The safety of the early removal of prophylactic drainage after liver resection based solely on predetermined criteria: a propensity score analysis

Early drain removal after hepatectomy based on bile leakage prediction using drainage fluid volume and direct bilirubin level

AIMS

This study aimed to determine the value of the drainage fluid volume and direct bilirubin level for predicting significant bile leakage (BL) after hepatectomy and establish novel criteria for early drain removal.

METHODS

Data from 351 patients who underwent hepatic resection at Gunma University in Japan between October 2018 and March 2022 were retrospectively analyzed. Clinical characteristics and surgical outcomes of patients with and without significant BL were compared. Criteria for early drain removal were determined and verified.

RESULTS

Bile leakage occurred in 27 (7.1%) patients; 8 (2.3%) had grade A leakage and 19 (5.4%) had grade B leakage. The optimal cut-off value for the drainage fluid direct bilirubin level on postoperative day (POD) 2 was 0.16 mg/dL, which had the highest area under the curve and negative predictive value (NPV). Patients with BL had significantly larger drainage volumes on POD 2. The best cut-off value was 125 mL because it had the greatest NPV. Patients in both the primary and validation (n = 90) cohorts with bilirubin levels less than 0.16 mg/dL and drainage volumes less than 125 mL did not experience leakage.

CONCLUSIONS

A drainage fluid volume less than 125 mL and direct bilirubin level less than 0.16 mg/dL on POD 2 are criteria for safe early drain removal after hepatectomy.

Early drain removal after live liver donor hepatectomy is safe - a randomized controlled pilot study

INTRODUCTION

The current study aimed to assess the safety of early drain removal after live donor hepatectomy (LDH).

METHODS

One hundred eight consecutive donors who met the inclusion criteria were randomized to early drain removal (EDR - postoperative day (POD) 3 - if serous and the drain bilirubin level was less than 3 mg/dl - "3 × 3" rule) and routine drain removal (RDR - drain output serous and less than 100 ml). The primary outcome was to compare the safety. The secondary outcome was to compare the postoperative morbidity.

RESULTS

Preoperative, intraoperative, and postoperative parameters except for the timing of drain removal were comparable. EDR was feasible in 46 out of 54 donors (85.14%) and none required re-intervention after EDR. There was significantly better pain relief with EDR (p = 0.00). Overall complications, pulmonary complications, and hospital stay were comparable on intention-to-treat analysis. However, pulmonary complications (EDR - 1.9% vs RDR - 16.3% P = 0.030), overall complications (18.8% vs 36.3%, P = 0.043), and hospital stay (8 vs 9, P = 0.014) were more in the RDR group on per treatment analysis. Bile leaks were seen in three donors (3.7% in the EDR group vs 1.9% in RDR, P = 0.558), and none of them required endo-biliary interventions. Re-exploration for intestinal obstruction was required for 3 donors in RDR (0% vs 5.7%; p = 0.079).

CONCLUSION

EDR by the "3 × 3" rule after LDH is safe and associated with better pain relief. On per treatment analysis, EDR was associated with significantly less hospital stay and lower pulmonary and overall complications.

CLINICAL TRIAL REGISTRY

Clinical Trials.gov - NCT04504487.

Advantages of skin closure with subcuticular suture for liver resection on postoperative and cosmetic outcomes: a propensity matched analysis

PURPOSE

The effects of subcuticular sutures on postoperative and cosmetic outcomes in patients who underwent liver resection have not been well studied. Here, we investigated the advantages of subcuticular suture compared to skin stapler regarding open liver resection.

METHODS

We assessed 342 patients who underwent liver resection at Nara Medical University between 2008 and 2015. They were divided into two groups: subcuticular suture and staple groups. Baseline characteristics and perioperative outcomes were retrospectively compared using one-to-one propensity score matching analysis.

RESULTS

In this period, 179 patients underwent skin closure with subcuticular sutures and 163 patients underwent skin closure with staples. After propensity matching, 85 pairs of cases were matched. The incidence of wound infection was similar in the two groups (3.5% in the subcuticular suture group and 9.4% in the staple group; p = 0.119). The length of hospital stay was significantly shorter in the subcuticular suture group than in the staple group (10 days vs 15 days; p < 0.001). In addition, the rate of patients who were discharged within 7 days after surgery was statistically higher in the subcuticular group (21.1% vs 3.5%, p = 0.001). Hypertrophic scar 6 months after surgery was significantly less frequent in the subcuticular group (9.4% vs 25.9%, p = 0.010).

CONCLUSION

Subcuticular sutures might be advantageous for liver surgery reducing length of hospital stay and proportion of hypertrophic scar.

Update on perioperative management of patients undergoing surgery for liver cancer

Hepatocellular carcinoma is often accompanied by chronic hepatitis or cirrhosis. Preoperative evaluation of liver function and postoperative nutritional management are critical in patients with hepatocellular carcinoma who undergo liver surgery. Although the incidence of postoperative complications and death has declined in Japan over the last 10 years, postoperative complications have not been fully overcome. Therefore, surgical procedures and perioperative management must be improved. Accurate preoperative evaluations of liver function, nutrition, inflammation, and body skeletal muscle are required. Determination of the optimal surgical procedure should consider not only tumor characteristics but also the physical reserve of the patient. Nutritional management of chronic liver disorders, especially maintaining protein synthesis for postoperative protein/energy, is important. Prophylactic antibiotics are recommended for short‐term use within 24 hours after surgery. Abdominal drainage is recommended for patients with cirrhosis who may develop large amounts of ascites, who are at risk of postoperative bleeding, or who may have bile leakage due to a large resection area. Postoperative exercise therapy may improve insulin resistance in patients with chronic liver damage. Implementation of an early/enhanced recovery after surgery program is recommended to reduce biological invasive responses and achieve early independence of physical activity and nutrition intake. We review the latest information on the perioperative management of patients undergoing liver resection for hepatocellular carcinoma.

Variation in Drain Management Among Patients Undergoing Major Hepatectomy

BACKGROUND

Although previous studies have suggested that drain management is highly variable, data on drain placement and timing of drain removal among patients undergoing hepatic resection remain scarce. The objective of the current study was to define the utilization of drain placement among patients undergoing major hepatic resection.

METHODS

The ACS NSQIP-targeted hepatectomy database was used to identify patients who underwent major hepatectomy between 2014 and 2017. Association between day of drain removal, timing of discharge, and drain fluid bilirubin on postoperative day (POD) 3 (DFB-3) was assessed. Propensity score matching (PSM) was used to compare outcomes of patients with a drain removed before and after POD 3.

RESULTS

Among 5330 patients, most patients had an abdominal drain placed at the time of hepatic resection (n = 3075, 57.7%). Of 2495 patients with data on timing of drain removal, only 380 patients (15.2%) had their drain removed by POD 3. Almost 1 in 6 patients (n = 441, 17.7%) were discharged home with the drain in place. DFB-3 values correlated poorly with POD of drain removal (R² = 0.0049). After PSM, early drain removal (≤ POD 3) was associated with lower rates of grade B or C bile leakage (2.1% vs. 7.1%, p = 0.008) and prolonged length of hospital stay (6.0% vs. 12.7%, p = 0.009) compared with delayed drain removal (> POD 3).

CONCLUSIONS

Roughly 3 in 5 patients had a drain placed at the time of major hepatectomy and only 1 in 7 patients had the drain removed early. This study demonstrated the potential benefits of early drain removal in an effort to improve the quality of care following major hepatectomy.

Drain Placement After Uncomplicated Hepatic Resection Increases Severe Postoperative Complication Rate: A Japanese Multi-institutional Randomized Controlled Trial (ND-trial)

OBJECTIVE

To assess the clinical impact of a no-drain policy after hepatic resection.

SUMMARY OF BACKGROUND DATA

Previous randomized controlled trials addressing no-drain policy after hepatic resection seem inconclusive because they did not adopt appropriate study design to validate its true clinical impact.

METHODS

This unblinded, randomized controlled trial was done at 7 Japanese institutions. Patients undergoing hepatic resection without biliary reconstruction were randomized to either D group or ND group. When the risk of postoperative bile leakage or hemorrhage were considered high, the patients were excluded during the operation. Primary endpoint was the postoperative complication of C-D grade 3 or higher within 90 postoperative days. A noninferiority of ND group to D group was assessed, and if it was confirmed, a superiority was assessed.

RESULTS

Between May 2015 and July 2017, a total of 400 patients were finally included in the per-protocol set analysis: 199 patients in D group and 201 patients in ND group. Intraoperatively, 37 patients were excluded from the final enrollment because of high risk of bile leakage or hemorrhage. Postoperative complication rate of C-D grade 3 or higher was 8.0% (16/199) in the D group and 2.5% (5/201) in the ND group. The risk difference was -5.5% (95% confidence interval: -9.9% to -1.2%) and fulfilled the prescribed noninferiority margin of 4%. No postoperative mortality was experienced in both groups. Bile leakage was diagnosed in 8.0% (16/199) of the D group and none in the ND group (P < 0.001). In none of the subgroups classified based on 8 potentially relevant factors, drain placement was favored in terms of C-D grade 3 or higher complication.

CONCLUSIONS

Drains should not be placed after uncomplicated hepatic resections.

Bacterial flora in the bile: Clinical implications and sensitivity pattern from a tertiary care centre

BACKGROUND

Though preoperative biliary drainage (PBD) has been suggested to be linked with increased perioperative morbidity it is still practiced commonly. We studied the association of PBD and positive biliary culture with surgical site infection and also analysed the common pathogens and their antibiotic sensitivity spectrum.

METHODS

Prospectively maintained data of patients who underwent various pancreatobiliary surgeries from 2017 to 2019 was analysed. Patients whose intraoperative bile culture reports were available were included in the study. Various factors associated with surgical site infection (SSI), microbial spectrum of bile culture and their sensitivity pattern were analysed.

RESULTS

Out of 68 patients whose bile culture report were available, PBD was done in 65% (n = 44). Among patients with infected bile (n = 51), biliary stent was present in 78.4% (n = 40). On univariate analysis, the factors associated with SSI were low albumin level (<3.5 mg%), long operative time (>6 h), duration of abdominal drain (>4 days), length of hospital stay, intraoperative bile spillage and infected bile. However, on multivariate analysis, only presence of drain for >4 days (p = 0.04) and positive bile culture (p = 0.02) was linked with increased risk of SSI. Most common organism isolated was E coli (73.2%), with 100% sensitivity to Colistin and Tigecycline shown by gram negative isolates.

CONCLUSION

Preoperative biliary stenting alone did not increase the risk of SSI, but the positive bile culture correlated with SSI irrespective of PBD. Most biliary pathogens were resistant to commonly used antibiotics and intraoperative bile culture will aid in providing appropriate antibiotic coverage.

Early drain removal after hepatectomy: an underutilized management strategy

BACKGROUND

Recent data suggest that routine drainage is unnecessary in patients undergoing hepatectomy, but many surgeons continue to utilize drains. We compared the outcomes of patients undergoing early versus routine drain removal after hepatectomy.

METHODS

Patients having drains placed during major (≥3 segments) or partial hepatectomy (≤2 segments) were identified in the 2014-16 ACS-NSQIP database. Propensity matching between early (POD 0-3) and routine (POD 4-7) drain removal and multivariable regressions were performed.

RESULTS

Early drain removal was performed in 661 (40%) of patients undergoing a partial hepatectomy and 211 (22%) of major hepatectomy patients. After matching, 719 early and 719 routine drain removal patients were compared. Early drain removal patients had lower overall (12 vs 19%, p < 0.001) and serious (9 vs 13%, p < 0.03) morbidity as well as fewer bile leaks (2.1% vs 5.0%, p < 0.003). Length of stay was two days shorter (4 vs 6 days, p < 0.01) and readmissions were less frequent (5.4 vs 8.1%, p = 0.02) for patients undergoing early drain removal.

CONCLUSION

Early drain removal is associated with fewer overall and serious complications, shorter length of stay and fewer readmissions. Early drain removal after hepatectomy is an underutilized management strategy.

Use of primary surgical drains in synchronous resection for colorectal liver metastases: a NSQIP analysis of current practice paradigm

BACKGROUND

There are no studies examining the use of subhepatic drains after simultaneous resection of synchronous colorectal liver metastases (sCRLM). This study aimed to (1) describe the current practices regarding primary drain placement, (2) evaluate drain efficacy in mitigating postoperative complications, and (3) determine impact of drain maintenance duration on patient outcomes.

METHODS

The ACS-NSQIP targeted data from 2014 to 2017 were analyzed. Propensity score of surgical drain versus no drain cohorts was performed. Main study outcomes were mortality, major morbidity, organ/space surgical site infection (SSI), secondary drain/aspiration procedure, and any septic events. Additional univariate/multivariate logistic analyses were performed to identify associations with drain placement and duration. Major hepatectomy was defined as formal right hepatectomy and any trisectionectomy.

RESULTS

584 combined liver and colorectal resection (CRR) cases were identified. Open partial hepatectomy with colectomy was the most common procedure (70%, n = 407). Nearly 40% of patients received surgical drains (n = 226). Major hepatectomy, lower serum albumin, and no intraoperative portal vein occlusion (Pringle maneuver) were significantly associated with drain placement (p < 0.05). In the matched cohort (n = 190 in each arm), patients with surgical drains experienced higher rates of major morbidity (30% vs 12%), organ/space SSI (16% vs 6%), postoperative drain/aspiration procedures (9% vs 3%), and sepsis/septic shock (12% vs 4%) (all p < 0.05). Patients with severely prolonged drain removal, defined as after postoperative day 13 (POD13), had higher risk of postoperative morbidity compared to those with earlier drain removal (p < 0.01). 30-day mortality rate was not significantly different between the two groups.

CONCLUSION

Primary surgical drains were placed in a substantial percentage of patients undergoing combined resection for sCRLM. This case-matched analysis suggested that surgical drains are associated with an increase in postoperative morbidity. Postoperative drain maintenance past 13 days is associated with worse outcomes compared to earlier removal.

Timing for removing prophylactic drains after liver resection: an evaluation of drain removal on the third and first postoperative days

Background

Conventionally, drains are removed from postoperative day (POD) 7 to POD 14 at our institute after hepatectomy (control group). This study was conducted to evaluate the outcomes of drain removal in the early postoperative period.

Methods

Recently, we defined criteria for the early removal of drains: (I) a drain-fluid bilirubin level of below 3 mg/dL; (II) a drain discharge volume of less than 500 mL/day; and (III) no macroscopic signs of bleeding or infection. For patients meeting these criteria, drains were removed on POD 3 between January 2012 and February 2013 (POD 3 group) and on POD 1 between February and December 2013 (POD 1 group). The outcomes of these groups were then retrospectively compared.

Results

The median duration of the postoperative hospital stay was shorter in the POD 3 group (11 days) than in the control group (14 days) (P<0.0001). The incidence of drain infection was lower in the POD 3 group (1.2%) than in the control group (5.7%). Meanwhile, the incidences of bile leakage and complications were higher in the POD 1 group than in the POD 3 group. However, the incidences were almost the same when patients whose drains were actually removed on the predefined POD were compared. The intraoperative findings were also considered when removing the drains.

Conclusions

Drain removal on POD 3 may reduce the length of the postoperative hospital stay and the incidence of drain infection without impairing safety. To remove drains safely on POD 1, however, the intraoperative findings should also be considered.

Significance of bacterial culturing of prophylactic drainage fluid in the early postoperative period after liver resection for predicting the development of surgical site infections

PURPOSES

The relationship between the results of bacterial drainage fluid cultures in the early postoperative period after liver resection and the development of surgical site infections (SSIs) is unclear. We evaluated the diagnostic value of bacterial cultures of drainage fluid obtained on postoperative day (POD) 1 after liver resection.

METHODS

The cases of all consecutive patients who underwent elective liver resection from January 2014 to December 2016 were analyzed. The association between a positive culture result and the development of SSIs was analyzed.

RESULTS

A total of 195 consecutive patients were studied. Positive drainage fluid cultures were obtained in 6 patients (3.1%). A multivariate analysis revealed that a positive drainage fluid culture was an independent risk factor for SSIs (odds ratio: 8.04, P = 0.035), and combined resection of the gastrointestinal tract was a risk factor for a positive drainage fluid culture (P = 0.006). Among the patients who did not undergo procedures involving the gastrointestinal tract, there was no association between drainage fluid culture positivity and SSIs.

CONCLUSIONS

The detection of positive culture results for drainage fluid collected on POD 1 after liver resection was associated with SSIs. However, among patients who did not undergo procedures involving the gastrointestinal tract, it was not a predictor of SSIs.

Does anatomic resection improve the postoperative outcomes of solitary hepatocellular carcinomas located on the liver surface?

BACKGROUND

It is unclear whether anatomic resection achieves better outcomes than nonanatomic resection in patients with hepatocellular carcinoma. This study aimed to compare the outcomes of anatomic resection and nonanatomic resection for hepatocellular carcinoma located on the liver surface via one-to-one propensity score-matching analysis.

METHODS

Data from all consecutive patients who underwent liver resection for primary solitary hepatocellular carcinoma at Nara Medical University Hospital, Japan, January 2007- December 2015 were retrieved. Superficial hepatocellular carcinomas were defined as hepatocellular carcinoma that extended to a depth of < 3 cm from the liver surface and measured < 5 cm in diameter. The prognoses of the patients with superficial hepatocellular carcinoma who underwent anatomic resection and nonanatomic resection were compared.

RESULTS

In this study 23 patients with superficial hepatocellular carcinoma underwent anatomic resection and 70 patients who underwent nonanatomic resection. The recurrence-free survival rate of the patients who underwent anatomic resection was better than that of the patients who underwent nonanatomic resection (P = .006), while no such difference was observed for nonsuperficial hepatocellular carcinoma. After the propensity score-matching procedure, the resected liver volume and operation time were the only background or clinical characteristics to exhibit significant differences between the anatomic resection (n = 20) and nonanatomic resection groups (n = 20). The recurrence-free survivial rate of the patients who underwent anatomic resection was significantly than that of the patients that underwent nonanatomic resections (P = .030), but overall survival did not differ significantly between the groups (P = .182).

CONCLUSION

Anatomic resection decreases the risk of tumor recurrence and improves recurrence-free survival compared with nonanatomic resection in patients with superficial hepatocellular carcinoma.

Risk factors and a simple model for predicting bile leakage after radical hepatectomy in patients with hepatic alveolar echinococcosis

Postoperative bile leakage (BL) is a major complication of hepatic alveolar echinococcosis (HAE). The purpose of this study was to identify the risk factors for BL and to establish a simple scoring system for predicting BL.A total of 152 patients with HAE were included in the study between May 2004 and December 2016. The patient's baseline data, laboratory blood tests, imaging features, and surgical management were collected. Univariate and multivariate analyses were used to screen for factors to predict BL. The cutoff values for those factors and predictive value of a model were determined by receiver operative characteristic curve (ROC) analysis.BL was detected in 22 of the 152 patients. Univariate analyses showed significant differences in the lesion diameter, levels of lactate dehydrogenase (LDH), alkaline phosphatase, aspartate aminotransferase, alanine aminotransferase and direct bilirubin (DBIL), inferior vena cava invasion, surface area of hepatectomy, blood loss and history of percutaneous transhepatic cholangial drainage between patients with and without BL. On multivariate analyses, DBIL > 7.1 μmol/L, LDH > 194 U/L, lesion diameter > 12 cm and a larger surface area of hepatectomy were independent predictors of BL. The resulting area under the ROC of the scoring model was 0.724 (95% CI, 0.646-0.793).The lesion diameter, DBIL, larger surface area of hepatectomy, and elevated LDH were the important factors affecting the occurrence of BL after surgery. The risk score model will help the clinician to assess BL before surgery. More studies are needed to confirm the scoring model and risk factors.