TRANSFUSION PRACTICE Cerebral Doppler velocimetry to predict fetal anemia after more than three intravenous fetal exchange transfusions Monika Hermann,1,2 Marie-Hélène Poissonnier,1 and Gilles Grangé1,2
BACKGROUND: We aimed to assess usefulness of the middle cerebral artery peak systolic velocity (MCA-PSV) in the prediction of fetal anemia after more than three intravenous fetal-exchange transfusions (IFET). STUDY DESIGN AND METHODS: A retrospective study was conducted over 6 years of 15 consecutive pregnancies with severe red blood cell fetomaternal alloimmunization requiring more than three IFETs. We evaluated correlation between MCA-PSV (expressed as multiples of the mean [MoM]) and pretransfusion hemoglobin (Hb) in the fetus (MoM). Analyses were also performed to assess the value of MCA-PSV to predict moderate to severe fetal anemia. RESULTS: Twenty-seven MCA-PSV measurements performed before the fourth to last IFET were coupled with pretransfusion Hb in the fetus. The median number of IFETs per fetus was five (range, four to eight). Five Hb samples found fetuses with severe (19%), seven with moderate (26%), and 15 with mild anemia (56%). There was a linear correlation between MCA-PSV(x) and Hb in the fetus(y): y = −0.21x + 0.93 (r = −0.50, p < 0.01). For the prediction of moderate to severe anemia the negative predictive value of MCA-PSV with a threshold of 1.5 MoM was 75%, positive predictive value 73%, specificity 80%, sensibility 67%, and positive likelihood ratio 3.33. The area under the receiver operating characteristic curve was 0.78 (95% confidence interval, 0.59-0.96; p < 0.001). For the prediction of severe anemia, MCA-PSV with a threshold of 1.5 MoM had 94% negative predictive value, 80% sensibility, and a positive likelihood ratio of 2.5. CONCLUSIONS: This study shows that a correlation between MCA-PSV and Hb in the fetus persists even after more than three IFETs. MCA-PSV measurements thus remain useful to monitor fetuses at risk of anemia.
2968
TRANSFUSION Volume 54, November 2014
V
ital statistics data indicate that there are approximately 800,000 pregnancies per year in France1,2 and probably 1,100,000 to 1,200,000 conceptions. Approximately 15% of these women are D–. Despite the prophylactic measures involving RhIG administration recommended by the French National College of Gynecology and Obstetrics in 2005,3,4 maternal red blood cell (RBC) alloimmunization remains the main cause of fetal anemia in France. The risk of hemolytic disease due to alloimmune antibodies is estimated at one per 28,000 fetuses, with 28 in utero or perinatal deaths1 each year in France. In utero surveillance of fetuses at risk thus remains essential. All RBC antibodies can be involved, but anti-D (Rhesus), anti-Kell, and anti-c are found most frequently. Before 2000, techniques to detect fetal anemia were either invasive (cordocentesis, amniocentesis), unreliable (antibody assays level in maternal blood, fetal cardiac rhythm), or late (sinusoidal fetal cardiac rhythm, signs of hydrops on ultrasound examination). Mari and coworkers demonstrated in 20005,6 that moderate to severe anemia can be detected noninvasively by Doppler ultrasonography from an increase in the peak velocity of systolic blood flow in the middle cerebral artery (MCAPSV) to more than 1.5 multiples of the mean (MoM) in
ABBREVIATIONS: IFET(s) = intravenous fetal-exchange transfusion(s); MCA-PSV = middle cerebral artery peak systolic velocity; MoM = multiples of the mean; ROC = receiver operating characteristic. From the 1Department of Obstetrics and Gynecology, Port-Royal Maternity, Groupe Hospitalier Cochin-Broca-Hôtel Dieu, Assistance Publique-Hopitaux de Paris; and 2Paris V René Descartes University, Paris, France. Address reprint requests to: Monika Hermann, Maternité de Port Royal, 53 Boulevard de l’Observatoire, 75014 Paris, France; e-mail: [email protected]. Received for publication February 13, 2014; revision received April 1, 2014, and accepted April 2, 2014. doi: 10.1111/trf.12714 © 2014 AABB TRANSFUSION 2014;54:2968-2973.
CEREBRAL DOPPLER AFTER FETAL TRANSFUSION
fetuses at risk of maternal RBC alloimmunization. Other studies7-13 over the past decade have confirmed these results: there is a strong correlation between the Doppler measurement of MCA-PSV and fetal anemia: R2 = 0.55 in the study by Mari and colleagues6 and 0.654 in the one by Carbonne and colleagues8 (p < 0.001). The threshold for intervention in Mari’s study was 1.5 MoM. Fetal anemia can be treated with intravenous (IV) fetal exchange transfusion (IFET). MCA-PSV is useful for timing the first,6 second14 (R2 = 0.51 for 1.69 MoM), and third15 (R2 = 0.43 for 1.50 MoM) IFETs, but after more than two IFETs, adult RBCs totally replace the fetal cells.16 Yet adult RBCs have different physical characteristics: they are smaller and less rigid and have less oxygen-binding capacity, whereas fetal RBCs carry more oxygen but release less to tissue. Fetal surveillance has not been tested for more than three IFETs and it is uncertain whether known curves to predict fetal anemia are still valid because of rheologic changes in fetal blood after IV transfusion with adult blood. The purpose of this study was to assess the value of MCA-PSV in predicting fetal anemia in fetuses that have undergone more than three IFETs.
MATERIALS AND METHODS This retrospective study was conducted at Saint-Vincent de Paul Hospital in Paris, France. We reviewed all records of pregnancies with IFET from January 1, 2004, through December 31, 2009 (n = 57). We excluded fetuses that had IFET for causes other than RBC alloimmunization (n = 12) and those with less than three IFETs (n = 20). Management of pregnancies at risk of RBC alloimmunization was done as described in Fig. 1 according to management described in literature.17,18 Antibody screening was performed (by indirect antiglobulin test) at the first prenatal visit, and any antibodies identified were quantified. If the maternal antibody was anti-D, fetal Rhesus genotyping was performed on maternal serum. D– fetuses did not require any particular further follow-up, while serum antibody level was monitored every 2 weeks for D+ fetuses. When this level reached 1 μg/L or more than 1/16, a weekly ultrasound examination included a Doppler measurement of the MCA-PSV. The sonographer also sought other signs of fetal anemia, such as hepatomegaly and hydrops. The technique to measure MCA-PSV was the one described by Mari and coworkers.6,19,20 Two trained physician-sonographers (MHP and ML) performed all ultrasound evaluations. IFET was scheduled when MCAPSV increased to higher than 1.5 MoM. The IFET procedure was done in an operating room, after administering local anesthesia to the mother and monitoring the fetal cardiac rhythm. The method was the one previously described by Poissonnier and coworkers.21 Cordocentesis was performed by ultrasonographically guided puncture
of the umbilical vein with an 18-gauge needle: for the first IFET, Rhesus incompatibility was first confirmed by blood group and Rhesus genotyping. At the beginning of each IFET, the severity of anemia was evaluated by fetal umbilical vein blood sampling with a point-of-care hemoglobin (Hb) test (Hemocue, Angelholm, Sweden), confirmed by laboratory results. When the Hb concentration in the fetus was less than 0.85 MoM, IFET was performed immediately after fetal anesthesia with vecuronium bromide (0.1 mg/kg), transfusing irradiated group O– compatible cytomegalovirus-negative packed RBCs. The KleihauerBetke test on the umbilical vein was used to detect residual fetal RBCs at the end of each IFET. IFET was performed every 1 to 4 weeks up to a gestational age of 34 weeks, when MCA-PSV exceeded 1.5 MoM or when it increased rapidly. After 34 weeks of gestation, labor was induced when MCA-PSV exceeded 1.5 MoM. If measurements remained approximately or less than 1 MoM, birth was scheduled at 36 weeks with induction of labor or cesarean delivery if vaginal birth was not possible. At birth, the neonatal blood was sampled to determine Hb at birth and detect irregular RBC antibodies. The main outcome measure was the correlation between MCA-PSV before IFET and Hb in the fetus at cordocentesis at the beginning of each procedure. We assessed MCA-PSV with a threshold of 1.5 diagnostic accuracy to predict fetal anemia. Values of Hb in the fetus and MCA-PSV are expressed as MoM to adjust for gestational age at the time of measurement according to the classification of Mari and colleagues:6 mild anemia corresponds to Hb concentrations in the fetus from 0.84 to 0.65 MoM, moderate anemia to Hb concentrations from 0.65 to 0.55 MoM, and severe anemia to Hb concentrations below 0.55 MoM. Statistical analyses were performed with computer software (XL-STAT Life, Addinsoft SARL, Paris, France; and Stata 12.1, StataCorp, College Station, TX). Correlations were calculated with Pearson’s test. Results were considered significant when p values were less than 0.05. Results are expressed as means (with their 95% confidence intervals [CIs]). Regression analysis was used to assess the correlation between MCA-PSV and Hb in the fetus.
RESULTS Over the 6-year study period, there were 13,758 births in Saint Vincent de Paul Hospital and 240 pregnancies monitored for risk of fetal anemia. At least one IFET was required for 57 of them, and 45 of these involved maternal RBC immunization. Of these pregnancies, 19 required more than two IFETs and 15 required more than three. Of these 15 fetuses, 13 were live born, while two died in utero. Both had persistent or recurrent hydrops, even after IFET, and the pathology examinations found it to be the only cause of death. Volume 54, November 2014 TRANSFUSION
2969
HERMANN ET AL.
Fig. 1. Management of D– women and fetuses at risk of anemia. GA = gestational age.
2970
TRANSFUSION Volume 54, November 2014
CEREBRAL DOPPLER AFTER FETAL TRANSFUSION
The antibody responsible for maternal alloimmunization was anti-D in 14 pregnancies and anti-Kell in one. MCA-PSV is as reliable for anti-Kell immunization as for anti-D immunization;19 therefore, management was the same for both. Anti-D were associated with other antibodies in 11 pregnancies: six anti-D + C, two anti-C + D + E, one anti-D + E, one anti-D + C + a, and one antiD + C + Kell. The 15 fetuses underwent a median of five IFETs (range, four to eight). The mean time between the last TABLE 1. Characteristics of transfused fetuses Data Number of IFETs per fetus Hb in the fetus before IFET MCA-PSV before IFET Gestational age at delivery
Mean (±SD) 5.6 (±1.4) 0.64 MoM (±0.12) 1.38 MoM (±0.29) 34+6 weeks (±2.4 days)
TABLE 2. Characteristics for MCA-PSV above 1.5 MoM after the third IFET
Characteristic Negative predictive value Positive predictive value Specificity Sensibility Positive likelihood ratio Negative likelihood ratio
Prediction of moderate to severe anemia* 75% 73% 80% 67% 3.33 0.42
* Hb in the fetus below 0.65 MoM. † Hb in the fetus below 0.55 MoM.
Prediction of severe anemia† 94% 36% 68% 80% 2.51 0.29
IFET and delivery was 21.5 days (95% CI, 10.3-32.7 days). The mean gestational age at birth was 34 + 6 weeks (95% CI, 32 + 3-37 + 1). The mean of the last measured MCA-PSV values before birth was 0.9 MoM (95% CI, 0.7-1.2 MoM). Overall, these fetuses underwent 87 cordocentesis procedures for suspected fetal anemia. Only one cordocentesis procedure was not followed by IFET; it was delayed for a few days because the ultrasound fetal cardiac rhythm showed poor fetal tolerance. All cordocentesis confirmed fetal anemia with Hb in the fetus below 0.84 MoM. Three fetuses, including the two who died in utero, had hydrops. The hydropic fetus who survived had a sinusoidal fetal cardiac rhythm that indicated the need for emergency cesarean delivery. In all, 39 IFETs were performed after the third IFET in these 15 fetuses. We recorded 27 pairs of pre-IFET MCAPSV and pretransfusion Hb measurements (some records were incomplete). The mean MCA-PSV before IFET was 1.38 MoM (95% CI, 1.09-1.66 MoM), and 11 measurements exceeded 1.5 MoM. Five pretransfusional blood samples (19%) concluded to severe anemia, seven to moderate (26%), and 15 (56%) to mild anemia. All Kleihauer-Betke tests were less than 0.01% after the second IFET. The mean Hb in the fetus before IFET was 0.64 MoM (±0.12; see Table 1). Of the 13 live-born babies, two had moderate anemia at birth, and three had mild anemia; the other eight were all born nonanemic (Hb > 0.84 MoM). Pretransfusion Hb (y) is correlated with its corresponding MCA-PSV(x), with r = −0.50, p < 0.01 (see Fig. 2). The relation is linear, expressed by y = −0.21x + 0.93. Table 2 respectively shows capacity of MCA-PSV with a threshold of 1.5 MoM to predict moderate to severe fetal anemia and severe anemia. The area under the receiver operating characteristic (ROC) curve to predict moderate to severe anemia is 0.78 (95% CI, 0.59-0.96; p < 0.001; see Fig. 3). To predict severe anemia, the area under the ROC curve is 0.65 (95% CI, 0.35-0.95).
DISCUSSION
Fig. 2. Linear regression of Hb in the fetus function of MCA-PSV after more than three IFETs. R2 = 0.254; p < 0.01.
Our study shows that MCA-PSV continues to be strongly negatively correlated with Hb in the fetus after more than three IFETs (r = −0.50, p < 0.01): the higher the MCA-PSV, the lower the expected Hb in the fetus. This is, to our knowledge, the first study to have assessed accuracy of MCA-PSV after more than three IFETs. Our findings are consistent with previous results in studies of anemic fetuses that had no, one, or two Volume 54, November 2014 TRANSFUSION
2971
HERMANN ET AL.
PSV measurements are found to be more than 1.5 MoM during fetal follow-up in between scheduled IFET. No fetuses followed by cerebral Doppler were born anemic, but no study has yet determined the gestational age until which this follow-up is reliable. Future research also needs to define the best term at which these fetuses should be delivered. In conclusion, our data show that the correlation between MCA-PSV and Hb in the fetus persists even after more than three IV fetal transfusions. Cerebral Doppler thus remains a useful tool for monitoring fetuses at risk of anemia undergoing serial IFET because of maternal alloimmunization.
Fig. 3. ROC curve to predict fetal moderate to severe anemia with MCA-PSV. Area under the curve = 0.78 (95% CI, 0.59-0.96).
IFETs.6,8,14,15 The diagnostic performance we found is, however, lower than that found by previous authors.6-10,12,13 This difference might have several explanations. Indeed, after more than three IFETs fetal blood has totally been replaced by adult blood with different physical properties; MCA-PSV measurements might thus be less reliable. Further, the study involved a small number of fetuses. A measurement bias could also affect our results: in some records, the reported MCA-PSV values available were measured 1 to 4 days before IFET; the real pretransfusion values were probably higher. Some IFETs were indeed scheduled in advance for practical reasons: only 45% of third or later IFETs were performed immediately after an MCA-PSV measurement above 1.5 MoM; the other serial IFETs were scheduled because of a rapid increase in the Doppler measurements or scheduled approximately 15 days after the previous IFET to take RBC hemolytic clearance into account. Such a bias might have been avoided with a prospective study that performed IFET only when MCA-PSV values reached 1.5 MoM. The correlation we found does not, however, prove that MCA-PSV is useful to schedule the timing of IFET beyond the second one. A prospective study of a larger group is needed for a more precise determination of the diagnostic accuracy of MCA-PSV for predicting fetal anemia after serial IFET. Awaiting these data, we suggest that MCA-PSV should still be used to monitor fetuses after serial IFET—a weekly follow-up should be maintained until 34 to 36 weeks. Although IFETs are often scheduled in advance taking into account RBC hemolytic clearance in fetuses transfused for severe alloimmunization, our result suggests that an IFET should be performed earlier if MCA2972
TRANSFUSION Volume 54, November 2014
ACKNOWLEDGMENT We thank Dr A. Cortey, head of the CNRHP
unit (Armand Trousseau Hospital, Paris, APHP), for preparing the blood to transfuse for the fetuses included in the study and analyzing all the blood samples.
CONFLICTS OF INTEREST The authors have disclosed no conflicts of interest.
REFERENCES 1. Branger B, Winer N. [Epidemiology of anti-D alloimmunization during pregnancy]. J Gynecol Obstet Biol Reprod (Paris) 2006;35(1 Suppl):1S87-S92. 2. INSEE. Bilan demographique. January 2009;1220. 3. Cortey A, Brossard Y. [Prevention of fetomaternal rhesus-D allo-immunization. Practical aspects]. J Gynecol Obstet Biol Reprod (Paris) 2006;35(1 Suppl):1S123-1S30. 4. CNGOF. Text of the guidelines for prevention of fetomaternal rhesus-D allo-immunization. J Gynecol Obstet Biol Reprod (Paris) 2006;35(1 Suppl):1S131-1S35. 5. Mari G, Adrignolo A, Abuhamad AZ, et al. Diagnosis of fetal anemia with Doppler ultrasound in the pregnancy complicated by maternal blood group immunization. Ultrasound Obstet Gynecol 1995;5:400-5. 6. Mari G, Deter RL, Carpenter RL, et al. Noninvasive diagnosis by Doppler ultrasonography of fetal anemia due to maternal red-cell alloimmunization. Collaborative Group for Doppler Assessment of the Blood Velocity in Anemic Fetuses. N Engl J Med 2000;342:9-14. 7. Abdel-Fattah SA, Soothill PW, Carroll SG, et al. Middle cerebral artery Doppler for the prediction of fetal anaemia in cases without hydrops: a practical approach. Br J Radiol 2002;75:726-30.
CEREBRAL DOPPLER AFTER FETAL TRANSFUSION
8. Carbonne B, Castaigne-Meary V, Cynober E, et al. [Use of peak systolic velocity of the middle cerebral artery in the management of fetal anemia due to fetomaternal erythrocyte alloimmunization]. J Gynecol Obstet Biol Reprod (Paris) 2008;37:163-9. 9. Delle Chiaie L, Buck G, Grab D, et al. Prediction of fetal anemia with Doppler measurement of the middle cerebral
14. Detti L, Oz U, Guney I, et al. Doppler ultrasound velocimetry for timing the second intrauterine transfusion in fetuses with anemia from red cell alloimmunization. Am J Obstet Gynecol 2001;185:1048-51. 15. Mari G, Zimmermann R, Moise KJ Jr., et al. Correlation between middle cerebral artery peak systolic velocity and fetal hemoglobin after 2 previous intrauterine transfusions.
artery peak systolic velocity in pregnancies complicated by
Am J Obstet Gynecol 2005;193(3 Pt 2):1117-20.
maternal blood group alloimmunization or parvovirus B19 infection. Ultrasound Obstet Gynecol 2001;18:232-6.
16. Welch R, Rampling MW, Anwar A, et al. Changes in hemorheology with fetal intravascular transfusion.
10. Kurmanavicius J, Streicher A, Wright EM, et al. Reference values of fetal peak systolic blood flow velocity in the middle cerebral artery at 19-40 weeks of gestation. Ultrasound Obstet Gynecol 2001;17:50-3. 11. Scheier M, Hernandez-Andrade E, Fonseca EB, et al. Prediction of severe fetal anemia in red blood cell alloimmunization after previous intrauterine transfusions. Am J Obstet Gynecol 2006;195:1550-6.
Am J Obstet Gynecol 1994;170:726-32. 17. Carbonne B, Castaigne V, Cynober E, et al. [Follow-up of pregnancies with red-cell allo-immunisation: state-of-the art]. Gynecol Obstet Fertil 2010;38:205-13. 18. Moise KJ Jr. Management of rhesus alloimmunization in pregnancy. Obstet Gynecol 2002;100:600-11. 19. Mari G. Middle cerebral artery peak systolic velocity for the diagnosis of fetal anemia: the untold story. Ultrasound
12. Teixeira JM, Duncan K, Letsky E, et al. Middle cerebral artery peak systolic velocity in the prediction of fetal anemia. Ultrasound Obstet Gynecol 2000;15:205-8.
Obstet Gynecol 2005;25:323-30. 20. Mari G, Abuhamad AZ, Cosmi E, et al. Middle cerebral artery peak systolic velocity: technique and variability.
13. Zimmerman R, Carpenter RJ Jr, Durig P, et al. Longitudinal measurement of peak systolic velocity in the fetal middle
J Ultrasound Med 2005;24:425-30. 21. Poissonnier MH, Picone O, Brossard Y, et al. Intravenous
cerebral artery for monitoring pregnancies complicated by red cell alloimmunisation: a prospective multicentre trial
fetal exchange transfusion before 22 weeks of gestation in early and severe red-cell fetomaternal alloimmunization.
with intention-to-treat. BJOG 2002;109:746-52.
Fetal Diagn Ther 2003;18:467-71.
Volume 54, November 2014 TRANSFUSION
2973
BACKGROUND: We aimed to assess usefulness of the middle cerebral artery peak systolic velocity (MCA-PSV) in the prediction of fetal anemia after more than three intravenous fetal-exchange transfusions (IFET). STUDY DESIGN AND METHODS: A retrospective study was conducted over 6 years of 15 consecutive pregnancies with severe red blood cell fetomaternal alloimmunization requiring more than three IFETs. We evaluated correlation between MCA-PSV (expressed as multiples of the mean [MoM]) and pretransfusion hemoglobin (Hb) in the fetus (MoM). Analyses were also performed to assess the value of MCA-PSV to predict moderate to severe fetal anemia. RESULTS: Twenty-seven MCA-PSV measurements performed before the fourth to last IFET were coupled with pretransfusion Hb in the fetus. The median number of IFETs per fetus was five (range, four to eight). Five Hb samples found fetuses with severe (19%), seven with moderate (26%), and 15 with mild anemia (56%). There was a linear correlation between MCA-PSV(x) and Hb in the fetus(y): y = −0.21x + 0.93 (r = −0.50, p < 0.01). For the prediction of moderate to severe anemia the negative predictive value of MCA-PSV with a threshold of 1.5 MoM was 75%, positive predictive value 73%, specificity 80%, sensibility 67%, and positive likelihood ratio 3.33. The area under the receiver operating characteristic curve was 0.78 (95% confidence interval, 0.59-0.96; p < 0.001). For the prediction of severe anemia, MCA-PSV with a threshold of 1.5 MoM had 94% negative predictive value, 80% sensibility, and a positive likelihood ratio of 2.5. CONCLUSIONS: This study shows that a correlation between MCA-PSV and Hb in the fetus persists even after more than three IFETs. MCA-PSV measurements thus remain useful to monitor fetuses at risk of anemia.
2968
TRANSFUSION Volume 54, November 2014
V
ital statistics data indicate that there are approximately 800,000 pregnancies per year in France1,2 and probably 1,100,000 to 1,200,000 conceptions. Approximately 15% of these women are D–. Despite the prophylactic measures involving RhIG administration recommended by the French National College of Gynecology and Obstetrics in 2005,3,4 maternal red blood cell (RBC) alloimmunization remains the main cause of fetal anemia in France. The risk of hemolytic disease due to alloimmune antibodies is estimated at one per 28,000 fetuses, with 28 in utero or perinatal deaths1 each year in France. In utero surveillance of fetuses at risk thus remains essential. All RBC antibodies can be involved, but anti-D (Rhesus), anti-Kell, and anti-c are found most frequently. Before 2000, techniques to detect fetal anemia were either invasive (cordocentesis, amniocentesis), unreliable (antibody assays level in maternal blood, fetal cardiac rhythm), or late (sinusoidal fetal cardiac rhythm, signs of hydrops on ultrasound examination). Mari and coworkers demonstrated in 20005,6 that moderate to severe anemia can be detected noninvasively by Doppler ultrasonography from an increase in the peak velocity of systolic blood flow in the middle cerebral artery (MCAPSV) to more than 1.5 multiples of the mean (MoM) in
ABBREVIATIONS: IFET(s) = intravenous fetal-exchange transfusion(s); MCA-PSV = middle cerebral artery peak systolic velocity; MoM = multiples of the mean; ROC = receiver operating characteristic. From the 1Department of Obstetrics and Gynecology, Port-Royal Maternity, Groupe Hospitalier Cochin-Broca-Hôtel Dieu, Assistance Publique-Hopitaux de Paris; and 2Paris V René Descartes University, Paris, France. Address reprint requests to: Monika Hermann, Maternité de Port Royal, 53 Boulevard de l’Observatoire, 75014 Paris, France; e-mail: [email protected]. Received for publication February 13, 2014; revision received April 1, 2014, and accepted April 2, 2014. doi: 10.1111/trf.12714 © 2014 AABB TRANSFUSION 2014;54:2968-2973.
CEREBRAL DOPPLER AFTER FETAL TRANSFUSION
fetuses at risk of maternal RBC alloimmunization. Other studies7-13 over the past decade have confirmed these results: there is a strong correlation between the Doppler measurement of MCA-PSV and fetal anemia: R2 = 0.55 in the study by Mari and colleagues6 and 0.654 in the one by Carbonne and colleagues8 (p < 0.001). The threshold for intervention in Mari’s study was 1.5 MoM. Fetal anemia can be treated with intravenous (IV) fetal exchange transfusion (IFET). MCA-PSV is useful for timing the first,6 second14 (R2 = 0.51 for 1.69 MoM), and third15 (R2 = 0.43 for 1.50 MoM) IFETs, but after more than two IFETs, adult RBCs totally replace the fetal cells.16 Yet adult RBCs have different physical characteristics: they are smaller and less rigid and have less oxygen-binding capacity, whereas fetal RBCs carry more oxygen but release less to tissue. Fetal surveillance has not been tested for more than three IFETs and it is uncertain whether known curves to predict fetal anemia are still valid because of rheologic changes in fetal blood after IV transfusion with adult blood. The purpose of this study was to assess the value of MCA-PSV in predicting fetal anemia in fetuses that have undergone more than three IFETs.
MATERIALS AND METHODS This retrospective study was conducted at Saint-Vincent de Paul Hospital in Paris, France. We reviewed all records of pregnancies with IFET from January 1, 2004, through December 31, 2009 (n = 57). We excluded fetuses that had IFET for causes other than RBC alloimmunization (n = 12) and those with less than three IFETs (n = 20). Management of pregnancies at risk of RBC alloimmunization was done as described in Fig. 1 according to management described in literature.17,18 Antibody screening was performed (by indirect antiglobulin test) at the first prenatal visit, and any antibodies identified were quantified. If the maternal antibody was anti-D, fetal Rhesus genotyping was performed on maternal serum. D– fetuses did not require any particular further follow-up, while serum antibody level was monitored every 2 weeks for D+ fetuses. When this level reached 1 μg/L or more than 1/16, a weekly ultrasound examination included a Doppler measurement of the MCA-PSV. The sonographer also sought other signs of fetal anemia, such as hepatomegaly and hydrops. The technique to measure MCA-PSV was the one described by Mari and coworkers.6,19,20 Two trained physician-sonographers (MHP and ML) performed all ultrasound evaluations. IFET was scheduled when MCAPSV increased to higher than 1.5 MoM. The IFET procedure was done in an operating room, after administering local anesthesia to the mother and monitoring the fetal cardiac rhythm. The method was the one previously described by Poissonnier and coworkers.21 Cordocentesis was performed by ultrasonographically guided puncture
of the umbilical vein with an 18-gauge needle: for the first IFET, Rhesus incompatibility was first confirmed by blood group and Rhesus genotyping. At the beginning of each IFET, the severity of anemia was evaluated by fetal umbilical vein blood sampling with a point-of-care hemoglobin (Hb) test (Hemocue, Angelholm, Sweden), confirmed by laboratory results. When the Hb concentration in the fetus was less than 0.85 MoM, IFET was performed immediately after fetal anesthesia with vecuronium bromide (0.1 mg/kg), transfusing irradiated group O– compatible cytomegalovirus-negative packed RBCs. The KleihauerBetke test on the umbilical vein was used to detect residual fetal RBCs at the end of each IFET. IFET was performed every 1 to 4 weeks up to a gestational age of 34 weeks, when MCA-PSV exceeded 1.5 MoM or when it increased rapidly. After 34 weeks of gestation, labor was induced when MCA-PSV exceeded 1.5 MoM. If measurements remained approximately or less than 1 MoM, birth was scheduled at 36 weeks with induction of labor or cesarean delivery if vaginal birth was not possible. At birth, the neonatal blood was sampled to determine Hb at birth and detect irregular RBC antibodies. The main outcome measure was the correlation between MCA-PSV before IFET and Hb in the fetus at cordocentesis at the beginning of each procedure. We assessed MCA-PSV with a threshold of 1.5 diagnostic accuracy to predict fetal anemia. Values of Hb in the fetus and MCA-PSV are expressed as MoM to adjust for gestational age at the time of measurement according to the classification of Mari and colleagues:6 mild anemia corresponds to Hb concentrations in the fetus from 0.84 to 0.65 MoM, moderate anemia to Hb concentrations from 0.65 to 0.55 MoM, and severe anemia to Hb concentrations below 0.55 MoM. Statistical analyses were performed with computer software (XL-STAT Life, Addinsoft SARL, Paris, France; and Stata 12.1, StataCorp, College Station, TX). Correlations were calculated with Pearson’s test. Results were considered significant when p values were less than 0.05. Results are expressed as means (with their 95% confidence intervals [CIs]). Regression analysis was used to assess the correlation between MCA-PSV and Hb in the fetus.
RESULTS Over the 6-year study period, there were 13,758 births in Saint Vincent de Paul Hospital and 240 pregnancies monitored for risk of fetal anemia. At least one IFET was required for 57 of them, and 45 of these involved maternal RBC immunization. Of these pregnancies, 19 required more than two IFETs and 15 required more than three. Of these 15 fetuses, 13 were live born, while two died in utero. Both had persistent or recurrent hydrops, even after IFET, and the pathology examinations found it to be the only cause of death. Volume 54, November 2014 TRANSFUSION
2969
HERMANN ET AL.
Fig. 1. Management of D– women and fetuses at risk of anemia. GA = gestational age.
2970
TRANSFUSION Volume 54, November 2014
CEREBRAL DOPPLER AFTER FETAL TRANSFUSION
The antibody responsible for maternal alloimmunization was anti-D in 14 pregnancies and anti-Kell in one. MCA-PSV is as reliable for anti-Kell immunization as for anti-D immunization;19 therefore, management was the same for both. Anti-D were associated with other antibodies in 11 pregnancies: six anti-D + C, two anti-C + D + E, one anti-D + E, one anti-D + C + a, and one antiD + C + Kell. The 15 fetuses underwent a median of five IFETs (range, four to eight). The mean time between the last TABLE 1. Characteristics of transfused fetuses Data Number of IFETs per fetus Hb in the fetus before IFET MCA-PSV before IFET Gestational age at delivery
Mean (±SD) 5.6 (±1.4) 0.64 MoM (±0.12) 1.38 MoM (±0.29) 34+6 weeks (±2.4 days)
TABLE 2. Characteristics for MCA-PSV above 1.5 MoM after the third IFET
Characteristic Negative predictive value Positive predictive value Specificity Sensibility Positive likelihood ratio Negative likelihood ratio
Prediction of moderate to severe anemia* 75% 73% 80% 67% 3.33 0.42
* Hb in the fetus below 0.65 MoM. † Hb in the fetus below 0.55 MoM.
Prediction of severe anemia† 94% 36% 68% 80% 2.51 0.29
IFET and delivery was 21.5 days (95% CI, 10.3-32.7 days). The mean gestational age at birth was 34 + 6 weeks (95% CI, 32 + 3-37 + 1). The mean of the last measured MCA-PSV values before birth was 0.9 MoM (95% CI, 0.7-1.2 MoM). Overall, these fetuses underwent 87 cordocentesis procedures for suspected fetal anemia. Only one cordocentesis procedure was not followed by IFET; it was delayed for a few days because the ultrasound fetal cardiac rhythm showed poor fetal tolerance. All cordocentesis confirmed fetal anemia with Hb in the fetus below 0.84 MoM. Three fetuses, including the two who died in utero, had hydrops. The hydropic fetus who survived had a sinusoidal fetal cardiac rhythm that indicated the need for emergency cesarean delivery. In all, 39 IFETs were performed after the third IFET in these 15 fetuses. We recorded 27 pairs of pre-IFET MCAPSV and pretransfusion Hb measurements (some records were incomplete). The mean MCA-PSV before IFET was 1.38 MoM (95% CI, 1.09-1.66 MoM), and 11 measurements exceeded 1.5 MoM. Five pretransfusional blood samples (19%) concluded to severe anemia, seven to moderate (26%), and 15 (56%) to mild anemia. All Kleihauer-Betke tests were less than 0.01% after the second IFET. The mean Hb in the fetus before IFET was 0.64 MoM (±0.12; see Table 1). Of the 13 live-born babies, two had moderate anemia at birth, and three had mild anemia; the other eight were all born nonanemic (Hb > 0.84 MoM). Pretransfusion Hb (y) is correlated with its corresponding MCA-PSV(x), with r = −0.50, p < 0.01 (see Fig. 2). The relation is linear, expressed by y = −0.21x + 0.93. Table 2 respectively shows capacity of MCA-PSV with a threshold of 1.5 MoM to predict moderate to severe fetal anemia and severe anemia. The area under the receiver operating characteristic (ROC) curve to predict moderate to severe anemia is 0.78 (95% CI, 0.59-0.96; p < 0.001; see Fig. 3). To predict severe anemia, the area under the ROC curve is 0.65 (95% CI, 0.35-0.95).
DISCUSSION
Fig. 2. Linear regression of Hb in the fetus function of MCA-PSV after more than three IFETs. R2 = 0.254; p < 0.01.
Our study shows that MCA-PSV continues to be strongly negatively correlated with Hb in the fetus after more than three IFETs (r = −0.50, p < 0.01): the higher the MCA-PSV, the lower the expected Hb in the fetus. This is, to our knowledge, the first study to have assessed accuracy of MCA-PSV after more than three IFETs. Our findings are consistent with previous results in studies of anemic fetuses that had no, one, or two Volume 54, November 2014 TRANSFUSION
2971
HERMANN ET AL.
PSV measurements are found to be more than 1.5 MoM during fetal follow-up in between scheduled IFET. No fetuses followed by cerebral Doppler were born anemic, but no study has yet determined the gestational age until which this follow-up is reliable. Future research also needs to define the best term at which these fetuses should be delivered. In conclusion, our data show that the correlation between MCA-PSV and Hb in the fetus persists even after more than three IV fetal transfusions. Cerebral Doppler thus remains a useful tool for monitoring fetuses at risk of anemia undergoing serial IFET because of maternal alloimmunization.
Fig. 3. ROC curve to predict fetal moderate to severe anemia with MCA-PSV. Area under the curve = 0.78 (95% CI, 0.59-0.96).
IFETs.6,8,14,15 The diagnostic performance we found is, however, lower than that found by previous authors.6-10,12,13 This difference might have several explanations. Indeed, after more than three IFETs fetal blood has totally been replaced by adult blood with different physical properties; MCA-PSV measurements might thus be less reliable. Further, the study involved a small number of fetuses. A measurement bias could also affect our results: in some records, the reported MCA-PSV values available were measured 1 to 4 days before IFET; the real pretransfusion values were probably higher. Some IFETs were indeed scheduled in advance for practical reasons: only 45% of third or later IFETs were performed immediately after an MCA-PSV measurement above 1.5 MoM; the other serial IFETs were scheduled because of a rapid increase in the Doppler measurements or scheduled approximately 15 days after the previous IFET to take RBC hemolytic clearance into account. Such a bias might have been avoided with a prospective study that performed IFET only when MCA-PSV values reached 1.5 MoM. The correlation we found does not, however, prove that MCA-PSV is useful to schedule the timing of IFET beyond the second one. A prospective study of a larger group is needed for a more precise determination of the diagnostic accuracy of MCA-PSV for predicting fetal anemia after serial IFET. Awaiting these data, we suggest that MCA-PSV should still be used to monitor fetuses after serial IFET—a weekly follow-up should be maintained until 34 to 36 weeks. Although IFETs are often scheduled in advance taking into account RBC hemolytic clearance in fetuses transfused for severe alloimmunization, our result suggests that an IFET should be performed earlier if MCA2972
TRANSFUSION Volume 54, November 2014
ACKNOWLEDGMENT We thank Dr A. Cortey, head of the CNRHP
unit (Armand Trousseau Hospital, Paris, APHP), for preparing the blood to transfuse for the fetuses included in the study and analyzing all the blood samples.
CONFLICTS OF INTEREST The authors have disclosed no conflicts of interest.
REFERENCES 1. Branger B, Winer N. [Epidemiology of anti-D alloimmunization during pregnancy]. J Gynecol Obstet Biol Reprod (Paris) 2006;35(1 Suppl):1S87-S92. 2. INSEE. Bilan demographique. January 2009;1220. 3. Cortey A, Brossard Y. [Prevention of fetomaternal rhesus-D allo-immunization. Practical aspects]. J Gynecol Obstet Biol Reprod (Paris) 2006;35(1 Suppl):1S123-1S30. 4. CNGOF. Text of the guidelines for prevention of fetomaternal rhesus-D allo-immunization. J Gynecol Obstet Biol Reprod (Paris) 2006;35(1 Suppl):1S131-1S35. 5. Mari G, Adrignolo A, Abuhamad AZ, et al. Diagnosis of fetal anemia with Doppler ultrasound in the pregnancy complicated by maternal blood group immunization. Ultrasound Obstet Gynecol 1995;5:400-5. 6. Mari G, Deter RL, Carpenter RL, et al. Noninvasive diagnosis by Doppler ultrasonography of fetal anemia due to maternal red-cell alloimmunization. Collaborative Group for Doppler Assessment of the Blood Velocity in Anemic Fetuses. N Engl J Med 2000;342:9-14. 7. Abdel-Fattah SA, Soothill PW, Carroll SG, et al. Middle cerebral artery Doppler for the prediction of fetal anaemia in cases without hydrops: a practical approach. Br J Radiol 2002;75:726-30.
CEREBRAL DOPPLER AFTER FETAL TRANSFUSION
8. Carbonne B, Castaigne-Meary V, Cynober E, et al. [Use of peak systolic velocity of the middle cerebral artery in the management of fetal anemia due to fetomaternal erythrocyte alloimmunization]. J Gynecol Obstet Biol Reprod (Paris) 2008;37:163-9. 9. Delle Chiaie L, Buck G, Grab D, et al. Prediction of fetal anemia with Doppler measurement of the middle cerebral
14. Detti L, Oz U, Guney I, et al. Doppler ultrasound velocimetry for timing the second intrauterine transfusion in fetuses with anemia from red cell alloimmunization. Am J Obstet Gynecol 2001;185:1048-51. 15. Mari G, Zimmermann R, Moise KJ Jr., et al. Correlation between middle cerebral artery peak systolic velocity and fetal hemoglobin after 2 previous intrauterine transfusions.
artery peak systolic velocity in pregnancies complicated by
Am J Obstet Gynecol 2005;193(3 Pt 2):1117-20.
maternal blood group alloimmunization or parvovirus B19 infection. Ultrasound Obstet Gynecol 2001;18:232-6.
16. Welch R, Rampling MW, Anwar A, et al. Changes in hemorheology with fetal intravascular transfusion.
10. Kurmanavicius J, Streicher A, Wright EM, et al. Reference values of fetal peak systolic blood flow velocity in the middle cerebral artery at 19-40 weeks of gestation. Ultrasound Obstet Gynecol 2001;17:50-3. 11. Scheier M, Hernandez-Andrade E, Fonseca EB, et al. Prediction of severe fetal anemia in red blood cell alloimmunization after previous intrauterine transfusions. Am J Obstet Gynecol 2006;195:1550-6.
Am J Obstet Gynecol 1994;170:726-32. 17. Carbonne B, Castaigne V, Cynober E, et al. [Follow-up of pregnancies with red-cell allo-immunisation: state-of-the art]. Gynecol Obstet Fertil 2010;38:205-13. 18. Moise KJ Jr. Management of rhesus alloimmunization in pregnancy. Obstet Gynecol 2002;100:600-11. 19. Mari G. Middle cerebral artery peak systolic velocity for the diagnosis of fetal anemia: the untold story. Ultrasound
12. Teixeira JM, Duncan K, Letsky E, et al. Middle cerebral artery peak systolic velocity in the prediction of fetal anemia. Ultrasound Obstet Gynecol 2000;15:205-8.
Obstet Gynecol 2005;25:323-30. 20. Mari G, Abuhamad AZ, Cosmi E, et al. Middle cerebral artery peak systolic velocity: technique and variability.
13. Zimmerman R, Carpenter RJ Jr, Durig P, et al. Longitudinal measurement of peak systolic velocity in the fetal middle
J Ultrasound Med 2005;24:425-30. 21. Poissonnier MH, Picone O, Brossard Y, et al. Intravenous
cerebral artery for monitoring pregnancies complicated by red cell alloimmunisation: a prospective multicentre trial
fetal exchange transfusion before 22 weeks of gestation in early and severe red-cell fetomaternal alloimmunization.
with intention-to-treat. BJOG 2002;109:746-52.
Fetal Diagn Ther 2003;18:467-71.
Volume 54, November 2014 TRANSFUSION
2973
