Postoperative course of C-reactive protein level after liver surgery—when to worry?

Abstract

Background

C-reactive protein (CRP) level is an established marker to detect inflammation and infectious complications after surgical procedures, but its use after liver surgery has not been well investigated. This study aimed to elucidate the role of CRP level after elective liver resection and to determine baseline values for a normal postoperative course, as well as the role of CRP level as a predictor of postoperative complications.

Methods

All liver resections performed between January 2008 and December 2023 were identified. Patients’ characteristics, intraoperative data and postoperative outcomes were analyzed. Postoperative surgical complications within 90 days were graded using the Clavien-Dindo classification system. The postoperative CRP values from postoperative day (POD) 1 to 10 were included.

Results

In total, 3,177 liver resections were identified during the study period, of which 85.1% (2,705) were performed for malignant liver disease. In 1,647 (51.8%) cases, there were no postoperative complications. The median CRP level in cases without complication (baseline) was 55 mg/L on POD 1, 130 mg/L on POD 2, 128 mg/L on POD 3, 101 mg/L on POD 4, 80 mg/L on POD 5, 70 mg/L on POD 7 and 120 mg/L on POD 10. The percentage of cases who suffered complications (highest grade) was 7.3% (n=232) with grade I, 14.5% (n=462) grade II, 14.5% (n=461) grade IIIa, 4.2% (n=132) grade IIIb, 3.7% (n=117) grade IVa, 0.4% (14) grade IVb, and 3.5% (n=112) with grade V complications. Infectious complications occurred in 28.9% (n=919) and severe infectious complications (≥ grade IIIa) in 20.4% (n=648) of the cases. The median CRP values in cases with infectious complication were significantly higher on POD 1 (59 mg/L; P=0.01), POD 2 (143 mg/L; P<0.001), POD 3 (164 mg/L; P<0.001), POD 4 (149 mg/L; P<0.001), POD 5 (134 mg/L; P<0.001) and POD 7 (114 mg/L; P=0.043). In multivariate analysis, only elevated CRP values on POD 5 (P=0.003) were predictors for infectious complications. The AUC for POD 5 was the highest at 0.72 and presented the highest predictive accuracy. The optimal cut-off value for CRP level POD 5 was 93 mg/L. The positive and negative predictive values (PPV/NPV) for the cut-off value POD 5 were 79.6% and 87.6%, respectively.

Conclusions

Postoperative elevated CRP level on POD 5 is a valuable predictive tool for early detection of infectious complications after liver surgery and can help in decision making for further diagnostic measurements.

Highlight box.

Key findings

• Postoperative elevated C-reactive protein (CRP) level on postoperative day (POD) 5 is a predictive tool for early detection of infectious complications after liver surgery.

What is known and what is new?

• CRP level is an established marker for inflammation and detection of infectious complications after surgical procedures, but its use after liver resection has not been well investigated.

• Postoperative elevated CRP level on POD 5 is a predictive tool for the early detection of infectious complications after liver surgery. The optimal cut-off value for CRP level on POD 5 was 93 mg/L.

What is the implication and what should change now?

• CRP-level on POD 5 can help in decision making for further diagnostic measurements and should be integrated into the daily clinical routine.

Introduction

Despite continuous improvement in surgical technique and perioperative management, liver surgery is still associated with considerable perioperative morbidity and mortality, approaching up to 50% and over 5%, respectively (1,2). Especially, infectious complications or bile leakages are associated with prolonged hospital stay, increased rate of mortality and worse oncological outcome (3,4). C-reactive protein level (CRP level) is a marker for inflammation, and its postoperative examination is routine after a wide array of procedures. Several studies have reported the use of CRP level as a biomarker for early detection of intra-abdominal infection and anastomotic leakages after colorectal surgery or esophagectomy (5-9). A meta-analysis by Cousin et al. identified CRP levels on postoperative day (POD) 3 and 5 as good predictive markers for infectious complications after elective colorectal surgery (6). Facy et al. reported that CRP level on POD 4 is a better predictive parameter for infectious complications than procalcitonin level after colorectal surgery (5). Furthermore, elevated CRP level, especially on POD 3, is a predictor for anastomotic leakage after esophagectomy, while CRP values below 8.3 mg/dL are a useful marker to almost exclude anastomotic leakage on POD 5 (7,8). One study evaluated the influence of open and minimal-invasive techniques for esophagectomy on postoperative CRP course (9). CRP levels were significantly higher after open esophagectomy, reflecting the trauma associated with the surgical approach. However, there are only limited data on the course of postoperative CRP level after liver surgery. What are the normal values after liver resection, and when should concerns be raised? Two studies recently reported that elevated postoperative CRP levels are a risk predictor for surgical site infection after liver resection (10,11). However, the postoperative course of CRP levels was not described in detail.

This study aimed to elucidate the role of CRP level after elective liver resection and to determine baseline values for a normal postoperative course, as well as the role of CRP level as a predictor of postoperative complications. We present this article in accordance with the STARD reporting checklist (available at https://tgh.amegroups.com/article/view/10.21037/tgh-2026-0019/rc).

Methods

Study design

All patients who underwent liver resection consecutively from January 2008 to December 2023 were identified from the prospective institutional database of the Department of General, Visceral and Transplantation Surgery, University Medical Center Mainz. Preoperative diagnostics usually included abdominal computed tomography (CT) or magnetic resonance imaging (MRI). In cases with anticipated complex liver resection, a three-dimensional CT-scan, liver volumetry or a 3D-print of the liver was performed for resection planning.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. In accordance with the federal state law (state hospital law §36 & §37) and the independent ethics committee of Rhineland-Palatinate, no ethical approval was necessary for this study as the data was retrospectively analyzed. All patients provided informed consent for the anonymous collection of data for scientific analyses.

Patients’ characteristics, American Society of Anesthesiologists (ASA) classification, cirrhosis/Child-Pugh score, comorbidities, intraoperative data, and postoperative outcomes were analyzed. Major hepatectomy was defined as resection of ≥ three liver segments, and extended hepatectomy as resection of five or more segments. The duration of operation was defined as time from skin incision to closure.

Postoperative surgical complications were graded using the Clavien-Dindo classification system (12). Biliary leakage was classified using the International Study Group of Liver Surgery (ISGLS) definition (13) and posthepatectomy liver failure was assessed according to the ISGLS definition (14). All complications that occurred during hospital stay and readmission to the hospital within three months were recorded. Mortality was defined as 90-day mortality rate. The postoperative CRP values from POD 1 to 5 were included (normal CRP level <5 mg/L) in the analysis for predictive accuracy as patients in good general condition and without complication are normally discharged after POD 5. Available values on POD 7 and ten were recorded, but not included in the predictive accuracy analysis due to selection bias. The number of cases with available CRP values for each POD are presented in Figure 1.

Figure 1.

Flow diagram of cases (total n=3,177) with available C-reactive protein values for each postoperative day. CRP, C-reactive protein; POD, postoperative day.

Surgical technique

Parenchymal dissection was performed by scratching the parenchyma with the scissor tip, using the clamp crushing technique, or using an ultrasound dissector for open surgery. Small vessels were sealed by bipolar forceps, and larger vessels were sealed using clips and sutures. The hilar structures and hepatic veins were transected and sutured over vascular clamps. An ultrasonic dissector or laparoscopic Cavitron ultrasonic surgical aspirator was used in laparoscopic surgery. Major vascular structures were closed using endo-GIA staplers. Intraoperative ultrasound was routinely performed.

Statistical analysis

Categorical variables are presented as numbers and percentages, and continuous variables are expressed as medians/ranges. Continuous variables were compared using the Mann-Whitney U test and categorical variables were compared via ² test. Binary logistic regression was used for multivariable analysis to identify independent risk factors for infectious complications. Receiver operating characteristic (ROC) curve analysis was used to assess the predictive accuracy of postoperative CRP level on postoperative infectious complications. Statistical significance was set at P values <0.05. SPSS Version 23 (IBM Corporation, USA) was used for statistical analysis.

Results

In total, 3,177 cases of liver resection were identified during the study period. Of these, 85.1% (n=2,705) were performed for malignant liver disease. The most frequent indication for liver resection was colorectal liver metastases (CRLM) in 1,265 cases (39.8%), followed by hepatocellular carcinoma (HCC) in 402 (12.7%) and intrahepatic cholangiocarcinoma (iCC) in 327 (10.3%) cases. Liver cirrhosis was present in 194 cases (6.1%), and in 985 (31%) cases, chemotherapy and/or targeted therapy or lately immunotherapy had been performed prior to resection. Preoperative cholestasis and/or cholangitis had been present in 6.9% and 3.1% of the cases, respectively. Patients with preoperative cholestasis were drained with percutaneous transhepatic bile duct drainage (PTBD) and/or endoscopic retrograde cholangiopancreatography (ERCP) and stenting prior to resection in most cases. The aim was to maintain the bilirubin level at ≤3.0 mg/dl at the time of operation Preoperative cholangitis was treated with antibiotics, and patients did not undergo surgery until infectious parameters were normalized.

Further patient information is presented in Table 1. A previous liver resection had been performed in 480 (15.1%) patients, and a minimal-invasive liver resection (laparoscopic, hand-assisted or robotic-assisted) had been performed in 345 (10.9%) cases. The types of resection are presented in Table 1. Hepaticojejunostomy was required in 374 (11.8%) and vascular reconstruction in 301 (9.5%) cases. Intraoperative blood transfusion was necessary in 257 (8.1%) cases. The median duration of operation was 200 (range, 29–815) minutes.

Table 1. Patients’ characteristics and intraoperative data (n=3,177).

Characteristics	Data
Age (years)	63 [16–92]
Male/female	1,800 (56.7)/1,377 (43.3)
BMI (kg/m2)	25.9 [15.1–54.9]
ASA
I	70 (2.2)
II	1,604 (50.5)
III	1,443 (45.4)
IV	60 (1.9)
Cirrhosis	194 (6.1)
Child A	188 (6.9)
Child B	6 (0.2)
Indications for liver resections
Colorectal liver metastases	1,265 (39.8)
Hepatocellular carcinoma	402 (12.7)
Intrahepatic cholangiocarcinoma	327 (10.3)
Perihilar cholangiocarcinoma	200 (6.3)
Gallbladder cancer	81 (2.5)
Non-colorectal liver metastases	384 (12.1)
Other malignant liver tumors	46 (1.4)
Benign indications	472 (14.9)
Previous liver resection	480 (15.1)
Minimal-invasive liver resection	345 (10.9)
Type of resection
Right hemihepatectomy	317 (10.0)
Left hemihepatectomy	251 (7.9)
Extended right hemihepatectomy	173 (5.4)
Extended left hemihepatectomy	117 (3.7)
Mesohepatectomy	81 (2.5)
ALPPS	31 (1.0)
Combination of segments	500 (15.7)
Left lateral sectionectomy	196 (6.2)
Bisegmentectomy	301 (9.5)
Segmentectomy	443 (13.9)
Non-anatomical resection	767 (24.1)
Hepaticojejunostomy	374 (11.8)
Vascular reconstruction	301 (9.5)
Intraoperative blood transfusion	257 (8.1)
Duration of operation (minutes)	200 [29–815]
Hospital stay (days)	10 [2–169]
Postoperative complication highest grade†
None	1647 (51.8)
Grade I	232 (7.3)
Grade II	462 (14.5)
Grade IIIa	461 (14.5)
Grade IIIb	132 (4.2)
Grade IVa	117 (3.7)
Grade IVb	14 (0.4)
Grade V	112 (3.5)
Types of infectious complications
Bile leakage infection	245 (7.7)
Perihepatic abscess	144 (4.5)
Cholangitis	86 (2.7)
Pneumonia	132 (4.2)
Pleural empyema	16 (0.5)
Urinary tract infection	78 (2.5)
Insufficiency gastrointestinal anastomosis	18 (0.6)
Gastrointestinal perforation	33 (1.0)
Wound infection	167 (5.3)

Data are presented as n (%) or median [range]. ^†, according to Clavien-Dindo classification (12). ALPPS, associating liver partition and portal vein ligation for staged hepatectomy; ASA, American Society of Anesthesiologists; BMI, body mass index.

Postoperative complications and CRP-level

In total, 1,647 (51.8%) of the cases had no complication. The percentage of cases who suffered complications (highest grade) was: 7.3% (n=232) with grade I, 14.5% (n=462) grade II, 14.5% (n=461) grade IIIa, 4.2% (n=132) grade IIIb, 3.7% (n=117) grade IVa, 0.4% (n=14) grade IVb, and 3.5% (n=112) with grade V complications. The median duration of IMC/ICU stay was one (range, 0–75) day and the median hospital stay was 10 (range, 2–169) days. Bile leakage was the most frequent complication, occurring in 15.4% (n=490) of the cases. Infectious complications occurred in 28.9% (n=919) and severe infectious complications (≥ grade IIIa) in 20.4% (n=648) of the cases.

The median CRP level in cases without complication (baseline) was 55 mg/L on POD 1, 130 mg/L on POD 2, 128 mg/L on POD 3, 101 mg/L on POD 4 and 80 mg/L on POD 5. CRP values on POD 7 and 10 were only available when patients were not discharged on POD 5 (minimum length of stay) and mostly in patients with complications. The median values were 70 mg/L on POD 7 and 120 mg/L POD 10 (Figure 2).

Figure 2.

C-reactive protein-values (mg/L; median) POD 1 (n=3,177), 2 (n=2,633), 3 (n=2,371), 4 (n=2,113), 5 (n=2,034), 7 (n=702) and 10 (n=522). CRP, C-reactive protein; POD, postoperative day.

In addition, the median CRP level in cases of minimal-invasive liver resection was significantly lower than that in open cases (POD 1 34 vs. 59 mg/L; P<0.001, POD 2 104 vs. 137 mg/L; P<0.001, POD 3 102 vs. 142 mg/L; P <0.001, POD 4 84 vs. 119 mg/L; P<0.001 and POD 5 83 vs. 94 mg/L; P=0.04). As patients with minimal-invasive liver resection had a median hospital stay of five days, CRP values for POD 7 (median 75 mg/L) and POD 10 (median 83 mg/L) were only available in 24 and 15 cases with complications, respectively. Figure 3 displays CRP values of open and minimal-invasive resections.

Figure 3.

C-reactive protein-values (mg/L; median) open (n=2,832) and minimal-invasive resections (n=345). POD, postoperative day.

The median CRP values for minor and major resections were 58 vs. 54 mg/L on POD 1 (P=0.15), 143 vs. 127 mg/L on POD 2 (P<0.001), 141 vs. 134 mg/L on POD 3 (P=0.16), 117 vs. 115 mg/L on POD 4 (P=0.54), 92 vs. 96 mg/L on POD 5 (P=0.28), 66 vs. 81 mg/L on POD 7 (P=0.41) and 137 vs. 119 mg/L on POD 10 (P=0.74). In cases of parenchymal sparing resections median CRP values were 61 vs. 53 mg/L on POD 1 (P<0.001), 153 vs. 126 mg/L on POD 2 (P<0.001), 148 vs. 133 mg/L on POD 3 (P=0.003), 121 vs. 114 mg/L on POD 4 (P=0.07), 96 vs. 93 mg/L on POD 5 (P=0.17), 70 vs. 71 mg/L on POD 7 (P=0.74) and 76 vs. 130 mg/L on POD 10 (P=0.63). CRP values in parenchymal and non-parenchymal sparing operations are displayed in Figure 4.

Figure 4.

C-reactive protein-values (mg/L; median) parenchymal (n=1,158) and non-parenchymal sparing resections (n=2,028) without (A) and with (B) infectious complications. POD, postoperative day.

Furthermore, CRP values in cases with CRLMs were 58 mg/L on POD 1, 140 mg/L on POD 2, 147 mg/L on POD 3, 120 mg/L on POD 4, 94 mg/L on POD 5, 80 mg/L on POD 7 and 107 mg/L on POD 10, whereas CRP values in cases with HCC were 45, 123, 142, 114, 94, 77 and 130 mg/L. In cases with iCC CRP values were 61, 131, 139, 132, 99, 96, 102 mg/L and in cases with perihilar cholangiocarcinoma (pCC) 54, 121, 136, 123, 113, 114 and 126 mg/L (Figure 5). The median CRP values in cases with CRLMs were significantly higher on POD 1 to 4 compared to other tumor entities (POD 1: P=0.043; POD 2: P=0.008; POD 3: P=0.002 and POD 4: P=0.02), representing the higher percentage of parenchymal sparing resections for this tumor entity.

Figure 5.

C-reactive protein-values (mg/L; median) after resection of CRLM (n=1,265), HCC (n=402), iCC (n=327) and pCC (n=200). CRLM, colorectal liver metastases; HCC, hepatocellular carcinoma; iCC, intrahepatic cholangiocarcinoma; pCC, perihilar carcinoma; POD, postoperative day.

When hepaticojejunostomy was required the CRP levels from POD 1 to 3 did not significantly differ from those without hepaticojejunostomy (POD 1 59 vs. 56 mg/L; P=0.55, POD 2 128 vs. 135 mg/L; P=0.38 and POD 3 141 vs. 138 mg/L; P=0.89). However, from POD 4 to 7 the CRP level in cases with hepaticojejunostomy was significantly higher (POD 4 128 vs. 114 mg/L, P=0.04; POD 5 114 vs. 92 mg/L, P<0.001 and POD 7 112 vs. 66 mg/L, P=0.003).

Moreover, the median CRP level in cases with infectious complication was significantly higher on POD 1 (59 mg/L; P=0.01), POD 2 (143 mg/L; P<0.001), POD 3 (164 mg/L; P<0.001), POD 4 (149 mg/L; P<0.001), POD 5 (134 mg/L; P<0.001) and POD 7 (114 mg/L; P=0.043).

The median CRP values in cases with no complications were not significantly different from median CRP values in cases with grade I and II complications (56 mg/L on POD 1, 131 mg/L on POD 2, 130 mg/L on POD 3, 104 mg/L on POD 4, 82mg/L on POD 5, 61 mg/L on POD 7 and 76 mg/L on POD 10). In contrast the CRP values in cases with complications grade ≥ IIIa were significantly higher with 161 mg/L on POD 3 (P<0.001), 149 mg/L on POD 4 (P<0.001), 134 mg/L on POD 5 (P<0.001), 127 mg/L on POD 7 (P=0.01) and 137 mg/L on POD 10 (P=0.02) (Figure 6).

Figure 6.

C-reactive protein-values (mg/L; median) with no complications (n=1,647), minor complications (Clavien-Dindo grade I + II; n=694) and major complications (Clavien-Dindo grade ≥ IIIa; n=836). POD, postoperative day.

Cases with infectious complications displayed significantly increased length of hospital stay (20 vs.8 days; P<0.001), rate of post-hepatectomy liver failure (9.2% vs. 2.4%; P<0.001) and mortality rate (7.5 vs. 1.9%; P<0.001).

In univariate analysis major hepatectomy, hepaticojejunostomy, preoperative cholangitis preoperative cholestasis, duration of operation, intraoperative blood transfusion and bile leakage were associated with infectious complications (P<0.001, respectively). However, in multivariate analysis only elevated CRP POD 5 (P=0.003) and postoperative bile leakage grade B+C (P<0.001) were independent risk factors for infectious complications (Table 2).

Table 2. Multivariate analysis for risk factors of postoperative infectious complications.

Infectious complications	Univariate analysis		Multivariate analysis
	Data	P	OR	95% CI	P
Major hepatectomy	56.9% vs. 36.4%	<0.001	1.2	0.5–2.7	0.71
Hepaticojejunostomy	15.0% vs. 2.0%	<0.001	0.6	0.1–2.6	0.49
Cholangitis preoperative	5.3% vs. 1.8%	<0.001	0.4	0.1–3.0	0.38
Cholestasis preoperative	13.1% vs. 4.3%	<0.001	3.8	0.7–21.1	0.13
Duration of operation	Median: 251 vs. 180 min	<0.001	1.0	1.0–1.1	0.42
Intraoperative blood transfusion	31.3% vs. 13.8%	<0.001	0.4	0.2–3.3	0.41
Bile leakage grade B + C	43% vs. 4.2%	<0.001	16.9	6.3–45.4	<0.001
CRP increase after POD 3 (>50 mg/L)	62% vs. 23%	<0.001	5.4	4.6–6.3	<0.001

CI, confidence interval; CRP, C-reactive protein; OR, odds ratio; POD, postoperative day.

ROC curve analysis was used to assess the predictive accuracy for infectious complications of CRP level from POD 1 to POD 5. The area under the curve (AUC) was 0.54 for POD 1 (95% CI: 0.52–0.56; P=0.74), 0.57 for POD 2 (95% CI: 0.54–0.59; P=0.31), 0.63 for POD 3 (95% CI: 0.6–0.66; P=0.07), 0.69 for POD 4 (95% CI: 0.66–0.71; P=0.003) and 0.72 for POD 5 (95% CI: 0.7–0.74; P<0.001). Optimal cut-off value on POD 3 was 114 mg/L. The AUC for POD 5 was the highest and presented the highest predictive accuracy (Figure 7).

Figure 7.

ROC curve analysis with sensitivity/specificity for C-reactive protein-level POD 5 (n=2,034). POD, postoperative day; ROC, receiver operating characteristic.

The optimal cut-off value for CRP level on POD 5 was 93 mg/L. The sensitivity and specificity for the cut-off value were 82.6% and 79.8%. The positive and negative predictive values for infectious complications (n=919) were 79.6% and 87.6%, respectively.

Discussion

This study aimed to elucidate the role of CRP values after elective liver resection and the association with postoperative infectious complications. CRP values in cases with minimal-invasive liver resection were lower compared to open resection. This might be the effect of less surgical trauma during minimal-invasive resection. However, in our study especially mainly minor resections were performed minimal-invasively, while complex resections, in particular those with hepaticojejunostomy and vascular resections, were performed in open technique. This for sure has led to a major bias not allowing to do any valid comparison between open and minimal-invasive liver surgery in the presented patient cohort. Matched studies are needed to describe the influence of the minimal-invasive techniques on CRP values and to compare it with open liver surgery in more detail.

One finding in our study are the somewhat higher CRP levels after parenchymal sparing resections. This might be explained by two different effects. First, multiple smaller resections result in a larger cutting surface/wound area of the liver than an anatomical resection. This might be one reason for elevated CRP levels. Second, in atypical resections blood vessels are less respected than in anatomical resections. Thus, atypical resections are associated with more impaired perfusion of liver parenchyma and more necrosis of liver cells in the area of resection margins (15,16). These findings do also explain that CRP values after resection of CRLMs were higher on POD one to four compared to other tumor entities. Most likely, this is reflecting the higher rate of parenchymal-sparing resections for this tumor entity and is not contributing to the tumor entity itself.

Infectious complications after liver surgery are a common occurrence and patient usually present with unspecific symptoms like pain or fever. However, infectious complications are a major contributing factor for an increased length of hospital stay and higher rates of post-hepatectomy liver failure as well as higher mortality rates (3,4). An early detection of infectious complications and rapid treatment of the complication with antibiotics and/or other interventions is necessary. In our study infectious complications occurred in 28.9% of the cases. The CRP levels on POD 1 to POD 7 in cases with infectious complications were significantly higher compared to cases without infectious complications. The peak-level of CRP for the latter cases was on POD 2. The peak-level in cases with infectious complications was on POD 3. The only independent risk factors for infectious complications in our study were the occurrence of bile leakage and an elevated CRP level on POD 5.

Preoperative cholestasis in combination with or without cholangitis were risk factors for postoperative infectious complications in univariate analysis, but not in the multivariate analysis. In our institution, in these patients preoperative biliary decompression usually is achieved after biliary stenting/drainage via ERCP and/or PTBD, and cholangitis is treated with antibiotics according to microbiological resistance testing (17). Despite cholestasis and cholangitis not being risk factors for postoperative infectious complications in multivariate analysis, this patient collective usually receives bile duct resection, which puts them at higher risk for developing complications, because of a higher rate of bile leakage associated with bilioenteric anastomosis (18).

Elevated CRP levels on POD 5 were a predictor for postoperative complications in our analysis. Early detection of infectious complications is integral to mitigate severity of infection and prolongation of hospital stay. Thus, postoperative elevated CRP levels may serve as a useful clinical tool after liver surgery as measuring CRP levels is part of the clinical routine after surgery. In patients with risk factors and elevated POD 5 CRP level, a diagnostic with CT-scan, and if necessary an empiric treatment with antibiotics seems to be reasonable. Moreover, elevated CRP level on POD 5 had the highest predictive accuracy for infectious complications in ROC curve analysis in our study and the optimal cut-off value was 93 mg/L. The negative predictive value to rule out infectious complications was 87.6% in our study. The study of Yasuda et al. identified CRP elevation on POD 3 to 5 as independent risk factor for organ/space surgical site infection in a patient cohort of 443 patients (10), but cut off values of CRP were not addressed. Pattou et al. investigated the predictive value of CRP values after liver resection in a cohort of 500 patients from three centers and identified CRP level on POD 3 and on POD 5 as a predictive tool (11). The cut off value of 87 mg/L on POD 5 was similar to our findings.

CRP level on POD 5 emerges as a the optimal day for predicting postoperative infectious complications mainly due to the fact that the biological half-life of CRP is around 19 hours after the inflammatory trigger is removed (19). The early days elevated CRP levels originate from the surgical trauma itself with inflammatory response syndrome and should decrease over time. On POD 5, CRP levels should be normalized, but the values stay high in cases with infectious complications.

The current study has several limitations. This study is of retrospective character and selection bias are present. Especially, values on POD 7 and 10 were only available for patients not discharged on POD 5 (minimal length of stay) and so mostly for patients with major resections and/or complications, which is why, CRP values on POD 10 are higher than on POD 5 or 7.

Conclusions

Postoperative elevated CRP-level on POD 5 is a valuable predictive tool for early detection of infectious complications after liver surgery. It can help in decision making for further diagnostic measurements in clinical daily routine.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. In accordance with the federal state law (state hospital law §36 & §37) and the independent ethics committee of Rhineland-Palatinate, no ethical approval was necessary for this study as the data was retrospectively analyzed. All patients provided informed consent for the anonymous collection of data for scientific analyses.

Data Sharing Statement

Available at https://tgh.amegroups.com/article/view/10.21037/tgh-2026-0019/dss

tgh-11-2026-0019-dss.pdf ^{(74.9KB, pdf)}

DOI: 10.21037/tgh-2026-0019