often delayed. Large sclerosant volume, high concentration, and advanced liver disease have been associated with increased risk of perforation. 3 The specific complication of esophago-pleural fistula has not been previously documented (to our knowledge), and was clearly necessary for the formation of this patient's empyema. Studies performed by the food industry have shown S cerevisiae to be virtually nonpathogenic.' Our review of the medical literature found only 15 cases of serious infection in humans, including six episodes of fungal sepsis and one of empyema.s. 11 Many of these patients had underlying debilitating diseases and a portal of entry for yeast. Two patients ingested brewers yeast daily,'·" and one case of S cerevisiae peritonitis followed an abdominal procedure performed by surgical personnel who were also amateur bakers. • Our patient prepared a chicken pot pie using brewer's yeast two weeks before hospital admission. In summary, we present a patient with an esophagopleural fistula as a complication of prophylactic variceal sclerotherapy. Esophago-pleuralfistula should be considered a potentially fatal complication in any patient after sclerotherapy who presents with a pleural effusion. Organisms normally considered harmless must be considered potential pathogens in patients compromised by such a fistula. ACKNOWLEDGMENTS: \\e would like to thank Gary Gilliland, M.D., Ph.D., for his advice, encouragement, and insightful critique of this manuscript. REFERENCES
1 Sales BJ, Kilby AE, Dietrich PA, Coffin LH , Krawitt EL. Pleural and mediastinal changes following endoscopic injection sclerotherapy of esophageal varices. Radiology 1983; 149:639-42 2 Bacon BR, Bailey-Newton RS , Connors AF. Pleural effusions after endoscopic variceal sclerotherapy. Gastroenterology 1985; 88:1910-14 3 Perino LE, Gholson CF, Golf JS. Esophageal perforation after
4
5 6 7
fiberoptic variceal sclerotherapy. J Clin Gastroenterol 1987; 9:286-89 Korula J, Pandya K, Yamada S. Perforation of esophagus after endoscopic variceal sclerotherapy: incidence and clues to pathogenesis. Dig Dis Sci 1989; 34:324-29 Holzschu DL, Chandler Fw, Ajello L, Ahearn DG. Evaluation of industrial yeast for pathogenicity. Sabouradia 1979; 17:71-8 Stein PD, Folkens AT, Hruska KA. Saccharomyces fungemia. Chest 1970; 58:173-74 Eng RHK, Drehmel R, Smith SM, Goldstein EJ. Saccharomyces cerevisiae infections in man. Sabouradia 1984; 22:403-07
8 Jensen DP, Smith DL. Fever of unknown origin secondary to brewers yeast ingestion. Arch Intern Med 1976; 136:332-33 9 Dougherty SH, Simmons RH . Postoperative peritonitis caused by Saccharomyces cerevisiae . Arch Surg 1982; 117:248-49 10 Rubinstein E, Noriega ER, Simberkolf MS, Holzman R, Rahal JJ. Fungal endocarditis: analysis of 24 cases and review of the literature. Medicine 1975; 54:331-44 11 Eschete ML, West BC. Saccharomyces cerevisiae septicemia. Arch Intern Med 1980; 140:1539 12 Sethi N, Mandell W. Saccharomyces fungemia in a patient with AIDS. NY State J Med 1988; 15:278-79
Pacemaker-Lead Puncture of the Tricuspid Valve* Successful Diagnosis and Treatment ftdro A. Rubio, M.D . , Ph.D., F.C.C .P.; and Mahdi S. Al-Bassam, M.D .
During insertion, a transvenous pacing lead pierced the tricuspid valve in a 66-year-old man, causing progressive tricuspid insufficiency and congestive heart failure. The defect remained undiagnosed for ten years and was then repaired. To our lrnowledge, this is the 6rst case in which this problem has been successfully treated rather than (Chut 1991; 99:1519-20) being diagnosed at autopsy.
P
acemaker-lead puncture of the tricuspid valve is an extremely rare complication of cardiac pacing. The four cases reported so far have all been diagnosed at autopsy., .. We describe a tricuspid puncture that occurred during lead implantation but was not diagnosed until ten years later. Surgical repair resulted in a successful outcome. CASE REPORT
A 76-year-old man was admitted to our hospital with severe shortness of breath, sharp, mostly right-sided chest pain, lower abdominal pain, and massive edema of both lower extremities that had worsened over a seven-year period and had not responded to treatment. At age 66 years, the patient had sustained an extensive myocardial infarction with heart block, for which a permanent ventricular-inhibited transvenous pacemaker was implanted . One year before the present hospital admission, he had undergone repair of an atrial septal defect and aortocoronary bypass to the right coronary artery. At the time of hospital admission, the patient had a blood pressure of 130/80 mm Hg, respirations of 20/min, and an irregular pulse of 82 beats/min. He also had an irregularly irregular cardiac rhythm, a grade 316 holosystolic ejection murmur heard best at the apex, and 3 + pedal edema. Other physical findings were unremarkable. Electrocardiography showed atrial fibrillation with a controlled ventricular rate and right bundle-branch block. Right- and leftsided cardiac catheterization disclosed severe coronary artery disease and severe tricuspid insufficiency. The patient underwent sequential aortocoronary bypass from the ascending aorta to the anterolateral obtuse marginal and left anterior descending arteries with a reversed saphenous vein. The previous right coronary bypass was found to be functioning well. Once the distal coronary anastomoses had been created, the right atrium was opened and the tricuspid valve was found to be dilated to a width of7 em. The pacemaker lead traversed and immobilized the posterior leaflet, to which it was attached by scar tissue (Fig 1). The lead was divided as low as possible in the right ventricle, and the left-chest segment was removed , but the right ventricular portion remained inextricable. Nevertheless, enough of the embedded wire was removed to allow free motion of the affected leaflet. The perforated leaflet and the anteroseptal commissure were then repaired, and the tricuspid insufficiency was corrected with a de Vega annuloplasty. The pericardium was closed with a patch (GoreTex). An epicardial pacemaker was inserted, and circulatory support was initiated with an intra-aortic balloon pump. Postoperatively, the patient did well, except for a right-sided pleural effusion that was resolved with pleurocentesis. He was discharged from the hospital on the 16th postoperative day. Although *From the Departments of Surgery (Dr. Rubio) and Cardiology (Dr. AI-Bassam), HCA Medical Center Hospital, Houston. CHEST I 99 I 6 I JUNE. 1991
1519
Fatal Airway Obstruction Caused by a Mucous Ball from a Transtracheal Oxygen Catheter* George G. Burton, M.D., F.C.C.P.;t Fred A. Wagshul, M.D.;t Diannah Henderson, R.N. ,§ and Samuel Wesley Kime, M.D.II
A 50-year-old man with pulmonary fibrosis and COPD presented with worsening cough, dyspnea, chest pain, and hypoxemia of no readily apparent etiology, approximately four weeks after insertion of a transtracheal oxygen therapy catheter. Despite vigorous bronchial hygiene therapy, the patient died. Autopsy revealed obstruction of the trachea by a large mucous ball. We point out the nonspeci6city of physical and radiologic findings associated with this condition and suggest that serial Row-volume loop analysis or earlier use of 6beroptic bronchoscopy might have been of assistance in premortem diagnosis of the mucous plug. (Chest 1991; 99:1520-23)
FIGURE l. The pa<.-emaker lead penetrated the posterior lea8et, which was scarred, preventing the leaHet from moving properly and the lead from being dislodged . his lower extremities had some residual edema, he responded well after metolazone was added to his accustomed regimen of ftuosemide and digoxin. DISCUSSION
A review of the literature shows that ours is the fifth reported case of pacing-lead perforation of the tricuspid valve and the first such case to have been diagnosed in a living patient. Moreover, our case also involved the longest delay (ten years) between lead implantation and diagnosis; in the other four cases, the maximal delay was only two years. Although leaflet penetration occurred immediately on lead insertion, early tricuspid insufficiency was minimal, so the problem was not diagnosed, despite medical treatment and open heart surgery elsewhere. By the time the patient was admitted to our hospital, he had been severely disabled for about seven years. Once the problem was diagnosed, surgical repair was straightforward. Physicians should suspect this unusual complication if a new systolic murmur develops after pacemaker implantation. Such murmurs were documented in three of the aforementioned cases, including ours. Once the correct diagnosis has been made, the patient's chances for survival should be excellent. REFERENCES
1 Fishenfeld J, Lamy Y. Laceration of the tricuspid valve by a pacemaker wire. Chest 1972; 61:697-98 2 Petterson SR, Singh JB, Reeves G, Kocot SL. Tricuspid valve perforation by endocardial pacing electrode. Chest 1973; 63:12526 3 Gould L, ReddyCVR, Yacob U, Teich M, DeMartino A, DePalma D, et al. Perforation of the tricuspid valve by a transvenous pacemaker. JAMA 1974; 230:86-7 4 Christie JL, Keelan MH Jr. Tricuspid valve perforation by a permanent pacing lead in a patient with cardiac amyloidosis: case report and brief literature review. PACE 1986; 9:124-26
1520
TTo1 T =transtracheal oxygen therapy; ITOT =Institute of Transtracheal Oxygen Therapy
he use of percutaneously inserted TTo,T catheters in T hypoxemic patients has been reported to result in less
dyspnea, greater acceptance by the patient of oxygen dependency, and physiologic, aesthetic, and financial benefits (the latter due to variable reductions in gas flow necessary to maintain adequate tissue oxygenation). Few serious and no fatal complications of TTo,T have yet been reported. We report a case of fatal obstruction of the trachea by a 7.0 x3.0x 3.0-cm mucous ball around a TTo,Tcatheter. CASE REPORT
A 50-year-old man was found to have pulmonary fibrosis at open lung biopsy in February 1987. In addition to his work as a metal grinder in a foundry, he had a 70-pack-year smoking history, and pulmonary function tests had shown obstructive and restrictive disease. At that time, he had significant exercise desaturation and was discharged to his home on oxygen therapy {1.0 Umin at rest and 2.0 Umin with exercise). On May 26, 1989, the patient was admitted with increasing dyspnea, fever, chills, and dusky cyanosis. His resting values on room air were PaO, of 46 mm Hg and PaCO, of 24 mm Hg. A ventilation-perfusion lung scan suggested low probability for pulmonary emboli. No pneumonia was found. His COPD was optimally treated, and his condition improved. On June 2, 1989, the stent of a Tio,T catheter (SCOOP; Transtracheal Systems) was inserted percutaneously, with careful instructions being given to the patient and his wife regarding its maintenance, including twice daily cleaning of the stoma. The Tio,T device was inserted by one of us, a physician formally trained in its use. One week later, on the basis of exercise saturation, an oxygen How of 2 Umin at rest and 3 U *From the Respiratory Services Department, Kettering Medical Center, Kettering, Ohio, and Wright State University School of Medicine, Dayton. tChief, Respiratory Services, Kettering Medical Center, and Clinical Professor of Medicine and AnesthesioiOf..'Y· tAssistant Clinical Professor of Medicine. §Supervisor, Pulmonary Rehabilitation Program, Kettering Medicine Center. II Associate Clinical Professor of Pathology., Reprint requests: Dr. Burton, Kettering Medical Center, 3535 Northern Blvd, Kettering, Ohio 45429 Fatal AJrway Obstruction (Burton et al)