Purpose: An individualized midline shield (MLS) has been advocated for delivering homogeneous radiotherapy for patients with invasive cervical carcinoma. Yet, many radiation oncologists continue to employ a standard block. In the latter instance, any deviation of the cranial-caudal central axis of he tandem from the patient's midline could result in dose inhomogeneity to tumor. A retrospective review of a single university medical center's experience with constructing the MLS was initiated to determine the outcome of using a standard block vs. a customized block that conforms to the "Point A" isodose line. In addition, participating radiation oncologists associated with the Gynecologic Oncology Group (GOG) were polled to assess if there exists a consensus regarding midline block utilization in the management of cervical cancer patients which could be compared to the institutional study.
Methods and materials: From January 1, 1990 through December 31, 1992, 32 patients with invasive cervical carcinoma who underwent low dose rate brachytherapy at a single institution were identified. Patients were grouped as having a standard block (18 cases), customized block (5 cases), or no block (9 cases). The "Point A" isodose distribution from the implant was superimposed onto the whole pelvic simulation film and quantitatively compared to the actual or a hypothetical standard block outlined on the same radiograph. In September of 1995, 56 member and affiliated institutions in the GOG were surveyed concerning their use of a MLS, and the results were tabulated in December of 1995.
Results: Approximately 72% of all cases (23 out of 32) at the single institution had tandem deviation ranging from 0-230 with a median of 50. This translated into a median percent overdosage to "Point A" Right of 15% and "Point A" Left of 12.5%. Although overall survival and incidence of chronic complications have not been affected by type of shielding, patient follow-up is limited with a median of 17.7 months (range: 4.2-58.9 months). Of the 56 surveyed radiation facilities in the GOG, 34 (61%) responded. One center was subsequently excluded as it performs only high dose rate brachytherapy. Of the evaluable respondents, 88% (29 out of 33) utilize a MLS in treating their patients with invasive cervical carcinoma. Of the latter group, 76% (22 out of 29), 21% (6 out of 29), and 3% (1 out of 29) employ a standard block, customized block, and a "step-wedge," respectively. For those using a standard block, 77% (17 out of 22) align the central cranial-caudal axis of the MLS along the corresponding midplane of the patient's pelvis on an anteroposterior radiograph rather than along the superior-inferior central axis of the tandem.
Conclusions: This study suggests that the use of a standard midline shield could result in potential tumor dose inhomogeneity and should be avoided. A national survey of major academic centers further suggests that the majority of these facilities also utilize a rectangular central block that is not positioned with respect to possible tandem deviation. Further investigation concerning the techniques of midline shield construction should be considered.