Comparison of turbidimetric and chemiluminescent immunoassays in fecal calprotectin testing

Introduction

Background

Calprotectin (CP) is a calcium- and zinc-binding protein belonging to the S100 protein family, with a molecular weight of 36 kDa (1). CP is present in cells and it is highly concentrated in the cytoplasm of neutrophils. After being released into the extracellular environment, CP can be detected in various biological fluids, including plasma, urine, saliva, intestinal secretions, and stool.

CP is considered an important marker of neutrophil turnover, and its levels have been shown to increase in several inflammatory conditions (2). When measured in feces, CP specifically reflects intestinal inflammation, whereas serum markers (e.g., C-reactive protein) indicate systemic inflammation without identifying its source. For this reason, fecal calprotectin (FC) has been included in various gastroenterology studies since the 2000s, exploring its role in different inflammatory conditions (2-4). Testing for FC can be useful in diagnosing inflammatory bowel disease (IBD) and distinguishing it from irritable bowel syndrome (IBS). Thanks to its sensitivity and high negative predictive value, a negative FC result reliably rules out IBD, thus diagnosing IBS without further testing or specialist consultation (5,6). However, a positive result is not specific to IBD, requiring specialist consultation and potential further investigations (7).

Rationale and knowledge gap

Nowadays, many different assays allowing FC testing exist, including enzyme-linked immunosorbent assay (ELISA), chemiluminescent immunoassay (CLIA), fluoro-enzyme immunoassays (FEIA) and particle-enhanced turbidimetric immunoassay (PETIA). Thanks to the ease of sample preparation and the reduced measuring time, CLIA and turbidimetric assays are the most widespread methods for FC testing (8). Although multiple assays exist, analytical differences between methods could influence FC values, potentially affecting clinical interpretation and patient management. Comparison between different assays may be warranted in specific situations, such as when a laboratory transitions to a new analytical platform, when patients are followed across centers employing distinct methods, or when historical results generated with alternative assays must be interpreted. In these instances, method-comparison studies can help ensure reliable results in routine laboratory practice, supporting continuity and reliability of clinical assessment. Moreover, it would be important to evaluate the advantages that a simple and accurate method may offer to laboratory organisation.

Current FC assays, often require multiple manual steps, including sample pre-treatment, batching, and hands-on pipetting. These procedures can lead to delays in result reporting, increased turnaround time, and higher risk of pre-analytical errors. Additionally, the need for batching and limited automation reduces laboratory flexibility and can create workflow bottlenecks, particularly in high-volume settings. Such inefficiencies may compromise timely clinical decision-making and increase operational costs.

In this context, the Calprest Turbo PETIA emerges as a rapid, automation-friendly option for FC testing, allowing the use of a dedicated instrument for fecal matrix rather than one that is also employed for serum immunochemical assays. These characteristics make it a valuable candidate for further investigation of its analytical performance and its potential contribution to improving routine workflows.

Objective

Since no study has directly compared the PETIA Calprest Turbo assay and the chemiluminescent Liaison Calprotectin assay, our aim was to evaluate the correlation between these two different methods for FC testing. Simultaneously, we aimed to assess the feasibility of employing a dedicated instrument for fecal testing, improving sample processing and laboratory organisation. We present this article in accordance with the MDAR reporting checklist (available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-56/rc).

Methods

This comparative study considered a total of 200 stool samples with routine CP assay request which were received between October 2024 and May 2025 at the Unit of Chemistry, Biochemistry and Molecular Biology of “A. Gemelli” Hospital Foundation IRCCS, in Rome. No clinical information was given on the tested population.

FC was quantitatively determined using the Calprest Turbo (Eurospital, Trieste, Italy) PETIA whose performances were compared to the Liaison Calprotectin (DiaSorin, Saluggia, Italy) CLIA, which is currently used in our laboratory for FC routine analysis. Assay specifications are outlined in Table 1.

We compared these assays considering FC concentrations <50 µg/g as negative, FC >120 µg/g as positive, and intermediate FC values as slightly positive, as indicated by the respective manufacturers.

Fresh stool samples were stored at 2–8 ℃ until examination with either Liaison Calprotectin and Calprest Turbo assays.

In order to minimise pre-analytical variability, all samples were extracted simultaneously in accordance with the respective manufacturers’ instructions. Although fecal extracts are stable when stored at room temperature, refrigerated, or frozen (Table 1), in the present study, CP was measured immediately after the extraction procedure, using both immunoassays on the same day, to further reduce pre-analytical variability.

Kit features

Liaison Calprotectin, DiaSorin

Liaison Calprotectin is a sandwich chemiluminescent assay involving two monoclonal antibodies (mAb) for capture and detection of FC: an isoluminol-conjugated mAb recognizes CP, which is previously bound to paramagnetic particles coated with a CP-specific mAb. The flash chemiluminescent reaction initiated by the starter reagents is measured by a photomultiplier as relative light units (RLUs) which are used to automatically calculate CP concentration, expressed in µg/g (Figure 1A).

Figure 1 DiaSorin and Eurospital technologies overview. (A) DiaSorin sandwich chemiluminescence immunoassay. Capture mAbs on paramagnetic particles recognise the fecal calprotectin which is in turn bound by isoluminol-conjugated detection mAbs. The addition of the starter reagents determines a chemiluminescent reaction which is measured in RLUs. (B) Eurospital particle-enhanced turbidimetric immunoassay. Calprotectin-specific antibodies on polystyrene latex particles recognise the analyte in the sample. The subsequent agglutination reaction increases the turbidity of the solution which is used to calculate calprotectin concentration. This figure was created using Microsoft PowerPoint. mAb, monoclonal antibody; RLU, relative light unit.

A two-point calibration was performed using lyophilised manufacturer’s calibrators, Liaison Calibrator 1 and 2 (Table 1), aligned automatically with the master curve.

The internal quality control was performed in each analytical session using the Liaison Calprotectin control set, level 1 and 2 (Table 1), in lyophilised form.

Calprest Turbo, Eurospital

Calprest Turbo is a PETIA based on the use of polystyrene latex particles coated with CP-specific antibodies. The binding between these particles and the CP in the fecal sample initiates an agglutination reaction, thereby increasing the turbidity of the solution. The degree of turbidity is directly proportional to CP concentration, which is expressed in µg/g and automatically calculated from the absorbance observed at 450 nm (Figure 1B).

We used a six-point calibration curve and two-level internal quality control samples, Calprest Turbo level 1 and 2 (Table 1), supplied as ready to use.

Sample preparation

Liaison Calprotectin, DiaSorin

After returning to room temperature, stool samples were prepared using the Liaison Calprotectin Quantitative Stool Extraction and Test (Q.S.E.T.) Device Plus kit. Through the sampling devices provided with grooves, 10.5 mg of feces were collected from different points of the sample. For liquid samples, 12 µL were collected using a pipette and directly transferred into the extraction tube. The samples were homogenised in a multi-tube vortex for 30 minutes, then analysed as a single batch on the Liaison XL analyser according to the manufacturer’s instructions.

Calprest Turbo, Eurospital

Stool samples were collected with the EasyCal Turbo kit; 24 mg of feces were collected from different point of the sample with grooved sampling devices. For liquid samples, 24 µL were collected and directly transferred into the extraction tube. In order to properly homogenise the content, each tube was vortexed for 60 seconds and centrifuged for 10 minutes at 4,000 ×g. Calibrators, controls and samples were analysed in a single batch on the Eu-Turbo Analyzer according to the manufacturer’s instructions.

Statistical analysis

All statistical analyses were performed using Medcalc Software version 23.2.1 (MedCalc Software Ltd., Ostend, Belgium). The number of samples included in the study was determined with a priori power calculation. Based on the literature, a clinically relevant mean difference (Δ) of 90 µg/g and a standard deviation (SD) of differences of 300 µg/g, derived from previously reported limits of agreement, were used (9,10). Using a significance level of 0.05 and a power of 0.80, the minimum required number of paired samples was calculated to be 70.

We used the Shapiro-Wilk test and the Wilcoxon signed-rank test to evaluate data distribution. The correlation between Liaison Calprotectin and Calprest Turbo was determined by nonparametric Passing-Bablok regression analysis while Cohen’s kappa coefficient was used to estimate the correlation degree. Cohen’s kappa values were interpreted as follows: ≥0.8 as excellent, 0.79–0.60 as substantial, 0.59–0.40 as moderate, 0.39–0.20 as slight, and <0.20 as poor (11). The significance level was determined by the associated P value <0.05.

No artificial intelligence was used in the writing of the manuscript, production of images or graphical elements, or in the collection and analysis of data.

Ethics

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Our study was conducted on residual samples from routine laboratory analyses; therefore, no informed consent was required.

Results

Liaison and Calprest are based on different assay methods, CLIA and PETIA respectively, with different calibration points and linearity ranges. We verified the precision of the two methods by performing replicates of Liaison Calprotectin control set (level 1: 33.4–59.4 µg/g; level 2: 158–282 µg/g) and Calprest Turbo quality controls (level 1: 145–270 µg/g; level 2: 240–446 µg/g). For Liaison we obtained an intra-assay coefficient of variation (CV) of 2.1% and an inter-assay CV of 6%; Calprest showed an intra-assay CV of 1.9% and an inter-assay CV of 3.7%. The intra-assay precision for both Liaison Calprotectin and Calprest Turbo was further verified by performing five replicates in the same analytical run of samples reporting different FC results: the negative sample was 15 µg/g, while the borderline and the positive samples were 72 and 250 µg/g, respectively.

For the three tested levels, Calprest Turbo and Liaison Calprotectin showed an overall intra-assay CV of 2.4% and 2.9%, respectively. Given the longer stability of the refrigerated extracted samples, inter-assay precision was evaluated only for Calprest Turbo. We calculated an overall inter-assay CV of 4.6% by performing replicates for five days of samples reporting negative, borderline and positive FC results. As evidenced by the Shapiro-Wilk test, our data did not follow a normal distribution, thus Wilcoxon signed-test rank was used. We analysed a total of 200 samples, considering separately 22 samples with FC concentrations >800 µg/g obtained with Liaison. Despite the unequivocal pathological significance of these values, we chose to analyse samples with FC concentrations above 800 separately, because Liaison performs dilutions for values exceeding 800 µg/g, while the Eurospital system only dilutes samples with FC values above 1,500 µg/g. Regarding the remaining 178 samples, no significant difference was observed between the median FC values obtained by Liaison Calprotectin and Calprest Turbo, respectively 37.5 µg/g [95% confidence interval (CI): 25.25–57.15] and 35.8 µg/g (95% CI: 20–40) (P=0.2). The Passing-Bablok analysis showed the following regression equation y = 15.108 + 0.683x, with 95% CI for intercept 12.86 to 16.78 and for slope 0.60 to 0.85 (Figure 2). The regression equation highlights the presence of a proportional difference between the two measurement methods. We observed that for low FC concentrations Calprest Turbo reports higher values than Liaison Calprotectin. Furthermore, Calprest Turbo tends to progressively underestimate CP values compared to Liaison, particularly at concentrations approaching the upper limit of 800 µg/g and above. Precision at FC concentrations >800 µg/g was not evaluated, as such values are unequivocally clinically positive and beyond the primary scope of this study, which focused on result agreement between methods.

Figure 2 Passing-Bablok regression plot. Regression line equation (solid blue line): y = 15.108 + 0.683x; 95% CI (dashed red line) for intercept 12.86 to 16.78 and for slope 0.60 to 0.85. The equality line is represented by the solid red line. CI, confidence interval.

Regarding those 22 samples showing FC concentrations >800 µg/g obtained with Liaison, the median CP values were significantly different, 775.9 µg/g (95% CI: 674.6–874.26) and 1,785 µg/g (95% CI: 1,497.38–4,799.37) and for Calprest Turbo and Liaison Calprotectin, respectively (P<0.001), showing a negative trend, proportional to the magnitude of the measurement. The bias appears to increase at higher concentrations.

Our primary aim was to assess the concordance of the results based on their clinical significance. We evaluated misclassification across three clinically relevant FC ranges: <50 µg/g (negative), >120 µg/g (positive), and intermediate values (slightly positive), according to the respective manufacturers. We did not analyze separate sub-ranges above 120 µg/g, as at these high values the numerical differences, including the negative bias of Calprest Turbo at concentrations above 800 µg/g, do not affect clinical classification or the assessment of inflammation severity.

The inter-rater reliability analysis revealed a substantial agreement between Liaison Calprotectin assay and Calprest Turbo assay with a Cohen’s kappa coefficient equal to 0.69 (95% CI: 0.61 to 0.78).

Table 2 shows the number of samples by diagnostic category and summarizes the classification consistency between the two methods by reporting the corresponding positive, negative, and overall agreement for clinically relevant FC ranges.

The results obtained from the study indicated an initial concordance rate of 81.5% between the methods, however, our analysis specifically focused on the 37 discordant cases (18.5%). Considering the diagnostic significance of FC levels, samples with slightly positive or positive results by Liaison and Calprest, and vice versa, were considered concordant. Accordingly, we recalculated the percentage of concordant and discordant cases taking into account the significance of FC levels. In the end, the true discordant cases turned out to be only 25, reducing the discordance rate to 12.5%. Out of these cases, only five were false negatives (2.5%). Consequently, based on the clinical interpretation, we obtained a concordance of 87.5% between the two methods.

Discussion

FC is a sensitive biomarker of intestinal inflammation, capable of differentiating between chronic IBD and functional disorders such as IBS, thereby reducing the need for colonoscopies.

The aim of the present study was to evaluate the performance of Calprest Turbo assay for FC measurement, which offers the advantage of operating on a dedicated analytical platform to the fecal matrix.

The analytical precision evaluation showed no substantial differences from the CV% values declared by the manufacturers, indicated in Table 1, and both methods displayed excellent accuracy in terms of repeatability and reproducibility. As reported by the manufacturers, the stability of refrigerated Liaison samples was limited to only 6 hours, therefore the inter-assay precision was assessed exclusively for Calprest extracts, which remained stable for up to 5 days at 2–8 ℃.

No significant difference was observed between the median FC values obtained by Liaison and Calprest Turbo for FC concentrations <800 µg/g. The negative bias shown by the Passing-Bablok regression analysis indicated that Calprest Turbo generally provides lower FC concentrations than Liaison, with a higher discrepancy for elevated values. The tendency of Calprest Turbo to underestimate CP concentrations was particularly evident for FC values >800 µg/g, although their pathological significance remained unchanged.

A significant bias at elevated FC concentrations has also been reported by previous studies comparing several methods for FC measurement, including the Liaison Calprotectin assay and PETIA-based methodologies (12-14). These studies reported an acceptable qualitative correlation but a significant bias for increasing FC concentrations obtained with the Liaison assay and the PETIA-based method. In contrast with our findings, these studies revealed that PETIA methods overestimated FC concentrations. In this regard, it should be noted that the PETIA method evaluated in these studies was not the Calprest Turbo assay. Consequently, the observed discrepancies could be explained, among other potential factors, by the use of antibodies directed against different complexes of the FC protein (14). This finding further highlights the heterogeneity of FC measures, based on the different methodology.

Furthermore, External Quality Assessment (EQA) reports demonstrate the different distribution of analytical results for the same EQA sample when processed using different methodologies. Consequently, beyond pre-analytical factors, it is important to compare laboratory data based on clinical interpretation criteria and to advise patients to consistently attend the same laboratory, where assays are performed using the same analytical method. With this approach a more consistent and reliable evaluation of results trend would be ensured.

These findings support the importance of selecting FC detection methods based on their specific analytical characteristics and intended clinical use. Importantly, clinicians should be fully aware of the limitations and particular features of the assay used in their laboratory when interpreting results.

In addition to the analytical concordance, our primary objective was to assess the agreement between the two methods based on the clinical significance of FC concentrations. Specifically, we achieved an initial concordance rate of 81.5%, subsequently increased to 87.5%, with only five false negatives. The increase in concordant cases was achieved by considering slightly positive results as positive, since slightly positive monitored patients are advised to repeat FC testing after 4–6 weeks. Similarly, screened patients without symptoms are suggested to repeat the test after 4–8 weeks or after 6–12 months if FC concentrations remain borderline, as indicated by the guidelines provided by the American College of Gastroenterology (ACG) and the European Crohn’s and Colitis Organisation (ECCO) (15,16).

We statistically analysed the results using Cohen’s kappa and obtained a value of k=0.69, confirming the substantial concordance between the two methods.

The few discordant results and in particular 2.5% of false negative cases could be attributed to pre-analytical variables such as different extraction procedures and to the intrinsic samples heterogeneity, which could contribute to the discrepancies in FC concentrations (17-19). Furthermore, as we did not have access to patients’ clinical data, we were unaware of the severity of their intestinal diseases, which may have contributed to the few cases of disagreement between the methods. It is important to note, however, that the aim of our study was not to assess the ability of FC to correlate with clinical, endoscopic, or histological severity. Nonetheless, incorporating patients’ clinical data in future studies could help to better explain any discrepancies between the two methods and determine whether discordant results are related to the patients’ clinical condition.

In addition to a good clinical correlation and relatively competitive costs, Calprest Turbo offers numerous practical advantages: sample preparation times are shorter (10 minutes) than Liaison ones (30 minutes) and the analytical response time is reduced from 45 minutes for Liaison to just 10 minutes for Calprest. This is crucial for allowing a faster and continuous monitoring of gastrointestinal diseases treatment.

Conclusions

In conclusion, Calprest Turbo showed an excellent reliability in terms of qualitative interpretation, thus representing a competitive FC assay method especially if used in combination with other fecal assays on the same instrument, considering that the Eu-Turbo Analyzer could allow the simultaneous detection of Helicobacter pylori, fecal occult blood and elastase with significantly reduced analysis time. Therefore, the implementation of other fecal assays in addition to CP and the use of a single instrument specifically for the fecal matrix could improve sample flow and organisational management of the laboratory. While this study did not allow evaluation of method consistency according to patients’ clinical diagnoses, the findings provide important insights into the performance and practical advantages of Calprest Turbo. These aspects offer valuable opportunities for future research, including the assessment of assay consistency in different clinical subgroups and laboratory settings, which could further enhance the generalizability and clinical applicability of FC measurements.

Acknowledgments

No artificial intelligence was used in the writing of the manuscript, production of images or graphical elements, or in the collection and analysis of data.

Footnote

Reporting Checklist: The authors have completed the MDAR reporting checklist. Available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-56/rc

Data Sharing Statement: Available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-56/dss

Peer Review File: Available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-56/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jlpm.amegroups.com/article/view/10.21037/jlpm-2025-1-56/coif). S.B. serves as an unpaid editorial board member of Journal of Laboratory and Precision Medicine from May 2025 to April 2027. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work 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. Our study was conducted on residual samples from routine laboratory analyses; therefore, no informed consent was required.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Andersson KB, Sletten K, Berntzen HB, et al. The leucocyte L1 protein: identity with the cystic fibrosis antigen and the calcium-binding MRP-8 and MRP-14 macrophage components. Scand J Immunol 1988;28:241-5. [Crossref] [PubMed]
2. Gisbert JP, McNicholl AG. Questions and answers on the role of faecal calprotectin as a biological marker in inflammatory bowel disease. Dig Liver Dis 2009;41:56-66. [Crossref] [PubMed]
3. Johne R, Fagerhol MK, Lyberg T, et al. Functional and clinical aspects of the mielomonocyte protein calprotectin. Mol Pathol 1997;50:113-23. [Crossref] [PubMed]
4. Jukic A, Bakiri L, Wagner EF, et al. Calprotectin: from biomarker to biological function. Gut 2021;70:1978-88. [Crossref] [PubMed]
5. Kristensen V, Klepp P, Cvancarova M, et al. Prediction of endoscopic disease activity in ulcerative colitis by two different assays for fecal calprotectin. J Crohns Colitis 2015;9:164-9. [Crossref] [PubMed]
6. Kumar A, Steinkasserer A, Berchtold S. Interleukin-10 influences the expression of MRP8 and MRP14 in human dendritic cells. Int Arch Allergy Immunol 2003;132:40-7. [Crossref] [PubMed]
7. Lin JF, Chen JM, Zuo JH, et al. Meta-analysis: fecal calprotectin for assessment of inflammatory bowel disease activity. Inflamm Bowel Dis 2014;20:1407-15. [Crossref] [PubMed]
8. Goll R, Heitmann R, Moe ØK, et al. Head to head comparison of two commercial fecal calprotectin kits as predictor of Mayo endoscopic sub-score and mucosal TNF expression in ulcerative colitis. PLoS One 2019;14:e0224895. [Crossref] [PubMed]
9. van Wassenaer EA, Diederen K, van Leeuwen EMM, et al. Can 2 Different Fecal Calprotectin Assays be Used Interchangeably in IBD Treatment? J Clin Gastroenterol 2022;56:e27-30. [Crossref] [PubMed]
10. Noebauer B, Ramic L, Konstantin A, et al. Analytical evaluation of a fully automated immunoassay for faecal calprotectin in a paediatric setting. Biochem Med (Zagreb) 2017;27:030710. [Crossref] [PubMed]
11. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159-74. [Crossref] [PubMed]
12. Handley SA, Dote NP, Wanandy T, et al. Verification of the Bühlmann fCAL turbo faecal calprotectin assay on the Binding Site Optilite benchtop analyser. Pract Lab Med 2023;36:e00318. [Crossref] [PubMed]
13. Lin L, Wyness SP, Jensen R, et al. Comparison of Next-Generation Assays for Fecal Calprotectin vs the PhiCal Assay. Am J Clin Pathol 2022;157:252-6. [Crossref] [PubMed]
14. Oyaert M, Boel A, Jacobs J, et al. Analytical performance and diagnostic accuracy of six different faecal calprotectin assays in inflammatory bowel disease. Clin Chem Lab Med 2017;55:1564-73. [Crossref] [PubMed]
15. Lichtenstein GR, Loftus EV, Afzali A, et al. ACG Clinical Guideline: Management of Crohn's Disease in Adults. Am J Gastroenterol 2025;120:1225-64. [Crossref] [PubMed]
16. Kucharzik T, Taylor S, Allocca M, et al. ECCO-ESGAR-ESP-IBUS Guideline on Diagnostics and Monitoring of Patients with Inflammatory Bowel Disease: Part 1. J Crohns Colitis 2025;19:jjaf106. [Crossref] [PubMed]
17. Juricic G, Brencic T, Tesija-Kuna A, et al. Faecal calprotectin determination: impact of preanalytical sample treatment and stool consistency on within- and between-method variability. Biochem Med (Zagreb) 2019;29:010707. [Crossref] [PubMed]
18. Oyaert M, Trouvé C, Baert F, et al. Comparison of two immunoassays for measurement of faecal calprotectin in detection of inflammatory bowel disease: (pre)-analytical and diagnostic performance characteristics. Clin Chem Lab Med 2014;52:391-7. [Crossref] [PubMed]
19. Padoan A, D'Incà R, Scapellato ML, et al. Improving IBD diagnosis and monitoring by understanding preanalytical, analytical and biological fecal calprotectin variability. Clin Chem Lab Med 2018;56:1926-35. [Crossref] [PubMed]