Neonatal Expanded Screening For Inherited Metabolic Diseases: Planning By Simulation

Inherited metabolic diseases are rare but may cause severe damages if not cared for in time. A neonatal screening program, including a first quick test to all newborns and a second accurate test for patients with positive results, is being planned in Veneto region, northeast Italy. A simulation model describing all operations has been built and implemented with the scope of giving suitable dimension to operating centres, in order to care for all revealed pathology within the maximum time before they may become dangerous for the involved patients. The model is pretty general and its application may be extended to changed situation of the same region or to other regions. INTRODUCTION The screening is the application of a test to all subjects of a population with the scope of identifying a disease at the moment when it is asymptomatic. A screening test has not the significance of a diagnostic test: it identifies a person who appears to be sound but probably suffers from a disease among people who do not; people with either positive or suspicious result shall be sent to the doctor for diagnosis and necessary therapy. The idea of an early disease diagnosis and treatment is simple; conversely the path, which brings on one side to care people for a previously undiagnosed disease and on the other side not to damage those who do not need any treatment, is not simple. Then in (Wilson and Junger 1968) leading criteria were fixed to identify pathologies for which a screening program is appropriate; such criteria were repeatedly updated and are synthetized in (Andermann et al. 2008) as follows: the screening program shall meet a recognized need; screening objectives shall be defined from the beginning; target population shall be defined; scientific evidence of screening program effectiveness is necessary; the plan shall integrate education, evaluation, clinical services and management plans; a quality control, which minimizes potential screening risks, is necessary; the plan shall assure well informed choices, confidentiality and autonomy respect; the plan shall promote screening access and equity to the whole population; evaluation plans shall be prepared from the beginning; screening total benefits shall outweigh damages. The scope of neonatal screening lays in identifying newborns with severe pathologies in order to promote appropriate interventions to avoid or to improve adverse outcomes (Wilcken and Wiley 2008). Biochemical mass test on newborns began in 1960 with the adoption of a screening for phenylketonuria, a rare congenital metabolism error which, if not treated, leads to a severe mental retardation. Gutrie in 1960 developed a methodology to measure phenylalanine concentration in blood; this test required a few blood drops taken from the heel and soaked into a blotting paper now known as Gutrie card (Marsden and Levy 2010); moreover it has the characteristic of being quickly effected on a large sample number. Such a test first became compulsory in Massachusetts in 1963, but soon in many states neonatal screening test plans took place. Note that the same sample may be used for many tests, so that other pathologies were introduced in the neonatal screen plans. In the 90’s the development of tandem mass spectrometry brought a great change to neonatal screening, as this method permits identification of a large number of metabolites from the same sample of a few blood drops on the blotting paper, so that the screening for 30-40 metabolic pathologies is possible. Pilot plans developed in Australia and New England studied tandem mass screening effectiveness and revealed a higher identification capacity if compared with clinical diagnoses (Zytkovicz et al. 2001, Wilcken et al. 2003); moreover results showed the advantage of better prognosis of identified and precociously treated patients. Neonatal screening programs with tandem mass technology were developed in Australia, Canada, Qatar and in the majority of U.S.A.. In Europe 24 states Proceedings 25th European Conference on Modelling and Simulation ©ECMS Tadeusz Burczynski, Joanna Kolodziej Aleksander Byrski, Marco Carvalho (Editors) ISBN: 978-0-9564944-2-9 / ISBN: 978-0-9564944-3-6 (CD) activated such plans, either applied to the whole country or limited to some regions (Bodamer et al. 2007). The set of screened metabolic diseases is different among the various states. Even if many papers were written about simulation in health care, no paper was found about simulation of neonatal screening except (Da Frè 2011). PROJECT FOR AN EXPANDED NEONATAL SCREENING IN VENETO REGION A project to execute an expanded neonatal screening in Veneto region (northeast Italy), 4,900,000 inhabitants, has been recently initiated. Such a project plans to create a network which includes two already operating centres, placed in Verona (West Veneto) and in Padova (East Veneto); the catchment area includes 42 birth centres, of which 15 in the provinces of Verona and Vicenza, which will refer to Verona, with about 19,000 newborns per year, and 27 in the provinces of Padova, Rovigo, Treviso, Venezia and Belluno, which will refer to Padova, with about 27,000 newborns per year. Probably the catchment area will expand to two neighbour regions in the future. Pahologies to be screened for are listed below; for each pathology there is written the official name, the usual abbreviation and the priority; high priority pathologies are characterized by +++, which specifies that such pathologies require beginning the appropriate therapy within the first week of life, middle priority pathologies by ++, as they require to begin the therapy within the second week of life, low priority by +, as they do not require any therapy within the first two weeks of life: Phenylketonuria (PKU) ++ Tyrosinemia, Type I (TYR I) +++ Citrullinemia (CIT) +++ Argininosuccinic Aciduria (ASA) +++ Arginase Deficiency (ARG) +++ Maple Syrup Urine Disease (MSUD) +++ Propionic Acidemia (PA) +++ Methylmalonyl-CoA Mutase (MMA) +++ Methylmalonic acidemia with homocystinuria type C/D (Cbl C/D) +++ Isovaleric acidemia (IVA) +++ Glutaric aciduria 1 (GA I) ++ Very long chain acyl-CoA dehydrogenase deficiency (VLCAD) +++ Long chain 3-hydroxyacyl-CoA dehydrogenase deficiency (LCHAD) +++ Medium chain acyl-CoA dehydrogenase deficiency (MCAD) ++ Short chain Acyl-CoA dehydrogenase deficiency (SCAD) +++ Glutaric aciduria 2 (GA II) ++ Carnitine palmitoyl transferase 1 deficiency (CPT I) ++ Carnitine palmitoyl transferase II deficiency (CPT II) +++ Carnitine-acylcarnitine translocase deficiency (CACT) +++ Carnitine uptake deficiency (CUD) ++ Glycogen Storage Disease Type II (Pompe Disease) (GSD II) ++ Fabry Disease (FD) + Mucopolysaccharidosis I, II, VI (MPS I, II, VI) + Actions related to neonatal screening for metabolic diseases are reported in Figure 1. Within two-three days from birth a sample is taken from every newborn; the sample is quickly dispatched to the first level centre of reference, where it is measured as soon as the machine is idle; if the first measuring gives a borderline result, then a second measuring is performed; if the result is uncertain a new sample is taken immediately, if it is related to a premature newborn a new sample is taken after fifteen days; if the result is clear and positive, the newborn is sent to the second level centre in Padova, to make a further test, in order to get an accurate diagnosis and to set up a therapy plan; the second level test is executed as soon as a free slot is found. Note that first level centres are planned to operate for six days a week.and the only second level centre for five days a week. THE PROBLEM The main concern of Region Health authorities is to find the right dimension for the two first level centres, placed in Verona and Padova, in terms of capacity, i.e., number of measured samples per day, and for the second level centre, placed in Padova, in terms of number of tests executed per week. It is obvious the capacity shall be decided in order not to impose dangerous delays in the processing of samples and positive patients, but not to require excessive expense by creating overdimensioned centres. Keep in mind that the study shall be open to reconsider centres dimensioning in the case either the population of the catchment area changes, or the catchment area expands. Given the variability of birth process and the flexibility we require to the model, we chose simulation as the most suitable technique to solve the problem; such a methodology permits us to know in advance the effects of our suggestions without implementing any new method in clinical practice; moreover many different alternatives can be tested witout any actual experiment. DATA ANALYSIS Data concerning birth process were obtained from the local birth records reporting births of year 2009. An accurate study revealed that the birth rate depends both on the month of the year and on the day of the week: the expected amount of births per day is given by a regression model expressed by the formula: μij = α + βi + δj where i=1,...,12 and j=1,...,7 Figure 1: Flow Chart of Actions Related to Screening according to the following parameters: α 105,48 β10 2,45 β 1 0,00 β 11 -6,13 β 2 -1,57 β 12 -3,59