Transesophageal echocardiographic guidance of transcatheter closure of atrial septal defect.

Transcatheter closure of atrial septal defect (ASD) was accomplished in 10 of 11 patients aged 13 months to 46 years (weight range 11 to 77 kg). Transesophageal echocardiography (TEE) was used simultaneously with fluoroscopic imaging in 4 of these patients aged 4.5 to 46 years (weight range 19 to 77 kg). TEE was used to ascertain defect size, position and number of defects and to ascertain appropriate seating of the defect occluder within the atrial defect. In 2 patients TEE-assisted transcatheter ASD closure was accomplished after previous attempts at transcatheter ASD closure, unaided by TEE, had been unsuccessful. The only unsuccessful ASD closure procedure occurred in the smallest patient in the series (an 11-kg 13-month-old), a child who was too small to undergo TEE using our 11-mm diameter endoscopic probe. The concomitant use of TEE with fluoroscopic imaging provides information that is unique and complementary and may improve the efficacy and safety of the transcatheter technique for ASD closure. The recent availability of a 7-mm diameter TEE probe will extend the use of TEE into the infant age group and may decrease the discomfort and potential morbidity of TEE in older patients.

Echocardiography of the right side of the heart.

Echocardiography has not only revealed the morphologic aspects of the right-sided cardiac structure but has, through the use of pulsed and color Doppler imaging, provided a whole new level of certainty in the diagnosis of cardiac disease. The structure and function of the right-sided chambers are now far better appreciated since this technique has become more widely used for early clinical diagnosis of right-sided cardiac disease.

HYDRA: a knowledge acquisition tool for expert systems that critique medical workup.

HYDRA is a computer-based knowledge acquisition tool under development to assist in the creation of expert systems which critique medical workup. To use HYDRA, a domain expert first outlines the recommended approaches to the workup of a chosen medical problem, using the Augmented Transition Network formalism. From this model, HYDRA produces a list of the various conditions for which critiquing comments may be required to react to all possible approaches that might be proposed by the user of the critiquing system. Domain-specific constraints can be used to restrict the number of conditions suggested. In this way, HYDRA assists the domain expert by providing a model for structuring the problem, and by breaking down the domain expert's work into a set of small, easily understood tasks.

Investigation of Doppler waveforms in porcine renal allografts: Doppler-pathologic correlation.

A porcine model was devised to investigate Doppler waveforms in dysfunctional renal allografts. NIH miniature pigs served as allografts donors and recipients. Renal transplantation was effected into the recipient pelvis while the left normotopic kidney was subjected to warm ischemia in order to induce acute tubular necrosis (ATN). Doppler demonstration of allograft arterial occlusion in six animals was confirmed at surgery. Increased pulsatility of intrarenal arterial signals constituted evidence of vascular rejection in two animals. Biopsies confirmed the diagnosis in both cases. Histologic changes of ATN were identified in two native kidneys subjected to ischemia. No waveform or pulsatility alterations were observed in these animals. The porcine model provides for the investigation of allograft Doppler waveforms in a controlled setting with free access to biopsy and operative pathologic correlation.

Exploring subjective vs. objective issues in the validation of computer-based critiquing advice.

Evaluation is an important part of the development of computer-based medical expert systems. Such evaluation may be particularly difficult when judging a critiquing system which responds to a proposed management strategy with a discussion of the advisability of that approach. DxCON is an expert system which produces a prose critique discussing the radiologic workup of obstructive jaundice. This paper describes DxCON, and its experimental validation by three independent judges. A central component of the validation involved allowing the judges to react to the system's advice in a quite flexible, unstructured fashion. This project provides a case study of how subjective issues impact both the design and implementation of a validation of a medical expert system whose output is explanatory prose.

A computer utility for automated retrieval of radiology reports.

OBJECTIVE: Many radiology information systems (RISs) installed during the 1980s have a user interface that is crude by today's standards. We explored improving this user interface by using the local processing power of the personal computer. In this article we describe a computer program that significantly enhances the ease of report retrieval by automating much of the interaction with our RIS, which is IDXrad. MATERIALS AND METHODS: The program, named DERVISH, runs under DOS on International Business Machines-compatible personal computers having an 80286 or faster processor, an enhanced graphics adapter or video gate array display, two serial ports, a mouse, and a bar code wand. It is written in the C++ programming language. Input may be keyboard, mouse, or bar code wand, and reports may be displayed singly or multiply and printed to a local printer or an RIS printer or copied to a disk file. DERVISH not only performs many of the routine tasks of an RIS session such a logging on and off but also provides a menu-driven environment for report display, including optional restriction of the report list to related prior reports. Keyboard use is minimized in favor of a mouse and a bar code wand. DERVISH emulates a standard RIS terminal for functions other than report retrieval. RESULTS: Originally intended to help retrieve the prior radiology reports of a patient at hand, DERVISH has found use in our quality management effort as well, simplifying comparison of reports from related radiologic techniques. An early version suffered poor acceptance because of its nonstandard interface and frequent malfunctions. The current version is presently in use in our emergency radiology department. CONCLUSION: DERVISH serves not only as a useful utility in itself but also as a demonstration of the ability of the local processing power of a desktop computer to improve the user interface of an aging RIS and to extend its functional life.

Duplex Doppler: identification of cavernous transformation of the portal vein.

Three cases of cavernous transformation of the portal vein are presented, which emphasize the value of duplex Doppler sonography in the recognition of abnormal vascular structures. In all three cases, cavernous transformation was unsuspected; in two, the initial sonographic or CT examinations were interpreted incorrectly. These cases suggest that the combination of characteristic pulsed-Doppler waveforms and the real-time appearance of cavernous transformation is virtually diagnostic.

The prospect of expert system-based cognitive support as a by-product of image acquisition and reporting.

In order for computer-based decision-support tools to find routine use in the everyday practice of clinical radiology, further development of user interface and knowledge content are required. In an ideal interface, the interaction between the radiologist and the computer would be minimized and painlessly integrated into existing work patterns. In this article, we explore some of the ways that pre-existing computer interactions in the processes of image acquisition and reporting can be used to feed case information into an expert system and thereby allow users to acquire advice from it in an automatic fashion. We describe interface models that we have developed in the domains of mammography and obstetric ultrasound, and discuss interface and content-related questions that have arisen from informal evaluations of these systems. In particular, the need for clinical outcome-relevant decision support and training level-appropriate decision support are discussed in detail.

Normal instability of the hip in the neonate: US standards.

In recent years, the use of real-time ultrasonography (US) has enabled dynamic evaluation of the infant hip through a range of motion and stress. Preliminary experience has suggested that a certain amount of instability in the hips of newborns is normal, but no standards have been established. In this study, a group of term neonates whose physical examinations were normal were examined with US on the 1st and 2d days of life. Each hip was imaged in the transverse plane in nonstressed and stressed positions, and movement of the femoral head under stress was quantitated. This displacement under stress was used to establish a normal range of hip instability in neonates. Patterns of hip laxity in boys and girls are identical, and in most infants hip instability diminishes between the 1st and 2d days of life. Our method of quantitating hip instability produces consistent results, with intraobserver 95% confidence intervals of +/- 1.2 mm for each measurement.

Optimizing radiologic workup: an artificial intelligence approach.

The increasing complexity of diagnostic imaging is presenting an ever expanding variety of radiologic test options to clinicians. As a result, it is becoming more difficult for referring physicians to select an appropriate sequence of tests. The current economic pressures on medicine make it particularly important that resources be used judiciously. Radiologic workup often involves a sequence of tests that lead from presenting signs and symptoms to a definitive diagnosis or intervention. This sequence ideally begins with simple, inexpensive, safe, non-invasive tests and progresses to more complex, expensive, and hazardous tests only if the simpler tests are insufficient to establish a diagnosis. DxCON is a developmental artificial intelligence-based computer system that gives advice to physicians about the optimum sequencing of radiologic tests. DxCON evaluates basic clinical information and a physician's proposed workup plan. The system then creates an analysis of the strengths and weaknesses of his plan. The domain chosen to explore computer-based workup advice is the radiologic workup of obstructive jaundice.

Liver transplant recipients: portable duplex US with correlative angiography.

Twenty patients, aged 4 months to 58 years, were evaluated for liver transplantation by duplex sonography, and 15 transplantations were completed; 42 postoperative examinations were performed. Sonographic findings were correlated with seven preoperative and five postoperative angiographic evaluations. Preoperative duplex US findings included tumors, portal vein occlusion, varices, biliary obstruction, and variant vascular anatomy. Postoperative findings included hepatic artery occlusion, portal vein occlusions (one with cavernous transformation), portal vein stenosis, biliary obstruction, intrahepatic and extrahepatic fluid collections, and air in the portal vein due to ischemic bowel. Use of angiography allowed confirmation of the vascular abnormalities and demonstrated evidence of rejection in patients with normal Doppler waveforms. Duplex sonography is a valuable portable technique for evaluating these patients and can be used in triage of patients requiring angiography.

Doppler signal quantitation in renal allografts: comparison in normal and rejecting transplants, with pathologic correlation.

Quantitative duplex Doppler sonography was performed in 55 renal transplant patients during 54 independent episodes of acute rejection, three episodes of chronic rejection, three episodes of acute tubular necrosis (ATN), and 23 occasions of normal graft function. Doppler signals were obtained from four arterial sites in each kidney. Nine patients, in whom signals were absent, were subsequently shown at nephrectomy to have absence of perfusion resulting from severe acute vascular rejection. In each patient with graft dysfunction, biopsy or nephrectomy was performed within 24 hours of the Doppler study. Arterial Doppler signals were quantified using a pulsatility index (PI). Acute rejection produced a significantly higher PI at each arterial site. Receiver-operator characteristics suggest that signals obtained from the segmental arteries are most sensitive to these changes. With a threshold PI of 1.5, the sensitivity of this technique for detection of acute renal allograft rejection is 75%; the specificity is 90%. In acute vascular rejection, the same PI yields a sensitivity of 79% and specificity of 90%. With a cutoff PI of 1.8, a specificity of 100% can be achieved.

Quantitative US attenuation in normal liver and in patients with diffuse liver disease: importance of fat.

Two methods are used to estimate ultrasound attenuation in liver. These were based on amplitude change and frequency change as a result of depth dependent attenuation. Evaluation of the two methods against a family of calibrated phantoms yielded correlation coefficients of 0.98 and 0.99, respectively. Liver attenuation in 26 control subjects was 0.50 and 0.52 dB/MHz/cm, respectively. Liver attenuation was estimated in 50 patients who later underwent liver biopsy. Comparison with quantitative histologic results showed that the presence of fat alone accounted for the increased attenuation associated with cirrhosis. Similar high attenuation values were found in patients with fatty infiltration. Fibrosis alone did not result in elevated liver attenuation. Cirrhotics without fatty infiltration had attenuation similar to that of the controls. Mechanisms of action are discussed.

Renal allografts in acute rejection: evaluation using duplex sonography.

Sixty-nine duplex sonographic studies were performed in 24 patients who had received renal allografts. After a prospective qualitative analysis of the Doppler waveforms, results were correlated with biopsy material and each patient's clinical course. Increased pulsatility of the Doppler waveform of intrarenal arterial flow constituted an abnormal study, indicating acute rejection. Overall sensitivity varied with the histologic form of rejection, with a 60% sensitivity for acute interstitial rejection with or without vascular rejection and an 82% sensitivity for acute vascular rejection. Overall specificity was 95% and 96%, respectively. Early rejection was also accurately detected in three patients less than 48 hours following kidney transplantation. Duplex sonography has a useful role in evaluating posttransplantation renal failure. Abnormal study results may obviate the need for biopsy and help in guiding clinical management.

Computed tomographic (CT) guided percutaneous fine-needle aspiration biopsy: the Yale experience.

Fifty-one CT-guided percutaneous fine-needle aspiration biopsies (PFNAB) were performed on 46 consecutive patients over 15 months. Cytologies were obtained to identify primary or secondary malignancy in the abdomen, pelvis, retroperitoneum, bone, and paraspinal region. Adequate cytologic material was obtained in 50 of 51 biopsies. There were 29 true-positive, 0 false-positive, 12 true-negative, and two false-negative cases with an overall accuracy rate of 95 percent. There was one minor complication, mesenteric hemorrhage, which did not require transfusion. Fifteen of the 51 biopsies were performed on outpatients. The procedure is an accurate, safe, and cost-effective nonsurgical means of diagnosing primary or secondary malignancy.

Observer variability in echo-Doppler measurements of portal flow in cirrhotic patients and normal volunteers.

The intraobserver and interobserver variability in measuring the portal vein flow by the echo-Doppler technique was evaluated in a blind controlled study. A total of 22 cirrhotic patients and 14 normal volunteers were examined by two skilled operators using duplex Doppler within a period of 1-3 mo (6 cirrhotics and 7 normal volunteers by both observers). Area, mean velocity, and flow were measured (4 measurements: A, B on day 1; C, D on day 2). The intraclass correlation coefficient was used to assess both the statistical and clinical significance of intraobserver and interobserver agreement for the measurements of these three parameters. The level of intraobserver agreement for each parameter on normal subjects and cirrhotics was obtained from the two measurements on the same day and from the two measurements at the same time on consecutive days. Overall agreement between the four measurements was also calculated. Levels of interobserver agreement were obtained by calculating separately the intraclass correlation coefficient from each of the four pairs by measurements made on the same subject by the two observers over the same period of 2 days. The coefficient of variation was also used to compare the variability in these measurements. Overall, intraobserver agreement on normal subjects varied from good to excellent for observer 1, and from fair to good for observer 2. On cirrhotic patients, observer 1 was excellent at all times for all parameters. Observer 2 had lower intraclass correlation coefficient values, especially for velocity on consecutive days. For the best of the two observers on the portal flow, the coefficient of variation in cirrhotic patients ranged from 2%-30% with a mean +/- SEM of 12% +/- 4%. No acceptable interobserver agreement was found between the two observers in either of the two samples of subjects. These results support the use of this technique mainly for the determination of rapid and large changes in portal hemodynamics within a short period of time. The technique seems to have low precision in monitoring chronic changes in portal hemodynamics.

Unusual causes of increased vascular impedance in renal transplants: duplex Doppler evaluation.

Duplex Doppler ultrasound (US) examination of the renal vasculature has proved valuable in assessing the kidney transplant. The normal renal allograft exhibits low-impedance arterial inflow similar to that seen in the normotopic kidney. The authors and others previously reported that a high vascular impedance, defined as either a pulsatility index (PI) greater than 1.8 or a resistive index greater than 0.9, indicates acute vascular rejection (AVR). Although AVR remains the most common cause of increased PI, the authors noted ten episodes among 180 serially followed-up transplants in which abnormal waveforms were clearly not due to rejection. Four other causes of increased vascular impedance are reported, including renal vein obstruction, severe acute tubular necrosis, pyelonephritis, and extrarenal compression of the graft. These new causes only slightly decrease the specificity of high vascular impedance for rejection. Furthermore, the cause can usually be recognized from the clinical history or other US findings.