Imputation of missing clock times – application to procalcitonin concentration time course after birth

McDonough M.H., Stocker S.L., Kippin T., Meiring W., Plaxco K.W.

AimPharmacokinetics have historically been assessed using drug concentration data obtained via blood draws and bench‐top analysis. The cumbersome nature of these typically constrains studies to at most a dozen concentration measurements per dosing event. This, in turn, limits our statistical power in the detection of hours‐scale, time‐varying physiological processes. Given the recent advent of in vivo electrochemical aptamer‐based (EAB) sensors, however, we can now obtain hundreds of concentration measurements per administration. Our aim in this paper was to assess the ability of these time‐dense datasets to describe time‐varying pharmacokinetic models with good statistical significance.MethodsWe used seconds‐resolved measurements of plasma tobramycin concentrations in rats to statistically compare traditional one‐ and two‐compartmental pharmacokinetic models to new models in which the proportional relationship between a drug's plasma concentration and its elimination rate varies in response to changing kidney function.ResultsWe found that a modified one‐compartment model in which the proportionality between the plasma concentration of tobramycin and its elimination rate falls reciprocally with time either meets or is preferred over the standard two‐compartment pharmacokinetic model for half of the datasets characterized. When we reduced the impact of the drug's rapid distribution phase on the model, this one‐compartment, time‐varying model was statistically preferred over the standard one‐compartment model for 80% of our datasets.ConclusionsOur results highlight both the impact that simple physiological changes (such as varying kidney function) can have on drug pharmacokinetics and the ability of high‐time resolution EAB sensor measurements to identify such impacts.

D’Souza S., Guhadasan R., Jennings R., Siner S., Paulus S., Thorburn K., Chesters C., Downey C., Baines P., Lane S., Carrol E.

Following surgery, it is difficult to distinguish a postoperative inflammatory reaction from infection. This study examined the predictive value of the biomarkers; procalcitonin, C-reactive protein, lactate, neutrophils, lymphocytes, platelets, and the biphasic activated partial thromboplastin time waveform in diagnosing bacterial infection following cardiac surgery.Prospective, observational study.A regional, PICU in the United Kingdom.Three-hundred sixty-eight children under the age of 16 admitted to the PICU for elective cardiac surgery were enrolled in the study.All biomarker measurements were determined daily until postoperative day 7. Children were assessed for postoperative infection until day 28 and divided into four groups: bacterial infection, culture-negative sepsis, viral infection, and no infection. We used the Kruskal-Wallis test, chi-square test, analysis of variance, and area under the curve in our analysis.In total, 71 of 368 children (19%) developed bacterial infection postoperatively, the majority being surgical site infections. In those with bacterial infection, procalcitonin was elevated on postoperative days 1-3 and the last measurement prior to event compared with those without bacterial infection. The most significant difference was the last measurement prior to event; 0.72 ng/mL in the bacterial infection group versus 0.13 ng/mL in the no infection group (for all groups; p < 0.001). Longitudinal profiles of all biomarkers were indistinct in the bacterial infection and nonbacterial infection groups except in those with culture-negative infections who had distinct procalcitonin kinetics on postoperative days 1-4. Children with culture-negative sepsis required longer ventilatory support and PICU stay and were more likely to develop complications than the other groups.None of the biomarkers studied within 3 days of infection distinguished between infection and postoperative inflammatory reaction. However, procalcitonin kinetics peaked on postoperative day 2 and fell more sharply than C-reactive protein kinetics, which peaked at postoperative day 3. The monitoring of procalcitonin kinetics following cardiac surgery may help guide rational antimicrobial use.

Lee J., Bang Y.H., Lee E.H., Choi B.M., Hong Y.S.

Nguyen T.H., Mouksassi M., Holford N., Al‐Huniti N., Freedman I., Hooker A.C., John J., Karlsson M.O., Mould D.R., Pérez Ruixo J.J., Plan E.L., Savic R., van Hasselt J.G., Weber B., Zhou C., et. al.

Gilroy D.W., Maini A.A., De Maeyer R.P., Segre E., Fullerton J.N.

Fukuzumi N., Osawa K., Sato I., Iwatani S., Ishino R., Hayashi N., Iijima K., Saegusa J., Morioka I.

Procalcitonin (PCT) levels are elevated early after birth in newborn infants; however, the physiological features and reference of serum PCT concentrations have not been fully studied in preterm infants. The aims of the current study were to establish an age-specific percentile-based reference curve of serum PCT concentrations in preterm infants and determine the features. The PCT concentration peaked in infants at 1 day old and decreased thereafter. At 1 day old, serum PCT concentrations in preterm infants <34 weeks’ gestational age were higher than those in late preterm infants between 34 and 36 weeks’ gestational age or term infants ≥37 weeks’ gestational age. Although the 50-percentile value in late preterm and term infants reached the adult normal level (0.1 ng/mL) at 5 days old, it did not in preterm infants. It took 9 weeks for preterm infants to reach it. Serum PCT concentrations at onset in late-onset infected preterm infants were over the 95-percentile value. We showed that the physiological feature in preterm infants was significantly different from that in late preterm infants, even in those <37 weeks’ gestational age. To detect late-onset bacterial infection and sepsis, an age-specific percentile-based reference curve may be useful in preterm infants.

Zhang W.R., Garg A.X., Coca S.G., Devereaux P.J., Eikelboom J., Kavsak P., McArthur E., Thiessen-Philbrook H., Shortt C., Shlipak M., Whitlock R., Parikh C.R.

Hur J., Yang H.T., Chun W., Kim J., Shin S., Kang H.J., Kim H.S.

Owen J.S., Fiedler-Kelly J.

Johansson Å.M., Karlsson M.O.

Missing covariate data is a common problem in nonlinear mixed effects modelling of clinical data. The aim of this study was to implement and compare methods for handling missing covariate data in nonlinear mixed effects modelling under different missing data mechanisms. Simulations generated data for 200 individuals with a 50% difference in clearance between males and females. Three different types of missing data mechanisms were simulated and information about sex was missing for 50% of the individuals. Six methods for handling the missing covariate were compared in a stochastic simulations and estimations study where 200 data sets were simulated. The methods were compared according to bias and precision of parameter estimates. Multiple imputation based on weight and response, full maximum likelihood modelling using information on weight and full maximum likelihood modelling where the proportion of males among the individuals lacking information about sex was estimated (EST) gave precise and unbiased estimates in the presence of missing data when data were missing completely at random or missing at random. When data were missing not at random, the only method resulting in low bias and high parameter precision was EST.

Holford N., Heo Y., Anderson B.

The pharmacokinetic behavior of medicines used in humans follows largely predictable patterns across the human age range from premature babies to elderly adults. Most of the differences associated with age are in fact due to differences in size. Additional considerations are required to describe the processes of maturation of clearance processes and postnatal changes in body composition. Application of standard approaches to reporting pharmacokinetic parameters is essential for comparative human pharmacokinetic studies from babies to adults. A standardized comparison of pharmacokinetic parameters obtained in children and adults is shown for 46 drugs. Appropriate size scaling shows that children (over 2 years old) are similar to adults. Maturation changes are generally completed within the first 2 years of postnatal life; consequently babies may be considered as immature children, whereas children are just small adults.

Auriti C., Fiscarelli E., Ronchetti M.P., Argentieri M., Marrocco G., Quondamcarlo A., Seganti G., Bagnoli F., Buonocore G., Serra G., Bacolla G., Mastropasqua S., Mari A., Corchia C., Prencipe G., et. al.

Objective To investigate the accuracy of procalcitonin (PCT) as a diagnostic marker of nosocomial sepsis (NS) and define the most accurate cut-off to distinguish infected from uninfected neonates.Setting Six neonatal intensive care units (NICUs).Patients 762 neonates admitted to six NICUs during a 28-month observational study for whom at least one serum sample was taken on admission.Main outcome measures Positive and negative predictive values at different PCT cut-off levels.Results The overall probability of an NS was doubled or more if PCT was >0.5 ng/ml. In very-low-birth-weight (VLBW) infants, a cut-off of >2.4 ng/ml gave a positive predictive value of NS near to 50% with a probability of a false-positive diagnosis of NS in about 10% of the patients.Conclusions In VLBW neonates, a serum PCT value >2.4 ng/ml prompts early empirical antibiotic therapy, while in normal-birth-weight infants, a PCT value ≤2.4 ng/ml carries a low risk of missing an NS.

Bohnhorst B., Lange M., Bartels D.B., Bejo L., Hoy L., Peter C.

To evaluate which clinical symptoms indicate proven neonatal bacterial infection (NBI) and whether measuring procalcitonin aside from C-reactive protein and interleukin 6 improves sensitivity and specificity in diagnosis.In a prospective observational study, clinical symptoms and procalcitonin, C-reactive protein and interleukin 6 were simultaneously determined from the 4th day of life in 170 preterm and term neonates at the first time of suspicion of NBI. Proven NBI was defined as a positive culture of otherwise sterile body fluids or radiologically verified pneumonia in combination with elevated inflammatory markers.Fifty-eight (34%) patients were diagnosed with proven late-onset NBI. In case of proven NBI, odds ratio and 95% confidence intervals were 2.64 (1.06-6.54) for arterial hypotension, 5.16 (2.55-10.43) for feeding intolerance and 9.18 (4.10-20.59) for prolonged capillary refill. Sensitivity of combined determination of C-reactive protein (>10 mg/L) and interleukin 6 (>100 pg/mL) was 91.4%, specificity 80.4%, positive predictive value 70.7% and negative predictive value 94.7%. The additional determination of procalcitonin (>0.7 ng/mL) resulted in 98.3%, 65.2%, 58.8% and 98.6%, respectively.Arterial hypotension, feeding intolerance and especially prolonged capillary refill indicate proven neonatal late-onset bacterial infection, even at the time of first suspicion. Additional measurement of procalcitonin does indeed improve sensitivity to nearly 100%, but is linked to a decline in specificity. Nevertheless, in the high-risk neonatal population, additional procalcitonin measurement can be recommended because all infants with NBI have to be identified.

Chiesa C., Natale F., Pascone R., Osborn J.F., Pacifico L., Bonci E., De Curtis M.

There is still no study evaluating the influence of gestational age (GA) per se on C reactive protein (CRP) and procalcitonin (PCT) reference intervals. We therefore investigated how length of gestation, age (hours), and prenatal and perinatal variables might influence the levels of CRP and PCT. We also determined 95% age-specific reference intervals for CRP and PCT in healthy preterm and term babies during the early neonatal period.One blood sample (one observation per neonate) was taken for CRP and PCT from each newborn between birth and the first 4 (for term), or 5 days (for preterm newborns) of life by using a high-sensitive CRP and PCT assays.Independently of gender and sampling time, GA had a significantly positive effect on CRP, and a significantly negative effect on PCT. Compared with healthy term babies, healthy preterm babies had a lower and shorter CRP response, and, conversely, an earlier, higher, and longer PCT response. CRP reference intervals were affected by a number of pro-inflammatory risk factors.Age- and GA-specific reference ranges for both CRP and PCT should be taken into account to optimize their use in the diagnosis of early-onset neonatal sepsis.

Sumpter A.L., Holford N.H.

To describe the pattern and variability of body weight with postmenstrual age (PMA) using nonlinear mixed effect modeling and to create a single mathematical function that can be used from prematurity to adulthood.PMA has been shown to predict functional properties of humans such as glomerular filtration rate and drug clearance. Widely used growth charts use postnatal age to predict weight in an idealized population and are not available as a mathematical function.We modeled 7164 body weight and PMA observations from a pooled database of 5031 premature neonates, infants, children, and adults. All subjects were participants in pharmacokinetic or renal function studies. PMA ranged from 23 weeks to 82 years. A mixed effect model was used to describe fixed (PMA, sex) and random between-subject variability.A model based on the sum of three sigmoid hyperbolic and one exponential functions described the data. Females were typically 12% lighter in weight. Part of the between-subject variability in weight decreased exponentially with a half-life of 3.5 PMA years, while the remainder stayed a constant fraction of the weight asymptote for each of the four functions.The change of weight with PMA and sex can be described with a simple equation. This is suitable for simulation of typical weight-age distributions and may be useful for evaluation of appropriate weight for age in children requiring medical treatment.