Impact of Extended Versus Limited Lymph Node Dissection on Surgical Outcome, Recurrence Patterns and Survival After Radical Cystectomy.
Background: The aim of our study is to evaluate the impact of extended versus limited lymph node dissection (LND) in patients undergoing radical cystectomy (RC) on survival, perioperative outcomes and pattern of recurrence.
Methods: We reviewed our charts to identify patients who underwent RC and LND with curative intent between January 2003 and November 2022. Standard open RC with limited or extended LND was routinely performed, depending on surgeon's preference. The upper limit of extended LND is usually the ureteral crossing with the common iliac artery, unless further extension to the aortic bifurcation or inferior mesenteric artery is clinically indicated. Whereas limited LND includes removal of external and internal iliac and obturator Lymph nodes (LNs). The primary outcome was to compare the 2 patient groups in terms of cancer-specific survival (CSS) and overall survival (OS). The secondary outcome was to assess the impact of the extent of LND on the pattern of recurrence (local and distant metastasis free-survival) and perioperative complications.
Results: Of 642 patients, 439 and 203 underwent limited and extended LND, respectively. In the extended LND group, the median number of LNs removed was 23 compared to 8 in the limited LND group (P < .001), which was associated with higher median positive LNs in the extended group (3 vs. 2, P = .05). Extended LND was associated with longer operative time (300 vs. 250 min., P < .001), but not with blood loss, postoperative hemoglobin drop, hospital stay, 90-days major complications and hospital readmission rates. Lymphocele requiring surgical intervention was higher in extended group (7.4% vs. 1.8%, P = .001). The median follow-up time of survivors was 41 months in the limited group and 52 months in the extended group, (P = .1). Overall, 127 (29%) and 52 (26%) patients in the limited and extended groups experienced clinical recurrence (P = .39). At multivariable Cox regression analysis, LND template was not associated with either local and distant metastasis-free survival or CSS, while resection of ≥16 LNs was an independent predictor of local recurrence-free survival (HR 0.54; P = .01) and CSS (HR 0.6; P = .002), regardless of the dissected template. Both extended LND (HR 0.63; P = .004) and resection of ≥16 LNs (HR 0.66; P = .003) were associated with improved OS.
Conclusions: The number of LNs removed appears to be more important than the LN template in patients undergoing RC. Resection of at least 16 LNs is associated with better cancer control and oncologic outcome.