LETTERS TO THE EDITOR
CONFLICT OF INTEREST
The Sew FUT2 mutation A385T does not result in a nonsecretor allele
The authors have disclosed no conflicts of interest.
Dennis Goldfinger, MD e-mail: [email protected] Joshua Sifuentes Alyssa Ziman, MD Division of Transfusion Medicine Department of Pathology and Laboratory Medicine David Geffen School of Medicine at UCLA Los Angeles, CA
REFERENCES 1. Pool JG, Gershgold EJ, Pappenhagen AR. High-potency antihaemophilic factor concentrate prepared from cryoglobulin precipitate. Nature 1964;203:312. 2. Ratnoff OD. Some complications of the therapy of classic hemophilia. J Lab Clin Med 1984;103:653-9. 3. Pierce GF, Lusher JM, Brownstein AP, et al. The use of purified clotting factor concentrates in hemophilia: influence of viral safety, cost, and supply on therapy. JAMA 1989;261:3434-8. 4. US Food and Drug Administration. User fee billable biologic products and potencies approved under Section 351 of the PHS Act. 2014 [cited 2014 Oct 30]. Available from: http://www.fda.gov/AboutFDA/CentersOffices/ OfficeofMedicalProductsandTobacco/CBER/ ucm122936.htm 5. Mainwarning RL, Brueckner GG. Fibrinogen transmitted hepatitis: a controlled study. JAMA 1966;195:437-41. 6. Cronberg S, Belfrage S, Nilsson IM. Fibrinogen-transmitted 7.
8. 9. 10. 11.
12. 13.
14.
hepatitis. Lancet 1963;I:967-9. Soloway HB, Bereznak CE. Plasma fibrinogen levels following cryoprecipitate infusion. Transfusion 1970;10: 326-8. Ness PM, Perkins HA. Cryoprecipitate as a reliable source of fibrinogen replacement. JAMA 1979;241:1690-1. Bove JR. Fibrinogen—is the benefit worth the risk? Transfusion 1978;18:129-36. Hile JP. Fibrinogen (human). Revocation of licenses. Fed Regist 1978;43:1131-2. Yang L, Stanworth S, Baglin T. Cryoprecipitate: an outmoded treatment? Transfus Med 2012;22: 315-20. Food and Drugs: Biologics. Code of Federal Regulations, title 21, parts 640.50-640.56, 1977. Roseff S, editor. Standards for blood banks and transfusion services. 29th ed. Bethesda (MD): American Association of Blood Banks; 2014. Ness PM, Perkins HA. Fibrinogen in cryoprecipitate and its relationship to factor VIII (AHF) levels. Transfusion 1980; 20:93-6.
In the paper by Tian and colleagues1 (and also in some of the references cited), the authors mistakenly define the FUT2 mutation A385T as a nonsecretor (se) mutation. The A385T mutation has been expressed and conclusively found to result in a functional (Se) FUT2 fucosyltransferase.2 This mutation results in an unstable FUT2 enzyme (Sew) and as a consequence is responsible for the salivary ABH partial-secretor and Le(a+b+) RBC phenotypes. Accordingly inferences in this article for the incidence of the nonsecretor phenotype will be incorrect, and instead most of these individuals will be predominantly salivary ABH partial-secretors and when Lewis positive of the Le(a+b+) phenotype. CONFLICT OF INTEREST The author has disclosed no conflicts of interest.
Stephen Henry, PhD e-mail: [email protected] School of Engineering Auckland University of Technology Auckland, New Zealand
REFERENCES 1. Tian L, Song N, Yao ZQ, et al. Sequence analysis of the human fucosyltransferase 1 and 2 genes in Tibetan blood donors: identification of three novel alleles. Transfusion 2014;54:1847-50. 2. Henry S, Mollicone R, Fernandez P, et al. Molecular basis for erythrocyte Le(a+b+) and the salivary ABH partial-secretor phenotypes. Expression of a FUT2 secretor allele with an A→T mutation at nucleotide 385 correlates with reduced α(1,2)fucosyltransferase activity. Glycoconj J 1996;13:985-93.
Thromboembolic events associated with immune globulin preparations: spontaneous pharmacovigilance data versus claims-based data In May 2014, Sridhar and colleagues1 published the results of a claims-based study that aimed at identifying the rate of same-day thrombotic and embolic events (TEEs)
*Dr. Schiff passed away unexpectedly since the original submission of this letter. We are saddened by the loss of a talented and accomplished colleague and acknowledge his tremendous work and contributions in developing more effective treatment options for patients with immunodeficiencies. Volume 54, December 2014 TRANSFUSION
3255
LETTERS TO THE EDITOR
TABLE 1. US distribution data for GGL and GGSD, spontaneous TEEs and Ph/Th, and reporting rates US Reporting rate (per 106 g) IG GGL* GGSD†
Units distributed (g)
Number of TEEs
Number of Ph/Th events
TEEs only
TEEs + Ph/Th
95,801,324 15,321,103
42 5
4 1
0.44 0.33
0.48 0.39
* For GGL, seven of 42 TEEs and none of four Ph/Th were same-day events (of note, 18/42 TEEs and 2/4 Ph/Th events were reported without latency information). † For GGSD, none of five TEEs and none of one Ph/Th events were same-day events (of note, 3/5 TEEs and 1/1 Ph/Th were reported without latency information).
associated with immune globulins (IGs). One of the study findings was an elevated but nonsignificant increase of the TEE rate for lyophilized IGs compared to the reference product GAMMAGARD LIQUID (GGL). Baxter’s GAMMAGARD S/D (GGSD) is one of the lyophilized IG products available in the United States and thus this raised the issue of whether the incidence of TEEs is higher in patients receiving GAMMAGARD S/D relative to GAMMAGARD LIQUID. To compare these findings with data obtained from spontaneous reports volunteered to Baxter Healthcare Corp., the marketing authorization holder of GGL and GGSD, Baxter’s pharmacovigilance database was searched for all spontaneous adverse event (AE) reports from the United States for GGL and GGSD received from January 1, 2006, through December 31, 2013. AE reports from the literature and solicited reports (e.g., from market research or patient support programs) were excluded to avoid confounding. TEEs associated with GGL and GGSD (irrespective of their causality or latency to event onset) were identified with the Standardized MedDRA Query (SMQ) “Embolic and thrombotic events” (e.g., myocardial infarctions, strokes, deep vein thrombosis, or pulmonary embolism) and events of phlebitis or thrombophlebitis (Ph/Th) with MedDRA SMQ “Thrombophlebitis.” US distribution data for GGL and GGSD were obtained from Baxter’s financial databases. The proportional reporting ratio (PRR) for GGSD versus GGL was calculated to test for disproportionate reporting and the statistical association using a chi-squared test on one degree of freedom with Yates’s correction.2 These methods are applied to test for disproportionalities in large pharmacovigilance databases, but also when performing head-to-head comparisons of medicinal products of the same class as in our analysis.3 US distribution data, the number of spontaneous US reports of TEEs and Ph/Th for GGL and GGSD, and their corresponding reporting rates for the period from January 1, 2006, through December 31, 2013, are displayed in Table 1. In contrast to the data published by Sridhar and colleagues, available pharmacovigilance data at Baxter do not show an elevated risk for GGSD-associated TEEs compared to GGL. Given the low absolute number of events reported to Baxter spontaneously (especially for GGSD, for 3256
TRANSFUSION Volume 54, December 2014
which one additional event would offset the observed differences in reporting rates), the presented reporting rates can be considered comparable. During the period analyzed, in addition to the number of TEEs and Ph/Th presented in Table 1, from spontaneous US reporting sources, a total of 1506 and 303 additional AE reports had been received for GGL and GGSD, respectively. A signal of disproportionate reporting can be assumed when the PRR is at least 2 and chi-square is at least 4 (and three or more cases of interest have been reported).3 The PRR for the analyzed spontaneous Baxter data for GGSD versus GGL is 0.7 (95% CI, 0.3-1.5), with chi-square (Yates corrected) calculated at 0.65 (p = 0.42). These results further confirm the notion that spontaneous US pharmacovigilance data available at Baxter do not support a higher risk of TEEs associated with GGSD compared to GGL. Various reasons may contribute to the different results of this analysis compared to the study published by Sridhar and colleagues, and some of these highlight in our opinion the current limitations of claims-based studies especially in rare disorders. 1.
2.
3.
The claims-based data analysis did not identify individual brands of lyophilized IGs. Given the distribution data for GGSD and GGL from January 1, 2008, through June 30, 2012 (approx. 5 tons GGSD and approx. 60 tons GGL), only approximately 300 persons studied by Sridhar and colleagues could have been exposed to GGSD (assuming an equal consumption of GGL and GGSD per person exposed). Therefore, the presented results of a potentially elevated but nonsignificant increase of the TEE rate for lyophilized IGs may represent the result of distinctly different rates for various lyophilized IGs. While it may be assumed that IGs are used equally across indications, IgA-depleted IGs such as GGSD may have a distinctly different usage pattern. IgAdepleted IGs are typically prescribed for patients with a history of adverse reactions to IG infusions that are thought to be associated with IgA but not at all confined to that one etiology. Sridhar and colleagues identified 10.5 venous sameday TEEs compared to 4.8 arterial same-day events
LETTERS TO THE EDITOR
per 1000 persons exposed to IG. This finding is in contrast to previous publications, which identified arterial events with higher rates than venous events.4 This discrepancy may indicate that the high number of venous TEEs observed by Sridhar and colleagues may have been largely driven by cases of phlebitis or thrombophlebitis. Unfortunately, separate numbers for these types of events have not been provided by the authors. Pooling events such as myocardial infarction or deep vein thrombosis with events such as phlebitis and thrombophlebitis, which may be attributable to potentially different risk factors, does not allow for making a clear distinction between events associated with IGs themselves versus events that may be associated with the devices used for administering IGs. Estimates for peripheral venous catheter– associated thrombophlebitis rates range from 2% to 80%, and the risk of these events is correlated with the site and size of the catheter.5 It is important to recognize that claims-based data and data obtained through spontaneous reporting schemes (SRSs) cannot be directly compared. While claims-based data describe a finite data set, spontaneous data are subject to several limitations, with underreporting of an unknown extent and incomplete reporting representing two major limitations of SRSs. While the study of claims-based data has added an additional but important tool to evaluating product-related risks, it has obvious limitations such as lack of detail regarding individual events. Nonetheless, without research such as that published by Sridhar and colleagues the magnitude of the risk of IG-associated TEEs might still be underestimated, but discrepancies such as those we observed require substantial additional research to further understand the limitations of claims-based data to avoid premature assumptions, as well as premature regulatory decisions. Moreover, in our opinion the presented data for GGL and GGSD highlight the importance of SRSs and the necessity to carefully examine SRS data, clinical data, and data from new tools such as commercial databases and carefully weigh all available evidence, including the limitations of each of these data sources. CONFLICT OF INTEREST The authors report that they are or were employed by Baxter Healthcare and own stock in the company.
Roger Berg, MD1 e-mail: [email protected] Richard Schiff, MD, PhD2* 1 Global Pharmacovigilance Baxter Innovations GmbH Vienna, Austria
2
Clinical Affairs Baxter Healthcare Corp. Westlake Village, CA
REFERENCES 1. Sridhar G, Ekezue BF, Izurieta HS, et al. Immune globulins and same-day thrombotic events as recorded in a large health care database during 2008 to 2012. Transfusion 2014; 54:2553-65. 2. Evans SJ, Waller PC, Davis S. Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf 2001;10:483-6. 3. Seong JM, Choi NK, Lee J, et al. Comparison of the safety of seven iodinated contrast media. J Korean Med Sci 2013;28: 1703-10. 4. Funk MB, Gross N, Gross S, et al. Thromboembolic events associated with immunoglobulin treatment. Vox Sang 2013; 105:54-64. 5. Cicolini G, Bonghi AP, Di Labio L, et al. Position of peripheral venous cannulae and the incidence of thrombophlebitis: an observational study. J Adv Nurs 2009; 65:1268-73.
Plasma exchange complications in patients treated for thrombotic thrombocytopenia purpura–hemolytic uremic syndrome: 2011 to 2014 Plasma exchange (PEX) is the standard treatment for patients with clinically suspected thrombotic thrombocytopenic purpura (TTP). Since the diagnosis of TTP may initially be uncertain, the decision to begin PEX in a patient with suspected TTP may be a difficult balance between the risk of severe TTP complications if treatment is delayed and the risk of severe PEX complications if treatment is begun. To document the frequency of PEX complications, the Oklahoma TTP-HUS Registry began in 1996 to prospectively document the outcomes of all PEX procedures for all consecutive patients in whom PEX was begun for an initial diagnosis of TTP, hemolytic uremic syndrome (HUS), or thrombotic microangiopathy (TMA).1 We have subsequently reported each 3-year cohort of patients; our most recent report described data on all 302 consecutive patients, June 26, 1996, to June 25, 2011.2 This report describes our subsequent experience with all 40
This project was supported in part by the National Institute of General Medical Sciences of the National Institutes of Health under Award U54GM104938. Volume 54, December 2014 TRANSFUSION
3257
CONFLICT OF INTEREST
The Sew FUT2 mutation A385T does not result in a nonsecretor allele
The authors have disclosed no conflicts of interest.
Dennis Goldfinger, MD e-mail: [email protected] Joshua Sifuentes Alyssa Ziman, MD Division of Transfusion Medicine Department of Pathology and Laboratory Medicine David Geffen School of Medicine at UCLA Los Angeles, CA
REFERENCES 1. Pool JG, Gershgold EJ, Pappenhagen AR. High-potency antihaemophilic factor concentrate prepared from cryoglobulin precipitate. Nature 1964;203:312. 2. Ratnoff OD. Some complications of the therapy of classic hemophilia. J Lab Clin Med 1984;103:653-9. 3. Pierce GF, Lusher JM, Brownstein AP, et al. The use of purified clotting factor concentrates in hemophilia: influence of viral safety, cost, and supply on therapy. JAMA 1989;261:3434-8. 4. US Food and Drug Administration. User fee billable biologic products and potencies approved under Section 351 of the PHS Act. 2014 [cited 2014 Oct 30]. Available from: http://www.fda.gov/AboutFDA/CentersOffices/ OfficeofMedicalProductsandTobacco/CBER/ ucm122936.htm 5. Mainwarning RL, Brueckner GG. Fibrinogen transmitted hepatitis: a controlled study. JAMA 1966;195:437-41. 6. Cronberg S, Belfrage S, Nilsson IM. Fibrinogen-transmitted 7.
8. 9. 10. 11.
12. 13.
14.
hepatitis. Lancet 1963;I:967-9. Soloway HB, Bereznak CE. Plasma fibrinogen levels following cryoprecipitate infusion. Transfusion 1970;10: 326-8. Ness PM, Perkins HA. Cryoprecipitate as a reliable source of fibrinogen replacement. JAMA 1979;241:1690-1. Bove JR. Fibrinogen—is the benefit worth the risk? Transfusion 1978;18:129-36. Hile JP. Fibrinogen (human). Revocation of licenses. Fed Regist 1978;43:1131-2. Yang L, Stanworth S, Baglin T. Cryoprecipitate: an outmoded treatment? Transfus Med 2012;22: 315-20. Food and Drugs: Biologics. Code of Federal Regulations, title 21, parts 640.50-640.56, 1977. Roseff S, editor. Standards for blood banks and transfusion services. 29th ed. Bethesda (MD): American Association of Blood Banks; 2014. Ness PM, Perkins HA. Fibrinogen in cryoprecipitate and its relationship to factor VIII (AHF) levels. Transfusion 1980; 20:93-6.
In the paper by Tian and colleagues1 (and also in some of the references cited), the authors mistakenly define the FUT2 mutation A385T as a nonsecretor (se) mutation. The A385T mutation has been expressed and conclusively found to result in a functional (Se) FUT2 fucosyltransferase.2 This mutation results in an unstable FUT2 enzyme (Sew) and as a consequence is responsible for the salivary ABH partial-secretor and Le(a+b+) RBC phenotypes. Accordingly inferences in this article for the incidence of the nonsecretor phenotype will be incorrect, and instead most of these individuals will be predominantly salivary ABH partial-secretors and when Lewis positive of the Le(a+b+) phenotype. CONFLICT OF INTEREST The author has disclosed no conflicts of interest.
Stephen Henry, PhD e-mail: [email protected] School of Engineering Auckland University of Technology Auckland, New Zealand
REFERENCES 1. Tian L, Song N, Yao ZQ, et al. Sequence analysis of the human fucosyltransferase 1 and 2 genes in Tibetan blood donors: identification of three novel alleles. Transfusion 2014;54:1847-50. 2. Henry S, Mollicone R, Fernandez P, et al. Molecular basis for erythrocyte Le(a+b+) and the salivary ABH partial-secretor phenotypes. Expression of a FUT2 secretor allele with an A→T mutation at nucleotide 385 correlates with reduced α(1,2)fucosyltransferase activity. Glycoconj J 1996;13:985-93.
Thromboembolic events associated with immune globulin preparations: spontaneous pharmacovigilance data versus claims-based data In May 2014, Sridhar and colleagues1 published the results of a claims-based study that aimed at identifying the rate of same-day thrombotic and embolic events (TEEs)
*Dr. Schiff passed away unexpectedly since the original submission of this letter. We are saddened by the loss of a talented and accomplished colleague and acknowledge his tremendous work and contributions in developing more effective treatment options for patients with immunodeficiencies. Volume 54, December 2014 TRANSFUSION
3255
LETTERS TO THE EDITOR
TABLE 1. US distribution data for GGL and GGSD, spontaneous TEEs and Ph/Th, and reporting rates US Reporting rate (per 106 g) IG GGL* GGSD†
Units distributed (g)
Number of TEEs
Number of Ph/Th events
TEEs only
TEEs + Ph/Th
95,801,324 15,321,103
42 5
4 1
0.44 0.33
0.48 0.39
* For GGL, seven of 42 TEEs and none of four Ph/Th were same-day events (of note, 18/42 TEEs and 2/4 Ph/Th events were reported without latency information). † For GGSD, none of five TEEs and none of one Ph/Th events were same-day events (of note, 3/5 TEEs and 1/1 Ph/Th were reported without latency information).
associated with immune globulins (IGs). One of the study findings was an elevated but nonsignificant increase of the TEE rate for lyophilized IGs compared to the reference product GAMMAGARD LIQUID (GGL). Baxter’s GAMMAGARD S/D (GGSD) is one of the lyophilized IG products available in the United States and thus this raised the issue of whether the incidence of TEEs is higher in patients receiving GAMMAGARD S/D relative to GAMMAGARD LIQUID. To compare these findings with data obtained from spontaneous reports volunteered to Baxter Healthcare Corp., the marketing authorization holder of GGL and GGSD, Baxter’s pharmacovigilance database was searched for all spontaneous adverse event (AE) reports from the United States for GGL and GGSD received from January 1, 2006, through December 31, 2013. AE reports from the literature and solicited reports (e.g., from market research or patient support programs) were excluded to avoid confounding. TEEs associated with GGL and GGSD (irrespective of their causality or latency to event onset) were identified with the Standardized MedDRA Query (SMQ) “Embolic and thrombotic events” (e.g., myocardial infarctions, strokes, deep vein thrombosis, or pulmonary embolism) and events of phlebitis or thrombophlebitis (Ph/Th) with MedDRA SMQ “Thrombophlebitis.” US distribution data for GGL and GGSD were obtained from Baxter’s financial databases. The proportional reporting ratio (PRR) for GGSD versus GGL was calculated to test for disproportionate reporting and the statistical association using a chi-squared test on one degree of freedom with Yates’s correction.2 These methods are applied to test for disproportionalities in large pharmacovigilance databases, but also when performing head-to-head comparisons of medicinal products of the same class as in our analysis.3 US distribution data, the number of spontaneous US reports of TEEs and Ph/Th for GGL and GGSD, and their corresponding reporting rates for the period from January 1, 2006, through December 31, 2013, are displayed in Table 1. In contrast to the data published by Sridhar and colleagues, available pharmacovigilance data at Baxter do not show an elevated risk for GGSD-associated TEEs compared to GGL. Given the low absolute number of events reported to Baxter spontaneously (especially for GGSD, for 3256
TRANSFUSION Volume 54, December 2014
which one additional event would offset the observed differences in reporting rates), the presented reporting rates can be considered comparable. During the period analyzed, in addition to the number of TEEs and Ph/Th presented in Table 1, from spontaneous US reporting sources, a total of 1506 and 303 additional AE reports had been received for GGL and GGSD, respectively. A signal of disproportionate reporting can be assumed when the PRR is at least 2 and chi-square is at least 4 (and three or more cases of interest have been reported).3 The PRR for the analyzed spontaneous Baxter data for GGSD versus GGL is 0.7 (95% CI, 0.3-1.5), with chi-square (Yates corrected) calculated at 0.65 (p = 0.42). These results further confirm the notion that spontaneous US pharmacovigilance data available at Baxter do not support a higher risk of TEEs associated with GGSD compared to GGL. Various reasons may contribute to the different results of this analysis compared to the study published by Sridhar and colleagues, and some of these highlight in our opinion the current limitations of claims-based studies especially in rare disorders. 1.
2.
3.
The claims-based data analysis did not identify individual brands of lyophilized IGs. Given the distribution data for GGSD and GGL from January 1, 2008, through June 30, 2012 (approx. 5 tons GGSD and approx. 60 tons GGL), only approximately 300 persons studied by Sridhar and colleagues could have been exposed to GGSD (assuming an equal consumption of GGL and GGSD per person exposed). Therefore, the presented results of a potentially elevated but nonsignificant increase of the TEE rate for lyophilized IGs may represent the result of distinctly different rates for various lyophilized IGs. While it may be assumed that IGs are used equally across indications, IgA-depleted IGs such as GGSD may have a distinctly different usage pattern. IgAdepleted IGs are typically prescribed for patients with a history of adverse reactions to IG infusions that are thought to be associated with IgA but not at all confined to that one etiology. Sridhar and colleagues identified 10.5 venous sameday TEEs compared to 4.8 arterial same-day events
LETTERS TO THE EDITOR
per 1000 persons exposed to IG. This finding is in contrast to previous publications, which identified arterial events with higher rates than venous events.4 This discrepancy may indicate that the high number of venous TEEs observed by Sridhar and colleagues may have been largely driven by cases of phlebitis or thrombophlebitis. Unfortunately, separate numbers for these types of events have not been provided by the authors. Pooling events such as myocardial infarction or deep vein thrombosis with events such as phlebitis and thrombophlebitis, which may be attributable to potentially different risk factors, does not allow for making a clear distinction between events associated with IGs themselves versus events that may be associated with the devices used for administering IGs. Estimates for peripheral venous catheter– associated thrombophlebitis rates range from 2% to 80%, and the risk of these events is correlated with the site and size of the catheter.5 It is important to recognize that claims-based data and data obtained through spontaneous reporting schemes (SRSs) cannot be directly compared. While claims-based data describe a finite data set, spontaneous data are subject to several limitations, with underreporting of an unknown extent and incomplete reporting representing two major limitations of SRSs. While the study of claims-based data has added an additional but important tool to evaluating product-related risks, it has obvious limitations such as lack of detail regarding individual events. Nonetheless, without research such as that published by Sridhar and colleagues the magnitude of the risk of IG-associated TEEs might still be underestimated, but discrepancies such as those we observed require substantial additional research to further understand the limitations of claims-based data to avoid premature assumptions, as well as premature regulatory decisions. Moreover, in our opinion the presented data for GGL and GGSD highlight the importance of SRSs and the necessity to carefully examine SRS data, clinical data, and data from new tools such as commercial databases and carefully weigh all available evidence, including the limitations of each of these data sources. CONFLICT OF INTEREST The authors report that they are or were employed by Baxter Healthcare and own stock in the company.
Roger Berg, MD1 e-mail: [email protected] Richard Schiff, MD, PhD2* 1 Global Pharmacovigilance Baxter Innovations GmbH Vienna, Austria
2
Clinical Affairs Baxter Healthcare Corp. Westlake Village, CA
REFERENCES 1. Sridhar G, Ekezue BF, Izurieta HS, et al. Immune globulins and same-day thrombotic events as recorded in a large health care database during 2008 to 2012. Transfusion 2014; 54:2553-65. 2. Evans SJ, Waller PC, Davis S. Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf 2001;10:483-6. 3. Seong JM, Choi NK, Lee J, et al. Comparison of the safety of seven iodinated contrast media. J Korean Med Sci 2013;28: 1703-10. 4. Funk MB, Gross N, Gross S, et al. Thromboembolic events associated with immunoglobulin treatment. Vox Sang 2013; 105:54-64. 5. Cicolini G, Bonghi AP, Di Labio L, et al. Position of peripheral venous cannulae and the incidence of thrombophlebitis: an observational study. J Adv Nurs 2009; 65:1268-73.
Plasma exchange complications in patients treated for thrombotic thrombocytopenia purpura–hemolytic uremic syndrome: 2011 to 2014 Plasma exchange (PEX) is the standard treatment for patients with clinically suspected thrombotic thrombocytopenic purpura (TTP). Since the diagnosis of TTP may initially be uncertain, the decision to begin PEX in a patient with suspected TTP may be a difficult balance between the risk of severe TTP complications if treatment is delayed and the risk of severe PEX complications if treatment is begun. To document the frequency of PEX complications, the Oklahoma TTP-HUS Registry began in 1996 to prospectively document the outcomes of all PEX procedures for all consecutive patients in whom PEX was begun for an initial diagnosis of TTP, hemolytic uremic syndrome (HUS), or thrombotic microangiopathy (TMA).1 We have subsequently reported each 3-year cohort of patients; our most recent report described data on all 302 consecutive patients, June 26, 1996, to June 25, 2011.2 This report describes our subsequent experience with all 40
This project was supported in part by the National Institute of General Medical Sciences of the National Institutes of Health under Award U54GM104938. Volume 54, December 2014 TRANSFUSION
3257
