35:439-443, 1975 3 Jones RC, Jahnke EJ: Coronary artery-atriovenbicular .&stula and venbicular septal defect due to penetrating wound ·of the heart. Circulation 32:995-1000, 1865_.~.,. _'.' 4 Storey CF, Kuzman WJ: Traumatic coronary artery-right abia1 &stula. Ann Thorac Surg 4:352-359, 1967 5 Tsagaris TJ, Bustamante RA: Coronary arteriovenous &stula and myocardial infarction due to trauma. Am J CardioI18:777-781, 1966 6 Sarot lA, Schechter DC, Weber DJ: Post-traumatic coronary arteriovenous fistulas with surgical cure. Ann Thorac Surg 6:172-177, 1968 7 Aaron BL, Doohen DH: Traumatic coronary artery-right abial &stula caused by a penetrating metal fragment. J Trauma 13:81-84, 1973 8 Cheng TO, Adkins PC: Traumatic aneurysm of left anterior descending coronary artery with fistulous opening into left venbic1e and left venbicular aneurysm after stab wound of chest. Am J Cardiol 31 :384-390, 1973 9 Siepser SL, Kalbnan AJ, Mills N, et a1: Coronary collateral flow after traumatic &stula between right coronary artery and right amum. N Engl J Med 287:754-756, 1972 10 North RL, BIalce HA, Nelson WP: Coronary artery right abia1 &stula secondary to bullet wound of the heart: Report of a case with successful surgical repair. Milit Med 136:267-270, 1971 11 Parmley LF, Marion WC, Mattingly TW: Penetrating wounds of the heart and aorta. Circulation 17:953-973, 1958 12 Gasul BM, Arcilla RA, Fell EH, et al: Congenital coronary arteriovenous fistula. Pediabics 25:531-560, 1960

Esophagopleural Fistula after Pneumonectomy· I. David Richardson, M.D.; Donald Cam."beU, M.D., F.C.C.P.; and I. Kent Trinkle, M.D., F.C.C.P. Rupture of the esophagus Into the space left alter pneumonedomy Is • nre and often fatal compUcation. Esop....ople...... 8ItuIas occurred In three patients following pneumonedomy. After previous methods failed, two patIe.... were _ treated by ...... a one",e procedure wldeh Included (1) suture closure of the 8ItuIa, (2) buttressing the repair with a viable, pedicled, two-rib IatefCOltlll-musele flap, and (3) perfo......... an estenslve thoracoplasty with a continuous drip infusion of neomycin. Such • procedure ofters the muimum opportunity for successful treatment of this catastrophic lesion. of the esophagus into the space left after Rupture pneumonectomy a rare and frequently fatal comis

plication. The three cases presented illustrate the difficultiesin the. management of this catastrophe. One patient died in the immediately postoperative period after a colon bypass procedure. The other two patients have

remained asymptomatic following fistula closure by the procedure advocated by Engleman and co-workers. 1 CASE REPoRTS CASE

1

A 61-year-old man underwent pleuropneumonectomy in

June 1968 for a destroyed right lung with empyema and persistently positive sputum cultures for acid-fast bacteria. During the difficult dissection, a small area of the middle portion of the esophagus was contused without mucosal perforation. The muscular wall was reinforced with sutures. Three weeks after surgery, the patient developed an esophagopleural fistula at the level of the repair. After initial drainage with a tube, a two-layer repair of the 1.5 X 1 em . perforation was performed with a gastrostomy and cervical loop esophagostomy. Within a few days the fistula recurred, and open-ehest drainage was performed. The patient was maintained by feedings through the gastrostomy tube, with the saliva diverted through the cervical esophagostomy. The fistula failed to close. In April 1969, a substernal right colonbypass procedure was performed. The patient died suddenly on the seventh postoperative day from a massive pulmonary embolus. Although oral feedings had not been started, the colon transplant and suture lines were intact. CASE

2

A 51-year-old man was admitted in 1972 with a destroyed right lung. The sputum culture was negative for tuberculosis and fungi, but a variety of pathogenic organisms was cultured. Right pneumonectomy was performed. The esophagus was not adherent and was not injured. Cultures of the resected lung revealed Bacteroides organisms, and appropriate antibiotic therapy was administered. The patient did well until three weeks after surgery, when a small bronchopleural6stula developed, requiring drainage with a tube. Neither transbronchoscopic cauterization nor secondary closure with pleural flaps was successful in obliterating the fistula. Open drainage of the pleural space (Eloesser2 flap) was perfonned; the patient was subsequently discharged, and the fistula closed. Approximately one year later, after a night of CCindigestion" and cCretching," the patient noted that oral fluids came out of the pleural drainage site. A large midesophageal fistula was demonstrated with a swallow of meglumine diatrizoate (Gastrografin) . Esophagoscopic and fiberoptic bronchoscopic examinations through the Eloesser flap showed a 1.5 em defect. The right pleural space was reopened, and a portion of the dense mediastinal scar was divided to permit a two-layer closure of the esophageal perforation. A pedicled flap of the fourth and fifth intercostal bundles was .sutured over the repaired esophagus. A nine-rib thoracoplasty and feeding jejunostomy were also performed. After surgery the patient did well, except for development of duodenal obstruction apparently due to a superior mesenteric arterial syndrome accentuated by high placement of the jejunostomy tube. Intravenous hyperalimentation was added to provide additional caloric intake. Barium swallow showed an intact esophagus. One year later, the patient was eating and gaining weight. CASE

3

°From the San Al)tonio State Chest Hospital and the Division of Cardiothoracic Surgery, the University of Texas Health Science Center, San A:ntonio. Reprint requests: Dr. Richardson, Department of Surgefll, 7703 Floyd Curl, San Antonio 78284

A 62-year-old woman underwent left pneumonectomy in 1971 for a destroyed lung with hemoptysis due to tuberculosis. After surgery the patient developed a bronchopleural fistula and empyema requiring open thoracostomy (Eloesser flap). The open thoracostomy gradually closed over the next

CHEST, 69: 6, JUNE, 1976

ESOPHAGOPLEURAL FlmU AFTER PNEUMONECTOMY 795

year; however, the patient steadfastly refused to have the flap revised. Consequently, the bronchopleura1 fistula remained open. Two years later, oral fluids and food began passing through the thoracostomy. Barium swallow and esophagoscopy showed a 2 em perforation of the esophagus at the level of the hilum, with no distal esophageal obstruction. Oral intake was stopped, and intravenous hyperalimentation was instituted. Two weeks later, left thoracotomy was performed. A small opening in the bronchial stump was sutured, and the esophagus was repaired with one layer of silk sutures. An eight-rib thoracoplasty obliterated the pleural space. Ten days later, the postoperative barium swallow showed a residual esophageal leak. At reoperation a wider exposure of the esophagus was obtained, and the perforation was closed in two layers. The repaired esophagus and the intact bronchial stump were covered with an intercostalmuscle pedicle. After surgery the patient continued to receive intravenous hyperalimentation for three weeks before resuming oral feedings. A barium swallow performed ten weeks after her second esophageal repair showed no evidence of a leak. The patient remains asymptomatic one year later. DISCUSSION

The most common cause of esophageal fistula after pneumonectomy is injury to the esophagus· during the removal of adherent lung, pleura, or lymph nodes. In this event the fistula manifests itself early in the postoperative course, as in our first case. Although causes of late esophageal rupture are less obvious, chronic infection of the pleural space is one likely mechanism. In two of our cases, the esophagus ruptured more than 15 months after open-chest drainage had been instituted for a postoperative bronchopleural fistula wi~ empyema. In the third case, chronic infection certainly seems culpable because the Eloesser flap had narrowed to the point that adequate drainage of the residual bronchopleural fistula was impossible. In the second case the esophagus apparently perforated after "retching:~ suggesting that the mechanism was similar to that of the Boerhaave syndrome, 3 without the buttressing effect of the IWlg. The esophagus thus ruptured at the area of least resistance, the empty right pleural cavity. Treatment of this condition is difficult, and procedures to close the fistula have a high rate of failure. Takaro et al· found that only 21 percent of their collected cases of esophagopleural fistulas were eventually cured, and the mortality was 51 percent. Immediate treatment consists of cessation of oral intake and drainage of the pleural cavity. Since the esophagus ustially perforates freely into the pleural space, immediate thoracotomy with mediastinal drainage is not indicated. If the patient is in poor nutritional balance, correction by intravenous hyperalimentation may be required. While nonoperative measures alone are rarely successful, occasional closures of small fistulas have been reported with only nutritional support and drainage. Urschel and co-workers 5 reported successful closure of several late fistulas using esophageal exclusion and diversion in continuity, in addition to drainage, nutritional support, and antibiotic therapy; however, none of their fistulas followed pneumonectomy, and all had the advantage of the buttressing IWlg. Direct.

798 RICHARDSON, CAMPBElL, TRINKLE

suture closure has a high incidence of failure unless the suture line is covered with a flap of viable tissue. Benjamin et ale reported two cases of successful secondary repair using viable pleural flaps. Successful resection and end-to-end anastomosis of the esophagus after failure of direct closure has also been achieved. Esophageal bypass procedures with stomach, small bowel, or colon have occasionally been successful, but the mortality is high with these extensive procedures. Successful closure of the fistula after thoracoplasty alone has been reported.· In 1965, Bryant and Eiseman1 demonstrated the value of the intercostal pedicle as an adjunct in aiding esophageal repair in experimental animals. Engleman and associates 1 subsequently performed a successful onestage procedure for the closure of a large esophageal defect. The operative technique included suture closure of the fistula, covering this repair with a pedicled two-rib intercostal-muscle Hap, and performing an extensive thoracoplasty with a continuous neomycin infusion of the pleural space. In our two successfully treated cases, we employed this technique after failure of direct suture closure. The recent reportS of another successfully treated case of esophageal rupture after pneumonectomy using an intercostal-muscle Hap indicates the value of the procedure in the treabnent of this unusual complication. AdjWlctive measures, such as intravenous hyperalimentation, which was used in our two successfuDy treated cases, allows the patient to achieve a state of anabolism and will undoubtedly improve the survival rates with this complication. If the perforation of the esophagus is operated early, before longstanding changes of inHammation and infection develop in the space left after pneumonectomy, then the addition of a thoracoplasty may not be necessary. Procedures such as that described by Clagett and Geraci9 may be successfully used; however, in the resections for tuberculosis where longstanding infection has occurred, we believe that thoracoplasty is essential. Nonetheless, the critical factor in the success of this excellentproeedure is the careful approximation of a pedicled flap of two intercostal-muscle bWldles with an intact blood supply over the repaired esophagus.

1 Engleman RM, Spencer Fe, Berg P: Postpneumonectomy esophageal fistula: Successful one-stage repair. J Thorac Cardiovasc Surg 59:811, 1970 2 Eloesser L: Of an operation for tuberculous empyema. Surg Gynecol Obstet 60: 1096, 1935 3 The Boerhaave syndrome, (editorial). JAMA 187:57, 1964 4 Takaro T, WaDcup HE, Okano T: Esophagopleural fistula as a complication of thoracic surgery. J Thorac Cardiovasc Surg 40:179, 1960 5 Urschel He, Razzuk MA, Wood RE, et al: Improved manage~ent of esophageal perforation: Exclusion and diversion in continuity. Ann Surg 179:587, 1974 6 Benjamin I, Olsen A, Ellis FH: Esophagopleurai fistula: A rare postpneumonectomy cOmplication. Ann Thorac Surg 7:139,1969

CHEST, 69: 6, JUNE, 1976

7 Bryant LR, Eiseman B: Experimental evaluation of intercoastal pedicle grafts in esophageal repair. J Thorae Cardiovasc Surg 50:626, 1965 8 Eftbimiadis M, Xanthakis D, Primikyrios N, llt ,al:.4te esophagopleural fistula after pneumonectomy for bronchial carcinoma. Chest 65:579-580, 1974 9 Clagett OT, Geraci JE: A procedure for the management of postpneumooectomy empyema. J Thorae Cardiovasc Surg 45:141,1963

Alternation of Left Ventricular Performance with Eledrical Alternans* Denni.t V. Cokkinos, M.D., F.C.C.P.; John N. Demopoulos, M.D.; Was T. Heimonas, M.D.; and Eutrlchias M. VOt'itlU, M.D.

Systolic time intervals (STIs) were measured in a patient with massive pericardial effusion and electrical alternans. De total electromechanical systole and the left ventricular ejection time (LVET) were sipificantly shorter in beats foRowing the smaller QRS complexes, wbDe the pre-ejedion time (pEP) was longer and the ntio of PEP/LVET Iarpr. 1bis beat-to-beat difference in the SI1a lIIIIestII tIwl an alternation in left ventricular perfol'DUlDCe accompanied the eIectrkal a1ternans. After evacnadon of the pericardial 8nid, electrical a1ternans and alternation dJappe8red.

m

I

t is currently believed that excessive rotation of the heart due to a great accumulation of pericardia! fluid accoWlts for electrical alternans. 1,2 °From the Second Professorial Medical Unit, Evangelismos Medical Center, Athens, Greece. Reprint requests: Dr. CokkiflO$, 5 Doryloiou, Athens 601,

Greece

II

We employed systolic time intervals (STIS)3 in the study of beat-to-beat changes in left ventricular performance in a patient with electrical alternans. This technique is widely accepted as a reliable method of evaluation of cardiac function 4 and has been used for the detection of phasic changes in cardiac performance. 5 ,8 CASE REPoRT

A 19-year-old girl had precordial pain accentuated with respiration, dyspnea, and evening temperature elevations. The blood pressure was 122nO rom Hg; the jugular venous pressure was elevated 2.0 em above the sternal angle at 45-. The cardiac impulse could not be palpated. The heart sounds were mufBed; no friction rub or murmur was heard. On a chest x-ray film, gross cardiac enlargement was seen. On the electrocardiogram, there was definite electrical alternans (Fig 1). The STIs were recorded on a multichannel recorder (Hewlett-Paclcard A4578). The paper speed was 100 rom/sec. As proposed by Weissler et al,4 the following intervals were measured: total electromechanical systole ( QS2 ), left ventricular ejection time (LVET), pre-ejection period (PEP), and the ratio of PEP/LVET. Thirty consecutive heart beats were recorded. Inspiratory movements were marked on the chart paper by a manual marker. The STIs following the larger QRS complexes were added as one group and compared to those of the beats following the smaller QRS complexes by use of Student's paired t-test. The QS2 and LVET intervals were significantly shorter in beats following the smaller QRS complexes, while the PEP was longer and the PEP/LVET ratio was larger (Fig 2 and Table 1). Respiratory motions did not seem to affect the LVET in a different manner from that found in normal subjects by Weissler et al. 7 Paradoxical pulse was not seen at any time in the LVET or blood pressure, nor was pulsus alternans detected by palpation or measurement of blood pressure. After removal of 400 ml of yellowish fluid, electrical alternans disappeared. The STI no longer showed differences

II

• •VI

.Vf

++++++++++++±

· tttI

+++++++++.+++-~

·t++-t-H-'t-+-IH-++-I V,

· · · · · 7...

....... : ... -.-- .. : . . . . ,. . , : '. . . : . . .••• -. -. --_ .. .... -._--._.--

V,

V.

V.

v,

~

-- : ;

I 1

-

V.

ft;'r+tr-HI-tHHHr-HH

CHEST, 69: 6, JUNE, 1976

FIGURE 1. Scalar ECG before fluid evacuation (double sensitivity). There is definite electrical alternans, best seen in leads I, aVL, aVF, VI, and Va.

ALTERNATION OF LEFT YENTRICUW PERFORMANCE 797

Esophagopleural fistula after pneumonectomy.

Rupture of the esophagus into the space left after pneumonectomy is a rare and often fatal complication. Esophagopleural fistulas occurred in three pa...
1MB Sizes 0 Downloads 0 Views