If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
No consensus was reached with regard to the effect of EDR on postoperative outcomes after pancreatic surgery. The meta-analysis was designed to explore the efficacy and safety of early drain removal (EDR).
Methods
Systematic literature search was performed. Data extraction and correction were performed by three researchers. For dichotomous and continuous outcomes, we calculated the pooled risk difference and mean difference with 95% confidence intervals, respectively. The heterogeneity of included studies was evaluated using Cochran's Q and I2 test. The stratified analyses of pancreaticoduodenectomy (PD) and distal pancreatectomy (DP) were performed.
Results
A total of 10 studies including 3 RCTs and 7 non RCTs were included for meta-analysis, among which 1780 patients with EDR and 5613 patients with late drain removal (LDR) were enrolled. The meta-analysis of both all the available studies and studies only with selected low risk patients indicated that EDR group had significantly lower incidences of Grade B/C postoperative pancreatic fistula (POPF) and total complications for both PD and DP. However, no advantages of EDR were observed in the meta-analysis of the 3 RCTs. In addition, EDR was associated with a lower incidence of intra-abdominal infection after PD. While for DP, EDR group had decreased risk of delayed gastric emptying and re-operation, and shorter postoperative in-hospital stay.
Conclusions
The meta-analysis demonstrates that EDR is effective and safe for both PD and DP considering POPF and total complications, especially for patients with low concentration of postoperative drain fluid amylase.
Introduction
Along with the rapid progress of surgical technique in recent years, the perioperative mortality of pancreatic surgery decreased significantly, whereas the incidences of postoperative complications are still high.
Survey on the current status of the diagnosis and treatment of pancreatic cancer in public tertiary hospitals in China: a cross-sectional questionnaire-based, observational study.
The postoperative pancreatic fistula (POPF) is one of the most important complications after pancreatic surgery, which significantly increases postoperative in-hospital stay and medical expenses.
The management of abdominal drainage plays a pivotal role in the postoperative recovery process of patients undergoing pancreatic resection. Observation and detection of the effusion around the operative area is considered as a crucial indicator for early identification of POPF, postpancreatectomy hemorrhage or intra-abdominal infection, so prophylactic drainage placement during pancreatic surgery is widely accepted.
However, a growing body of study raised concerns about the disadvantages of intraperitoneal drainage for both pancreaticoduodenectomy (PD) and distal pancreatectomy (DP). In 2001, Conlon et al.
conducted the first randomized controlled trial (RCT) and reported that the drainage placement after pancreatic resection failed to reduce postoperative complications, but increased the incidence of intra-abdominal collections and infection. Thereafter, some RCTs and meta-analyses proved the safety and feasibility of omission of drainage for pancreatic resection.
The value of drains as a fistula mitigation strategy for pancreatoduodenectomy: something for everyone? Results of a randomized prospective multi-institutional study.
However, one RCT had been stopped by the Data Safety Monitoring Board due to the significant increase in mortality (3%–12%) for patients undergoing PD without intraperitoneal drainage.
Currently, routine drainage placement during pancreatic surgery is accepted in most pancreatic centers. More attention was paid to the early drain removal (EDR) on the premise of safety, in order to accelerate the recovery of patients, or even to reduce the incidences of drainage-related complications.
conducted the first RCT to compare EDR and late drain removal (LDR) for patients underwent standard pancreatic resections and at low risk of POPF, and reported that a prolonged period of drain insertion increased complications and in-hospital stay. Subsequently, Dai et al.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
conducted single and multiple-center RCTs to evaluate the safety of EDR in patients with low or middle risk of POPF, which demonstrated EDR is safe in selected patients. The American College of Surgeons’ National Surgical Quality Improvement Program (ACS-NSQIP) was utilized to explore the effect of EDR on postoperative complications for PD and DP, and the results showed that EDR after both PD and DP was associated with better outcomes.
However, no consensus was wildly reached about the influences of EDR, and there was no meta-analysis for the efficacy and safety evaluation of EDR after pancreatic resection. Furthermore, selection criteria of low risk patients and the time-point of EDR varied in previous studies. Thus, the efficacy and safety of EDR following pancreatic resections remains to be investigated.
Here, we conducted the first meta-analysis to evaluate the efficacy and safety of EDR. The stratified analyses of PD and DP, and subgroup analyses of RCTs and non RCTs were performed. In particular, if the meta-analysis of RCTs showed inconsistent results with that of all available studies, further meta-analysis for studies that only recruited patients with low concentration of postoperative DFA, who were considered with low risk of POPF, was conducted.
Methods
Literature search
Systematic literature search was conducted by two researchers at PubMed, Embase, MEDLINE, Web of Science, Cochrane Central Register of Controlled Trials until 29 September 2021. For PubMed, the search term was (drainage [Title] OR drain [Title] OR drains [Title] OR peritoneal drainage [Title]) AND (pancreas OR pancreatectomy OR pancreaticoduodenectomy OR Whipple OR distal pancreatectomy OR pancreatic resection). We also used synonyms for drainage (e.g. suction). The search strategy was appropriately translated for the other databases. In addition, references of all retrieved studies were manually searched to identify additional eligible studies. Only English literature was included in this study.
Inclusion and exclusion criteria
Inclusion criteria: (a) the types of pancreatic surgery were clearly defined as PD and/or DP; (b) postoperative outcomes between EDR and LDR were compared; (c) prospective or retrospective studies; (d) provided data on any complications. Exclusion criteria: (a) review or case report; (b) related postoperative outcomes were not reported; (c) redundant publications.
Data extraction and quality assessment
Data extraction and correction were performed by three researchers. Retrieved variables included: (a) general characteristics: author, publication year, country, cohort span, low risk patient selection criteria, and the definition of early/late drain removal; (b) demographic and surgical characteristics: age, gender, body mass index (BMI), diabetes, intraoperative RBC transfusion, operation time, pancreatic texture, diameter of main pancreatic duct; (c) primary outcomes: Grade B/C POPF, total complications; (d) secondary outcomes: postoperative hemorrhage, intra-abdominal infection, postoperative in-hospital stay, delayed gastric emptying (DGE), biliary fistula, intra-abdominal fluid collections, wound infection, pulmonary complications, intervention, re-admission, re-operation. The quality of included studies was assessed based on Cochrane Collaboration Handbook for RCTs and Newcastle–Ottawa scale for non RCTs. All cases were re-grouped according to the types of surgery, and the primary and secondary outcomes between EDR and LDR were compared in the total cohort termed pancreatectomy (including both PD and DP), PD, and DP respectively. Any discrepancies among researchers were resolved by consensus and arbitration by a panel of investigators (YSM, YMY, and XDT).
Statistical analysis
The quantitative synthesis of variables was conducted using Review Manager (Version 5.3, Cochrane Collaboration, Oxford, United Kingdom). For dichotomous outcomes (e.g. Grade B/C POPF, total complications, postoperative hemorrhage, intra-abdominal infection, DGE, re-admission, re-operation, biliary fistula, intra-abdominal fluid collections, wound infection, pulmonary complications, and intervention), we calculated the pooled risk difference, risk ratio (RR) with 95% confidence intervals. For continuous outcomes (e.g. postoperative in-hospital stay, age, and BMI), we calculated the pooled mean difference, weighted mean difference (WMD) with corresponding 95% confidence intervals. The heterogeneity of included studies was evaluated using Cochran's Q and I2 test. If I2 > 50%, random effects model was used, on the contrary fixed effects model was used if I2 ≤ 50%. The publication bias was evaluated by funnel plot. Statistical significance was defined as P < 0.05.
Results
Characteristics of included studies
Systematic literature search was conducted as shown in Fig. 1. a total of 10 studies were included for meta-analysis based on inclusion and exclusion criteria, including 3 RCTs and 7 non RCTs.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
General characteristics of included studies were summarized in Table 1. According to the Newcastle–Ottawa Scale (NOS) score, all non RCTs studies were considered to be of high quality (NOS score ≥ 6). In total, 1780 patients with EDR and 5613 patients with LDR were enrolled. The selection criteria of enrolled patients were summarized in Table 1. In all RCTs and 2 non RCTs, only patients with low concentration of DFA were enrolled (DFA ≤5000 U/L on POD 1 in 1 RCT
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Characteristics of enrolled patients: quantitative synthesis
Characteristics of enrolled patients for each study were retrieved by three researchers (Table 2). There were no significant differences in age, BMI, diabetes, intraoperative RBC transfusion, operation time, pancreatic texture, small main pancreatic duct between EDR and LDR groups through meta-analysis (Supplementary Fig. S1A-B and E-P). For meta-analysis of all the 10 studies, EDR group had slightly more female patients than LDR group (54.31% vs. 53.29; RR = 1.08; P = 0.01), whereas for meta-analysis of 3 RCTs, no significant difference in the gender ratio was observed (Supplementary Fig. S1C-D). In general, these two groups were relatively homogeneous for demographic, surgical, and pathological characteristics.
Table 2Characteristics of enrolled patients for each study in meta-analysis
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
compared Grade B/C POPF between EDR and LDR groups. For pancreatectomy, including both PD and DP, the meta-analysis of all studies indicated that EDR group had lower Grade B/C POPF rate than LDR group (RR = 0.23, 95% CI: 0.15–0.37; P < 0.00001; I2 = 50%) (Fig. 2A). However, the meta-analysis of 3 RCTs showed that there was no statistical difference in Grade B/C POPF rate between two groups (RR = 0.47, 95% CI: 0.06–3.46; P = 0.46; I2 = 70%). The same results were observed when only including PD (all studies: RR = 0.33, 95% CI: 0.18–0.60; P = 0.0003; I2 = 15%; RCTs: RR = 0.69, 95% CI: 0.27–1.78; P = 0.44; I2 = 0%) or DP (all studies: RR = 0.17, 95% CI: 0.13–0.24; P < 0.00001; I2 = 12%; RCTs: RR = 3.00, 95% CI: 0.13–69.52; P = 0.49) (Fig. 2b and c). There was no evidence of publication bias (Fig. 2d–f). In order to obtain robust results, we further conducted meta-analysis for studies
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
that only recruited patients with low concentration of postoperative DFA. The same trends were observed (pancreatectomy: RR = 0.31, 95% CI: 0.10–0.95; P = 0.04; PD: RR = 0.37, 95% CI: 0.18–0.77; P = 0.0003), indicating that EDR was associated with a reduced rate of POPF in selected low risk patients. (Supplementary Fig. S2A-D).
Table 3Primary and secondary outcomes of included studies in meta-analysis
Figure 2Quantitative synthesis of the primary outcomes (a–c) Quantitative synthesis of Grade B/C POPF for pancreatectomy (a), PD (b), and DP (c). Subgroup analyses of RCTs and non RCTs were conducted (d–f) Funnel plots of Grade B/C POPF (g–i) Quantitative synthesis of total complications for pancreatectomy (g), PD (h), and DP (i) (j–l) Funnel plots of total complications
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
compared total complications between EDR and LDR groups. For pancreatectomy, including PD and DP, the meta-analysis of all studies indicated EDR group had lower total complications rate compared with LDR group (RR = 0.63, 95% CI: 0.49–0.80; P = 0.0001; I2 = 83%) (Fig. 2g). However, the meta-analysis of RCTs showed that there was no statistical difference in total complications rate between two groups (RR = 0.61, 95% CI: 0.37–1.03; P = 0.06; I2 = 82%) (Fig. 2g). The same results were obtained when only PD was included (all studies: RR = 0.74, 95% CI: 0.61–0.90; P = 0.003; I2 = 56%; and RCTs: RR = 0.66, 95% CI: 0.31–1.37; P = 0.26; I2 = 85%) (Fig. 2h). For DP, the meta-analysis of 3 studies including 1 RCT
indicated that EDR group had lower total complications rate (RR = 0.44, 95% CI: 0.37–0.53; P < 0.00001; I2 = 0%) (Fig. 2i). No obvious publication bias was found (Fig. 2j-l). The same results were observed when only patients with low concentration of DFA were enrolled (pancreatectomy: RR = 0.63, 95% CI: 0.49–0.82; P = 0.0006; PD: RR = 0.69, 95% CI: 0.53–0.90; P = 0.007) (Supplementary Fig. S2E-H).
In summary, our meta-analyses indicated that EDR was associated with a significantly lower risk of Grade B/C POPF and total complications for both PD and DP. These results were further verified when studies that only recruited patients with low risk of POPF were included for meta-analysis.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
compared postoperative hemorrhage between EDR and LDR groups. The meta-analysis of both all studies and RCTs showed that there was no significant difference in postoperative hemorrhage for pancreatectomy (all studies: RR = 0.54, 95% CI: 0.29–1.02; P = 0.06; I2 = 0%; RCTs: RR = 0.78, 95% CI: 0.29–2.06; P = 0.45; I2 = 0%) and PD (all studies: RR = 0.82, 95% CI: 0.36–1.85; P = 0.63; I2 = 0%; RCTs: RR = 0.87, 95% CI: 0.31–2.44; P = 0.27; I2 = 17%) (Fig. 3a–d). There is currently not sufficient data to perform meta-analysis of the influence of EDR on postoperative hemorrhage risk for DP.
Figure 3Quantitative synthesis of the secondary outcomes (a–b) Quantitative synthesis of postoperative hemorrhage for pancreatectomy (a), PD (b). Subgroup analyses of RCTs and non RCTs were conducted (c–d) Funnel plots of postoperative hemorrhage (e) Quantitative synthesis of intra-abdominal infection for PD (f) Funnel plot of intra-abdominal infection (g–i) Quantitative synthesis of delayed gastric emptying for pancreatectomy (g), PD (h), and DP (i) (j–l) Funnel plots of delayed gastric emptying
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
reported intra-abdominal infection rate of patients undergoing PD, and the meta-analysis showed that EDR was associated with a significant decrease of intra-abdominal infection rate (PD: RR = 0.36, 95% CI: 0.21–0.62; P = 0.0002; I2 = 15%) (Fig. 3e and f).
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
compared re-operation rate between EDR and LDR groups. The meta-analysis of all these studies showed EDR had lower re-operation rate than LDR for pancreatectomy (RR = 0.50, 95% CI: 0.34–0.72; P = 0.0002; I2 = 17%) and DP (RR = 0.35, 95% CI: 0.13–0.94; P = 0.04; I2 = 74%), but not for PD (RR = 0.76, 95% CI: 0.35–1.61; P = 0.47; I2 = 0%). When only RCTs were meta-analyzed, the result showed comparable re-operation rate for pancreatectomy, PD (Fig. 4a–f). Meta-analysis of studies including only patients with low concentration of postoperative DFA showed that no significant difference was found for pancreatectomy and PD.
Figure 4Quantitative synthesis of the other secondary outcomes (a–b) Quantitative synthesis of re-operation for pancreatectomy (a), PD (b), DP (c). Subgroup analyses of RCTs and non RCTs were conducted (d–f) Funnel plots of re-operation (g–h) Quantitative synthesis of postoperative in-hospital stay for pancreatectomy (g), DP (h). Subgroup analyses of RCTs and non RCTs were conducted (i–j) Funnel plots of postoperative in-hospital stay
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
compared postoperative in-hospital stay between EDR and LDR groups. Meta-analyses for both all the studies and RCTs showed that EDR was associated with a decrease of postoperative in-hospital stay for pancreatectomy (all studies: WMD = −2.62, 95% CI: −3.74 to −1.50; P < 0.00001; RCTs: WMD = −1.68, 95% CI: −2.72 to −0.65; P = 0.001) and DP (all studies: WMD = −2.77, 95% CI: −4.38 to −1.16; P = 0.0008). While currently no sufficient data was available to perform meta-analysis on the influence of EDR on the postoperative in-hospital stay for PD (Fig. 4g–j).
Biliary fistula and wound infection rates were compared in 2 RCTs
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
The meta-analysis of all studies indicated that EDR had lower re-admission rate than LDR for pancreatectomy (RR = 0.62, 95% CI: 0.50–0.78; P < 0.001), PD (RR = 0.71, 95% CI: 0.52–0.97; P = 0.035) and DP (RR = 0.52, 95% CI: 0.42–0.64; P < 0.00001). However, the meta-analysis of RCTs showed there was no statistical difference in re-admission rate between the 2 groups (pancreatectomy: RR = 0.77, 95% CI: 0.16–3.71; P = 0.12; PD: RR = 1.17, 95% CI: 0.40–3.41; P = 0.17). Moreover, re-admission rates of EDR and LDR were comparable for pancreatectomy and PD when only patients with low concentration of postoperative DFA were enrolled (pancreatectomy: RR = 0.85, 95% CI: 0.54–1.34; P = 0.49; PD: RR = 0.93, 95% CI: 0.61–1.42; P = 0.74) (Supplementary Table S1).
There was no significant difference in intervention rate between the 2 groups (pancreatectomy: RR = 0.75, 95% CI: 0.54–1.05; P = 0.09; PD: RR = 0.86, 95% CI: 0.39–1.92; P = 0.72). However, EDR was associated with a decrease of total medical expenses in 3 RCTs
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Taken together, our meta-analysis showed that most of these secondary outcomes were comparable between the EDR and LDR groups for patients after pancreatic surgery. EDR was associated with a lower incidence of intra-abdominal infection after PD. While for DP, EDR group had decreased rates of DGE, re-operation and shorter postoperative in-hospital stay. Currently, no consensus could be reached considering the re-admission rate.
Discussion
In this study, we conducted the first meta-analysis to evaluate the efficacy and safety of EDR for patients undergoing pancreatic resections. The stratified analyses of PD and DP were performed. The meta-analyses of all the available studies and those enrolled only low risk patients showed EDR was associated with significantly lower incidences of grade B/C POPF and total complications for both PD and DP. Moreover, EDR was associated with lower rate of intra-abdominal infection for PD. While it reduced the rates of DGE and re-operation, and shorten the postoperative in-hospital stay for DP. There were no significant differences in postoperative hemorrhage, biliary fistula, wound infection, intra-abdominal fluid collections, pulmonary complications, and intervention rate between EDR and LDR groups. Limited evidence showed that EDR might be associated with a decrease of total medical expenses for patients after pancreatic surgery.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
prospectively assigned the patients following PD into EDR and LDR groups and compared postoperative complications. The results demonstrated that the rates of POPF, intra-abdominal infection and fluid collections were significantly lower in EDR group. In 2010, Bassi et al.
conducted the first RCT to compare EDR and LDR for patients undergoing pancreatectomy, and reported that EDR decreased the incidences of POPF, in-hospital stay, abdominal, and pulmonary complications. Nevertheless, this single-center RCT had limited sample size, and the influences of EDR on patients undergoing PD and DP were not investigated respectively. Thereafter, multiple non RCTs
supported the advantages of EDR after pancreatic surgery, while the conclusions were inconsistent. In order to further evaluate the efficacy of EDR for PD and DP, Dai et al.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
conducted single and multiple-center RCTs in recent years and found no significant differences in POPF and other complications between EDR and LDR groups for both PD and DP. Hence, whether EDR could decrease the complication rate of patients undergoing PD and DP remains to be investigated.
In regard to Grade B/C POPF, the current meta-analysis of all the 10 available studies indicated that EDR was superior to LDR for both PD and DP, whereas the meta-analysis of 3 RCTs showed no significant differences. This might be attributed to the inherent drawback of selection bias for non RCTs, especially considering that 5 of the 7 non RCT didn't propose clear selection criteria for patients' enrollment. In these studies, patients who had high risk of POPF were inevitably inclined to retain the drains longer, thus it was easier to conclude that EDR was superior than LDR. To avoid this selection bias, further analysis for studies that only recruited patients with low concentration of postoperative DFA was conducted, in case that meta-analysis of RCTs showed inconsistent results with that of all available studies. It should be noted that the selection criteria of participants were also inconsistent among the 3 RCTs. The selection strategy of DFA on POD 1 ≤ 5000 U/L) was used in 1 RCTs
and a lower Grade B/C POPF rate was observed compared with all studies (6.95% vs. 16.04%). While in the single and multiple-center RCT performed by Dai, the stricter selection criteria (DFA on POD 1 and 3 ≤ 5000 U/L and drain output within POD 3 ≤ 300 ml per day) was used, and the Grade B/C POPF rates were even lower as 1.39% and 5.13%, respectively. Both of these 2 RCTs performed by Dai and colleagues
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
showed no differences of POPF and total complications between EDR and LDR, which contributed to the equivalence conclusion of meta-analysis of RCTs. However, our results from 5 studies that enrolled only low risk patients showed advantages of EDR in terms of grade B/C POPF and total complications. The possible reason is that in Dai's studies, too strict inclusion criteria led to a similarly lower rate of POPF in the control groups (6.41% and 0%), hence EDR failed to further reduce this risk.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
In addition to POPF, the major concerns of EDR are intra-abdominal fluid collection and consequent intra-abdominal infection, hemorrhage and DGE. The meta-analysis showed that EDR was associated with a lower incidence of intra-abdominal infection after PD. While for DP, EDR group had decreased rates of DGE and re-operation, and shorter postoperative in-hospital stay. There was only a multicenter RCT
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
compared postoperative in-hospital stay after PD, which indicated that EDR reduced postoperative in-hospital stay compared with LDR (15 d vs. 16.7 d, P = 0.010), thus no sufficient data was available to perform meta-analysis. Additionally, the meta-analysis of re-admission rate in all the available studies (preferred EDR in both PD and DP) was different from that of the 3 RCTs. While the results from studies that only enrolled patients with low concentration of postoperative DFA also showed no difference between the 2 groups, which was in consist with meta-analysis of RCTs. Taken together, more well-designed studies are mandatory to further evaluate the effectiveness of EDR following PD or DP.
There are several limitations in this study. First, funnel plots showed that there exists publication bias for some outcomes in meta-analysis. Second, studies included in the current meta-analysis had large time, area and race span, and the definitions of low risk patients and primary and secondary outcomes varied among the studies, contributing to the heterogeneity of meta-analysis. Finally, there are only 3 RCTs so far, thus more RCTs to evaluate the efficacy and safety of EDR are needed.
Conclusion
Our meta-analysis demonstrates that EDR is effective and safe for both PD and DP considering POPF and total complications, especially for patients with low concentration of postoperative DFA. More well-designed RCTs are mandatory to further evaluate the influence of EDR following pancreatectomy.
Funding
This study was supported by The Natural Science Foundation of China (NO.82171722, 81871954), Beijing Municipal Natural Science Foundation (NO.7212111), and Tianjin Key Medical Discipline (Specialty) Construction Project.
Authors' contributions
Conceptualization, KC, XDT; Literature Search, KC, ZHL, BHY; Data Collection, KC, BHY, YSM; Formal Analysis, KC; Validation, KC, BHY, SPZ, ZJS, and XDT; Investigation and Visualization, KC; Methodology, KC and XDT; Writing – original draft, KC; Project administration, YMY; Writing – review & editing XDT; Supervision, XDT and YMY. All authors read and approved the final version of the manuscript.
Consent for publication
All authors agree to publish this article.
Availability of data and material
All data are available upon request.
Conflict of interest
None declare.
Acknowledgements
We thank all staffs for the effort of conducting EDR studies, medical statistics room of the Peking University First Hospital.
Appendix A. Supplementary data
The following are the Supplementary data to this article.
Survey on the current status of the diagnosis and treatment of pancreatic cancer in public tertiary hospitals in China: a cross-sectional questionnaire-based, observational study.
The value of drains as a fistula mitigation strategy for pancreatoduodenectomy: something for everyone? Results of a randomized prospective multi-institutional study.
Early drain removal is safe in patients with low or intermediate risk of pancreatic fistula after pancreaticoduodenectomy: a multicenter, randomized controlled trial.
Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients.
00044-8&id=gr1.jpg){kind=link}
00044-8&id=gr2.jpg){kind=link}
00044-8&id=gr3.jpg){kind=link}
00044-8&id=gr4.jpg){kind=link}