Prediction of acidemia at birth by Doppler assessment of fetal cerebral transverse sinus in pregnancies with placental insufficiency

Transcription

1 Ultrasound Obstet Gynecol 2009; 33: Published online 6 October 2008 in Wiley InterScience ( DOI: /uog.6130 Prediction of acidemia at birth by Doppler assessment of fetal cerebral transverse sinus in pregnancies with placental insufficiency M. M. BARBOSA, F. H. C. CARVALHO, E. ARAUJO JÚNIOR, L. M. M. NARDOZZA, R. M. SANTANA, M. R. TORLONI and A. F. MORON Department of Obstetrics, Sao Paulo Federal University (Unifesp/EPM), Sao Paulo, SP, Brazil KEYWORDS: acidemia; birth; Doppler; placental insufficiency; ultrasound ABSTRACT INTRODUCTION Objectives To evaluate the prediction of acidemia at birth using cerebral transverse sinus (CTS) Doppler velocimetry and to determine the best parameter and cut-off values for its prediction in pregnancies complicated with placental insufficiency. Methods This was a prospective cross-sectional study involving 69 pregnant women (26 40 weeks gestation) with placental insufficiency managed in two Brazilian hospitals. Doppler assessment of the CTS was carried out in the last 24 h before delivery, and the peak ventricular systolic (S-wave) and diastolic (D-wave) velocities as well as the atrial systolic velocity (A-wave) were recorded and the pulsatility index for veins (PIV) was calculated. At birth, arterial and venous umbilical cord blood samples were collected to determine acid base and ph status. A receiver operating characteristics (ROC) curve was constructed for each Doppler parameter with birth acidemia as the dependent variable. Sensitivity, specificity, positive and negative predictive values, accuracy and falsepositive and false-negative rates were calculated for the parameters considered to be good predictors of acidemia. Results The S, D and A peak velocities and the S/A ratio were not good predictors of acidemia at birth. The PIV and the (S A)/S ratio were good predictors of acidemia (area under the ROC curve = (P = 0.009) and (P = 0.009), respectively). The cut-off values were PIV = and (S A)/S = 0.703). Conclusions The PIV and the (S A)/S ratio of the CTS were good predictors of acidemia at birth in this highrisk population with placental insufficiency. Copyright 2008 ISUOG. Published by John Wiley & Sons, Ltd. Fetuses exposed to placental insufficiency experience progressive oxygen and nutritional deficits, which ultimately lead to hypoxia, hypercapnia and acidosis. It can lead to fetal growth restriction, which affects approximately 8% of all pregnancies and is associated with increased perinatal mortality and morbidity 1. Growth-restricted neonates have increased rates of meconium aspiration, hematological and metabolic disorders, cognitive dysfunction and cerebral palsy. Most growth-restricted fetuses weigh < 2500 g at delivery and the long-term effects of low birth weight include increased risks for coronary disease, hypertension and diabetes mellitus in adult life 2. Animal studies 3,4 have shown that fetal hypoxia stimulates chemoreceptors that trigger hemodynamic modifications aimed at protecting areas essential to fetal survival, i.e. the brain, heart and adrenals. This selective vasodilatation, combined with compensatory vasoconstriction of other areas, including kidneys, lungs and intestines, is known as fetal brain sparing or centralization and has been shown to occur in human fetuses also 5. Fetal arterial Doppler studies are useful in identifying the typical phases of vascular redistribution seen in fetal centralization. As uterine and placental functions deteriorate, abnormal blood flow is detected in the umbilical artery (UA), in the descending aorta and in the middle cerebral artery (MCA) 6,7. In the final stages of centralization, end-diastolic ventricular pressure increases, blood ejection during atrial contraction decreases and abnormalities in the fetal venous flow become noticeable on venous Doppler studies 8. The abnormalities detected on venous Doppler studies lead to the clinical recognition of hemodynamic decompensation. Correspondence to: Dr E. Araujo Júnior, Rua Carlos Weber, 950 apto. 113 Visage, Alto da Lapa, São Paulo SP, Brazil ( araujojred@terra.com.br) Accepted: 19 March 2008 Copyright 2008 ISUOG. Published by John Wiley & Sons, Ltd. ORIGINAL PAPER

2 Doppler assessment of cerebral transverse sinus 189 With the reduction in cardiac output, central venous pressure rises, causing an increase in inferior vena cava reverse flow and absent or reversed flow in the ductus venosus (DV) during atrial systole. This progressive deterioration of blood flow in the precordial veins presumably reflects the aggravation of myocardial dysfunction and is believed to be related directly to the severity of acidemia At present, Doppler studies of the DV seem to be one of the best methods with which to monitor severely growthrestricted fetuses 10,11. This evaluation gives important information regarding fetal compromise and helps in deciding the ideal timing for delivery to minimize the perinatal morbidity and mortality of severely growthrestricted fetuses. Carvalho et al. 12 reported the S/A and (S A)/S peak velocity ratios and the pulsatility index for veins (PIV) of the DV to be good predictors of acidemia at birth. The cerebral transverse sinus (CTS) shows a triphasic pulsatile forward flow velocity pattern similar to that described in other fetal venous vessels. The first studies evaluating its blood flow to assess fetal well-being were published at the end of the last decade 13,14. However, none of these studies aimed to determine the Doppler parameters of the CTS best suited for the prediction of acidemia at birth in pregnancies with placental insufficiency. Our aim, therefore, was to evaluate, in pregnancies with placental insufficiency, the prediction of acidemia at birth using various Doppler parameters of the CTS. Specifically, CTS peak velocities (S, D and A), ratios (S/A and (S A)/S) and the PIV were evaluated and their best cut-offs were calculated. METHODS This was a prospective cross-sectional study involving 69 pregnant women with placental insufficiency. Fortyseven of these patients were also included in two previous publications on the prediction of acidemia at birth using DV Doppler 12 and venous arterial Doppler ratios (DV pulsatility index (PI)/MCA-PI) 15. All participants were managed in two Brazilian public teaching hospitals, one located in the city of São Paulo (São Paulo Federal University Hospital) and one in the city of Fortaleza (Maternidade Escola Assis Chateaubriand). All sonographic examinations were performed by one of two operators (F.H.C.C. and M.M.B.) between March 2001 and March The study was approved by the ethics committee of both universities and followed the principles of the Declaration of Helsinki for research involving human subjects. All patients gave written informed consent to participate. The diagnosis of placental insufficiency was based on the presence of UA-PI > 95 th percentile for gestational age according to the criteria proposed by Arduini and Rizzo 16. Fetal centralization was diagnosed when MCA-PI UA-PI 17, or when MCA-PI < 5 th percentile for gestational age 16. Inclusion criteria were: singleton pregnancy with gestational age of at least 26 weeks (determined through reliable last menstrual period or sonography performed before the 20 th week), presence of bilateral uterine artery notches after 26 weeks gestation and absence of structural fetal anomalies. Exclusion criteria were: lack of information regarding delivery, no venous cord blood ph and blood gas analysis, diagnosis of placental abruption before or during delivery, infant having any structural or chromosomal anomalies and administration of general anesthesia during delivery. Neonates with weights below the 10 th percentile for gestational age were classified as small-for-gestational age (SGA) according to Lubchenco et al. 18 According to the standard management protocol of the two participating centers, after the diagnosis of placental insufficiency, all patients were admitted to the hospital until delivery. Only the Doppler evaluations performed during the last 24 h before delivery were included in this study. All sonographic examinations were performed using a Voluson 730 Pro (GE Medical Systems, Zipf, Austria) or a Versa Pro (Siemens, Erlangen, Germany) ultrasound machine with color pulsed Doppler capability and equipped with a 3.5-MHz transducer. For sonography, patients were in semi-fowler position and the fetal head was scanned using two-dimensional ultrasound in an axial transcerebellar plane. Color flow mapping and pulsed Doppler flow velocimetry were performed on the CTS bilaterally, adjacent to the internal edge of the calvarium (Figure 1). Doppler measurements of three consecutive and symmetrical waves were recorded in the absence of fetal breathing and activity and uterine contractions. The average of these three measurements was used for all the study parameters. The insonation angle was < 30 for arteries and always set at 0 for CTS measurements. A 100-Hz filter was used and the sample size was adjusted according to the vessel diameter. The peak ventricular systolic (S-wave) and diastolic (Dwave) velocities as well as the atrial systolic velocity (A-wave) were recorded. The PIV (PIV = S A/timeaveraged maximum velocity) and the S/A and (S A)/S Figure 1 Identification of the cerebral transverse sinus using color Doppler flow mapping in an axial plane near the cerebellum, adjacent to the inner border of the fetal calvarium.

3 190 Barbosa et al. ratios were calculated automatically by the ultrasound machine s software. At delivery, before the first breathing movement, a cm segment of the umbilical cord was doubleclamped and approximately 1 ml arterial and 1 ml venous blood were obtained for ph and blood gas analysis, using G needles and disposable 3 ml plastic insulin syringes (Becton Dickinson, Franklin Lakes, NJ, USA) which had been rinsed out previously with heparin. Blood samples were sent immediately to the laboratory and were analyzed within 5 10 min after collection. An ABL-5 analyzer (Radiometer, Westlake, OH, USA) was used to determine the arterial and venous ph, pco 2 (partial pressure of carbon dioxide) and base excess (BE). Acidemia at birth was defined as UA ph < 7.20 in the absence of uterine contractions and < 7.15 in the presence of contractions. Only cases of metabolic or mixed acidosis were considered: BE < 10 as well as pco 2 > 60 mmhg in the absence of contractions, and BE < 11 as well as pco 2 > 65 mmhg in the presence of contractions 19. The SPSS 10.0 program (SPSS Inc., Chicago, IL, USA) was used for statistical analyses. The independent variables were the CTS velocities for the S-, D- and A-waves, the PIV and the S/A and (S A)/S ratios. The dependent variable was the presence of metabolic or mixed acidosis at birth. The sensitivity and specificity of various Doppler cut-offs were calculated and a receiver operating characteristics (ROC) curve was constructed for each parameter analyzed. If the parameter was considered suitable for the prediction of acidemia at birth (area under the curve (AUC) > 50%, P < 0.05), the best cut-off was selected. This choice was based on the positive and negative predictive values (PPV and NPV), false-positive and false-negative rates, accuracy and positive and negative likelihood ratios (+LR and LR). RESULTS A total of 69 women with placental insufficiency fulfilled the inclusion criteria. Maternal age ranged from 15 to 44 (mean ± SD, 28 ± 7; median, 28) years and gestational age ranged from 26 to 40 (mean ± SD, 32.1 ± 3.8; median, 32) weeks. Delivery was by Cesarean section in 91.3% of the patients and the birth weight ranged from 500 to 2935 (mean ± SD, ± 645.8; median, 1315) g. Fifty-seven (82.6%) of the infants were SGA and 12 (17.4%) were appropriate-for-gestational age. Over one third (39.1%) of the infants had a 1-min Apgar score < 7 and eight (11.6%) had a 5-min score < 7. The ph at birth ranged from 6.76 to 7.41 (mean ± SD, 7.24 ± 0.10) and BE ranged from 23.9 to 0 (mean ± SD, 7.03 ± 5.07). Twenty-one (30.4%) neonates had metabolic or mixed acidemia at birth and their mean gestational age was 29.6 ± 2.8 weeks. None of the 69 infants was diagnosed with respiratory acidosis. Analysis of the ROC curve indicated that the peak velocities (S, D and A) were not good predictors of acidemia at birth in this population. The AUCs for Sensitivity Specificity Figure 2 Receiver operating characteristics curve for the pulsatility index for veins (PIV) of the cerebral transverse sinus in the prediction of acidemia at birth. Area under the curve = (95% CI, ). S, D and A peak velocities were (95% CI, ; P = 0.112), (95% CI, ; P = 0.118) and (95% CI, ; P = 0449), respectively. PIV was a good predictor of acidemia at birth, with an AUC of (95% CI, ; P = 0.009). The best PIV cut-off value was determined through a table with the coordinates of the ROC curve using sensitivity and false-positive rates (1 specificity) for the PIV values observed. A PIV cut-off value of had a sensitivity of 66.7%, specificity of 77.1%, PPV of 56.0%, NPV of 84.1%, false-positive rate of 22.9%, false-negative rate of 33.3%, accuracy of 73.9%, +LR of 2.91 and LR of 0.43 for the prediction of acidemia at birth (Figure 2). The (S A)/S ratio was a good predictor of acidemia at birth, with an AUC of (95% CI, ; P = ) An (S A)/S ratio cut-off value of had a sensitivity of 52.4%, specificity of 79.2%, PPV of 52.4%, NPV of 79.2%, false-positive rate of 20.8%, false-negative rate of 47.6%, accuracy of 71.0%, +LR of 2.51 and LR of 0.60 (Figure 3). The S/A ratio also had an AUC > 50% but it was not statistically significant (P = 0.738). DISCUSSION In cases of severe placental insufficiency, venous Doppler velocimetry allows a precise diagnosis of fetal acidemia, which is especially important in the reduction of perinatal mortality and morbidity in very preterm pregnancies. Therefore, fetal venous Doppler studies are useful in monitoring the severity of circulatory alterations as well as in improving perinatal outcome 20.

4 Doppler assessment of cerebral transverse sinus 191 Sensitivity Specificity Figure 3 Receiver operating characteristics curve for the (S A)/S ratio of the cerebral transverse sinus in the prediction of acidemia at birth. Area under the curve = (95% CI, ). Analysis of umbilical cord blood ph and acid base status is considered the gold standard for evaluating the biochemical status of fetuses and neonates. Evaluation of the acid base status at delivery through umbilical cord blood sampling has been shown to be superior to the Apgar score in the detection of intrapartum asphyxia and it should be part of routine obstetric care 21,22.Inthis study, the cut-off used to define acidemia (UA ph < 7.20) was selected to include all fetuses with acidemia, rather than only the more severe cases (ph < 7.0) which would have the highest incidence of neurological morbidity and neonatal mortality 23,24. The DV has been investigated extensively in search of Doppler parameters to predict adverse perinatal outcome 9,25. Baschat et al. 20 evaluated 224 growthrestricted fetuses and reported an association between neonatal mortality and absent or reversed A-wave in the DV and the presence of pulsations in the umbilical vein. Although some studies have tried to use DV Doppler velocimetry to predict acidemia at birth, no consensus has been reached as to which parameter should be used as an ideal cut-off point. Carvalho et al. 12 evaluated 47 pregnancies with placental insufficiency using DV Doppler velocimetry and reported that the S/A and (S A)/S ratios and the PIV were the best parameters with which to predict acidemia at birth, reporting the best cut-offs for its prediction to be 2.67 for S/A, 0.63 for (S A)/S and 0.76 for PIV. Hofstaetter et al. 26 also found the (S A)/S ratio and the PIV to be good predictors of birth acidemia, with similar cut-off values of 0.70 for (S A)/S and 0.90 for PIV. The CTS runs along the tentorium cerebelli, adjacent to the occipital bone, and receives all the venous drainage from the external surface of the brain. Its triphasic waveform reflects the ventricular systole and diastole and the atrial contraction of the cardiac cycle. The first publication on CTS Doppler was a cross-sectional study that analyzed 126 pregnancies between 20 and 42 weeks of gestation and established the normal values for the peak velocities and resistance index of this vessel 13. The authors reported that the peak velocities increased throughout pregnancy, while the resistance index and PI increased between 20 and 28 weeks and then decreased until term. Senat et al. 14 conducted a prospective longitudinal study on 75 growth-restricted fetuses (at weeks) and compared CTS and DV Doppler velocimetry findings with the short-term variation in computerized cardiotocography. They observed a strong association between the PIV of the CTS and that of the DV while the PIV was negatively correlated with the computerized cardiotocography short-term variation. There have been no previous publications on the prediction of acidemia at birth using CTS Doppler velocimetry. Similar to the DV findings of Carvalho et al. 12, we found for the CTS that the peak velocities of the S-, D- and A-waves were not good predictors of birth acidemia. This could be because all CTS peak velocities are lower compared with the DV peak velocities at the same gestational age. However, Hofstaetter et al. 26 reported that an A-wave velocity 17 cm/s was capable of predicting birth acidemia with a sensitivity of 79% and a specificity of 68%. Using as the cut-off value, we found the (S A)/SratiooftheCTStobea good predictor of acidemia at birth. This cut-off is quite similar to that proposed for the DV in previous studies (0.63 and 0.70) 12,26. Surprisingly, the parameter used most frequently for DV velocimetry 27, the S/A ratio, was not a good predictor of acidemia at birth. The A-wave peak velocity in the CTS was lower than that in the DV and, although there was an increase of this velocity with advancing gestational age, it was less evident than in the DV, compared with in the study of Carvalho et al. 12.This may in part explain why the S/A ratio for the CTS was not statistically significant 12,20,26. Our findings using angle-independent indices were encouraging. Using a cut-off of 0.855, the PIV was a good predictor of birth acidemia, with a sensitivity of 66.7%, a specificity of 77.1%, a PPV of 56%, a NPV of 84.1% and an AUC of The high specificity and especially the NPV of 84.1% are clinically important when deciding the best moment to deliver extremely premature fetuses. A CTS-PIV below this cut-off indicates a low risk for acidemia at birth. When determining the cut-off from the ROC curve, we placed more value on specificity than on sensitivity, so that a negative test result would assure us that the fetus was not at risk for acidemia and a positive one would indicate a high probability of the fetus being acidemic, with a low false-positive rate. The accuracies of 73.9% (PIV) and 71.0% ((S A)/S) indicate a high rate of agreement between this test and the gold standard of cord blood ph, even in this high-risk population of 69 patients. The +LR of 2.91 means that fetuses with CTS Doppler values above the proposed cut-offs were almost three times more

5 192 Barbosa et al. likely to be acidemic at birth than were fetuses with measurements below these cut-offs. Similarly, the LR of 0.43 means that fetuses with CTS Doppler values below these cut-offs were over twice as likely to be non-acidemic at birth than were those with values above these cut-offs. However, these results and the proposed CTS Doppler cut-offs should be interpreted with caution due to the small sample size of this study. These findings need to be confirmed in future studies involving more patients with placental insufficiency. Only then can we evaluate the role of CTS Doppler velocimetry in the management of pregnancies with placental insufficiency. In conclusion, there is currently no consensus on the best venous Doppler parameter to use in the prediction of birth acidemia. We found the (S A)/S and PIV of the CTS to be good predictors of acidemia at birth and believe that the inclusion of CTS Doppler in the management of very preterm (< 32 weeks) pregnancies with severe placental insufficiency could help in the decision as to when these fetuses should be delivered. ACKNOWLEDGMENT We would like to thank the CAPES for financial support. REFERENCES 1. Regnault TR, Galan HL, Parker TA, Anthony RV. Placental development in normal and compromised pregnancies a review. Placenta 2002; 23 (Suppl A): S119 S Barker DJ, Gluckman PD, Godfrey KM. Fetal nutrition and cardiovascular disease in adult life. Lancet 1993; 341: Kjellmer I, Karlsson K, Olsson T, Rosen KG. Cerebral reactions during intrauterine asphyxia in the sheep. I. Circulation and oxygen consumption in the fetal brain. Ped Res 1974; 8: Dawes GS, Duncan SL, Lewis BV, Merlet CL, Owen- Thomas JB, Reeves JT. Cyanide stimulation of the systemic arterial chemoreceptors in foetal lambs. J Physiol 1969; 201: Wladimiroff JW, Wijngaard JA, Degani S, Noordam MJ, van Eyck J, Tonge HM. Cerebral and umbilical arterial blood flow velocity waveforms in normal and growth-retarded pregnancies. Obstet Gynecol 1987; 69: Harrington K, Carpenter RG, Nguyen M, Campbell S. Changes observed in Doppler studies of the fetal circulation in pregnancies complicated by pre-eclampsia or the delivery of a small-for-gestational-age baby. I. Cross-sectional analysis. Ultrasound Obstet Gynecol 1995; 6: Soothill PW, Ajayi RA, Campbell S, Nicolaides KH. Prediction of morbidity in small and normally grown fetuses by fetal heart rate variability, biophysical profile score and umbilical artery Doppler studies. Br J Obstet Gynaecol 1993; 100: Hecher K, Campbell S, Doyle P, Harrington K, Nicolaides K. Assessment of fetal compromise by Doppler ultrasound investigation of fetal circulation. Arterial, intracardiac, and venous blood flow velocity studies. Circulation 1995; 91: Baschat AA, Gembruch U, Reiss I, Gortner L, Weiner CP, Harman CR. Relationship between arterial and venous Doppler and perinatal outcome in fetal growth restriction. Ultrasound Obstet Gynecol 2000; 16: Ferrazzi E, Bozzo M, Rigano S, Bellotti M, Morabito A, Pardi G, Battaglia FC, Galan HL. Temporal sequence of abnormal Doppler changes in the peripheral and central circulatory systems of the severely growth-restricted fetus. Ultrasound Obstet Gynecol 2002; 19: Romero R, Kalache KD, Kadar N. Timing the delivery of the preterm severely growth-restricted fetus: venous Doppler, cardiotocography or biophysical profile? Ultrasound Obstet Gynecol 2002; 19: Carvalho FHC, Moron AF, Mattar R, Santana RM, Murta CGV, Barbosa MM, Torloni MR, Vasques FAP. Ductus venosus Doppler velocimetry in the prediction of acidemia at birth: which is the best parameter? Prenat Diagn 2005; 25: Laurichesse-Delmas H, Grimaud O, Moscoso G, Ville Y. Color Doppler study of the venous circulation in the fetal brain and hemodynamic study of the cerebral transverse sinus. Ultrasound Obstet Gynecol 1999; 13: Senat MV, Schwarzler P, Alcais A, Ville Y. Longitudinal changes in the ductus venosus, cerebral transverse sinus and cardiotocogram in fetal growth restriction. Ultrasound Obstet Gynecol 2000; 16: Carvalho FHC, Moron AF, Mattar R, Murta CGV, Santana RM, Barbosa MM, Torloni MR, Vasques FAP. Venous-arterial Doppler ratios in the prediction of acidemia at birth in pregnancies with placental insufficiency. Fetal Diagn Ther 2006; 21: Arduini R, Rizzo G. Normal values of pulsatility index from fetal vessels: a cross sectional study on healthy fetuses. JPerinatMed1990; 18: Arbeille P, Roncin A, Berson M, Patat F, Pourcelot L. Exploration of the fetal cerebral blood flow by Doppler ultrasound in normal and pathological pregnancies. Ultrasound Med Biol 1987; 13: Lubchenco LO, Hansman C, Dressler M, Boyd E. Intrauterine growth as estimated from liveborn birth-weight data at 24 to 42 weeks of gestation. Pediatrics 1963; 32: Vintzileos AM, Egan JFX, Campbell WA, Rodis JF, Scorza WE, Fleming AD, McLean DA. Asphyxia at birth as determined by cord blood ph measurements in preterm and term gestations: correlation with neonatal outcome. J Matern Fetal Med 1992; 1: Baschat AA, Gembruch U, Weiner CP, Harman CR. Qualitative venous Doppler waveform analysis improves prediction of critical perinatal outcomes in premature growth-restricted fetuses. Ultrasound Obstet Gynecol 2003; 22: Thorp JA, Sampson JE, Parisi VM, Creasy RK. Routine umbilical cord blood gas determinations? Am J Obstet Gynecol 1989; 161: Johnson JWC, Richards DS, Wagaman RA. The case for routine umbilical cord acid-base studies at delivery. Am J Obstet Gynecol 1990; 162: Gilstrap LC 3 rd, Leveno KJ, Burris J, Williams ML, Little BB. Diagnosis of birth asphyxia on the basis of fetal ph, Apgar score, and newborn cerebral dysfunction. Am J Obstet Gynecol 1989; 161: Goldaber KG, Gilstrap LC 3 rd, Leveno KJ, Dax JS, McIntire DD. Pathologic fetal acidemia. Obstet Gynecol 1991; 78: Figueras F, Martínez JM, Puerto B, Coll O, Cararach V, Vanrell JA. Contraction stress test versus ductus venosus Doppler evaluation for the prediction of adverse perinatal outcome in growth-restricted fetuses with non-reassuring non-stress test. Ultrasound Obstet Gynecol 2003; 21: Hofstaetter C, Gudmundsson S, Hansmann M. Venous Doppler velocimetry in the surveillance of severely compromised fetuses. Ultrasound Obstet Gynecol 2002; 20: Harman CR, Baschat AA. Arterial and venous Doppler in IUGR. Clin Obstet Gynecol 2003; 46: