Band 55, Heft 7

Different imaging tools, circulating endocrine markers, and fine-needle aspiration (FNA) cytology are of great importance in the diagnosis and follow-up of different thyroid and parathyroid diseases. Sometimes, however, they are conflicting or inconclusive: interestingly, measuring endocrine markers (i.e. thyroglobulin, calcitonin, parathyroid hormone) in fluids from FNA proved to be a very useful complementary diagnostic tool in such cases. The determination of endocrine markers in fluids other than serum/plasma has been developed in the last years. Although studies have reported overall satisfactory results, a good standardization of procedures has not yet been reached, and further efforts should be made in order to better define pre-analytical, analytical, and post-analytical aspects. Here we reviewed critically the literature on the measurement of FNA endocrine markers, focusing on laboratory issues, such as preparation of the sample, choice of solution, and technical features of determination of these markers. Indeed, information for use of FNA-Tg, FNA-CT, and FNA-PTH in clinical practice was also provided.

Pharmacogenomics has significantly added to our understanding of drug responses in clinical pharmacology, changing the paradigm of treatment decisions. Interrogations of both inherited and somatic variations for therapeutic purposes are increasingly being adopted in clinics, where quality control (QC) materials are required. However, for many pharmacogenomic tests, the acquisition of well-characterized QC materials is often difficult or impossible. In this review, several sources of appropriate QC materials for therapy-associated genetic testing are discussed. Among them, the novel methods for producing renewable controls that resemble patient samples are highlighted. Owing to technological complexity, more efforts are needed to develop proper controls for next-generation sequencing-based assay.

Over the past decades, there have been great advances in the understanding of the pathogenesis of venous thromboembolism (VTE) in patients with inherited and acquired thrombophilia [mainly antiphospholipid antibody syndrome (APS)]. However, a number of questions remain unanswered. Prognostic markers capable of estimating the individual VTE risk would be of great use. Microparticles (MPs) are sub-micron membrane vesicles constitutively released from the surface of cells after cellular activation and apoptosis. The effects of MPs on thrombogenesis include the exposure of phopshatidylserine and the expression of tissue factor and MPs have been described in clinical studies as possible diagnostic and prognostic biomarkers for VTE. This review will provide a novel perspective on the current knowledge and research trends on the possible role of MPs in hereditary thrombophilia and APS. Basically, the published data show that circulating MPs may contribute to the development of VTE in thrombophilic carriers, both in mild and severe states. Moreover, the presence of endothelial-MPs and platelet-MPs has been described in antiphospholipid syndrome and seems to be directly linked to antiphospholipid antibodies and not to other underlying autoimmune disorders or the thrombotic event itself. In conclusion, circulating MPs may constitute an epiphenomenon of thrombophilia itself and could be up-regulated in acute particular conditions, promoting a global prothrombotic state up to the threshold of the clinical relevant thrombotic event.

Among the red cell membrane disorders, hereditary spherocytosis (HS) is one of the most common causes of inherited hemolytic anemia. HS results from the deficiency or dysfunction of red blood cell membrane proteins, such as α spectrin, β spectrin, ankyrin, anion channel protein (Band-3 protein), protein 4.1 and protein 4.2. Conventionally, HS diagnosis is established through a series of tests, which include spherocytes identification in peripheral smear, reticulocyte count, osmotic fragility, etc. Currently, different hematological analyzers provide erythrocyte indicators that estimate the presence of spherocytes and correlate that with HS, which can be useful for disease screening. The most traditional method is the osmotic fragility (OF) test, which is labor-intensive and time-consuming to perform and presents low sensitivity and specificity values. Thus, new methods have been developed for HS diagnosis, such as flow cytometry. Current guidelines recommend the use of flow cytometry as a screening test for HS diagnosis using the eosin-5′-maleimide (EMA) binding test. Thus, HS diagnosis is the result of a collaboration between clinicians and laboratories, who should take into account the family history and the exclusion of other causes of secondary spherocytosis.

External Quality Assurance (EQA) is vital to ensure acceptable analytical quality in medical laboratories. A key component of an EQA scheme is an analytical performance specification (APS) for each measurand that a laboratory can use to assess the extent of deviation of the obtained results from the target value. A consensus conference held in Milan in 2014 has proposed three models to set APS and these can be applied to setting APS for EQA. A goal arising from this conference is the harmonisation of EQA APS between different schemes to deliver consistent quality messages to laboratories irrespective of location and the choice of EQA provider. At this time there are wide differences in the APS used in different EQA schemes for the same measurands. Contributing factors to this variation are that the APS in different schemes are established using different criteria, applied to different types of data (e.g. single data points, multiple data points), used for different goals (e.g. improvement of analytical quality; licensing), and with the aim of eliciting different responses from participants. This paper provides recommendations from the European Federation of Laboratory Medicine (EFLM) Task and Finish Group on Performance Specifications for External Quality Assurance Schemes (TFG-APSEQA) and on clear terminology for EQA APS. The recommended terminology covers six elements required to understand APS: 1) a statement on the EQA material matrix and its commutability; 2) the method used to assign the target value; 3) the data set to which APS are applied; 4) the applicable analytical property being assessed (i.e. total error, bias, imprecision, uncertainty); 5) the rationale for the selection of the APS; and 6) the type of the Milan model(s) used to set the APS. The terminology is required for EQA participants and other interested parties to understand the meaning of meeting or not meeting APS.

Background: Differences between the designs of hepatitis C virus (HCV) viral load assays can result in genotype-related variability in RNA quantification. We tested paired aliquots of plasma specimens from HCV-infected individuals using two versions (v1.0 and v2.0) of the Roche COBAS AmpliPrep/COBAS TaqMan HCV Test (CAP/CTM HCV) and noted variability between results for a subset of specimens; we then sought to determine whether discrepant results were more prevalent among specific HCV genotypes. Methods: Archived and prospectively-collected plasma samples from 114 unique patients were tested using CAP/CTM HCV v1.0 and v2.0. The HCV genotype result for each patient was determined by retrospectively reviewing laboratory records. Results: All (46/46) specimens with quantifiable viral loads from patients with genotype 1 or 2 infection had CAP/CTM HCV v1.0 and v2.0 results that were within 0.5 log 10 IU/mL; in contrast, only 3/11 (27.3%) from patients with HCV genotype 3 (mean difference, 0.56 log 10 IU/mL higher with v2.0) and 0/3 (0%) from patients with HCV genotype 4 (mean difference, 0.91 log 10 IU/mL higher with v2.0) had results within 0.5 log 10 IU/mL. Among specimens with detectable HCV RNA below the lower limit of quantification with v1.0, greater proportions of genotype 3 (4/7, 57.1%) and genotype 4 (3/4, 75.0%) specimens than genotype 1 or 2 specimens (6/30, 20.0%) had v2.0 results within the quantifiable range. Conclusions: In patients infected with HCV genotype 3, sequential CAP/CTM HCV viral load results should be compared with caution and interpreted in the context of the specific assay version used.

Background: Despite the importance of manual pipetting of fluids such as water, solutions, buffers, reagents, or biological samples in daily laboratory practice, the intra- and inter-individual imprecision of this activity has not been recently described in scientific publications. Methods: Twenty laboratory operators were randomly enrolled for this study. Imprecision of manual pipetting was estimated by asking each laboratory professional to dispense 1 mL, 100 μL or 10 μL of distilled water for 10 consecutive times with three certified pipettes into a 50-mL plastic container placed into a gravimetric balance. The weight of the water dispensed was systematically recorded for each of the 10 repeated attempts, and the inter- and intra-operator imprecision was finally calculated and expressed as coefficient of variation (CV%). Results: The mean intra-individual imprecision was 5.7% (range, 0%–11.8%) for pipetting 10 μL, 0.8% (range, 0.4%–1.9%) for pipetting 100 μL, and 0.2% (range, 0.1%–0.5%) for pipetting 1 mL. Overall, the mean inter-individual imprecision was 8.1% for pipetting 10 μL, 1.1% for pipetting 100 μL and 0.4% for pipetting 1 mL. A significantly inverse correlation was found between intra-individual pipetting imprecision and the amount of water dispensed (r = –0.80; p<0.001). No significant correlation was observed between individual pipetting performance and sex, age, qualification, and years of experience in the laboratory. Conclusions: The results of this study show that manual pipetting is plagued by a considerable intra- and inter-individual imprecision, which is inversely correlated with the amount of fluid dispensed.

Background: To maintain the best performance a frozen serum sample should be thawed once to prevent repeated freeze-thaw cycles. Archival biobanks often have one tube of a sample available, causing repeated freeze-thaw cycles when the sample is used in multiple research projects. In this study, we investigated potential effects of freeze-thaw cycles on several biochemical components in serum. Methods: Serum from 40 fasting donors of both genders, aged 30–60 years, were frozen at –25 °C. Aliquots of the 40 different samples went through 1, 2, 3, 4, 5 and 10 thaws, respectively. They were analyzed after 3 month of storage for 15 serum components including electrolytes and metabolites, proteins and enzymes, lipids, hormones and vitamins. One-way analyses of variance (ANOVA) with repeated measurements and equivalence tests were used to examine differences in component levels. Results: Albumin, aspartate-aminotransferase (ASAT), cholesterol, creatinine, C-reactive protein, glucose, immunoglobulin G, potassium, testosterone, triglycerides, urea and vitamin B12 levels did not show significant difference for pairwise comparisons after 10 repeated thaws. Although albumin, ASAT, bilirubin, potassium, sodium, testosterone and thyroid stimulating hormone (TSH) showed overall statistically significant changes in serum levels, only bilirubin, sodium and TSH were significant for the pairwise comparisons investigated. Clinical significance were shown for albumin, ASAT, bilirubin, sodium and testosterone. Conclusions: Twelve components (albumin, ASAT, cholesterol, creatinine, C-reactive protein, glucose, immunoglobulin G, potassium, testosterone, triglycerides, urea and vitamin B12) were robust to 10 repeated thaws compared to baseline level. Three components (bilirubin, sodium and TSH) showed statistical significant difference for pairwise comparisons, however, TSH was not clinically affected.

Background: Irreproducibility of scientific results constitutes an undesirably onerous economic burden and is in many cases caused by low-quality materials. Therefore, researchers are increasingly devoting their attention to the bioresources they use. In turn, those bioresources are required to validate their preanalytical processes in order to ensure best possible quality. The present study thus aimed to evaluate the impact of repeated temperature fluctuations, as they occur in most research biobanks due to repetitive opening and closing of freezer doors, on the stability of 26 biochemical analytes. Methods: Serum of 43 individuals was randomly assigned to a fluctuation (n=21) and a control group (n=22). Serum of the fluctuation group underwent controlled temperature fluctuations (30 fluctuations <−75°C – <−65°C – <−75°C under real-life freezer conditions within 21 days). Control sera were stored at constant conditions. After 10, 20, and 30 fluctuations, results derived from the fluctuation group were compared to baseline and to the control group by means of general linear models. Results: Sixteen biomarkers showed statistically significant changes over time, whereas only seven of those presented with diagnostically/clinically relevant changes at certain time points (aspartate aminotransferase, amylase, calcium, uric acid, creatinine, inorganic phosphate and total protein). However, there was no difference between the fluctuation and the control group. Conclusions: Some serum analytes are influenced by storage, even at temperatures as low as <−70°C. In contrast, we found no evidence that complex temperature fluctuations produced by storage of and access to biospecimens in biobank freezers generate any additional variability.

Background: Copeptin, part of the vasopressin precursor, is increasingly used as marker for vasopressin and is claimed to have better ex vivo stability. However, no study has directly compared the ex vivo stability of copeptin and vasopressin. Methods: Blood of ten healthy volunteers was collected in EDTA tubes. Next, we studied the effect of various pre-analytical conditions on measured vasopressin and copeptin levels: centrifugation speed, short-term storage temperature and differences between whole blood and plasma, long-term storage temperature and repeated freezing and thawing. The acceptable change limit (ACL), indicating the maximal percentage change that can be explained by assay variability, was used as cut-off to determine changes in vasopressin and copeptin. Results: The ACL was 25% for vasopressin and 19% for copeptin. Higher centrifugation speed resulted in lower vasopressin levels, whereas copeptin concentration was unaffected. In whole blood, vasopressin was stable up to 2 h at 25°C and 6 h at 4°C. In plasma, vasopressin was stable up to 6 h at 25°C and 24 h at 4°C. In contrast, copeptin was stable in whole blood and plasma for at least 24h at both temperatures. At –20°C, vasopressin was stable up to 1 month and copeptin for at least 4 months. Both vasopressin and copeptin were stable after 4 months when stored at –80°C and –150°C. Vasopressin concentration decreased after four freeze-thaw cycles, whereas copeptin concentration was unaffected. Conclusion: Vasopressin levels were considerably affected by pre-analytical conditions, while copeptin levels were stable. Therefore, a strict sample handling protocol for measurement of vasopressin is recommended.

Background: Diurnal variations and physiologic changes of urinary proteome have been suggested in the urinary proteomics field. However, no clear evidence has been demonstrated. The present study thus aimed to define changes in urinary proteome by physiological stimuli, i.e. caffeine intake and excessive water drinking, both of which cause physiologic diuresis. Methods: Urine samples were collected from 30 healthy individuals under three different conditions: (i) morning void as the control; (ii) after drinking a cup of coffee; and (iii) after drinking 1 L of water within 20 min. Thereafter, differentially excreted proteins were analyzed by 2-DE proteomics approach and validated by Western blotting and ELISA. Results: Spot matching, quantitative intensity analysis, and ANOVA followed by Tukey’s post-hoc multiple comparisons and the Bonferroni correction revealed significant differences in levels of five protein spots among three different conditions. These proteins were identified by quadrupole time-of-flight mass spectrometry (Q-TOF MS) and/or MS/MS analyses as kininogen 1 isoform 3, β-actin, prostaglandin D synthase (PGDS), fibrinogen α-chain and immunoglobulin light chain. Among these, the decreased level of immunoglobulin was successfully validated by Western blotting and ELISA. Conclusions: These data indicated that caffeine intake and excessive water drinking could affect urinary excretion of some proteins and may affect urinary proteome analysis.

Background: Measurements of autoantibodies to interferon-ω (IFN-ω) in patients with autoimmune polyglandular syndrome type 1 (APS-1) were performed using a new immunoprecipitation assay (IPA) based on 125 I-labeled IFN-ω. Methods: We have developed and validated a new IPA based on 125 I-labeled IFN-ω. Sera from 78 patients (aged 3–78 years) with clinically diagnosed APS-1, 35 first degree relatives, 323 patients with other adrenal or non-adrenal autoimmune diseases and 84 healthy blood donors were used in the study. In addition, clinical features and autoimmune regulator ( AIRE ) genotype for the APS-1 patients were analyzed. Results: Sixty-six (84.6%) of 78 APS-1 patients were positive for IFN-ω Ab using 125 I-labeled IFN-ω IPA. IFN-ω Ab was the most prevalent of the six different autoantibodies tested in this group of APS-1 patients. All 66 IFN-ω Ab-positive APS-1 patients had AIRE mutations and 7 IFN-ω Ab-negative patients had no detectable AIRE mutations, whereas 3 (3.8%) patients were discrepant for IFN-ω Ab positivity and AIRE mutation results. Out of autoimmune controls studied, two patients were positive for IFN-ω Ab. Positivity and levels of IFN-ω Ab in the APS-1 patients studied were similar irrespective of patient’s clinical phenotype and AIRE genotype. Furthermore, IFN-ω Ab levels did not change over time (up to 36 years of disease duration) in 8 APS-1 patients studied. Conclusions: We have developed a novel, highly sensitive and specific assay for measurement of IFN-ω Ab. It provides a simple and convenient method for the assessment of patients with APS-1 and selecting patients suspected of having APS-1 for AIRE gene analysis.

Background: Middle- and long-term biological variation data for hematological parameters have been reported in the literature. Within-day 24-h variability profiles for hematological parameters are currently lacking. However, comprehensive hour-to-hour variability data are critical to detect diurnal cyclical rhythms, and to take into account the ‘time of sample collection’ as a possible determinant of natural fluctuation. In this study, we assessed 24-h variation profiles for 20 hematological parameters. Methods: Blood samples were collected under standardized conditions from 24 subjects every hour for 24 h. At each measurement, 20 hematological parameters were determined in duplicate. Analytical variation (CV A ), within-subject biological variation (CV I ), between-subject biological variation (CV G ), index of individuality (II), and reference change values (RCVs) were calculated. For the parameters with a diurnal rhythm, hour-to-hour RCVs were determined. Results: All parameters showed higher CV G than CV I . Highest CV G was found for eosinophils (46.6%; 95% CI, 34.9%–70.1%) and the lowest value was mean corpuscular hemoglobin concentration (MCHC) (3.2%; 95% CI, 2.4%–4.8%). CV I varied from 0.4% (95% CI, 0.32%–0.42%) to 20.9% (95% CI, 19.4%–22.6%) for red cell distribution width (RDW) and eosinophils, respectively. Six hematological parameters showed a diurnal rhythm. Conclusions: We present complete 24-h variability profiles for 20 hematological parameters. Hour-to-hour reference changes values may help to better discriminate between random fluctuations and true changes in parameters with rhythmic diurnal oscillations.

Background: The study aim was to validate Beckman Coulter’s fully automated Access Immunoassay System (BC Access assay) for anti-Müllerian hormone (AMH) and compare it with Beckman Coulter’s Modified Manual Generation II assay (BC Mod Gen II), with regard to cycle AMH fluctuations and antral follicle counts. Methods: During one complete menstrual cycle, transvaginal ultrasound was performed on regularly menstruating women (n=39; 18–40years) every 2 days until the dominant ovarian follicle reached 16mm, then daily until observed ovulation; blood samples were collected throughout the cycle. Number and size of antral follicles was determined and AMH levels measured using both assays. Results: AMH levels measured by the BC Access assay vary over ovulatory menstrual cycles, with a statistically significant pre-ovulatory decrease from –5 to +2 days around objective ovulation. Mean luteal AMH levels were significantly lower (–7.99%) than mean follicular levels but increased again towards the end of the luteal phase. Antral follicle count can be estimated from AMH (ng/mL, BC Access assay) concentrations on any follicular phase day. BC Access assay-obtained AMH values are considerably lower compared with the BC Mod Gen II assay (–19% on average); conversion equation: AMH BC Access (ng/mL)=0.85 [AMH BC Mod Gen II (ng/mL)] 0.95 . Conclusions: AMH levels vary throughout the cycle, independently of assay utilised. A formula can be used to convert BC Access assay-obtained AMH levels to BC Mod Gen II values. The number of antral follicles can be consistently estimated from pre-ovulatory AMH levels using either assay.

Background: Primary hyperparathyroidism (pHPT) is associated with low-grade inflammation, left ventricular hypertrophy and increased cardiovascular mortality, but the association between inflammatory markers and parameters of adverse cardiac remodeling is unknown. We investigated the relationship between C-reactive protein (CRP), the essential amino acid tryptophan and its pro-inflammatory derivatives kynurenine and quinolinic acid (QUIN) with echocardiographic parameters. Methods: Cross-sectional baseline data from the “Eplerenone in Primary Hyperparathyroidism” trial were analyzed. Patients with any acute illness were excluded. We assessed associations between CRP, serum levels of tryptophan, kynurenine and QUIN and left ventricular mass index (LVMI), left atrial volume index (LAVI) and E/e′. Results: Among 136 subjects with pHPT (79% females), 100 (73%) had arterial hypertension and the prevalence of left ventricular hypertrophy was 52%. Multivariate linear regression analyses with LVMI, LAVI and E/e′ as respective dependent variables, and C-reactive protein and tryptophan, kynurenine and QUIN as respective independent variables were performed. Analyses were adjusted for age, sex, blood pressure, parathyroid hormone, calcium and other cardiovascular risk factors. LVMI was independently associated with CRP (adjusted β-coefficient=0.193, p=0.030) and QUIN (β=0.270, p=0.007), but not kynurenine. LAVI was related with CRP (β=0.315, p<0.001), kynurenine (β=0.256, p=0.005) and QUIN (β=0.213, p=0.044). E/e′ was related with kynurenine (β=0.221, p=0.022) and QUIN (β=0.292, p=0.006). Tryptophan was not associated with any of the remodeling parameters. [Correction added after online publication (22 April 2017: The sentence “Among 136 subjects with pHPT (79% females), 100 (73%) had left ventricular hypertrophy.” was corrected to “Among 136 subjects with pHPT (79% females), 100 (73%) had arterial hypertension and the prevalence of left ventricular hypertrophy was 52%.”] Conclusions: Cardiac remodeling is common in pHPT and is associated with low-grade inflammation and activation of the tryptophan-kynurenine pathway. The potential role of kynurenine and QUIN as cardiovascular risk factors may be further investigated in future studies.

Background: The detection of infectious bacteria in blood culture samples is important for diagnosis and treatment, but this requires 1–2 days at least, and is not adequate as a rapid test. Therefore, we have investigated the diagnostic ability and the optimal cutoff value of procalcitonin (PCT) and C-reactive protein (CRP) for predicting the bacteremias using receiver operating characteristic (ROC) curves and relative cumulative frequency distribution (RCD) curves. Methods: A case-control study was performed in inpatients (852 subjects: 426 positive cultures and 426 negative cultures) from January 1 to December 31, 2014. We retrospectively investigated their blood culture and blood chemistry findings recorded in this period using electronic medical records. Results: Area under the ROC curve of PCT and CRP were 0.79 and 0.66, respectively. The optimal cutoff values were 0.5 μg/L with a sensitivity of 70% and specificity of 70% for PCT and 50.0 mg/L with a sensitivity of 63% and specificity of 65% for CRP. When the optimal cutoff value was treated as a reference, the odds ratio (OR) was 71.11 and the hazard ratio (HR) was 6.27 for PCT >2.0 μg/L, and the risk of blood culture positivity was markedly elevated. PCT levels were significantly higher in the population with Gram-negative rod (GNR) infections than in the population with Gram-positive coccal (GPC) infections. Conclusions: The elevation of CRP and PCT were significantly associated with bacteremias. PCT was superior to CRP as a diagnostic indicator for predicting bacteremias, for discriminating bacterial from nonbacterial infections, and for determining bacterial species.

Background: Early recognition of patients who have developed anastomotic leakage (AL) after colorectal surgery is crucial for the successful treatment of this complication. The aim of this study was to assess the usefulness of the assessment of procalcitonin (PCT) and interleukin-6 (IL-6) levels in the prognosis of AL. Methods: This observational study included 157 patients who underwent major elective colorectal surgery. The most common indications for surgery were cancer and inflammatory bowel diseases. Serum samples were obtained directly before surgery (D0) and 1 day (D1) after surgery, and the relationships between the serum concentrations of PCT and IL-6 and development of AL were assessed. Results: In total, 10.2% of patients developed post-surgical infections due to AL. PCT levels on D1 were significantly higher in patients who developed AL [2.73 (1.40–4.62)] than in those who recovered without complications [0.2 (0.09–0.44)]. The area under the ROC curve for PCT on D1 was 0.94, 95% CI (0.89–0.98). The sensitivity and specificity of the prediction of an infection were 87% and 87%, respectively, for PCT on D1, which was higher than 1.09 ng/mL. The increase in PCT concentration between D0 and D1 was significantly higher in patients with AL (p<0.001). Patients who developed AL had higher concentrations of IL-6 on D1, but the difference was not significant (p=0.28). Conclusions: This study confirms that surgical trauma increases serum PCT concentrations and that the concentration of PCT on D1 can predict AL after colorectal surgery. However, IL-6 is not a good early marker for developing AL.

Background: As part of the immune defense during infection, an increase in enzyme activity of indoleamine 2,3-dioxygenase (IDO) leads to a breakdown of tryptophan to kynurenine. In previous animal studies, therapeutic antagonism of IDO resulted in reduced sepsis mortality. We investigated the prognostic ability of tryptophan, serotonin, kynurenine and IDO (represented by the ratio of kynurenine/tryptophan) to predict adverse clinical outcomes in patients with community-acquired pneumonia (CAP). Methods: We measured tryptophan, serotonin and kynurenine on admission plasma samples from CAP patients included in a previous multicenter trial by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). We studied their association with inflammation (C-reactive protein), infection (procalcitonin) and clinical outcome. Results: Mortality in the 268 included patients was 45% within 6 years of follow-up. IDO and kynurenine showed a strong positive correlation with markers of infection (procalcitonin) and inflammation (C-reactive protein) as well as sepsis and CAP severity scores. Tryptophan showed similar, but negative correlations. In a multivariate regression analysis adjusted for age and comorbidities, higher IDO activity and lower tryptophan levels were strongly associated with short-term adverse outcome defined as death and/or ICU admission within 30 days with adjusted odds ratios of 9.1 [95% confidence interval (CI) 1.4–59.5, p=0.021] and 0.11 (95% CI 0.02–0.70, p=0.021). Multivariate analysis did not reveal significant associations for kynurenine and serotonin. Conclusions: In hospitalized CAP patients, higher IDO activity and lower tryptophan levels independently predicted disease severity and short-term adverse outcome. Whether therapeutic modulation of IDO has positive effects on outcome needs further investigation.