Clinical Outcome After Pulmonary Resection for Lung Cancer Patients on Hemodialysis

Clinical Outcome After Pulmonary Resection for

Lung Cancer Patients on Hemodialysis GENERAL THORACIC Toshiro Obuchi, MD, Wakako Hamanaka, MD, Yasuhiro Yoshida, MD, Jun Yanagisawa, MD, Daisuke Hamatake, MD, Takeshi Shiraishi, MD, and Akinori Iwasaki, MD Department of Thoracic, Breast, Endocrine, and Pediatric Surgery, Fukuoka University Hospital, Fukuoka, Japan

Background. The number of operations for patients tients had been treated for previous carcinoma. The with malignant tumors receiving long-term hemodialysis histopathologic diagnoses were adenocarcinoma in 9 has been increasing; however, there are only few reports patients and squamous cell carcinoma in 2. Procedures about pulmonary resection for the patients with lung included lobectomy in 9, pneumonectomy in 1, and cancer. wedge resection in 1. There were no in-hospital deaths. Methods. Between 1995 and 2009, 11 hemodialysis Postoperative morbidity included 2 cases of pneumonia patients (6 men, 5 women; mean age, 66.4 years) with and 1 of chylothorax. At the time of our investigation, 6 non-small cell lung cancer underwent pulmonary resec- patients were dead; 2 of cancer and 4 of noncancer causes. tion at our institution. We retrospectively evaluated their The overall 5-year survival rate of 11 patients was 28.0%. postoperative clinical outcomes and long-term results. Conclusions. Hemodialysis is not a contraindication to Results. The underlying kidney conditions included lung resection, despite the high morbidity rate. Surgical nephrosclerosis in 3, diabetic nephropathy in 3, glomer- treatments, including lobectomy, remain one of effective ulonephritis in 1, and polycystic kidney in 1; 3 patients treatments for patients on hemodialysis with lung cancer. had undergone nephrectomy. The median duration of (Ann Thorac Surg 2009;88:1745–8) hemodialysis preoperatively was 5.0 years. Three pa- © 2009 by The Society of Thoracic Surgeons

wing to technologic progress made in hemodialysis, years (range, 0.5 to 23 years), with a mean duration of 6.7 O the number of patients in a long-term hemodialy- years. sis has been increasing; therefore, it is reasonable to Pulmonary resections were performed through a pos- assume that the number of hemodialysis patients with terolateral thoracotomy or using the video-assisted tho- lung cancer in need of pulmonary resection should racic surgical (VATS) technique, which are the standard increase [1–3]. However, only few reports have detailed procedures in our institution. the clinical outcomes after pulmonary resection for he- In all patients, hemodialysis through the shunt was modialysis patients with non-small cell lung cancer routine 3 days a week, on Tuesday, Thursday, and Satur- [4–6]. As such, we evaluated the clinical outcomes and day, using heparin sodium. All operations were per- long-term results of these patients. formed on Wednesday so that the patients underwent hemodialysis on the day before the operation and the day Patients and Methods after. No additional hemodialysis was done on the operative day. For the perioperative hemodialysis, nafamostat This study protocol was examined and approved by the mesylate was used instead of heparin sodium until the department Research Review Board. Before the opera- chest drainage tube was removed and hemostasis was tion, we had obtained a written consent from all patients conﬁrmed. For priming the circuit of hemodialysis, 20 mg giving us permission to perform research using their of nafamostat mesylate was used, and 30 mg/h of nafa- data, with an understanding that their privacy would be mostat mesylate was continuously added into the circuit protected. The board concluded that this study does not during hemodialysis. violate patient privacy and approved the consent form. Antibiotics were routinely given intravenously, intra- Between January 1995 and January 2009, 11 patients (6 operatively, and postoperatively. We administered 1 men, 5 women) with non-small cell lung cancer who were gram/d of an antibiotic such as piperacillin, cefazolin, or also receiving hemodialysis underwent lung resection at cefotiam. On the operative day, the antibiotic was intra- our institution. We retrospectively reviewed their data, venously dripped at the start of the operation. Postoper- and their characteristics are reported in Table 1. The atively, the antibiotic was given for a few days. median preoperative duration of hemodialysis was 5.0 We investigated the 11 patients in terms of respiratory Accepted for publication Aug 6, 2009. functions, serum levels of urea nitrogen and creatinine, Address correspondence to Dr Obuchi, Jonan-ku, Nanakuma, 7 chome, underlying kidney condition, histology, status of lung 45-1, Fukuoka, 814-0180, Japan; e-mail: [email protected]. cancer and history, comorbidity, history of smoking,

© 2009 by The Society of Thoracic Surgeons 0003-4975/09/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2009.08.010 EEA THORACIC GENERAL 1746 OBUCHI ET AL Ann Thorac Surg OUTCOMES OF HEMODIALYSIS WITH LUNG CANCER 2009;88:1745–8

Table 1. Patient Characteristics were determined as IA in 6 patients, IB in 3, IIB in 1, and IIIA in 1. Variable Result Treatment history revealed that 3 patients had already Patients, No. 11 been treated for previous primary carcinoma, comprising Gender, No. colon carcinoma in 2 and renal carcinoma in 1. Also, 6 Male 6 patients were current smokers at the time of admission. Female 5 Procedures included lobectomy in 9 patients, pneumo- Age, mean (range), y 66.4 (51–89) nectomy in 1, and wedge resection for a tiny localized Follow-up, mean (range), d 995.6 (106–4975) adenocarcinoma, Noguchi type B, in 1. Operations were PathologicTNMstage, No. done through posterolateral thoracotomy in 7 patients IA 6 and by VATS in 4. The operative time and the intraop- IB 3 erative blood loss are reported in Table 3. The mean duration of postoperative thoracic drainage was 4.2 days, IIB 1 and the mean duration of postoperative hospitalization IIIA 1 was 15.6 days. There were no hospital deaths. Postoper- Histologic type, No. ative complications included pneumonia in 2 patients Adenocarcinoma 9 and chylothorax in 1. The morbidity rate was 27.3%. Squamous cell carcinoma 2 Six patients had died at the time of the investigation. Surgical procedure, No. Two died of lung cancer and the remaining 4 died of Lobectomy 9 noncancer causes such as acute myocardial infarction or Pneumonectomy 1 subarachnoid hemorrhage. The overall 5-year survival Wedge resection 1 rate was 28.0% (Fig 1), and the 5-year survival rate for Surgical approach, No. stage I patients was 37.5% (Fig 2). Open thoracotomy 7 VATS 4 Respiratory function, mean Ϯ SD Comment Vital capacity, L 2.91 Ϯ 0.70 Patients receiving hemodialysis have been reported to % Vital capacity 102.5 Ϯ 22.5 have a potentially higher risk of cancer, and they un- Ϯ FEV1,L 2.04 0.36 dergo more operations in their lifetime than others [2–4]. Ϯ FEV1,% 78.7 11.1 Maisonneuve and colleagues [3] reported high risks Serum values after hemodialysis associated with cancer of the kidney, bladder, and endo- Urea nitrogen, mean Ϯ SD, mg/dL 29.4 Ϯ 16.4 crine organs in patients undergoing hemodialysis, but no Creatinine, mean Ϯ SD, mg/dL 4.00 Ϯ 2.39 increase in the incidence of cancer of the lung, colon or Hemodialysis duration, median (range), y 5.0 (0.5–23) rectum, breast, and stomach. Kantor and colleagues [5] also reported that no increase in the number of patients ϭ ϭ FEV1 forced expiratory volume in 1 second; SD standard devia- with lung cancer was found in hemodialysis patients, so ϭ tion; VATS video-assisted thoracic surgery. the relative risk for these patients was 0.8. However, patients receiving hemodialysis are generally thought to surgical procedure, operative time, volume of intraoper- have weakened anticancer immune systems [4]. In fact, 3 ative blood loss, duration of postoperative thoracic drain- patients (27.3%) in our study had a treatment history of age, duration of postoperative hospitalization, postoper- another primary carcinoma. It is unclear whether this ative complication, patient prognosis, and cause of death. rate of double cancer (27.3%) in our limited experience Data for the 11 patients were statistically analyzed has some clinical meaning, but this rate seems relatively using StatMate software (ATMS Inc, Tokyo, Japan). Us- high. ing this software, we plotted Kaplan-Meier curves. At our institution, the patients on hemodialysis could safely undergo even major lung resections with our usual procedures, under strict patient selection. It was reason- Results able that the mean volume of intraoperative blood loss The mean duration of follow-up was 995.6 days (range, was about 150 grams, and the mean duration of thoracic 106 to 4975 days). The mean serum levels of urea nitrogen drainage was about 4 days. We think that perioperative and creatinine after hemodialysis were 29.4 mg/dL and usage of nafamostat mesylate, which is an ultra-short- 4.0 mg/dL, respectively. acting anticoagulant agent used in hemodialysis, is more The underlying kidney conditions included nephro- effective than heparin sodium in controlling intraopera- sclerosis in 3 patients, diabetic nephropathy in 3, glomer- tive and postoperative bleeding, as Tsuchida and col- ulonephritis in 1, and polycystic kidney in 1 (Table 2). In leagues [6] also reported. The administration of hemodi- addition, 3 patients had undergone nephrectomy. The alysis with nafamostat mesylate instead of heparin histologic diagnosis was adenocarcinoma in 9 patients sodium did not negatively affect hemostasis periopera- and squamous cell carcinoma in 2. Although preoperatively for the present 11 patients. tive clinical stage was evaluated to be IA or IB for all A high morbidity rate after pulmonary resection for patients, the postoperative pathologic stages of cancer patients on hemodialysis has been reported [6].We Ann Thorac Surg OBUCHI ET AL 1747 2009;88:1745–8 OUTCOMES OF HEMODIALYSIS WITH LUNG CANCER

Table 2. Preoperative Characteristics of the 11 Hemodialysis Patients

Underlying Renal Smoking Pt Age Sex Histology Stage History, Comorbidity Condition Pack-years GENERAL THORACIC 1 51 F Ad IA ... Glomerulonephritis 0 2 68 F Ad IB ... Nephrosclerosis 0 3 74 M Ad IIIA ... Polycystic kidney 60 4 89 M Ad IA ... Nephrosclerosis 60 5 71 M SCC IB DM, PCI, AF Diabetic nephropathy 40 6 60 M Ad IA DM Diabetic nephropathy 30 7 58 F Ad IA Renal cancer Postnephrectomy 0 8 68 M SCC IIB Renal tuberculosis Postnephrectomy 48 9 55 F Ad IA Colon cancer, renal hematoma Postnephrectomy 0 10 73 M Ad IB Colon cancer, DM, PCI Diabetic nephropathy 99 11 63 F Ad IA ... Nephrosclerosis 0

Ad ϭ adenocarcinoma; AF ϭ atrial fibrillation; DM ϭ diabetes mellitus; F ϭ female; M ϭ male; PCI ϭ percutaneous coronary intervention; SCC ϭ squamous cell carcinoma. found complications in 3 of 11 patients. For 2 of those 3 also difficult to choose other methods of treatment, such patients, complications of pneumonia resulted in pro- as chemotherapy or best supportive care, in the case like longed durations of postoperative hospital stays. The this patient, who had early-stage cancer and was in good 27.3% morbidity rate in our series was high, even though performance status. Radiotherapy is also a treatment no fatal complications occurred. Thus, hemodialysis pa- option, but it is not easy to choose radiotherapy when a tients undergoing lung resection should be closely mon- better curability can be expected from surgery. Although itored postoperatively for symptoms of complications lung wedge resection is usually a good surgical treatment such as pneumonia. option, depending on the location of tumor, lobectomy Japan is one of the countries with the longest life was the only option in this patient. Generally, lobectomy expectancy at birth in the world, and because the elderly has its advantage in the low recurrence rate after the population is increasing as whole, the number of elderly procedure, with simple postoperative follow-up and patients with lung cancer has been increasing. Currently, without the need for continuous treatments, especially lobectomies are being performed even on extremely for those with stage I lung cancer. With respect to our elderly patients, such as octogenarians, provided that 89-year-old patient, the reason why we performed lobec- they are in good performance status, the curability is tomy was that first, he was healthy for his age and appropriate, and their quality of life is not negatively expressed a strong desire to have lobectomy over other affected [7]. options. Second, the average life span of 89-year-old Our study included an 89-year-old patient who under- Japanese men is 4.70 years. In fact, he was discharged 12 went lobectomy through a posterolateral thoracotomy. days after lobectomy without any complications. At least Although it is generally difficult to evaluate the operative in this case, we think lobectomy was one of reasonable indication of someone who is as old as this patient, it is options.

Table 3. Postoperative Characteristics of the 11 Hemodialysis Patients

Operation Blood Duration of Post-op Time Loss Drainage LOS Survival Pt Procedure (min) (g) (days) (days) Complication (mon) Follow-up Cause of Death

1 Lobectomy 220 105 5 10 . . . 165.8 Alive ... 2 Lobectomy 260 75 4 15 . . . 12.5 Dead Local recurrence 3 VATS lobectomy 350 215 3 15 . . . 16.3 Dead Distant metastasis 4 Lobectomy 205 100 2 12 . . . 24.1 Dead Heart failure 5 Lobectomy 270 315 4 10 . . . 18.3 Alive ... 6 VATS lobectomy 130 40 5 13 . . . 35.1 Dead Unknown (noncancer) 7 Lobectomy 170 110 4 18 . . . 5.2 Dead SAH 8 Pneumonectomy 200 374 1 27 Pneumonia 27.6 Dead AMI 9 VATS wedge resection 35 5 1 3 . . . 39.4 Alive ... 10 Lobectomy 170 150 8 36 Pneumonia 17.1 Alive ... 11 VATS lobectomy 150 100 9 13 Chylothorax 3.5 Alive ...

AMI ϭ acute myocardial infarction; LOS ϭ length of stay; SAH ϭ subarachnoid hemorrhage; VATS ϭ video-assisted thoracic surgery. EEA THORACIC GENERAL 1748 OBUCHI ET AL Ann Thorac Surg OUTCOMES OF HEMODIALYSIS WITH LUNG CANCER 2009;88:1745–8

In our limited study, the 5-year survival rate was 28.0% for the 11 hemodialysis patients with lung cancer undergoing pulmonary resection and 37.5% for the 9 patients with stage I lung cancer. Although it is problematic to make a simple comparison between our results and the results of patients not receiving hemodialysis [8], it is still necessary to ﬁnd a clinical meaning from our results. One possible explanation is that arteriosclerosis, which is largely a cardiovascular disease, has a great effect on morbidity and mortality for hemodialysis patients [9]. Ohtake and colleagues [9] reported that cardiac death accounts for almost 40% of total deaths among hemodialysis patients, and coronary angiography showed signif- Fig 2. Long-term survival after pulmonary resection is shown for he- icant coronary artery stenosis in 53.3% of 30 asymptom- modialysis patients with stage I lung cancer. The 5-year survival rate atic chronic kidney disease patients at the start of was 37.5%. hemodialysis. Among the many complications induced by long-term hemodialysis are infections, chronic heart failure, and References arteriosclerosis of cerebral and cardiac arteries. In fact, 4 of the 6 deaths in our study were of noncancer causes, 1. Pinson CW, Schuman ES, Gross GF, et al. Surgery in including arteriosclerosis. The prognosis for those pa- long-term dialysis patients: Experience with more than 300 tients after lung resection might have been inﬂuenced by cases. Am J Surg 1986;151:567–71. their underlying complications. Nevertheless, surgical 2. Cuckovic´ C, Djukanovic´ L, Jankovic´ S, et al. Malignant treatments are still effective for treating lung cancer. tumors in hemodialysis patients. Nephron 1996;73:710–2. In conclusion, our study revealed that hemodialysis is 3. Maisonneuve P, Agodoa L, Gellert R, et al. Cancer in patients on dialysis for end-stage renal disease: an interna- not a contraindication to lung resection, despite high tional collaborative study. Lancet 1999;354:93–9. morbidity rate; surgical intervention is still an effective 4. Ciriaco P, Casiraghi M, Melloni G, et al. Pulmonary resection treatment for patients on hemodialysis with lung cancer. for non-small-cell lung cancer in patients on hemodialysis: In our limited study, however, we were unable to refer to clinical outcome and long-term results. World J Surg 2005; differences in effectiveness between surgical intervention 29:1516–9. and radiotherapy [10] or between lobectomy and limited 5. Kantor AF, Hoover RN, Kinlen LJ, et al. Cancer in patients receiving long-term dialysis treatment. Am J Epidemiol lung resection [11]. We think that further studies are 1987;126:370–6. needed to establish the therapeutic tactics for hemodial- 6. Tsuchida M, Yamato Y, Aoki T, et al. Complications associ- ysis patients with lung cancer, especially early-stage ated with pulmonary resection in lung cancer patients on cancer. dialysis. Ann Thorac Surg 2001;71:435–8. 7. Togashi K, Koike T. Surgical treatment for patients aged 80 years or older with primary lung cancer and differences in outcomes according to sex. Kyobu Geka 2008;61:347–51. 8. Asamura H, Goya T, Koshiishi Y, et al. A Japanese Lung Cancer Registry study: prognosis of 13,010 resected lung cancers. J Thorac Oncol 2008;3:46–52. 9. Ohtake T, Kobayashi S, Moriya H, et al. High prevalence of occult coronary artery stenosis in patients with chronic kidney disease at the initiation of renal replacement therapy: an angiographic examination. J Am Soc Nephrol 2005;16: 1141–8. 10. Qiao X, Tullgren O, Lax I, Sirzén F, Lewensohn R. The role of radiotherapy in treatment of stage I non-small cell lung cancer. Lung Cancer 2003;41:1–11. 11. Chamogeorgakis T, Ieromonachos C, Georgiannakis E, Mal- lios D. Does lobectomy achieve better survival and recur- Fig 1. Kaplan-Meier curves show overall survival after pulmonary rence rates than limited pulmonary resection for T1N0M0 resection for hemodialysis patients with non-small cell lung cancer. non-small cell lung cancer patients? Interact Cardiovasc The 5-year survival rate was 28.0%. Thorac Surg 2009;8:364–72.