Blood samples were collected from 30 presumably healthy volunteers with normal APTT. The samples were collected by venipuncture into 4.5 ml test tubes containing 0.5 ml of 0.129 M buffered sodium citrate. After centrifugation at 4000 g for 20 minutes, the plasma was pooled and stored in small aliquots at -70°C.

Solutions were made of hexadimethrinebromide (Polybrene^®, Sigma Chemical Company, Saint Louis MO, USA), poly-_L-arginine (poly-_L-arginine hydrochloride, MW 15,000–70,000, Sigma), poly-_L-lysine (poly-_L-lysine hydrobromide, MW 15,000–30,000, Sigma), poly-_DL-lysine (poly-_DL-lysine hydrobromide, MW 1,000–4,000, Sigma), poly-_L-histidine (poly-_L-histidine hydrochloride, MW 15,000–50,000, Sigma), or protamine sulphate (Protamine^® Sulphate, LEO, Ballerup, Denmark) and 0.15 M NaCl to concentrations of 0.66 mg/ml.

For the testing of the different heparin neutralisers, each of them was added to 0.5 ml NP to final concentrations varying from 0.0 to 6.6 μg/ml. Clotting times were measured with APTT-reagents containing crude cephalin diluted 1/100, 1/200, 1/800 or 1/3200. Clotting times with each heparin neutraliser were also measured after the addition of UFH 0.5 U/ml (final concentration) to NP.

The clotting time of 1:1 mixture of patient plasma and NP is determined with low and high PL concentrations, respectively. The ratio between these two coagulation times is normalised by dividing with the corresponding ratio of the NP. This final ratio is the Lupus Ratio (LR) of that plasma (Fig. 1).

APTT reagents were made from dilutions of crude cephalin from bovine brain (a generous gift from Axis-Shield PoC AS, Oslo, Norway). A constant concentration (final concentration 6 mg/l) of ellagic acid was used as activator. The reagents for the APTT-based Lupus Ratio test contain crude cephalin diluted 1/100 and 1/3200 in Owren's buffer, respectively 4. In-house reagents for the RVVT-based Lupus Ratio test contain crude cephalin diluted 1/200 and 1/10000 in Owren's buffer. Russell's Viper Venom (Sigma-Aldrich Norway A/S, Oslo, Norway) was diluted in imidazole buffer with 1% BSA and aliquots were stored at -20C 49.

After earlier testing of stability of reagents and reproducibility of clotting times, the cephalin dilutions used in the RVVT assay are made one day ahead of their use and are stable for at least one week. The RVV was kept on ice and was stable for 4 hours after thawing. The APTT reagents were autoclaved and sealed and were then stable for months.

The reference limits have been calculated earlier from a reference population of 120 healthy blood donors 5. The 97.5 percentile was chosen as the upper reference limit. The upper reference limit for the APTT-based assay was 1.05; for the RVVT-based assay it was 1.11.

Based on serial dilutions of plasma from patients with strong LA activity, we have previously chosen arbitrary limits for semi-quantification of LAs into weak, moderate or strong. This semi-quantification has been shown to have good reproducibility 5.

Plasma from twenty patients was collected in the same way as described for NP. The patients included twelve patients with previously diagnosed LA and eight patients with previous negative tests for LA.

The APTT-based LR test was performed on patient plasma before and after the addition of 1/50 volume of polybrene or NaCl, with a final concentration of polybrene of 7.9 μg/ml. Patient plasma with polybrene was then tested before and after the addition of 1 U/ml UFH (final concentration) or NaCl.

Polybrene with a final concentration of 3.96 μg/ml, which should be able to neutralise up to 0.6 U/ml heparin, and/or UFH 0.5 U/ml (final concentration) was added to normal plasma. RVVT-based Lupus Ratios were then measured.

Polybrene (final concentration 3.96 μg/ml) and/or UFH was also added to patient plasma from both patients with known LA (n = 12) and plasma from patients without LA (n = 13).

Plasma from eleven LA positive patients had been collected earlier in a study on tissue factor pathway inhibitor in LA positive patients 10. Informed consent was obtained from all patients before inclusion. Blood samples were drawn before and exactly five minutes after the intravenous injection of 5000 IU UFH. Aliquots were stored at -70°C. The heparin concentration in postheparin plasma, measured as antiXa activity (Coatest^®, LMW heparin/heparin, Chromogenix AB, Mölndal, Sweden), ranged from 0.50 to 1.37 U/ml (individual results not shown). Polybrene (Sigma) was added to NP to a concentration of 15.84 μg/ml. Patient plasma was mixed 1:1 with polybrene-containing NP. The LRs of these mixtures were measured.

The APTT of plasma heparinised in vitro was measured with the Thrombotrack instrument from NycoMed Pharma. The remaining studies were performed with STA Compact coagulation analyser from Stago Diagnostica.

Poly-_L-histidine and poly-_DL-lysine were not able to neutralise heparin and were excluded from further testing.

The remaining four cations tested all showed a tendency to prolongation of the APTT, particularly with the APTT-reagent with low phospholipid concentration (table I). The effect of poly-_L-arginine seemed less predictable than that of the others as it could both prolong and shorten the APTT (table I). All four cations could neutralise completely the heparin effect in the APTT assays (figure 2).

Poly-L-arginine and polybrene were marginally more effective. As the stability of polybrene is better documented, it was chosen as heparin neutraliser.

Results obtained with a thrombin time assay indicated that 7.9 μg polybrene neutralised completely 1.2 units of UFH (results not shown). This concentration of polybrene was used in the following experiments. In the LR assay test plasma is mixed with NP prior to assay. For practical purposes, polybrene was added to the NP batch to a final concentration of 7.9 × 2 = 15.8 μg/ml, in order to obtain the same polybrene:heparin ratio as when polybrene was added directly to test plasma.

The median preheparin APTT-based LR in the eleven patients that received one injection with UFH was 1.40 (table 5). LR values with polybrene, both in heparinised and preheparin samples, were very similar to baseline LR values (median LR with polybrene 1.34; median LR in postheparin samples 1.32). In one patient the semi-quantification changed from moderate to low positive after the injection of heparin (table 5).