Fetomaternal hemorrhage following trauma

December 15, 1985 Am J Obstet Gynecol

Leveno, Cunningham, and Pritchard

4. Porreca RP. High cesarean section rate: a new perspective. Obstet Gynecol 1985;65:307-12. 5. O'Driscoll K, Foley M, MacDonald D. Active management of labor as an alternative to cesarean section for dystocia. Obstet Gynecol 1984;63:485-90. 6. MacDonald D. National Maternity Hospital, Dublin: clinical report for the year 1983. Dublin: Cahill Printers, 1984. 7. Bottoms SF, Rosen MG, Sokol RJ. The increase in cesarean birth rate. N Eng! J Med 1980;302:559-63.

8. Rosen MG. An introduction to the process of consensus. AMJ 0BSTET GYNECOL 1981;139:901-9. 9. Pritchard JA, MacDonald PC. Williams Obstetrics. 15th ed. New York: Appleton-Century-Croft s, 1976:904. 10. Committee on obstetrics: maternal and fetal medicine. Guidelines for vaginal delivery after a previous cesarean birth. ACOG Newsletter 1985;(February)8. 11. Graham AR. Trial labor following previous cesarean section. AMJ 0BSTET GYNECOL 1984;149:35-44.

Fetomaternal hemorrhage following trauma Peter G. Rose, M.D., Patricia L. Strohm, M.T., and Frederick P. Zuspan, M.D. Columbus, Ohio Fetomaternal hemorrhage can result from different types of trauma and may be followed by fetal anemia, fetal death, or isoimmunization. We prospectively studied the frequency and volume of fetomaternal hemorrhage, fetal well-being, abruptio placentae, and fetal outcome in 32 pregnant patients suffering recent trauma. Fetomaternal hemorrhage occurred in nine of 32 trauma patients (28%) with a mean volume of 16 ml ± 14.3(SD). There was a statistically significant difference in the frequency and mean volume of fetomaternal hemorrhage in this group over that in gestational-age-matched controls. Neither the nature of the trauma nor the gestational age was related to the frequency or volume of fetomaternal hemorrhage. The outcome in three of the nine trauma patients who sustained fetomaternal hemorrhage was poor; fetal anemia, paroxysmal atrial tachycardia, and fetal death occurred in each one. Maternal trauma remains a significant cause of maternal and fetal morbidity and death, and the use of the Kleihauer-Betke analysis is indicated to identify fetomaternal hemorrhage. Rh-immune globulin therapy should be given to Rh-negative patients with fetomaternal hemorrhage. (AM J OssTET GvNECOL 1985;153:844-7.)

Key words: Pregnancy, trauma, fetomaternal hemorrhage

Fetomaternal hemorrhage can result from abdominal trauma and obstetric manipulation.'-" The consequences may include fetal anemia, fetal death, and isoimmunization if the mother is Rh-negative. Fetomaternal hemorrhage has been periodically reported 4 following catastrophic maternal trauma,"· but no study has prospectively evaluated the relationship of fetomaternal hemorrhage to trauma. We undertook a prospective study of the incidence and significance of fetomaternal hemorrhage in pregnant women who had suffered abdominal or multiple trauma by means of the Kleihauer-Betke analysis, fetal heart rate monitoring, and ultrasonic examination of the placenta and evaluated the pregnancy outcome.

From the Department of Obstetrics and Gynecology and the Transfusion Service, The Ohio State University. Received for publication August 23, 1985; accepted September 23, 1985. Reprint requests: Peter G. Rose, M.D., Department of Obstetrics and Gynecology, 410 West lOth St., Columbus, OH 43210.

844

Material and methods

During a 12-month period 32 pregnant patients suffering recent abdominal and/or multiple trauma were studied to detect and measure fetomaternal hemorrhage. Peripheral blood samples were obtained upon admission (in all cases, within 48 hours of the trauma), and analyzed for fetal red blood cells with the use of a modified Kleihauer-Betke technique (BMC fetal hemoglobin analysis, Biodynamics Inc., Indianapolis, Indiana). Peripheral blood smear slides, each of which contained approximately 50,000 red blood cells, were examined for fetal red blood cells. When fetal red blood cells were detected, 1000 cells were counted, and the estimated fetal bleed was calculated with the following formula: ad'usted maternal Feta~ red blood Stained cells = cells m maternal X ~ . red cell volume U nstamed cells circulation The accuracy of the Kleihauer-Betke technique depends on the number of red blood cells counted. AI-

Volume 153 Number 8

though research laboratories have counted 350,000 red blood cells per slide, most clinical laboratories count only I 000 cells in the manner described. Scanning 50,000 red blood cells makes the technique very sensitive for determining the presence of fetal red blood cells, whereas counting 1000 cells decreases the accuracy in determining the volume of fetal-maternal hemorrhage. We chose the latter method because of its wider clinical use in order to assure greater clinical applicability. Red blood cells incompletely stained with O.I% erythrosin were defined as adult cells, a conclusion confirmed with hemoglobin electrophoresis. Rh-negative patients with documented fetomaternal hemorrhage received an appropriate dose of Rh-immune globulin. All patients who were at more than 20 weeks' gestation also underwent continuous fetal heart rate monitoring for a minimum of 30 minutes at the time of initial evaluation. Fetal heart monitor tracings were analyzed for baseline heart rate, the presence or absence of accelerations, and the presence of decelerations. Immediate ultrasonic examination of the placental location was performed to evaluate possible abruptio placentae. At the time of labor, all patients underwent either external or internal electronic fetal heart rate monitoring. Each infant was given a physical examination by the pediatric service at the time of delivery and at discharge. Based on initial physical examination, infants were assigned to the normal nursery or the neonatal intensive care unit for evaluation. A control population of 32 gestational age-matched patients without a history of trauma during pregnancy were obtained from our obstetric clinic, and peripheral blood samples were evaluated by Kleihauer-Betke analysis as described (Table I). Data were analyzed by the X2 test or Fisher's exact test and continuous data were analyzed by a I- or 2- sample t test.

Results Of 32 patients who suffered abdominal or multiple trauma seen during a I2-month period, nine (28%) suffered fetomaternal hemorrhage. The nature of the trauma consisted of falling to the floor (seven), falling down a flight of stairs (six), motor vehicle accidents with or without direct abdominal trauma (II), and direct abdominal trauma alone (eight) (Table II). Although the severity of the injuries varied, all were limited to contusions or superficial lacerations. No pregnant patients were hypotensive of tachycardiac during evaluation and none required operative care. The type of injury was not positively correlated with either the incidence or the volume of fetomaternal hemorrhage. The gestational age of the 32 patients varied from 8 to 38 weeks, with a mean of 27.0 weeks. Patients with fetomaternal hemorrhage had a mean gestational

Fetomaternal hemorrhage following trauma

845

Table I. Fetomaternal hemorrhage in trauma and gestational age-matched control groups Group

Trauma Fetomaternal hemorrhage No fetomaternal hemorrhage Gestational age-matched control Fetomaternal hemorrhage No fetomaternal hemorrhage

Sample size

32 9

23 32 2 30

age of 29.0 ± 4.5 weeks versus 26.3 ± 7.6 weeks for patients who did not suffer fetomaternal hemorrhage. This difference was not statistically significant (p = 0.3). The incidence of fetomaternal hemorrhage (28%) was significantly greater than the 6% in the control group (t = 5.38, df = I, p < 0.025). The volume of fetomaternal hemorrhage varied from 5 to 40 ml, with a mean of I6 mi. This represents I4% to 34% of the fetal blood volume with a mean of 2I%. The mean amount of fetal bleeding of I6 ml in the trauma group was significantly greater (t = 3.5, df = 8, p < 0.02) than the maximum amount of spontaneous fetal bleeding reported by Jorgensen' (I ml) and significantly greater (t = 2.3I, df = 8, p < 0.05) than the maximum volume (5 ml) of spontaneous fetomaternal hemorrhage in the control group. The frequency of Rh-negative blood type among trauma patients was 9%, comparable to that in the general population of central Ohio. 8 No relationship between Rh type and the incidence of fetomaternal hemorrhage was noted. Rh-negative patients with documented fetomaternal hemorrhage received Rhimmune globulin. No fetal heart rate abnormalitjes or ultrasonically diagnosed cases of abruptio placentae were noted in any of these patients. Three of the nine patients with fetomatern'al hemorrhage in the trauma group experienced complications. One patient, who sustained a motor vehicle accident at 24 weeks, presented 2 days later with an intrauterine fetal death. Fetal red blood cell analysis demonstrated a IO ml hemorrhage. Ultrasound examination did not show abruptio placentae, por was it noted on placental inspection at delivery. The patient was Rh-negative, and Rh-immune gjobulin was given. A second patient, who was kicked in the abdomen at 32 weeks' gestation, experienced a 5 ml fetoma,ternal hemorrhage and was delivered of a term infant with meconium-stained fluid. The infant, who had an otherwise uncomplicated delivery without excessive blood loss, was anemic. A third patient, with fetal bleeding of I 0 ml, was noted while in labor to have fetal paroxysmal atrial tachycardia which resolved shortly after delivery. The remaining six patients with fetoma~ernaJ hem-

846

Rose, Strohm, and Zuspan

December 15, 1985 Am J Obstet Gynecol

Table II. Patients suffering abdominal or multiple trauma

Patient

Estimated gestational age (Wk)

Fetal bleeding

Trauma

Rh type

Fetal outcome

No fetomatemal hemorrhage

A. C. R. D. L.C. T. s. L. H. B. W. S.C. L. S. P. C. D. T. F. B. L. H. L. 0.

s. w.

J.W. B.S. P. K. L.T. T. R. V. N. R. M. T. R. D. F. Fetomatemal hemorrhage K.V.

S. R.

v. s.

C. N. C. B. A. H. B. W. R. T. L.J.

8 12 18 18 20 21 23 24 25 26 26 26 28 29 30 32 32 32 33 .33 34 36 38

MVA, LOC Abdominal trauma MVA Abdominal trauma Abdominal trauma MVA MVA Fall, 5 steps Fall, 7 steps Fall Abdominal trauma Abdominal trauma MVA Abdominal trauma MVA MVA Fall, 7 steps Fall MVA Fall Fall Fall Fall

23 24 25 28 30 31 31 32 37

MVA MVA Fall, 15 steps Fall, 6 steps Fall MVA Fall, 7 steps Abdominal trauma Abdominal trauma

MVA = Motor vehicle accident; LOC PAT = paroxysmal atrial tachycrdia.

=

loss of consciousness; lost

orrhage had an uncomplicated subsequent prenatal course and delivery of healthy infants despite fetomaternal hemorrhage. Although an absolute causal relationship cannot be established between fetomaternal hemorrhage and fetal outcome, analysis relating fetal outcome to prior fetomaternal hemorrhage by Fisher's exact test yielded a value of p = 0.06 (two-tailed test). While this is not significant in the view of the classically accepted 0.05 value, a strong trend exists.

Comment Maternal trauma remains a significant cause of maternal and fetal morbidity and death. The percentage of maternal deaths in North Carolina resulting from motor vehicle accidents doubled from 5.7 in 1970 to 1l.4in1975." Fort and Harlin 10 studied 240 cases of noncatastrophic trauma in 210 gravid patients. These traumatic events included abdominal assault, falls, and motor vehicle accidents requiring only observation or minor laceration repair. The percentage of maternal trauma in-

5 10 5 10 10 40 20 5 40

= lost

ml ml ml ml ml ml ml ml ml

to follow-up; IUFD

+ + + + + + + + + + + + + + + + + + + + + +

Good Good Lost Good Good Good Good Good Good Good Good Good Lost Good Good Good Good Good Good Good Good Good Good

+

Good IUFD Good PAT Good Good Good Anemia Good

+ + + + + +

=

intrauterine fetal death;

creased with each trimester: 8.8, 39.6, and 51.7, respectively, in the first, second, and third trimesters. Eight of these 210 patients suffered fetal losses. The presence or absence of symptoms following trauma was not of prognostic significance to pregnancy outcome, as was the case in our study. Bickers and Wennberg 2 and Galle and Anderson 9 demonstrated maternal trauma to be associated with fetal distress as determined by fetal heart monitoring. These articles demonstrate that maternal vital signs, on occasion, have not reflected uteroplacental circulation or fetal well-being. 2 · 9 Maternal cardiovascular shock may be less apparent because of an expanded blood volume as well as peripheral and uterine vasoconstriction. Clearly, fetal heart monitoring has a role in evaluation of patients after trauma. Documentation of fetomaternal hemorrhage is important in the evaluation of pregnant trauma patients.' Patients with documented fetomaterna\ hemorrhage may have benign fetal heart monitor tracings initially yet be at risk to develop later fetal distress. Repeated

Volume 153 Number 8

Kleihauer-Betke analysis can identify patients with continued fetomaternal hemorrhage. In the presence of maternal trauma, the assumption should be made that a fetal hemorrhage may have occurred. Only by use of the Kleihauer-Betke analysis can this be excluded. Ultrasonography did not aid in the diagnosis or management of fetomaternal hemorrhage in this study. Jorgensen, 1 counting 50,000 red blood cells/slide, reported spontaneous fetomaternal hemorrhage occurring in 8% of 151 second-trimester and 12% of 186 third-trimester pregnancies. However, 75% of the second-trimester and 83% of the third-trimester hemorrhages were less than 0.1 ml of fetal blood, with the remainder of the patients in both groups having 0.1 to 1 ml of bleeding. The magnitude of the fetomaternal hemorrhage in our study, ranging from 5 to 40 ml and averaging 16 ml, as noted previously, shows a statistical relationship between the patient's trauma and fetomaternal hemorrhage. As little as 0.01 to 0.03 ml of Rh-positive blood as a single injection will sensitize 70% of Rh-negative patients." The remaining 30% ofRh-negative patients are nonresponders and may not be sensitized even after repeated injections over many years. 11 All patients in our study sustained fetomaternal hemorrhage significant enough to sensitize Rh-negative mothers. The following protocol is now recommended for patients with maternal trauma. The degree of trauma and the absence of overt damage should not preclude this evaluation, which includes: (1) continuous heart rate monitoring (for a minimum of 30 minutes); (2) Kleihauer-Betke analysis to document fetomaternal hem-

Fetomaternal hemorrhage following trauma

847

orrhage; (3) Rh determination of the mother; (4) repeat Kleihauer-Betke analysis, if fetomaternal hemorrhage exists, to rule out chronic fetomaternal hemorrhage; repeated fetal monitoring if chronic fetomaternal hemorrhage exists; (5) an appropriate dose of Rh-immune globulin therapy for Rh-negative mothers with documented fetal hemorrhage. REFERENCES I. Jorgensen ]. Feto-maternal bleeding during pregnancy

and at delivery. Acta Obstet Gynecol Scand 1977;56:487. 2. Bickers RG,.Wennberg R. Feto-maternal transfusion following trauma. Obstet Gynecol 1983;61:258. 3. Renaer M, Van de Putte I, Vermylen C. Massive fetomaternal hemorrhage as a cause of perinatal mortality and morbidity. Eur J Obstet Gynecol Reprod Bioi 1976;6:125. . 4. Gjide P, Rasmussen TB,JirgensenJ. Feto-maternal bleeding during attempts at external version. Br J Obstet Gynaecol1980;87:571. 5. Marcus RG, Crewe-Brown H, Krawitz S, eta!. Feto-maternal haemorrhage following successful and unsuccessful attempts at external cephalic version. Br J Obstet Gynaecol 1975;82:578. 6. Lele AS, Carmody PJ, Hurd ME, et a!. Feto-maternal bleeding following diagnostic amniocentesis. Obstet Gynecol 1982;60:60. 7. Dawes GS. Fetal and neonatal physiology. Chicago: Year Book Medical Publishers, 1968: 177. 8. McDowell M. (Personal Communication), American Red Cross, Columbus, Ohio 9. Galle PC, Anderson SG. A case of automobile trauma during pregnancy. ObstetGynecol 1979;54:467. I 0. Fort AT, Harlin RS. Pregnancy outcome after noncatastrophic maternal trauma during pregnancy. Obstet Gynecol 1970;35:912. 11. Mollison PL. Clinical aspects of Rh immunization. Am J Clin Pathol 1973;60:287. 12. ScottJR, WarenskiJC. Tests to detect and quantitate fetomaternal bleeding. Clin Obstet Gynecol 1982;25:277.