Clinical impact of perirenal thickness on short‐ and long‐term outcomes of gastric cancer after curative surgery

Abstract Background A variety of factors for short‐ and long‐term outcomes have been reported after radical resection for gastric cancer (GC). Obesity and emaciation had been reported to be a cause of poor short‐ and long‐term outcomes with gastrointestinal cancer. However, the indicators are still controversial. The purpose of this study was to evaluate the relationship between perirenal thickness (PT) and short‐ and long‐term outcomes after radical surgery for GC. Methods We analyzed the data of 364 patients with GC who underwent radical surgery. We evaluated the distance from the anterior margin of the quadratus lumborum muscle to the dorsal margin of the left renal pole using computed tomography (CT) as an indicator of PT. The association between PT and clinicopathological factors and short‐ and long‐term outcomes was evaluated. Results The PT data were divided into low, normal, and high groups by gender using the tertile value. We found that the PT low group was 121 patients, normal group was 121 patients, and high group was 122 patients. Multivariate analyses showed that the high PT group was an independent risk factor for a short‐outcome after curative surgery in GC patients (odds ratio 2.163; 95% confidence interval [CI] 1.156–4.046; P = .016). And the low PT group was an independent risk factor for overall survival (hazard ratio 2.488; 95% CI 1.400–4.421; P = .0019) and relapse‐free survival (hazard ratio 2.342; 95% CI 1.349–4.064; P = .0025) after curative surgery in GC patients. Conclusion Perirenal thickness is a simple and useful factor for predicting short‐ and long‐term outcomes after radical surgery for GC.


| INTRODUC TI ON
Gastric cancer (GC) is the third-most common cancer and the second-most common cause of death worldwide. 1 Radical surgery with lymphadenectomy offers a curative result for patients with resectable GC, and a variety of factors for short-and long-term outcomes have been reported.
Obesity is associated with postoperative complications and worse prognosis in patients with gastrointestinal cancer, and is regarded as a major technical limiting factor for surgery because of the substantial surgical difficulties caused by abundant visceral fat (VF) and a narrow operating field. Many reports have demonstrated that obesity and abundant VF are associated with worse surgical outcomes after various surgical procedures. In these reports, obesity was associated with increased surgery time, blood loss, and postoperative complications. 2,3 Moreover, the negative impact of postoperative complications on overall survival (OS) and relapse-free survival (RFS) after radical surgery for GC has been reported and discussed. [4][5][6] Similarly, emaciation has been reported to be a cause of poor short-and long-term outcomes from low nutrition and sarcopenia. [7][8][9] The body mass index (BMI) is one such indicator of obesity and emaciation; however, it is controversial. [10][11][12] Hiki et al previously reported that the thickness of the adipose tissue dorsal to the left kidney correlated with the amount of VF and short outcome after laparoscopic distal gastrectomy for early GC. 13 However, there is no report on the association between longterm outcome and perirenal thickness (PT). We hypothesized that when there is much PT, postoperative complications increase and when there is little PT, the long-term prognosis worsens from low nutrition and sarcopenia compared to when the PT is normal. The purpose of this study was to evaluate the relationship between perirenal thickness and short-and long-term outcomes after surgery for GC. The primary outcome was overall survival. The secondary outcomes were postoperative complications rate and RFS.

| Study subjects and outcomes
OS was defined as the interval from surgery to the date of death from any cause, or the last follow-up in living patients. RFS was defined as the interval from surgery to either the first recurrence or death from any cause.

| Definition of postoperative complications
The severity of postoperative complications was evaluated using the Clavien-Dindo classification system. 15 We defined postoperative complications in this study as conditions classified as Clavien-Dindo grade II or higher, severe postoperative complications as Clavien-Dindo grade IIIa or higher, and inflammatory complications such as anastomotic leakage, pancreatic fistula, intraabdominal infection, infection at the surgical site, and pneumonia.

| Measurement of the perirenal thickness
We evaluated the distance from the anterior margin of the quadratus lumborum muscle to the dorsal margin of the left renal pole using CT of the transverse section, 5-mm slice, and venous phase as in our previous article. 13 We took our measurements using the left kidney because the position of the right kidney may be influenced by the position of the liver. We decided to take our measurement at the level where the left renal vein exits the kidney, and measured the distance from the anterior margin of the quadratus lumborum muscle to the dorsal margin of the left renal pole as an indicator of PT.
To clarify the relationship between PT and short-and long-term outcomes, we divided into low, normal, and high groups for PT date by using the tertile value. Since there are differences in the amount of VF between male and female, the present study was also analyzed separately for male and female.

| Statistics
The clinicopathological characteristics and laboratory data of the two groups were compared using the chi-squared test for categorical variables and the Mann-Whitney U-test for continuous variables. P < .05 was considered statistically significant. In the univariate and multivariate analyses, continuous variables, including the preoperative laboratory data, surgical duration time, and amount of bleeding, were divided by the median. We used the Cox proportional hazards model to assess the effects of covariates during both the univariate and multivariate analyses; the categorical covariates are listed in Table 3. We conducted multivariate analysis using factors with P < .05 in the univariate analysis. All tests were analyzed using JMP software (SAS Institute, Cary, NC, USA). In total, we found that the PT low group was 121 patients, the PT normal group was 121 patients, and the PT high group was 122 patients.

| Patients' and baseline characteristics
In terms of the preoperative factors, there was a significantly higher number of patients who were male in the PT high group. In terms of surgical factors, there was a significantly longer surgical time, higher amount of blood loss, and higher frequency of postoperative complications (including severe and inflammatory complications) in the PT high group. However, there was no significant difference in cStage, pStage, between the groups ( Table 1).

| Risk factors for postoperative complications
Variables that achieved a probability value of .05 in the univariate analysis were included in a subsequent multivariate analysis to identify risk factors for postoperative complications. In the multivariate   Figure 1A,B, respectively. In the multivariate analysis, the PT low was selected as an independent predictive factor for OS (hazard ratio 2.488; 95% CI 1.400-4.421; P = .0019, Table 3) and RFS (hazard ratio 2.342; 95% CI 1.349-4.064; P = .0025, Table 4) in GC patients who underwent curative gastrectomy. However, PT high was not selected as an independent predictive factor for OS and RFS in this cohort.

| DISCUSS ION
In the present study we demonstrated that PT is a simple and useful predictor for the short-and long-term outcomes after curative surgery in GC patients. Our results indicated that the high PT group was associated with a higher frequency of postoperative complications, and the low PT group was associated with a potentially worsening prognosis. To the best of our knowledge, this is the first report to reveal an association between PT and short-and long-term outcomes after curative surgery in GC patients.
We found that the high PT group was associated with shortterm outcomes after radical surgery for GC. BMI is widely used as an indicator of obesity in clinical settings. A higher BMI has been associated with increased complications in several studies. 3,[16][17][18] However, another study demonstrated that there was no difference in surgical outcomes on the basis of BMI. 19 These results show that the issue of whether BMI is correlated with short-term outcomes remains controversial. Because BMI is calculated using body weight and height, it considers not only the VF but also the subcutaneous fat. This makes it impossible to distinguish between the two using this indicator alone. Therefore, we considered that BMI may not be suitable for measuring obesity in surgical patients and that VF volume is more indicative of obesity. However, the precise measurement of the entire VF volume requires specialized software. The method of measuring VF using CT scans and an image processing 3-D image analysis system (eg: SYNAPSE VINCENT, DICOM images) is complex and is not suited to routine medical practice. Therefore, we focused on PT as a simple indicator of VF. This study revealed that CT assessment of PT, which was validated as a surrogate of VF, was superior to BMI for GC patients. The simplicity with which PT can be measured makes it a clinically valuable tool. Using PT, surgical teams can assess the risk of complications before surgery.
Several recent studies revealed that the occurrence of postoperative complications could have a significantly negative impact on the prognosis of patients with GC. [4][5][6]20 Therefore, it is important to prevent the occurrence of complications, and prediction of morbidities before surgery is considered an important task, especially in patients with malignancy. Indeed, this study revealed that postoperative complications were an independent prognostic factor after radical surgery for GC. In this study, high PT was an independent predictor of short-term postoperative outcome, while it tended to be associated with a worse prognosis but was not an independent prognostic factor. Analysis of subgroups suggests that even high PT may not worsen the long-term outcome without postoperative complications ( Figure S1). This suggests that the high level of recruitable energy from intraabdominal fat tissue may have prevented the prognosis from worsening. 21 We found that the low PT group was associated with long-term outcomes after radical surgery for GC. The relationship between prognosis and the low PT content may be attributed to several factors. store, which they can access in times of negative energy balance. 21 Conversely, the basic physical capacity is much reduced in patients with low VF reserves and should present as a poor prognostic feature. 22 Previous studies reported various systems for predicting surgical risks. The physiological and operative severity score for the enumeration of mortality and morbidity (POSSUM), 23  This study had several limitations. First, this study was a retrospective analysis, and thus might be subject to several biases.
Second, the cutoff values used in this study may differ from those applied at other institutions; the present study analyzed continuous variables of only two institutions. Different institutions may have different cutoff values. Therefore, a prospective multiinstitutional study is desirable to validate the present findings.
Interobserver reliability for these types of measurements needs to be demonstrated in further multi-institutional studies. Third, this study only focused on the score at one point before surgery, and we also need to analyze cases with exercise and/or nutritional interventions in the future.
In conclusion, this study is the first to reveal the association between PT and short-and long-term outcomes after curative surgery in GC patients. When we are able to undertake appropriate preoperative evaluations using PT, it will be possible to minimize complications to the greatest possible extent and may improve prognosis.

ACK N OWLED G M ENTS
We thank H. Nikki March, PhD, Edanz Group (https://en-autho rservi ces.edanz.com/ac) for editing a draft of this article.