Goemans, B.F., Zwaan, C.M., Miller, M., Zimmermann, M., Harlow, A., Meshinchi, S., Loonen, A.H., Hahlen, K., Reinhardt, D., Creutzig, U., Kaspers, G.J. & Heinrich, M.C. (2005) Mutations in KIT and RAS are frequent events in pediatric core binding factor acute myeloid leukemia. Leukemia, in press. Loh, M.L., Reynolds, M.G., Vattikuti, S., Gerbing, R.B., Alonzo, T.A., Carlson, E., Cheng, J.W., Lee, C.M., Lange, B.J. & Meshinchi, S. (2004) PTPN11 mutations in pediatric patients with acute myeloid leukemia: results from the Children’s Cancer Group. Leukemia, 18, 1831–1834. Tartaglia, M., Kalidas, K., Shaw, A., Song, X., Musat, D.L., van der Burgt, I., Brunner, H.G., Bertola, D.R., Crosby, A., Ion, A., Kucherlapati, R.S., Jeffery, S., Patton, M.A. & Gelb, B.D. (2002) PTPN11 mutations in Noonan syndrome: molecular spectrum, genotype-phenotype correlation, and phenotypic heterogeneity. American Journal of Human Genetics, 70, 1555–1563. Tartaglia, M., Martinelli, S., Iavarone, I., Cazzaniga, G., Spinelli, M., Giarin, E., Petrangeli, V., Carta, C., Masetti, R., Arico, M., Locatelli, F., Basso, G., Sorcini, M., Pession, A. & Biondi, A. (2005) Somatic PTPN11 mutations in childhood acute myeloid leukaemia. British Journal of Haematology, 129, 333–339. Keywords: PTPN11 mutation, FAB M5 AML, paediatric AML, geographical differences. doi: 10.1111/j.1365-2141.2005.05685.x Diagnostic performance of D-dimer is lower in elderly outpatients with suspected deep venous thrombosis We read with interest the paper by Schutgens et al (2005). Their main conclusion regarding a lower diagnostic value of D-dimer (DD) for the exclusion of deep venous thrombosis (DVT) in the elderly was in keeping with previously published experience (van der Graaf et al, 2000; Aguilar et al, 2002); as the authors found that the mean DD plasma concentrations rise and the specificity of this test decreases with age. Following consideration of the results reported by Schutgens et al (2005) we decided to evaluate the diagnostic value of DD in different age groups of a prospectively followed series of outpatients presenting with clinically suspected DVT. Exclusion criteria were age below 18 years, pregnancy, ongoing oral anticoag- ulant treatment, the use of therapeutic doses of low molecular weight heparin during the previous 48 h or protocol violation. All patients underwent an initial clinical evaluation accord- ing to the Wells modified score (Wells et al, 2003) and DD testing using a quantitative immunoturbidimetric test (STA LiatestÒ D-Di; Diagnostica Stago, Asnie ´res sur Seine, France) assayed in a STA-Compact coagulometer (Diagnostica Stago). A DD concentration <0Æ4 ng/ml was considered to be normal as previously validated (van der Graaf et al, 2000; Aguilar et al, 2002); results were available in 10 min. Subsequent manage- ment has been previously reported; briefly, patients initially scored as unlikely to have DVT and presenting with a normal DD level were discharged without any further diagnostic test. All patients scored as likely to have DVT underwent lower- limb Doppler compression venous ultrasound (CVU); only if this initial test was negative and DD levels were elevated was a second CVU performed a week later (Keeling et al, 2004). A 3-month follow-up period was set up for patients discharged without DVT confirmation. Parameters evaluated and age group distribution were similar to those reported by Schutgens et al (2005) in order to allow a more reliable comparison between the data obtained from both studies. EPIDAT (version 3.0) and STATGRAPHIC PLUS (version 2.1) software support was used for statistical analysis. The most relevant parameters that were evaluated are summarised in Table I. Our study population was older and Table I. Parameters evaluated and related to diagnostic performance of DD in different age groups. Age (years) n Prevalence of DVT Sensitivity Specificity NPV Unlikely + DD normal Elevated DD 18–47 64 19 100 (75Æ7–100) 56Æ9 (43Æ3–69Æ5) 100 (88Æ3–100) 45Æ3 (29/64) 53Æ1 (34/64) 48–60 68 14Æ9 100 (72Æ2–100) 35Æ1 (24–48Æ1) 100 (83Æ9–100) 30Æ9 (21/68) 69Æ1 (47/68) 61–74 173 24Æ6 100 (91Æ6–100) 30Æ2 (23–38Æ6) 100 (91–100) 22Æ1 (38/172) 76Æ7 (132/172) ‡75 215 35Æ5 100 (95Æ2–100) 16Æ7 (11Æ4–23Æ8) 100 (85Æ7–100) 9Æ8 (21/215) 89Æ3 (192/215) Total 520 27Æ1 100 (75Æ7–100) 29Æ5 (25Æ1–34Æ3) 100 (96Æ7–100) 21 (109/520) 78Æ1 (406/520) Values expressed as a percentage with 95% confidence interval given in parentheses. DVT, deep venous thrombosis; NPV, negative predictive value; DD, D-dimer. Correspondence ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 795–805 803