identify fetuses that might suffer in utero hypoxia and postnatal acidosis. All efforts at prenatal visualization of UC and/or placental lesions are thus warranted, since fetal thrombophilia alone is not usually associated with fetal vascular lesions of the placenta (although it may be an underlying risk factor) [46, 57]. Also, oligohydramnios predisposing to UC compression can be relieved by amnioinfusion [45].

Finally, comparing our results with those of other reports may be difficult because various authors may use different diagnostic criteria for placental lesions, particularly in groups labeled “other placental abnormalities.” Some authors exclude FTV as a cause of death if UC blood flow restriction is present [7]. But others take an opposite view and stress that the placental histology provides useful information about fetal acidemia with low-risk term pregnancy in clinically undetectable cases [58]. Previous studies that linked abnormal UC coiling with clinical outcome are generally too small and/or selective to allow meaningful conclusions or to apply to low-risk populations [59].

In conclusion, the clinical signs of UC compromise alone cannot and do not predict the presence or absence of histological placental lesions because they usually occur acutely and too close to delivery for the histological features of FTV to develop. Although UC compromise is an important factor in perinatal morbidity and mortality, confirmation of clinical UC compromise should not be expected on placental examination if no morphological UC abnormality or abnormal UC insertion has been found. However, when clinical UC compromise and morphological UC abnormality or abnormal UC insertion do coexist, they significantly strengthen the case that UC compromise indeed plays a role in perinatal morbidity and mortality.