Are shortened aPTT values always to be attributed only to preanalytical problems?

OBJECTIVES: It has been recognized that shortened activated partial thromboplastin time (aPTT) may be caused by various preanalytical conditions. As coagulation Factor VIII is included in the in vitro intrinsic coagulation cascade measured by aPTT, we hypothesized that the shortened aPTT could be a result of elevated FVIII activity. We aimed to inspect the connection of elevated FVIII with shortened aPTT, and the possible effect inflammation has on routine laboratory parameters. METHODS: 40 patients from various hospital departments with aPTT measurement below the lower limit of the reference interval (<23.0 s) were included in the study. To compare the obtained results with aPTT measurements in the non-inflammatory state, samples from 25 volunteers (laboratory personnel) were collected. White blood cell count, C-reactive protein, aPTT, and FVIII values were measured in the control group. RESULTS: Only two samples among 40 patients with shortened aPTT (5 %) were clotted. Out of the remaining 38, 26 had FVIII activity above 150 % (upper limit of a reference interval), median value of 194 % (IQR: 143-243 %). Seven samples in the control group had shortened aPTT results (36 %). However, all coagulation samples were clot and hemolysis-free. Multiple regression identified only FVIII activity as an independent variable in predicting aPTT values (p=0.001). CONCLUSIONS: Our results support the thesis that shortened aPTT is rarely a consequence of preanalytical problems. Elevated FVIII activity causes shortened aPTT, not only in the inflammatory state but also in individuals with concentration of inflammatory markers within reference intervals.

Mixing studies for lupus anticoagulant: does it matter how we mix?

Although clear and detailed recommendation regarding the lupus anticoagulant mixing test exist, various sources of NPP are used. We decided to inspect the possible differences in mixing studies depending on the mixing media. Four types of mixing media were prepared for 45 random remnant plasma samples: standard human plasma, control plasma N, previously analyzed patient with normal coagulation values, and home-made normal pool plasma (NPP). Samples were analyzed by using Siemens Dade Actin FSL Activated PTT Reagent on BCS XP analyzer. The median aPTT values of mixing studies with commercial lyophilized NPP, with commercial IQC, as well as with a patient did not differ (26.6, 26.3, and 26.8 s, respectively). Median value of a mixing study with home-made NPP was significantly higher from the rest of the group (27.9 s) ( P  < 0.05). According to the obtained results, we decided to employ the commercial lyophilized NPP for future lupus anticoagulant mixing studies.

Verification of a 6-part differential haematology analyser Siemens Advia 2120i.

Introduction: The aim of this study was to perform a comprehensive verification of a 6-part differential haematology analyser Siemens Advia 2120i (Erlangen, Germany), prior to its routine implementation. Materials and methods: Our verification protocol included: precision (within- and between-run), estimated bias (%) as measure of trueness, which was calculated from observed and manufacturers' declared value, analytical measuring interval (AMI), carryover, confirmation of a limit of blank (LoB), determination of a limit of detection (LoD) and limit of quantitation (LoQ). The K2 ethylenediaminetetraacetic acid (EDTA) patients' leftover samples were used for verification of analyser Advia 2021i. Acceptance criteria were based on manufacturer technical specifications (Siemens), 2016 state-of-the-art criteria (Vis and Huisman), and EFLM Biological Variation Database. Results: The within- and between-run precision were acceptable for all parameters and the lowest coefficients of variation were observed for mean corpuscular volume (MCV) (0.3% and 0.6%, respectively). Estimated bias was within the acceptance criteria for all parameters except for MCV (the estimated bias was 2.2% (acceptance criteria 2.0%). AMI was confirmed for all tested parameters (r > 0.99). The carryover estimates ranged from 0.1% for platelet count (Plt) to 0.6% for red blood cell count and were within the manufacturers' specifications (≤ 1%). Manufacturers' claims for LoB were confirmed for leukocytes, erythrocytes, haemoglobin, and platelets. The estimated LoD and LoQ were 0.05 x109/L and 0.1 x109/L for white blood cell count, while for Plt values were 2 x109/L and 3 x109/L, respectively. Conclusions: Analytical performance of the Siemens Advia 2120i meets predefined quality goals and is suitable for routine use in a clinical laboratory.

Serum and urine osmolality: 8 hours, 24 hours and 1-month sample stability.

OBJECTIVES: The body of literature varies significantly regarding serum and urine osmolality stability. Therefore, our aim was to investigate the stability of serum and urine osmolality at different temperatures (room temperature (RT) 4-8 °C, -20 °C) and time conditions (8 h, 24 h, 1 month). METHODS: The stability study was conducted following the CRESS guidelines, including 40 serum and urine samples. Samples were aliquoted into three aliquots and stored as follows: primary tube stored at RT for 8 h; two capped aliquots stored at 4-8 °C for 8 h and 24 h; one aliquot stored at -20 °C for 1 month. To minimize imprecision error, serum and urine osmolality were measured by the freezing point depression method in triplicate on OSMOMAT 3000 (Gonotech, Germany) analyzer. Percentage difference (PD%) against baseline measurement was calculated. Deviations were assessed against a reference change value of 5.0%. RESULTS: The PD% for serum and urine osmolality was below 2.0% for all time/temperature conditions. For serum samples: primary tube after 8 h at RT PD% (95% CI) = 0.0% (-0.3, 0.2%); 8 h at 4-8 °C PD% (95% CI) = -0.4% (-0.7, 0.0%); 24 h at 4-8 °C PD% (95% CI) = -0.7% (-0.7, -0.6%); 1 month at -20 °C PD% (95% CI) = -2.1% (-2.4, -1.5%). For urine samples: after 8 h at RT PD% (95% CI) =0.6% (0.2, 0.9%); 8 h at 4-8 °C PD% (95% CI) = -0.2% (-0.5, 0.1%); 24 h at 4-8 °C PD% (95% CI) = -0.2% (-0.5, 0.0%); 1 month at -20 °C PD% (95% CI) = -2.0% (-3.0, -1.0%). CONCLUSIONS: Changes in osmolality for tested conditions for serum and urine samples, were within acceptance criteria. Reflex and add-on osmolality testing can be performed within the same day in samples kept at RT for 8 h in primary tube and within 24 h, in aliquoted refrigerated samples, without compromising the reliability of test results. For longer storage, samples should be kept at -20 °C.

Comparison of three different protocols for obtaining hemolysis.

OBJECTIVES: Hemolysis is associated with erroneous or delayed results. Objectives of the study were to compare four different methods for obtaining hemolysis in vitro on three different analyzers. METHODS: Hemolysis was prepared with addition of pure hemoglobin into serum pool, osmotic shock, aspiration through blood collection needle, freezing/thawing of whole blood. Biochemistry parameters were measured in duplicate at Architect c8000 (Abbott, Abbott Park, USA), Beckman Coulter AU680 (Beckman Coulter, Brea, USA) and Cobas 6000 c501 (Roche, Mannheim, Germany), according to manufacturers' declarations. Cut-off value was defined as the highest value of H index with corresponding bias lower than acceptance criteria. RESULTS: We were not able to obtain results with freezing protocol. On all three platforms, lowest number of analytes were sensitive to hemolysis at H=0.5 using method of adding free hemoglobin. When osmotic shock was used, cut-off values for the most analytes were generally met at lower values. Hemolysis significantly interfered with measurement of potassium and lactate dehydrogenase (LD) at H=0.5 on all platforms. The most of the tested analytes had the lowest acceptable H index when aspiration method was used. At the low level of hemolysis (H=0.8) glucose, sodium, potassium, chloride, phosphate, and LD were affected on all analyzers, with some additional analytes depending on the manufacturer. CONCLUSIONS: Hemolysis interference differs on different analyzers and according to protocol for obtaining hemolysis. Aspiration method was generally the most sensitive to hemolysis interference, while addition of free Hb was the most resistant.

Verification of Atellica 1500 and comparison with Iris urine analyser and urine culture.

INTRODUCTION: The aims of study were to assess: 1) performance specifications of Atellica 1500, 2) comparability of Atellica 1500 and Iris, 3) the accuracy of both analysers in their ability to detect bacteria. MATERIALS AND METHODS: Carryover, linearity, precision, reproducibility, and limit of blank (LoB) verification were evaluated for erythrocyte and leukocyte counts. ICSH 2014 protocol was used for estimation of carryover, CLSI EP15-A3 for precision, and CLSI EP17 for LoB verification. Comparison for quantitative parameters was evaluated by Bland-Altman plot and Passing-Bablok regression. Qualitative parameters were evaluated by Weighted kappa analysis. Sixty-five urine samples were randomly selected and sent for urine culture which was used as reference method to determine the accuracy of bacteria detection by analysers. RESULTS: Analytical specifications of Atellica 1500 were successfully verified. Total of 393 samples were used for qualitative comparison, while 269 for sediment urinalysis. Bland-Altman analysis showed statistically significant proportional bias for erythrocytes and leukocytes. Passing-Bablok analysis for leukocytes pointed to significant constant and minor proportional difference, while it was not performed for erythrocytes due to significant data deviation from linearity. Kappa analysis resulted in the strongest agreements for pH, ketones, glucose concentrations and leukocytes, while the poorest agreement for bacteria. The sensitivity and specificity of bacteria detection were: 91 (59-100)% and 76 (66-87)% for Atellica 1500 and 46 (17-77)% and 96 (87-100)% for Iris. CONCLUSION: There are large differences between Atellica 1500 and Iris analysers, due to which they are not comparable and can not be used interchangeably. While there was no difference in specificity of bacteria detection, Iris analyser had greater sensitivity.

Reporting of activated partial thromboplastin time (aPTT): Could we achieve better comparability of the results?

INTRODUCTION: Activated partial thromboplastin time (aPTT) is determined and reported as clotting time in seconds aPTT(s), but it is presumed that reporting results as patient-to-normal clotting time ratio, aPTT(r), could minimize within-laboratory variability. The aim of study was to investigate differences in reporting aPTT results that can affect comparability of the results among Croatian laboratories and suggest further steps for its harmonization. MATERIALS AND METHODS: The questionnaire on aPTT reporting practice was distributed to 83 laboratories through Survey Monkey application in March 2019 as the part of the first regular round of Croatian Centre for Quality Assessment in Laboratory Medicine proficiency testing. RESULTS: The survey response rate was 0.49. Majority of laboratories report aPTT results as both, seconds and ratio. Participants reported use of 23 different aPTT(s) reference intervals along with 17 different combinations of reagent/coagulometer and 25 aPTT(s) denominators of different origin for aPTT(r) calculation. Despite the same aPTT(s) results, the use of different denominators caused a dispersion of aPTT(r) results that can lead to exceeding external quality assessment performance criteria of 7%, particularly when results were compared for the same reagent group only. By applying aPTT(s) reference interval mean as denominator for calculation of aPTT(r) reference interval better concordance to harmonized one was obtained (17 vs. 27; χ2 = 3.972; P = 0.046). CONCLUSION: In order to improve comparability of the results, laboratories are advised to use mean of aPTT(s) reference interval as denominator for aPTT(r) calculation. Type of coagulometer need to be considered when evaluating aPTT proficiency test results and its currently acceptable limit of performance evaluated accordingly.

Women in sports: the applicability of common national reference intervals for inflammatory and coagulation biomarkers (HemSter Study).

INTRODUCTION: Intensive physical activity causes functional and metabolic changes in the athlete's organism. The study aimed to verify the common national available reference intervals (RIs) for common inflammatory and screening coagulation tests in a population of healthy young female athletes. MATERIALS AND METHODS: One hundred and twenty-one female athletes (age range: 16-34), from various sports disciplines (water polo, handball, volleyball, football, basketball), were included in the study. All participants completed the international physical activity short-form questionnaire. Blood samples were collected between 8-10 am, after an overnight fast, before any physical activity. Reference intervals were determined according to Clinical & Laboratory Standards Institute EP28-A3C Guidelines. RESULTS: Calculated RIs for white blood cell count (WBC), prothrombin time (PT), and activated partial thromboplastin time (APTT) ratio were in accordance with the common national RIs. Calculated RI for C-reactive protein (CRP) was lower (< 2.9 mg/L) than the proposed cut-off for a healthy population (< 5.0 mg/L). Reference interval for fibrinogen was higher (1.9-4.4 g/L), than the available RIs (1.8-3.5 g/L). D-dimer cut-off value was set at 852 µg/L fibrinogen equivalent units (FEU), higher than the proposed 500 µg/L FEU for venous thromboembolism (VTE) exclusion. CONCLUSIONS: The applicability of the available RIs for WBC count, PT, and APTT-ratio was confirmed. However, RIs for CRP and fibrinogen differed significantly than the available common national RIs for the healthy non-athletes' population. A higher cut-off for D-dimers should be extensively verified before implementation for VTE diagnosis exclusion in a group of healthy young female athletes.

Women in sports: The applicability of reference intervals for 6 commercially available testosterone immunoassays (HemSter Study).

BACKGROUND: Testosterone levels in female athletes are increased due to their physical activity and correlate with their exercise volume. We therefore hypothesized that the reference intervals (RIs) derived from the general population are not applicable for female athletes. The aim of this study was to evaluate the applicability of the given RIs for 6 commercially available testosterone immunoassays in a group of female athletes. METHODS: Our study included 121 female athletes from various sporting disciplines (water polo, handball, volleyball, football, and basketball). The physical activity score was assessed by the Short Form of the International Physical Activity Questionnaire. Total testosterone was measured in serum samples by the reference LC-MS/MS method and six different immunoassays (Abbott Architect 2nd Generation Testosterone, Beckman Coulter Access Testosterone, Roche Elecsys Testosterone II, Siemens Atellica® IM Testosterone II (TSTII), Siemens IMMULITE 2000 Total Testosterone, and Snibe MAGLUMI™ Testosterone). RESULTS: There were statistically significant differences in age (P = 0.042), weight (P = 0.001), height (P < 0.001), and BMI (P < 0.001) between athletes across different sports. Their quantitative measurements of physical activity and testosterone concentration did not differ significantly between subgroups of various sports, P = 0.167 and P = 0.181, respectively. All immunoassays had a positive absolute and relative bias, in comparison with the LC-MS/MS. The manufacturer's RI was not verified for Abbott Architect, Beckman Coulter Access, and Roche Elecsys Testosterone methods, with the highest percentage of athletes above RI for Beckman Coulter (30%). CONCLUSIONS: We demonstrated that the upper reference limit provided was too low for some young female athletes. Clinical laboratories should consider implementation of the new proposed RIs.

How well do Croatian laboratories adhere to national recommendations for laboratory diagnostics of chronic kidney disease (CKD)?

Background In 2014, the Joint Croatian Working Group (JCWG) for laboratory diagnostic of chronic kidney disease (CKD) conducted a survey across medical-biochemistry laboratories which demonstrated a large heterogeneity in this area of laboratory medicine. To ensure the tools for the standardization process, in 2017 the JCWG-CKD published the first Croatian recommendations for laboratory diagnostics of CKD. To assess the implementation process, we have repeated a survey to explore how well laboratories adhere to the recommendations. Methods An invitation to the survey was sent to all Croatian medical-biochemistry laboratories (n = 196). The questionnaire was designed in a form of 19 questions and statements, with possible multiple answers. Results The response rate was 98/196 (50.0%). The predominant method for serum creatinine measurement was the standardized compensated Jaffe method (79.2%). There was substantial decrease in the number of laboratories which measure creatinine with the non-standardized uncompensated Jaffe method, compared with the initial 2014 assessment; 7% vs. 40%, respectively. The number of the laboratories that did not report estimated glomerular filtration rate (eGFR) values decreased almost by half compared to the initial data (37.6% vs. 74.4%). However, compared to the 2014 initial assessment, a similar number of laboratories (54/98 vs. 58/80) did not measure urine albumin or protein. Conclusions The collected data showed a substantial improvement in the standardization of the serum creatinine measurement, as well as in the reporting of eGFR. However, albuminuria or proteinuria assessment is still not implemented nationwide, mainly in primary health care laboratories. This demonstrates the importance of promoting and monitoring implementation of guidelines after publication.

Retrospective Evaluation of the Diagnostic Accuracy of the Modified Alvarado Scoring System (MASS) in a Croatian Hospital.

OBJECTIVE: Diagnosing acute appendicitis (AA) is challenging and this has stimulated surgeons to develop scoring systems that could potentially decrease the rate of misdiagnosis in patients with suspected appendicitis. One of the most widely used today is the Modified Alvarado scoring system (MASS), however its sensitivity and specificity varies a great deal between studies. As a result, we wanted to assess the diagnostic accuracy of MASS retrospectively at our institution to achieve the highest possible value of sensitivity and decrease the number of false negative patients. MATERIALS AND METHODS: We retrospectively calculated MASS for all subsequent patients who had undergone an appendectomy at our institution between July 2015 and March 2017. RESULTS: In 118 out of 146 operated patients, AA was confirmed intraoperatively. There was a statistically significant difference between the average MASS score in the positive and negative appendectomy groups (6 v. 4, respectively, P<0.001), with a significantly higher number of females among the negative appendectomies (P<0.001). When lowering the cut-off to a value as low as ≥3, the sensitivity of the MASS score increased to 97.45% (95% CI: 92.7 - 99.5), thus obtaining a very low false negative rate of merely 2.55%. CONCLUSION: This retrospective diagnostic accuracy study confirmed the higher average MASS score in the group of patients with confirmed AA diagnosis. A MASS score above the proposed low cut-off value (≥3) can be a useful tool to help surgeons ruling in patients with AA in order to reduce the risk of missing diagnosis.

Escherichia coli spheroplasts in a Croatian patient misclassified by two urine sediment analysers as erythrocytes: case report.

INTRODUCTION: It has already been reported that subinhibitory concentrations of ß-lactam antibiotics can cause abnormal changes of bacterial forms, such as spheroplasts. Herein we report a case of Croatian male patient with Escherichia coli spheroplasts present in urine after treatment with tazobactam, on the tenth day of hospitalization. The aim of this report is to emphasize the inability of imaging based automated urine analysers to recognize some relatively uncommon forms of bacterial presentation in urine sediment. MATERIALS AND METHODS: During routine urine analysis, unusual particles were observed in patient urine. Urine sediment was examined by two urine analysers: Atellica 1500 (Siemens, Germany) and Iris iQ200 (Beckman Coulter, USA). Additionally, urine was sent for culture testing to Microbiology department. RESULTS: Both urine analysers didn't indicate presence of bacteria in urine sediment. Unusual particles observed on the tenth day were classified as erythrocytes by both instruments. Dipstick test showed blood trace and microscopic analysis revealed bacteria in urine. Urine culture was positive for Escherichia coli. Careful examination of urine sediment has confirmed that shapes present in urine were abnormal bacterial forms called spheroplasts. CONCLUSIONS: Imaging based automated urine analysers are not able to recognize bacterial spheroplasts in urine sediment misclassifying it as erythrocytes. Microscopic examination remains the gold standard for urines with blood trace or negative blood, in which erythrocytes are reported by urine analyser in urine sediment. Failure to identify and follow up such cases may lead to inaccurate treatment decisions and puts patient safety at risk.

Filling accuracy and imprecision of commercial evacuated sodium citrate coagulation tubes.

Current recommendations advocate that blood tubes for coagulation testing should be filled not less than 90% of their nominal filling volume, since under- or over-filling >10% may generate unreliable results of some hemostasis assays. This study was hence aimed to explore filling accuracy and precision of commercial blood tubes. Between-lot variations of 3 different lots (20 tubes per lot) of 3.2% citrate blood tubes manufactured by Becton Dickinson, Greiner and Kima were studied. One additional lot from each manufacturer was assessed in triplicate (three series of 20 tubes), to assess within-lot variation. All tubes were first weighed empty and then filled with distilled water by a syringe, under ideal filling conditions. Filled tubes were weighed again, in duplicate. For each 20 tubes series, mean bias (deviation from the ideal tube filling volume) and imprecision (coefficient of variation; CV%) were calculated. All biases were within ±10%. Within-lot and between-lot variation in filling volume was acceptable, and comprised between 0.4 and 2.4%. Greiner tubes were the most accurate (bias, -1.0 to 2.4%), followed by Kima (bias, -7.8 to -5.9%) and Becton Dickinson (bias, -9.6 to 3.3%) tubes. The highest between-lot difference was noted for Becton Dickinson tubes (up to 12.9%), followed by Greiner and Kima tubes (up to 3.4 and 1.8%, respectively). Although coagulation tubes filling accuracy was within ±10% for all three tested manufacturers, the overall bias was found to be variable among manufacturers and lots. Major effort shall be made by blood tube manufacturers for improving standardization of their products.

The diagnostic accuracy of clinical and laboratory parameters in the diagnosis of acute appendicitis in the adult emergency department population - a case control pilot study.

INTRODUCTION: The evaluation of patients with suspected appendicitis strives to identify all patients with presenting symptoms while minimizing negative appendectomy rate. The aim of the study was to identify the optimal combination of clinical and laboratory parameters that should facilitate the emergency department surgeon's definite decision. MATERIALS AND METHODS: The study group comprised 120 patients with suspicion of acute appendicitis (AA). In 60 patients the AA diagnosis was confirmed intraoperatively and by histological analysis. Clinical parameters included: appetite, vomiting, diarrhea, dysuria, signs of localized peritonitis and pain migration. Measured laboratory parameters were: C-reactive protein (CRP), complete blood count (CBC) and the urine test strip. RESULTS: The control group of patients were more likely to present following symptoms: no changes in appetite (P < 0.001), diarrhea (P = 0.009) and dysuria (P = 0.047). CRP and white blood cell count (WBC) were significantly higher in the group with confirmed AA compared to the control group (44.7 vs. 6.6, and 13.6 ± 3.9 vs. 9.0 ± 3.4, respectively; P < 0.001). The multivariate logistic regression analysis identified lack of appetite (P = 0.013), absence of diarrhea (P = 0.004), and positive finding of signs of localized peritonitis (P = 0.013), as well as WBCs (P < 0.001) and negative urine test strip results (P = 0.009) as statistically significant predictors of AA. The highest percentage of correctly classified cases (82%) was achieved by combination of common clinical exam and basic inexpensive laboratory parameters (WBCs and urine test strip). CONCLUSIONS: Acute appendicitis in the emergency setting may be successfully ruled in based on elevated WBCs and negative urine test strip in combination with signs of localized peritonitis, lack of appetite and absence of diarrhea. Since CRP did not contribute to the overall diagnostic accuracy, its use in AA diagnostic protocols is of no value.