Diagnostic performance of serum interleukin-6 in pediatric acute appendicitis: a systematic review

vier Arredondo Montero J a · Carlos Bardají Pascual · Mónica Bronte Anaut · Natalia López-Andrés ·Giuseppa Antona · Nerea Martín-Calvo ,

Abstract

Keywords Acute appendicitis · Children · Interleukin-6 · Pediatric · Sensitivity · Specificity · Systematic review

lntroduction

Acute appendicitis is the most frequent urgent surgical pathology in pediatric patients [ 1]. Despite the advances of the last decades in relation to radiological and analytical studies in the context of the pediatric acute abdomen,the diagnosis of acute appendicitis continues to pose a significant challenge due to the wide variety of symptoms and signs that may mimic other conditions.

In relation to radiological studies, most centers have standardized the use of ultrasound as the test of choice in clinical practice. Its main advantages are its non-invasive nature, the absence of ionizing radiation, and the possibility of repeating serial examinations in emergency departments.However, it is an operator-dependent test, and in many cases,the appendicular disposition does not allow its adequate identification during the study. Computed axial tomography(CT) is reserved for selected cases, given the high radiation dose involved for particularly vulnerable patients [ 1].

The standard treatment of the acute appendicitis is surgical removal of the inflamed appendix [ 1]. However, in recent years colonoscopy has acquired both a diagnostic and therapeutic role in pediatric acute appendicitis, through the socalled endoscopic retrograde appendicitis therapy (ERAT).Recent randomized clinical studies show that ERAT is a safe and feasible procedure in uncomplicated appendicitis as an alternative to the traditional surgical approach [ 2].Conservative management with antimicrobial therapy as an alternative to surgery for early uncomplicated appendicitis has recently been studied in the pediatric population, but antimicrobial therapy is still controversial owing to its considerable failure rate, over 32% [ 1].

An incorrect diagnosis leads to removal of a normal appendix, known as a “negative appendicectomy”, in up to 6% of operations in recent pediatric series [ 3]. Negative appendicectomies expose patients to the unnecessary risk of surgical complications with health and economic implications [ 4, 5]. On the other hand, a delay in the diagnosis is often accompanied by a delay in the intervention, which entails a high risk of complications as abscess, perforation,peritonitis and even death [ 6].

During previous decades, several serum biomarkers have emerged as possible tools to improve the diagnostic performance of acute appendicitis. While most of these biomarkers have been of little or limited utility, others have shown promising results [ 7]. Blood biomarkers currently used for the diagnosis of acute appendicitis include C-reactive protein (CRP), white cell count (WCC) and neutrophil count.Recent studies have revealed that ratios derived from common markers in clinical practice of classically used markers, such as the neutrophil/lymphocyte index or the delta neutrophil index, showed promising results [ 8, 9]. Extensive research exists for the validation of calprotectin and interleukin (IL)-1 to IL-10 in clinical studies; and there is a relentless search for new biomarkers, such as leucine rich α2 glycoprotein (LRG), high-mobility group box 1 (HMBG-1) and pentraxin-3 (PTX-3), but studies are still limited by small sample sizes [ 10– 14].The diagnostic performance of IL-6, a pro-inflammatory cytokine secreted by macrophages, T cells, endothelial cells,and fibroblasts, in the treatment of acute appendicitis has been studied extensively in adult populations [ 15, 16]. In the last two decades, growing evidence suggests that it could be a valuable diagnostic tool in pediatric acute appendicitis as well. In this context, we aimed to perform a systematic review of the literature published between 2000 and 2021 on the diagnostic performance of IL-6 in pediatric acute appendicitis.

Methods

Literature search and selection

We followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidance. We specifically designed and implemented a review protocol that was registered in the international prospective register of systematic reviews (PROSPERO ID CRD42020216557).Inclusion and exclusion criteria are listed in Table 1. Eligible studies were identified by searching in main existing medical bibliography databases (Pubmed, Medline OVID,Cochrane library). Search terms used for medical subject headings and key words were: (appendicitis OR appendectomy) AND (pediatric OR children) AND (interleukin). The search was last executed on 24.08.2021.Our initial search returned 68 results that were imported to COVIDENCE, an online software for systematic review management (Covidence.org). Twenty-six of the references were excluded due to duplication. Two independent reviewers screened the title and abstract of the remaining 42 articles following the inclusion and exclusion criteria defined in the protocol. Disagreement was resolved by discussion. Finally,9 articles were reviewed in full text and were selected for data extraction. The flow chart of the literature search and the selection of articles is shown in Fig. 1.

Table 1 Inclusion and exclusion criteria

Inclusion criteria Prospective observational clinical studies measuring the accuracy of interleukin-6 relative to the reference standards for the diagnosis of appendicitis in pediatric population Diagnostic validation studies measuring the accuracy of interleukin-6 relative to the reference standards for the diagnosis of appendicitis in pediatric population Exclusion criteria Retrospective studies Case reports Reviews Languages other than English or Spanish Studies with no surgical intervention Studies with no population of interest Patients older than 21 years Studies conducted in immunocompromised patients Studies specifically conducted on patients with systemic conditions causing abdominal pain such as porphyria, sickle cell anemia and familial Mediterranean fever Studies conducted in patients with metastatic neoplastic disease and invasive abdominal neoplastic disease

Quality assessment

An analysis of the selected articles to ensure their methodological quality and to assess the risk of bias according to the QUADAS2 standards was done. We evaluated patient selection, index test, reference standard and flow and timing.

Data extract and synthesis

Fig. 1 Flow diagram of the search and selection process

Two independent reviewers extracted the relevant data from the selected articles following a standardized procedure. Extracted data included year of publication, type of study, study population (age and sample size), control group definition, mean and standard deviation (or median and range) for serum IL-6 values, statistical p value for the between group comparison, IL-6 cut-off value (if established), and sensitivity and specificity associated with it. There were no disagreements or conflicts between the reviewers after collating the extracted data.

Results

Extracted data of the selected articles are summarized in Table 2. All studies were carried out between 2006 and 2020[ 11, 17– 24]. Of the selected studies, 2 were Turkish [ 17, 21],1 was German [ 18], 1 was Slovenian [ 19], 1 was from the USA [ 20], 2 were Latvian [ 22, 24], 1 Canadian [ 11] and 1 Iranian [ 23]. All studies were prospective [ 11, 17– 24]. Eight of the studies involved exclusively pediatric population [ 11,17– 22, 24], while 1 of them included both pediatric and adult population [ 23].

This study includes a sample size of 1108 pediatric patients, of whom 600 had a confirmed diagnosis of acute appendicitis. 475 controls are included. It should be noted that in the study performed in a mixed population, it is not made clear what percentage of the pediatric patients presented acute appendicitis and what percentage of them were classified as controls. The age range is 0–21 years. The gender distribution is 641 males (57.85%) and 467 females(42.15%).

The risk of bias in relation to the selection of patients was considered low in all the evaluated articles. The risk of bias in relation to the index test was considered low in all articles evaluated. The risk of bias in relation to the reference standard was considered low in all articles evaluated. The risk of bias in relation to flow and timing was considered low in all articles evaluated. The results of the QUADAS2 analysis are shown in Fig. 2.

The definition of the “event” as the histopathological confirmation of acute appendicitis in the surgical specimen was consistent through the analyzed studies. This was not the case for the definition of the “control group”, which was constituted either by patients with non-surgical abdominal pain, mesenteric lymphadenitis, undergoing elective outpatient surgery, or with histopathological confirmed white appendectomies.

Nine publications provided serum values of IL-6 either in pg/mL or ng/L [ 11, 17– 24]. Median (range) was the most commonly used estimate to present the results. Six out of the 9 selected studies used median (range) [ 19– 22, 24],whereas 3 used mean (standard deviation) [ 11, 17, 18]. One publication did not provide measured serum values of IL-6[ 23]. Eight studies defined a specific cut-off point for serum values of IL-6 and presented results of sensitivity (from 48.1 to 96.4%) and specificity (from 10.3 to 100%). The proposed cut-off point ranged between 4.3 and 68.9 pg/mL(ng/L). Nine publications found significantly higher levels of IL-6 in the acute appendicitis group compared with the control group (whatever was the definition of control) [ 11,17– 22, 24]. The remaining publication did not report serum levels of IL-6 [ 23]. Most of the studies provided a single measure of serum levels of IL-6, but 3 studies presented stratified values of IL-6 by the histopathological appearance of the appendix (early vs. suppurative vs. perforated and non-perforated vs. perforated). Those articles showed that the higher the degree of evolution of the acute appendicitis,the higher the serum level of IL-6, which reached a peak in those patients with perforated acute appendicitis.

Table 2 Summary of the publications included in the systematic review

10.3 100 69.7 Depending on the abdominal pain onset: < 24 h: 68.5 24–48 h: 71.1 > 48 h: 58.5 71.4 Sensitivity (%) Specificity (%)96.4 75.8 73.5< 24 h:82.1 24–48 h: 95.6 > 48 h: 91.7 76.5 Cut off (pg/mL)5.4 P< 0.01 20.21< 0.05 12.2 0.02< 0.001 Depending on the abdominal pain onset: < 24 h: 11.31 24–48 h: 5.7 > 48 h: 2.5< 0.001 8 IL-6 serum values (control group) (pg/mL)234.0 ± 313.3 a 3.8 ± 4.3 a 5.0 (2.0–19.9) b 4.4 (1.3–16.1) b 3 (2–18.4) b Control group size 44 22 33 186 15 IL-6 serum values (acute appendicitis group) (pg/mL)743.5 ± 1581.7 a 15.3 ± 23.2 (5.9) [0.0–112.0] a (early appendicitis) 29.5 ± 34.4 (18.5) [0.0–109.8] a (suppu-rative appendicitis) 70.3 ± 104.9 (32.5) [9.0–472.0] a (perfo-rated appendicitis)11.8 (5.0–275.0) b 21.0 (9.8–37.0) b (non-perforated appendi-citis) 122.3 (42.2–388.6) b (perforated appen-dicitis)14.4 (2.7–35) b (non-perforated appendi-citis) 33 (6.8–1000) b (perfo-rated appendicitis)61 189 49 94 34 Title (research group) Total sample size Acute appen-dicitis sample size 105 211 82 280 49 Sequential cytokine levels in the diagno-sis of appendicitis (Türkyilmaz et al.) (2006)Diagnostic value of blood inflammatory markers for detection of acute appendicitis in children (Sack et al.) (2006)Interleukin-6 and lipopolysaccharide-binding protein in acute appendicitis in children (Groselj-Grenc et al.) (2006)Discriminative accu-racy of novel and traditional biomark-ers in children with suspected appendici-tis adjusted for dura-tion of abdominal pain (Kharbanda et al.) (2011)Are neutrophil CD64 expression and interleukin-6 early useful markers for diagnosis of acute appendicitis? (Ozguner et al.) (2011)

Table 2 (continued)

- Data not available. a Mean, standard deviation. b Median. c This is the only mixed population study included in the review. The number in parentheses reflects the sample size of the pediatric subpopulation included in the study

76,9 90.1 55 86.2 Sensitivity (%) Specificity (%)67,7 48.1 76 72.6 Cut off (pg/mL)4.3 5 P 0.005< 0 001 68.9< 0.001 24.64 IL-6 serum values (control group) (pg/mL)3.2 (3.2–3.2) b (mesen-teric lymphadenitis) 3.2 (3.2–11.7) b (ambulatory surgery group)28.0 (44.3) a 10.93 (5.89–18.53) b Control group size 43 83 20 IL-6 serum values (acute appendicitis group) (pg/mL)8 (3.2–97.6) b 1113.8 (6165.1) a 32.91 (21.45–95.36) b 29 31 79 60 (-) c 63 Title (research group) Total sample size Acute appen-dicitis sample size 74 185 80 (30) c 92 Role of serum cytokines in acute appendicitis and acute mesenteric lymphadenitis among children (Zviedre et al.) (2016)Cytokines and chemokines in pedi-atric appendicitis: a multiplex analysis of inflammatory protein mediators (Naqvi et al.) (2018)The diagnostic values of procalcitonin and interleukin-6 in acute appendicitis (Haghi et al.) (2019)Determining acute complicated and uncomplicated appendicitis using serum and urine biomarkers: interleu-kin-6 and neutrophil gelatinase-associated lipocalin (Kakar et al.) (2020)

Fig. 2 Results of the quality assessment according to the QUADAS2 criteria

Discussion

Pediatric acute appendicitis continues to pose a diagnostic challenge today. Aiming to pursue more accurate clinical diagnoses, the trend in the last two decades has been to explore the diagnostic performance of inflammatory biomarkers. In this review, we found that, although most of the included studies reported significant differences in serum levels of IL-6 between the compared groups, the reported values of sensitivity and specificity were low and inconsistent, suggesting that IL-6 should not be considered as a single test for differential diagnosis of a case of acute abdomen.

IL-6 is a multifunctional inflammatory marker whose formation depends on tumor necrosis factor-alpha (TNF-α)and interleukin-1 beta (IL-1β) [ 25]. Because IL-6 is usually greatly increased in septic patients, recent literature has revealed that it may be useful for the diagnosis of bacterial infections [ 26– 28].

The exact mechanism by which IL-6 increases in the context of acute appendicitis has not been established clearly;however, from a pathophysiological point of view, it seems to be due to the inflammatory process and neutrophil recruitment following the bacterial translocation that occurs after the parietal destructuring of the appendix [ 29, 30]. Other factors that explain the increase of IL-6 in acute appendicitis are the fact that it is an inflammatory process that involves many other pro-inflammatory cytokines, and the potential septic component that some patients present.

Among the biomarkers that are currently postulated as candidates to play a role in the diagnosis of acute appendicitis, IL-6 -together with calprotectin- are the most accessible from a logistical and economic point of view. Therefore,the potential advantages of IL-6 as a diagnostic biomarker of pediatric acute appendicitis would be its availability in most tertiary centers worldwide and its ease of processing.The most remarkable drawbacks would be its economic cost(higher than the markers used in routine clinical practice)and its low positive predictive value when considered alone.

Publications analyzing the cost both in economic and in processing time of determining serum values of IL-6 with the ELISA technique are few. Acharya et al. performed a decision analysis using a multi-criteria, decision-making methodology to assess the cost–benefit of IL-6 regarding diagnostic performance in appendicitis [ 31]. The authors concluded that,despite its acceptable diagnostic performance, determining the serum values of IL-6 had an important economic cost(£15.5 per determination) and required a considerable processing time (168 hours per determination) [ 31].

Although the articles included in this systematic review found significant difference in IL-6 values between groups,they failed to define a single cut-off to be used as a diagnostic criterion. In this systematic review, we observed a very wide range in the proposed cut-offs (from 4.3 to 68.9 pg/mL),resulting in very different values of sensitivities (from 48.1 to 96.4%) and specificities (from 10.3 to 100%). Although most of the authors stated that they opted for the cut-off point with the highest sensitivity and specificity values after a ROC curve, we attribute the observed heterogeneity to the fact that normal range of IL-6 is tremendously wide as well.

Two of the publications included in the review presented positive and negative predictive values (PPV and NPV) of IL-6 for the diagnosis of acute appendicitis [ 17, 19]. The PPV ranged from 62 to 65.9%, and the NPV, from 75 to 76.7%, which reinforces the idea that the diagnostic performance of IL-6 alone is modest. İsmet et al. correlated the preoperative values of IL-6 with the duration of the abdominal pain, and found that the maximum diagnostic yield of IL-6 was achieved 24–48 h after the onset of the abdominal pain [ 21]. Although this novel finding needs to be confirmed,it suggests the potential utility of serial IL-6 in the differential diagnosis of abdominal pain in children.

In the analyzed studies, we observed a marked consistency regarding the existence of significant differences of serum IL-6 values between groups. This, together with the biological plausibility beyond the elevation of IL-6 in response to an acute inflammation, and the possible gradient of IL-6 in relation to the time of evolution, suggest the existence of a true biological effect.

Although we did not identify important sources of bias using the QUADAS2 index [ 32], the results of the analyzed studies are limited by the small sample sizes, a suboptimal control of confounding, and the use of non-parametric statistical test, which did not allow the calculation of confidence intervals. Besides, the potential effect modification of sex and age has been studied in adult populations, but to the best of our knowledge, it has not been assessed in children [ 33].Considering that in pediatric population the normal values of some biomarkers are highly associated with sex and age,we missed stratified analyses considering a potential interaction with sex or age.

This review is not exempt of limitations. First, owing to the high heterogeneity between the analyzed studies, we were unable to provide an overall estimate for the diagnostic performance of IL-6 in pediatric acute appendicitis.Although we found a reasonable homogeneity in the study designs, in the sociodemographic distribution of the samples and in the methodology used to measure IL-6 by ELISA, we believe that factors that may have contributed to the potential heterogeneity of the series are the clinical and semiological variability in the presentation of the disease depending on the age range of the patients, the lack of consensus in the grouping of cases according to the type of appendicitis they presented (depending on the work, reference is made to suppurative, phlegmonous, gangrenous, perforated or complicated/non-complicated appendicitis) and the absence of a defined and standardized cut-off in all the studies.

Second, even though we searched in the most important sources of medical literature, we cannot deny the existence of other studies with different results that would not have been included in this review. Lastly, we did not perform any test aimed to detect the publication bias. Nevertheless, our study has several strengths because we carefully followed the PRISMA recommendations [ 34]. Our research updates and adds to the existing literature because previous evidence included a very limited number of studies, was not specific of pediatric population, and focused on the cost–benefit of IL-6 in acute appendicitis [ 31]. To the best of our knowledge, this is the first systematic review on the diagnostic performance of IL-6 in pediatric acute appendicitis establishing a general framework of knowledge for future targeted validation studies.

In conclusion, higher levels of serum IL-6 were observed in the group of children with acute appendicitis. However,the existing evidence is insufficient to define an optimal cutoff of IL-6 with acceptable predictive values. Therefore, the performance of IL-6 alone for the diagnostic of acute appendicitis in pediatric population is only modest. Nevertheless,serum values of IL-6 seemed to correlate with the time of evolution and the severity of the appendicitis.

Considering the emerging use of non-operative management of acute appendicitis, this could be a useful tool for diagnostically stratifying appendicitis into complicated and uncomplicated. Further studies are needed to confirm this classificatory potential. These studies also should consider clinical variables (e.g., fever) and their potential correlation with IL-6 levels. It would also be of interest, if this ability to stratify wasconfirmed, to develop a clinical and analytical pediatric appendicitis with IL-6 serum values score that could be standardized. Multicenter studies with a larger cohort also would make it possible to establish an optimal cut-off for the use of this molecule in routine clinical practice.

Author contributions

JAM conceptualization and study design; literature search and selection, data curation and extraction, formal analysis;investigation; methodology; project administration; resources; validation; visualization; writing–original draft; writing–review and editing. CBP literature search and selection, data curation and extraction,formal analysis; investigation; writing–original draft. MBA literature search and selection, data curation and extraction, formal analysis; investigation; writing–original draft. NLA literature search and selection, data curation and extraction, formal analysis; investigation;writing–original draft. GA literature search and selection, data curation and extraction, formal analysis; investigation; writing–original draft. NMC conceptualization and study design; literature search and selection, data curation and extraction, formal analysis; investigation;methodology; project administration; resources; validation; visualization; writing–original draft; writing–review and editing.

Funding

This review did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. None of the authors have external funding to declare. No financial or nonfinancial benefits have been received or will be received from any party related directly or indirectly to the subject of this article.

Data availability

The data used to carry out this systematic review are available upon request from the review authors.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This study did not involve the participation of human or animal subjects, and therefore was exempt from formal assessment by the ethics committee for clinical research of our center.