Scanner-Based Protocol-Driven Ultrasound: An Effective Method to Improve Efficiency in an Ultrasound Department
Abstract
OBJECTIVE. For ultrasound, a wide variation is often observed among the number and sequence of images acquired for a particular examination type. Scanner-based protocols are preset pathways in the ultrasound machine that guide a sonographer through the required study images. These protocols can streamline image acquisition by improving consistency and efficiency of ultrasound examinations. This study evaluated whether implementation of scanner-based protocol-driven ultrasound improves efficiency by decreasing the scanning duration and number of images acquired.
MATERIALS AND METHODS. Retrospective evaluation of 437 carotid Doppler examinations, 395 complete abdominal ultrasound examinations with Doppler imaging, and 413 bilateral lower extremity venous Doppler examinations for deep venous thrombosis (DVT) performed by five sonographers before and after implementation of scanner-based protocol-driven ultrasound was performed. The scanning duration and number of images acquired for each study were recorded. Statistical analysis compared the scanning duration and number of images acquired before and after implementation of protocol-driven ultrasound. A p value of < 0.05 was considered significant.
RESULTS. A significant decrease in scanning duration occurred for both carotid Doppler ultrasound examinations (decrease by 12.4% [2.7 minutes], p < 0.0001) and complete abdominal ultrasound examinations with Doppler imaging (decrease by 7.5% [2.0 minutes], p = 0.0054) after implementation of protocol-driven ultrasound. The decrease in scanning duration was not significant for lower extremity DVT Doppler examinations (p = 0.4192). In addition, there was a significant decrease in the overall number of images obtained for all three types of studies.
CONCLUSION. Scanner-based protocol-driven ultrasound is an effective method that streamlines image acquisition and significantly improves efficiency in an ultrasound department while ensuring consistency and adherence to accreditation guidelines.
Individual institutions often have ultrasound protocols containing necessary images for a particular type of study, usually in keeping with guidelines established by accreditation agencies such as the American College of Radiology (ACR) [1]. However, ultrasound is an operator-dependent modality, and each sonographer has an individual style of performing an ultrasound examination. For the more complex protocols or for patients with numerous positive findings, there is opportunity for the sonographer to overlook acquisition of a specific image. For instance, a dedicated image of the inferior vena cava in a complex abdominal ultrasound may not be included if the sonographer is focused on obtaining images of multiple liver masses. In our experience, sonographers new to our institution may need time to adjust to the new protocols, and there is an increased risk for a required image to be omitted from the examination. Differences in levels of experience and confidence among sonographers may result in wide variations in the number and sequence of images obtained for the same type of examination.
Scanner-based ultrasound protocols are preset pathways that guide a sonographer through the list of images that are required for a particular study. Many modern ultrasound scanners have the capability to create user-generated protocols for specific examinations. The protocols are prospectively created to meet the needs and preferences of an ultrasound department and are useful to ensure adherence to national standards such as those used for ACR accreditation. Once an ultrasound department finalizes the list and sequence of images required for a particular study, it can enlist the help of the manufacturers' application specialists to assist uploading the protocol. The protocol can then be distributed to all machines to promote a standardized set of images and order of acquisition, which can aid physicians during interpretation. Within the scanner, a set of prompts with preset labels prompts the sonographer to select for the next image in the required dataset. The protocol can be paused at any time to allow the sonographer to choose additional images, but the protocol will resume with additional prompts until the entire dataset is obtained. The use of scanner-based ultrasound protocols can streamline image acquisition to increase efficiency and consistency of ultrasound studies [2, 3]. Considering current external and governmental mandates regarding quality and the need for efficiency because of reimbursement constraints, there is an ongoing need for research into ways to reduce the variability and duration of imaging studies while maintaining the required components.
The purpose of this study was to evaluate the effect of implementation of scanner-based protocols on three different but commonly performed types of ultrasound examinations: carotid Doppler examinations, complete abdominal ultrasound examinations with Doppler imaging, and bilateral lower extremity deep venous thrombosis (DVT) Doppler ultrasound examinations. We hypothesized that implementation of scanner-based ultrasound protocols could decrease the scanning duration and number of images obtained to complete diagnostic quality ultrasound examinations that meet ACR accreditation guidelines.
Materials and Methods
This HIPAA-compliant retrospective study was performed after being approved by the institutional review board (IRB). The IRB did not require informed consent for this retrospective evaluation.
Study Population
A retrospective evaluation of 437 carotid Doppler ultrasound examinations, 395 complete abdominal ultrasound examinations with Doppler imaging, and 413 bilateral lower extremity DVT Doppler examinations performed from January 2009 through December 2012 by five experienced registered diagnostic medical sonographers was performed. At the start of the study period in 2009, the experience level of the sonographers was as follows: sonographer 1, 8 years; sonographer 2, 28 years; sonographer 3, 13 years; sonographer 4, 31 years; and sonographer 5, 18 years. All studies were performed at the outpatient imaging center of a tertiary care academic center to decrease heterogeneity in patients' abilities to cooperate due to illness; approximately 16,000 ultrasound examinations per year are performed at this center. Two hundred nineteen carotid Doppler examinations, 197 abdominal ultrasound examinations with Doppler imaging, and 209 lower extremity DVT examinations were performed before the introduction of scanner-based protocol-driven ultrasound. Two hundred eighteen carotid Doppler examinations, 198 abdominal ultrasound examinations with Doppler imaging, and 204 lower extremity DVT examinations were performed at least 6 months after implementation of protocol-driven ultrasound. No examination from the first 6 months after implementation of the scanner-based protocol-driven ultrasound was included to allow time for the sonographers to acclimate to the new workflow. Examinations were selected in consecutive order. The inclusion criterion was any patient with one of the three ultrasound studies performed by any of the five sonographers during the study period. Studies that did not have images archived were excluded.
Protocol-Driven Ultrasound at Our Institution
In scanner-based protocol-driven ultrasound, the protocol in the ultrasound scanner guides the sonographer through the sequence of images required for a particular examination. When a sonographer begins the study and selects a patient to scan, a list of protocols available for various studies populates the screen. The sonographer then launches the desired protocol: in this study, a complete abdominal ultrasound examination with Doppler imaging, carotid Doppler examination, or lower extremity DVT ultrasound examination. After the desired protocol is launched, a screen prepopulated with the image annotation for the first required image for that particular study appears on the screen (e.g., transverse right proximal common carotid artery for the carotid Doppler protocol). Once the sonographer saves the required image, the annotation for the next required image appears on the screen. The left side of the screen has the entire list of required images for that particular study, each with an adjacent check box. Check marks appear adjacent to the list of required images as the sonographer progresses through the study (Fig. 1). The sonographer can exit and reenter the protocol at any point to obtain additional images as needed.
Fig. 1A —Scanner-based protocol-driven ultrasound.
A, Sonographer selects type of examination from list of preloaded protocols to launch protocol. In this example, Carotid University of Alabama at Birmingham protocol is selected.
Fig. 1B —Scanner-based protocol-driven ultrasound.
B, As sonographer proceeds through examination, check marks (arrow) appear on left side of screen in front of required list of images as they are obtained. Annotation for next required image automatically appears at bottom of screen (arrowhead) as soon as image is saved. In this example, next image is right common carotid artery proximal longitudinal.
After completion of each study, the sonographer reviews the images with a radiologist or physician extender to ensure that diagnostic quality images have been obtained and that the dataset contains the required images according to the accrediting body and local preference.
Scanning Duration and Number of Images
The ultrasound examinations before implementation of protocol-driven ultrasound were acquired using iU22 (Philips Healthcare Ultrasound) and Sequoia (Siemens Healthcare Ultrasound) scanners. The postprotocol ultrasound examinations were performed using iU22 scanners.
To assess the efficiency of the ultrasound examinations performed before and after the implementation of protocol-driven ultrasound, we recorded the scanning duration and number of images in each complete abdominal ultrasound examination with Doppler imaging, carotid Doppler examination, and lower extremity DVT examination. The amount of time required for each ultrasound examination was calculated from the time stamps on the first and last images. If the study had a break of more than 4 minutes between two images, the duration of the break was subtracted from the total duration of the study. The time for breaks lasting more than 4 minutes was subtracted because such a long break between images would likely have occurred because the sonographer or patient left the room during the study. A break could occur because the sonographer left the room to get the images checked by the physician and then returned to acquire additional images. Any studies that were discontinued before completion were also excluded, although discontinuation of a study is extremely rare in the out-patient setting. Determination of adequate study quality and of the completeness of the image data-set was made by a radiologist or physician extender at the time of the ultrasound study. If an examination was not completed is accordance with ACR guidelines, the sonographer was asked to obtain the required images. Figure 2 is a schematic diagram of the workflow.
Statistical Analysis
Statistical analysis was performed by a biostatistician. Type 3 tests of fixed effects were used to compare the scanning duration and number of images acquired before and after implementation of protocol-driven ultrasound for each individual sonographer and for the entire group of sonographers. A p value of < 0.05 was prospectively considered to be statistically significant.
Results
The numbers of examinations performed by each sonographer before and after implementation of the scanner-based protocols are similar, as shown in Table 1.
| Sonographer | No. of Examinations in the Preprotocol Period | No. of Examinations in the Postprotocol Period | ||||
|---|---|---|---|---|---|---|
| Carotid Doppler Imaging | Abdominal Ultrasound With Doppler Imaging | Bilateral Lower Extremity DVT Doppler Imaging | Carotid Doppler Imaging | Abdominal Ultrasound With Doppler Imaging | Bilateral Lower Extremity DVT Doppler Imaging | |
| Sonographer 1 | 70 | 40 | 40 | 25 | 40 | 36 |
| Sonographer 2 | 48 | 38 | 51 | 51 | 38 | 43 |
| Sonographer 3 | 46 | 40 | 41 | 58 | 40 | 46 |
| Sonographer 4 | 25 | 40 | 40 | 34 | 40 | 41 |
| Sonographer 5 | 30 | 39 | 37 | 50 | 40 | 38 |
| Total | 219 | 197 | 209 | 218 | 198 | 204 |
Note—DVT = deep venous thrombosis.
Scanning Duration
Compared with the results for the examinations performed before implementation of protocol-driven ultrasound, the mean scanning duration of the carotid Doppler ultrasound examinations decreased by 12.4%, (p < 0.0001) and the mean scanning duration of the complete abdominal ultrasound examinations with Doppler imaging decreased by 7.5% (p = 0.0054) after implementation. There was a 4.5% decrease in the scanning duration of lower extremity DVT ultrasound examinations after implementation of protocol-driven ultrasound, but this difference is not statistically significant (p = 0.4192).
For the carotid Doppler ultrasound examinations (Table 2), four of the five sonographers significantly reduced their individual scanning times after implementation of protocol-driven ultrasound. One sonographer's (sonographer 5) scanning time increased (mean, 0.2 minute) after implementation of protocol-driven ultrasound, although this change was not statistically significant. This sonographer had the lowest mean scanning durations for all three examination types before implementation of protocol-driven ultrasound.
| Sonographer | Scanning Duration (min) | Difference in Scanning Duration Between Preprotocol and Postprotocol Examinations (min) | p | |
|---|---|---|---|---|
| Before the Protocol | After the Protocol | |||
| Sonographer 1 | 17.0 | 12.4 | 4.6 | < 0.0001 |
| Sonographer 2 | 30.3 | 25.1 | 5.2 | < 0.0001 |
| Sonographer 3 | 20.0 | 17.4 | 2.6 | 0.0055 |
| Sonographer 4 | 29.0 | 22.4 | 6.6 | < 0.0001 |
| Sonographer 5 | 15.3 | 15.5 | −0.2 | 0.8939 |
| All studies | 21.7 | 19.0 | 2.7 | < 0.0001 |
Four of the five sonographers reduced their individual scanning times for complete abdominal ultrasound examinations with Doppler imaging after implementation of protocol-driven ultrasound, but only one sonographer's time reduction was statistically significant (Table 3). One sonographer's (sonographer 5) scanning time again increased slightly (0.2 minute) after implementation of protocol-driven ultrasound, but this difference was not statistically significant.
| Sonographer | Scanning Duration (min) | Difference in Scanning Duration Between Preprotocol and Postprotocol Examinations (min) | p | |
|---|---|---|---|---|
| Before the Protocol | After the Protocol | |||
| Sonographer 1 | 20.7 | 18.6 | 2.1 | 0.1747 |
| Sonographer 2 | 34.7 | 34.2 | 0.5 | 0.7520 |
| Sonographer 3 | 20.9 | 19.3 | 1.6 | 0.2995 |
| Sonographer 4 | 37.0 | 31.4 | 5.6 | 0.0002 |
| Sonographer 5 | 19.3 | 19.5 | −0.2 | 0.8928 |
| All studies | 26.5 | 24.5 | 2.0 | 0.0054 |
For the lower extremity DVT ultrasound examinations, all sonographers individually reduced their scanning times, but these decreases were not statistically significant for any of them (Table 4).
| Sonographer | Scanning Duration (min) | Difference in Scanning Duration Between Preprotocol and Postprotocol Examinations (min) | p | |
|---|---|---|---|---|
| Before the Protocol | After the Protocol | |||
| Sonographer 1 | 11.5 | 11.3 | 0.2 | 0.9097 |
| Sonographer 2 | 19.0 | 18.7 | 0.3 | 0.8570 |
| Sonographer 3 | 14.2 | 12.3 | 1.9 | 0.2559 |
| Sonographer 4 | 21.4 | 20.9 | 0.5 | 0.7661 |
| Sonographer 5 | 10.9 | 10.7 | 0.2 | 0.9106 |
| All studies | 15.6 | 14.9 | 0.7 | 0.4192 |
All sonographers included were highly experienced (8–31 years of experience), and the results were not affected by sonographer experience.
Number of Images
After implementation of protocol-driven ultrasound, there were significant decreases in the overall mean number of images obtained for carotid Doppler examinations (6.2 fewer images; p < 0.0001), abdominal ultrasound examinations with Doppler imaging (9.2 fewer images, p < 0.0001), and bilateral lower extremity DVT examinations (4.0 fewer images, p < 0.0001) (Table 5).
| Sonographer | Mean No. of Images | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Carotid Doppler Ultrasound Examinations | Abdominal Ultrasound Examinations With Doppler Imaging | Lower Extremity DVT Examinations | |||||||
| Preprotocol Period | Postprotocol Period | Differencea (p) | Preprotocol Period | Postprotocol Period | Differencea (p) | Preprotocol Period | Postprotocol Period | Differencea (p) | |
| Sonographer 1 | 51.8 | 38.1 | 13.6 (< 0.0001) | 65.2 | 48.8 | 16.4 (< 0.0001) | 34.9 | 31.2 | 2.8 (0.0051) |
| Sonographer 2 | 51.5 | 41.2 | 10.3 (< 0.0001) | 61.0 | 54.7 | 6.3(0.0157) | 44.2 | 36.2 | 6.8 (< 0.0001) |
| Sonographer 3 | 41.6 | 39.0 | 2.6(0.0156) | 55.0 | 49.0 | 6.0(0.0195) | 33.2 | 30.7 | 2.5 (0.0396) |
| Sonographer 4 | 50.4 | 45.9 | 4.5 (0.0018) | 68.4 | 60.4 | 8.0 (0.0018) | 38.4 | 35.3 | 3.1 (0.0152) |
| Sonographer 5 | 42.0 | 45.4 | −3.4 (0.0072) | 67.3 | 58.1 | 9.2 (0.0004) | 34.5 | 33.5 | 1.0 (0.4442) |
| All studies | 48.1 | 41.9 | 6.2 (< 0.0001) | 63.4 | 54.2 | 9.2 (< 0.0001) | 37.4 | 33.4 | 4.0 (< 0.0001) |
Note—DVT = deep venous thrombosis.
a
Difference in the mean number of images between the preprotocol period and the postprotocol period.
Discussion
Although increasing attention is being paid to optimizing workflow in an effort to improve productivity and efficiency of imaging departments, there is a paucity of workflow optimization data in the literature regarding ultrasound [1, 4–6]. Workflow optimization can take many forms, involving scheduling, the initial patient encounter, image acquisition, and physician interpretation [6–10]. The use of scanner-based ultrasound protocols improved the efficiency of one aspect of work-flow in our ultrasound department for two of the three types of examinations by significantly reducing the time required for each study. Although not statistically significant, a decrease in scanning duration was also noted for the third type of examination evaluated (i.e., lower extremity DVT Doppler examinations).
The use of preloaded protocols simplifies the workflow for sonographers. Throughout a study, the sonographer is aware of exactly which image needs to be obtained because the annotation for the next required image automatically prepopulates the screen as soon as an image is saved. These prompts can substantially reduce the number of keystrokes for the sonographer and ensure that all required images are obtained [2].
Although a decrease of a few minutes in examination time may not seem striking to the casual observer, it has greater importance when that time saving is reproduced hundreds of times each month. For a sonographer who averages 250 examinations per month, a 2-minute difference would yield 500 minutes saved or the equivalent of the scanning time for 20–30 additional examinations.
In our study, there were significant reductions in the mean number of images obtained per study after implementation of scanner-based protocols. These reductions can be explained by the fact that scanner-based protocols guide sonographers through a study, so the likelihood of obtaining redundant images or images not required for that particular study decreases. However, the sonographer does not lose the ability to obtain additional images to document an abnormality because he or she can exit and reenter the protocol at any point. In other words, the protocol-required images are the minimum—not the maximum—number of images that can be obtained for an individual study. In our institution, examinations are routinely checked by a radiologist or physician extender before the patient leaves the clinic. This check ensured that the reduction in the number of images obtained in the post-protocol period did not compromise the diagnostic quality of the examinations.
Although all the included sonographers had considerable experience, a wide variation in scanning time for the same type of examination was noted among the sonographers. Implementation of scanner-based protocols did not decrease the scanning time of our fastest sonographer, but it had an impact on the four other sonographers with longer scanning times. Most practices likely have a variation among sonographers; although implementation of scanner-based protocols may not be as beneficial to the fastest sonographers, it can help increase the overall efficiency of the department. In addition, we believe that the potential benefits of implementation of scanner-based protocols in terms of time efficiency and examination quality may be even greater for junior sonographers as they become familiar with a department's standards.
Although most ultrasound departments have written ultrasound protocols, adherence to and understanding of these protocols are not uniform among all sonographers. One of the advantages of scanner-based protocols is that by guiding sonographers through a study, the software promotes adherence to accreditation and departmental guidelines. It also presents the radiologist with a consistent sequence of images, which facilitates image interpretation. Although we did not specifically evaluate this aspect of implementation, the reduction in the number of keystrokes required by sonographers improves ergonomics and could potentially reduce some work-related injuries for sonographers.
The main limitation of this study is its retrospective design. There are factors that may have occurred at the time of scanning that cannot be fully accounted for by the images and times, but the number of evaluated studies helps to reduce the impact of any outliers. Another limitation is that only three types of ultrasound examination were evaluated. Of the three types evaluated, lower extremity DVT Doppler ultrasound examinations did not show a significant decrease in scanning duration after implementation of scanner-based protocols. Therefore, the impact of scanner-based protocol-driven ultrasound on different types of examinations is varied and was not fully assessed in this study. Last, the results of our assessments of patients in the outpatient setting may not translate to sicker inpatients who can present additional challenges that can increase scanning duration.
In conclusion, the introduction of scanner-based ultrasound protocols is an effective method that streamlines image acquisition and significantly improves the efficiency of image acquisition in an ultrasound department. Certain types of ultrasound studies, such as carotid Doppler examinations, may benefit more than others from using this technique.
Footnote
Based on a presentation at the Radiological Society of North America 2013 annual meeting, Chicago, IL.
References
1.
Robbin ML, Lockhart ME, Weber TM, et al. Ultrasound quality and efficiency: how to make your practice flourish. J Ultrasound Med 2011; 30:739–743
2.
Brandli L. Benefits of protocol-driven ultrasound exams. Radiol Manage 2007; 29:56–59
3.
Marshall H, Dimitrijevic E, Mendelson E, Weber S. Preprogrammed protocol for survey breast ultrasound: does it promote workflow efficiency? (abstract) In: 108th ARRS annual meeting abstract book. Leesburg, VA: ARRS, 2009:A36
4.
Reiner B, Siegel E, Carrino JA. Workflow optimization: current trends and future directions. J Digit Imaging 2002; 15:141–152
5.
Coffin CT. The continuous improvement process and ergonomics in ultrasound department. Radiol Manage 2013; 35:22–25; quiz, 26–27
6.
Badano LP, Nucifora G, Stacul S, et al. Improved workflow, sonographer productivity, and cost-effectiveness of echocardiographic service for inpatients by using miniaturized systems. Eur J Echocardiogr 2009; 10:537–542
7.
Espinoza J, Good S, Russell E, Lee W. Does the use of automated fetal biometry improve clinical work flow efficiency? J Ultrasound Med 2013; 32:847–850
8.
Sanguinetti K. Implementing volume ultrasound workflow. Radiol Manage 2008; 30:54–56
9.
Li MF, Tsai JC, Chen WJ, Lin HS, Pan HB, Yang TL. Redefining the sonography workflow through the application of a departmental computerized workflow management system. Int J Med Inform 2013; 82:168–176
10.
Lu L, Li J, Gisler P. Improving financial performance by modeling and analysis of radiology procedure scheduling at a large community hospital. J Med Syst 2011; 35:299–307
Information & Authors
Information
Published In
Copyright
© American Roentgen Ray Society.
History
Submitted: May 12, 2015
Accepted: September 4, 2015
First published: February 11, 2016
Keywords
Authors
Metrics & Citations
Metrics
Citations
Export Citations
To download the citation to this article, select your reference manager software.