The Impact of the Surgical Apgar Score On Oncological Outcomes in Patients With Colorectal Cancer: A Single-Center Retrospective Cohort Study

Background: The surgical Apgar score (SAS) predicts postoperative complications (POCs) following gastrointestinal cancer surgery. Recently, the SAS was reported to be a predictor of not only POCs but also the prognosis. However, the impact of the SAS on oncological outcomes in patients with colorectal cancer (CRC) has not been fully examined. The present study therefore explored the oncological signicance of the SAS in patients with CRC. Methods: We retrospectively analyzed 639 patients who underwent radical surgery for CRC. The SAS was calculated based on three intraoperative parameters: estimated blood loss, lowest mean arterial pressure and lowest heart rate. The optimal cut-off value of the SAS was determined by receiver operating characteristic curves. All patients were classied into 2 groups based on the SAS ( ≤ 6 and >6). The association of the SAS with the recurrence-free survival (RFS), overall survival (OS) and cancer-specic survival (CSS) was analyzed. Results: Univariate analyses revealed that a lower SAS ( ≤ 6) was signicantly associated with a worse RFS, OS and CSS. A multivariate analysis revealed that age ≥ 75 years old, Charlson comorbidity index ≥ 1, ASA-Physical Status ≥ 3, SAS ≤ 6, histologically undifferentiated tumor type and an advanced pStage were independent factors for the OS, and an SAS ≤ 6 and advanced pStage were independent factors for the CSS. Conclusions: A lower SAS ( ≤ 6) was an independent prognostic factor for not only the OS but also the CSS in patients with CRC, suggesting that the SAS might be a useful biomarker predicting oncological outcomes in patients with CRC. survival; CSS, Cancer-specic survival; GIST, gastrointestinal stromal tumor; BMI, body mass index; GPS, Glasgow prognostic score; CRP, C-reactive protein; ASA-PS, American Society of Anesthesiologists classication of physical status; pT, Pathological T; pN, pathological N; pStage, Pathological TNM stage; CCI, Charlson comorbidity index; ROC, Receiver operating characteristic; CDC, Clavien-Dindo classication; JSCCR, Japanese Society for Cancer of the Colon and Rectum; CT, Computed tomography, FOLFOX, 5-uorouracil/leucovorin plus oxaliplatin; CapeOX, Capecitabine plus oxaliplatin; HR, Hazard ratio; CI, Condence interval; 5-FU, 5-uorouracil; LV, Leucovorin.


Introduction
Colorectal cancer (CRC) was estimated to account for more than 1.9 million new colorectal cancer cases and 935,000 deaths in 2020, ranking third in terms of incidence but second in terms of mortality globally [1]. Although surgical resection is standard treatment for local and regional CRC worldwide, the mortality from CRC remains unsatisfactory.
Notably, among patients who undergo curative surgery for CRC, approximately one-third will develop disease recurrence, underscoring the importance of developing biomarkers to identify patients who may require postoperative intensi cation of treatment [2]. Postoperative complications (POCs) are reportedly signi cantly associated with a poor prognosis in CRC [3]. Therefore, predicting and preventing POCs might be one way to increase the survival in CRC.
The surgical Apgar score (SAS) system was developed by Gawande et al. to predict POCs in general surgery in 2007 [4]. The SAS consists of three intraoperative parameters: the estimated blood loss (EBL), lowest mean arterial pressure (LMAP) and lowest heart rate (LHR). The SAS has been validated as a predictor of POCs in CRC surgeries [5]. Previously, we reported that the SAS was a valuable predictor of severe complications after CRC surgery in elderly patients [6]. One reason why the SAS is able to predict POCs is that it re ects the intraoperative hemodynamic stability in patients with gastrointestinal cancer. Recent studies have highlighted the signi cant impact of the SAS on not only POCs but also the overall survival (OS) in gastrointestinal cancer [7,8]. However, the impact of the SAS on oncological outcomes in patients with CRC has not been fully examined.
We hypothesized that the SAS, which re ects intraoperative hemodynamics, would affect not only the OS but also the oncological long-term outcomes, such as the recurrence-free survival (RFS) and cancer-speci c survival (CSS), in CRC patients. The present study therefore assessed the impact of the SAS on oncological outcomes after radical surgery in CRC patients.

Patients
We retrospectively analyzed consecutive patients who underwent radical surgery under general anesthesia for CRC at the Department of Gastroenterological Surgery, Osaka City University Hospital from January 2008 to December 2014. We excluded patients with pathological Stage 0 or , non-curative (R1 or R2) resection, preoperative treatment (chemotherapy and/or radiotherapy), synchronous surgeries for other cancers and histologically atypical tumors, such as squamous cell carcinoma, small-cell carcinoma, gastrointestinal stromal tumor (GIST) or melanoma. The following clinical and surgical data were collected from electronic medical records: age, gender, body mass index (BMI), the presence of current smoking, the Glasgow prognostic score (GPS) based on preoperative serum albumin and C-reactive protein (CRP) levels [9], the American Society of Anesthesiologists classi cation of physical status (ASA-PS), tumor location (colon and rectum), pathological T (pT) stage, pathological N (pN) stage, pathological TNM stage (pStage), histological tumor type (differentiated type; well-or moderately differentiated adenocarcinoma and undifferentiated type; poorly differentiated and mucinous adenocarcinoma), operative procedure (laparoscopy and open surgery), operation time, intraoperative EBL, transfusion, intraoperative LMAP and intraoperative LHR. Comorbidities were evaluated according to the Charlson comorbidity index (CCI) [10]. The pathological TNM stage was determined based on the 8 th edition of the Union for International Cancer Control TNM classi cation of malignant tumors [11].

SAS
We used the original the SAS scoring system to calculate the SAS [4]. The three intraoperative SAS parameters (EBL, LMAP and LHR) were extracted from electronic anesthesia records. The score is the sum of the points from each category ( Table 1). The cut-off value of the SAS was determined as the point on the receiver operating characteristic (ROC) curve predicting severe POCs, de ned as grade ≥III according to the Clavien-Dindo classi cation (CDC) [12], at which the Youden index was maximal. All patients were classi ed into one of two groups based on this cut-off value.

Treatment strategy
Our treatment strategy for CRC is based on the Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines [13]. All patients underwent various radiological tests for the preoperative diagnosis and staging, such as colonoscopy and contrast-enhanced computed tomography (CT). Radical surgery was de ned as no residual tumor cells microscopically at the stump of the surgical specimen with an adequate surgical margin. General anesthesia was mainly performed by intravenous anesthesia, and the anesthesiologists were involved in the anesthesia management of all cases. Adjuvant chemotherapy was performed for patients with pathological stage II/III disease. Patients received monotherapy using an oral pro-drug based on 5-FU, such as capecitabine or combination therapy with 5-FU and oxaliplatin, such as 5-uorouracil/leucovorin plus oxaliplatin (FOLFOX) or capecitabine plus oxaliplatin (CapeOX).

POCs and the prognosis
Severe POCs were de ned as grade ≥III according to the CDC that developed within 30 days after surgery. The prognosis was analyzed based on the information in the electronic medical record. Patients were followed up at intervals of three to six months until the end of this study or death. The OS, RFS and CSS were calculated from the start date of the operation to the date of last follow-up or death, to the con rmed date of recurrence or death and to the date of last follow-up or death due to CRC, respectively.

Statistical analyses
The data of continuous variables are presented as the median (interquartile range [IQR]). The cut-off value of the SAS was calculated by the ROC curve for severe POCs. A univariate analysis was performed by the Mann-Whitney U test for continuous variables and by the chi-squared test for categorical variables. The survival probabilities (the OS, RFS and CSS) were calculated by Kaplan-Meier survival curves and statistically compared by the log-rank test. Univariate and multivariate analyses using the Cox proportional hazard model were performed to identify signi cant prognostic factors for the OS and CSS. Hazard ratios (HRs) and 95% con dence intervals (CIs) were calculated. Values of p<0.05 were considered signi cant. All data analyses were conducted using the JMP ® 13 software program (SAS Institute Inc., Cary, NC, USA).

Ethics
The Ethics Committee at Osaka City University approved this retrospective study of clinical data study, which was conducted in according with the principles of the Declaration of Helsinki.

Results
Patients' characteristics A total of 639 (colon cancer in 460 cases and rectal cancer in 179 cases) patients were enrolled in this study. Severe POCs of CDC grade ≥III were observed in 102 patients (16.0%). According to the ROC curve analysis, patients were divided into two groups based on the cut-off value of the SAS. The patients with an SAS ≤6 (n=190, 29.7%) were assigned to the low-SAS group, and those with an SAS ≥7 (n=449, 70.3%) were assigned to the high-SAS group.

Postoperative outcomes
The low-SAS group more frequently included patients with severe POCs (CDC grade ≥III) (p<0.001) and who had a signi cantly longer postoperative stay (p<0.001) than the high-SAS group ( Table 2).

The prognosis
The median follow-up time was 63.4 (IQR, 54.8−83.0) months for all patients. Recurrence was observed in 96 cases (15.0%). Death due to CRC was observed in 61 cases (9.5%). A total of 142 deaths (22.2%) were observed. The 5-year OS, RFS and CSS rates for the entire study population were 82.4%, 86.1% and 91.8%, respectively. Kaplan-Meier survival curves comparing the OS, RFS and CSS between the two groups are shown in Figure 1A-C. The OS, RFS and CSS rates in the low-SAS group were signi cantly lower than those in the high-SAS group (p<0.001, p=0.003, and p<0.001, respectively).

Univariate and multivariate analyses for the OS and CSS
The results of univariate and multivariate analyses for the OS and CSS are summarized in Tables Figure 2A, 2B. The OS rates in the low-SAS group were signi cantly lower than those in the high-SAS group among the patients with and without severe POCs (p=0.02 and p=0.016 respectively). A subgroup analysis according to the pStage (I, II and III) was also conducted. The Kaplan-Meier survival curves comparing the OS based on the SAS in patients with pStage I, II and III disease are shown in Figure 3A−C. The OS rates in the low-SAS group were signi cantly lower than those in the high-SAS group among patients with pStage II and III disease (p=0.048 and p=0.016 respectively), while no signi cant difference was seen among the patients with pStage I disease (p=0.172).

Discussion
In this study, we evaluated the SAS in 639 patients who underwent radical surgery for CRC. We identi ed a lower SAS (≤6) as an independent prognostic factor for the OS and CSS. Nakagawa et al. previously reported that the SAS predicted not only POCs but also the OS in esophageal cancer patients [7], and Yamada et al. reported that the SAS predicted the OS in gastric cancer patients [8]. However, the association between the SAS and oncological outcomes in CRC patients has been unclear. To our knowledge, this is the rst study to clarify the impact of the SAS on the RFS and CSS in CRC patients. Our results suggested that the SAS might be a useful biomarker predicting oncological outcomes after radical surgery in CRC patients.
In this study, an older age (≥75), a higher CCI (≥1), a higher ASA-PS (≥3), a lower SAS (≤6), histologically undifferentiated tumor type and advanced pStage (≥III) were identi ed as independent factors for the OS. Our results were consistent with those of previous studies [14][15][16]. However, the impact of SAS on the OS has not been fully examined in CRC. An explanation concerning the correlation of the SAS with the OS has been considered. First, the SAS, consists of EBL, LMAP and LHR, re ects intraoperative hemodynamics. Previous studied reported that signi cant blood loss, intraoperative hypotension and a higher heart rate were associated with a poor prognosis in CRC [17][18][19]. These studies further indicated that hemodynamic instability might affect the survival in CRC. Second, the SAS re ects surgical stress, as signi cant blood loss, a large incision and prolonged operation time result in a low SAS. In the present study, a lower SAS was more frequent in patients with more blood loss, open surgery and a longer operation time. Our results were consistent with those of the previous study [20]. Finally, a low SAS was associated with POCs. POCs affect the prognosis in CRC because of marked postoperative in ammation and a poor immunological status [21,22]. In the present study, a lower SAS was signi cantly associated with severe POCs. However, regardless of POCs, a lower SAS was signi cantly associated with a poor OS. Our ndings therefore suggest that the SAS might be a useful prognostic marker either with or without POCs in CRC patients.
The oncological signi cance of the SAS has been poorly documented in CRC patients. A large amount of intraoperative blood loss has been reported to be associated with both tumor cell spillage and immunosuppression, thus leading to cancer recurrence [17]. In addition, a poor intravascular blood ow induces the arrest, adhesion and extravasation of circulating tumor cells preceding metastasis [23]. Furthermore, cancer progression exacerbates the cardiac function [24]. Tumors induce cardiac atrophy and dysfunction through the release of proin ammatory cytokines [25]. In the present study, a lower SAS was signi cantly associated with an advanced pT, pN and pStage. A lower SAS was signi cantly associated with a worse RFS and CSS. In particular, a lower SAS was an independent factor for the CSS. These ndings suggest that the SAS might be a biomarker re ecting not only the intraoperative hemodynamics but also cancer progression in CRC patients.
Postoperative adjuvant chemotherapy using doublet therapy of 5-uorouracil (5-FU) and folic acid (leucovorin, LV) or capecitabine with oxaliplatin (FOLFOX or CapeOX) has been widely considered the standard treatment for patients with stage III CRC after curative resection [26,27]. However, 20%−30% of patients with stage III CRC develop recurrence despite receiving adjuvant chemotherapy [28]. This indicates that there remains room for improvement in the outcomes of such patients. Risk factors for recurrence that can help determine the regimen and duration of adjuvant chemotherapy have not been fully validated. In the present study, a subgroup analysis showed that a lower SAS was signi cantly associated with a worse OS in patients with pStage II and III CRC. These ndings suggest that the SAS might be a prognostic biomarker, regardless of the Stage, and may be useful for determining the indication and regimen of adjuvant chemotherapy in CRC patients.
Several limitations associated with the present study warrant mention. First, this study was a retrospective study conducted at a single institution and included patients who underwent both laparoscopic and open surgery, which might have contributed to selection bias. Second, data on anesthesia management, such as the volume of infusions, sedatives and analgesics, were insu cient. Finally, the optimal SAS cut-off value has not yet been determined. The cut-off value in the present study was determined by ROC curve analyses for severe POCs.

Conclusion
A lower SAS (≤6) was an independent prognostic factor for the OS and CSS after radical surgery in CRC patients. Our results suggest that the SAS might be a useful biomarker predicting oncological outcomes in CRC.

Declarations
Ethics approval and consent to participate The Ethics Committee at Osaka City University approved this retrospective study of clinical data study, which was conducted in according with the principles of the Declaration of Helsinki.

Consent for publication
Informed consent was obtained from all individual participants included in the study.

Availability of data and materials
The datasets generated during and/or analyzed during the current study are not publicly available due to hospital regulations.

Competing interests
The authors have no con icts of interest to declare.

Funding
There are no resources of funding to be reported or declared.
Authors' contributions AS and TF contributed signi cantly to the study design, data analysis, and drafted the manuscript. HN, MS, YI, MS and YO participated in data collection and assisted with data interpretation. KM and MO critically reviewed and revised the manuscript. All authors read and approved the nal manuscript.  The prognosis based on the surgical Apgar score (SAS). A: The overall survival (OS). The 5-year OS rates in the low-SAS group (≤6) were signi cantly lower than those in the high-SAS group (≥7) (p <0.001). B: The recurrence-free survival (RFS). The 5-year RFS rates in the low-SAS group (≤6) were signi cantly lower than those in the high-SAS group (≥7) (p=0.003). C: The cancer-speci c survival (CSS). The 5-year CSS rates in the low-SAS group (≤6) were signi cantly lower than those in the high-SAS group (≥7) (Kaplan-Meier; p <0.001).

Figure 2
The overall survival (OS) in patients with or without severe complications. A: The OS in 537 patients without severe complications. The OS rates in the low-SAS group (≤6) were signi cantly lower than that in the high-SAS group (≥7) (p=0.016). B: The OS in 102 patients with severe complications. The OS rates in the low-SAS group (≤6) were signi cantly lower than that in the high-SAS group (≥7) (p=0.02).