Homocystinuria is an autosomal recessive disorder usually caused by deficiency of cystathionine -synthase, leading to grossly increased plasma and urine concentrations of homocysteine. There is considerable evidence that early detection and treatment can prevent the clinical consequences of the enzyme deficiency (1, 2 ); therefore, screening for the disorder has been advocated (1 ). Many cases of homocystinuria have secondary hypermethioninemia. Neonatal screening for homocystinuria by measurement of increased methionine concentrations in dried blood spots (DBS) has been performed in some centers but has poor sensitivity. Approximately 20% of cases are missed, partly because of low methionine concentrations in breast milk and some infant formulas (3 ). The measurement of homocysteine in DBS has not been used as a screen for homocystinuria, partly because of uncertainty about the suitability of DBS samples. Homocysteine concentrations are unstable in whole blood stored at room temperature and increase by ϳ1.0 mol/L per hour (4 ). This is attributable to in vitro erythrocyte transmethylation reactions that lead to continuous production and release of homocysteine (5 ). It is not known whether homocysteine is released from erythrocytes in blood that has been spotted on filter paper and dried, either during the spotting and drying process or during storage at room temperature. A recent report assessed stability in screening cards that were stored at 4°C, which does not reflect routine practice (6 ). Another investigated the stability of DBS samples, using whole blood to which large concentrations of aqueous homocysteine calibrator had been added, thus potentially masking any increase in homocysteine attributable to release from erythrocytes (7 ). The lack of information on the stability of homocysteine in DBS as applied to neonatal screening prompted the following investigation.Blood samples were collected from adult inpatients into tubes containing potassium EDTA. Samples were mixed gently, and blood was spotted on filter paper cards (standard neonatal screening cards) and dried in air. The cards were stored at room temperature until analysis. Immediately after the blood spots were prepared, the remaining samples were centrifuged, and the plasma was removed and stored at Ϫ20°C until analysis. Ethical approval was obtained from the UK South West Local Research Ethics Committee.We measured total homocysteine in DBS by HPLC, using a modification of a method for plasma total homocysteine (8 ). We punched 6-mm DBS into flat-bottomed tubes, added 80 L of 2.16 mol/L cysteamine (internal standard) and 15 L of 100 mL/L tributylphosphine in dimethylformamide to the tubes, and incubated the tubes at 4°C for 30 min. We used 7-fluorobenzo-2-oxa-1,3-diazole-4-sulfuric acid to derivatize the thiols. Separation was performed by reversed-phase HPLC with fluorometric detection (Shimadzu RF-10AXL). The method was standardized by use of 15 L of an aqueous homocysteine calibrator (20 mol/L) in place of DBS. The DBS method was linear u...