Dalal I, Somekh E, Bilker-Reich A, Boaz M, Gorenstein A, Serour F. Serum and Peritoneal Inflammatory Mediators in Children With Suspected Acute Appendicitis. Arch Surg. 2005;140(2):169-173. doi:10.1001/archsurg.140.2.169
Inflammatory markers differ between subjects with appendicitis and controls. Markers of inflammation differ in serum compared with intraperitoneal fluid. Among subjects with appendicitis, inflammatory markers differ between subjects with and without perforation.
Twenty-four children who underwent an appendectomy. Group A (n = 19) consisted of patients with appendicitis and group N (n = 5) of patients with normal appendixes.
Main Outcome Measures
Serum and peritoneal levels of interleukin (IL)8, IL-10, granulocyte colony-stimulating factor, interferon γ soluble intercellular adhesion molecule-1, matrix metalloproteinase-9, and tissue inhibitor of metalloproteinases-1 were measured by enzyme-linked immunosorbent assay.
Age, sex, complete blood count, C-reactive protein level, and serum cytokines did not significantly differ by group. Peritoneal concentrations of interleukin-8 (mean ± SD, 1416.8 ± 1436 pg/mL vs 48 ± 74.4 pg/mL, P = .001), IL-10 (mean ± SD, 3085 ± 5893 pg/mL vs 84 ± 46 pg/mL, P = .02), matrix metalloproteinase-9 (mean ± SD, 1784 ± 1225.1 ng/mL vs 435 ± 563 ng/mL, P = .03), and tissue inhibitor of metalloproteinases-1 (mean ± SD, 8939.2 ± 7312.2 ng/mL vs 602.1 ± 345.6 ng/mL, P<.001) were significantly different in group A compared with group N. When compared by perforation (n = 8 with perforation vs n = 11 without perforation), peritoneal granulocyte colony-stimulating factor levels were elevated in subjects with perforation (mean ± SD, 4.3 ± 14.4 pg/mL vs 62.7 ± 79.2 pg/mL, P = .02). Although serum tissue inhibitor of metalloproteinases-1 was not different between groups N and A, it was significantly different between group N and patients with a perforated appendicitis (mean ± SD, 205.9 ± 43.8 ng/mL vs 3068.9 ± 5122.4 ng/mL, P = .04).
Presently, it is not practical to differentiate appendicitis in a pediatric population from other causes of abdominal pain based on the detection of systemic inflammatory response markers.
Acute appendicitis is the most common cause of acute abdominal surgery in children.1 The risk of developing acute appendicitis that progresses to perforation is higher in children than in adults, with a reported perforation incidence ranging from 20% to 50%.2 Children may have many different symptoms, making acute appendicitis the most commonly misdiagnosed surgical disease in the United States.3 The difficulty of diagnosis, as expressed by the 15% to 40% of unnecessary appendectomies,3 has motivated some authors to search for better diagnostic tools. The use of sonography, computed tomography, and laparoscopic procedures has only slightly decreased the percentage of unnecessary operations,3,4 leading other investigators to focus on the inflammatory process.
Acute appendicitis is characterized by a sequence of events starting with initial obstruction of the appendiceal lumen and impaired blood flow, which lead to destruction of the local epithelium barrier, bacterial invasion, and subsequently leukocyte infiltration. The process of homing and migration of leukocytes to their target tissues involves complex interactions mediated by a variety of soluble proteins, such as C-reactive protein, cytokines, chemokines, adhesion molecules, proteases, and many cells, including polymorphonuclear leukocytes, T cells, monocytes, and natural killer cells. Few investigators have attempted to characterize the inflammatory response involved in acute appendicitis. We previously described the recruitment of unique CD19 cells including CD5+ B cells to the appendix in the process of acute appendicitis in a pediatric population.5 Other studies have tried to characterize the inflammatory response in acute appendicitis with inconclusive results.6- 10 Recently, 2 studies in adults have highlighted the diagnostic value of cytokine levels in acute appendicitis.11,12 Until now, there have been no such data in pediatric patients.
Since the magnitude of the inflammatory response to lipopolysaccharide administration in both humans and mice was found to be inversely correlated with age,13,14 we conducted this study to expand our understanding of the systemic and local inflammatory processes involved in acute appendicitis in a homogeneous pediatric group. We simultaneously analyzed several inflammatory mediators in the serum and peritoneal fluid of a cohort of children operated on for suspected acute appendicitis. We chose multiple mediators, including interleukin (IL)-8, IL-10, granulocyte colony-stimulating factor (G-CSF), interferon-γ (IFN-γ), soluble intercellular adhesion molecule-1 (sICAM-1), matrix metalloproteinase-9 (MMP-9), and tissue inhibitor of metalloproteinases-1 (TIMP-1), because they represent different arms of the inflammatory response.
Twenty-four patients, 17 males and 7 females aged 2.7 to 16 years (mean ± SD, 10.9 ± 3.6 years), were included in the study. All of them had clinical symptoms suggestive of acute appendicitis and underwent a formal open or laparoscopic appendectomy. Based on the appendiceal histological findings detected by an experienced pathologist, we identified 2 groups. Group A consisted of 19 patients with proven acute appendicitis, and group N included 5 children with normal appendixes. The study was approved by the institutional ethics committee, and informed consent was obtained from all parents or legal guardians.
Peripheral venous blood samples were drawn preoperatively in the emergency department for complete blood cell counts and C-reactive protein levels. Immediately after the peritoneal cavity was opened, a sample of intraperitoneal fluid with no blood contamination was collected into a syringe. The remaining serum and peritoneal fluid samples were centrifuged at 3000 revolutions per minute for 5 minutes and then frozen in sterile tubes at –70°C until cytokine measurement. The appendix was sent for histological examination.
Levels of IL-8, IL-10, G-CSF, IFN-γ, sICAM-1, MMP-9, and TIMP-1 were measured in duplicate with commercially available enzyme-linked immunosorbent assay kits (Research and Diagnostic Systems Inc, Minneapolis, Minn) according to the quantitative sandwich immunoassay technique. In each assay, the appropriate recombinant human cytokines were used to generate the standard curves. All samples were tested with the appropriate dilution according to the manufacturer’s instructions. The minimum detectable levels of these assay kits were IL-8, 10 pg/mL; IL-10, 3.9 pg/mL; G-CSF, 0.4 pg/mL; IFN-γ, 8 pg/mL; sICAM-1, 0.35 ng/mL; MMP-9, 0.156 ng/mL; and TIMP-1, 0.08 ng/mL.
Analysis of data was carried out using SPSS 9.0 statistical analysis software (SPSS Inc, Chicago, Ill). Descriptive statistics for continuous variables such as laboratory parameters were calculated and are reported as mean ± SD. Categorical variables were described using frequency distributions. Distributions of continuous variables were tested for normality using the Kolmogorov-Smirnov test. Variables with distributions differing significantly from normal, such as cytokines, were compared by source using the Mann-Whitney nonparametric U test. The t test for independent samples was used to detect differences in the means of normally distributed continuous variables using source as the categorical variable. One-way analysis of variance or the Kruskal-Wallis test was used to compare laboratory and demographic parameters across 3 categories of participants. Post hoc pairwise testing was carried out with the Bonferroni test or the Mann-Whitney U test, and χ2 (exact when possible) was used to detect differences in categorical variables by source. Associations between variables were described using Spearman ρ because of skewing. All tests were 2-sided and considered significant at P<.05.
Age and sex were not significantly different between the 2 groups. Also similar by group were mean white blood cell, neutrophil, platelet, and C-reactive protein values (Table 1).
There were no significant differences in serum levels of the different cytokines between the 2 groups. However, peritoneal concentrations of IL-8, IL-10, MMP-9, and TIMP-1 were significantly higher in group A than in group N (Table 2). No significant correlation was found between white blood cell counts or C-reactive protein levels and the various cytokines, except for white blood cell counts and serum MMP-9 (r = 0.53, P = .008).
To further characterize our patients with acute appendicitis, we divided them into 2 subgroups according to the histological findings of nonperforated appendicitis (group NPA, n = 11) or perforated appendicitis (group PA, n = 8), and all epidemiological and laboratory parameters were reanalyzed. Age did not significantly differ between these 2 groups, while male to female ratios were 6:5 and 7:1 in the NPA and PA groups, respectively.
When all 3 groups (NPA, PA, and N) were compared, no significant differences were found in mean white blood cell count, neutrophil count, platelet count, and C-reactive protein values. Peritoneal IL-10 levels differed significantly only between group N and group PA (mean ± SD, 84 ± 46 pg/mL vs 5726.8 ± 8305.4 pg/mL, P = .03). Peritoneal G-CSF levels were also found to be significantly different between group NPA and group PA (mean ± SD, 4.3 ± 14.4 pg/mL vs 62.7 ± 79.2 pg/mL, P = .02). Peritoneal TIMP-1 levels differed significantly between group N and group NPA (mean ± SD, 602.1 ± 345.6 ng/mL vs 11816 ± 7584 ng/mL, P = .005). Although serum TIMP-1 levels were not different between groups N and A, they were significantly different between groups N and PA (mean ± SD, 205.9 ± 43.8 ng/mL vs 3068.9 ± 5122.4 ng/mL, P = .04). Regarding the other inflammatory mediators tested (serum IL-10, serum G-CSF, and both serum and peritoneal fluid of IL-8, MMP-9, IFN-γ, and sICAM-1), no significant differences were found between the 3 groups.
In this study, we prospectively tested different facets of systemic and local immune mediators, including proinflammatory (IL-8, G-CSF, IFN-γ, sICAM-1, and MMP-9) and anti-inflammatory (IL-10 and TIMP-1) mediators, in a cohort of pediatric patients with symptoms suggestive of acute appendicitis. Our results demonstrate significantly increased levels of IL-8, MMP-9, IL-10, and TIMP-1 in the peritoneal fluid, but not in the serum, of children with histologically proven acute appendicitis compared with children with normal appendixes. By further analysis of our results based on histological findings of perforated vs nonperforated appendicitis, we were able to show that peritoneal IL-10 and TIMP-1 levels were significantly different between groups N and PA and between groups N and NPA, while peritoneal G-CSF was significantly different only between groups NPA and PA. The only mediator found to be significantly different in the serum was TIMP-1 (between groups N and PA).
Cytokines are a group of polypeptides and glycoproteins with a molecular weight from 8 to 30 kd. Lymphokines, monokines, ILs, tumor necrosis factors, IFNs, colony-stimulating factors, and a variety of other growth factors are among these. The cytokines usually are short-lived, essential mediators of immunity and inflammation that are induced in response to cellular activation rather than synthesized constitutively.15
Interleukin-8 is a chemokine produced by activated monocytes/macrophages and is considered a highly selective neutrophil chemoattractant.7,15 The effect of IL-8 on neutrophils is not restricted to the influx of neutrophils into inflamed tissue; at high concentrations, it can also activate neutrophils to degranulate their intracellular stores, thereby promoting the inflammatory reaction. Zeillemaker et al7 found that peritoneal fluid IL-8 levels varied significantly between patients with nonperforated acute appendicitis and those with perforated appendixes. In an in vitro model for bacterial peritonitis, Zeillemaker and colleagues were able to show that various species of pathogenic bacteria can stimulate cultured human mesothelium to secrete high levels of IL-8, suggesting that these cells are the main source of IL-8 production in acute appendicitis.7 Similar results were reported by Rivera-Chavez et al.12 However, Yoon et al11 recently found that preoperative serum levels of IL-6 and IL-8 were higher in patients with perforated vs nonperforated appendicitis. They speculated that serum levels of these cytokines might correlate with the degree of inflammation and help to differentiate disease severity. In contrast, we were able to show significant differences only in the peritoneal fluid of patients with acute appendicitis.
Interleukin-10 is produced by activated macrophages and helper T cells, and its major function is to inhibit activated macrophages and therefore maintain homeostatic control of innate and cell-mediated immune reactions.15 We found high levels of IL-10 only in the peritoneal fluid of patients with acute appendicitis. Similarly, Yoon et al11 reported that plasma levels of IL-10 together with IL-2 and IL-1b were not significantly different between the 2 groups of perforated and nonperforated appendicitis. In contrast, Rivera-Chavez et al12 found high levels of IL-10, together with other anti-inflammatory mediators such as IL-6, IL-1 receptor antagonist, and IL-4, in the plasma and peritoneal fluids of patients with pathologically proven appendicitis. Furthermore, patients with perforated appendicitis had higher plasma levels of IL-6 and IL-10 and lower plasma levels of IL-12 and IFN-γ. Rivera-Chavez et al12 concluded that localized infection like appendicitis could induce a systemic response that is predominantly anti-inflammatory.
Granulocyte colony-stimulating factor is produced by activated T cells, macrophages, and endothelial cells at sites of infection, and it acts on bone marrow to increase the production and mobilization of neutrophils to replace those consumed in inflammatory reactions.15 Cluitmans et al16 reported that G-CSF gene expression was detected in inflamed but not in normal appendixes. Moreover, plasma G-CSF levels were significantly elevated in patients with appendicitis compared with patients without inflamed appendixes.16 Cluitmans and colleagues concluded that G-CSF is produced at the inflammatory site and transported to the bone marrow via the blood stream to stimulate granulopoiesis, and in fact, G-CSF acts in an endocrine fashion.16 We found no difference in serum levels of G-CSF between patients with and without appendicitis. However, G-CSF levels were significantly elevated in the peritoneal fluid of patients with perforated compared with nonperforated appendicitis. It can be speculated that this discrepancy is a result of sample timing or of the magnitude of the inflammatory response.
Interferon γ is produced by T and natural killer lymphocytes whose principal function is to activate macrophages in both innate immune responses and adaptive cell-mediated immune responses.15 Lower IFN-γ concentrations were found in the peritoneal aspirates of patients with perforated vs nonperforated appendicitis, although relatively few patients in either group had detectable IFN-γ in peritoneal fluid.12
Another crucial step in the process of recruitment of neutrophils to areas of infection is up-regulation in the synthesis of adhesion molecules, such as sICAM-1. Soluble intercellular adhesion molecule-1 is produced by antigen-presenting cells and vascular endothelial cells, and it is critical to the migration of neutrophils from the vascular compartment to the inflamed organ.15 Levels of sICAM-1 were found to be elevated in some inflammatory processes, such as meningitis17 and acute respiratory distress syndrome.18 We were not able to show any difference in IFN-γ and sICAM-1 between the 3 groups in both peritoneal fluid and serum levels.
The MMPs are a group of zinc-dependent enzymes involved in the degradation of the extracellular matrix and basement-membrane components in both normal and pathological processes.17,19 Their potentially hazardous activity is controlled through binding to specific endogenous tissue inhibitors, TIMPs.20,21 The balance between MMPs and TIMPs plays an important role in normal physiologic events, such as tissue repair, embryogenesis, extravillous trophoblast invasion, and menstruation,22 and in pathologic processes, such as rheumatoid arthritis and tumor invasion.23 Animal models have suggested that degranulation of neutrophils is in fact the primary source of MMP-9.17 In addition, MMP-9 has been shown to facilitate leukocyte extravasation and permeation of serum proteins into surrounding tissues.19 Tissue inhibitor of metalloproteinases-1 is secreted as heterodimeric complexes with MMP-9 and modulates its proteolytic activity.20,24 We found significantly high levels of MMP-9 (4 fold) and TIMP-1 (15 fold) in the peritoneal fluid but not in the serum of patients with acute appendicitis as compared with patients with histologically normal appendixes. However, only patients with perforated appendicitis had high serum levels of TIMP-1 as compared with patients with nonperforated appendicitis, suggesting that TIMP-1 leaks to the systemic compartment only in progressive inflammatory response. In fact, it seems that the degree of inflammation observed in serum reflects the severity of the local inflammatory response. Gattorno et al25 showed that patients with juvenile idiopathic arthritides had significantly high levels of MMP-3 and TIMP-1 in the serum and synovial fluid. They hypothesized that these proteases may represent a crucial event for the development and perpetuation of tissue damage. Shapiro et al17 reported that MMP-2 and IL-8 concentrations were elevated in the cerebrospinal fluid of both patients with bacterial meningitis and patients with viral meningitis. In addition, Shapiro and colleagues found that MMP-9 and sICAM-1 were not detected in control cerebrospinal fluid samples but were observed in viral-infected samples and further elevated in bacterial-infected samples.17
Before conclusions can be made, several points must be considered. The wide variability in cytokine levels observed in our study population made between-subject differences difficult to detect. Nevertheless, findings of the present exploratory study indicate a need for further investigations adequately powered to detect by-group differences or lack thereof. Additionally, measurements have known limitations with regard to correlation between biologic activity and immunoassay results. Sample timing is another crucial factor since many mediators have different kinetic patterns; therefore, a single measurement does not necessarily reflect the exact role of the measured cytokine. Finally, the degree of the inflammatory response seems to be age related, and thus one cannot extrapolate results obtained from adult populations.
In conclusion, statistically significant differences in serum levels of all measured inflammatory mediators were not detected between patients with and without appendicitis. This, together with elevated levels of many mediators in the peritoneal fluid of patients with acute appendicitis, suggests that in most patients with acute appendicitis, the cytokine profile is consistent with a localized inflammatory reaction. Only in the advanced pathologic process of appendicitis (such as perforation) could a systemic reaction be detected. Our results suggest that, at present, it would be very difficult and probably impractical to rely on systemic inflammatory response markers to differentiate acute appendicitis from other causes of abdominal pain. Surgeons’ clinical skills remain the mainstay of the decision of whether to operate on a patient with suspected acute appendicitis.
Correspondence: Francis Serour, MD, Department of Pediatric Surgery, The E. Wolfson Medical Center, PO Box 5, Holon 58100, Israel (firstname.lastname@example.org).
Accepted for Publication: June 30, 2004.