10 Involution of Placental Site: Retained Placenta

(1)

10

Involution of Placental Site:

Retained Placenta

trium against the noncontractile placenta. One may spec- ulate that detachment might be easier when the placenta is more turgid and still ﬁlled with fetal blood, but it is apparently not the case. Walsh (1968) found that when the cord is clamped early and more fetal blood remains in the placenta, postpartum hemorrhage is signiﬁcantly more common than when this blood was allowed to drain into the fetus.

Immediately following delivery, the uterine surface becomes covered with fibrin and blood clot; the contraction of the uterus clamps the maternal vessels and this stops uterine bleeding. Ludwig (1971), and Ludwig and Metzger (1971) have made elegant electron-microscopic observations of this area postpartum. In their scanning electron-microscopic study, they speak of a “wallpaper- ing” of the endometrial surface with a delicate meshwork of cross-linked fibrin that includes deformed erythro- cytes. They had shown earlier that, with the first contraction after expulsion of the placenta, this fibrin is deposited and that it aids hemostasis. There is pronounced furrow- ing of the inner surface of the uterus following delivery that is bridged by the “tapestry” of fibrin. Mukaida et al.

(1975), who also investigated this topic, came to similar conclusions.

Friedländer (1870) found that most of the endometrium degenerates after birth and that regeneration takes place from the glands and stroma in the spongy layer of the endometrium; this process is largely completed by 4 weeks postpartum, at least in the area away from the implantation site. The endometrium of the implantation site itself is not reformed for several additional weeks after delivery. The controversy over what constitutes physiologic changes after birth did not cease until Williams (1931) published his comprehensive study of normal uteri that had been removed up to 4 months after delivery. He showed that the detachment of the placental site pro- ceeded with little inﬂammation or necrosis. The cessation of blood ﬂow through the endometrial vessels was largely accomplished by contraction of the uterus itself. In the

313

Involution of the Placental Site

Pathologists rarely obtain a postpartum uterus to enable a detailed study of the involutional changes that take place at the former site of implantation. Therefore, involution of the normal placental site has been studied by only a few investigators. Normally, the postpartum lochia contains the decidual remnants, including perhaps some of the degenerating remnants of the vasculature that had previously undergone the so-called physiologic changes of pregnancy. Only when significant postpartum hemorrhage occurs and hysterectomy is performed, is the pathologist asked to seek the cause of the bleeding. He or she may then find remains of villi, incompletely thrombosed vessels, placental polyps, and some degree of inflammatory reaction. Frisoli (1981) suggested that retained placental tissue is found in about 50% of such cases at curettage. The curettings and postpartum uteri are difficult tissues to study objectively because most pathologists have little experience with the normal, complex process of placental site involution. Williams (1931), in a classic paper, attempted to rectify this situation. His study should be read before any interpretation of such a postpartum uterus is undertaken. It must also be recognized that in 85% of normal, delivered placentas the decidua basalis shows foci of polymorphonuclear leu- kocyte infiltration (Schneider, 1970). These cells are part of an apparently normal process of implantation and are not considered an expression of deciduitis or infection. Nevertheless, postpartum hemorrhage and subinvolution are nearly always associated with significant inflammation, if only because the cervix remains open and patulous.

The placenta separates from the uterus in the decidua basalis, deep to Nitabuch’s fibrin layer. This fibrin layer is usually present in the floor of the delivered placenta but, contrary to some opinions, it does not serve as the cleav- age plane. Placental separation probably occurs largely because of the shearing action of the underlying myome-

(2)

314 10. Involution of Placental Site: Retained Placenta

absence of contraction, as in cases following former overdistention (e.g., in multiple pregnancy, hydramnios), these vessels may bleed extensively. This uterine atony presents a life-threatening situation that demands emer- gency attention. Normally then, these vessels are clamped by uterine contraction; they thrombose subsequently (Fig. 10.1), and the vascular placental site eventually becomes a mass of hyaline plaques (Fig. 10.2). The vessels

Figure 10.1. Poorly thrombosed uterine vessels of the placental site 2 weeks after delivery of a hydatidiform mole.

H&E ¥160.

Figure 10.2. Agglomeration of hyaline masses from placental site vessels 6 weeks after normal delivery. H&E ¥64.

become “organized” by the ingrowth of ﬁbrocytes and endothelium, and they are later recanalized. This process takes many weeks. One can then readily identify former implantation sites by the presence of these unusual hyalinized vessels and by the remains of some placental site giant cells. These residua of implantation have been demonstrated particularly impressively in several monkey species, in which they form macroscopically visible plaques (Bronson et al., 1972). Although the changes in monkeys are, in general, similar, they differ by involving usually both uterine sides, as their placenta is commonly bidiscoid. Moreover, they are much more massive and persist much longer than the approximately 7 weeks considered to be the normal for women. Also, some calciﬁca- tion occurs in the hyaline areas of simian placental site involution, which is not often the case in humans.

The placental site is then “exfoliated,” as Williams (1931) called it, rather than being absorbed into the myometrial tissue. At least the portions that are central to the myometrium are undermined. Williams considered that the decidual site is undermined by regrowth and by extension and downgrowth of endometrium that origi- nates from remaining endometrial glands and stroma.

He described it to be a “conservative” effort of nature to thus dispose of the thrombosed vessels. It must be borne in mind, however, that it applies only to the decidual portions of the placental site; the myometrial part has a much different regressive curve. One mystery of postpartum uterine involution is the rapidity with which the muscle mass is reduced. The average uterus weighs about 1000 g after term delivery and, largely through the dissolution of cytoplasm, it shrinks to less than 100 g, usually within 2 months. An appreciation of this massive involution of myometrium postpartum is especially important in our understanding of the “healing” of cesarean section incisions. Virtually no myometrium repairs the incisional defect. Only a thin ﬁbrous scar approximates the muscle

(3)

layers. Thus, in subsequent pregnancies the probability of dehiscence exists (see Chapter 9).

The nature of the vascular changes that take place at the placental site postpartum was the topic of a study by Maher (1959). He observed that the elastic membranes of placental site vessels lose their staining qualities for 25 to 30 days. After that, the elastica is restored, but there is persistence of abnormalities in the pattern of the elastic membrane for a long time. Maher described striking involutional and regenerative changes of the placental site vessels and compared them with the much less prominent effect of pregnancy in those portions of the uterus and fallopian tube that are remote from the implantation site.

Anderson and Davis (1968) made the next major con- tribution to the study of placental site involution. They believed that vascular constriction and occlusion cause necrosis and sloughing of the former decidua basalis.

They confirmed the findings of Sharman (1953), which were made from 626 postpartum biopsies and 11 uteri, removed 1 to 9 months after delivery. Anderson and Davis suggested that the absence of necrosis observed by Williams (1931) was due to his lack of appropriate mate- rial from days 4 to 7 postpartum. They corrected this deficiency and measured the placental site. Immediately after birth of the infant, the placental site was 18 cm; after the placenta detached, it measured 9 cm (largely because of uterine contraction), and it was 4 mm in thickness.

At 8 days postpartum, the placental site measured only 4.5 cm in diameter, and by 8 weeks it had shrunken to 2 cm. The latest time at which a formerly normal placental site was visible macroscopically was 11 weeks after delivery. In contrast to Williams’ opinion, a 6-mm-deep area of necrosis of the placental site, extending onto the myometrium, was observed from 2 to 6 days. A mixed “granulocy- tic and mononuclear inflammatory infiltrate occurred in the slough and the viable endomyometrium.” Anderson and Davis described the “physiologic adenomyosis” that occurs in the superficial myometrium, particularly in the center of the former placental site, and they delineated the development of pleomorphic nucleated cells of the endometrial glands. In their experience, placental site giant cells persisted for several days, were confined to the

superﬁcial layers, and did not extend deeply into the myometrium. Villi were found in four early specimens from apparently uncomplicated pregnancies. None of these elements normally persisted past 4 weeks. Vessels were described as thrombosed and showing “endophlebitis.”

Endothelial proliferation was a particularly prominent feature in the veins. Toward the end of the ﬁrst week, the veins showed the beginning of organization of clots;

arterial thrombosis was not a prominent feature. Because these normal changes are often misinterpreted as possi- bly being pathologic, a summary of the features of these sequential events is presented in Table 10.1.

Anderson and Davis (1968) were unable to resolve the nature of the placental site giant cells but favored a trophoblastic origin for these cellular elements. They stated emphatically that involution of the placental site does not lead to ﬁbrosis of the uterus. They insisted that the term subinvolution must be understood within the framework of these normal involutional processes.

Sharman (1953) found that mitoses of the endometrium first occurred 8 days after delivery. By 16 days, the endometrium was structurally intact again. He also observed an “appreciable number” of plasma cells to be present 6 weeks after birth in 50% of his cases (37% had plasma cells at 3 to 4 months postpartum). This result is surprising, as it implies infection in one half of postpartum patients. It must be reiterated, however, that the inflammatory reaction of the normal placental site involutional process is usually sparse, and in most women it is confined to the first few days. Moreover, this infiltrate is made up of granulocytes and macrophages; plasma cells appear only late in this process.

In step with the involution of placental bed trophoblast are the endocrine signals of pregnancy. Human chorionic gonadotropin (hCG) disappears within 2 weeks from the maternal serum completely, regulated largely by its half-life of 9 to 37 hours (Midgley & Jaffe, 1968; Jaffe et al., 1969). Human placental lactogen (hPL), on the other hand, having a shorter half-life of up to 30 minutes, disappears within a day (Geiger, 1973). The disappearance of other pregnancy markers is less well deﬁned (Tulchinsky & Ryan, 1980). Korhonen et al. (1997) studied the hCG decline in some detail in six women following

Table 10.1. Structural alterations in the postpartum placental site

Time Size (cm) Slough Glands Decidua Veins Arteries

Term 9 Hemorrhage Present Viable Hyaline Fibrinoid

1–3 d 7–8 Early necrosis Prominent Necrosis Thrombosis Obliterated 3–5 d 6 With reaction Atypia Necrosis Organization Hyalinized 5–8 d 4.5 Demarcated Increased Regression Organization Hyalinized

4–20 w 2 Hemosiderin Inactive Absent Recanalized Remnants of hyaline Source: From Anderson and Davis (1968), with permission.

(4)

316 10. Involution of Placental Site: Retained Placenta normal delivery. From levels of around 35,000 IU/mL at

the end of pregnancy, rapid decline leads to virtual elimi- nation (5 IU/mL) by 21 days after delivery, a-hCG having a much more rapid decline than b-hCG. Inasmuch as one may see extravillous trophoblast (X cells) for some time following delivery in the hyalinizing placental site, however, it is likely that some of the large quantity of major basic protein (MBP) that exists during gestation can be detected for longer periods, but measurements have not been made.

Subinvolution

At times the uterus does not involute properly, and the clinical diagnosis of “uterine subinvolution” is enter- tained. What are the pathologic equivalents of this condition? Khong and Khong (1993) are the most recent authors to have examined this phenomenon, with the evaluation of 169 specimens from “delayed postpartum hemorrhage” that followed a singleton delivery. They provided some reason why such studies should be divided into periods before and after 6 weeks postpartum, not an important issue for pathologists, we believe. Retained placental tissue was most common (45 cases), to be followed by “subinvolution,” that is, widely patent uter- oplacental arteries in 30 patients. Endometritis (seven patients) was uncommon and most of the other patients had essentially normal histologic ﬁndings. Placental tissue may have been retained because it resulted from placenta accreta, or because the fragmentation of a manually removed placenta was not appreciated at delivery, and a portion of a cotyledon was left behind.

Normally, the uterus shrinks rapidly, with dissolution of the myometrium. Within 2 weeks, its size reduces from about 1000 g to nearly 100 g, and the endometrium and placental site are cast off. With subinvolution this regression is delayed; the uterus remains boggy and edematous and may actually be retroﬂexed and congested. Often there is bleeding and some remaining placental tissue (“placental polyp”), or an infection may cause uterine subinvolution. Thorsteinsson and Kempers (1970) studied the records of 148 patients with delayed postpartum bleeding, occuring in 1% of deliveries. They found the bleeding most commonly resulted from subinvolution of the placental site. Curettage was done in 94 patients, and 78 were found to harbor microscopic fragments of placental tissue; seven additional patients had grossly visible placental remains. Curettage cured most patients. It is interesting to note here also that most of the patients (135) were said to have had normal placentas at delivery.

According to Stamm (1961), incompletely removed placentas have been the cause of litigation. For that reason, others had suggested that placentas be injected for completeness. Stamm, however, believed that the

evaluation for completeness of the placenta through the injection of milk is inaccurate, and that it should not become a routine procedure. Tears in the placenta often occur, which makes it difﬁcult to ascertain with certainty that removal was complete. Stamm suggested that manual exploration of the uterus is necessary when one is uncer- tain whether a placenta has been removed completely.

Placental Polyps

Placental polyps were described in three patients by Hoberman and his colleagues (1963), who also reviewed the sparse English-language literature on this topic. Pla- cental polyps are usually made up of remnants of villous tissue that have become encased in layered clot. At times, one ﬁnds that the villi are directly attached to the myometrium; they are thus focal placentas accretas, but because of degenerative changes this diagnosis is often impossible to verify.

When these “polyps” are removed, the symptoms usually abate. If they are not removed, life-threatening bleeding may occur. This possibility was emphasized in the study of placental polyps undertaken by Dyer and Bradburn (1971). They decried that this important com- plication of the puerperium is rarely mentioned in modern textbooks. Dyer and Bradburn described the “immediate type” of polyp, which occurs during the ﬁrst month after delivery, and the “delayed type,” where the tissue may be discovered months or even years later. They acknowl- edged that accretas may be part of the problem but believed that other, unrecognized factors play a role as well. Finally, they speculated that such retained fragments are the possible sites of future trophoblastic tumors. In our opinion, there is little evidence for the latter specula- tion, even though the placental site trophoblastic tumor has become a delineated entity (see Chapter 23). It is true, however, that an exaggerated placental site may pose as tumor and the differential diagnosis is not always easy (Benirschke, 1997). Philippe et al. (1968) claimed that a deﬁcient arterial occlusion at the site of involution is a probable cause of postpartum hemorrhage. After studying and depicting the normal vascular occlusions, these authors showed hugely distended vessels with poorly organized thrombi, as well as increased hyaline masses, to be a cause of the bleeding that may take place later, when early postpartum hemorrhage has occurred.

Philippe and his colleagues speculated that these vascular changes may result from inadequate hormonal effects on these vessels during the immediate postpartum period.

They also suggested that therapy with 17-hydroxypro- gesterone caproate may resolve the condition and that hysterectomy is not always indicated. Rutherford and Hertig (1945) showed placental polyps in three postpartum patients who had neither placental remains nor

(5)

inflammation, but had hugely dilated, improperly occluded, placental site blood vessels. Perhaps this situation constitutes the “true” subinvolution, a process whose etiology is yet to be clarified. Most recently, late bleeding has been associated with the discovery of hyalinized masses (placental-site nodules); when stained with anti- bodies to hPL, these masses were found to contain posi- tively staining placental-site giant cells (Young et al., 1990) (see Chapter 23). The authors cautioned that these masses are benign remnants of placentation that must not be mistaken for tumor. Suffice it to say, retained placental villi are not always present in patients who experience postpartum hemorrhage. Incomplete occlusion of vessels alone may be the cause of subinvolution, as also may be an idiopathic failure of trophoblastic placental site cells to involute.

In addition to these features of subinvolution, there is the postpartum endometritis, “puerperal fever,” an infection caused by a variety of microorganisms. Its prototype, however, is caused by group A streptococci. The histologic features may be similar to those seen with other subinvolutional events, although there are additionally pronounced acute inﬂammation, inﬁltration with plasma cells, phlebothrombosis, and often colonies of bacterial organisms. All are life-threatening processes that may also lead to pulmonary embolism. Various aspects of a placental polyp, symptomatic 9 years after abortion of the last known pregnancy, were reported by Lawrence et al.

(1988). At that time the patient had undergone steriliza- tion. The polyp was composed mostly of “necrotic and hyalinized villi, without identiﬁable trophoblast.” A few syncytial cells exhibited hCG immunohistochemically, and abundant X cells were present with the hPL reaction.

Regrettably, the authors did not report on the status of the ovaries so as to rule out a possible recent pregnancy.

They also reviewed the few case reports and case series

Figure 10.3. Myometrial remnants of a former placental site with hyalinization, 2 years after normal delivery and death for unrelated reasons. H&E ¥160.

from the literature, which included a placental polyp 21 years after pregnancy.

In our experience, the hyaline scars of former placental sites may persist in the uterus for much longer than has been indicated in most of the literature cited here, much as the corpus albicans persists for years in the ovary. The myometrial remnants of former placental site vessels are shown in Figures 10.3 and 10.4, assuredly 2 years after the

Figure 10.4. Finding similar to that in Figure 10.3 but deeper in the myometrium. H&E ¥160.

(6)

last pregnancy. Figure 10.5 shows the uterus of a patient with retained placental tissue who died 16 weeks postpartum. Numerous trophoblastic cells can be seen throughout the myometrium of the placental site. Most, but not all, villi were completely necrotic in this patient (Fig. 10.6).

Somewhat related to these rare pathological ﬁndings are the observations made by Craven and Ward (1997)

Figure 10.6. Live and necrotic villi forming the placental polyp of the uterus shown in Figure 10.5, 16 weeks after delivery. H&E ¥64.

Figure 10.5. Composite of implantation site with endometrium above. This patient had retained placental tissue for 16 weeks (see Fig. 10.6), and many of the placental site giant cells (dark)

persist deep in the myometrium. No vascular thrombosis had occurred. H&E ¥16.

on placental bed biopsy. This was done after cesarean section delivery of a multiparous patient for a study of preeclampsia. Microscopically, squamous epithelial masses, undoubtedly of vernix origin, were embedded in endomyometrium and maternal vessels. There was intense inﬂammation also present, and the authors considered this to be the result of membrane rupture.

(7)

Involution of a Remaining Placenta

Involution of a remaining placenta is conveniently dealt with at this point because some of its features resemble the findings of retained placental tissue and placental polyps. When fetectomy had been performed in rhesus monkeys between the 10th and 23rd weeks of pregnancy, it was alleged that the placenta was later delivered “on time” (Van Wagenen & Newton, 1943), that is, at term for this species. Lewis and Hertz (1966) found that most of the monkeys developed degenerated placentas when fetectomy was done between the 14th and 35th days of gestation. They had expected to find hydatid swelling, perhaps molar transformation, but it did not occur. In similar, larger experiments, Panigel and Myers (1972) also did not find hydatid changes following either fetectomy or ligation of the vessels that extend to the monkey’s second placental disk (the “bridging vessels”). They observed that some villi remained structurally normal for several weeks, although their fetal capillaries and red blood cells soon disintegrated. The maintenance of some fine structural detail in these surviving villi is indeed remarkable. The villi become usually more compacted, rather than hydropic. Detaching a disk of the placenta and thereby interrupting the maternal perfusion in monkeys, however, led to placental infarction (Myers

& Panigel, 1973). In many ways the villous changes found after prolonged retention of the placenta resemble those encountered following fetal demise (Fox, 1968).

They are initiated by syncytial knotting, stromal ﬁbrosis with loss of capillaries, and ﬁbrosis of major stem vessels.

Calciﬁcation and complete villous hyalinization are end stages.

When investigation of the factors that initiate labor required reexamination of the role of fetectomy, specifi - cally as it relates to the role played by the fetal adrenal and pituitary glands, Lanman et al. (1975) restudied the placental changes following fetectomy in the rhesus monkey. In the 13 animals examined by these authors, fetectomy was performed between 65 and 143 days, term being 168 days. Nine of the placentas remained in situ beyond the expected date of delivery, two delivered near term, and two delivered prematurely. Histologic study revealed empty vessels, stromal fibrosis of villi, hyalinization, and calcification 17 to 108 days after fetectomy. X- cell hyperplasia was prominent in two placentas. The placental weights were similar to what was expected at the time of fetectomy, but they were much smaller than expected for the gestational age of the placenta. Some- what different results were obtained by Nathanielsz et al.

(1992). These authors were unable to obtain the placentas for dissection but, from the endocrine changes that occurred after fetectomy, they decided that a functional fetal circulation in necessary for normal gestational control.

The anatomic results just reviewed have also been the experience with retained placentas in women whose fetus has died or when only one fetus of twins survives. The placental tissue remains structurally intact for a long time.

It first loses its fetal vasculature and undergoes fibrosis, and much fibrin accumulates, particularly in the intervillous space. The trophoblast continues to be perfused by the maternal circulation and so survives longer. The placenta eventually atrophies and comes to resemble an infarct. The result is not a true infarct, because an infarct occurs only after interruption of the maternal intervillous circulation.

References

Anderson, W.R. and Davis, J.: Placental site involution. Amer. J.

Obstet. Gynecol. 102:23–33, 1968.

Benirschke, K.: Critical commentary to: Placental site trophoblastic tumor (PSTT) initially misdiagnosed as cervical car- cinoma (by L.C. Horn, et al.) Pathol. Res. Pract. 193:233–234, 1997.

Bronson, R., Volk, T.L. and Ruebner, B.H.: Involution of placental site and corpus luteum in the monkey. Amer. J. Obstet.

Gynecol. 113:70–75, 1972.

Craven, C.M. and Ward, K.: Premature rupture of the amniotic membranes diagnosed by placental bed biopsy. Arch. Pathol.

Lab. Med. 121:167–168, 1997.

Dyer, I. and Bradburn, D.M.: An inquiry into the etiology of placental polyps. Amer. J. Obstet. Gynecol. 109:858–867, 1971.

Fox, H.: Morphological changes in the human placenta follow- ing fetal death. J. Obstetr. Gynaecol. Brit. Cwlth. 75:839–843, 1968.

Friedländer, C.: Physiologisch-anatomische Untersuchungen über den Uterus. Simmel and Co., Leipzig, 1870.

Frisoli, G.: Physiology and Pathology of the Puerperium. Chapter 93 in, Principles and Practice of Obstetrics & Perinatology. L.

Iffy and H.A. Kaminetzky, eds. John Wiley, New York, 1981.

Geiger, W.: Radioimmunological determination of human chorionic gonadotropin, human placental lactogen, growth hormone and thyrotropin in the serum of mother and child during the early puerperium. Horm. Metab. Res. 5:342–346, 1973.

Hoberman, L.K., Hawkinson, J.A. and Beecham, C.T.: Placental polyp: report of 3 cases. Obstet. Gynecol. 22:25–29, 1963.

Jaffe, R.B., Lee, P.A. and Midgley, A.R.: Serum gonadotropins before, at the inception of, and following human pregnancy.

J. Clin. Endocrinol. Metabol. 29:1281–1283, 1969.

Khong, T.Y. and Khong, T.K.: Delayed postpartum hemorrhage:

a morphologic study of causes and their relation to other pregnancy disorders. Obstet. Gynecol. 82:17–22, 1993.

Korhonen, J., Alfthan, H., Ylöstalo, P., Veldhuis, J. and Stenman, U.-H.: Disappearance of human chorionic gonadotropin and its a- and b-subunits after term pregnancy. Clin. Chem.

43:2155–2163, 1997.

Lanman, J.T., Mitsudo, S.M., Brinson, A.O. and Thau, R.B.:

Fetectomy in monkeys (Macaca mulatta); retention of the placenta past term. Biol. Reprod. 12:522–525, 1975.

(8)

Lawrence, W.D., Qureshi, F. and Bonakdar, M.I.: “Placental polyp”: light microscopic and immunohistochemical observa- tions. Hum. Pathol. 19:1467–1470, 1988.

Lewis, J. and Hertz, R.: Effects of early embryectomy and hormonal therapy on the fate of the placenta in pregnant rhesus monkeys. Proc. Soc. Exp. Biol. Med. 123:805–809, 1966.

Ludwig, H.: Surface structure of the human term placenta and of the uterine wall post partum in the screen scan electron microscope. Amer. J. Obstet. Gynecol. 111:328–344, 1971.

Ludwig, H. and Metzger, H.: Das uterine Placentarbett post partum im Rasterelektronenmikroskop, zugleich ein Beitrag der extravasalen Fibrinbildung. Arch. Gynäkol. 210:251–266, 1971.

Maher, J.A.: Morphologic and histochemical changes in post- partum uterine blood vessels. AMA Arch. Pathol. 67:175–180, 1959.

Midgley, A.R. and Jaffe, R.B.: Regulation of human gonadotropins. II. Disappearance of human hCG following delivery. J.

Clin. Endocrinol. Metabol. 28:1712–1718, 1968.

Mukaida, T., Yoshida, K. and Soma, H.: A surface ultrastructural study of the placental separation site. J. Clin. Electron Microsc.

8:5–6, 1975.

Myers, R.E. and Panigel, M.: Experimental placental detachment in the rhesus monkey: Changes in villous ultrastructure.

J. Med. Primatol. 2:170–189, 1973.

Nathanielsz, P.W., Figueroa, J.P. and Honnebier, M.B.O.M.: In the rhesus monkey placental retention after fetectomy at 121 to 130 days’ gestation outlasts the normal duration of preg- nancy. Amer. J. Obstet. Gynecol. 166:1529–1535, 1992.

Panigel, M. and Myers, R.E.: Histological and ultrastructural changes in rhesus monkey placenta following interruption of

fetal placental circulation by fetectomy or interplacental umbilical vessel ligation. Acta Anat. (Basel) 81:481–506, 1972.

Philippe, E., Ritter, J., Renaud, R., Dellenbach, P., Fonck-Cussac, Y. and Gandar, R.: Les métrorrhagies tardives du post-partum par anomalie d’involution des artères utéro-placentaires. Rev.

Fr. Gynécol. 63:255–262, 1968.

Rutherford, R.N. and Hertig, A.T.: Noninvolution of the placen- tal site. Amer. J. Obstet. Gynecol. 49:378–384, 1945.

Schneider, L.: Über Vorkommen und Bedeutung leukocytärer Inﬁltrate im Ablösungsbereich der spontan geborenen Pla- centa. Arch. Gynäkol. 208:247–254, 1970.

Sharman, A.: Post-partum regeneration of the human endome- trium. J. Anat. 87:1–10, 1953.

Stamm, H.: Alte und neue Probleme bei Plazentarpolypen.

Gynaecologia 151:252–260, 1961.

Thorsteinsson, V.T. and Kempers, R.D.: Delayed postpartum bleeding. Amer. J. Obstet. Gynecol. 107:565–571, 1970.

Tulchinsky, D. and Ryan, K.J.: Maternal-Fetal Endocrinology.

W.B. Saunders, Philadelphia, 1980.

van Wagenen, G. and Newton, W.H.: Pregnancy in the monkey after removal of the fetus. Surg. Gynecol. Obstet. 77:539–549, 1943.

Walsh, S.Z.: Maternal effects of early and late clamping of the umbilical cord. Lancet 1:996–997, 1968.

Williams, J.W.: Regeneration of the uterine mucosa after delivery, with especial reference to the placental site. Amer. J.

Obstet. Gynecol. 22:664–696, 793–796, 1931.

Young, R.H., Kurman, R.J. and Scully, R.E.: Placental site nodules and plaques. An analysis of 20 cases. Amer. J. Surg.

Pathol. 14:1001–1009, 1990.