br Results Of the U S

Results Of the 51 U.S. jurisdictions surveyed (Table 1), 33 reported that hyperargininemia is one of the conditions for which all newborns are required to be screened, with the earliest screening reported in Massachusetts in 1999. Of the 18 other jurisdictions, an additional 5 reported that hyperargininemia would likely be detected by the screening algorithm currently in use for other metabolic conditions, and screen positive cases would be followed up accordingly, even though screening is not required. Thirteen state screening programs reported no screening for hyperargininemia (Alabama, Arizona, Arkansas, Florida, Kansas, Maryland, Montana, Nebraska, South Carolina, Virginia, Washington, West Virginia and Wisconsin). One jurisdiction indicated that arginine levels were not officially reported as part of the program, even though they were observable, since hyperargininemia was not included in the program's current screening mandate. While most survey respondents noted that an elevation of Arg was the trigger leading to further investigation of the possibility of hyperargininemia, the analytical cutoffs requiring additional follow-up (screen positive) varied widely. Multiple cutoffs related to age at time of screening and to birth weights were reported by some programs. Most reported using two Arg cutoff levels leading to two different follow-up pathways, repeat filter paper screening or referral for clinical evaluation. A few programs reported that elevated Arg values above a single cutoff simply resulted in clinical referral (i.e. a repeat filter paper was not part of the general follow-up protocol, although a repeat specimen after clinical referral might occur). The range of Arg values considered actionable (requiring additional follow-up of any type) varied from 20μmol/L to 130μmol/L. Some programs chose lower cutoffs for Arg initially and used a second-tier ratio or combination of ratios of other ProteOrange to ultimately reduce the number of patients recalled. Arg/Orn was the most popular second-tier discriminator with cutoffs for actionable levels varying from 0.45 to 1.5. Other ratios in use included Arg/Ala, Arg/Phe, Cit/Arg and Arg/[Phe×Leu]. Using archived analytical MS/MS data from confirmed cases in California, we constructed a table (Table 2) to illustrate the results that would be obtained using the various reported screening protocols. As a point of reference, 5.4 million specimens were screened during this time period and 138 were originally found to have both Arg≥50μmol/L and Arg/Orn ≥1.40, the criteria for screening positivity during this time period. Clinical evaluation resulted in the 9 cases referenced. In order to assess the potential impact of some of these protocols on screening follow-up, we developed a table (Table 3) to show the percentage of newborns that would require additional follow-up based a proposed protocol using 2015 California NBS data. We chose consider Arg alone, Arg in combination with Arg/Orn, and Arg in combination with Arg/[Phe×Leu]. We also included the R4S Tool Runner multivariate/metabolic profile analysis [8], [9]. While the California hyperargininemia protocol resulted in 0.01% follow-up, two other protocols (utilization Arg and Arg/[Phe×Leu] and R4S Tool Runner appeared to be at least as good.
Discussion Ideally, newborn screening for a particular disorder would be uniform between jurisdictions and have a very low false negative rate and a low recall (false positive) rate. Screening algorithms would be based on sufficient case detection evidence to validate the algorithm with modification as new data accumulates. It is clear from the information presented here that NBS for hyperargininemia in the U.S. is far from ideal with significant variability between state screening programs. Variations appear to be based on anecdotal findings without a solid scientific basis (Table 1). Reliable national NBS incidence data beyond that reported here do not exist.