Prediction of Osteoporotic Hip Fracture Outcome: Comparative Accuracy of 27 Immune-Inflammatory-Metabolic Markers and Related Conceptual Issues.

Objectives: This study, based on the concept of immuno-inflammatory-metabolic (IIM) dysregulation, investigated and compared the prognostic impact of 27 indices at admission for prediction of postoperative myocardial injury (PMI) and/or hospital death in hip fracture (HF) patients. Methods: In consecutive HF patient (n = 1273, mean age 82.9 ± 8.7 years, 73.5% females) demographics, medical history, laboratory parameters, and outcomes were recorded prospectively. Multiple logistic regression and receiver-operating characteristic analyses (the area under the curve, AUC) were used to establish the predictive role for each biomarker. Results: Among 27 IIM biomarkers, 10 indices were significantly associated with development of PMI and 16 were indicative of a fatal outcome; in the subset of patients aged >80 years with ischaemic heart disease (IHD, the highest risk group: 90.2% of all deaths), the corresponding figures were 26 and 20. In the latter group, the five strongest preoperative predictors for PMI were anaemia (AUC 0.7879), monocyte/eosinophil ratio > 13.0 (AUC 0.7814), neutrophil/lymphocyte ratio > 7.5 (AUC 0.7784), eosinophil count < 1.1 × 109/L (AUC 0.7780), and neutrophil/albumin × 10 > 2.4 (AUC 0.7732); additionally, sensitivity was 83.1-75.4% and specificity was 82.1-75.0%. The highest predictors of in-hospital death were platelet/lymphocyte ratio > 280.0 (AUC 0.8390), lymphocyte/monocyte ratio < 1.1 (AUC 0.8375), albumin < 33 g/L (AUC 0.7889), red cell distribution width > 14.5% (AUC 0.7739), and anaemia (AUC 0.7604), sensitivity 88.2% and above, and specificity 85.1-79.3%. Internal validation confirmed the predictive value of the models. Conclusions: Comparison of 27 IIM indices in HF patients identified several simple, widely available, and inexpensive parameters highly predictive for PMI and/or in-hospital death. The applicability of IIM biomarkers to diagnose and predict risks for chronic diseases, including OP/OF, in the preclinical stages is discussed.

A novel approach to community CPR and AED outreach focused on underserved learner communities.

Creating a sustainable community cardiopulmonary resuscitation (CPR) and automated external defibrillator (AED) program that reaches underserved communities poses a challenge for the emergency medical services (EMS) community. Attendance, funding, and resources have all been linked to struggles surrounding community CPR/AED programs. Through our experience in conducting CPR/AED trainings in underserved regions of eastern North Carolina, we propose a method of effectively utilizing existing organizations and institutions of learning to expand and maintain a sustainable community CPR/AED program. Furthermore, we demonstrate 10 cornerstones in developing relationships within the community to increase attendance and participation in diverse communities.

Shrinkage-based Random Local Clocks with Scalable Inference.

Molecular clock models undergird modern methods of divergence-time estimation. Local clock models propose that the rate of molecular evolution is constant within phylogenetic subtrees. Current local clock inference procedures exhibit one or more weaknesses, namely they achieve limited scalability to trees with large numbers of taxa, impose model misspecification, or require a priori knowledge of the existence and location of clocks. To overcome these challenges, we present an autocorrelated, Bayesian model of heritable clock rate evolution that leverages heavy-tailed priors with mean zero to shrink increments of change between branch-specific clocks. We further develop an efficient Hamiltonian Monte Carlo sampler that exploits closed form gradient computations to scale our model to large trees. Inference under our shrinkage clock exhibits a speed-up compared to the popular random local clock when estimating branch-specific clock rates on a variety of simulated datasets. This speed-up increases with the size of the problem. We further show our shrinkage clock recovers known local clocks within a rodent and mammalian phylogeny. Finally, in a problem that once appeared computationally impractical, we investigate the heritable clock structure of various surface glycoproteins of influenza A virus in the absence of prior knowledge about clock placement. We implement our shrinkage clock and make it publicly available in the BEAST software package.

Scalable Bayesian Divergence Time Estimation With Ratio Transformations.

Divergence time estimation is crucial to provide temporal signals for dating biologically important events from species divergence to viral transmissions in space and time. With the advent of high-throughput sequencing, recent Bayesian phylogenetic studies have analyzed hundreds to thousands of sequences. Such large-scale analyses challenge divergence time reconstruction by requiring inference on highly correlated internal node heights that often become computationally infeasible. To overcome this limitation, we explore a ratio transformation that maps the original $N-1$ internal node heights into a space of one height parameter and $N-2$ ratio parameters. To make the analyses scalable, we develop a collection of linear-time algorithms to compute the gradient and Jacobian-associated terms of the log-likelihood with respect to these ratios. We then apply Hamiltonian Monte Carlo sampling with the ratio transform in a Bayesian framework to learn the divergence times in 4 pathogenic viruses (West Nile virus, rabies virus, Lassa virus, and Ebola virus) and the coralline red algae. Our method both resolves a mixing issue in the West Nile virus example and improves inference efficiency by at least 5-fold for the Lassa and rabies virus examples as well as for the algae example. Our method now also makes it computationally feasible to incorporate mixed-effects molecular clock models for the Ebola virus example, confirms the findings from the original study, and reveals clearer multimodal distributions of the divergence times of some clades of interest.

Calcifications and Cystic Morphology on Preoperative Imaging Predict Survival After Resection of Pancreatic Neuroendocrine Tumors.

BACKGROUND: Radiographic calcifications and cystic morphology are associated with higher and lower tumor grade, respectively, in pancreatic neuroendocrine tumors (PNETs). Whether calcifications and/or cystic morphology could be used preoperatively to predict post-resection survival in patients with PNETs remains elusive. METHODS: Patients undergoing curative-intent resection of well-differentiated PNETs from 2000 to 2017 at eight academic institutions participating in the US Neuroendocrine Tumor Study Group were identified. Preoperative cross-sectional imaging reports were reviewed to identify the presence of calcifications and of a cystic component occupying >50% of the total tumor area. Clinicopathologic characteristics and recurrence-free survival (RFS) were compared. RESULTS: Of 981 patients studied, 18% had calcifications and 17% had cystic tumors. Tumors with calcifications were more commonly associated with Ki-67 ≥3% (47% vs. 33%; p = 0.029), lymph node metastasis (36% vs. 24%; p = 0.011), and distant metastasis (13% vs. 4%; p < 0.001). In contrast, cystic tumors were less commonly associated with lymph node metastasis (12% vs. 30%; p < 0.001). Five-year RFS after resection was most favorable for cystic tumors without calcifications (91%), intermediate for solid tumors without calcifications (77%), and least favorable for any calcified PNET (solid 69%, cystic 67%; p = 0.043). Calcifications remained an independent predictor of RFS on multivariable analysis (p = 0.043) controlling for nodal (p < 0.001) and distant metastasis (p = 0.001). CONCLUSIONS: Easily detectable radiographic features, such as calcifications and cystic morphology, can be used preoperatively to stratify prognosis in patients with PNETs and possibly inform the decision to operate or not, as well as guide the extent of resection and potential use of neoadjuvant therapy.

Comparison of Prognostic Value of 10 Biochemical Indices at Admission for Prediction Postoperative Myocardial Injury and Hospital Mortality in Patients with Osteoporotic Hip Fracture.

Aim: To evaluate the prognostic impact at admission of 10 biochemical indices for prediction postoperative myocardial injury (PMI) and/or hospital death in hip fracture (HF) patients. Methods: In 1273 consecutive patients with HF (mean age 82.9 ± 8.7 years, 73.5% women), clinical and laboratory parameters were collected prospectively, and outcomes were recorded. Multiple logistic regression and receiver-operating characteristic analyses (the area under the curve, AUC) were preformed, the number needed to predict (NNP) outcome was calculated. Results: Age ≥ 80 years and IHD were the most prominent clinical factors associated with both PMI (with cardiac troponin I rise) and in-hospital death. PMI occurred in 555 (43.6%) patients and contributed to 80.3% (49/61) of all deaths (mortality rate 8.8% vs. 1.9% in non-PMI patients). The most accurate biochemical predictive markers were parathyroid hormone > 6.8 pmol/L, urea > 7.5 mmol/L, 25(OH)vitamin D < 25 nmol/L, albumin < 33 g/L, and ratios gamma-glutamyl transferase (GGT) to alanine aminotransferase > 2.5, urea/albumin ≥ 2.0 and GGT/albumin ≥ 7.0; the AUC for developing PMI ranged between 0.782 and 0.742 (NNP: 1.84−2.13), the AUC for fatal outcome ranged from 0.803 to 0.722, (NNP: 3.77−9.52). Conclusions: In HF patients, easily accessible biochemical indices at admission substantially improve prediction of hospital outcomes, especially in the aged >80 years with IHD.

Scalable Bayesian phylogenetics.

Recent advances in Bayesian phylogenetics offer substantial computational savings to accommodate increased genomic sampling that challenges traditional inference methods. In this review, we begin with a brief summary of the Bayesian phylogenetic framework, and then conceptualize a variety of methods to improve posterior approximations via Markov chain Monte Carlo (MCMC) sampling. Specifically, we discuss methods to improve the speed of likelihood calculations, reduce MCMC burn-in, and generate better MCMC proposals. We apply several of these techniques to study the evolution of HIV virulence along a 1536-tip phylogeny and estimate the internal node heights of a 1000-tip SARS-CoV-2 phylogenetic tree in order to illustrate the speed-up of such analyses using current state-of-the-art approaches. We conclude our review with a discussion of promising alternatives to MCMC that approximate the phylogenetic posterior. This article is part of a discussion meeting issue 'Genomic population structures of microbial pathogens'.

Development and Validation of a Modified Eighth AJCC Staging System for Primary Pancreatic Neuroendocrine Tumors.

OBJECTIVE: To improve the prognostic accuracy of the eighth edition of AJCC staging system for pNETs with establishment and validation of a new staging system. BACKGROUND: Validation of the updated eighth AJCC staging system for pNETs has been limited and controversial. METHODS: Data from the SEER registry (1975-2016) (n = 3303) and a multi-institutional database (2000-2016) (n = 825) was used as development and validation cohorts, respectively. A mTNM was proposed by maintaining the eighth AJCC T and M definitions, and the recently proposed N status as N0 (no LNM), N1 (1-3 LNM), and N2 (≥4 LNM), but adopting a new stage classification. RESULTS: The eighth TNM staging system failed to stratify patients with stage I versus IIA, stage IIB versus IIIA, and overall stage I versus II relative to long-term OS in both database. There was a monotonic decrement in survival based on the proposed mTNM staging classification among patients derived from both the SEER (5-year OS, stage I 87.0% vs stage II 80.3% vs stage III 72.9% vs stage IV 57.2%, all P < 0.001), and multi-institutional (5-year OS, stage I 97.6% vs stage II 82.7% vs stage III 78.4% vs stage IV 50.0%, all P < 0.05) datasets. On multivariable analysis, mTNM staging remained strongly associated with prognosis, as the hazard of death incrementally increased with each stage among patients in the 2 cohorts. CONCLUSION: A mTNM pNETs clinical staging system using N0, N1, N2 nodal categories was better at stratifying patients relative to long-term OS than the eighth AJCC staging.

Identifying Risk Factors and Patterns for Early Recurrence of Pancreatic Neuroendocrine Tumors: A Multi-Institutional Study.

BACKGROUND: Identifying patients at risk for early recurrence (ER) following resection for pancreatic neuroendocrine tumors (pNETs) might help to tailor adjuvant therapies and surveillance intensity in the post-operative setting. METHODS: Patients undergoing surgical resection for pNETs between 1998-2018 were identified using a multi-institutional database. Using a minimum p-value approach, optimal cut-off value of recurrence-free survival (RFS) was determined based on the difference in post-recurrence survival (PRS). Risk factors for early recurrence were identified. RESULTS: Among 807 patients who underwent curative-intent resection for pNETs, the optimal length of RFS to define ER was identified at 18 months (lowest p-value of 0.019). Median RFS was 11.0 months (95% 8.5-12.60) among ER patients (n = 49) versus 41.0 months (95% CI: 35.0-45.9) among non-ER patients (n = 77). Median PRS was worse among ER patients compared with non-ER patients (42.6 months vs. 81.5 months, p = 0.04). On multivariable analysis, tumor size (OR: 1.20, 95% CI: 1.05-1.37, p = 0.007) and positive lymph nodes (OR: 4.69, 95% CI: 1.41-15.58, p = 0.01) were independently associated with ER. CONCLUSION: An evidence-based cut-off value for ER after surgery for pNET was defined at 18 months. These data emphasized the importance of close follow-up in the first two years after surgery.

Long-Term Outcomes after Spleen-Preserving Distal Pancreatectomy for Pancreatic Neuroendocrine Tumors: Results from the US Neuroendocrine Study Group.

BACKGROUND: The adoption of spleen-preserving distal pancreatectomy (SPDP) for malignant disease such as pancreatic neuroendocrine tumors (pNETs) has been controversial. The objective of the current study was to assess the impact of SPDP on outcomes of patients with pNETs. METHODS: Patients undergoing a distal pancreatectomy for pNET between 2002 and 2016 were identified in the US Neuroendocrine Tumor Study Group database. Propensity score matching (PSM) was used to compare short- and long-term outcomes of patients undergoing SPDP versus distal pancreatectomy with splenectomy (DPS). RESULTS: Among 621 patients, 103 patients (16.6%) underwent an SPDP. Patients who underwent SPDP were more likely to have lower BMI (median, 27.5 [IQR 24.0-31.2] vs. 28.7 [IQR 25.7-33.6]; p = 0.005) and have undergone minimally invasive surgery (n = 56, 54.4% vs. n = 185, 35.7%; p < 0.001). After PSM, while the median total number of lymph nodes examined among patients who underwent an SPDP was lower compared with DPS (3 [IQR 1-8] vs. 9 [5-13]; p < 0.001), 5-year overall survival (OS) and recurrence-free survival (RFS) were comparable (OS: 96.8 vs. 92.0%, log-rank p = 0.21, RFS: 91.1 vs. 84.7%, log-rank p = 0.93). In addition, patients undergoing SPDP had less intraoperative blood loss (median, 100 mL [IQR 10-250] vs. 150 mL [IQR 100-400]; p = 0.001), lower incidence of serious complications (n = 13, 12.8% vs. n = 28, 27.5%; p = 0.014), and shorter length of stay (median: 5 days [IQR 4-7] vs. 6 days [IQR 5-13]; p = 0.049) compared with patients undergoing DPS. CONCLUSION: SPDP for pNET was associated with acceptable perioperative and long-term outcomes that were comparable to DPS. SPDP should be considered for patients with pNET.

New Nodal Staging for Primary Pancreatic Neuroendocrine Tumors: A Multi-institutional and National Data Analysis.

OBJECTIVE: To determine the prognostic role of metastatic lymph node (LN) number and the minimal number of LNs for optimal staging of patients with pancreatic neuroendocrine tumors (pNETs). BACKGROUND: Prognosis relative to number of LN metastasis (LNM), and minimal number of LNs needed to evaluate for accurate staging, have been poorly defined for pNETs. METHODS: Number of LNM and total number of LN evaluated (TNLE) were assessed relative to recurrence-free survival (RFS) and overall survival (OS) in a multi-institutional database. External validation was performed using Surveillance, Epidemiology and End Results (SEER) registry. RESULTS: Among 854 patients who underwent resection, 233 (27.3%) had at least 1 LNM. Patients with 1, 2, or 3 LNM had a comparable worse RFS versus patients with no nodal metastasis (5-year RFS, 1 LNM 65.6%, 2 LNM 68.2%, 3 LNM 63.2% vs 0 LNM 82.6%; all P < 0.001). In contrast, patients with ≥4 LNM (proposed N2) had a worse RFS versus patients who either had 1 to 3 LNM (proposed N1) or node-negative disease (5-year RFS, ≥4 LNM 43.5% vs 1-3 LNM 66.3%, 0 LNM 82.6%; all P < 0.05) [C-statistics area under the curve (AUC) 0.650]. TNLE ≥8 had the highest discriminatory power relative to RFS (AUC 0.713) and OS (AUC 0.726) among patients who had 1 to 3 LNM, and patients who had ≥4 LNM in the multi-institutional and SEER database (n = 2764). CONCLUSIONS: Regional lymphadenectomy of at least 8 lymph nodes was necessary to stage patients accurately. The proposed nodal staging of N0, N1, and N2 optimally staged patients.

A telephone-based surgical transitional care program with improved patient satisfaction scores and fiscal neutrality.

BACKGROUND: Limited data exist regarding the downstream effects of surgical transitional care programs. We explored the impact of such programs on patient satisfaction and fiscal metrics. METHODS: A telephone-based surgical transitional care program enrolled patients undergoing complex abdominal surgery between 2015 to 2017. A matched cohort undergoing similar procedures between 2010 to 2015 were used as controls. Press Ganey scores were used to reflect patient satisfaction. Hospital costs, reimbursements, and margins were analyzed for index hospitalizations and readmissions within 90 days of surgery. RESULTS: There were 607 patients in the control group and 608 in the transitional care program; survey response rates were 37% and 35%, respectively. Transitional care patients rated their understanding of personal responsibilities in post-discharge care higher than controls (59% vs 69%, P = .02). Transitional care patients felt they received better educational materials about their condition or treatment (55% vs 68%, P < .01) and rated their global hospital experience higher (46% vs 57%, P = .02). The aggregate (index plus readmission) cost was greater for the transitional care ($22,814 vs $25,827, P < .01), but there was no difference in aggregate margin ($7,027 vs $4,698, P = .25). Multivariable adjustment yielded similar results for the aggregate cost (ref vs $2,232, P = .03) and margin (ref vs $1,299, P = .23). CONCLUSION: The use of this dedicated abdominal surgery transitional care program is associated with improved Press Ganey patient education and global rating scores. The cost to support this program did not adversely affect the hospital margin when considering all factors. These data support broader investment in patient centered initiatives that may significantly enhance patient experience.

Relaxed Random Walks at Scale.

Relaxed random walk (RRW) models of trait evolution introduce branch-specific rate multipliers to modulate the variance of a standard Brownian diffusion process along a phylogeny and more accurately model overdispersed biological data. Increased taxonomic sampling challenges inference under RRWs as the number of unknown parameters grows with the number of taxa. To solve this problem, we present a scalable method to efficiently fit RRWs and infer this branch-specific variation in a Bayesian framework. We develop a Hamiltonian Monte Carlo (HMC) sampler to approximate the high-dimensional, correlated posterior that exploits a closed-form evaluation of the gradient of the trait data log-likelihood with respect to all branch-rate multipliers simultaneously. Our gradient calculation achieves computational complexity that scales only linearly with the number of taxa under study. We compare the efficiency of our HMC sampler to the previously standard univariable Metropolis-Hastings approach while studying the spatial emergence of the West Nile virus in North America in the early 2000s. Our method achieves at least a 6-fold speed increase over the univariable approach. Additionally, we demonstrate the scalability of our method by applying the RRW to study the correlation between five mammalian life history traits in a phylogenetic tree with $3650$ tips.[Bayesian inference; BEAST; Hamiltonian Monte Carlo; life history; phylodynamics, relaxed random walk.].

Indications and outcomes of enucleation versus formal pancreatectomy for pancreatic neuroendocrine tumors.

BACKGROUND: Pancreatoduodenectomy (PD) or distal pancreatectomy (DP) are common procedures for patients with a pancreatic neuroendocrine tumor (pNET). Nevertheless, certain patients may benefit from a pancreas-preserving resection such as enucleation (EN). The aim of this study was to define the indications and differences in long-term outcomes among patients undergoing EN and PD/DP. METHODS: Patients undergoing resection of a pNET between 1992 and 2016 were identified. Indications and outcomes were evaluated, and propensity score matching (PSM) analysis was performed to compare long-term outcomes between patients who underwent EN versus PD/DP. RESULTS: Among 1034 patients, 143 (13.8%) underwent EN, 304 (29.4%) PD, and 587 (56.8%) DP. Indications for EN were small size (1.5 cm, IQR:1.0-1.9), functional tumors (58.0%) that were mainly insulinomas (51.7%). After PSM (n = 109 per group), incidence of postoperative pancreatic fistula (POPF) grade B/C was higher after EN (24.5%) compared with PD/DP (14.0%) (p = 0.049). Median recurrence-free survival (RFS) was comparable among patients who underwent EN (47 months, 95% CI:23-71) versus PD/DP (37 months, 95% CI: 33-47, p = 0.480). CONCLUSION: Comparable long-term outcomes were noted among patients who underwent EN versus PD/DP for pNET. The incidence of clinically significant POPF was higher after EN.

Epidemiological hypothesis testing using a phylogeographic and phylodynamic framework.

Computational analyses of pathogen genomes are increasingly used to unravel the dispersal history and transmission dynamics of epidemics. Here, we show how to go beyond historical reconstructions and use spatially-explicit phylogeographic and phylodynamic approaches to formally test epidemiological hypotheses. We illustrate our approach by focusing on the West Nile virus (WNV) spread in North America that has substantially impacted public, veterinary, and wildlife health. We apply an analytical workflow to a comprehensive WNV genome collection to test the impact of environmental factors on the dispersal of viral lineages and on viral population genetic diversity through time. We find that WNV lineages tend to disperse faster in areas with higher temperatures and we identify temporal variation in temperature as a main predictor of viral genetic diversity through time. By contrasting inference with simulation, we find no evidence for viral lineages to preferentially circulate within the same migratory bird flyway, suggesting a substantial role for non-migratory birds or mosquito dispersal along the longitudinal gradient.

Helicobacter pylori Related Diseases and Osteoporotic Fractures (Narrative Review).

Osteoporosis (OP) and osteoporotic fractures (OFs) are common multifactorial and heterogenic disorders of increasing incidence. Helicobacter pylori (H.p.) colonizes the stomach approximately in half of the world's population, causes gastroduodenal diseases and is prevalent in numerous extra-digestive diseases known to be associated with OP/OF. The studies regarding relationship between H.p. infection (HPI) and OP/OFs are inconsistent. The current review summarizes the relevant literature on the potential role of HPI in OP, falls and OFs and highlights the reasons for controversies in the publications. In the first section, after a brief overview of HPI biological features, we analyze the studies evaluating the association of HPI and bone status. The second part includes data on the prevalence of OP/OFs in HPI-induced gastroduodenal diseases (peptic ulcer, chronic/atrophic gastritis and cancer) and the effects of acid-suppressive drugs. In the next section, we discuss the possible contribution of HPI-associated extra-digestive diseases and medications to OP/OF, focusing on conditions affecting both bone homeostasis and predisposing to falls. In the last section, we describe clinical implications of accumulated data on HPI as a co-factor of OP/OF and present a feasible five-step algorithm for OP/OF risk assessment and management in regard to HPI, emphasizing the importance of an integrative (but differentiated) holistic approach. Increased awareness about the consequences of HPI linked to OP/OF can aid early detection and management. Further research on the HPI-OP/OF relationship is needed to close current knowledge gaps and improve clinical management of both OP/OF and HPI-related disorders.

Correction to: Trends in the Number of Lymph Nodes Evaluated Among Patients with Pancreatic Neuroendocrine Tumors in the United States: A Multi-Institutional and National Database Analysis.

In the original article, Ryan C. Fields' middle initial is missing.

Natural history and cost analysis of surgical bypass versus endoscopic stenting for the palliative management of malignant gastric outlet obstruction.

BACKGROUND: Malignant gastric outlet obstruction (GOO) is managed with palliative surgical bypass or endoscopic stenting. Limited data exist on differences in cost and outcomes. METHODS: Patients with malignant GOO undergoing palliative gastrojejunostomy (GJ) or endoscopic stent (ES) were identified between 2012 and 2015 using the MarketScan® Database. Median costs (payments) for the index procedure and 90-day readmissions and re-intervention were calculated. Frequency of treatment failure-defined as repeat surgery, stenting, or gastrostomy tube-was measured. RESULTS: A total of 327 patients were included: 193 underwent GJ and 134 underwent ES. Compared to GJ, stenting resulted in lower total median payments for the index hospitalization and procedure-related 90-day readmissions ($18,500 ES vs. $37,200 GJ, p = 0.032). For patients treated with ES, 25 (19%) required a re-intervention for treatment-failure, compared to 18 (9%) patients who underwent GJ (p = 0.010). On multivariable analysis, stenting remained significantly associated with need for secondary re-intervention compared to GJ (HR for ES 2.0 [1.1-3.8], p 0.028). CONCLUSION: In patients with malignant GOO, endoscopic stenting results in significant 90-day cost saving, however was associated with twice the rate of secondary intervention. The decision for surgical bypass versus endoscopic stenting should consider patient prognosis, anticipated cost, and likelihood of needing re-intervention.

Resection of pancreatic neuroendocrine tumors: defining patterns and time course of recurrence.

BACKGROUND: To define recurrence patterns and time course, as well as risk factors associated with recurrence following curative resection of pNETs. METHOD: Patients who underwent curative-intent resection for pNET between 1997 and 2016 were identified from the US Neuroendocrine Tumor Study Group. Data on baseline and tumor-specific characteristics, overall survival (OS), timing and first-site of recurrence, predictors and recurrence management were analyzed. RESULTS: Among 1020 patients, 154 (15.1%) patients developed recurrence. Among patients who experienced recurrence, 76 (49.4%) had liver-only recurrence, while 35 (22.7%) had pancreas-only recurrence. The proportion of liver-only recurrence increased from 54.3% within one-year after surgery to 61.5% from four-to-six years after surgery; whereas the proportion of pancreas-only recurrence decreased from 26.1% to 7.7% over these time periods. While liver-only recurrence was associated with tumor characteristics, pancreas-only recurrence was only associated with surgical margin status. Patients undergoing curative resection of recurrence had comparable OS with patients who had no recurrence (median OS, pancreas-only recurrence, 133.9 months; liver-only recurrence, not attained; no recurrence, 143.0 months, p = 0.499) CONCLUSIONS: Different recurrence patterns and timing course, as well as risk factors suggest biological heterogeneity of pNET recurrence. A personalized approach to postoperative surveillance and treatment of recurrence disease should be considered.