Pathologic risk factors for predicting residual disease in subsequent hysterectomy following LEEP conization.

The loop electrosurgical excision procedure (LEEP) is widely used for cervical conization. It is unclear if margin status is the single most important factor for predicting residual or recurrent disease. We therefore sought to determine supplementary histologic factors that can be used to predict residual disease in subsequent hysterectomy specimens in order to select patients at high risk for residual and recurrent disease.The study group comprised 172 women who had undergone LEEP conization and subsequent hysterectomy. Histological findings were assessed and compared in patients with and without residual disease.A univariate analysis revealed that percentage of involved pieces (51% vs. 32%, p<0.001), percentage of involved mucosal length (30% vs. 17%, p=0.001), positive endocervical biopsy (77% vs. 45%, p<0.001), bulbous growth (73% vs. 44%, p=0.001), pseudogland or central necrosis (73% vs. 26%, p<0.001), and severe degree of nuclear atypia (17% vs. 4%, p=0.043) were significantly more frequent in patients with residual tumors among 140 resection margin positive patients, but mitotic count, severe inflammatory cell infiltration, endocervical glandular extension, and cytoplasmic maturation did not differ significantly between the two groups. Multivariate analysis revealed that positive endocervical biopsy, bulbous growth, and central necrosis were associated with residual disease in the resection margin positive patients.In patients with positive LEEP margins in young women who want to preserve her fertility, negative endocervical biopsy, and the absence of bulbous growth or central necrosis can be used as selection criteria for patients who can defer the additional surgery and be closely followed with colposcopic and cytologic examination. Even in samples with negative LEEP margins, the patients who have those histologic parameters may need more careful monitoring for residual tumor or tumor recurrence.