T R A N S F U S I O N C O M P L I C AT I O N S Screening of blood donors for chronic Coxiella burnetii infection after large Q fever outbreaks Ed Slot,1 Boris M. Hogema,1 Michel Molier,1 and Hans L. Zaaijer1,2
BACKGROUND: The Netherlands experienced major Q fever outbreaks from 2007 through 2009. An increasing number of human chronic Q fever cases has been reported in the affected area. Blood donors unaware of chronic Coxiella burnetii infection might be infectious for transfusion recipients. Local blood donations were screened for serologic signs of chronic Coxiella infection. STUDY DESIGN AND METHODS: From August 2012 through January 2013, a total of 2490 serum samples were collected from all consenting blood donors in the most affected Q fever outbreak area and screened for Phase II anti-Coxiella immunoglobulin G antibodies using an enzyme-linked immunosorbent assay (ELISA). (Phase II antibodies are considered indicative for resolved or ongoing Coxiella infection.) Reactive samples were confirmed by quantitative immunofluorescent Phase I and II antibody testing. A Phase I antibody titer of at least 1024 was considered indicative for chronic Coxiella infection. For 179 donors archived samples from 2009 and 2010 were available to study the long-term course of Coxiella antibodies. RESULTS: A total of 110 of 2490 donors were confirmed positive for Phase II Coxiella antibodies (4.4%; 95% confidence interval, 3.7%-5.3%), of which 79 were reactive for Phase I antibodies, with a maximum titer of 256. In 15 of 24 donors (62.5%), testing positive for Phase II antibodies in 2009 and 2010, ELISA reactivity had declined to negativity in 2012 and 2013. CONCLUSION: After large Q fever outbreaks in the Netherlands, no sign of potentially infectious chronic Coxiella infection was found among blood donors in the most affected area. Using an ELISA for detection, Coxiella antibodies in previously exposed donors waned quickly.
Q
fever is a zoonotic disease caused by the bacterium Coxiella burnetii. The bacterium can cause acute and chronic infections, with and without clinical symptoms. Domesticated ruminants are commonly recognized as the main reservoir for human infection. From 2007 through 2009 the Netherlands experienced seasonal Q fever outbreaks of unprecedented scale, resulting in more than 3500 notified cases of Q fever in humans.1 Abortion waves in dairy goat farms have been found to be the primary source of the Q fever outbreaks, whereas nondairy sheep farms were involved to a lesser extent.2,3 After veterinary control measures, including compulsory vaccination of goats and dairy sheep, large-scale culling of pregnant goats at infected farms and the implementation of a nationwide monitoring program to timely detect C. burnetii DNA in bulk tank milk of dairy goat farms, the Q fever outbreaks subsided by the end of 2009. In recent years an increased number of chronic Q fever cases has been reported by hospitals located in the former outbreak area; most of the patients diagnosed with chronic Q fever did not recall acute Q fever and were unaware of being infected with C. burnetii.4 It is unknown whether, and to what extent, chronic Coxiella infection among donors poses a threat to the safety of blood products in the Netherlands.
ABBREVIATION: IFA = immunofluorescence assay. From the 1Departments of Blood-borne Infections and Virology, Sanquin Blood Supply Foundation; and 2Medical Microbiology (CINIMA), Academic Medical Center, Amsterdam, the Netherlands. Address reprint requests to: Ed Slot, Department of Blood-borne Infections, Sanquin Blood Supply Foundation, Plesmanlaan 125, 1066CX Amsterdam, the Netherlands; e-mail: [email protected]. Funded by Sanquin Blood Supply Foundation Received for publication November 4, 2013; revision received March 20, 2014, and accepted March 30, 2014. doi: 10.1111/trf.12749 © 2014 AABB TRANSFUSION 2014;54:2867-2870. Volume 54, November 2014 TRANSFUSION
2867
SLOT ET AL.
C. burnetii occurs in two antigenic phases, called Phase I and Phase II. The diagnosis of C. burnetii infection relies mainly on serology, namely, the detection of antibodies to Phase I and Phase II antigens. Using an immunofluorescence assay (IFA) as the preferred method for antibody detection, a persisting immunoglobulin G (IgG) antibody titer of at least 1024 to Phase I antigens is considered to be indicative for chronic infection. Final diagnosis of chronic Q fever relies on confirmation with polymerase chain reaction (PCR) and/or assessment of clinical symptoms, risk factors, and echocardiographic imaging.5-7 Chronic Q fever predominantly occurs in patients with underlying valvular heart disease, vascular aneurysms, vascular grafts, or immune deficiencies and in pregnant women.4,8 However, chronic Q fever also occurs in individuals who were initially healthy and cannot recall an episode of acute Q fever.4 Since no reliable data exist on the incidence of chronic C. burnetii infection after acute, subclinical infection in the Netherlands, the number of Dutch citizens at risk for developing chronic Q fever is unknown.9 Because chronic Q fever is an intravascular infection, C. burnetii may be present in circulating blood. Indeed blood from chronic Q fever patients is often positive in the PCR for C. burnetii DNA.5 Thus, until chronically infected donors show clinical symptoms of chronic Q fever, they may pose a threat to the safety of blood. Therefore, 3 years after the largest community outbreaks of Q fever ever recorded in history, we screened blood donors living in the area with the highest incidence of acute Q fever during the outbreak years for serologic signs of chronic C. burnetii infection.
MATERIALS AND METHODS From August 2012 through January 2013, serum samples from 2490 consenting blood donors were collected at two blood collection sites, Oss and Veghel, located in the center of the area with the highest Q fever incidence between 2007 and 2010 (see Fig. 1). All 2557 donors who showed up to donate blood at the two collection sites were asked to participate. Sixty-seven donors did not provide consent to participate in research, resulting in a participation rate of 97.2%. The participating 2490 donors represent 58.0% of the total pool of blood donors registered at the two collection sites. Using an automated platform (Freedom EVOlyzer, Tecan Benelux BVBA, Giessen, Germany), the collected samples were tested for presence of IgG antibodies to Phase II of C. burnetii by an enzyme-linked immunosorbent assay (ELISA classic C. burnetii Phase II IgG, Serion, Mediphos, Renkum, the Netherlands). Borderline and positive samples were subsequently tested for IgG antibody titers to Phase I and Phase II using an IFA (Q fever IFA IgG, Focus Diagnostics, Cypress, CA). A Phase I 2868
TRANSFUSION Volume 54, November 2014
A
B
Collection site A (Oss)
Collection site B (Veghel)
n = 2490
IgG phase II (ELISA)
IgG phase I (IFA)
IgG phase I titer (IFA)
95.6% nonreactive
4.4% confirmed reactive
28.2% nonreactive
71.8% reactive
97.5% ≤ 64
2.5% > 64
0.0% > 256
Fig. 1. (A) Origin of donor samples. Blood collection sites Oss (A) and Veghel (B) are located in the area where Q fever outbreaks occurred from 2007 through 2009. Blue circles indicate surrounding areas serving the collection sites. Darker colors indicate higher incidence of acute Q fever in 2009 (adapted from Roest et al.2). (B) C. burnetii Phase II and Phase I IgG antibody reactivity in donors from collection sites A and B.
IgG IFA titer of at least 1024 was considered indicative for potential chronic Coxiella infection.5 All assays were performed in accordance with the manufacturer’s instructions. Test results of a previous study, performed in 2009 and 2010 among donors living in the postal code areas with the highest number of acute Q fever notifications per 1000 inhabitants in 2009,10 were used to compare the antiCoxiella seroprevalence in 2009 to 2010 and in 2012 to 2013, as well as to study the long-term course of Coxiella antibodies in serial donor samples. A total of 397 donors who participated in this study were residing in the area where the 2009 to 2010 study was conducted. Samples from 179 donors participating in this study had also been included in the 2009 to 2010 study. Two ELISA kit lots were used for donor sample screening. The initially used kit lot (SKA.CD) showed an abnormal high number of borderline and positive test results
SCREENING DONORS FOR CHRONIC Q FEVER
3.0
TABLE 1. Age and sex distribution of participating donors Female 207 210 317 366 194 1294
Male 88 141 299 414 254 1196
2.5 Total 295 351 616 780 448 2490
2.0 OD/CO
Age group (years) 18-29 30-39 40-49 50-59 60-69 Total
1.5 1.0 0.5 0.0
TABLE 2. C. burnetii Phase I IgG IFA titers in blood donors, positive for Phase II IgG (n = 110) Phase I IgG IFA titer Negative (256
Number (%) of donors 31 (28.2) 75 (68.2) 2 (1.8) 1 (0.9) 1 (0.9) 0 (0.0)
that could not be confirmed by IFA, deviant from prior experience with other lots. These findings indicated a poor specificity of lot SKA.CD. This was later confirmed by the manufacturer and was also noted in a study of postmortem samples of tissue donors.11 The problem only related to the specificity; the sensitivity of lot SKA.CD was at least as good as that from other lots. To prevent inclusion of false-positive data, all samples with a borderline or positive ELISA result in lot SKA.CD were retested using a newly produced kit lot. Retesting confirmed the reduced specificity of the first lot used; initial borderline and positive results were excluded from final analysis. Archived samples that tested confirmed positive for IgG Phase II antibodies to C. burnetii in 2010 were retested using the new lot, confirming that the sensitivity of the new lot was consistent with the lots used in the previous study.
RESULTS To monitor chronic Coxiella infection in blood donors after large outbreaks of Q fever, donor samples were collected in 2012 at two blood collection sites in the center of the area with high Q fever incidence between 2007 and 2010 (see Table 1 and Figure 1). Screening by ELISA for IgG Phase II antibodies to C. burnetii showed that 110 of 2490 donors were reactive (4.4%; 95% confidence interval [CI], 3.7%-5.3%). For all 110 donors Phase II antibody reactivity was confirmed by IFA, in line with previous findings showing that the IFA is the more sensitive test.12 In 79 of the 110 donors (71.8%; 95% CI, 62.3%-79.8%) Phase I IgG antibodies were detected by IFA (see Table 2). Four donors showed a Phase I titer of more than 32; no sample exceeded a Phase I titer of 256. A total of 397 donors of this study live in the area where we conducted a study on acute Q fever in 2009 to
2009
2010
2011
2012
2013
Fig. 2. Long-term course of Phase II IgG Coxiella antibodies as detected by ELISA (Serion) in blood donors who were confirmed positive in 2009 to 2010 (n = 24). The color of the symbols indicates the qualitative results (black = positive; gray = borderline; white = negative). Nota bene: the borderline cutoff value is kit lot dependent. OD/CO = sample-to-cutoff ratio.
2010.10 Twenty-two of these 397 donors (5.5%; 95% CI, 3.6%-8.4%) tested positive for Phase II IgG antibodies, while in the previous study we observed a significantly higher seroprevalence of 12.2% (95% CI, 9.6%-15.3%). A total of 179 donors were included in both studies. Twentyfour of these 179 donors (13.4%; 95% CI, 8.9%-19.5%) tested confirmed positive in 2009 to 2010. Only nine of these 24 donors were still seroreactive by ELISA for Phase II IgG antibodies in 2012 to 2013; in 15 of 24 donors (62.5%; 95% CI, 40.8%-80.5%) Phase II IgG antibodies had declined to negativity (see Fig. 2). To confirm this waning of antibodies and to exclude artifacts caused by varying sensitivity of antibody detection, the 24 positive samples from 2009 to 2010 were retested in 2012. Negligible differences in sample-to-cutoff ratios were observed, showing that the sensitivity of antibody detection in 2010 and in the current study is comparable. The Phase II IgG antibody titers also declined strongly when determined by IFA. However, only three of the 23 serial donor samples that could be tested (of 24 donors) turned negative in the follow-up sample, demonstrating the superior sensitivity of IFA over the ELISA for Q fever antibody detection.
DISCUSSION From 2007 through 2009 the Netherlands experienced seasonal Q fever outbreaks of an unprecedented scale. Primary infection with C. burnetii, both symptomatic and asymptomatic, may result in chronic Coxiella infection, which may become manifest years after primary infection. Hence in a previously affected area the safety of blood may be compromised by donors unknowingly incubating for chronic Q fever. This study is the first to examine asymptomatic chronic C. burnetii infection in blood Volume 54, November 2014 TRANSFUSION
2869
SLOT ET AL.
donors in a former outbreak area. Among 2490 at-risk donors tested no serologic evidence was found for chronic Coxiella infection (0%; 95% CI, 0%-0.2%), confirming that chronic C. burnetii infection is not a common condition in exposed, healthy adults. Of course the study results cannot be extrapolated to children, persons with medical conditions, and elderly persons. A striking decline of ELISA Phase II IgG antibody reactivity was found soon after initial infection. In a previous study we observed a seroprevalence of 12.2% among blood donors during the 2009 Q fever outbreak,10 while 3 years later only 4.4% seroreactivity remained. Longitudinal comparison of Phase II IgG ELISA reactivity in 24 donors who participated in both studies confirmed this finding. Consequently Phase II Coxiella IgG antibody detection by ELISA rapidly becomes unsuitable for conducting seroprevalence studies after Coxiella exposure; testing by immunofluorescence seems more appropriate. Blood donors who are unaware of infection may transmit C. burnetii to recipients, but transmission by blood products has not been demonstrated indisputably. Only one probable case of C. burnetii transmission via transfusion of blood products has been reported; this case was related to a donor with acute infection.13 Transmission of C. burnetii via transfusion of blood products could not be demonstrated in a lookback exercise, concerning transfused blood products from 13 Dutch donors with confirmed acute C. burnetii infection around the time of donation.14 In another Dutch study a possible case of transmission could neither be confirmed nor ruled out.10 There are no reports of donors with chronic C. burnetii infection who infected their recipients. It seems likely that blood donors who are asymptomatically infected with C. burnetii, both in the acute and in the chronic state, only pose a limited threat to the safety of blood. In conclusion, Phase II IgG Coxiella antibodies, as detected by an ELISA, waned rapidly in the years after acute infection. Serologic signs of chronic C. burnetii infection were absent in blood donors living in an area with recent severe outbreaks of Q fever, which supports the EU directive allowing the reentry of donors 2 years after resolved infection.15 As an additional safety measure such reentry should only be implemented for donors testing negative for Phase I IgG Coxiella antibodies.
REFERENCES 1. van der Hoek W, Dijkstra F, Schimmer B, et al. Q fever in the Netherlands: an update on the epidemiology and control measures. Euro Surveill 2010;15:19520. 2. Roest HI, Tilburg JJ, van der Hoek W, et al. The Q fever epidemic in The Netherlands: history, onset, response and reflection. Epidemiol Infect 2011;139:1-12. 3. Dijkstra F, van der Hoek W, Wijers N, et al. The 2007-2010 Q fever epidemic in The Netherlands: characteristics of notified acute Q fever patients and the association with dairy goat farming. FEMS Immunol Med Microbiol 2012; 64:3-12. 4. Wegdam-Blans MC, Stokmans RA, Tjhie JH, et al. Targeted screening as a tool for the early detection of chronic Q fever patients after a large outbreak. Eur J Clin Microbiol Infect Dis 2013;32:353-9. 5. Wegdam-Blans MC, Kampschreur LM, Delsing CE, et al. Chronic Q fever: review of the literature and a proposal of new diagnostic criteria. J Infect 2012;64:247-59. 6. Kampschreur LM, Oosterheert JJ, Koop AM, et al. Microbiological challenges in the diagnosis of chronic Q fever. Clin Vaccine Immunol 2012;19:787-90. 7. Kampschreur LM, Wever PC, Wegdam-Blans MC, et al. Defining chronic Q fever: a matter of debate. J Infect 2012; 65:362-3. 8. Kampschreur LM, Dekker S, Hagenaars JC, et al. Identification of risk factors for chronic Q fever, the Netherlands. Emerg Infect Dis 2012;18:563-70. 9. van der Hoek W, Schneeberger PM, Oomen T, et al. Shifting priorities in the aftermath of a Q fever epidemic in 2007 to 2009 in The Netherlands: from acute to chronic infection. Euro Surveill 2012;17:20059. 10. Hogema BM, Slot E, Molier M, et al. Coxiella burnetii infec-
11.
12.
13.
ACKNOWLEDGMENTS
14.
We thank Annemieke de Fijter and the staff of Sanquin Blood Collection Centers for collecting the blood samples. We thank Ed Bakker for performing the serologic screening and Bert Mesman for data extraction from the blood bank information system.
15.
CONFLICT OF INTEREST The authors have disclosed no conflicts of interest. 2870
TRANSFUSION Volume 54, November 2014
tion among blood donors during the 2009 Q-fever outbreak in the Netherlands. Transfusion 2012;52:144-50. Van Wijk MJ, Maas DW, Renders NH, et al. Screening of post-mortem tissue donors for Coxiella burnetii infection after large outbreaks of Q fever in The Netherlands. BMC Infect Dis 2014;14:6. Herremans T, Hogema BM, Nabuurs M, et al. Comparison of the performance of IFA, CFA, and ELISA assays for the serodiagnosis of acute Q fever by quality assessment. Diagn Microbiol Infect Dis 2013;75:16-21. Anonymous. Q-fever transmitted by blood transfusion— United States. CDC public health notes 1977. van Kraaij MG, Slot E, Hogema BM, et al. Lookback procedures after postdonation notifications during a Q fever outbreak in the Netherlands. Transfusion 2013;53:716-21. Commission of the European Communities. Commission Directive 2004/33/EC of 22 March 2004 implementing Directive 2002/98/EC of the European Parliament and of the Council as regards certain technical requirements for blood and blood components. OJ 2004;L91:25-39.
BACKGROUND: The Netherlands experienced major Q fever outbreaks from 2007 through 2009. An increasing number of human chronic Q fever cases has been reported in the affected area. Blood donors unaware of chronic Coxiella burnetii infection might be infectious for transfusion recipients. Local blood donations were screened for serologic signs of chronic Coxiella infection. STUDY DESIGN AND METHODS: From August 2012 through January 2013, a total of 2490 serum samples were collected from all consenting blood donors in the most affected Q fever outbreak area and screened for Phase II anti-Coxiella immunoglobulin G antibodies using an enzyme-linked immunosorbent assay (ELISA). (Phase II antibodies are considered indicative for resolved or ongoing Coxiella infection.) Reactive samples were confirmed by quantitative immunofluorescent Phase I and II antibody testing. A Phase I antibody titer of at least 1024 was considered indicative for chronic Coxiella infection. For 179 donors archived samples from 2009 and 2010 were available to study the long-term course of Coxiella antibodies. RESULTS: A total of 110 of 2490 donors were confirmed positive for Phase II Coxiella antibodies (4.4%; 95% confidence interval, 3.7%-5.3%), of which 79 were reactive for Phase I antibodies, with a maximum titer of 256. In 15 of 24 donors (62.5%), testing positive for Phase II antibodies in 2009 and 2010, ELISA reactivity had declined to negativity in 2012 and 2013. CONCLUSION: After large Q fever outbreaks in the Netherlands, no sign of potentially infectious chronic Coxiella infection was found among blood donors in the most affected area. Using an ELISA for detection, Coxiella antibodies in previously exposed donors waned quickly.
Q
fever is a zoonotic disease caused by the bacterium Coxiella burnetii. The bacterium can cause acute and chronic infections, with and without clinical symptoms. Domesticated ruminants are commonly recognized as the main reservoir for human infection. From 2007 through 2009 the Netherlands experienced seasonal Q fever outbreaks of unprecedented scale, resulting in more than 3500 notified cases of Q fever in humans.1 Abortion waves in dairy goat farms have been found to be the primary source of the Q fever outbreaks, whereas nondairy sheep farms were involved to a lesser extent.2,3 After veterinary control measures, including compulsory vaccination of goats and dairy sheep, large-scale culling of pregnant goats at infected farms and the implementation of a nationwide monitoring program to timely detect C. burnetii DNA in bulk tank milk of dairy goat farms, the Q fever outbreaks subsided by the end of 2009. In recent years an increased number of chronic Q fever cases has been reported by hospitals located in the former outbreak area; most of the patients diagnosed with chronic Q fever did not recall acute Q fever and were unaware of being infected with C. burnetii.4 It is unknown whether, and to what extent, chronic Coxiella infection among donors poses a threat to the safety of blood products in the Netherlands.
ABBREVIATION: IFA = immunofluorescence assay. From the 1Departments of Blood-borne Infections and Virology, Sanquin Blood Supply Foundation; and 2Medical Microbiology (CINIMA), Academic Medical Center, Amsterdam, the Netherlands. Address reprint requests to: Ed Slot, Department of Blood-borne Infections, Sanquin Blood Supply Foundation, Plesmanlaan 125, 1066CX Amsterdam, the Netherlands; e-mail: [email protected]. Funded by Sanquin Blood Supply Foundation Received for publication November 4, 2013; revision received March 20, 2014, and accepted March 30, 2014. doi: 10.1111/trf.12749 © 2014 AABB TRANSFUSION 2014;54:2867-2870. Volume 54, November 2014 TRANSFUSION
2867
SLOT ET AL.
C. burnetii occurs in two antigenic phases, called Phase I and Phase II. The diagnosis of C. burnetii infection relies mainly on serology, namely, the detection of antibodies to Phase I and Phase II antigens. Using an immunofluorescence assay (IFA) as the preferred method for antibody detection, a persisting immunoglobulin G (IgG) antibody titer of at least 1024 to Phase I antigens is considered to be indicative for chronic infection. Final diagnosis of chronic Q fever relies on confirmation with polymerase chain reaction (PCR) and/or assessment of clinical symptoms, risk factors, and echocardiographic imaging.5-7 Chronic Q fever predominantly occurs in patients with underlying valvular heart disease, vascular aneurysms, vascular grafts, or immune deficiencies and in pregnant women.4,8 However, chronic Q fever also occurs in individuals who were initially healthy and cannot recall an episode of acute Q fever.4 Since no reliable data exist on the incidence of chronic C. burnetii infection after acute, subclinical infection in the Netherlands, the number of Dutch citizens at risk for developing chronic Q fever is unknown.9 Because chronic Q fever is an intravascular infection, C. burnetii may be present in circulating blood. Indeed blood from chronic Q fever patients is often positive in the PCR for C. burnetii DNA.5 Thus, until chronically infected donors show clinical symptoms of chronic Q fever, they may pose a threat to the safety of blood. Therefore, 3 years after the largest community outbreaks of Q fever ever recorded in history, we screened blood donors living in the area with the highest incidence of acute Q fever during the outbreak years for serologic signs of chronic C. burnetii infection.
MATERIALS AND METHODS From August 2012 through January 2013, serum samples from 2490 consenting blood donors were collected at two blood collection sites, Oss and Veghel, located in the center of the area with the highest Q fever incidence between 2007 and 2010 (see Fig. 1). All 2557 donors who showed up to donate blood at the two collection sites were asked to participate. Sixty-seven donors did not provide consent to participate in research, resulting in a participation rate of 97.2%. The participating 2490 donors represent 58.0% of the total pool of blood donors registered at the two collection sites. Using an automated platform (Freedom EVOlyzer, Tecan Benelux BVBA, Giessen, Germany), the collected samples were tested for presence of IgG antibodies to Phase II of C. burnetii by an enzyme-linked immunosorbent assay (ELISA classic C. burnetii Phase II IgG, Serion, Mediphos, Renkum, the Netherlands). Borderline and positive samples were subsequently tested for IgG antibody titers to Phase I and Phase II using an IFA (Q fever IFA IgG, Focus Diagnostics, Cypress, CA). A Phase I 2868
TRANSFUSION Volume 54, November 2014
A
B
Collection site A (Oss)
Collection site B (Veghel)
n = 2490
IgG phase II (ELISA)
IgG phase I (IFA)
IgG phase I titer (IFA)
95.6% nonreactive
4.4% confirmed reactive
28.2% nonreactive
71.8% reactive
97.5% ≤ 64
2.5% > 64
0.0% > 256
Fig. 1. (A) Origin of donor samples. Blood collection sites Oss (A) and Veghel (B) are located in the area where Q fever outbreaks occurred from 2007 through 2009. Blue circles indicate surrounding areas serving the collection sites. Darker colors indicate higher incidence of acute Q fever in 2009 (adapted from Roest et al.2). (B) C. burnetii Phase II and Phase I IgG antibody reactivity in donors from collection sites A and B.
IgG IFA titer of at least 1024 was considered indicative for potential chronic Coxiella infection.5 All assays were performed in accordance with the manufacturer’s instructions. Test results of a previous study, performed in 2009 and 2010 among donors living in the postal code areas with the highest number of acute Q fever notifications per 1000 inhabitants in 2009,10 were used to compare the antiCoxiella seroprevalence in 2009 to 2010 and in 2012 to 2013, as well as to study the long-term course of Coxiella antibodies in serial donor samples. A total of 397 donors who participated in this study were residing in the area where the 2009 to 2010 study was conducted. Samples from 179 donors participating in this study had also been included in the 2009 to 2010 study. Two ELISA kit lots were used for donor sample screening. The initially used kit lot (SKA.CD) showed an abnormal high number of borderline and positive test results
SCREENING DONORS FOR CHRONIC Q FEVER
3.0
TABLE 1. Age and sex distribution of participating donors Female 207 210 317 366 194 1294
Male 88 141 299 414 254 1196
2.5 Total 295 351 616 780 448 2490
2.0 OD/CO
Age group (years) 18-29 30-39 40-49 50-59 60-69 Total
1.5 1.0 0.5 0.0
TABLE 2. C. burnetii Phase I IgG IFA titers in blood donors, positive for Phase II IgG (n = 110) Phase I IgG IFA titer Negative (256
Number (%) of donors 31 (28.2) 75 (68.2) 2 (1.8) 1 (0.9) 1 (0.9) 0 (0.0)
that could not be confirmed by IFA, deviant from prior experience with other lots. These findings indicated a poor specificity of lot SKA.CD. This was later confirmed by the manufacturer and was also noted in a study of postmortem samples of tissue donors.11 The problem only related to the specificity; the sensitivity of lot SKA.CD was at least as good as that from other lots. To prevent inclusion of false-positive data, all samples with a borderline or positive ELISA result in lot SKA.CD were retested using a newly produced kit lot. Retesting confirmed the reduced specificity of the first lot used; initial borderline and positive results were excluded from final analysis. Archived samples that tested confirmed positive for IgG Phase II antibodies to C. burnetii in 2010 were retested using the new lot, confirming that the sensitivity of the new lot was consistent with the lots used in the previous study.
RESULTS To monitor chronic Coxiella infection in blood donors after large outbreaks of Q fever, donor samples were collected in 2012 at two blood collection sites in the center of the area with high Q fever incidence between 2007 and 2010 (see Table 1 and Figure 1). Screening by ELISA for IgG Phase II antibodies to C. burnetii showed that 110 of 2490 donors were reactive (4.4%; 95% confidence interval [CI], 3.7%-5.3%). For all 110 donors Phase II antibody reactivity was confirmed by IFA, in line with previous findings showing that the IFA is the more sensitive test.12 In 79 of the 110 donors (71.8%; 95% CI, 62.3%-79.8%) Phase I IgG antibodies were detected by IFA (see Table 2). Four donors showed a Phase I titer of more than 32; no sample exceeded a Phase I titer of 256. A total of 397 donors of this study live in the area where we conducted a study on acute Q fever in 2009 to
2009
2010
2011
2012
2013
Fig. 2. Long-term course of Phase II IgG Coxiella antibodies as detected by ELISA (Serion) in blood donors who were confirmed positive in 2009 to 2010 (n = 24). The color of the symbols indicates the qualitative results (black = positive; gray = borderline; white = negative). Nota bene: the borderline cutoff value is kit lot dependent. OD/CO = sample-to-cutoff ratio.
2010.10 Twenty-two of these 397 donors (5.5%; 95% CI, 3.6%-8.4%) tested positive for Phase II IgG antibodies, while in the previous study we observed a significantly higher seroprevalence of 12.2% (95% CI, 9.6%-15.3%). A total of 179 donors were included in both studies. Twentyfour of these 179 donors (13.4%; 95% CI, 8.9%-19.5%) tested confirmed positive in 2009 to 2010. Only nine of these 24 donors were still seroreactive by ELISA for Phase II IgG antibodies in 2012 to 2013; in 15 of 24 donors (62.5%; 95% CI, 40.8%-80.5%) Phase II IgG antibodies had declined to negativity (see Fig. 2). To confirm this waning of antibodies and to exclude artifacts caused by varying sensitivity of antibody detection, the 24 positive samples from 2009 to 2010 were retested in 2012. Negligible differences in sample-to-cutoff ratios were observed, showing that the sensitivity of antibody detection in 2010 and in the current study is comparable. The Phase II IgG antibody titers also declined strongly when determined by IFA. However, only three of the 23 serial donor samples that could be tested (of 24 donors) turned negative in the follow-up sample, demonstrating the superior sensitivity of IFA over the ELISA for Q fever antibody detection.
DISCUSSION From 2007 through 2009 the Netherlands experienced seasonal Q fever outbreaks of an unprecedented scale. Primary infection with C. burnetii, both symptomatic and asymptomatic, may result in chronic Coxiella infection, which may become manifest years after primary infection. Hence in a previously affected area the safety of blood may be compromised by donors unknowingly incubating for chronic Q fever. This study is the first to examine asymptomatic chronic C. burnetii infection in blood Volume 54, November 2014 TRANSFUSION
2869
SLOT ET AL.
donors in a former outbreak area. Among 2490 at-risk donors tested no serologic evidence was found for chronic Coxiella infection (0%; 95% CI, 0%-0.2%), confirming that chronic C. burnetii infection is not a common condition in exposed, healthy adults. Of course the study results cannot be extrapolated to children, persons with medical conditions, and elderly persons. A striking decline of ELISA Phase II IgG antibody reactivity was found soon after initial infection. In a previous study we observed a seroprevalence of 12.2% among blood donors during the 2009 Q fever outbreak,10 while 3 years later only 4.4% seroreactivity remained. Longitudinal comparison of Phase II IgG ELISA reactivity in 24 donors who participated in both studies confirmed this finding. Consequently Phase II Coxiella IgG antibody detection by ELISA rapidly becomes unsuitable for conducting seroprevalence studies after Coxiella exposure; testing by immunofluorescence seems more appropriate. Blood donors who are unaware of infection may transmit C. burnetii to recipients, but transmission by blood products has not been demonstrated indisputably. Only one probable case of C. burnetii transmission via transfusion of blood products has been reported; this case was related to a donor with acute infection.13 Transmission of C. burnetii via transfusion of blood products could not be demonstrated in a lookback exercise, concerning transfused blood products from 13 Dutch donors with confirmed acute C. burnetii infection around the time of donation.14 In another Dutch study a possible case of transmission could neither be confirmed nor ruled out.10 There are no reports of donors with chronic C. burnetii infection who infected their recipients. It seems likely that blood donors who are asymptomatically infected with C. burnetii, both in the acute and in the chronic state, only pose a limited threat to the safety of blood. In conclusion, Phase II IgG Coxiella antibodies, as detected by an ELISA, waned rapidly in the years after acute infection. Serologic signs of chronic C. burnetii infection were absent in blood donors living in an area with recent severe outbreaks of Q fever, which supports the EU directive allowing the reentry of donors 2 years after resolved infection.15 As an additional safety measure such reentry should only be implemented for donors testing negative for Phase I IgG Coxiella antibodies.
REFERENCES 1. van der Hoek W, Dijkstra F, Schimmer B, et al. Q fever in the Netherlands: an update on the epidemiology and control measures. Euro Surveill 2010;15:19520. 2. Roest HI, Tilburg JJ, van der Hoek W, et al. The Q fever epidemic in The Netherlands: history, onset, response and reflection. Epidemiol Infect 2011;139:1-12. 3. Dijkstra F, van der Hoek W, Wijers N, et al. The 2007-2010 Q fever epidemic in The Netherlands: characteristics of notified acute Q fever patients and the association with dairy goat farming. FEMS Immunol Med Microbiol 2012; 64:3-12. 4. Wegdam-Blans MC, Stokmans RA, Tjhie JH, et al. Targeted screening as a tool for the early detection of chronic Q fever patients after a large outbreak. Eur J Clin Microbiol Infect Dis 2013;32:353-9. 5. Wegdam-Blans MC, Kampschreur LM, Delsing CE, et al. Chronic Q fever: review of the literature and a proposal of new diagnostic criteria. J Infect 2012;64:247-59. 6. Kampschreur LM, Oosterheert JJ, Koop AM, et al. Microbiological challenges in the diagnosis of chronic Q fever. Clin Vaccine Immunol 2012;19:787-90. 7. Kampschreur LM, Wever PC, Wegdam-Blans MC, et al. Defining chronic Q fever: a matter of debate. J Infect 2012; 65:362-3. 8. Kampschreur LM, Dekker S, Hagenaars JC, et al. Identification of risk factors for chronic Q fever, the Netherlands. Emerg Infect Dis 2012;18:563-70. 9. van der Hoek W, Schneeberger PM, Oomen T, et al. Shifting priorities in the aftermath of a Q fever epidemic in 2007 to 2009 in The Netherlands: from acute to chronic infection. Euro Surveill 2012;17:20059. 10. Hogema BM, Slot E, Molier M, et al. Coxiella burnetii infec-
11.
12.
13.
ACKNOWLEDGMENTS
14.
We thank Annemieke de Fijter and the staff of Sanquin Blood Collection Centers for collecting the blood samples. We thank Ed Bakker for performing the serologic screening and Bert Mesman for data extraction from the blood bank information system.
15.
CONFLICT OF INTEREST The authors have disclosed no conflicts of interest. 2870
TRANSFUSION Volume 54, November 2014
tion among blood donors during the 2009 Q-fever outbreak in the Netherlands. Transfusion 2012;52:144-50. Van Wijk MJ, Maas DW, Renders NH, et al. Screening of post-mortem tissue donors for Coxiella burnetii infection after large outbreaks of Q fever in The Netherlands. BMC Infect Dis 2014;14:6. Herremans T, Hogema BM, Nabuurs M, et al. Comparison of the performance of IFA, CFA, and ELISA assays for the serodiagnosis of acute Q fever by quality assessment. Diagn Microbiol Infect Dis 2013;75:16-21. Anonymous. Q-fever transmitted by blood transfusion— United States. CDC public health notes 1977. van Kraaij MG, Slot E, Hogema BM, et al. Lookback procedures after postdonation notifications during a Q fever outbreak in the Netherlands. Transfusion 2013;53:716-21. Commission of the European Communities. Commission Directive 2004/33/EC of 22 March 2004 implementing Directive 2002/98/EC of the European Parliament and of the Council as regards certain technical requirements for blood and blood components. OJ 2004;L91:25-39.
