Papillary proliferation of the endometrium (PPE) without cytologic atypia is uncommon and has only been studied in detail by Lehman and Hart in 2001. On histologic examination, PPE ranges from simple papillae with fibrovascular cores, often involving the surface of endometrial polyps, to complex intracystic proliferations; some consider the latter to be analogous to nonatypical complex hyperplasia. To further characterize PPE, with emphasis on the risk of and features associated with concurrent or subsequent neoplasia, the clinicopathologic features of 59 cases without cytologic atypia were studied. The cases were classified into 2 groups according to the degree of architectural complexity and extent of proliferation. Group 1 consisted of those with localized simple papillae. Simple papillae were defined as those with short, predominantly nonbranching stalks; those with occasional secondary branches and/or detached papillae were also included in this group. Localized proliferations were those with 1 or 2 foci involving the surface or the subjacent glands of polyps or nonpolypoid endometrium. Group 2 consisted of those with complex papillae and/or those with diffuse and crowded intracystic papillae. Complex papillae were those with either short or long stalks, with frequent secondary and complex branches. Diffuse proliferation was defined as presence of 3 or more foci within a specimen or involvement of >50% of the endometrial polyp by simple or complex PPE. Any coexistent or subsequent hyperplasia of conventional type (World Health Organization classification) or adenocarcinoma was recorded. The age of patients ranged from 23 to 82 years (median, 53 y); 36 (61%) were postmenopausal. The majority presented with abnormal vaginal bleeding. Sixteen patients (27%) were receiving hormonal preparations including 5 who were treated with a progestogen for preexisting endometrial hyperplasia or low-grade endometrioid adenocarcinoma. The histologic diagnosis of PPE was made in 49 biopsies and in 10 hysterectomy specimens. Thirty-six cases (61%) were classified as group 1 and 23 (39%) as group 2. In 47 cases (80%), there was a coexisting endometrial polyp, 39 (66%) of which were involved by the PPE. Fifty-three cases (90%) had coexisting epithelial metaplastic changes, 41 (77%) of which were involved by the PPE. The most common type of metaplasia was mucinous (41 of 59 cases, or 69%). Follow-up information was known for 46 patients (78%). Coexistent or subsequent nonatypical and atypical hyperplasia was found in 8 (17%) and 6 cases (13%), respectively. In 6 of the 46 cases (13%), a low-grade endometrioid adenocarcinoma was present either in the original specimen or during follow-up. In contrast to group 1 PPE, those with group 2 features were significantly associated with concurrent or subsequent premalignant lesions (nonatypical and atypical hyperplasia) or carcinoma (P<0.0001). This study indicates that localized and architecturally simple PPEs confined to a completely removed polyp are usually associated with a benign outcome and may be appropriately labeled as "benign papillary proliferation of the endometrium." Lesions with architecturally complex papillae, especially when extensive, have an increased risk of concurrent or subsequent endometrial hyperplasia and carcinoma and should probably be regarded as analogous to atypical complex hyperplasia, and the term "complex papillary hyperplasia" is appropriate. As the distinction between simple and complex PPE may be difficult in small endometrial aspirational samples, consideration for curettage should be given to ascertain whether the lesion has been completely removed.