- Email: [email protected]
Cardiac operation with associated pulmonary resection
Cardiac operation with associated pulmonary resection From February 1988 to May 1992, a total of 11 patients aged 52 to 81 years underwent concomitant cardiac operation and pulmonary resection for bronchogenic carcinoma at our institution. All pulmonary lesions were incidental findings on preoperative chest x-ray films. Diagnosis was obtained in six patients before resection. The operation was performed through a midline sternotomy with all patients requiring cardiopulmonary bypass. Pulmonary procedures included two wedge resections, seven lobectomies, and two double lobectomies. Seven patients underwent lobectomy while supported by bypass, with the lungs collapsed, during rewarming. Total bypass time for these patients averaged 143 minutes. Pathologic examination showed all lesions to be non-small-cell malignant tumors; four adenocarcinomas, four squamous cell carcinomas, two bronchoalveolar carcinomas, and one undifferentiated carcinoma. Nine were stage I and two were stage II. One of the wedge resections showed malignant disease involving the surgical margin that later required completion lobectomy. There were no operative deaths and no major postoperative complications. Postoperative hospital stays ranged from 6 to 17 days (mean 10 days) except for one patient who required a prolonged hospitalization because of a complication after thoracentesis on the side opposite the pulmonary resection. Concomitant cardiac operations with lobectomy can be safely performed during cardiopulmonary bypass without significantly prolonging pump time. Our observations suggest that concomitant cardiac surgery with pulmonary resection is a safe and effective technique with minimal morbidity and short hospital stay. (J THoRAe CARDIOVASC SURG 1993;105:912-7)
Taro Yokoyama, MD, PhD, Marvin J. Derrick, MD, and Anthony W. Lee, MD, Los Angeles and Burbank. Calif.
I
t is not unusual to have both surgical disease of the heart and pulmonary neoplastic disease simultaneously. In the past, management of these patients presented a diagnostic as well as a therapeutic problem. Resecting the pulmonary lesion without addressing the cardiac problem posed a significant risk to the patient. 1-3 Increasing experience suggests that both problems can be addressed during the same operation. Several reports in the literature have suggested that concomitant cardiac operations and noncardiac proceFrom thePacific Cardiothoracic Surgery Group, St. Vincent Medical Center, Los Angeles, Calif., and Saint Joseph Medical Center, Burbank, Calif. Read at the Eighteenth Annual Meeting of The Western Thoracic Surgical Association, Kauai, Hawaii, June 24-27, 1992. Address for reprints: Taro Yokoyama, MD, PhD, 201 S.Alvarado St., Suite 626, Los Angeles, CA 90057. Copyright © 1993 byMosby-Year Book, Inc. 0022-5223/93 $1.00 +.10
912
12/6/44677
dures can be successfully and safely performed.v" despite the risk of bleeding caused by altered coagulation mechanisms. The combined procedure offers distinct advantages over operations done in a staged fashion. One obvious advantage of the combined procedure for the patient is elimination of the pain and risk associated with a second major operation. This report presents our experience with concomitant cardiac operations and pulmonary resection in II patients with bronchogenic carcinoma. Patients From February 1988 to May 1992, a totalof 2909 adultcardiacsurgical procedures necessitating cardiopulmonary bypass were performed at Saint Joseph Hospital in Burbank andSaint Vincent Hospital in LosAngeles. Duringthat period II patients underwent combined pulmonary resection of a bronchogenic carcinoma with a cardiac procedure that necessitated cardiopulmonary bypass. Excluded were patients who underwent concomitant pulmonary resections for benign disease or iatrogenic intraoperative pulmonary complications.
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 5
Yokoyama, Derrick, Lee
Table I. Cardiac procedures performed with concomitant pulmonary resection Procedure
CABG Redo CABG AVR AVR and CABG Total
No. of patients 7
2 I I II
9 13
Table II. Lung disease
Percent 64 18 9 9 100
Disease
No. of patients
Percent
Adenocarcinoma Squamous cell carcinoma Bronchoalveolar U ndilferentiated Total
4 4 2 I II
36 36 18 10 100
CARG. Coronary artery bypass grafting; A YR. aortic valve replacement.
The charts on each of these patients were reviewed to ascertain the operation performed, cell type, TNM classification, stage, timing of pulmonary resection in relation to heparinization, bypass time, and hospital stay. Follow-up was done by contacting the patient and his or her primary physician by telephone to ascertain the postoperative status. The ages of the patients ranged from 52 to 81 years (mean age 70.4 years). Eight were men and three were women. All patients had symptomatic cardiac disease and were in New York Heart Association class III or IV at the time of the operation. Nine patients had coronary artery disease and two had valvular heart disease (Table I). Patients with coronary artery disease required from one to five bypasses. All procedures were performed on an elective basis. In each patient an asymptomatic pulmonary lesion was found on routine preoperative chest x-ray films. The lesion was a nodule in seven patients and an infiltrate in four patients. A preoperative diagnosis of bronchogenic carcinoma was made in six patients before the operation, either by percutaneous needle biopsy in five patients or by broncoscopic biopsy in one. The lesions in the remaining five patients were highly suggestive of carcinoma. A computed tomographic scan was obtained in all patients before the operation. No evidence of mediastinal adenopathy was identified. Pulmonary procedures performed included two wedge resections, seven lobectomies, and two bilobectomies. Extensive mediastinal and hilar lymph node sampling was done at the time of the operation. All of the procedures were performed through a median sternotomy. A double-lumen endotracheal tube was used in each case. Standard pulmonary techniques were used for all resections and a mechanical stapler was used for all bronchial closures. Wedge resections were performed before bypass prior to heparinization by the standard technique with a GIA stapler (Auto Suture Company Division, United States Surgical Corporation, Norwalk, Conn.). Patients without a preoperative diagnosis of bronchogenic carcinoma had intraoperative frozen section confirmation of the neoplasm before lobectomy. Dissection of the hilum for lobar resection was performed before bypass prior to heparinization. In a majority of the patients, the lobe was removed at the conclusion of the cardiac procedure while the patient was still supported by bypass with the lung collapsed during rewarming.
Results There were no operative deaths among the II patients in our study. In addition, there were no major postoperative complications related to the cardiac procedure or
pulmonary resection. Postoperative stay ranged from 6 to 17 days (mean 10 days), except for one patient who remained hospitalized for 29 days. In this patient a hemothorax developed after thoracentesis on the side opposite the pulmonary resection. Tube thoracostomy failed to adequately drain the blood, and a thoracotomy with decortication was eventually required. Another patient, who had a double lobectomy, remained hospitalized for 17 days primarily because of repeated collapse of the remaining right lower lobe as a result of mucous plugging. Treatment with a bronchoscope was required on two separate occasions. A third patient was hospitalized for 16 days primarily because of severe postoperative bronchospasm, which eventually responded to vigorous medical therapy. Two patients had transient atrial fibrillation after the operation; both had lobectomies. All patients were discharged home in normal sinus rhythm. One patient had postoperative periods of sinus arrest with junctional escape after aortic valve replacement. Because of this and in conjunction with her preoperative evidence of conduction system disease, a permanent pacemaker was placed. None ofthe patients had prolonged air leaks, excessivechest tube drainage, pleural space infections, or wound infections. A small air pocket developed posteriorly after chest tube removal in one patient who had left upper lobe resection. The patient refused aspiration of this pocket. On postoperative follow-up, the small pocket of air was completely reabsorbed. Pathologic examination showed a variety of malignancies (Table II). There were four adenocarcinomas, four squamous cell carcinomas, two bronchoalveolar cell carcinomas, and one undifferentiated carcinoma. Nine of these malignant tumors were stage I and two were stage II. Mediastinal lymph node sampling was performed in each case and no evidence of N2 disease was found. Pulmonary resection was believed to be appropriate for the pathologic condition in all patients studied. Pulmonary procedures included two wedge resections, seven lobectomies, and two bilobectomies. One patient, who had a wedge resection performed for a T1 NO MO adenocarcinoma, was found to have evidence of cancer close to the margin on final pathologic analysis. She underwent
914
The Journal of Thoracic and Cardiovascular Surgery May 1993
Yokoyama, Derrick, Lee
completion lobectomy on a separate hospital admission. No evidence of cancer was found in the remaining lung or evidence of hilar lymph node disease. The only patient who underwent a definitive wedge resection was a 79year-old patient with a small coin lesion in the right lower lobe. Results of initial frozen section analysis were reported as carcinoma, probably metastatic versus carcinoid. The lesion was peripheral and measured 13 mm in greatest diameter. A postoperative metastatic work-up showed no abnormalities. The final pathology report indicated an undifferentiated carcinoma with good margins. Studies have suggested that peripherally located TI NO MO lesions less than 2 em in diameter have a better prognosis and may be treated adequately with conservative resection'? This combined with the patient's age and general debilitation suggested that no further surgical intervention was necessary. In seven of the nine patients who underwent lobectomy, resection was performed at the conclusion of the cardiac operation during rewarming when the patient was still supported by cardiopulmonary bypass. Individual patient bypass time ranged from 88 to 225 minutes. Average bypass time for the seven patients who underwent lobar resection on bypass was 143 minutes. Because much of the hilar dissection was performed before bypass (before heparinization), final ligation of the pulmonary vessels with stapling of the interlobar fissure and bronchus added little to the total pump time. On bypass, with the heart decompressed and lungs collapsed, exposure of the hilum was greatly facilitated. Bleeding was not found to be a significant problem during heparinization so long as careful dissection was undertaken. Complete follow-up of all patients ranged from I to 22 months (average 12.1 months). Four patients are presently alive and free of any evidence of malignant disease at postoperative intervals of 1 to 18 months. A 76-yearold man who underwent right middle and lower lobectomy for a T2 NO MO squamous cell carcinoma has recurrence locally in the chest 19 months after the operation. He is presently undergoing radiation therapy for extensive disease involving the remaining right upper lobe and trachea. There was one death unrelated to cardiac or pulmonary disease. This occurred in a 69-year-old patient with stage II squamous cell carcinoma who died of bulbar palsy 21 months after the operation. There was no evidence of recurrence of his cancer. One patient had a sudden cardiac death at 5 months. He was also free of any recurrent lung cancer. Four patients have died of recurrent metastatic lung cancer 5 to 22 months after resection. All of these patients had T I NO MO disease that included bronchoalveolar cell carcinoma, adenocarcinoma, squamous cell carcinoma, and undifferentiated carcinoma.
Pulmonary procedures on these patients included two lobectomies, one bilobectomy, and one wedge resection.
Discussion In selected patients, median sternotomy has proved to be a reasonable alternative to posterolateral or anterior thoracotomy.s-" It provides excellent exposure for the resection of metastatic tumors and also has advantages in the approach to bilateral bullous disease.!? In addition, several reports have advocated sternotomy for the treatment of primary pulmonary neoplasms.f- 9 Because it is not unusual to have both concomitant surgical disease of the heart and pulmonary neoplastic disease, it would appear logical to consider treating the two problems during the same operation. Traditionally, pulmonary resection has not been routinely performed with cardiac procedures that require extracorporeal circulation, primarily because of a concern for intrapulmonary bleeding, a risk that comes from heparinization and from altered coagulation mechanisms caused by cardiopulmonary bypass. Intraparenchymal pulmonary hemorrhage after wedge resection during cardiopulmonary bypass has led to acute respiratory insufficiency and death.P Another concern has been the spread of local infection after resection of fungal infiltrates or granuloma, possibly the result of altered immune mechanisms related to cardiopulmonary bypass. I I Finally, there is the technical concern over whether adequate mediastinal lymph node sampling for staging can be performed by this approach. Advantages of combined cardiac and pulmonary operations outweigh such concerns. Patients who undergo the procedure avoid the pain and risk associated with a staged operation. The patients are also spared the cost of a second hospitalization. Furthermore, a combined procedure allows expeditious treatment of the pulmonary neoplasm, a factor that could be important in tumors with rapid doubling times. Median sternotomy appears to be well tolerated by patients. Cooper, Nelems, and Pearson'? have demonstrated faster recovery of pulmonary function in patients who had a median sternotomy versus a standard posterolateral thoracotomy. Postoperative pain has also been shown to be reduced.f This can be an important factor in patients with borderline pulmonary function, a situation frequently seen in patients with bronchogenic carcinoma. The decision to proceed with a combined procedure must be weighed carefully with each patient. Factors to consider include pulmonary reserve, age, presence and degree of left ventricular dysfunction, bypass time, and functional status. Technical considerations must also be taken into account. Large central tumors that require
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 5
pneumonectomy, superior sulcus tumors, extensive adhesions, or tumors with chest wall involvement should probably be resected at a later time in a staged fashion. This does not preclude wedge resection or mediastinal lymph node sampling, which may alter the prognosis and determine whether the patient has resectable disease. Our technique includes using a double-lumen endotracheal tube. We believe that a critical factor necessary in performing a satisfactory pulmonary resection through a median sternotomy is selective unilateral ventilation. Deflation of the lung allows an unobstructed view of the hilum and mediastinum and a more expeditious dissection. After the chest is opened, the lesion is inspected and technical considerations for resection are considered. Division of the inferior pulmonary ligament will allow access to more remote areas of the lung. A wedge resection or mediastinal lymph node dissection can be performed at this time. Frozen section analysis of these specimens may alter the proposed treatment plan. If concomitant resection is indicated, then dissection of the hilum is undertaken with identification of the pulmonary artery branches and pulmonary vein. If exposure is good, then the lobar resection can be completed. Otherwise, the patient is heparinized and supported by cardiopulmonary bypass with resection being performed at the conclusion of the cardiac procedure during rewarming. Exposure is optimized with the lung deflated and the heart decompressed during bypass. If hilar dissection has been performed before bypass, the actual pulmonary resection during rewarming can take anywhere from 10 to 20 minutes. If at any time during the initial mediastinal lymph node dissection or hilar dissection the patient shows signs of hemodynamic instability or cardiac ischemia, the pulmonary resection can be abandoned and cardiopulmonary bypass can be begun immediately. Exposure through a median sternotomy can be satisfactory; however, experience is often required if the surgeon is to become comfortable with this particular approach. In most cases, if the patient is in hemodynamically stable condition, visualization of the hilum during one-lung ventilation is satisfactory. The only exception to this is exposure of the left lower lobe hilum, which can be particularly difficult because retraction of the heart compromises hemodynamics. Under these circumstances exposure can be greatly facilitated by supporting the patient with cardiopulmonary bypass and decompressing the heart. Mediastinal lymph node sampling can be performed through a median sternotomy, although there is no question that complicated hilar or mediastinal dissection is more easily done via a posterolateral thoracotomy. Subcarinal and posterior mediastinal lymph nodes are par-
Yokoyama, Derrick, Lee
9I5
ticulariy difficult to sample. One must remember that lymph node metastasis in lung cancer is evidence of systemic disease as it is in colon cancer, breast cancer, melanoma, and other malignant tumors. Incomplete mediastinallymph node dissection may affect prognosis but should not have any effect on survival. Despite the fact that operative survival was excellent, short-term survival in this small series was less than expected. Of II patients treated, 4 patients have died of metastatic carcinoma 5 to 22 months after resection, and I patient has recurrent disease at 19 months. Because the number of patients in this study is small, definitive conclusions cannot be made. The possibility that cardiopulmonary bypass has some transient effect on the immune system is certainly possible. Further studies will be needed to address this issue. Our observations and the literature suggest that combined cardiac operations and pulmonary resection can be safely performed and an adequate cancer operation can be undertaken in selected patients with stage I or stage II pulmonary neoplastic disease.o 6 Experience by others has shown a small risk of intraparenchymal pulmonary hemorrhage when pulmonary wedge resection is performed during bypass.l Because of this, caution is recommended when one is considering deep wedge resections during bypass. It may prove safer to perform wedge resections either before bypass before heparinization or after bypass after reversal of heparin with protamine sulfate. In conclusion, we find that combined cardiac operations and pulmonary resection for neoplastic disease have the advantage of correcting both problems simultaneously and saving the patient from the pain and the risk of a second major procedure. Exposure through a median sternotomy has been satisfactory and can be further facilitated by one-lung ventilation and cardiopulmonary bypass. If patients are properiy screened and technical considerations are taken into account, a concomitant pulmonary resection with staging can be safely performed with minimal morbidity and a short hospitalization. We acknowledge the help of Lucy Gonzales and Marianne Singleton in manuscript preparation. REFERENCES I. Foster ED, DavisKB, Carpenter JA, et al. Risk of noncardiac operationin patients withdefined coronaryartery disease: The Coronary Artery Surgery Study (CASS) registry experience. Ann Thorac Surg 1986;41 :42-50. 2. Steen PA, Tinker JH, Tarhan S. Myocardial reinfarction after anesthesia and surgery. JAMA 1978;239:2566-70. 3. Weitz HH, Goldman L. Noncardiac surgeryin the patient with heart disease. Med Clin North Am 1987;71:413-31.
9I6
Yokoyama, Derrick, Lee
4. Dalton ML, Parker TM, Mistrot JJ, Bricker DL. Concomitant coronary artery bypass and major noncardiac surgery. J THORAC CARDIOVASC SURG 1978;75:621-4. 5. Piehler JM, Trastek VF, Pairolero PC, et al. Concomitant cardiac and pulmonary operations. J THORAC CARDlOVASC SURG 1985;90:662-7. 6. Canver CC, Bhayana IN, Lajos TZ, et al. Pulmonary resection combined with cardiac operations. Ann Thorac Surg 1990;50:796-9. 7. Read RC, Yoder G, Schaeffer RC. Survival after conservative resection for TJNoMo non-small cell lung cancer. Ann Thorac Surg 1990;49:391-400. 8. Asaph JW, Keppel JF. Midline sternotomy for the treatment of primary pulmonary neoplasms. Am J Surg 1984; 147:589-92. 9. Urschel HC, Razzuk MA. Median sternotomy as a standard approach for pulmonary resection. Ann Thorac Surg 1986;41:130-4. 10. Cooper JD, N elems JM, Pearson FG. Extended indications for median sternotomy in patients requiring pulmonary resection. Ann Thorac Surg 1978;26:413-20. 11. Peters RM, Swain JA. Management of the patient with emphysema, coronary artery disease, and lung cancer. Am J Surg 1982;143:701-5.
Discussion Dr. Bradley J. Harlan (Sacramento, Calif). Dr. Derrick addresses an important problem in management of the patient with surgically treatable cardiac disease and suspected or proved malignant pulmonary disease. He has shown that pulmonary resection, at the time of cardiac surgery, can be performed with low mortality and low morbidity, observations that confirm the conclusions of previously published studies. Dr. William Logue and my partners, Drs. Junod and Miller, have recently reviewed our experience in a similar group of patients. In 19 patients spanning an II-year period there were one death and no hemorrhagic complications or wound infections, results similar to those of Dr. Derrick. Nine of these patients had lobectomies and 10 had wedge resections. The series includes benign as well as malignant lesions, and this brings me to my first question. Why did you not include patients with benign lesions? Do you know how many patients had combined operations for benign lesions and whether the morbidity in this group of patients was as low as in your group with malignant lesions? Dr. Derrick. The majority of our patients had malignant disease. Three other patients underwent wedge resections for benign disease but were not included in this study. Because the number of patients with benign disease was so small, we decided to limit our paper to patients with malignant disease. All three patients who underwent combined cardiac operations with wedge resections for benign disease had uneventful postoperative courses without any significant complications. Dr. Harlan. I have only one more comment and question. We begin the dissection of the hilar vessels after the completion of the cardiac procedure with the patient supported by normothermic bypass and use a single-lumen endotracheal tube. The collapsed lung facilitates exposure and dissection and there is no risk of inducing cardiac ischemia as there is with your method. Do you know whether ischemia occurred before bypass in any
The Journal of Thoracic and Cardiovascular Surgery May 1993
of these patients and whether dissection was therefore abandoned until after the cardiac procedure was performed? Why not use the more simplified technique? Dr. Derrick. One patient in our series had ST changes consistent with ischemia during dissection of the hilum. This was a 52-year-old patient having redo coronary bypass grafting. Bypass was instituted immediately and the remainder of the dissection was performed at the conclusion of the cardiac operation while the patient was still supported by bypass. His bypass time was 225 minutes, which was the longest in our series. Despite this, the patient did quite well. There are many methods of conducting a combined cardiac operation with pulmonary resection. These have been discussed in the literature. I believe it is really a matter of personal preference. We have found that hilar dissection performed before heparinization is associated with less bleeding. One-lung ventilation obviously facilitates our dissection. Dissection at the conclusion of the operation is often associated with more oozing because of platelet dysfunction. There are three factors that have to be considered when doing this kind of procedure. Probably the most important is selecting the patients properly. One must first consider patient characteristics, including age, left ventricular function, and pulmonary status. Then one must consider technical problems such as location of the lesion and the presence or absence of adhesions. Finally, one must consider the surgeon's experience with the midline sternotomy approach. All three of these factors must be considered when one is deciding whether or not to perform a combined cardiac operation with pulmonary resection or a staged procedure. Dr. Richard Anderson (Seattle, Wash.). I am certainly impressed with these results. I have long been reluctant to combine pulmonary resection and cardiopulmonary bypass for the reasons that you mentioned. Your results give me confidence to do this more frequently. We have generally done these operations sequentially. Can you give us some idea how many patients were operated on sequentially over the same time period and the reasons for staging? Dr. Derrick. I do not know the exact number of patients but I am sure it was less than five. Typically, patients had staged procedures if they had poor pulmonary reserve, if they required an emergency cardiac operation, or if their cardiac procedure alone posed a high risk of mortality. Dr. Anderson. The other thing that worries me about doing a resection involvingtransection of a bronchus is that the patient may require prolonged ventilation after the operation. Pulmonary function obviously must have an effect. Are there situations in which you would not do a lobectomy because of poor pulmonary function? Dr. Derrick. Clearly there are situations in which lobectomy would be dangerous because of poor pulmonary function. In addition, it is important that the surgeon consider the difficulty and complexity of the cardiac operation. If one is anticipating a prolonged pump run in a patient who already has borderline pulmonary function, I think the advisability of doing a lobar resection warrants careful consideration. A wedge resection alone should be safe in all patients and can give valuable information for decisions pertaining to a second staged operation. Dr. John R. Benfield (Sacramento, Calif). A number of years ago we presented a series of 70 consecutive intraoperative needle aspiration diagnoses of pulmonary lesions (J THORAC
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 5
CARDIOVASC SURG 1983;85:404-8). The reason I bring this forward again is that I notice that you relied largely on wedge resections or other forms of frozen section biopsy. Did you use intraoperative needle diagnosis and do you intend to do that in the future? Our cytologist, Dr. Raymond L. Teplitz, was able to differentiate cancer from noncancerous lesions reliably among the cancers that we subsequently resected. He and his associates were able to give a correct cell type diagnosis in 85% of the cases. Therefore you might consider the use of intraoperative pine needle biopsies in the future. Dr. Derrick. There is definitely a place for intraoperative needle aspiration diagnosis of pulmonary lesions. We have found in this series that a stapled wedge resection has given us
Yokoyama, Derrick, Lee
917
the information that we have required. Intraoperative fine needle aspiration, however, can be especially helpful for lesions in which biopsy by standard methods is difficult or hazardous. It is important that the pathologist at your institution be skilled in interpreting these samples. Our group plans to use this technique more in the future. Dr. Walter Dembitsky (San Diego, Calif). Since we are speaking in anecdotes, I do know of an interesting case in which a wedge biopsy was done for an infiltrate in the lung. After the operation, the patient had disseminated fungal infection with growths in the skull, sternum, and long bones. Thus one must be cautious in obtaining biopsy specimens of infiltrates and then putting patients on a cardiopulmonary circuit. Dr. Derrick. This is true, I agree.
Bound volumes available to subscribers Boundvolumesof THEJOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1993issues from the Publisher, at a cost of $75.00 for domestic, $98.25 for Canadian, and $93.00 for international subscribers for Vol. 105 (JanuaryJune) and Vol. 106 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertisingis removed.Copies are shipped within 60 days after publicationof the last issueof the volume.The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby, SubscriptionServices,11830Westline Industrial Drive,St. Louis,Missouri63146-3318,USA; phone I (800) 453-4351or(314) 453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular
JOURNAL
subscription.
Taro Yokoyama, MD, PhD, Marvin J. Derrick, MD, and Anthony W. Lee, MD, Los Angeles and Burbank. Calif.
I
t is not unusual to have both surgical disease of the heart and pulmonary neoplastic disease simultaneously. In the past, management of these patients presented a diagnostic as well as a therapeutic problem. Resecting the pulmonary lesion without addressing the cardiac problem posed a significant risk to the patient. 1-3 Increasing experience suggests that both problems can be addressed during the same operation. Several reports in the literature have suggested that concomitant cardiac operations and noncardiac proceFrom thePacific Cardiothoracic Surgery Group, St. Vincent Medical Center, Los Angeles, Calif., and Saint Joseph Medical Center, Burbank, Calif. Read at the Eighteenth Annual Meeting of The Western Thoracic Surgical Association, Kauai, Hawaii, June 24-27, 1992. Address for reprints: Taro Yokoyama, MD, PhD, 201 S.Alvarado St., Suite 626, Los Angeles, CA 90057. Copyright © 1993 byMosby-Year Book, Inc. 0022-5223/93 $1.00 +.10
912
12/6/44677
dures can be successfully and safely performed.v" despite the risk of bleeding caused by altered coagulation mechanisms. The combined procedure offers distinct advantages over operations done in a staged fashion. One obvious advantage of the combined procedure for the patient is elimination of the pain and risk associated with a second major operation. This report presents our experience with concomitant cardiac operations and pulmonary resection in II patients with bronchogenic carcinoma. Patients From February 1988 to May 1992, a totalof 2909 adultcardiacsurgical procedures necessitating cardiopulmonary bypass were performed at Saint Joseph Hospital in Burbank andSaint Vincent Hospital in LosAngeles. Duringthat period II patients underwent combined pulmonary resection of a bronchogenic carcinoma with a cardiac procedure that necessitated cardiopulmonary bypass. Excluded were patients who underwent concomitant pulmonary resections for benign disease or iatrogenic intraoperative pulmonary complications.
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 5
Yokoyama, Derrick, Lee
Table I. Cardiac procedures performed with concomitant pulmonary resection Procedure
CABG Redo CABG AVR AVR and CABG Total
No. of patients 7
2 I I II
9 13
Table II. Lung disease
Percent 64 18 9 9 100
Disease
No. of patients
Percent
Adenocarcinoma Squamous cell carcinoma Bronchoalveolar U ndilferentiated Total
4 4 2 I II
36 36 18 10 100
CARG. Coronary artery bypass grafting; A YR. aortic valve replacement.
The charts on each of these patients were reviewed to ascertain the operation performed, cell type, TNM classification, stage, timing of pulmonary resection in relation to heparinization, bypass time, and hospital stay. Follow-up was done by contacting the patient and his or her primary physician by telephone to ascertain the postoperative status. The ages of the patients ranged from 52 to 81 years (mean age 70.4 years). Eight were men and three were women. All patients had symptomatic cardiac disease and were in New York Heart Association class III or IV at the time of the operation. Nine patients had coronary artery disease and two had valvular heart disease (Table I). Patients with coronary artery disease required from one to five bypasses. All procedures were performed on an elective basis. In each patient an asymptomatic pulmonary lesion was found on routine preoperative chest x-ray films. The lesion was a nodule in seven patients and an infiltrate in four patients. A preoperative diagnosis of bronchogenic carcinoma was made in six patients before the operation, either by percutaneous needle biopsy in five patients or by broncoscopic biopsy in one. The lesions in the remaining five patients were highly suggestive of carcinoma. A computed tomographic scan was obtained in all patients before the operation. No evidence of mediastinal adenopathy was identified. Pulmonary procedures performed included two wedge resections, seven lobectomies, and two bilobectomies. Extensive mediastinal and hilar lymph node sampling was done at the time of the operation. All of the procedures were performed through a median sternotomy. A double-lumen endotracheal tube was used in each case. Standard pulmonary techniques were used for all resections and a mechanical stapler was used for all bronchial closures. Wedge resections were performed before bypass prior to heparinization by the standard technique with a GIA stapler (Auto Suture Company Division, United States Surgical Corporation, Norwalk, Conn.). Patients without a preoperative diagnosis of bronchogenic carcinoma had intraoperative frozen section confirmation of the neoplasm before lobectomy. Dissection of the hilum for lobar resection was performed before bypass prior to heparinization. In a majority of the patients, the lobe was removed at the conclusion of the cardiac procedure while the patient was still supported by bypass with the lung collapsed during rewarming.
Results There were no operative deaths among the II patients in our study. In addition, there were no major postoperative complications related to the cardiac procedure or
pulmonary resection. Postoperative stay ranged from 6 to 17 days (mean 10 days), except for one patient who remained hospitalized for 29 days. In this patient a hemothorax developed after thoracentesis on the side opposite the pulmonary resection. Tube thoracostomy failed to adequately drain the blood, and a thoracotomy with decortication was eventually required. Another patient, who had a double lobectomy, remained hospitalized for 17 days primarily because of repeated collapse of the remaining right lower lobe as a result of mucous plugging. Treatment with a bronchoscope was required on two separate occasions. A third patient was hospitalized for 16 days primarily because of severe postoperative bronchospasm, which eventually responded to vigorous medical therapy. Two patients had transient atrial fibrillation after the operation; both had lobectomies. All patients were discharged home in normal sinus rhythm. One patient had postoperative periods of sinus arrest with junctional escape after aortic valve replacement. Because of this and in conjunction with her preoperative evidence of conduction system disease, a permanent pacemaker was placed. None ofthe patients had prolonged air leaks, excessivechest tube drainage, pleural space infections, or wound infections. A small air pocket developed posteriorly after chest tube removal in one patient who had left upper lobe resection. The patient refused aspiration of this pocket. On postoperative follow-up, the small pocket of air was completely reabsorbed. Pathologic examination showed a variety of malignancies (Table II). There were four adenocarcinomas, four squamous cell carcinomas, two bronchoalveolar cell carcinomas, and one undifferentiated carcinoma. Nine of these malignant tumors were stage I and two were stage II. Mediastinal lymph node sampling was performed in each case and no evidence of N2 disease was found. Pulmonary resection was believed to be appropriate for the pathologic condition in all patients studied. Pulmonary procedures included two wedge resections, seven lobectomies, and two bilobectomies. One patient, who had a wedge resection performed for a T1 NO MO adenocarcinoma, was found to have evidence of cancer close to the margin on final pathologic analysis. She underwent
914
The Journal of Thoracic and Cardiovascular Surgery May 1993
Yokoyama, Derrick, Lee
completion lobectomy on a separate hospital admission. No evidence of cancer was found in the remaining lung or evidence of hilar lymph node disease. The only patient who underwent a definitive wedge resection was a 79year-old patient with a small coin lesion in the right lower lobe. Results of initial frozen section analysis were reported as carcinoma, probably metastatic versus carcinoid. The lesion was peripheral and measured 13 mm in greatest diameter. A postoperative metastatic work-up showed no abnormalities. The final pathology report indicated an undifferentiated carcinoma with good margins. Studies have suggested that peripherally located TI NO MO lesions less than 2 em in diameter have a better prognosis and may be treated adequately with conservative resection'? This combined with the patient's age and general debilitation suggested that no further surgical intervention was necessary. In seven of the nine patients who underwent lobectomy, resection was performed at the conclusion of the cardiac operation during rewarming when the patient was still supported by cardiopulmonary bypass. Individual patient bypass time ranged from 88 to 225 minutes. Average bypass time for the seven patients who underwent lobar resection on bypass was 143 minutes. Because much of the hilar dissection was performed before bypass (before heparinization), final ligation of the pulmonary vessels with stapling of the interlobar fissure and bronchus added little to the total pump time. On bypass, with the heart decompressed and lungs collapsed, exposure of the hilum was greatly facilitated. Bleeding was not found to be a significant problem during heparinization so long as careful dissection was undertaken. Complete follow-up of all patients ranged from I to 22 months (average 12.1 months). Four patients are presently alive and free of any evidence of malignant disease at postoperative intervals of 1 to 18 months. A 76-yearold man who underwent right middle and lower lobectomy for a T2 NO MO squamous cell carcinoma has recurrence locally in the chest 19 months after the operation. He is presently undergoing radiation therapy for extensive disease involving the remaining right upper lobe and trachea. There was one death unrelated to cardiac or pulmonary disease. This occurred in a 69-year-old patient with stage II squamous cell carcinoma who died of bulbar palsy 21 months after the operation. There was no evidence of recurrence of his cancer. One patient had a sudden cardiac death at 5 months. He was also free of any recurrent lung cancer. Four patients have died of recurrent metastatic lung cancer 5 to 22 months after resection. All of these patients had T I NO MO disease that included bronchoalveolar cell carcinoma, adenocarcinoma, squamous cell carcinoma, and undifferentiated carcinoma.
Pulmonary procedures on these patients included two lobectomies, one bilobectomy, and one wedge resection.
Discussion In selected patients, median sternotomy has proved to be a reasonable alternative to posterolateral or anterior thoracotomy.s-" It provides excellent exposure for the resection of metastatic tumors and also has advantages in the approach to bilateral bullous disease.!? In addition, several reports have advocated sternotomy for the treatment of primary pulmonary neoplasms.f- 9 Because it is not unusual to have both concomitant surgical disease of the heart and pulmonary neoplastic disease, it would appear logical to consider treating the two problems during the same operation. Traditionally, pulmonary resection has not been routinely performed with cardiac procedures that require extracorporeal circulation, primarily because of a concern for intrapulmonary bleeding, a risk that comes from heparinization and from altered coagulation mechanisms caused by cardiopulmonary bypass. Intraparenchymal pulmonary hemorrhage after wedge resection during cardiopulmonary bypass has led to acute respiratory insufficiency and death.P Another concern has been the spread of local infection after resection of fungal infiltrates or granuloma, possibly the result of altered immune mechanisms related to cardiopulmonary bypass. I I Finally, there is the technical concern over whether adequate mediastinal lymph node sampling for staging can be performed by this approach. Advantages of combined cardiac and pulmonary operations outweigh such concerns. Patients who undergo the procedure avoid the pain and risk associated with a staged operation. The patients are also spared the cost of a second hospitalization. Furthermore, a combined procedure allows expeditious treatment of the pulmonary neoplasm, a factor that could be important in tumors with rapid doubling times. Median sternotomy appears to be well tolerated by patients. Cooper, Nelems, and Pearson'? have demonstrated faster recovery of pulmonary function in patients who had a median sternotomy versus a standard posterolateral thoracotomy. Postoperative pain has also been shown to be reduced.f This can be an important factor in patients with borderline pulmonary function, a situation frequently seen in patients with bronchogenic carcinoma. The decision to proceed with a combined procedure must be weighed carefully with each patient. Factors to consider include pulmonary reserve, age, presence and degree of left ventricular dysfunction, bypass time, and functional status. Technical considerations must also be taken into account. Large central tumors that require
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 5
pneumonectomy, superior sulcus tumors, extensive adhesions, or tumors with chest wall involvement should probably be resected at a later time in a staged fashion. This does not preclude wedge resection or mediastinal lymph node sampling, which may alter the prognosis and determine whether the patient has resectable disease. Our technique includes using a double-lumen endotracheal tube. We believe that a critical factor necessary in performing a satisfactory pulmonary resection through a median sternotomy is selective unilateral ventilation. Deflation of the lung allows an unobstructed view of the hilum and mediastinum and a more expeditious dissection. After the chest is opened, the lesion is inspected and technical considerations for resection are considered. Division of the inferior pulmonary ligament will allow access to more remote areas of the lung. A wedge resection or mediastinal lymph node dissection can be performed at this time. Frozen section analysis of these specimens may alter the proposed treatment plan. If concomitant resection is indicated, then dissection of the hilum is undertaken with identification of the pulmonary artery branches and pulmonary vein. If exposure is good, then the lobar resection can be completed. Otherwise, the patient is heparinized and supported by cardiopulmonary bypass with resection being performed at the conclusion of the cardiac procedure during rewarming. Exposure is optimized with the lung deflated and the heart decompressed during bypass. If hilar dissection has been performed before bypass, the actual pulmonary resection during rewarming can take anywhere from 10 to 20 minutes. If at any time during the initial mediastinal lymph node dissection or hilar dissection the patient shows signs of hemodynamic instability or cardiac ischemia, the pulmonary resection can be abandoned and cardiopulmonary bypass can be begun immediately. Exposure through a median sternotomy can be satisfactory; however, experience is often required if the surgeon is to become comfortable with this particular approach. In most cases, if the patient is in hemodynamically stable condition, visualization of the hilum during one-lung ventilation is satisfactory. The only exception to this is exposure of the left lower lobe hilum, which can be particularly difficult because retraction of the heart compromises hemodynamics. Under these circumstances exposure can be greatly facilitated by supporting the patient with cardiopulmonary bypass and decompressing the heart. Mediastinal lymph node sampling can be performed through a median sternotomy, although there is no question that complicated hilar or mediastinal dissection is more easily done via a posterolateral thoracotomy. Subcarinal and posterior mediastinal lymph nodes are par-
Yokoyama, Derrick, Lee
9I5
ticulariy difficult to sample. One must remember that lymph node metastasis in lung cancer is evidence of systemic disease as it is in colon cancer, breast cancer, melanoma, and other malignant tumors. Incomplete mediastinallymph node dissection may affect prognosis but should not have any effect on survival. Despite the fact that operative survival was excellent, short-term survival in this small series was less than expected. Of II patients treated, 4 patients have died of metastatic carcinoma 5 to 22 months after resection, and I patient has recurrent disease at 19 months. Because the number of patients in this study is small, definitive conclusions cannot be made. The possibility that cardiopulmonary bypass has some transient effect on the immune system is certainly possible. Further studies will be needed to address this issue. Our observations and the literature suggest that combined cardiac operations and pulmonary resection can be safely performed and an adequate cancer operation can be undertaken in selected patients with stage I or stage II pulmonary neoplastic disease.o 6 Experience by others has shown a small risk of intraparenchymal pulmonary hemorrhage when pulmonary wedge resection is performed during bypass.l Because of this, caution is recommended when one is considering deep wedge resections during bypass. It may prove safer to perform wedge resections either before bypass before heparinization or after bypass after reversal of heparin with protamine sulfate. In conclusion, we find that combined cardiac operations and pulmonary resection for neoplastic disease have the advantage of correcting both problems simultaneously and saving the patient from the pain and the risk of a second major procedure. Exposure through a median sternotomy has been satisfactory and can be further facilitated by one-lung ventilation and cardiopulmonary bypass. If patients are properiy screened and technical considerations are taken into account, a concomitant pulmonary resection with staging can be safely performed with minimal morbidity and a short hospitalization. We acknowledge the help of Lucy Gonzales and Marianne Singleton in manuscript preparation. REFERENCES I. Foster ED, DavisKB, Carpenter JA, et al. Risk of noncardiac operationin patients withdefined coronaryartery disease: The Coronary Artery Surgery Study (CASS) registry experience. Ann Thorac Surg 1986;41 :42-50. 2. Steen PA, Tinker JH, Tarhan S. Myocardial reinfarction after anesthesia and surgery. JAMA 1978;239:2566-70. 3. Weitz HH, Goldman L. Noncardiac surgeryin the patient with heart disease. Med Clin North Am 1987;71:413-31.
9I6
Yokoyama, Derrick, Lee
4. Dalton ML, Parker TM, Mistrot JJ, Bricker DL. Concomitant coronary artery bypass and major noncardiac surgery. J THORAC CARDIOVASC SURG 1978;75:621-4. 5. Piehler JM, Trastek VF, Pairolero PC, et al. Concomitant cardiac and pulmonary operations. J THORAC CARDlOVASC SURG 1985;90:662-7. 6. Canver CC, Bhayana IN, Lajos TZ, et al. Pulmonary resection combined with cardiac operations. Ann Thorac Surg 1990;50:796-9. 7. Read RC, Yoder G, Schaeffer RC. Survival after conservative resection for TJNoMo non-small cell lung cancer. Ann Thorac Surg 1990;49:391-400. 8. Asaph JW, Keppel JF. Midline sternotomy for the treatment of primary pulmonary neoplasms. Am J Surg 1984; 147:589-92. 9. Urschel HC, Razzuk MA. Median sternotomy as a standard approach for pulmonary resection. Ann Thorac Surg 1986;41:130-4. 10. Cooper JD, N elems JM, Pearson FG. Extended indications for median sternotomy in patients requiring pulmonary resection. Ann Thorac Surg 1978;26:413-20. 11. Peters RM, Swain JA. Management of the patient with emphysema, coronary artery disease, and lung cancer. Am J Surg 1982;143:701-5.
Discussion Dr. Bradley J. Harlan (Sacramento, Calif). Dr. Derrick addresses an important problem in management of the patient with surgically treatable cardiac disease and suspected or proved malignant pulmonary disease. He has shown that pulmonary resection, at the time of cardiac surgery, can be performed with low mortality and low morbidity, observations that confirm the conclusions of previously published studies. Dr. William Logue and my partners, Drs. Junod and Miller, have recently reviewed our experience in a similar group of patients. In 19 patients spanning an II-year period there were one death and no hemorrhagic complications or wound infections, results similar to those of Dr. Derrick. Nine of these patients had lobectomies and 10 had wedge resections. The series includes benign as well as malignant lesions, and this brings me to my first question. Why did you not include patients with benign lesions? Do you know how many patients had combined operations for benign lesions and whether the morbidity in this group of patients was as low as in your group with malignant lesions? Dr. Derrick. The majority of our patients had malignant disease. Three other patients underwent wedge resections for benign disease but were not included in this study. Because the number of patients with benign disease was so small, we decided to limit our paper to patients with malignant disease. All three patients who underwent combined cardiac operations with wedge resections for benign disease had uneventful postoperative courses without any significant complications. Dr. Harlan. I have only one more comment and question. We begin the dissection of the hilar vessels after the completion of the cardiac procedure with the patient supported by normothermic bypass and use a single-lumen endotracheal tube. The collapsed lung facilitates exposure and dissection and there is no risk of inducing cardiac ischemia as there is with your method. Do you know whether ischemia occurred before bypass in any
The Journal of Thoracic and Cardiovascular Surgery May 1993
of these patients and whether dissection was therefore abandoned until after the cardiac procedure was performed? Why not use the more simplified technique? Dr. Derrick. One patient in our series had ST changes consistent with ischemia during dissection of the hilum. This was a 52-year-old patient having redo coronary bypass grafting. Bypass was instituted immediately and the remainder of the dissection was performed at the conclusion of the cardiac operation while the patient was still supported by bypass. His bypass time was 225 minutes, which was the longest in our series. Despite this, the patient did quite well. There are many methods of conducting a combined cardiac operation with pulmonary resection. These have been discussed in the literature. I believe it is really a matter of personal preference. We have found that hilar dissection performed before heparinization is associated with less bleeding. One-lung ventilation obviously facilitates our dissection. Dissection at the conclusion of the operation is often associated with more oozing because of platelet dysfunction. There are three factors that have to be considered when doing this kind of procedure. Probably the most important is selecting the patients properly. One must first consider patient characteristics, including age, left ventricular function, and pulmonary status. Then one must consider technical problems such as location of the lesion and the presence or absence of adhesions. Finally, one must consider the surgeon's experience with the midline sternotomy approach. All three of these factors must be considered when one is deciding whether or not to perform a combined cardiac operation with pulmonary resection or a staged procedure. Dr. Richard Anderson (Seattle, Wash.). I am certainly impressed with these results. I have long been reluctant to combine pulmonary resection and cardiopulmonary bypass for the reasons that you mentioned. Your results give me confidence to do this more frequently. We have generally done these operations sequentially. Can you give us some idea how many patients were operated on sequentially over the same time period and the reasons for staging? Dr. Derrick. I do not know the exact number of patients but I am sure it was less than five. Typically, patients had staged procedures if they had poor pulmonary reserve, if they required an emergency cardiac operation, or if their cardiac procedure alone posed a high risk of mortality. Dr. Anderson. The other thing that worries me about doing a resection involvingtransection of a bronchus is that the patient may require prolonged ventilation after the operation. Pulmonary function obviously must have an effect. Are there situations in which you would not do a lobectomy because of poor pulmonary function? Dr. Derrick. Clearly there are situations in which lobectomy would be dangerous because of poor pulmonary function. In addition, it is important that the surgeon consider the difficulty and complexity of the cardiac operation. If one is anticipating a prolonged pump run in a patient who already has borderline pulmonary function, I think the advisability of doing a lobar resection warrants careful consideration. A wedge resection alone should be safe in all patients and can give valuable information for decisions pertaining to a second staged operation. Dr. John R. Benfield (Sacramento, Calif). A number of years ago we presented a series of 70 consecutive intraoperative needle aspiration diagnoses of pulmonary lesions (J THORAC
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 5
CARDIOVASC SURG 1983;85:404-8). The reason I bring this forward again is that I notice that you relied largely on wedge resections or other forms of frozen section biopsy. Did you use intraoperative needle diagnosis and do you intend to do that in the future? Our cytologist, Dr. Raymond L. Teplitz, was able to differentiate cancer from noncancerous lesions reliably among the cancers that we subsequently resected. He and his associates were able to give a correct cell type diagnosis in 85% of the cases. Therefore you might consider the use of intraoperative pine needle biopsies in the future. Dr. Derrick. There is definitely a place for intraoperative needle aspiration diagnosis of pulmonary lesions. We have found in this series that a stapled wedge resection has given us
Yokoyama, Derrick, Lee
917
the information that we have required. Intraoperative fine needle aspiration, however, can be especially helpful for lesions in which biopsy by standard methods is difficult or hazardous. It is important that the pathologist at your institution be skilled in interpreting these samples. Our group plans to use this technique more in the future. Dr. Walter Dembitsky (San Diego, Calif). Since we are speaking in anecdotes, I do know of an interesting case in which a wedge biopsy was done for an infiltrate in the lung. After the operation, the patient had disseminated fungal infection with growths in the skull, sternum, and long bones. Thus one must be cautious in obtaining biopsy specimens of infiltrates and then putting patients on a cardiopulmonary circuit. Dr. Derrick. This is true, I agree.
Bound volumes available to subscribers Boundvolumesof THEJOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1993issues from the Publisher, at a cost of $75.00 for domestic, $98.25 for Canadian, and $93.00 for international subscribers for Vol. 105 (JanuaryJune) and Vol. 106 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertisingis removed.Copies are shipped within 60 days after publicationof the last issueof the volume.The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby, SubscriptionServices,11830Westline Industrial Drive,St. Louis,Missouri63146-3318,USA; phone I (800) 453-4351or(314) 453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular
JOURNAL
subscription.