Systematic review and meta-analysis of prophylactic abdominal drainage after pancreatic resection

Table 1 Definitions of clinical outcomes in each included study

Author	Mortality	Morbidity	POPF	Abdominal abscess
Heslin et al[6]	NA	NA	Drain output at a rate of ≥ 30 mL/d or more and lasting for more than 7 d	Abdominal collection associated with fever and a positive culture requiring either percutaneous or operative drainage yielding positive cultures
Conlon et al[8]	Deaths within 30 d of surgery	NA	Drain output on postoperative day 5 or > 30 mL and amylase level > 150 IU/L and/or three times greater than the serum value	Abdominal collection associated with fever and a positive culture requiring either surgical or radiologic drainage
Fisher et al[7]	Deaths within 30 d of surgery.	CTCAE (v4.0)[35]	ISGPF[36]	Abdominal collection with a positive Gram stain or cultures
Paulus et al[26]	NA	NA	ISGPF[36]	Abdominal collection associated with fever, abnormal blood routine test, and positive cultures
Adham et al[29]	Deaths within 90 d of surgery	Clavien classification[37]	ISGPF[36]	Abdominal collection associated with fever and a positive culture requiring surgical drain or interventional treatment
Correa-Gallego et al[9]	Deaths within 90 d of surgery	CTCAE (v4.0)[35]	Clinical signs and symptoms with amylase-rich drainage > 50 mL/d beyond postoperative day 10	Clinical signs and symptoms or radiologic diagnosis of abdominal abscess or peritonitis
Lim et al[27]	Clavien classification[37]	Clavien classification[37]	ISGPF[36]	NA
Mehta et al[28]	Deaths within 30 d of surgery	Clavien classification[37]	ISGPF[36]	NA
Van Buren et al[10]	Deaths within 90 d of surgery	CTCAE (v4.0)[35]	ISGPF[36]	NA

NA: Not available; CTCAE (v4.0): Common terminology criteria for adverse effects, version 4.0; ISGPF: International Study Group of Pancreatic Fistula.

Table 2 Characteristics of included studies

Author, year	Country	Design	No. of patients	Group	Age (yr)	Male:female	Operation type: No. of patients
Heslin et al[6]	United States	OCS	89	Drain	65 ± 2	18:20 (58.1)	PD: 51
1998				No drain	65 ± 2	32:19	PD: 38
Conlon et al[8]	United States	RCT	179	Drain	66 (23-81)	46:42 (49.7)	PD: 73, DP: 15
2001				No drain	69 (33-87)	43:48	PD: 66, DP: 25
Fisher et al[7]	United States	OCS	228	Drain	63 (53-72)	78:101 (40.7)	PD: 123, DP: 56
2011				No drain	59 (51-70)	19:40	PD: 30, DP: 17
Paulus et al[26]	United States	OCS	59	Drain	52 (44-66)	NA	DP: 39
2012				No drain	58 (52-68)	NA	DP: 30
Adham et al[29]	France	OCS	242	Drain	61.5 (20-85)	66:64 (52.4)	PD: 79, DP: 29, Others: 22
2013				No drain	66.5 (19-85)	61:51	PD: 69, DP: 37, Others: 6
Correa-Gallego et al[9]	United States	OCS	739 (Subgroup A of PD)	Drain	NA	NA	PD: 386
2013				No drain	NA	NA	PD: 353
			350 (Subgroup B of DP)	Drain	NA	NA	DP: 154
				No drain	NA	NA	DP: 196
Lim et al[27]	France	OCS	54	Drain	62 (40-76)	8:19 (29.6)	PD: 27
2013				No drain	62 (38-78)	8:19	PD: 27
Mehta et al[28]	United States	OCS	709	Drain	60	130:121	PD: 251
2013				No drain	62.5	232:236	PD: 458
Van Buren et al[10]	United States	RCT	137	Drain	62.1 ± 11.7	37:31	PD: 68
2013				No drain	64.3 ± 12.6	38:31	PD: 69

OCS: Observational comparative study; PD: Pancreaticoduodenectomy; DP: Distal pancreatectomy; Others: Enucleation and Central pancreatectomy; NA: Not available.

Table 3 Comparability between drained patients and non-drained patients

Author	Comorbidity	Preoperative treatment	Preoperative biochemical test	Pathology	Length of operation	Estimated blood loss	Texture of pancreas	Diameter of pancreatic duct
Heslin et al[6]	Comparable	Comparable	Comparable	Comparable	NA	Comparable	NA	NA
Conlon et al[8]	NA	Comparable	NA	Comparable	Comparable	Comparable	NA	NA
Fisher et al[7]	Significant difference	NA	Significant difference	Comparable	Comparable	Significant difference	Comparable	Comparable
Paulus et al[26]	NA	NA	NA	Comparable	Comparable	Comparable	Comparable
Adham et al[29]	Comparable	Comparable	Comparable	Comparable	NA	NA	NA	NA
Correa-Gallego et al[9]
PD subgroup	NA	Comparable	NA	Comparable	Significant difference	Significant difference	Significant difference	Comparable
DP subgroup	NA	NA	NA	Comparable	Significant difference	Significant difference	NA	Significant difference
Lim et al[27]	Comparable	Comparable	Comparable	Comparable	Comparable	Comparable	Comparable	Comparable
Mehta et al[28]	Comparable	Comparable	Comparable	Comparable	Significant difference	Significant difference	NA	Significant difference
Van Buren et al[10]	Comparable	Comparable	Comparable	Comparable	Comparable	Comparable	Comparable	Comparable

NA: Not available; PD: Pancreaticoduodenectomy; DP: Distal pancreatectomy.

Table 4 Quality of assessment of included studies

Cohort studies	Representativeness of the exposed cohort	Selection of the non-exposed cohort	Ascertainment of exposure	Comparability between the two cohorts	Assessment of outcome	Length of follow-up
Heslin et al[6]	Potential selection bias	Same patient base	Surgical record	No restriction/matching	Independent assessment	NM
Paulus et al[26]	Representative	Same patient base	Surgical record	No restriction/matching	Surgical record	NM
Fisher et al[7]	Representative	Different patient base	Surgical record	No restriction/matching	Surgical record	30 d
Adham et al[29]	Representative	Same patient base	Surgical record	No restriction/matching	Surgical record	90 d
Correa-Gallego et al[9]	Representative	Same patient base	Surgical record	No restriction/matching	Surgical record	90 d
Metha et al[28]	Representative	Same patient base	Surgical record	No restriction/matching	Surgical record	90 d
Case-control study	Representativeness of the cases	Selection of Controls	Ascertainment of exposure	Comparability of cases and controls	Assessment of outcome	Definition of Controls and cases
Lim et al[27]	Potential selection bias	Hospital control	Surgical record	One to one matching	Surgical record	Surgical record
RCTs	Random sequence generation	Allocation concealment	Blinding of participants and personnel	Blinding of outcome assessment	Incomplete outcome data	Selective reporting
Conlon et al[8]	Low risk	Low risk	High risk	Low risk	Unclear risk	Low risk
Van Buren et al[10]	Low risk	Low risk	High risk	Unclear risk	Low risk	Low risk

RCT: Randomized controlled trial.

Table 5 Summary of results

Outcome of interest	Studies	Patients		Results		Pooled estimates	P value	P value for HG	I2
		No drainage	Drainage	No drainage	Drainage	(95%CI)
Mortality
Overall analysis	7	1353	1283	2.96%	1.87%	1.56 (0.93-2.62)	0.09	0.31	15%
Restricted analysis of RCTs	2	160	150	6.25%	2.56%	2.55 (0.79-8.30)	0.12	0.25	26%
Subgroup analysis of PD	5	954	911	3.35%	1.32%	2.39 (1.22-4.69)	0.01	0.52	0%
Overall morbidity
Overall analysis	9	1421	1373	43.54%	52.59%	0.69 (0.52-0.92)	0.01	0.01	58%
Restricted analysis of RCTs	2	160	150	66.88%	67.31%	1.00 (0.58-1.72)	1.00	0.26	20%
Subgroup analysis of PD	5	945	783	46.35%	51.34%	0.69 (0.56-0.84)	< 0.01	0.23	28%
Subgroup analysis of DP	2	226	193	27.88%	33.68%	1.29 (0.24-6.81)	0.76	< 0.01	89%
POPF
Overall analysis	7	1292	1234	13.78%	27.55%	0.55 (0.42-0.72)	< 0.01	0.07	46%
Subgroup analysis of PD	4	907	732	13.34%	26.23%	0.46 (0.35-0.59)	< 0.01	0.24	28%
Subgroup analysis of DP	2	226	193	16.81%	24.87%	0.39 (0.07-2.21)	0.29	0.17	46%
CR-PF
Overall analysis	6	743	694	9.02%	13.26%	0.72 (0.33-1.59)	0.42	< 0.01	69%
Subgroup analysis of PD	3	554	346	8.84%	15.32%	0.61 (0.14-2.66)	0.51	< 0.01	81%
Abdominal abscess
Overall analysis	7	414	582	11.84%	8.59%	1.29 (0.84-1.98)	0.25	0.34	11%
Restricted analysis of RCTs	2	160	150	15.00%	7.70%	1.95 (0.53-7.16)	0.32	0.09	65%
Subgroup analysis of PD	3	134	146	14.18%	6.85%	2.12 (0.95-4.72)	0.07	0.13	50%
Interventional radiology drainage
Overall analysis	8	1309	1243	11.38%	12.31%	1.05 (0.69-1.62)	0.81	0.03	52%
Restricted analysis of RCTs	2	160	150	14.38%	10.90%	1.35 (0.26-6.97)	0.72	0.02	81%
Subgroup analysis of PD	5	945	783	10.16%	12.52%	0.87 (0.65-1.19)	0.39	0.13	43%
Subgroup analysis of DP	2	226	193	18.14%	20.73%	1.03 (0.38-2.80)	0.95	0.13	57%
Reoperation
Overall analysis	9	1421	1373	4.71%	4.73%	1.01 (0.70-1.47)	0.95	0.59	0%
Restricted analysis of RCTs	2	160	150	16.67%	6.41%	1.11 (0.17-7.29)	0.91	0.07	70%
Subgroup analysis of PD	5	945	783	4.02%	2.68%	1.26 (0.73-2.17)	0.41	0.51	0%
Subgroup analysis of DP	2	226	193	3.54%	6.22%	0.80 (0.29-2.17)	0.66	0.46	0%
Length of hospital stay
Overall analysis	9	1421	1373	-	-	-0.96 [-1.74-(-0.18)]	0.02	< 0.01	92%
Restricted analysis of RCTs	2	160	150	-	-	0.78 (-0.40-1.97)	0.19	0.49	0%
Subgroup analysis of PD	5	945	783	-	-	-0.75 (-1.73-0.24)	0.14	< 0.01	85%
Subgroup analysis of DP	2	226	193	-	-	-2.10 [-2.46-(-1.73)]	< 0.01	0.29	11%

PD: Pancreaticoduodenectomy; DP: Distal pancreatectomy; RCT: Randomized clinical trial.