August 2018: Diagnosing A Bleeding Disorder

by Donna Castellone, MS, MT (ASCP) SH • August 01, 2018



The laboratory is provided with minimal information regarding patients. The best we can be assured of when the test is ordered is, maybe, and I do mean maybe, the anticoagulant he or she is on, and the patient's sex (which I have noted to be wrong at times). This makes diagnosing a bleeding disorder in the laboratory a challenge. So, lets start with the clinician.

The first assessment and one of the most important tools in the diagnosis of a bleeding disorder is a complete family history. This would include at what age bleeding started, medications, information on family members who have bleeding symptoms, and location or type of bleeding. All of this information will help to determine if the disorder is acquired or congenital. Also, it can help point the direction of testing to primary hemostasis or secondary hemostasis. Both prescription medications and over the counter medications including vitamins and supplements should be investigated. There is a large laundry list of medications that can contribute to the incidence of bleeding.

There are also several standardized bleeding scoring systems that aids in distinguishing between the presence and/or absence of a disorder by assigning a numerical value to symptoms. The higher the number the more likely the patient has a bleeding disorder. It has also been noted that physical examination may only provide limited information in diagnosing a bleeding disorder since the presence of bruising or bleeding may be limited at the time of examination.

What happens next? If you are looking to evaluate primary hemostasis, there are limited methods to screen patients for platelet disorders. The PFA 100 is a global assessment of platelet function using whole blood. A prolongation of this screen should prompt a further investigation of platelet function. It is important to remember that the test is impacted by the number of platelets as well as the hemoglobin level of the patient. All of this should be considered in relation to results. The sensitivity of the test improves when both the ADP and the EPI cartridge are run. The Verify Now test is used in patients to evaluate the effects of aspirin, clopidogrel or abciximab. Specific cartridges are used to test the effectiveness of these drugs on platelet function. It doesn’t provide any evaluation of other types of disorders in primary hemostasis. The bleeding time test is an old screen that looked at the ability of the platelets to clot at the site of injury. This test has mostly been removed from practice, since it was riddled with technique issues and did not predict bleeding. The gold standard test is platelet aggregation by light transmittance. This uses platelet rich plasma and several agonists to induce platelet aggregation. Responses can correlate to platelet function disorders. This can be enhanced by performing lumi aggregation which looks at granular release providing further information on platelet function. This test is performed in specialized coagulation laboratories and is not well standardized as different laboratories use different concentrations of agonists which can impact how the platelets respond. Thromboelastography may also provide information on platelet function but can be complicated to perform and interpret.

The first line of laboratory testing in a patient suspected with a bleeding disorder of secondary hemostasis should be the prothrombin time (PT), and the activated partial thromboplastin time (aPTT). These are global screening assays that provide information on possible coagulation factors deficiencies. What does a normal screening test tell the clinician? Basically you are assuring that their patient has “normal” levels of all of the factors they are screening for. That means they should have at least 50% of all levels of factors…..or does it? The definition of a sensitive reagent is that the reagent will prolong on a screening test when the factor level is < 30%. That means that you can have a screening reagent that may be insensitive to certain factor deficiencies resulting in a normal PT or aPTT. It is important to know how your reagents perform. This should be evaluated with each new lot of coagulation reagents.

Let’s assume our reagents work properly. A prolongation of just the PT will point to a factor VII deficiency, while a prolongation of the aPTT can mean further investigation of factors VIII, IX and XI. If both the PT and the aPTT are prolonged common pathway factors II, V, X or fibrinogen may be the cause. The aPTT will also be prolonged when there is a factor XII deficiency and or in a lupus anticoagulant, however these patients do not bleed.

The most prevalent congenital bleeding disorders are FVIII (1:5000-10,000 males) followed by FIX and FXI (1:100,000) with FX (1:500,000), FV and FXII (1:1,000,000) and FII as the rarest deficiencies (adapted from: Hereditary bleeding disorders - general. PathologyOutlines.com website: http://www.pathologyoutlines.com/topic/coagulationhereditarybleedinggeneral.html.)

When the PT and/or aPTT is prolonged, further investigation will aid in determining the abnormality. To eliminate a prolonged aPTT due to heparin or direct thrombin inhibitors, a thrombin time (TT) should be performed. If the TT is prolonged, a reptilase test may be performed to distinguish between anticoagulation and a fibrinogen deficiency. If the reptilase time is normal, the sample is contaminated with an anticoagulant and no further testing should be performed. If the TT is normal, a mixing study would be the next line of testing and will help determine if the cause of prolongation is due to a factor deficiency or an inhibitor. This test will use equal parts of the patients’ plasma that will be mixed with pooled normal plasma (PNP). It is important for the PNP to be well defined to ensure that the PNP used contains normal levels of all factors. If the patient is deficient in a factor the PNP will replace the missing factor and the PT of aPTT test will normalize. This indicates a factor deficiency. If the mixing study is still prolonged after the mix, it means that something in the patient’s plasma is inhibiting the PNP and therefore unable to correct. Several different criteria are used to determine if a mixing study has corrected. Laboratories may compare results to the PNP and allow results to be within several seconds of the PNP to determine a correction, or they may consider a correction if the mixing study result falls back into the normal range. There are no published guidelines or standards for laboratories to follow.

Upon correction of a mixing study, specific factor assays are performed. There are many variables in performing factor assays: the calibrator used as well as the calibrator assigned value, the type of standard curve that is used (polynomial, linear), the number of points used in the construction of the standard curve (CAP1 states a minimum of 3 points must be used), how far in seconds the points on the curve are (rule of thumb is about 7 seconds between the points- or your curve will be too flat and you will see large differences in values with small numerical changes), the deficient plasma (lyophilized, fresh frozen- congenital or immunodepleted), the buffer and the concentration of the calcium chloride. This in addition to the reagents that are used which may have different activators (ellagic acid, kaolin, micronized silica or silica) as well as the type of phospholipids. All of this plays a role in your results and we haven’t even run a test as of yet! Then we need to consider the dilutions that are run, (CAP1 states a minimum of 3 dilutions) Most importantly, all values must be on the standard curve with no extrapolation. What about those low values? Can your analyzer curve accurately identify them, or do you need a separate curve? There is a big difference between a factor assay of 1% and < 1%. In many cases the analyzer has trouble at that very low range due to inherent “noise” at low levels. A good practice when running low curves is to also run a buffer blank- if the buffer runs at 160 seconds and records a value of 1%, and the patient is at 165 seconds also at 1%, it can be deduced that there is no activity in the buffer blank, therefore the patient sample of 165 seconds is in reality <1%. Another check for low levels of factor assays is to perform either a PT or an aPTT which should correlate with your levels of the factors tested. Remember they are inversely proportional, the lower the level of factor the more prolonged the PT and/or aPTT.

The final piece to the puzzle is pre-analytical variables which are extremely important in coagulation testing. It is important to remember that both Factor VIII and V are heat labile, and will lose stability quickly. If you work with samples that have been aliquoted, you may have samples that give falsely decreased levels of factors due to improper handling of samples.

Based on all of these issues, all abnormal testing should be repeated to ensure results truly reflect the patient factor levels and rule out any errors due to pre-analytical variables or errors in testing. Diagnosing a bleeding disorder is a complicated process and many issues need to be investigated to ensure that the coagulation results aid the clinician in identifying the cause of the bleeding disorder.


1 CAP Hematology/Coagulation Checklists from College of American Pathology. 2017.