ELSEVIER
ScienceDirect
EJSO 36 (2010) 699-704
EJSO the Journal of Cancer Surgery www.ejso.com
Prognostic factors for survival after surgery for adrenal metastasis
A. Muth*, F. Persson, S. Jansson, V. Johanson, H. Ahlman, B. Wängberg Endocrine Surgery Unit, Institute of Clinical Sciences, Sahlgrenska Academy, SE-413 45 Gothenburg, Sweden Accepted 5 April 2010 Available online 7 May 2010
Abstract
Aim: To better define the indications for adrenalectomy for adrenal metastasis we have analysed factors predicting survival in our institu- tional series.
Methods: A consecutive series of 30 patients undergoing adrenalectomy for metastasis (1996-2007), excluding patients with simultaneous ipsilateral renal cell carcinoma (RCC), was studied. Metastases were regarded as synchronous (<6 mo), or metachronous (>6 mo), depending on the interval after primary surgery. Survival was calculated from time of adrenalectomy and factors influencing survival were identified. Results: The tumour diagnoses were RCC n = 9, malignant melanoma n = 5, non-small-cell lung cancer n = 5, colorectal carcinoma n = 4, foregut carcinoid n = 2, adrenocortical carcinoma, breast cancer, hepatocellular carcinoma, urothelial carcinoma, and liposarcoma (one each); nine adrenal metastases were synchronous and 21 metachronous. Ten patients had undergone previous surgery for extra-adrenal metastases. Out of 30 adrenalectomies 10 were laparoscopic (LAdx) and 20 open (OAdx) procedures without surgical complications. The local recurrence rate was low: LAdx 1/10, OAdx 1/20, and the median survival was 23 months. Independent prognosticators of favourable survival were adrenalectomy for potential cure (p = 0.01), no previous metastasis surgery (p = 0.02), and tumour type (p = 0.043), with better prognosis for patients with adrenal metastasis from colorectal carcinoma and RCC and worse prognosis in non-small-cell lung cancer and malignant melanoma. Conclusions: Surgery for adrenal metastasis is safe and the indication for this procedure in an individual patient can be supported by several prognostic factors. The survival benefit in patients with adrenalectomy for potential cure indicates a therapeutic value of adrenalectomy in selected patients.
@ 2010 Elsevier Ltd. All rights reserved.
Keywords: Adrenal gland neoplasm/secondary; Adrenal gland neoplasm/surgery; Adrenalectomy; Follow-up studies
Introduction
Metastases to the adrenal glands are present in 13-27 percent of disseminated malignancies at autopsy1,2 with the highest figures for patients with pulmonary or renal pri- mary tumours. However, isolated adrenal metastasis only occurs in less than one percent of these cases.3 Adrenalec- tomy can render the patient tumour-free in case of isolated lesions, but the procedure can also be part of a staged
treatment programme when multiple metastatic sites have been identified. Laparoscopic adrenalectomy (LAdx) has been claimed to be a safe alternative in suspected or con- firmed malignancy.4,5 Only few series reporting on indica- tions and results of surgery for adrenal metastases have so far been published.6-12 We have analysed our institutional series to identify prognostic factors for survival and further define the indications for surgical treatment.
Patients and methods
Patients
A consecutive series of patients undergoing adrenalec- tomy for metastasis to the adrenal gland (1996-2007) was studied, i.e. from the introduction of LAdx at our cen- tre. Patients with direct extension of a primary tumour into the adrenal gland, or renal cell carcinoma (RCC) with ipsi- lateral synchronous adrenal metastasis, were not studied.
* Previous communication: In part presented as a poster at the European Society of Endocrine Surgeons’ 3rd BIENNIAL CONGRESS, April 24-26, 2008, Barcelona.
* Corresponding author at: Department of Surgery, Sahlgrenska Univer- sity Hospital, SE-413 45 Gothenburg, Sweden. Tel .: +46 31 342 10 00; fax: +46 31 342 46 00.
E-mail addresses: andreas.muth@vgregion.se (A. Muth), fredrik.l. persson@vgregion.se (F. Persson), svante.jansson@vgregion.se (S. Jansson), viktor.johanson@vgregion.se (V. Johanson), hakan.ahlman@surgery.gu.se (H. Ahlman), bo.wangberg@surgery.gu.se (B. Wängberg).
The clinical information was gathered through patient re- cords from our unit and referring hospitals. All histopatho- logical reports were reviewed. Information on causes of death was retrieved from the Swedish Cause of Death Reg- ister. The study was approved by the Regional Ethical Re- view Board in Gothenburg.
Definitions
Metastases were regarded as synchronous if detected within 6 months after primary surgery. Metastases discov- ered more than 6 months after primary surgery were defined as metachronous. Disease-free interval (DFI) was defined as the period of time the patient was tumour-free prior to de- tection of the adrenal metastasis. As a consequence, patients with an adrenal metastasis detected more than 6 months af- ter primary surgery, but treated for another tumour manifes- tation within 6 months, thus have a metachronous adrenal metastasis with DFI less than 6 months. Disease-free and overall survival were calculated from time of adrenalectomy up to tumour recurrence or death. Adrenalectomy for poten- tial cure was defined as a local R0 resection with no evi- dence of residual tumour at other sites.
Statistical analysis
Overall and disease-free survival was calculated accord- ing to the Kaplan-Meier method. Patient gender, age at surgery, clinical presentation (synchronous or metachro- nous tumour, DFI, previous metastasectomy), type of oper- ation (open or laparoscopic), completeness of resection, size and histopathological diagnosis of the metastasis were evaluated using overall survival as the main outcome measure. Univariate comparisons of survival in different groups were performed with the log-rank test. Multiple stepwise Cox-regression13 was used to construct a model relating survival to risk factors. The results given are from the final model. A p-value < 0.05 was considered sta- tistically significant.
Results
Patient characteristics
Thirty patients with adrenal metastasis (12 female, 18 male, mean (median) age 60.6 (62.5) years, range 30-79 years) were treated and fulfilled the inclusion criteria at our centre during the study period (Table 1). The diagnoses were RCC n = 9, malignant melanoma n = 5, non-small cell lung cancer (NSCLC) n = 5, colorectal carcinoma n = 4, foregut carcinoid n = 2, adrenocortical carcinoma (metastasis to contralateral adrenal), breast cancer, hepato- cellular carcinoma, urothelial carcinoma, and liposarcoma (one each).
Clinical presentation
Nine patients had synchronous and 21 had metachro- nous adrenal metastasis; 10/21 had been subject to previous surgery for metastases. The mean (median) interval be- tween primary surgery and detection of metastasis in the metachronous group was 80 (48) months, range 10-460, and mean (median) DFI was 10 (26) months, range 0-117. Nine patients had a DFI >12 months.
Staging and diagnosis
Preoperative work-up included computerized tomogra- phy (CT) of the abdomen (n = 28), in 10 patients supple- mented with abdominal magnetic resonance imaging (MRI). Three patients were staged with 18F-fluorodeoxy- glucose-PET-CT. Fine-needle aspiration cytology of the adrenal mass was performed in seven patients, confirming the diagnosis in five.
Surgical treatment
The choice of surgical technique was up to the individual surgeon. LAdx was preferred in isolated adrenal metastasis, while open adrenalectomy (OAdx) was considered in multi- focal disease or severe peritoneal adhesions. 20 OAdx and 10 LAdx were performed, the latter ones using the lateral trans- peritoneal approach. Three LAdx were converted to OAdx due to technical difficulties. OAdx was in eight patients com- bined with other surgical procedures: nephrectomy/renal re- section in three, liver resection in two, splenectomy, cholecystectomy or bilateral adx in one patient each. The mean hospital stay after LAdx was four days, after OAdx ten days (p = 0.05). The local recurrence rate was low in both groups (OAdx: 1/20, LAdx: 1/10, converted to OAdx) during a mean observation period of 35 and 16 months, re- spectively. No port-site metastases were observed. Adrenal- ectomy for potential cure was achieved in 8/10 patients treated with LAdx and in 11/20 treated with OAdx.
There was no mortality associated with the surgical pro- cedure. Two patients had non-fatal procedure-related com- plications (stroke and atrial fibrillation/heart failure, respectively). The first patient died four months postopera- tively of a myocardial infarction, the other of progressive tumour disease 36 months after adrenalectomy.
Follow-up, recurrence and survival
Complete follow-up was available in all patients. Me- dian overall survival (95% CI) was 23 (15-31) months with a 5-year actuarial survival rate of 22.5%. Median dis- ease-free survival (95% CI) was 6 (0.9-11) months (Fig. 1A). Three patients are alive with no evidence of dis- ease at 101, 60 and two months of follow-up. Five patients are alive with disease after 120, 52, 34, 24 and 16 months of follow-up. Twenty patients died of their tumour disease,
| No | Sex | Age | Presentation | DFI (mo.) | Previous metastasectomy | Technique | Histopathology | Adx for potential cure | DFS (mo.) | FU (mo.) | Status |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | M | 33 | Metachronous | 10 | Yes | OAdx | Malignant melanoma | No | 0 | 11 | DOD |
| 2 | M | 42 | Metachronous | 2 | Yes | LAdx (conv.) | Urothelial carcinoma | No | 0 | 15 | DOD |
| 3 | M | 51 | Synchronous | 0 | No | OAdx | RCC | Yes | 93 | 120 | AWD |
| 4 | M | 30 | Metachronous | 10 | Yes | LAdx | Myxoid liposarcoma | Yes | 3 | 38 | DOD |
| 5 | M | 68 | Metachronous | 5 | Yes | LAdx | Colorectal carcinoma | Yes | 12 | 23 | DOD |
| 6 | M | 45 | Synchronous | 0 | No | LAdx | Hepatocellular carcinoma | Yes | 19 | 20 | DOD |
| 7 | F | 58 | Synchronous | 6 | No | OAdx | RCC | Yes | 13 | 26 | DOD |
| 8 | M | 62 | Metachronous | 52 | No | OAdx | Foregut carcinoid | No | 0 | 25 | DOD |
| 9 | F | 63 | Synchronous | 0 | No | OAdx | NSCLC | No | 0 | 15 | DOD |
| 10 | M | 64 | Metachronous | 117 | No | OAdx | RCC | Yes | 101 | 101 | NED |
| 11 | M | 57 | Synchronous | 0 | No | OAdx | RCC | Yes | 36 | 76 | DOD |
| 12 | F | 65 | Metachronous | 21 | No | LAdx (conv.) | NSCLC | Yes | 13 | 24 | DOD |
| 13 | M | 78 | Synchronous | 0 | No | LAdx | NSCLC | No | 0 | 6 | DOC |
| 14 | M | 75 | Synchronous | 0 | No | OAdx | RCC | No | 0 | 4 | DOD |
| 15 | F | 65 | Metachronous | 28 | No | OAdx | RCC | Yes | 12 | 51 | DOD |
| 16 | M | 62 | Synchronous | 0 | No | OAdx | Foregut carcinoid | No | 0 | 12 | DOD |
| 17 | F | 60 | Metachronous | 68 | No | OAdx | Breast cancer | Yes | 60 | 60 | NED |
| 18 | F | 59 | Metachronous | 10 | No | LAdx | NSCLC | Yes | 7 | 12 | DOD |
| 19 | F | 73 | Metachronous | 47 | No | OAdx | RCC | Yes | 14 | 36 | DOD |
| 20 | M | 57 | Metachronous | 37 | Yes | OAdx | Colorectal carcinoma | Yes | 22 | 52 | AWD |
| 21 | F | 78 | Metachronous | 38 | Yes | OAdx | Malignant melanoma | Yes | 6 | 19 | DOD |
| 22 | M | 58 | Metachronous | 0 | No | OAdx | RCC | No | 0 | 14 | DOC |
| 23 | M | 65 | Metachronous | 0 | Yes | OAdx | Malignant melanoma | No | 0 | 34 | AWD |
| 24 | F | 68 | Synchronous | 0 | No | OAdx | NSCLC | No | 0 | 2 | DOD |
| 25 | M | 67 | Metachronous | 0 | No | OAdx | Malignant melanoma | No | 0 | 16 | DOD |
| 26 | F | 57 | Metachronous | 5 | Yes | LAdx | Colorectal carcinoma | Yes | 7 | 24 | AWD |
| 27 | M | 79 | Metachronous | 3 | Yes | LAdx | Malignant melanoma | Yes | 3 | 4 | DOD |
| 28 | F | 71 | Metachronous | 7 | No | OAdx | Colorectal carcinoma | Yes | 8 | 16 | AWD |
| 29 | F | 44 | Metachronous | 0 | Yes | OAdx | Adrenocortical carcinoma | Yes | 1 | 11 | DOD |
| 30 | M | 64 | Metachronous | 85 | No | LAdx (conv.) | RCC | Yes | 2 | 2 | NED |
two of other causes (traffic accident and myocardial infarc- tion, respectively).
Adrenalectomy for potential cure was associated with prolonged survival both in univariate and multivariate anal- yses, with a hazard ratio of 4.9 for patients with residual metastases (p = 0.01) (Table 2, Fig. 1B). Previous metas- tasis surgery was a significant independent risk factor for worse prognosis with a hazard ratio of 5.8 in multivariate analysis (p = 0.02) (Table 2, Fig. 1C). The mean interval (95% CI) between primary surgery and adrenalectomy was 47 (2-93) months in patients without vs. 77 (31-123) months in patients with previous surgery for me- tastases (p = 0.41). Significant differences in survival with regard to tumour type were seen (p = 0.043), with longer survival for patients with colorectal carcinoma or renal cell carcinoma and shorter for those with NSCLC or malig- nant melanoma (Table 2, Fig. 1D). The hazard ratio for death after metastasis surgery in patients with NSCLC vs. colorectal carcinoma was 37.6 (p = 0.008). There was no statistically significant difference in survival between pa- tients with synchronous or metachronous metastasis. How- ever, in the metachronous group DFI > 12 months was
associated with a 21 months longer median survival (log- rank test, p = 0.03). No correlation between survival and gender, age at surgery, surgical technique (LAdx vs. OAdx) or size of the metastasis (>45 mm vs. < 45 mm) was demonstrated.
Discussion
Main findings
In our consecutive 12-year series of 30 patients undergo- ing surgery for adrenal metastasis the overall median sur- vival was 23 months with a 5-year actuarial survival rate of 22.5%. This is in agreement with other investigators, e.g. Lo et al.º reported a 2-year survival rate of 40% in a se- ries of 52 patients. Strong et al.,12 updating a previously published series,9 reported an actuarial survival rate of 37% (29/78) at 2 years, and of 17% (6/37) at 8 years; the series consisted of 92 patients. Castillo et al.14 reported a mean survival of 26 months in a series of 22 cases. Sebag et al,11 found a median survival of 23 months in 16 patients; one-third was alive and disease-free at 5 years.
A
B
1.0
- overall survival
1.0
+ adx for potential cure +
Fraction survival
0.8
--- disease-free survival
Fraction survival
0.8
-L adx for potential cure -
0.6
u
0.6
P=0.01
0.4
0.4-
0.2
0.2-
0.0
0.0
0
1
2
3
4
5
0
1
2
3
4
5
Follow-up (years)
Follow-up (years)
OS
30
23
13
8
7
4
Adx +
19
16
11
8
7
4
DFS
19
11
5
4
3
4
Adx -
11
7
2
C
D
1.0
L
previous metastasectomy
1.0
··· · colorectal carcinoma
— no previous metastasectomy
P=0.043
… malignant melanoma
Fraction survival
0.8.
Fraction survival
0.8
NSCLC
P=0.02
៛
0.6
0.6
renal cell carcinoma
I-other
0.4
M
0.4
i
0.2-
5
1
0.2-
0.0
1
2
3
5
0.0
0
4
0
1
2
3
4
5
Follow-up (years)
Follow-up (years)
PM +
10
9
4
3
2
1
CRC
4
4
2
1
1
PM -
20
16
9
5
5
3
MM
5
3
1
NSCLC
5
3
1
RCC
9
7
6
5
4
4
Other
7
6
3
2
2
In our study adrenalectomy for potential cure, previous me- tastasis surgery and tumour type were all independent prog- nostic factors for survival (Table 2). The significance of adrenalectomy for potential cure as a prognostic factor has pre- viously been identified by some,6,7 but not other, authors.9,1 9,12
Impact of surgery on survival
Whether adrenalectomy alters the course of disease is not easily assessed in a setting of multiple metastatic sites subject to multi-modal treatment. Randomized studies have not been performed. In a small case-control study Luketich & Burt15 demonstrated a prolonged median survival in patients with NSCLC and synchronous solitary adrenal metastasis under- going chemotherapy and adrenalectomy vs. chemotherapy alone (31 vs. 8.5 months). In a review of patients with mela- noma metastases, Mittendorf et al.16 found longer survival in surgically treated cases compared to the entire patient group. The bias in the selection of adrenalectomy candidates is ob- vious. However, our finding that adrenalectomy for potential cure was an independent factor for good prognosis supports a positive survival effect of adrenalectomy.
Patient selection
To select patients suited for adrenalectomy leading to po- tential cure careful preoperative work-up, and optimal imag- ing, is necessary. With the refinement of CT diagnostics for
adrenal lesions,17 dedicated adrenal CT is our method of choice for characterisation of the tumour. In line with this we no longer routinely use fine-needle aspiration cytology. 18F-FDG-PET-CT was only performed in three patients in this series, but is now our standard procedure for staging.
Surgical methods
No difference in survival was seen with regard to the surgical technique in this series, in line with previous stud- ies.9,12 The difference in potential curative adrenalectomy rate between LAdx and OAdx probably reflects our prefer- ence for open surgery in patients with multi-focal disease. The rate of local recurrence was low in both adrenalectomy groups. However, LAdx was associated with a much shorter hospital stay than OAdx and for patients with isolated me- tastasis we now prefer LAdx.
Prognostic factors for survival
Tumour type was early suggested to be a prognostic fac- tor leading to better survival in patients with metastases from adenocarcinoma.º In our series tumour type was a sig- nificant prognosticator with the most favourable outcome for patients with colorectal carcinoma; the shortest survival was seen in patients with NSCLC (Fig. 1D).
Previous surgery for metastases was an independent neg- ative prognostic factor in our series, i.e. the adrenal
| Variable | Mean survival in months (95% CI) | Univariate Analysis (p-value) | Multivariate analysis (p-value) | Hazard ratio |
|---|---|---|---|---|
| Gender | 0.86 | n.s. | ||
| Male (n = 18) | 28 (18-39) | |||
| Female (n = 12) | 28 (17-39) | |||
| Tumour type | 0.023 | 0.043 | ||
| Colorectal cancer (n = 4) | 42 (27-58) | 1.0ª | ||
| Renal cell carcinoma (n = 9) | 56 (26-85) | 0.10ª | 8.5ª | |
| Malignant melanoma (n = 5) | 17 (8-26) | 0.13ª | 6.2ª | |
| Non-small cell lung cancer (n = 5) | 12 (4-19) | 0.008ª | 37.6ª | |
| Other (n = 7) | 26 (14-38) | 0.16ª | 4.7ª | |
| Presentation | 0.37 | n.s. | ||
| Synchronous (n = 9) | 31 (6-56) | |||
| Metachronous (n = 21) | 40 (23-54) | |||
| Disease-free interval | 0.028 | n.s. | ||
| Metachronous DFI<12 months (n = 12) | 21 (14-29) | |||
| Metachronous DFI>12 months (n = 9) | 57 (33-82) | |||
| Previous metastasectomy | 0.53 | 0.02 | 5.8 | |
| No (n = 20) | 42 (23-60) | |||
| Yes (n = 10) | 24 (14-35) | |||
| Size of metastasis | 1.00 | n.s. | ||
| <45 mm (n = 12) | 40 (16-64) | |||
| >45 mm (n = 18) | 36 (19-52) | |||
| Surgical technique | 0.14 | n.s. | ||
| Laparoscopic adrenalectomy (n = 10) | 20 (12-28) | |||
| Open adrenalectomy (n = 20) | 46 (26-66) | |||
| Adrenalectomy for potential cure | 0.02 | 0.01 | 4.9b | |
| Achieved (n = 19) | 51 (31-72) | |||
| Not achieved (n = 11) | 14 (9-19) |
a Versus colorectal carcinoma.
b Adrenalectomy for potential cure not achieved vs. achieved.
metastasis might represent a late manifestation of tumour disease (Fig. 1C). However, the time interval between pri- mary surgery and detection of the adrenal metastasis was not significantly longer in these patients than in those with- out previous metastasis surgery. Ten out of 30 patients in our series had undergone previous surgery for metastases compared to 22% in the report by Sarela et al.9 and 56% reported by Sebag et al.11
DFI can be regarded as a surrogate marker for tumour ag- gressiveness. It seems reasonable to assume that patients with a long disease-free period have a more indolent course and therefore a better prognosis. In a study on adrenal metastases of NSCLC by Mercier et al.,10 and in the first two reports from the Memorial Sloan-Kettering Cancer Center,7,9 meta- chronous tumours were associated with better prognosis. In the updated MSKCC series this association was no longer ev- ident12 and has not been confirmed by other authors.6,8,11 In the literature, DFI has been defined as the time interval be- tween primary surgery and detection of the adrenal metasta- sis, not taking surgery for other tumour manifestations into account. We used a strict definition of DFI as the period of time the patient was objectively tumour-free prior to the rec- ognition of adrenal metastasis. With this definition DFI >12
months was a positive prognostic factor in patients with metachronous metastasis.
In contrast to the present and other reports7,9,11 Strong et al.12 found that adrenal metastases with large diameter (>45 mm) were associated with worse prognosis. Almost half (42%) of their patients had lung cancer and tumour size may more accurately correlate to tumour aggressive- ness in less heterogenous patient series.
Analyses of prognostic factors in case-series are ham- pered by selection bias and many variables are closely inter-related. The small number of patients in the non-rand- omised studies published further adds to the complexity. A larger pooled analysis using strict definitions and stand- ardised protocols would help to identify the optimal indica- tions for adrenalectomy for adrenal metastasis. In larger patient materials more precise identification of tumour characteristics, including molecular markers, could lead us to tailored treatment of individual patients.
Conclusions
Adrenalectomy should always be considered for isolated adrenal metastasis but also in patients undergoing multi-
modal treatment for metastatic disease. The median sur- vival in this series was 23 months with low perioperative morbidity. Factors associated with longer survival were tu- mour type (best for colorectal carcinoma), no prior surgery for metastases, long disease-free interval and potentially curative adrenalectomy at time of surgery.
Conflict of interest statement
None declared.
Acknowledgements
This study was funded in part by a generous unrestricted research grant from the Region of West Sweden.
The expert assistance of Kjell Pettersson, Department of Economics, School of Business, Economics and Law, Uni- versity of Gothenburg, is greatly appreciated for statistical advice and calculations.
References
1. Glomset D. The incidence of metastasis of malignant tumors to the ad- renals. Am J Cancer 1938;32(1):57-61.
2. Abrams HL, Spiro R, Goldstein N. Metastases in carcinoma; analysis of 1000 autopsied cases. Cancer 1950;3(1):74-85.
3. Lam KY, Lo CY. Metastatic tumours of the adrenal glands: a 30-year experience in a teaching hospital. Clin Endocrinol (Oxf) 2002;56(1): 95-101.
4. Kebebew E, Siperstein AE, Clark OH, Duh QY. Results of laparo- scopic adrenalectomy for suspected and unsuspected malignant ad- renal neoplasms. Arch Surg 2002;137(8):948-51. discussion 952-3.
5. Moinzadeh A, Gill IS. Laparoscopic radical adrenalectomy for malig- nancy in 31 patients. J Urol 2005;173(2):519-25.
6. Lo CY, van Heerden JA, Soreide JA, et al. Adrenalectomy for meta- static disease to the adrenal glands. Br J Surg 1996;83(4):528-31.
7. Kim SH, Brennan MF, Russo P, Burt ME, Coit DG. The role of sur- gery in the treatment of clinically isolated adrenal metastasis. Cancer 1998;82(2):389-94.
8. Porte H, Siat J, Guibert B, et al. Resection of adrenal metastases from non-small cell lung cancer: a multicenter study. Ann Thorac Surg 2001;71(3):981-5.
9. Sarela AI, Murphy I, Coit DG, Conlon KC. Metastasis to the adrenal gland: the emerging role of laparoscopic surgery. Ann Surg Oncol 2003;10(10):1191-6.
10. Mercier O, Fadel E, de Perrot M, et al. Surgical treatment of solitary adrenal metastasis from non-small cell lung cancer. J Thorac Cardio- vasc Surg 2005;130(1):136-40.
11. Sebag F, Calzolari F, Harding J, Sierra M, Palazzo FF, Henry JF. Iso- lated adrenal metastasis: the role of laparoscopic surgery. World J Surg 2006;30(5):888-92.
12. Strong VE, D’Angelica M, Tang L, et al. Laparoscopic adrenalectomy for isolated adrenal metastasis. Ann Surg Oncol 2007;14(12): 3392-400.
13. Cox DR, Oakes D. Analysis of survival data. London: Chapman and Hall; 1984. pp. 91-107.
14. Castillo OA, Vitagliano G, Kerkebe M, Parma P, Pinto I, Diaz M. Lap- aroscopic adrenalectomy for suspected metastasis of adrenal glands: our experience. Urology 2007;69(4):637-41.
15. Luketich JD, Burt ME. Does resection of adrenal metastases from non-small cell lung cancer improve survival? Ann Thorac Surg 1996;62(6):1614-6.
16. Mittendorf EA, Lim SJ, Schacherer CW, et al. Melanoma adrenal me- tastasis: natural history and surgical management. Am J Surg 2008;195 (3):363-8. discussion 368-9.
17. Szolar DH, Kammerhuber FH. Adrenal adenomas and non adenomas: assessment of washout at delayed contrast-enhanced CT. Radiology 1998;207(2):369-75.
18. Israel O, Kuten A. Early detection of cancer recurrence: 18F-FDG PET/CT can make a difference in diagnosis and patient care. J Nucl Med 2007;48(Suppl. 1):28S-35S.
19. Czernin J, Allen-Auerbach M, Schelbert HR. Improvements in cancer staging with PET/CT: literature-based evidence as of September 2006. J Nucl Med 2007;48(Suppl. 1):78S-88S.