Märit Halmin

Social differences in breast cancer survival in relation to patient management within a National Health Care System (Sweden)

Epidemiologic studies have shown that cancer survival is poorer in low compared with high socioeconomic groups. We investigated whether these differences were associated with disparities in tumour characteristics and management. This cohort study was based on 9,908 women aged 20–79 years at diagnosis with primary breast cancer identified in a Swedish population‐based clinical register. Information on socioeconomic standing was obtained from a social database. The 5‐year cause‐specific survival (CSS) and mortality hazard ratios (HR) were estimated by Cox proportional hazard models to assess differences in survival between socioeconomic groups while adjusting for diagnostic intensity, tumour characteristics and treatment. Following adjustment for age, year and stage at diagnosis, the risk of dying of breast cancer was 35% lower among women with high education compared with that of low education (HR = 0.65, 95% CI 0.53–0.80). When compared with women with high education, a lower percentage of women with low education had been investigated for proliferation (84 vs. 76%) or hormone receptor status (89 vs. 81%), had tumours ≤20 mm (68 vs. 64%), were treated at a main hospital (75 vs. 68%) and had received radiation treatment (80 vs. 67%) or chemotherapy (31 vs. 18%). However, these proportional differences could not explain the observed social gradient in survival. To minimize social differences in breast cancer survival, further research should address not only factors leading to inequities in management but also focus on patient factors such as health awareness, comorbidity burden and compliance to adjuvant treatment.

Long-term inequalities in breast cancer survival a ten year follow-up study of patients managed within a National Health Care System (Sweden)

Introduction. Converging epidemiological evidence based on studies of different designs in a variety of populations and settings show that cancer survival tends to be poorer in low compared to high socioeconomic groups. In an extension of an earlier register-based study, we examined the influence of socioeconomic factors on long-term survival in women with a first diagnosis of invasive breast cancer in 1993 in Sweden, a country with a policy of providing equal access to health care to all at nominal cost within a National Health Care System. Material and methods. The study was based on data set generated by record linkages between the Swedish Cancer Register, Census databases and the Cause of Death Register. Four different categorical variables were used as indicators of socioeconomic standing. Cox proportional hazard regression models were used to estimate the effects of socioeconomic status on risk of death. Results. Of 4 645 eligible women with breast cancer, 1 016 had died from breast cancer at the end of follow-up on December 31, 2003. After adjustment for tumour size and age at diagnosis, risk of death was 19% lower among women belonging to a household of high compared to low socioeconomic status (HR high versus low 0.81; 95% CI: 0.67–0.97). Discussion. These findings indicate that social inequalities in breast cancer survival persist at least up to ten years after an initial diagnosis. While social gradients detected shortly after diagnosis may mainly reflect an influence of socioeconomic differences in overall health status and frailty, differentials persisting beyond five years rather point to a long-term influence of disparities in management of both primary tumours and recurrences. Further studies are needed to explore whether the present findings reflect amendable inequalities in access to state-of- the-art treatment. For all calendar periods, observed survival in the most privileged groups sets the goal for what is achievable for all breast cancer patients.

Effect of Plasma-to-rbc Ratios in Trauma Patients: A Cohort Study With Time-dependent Data*

Background:A widespread approach today is to transfuse bleeding trauma patients with RBC concentrates and plasma at a 1:1 ratio. This regime is supported by a range of observational studies showing lower mortality in bleeding patients receiving equal volumes of plasma and RBCs. The rationale for this practice is still unclear with several studies failing to show any survival benefits of increased plasma use, perhaps due to a failure to account for the timing of transfused units. Objective:To study the association between plasma-to-RBC ratios and risk of death in trauma patients, using appropriate methods. Design, Settings, and Participants:In a retrospective cohort study, we assembled data on 741 transfused trauma patients at a large trauma center. Measures of transfusion therapy were assessed entirely time dependently, and relative risk of death was compared between patients receiving low to high plasma-to-RBC ratio (< 0.85 vs > 0.85). Measurements and Results:In the time-dependent analyses, we saw no significant association between a low plasma ratio and the risk of death. However, age more than 75 years, injury severity score greater than 33, Glasgow Coma Scale less than 8, and systolic blood pressure lower than 90 mm Hg were all significantly associated with increased risk of death. Conversely, when the analyses were conducted with conventional methods, a strong protective effect of high plasma ratios was seen. Conclusions:The key finding in our study is the strikingly different results produced by time-dependent analyses and the conventional analyses when studying survival and plasma-to-RBC ratio, supporting recent claims that prior studies showing benefit of high plasma ratios might have suffered from survival bias. There is a great need for further studies on the subject to enable improvements in treatment of massively bleeding trauma patients.

Epidemiology of Massive Transfusion: A Binational Study From Sweden and Denmark*

Objective:There is an increasing focus on massive transfusion, but there is a paucity of comprehensive descriptions of the massively transfused patients and their outcomes. The objective of this study is to describe the incidence rate of massive transfusion, patient characteristics, and the mortality of massively transfused patients. Design:Descriptive cohort study. Setting:Nationwide study with data from Sweden and Denmark. Patients:The study was based on the Scandinavian Donations and Transfusions database, including all patients receiving 10 or more red cell concentrate transfusions in Sweden from 1987 and in Denmark from 1996. A total of 92,057 patients were included. Patients were followed until the end of 2012. Measurements and Main Results:Descriptive statistics were used to characterize the patients and indications. Post transfusion mortality was expressed as crude 30-day mortality and as long-term mortality using the Kaplan-Meier method and using standardized mortality ratios. The incidence of massive transfusion was higher in Denmark (4.5 per 10,000) than in Sweden (2.5 per 10,000). The most common indication for massive transfusion was major surgery (61.2%) followed by trauma (15.4%). Massive transfusion due to obstetrical bleeding constituted only 1.8%. The overall 5-year mortality was very high (54.6%), however with large differences between indication groups, ranging from 91.1% among those transfused for a malignant disease without surgery to 1.7% among patients transfused for obstetrical bleeding. The early standardized mortality ratios were high and decreased thereafter, but remained elevated throughout the time period. Conclusions:This large-scale study based on nationwide data from Sweden and Denmark describes the complete range of massive transfusion. We report a nonnegligible incidence and both a high absolute mortality and high standardized mortality ratio. The general pattern was similar for Sweden and Denmark, and we believe that similar patterns may be found in other high-resource countries. The study provides a relevant background for clinicians and researchers for designing future studies in this field.

Lack of association between blood donor age and survival of transfused patients

To the editor: The possible rejuvenating effects of transfusions from young donors to older patients have generated considerable interest recently, following publication of a study showing improvements in muscle regeneration when older mice were transfused with blood from younger mice.[1][1]

Length of Storage of Red Blood Cells and Patient Survival After Blood Transfusion: A Binational Cohort Study

Red blood cells (RBCs) undergo various physiologic changes during storage (14), but whether the changes observed in vitro are clinically significant or are associated with mortality in the recipient is under debate (5). Many observational studies investigated the association between the duration of RBC storage and adverse outcomes, including mortality (631). Results are conflicting, with some studies finding prolonged storage time to be an independent risk factor for adverse outcomes (30), others showing no such association (1419), and still others revealing detrimental effects from fresh blood (31). These study results may have been confounded by differences among patients receiving blood of different storage lengths (32, 33). Four recently completed randomized trials observed no association between storage time and risk for adverse outcomes (3437). However, these studies were not sufficiently powered to exclude small yet clinically relevant effects, did not include adequate numbers of units near the end of expiration, and included only specific patient groups, possibly limiting their generalizability (3840). Also, summarizing data in meta-analyses has been difficult because of methodological heterogeneity (39, 41). Consequently, we performed a large, updated binational cohort study with more recent data to investigate the association between RBC storage and risk for death. Methods Setting Sweden and Denmark are northern European countries with 9.9 million and 5.7 million inhabitants, respectively. Both countries have public health care systems with universal coverage, and life expectancy there is high. Transfusion services in Sweden and Denmark are part of the public health care system. The national guidelines for component preparation and transfusion practice are similar, except the maximum storage time is 42 days in Sweden and 35 days in Denmark (25). Whole blood is collected in commercially available blood bags containing citratephosphatedextrose. The RBCs are suspended in salineadenineglucosemannitol solution. Although the practice of leukocyte reduction has varied among hospitals and over time, in both countries all blood components currently are leukocyte reduced before storage. Data Sources Analyses were based on the Scandinavian Donations and Transfusions (SCANDAT2) database, which was described in greater detail previously (42). In brief, electronic registration of data on blood donors, blood donations, blood components, and transfusion recipients was initiated in 1968 in Sweden and 1983 in Denmark. Although initially only a few blood centers had electronic registration the fraction of centers recording data electronically increased gradually over time and has been nearly nationwide since 1996 in Sweden and 1998 in Denmark. All persons in the SCANDAT2 database are identified by unique national registration numbers, enabling linkage to nationwide health outcomes registers, including each countrys nationwide patient registry that provides detailed data on hospital care. Linkage to population registers provided dates of birth and, if applicable, death and emigration. The analyses were performed as a retrospective cohort study with transfusion data obtained from the SCANDAT2 database as well as covariate and outcomes data from each countrys nationwide patient and death registers. We included all patients between the ages of 15 and 90 years who received at least 1 RBC unit between 2003 and 2012. Patients younger than 15 years were excluded because of specific transfusion practices in pediatric care, and those older than 90 years were excluded because of their intrinsically poor prognosis. All recipients of autologous transfusions or transfusions of RBCs with unknown storage time were excluded. A priori, we assessed mortality over 30 days and 1 year of follow-up. Discrete Exposure Group Approach We first conducted a set of analyses that followed an approach similar to that of 2 previous studies based on the SCANDAT database (14, 25). All transfusions were grouped into episodes, with a 7-day exposure periodduring which we ascertained data on all transfusions. Follow-up started on the day of the last transfusion of the exposure ascertainment period. All patients then were assigned to discrete exposure groups of those who had received units stored for 0 to 9, 10 to 19, 20 to 29, or 30 to 42 days, as well as a mixed group of patients who had received RBC units from more than 1 storage time category. Although grouping patients into discrete RBC storage time categories ensured that no overlap occurred among patients with regard to the main exposure of interest, this approach required the allocation of a substantial number of patients into a mixed category, which diluted power. Also, because the probability that a patient received units from only 1 storage time category decreased with each additional unit, many of the most severely ill patientsthat is, those who needed the most transfusionslikely ended up in the mixed category. Therefore, because most transfused RBC units were stored for only a moderate duration (see the histograms of RBC storage of transfused units in the Appendix Figure), the probability of receiving blood from a different exposure category was greater for patients who received blood units nearer the end of storage than for those who received fresh blood. For example, a patient who on day 1 received 2 RBC units stored for 30 to 42 days and on day 2 needed additional units was more likely to receive fresher blood and therefore transfer to the mixed category than a patient who on day 1 received 2 units stored for 10 to 19 days. In effect, this means the sickest patients in each category gradually shifted to the mixed category, as they required additional transfusions, thus potentially decreasing the mortality rate in the 4 main exposure groups and increasing it in the mixed category. As such, we hypothesized that the rate of patients entering the mixed category would differ among the exposure categories, which might adversely affect the validity of the analyses. To verify this phenomenon, we estimated the probability of receiving transfusions from a different storage time category on day 2 relative to the storage time of all units transfused on day 1. For details, see the Appendix. Appendix Figure. Distribution and mean storage time of RBCs, by blood group. RBC = red blood cell. Time-Dependent Approach Recognizing the aforementioned phenomenon, we devised a second analysis. We defined exposure as the number of transfused old units (stored for 30 days) or very old units (stored for 35 days), and these variables were treated time dependently so that patients could change exposure throughout follow-up. For example, in the main analysis, a patient receiving a total of 8 RBC units4 stored for 20 days on day 1 and 4 stored for 33 days on day 2would be placed in the mixed group. However, in the time-dependent analysis, the patient would contribute 1 person-day of follow-up with 0 old units and for the remainder of the follow-up would be categorized as having received 4 old units. The purpose of this analytic approach was to allow patients to become more exposed with time, thereby preventing patients who received additional transfusions from being allocated to the mixed storage time category. Instrumental Variable Approach Finally, because we recognized that the mean storage time of transfused blood units is strongly related to patient blood group (Appendix Figure), and that blood group in turn should at most be only weakly associated with mortality, we performed an instrumental variable analysis. This type of analysis is used to remove the effect of hidden bias in observational studies (43, 44) through the use of an instrumental variable, which is associated with the exposure of interest but has no independent effect on the outcome (45). Here, we used rhesus (RhD) status as an instrument, which in this setting should be a naturally randomized factor. Clinical characteristics, such as morbidity and disease severity, should be similar in patients with blood group Aand those with A+. As such, although RhD status strongly predicts mean storage (Appendix Figure), it should not independently predict mortality. In this analysis, we excluded patients born outside Sweden or Denmark, because other genetic variance might have confounded the results. Statistical Analysis In the first 2 sets of analyses, we estimated the 30-day and 1-year hazard ratios (HRs) of death in relation to length of RBC storage by using Cox proportional hazard regression models. In the discrete exposure group models, exposure was considered a categorical term. Recognizing that fresh units also have been implicated as having deleterious effects on patient survival (31), we used the 10- to 19-day storage time category as a reference. Follow-up began on the day of the last transfusion in the exposure ascertainment period. In the time-dependent models, we used variables for the total number of old units (stored 30 days) or very old units (stored 35 days) as exposure. These variables were fitted as linear terms or categorized as 0, 1 to 3, 4 to 6, or more than 6 old or very old units. Here, follow-up started on the day of the first transfusion. Except for the time-dependent model allowing exposure to change throughout follow-up, the statistical models for the first 2 sets of analyses were similar. In addition to parameters for exposure, both statistical models also included the cumulative number of RBC transfusions (as a restricted cubic spline with 6 knots), ABORhD blood group (as a categorical variable), year of first transfusion (as a categorical variable), age (as a restricted cubic spline with 5 knots), sex (as a categorical variable), whether the patient had previously received a transfusion (as a binary variable), weekday of first transfusion (as a categorical variable), number of platelets and plasma transfusions (as binary factors),

Relative effects of plasma, fibrinogen concentrate, and factor XIII on ROTEM coagulation profiles in an in vitro model of massive transfusion in trauma

Abstract Massive traumatic haemorrhage is aggravated through the development of trauma-induced coagulopathy, which is managed by plasma transfusion and/or fibrinogen concentrate administration. It is yet unclear whether these treatments are equally potent in ensuring adequate haemostasis, and whether additional factor XIII (FXIII) administration provides further benefits. In this study, we compared ROTEM whole blood coagulation profiles after experimental massive transfusion with different transfusion regimens in an in vitro model of dilution- and transfusion-related coagulopathy. Healthy donor blood was mixed 1 + 1 with six different transfusion regimens. Each regimen contained RBC, platelet concentrate, and either fresh frozen plasma (FFP) or Ringer’s acetate (RA). The regimens were further augmented through addition of a low- or medium-dose fibrinogen concentrate and FXIII. Transfusion with FFP alone was insufficient to maintain tissue-factor activated clot strength, coincidental with a deficiency in fibrin-based clot strength. Fibrinogen concentrate conserved, but did not improve coagulation kinetics and overall clot strength. Only combination therapy with FFP and low-dose fibrinogen concentrate improved both coagulation kinetics and fibrin-based clot strength. Administration of FXIII did not result in an improvement of clot strength. In conclusion, combination therapy with both FFP and low-dose fibrinogen concentrate improved clotting time and produced firm clots, representing a possible preferred first-line regimen to manage trauma-induced coagulopathy when RBC and platelets are also transfused. Further research is required to identify optimal first-line transfusion fluids for massive traumatic haemorrhage.