Thrombophilia testing and diagnostic dilemma – a tertiary centre experience

To the Editor,

Deficiencies of antithrombin, proteins C and S have been linked to venous thrombosis [1–3]. Increased fibrinogen, prothrombin, factors VIII, IX, XI, impaired fibrinolysis [4], factor V Leiden and mutations in prothrombin gene are some of the other risk factors for thrombosis.

Results of thrombophilia testing may be confounded by administration of oral vitamin K antagonist anticoagulants [5]. This can be detected by elevated prothrombin time (PT) and low levels of natural anticoagulants should be confirmed by one or more separate tests [5]. Currently, there is no single test to identify all thrombophilias and some are affected by intake of drugs such as warfarin [6]. Furthermore, in acute thrombosis levels of natural anticoagulants transiently decrease and testing may be delayed for 6 weeks to allow the acute phase reactants to return to baseline [6]. In contrast, in another study by Kovacs MJ, abnormal levels of proteins C and S detected within 24 h of diagnosis of venous thromboembolism continued to remain abnormal on retesting after 3 months [7]. They concluded that overall, 97.8% of initial protein C and S testing were accurate. The prevalence of antithrombin deficiency in the general population is rare (0.02%) as compared to protein C (0.2%) and protein S (0.03%–0.13%) deficiency [8].

Blood samples from patients undergoing thrombophilia workup were collected in 3.2% trisodium citrate (9:1 ratio) and centrifuged within 60 min of collection at 2500 g for 15 min. PT within 11–13 s and APTT within 23–33 s were considered normal. Antithrombin, protein C and S were analyzed using STA compact analyzer (Diagnostica Stago) whereas activated protein C resistance was evaluated using Sysmex analyzer (Siemens). Plasma samples for antithrombin, protein C, protein S assay and detection of activated protein C resistance were pooled together and preserved at –20°C and tests carried out within 21 days of sample collection. Lupus anticoagulant was detected using dilute Russell’s viper venom test (Staclot Diagnostica Stago) and was considered positive when the ratio of screen and confirm was more than 1.2. Antithrombin was measured by a chromogenic assay (STA-STACHROM ATIII, Diagnostica Stago). Values of antithrombin between 80% and 120% are considered normal. Protein C was measured by a clotting-based assay (STA-STACLOT PROTEIN C, Diagnostica Stago) and Protein S was evaluated by a clotting based assay (STA-STACLOT PROTEIN S, Diagnostica Stago). Protein C of 70%–130% and Protein S of 65%–140% was considered normal. Activated protein C resistance is an APTT based screening assay (Pro C Global, Siemens). In our laboratory activated protein C resistance was considered negative when this ratio (ratio of APTT with and without addition of activated protein C) was more than 0.7 and positive when it is <0.7. In case the ratio is 0.7, the test is performed again on a repeat sample.

In the study period, 75 patients underwent workup for thrombophilia. There were 49 males and 26 female patients with age varying from 17 to 68 years.

Levels of natural anticoagulants: 30 (40%) samples had low anticoagulant whereas, 45 (60%) had normal or increased levels of all three anticoagulants. Fifteen samples had low levels of one anticoagulant only. Antithrombin alone was low in 7 (9.3%), protein C in 5 (6.6%) and protein S in 3 (4%). Combined protein C and antithrombin deficiency was seen in 1 (1.3%), antithrombin and protein S in 2 (2.6%) and all three in 7 (9.3%). Low protein C and S were seen in 5 (6.6%). Activated protein C resistance was positive in 3 cases, one of whom had low levels of all 3 natural anticoagulants, second had low protein C and S and the third had normal levels of all three anticoagulants.

Increased antithrombin (122%–127%) was seen in 3 (4%), increased protein C (133%–226%) in 13 (17.3%) and increased protein S (145%–175%) in 4 (5.3%).

PT and APTT: 17 (22.6%) patients with low level of natural anticoagulants had elevated PT (14–36 s) and 12 (16%) had elevated (Figure 1) APTT (34–50 s). In comparison, 5 (33.3%) patients with no deficiency had elevated PT (14–18 s) and elevated APTT (34–53 s) in 5 (33.3%).

Figure 1

Frequency histogram showing distribution of APTT in patients with anticoagulant deficiency.

x-axis: APTT time; y-axis, number of patients.

Lupus anticoagulant: Lupus anticoagulant was detected in 5 (33.3%) of which 2 had low antithrombin, proteins C and S, 1 had low proteins C and S and 1 had low protein C alone.

Fisher’s test: Reduced levels of natural anticoagulants (antithrombin and protein C) was compared with increased PT and APTT and results are summarized in Table 1. The p-value is more than 0.05. From our study it appears that 60% of patients have normal levels of natural anticoagulants. Combined antithrombin, protein C and S deficiency and deficiency of antithrombin (9.3%) alone were the commonest deficiencies seen in our study. Shortening of APTT has been noticed in thrombophilia. Paradoxically, we had elevated APTT in 29.3% of our cases and elevated PT in 22.6%. Lupus anticoagulant was detected in only 4% of the patients with elevated APTT. Hence elevated PT and APTT could indicate the confounding affects of vitamin K antagonists, heparin or inappropriate timing of assay following cessation of therapeutic anticoagulant, reduced coagulation factors, disseminated intravascular coagulation or liver disease. Diagnosing deficiency of anticoagulants in these patients with elevated PT and APTT is inappropriate unless confirmed by retesting after stopping therapeutic anticoagulants. Fisher test further indicates the difficulty in interpretation of these assays as thrombophilia due to antithrombin or protein C deficiency is unlikely to be related to elevated PT and APTT.

We conclude that inspite of sufficient recent literature [9, 10] on the accurate timing for thrombophilia testing, inappropriate laboratory testing for thrombophilia continues until the present day.