Introduction
According to World Health Organization the most common causes of death in 2019 were ischaemic heart disease and stroke. These are both often a result of thrombosis. In 2019 these two causes of death together were responsible for more than a quarter of deaths (WHO, 2020). Therefore, it is important that there is a quick and precise diagnostic marker for thrombosis. Currently, D-dimer is being used as a diagnostic marker. However, it has its many drawbacks (Kabrhel, et. al., 2010). Thus, the main purpose of this study was to identify whether prothrombin fragment 1+2 could be used as a potential diagnostic marker for thrombosis or acute bleeding. Two goals were set for this research paper. One of these goals was to find the correlation between D-dimer and prothrombin fragments 1+2 concentrations. The other one was to identify whether there was a difference between the levels of prothrombin fragment 1+2 when the blood samples were tested immediately after being drawn compared to testing the blood samples after they were frozen and thawed. This study was conducted in collaboration with St. Anne’s University Hospital in Brno, Czech Republic under the supervision of RNDr. Magda Popelová, PhD. It had financial support from the South Moravian Region of the Czech Republic.
The appearance of D-dimer and prothrombin fragment 1+2 in haemostasis
In general, there are 4 stages to haemostasis: constriction of the blood vessel, platelet clotting, coagulation cascade and fibrinolysis (Mourek, 2012, Kittnar, 2011, Penka M, Penka I, Gumulec, 2014). Prothrombin fragment 1+2 appears in the coagulation cascade when prothrombin is converted into thrombin (Haeberli, 1999). Meanwhile, D-dimer appears in fibrinolysis when fibrin is cleaved by plasmin into multiple small polymers called fibrin degradation products (FDPs). One of these FDPs is D-dimer (Bounds, 2023, Pecka, 2004). Since both molecules are produced during haemostasis, they should both be viable diagnostic markers for thrombosis (just as D-dimer is now). As both can be measured to see when their concentrations will go above the expected norms.
The drawbacks of using D-dimer as a diagnostic marker
D-dimer is known for giving false-positives when used as a diagnostic marker. Therefore, it can only be used to exclude the possibility that the patient is undergoing thrombosis. Some of the possible factors that can raise the D-dimer concentration are: sex, age, race, being after a certain surgery, some malignant tumours, inactivity (for example having a cast), pregnancy, certain diseases, smoking, recent trauma or certain infections (Kabrhel, et. al., 2010, Pulivarthi, Gurram, 2014).
The possible substitutes for D-dimer
There are a few substitutes for D-dimer that are currently being tested for their viability. This research paper along with some other studies focused on prothrombin fragment 1+2 (Sadílek, et. al., 2023). Soluble fibrin, fibrin monomer and thrombin-antithrombin are examples of some other molecules that are currently being researched as potential diagnostic markers (Ota, 2008, McFarland, 2018, Singh, 2015, LaPelusa, 2020). Some studies have also conducted research on the viability of these markers when they are measured from a urine sample (Borris, 2007, Wexels, 2017).
What is prothrombin fragment 1+2
Prothrombin fragment 1+2 is a molecule created during the coagulation cascade. It is a polypeptide fragment which compared to D-dimer is created earlier in haemostasis. This could possibly be used to detect thrombosis faster. It also should not produce false positives and thus should be more efficient at indicating thrombosis (Sadílek, et. al., 2023)
Methodology
Blood samples were collected from the hospital’s patients that were admitted with suspected thrombosis. These samples were assigned a number in the register to keep track of them. Afterwards, the samples were centrifuged to obtain plasma without blood cells. For clinical purposes they were immediately input into Atellica COAG 360 System (a haematological machine designed by Siemens Healthineers) to measure the D-dimer concentrations. However, the measurement of prothrombin fragment 1+2 concentrations required a special reagent that was expensive and had a short expiry date after it was thawed. Due to these limitations, it was decided to freeze the plasma samples at -70 ⁰C in cryo tubes to store them. Eventually, once enough samples were collected to fully use up one reagent, the samples were thawed in a water bath set to 37 ⁰C for about 5 minutes before setting them aside in the laboratory temperature for 10 minutes. They were then inputed into the machine. The machine is fully automatic and uses chemiluminescence immunoassay to measure the prothrombin fragment 1+2 concentrations. It generated a spreadsheet with the measured values for our assigned numbers in the register. The machine for this testing was calibrated and checked for its accuracy using controls all according to the instruction manual.
In total 200 samples were collected (121 women, 79 men – ranging from 21-92 years old) for the first goal of this study and 50 additional samples for the second goal that was set. No data on sex or age is available for the second goal.
Results
In this study two statistical methods were mainly used: Pearson correlation coefficient (r) and t-test. For the first goal the correlation between D-dimer and prothrombin fragment 1+2 concentrations was 0.536. It was then checked whether each correlation was significant. This particular correlation would be significant for n=200 and p=0.01 if our r value was bigger or equal to 0.181. Therefore, this correlation is significant.
Fig. 1 Graph of correlation between D-dimer and prothrombin fragment 1+2 concentrations (frozen)
It was then decided to check how many samples had a matching result when D-dimer and prothrombin fragment 1+2 were used as a diagnostic marker. A matching result was defined as both markers indicating a positive result or a negative one. Thus, a non-matching result was one where one marker turned out positive and the other one negative. The percentage of matching results was 68%. It is likely that the 32% of non-matching results was caused by the false positives in D-dimers.
Fig. 2 Proportion of matching and non-matching results when D-dimer and prothrombin fragment 1+2 are used as a diagnostic marker
For the second goal there were 50 samples so n=50. The correlation was once again calculated between D-dimer and prothrombin fragment 1+2 concentrations. The r value ended up being 0.353 and for it to be significant at p=0.01 our r would have to be greater or equal to 0.354. Therefore, it was concluded to be on the boundary of being significant at p=0.01 but would definitely be significant at p=0.05.
Fig. 3 Graph of correlation between D-dimer and prothrombin fragment 1+2 (fresh)
At last, a correlation was calculated between prothrombin fragment 1+2 concentrations measured immediately after obtaining the blood sample and concentrations measured after freezing and then thawing the same blood sample. The r value ended up being 0.999 which indicated a basically linear relationship. As a result, the function (y = 0.9416x+6.1518) that was obtained while making the graph was used to try and predict the concentrations of prothrombin fragment 1+2 after freezing and then thawing the same samples. This was done in an attempt to estimate how much prothrombin fragment 1+2 denatures while being handled and while frozen.
In the beginning a t-test was conducted between the concentrations of prothrombin fragment 1+2 measured immediately after the samples were obtained and then when they were thawed. This t-test proved that these concentrations were considered statistically different and thus it was necessary to consider the change in their values. Afterwards, a t-test was conducted between the actual concentrations that were obtained after the samples were thawed and measured and our predicted concentrations. This t-test did not present sufficient proof to reject our hypothesis that these two values are statistically identically. Therefore, it was concluded that this function can be safely used to predict the concentrations of samples that were frozen at -70 ⁰C for a maximum period of one week. With the condition that the prothrombin fragment 1+2 concentrations are between 57.02 to 916.06 pmol/l.
Conclusion
In regards, to the first goal of this study, a significant positive correlation was observed between D-dimer and prothrombin fragment 1+2 concentrations. The r value that was obtained in this research paper is supported by another study that had 115 patients and calculated an r value of 0.542 (Al-Samkari, et. al, 2020). An investigation was then conducted into how many samples had a matching result when using both markers. Although this was more than half it could be higher if the false positives of D-dimer were removed. When the correlation was recalculated for only the samples whose results matched, a more significant positive correlation was obtained. All of these are an indicator of the possibility that prothrombin fragment 1+2 concentration could be utilised as a diagnostic marker for thrombosis instead of D-dimer concentration.
For the second goal of this study, the correlation was calculated between the concentrations of D-dimer and prothrombin fragment 1+2 measured immediately after the sample was obtained and once again, a significant positive correlation was observed. Afterwards, the correlation between the concentrations of prothrombin fragment 1+2 measured immediately after the samples were obtained and after freezing and thawing the samples was calculated. This relationship turned out to be basically linear. Afterwards, by conducting t-tests the possibility of using correlation functions to predict how prothrombin fragment 1+2 denatures at different temperatures and periods of time was confirmed. This could be beneficial in making diagnosis of thrombosis more precise as it presents an option of making separate diagnostic boundaries for each scenario. This study’s diagnostic boundary of 307.1 pmol/l was input into the function that was obtained while creating the graph. A new possible diagnostic boundary of 295.32 pmol/l was calculated. The decrease in the prothrombin fragment 1+2 concentrations of the samples after thawing them was on average 2.14%.
Sources:
Al‐Samkari, H., Song, F., Van Cott, E.M., Kuter, D.J., Rosovsky, R. (2020) Evaluation of the Prothrombin Fragment 1.2 in Patients with Coronavirus Disease 2019 (COVID ‐19). American Journal of Hematology 95 (12), 1479 - 1485.
Borris, L.C., Breindahl, M., Ryge, C., Sommer, H.M., Lassen, M.R. (2007) Prothrombin Fragment 1 + 2 in Urine as an Indicator of Sustained Coagulation Activation after Total Hip Arthroplasty. Thrombosis Research 121 (3) 369 - 376.
D-Dimer; www.ncbi.nlm.nih.gov; date of access: 06. 02. 2024
Haeberli, A. (1999) Prothrombin fragment F1+2. Laboratory Techniques in Thrombosis — a Manual. 1 (2) 217 - 222.
Kabrhel, C., Mark Courtney, D., Camargo, C.A., Plewa, M.C., Nordenholz, K.E., Moore, C.L., Richman, P.B., Smithline, H.A., Beam, D.M., Kline, J.A. (2010) Factors Associated with Positive D-dimer Results in Patients Evaluated for Pulmonary Embolism. Academic Emergency Medicine 17 (6) 589 - 597.
Kittnar, O., et. al. (2011) Lékařská Fyziologie. Grada Publishing.
McFarland, C.P., Lind, S.E. (2018) Thrombin Generation Biomarkers Decline With Parenteral Anticoagulation—An Overlooked Means of Anticoagulation Monitoring?. Clinical and Applied Thrombosis/Hemostasis 24 (5) 708–717.
Mourek, J. (2012) Fyziologie - Učebnice pro Studenty Zdravotnických Oborů – 2nd edition. Grada Publishing.
Ota, S., et. al. (2008) Elevated Levels of Prothrombin Fragment 1 + 2 Indicate High Risk of Thrombosis. Clinical and Applied Thrombosis/Hemostasis 14 (3) 279 - 285.
Pecka, M. (2004) Laboratorní Hematologie v Přehledu:Fyziologie a Patofyziologie Hemostázy. 1st Edition, FINIDR.
Penka, M., Penka, I., Gumulec, J., et. al. (2014) Krvácení. Grada Publishing.
Physiology, Hemostasis; www.ncbi.nlm.nih.gov; date of acesss: 12. 01. 2024
Sadílek, P., Jančíková, L., Fátorová, I. (2022) Fragmenty Protrombinu 1+2 Včera, Dnes a Zítra. Česko - Slovenská Konference Laboratorní Hematologie 20 (1) 37
Siemens Healthineers (2021) Manuals for Innovance Loci
Singh, N., Pati, H.P., Tyagi, S., Upadhyay, A.D., Saxena, R. (2015) Evaluation of the Diagnostic Performance of Fibrin Monomer in Comparison to D-Dimer in Patients with Overt and Nonovert Disseminated Intravascular Coagulation. Clinical and Applied Thrombosis/Hemostasis 23 (5) 460 - 465.
Wexels, F., Dahl, O.E., Pripp, A.H., Seljeflot I. (2017) Thrombin Generation in Patients with Suspected Venous Thromboembolism. Clinical and Applied Thrombosis/Hemostasis 23 (5) 416 - 421.
WHO; www.who.int; date of access: 20. 07. 2024
Kommentare