- Email: [email protected]
Concomitant cardiac and pulmonary operation: The role of cardiopulmonary bypass
Concomitant Cardiac and Pulmonary Operation: The Role of Cardiopulmonary Bypass Karl S. Ulicny, Jr, MD, Victor Schmelzer, MD, John B. Flege, Jr, MD, Joseph C. Todd, MD, Donald L. Mitts, MD, David B. Melvin, MD, and Creighton B. Wright, MD Division of Cardiothoracic Surgery, Department of Surgery, The Jewish Hospital of Cincinnati, and Division of Cardiothoracic Surgery, Department of Surgery, The Christ Hospital of Cincinnati, Cincinnati, Ohio
To assess the safety and efficacy of concomitant pulmonary resection and cardiac operation requiring cardiopulmonary bypass, the records of 19 patients were reviewed. Eighteen patients (94.7%) presented with cardiac symptoms and were found to have pulmonary pathology of indeterminate etiology. Pulmonary resections were performed through a median sternotomy in all but 1patient, who underwent posterolateral thoracotomy and right middle lobectomy after repositioning because dense adhesions prevented adequate dissection through the initial incision. A total of 24 resections were performed. Sixteen (66.7%) were performed on cardiopulmonary bypass. Six wedge resections (25.0%) were performed before bypass. Two lobectomies (8.3%) were performed after infusion of protamine sulfate. Nine patients (47.4%) had benign pathology, 7 (36.8%)had primary carcinoma, and 3 (15.8%) had metastatic disease. Bleeding complications occurred in 15.8% of patients (3/19). There was 1 perioperative death (5.3%), which was due to adult respiratory distress syndrome after intraoperative hemorrhage followed lobectomy for bullous disease. Another patient required lateral extension of the sternotomy during an episode of exsanguinating intraparenchymal pul-
monary hemorrhage, which resulted in lobectomy, as well as costochondral and sternal osteomyelitis. A third patient required exploration for bleeding at the staple line. Postoperative complications occurred in 7 patients (36.8%)and were predominantly respiratory (517, 71.4%) ( p = 0.006). The median postoperative hospitalization was 15 days. Although comparison of patients who underwent pulmonary resection during bypass with those who had resection either before heparinization or after protamine infusion showed no significant difference with respect to age, incidence of malignancy, operation performed, complications, postoperative hospitalization, or survival, this was probably due to the small number of patients in the study. Survival correlated only with the diagnosis of malignancy ( p = 0.042). Pulmonary resection performed on cardiopulmonary bypass leads to excessive bleeding and pulmonary complications and perhaps to excessive hospitalization. If concomitant correction of both cardiac and pulmonary conditions must be performed, pulmonary resection should be accomplished after reversal of anticoagulation to prevent excessive bleeding. (Ann Thorac Surg 1992;54:289-95)
P
Concomitant operations might allow prompt correction of both conditions, thereby sparing the patient a second major thoracic procedure with its attendant risks and expense. In the most recent study [6] to address the question of concomitant operations, no pulmonary resection was performed during anticoagulation. The purpose of this study was to assess the safety and efficacy of concomitant pulmonary and cardiac operations with respect to the timing of pulmonary resection in relation to cardiopulmonary bypass (CPB).
ulmonary disease and cardiovascular disease have several common risk factors. When they occur simultaneously in a patient, this creates a complex therapeutic challenge. Patients usually present with cardiac symptoms and are subsequently found to have pulmonary pathology. Less often, patients are found to have major cardiac pathology during evaluation for pulmonary resection [l]. Traditionally, the two diseases have been treated in a staged fashion depending on clinical priority. Reluctance to perform combined procedures has been based on concerns about adequate pulmonary exposure, as well as bleeding complications secondary to the systemic heparinization required for the cardiac portion of the operation [l]. Although median sternotomy has been shown to be safe and efficacious for pulmonary resection [2-61, the effects of systemic heparization are not known. Accepted for publication Jan 9, 1992 Address reprint requests to Dr Schmelzer, 2123 Auburn Ave, Cincinnati, OH 45219.
0 1992 by The Society of Thoracic Surgeons
Material and Methods
Patient Population The clinical records of all patients who underwent concomitant pulmonary and cardiac operations that required CPB from February 1980 through January 1990 were reviewed. The patient population consisted of 19 patients (15 men and 4 women) with a mean age (* the standard 1.8 years (range, 52 to 78 error of the mean) of 63.3
*
0003-4975/92/$5.00
290
Ann Thorac Surg 1992;54:289-95
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
years) (Table 1). Eighteen patients (94.7%)presented with unstable cardiac symptoms (New York Heart Association class I11 or IV) resulting from coronary artery disease (n = 13),valvular disease (n = 4), and left atrial tumor (n = 1). One patient presented with massive hemoptysis secondary to an Aspergillus cavitary lesion, and unstable angina developed during the evaluation. Another patient was scheduled for coronary artery bypass, which was postponed because of an asymptomatic left lower lobe infiltrate. Pulmonary lesions were identifiable in all but 1 patient on preoperative chest roentgenograms. Fourteen patients (73.7%)had a new solitary nodule, 2 (10.5%)had multiple nodules, 1 had an infiltrate, and 1 had a cavitary lesion.
One patient underwent left upper lobe wedge resection before bypass to prevent injury to a planned left IMA graft. Wedge resection of the right upper lobe lesion was later performed on bypass.
Anatomy of Resection
Results
Twenty-four pulmonary resections were performed in the 19 patients (Table 2). Standard surgical techniques were employed for all resections, and mechanical staplers were used for all bronchial closures and all wedge resections. The 14 wedge resections were performed in 12 patients; 1 later required lobectomy to control intraparenchymal hemorrhage. A second patient underwent lobectomy after wedge resection documented malignancy. Six patients underwent lobectomy as the primary pulmonary procedure. One patient underwent left upper lobectomy and right upper lobe wedge resection. Seventeen operations (89.5%) in the 19 patients were performed through a median sternotomy alone. One patient, who underwent wedge biopsy of a persistent left lower lobe infiltrate before bypass, required lateral extension of the sternotomy into the fifth interspace to control exsanguinating intraparenchymal hemorrhage, which occurred after heparinization. In a second patient, dense adhesions around a middle lobe Aspergillus cavitary lesion prohibited hilar dissection. Because he had presented with massive hemoptysis, the patient was repositioned and a posterolateral thoracotomy was performed after completion of coronary artery bypass grafting.
There was 1 perioperative death (within 30 days of operation or within the same hospitalization) (5.3%), which occurred in a 72-year-old man who underwent aortic valve replacement and mitral valve repair. In the dissection of a densely adherent IMA graft (previous operation), violation of numerous left upper lobe bullae occurred, necessitating lobectomy, which was performed after CPB. Despite reversal of anticoagulation, severe diffuse bleeding with hemodynamic instability ensued. The patient died on postoperative day 2 of adult respiratory distress syndrome affecting the right lung. Most of the major complications were pulmonary (p = 0.006). Six additional patients (31.6%) had postoperative complications related to pulmonary resection. One patient required reoperation for bleeding at the staple line after wedge resection performed on CPB. Five patients (26.3%) required prolonged ventilatory support. Four required 4 to 7 days of initial respiratory support, 1 of whom had sequelae from previous poliomyelitis, and the fifth required reintubation on postoperative day 6 for an additional 6 days. Three of them underwent lobectomy, 1 had a segmentectomy, and 1, a wedge resection. Two patients (10.5%), 1 with lobectomy and 1 with wedge resection, had pneumonia postoperatively. There were no prolonged air leaks. One patient required chest tubes for 7 days because of excessive drainage. In the patient who underwent lobectomy for intraparenchymal hemorrhage, osteochondritis of the fifth rib and sternal osteomyelitis developed and myocutaneous flap reconstruction was performed. Other complications are listed in Table 1. All atrial (n = 5) and ventricular (n = 4) dysrhythmias responded pharmacologically. The median postoperative hospitalization was 15 days (mean period, 19.1 2 2.7 days; range, 8 to 51 days). The occurrence of major postoperative, complications was not significantly related to resection on CPB, operation performed, or age of the patient. Nine patients (47.4%)had benign pulmonary disease, 5 patients had bronchogenic carcinoma (26.3%), 2 had carcinoid (10.5%), and 3 (15.8%) had metastatic processes (Table 4). Four patients with primary adenocarcinoma underwent lobectomy for cure. The fifth primary malignancy was treated with wedge resection. The 3 patients with pulmonary metastases were a 63-year-old woman
Timing of Resection Pulmonary resections were performed at various times in relation to CPB (Table 3 ) . Three wedge resections were accomplished in 2 patients before bypass and anticoagulation. In the first patient, a right upper lobe wedge resection was performed before placement of bilateral internal mammary artery (IMA) grafts, and in the second, bilateral wedge resections were performed before resection of a left atrial leiomyosarcoma to document histology. Six wedge resections, five lobectomies, and one segmentectomy were performed after the cardiac procedure while the patient was still on CPB. A large, nondeflatable right upper lobe bullous region was resected before bypass, and the patient then underwent left lower lobectomy after grafting while on bypass to facilitate exposure. One left lower lobectomy was emergently performed on bypass to control intraoperative hemorrhage that ensued after an initial wedge resection performed before anticoagulation. One left upper lobe wedge resection was performed after vein grafting but before grafting of the IMA.
Statistical Analysis
Discrete variables were analyzed by x2 or Fisher’s exact tests, and one-way analysis of variance was used to examine differences between continuous variables. Survival was determined by actuarial analysis using the methods of Kaplan and Meier [7, 81 with the day of operation as the starting time. Differences resulting in a p value of less than 0.05 were considered significant.
CVG x 3, LLL WR CVG x 4, RUL WR, RU lobectomy LLL WR, RUL WR, L atrial tumor CVG x 5, RM lobectomy MVR, RLL WR
M
M F M
F M
M
M
M M
56
61 59 61
52 74
67
58
68
69
58
77 78
62
72
54
4
5 6 7
8 9
10
11
12
13
14
15 16
17
18
19
RUL WR, Bilateral IMA LUL WR, CVG x 2, LIMA x 1, RUL WR CVG x 2, LLL WR, LIMA x 1 MVR, AVR, LU lobectomy CVG x 4, LIMA x 1, LUL WR
LLL WR, CVG x 5, LL lobectomy
CVG x 5, LLL WR
14
16
... ... 12 26 14 10 17
28 14
8
... ... ...
... ...
... 22
...
...
...
15
Atrial fibrillatiodventricular tachycardia, acalculous cholecystitis
Death, ARDS
Atrial fibrillation, pericarditis
... Atrial fibrillation
Reoperation for staple line bleeding Intraop pulmonary hemorrhage, IMA disruption, osteomyelitis
Ventricular ectopy, herpes zoster Atrial fibrillation
Pericarditis Pericarditis, ventricular ectopy Pneumonia
...
Atrial fibrillation Pericarditis
10
...
1
... UGI bleeding, ventricular ectopy Reoperation for sternal bleeding Pneumonia, high chest-tube output
Complications
10
51
... ... ...
18
...
... ...
...
6
17 41
(4
Postop Hospitalization
...
Reintubation Day
1
1 1
7
2
1
1
4
1 3
3 1 7
4
1
1 1/12
Extubation Day
adenocarcinoma
? Metastatic
Bullous disease
Carcinoid
His toplasma Inflammatory
Undifferentiated large cell carcinoma, stage I1 Pneumonia
Carcinoid
Aspergillus
L atrial leiomyosarcoma
Histoplasma Adenocarcinoma, stage I
Bullous disease (R), adenocarcinoma (L), stage I Adenocarcinoma, stage I1 Histoplasma Hamartoma
Tuberculosis
Spindle cell sarcoma Adenocarcinoma, stage I1
Pathology
AWD, 10 mo; 7 RUL metastasis
...
DF, 47 mo DF, 47 mo; coronary disease DF, 29 mo
DF, 55 mo; CHF
DDF, 65 mo; redo MVIVAVR DOD, 3.5 mo
DF, 69 mo
DOD, 18 mo
DOD, 21 mo DF, 93 mo DF, 90 mo; coronary reoperation X 2 DF, 86 mo DF, 83 mo; CHF
DF, 94 mo
DF. 127 mo
DOD, 89 mo DOD, 13 mo
Follow-up
ARDS = adult respiratory distress syndrome; AVR = aortic valve replacement; AWD = alive with disease; CHF = congestive heart failure; CVG = coronary venous grafting; DDF = died DF = disease-free; DOD = died of disease; IMA = internal mammary artery graft; LIMA = left internal mammary artery praft; LL = left lower; LLL = left lower lobe; LU disease-free; = left upper; LUL = left upper lobe; MVR = mitral valve replacement repair; RL = right lower; RLL = right lower lobe; RM = right middle; RML = right middle lobe; RU = right upper; RUL = right upper lobe; UGI = upper gastrointestinal; WR = wedge resection.
M
M
M
M
F
M
AVR, RU lobectomy CVG X 5, RUL WR CVG x 3, MVR, LLL segmentectomy
M
53
3
CVG x 4, RL lobectomy CVG x82, RM lobectomy Redo CVG x 3, LL lobectomy RUL WR, CVG x 2, LU lobectomy
F M
63 61
1 2
Operations
(Y)
Sex
Age
Patient No.
Table 1. Summary of Data on Patient Population
292
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
Ann Thorac Surg 1992;54:289-95
Table 2. Type and Location of 24 Pulmonary Resections in 19 Patients Undergoing Concomitant Cardiac and Pulmonary Procedures' Location
Wedge
Resection Segmentectomy Lobectomy
Table 4. Pulmonary Pathology in 19 Patients Undergoing Concomitant Cardiac and Pulmonaw Procedures" No. of Patients
Pathology Total
~~
Benign
Right
Upper
Hamartoma
1 (5.3)
6
0
2
8 (33.3)
Histoplasma
3 (15.8)
Middle lobe Lower lobe Left Upper lobe Lower
0
0
2
2 (8.3)
Tuberculosis Aspergillus
1
0
1
2 (8.3)
Pneumonia Bullae
1 (5.3) l(5.3) l(5.3) 2b (10.5) l(5.3)
lobe
lobe Total
a
2
0
2
4 (16.7)
5
1
2
8 (33.3)
1 (4.2)
9 (37.5)
14 (58.3)
Inflammatory Malignant Primary
Adenocarcinoma Undifferentiated large cell Metastatic Spindle cell sarcoma Atrial leiomyosarcoma Adenocarcinoma
24
Numbers in parentheses are percentages.
with a solitary right lower lobe metastasis from a spindle cell sarcoma resected from the right arm 20 years earlier, a 67-year-old man with a left atrial leiomyosarcoma and two pulmonary metastases, and a 54-year-old man with poorly differentiated adenocarcinoma of probable gastrointestinal origin. Follow-up, conducted by telephone interview and review of all medical records, was current for all patients and averaged 57.8 8.5 months (range, 3.5 to 127 months). Of the 8 patients with only benign disease who survived operation, all are currently alive 47 to 127 months after operation without evidence of dissemination caused by concomitant operation. Of the 5 patients with bronchogenic carcinoma, 2 are currently alive without disease at 83 and 94 months, and 3 died of disease at 3.5, 13, and 21 months after operation. One patient with carcinoid is currently alive without evidence of recurrence 29 months after operation. The second died without evidence of recurrence 65 months after resection after aortic and repeat mitral valve replacement. Of the 3 patients with metastatic disease, 1 is currently alive with recurrence 10 months after operation, and 2 died of disease at 18 and 89 months postoperatively.
*
4b (21.1) 2 (10.5) l(5.3)
Carcinoid
1 (5.3) l(5.3) 1 (5.3)
Numbers in parentheses are percentages. had both bullous disease and adenocawinoma.
a
One patient (patient 4)
The observed survival rate for all patients undergoing combined procedures was 73.7% at 5 years and 68.4% at 10 years (Fig 1). Survival was not affected by age, occurrence of major postoperative complications, resection on CPB, or operation received. Survival did correspond to the diagnosis of malignancy ( p = 0.042). For all patients with cancer, the 5- and 10-year survival rates were 60.0% and 50.0%, respectively. Patients with bronchogenic carcinoma fared worse than those with metastatic disease; 5-year survival rates were 40.0% and 50.0%, respectively ( p = not significant).
Comment Concomitant pulmonary and cardiac operations have not traditionally been performed. The most clinically important disease was addressed first, followed by a second operation at a later date for the remaining problem.
Table 3. Timing of Pulmonary Resection in Relation to Cardiopulmonary Bypass in 19 Patients Undergoing Concomitant Cardiac and Pulmonary Procedures' Timing Before bypass Before bypass and during bypass after cardiac procedure On bypass during cardiac procedure On bypass after cardiac procedure After bypass Total Numbers in parentheses are percentages.
Wedge Resection
Multiple Wedge Resection
Wedge Resection and Lobectomy
Lobectomy
Segmentectomy
Total
1 0
1 1
2
0 0
0 0
2 (10.5) 3 (15.8)
1 5 0
0 0 0
0 1 0
0 4 2
0 1 0
l(5.3) 11 (57.9) 2 (10.5)
7 (36.8)
2 (10.5)
3 (15.8)
6 (31.6)
1 (5.3)
19
0
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
Ann Thorac Surg 1Y92:54:28Y-Y5
293
Fig 1 . Observed percent survival of patients undergoing concomitant cardiac and pulmonary procedures. Bronchogenic carcinoma (CAI excludes the diagnosis of carcinoid. Metastatic disease includes adenocarcinoma ( n = Z), spindle cell sarcoma (n = l ) , and atrial leiomyosarcoma ( n = I).
Percent ALLPATIENTS
+ ALLBENIGN
-113
20-
0
I
.
.
.
.
.
-
Prolongation of the operative time for concomitant procedures was believed to increase the operative mortality. The second procedure was delayed for 4 to 6 weeks to allow the patient time to gain weight and muscle mass [9]. The traditional approach to bilateral pulmonary disease has been staged lateral thoracotomies, 7 to 90 days apart, depending on the disease progression and the condition of the patient [lo]. Concomitant procedures would be attractive if both could be performed through a median sternotomy and if both disease processes could be managed safely. In this study, 78.9% of patients (15/19 patients) underwent pulmonary resection during CPB. The only other reports of concomitant operations are by Piehler and colleagues [l]and Canver and associates [6]. The majority of the 43 resections in the former study were performed either before (39.5%) or after (48.8%) bypass; only five (11.6%) were performed on bypass. Similarly, Canver’s group carried out pulmonary resection before heparinization in 57.1% of patients and after protamine reversal in 42.9%. In contrast, the majority of pulmonary resections (63.2%) in this review were performed under full anticoagulation. An additional 15.8% of operations consisted of two resections, one of which was performed before bypass and the other on bypass after the cardiac procedure. The remainder were performed either before (10.5%) or after (10.5%) anticoagulation. Some patients required a combined approach because of the nature of their disease. The patient with Aspergillus, for example, had life-threatening hemoptysis and severe unstable angina. Another patient confronted the surgeon with left upper lobe bullous disease densely adherent to a previously placed IMA graft that required mobilization. Another, with left upper lobe adenocarcinoma, required resection of nondeflatable bullous disease for exposure. One patient was seen with bilateral pulmonary metastases from a left atrial leiomyosarcoma. It could be argued that the remaining patients should have been operated on in a staged fashion, but data supporting this concept have been lacking. Nevertheless, this group of patients allows
I
.
ALLCANCER BRONCHOOENICCA YETASATICCA
-
I
comparison of both the advantages and the disadvantages of pulmonary resection before, during, and after bypass. As statistical comparison is difficult in this small group of patients because of the lack of statistical power, some potentially large differences between resection on CPB and survival, hospitalization, and complications might not become significant. The major avoidance of concomitant procedures was based on the potential for intraparenchymal pulmonary hemorrhage resulting from the heparin-induced coagulopathy necessary for extracorporeal circulation [ 11. Based on a single death due to intraparenchymal pulmonary hemorrhage, Piehler and colleagues [l]concluded that pulmonary resection on CPB might be associated with increased risk. Bleeding complications occurred in 15.8% of patients in this series. Intraparenchymal pulmonary hemorrhage occurred after heparinization in a patient in whom a wedge biopsy was performed before anticoagulation for a persistent nodular infiltrate of indeterminate etiology. Although it did not result in mortality, it did necessitate a lobectomy for benign disease. A second patient required exploration for bleeding at the staple line after a wedge resection performed on bypass. A third patient died of hemorrhage-related adult respiratory distress syndrome after lobectomy for control of bullae unintentionally violated during dissection of a densely adherent IMA graft. The instance of postoperative sternal bleeding was unrelated to the pulmonary resection. Although statistical analysis showed no correlation between timing of resection with respect to CPB and anticoagulation and to the risk of subsequent complications, the number of patients in this series is again too small to give adequate statistical power. A second concern has been the possible dissemination of infection after resection of infectious nodules, which might be exacerbated by the humoral and cellular immunosuppressive effects of CPB [9]. In this series, there was 1 case of sternal infection from the same organism ( E n terobacter sp) cultured from a left lower lobe infiltrate treated with lobectomy for intraparenchymal pulmonary
294
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
hemorrhage. There was no instance of dissemination of Aspergillus, Histoplasma, or tuberculosis related to the concomitant procedures, nor was there an apparent decrease in survival of patients undergoing resection for malignancy attributable to immunosuppression after bypass [9] and after operation [lo]. Five-year survival was 100% for patients who were discharged after resection for benign disease. The 5-year survival of 40% for patients with primary pulmonary malignancy in this small series compares well with the 26% to 35% in the literature [ll-181. Two patients were in stage I and 3 were in stage 11.
A third concern has been that of limited pulmonary exposure through a median sternotomy [l].Median sternotomy provides adequate exposure for most pulmonary resections [14, 6, 101. It has been advocated by some surgeons as the incision of choice for certain types of pulmonary resection [2-51, especially bilateral pulmonary metastases [ 3 ] and bilateral bullous disease [4]. Exposure for a sleeve left upper lobe resection is difficult, as is exposure of the distal left mainstem bronchus, because of the required retraction of the heart, which can induce hemodynamic instability [lo]. Hilar lymph node dissection through a median sternotomy can be tedious and limited, especially if adhesions are present. Exposure was a problem in 3 patients in the present series. In 1, a posterolateral thoracotomy was required because hilar exposure was limited by dense adhesions from Aspergillus. Concomitant operation was necessary, as both massive hemoptysis and unstable angina were present. In the patient with intrapulmonary hemorrhage, limited exposure to the left lower lobe required extension of the sternotomy into the left fifth interspace. Despite this extension, there was difficulty in exposure because of the previously placed left IMA graft, which was avulsed and required revision. In the third patient, dense adhesions between large left upper lobe bullae, the parietal pleura, and the IMA made exposure of the aortic root for valve replacement difficult. The ensuing large intraoperative blood loss and hemodynamic instability resulted in fatal adult respiratory distress syndrome. In the remaining patients, access to all lobes was possible, whether for wedge resection or lobectomy. Anesthesia techniques were similar to those used in routine cardiac operations. Double-lumen endotracheal tubes were not used. We think this was a tactical error [6]. Although CPB can be useful to facilitate exposure by simultaneous lung deflation, CPB times are thereby extended and morbidity increased. The use of double-lumen endotracheal tubes might obviate the need for prolonged CPB. If a combined procedure is required, consideration to pulmonary resection before IMA grafting should be made to avoid injury to this graft during resection. Median sternotomy causes less deleterious changes in lung volumes, flows, and compliance than does lateral thoracotomy [2, 10, 191 and therefore might be the preferable exposure for patients with limited pulmonary reserve [l].Both incisions result in a 50% or greater loss of vital capacity and peak airway flow on postoperative day 2 [2]. At 1 week postoperatively, peak flow returns to 80%
Ann Thorac Surg 1992;54:289-95
of preoperative values after median sternotomy compared with 64% after thoracotomy, and vital capacity returns to 71% of preoperative levels after median sternotomy compared with 58% for thoracotomy [2]. These facts must be judged in relation to the detrimental effects of CPB on pulmonary function [20-221. Chest wall compliance is decreased more with cardiac procedures than with operations performed without CPB through a similar incision [20]. It is unclear whether the pulmonary complications encountered in this series are definitely attributable to concomitant procedures, as there are no controls in this retrospective review. However, 5 patients required prolonged ventilatory support. Exclusion of the patient with poliomyelitis results in a respiratory complication rate of 21% (4/19 patients). The median postoperative hospitalization was 15 days in this series. Although this is longer than that seen in patients undergoing routine open-heart or pulmonary procedures, it is likely that the hospitalization resulting from a concomitant approach is equivalent to or shorter than the combined hospitalization of staged procedures. In conclusion, concomitant correction of both cardiac and pulmonary conditions is associated with an increase in hemorrhagic and pulmonary complications. Based on the high complication rate with pulmonary resection on CPB, we recommend that pulmonary resection usually be staged sometime after the cardiac operation. If this is not feasible, the pulmonary resection should be performed after reversal of anticoagulation to reduce the risk of bleeding after heparinization. Concomitant procedures should be avoided if possible in the presence of dense pleural or hilar adhesions and when the pulmonary pathology resides in the left lower lobe. Double-lumen endotracheal tubes might facilitate exposure and decrease the need for extended CPB. This study cannot make conclusions regarding the morbidity and mortality of patients undergoing staged procedures. We thank Jackie Dorsey for her energy and dedication in assistance with data collection, patient follow-up, and preparation of
the manuscript.
References 1. Piehler JM, Trastek VF, Pairolero PC, et al. Concomitant cardiac and pulmonary operations. J Thorac Cardiovasc Surg
1985;90:662-7. 2. Cooper JD, Nelems JM, Pearson FG. Extended indications for median sternotomy in patients requiring pulmonary resection. Ann Thorac Surg 1978;26:41>20. 3. Takita H, Merrin C , Didolkar MS, Douglass HO, Edgerton F. The surgical management of multiple lung metastases. Ann Thorac Surg 1977;24:359-64. 4. Mercier C, Page A, Verdant A, Cossette R, Dontigny L, Pelletier LC. Outpatient management of intercostal tube drainage in spontaneous pneumothorax. Ann Thorac Surg 1976;25:163-5 5. Dalton ML, Parker TM, Mistrot I, Bricker DL. Concomitant coronary artery bypass and major noncardiac surgery. J Thorac Cardiovasc Surg 1978;75:621-4. 6. Canver CC, Bhayana JN, Lajos TZ, et al. Pulmonary resection combined with cardiac operations. Ann Thorac Surg 1990;50: 7969.
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
Ann Thorac Surg 1992;54289-95
7. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-61. 8. Grunkemeier GL, Starr A. Actuarial analysis of surgical results: rationale and method. Ann Thorac Surg 1977;24: 404-8. 9. Peters RM, Swain JA. Management of the patient with emphysema, coronary artery disease, and lung cancer. Am J Surg 1982;143:701-5. 10. Meng RL, Jensik RJ, Kittle CF, Faber LP. Median sternotomy for synchronous bilateral pulmonary operations. J Thorac Cardiovasc Surg 1980;80:1-7. 11. Peters RM. The role of limited resection in carcinoma of the lung. Am J Surg 1982;143:70&10. 12. Hoffman TH, Ransdell HT. Comparison of lobectomy and wedge resection for carcinoma of the lung. J Thorac Cardiovasc Surg 1980;79:211-5. 13. Shields TW, Higgins GA Jr. Minimal pulmonary resection in treatment of carcinoma of the lung. Arch Surg 1974;lOS: 420-2. 14. LeRoux BT. Management of bronchial carcinoma by segmental resection. Thorax 1972;27704. 15. Paulson DL, Urschel HC Jr. Selectivity in the surgical treat-
16. 17. 18. 19. 20.
21. 22.
295
ment of bronchogenic carcinoma. J Thorac Cardiovasc Surg 1971;62:554-62. Vincent RG, Takita H, Lane WW, Gutierrez AC, Pickren JW. Surgical therapy of lung cancer. J Thorac Cardiovasc Surg 1976;71:581-91. Weiss W. Operative mortality and five year survival rates in men with bronchogenic carcinoma. Chest 1974;66:48>7. Jensik RJ, Faber LP, Milloy FJ, Monson DO. Segmental resection for lung cancer. A fifteen year experience. J Thorac Cardiovasc Surg 1973;66:56>72. Peters RM, Wellons HA, Htwe TM. Total compliance and work of breathing after thoracotomy. J Thorac Cardiovasc Surg 1969;5734%55. Ellison LT, Duke JF, Ellison RG. Pulmonary compliance following open-heart surgery and its relationship to ventilation and gas exchange. Circulation 1967;35(Suppl 1):217-25. Andersen ND, Ghia J. Pulmonary function, cardiac status, and postoperative course in relation to cardiopulmonary bypass. J Thorac Cardiovasc Surg 1970;59:474-83. Howatt WF, Talner NS, Sloan H, Demuth GR. Pulmonary function changes following repair of heart lesions with the aid of extracorporeal circulation. J Thorac Cardiovasc Surg 1962;43:649-57.
To assess the safety and efficacy of concomitant pulmonary resection and cardiac operation requiring cardiopulmonary bypass, the records of 19 patients were reviewed. Eighteen patients (94.7%) presented with cardiac symptoms and were found to have pulmonary pathology of indeterminate etiology. Pulmonary resections were performed through a median sternotomy in all but 1patient, who underwent posterolateral thoracotomy and right middle lobectomy after repositioning because dense adhesions prevented adequate dissection through the initial incision. A total of 24 resections were performed. Sixteen (66.7%) were performed on cardiopulmonary bypass. Six wedge resections (25.0%) were performed before bypass. Two lobectomies (8.3%) were performed after infusion of protamine sulfate. Nine patients (47.4%) had benign pathology, 7 (36.8%)had primary carcinoma, and 3 (15.8%) had metastatic disease. Bleeding complications occurred in 15.8% of patients (3/19). There was 1 perioperative death (5.3%), which was due to adult respiratory distress syndrome after intraoperative hemorrhage followed lobectomy for bullous disease. Another patient required lateral extension of the sternotomy during an episode of exsanguinating intraparenchymal pul-
monary hemorrhage, which resulted in lobectomy, as well as costochondral and sternal osteomyelitis. A third patient required exploration for bleeding at the staple line. Postoperative complications occurred in 7 patients (36.8%)and were predominantly respiratory (517, 71.4%) ( p = 0.006). The median postoperative hospitalization was 15 days. Although comparison of patients who underwent pulmonary resection during bypass with those who had resection either before heparinization or after protamine infusion showed no significant difference with respect to age, incidence of malignancy, operation performed, complications, postoperative hospitalization, or survival, this was probably due to the small number of patients in the study. Survival correlated only with the diagnosis of malignancy ( p = 0.042). Pulmonary resection performed on cardiopulmonary bypass leads to excessive bleeding and pulmonary complications and perhaps to excessive hospitalization. If concomitant correction of both cardiac and pulmonary conditions must be performed, pulmonary resection should be accomplished after reversal of anticoagulation to prevent excessive bleeding. (Ann Thorac Surg 1992;54:289-95)
P
Concomitant operations might allow prompt correction of both conditions, thereby sparing the patient a second major thoracic procedure with its attendant risks and expense. In the most recent study [6] to address the question of concomitant operations, no pulmonary resection was performed during anticoagulation. The purpose of this study was to assess the safety and efficacy of concomitant pulmonary and cardiac operations with respect to the timing of pulmonary resection in relation to cardiopulmonary bypass (CPB).
ulmonary disease and cardiovascular disease have several common risk factors. When they occur simultaneously in a patient, this creates a complex therapeutic challenge. Patients usually present with cardiac symptoms and are subsequently found to have pulmonary pathology. Less often, patients are found to have major cardiac pathology during evaluation for pulmonary resection [l]. Traditionally, the two diseases have been treated in a staged fashion depending on clinical priority. Reluctance to perform combined procedures has been based on concerns about adequate pulmonary exposure, as well as bleeding complications secondary to the systemic heparinization required for the cardiac portion of the operation [l]. Although median sternotomy has been shown to be safe and efficacious for pulmonary resection [2-61, the effects of systemic heparization are not known. Accepted for publication Jan 9, 1992 Address reprint requests to Dr Schmelzer, 2123 Auburn Ave, Cincinnati, OH 45219.
0 1992 by The Society of Thoracic Surgeons
Material and Methods
Patient Population The clinical records of all patients who underwent concomitant pulmonary and cardiac operations that required CPB from February 1980 through January 1990 were reviewed. The patient population consisted of 19 patients (15 men and 4 women) with a mean age (* the standard 1.8 years (range, 52 to 78 error of the mean) of 63.3
*
0003-4975/92/$5.00
290
Ann Thorac Surg 1992;54:289-95
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
years) (Table 1). Eighteen patients (94.7%)presented with unstable cardiac symptoms (New York Heart Association class I11 or IV) resulting from coronary artery disease (n = 13),valvular disease (n = 4), and left atrial tumor (n = 1). One patient presented with massive hemoptysis secondary to an Aspergillus cavitary lesion, and unstable angina developed during the evaluation. Another patient was scheduled for coronary artery bypass, which was postponed because of an asymptomatic left lower lobe infiltrate. Pulmonary lesions were identifiable in all but 1 patient on preoperative chest roentgenograms. Fourteen patients (73.7%)had a new solitary nodule, 2 (10.5%)had multiple nodules, 1 had an infiltrate, and 1 had a cavitary lesion.
One patient underwent left upper lobe wedge resection before bypass to prevent injury to a planned left IMA graft. Wedge resection of the right upper lobe lesion was later performed on bypass.
Anatomy of Resection
Results
Twenty-four pulmonary resections were performed in the 19 patients (Table 2). Standard surgical techniques were employed for all resections, and mechanical staplers were used for all bronchial closures and all wedge resections. The 14 wedge resections were performed in 12 patients; 1 later required lobectomy to control intraparenchymal hemorrhage. A second patient underwent lobectomy after wedge resection documented malignancy. Six patients underwent lobectomy as the primary pulmonary procedure. One patient underwent left upper lobectomy and right upper lobe wedge resection. Seventeen operations (89.5%) in the 19 patients were performed through a median sternotomy alone. One patient, who underwent wedge biopsy of a persistent left lower lobe infiltrate before bypass, required lateral extension of the sternotomy into the fifth interspace to control exsanguinating intraparenchymal hemorrhage, which occurred after heparinization. In a second patient, dense adhesions around a middle lobe Aspergillus cavitary lesion prohibited hilar dissection. Because he had presented with massive hemoptysis, the patient was repositioned and a posterolateral thoracotomy was performed after completion of coronary artery bypass grafting.
There was 1 perioperative death (within 30 days of operation or within the same hospitalization) (5.3%), which occurred in a 72-year-old man who underwent aortic valve replacement and mitral valve repair. In the dissection of a densely adherent IMA graft (previous operation), violation of numerous left upper lobe bullae occurred, necessitating lobectomy, which was performed after CPB. Despite reversal of anticoagulation, severe diffuse bleeding with hemodynamic instability ensued. The patient died on postoperative day 2 of adult respiratory distress syndrome affecting the right lung. Most of the major complications were pulmonary (p = 0.006). Six additional patients (31.6%) had postoperative complications related to pulmonary resection. One patient required reoperation for bleeding at the staple line after wedge resection performed on CPB. Five patients (26.3%) required prolonged ventilatory support. Four required 4 to 7 days of initial respiratory support, 1 of whom had sequelae from previous poliomyelitis, and the fifth required reintubation on postoperative day 6 for an additional 6 days. Three of them underwent lobectomy, 1 had a segmentectomy, and 1, a wedge resection. Two patients (10.5%), 1 with lobectomy and 1 with wedge resection, had pneumonia postoperatively. There were no prolonged air leaks. One patient required chest tubes for 7 days because of excessive drainage. In the patient who underwent lobectomy for intraparenchymal hemorrhage, osteochondritis of the fifth rib and sternal osteomyelitis developed and myocutaneous flap reconstruction was performed. Other complications are listed in Table 1. All atrial (n = 5) and ventricular (n = 4) dysrhythmias responded pharmacologically. The median postoperative hospitalization was 15 days (mean period, 19.1 2 2.7 days; range, 8 to 51 days). The occurrence of major postoperative, complications was not significantly related to resection on CPB, operation performed, or age of the patient. Nine patients (47.4%)had benign pulmonary disease, 5 patients had bronchogenic carcinoma (26.3%), 2 had carcinoid (10.5%), and 3 (15.8%) had metastatic processes (Table 4). Four patients with primary adenocarcinoma underwent lobectomy for cure. The fifth primary malignancy was treated with wedge resection. The 3 patients with pulmonary metastases were a 63-year-old woman
Timing of Resection Pulmonary resections were performed at various times in relation to CPB (Table 3 ) . Three wedge resections were accomplished in 2 patients before bypass and anticoagulation. In the first patient, a right upper lobe wedge resection was performed before placement of bilateral internal mammary artery (IMA) grafts, and in the second, bilateral wedge resections were performed before resection of a left atrial leiomyosarcoma to document histology. Six wedge resections, five lobectomies, and one segmentectomy were performed after the cardiac procedure while the patient was still on CPB. A large, nondeflatable right upper lobe bullous region was resected before bypass, and the patient then underwent left lower lobectomy after grafting while on bypass to facilitate exposure. One left lower lobectomy was emergently performed on bypass to control intraoperative hemorrhage that ensued after an initial wedge resection performed before anticoagulation. One left upper lobe wedge resection was performed after vein grafting but before grafting of the IMA.
Statistical Analysis
Discrete variables were analyzed by x2 or Fisher’s exact tests, and one-way analysis of variance was used to examine differences between continuous variables. Survival was determined by actuarial analysis using the methods of Kaplan and Meier [7, 81 with the day of operation as the starting time. Differences resulting in a p value of less than 0.05 were considered significant.
CVG x 3, LLL WR CVG x 4, RUL WR, RU lobectomy LLL WR, RUL WR, L atrial tumor CVG x 5, RM lobectomy MVR, RLL WR
M
M F M
F M
M
M
M M
56
61 59 61
52 74
67
58
68
69
58
77 78
62
72
54
4
5 6 7
8 9
10
11
12
13
14
15 16
17
18
19
RUL WR, Bilateral IMA LUL WR, CVG x 2, LIMA x 1, RUL WR CVG x 2, LLL WR, LIMA x 1 MVR, AVR, LU lobectomy CVG x 4, LIMA x 1, LUL WR
LLL WR, CVG x 5, LL lobectomy
CVG x 5, LLL WR
14
16
... ... 12 26 14 10 17
28 14
8
... ... ...
... ...
... 22
...
...
...
15
Atrial fibrillatiodventricular tachycardia, acalculous cholecystitis
Death, ARDS
Atrial fibrillation, pericarditis
... Atrial fibrillation
Reoperation for staple line bleeding Intraop pulmonary hemorrhage, IMA disruption, osteomyelitis
Ventricular ectopy, herpes zoster Atrial fibrillation
Pericarditis Pericarditis, ventricular ectopy Pneumonia
...
Atrial fibrillation Pericarditis
10
...
1
... UGI bleeding, ventricular ectopy Reoperation for sternal bleeding Pneumonia, high chest-tube output
Complications
10
51
... ... ...
18
...
... ...
...
6
17 41
(4
Postop Hospitalization
...
Reintubation Day
1
1 1
7
2
1
1
4
1 3
3 1 7
4
1
1 1/12
Extubation Day
adenocarcinoma
? Metastatic
Bullous disease
Carcinoid
His toplasma Inflammatory
Undifferentiated large cell carcinoma, stage I1 Pneumonia
Carcinoid
Aspergillus
L atrial leiomyosarcoma
Histoplasma Adenocarcinoma, stage I
Bullous disease (R), adenocarcinoma (L), stage I Adenocarcinoma, stage I1 Histoplasma Hamartoma
Tuberculosis
Spindle cell sarcoma Adenocarcinoma, stage I1
Pathology
AWD, 10 mo; 7 RUL metastasis
...
DF, 47 mo DF, 47 mo; coronary disease DF, 29 mo
DF, 55 mo; CHF
DDF, 65 mo; redo MVIVAVR DOD, 3.5 mo
DF, 69 mo
DOD, 18 mo
DOD, 21 mo DF, 93 mo DF, 90 mo; coronary reoperation X 2 DF, 86 mo DF, 83 mo; CHF
DF, 94 mo
DF. 127 mo
DOD, 89 mo DOD, 13 mo
Follow-up
ARDS = adult respiratory distress syndrome; AVR = aortic valve replacement; AWD = alive with disease; CHF = congestive heart failure; CVG = coronary venous grafting; DDF = died DF = disease-free; DOD = died of disease; IMA = internal mammary artery graft; LIMA = left internal mammary artery praft; LL = left lower; LLL = left lower lobe; LU disease-free; = left upper; LUL = left upper lobe; MVR = mitral valve replacement repair; RL = right lower; RLL = right lower lobe; RM = right middle; RML = right middle lobe; RU = right upper; RUL = right upper lobe; UGI = upper gastrointestinal; WR = wedge resection.
M
M
M
M
F
M
AVR, RU lobectomy CVG X 5, RUL WR CVG x 3, MVR, LLL segmentectomy
M
53
3
CVG x 4, RL lobectomy CVG x82, RM lobectomy Redo CVG x 3, LL lobectomy RUL WR, CVG x 2, LU lobectomy
F M
63 61
1 2
Operations
(Y)
Sex
Age
Patient No.
Table 1. Summary of Data on Patient Population
292
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
Ann Thorac Surg 1992;54:289-95
Table 2. Type and Location of 24 Pulmonary Resections in 19 Patients Undergoing Concomitant Cardiac and Pulmonary Procedures' Location
Wedge
Resection Segmentectomy Lobectomy
Table 4. Pulmonary Pathology in 19 Patients Undergoing Concomitant Cardiac and Pulmonaw Procedures" No. of Patients
Pathology Total
~~
Benign
Right
Upper
Hamartoma
1 (5.3)
6
0
2
8 (33.3)
Histoplasma
3 (15.8)
Middle lobe Lower lobe Left Upper lobe Lower
0
0
2
2 (8.3)
Tuberculosis Aspergillus
1
0
1
2 (8.3)
Pneumonia Bullae
1 (5.3) l(5.3) l(5.3) 2b (10.5) l(5.3)
lobe
lobe Total
a
2
0
2
4 (16.7)
5
1
2
8 (33.3)
1 (4.2)
9 (37.5)
14 (58.3)
Inflammatory Malignant Primary
Adenocarcinoma Undifferentiated large cell Metastatic Spindle cell sarcoma Atrial leiomyosarcoma Adenocarcinoma
24
Numbers in parentheses are percentages.
with a solitary right lower lobe metastasis from a spindle cell sarcoma resected from the right arm 20 years earlier, a 67-year-old man with a left atrial leiomyosarcoma and two pulmonary metastases, and a 54-year-old man with poorly differentiated adenocarcinoma of probable gastrointestinal origin. Follow-up, conducted by telephone interview and review of all medical records, was current for all patients and averaged 57.8 8.5 months (range, 3.5 to 127 months). Of the 8 patients with only benign disease who survived operation, all are currently alive 47 to 127 months after operation without evidence of dissemination caused by concomitant operation. Of the 5 patients with bronchogenic carcinoma, 2 are currently alive without disease at 83 and 94 months, and 3 died of disease at 3.5, 13, and 21 months after operation. One patient with carcinoid is currently alive without evidence of recurrence 29 months after operation. The second died without evidence of recurrence 65 months after resection after aortic and repeat mitral valve replacement. Of the 3 patients with metastatic disease, 1 is currently alive with recurrence 10 months after operation, and 2 died of disease at 18 and 89 months postoperatively.
*
4b (21.1) 2 (10.5) l(5.3)
Carcinoid
1 (5.3) l(5.3) 1 (5.3)
Numbers in parentheses are percentages. had both bullous disease and adenocawinoma.
a
One patient (patient 4)
The observed survival rate for all patients undergoing combined procedures was 73.7% at 5 years and 68.4% at 10 years (Fig 1). Survival was not affected by age, occurrence of major postoperative complications, resection on CPB, or operation received. Survival did correspond to the diagnosis of malignancy ( p = 0.042). For all patients with cancer, the 5- and 10-year survival rates were 60.0% and 50.0%, respectively. Patients with bronchogenic carcinoma fared worse than those with metastatic disease; 5-year survival rates were 40.0% and 50.0%, respectively ( p = not significant).
Comment Concomitant pulmonary and cardiac operations have not traditionally been performed. The most clinically important disease was addressed first, followed by a second operation at a later date for the remaining problem.
Table 3. Timing of Pulmonary Resection in Relation to Cardiopulmonary Bypass in 19 Patients Undergoing Concomitant Cardiac and Pulmonary Procedures' Timing Before bypass Before bypass and during bypass after cardiac procedure On bypass during cardiac procedure On bypass after cardiac procedure After bypass Total Numbers in parentheses are percentages.
Wedge Resection
Multiple Wedge Resection
Wedge Resection and Lobectomy
Lobectomy
Segmentectomy
Total
1 0
1 1
2
0 0
0 0
2 (10.5) 3 (15.8)
1 5 0
0 0 0
0 1 0
0 4 2
0 1 0
l(5.3) 11 (57.9) 2 (10.5)
7 (36.8)
2 (10.5)
3 (15.8)
6 (31.6)
1 (5.3)
19
0
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
Ann Thorac Surg 1Y92:54:28Y-Y5
293
Fig 1 . Observed percent survival of patients undergoing concomitant cardiac and pulmonary procedures. Bronchogenic carcinoma (CAI excludes the diagnosis of carcinoid. Metastatic disease includes adenocarcinoma ( n = Z), spindle cell sarcoma (n = l ) , and atrial leiomyosarcoma ( n = I).
Percent ALLPATIENTS
+ ALLBENIGN
-113
20-
0
I
.
.
.
.
.
-
Prolongation of the operative time for concomitant procedures was believed to increase the operative mortality. The second procedure was delayed for 4 to 6 weeks to allow the patient time to gain weight and muscle mass [9]. The traditional approach to bilateral pulmonary disease has been staged lateral thoracotomies, 7 to 90 days apart, depending on the disease progression and the condition of the patient [lo]. Concomitant procedures would be attractive if both could be performed through a median sternotomy and if both disease processes could be managed safely. In this study, 78.9% of patients (15/19 patients) underwent pulmonary resection during CPB. The only other reports of concomitant operations are by Piehler and colleagues [l]and Canver and associates [6]. The majority of the 43 resections in the former study were performed either before (39.5%) or after (48.8%) bypass; only five (11.6%) were performed on bypass. Similarly, Canver’s group carried out pulmonary resection before heparinization in 57.1% of patients and after protamine reversal in 42.9%. In contrast, the majority of pulmonary resections (63.2%) in this review were performed under full anticoagulation. An additional 15.8% of operations consisted of two resections, one of which was performed before bypass and the other on bypass after the cardiac procedure. The remainder were performed either before (10.5%) or after (10.5%) anticoagulation. Some patients required a combined approach because of the nature of their disease. The patient with Aspergillus, for example, had life-threatening hemoptysis and severe unstable angina. Another patient confronted the surgeon with left upper lobe bullous disease densely adherent to a previously placed IMA graft that required mobilization. Another, with left upper lobe adenocarcinoma, required resection of nondeflatable bullous disease for exposure. One patient was seen with bilateral pulmonary metastases from a left atrial leiomyosarcoma. It could be argued that the remaining patients should have been operated on in a staged fashion, but data supporting this concept have been lacking. Nevertheless, this group of patients allows
I
.
ALLCANCER BRONCHOOENICCA YETASATICCA
-
I
comparison of both the advantages and the disadvantages of pulmonary resection before, during, and after bypass. As statistical comparison is difficult in this small group of patients because of the lack of statistical power, some potentially large differences between resection on CPB and survival, hospitalization, and complications might not become significant. The major avoidance of concomitant procedures was based on the potential for intraparenchymal pulmonary hemorrhage resulting from the heparin-induced coagulopathy necessary for extracorporeal circulation [ 11. Based on a single death due to intraparenchymal pulmonary hemorrhage, Piehler and colleagues [l]concluded that pulmonary resection on CPB might be associated with increased risk. Bleeding complications occurred in 15.8% of patients in this series. Intraparenchymal pulmonary hemorrhage occurred after heparinization in a patient in whom a wedge biopsy was performed before anticoagulation for a persistent nodular infiltrate of indeterminate etiology. Although it did not result in mortality, it did necessitate a lobectomy for benign disease. A second patient required exploration for bleeding at the staple line after a wedge resection performed on bypass. A third patient died of hemorrhage-related adult respiratory distress syndrome after lobectomy for control of bullae unintentionally violated during dissection of a densely adherent IMA graft. The instance of postoperative sternal bleeding was unrelated to the pulmonary resection. Although statistical analysis showed no correlation between timing of resection with respect to CPB and anticoagulation and to the risk of subsequent complications, the number of patients in this series is again too small to give adequate statistical power. A second concern has been the possible dissemination of infection after resection of infectious nodules, which might be exacerbated by the humoral and cellular immunosuppressive effects of CPB [9]. In this series, there was 1 case of sternal infection from the same organism ( E n terobacter sp) cultured from a left lower lobe infiltrate treated with lobectomy for intraparenchymal pulmonary
294
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
hemorrhage. There was no instance of dissemination of Aspergillus, Histoplasma, or tuberculosis related to the concomitant procedures, nor was there an apparent decrease in survival of patients undergoing resection for malignancy attributable to immunosuppression after bypass [9] and after operation [lo]. Five-year survival was 100% for patients who were discharged after resection for benign disease. The 5-year survival of 40% for patients with primary pulmonary malignancy in this small series compares well with the 26% to 35% in the literature [ll-181. Two patients were in stage I and 3 were in stage 11.
A third concern has been that of limited pulmonary exposure through a median sternotomy [l].Median sternotomy provides adequate exposure for most pulmonary resections [14, 6, 101. It has been advocated by some surgeons as the incision of choice for certain types of pulmonary resection [2-51, especially bilateral pulmonary metastases [ 3 ] and bilateral bullous disease [4]. Exposure for a sleeve left upper lobe resection is difficult, as is exposure of the distal left mainstem bronchus, because of the required retraction of the heart, which can induce hemodynamic instability [lo]. Hilar lymph node dissection through a median sternotomy can be tedious and limited, especially if adhesions are present. Exposure was a problem in 3 patients in the present series. In 1, a posterolateral thoracotomy was required because hilar exposure was limited by dense adhesions from Aspergillus. Concomitant operation was necessary, as both massive hemoptysis and unstable angina were present. In the patient with intrapulmonary hemorrhage, limited exposure to the left lower lobe required extension of the sternotomy into the left fifth interspace. Despite this extension, there was difficulty in exposure because of the previously placed left IMA graft, which was avulsed and required revision. In the third patient, dense adhesions between large left upper lobe bullae, the parietal pleura, and the IMA made exposure of the aortic root for valve replacement difficult. The ensuing large intraoperative blood loss and hemodynamic instability resulted in fatal adult respiratory distress syndrome. In the remaining patients, access to all lobes was possible, whether for wedge resection or lobectomy. Anesthesia techniques were similar to those used in routine cardiac operations. Double-lumen endotracheal tubes were not used. We think this was a tactical error [6]. Although CPB can be useful to facilitate exposure by simultaneous lung deflation, CPB times are thereby extended and morbidity increased. The use of double-lumen endotracheal tubes might obviate the need for prolonged CPB. If a combined procedure is required, consideration to pulmonary resection before IMA grafting should be made to avoid injury to this graft during resection. Median sternotomy causes less deleterious changes in lung volumes, flows, and compliance than does lateral thoracotomy [2, 10, 191 and therefore might be the preferable exposure for patients with limited pulmonary reserve [l].Both incisions result in a 50% or greater loss of vital capacity and peak airway flow on postoperative day 2 [2]. At 1 week postoperatively, peak flow returns to 80%
Ann Thorac Surg 1992;54:289-95
of preoperative values after median sternotomy compared with 64% after thoracotomy, and vital capacity returns to 71% of preoperative levels after median sternotomy compared with 58% for thoracotomy [2]. These facts must be judged in relation to the detrimental effects of CPB on pulmonary function [20-221. Chest wall compliance is decreased more with cardiac procedures than with operations performed without CPB through a similar incision [20]. It is unclear whether the pulmonary complications encountered in this series are definitely attributable to concomitant procedures, as there are no controls in this retrospective review. However, 5 patients required prolonged ventilatory support. Exclusion of the patient with poliomyelitis results in a respiratory complication rate of 21% (4/19 patients). The median postoperative hospitalization was 15 days in this series. Although this is longer than that seen in patients undergoing routine open-heart or pulmonary procedures, it is likely that the hospitalization resulting from a concomitant approach is equivalent to or shorter than the combined hospitalization of staged procedures. In conclusion, concomitant correction of both cardiac and pulmonary conditions is associated with an increase in hemorrhagic and pulmonary complications. Based on the high complication rate with pulmonary resection on CPB, we recommend that pulmonary resection usually be staged sometime after the cardiac operation. If this is not feasible, the pulmonary resection should be performed after reversal of anticoagulation to reduce the risk of bleeding after heparinization. Concomitant procedures should be avoided if possible in the presence of dense pleural or hilar adhesions and when the pulmonary pathology resides in the left lower lobe. Double-lumen endotracheal tubes might facilitate exposure and decrease the need for extended CPB. This study cannot make conclusions regarding the morbidity and mortality of patients undergoing staged procedures. We thank Jackie Dorsey for her energy and dedication in assistance with data collection, patient follow-up, and preparation of
the manuscript.
References 1. Piehler JM, Trastek VF, Pairolero PC, et al. Concomitant cardiac and pulmonary operations. J Thorac Cardiovasc Surg
1985;90:662-7. 2. Cooper JD, Nelems JM, Pearson FG. Extended indications for median sternotomy in patients requiring pulmonary resection. Ann Thorac Surg 1978;26:41>20. 3. Takita H, Merrin C , Didolkar MS, Douglass HO, Edgerton F. The surgical management of multiple lung metastases. Ann Thorac Surg 1977;24:359-64. 4. Mercier C, Page A, Verdant A, Cossette R, Dontigny L, Pelletier LC. Outpatient management of intercostal tube drainage in spontaneous pneumothorax. Ann Thorac Surg 1976;25:163-5 5. Dalton ML, Parker TM, Mistrot I, Bricker DL. Concomitant coronary artery bypass and major noncardiac surgery. J Thorac Cardiovasc Surg 1978;75:621-4. 6. Canver CC, Bhayana JN, Lajos TZ, et al. Pulmonary resection combined with cardiac operations. Ann Thorac Surg 1990;50: 7969.
ULICNY ET AL CARDIAC AND PULMONARY OPERATION
Ann Thorac Surg 1992;54289-95
7. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-61. 8. Grunkemeier GL, Starr A. Actuarial analysis of surgical results: rationale and method. Ann Thorac Surg 1977;24: 404-8. 9. Peters RM, Swain JA. Management of the patient with emphysema, coronary artery disease, and lung cancer. Am J Surg 1982;143:701-5. 10. Meng RL, Jensik RJ, Kittle CF, Faber LP. Median sternotomy for synchronous bilateral pulmonary operations. J Thorac Cardiovasc Surg 1980;80:1-7. 11. Peters RM. The role of limited resection in carcinoma of the lung. Am J Surg 1982;143:70&10. 12. Hoffman TH, Ransdell HT. Comparison of lobectomy and wedge resection for carcinoma of the lung. J Thorac Cardiovasc Surg 1980;79:211-5. 13. Shields TW, Higgins GA Jr. Minimal pulmonary resection in treatment of carcinoma of the lung. Arch Surg 1974;lOS: 420-2. 14. LeRoux BT. Management of bronchial carcinoma by segmental resection. Thorax 1972;27704. 15. Paulson DL, Urschel HC Jr. Selectivity in the surgical treat-
16. 17. 18. 19. 20.
21. 22.
295
ment of bronchogenic carcinoma. J Thorac Cardiovasc Surg 1971;62:554-62. Vincent RG, Takita H, Lane WW, Gutierrez AC, Pickren JW. Surgical therapy of lung cancer. J Thorac Cardiovasc Surg 1976;71:581-91. Weiss W. Operative mortality and five year survival rates in men with bronchogenic carcinoma. Chest 1974;66:48>7. Jensik RJ, Faber LP, Milloy FJ, Monson DO. Segmental resection for lung cancer. A fifteen year experience. J Thorac Cardiovasc Surg 1973;66:56>72. Peters RM, Wellons HA, Htwe TM. Total compliance and work of breathing after thoracotomy. J Thorac Cardiovasc Surg 1969;5734%55. Ellison LT, Duke JF, Ellison RG. Pulmonary compliance following open-heart surgery and its relationship to ventilation and gas exchange. Circulation 1967;35(Suppl 1):217-25. Andersen ND, Ghia J. Pulmonary function, cardiac status, and postoperative course in relation to cardiopulmonary bypass. J Thorac Cardiovasc Surg 1970;59:474-83. Howatt WF, Talner NS, Sloan H, Demuth GR. Pulmonary function changes following repair of heart lesions with the aid of extracorporeal circulation. J Thorac Cardiovasc Surg 1962;43:649-57.