Ojima T, Iwahashi M, Nakamori M, Nakamura M, Naka T, Ishida K, Ueda K, Katsuda M, Iida T, Tsuji T, Yamaue H. Influence of Overweight on Patients With Gastric Cancer After Undergoing Curative GastrectomyAn Analysis of 689 Consecutive Cases Managed by a Single Center. Arch Surg. 2009;144(4):351-358. doi:10.1001/archsurg.2009.20
Overweight (body mass index [calculated as weight in kilograms divided by height in meters squared], ≥25.0) has an effect on surgical results, postoperative complications, and long-term survival in patients with gastric cancer who underwent curative gastrectomy.
Retrospective study from January 1, 1992, through December 31, 2002.
Wakayama Medical University Hospital.
This study included 689 patients who underwent curative gastrectomy (R0). Patients who underwent laparoscopic gastrectomy, gastrectomy with pancreaticoduodenectomy, gastrectomy with another organ resection (liver, colon, or ovary), or gastrectomy with thoracotomy were not included.
Main Outcome Measures
Duration of operation, amount of blood loss, incidence of postoperative complications, and survival analysis.
The mean (SD) duration of the operation was longer in the overweight group (315  minutes) than in the normal-weight group (277  minutes) (P < .001). The mean (SD) intraoperative blood loss was larger in the overweight group (882  mL) than in the normal-weight group (536  mL) (P < .001). The rates of postoperative complications (anastomotic leakage, pancreatic fistula, and intra-abdominal abscess) were significantly higher in the overweight group (P < .05). Multivariate logistic regression analysis identified that postoperative complications were significantly associated with being overweight (P = .01) and with undergoing pancreatectomy (P = .03). Disease-specific and overall survival did not show any significant difference between the 2 groups.
Being overweight is not a poor risk factor for survival in patients with gastric cancer, although it is independently predictive of postoperative complications.
In a Japanese randomized controlled trial, Japan Clinical Oncology Group Study 9501 (JCOG9501), there was no difference in the incidence of major postoperative complications between the standard D2 dissection and D2 with extended para-aortic dissection (D3) in patients with gastric cancer (hereinafter referred to as gastric cancer patients) undergoing gastrectomy.1 However, 2 large European randomized controlled trials that compared D1 and D2 dissections reported an increase in surgical morbidity and mortality in the D2 dissection group.2,3 These studies failed to show a survival benefit in the D2 dissection group. Japanese patients are generally slender, with a lower body mass index (BMI) (calculated as weight in kilograms divided by height in meters squared) than white patients. These differences in patient physique may partly explain the high mortality and morbidity. In fact, the JCOG9501 data showed that overweight patients with a BMI greater than 25.0 are at increased risk for the postoperative complications of abdominal abscess and pancreatic fistula after gastrectomy with D2 dissection.4
The following are critiques of the JCOG9501 trial. First, there was an extremely low number of patients with concomitant disease because the eligibility criteria for this trial were severe. Therefore, the JCOG9501 trial results do not reflect the risk of the postoperative complications of overweight patients with serious concomitant disease. Second, this trial did not include D1 dissection. Japan, compared with Western countries, has a higher number of patients with early gastric cancer. Approximately half of the gastric cancer patients have mucosal or submucosal gastric cancer at the time of diagnosis.5 According to the Japanese Research Society for Gastric Cancer Rules, the extent of lymphadenectomy should be a modified D1 for mucosal gastric cancer.6 In Western countries, a traditional D1 dissection is performed on 95% of gastric cancer patients, regardless of the tumor stage.7 Therefore, we considered it essential to clarify the effects of overweight on postoperative complications in consecutive gastric cancer patients.
The survival of gastric cancer patients is related to various factors, such as the histological aggressiveness of the tumor, patient age, and the preoperative medical and nutritional conditions of the patient.8 The relationship between being overweight and the overall prognosis in gastric cancer patients is an important issue to resolve, and this relationship remains controversial.9- 12
For these reasons, our study was conducted to investigate the effects of overweight on the duration of operation, amount of blood loss, incidence of postoperative complications, and survival rate in gastric cancer patients who underwent curative gastrectomy.
From January 1, 1992, through December 31, 2002, a total of 1058 patients underwent surgery for gastric cancer at Wakayama Medical University Hospital. This study included 689 patients who underwent curative gastrectomy (International Union Against Cancer R0 resection), which is defined as an absence of residual tumor microscopically.13 Mean (SD) patient age was 63.6 (12.3) (range, 24-95) years. There were 497 men and 192 women. These 689 patients were followed up for at least 5 years or until death. Patients with cancer in another organ or patients who underwent laparoscopic gastrectomy, gastrectomy with pancreaticoduodenectomy, gastrectomy with an additional organ resection (liver, colon, or ovary), or gastrectomy with thoracotomy were excluded. None of the patients received preoperative neoadjuvant chemotherapy. Patient height and body weight were measured preoperatively, and the BMI standard calculation was as recommended by the National Institutes of Health Consensus Development Conference in 1985 as an accurate index for the prediction of medically significant obesity.14 Patients with a BMI of 25.0 or greater were classified as overweight by World Health Organization criteria.15 According to National Institutes of Health criteria, a BMI of 25.0 to 29.9 is classified as overweight, and a BMI of 30.0 or greater is classified as obesity.16 In this study, patients were assigned to the following 2 groups according to their BMI: BMI of less than 25.0 (BMI <25.0 group) and BMI of 25.0 or greater (BMI ≥25.0 group).
Patients with clinically diagnosed hypertension and patients with cardiovascular disease, such as angina pectoris, or old myocardial infarction were defined as having cardiovascular disease. Patients with abnormal pulmonary function on spirograms (vital capacity ratio, <0.7; or the ratio of forced expiratory volume in 1 second to forced vital capacity, <0.6) were defined as having comorbidity with pulmonary disease.17 Patients with an estimated creatinine clearance rate lower than 60 mL/min or a rising serum creatinine level (>2 mg/dL [to convert to micromoles per liter, multiply by 88.4]) were defined as having renal dysfunction.18 Patients with liver cirrhosis (defined using the Child-Pugh classification), patients receiving treatment for liver disease, and patients with a serum aspartate aminotransferase level of greater than 2 times the upper limit of normal serum levels were defined as having liver dysfunction.19 Diabetes mellitus was noted if the patient had a fasting blood glucose concentration of more than 126 mg/dL (to convert to millimoles per liter, multiply by 0.0555) or was receiving antidiabetic therapy. Otherwise, the results of a 75-g oral glucose tolerance test were used to diagnose diabetes mellitus.20,21 Anemia was defined as a total hemoglobin level of less than 11 g/dL (to convert to grams per liter, multiply by 10).22
Standard radical open gastrectomy was performed in all 689 patients. Distal gastrectomy was performed in 398 patients; total gastrectomy, in 258 patients; and proximal gastrectomy, in 33 patients. The extent of lymph node dissection was adjusted for the location of the primary tumor according to the general rules of the Japanese Research Society for Gastric Cancer.6 In Japan, systemic D2 lymph node dissection is standard. In fact, D2 and D3 lymph node dissections were performed in almost three-quarters of our cases (74.3%). Tumor invasion (T) and lymph node classifications (N) followed the International Union Against Cancer criteria.13
To evaluate their anastomotic condition, most patients underwent an upper gastrointestinal tract water-soluble contrast study after postoperative day 5 or 7. Leakage at the anastomosis site was defined as leakage of contrast medium. Pancreatic fistula, according to the criteria of the International Study Group of Pancreatic Surgeons, was defined as any measurable drainage from an intraoperatively placed drain on or after postoperative day 3, with an amylase content greater than 3 times the upper limit of normal serum amylase level (>300 IU/L [to convert to microkatals per liter, multiply by 0.0167]).23 The International Study Group of Pancreatic Surgeons has proposed a consensus definition and clinical grading for postoperative pancreatic fistula, which was defined as follows: grade A, called “transient fistula,” has no clinical impact; grade B required a change in management or adjustment in the clinical pathway; and grade C required a major change in clinical management or deviation from the normal clinical pathway.23,24 In this study, grades B and C were regarded as clinically significant pancreatic fistula. Intra-abdominal abscess was defined as intra-abdominal fluid collection with positive culture results identified by ultrasonography or computed tomography and associated with persistent fever and elevations of white blood cell and serum C-reactive protein levels. Diagnosis of postoperative pneumonia was obtained via computed tomography and hematological tests. Surgical mortality included in-hospital deaths within 30 days after surgery.
We used commercially available software (StatView 5.0; Abacus Concepts, Inc, Berkeley, California) for all statistical analyses. Quantitative results are expressed as mean (SD). Statistical comparison between the BMI ≥25.0 and BMI <25.0 groups was performed with χ2 statistics or the Fisher test. Univariate and multivariate logistic regression analyses were performed to identify risk factors influencing postoperative complications (anastomotic leakage, pancreatic fistula, and intra-abdominal abscess). Risk factors with a univariate P < .05 were included in the multivariate analysis. Survival curves were computed using the Kaplan-Meier method and compared by means of the log-rank test. P < .05 was considered significant.
One hundred sixteen patients were classified as overweight (BMI, ≥25.0), and 573 patients were classified as normal weight (BMI, <25.0). Patient characteristics, disease, and surgical treatments were stratified according to the BMI group. There were no differences between the 2 groups in age, sex distribution, distribution of tumor stage and lymph node status, tumor size, and histological and macroscopic types of tumor. The frequency of total gastrectomy was higher in the BMI ≥25.0 group (46.6%) than in the BMI <25.0 group (35.6%) (P = .01) (Table 1). The frequency of splenectomy was higher in the BMI ≥25.0 group (41.4%) than in the BMI < 25 group (27.2%) (P = .004), although there were no significant differences in the frequency of pancreatectomy (Table 1). There were no obvious differences in the extent of lymph node dissection between the 2 groups. The mean numbers of resected lymph nodes for D1, D2, or D3 dissections did not significantly differ between the 2 groups. Regarding the type of skin incision, the frequency of transverse incision was higher in the BMI ≥25.0 group (16.4% vs 4.4%; P < .001)(Table 1). In addition, there were no differences between the groups in the frequency of patients treated with adjuvant chemotherapy.
The BMI ≥25.0 group showed a higher rate of comorbidity with diabetes mellitus compared with the BMI <25.0 group (P = .02) (Table 2). The prevalence of additional concomitant diseases did not significantly differ between the 2 groups.
The mean duration of operation was longer in the BMI ≥25.0 group (315  minutes) than in the BMI <25.0 group (277  minutes) (P < .001) (Table 3). The mean blood loss was larger in the BMI ≥25.0 group (882  mL) than in the BMI <25.0 group (536  mL) (P < .001), and significantly more patients in the BMI ≥25.0 group required blood transfusion (P = .006) (Table 3). Furthermore, we compared the duration of operation, blood loss, and the numbers of patients requiring blood transfusion according to type of gastrectomy. Differences between the 2 groups are listed in Table 3.
Details of the major postoperative complications are listed in Table 4. Regarding surgical complications, more patients in the BMI ≥25.0 group showed anastomotic leakage (4.3% in the BMI ≥25.0 group vs 0.5% in the BMI <25.0 group; P = .005) (Table 4). The rate of pancreatic fistula was significantly higher in the BMI ≥25.0 group (11.2%) than in the BMI < 25 group (2.4%) (P = .001) (Table 4). The rate of intra-abdominal abscess was significantly higher in the BMI ≥25.0 group (5.2%) than in the BMI <25.0 group (1.7%) (P = .04) (Table 4). There was no difference in the rates of wound infection between the 2 groups. In addition, there was no difference in the rate of nonsurgical postoperative complications, such as postoperative pneumonia and deterioration of liver function, between the 2 groups. One patient in the BMI <25.0 group died in the hospital.
Univariate and multivariate analysis were performed to identify risk factors for postoperative complications (anastomotic leakage, pancreatic fistula, and intra-abdominal abscess). Table 5 shows the results of 16 variables univariately examined as potential risk factors for the 44 patients with postoperative complications vs the 645 patients without postoperative complications. Eleven of 16 factors differed significantly between these groups (P < .05) (Table 5). The multivariate logistic regression analysis identified that postoperative complications were significantly associated with being overweight (BMI ≥25.0 group) and undergoing pancreatectomy, with odds ratios of 2.69 (95% confidence interval, 1.26-5.54) and 3.29 (95% confidence interval, 1.14-9.55), respectively (Table 5). Furthermore, we studied risk factors for postoperative complications according to the extent of lymph node dissection. In the D1 and D2 subgroups, only a BMI of 25.0 or greater was independently predictive of developing postoperative complications (odds ratios of 12.50 in the D1 subgroup and 2.93 in the D2 subgroup) (Table 6). However, univariate and multivariate logistic regression analysis could not identify a BMI of 25.0 or greater as a risk factor for postoperative complications in the D3 subgroup.
In disease-specific survival and overall, there were no significant differences between the BMI ≥25.0 and BMI <25.0 groups in patients overall (both P = .11). When patients were stratified by stage, there still were no significant differences between the 2 groups (P > .05 for all comparisons of disease-specific and overall survival) (Table 7).
The duration of operation for overweight patients was longer than that for normal-weight patients, and the amount of intraoperative blood loss for overweight patients was larger than that for normal-weight patients. This is in accordance with other reports.9,11 However, Gretschel et al25 found no significant correlation between BMI and blood loss or duration of operation. Imai et al26 have shown that duration of operation was the only significant risk factor for postoperative complications after gastrectomy. According to Hawn et al,27 a higher BMI was associated with duration of operation in general surgery, but it was not associated with postoperative complications. In this study, the duration of operation and the amount of intraoperative blood loss were not directly identified as risk factors for postoperative complications, although being overweight was directly recognized as a potential risk factor. Our results are in accord with the results of clinical trial JCOG9501, that being overweight increased the risk of postoperative complications in patients undergoing a D2 but not a D3 dissection.4 In our study, the incidence of postoperative complications in normal-weight patients was 3.6% in the D2 dissection subgroup. However, this incidence has increased to 14.0% in normal-weight patients undergoing a D3 dissection (data not shown). Therefore, we believe that being overweight does not independently predict the development of postoperative complications in the D3 subgroup. For D2 dissection, other researchers also evaluated the relationship between overweight patients and the rate of postoperative complications.9,11 However, few such reports on the effects of overweight on postoperative complications in patients after D1 gastrectomy have been published. In our D1 dissection subgroup, only being overweight was identified as an adverse predictor of postoperative complications after gastrectomy. Therefore, surgical care is needed when performing gastrectomy with not only D2 but also D1dissection in overweight patients.
On the other hand, cachectic patients are often in a nutritionally poor or insufficient condition, and a poor preoperative nutritional condition is an important factor relating to morbidity and mortality.10 In our series, 35 patients were classified as cachectic patients (BMI, <17.5), and 538 patients were classified as normal-weight patients (BMI, ≥17.5 to <25.0). However, there was no difference between the 2 groups in the rates of postoperative complications (data not shown).
In gastrectomy28 and in colorectal surgery,29 diabetes mellitus has been reported to be one of the major risk factors contributing to the development of organ/space surgical site infection. The critical role played by polymorphonuclear neutrophils in the host defense mechanism against infection has encouraged the study of various aspects of neutrophil function in diabetic patients.30,31 Overweight patients are at increased risk for the development of diabetes mellitus.32 In fact, in our study, the incidence of diabetes was significantly higher in overweight compared with normal-weight patients. However, diabetes did not correlate with the development of postoperative complications. This is probably because the blood glucose level was well controlled during the perioperative period in our series. Indeed, poor control of blood glucose level impairs polymorphonuclear neutrophil functions including phagocytosis and bacterial-killing activities.28 However, further detailed studies of this issue may be necessary.
Other researchers reported a significant reduction in the total number of nodes removed after D2 dissection in overweight compared with normal-weight patients.4,9 However, we found no correlation between BMI and the number of resected lymph nodes. Anatomical dissections by Wagner et al33 demonstrated that a mean of 27 lymph nodes could be recovered during a D2 dissection. The German Gastric Cancer Study set the criteria for a D2 lymphadenectomy as more than 25 lymph nodes removed.34 We showed a mean lymph node count of 30 in patients who underwent a D2 lymphadenectomy, having performed meticulous lymphadenectomy to eradicate local disease in both overweight and normal-weight gastric cancer patients. According to our disease-specific survival analysis, there was no apparent relationship between overweight and prognosis in gastric cancer patients. Dhar et al9 reported that overweight patients more frequently underwent an apparently unsuccessful lymphadenectomy compared with patients with low BMI and had a higher ratio of diseased to removed lymph nodes. In this study, the possibility exists that extending survival in overweight patients correlates with radical lymphadenectomy because it has been recommended to excise regional lymph nodes to achieve excellent survival in gastric cancer patients.35
The presence of comorbid disease associated with being overweight, such as cardiovascular diseases, liver dysfunction, pulmonary diseases, and diabetes, may negatively affect the prognosis of postoperative patients.36 Furthermore, obesity and overweight are risk factors for several human malignant neoplasms, including endometrial, renal, esophageal, breast, and colon cancers.37 However, in this study, there were no significant differences between overweight and normal-weight patients in overall survival. Possible reasons for this discrepancy may be that the proportion of overweight patients in this study was low (16.8%) and that there were few obese patients (BMI, ≥30.0 [1.2%]). Another possible reason is that the rates of all comorbidities were not significantly different between overweight and normal-weight patients.
In conclusion, being overweight increased the risk of postoperative complications in gastric cancer patients undergoing gastrectomy. However, being overweight did not predict poor survival. Greater care is needed when performing gastrectomy with radical lymph node dissection for gastric cancer in overweight patients.
Correspondence: Hiroki Yamaue, MD, Second Department of Surgery, Wakayama Medical University, School of Medicine, 811-1, Kimiidera, Wakayama 641-8510, Japan (firstname.lastname@example.org).
Accepted for Publication: April 27, 2008.
Author Contributions:Study concept and design: Ojima and Yamaue. Acquisition of data: Ojima, Naka, Ishida, Katsuda, Iida, and Tsuji. Analysis and interpretation of data: Ojima, Iwahashi, Nakamori, Nakamura, and Ueda. Drafting of the manuscript: Ojima, Iwahashi, Nakamori, Nakamura, Naka, Ishida, Ueda, Katsuda, Iida, and Tsuji. Critical revision of the manuscript for important intellectual content: Yamaue. Statistical analysis: Yamaue. Administrative, technical, and material support: Iwahashi, Nakamori, Nakamura, and Yamaue. Study supervision: Yamaue.
Financial Disclosure: None reported.