Research Letters 270 www.thelancet.com Vol 364 July 17, 2004 Fetal death in the mid-trimester of pregnancy is a clinical challenge because the cause is usually unknown and no biochemical or biophysical prediction methods. 1 S100B is an acidic calcium-binding protein that is highly concentrated in nervous tissue; it has a half-life of about 1 h and is mainly eliminated by the kidneys. 2 The fetal nervous system is the source of a large proportion of S100B present in amniotic fluid, 2–4 and detection of high concentrations in amniotic fluid surrounding anencephalic fetuses has been deemed an indicator of fetal CNS damage. 4 Raised serum and urine S100B concentrations are a consolidated marker of brain damage in adults, 2 paediatric patients, 3 and during the perinatal period. 3,5 We aimed to assess whether S100B concentrations in amniotic fluid at the time of amniocentesis represent a useful method to identify women at risk of having intrauterine fetal death. We also measured amounts of  fetoprotein, a glycoprotein produced by the yolk sac and fetal gastrointestinal tract, which is used in screening for neural-tube defects. 6 We did a case-control study with amniotic fluid that we obtained from women undergoing amniocentesis for genetic indications, who we recruited consecutively from Jan 1, 1999, to May 31, 2002, at our tertiary referral centres for obstetric care. Of these women, cases were those who had a spontaneous fetal death in the mid-trimester. We estimated gestational age from the last menstrual period and ultrasonographic fetal biometry at amniocentesis. We judged fetal growth to be appropriate by a normal ultrasound biometry 7 and postnatal confirmation. All cases had a normal karyotype, and fetuses had no detectable anomalies. Controls—matched for gestational age and parity— were from the same period as the cases and were those who delivered a healthy infant at 37 weeks or more, of appropriate weight for gestational age, and without complications. The control group included multiple pregnancies; women who subsequently developed intrauterine growth retardation or who had pregnancy- induced hypertension or pre-eclampsia; those with diabetes; fetal malformations (ventricular malformation and cystic hygroma); chromosomal abnormalities (trisomy 21 and trisomy 18); and women who were exposed to alcohol or tobacco smoke. We obtained informed consent from all women before inclusion in the study and approval from our local human investigation committee. Fetal death in the mid-trimester of pregnancy is unexplained and no reliable markers are available to identify at- risk women. We aimed to assess use of  fetoprotein and S100B concentrations in amniotic fluid as markers. We did a case-control study in 758 healthy women undergoing amniocentesis at mid-gestation, of whom 12 had a spontaneous intrauterine fetal death before 28 weeks, and 746 matched controls. Concentrations were corrected for gestational age by conversion to multiples of median (MoM) of healthy controls of the same gestational age. Concentrations of S100B, but not  fetoprotein, were significantly higher (p<0·0001) in women who later had spontaneous fetal death (median 4·431 MoM [95%CI 3·605–6·197]) than in controls (1·000 MoM [1·062–1·121]). Sensitivity, specificity, and positive and negative predictive values of S100B as a diagnostic test were 100%, suggesting that measurement of this protein at amniocentesis could be useful to identify at-risk women. Amniotic fluid S100B protein in mid-gestation and intrauterine fetal death Pasquale Florio, Fabrizio Michetti, Matteo Bruschettini, Mario Lituania, Pierluigi Bruschettini, Filiberto M Severi, Felice Petraglia, Diego Gazzolo Lancet 2004; 364: 270–72 Department of Paediatrics, Obstetrics, and Reproductive Medicine, University of Siena, Policlinico “Le Scotte”, viale Bracci, 53100 Siena, Italy (P Florio MD, F M Severi MD, F Petraglia MD); Institute of Anatomy and Cell Biology, Catholic University, Rome, Italy (F Michetti MD); and Department of Paediatrics (M Bruschettini MD, P Bruschettini MD, D Gazzolo MD) and Obstetrics and Gynaecology (M Lituani MD), G Gaslini Children’s Hospital, University of Genoa, Italy Correspondence to: Dr Felice Petraglia petraglia@unisi.it Amniotic fluid  fetoprotein (MoM) 0 3·0 0·5 1·0 1·5 2·5 2·0 A Amniotic fluid S100B (MoM) Healthy controls Cases with fetal death 0 12·5 10·0 7·5 5·0 2·5 B Figure:  fetoprotein (A) and S100B (B) concentrations in amniotic fluid Solid lines represent median values. Dotted line represents the cut-off of 2·5 MoM.