Predictive Ability of C-Reactive Protein in Detecting Short-Term Complications After Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy: A Retrospective Cross-Sectional Study

Background Cytoreductive surgery (CRS) with hyperthermic intraperitoneal chemotherapy (HIPEC) is a potentially curative treatment for peritoneal carcinomatosis. Objective The aim of this study was to determine the predictive value of postoperative inflammatory biomarkers in assessing complications after CRS and HIPEC. Methods A prospective database of 181 patients, who underwent CRS-HIPEC between March 2014 through April 2018 in the Erasmus MC, was retrospectively analyzed. Postoperative complications were defined according to the serious adverse event (SAE) grading system. Levels of C-reactive protein (CRP) and white blood cell (WBC) count were compared between patients with SAE grade < 3 and SAE grade ≥ 3. The area under the receiver operating characteristic curve (AUC) was calculated for CRP and WBC against SAE ≥ 3 and various intra-abdominal complications. Results SAE ≥ 3 postoperative complications occurred in 50 patients. From the second until the fifth postoperative day (POD), CRP levels were significantly higher (p = 0.023, p < 0.001, p = 0.002, and p = 0.002, respectively) in these patients. CRP concentrations above 166 mg/L on POD3 (AUC 0.75) and 116 mg/L on POD4 (AUC 0.70) were associated with the highest risk of an SAE ≥ 3. Postoperative WBC levels were not significantly different between patients with SAE < 3 and SAE ≥ 3 complications. Conclusion Data from our hospital suggest that CRP levels that continue to rise after POD2 or that are ≥ 166 mg/L at POD3 or ≥ 116 mg/L at POD4, indicate a considerable risk for developing high-grade SAEs. The cut-off values we found can potentially be used as a threshold for additional diagnostic interventions, after they have been validated in external data.

(p = 0.023, p \ 0.001, p = 0.002, and p = 0.002, respectively) in these patients. CRP concentrations above 166 mg/L on POD3 (AUC 0.75) and 116 mg/L on POD4 (AUC 0.70) were associated with the highest risk of an SAE C 3. Postoperative WBC levels were not significantly different between patients with SAE \ 3 and SAE C 3 complications. Conclusion. Data from our hospital suggest that CRP levels that continue to rise after POD2 or that are C 166 mg/L at POD3 or C 116 mg/L at POD4, indicate a considerable risk for developing high-grade SAEs. The cutoff values we found can potentially be used as a threshold for additional diagnostic interventions, after they have been validated in external data.
Cytoreductive surgery combined with intraoperative hyperthermic intraperitoneal chemotherapy (CRS-HIPEC) has been considered a potentially curative therapeutic modality for patients presenting with peritoneal carcinomatosis (PC). [1][2][3] This extensive surgical treatment has been associated with improved survival outcomes for selected patients with PC from colorectal cancer (CRC) and pseudomyxoma peritonei (PMP), with 5-year survival rates of approximately 30% and 74% for CRC and PMP, respectively. [4][5][6] However, it has also been associated with considerable postoperative morbidity and mortality, with estimates of approximately 30% and 2-3%, respectively. [7][8][9] When attempting to reduce postoperative morbidity and mortality, early recognition of high-grade serious adverse events (SAEs) could be of great significance. C-reactive protein (CRP) is an acute-phase inflammation protein secreted primarily by liver hepatocytes, smooth muscle cells, and adipocytes, among others. 10 With its halflife being only 19 h and its increase being proportional to the degree of the inflammation process, CRP has established itself as an inexpensive, highly sensitive but nonspecific biomarker of systemic inflammatory response, 11,12 and has been identified as a potential predictive marker of postoperative complications after abdominal surgery. 13,14 Intra-abdominal complications, mainly septic complications or anastomotic leakage, are associated with mortality, reoperation, increased hospital stay, and higher costs. 15 Research on the utility and predictive value of biomarkers, such as CRP and WBC (white blood cell) levels, after CRS and HIPEC has been limited. 16 The aim of this cross-sectional retrospective study was to determine the predictive value of postoperative CRP and WBC levels in identifying complications after CRS and HIPEC in patients with PC from CRC or PMP.

Study Population
All patients with PC from CRC or PMP who underwent CRS-HIPEC in the Erasmus Medical Center between March 2014 through April 2018 were included in this study. A prospective database was built based on patients' chart review by using the electronic medical record system at this institution. Patients with recurrent peritoneal disease who underwent a second CRS-HIPEC procedure in the aforementioned time interval, were also included.

Perioperative Course
CRS-HIPEC procedures were performed by a specialized surgical team and in accordance with Dutch CRS and HIPEC protocols. 17 After abdominal access via laparotomy, a thorough assessment of the extent of peritoneal disease (only in cases with PC from colorectal and appendiceal cancer) was conducted to determine the Peritoneal Cancer Index (PCI) score according to Jacquet and Sugarbaker. 18 If the PCI score was under 20 and/or the specialized surgeons deemed the peritoneal disease resectable, the greater omentum, primary tumor (if still present), affected visceral abdominal organs, affected parietal surfaces, and all peritoneal implants were resected. Administration of HIPEC was by way of the open (coliseum) technique in which the abdominal cavity was filled with an iso-osmotic glucose/electrolyte dialysis (Dianeal Ò ) carrier solution, with either mitomycin-C or oxaliplatin being added to the perfusate as chemotherapeutic agent, once the desired abdominal temperature of [ 40°C was reached. After the HIPEC perfusion, intestinal bowel anastomoses and/or a stomy procedure was performed if necessary.
Postoperatively, patients were treated following standard of care for CRS-HIPEC procedures. Laboratory tests and diagnostic imaging modalities, such as computed tomography (CT) scans, were liberally used when deemed necessary. Postoperative complications were retrospectively classified according to the SAE grading system: SAE = 1 denotes an asymptomatic or mild complication (intervention not indicated); SAE = 2 denotes a moderate complication (local or non-invasive intervention indicated); SAE = 3 denotes a severe complication (significant but not immediately life-threatening, radiological or surgical intervention indicated); SAE = 4 denotes a lifethreatening complication (reoperation and/or prolonged intensive care unit [ICU] stay indicated); and SAE = 5 denotes in-hospital death related to the adverse event. 17,19 Intra-abdominal gastrointestinal complications included anastomotic leakage, bowel perforation or ischemia/ necrosis. The postoperative period was defined as the duration of the entire hospital stay following CRS and HIPEC, regardless of length.

Laboratory Data
Laboratory results (including postoperative biomarkers) of all patients who underwent CRS-HIPEC were recorded on arrival to the ICU and then daily during the patient's usually brief stay (1-3 days). When transferred to the ward, CRP levels were drawn in addition to a complete blood count (CBC), including white blood cell (WBC) count and blood chemistry in patients, usually three times a week (according to the Erasmus MC CRS-HIPEC protocol). CRP and WBC levels were routinely measured on postoperative days (PODs) 1, 2, 3, 4, and/or 5. Laboratory data were gathered retrospectively.

Statistical Analysis
Quantitative variables are presented as median with interquartile range (IQR), while categorical variables are presented as counts with percentages. Daily postoperative CRP values and WBC count between the SAE \ 3 and SAE C 3 groups were compared using the Mann-Whitney U-test (non-parametric). All tests were performed twosided and results were considered significantly different when the p value was \ 0.05. Diagnostic accuracy of CRP and WBC values on consecutive PODs was analyzed using the receiver operating characteristic (ROC) curve by calculating separate cut-off levels for CRP and WBC with optimal sensitivity and specificity. Outcomes assessed were intra-abdominal gastrointestinal complications, intra-abdominal abscess, and SAE C 3. Areas under the receiver operating characteristic curves (AUCs) were used to compare ROC curves. Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) version 21.0 (IBM Corporation, Armonk, NY, USA).

Ethical Considerations
All study procedures were performed according to the Erasmus MC Research Codes and with permission of the local Medical Ethics Review Committee (MEC-2018-1286).

DISCUSSION
The aim of this retrospective study was to determine the value of early postoperative inflammatory biomarker levels in identifying patients at risk of developing high-grade SAEs (SAE grade C 3) following CRS-HIPEC. To our knowledge, this is the first study to consider postoperative CRP and WBC levels as a diagnostic tool for identifying high-grade complications following CRS-HIPEC in patients with CRC, appendiceal cancer, and PMP. In cases of SAE grade \ 3 complications, CRP concentrations peaked on POD 2, and peaked at POD 3 in patients with  SAE C 3 complications. The differences in CRP concentrations were significant from POD 2 until POD 5 between the aforementioned two groups. Most (78%) postoperative SAE C 3 complications were diagnosed after peak CRP concentrations.
In this study, POD 3 and POD 4 were chosen as the time points for calculating AUCs, considering the compelling research 20 that suggests that postoperative CRP reaches a peak at POD 3 or 4, with better predictive accuracy than CRP on PODs 1 or 2. CRP values on POD 3 had moderate diagnostic accuracy (AUCs [ 0.70) for predicting SAE C 3, with cut-off values of 166 mg/L on POD 3 (sensitivity 61.1%; specificity 84.5%). Gans et al. 20 reported a similar CRP cut-off value of 159 mg/L on POD 3 (sensitivity 77%; specificity 77%) in a meta-analysis regarding 'major abdominal surgery'. The data observed in our tertiary center suggest that high CRP levels on PODs 3 and 4 indicate a considerable risk for developing high-grade  These aforementioned results suggest that CRP might be utilized to identify patients who are at high risk of developing postoperative SAE C 3 complications. In clinical practice, besides CRP, other variables such as heart rate, temperature, blood pressure, and urinary output are taken into account in decision making for further diagnostics or a reoperation. Taking the inexpensiveness of the laboratory test into account (less than €4 in The Netherlands), also makes this biomarker even more attractive for postoperative monitoring. The current study exclusively analyzed CRP levels and did not consider other clinical parameters. These clinical parameters can influence the pretest probability of developing high-grade SAEs, thereby improving the predictive value of CRP. Hence, if CRP levels are C 166 at POD 3, clinicians might pay better attention to other clinical parameters. Consequently, the early detection rate of high-grade SAEs might increase. Unfortunately, it was not possible to predict specific complications based on CRP levels alone. Although CRP levels were significantly increased in patients who developed intra-abdominal abscesses or gastrointestinal leakage, the predictive value of CRP for these specific complications was low. This can be explained first by the fact that CRP is a non-specific biomarker and therefore the predictive ability for specific complications is low. Second, the number of events per complication was too small for conclusive statistical analysis; thus, elevated CRP can increase caution for the high possibility that serious complications are evolving, but it cannot precisely predict what complication will develop. b Postoperative WBC count after CRS and HIPEC. POD postoperative day, CRP C-reactive protein, SAE serious adverse event, WBC white blood cell count WBC count did not differ significantly between the SAE \ 3 and SAE C 3 groups; however, a significant increase in WBC levels from POD 3 to POD 4 was observed in patients with gastrointestinal complications, with a corresponding ROC curve demonstrating moderate diagnostic accuracy. Nonetheless, WBC appears to be less useful, in general, than CRP for detecting high-grade SAEs. A possible explanation might be the low WBC levels due to extensive blood loss and dilution from intravenous fluid administration after CRS-HIPEC. Studies have also shown suppression of the cellular immune response after major surgery, trauma, or injury. 21,22 For CRS-HIPEC procedures specifically, mild leukopenia has been reported as a result of systemic uptake of intraperitoneal chemotherapy. 23,24 These reasons might explain why WBC does not seem to be a reliable predictor of early postoperative complications in patients undergoing CRS-HIPEC. This observation has been previously reported for colorectal surgery. 25 The current study found significant differences between the SAE \ 3 and SAE C 3 groups in relation to sex, number of cholecystectomies, and blood loss. In previously  published literature, the effect of sex on postoperative outcomes has been debated. Some earlier studies demonstrated that male patients have a higher risk of complications following colorectal surgery (open and laparoscopic), 26 and that higher rates of anastomotic leakages were associated with male sex. 27 However, these aforementioned associations with sex have not been demonstrated in other (retrospective) cohort studies of patients undergoing CRS and HIPEC. [28][29][30] In the current data, no explanation could be found for this observation. In the SAE C 3 group, significantly more cholecystectomies were performed during CRS. Cholecystectomy has not been earlier described as a risk factor for developing highgrade SAEs after CRS and HIPEC. As most of the cholecystectomies were performed in patients with PMPs, it is very likely that cholecystectomy is an indicator of the extent of disease spread, and thus the extent of surgery. The significant difference in the proportion of cholecystectomies between the SAE \ 3 and SAE C 3 groups may be explained via this underlying mechanism. Lastly, median blood loss was significantly higher in the SAE C 3 group. This observation was expected as extensive blood loss has been associated with postoperative morbidity in both general colorectal surgery and the CRS and HIPEC procedures. 26,31,32 Perioperative blood loss may, to some extent, reflect the extensiveness of the procedure; there will generally be more blood loss in larger procedures, which results in an increased risk of developing high-grade SAEs. [28][29][30] Limitations There are some limitations to the current study, including, first, the retrospective nature of data collection (including laboratory markers) and the limited study sample size, and, second, the amount of (possibly non-random) missing laboratory data, particularly on POD 4. This is explained by the postoperative HIPEC protocol in the Erasmus MC, which states that laboratory testing should be performed daily on the first, second, and third PODs, and afterwards three times per week on the ward. Third, only the 'early' (until POD 5) CRP levels were analyzed in this study, since most CRP values after POD 5 were more likely to be 'missing not at random': missingness related to a speedy recovery, and thus unnecessary laboratory testing and/or hospital discharge ('confounding by indication'). However, this observation was not considered an issue for this particular study, considering its aim was to evaluate CRP as a biomarker for early detection of SAE C 3 complications. In addition, Medina Fernandez et al. 16 suggested that CRP cut-off values might only be of value in the first postoperative week, as their results found CRP levels in the second postoperative week to be not significantly different between patients who developed infectious complications and those who did not, following CRS-HIPEC for ovarian PC.

CONCLUSION
With a cut-off value of 166 mg/L on POD 3 after CRS-HIPEC, CRP is a good screening test with high specificity in differentiating between SAE \ 3 and SAE C 3 complications. Following CRS-HIPEC, postoperative CRP levels might not only aid in patient selection to prevent overuse of imaging but also for earlier and safe hospital discharge. More prospective studies are needed to more accurately determine the predictive ability of early postoperative CRP levels, in combination with clinical parameters, after CRS-HIPEC. OPEN ACCESS This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.