Arguments for an individualized closure of the pancreatic remnant after distal pancreatic resection

Table 1 Patients and operations, univariate and multivariate analysis of risk factors for the occurrence of postoperative pancreatic fistula B/C, median (range) n (%)

Parameter		No POPF (n = 96)	POPF (n = 30)	Odds ratio	P uni-variate	P multi-variate
Patients
Age (yr)	61 (24-83)	61 (24-82)	61 (24-83)	1.002	0.859
Sex (M:F)	49:77	38:58	11:19	0.884	0.775
BMI (kg/m2)	24 (16-41)	24 (16-34)	27 (17-41)	1.181	0.001	0.009
Diabetes	24 (19.0)	20	4	0.585	0.365
Alcohol	23 (18.3)	19	4	0.623	0.427
Crea (mg/dL)	0.80 (0.40-1.87)	0.80 (0.40-1.87)	0.80 (0.46-1.40)	0.806	0.827
WBC (tsd/μL)	6.9 (2.6-18.5)	6.9 (2.6-18.5)	6.8 (3.2-17.7)	0.978	0.759
Hb (g/dL)	13.2 (7.8-16.8)	13.1 (7.8-16.8)	13.4 (10.7-16.6)	1.189	0.174
Bili (mg/dL)	0.6 (0.2-1.8)	0.6 (0.2-1.4)	0.6 (0.2-1.8)	0.635	0.545
Operations
Period 94-01	43 (34)	37	6	2.508	0.067
Period 02-09	83 (66)	59	24
OP time	270 (125-570)	270 (125-570)	269 (157-510)	0.996	0.154
DC-HS	47 (37.3)	30	17	2.877	0.014	0.030
DC-S	18 (14.3)	16	2	0.357	0.188
PJ	52 (41.3)	43	9	0.528	0.154
SMP	9 (7.1)	7	2	0.908	0.908
Splenectomy	109 (86.5)	84	25	0.714	0.561
Multivisceral	26 (20.6)	23	3	0.353	0.111
Histopathology
PDAC	38 (28.6)	32	4	0.308	0.042
CP	40 (31.7)	32	8	0.727	0.495
CNP	21 (16.7)	12	9	3.000	0.029
NET	16 (12.7)	12	4	1.077	0.905
OTH	13 (10.3)	8	5	2.200	0.199
High-risk	50 (39.5)	32	18	3.000	0.011	0.168

Univariate and multivariate analysis: binary logistic regression. POPF: Postoperative pancreatic fistula (International Study Group for Pancreatic Surgery definition); CI: Confidence interval; BMI: Body mass index; DC-HS: Direct closure hand suture; DC-S: Direct closure stapler; PJ: Pancreatojejunostomy; SMP: Seromuscular patch; PDAC: Pancreatic ductal adenocarcinoma; CP: Chronic pancreatitis; CNP: Cystic neoplasia of the pancreas; NET: Neuroendocrine tumor; OTH: Other pancreatic pathology. CNPs were: 2 intraductal papillary mucinous neoplasia, 2 solid pseudopapillary neoplasia, 11 serous cystic adenoma, 6 mucinous cystic neoplasia. High-risk pancreas is defined as pathology with odds ratio > 1 i.e., other than PDAC or CP, periods of time are 1994-2001 vs 2002-2009.

Table 2 Factors influencing postoperative pancreatic fistula rate in different risk groups n (%)

Technique	Low-risk pancreas1			High-risk pancreas2
	POPF	CC	P value	POPF	CC	P value
Hand suture	8/26 (30.8)	0.296	0.009	9/21 (42.9)	0.122	0.400
PJ	4/36 (11.1)	-0.122	0.295	5/16 (31.3)	-0.068	0.639
Stapler	0/11 (0)	-0.178	0.124	2/7 (28.6)	-0.062	0.667
SM patch	0/3 (0)	-0.088	0.451	2/6 (33.3)	-0.021	0.888
Risk factor
BMI (kg/m2)	24 (17-30) vs 23 (16-32)3	0.178	0.125	29 (21-41) vs 25 (17-34)3	0.349	0.013

CC: Correlation coefficient; POPF: Postoperative pancreatic fistula (Grade B or C of the International Study Group for Pancreatic Surgery definition); BMI: Body mass index, high-risk pancreas defined as pathology with odds ratio > 1 (see Table 1); PJ: Pancreatojejunostomy; SM: Patch seromuscular patch.

¹POPF rate: 12/76 (15.8%);

²POPF rate: 18/50 (36.0%);

³Median (range) in POPF vs no POPF. The analyses were carried out for each risk group separately. P value given for correlation with occurrence of POPF (two-sided Spearman rank test).

Table 3 Postoperative pancreatic fistula is the main factor of perioperative morbidity after distal pancreatic resection

Parameter description for all patients (n = 126)		Groups		Correlation
Parameter	n (%)	No POPF (n = 96)	POPF (n = 30)	Coefficient	P value
POPF	30 (23.8)	96	30	1.000	NA
Overall morbidity	76 (60.3)	46	30	0.453	< 0.001
Surgical morbidity	55 (43.7)	25	30	0.635	< 0.001
Severe morbidity	19 (15.1)	11	8	0.181	0.043
Inta-abdominal abscess	17 (13.5)	5	12	0.434	< 0.001
Septic shock	3 (2.4)	1	2	0.157	0.079
PPH	8 (6.3)	4	4	0.160	0.073
Reoperation	18 (14.3)	10	8	0.198	0.026
Overall mortality	2 (1.6)	1	1	0.078	0.385
Hospital stay (d), median (range)	15 (8-143)	14 (8-143)	32 (11-108)	0.552	< 0.001

POPF: Postoperative pancreatic fistula of Grade B or C [International Study Group for Pancreatic Surgery (ISGPS) definition]; PPH: postpancreatectomy hemorrhage (ISGPS definition), severe morbidity includes complications leading to sepsis, reintubation or reoperation. Septic shock is defined as sepsis with iv catecholamin requirement. Correlations derived from two-sided Spearman’s rank correlation test. NA: Not applicable.

Table 4 Distal pancreatic resection - changing indications, demographics and operative techniques n (%)

Parameter	1994-2001 (n = 43)	2002-2009 (n = 83)	P value
Patients
Age (yr, median)	51	64	0.001
BMI (kg/m2, median)	25	24	0.713
Histology
PDAC	8 (19)	28 (34)	0.097
CP	22 (51)	18 (22)	0.001
CNP	3 (7)	18 (22)	0.044
NET	5 (12)	11 (13)	1.000
Other	5 (12)	8 (10)	0.762
High-risk pancreas	13 (30)	37 (45)	0.129
Operations
Multivisceral resections	9 (21)	17 (21)	1.000
Hand suture closure	13 (30)	34 (41)	0.252
Pancreatojejunostomy	30 (70)	22 (27)	< 0.001
Stapler closure	0 (0)	18 (22)	< 0.001
Seromuscular patch	0 (0)	9 (11)	0.027
Perioperative parameters
POPF B/C	6 (14)	24 (29)	0.078
OHS (d)	13	15	0.036

POPF: Postoperative pancreatic fistula of Grade B or C (International Study Group for Pancreatic Surgery definition); BMI: Body mass index; PDAC: Pancreatic ductal adenocarcinoma; CNP: Cystic neoplasia of the pancreas; NET: Neuroendocrine tumor; OTH: Other pancreatic pathology, high-risk pancreas defined as not PDAC or CP; OHS: Overall hospital stay, P value given for two-sided Fisher’s exact test.