Plasma presepsin in determining gastric leaks following bariatric surgery

Introduction

Laparoscopic sleeve gastrectomy (LSG) was first described by Gagner in 2003 and has become one of the most effective surgical interventions for the prevention of morbid obesity in the long term [1]. Leakage and bleeding from the staple line are the complications with the greatest effect on morbidity and mortality following LSG surgery [2], [3]. Staple line leaks after LSG have been reported as 0.5%–20% in several studies [4], [5], [6]. Despite advances in stapler technology, anastomotic leaks remain a serious problem for surgeons [7], [8]. The gastroesophageal junction and proximal gastric region, known as the His angle, is the most common area of leakage [9], [10]. Technical insufficiencies, insufficient blood circulation, sepsis, and local ischemia resulting from inadequate oxygenation are common causes of stapler leaks [11], [12]. Early leaks manifest with sudden onset of abdominal pain, tachycardia and fever. Early detection of leakage is an important problem, especially in patients with late-onset leaks or inpatients who cannot be diagnosed after the abdominal examinations [13]. The most commonly used laboratory parameters for detecting leaks are white blood count (WBC) (leukocyte) and C-reactive protein (CRP) values [14]. The soluble CD14 subtype (sCD14-ST) presepsin was identified by Yaegashi et al. [15]. Cluster of differentiation 14 (CD14) is a multifunctional glycoprotein mostly expressed on monocyte/macrophage membrane surfaces. There is also minimal distribution of CD14 on the cell surface of neutrophils (mCD14), which functions as a specific receptor for complexes of lipopolysaccharides (LPSs) and LPS-binding proteins (LPBs). When CD14 is released from the cell membrane, the LPS-LBP-CD14 complex is released into the circulation providing soluble CD14 (sCD14), which is also directly secreted by hepatocytes [16], [17]. In conditions of inflammation, plasma protease activity produces sCD14 fragments. The 64-amino acid N-terminal fragments form the sCD14 subtype (sCD14-ST), which has recently been renamed presepsin [18]. In several recent studies, the role of presepsin has been evaluated as a biomarker in adults. According to the results of those studies, presepsin seems to provide advantages of an earlier increase in blood levels, higher sensitivity and specificity, and prognostic value during inflammatory stress compared to the most well-known and widely-used markers of CRP and procalcitonin (PCT) [19], [20], [21].

The most important factor in reducing morbidity and mortality caused by stapler line leaks is early diagnosis [22], [23]. The aim of this study was to determine the utility of plasma presepsin levels in the detection of stapler leaks and in monitoring leakage together with WBC, CRP and neutrophil-lymphocyte ratio (NLR).

Materials and methods

Results

In the comparison of preoperative WBC, CRP, NLR and presepsin values, no statistically significant differences were found between the patient and control groups. The mean plasma presepsin value of the control group was determined to be 273 pg/mL (Table 1).

No statistically significant difference was determined between the patients with and without leakage in respect of preoperative leukocyte counts (p>0.05). WBC values on postoperative days 1, 3 and 5 were higher in cases with leakage (p=0.035, p=0.001, p=0.001) respectively. There is significant difference between leukocyte preop, day 1, day 3, day 5 (p¹<0.001). The mean of day 1 is higher in staple line leak (−). The mean of day 5 is higher in staple line leak (+) (Table 2, Figure 1).

Figure 1:

Mean leukocyte values of patient groups.

No statistically significant difference was determined in respect of preoperative CRP values (p>0.05). CRP values on postoperative days 1, 3 and 5 were higher in cases with leakage (p=0.001, p=0.001, p=0.001, respectively). There is significant difference between CRP preop, day 1, day 3, day 5 (p¹<0.001). The mean of day 3 is higher in staple line leak (−). The mean of day 5 is higher in staple line leak (+) (Table 3, Figure 2).

Figure 2:

Mean C-reactive protein values of patient groups.

No statistically significant difference was determined in respect of preoperative NLR values (p>0.05). NLR values on postoperative days 1, 3 and 5 were higher in cases with leakage (p=0.001, p=0.001, p=0.001), respectively. There is significant difference between NLR preop, day 1, day 3, day 5 (p¹<0.001). The mean of day 3 is higher in staple line leak (−). The mean of day 5 is higher in staple line leak (+) (Table 4, Figure 3).

Figure 3:

Mean neutrophil-lymphocyte ratio of patient groups.

No statistically significant difference was determined in respect of preoperative presepsin values (p>0.05). Presepsin values on postoperative days 1, 3 and 5 were higher in cases with leakage (p=0.001, p=0.001, p=0.001, respectively). There is significant difference between preop, day 1, day 3, day 5 (p¹<0.001). The mean of day 5 is higher in staple line leak (−) and staple line leak (+) (Table 5, Figure 4).

Figure 4:

Mean presepsin values of patient groups.

Cut-off values of WBC, CRP, NLR and presepsin were determined as 13.2×10³/μL, 3.4 mg/dL, 4.3 and 461 pg/mL, respectively.

Sensitivity and specificity values were determined as 77.78% and 63.57%, respectively for WBC, 88.89% and 87.97% for CRP, 77.78% and 97.94% for NLR, and 100% and 96.91% for presepsin. PPV, NPV and the area below the ROC curve of WBC were calculated as 6.2, 98.9 and 71%, respectively, of CRP at 18.6, 99.6 and 94%, of NLR at 53.8, 99.3 and 88%, and of presepsin at 50, 100 and 99% (Table 6).

The comparisons of postoperative day 1 values of the patients with staple line leakage are summarized in Table 7. The presepsin value was more significant than the other laboratory parameters (Leuc, p<0.05; CRP, NLR). The areas under the ROC curve were PRE>CRP>NLR>LEUC. There was also only significant difference between the LEU and PRE according to AUC data (p<0.008 Bonferroni correction).

Stapler line leakage developed in a total of 9 (3%) patients (6 females, 3 males; median age: 40 years (range: 19–56 years); mean BMI: 43.2). These patients were treated with endoscopic stenting and percutaneous drainage. The intra-abdominal drainage status was seropurulent in 5 of 9 patients and the others were serous. Other than intra-abdominal abscess, no major complications developed in any patients.

Discussion

Early detection of leakage is the most important criterion for decreasing the mortality rate (1–10%), which may occur as a result of abscess, peritonitis, sepsis, or multi organ failure after stapler line leakage [25], [26]. A delayed diagnosis may result in prolonged hospital stay, re-hospitalization and increased treatment costs [27]. In all cases, the early detection of postoperative leaks is of critical importance [28]. Tachycardia is the most important early clinical sign of leakage. It was observed that tachycardia and presepsin levels were correlated (data not shown). Physical examination (tachycardia etc.), inflammatory parameters and computed tomography are used to detect anastomotic leaks [26], [28], [29]. CRP, which is synthesized from the liver and has a half-life of 19 h, is an inflammatory marker commonly used in the diagnosis of intra-abdominal infections. In the postoperative period, CRP has 74% sensitivity and 75% specificity in the detection of anastomotic leakage [30]. In the current study, a CRP value >3.4 mg/dL on the first day was found to be significant.

The neutrophil lymphocyte ratio (NLR) is a parameter that is easily calculated by the ratio of neutrophil and lymphocyte numbers determined in the whole blood count. This is accepted as an inflammation marker, and the efficacy has been evaluated in various diseases in recent studies [31], [32]. There are studies advocating that NLR is a useful marker for determining the degree of inflammation (non-complicated/complicated appendicitis) in acute appendicitis, which is the most common cause of emergency surgical pathology [33]. These results support the hypothesis of the utility of the NLR in the identification and follow-up of an acutely deteriorating table. In the current study, NLR≥4.3 on postoperative day 1 was found to be significant.

Sargentini et al. showed that in the diagnosis of sepsis, presepsin, which has moderate diagnostic value alone, has increased diagnostic value and may be used as a sepsis biomarker when used as an acute phase reactant with other inflammatory parameters [34]. In a meta-analysis, presepsin, as an effective biomarker in the diagnosis of sepsis, was reported to have 83% specificity and 78% sensitivity [35]. In the current prospective study of presepsin in acute inflammatory pathologies, the median plasma level of presepsin in the control group was found to be 273 pg/mL, which was similar to the normal value of presepsin as described previously.

Patients with stapler fistula had similar statistically significantly elevated levels of presepsin, CRP and NLR due to intra-abdominal infection on the first day postoperatively, which predicted complications and these values were in accordance with the levels on the postoperative 3rd and 5th days.

According to comparisons of the days by each group in stapler line leak (−) group, mean LEU levels were reached its maximum level at day 1 and then it was shown tendency to decreasing. Also maximum mean CRP and NLR levels were found at 3rd day. This findings may be misleading to determining the leaks, furthermore ROC analysis have shown that the presepsin is more sensitive and effective marker.

As morbidly obese patients are difficult to examine, these parameters play an even more important role in the detection of intra-abdominal infections.

The postoperative day 1 presepsin value was determined to be more significant than the other laboratory parameters (Leu, CRP, NLR), according to the ROC analysis of the patients with staple line leak, indicating that presepsin may be more effective in early-term clinical practice.

Due to the diversity of intra-abdominal flora, early intervention on day 1 for leakage after obesity surgery has been shown to reduce the risk of mortality [36], whereas an average of 2.5 days delay in intervention increases mortality by 24–39% [37].

One of the negative aspects of the study is the high rate of leakage. This is because some of the operations were performed during the training of the assistants under the supervision of experienced bariatric surgeons [38], [39].

The results of this study indicated that elevated levels of presepsin can be more useful in the detection and follow-up of leaks following LSG compared to other inflammatory parameters such as CRP and NLR especially on postoperative days 1 and 3. The increase in presepsin level, especially on the first postoperative day, is important in the detection of possible postoperative complications which are not seen clinically. Further research will provide a better understanding of the relationship between presepsin and complications in morbid obesity surgery.