Fetal Diaphragmatic Hernia: Echocardiography and Clinical Outcome By Karen J. VanderWall, Thomas Kohl, N. Scott Adzick, Norman H. Silverman, Julien I. Hoffman, and Michael R. Harrison San Francisco, California l It is difficult to predict survival of fetuses diagnosed prenatally with congenital diaphragmatic hernia. Some stud- ies suggest that left heart underdevelopment is associated with poor outcome, but fetal echocardiographic variables have not been conclusively proven to be good predictors of postnatal survival. The authors reviewed detailed fetal echo- cardiographic studies in twelve fetuses with congenital diaphragmatic hernia. Ten echocardiographic variables, in- cluding left and right ventricular width, left ventricular volume, and left ventricular mass, were examined from a four-chamber view, corrected for gestational age, and com- pared with normal data. The results of this study showed no significant differences between survivors and nonsurvivors in the ten variables analyzed. Although left heart dimensions and left ventricular volume in fetuses with congenital dia- phragmatic hernia were below the expected normal range, these results did not predict postnatal outcome. Copyright o 1997 by W.6. Saunders Company INDEX WORDS: Congenital diaphragmatic hernia, fetal echo- cardiography, left heart hypoplasia. A LTHOUGH cogenital diaphragmatichernia (CDH) is now frequently diagnosedbefore birth, predicting outcome after birth hasbeen difficult. Some studieshave suggested that the heart may be abnormal in fetuseswith CDH, and that left heart hypoplasia may predict poor outcome.‘-3 Fetal echocardiography is quite accurate in identifying associatedcardiac abnormalities, but a de- tailed assessment of cardiac structure and function is usually not done. We reviewed detailed echocardio- graphic studies on fetuses diagnosed with CDH to identify variables that might predict outcome after birth. We then compared the echocardiographic variables from the CDH fetuses with echocardiographic variables from normal fetuses. MATERIALS AND METHODS Thirty of the patients with a fetal diagnosis of CDH who were referred 10 the Fetal Treatment Center at the University of California, San Francisco (UCSF) from 1990 to 1995 had detaIled fetal echocarcho- graphic evaluations. We excluded fetuses with other cardiac anomalies and cases mvolvmg elective termmation of pregnancy, in utero fetal demise. or fetal surgery. Of the 30 patients reviewed. twelve fetuses (range. 17 to 2.5 weeks’ gestation) had fetal echocardiographic studies that were adequate for careful measurement and analysis. One examiner (T.K.) who was unaware of clinical outcome reviewed the echocardiograpluc videotaped exams. Echocardiographic variables were examined from a four-chamber view of the heart (Fig 1). We measured the maximum length of the left and right ventricular chamber Journal of Pediatric Surgery, Vol32, No 2 (February). 1997: pp 223-226 during diastole (LV length, RV length), maximum width of the left and right ventricular wall during diastole (LV width. RV width). and left ventricular wall. right ventricular wall and septal thickness during drastole (LV wall, RV wall, VS). The left-to-right ventricular width ratio (measunng the widths of the ventricles m end-diastole) and the left ventricular volume (LV volume) were also calculated. Left ventricular mass (LV mass) was calculated at end-dlastole (LV volume of muscle and cavity - LV volume of the cavity) X l.05].1 Postnatal care took place at UCSF or other appropnate tertiary centers with extracorporeal membrane oxygenation (ECMO) capability. The principal outcome vanable was survival, wluch was defined as achievmg the age of 2 months or &charge from the hospital. StatIstical analysis was performed by the paired Student’s 1 test with P values of less than .05 consldered sigmficant. Regression analysis curves with mean and confidence levels for individual pomts were constructed for each variable and data were compared with normal controls at corresponding gestatlonal ages.5 A one-sided Fisher’s exact test was used to compare the CDH group with the normal controls. RESULTS The overall survival was 50% (6 of 12). All diagnoses of CDH were madeat or before 25 weeks’ gestation.The average gestational age was similar (23 weeks for survivors, 22.8 weeks for nonsurvivors). Of the ten variables evaluated, none predicted survival (Table 1). Although the LV width and LV masswere less in the nonsurvivor group, the differences were not significant (P > .5). When the echocardiographic variables from the CDH group (n = 12) were compared with fetal echocardio- graphic variables from normal controls, the LV width and RV width were significantly less than controls at the same gestational age (one-sided Fisher’s exact test, P = .034 and P = .0226, respectively; Fig 2). The LV length and RV length were not significantly different from controls (data not shown). The LV volume and LV mass were also significantly less than the control group (P = 0.0102 and P = 0.0137. respectively; Fig 3). From the Departments of Surgery Pedratnc Echocardiography, and Pediatrics. Urziversuy of California, San Francisco. CA Presented at tlze 27th Annual Meeting of the Amerrcan Pediatric Surgical Assoczation, San Diego, California. May 20-23. 1996. Address reprznt requests to Karen J. VanderWall. MD, The Fetal Treatment Center: 513 PurnassusAve, Room 1601 HSW San Francisco. CA 93133-0570. Copyright o I997 by W B. Saurzders Cornpuny 0022-3468/97/3202-0015$03.00/0 223