Abstract
Background and objectives: Hemodialysis patients with type 2 diabetes exhibit an excessive cardiovascular risk and regularly receive heparin. We tested whether antibodies to the platelet factor 4–heparin complex (PF4-H-AB) contribute to outcome.
Design, setting, participants, & measurements: In 1255 hemodialysis patients with type 2 diabetes, the German Diabetes Dialysis Study evaluated the effect of atorvastatin (20 mg/d) versus placebo. In a post hoc analysis, the association among PF4-H-ABs, biochemistry, and prespecified, centrally adjudicated end points (combined cardiovascular end point [CVE], all-cause mortality, sudden death, myocardial infarction, stroke) was investigated.
Results: During 4 years, 460 patients reached the CVE; 605 died, 159 of sudden death. Myocardial infarction and stroke occurred in 199 and 97 patients, respectively. Positive PF4-H-AB status was found in 231 (18.7%) of 1236 tested patients and was associated with lower albumin, higher C-reactive protein, and arrhythmia. In a multivariate model adjusted for demographics, comorbidities, and biochemistry, PF4-H-ABs were associated with sudden death. No significant association between PF4-H-ABs and all-cause mortality, myocardial infarction, stroke, or the CVE was observed. Detecting an interaction between acetylsalicylic acid and PF4-H-ABs regarding sudden death and mortality, we found that the association between PF4-H-ABs and outcomes was restricted to patients with acetylsalicylic acid use, most likely because of indication bias.
Conclusions: In hemodialysis patients who have type 2 diabetes and are treated with acetylsalicylic acid, PF4-H-ABs are associated with sudden and all-cause death. Further studies are needed to elucidate this association.
Annual mortality rates in hemodialysis patients with type 2 diabetes are extremely high (250 to 300 per 1000 patient-years) (1). Treatment of established cardiovascular risk factors shows only limited benefit; therefore, the existence of other nontraditional risk factors has been assumed (2,3).
One of these emerging risk factors might be antibodies to the platelet factor 4–heparin complex (PF4-H-ABs), which can lead to platelet activation and loss of antithrombogenic endothelial surface properties (4). According to the “Iceberg Model” (5), only a minority of patients with increased PF4-H-ABs at the apex of the “iceberg” will develop the full-blown picture of heparin-induced thrombocytopenia type 2 with arterial and venous thromboses. Investigations have demonstrated that patients who are at the broad base of this iceberg and have no signs of thrombosis and have normal platelet counts might also have an increased cardiovascular risk (6).
Hemodialysis patients are repeatedly exposed to heparin and, therefore, may have an increased risk for formation of PF4-H-ABs. Indeed, some investigators found PF4-H-ABs to be increased and associated with cardiovascular events (7–10).
We performed a post hoc analysis of the German Diabetes Dialysis Study (Die Deutsche Diabetes Dialyse Studie [4D Study]) to evaluate the efficacy and safety of atorvastatin in 1255 hemodialysis patients with type 2 diabetes (11). The main question was whether the extreme cardiovascular morbidity and mortality in these patients could be attributed to circulating PF4-H-ABs.
Materials and Methods
Study Design and Participants
Design and methods of the 4D Study have previously been reported (11,12). The 4D Study was a randomized, multicenter trial of 1255 patients who had type 2 diabetes, were 18 to 80 years of age, and had a previous duration of hemodialysis of <2 years. Between March 1998 and October 2002, patients were recruited in 178 dialysis units throughout Germany. After a run-in period of 4 weeks, patients were randomly assigned to receive double-blind treatment with either 20 mg of atorvastatin once daily (n = 619) or placebo (n = 636). Data were recorded at 4 weeks and every 6 months. At each follow-up, a blood sample was taken and information was recorded about any suspected study end point or other serious adverse experience. Further details related to study end points were sought from family doctors, emergency physicians, hospitals, and local health authorities.
Outcome Measures
End points were evaluated by a specialized committee blinded to study treatment, according to prespecified criteria (11–13). The primary study outcome was a composite of death from cardiac causes, myocardial infarction (MI), and stroke, whichever occurred first (composite cardiovascular end point [CVE]). Sudden death, stroke, MI, and death from any cause were defined as secondary outcomes. These five end points were the outcome measures in this post hoc analysis.
Laboratory Procedures
All laboratory measurements of the initial clinical trial and the post hoc analyses of PF4-H-ABs and high-sensitivity C-reactive protein (CRP) were performed centrally at the Department of Clinical Chemistry, University of Freiburg (Freiburg, Germany). Total PF4-H-ABs (IgG, IgA, and IgM) were measured by a solid-phase ELISA in duplicate (PF4 ENHANCED; GTI, Waukesha, WI). Results were reported as OD. An OD >0.4 was considered positive for statistical analyses. Interassay coefficients of variance for PF4-H-ABs were <7%. Blood samples were taken before start of dialysis and administration of heparin or additional drugs.
Statistical Analysis
Patient data are presented according to baseline PF4-H-AB status (negative or positive). P values for comparison between groups were derived from a general linear model for continuous and a logistic regression analysis for categorical variables adjusted for age and gender, as appropriate.
Kaplan-Meier estimates for cumulative incidences were calculated for all outcomes. The Cox proportional hazards regression model was used to estimate relative risks and corresponding 95% confidence intervals (CIs). The association of baseline PF4-H-ABs with outcome was analyzed as a continuous variable (logarithmically transformed, because values were not normally distributed) and as categorical variables: (1) Negative or positive (patients with negative PF4-H-AB status served as reference) and (2) quartiles of PF4-H-Abs (the first quartile served as the reference). The interaction of PF4-H-AB status and study treatment regarding the five end points was tested within the Cox regression model.
The following explanatory variables were considered for inclusion in the various Cox proportional hazards regression models: Treatment assignment, gender, age, phosphate, LDL, hemoglobin, glycated hemoglobin, CRP, smoking, systolic and diastolic BP, body mass index, ultrafiltration volume, duration of dialysis, hemodialysis shunt, history of stroke/transient ischemic attack, coronary artery disease (MI, coronary artery bypass grafting, percutaneous coronary intervention, and angiographically documented coronary artery disease), arrhythmia, peripheral vascular disease, and congestive heart failure (predominantly New York Heart Association class II). A stepwise selection procedure was used to determine the variables that were included in the final model (treatment assignment was always included), separately for each end point. This procedure starts examining the variable with the largest adjusted χ2 statistic and adds variables to the model if P ≤ 0.25 and retains them if P ≤ 0.15 in the following steps. In the final analysis P ≤ 0.05 was considered to be significant (in an exploratory sense). All p values are reported two-sided. Analyses were done using SAS 9.1.3 (SAS Institute, Chicago, IL).
Results
Of 1255 patients who took part in the 4D Study, 1236 (626 on placebo and 610 on atorvastatin) had a baseline PF4-H-AB measurement. The mean follow-up-period of the 4D Study was 3.96 years (median 4.00 years) for patients on atorvastatin and 3.91 years (median 4.08 years) for those on placebo. During follow-up, 605 patients died, 159 of sudden death. Furthermore, 460 patients reached the CVE (cardiac death, nonfatal MI, and stroke), with MI and stroke occurring in 199 and 97 patients, respectively. Increased PF4-H-ABs were found in 231 (18.7%) of 1236 patients analyzed.
Patient Characteristics
Patients with positive PF4-H-AB status showed lower albumin and higher CRP concentrations compared with those with negative PF4-H-AB status. Furthermore, they more often experienced arrhythmias, were less likely to present with a hemodialysis shunt, and were on maintenance hemodialysis for a shorter period (Table 1).
Table 1.
Patient characteristics according to baseline PF4-H-AB status
| Variable | Group 1: Negative PF4-H-AB Test (n = 1005) | Group 2: Positive PF4-H-AB Test (n = 231) | Pa |
|---|---|---|---|
| Age (years; mean [SD]) | 65.5 ± 8.3 | 66.3 ± 8.2 | 0.299 |
| Gender (male/female) | 548/457 | 118/113 | 0.495 |
| Ever smoked (% [n]) | 42 (418) | 36 (83) | 0.213 |
| BMI (kg/m2; mean [SD]) | 27.4 ± 4.8 | 28.0 ± 4.7 | 0.067 |
| SBP (mmHg; mean [SD]) | 146 ± 22 | 145 ± 22 | 0.460 |
| DBP (mmHg; mean [SD]) | 76 ± 11 | 76 ± 11 | 0.817 |
| Ultrafiltration volume (kg; median [25th to 75th percentiles]) | 2.0 (1.5 to 3.0) | 2.0 (1.0 to 3.0) | 0.443 |
| Hemodialysis shunt (% [n]) | 94 (944) | 90 (208) | 0.045 |
| Central venous dialysis catheter (% [n]) | 6 (60) | 10 (23) | 0.045 |
| Time on dialysis (months; mean [SD]) | 8.6 ± 7.1 | 6.7 ± 5.8 | <0.001 |
| History ofb | |||
| arrhythmia (% [n]) | 17 (174) | 25 (57) | 0.017 |
| MI, CABG, PCI, or CHDc (% [n]) | 29 (293) | 30 (69) | 0.811 |
| CHF (% [n])d | 34 (345) | 39 (91) | 0.210 |
| stroke or TIA | 18 (158) | 16 (37) | 0.383 |
| peripheral vascular disease (% [n]) | 45 (453) | 42 (98) | 0.512 |
| Hemoglobin (g/dl; mean [SD]) | 10.9 ± 1.3 | 10.8 ± 1.4 | 0.394 |
| Platelets (1000/μl; mean [SD]) | 257 ± 80 | 259 ± 84 | 0.608 |
| HbA1c (%; mean [SD]) | 6.74 ± 1.27 | 6.63 ± 1.22 | 0.225 |
| Phosphate (mg/L; mean [SD]) | 6.02 ± 1.59 | 6.08 ± 1.66 | 0.453 |
| Albumin (g/dl; mean [SD]) | 3.83 ± 0.31 | 3.77 ± 0.28 | 0.016 |
| CRP (mg/L; median [25th to 75th percentiles]) | 4.9 (2.2 to 12.0) | 5.6 (2.6 to 14.2) | 0.042 |
| LDL cholesterol (mg/dl; mean [SD]) | 125 ± 29 | 128 ± 32 | 0.211 |
To convert hemoglobin values to millimoles per liter, multiply by 0.62. To convert platelet values to platelets per liter, multiply by 106. To convert values for phosphate to millimoles per liter, multiply by 0.32. To convert values for LDL cholesterol to millimoles per liter, multiply by 0.03. BMI, body mass index; CABG, coronary artery bypass grafting surgery; CHD, coronary heart disease; CHF, congestive heart failure; DBP, diastolic BP; HbA1c, glycated hemoglobin; PCI, percutaneous coronary intervention; SBP, systolic BP; TIA, transient ischemic attack.
P values for comparison between groups of patients with positive and negative PF4-H-AB test at baseline were derived from a general linear model for continuous variables or logistic regression for categorical variables both adjusted for age and gender, as appropriate.
Types of disease and intervention are not mutually exclusive.
Documented by coronary angiography.
Predominantly New York Heart Association class II.
Baseline PF4-H-ABs and Outcome
Analyses with the PF4-H-ABs as continuous variable showed a significant increase in the risk for sudden death per unit increase in log PF4-H-ABs (hazard ratio [HR] 1.41; 95% CI 1.03 to 1.92; P = 0.03). Accordingly, patients with a positive PF4-H-AB status at baseline tended to be at higher risk for sudden death (HR 1.39; 95% CI 0.95 to 2.03; P = 0.09) compared with those with negative PF4-H-AB status at baseline. No statistically significant association between the PF4-H-ABs (continuous variable and categorization into positive or negative) and any other outcome measure was observed (Table 2).
Table 2.
Risk for all-cause death, CVE, sudden death, stroke, and MI by baseline PF4-H-ABs
| Parameter | Patients with |
|
|---|---|---|
| Negative PF4-H-AB Status (n = 1005) | Positive PF4-H-AB Status (n = 231) | |
| All-cause death | ||
| no. of events during study | 486 | 119 |
| Kaplan-Meier estimate (95% CI)a | 0.49 (0.45 to 0.52) | 0.54 (0.47 to 0.62) |
| adjusted HR (95% CI; P)b | c | 1.15 (0.94 to 1.41; 0.18) |
| adjusted HR (95% CI; P)b with PF4-H-ABs as continuous variable | 1.12 (0.95 to 1.32; 0.19) | |
| CVE (MI, cardiac death, and stroke) | ||
| no. of events during study | 372 | 88 |
| Kaplan-Meier estimate (95% CI)a | 0.43 (0.39 to 0.47) | 0.44 (0.36 to 0.52) |
| adjusted HR (95% CI; P)b | c | 1.09 (0.86 to 1.38; 0.48) |
| adjusted HR (95% CI; P)b with PF4-H-ABs as continuous variable | 1.08 (0.89 to 1.30; 0.44) | |
| Sudden death | ||
| no. of events during study | 124 | 35 |
| Kaplan-Meier estimate (95% CI)a | 0.16 (0.13 to 0.19) | 0.19 (0.12 to 0.25) |
| adjusted HR (95% CI; P)b | c | 1.39 (0.95 to 2.03; 0.09) |
| adjusted HR (95% CI; P)b with PF4-H-ABs as continuous variable | 1.41 (1.03 to 1.92; 0.03) | |
| Stroke | ||
| no. of events during study | 82 | 15 |
| Kaplan-Meier estimate (95% CI)a | 0.10 (0.08 to 0.13) | 0.08 (0.04 to 0.13) |
| adjusted HR (95% CI; P)b | c | 0.79 (0.46 to 1.38; 0.41) |
| adjusted HR (95% CI; P)b with PF4-H-ABs as continuous variable | 0.85 (0.54 to 1.32; 0.46) | |
| MI | ||
| no. of events during study | 163 | 36 |
| Kaplan-Meier estimate (95% CI)a | 0.22 (0.19 to 0.25) | 0.23 (0.15 to 0.31) |
| adjusted HR (95% CI; P)b | c | 1.03 (0.71 to 1.48; 0.89) |
| adjusted HR (95% CI; P)b with PF4-H-ABs as continuous variable | 0.97 (0.72 to 1.30; 0.81) | |
Unadjusted Kaplan-Meier estimates at the end of year 4.
Explanatory variables were selected by a stepwise process with adjustment for treatment assignment (was always included in the final model), gender, age, phosphate, LDL, hemoglobin, glycated hemoglobin, CRP, ever smoked, systolic and diastolic BP, body mass index, ultrafiltration volume, duration of dialysis, hemodialysis shunt, history of stroke/transient ischemic attack, coronary artery disease (MI, coronary artery bypass grafting, percutaneous coronary intervention, and angiographically documented coronary artery disease), arrhythmia, peripheral vascular disease, and congestive heart failure (predominantly New York Heart Association class II).
Patients with negative PF4-H-AB status (OD <0.4) at baseline served as the reference.
The definition of a positive antibody status (OD >0.4) is based on correlations with the syndrome of heparin-induced thrombocytopenia. The appropriate threshold in a population without thrombocytopenia is not known; therefore, we further analyzed the association between quartiles of PF4-H-ABs and outcome (Table 3), and we looked for the optimal cutoff value to select a high-risk group regarding sudden death, which was an OD of 0.341 according to the minimal P value in the multivariate Cox model (HR 1.54; 95% CI 1.11 to 2.14; P = 0.01) (Figure 1).
Table 3.
Risk for all-cause death, CVE, sudden death, stroke, and MI by quartiles of baseline PF4-H-ABs
| Parameter | Quartile 1 ≤0.208 PF4-H-ABs (n = 310) | Quartile 2 >0.208 to ≤0.268 PF4-H-ABs (n = 309) | Quartile 3 >0.268 to ≤0.362 PF4-H-ABs (n = 309) | Quartile 4 >0.362 PF4-H-ABs (n = 308) |
|---|---|---|---|---|
| All-cause death | ||||
| no. of events during study | 152 | 152 | 142 | 159 |
| adjusted HR (95% CI)a | b | 1.27 (1.01 to 1.59) | 1.12 (0.88 to 1.41) | 1.27 (1.02 to 1.60) |
| P | 0.05 | 0.36 | 0.04 | |
| CVE (MI, cardiac death, and stroke) | ||||
| no. of events during study | 120 | 108 | 107 | 125 |
| adjusted HR (95% CI)a | b | 1.12 (0.86 to 1.46) | 1.07 (0.82 to 1.39) | 1.28 (1.00 to 1.66) |
| P | 0.41 | 0.62 | 0.06 | |
| MI | ||||
| no. of events during study | 55 | 45 | 44 | 55 |
| adjusted HR (95% CI)a | b | 0.97 (0.65 to 1.45) | 0.96 (0.64 to 1.44) | 1.23 (0.84 to 1.79) |
| P | 0.89 | 0.85 | 0.29 | |
| Sudden death | ||||
| no. of events during study | 39 | 31 | 40 | 49 |
| adjusted HR (95% CI)a | b | 0.98 (0.61 to 1.57) | 1.12 (0.72 to 1.75) | 1.50 (0.98 to 2.30) |
| P | 0.92 | 0.62 | 0.06 | |
| Stroke | ||||
| no. of events during study | 29 | 26 | 18 | 24 |
| adjusted HR (95% CI)a | b | 1.11 (0.65 to 1.90) | 0.75 (0.41 to 1.36) | 0.94 (0.55 to 1.63) |
| P | 0.70 | 0.34 | 0.84 |
Explanatory variables were selected by a stepwise process (atorvastatin treatment was always included in the final model) with adjustment for treatment assignment (was always included in the final model), gender, age, phosphate, LDL, hemoglobin, glycated hemoglobin (HbA1c), CRP, ever smoked, systolic and diastolic BP, body mass index, ultrafiltration volume, duration of dialysis, hemodialysis shunt, history of stroke/transient ischemic attack, coronary artery disease (MI, coronary artery bypass grafting, percutaneous coronary intervention, and angiographically documented coronary artery disease), arrhythmia, peripheral vascular disease, and congestive heart failure (predominantly New York Heart Association class II).
The first quartile was used as the reference for each of the other three.
Figure 1.
Kaplan-Meier estimate for time to sudden death according to baseline PF4-H-AB status (OD >0.341 was chosen to define positive PF4-H-AB testing).
Furthermore, we performed subgroup analyses of patients with high (more than the median) and low (less than or equal to the median) platelets, which showed that the HR of PF4-H-AB positive versus PF4-H-AB negative for sudden death tended to be higher in patients with low platelets (patients with platelets <249 500/μl: HR 1.42 [95% CI 0.82 to 2.46; P = 0.21]; patients with platelets >249 500/μl: HR 1.16 [95% CI 0.68 to 1.98; P = 0.58]) compared with those with high platelets.
Regarding MI, we found a significant interaction between PF4-H-AB status (negative or positive) and treatment (P = 0.04); therefore, the association between PF4-H-AB status and MI was evaluated in each treatment group separately: No significant association between positive PF4-H-AB test and the risk for MI was found (placebo-treated patients only: HR 1.50 [95% CI 0.94 to 2.38; P = 0.09]; atorvastatin-treated patients only: HR 0.64 [95% CI 0.35 to 1.18; P = 0.15]). There was no significant interaction between treatment and PF4-H-AB status with respect to the other outcome measures.
No significant influence of low molecular weight heparin, acetylsalicylic acid or anticoagulants on sudden death was observed; however, we found an interaction among acetylsalicylic acid use at baseline, PF4-H-ABs, and outcome. The P values for the interaction were as follows: P < 0.01 for all-cause death, P = 0.12 for CVE, P = 0.01 for sudden death, P = 0.47 for stroke, and P = 0.57 for MI. Consequently, the associations between PF4-H-ABs, sudden death, and all-cause mortality were analyzed separately for patients with and without acetylsalicylic acid use at baseline. Patients who were on acetylsalicylic acid and showed a positive PF4-H-AB status were at higher risk for dying compared with those who had negative PF4-H-AB status and were on acetylsalicylic acid (all-cause death: HR 1.72 [95% CI 1.33 to 2.24; P < 0.01]; sudden death: HR 2.20 [95% CI 1.36 to 3.56; P < 0.01]). In patients who were not treated with acetylsalicylic acid, the risk for dying (all-cause mortality and sudden death) did not differ significantly between those with positive and those with negative PF4-H-AB status (all cause death: HR 0.77 [95% CI 0.57 to 1.04; P = 0.08]; sudden death: HR 0.75 [95% CI 0.39 to 1.44; P = 0.38]).
Discussion
We tested the hypothesis that PF4-H-ABs were linked to the excessive cardiovascular risk in patients who had type 2 diabetes and were on long-term hemodialysis in a large prospective patient cohort with a high incidence of prespecified and centrally adjudicated end points during 4 years of follow-up. For these patients, PF4-H-ABs at baseline were found to be predictive of sudden death. No statistically significant association between PF4-H-AB status and combined cardiovascular events, MI, all-cause death, and stroke was detected. PF4-H-ABs were a common finding and associated with signs of inflammation.
Sudden death was the major component of mortality in the 4D Study. Similar information has been available from the 2007 data set of the US Renal Data System for the general hemodialysis population showing that only 9% of deaths have been caused by acute MI or atherosclerotic heart disease, whereas 27% have been classified as cardiac arrest or arrhythmia (1). Corresponding data from the 4D Study were 11% for coronary heart disease deaths and 26% for sudden deaths (13).
In this analysis, no significant association between PF4-H-ABs and MI or stroke was identified; therefore, it was unlikely that macrovascular disease (e.g., MI) has been responsible for the increase in the risk for sudden death. Consequently, alternative causes came into focus. Indeed, microvessel disease and cardiac fibrosis were found to be a typical abnormality of cardiac structure in uremia (14–16). Wall thickening of intramyocardial arteries and inadequate capillary density were described to restrict compensatory vasodilation and the ability of the heart to cope with increased oxygen demand (17). This might have been of special importance in the presence of PF4-H-Abs, which are known to induce platelet activation (6) and to act via the F(ab)2 fragment of IgG PF4-H-ABs binding to and activating microvascular endothelial cells (18), thereby potentially aggravating the already impaired cardiac perfusion and promoting cardiac fibrosis. Cardiac fibrosis had both mechanical and electrical sequelae affecting cardiovascular prognosis (14). Fibrosis was found to reduce the ventricular compliance and to promote arrhythmia. Fibrous tissue with high electrical resistance, interposed between cardiomyocytes, has been described to cause local delay in the spread of the action potential favoring the development of reentry types of arrhythmias (19,20).
Data with respect to macrovascular disease and all-cause death were not unequivocal. Previous studies found that patients who had acute coronary syndrome and tested positive for PF4-H-ABs were more likely to experience an MI (21) and a combined end point of death, revascularization, MI, and stroke (21,22); however, for patients who underwent percutaneous coronary intervention and for patients who were on dialysis, no significant increase in thrombotic events in antibody-positive patients was found (23,24). With respect to all-cause mortality, two studies showed PF4-H-ABs to be predictive of death in hemodialysis patients (7,10). The Choices for Healthy Outcomes in Caring for End-Stage Renal Disease study (24) did not confirm these findings. In this analysis, a significant association between PF4-H-ABs and all-cause death was found when PF4-H-ABs were categorized into quartiles that identified patients within the first quartile being at lowest risk for all-cause death. No significant association was detected when PF4-H-ABs were analyzed as a continuous or categorical (positive or negative PF4-H-AB status) variable; however, we detected an interaction between acetylsalicylic acid use and PF4-H-ABs regarding sudden and all-cause death most likely as a result of indication bias. Patients who are treated with acetylsalicylic acid are supposed to be at higher risk. In these patients, adding another risk factor might result in a significant increase in risk.
In line with previous work (7,25), we identified increased PF4-H-ABs in a high percentage (18.7%) of patients. We underline that the study took place before the problems with chondroitin sulfate, but the actual purity of heparin that was used from 1998 through 2002 remains unclear. It is important, because oversulfated chondroitin sulfate is associated with an increased prevalence of PF4-H-ABs in hemodialysis patients (26).
Patients with positive PF4-H-AB status presented with lower albumin and higher CRP concentrations compared with those with negative PF4-H-AB status. PF4-H-ABs may have elicited an inflammatory response (4,27). Alternatively, they were markers of a general inflammatory process. Evidence was found to support this: Higher levels of CRP, myeloperoxidase, and soluble adhesion molecules have been demonstrated in patients with heparin antibodies (28,29). In this context, the higher prevalence of dialysis catheters in patients with positive PF4-H-AB status might as well have played a role, because catheter use was associated with lower albumin (mean albumin 3.71 versus 3.83 g/dl; P < 0.01) and higher CRP (mean logCRP 1.94 versus 1.64 mg/L; P = 0.03). It remains unclear whether this results from selection bias or from catheter use; however, increased markers of microinflammation have generally been associated with a higher risk for mortality, cardiovascular events, and also sudden death (30–32). Our multivariate analysis was adjusted for this.
In patients with positive PF4-H-AB status, we found a lower prevalence of hemodialysis shunts. Whether this was due to a higher rate of preceding fistula thromboses (9,33) or a result of the permanent exposition to heparin to prevent thrombus formation within the catheter is unknown.
Our finding of more arrhythmias in patients with positive PF4-H-AB status could have pointed to a myocardial microcirculation already compromised by platelet clots (34).
Reasons for the negative correlation between PF4-H-AB status and the time that patients had been on hemodialysis (a similar association has been described [9]) need to be elucidated further; however, this was in line with our finding that increasing PF4-H-ABs were associated with a higher risk for sudden death. This was the case at each time point. Consequently, patients within the antibody-positive group might have died earlier and stayed shorter on dialysis. Another reason might have been that with repeat exposure, tolerance to heparin developed.
Our study had several limitations. We performed a post hoc analysis within a selected cohort of German patients who had type 2 diabetes and were on hemodialysis. Results may not necessarily be generalizable to other populations, and cause cannot be inferred from these associations. Although our analyses have been adjusted carefully for multiple variables, residual confounding cannot be excluded. Furthermore, we detected an interaction between acetylsalicylic acid and PF4-H-ABs regarding sudden death. The association between PF4-H-ABs and sudden death was restricted to patients with acetylsalicylic acid use, most likely because of indication bias. Patients who are treated with acetylsalicylic acid are supposed to be at higher risk; adding another risk factor might result in a significant increase in risk. An effect of acetylsalicylic acid cannot be excluded; however, there is no plausible pathophysiologic explanation. Our assay analyzed the PF4-H-ABs without detecting the IgG-specific PF4-H-ABs. Some investigators recommended the combination of a functional test of platelet activation in conjunction with measurements of PF4-H-ABs to increase specificity (35). Because of the post hoc study design, we did not have functional data on platelet activation because fresh platelet preparations were not available; however, these limitations should have been balanced by the large study population, the high incidence of prespecified and centrally adjudicated end points, and the long follow-up period.
Conclusions
This study found PF4-H-ABs to be associated with sudden and all-cause death in patients who had type 2 diabetes, were on hemodialysis, and were treated with acetylsalicylic acid, whereas it did not support an association with MI and stroke. These results further endorse a possible link between increased PF4-H-ABs and microinflammation. PF4-H-ABs are a common finding in hemodialysis patients with type 2 diabetes. Whether treating these patients with nonheparin anticoagulants would decrease the risk for sudden death needs to be tested in an intervention trial.
Disclosures
None.
Acknowledgments
We thank Mitsubishi Pharma for funding the PF4-H-AB reagents. We thank all 4D Study investigators and nurses (for the complete list, go to http://www.uni-wuerzburg.de/nephrologie); without their collaboration, this article would not have been written. Special thanks go to the event committee (J. Mann (chair), J. Bommer, P. Schanzenbächer, P. Schollmeyer, and M. Schartl) the laboratory staff at the Universities of Freiburg and Würzburg, and the interdisciplinary center of clinical research at the University of Würzburg.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
References
- 1.US Renal Data System: USRDS 2007 Annual Data Report, Bethesda, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2007 [Google Scholar]
- 2.Johnson DW, Craven AM, Isbel NM: Modification of cardiovascular risk in hemodialysis patients: An evidence-based review. Hemodial Int 11: 1– 14, 2007 [DOI] [PubMed] [Google Scholar]
- 3.Ronco C, Bowry S, Tetta C: Dialysis patients and cardiovascular problems: Can technology help solve the complex equation? Blood Purif 24: 39– 45, 2006 [DOI] [PubMed] [Google Scholar]
- 4.Walenga JM, Jeske WP, Prechel MM, Bakhos M: Newer insights on the mechanism of heparin-induced thrombocytopenia. Semin Thromb Hemost 30[ Suppl 1]: 57– 67, 2004 [DOI] [PubMed] [Google Scholar]
- 5.Warkentin T, Kelton J: Interaction of heparin with platelets, including heparin-induced thrombocytopenia. In: Low-Molecular-Weight Heparins in Prophylaxis and Therapy of Thromboembolic Diseases, edited by Bounameaux H.New York, Marcel Dekker, 1994, pp 75– 127 [Google Scholar]
- 6.Stribling WK, Slaughter TF, Houle TT, Sane DC: Beyond the platelet count: Heparin antibodies as independent risk predictors. Am Heart J 153: 900– 906, 2007 [DOI] [PubMed] [Google Scholar]
- 7.Carrier M, Rodger MA, Fergusson D, Doucette S, Kovacs MJ, Moore J, Kelton JG, Knoll GA: Increased mortality in hemodialysis patients having specific antibodies to the platelet factor 4-heparin complex. Kidney Int 73: 213– 219, 2008 [DOI] [PubMed] [Google Scholar]
- 8.Mureebe L, Coats RD, Silliman WR, Shuster TA, Nichols WK, Silver D: Heparin-associated antiplatelet antibodies increase morbidity and mortality in hemodialysis patients. Surgery 136: 848– 853, 2004 [DOI] [PubMed] [Google Scholar]
- 9.Nakamoto H, Shimada Y, Kanno T, Wanaka K, Matsuo T, Suzuki H: Role of platelet factor 4-heparin complex antibody (HIT antibody) in the pathogenesis of thrombotic episodes in patients on hemodialysis. Hemodial Int 9[ Suppl 1]: S2– S7, 2005 [DOI] [PubMed] [Google Scholar]
- 10.Pena de la Vega L, Miller RS, Benda MM, Grill DE, Johnson MG, McCarthy JT, McBane RD: Association of heparin-dependent antibodies and adverse outcomes in hemodialysis patients: A population-based study. Mayo Clin Proc 80: 995– 1000, 2005 [DOI] [PubMed] [Google Scholar]
- 11.Wanner C, Krane V, Marz W, Olschewski M, Asmus HG, Kramer W, Kuhn KW, Kutemeyer H, Mann JF, Ruf G, Ritz E: Randomized controlled trial on the efficacy and safety of atorvastatin in patients with type 2 diabetes on hemodialysis (4D study): Demographic and baseline characteristics. Kidney Blood Press Res 27: 259– 266, 2004 [DOI] [PubMed] [Google Scholar]
- 12.Wanner C, Krane V, Ruf G, Marz W, Ritz E: Rationale and design of a trial improving outcome of type 2 diabetics on hemodialysis. Die Deutsche Diabetes Dialyse Studie Investigators. Kidney Int Suppl 71: S222– S226, 1999 [DOI] [PubMed] [Google Scholar]
- 13.Wanner C, Krane V, Marz W, Olschewski M, Mann JF, Ruf G, Ritz E: Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis. N Engl J Med 353: 238– 248, 2005 [DOI] [PubMed] [Google Scholar]
- 14.Tornig J, Amann K, Ritz E, Nichols C, Zeier M, Mall G: Arteriolar wall thickening, capillary rarefaction and interstitial fibrosis in the heart of rats with renal failure: The effects of ramipril, nifedipine and moxonidine. J Am Soc Nephrol 7: 667– 675, 1996 [DOI] [PubMed] [Google Scholar]
- 15.Amann K, Breitbach M, Ritz E, Mall G: Myocyte/capillary mismatch in the heart of uremic patients. J Am Soc Nephrol 9: 1018– 1022, 1998 [DOI] [PubMed] [Google Scholar]
- 16.Ritz E, Wanner C: The challenge of sudden death in dialysis patients. Clin J Am Soc Nephrol 3: 920– 929, 2008 [DOI] [PubMed] [Google Scholar]
- 17.Dikow R, Kihm LP, Zeier M, Kapitza J, Tornig J, Amann K, Tiefenbacher C, Ritz E: Increased infarct size in uremic rats: reduced ischemia tolerance? J Am Soc Nephrol 15: 1530– 1536, 2004 [DOI] [PubMed] [Google Scholar]
- 18.Blank M, Shoenfeld Y, Tavor S, Praprotnik S, Boffa MC, Weksler B, Walenga MJ, Amiral J, Eldor A: Anti-platelet factor 4/heparin antibodies from patients with heparin-induced thrombocytopenia provoke direct activation of microvascular endothelial cells. Int Immunol 14: 121– 129, 2002 [DOI] [PubMed] [Google Scholar]
- 19.Chen PS, Chou CC, Tan AY, Zhou S, Fishbein MC, Hwang C, Karagueuzian HS, Lin SF: The mechanisms of atrial fibrillation. J Cardiovasc Electrophysiol 17[ Suppl 3]: S2– S7, 2006 [DOI] [PubMed] [Google Scholar]
- 20.Khan R, Sheppard R: Fibrosis in heart disease: Understanding the role of transforming growth factor-beta in cardiomyopathy, valvular disease and arrhythmia. Immunology 118: 10– 24, 2006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Williams RT, Damaraju LV, Mascelli MA, Barnathan ES, Califf RM, Simoons ML, Deliargyris EN, Sane DC: Anti-platelet factor 4/heparin antibodies: An independent predictor of 30-day myocardial infarction after acute coronary ischemic syndromes. Circulation 107: 2307– 2312, 2003 [DOI] [PubMed] [Google Scholar]
- 22.Mattioli AV, Bonetti L, Sternieri S, Mattioli G: Heparin-induced thrombocytopenia in patients treated with unfractionated heparin: Prevalence of thrombosis in a 1 year follow-up. Ital Heart J 1: 39– 42, 2000 [PubMed] [Google Scholar]
- 23.Gluckman TJ, Segal JB, Fredde NL, Saland KE, Jani JT, Walenga JM, Prechel MM, Citro KM, Zidar DA, Fox E, Schulman SP, Kickler TS, Rade JJ: Incidence of antiplatelet factor 4/heparin antibody induction in patients undergoing percutaneous coronary revascularization. Am J Cardiol 95: 744– 747, 2005 [DOI] [PubMed] [Google Scholar]
- 24.Asmis LM, Segal JB, Plantinga LC, Fink NE, Kerman JS, Kickler TS, Coresh J, Gardner LB: Heparin-induced antibodies and cardiovascular risk in patients on dialysis. Thromb Haemost 100: 498– 504, 2008 [PubMed] [Google Scholar]
- 25.Palomo I, Pereira J, Alarcon M, Diaz G, Hidalgo P, Pizarro I, Jara E, Rojas P, Quiroga G, Moore-Carrasco R: Prevalence of heparin-induced antibodies in patients with chronic renal failure undergoing hemodialysis. J Clin Lab Anal 19: 189– 195, 2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Adiguzel C, Bansal V, Litinas E, Cunanan J, Iqbal O, Nelson K, Kannan M, Hoppensteadt D, Fareed J: Increased prevalence of antiheparin platelet factor 4 antibodies in patients may be due to contaminated heparin. Clin Appl Thromb Hemost 15: 145– 151, 2009 [DOI] [PubMed] [Google Scholar]
- 27.Xiao Z, Visentin GP, Dayananda KM, Neelamegham S: Immune complexes formed following the binding of anti-platelet factor 4 (CXCL4) antibodies to CXCL4 stimulate human neutrophil activation and cell adhesion. Blood 112: 1091– 1100, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Khairy M, Lasne D, Brohard-Bohn B, Aiach M, Rendu F, Bachelot-Loza C: A new approach in the study of the molecular and cellular events implicated in heparin-induced thrombocytopenia: Formation of leukocyte-platelet aggregates. Thromb Haemost 85: 1090– 1096, 2001 [PubMed] [Google Scholar]
- 29.Mascelli MA, Deliargyris EN, Damaraju LV, Barnathan ES, Califf RM, Simoons ML, Sane DC: Antibodies to platelet factor 4/heparin are associated with elevated endothelial cell activation markers in patients with acute coronary ischemic syndromes. J Thromb Thrombolysis 18: 171– 175, 2004 [DOI] [PubMed] [Google Scholar]
- 30.Zimmermann J, Herrlinger S, Pruy A, Metzger T, Wanner C: Inflammation enhances cardiovascular risk and mortality in hemodialysis patients. Kidney Int 55: 648– 658, 1999 [DOI] [PubMed] [Google Scholar]
- 31.Krane V, Winkler K, Drechsler C, Lilienthal J, Marz W, Wanner C: Effect of atorvastatin on inflammation and outcome in patients with type 2 diabetes mellitus on hemodialysis. Kidney Int 74: 1461– 1467, 2008 [DOI] [PubMed] [Google Scholar]
- 32.Koenig W, Sund M, Frohlich M, Fischer HG, Lowel H, Doring A, Hutchinson WL, Pepys MB: C-Reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men: Results from the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. Circulation 99: 237– 242, 1999 [DOI] [PubMed] [Google Scholar]
- 33.Carrier M, Knoll GA, Kovacs MJ, Moore JC, Fergusson D, Rodger MA: The prevalence of antibodies to the platelet factor 4-heparin complex and association with access thrombosis in patients on chronic hemodialysis. Thromb Res 120: 215– 220, 2007 [DOI] [PubMed] [Google Scholar]
- 34.Narula AS, Jha V, Bali HK, Sakhuja V, Sapru RP: Cardiac arrhythmias and silent myocardial ischemia during hemodialysis. Ren Fail 22: 355– 368, 2000 [DOI] [PubMed] [Google Scholar]
- 35.Eichler P, Raschke R, Lubenow N, Meyer O, Schwind P, Greinacher A: The new ID-heparin/PF4 antibody test for rapid detection of heparin-induced antibodies in comparison with functional and antigenic assays. Br J Haematol 116: 887– 891, 2002 [DOI] [PubMed] [Google Scholar]