Opioids such as morphine are the cornerstone of pain treatment. The challenge of measuring the concentrations of morphine and its active metabolites in order to assess human pharmacokinetics and monitor therapeutic drugs in children requires assays with high sensitivity in small blood volumes. We developed and validated a semi-automated LC-MS/MS assay for the simultaneous quantification of morphine and its active metabolites morphine 3β-glucuronide (M3G) and morphine 6β-glucuronide (M6G) in human plasma and in dried blood spots (DBS). Reconstitution in water (DBS only) and addition of a protein precipitation solution containing the internal standards were the only manual steps. Morphine and its metabolites were separated on a Kinetex 2.6-μm PFP analytical column using an acetonitrile/0.1% formic acid gradient. The analytes were detected in the positive multiple reaction mode. In plasma, the assay had the following performance characteristics: range of reliable response of 0.25-1000 ng/mL (r(2) > 0.99) for morphine, 1-1,000 ng/mL (r(2) > 0.99) for M3G, and 2.5-1,000 ng/mL for M6G. In DBS, the assay had a range of reliable response of 1-1,000 ng/mL (r(2) > 0.99) for morphine and M3G, and of 2.5-1,000 ng/mL for M6G. For inter-day accuracy and precision for morphine, M3G and M6G were within 15% of the nominal values in both plasma and DBS. There was no carryover, ion suppression, or matrix interferences. The assay fulfilled all predefined acceptance criteria, and its sensitivity using DBS samples was adequate for the measurement of pediatric pharmacokinetic samples using a small blood of only 20-50 μL.

Preterm and term neonates often require surgical procedures and analgesia. However, our knowledge about neonatal pharmacokinetics of fentanyl, the most commonly used drug for these procedures, and its metabolites is still incomplete. To facilitate pharmacokinetic studies of fentanyl and its metabolites in neonates and other children, we developed and validated an LC-MS/MS method based on minimally invasive, low blood volume sampling. LC-MS/MS was used for the simultaneous analysis of fentanyl, despropionyl fentanyl (DPF), and norfentanyl from dried blood samples (DBS) collected on filter paper. Positive ions were monitored using multiple reaction monitoring. Since the standard matrix for measuring fentanyl blood concentrations is plasma, the assay was developed and validated in plasma, whole blood, and then DBS. Our method was able to measure clinically relevant levels of fentanyl and its metabolites. In DBS, the lower limits of quantification were 100 pg/mL for fentanyl with a range of reliable response from 0.1 to 100 ng/mL (r(2)>0.99) and 250 pg/mL for both DPF and norfentanyl with a range of reliable response from 0.25 to 100 ng/mL (r(2)>0.99). In plasma and in DBS inter-day accuracy and precisions of fentanyl met predefined acceptance criteria and also indicated comparable assay performance in both matrices.

Modern analytics applications use a diverse mix of libraries and functions. Unfortunately, there is no optimization across these libraries, resulting in performance penalties as high as an order of magnitude in many applications. To address this problem, we proposed Weld, a common runtime for existing data analytics libraries that performs key physical optimizations such as pipelining under existing, imperative library APIs. In this work, we further develop the Weld vision by designing an automatic adaptive optimizer for Weld applications, and evaluating its impact on realistic data science workloads. Our optimizer eliminates multiple forms of overhead that arise when composing imperative libraries like Pandas and NumPy, and uses lightweight measurements to make data-dependent decisions at runtime in ad-hoc workloads where no statistics are available, with sub-second overhead. We also evaluate which optimizations have the largest impact in practice and whether Weld can be integrated into libraries incrementally. Our results are promising: using our optimizer, Weld accelerates data science workloads by up to 23× on one thread and 80× on eight threads, and its adaptive optimizations provide up to a 3.75× speedup over rule-based optimization. Moreover, Weld provides benefits if even just 4-5 operators in a library are ported to use it. Our results show that common runtime designs like Weld may be a viable approach to accelerate analytics.

BACKGROUND: There are few comparative data on the analgesic options used to manage patients undergoing minimally invasive repair of pectus excavatum (MIRPE). The Society for Pediatric Anesthesia Improvement Network was established to investigate outcomes for procedures where there is significant management variability. For our first study, we established a multicenter observational database to characterize the analgesic strategies used to manage pediatric patients undergoing MIRPE. Outcome data from the participating centers were used to assess the association between analgesic strategy and pain outcomes. METHODS: Fourteen institutions enrolled patients from June 2014 through August 2015. Network members agreed to an observational methodology where each institution managed patients based on their institutional standards and protocols. There was no requirement to standardize care. Patients were categorized based on analgesic strategy: epidural catheter (EC), paravertebral catheter (PVC), wound catheter (WC), no regional (NR) analgesia, and intrathecal morphine techniques. Primary outcomes, pain score and opioid consumption by postoperative day (POD), for each technique were compared while adjusting for confounders using multivariable modeling that included 5 covariates: age, sex, number of bars, Haller index, and use of preoperative pain medication. Pain scores were analyzed using repeated-measures analysis of variance with Bonferroni correction. Opioid consumption was analyzed using a multivariable quantile regression. RESULTS: Data were collected on 348 patients and categorized based on primary analgesic strategy: EC (122), PVC (57), WC (41), NR (120), and intrathecal morphine (8). Compared to EC, daily median pain scores were higher in patients managed with PVC (POD 0), WC (POD 0, 1, 2, 3), and NR (POD 0, 1, 2), respectively (P < .001–.024 depending on group). Daily opioid requirements were higher in patients managed with PVC (POD 0, 1), WC (POD 0, 1, 2), and NR (POD 0, 1, 2) when compared to patients managed with EC (P < .001). CONCLUSIONS: Our data indicate variation in pain management strategies for patients undergoing MIRPE within our network. The results indicate that most patients have mild-to-moderate pain postoperatively regardless of analgesic management. Patients managed with EC had lower pain scores and opioid consumption in the early recovery period compared to other treatment strategies.

Virtual reality (VR), a computer-generated simulation of a 3-dimensional environment, is a relatively new method of providing distraction before and during procedures. We describe the use of a VR multiuser application, Oculus Rooms, to calm an anxious 10-year-old boy during transportation to the operating room and induction of anesthesia. The use of VR could lessen the perioperative anxiety of children by maintaining a virtual child-parent connection while avoiding the potential drawbacks to having parents actually in the operating room before and during induction of anesthesia. The success of this novel technique has important implications for future clinical trials and practice.