Complications in Vascular Surgery

Jonathan B. Towne, Larry H. Hollier, Jonathan B. Towne, Larry H. Hollier

• Publisher: CRC Press
• Year: 2004
• eBook ISBN: 9781135526061
• Topic: Medicine
• Subtopic: Surgery & Surgical Medicine

Substantially revised to reflect the most recent surgical techniques and practices, this reference describes the most effective strategies to prevent, identify, and manage complications in vascular surgery-guiding surgeons through patient selection; instances of entrapment, malpositioning, and rupture; emerging endovascular treatments; and specific

1
Pitfalls of Noninvasive Vascular Testing

Dennis F.Bandyk
University of South Florida College of Medicine, Tampa, Florida, U.S.A.

Diagnostic testing of patients with vascular disease requires a thorough understanding of the instrumentation, arterial and venous anatomy, and hemodynamics of blood circulation. Although physical examination and vascular imaging studies—such as contrast arteriography, magnetic resonance angiography, and contrast-enhanced computed tomography—are indispensable in the management of peripheral vascular disease, noninvasive vascular testing retains a prominent role in patient evaluation both prior to and following intervention. Methods that use Doppler ultrasound—in particular duplex ultrasonography—and plethysmography form the cornerstone of noninvasive vascular testing. Testing is used to detect and grade the severity of cerebrovascular, peripheral arterial, and peripheral venous disease and thereby assists in disease management. The accuracy of duplex ultrasonography coupled with indirect physiological testing methods is superior to clinical evaluation alone and in many patients provides sufficient anatomical information of disease extent and severity to proceed with surgical or endovascular intervention without confirmatory imaging studies. Newer enhancements of duplex ultrasonography—such as power Doppler angiography, sonographic composite imaging, and the use of contrast agents—have further improved anatomic resolution, which allows better characterization of the extent and morphology of vascular disease.

Noninvasive vascular testing methods measure biophysical properties of the circulation (e.g., pressure, pulse contour, blood-flow velocity, turbulence-disturbed flow); these measurements can be used for disease localization and classification of severity. The diagnostic accuracy of each technique depends on the precision and reproducibility of the measurements. The measurements of blood pressure and velocity recorded by vascular laboratory instrumentation should not always be assumed to be accurate, since testing can be affected by a number of factors, including biological variability, instrumentation employed for testing, the skill and bias of the examiner, the pathological process being studied, and conditions under which the measurements are recorded. Interpretation of noninvasive testing studies requires an appreciation of the limitations, pitfalls, and artifacts associated with the various diagnostic techniques. Errors in any of these areas can result in an incorrect diagnosis and subsequent clinical decision making.

In this chapter, the common pitfalls of noninvasive vascular testing related to instrumentation, testing protocols, and diagnostic interpretation are reviewed and techniques to minimize their occurrence are outlined.

I. CLASSIFICATION OF THE PITFALLS OF NONINVASIVE TESTING

Diagnostic errors associated with noninvasive vascular testing can result from procedural, interpretative, or statistical pitfalls. Procedural pitfalls can be due to the instrumentation, to deviations from the testing protocol, or to biological variability of the measurement. Interpretative pitfalls can decrease diagnostic accuracy when the measurement or physician’s interpretation does not agree with a recognized “gold standard,” such as arteriography. Interpretative errors can occur despite the recording of a precise and reproducible measurement. For example, the use of velocity criteria to grade internal carotid artery stenosis not previously validated by comparison with angiographic results can cause consistent overclassification of disease severity.

The accuracy of diagnostic testing is commonly expressed in terms of descriptive statistical parameters such as sensitivity, specificity, and positive and negative predictive values (PPV and NPV). These parameters are useful to compare the diagnostic accuracy between different testing methods or threshold criteria, but they are subject to statistical pitfalls introduced by the reliability of the gold standard used for correlation as well as disease prevalence and clinical status (symptomatic, asymptomatic) of the study population. For example, the sensitivity (ability to detect the presence of a disease state) and specificity (ability to recognize the absence of a disease state) of a specific test are not affected the prevalence of the particular disease state in the study population used to calculate diagnostic accuracy. But predictive values (PPV, NPV), which are better measures of clinical usefulness in the management of an individual patient, are highly dependent on disease prevalence. It is recommended that testing with the highest sensitivity be used to screen patients to rule out a particular disease state (1). For an individual patient, a test with a high (>90%) NPV can be used to exclude the disease state. Similarly, testing with high specificity and PPV should be used to confirm the presence of disease or to proceed with intervention—i.e., performing carotid endarterectomy based on duplex ultrasonography. It is impossible to eliminate all sources of measurement error associated with noninvasive vascular testing because of inherent variability in instrumentation, anatomy, and the examination technique. Despite this caveat, noninvasive vascular diagnostics, in particular duplex ultrasonography, demonstrate sufficient accuracy to permit medical, surgical, or endovascular intervention with a high degree of clinical certainty.

The noninvasive vascular laboratory typically employs a number of different instruments for testing in the areas of cerebrovascular, peripheral arterial, peripheral venous, and abdominal visceral disease. The most widely used instrumentation relies on ultrasound to interrogate vessels for patency and flow. Duplex ultrasonography is necessary instrumentation in an accredited vascular laboratory, and the clinical applications of this technique are numerous in the evaluation cerebrovascular, peripheral venous, peripheral arterial, and visceral vascular disease (Table 1). Plethysmographic instruments, such as the pulse volume recorder or photoplethysmograph, coupled with the pressure transducer (aneroid manometer), provide indirect hemodynamic measurements of peripheral arterial and venous flow and pressure. Compared with duplex ultrasound, these indirect techniques are vulnerable to their own unique diagnostic pitfalls.

Table 1 Clinical Applications of Duplex Ultrasonography

II. PROCEDURAL PITFALLS

A. Instrumentation

Instrumentation in good operating condition and calibration is required for noninvasive vascular testing. Measurements of limb blood pressure (ankle-brachial systolic pressure index, or ABI), treadmill walking time, pulsatility index, and duplex-acquired velocity spectral parameters (peak systolic velocity, end-diastolic velocity) demonstrate reproducibility comparable to that of other common clinical (pulse rate, hemoglobin, serum creatinine) measurements. At a 95% confidence level, a significant change between two measurements has been found to be greater than 14% for ABI, greater than 120 for treadmill walking time, greater than 0.4 for pulsatility index, and greater than 20% for duplex-acquired velocity spectra and volume flow values (2).

Use of pressure cuffs of improper width relative to limb girth or noncalibrated manometers can result in erroneous measurement of segmental limb systolic blood pressure. The interpretation of the high-thigh pressure measurement to evaluate aortoiliac and common femoral artery inflow occlusive disease is highly dependent on the size of the limb to relative to the width of the cuff used. Theoretically, cuff width should be 20% greater than the diameter of the limb (3). When a narrow (10- to 12-cm) thigh cuff is used, normal high-thigh pressure is at least 20–30 mmHg higher than brachial pressure because of the artifact produced by the relatively narrow cuff (e.g., normal high-thigh systolic pressure index is >1.2; while the normal value of ankle-brachial systolic pressure index is >0.95) (1).

Careful attention to instrument calibration is particularly important when plethysmographic techniques are used for pulse volume recordings (PVR air plethysmography) or to measure lower limb venous outflow (air or impedance plethysmography). Standardization of cuff inflation pressure (approximately 65 mmHg) is mandatory to obtain reliable, reproducible air plethysmographic waveforms for the detection of arterial occlusive disease. Fortunately, modern computer-based PVR instruments include an internal calibration system that virtually eliminates operator errors in cuff application and technique. Improper cuff or photocell application can also produce artifacts and contribute to erroneous measurements of digital pressure and venous recovery time.

When ultrasound is used to image blood vessels or interrogate blood flow patterns within them, a variety of pitfalls can result in erroneous data or a study that cannot be interpreted. These problems can be minimized if the operator has a thorough understanding of ultrasound physics and the design features of the ultrasound system. A number of factors relating to the scan-head design, Doppler system, frequency analyzer, and display devices can affect imaging resolution and velocity spectral data (Table 2). Inappropriate selection of the transducer’s ultrasound frequency is a common pitfall of duplex ultrasonography. High (7- to 15-MHz) B-mode imaging frequencies allow superior lateral and depth resolution but are strongly attenuated by tissue, thereby limiting imaging to only superficial (1- to 5-cm) vessels. Selection of an optimum transducer frequency should be based on the depth of the vessel examined and the composition of overlying tissue. Table 3 lists the transducer ultrasound frequency to obtain the strongest Doppler signal from vessels imaged through different tissues; these frequencies are based on equations that account for ultrasound scatter, tissue attenuation, and vessel depth (4). For example, duplex testing of the carotid artery in an obese patient that lies under fat and muscle at a depth of 7 cm may require a transducer frequency of 3 MHz to record a strong Doppler signal.

Variation in beam pattern and focusing can result in lateral image and refractive distortion. Because ultrasound is a wave, the shape of the beam varies at distance from the transducer. By increasing the transducer bandwidth, unwanted variations in the beam pattern are smoothed out. Adjusting the focus (focal point) to just below the area of interest is important to minimize lateral image distortion. In general, steering the ultrasound beam perpendicular to vessel walls provides superior depth resolution. Refractive distortion results when the ultrasound travels through and crosses a boundary from one tissue to another with a different ultrasound propagation speed. This can result in errors in dimensions and the number of objects in the lateral direction of any ultrasound image. Duplication of the subclavian artery is an example of a refractive distortion caused by reflection of ultrasound from the pleura. Duplications appear in both B-mode and color images, and spectral waveforms can be obtained from both images. The only defense against misdiagnosis is a knowledge of anatomy and anatomical anomalies and the concept of refractive distortion.

Table 2 Factors Affecting Ultrasonic Imaging and Doppler Data of Duplex Ultrasonography

Table 3 Ultrasound Frequency for the Strongest Doppler Signal Relative to Vessel Depth and Type of Overlying Tissue

Ultrasound beam steering can also introduce error measurement of blood-flow velocity. An experimental study using a velocity-calibrated string phantom demonstrated significant overestimation of recorded velocity when a multielement scan head was used in steered versus unsteered modes (4). Differences in the range of 20–50% were recorded when end elements of the scan head were used to record the Doppler signal. Errors of this magnitude are worrisome, since measurements of peak systolic and end-diastolic velocity are used clinically to recommend intervention for internal carotid artery or vein bypass graft stenosis. The reasons for velocity overestimation are complex and related to characteristics of the ultrasound system, including transducer beam width, aperture size, transmitting frequency, and angle of Doppler beam insonation. Manufacturers should be encouraged to include velocity calibration in routine instrument maintenance using test phantoms. In clinical situations where peak velocity measurements approach thresholds levels, a linear-array scan head should be used in the steered configuration with the Doppler cursor positioned at the end of the transducer array, so that pulse Doppler signal recording will be at the lowest angle of incidence to the axis of flow. Another strategy is to repeat the study using a phased-array transducer and compare the values to those obtained with the linear-array transducer.

Duplex ultrasound systems continue to undergo rapid evolution in terms of image resolution, color Doppler imaging, cost, and size. Some instruments provide velocity measurements from peripheral arteries without any assumption about the Doppler examination angle (5). When color-flow imaging was introduced, it was hoped that diagnostic accuracy would improve. A comparison between color Doppler velocity and spectral waveform velocity demonstrates that values can differ due to angle adjustments or differences in signal processing. The definition of velocity is obscure, since duplex instrumentation records blood cell movement from a volume or voxel and the velocity components displayed in the spectral display represent amplitude of multiple velocity vectors. A reproducible value can be obtained only when a consistent examination angle is used. This is important clinically for the diagnosis of disease progression. Serial duplex examinations should be performed with the same instrument and Doppler signals recorded at the same Doppler angle from an identical image as that acquired in the prior study. Because of the confusion produced by real-time color imaging and color map aliasing, some instruments provide a feature called “color power angiography,” which shows pixels of blood motion in color without showing direction. The resultant image permits the selection of pulsed Doppler recording sites at and downstream from the location of maximum luminal stenosis. This minimizes the likelihood that regions of slow blood flow or vessel segments insonated at high (80 -to 90-degree) Doppler beam angles will be coded as showing no flow. When regions of no flow are identified, it is essential to scan from at least two lines of sight to improve the angle at which scan lines intersect with blood flow and thereby increase the likelihood of color-coding blood flow if present. As a rule, if the tissue (i.e., blood) velocity is less than 1 cm/s, the velocity is considered to be zero and color is not shown. If a wall filter is activated, velocities below about 10 cm/s are not shown.

In “real-time” color Doppler ultrasound, the image is not formed instantly but processed from the Doppler data from left to right over 30–50 ms. This produces a time distortion in all color-flow images. The speed of acquisition is approximately 150 cm/s which is comparable to speeds in the vascular system: wall motion <1 cm/s, average arterial blood velocity=30 cm/s, pulse propagation speed=1000 cm/s. Careful inspection of a single color-flow ima...

Table of contents

• COVER PAGE
• TITLE PAGE
• COPYRIGHT PAGE
• PREFACE TO THE SECOND EDITION
• PREFACE TO THE FIRST EDITION
• Contributors
• 1: PITFALLS OF NONINVASIVE VASCULAR TESTING
• 2: CARDIOPULMONARY COMPLICATIONS RELATED TO VASCULAR SURGERY
• 3: RENAL FAILURE AND FLUID SHIFTS FOLLOWING VASCULAR SURGERY
• 4: INTIMAL HYPERPLASIA: THE MECHANISMS AND TREATMENT OF THE RESPONSE TO ARTERIAL INJURY
• 5: THE HEALING CHARACTERISTICS, DURABILITY, AND LONG-TERM COMPLICATIONS OF VASCULAR PROSTHESES
• 6: ANASTOMOTIC ANEURYSMS
• 7: HYPERCOAGULABLE STATES AND UNEXPLAINED VASCULAR GRAFT THROMBOSIS
• 8: COMPLICATIONS AND FAILURES OF ANTICOAGULANT AND ANTITHROMBOTIC THERAPY
• 9: GASTROINTESTINAL AND VISCERAL ISCHEMIC COMPLICATIONS OF AORTIC RECONSTRUCTION
• 10: SPINAL CORD ISCHEMIA
• 11: IMPOTENCE FOLLOWING AORTIC SURGERY
• 12: COMPLICATIONS FOLLOWING RECONSTRUCTIONS OF THE PARARENAL AORTA AND ITS BRANCHES
• 13: COMPLICATIONS OF MODERN RENAL REVASCULARIZATION
• 14: THE DIAGNOSIS AND MANAGEMENT OF AORTIC BIFURCATION GRAFT LIMB OCCLUSIONS
• 15: PROBLEMS RELATED TO EXTRA-ANATOMIC BYPASS—INCLUDING AXILLOFEMORAL, FEMOROFEMORAL, OBTURATOR, AND THORACOFEMORAL BYPASSES
• 16: VASCULAR GRAFT INFECTIONS: EPIDEMIOLOGY, MICROBIOLOGY, PATHOGENESIS, AND PREVENTION
• 17: AORTIC GRAFT INFECTIONS
• 18: DETECTION AND MANAGEMENT OF FAILING AUTOGENOUS GRAFTS
• 19: AN APPROACH TO TREATMENT OF INF RAINGUINAL GRAFT OCCLUSIONS
• 20: WOUND COMPLICATIONS FOLLOWING VASCULAR RECONSTRUCTIVE SURGERY
• 21: COMPLICATIONS IN THE MANAGEMENT OF THE DIABETIC FOOT
• 22: COMPLICATIONS OF LOWER EXTREMITY AMPUTATION
• 23: COMPLICATIONS OF VASCULAR ACCESS
• 24: COMPLICATIONS OF THORACIC OUTLET SURGERY
• 25: STROKE AS A COMPLICATION OF NONCEREBROVASCULAR SURGERY
• 26: COMPLICATIONS OF REPAIR OF THE SUPRA-AORTIC TRUNKS AND THE VERTEBRAL ARTERIES
• 27: PREVENTION OF TRANSIENT ISCHEMIC ATTACKS AND ACUTE STROKES AFTER CAROTID ENDARTERECTOMY: A CRITIQUE OF TECHNIQUES FOR CEREBROVASCULAR PROTECTION DURING CAROTID ENDARTERECTOMY
• 28: NONSTROKE COMPLICATIONS OF CAROTID ENDARTERECTOMY
• 29: RADIATION EXPOSURE AND CONTRAST TOXICITY
• 30: COMPLICATIONS IN PERIPHERAL THROMBOLYSIS
• 31: COMPLICATIONS OF SCLEROTHERAPY
• 32: COMPLICATIONS OF SUBFASCIAL ENDOSCOPIC PERFORATOR VEIN SURGERY AND MINIMALLY INVASIVE VEIN HARVESTS
• 33: COMPLICATIONS OF VENOUS ENDOVASCULAR LYSIS AND STENTING (ILIAC, SUBCLAVIAN)
• 34: COMPLICATIONS OF ENDOVASCULAR INTERVENTION FOR AV ACCESS GRAFTS
• 35: COMPLICATIONS OF VENA CAVA FILTERS
• 36: COMPLICATIONS OF PERCUTANEOUS TREATMENT OF ARTERIOVENOUS MALFORMATIONS
• 37: ENDOVASCULAR COMPLICATIONS OF ANGIOPLASTY AND STENTING
• 38: COMPLICATIONS OF CAROTID STENTING
• 39: ENDOVASCULAR ACCESS COMPLICATIONS
• 40: DEVICE FAILURE
• 41: ENDOLEAK
• 42: COMPLICATIONS FOLLOWING ENDOVASCULAR THORACIC AORTIC ANEURYSM REPAIR
• 43: COMPLICATIONS OF ANGIOGENESISTHERAPY
• ABOUT THE EDITORS