In this multicenter database analysis of patients undergoing PD, we retrospectively analyzed what factors were evaluated using Ideal Outcome (IO, defined as the absence of in-hospital mortality, severe complications as classified by Clavien-Dindo [grade ≥ 3], postoperative pancreatic fistula grade B or C, necessity for reoperation, extended hospital stay, and readmission within 30 days of discharge). We found that younger age, absence of invasive preoperative procedures such as PTBD, receipt of neoadjuvant chemotherapy and certain histological diagnoses correlated with a greater chance of achieving IO among these patients.
Age has already been established as an important factor influencing better outcomes and quality of life (QOL) following PD [14]. Although the short-term outcomes of PD in elderly patients are similar to those in younger patients, with no significant differences in overall morbidity, mortality, or pancreatic fistula rates, long-term outcomes show that younger patients have better disease-free and overall survival rates than elderly patients [14, 15]. Our analysis was in concordance with previous research, with age > 70 years showing lesser odds of achieving IO in both the general cohort and the PDAC-specific group. These findings reinforce the potential use of IO to stratify patients and predict how they might fare postoperatively.
Percutaneous transhepatic biliary drainage (PTBD) before PD is a feasible and effective procedure for treating biliopancreatic leakage and post-ERCP cholangitis. It has been shown to have high technical and clinical success rates in resolving leakage and reducing complications [16–18]. However, research has also reported increased morbidity and mortality in some cases with the use of PTBD. In a comparison of palliative patients who underwent PTBD, Rees et al. reported a higher mortality rate in their case group [19]. PTBD can increase the risk of infections, including cholangitis, induce local inflammation and fibrosis, and complicate subsequent surgical procedures by causing changes in the biliary tract [20, 21]. Thus, its application is often reserved for specific scenarios where ERCP has failed or is not technically feasible, highlighting a selection bias towards patients with higher risk profiles. Turan et al. pointed out that PTBD procedures are usually performed as a rescue procedure after unsuccessful ERCP. They further suggested that subsequent complications could also be attributed to the previously executed ERCP, which may have played a part in the notable 30-day morbidity and mortality rates [22]. It is also our practice to perform PTBD only after failure of ERCP. Those patients are usually at higher risk of cholangitis and infections.
Some studies suggest that neoadjuvant chemotherapy may improve outcomes after PD for PDAC, including potentially enhancing overall survival rates [23, 24]. A retrospective analysis of the American College of Surgeons NSQIP database over 5 years demonstrated that neoadjuvant chemotherapy was associated with a decrease in serious morbidity, clinically relevant pancreatic fistulas, organ space infections, need for percutaneous drainage, reoperations, and prolonged hospital stays [25]. Another review by Hyman et al. concluded that the use of neoadjuvant therapy before performing PD for PDAC did not adversely affect post-surgical outcomes [25]. Significant associations between the utility of neoadjuvant chemoradiation and IO can be beneficial in guiding future research and clinical optimization when encountering high-risk patients with PD. Patients undergoing neoadjuvant chemotherapy often require concurrent preoperative biliary interventions, such as stenting, which complicates the assessment of their IO in PD. This overlap necessitates a deeper investigation into how these combined treatments influence IO, considering factors like disease severity and individual responses to chemotherapy. Given these associations, the utility of neoadjuvant chemoradiation in improving IO rates is noteworthy. However, it is critical to emphasize that neoadjuvant therapy should be guided by oncological considerations, particularly in patients with borderline or locally advanced pancreatic cancer, where its use has been associated with significant survival benefits compared to upfront surgery [26]. Additionally, the choice of neoadjuvant therapy can influence the rate of achieving negative resection margins, reduce nodal involvement, and ultimately improve overall survival [27]. In light of these findings, integrating neoadjuvant therapy into the treatment protocols for high-risk patients should be considered not only for its potential to enhance surgical outcomes but also to ensure that these patients are more likely to receive adjuvant therapy, thereby improving long-term survival outcomes [28].
A study comparing outcomes between open, laparoscopic, and robotic pancreaticoduodenectomies using NSQIP data found that laparoscopic PD (LPD) was independently associated with decreased 30-day morbidity compared with open PD (OPD), with no difference in 30-day mortality [29]. Similarly, a network meta-analysis indicated that while the operation time for OPD was shorter, robotic PD was associated with significantly less blood loss than both LPD and OPD, and both minimally invasive approaches (robotic and laparoscopic) were linked to reduced rates of delayed gastric emptying compared to OPD [30]. Although our analysis did not reveal a significant association between the type of operation (open, laparoscopic, robotic) and IO, this is consistent with recent high-level evidence demonstrating the non-inferiority of minimally invasive pancreatoduodenectomy (MIPD) compared to OPD [31, 32]. Several randomized controlled trials and meta-analyses have shown that MIPD offers similar oncologic and perioperative outcomes as OPD, including comparable survival rates and R0 resection margins, which may explain the lack of significant differences in IO observed in our study [33, 34]. Additionally, surgeons performing MIPD may select patients with less complex cases or better overall health, which could lead to outcomes comparable to those of the traditional open approach. This selection bias, coupled with the comprehensive nature of IO as a measure that includes various aspects of surgical success, might dilute the specific impact of any surgical approach. Our findings align with the broader literature, supporting the view that MIPD can be considered a viable alternative to OPD without compromising patient outcomes [35].
Our study has some limitations. While neoadjuvant chemotherapy was a significant factor in predicting IO for the entire cohort, it was not significant when considering only PD patients with PDAC. This discrepancy highlights that many patients included in our study (e.g., those with cysts, pancreatitis) are not candidates for chemotherapy, while those who might benefit the most (PDAC patients) did not show a significant association. Moreover, our findings are based on data from a single center, which limits the generalizability of our results. The small number of cases in certain categories (e.g., laparoscopic and robotic procedures, neuroendocrine tumors) further complicates the ability to draw definitive conclusions from the multivariate analysis. Future research should include prospective studies to validate the findings of our retrospective analysis. Expanding the scope of investigation to include a broader range of patient cohorts and integrating frailty indices could offer deeper insights into patient selection and preoperative optimization. Further studies should also explore the impact of long-term oncologic outcomes and quality of life measures on IO, providing a more holistic view of patient recovery and success post-PD. Additionally, refining the IO criteria to incorporate patient-centric outcomes and integrating these measures into clinical decision-making tools could standardize care and improve overall outcomes for PD patients.
In conclusion, our comprehensive retrospective analysis across multiple Northwell Health Centers highlights the potential validity of the Ideal Outcome (IO) as a prognostic measure in assessing the quality-of-care post-pancreatoduodenectomy (PD). Integrating IO into clinical decision-making tools and practice guidelines could also be explored, aiming to standardize care and improve overall outcomes for PD patients. While our findings align with the existing literature regarding the impact of these factors on PD outcomes, they also provide new insights, particularly regarding the potential role of IO as a robust composite measure.