Correspondence.

(Letters to the Editor are welcomed. They may report new clinical or laboratory observations and new developments in medical care or may contain comments on recent contents of the Journal. They will be published, if found suitable, as space permits. Like other material submitted for publication, letters must be typewritten, double-spaced, and submitted in duplicate. They must not exceed two typewritten pages in length. No more than five references and one figure or table may be used. See "Information for Authors" for format of references, tables, and figures. Editing, possible abridgment, and acceptance remain the prerogative of the Editors.)

Actinomycosis Presenting as an Oral Ulcer in a Neutropenic Patient

To the Editor: We report a case of actinomycosis presenting as an oral ulcer in a neutropenic patient. We believe it is the first reported case of actinomycosis presenting in this manner. The patient was a 77-year-old African American woman who presented to the University of Illinois Medical Center with the complaints of throat pain and inability to eat for 1 to 2 weeks. The patient had been diagnosed with myelodysplastic syndrome (refractory anemia with excess blasts) approximately 5 years before, and at the time of presentation was pancytopenic and neutropenic. On examination, she was found to have an ulcer measuring 2 x 2 cm with a whitish base and erythematous margins in the back of her soft palate. The rest of the physical examination was normal, with no cervical lymphadenopathy or sinus tenderness. At that time, a presumptive diagnosis of herpetic infection was made. Cultures of the ulcer site failed to grow any viruses, bacteria, or fungi. After 3 weeks of treatment with oral and intravenous acyclovir and oral fluconazole, she presented with increased pain and increase in the size of the ulcer. A biopsy of the lesion was performed, which revealed squamous mucosa with granulation tissue, acute and chronic inflammation, and numerous organisms morphologically suggestive of Actinomyces species. The Brown-Brenn stain disclosed granules composed of gram-positive filamentous organisms characteristic of Actinomyces. Computed tomography of the sinuses and head did not reveal any sinus or central nervous system invasion. Since the patient was allergic to penicillin, she was initially treated with intravenous clindamycin, and later given oral doxycycline when she was discharged to home. This resulted in the ulcer almost completely resolving after 3 weeks. A few weeks, later she was readmitted with neutropenic fever and sepsis and died.

Actinomycosis presenting acutely as a superficial oral ulcer is extremely rare, especially when there is no history of tissue injury. The usual differential diagnosis of an oral ulcer in immunocompromised patients includes herpetic and candidal infections, secondary syphilis, gonorrhea, primary human immunodeficiency virus infection, and aphthous stomatitis. Actinomycosis is generally a chronic disease, and usually presents in the form of an abscess, soft tissue swelling, or mass lesion in the oral or cervicofacial region. The most common site of occurence is the perimandibular region, affecting the angle of the jaw, cheek, submental region, retromandibular region, or tempromandibular region. (1) Actinomycosis is frequently misdiagnosed; only 7% of patients with this disease are diagnosed correctly at clinical presentation. The diagnosis of actinomycosis can be made by culturing the organism, keeping in mind that Actinomyces organisms are rather delicate and difficult to culture, especially if they are expose d to air. Thus, we propose that this infection should be kept in mind in the treatment of ulcerations in immunocompromised patients, and that a biopsy should be attempted when the ulcers do not heal after treatment with antiviral and antifungal therapy. The therapy for actinomycosis consists of high and prolonged doses of antibiotics. Penicillin is the drug of choice but, if contraindicated, clindamycin and tetracycline are useful alternatives.

Amit Verma, MD

Wendy Stock, MD

Section of Hematology-Oncology

Marcy Lait, MD

Department of Medicine

Department of Otolaryngology

Karen Ferrer, MD

Department of Pathology

John Quinn, MD

Section of Infectious Diseases

Leonidas C. Platanias, MD, PhD

Department of Medicine

Section of Hematology-Oncology

University of Illinois at Chicago Medical Center

Department of Medicine

840 S Wood St

Chicago IL 60612

Reference

(1.) Neville BW, Damm DD, Allen CM, et al: Oral and Maxillofacial Pathology. Philadelphia, WB Saunders Co, 1995, pp 156-157

Residents: Are You Asleep at the Wheel?

To the Editor: I had the privilege of attending the Resident Physician Leadership Symposium (RPLS) from April 9 through April 11, 2002. This conference was created by the C. Everett Koop Institute and is held yearly at Dartmouth College. Nineteen residents from a variety of specialties and an assortment of programs around the country attended, selected from an applicant pool of approximately 80 people. We spent 3 days listening to talks, participating in workshops, and designing projects. Topics included health policy, population health, and computers in medicine, to name a few. Two talks in particular gave me pause. These topics were the Accreditation Council for Graduate Medical Education (ACGME) and resident work-hours legislation. These 2 topics stirred up the residents at the conference; we found ourselves both concerned and excited at the same time. I hope that hearing about them will do the same for you, my fellow residents.

Dr. Betty Chang, the resident representative to the ACOME, gave a talk entitled "What Are the ACGME Core Competencies and Why Should You Care?" The ACGME is a familiar string of letters that likely does not invoke much reaction from residents. This organization is responsible for the decision of whether a residency program is accredited. The ACGME is made up largely of physicians and represents the interests of various organizations, such as the American Medical Association, American Board of Medical Specialties, American Hospital Association, American Association of Medical Schools, and Council of Medical Specialty Societies. In my view, up to this point, the ACGME had consisted of the organization that sent the guy in a suit to my program this year. He talked to us about our experiences and took copious notes. We were forewarned that this particular guy could do damage to our program if we said too much. Truthfully, I just have not paid much attention to that 5-letter acronym before this conference. Judging by the comments of those around me at the RPLS, neither had the other residents.

It is time for all of us to pay attention to the ACGME, however, and this is why. The ACGME has recently defined 6 core competencies to be covered during every resident's training. These core competencies are as follows:

Patient care includes making informed decisions for patient care, performing procedures, counseling patients, and using information and technology in patient care.

Medical knowledge includes knowledge about current and evolving practices and the know-how to use this information.

Professionalism includes ethics, personal responsibility, and issues of cultural sensitivity.

Systems-based practice relates to how well you function in the system in which you learn and how well you understand the system in the context of the larger health care system.

Practice-based improvement includes how well you keep up with current literature and technology, as well as integrating these new techniques and technologies into your practice.

Interpersonal communication skills includes how well you communicate with patients, families, coworkers, and other health care personnel.

These core competencies will be integrated into your program beginning June 2002. Different specialties will embrace these to a greater or lesser degree, and some will gradually phase in the competencies over time. What does this mean for you? It means potentially more required training It is no longer enough to learn disease processes, chief complaints, and basic therapeutics. To be prepared for the future of medicine, we need to be well-versed in all of these 6 competencies.

The second talk addressed the topic of current resident work-hour legislation. Is there any resident who has not at some time thought about work hours? This issue is easily recognized by most residents and strikes close to home. A fourth-year medical student gave this address. This particular student, Jaya Agrawal, is president of the American Medical Student Association (AMSA) and has been instrumental in getting a bill introduced into Congress that would limit resident work hours.

Agrawal did not focus on this during her talk, but I learned later that the Committee of Interns and Residents (CIR) also supports this bill. I had never heard of the CIR until I researched this bill. Their Web site states that they are the "largest housestaff union in the country," and that their membership includes 10,000 residents from California, Florida, Massachusetts, New Jersey, Washington, DC, and New York. No one at the RPLS conference had heard of this organization, and neither had anyone in my program in Wisconsin.

Agrawal quoted various (1-4) that illustrated residents' decreased cognitive functioning after sleep deprivation, increased depression with sleep deprivation, and increased risk of car accidents post-call. She spent most of her talk discussing the bill that is pending in Congress, however. The bill, HR 3236, would limit the resident work week to 80 hours. It would limit the length of any one shift to 24 consecutive hours, limit the shift in any emergency department to 12 hours, provide residents with a minimum of 10 hours off between shifts, at least 1 of 7 days off, 1 full weekend off per month, and on-call nights no more than every third night. This means less work hours.

Think about this subtle irony for a moment. The ACGME wishes to put more education and more training into your program. Agrawal, AMSA, CIR, and many other medical student activists would like to provide you with less work hours. What is the common theme in these 2 conflicting efforts?

The theme is that both these efforts are proceeding with little resident awareness, involvement, or activism. The premise is that both of these efforts represent federal or outside intervention in your career as a resident. Both efforts come with the best of intentions: to create professional and caring physicians who do not wreck their cars post-call. However, are federal or outside organizations' one-size-fits-all rules the answer? Whichever way you answered, ask yourself if anyone else knows that you believe that way.

It seems that we as residents have literally and figuratively been asleep at the wheel. Remember being a medical student in AMSA, or a premedical volunteer at the local hospital, or a student representative in your high school? You had a voice then, you had idealism out the wazoo. As residents, we seem to have lost this voice, and this idealism. Have the 36-hour shifts beaten it out of us? To a degree, they have. However, our pattern of learned helplessness is not the answer. It is time to take action. It is time to be aware of, get involved in, and lobby for or against these issues. Talk to your program director, your chairperson, your chief resident, your fellow residents. Bring up these tough issues about what is lacking in your education and your program. Look up the CIR and see if it has anything to offer you or your program. Figure out the best way to add these core competencies to your program. Get a resident group together at your hospital; try to include representatives from each specialty. Meet ever y other month. Find out your shared concerns. Work on ways to get the noneducational aspects of your job minimized so that you have more time for professionalism or systems-based learning, and so that you can get out of the hospital post-call. Work on ways to make changes locally; otherwise these changes will only come from outside organizations with minimal resident opinion or voice.

For more information on the ACGME Core Competencies, visit the Web site at http://www.acgme.org. Consider contacting Betty Chang, your only resident representative in ACGME, with your questions or opinions. For information on the pending work-hours legislation, visit the AMSA Web site at http://www.amsa.org, or contact Jaya Agrawal. For information on CIR, visit their Web site at http://www.cirdocs.org. To learn about attending a future RPLS, check out the Web site at http://www.dartmouth.edu/dms/koop/rpls/index. html.

Megan A. Moreno, MD

Pediatric Residency Program

University of Wisconsin-Madison 

3009 Stevens St

Madison, WI 53705

References

(1.) Lamberg L: Long hours, little sleep: bad medicine for physicians-in-training? J AMA 2002; 287:303-306

(2.) Steele MT, Ma OJ, Watson WA, et al: The occupational risk of motor vehicle collisions for emergency medicine residents. Acad Emerg Med 1999; 6:1050-1053

(3.) Hart RP, Buchsbaum DG, Wade JB, et al: Effect of sleep deprivation on first-year residents response times, memory, and mood. J Med Educ 1987; 62:940-942

(4.) Ford CV, Wentz DK: Internship: what is stressful? South Med J 1986; 79:595-599

Right Atrial Mobile Thrombus and Pulmonary Thromboembolism Complicating Hemorrhagic Campylobacter jejuni Colitis

To the Editor: Right-sided heart thrombi, located mostly in the right atrium, are rare clinical entities that can lead to pulmonary thromboembolism (PTE), with mortality rates in excess of 40%. (1,2) Campylobacter jejuni, a common cause of diarrheal illness, also has a predilection for vascular endothelium, causing thrombophlebitis, deep vein thrombosis (DVT), bacteremia, and endocarditis. (3,4) We report a case of DVT, right atrial thrombus (RAT), and massive PTE complicating hemorrhagic C jejuni colitis in a nonimmunocompromised host.

A 65-year-old man with hypertension and hypercholesterolemia developed an acute diarrheal illness while on vacation in the United Kingdom, and promptly flew back to the United States. The night before admission he had become short of breath, with left arm discomfort but no chest pain. At the time of admission, he was dehydrated and afebrile; his pulse rate was 90 beats/minute, pulse oximeter reading 96% on room air, and blood pressure level 95/68 mm Hg. Results of physical examination were unremarkable. Chest radiograph and electrocardiogram (EKG) were normal. Results of initial laboratory evaluation revealed normal complete blood count and electrolyte levels; hepatic function tests were also normal. Blood urea nitrogen level was 36 mg/dL, serum creatinine level was 2.3 mg/dL, and urine protein level was 30 mg/dL. He was started on intravenous normal saline, aspirin (325 mg daily) for cardiac ischemia, attapulgite (Kaopectate) suspension as needed, and oral ciprofloxacin. Serial CK-MB estimations in the first 24 hours were unremarkable, and there were no EKG changes. The initial troponin I level was elevated at 2.8 ng/mL (normal, 0-2.0), but it did not increase further.

The patient felt better by day 2 and was allowed out of bed. Later the same evening, he became increasingly short of breath and fainted. He was pale, diaphoretic, cold, clammy and short of breath. Blood pressure level was 80/60 mm Hg and results of arterial blood gas tests showed pH of 7.42, pC[O.sub.2] of 23.9 mm Hg, p[O.sub.2] of 70.7 mm Hg, with bicarbonate level of 15 mEq/L on room air. Electrocardiogram demonstrated a new right bundle-branch block, premature ventricular contractions, and new ST-T wave changes. He was resuscitated with intravenous normal saline and increased supplemental oxygen. His blood pressure quickly improved and he was transferred to the cardiac care unit. A bedside EKG showed concentric left ventricular hypertrophy, a mobile, pedunculated right-atrium (RA) mas attached to the interatrial septum, which moved in and out of the tricuspid valve (Fig. 1). The RA and right ventricle were enlarged, left ventricular ejection fraction was 71%, and the estimated pulmonary artery pressure was 65 mm Hg. Fibrinogen split products were increased. Treatment with intravenous heparin was started.

Examination of a stool sample showed numerous leukocytes, and by day 2, the specimen was growing Campylobacter species. Due to reports of quinolone resistance by Campylobacter species, ciprofloxacin was replaced with erythromycin in the patient's regimen. Tests performed on the stool sample were negative for Clostridium difficile toxin. Blood cultures grew no organisms. A Doppler ultrasound examination of the lower extremities identified a thrombus in the proximal right peroneal vein and valvular incompetence of the left popliteal vein. The patient developed bright red rectal bleeding and his hematocrit level dropped from 42.7% to 26%. Platelet count was 145,000/[mm.sup.3], but prothrombin time was 31 seconds. Heparin and aspirin were promptly discontinued, intravenous famotidine therapy was started, and blood transfusion was performed. Enhanced chest computed tomography (CT) scan showed multiple filling defects within the main pulmonary artery and within the right and left pulmonary arteries and their branch es, consistent with massive pulmonary embolus (Fig. 2). Simultaneous enhanced abdominal CT showed thickening and edema of the colon, extending from the cecum to the hepatic flexure, compatible with colitis (Fig. 3). The stool pathogen was confirmed as C. jejuni. Results of four double sets of blood cultures remained negative after 1 week. A venogram on day 3 showed no thrombi in the iliac veins or the inferior vena cava (IVC). In view of the continued need for anticoagulation, heparin was restarted on day 3, with a tight target PTT of 50 to 60 seconds. Colonoscopy revealed a large amount of residual blood in the colon, and cecal ulcerations compatible with infectious colitis. Biopsy was not done because of risk of further bleeding. A transesophageal echocardiogram (TEE) the next day confirmed a large RA thrombus protruding from the IVC, and the eustachian valve crossing the IVC and filling about half of the RA. The EKG had reverted to a normal sinus rhythm with normal intraventricular conduction. Hypercoagula bility work-up revealed normal levels of antithrombin III and anticardiolipin antibodies, immunoglobulin M (IgM), and immunoglobulin G (IgG). Test results for factor V Leiden R506Q mutation and rapid plasma reagin were negative. He improved without any further complications. Coumadin was started after 5 days of heparin, and the patient was discharged 2 weeks after admission. A repeat TEE four months later revealed normal cardiac chambers and a prominent eustachian valve, but no thrombus.

Campylobacter species, a heterogenous group of microaerophilic, curved, gram-negative rods, are known to have a predilection for vascular endothelium, causing thrombophlebitis, DVT, bacteremia and endocarditis. (3,4) It is commonly thought that C. jejuni produces only localized, self-limiting gastrocolitis.

Despite several attempts, we failed to document C. jejuni bacteremia, probably because C. jejuni is difficult to culture. (3) In the absence of any identifiable hypercoagulable state in this patient, we think that the peroneal DVT, RAT and PE could only be causatively linked to the severe hemorrhagic C. jejuni colitis in our patient. The contribution of the trans-Atlantic flight to thrombogenesis is only conjecture. It is also unclear whether to ascribe any pathogenetic role to the eustachian valve crossing the IVC. This is the first report of DVT, RAT, and PTE complicating severe hemorrhagic C. jejuni colitis in a previously nonimmunocompromised man.

Right heart thrombus is an uncommon phenomenon, and management remains controversial. (2) The mainstay of diagnosis for right heart thrombus is transthoracic echocardiography (TTE), but TEE may allow for better visualization and further identification of extension to the left side of the heart. (1,2) Recently, cardiac ultrafast CT was shown to be superior to conventional TTE with pulsed color-Doppler echocardiography in the evaluation of RAT in patients with chronic atrial fibrillation. (5) Treatment options available for RAT include intravenous heparinization, thrombolytic therapy, surgical thrombectomy with cardiopulmonary bypass, and percutaneous mechanical thrombectomy. (1,2) Clearly, patient selection needs to be individualized according to clot size and morphology, pre-existing pulmonary embolism, patient's cardiopulmonary reserve, comorbid conditions, and available local expertise.

Macaulay Onuigbo, MD, MSc

Division of Nephrology

University of Maryland School of Medicine

22 S Greene St, Room N3W143

Baltimore, MD 21201-1595

Bertram Lacey, MD

James Ricely, DO

Department of Medicine

Greater Baltimore Medical Center 

6565 N Charles St, Suite 203

Baltimore, MD 21204

References

(1.) The European Working Group on Echocardiography: The European Cooperative Study on the clinical significance of right heart thrombi. Eur Heart J 1989; 10:1046-1059

(2.) Thompson CA, Skelton TN: Thromboembolism in the right side of the heart. South Med J 1999; 92:826-830

(3.) Carbone KM, Heinrich MC, Quinn TC: Thrombophlebitis and cellulitis due to Campylobacter fetus ssp. Fetus. Report of four cases and a review of the literature. Medicine (Baltimore) 1985; 64:244-250

(4.) Farrugia DC, Eykyn SJ, Smyth EG: Campylobacter fetus endocarditis: Two case reports and review. Clin Infect Dis 1994; 18:443-446

(5.) Kitayama H, Kiuchi K, Endo T, et al: Value of cardiac ultra-fast computed tomography for detecting right atrial thrombi in chronic atrial fibrillation. Am J Cardiol 1997; 79:1292-1295

Unilateral Hyperlucency and Lung Abscess in a 20-Year-Old Woman

To the Editor. In 1953, Swyer and James (1) described unilateral, hyperlucent lung syndrome in a 6-year-old boy. The following year, MacLeod reported a similar syndrome in 9 patients, (2) hence the name Swyer-James-MacLeod syndrome (SJMS). The hallmarks of SJMS are pulmonary hyperlucency without air trapping, the absence of an obstructing tumor or foreign body, evidence of frequent pulmonary infections, and markedly decreased perfusion and ventilation in the affected lung. (3-5)

A 20-year-old Austrian woman traveling in the United States was seen in our emergency department. She complained of a 3-week history of fever, chills, cough productive of yellow sputum without hemoptysis, pleuritic left-sided chest pain, night sweats, and a 10-pound weight loss. She denied wheezing, but became short of breath on mild to moderate exertion. She had a history of protracted chest infection at the age of 13 years; a chest x-ray had been abnormal at that time and bronchoscopy had revealed that her left lung was underdeveloped. Her father had a history of pulmonary tuberculosis that had been treated successfully 10 years earlier. The patient smoked a pack of cigarettes a day for 5 years and drank 5 to 6 alcoholic drinks on weekends. There was a history of extensive travels in Europe, but she denied any high-risk behavior that might lead to infection with human immunodeficiency virus (HIV). She denied use of recreational drugs, and had no drug allergies. Before coming to the United States, she had co mpleted a 7-day course of oral antibiotic therapy.

This tall, thin, acutely ill white female was dehydrated and tachypneic, with a temperature of 103.7[degrees]F, pulse rate of 100 beats/min, and blood pressure level 90/50 mm Fig with orthostasis. Pulse oximetry was 96% on room air. The left hemithorax was dull to percussion and there were inspiratory crepitations in the left mid-lung zones. No rhonchi or wheezes were heard; there were no heart murmurs. Hematocrit level was 32%, white blood cell count was 13.9 x [10.sup.9]/L (29% band forms and 79% neutrophils), and platelet count was 305,000/[micro]L; her alkaline phosphatase level was mildly elevated at 156 IU/L. A chest radiograph revealed a unilateral hyperlucency occupying the superior three fifths of the left lung, and cavity formation with consolidation in the left lingula (Fig. 1). Computed tomography (CT) scan of the chest demonstrated a normal right lung, hyperaeration of the left lung with mediastinal shift to the right, multiple cystic structures (the largest measuring 5 x 4 cm), and left lingular consolidation, but no centrally obstructing endobronchial masses (Fig. 2).

Treatment for lung abscess with intravenous levofloxacin and clindamycin was started empirically, and she was given intravenous fluids for rehydration. She did not improve after several days of intravenous antibiotics, even with the addition of parenteral gentamicin. All cultures for bacteria, mycobacteria, and fungi remained negative. Purified protein derivative (PPD) skin test and HIV test results were negative. She underwent fiberoptic bronchoscopy, which showed a normal right lung but purulent material throughout the left bronchial tree, with small, slit-like airways in the left lung. No endobronchial obstructing lesion was apparent. Bronchial washings were sent for bacteriologic, fungal, and mycobacteriologic cultures and cytology. Four days after admission, a surgical consultation was done because of her persisting symptoms and increasing polymorphonuclear leukocytosis. The patient initially refused all surgical options until a ventilation/perfusion (V/Q) scan showed a normal right lung and absence of b oth ventilation and perfusion throughout the superior three fourths of the left lung (Fig. 3); lung quantitative function analysis during the V/Q scan showed very minimal contribution from the left lung. She then chose to undergo a left pneumonectomy. Intra-operative bronchoscopy findings were similar to the earlier study; findings included an abnormally wide left third intercostal space, a swollen, ischemic-appearing left lung with a walled-off, ruptured lung abscess, an abnormally small left main bron-chus, a very small left pulmonary artery, and a totally consolidated left lower lobe. Pathologic examination of the pneumonectomy specimen revealed an intralobar sequestration involving most of the left lower lobe. The left upper lobe showed severe acute and chronic inflammation, with multiple acute and chronic abscesses and extensive replacement of the alveolar air spaces by plugs of chronically inflamed granulation tissue. The sequestered left lower lobe showed no inflammatory changes. She had an uneventful postoperative recovery and subsequently returned to Austria feeling well. Results of all microbiologic studies, including cultures carried out on repeat samples of sputum, blood, and bronchial washings, 4 weeks of fungal culture, and 6 weeks of mycobacteriological culture, were negative.

Our patient had features consistent with the SJMS. (3-5) The current hypothesis is that the SJMS is a form of acquired, nondestructive, segmental emphysema. (4) Repeated childhood infections with viruses or mycoplasma have been implicated in this syndrome. Our patient had a prolonged chest infection at 13 years of age that may have predisposed her to the subsequent development of SJMS. We believe that a congenital anomaly in the left lung may have led to the subsequent infection-associated, acquired, segmental emphysema. The differential diagnosis for SJMS includes congenital lobar emphysema and a host of acquired causes; acquired causes include lung abscess (post-pneumonic or postaspiration), bronchiectasis, cavitating lesions (tuberculosis and Wegener's granulomatosis), cystic states (tension bronchogenic cysts), pulmonary vascular disease, and central obstructing lesions (tumors and foreign bodies). The surgical therapy advocated for SJMS is pneumonectomy, but conservation of pulmonary tissue by segmentect omy or lobectomy, when feasible, has been recommended recently.

Macaulay Onuigbo MD, MSc

Division of Nephrology

University of Maryland School of Medicine

22 S Greene St

Room N3W143

Baltimore, MD 21201-1595

Xiaodan Qu, MD

Eugene Obah, MD

Department of Medicine

Greater Baltimore Medical Center

References

(1.) Swyer PR, James GW: A case of unilateral pulmonary emphysema. Thorax 1953; 8:133-136

(2.) MacLeod WM: Abnormal translucency of one lung. Thorax 1954; 9:147-153

(3.) Moore ADA, Godwin JD, Dietrich PA, et al: Swyer-James syndrome: CT findings in eight patients. AJR Am J Roentgenol 1992; 158:1211-1215

(4.) Ohri SK, Rutty G, Fountain SW: Acquired segmental emphysema: the enlarging spectrum of Swyer-James-MacLeod's syndrome. Ann Thorac Surg 1993; 56:120-124

(5.) Lucaya J, Gartner S, Garcia-Pena P, et al: Spectrum manifestations of Swyer-James-MacLeod syndrome. J Comput Assist Tomogr 1998; 22:592-597