The assessment of abnormal pelvic blood flow by transvaginal color and pulsed Doppler

Ultrasound in Med & Biol. Vol. 16, No. 5, pp. 437-442, 1990 0301-5629/90 $3.00 + .00 Printed in the U.S.A. © 1990 Pergamon Press plc

OOriginal Contribution

T H E A S S E S S M E N T O F A B N O R M A L P E L V I C B L O O D F L O W B Y

T R A N S V A G I N A L C O L O R A N D P U L S E D D O P P L E R

ASIM K U R J A K , IVICA Z A L U D , 7,ARKO ALFIREVI(~ a n d D A V O R J U R K O V I ~ Department of Obstetrics and Gynecology, Medical Faculty, University of Zagreb, Yugoslavia

(Received 25 September 1989; in final form 12 December 1989)

Abstract--Transvaginal color Doppler was used to analyze a group of 19 healthy, 48 infertile and 8 postmenopausal women; 67 patients with uterine and 151 patients with adnexal masses, and 19 patients with a suspected ectopic pregnancy. The ultrasonographer had not been informed of other clinical findings and indications for operative treatment. A 5 MHz transvaginai color Doppler probe was used to visualize the pelvic anatomy and blood flow. Each arterial blood flow velocity was classified into either of two types. In the normal flow pattern, the diastolic flow component was absent or small and the value of resistance index (RI) was >0.50. In the abnormal flow pattern, a large diastolic flow component was evident and the value of RI was <0.50. Transvaginal color and pulsed Doppler studies in confirmed malignant uterine masses had moderate accuracy (90.2%), with a sensitivity (55.6%) and specificity (100%); and in confirmed malignant adnexal masses high accuracy (100%), sensitivity (100%) and specificity (100%). Comparison of blood flow characteristics within benign and malignant tumor tissue revealed less impedance and higher velocity in cases of malignancy. Blood flow studies in both uterine arterias have not shown clinically important values. This method can be also used in the diagnosis of ectopic pregnancy.

Key Words: Transvaginal sonography, Color Doppler, Pelvic circulation, Pelvic tumors, Ectopic pregnancy.

I N T R O D U C T I O N

E n d o s o n o g r a p h y has evo lved into an i m p o r t a n t imaging modal i ty which has significantly changed the profile o f diagnostic ul t rasound. The detect ion of space-occupying lesions was previously the ma jo r clinical problem, especially in the assessment of gynecological patients (Kurjak and Jurkovi~ 1987). Using the endovaginal approach the pelvic structures are now clearly defined and their characterizat ion has become the pr imary goal o f our recent investigations. The necessity for tissue characterization (malignant vs. benign lesion) in the assessment of gynecological patients correlates with the clinician's growing aware- ness of the diagnostic value of blood flow studies by Doppler ultrasound. The combinat ion of high quality B -mode images, pulsed wave D o p p l e r and color Doppler in the same vaginal probe allows simulta- neous visualization of structural and flow informa- tion constituting an exciting recent development.

We have already published our prel iminary data

Address correspondence to: Prof. Asim Kurjak, M.D., Ph.D., Department of Obstetrics and Gynecology, Medical Faculty, Uni- versity of Zagreb, Pavleka Migkine 64, 41.000 Zagreb, Yugoslavia.

with this modali ty showing that transvaginal color Doppler could be the imaging modali ty of choice in the evaluat ion of pelvic masses and ectopic pregnancy (Kurjak and Jurkovi6 1989). This paper is our final report on pelvic blood flow results obtained in 3 i 2 gynecological patients.

S U B J E C T S A N D M E T H O D S

Various groups o f subjects have been studied consisting of 19 healthy fertile women, 48 infertile and 8 postmenopausal women, as well as 67 patients with uterine and 151 patients with adnexal masses, and 19 patients with suspected ectopic pregnancies. In the subgroup of 94 patients with ultrasonically proven uterine and/or adnexal masses, transvaginal color Doppler studies were performed before laparoscopy and/or laparotomy. The ultrasonographer had not been informed about the clinical findings or indications for operative treatment, apart f rom the B- mode ultrasound finding and data of the last men- strual period. In this subgroup of patients with pelvic tumors, sonographic findings were compared to the operative and pathohistological diagnosis (final diagnosis).

437

438 Ultrasound in Medicine and Biology

A 5 M H z transvaginal color Doppler probe was used to visualize the pelvic ana tomy and blood flow. In addition to color Doppler, the machine (Aloka co lor D o p p l e r SSD 350, Aloka Co., J a p a n ) is equipped with a conventional pulsed Doppler system which can be used s imul taneous ly with B-mode imaging. The spatial-peak temporal average intensity was approximately 110 m W / c m 2. Estimated atten- uation by the maternal tissues reduced this to less than 50 m W / c m 2. Each examinat ion lasted less than 20 min.

Transvaginal color Doppler was used in an at- t empt to visualize blood flow in the major pelvic vessels, especially the uterine artery, and vessels within the uterine, ovarian or t umor tissue. Color signals from the both uterine arteries could be seen in all patients just lateral to the cervix at the level of the cervicocorporeal junct ion of the uterus. This method was used to visualize the regional flow pattern and to select an appropriate area for spectral analysis. Pe- ripheral impedance was then assessed using resistance or Pourcelot index (RI) by pulsed Doppler (Pourcelot 1974). This angle independent index is believed to be a good assessor of downstream vascular resistance. Each arterial blood flow velocity was classified into one of two types. In the normal flow pattern, the diastolic flow componen t was absent or small and the value of RI was >0.50. In the abnormal flow pattern, a large diastolic flow componen t was evident and the value of RI was <0.50. On each record, five separate cardiac cycles were examined to determine the mean RI. Signals from various areas of the same structures in each patient were recorded and the highest value of RI was used for further analysis. Student 's test was used for statistical analysis.

R E S U L T S

Blood flow in the uterine artery Both the left and the right uterine arteries were

visualized using transvaginal color Doppler in 312 patients. Artery and vein were distinguished accord-

Table 1. Resistance index (RI) in the uterine artery.

Uterine artery (RI)

Patients N Left Right

Healthy fertile 19 0.86 +_ 0.04 0.85 _+ 0.07 Infertile 48 0.85 _+ 0.06 0.86 + 0.05 Postmenopausal 8 0.87 _+ 0.09 0.89 _+ 0.04 Uterine mass 67 0.81 _+ 0.08 0.82 _+ 0.08 Adnexal mass 159 0.83 + 0.08 0.82 _+ 0.09 Ectopie pregnancy 11 0.77 +_ 0.09 0.73 _+ 0.08 Total 312

Volume 16, Number 5, 1990

Table 2. Blood flow in the uterine mass (N = 41).

Color flow

Not PHD N Present present RI a

Adenocarcinoma endometrii 5 5 0

Carcinoma cervicis in situ 4 0 4

Myoma uteri 29 6 23 Endometriotic cyst 3 0 3 Total 41 11 30

0.38 +_ 0.06

0.59 _+ 0.08

RI" = resistance index (mean and SD).

ing to pulsat ion and brightness of the color flow. More detailed analysis of blood flow in the uterine a r te ry was p e r f o r m e d using pulsed Dopple r . It showed flow velocity waveforms with high pulsatility and very low diastolic flow. Some differences in pe- r ipheral impedance a m o n g subgroups of pat ients were noticed (Table 1). However, a statistically significant difference could not be shown between the left and the fight uterine artery, even in patients with unilateral pelvic masses (t -- 0.56, p > 0.05). It is obvious that resistance index (RI) is the lowest in the group of patients with proven ectopic pregnancy (RI = 0.77 + 0.09 and 0.73 + 0.08) and the highest in postmenopausal women (RI = 0.87 _ 0.09 and 0.89 +_ 0.04) (t = 6.79, p < 0.01). Very similar results to postmenopausal women were found in healthy fertile women (RI = 0.86 + 0.04 and 0.85 + 0.07) and in- ferti le w o m e n with n o r m a l pelvic a n a t o m y (RI = 0.85 + 0.06 and 0.86 _+ 0.05) and, no significant difference was found between them. Lower RI was found in patients with uterine (RI : 0.81 + 0.08 and 0.82 + 0.08) and adnexal tumors (RI = 0.83 _+ 0.08 and 0.82 + 0.09) than in the group of healthy fertile women, but no statistically significant difference was again present.

Blood flow in pelvic tumors Within a subgroup of women with uterine and

adnexal tumors, 83 patients were scheduled for operative treatment. Pathohistological diagnoses in these patients are listed in Tables 2 and 3. There were 9 cases of uterine and 10 cases of adnexal malignancy, as well as 32 cases of benign uterine and 32 cases of benign adnexal tumor. Color flow inside the endo- met r ium and m y o m e t r i u m was detected in all cases with endometr ial carcinoma. The peripheral impedance was very low (RI = 0.38 + 0.06).

In 4 cases of cervical carc inoma we did not detect any abnormal i ty of blood flow. It can be specu- lated that newly formed vessels are too small and the

Pelvic blood flow by transvaginal color and pulsed Doppler • A. KURJAK et al.

Table 3. Blood flow in the adnexal mass (N = 42).

439

PHD

Color flow

N Present Not present RI a

Cystadenocarcinoma serosum 4 4 0 Cystadenocarcinoma mucinosum 3 3 0 Metastatic follicular Ca of the thyroid gland 1 1 0 Metastatic rectal Ca 1 1 0 Granulosa cell tumor 1 1 0 Dermoid cyst 4 0 4 Simple ovarian cyst 15 0 15 Hydrosalpinx 7 1 6 Cystadenoma serosum 2 0 2 Cystadenoma mucinosum 1 0 ! Cystadenoma papilliferum 1 0 1 Endometriosis 2 0 2 Total 42 11 31

0.34 _+ 0.09 0.35 _+ 0.06

0.25 0.30 0.45

0.71

RI a - resistance index (mean and SD).

velocity and volume flow are below the resolution power of the equipment . In 6 out of 29 cases with myoma , color flow was detected in the border of the tumor . The velocities of blood flow in the m yomas were high and similar to the flow in the uterine artery (30-50 cm/s). However, when the peripheral impedance was compared in cases of m y o m a uteri and car- c inoma endometri i , significantly lower RI was obtained in cases of carc inoma endometri i (t = 6.90, p < 0.01). T ransvag ina l color and pulsed Dopp le r s tudy in conf i rmed mal ignan t uterine masses had moderate accuracy (90.2%), sensitivity (55.6%) and specificity (100%) (Table 4). This was due to the un- detected color flow (false-negative finding) in all 4 cases of cervical carcinoma. In 23 cases of uterine m y o m a and 3 cases ofendometr io t ic cyst, blood flow could not be seen inside the mass.

In all cases of adnexal malignancy, color flow was detected inside the tumor. Newly formed vessels showed very p r o m i n e n t blood flow with high vel- oci ty ( 2 0 - 6 0 cm/ s ) and very low res is tance (RI = 0 .25-0 .45) . D e m o n s t r a t i o n of smal l vascu la r branches inside the solid part of the t umor was a regular finding (Figs. 1 and 2). This flow pattern was

Table 4. Evaluation of diagnostic criteria for the assessment of uterine mass (N = 41).

Final diagnosis

Color flow and RI Malignant Benign

Abnormal flow pattern 5 0 Normal flow pattern 4 32 Total 9 32

PPV = 100%; NPV = 88.9%; Sensitivity = 55.6%; Specificity = 100%; Accuracy = 90.2%.

less obvious only in one case of metastatic follicular carc inoma of the thyroid gland, but the RI fell within the expected range. Vascular izat ion of the t u m o r could not be detected in benign adnexal masses, with the exception of one case of hydrosalpinx with flow seen at the border of the cystic structure. However, the peripheral impedance in that case was significantly higher (RI = 0.71) than in malignant masses (t = 7.28, p < 0.01). In 7 patients with moderately enlarged cystic ovaries and who did not undergo opera- tion, blood flow was detected inside the ovary with decreased resistance (RI = 0.52 + 0.04) and no prom- inent color . The e x a m i n a t i o n was repea ted one month later in the proliferative phase of the men- strual cycle and no flow was observed. The same findings were found in ovaries with a corpus luteum cyst. The findings in both groups of patients were not con- sidered as pelvic pathology and therefore they were not included in the analysis. Waveform analysis of signals obtained from "ho t " color coded areas within adnexal masses showed excellent results (Table 5). All malignant tumors had RI less than 0.50 and all benign tumors had RI higher than 0.50. Accuracy of this method was 100%, as well as sensitivity and specificity. Color flow and RI of 0.50 is used as a cut-off poin t to differentiate mal ignant vs. benign lesion. Using such criteria, 7 pr imary ovarian malignancies were diagnosed (1 at stage I-A, and 6 at stage III).

Ectopic pregnancy Nineteen patients with suspected ectopic preg-

nancy were examined (Table 6). Blood flow within pelvic masses was detected using color Doppler in 10 patients (Figs. 3 and 4). Laparoscopy and/or laparotomy confirmed an ectopic pregnancy in 11 out of 19 patients. Two false-negative color flows occurred in

440 Volume 16, Number 5, 1990 Ultrasound in Medicine and Biology

Table 5. Evaluation of diagnostic criteria for the assessment of adnexal mass (N = 42).

Final diagnosis

Color flow and RI Malignant Benign

Abnormal flow pattern 10 0 Normal flow pattern 0 32 Total 10 32

PPV - 100%: NPV - 100%: Sensitivity - 100%: Specificity - 100%; Accuracy - 100%.

cases of tubal abortion and very likely with an inac- tive trophoblast. One false-positive finding was seen in the case of a corpus luteum cyst. Waveform analysis showed moderate velocity of blood flow (12-35 cm/s) and very low peripheral resistance with increased diastolic flow within pelvic masses. In all cases (including one false-positive case) the RI was lower than 0.50. T ransvag ina l color and pulsed Doppler study of flow within ectopic pregnancies had high accuracy (84.2%), sensitivity (81.8%) and specificity (87.5%).

D I S C U S S I O N

Many clinical problems remain in differentiating benign and malignant tumors in vivo. However, there are known features that characterize this difference. Many of them, such as mitotic index and pleomor- phism, are histologic and are not amenable to current imaging methods. One feature of malignancy is the bizzare vascular morphology that characterizes many malignant tumors. Flow in these abnormal channels gives rise to characteristics that can be detected by Doppler techniques.

T u m o r vessels have a relative paucity of smooth muscle in their walls in comparison with to their cali- ber (Gammil l et al. 1976). Due to their lack of normal muscle components , the amount of vasoconstriction by sympathetic stimulation cannot be increased. This could be the basis for the differentiation between benign and m a l i g n a n t t u m o r vascula r i ty ( A b r a m s 1982).

The importance of t umor vascularity was first recognized by Judah Folkman in 1972, when he pro- posed the hypothesis that increased cell population must be preceded by the production of new vessels (Folkman 1972). These vessels must be formed very early on in the development of the t umor and therefore are early markers of the presence of a tumor. Several substances are identified as angiogenic factors (Folkman et al. 1971). Such factors are secreted by the t umor and induce the formation of new vessels by

the host tissue. As noted in morphologic studies, these vessels have abnormal vascular morphology, including the f requent presence of a r t e r iovenous anastomoses and vascular walls deficient in muscular elements. The lack of muscular elements is reflected in low impedance to flow leading to high diastolic flow and, in some tumors, the absence of systolic/diastolic variation.

It seems that Doppler studies and transvaginal color Doppler sonography can produce better characterization of pelvic t umor vascularity (Taylor et al. 1985, 1988; Taylor and Morse 1988; Kurjak et al. 1988, 1989a). Our own results in the group of patients with pelvic masses have shown that it is easy to demonstra te blood flow in small vascular branches within the tissue in pathological conditions. These vessels are detectable by ultrasonic equipment due to their higher blood velocity and consequently the increased volume of the flow. Being very thin and ran- domly dispersed within the tissue, such vessels are difficult to find unless color Doppler is used.

Our own study has shown that t umor vascularity can be successfully used for the characterization of pelvic tumors. Sensitivity and accuracy are accept- ably high and it seems that we'should be more con- cerned about false-negative than about false-positive findings.

The goal of transvaginal sonography should be to identify tumors in ovaries that are not significantly enlarged (Fleisher 1988). Transvaginal color Doppler may be sufficiently sensitive for early detection of ovarian cancer which makes up about 25% of all gy- necologic ma l ignanc ie s (Si lverberg and Lube ra 1988). Despite advances in the t reatment and management of the disease, the overall five-year survival rate remains at 20% to 30%. This is primarily the result of the advanced stage of the disease when detected. Preoperative diagnosis of ovarian carcinoma in pos tmenopausa l women is frequently based on palpation of a pelvic mass with or without a fluid wave. However, adnexal assessment of size and con- sistency during the pelvic examinat ion is very subjec- tive and often imprecise. Campbell et al. (1982) pre-

Table 6. Ectopic p r e g n a n c y (N = 19).

Ectopic pregnancy

Color flow Confirmed Not confirmed

Present 9 1 Not present 2 7

PPV - 90%; NPV 77.8%: Sensitivity - 81.8%; Specificity 87.5%; Accuracy - 84.2%.

Pelvic blood flow by transvaginal color and pulsed Doppler • A. KURJAK et al. 441

Fig. 1. Increased vascularity of solid adnexal tumor as demonstrated by transvaginal color Doppler.

sented the first encouraging results from a screening program for ovarian cancer using an ultrasound tech- nique. Other authors tried to assess the efficacy of transvaginal (Rodriguez et al. 1988) or t ransabdomi- nal sonography in determining the size and morphologic findings of the postmenopausal ovary (Andolfe t al. 1986; Rulin and Preston 1987; Goldstein et al. 1989).

Our present exper ience has shown that ultrasound examina t ion by a t ransvaginal p robe com- bined with pulsed and color Doppler assessment of the pelvic circulation may increase the reliability of ultrasound diagnosis in certain pathological conditions (Kurjak et al. 1989c). It is our belief that only color Doppler visualization and waveform analysis of blood flow can be of significant help in distinguishing benign and malignant tumors.

Fig. 3. Adnexal mass suspected to be an ectopic pregnancy.

Our study suggested that increased tumor vascularization and the lower RI can be used as evidence of neovascularity in pelvic tumors. The rate of accuracy is very high in cases of adnexal malignancy (100%) but not so high in cases o f u te r ine m a l i g n a n c y (90.2%) because o f the imposs ibi l i ty o f detect ing color flow in cervical carcinoma.

Color Dopp le r signals and addi t ional pulsed Doppler analysis support the clinician's suspicion of a malignant or benign tumor. All cases in which a malignant pelvic t umor was confirmed showed a lower RI than 0.50. Contrary to this, none of the confirmed benign t u m o r s showed such a RI. Benign pelvic tumors predominant ly showed no vascularization inside the t umor tissue which could be detected using t r ansvag ina l co lor Dopp le r . Howeve r , i f co lor Dopp le r flow was detected, addi t iona l wave fo rm

Fig. 2. High velocity and low resistance waveform (right) that was obtained from newly formed vessels within tumor

tissue.

Fig. 4. The same patient examined by transvaginal color Doppler. Color flow indicated ectopic trophoblast and pulsed Doppler analysis (right) showed typical low resis-

tance flow.

442 Ultrasound in Medicine and Biology Volume 16, Number 5, 1990

analysis showed much higher resistance than in malignant cases (RI > 0.50).

Study of flow may have more immediate applications in the diagnosis of ectopic pregnancy. Sono- graphic findings of an ectopic pregnancy include an empty uterus and an adnexal mass. However, these are nonspecific features, which can occur in abortion and early normotopic pregnancies, as well as in non- gravid women. Color and pulsed Doppler can help to characterize the nature of the adnexal mass thus en- abling preoperative diagnosis when an ectopic fetus and its characteristic heartbeat cannot be visualized. Transvaginal sonography enables identification of the ectopic trophoblast and enables differentiation from flow in the corpus luteum. In most individual patients it was possible to separate a lower velocity luteal flow in the ovary from a higher velocity signal, usually situated more medially and within the tube (Kurjak et al. 1989b; Taylor et al. 1989). The fetus invades the maternal tissue, and it is possible to get very high blood flows from the maternal arteries into the spaces around them. It is therefore worth using this vascular signature of trophoblastic flow as a way of tissue characterization of the adnexal mass. Arte- rial flow produced by trophoblastic invasion of the maternal tissue, would account for the high-velocity, low-impedance signals.

To sum up, our study has shown that transvaginal color Doppler is a very promising tool in tissue characterization of pelvic masses. Because of its high accuracy, sensitivity and specificity, it is the imaging of choice in diagnostic procedures. Blood flow studies in both uterine arterias have shown certain variations depending on pelvic pathology, although it seems that these data are not of great clinical value. Our results have stressed the importance of blood flow visualization within the pelvic mass. Using transvaginal color Doppler and additional waveform analysis, it is possible to obtain important data which can help c l in ic ians to differentiate benign and mal ignant tumors and ectopic pregnancy at the earliest possible stage.

REFERENCES

Abrams, H. L Renal tumor versus renal cyst. In: Angiography. Boston: Little-Brown; 1982:1123-1124.

Andolf, E.; Svalenius, E.; Astedt, B. Ultrasonography for early detection of ovarian carcinoma. Br. J. Obstet. Gynaecol. 93:1286-1289; 1986.

Campbell, S.; Goessons, L.: Goswamy, R. K.; Whitehead, M. Real-time ultrasonography for determination of ovarian morphology and volume: A possible early screening test for ovarian cancer? Lancet I:425-426; 1982.

Fleisher, A. C. Transvaginal sonography helps find ovarian cancer. Diagn. Imag. 10:124-128; 1988.

Folkman, J. Anti-angiogenesis: New concept for therapy of solid tumors. Ann. Surg. 175:409-416; 1972.

Folkman, J.; Nerler, E.; Abernathy, C.; Williams, G. Isolation of a tumor factor responsible for angiogenesis. J. Exp. Med. 33:275-280; 1971.

Gammill, S. L; Shipkey, R. B.: Himmelfarb, E. H. Roentgenol- ogy-pathology correlative study of neovascularity. A JR 126:376-385; 1976.

Goldstein, S. R.; Subramanyam, B.; Snyder, J. R.; Belier, U.: Raghavendra, B. N.; Beckman, E. M. The postmenopausal cystic adnexal mass: The potential role of ultrasound in conser- vative management. Obstet. Gynecol. 73:8-10; 1989.

Kurjak, A.; Alfirevic, Z.; Miljan, M. Conventional and color Doppler in the assessment of fetal and maternal circulation. Ultrasound Med. Biol. 14:337-354; 1988.

Kurjak, A.; Jurkovic, D. The value of ultrasound in the initial assessment of gynecological patient. Ultrasound Med. Biol. 13:401-419; 1987.

Kurjak, A.; Jurkovic, D. Transvaginal color Doppler in the assessment of pelvic massess. 6th World Congress In Vitro Fertiliza- tion and Alternate Assisted Reproduction. Jerusalem (Book of abstracts, p. 26); 1989.

Kurjak, A.: Jurkovic, D.; Alfirevic, Z.; Zalud, I. Transvaginal color Doppler. J. Clin. Ultrasound 1989a.

Kurjak, A.: Miljan, M.; Jurkovic, D.; Alfirevic, Z.; Zalud, 1. Color Doppler in the assessment of fetomaternal circulation. Rech. Gynecol. 1:269-274:1989b.

Kurjak, A.; Zalud, I.: Jurkovic, D.; Alfirevic, Z.; Miljan, M. Trans- vaginal color Doppler in the assessment of pelvic circulation. Acta Obstet. Gynecol. Scand. 68:131-135; 1989c.

Pourcelot, L. Applications cliniques de l'examen Doppler transcu- lane. In: Peronneau, P., ed. Velocimetrc ultrasonore Doppler 34. Paris: Seminare INSERM; 1974:213-240.

Rodriguez, M. H.; Platt, L. D.; Medearis, A. L.; Lacarra, M.; Lobo, R. A. The use of transvaginal sonography for evaluation of postmenopausal ovarian size and morphology. Am. J. Obstet. Gynecol. 159:810-814; 1988.

Rulin, M. C.; Preston, A. L. Adnexal masses in postmenopausal women. Obstet. Gynecol. 70:578-581; 1987.

Silverberg, E.; Lubera, J. Cancer statistic. CA 38:9: 1988. Taylor, K. J. W.; Burns, P. N.; Wells, P. N. T.: Conway, D. 1.; Hull,

M. G. R. Ultrasound Doppler flow studies for the ovarian and uterine arteries. Br. J. Obstet. Gynaecol. 92:240-246; 1985.

Taylor, K. J. W.; Morse, S. S. Doppler detects vascularity of some malignant tumors. Diagn. imag. 10:132-136; 1988.

Taylor, K. J. W.; Ramos, I.; Carter, D.; Morse, S. S.; Snower, D.; Fortune, K. Correlation of Doppler US Tumor Signals with neovascular morphologic features. Radiology 166:57-62; 1988.

Taylor, K. J. W.; Ramos, I. M.; Feyock, A. L.; Shower, D. P.; Carter, D.; Shapiro, B. S.; Meyer, W. R.; DeCherney, A. H. Ectopic pregnancy: Duplex Doppler evaluation. Radiology 173:93-97; 1989.

The assessment of abnormal pelvic blood flow by transvaginal color and pulsed Doppler

Documents

Transcript of The assessment of abnormal pelvic blood flow by transvaginal color and pulsed Doppler