American Journal of Emergency Medicine (2008) 26, 81 - 85

Diagnostics

ED endovaginal pelvic ultrasound in nonpregnant women with Right lower quadrant pain

Vivek S. Tayal MD^a,*, Mark Bullard MD^a, Doug R. Swanson MD^a, Christian J. Schulz PA^a,

Katrina N. Bacalis PA^a, Susan A. Bliss MD^b, H. James Norton PhD^c

^aDepartment of Emergency Medicine, Carolinas Medical Center, Box 32861, Charlotte, NC 28232, USA ^bDepartment of Obstetrics and Gynecology, Carolinas Medical Center, Charlotte, NC 28232, USA ^cDepartment of Biostatistics, Carolinas Medical Center, Charlotte, NC 28232, USA

Received 29 January 2007; revised 9 February 2007; accepted 9 February 2007

Abstract

Introduction: We hypothesized that emergency physician-performed endovaginal ultrasound (EVUS) would change diagnostic decision making in nonpregnant women with right lower quadrant (RLQ) pain. Methods: A prospective cohort of female patients was enrolled at an urban emergency department (ED). Inclusion criteria were RLQ pain, hemodynamic stability, and a strong suspicion for appendicitis or right adnexal pathology. Treating physicians were queried regarding pre- and post-ED EVUS probability of disease, differential diagnoses, consultation, and management. Positive findings included large cysts or multitissue densities, tubal dilation, uterine enlargement/mass, and extensive peritoneal fluid.

Results: With a positive ED EVUS, mean physician probability increased for gynecologic (24%) and decreased for both surgical (14%) and medical (20%) disease. With a negative ED EVUS, mean physician probability increased for surgical disease (5.3%) and decreased for gynecologic disease (18.6%).

Conclusion: Emergency department EVUS changes physician diagnostic decision making in nonpregnant women with undifferentiated RLQ pain.

D 2008

Introduction

Nonpregnant female patients with acute abdominal pain have traditionally required a lengthier and more extensive evaluation because of the different nature of their anatomy and pathologic processes. With the reproductive tract’s

Presented at the annual meeting of the Society of Academic Emergency Medicine in New York, NY, in May 2005.

* Corresponding author. Tel.: +1 704 355 3181; fax: +1 704 355 7047.

E-mail address: [email protected] (V.S. Tayal).

adjacent location to the abdominal cavity, referred pain from multiple spaces and organs can be referred to the right lower quadrant (RLQ). In addition to the traditional history, physical examination, and laboratory evaluation, other management actions that may be considered include surgical and gynecologic consultation, sonographic evalu- ation, computed tomography (CT), and immediate surgical procedures [1].

After observing this prolonged, extensive, and con- suming pattern of evaluation in many of our nonpregnant female patients with RLQ abdominal pain, we suspected

0735-6757/$ - see front matter D 2008 doi:10.1016/j.ajem.2007.02.029

3. Study protocol

Table 1 Study population clinical characteristics

CI indicates confidence interval.

Demographic	Percentage with 95% CI	No. of patients
History
Prior history of sexually	5 (2-17)	2
transmitted disease
Prior abdominal surgery	20 (11-35)	8
Prior gynecologic surgery	43 (28-58)	17
Prior pregnancy	80 (65-89)	32
Prior cesarean delivery Examination	10 (4-23)	4
RLQ abdominal tenderness	100 (91-100)	40
Deep abdominal tenderness	57.5 (42-72)	23
Rebound	10 (4-23)	4
Rovsing sign	7.5 (3-20)	3
Heel tap	5 (2-17)	2
Obturator	7.5 (3-20)	3
Vaginal discharge	10 (4-23)	4
Cervical motion tenderness	10 (4-23)	4
Fundal tenderness	15 (7-29)	6
Left adnexal tenderness-	5 (1-17)	2
pelvic exam
Right adnexal tenderness-	62.5 (47-76)	25
pelvic exam

Patients underwent history and physical examination including pelvic examination. Treating emergency physi- cians included emergency medicine faculty physicians, emergency medicine residents, and emergency medicine physician assistants. These providers were queried both pre- and post-ED EVUS on the following: probability of surgical, gynecologic, and medical disease (1-100 scale); need for gynecology, surgical, and Medical consultations; further necessary imaging needed (ultrasound from gyne- cology or radiology, CT noncontrast abdomen/pelvis, CT with intravenous and oral contrast abdomen/pelvis); and need for immediate operative therapy. Treating providers were also asked to list their top 5 differential diagnoses with graded weight (5 points assigned to highest, 1 point assigned to lowest).

that our sonographic skills with first-trimester pelvic ultrasound could be used to address the issue of contrib- utory pelvic pathology [1-3]. We hypothesized that the use of endovaginal ultrasound (EVUS) by emergency physicians would change diagnostic decision making in the evaluation of adult nonpregnant female patients with RLQ pain.

Methods

This study was a prospective, interventional convenience sample of nonpregnant adult women with significant right lower pain performed at an urban, regional emergency department (ED) with annual volume of greater than 100000 over a period of 21 months. This study was approved by the institutional review board, and informed consent was obtained from each patient.

Inclusion criteria included the following: age N17 years, RLQ pain at or near the McBurney point, hemodynamic stability (systolic blood pressure N90 mmHg, heart rate b140 beats/min, respiratory rate b40 breaths/min), no other clearly identifiable diagnosis (eg, pyelonephritis, diverticu- litis, Pelvic inflammatory disease); and both appendicitis and adnexal pathology were strongly considered in the phys- ician’s differential diagnosis. Exclusion criteria included the following: pregnancy, recent gynecologic procedure (within the last 2 weeks), recent abdominal surgical procedure (within the last 2 weeks), history of right oophorectomy/ salpingectomy, or history of appendectomy.

Ultrasound protocol

A preselected group of 3 emergency physicians and

2 physician assistants with prior credentialing in first- trimester pelvic ultrasound was trained in nonpregnant EVUS by the emergency medicine ultrasound director and the institutional gynecologic director of ultrasound. The 5 pro- viders were already familiar with the techniques of pelvic ultrasound, so the educational objectives were focused on standard organ measurements of the female pelvis and abnormalities of the nonpregnant female reproductive tract. This training included 2 hours of lecture and hands-on proctoring. Each sonologist was required to perform 5 train- ing ultrasounds, which were reviewed by both ultrasound directors before official patient enrollment in the study.

The following pelvic ultrasound protocol was used after informed consent. A Shimadzu 450XL (Kyoto, Japan) with a 5- to 8-MHz endovaginal transducer was used for all ultrasound examinations. Coronal and sagittal scans of the uterus and ovaries were performed, and measurements in 3 dimensions were recorded. Volumes were determined by the prolate ellipsoid formula (length x height x width x 0.523). Abnormalities of the uterus and adnexa were noted. The cul-de-sac was surveyed for fluid. Significant gyne- cologic findings, as reached by consensus between the 2 ultrasound directors, were chosen on the basis of their diagnostic implications as follows: a large cystic structure N4 cm in any one dimension, multitissue density structure, tubal dilation, uterine enlargement or mass, or peritoneal fluid past the uterine body.

Demographic data were obtained from the patient’s medical record.

Statistics

Descriptive statistics, including means and standard deviations, or counts and percentages, were calculated to determine the change in pretest probability of disease

Table 2 Comparison of gynecologic ultrasound findings
	ED EVUS positive	ED EVUS negative	All patients
No. of patients	12	28	40
Uterus length, mean (cm)	9.1 F 1.2	8.0 F 1.7	8.31 F 1.7
Uterus height, mean (cm)	5.0 F 1.6	3.8 F 1.0	4.2 F 1.3
Uterus width, mean (cm)	6.2 F 1.3	4.9 F 1.3	5.2 F 1/4
Right ovarian volume, mean (cm3)	31.6 F 26.5	9.6 F 6.0	15.5 F 17.3
Left ovarian volume, mean (cm3)	20.7 F 31.9	8.5 F 5.3	10.8 F 7.7
Large cystic mass	9	0	9
Multitissue density	8	0	8
Tubal dilation	4	0	2
Uterine enlargement	2	0	4
Significant peritoneal fluid	3	0	1

stratified by presence or absence of ED EVUS findings. Pre- and posttest probabilities were compared using Wilcoxon Signed Rank Tests. A P value of less than 0.05 was considered statistically significant. The SAS (Cary, NC) software was used for all analyses.

Results

Forty patients were enrolled, with a mean age of 34 years (SD = 10, range 18-73) and with ethnic identification of 50% African American, 38% white, 10% Hispanic, and 2% Asian. These 40 patients have the following mean vital signs: temperature 98.28F (SD = 1.4, range 96.6-103.7), pulse 91 (SD = 14.9, range 68-145), and systolic blood pressure 124 mm Hg (SD = 14, range 104-154). Table 1 reflects the study population’s historical and physical examination characteristics.

Pelvic EVUS resulted in normal findings in most cases. Ovarian cysts and uterine fibroids were the most common abnormalities (Table 2). Twelve patients had significant

Fig. 1 Change in physician probability regarding different disease categories stratified by ultrasound findings.

pelvic findings as defined above. In comparison of patients with significant pelvic findings (ED EVUS positive) and those without significant pelvic findings (ED EVUS nega- tive), there was no significant difference in uterine measure- ments; however, mean ovarian size differed significantly between ED EVUS groups (right ovarian size P b .001, left ovarian size P b .001).

In patients with a positive ED EVUS for significant gynecologic findings, there was a mean decrease of 14.2% ( P = .002) in the physician’s perceived probability for surgical disease and 20.2% ( P = .001) for medical disease, whereas there was an increase of 24.2% ( P = .133) in the physician’s probability for gynecologic disease. In patients with a negative ED EVUS, there was a mean increase of

5.35 ( P = .048) in the physician’s probability for surgical disease, whereas there was a decrease of 18.6% ( P b .001) in the probability for gynecologic disease and 2.5% ( P = .58) for medical disease (Fig. 1).

Emergency department EVUS reduced physician proba- bility of gynecologic disease in 24 (60%) patients, increased it in 10 patients (25%), and produced no change in 6 patients (15%). There was a mean reduction of 25% ( P b .001) in weighted differential diagnoses with regard to gynecologic disease after ED EVUS.

In regard to further management, ED EVUS changed consultation and management minimally (Table 3). Con- trasted CT imaging of the abdomen and pelvis increased

Table 3 Change in consultation and management after ED EVUS

Emergency physician management plan No. of patients (%)

with change post-ED EVUS

Surgical consult -2 -5.0%

Gynecologic consult -1 -2.5%

Medical consult +1 +2.5%

Ultrasound from gynecology service -2 -5.0%

Ultrasound from radiology service +2 +5.0%

CT scan abdomen/pelvis noncontrast -5 -12.5%

CT scan abdomen/pelvis contrast +5 +12.5%

Immediate laparotomy/laparoscopy 0 0

after ED EVUS and was eventually used in 75% of enrolled patients. These CT scans showed the following findings not found on ED EVUS: 4 cases of appendicitis, 2 cases of enlarged mesenteric lymph nodes, and 1 case of inguinal deep vein thrombosis.

Eventual final discharge diagnoses for these 40 patients included appendicitis (4), gynecologic disease (14) (pelvic inflammatory disease [1], ovarian cysts [9], endometrioma [1], uterine fibroids [3]), 4 patients with medical disease (pyelonephritis [1], gastroenteritis [2], inguinal deep vein thrombosis [1]), and 18 patients with nonspecific abdom- inal pain.

Discussion

The evaluation and management of the nonpregnant female patient with RLQ pain encompass consideration of numerous diagnoses, consultants, laboratory tests, imaging tests, observation, surgery, and possibly procedures. As we described before, we believed we could change a compo- nent of this typical course by implementing pelvic ultra- sound as a step after physical examination and before further management. Of note, we designed this study in an era when consultation was more frequent and when imaging, particularly CT scanning, was used less in our ED. Our experience suggested that the use of EVUS, as opposed to transabdominal scanning, would be the pelvic sonographic modality of choice due to its exquisite anatomical definition of uterine and adnexal pathology. Endovaginal pelvic ultrasound usually complements the pelvic examination and is conveniently performed rapidly

after the pelvic examination at the bedside [4].

The predominance of normal pelvic ultrasounds was not surprising in this young cohort, most of whom had no history of gynecologic disease. Positive ED EVUS did account for the discovery of ovarian cysts and fibroids, which made up a third of discharge diagnoses. Although the groups differed in all uterine dimensions, and greatly in ovarian size, only the differences in ovarian size were statistically significant.

As expected with the anatomical detail shown by ED EVUS, there were large differences in physician probability of gynecologic disease. The directions of changes were expected; positive ED EVUS findings diminished surgical and medical probabilities and naturally increased gyneco- logic probabilities. Interestingly, ED EVUS did have an effect on the treating physician’s probability of all queriED diseases such as surgical and medical beyond gynecologic etiology. This finding was unexpected, as our ultrasound was confined to the pelvic anatomy and we did not use other more specific sonographic examinations such as renal or venous ultra- sound. The implications of ED EVUS on the physician’s probability of disease may extend beyond the pelvic rim.

During this era, CT of the abdomen and pelvis became more popular and routine, almost replacing both gynecologic and Surgical consultation [5,6]. Our study showed that most

physicians wanted further imaging evaluation of the bowel, specifically the appendix, and turned to the readily available option of CT. It became clear that although many considered the ED EVUS useful, it could not eliminate the use of CT to identify appendicitis. It remains to be determined whether the approach with initial use of EVUS with optional CT scanning would be a superior strategy to just CT scanning, especially with regard to cost, time, and ionizing nature of CT.

In the future, a study of the combination of pelvic ultrasound, laboratory tests, and observation vs CT scanning may be a viable comparison [7]. As more of the population is exposed to ionizing radiation from ubiquitous CT scanning, questions will arise as to the need for a CT on every patient with RLQ pain [6,8,9]. Pelvic ultrasound in different stratifications of pretest probability of disease in patients with significant pelvic or abdominal abnormalities may also be reasonable avenues for further research.

Limitations

This was an observational, convenience sample of enrolled patients with limited inclusion criteria who had sonography done by a select group of trained physicians and physician assistants. Use by personnel with other levels of training may have influenced the study in both positive and negative ways. The study inclusion criteria were meant to look at patients with high suspicion of appendicitis, and pelvic ultrasound may be useful in a more generalized cohort of patients with generalized lower quadrant pain. The changes in decision making may have been influenced by other factors including education level, departmental crowd- ing, availability of other resources, and patient willingness to pursue other testing. This study was limited to findings with gray-scale endovaginal pelvic ultrasound; other sonographic techniques such as Doppler ultrasound, appendiceal com- pression ultrasound, renal ultrasound, or venous ultrasound may also influence decision making in this type of patient. In this limited cohort, ED EVUS did not identify Tubo-ovarian abscesses, Pelvic abscesses, or a dilated appendix typical of sonographic appendicitis. Finally, the lack of randomization and of significant numbers of patients precludes any definite recommendation of the use of pelvic ultrasound in this manner.

Conclusion

Emergency department EVUS changes physician diag- nostic decision making in nonpregnant women with undifferentiated RLQ pain.

References

1. Lambert MJ, Villa M. Gynecologic ultrasound in emergency medicine. In: Rosen CL, Wolfe RE, editors. Ultrasound in emergency medicine. Philadelphia7 Saunders; 2004. p. 683 - 96.
2. Tayal VS, Cohen H, Norton HJ. Outcome of patients with an indeterminate emergency department first-trimester pelvic ultrasound to rule out Ectopic pregnancy. Acad Emerg Med 2004;11:912 - 7.
3. Haider Z, Condous G, Ahmed S, Kirk E, Bourne T. Transvaginal sonographic diagnosis of appendicitis in acute pelvic pain. J Ultrasound Med 2006;25:1243 - 4.
4. Bulas DI, Ahlstrom PA, Sivit CJ, Blask RN, O’Donnell RM. Pelvic inflammatory disease in the adolescent: comparison of transabdominal and transvaginal sonographic evaluation. Radiology 1992;183:435 - 9.
5. Regimbeau JM, Panis Y, Le P, Soyer P, Kardache M, Rymer R, et al. Should routine Spiral computed tomography be performed in adult patients with acute abdominal pain of the right lower quadrant? Int J Colorectal Dis 2003;18:126 - 30.
6. Perez J, Barone JE, Wilbanks TO, Jorgensson D, Corvo PR. Liberal use of Computed tomography scanning does not improve diagnostic accuracy in appendicitis. Am J Surg 2003;185:194 - 7.
7. Roberts M, Valley V, Bradley K, Summers R. Combined use of C-reactive protein and endovaginal ultrasonography in the emergency department diagnosis of complicated pelvic inflammatory disease. Ann Emerg Med 2002;40:S75.
8. Torbati SS, Guss DA. Impact of helical computed tomography on the outcomes of emergency department patients with suspected appendi- citis. Acad Emerg Med 2003;10:823 - 9.
9. Vadeboncoeur TF, Heister RR, Behling CA, Guss DA. Impact of helical computed tomography on the rate of negative appendicitis. Am J Emerg Med 2006;24:43 - 7.