Patients with abdominal aortic aneurysm who underwent either open abdominal surgery or endovascular aneu- rysm repair at Juntendo University Hospital were studied. Peripheral blood sample was used, and it was handled according to the Declaration of Helsinki. The study was approved by the Human Ethics Committee of Juntendo University. A written consent was signed by each patient prior to the surgery.

Patients treated for abdominal aortic aneurysm from April to December 2012 were divided into two groups. The first group underwent open abdominal surgery (OP), and the other group underwent endovascular aneurysm re- pair (SG). Cases were omitted if the patient received preoperative steroid treatment, the surgery combined mul- tiple procedures such as open heart surgery, or the patient had inflammatory diseases such as vasculitis. Pre- operative peripheral blood was taken to measure the baseline levels of endotoxin (Et), body temperature, white blood cell count (WBC), CRP, tumor necrosis factor α (TNF-α), and interleukin 6 (IL-6). Among them, WBC and CRP were measured in central laboratory of Juntendo University, and TNF-α, IL-6 were measured by out- sourcing laboratory (SRL Inc, Tokyo). These values were also measured postoperatively until postoperative day 5 to calculate the rate of change. Patient information such as age, gender, BMI, medication, complications, pre- vious surgical history including procedure, device used, operative time, and postoperative course were recorded and compared between the two groups.

Spectral’s Endotoxin Activity Assay (EAA, Toray Medical Co. Ltd., Tokyo) was used for the measurement of whole blood endotoxin. The presence of endotoxin or lipopolysaccharide (LPS), an outer membrane component of gram-negative bacteria, can result in immune complex formation in the presence of a specific anti-LPS anti- body to yield complement fixation. Then, the opsonized zymosan primes neutrophils in the blood to trigger their oxidative burst.

A chemilumine scence assay was performed according to the manufacture’s instruction. Actual measurements were made of light emissions over a total period of 15 min in a tubeluminometer (Auto Lumat LB953, E.G. & G, Berthold, Wildbat, Germany) and quantitative analysis was completed within 2 hours of sample collection. The relative light units (RLU) measured by the instrument was converted by calculation into an endotoxin activity (EA) value.

Each measurement was recorded as average ± standard error. Comparison between the 2 groups was made using Statcel 3 software (OMS publishing Inc, Tokyo). Statistical significance was determined using the Student’s t-test, and P < 0.05 was considered significantly different. The comparison of the peak level of Et with preopera- tive value in each patient was analyzed by the paired t-test.

During the study period, 17 cases of abdominal aortic aneurysms were treated. Open abdominal surgeries (OP group) were performed in 9 of these cases, and endovascular aneurysm repairs (SG group) were performed in 8 cases. Patients with preoperative endotoxin (Et) above the low level cutoff value (0.4) were excluded from the study. The study was performed in 14 remaining cases, 7 cases in each group, which met none of the other ex- clusion criteria.

Preoperative patient background is shown in Table 1. The OP group consisted of all male patients, and the SG group consisted of 4 male and 3 female patients. The average age of patients were 70.6 ± 8.5 (OP) and 71.1 ± 7.3 (SG) and did not show significant difference. There were no significant differences in preoperative BMI, diabetes, hypertension, hyperlipidemia and smoking history. Serum creatinine (Cre) were significantly higher in the OP group than in the SG group (1.63 ± 0.89 vs 0.73 ± 0.12, P = 0.02). No significant difference was seen in preoperative endotoxin (Et) between the groups (0.17 ± 0.07 vs 0.19 ± 0.06, P = 0.47). There were no significant differences in other variables such as body temperature, WBC, CRP, TNF-α, and IL-6 as illustrated in Table 2. The OP group consisted of 5 patients with abdominal aortic aneurysm and 2 with arteriosclerosis obliterans, and all patients in the SG group had abdominal aortic aneurysm. Operative time and postoperative hospital stay were significantly longer in the OP group, and there were no significant difference in blood transfusion or ICU stay (Table 3).

^*BMI: body mass index, DM: diabetes mellitus, HT: hypertension, HL: hyperlipidemia, creatinine: serum creatinine.

^*Et: endotoxin, BT: body temperature, WBC: white blood cells, CRP: C-reactive protein TNF-α: tumor necrosis factor-alpha, IL-6: interleukin-6.

Figure 1 shows the sequential changes in EA levels. Although no significant difference between the 2 groups was seen on the day of surgery to 1 day after surgery, the value increased significantly in the OP group 3 days after surgery (0.38 ± 0.06 vs 0.26 ± 0.08, OP vs SG, P = 0.01). This value in the OP group decreased on day 5, and no significant difference was seen compared with the SG group.

The peak level of EA in each patient was compared with the preoperative baseline (Figure 2). The EA values were increased in all patients of the OP group. In contrast, while the peak Et value in the SG group showed an increase compared to the baseline in most cases, some of them were unchanged or slightly decreased. The mean value increased significantly after surgery in both OP (P = 0.002) and SG groups (P = 0.035), and the slope was steeper in the OS group than the SG group.

WBC, neutrophil count, CRP, TNF-α, IL-6, and body temperature were measured as inflammatory markers (Figure 3). WBC and neutrophil count were both significantly higher in the OP group immediately after the surgery (P < 0.01), but did not show significant differences subsequently. No significant difference in serum CRP levels between the 2 groups was observed during the study. The OP group showed higher levels of serum TNF-α compared to those in the SG group throughout the study, but the difference was not significant. Serum IL-6 reached peak level on postoperative day 1, and then subsequently decreased in both groups. The level was slightly higher in the OP group on postoperative day 3, but the difference was not significant (P = 0.86). Change in body temperature was slightly higher in the SG group and prolonged fever occurred, but the increase was not significant (P < 0.09).

Previous reports indicate that high levels of postoperative endotoxin lead to prolonged ICU and hospital stay. However, there are few reports which track the change in post operative EA levels for up to five days after treatment. Klein et al. performed the sequential determinations of perioperative EA levels in patients who un- derwent open heart surgeries with cardiopulmonary bypass by using the EAA method, and concluded that high levels of endotoxin increase the risk of infection [3] . However, the measurements were only performed imme- diately after aortic clamping and on postoperative day 1 (within 24 hours). Previous studies demonstrated that the endotoxin level was the highest during cardiopulmonary bypass, and then it decreased with time [5] [23] .

Contrastively, in our current study, endotoxin levels did not present significant increase until the first day of surgery in either group, but all cases in the OP group showed an increase until the third day of surgery. The en- dotoxin levels in three of these cases exceeded 0.4, the low level cutoff set by the MEDIC Study [12] .

Intestinal handling is thought to be the reason behind the elevation of postoperative EA levels in the OP group. All cases in the OP group involved infrarenal aortic clamping, and there were no remarkable events in the intes- tinal blood supply from the celiac artery and the superior mesenteric artery. As in other general operative pro- cedures, surgical view was kept clear using devices such as a retractor. It has been previously suggested that in- testinal translocation could increase blood endotoxin [23] , and it is suspected that intestinal retraction causes this increase on postoperative day 3 when the diet is generally started in patients who underwent open abdominal surgery. In addition, the traditional Limulus test was performed with Limulus Amebocyte Lysate (LAL) assay kit (Wako pure chemical industrial, Tokyo, Japan) in serum blood samples by outsourcing laboratory (SRL Inc.), which showed the highest level of postoperative endotoxin in the OP group, but the levels did not exceed the reference value (1.0 pg/mL) at any point after surgery. The tendency of postoperative EA levels to increase a few days after surgery became evident from our analysis, but the mechanism behind it is still unclear and re- quires further investigation.

Meanwhile, there were 3 cases with preoperative endotoxin levels of 0.4 or above which were excluded from the study. Of these cases, preoperative pneumonia and pulmonary metastasis were suspected in one case, and the other two did not display any evidence of infection though Buerger’s disease was present in one of them. Taking into account the MEDIC Study which reports endotoxin levels of 0.4 - 0.6 were seen in approximately 7% of healthy volunteers, there could be a problem of nonspecific elevations in three excluded cases, or there may be some clinical significance other than infection.

Another interesting result from this study was that observed elevations in blood endotoxin were not dependent on WBC count, CRP and inflammatory cytokine levels such as serum TNF-α or IL-6. Previous studies have pointed out that elevated endotoxin stimulates the production of inflammatory markers such as TNF-α, IL-1β, and IL-6 [24] . In this study, only endotoxin levels were significantly different between OP and SG groups while other inflammatory markers did not present such differences. Since no sign of infection other than endotoxin was seen in these measurements, the possibility has been suggested that blood endotoxin assay would be useful as an inflammatory marker independent of other factors. The study revealed no relationship between endotoxin levels and postoperative complications, since there were a limited number of cases, no sigh of postoperative in- fections and complications were present in any of the cases, and endotoxin levels did not exceed 0.6. However, there have been numerous reports in the past which point out a possible relationship between the two. In addi- tion, the EAA method used in this study is more simplified and requires less time compared to the conventional method described above. Furthermore, since only EAA showed differences between open surgery and endovas- cular treatment on post-operatively period, it may be useful as a sensitive bio-marker to detect inflammation for abdominal aortic surgery. However, even in the case which exceeded baseline value (0.4) did not show any sign of infarction or other complication, further evaluation is necessary to set a cut-off value.