Diaa M. El-Mowafi - Zagagig University, Egypt

Placental Localization by Transperineal Sonography in Antepartum Hemorrhage

Diaa El-Mowafi, Abdel-Fattah Hegazi
Department of Obstetrics and Gynecology, Benha Faculty of Medicine

Correspond Author: Diaa El-Mowafi,M.D. 4 Ghazza St., El-Hossania, El-Mansoura 35111, Egypt. Tel. +2-050-363308 Fax +2-050-332771

Abstract

Transperineal and transabdominal sonography were done for 180 patients attending Benha University Hospitals at 28-40 weeks gestational age with an antepartum hemorrhage in the period between July, 1995 and October, 1996.

Transabdomianl sonograms (TAS) were not conclusive for detection or exclusion of placenta praevia (PP) in 128 patients (71.1%) due to unsatisfactory visualization of the cervix. Four patients of those were found to have PP during delivery. The remaining 52 patients (28.9%) were diagnosed by (TAS) as having PP which confirmed at delivery.

Transperineal sonography (TPS) successfully visualized the internal surface of the cervix in all our 180 patients, allowing exclusion of PP in 122 cases (67.8%), none of them showed evidence of PP at delivery. TPS showed PP in 58 patients (32.2%), where it was confirmed in 56 patients only at delivery. TAS showed 92.8% sensitivity, 94.3% specificity, 88.1% positive predictive value,96.6% negative predictive value, with an overall accuracy of 93.8%. TPS showed 98.5% sensitivity, 97.5% specificity, 94.8% positive predictive value, 99.1% negative predictive value with an overall accuracy of 97.7%.

Introduction

Although sonography is the method of choice for evaluating patients with antepartum hemorrhage, still it gives false positive results due to distortion of the lower uterine segment by uterine contraction, an over distended urinary bladder or the presence of subchorionic hematoma overlying the cervix simulating placental tissue ( Bowie et al., 1978, Artis et al., 1985, Townsend et al., 1986 and Gallagher et al., 1987). To avoid false positive interpretation, both the lower edge of the placenta and the cervix must be visualized (Barbara et al., 1992).

Although TAS is usually successful in depicting the cervix during the first and second trimester of pregnancy, an additional technique is needed to complement TAS during the third trimester ( Brown et al., 1986). Techniques had been developed to evaluate the presenting part of the foots from the pelvis but this can be uncomfortable for the patient and are often unsuccessful late in the third trimester ( Jeffry and Laing, 1981).

Another approach was to use endovaginal sonography. Initial studies were promising, but endovaginal probe must be inserted and adjusted cautiously to avoid evoking an attack of bleeding in a patient with possible PP (Robert et al., 1990). TPS seems to be more convenient and possibly safer in imaging the cervix and lower uterine segment as vaginal penetration is avoided (Barbara et al., 1991).

Patients and Methods

This study was carried out on 200 patients attending with antepartum hemorrhage in the maternity clinic or emergency department in Benha University Hospitals. Twenty patients were missed during follow up and/or labor, so they have been excluded from the study and 180 patients were included in the results.

All patients were subjected to complete history taking, full general and abdominal examination. TAS and TPS were done for every patient using 3.5 MHz mechanical sector real-time scanner (Aloka, SSD 720 - Japan).

Transabdominal Scanning:

The examination was performed while the mother in the supine position with her urinary bladder partially filled so as not to distort the configuration of the cervix or lower uterine segment.

The gestational age was determined by measurement of the biparietal diameter (BPD) and femur length (FL) according to the criteria described by Handlock (1990).

The placenta was localized and considered praevia if placental tissue was overlying any part of the cervix. It is complete PP if placental tissues are covering the entire surface of the internal os, marginal when covering part of the cervix without encroaching on the internal os and partial when covering a portion of the internal os without covering the entire os. The last two grades were grouped together in data analysis as they are difficult to distinguish from each other by sonography. TAS was considered inadequate to exclude PP if the inferior edge of the placenta could not be imaged or the cervix could not be visualized.

Abruptio placenta was diagnosed if there is placental infarction, marginal necrosis or retroplacental blood clot as described by Nyberg et al., (1987).

Transperineal Scanning:

Ultrasound gel was placed on the head of the same abdominal transducer. A disposable condom was used as a protective covering applied to the transducer head secured with a rubber band and an ultrasound gel was applied to the other surface of to the protective covering. The examination was carried out while the patient lied supine with the thighs abducted sufficiently to allow placement and lateral angulation of the transducer. The transducer was placed directly over the perineum in a sagital orientation between the labia majora, usually directly over the labia minora, but occasionally between them. The center of the transducer was located posterior to the urethra and anterior to the vagina (Fig. 1). When the cervix and lower uterine segment were visualized, the transducer was slowly angled medially and laterally to image the entire internal surface of the cervix. Criteria for diagnosis of PP were the same as in transabdominal scanning.

PP was excluded if at least one of the following was present:

(1) The lower edge of the placenta seen separate from the cervix.

(2) Amniotic fluid between the presenting part and the cervix without interposed placental tissue.

(3) The presenting part immediately overlying the cervix without intervening placental tissue.

One hundred and fifteen cases were delivered at 36-40 weeks gestation either by cesarean section (48) or by normal vaginal delivery (67). The indication for caesarian section was severe attack of vaginal bleeding either because of PP (40) or abruptio placentae (8) cases. The other 65 patients were followed up by conservative management and repeated ultrasonography till delivery. Sixteen of those patients were delivered by cesarean section and 49 by vaginal delivery. The indication for cesarean section was PP. Absence of PP was confirmed at delivery by absence of placental tissue overlying the cervix at cesarean section or if successful vaginal delivery was accomplished without abnormal vaginal bleeding attributable to PP ( Barbara et al., 1992).

Results

The study included 180 patients of mean age 28± 4.2 (S.D) years and of mean gestational age 36.8± 5.5 (S.D.) weeks. Fifty-six patients had PP (G1) while the remaining 124 patients had no PP at delivery (G2). Data analysis of the later group revealed that 68 patients had abruptio placentae diagnosed by TAS only as TPS has afield limited to the lower uterine segment, cervix and vagina. Eight patients had a local cause of bleeding detected by speculum examination after exclusion of PP, while in 47 patients no definite cause of bleeding could be identified. Table 1 shows no significant difference in the clinical criteria of both groups (G1&G2).

TPS successfully depicted the internal surface of the cervix in all our 180 patients, allowing determination of the presence or absence of overlying placental tissue. While TAS was inconclusive for PP in 128 patients (71.1%), because the cervix was obscured by the presenting part of the foots. The difference between the two methods in visualization of the internal surface of the cervix was statistically very highly significant (P<0.001) (Table 2).

The inferior edge of the placenta was seen in 164 patients (91.1%) by TAS versus 57 patients (31.6%) by TPS. The difference was statistically highly significant (P<0.01) (Table 3).

TPS excluded PP in 122 patients (67.6%) and showed it in the remaining 58 patients (32.3%) (Table 4). The exclusion of PP was based upon : (1) No intervening placental tissue between the presenting part and the cervix in 69 patients (56.5%). (5) Visualization of amniotic fluid between the presenting part and the cervix without interposed placental tissues in 53 patients (43.5%). Delivery records confirmed the absence of PP in the all-122 patients.

Twenty-nine of the 58 patients in whom TPS showed PP had complete PP and 29 had marginal partial PP. There was no significant difference (P>0.05) between delivery outcome and TPS diagnosis of PP (Table 4).

TAS was suggestive of PP in 59 patients, 33 (18.3%) of them had a complete PP and 26 (14.4) had marginal or partial PP. There was no significant difference (P>0.05) between delivery outcome and TAS diagnosis (Table 5).

The overall accuracy of TPS was 97.7% versus 93.8% for TAS (Table 6).

Discussion

Sonography for patients with third trimester vaginal bleeding requires sufficient visualization of both the inferior edge of the placenta and the cervix to exclude or verify PP as a cause of such bleeding ( Barbara et al,. 1992). In this study, the internal surface of the cervix was clearly visualized in all cases (100%) by the perineal approach while it was seen in only 20% (36/180) of cases by the abdominal approach. Many authors agree with these results (Barbara et al., 1991, Barbara et al., 1992 and Ricky et al., 1995 ). Although Barbara et al., (1991), reported a potential difficulty with TPS that is the presence of rectal gas obscuring the region of the internal os. However, this dose not pose a significant problem when evaluating our patients because it is the internal surface of the cervix which is important when imaging is looking for overlying placental tissue.

In our study, the inferior edge of the placenta was visualized in 91% of cases using transabdominal technique, while it is visualized in 31.6% only by transperineal approach. These agree with the results of Barbara et al., (1992). Their explanation was that the lower edge of the placenta is usually beyond the field of view of transperineal approach.

Dietz et al., (1991), described a comparison between TAS and TPS for localization of low-lying placenta. They found significant (P<0.001) discrepancies between the two methods. Perineal scanning showed a higher rate of diagnosis (32%) than abdominal scanning (11%). They explained this difference by the influence of bladder distention with the use of abdominal scanning. An analysis of pregnancy outcome in their results yielded a positive predictive value of 78% with abdominal ultra sound and 86% with perineal ultrasound. Our study showed higher rates of accuracy with positive predictive value of 88.1% for abdominal ultrasound and 94.8% for transperineal scanning. This may be attributed to the fact that Dietz and his colleagues used a fall bladder technique with abdominal scanning and the patients were examined during the second and third trimesters.

Zilianti et al., (1991) described transperineal sonographic findings in a variety of clinical conditions, only 20 of their 184 patients were examined for possible PP. In addition, the authors did not attempt to evaluate the various grades of PP as our study did. In spite of that, we agreed with their conclusion in that TPS is useful in evaluation of possible PP.

Conclusion

TPS is a safe, accurate and rapid technique to complement TAS for evaluation of patients with third trimester bleeding. It has patient’s acceptance and tolerance, without the need for vaginal penetration or manipulation of the cervix, which is very important in such cases.

This technique is important in obstetric practice and needs further study to validate its potential application in other obstetric problems as premature rupture of membranes, cord presentation and cervical incompetence.

References

1. Artis, A.A., Bowie, J.D., Rosenberg, E.R., and Rauch, R.F.: The fallacy of placental migration: effect of sonographic techniques. Am J Rad; 144: 79,1985.
2. Barbara, S.H., James, D.B., Therese, M.W.,Barbara, A.C., Mark, A.K., and Sheryl, G.J. : Sonography of the cervix during the third trimester of pregnancy, value fo transperineal approach. Am J Rad; 157: 73, 1991.
3. Barbara, S.H., James, D.B., Barabra, A.C., Mark, A.C., and Therse, M.W. : Diagnosis of placenta praevia during the third trimester:Role of transperineal sonography. Am J Rad 159: 83,1992.
4. Bowie, J.D., Rochester, D., and Dobkin, A.V.: Accuracy of placental localization by ultrasound. Radiology; 128: 177,1978.
5. Brown, J.E., Thieme, G.A., Shah, M., Fleischer, A.C., and Poehm, F.H.: Transabdominal and transvaginal endosonography: evaluation of the cervix and lower uterine segment in pregnancy. Am J Obstet Gynecol: 155: 721,1986.
6. Dietz, H.P., Boos, R., Gnirs, J., and Schmidt, W. : Vergelich der perineal sonographie und der abdominalen. Ultrasch alluntersuchung bei placenta praevia. Geburtsh U Frauenheilk. 51: 357,1991.
7. Gallagher, P., Fagan, C.J., Bedi, D.G., Winsett, M.Z., and Reyes, R.N.: Potential placenta praevia: Definition, Frequency, And Significance. Am J Rad 149: 1013,1987.
8. Hadlock, F.P.: Sonographic estimation of fetal age and weight. Radiol Clin N A 28: 39,1990.
9. Jeffery, R.P. and Laing, F.C.: Sonography of the low-lying placenta: value of trendelenberg and traction scans. Am J Rad 137: 547,1981.
10. Nyberg, D.A., Mack, L.A. and Benedetti, T.J. : Placental abruption and placental hemorrhage: Correlation of sonographic findings with foetal outcome. Radiol 164: 357,1987.
11. Richey, S.D., Kirk, D.R., Scott, W.R., Susan, M.R., Susan, M.C., and Diane M.T. : The correlation between transperineal sonography and digital examination in the evaluation of the third trimester cervix. Obstet Gynecol 85 (5): 745,1995.
12. Robert, A.L., Erik, C.A., Marinus, J.C., and Jurity, W.W.: Accuracy and safety of transvaginal sonographic placental localization Obstet Gynecol 76: 729,1990.
13. Towsend, R.R.: Technical factors responsible for placental migration. Sonographic assesment. Radiol 160: 105,1986.

 

Table (1) Clinical Criteria of Placenta Praevia Group (G1) and Non Placenta Praevia Group (G2)

Clinical Criteria	G1 n=56	G2 n=124	P
	Mean± SD (SEM)	Mean ± SD (SEM)
Age	28.67 ± 3.68 (0.50)	27.82 ± 4.52 ( 0.39)	N.S.
Gestational age	38.7 ± 1.96 (0.28)	37.3 ± 1.87 (0.24)	N.S.
Parity	2.83 ± 1.6 (0.15)	2.72 ± 1.49 (0.11)	N.S.
Gravidity	3.7 ± 2.1 (0.21)	2.9 ± 1.91 (0.19)	N.S.
Fundal level	38.4 ± 1.94(0.26)	36.02 ± 2.64 (0.24)	N.S.
B.P.D	8.93 ± 0.54 (0.03)	8.58 ± 0.78 (0.07)	N.S.
F.L	7.4 ± 0.32 (0.04)	7.02 ± 0.47 (0.04)	N.S.

P> 0.05

Table (2) Comparison Between TPS and TAS in Visualization of Internal Surface of the cervix

	TPS		TAS
	V	NV	V	NV
N (%)	180 (100%)	0 (0%)	52 (28.8%)	128 (71.1%)
Mean± SD	0.11± 0.0		0.20± 0.4
T	26.7582
P	Very highly significant

P< 0.001
V= Visualized
NV= Not visualized

 Table (3) Comparison between TPS and TAS in Visualization of Inferior Edge of the Placenta

	TPS		TAS
	V	NV	V	NV
N(%)	57 (31.67%)	123 (68.33%)	164 (91.11%)	16 (8.89%)
Mean± SD	0.3167± 0.4665		0.911± 0.2854
T	14.5843
P	Highly significant

P< 0.01
V= Visualized
NV= Not visualized

Table (4) Validity of TPS in Diagnosis of PP

P> 0.05
G1= Non PP. G2= PP. CPP=Complete PP. (PorM)PP= Partial or Marginal PP.

Table (5) Validity of TAS in Diagnosis of PP

P> 0.05
G1= Non PP. G5= PP. CPP= Complete PP. (PorM)PP= Partial or marginal PP.

Table (6) Comparison between Accuracy of TPS and TAS in Diagnosis of PP

	TPS	TAS
Sensitivity	98.21%	92.85%
Specificity	97.58%	94.35%
Positive predictive value	94.82%	88.1%
Negative predictive value	99.18%	96.69%
Accuracy	97.77%	93.88%