(P 287) INTERNAL HEMIPELVECTOMY: A SINGLE INSTITUTION'S LEARNING CURVE AND LONGITUDINAL EXPERIENCE

Objective: Tumors within the pelvis present many problems for orthopedic surgeons due to their proximity to surrounding neurovascular structures and the risk of spinopelvic destabilization during tumor resection. While historically external hemipelvectomy was the treatment of choice, internal hemipelvectomy has become more commonly preferred as a limb salvage technique. This transition has been supported by improvements in technology including 3-dimensional (3D) cutting guides and computer navigation technology. Although there have been some smaller studies documenting surgical outcomes and prognostic factors for internal hemipelvectomy resections, there are few longitudinal studies that evaluate the effect of surgeon case volume and how these changes have affected surgeon decision making.

The aim of this clinical study was to assess how an individual surgeon's experiences and advancements in techniques and treatment options in the approach to internal hemipelvectomy have impacted patient outcomes at a single institution.

Methods: This study retrospectively examined a single tertiary academic institution’s consecutive longitudinal experience with internal hemipelvectomy for primary sarcoma or pelvic metastases over a twenty-six-year period between the years 1994 and 2020. Two grouping methods were utilized. The first stratified patients into cohorts based on the date of surgery with three distinct “eras” (“early”, “middle”, and “modern”), which reflect the implementation of new techniques including 3D computer navigation and cutting guide technology into our clinical practice. The second method of cohort selection grouped patients based on each surgeon’s case experience with internal hemipelvectomy (“inexperienced”, “developing”, and “experienced”). Primary endpoints included margin status, complication profiles, and long-term oncologic outcomes. Whole group multivariate analysis was used to evaluate variables predicting blood loss, operative time, tumor-free survival, and mortality.

Results: A total of 72 patients who underwent internal hemipelvectomy were identified. Of these patients, 24 had surgery between 1994 and 2007 (early), 28 between 2007 and 2015 (middle), and 20 between 2016 and 2020 (modern). Twenty-eight patients had surgery while the surgeon was still inexperienced, 24 while developing, and 20 when experienced. Evaluation by era (Table 1) demonstrated that a greater proportion of patients were indicated for surgery for oligometastatic disease in the modern era (0% vs 14.3% vs 35%, p = .022). Fewer modern cases utilized freehand resection (100% vs 75% vs 55%, p = .012), while instead opting for more computer navigation (0% vs 25% vs 20%, p = .012), and customized 3D-printed cutting guides (0% vs 0% vs 25%, p = .002). Similarly, there was a decline in the rate of massive blood loss (>2 liters) observed (72.2% vs 30.8% vs 35%, p = .016), and interdisciplinary collaboration with a general surgeon for pelvic dissection became more common (4.2% vs 32.1% vs 85%, p < .001). Local recurrence was less prevalent in patients treated in middle and modern eras (50% vs 15.4% vs 25%, p = .045). When evaluating by case experience (Table 2), surgeries performed by experienced surgeons were less frequently complicated by massive blood loss (66.7% vs 40% vs 20%, p = .007) and more often involved a general surgeon for pelvic dissection (17.9% vs 37.5% vs 65%, p = .004). Whole group multivariate analysis demonstrated that increased surgeon case experience predicted lower intraoperative blood loss (p = .040). However, surgeon experience had no significant effect on operative time (p = .125), tumor-free survival (p = .501), or overall patient survival (p = .735).

Conclusion: Our findings demonstrate a transition to less freehand surgical approaches and more 3D navigation and customized 3D cutting guides over the years. There is a preference towards interdisciplinary involvement for intraoperative assistance, as well as a willingness to be more aggressive in treating metastatic disease surgically. Although there does not appear to be a change in perioperative outcomes over the years, this is likely due a willingness to treat patients at later stages of disease and patients with more complex pathology, potentially masking the benefits of experience and technological advances.