INTRODUCTION
Coagulation tests are usually requested by clinicians for a variety of reasons. It could be part of an investigation for a pa- tient with bleeding tendencies or a rou- tine test prior to a surgical procedure. With the increasing trend of routine coa- gulation testing, the haemostasis laboratory is often burdened with the task of investigating an isolated prolonged APTT. In most cases, these turned out to be inconclusive or due to factor XII defi- ciency which is of no clinical significance. In this case report we illustrate a patient with chronic myelomonocytic leukaemia who presented with multiple subcutane- ous abscesses. The initial investigation showed a prolonged APTT, a low factor IX level and the presence of an inhibitor. She was diagnosed as an acquired factor IX inhibitor. On further testing this turned out to be a lupus anticoagulant. We dis- cuss the investigation of a prolonged APTT and how to differentiate a specific factor inhibitor from a lupus anticoagu- lant. This is the first case report that de- monstrates correction of an apparent low factor level with phopholipid mixing studies in the presence of a lupus anticoa- gulant.
CASE REPORT
A 51 year-old lady was diagnosed with chronic myelomonocytic leukaemia in August 2004 when she presented with epigastric pain and was noted to have an enlarged liver and spleen with monocyto- sis. She remained well following treatment with low dose cytarabine until three years later when she presented with multiple large abscesses over the buttocks, thigh and abdomen. She was seen by the orthopaedic surgeon and was planned for drainage and saucerisation of the ab- scesses. She had no bleeding symp- toms. Laboratory results included a hae- moglobin of 10.4g/dL, a white blood cell count of 5.6x109/L, and a platelet count of 41x109/L. Prothrombin time was 12.2 seconds (normal, 10.2-13.2 sec); APTT, 72.2 seconds (normal, 21-32 sec) and fibrinogen, 362 mg/dL (normal 150-450 mg/dL). Factor VIII:C activity was 72.9%; factor IX:C activity was 4.6%, with factor IX inhibitor of 4.0 Bethesda units (BU). In view of the results, she was diagnosed as having an acquired factor IX inhibitor. She was treated with recombinant acti- vated factor VII at 90mcg/kg and under- went the surgery without any bleeding or thrombotic complications. She was also given a course of antibiotics. However, three weeks later she returned with re- currence of the abscesses in the left axilla and back.
The patient was now referred to our hematology department for further man- agement. She had no bleeding symptoms. A repeat coagulation test was done. The results showed PT was 14.7 seconds (normal, 11.0-14.5 sec), APTT was 120.0 seconds (normal, 28-40 seconds) and the mixing studies with normal plasma showed a persistent prolongation of the APTT at 96.5 seconds suggesting the presence of an inhibitor. Factor VIII:C was 22% (normal, 60%- 150%), factor IX:C was <1.0% (normal, 60%-150%), factor XI was <1.0% and factor XII <1.0%. Serial dilutions for fac-tor levels were assayed to look for speci- ficity of the inhibitor (Table 1). A chromogenic factor VIII assay was also done and this showed a factor VIII level of 120%. To demonstrate that the low factor levels were due to the presence of a lupus anticoagulant (LA), phospholipid was added to the factor assays. The factor levels rose confirming that the apparent low factor levels were due to the interference by the LA (Table 2). The platelet count was 80x109/L with the presence of giant platelets and platelet clumps. Further testing for LA was carried out. The dilute Russell’s viper venom test (DRVVT) was prolonged at 88.2 seconds (normal range 26-36 seconds) and corrected to 42.3 seconds with the addition of phospholipid, giving a normalized DRVVT ratio of 2.6 (normal range up to 1.2) (Table 3). The hexagonal II phase phosphatidyle- thanolamine assay (Staclot-LA), showed a positive test for LA at 54 seconds (neg- ative ≤ 8 sec) (Table 4). The diagnosis of acquired factor IX inhibitor was therefore revised and the prolonged APTT was attributed to the presence of a LA. The patient underwent a second surgical drainage and saucerisation of the ab- scesses, this time without any factor cover. There was no bleeding complica- tion. Her blood cultures were negative but the indirect fluorescent antibody titer for meloidosis was positive (1:160). She responded to treatment with intravenous ceftazidime and oral cotrimoxazole. She has had no recurrence of the abscesses and is on long-term treatment with cotrimoxazole. A repeat test 6 months later showed persistence of the LA.
MATERIALS AND METHODS
The PT, APTT and factor assays were determined by using an automated coa- gulation analyzer (ACL 9000, Instrument Laboratory, USA). The PT reagents in- clude fibrinogen HS and calcium throm- boplastin (Rabbit-brain tissue), the APTT reagents consist of phospholipids (syn- thetic derived for optimal platelet lite ac- tivity) and calcium chloride (0.020M). All these reagents were supplied by the company from the Instrument Laboratory Sdn. Bhd. Mixing studies were performed using pooled plasma from 50 healthy normal individuals recruited from our hospital. The factor deficient plasma was commercially obtained from the Instru- ment Laboratory (USA). The presence of LA was detected by the dilute Russell’s viper venom test (DRVVT) and the Staclot LA. The phospholipid from the Staclot LA was used in the phospholipid mixing studies.
DISCUSSION
The activated partial thromboplastin time (APTT) is a non-specific screening test of the intrinsic system while the prothrombin time (PT) is a screening test of the ex- trinsic system. A prolonged APTT with a normal PT indicates a possible deficiency of factors VIII, IX, XI or XII or the pres- ence of inhibitors. These inhibitors are either specific to these factors or are non- specific, the most common of which is the lupus anticoagulant (LA) (Chng et al. 2005). No single test is definitive for a LA. As a result, a variety of tests and also the clinical history are used in an attempt to establish the diagnosis (Lupus anti- coagulant working party 1991). In the present case, a misdiagnosis of factor IX inhibitor was made based on the initial test showing a low factor IX with normal factor VIII levels. The clinical history is very important to help us decide what further tests need to be done. A specific factor inhibitor usually has bleeding ma- nifestations. In this case the patient had no bleeding symptoms, so one should suspect the presence of a LA. Lupus anticoagulant can affect the apparent level of coagulation factors by interfering with the phospholipid-dependent one-stage assay based on APTT. In our experience factor IX and XI assays are affected mar- kedly than factor VIII. To overcome this, dilution assays on the patient’s plasma should have been carried out to dilute out the effect of the lupus inhibitor (Kasper 1991).
In our laboratory, we found multiple factor deficiencies which made it easier as we then knew we were most likely dealing with a nonspecific inhibitor such as a LA. The dilution assays however did not show a rise in the apparent factor VIII, IX, XI and XII levels as would be ex- pected in the presence of a LA with factor levels remaining low in the presence of a specific inhibitor (Kasper 1991). This could be explained by the interference of a very strong lupus inhibitor. The factor VIII level did rise a little but fell to < 1.0% at dilutions above 1/50, most likely due to dilution of the factor itself. To further prove that the apparent low factor levels were due to the presence of the LA, we added phospholipid to the patient’s plasma and repeated the factor assays. Addition of phospholipid neutralizes the effect of LA and will correct the clotting times in the APTT-based factor assays (Tripodi 2007). This was clearly shown by the normalization of the apparent low factor levels with addition of phospholi- pid. It should be noted that addition of phospholipid to normal plasma produces a small dilution effect reducing the factor level slightly. We also did the chromo- genic assay for factor VIII and this was normal. Chromogenic factor assays are based on chromogenic substrates spe- cific for the factor assayed and are not phospholipid-dependent, hence will not be affected by the LA (Kazmi et al. 1998). The DRVVT and Staclot tests confirmed the presence of a LA in our patient. The DRVVT is very specific for LA and is not prolonged in the presence of a specific inhibitor to factor VIII or IX as the DRVVT directly activates factor X (Tropodi et al. 2005).
This patient was treated initially with re- combinant activated factor VII when she was misdiagnosed with factor IX inhibitor. This was unnecessary and inappropriate. Not only was it expensive, it could have been dangerous to the patient as she could have had a thrombosis. Patients with LA have an increased risk for throm- bosis and infusion of activated factor concentrates will increase this risk fur- ther. This case report illustrates the im- portance of correct interpretation of factor assays in the presence of a LA to avoid misdiagnosis and inappropriate treat- ment. In summary, a specific factor inhi- bitor can be differentiated from a lupus anticoagulant by: (1) the clinical history: patients with a specific factor inhibitor usually bleed, while patients with LA are usually asymptomatic or they may present with a thrombosis; they rarely bleed; (2) dilution assays: factor levels with a specific inhibitor should not rise with serial dilutions of patient’s plasma; (3) DRVVT: a specific factor inhibitor to factor VIII or IX will not prolong the DRVVT; (4) mixing with phospholipids: a specific factor inhibitor should not be cor- rected with phospholipids and (5) chro- mogenic factor assay: factors assayed by the chromogenic method will remain low in the presence of a specific factor inhi- bitor (Douglas 1994).
CONCLUSION
A lupus anticoagulant can affect the ap- parent levels of the APTT-based factor assays leading to a misdiagnosis of a specific factor inhibitor. It is important to distinguish LA interference from factor deficiency or the presence of a specific factor inhibitor to avoid misdiagnosis be- cause the implications and the treatment differ between these groups of patients.
ACKNOWLEDGEMENTS
We would like to thank the Director Gen- eral of Health Malaysia for permission to publish this paper and to the director of Ampang Hospital, Datin Dr. Aishah Maki- nuddin and all staff of the Hematology Department for supporting this work. Our appreciation is extended to Dr Mike Laffan, Professor of Haemostasis and Thrombosis, Imperial College London for his invaluable input and advice on writing this manuscript.