Postoperative Ratio of C-Reactive Protein to Albumin as a Predictive Marker in Patients with Crohn's Disease Undergoing Bowel Resection

Background The ratio of C-reactive protein (CRP) to albumin (CAR) has a significant correlation with postoperative complications and acts as a predictor in patients with pancreatic cancer and colorectal cancer. However, whether the CAR can be used to predict complications in Crohn's disease (CD) patients after surgery has not yet been reported. Methods A total of 534 CD patients undergoing surgery between 2016 and 2020 were enrolled. The risk factors of postoperative complications were assessed by univariate and multivariate analyses. The cutoff values and the accuracy of diagnosis for the CAR and postoperative CRP levels were examined with receiver operating characteristic (ROC) curves. Results The rate of postoperative complications was 32.2%. The postoperative CAR (OR 13.200; 95% CI 6.501-26.803; P < 0.001) was a significant independent risk factor for complications. Compared with the CRP level on postoperative day 3, the CAR more accurately indicated postoperative complications in CD patients (AUC: 0.699 vs. 0.771; Youden index: 0.361 vs. 0.599). ROC curves showed that the cutoff value for the CAR was 3.25. Patients with a CAR ≥ 3.25 had more complications (P < 0.001), a longer postoperative stay (15.5 ± 0.6 d vs. 9.0 ± 0.2 d, P < 0.001), and more surgical site infections (48.2% vs. 5.7%, P < 0.001) than those with a CAR < 3.25. Conclusions Compared to the CRP level, the CAR can more accurately predict postoperative complications and can act as a predictive marker in CD patients after surgery.


Introduction
Drug-induced remission can often be achieved in Crohn's disease (CD) patients, but as the disease progresses, patients eventually must undergo surgical treatment [1,2]. The common postoperative complications in CD patients are postoperative bleeding, anastomotic leakage, abdominal abscess, intestinal obstruction, and short bowel syndrome, which can increase the treatment costs, prolong hospital stays, and reduce the long-term survival rate [3]. Therefore, the timely detection of postoperative outcomes is very important to improve the prognosis of CD patients.
Destruction of the intestinal structure leads to gradual worsening of the nutritional status of CD patients. Malnutri-tion and anemia contribute to an elevated incidence of postoperative complications [4,5]. At present, a variety of predictors have been proposed to predict outcomes after surgery, such as the C-reactive protein (CRP), procalcitonin (PCT), platelet-to-lymphocyte ratio (PLR), neutrophil to lymphocyte ratio (NLR), body mass index (BMI), sarcopenia, and albumin (ALB) levels [6][7][8][9][10]. However, most predictors are assessed before surgery, and they cannot reflect inflammation caused by surgical stress. The systemic inflammatory response after surgery obviously affects postoperative outcomes [11]. Additionally, most CD patients will undergo some type of preoperative optimization, such as nutrition therapy before surgery. Thus, the PLR, NLR, BMI, sarcopenia, and other predictors will be improved before surgery, and these predictors are always only related to nutritional status but ignore the surgical stress.
Many studies have confirmed that CRP can be used as an inflammatory index to reflect the degree of trauma and inflammatory state [12]. The CRP level on postoperative day (POD) 3 or 4 was reported to be the best predictor of postoperative complications [13,14]. In addition, the plasma level of ALB reflects not only the nutritional status of the body, but also inflammation from surgical trauma. Galata et al. [15] found that the preoperative ALB level was an independent predictor of major postoperative complications in CD patients after colorectal surgery. Ghoneima et al. [16] found that the preoperative CRP, haemoglobin (Hb), and ALB levels can act as predictors of septic complications in CD patients after surgery. Recently, some scholars have proposed that the CRP/ALB ratio (CAR) can predict postoperative complications in a timely manner in colorectal cancer, and its predictive value is better than that of CRP alone [17,18]. Thus, the postoperative CAR including ALB and CRP not only reflects nutritional status, but also is associated with the systemic inflammatory response after surgery. However, few studies have assessed the role of the CAR in predicting postoperative complications in CD patients.
In the current study, we investigated the relationship between the CAR and postoperative outcomes in CD patients and compared the diagnostic accuracy of the CAR with that of the postoperative CRP level.

Methods and Materials
2.1. Patients. The clinical records of consecutive CD patients undergoing surgery were collected and retrospectively reviewed. The inclusion criteria included are as follows: (1) CD was diagnosed according to the European Crohn's and Colitis Organization (ECCO) guidelines [19], and (2) bowel resection was performed. The exclusion criteria included (1) incomplete laboratory data, (2) multiple organ failure, (3) closing of the ileostomy or colostomy, (4) emergency surgery, and (5) ALB infusion before the operation or within 2 days after the operation. This study was approved by the ethics committee of the School of Medicine, Zhejiang University.

Data
Collection. The data included the patients' baseline characteristics (such as BMI and comorbidities), intraoperative data, and laboratory data (preoperative Hb level, preoperative ALB level, CRP level on POD1 and 3, ALB level on POD1, postoperative CAR, preoperative CRP level, preoperative erythrocyte sedimentation rate (ESR), preoperative white blood cell (WBC) count, preoperative red blood cell (RBC) count, preoperative platelet (PLT) count, and preoperative lymphocyte count) from the database. The CAR was defined as follows: CRP on POD1/ALB on POD1 × 100%.

Definition of Outcomes.
Our study focused on the relationship between the CAR and postoperative complications in CD patients. Postoperative complications were defined as those that occurred within 30 days after surgery or before hospital discharge according to the Clavien-Dindo classifica-tion system [20]. Mild complications included Clavien-Dindo grade I or II complications, while major complications were Clavien-Dindo grades III-IV complications. The postoperative stay and surgical site infections (SSIs) were also collected from the database retrospectively. SSIs included surface incisional infections and deep space infections.
2.4. Statistical Analysis. SPSS 22.0 was used to analyze all data. Continuous data are reported as means ± standard deviations or medians (interquartile ranges), whereas categorical variables are described as numbers (percentages). Continuous data were analyzed using Student's t-tests, while categorical variables were analyzed by Pearson's χ 2 test. The critical cutoff value for the CAR was calculated based on the ROC curve and Youden index. The potential independent risk factors for predicting postoperative outcomes were identified. Risk factors with a value of P < 0:1 were evaluated in the multivariate analysis. A P value <0.05 was considered statistically significant.

Discussion
This study showed that the CAR was a reliable and accurate indicator of postoperative outcomes in CD patients. The CAR on POD1 was a better predictor of postoperative complications than the CRP on POD3. Moreover, CD patients with a CAR ≥ 3:25 had more postoperative complications, longer postoperative stays, and more SSIs.
With the progression of the disease, many CD patients eventually require surgery [21,22]. Costa et al. [1] reported that 70% of CD patients eventually needed surgical intervention, and postoperative complications were very common. Therefore, it is very important to identify an accurate method   Many studies have confirmed that the inflammatory response after surgery is a risk factor for postoperative complications [23], including factors such as CRP, serum amyloid A, and IL-6 [24][25][26]. ALB is also considered to be an indicator of short-term and long-term postoperative outcomes in CD patients [27,28]. In addition, some newer predictive scores depend on inflammation, including the modified Glasgow Outcome Score (MGPS), NLR, and CAR [29][30][31][32], and can also be used to predict complications in patients undergoing colorectal surgery. Haruki et al. [33] suggested that the CAR was an independent risk factor of poor long-term outcomes of pancreatic resection. A meta-analysis by Wang et al. [34] showed that in patients with colorectal cancer, an increased CAR was associated with a poor outcome. They suggested that the CAR was a predictive factor that could be used to classify colorectal patients according to risk. Our results also demonstrated that the CAR could predict postoperative complications in CD patients.
CRP is an important index for evaluating the activity of CD, and it can also predict postoperative complications [35]. However, the postoperative complications in CD patients are not only related to inflammation but also closely related to nutritional status, which is reflected in the ALB level. Low ALB levels negatively affect the prediction of patient outcomes [36]. Therefore, for patients with CD, it is far from sufficient to use only the CRP level to predict their postoperative outcome, as indicated in a study by Easton and Balogh [37], who showed some drawbacks of using the CRP level. Hence, the CAR, which incorporates both the CRP and ALB levels, can predict the outcome of CD patients after bowel resection, and its accuracy is better than that of the CRP level according to the current study. To our knowledge, this study is the first to explore and compare the predictive ability of the CAR for postoperative complications in CD patients.
This result is not unexpected because the CRP level represents the degree of inflammation in patients, which is a risk factor for poor wound healing and infection and is related to a poor outcome [38][39][40]. The ALB level indicates the nutritional status of patients. Hypoproteinemia is associated with inflammation or previous malnutrition [41], which can lead to muscle atrophy and respiratory and immune dysfunction, thus prolonging the postoperative recovery time and increasing the incidence of postoperative complications in CD patients [42]. Acute stress can cause damage to vascular endothelial function, allowing ALB to move into the interstitial space, which causes tissue edema and insufficient perfusion, leading to a series of complications. Therefore, a higher CRP level and lower ALB level result in a higher Patients with a very high postoperative CAR should be intensively monitored for early detection of complications. Thus, these results have important implications for clinicians to optimally implement prophylactic strategies during the early postoperative period to improve outcomes in CD patients after bowel resection, including ALB infusion, prolongation of antibiotic administration, and other examinations to detect complications. Surgeons are advised to be aware of the CAR during the early postoperative period, even for patients with normal preoperative ALB and CRP levels.
Despite our interesting findings, there are still several limitations of the current study. First, this single-center study included a homogeneous cohort of patients who underwent surgery with a fixed surgical team. Second, the retrospective nature of this study meant that selection bias could have occurred. Therefore, prospective multicenter studies are warranted to confirm role of the CAR in predicting the short-term and long-term prognoses of CD patients after surgery.

Conclusions
Our results showed that the CAR, a new and feasible tool, has a significant correlation with postoperative complications and can serve as a predictive marker in CD patients undergoing bowel resection. Compared with the CRP level, the CAR is more accurate for the prediction of postoperative complications and could help clinicians evaluate the precise risk level and nutritional status of patients earlier. When patients have a CAR greater than 3.25, clinicians should be vigilant, continuously monitor the occurrence of postoperative complications, and provide timely interventions to improve their outcome.

Abbreviations
Hb: Haemoglobin ALB: Albumin CRP: C-reactive protein PLT: Platelet CAR: The CRP/ALB ratio CD: Crohn's disease PLR: Platelet-to-lymphocyte ratio NLR: Neutrophil to lymphocyte ratio BMI: Body mass index ESR: Erythrocyte sedimentation rate WBC: White blood cell RBC: Red blood cell SSIs: Surgical site infections.

Data Availability
The datasets were available from the corresponding author (gxlmed@zju.edu.cn).

Ethical Approval
This study was approved by the ethics committee of Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University.

Consent
Written informed consent was obtained from all participants.

Conflicts of Interest
The authors declare that they have no conflicts of interest.

Authors' Contributions
Xiaolong Ge and Hangfen Zhao contributed to the data analysis and manuscript writing. Huaying Liu and Wei Zhou contributed to data collection and study design. Weilin Qi, Wei Liu, Lingna Ye, Qian Cao, and Xianfa Wang were involved in data collection, data analysis, and manuscript editing. Hangfen Zhao and Huaying Liu contributed equally to this paper.