Impact of chronic kidney disease stage on morbidity after gastrectomy for gastric cancer

Abstract Aim The outcomes of gastrectomy for gastric cancer in patients at each severity of chronic kidney disease (CKD) remain unknown. Methods We retrospectively analyzed the outcomes of 560 patients who underwent distal or total gastrectomy for gastric cancer between 2009 and 2018. We classified the patients into four groups based on estimated glomerular filtration rate: stage 1/2 (normal to mild, n = 375), stage 3a (mild to moderate, n = 122), stage 3b (moderate to severe, n = 43), and stage 4/5 (severe to end‐stage, n = 20) CKD. The relationship between CKD stage and the incidence of postoperative morbidity was analyzed. Results CKD was a predictor of overall morbidity independent of age, gender, American Society of Anesthesiologists Performance Status, pulmonary comorbidity, extent of lymphadenectomy, and operation time in a multivariate analysis. The incidences of overall and severe morbidity were significantly increased with CKD progression (both P < .001). Compared to stage 1/2 CKD, the odds of overall morbidity were significantly elevated in stage 3a (odds ratio [OR] 1.87, P = .007), stage 3b (OR 3.86, P < .001), and stage 4/5 (OR 8.60, P < .001). The risk of procedure‐related morbidity was strikingly increased in stage 3b (OR 2.93, P = .004). The risk of procedure‐unrelated morbidity elevated markedly in stage 3a (OR 2.77, P = .001). A significant graded association between CKD progression and overall morbidity was also revealed in elderly patients (P = .001). Conclusions The severity of CKD predicts the likelihood and type of morbidity after gastrectomy and can guide surgical decision‐making for patients with gastric cancer.


| INTRODUC TI ON
Gastric cancer is one of the leading causes of cancer deaths throughout the world and continues to increase as populations in developed countries age. 1 Gastrectomy is the mainstay of treatment for gastric cancer, but its postoperative morbidity impairs the patient's physical condition and often leads to an unfavorable clinical course.
Infectious morbidity was also suggested to worsen long-term outcomes after gastrectomy. [2][3][4] The identification of a high-risk candidate who is likely to develop postoperative morbidity is crucial to the planning of surgical treatment for gastric cancer.
Chronic kidney disease (CKD) is increasingly prevalent globally. 5,6 In Japan, the prevalence rate of end-stage renal disease is the highest among the world, 7 and the number of patients with CKD reached 13 million in 2005. 8 A recent large analysis revealed that the incidence risk of gastric cancer in patients with CKD was significantly higher than in the general population. 9 Surgery for gastric cancer patients with CKD is expected to increase in the future. Renal dysfunction adversely affects a patient's postoperative course due to impairment of physical function including wound healing, immunity, body fluid regulation, and hemostasis. The presence of end-stage renal disease has been shown to be accompanied by a higher risk of morbidities after gastroenterological cancer treatment, including gastric cancer treatments. [10][11][12][13][14][15] However, the impact of mild to moderate CKD on the incidence and type of morbidity after gastrectomy is unclear. 16 The glomerular filtration rate (GFR) is used in both the definition and the staging of CKD. 17 The estimated GFR (eGFR) tends to decline with increasing age, and the CKD population based on GFR increases among elderly patients. However, the influence of CKD that is detectable with the use of eGFR in elderly patients undergoing gastrectomy is unknown. We conducted the present study to retrospectively analyze the association between the severity of CKD and the development of morbidity after gastrectomy for gastric cancer, and to elucidate the effect of CKD on surgical outcomes in elderly gastric cancer patients.

| Patients
From January 2009 to August 2018, a total of 776 patients underwent a gastrectomy for gastric cancer at Kobe University Hospital.
Of these, the patients who underwent a proximal gastrectomy, which had less postoperative morbidities in comparison with other type of gastrectomy, and those who received preoperative chemotherapy, which has a potential to increase morbidities, were excluded. We analyzed the cases of the remaining 560 patients who underwent a distal or total gastrectomy with curative intent, using the patients' clinicopathological and treatment data extracted from medical charts. Each patient's preoperative physical status was assessed by age, gender, body mass index (BMI), serum albumin concentration, American Society of Anesthesiologists Performance Status (ASA-PS), Charlson comorbidity index (CCI), and the presence or absence of diabetes mellitus, pulmonary disease, cardiovascular disease, and a history of anticoagulant agent(s) use. Tumor stage was classified according to the Japanese Classification of Gastric Carcinoma. 18 Surgical procedures were performed according to the Japanese Gastric Cancer Treatment Guidelines. 19 Postoperative morbidities diagnosed as grade II or higher on the Clavien-Dindo classification were defined as procedure-related or procedure-unrelated morbidities. 20 Severe morbidities were defined as morbidities diagnosed as grade III or higher. Procedure-related morbidities included anastomotic leakage, pancreatic fistula, intraabdominal abscess, intra-abdominal bleeding, anastomotic bleeding, delayed gastric emptying, anastomotic stricture, small bowel obstruction, and wound infection. The procedure-unrelated morbidities were categorized as follows: cardiovascular morbidity (including arrhythmia, ischemic heart disease, heart failure, and stroke), pulmonary morbidity (including pneumonia, atelectasis, pleural effusion, and edema), other-site infectious morbidity (including cholecystitis, urinary tract infection, enterocolitis, and catheter-related infection), and hemostatic morbidity (including venous thromboembolism and intestinal bleeding). All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Declaration of Helsinki. line. 17 Based on the patient's eGFR measured at his or her first visit to our hospital, we classified the patients into the following four groups: stage 1/2 CKD (normal to mild CKD: eGFR ≥ 60 ml/min/1.73 m 2 ), stage 3a CKD (mild to moderate CKD: eGFR 45 to < 60 ml/ min/1.73 m 2 ), stage 3b CKD (moderate to severe CKD: eGFR 30

| Statistical analyses
Statistical comparisons among different groups were performed with the Mann-Whitney U-test and Spearman rank correlation for categorical variables, and with an analysis of variance (ANOVA) or Kruskal-Wallis test for continuous variables, as appropriate. Potential risk factors for postoperative morbidities were evaluated by a logistic regression analysis. Factors with a probability level ≤ 0.05 were adopted for the subsequent multivariate analysis. Independent risk factors were considered appropriate at a probability level < 0.05. All statistical analyses were performed using JMP statistical software, version 11 (SAS, Cary, NC, USA).

| Clinicopathological characteristics
The study cohort consisted of 415 males (74.1%) and 145 females In the stage 4/5 CKD group, eight patients were on hemodialysis preoperatively. The patient characteristics of the four groups are summarized in Table 1.
The stage 1/2 group had a significantly younger mean age and a significantly lower proportion of males compared to the other three groups (P < .001 and P = .016, respectively). The serum albumin concentrations in the stage 3b and 4/5 groups were significantly lower than those in the stage 1/2 and 3a groups (P < .001). Patients with higher ASA-PS and with higher CCI values were significantly more common in the stage 3b and 4/5 groups than in the stage 1/2 and 3a groups (both P < .001). Patients with advanced stage CKD showed significant increases of concomitant diabetes mellitus and cardiovascular disease (P = .012 and P < .001, respectively), but not pulmonary disease. Patients with stages 1 or 2 were less likely to take anticoagulant agents than those with the other three stages (P < .001). The stage 3a, stage 3b, and stage 4/5 groups had a histologically differentiated tumor significantly more frequently than the stage 1/2 group (P = .008).
With respect to surgical procedures, laparoscopic-approach surgery significantly decreased with stage progression (P = .004). The extent of gastrectomy or lymph node dissection did not differ among the four groups.

| Postoperative morbidity
The patients' CKD stages and postoperative morbidity data are summarized in Table 2 (Figure 1).

| Elderly patients
Elderly patients (i.e. those aged ≥ 75 years) accounted for 35% (n = 196) of the total cohort, and 54.6% (n = 107) of the elderly patients had CKD. Overall morbidity significantly increased with CKD stage progression among the elderly patients (P = .01, Table 4). Both procedure-related and procedure-unrelated morbidities were significantly associated with CKD stage progression in this patient group (P = .047 and P = .007, respectively). Severe morbidity showed a tendency to increase with CKD stage progression, but the difference did not reach significance.

| D ISCUSS I ON
As most previous articles merely reported that the incidence of morbidity after gastrectomy increased in gastric cancer patients with CKD, the effects of the severity of CKD on what type of morbidity remain unclear. To our knowledge, the present study is the first report that demonstrates a graded association between distinct CKD stages and the incidence and type of morbidity after gastrectomy.
A possible explanation for the strong positive correlation between CKD progression and adverse outcomes is that the severity of CKD may reflect a patient's health impairment as well as his or her renal dysfunction. We observed that CKD stage progression was associated with higher ASA-PS, higher CCI values, and the presence of comorbidities. This finding is in accord with a report demonstrating that frailty was strongly associated with all stages of CKD. 22 Moreover, CKD is considered to be a cause of sarcopenia, and malnutrition is often concurrent. [23][24][25][26] Considering these findings, the progression of and altered drug metabolism. 30 Therefore, nutritional intervention for CKD patients has a dilemma for prevention of morbidity and CKD progression. The efficacy of nutritional therapy with protein restriction for CKD patients who undergoing gastrectomy is uncertain.
The severity of CKD is defined by the GFR value. The GFR tends to decline with aging. In the present patient cohort, the proportion of CKD patients among elderly patients was much higher than that in non-elderly patients (54.6% vs 21.4%). It is doubtful whether a decline in the GFR represents CKD in the elderly. CKD is known to increase the all-death rate and the rate of death due to cardiovascular causes in the elderly as well as the non-elderly, and a decline in the GFR is an independent risk factor for cardiovascular death separate from age. 32 Similarly, CKD was identified as a risk factor for developing postoperative morbidity independently of age in our multivariate analysis. We also observed a graded association between CKD stage progression and postoperative morbidity in elderly patients. GFRbased CKD can be a graded risk factor for adverse outcomes after gastrectomy across age groups.
This retrospective study at a single institution had several lim- The confirmation of these findings requires a larger study, but our results provide a strategic concept for gastric cancer surgery in CKD patients.
In conclusion, the severity of CKD can be used to predict the likelihood and type of morbidity after gastrectomy across age categories. The risk for procedure-unrelated morbidity was evident even at the mild stage of CKD. The risk for both procedure-related and -unrelated morbidities increased notably at the moderate stage of CKD with an eGFR < 45 ml/min/1.73 m 2 . CKD progression for the TA B L E 4 Postoperative morbidities stratified by CKD stage in the elderly risk stratification of patients undergoing gastric cancer surgery can guide the decision of surgical procedures.

CO N FLI C T O F I NTE R E S T
Authors declare no conflict of interests for this article.

E TH I C A L A PPROVA L
The protocol for this research project has been approved by a suitably constituted Ethics Committee of the institution and it con-