Factors associated with prolonged hospitalization in patients undergoing pancreatoduodenectomy

Severity of Pancreatic Leak in Relation to Gut Restoration After Pancreaticoduodenectomy: The Role of the Roux-en-Y Configuration

Background

Pancreatic leak after pancreaticoduodenectomy and gut restoration via a single jejunal loop remains the crucial predictor of patients' outcome. Our reasoning that active pancreatic enzymes may be more disruptive to the pancreatojejunostomy prompted us to explore a Roux-en-Y configuration for the gut restoration, anticipating diversion of bile salts away from the pancreatic stump. Our study aims at comparing two techniques regarding the severity of postoperative pancreatic fistula (POPF) and patients' outcome.

Methods

The files of 415 pancreaticoduodenectomy patients were retrospectively reviewed. Based on gut restoration, the patients were divided into: cohort A (n = 105), with gut restoration via a single jejunal loop, cohort B (n = 140) via a Roux-en-Y technique assigning the draining of pancreatic stump to the short limb and gastrojejunostomy and bile (hepaticojejunostomy) flow to long limb, and cohort C (n = 170) granting the short limb to the gastric and pancreatic anastomosis, whereas hepaticojejunostomy was performed to the long limp. The POPF-related morbidity and mortality were analyzed.

Results

Overall POPF in cohort A versus cohorts B and C was 19% versus 12.1% and 9.4%, respectively (P = 0.01 A vs B + C). POPF-related morbidity in cohort A versus cohorts B and C was 10.5% versus 7.3% and 6.3%, respectively (P = 0.03 A vs B+C). POPF-related total hospital mortality in cohorts A versus B and C was 1.9% versus 0.8% and 0.59%, respectively (P = 0.02 A vs B+C).

Conclusion

Roux-en-Y configuration showed lower incidence and severity of POPF. Irrespective of technical skill, creating a gastrojejunostomy close to pancreatojejunostomy renders the pancreatic enzymes less active by leaping the bile salts away from the pancreatic duct and providing a lower pH.

Generation and validation of a formula to calculate hemoglobin loss on a cohort of healthy adults subjected to controlled blood loss

Background

The ability to approximate intra-operative hemoglobin loss with reasonable precision and linearity is prerequisite for determination of a relevant surgical outcome parameter: This information enables comparison of surgical procedures between different techniques, surgeons or hospitals, and supports anticipation of transfusion needs. Different formulas have been proposed, but none of them were validated for accuracy, precision and linearity against a cohort with precisely measured hemoglobin loss and, possibly for that reason, neither has established itself as gold standard. We sought to identify the minimal dataset needed to generate reasonably precise and accurate hemoglobin loss prediction tools and to derive and validate an estimation formula.

Methods

Routinely available clinical and laboratory data from a cohort of 401 healthy individuals with controlled hemoglobin loss between 29 and 233 g were extracted from medical charts. Supervised learning algorithms were applied to identify a minimal data set and to generate and validate a formula for calculation of hemoglobin loss.

Results

Of the classical supervised learning algorithms applied, the linear and Ridge regression models performed at least as well as the more complex models. Most straightforward to analyze and check for robustness, we proceeded with linear regression. Weight, height, sex and hemoglobin concentration before and on the morning after the intervention were sufficient to generate a formula for estimation of hemoglobin loss. The resulting model yields an outstanding R² of 53.2% with similar precision throughout the entire range of volumes or donor sizes, thereby meaningfully outperforming previously proposed medical models.

Conclusions

The resulting formula will allow objective benchmarking of surgical blood loss, enabling informed decision making as to the need for pre-operative type-and-cross only vs. reservation of packed red cell units, depending on a patient’s anemia tolerance, and thus contributing to resource management.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-021-02783-9.

Impact of Complications After Pancreatoduodenectomy on Mortality, Organ Failure, Hospital Stay, and Readmission: Analysis of a Nationwide Audit

OBJECTIVE

To quantify the impact of individual complications on mortality, organ failure, hospital stay, and readmission after pancreatoduodenectomy.

SUMMARY OF BACKGROUND DATA

An initial complication may provoke a sequence of adverse events potentially leading to mortality after pancreatoduodenectomy. This study was conducted to aid prioritization of quality improvement initiatives.

METHODS

Data from consecutive patients undergoing pancreatoduodenectomy (2014-2017) were extracted from the Dutch Pancreatic Cancer Audit. Population attributable fractions (PAF) were calculated for the association of each complication (ie, postoperative pancreatic fistula, postpancreatectomy hemorrhage, bile leakage, delayed gastric emptying, wound infection, and pneumonia) with each unfavorable outcome [ie, in-hospital mortality, organ failure, prolonged hospital stay (>75th percentile), and unplanned readmission), whereas adjusting for confounders and other complications. The PAF represents the proportion of an outcome that could be prevented if a complication would be eliminated completely.

RESULTS

Overall, 2620 patients were analyzed. In-hospital mortality occurred in 95 patients (3.6%), organ failure in 198 patients (7.6%), and readmission in 427 patients (16.2%). Postoperative pancreatic fistula and postpancreatectomy hemorrhage had the greatest independent impact on mortality [PAF 25.7% (95% CI 13.4-37.9) and 32.8% (21.9-43.8), respectively] and organ failure [PAF 21.8% (95% CI 12.9-30.6) and 22.1% (15.0-29.1), respectively]. Delayed gastric emptying had the greatest independent impact on prolonged hospital stay [PAF 27.6% (95% CI 23.5-31.8)]. The impact of individual complications on unplanned readmission was smaller than 11%.

CONCLUSION

Interventions focusing on postoperative pancreatic fistula and postpancreatectomy hemorrhage may have the greatest impact on in-hospital mortality and organ failure. To prevent prolonged hospital stay, initiatives should in addition focus on delayed gastric emptying.

Monitoring gastric myoelectric activity after pancreaticoduodenectomy for diet "readiness"

Postoperative delayed gastric emptying (DGE) is a frustrating complication of pancreaticoduodenectomy (PD). We studied whether monitoring of postoperative gastric motor activity using a novel wireless patch system can identify patients at risk for DGE. Patients ( n = 81) were prospectively studied since 2016; 75 patients total were analyzed for this study. After PD, battery-operated wireless patches (G-Tech Medical) that acquire gastrointestinal myoelectrical signals are placed on the abdomen and transmit data by Bluetooth. Patients were divided into early and late groups by diet tolerance of 7 days [enhanced recovery after surgery (ERAS) goal]. Subgroup analysis was done of patients included after ERAS initiation. The early and late groups had 50 and 25 patients, respectively, with a length of stay (LOS) of 7 and 11 days ( P < 0.05). Nasogastric insertion was required in 44% of the late group. Tolerance of food was noted by 6 versus 9 days in the early versus late group ( P < 0.05) with higher cumulative gastric myoelectrical activity. Diminished gastric myoelectrical activity accurately identified delayed tolerance to regular diet in a logistical regression analysis [area under the curve (AUC): 0.81; 95% confidence interval (CI), 0.74-0.92]. The gastric myoelectrical activity also identified a delayed LOS status with an AUC of 0.75 (95% CI, 0.67-0.88). This stomach signal continued to be predictive in 90% of the ERAS cohort, despite earlier oral intake. Measurement of gastric activity after PD can distinguish patients with shorter or longer times to diet. This noninvasive technology provides data to identify patients at risk for DGE and may guide the timing of oral intake by gastric "readiness." NEW & NOTEWORTHY Limited clinical indicators exist after pancreaticoduodenectomy to allow prediction of delayed gastric emptying (DGE). This study introduces a novel, noninvasive, wireless patch system capable of accurately monitoring gastric myoelectric activity after surgery. This system can differentiate patients with longer or shorter times to a regular diet as well as provide objective data to identify patients at risk for DGE. This technology has the potential to individualize feeding regimens based on gastric activity patterns to improve outcomes.

Independent Predictors of Increased Operative Time and Hospital Length of Stay Are Consistent Across Different Surgical Approaches to Pancreatoduodenectomy

BACKGROUND

While minimally invasive approaches are increasingly being utilized for pancreatoduodenectomy (PD), factors associated with prolonged operative time (OpTime) and hospital length of stay (LOS) remain poorly defined, and it is unclear whether these factors are consistent across surgical approaches.

METHODS

The ACS-NSQIP targeted pancreatectomy database from 2014 to 2016 was used to identify all patients who underwent open (OPD), laparoscopic (LPD), or robotic (RPD) pancreatoduodenectomy. Multivariable linear regression analyses were used to evaluate predictors of OpTime and LOS, as well as quantify the changes observed relative to each surgical approach.

RESULTS

Among 10,970 patients, PD procedure types varied: 9963 (92%) open, 418 (4%) laparoscopic, and 409 (4%) robotic. LOS was longer for the open and laparoscopic approaches (11 vs. 11 vs. 10 days, P = 0.0068), whereas OpTime was shortest for OPD (366 vs. 426 vs. 435 min, P < 0.0001). Independent predictors of a prolonged OpTime were ASA class ≥ 3 (P = 0.0002), preoperative XRT (P < 0.0001), pancreatic duct < 3 mm (P = 0.0001), T stage ≥ 3 (P = 0.0108), and vascular resection (P < 0.0001) for OPD; T stage ≥ 3 (P = 0.0510) and vascular resection (P = 0.0062) for LPD; and malignancy (P = 0.0460) and conversion to laparotomy (P = 0.0001) for RPD. Independent predictors of increased LOS were age ≥ 65 years (P = 0.0002), ASA class ≥ 3 (P = 0.0012), hypoalbuminemia (P < 0.0001), and preoperative blood transfusion (P < 0.0001) for OPD as well as an OpTime > 370 min (all p < 0.05) and specific postoperative complications (all p < 0.05) for all surgical approaches.

CONCLUSIONS

Perioperative risk factors for prolonged OpTime and hospital LOS are relatively consistent across open, laparoscopic, and robotic approaches to PD. Particular attention to these factors may help identify opportunities to improve perioperative quality, enhance patient satisfaction, and ensure an efficient allocation of hospital resources.