Gastric Cancer Patients with End-stage Renal Disease Who Underwent Radical Gastrectomy

Discussion

Radical resection of GC remains the treatment of choice and provides a chance of cure, even if extended resection may be required (11). Age, tumor depth, nodal status, distant metastasis and resection margin are consistently reported as independent prognostic factors for GC patients undergoing curative resection (11). Our previous study confirmed that old age, lymph node metastasis, serosal invasion, peritoneal seeding, positive resection margin, liver metastasis, tumor size and lymphatic invasion were independent prognostic factors for GC patients undergoing radical resection. The 5-year survival rate for all resected patients was 45.6% and 57.0% for curative resection (5, 13). Although several large series reported the prognostic analysis on GC undergoing radical gastrectomy (14-16), the influence of associated ESRD on GC patients who underwent curative resection has not been well-clarified.

Impaired renal function may be combined with several medical diseases and have potential risks for operation. Mori et al. reported the safety and outcome of gastrectomy for GC patients with nonuremic renal failure (17). These patients could achieve similar operation time, blood loss, complications, hospital stay and survival rates to those without renal function impairment (17).

Regarding patients with uremic renal failure, higher risks for operative procedures, due to associated multiple systemic diseases, are always considered as an obstacle to any invasive procedure (6).

The incidence of cancer in ESRD patient is about three-times higher than that of the normal population but the incidence of GC in ESRD patients is not significantly increased over that of the normal population (9), except elderly patients. As demonstrated in this study, the incidence of GC with ESRD was 1.28% (26/2,021). Because of the lower incidence, most surgeons have limited information for this kind of patients.

In the present report, it is worth noting that the ESRD patients with GC had distinct clinicopathological charactistics, including lower hemoglobin levels, high preoperative comorbidity, more early-stage tumors, less lymph node metastasis, less perineural invasion, more well-differentiated tumors and more intestinal types. Stages I and II GC comprised of 53.8% (n=14/26) and 26.9% (n=7/26) of ESRD patients with GC respectively. In Taiwan, ESRD patients usually have regular hemodialysis three-times a week. Most patients of GC with ESRD were diagnosed with anemia or dizziness. Aggressive examination using panendoscope or colonofiberscope to discover the reason for anemia might partly explain the higher rate of early-staged GC in the ESRD-GC group patients of our study. This phenomenon is well-demonstrated by Japan's screen program for GC in Japan. In Japan, early-stage gastric cancer accounts for 72.6% of GC patients under going gastrectomy due to their screening system (18). In our study, early-stage gastric cancer in the ESRD-GC group (stages I and II) was about 80.7%. The incidence is similar to that in Japan (18). The early diagnosis of GC with ESRD might be attributed to the regular hemoglobin follow-up and aggressive investigation for blood loss in the gastro intestinal tract.

Unfortunately, the surgical morbidity and hospital mortality rates are higher for GC-ESRD patients than the GC group without ESRD (30.8%, 14.7% versus 11.5 %, 2.8%, respectively). The most common causes of complication's or mortality are leakage and post-operative pancreatitis. The TNM stage of the three cases with mortality was T4aN0M0, T4aN1M0 and T4bN3bM0, respectively. For these patients with advanced disease, extensive lymph node dissection inducing blood supply insufficiency, especially for patients with renal failure, would produce poor healing process in the duodenum stump and pancreas capsule might explain technical difficulty. Impaired renal function combined with several medical diseases may partly contribute to the higher surgical morbidity or even mortality. For surgeons, the surgical technique for extensive dissection and degree of tumor invasion might mainly explain the reasons of complications. Post-operative pancreatitis is mostly caused by pancreatic tissue injury, while LN dissection or pancreatic anterior capsule dissection and leakage are determined by the quality of blood supply and invasion of tumor. Blood supply should be preserved if the invasiveness of the tumor is severe. Although it might be difficult to prevent injury of pancreas and blood supply of duodenum in advanced patients, delicate surgical skill is required to minimize the injury. Neoadjuvant chemotherapy for GC-ESRD in advanced stages to decrease morbidity and even mortality by tumor down-staging might be helpful; however, no report has been published to date. On the other hand, to balance surgical risk and oncological clearance, GC-ESRD patients with early-stage cancer status may undergo treatment to a lesser extent, such as endoscopic mucosal resection, endoscopic submucosal dissection or laparoscopic wedge resection, etc.

Regarding the impact of ESRD on survival for GC after radical gastrectomy, ESRD-GC and GC group patients had similar survival rates. However, the ESRD-GC group of patients tended to have better survival, although the trend was not significant due to the limited number of cases.

This information is important to surgeons when they encounter patients with GC associated with ESRD. The majority of the ESRD-GC group patients had early-stage GC but the surgical risk is high in advanced patients. Newman et al. reported that emergent abdominal surgery in patients (n=21) with chronic renal failure (CRF) was associated with poor survival rates. Surgical (30-day) mortality for patients with CRF receiving emergent surgery was 38% and 0% for patients with CRF receiving elective surgery (5 patients) (8). Martínez et al. emphasized that morbidity and mortality after GC surgery was influenced by preoperative conditions of patients. Statistically significant risk factors for mortality were the Goldman cardiac risk index, albumin, creatinine and total lymphocyte count (7). Thus, adequate pre-operative preparation for ESRD patients is important and can improve the post-operative survival rate. We should acomplish adequate pre-operative preparation, perform delicate surgery and excellent post-operative care to minimize and even eliminate surgical complications and mortality. This way, the ESRD-GC patients will have similar survival; even better survival than that for normal populations.

Although our results may support both benefit and feasibility for radical gastrectomy for GC-ESRD, especially in early-stage patients, several problems need to be addressed. First, the case number of the study group is quite small. Second, this study is a retrospective case control study and, therefore, selection bias is inevitable. Third, we cannot compare treat ment efficacy for ESRD-GC patients between surgery and conservative treatment due to lack of complete data, because some patients without surgery chose no treatment and are thus lost for follow-up. Sometimes, surgeons may consider patients with GC-ESRD for radical gastrectomy and this is probably less feasible due to multiple comorbidities and change to conservative treatment, especially for surgery in the presence of advanced disease stage. However, this is the first study to address this specific clinical issue.

In conclusion, radical gastrectomy is feasible and beneficial for GC-ESRD patients, especially in early stage, with thorough pre-operative evaluation and postoperative care. However, radical gastrectomy for the treatment of GC in the ESRD-patient group should be more cautiously considered, otherwise any benefit will be compromised by higher postoperative complications and even mortality.