Introduction

51Cr is a radioactive isotope of chromium having a half-life of 27.7 days. It decays 91 % of the time by electron capture directly to the ground state of 51V without gamma rays emission. Only 9 % of 51Cr decays directly to the excited state of the daughter (51mV), which then further decays by isomeric transition to the ground state and emitting a 320 keV gamma ray during the process [1, 2].

At present, 51Cr is used in many application fields, in particular, in medicine [3, 4]. On one side, due to its chemical properties, 51Cr is an excellent tool for the studies with labeled red blood cells in order to measure their survival time, mass or volume. In sequestration studies labeled platelets are used for the diagnosis of gastrointestinal bleeding [5]. On the other side, gamma emissions from 51Cr present an external dose hazard [6]. The retention of 51Cr in the body (biological half-life) is very dependent on its chemical form (Cr (III) or Cr (IV)). According to ICRP [7], 5 % of the uptake is transferred to bones (and retained with a biological half-life of 1000 days), 30 % is directly excreted and 65 % is distributed to other organs and tissues in the body.

Nowadays, 51Cr activity is usually determined by conventional γ-spectrometry [8–10]. In spite of the fact that the method needs almost no (or minimum) sample preparation steps, careful attention for the calibration geometry should be paid. For instance, the calibration should be performed with a homogenised calibration standard (for blood samples this is not always easy). The limits of detection achieved by γ-spectrometry for 51Cr are normally in the range of 10 Bq g−1 depending on the type of detectors. Moreover, for such analysis relatively long counting times are required [11, 12]. However, for some clinical studies, where the incorporated doses are sometimes restricted, a method with better detection capability would be preferable.

Since 51Cr decays by electron capture, it could also be detected using liquid scintillation counting (LSC) [13, 14]. Due to its simplicity, sensitivity, and relative simple sample preparation, this method could be a suitable alternative for γ-spectrometry in order to determine the 51Cr in blood samples.

In the present study we developed a robust analytical method for the measurement of 51Cr in blood samples in routine mode by means of LSC. The method was validated with spiked blood samples from different subjects.

Experimental

Reagents and sample solutions

All acids and reagents used for the development and validation of the current testing method were of analytical grade or better: 65 % v v−1 HNO3 (Suprapur, Merck, Darmstadt, Germany); 10 % v v−1 H2O2 (Suprapur, Merck, Darmstadt, Germany). A standard certified solution of 51Cr (Cr (III) in 0.2 M HCl) was supplied by Eckert & Ziegler Nuclitec GmbH Braunschweig, Germany (Certificate Nr. 85449-443). For all dilutions, ultrapure water (18.2 MΩ cm) from Merck Millipore (Darmstadt, Germany) water system was used.

For all experiments in this work, blood samples from different subjects spiked with a known amount of 51Cr standard solution were used.

Instrumentation and methods

In order to reduce the influence of matrix on the LSC measurements as well as to solubilise the Cr, all samples were digested using a micro-wave digestion system MARS (CEM, Germany). one ml of blood sample, one ml of HNO3 and one ml H2O2 were mixed and digested at 180°C for 30 min. Afterwards, the sample solution was neutralised with two ml of NH3. No residue of the sample was observed.

The measurements in this study have been performed using a LSC HIDEX 300SL (FCI, Mainz, Germany). This instrument incorporates a triple-PMT detector technology facilitating absolute activity counting without external radioactive source using triple-to-double coincidence ratio (TDCR) [15, 16]. For the measurements Ultima Gold LLT cocktail was added to digested blood samples in a sample to cocktail ratio of 5–10. High density polyethylene plastic vials from HDPE of 20 ml were used in order to ensure a low background as well as a high counting efficiency. The measurements were done without quench curve corrections. Typical TDCR values for 51Cr in blood samples observed in the current study were in the range of 0.5–0.6. The samples were counted 3600 s.

For comparative measurements of 51Cr activity, γ-spectrometric system equipped with HPGe detectors (Ortec, UK) as well as whole body counter with HPGe detectors (Canberra, USA) installed at Competent Incorporation Monitoring Body Jülich (Jülich, Germany) were used.

Results and discussion

Sensitivity

To study the sensitivity of the developed method laboratory standards were prepared by spiking of one ml aliquots of a same blood sample with known activities of 51Cr. In total, eight samples were prepared, comprising of one blank, one sample of 67 ± 4 Bq, three samples of 200 ± 10 Bq, one sample of 670 ± 20 Bq, one sample of 1000 ± 30 Bq and one sample of 1370 ± 35 Bq of 51Cr. The measured instrument response is shown on the Fig. 1.

Fig. 1
figure 1

Sensitivity response for 51Cr activity obtained with LSC Hidex 300SL

Full size image

A linear function fits the measured activity data with the regression line y = 0.0347x − 0.0011 very well and a correlation coefficient of R 2 = 0.991 was obtained.

A detection capability of the method was determined in accordance to DIN ISO 11929 (assuming k α = 3 and k β = 1645). A detection limit of 0.5 Bq g−1 for 51Cr in blood samples was achieved with a counting time of 3600 s.

Accuracy

To determine the accuracy, four test samples (accuracy samples) with known 51Cr activities (67 ± 4, 200 ± 10, 670 ± 20, and 1000 ± 30 Bq), prepared independently from the sensitivity samples, were measured. For further quality assurance the same experiments were additionally repeated after 1 and 3 days. The results are presented in Table 1.

Table 1 Difference in 51Cr activity from its reference value determined with the developed LSC procedure in spiked blood samples
Full size table

A very good correlation between the added amount of 51Cr and the measured values was observed. The difference from the reference value was between −12 and 5 %.

Precision

In order to access the precision of the measurements a coefficient of variation (CV) for the accuracy samples was evaluated using Eq. (1):

$${\text{CV = }}\frac{\text{SD}}{\text{MEAN}}{ \times } 1 0 0$$
(1)

where CV is the coefficient of variation, SD is the standard deviation of the measurements and MEAN is the mean value of the measurements

The same samples were measured 10 times and the coefficient of variation was calculated. The results summarized in Table 2 show that the precisions (CV) of the measurements ranged from 1.3 to 5.3 % and were dependent on the activity of the samples.

Table 2 Coefficients of variation for the different activities of 51Cr measured by developed LSC procedure
Full size table

To determine a “between run precision” (agreement between individual experimental results obtained with the same method on identical test material or samples, under the same conditions) [17] all the accuracy samples were measured again after 1 and 3 days (see Table 2). As shown in the table no significant difference in the precisions was observed.

Stability

The stability (time-resolved) of 51Cr measurements by means of LSC was determined for two blood samples spiked with the activities of 51Cr of 775 ± 30 and 1510 ± 70 Bq, respectively. These samples were repeatedly measured during a selected period of time. The observed curves for the two measured samples were corrected for the physical decay of 51Cr. After the correction (see Fig. 2b) the long term precision of the measurements was calculated to be 0.7 and 0.8 % for the 775 ± 30  and 1510 ± 70 Bq of 51Cr samples, respectively. This proved an additional insurance of reliable determination of 51Cr in blood samples.

Fig. 2
figure 2

Response of the LSC instrument for two laboratory-prepared standards with the activity of 51Cr of 775 ± 30 and 1510 ± 70 Bq, respectively, measured during the selected period of time. The results are corrected with respect to the decay of 51Cr

Full size image

Comparative measurements

To further validate the developed method comparative determination of 51Cr activity in two synthetically prepared blood samples using three independent techniques: LSC, conventional γ-spectrometry and whole body counter (WBC) were performed. These techniques utilize different analytical principles for 51Cr analysis (LSC measure the electron capture of 51Cr, while γ-spectrometry and WBC analyze its γ-emission). The data of these measurements are presented in Fig. 3. A good agreement between the results of all testing methods was obtained. They corresponded to the expected activities of 51Cr in both measured samples within the measurements error of applied techniques.

Fig. 3
figure 3

Comparative measurement of 51Cr activity in laboratory-prepared blood samples 1 and 2 using three different independent testing methods: LSC, γ-spectrometry and WBC

Full size image

Conclusions

In this work we demonstrated the capability of the developed testing procedure for the determination of 51Cr in blood samples by means of LSC. The method meets all quality requirements for the reliable determination of 51Cr in blood samples. Performed studies on sensitivity, accuracy, precision and stability fit the acceptance criteria and confirm the developed procedure being applicable for the measurements of 51Cr in blood samples. Comparative measurements using different (independent) methodologies such as conventional γ-spectrometry as well as WBC additionally prove the competency of developed LSC procedure for the determination of 51Cr in blood samples. The uncertainty for these measurements was not exceeding 12 %. A limit of detection of 0.5 Bq g−1 for 51Cr in blood samples was obtained.