Article 1: Hand-held ultrasound as a screening tool for abdominal aortic aneurysm

Transcription

1 Article 1: Hand-held ultrasound as a screening tool for abdominal aortic aneurysm You are performing a comprehensive physical examination on a 75-year-old man who is seeing you for a first-time check up. He smokes cigarettes, has hypertension and diabetes that are controlled with medication, and had a mild myocardial infarction several years ago. During his examination you note a small pulsatile mass in his periumbilical region. You recommend a CT scan of the abdomen. As your worried patient leaves your office to drive to the radiology center across town, an eager sales representative is waiting to show you yet another piece of allegedly essential office equipment a portable ultrasound machine. I don t think this ultrasound machine will be of much help to me, you say guardedly but he hands you a copy of this article. 1. What is the purpose of this article? 2. Describe the study population and how it was selected. What characteristics of the population and the way it was selected could affect the analysis of the results? 3. What is the gold standard in this study? Could the selection of the gold standard affect the validity of the study?

2 4. Draw the 2x2 table and calculate the sensitivity, specificity, positive predictive value, and negative predictive value using the prevalence data provided in the article. The prevalence of AAA in the general population among patients aged >60 is likely between 2-8% (source: U.S. Preventive Services Task Force. Guide to Clinical Preventive Services, 2nd Edition. Washington, DC: U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion, [ Assume the prevalence changes to 5%, the total population is 1000, and the sensitivity and specificity remain the same; recalculate the predictive values. What are the implications of these results? 5. What are the likelihood ratios for this test and what is their significance? 6. What are the limitations of this study?

3 The American Journal of Surgery 186 (2003) Scientific paper A prospective study of a hand-held ultrasound device in abdominal aortic aneurysm evaluation Peter H. Lin, M.D.*, Ruth L. Bush, M.D., Sally A. McCoy, N.P., Deborah Felkai, R.V.T., Terrance K. Pasnelli, M.S., John C. Nelson, M.D., Kenneth Watts, M.D., Russell C. Lam, M.D., Alan B. Lumsden, M.D. Division of Vascular Surgery and Endovascular Therapy, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston VAMC (112), 2002 Holcomb Blvd., Houston, TX 77030, USA Manuscript received June 2, 2003; revised manuscript July 23, 2003 Abstract Purpose: Effective treatment of abdominal aortic aneurysm (AAA) requires both early detection and timely repair to reduce aneurysmrelated mortality. The purpose of this prospective study was to evaluate the utility of a hand-held ultrasonography (US) device in AAA screening in a Veterans Affairs vascular surgical service. Methods: During a 16-month period, patients with risk factors for AAA were evaluated in a blinded fashion with a hand-held US device performed by physicians. A conventional abdominal duplex US examination was also performed by a certified vascular ultrasonographer. Results of the hand-held US was compared with the conventional duplex US examination. Results: A total of 104 patients were evaluated (97 men, mean age years). The mean times for hand-held and conventional duplex US examinations were minutes and minutes (not significant), respectively. Using the conventional duplex US as a reference, the sensitivity and specificity of the hand-held device in detecting a AAA were 93% and 97%, respectively. The positive and negative predictive values of the hand-held device were 89% and 98%, respectively. The likelihood ratios of positive and negative tests of the hand-held US device examination were 82 and 0.14, respectively. The diagnostic accuracy of the hand-held US device as compared with the conventional duplex US was 98%. Conclusions: A hand-held portable US device is effective and accurate in AAA screening with results comparable to the conventional abdominal duplex examination. Moreover, hand-held portable US for AAA screening can be performed expeditiously during physical examination. It should be used as an extension in routine physical examination in vascular patients Excerpta Medica, Inc. All rights reserved. Keywords: Ultrasonography; Abdominal aortic aneurysm; Physical examination Abdominal aortic aneurysm (AAA) is the tenth leading cause of death in men in the United States. The majority of AAA-related deaths are due to aneurysm rupture. Management of ruptured AAA is commonly associated with high operative mortality, which ranges between 53% and 82% [1 4]. It is estimated that these mortality rates are frequently understated as many patients die of aneurysm rupture before reaching the hospital [5,6]. Of patients who presented to the hospital with a ruptured AAA, nearly one third of them had no prior knowledge of their aortic aneurysm [7]. * Corresponding author. Tel.: address: plin@bcm.tmc.edu Effective treatment of AAA involves early detection of AAA before aneurysms rupture. Traditional AAA screening strategy relies on thorough physical examination coupled with abdominal duplex ultrasonography (US) examination performed by trained ultrasonographers in a dedicated radiology or vascular laboratory. Recent advances in diagnostic ultrasonography have led to the development of a relatively inexpensive hand-held ultrasound unit that has wide clinical applications [8,9]. The utility of this small hand-held US device may significantly enhance a physicians ability to detect a AAA in a physical examination, analogous to what a stethoscope can provide that enables physicians to detect heart murmurs. The purpose of this prospective study was to evaluate the /03/$ see front matter 2003 Excerpta Medica, Inc. All rights reserved. doi: /j.amjsurg

4 456 P.H. Lin et al. / The American Journal of Surgery 186 (2003) utility of a portable hand-held US device in AAA screening in a vascular surgery clinic of a large VA hospital. The results of the hand-held US study were validated with a conventional abdominal duplex US performed by a certified vascular technologist. The study was designed to test the hypothesis that a hand-held US unit would enable physicians to adequately visualize the aorta and accurately detect a AAA. Methods During a 16-month period ending in April 2003, patients with risk factors for AAA referred to the vascular surgery clinic at the Houston VA Medical Center were prospectively studied. Inclusion criteria for AAA screening included two or more of the following risk factors: age greater than 60 years, heavy smoking history (more than 20-pack per day), hypertension, personal history or family history of AAA, carotid artery disease, diabetes, hyperlipidemia, coronary artery disease, claudication (more than 100 yards), ischemic rest pain, or ischemic ulceration. Patients were evaluated concomitantly with a hand-held portable US unit performed by physicians as well as a conventional abdominal duplex evaluation performed by a certified vascular ultrasonographer who was blinded to the portable US data. The portable US evaluation was performed as a part of the routine physical examination in the vascular surgical clinic, whereas the conventional abdominal duplex US was performed either during the same clinic visit or as a scheduled appointment at a different time. The portable US device (SonoSite 180 plus; SonoSite, Bothell, Washington; Fig. 1) is a small, hand-held unit with a 2.5-MHz curved linear array transducer, which is connected to a compact console weighing slightly over 4 pounds. The unit has two-dimensional gain and depth control settings similar to a conventional duplex ultrasound unit. All abdominal US evaluations using the SonoSite device were performed in the vascular clinic either by the vascular surgery staff or senior surgical residents with staff supervision. The conventional duplex ultrasound (CDU) of the abdomen was performed using a Toshiba unit (PowerVision Ultrasound, New York, New York) with a standard ultrasound transducer that was performed by a certified vascular technologist. All screening ultrasound examinations in this study were primarily focused on the identification of the infrarenal aorta and assessment of its maximal dimension. The maximal aortic diameter was measured in both the sagittal and transverse anterior-posterior dimension. No attempts were made to visualize other intraabdominal organs such as the gallbladder or liver. Measurements of the maximum transverse and longitudinal aortic dimension were performed using still-frame images and built-in measuring calipers. An aortic aneurysm is defined as a segmental enlargement of the abdominal aorta equal or greater than 30 mm in its Fig. 1. The portable ultrasonography SonoSite 180 plus model used in this study. greatest transverse diameter. The amount of time required for each abdominal ultrasound evaluation was also recorded. Descriptive statistics were reported as mean SD. The diagnostic accuracy and predictive value of the SonoSite examination was compared with the CDU result, which was considered the standard screening assessment of AAA. For statistical analysis, a comparison of risk factors between those with AAA and without AAA based on CDU examination was analyzed using the Student t test. The test results were considered significant at a P value of less than The aortic dimensions determined by either the hand-held SonoSite ultrasound unit and CDU were compared with each other using the Bland-Altman analysis. In this plot, the differences between the measurements of the two imaging modalities were plotted against their average. Results A total of 111 patients were enrolled in the study. Seven patients were excluded owing to lack of the conventional abdominal duplex US examination for comparison because they either refused or were lost to follow-up. The remaining 104 patients formed the basis of data analysis in this study. The mean age of the patients was years with a range of 52 to 84 years. Table 1 lists the pertinent epidemiologic characteristics and cardiovascular risk factors in patients screened for aortic aneurysm based on CDU exam-

5 P.H. Lin et al. / The American Journal of Surgery 186 (2003) Table 1 Comparison of relative risk factors in patients with and without abdominal aortic aneurysm (AAA) based on conventional duplex ultrasonography screening examination Characteristic AAA ( 3 cm) n 15 No AAA n 89 P value Sex (male) 92% 93% NS White race 84% 80% NS Black race 9% 13% NS Chronic obstructive 24% 20% NS pulmonary disease Family history of AAA 17% 13% NS Diabetes 24% 20% NS Smoking history 61% 73% NS Hypertension 69% 40% NS Hyperlipidemia 38% 27% NS Coronary artery disease 47% 33% NS Peripheral arterial 66% 40% NS disease Cerebrovascular disease 24% 20% NS NS not significant. ination, which identified 89 patients with AAA and 15 patients without AAA. No significant differences were noted in the incidence of relative risk factors between those with aneurysms and those without aneurysms (Table 1). Table 2 indicates the reasons in which these patients were referred to the vascular clinic visit. Patients with a known history of AAA were more likely to have an aortic aneurysm compared with those without an aneurysm based on CDU screening examination (11% versus 0%, P 0.02). There was no difference in the remaining indications of clinic visit between patients with and without aortic aneurysm (Table 2). Visualization of the infrarenal aorta was achieved in all patients in both SonoSite and CDU modalities. The mean times for hand-held SonoSite and CDU examinations were minutes and minutes (not significant), respectively. The results of mean aortic measurement using Table 2 Comparison of indication of clinic visit in patients with and without abdominal aortic aneurysm (AAA) based on conventional duplex ultrasonography screening examination Reason for clinic visit AAA ( 3 cm) n 15 No AAA n 89 P value Claudication 35% 47% NS Ischemic rest pain 13% 20% NS Gangrene/ischemic 8% 13% NS ulcer Carotid artery disease 18% 7% NS Known history of AAA 11% 0% 0.02 Hemodialysis access 12% 7% NS evaluation Other 2% 7% NS NS not significant. Table 3 Comparison of aortic measurement by SonoSite and conventional duplex ultrasonography (CDU) examinations Examination modality Mean aortic diameter (mm) either the SonoSite or CDU modality are summarized in Table 3. No statistically difference in aortic dimension was found between the two imaging modalities. The comparison of the maximal aortic diameter equal to or greater than 30 mm when examined by the SonoSite or CDU modalities is demonstrated in Table 4. Eighty-six patients were found to have an aortic diameter less than 30 mm as determined by both the SonoSite and CDU examinations. In contrast, 14 AAAs (maximal aortic diameter equal to or greater than 30 mm) were identified by both ultrasound modalities. One patient was found to have an aortic diameter less than 30 mm by SonoSite unit, whereas the CDU examination determined it to be greater than 30 mm. In contrast, 3 patients were found to have aortic aneurysms of equal to or greater than 30 mm by SonoSite examination, whereas the CDU examination determined that they were less than 30 mm. By using the CDU calculation as the reference measurement, the sensitivity and specificity of the SonoSite evaluation were 93% (95% confidence interval [CI]: 0.79 to 0.96) and 97% (95% CI: 0.93 to 0.99), respectively. The positive predictive value of the SonoSite examination was 89% (95% CI: 0.78 to 0.96), and the negative predictive value was 98% (95% CI: 0.95 to 1.00). The likelihood ratios of positive and negative tests of the SonoSite examination were 82 and 0.14, respectively. The diagnostic accuracy of the SonoSite study as compared with the CDU reference was 98% (95% CI: 0.94 to 1.00). Among the 17 patients with an aortic diameter equal to or greater than 30 mm as determined by the SonoSite evaluation, 9 of them underwent additional imaging studies with either abdominal computed tomography (CT [n 6]) or magnetic resonance angiography (MRA [n 3]). All 8 patients had aortic diameters that were confirmed to be greater than 30 mm. We also assessed the correlation of CDU results of patients with aortic aneurysm with additional CT scan or MRA studies. Among the 15 patients with an aortic diameter equal to or greater than 30 mm as as- SD Range (mm) SonoSite CDU Table 4 Maximal aortic diameter as determined by the SonoSite and conventional duplex ultrasonography (CDU) examinations CDU ( 30 mm) CDU ( 30 mm) Total SonoSite ( 30 mm) 86 (83%) 1 (1%) 87 (84%) SonoSite ( 30 mm) 3 (3%) 14 (13%) 17 (16%) Total 89 (86%) 15 (14%) 104 (100%)

6 458 P.H. Lin et al. / The American Journal of Surgery 186 (2003) Fig. 2. Scatter plot of the maximum aortic diameter as determined by the hand-held ultrasonography device and conventional duplex ultrasonography modalities. sessed by the CDU examination, 7 patients had additional imaging studies (abdominal CT, n 6) or MRA (n 1). All 7 patients had aortic dimensions that were confirmed to be either equal to or greater than 30 mm in diameter. The scatter plot of the maximum aortic diameter as determined by the SonoSite and CDU modalities is shown in Fig. 2. The scatter plot reveals a highly correlated AAA assessment between the two imaging modalities. The degree of the difference as measured by these two ultrasound examinations in the Bland-Altman scatter plot is shown in Fig. 3, which demonstrates a high correlation between the two imaging modalities in evaluating aortic aneurysms greater than 30 mm in diameter. Comments This study demonstrated the benefit of a portable handheld US device in screening for AAA with diagnostic accuracy comparable to the conventional abdominal duplex Fig. 3. Bland-Altman plot displaying the degree of the difference as measured by these two ultrasonography examinations versus the reference value based on the conventional duplex ultrasonography measurement. US evaluation. In addition, the diagnostic accuracy of this portable US unit when performed by physicians is similar to that of the conventional duplex US examination performed by a certified ultrasonographer. The hand-held portable unit can be used expeditiously to determine the aortic dimension during a routine physical examination without the potential delay associated with scheduling a formal abdominal duplex study. Our finding underscores the importance of using such a portable US unit in evaluating a vascular patient for AAA during a physical examination. The high diagnostic accuracy of the hand-held SonoSite device in detecting AAA underscores the importance of using this diagnostic modality in screening AAA. The clinical benefit of using such a portable ultrasound device in screening aortic aneurysm has similarly been reported in several clinical investigations [10,11]. Similar to our findings, these studies noted high diagnostic accuracy in evaluating patients with abdominal aortic aneurysms. The findings of our study strongly support the concept that this portable diagnostic modality should be used to evaluate vascular patients in routine physical examination. Clinicians have traditionally relied on various hand-held adjunctive tools to aid physical examination for detecting certain illnesses. For instance, a stethoscope is an essential device to detect heart murmur or assess lung sounds. An otoscope enables an otolaryngologist to assess potential ear infections in pediatric patients. An ophthalmoscope is critical to allow a physician to perform a fundoscopic examination. These hand-held diagnostic tools have greatly enhanced physicians ability to accurately detect clinical pathology and broaden their ability to make treatment recommendations based on the diagnostic accuracy of these portable devices. We believe that the SonoSite hand-held US unit should be an integral part of a physical examination when assessing for a potential AAA. The findings of our study have prompted efforts at our own institution to implement hand-held ultrasonography in physical examination in the medical school curriculum. Current medical students at our institution are taught to use the hand-held US unit when performing an abdominal examination. Advances in US technology has led to the production of a smaller and lighter portable US unit. The new miniaturized model (ilook 25 model), which weights only 3 pounds, undoubtedly enhances the portable benefit of this imaging equipment. These portable US devices have already been shown to play a critical role in physical examination in certain clinical scenarios [12,13]. For instance, a portable US device is ideally suited to assess biliary echogenicity to determine the origin of right upper quadrant abdominal pain. The hand-held US device is similarly suited to detect free abdominal fluid in trauma patients who sustain a stab wound to the abdomen, in lieu of more invasive diagnostic peritoneal lavage or expensive abdominal computed tomography. Clearly the benefit of a conventional duplex US evaluation in these settings has been firmly established [14 16]. The portable hand-held US device may

7 P.H. Lin et al. / The American Journal of Surgery 186 (2003) permit the physician to perform a more rapid screening evaluation to provide immediate diagnostic information, particularly when a formal duplex scanning is not immediately available due in part to off hours or unavailability of an ultrasonographer. There are several limitations of our study. The relatively small patient sample in our study may not adequately reflect the true sensitivity or specificity of the portable US evaluation. Moreover, the diagnostic accuracy of the hand-held US study in the majority of our patients was compared with a conventional duplex US study, rather than an abdominal CT scan which is commonly regarded as the gold standard in preoperative AAA evaluation. In addition, the hand-held US evaluations in this study were performed either by the vascular staff or surgical residents, which were compared with the conventional duplex scan that was performed by an experienced ultrasonographer. The disparity in ultrasound experience between the physician and ultrasound technician may in part be attributed to the interobserver variability of these two imaging modalities. There was indeed a greater learning curve in surgical residents who have no prior experience in vascular ultrasonography to learn to use the portable US device. Many studies have clearly illustrated that the accuracy of an US examination is dependent on the operator s experience [17,18]. Previous efforts to incorporate a portable US examination in the general surgical resident s training on trauma patient evaluation have yielded successful results with high diagnostic accuracy [19 22]. With proper training and supervision, portable US examination for AAA screening undoubtedly can be performed in the hands of surgical residents with high diagnostic accuracy. In summary, our study demonstrates the feasibility of utilizing a portable hand-held US device in AAA screening with a high degree of accuracy. This diagnostic modality can be performed quickly and yields potential valuable information in screening patients for AAA. The continual efforts in the ultrasound industry to refine and miniaturize these portable units may one day make them an integral diagnostic tool in a comprehensive physical examination, much the way that otoscopes or ophthalmoscopes do. More emphasis in hands-on ultrasound training will undoubtedly be beneficial in medical education and residency training so physicians can readily incorporate this diagnostic modality into their future clinical practice. References [1] Scott RA, Tisi PV, Ashton HA, Allen DR. Abdominal aortic aneurysm rupture rates: a 7-year follow-up of the entire abdominal aortic aneurysm population detected by screening. J Vasc Surg 1998;28: [2] Gloviczki P, Pairolero PC, Mucha P, et al. Ruptured abdominal aortic aneurysms: repair should not be denied. J Vasc Surg 1992;15: [3] Noel AA, Gloviczki P, Cherry KJ, et al. Ruptured abdominal aortic aneurysms: the excessive mortality rate of conventional repair. J Vasc Surg 2001;34:41 6. [4] Prance SE, Wilson YG, Cosgrove CM, et al. Ruptured abdominal aortic aneurysms: selecting patients for surgery. Eur J Vasc Endovasc Surg 1999;17: [5] Belkin M, Donaldson MC, Whittemore AD. Abdominal aortic aneurysms. Curr Opin Cardiol 1994;9: [6] Johansen K. Ruptured abdominal aortic aneurysm: how should recent outcome studies impact current practices? Semin Vasc Surg 1995;8: [7] Law M. Screening for abdominal aortic aneurysms. Br Med Bull 1998;54: [8] Burdjalov V, Srinivasan P, Baumgart S, Spitzer AR. Handheld, portable ultrasound in the neonatal intensive care nursery: a new, inexpensive tool for the rapid diagnosis of common neonatal problems. J Perinatol 2002;22: [9] Kirkpatrick AW, Simons RK, Brown R, et al. The hand-held FAST: experience with hand-held trauma sonography in a level-i urban trauma center. Injury 2002;33: [10] Vourvouri EC, Poldermans D, Schinkel AF, et al. Abdominal aortic aneurysm screening using a hand-held ultrasound device. A pilot study. Eur J Vasc Endovasc Surg 2001;22: [11] Bruce CJ, Spittell PC, Montgomery SC, et al. Personal ultrasound imager: abdominal aortic aneurysm screening. J Am Soc Echocardiogr 2000;13: [12] Rozycki GS, Newman PG. Surgeon-performed ultrasound for the assessment of abdominal injuries. Adv Surg 1999;33: [13] Eachempati SR, Barie PS. Minimally invasive and noninvasive diagnosis and therapy in critically ill and injured patients. Arch Surg 1999;134: [14] Lefrant JY, Cuvillon P, Benezet JF, et al. Pulsed Doppler ultrasonography guidance for catheterization of the subclavian vein: a randomized study. Anesthesiology 1998;88: [15] Poelaert J, Schmidt C, Colardyn F. Transoesophageal echocardiography in the critically ill. Anaesthesia 1998;53: [16] Roewer N, Greim CA. Echocardiography in intensive care medicine. Acta Anaesthesiol Scand 1997;111(suppl):87 9. [17] Drost WT, Mattoon JS, Samii VF, et al. A retrospective study into the effects of operator experience on the accuracy of ultrasound in the diagnosis of gastric neoplasia in dogs. Vet Radiol Ultrasound 2001; 42:358. [18] Rowland JL, Kuhn M, Bonnin RL, et al. Accuracy of emergency department bedside ultrasonography. Emerg Med (Fremantle) 2001; 13: [19] Rozycki GS. Surgeon-performed ultrasound: its use in clinical practice. Ann Surg 1998;228: [20] Rozycki GS, Ballard RB, Feliciano DV, et al. Surgeon-performed ultrasound for the assessment of truncal injuries: lessons learned from 1540 patients. Ann Surg 1998;228: [21] Rozycki GS, Ochsner MG, Jaffin JH, Champion HR. Prospective evaluation of surgeons use of ultrasound in the evaluation of trauma patients. J Trauma 1993;34: [22] Rozycki GS, Pennington SD, Feliciano DV. Surgeon-performed ultrasound in the critical care setting: its use as an extension of the physical examination to detect pleural effusion. J Trauma 2001;50: