Visceral obesity determined by CT as a predictor of short-term postoperative complications in ovarian cancer

Objective: To explore the association between visceral obesity and short-term postoperative complications in patients with advanced ovarian cancer undergoing cytoreductive surgery. Methods: Medical records were reviewed for patients with advanced epithelial ovarian cancer. Visceral fat area, subcutaneous fat area and total fat area were measured on a single slice at the level of L3/4 of a preoperative CT scan. The Receiver Operating Characteristic (ROC) curve was used to calculate the optimal cut-off value for visceral fat area. Relationships between visceral fat area and characteristics of ovarian cancer patients were analyzed. Univariable and multivariable Logistic regression analyseswere performed to investigate the value of perioperative characteristics of patients on short-term complications. Results: According to the ROC curve, the best cut-off value of VFA was 93 cm². Of the 130 patients, 53.8% (70/130) were presented visceral obesity. Patients with visceral obesity were older than those with non-visceral obesity (58.4 years old vs. 52.1 years old, p < 0.001). The proportion of patients with hypertension was higher (35.7% vs. 13.3%, p = 0.003). The total fat area and subcutaneous fat area were higher in patients with visceral obesity (294.3 ± 75.5 vs. 176.2 ± 68.7, p < 0.001; 158.9 ± 54.7 vs. 121.7 ± 52.6, p < 0.001). Compared with patients in the non-visceral obese group, patients in the visceral obese group were more likely to have postoperative fever (21/70 30.0% vs. 8/60 1.25%, p = 0.023), leading to a longer length of hospital stay (21 days vs. 17 days, p = 0.009). Time from surgery to adjuvant chemotherapy for patients with visceral obesity has been delayed (24 days vs. 19 days, p = 0.037). Multivariate analysis showed that visceral obesity (OR = 6.451, p < 0.001) and operation time (OR = 1.006, p < 0.001) were independent predictors of postoperative complications. Conclusion: Visceral obesity is an important risk factor for short-term postoperative complications in patients with advanced ovarian cancer undergoing cytoreductive surgery.


Introduction
Ovarian cancer is the most lethal gynecologic malignancies [1].Cytoreductive surgery and chemotherapy are considered the standard treatments for advanced ovarian cancer.Removing all visible tumors during cytoreductive surgery provides the best survival outcome for patients [2].However, patients with advanced ovarian cancer often have extensive metastases in the abdominal and pelvic cavity, including the omentum, intestine, peritoneum, appendix, liver, abdominal and pelvic lymph nodes [3].Complete removal of tumors may only be achievable through more aggressive surgery, which may result in higher complication rates, including tissue damage, massive bleeding, prolonged operating time, etc [4].Therefore, identifying risk factors that can predict short-term complications will provide value for the management of perioperative patients.
Hughes et al. recently reported a signi cant difference in the overall complication incidence of obese patients and nonobese patients after major abdominal surgery in gastric, rectal and liver cancers [5].
Apart from increasing di culties during cytoreductive surgery, obesity may also contribute to more postoperative complications and higher costs [6].Body mass index (BMI) is the most common indicator of obesity.However, it cannot be used to describe the distribution of abdominal fat.Recently, visceral fat area (VFA), total fat area (TFA) and subcutaneous fat area (SFA) have been proposed to assess fat distribution in obese patients [7].Takeuchi et al. found that patients with high visceral fat areas were associated with postoperative complications in gastric cancer [8].However, whether visceral obesity could predict short-term postoperative complications in patients with ovarian cancer remains unclear.
Computed tomography (CT) scans are broadly used to assess the condition of patients with ovarian cancer and are more accurate and sensitive for assessing fat distribution [9][10].
Thus, the aim of this study was to determine the relationship between visceral obesity determined by CT scans and the short-term postoperative complications of ovarian cancer.Visceral obesity is an important predictor of short-term complications after cytoreductive surgery for advanced ovarian cancer.

Methods
Patients with ovarian cancer undergoing cytoreductive surgery between January 2018 and December 2020 in the First A liated Hospital of Nanjing Medical University were included, and a preoperative abdominal CT scan within 3 weeks was included.Those who had incomplete perioperative data or underwent any treatment after CT scan were excluded.All complications occurring within 30 days after surgery were included and classi ed according to the Clavien-Dindo classi cation (CDC) [11].Grade I complications can be treated with antipyretics, diuretics, analgesics, and uid replacement in the clinic, which were regarded as not clinically signi cant complications.This study only analyzed complications of grade II and above.If one patient had multiple complications with different grades, the classi cation of complications was based on the highest complication rate.Extent and complexity of surgery was categorized using a previously described surgical complexity score (SCS) [12], which were de ned as low, intermediate, and high (scores 1 to 3, 4 to 7, and ≥ 8, respectively).This study was approved by the Institutional Review Board at the First A liated Hospital of Nanjing Medical University.
Image analysis was performed using ImageJ software (ImageJ; The National Institutes of Health, 105 Washington, MD, USA; version 1.47).Preoperative CT scan images of 130 patients with ovarian cancer at the level of L3-L4 were completely obtained.In this study, VFA, TFA and SFA were measured using a tissue attenuation range of -190 ~ -30 HU [13].(Fig. 1) Statistical analysis: SPSS statistics v24 was used for all statistical analysis.The correlation between VFA and BMI was analyzed by Spearman correlation.The best cut-off value of VFA was determined by the Receiver Operating Characteristic curve (ROC curve).Continuous data are reported as the mean ± SD, and categorical data are reported as numbers (percentage).Comparisons between groups were made by Student's test or Mann-Whitney U test for continuous data and chi-squared test or Fisher's exact test for categorical variable data.A p-value < 0.05 was considered statistically signi cant.Univariate and multivariate logistic regression analyses were performed to explore the predictors of postoperative complications in patients with advanced ovarian cancer undergoing cytoreductive surgery.The variables with a p-value < 0.05 in the univariate analysis were evaluated in a multivariate logistic regression analysis.

Results
A total of 130 patients were included in this study.We rst analyzed the correlation between BMI and VFA, and BMI and VFA were found to have a positive correlation (r 2 = 0.202, p < 0.001) (Fig. 2).Based on shortterm postoperative complications, ROC curve was used to analyze, the curve AUC was 0.702, and the best cut-off value for VFA was 93 cm 2 (sensitivity was 70.8%, speci city was 69.0%).
The clinicopathological characteristics of the 130 patients are summarized in Table 1.Their median age was 54 years (range, 24-80 years).There is a lack of visceral obesity standards for the Chinese population based on CT measurements.According to the ROC curve, 93 was determined to be the best cut-off value to divide patients into two groups, visceral obesity and non-visceral obesity.Patients with visceral obesity were older than those with non-visceral obesity (58.4 years old vs. 52.1 years old, p < 0.001).TFA and SFA were much higher in the visceral obesity group (294.3 ± 75.5 vs. 176.2± 68.7, p < 0.001; 158.9 ± 54.7 vs. 121.7 ± 52.6, p < 0.001).Patients with visceral obesity were more likely to have hypertension (35.7% vs. 13.3%, p = 0.003).In patients with visceral obesity, lipoprotein a (LPa) was lower (206.7 ± 225.5 vs.305.5 ± 276.9, P = 0.037), while triglycerides (TG) (1.8 ± 0.8 vs. 1.3 ± 0.6, p = 0.012) was higher.However, serum CA125 and HE4 levels were not signi cantly different between the two groups (Table 2).Due to the large scope of cytoreductive surgery for advanced ovarian cancer, postoperative complications will increase.We found that 39% (51/130) of patients with advanced ovarian cancer had ≥ 2 postoperative complications.In addition, we analyzed the correlation between visceral obesity and postoperative complications in patients with stage III-IV disease, and the number of postoperative complications in patients with visceral obesity increased signi cantly (p < 0.001).Patients with visceral obese prolonged the time from surgery to adjuvant chemotherapy (p = 0.037).Patients in the visceral obese group were more likely to have postoperative fever (p = 0.023) (Table 3).Postoperative treatment and rehabilitation of patients with fever were reported in Table 4.More types of antibiotics were used in patients with postoperative fever.And the using time of antibiotics and length of hospital stay prolonged signi cantly (p < 0.001, p < 0.001, p = 0.009) (Table 4).To investigate whether the characteristics of patients with visceral obesity have independent predictive value for the occurrence of complications, univariate and multivariate logistic regression analyses were performed.Univariate analysis showed that visceral obesity, operation time and intraoperative blood loss were signi cantly associated with postoperative complications.In multivariate analysis, visceral obesity (OR = 6.451, p < 0.001) and operation time (OR = 1.006, p = 0.009) were independent predictors of postoperative complications (Table 5).The AUCs of VFA, SFA, TFA, and BMI were 0.702, 0.507, 0.609, and 0.572, respectively.Compared with SFA, TFA and BMI, VFA can better assess visceral obesity and can better predict short-term complications after ovarian cancer surgery.(Fig. 3)

Discussion
Epidemiological evidence con rms that obesity is a risk factor for the onset of a variety of cancers [14][15].To determine obesity, BMI is typically used, and patients with a BMI of ≥ 30 kg/m² are de ned as obese.However, BMI does not re ect the whole-body fat distribution [16].Traditionally, intra-abdominal fat is indirectly measured by waist circumference, hip circumference or waist-to-hip ratio [17].Whereas, these parameters do not re ect the distribution of body fat either.In the present study, CT was used to measure the visceral fat area directly.Our results showed that patients could have visceral obesity even with a normal BMI.Nevertheless, an appropriate range of CT values for adipose tissue segmentation has not yet been determined.In a recent study, visceral obesity in females was de ned as > 80.1 cm², using metabolic syndrome (MetSyn) as an indicator of obesity-associated dysmetabolism in obesityassociated cancer [18][19].Heus et al. found that the visceral fat area threshold was 100 cm² and 130 cm², and 100 cm² had a better correlation with postoperative complications [20].In the present study, ROC curve was used to de ne the best cutoff value.The incidence of complications after cytoreductive surgery increased signi cantly in patients with visceral obesity.This suggests that providers should design strategies to reduce complications and be more aware of the possibility of complications.
The occurrence of postoperative complications is an important unfavorable factor for the rapid recovery of patients [21].It is particularly critical to identify the factors predicting postoperative complications.
Severe complications can be the result of complex procedures.These complications can result in delays of the start of adjuvant therapy, which can worsen the condition.In a study of 369 patients who had an abdominal or laparoscopic procedure for proven or suspected gynecological cancer, Kondalsamy-Chennakesavan et al. demonstrated that surgical complexity was one of the independently predictive factors of an adverse events [22].In our study, the complexity of the operation was scored by SCS.Univariate analysis also showed that SCS weas signi cantly associated with postoperative complications, while result of multivariate analysis was not.The reason for this inconsistency may be due to small size of enrolled patients.
Whether visceral obesity can be used to predict postoperative complications in patients has been controversial.C. Heus et al. found that patients with visceral obesity had an increased risk of postoperative complications [23].In contrast, the study by Rutten et al. showed that there was no correlation between visceral obesity and postoperative complications [24].It is worth noting that the characteristics of the patients included in these two studies were not consistent.Moreover, the operation period was also inconsistent, and the selected CT scan images and method of measurement were also different.In this study, VFA was manually traced on a single transverse slice at the level of L3-L4.
Importantly, visceral obesity de ned as a VFA of ≥ 93 cm 2 was identi ed as an independent prognostic factor associated with postoperative complications.In addition, Boutin et al. found that there was no signi cant relationship between adipose tissue distribution and postoperative complications in patients with soft tissue sarcoma [25].The results from three studies on complications of radical resection of gastrointestinal tumors revealed that an increase in visceral fat could lead to a prolonged operation time, poor recovery and postoperative complications [26][27][28].These studies suggest that visceral obesity may only have predictive value in certain types of tumors.
Postoperative fever is common in patients undergoing cytoreductive surgery.A postoperative body temperature rise (< 38.3°C) does not require special treatment [29].In this study, we found that patients with visceral obesity were more likely to have postoperative fever and a postoperative body temperature ≥ 38.3°C.Adipose tissue is the primary site for storing and mobilizing lipids: it is also associated with endocrine and metabolic functions and contains multiple immune cells [30].Weisberg et al. demonstrated that the number of macrophages increases during obesity, and the expression of TNF-α and IL-6 are induced, which activate the in ammatory pathway [31].Similarly, the low levels of adiponectin in obese patients may increase postoperative insulin resistance and induce in ammation [32].Therefore, the increase in postoperative fever in patients with visceral obesity may be related to the greater release of these in ammatory factors, but this needs further study.
In this study, preoperative CT images were used to assess the incidence of postoperative complications.
Visceral obesity was de ned as > 93 cm 2 , which improved the applicability in clinical practice.Our study con rmed that BMI is a weak indicator for short-term surgical and recovery outcomes after cytoreductive surgery.However, this study also has some limitations.First, it was a single-center and retrospective study that produced inherent and unavoidable biases.Second, In China, due to the lack of rehabilitation institutions to assist in the management of postoperative patients, most patients need to recover in the hospital after surgery, which leading to a relatively long hospital stay.At the same time, the length of hospital stay is prolonged also because of the immediate postoperative chemotherapy.Meanwhile, it may help us observe mild complications.Last, the sample size was small.The number of samples needs to be increased to further prove that visceral obesity is an independent predictor of short-term complications of cytoreductive surgery for ovarian cancer.

Declarations
This Correlation between BMI and VFA.BMI, Body mass index.Visceral fat area.
The AUCs of VFA, SFA, TFA, and BMI

Table 4
Postoperative treatment and rehabilitation of Patients with high fever

Table 5
Univariate and Multivariate Logistic Regression Analysis study was supported by the National Natural Science Foundation(No.81872119, 82103286),Traditional Chinese Medicine Science and Technology Development Project in Jiangsu Province(ZD202014),Jiangsu Province Policy Guidance Program (International Science and Technology Cooperation/Hong Kong, Macao and Taiwan Science and Technology Cooperation) (BZ2020009), 333 High-level Talents Training Project in Jiangsu Province (BRA2019097).Innovative Team of Jiangsu Province(CXTDA2017008), Youth Scientist Fund of Jiangsu Basic Research Program (BK20210971), Natural Science Research Program of Higher Education Institutions in Jiangsu Province (20KJB320016), Scienti c Research Project of Jiangsu Province Maternal and Child Health Association (No.FYX201909)