Early prediction of postoperative infection using inflammatory markers after cytoreductive surgery for peritoneal carcinomatosis

Abstract

BACKGROUND

Major postoperative complications have proved to be an independent adverse prognostic factor for long-term survival in patients undergoing cytoreductive surgery (CRS) with or without hyperthermic intraperitoneal chemotherapy (HIPEC). C-reactive protein (CRP) is an inflammatory marker that is reportedly a useful tool for the early prediction of postoperative complications, as is the neutrophil-to-lymphocyte ratio (NLR). In patients with peritoneal carcinomatosis, postoperative CRP levels on days 2 to 4 are predictors of early complications after CRS plus HIPEC.

AIM

To determine the usefulness of CRP and NLR for the early detection of overall postoperative infections (OPIs) after CRS +/- HIPEC.

METHODS

Patients treated on a peritoneal carcinomatosis program at a tertiary care hospital, in whom complete or optimal cytoreduction was achieved, were analyzed retrospectively. A total of 111 patients were included in this study. CRP and NRL values prior to surgery and during the first four postoperative days (PODs) were recorded, along with immunonutrition intake. Their association with OPI and intra-abdominal infections during the first week after surgery was evaluated.

RESULTS

Of the 111 patients included, 19 presented OPI and 8 intra-abdominal infections. Patients with infections had a higher number of digestive anastomoses than those without (1 vs 0.5, P = 0.053 and 1.2 vs 0.6, P = 0.049) and longer length of stay (19 vs 14.9 days, P = 0.022 and 22.3 vs 15.1 days, P = 0.006). CRP values above 118 mg/L on POD3 yielded a sensitivity of 66.7% and a specificity of 74.2% to detect OPI. No differences in NLR values were observed. Patients with immunonutrition intake had higher CRP levels regardless of whether they presented OPI. Subsequently, on POD3 and POD4, patients with OPI presented with higher levels of CRP than patients without infection, regardless of the immunonutrition intake.

CONCLUSION

CRP levels are useful to detect early OPI in patients with peritoneal carcinomatosis undergoing CRS. A cut-off value of 118 mg/L on POD3 yields the best sensitivity and specificity.

Key Words: Peritoneal metastasis; Peritoneal carcinomatosis; Cytoreductive surgery; Hypherthermic intraperitoneal chemotherapy; C-reactive protein; Neutrophil-to-lymphocyte ratio; Postoperative complications

Core Tip: In this study, inflammatory markers C-reactive protein (CRP) and neutrophil-to-lymphocyte ratio were evaluated retrospectively in patients undergoing cytoreductive surgery and hyperthermic intraperitoneal chemotherapy to establish the association of theses markers with overall postoperative infections (OPIs). The effect of preoperative immunonutrition intake on these markers was also analyzed. The main finding of the study was the difference in CRP kinetics between patients administered or not administered preoperative immunonutrition. We also conclude that tracking CRP kinetics may be more helpful for predicting OPIs than establishing a specific cut-off point. By contrast, neutrophil-to-lymphocyte ratio levels did not accurately predict OPI in these patients.

INTRODUCTION

In the last two decades, cytoreductive surgery (CRS) with or without hyperthermic intraperitoneal chemotherapy (HIPEC) has been the only treatment able to improve long-term survival in patients with peritoneal metastasis originating in the mesothelium or from tumors in the appendix, colon, and ovary[1-3]. Currently, with the improvements in the technique and its consolidation as the treatment of choice at high-volume centers, the associated postoperative mortality and morbidity rates have fallen to 1%-6% and 12%-34%, respectively[4]. These figures are comparable to those recorded in other major oncologic abdominal procedures such as hepatectomy, pancreatoduodenectomy, and esophagectomy[3,5-7].

Major postoperative complications have proven to be independent adverse prognostic factors for long-term survival in patients undergoing CRS[7-10]. The mechanism for their development remains uncertain, but surgical stress and systemic inflammatory response may play an important role by impairing cell-mediated immunity and natural killer cell function[11,12]. This immune system dysfunction may impair the control of microscopic residual disease after surgery[13,14]. Therefore, postoperative complications may increase tumor recurrence and worsen long-term survival, and in fact these effects have been observed in patients with esophageal, colorectal, hepatic and pancreatic neoplasms undergoing surgical resection[15-18].

Inflammation can result in immunological dysfunction, potentially increasing the risk of complications after surgery[16,19]. Furthermore, preoperative inflammation is known to shorten long-term survival in these patients[20,21]. Thus, reducing preoperative inflammation might improve postoperative outcomes and cancer prognosis[22]. C-reactive protein (CRP) is an inflammatory marker that has been reported to be a useful tool for the early prediction of postoperative complications[23-25]. In patients with peritoneal carcinomatosis, postoperative CRP levels on days 2 to 4 are predictors of early complications after CRS plus HIPEC[26-28]. In addition, rising CRP concentrations have been associated with poorer survival outcome in patients with cancer[29,30]. For its part, the neutrophil-to-lymphocyte ratio (NLR) has also been used as a predictive marker of postoperative complications in colorectal and esophagogastric surgery[30-32].

Immunonutrition formulas enriched with arginine, glutamine, omega 3 and nucleotides, among other nutrients, may reduce the postoperative inflammatory response and boost lymphocyte production and function[33]. Its effect on major digestive oncologic surgery has been studied in several randomized trials and meta-analyses, which have shown reductions in postoperative infectious complications and hospital length of stay[34-36], and in another study, in overall postoperative complications[37]. Recently, a randomized control trial assessing immunonutrition in patients undergoing CRS and HIPEC[38], showed that those with perioperative immunonutrition intake had shorter hospitalization, fewer wound infections and fewer major postoperative complications than those receiving standard nutrition, although the differences were not statistically significant. The aim of our study was to evaluate CRP and NLR levels in relation to the occurrence of postoperative overall and intra-abdominal infections (IAIs) in patients following CRS with or without HIPEC.

MATERIALS AND METHODS

Design and setting

A retrospective cohort study was conducted of all patients with peritoneal metastases undergoing a complete or optimal cytoreduction after CRS with or without HIPEC, between November 2014 and December 2020 at the Peritoneal Carcinomatosis Unit of the University of Elche General Hospital. Written informed consent was obtained prior to surgery and the study was approved by our institution’s Ethics Committee (Approval No. PI 21/2018).

Patient data

Demographic variables obtained included age, sex, American Society of Anaesthesiologists classification, carcinomatosis origin, neoadjuvant chemotherapy and immunonutrition intake. Two groups of patients were established on the basis of whether or not they had received immunonutrition. A complete blood test including blood count with NLR, CRP levels and albuminemia was obtained prior to surgery and was repeated during the early postoperative period.

Perioperative course

Prescription of immunonutrition with an Atempero^® 200 mL hyperproteic shake containing immunonutrients such as L-arginine, omega-3 fatty acids, and nucleotides was introduced in October 2018. Prior to this date, patients did not receive immunonutrition as standard practice. Immunonutrition was administered twice a day for 7 days prior to surgery. A high protein diet was also recommended before surgery. Antibiotic and antithrombotic prophylaxis was administered in accordance with the protocol.

The peritoneal tumor burden was quantified according to the peritoneal carcinomatosis index, and cytoreductive completeness was scored at the end of the surgical procedure. Drugs used during the HIPEC procedure were mitomycin-C 35 mg/m² for 60 minutes and oxaliplatin in patients with peritoneal metastasis of colorectal origin or peritoneal pseudomyxoma. Oxaliplatin was only administered to 8 patients, and was not used after 2016. Paclitaxel 60 mg/m² was administered for 60 minutes in women with ovarian peritoneal metastasis. HIPEC was initially delivered with an open-coliseum technique, but in January 2018 a closed technique was introduced (PRS Combat, Galmaz Biothech, Madrid, Spain).

The primary outcome was the accuracy of CRP and NLR for detecting overall postoperative infection (OPI), and the secondary outcome was their ability to detect IAI, both diagnosed in the first seven postoperative days (PODs). OPI included surgical site, urinary tract and catheter-related infections, and pneumonia. IAI included anastomotic leakage confirmed by computed tomography (CT) or at reoperation, the presence of pus or enteric contents within the drains, and an abdominal or pelvic collection percutaneously drained with a positive culture. Adverse events were recorded according to the Clavien-Dindo classification[39]. Length of stay was also recorded.

Follow-up

All patients were seen by surgeons at the outpatient clinic within 1 month of discharge. Then, follow-up visits were carried out by surgeons and oncologists every 3 months during the first 2 years after surgery, then every 6 months until 5 years after surgery, and once a year thereafter. The follow-up visit consisted of physical examination, measurement of serum tumor markers and a thoraco-abdominal CT scan.

Statistical analysis

Continuous variables are described as medians and interquartile ranges. Categorical variables are described as frequencies and percentages. Levene’s test was used to assess the equality of variances, the Mann-Whitney U test and Student’s t test to compare medians, and the χ² test to compare proportions. The threshold for statistical significance was set at P < 0.05. To assess the evolution of CRP and NLR values in patients with and without OPI and IAI, a linear general model (repeated measures analysis of variance) was used. A non-parametric receiver operating characteristic curve was performed to describe the sensitivity and specificity of CRP as a predictor of overall and IAI and to determine the most discriminating cut-off value.

RESULTS

A total of 111 patients were included, 82 women (73.9%) and 29 men (26.1%). Baseline and surgical characteristics are described in Table 1. Immunonutrition was administered to 49 patients (44.1%). Median peritoneal carcinomatosis index was 10 and complete cytoreduction was achieved in 87.4% of patients. Nineteen patients (17.1%) had OPI, and eight (7.2%) presented IAI. Patients with OPI had a higher number of digestive anastomoses (1 vs 0.5, P = 0.053) and a longer length of stay (19 vs 14.9 days, P = 0.022) than patients without OPI. Similar findings were observed among patients with IAI compared with those without IAI (number of digestive anastomoses 1.2 vs 0.6, P = 0.049, in hospital stay 22.3 vs 15.1 days, P = 0.006). No differences were found in reintervention and mortality rates in either the OPI or the IAI group.

Table 1 Baseline and surgical characteristics according to morbidity of all 111 patients in the study, n (%).

Characteristic	All patients (n = 111)	Overall postoperative infection			Intra-abdominal infection
		No (n = 92)	Yes (n = 19)	P value	No (n = 103)	Yes (n = 8)	P value
Age (years), mean ± SD	60.7 ± 10.3	60.7 ± 10.5	60.8 ± 9.6	0.957	60.7 ± 10.3	60.0 ± 11.9	0.849
Sex
Men	29 (26.1)	24 (82.8)	5 (17.2)	0.984	26 (89.7)	3 (10.3)	-
Women	82 (73.9)	68 (82.9)	14 (17.1)	-	77 (93.9)	5 (6.1)	0.447
ASA score, mean ± SD	2.2 ± 0.6	2.2 ± 0.6	2.4 ± 0.6	0.262	2.2 ± 0.6	2.4 ± 0.7	0.473
Neoadjuvant chemotherapy
No	51 (45.0)	45 (88.2)	6 (11.8)	0.167	48 (94.1)	3 (5.9)	0.619
Yes	60 (54.1)	47 (78.3)	13 (21.7)	-	55 (91.7)	5 (8.3)	-
Immunonutrition
No	62 (55.9)	53 (85.5)	9 (14.5)	0.413	59 (95.2)	3 (4.8)	0.278
Yes	49 (44.1)	39 (79.6)	10 (20.4)	-	44 (89.8)	5 (10.2)	-
Carcinomatosis origin
Ovarian	55 (49.5)	48 (87.3)	7 (12.7)	-	54 (98.2)	1 (1.8)	-
Colorectal	34 (30.6)	24 (70.6)	10 (29.4)	-	28 (82.4)	6 (17.7)	-
Pseudomyxoma	13 (11.7)	13 (100.0)	0 (0.0)	-	13 (100.0)	0 (0.0)	-
Gastric	2 (1.8)	1 (50.0)	1 (50.0)	1	2 (100.0)	0 (0.0)	1
Endometrial	2 (1.8)	2 (100.0)	0 (0.0)	-	2 (100.0)	0 (0.0)	-
Mesothelioma	1 (0.9)	1 (100.0)	0 (0.0)	-	1 (100.0)	0 (0.0)	-
Primary	1 (0.9)	1 (100.0)	0 (0.0)	-	1 (100.0)	0 (0.0)	-
Neuroendocrinal	1 (0.9)	1 (100.0)	0 (0.0)	-	1 (100.0)	0 (0.0)	-
Others	2 (1.8)	1 (50.0)	1 (50.0)	-	1 (50.0)	1 (50.0)	-
PCI, median ± SD	10.4 ± 6.5	10.5 ± 6.7	9.5 ± 5.4	0.520	10.4 ± 6.6	10.0 ± 5.3	0.874
Complete cytoreduction	97 (87.4)	78 (80.4)	19 (19.6)	0.069	89 (91.8)	8 (8.3)	0.265
Number of visceral resections, mean ± SD	1.6 ± 1.2	1.5 ± 1.2	2.0 ± 1.2	0.161	1.6 ± 1.2	2.0 ± 1.1	0.324
Number of digestive anastomoses, mean ± SD	0.6 ± 0.6	0.5 ± 0.7	1.0 ± 0.7	0.053	0.6 ± 0.6	1.2 ± 0.5	0.049
HIPEC
No	23 (20.7)	17 (73.9)	6 (26.1)	0.200	20 (87.0)	3 (13.0)	0.224
Yes	88 (79.3)	75 (85.2)	13 (14.8)	-	83 (94.3)	5 (5.7)	-
In hospital stay, median ± SD	15.6 ± 7.2	14.9 ± 7.1	19.0 ± 6.6	0.022	15.1 ± 7.0	22.3 ± 6.3	0.006
Reintervention	11 (9.9)	5 (45.5)	6 (54.5)	0.000	7 (63.6)	4 (36.4)	0.000
Mortality	2 (1.8)	1 (50.0)	1 (50.0)	0.213	2 (100.0)	0 (0.0)	0.691

¹The test has been omitted due to the small number of cases in the categories.

ASA: American Society of Anaesthesiologists; HIPEC: Hyperthermic intraperitoneal chemotherapy; PCI: Peritoneal carcinomatosis index; SD: Standard deviation.

CRP

Daily CRP levels on POD1 to POD4 were analyzed in relation to the presence or absence of OPI or IAI (Table 2). CRP levels peaked in both groups on POD2 and then fell on POD3 and POD4 in the IAI group, but not in the OPI group (Figure 1A). Patients with OPI had higher CRP levels than those without OPI on all PODs, the difference being significant on POD2 (87.0 vs 145.0, P = 0.013), POD3 (70.4 vs 153.5, P = 0.001), and POD4 (80.5 vs 151.5, P = 0.047). Patients with IAI showed higher CRP levels on POD2, POD3, and POD4 than patients who did not, but the difference was not significant. The area under the receiver operating characteristic curve (AUC) of CRP values on the first four PODs are recorded in Table 3. AUCs were significantly higher on POD2, POD3, and POD4 in patients with OPI compared to those without. The highest AUC ratio was obtained on POD3 in patients with OPI: 0.7 (95% confidence interval [CI]: 0.6-0.9, P = 0.005) with an optimal cut-off value of 118 mg/L, yielding a sensitivity of 66.7% and a specificity of 74.2%.

Figure 1 Daily median biomarker levels in post. A: Daily median levels of C-reactive protein (CRP) and neutrophil-to-lymphocyte ratio in patients with and without overall postoperative infection; B: Daily median levels of CRP in patients with and without overall postoperative infection according to immunonutrition (IMN) intake.

Table 2 Daily levels of C-reactive protein and neutrophil-to-lymphocyte ratio in all patients according to the presence or absence of overall postoperative infection and in the presence or absence of intra-abdominal infection.

Characteristic	All patients (n = 111)	Overall postoperative infection			Intra-abdominal infection
		No (n = 92)	Yes (n = 19)	P value	No (n = 103)	Yes (n = 8)	P value
Day 1
CRP (mg/L)	79.5 (58.0-102.0)	77.0 (58.0-101.0)	91.0 (51.0-119.0)	0.390	77.5 (58.0-101.3)	89.5 (73.0-165.5)	0.114
NLR (mg/L)	10.4 (6.2-16.2)	10.2 (6.1-15.8)	10.4 (6.6-23.4)	0.682	9.9 (6.2-15.2)	15.9 (4.0-28.0)	0.228
Day 2
CRP (mg/L)	91.5 (65.3-147.0)	87.0 (60.0-127.0)	145.0 (81.0-187.0)	0.013	88.1 (62.3-143.5)	143.0 (76.5-195.5)	0.195
NLR (mg/L)	10.4 (7.2-14.5)	10.4 (7.1-14.3)	9.6 (7.2-15.6)	0.377	10.3 (7.2-14.3)	13.3 (5.9-17.7)	0.707
Day 3
CRP (mg/L)	75.0 (47.0-140.0)	70.4 (44.5-120.5)	153.5 (100.3-240.3)	0.001	73.0 (47.0-136.0)	119.0 (38.0-167.8)	0.442
NLR (mg/L)	10.0 (7.0-12.2)	10.5 (7.1-12.1)	8.4 (6.1-17.1)	0.674	10.0 (7.0-12.1)	9.9 (4.9-33.4)	0.783
Day 4
CRP (mg/L)	99.5 (48.5-155.3)	80.5 (46.5-128.0)	151.5 (90.8-204.6)	0.047	96.5 (49.3-156.8)	123.5 (40.1-163.5)	0.918
NLR (mg/L)	8.4 (6.2-11.1)	8.1 (6.0-10.4)	9.3 (6.7-18.5)	0.120	8.2 (6.2-10.6)	8.73 (6.4-24.2)	0.294

Results are presented as medians and interquartile ranges. CRP: C-reactive protein; NLR: Neutrophil-to-lymphocyte ratio.

Table 3 Daily area under the receiver operating characteristic curve of C-reactive protein and neutrophil-to-lymphocyte ratio in patients with overall postoperative infection and with intra-abdominal infection.

Characteristic	AUC for OPI (95%CI)	P value	AUC for IAI (95%CI)	P value
Day 1
CRP (mg/L)	0.6 (0.4-0.7)	0.453	0.6 (0.5-0.8)	0.201
NLR (mg/L)	0.5 (0.4-0.7)	0.716	0.6 (0.3-0.8)	0.448
Day 2
CRP (mg/L)	0.7 (0.5-0.8)	0.019	0.7 (0.5-0.9)	0.129
NLR (mg/L)	0.5 (0.4-0.7)	0.946	0.6 (0.3-0.8)	0.557
Day 3
CRP (mg/L)	0.7 (0.6-0.9)	0.005	0.6 (0.3-0.8)	0.550
NLR (mg/L)	0.5 (0.3-0.6)	0.674	0.5 (0.3-0.8)	0.783
Day 4
CRP (mg/L)	0.7 (0.5-0.9)	0.038	0.5 (0.3-0.8)	0.831
NLR (mg/L)	0.6 (0.5-0.8)	0.120	0.6 (0.4-0.9)	0.294

Results are presented as medians and interquartile ranges. AUC: Area under the receiver operating characteristic curve; CI: Confidence interval; CRP: C-reactive protein; IAI: Intra-abdominal infection; NLR: Neutrophil-to-lymphocyte ratio; OPI: Overall postoperative infection.

NLR

Daily NLR levels on the first four PODs in relation to the presence or absence of OPI or IAI are shown in Table 2. Among patients with OPI, NLR levels rose on POD1, then fell on POD2 and POD3, and peaked again on POD4, while among patients without OPI NLR levels were stable on POD1 to POD3 and then descended on POD4 (Figure 1A). No statistical differences were found between patients exhibiting OPI or IAI compared to patients without these infections. On POD1, patients with IAI presented a higher NLR than patients without IAI, although the differences were not statistically significant (15.9 vs 9.9, P = 0.228).

The AUC for NLR values obtained on POD1 to POD4 are shown in Table 3. The differences between values for the first four PODs were not significant. The best AUC ratio was obtained in patients with OPI on POD4: 0.6 (95%CI: 0.5-0.8, P = 0.120) with an optimal cut-off value of 8.7, yielding a sensitivity of 56.3% and a specificity of 56.2%.

Effect of preoperative immunonutrition intake on CRP levels

CRP levels varied depending on the immunonutrition intake (Table 4). Patients with immunonutrition intake showed higher CRP levels on POD2 regardless of whether they presented OPI. In those with immunonutrition intake and OPI, the CRP peak was higher on POD2, but CRP levels were lower on POD3 and POD4 compared to patients without immunonutrition intake. Meanwhile, patients with IAI exhibited higher CRP levels despite immunonutrition intake on POD1 to POD3, but on POD4, patients without immunonutrition intake showed higher CRP levels, although the differences were not significant (Table 5).

Table 4 Overall postoperative infection: Daily levels of C-reactive protein depending on immunonutrition intake.

Characteristic	Immunonutrition
	No (n = 62)	Yes (n = 49)	P value
CRP day 1
Overall postoperative infection	60.0 (39.0-121.0)	96.4 (82.3-126.0)	0.282
No overall postoperative infection	77.0 (54.5-95.0)	80.0 (63.0-106.7)	0.346
CRP day 2
Overall postoperative infection	88.0 (61.5-204.0)	163.0 (117.0-186.3)	0.535
No overall postoperative infection	84.0 (52.8-109.0)	103.0 (76.7-158.8)	0.005
CRP day 3
Overall postoperative infection	165.0 (72.5-268.0)	125.0 (88.0-167.5)	0.313
No overall postoperative infection	64.0 (40.5-96.5)	102.0 (65.3-143.8)	0.008
CRP day 4
Overall postoperative infection	154.5 (45.3-187.0)	146.0 (108.3-205.7)	0.811
No overall postoperative infection	65.3 (36.5-122.3)	97.0 (62.0-130.0)	0.243

Results are presented as medians and interquartile ranges. CRP: C-reactive protein.

Table 5 Intra-abdominal infection: Daily levels of C-reactive protein depending on immunonutrition intake.

Characteristic	Immunonutrition
	No (n = 62)	Yes (n = 49)	P value
CRP day 1
Intra-abdominal infection	113.0	86.0 (61.0-138.5)	0.487
No intra-abdominal infection	67.0 (50.0–95.0)	81.0 (65.0-113.0)	0.112
CRP day 2
Intra-abdominal infection	199.0	120.0 (78.5-175.5)	0.510
No intra-abdominal infection	84.0 (53.5-111.0)	118.0 (77.0-170.0)	0.002
CRP day 3
Intra-abdominal infection	114.0	124.0 (41.5-167.5)	0.642
No intra-abdominal infection	67.0 (41.0-114.0)	114.4 (65.0-172.0)	0.047
CRP day 4
Intra-abdominal infection	48.0	142.0 (60.8-177.0)	0.377
No intra-abdominal infection	76.0 (38.0-156.0)	102.0 (65.0-172.0)	0.278

Results are presented as medians and interquartile ranges. CRP: C-reactive protein.

On the other hand, patients with immunonutrition intake who did not present OPI showed higher levels of CRP than those without immunonutrition on POD2 and POD3 (103 vs 84, P = 0.005, and 102 vs 64, P = 0.008), and even on POD4 although the difference was not statistically significant. Similarly, patients with immunonutrition intake who did not present IAI showed higher CRP values on POD2 and POD3 (118 vs 84, P = 0.002, and 114 vs 67, P = 0.047) (Table 5).

The kinetics of CRP in patients with OPI is depicted in Figure 1B. Patients with immunonutrition intake had higher values of CRP on POD1 and POD2, whether or not they presented OPI. Subsequently, on POD3 and POD4, patients with OPI presented higher levels of CRP than patients without OPI, regardless of immunonutrition intake. Among patients with OPI, the CRP peak value was recorded on POD2 in those with immunonutrition intake and on POD3 in those without (Figure 1B). Among patients without OPI, CRP levels on POD2, POD3, and POD4 were higher in patients with immunonutrition intake than in those without.

DISCUSSION

This study demonstrates the usefulness of CRP levels for early detection of OPI occurring during the first week in patients with peritoneal metastases who had undergone CRS. Its main finding is the difference in CRP kinetics depending on whether or not patients were administered preoperative immunonutrition. CRP levels rise after an inflammatory aggression and the intensity of the inflammatory event will determine the amount of CRP produced[27]. In general, CRP concentration doubles every 8 hours and reaches its highest value after 36 to 50 hours[40]. It then recedes, gradually on the following POD as surgery-related inflammation diminishes, although CRP may rise again after POD10 in patients who receive HIPEC[28]. Moreover, this kinetics is also altered in the case of an underlying infection. Knowledge of this kinetics enables the use of CRP as an early detector of postoperative infections; nevertheless, a wide range of different cut-off points have been proposed to identify patients who might be at risk of developing a postoperative infection depending on the surgery performed[23,24,26,28].

In our study, CRP levels peaked on POD2 and receded gradually on POD3 and POD4 in patients without OPI, but remained high in patients with OPI. The optimal cut-off value for detecting early OPI was 118 mg/L on POD3. Only a trend towards higher CRP values during the first three PODs was found in patients who presented IAI. In colorectal surgery, a CRP cut-off between 94-125 mg/L on POD4 has been proposed as a sign of alarm, ruling out patient discharge[23,24]. Similarly, CRP concentrations above 166 mg/L on POD3 or above 116 mg/L on POD4, as well as maintained high CRP levels on POD3 and POD4, have been associated with a higher risk of postoperative complications after CRS[26,28]. It has also been observed that the CRP peak occurs on POD2 and reaches values between 136 mg/L and 186 mg/L[26,28], and this pattern has been proposed for use as an individual reference point in the monitoring of postoperative evolution[32]. According to our results, patients with CRP values > 118 mg/dL on POD3 should be carefully evaluated and medical discharge is not recommended until OPI is ruled out. These findings are in line with those reported in previous work[23-26,28].

Interestingly, in our study, patients with immunonutrition intake showed higher values of CRP than those without, regardless of whether or not they presented OPI. It is noteworthy that CRP levels on POD2, POD3, and POD4 in patients who did not suffer OPI were higher in those with immunonutrition intake than in those without. Considering that immunonutrition formulas are designed to modulate our immune system response, these findings may seem paradoxical. After an injury, protein synthesis switches from constitutive and visceral proteins to proinflammatory cytokines such as tumor necrosis factor alpha (TNF-a), interleukin 1 (IL-1), and IL-6, the last of which stimulates CRP production[41]. Omega-3 fatty acids and nucleotides, which are present in immunonutrition formulas, restore normal homeostasis postoperatively by reducing proinflammatory mediators such as IL-6 and TNF-a[42]. Therefore, patients with immunonutrition intake were expected to have lower levels of CRP than those without, as some researchers have reported [41-43]. On the other hand, several studies have suggested that CRP may have the capacity to induce the production of IL-1, an anti-inflammatory cytokine with an active role in the regulation of inflammation[44]. A recent meta-analysis of patients undergoing esophagectomy concluded that perioperative enteral immunonutrition did not reduce postoperative levels of CRP or IL-6[45].

The current European Society of Clinical Nutrition and Metabolism guidelines[46] recommend the administration of specific formulas enriched with immunonutrients in malnourished patients undergoing a major cancer surgery and, therefore, many Enhanced Recovery After Surgery protocols include the recommendation of using immunonutrition formulas during the preoperative or perioperative period[47,48]. Although the use of immunonutrition has established itself as a standard practice worldwide, few studies have assessed its actual effect on laboratory values, and the results reported vary widely. One meta-analysis published in 2018 analyzing the effect of parenteral immunonutrition on patients undergoing colon cancer surgery concluded that it lowered IL-6 and CD8 levels while raising those of CD3, CD4/CD8 and CD4 T lymphocytes[49]. Meanwhile, another meta-analysis published in 2023 involving immunonutrition in patients undergoing pancreaticoduodenectomy did not find significant differences in the levels of IL-6 and CRP between patients with and without immunonutrition[48]. These findings differ from ours, since our patients showed higher levels of CRP regardless of the presence or absence of OPI. Therefore, further investigation of the effect of immunonutrition on inflammation markers and cytokines is probably needed.

In this regard, we focused our analysis on CRP levels during the first four PODs[50] since their elevation has been consistently correlated with the presence of postoperative complications occurring in the first week[23-25,27,28]. In fact, after POD14, patients undergoing CRS without complications continue to present CRP values 10 times above the normal cut-off point[28]. Hence, after the first postoperative week, the usefulness of CRP for detecting postoperative complications after CRS and HIPEC falls notably. As for NLR, several studies have highlighted its utility for detecting postoperative complications in major abdominal surgery[31,32]. In our study, no differences in NLR patterns were observed according to the presence of absence of OPI and IAI; nor were there differences in the NLR kinetics during the early postoperative period between patients with and without OPI. Therefore, we cannot confirm the usefulness of this marker for monitoring postoperative evolution in patients with peritoneal metastases undergoing CRS and HIPEC. In a study carried out on patients undergoing CRS, patients with infectious complications showed NLR levels above 7-8 throughout the postoperative period, while patients without complications (or with non-infectious complications) showed levels around 5. In fact, that study concluded that only CRP was useful for detecting infectious complications[28].

The key insight of our study is the attention it draws to the dynamic pattern of CRP kinetics as a marker of OPI, and, therefore, to the risk of relying solely on absolute CRP values, which can fluctuate depending on the administration of preoperative immunonutrition. This finding underscores the value of using the patient’s CRP level on POD2 as an individualized baseline. By comparing this baseline with the CRP level on POD4, clinicians can gain a more reliable indicator of potential complications: If the CRP level is higher on POD4 than on POD2, vigilant monitoring and consideration of an early CT scan are warranted. This proactive approach may facilitate the timely detection and management of postoperative complications, ultimately improving patient outcomes. Our study has several limitations. It is a retrospective analysis, in which data on CRP levels on POD4 were lacking in a number of patients. Moreover, the low number of patients with IAI precluded an accurate assessment of CRP and NLR as predictive factors for early IAI detection.

CONCLUSION

In conclusion, CRP levels were useful for detecting early OPI in patients with peritoneal carcinomatosis undergoing CRS. A cut-off value of 118 mg/L on POD3 yielded the best sensitivity and specificity, but tracking CRP kinetics may be more helpful for predicting OPI. By contrast, NLR levels did not accurately predict OPI in these patients. Preoperative immunonutrition intake raised early postoperative CRP levels both in patients with OPI and those without.