Prognosis after neoadjuvant chemoradiation or chemotherapy for locally advanced gastro-oesophageal junctional adenocarcinoma

BACKGROUND

Trials typically group cancers of the gastro-oesophageal junction (GOJ) with oesophageal or gastric cancer when studying neoadjuvant chemoradiation and perioperative chemotherapy, so the results may not be fully applicable to GOJ cancer. Because optimal neoadjuvant treatment for GOJ cancer remains controversial, outcomes with neoadjuvant chemoradiation versus chemotherapy for locally advanced GOJ adenocarcinoma were compared retrospectively.

METHODS

Data were collected from all patients who underwent neoadjuvant treatment followed by surgery for adenocarcinoma located at the GOJ at a single high-volume institution between 2002 and 2017. Postoperative major complications and mortality were compared between groups using Fisher's exact test. Overall survival (OS) and disease-free survival (DFS) were assessed by log rank test and multivariable Cox regression analyses. Cumulative incidence functions were used to estimate recurrence, and groups were compared using Gray's test.

RESULTS

Of 775 patients, 650 had neoadjuvant chemoradiation and 125 had chemotherapy. These groups were comparable in terms of clinical tumour and lymph node categories, although the chemoradiation group had greater proportions of white men, complete pathological response to chemotherapy, and smaller proportions of diffuse cancer, poor differentiation, and neurovascular invasion. Postoperative major complications (20.0 versus 17.6 per cent) and 30-day mortality (1.7 versus 1.6 per cent) were not significantly different between the chemoradiation and chemotherapy groups. After adjustment, type of therapy (chemoradiation versus chemotherapy) was not significantly associated with OS (hazard ratio (HR) 1.26, 95 per cent c.i. 0.96 to 1.67) or DFS (HR 1.27, 0.98 to 1.64). Type of recurrence (local, regional, or distant) did not differ after neoadjuvant chemoradiation versus chemotherapy.

CONCLUSION

In patients undergoing surgical resection for locally advanced adenocarcinoma of the GOJ, OS and DFS did not differ significantly between patients who had neoadjuvant chemoradiation compared with chemotherapy.

Outcomes of induction chemotherapy followed by chemoradiation using intensity-modulated radiation therapy for esophageal adenocarcinoma

This study looks at toxicity and survival data when chemoradiation (CRT) is delivered using intensity-modulated radiation therapy (IMRT) after induction chemotherapy. Forty-one patients with esophageal adenocarcinoma treated with IMRT from March 2007 to May 2009 at Memorial Sloan-Kettering Cancer Center were analyzed. All patients received induction chemotherapy prior to CRT. Thirty-nine percent (n = 16) of patients underwent surgical resection less than 4 months after completing CRT. Patients were predominantly male (78%), with a median age of 68 years (range 32-85 years). The majority of acute treatment-related toxicity was hematologic or gastrointestinal, with 17% of patients having grade 3+ hematologic toxicity and 12% of patients having grade 3+ gastrointestinal toxicity. Only two patients developed grade 2-3 pneumonitis (5%) and 5 patients experienced post-operative pulmonary complications (29%). Eight patients (20%) required a treatment break. With a median follow up of 41 months for surviving patients, 2-year overall survival was 61%, and the cumulative incidences of local failure (LF) and distant metastases were 40% and 51%, respectively. This rate of LF was reduced to 13% in patients who underwent surgical resection. Surgery and younger age were significant predictors of decreased time to LF on univariate analysis. Induction chemotherapy followed by CRT using IMRT in the treatment of esophageal cancer is well tolerated and is not associated with an elevated risk of postoperative pulmonary complications. The use of IMRT may allow for integration of more intensified systemic therapy or radiation dose escalation for esophageal adenocarcinoma, ultimately improving outcomes for patients with this aggressive disease.

Clinical tools do not predict pathological complete response in patients with esophageal squamous cell cancer treated with definitive chemoradiotherapy

Chemoradiotherapy for locally advanced esophageal squamous cell carcinoma is associated with high rates of pathological complete response. A pathological complete response is recognized to be an important predictor of improved survival, to the extent that observation rather than surgery is advocated by some in patients with presumed pathological complete response based on their clinical response. The goal of this study was to look at the ability of clinical variables to predict pathological complete response after chemoradiotherapy for locally advanced esophageal squamous cell carcinoma. We reviewed retrospectively patients with locally advanced esophageal squamous cell carcinoma who underwent chemoradiotherapy followed by surgery and compared those with pathological complete response to patients with residual disease. Between January 1996 and December 2010, 116 patients met inclusion criteria. Fifty-six percent of patients had a pathological complete response and a median survival of 128.1 months versus 28.4 months in patients with residual disease. When compared with patients with residual disease, patients with a pathological complete response had a lower post-neoadjuvant positron emission tomography (PET) maximum standardized uptake value (SUVmax), a larger decrease in PET SUVmax, a less thick tumor on post-chemoradiotherapy computed tomography and a higher rate of normal appearing post-chemoradiotherapy endoscopy with benign biopsy of the tumor bed. However, none of these characteristics alone was able to correctly identify patients with a pathological complete response, and none has significant specificity. Although the rate of pathological complete response after chemoradiotherapy is high in patients with esophageal squamous cell carcinoma, the ability of identifying patients with pathological complete response is limited. A reduction of the PET SUVmax by >70%, a normal appearing endoscopic examination, and no residual disease on biopsy all were seen in >65% of the patients with a pathological complete response. Even if these findings were unable to confirm the absence of residual disease in the primary tumor, they can help guide expectant management in high-risk patients.

The association of PET response with complete pathological response (CPR) and residual nodal disease (RND) after induction chemoradiotherapy (CRT) and resection of esophageal cancer: A review of 493 cases

4552

Background: RND and lack of CPR of the primary tumor correlate with poor survival after induction CRT and resection of esophageal cancer. PET response to CRT (SUVmax change and post-induction SUVmax) is used by some clinicians as an indicator of CPR and RND in order to stratify patients after CRT to observation alone vs completion resection. We aimed to investigate the association of PET response with CPR and RND after induction CRT and resection of esophageal cancer. Methods: An IRB-approved retrospective review of an institutional surgical database identified patients who underwent resection of esophageal squamous cell (SCC) and adeno carcinoma (AC) following CRT. The database was locked on Sept 30, 2008. Categorical variables were analyzed by chi square, continuous variables by t-test, and survival by the Kaplan-Meier method. Results: From 1/96 to 3/08, 493 patients were identified, 82% were male. Median age was 62, chemotherapy cisplatin-based in 87%, mean radiation 50 Gy, in-hospital mortality 4.1% and R0 resection rate 88%. Pathology revealed AC in 80%, lack of CPR in 73% and RND in 35%. While in AC patients CPR and lack of RND were both associated with prolonged survival, PET response was not associated with either. In SCC patients, prolonged survival was associated with CPR but not with lack of RND. In SCC, PET response was associated with CPR but not RND. In these patients, reduction in SUVmax <50, 50–75 and >75% was associated with CPR rates of 29, 44 and 85% respectively (p=0.02). Conclusions: These results do not support the use of PET response to justify observation alone after CRT in esophageal AC. With respect to SCC, though exploratory, these provocative results support further study of the use of PET response to predict CPR.

[Table: see text]

No significant financial relationships to disclose.

Phase II study of induction and adjuvant bevacizumab in patients with stage IB-IIIA non-small cell lung cancer (NSCLC) receiving induction docetaxel and cisplatin

7531

Background: VEGF supports growth of NSCLC and is blocked by bevacizumab. Bevacizumab with chemotherapy is an established treatment for advanced NSCLC. We conducted this study to assess the efficacy and toxicity of induction bevacizumab (Bev) with cisplatin (C) and docetaxel (D) in patients with resectable NSCLC. Methods: All patients (pts) had resectable Stage IB-IIIA NSCLC and received up to 4 cycles of D 75 mg/m2 and C 75 mg/m2 followed by surgery. Pts with non-squamous NSCLC also received 3 cycles of Bev 15 mg/kg along with induction DC (Bev-DC). Bev was not given in pre-operative cycle 4. Pts with squamous or central tumors or hemoptysis received 4 cycles of DC only (DC). All resected pts were eligible for adjuvant Bev 15 mg/kg q3 weeks for 1 year. The primary endpoint was rate of downstaging from pre-operative clinical stage to pathologic stage. Results: From Aug 2005 - Nov 2008, 47 pts were enrolled: 27 women; median age 62; Stage 1B -15%, Stage II-17%, Stage IIIA-68%. Of 36 pts given Bev-DC, 15/30 (50%) who have completed surgery were downstaged. Of 11 pts given DC, 3/11 (27%) were downstaged. In Bev-DC group, 22/33 (67%) completed all 4 cycles of DC without a dose reduction and 28/33 pts (85%) completed all 3 planned cycles of Bev. Pts received <3 cycles due to: hemoptysis (n=3), consent withdrawal (n=1), hypertension (n=1), and sepsis (n=1). For DC, 6/11 pts (55%) completed all 4 cycles of DC without a dose reduction. 42/43 patients were resected. R0 resection rate: 29/31 for Bev-DC and 9/11 for DC. Grade 3/4 surgical complications were seen in 5/31 pts (16%) in Bev-DC and 1/11 (9%) in DC. Only 17/41 (41%) who completed surgery received adjuvant bev (median 6 cycles), and 7/41 (17%) completed all adjuvant Bev. Pts did not receive adjuvant Bev due to: POD prior to adjuvant Bev (n=9), POD during adjuvant Bev (n=6), toxicity (n=4), consent withdrawal (n=1), RT for unresectable disease (n=1), surgical complications (n=4), chemotherapy complications (n=4). No treatment related deaths. Conclusions: Induction chemotherapy with Bev-DC in NSCLC is feasible and is associated with an improved rate of downstaging compared with historical controls (33%). In patients treated with DC-Bev, grade 3/4 surgical complications were increased by 7%.

[Table: see text]

Minimal residual local disease predicts improved survival after chemoradiotherapy in patients with squamous cell carcinoma of the esophagus

15070

Background: Our recent analyses (JCO, in press) showed that residual nodal disease but not T-stage predicted survival after chemo-radiotherapy (CRT) and surgery for esophageal adenocarcinoma (AC). In this study, we investigated prognostic factors for esophageal squamous cell carcinoma (SCC) after CRT. Methods: Retrospective review of patients with esophageal SCC who had CRT and esophagectomy. Data collected: demographics, CRT details, pathologic findings, and survival. Statistical methods included recursive partitioning (RP) and Kaplan-Meier (KM) analyses. Results: Patients included in the study were treated between 1/1996 and 2/2006. Follow up was thru 5/06. 91 patients were appropriate for analysis. There were 56 men (61.5%) and 35 women (38.5%). 72 (79.1%) patients had clinical regional disease prior to treatment, while the rest had locally advanced disease. Median radiation dose was 5040 cGy, and 78 (85.7%) patients received cisplatin based chemotherapy. 49 (53.8%) patients had a complete local pathologic response (pCR), including 10/91 (10.9%) who had a pCR with residual nodal disease. 42 (46.2%) patients had residual local disease. RP analysis identified 3 prognostic groups: a) Group 1 (n=52), patients with minimal residual local disease (pCR & T1- regardless of nodal status), b) Group 2 (n=28), patients with residual T2 disease (N0 and N1) as well as patients with T3–4N0 disease, and c) Group 3 (n=11), patients with residual T3–4N1 disease. 3-year survival by KM analysis was 68.4% in group 1, 45.6% in group 2, and 0 % in group 3 (p<0.001). Conclusions: Unlike adenocarcinoma of the esophagus where residual nodal disease after CRT is the most significant predictor of survival, in SCC of the esophagus, the presence of minimal residual local disease after CRT, regardless of nodal status, predicts the best survival. The implications of these findings might include establishing different endpoints to assess response to treatment and prognostic criteria.

No significant financial relationships to disclose.

Female gender is an independent predictor of superior perioperative survival following non-small cell lung cancer (NSCLC) resection

7150

Background: Selecting patients who will benefit from surgical resection of NSCLC, especially following induction therapy, can be challenging. We analyzed our modern experience to determine factors associated with superior operative outcomes. Methods: A retrospective review of a prospectively maintained database of consecutive NSCLC patients who underwent surgical exploration for resection from January 1999 through June 2005 was performed. Factors evaluated included age, sex, co-morbidities, use of induction therapy, previous thoracic operations, extent of lung resection, and in-hospital mortality. Associations between categorical variables were evaluated with a chi-squared test and for continuous variables with an analysis of variance. Multivariate analysis was done with a logistic regression model. Results: A total of 2,524 patients underwent attempt at surgical resection, 1,398 of whom were female (55.4%). One hundred and four (4.1%) patients had exploration only. The majority of patients underwent lobectomy (1,692/2,524, 67.1%), and 6.1% (154/2,524) had a pneumonectomy. Induction therapy was given in 492 patients (19.5%). Univariate analysis showed that male gender (2.4% vs 0.93%, respectively, p = 0.003), presence of cardiac co-morbidity (p = 0.05), poor diffusion capacity (p = 0.003), and greater extent of resection (p = 0.01) were associated with increased in-hospital mortality. Multivariate analysis controlling for gender, age, diffusion capacity, cardiac, and diabetic co-morbidity, as well as prior lung cancer operation, extent of resection and use of induction therapy demonstrated that increased age, decreased diffusion capacity and greater anatomic lung resection were risk factors for higher in-hospital mortality. However, female sex was an independent predictor of lower in-hospital mortality (OR 0.41, p = 0.01). Conclusions: In addition to other previously described predictors of poor surgical outcome, such as advanced age, poor lung function and greater extent of lung removal, we observed that female gender appears to be associated with better in-hospital survival following surgical resection of NSCLC.

No significant financial relationships to disclose.

A phase II trial of surgical resection and adjuvant high-dose hemithoracic radiation for malignant pleural mesothelioma

BACKGROUND

Surgical resection of malignant pleural mesothelioma is reported to have up to an 80% rate of local recurrence. We performed a phase II trial of high-dose hemithoracic radiation after complete resection to determine feasibility and to estimate rates of local recurrence and survival.

METHODS

Patients were eligible if they had a resectable tumor, as determined by computed tomographic scanning, and adequate cardiopulmonary function for extrapleural pneumonectomy or pleurectomy/decortication. After complete resection, patients received hemithoracic radiation (54 Gy) and then were followed up with serial computed tomographic scanning.

RESULTS

From 1995 to 1998, 88 patients (73 men and 15 women; median age, 62.5 years) were entered into the study. The operations performed included 62 extrapleural pneumonectomies (70%) and 5 pleurectomies/decortications; procedures for exploration only were performed in 21 patients. Seven (7.9%) patients died postoperatively. Adjuvant radiation administered to 57 patients (54 undergoing extrapleural pneumonectomy and 3 undergoing pleurectomy/decortication) at a median dose of 54 Gy was well tolerated (grade 0-2 fatigue, esophagitis), except for one late esophageal fistula. The median survival was 33.8 months for stage I and II tumors but only 10 months for stage III and IV tumors (P =.04). For the patients undergoing extrapleural pneumonectomy, the sites of recurrence were locoregional in 2, locoregional and distant in 5, and distant only in 30.

CONCLUSION

Hemithoracic radiation after complete surgical resection at a dose not previously reported is feasible. This approach dramatically reduces local recurrence and is associated with prolonged survival for early-stage tumors. Stage III disease has a high risk of early distant relapse and should be considered for trials of systemic therapy added to this regimen of resection and radiation.

Morbidity and mortality after neoadjuvant therapy for lung cancer: the risks of right pneumonectomy

BACKGROUND

The risks of complications in patients undergoing thoracotomy after neoadjuvant therapy for nonsmall cell lung cancer remain controversial. We reviewed our experience to define it further.

METHODS

All patients undergoing thoracotomy after induction chemotherapy from 1993 through 1999 were reviewed. Univariate and multivariate methods for logistic regression model were used to identify predictors of adverse events.

RESULTS

Induction chemotherapy included mitomycin, vinblastine, and cisplatin (179 patients), carboplatin and paclitaxel (152 patients), and other combinations (139 patients). Eighty-five patients (18%) received preoperative radiation. Operations were pneumonectomy (97 patients), lobectomy (297 patients), lesser resection (18 patients), and exploration only (58 patients). Total mortality was 7 of 297 (2.4%) and 11 of 97 (11.3%) for all lobectomies and pneumonectomies, respectively, but mortality was 11 of 46 (23.9%) for right pneumonectomy. Complications developed in 179 patients (38%). By multiple regression analysis, right pneumonectomy (p = 0.02), blood loss (p = 0.01), and forced expiratory volume in one second (percent predicted) (p = 0.01) predicted complications. No factor emerged to explain this high right pneumonectomy mortality rate.

CONCLUSIONS

Pulmonary resection after neoadjuvant therapy is associated with acceptable overall morbidity and mortality. However, right pneumonectomy is associated with a significantly increased risk and should be performed only in selected patients.

Effects of diltiazem prophylaxis on the incidence and clinical outcome of atrial arrhythmias after thoracic surgery

OBJECTIVES

We sought to determine whether early prophylaxis with an L -type calcium channel blocker reduces the incidence and morbidity associated with atrial fibrillation/flutter and supraventricular tachyarrhythmia after major thoracic operations.

METHODS

In this randomized, double-blind, placebo-controlled study, 330 patients were given either intravenous diltiazem (n = 167) or placebo (n = 163) immediately after lobectomy (> or =60 years) or pneumonectomy (> or =18 years) and orally thereafter for 14 days. The primary end point with respect to efficacy was a sustained (> or =15 minutes) or clinically significant atrial arrhythmia during treatment.

RESULTS

Postoperative atrial arrhythmias (atrial fibrillation/flutter = 60; supraventricular tachyarrhythmias = 5) occurred in 25 (15%) of the 167 patients in the diltiazem group and 40 (25%) of the 163 patients in the placebo group (P = .03). When compared with placebo, diltiazem nearly halved the incidence of clinically significant arrhythmias (17/167 [10%] vs. 31/163 [19%], P = .02). The 2 groups did not differ in the incidence of other major postoperative complications or overall duration or costs of hospitalization. No serious adverse effects caused by diltiazem were seen.

CONCLUSIONS

After major thoracic operations, prophylactic diltiazem reduced the incidence of clinically significant atrial arrhythmias in patients considered at high risk for this complication.

Factors determining outcome after surgical resection of T3 and T4 lung cancers of the superior sulcus

BACKGROUND

The treatment of superior sulcus lung cancers is evolving and preoperative chemotherapy is increasingly used. To establish a historical benchmark against which new therapies can be assessed, we reviewed our 24-year experience with patients undergoing thoracotomy for lung cancers of the superior sulcus.

METHODS

Data were acquired through retrospective chart review. Overall survival was calculated by the method of Kaplan and Meier, and prognostic factors were examined by log rank and Cox proportional hazards modeling.

RESULTS

From 1974 to 1998, 225 patients underwent thoracotomy. The patients included 144 men (64%) and 81 women with a median age of 55 years. The majority of patients (55%) received preoperative radiation, but 35% did not have any preoperative treatment. Tumor stages were IIB (T3 N0) in 52%, IIIA in 15%, and IIIB in 27% of patients. Complete resection was achieved in 64% of T3 N0 tumors, 54% of T3 N2 tumors, and 39% of T4 N0 tumors. Operative mortality was 4%. Median survival was 33 months for stage IIB and 12 months for both stages IIIA and IIIB. Actuarial 5-year survivals were 46% for stage IIB, 0% for stage IIIA, and 13% for stage IIIB. By univariate and multivariable analyses, T and N status and complete resection had a significant impact on survival. Locoregional disease was the most common form of relapse.

CONCLUSIONS

Our results provide a benchmark against which new treatment regimens can be evaluated. Control of locoregional disease remains the major challenge in treating lung cancers of the superior sulcus. The potential benefit of preoperative chemotherapy or chemoradiotherapy must be assessed by whether it leads to higher rates of complete resection and a lower risk of local relapse.

Extent of chest wall invasion and survival in patients with lung cancer

BACKGROUND

The long-term survival after operation of patients with lung cancer involving the chest wall is known to be related to regional nodal involvement and completeness of resection, but it is not known whether the depth of chest wall involvement or the type of resection (extrapleural or en bloc) affects either the rate of local recurrence or survival.

METHODS

We retrospectively reviewed the Memorial Sloan-Kettering Cancer Center experience between 1974 and 1993 of 334 patients undergoing surgical exploration for lung cancer involving the chest wall or parietal pleura.

RESULTS

Of 334 patients who underwent exploration, 175 had apparently complete (R0) resections, 94 had incomplete (R1 or R2) resections, and 65 underwent exploration without resection. The overall 5-year survival of R0 patients was 32%, of R1 or R2 patients 4%, and of patients undergoing exploration without resection 0%. In the patients undergoing R0 resections, the extent of chest wall involvement was limited to the parietal pleura in 80 patients, and extended into the ribs or soft tissues in 95. The 5-year survival of R0 patients with T3 N0 M0 disease was 49%, T3 N1 M0 disease 27%, and T3 N2 M0 disease 15% (p < 0.0003). Independent of lymph node involvement, a survival advantage was observed in R0 patients if the chest wall involvement was limited to parietal pleura only, rather than invading into the chest wall musculature or ribs.

CONCLUSIONS

Survival of patients with lung cancer invading the chest wall after resection with curative intent is highly dependent on the extent of nodal involvement and the completeness of resection, and much less so on the depth of chest wall invasion.

Factors influencing ten-year survival in resected stages I to IIIa non-small cell lung cancer

OBJECTIVE

The purpose of this study was to determine (in survivors of 5 years after resection of their lung cancer) whether age, sex, histologic condition, and age have any influence on furthering survival beyond 5 years.

METHODS

From 1973 to 1989, 686 patients were alive and well 5 years after complete resection of their lung cancers. Survival analysis was carried out with only deaths from lung cancer treated as deaths. Deaths from other causes were treated as withdrawals. Multivariate Cox regression was used to test the relationship of survival to age, sex, histologic condition, and stage.

RESULTS

The population in this study had the following characteristics at the time of operation: The male/female ratio was 1.38:1, and the median age was 61 years. The histologic condition of their lung cancer was adenocarcinoma in 412 patients, squamous cell in 244 patients, large cell carcinoma in 29 patients, and small cell carcinoma in 1 patient. The stage of the disease was stage IA in 263 patients, IB in 261 patients, IIA in 12 patients, IIB in 68 patients, and IIIA in 82 patients. The extent of resection was a lobectomy or bilobectomy in 579 patients, pneumonectomy in 55 patients, and wedge resection or segmentectomy in 52 patients. A recurrence or a new lung primary occurrence was considered as failure to remain free of lung cancer. The median follow-up on all patients was 122 months from initial treatment. Of the 686 patients, 26 patients experienced the development of late recurrence and 36 new cancers, beyond 5 years. Overall survival for 5 additional years after a 5-year check point was 92.4%. Likewise, survival by nodal status was 93% for N0 tumors, 95% for N1 tumors, and 90% for N2 tumors. Survival by stage was 93% for stage I tumors and 91% for stage II or IIIA tumors.

CONCLUSIONS

In patients with surgically treated lung cancer, neither age, sex, histologic condition, nor stage is a predictor of the risk of late recurrence or new lung cancer. The only prognostic factor appears to be the survival of the patient free of lung cancer for 5 years from the initial treatment, with a resultant favorable outlook to remain well for 10 or more years.

Pulmonary metastasectomy for testicular germ cell tumors: a 28-year experience

BACKGROUND

The role of surgery in patients with pulmonary metastatic germ cell tumors has been evolving since the 1970s. To evaluate the results of pulmonary resection, we reviewed our 28-year experience.

METHODS

Between July 1967 and May 1995, 157 patients with testicular germ cell tumors underwent pulmonary resections for suspected metastases. Their clinical and pathological data were reviewed. Kaplan-Meier and Cox regression models were used to analyze prognostic factors for survival after resection of metastatic disease.

RESULTS

All patients were male with median age of 27 years (range 15-65). Complete resection was accomplished in 155 (99%) patients. Viable carcinoma was present in 44% (70) of the patients. Forty-one (26%) patients had metastases to other sites after pulmonary metastasectomy. The overall actuarial survival 5 years after pulmonary resection was 68% for the entire group and 82% for patients diagnosed after 1985. On multivariate analysis, the adverse prognostic factors were metastases to nonpulmonary visceral sites (p = 0.0069) and the presence of viable carcinoma in the resected specimen (p < 0.0001).

CONCLUSIONS

With current chemotherapy regimens, almost 85% of the patients with testicular germ cell tumors undergoing complete resection of their pulmonary metastases can be expected to achieve long-term survival.

Lobectomy improves ventilatory function in selected patients with severe COPD

BACKGROUND

Patients often undergo limited resection instead of lobectomy for non-small cell lung cancer because of a low preoperative forced expiratory volume in 1 second (FEV1). Our goal is to define criteria that will preoperatively identify a group of patients who will not lose further function after lobectomy.

METHODS

Patients who underwent lobectomy with a preoperative FEV1 of less than 80% of predicted were retrospectively identified. Data collected included preoperative and postoperative pulmonary function tests, age, sex, the lobe resected, and preoperative ventilation-perfusion scan result.

RESULTS

Thirty-two patients were included in this study. The median preoperative FEV1 was 60% of predicted (1.65 L) and the mean change in FEV1 was a loss of 7.8% after lobectomy. The patients were divided into two groups. Group 1 (n = 13) had a preoperative FEV1 of less than or equal to 60% of predicted (median, 49%; 1.35 L) combined with an FEV1 to forced vital capacity ratio of less than or equal to 0.6. Group 2 (n = 19) includes all other patients (median preoperative FEV1, 69% of predicted; 1.87 L). The mean changes in FEV1 after lobectomy were +3.7% and -15.7% for groups 1 and 2, respectively (p < 0.005). A chronic obstructive pulmonary disease index was defined and then calculated for each patient. The relationship between this index and the change in FEV1 after lobectomy for all 32 patients appears linear (r = -0.43; p = 0.015).

CONCLUSIONS

Patients with a very low preoperative FEV1 and FEV1 to forced vital capacity ratio are less likely to lose ventilatory function after lobectomy and may actually improve it.

Proposed revision of the staging classification for esophageal cancer

OBJECTIVES

This study analyzed survival with respect to lymph node involvement to develop a new staging system for patients with esophageal cancer that accurately reflects prognosis.

METHODS

The records of patients undergoing resection of primary esophageal cancer from 1989 to 1993 were reviewed. The data collected included patient age and sex, tumor histologic characteristics and location, the use of preoperative or postoperative radiation and chemotherapy, the type of resection, the depth of tumor invasion, the number and location of benign and malignant lymph nodes in the resected specimen, the disease status at last follow-up, and the first site of relapse. With an anatomically specific lymph node map, tumors designated in the current American Joint Committee on Cancer system as M1 because of extensive lymph node metastases were reclassified as N2, reserving the M1 category for visceral metastases. Survival was analyzed by the Kaplan-Meier method, and prognostic factors were assessed by log-rank and Cox regression analyses.

RESULTS

There were 216 patients (159 men, 57 women) with a median age of 63.5 years. Adenocarcinoma of the distal esophagus or gastroesophageal junction was the most common tumor (127 patients, 59%) and Ivor Lewis esophagogastrectomy was the most frequently performed operation. Both lymph node location (N1 versus N2) and number (0 vs 1 to 3 vs 4 or more) significantly influenced survival.

CONCLUSIONS

A new staging system that adds an N2 M0 descriptor and reclassifies stage groupings reflects prognosis more accurately than does the current American Joint Committee on Cancer staging system. The number of positive lymph nodes is also an important stratification factor.

Primary sarcomas of the mediastinum: results of therapy

Primary sarcomas of the mediastinum are rare, and data concerning treatment and results of therapy are sparse.

OBJECTIVE

To assess presentation, management, prognostic factors, and survival in mediastinal sarcomas.

METHODS

We reviewed our experience with 47 patients with the diagnosis of primary sarcoma of the mediastinum. Data were collected from a computerized institutional database and medical records. Survival was analyzed by Kaplan-Meier method and comparisons of survival by log rank test.

RESULTS

The median age of 47 patients with mediastinal sarcoma was 39 years (range 2.5 to 69 years), with a male/female ratio of 1.6. The most common complaints were chest/shoulder pain (38%) and dyspnea (23%). The most common tumor types were malignant peripheral nerve tumor (26%), spindle cell sarcoma (15%), leiomyosarcoma (9%), and liposarcoma (9%). Operation was the primary treatment modality in 72% of cases (n = 34); 22 sarcomas (47%) were completely resected. The overall 5-year survival was 32%. High-grade lesions had a significantly decreased survival (5-year survival = 27%) compared with low-grade tumors (5-year survival = 66%) (p = 0.05). The overwhelming factor determining survival was the ability to completely resect the tumors (5-year survival 49% for complete resection; 3-year survival 18% for incomplete or no resection) (p = 0.0016). Despite complete resection, local recurrence occurred in 64% of cases.

CONCLUSION

Because the overall survival for patients with mediastinal sarcomas is 32% and the local recurrence is 64% for tumors completely resected, aggressive adjuvant therapy should continue to be systematically explored.

Thymic carcinoma: current staging does not predict prognosis

BACKGROUND

Thymic carcinomas are currently staged by Masaoka classification, a staging system for thymomas. We retrospectively evaluated surgical patients with thymic carcinoma to determine prognostic factors and to evaluate the usefulness of Masaoka staging in this disease.

METHODS

Our computerized tumor registry yielded 118 patients with thymoma. Review of pathologic material revealed 43 cases of thymic carcinoma. Collection of data was by review of hospital and physician charts and telephone contact with patients. Analysis of prognostic factors was performed in patients undergoing complete resection by the method of Kaplan-Meier and Cox proportional hazards regression.

RESULTS

Between 1949 and 1993, 43 patients underwent surgery for thymic carcinoma. Overall survival was 65% at 5 years and 35% at 10 years. Overall recurrence was 65% at 5 years and 75% at 10 years. On univariate analysis, survival was not dependent on age, sex, tumor size, or Masaoka stage but was dependent on innominate vessel invasion. By multivariate analysis, survival was dependent only on innominate vessel invasion.

CONCLUSIONS

Patients with thymic carcinoma have a high rate of recurrence. Tumor invasion of the innominate vessels is associated with a particularly poor prognosis. Although Masaoka staging is useful in staging patients with thymoma, it does not appear to predict outcome for patients with thymic carcinoma.

Radiographic appearances of esophageal stents

Among patients with esophageal carcinoma and associated dysphagia, more than 60% have unresectable disease at presentation. In such cases, the goal of treatment is primarily palliation of the dysphagia; treatment options include surgery, radiation therapy, laser ablation, and placement of stents. Beginning with the first stent-made of boxwood and silver in 1885-evolution in design led to the creation of large-diameter, rigid plastic stents placed at laparotomy and eventually placed by means of endoscopy. However, complications such as perforation, hemorrhage, dislodgment, pressure necrosis, and occlusion were frequently encountered with these stents. The development of small-diameter, expandable metal stents eliminated some of these complications. Metal stents have greatly reduced procedure-related morbidity and mortality, but complications such as perforation, malposition, migration, tumor ingrowth and overgrowth, food obstruction, and tracheoesophageal fistula persist. An efficacious and increasingly used method of treating malignant dysphagia, esophageal stent placement must undergo further improvements to reduce the frequency of complications, particularly migration and tumor ingrowth.

Complications of abdominal right-thoracic (Ivor Lewis) esophagectomy

Esophageal resection and reconstruction are associated with significant operative and postoperative morbidity and mortality. Careful evaluation of the patient's cardiopulmonary status; proper preparation of the patient with smoking cessation, exercise, and cardiopulmonary rehabilitation; assessment of the stage of disease; selection of a suitable operative technique; and meticulous attention to technical details help reduce the incidence of complications and ensure a successful outcome in this technically challenging procedure.

Herniation of the transverse colon after esophagectomy: is retrocardiac air a normal postoperative finding?

OBJECTIVE

This study describes the radiographic and CT imaging features of colonic herniation complicating esophagogastrectomy in three patients, all of whom required surgery.

CONCLUSION

After routine esophagogastrectomy with gastric pull-through for esophageal carcinoma, the intrathoracic gastric pull-through most often lies in a right paraspinal location. Once a left-sided gastric pull-through has been excluded, radiologists must be suspicious for left retrocardiac air lucency that possibly represents herniation of the nearby colon. Early diagnosis is important for the prevention of bowel strangulation, especially of the ischemia-susceptible splenic flexure.

Effects of diltiazem versus digoxin on dysrhythmias and cardiac function after pneumonectomy

BACKGROUND

This prospective study was designed to determine whether diltiazem is superior to digoxin for the prophylaxis of supraventricular dysrhythmias (SVD) after pneumonectomy or extrapleural pneumonectomy (EPP) and to assess the influence of these drugs on perioperative cardiac function.

METHODS

Seventy consecutive patients without previous SVD were randomly allocated immediately after pneumonectomy or EPP to receive diltiazem (n = 35) or digoxin (n = 35). Diltiazem-treated patients received a slow intravenous loading dose of 20 mg, followed by 10 mg intravenously every 4 hours for 24 to 36 hours, then 180 to 240 mg orally daily for 1 month. Digoxin-treated patients received a 1-mg intravenous loading in the first 24 to 36 hours, then 0.125 to 0.25 mg orally daily for 1 month. A concurrent prospective cohort of 40 patients without previous SVD, who did not participate in the study and underwent pneumonectomy or EPP without prophylaxis, served as a comparison group for SVD occurrence. Serial Doppler echocardiograms were performed to assess cardiac function and all patients were continuously monitored with Holter recorders for 3 days. Data were analyzed by intent-to-treat.

RESULTS

In patients undergoing standard or intrapericardial pneumonectomy, diltiazem prevented the overall incidence of postoperative SVD when compared with digoxin, 0 of 21 patients versus 8 of 25 patients, respectively, p < 0.005. When EPP patients were included in the analysis, diltiazem decreased the incidence of all SVD from 11 of 35 patients (31%) to 5 of 35 patients (14%) when compared with digoxin, p = 0.09. Digoxin-treated patients had a similar incidence of all SVD (31%) as concurrent controls (11 of 40 patients [28%]). The two treated groups did not differ in right or left atrial size, left ventricular ejection fraction, or right heart pressure. When all patients were combined, those in whom SVD developed were significantly older (65 +/- 12 years versus 55 +/- 11 years, p = 0.004) and had a longer median hospital stay (9 versus 6 days, p = 0.03), when compared with those in whom SVD did not develop, respectively. The subset of patients undergoing EPP had a greater incidence of atrial fibrillation and electrocardiographic changes suggestive of postoperative pericarditis than all other pneumonectomy patients.

CONCLUSIONS

Diltiazem was both safe and more effective than digoxin in reducing the overall incidence of SVD after standard or intrapericardial pneumonectomy. Digoxin therapy had no effect on the incidence of postoperative SVD and is not recommended for prophylaxis of SVD. Dysrhythmias after pneumonectomy or EPP occur in older patients and are associated with a greater length of hospital stay.

Barium sulfate: a new (old) contrast agent for diagnosis of postoperative esophageal leaks

PURPOSE

To determine whether esophagography with use of barium alone is safe for detection of postoperative leaks of the esophagus.

MATERIALS AND METHODS

A review was performed of 29 postoperative esophagograms that showed 29 leaks in 12 patients. All studies were performed with a 50% dilution of barium sulfate. Leak volumes were calculated as the product (in cubic centimeters) of the length, width, and height. The safety and efficacy of barium were determined on the basis of development of mediastinitis and retention of barium in the mediastinum that would interfere with subsequent patient care.

RESULTS

In 11 of the 12 patients, follow-up studies were performed 4-48 days (mean, 10.2 days) after diagnosis. Leaks were 0.25-375 cm3 (mean, 31.4 cm3). In five of 17 (29%) follow-up procedures, small amounts of residual barium were detectable on the scout radiograph, none of which interfered with interpretation of the new images. During a follow-up period of 7-448 days (mean, 226 days), no cases of mediastinitis were found.

CONCLUSION

Esophagography can be performed safely with barium to rule out an anastomotic esophageal leak.

Role of video-assisted thoracic surgery in the treatment of pulmonary metastases: results of a prospective trial

BACKGROUND

A retrospective review revealed a 42% error rate between computed tomographic scan reports and thoracotomy findings; therefore, a prospective study was designed to compare the value of computed tomographic scans, video-assisted thoracoscopic exploration, and open thoracotomy in the management of pulmonary metastases.

METHODS

Eligibility included any patient with only one or two ipsilateral pulmonary metastases identified on computed tomographic scan who was being considered for surgical resection. Initially video-assisted thoracic surgery was performed and all lesions identified were resected. A thoracotomy adequate for complete lung palpation was then carried out and any additional lesions found were removed.

RESULTS

Eighteen patients of a planned 50 were treated before closure of the study. Four patients (22%) had no additional lesions found at thoracotomy. The primary sites of tumor were colon (10), breast (3), and one patient each skin (squamous), cervix, kidney, melanoma, and sarcoma. Four patients (22%) did have additional lesions at thoracotomy, which were benign. In the remaining 10 patients (56%) additional malignant lesions were found at thoracotomy after video-assisted thoracoscopic exploration. After 18 patients were entered, analysis of the early results disclosed a 56% failure rate of a computed tomographic scan and video-assisted thoracic surgery to detect all lesions. Being within the 95% confidence interval (32% to 78%), the study was abandoned.

CONCLUSIONS

We conclude that video-assisted thoracic surgery should be used only as a diagnostic tool in managing lung metastasis. A thoracotomy is required to achieve complete resection, which is the major survival prognosticator for satisfactory long-term results.

Symptomatic tachydysrhythmias after esophagectomy: incidence and outcome measures

BACKGROUND

Supraventricular tachydysrhythmias (SVT) after esophageal operations for carcinoma occur frequently and may be associated with increased morbidity. Prospective data on the etiology, incidence, and importance of these dysrhythmias are sparse.

METHODS

In 100 consecutive patients undergoing esophagectomy without prior history of atrial dysrhythmias or receiving antiarrhythmics, we prospectively examined the effects of predefined risk factors by history and pulmonary function on the 30-day incidence of symptomatic postoperative SVT, need for intensive care unit admission, and mortality rate.

RESULTS

Symptomatic postoperative SVT occurred in 13 (13%) of the 100 patients studied at a median of 3 days after operation and was accompanied by hypotension in 9/13 (69%). Univariate correlates of SVT were older age (p = 0.03), perioperative use of theophylline (p = 0.044), and a low carbon monoxide diffusion capacity (measured in 56% of patients) on preoperative pulmonary function. Patients in whom SVT developed had a higher rate of intensive care unit admission (p = 0.0001) and a longer hospital stay (p = 0.036). Although patients in whom SVT developed had a higher (p = 0.013) 30-day mortality rate, SVT was not the direct cause of death.

CONCLUSIONS

These prospective data show that the true incidence of symptomatic SVT within 30 days of esophagectomy is lower than previously reported. Occurrence of SVT was associated with significant morbidity. Older age was the strongest predictor of SVT after esophagectomy. In high-risk patients, continued monitoring (48 to 72 hours) and early interventions to decrease the incidence of postoperative SVT may improve overall surgical outcomes.

Erythromycin stimulates gastric emptying after esophagectomy with gastric replacement: a randomized clinical trial

Delayed gastric emptying after esophagogastrectomy can pose a significant early postoperative problem. Because erythromycin, which stimulates the gastric antral and duodenal motilin receptor, has been shown to significantly increase gastric emptying in patients with diabetic gastroparesis, we decided to evaluate its effect on gastric emptying after esophagogastrectomy.

METHODS

Twenty-four patients (18 men and six women, age range 41 to 79 years, median 66 years) were randomized to receive either erythromycin lactobionate (200 mg in 50 ml normal saline solution intravenously) (n = 13) or placebo (50 ml normal saline solution intravenously (n = 11) 11 days after esophagogastrectomy (with pyloric drainage procedure). After erythromycin or placebo had been infused over a 15-minute period, patients ingested a solid meal (scrambled egg with bread) labeled with technetium 99m sulfur colloid (500 microCi) over approximately 15 minutes. Dynamic images of the stomach were then acquired over 90 minutes in the supine position by gamma imaging. Results were expressed as percentage of counts retained in the stomach (percent gastric retention) over time.

RESULTS

There were no side effects of erythromycin. In the placebo group, the mean percent of radiolabeled meal retained in the stomach after 90 minutes was 88%, which was significantly greater than in the erythromycin group, 37% (p < 0.001). In addition, analysis of covariance demonstrated that the rate of gastric emptying (slope of the line) was significantly greater in the erythromycin-treated group than in the placebo group (p < 0.0001).

CONCLUSION

Early satiety after esophagogastrectomy may be due to delayed gastric emptying and not due to a decrease in the gastric reservoir. Intravenous erythromycin significantly improves gastric emptying in patients after esophagogastrectomy by stimulating gastric motility.

Vocal cord medialization for unilateral paralysis associated with intrathoracic malignancies

Patients with unilateral vocal cord paralysis from intrathoracic malignancies may have significant dysfunctions of speech, swallowing, ventilation, and effective coughing as a result of inadequate compensation of the nonparalyzed cord. In patients with already compromised pulmonary function, aspiration can be a life-threatening event. Sixty-three patients with intrathoracic malignancies required surgical correction of vocal cord paralysis. Primary pathology included lung cancer (49), esophageal cancer (nine), and miscellaneous tumors (five). Symptoms included hoarseness (62), dyspnea (21), aspiration (26), weight loss (19), dysphagia (14), and pneumonia (14). The surgical procedures included medial displacement of the vocal cord with silicone elastomer (48), temporary Gelfoam injection (seven), and Teflon (polytetrafluoroethylene) injection (eight) to move the affected cord to a medial position. In 11 patients, the operation was performed in the acute postoperative setting to improve pulmonary toilet. Symptomatic improvement was noted in the following proportions of affected patients: hoarseness, 92%; dyspnea, 90%; dysphagia, 93%; aspiration, 92%; pneumonia, 93%; and weight loss, 47%. Overall success rate of the intervention was 57 of 63 patients (90%). All 11 patients treated in the acute setting had immediate improvement. A variety of complications occurred in 17% of patients. Surgical management of vocal cord paralysis in patients with intrathoracic malignancies prevents life-threatening pulmonary complications in the acute postoperative setting. In chronic situations, it provides patients with improved speech, swallowing, and pulmonary function, resulting in improved quality of life, even for patients not cured of their disease.

Predictors of survival in malignant tumors of the sternum

From 1930 to 1994, 54 patients with primary malignant tumors of the sternum were seen. Fifty patients were first seen with a mass, and one half of them also had pain in the sternal region. Two patients had no symptoms at presentation. Among 39 solid tumors were 26 chondrosarcomas, 10 osteosarcomas, 1 fibrosarcoma, 1 angiosarcoma, and 1 malignant fibrous histiocytoma. Of these, 25 were low-grade and 14 were high-grade tumors. Among 15 small cell tumors were 8 plasmacytomas, 6 malignant lymphomas, and 1 Ewing's sarcoma. Partial or subtotal sternectomy was done in 37 patients and total sternectomy in 3. Of the remaining 14 patients, 3 had local excision; 10 had external radiation or chemotherapy without operation, or both; and 1 had no treatment. All but one patient treated by wide resection (N = 40) had some form of skeletal reconstruction of the chest wall defect. Thirty-one (78%) underwent repair with Marlex mesh, and in 25 this was combined with methyl methacrylate. The skin edges were closed per primum in 32 patients; 8 required muscle, omentum, or skin flaps. Resection in chondrosarcomas yielded a 5-year survival (Kaplan-Meier) of 80% (median follow-up, 17 years). The 5-year survival in osteosarcomas was 14%. Resection was curative in 64% of low-grade sarcomas but in only 7% of high-grade sarcomas. In small cell tumors, resection and radiation were helpful for local control; all failures were a result of distant metastases. We conclude that primary sarcomas of the sternum though uncommon are potentially curable by wide surgical excision. With rigid prostheses to repair the skeletal defects, the surgical complication rates are low. Overall survival after complete surgical resection is related to tumor histologic type and grade.

Impact of clinicopathologic parameters on patient survival in carcinoma of the cervical esophagus

BACKGROUND

The survival of patients with carcinoma of the cervical esophagus remains poor in spite of multimodality treatment and technical improvements in surgical resection and reconstruction. This study was undertaken to update our experience with cervical esophageal carcinoma and to identify factors that had an impact on patient survival and quality of life.

PATIENTS AND METHODS

Clinical data encompassing 132 variables were collected on 67 patients with cervical esophageal carcinoma from 1980 to 1993. Statistical analysis was performed: independent Student's t-tests, Cox regression, Kaplan-Meier curves, and log rank analyses were used in the statistical evaluation. The mean age of the patients was 63 years (range 31 to 88). Dysphagia was the primary symptom in 86% of patients; 80% had received no prior treatment. The most common abnormal finding (21%) on physical examination was a neck mass.

RESULTS

Curative resection was performed in 22 patients, 7 had palliative procedures, and 7 were found to be inoperable at exploration and received palliative treatment. Radiation with or without chemotherapy was definitive treatment for 10 patients, whereas 4 patients were treated with chemotherapy alone for cure, and 17 patients received palliative treatment. The mean survival following diagnosis was 17 months (range 1 to 96). Cumulative 5-year survival was 12%.

CONCLUSIONS

Persistent disease, chemotherapy prior to presentation, and chemotherapy for cure remained as statistically significant parameters associated with decreased survival by multivariate analysis. There was a trend toward improved survival in patients treated with surgical resection.

Thymoma: a multivariate analysis of factors predicting survival

BACKGROUND

Despite complete surgical excision, malignant thymomas often recur with resultant death. We reviewed our series to determine which factors independently predict survival after surgical resection.

METHODS

A retrospective analysis of patients operated on for thymoma between 1949 and 1993 at Memorial Sloan-Kettering Cancer Center was performed. Clinical data were collected from chart review. Only patients with a pathology report confirming the diagnosis of thymoma were included in this analysis. Kaplan-Meier survival curves were generated and comparisons of survival analyzed by log rank test. Multivariate analysis was performed by the Cox proportional hazard model.

RESULTS

One hundred eighteen patients with thymoma underwent operation. There were 86 complete resections (73%), 18 partial resections (15%), and 14 biopsies (12%). By Masaoka staging, 25 patients were stage I (21%), 41 stage II (35%), 43 stage III (36%), and 9 stage IVa (8%). Overall survival was 77% at 5 years and 55% at 10 years. Tumor recurred in 25 (29%) of 86 completely resected thymomas. Stage of disease (p = 0.03) was the only independent prognostic factor affecting recurrence. By multivariate analysis, stage (p = 0.003), tumor size (p = 0.0001), histology (p = 0.004), and extent of surgical resection (p = 0.0006) were independent predictors of long-term survival.

CONCLUSIONS

Patients with stage I disease require no further therapy after complete surgical resection. Neoadjuvant therapy should be considered for patients with large tumors and invasive disease.

Ivor Lewis esophagectomy

Lewis described a technique for resection of cancer of the midthoracic esophagus in a staged manner in 1946. The same procedure done at one stage has remained a standard technique for resection of a carcinoma involving the thoracic esophagus.

Echoplanar MR imaging for characterization of adrenal masses in patients with malignant neoplasms: preliminary evaluation of calculated T2 relaxation values

OBJECTIVE

We undertook this study to assess the utility of echoplanar MR imaging for distinguishing benign from malignant adrenal masses in patients with known malignant neoplasms.

MATERIALS AND METHODS

Thirty consecutive patients with 31 adrenal masses and a known malignant neoplasm underwent breath-hold echoplanar MR imaging with a repetition time of 6000 msec and four echo times (40, 80, 120, 160 msec) on a 1.5-T unit before biopsy. Subsequently, 10 masses were shown to be malignant at histologic examination, 12 masses were benign at histologic examination, and nine were thought to be benign because they had not changed in size at follow-up imaging. Mean lesion size was 2.4 +/- 2.1 cm. T2 calculations using regions of interest in the liver and adrenal mass were performed in each patient.

RESULTS

The mean calculated T2 value of benign adrenal masses was 70.3 msec (SD, 11.6 msec) versus 104.6 msec (SD, 35.2 msec) for malignant adrenal masses (p = .013). Using a cutoff T2 value of 84 msec, 19 (90%) of 21 benign masses and nine (90%) of 10 malignant masses were correctly classified. The mean adrenal/liver T2 ratio was 1.4 (SD, 0.25) for benign lesions, and 2.1 (SD, 0.78) for malignant lesions (p = .017). Using a cutoff ratio of 1.60, 19 (90%) of 21 benign lesions and eight (80%) of 10 malignant lesions would have been correctly classified.

CONCLUSION

This preliminary work suggests that obtaining T2 calculations from echoplanar MR images of adrenal masses is a useful technique for distinguishing benign from malignant adrenal masses in patients at risk for adrenal metastasis.

Lung resection for colorectal metastases. 10-year results

BACKGROUND

Metastasectomy for colorectal carcinoma to the lung is controversial. We analyzed results of this approach to justify its credibility.

METHODS

We studied 144 patients by retrospective review after complete resection of lung metastases from colorectal cancer from 1965 through 1988. Patient selection and prognostic factors influencing survival were analyzed. Survival was analyzed by the Kaplan-Meier method, and comparisons were made by log-rank analysis.

RESULTS

A total of 170 thoracotomies were performed in 144 patients. The overall 5- and 10-year survival was 40% and 30%, respectively. Those patients undergoing complete resection of their metastases survived significantly longer than those undergoing incomplete resections.

CONCLUSION

It appears that resection of pulmonary metastases from colorectal carcinoma translates into long-term survival benefit.

Incidence of local recurrence and second primary tumors in resected stage I lung cancer

From 1973 to 1985, 598 patients underwent resection for stage I non-small-cell lung cancer. There were 291 T1 lesions and 307 T2 lesions. The male/female ratio was 1.9:1. The histologic type was squamous carcinoma in 233 and nonsquamous carcinoma in 365. Lobectomy was performed in 511 patients (85%), pneumonectomy in 25 (4%), and wedge resection or segmentectomy in 62 (11%). A mediastinal lymph node dissection was carried out in 560 patients (94%) and no lymph node dissection in 38 (6%). Fourteen postoperative deaths occurred (2.3%). Ninety-nine percent of the patients were observed for a minimum of 5 years or until death with an overall median follow-up of 91 months. The overall 5- and 10-year survivals (Kaplan-Meier) were 75% and 67%, respectively. Survival in patients with T1 N0 tumors was 82% at 5 years and 74% at 10 years compared with 68% at 5 years and 60% at 10 years for patients with T2 tumors (p < 0.0004). The overall incidence of recurrence was 27% (local or regional 7%, systemic 20%) and was not influenced by histologic type. Second primary cancers developed in 206 patients (34%). Of these, 70 (34%) were second primary lung cancers. Despite complete resection, 31 of 62 patients (50%) who had wedge resection or segmentectomy had recurrence. Five- and 10-year survivals after wedge resection or segmentectomy were 59% and 35%, respectively, significantly less than survivals of those undergoing lobectomy (5 years, 77%; 10 years, 70%). The 5- and 10-year survivals in the 38 patients who had no lymph node dissection were reduced to 59% and 32%, respectively. Apart from the favorable prognosis observed in this group of patients, three facts emerge as significant: (1) Systematic lymph node dissection is necessary to ensure that the disease is accurately staged; (2) lesser resections (wedge/segment) result in high recurrence rates and reduced survival regardless of histologic type; and (3) second primary lung cancers are prevalent in long-term survivors.

Management of non-small cell lung cancer with direct mediastinal involvement

The results of surgical treatment were analyzed for 102 patients with non-small cell lung cancer invading the mediastinum by direct extension (T3 and T4), but those who had N2 disease were excluded to eliminate the adverse prognostic effect of this nodal subset. The histologic type was squamous cell carcinoma in 55 patients, adenocarcinoma in 40, and large cell carcinoma in 7. There were 58 T3 tumors invading the mediastinal pleura or fat, phrenic nerve, vagus nerve, pericardium, or pulmonary vessels and 44 T4 lesions invading the aorta, vena cava, esophagus, trachea, spine, or atrium. Resection included lobectomy (33 patients), pneumonectomy (32 patients), and limited resection (6 patients). Complete resection was possible in 46 patients and incomplete or no resection was possible in 56. The interstitial implantation of radioactive sources to control residual tumor also was undertaken in 43 patients. The operative mortality was 6%. The overall survival (Kaplan-Meier) was 19% at 5 years (median survival time, 18 months). Factors found to be significantly affect survival were complete resectability and the histologic type. With complete resection, the 5-year survival was 30% (p = 0.005). The 5-year survival in patients with adenocarcinoma or large-cell carcinoma was 30%, compared with 14% in patients with squamous cell carcinoma (p = 0.002). The extent of mediastinal involvement (T3 versus T4) influenced resectability and survival, and this approached statistical significance (p = 0.055). We conclude that most patients with non-small cell carcinoma and mediastinal invasion do poorly with primary surgical treatment.

Esophageal involvement in Wegener's granulomatosis

Wegener's granulomatosis (WG) is characterized by granulomatous vasculitis, renal disease, and upper and lower respiratory tract disease. Although most organ systems can be involved, gastrointestinal (GI) manifestations are notably uncommon. We describe a patient with WG whose presentation was unique for the prominence of odynophagia. Esophagoscopy revealed erosive esophagitis, which on biopsy was shown to be due to direct involvement by the underlying vasculitis. This is first antermortem documentation of esophageal disease secondary to WG. The GI manifestations of WG are reviewed.

Randomized trial comparing postoperative chemotherapy with vindesine and cisplatin plus thoracic irradiation with irradiation alone in stage III (N2) non-small cell lung cancer

This prospective randomized trial was performed to determine whether postoperative chemotherapy with vindesine and cisplatin could lengthen time to progression and overall survival in stage III (T1-3N2M0) non-small cell lung cancer (NSCLC) patients. Seventy-two patients were entered; 36 were randomized to receive chemotherapy. Patients were stratified by extent of resection (complete vs. incomplete) and histology (squamous vs. nonsquamous). All had surgery and mediastinal irradiation 6-7 weeks post-thoracotomy. Incompletely resected patients had intraoperative 125I and/or 192Ir implantation. Vindesine (3 mg/m2) weekly x 5, then every 2 weeks x 8, and cisplatin (120 mg/m2) days 1, 29, 71, 113 were planned for those randomized to chemotherapy. No difference in time to progression (median 9.2 months for radiation + chemotherapy vs. 9.0 months for radiation, P = 0.35) or overall survival (16.3 months for radiation + chemotherapy vs. 19.1 months for radiation, P = 0.42) was found. Postoperative vindesine and cisplatin did not prolong time to progression or survival in this population of stage III NSCLC.

The clamshell incision: an improved approach to bilateral pulmonary and mediastinal tumor

Median sternotomy has been the accepted approach for dealing with mediastinal tumors or bilateral pulmonary disease, but exposure to the lower lobes and to mediastinal tumors extensively involving a hemithorax is often limited. Based on the reported experience from double-lung transplantation, we explored the use of clamshell incisions for these difficult problems. From March 1991 to December 1993, we prospectively studied the utility of clamshell incisions in 90 patients for the following indications: bilateral pulmonary metastases (62 patients), primary lung carcinoma with mediastinal involvement (13 patients), primary tumors of the mediastinum (14 patients), and mesothelioma (1 patient). Bilateral anterior thoracotomies with a transverse sternotomy (clamshell incision) were employed in 71 patients and a unilateral anterior thoracotomy with partial or complete median sternotomy (hemiclamshell incision) was used in 19 patients. For closure, we used pericostal sutures and sternal wires, usually augmented by sternal K-wire stents or Steinmann pins to prevent sternal override. Exposure to all areas of the mediastinum, pericardium, pleura, and lung was excellent. Specifically, the clamshell incision afforded markedly better access to lower lobe disease and hemithoracic extension of mediastinal disease than that possible with median sternotomy. There were no deaths or significant morbidity, and all patients tolerated the incisions well without mechanical respiratory difficulties. There was one wound infection. There was no late sternal override and the cosmetic results were found to be excellent during a follow-up of 2 to 33 months. We conclude that clamshell incisions constitute an improved surgical approach for the management of bilateral pulmonary or combined pulmonary and mediastinal disease.

Influence of surgical resection and brachytherapy in the management of superior sulcus tumor

We analyzed the results of surgical treatment of all patients presenting with untreated superior sulcus tumors between 1974 to 1991 inclusive at our institution. Most patients received preoperative radiotherapy. We attempted to analyze the influence of surgical resection and intraoperative brachytherapy in obtaining locoregional control and disease-free survival. One hundred twenty-four patients underwent thoracotomy and 100 patients underwent resection. The overall 5-year survival rate was 26% for all patients and 30% for resected patients. Those patients receiving a complete resection achieved a 41% 5-year survival. The best single group were those patients undergoing a lobectomy (versus wedge resection) and en-bloc chest wall resection (60% 5-year survival). We were unable to demonstrate an advantage for the use of intraoperative brachytherapy in those patients with complete resection. For those patients with incomplete resection, the use of brachytherapy combined with preoperative or postoperative external radiation therapy resulted in a 9% 5-year survival. Locoregional failure was significant both in patients with complete resection and in patients with incomplete resection. Adverse prognostic factors included Horner's syndrome, N2 and N3 disease, T4 disease, and incomplete resections. In superior sulcus tumors, every attempt to completely resect the tumor by en-bloc chest wall resection combined with lobectomy and adequate nodal staging remains the surgical treatment of choice together with either preoperative, postoperative, or "sandwich" external radiation therapy.

Contribution of videothoracoscopy to the management of the cancer patient

BACKGROUND

Videothoracoscopy has rapidly become a popular procedure, but its technical feasibility has been emphasized without critically evaluating its role in the management of thoracic disease. To assess the value of videothoracoscopy in the diagnosis and staging of the cancer patient and to determine if it has added to our previous standard approach of thoracoscopy performed with a mediastinoscope without video, we established a prospective database when we initiated videothoracoscopy in January 1992.

METHODS

Patients were offered videothoracoscopy as an alternative to thoracotomy only if other standard approaches (e.g., needle biopsy, mediastinoscopy) were inadequate to diagnose or stage cancer or to restage patients after therapy. Parameters entered and analyzed in a prospective database were patient name; age; sex; past history; indications for videothoracoscopy; procedure type; surgical technique; whether conversion to thoracotomy was necessary, and if so, why; complications; and pathology. A complete case list of thoracoscopies performed in 1991 before videothoracoscopy was available provided historical comparison.

RESULTS

From January 1 to December 31, 1991, 82 patients underwent thoracoscopy using a mediastinoscope for diagnosis and therapy of pleural disease. From January 1 to July 31, 1992, 160 patients (male:female = 81:79; mean age 56 years) had videothoracoscopy; 72 of 160 patients (44%) had procedures that previously would have required thoracotomy: 69 lung wedge resections, one pericardial window, one pleurectomy, one mediastinal node sampling. No major resectional procedures (e.g., lobectomy, esophagectomy) were performed by videothoracoscopy. Twenty-two percent of all patients (35 of 160), and 23% of wedge resection patients (16 of 69) required conversion to thoracotomy because videothoracoscopy was inadequate for diagnosis or staging. Reasons for conversion (multiple reasons in five patients) included further resection required in 23 patients; inability to evaluate lesion in 11; adhesions in five; and inability to tolerate one lung ventilation in two. The chest tube was in place postoperatively for a mean of 2.3 days. Thirty-day postoperative complications included ventilation for > 48 h in one patient; prolonged air leak in one; pneumonia in one; arrhythmia in one; and death from progressive disease in two.

CONCLUSIONS

Although the role of videothoracoscopy in the treatment of primary thoracic malignancies and pulmonary metastases is still undefined, this early experience indicates that videothoracoscopy often enhances the ability to diagnose and stage patients by obviating thoracotomy.

Prospective evaluation of unilateral adrenal masses in patients with operable non-small-cell lung cancer. Impact of magnetic resonance imaging

We designed a prospective study to evaluate the accuracy of magnetic resonance imaging in distinguishing a benign from a malignant adrenal mass in patients with otherwise operable non-small-cell lung cancer.

METHODS

Potentially operable non-small-cell lung cancer was prospectively staged. If a unilateral adrenal mass was found by computed tomographic scanning, respiratory compensated and cardiac gated thin section magnetic resonance imaging of the adrenal glands was done. One radiologist interpreted the magnetic resonance imaging scan blinded and, on the basis of the relative signal strengths of the T1- and T2-weighted images, judged whether the adrenal mass was benign or malignant. The patients then underwent a percutaneous needle biopsy of the adrenal mass, if technically feasible. If the result of the needle biopsy was nondiagnostic or if the biopsy was not feasible, an adrenalectomy through a posterior approach was performed.

RESULTS

Twenty-seven patients with a unilateral adrenal mass entered the study-11 men and 16 women whose ages ranged from 42 to 75 years (median 58 years). Four patients had epidermoid and 23 adenocarcinoma of the lung. The clinical locoregional stage was I in 9, II in 1, IIIA in 16, and IIIB in 1. Twenty-five completed the magnetic resonance imaging procedure. Five adrenal masses (19%) were metastatic non-small-cell lung cancer (adenocarcinoma = 4, epidermoid = 1); 22 masses (81%) were benign (adenoma = 20, hyperplasia = 2). There were no significant differences in age, sex, histologic type, or locoregional stage between those with a benign versus a malignant mass. However, the malignant masses were significantly larger (3.8 +/- 1.9 cm; range 2.5 to 7.1; median 3.1) than the benign masses (2.0 +/- 0.4 cm, range 1.2 to 2.8; median 2.0) (p < 0.001). Among those having magnetic resonance imaging (n = 25), the technique correctly predicted a malignant mass in the four patients with a histologically confirmed metastasis from non-small-cell lung cancer. However, among the 21 histologically benign masses, the magnetic resonance imaging was interpreted as benign in 5, malignant in 14, and indeterminate in 2. Therefore, although the false-negative rate was 0%, the false-positive rate was 67%.

CONCLUSION

Most adrenal masses in patients with otherwise operable non-small-cell lung cancer are benign. Currently available magnetic resonance imaging methods cannot replace biopsy.

Treatment and prognosis in bronchial carcinoids involving regional lymph nodes

From 1953 to 1992, 25 patients were surgically treated for bronchial carcinoids with metastases to regional lymph nodes (N1 or N2). The tumors were located centrally, involving main or lobar bronchi in 12 patients and were peripheral in 13. Histologically, 12 of the carcinoids were classified as typical and 13 as atypical (neuroendocrine carcinoma). Pneumonectomy was performed in 11 patients, sleeve lobectomy in one, lobectomy in seven and bilobectomy in six. A formal mediastinal lymph node dissection was done in 20 patients. At final staging, 10 had N1 disease and 15 had N2. No adjuvant treatment was given to the 10 patients with N1 disease. External radiation therapy was given after the operation to 9 of 15 patients with N2 disease. The overall 5-year survival (Kaplan-Meier) was 75% (median 62 months). No difference in survival was found between patients with N1 or N2 disease. However, survival and recurrence rate differed between typical and atypical carcinoids. In typical carcinoids, the 5-year survival was 92% and, in atypical carcinoids, it was 60% (p = 0.02). We conclude that complete resection for bronchial carcinoids results in long-term survival despite the presence of regional lymph node metastases. Recurrence appears to depend more on cell type than nodal status. Postoperative radiation therapy does not appear to be beneficial.

Accuracy of lung imaging in metastases with implications for the role of thoracoscopy

Thoracoscopy for wedge resection of lung metastases is rapidly increasing in frequency. This technique precludes bimanual palpation of the lung to locate additional lesions not seen on the surface. Finger palpation is inadequate. Implications regarding the failure to identify all metastases and the negative impact on long-term survival led us to review retrospectively the correlation between pathologic findings and imaging reports. One hundred forty-four patients who had resection of lung metastases from colorectal cancer were studied. All had chest roentgenograms and 72 had computed tomographic scans as well. Chest roentgenogram and computed tomographic reports differed in the number of nodules reported in 17 of 72 patients (24%). In 3 of 17 patients chest roentgenogram showed more nodules than computed tomography. Chest roentgenogram differed from pathologic findings at surgery in 57 of 144 patients (39%). Twenty-six of 57 patients (46%) had more lesions than chest roentgenogram detected and 31 had fewer. Computed tomographic scans differed from pathologic findings in 30 of 72 patients (42%). If one or two lesions were imaged, 12 patients had fewer cancers (some lesions were benign) and 18 had more cancers than computed tomography reported; computed tomographic scans erred 28% of the time. The inability to adequately palpate the entire lung using the thoracoscope alone markedly impairs the surgeon's ability to know if a resection of all lesions has been done. The validity of using thoracoscopy resection in the management of metastatic disease is seriously questioned other than for diagnosis.