Systematic Review and Meta-analysis of Restrictive Perioperative Fluid Management in Pancreaticoduodenectomy. World J Surg. 2018;42:2938-2950. [PMID: 29464346 DOI: 10.1007/s00268-018-4545-6]

Pilot Study to Optimize Goal-directed Hemodynamic Management During Pancreatectomy

INTRODUCTION

Intraoperative goal-directed hemodynamic therapy (GDHT) is a cornerstone of enhanced recovery protocols. We hypothesized that use of an advanced noninvasive intraoperative hemodynamic monitoring system to guide GDHT may decrease intraoperative hypotension (IOH) and improve perfusion during pancreatic resection.

METHODS

The monitor uses machine learning to produce the Hypotension Prediction Index to predict hypotensive episodes. A clinical decision-making algorithm uses the Hypotension Prediction Index and hemodynamic data to guide intraoperative fluid versus pressor management. Pre-implementation (PRE), patients were placed on the monitor and managed per usual. Post-implementation (POST), anesthesia teams were educated on the algorithm and asked to use the GDHT guidelines. Hemodynamic data points were collected every 20 s (8942 PRE and 26,638 POST measurements). We compared IOH (mean arterial pressure <65 mmHg), cardiac index >2, and stroke volume variation <12 between the two groups.

RESULTS

10 patients were in the PRE and 24 in the POST groups. In the POST group, there were fewer minimally invasive resections (4.2% versus 30.0%, P = 0.07), more pancreaticoduodenectomies (75.0% versus 20.0%, P < 0.01), and longer operative times (329.0 + 108.2 min versus 225.1 + 92.8 min, P = 0.01). After implementation, hemodynamic parameters improved. There was a 33.3% reduction in IOH (5.2% ± 0.1% versus 7.8% ± 0.3%, P < 0.01, a 31.6% increase in cardiac index >2.0 (83.7% + 0.2% versus 63.6% + 0.5%, P < 0.01), and a 37.6% increase in stroke volume variation <12 (73.2% + 0.3% versus 53.2% + 0.5%, P < 0.01).

CONCLUSIONS

Advanced intraoperative hemodynamic monitoring to predict IOH combined with a clinical decision-making tree for GDHT may improve intraoperative hemodynamic parameters during pancreatectomy. This warrants further investigation in larger studies.

Intraoperative fluid management is not predictive of AKI in major pancreatic surgery: a retrospective cohort study

BACKGROUND

Pancreatic surgery is associated with a significant risk for acute kidney injury (AKI) and clinically relevant postoperative pancreatic fistula (CR-POPF). This investigation evaluated the impact of intraoperative volume administration, vasopressor therapy, and blood pressure management on the primary outcome of AKI and the secondary outcome of a CR-POPF after pancreatic surgery.

METHODS

This retrospective single-center cohort investigated 200 consecutive pancreatic surgeries (January 2018-December 2021). Patients were categorized for the presence/absence of AKI (Kidney Disease Improving Global Outcomes) and CR-POPF. After univariate analysis, multivariable models were constructed to control for the univariate cofactor differences in the primary and secondary outcomes.

RESULTS

AKI was identified in 20 patients (10%) with significant univariate differences in demographics (body mass index and gender), comorbidities, indices of chronic renal insufficiency, and an increased AKI Risk score. Surgical characteristics, intraoperative fluid, vasopressor, and blood pressure management were similar in patients with and without AKI. Patients with AKI had increased blood loss, lower urine output, and packed red blood cell administration. After multivariate analysis, male gender (OR = 7.9, 95% C.I. 1.8-35.1) and the AKI Risk score (OR = 6.3, 95% C.I. 2.4-16.4) were associated with the development of AKI (p < 0.001). Intraoperative and postoperative volume, vasopressor administration, and intraoperative hypotension had no significant impact in the multivariate analysis. CR-POPF occurred in 23 patients (11.9%) with no significant contributing factors in the multivariate analysis. Patients who developed AKI or a CR-POPF had an increase in surgical complications, length of stay, discharge to a skilled nursing facility, and mortality.

CONCLUSION

In this analysis, intraoperative volume administration, vasopressor therapy, and a blood pressure < 55 mmHg for more than 10 min were not associated with an increased risk of AKI. After multivariate analysis, male gender and an elevated AKI Risk score were associated with an increased likelihood of AKI.

The impact of intra- and postoperative fluid balance in pancreatic surgery - A retrospective cohort study

BACKGROUND/OBJECTIVES

The aim of this study was to evaluate the impact of perioperative fluid administration in pancreatic surgery.

METHODS

Patients who underwent pancreatic resections were identified from our institution's prospectively maintained database. Fluid balances were recorded intraoperatively and at 24hr postoperatively. Patients were stratified into tertiles of fluid administration (low, medium, high). Adjusted multivariable analysis was performed and outcome measures were postoperative complications.

RESULTS

A total of 211 patients were included from 2012 to 2017. Complication rates were POPF(B/C) 19.4%, DGE(B/C) 14.7%, PPH(C) 10.0% and CDC ≥ IIIb 26.1%. In multivariable analysis, high perioperative fluid balance was an independent risk factor associated with POPF (OR = 10.5, 95%CI 2.7-40.7, p = .001), CDC (OR = 2.5, 95%CI 1.2-5.3, p < .002), DGE (OR = 2.3, 95%CI 1.0-5.2, p = .017), PPH (OR = 6.7 95%CI 2.2-20.0, p = .038) and reoperation (OR = 3.1, 95%CI 1.6-6.2, p = .006). In multivariable analysis with intraoperative and postoperative fluid balances as separate predictors, intraoperative (OR = 2,5, 95%CI 1.2-5.5, p = .04) and postoperative fluid balance (OR = 2.5, 95%CI 1.2-5.5, p = .02) were predictors of POPF. Postoperative fluid balance was the only predictor for mortality (OR = 4.5, 95%CI 1.0-18.9, p = .041) and predictor for CDC (OR = 2.0, 95%CI 1.0-4.0, p = .043) and OHS days (OR = 6.9, 95%CI 0.03-13.7, p = .038).

CONCLUSIONS

High postoperative fluid balance in particular is associated with postoperative morbidity. Maintaining a fluid-restrictive strategy postoperatively should be recommended for patients undergoing pancreatic surgery.

Goal-directed fluid management associates with fewer postoperative fluid collections in pancreatoduodenectomy patients

BACKGROUND

The association between perioperative fluid management and complications in pancreatoduodenectomy patients remains controversial. We explored the association between fluid management and radiological signs of complications.

METHODS

We examined pancreatoduodenectomy patients operated between July 2014 and December 2015 (n = 125) and between January 2017 and June 2018 (n = 124). The first cohort received intraoperative fluid management according to a goal-directed strategy and the second cohort was treated conventionally. We analyzed fluid administration, edema visible in computed tomography (CT) scans seven days postoperatively, and radiological signs of complications occurring up to 30 days. We performed multivariable logistic regression analyses to identify risk factors for fluid collections.

RESULTS

No statistically significant difference in postoperative edema via CT scans emerged between the fluid management groups. However, the intraperitoneal space expanded in patients with severe Clavien-Dindo complications compared with patients experiencing mild or no complications (19.1% (IQR 10.4-40.5) vs 2.5% (IQR -7.9-16.6), p = 0.004). Fluid collections were less frequent in the goal-directed group than in the conventional fluid management group (16.8% vs 34.7%, p = 0.001). Risk factors for fluid collections included main pancreatic duct size ≤3 mm, less intraoperative fluid volume accompanying conventional fluid management, a lower postoperative urine output, and postoperative congestive heart failure. The goal-directed group received more intraoperative fluids than the conventional fluid management group and postoperative urine output was higher in the goal-directed group on postoperative days 1-3.

CONCLUSIONS

Optimization of intraoperative fluid management through target-controlled strategies and early diuresis were associated with a lower frequency of fluid collections in postoperative CT.

Effects of intraoperative fluid balance during pancreatoduodenectomy on postoperative pancreatic fistula: an observational cohort study

BACKGROUND

Perioperative fluid management during major abdominal surgery has been controversial. Postoperative pancreatic fistula (POPF) is a critical complication of pancreaticoduodenectomy (PD). We conducted a retrospective cohort study to analyze the impact of intraoperative fluid balance on the development of POPF.

METHODS

This retrospective cohort study enrolled 567 patients who underwent open pancreaticoduodenectomy, and the demographic, laboratory, and medical data were recorded. All patients were categorized into four groups according to quartiles of intraoperative fluid balance. Multivariate logistic regression and restricted cubic splines (RCSs) were used to analyze the relationship between intraoperative fluid balance and POPF.

RESULTS

The intraoperative fluid balance of all patients ranged from -8.47 to 13.56 mL/kg/h. A total of 108 patients reported POPF, and the incidence was 19.0%. After adjusting for potential confounders and using restricted cubic splines, the dose‒response relationship between intraoperative fluid balance and POPF was found to be statistically insignificant. The incidences of bile leakage, postpancreatectomy hemorrhage, and delayed gastric emptying were 4.4%, 20.8%, and 14.8%, respectively. Intraoperative fluid balance was not associated with these abdominal complications. BMI ≥ 25 kg/m², preoperative blood glucose < 6 mmol/L, long surgery time, and lesions not located in the pancreas were independent risk factors for POPF.

CONCLUSION

The study did not find a significant association between intraoperative fluid balance and POPF. Well-designed multicenter studies are necessary to explore the association between intraoperative fluid balance and POPF.

Optimal Perioperative Fluid Therapy Associates with Fewer Complications After Pancreaticoduodenectomy

BACKGROUND

Optimal fluid management in pancreaticoduodenectomy patients remains contested. We aimed to examine the association between perioperative fluid administration and postoperative complications.

METHODS

We studied 168 pancreaticoduodenectomy patients operated in 2015 (n = 93) or 2017 (n = 75) at Helsinki University Hospital. In 2015, patients received intraoperative fluids following a goal-directed approach and, in 2017, according to anesthesiologist's clinical practice (conventional fluid management). We analyzed the differences in perioperative fluid administration between the groups, specifically examining the occurrence of severe complications (Clavien-Dindo ≥ III), pancreatic fistulas, cardiovascular complications, and the length of hospital stay.

RESULTS

The goal-directed group received more intraoperative fluids than the conventional fluid management group (12.0 ml/kg/h vs. 8.3 ml/kg/h, p < 0.001). Urine output (770 ml vs. 575 ml, p = 0.004) and intraoperative fluid balance (9.4 ml/kg/h vs. 6.3 ml/kg/h, p < 0.001) were higher in the goal-directed group than in the conventional fluid management group. Severe surgical complications (19.4% vs. 38.7%, p = 0.009) as well as clinically relevant pancreatic fistulas (1.1% vs. 10.7%, p = 0.011) occurred more frequently in patients receiving conventional fluid management. Moreover, the conventional fluid management group experienced longer hospital stays (9.0 vs. 11.5 days, p = 0.02). Lower intraoperative fluid volume accompanying conventional fluid management was associated with a higher risk of severe postoperative complications compared with higher volume in the goal-directed group (odds ratio 2.58 (95% confidence interval 1.04-6.42), p = 0.041).

CONCLUSIONS

The goal-directed group experienced severe complications less frequently. Our findings indicate that optimizing the intraoperative fluid administration benefits patients, while adopting a too-restrictive approach represents an inferior choice.

Prevention and Treatment of Grade C Postoperative Pancreatic Fistula

Postoperative pancreatic fistula (POPF) is a troublesome complication after pancreatic surgeries, and grade C POPF is the most serious situation among pancreatic fistulas. At present, the incidence of grade C POPF varies from less than 1% to greater than 9%, with an extremely high postoperative mortality rate of 25.7%. The patients with grade C POPF finally undergo surgery with a poor prognosis after various failed conservative treatments. Although various surgical and perioperative attempts have been made to reduce the incidence of grade C POPF, the rates of this costly complication have not been significantly diminished. Hearteningly, several related studies have found that intra-abdominal infection from intestinal flora could promote the development of grade C POPF, which would help physicians to better prevent this complication. In this review, we briefly introduced the definition and relevant risk factors for grade C POPF. Moreover, this review discusses the two main pathways, direct intestinal juice spillover and bacterial translocation, by which intestinal microbes enter the abdominal cavity. Based on the abovementioned theory, we summarize the operation techniques and perioperative management of grade C POPF and discuss novel methods and surgical treatments to reverse this dilemma.

The clinical impact of the systolic volume variation guided intraoperative fluid administration regimen on surgical outcomes after pancreaticoduodenectomy: a retrospective cohort study

BACKGROUND

Pancreaticoduodenectomy is associated with high morbidity. Many preoperative variables are risk factors for postoperative complications, but they are primarily non-modifiable. It is not clear whether an intraoperative goal-directed fluid regimen might be associated with fewer postoperative surgical complications compared to current conservative, non-goal-directed fluid practices. We hypothesize that the use of Systolic Volume Variation (SVV)-guided intraoperative fluid administration might be beneficial.

METHODS

Data from 223 patients who underwent pancreaticoduodenectomy in our institution between 2015 and 2019 were reviewed. Patients were classified into two groups based on the use of intraoperative use of SVV to guide the administration of fluids. The decision to use SVV or not was made by the attending anesthesiologist. Subjects were classified into SVV-guided intraoperative fluid therapy (SVV group) and non-SVV-guided intraoperative fluid therapy (non-SVV group). Uni and multivariate regression analyses were conducted to determine if SVV-guided fluid therapy was significantly associated with a lower incidence of postoperative surgical complications, such as Postoperative Pancreatic Fistula (POPF), Delayed Gastric Emptying (DGE), among others, after adjusting for confounders.

RESULTS

Baseline, demographic, and intraoperative characteristics were similar between SVV and non-SVV groups. In the multivariate analysis, the use of SVV guidance was significantly associated with fewer postoperative surgical complications (OR = 0.48; 95% CI 0.25-0.91; p = 0.025), even after adjusting for significant covariates, such as perioperative use of epidural, pancreatic gland parenchyma texture, and diameter of the pancreatic duct.

CONCLUSIONS

VV-guided intraoperative fluid administration might be associated with fewer postoperative surgical complications after pancreaticoduodenectomy.

Restrictive intraoperative fluid intake in liver surgery and postoperative renal function: A propensity score matched study

BACKGROUND

Postoperative acute kidney injury (AKI) is a common complication in hepatic surgery. In hepatic surgery, relative hypovolemia may help to limit blood loss, but the consequences of restrictive fluid intake are unknown. The goal of this study was to determine the influence of intraoperative fluid intake on the incidence of AKI and its consequences.

METHODS

Data from 397 consecutive patients who underwent liver resection were prospectively recorded and retrospectively analyszed. We compared the incidence of postoperative acute kidney failure in patients given restrictive (≤ 5 mL/kg/h) versus liberal (> 5 mL/kg/h) fluid therapy. We calculated a 1:1 match propensity score using logistic regression to estimate the likelihood of patients receiving restrictive or liberal intraoperative fluid intakes. The association between the intraoperative fluid intake strategy and occurrence of postoperative AKI were tested using a Cox frailty model on the database of matched patients.

RESULTS

Postoperative AKI was diagnosed in 133 of the 397 patients. Fluid intake strategy was restrictive for 121 patients and liberal for 276 patients. After propensity score matching to balance confounding factors, the liberal strategy was associated with a significantly lower risk for postoperative AKI compared to the restrictive strategy (Hazard Ratio 0.40 [0.29; 0.56], P<0.001). Patients with postoperative AKI had longer hospital stays and higher mortality. There were no cases of further blood loss in the liberal fluid intake group.

CONCLUSIONS

A restrictive fluid intake strategy is a risk factor for developing postoperative AKI, with serious consequences, without reducing blood loss in liver surgery.

MEasuring the impact of Anesthetist-administered medications volumeS on intraoperative flUid balance duRing prolonged abdominal surgEry (MEASURE Study)

BACKGROUND

The contribution of intraoperative anesthetist-administered medications (IAAMs) to the total volume of intraoperative intravenous (IV) fluid therapy and their association with postoperative outcomes has never been formally investigated.

METHODS

We performed a retrospective study of adult patients undergoing pancreaticoduodenectomy. The volume of IAAMs, crystalloids and colloids, blood and blood products, blood loss, urine output and intraoperative fluid balance were collected. The contribution of IAAMs to the total intraoperative IV fluid volume and postoperative complications was evaluated.

RESULTS

A total of 152 consecutive patients were included. The median volume of IAAMs was 363.8 mL (interquartile range [IQR], (241.0-492.5) delivered at a median rate of 0.61 mL kg hr^-1 (0.40-0.87) over a median duration of surgery of 489 minutes (416.3-605.3). This increased the total administered fluid volume by 5.2% (95% confidence intervals [CI]: 4.6, 5.9%) (Cohen's d=1.33, P<0.001). The volume of IAAMs was comparable to the intraoperative colloid volume administered (median colloid volume, 400 mL). Overall, fluid volumes correlated significantly with the severity of complications (P=0.011), and the correlation strength increased when the IAAMs volume was included (P=0.005). On addition of IAAMs, the area under the receiver operator characteristic curve for prediction of postoperative complications increased from 0.580 (95%CI: 0.458, 0.701) to 0.603 (95%CI: 0.483, 0.723), P=0.041).

CONCLUSIONS

IAAMs significantly increased the total administered fluid volume during pancreaticoduodenectomy. Their inclusion increases the accuracy of postoperative complications predictions. These findings support their inclusion in fluid volumes and balances in future interventional studies.

Risk-Stratified Pancreatectomy Clinical Pathway Implementation and Delayed Gastric Emptying

BACKGROUND

Delayed gastric emptying (DGE) is a frequent complication after pancreaticoduodenectomy (PD) that impairs recovery and quality of life. The purpose of this study was to assess the impact risk-stratified pancreatectomy clinical pathways (RSPCPs) had on delayed gastric emptying (DGE) and identify factors associated with DGE in a contemporary period.

METHODS

A single-institution, prospective database was queried for consecutive PDs during July 2011-November 2019. Using international definitions, DGE rates were compared between periods before and after RSPCPs were implemented in 2016, classifying patients according to their postoperative pancreatic fistula (POPF) risk. Risk factors were analyzed to identify modifiable targets.

RESULTS

Among 724 elective PDs, 552 (76%) were for adenocarcinoma and 172 (24%) for other diagnoses. Of the 197 (27%) patients with DGE, 119 (16%) had type A, 41 (6%) type B, and 38 (5%) type C. In the overall cohort, DGE rates were higher with pylorus-preserving vs. classic hand-sewn reconstruction (odds ratio [OR] - 1.84; p < 0.001), postoperative abscess (OR - 2.54; p = 0.003), and non-white patients (p = 0.007), but lower after implementation of RSPCPs (OR - 0.34, p < 0.001). In the 374 patients treated with RSPCPs, only 17% (n = 65/374) developed DGE. Patients with protocol-compliant NGT removal ≤ 48 h were less likely to experience DGE (OR - 1.46, p = 0.042).

CONCLUSION

Our data suggest that implementation of preoperatively assigned RSPCPs as a care bundle was the most important factor in decreasing DGE. These gains were accentuated in patients who underwent early nasogastric tube removal and had a classic hand-sewn gastro-jejunostomy reconstruction. Application of these modifiable factors is generalizable with low implementation barriers.

Goal-directed therapy with bolus albumin 5% is not superior to bolus ringer acetate in maintaining systemic and mesenteric oxygen delivery in major upper abdominal surgery: A randomised controlled trial

BACKGROUND

Goal-directed therapy (GDT) is increasingly used in abdominal surgery. Whether crystalloids can exert the same effect as colloid, and how this may affect perfusion, is still unclear. The effect of GDT on the systemic oxygen delivery index (sDO2I) and the mesenteric oxygen delivery index (mDO2I) can be quantified by measuring cardiac index and flow in the superior mesenteric artery, respectively.

OBJECTIVE

The aim of this study was to test the hypothesis that intra-operative GDT with bolus human albumin 5% is superior to GDT with bolus ringer acetate in maintaining sDO2I and mDO2I in elective major upper gastrointestinal cancer surgery.

DESIGN

Randomised controlled double blinded trial.

SETTING

Odense University Hospital, Denmark, from May 2014 to June 2015.

PATIENTS

A total of 89 adults scheduled for elective major upper gastrointestinal cancer surgery were randomised and data from 60 were analysed.

EXCLUSION CRITERIA

contraindications for using the LiDCOplus system, known allergy to albumin, pre-operative renal failure, pancreatic cancer and pre-operative down staging using chemotherapy and/or radiation therapy, pregnancy.

INTERVENTIONS

Patients were randomised to intra-operative GDT with either bolus human albumin or ringer acetate 250 ml, guided by pulse pressure variation and stroke volume.

MAIN OUTCOME MEASURES

Changes in sDO2I and mDO2I. Secondary outcomes were changes in other haemodynamic variables, fluid balance, blood transfusions, fluid-related complications and length of stay (LOS) in ICU and hospital.

RESULTS

Median [IQR] sDO2I was 522 [420 to 665] ml min m in the ringer acetate group and 490 [363 to 676] ml min m in the human albumin group, P = 0.36. Median [IQR] mDO2I was 12.1 [5.8 to 28.7] ml min m in the ringer acetate group and 17.0 [7.6 to 27.5] ml min m in the human albumin group, P = 0.17. Other haemodynamic comparisons did not differ significantly. More trial fluid was administered in the ringer acetate group. We found no significant difference in transfusions, complications or LOS.

CONCLUSION

Bolus human albumin 5% was not superior to bolus ringer acetate in maintaining systemic or mesenteric oxygen delivery in elective major upper gastrointestinal cancer surgery, despite the administration of larger volumes of trial fluid in the ringer acetate group. No significant difference was seen in fluid-related complications or LOS.

TRIAL REGISTRATION

https://eudract.ema.europa.eu/ Identifier: 2013-002217-36.

Working towards an ERAS Protocol for Pancreatic Transplantation: A Narrative Review

Enhanced recovery after surgery (ERAS) initially started in the early 2000s as a series of protocols to improve the perioperative care of surgical patients. They aimed to increase patient satisfaction while reducing postoperative complications and postoperative length of stay. Despite these protocols being widely adopted in many fields of surgery, they are yet to be adopted in pancreatic transplantation: a high-risk surgery with often prolonged length of postoperative stay and high rate of complications. We have analysed the literature in pancreatic and transplantation surgery to identify the necessary preoperative, intra-operative and postoperative components of an ERAS pathway in pancreas transplantation.

Perioperative fluids and complications after pancreatoduodenectomy within an enhanced recovery pathway

Optimized fluid management is a key component of enhanced recovery (ERAS) pathways. Implementation is challenging for pancreatoduodenectomy (PD) and clear guidance is missing in the respective protocol. The aim of this retrospective study was to evaluate the influence of perioperative intravenous (IV) fluid administration on postoperative complications. 164 consecutive patients undergoing PD within ERAS between October 2012 and June 2017 were included. Perioperative IV fluid and morbidity (Clavien classification and comprehensive complication index (CCI)) were assessed. A threshold of more than 4400 ml IV fluid during the first 24 h could be identified to predict occurrence of complications (area under ROC curve 0.71), with a positive and negative predictive value of 93 and 23% respectively. More than 4400 ml intravenous fluids during the first 24 h was an independent predictor of overall postoperative complications (adjusted odds ratio 4.40, 95% CI 1.47–13.19; p value = 0.008). Patients receiving ≥ 4400 ml were associated with increased overall complications (94 vs 77%; p value < 0.001), especially pulmonary complications (31 vs 16%; p value = 0.037), as well as a higher median CCI (33.7 vs 26.2; p value 0.041). This threshold of 4400 ml intravenous fluid might be a useful indicator for the management following pancreatoduodenectomy.

Risk Stratification for the Intensive Care Unit Following Pancreaticoduodenectomy

INTRODUCTION

In view of the limited capacities in intensive care units and the increasing economic burden, identification of risk factors could allow better and more efficient planning. Therefore, the aim of this study was to assess independent risk factors for the duration of intensive care unit stay after pancreatoduodenectomy (PD).

METHODS

147 patients who underwent pancreatoduodenectomy in the time period from 2013 to 2015 were identified from a prospective database and a retrospective analysis was performed. The primary endpoint was length of time spent in the ICU. A retrograde analysis was performed using univariate and multivariate regression analysis. All pre-, intra- and postoperative parameters were considered in the analysis.

RESULTS

The median time spent in the intensive care unit (ICU) is one day. The univariate analysis demonstrated increased pack years, cerebrovascular events, anticoagulation, elevated creatinine and CA 19-9 as preoperative risk factors. In multivariate analysis, antihypertensive medication (AHT; OR 2.46; 95% CI 1.57 - 3.87; p = 0.05), operation time (OR 1.01; 95% CI 1.00 - 1.01; p = 0.03), extended LAD (OR 5.46; 95% CI 2.77 - 10.75; p = 0.01) and severe PPH (OR 4.01; 95% CI 2.07 - 7.76; p = 0.04) are significant risk factors for longer ICU stay.

DISCUSSION

Patients with cardiovascular risk factors and elevated preoperative creatinine level are at greater risk for a prolonged ICU stay. Risk and benefit of an extended LAD should be weighed during the operation. Median duration on ICU/IMC after PD is one day or less for patients without risk factors. Whether routine monitoring in the ICU/IMC after PD is necessary must be clarified in further studies.

Risk prediction platform for pancreatic fistula after pancreatoduodenectomy using artificial intelligence

BACKGROUND

Despite advancements in operative technique and improvements in postoperative managements, postoperative pancreatic fistula (POPF) is a life-threatening complication following pancreatoduodenectomy (PD). There are some reports to predict POPF preoperatively or intraoperatively, but the accuracy of those is questionable. Artificial intelligence (AI) technology is being actively used in the medical field, but few studies have reported applying it to outcomes after PD.

AIM

To develop a risk prediction platform for POPF using an AI model.

METHODS

Medical records were reviewed from 1769 patients at Samsung Medical Center who underwent PD from 2007 to 2016. A total of 38 variables were inserted into AI-driven algorithms. The algorithms tested to make the risk prediction platform were random forest (RF) and a neural network (NN) with or without recursive feature elimination (RFE). The median imputation method was used for missing values. The area under the curve (AUC) was calculated to examine the discriminative power of algorithm for POPF prediction.

RESULTS

The number of POPFs was 221 (12.5%) according to the International Study Group of Pancreatic Fistula definition 2016. After median imputation, AUCs using 38 variables were 0.68 ± 0.02 with RF and 0.71 ± 0.02 with NN. The maximal AUC using NN with RFE was 0.74. Sixteen risk factors for POPF were identified by AI algorithm: Pancreatic duct diameter, body mass index, preoperative serum albumin, lipase level, amount of intraoperative fluid infusion, age, platelet count, extrapancreatic location of tumor, combined venous resection, co-existing pancreatitis, neoadjuvant radiotherapy, American Society of Anesthesiologists’ score, sex, soft texture of the pancreas, underlying heart disease, and preoperative endoscopic biliary decompression. We developed a web-based POPF prediction platform, and this application is freely available at http://popfrisk.smchbp.org.

CONCLUSION

This study is the first to predict POPF with multiple risk factors using AI. This platform is reliable (AUC 0.74), so it could be used to select patients who need especially intense therapy and to preoperatively establish an effective treatment strategy.

Pre-, peri- and post-operative factors for the development of pancreatic fistula after pancreatic surgery

BACKGROUND

The most hazardous complication to pancreatic surgery is the development of a post-operative pancreatic fistula (POPF). Appropriate understanding of the underlying pathophysiology, risk factors and perioperative mechanisms may allow for better management and use of preventive measures.

METHODS

Systematic literature search using the English PubMed literature up to April 2019, with emphasis on the past 5 years.

RESULTS

Several risk scores have been developed but none are perfect in predicting POPF risk. A conceptual framework of factors that contribute to the pathophysiology of pancreatic fistulae is still developing but incomplete. Recognized factors include those related to the patient, the pathology and the perioperative care. Interventions such as use of drains, stents and various drugs to mediate risk is still debated. Emerging data suggest that both the microbiome and the inflammation in the post-operative phase may play important roles in risk for POPF. Available risk scores allow for stratification of risk and mitigation strategies tailored to reduce this. However, accurate estimation of risk remains a challenge and mechanisms are only partially understood.

CONCLUSIONS

The pathophysiology of POPF remains poorly understood. Current models only partially explain risks or associated mechanisms. Novel areas of investigation need to be explored for better prediction.

Early postoperative pancreatitis following pancreaticoduodenectomy: what is clinically relevant postoperative pancreatitis?

BACKGROUND/OBJECTIVES

Postoperative pancreatitis (POP) has recently been shown to be the cause of pancreatic fistula (POPF) following pancreaticoduodenectomy (PD). The aim of the present study was to document the perioperative outcome associated with POP and determine potential risk factors for POP.

METHODS

Patients undergoing PD between 2009 and 2015 were identified from the prospective data base at a single center. The previous suggested definition of POP by Connor was used. Complications were graded according to the Clavien-Dindo classification and by the grading proposed for POP. Risk factors for POP were analyzed by univariate and multivariate analysis.

RESULTS

Of 190 patients, a total of 100 patients (53%) developed POP of whom 22 (12%) and 13 (7%) had grade B and grade C complications, respectively. Elevated serum CRP-levels on postoperative day (POD) 2 and elevated serum lipase on POD 1 were associated with onset of cr-POP.

CONCLUSION

The proposed definition of POP constitutes a valuable tool to assess a serious pancreatic-surgery associated complication. Routine serum CRP and serum lipase levels on the first two postoperative days enable sufficient discrimination of clinically relevant POP.

Intraoperative Fluid Administration and Surgical Outcomes Following Pancreaticoduodenectomy: External Validation at a Tertiary Referral Center

BACKGROUND

While intraoperative fluid overload is associated with higher complication rates following surgery, data for pancreaticoduodenectomy are scarce and heterogeneous. We evaluated multiple prior definitions of restrictive and liberal fluid regimens and analyzed whether these affected surgical outcomes at our tertiary referral center.

METHODS

Studies evaluating different intraoperative fluid regimens on outcomes after pancreatic resections were retrieved. After application of all prior definitions of restrictive and liberal fluid regimens to our patient cohort, relative risks of each outcome were calculated using all reported infusion regimens.

RESULTS

Five hundred and seven pancreaticoduodenectomies were included. Nine different fluid regimens were evaluated. Two regimens utilized absolute volume cutoffs, and the remaining evaluated various infusion rates, ranging from 5 to 15 mL/kg/h. Total volume administration of >5000 mL and >6000 mL was associated with increased complications (RR 1.25 and RR 1.17, respectively) and >6000 mL with increased sepsis (RR 2.14). Conversely, a rate of <5 mL/kg/h was associated with increased risk of postoperative pancreatic fistula (POPF, RR 3.16) and sepsis (RR 3.20), <6.8 mL/kg/h with increased major morbidity (RR 1.64) and sepsis (RR 2.27), and <8.2 mL/kg/h with increased POPF (RR 2.16). No effects were observed on pulmonary complications, surgical site infections, length of stay, or mortality.

CONCLUSIONS

In an uncontrolled setting with no standard intraoperative or postoperative care map, the volume of intraoperative fluid administration appears to have limited impact on early postoperative outcomes following pancreaticoduodenectomy, with adverse outcomes only seen at extreme values.

Clinical guidelines for perioperative hemodynamic management of non cardiac surgical adult patients

Perioperative hemodynamic management, through monitoring and intervention on physiological parameters to improve cardiac output and oxygen delivery (goal-directed therapy, GDT), may improve outcome. However, an Italian survey has revealed that hemodynamic protocols are applied by only 29.1% of anesthesiologists. Aim of this paper is to provide clinical guidelines for a rationale use of perioperative hemodynamic management in non cardiac surgical adult patients, oriented for Italy and updated with most recent studies. Guidelines were elaborated according to NICE (National Institute for Health and Care Excellence) and GRADE system (Grading of Recommendations of Assessment Development and Evaluations). Key questions were formulated according to PICO system (Population, Intervention, Comparators, Outcome). Guidelines and systematic reviews were identified on main research databases and strategy was updated to June 2018. There is not enough good quality evidence to support the adoption of a GDT protocol in order to reduce mortality, although it may be useful in high risk patients. Perioperative GDT protocol to guide fluid therapy is recommended to reduce morbidity. Continuous monitoring of arterial pressure may help to identify short periods of hemodynamic instability and hypotension. Fluid strategy should aim to a near zero balance in normovolemic patients at the beginning of surgery, and a slight positive fluid balance may be allowed to protect renal function. Drugs such as inotropes, vasocostrictors, and vasodilatator should be used only when fluids alone are not sufficient to optimize hemodynamics. Perioperative GDT protocols are associated with a reduction in costs, although no economic study has been performed in Italy.