Abstract
Background:
Recently, subclassification of pancreatoduodenectomy in 4 differing types has been reported, because additional major vascular and multivisceral resections have been shown to be associated with an increased risk of postoperative morbidity and mortality.
Objective:
To classify distal pancreatectomy (DP) based on the extent of resection and technical difficulty and to evaluate postoperative outcomes with regards to this classification system.
Methods:
All consecutive patients who had undergone DP between 2001 and 2020 in a high-volume pancreatic surgery center were included in this study. DPs were subclassified into 4 distinct categories reflecting the extent of resection and technical difficulty, including standard DP (type 1), DP with venous (type 2), multivisceral (type 3), or arterial resection (type 4). Patient characteristics, perioperative data, and postoperative outcomes were analyzed and compared among the 4 groups.
Results:
A total of 2135 patients underwent DP. Standard DP was the most frequently performed procedure (64.8%). The overall 90-day mortality rate was 1.6%. Morbidity rates were higher in patients with additional vascular or multivisceral resections, and 90-day mortality gradually increased with the extent of resection from standard DP to DP with arterial resection (type 1: 0.7%; type 2: 1.3%; type 3: 3%; type 4: 8.7%; P<0.0001). Multivariable analysis confirmed the type of DP as an independent risk factor for 90-day mortality.
Conclusions:
Postoperative outcomes after DP depend on the extent of resection and correlate with the type of DP. The implementation of the 4-type classification system allows standardized reporting of surgical outcomes after DP improving comparability of future studies.
Keywords: distal pancreatectomy, morbidity, mortality, risk stratification
Distal pancreatectomy (DP) entails the resection of the body and tail of the pancreas left of the mesenterico-portal venous axis and remains the standard treatment for various benign and malignant pathologies. It is performed less often than resections of the pancreatic head because of a lower incidence and later diagnosis of diseases in this part of the organ. Tumors are generally diagnosed in more advanced stages and often involve adjacent organs or major vessels. Over the last 2 decades, progress in neoadjuvant therapy and surgical techniques have led to an expansion of indications and a more aggressive surgical approach. Extended resections, including major vascular1–4 and multivisceral resections, have been introduced to achieve radical resection of locally advanced, previously deemed unresectable tumors, while maintaining acceptable complication rates.5
In general, extended resections are associated with an increased risk of perioperative morbidity and mortality,6–9 albeit robust data on outcomes after DP are limited. The International Study Group of Pancreatic Surgery (ISGPS) defined standard and extended resections in pancreatic surgery in 2014.10 However, the proposed classification does not further differentiate between additional vascular and/or organ resection and is seldomly applied for evaluation of postoperative morbidity and mortality. The term DP covers a spectrum of procedures with a broad range of technical difficulty. Recently, for pancreatoduodenectomy and total pancreatectomy simply applicable, 4-stage categorization systems based on the extent, complexity and technical aspects of surgery have been described and shown to significantly correlate with perioperative outcomes.11,12 Yet, to date, there is no similar standardized classification system for DP, making reports of perioperative outcomes after DP hardly comparable.
Therefore, the aim of the present study was to establish a classification of different types of DP with regard to the extent of resection, complexity and technical difficulty and assess type-specific postoperative outcomes.
METHODS
Data of 2135 consecutive patients, who underwent elective DP at Heidelberg University Hospital, Heidelberg, Germany within a 20-year period (October 2001 until December 2020), were extracted from a prospectively maintained database. Each patient conferred written informed consent for data collection and analysis. The Ethics Committee of the Medical Faculty of the University of Heidelberg authorized the study (S011/2015). The findings of our study are reported in accordance with the revised STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement.13 Patient baseline characteristics and perioperative data included age, sex, body mass index, American Society of Anesthesiologists performance status (ASA), histopathologic findings, duration of surgery and estimated intraoperative blood loss. Technical details of the procedure such as surgical access, transection technique and abdominal drain placement as well as kind of additional vascular or organ resection were extracted. Short-term postoperative outcomes encompassed postoperative pancreas-specific morbidity, including postoperative pancreatic fistula (POPF), postpancreatectomy hemorrhage (PPH), delayed gastric emptying (DGE) and chyle leak, all defined according to the classification of the ISGPS, and relaparotomy as well as 90-day mortality.4,14–16 The overall postoperative outcome of each patient was graded according to the Clavien-Dindo classification (CDC).17 In addition, textbook outcome as a multidimensional measure for surgical quality, reflecting the ideal outcome, was determined.18–20 Textbook outcome in pancreatic surgery according to the Dutch Pancreatic Cancer Group was defined by the absence of POPF, bile leak, PPH (all ISGPS grade B/C), severe complications (CDC ≥III), readmission and in hospital mortality.21 Based on the extent, complexity and technical difficulty of DP, the following 4-type classification system, analogous to the recently published categorization of pancreatoduodenectomy and total pancreatectomy11,12 has been defined: type 1: standard DP with or without splenectomy, type 2: DP with additional portal vein/superior mesenteric vein and/or left renal vein resection, type 3: DP with additional organ resection, and type 4: DP with additional major arterial resection (except splenic artery). Procedures with features of >1 type were placed in the higher category (eg, DP with both venous and multivisceral resection was classified as type 3 rather than type 2). The definition of multivisceral DP corresponded to the proposal of the ISGPS,10 that is, DP together with any type of resection of the stomach, colon, and/or mesocolon with relevant vascular structures of the transverse mesocolon (middle or left colic vessels), small bowel, left adrenal gland, left kidney and/or renal vessels, crus or part of diaphragm, and liver. DP with excisional biopsies of the liver and stomach were categorized as standard procedure.
Statistical Analysis
The software SAS, release 9.4 (SAS Institute) was used for data management and statistical analysis. The proposed classification system with 4 types was subsequently validated for postoperative morbidity and mortality. Patient characteristics, perioperative parameters, and postoperative outcomes were summarized using absolute and relative frequencies for categorical parameters and the median with interquartile range (IQR) for continuous parameters. For comparisons, either the Fisher exact test or the χ2 test was performed, as appropriate, together with the Kruskal-Wallis test. All tests were used to compare the 4 types as overall testing and the resulting P values were interpreted together with the observed differences of the respective parameter. Univariable risk factor analysis for 90-day mortality and postoperative morbidity was performed using logistic regression analysis. The odds ratio (OR) and the corresponding 95% CI are presented. Subsequently, we examined a multivariable logistic regression model including risk factors with a univariable P value <0.2. Variable selection and model building were performed using the Akaike information criterion and no missing values were imputed. A 2-sided value of P<0.05 was considered statistically significant. Because of the exploratory nature of the analyses performed, results were interpreted carefully, and P values were used descriptively.
RESULTS
Baseline Characteristics and Surgical Details
A total of 2135 patients underwent elective DP. The median age was 62.1 (IQR, 52.2–70.7) years. In all, 1124 (52.6%) patients were female. The most frequent indication for DP was pancreatic ductal adenocarcinoma (37.7%), followed by neuroendocrine tumors (14.1%), chronic pancreatitis (10.5%), intraductal papillary mucinous neoplasms (8.7%), serous cystic neoplasms (8.3%), mucinous cystic neoplasms (4.6%), and others (metastases, rare pancreatic carcinomas, sarcomas, pseudopapillary tumors; 16%). Neoadjuvant chemotherapy and radiotherapy had been administered in 255 (12%) and 82 (3.8%) patients, respectively. Surgery was mostly performed using an open approach [open: 1932 (90.5%); laparoscopic: 110 (5.1%); robotic: 93 (4.4%)]. The median operation time was 194 (IQR, 147–250) minutes with a median estimated blood loss of 500 (IQR, 300–1000) mL. Splenectomy was performed in 1753 (82.1%) patients (Table 1).
TABLE 1.
Patient Characteristics According to the Type of DP
| n (%) | ||||||
|---|---|---|---|---|---|---|
| Characteristics | Total | Type 1 | Type 2 | Type 3 | Type 4 | P |
| N | 2135 | 1383 | 79 | 604 | 69 | |
| Age [median (IQR)] (yr) | 62.1 (52.2–70.7) | 61.5 (51.7–70.5) | 66.2 (57.9–73.6) | 62.4 (53.3–70.9) | 62.3 (51.5–70.1) | 0.0148 |
| Sex | 0.0015 | |||||
| Male | 1011 (47.4) | 618 (44.7) | 34 (43.0) | 318 (52.6) | 41 (59.4) | |
| Female | 1124 (52.6) | 765 (55.3) | 45 (57.0) | 286 (47.4) | 28 (40.6) | |
| BMI [median (IQR)] | 24.8 (22.3–27.8) | 25.1 (22.4–28.1) | 25.6 (22.5–28.4) | 24.5 (22.1–27.1) | 24.3 (21.5–27.0) | 0.0064 |
| ASA | <0.0001 | |||||
| I | 114 (5.5) | 83 (6.1) | 4 (5.3) | 20 (3.4) | 7 (10.5) | |
| II | 1280 (61.2) | 873 (64.4) | 40 (52.6) | 331 (55.9) | 36 (53.7) | |
| III | 687 (32.9) | 394 (29.1) | 32 (42.1) | 237 (40.0) | 24 (35.8) | |
| IV | 10 (0.5) | 6 (0.4) | 0 (0) | 4 (0.7) | 0 (0) | |
| NA | 44 | 27 | 3 | 12 | 2 | |
| Spleen | <0.0001 | |||||
| With splenectomy | 1753 (82.1) | 1071 (77.4) | 76 (96.2) | 541 (89.6) | 65 (94.2) | |
| Without splenectomy | 382 (17.9) | 312 (22.6) | 3 (3.8) | 63 (10.4) | 4 (5.8) | |
| Surgical approach | <0.0001 | |||||
| Open | 1932 (90.5) | 1185 (85.7) | 79 (100) | 599 (99.2) | 69 (100) | |
| Laparoscopic | 110 (5.1) | 107 (7.7) | 0 (0) | 3 (0.5) | 0 (0) | |
| Robotic | 93 (4.4) | 91 (6.6) | 0 (0) | 2 (0.3) | 0 (0) | |
| Operation time [median (IQR)] (min) | 194 (147–250) | 170 (135–219) | 236 (195–300) | 232 (183–300) | 267 (230–328) | <0.0001 |
| Estimated blood loss [median (IQR)] (mL) | 500 (300–1000) | 450 (200–700) | 1000 (600–1800) | 800 (500–1300) | 1450 (750–2500) | <0.0001 |
| Histologic findings | <0.0001 | |||||
| Pancreatic carcinoma | 804 (37.7) | 416 (30.1) | 70 (88.6) | 257 (42.5) | 61 (84.4) | |
| NET | 302 (14.1) | 225 (16.3) | 6 (7.6) | 69 (11.4) | 2 (2.9) | |
| SCN | 178 (8.3) | 166 (12.0) | 0 (0) | 12 (2.0) | 0 (0) | |
| MCN | 99 (4.6) | 87 (6.3) | 1 (1.3) | 11 (1.8) | 0 (0) | |
| Chronic pancreatitis | 225 (10.5) | 185 (13.4) | 1 (1.3) | 39 (6.5) | 0 (0) | |
| IPMN | 185 (8.7) | 168 (12.1) | 0 (0) | 17 (2.8) | 0 (0) | |
| Other | 342 (16.0) | 136 (9.8) | 1 (1.3) | 199 (33.0) | 6 (8.7) | |
| Neoadjuvant CTx | 255 (12.0) | 67 (4.8) | 14 (17.7) | 137 (22.7) | 37 (53.6) | <0.0001 |
| Neoadjuvant RTx | 82 (3.8) | 28 (2.0) | 3 (3.8) | 38 (6.3) | 13 (18.8) | <0.0001 |
BMI indicates body mass index; CTx, chemotherapy; IPMN, intraductal papillary mucinous neoplasm; MCN mucinous cystic neoplasms; NA, not available; NET, neuroendocrine tumor; RTx, radiotherapy; SCN, serous cystic neoplasm.
Most patients underwent standard DP [type 1: 1383 (64.8%)]. DP with venous resection (type 2) was performed in 79 (3.7%) patients. Portal vein/superior mesenteric vein resection was necessary in 78 and left renal vein resection in 2 patients, including tangential resections with venorrhaphy (ISGPS type 1) in 29 patients, resections with patch reconstruction (ISGPS type 2) in 9 patients, end-to-end anastomosis (ISGPS type 3) in 31 patients, and resections with graft interposition (ISGPS type 4) in 10 patients. After resection of left renal vein, no venous reconstruction was performed in 1 patient.22 Multiple additional vein resections in 1 patient were possible. DP with multivisceral resection (type 3) was performed in 604 patients (28.3%). Resected organs included the left adrenal gland (n=250), stomach (n=222), colon (n=189), liver (n=107), left kidney (n=106), small bowel (n=76), diaphragm (n=41) and others (adnexa, uterus, testis, esophagus, abdominal wall, lung; n=44). Multiple additional organ resections in 1 patient were possible. Additional portal vein/superior mesenteric vein resection in type 3 DP was performed in 42, vena cava resection in 2, and left renal vein resection in 3 patients, respectively. DP with major arterial resection (type 4) was done in 69 (3.2%) patients. Resection of the celiac trunk, hepatic artery, a segment of the aorta and the superior mesenteric artery was performed in 60, 11, 2, and 1 patients, respectively. Multiple additional artery resections in 1 patient were possible. Additional portal/superior mesenteric vein and/or left renal vein resection was performed in 30 type 4 DP patients, respectively. Patients with type 4 DP also underwent multivisceral resection (n=45) including resection of the left adrenal gland (n=30), stomach (n=18), colon (n=10), small bowel (n=8), liver (n=5), kidney (n=4), and diaphragm (n=4). Details according to the 4 different types of DP are reported in Table 1. Malignant diseases and neoadjuvant regimes were more common in patients undergoing extended DP. In more extended DP, an open approach was preferred (P<0.0001), duration of surgery was longer (P<0.0001), and estimated blood loss was higher (P<0.0001).
Postoperative surgical complications with a severity of ≥III according to the CDC were observed in 601 (28.1%) patients, including grade B/C POPF [n=498 (23.3%)], grade C PPH [n=58 (2.7%)], grade B/C DGE [n=135 (6.3%)] and chyle leakage [n=135 (6.3%)]. Relaparotomy was required in 198 (9.3%) patients. Completion pancreatectomy was necessary in 8 patients due to R1 situation and 7 patients due to severe POPF and/or PPH. The overall 90-day mortality rate for the entire patient cohort was 1.6%. 1453 (68.1%) patients had textbook outcome. Postoperative outcomes are summarized in Table 2.
TABLE 2.
Postoperative Outcomes According to the Extent of DP
| n (%) | ||||||
|---|---|---|---|---|---|---|
| Outcomes | Total | Type 1 | Type 2 | Type 3 | Type 4 | P |
| N | 2135 | 1383 | 79 | 604 | 69 | |
| 90-d mortality | 34 (1.6) | 9 (0.7) | 1 (1.3) | 18 (3.0) | 6 (8.7) | <0.0001 |
| Surgical morbidity | ||||||
| POPF (grade B+C) | 498 (23.3) | 279 (20.2) | 26 (32.9) | 176 (29.1) | 17 (24.6) | <0.0001 |
| PPH (grade C) | 58 (2.7) | 28 (2.0) | 4 (5.1) | 21 (3.5) | 5 (7.2) | 0.0101 |
| DGE (grade B+C) | 135 (6.3) | 44 (3.2) | 6 (7.6) | 72 (11.9) | 13 (18.8) | <0.0001 |
| Chyle leak | 135 (6.3) | 65 (4.7) | 8 (10.1) | 51 (8.4) | 11 (15.9) | <0.0001 |
| CDC≥III | 601 (28.1) | 309 (22.3) | 32 (40.5) | 232 (38.4) | 28 (40.6) | <0.0001 |
| Relaparotomy | 198 (9.3) | 85 (6.1) | 11 (13.9) | 85 (14.1) | 17 (24.6) | <0.0001 |
| Completion pancreatectomy | 16 (0.7) | 8 (0.6) | 2 (2.5) | 4 (0.7) | 2 (2.9) | 0.0446 |
| ICU stay ≥2 d | 276 (12.9) | 88 (6.4) | 21 (26.6) | 132 (21.9) | 35 (50.7) | <0.0001 |
| Textbook outcome* | 1453 (68.1) | 1013 (73.3) | 45 (57.0) | 356 (58.9) | 39 (56.5) | <0.0001 |
| Length of postoperative hospital stay [median (IQR)] (d) | 10 (8–16) | 9 (8–13) | 13 (10–25) | 14 (10–22) | 15 (11–24) | <0.0001 |
According to the definition for pancreatic surgery of the Dutch Pancreatic Cancer Group. Multiple entries are possible.
ICU indicate intensive care unit.
The number of DPs increased continuously over the study period [2001–2004: 130 (6.1%); 2005–2008: 357 (16.7%); 2009–2012: 424 (19.9%); 2013–2016: 580 (27.2%); 2017–2020: 643 (30.1%)], concomitant with an increasing number of extended resections.
Postoperative Outcomes Stratified by Type of DP
Overall postoperative morbidity increased significantly with the complexity of DP (Table 2), with more severe complications (CDC ≥III) occurring in extended compared with standard DP (type 1: 22.3%; type 2: 40.5%; type 3: 38.4%; type 4: 40.6%; P<0.0001). POPF grade B/C appeared more often after extended resections (type 1: 20.2%; type 2: 32.9%; type 3: 29.1%; type 4: 24.6%; P<0.0001), coming along with higher rates of PPH grade C (type 1: 2.0%; type 2: 5.1%; type 3: 3.5%; type 4: 7.2%; P=0.0101). Even though type 2 DP showed the highest rate of POPF, type 4 DP was associated with more severe PPHs. Likewise, the rates of DGE and chyle leakage were associated with type of DP (type 1: 3.2%; type 2: 7.6%; type 3: 11.9%; type 4: 18.8%; P<0.0001 and type 1: 4.7%; type 2: 10.1%; type 3: 8.4%; type 4: 15.9%; P<0.0001), respectively. Also, relaparotomy was more often necessary following extended DP, mostly after arterial resections (type 1: 6.1%; type 2: 13.9%; type 3: 14.1%; type 4: 24.6%; P<0.0001), and the completion pancreatectomy rate correlated with additional vascular resections (type 1: 0.6%; type 2: 2.5%; type 3: 0.7%; type 4: 2.9%; P=0.0446). Accordingly, patients, who underwent type 2 to 4 DP, stayed 3.5 to 7.8 times longer in the intensive care unit, with an increasing rate of patients with an intensive care unit stay ≥2 days (type 1: 6.4%; type 2: 26.6%; type 3: 21.9%; type 4: 50.7%; P<0.0001). Similarly, the median length of hospital stay increased throughout the 4 DP categories [type 1: 9 (IQR, 8–13) days; type 2: 13 (IQR, 10–25) days; type 3: 14 (IQR, 10–22) days; type 4: 15 (IQR, 11–24) days; P<0.0001]. Ninety-day mortality rates were strongly associated with type of resection (type 1: 0.7%; type 2: 1.3%; type 3: 3.0%; type 4: 8.7%; P<0.0001). Vice versa, higher postoperative morbidity and mortality rates led to lower rates of textbook outcome (type 1: 73.3%; type 2: 57.0%; type 3: 58.9%; type 4: 56.5%; P<0.0001; Table 2).
Multivariable analysis revealed the following independent risk factors for 90-day mortality: ASA stages III and IV (OR=5.031, 95% CI: 2.083–14.059; P=0.0007), age ≥70 years (OR=2.462, 95% CI: 1.139–5.428; P=0.0225), chronic heart disease (OR=3.064, 95% CI: 1.256–7.155; P=0.0109), estimated blood loss ≥1000 mL (OR=2.413, 95% CI: 1.124–5.291; P=0.0246), and type of DP (type 3 vs type 1/2: OR=3.172, 95% CI: 1.421–7.431; P=0.0057; type 4 vs type 1/2: OR=9.017, 95% CI: 2.478–29.758; P=0.0004). Details of univariable and multivariable analyses are shown in Tables 3 and 4.
TABLE 3.
Univariable Analysis of Risk Factors With Regard to 90-day Mortality
| Parameters | Category | n (%) | Events [n (%)] | OR | 95% CI | P |
|---|---|---|---|---|---|---|
| Sex | Male | 1011 (47.4) | 13 (1.2) | 1 | ||
| Female | 1124 (52.6) | 21 (2.1) | 0.552 | 0.268–1.094 | 0.0892 | |
| Age | <50 yr | 437 (20.5) | 1 (0.2) | 0.183 | 0.010–0.912 | 0.0301 |
| 50–69 yr | 1128 (52.8) | 14 (1.2) | 1 | |||
| ≥70 yr | 570 (26.7) | 19 (3.3) | 2.744 | 1.373–5.617 | 0.0011 | |
| ASA classification | ASA I/II | 1394 (66.7) | 8 (0.6) | 1 | ||
| ASA III/IV | 697 (33.3) | 26 (3.7) | 6.713 | 3.160–15.937 | <0.0001 | |
| Diabetes mellitus | No | 1756 (82.2) | 26 (1.5) | 1 | ||
| Yes | 379 (17.8) | 8 (2.1) | 1.435 | 0.603–3.057 | 0.3766 | |
| CHD/heart failure | No | 1986 (93.0) | 24 (1.2) | 1 | ||
| Yes | 149 (7.0) | 10 (6.7) | 5.881 | 2.638–12.206 | <0.0001 | |
| Previous history of pancreatitis | No | 1923 (90.1) | 33 (1.7) | 1 | ||
| Yes | 212 (9.9) | 1 (0.5) | 0.271 | 0.015–1.269 | 0.2000 | |
| BMI | <25 | 1078 (51.3) | 13 (1.2) | 1 | ||
| 25 to <30 | 716 (34.0) | 14 (2.0) | 1.634 | 0.759–3.540 | 0.2061 | |
| ≥30 | 309 (14.7) | 5 (1.6) | 1.347 | 0.430–3.604 | 0.5739 | |
| Neoadjuvant therapy | No | 1861 (87.2) | 30 (1.6) | 1 | ||
| Yes | 274 (12.8) | 4 (1.5) | 0.904 | 0.267–2.313 | 0.8510 | |
| Carcinoma | No | 1077 (50.4) | 13 (1.2) | 1 | ||
| Yes | 1058 (59.6) | 21 (2.0) | 1.657 | 0.835–3.414 | 0.1553 | |
| DP type | Type 1 | 1383 (64.8) | 9 (0.7) | 1 | ||
| Type 2 | 79 (3.7) | 1 (1.3) | 1.957 | 0.106–10.612 | 0.5266 | |
| Type 3 | 604 (28.3) | 18 (3.0) | 4.689 | 2.147–11.000 | 0.0002 | |
| Type 4 | 69 (3.2) | 6 (8.7) | 14.540 | 4.747–41.593 | <0.0001 | |
| Splenectomy | No | 382 (17.9) | 4 (1.0) | 1 | ||
| Yes | 1753 (82.1) | 30 (1.7) | 1.645 | 0.645–5.564 | 0.3522 | |
| Operation time | ≤195 min | 1105 (51.8) | 7 (0.6) | 1 | ||
| 196–240 min | 441 (20.7) | 10 (2.3) | 3.639 | 1.389–10.080 | 0.0092 | |
| >240 min | 587 (27.5) | 17 (2.9) | 4.678 | 2.006–12.158 | 0.0006 | |
| Estimated blood loss | <500 mL | 848 (40.4) | 13 (0.8) | 1 | ||
| 500–1000 mL | 818 (38.9) | 13 (0.8) | 1.764 | 0.607–5.763 | 0.3106 | |
| >1000 mL | 435 (20.7) | 19 (3.4) | 6.846 | 2.710–20.849 | 0.0002 |
BMI indicate body mass index; CHD, coronary heart disease.
TABLE 4.
Multivariable Analysis of Risk Factors With Regard to 90-day Mortality in 2056 Patients (79 Patients With Missing Values Excluded)
| Parameters | Category | OR | 95% CI | P |
|---|---|---|---|---|
| ASA classification | ASA III/IV vs ASA I/II | 5.031 | 2.083–14.059 | 0.0007 |
| DP type | Type 3 vs type 1/2 | 3.172 | 1.421–7.431 | 0.0057 |
| Type 4 vs type 1/2 | 9.017 | 2.478–29.758 | 0.0004 | |
| Age | ≥70 vs <70 yr | 2.462 | 1.139–5.428 | 0.0225 |
| CHD/heart failure | Yes vs no | 3.064 | 1.256–7.155 | 0.0109 |
| Estimated blood loss | ≥1000 vs <1000 mL | 2.413 | 1.124–5.291 | 0.0246 |
| Not included | ||||
| Sex | Male vs female | 0.8222 | ||
| Previous history of pancreatitis | Yes vs no | 0.2802 | ||
| Operation time | >195 vs ≤195 min | 0.1682 | ||
| Carcinoma | Yes vs no | 0.0878 | ||
CHD indicates coronary heart disease.
Consistently, ASA stages III/IV (OR=1.322, 95% CI: 1.078–1.620; P=0.0072), estimated blood loss ≥1000 mL (OR=2.018, 95% CI: 1.501–2.715; P<0.0001), operation time >240 versus ≤195 minutes (OR 3.172, 95% CI: 1.421–7.431; P=0.0036), and type of DP (type 2/3/4 vs type 1: OR=1.385, 95% CI: 1.114–1.719; P=0.0032), were independent risk factors for an outcome other than textbook by univariable and multivariable analyses (Tables 5, 6).
TABLE 5.
Univariable Analysis of Risk Factors With Regard to the Morbidity Defined as Missing a Textbook Outcome
| Parameters | Category | n (%) | Event [n (%)] | OR | 95% CI | P |
|---|---|---|---|---|---|---|
| Sex | Male | 1011 (47.4) | 349 (34.5) | 1 | ||
| Female | 1124 (52.6) | 333 (29.6) | 0.799 | 0.665–0.958 | 0.0155 | |
| Age | <50 yr | 437 (20.5) | 130 (29.7) | 0.875 | 0.687–1.110 | 0.2854 |
| 50–69 yr | 1128 (52.8) | 369 (32.6) | 1 | |||
| ≥70 yr | 570 (26.7) | 184 (32.3) | 0.984 | 0.793–1.220 | 0.6359 | |
| ASA classification | ASA I/II | 1394 (66.7) | 405 (29.1) | 1 | ||
| ASA III/IV | 697 (33.3) | 263 (37.7) | 1.480 | 1.221–1.792 | <0.0001 | |
| Diabetes mellitus | No | 1756 (82.2) | 548 (31.2) | 1 | ||
| Yes | 379 (17.8) | 134 (35.4) | 1.206 | 0.953–1.521 | 0.1165 | |
| CHD/heart failure | No | 1986 (93.0) | 620 (31.22) | 1 | ||
| Yes | 149 (7.0) | 62 (41.6) | 1.570 | 1.115–2.200 | 0.0091 | |
| Previous history of pancreatitis | No | 1923 (90.1) | 616 (32.0) | 1 | ||
| Yes | 212 (9.9) | 66 (31.1) | 0.959 | 0.703–1.297 | 0.7907 | |
| BMI | <25 | 1078 (51.3) | 324 (30.1) | 1 | ||
| 25 to <30 | 716 (34.0) | 237 (33.1) | 1.151 | 0.940–1.410 | 0.1732 | |
| ≥30 | 309 (14.7) | 111 (35.9) | 1.305 | 0.998–1.700 | 0.0504 | |
| Neoadjuvant therapy | No | 1861 (87.2) | 591 (31.8) | 1 | ||
| Yes | 274 (12.8) | 91 (33.2) | 1.069 | 0.813–1.396 | 0.6298 | |
| Carcinoma | No | 1077 (50.4) | 309 (28.7) | 1 | ||
| Yes | 1058 (59.6) | 373 (35.3) | 1.353 | 1.128–1.625 | 0.0012 | |
| DP type | Type 1 | 1383 (64.8) | 370 (26.8) | 1 | ||
| Type 2 | 79 (3.7) | 34 (43.0) | 2.069 | 1.297–3.272 | 0.0020 | |
| Type 3 | 604 (28.3) | 248 (41.1) | 1.907 | 1.559–2.332 | <0.0001 | |
| Type 4 | 69 (3.2) | 30 (43.5) | 2.106 | 1.281–3.431 | 0.0029 | |
| Splenectomy | No | 382 (17.9) | 104 (27.2) | 1 | ||
| Yes | 1753 (82.1) | 578 (33.0) | 1.315 | 1.031–1.688 | 0.0294 | |
| Operation time | ≤195 min | 1105 (51.8) | 277 (25.1) | 1 | ||
| 196–240 min | 441 (20.7) | 153 (34.7) | 1.588 | 1.250–2.015 | 0.0001 | |
| >240 min | 587 (27.5) | 252 (42.9) | 2.248 | 1.818–2.782 | <0.0001 | |
| Estimated blood loss | <500 mL | 848 (40.4) | 199 (23.5) | 1 | ||
| 500–1000 mL | 818 (38.9) | 270 (33.0) | 1.668 | 1.339–2.081 | <0.0001 | |
| >1000 mL | 435 (20.7) | 202 (46.4) | 2.934 | 2.285–3.774 | <0.0001 |
BMI indicate body mass index; CHD, coronary heart disease.
TABLE 6.
Multivariable Analysis of Risk Factors With Regard to the Morbidity Defined as Missing a Textbook Outcome After DP in 1981 Patients (154 Patients With Missing Values Excluded)
| Parameters | Category | OR | 95% CI | P |
|---|---|---|---|---|
| Estimated blood loss | ≥500 vs <500 mL | 1.435 | 1.137–1.812 | 0.0024 |
| ≥1000 vs <500 mL | 2.018 | 1.501–2.715 | <0.0001 | |
| Operation time | >195 vs ≤195 min | 2.413 | 1.124–5.291 | 0.0211 |
| >240 vs ≤195 min | 3.172 | 1.421–7.431 | 0.0036 | |
| DP type | Type 2/3/4 vs type 1 | 1.385 | 1.114–1.719 | 0.0032 |
| ASA classification | ASA III/IV vs ASA I/II | 1.322 | 1.078–1.620 | 0.0072 |
| Not included | ||||
| Sex | Male vs female | 0.9599 | ||
| Splenectomy | Yes vs no | 0.8653 | ||
| Diabetes mellitus | Yes vs no | 0.8562 | ||
| Carcinoma | Yes vs no | 0.8259 | ||
| BMI | ≥30 vs <30 | 0.2167 | ||
| CHD/heart failure | Yes vs no | 0.1172 | ||
BMI indicates body mass index; CHD, coronary heart disease.
DISCUSSION
To our knowledge, this is the largest retrospective single center study analyzing outcomes after DP with special attention to the extent of surgery. Overall morbidity and mortality rates were comparable to previously published cohorts from high-volume centers.23–26 Yet, these earlier studies did not take into account the extent and different degrees of surgical complexity. Just recently, the concept of extended versus standard pancreatic resections has gained general attention, and the ISGPS proposed a consensus definition for extended pancreatectomy, which, however, subsumes a broad spectrum of surgical procedures differing in complexity.10 Mihaljevic et al11 and Loos et al12 further dissected the term “extended” pancreatectomy and proposed a 4-stage classification system for pancreatoduodenectomy and total pancreatectomy, respectively, which allows the distinction of surgical procedures in terms of additional vascular (venous vs arterial) and/or organ resections. Both studies showed that postoperative morbidity and mortality correlate with the extent of surgery and clearly differ between the 4 types. The same investigation has not been made for DP, yet. Previous studies comparing standard versus extended DP, mostly involving multivisceral resections, reported disparate results concerning short-term outcomes. While several retrospective cohort studies showed no significant differences in morbidity and mortality after extended DP compared with standard DP,27–29 other studies reported increased complication rates, especially higher incidence of POPF for additional organ resections.30,31 Yet, in a large multi-national, retrospective study both concomitant splenectomy and the need for vascular resection, but not multivisceral resections, were associated with a greater risk of POPF.26 In a large systematic review and meta-analysis, Chong et al32 demonstrated no difference in POPF rates between multivisceral and standard DP, and no association of (unspecified) vascular resection with POPF.
In the present study, we investigated in detail whether different degrees of surgical complexity are associated with different short-term outcomes after DP. As shown for pancreatoduodenectomy and total pancreatectomy,11,12 our proposed 4-type classification of DP based on the extent of surgery and technical complexity finds a strong correlation with postoperative morbidity and mortality, confirming the 4 types as different risk categories. We can clearly demonstrate that DP is not a homogenous surgical procedure and that it makes sense to differentiate between well-defined types of surgery with different risk profile. For example, twice the number of patients suffered from serious adverse events once extended DP was performed, as well as the rate of relaparotomies was significantly higher, especially after arterial resections. Grade B/C POPF, severe PPH, DGE, and chyle leakage were also more frequent after types 2 to 4, compared with type 1 (standard DP). There is a close correlation of 90-day mortality with surgical complexity with the highest rate after type 4. Univariable and multivariable analyses confirmed arterial resection (type 4) as an independent risk factor for 90-day mortality, as well as for missing a textbook outcome. Thus, the proposed simply applicable classification of DP allows for a better surgical risk stratification, which could be useful for clinical decision-making and comparability of results by standardized reporting. This study has several limitations, mainly due to its design as a retrospective single center observational study over a long period of time. The results come from a high-volume center for pancreatic surgery. Since hospital volume is known to be an essential factor for postoperative outcome after major pancreatic surgery,33–36 morbidity and mortality might be different in hospitals with lower case load.37 However, more complex surgery is usually performed at large centers and the results may be representative for other national and international specialized pancreatic surgery centers.38 Still, generalizability is limited and it would be desirable to validate the findings in a multicenter study. Given the retrospective study design, selection bias due to differences in patient baseline characteristics among the 4 different types of DP needs to be considered, and varying outcomes might not only be explained by the surgical procedure itself. Although we were not able to control for confounders, we aimed to include known postoperative short-term outcome-related factors into our analyses.
In conclusion, our 4-type classification system proved that not all DP are comparable in terms of short-term outcomes. It is reasonable to discriminate DPs in terms of the extent and complexity of surgery. The proposed classification is simple and reproducible and allows for a better risk assessment for the individual patient and better comparability of future studies.
Footnotes
M.L. and C.E.M. contributed equally.
M.L. and C.E.M. have made substantial contributions to the conception and design of the study, acquisition of data, analysis and interpretation of data. Both authors contributed equally to this work. A.T.L.X. and M.H. have made substantial contributions to the acquisition, analysis and interpretation of data. U.H., S.R., and T.H. have made substantial contributions to the conception and design of the study, analysis and interpretation of data. A.M., C.B., M.S., M.A.-S., and M.W.B. have made substantial contributions to the conception and design of the study and interpretation of data. All authors have participated in drafting the article or revising it critically for important intellectual content and all authors have given final approval of this version to be published.
The authors report no conflicts of interest.
Contributor Information
Martin Loos, Email: martin.loos@med.uni-heidelberg.de.
Claudia E. Mack, Email: claudia.mack@med.uni-heidelberg.de.
An Ting L. Xu, Email: an.xu@stud.uni-heidelberg.de.
Matthias Hassenpflug, Email: matthias.hassenpflug@med.uni-heidelberg.de.
Ulf Hinz, Email: ulf.hinz@med.uni-heidelberg.de.
Arianeb Mehrabi, Email: arianeb.mehrabi@med.uni-heidelberg.de.
Christoph Berchtold, Email: Christoph.Berchtold@med.uni-heidelberg.de.
Martin Schneider, Email: Martin.Schneider@med.uni-heidelberg.de.
Mohammed Al-Saeedi, Email: Mohammed.Al-Saeedi@med.uni-heidelberg.de.
Susanne Roth, Email: Susanne.Roth@med.uni-heidelberg.de.
Thilo Hackert, Email: t.hackert@uke.de.
Markus W. Büchler, Email: markus.buechler@med.uni-heidelberg.de.
REFERENCES
- 1. Loos M, Kester T, Klaiber U, et al. Arterial resection in pancreatic cancer surgery: effective after a learning curve. Ann Surg. 2022;275:759–768. [DOI] [PubMed] [Google Scholar]
- 2. Tee MC, Krajewski AC, Groeschl RT, et al. Indications and perioperative outcomes for pancreatectomy with arterial resection. J Am Coll Surg. 2018;227:255–269. [DOI] [PubMed] [Google Scholar]
- 3. Kinny-Köster B, Habib JR, van Oosten AF, et al. Conduits in vascular pancreatic surgery: analysis of clinical outcomes, operative techniques and graft performance. Ann Surg. 2022;278:e94–e104. [DOI] [PubMed] [Google Scholar]
- 4. Besselink MG, van Rijssen LB, Bassi C, et al. Definition and classification of chyle leak after pancreatic operation: a consensus statement by the International Study Group on Pancreatic Surgery. Surgery. 2017;161:365–372. [DOI] [PubMed] [Google Scholar]
- 5. Strobel O, Neoptolemos J, Jager D, et al. Optimizing the outcomes of pancreatic cancer surgery. Nat Rev Clin Oncol. 2019;16:11–26. [DOI] [PubMed] [Google Scholar]
- 6. Hartwig W, Hackert T, Hinz U, et al. Multivisceral resection for pancreatic malignancies: risk-analysis and long-term outcome. Ann Surg. 2009;250:81–87. [DOI] [PubMed] [Google Scholar]
- 7. Kleeff J, Diener MK, Z’Graggen K, et al. Distal pancreatectomy: risk factors for surgical failure in 302 consecutive cases. Ann Surg. 2007;245:573–582. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Merath K, Chen Q, Bagante F, et al. Synergistic effects of perioperative complications on 30-day mortality following hepatopancreatic surgery. J Gastrointest Surg. 2018;22:1715–1723. [DOI] [PubMed] [Google Scholar]
- 9. Hicks CW, Tosoian JJ, Craig-Schapiro R, et al. Early hospital readmission for gastrointestinal-related complications predicts long-term mortality after pancreatectomy. Am J Surg. 2015;210:636–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hartwig W, Vollmer CM, Fingerhut A, et al. Extended pancreatectomy in pancreatic ductal adenocarcinoma: definition and consensus of the International Study Group for Pancreatic Surgery (ISGPS). Surgery. 2014;156:1–14. [DOI] [PubMed] [Google Scholar]
- 11. Mihaljevic AL, Hackert T, Loos M, et al. Not all Whipple procedures are equal: proposal for a classification of pancreatoduodenectomies. Surgery. 2021;169:1456–1462. [DOI] [PubMed] [Google Scholar]
- 12. Loos M, Al-Saeedi M, Hinz U, et al. Categorization of differing types of total pancreatectomy. JAMA Surg. 2022;157:120–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370:1453–1457. [DOI] [PubMed] [Google Scholar]
- 14. Bassi C, Marchegiani G, Dervenis C, et al. The 2016 update of the International Study Group (ISGPS) definition and grading of postoperative pancreatic fistula: 11 years after. Surgery. 2017;161:584–591. [DOI] [PubMed] [Google Scholar]
- 15. Wente MN, Veit JA, Bassi C, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery. 2007;142:20–25. [DOI] [PubMed] [Google Scholar]
- 16. Wente MN, Bassi C, Dervenis C, et al. Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS). Surgery. 2007;142:761–768. [DOI] [PubMed] [Google Scholar]
- 17. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240:205–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Merath K, Chen Q, Bagante F, et al. Textbook outcomes among medicare patients undergoing hepatopancreatic surgery. Ann Surg. 2020;271:1116–1123. [DOI] [PubMed] [Google Scholar]
- 19. Chen Q, Diaz A, Beane J, et al. Achieving an optimal textbook outcome following pancreatic resection: the impact of surgeon specific experience in achieving high quality outcomes. Am J Surg. 2022;225:499–503. [DOI] [PubMed] [Google Scholar]
- 20. Hyer JM, Beane JD, Spolverato G, et al. Trends in textbook outcomes over time: are optimal outcomes following complex gastrointestinal surgery for cancer increasing? J Gastrointest Surg. 2022;26:50–59. [DOI] [PubMed] [Google Scholar]
- 21. van Roessel S, Mackay TM, van Dieren S, et al. Textbook outcome: nationwide analysis of a novel quality measure in pancreatic surgery. Ann Surg. 2020;271:155–162. [DOI] [PubMed] [Google Scholar]
- 22. Hackert T, Klaiber U, Hinz U, et al. Portal vein resection in pancreatic cancer surgery: risk of thrombosis and radicality determine survival. Ann Surg. 2022;277:e1291–1298. [DOI] [PubMed] [Google Scholar]
- 23. Daniel FE, Tamim HM, Hosni MN, et al. Short-term surgical morbidity and mortality of distal pancreatectomy performed for benign versus malignant diseases: a NSQIP analysis. Surg Endosc. 2020;34:3927–3935. [DOI] [PubMed] [Google Scholar]
- 24. Lof S, van der Heijde N, Abuawwad M, et al. Robotic versus laparoscopic distal pancreatectomy: multicentre analysis. Br J Surg. 2021;108:188–195. [DOI] [PubMed] [Google Scholar]
- 25. Nathan H, Cameron JL, Goodwin CR, et al. Risk factors for pancreatic leak after distal pancreatectomy. Ann Surg. 2009;250:277–281. [DOI] [PubMed] [Google Scholar]
- 26. Ecker BL, McMillan MT, Allegrini V, et al. Risk factors and mitigation strategies for pancreatic fistula after distal pancreatectomy: analysis of 2026 resections from the International, Multi-institutional Distal Pancreatectomy Study Group. Ann Surg. 2019;269:143–149. [DOI] [PubMed] [Google Scholar]
- 27. Irani JL, Ashley SW, Brooks DC, et al. Distal pancreatectomy is not associated with increased perioperative morbidity when performed as part of a multivisceral resection. J Gastrointest Surg. 2008;12:2177–2182. [DOI] [PubMed] [Google Scholar]
- 28. Malinka T, Klein F, Andreou A, et al. Distal pancreatectomy combined with multivisceral resection is associated with postoperative complication rates and survival comparable to those after standard procedures. J Gastrointest Surg. 2018;22:1549–1556. [DOI] [PubMed] [Google Scholar]
- 29. Panzeri F, Marchegiani G, Malleo G, et al. Distal pancreatectomy associated with multivisceral resection: results from a single centre experience. Langenbecks Arch Surg. 2017;402:457–464. [DOI] [PubMed] [Google Scholar]
- 30. Ferrone CR, Warshaw AL, Rattner DW, et al. Pancreatic fistula rates after 462 distal pancreatectomies: staplers do not decrease fistula rates. J Gastrointest Surg. 2008;12:1691–1697; discussion 1697–1698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Yang F, Jin C, Hao S, et al. Drain contamination after distal pancreatectomy: incidence, risk factors, and association with postoperative pancreatic fistula. J Gastrointest Surg. 2019;23:2449–2458. [DOI] [PubMed] [Google Scholar]
- 32. Chong E, Ratnayake B, Lee S, et al. Systematic review and meta-analysis of risk factors of postoperative pancreatic fistula after distal pancreatectomy in the era of 2016 International Study Group pancreatic fistula definition. HPB (Oxford). 2021;23:1139–1151. [DOI] [PubMed] [Google Scholar]
- 33. Diaz A, Paredes AZ, Hyer JM, et al. Variation in value among hospitals performing complex cancer operations. Surgery. 2020;168:106–112. [DOI] [PubMed] [Google Scholar]
- 34. Krautz C, Nimptsch U, Weber GF, et al. Effect of hospital volume on in-hospital morbidity and mortality following pancreatic surgery in Germany. Ann Surg. 2018;267:411–417. [DOI] [PubMed] [Google Scholar]
- 35. Hata T, Motoi F, Ishida M, et al. Effect of hospital volume on surgical outcomes after pancreaticoduodenectomy: a systematic review and meta-analysis. Ann Surg. 2016;263:664–672. [DOI] [PubMed] [Google Scholar]
- 36. Ratnayake B, Pendharkar SA, Connor S, et al. Patient volume and clinical outcome after pancreatic cancer resection: a contemporary systematic review and meta-analysis. Surgery. 2022;172:273–283. [DOI] [PubMed] [Google Scholar]
- 37. Güller U, Warschkow R, Ackermann CJ, et al. Lower hospital volume is associated with higher mortality after oesophageal, gastric, pancreatic and rectal cancer resection. Swiss Med Wkly. 2017;147:w14473. [DOI] [PubMed] [Google Scholar]
- 38. Ahola R, Sand J, Laukkarinen J. Centralization of pancreatic surgery improves results: review. Scand J Surg. 2020;109:4–10. [DOI] [PubMed] [Google Scholar]