a b s t r a c t

Background: For patients with early intrauterine pregnancy , the sonographic signs of the gestation may be below the resolution of transabdominal ultrasound (TAU); however, it may be identified by transvaginal ultra- sound (TVU). We sought to determine how often TVU performed in the emergency department (ED) reveals a viable IUP after a nondiagnostic ED TAU and the impact of ED TVU on patient length of stay (LOS).

Methods: This was a retrospective cohort study of women presenting to our ED with complications of early preg- nancy from January 1, 2007 to February 28, 2009 in a single urban adult ED. Abstractors recorded clinical and im- aging data in a database. Patient imaging modality and results were recorded and compared with respect to ultrasound (US) findings and LOS. Results: Of 2429 subjects identified, 795 required TVU as part of their care. Emergency department TVU was per- formed in 528 patients, and 267 went to radiology (RAD). Emergency department TVU identified a viable IUP in 261 patients (49.6%). Patients having initial ED US had shorter LOS than patients with initial RAD US (median 4.0 vs 6.0 hours; P b .001). Emergency department LOS was shorter for women who had ED TVU performed com- pared with those sent for RAD TVU regardless of the findings of the US (median 4.9 vs 6.7 hours; P b .001). There was no increased LOS for patients who needed further RAD US after an indeterminate ED TVU (7.0 vs 7.1 hours; P = .43). There was no difference in LOS for those who had a viable IUP confirmed on ED TAU vs ED (median 3.1 vs 3.2 hours, respectively; P b .32).

Conclusion: When an ED TVU was performed, a viable IUP was detected 49.6% of the time. Emergency department LOS was significantly shorter for women who received ED TVU after indeterminate ED TAU compared with those sent to RAD for TVU, with more marked Time Savings among those with live IUP diagnosed on ED TVU. For pa- tients who do not receive a definitive diagnosis of IUP on ED TVU, this approach does not result in increased LOS.

(C) 2015

Introduction

Complications of first trimester pregnancy are a common emergency department (ED) problem. Miscarriage occurs in 20% to 25% and Ectopic pregnancy in 1.5% to 2% of pregnancies [1,2]. Studies have shown that ED physicians can accurately and rapidly use bedside ultrasound (US) to detect an intrauterine pregnancy (IUP) [3-5]. In addition, studies have shown that pelvic US performed by emergency physicians de- creases ED length of stay [6-8]. To our knowledge, no study has

? Disclosures: None of the authors listed on this study has outside funding or support of any kind for the work or any conflicts of interest to disclose.

?? Presentation of data: Society for Academic Emergency Medicine (SAEM) oral presen-

tation in Arizona, 2011.

* Corresponding author.

E-mail addresses: [email protected], [email protected] (N.L. Panebianco).

determined the added utility of ED Transvaginal ultrasound (TVU) after an indeterminate ED transabdominal ultrasound (TAU). In this study, we sought to determine how often a viable IUP is diagnosed on ED TVU when the ED TAU is nondiagnostic. In addition, we examined how ED TAU/TVU affected patient LOS.

Transabdominal pelvic US is relatively easily learned in conjunction with other sonographic applications using a general purpose abdominal transducer. Transvaginal ultrasonography is performed using an endocavitary transducer that provides higher resolution images due to its proximity to the target organs. This allows for detection of an IUP ear- lier in gestation as well as more detailed examination of the adnexa. However, TVU involves the cost of a dedicated transducer that is used for relatively few other applications and that requires special disinfec- tion and storage. Emergency department TVU also requires additional training, time to perform, and the presence of a chaperone. In addition to these impediments, ED physicians and their patients may be

http://dx.doi.org/10.1016/j.ajem.2015.02.023

0735-6757/(C) 2015

concerned about the possible need for a repeated endocavitary examination by radiologists after an indeterminate ED TVU examination. For this reason, some EDs use a strategy in which all patients with an indeterminate ED TAU are referred to radiology (RAD) for the TVU examination. However, this strategy may result in longer ED LOS for those patients who could be discharged home immediately after a diagnostic TVU performed in ED.

Because of these relative resource burdens and potential benefits, it would be helpful to determine how frequently TVU is used and how often it is diagnostic among typical ED patients. If it is found that ED TVU yields a diagnosis of viable IUP in a substantial proportion of pa- tients whose ED TAU is indeterminate, and the time savings in each case is large; it may justify the additional costs, training, physician time, and potential discomfort to the patient.

Methods

Study design and setting

This was a retrospective cohort study of data from January 2007 to February 2009 in a single urban adult ED with approximately 65000 an- nual visits in a tertiary care university medical center. The study ED has a 4-year residency program and a US fellowship. Our ED residents re- ceive 4 weeks of dedicated US training, and all meet the minimum edu- cational standards as outlined in the American College of Emergency Physicians (ACEP) Emergency US Guidelines for ED physicians practi- cing emergency US [9]. Resident-performed US examinations are reviewed by an attending physician at the time it is performed or re- ceive radiologic imaging to confirm resident findings. Attending physi- cians are credentialed to perform US by the Credentialing Committee at the University of Pennsylvania. At the time of this study, our ED had 2 US fellowship-trained attending physicians, 2 US fellows, and 75% of the faculty were credentialed in ED TVU. All ED-performed USs are reviewed by the Division of Emergency US on a weekly basis for quality and accuracy. This study was approved by the Institutional Review Board at The University of Pennsylvania.

Selection of patients

Female patients presenting to the ED between the ages of 15 and 55 years with complaints of either abdominal pain, pelvic pain, or Vaginal bleeding receive a urine pregnancy test in triage or as soon as possible thereafter. Patients in this group whose pregnancy test is positive are identified in the triage note of our electronic medical record (EMR) as having first trimester pain/bleeding. The cohort for this study was iden- tified by retrospectively querying the EMR for these patients.

Description of clinical practice and patient flow

In our ED, the ultimate goal of our evaluation is to place patients in one of the following Diagnostic categories by the time of disposition: definite viable IUP, spontaneous abortion, intrauterine fetal demise (IUFD), ectopic pregnancy, molar pregnancy, or indeterminate. The al- gorithm for management of patients with complications of first trimes- ter pregnancy was developed conjointly by the Departments of Emergency Medicine and Obstetrics and Gynecology. If a patient does not have a “definite viable IUP” on RAD US, the gynecology service is consulted, while the patient is in the ED. Patients who are discharged with a final “indeterminate” diagnosis have close outpatient follow-up with repeat serial quantitative ?-human chorionic gonadotropin (?- hCG) and sonography as indicated, until the patient is identified as hav- ing one of the other definitive diagnoses listed above. For most patients, the first step in the ED evaluation is Emergency bedside ultrasound (EMBU) performed by the clinical team caring for the patient. The first US examination is ED TAU. A minority of patients do not get EMBU but go directly to RAD US, largely related to attending comfort with staffing the examination. After ED TAU, if there is no evidence of definite

viable IUP (defined at our institution as presence of a yolk sac, fetal heart motion, or fetal pole), the patient receives a TVU. Based on attending physician preference, this is performed either by the clinical team caring for the patient (ED TVU) or in the RAD department (RAD US). If the ED TVU fails to reveal a definite IUP, patients go to RAD US for a full exami- nation. The exception to this is in cases, where the patient is unstable. In these cases, a gynecology consultation and definitive care proceeds based on the ED US. At our institution during the study period, RAD personnel (US technicians and/or radiologists) were available to perform RAD US around the clock, however, at night had to be called in from home. For the most part, the preference of ED attending physi- cians to perform ED TVU is based on their training and experience with pelvic ultrasonography.

Data abstraction

Before data abstraction, 1 research associate and 2 US fellows were provided 1-hour training by the study principal investigator to define study variables. Variables abstracted from the EMR included in-room time; time of disposition (discharge or order to admit); age; race; vital signs; gravity; parity, if the patient was bleeding; days of bleeding; site of US examination (in ED, in RAD suite, or both); US modality; result(s) of the US(s); and final ED diagnosis. The abstractors were instructed to document the EMBU/RAD result as reported in the record. Any questionable EMBU/RAD reports were arbitrated by the study prin- cipal investigator. In our EMR, the chart is automatically time stamped when the patient is placed in a room and at the time of disposition. For the purposes of this study, a RAD report that described the presence of a yolk sac, fetal pole, or fetal heart motion in the uterus was consi- dered a “definite IUP.” Reports of definite IUFD, molar pregnancy, or retained products were recorded as such. All other reports (eg, “possible ectopic,” “cannot rule out ectopic,” and “nonspecific endometrial sac”) were recorded as “indeterminate.” Length of stay was defined as the period between the in-room time to the time of disposition.

Statistical analysis

Summary statistics such as frequencies and percentages for categori- cal data and means with SDs or medians with interquartile range as ap- propriate for continuous data was used to describe the population with regard to demographics and symptoms. To determine if there were baseline differences in presenting complaint between the groups that did and did not receive ED TVU, ?² test was used. To assess differences in LOS, Student t test was used to compare (1) patients going directly to RAD (no EMBU) vs those receiving EMBU (TAU and/or TVU) and (2) pa- tients with no IUP on TAU receiving ED TVU vs those receiving RAD TVU. Additional analyses of LOS comparing subgroups of patients with no IUP on TAU were performed using analysis of variance (ANOVA). To adjust for multiple post hoc pairwise comparisons within the ANOVA, a P value b .01 was used. Because LOS was not normally distributed, a log transformation was performed before either the t test or the ANOVA. In addition, in the LOS analyses, patients were excluded if they left with- out treatment complete (LWTC), records were incomplete as to the type of ED US they received, or if they went to RAD for an alternate diagnosis (eg, appendicitis). All analyses were performed using SAS statistical software (version 9.3; SAS Institute, Cary, NC). A P value b .05 was con- sidered statistically significant unless otherwise noted.

Results

Between January 2007 and February 2009, 2429 unique records were obtained of patients presenting with first trimester pregnant/ bleeding (Fig. 1). Of these, 72 women went directly to RAD with no ED US. One hundred fourteen patients were excluded because the med- ical record does not clearly identify whether they had ED TAU only or ED TAU plus ED TVU. Emergency department TAU identified a viable IUP in 1422 patients. Of these, 1352 were medically managed, and 70 had a RAD US for other reasons. From qualitative data review, other reasons

Table 1

Total Patients

N = 2429

Went directly to radiology with no EMBU

N = 72 LOS: 6.0h

N = 2243

IUP on EMBU TAU

N = 1422 LOS: 3.2h

RAD US for other reasons

N = 70

IUP discharge

N = 1352

No IUP on EMBU TAU N = 821

No TVU performed

N = 26

18 LWTC/AMA

8 Clinically Managed

(7 IUFD, 1 Ectopic)

TVU performed

N = 795

Demographics data for those who had an ED TVU after nondiagnostic ED TAU vs those who went to RAD after a nondiagnostic ED TAU^a

ED TVU (n = 528)

TAU to RAD P

(n = 267)

n % n %

	N = 2357 Race					.89
	Black/African American	416	88%	192	88%
	White	27	6%	11	5%
Unspecified EMBU	Other	31	6%	16	6%
TAU vs TVU N = 114 (excluded)	Age (mean +- SD) Vital signs (mean +- SD)	24.4 +- 6.0		25.2 +- 6.8		.15

Heart rate	86.5 +- 13.6		87.1 +- 16.0		.63
Systolic BP	121.6 +- 18.0		119.5 +- 19.5		.13
Diastolic BP	72.2 +- 12.1		71.2 +- 12.1		.29
Respiratory rate	16.8 +- 3.9		17.0 +- 6.8		.59
Temperature History	98.4 +- 0.6		98.4 +- 0.5		.93
Gravida 0-2	279	53%	136	51%	.78
3-7	228	43%	118	44%
N 7 Parity 0-1	21 384	4% 73%	13 192	5% 72%	.93
2-6	141	27%	73	27%
N 6	3	b1%	2	b 1%
Pain	432	82%	223	83%	.55
Bleeding	314	59%	195	73%	.0002
Bleeding N 2 d	98	32%	68	35%	.50

^a Numbers may not add up to totals due to missing values.

Fig. 1. Flow diagram of how patients were selected for analysis.

included rule out torsion, rule out appendicitis, rule out cholecystitis, and rule out ruptured ovarian cyst. Eight hundred twenty one patients did not have a definite IUP identified by TAU and were therefore eligible for ED TVU. Of these, 18 LWTC or against medical advice before further imaging, and 8 were managed without further imaging leaving 795 for analysis.

General demographics

Among the 795 patients, the median age was 23 years, and most pa- tients were African American (88%). Fifty-two percent of patients had a gravity of 1 to 2, and 72% had parity 0 to 1. Eighty-two percent of pa- tients presented with pain and 64% with vaginal bleeding (Table 1).

Results of ED TVU

Of the 528 women who had an ED TVU, 261 (49.4%) had a viable IUP diagnosed. Of the 267 with no viable IUP identified on ED TVU, the ED physician’s impression based on the US was indeterminate in 209 (39.4%), IUFD in 35 (6.6%), ectopic in 20 (3.8%), molar in 1 (0.2%), and spontaneous abortion in 2 (0.4%) (Fig. 2). Eighteen patients did not go on to RAD US because they LWTC or received definitive without further diag- nostic imaging, leaving 249 patients who went to RAD US after ED TVU.

Of the 249 patients, RAD US identified 33 viable IUP (13.3%). In the non-IUP group, there were 157 indeterminate (63.1%), 36 IUFD (14.5%), 20 ectopic (8.0%), 1 molar (0.4%), and 2 spontaneous abortions (0.8%).

Results of RAD US

After ED TAU revealed no IUP, 267 patients were sent directly to RAD US without ED TVU. Of these 267 patients, 56 had a viable IUP (21.0%). Of the 211 with no viable IUP, 125 had an indeterminate US (46.8%), 58 IUFD (21.7%), 22 ectopic (8.2%), 2 molar (0.8%), and 4 spontaneous

abortion (1.5%) (Fig. 2).

Comparison of groups

The women in the ED TVU vs RAD US groups experienced abdominal pain at similar rates (ED TVU, 82% vs RAD US, 83%; P = .55) (Table 1). Patients who went directly to RAD after ED TAU were more likely to be bleeding than patients who received ED TVU (ED TVU, 59% vs RAD US, 73%; P = .0002). Although women in the RAD US group were more likely to have bleeding, there was no difference in extended bleed- ing (>=3 days) between the 2 groups (bleeding, >=3 days: ED TVU, 32% vs RAD US, 35%; P = .50). Patients were significantly more likely to have a live IUP diagnosed on their ED TVU than those who went to RAD US after an initial indeterminate ED TAU (49.6% vs 21.0%; P b .0001). Conversely, the rate of final diagnosis of “indeterminate,” “IUFD,” and “ectopic” was higher in the patients who went to RAD US after ED TAU vs those who re- ceived an ED TVU (indeterminate, 46.8% vs 39.4%; IUFD, 21.7% vs 6.6%; and ectopic, 8.2% vs 3.8%) (Table 2). With regard to the patients who had no viable IUP on ED TVU, the ED TVU diagnosis had excellent crude agreement on all diagnoses except in the “indeterminate” group, which makes sense in the context of our practice environment (Table 3.). We coach physicians to be conservative when making a diagnosis other than definitive IUP and defer to RAD when unsure. In the most dangerous category of ectopic pregnancy, the ED TVU and RAD had 100% agreement.

Time comparisons

Patients presenting with complications of first trimester pregnancy who went directly to RAD without any bedside US had significantly lon- ger LOS compared with patients receiving any EMBU (6.0 vs 4.0 hours; P b .0001) (Fig. 3, Comparison A). Length of stay of patients who had a de- finite IUP identified on TAU (n = 1352) had a LOS of only 3.2 hours.

After indeterminate TAU, LOS for patients sent directly to RAD US without receiving ED TVU was significantly longer compared with those who received ED TVU (6.7 vs 4.9 hours, respectively; P b .0001) (Fig. 3, Comparison B). Patients diagnosed with a viable IUP on ED TVU had a significantly shorter LOS than those with the same result di- agnosed by RAD (3.6 vs 5.5 hours, respectively; P = .001) (Fig. 3, Com- parison C). After indeterminate TAU, an ED TVU showing a definite IUP was associated with only marginal increased LOS compared with those

ED TVU performed?

” Yes” N = 528

“No” N = 267

Radiology Ultrasound

Definite IUP N = 261 LOS: 3.6h

No IUP N = 267

Indeterminate=209 IUFD =35

Ectopic=20 Mole=1 SAB=2

Definite IUP

N = 56

LOS: 5.5

No IUP N = 211

Indeterminate:125 (LOS:7.1)

IUFD=58

Ectopic =22

Molar pregnancy=2 SAB=4

Symptomatic care and discharged N = 211

Symptomatic care and discharged

Radiology Other reasons N = 50

Excluded N = 18 9 LWTC

Definitive care N = 9 IUFD=3

Ectopic=4 SAB=2

Radiology Ultrasound N = 249

Definite IUP

N = 33

No IUP=216

Indeterminate=157 (LOS:7.0) IUFD = 36

Ectopic =20

Mole =1

SAB = 2

Symptomatic care and discharged

Fig. 2. Patient flow, diagnostic outcome, and length of stay for various diagnostic imaging strategies. Abbreviation: SAB, spontaneous abortion.

patients whose IUP had been diagnosed on ED TAU alone (3.6 vs 3.2 hours). The greatest LOS was for patients with a final “indeterminate” diagnosis regardless whether they had an ED TVU performed or went directly to RAD US after an indeterminate ED TAU (7.0 vs 7.1 hours, re- spectively; P = .43).

Discussion

The principal findings of this study are that when an ED physician chooses to perform an ED TVU after an indeterminate ED TAU, an IUP is diagnosed almost half the time, and performing the ED TVU is associated

Table 2

Ultrasound results by location

ED TVU (n = 528)

TAU to RAD (n = 267)

Difference (95% CI)

Table 3

Crude agreement between RAD and ED TVU. This assessed often the ED diagnosis agreed with the RAD diagnosis

RAD result ED TVU

	n	%		n	%				IUFD	Indeterminate	Ectopic	Molar
IUP	262	49.6%		56	21.0%	28.6% (21.9% to 34.9%)		IUP	0	33	0	0
Indeterminate	208	39.4%		125	46.8%	-7.4% (-14.7% to -0.2%)		IUFD	27	9	0	0
IUFD	35	6.6%		58	21.7%	-15.1% (-20.8% to -10.0%)		Indeterminate	1	156	0	0
Ectopic	20	3.8%		22	8.2%	-4.4% (-8.6% to -1.1%)		Ectopic	0	4	16	0
Molar	1	0.2%		2	0.8%	-0.6% (-2.5% to .4%)		Molar	0	0	0	1
Spontaneous AB	2	0.4%		4	1.5%	-1.1% (-3.4% to 0.2%)		Spontaneous AB	0	2	0	0

?² = 84.3, P b

				Total	28	204	16	1
.0001. AB, abortion.				Crude agreement	96%	76%	100%	100%

Abbreviation:

ED TAU: live IUP (n = 1,352)

3.2

4.0

A

6.0

4.9

B

6.7

3.6

C

5.5

5.8

D

7.0

7.1

EMBU (n = 2,217)

RAD US direct (n = 72)

ED TAU: no definite IUP then ED TVU (n = 528)

ED TAU: no definite IUP then RAD US (n = 267)

ED TAU: no definite IUP then ED TVU: IUP (n = 261)

ED TAU: no definite IUP then RAD US: IUP (n = 56)

ED TAU/TVU: no definite IUP then RAD US: IUP (n = 33)

ED TAU/TVU: no definite e IUP then RAD US: no definite IUP

(n = 216)

ED TAU: no definite IUP then RAD US: no definite IUP

(n = 211)

2 3 4 5 6 7 8

Room to ED Discharge (Hours)

Comparison	P value
A	<.0001
B	<.0001
C	<.0001
D*	<.01

*2 comparisons of ED TAU/TVU no IUP to other 2 groups

Fig. 3. Length of stay comparisons for various diagnostic categories.

with significantly shorter LOS for almost all diagnoses. Performing an ED TVU does not significantly add to ED LOS for any diagnosis.

Although it is well established that TVU provides greater imaging resolution of the gravid uterus, in a survey of 471 ED residents, TVU was 1 of the top 5 US examinations they were least comfortable performing [10]. The burden of purchasing an endocavitary transducer, educating individuals in its use, onerous regulations in sanitizing the probe, and taboo regarding subjecting women to the potential of multi- ple invasive examinations with limited data on the added utility of ED- physician performed TVU compared with TAU may be driving this response. Further study is needed to elicit the barriers to training and performing ED TVU.

Regarding the perception that patients find TVU uncomfortable or embarrassing, a study by Braithwaite and Economides [11] found that women view Transvaginal sonography as acceptable (98.1%), not embarrassing (88.9%), not painful (98.6%), and not stressful (90.6%). Barnhart et al [12] reported in a small study that 88.2% of women find TVU acceptable. With respect to managing expectations, they found that 54.6% of women found the vaginal US to be as comfortable as expected, 44.7% found it more comfortable than expected, and only 0.7% of women found it to be worse than expected.

Several studies have shown that ED physicians can accurately diagnose an IUP and that ED physician-performed US improves ED LOS [6-8]. The current study confirms previous studies on this topic by demonstrating a marked decrease in LOS for patients with an identifiable IUP on ED US. It adds to previous literature by analyzing the impact on LOS of ED US in a

clinical pathway that involves both RAD and ED ultrasonography [6-8]. Concerns might be raised regarding the generalizability of the current study. The time savings noted in this study may not apply to all practice set- tings and must be considered relative to one’s own practice environment. The current study was conducted at an institution with 24-hour US techni- cian and RAD coverage. In locations without 24-hour RAD coverage, the difference in LOS between the 2 groups might be even more pronounced. A further concern might be that patients who went to RAD US after an indeterminate TAU in the ED might have been in some way a sicker or more complicated group than those who had an ED TVU, or that they might have had higher rates of gynecologic consultation, thereby leading to longer LOS. It is true that, in the present study, the RAD US group had a higher proportion of patients with vaginal bleeding and a lower rate of “definite viable IUP” diagnosis. A previous study has shown that the pre- sence of vaginal bleeding in early pregnancy is associated with decreased diagnostic accuracy of US, and this in part may explain the high preva- lence of vaginal bleeding in the group sent directly to RAD US after an in- determinate ED TAU [13]. It is possible that the combination of clinical and ED TAU findings gives physicians clues regarding the likelihood that the TVU will reveal the findings to make the diagnosis of “definite IUP.” However, the statistically significant difference in bleeding between the groups that received ED TVU after indeterminate ED TAU and those that were sent for RAD US might actually not represent a “bias” so much as as- tute clinical practice in that it reflects the actual practice of experienced clinicians. Future study to determine why an ED TVU is performed in

some patients and not others is warranted.

Quantitative ?-hCG levels were not collected for analysis. In our insti- tution, a ?-hCG is often not drawn, unless the ED TAU (or ED TVU if per- formed) reveals no IUP, as the value does not change our protocol of whether to perform an US and is only useful if there is no definitive IUP. At our institution, women without a definitive IUP on ED US receive a RAD US regardless of the ?-hCG; however, a recent study by Kus and Juliano [13] suggests that, in low-risk patients, women with a ?-hCG less than 3000 mlU/mL with an indeterminate ED US may be safely sent home with close gynecologic follow-up.

Overall, the emergency physician’s diagnosis was very similar to those of radiologists. It is to be anticipated that Emergency sonologists will make an indeterminate diagnosis more frequently than radiolo- gists. Regarding the most concerning form of error, this study cannot ab- solutely exclude the possibility that some cases of ectopic pregnancy were overlooked on EMBU and discharged from the ED. However, our hospital is the major provider of obstetrical services in the area, and most of our patients return to this institution any time they have a prob- lem. Furthermore, our quality assurance process did not identify any images that appeared to demonstrate an ectopic when the ED reading was of an IUP. In this respect, our quality assurance Review process is no different from that of a department of RAD.

Neither operator experience, nor personal attending emergency physi- cian practice preferences were controlled in the study, which was conduct- ed in an ED with an active US program for both residents and fellows. These factors may make it difficult to generalize the study findings, al- though they may also reflect the reality of ultrasonography in our specialty in which there is a wide range of approaches and skill sets that is consid- ered acceptable. Along these lines, it is likely that different practitioners would have rendered different US results based on their proficiency, with less experienced sonologists more likely to call scans “indeterminate” and obtain a RAD scan. This being said, there are numerous studies demon- strating ED sonologists’ ability to perform this task accurately [2-4].

To our knowledge, no previous study has sought to determine the added utility of ED TVU after a nondiagnostic ED TAU. It is likely that in- dividual emergency physicians will determine on a case-by-case basis whether the time savings demonstrated in this study are likely to be re- alized in their practice setting. However, this study provides some evi- dence that there is justification for the added cost, training, and probe maintenance associated with providing an ED TVU examination.

Conclusion

In summary, when an ED physician chooses to perform an ED TVU after no IUP is identified on ED TAU, an IUP is detected 49.4% of the time. Emergency department LOS was significantly shorter for women who received ED TVU after indeterminate ED TAU compared with those sent to RAD for TVU, with more marked time savings among those with live IUP diagnosed on ED TVU. Performing an ED TVU is not associated with an increased LOS.

References

1. Barnhart KT. Ectopic pregnancy. N Engl J Med 2009;4:379-87.
2. Clinical policy. Critical issues in the initial evaluation and management of patients presenting to the emergency department in early pregnancy. Ann Emerg Med 2003;41:123-33.
3. Durham B, Lane B, Burbridge L, Balasubramaniam S. Pelvic ultrasound performed by emergency physicians for the detection of ectopic pregnancy in complicated first- trimester pregnancies. Ann Emerg Med 1997;29(3):338-47.
4. Blivas M, Sierzenski P, Plecque D, Lambert M. Do emergency physicians safe time when locating a live intrauterine pregnancy with bedside ultrasound? Acad Emerg Med 2000;7(9):988-93.
5. Wattimena J, Pelosi M, Cheney K, Green T, Black KI. Factors affecting length of stay for women presenting with early pregnancy complications to a public hospital emergency department. Emerg Med Australas 2013;25(1):22-7.
6. Shih CH. Effect of emergency physician-performed pelvic sonography on length of stay in the emergency department. Ann Emerg Med 1997;29(3):348-51 [discussion 352].
7. Blaivas M, Sierzenski P, Plecque D, Lambert M. Do emergency physicians safe time when locating a live intrauterine pregnancy with bedside ultrasonography? Acad Emerg Med 2000;7(9):988-93.
8. Burgherm S, Tandy T, Dawdy MR. Transvaginal ultrasonography by emergency phy- sicians decreases patient time in the emergency department. Acad Emerg Med 1998;5(8):802-7.
9. Kappor M, Greenstein J, Davis M, Das D, Panebianco N. Emergency medicine resi- dents perspective on ultrasound training. Ann Emerg Med 2014;64(4):S24-5.
10. Mohammed F, Basama S, Crosfill F, Price A. Women’s perception of transvaginal sonography in the first trimester; in an early pregnancy assessment unit. Arch Gynecol Obstet 2004;269:117-20.
11. Braithwaite JM, Economides DL. Acceptability by patients of transvaginal sonogra- phy in the elective assessment of the first-trimester fetus. Ultrasound Obstet Gynecol 1997;9:91-3.
12. Barnhart KT, Fay CA, Suescum M, Sammel MD, Appleby D, Shaunik A, et al. Clinical factors affecting the accuracy of ultrasonography in symptomatic first-trimester pregnancy. Obstet Gynecol 2011;117(2, Part 1):299-306.
13. Kus M, Juliano M. Do all women with indeterminate pregnancies need a formal ultrasound before discharge from the emergency department? Mil Med 2014; 179(11):1263-5.