A total of 555 patients were included in this study, with 53 (9.5%) in the young group (age, < 50 years) and 502 (90.5%) in the old group (age, ≥ 50 years) (Table 1). In terms of demographics, there were no significant differences in sex, BMI, and tumor size between the two groups. However, a higher proportion of patients in the young group had an ASA physical status classification of ≥ 3 (9.4% vs. 38.0%, p < 0.001). Periampullary adenocarcinomas were less common in the young group than in the old group (32.1% vs. 76.5%, p < 0.001), whereas solid and pseudopapillary tumors (9.4% vs. 1.0%) and neuroendocrine tumors (15.1% vs. 3.6%) were more common in the young group (p < 0.001). Among the 401cases of periampullary adenocarcinoma, the distributions of pancreatic head (41.2% vs. 48.4%) and other periampullary adenocarcinoma (58.8% vs. 51.6%) between the young and old groups were similar (p = 0.626). Soft pancreatic parenchyma (77.4% vs. 62.5%, p = 0.033) and non-dilated (≤ 3 mm) pancreatic ducts (77.4% vs. 46.3%, p < 0.001) were more frequently observed in the young group.
Table 1
Demographics for patients with periampullary lesions undergoing robotic pancreaticoduodenectomy
| Total | Age < 50 y/o | Age ≥ 50 y/o | P value |
Patients, n (%) | 555 | 53 (9.5%) | 502 (90.5%) | |
Age, year old | | | | < 0.001 |
Median (range) | 67 (13–97) | 42 (13–49) | 68 50–97) | |
Mean ± SDa | 66 ± 12 | 40 ± 9 | 69 ± 9 | |
Sex | | | | 0.512 |
female | 259 (46.7%) | 27 (50.9%) | 232 (46.2%) | |
male | 296 (53.3%) | 26(49.1%) | 270 (53.8%) | |
BMIb, kg/m2 | | | | 0.628 |
Median (range) | 23.5 (15.4–36.2) | 23.1 (16.7–34.1) | 23.5 (15.4–36.2) | |
Mean ± SD | 23.7 ± 3.5 | 23.9 ± 4.1 | 23.7 ± 3.4 | |
ASAc physical status classification | | | | < 0.001 |
< 3 | 359 (64.7%) | 48 (90.6%) | 311 (62.0%) | |
≥ 3 | 196 (35.3%) | 5 (9.4%) | 191 (38.0%) | |
Periampullary lesions | | | | < 0.001 |
Pancreatic head adenocarcinoma | 193 (34.8%) | 7 (13.2%) | 186 (37.1%) | |
Ampullary adenocarcinoma | 139 (25.0%) | 6 (11.3%) | 133 (25.5%) | |
Distal CBDd adenocarcinoma | 43 (7.7%) | 0 (0.0%) | 43 (8.6%) | |
Duodenal adenocarcinoma | 26 (4.7%) | 4 (7.5%) | 22 (4.4%) | |
IPMNe | 43 (7.7%) | 4 (7.5%) | 39 (7.8%) | |
Neuroendocrine tumor | 26 (4.7%) | 8 (15.1%) | 18 (3.6%) | |
Solid and pseudopapillary tumor | 10 (1.8%) | 5 (9.4%) | 5 (1.0%) | |
Chronic pancreatitis | 16 (2.9%) | 5 (9.4%) | 11 (2.2%) | |
Other malignant tumor | 33 (5.9%) | 7 (13.2%) | 26 (5.2%) | |
Other benign tumor | 26 (4.7%) | 7 (13.2%) | 19 (3.8%) | |
Periampullary adenocarcinomas | | | | < 0.001 |
Yes | 401 (72.3%) | 17 (32.1%) | 384 (76.5%) | |
No | 154 (27.7%) | 36 (57.9%) | 118 (23.5%) | |
Periampullary adenocarcinomas | | | | 0.626 |
Pancreatic head adenocarcinomas | 193 (48.1%) | 7 (41.2%) | 186 (48.4%) | |
Other periampullary adenocarcinoma | 208 (51.9%) | 10 (58.8%) | 198 (51.6%) | |
Pancreatic parenchyma | | | | 0.033 |
soft | 355 (64.0%) | 41 (77.4%) | 314 (62.5%) | |
hard | 200 (36.0%) | 12 (22.6%) | 188 (37.5%) | |
Pancreatic duct | | | | < 0.001 |
non-dilated ≤ 3 mm | 270 (49.3%) | 41 (77.4%) | 229 (46.3%) | |
dilated > 3mm | 278 (50.7%) | 12 (22.6%) | 266 (53.7%) | |
Tumor size, cm | | | | 0.263 |
Median (range) | 3.0 (0.5–11.0) | 3.0 (1.0–8.5) | 3.0 (0.5–11.0) | |
Mean ± SD | 3.1 ± 1.4 | 3.3 ± 1.7 | 3.1 ± 1.4 | |
aSD: standard deviation; bBMI: body mass index; cASA: American Society of Anesthesiologists; dCBD: common bile duct; eIPMN: intraductal papillary mucinous neoplasm |
Regarding surgical outcomes (Table 2), there were no significant differences in operation time (median, 7.8 vs. 8.3 h; p = 0.508), intraoperative blood loss (median, 100 vs. 160 mL: p = 0.681), surgical radicality (R0 resection, 92.5% vs. 85.1%; p = 0.217), lymph node yield (median, 17 vs. 18; p = 0.681), lymph node involvement (50.0% vs. 56.1%, p = 0.798), stage 1 + 2 (58.8% vs. 70.6%, p = 0.292), conversion rate (5.7% vs. 8.4%, p = 0.492), and vascular resection rate (3.8% vs. 3.8%, p = 0.997) between the young and old groups. However, most of these surgical outcomes were favorable in the young group. The LOS was shorter in the young group, with a median of 16 days, than in the old group, with a median of 20 days (p = 0.033). After multivariate analysis by binary logistic regression (Fig. 1), pancreatic head adenocarcinoma (+), morbidity (+), POPF (+), and chyle leakage (+) were independent factors to predict longer length of stay (LOS) after RPD, and age itself was not.
Table 2
Surgical outcomes after robotic pancreaticoduodenectomy
| Total | Age < 50 y/o | Age ≥ 50 y/o | P value |
Patients, n | 555 | 53 (9.5%) | 502 (90.5%) | |
Operation time, hour | | | | 0.508 |
Median (range) | 8.0 (3.3–16.3) | 7.8 (4.0–13.5) | 8.3 (3.3–16.3) | |
Mean ± SDa | 8.4 ± 2.3 | 7.9 ± 2.3 | 8.4 ± 2.3 | |
Blood loss, c.c. | | | | 0.681 |
Median (range) | 160 (0–6000) | 100 (0–4600) | 160 (0–6000) | |
Mean ± SD | 239 ± 396 | 261 ± 666 | 237 ± 357 | |
Surgical radicality | | | | 0.217 |
R0 | 476 (85.8%) | 49 (92.5%) | 427 (85.1%) | |
R1 | 57 (10.3%) | 4 (7.5%) | 53 (10.6%) | |
R2 | 22 (4.0%) | 0 | 22 (4.4%) | |
Lymph node yield | | | | 0.351 |
Median (range) | 18 (12–49) | 17 (12–37) | 18 (12–49) | |
Mean ± SD | 19 ± 6 | 18 ± 6 | 19 ± 5 | |
Lymph node involvement | 218 (55.9%) | 8 (50.0%) | 210 (56.1%) | 0.798 |
Stage | | | | 0.292 |
1 + 2 | 281 (70.1% | 10 (58.8%) | 271 (70.6%) | |
3 + 4 | 120 (29.9%) | 7 (41.2%) | 113 (29.4%) | |
Conversion to open, n (%) | 45 (8.1%) | 3 (5.7%) | 42 (8.4%) | 0.492 |
Vascular resection, n (%) | 21 (3.8%) | 2 (3.8%) | 19 (3.8%) | 0.997 |
LOSb, day | | | | 0.033 |
Median (range) | 19 (6–118) | 16 (6–46) | 20 (6–118) | |
Mean ± SD | 23 ± 14 | 19 ± 9 | 23 ± 14 | |
aSD: standard deviation; bLOS: length of stay |
The overall surgical mortality for the entire cohort was 1.5%, with no surgical mortality in the young group compared with 1.6% in the old group (p = 0.352). The DGE rate was 4.3% for all patients, with 1.9% in the young group and 4.6% in the old group (p = 0.359). The overall POPF rate was 7.9%, with 7.5% in the young group and 8.0% in the old group (p = 0.914). There were also no significant differences in surgical morbidity, Clavien–Dindo surgical complications, severity of complications, PPH, chyle leakage, bile leakage, and wound infection between the young and old groups (Table 3).
Table 3
Surgical risks after robotic pancreaticoduodenectomy
| Total | Age < 50 y/o | Age ≥ 50 y/o | P value |
Patients, n | 555 | 53 (9.5%) | 502 (90.5%) | |
Surgical mortality | 8 (1.5%) | 0 | 8 (1.6%) | 0.352 |
Surgical morbidity | 312 (56.2%) | 28 (52.8%) | 284 (56.6%) | 0.601 |
Surgical complication | | | | 0.888 |
Clavien–Dindo 0 | 236 (42.5%) | 23 (43.4%) | 213 (42.4%) | |
Clavien–Dindo I | 191 (34.4%) | 18 (34.0%) | 173 (34.5%) | |
Clavien–Dindo II | 52 (9.4%) | 5 (9.4%) | 47 (9.4%) | |
Clavien–Dindo III | 62 (11.2%) | 7 (13.2%) | 55 (11.0%) | |
Clavien–Dindo IV | 5 (0.9%) | 0 | 5 (1.0%) | |
Clavien–Dindo V (death) | 9 (1.6%) | 0 | 9 (1.8%) | |
Severity of complication, n = 319 | | | | 0.947 |
Minor (Clavien–Dindo I-II) | 243 (76.2%) | 23 (76.7%) | 220 (76.1%) | |
Major (Clavien–Dindo ≥ III) | 76 (23.8%) | 7 (23.3%) | 69 (23.9%) | |
POPFa (ISGPFb grade B and C) | | | | |
Overall | 44 (7.9%) | 4 (7.5%) | 40 (8.0%) | 0.914 |
Parenchyma of pancreas | | | | |
soft | 37 (10.4%) | 4 (9.8%) | 33 (10.5%) | 0.882 |
hard | 7 (3.5%) | 0 | 7 (3.7%) | 0.496 |
Diameter of pancreatic duct | | | | |
non-dilated ≤ 3 mm | 30 (11.1%) | 4 (9.8%) | 26 (11.4%) | 0.746 |
dilated > 3 mm | 14 (5.0%) | 0 | 14 (5.3%) | 0.415 |
DGEc (ISGPSd grade B and C) | 24 (4.3%) | 1 (1.9%) | 23 (4.6%) | 0.359 |
PPHe (ISGPSd grade B and C) | 32 (5.8%) | 4 (7.5%) | 28 (5.6%) | 0.559 |
Chyle leakage | 140 (25.2%) | 14 (26.4%) | 126 (25.1%) | 0.834 |
Bile leakage | 10 (1.8%) | 1 (1.9%) | 9 (1.8%) | 0.961 |
Wound infection | 28 (5.0%) | 1 (1.9%) | 27 (5.4%) | 0.269 |
aPOPF: postoperative pancreatic fistula, bISGPF: International Study Group of Pancreatic Fistula; cDGE: delayed gastric emptying; dISGPS: International Study Group of Pancreatic Surgery; ePPH: postpancreatectomy hemorrhage |
In terms of survival outcomes, the 5-year survival rate for the entire cohort of overall periampullary adenocarcinoma was 48.1% (Table 4). The young group showed a significantly higher 5-year survival rate for overall periampullary adenocarcinoma than the old group (76.4% vs. 46.7%, p = 0.047) (Fig. 2). However, there was no significant difference in survival between the two groups when analyzing the subgroups of pancreatic head adenocarcinoma (5-year survival, 62.5% vs. 31.4%; p = 0.171) and ampullary adenocarcinoma (5-year survival, 100% vs. 61.4%; p = 0.159). After multivariate analysis using Cox proportional hazards regression model (Fig. 3), pancreatic head cancer (+), lymph node (LN) involvement (+), and late stage 3 + 4 (+) were the independent prognostic factors to predict unfavorable survival after robotic pancreaticoduodenectomy, and age was not an independent prognostic predictor.
Table 4
Survival outcomes for periampullary adenocarcinomas after robotic pancreaticoduodenectomy
Periampullary adenocarcinoma | Median, (mon.) | Range, (mon.) | Mean ± SDa, (mon.) | 1-year survival | 3-year survival | 5-year survival | P value |
Overall periampullary | | | | | | | |
Total, n = 394 | 20.4 | 0.2–107.6 | 28.7 + 23.3 | 85.4% | 57.1% | 48.1% | 0.047 |
Age < 50 y/o, n = 17 | 35.3 | 8.9–82.9 | 40.1 ± 24.2 | 100% | 76.4% | 76.4% | |
Age ≥ 50 y/o, n = 377 | 20.2 | 0.2–107.6 | 28.2 ± 23.2 | 84.7% | 56.2% | 46.7% | |
Pancreatic head | | | | | | | |
Total, n = 191 | 16.6 | 0.8–98.1 | 23.0 ± 19.8 | 77.8% | 40.4% | 32.9% | 0.171 |
Age < 50 y/o, n = 7 | 24.6 | 8.9–67.3 | 34.4 ± 22.4 | 100% | 62.5% | 62.5% | |
Age ≥ 50 y/o, n = 184 | 16.5 | 0.8–98.1 | 22.6 ± 19.6 | 76.9% | 39.4% | 31.4% | |
Ampullary | | | | | | | |
Total, n = 136 | 28.1 | 0.2–107.6 | 35.7 ± 26.3 | 91.3% | 73.9% | 63.1% | 0.159 |
Age < 50 y/o, n = 6 | 43.4 | 11.7–75.6 | 41.9 ± 29.0 | 100% | 100% | 100% | |
Age ≥ 50 y/o, n = 130 | 28.1 | 0.2–107.6 | 35.7 ± 26.3 | 90.9% | 72.8% | 61.4% | |
aSD: standard deviation; |