Article, Urology

The impact of goal-directed transvaginal ultrasonography on clinical decision-making for emergency physicians

a b s t r a c t

The aim of study was to determine the impact of “goal-directed transvaginal ultrasonography” (TVUSG) on real- time clinical decision making of Attending emergency physicians evaluating their level of certainty for prelimi- nary diagnosis, admission, surgery, treatment, additional laboratory, and discharge in patients presenting with acute pelvic pain to the emergency department (ED).

This prospective cross-sectional clinical study was conducted on sexually active female patients older than 18 years who presented with acute pelvic pain in the ED. The level of certainty of clinical decision making as mentioned above was measured by a Visual analogue scale from 0 to 100 mm with 100 mm being most certain before and after TVUSG. Statistical analysis was performed on 88 patients. The mean age was 31.7 +-8.3 years with a median of 30 years. Among Clinical decisions, there was a significant difference between pre-TVUSG and post-TVUSG certainty of the decision to perform preliminary diagnoses derived from patient’s history and physical examination but not in the other outcomes (treatment, admission, surgery, and discharge). (P = .05). Of the patients included in the study, 11 (12.5%) were admitted to hospital, and 2 (2.3%) of them were operated on. The remaining 75 (85.2%) patients were discharged from the ED; of the patients that had been discharged, 18 (20.5%) patients later consulted another physician, and no further pathology could be discovered.

In conclusion, US performed by attending emergency physicians may affect the certainty of their decisions in patients presenting with acute pelvic pain. This effect statistically significantly on the decision to determine preliminary diagnosis.

(C) 2015

Introduction

The underlying reason for the success of bedside ultrasonography (USG), which can be performed as a part of physical examination of pa- tients admitted at the emergency department (ED) with abdominal pain, is that the procedure is a practical, noninvasive, inexpensive, and repeatable diagnosis method. “Goal-directed USG” can be described as an assessment of patient condition that is made using real-time USG im- ages after the emergency medicine specialist has obtained all clinical in- dications of the patient in light of medical history, physical examination, and the resulting prediagnoses. These tools are referred to as “the stethoscope of emergency medicine specialists” chiefly in the United States and recently in list your country.

Ultrasonography is used fairly extensively in emergency medicine. Trauma, ophthalmologic, gastrointestinal, and pelvic emergencies

? This study was supported by Akdeniz University Research Projects Unit.

* Corresponding author at: Department of Emergency Medicine, Akdeniz University Faculty of Medicine, Dumlupinar Bulvari 07059 Antalya, Turkey. Tel.: +90 242 2496178;

fax: +90 242 2274490.

E-mail address: [email protected] (F. Bektas).

constitute only a part of them. To assess the impact of goal-directed USG on clinical decision making of emergency physicians, each applica- tion has to be evaluated separately. Two previously published studies on the subject have shown that the procedure had a positive effect on survival of patients visiting the ED [1,2].

Although bedside “goal-directed transvaginal ultrasonography” (TVUSG) has become a part of gynecological examination of patients admitted to the ED with complaints of acute pelvic pain, its impact on clinical decisions of emergency medicine specialists has not been investigated.

The aim of this study is to assess the effect on real-time clinical decision making of emergency medicine specialists when bedside goal-directed TVUSG is performed on patients visiting the ED with acute pelvic pain.

Methods

This single-center, prospective clinical study was conducted on sexually active female patients older than 18 years that presented with acute pelvic pain to the tertiary care university ED, which is a third-level university hospital that receives approximately 90000 visits to the ED annually, between June 2011 and December 2011. The study

http://dx.doi.org/10.1016/j.ajem.2015.03.063

0735-6757/(C) 2015

896 N. Sayrac et al. / American Journal of Emergency Medicine 33 (2015) 895898

Acute Pelvic Pain (n = 238)

Excluded (n = 149)

Virgin (n = 57). Pregnant (n = 34).

Not informed consent (n = 34).

Previous pelvic or rectal surgery (n = 22). Patients receiving infertility treatment (n = 2).

Included (n = 89)

Not Follow up (n = 1)

Recorded

Figure. Patient flowchart.

Analysis (n = 88)

Analysis

Follow Up

had been given confirmation by the Research Ethics Committee of University beforehand. Prior to the study, academic staff, nurses, and tri- age professionals working at the ED were informed about the workflow. The study was carried out on weekdays, between 8:00 AM and mid- night, while 2 emergency physicians were working at the ED. Inclusion criteria for female patients were defined as being older than 18 years, being sexually active, and having complaints of lower abdominal or pel- vic pain for less than 3 months. Exclusion criteria for the study are given

in the Figure.

medical histories of the patients meeting inclusion criteria were taken by the responsible emergency medicine resident. After performing a full physical examination, the emergency medicine resident planned the necessary examination and treatment appropriate to the prediagnosis in accordance with daily Patient evaluation procedures. Following this assessment, study procedure was explained to the patients meeting inclu- sion criteria; and those of who agreed to give written informed consent were included in the study.

Prior to the onset of the study, emergency medicine residents working at the ED were given a 1-hour theoretical course in acute pelvic pain, diagnosis, pathophysiology, clinical symptoms, imaging (TVUSG), differential diagnosis, and treatment and a 2-hour practical course. Fol- lowing theoretical training, emergency medicine residents involved in the study were subjected to preliminary study covering 10 patients, specifying which particular indications and symptoms to pay attention to during gynecologic examination. This preliminary study also covered application techniques of TVUSG, structures to notice during USG, diagno- ses, and pathologic symptoms. Afterwards, data collection was initiated.

General physical and gynecological examinations were performed on the patients involved, and results were recorded. Indications and patho- logical symptoms checked throughout gynecological examination are as follows: cervical dilatation, pain during cervical motion, cervical discharge (blood/pus/clear), vaginal-cervical erosion/inflammation/tumor, uterine/ ovarian/pelvic mass, and uterine and ovarian tenderness. Clinical deci- sions before application of TVSUG were recorded by nurses present dur- ing the examination. Certainty of the physician in clinical decisions prior to USG was marked on the form using a 100-mm visual analogue scale

(VAS). Clinical decisions of the physician were classified as prediagnosis, evaluation, treatment, hospital admission, emergency surgery, and discharge. Transvaginal USG was performed following gynecological examination in lithotomy position on the examination table with a 5- to 7.5-MHz vaginal probe (Medison Digital Sonoace 5500; Medison America, Inc, Cypress, CA) and scanned for diagnosis and pathologies given in Table 1.

After TVSUG was performed, certainty of the physician in said clini- cal decisions subsequent to USG was marked again on the 100-mm VAS; and the study was concluded. Positions of each mark made by physi- cians were recorded, and these measurements were arranged in 3 groups ranking clinical probability: 0 to 30 mm (low), 31 to 70 mm (moderate), and 71 to 100 mm (high). The effects of performing TVSUG on certainty of a physician are categorized as follows:

Decision making of the physician in clinical decision may change. This statement is regarded as the primary outcome measure of the study.
  • Pelvic pathologies can be ruled out.
  • A different pathology from the physician’s clinical decision may be encountered.
  • Clinical decision making of the physician may be unaltered.
  • After the study was concluded, standard emergency medicine proce- dures were resumed for daily treatment of patients with acute pelvic pain. Radiology consultation was requested for patients in need of con- ventional USG subsequent to TVUSG. If required, emergency consultation

    Table 1

    Findings of TVUSG

    Intrauterine pregnancy Transvaginal probe tenderness Free fluid in Douglas

    Uterine fibroid

    Tuba-ovarian abscess Ovarian solid mass Ovarian cyst

    Ectopic pregnancy

    N. Sayrac et al. / American Journal of Emergency Medicine 33 (2015) 895898 897

    Table 2

    Vital signs of study patients

    Vital sign

    Mean

    Systolic (mm Hg)

    120

    Diastolic (mm Hg)

    71

    Fever (?C)

    36.6

    Pulse (beat/min)

    85

    for patients with obstetrics and gynecology consultation was provided; patients not needing emergency consultation were directed to obstetrics and gynecology polyclinic for controls, and follow-up treatments were provided there. Further tracking data of the patients involved in the study were collected by a faculty member of the Department of Obstetrics and Gynaecology who was associated with the study. Final diagnoses of the patients were made using results of abdominal computed tomogra- phy and conventional USG, operative findings, and clinical follow-up. All the patients were contacted via telephone 1 week later and asked if their pain persisted, if they consulted another physician, and if they had an operation. Resulting data were recorded.

    Data collected in the study were entered into the computer program Statistical Package for the Social Sciences (SPSS, Chicago, IL) 16.0, and statistical analysis was performed. Continuous variables are mean (standard deviation); failing to comply with a normal distribution was expressed as median (interquartile range). Categorical data were expressed as a percentage. Comparison of the dependent 2 groups for frequency data was by McNemar test. P value b .05 was considered statistically significant.

    Results

    Out of 248 patients that visited the ED with acute pelvic pain, 89 pa- tients meeting the criteria were included in the study. One patient was later excluded because of unavailable data. Statistical analysis was per- formed on 88 patients. Patient flowchart used in the study is given in the Figure. The youngest of the patients was 19 years old, whereas the oldest was 57. Mean age was 31.7 +- 8.3 with a median of 30. Mean height and weight of patients were 162.5 cm and 62.9 kg, respectively.

    Vital signs (Table 2) and results of physical examinations (Table 3) of the patients at the time of arrival at the ED are presented.

    Ovarian cysts in 21 patients and fluid collection in Douglas pouches in 17 patients were discovered by TVUSG (Table 4).

    Eleven (12.5%) patients involved in the study were admitted to the hospital, whereas 75 (85.2%) patients were discharged with suggestions (Table 5). Two (2.3%) of the admitted patients were operated on. Eigh- teen (20.5%) of the discharged patients consulted another physician, but additional pathology was not detected in patients. Final diagnoses of the patients are given in Table 6.

    To investigate the impact of TVUSG on clinical decisions, positions of the marks made by physicians on the VAS were determined using rulers and arranged into subgroups. This categorization showed that TVUSG had an effect on clinical decisions of emergency specialists. This effect

    Table 3

    Physical examination findings of study patients

    Physical examination

    + (%)

    – (%)

    Pelvic tenderness

    88 (100)

    Abdominal defense

    7 (8)

    81 (92)

    Rebound

    11 (12.5)

    77 (87.5)

    Ovarian tenderness

    58 (65.9)

    30 (34.1)

    Uterine tenderness

    34 (38.6)

    54 (61.4)

    Cervical motion tenderness

    53 (60.2)

    35 (39.8)

    Cervical discharge

    56 (63.6)

    32 (36.4)

    Uterine mass

    2 (2.3)

    86 (97.7)

    Cervical erosions

    19 (21.6)

    69 (78.4)

    Cervical dilatations

    3 (3.4)

    85 (96.6)

    Table 4

    Transvaginal USG findings of study patients

    USG findings

    +

    Suspect

    Total

    Intrauterine pregnancy

    87

    1

    88

    Ectopic pregnancy

    1

    83

    4

    88

    Ovarian cyst

    21

    58

    9

    88

    Ovarian solid mass

    1

    81

    6

    88

    Tuba-ovarian abscess

    79

    9

    88

    Uterine fibroid

    3

    75

    10

    88

    Free fluid in Douglas

    17

    62

    9

    88

    Transvaginal probe tenderness

    45

    33

    10

    88

    was especially pronounced for prediagnoses of physicians built on pa- tient history and physical examination (systemic and gynecological). Clinical decisions after the result of this categorization are given in Table 7.

    Limitations

    Include all female patients older than 18 years that visited the ED with complaints of pelvic pain in this study. After patients meeting ex- clusion criteria and those with unavailable data were excluded, 88 pa- tients in total remained available for statistical analysis. We believe that as the number of patients involved increases, the marginal statisti- cal significance of clinical decisions especially regarding prediagnoses made after TVUSG would become stronger. However, it must be noted that the opposite result can also be obtained.

    Another limitation of this study was that the Image quality of the USG device used was on a lower level than devices supporting advanced technology. If the study is replicated using a more advanced device, leading to better imaging of anatomical structures and identification of associated pathologies, we believe that Confidence levels of physicians in clinical decisions would likely improve.

    Discussion

    This prospective clinical study showed that for patients visiting the ED with isolated pelvic pain, performing TVUSG after patient history was taken and physical examination was done had a positive effect only on prediagnosis among clinical decisions of the emergency physi- cians (prediagnosis, evaluation, treatment, hospitalization, emergency surgery, and discharge from the ED). The physician taking patient histo- ry and performing physical examination obtains USG images either supporting or conflicting with the clinical decisions made prior to USG. This is called goal-directed USG and affects survival of the patient at the ED. This study showed that bedside TVUSG performed as a part of physical examination had an effect on surveys of the patients at the ED (P = .05). Transvaginal USG can alter prediagnosis of the patient made by the physician that takes patient history and performs physical examination. However, the same cannot be stated for clinical decisions regarding discharge, admittance, emergency surgery, or treatment at the ED (P N .05). Confirmatory tests help emergency medicine physi- cians confirm or eliminate diagnoses of patients associated with low or high clinical probability. In this study, there were 51 patients in total with prediagnoses ranking low and high clinical probability. Most of these patients had highly clinical probability (n = 40, 45.5%), whereas 11 (12.5%) patients exhibited low clinical probability. Transvaginal USG performed by emergency medicine specialists helped

    Table 5

    Outcomes of study patients

    Outcome n (%)

    Admit to hospital

    11 (12.5)

    Discharged from ED

    77 (85.2)

    Total

    88 (100)

    898 N. Sayrac et al. / American Journal of Emergency Medicine 33 (2015) 895898

    Table 6

    Final diagnosis of patients

    Final diagnosis n (%)

    Pelvic inflammatory disease 30 (34.9)

    Nonspecific pelvic pain 17 (19.3)

    Cystitis 12 (13.5)

    Ovarian cyst rupture 5 (5.6)

    Dysfunctional uterine hemorrhage 5 (5.6)

    Dysmenorrheal syndrome 3 (3.4)

    Vaginitis 3 (3.4)

    Cervicitis 2 (2.2)

    Complicated ovarian cyst 2 (2.2)

    Tuba-ovarian abscess 2 (2.2)

    Herpes genitalis 1 (1.1)

    Abortion 1 (1.1)

    Other 5 (5.5)

    them confirm prediagnoses with high clinical probability; however prediagnoses with low clinical probability were not supported by TVUSG. Out of 11 prediagnoses with low probability, 5 remained low, whereas 3 prediagnoses each were raised to moderate and high proba- bility; additional laboratory and imaging work was required for reaching prediagnoses. Such cases are potentially of great significance because they lead to distinct changes in clinical decisions influenced by TVUSG.

    Forty-two percent (n = 37) of the patients involved in the study re- ceived prediagnoses with moderate clinical probability. Twenty of these prediagnoses were raised to high probability after TVUSG, whereas 8 declined to low. In general, such patients constitute the most difficult group for emergency medicine physicians.

    This study showed that although TVUSG impacted clinical decisions of emergency medicine physicians regarding prediagnoses, its effect on de- cisions regarding admittance, discharge, and emergency surgery was not statistically significant but still clinically of some importance. Patients considered for hospitalization were admitted, and those to be discharged were indeed discharged. Literature on the impact of USG on clinical deci- sion making of physicians at the ED is found to be limited to 2 studies. Bektas et al [1] studied patients calling at the ED with isolated upper right quadrant pain. Performing upper right quadrant USG after taking of patient history and physical examination was shown to positively affect clinical decisions of physicians (P = .01). In contrast, decisions of emer- gency medicine specialists regarding discharge, hospitalization, emergen- cy surgery, and treatment at the ED were unaltered (P N .05). The study

    made by Bassler et al [2] involved all patients visiting the ED with com- plaints of abdominal pain. Because there was no discrimination between quadrants or any specific complaints of pain, the patient sample was not homogenous. Consequently, it is not surprising that USG may reveal un- expected ultrasonographic results unrelated to the patient’s original com- plaint and thus affect decisions of emergency medicine specialists. Regardless of this major limitation, said study concluded that USG per- formed at the ED impacted diagnoses and decisions on discharge and treatment. In addition, confidence in clinical decisions made on the pa- tients involved was registered on a numeric scale from 1 to 10 and even- tually grouped into 3 rankings: low (1-3), moderate (4-7), and high (8- 10) confidence. Instead of the numeric scale used in the said study, a VAS would have produced more explicit results [2].

    When patients visit the ED because of pelvic pain, exhibiting acute abdominal symptoms, conventional USG investigation is technically limited by the presence of bowel gas, insufficiently full bladders, and obesity. These limitations do not apply to TVUSG [3,4]. Moreover, an empty bladder leads to better TVUSG performance. In this study, USG imaging was performed by veteran emergency medicine specialists with at least 10 years of USG experience. Ultrasonography is a personal- ized diagnosis tool. The more experience in application a user has, the more accurate and reliable imaging is. Ultimately, TVUSG is affected less by technical factors related to patients, such as bowel gas, obesity, or abdominal rigidity.

    In conclusion, although goal-directed TVUSG performed at the ED on patients with acute pelvic pain impacted clinical decisions of emergency medicine specialists on prediagnoses, no statistically significant effect was observed on clinical decisions regarding admittance, discharge, or emergency surgery. Transvaginal USG helps emergency medicine physi- cians better examine and diagnose patients visiting the ED with acute pel- vic pain following bimanual examination, and the procedure could be considered as the “eyes of hands” for emergency medicine physicians.

    References

    1. Bektas F, Eken C, Soyuncu S, Kusoglu L, Cete Y. Contribution of goal-directed ultrasonog- raphy to clinical decision-making for emergency physicians. Emerg Med J 2009;26(3): 169-72.
    2. Bassler D, Snoey ER, Kim J. Goal-directed Abdominal ultrasonography: impact on real- time decision making in the emergency department. J Emerg Med 2003;24(4):375-8.
    3. Ma OJ, MD, Mateer JR, MD, RDMS, Blaivas M, MD, RDMS. Gynecologic consepts. In: Emergency ultrasound 2th ed. chapter 14; 2008.
    4. American College of Emergency Physicians: policy statement: emergency ultrasound imaging criteria compendium. Approved by ACEP Board of Directors April 2006; 2006.

      Table 7

      Levels of change in certainty of decisions before and after TVUSG according to categorization of VAS values

      Certainty of decision Post-TVUSG level of certainty P

      Pre-TVUSG level of certainty

      Low, n

      Moderate, n

      High, n

      Total

      Prediagnosis

      Low, n (%)

      5 (45.5)

      3 (27.3)

      3 (27.3)

      11

      .05

      Moderate, n (%)

      8 (21.6)

      9 (24.3)

      20 (54.1)

      37

      High, n (%)

      4 (10.0)

      8 (20.0)

      28 (70.0)

      40

      Additional diagnostic studies

      Low, n (%)

      13 (72.2)

      4 (22.2)

      1 (5.6)

      18

      .16

      Moderate, n (%)

      10 (27.0)

      15 (40.5)

      12 (32.4)

      37

      High, n (%)

      4 (12.1)

      8 (24.2)

      21 (63.6)

      33

      Treatment

      Low, n (%)

      15 (75.0)

      3 (15.0)

      2 (10.0)

      20

      .27

      Moderate, n (%)

      6 (17.6)

      13 (38.2)

      15 (44.1)

      34

      High, n (%)

      2 (5.9)

      7 (20.6)

      25 (73:5)

      34

      Admission

      Low, n (%)

      57 (87.7)

      6 (9.2)

      2 (3.1)

      65

      .18

      Moderate, n (%)

      6 (42.9)

      1 (7.1)

      7 (50.0)

      14

      High, n (%)

      1 (11.1)

      1 (11.1)

      7 (77.8)

      9

      Emergency surgery

      Low, n (%)

      68 (89.5)

      4 (5.3)

      4 (5.3)

      76

      .42

      Moderate, n (%)

      2 (25.0)

      4 (50.0)

      2 (25.0)

      8

      High, n (%)

      1 (25.0)

      1 (25.0)

      2 (50.0)

      4

      Discharged from ED

      Low, n (%)

      6 (75.0)

      0 (0.0)

      2 (25.0)

      8

      .10

      Moderate, n (%)

      6 (37.5)

      3 (18.8)

      7 (43.8)

      16

      High, n (%)

      2 (3.1)

      6 (9.4)

      56 (87.5)

      62

      VAS values reflecting certainty of decisions of attending emergency physicians were categorized as follows: low (0-30 mm), moderate (31-70 mm), and high (71-100 mm).