- Email: [email protected]
Palliative Percutaneous Nephrostomy in Recurrent Cervical Cancer: A Retrospective Analysis of 50 Consecutive Cases
Vol. 36 No. 2 August 2008
Journal of Pain and Symptom Management
185
Original Article
Palliative Percutaneous Nephrostomy in Recurrent Cervical Cancer: A Retrospective Analysis of 50 Consecutive Cases Rodrigo Dienstmann, MD, Cristhiane da Silva Pinto, MD, Margarida Tutungi Pereira, MD, Isabele Avila Small, MD, and Carlos Gil Ferreira, PhD Division of Clinical Cancer Research (R.D., I.A.S., C.G.F.) and Division of Palliative Care (C.S.P., M.T.P), National Cancer Institute, Rio de Janeiro, Brazil
Abstract Cervical cancer is a public health problem in Brazil, with annual incidence rates of 20e40 cases/100,000 women. Most patients with recurrent disease have symptoms from locoregional disease and may develop renal failure. This study aims to evaluate the outcome of patients with recurrent cervical cancer who underwent percutaneous nephrostomy (PN). We reviewed the medical records of 50 such patients who were referred to the Palliative Care Unit of the Brazilian National Cancer Institute from January 2002 to October 2006. Median age was 44 years (range, 26e67 years). Half the patients had improvement in pain or uremic symptoms, and seven (14%) had improved performance status (PS) after the procedure. Thirty patients (60%) had improvement of renal function; median creatinine levels before and after PN were 6.4 and 3.7 mg/dL, respectively (P < 0.05). Median overall survival after PN was 8.9 weeks (95% confidence interval [CI]: 7.4e10.3). Median survival was 9.9 weeks (95% CI: 8.7e11.0) in 40 patients with baseline PS 1e3 and one week (95% CI: 0.1e1.9) in 10 patients with PS 4 (log rank, P < 0.0001). Median survival in patients with and without improvement of renal function after PN was 10.0 weeks (95% CI: 8.6e11.3) and 2.6 weeks (95% CI: 0e11.3), respectively (log rank, P ¼ 0.01). Twenty-nine patients (58%) died from renal failure. Complications were mainly urinary tract infection (n ¼ 10), catheter loss (n ¼ 9), and bleeding (n ¼ 1). These data suggest that PN can be of clinical benefit for carefully selected patients with recurrent cervical cancer. J Pain Symptom Manage 2008;36:185e190. Ó 2008 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Cervical cancer, nephrostomy, palliative care, renal failure
Introduction Address correspondence to: Rodrigo Dienstmann, MD, Clinical Cancer Research, Instituto Nacional de Caˆncer of Brazil, Rua Andre´ Cavalcanti, 37 segundo andar, Rio de Janeiro 20231-050, Brazil. E-mail: [email protected] Accepted for publication: September 18, 2007. Ó 2008 U.S. Cancer Pain Relief Committee Published by Elsevier Inc. All rights reserved.
Cervical cancer is a public health problem in Brazil, with an annual incidence rate of 20e40 cases/100,000 women. More than 18,000 cases are estimated for 2008. It is the second most common malignancy in women, after breast cancer. In addition, it is the fourth leading cause of cancer death in Brazilian women.1 0885-3924/08/$esee front matter doi:10.1016/j.jpainsymman.2007.09.010
186
Dienstmann et al.
Most patients present with locally advanced disease. At the Brazilian National Cancer Institute, two-thirds of women have stage III/IV disease at diagnosis (unpublished data). These patients receive cisplatin-based chemotherapy in combination with radiotherapy as definitive treatment and have a high risk of developing progressive disease during follow up. Clinical options for residual or recurrent disease are limited. Currently, there is no established treatment for refractory cervical cancer. Chemotherapy in this setting yields a limited success rate and is very toxic.2e7 Response rates are lower than 15%, median estimated progression-free survival is about two months, and median overall survival is approximately six months.4,5 Pelvic pain, which may be difficult to control, and obstructive renal failure secondary to urinary tract invasion are frequently seen in patients with locally advanced recurrent cervical cancer. Percutaneous nephrostomy (PN) is a clinical option as a palliative measure in this situation. It is associated with survival benefit in patients with untreated cervical cancer or those with viable treatment options, as previously shown in retrospective studies.8,9 However, when used in patients with refractory/ metastatic disease or those without treatment options, clinical benefit has been limited.8,10e12 In addition, complications of the procedure have been reported in 40% of the patients.13,14 There are no clear guidelines to predict which patients in the palliative care scenario derive advantage from PN in terms of survival and quality of life. Factors influencing the nonrecovery of renal function after the relief of urinary tract obstruction in women with cervical cancer include older age and decreased renal cortical thickness.15 This study aims to evaluate the outcome of patients with recurrent cervical cancer in the palliative care setting who underwent PN and to identify those who can potentially benefit from this intervention.
Methods We retrospectively reviewed the medical records of all patients who underwent PN following referral to the Palliative Care Unit of the Brazilian National Cancer Institute between January 2002 and October 2006 with the diagnosis of cervical cancer. At our
Vol. 36 No. 2 August 2008
institution, patients receiving only palliative care are referred to the Palliative Care Unit. Those patients assessed in our study underwent urinary diversion when no chemotherapy or radiotherapy options were available. The main objective of this study was to estimate overall survival of these patients, defined as the date between the procedure and death. Secondary objectives were (a) to evaluate the clinical benefit of the intervention, defined as improvement in uremic symptoms (according to medical records), pelvic/lumbar pain (using a Visual Analog Scale) or performance status (PS); (b) to assess the proportion of patients with improvement in renal function after the procedure; (c) to estimate the proportion of patients discharged from hospital after the procedure; (d) to evaluate the morbidity and complications of PN (specifically catheter loss/ obstruction, bleeding and urinary tract infection); and (e) to evaluate the cause of death in these patients. Performance status was measured according to Eastern Cooperative Oncology Group/Zubrod criteria. Improvement in renal function was defined as consistent decrease in creatinine levels or stabilization of creatinine levels in the normal range after the intervention. All patients had ureteral obstruction diagnosed through ultrasound. PN was unilateral in all cases and the side cleared from the obstruction was the one with less pyelocaliceal dilatation and greater cortical thickness, according to previous published data.16,17 All analyses were performed using SSPS statistical software. Survival curves for overall survival data were based on the KaplaneMeier method. Comparisons of overall survival according to PS and renal function at the time of intervention were done using log-rank tests. A P-value lower than 0.05 was considered significant. Median survival times are presented in weeks followed by 95% confidence interval (95% CI). The Ethics Committee of our institution approved the study.
Results Fifty patients received palliative PN due to urinary tract obstruction. Median age was 44 years (range, 26e67 years). Race was white in 23 patients (46%) and black in the remaining. All patients received previous treatment with
Vol. 36 No. 2 August 2008
Percutaneous Nephrostomy in Cervical Cancer
surgery or chemoradiotherapy followed by brachytherapy as definitive therapy. Additional palliative radiotherapy, with or without chemotherapy, was performed based on previous treatment. Median time from diagnosis until referral to the Palliative Care Unit was 12 months (range, 2e51 months). At the time of indication for palliative care, 50 had locally advanced disease and seven also presented with metastatic disease (liver, lung, or bone). Table 1 describes the PS of these patients prior to PN. Thirty-seven patients (74%) had uremic symptoms at the time of PN. Half of them (18 of 37) had improvement of these symptoms after the intervention. Lumbar pain related to hydronephrosis was reported in 16 patients (32%). This symptom resolved in eight patients (50%) immediately after the procedure. Placement of another catheter was not considered in those patients without improvement of uremic symptoms or lumbar pain if the renal cortical thickness on the side not cleared initially from the obstruction was smaller or another invasive intervention was not justified. Therefore, symptoms were managed with appropriate medications. Only seven patients (14%) had improvement in PS after PN. Eleven patients (22%) were not discharged from hospital until death. Most of them (seven of 11) had PS 4 at the time of PN. Thirty patients (60%) had improvement in renal function. Most patients (44 of 50) had bilateral hydronephrosis at initial ultrasound. Median creatinine levels before and after PN were 6.4 mg/dL (range, 0.8e19.4 mg/dL) and 3.7 mg/dL (range, 0.4e21.9 mg/dL), respectively (P < 0.001). Two patients had normal renal function (as measured by creatinine levels) prior to nephrostomy (another five had creatinine levels higher than upper normal limit but lower than 2.0 mg/dL). PN was ‘‘prophylactic’’ in these patients once creatinine levels were increasing progressively. No patient had hemodialysis before or after the procedure.
Median overall survival after PN was 8.9 weeks (95% CI: 7.4e10.3). Median survival according to PS is described in Table 2 and illustrated in Fig. 1. Once those with PS 1e3 had comparable survival curves, comparisons were made considering these patients as a group. As shown in Fig. 2, median survival was 9.9 weeks (95% CI: 8.7e11.0) in these 40 patients, compared with one week (95% CI: 0.1e1.9) in 10 patients with PS 4; (log rank, P < 0.0001). Median survival in patients with and without improvement of renal function after PN was 10.0 weeks (95% CI: 8.6e11.3) and 2.6 weeks (95% CI: 0e11.3), respectively (log rank, P ¼ 0.01), as illustrated in Fig. 3. In an attempt to identify subgroups of patients that could potentially have greater benefit from the intervention in terms of prolonged survival, we performed an exploratory analysis evaluating survival curves of patients according to renal function prior to PN. Patients were analyzed according to baseline creatinine levels (lower or higher than 2.0 mg/dL). As shown in Fig. 4, survival was not affected by renal function prior to nephrostomy. Median survival of those patients with concomitant metastatic disease (seven patients) was 8.1 weeks (95% CI: 0.1e23.2), not significantly different from patients with locoregional disease only (9.4 weeks, 95% CI: 8.4e10.5; log rank, P ¼ 0.8). Complications of PN are described in Table 3. Twenty-two patients (44%) had some form of morbidity related to the intervention. Twentynine patients (58%) died in uremia (presence of significant uremic symptoms attributed to raised urea and decreased renal function). Of the 30 patients with improvement in renal function, 11 died due to renal failure. Eighteen patients (36%) died due to disease progression without renal failure. Three patients died at home and it was not possible to evaluate renal function/uremic symptoms at the time of Table 2 Median Overall Survival (in Weeks) According to Performance Status (PS) at the Time of Percutaneous Nephrostomy
Table 1 Performance Status (PS) at the Time of Palliative Percutaneous Nephrostomy PS
n
Percent
0 1 2 3 4
0 5 7 28 10
0 10 14 56 20
187
PS 1 2 3 4 Overall
Survival (weeks)
95% CI
12.1 11.1 9.7 1.0 8.9
6.3e18.0 1.2e21.0 7.8e11.6 0.1e1.9 7.4e10.3
188
Dienstmann et al.
Vol. 36 No. 2 August 2008
Fig. 1. Median overall survival according to Performance Status (PS) of 50 patients with cervical cancer who underwent palliative percutaneous nephrostomy.
Fig. 3. Median survival according to improvement of renal function (RF) after percutaneous nephrostomy (RF improved: 30 patients, RF not improved: 20 patients; log rank, P ¼ 0.01).
death. Six patients (12%) had urinary septicemia requiring hospital admission, and in two cases, this was the actual cause of death.
chemotherapy in the palliative setting are associated with dismal prognosis and short survival. PN as a palliative measure to relieve hydronephrosis and renal failure is a common indication in our institution. However, the indications for this procedure have been questioned in patients with poor prognosis and the real benefit of this intervention has not been previously studied in patients with cervical cancer who are receiving palliative care exclusively. Our retrospective study shows that nephrostomy was associated with clinical benefit in about half of the patients, as measured by
Discussion Cervical cancer is an epidemic disease in Brazil. Screening programs are inefficient and most patients present with locally advanced disease at the time of diagnosis, which increases the likelihood of locoregional relapse. The poor PS of these women at disease progression and the limited benefit of
Fig. 2. Median survival according to Performance Status (PS 1e3: 40 patients, PS 4: 10 patients; log rank, P < 0.0001).
Fig. 4. Median survival according to renal function prior to percutaneous nephrostomy (creatinine [Cr] levels > 2 mg/dL: 43 patients, Cr < ¼ 2 mg/dL: 7 patients; log rank, P ¼ 0.073).
Vol. 36 No. 2 August 2008
Percutaneous Nephrostomy in Cervical Cancer
Table 3 Complications of Palliative Percutaneous Nephrostomy in 50 Patients Complication Urinary tract infection Catheter loss Bleeding Obstruction Pain
Frequency
Percent
10 9 1 1 1
20 18 2 2 2
recovery from uremic symptoms, lumbar pain, or renal function. Nevertheless, only 14% of women had improvement in PS and 22% never left the hospital after the procedure. Shekarriz et al. showed that 15% of 103 patients with advanced malignancies who underwent palliative urinary diversion were never discharged from hospital.18 Romero et al. presented an even larger number, with 39% of 43 patients dying during the hospitalization period.13 The median survival in our study was nine weeks. This survival is difficult to put into perspective, because previous studies have evaluated a limited number of patients. Watkinson et al. studied the role of PN in the management of renal failure resulting from advanced abdomino-pelvic malignancy. In the subgroup of patients with relapsed disease and no conventional treatment options (n ¼ 18), median survival was estimated at only 38 days (five weeks).8 Chan et al. observed a median survival of 51 days in eight women with recurrent cervical cancer and obstructive uropathy who underwent PN.11 Emmert et al. described no survival longer than two months in three patients with progressive localized primary cervical cancer who underwent PN.12 Moreover, patients with metastatic cervical cancer at the time of intervention were also not satisfactorily served by nephrostomy.12 In our study, patients with metastatic disease (n ¼ 7) had the same median survival of those with locally advanced disease. However, according to our data, in the subgroup of patients with better PS (1e3), median survival appears to be longer, approaching 10 weeks. These patients might benefit from the intervention in terms of prolonged survival, once they have the chance to receive adequate symptom palliation. In marked contrast, no worthwhile benefit was obtained when nephrostomy was indicated as a palliative measure in patients with PS 4. We also showed that patients with improvement in renal function after percutaneous
189
urinary diversion have a longer median survival, which is probably related to less advanced disease and better PS at the time of indication of nephrostomy. However, decline of creatinine levels does not guarantee improvement of uremic symptoms, pain, or quality of life in patients with advanced neoplasia in palliative care. Most patients in our study died in renal failure. Even those with an initial improvement of renal function eventually have recurrent obstructive nephropathy and uremia (11 of 30 patients in our series). This is in line with the study by Harrington et al.,14 which reported the same outcome in 20 of 42 patients. This procedure is usually relatively safe, simple, and fast in noncancer patients.19 Thus, many experts could feel a strong wish to perform PN in patients with cancer-derived obstruction before properly assessing each patient’s individual situation. In this regard, it should be highlighted that our study showed that 44% of the patients have some sort of morbidity related to the procedure, mostly catheter loss (requiring additional interventions) and urinary tract infection. Previous studies reported a complication rate of 40%e70%, and this should be kept in mind at the time of indication for this invasive intervention.13,14,16,18 The pain and suffering caused by prolonging the lifetime of these patients must be carefully considered before performing PN. Previous studies reported that more than 80% of malignant uropathy patients with nephrostomy tubes show cancerrelated symptoms, despite implementation of the diversion.18,20,21 The main factor that should guide the management is patient desire after an honest discussion with the attending physician. Some patients may wish to prolong life even for a short time due to emotional, legal, or financial reasons, and this wish must be respected. If PN is conducted for a terminally ill woman with cervical cancer and prolongs her survival by relieving obstructive uropathy, the disease will progress naturally and complications such as recto-vaginal fistulae formation, refractory tenesmus, or digestive tract obstruction are likely to happen. These conditions are both physically and emotionally distressing but not, in themselves, life threatening, as reported by Chang.22 Death caused by renal failure is relatively peaceful in this setting. For that reason, before PN,
190
Dienstmann et al.
patients, and their families must be completely informed about the palliative role of urinary diversion, the potential complications of the procedure, the expected outcomes and anticipated disease prognosis. We conclude that most patients with advanced cervical cancer have poor performance status and short survival after PN. The indication for invasive urinary diversion in terminal cancer patients should be individualized on the basis of expectations for prolonged functional palliation. There are no clear guidelines that could help us in the decision of which patients would benefit from such intervention. Our data suggest that PN can be of clinical benefit for carefully selected patients with recurrent cervical cancer, namely those with PS 1e3.
Acknowledgments The authors would like to thank Dr. Jose´ Bines for manuscript review.
References 1. Instituto Nacional de Caˆncer (INCA, Rio de Janeiro). Caˆncer no Brasil: dados dos registros de base populacional. Available at http://www.inca.gov.br/ estimativa/2008. Accessed March 21, 2008. 2. Sutton GP, Blessing JA, DiSaia PJ, McGuire WP. Phase II study of ifosfamide and mesna in nonsquamous carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1993;49:48e50. 3. Verschraegen CF, Levy T, Kudelka AP, et al. Phase II study of irinotecan in prior chemotherapy-treated squamous cell carcinoma of the cervix. J Clin Oncol 1997;15:625e631. 4. Bookman MA, Blessing JA, Hanjani P, Herzog TJ, Andersen WA. Topotecan in squamous cell carcinoma of the cervix: a Phase II study of the Gynecologic Oncology Group. Gynecol Oncol 2000;77:446e449. 5. Muderspach LI, Blessing JA, Levenback C, Moore JL Jr. A Phase II study of topotecan in patients with squamous cell carcinoma of the cervix: a gynecologic oncology group study. Gynecol Oncol 2001;81:213e215. 6. Muggia FM, Blessing JA, Method M, et al. Evaluation of vinorelbine in persistent or recurrent squamous cell carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92: 639e643. 7. Homesley HD, Vaccarello L, Lowendowski G, Elbendary AA. A phase II trial of weekly one hour paclitaxel as second-line therapy for endometrial
Vol. 36 No. 2 August 2008
and cervical cancer. Proc Am Soc Clin Oncol 2006;24(18S). [abstract 15013]. 8. Watkinson AF, A’Hern RP, Jones A, King DM, Moskovic EC. The role of percutaneous nephrostomy in malignant urinary tract obstruction. Clin Radiol 1993;47:32e35. 9. Soper JT, Blaszczyk TM, Oke E, Clarke-Pearson D, Creasman WT. Percutanous nephrostomy in gynecologic oncology patients. Am J Obstet Gynecol 1988; 158:1126e1131. 10. Gadducci A, Madrigali A, Facchini V, Fioretti P. Percutaneous nephrostomy in patients with advanced or recurrent cervical cancer. Clin Exp Obstet Gynecol 1994;21:71e73. 11. Chan S, Robinson AC, Johnson RJ. Percutaneos nephrsotomy: its value in obstructive uropathy complicating carcinoma of cervix uterus. Clin Oncol (R Coll Radio) 1990;2:156e158. 12. Emmert C, Rassler J, Kohler U. Survival and quality of life after percutaneous nephrostomy for malignant ureteric obstruction in patients with terminal cervical cancer. Arch Gynecol Obstet 1997; 259:147e151. 13. Romero FR, Broglio M, Pires SR, et al. Indications for percutaneous nephrostomy in patients with obstructive uropathy due to malignant urogenital neoplasias. Int Braz J Urol 2005;31:117e124. 14. Harrington KJ, Pandha HS, Kelly SA, et al. Palliation of obstructive nephropathy due to malignancy. Br J Urol 1995;76:101e107. 15. Lutaif NA, Yu L, Abdulkader RC. Factors influencing the non-recovery of renal function after the relief of urinary tract obstruction in women with cancer of the cervix. Ren Fail 2003;25:215e223. 16. Kinn AC, Ohlsen H. Percutaneous nephrostomyda retrospective study focused on palliative indications. APMIS Suppl 2003;109:66e70. 17. Chung SY, Stein RJ, Landsittel D, et al. 15-year experience with the management of extrinsic ureteral obstruction with indwelling ureteral stents. J Urol 2004;172:592e595. 18. Shekarriz B, Shekarriz H, Upadhyay J, et al. Outcome of urinary diversion in the treatment of advanced malignancies. Cancer 1999;85:998e1003. 19. Lewis S, Patel U. Major complications after percutaneous nephrostomydlessons from a department audit. Clin Radiol 2004;59:171e179. 20. Meyer JE, Green TH Jr, Yatsuhashi M. Palliative urinary diversion in carcinoma of the cervix. Obstet Gynecol 1980;55:95e98. 21. Little B, Ho KJ, Gawley S, Young M. Use of nephrsotomy tubes in ureteric obstruction from incurable malignancy. Int J Clin Pract 2003;57:180e181. 22. Chang HL, Lim HW, Su FH, Tsai ST, Wang YW. Win or lose? Percutaneous nephrostomy for a terminal-stage cervical-cancer patient featuring obstructive uropathy. J Palliat Care 2006;22:57e60.
Journal of Pain and Symptom Management
185
Original Article
Palliative Percutaneous Nephrostomy in Recurrent Cervical Cancer: A Retrospective Analysis of 50 Consecutive Cases Rodrigo Dienstmann, MD, Cristhiane da Silva Pinto, MD, Margarida Tutungi Pereira, MD, Isabele Avila Small, MD, and Carlos Gil Ferreira, PhD Division of Clinical Cancer Research (R.D., I.A.S., C.G.F.) and Division of Palliative Care (C.S.P., M.T.P), National Cancer Institute, Rio de Janeiro, Brazil
Abstract Cervical cancer is a public health problem in Brazil, with annual incidence rates of 20e40 cases/100,000 women. Most patients with recurrent disease have symptoms from locoregional disease and may develop renal failure. This study aims to evaluate the outcome of patients with recurrent cervical cancer who underwent percutaneous nephrostomy (PN). We reviewed the medical records of 50 such patients who were referred to the Palliative Care Unit of the Brazilian National Cancer Institute from January 2002 to October 2006. Median age was 44 years (range, 26e67 years). Half the patients had improvement in pain or uremic symptoms, and seven (14%) had improved performance status (PS) after the procedure. Thirty patients (60%) had improvement of renal function; median creatinine levels before and after PN were 6.4 and 3.7 mg/dL, respectively (P < 0.05). Median overall survival after PN was 8.9 weeks (95% confidence interval [CI]: 7.4e10.3). Median survival was 9.9 weeks (95% CI: 8.7e11.0) in 40 patients with baseline PS 1e3 and one week (95% CI: 0.1e1.9) in 10 patients with PS 4 (log rank, P < 0.0001). Median survival in patients with and without improvement of renal function after PN was 10.0 weeks (95% CI: 8.6e11.3) and 2.6 weeks (95% CI: 0e11.3), respectively (log rank, P ¼ 0.01). Twenty-nine patients (58%) died from renal failure. Complications were mainly urinary tract infection (n ¼ 10), catheter loss (n ¼ 9), and bleeding (n ¼ 1). These data suggest that PN can be of clinical benefit for carefully selected patients with recurrent cervical cancer. J Pain Symptom Manage 2008;36:185e190. Ó 2008 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Cervical cancer, nephrostomy, palliative care, renal failure
Introduction Address correspondence to: Rodrigo Dienstmann, MD, Clinical Cancer Research, Instituto Nacional de Caˆncer of Brazil, Rua Andre´ Cavalcanti, 37 segundo andar, Rio de Janeiro 20231-050, Brazil. E-mail: [email protected] Accepted for publication: September 18, 2007. Ó 2008 U.S. Cancer Pain Relief Committee Published by Elsevier Inc. All rights reserved.
Cervical cancer is a public health problem in Brazil, with an annual incidence rate of 20e40 cases/100,000 women. More than 18,000 cases are estimated for 2008. It is the second most common malignancy in women, after breast cancer. In addition, it is the fourth leading cause of cancer death in Brazilian women.1 0885-3924/08/$esee front matter doi:10.1016/j.jpainsymman.2007.09.010
186
Dienstmann et al.
Most patients present with locally advanced disease. At the Brazilian National Cancer Institute, two-thirds of women have stage III/IV disease at diagnosis (unpublished data). These patients receive cisplatin-based chemotherapy in combination with radiotherapy as definitive treatment and have a high risk of developing progressive disease during follow up. Clinical options for residual or recurrent disease are limited. Currently, there is no established treatment for refractory cervical cancer. Chemotherapy in this setting yields a limited success rate and is very toxic.2e7 Response rates are lower than 15%, median estimated progression-free survival is about two months, and median overall survival is approximately six months.4,5 Pelvic pain, which may be difficult to control, and obstructive renal failure secondary to urinary tract invasion are frequently seen in patients with locally advanced recurrent cervical cancer. Percutaneous nephrostomy (PN) is a clinical option as a palliative measure in this situation. It is associated with survival benefit in patients with untreated cervical cancer or those with viable treatment options, as previously shown in retrospective studies.8,9 However, when used in patients with refractory/ metastatic disease or those without treatment options, clinical benefit has been limited.8,10e12 In addition, complications of the procedure have been reported in 40% of the patients.13,14 There are no clear guidelines to predict which patients in the palliative care scenario derive advantage from PN in terms of survival and quality of life. Factors influencing the nonrecovery of renal function after the relief of urinary tract obstruction in women with cervical cancer include older age and decreased renal cortical thickness.15 This study aims to evaluate the outcome of patients with recurrent cervical cancer in the palliative care setting who underwent PN and to identify those who can potentially benefit from this intervention.
Methods We retrospectively reviewed the medical records of all patients who underwent PN following referral to the Palliative Care Unit of the Brazilian National Cancer Institute between January 2002 and October 2006 with the diagnosis of cervical cancer. At our
Vol. 36 No. 2 August 2008
institution, patients receiving only palliative care are referred to the Palliative Care Unit. Those patients assessed in our study underwent urinary diversion when no chemotherapy or radiotherapy options were available. The main objective of this study was to estimate overall survival of these patients, defined as the date between the procedure and death. Secondary objectives were (a) to evaluate the clinical benefit of the intervention, defined as improvement in uremic symptoms (according to medical records), pelvic/lumbar pain (using a Visual Analog Scale) or performance status (PS); (b) to assess the proportion of patients with improvement in renal function after the procedure; (c) to estimate the proportion of patients discharged from hospital after the procedure; (d) to evaluate the morbidity and complications of PN (specifically catheter loss/ obstruction, bleeding and urinary tract infection); and (e) to evaluate the cause of death in these patients. Performance status was measured according to Eastern Cooperative Oncology Group/Zubrod criteria. Improvement in renal function was defined as consistent decrease in creatinine levels or stabilization of creatinine levels in the normal range after the intervention. All patients had ureteral obstruction diagnosed through ultrasound. PN was unilateral in all cases and the side cleared from the obstruction was the one with less pyelocaliceal dilatation and greater cortical thickness, according to previous published data.16,17 All analyses were performed using SSPS statistical software. Survival curves for overall survival data were based on the KaplaneMeier method. Comparisons of overall survival according to PS and renal function at the time of intervention were done using log-rank tests. A P-value lower than 0.05 was considered significant. Median survival times are presented in weeks followed by 95% confidence interval (95% CI). The Ethics Committee of our institution approved the study.
Results Fifty patients received palliative PN due to urinary tract obstruction. Median age was 44 years (range, 26e67 years). Race was white in 23 patients (46%) and black in the remaining. All patients received previous treatment with
Vol. 36 No. 2 August 2008
Percutaneous Nephrostomy in Cervical Cancer
surgery or chemoradiotherapy followed by brachytherapy as definitive therapy. Additional palliative radiotherapy, with or without chemotherapy, was performed based on previous treatment. Median time from diagnosis until referral to the Palliative Care Unit was 12 months (range, 2e51 months). At the time of indication for palliative care, 50 had locally advanced disease and seven also presented with metastatic disease (liver, lung, or bone). Table 1 describes the PS of these patients prior to PN. Thirty-seven patients (74%) had uremic symptoms at the time of PN. Half of them (18 of 37) had improvement of these symptoms after the intervention. Lumbar pain related to hydronephrosis was reported in 16 patients (32%). This symptom resolved in eight patients (50%) immediately after the procedure. Placement of another catheter was not considered in those patients without improvement of uremic symptoms or lumbar pain if the renal cortical thickness on the side not cleared initially from the obstruction was smaller or another invasive intervention was not justified. Therefore, symptoms were managed with appropriate medications. Only seven patients (14%) had improvement in PS after PN. Eleven patients (22%) were not discharged from hospital until death. Most of them (seven of 11) had PS 4 at the time of PN. Thirty patients (60%) had improvement in renal function. Most patients (44 of 50) had bilateral hydronephrosis at initial ultrasound. Median creatinine levels before and after PN were 6.4 mg/dL (range, 0.8e19.4 mg/dL) and 3.7 mg/dL (range, 0.4e21.9 mg/dL), respectively (P < 0.001). Two patients had normal renal function (as measured by creatinine levels) prior to nephrostomy (another five had creatinine levels higher than upper normal limit but lower than 2.0 mg/dL). PN was ‘‘prophylactic’’ in these patients once creatinine levels were increasing progressively. No patient had hemodialysis before or after the procedure.
Median overall survival after PN was 8.9 weeks (95% CI: 7.4e10.3). Median survival according to PS is described in Table 2 and illustrated in Fig. 1. Once those with PS 1e3 had comparable survival curves, comparisons were made considering these patients as a group. As shown in Fig. 2, median survival was 9.9 weeks (95% CI: 8.7e11.0) in these 40 patients, compared with one week (95% CI: 0.1e1.9) in 10 patients with PS 4; (log rank, P < 0.0001). Median survival in patients with and without improvement of renal function after PN was 10.0 weeks (95% CI: 8.6e11.3) and 2.6 weeks (95% CI: 0e11.3), respectively (log rank, P ¼ 0.01), as illustrated in Fig. 3. In an attempt to identify subgroups of patients that could potentially have greater benefit from the intervention in terms of prolonged survival, we performed an exploratory analysis evaluating survival curves of patients according to renal function prior to PN. Patients were analyzed according to baseline creatinine levels (lower or higher than 2.0 mg/dL). As shown in Fig. 4, survival was not affected by renal function prior to nephrostomy. Median survival of those patients with concomitant metastatic disease (seven patients) was 8.1 weeks (95% CI: 0.1e23.2), not significantly different from patients with locoregional disease only (9.4 weeks, 95% CI: 8.4e10.5; log rank, P ¼ 0.8). Complications of PN are described in Table 3. Twenty-two patients (44%) had some form of morbidity related to the intervention. Twentynine patients (58%) died in uremia (presence of significant uremic symptoms attributed to raised urea and decreased renal function). Of the 30 patients with improvement in renal function, 11 died due to renal failure. Eighteen patients (36%) died due to disease progression without renal failure. Three patients died at home and it was not possible to evaluate renal function/uremic symptoms at the time of Table 2 Median Overall Survival (in Weeks) According to Performance Status (PS) at the Time of Percutaneous Nephrostomy
Table 1 Performance Status (PS) at the Time of Palliative Percutaneous Nephrostomy PS
n
Percent
0 1 2 3 4
0 5 7 28 10
0 10 14 56 20
187
PS 1 2 3 4 Overall
Survival (weeks)
95% CI
12.1 11.1 9.7 1.0 8.9
6.3e18.0 1.2e21.0 7.8e11.6 0.1e1.9 7.4e10.3
188
Dienstmann et al.
Vol. 36 No. 2 August 2008
Fig. 1. Median overall survival according to Performance Status (PS) of 50 patients with cervical cancer who underwent palliative percutaneous nephrostomy.
Fig. 3. Median survival according to improvement of renal function (RF) after percutaneous nephrostomy (RF improved: 30 patients, RF not improved: 20 patients; log rank, P ¼ 0.01).
death. Six patients (12%) had urinary septicemia requiring hospital admission, and in two cases, this was the actual cause of death.
chemotherapy in the palliative setting are associated with dismal prognosis and short survival. PN as a palliative measure to relieve hydronephrosis and renal failure is a common indication in our institution. However, the indications for this procedure have been questioned in patients with poor prognosis and the real benefit of this intervention has not been previously studied in patients with cervical cancer who are receiving palliative care exclusively. Our retrospective study shows that nephrostomy was associated with clinical benefit in about half of the patients, as measured by
Discussion Cervical cancer is an epidemic disease in Brazil. Screening programs are inefficient and most patients present with locally advanced disease at the time of diagnosis, which increases the likelihood of locoregional relapse. The poor PS of these women at disease progression and the limited benefit of
Fig. 2. Median survival according to Performance Status (PS 1e3: 40 patients, PS 4: 10 patients; log rank, P < 0.0001).
Fig. 4. Median survival according to renal function prior to percutaneous nephrostomy (creatinine [Cr] levels > 2 mg/dL: 43 patients, Cr < ¼ 2 mg/dL: 7 patients; log rank, P ¼ 0.073).
Vol. 36 No. 2 August 2008
Percutaneous Nephrostomy in Cervical Cancer
Table 3 Complications of Palliative Percutaneous Nephrostomy in 50 Patients Complication Urinary tract infection Catheter loss Bleeding Obstruction Pain
Frequency
Percent
10 9 1 1 1
20 18 2 2 2
recovery from uremic symptoms, lumbar pain, or renal function. Nevertheless, only 14% of women had improvement in PS and 22% never left the hospital after the procedure. Shekarriz et al. showed that 15% of 103 patients with advanced malignancies who underwent palliative urinary diversion were never discharged from hospital.18 Romero et al. presented an even larger number, with 39% of 43 patients dying during the hospitalization period.13 The median survival in our study was nine weeks. This survival is difficult to put into perspective, because previous studies have evaluated a limited number of patients. Watkinson et al. studied the role of PN in the management of renal failure resulting from advanced abdomino-pelvic malignancy. In the subgroup of patients with relapsed disease and no conventional treatment options (n ¼ 18), median survival was estimated at only 38 days (five weeks).8 Chan et al. observed a median survival of 51 days in eight women with recurrent cervical cancer and obstructive uropathy who underwent PN.11 Emmert et al. described no survival longer than two months in three patients with progressive localized primary cervical cancer who underwent PN.12 Moreover, patients with metastatic cervical cancer at the time of intervention were also not satisfactorily served by nephrostomy.12 In our study, patients with metastatic disease (n ¼ 7) had the same median survival of those with locally advanced disease. However, according to our data, in the subgroup of patients with better PS (1e3), median survival appears to be longer, approaching 10 weeks. These patients might benefit from the intervention in terms of prolonged survival, once they have the chance to receive adequate symptom palliation. In marked contrast, no worthwhile benefit was obtained when nephrostomy was indicated as a palliative measure in patients with PS 4. We also showed that patients with improvement in renal function after percutaneous
189
urinary diversion have a longer median survival, which is probably related to less advanced disease and better PS at the time of indication of nephrostomy. However, decline of creatinine levels does not guarantee improvement of uremic symptoms, pain, or quality of life in patients with advanced neoplasia in palliative care. Most patients in our study died in renal failure. Even those with an initial improvement of renal function eventually have recurrent obstructive nephropathy and uremia (11 of 30 patients in our series). This is in line with the study by Harrington et al.,14 which reported the same outcome in 20 of 42 patients. This procedure is usually relatively safe, simple, and fast in noncancer patients.19 Thus, many experts could feel a strong wish to perform PN in patients with cancer-derived obstruction before properly assessing each patient’s individual situation. In this regard, it should be highlighted that our study showed that 44% of the patients have some sort of morbidity related to the procedure, mostly catheter loss (requiring additional interventions) and urinary tract infection. Previous studies reported a complication rate of 40%e70%, and this should be kept in mind at the time of indication for this invasive intervention.13,14,16,18 The pain and suffering caused by prolonging the lifetime of these patients must be carefully considered before performing PN. Previous studies reported that more than 80% of malignant uropathy patients with nephrostomy tubes show cancerrelated symptoms, despite implementation of the diversion.18,20,21 The main factor that should guide the management is patient desire after an honest discussion with the attending physician. Some patients may wish to prolong life even for a short time due to emotional, legal, or financial reasons, and this wish must be respected. If PN is conducted for a terminally ill woman with cervical cancer and prolongs her survival by relieving obstructive uropathy, the disease will progress naturally and complications such as recto-vaginal fistulae formation, refractory tenesmus, or digestive tract obstruction are likely to happen. These conditions are both physically and emotionally distressing but not, in themselves, life threatening, as reported by Chang.22 Death caused by renal failure is relatively peaceful in this setting. For that reason, before PN,
190
Dienstmann et al.
patients, and their families must be completely informed about the palliative role of urinary diversion, the potential complications of the procedure, the expected outcomes and anticipated disease prognosis. We conclude that most patients with advanced cervical cancer have poor performance status and short survival after PN. The indication for invasive urinary diversion in terminal cancer patients should be individualized on the basis of expectations for prolonged functional palliation. There are no clear guidelines that could help us in the decision of which patients would benefit from such intervention. Our data suggest that PN can be of clinical benefit for carefully selected patients with recurrent cervical cancer, namely those with PS 1e3.
Acknowledgments The authors would like to thank Dr. Jose´ Bines for manuscript review.
References 1. Instituto Nacional de Caˆncer (INCA, Rio de Janeiro). Caˆncer no Brasil: dados dos registros de base populacional. Available at http://www.inca.gov.br/ estimativa/2008. Accessed March 21, 2008. 2. Sutton GP, Blessing JA, DiSaia PJ, McGuire WP. Phase II study of ifosfamide and mesna in nonsquamous carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1993;49:48e50. 3. Verschraegen CF, Levy T, Kudelka AP, et al. Phase II study of irinotecan in prior chemotherapy-treated squamous cell carcinoma of the cervix. J Clin Oncol 1997;15:625e631. 4. Bookman MA, Blessing JA, Hanjani P, Herzog TJ, Andersen WA. Topotecan in squamous cell carcinoma of the cervix: a Phase II study of the Gynecologic Oncology Group. Gynecol Oncol 2000;77:446e449. 5. Muderspach LI, Blessing JA, Levenback C, Moore JL Jr. A Phase II study of topotecan in patients with squamous cell carcinoma of the cervix: a gynecologic oncology group study. Gynecol Oncol 2001;81:213e215. 6. Muggia FM, Blessing JA, Method M, et al. Evaluation of vinorelbine in persistent or recurrent squamous cell carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92: 639e643. 7. Homesley HD, Vaccarello L, Lowendowski G, Elbendary AA. A phase II trial of weekly one hour paclitaxel as second-line therapy for endometrial
Vol. 36 No. 2 August 2008
and cervical cancer. Proc Am Soc Clin Oncol 2006;24(18S). [abstract 15013]. 8. Watkinson AF, A’Hern RP, Jones A, King DM, Moskovic EC. The role of percutaneous nephrostomy in malignant urinary tract obstruction. Clin Radiol 1993;47:32e35. 9. Soper JT, Blaszczyk TM, Oke E, Clarke-Pearson D, Creasman WT. Percutanous nephrostomy in gynecologic oncology patients. Am J Obstet Gynecol 1988; 158:1126e1131. 10. Gadducci A, Madrigali A, Facchini V, Fioretti P. Percutaneous nephrostomy in patients with advanced or recurrent cervical cancer. Clin Exp Obstet Gynecol 1994;21:71e73. 11. Chan S, Robinson AC, Johnson RJ. Percutaneos nephrsotomy: its value in obstructive uropathy complicating carcinoma of cervix uterus. Clin Oncol (R Coll Radio) 1990;2:156e158. 12. Emmert C, Rassler J, Kohler U. Survival and quality of life after percutaneous nephrostomy for malignant ureteric obstruction in patients with terminal cervical cancer. Arch Gynecol Obstet 1997; 259:147e151. 13. Romero FR, Broglio M, Pires SR, et al. Indications for percutaneous nephrostomy in patients with obstructive uropathy due to malignant urogenital neoplasias. Int Braz J Urol 2005;31:117e124. 14. Harrington KJ, Pandha HS, Kelly SA, et al. Palliation of obstructive nephropathy due to malignancy. Br J Urol 1995;76:101e107. 15. Lutaif NA, Yu L, Abdulkader RC. Factors influencing the non-recovery of renal function after the relief of urinary tract obstruction in women with cancer of the cervix. Ren Fail 2003;25:215e223. 16. Kinn AC, Ohlsen H. Percutaneous nephrostomyda retrospective study focused on palliative indications. APMIS Suppl 2003;109:66e70. 17. Chung SY, Stein RJ, Landsittel D, et al. 15-year experience with the management of extrinsic ureteral obstruction with indwelling ureteral stents. J Urol 2004;172:592e595. 18. Shekarriz B, Shekarriz H, Upadhyay J, et al. Outcome of urinary diversion in the treatment of advanced malignancies. Cancer 1999;85:998e1003. 19. Lewis S, Patel U. Major complications after percutaneous nephrostomydlessons from a department audit. Clin Radiol 2004;59:171e179. 20. Meyer JE, Green TH Jr, Yatsuhashi M. Palliative urinary diversion in carcinoma of the cervix. Obstet Gynecol 1980;55:95e98. 21. Little B, Ho KJ, Gawley S, Young M. Use of nephrsotomy tubes in ureteric obstruction from incurable malignancy. Int J Clin Pract 2003;57:180e181. 22. Chang HL, Lim HW, Su FH, Tsai ST, Wang YW. Win or lose? Percutaneous nephrostomy for a terminal-stage cervical-cancer patient featuring obstructive uropathy. J Palliat Care 2006;22:57e60.