Paper EPR RETROSPECTIVE DOSIMETRY WITH FINGERNAILS: REPORT ON FIRST APPLICATION CASES Francois Trompier,* François Queinnec,* Eric Bey,† Thierry De Revel,‡ Jean Jacques Lataillade,§ Isabelle Clairand,* Marc Benderitter,* and Jean-François Bottollier-Depois* Abstract—For localized irradiation to hands, in case of sources accidentally handled, it is very difficult to estimate the dose distribution by calculation. Doses may reach several tens of grays, and the dose distribution is usually very heterogeneous. Until recently, doses in such situations could be estimated only by analysis of bone biopsies using Electron Paramagnetic Resonance (EPR) spectroscopy. This technique was used previously on surgical wastes or after amputation of a finger. In this case, the dose information was available in one or a few locations on the hand only, due to the limited number of biopsy fragments usually collected. The idea to measure free radicals (FRs) induced by radiation in nails to estimate a dose is not new, but up to now, no application cases were reported. As a matter of fact, the EPR analysis of nails is complex due to the presence of intrinsic signals and parasitic signals induced by the mechanical stress (when nails are collected), which overlaps the radio-induced components. In addition, the radio-induced FRs identified up to now are unstable and very sensitive to humidity. In these conditions, it was difficult to foresee any application for dosimetry with fingernails. Recently, stable radio-induced FRs in nails has been identified and an associated protocol for dose assessment developed. This protocol has been applied by the Institut de Radioprotection et de Sûreté Nucléaire on fingernail samples from victims of three different radiological accidents that occurred between 2008 and 2012 in different places. Health Phys. 106(6):798–805; 2014 Key words: accidents, handling; biodosimetry; dose assessment; health effects
INTRODUCTION ELECTRON PARAMAGNETIC resonance (EPR) spectroscopy is a versatile and key tool for dose assessment of victims *Institut de Radioprotection et de Sûreté Nucléaire, IRSN, BP17, 92262 Fontenay-aux-Roses, France; †Hôpital d’Instruction des Armées Percy, Service d’Hématologie, BP 410, 92141 Clamart Cedex, France; ‡Hôpital d’Instruction des Armées Percy, Service de Chirurgie Plastique, BP 410, 92141 Clamart Cedex, France; and §Hôpital d’Instruction des Armées Percy, CTS Jean Julliard, BP 410, 92141 Clamart Cedex, France. The authors declare no conflicts of interest. For correspondence contact: F. Trompier, Institut de Radioprotection et de Sûreté Nucléaire, IRSN, BP17, 92262 Fontenay-aux-Roses, France, or email at [email protected]. (Manuscript accepted 23 January 2014) 0017-9078/14/0 Copyright © 2014 Health Physics Society DOI: 10.1097/HP.0000000000000110
of severe radiological accident. EPR spectroscopy is used in the frame of a multi-technique approach together with biological dosimetry and dose reconstruction tools such as numerical simulation. In this context, EPR has been used mainly on bone biopsies, specifically in the case of localized irradiation (Regulla and Deffner 1989; Schauer et al. 1996; Wu et al. 1998; Clairand et al. 2006; Trompier et al. 2007a; Clairand et al. 2008; IAEA 2004). As a matter of fact, in the case of a localized irradiation (i.e., source handlings, radiotherapy, and radiology accidents or sources in a clothes pocket), the local dose can be several orders of magnitude higher than the whole body dose. The assessment of the maximum of dose is therefore required to define the best therapeutic strategies (Tamarat et al. 2012; Benderitter et al. 2010; Bey et al. 2010). In some scenarios with a good knowledge of the accident parameters, it is possible to assess the localized dose by numerical simulation (radiotherapy and radiology accidents, sources in a clothes pocket), but in the case of sources handling, which is a very common case of accidents, the numerical approach reaches its limit due to the difficulty of reconstructing the irradiation configuration, and since the dose distribution is characterized by high gradients, especially before the appearance of any clinical signs. EPR dosimetry on bone cannot be done easily in the early management phase of victims because of the invasiveness of the sampling. Therefore, the possibility of evaluating dose on nails was already considered several decades ago (Dalgarno and McClymont 1989; Symons et al. 1995). Nails are easy to collect and give an estimation of the dose distribution if nails from each finger or toe can be analyzed independently. Great efforts have been made to understand the radicals mechanism in nails and to establish EPR nails dosimetry with different approaches in recent years (Romanyukha et al. 2007a, 2010; Reyes et al. 2008, 2009, 2012; Trompier et al. 2007 b, 2009; Black and Swarts 2010; Wilcox et al. 2010; He et al. 2011). Nevertheless, up to now, no application cases were ever reported, to the authors’ knowledge. Several difficulties were indeed major obstacles to the use www.health-physics.com
798
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
EPR retrospective dosimetry with fingernails c F. TROMPIER
of nails in dosimetry. Chandra and Symons (1987) have reported the presence of parasitic EPR signals induced by the mechanical stress when nails are cut, which overlaps the radio-induced signals (RIS). Trompier et al. (2007 b) have reported the effect of humidity on the stability of both the RIS and the so-called mechanic induced signals (MIS). The fading of the RIS and MIS is correlated to the level of humidity: the highest is the humidity content, and the fastest is the decay. As a consequence, hand washing eliminates the RIS and MIS, at least the components of the RIS and MIS that were identified at that time. As a conclusion, the dose estimation on nails by EPR could be envisaged only if the nails could be collected in a very short time after the irradiation, which considerably limits the field of application considering that the usual delay in identifying the accident and collecting and transporting the samples is at minimum on the order of days and more frequently of weeks. These two major obstacles are very probably the reason explaining the absence of application cases, whereas the accidents with local irradiation to hands are quite frequent (i.e., accidents with gammagraphy source). Recently, the finding of a stable RIS in nails, labeled RIS5, has made it possible to foresee application of nails dosimetry (Trompier et al. 2014). This newly identified RIS component, if stable, is, however, of weak intensity and has similar spectroscopic characteristics as the so called “background signal” (BKG), which is an intrinsic signal observed before irradiation. Regarding the thermal stability of the RIS5, the intensity of the RIS5 does not vary more than 20% with a isochronal thermal annealing up to 180°C, which indicates a thermal similar to the intrinsic signals; whereas for all other free radicals, the signal intensity significantly decreases at 60°C. In addition, repeated water soakings do not affect the RIS5 intensity (Trompier et al. 2014). Moreover, the RIS5 has a relation between the signal intensity and
ET AL.
799
the dose that is not linear but presents saturation behavior and a decrease of the intensity once the maximum intensity is reached. Although the variability of the saturation behavior has not been studied closely yet, the first data seem to show that for the same individual, the dose saturation occurs at the same dose for nails of different fingers collected at the same period or collected at different time periods (over 3 mo). The dose saturation behavior makes difficult the use of classical approaches used in dosimetry (calibration curve or dose additive method), which are based on the relation between the signal intensity and the dose. With RIS5, a given intensity can be related to two different values of dose. Therefore, a new approach has been developed that is based on the dose saturation characteristic. This new approach is dedicated to the identification and estimation of high doses of radiation on fingers, which is currently the case with localized irradiation to hands. This new approach has been used by IRSN for the analysis of nail samples collected from different victims of three different radiological accidents that occurred between 2008 and 2012. The aim of this paper is not to summarize all the measurements performed by IRSN for these cases, but to highlight those that were the most pertinent in the victim’s management or those that could be compared with other dose estimation (dose estimation on bone by EPR). Besides the three cases, in the same period, IRSN has participated with a joint expertise on nails with the Naval Dosimetry Center for an accident that occurred in October 2011 in Dayton, OH, with 130 kVp x-ray used for nondestructive testing. DESCRIPTION OF THE ACCIDENTS These three accidents are related to the handling of sources of 192Ir for a radiography camera used for
Fig. 1. Picture of a radiography camera: (a) the source container, (b) the ejection tube with the collimator in place for a radiography session (IAEA 2012). www.health-physics.com
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
800
Health Physics
gammagraphy application on industrial sites to check the quality of welding on pipes and tubes (Fig. 1). The first accident occurred in the city of Rades located in Tunisia on 23 March 2008. During a session of radiography, the 192Ir source (2.96 TBq) of the radiography camera stays blocked in the ejection tube. One of the manipulators, after having disconnected the ejection tube, found the source holder containing the source without identifying it as the source (Fig. 2). From his testimony, he handled the source for a duration estimated to be about 10 min, mainly in the left hand. The victim was transferred to the Hôpital d’Instruction des Armées Percy (HIAP, Clamart, France) on 2 May. The whole body dose was estimated on 5 May by biological dosimetry at 0.25 Gy (0.1–0.46 Gy) (Martin et al. 2008), which is consistent with the 0.28 Gy whole body dose estimation obtained by numerical reconstitution of the accident (Huet et al. 2008). The victim described erythema on hands with a feeling of warmth that remained for 2 wk followed by severe skin lesions (humid desquamation) (Fig. 3). Nails from the left thumb, index, and middle fingers fell out and were collected and stored at low temperature (−32°C). In 2009, the necrosis of the extremity of the bones of the first phalanx of the most exposed fingers (left index and middle finger) required surgical removal of the distal phalanxes of the extremity. The second accident occurred in Gabon on 8 November 2010 during a radiography procedure. The 192Ir source (3.24 TBq) was disconnected from the remote control. As a consequence, the source stayed in the ejection tube between two radiographies, leading to the overexposure of three workers. The ejection tube and the collimator were manipulated by the operators with the source blocked inside, leading to a whole body exposure associated with a severe irradiation to hands. The most exposed patient was transferred to the HIAP to be treated for skin lesions on the hands. Nails were collected on 29 December 2010. In 2012, the most exposed victim came back to France at the HIAP because of a sharp and persistent pain in the left thumb and signs of osteoradionecrosis. Surgical removal of the distal phalanx of the left thumb was consequently performed.
Fig. 2. Picture of two different types of sources and source holders used in radiography cameras (IRSN).
June 2014, Volume 106, Number 6
Fig. 3. Picture of the lesion observed on the hand of the main victim of the Rades accident (IRSN).
The third accident occurred in Chilca, Peru, on 12 January 2012, when the 192Ir source (3.6 TBq) in a radiography camera used to test pipe joints got disengaged and stuck inside the collimator during a 1.5‐h radiography session. As a result, five persons were overexposed to various levels of severity. Three of the workers manipulated the collimator with the source in irradiation position or worked in close vicinity to the source. In addition to a partial body irradiation, these victims were also exposed to a high level of dose to their hands. In the first days after the accident, only one worker exhibited a skin lesion (blister) localized on the left index finger. In Peru, nails were collected from the three victims as well as tooth enamel minibiopsies for the most exposed worker and shipped to EPR laboratory of the Institut de Radioprotection et de Sûreté Nucléaire (IRSN). For the most exposed person, the dose estimated by EPR analysis of nails and tooth enamel minibiopsies showed that the irradiation was more severe than expected. Consequently, this patient suffering from combined middle radio-induced aplasia and severe radiological burns; he was transferred to HIAP in France in February 2012. After his transfer to France, additional skin lesions have appeared on other fingers; the prognosis for hand lesions was bad. For the two other patients, skin lesions have also appeared lately on fingers.
www.health-physics.com
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
EPR retrospective dosimetry with fingernails c F. TROMPIER
MATERIALS AND METHODS Samples Nails were collected from the victims at different time periods after the three accidents but in a time window shorter than 2 mo. For the accident in Rades (Tunisia), the whole nails of the left index and middle finger fell out 2 mo after the irradiation. The nails were clipped from the free edge of each nail. The bone samples from the left index and the left middle finger (identified as G2 and G3) were collected 1 y later. The bone samples’ masses were, respectively, 43 and 20 mg for G2 and G3. The analysis of the bones was performed in the days following the surgery with X-band EPR technique, whereas the nails were analyzed by Q-band EPR in January 2010 (about 2 y after the accident), when a first protocol for nails was available. In the meantime, the nails were stored at −32°C. For the accident in Gabon, fingernails of the victims were collected 3 wk after the accident and stored at room temperature. The Q-band EPR measurements were performed in the 3 wk following the receipt of the collected nails. For the accident in Chilca (Peru), the nails from the three main victims were collected in Peru 3 wk after the accident and transported to France by the IAEA mission team. Samples were stored at room temperature and Q-band EPR measurements performed within the 3 wk following receipt of the samples. For the victim transferred to France at the HIAP, an amputation of the first phalanx of the left index finger was finally performed at the end of March 2012.The phalanx was collected for EPR analysis. Four mini-biopsies of a few milligrams of bones were collected at different locations of the phalanx and measured by Qband EPR to estimate the heterogeneity of the dose distribution in the phalanx bone. Sample preparation The nail clippings were, when needed, additionally cut to fit in the 3‐mm external diameter tube. Before measurements, the samples were rinsed in distilled water for 10 min and then dried in a vacuum dryer with silicagel for 15 h. For measurements, the mass of nail samples ranged between 2–5 mg. Concerning bones, the samples were cleaned in distilled water using an ultrasonic bath with a total duration of less than 30 min. For Q-band EPR measurements, the mass of bone samples ranged between 2–5 mg, whereas for X-band, the whole mass of bone samples available was used. EPR measurements EPR measurements were performed at two different microwave frequencies (X- and Q-band). EPR spectra were recorded at room temperature. For the Q-band spectrometer, measurements were performed with a Bruker EMX+
ET AL.
801
spectrometer supplied with a ER5106QT/W resonator and with a Bruker EMX supplied with a SHQ resonator for X-band. The X-band is the frequency commonly used in dosimetry application, whereas Q-band was only recently proposed for dosimetry application with very few cases (Romanyukha et al. 2007b; De et al. 2013). The advantages of Q-band for nails and calcified tissues (enamel and bones) are that it tends to be more sensitive than X-band for small mass samples (
INTRODUCTION ELECTRON PARAMAGNETIC resonance (EPR) spectroscopy is a versatile and key tool for dose assessment of victims *Institut de Radioprotection et de Sûreté Nucléaire, IRSN, BP17, 92262 Fontenay-aux-Roses, France; †Hôpital d’Instruction des Armées Percy, Service d’Hématologie, BP 410, 92141 Clamart Cedex, France; ‡Hôpital d’Instruction des Armées Percy, Service de Chirurgie Plastique, BP 410, 92141 Clamart Cedex, France; and §Hôpital d’Instruction des Armées Percy, CTS Jean Julliard, BP 410, 92141 Clamart Cedex, France. The authors declare no conflicts of interest. For correspondence contact: F. Trompier, Institut de Radioprotection et de Sûreté Nucléaire, IRSN, BP17, 92262 Fontenay-aux-Roses, France, or email at [email protected]. (Manuscript accepted 23 January 2014) 0017-9078/14/0 Copyright © 2014 Health Physics Society DOI: 10.1097/HP.0000000000000110
of severe radiological accident. EPR spectroscopy is used in the frame of a multi-technique approach together with biological dosimetry and dose reconstruction tools such as numerical simulation. In this context, EPR has been used mainly on bone biopsies, specifically in the case of localized irradiation (Regulla and Deffner 1989; Schauer et al. 1996; Wu et al. 1998; Clairand et al. 2006; Trompier et al. 2007a; Clairand et al. 2008; IAEA 2004). As a matter of fact, in the case of a localized irradiation (i.e., source handlings, radiotherapy, and radiology accidents or sources in a clothes pocket), the local dose can be several orders of magnitude higher than the whole body dose. The assessment of the maximum of dose is therefore required to define the best therapeutic strategies (Tamarat et al. 2012; Benderitter et al. 2010; Bey et al. 2010). In some scenarios with a good knowledge of the accident parameters, it is possible to assess the localized dose by numerical simulation (radiotherapy and radiology accidents, sources in a clothes pocket), but in the case of sources handling, which is a very common case of accidents, the numerical approach reaches its limit due to the difficulty of reconstructing the irradiation configuration, and since the dose distribution is characterized by high gradients, especially before the appearance of any clinical signs. EPR dosimetry on bone cannot be done easily in the early management phase of victims because of the invasiveness of the sampling. Therefore, the possibility of evaluating dose on nails was already considered several decades ago (Dalgarno and McClymont 1989; Symons et al. 1995). Nails are easy to collect and give an estimation of the dose distribution if nails from each finger or toe can be analyzed independently. Great efforts have been made to understand the radicals mechanism in nails and to establish EPR nails dosimetry with different approaches in recent years (Romanyukha et al. 2007a, 2010; Reyes et al. 2008, 2009, 2012; Trompier et al. 2007 b, 2009; Black and Swarts 2010; Wilcox et al. 2010; He et al. 2011). Nevertheless, up to now, no application cases were ever reported, to the authors’ knowledge. Several difficulties were indeed major obstacles to the use www.health-physics.com
798
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
EPR retrospective dosimetry with fingernails c F. TROMPIER
of nails in dosimetry. Chandra and Symons (1987) have reported the presence of parasitic EPR signals induced by the mechanical stress when nails are cut, which overlaps the radio-induced signals (RIS). Trompier et al. (2007 b) have reported the effect of humidity on the stability of both the RIS and the so-called mechanic induced signals (MIS). The fading of the RIS and MIS is correlated to the level of humidity: the highest is the humidity content, and the fastest is the decay. As a consequence, hand washing eliminates the RIS and MIS, at least the components of the RIS and MIS that were identified at that time. As a conclusion, the dose estimation on nails by EPR could be envisaged only if the nails could be collected in a very short time after the irradiation, which considerably limits the field of application considering that the usual delay in identifying the accident and collecting and transporting the samples is at minimum on the order of days and more frequently of weeks. These two major obstacles are very probably the reason explaining the absence of application cases, whereas the accidents with local irradiation to hands are quite frequent (i.e., accidents with gammagraphy source). Recently, the finding of a stable RIS in nails, labeled RIS5, has made it possible to foresee application of nails dosimetry (Trompier et al. 2014). This newly identified RIS component, if stable, is, however, of weak intensity and has similar spectroscopic characteristics as the so called “background signal” (BKG), which is an intrinsic signal observed before irradiation. Regarding the thermal stability of the RIS5, the intensity of the RIS5 does not vary more than 20% with a isochronal thermal annealing up to 180°C, which indicates a thermal similar to the intrinsic signals; whereas for all other free radicals, the signal intensity significantly decreases at 60°C. In addition, repeated water soakings do not affect the RIS5 intensity (Trompier et al. 2014). Moreover, the RIS5 has a relation between the signal intensity and
ET AL.
799
the dose that is not linear but presents saturation behavior and a decrease of the intensity once the maximum intensity is reached. Although the variability of the saturation behavior has not been studied closely yet, the first data seem to show that for the same individual, the dose saturation occurs at the same dose for nails of different fingers collected at the same period or collected at different time periods (over 3 mo). The dose saturation behavior makes difficult the use of classical approaches used in dosimetry (calibration curve or dose additive method), which are based on the relation between the signal intensity and the dose. With RIS5, a given intensity can be related to two different values of dose. Therefore, a new approach has been developed that is based on the dose saturation characteristic. This new approach is dedicated to the identification and estimation of high doses of radiation on fingers, which is currently the case with localized irradiation to hands. This new approach has been used by IRSN for the analysis of nail samples collected from different victims of three different radiological accidents that occurred between 2008 and 2012. The aim of this paper is not to summarize all the measurements performed by IRSN for these cases, but to highlight those that were the most pertinent in the victim’s management or those that could be compared with other dose estimation (dose estimation on bone by EPR). Besides the three cases, in the same period, IRSN has participated with a joint expertise on nails with the Naval Dosimetry Center for an accident that occurred in October 2011 in Dayton, OH, with 130 kVp x-ray used for nondestructive testing. DESCRIPTION OF THE ACCIDENTS These three accidents are related to the handling of sources of 192Ir for a radiography camera used for
Fig. 1. Picture of a radiography camera: (a) the source container, (b) the ejection tube with the collimator in place for a radiography session (IAEA 2012). www.health-physics.com
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
800
Health Physics
gammagraphy application on industrial sites to check the quality of welding on pipes and tubes (Fig. 1). The first accident occurred in the city of Rades located in Tunisia on 23 March 2008. During a session of radiography, the 192Ir source (2.96 TBq) of the radiography camera stays blocked in the ejection tube. One of the manipulators, after having disconnected the ejection tube, found the source holder containing the source without identifying it as the source (Fig. 2). From his testimony, he handled the source for a duration estimated to be about 10 min, mainly in the left hand. The victim was transferred to the Hôpital d’Instruction des Armées Percy (HIAP, Clamart, France) on 2 May. The whole body dose was estimated on 5 May by biological dosimetry at 0.25 Gy (0.1–0.46 Gy) (Martin et al. 2008), which is consistent with the 0.28 Gy whole body dose estimation obtained by numerical reconstitution of the accident (Huet et al. 2008). The victim described erythema on hands with a feeling of warmth that remained for 2 wk followed by severe skin lesions (humid desquamation) (Fig. 3). Nails from the left thumb, index, and middle fingers fell out and were collected and stored at low temperature (−32°C). In 2009, the necrosis of the extremity of the bones of the first phalanx of the most exposed fingers (left index and middle finger) required surgical removal of the distal phalanxes of the extremity. The second accident occurred in Gabon on 8 November 2010 during a radiography procedure. The 192Ir source (3.24 TBq) was disconnected from the remote control. As a consequence, the source stayed in the ejection tube between two radiographies, leading to the overexposure of three workers. The ejection tube and the collimator were manipulated by the operators with the source blocked inside, leading to a whole body exposure associated with a severe irradiation to hands. The most exposed patient was transferred to the HIAP to be treated for skin lesions on the hands. Nails were collected on 29 December 2010. In 2012, the most exposed victim came back to France at the HIAP because of a sharp and persistent pain in the left thumb and signs of osteoradionecrosis. Surgical removal of the distal phalanx of the left thumb was consequently performed.
Fig. 2. Picture of two different types of sources and source holders used in radiography cameras (IRSN).
June 2014, Volume 106, Number 6
Fig. 3. Picture of the lesion observed on the hand of the main victim of the Rades accident (IRSN).
The third accident occurred in Chilca, Peru, on 12 January 2012, when the 192Ir source (3.6 TBq) in a radiography camera used to test pipe joints got disengaged and stuck inside the collimator during a 1.5‐h radiography session. As a result, five persons were overexposed to various levels of severity. Three of the workers manipulated the collimator with the source in irradiation position or worked in close vicinity to the source. In addition to a partial body irradiation, these victims were also exposed to a high level of dose to their hands. In the first days after the accident, only one worker exhibited a skin lesion (blister) localized on the left index finger. In Peru, nails were collected from the three victims as well as tooth enamel minibiopsies for the most exposed worker and shipped to EPR laboratory of the Institut de Radioprotection et de Sûreté Nucléaire (IRSN). For the most exposed person, the dose estimated by EPR analysis of nails and tooth enamel minibiopsies showed that the irradiation was more severe than expected. Consequently, this patient suffering from combined middle radio-induced aplasia and severe radiological burns; he was transferred to HIAP in France in February 2012. After his transfer to France, additional skin lesions have appeared on other fingers; the prognosis for hand lesions was bad. For the two other patients, skin lesions have also appeared lately on fingers.
www.health-physics.com
Copyright © 2014 Health Physics Society. Unauthorized reproduction of this article is prohibited.
EPR retrospective dosimetry with fingernails c F. TROMPIER
MATERIALS AND METHODS Samples Nails were collected from the victims at different time periods after the three accidents but in a time window shorter than 2 mo. For the accident in Rades (Tunisia), the whole nails of the left index and middle finger fell out 2 mo after the irradiation. The nails were clipped from the free edge of each nail. The bone samples from the left index and the left middle finger (identified as G2 and G3) were collected 1 y later. The bone samples’ masses were, respectively, 43 and 20 mg for G2 and G3. The analysis of the bones was performed in the days following the surgery with X-band EPR technique, whereas the nails were analyzed by Q-band EPR in January 2010 (about 2 y after the accident), when a first protocol for nails was available. In the meantime, the nails were stored at −32°C. For the accident in Gabon, fingernails of the victims were collected 3 wk after the accident and stored at room temperature. The Q-band EPR measurements were performed in the 3 wk following the receipt of the collected nails. For the accident in Chilca (Peru), the nails from the three main victims were collected in Peru 3 wk after the accident and transported to France by the IAEA mission team. Samples were stored at room temperature and Q-band EPR measurements performed within the 3 wk following receipt of the samples. For the victim transferred to France at the HIAP, an amputation of the first phalanx of the left index finger was finally performed at the end of March 2012.The phalanx was collected for EPR analysis. Four mini-biopsies of a few milligrams of bones were collected at different locations of the phalanx and measured by Qband EPR to estimate the heterogeneity of the dose distribution in the phalanx bone. Sample preparation The nail clippings were, when needed, additionally cut to fit in the 3‐mm external diameter tube. Before measurements, the samples were rinsed in distilled water for 10 min and then dried in a vacuum dryer with silicagel for 15 h. For measurements, the mass of nail samples ranged between 2–5 mg. Concerning bones, the samples were cleaned in distilled water using an ultrasonic bath with a total duration of less than 30 min. For Q-band EPR measurements, the mass of bone samples ranged between 2–5 mg, whereas for X-band, the whole mass of bone samples available was used. EPR measurements EPR measurements were performed at two different microwave frequencies (X- and Q-band). EPR spectra were recorded at room temperature. For the Q-band spectrometer, measurements were performed with a Bruker EMX+
ET AL.
801
spectrometer supplied with a ER5106QT/W resonator and with a Bruker EMX supplied with a SHQ resonator for X-band. The X-band is the frequency commonly used in dosimetry application, whereas Q-band was only recently proposed for dosimetry application with very few cases (Romanyukha et al. 2007b; De et al. 2013). The advantages of Q-band for nails and calcified tissues (enamel and bones) are that it tends to be more sensitive than X-band for small mass samples (
