By Rodney T. Miller, M.D., Director of Immunohistochemistry  


Hydropic Degeneration vs. Partial Mole vs. Complete Mole

Microscopic examination of products of conception is a common task of many surgical pathologists. It is well known that on occasion it is difficult to distinguish degenerative changes in a nonmolar placenta (so-called “hydropic abortion”) from partial hydatidiform mole and complete hydatidiform mole. Attesting to this fact is a 1993 study by Howat and associates that found 5 of 7 pathologists could reach diagnostic agreement in only 70% (35 of 50) of hydropic placentas when employing histologic criteria alone, underscoring the imperfect nature of the histologic criteria employed for distinguishing these entities. Particular difficulties were encountered in the identification of partial mole. Another study by Takahashi employing restriction fragment-linked polymorphism analysis discovered that 20% of 10 hydropic placental specimens were misclassified using histopathologic criteria. Because of differences in DNA content, DNA ploidy analysis by flow cytometry can assist in the accurate classification of these specimens, although those techniques may not be readily available in some locations. This month, we review several papers that discuss the utility of immunohistochemical stains for Ki-67 and p57KIP2 in addressing this problem.


In a 1996 paper published in the International Journal of Gynecological Pathology, Schammel and Bocklage investigated the utility of P53, PCNA, and Ki-67 immunostains in distinguishing hydropic molar from nonmolar placentas. They studied 23 complete moles, 14 partial moles, 8 moles (not further classified), and 15 hydropic, nonmolar placentas. Diagnosis of each case was based on knowledge of both the histologic findings as well as the flow cytometry DNA ploidy data. PCNA and P53 expression did not allow accurate discrimination between molar and non-molar placentas. However, they found that Ki-67 immunostains differed significantly between the molar and nonmolar placentas. All partial or complete molar specimens contained at least one medium-sized villus that had >70% Ki-67-positive cytotrophoblastic cells rimming the periphery of the villi, whereas the maximum Ki-67 rim percentage (KRP) observed in the hydropic abortion specimens was 22%. We have employed KRP in this situation for several years, and we have found it useful in helping to separate such cases. It is important to adhere strictly to the definition of a “medium-sized villus”, which is defined by the authors as a villus that just spans a 200X microscopic field using an Olympus microscope (20x objective with 10x ocular lenses). If this definition is not strictly adhered to, there may be a tendency to overdiagnose molar specimens. Ki-67 immunostains did not allow distinction of partial moles from complete moles, since both of these categories had a high KRP.

Recently, Castrillon et al and later Fukunaga reported that immunostains for p57KIP2 protein, a cyclin-dependent kinase inhibitor, can assist in distinguishing diploid complete moles from diploid spontaneous hydropic abortions and triploid partial moles. Similar to normal placentas, diploid spontaneous hydropic abortions and triploid partial moles demonstrate a high frequency of expression of this protein in the nuclei of cytotrophoblasts, intermediate trophoblasts, and villous stromal cells (but not syncytiotrophoblasts). However, complete moles show absent or low-level immunostaining in these same elements (low-level immunostaining was defined as 1-10% positive cells). Normal decidual stromal cells are strongly stained, serving as convenient internal positive controls.


By combining Ki-67 with p57KIP2 immunostaining, it is possible to accurately classify many of these specimens, using only these 2 immunostains. Hydropic abortions show a low Ki-67 KRP, but demonstrate high expression of p57KIP2. Partial moles show a high KRP with high expression of p57KIP2, and complete moles show high KRP with low expression of p57KIP2. These results are summarized in the table below.


Obviously Ki-67 has been available for many years in the ProPath Immunohistochemistry Laboratory, but p57KIP2 is also now available for those who want to take advantage of this antibody to address this common problem in surgical pathology.  Hydrop. Degen Partial Mole Comp. Mole H&E Ki-67 p57 Ki-67 Rim% (med size villi) p57 in cytotrophoblasts & villous stromal cells Hydropic Degen Low (<22%) High Partial Mole High (>70%) High Complete Mole High (>70%) Low


1. Howat AJ, Beck S, Fox H et al: Can Histopathologists Reliably Diagnose Molar Malignancy? Journal of Clinical Pathology 46: 599-602, 1993.

2. Takahashi H, Kanazawa K, Ikarashi T et al: Discrepancy in the Diagnoses of Hydatidiform Mole by Macroscopic and Microscopic Findings and the Deoxyribonucleic Acid Fingerprint Method. American Journal of Obstetrics and Gynecology 163:112-113, 1990.

3. Schammel DP, Bocklage T: p53, PCNA, and Ki-67 In Hydropic Molar and Nonmolar Placentas: An Immunohistochemical Study. International Journal of Gynecological Pathology 15:158-166, 1996.

4. Castrillon DH, Sun D, Weremowicz S et al: Discrimination of Complete Hydatidiform Mole from Its Mimics by Immunohistochemistry of the Paternally Imprinted Gene Product p57KIP2. American Journal of Surgical Pathology 25 (10): 1225-1230, 2001.

5. Fukunaga M: Immunohistochemical Characterization of p57KIP2 Expression in Early Hydatidiform Moles. Human Pathology 33: 1188-1192, 2002


Date of last revision: March 2003.