Ann Surg Oncol (2014) 21:1115–1122 DOI 10.1245/s10434-013-3409-4
ORIGINAL ARTICLE – GASTROINTESTINAL ONCOLOGY
A Body Shape Index Has a Good Correlation with Postoperative Complications in Gastric Cancer Surgery Bang Wool Eom, MD1, Jungnam Joo, PhD2, Hong Man Yoon, MD1, Keun Won Ryu, MD, PhD1, Young-Woo Kim, MD, PhD1, and Jun Ho Lee, MD, PhD1,3 Center for Gastric Cancer, National Cancer Center, Goyang-si, Republic of Korea; 2Biometric Research Branch, Division of Cancer Epidemiology and Prevention, Research Institute & Hospital, National Cancer Center, Goyang-si, Republic of Korea; 3 Present Address: Department of Surgery, Sungkyunkwan University School of Medicine, Seoul, Gangnam-gu, Republic of Korea 1
ABSTRACT Background. The relationship between obesity and surgical complications has been controversial. A Body Shape Index (ABSI) is a newly developed anthropometric index based on waist circumference adjusted for height and weight. The aim of this study was to investigate the relationship between ABSI and surgical complications. Methods. From November 2001 to September 2012, 4,813 patients underwent curative resection for gastric cancer. ABSI was defined as waist circumference divided by (BMI2/3height1/2). Data of clinicopathologic characteristics and morbidity were collected by retrospective review. Binary logistic regression was used for multivariable analyses to determine whether ABSI was independently associated with postoperative complications. Results. The incidence of overall surgical complications was 13.4 %, and the most common complication was ileus (2.8 %). In the multivariable analysis, ABSI was an independent factor for overall complications [odds ratio (OR), 1.22; 95 % confidence interval (CI) 1.01–1.48; P = 0.041). However, BMI showed no statistical significance (OR, 1.03; 95 % CI 1.00–1.06; P = 0.063). In the subgroup analyses, ABSI was significantly associated with overall complications regarding open gastrectomy (OR, 1.26; 95 % CI 1.01–1.57; P = 0.039). Regarding laparoscopy-
The study was approved by the institutional review board of the Korean National Cancer Center. (No. NCCNCS-12-571) Ó Society of Surgical Oncology 2013 First Received: 26 June 2013; Published Online: 4 December 2013 J. H. Lee, MD, PhD e-mail: [email protected]
assisted gastrectomy, ABSI had no significant effect on overall complications (P = 0.844). Conclusions. ABSI shows good correlation with surgical complications in patients with gastric cancer. Further studies are needed for the various clinical roles of ABSI, and the results could be helpful to determine the effect of abdominal obesity on gastric cancer surgery and the clinical usefulness of ABSI. The prevalence of overweight and obesity is rapidly increasing, and obesity is the fifth leading risk factor for global death according to the World Health Organization. Obesity is well known to be associated with several chronic diseases, including diabetes, hypertension, coronary heart disease, arthritis, and certain forms of cancer.1,2 Previous large-scale studies showed that overweight and obesity are an independent risk factors for all-cause mortality.3–5 In addition to the high comorbidity rate in obese patients, many surgeons, particularly abdominal surgeons, have felt burdened regarding obese patients because of the accumulated visceral fat tissue that can make operations technically difficult. Many studies have evaluated the relationship between obesity and surgical outcomes; however, the effect of obesity on surgical complications remains controversial. One cause may be that obesity, which was measured by body mass index (BMI), does not distinguish between muscle and fat accumulation or between central and general obesity.6–8 Ethnic differences regarding the distribution of central obesity may also have affected these contradictory results.9 Recently, a new anthropometric measure incorporating body shape, A Body Shape Index (ABSI), was developed by Krakauer and Krakauer.10 The researchers found that ABSI was a substantial risk factor for premature mortality in the general
1116
population and is expressed as the excess risk from a high waist circumference (WC). However, there are no data regarding whether ABSI is associated with surgical complications. Here, we aimed to evaluate whether ABSI had an effect on surgical complications in patients who underwent curative gastrectomy for gastric cancer. Gastrectomy with lymph node dissection is considerably influenced by abdominal shape and abdominal fat accumulation.11–13 METHODS
B. W. Eom et al.
disease), preoperative laboratory findings (hemoglobin, albumin, and creatinine), operation type, and whether a patient received adjuvant chemotherapy were identified. We classified operation type according to the surgical approach (open vs. laparoscopy) and extent of gastrectomy (subtotal vs. total gastrectomy). The cancers were staged by using the Union Internationale contre le Cancer tumor-node-metastasis system, 7th edition.15 Complication data were collected from operation day to last follow-up, which included both immediate postoperative complications and late complications such as adhesive ileus and incisional hernia.
Patients Statistical Analysis The medical records of consecutive patients undergoing gastric surgery from November 2001 to September 2012 were reviewed by using the database of the Center for Gastric Cancer at the National Cancer Center, Korea. Patients who underwent curative gastrectomy for gastric cancer were included in the current study regardless of operation type. We excluded those who received neoadjuvant chemotherapy, those who underwent palliative surgery, those who had distant metastasis, and those who had other cancers preoperatively or postoperatively. Additionally, patients who had one or more missing records among weight, height, and WC were excluded. In total, 4,813 patients were available for evaluating in this study. Patients underwent subtotal or total gastrectomy with more than D1 ? b (Nos. 7, 8, and 9) lymph node dissection for early gastric cancer and D2 dissection for advanced gastric cancer, as defined by the Japanese Gastric Cancer Association.14 After surgery, all patients were followed up regularly for 5 years, and most patients with stage II or more advanced gastric cancer received adjuvant chemotherapy. This study was approved by the Institutional Review Board at the National Cancer Center (No. NCCNCS-12-571). BMI and ABSI Patients’ height and body weight were examined preoperatively, and BMI was calculated as weight in kilograms divided by the square of height in meters (kg/ m2). WC in meters was measured with the tape measure positioned just above the uppermost lateral border of the ilium in the supine position. ABSI was calculated according to WC adjusted for height and weight as follows.10: ABSI ¼ WC ðmÞ weight ðkgÞ2=3 heightðmÞ5=6 WC ¼ ðm11=6 kg2=3 Þ: 2=3 BMI height1=2
All continuous values are presented as the mean ± SD. Statistical analysis was conducted by using the Chi square test for categorical variables and Student’s t test for continuous variables. A binary logistic regression model was used for multivariable analyses to determine whether ABSI had an independent effect on surgical complications. The prognostic values of covariates were expressed by calculating the odds ratios (ORs) and 95 % confidence intervals (CIs). A Pvalue less than 0.05 was considered statistically significant, and all statistical analyses were performed by using SAS software (version 9.1.3; SAS Institute, Cary, NC). RESULTS Demographics and Complications Table 1 summarizes patients’ clinicopathologic characteristics. The mean age of all patients was 57.8 ± 11.9 years, and 67.0 % were male. Approximately 64 % of patients had BMI in the normal range. The laparoscopic approach included 1,446 laparoscopy-assisted gastrectomies, 169 robot-assisted gastrectomies, and 12 hand-assisted laparoscopic surgeries. The most common tumor location was the lower third of the stomach, and more than half the patients had early gastric cancer. Among a total of 4,813 patients, 643 (13.4 %) patients had postoperative complications. We classified them as intraabdominal, wound, systemic, and late-functional complications (Table 2). The most common complication was adhesive ileus (2.8 %), followed by abscess/fluid collection (2.4 %), stricture (2.1 %), and acute wound problem (1.4 %). Mortality, defined as death within 30 days after operation, occurred in 12 patients (0.25 %).
Data Collection
Effect of ABSI on Complications
Clinicopathologic characteristics, comorbidities (hypertension, diabetes mellitus, heart disease, and pulmonary
Univariable and multivariable analyses were performed to evaluate the relationship between surgical complications
A Body Shape Index and Surgical Complications TABLE 1 Clinicopathologic characteristics (N = 4,813)
1117
Factor
Subgroup
Result
Age, y (mean ± SD) Sex, n (%)
57.8 ± 11.9 Male
3,226 (67.0)
Female
1,587 (33.0)
BMI, kg/m2 (mean ± SD)
23.7 ± 3.2 \18.5 kg/m2 (underweight)
BMI, n (%)
2
18.5–24.9 kg/m (normal)
3,072 (63.8)
25–29.9 kg/m2 (overweight)
1,413 (29.3)
C30 kg/m2 (obese) ABSI, m
11/6
-2/3
kg
(mean ± SD) Present
1,275 (26.5) 563 (11.7)
Diabetes mellitus, n (%)
Present
Heart disease,a n (%)
Present
140 (2.9)
Pulmonary disease,b n (%)
Present
88 (1.8)
Hemoglobin, n (%)
\10 g/dL
234 (4.9)
Albumin, n (%)
C10 g/dL \3.5 g/dL
4,579 (95.1) 152 (3.2)
C3.5 g/dL
4,659 (96.8)
B1.4 mg/dL
4,770 (99.1)
[1.4 mg/dL
43 (0.9)
Open
3,186 (66.2)
Tumor size, cm (mean ± SD) Surgical approach, n (%) Extent of gastrectomy, n (%) Location, n (%)
Lauren classification, n (%)
pT, n (%) BMI body mass index, ABSI a body shape index Heart disease included ischemic heart disease, arrhythmia, and valvular disease
pN, n (%)
b
Pulmonary disease included asthma, chronic obstructive pulmonary disease, bronchiectasis, and pneumoconiosis
129 (2.7) 0.079 ± 0.005
Hypertension, n (%)
Creatinine, n (%)
a
199 (4.1)
Adjuvant chemotherapy, n (%)
and BMI or ABSI (Table 3). In the univariable analysis, neither BMI nor ABSI had a significant effect on overall surgical complications. However, after adjusting 17 baseline variables (age; sex; hypertension; diabetes mellitus; heart disease; pulmonary disease; preoperative hemoglobin, albumin, and creatinine
4.5 ± 2.8 Laparoscopic
1,627 (33.8)
Subtotal
3,478 (72.3)
Total
1,335 (27.7)
Upper
655 (13.6)
Middle
1,412 (29.3)
Lower
2,068 (43.0)
Combined
678 (14.1)
Intestinal
2,216 (45.9)
Diffuse
2,004 (41.5)
Mixed Unidentified
432 (8.9) 161 (3.3)
1
2,603 (54.1)
2
663 (13.8)
3
906 (18.8)
4
641 (13.3)
0
3,096 (64.3)
1
639 (13.3)
2
485 (10.1)
3
593 (12.3)
No
3,623 (75.3)
Yes
1,190 (24.7)
levels; surgical approach; extent of gastrectomy; tumor size; tumor location; Lauren classification; pT; pN; and adjuvant chemotherapy) in the multivariable analysis, we found that ABSI showed an independent correlation with overall surgical complications (P = 0.041). For each 0.01 increase in ABSI, the odds of complication occurrence
1118
B. W. Eom et al.
TABLE 2 Postoperative complications (N = 4,813) Complication
n
%
No
4,170
86.6
40
0.8
53
1.1
Intraabdominal complications Bleeding Leakage Abscess/fluid collection
117
2.4
Stricture
102
2.1
Adhesive ileus
137
2.8
Gastric stasis
26
0.5
Internal hernia
7
0.1
Bowel perforation
7
0.1
Othersa Wound complications
7
0.1
Acute wound problem
68
1.4
Incisional hernia
24
0.5
Pneumonia
28
0.6
Embolism/thrombosis
11
0.2
Pleural effusion
3
0.1
Pneumothorax
3
0.1
Heart disease
4
0.1
Cerebral infarction
1
0.02
5
0.1
Risk of Complications According to ABSI or BMI
Systemic complications
Late functional complication Cholecystitis
overall complications (OR, 1.02; 95 % CI 0.99–1.06; P = 0.179). For intraabdominal complications, neither ABSI nor BMI was a significant risk factor (OR, 1.23; 95 % CI 0.96–1.58; P = 0.096 and OR, 1.01; 95 % CI 0.97–1.04; P = 0.729, respectively). Regarding laparoscopy-assisted gastrectomy, neither ABSI nor BMI showed a significant effect for overall complications (OR, 1.04; 95 % CI 0.70–1.55; P = 0.844 and OR, 1.05, 95 % CI 0.99–1.12; P = 0.088, respectively). Also, for intraabdominal complications, neither ABSI nor BMI had any significant effect (OR, 1.14; 95 % CI 0.75–1.76; P = 0.535 and OR 1.03; 95 % CI 0.96–1.10; P = 0.388, respectively).
a
Others included lymphorrhea (n = 3), splenic infarction (n = 2), and pseudomembranous colitis (n = 2)
increased by 22 %. However, BMI showed no statistical significance in overall complications (P = 0.063). For the rest, male sex, preoperative low hemoglobin level, open approach, total gastrectomy, and advanced N classification were significant risk factors for surgical complications. Next, we focused on intraabdominal complications, because ABSI or BMI was suggested to be more related to intraabdominal complications than to overall complications. In the multivariable analysis, neither ABSI nor BMI had statistical significance; however, the P-value of ABSI differed widely from that of BMI (P = 0.061 for ABSI; P = 0.501 for BMI; Table 4). Independent risk factors for intraabdominal complications were male sex, total gastrectomy, advanced N classification, and no adjuvant chemotherapy. Subgroup Analyses According to Surgical Approach Additionally, we performed multivariable subgroup analyses according to surgical approach. Regarding open gastrectomy, ABSI was significantly associated with overall complications (OR, 1.26; 95 % CI 1.01–1.57; P = 0.039). However, BMI had no significant effect on
We estimated surgical complication rates according to quartiles of ABSI or BMI (Fig. 1a). The overall complication rate increased as ABSI increased. However, the complication rate according to BMI was irregular. For intraabdominal complications, the complication rate increased gradually according to ABSI; however, no difference between the second and third quartiles of BMI was found, and the complication rate increased sharply in the fourth quartile of BMI (Fig. 1b). DISCUSSION The present study is the first to evaluate whether ABSI is associated with surgical complications worldwide. By multivariable analysis, we found that ABSI was an independent risk factor for overall complications in patients with gastric cancer. Compared with BMI, ABSI is better correlated with surgical complications. Surgical complications result in various clinical problems such as increased hospital duration, increased cost, longterm sequelae, and severe complications that can become life-threatening. Previous studies have evaluated the effect of BMI on surgical complications in various abdominal surgeries, and contradictory results have been shown.16–22 Several studies have shown that obesity increases postoperative complications such as anastomosis leak, wound infection, and pulmonary embolism.16–19 However, other studies have reported no difference in morbidity between obese and nonobese patients.20–22 Similar results were also shown in gastric cancer specifically. Higher rates of complications, larger amounts of blood loss, and longer operating times were found in high-BMI patients in some studies.23–25 However, other studies reported that high BMI itself might not increase operative morbidity.26–28 Conversely, some study groups have estimated abdominal shape–related measures such as WC, waist-to-hip
A Body Shape Index and Surgical Complications
1119
TABLE 3 Univariable and multivariable analysis of factors for overall complications Factor
Subgroup
Univariable Hazard ratio
Age Sex
Female vs. male
ABSI BMI
Multivariable (including BMI) Multivariable (including ABSI) P-value Hazard ratio
P–value
Hazard ratio
P
1.01 (1.00, 1.02)
0.012 1.00 (1.00, 1.02)
0.098
1.00 (1.00, 1.01)
0.77 (0.64, 0.93)
0.005 0.79 (0.65, 0.96)
0.019
0.73 (0.59, 0.91)
0.326 0.004
1.13 (0.96, 1.34)a
0.149
1.22 (1.01, 1.48)a
0.041
1.01 (0.98, 1.03)
0.571 1.03 (1.00, 1.06)
0.063
Hypertension
Yes vs. no
1.03 (0.85, 1.24)
0.798 0.97 (0.78, 1.20)
0.747
1.01 (0.82, 1.24)
0.965
Diabetes mellitus
Yes vs. no
0.96 (0.74, 1.25)
0.77
0.93 (0.70, 1.22)
0.585
0.92 (0.70, 1.22)
0.569
Heart diseaseb Pulmonary diseasec
Yes vs. no Yes vs. no
1.57 (1.03, 2.42) 2.07 (1.26, 3.40)
0.038 1.49 (0.94, 2.36) 0.004 1.62 (0.94, 2.79)
0.091 0.085
1.50 (0.95, 2.38) 1.57 (0.91, 2.71)
0.083 0.105
Hemoglobin
\10.0 vs. C 10.0 g/dL 2.03 (1.48, 2.79) \0.001 1.52 (1.06, 2.16)
0.022
1.50 (1.05, 2.13)
0.026
Albumin
\3.5 vs. C 3.5 g/dL
1.92 (1.30, 2.84)
0.001 1.27 (0.81, 2.01)
0.295
1.21 (0.77, 1.90)
0.419
Creatinine
C1.4 vs. \ 1.4 mg/dL 1.49 (0.69, 3.22)
0.313 1.06 (0.45, 2.48)
0.896
1.08 (0.46, 2.52)
0.861
Laparoscopic approach
Versus open
0.50 (0.41, 0.61) \0.001 0.70 (0.56, 0.89)
0.003
0.70 (0.56, 0.88)
0.003
Total gastrectomy
Versus subtotal
2.77 (2.34, 3.29) \0.001 2.21 (1.68, 2.90)
\0.001
2.21 (1.69, 2.91)
\0.001
1.09 (1.06, 1.12) \0.001 0.98 (0.95, 1.02)
0.375
0.98 (0.95, 1.02)
0.376
Tumor size Location
Lauren Classification pT
pN
Upper
1
1
1
Middle
0.41 (0.32, 0.53) \0.001 0.83 (0.60, 1.14)
0.243
0.83 (0.60, 1.14)
0.244
Lower
0.41 (0.32, 0.51) \0.001 0.87 (0.61, 1.22)
0.414
0.87 (0.61, 1.23)
0.422
Combined
0.85 (0.65, 1.11)
0.904
0.97 (0.71, 1.32)
0.825
Intestinal
1
0.231 0.98 (0.72, 1.34) 1
1
Diffuse
1.10 (0.92, 1.31)
0.313 1.13 (0.92, 1.38)
0.243
1.14 (0.93, 1.39)
0.223
Mixed
1.10 (0.81, 1.48)
0.544 1.16 (0.85, 1.59)
0.348
1.16 (0.85, 1.58)
0.355
1 2
1 1.40 (1.08, 1.79)
0.01
1 1.08 (0.81, 1.43)
0.61
1 1.06 (0.80, 1.41)
0.668
3
1.45 (1.15, 1.80)
0.001 0.99 (0.74, 1.33)
0.96
0.98 (0.73, 1.31)
0.893
0.05
1.40 (0.98, 1.99)
0.065
4
2.52 (2.01, 3.15) \0.001 1.43 (1.00, 2.04)
0
1
1
1.22 (0.94, 1.57)
0.129 1.15 (0.86, 1.54)
0.354
1.15 (0.86, 1.55)
2
1.52 (1.17, 1.99)
0.002 1.29 (0.92, 1.80)
0.135
1.31 (0.94, 1.83)
0.116
3
2.13 (1.70, 2.67) \0.001 1.60 (1.14, 2.24)
0.007
1.59 (1.13, 2.23)
0.007
1.38 (1.15, 1.66)
0.046
0.77 (0.59, 1.00)
0.051
Adjuvant chemotherapy Yes vs. no
1
0.001 0.76 (0.59, 1.00)
1 0.343
ABSI a body shape index, BMI body mass index a
Odds ratio for each 0.01 increase in ABSI
b
Heart disease included ischemic heart disease, arrhythmia, and valvular disease
c
Pulmonary disease included asthma, chronic obstructive pulmonary disease, bronchiectasis, and pneumoconiosis
ratio, abdominal anterior to posterior diameter, and visceral fat area.11–13,29–35 WC and waist-to-hip ratio have been identified to be associated with all-cause mortality in some large-scale studies.29–31 A few studies showed that visceral fat area was a more useful variable than BMI in terms of surgical complications such as pancreatic fistula and intraoperative bleeding.11,12,32–34 Moreover, Yamamoto et al.13 showed that celiac artery depth ratio was independently related with postoperative pancreatic fistula. However, few data are available concerning the relationship between WC or waist-to-hip ratio and postoperative complications. Additionally, visceral fat area and celiac
artery depth ratio require specialized software and a considerable amount of time and energy. The new anthropometric measure ABSI is calculated on the basis of WC adjusted for height and weight.10 Krakauer and Krakauer found that mean ABSI increased steadily from mid life into old age and was consistently higher in men than women after young adulthood. However, ageand sex-specific BMI and WC had different trends than ABSI, decreasing after approximately 60 years of age.10 Moreover, the mean WC was higher in men, whereas the mean BMI was higher in women. These different characteristics of ABSI compared with WC and BMI can be
1120
B. W. Eom et al.
TABLE 4 Univariable and multivariable analysis of factors for intraabdominal complications Factor
Subgroup
Univariable Hazard ratio
Age Sex
Female vs. male
ABSI BMI
Multivariable (including BMI) Multivariable (including ABSI) P-value Hazard ratio
P–value
Hazard ratio
P-value
1.00 (0.99, 1.01)
0.786 1.00 (0.99, 1.01)
0.6
1.00 (0.99, 1.01)
0.313
0.76 (0.62, 0.93)
0.008 0.75 (0.60, 0.94)
0.013
0.70 (0.55, 0.89)
0.003
1.07 (0.89, 1.29)a
0.476
1.23 (0.99, 1.52)
0.061
1.00 (0.97, 1.03)
0.967 1.01 (0.98, 1.04)
0.501
Hypertension
Yes vs. no
0.98 (0.80, 1.22)
0.881 0.98 (0.77, 1.25)
0.854
0.99 (0.78, 1.25)
0.927
Diabetes mellitus
Yes vs. no
1.09 (0.82, 1.44)
0.557 1.11 (0.82, 1.50)
0.512
1.10 (0.81, 1.49)
0.551
Heart diseaseb Pulmonary diseasec
Yes vs. no Yes vs. no
1.30 (0.78, 2.14) 2.00 (1.13, 3.41)
0.315 1.32 (0.77, 2.27) 0.016 1.54 (0.83, 2.83)
0.306 0.169
1.33 (0.78, 2.27) 1.51 (0.82, 2.77)
0.303 0.189
Hemoglobin
\10.0 vs. C10.0 g/dL 1.74 (1.22, 2.50)
0.002 1.35 (0.90, 2.01)
0.15
1.34 (0.89, 2.00)
0.157
Albumin
\3.5 vs. C3.5 g/dL
1.50 (0.94, 2.38)
0.086 0.94 (0.54, 1.63)
0.812
0.90 (0.52, 1.56)
0.706
Creatinine
C1.4 vs. \1.4 mg/dL 1.70 (0.75, 3.85)
0.201 1.31 (0.53, 3.24)
0.564
1.32 (0.53, 3.28)
0.545
Laparoscopic approach
Versus open
0.55 (0.44, 0.68) \0.001 0.83 (0.64, 1.07)
0.154
0.83 (0.64, 1.07)
0.156
Total gastrectomy
Versus subtotal
3.16 (2.62, 3.82) \0.001 2.63 (1.94, 3.56)
\0.001
2.64 (1.95, 3.57)
\0.001
1.10 (1.07, 1.13) \0.001 1.00 (0.96, 1.05)
0.891
1.00 (0.96, 1.05)
0.883
Tumor size Location
Lauren Classification pT
pN
Upper
1
1
Middle
0.40 (0.30, 0.52) \0.001 0.89 (0.63, 1.26)
0.506
0.89 (0.63, 1.26)
0.504
Lower
0.36 (0.28, 0.46) \0.001 0.90 (0.62, 1.32)
0.598
0.90 (0.62, 1.32)
0.602
Combined
0.83 (0.62, 1.10)
0.628
0.91 (0.65, 1.28)
0.588
Intestinal
1
0.193 0.92 (0.65, 1.29)
1
1
1
Diffuse
1.22 (1.00, 1.48)
0.055 1.13 (0.90, 1.42)
0.304
1.13 (0.90, 1.43)
0.285
Mixed
1.37 (1.00, 1.89)
0.052 1.38 (1.00, 1.92)
0.059
1.38 (0.99, 1.93)
0.057
1 2
1 1.45 (1.10, 1.92)
1 0.009 1.13 (0.83, 1.55)
0.445
1 1.13 (0.82, 1.54)
0.46
3
1.51 (1.18, 1.93)
0.001 1.03 (0.74, 1.43)
0.852
1.03 (0.74, 1.42)
0.88
0.147
1.34 (0.90, 1.99)
0.15
4
2.47 (1.93, 3.16) \0.001 1.34 (0.90, 2.00)
0
1
1
1.36 (1.04, 1.79)
0.028 1.27 (0.92, 1.75)
0.149
1.27 (0.92, 1.76)
2
1.52 (1.13, 2.04)
0.006 1.32 (0.91, 1.92)
0.143
1.34 (0.92, 1.95)
0.127
3
2.22 (1.73, 2.84) \0.001 1.70 (1.17, 2.45)
0.006
1.70 (1.17, 2.47)
0.006
1.46 (1.20, 1.79) \0.001 0.73 (0.55, 0.98)
0.037
0.73 (0.55, 0.98)
0.037
Adjuvant chemotherapy Yes vs. no
1
1 0.145
ABSI a body shape index, BMI body mass index a
Odds ratio for each 0.01 increase in ABSI
b
Heart disease included ischemic heart disease, arrhythmia, and valvular disease
c
Pulmonary disease included asthma, chronic obstructive pulmonary disease, bronchiectasis, and pneumoconiosis
helpful for understanding ABSI. Briefly, a high ABSI might correspond with a greater fraction of visceral fat compared with peripheral tissue at a given height and weight.36 ABSI was recently developed, and few data are available for it. Krakauer and Krakauer reported that ABSI can be a substantial risk factor for premature mortality in the general US population.10 Since then, identified that ABSI was a better predictor of blood pressure than BMI or WC in Portuguese adolescents Duncan et al.37 However, did not show the superiority of ABSI in predicting diabetes
mellitus in the Chinese population compared with BMI or WC He et al.36 In this study, ABSI showed a good relationship with surgical complications in gastric cancer patients. One possible reason is that radical gastrectomy with D2 lymph node dissection is strongly affected by abdominal fat accumulation. There are many major vessels in the gastrectomy field, and excessive adipose tissue conceals these vascular structures, resulting in bleeding or adjacent organ injury. Excess fat tissue also makes lymph node dissection more difficult and induces incomplete lymph node
A Body Shape Index and Surgical Complications FIG. 1 (a) Overall complication rates according to quartiles of ABSI (blue) and BMI (yellow). (b) Intraabdominal complication rates according to quartiles of ABSI (blue) and BMI (yellow)
1121
(a)
(b)
20
8
15
6
4
10
ABSI BMI
ABSI BMI 2
5
0−25
25−50
dissection and a long operating time. Moreover, the surgical approach for anastomosis is risky in patients with a long anterior to posterior diameter that can lead to anastomosis leakage or stricture. These factors would affect the significance of ABSI. Additionally, according to subgroup analyses, ABSI had a significant effect on surgical complications in open surgery compared with laparoscopic surgery. Particularly, ABSI had a stronger statistical relationship with overall complications compared with BMI in open surgery. These results coincided with many previous studies. Many previous studies have reported increased morbidity in open gastric surgery and no significant difference in laparoscopic surgery.23,24,27,34,35 According to the study by Makino et al., laparoscopy-assisted distal gastrectomy is not influenced by obesity, whereas open distal gastrectomy may be influenced by obesity, a result that is identical to that in the current study.34 Therefore, it seems that abdominal obesity should be considered for surgical complications more in open gastrectomy than in laparoscopic surgery. Potential limitations of the current study include misclassification bias resulting from retrospective characteristics. Selection bias also cannot be excluded because patients with missing data were excluded. Furthermore, although we statistically controlled relevant confounders, some confounders can be missed in the multivariable analysis. Finally, surgical complications were not classified systemically, such as with the severity grading system, because of the lack of clinical data.38 Nevertheless, a large-scale cohort and a multivariable statistical method might compensate for these limitations. In conclusion, as a new anthropometric measure, ABSI shows good correlation with surgical complications in patients with gastric cancer. Surgeons can consider not only the patient’s sex, preoperative laboratory findings,
50−75
75−100
0−25
25−50
50−75
75−100
surgical method, and clinical stage, but also abdominal obesity, as risk factors for surgical complications by using ABSI. We are monitoring patients’ survival to identify the effect of ABSI on survival. Further studies are also needed for the various clinical roles of ABSI, and the results could be helpful to determine the effect of abdominal obesity on gastric cancer surgery and the clinical usefulness of ABSI. ACKNOWLEDGMENT This work was supported by a grant of the National Cancer Center (NCC-1110531-1,2).
REFERENCES 1. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA. 2003;289:76–79. 2. Kopelman PG. Obesity as a medical problem. Nature. 2000;404:635–643. 3. Pischon T, Boeing H, Hoffmann K, Bergmann M, Schulze MB, Overvad K et al. General and abdominal adiposity and risk of death in Europe. N Engl J Med. 2008;359:2105–2120. 4. Berrington de Gonzalez A, Hartge P, Cerhan JR, Flint AJ, Hannan L, MacInnis RJ et al. Body-mass index and mortality among 1.46 million white adults. N Engl J Med. 2010;363:2211–2219. 5. Boggs DA, Rosenberg L, Cozier YC, Wise LA, Coogan PF, RuizNarvaez EA et al. General and abdominal obesity and risk of death among black women. N Engl J Med. 2011;365:901–908. 6. Nevill AM, Stewart AD, Olds T, Holder R. Relationship between adiposity and body size reveals limitations of BMI. Am J Phys Anthropol. 2006;129:151–156. 7. Heymsfield SB, Scherzer R, Pietrobelli A, Lewis CE, Grunfeld C. Body mass index as a phenotypic expression of adiposity: quantitative contribution of muscularity in a population-based sample. Int J Obes. 2009;33:1363–1373. 8. Go´mez-Ambrosi J, Silva C, Galofre´ JC, Escalada J, Santos S, Milla´n D et al. Body mass index classification misses subjects with increased cardiometabolic risk factors related to elevated adiposity. Int J Obes. 2012;36:286–294. 9. Agyemang C, van Valkengoed IG, van den Born BJ, Bhopal R, Stronks K. Heterogeneity in sex differences in the metabolic syndrome in Dutch white, Surinamese African and South Asian populations. Diabet Med. 2012;29:1159–1164.
1122 10. Krakauer NY, Krakauer JC. A new body shape index predicts mortality hazard independently of body mass index. PLoS One. 2012;7:e39504. 11. Lee JH, Paik YH, Lee JS, et al. Abdominal shape of gastric cancer patients influences short-term surgical outcomes. Ann Surg Oncol. 2007;14:1288–1294. 12. Tanaka K, Miyashiro I, Yano M, et al. Accumulation of excess visceral fat is a risk factor for pancreatic fistula formation after total gastrectomy. Ann Surg Oncol. 2009;16:1520–1555. 13. Yamamoto N, Oshima T, Sato T, Makino H, Nagano Y, Fujii S et al. Upper abdominal body shape is the risk factor for postoperative pancreatic fistula after splenectomy for advanced gastric cancer: a retrospective study. World J Surg Oncol. 2008;6:109. 14. Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 2nd English ed. Gastric Cancer. 1998;1: 10–24. 15. Edge SB, Byrd DR, Compton CC. AJCC cancer staging manual. 7th ed. New York: Springer-Verlag; 2010. 16. Pikarsky AJ, Saida Y, Yamaguchi T, et al. Is obesity a high-risk factor for laparoscopic colorectal surgery? Surg Endosc. 2002;16:855–858. 17. Merkow RP, Bilimoria KY, McCarter MD, Bentrem DJ. Effect of body mass index on short-term outcomes after colectomy for cancer. J Am Coll Surg. 2009;208:53–61. 18. Mathur AK, Ghaferi AA, Osborne NH, et al. Body mass index and adverse perioperative outcomes following hepatic resection. J Gastrointest Surg. 2010;14:1285–1291. 19. House MG, Fong Y, Arnaoutakis DJ, Sharma R, Winston CB, Protic et al. Preoperative predictors for complications after pancreaticoduodenectomy: impact of BMI and body fat distribution. J Gastrointest Surg. 2008;12:270–278. 20. Dindo D, Muller MK, Weber M, Clavien PA. Obesity in general elective surgery. Lancet. 2003;361:2032–2035. 21. Al-Refaie WB, Parsons HM, Henderson WG, et al. Body mass index and major cancer surgery outcomes: lack of association or need for alternative measurements of obesity? Ann Surg Oncol. 2010;17:2264–2273. 22. Balentine CJ, Enriquez J, Cruz G, et al. Obesity does not increase complications following pancreatic surgery. J Surg Res. 2011;170:220–225. 23. Inagawa S, Adachi S, Oda T, Kawamoto T, Koike N, Fukao K. Effect of fat volume on postoperative complications and survival rate after D2 dissection for gastric cancer. Gastric Cancer. 2000;3:141–144. 24. Kodera Y, Sasako M, Yamamoto S, Sano T, Nashimoto A, Kurita A. Identification of risk factors for the development of complications following extended and superextended lymphadenectomies for gastric cancer. Br J Surg. 2005;92:1103–1109. 25. Kunisaki C, Makino H, Takagawa R, et al. Predictive factors for surgical complications of laparoscopy-assisted distal gastrectomy for gastric cancer. Surg Endosc. 2009;23:2085–2093.
B. W. Eom et al. 26. Gretschel S, Christoph F, Bembenek A, Estevez-Schwarz L, Schneider U, Schlag PM. Body mass index does not affect systematic D2 lymph node dissection and postoperative morbidity in gastric cancer patients. Ann Surg Oncol. 2003;10:363–368. 27. Lee HJ, Kim HH, Kim MC, et al. The impact of a high body mass index on laparoscopy assisted gastrectomy for gastric cancer. Surg Endosc. 2009;23:2473–2479. 28. Oki E, Sakaguchi Y, Ohgaki K, et al. The impact of obesity on the use of a totally laparoscopic distal gastrectomy in patients with gastric cancer. J Gastric Cancer. 2012;12:108–112. 29. Simpson JA, MacInnis RJ, Peeters A, Hopper JL, Giles GG, English DR. A comparison of adiposity measures as predictors of all-cause mortality: the Melbourne Collaborative Cohort Study. Obesity. 2007;15:994–1003. 30. Seidell JC. Waist circumference and waist/hip ratio in relation to all-cause mortality, cancer and sleep apnea. Eur J Clin Nutr. 2010;64:35–41. 31. Petursson H, Sigurdsson JA, Bengtsson C, Nilsen TI, Getz L. Body configuration as a predictor of mortality: comparison of five anthropometric measures in a 12 year follow-up of the Norwegian HUNT 2 study. PLoS One. 2011;6:e26621. 32. House MG, Fong Y, Arnaoutakis DJ, Sharma R, Winston CB, Protic M et al. Preoperative predictors for complications after pancreaticoduodenectomy: impact of BMI and body fat distribution. J Gastrointest Surg. 2008;12:270–278. 33. Tsujinaka S, Konishi F, Kawamura YJ, et al. Visceral obesity predicts surgical outcomes after laparoscopic colectomy for sigmoid colon cancer. Dis Colon Rectum. 2008;51:1757–1765. 34. Makino H, Kunisaki C, Izumisawa Y, et al. Effect of obesity on laparoscopy-assisted distal gastrectomy compared with open distal gastrectomy for gastric cancer. J Surg Oncol. 2010;102:141–147. 35. Ojima T, Iwahashi M, Nakamori M, M Nakamura, K Takifuji, M Katsuda et al. The impact of abdominal shape index of patients on laparoscopy-assisted distal gastrectomy for early gastric cancer. Langenbecks Arch Surg. 2012;397:437–445. 36. He S, Chen X. Could the new body shape index predict the new onset of diabetes mellitus in the Chinese population? PLoS One. 2013;8:e50573. 37. Duncan MJ, Mota J, Vale S, Santos MP, Ribeiro JC. Associations between body mass index, waist circumference and body shape index with resting blood pressure in Portuguese adolescents. Ann Hum Biol. 2013;40:163–167. 38. Strasberg SM, Linehan DC, Hawkins WG. The accordion severity grading system of surgical complications. Ann Surg. 2009; 250: 177–186.
ORIGINAL ARTICLE – GASTROINTESTINAL ONCOLOGY
A Body Shape Index Has a Good Correlation with Postoperative Complications in Gastric Cancer Surgery Bang Wool Eom, MD1, Jungnam Joo, PhD2, Hong Man Yoon, MD1, Keun Won Ryu, MD, PhD1, Young-Woo Kim, MD, PhD1, and Jun Ho Lee, MD, PhD1,3 Center for Gastric Cancer, National Cancer Center, Goyang-si, Republic of Korea; 2Biometric Research Branch, Division of Cancer Epidemiology and Prevention, Research Institute & Hospital, National Cancer Center, Goyang-si, Republic of Korea; 3 Present Address: Department of Surgery, Sungkyunkwan University School of Medicine, Seoul, Gangnam-gu, Republic of Korea 1
ABSTRACT Background. The relationship between obesity and surgical complications has been controversial. A Body Shape Index (ABSI) is a newly developed anthropometric index based on waist circumference adjusted for height and weight. The aim of this study was to investigate the relationship between ABSI and surgical complications. Methods. From November 2001 to September 2012, 4,813 patients underwent curative resection for gastric cancer. ABSI was defined as waist circumference divided by (BMI2/3height1/2). Data of clinicopathologic characteristics and morbidity were collected by retrospective review. Binary logistic regression was used for multivariable analyses to determine whether ABSI was independently associated with postoperative complications. Results. The incidence of overall surgical complications was 13.4 %, and the most common complication was ileus (2.8 %). In the multivariable analysis, ABSI was an independent factor for overall complications [odds ratio (OR), 1.22; 95 % confidence interval (CI) 1.01–1.48; P = 0.041). However, BMI showed no statistical significance (OR, 1.03; 95 % CI 1.00–1.06; P = 0.063). In the subgroup analyses, ABSI was significantly associated with overall complications regarding open gastrectomy (OR, 1.26; 95 % CI 1.01–1.57; P = 0.039). Regarding laparoscopy-
The study was approved by the institutional review board of the Korean National Cancer Center. (No. NCCNCS-12-571) Ó Society of Surgical Oncology 2013 First Received: 26 June 2013; Published Online: 4 December 2013 J. H. Lee, MD, PhD e-mail: [email protected]
assisted gastrectomy, ABSI had no significant effect on overall complications (P = 0.844). Conclusions. ABSI shows good correlation with surgical complications in patients with gastric cancer. Further studies are needed for the various clinical roles of ABSI, and the results could be helpful to determine the effect of abdominal obesity on gastric cancer surgery and the clinical usefulness of ABSI. The prevalence of overweight and obesity is rapidly increasing, and obesity is the fifth leading risk factor for global death according to the World Health Organization. Obesity is well known to be associated with several chronic diseases, including diabetes, hypertension, coronary heart disease, arthritis, and certain forms of cancer.1,2 Previous large-scale studies showed that overweight and obesity are an independent risk factors for all-cause mortality.3–5 In addition to the high comorbidity rate in obese patients, many surgeons, particularly abdominal surgeons, have felt burdened regarding obese patients because of the accumulated visceral fat tissue that can make operations technically difficult. Many studies have evaluated the relationship between obesity and surgical outcomes; however, the effect of obesity on surgical complications remains controversial. One cause may be that obesity, which was measured by body mass index (BMI), does not distinguish between muscle and fat accumulation or between central and general obesity.6–8 Ethnic differences regarding the distribution of central obesity may also have affected these contradictory results.9 Recently, a new anthropometric measure incorporating body shape, A Body Shape Index (ABSI), was developed by Krakauer and Krakauer.10 The researchers found that ABSI was a substantial risk factor for premature mortality in the general
1116
population and is expressed as the excess risk from a high waist circumference (WC). However, there are no data regarding whether ABSI is associated with surgical complications. Here, we aimed to evaluate whether ABSI had an effect on surgical complications in patients who underwent curative gastrectomy for gastric cancer. Gastrectomy with lymph node dissection is considerably influenced by abdominal shape and abdominal fat accumulation.11–13 METHODS
B. W. Eom et al.
disease), preoperative laboratory findings (hemoglobin, albumin, and creatinine), operation type, and whether a patient received adjuvant chemotherapy were identified. We classified operation type according to the surgical approach (open vs. laparoscopy) and extent of gastrectomy (subtotal vs. total gastrectomy). The cancers were staged by using the Union Internationale contre le Cancer tumor-node-metastasis system, 7th edition.15 Complication data were collected from operation day to last follow-up, which included both immediate postoperative complications and late complications such as adhesive ileus and incisional hernia.
Patients Statistical Analysis The medical records of consecutive patients undergoing gastric surgery from November 2001 to September 2012 were reviewed by using the database of the Center for Gastric Cancer at the National Cancer Center, Korea. Patients who underwent curative gastrectomy for gastric cancer were included in the current study regardless of operation type. We excluded those who received neoadjuvant chemotherapy, those who underwent palliative surgery, those who had distant metastasis, and those who had other cancers preoperatively or postoperatively. Additionally, patients who had one or more missing records among weight, height, and WC were excluded. In total, 4,813 patients were available for evaluating in this study. Patients underwent subtotal or total gastrectomy with more than D1 ? b (Nos. 7, 8, and 9) lymph node dissection for early gastric cancer and D2 dissection for advanced gastric cancer, as defined by the Japanese Gastric Cancer Association.14 After surgery, all patients were followed up regularly for 5 years, and most patients with stage II or more advanced gastric cancer received adjuvant chemotherapy. This study was approved by the Institutional Review Board at the National Cancer Center (No. NCCNCS-12-571). BMI and ABSI Patients’ height and body weight were examined preoperatively, and BMI was calculated as weight in kilograms divided by the square of height in meters (kg/ m2). WC in meters was measured with the tape measure positioned just above the uppermost lateral border of the ilium in the supine position. ABSI was calculated according to WC adjusted for height and weight as follows.10: ABSI ¼ WC ðmÞ weight ðkgÞ2=3 heightðmÞ5=6 WC ¼ ðm11=6 kg2=3 Þ: 2=3 BMI height1=2
All continuous values are presented as the mean ± SD. Statistical analysis was conducted by using the Chi square test for categorical variables and Student’s t test for continuous variables. A binary logistic regression model was used for multivariable analyses to determine whether ABSI had an independent effect on surgical complications. The prognostic values of covariates were expressed by calculating the odds ratios (ORs) and 95 % confidence intervals (CIs). A Pvalue less than 0.05 was considered statistically significant, and all statistical analyses were performed by using SAS software (version 9.1.3; SAS Institute, Cary, NC). RESULTS Demographics and Complications Table 1 summarizes patients’ clinicopathologic characteristics. The mean age of all patients was 57.8 ± 11.9 years, and 67.0 % were male. Approximately 64 % of patients had BMI in the normal range. The laparoscopic approach included 1,446 laparoscopy-assisted gastrectomies, 169 robot-assisted gastrectomies, and 12 hand-assisted laparoscopic surgeries. The most common tumor location was the lower third of the stomach, and more than half the patients had early gastric cancer. Among a total of 4,813 patients, 643 (13.4 %) patients had postoperative complications. We classified them as intraabdominal, wound, systemic, and late-functional complications (Table 2). The most common complication was adhesive ileus (2.8 %), followed by abscess/fluid collection (2.4 %), stricture (2.1 %), and acute wound problem (1.4 %). Mortality, defined as death within 30 days after operation, occurred in 12 patients (0.25 %).
Data Collection
Effect of ABSI on Complications
Clinicopathologic characteristics, comorbidities (hypertension, diabetes mellitus, heart disease, and pulmonary
Univariable and multivariable analyses were performed to evaluate the relationship between surgical complications
A Body Shape Index and Surgical Complications TABLE 1 Clinicopathologic characteristics (N = 4,813)
1117
Factor
Subgroup
Result
Age, y (mean ± SD) Sex, n (%)
57.8 ± 11.9 Male
3,226 (67.0)
Female
1,587 (33.0)
BMI, kg/m2 (mean ± SD)
23.7 ± 3.2 \18.5 kg/m2 (underweight)
BMI, n (%)
2
18.5–24.9 kg/m (normal)
3,072 (63.8)
25–29.9 kg/m2 (overweight)
1,413 (29.3)
C30 kg/m2 (obese) ABSI, m
11/6
-2/3
kg
(mean ± SD) Present
1,275 (26.5) 563 (11.7)
Diabetes mellitus, n (%)
Present
Heart disease,a n (%)
Present
140 (2.9)
Pulmonary disease,b n (%)
Present
88 (1.8)
Hemoglobin, n (%)
\10 g/dL
234 (4.9)
Albumin, n (%)
C10 g/dL \3.5 g/dL
4,579 (95.1) 152 (3.2)
C3.5 g/dL
4,659 (96.8)
B1.4 mg/dL
4,770 (99.1)
[1.4 mg/dL
43 (0.9)
Open
3,186 (66.2)
Tumor size, cm (mean ± SD) Surgical approach, n (%) Extent of gastrectomy, n (%) Location, n (%)
Lauren classification, n (%)
pT, n (%) BMI body mass index, ABSI a body shape index Heart disease included ischemic heart disease, arrhythmia, and valvular disease
pN, n (%)
b
Pulmonary disease included asthma, chronic obstructive pulmonary disease, bronchiectasis, and pneumoconiosis
129 (2.7) 0.079 ± 0.005
Hypertension, n (%)
Creatinine, n (%)
a
199 (4.1)
Adjuvant chemotherapy, n (%)
and BMI or ABSI (Table 3). In the univariable analysis, neither BMI nor ABSI had a significant effect on overall surgical complications. However, after adjusting 17 baseline variables (age; sex; hypertension; diabetes mellitus; heart disease; pulmonary disease; preoperative hemoglobin, albumin, and creatinine
4.5 ± 2.8 Laparoscopic
1,627 (33.8)
Subtotal
3,478 (72.3)
Total
1,335 (27.7)
Upper
655 (13.6)
Middle
1,412 (29.3)
Lower
2,068 (43.0)
Combined
678 (14.1)
Intestinal
2,216 (45.9)
Diffuse
2,004 (41.5)
Mixed Unidentified
432 (8.9) 161 (3.3)
1
2,603 (54.1)
2
663 (13.8)
3
906 (18.8)
4
641 (13.3)
0
3,096 (64.3)
1
639 (13.3)
2
485 (10.1)
3
593 (12.3)
No
3,623 (75.3)
Yes
1,190 (24.7)
levels; surgical approach; extent of gastrectomy; tumor size; tumor location; Lauren classification; pT; pN; and adjuvant chemotherapy) in the multivariable analysis, we found that ABSI showed an independent correlation with overall surgical complications (P = 0.041). For each 0.01 increase in ABSI, the odds of complication occurrence
1118
B. W. Eom et al.
TABLE 2 Postoperative complications (N = 4,813) Complication
n
%
No
4,170
86.6
40
0.8
53
1.1
Intraabdominal complications Bleeding Leakage Abscess/fluid collection
117
2.4
Stricture
102
2.1
Adhesive ileus
137
2.8
Gastric stasis
26
0.5
Internal hernia
7
0.1
Bowel perforation
7
0.1
Othersa Wound complications
7
0.1
Acute wound problem
68
1.4
Incisional hernia
24
0.5
Pneumonia
28
0.6
Embolism/thrombosis
11
0.2
Pleural effusion
3
0.1
Pneumothorax
3
0.1
Heart disease
4
0.1
Cerebral infarction
1
0.02
5
0.1
Risk of Complications According to ABSI or BMI
Systemic complications
Late functional complication Cholecystitis
overall complications (OR, 1.02; 95 % CI 0.99–1.06; P = 0.179). For intraabdominal complications, neither ABSI nor BMI was a significant risk factor (OR, 1.23; 95 % CI 0.96–1.58; P = 0.096 and OR, 1.01; 95 % CI 0.97–1.04; P = 0.729, respectively). Regarding laparoscopy-assisted gastrectomy, neither ABSI nor BMI showed a significant effect for overall complications (OR, 1.04; 95 % CI 0.70–1.55; P = 0.844 and OR, 1.05, 95 % CI 0.99–1.12; P = 0.088, respectively). Also, for intraabdominal complications, neither ABSI nor BMI had any significant effect (OR, 1.14; 95 % CI 0.75–1.76; P = 0.535 and OR 1.03; 95 % CI 0.96–1.10; P = 0.388, respectively).
a
Others included lymphorrhea (n = 3), splenic infarction (n = 2), and pseudomembranous colitis (n = 2)
increased by 22 %. However, BMI showed no statistical significance in overall complications (P = 0.063). For the rest, male sex, preoperative low hemoglobin level, open approach, total gastrectomy, and advanced N classification were significant risk factors for surgical complications. Next, we focused on intraabdominal complications, because ABSI or BMI was suggested to be more related to intraabdominal complications than to overall complications. In the multivariable analysis, neither ABSI nor BMI had statistical significance; however, the P-value of ABSI differed widely from that of BMI (P = 0.061 for ABSI; P = 0.501 for BMI; Table 4). Independent risk factors for intraabdominal complications were male sex, total gastrectomy, advanced N classification, and no adjuvant chemotherapy. Subgroup Analyses According to Surgical Approach Additionally, we performed multivariable subgroup analyses according to surgical approach. Regarding open gastrectomy, ABSI was significantly associated with overall complications (OR, 1.26; 95 % CI 1.01–1.57; P = 0.039). However, BMI had no significant effect on
We estimated surgical complication rates according to quartiles of ABSI or BMI (Fig. 1a). The overall complication rate increased as ABSI increased. However, the complication rate according to BMI was irregular. For intraabdominal complications, the complication rate increased gradually according to ABSI; however, no difference between the second and third quartiles of BMI was found, and the complication rate increased sharply in the fourth quartile of BMI (Fig. 1b). DISCUSSION The present study is the first to evaluate whether ABSI is associated with surgical complications worldwide. By multivariable analysis, we found that ABSI was an independent risk factor for overall complications in patients with gastric cancer. Compared with BMI, ABSI is better correlated with surgical complications. Surgical complications result in various clinical problems such as increased hospital duration, increased cost, longterm sequelae, and severe complications that can become life-threatening. Previous studies have evaluated the effect of BMI on surgical complications in various abdominal surgeries, and contradictory results have been shown.16–22 Several studies have shown that obesity increases postoperative complications such as anastomosis leak, wound infection, and pulmonary embolism.16–19 However, other studies have reported no difference in morbidity between obese and nonobese patients.20–22 Similar results were also shown in gastric cancer specifically. Higher rates of complications, larger amounts of blood loss, and longer operating times were found in high-BMI patients in some studies.23–25 However, other studies reported that high BMI itself might not increase operative morbidity.26–28 Conversely, some study groups have estimated abdominal shape–related measures such as WC, waist-to-hip
A Body Shape Index and Surgical Complications
1119
TABLE 3 Univariable and multivariable analysis of factors for overall complications Factor
Subgroup
Univariable Hazard ratio
Age Sex
Female vs. male
ABSI BMI
Multivariable (including BMI) Multivariable (including ABSI) P-value Hazard ratio
P–value
Hazard ratio
P
1.01 (1.00, 1.02)
0.012 1.00 (1.00, 1.02)
0.098
1.00 (1.00, 1.01)
0.77 (0.64, 0.93)
0.005 0.79 (0.65, 0.96)
0.019
0.73 (0.59, 0.91)
0.326 0.004
1.13 (0.96, 1.34)a
0.149
1.22 (1.01, 1.48)a
0.041
1.01 (0.98, 1.03)
0.571 1.03 (1.00, 1.06)
0.063
Hypertension
Yes vs. no
1.03 (0.85, 1.24)
0.798 0.97 (0.78, 1.20)
0.747
1.01 (0.82, 1.24)
0.965
Diabetes mellitus
Yes vs. no
0.96 (0.74, 1.25)
0.77
0.93 (0.70, 1.22)
0.585
0.92 (0.70, 1.22)
0.569
Heart diseaseb Pulmonary diseasec
Yes vs. no Yes vs. no
1.57 (1.03, 2.42) 2.07 (1.26, 3.40)
0.038 1.49 (0.94, 2.36) 0.004 1.62 (0.94, 2.79)
0.091 0.085
1.50 (0.95, 2.38) 1.57 (0.91, 2.71)
0.083 0.105
Hemoglobin
\10.0 vs. C 10.0 g/dL 2.03 (1.48, 2.79) \0.001 1.52 (1.06, 2.16)
0.022
1.50 (1.05, 2.13)
0.026
Albumin
\3.5 vs. C 3.5 g/dL
1.92 (1.30, 2.84)
0.001 1.27 (0.81, 2.01)
0.295
1.21 (0.77, 1.90)
0.419
Creatinine
C1.4 vs. \ 1.4 mg/dL 1.49 (0.69, 3.22)
0.313 1.06 (0.45, 2.48)
0.896
1.08 (0.46, 2.52)
0.861
Laparoscopic approach
Versus open
0.50 (0.41, 0.61) \0.001 0.70 (0.56, 0.89)
0.003
0.70 (0.56, 0.88)
0.003
Total gastrectomy
Versus subtotal
2.77 (2.34, 3.29) \0.001 2.21 (1.68, 2.90)
\0.001
2.21 (1.69, 2.91)
\0.001
1.09 (1.06, 1.12) \0.001 0.98 (0.95, 1.02)
0.375
0.98 (0.95, 1.02)
0.376
Tumor size Location
Lauren Classification pT
pN
Upper
1
1
1
Middle
0.41 (0.32, 0.53) \0.001 0.83 (0.60, 1.14)
0.243
0.83 (0.60, 1.14)
0.244
Lower
0.41 (0.32, 0.51) \0.001 0.87 (0.61, 1.22)
0.414
0.87 (0.61, 1.23)
0.422
Combined
0.85 (0.65, 1.11)
0.904
0.97 (0.71, 1.32)
0.825
Intestinal
1
0.231 0.98 (0.72, 1.34) 1
1
Diffuse
1.10 (0.92, 1.31)
0.313 1.13 (0.92, 1.38)
0.243
1.14 (0.93, 1.39)
0.223
Mixed
1.10 (0.81, 1.48)
0.544 1.16 (0.85, 1.59)
0.348
1.16 (0.85, 1.58)
0.355
1 2
1 1.40 (1.08, 1.79)
0.01
1 1.08 (0.81, 1.43)
0.61
1 1.06 (0.80, 1.41)
0.668
3
1.45 (1.15, 1.80)
0.001 0.99 (0.74, 1.33)
0.96
0.98 (0.73, 1.31)
0.893
0.05
1.40 (0.98, 1.99)
0.065
4
2.52 (2.01, 3.15) \0.001 1.43 (1.00, 2.04)
0
1
1
1.22 (0.94, 1.57)
0.129 1.15 (0.86, 1.54)
0.354
1.15 (0.86, 1.55)
2
1.52 (1.17, 1.99)
0.002 1.29 (0.92, 1.80)
0.135
1.31 (0.94, 1.83)
0.116
3
2.13 (1.70, 2.67) \0.001 1.60 (1.14, 2.24)
0.007
1.59 (1.13, 2.23)
0.007
1.38 (1.15, 1.66)
0.046
0.77 (0.59, 1.00)
0.051
Adjuvant chemotherapy Yes vs. no
1
0.001 0.76 (0.59, 1.00)
1 0.343
ABSI a body shape index, BMI body mass index a
Odds ratio for each 0.01 increase in ABSI
b
Heart disease included ischemic heart disease, arrhythmia, and valvular disease
c
Pulmonary disease included asthma, chronic obstructive pulmonary disease, bronchiectasis, and pneumoconiosis
ratio, abdominal anterior to posterior diameter, and visceral fat area.11–13,29–35 WC and waist-to-hip ratio have been identified to be associated with all-cause mortality in some large-scale studies.29–31 A few studies showed that visceral fat area was a more useful variable than BMI in terms of surgical complications such as pancreatic fistula and intraoperative bleeding.11,12,32–34 Moreover, Yamamoto et al.13 showed that celiac artery depth ratio was independently related with postoperative pancreatic fistula. However, few data are available concerning the relationship between WC or waist-to-hip ratio and postoperative complications. Additionally, visceral fat area and celiac
artery depth ratio require specialized software and a considerable amount of time and energy. The new anthropometric measure ABSI is calculated on the basis of WC adjusted for height and weight.10 Krakauer and Krakauer found that mean ABSI increased steadily from mid life into old age and was consistently higher in men than women after young adulthood. However, ageand sex-specific BMI and WC had different trends than ABSI, decreasing after approximately 60 years of age.10 Moreover, the mean WC was higher in men, whereas the mean BMI was higher in women. These different characteristics of ABSI compared with WC and BMI can be
1120
B. W. Eom et al.
TABLE 4 Univariable and multivariable analysis of factors for intraabdominal complications Factor
Subgroup
Univariable Hazard ratio
Age Sex
Female vs. male
ABSI BMI
Multivariable (including BMI) Multivariable (including ABSI) P-value Hazard ratio
P–value
Hazard ratio
P-value
1.00 (0.99, 1.01)
0.786 1.00 (0.99, 1.01)
0.6
1.00 (0.99, 1.01)
0.313
0.76 (0.62, 0.93)
0.008 0.75 (0.60, 0.94)
0.013
0.70 (0.55, 0.89)
0.003
1.07 (0.89, 1.29)a
0.476
1.23 (0.99, 1.52)
0.061
1.00 (0.97, 1.03)
0.967 1.01 (0.98, 1.04)
0.501
Hypertension
Yes vs. no
0.98 (0.80, 1.22)
0.881 0.98 (0.77, 1.25)
0.854
0.99 (0.78, 1.25)
0.927
Diabetes mellitus
Yes vs. no
1.09 (0.82, 1.44)
0.557 1.11 (0.82, 1.50)
0.512
1.10 (0.81, 1.49)
0.551
Heart diseaseb Pulmonary diseasec
Yes vs. no Yes vs. no
1.30 (0.78, 2.14) 2.00 (1.13, 3.41)
0.315 1.32 (0.77, 2.27) 0.016 1.54 (0.83, 2.83)
0.306 0.169
1.33 (0.78, 2.27) 1.51 (0.82, 2.77)
0.303 0.189
Hemoglobin
\10.0 vs. C10.0 g/dL 1.74 (1.22, 2.50)
0.002 1.35 (0.90, 2.01)
0.15
1.34 (0.89, 2.00)
0.157
Albumin
\3.5 vs. C3.5 g/dL
1.50 (0.94, 2.38)
0.086 0.94 (0.54, 1.63)
0.812
0.90 (0.52, 1.56)
0.706
Creatinine
C1.4 vs. \1.4 mg/dL 1.70 (0.75, 3.85)
0.201 1.31 (0.53, 3.24)
0.564
1.32 (0.53, 3.28)
0.545
Laparoscopic approach
Versus open
0.55 (0.44, 0.68) \0.001 0.83 (0.64, 1.07)
0.154
0.83 (0.64, 1.07)
0.156
Total gastrectomy
Versus subtotal
3.16 (2.62, 3.82) \0.001 2.63 (1.94, 3.56)
\0.001
2.64 (1.95, 3.57)
\0.001
1.10 (1.07, 1.13) \0.001 1.00 (0.96, 1.05)
0.891
1.00 (0.96, 1.05)
0.883
Tumor size Location
Lauren Classification pT
pN
Upper
1
1
Middle
0.40 (0.30, 0.52) \0.001 0.89 (0.63, 1.26)
0.506
0.89 (0.63, 1.26)
0.504
Lower
0.36 (0.28, 0.46) \0.001 0.90 (0.62, 1.32)
0.598
0.90 (0.62, 1.32)
0.602
Combined
0.83 (0.62, 1.10)
0.628
0.91 (0.65, 1.28)
0.588
Intestinal
1
0.193 0.92 (0.65, 1.29)
1
1
1
Diffuse
1.22 (1.00, 1.48)
0.055 1.13 (0.90, 1.42)
0.304
1.13 (0.90, 1.43)
0.285
Mixed
1.37 (1.00, 1.89)
0.052 1.38 (1.00, 1.92)
0.059
1.38 (0.99, 1.93)
0.057
1 2
1 1.45 (1.10, 1.92)
1 0.009 1.13 (0.83, 1.55)
0.445
1 1.13 (0.82, 1.54)
0.46
3
1.51 (1.18, 1.93)
0.001 1.03 (0.74, 1.43)
0.852
1.03 (0.74, 1.42)
0.88
0.147
1.34 (0.90, 1.99)
0.15
4
2.47 (1.93, 3.16) \0.001 1.34 (0.90, 2.00)
0
1
1
1.36 (1.04, 1.79)
0.028 1.27 (0.92, 1.75)
0.149
1.27 (0.92, 1.76)
2
1.52 (1.13, 2.04)
0.006 1.32 (0.91, 1.92)
0.143
1.34 (0.92, 1.95)
0.127
3
2.22 (1.73, 2.84) \0.001 1.70 (1.17, 2.45)
0.006
1.70 (1.17, 2.47)
0.006
1.46 (1.20, 1.79) \0.001 0.73 (0.55, 0.98)
0.037
0.73 (0.55, 0.98)
0.037
Adjuvant chemotherapy Yes vs. no
1
1 0.145
ABSI a body shape index, BMI body mass index a
Odds ratio for each 0.01 increase in ABSI
b
Heart disease included ischemic heart disease, arrhythmia, and valvular disease
c
Pulmonary disease included asthma, chronic obstructive pulmonary disease, bronchiectasis, and pneumoconiosis
helpful for understanding ABSI. Briefly, a high ABSI might correspond with a greater fraction of visceral fat compared with peripheral tissue at a given height and weight.36 ABSI was recently developed, and few data are available for it. Krakauer and Krakauer reported that ABSI can be a substantial risk factor for premature mortality in the general US population.10 Since then, identified that ABSI was a better predictor of blood pressure than BMI or WC in Portuguese adolescents Duncan et al.37 However, did not show the superiority of ABSI in predicting diabetes
mellitus in the Chinese population compared with BMI or WC He et al.36 In this study, ABSI showed a good relationship with surgical complications in gastric cancer patients. One possible reason is that radical gastrectomy with D2 lymph node dissection is strongly affected by abdominal fat accumulation. There are many major vessels in the gastrectomy field, and excessive adipose tissue conceals these vascular structures, resulting in bleeding or adjacent organ injury. Excess fat tissue also makes lymph node dissection more difficult and induces incomplete lymph node
A Body Shape Index and Surgical Complications FIG. 1 (a) Overall complication rates according to quartiles of ABSI (blue) and BMI (yellow). (b) Intraabdominal complication rates according to quartiles of ABSI (blue) and BMI (yellow)
1121
(a)
(b)
20
8
15
6
4
10
ABSI BMI
ABSI BMI 2
5
0−25
25−50
dissection and a long operating time. Moreover, the surgical approach for anastomosis is risky in patients with a long anterior to posterior diameter that can lead to anastomosis leakage or stricture. These factors would affect the significance of ABSI. Additionally, according to subgroup analyses, ABSI had a significant effect on surgical complications in open surgery compared with laparoscopic surgery. Particularly, ABSI had a stronger statistical relationship with overall complications compared with BMI in open surgery. These results coincided with many previous studies. Many previous studies have reported increased morbidity in open gastric surgery and no significant difference in laparoscopic surgery.23,24,27,34,35 According to the study by Makino et al., laparoscopy-assisted distal gastrectomy is not influenced by obesity, whereas open distal gastrectomy may be influenced by obesity, a result that is identical to that in the current study.34 Therefore, it seems that abdominal obesity should be considered for surgical complications more in open gastrectomy than in laparoscopic surgery. Potential limitations of the current study include misclassification bias resulting from retrospective characteristics. Selection bias also cannot be excluded because patients with missing data were excluded. Furthermore, although we statistically controlled relevant confounders, some confounders can be missed in the multivariable analysis. Finally, surgical complications were not classified systemically, such as with the severity grading system, because of the lack of clinical data.38 Nevertheless, a large-scale cohort and a multivariable statistical method might compensate for these limitations. In conclusion, as a new anthropometric measure, ABSI shows good correlation with surgical complications in patients with gastric cancer. Surgeons can consider not only the patient’s sex, preoperative laboratory findings,
50−75
75−100
0−25
25−50
50−75
75−100
surgical method, and clinical stage, but also abdominal obesity, as risk factors for surgical complications by using ABSI. We are monitoring patients’ survival to identify the effect of ABSI on survival. Further studies are also needed for the various clinical roles of ABSI, and the results could be helpful to determine the effect of abdominal obesity on gastric cancer surgery and the clinical usefulness of ABSI. ACKNOWLEDGMENT This work was supported by a grant of the National Cancer Center (NCC-1110531-1,2).
REFERENCES 1. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA. 2003;289:76–79. 2. Kopelman PG. Obesity as a medical problem. Nature. 2000;404:635–643. 3. Pischon T, Boeing H, Hoffmann K, Bergmann M, Schulze MB, Overvad K et al. General and abdominal adiposity and risk of death in Europe. N Engl J Med. 2008;359:2105–2120. 4. Berrington de Gonzalez A, Hartge P, Cerhan JR, Flint AJ, Hannan L, MacInnis RJ et al. Body-mass index and mortality among 1.46 million white adults. N Engl J Med. 2010;363:2211–2219. 5. Boggs DA, Rosenberg L, Cozier YC, Wise LA, Coogan PF, RuizNarvaez EA et al. General and abdominal obesity and risk of death among black women. N Engl J Med. 2011;365:901–908. 6. Nevill AM, Stewart AD, Olds T, Holder R. Relationship between adiposity and body size reveals limitations of BMI. Am J Phys Anthropol. 2006;129:151–156. 7. Heymsfield SB, Scherzer R, Pietrobelli A, Lewis CE, Grunfeld C. Body mass index as a phenotypic expression of adiposity: quantitative contribution of muscularity in a population-based sample. Int J Obes. 2009;33:1363–1373. 8. Go´mez-Ambrosi J, Silva C, Galofre´ JC, Escalada J, Santos S, Milla´n D et al. Body mass index classification misses subjects with increased cardiometabolic risk factors related to elevated adiposity. Int J Obes. 2012;36:286–294. 9. Agyemang C, van Valkengoed IG, van den Born BJ, Bhopal R, Stronks K. Heterogeneity in sex differences in the metabolic syndrome in Dutch white, Surinamese African and South Asian populations. Diabet Med. 2012;29:1159–1164.
1122 10. Krakauer NY, Krakauer JC. A new body shape index predicts mortality hazard independently of body mass index. PLoS One. 2012;7:e39504. 11. Lee JH, Paik YH, Lee JS, et al. Abdominal shape of gastric cancer patients influences short-term surgical outcomes. Ann Surg Oncol. 2007;14:1288–1294. 12. Tanaka K, Miyashiro I, Yano M, et al. Accumulation of excess visceral fat is a risk factor for pancreatic fistula formation after total gastrectomy. Ann Surg Oncol. 2009;16:1520–1555. 13. Yamamoto N, Oshima T, Sato T, Makino H, Nagano Y, Fujii S et al. Upper abdominal body shape is the risk factor for postoperative pancreatic fistula after splenectomy for advanced gastric cancer: a retrospective study. World J Surg Oncol. 2008;6:109. 14. Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 2nd English ed. Gastric Cancer. 1998;1: 10–24. 15. Edge SB, Byrd DR, Compton CC. AJCC cancer staging manual. 7th ed. New York: Springer-Verlag; 2010. 16. Pikarsky AJ, Saida Y, Yamaguchi T, et al. Is obesity a high-risk factor for laparoscopic colorectal surgery? Surg Endosc. 2002;16:855–858. 17. Merkow RP, Bilimoria KY, McCarter MD, Bentrem DJ. Effect of body mass index on short-term outcomes after colectomy for cancer. J Am Coll Surg. 2009;208:53–61. 18. Mathur AK, Ghaferi AA, Osborne NH, et al. Body mass index and adverse perioperative outcomes following hepatic resection. J Gastrointest Surg. 2010;14:1285–1291. 19. House MG, Fong Y, Arnaoutakis DJ, Sharma R, Winston CB, Protic et al. Preoperative predictors for complications after pancreaticoduodenectomy: impact of BMI and body fat distribution. J Gastrointest Surg. 2008;12:270–278. 20. Dindo D, Muller MK, Weber M, Clavien PA. Obesity in general elective surgery. Lancet. 2003;361:2032–2035. 21. Al-Refaie WB, Parsons HM, Henderson WG, et al. Body mass index and major cancer surgery outcomes: lack of association or need for alternative measurements of obesity? Ann Surg Oncol. 2010;17:2264–2273. 22. Balentine CJ, Enriquez J, Cruz G, et al. Obesity does not increase complications following pancreatic surgery. J Surg Res. 2011;170:220–225. 23. Inagawa S, Adachi S, Oda T, Kawamoto T, Koike N, Fukao K. Effect of fat volume on postoperative complications and survival rate after D2 dissection for gastric cancer. Gastric Cancer. 2000;3:141–144. 24. Kodera Y, Sasako M, Yamamoto S, Sano T, Nashimoto A, Kurita A. Identification of risk factors for the development of complications following extended and superextended lymphadenectomies for gastric cancer. Br J Surg. 2005;92:1103–1109. 25. Kunisaki C, Makino H, Takagawa R, et al. Predictive factors for surgical complications of laparoscopy-assisted distal gastrectomy for gastric cancer. Surg Endosc. 2009;23:2085–2093.
B. W. Eom et al. 26. Gretschel S, Christoph F, Bembenek A, Estevez-Schwarz L, Schneider U, Schlag PM. Body mass index does not affect systematic D2 lymph node dissection and postoperative morbidity in gastric cancer patients. Ann Surg Oncol. 2003;10:363–368. 27. Lee HJ, Kim HH, Kim MC, et al. The impact of a high body mass index on laparoscopy assisted gastrectomy for gastric cancer. Surg Endosc. 2009;23:2473–2479. 28. Oki E, Sakaguchi Y, Ohgaki K, et al. The impact of obesity on the use of a totally laparoscopic distal gastrectomy in patients with gastric cancer. J Gastric Cancer. 2012;12:108–112. 29. Simpson JA, MacInnis RJ, Peeters A, Hopper JL, Giles GG, English DR. A comparison of adiposity measures as predictors of all-cause mortality: the Melbourne Collaborative Cohort Study. Obesity. 2007;15:994–1003. 30. Seidell JC. Waist circumference and waist/hip ratio in relation to all-cause mortality, cancer and sleep apnea. Eur J Clin Nutr. 2010;64:35–41. 31. Petursson H, Sigurdsson JA, Bengtsson C, Nilsen TI, Getz L. Body configuration as a predictor of mortality: comparison of five anthropometric measures in a 12 year follow-up of the Norwegian HUNT 2 study. PLoS One. 2011;6:e26621. 32. House MG, Fong Y, Arnaoutakis DJ, Sharma R, Winston CB, Protic M et al. Preoperative predictors for complications after pancreaticoduodenectomy: impact of BMI and body fat distribution. J Gastrointest Surg. 2008;12:270–278. 33. Tsujinaka S, Konishi F, Kawamura YJ, et al. Visceral obesity predicts surgical outcomes after laparoscopic colectomy for sigmoid colon cancer. Dis Colon Rectum. 2008;51:1757–1765. 34. Makino H, Kunisaki C, Izumisawa Y, et al. Effect of obesity on laparoscopy-assisted distal gastrectomy compared with open distal gastrectomy for gastric cancer. J Surg Oncol. 2010;102:141–147. 35. Ojima T, Iwahashi M, Nakamori M, M Nakamura, K Takifuji, M Katsuda et al. The impact of abdominal shape index of patients on laparoscopy-assisted distal gastrectomy for early gastric cancer. Langenbecks Arch Surg. 2012;397:437–445. 36. He S, Chen X. Could the new body shape index predict the new onset of diabetes mellitus in the Chinese population? PLoS One. 2013;8:e50573. 37. Duncan MJ, Mota J, Vale S, Santos MP, Ribeiro JC. Associations between body mass index, waist circumference and body shape index with resting blood pressure in Portuguese adolescents. Ann Hum Biol. 2013;40:163–167. 38. Strasberg SM, Linehan DC, Hawkins WG. The accordion severity grading system of surgical complications. Ann Surg. 2009; 250: 177–186.
