IgM-Anticardiolipin Antibody and Vascular Access Thrombosis in Chronic Hemodialysis Patients

Abstract

Aim: Vascular access failure is a major cause of morbidity in chronic hemodialysis (HD) patients. Elevated immunoglobulin-M anticardiolipin antibody (IgM-aCL) titer is associated with stenosis of vascular access in HD patients. The clinical significance of elevated IgM-aCL titer relative to recurrent vascular access thrombosis (VAT) in patients with HD is less clear. However, little information has been available until now about the clinical influence of elevated IgM-aCL titer with recurrent VAT in HD patients from Western countries, and no report exists for Taiwan. This study attempted to determine whether elevated IgM-aCL titer was associated with recurrent VAT in HD patients. Methods: This study enrolled 483 patients undergoing HD. IgM-aCL titer and hepatitis C marker were measured for all subjects. Results: Elevated IgM-aCL titer was present in 17.4% (84/483) of patients. There was no association recurrent VAT between elevated and normal IgM-aCL titers (P = 0.90). Presence of hepatitis C had significant differences between elevated and normal IgM-aCL titers (P = 0.027). Conclusions: We found no significant differences in recurrent VAT between elevated and normal IgM-aCL titer in chronic HD patients. Our results suggest recurrent VAT of synthetic or native fistula may not be caused by elevated IgM-aCL titer in these patients. Presence of hepatitis C may be a cofactor.

• vascular access thrombosis
• IgM-anticardiolipin antibody
• hemodialysis

Introduction

Vascular access is necessary for HD. Elevated IgM-aCL titer is associated with stenosis of vascular access in HD patients and does not predict VAT.[1] The clinical significance of elevated IgM-aCL titer relative to recurrent VAT in HD patients is less clear. However, little information[1&2] has been available until now about the clinical influence of elevated IgM-aCL titer with VAT in HD from Western countries, and no report exists for Taiwan. This study attempted to determine whether elevated IgM-aCL titer was associated with recurrent VAT in chronic HD patients.

Patients and Methods

A total of 483 patients who underwent HD were included in this study. This study conducted a cross-sectional analysis of patients with ESRD receiving HD (more than 6 months duration) in a centralized dialysis unit. All 483 patients enrolled in April 2000 underwent midweek predialysis blood sample testing for IgM-aCL titer and hepatitis C marker. Moreover, the number ofoccurrences of recurrent VAT (two or more thromboses) in each patient since HD was retrospectively and prospectively reviewed. Approximately 27.7% of the patients had synthetic graft, and 72.3% had native fistula. Regarding the membrane materials of the hollow fiber, 28.8% of the patients sampled used cellulose diacetate, 26.1% used polysulfone, 23.0% used polyamide, 20.3% used cellulose triacetate, and 1.8% used polymethylmethacrylate. Approximately 50% (239/483) of patients reused their hollow fibers. These patients who reused their hollow fibers did so an average of about six times. Average blood flow was 230–280 mL/min. Moreover, average dialysis length was 4 h, and the adequacy of dialysis was assessed on a monthly basis using the Kt/V. The underlying causes of ESRD in this population were as follows: Around 58.2% of the patients had chronic glomerulonephritis or interstitial nephritis, 22.2% had diabetic nephropathy, and the remaining had other diseases. Exclusion criteria included refusal or inability to give informed consent and the requirement for short-term (that is, < 6 months) dialysis or the use of antiplatelet drugs or oral anticoagulant agents. All study subjects provided written informed consent.

The presence of risk factors for VAT was elicited using the results of the patient questionnaire and was confirmed by careful inpatient and outpatient chart review. The charts contained records of all outpatient results, including cardiovascular Doppler, angioplasty, surgery, previous hospital admissions, and VAT history. The use of interdialytic anticoagulant (such as heparin) was also recorded. Information on age, type of shunt (synthetic graft versus native fistula), type of hollow fiber used (reuse versus no reuse), current smoker, etiology of renal failure, and hepatitis C was recorded. A patient was considered a smoker if they were a current smoker, an ex-smoker if they had smoked previously, and a nonsmoker if they had never smoked. Moreover, diabetes mellitus was considered to be present if the patient was on oral hypoglycemics, insulin, or a diabetic diet, or if the patient's fasting plasma glucose exceeded 140 mg/dL. VAT was defined as a lack of blood flow caused by palpitation and auscultation. The presence of thrombus was confirmed by surgery or angiography.

Laboratory Testing

Blood for IgM-aCL titer assay was drawn from the arterial port before starting HD. The standardization of the Varelisa cardiolipin IgM antibodies assay was carried out against an established standard serum (University of Louisville, KY, USA, Division of Rheumatology, Department of Medicine). The results are calibrated in GPL-U/mL: 1 GPL-U corresponds to the binding activity of 1 µg/mL of cardiolipin IgM antibody preparation that was purified from a standard serum by affinity chromatography. Normal values were set at less than 6 GPL-U/mL.

We randomly selected 100 normal subjects at our hospital. The average values were around 3.0 GPL-U/mL. Abbott systems HCV version 3.0 is a microparticle enzyme immunoassay for the qualitative detection of antibody to hepatitis C virus (anti-HCV) in human serum or plasma.

Statistical Analyses

All continuous data were tabulated as mean ± SEM. Paired and independent t tests were used for within-group and between-group comparisons, respectively. The level of statistical significance was set at a p of less than 0.05. (Ninety-five percent CIs were used for means, and binomial exact 95% CIs were used for proportions.)

The secondary hypothesis of this study, namely, the presence of an association between recurrent VAT and gender, smoking status, use of erythropoietin, interdialytic use of anticoagulants, mean duration of HD, presence of hepatitis C, etiology, recurrent VAT, type of shunt, hollow-fiber material, and reuse, was tested using the chi-square test or Fisher's exact test. Multiple logistic regressions were conducted using the presence of VAT as the outcome variable. Predictor variables included various risk factors (such as age, IgM-aCL titer, current smoker, diabetes mellitus, type of shunt, type of hollow fiber used, and hepatitis C).

Data were entered and analyzed in the SAS system (Statistical Analysis System, version 8.2, SAS Institute, Durham, NC, USA). Analysis of variance was used for comparison between groups.

Results

Tables 1 and 2 list the baseline patient characteristics for those with elevated and normal IgM-aCL titers. The mean duration of HD was 67.6 ± 71.2 months (from 13–199 months). The causes of end-stage renal disease included chronic glomerulonephritis or interstitial nephritis for 58.2% of patients, diabetes nephropathy for 22.2%, obstructive uropathy for 3.2%, systemic lupus erythematous (SLE) for 2.6%, polycystic kidney disease for 1.6%, transitional cell carcinoma for 1.0%, drug poisoning for 1.0%, and unknown for 3.0%. RecurrentVAT had no significant difference between elevated and normal IgM-aCL titers (P = 0.90). The presence of hepatitis C showed significant difference between elevated and normal IgM-aCL titers (P = 0.027). There were no statistically significant differences between elevated and normal IgM-aCL titers, including age, sex, smoking status, use of erythropoietin, interdialytic use of anticoagulants, mean duration of dialysis, etiology, hollow-fiber membrane materials, and reuse. Patients (15.5%; 13/84) with elevated IgM-aCL titer exceeded 12 GPL U/mL. Only one patient exceeded 100 GPL U/mL and had diabetic nephropathy, but no more recurrent VAT was found.

Table 1 Baseline Patient Characteristics for Those with Elevated and Normal IgM-Anticardiolipin Antibody Titers.

Patient characteristics	Elevated IgM-aCL titers (95% CI)	Normal IgM-aCL titers (95% CI)	P value
Number of patient	84	399
Age mean in years	55.7 (52.4–57.2)	57.5 (56.5–59.0)	0.07
Sex male	46.4% (35.5–57.7)	38.6% (33.8–43.6)	0.18
Type of shunt
Native fistula	66.7% (55.5–76.6)	73.4% (68.8–77.7)	0.21
Synthetic graft	33.3% (23.4–44.5)	26.6% (22.3–31.2)
Smoking status
Never	81.0% (70.9–88.7)	84.2% (80.3–87.7)	0.46
Ex-smoker or smoker	19.0% (11.3–29.1)	15.8% (12.3–19.7)
Use of erythropoietin	95.2% (88.3–98.7)	95.2% (92.7–97.1)	1.00
Interdialytic use of anticoagulants	86.9% (77.8–93.3)	92.5% (89.4–94.9)	0.10
Etiology of end-stage renal disease
Diabetes mellitus	16.7% (0.9–26.4)	23.1% (19.3–27.8)	0.1
Glomerulonephritis	61.9% (50.7–72.2)	59.9% (54.9–64.7)
Hypertension	7.1% (2.6–14.9)	9.3% (6.6–12.6)
Lupus nephritis	6.0% (2.0–13.4)	2.0% (0.9–3.9)
Obstructive uropathy	7.1% (2.7–14.9)	2.5% (1.2–4.6)
Polycystic kidney disease	1.2% (0–6.5)	1.8% (0.7–3.6)
Other	0	1.3% (0.4–2.9)

Table 2 Baseline Patient Characteristics for Those with Elevated and Normal IgM-Anticardiolipin Antibody Titers.

Patient characteristics	Elevated IgM-aCL titers (95% CI)	Normal IgM-aCL titers (95% CI)	P value
Number of patients	84	399
Mean duration of hemodialysis (month)	64.1 (55.3–72.9 )	68.4 (64.3–72.5)	0.39
Recurrent vascular access thrombosis	38.1% (27.7–49.3)	38.9% (34.0–43.8)	0.90
Reuse of hollow fibers	44.1% (33.2–55.3)	50.6% (45.6–55.6)	0.27
Presence of hepatitis C	6.0% (2.0–13.4)	15.0% (11.7–18.9)	0.027
Membrane materials of hollow fiber
Cellulose diacetate	32.1% (22.4–43.2)	27.8% (23.5–32.5)	0.06
Cellulose triacetate	19.1% (11.3–29.4)	21.1% (17.2–25.4)
Polysulfone	16.8% (16.2–35.6)	26.1% (21.8–30.7)
Polyamide	19.1% (11.3–29.1)	24.3% (20.2–28.8)
Polymethylmethacrylate	4.8% (1.3–11.8)	0.8% (0.2–2.2)

Multiple logistic regression analysis, using VAT history as the outcome variable and controlling for previously postulated risk factors for VAT as predictor variables, did not appear to be independently associated with VAT (Table 3). The only variable associated with an increased risk of VAT was the type of shunt placed (odds ratio 18.31, 95% CI, 8.58–39.09, P < 0.0001 for patients with a synthetic graft compared with those with a native fistula). A trend toward an association between presence of hepatitis C and increased rates of VAT was seen. The odds ratio of patients with presence of hepatitis C increasing rates of VAT was 1.74 (95% CI 0.95–3.20, P = 0.07 for patients with positive as compared with those with a negative).

Table 3 Multiple Logistic Regression Using Previously Reported Risk Factors to Predict for One or More Episodes of Vascular Access Thrombosis.

Risk factors	Odd ratio	95% CI	P value
Age years	0.99	0.96–1.04	0.06
IgM-aCL (positive as compared with negative)	1.56	0.89–2.74	0.12
Current smoker	1.30	0.74–2.28	0.36
Diabetes mellitus	0.83	0.49–1.40	0.48
Type of shunt (synthetic graft as compared with native fistula)	18.31	8.58–39.09	< 0.001
Type of hollow-fiber use (reuse as compared with no reuse)	1.10	0.72–1.69	0.65
Hepatitis C (positive as compared with negative)	1.74	0.95–3.20	0.07

Discussion

VAT, a widespread problem with native fistula or synthetic graft, contributes markedly to morbidity in HD patients. Previous studies showed elevated aCL titers, varying from 15.5%–29%.[1-7] Previous studies reported an association of higher incidence of elevated aCL titers in patients dialyzed using cuprophan membrane.[8-10] The incidence of elevated aCL titer increases with HD, probably due to some kind of membrane bioincompatibility. In this study, 17.4% of patients displayed elevated IgM-aCL titer and mild higher prevalence than previously reported (10%).[1&2] No cellulose membrane (cuprophan) was used in our HD patients. Moreover, 51.4% used noncellulose membrane (synthetics), while 48.6% of the present sample used substituted cellulose membrane. Polysulfone, polyamide, and polymethylmethacrylate belong to the synthetics membrane group. Approximately 51.4% of hollow fibers belong to biocompatible membranes in the present sample of patients. I think that the effect of elevated IgM-aCL titer was less influenced due to the majority use of biocompatible membranes.

Little information about the relationship between elevated IgM-aCL titer and recurrent VAT in HD patients is clear.[1&2] This study found a tendency toward recurrent VAT in 38.9% (188/483) of the sample patients. Approximately 82.6% (399/483) of these patients had normal IgM-aCL titer. We found no association recurrent VAT between elevated and normal IgM-aCL titers in HD patients, resembling previous reports.[1] The present finding of IgM-aCL not associated with VAT in HD may underlie the pathogenicity of IgM-aCL and possible other multiple factors. Moreover, repeat vascular injury or exposure to artificial cellulose membranes may explain the elevated aCL levels, which resemble a previously described effect on platelet–endothelium interaction.[11] A previous study reported that 40% of the IgM-aCL titer was presented at levels at least twice the upper limit of normal.[1] In our study. 15.5% (13/84) of patients with elevated IgM-aCL titer exceeded 12 GPL U/mL. Only one patient exceeded 100 GPL U/mL and had diabetic nephropathy, but no more recurrent VAT was found.

Previous studies have disagreed about the significance of antiphospholipid antibodies as a risk factor for access thrombosis.[8-10], [12-18] Chew et al. reported that they did not find an increase in risk for thrombotic events or access repair in stable dialysis patients with fistula followed for 1 year.[9] In this study, only 2.6% SLE had 11 patients and were without features of antiphospholipid syndrome. Three patients had elevated IgM-aCL titer, and eight had normal IgM-aCL titer. Two patients had recurrent VAT, and nine had no recurrent VAT, followed more than 1 year, and similar to previous studies.[9] Moreover, only 11 patients had SLE, and the remainder displayed no autoimmune disease or human immunodeficiency virus antibodies. None of the patients with elevated IgM-aCL concentration were receiving chlorpromazine.

A previous study reported[1] that 81 HD patients with synthetic graft (40%) displayed normal IgM-aCL titer, compared to 10 HD patients with synthetic graft (40%) who displayed elevated IgM-aCL titer. The presence of IgM-aCL was not predicted VAT. In the present study, 135 patients with synthetic graft had 20.7% elevated IgM-aCL titer. The other sample of 348 patients with native fistula contained 16.1% elevated IgM-aCL titer. Notably, the presence of IgM-aCL had mild higher prevalence in patients with synthetic grafting, but there was no association type of shunt between elevated and normal IgM-aCL titers.

No previous study has considered the reuse of hollow-fiber and elevated IgM-aCL titers in HD patients. The only similar work was a prospective study on the intradialytic symptoms associated with dialyzer reuse.[19] In the present study, approximately 49.5% (239/483) of patients reused hollow fibers. The reuse of hollow fibers took place an average of about six times. Dialyzers were reprocessed with an automated machine using peroxyacetic acid and hydrogen peroxide as sterilants. This studyfound similar frequency to overall and specific symptoms for both new and reused hollow fibers. Moreover, there was no associated hollow-fiber reuse between elevated and normal IgM-aCL titers.

Previous studies considered that the presence of hepatitis C was associated with elevated IgM-aCL titers.[20-22] In the present study, approximately 13.5% (65/483) of patients showed the presence of hepatitis C. Moreover, the presence of hepatitis C caused significantly elevated and normal IgM-aCL titers in HD patients. Five of 65 patients (7.7%) had elevated IgM-aCL titer. Only one patient exceeded 12 GPL U/mL, and the other 64 patients showed low titer. I think that the presence of hepatitis C may be a cofactor and will need to be studied more to confirm this observation.

In summary, the prevalence of elevated IgM-aCL titer was 17.4% among HD patients. We found no significant differences in recurrent VAT between elevated and normal IgM-aCL titers in chronic HD patients. Therefore, we conclude that recurrent VAT of synthetic or native fistula may not be caused by elevated IgM-aCL titer in these patients. Presence of hepatitis C may be a cofactor, and more prospective studies are needed to confirm this observation.