numero rivista e pagine: HSR Proceedings in Intensive Care and Cardiovascular Anesthesia 2010; 2(2): 119-127
PDF version

Perioperative evaluation of primary hemostasis in patients undergoing mitral valve repair

Authors: F. Pappalardo*1, P. Della Valle2, G. Maj1, A. Franco1, A. Lattuada3, G. Landoni1, A. Zangrillo1, A. D'Angelo2

1Department of Anesthesia and Intensive Care, UniversitÓ Vita-Salute San Raffaele, Milan, Italy
2Coagulation Service and Thrombosis Research Unit, UniversitÓ Vita-Salute San Raffaele, Milan, Italy
3Coagulation Service and Thrombosis Research Unit, Ospedale Sacco, Milan, Italy

Corresponding author: * Corresponding author:
Federico Pappalardo, MD
Department of Cardiovascular Anesthesia and Intensive Care
UniversitÓ Vita-Salute San Raffaele
Via Olgettina, 60 - 20132 Milan, Italy
E-mail: pappalardo.federico@hsr.it



No data exist on the prevalence of primary hemostatic defects and acquired von Willebrand disease in mitral valve prolapse with severe regurgitation.


Primary hemostasis was evaluated by PFA-100, von Willebrand Factor Antigen (vWF:Ag) and Ristocetin cofactor (vWF:RiCof) assays in a prospective observational trial. Sixty-five consecutive patients with mitral regurgitation (study group) or aortic stenosis (control group) who were operated for mitral valve repair or aortic valve replacement were enrolled in the study.


There were no differences in Closure Time in the two groups at all time points. The concentration of plasma vWF: Ag was within normal limits in all patients preoperatively; after surgery, a significant increase was observed in both groups from baseline (199 +/- 144 mcg/dL vs. 295 +/-141 mcg/dL in the study group, p=0.002; 243 +/- 141 mcg/dLl vs 338 +/- 154 mcg/dL in the control group, p=0.009). The ratio of vWF:RiCof to vWF:Ag was slightly decreased preoperatively in both groups (ratio= 0.91) and showed a marked increase in the postoperative period (ratio=0.22) as, probably, new hemostatically effective large multimeric forms of vWF were released.


Patients who present for surgery with a valvular pathology with high shear stress have some degree of primary hemostasis defect; nevertheless, the potent stimulus of surgery and the correction of the underlying disease allow quick restoration of vWF activity and normalization of PFA-100.


Keywords: von Willebrand, mitral surgery, aortic surgery, platelect function, cardiac surgery


Acquired von Willebrand syndrome in high intracardiac shear stress conditions, such as aortic stenosis [1,2,3,4,5,6,7,8], ventricular septal defect [9] and patent ductus arteriosus [10], has been clearly defined in its epidemiology, pathophysiology and clinical correlations. Similarly, patients with other cardiac defects might occasionally exhibit such abnormality.
Platelet and blood clotting activation has been shown in patients with mitral valve prolapse but no data exist on the prevalence of primary hemostatic defects and acquired von Willebrand disease in mitral valve prolapse with severe regurgitation [11].
The fluid dynamics of a regurgitant valve are, in many ways, similar to the fluid dynamics of a stenotic valve; however, the anatomy of the regurgitant orifice and the dynamics of the regurgitant flow are complex and not fully understood. The shape and the direction of the regurgitant jet depends on multiple factors, including the anatomy and orientation of the regurgitant orifice, the driving force across the valve, and the size and compli-ance of the receiving chamber.
Given these facts, we hypothesized that primary hemostasis defects could be a feature in patients with mitral valve prolapse.
Aim of this study was to evaluate the prevalence of primary hemostasis disorders in patients with severe mitral regurgitation due to prolapse without an history of pathological mucocutaneous bleeding, by means of PFA-100 and to define their pathophysiology and clinical implications.



65 consecutive patients undergoing open heart surgery for valve operations were prospectively evaluated; 41 patients underwent mitral valve repair for severe mitral regurgitation (study group), 24 of them underwent aortic valve replacement for severe aortic stenosis (control group). 5 patients were excluded from the analysis for postoperative re-exploration with evident source of surgical bleeding.
Patients were not included if operated on urgency-emergency priority or with a minimally invasive approach or had chronic atrial fibrillation. No patient received warfarin or antiplatelet drugs within 7 days before surgery; patients on intravenous heparin had it stopped at least 6 hours preoperatively. Patients were not considered eligible if the preoperative platelet count was <80,000 x 109/l, if they had preoperative PT and/or APTT ratios >1.2 or if they had a history of hematological or hepatic disease (e.g., active hepatitis or cirrhosis), or chronic renal failure requiring dialysis. All patients signed an informed consent, according to the Declaration of Helsinki, after approval of the protocol by the Ethics Committee of our Institution.

The PFA-100 provides an ex vivo quantitative measure of platelet function under high shear conditions simulating primary haemostatic events following vascular injury.
The platelet function analyzer (PFA-100, Dade Behring, Inc, Deerfield, Ill.) performs analyses on citrated whole blood samples (Vacutainer tubes, Becton Dickinson; citrate 129 mmol/L; 3.8%) and measures the time needed for a platelet plug to form after activation of platelets by collagen and adenosin diphosphate (CADP, 50 μg) or collagen and epinephrine (CEPI, 10 μg). The sample is aspirated under constant negative pressure through a 200 μm capillary into a small reservoir, by passing through an 150 μm opening coated with collagen/epinephrine (CEPI) or collagen/ADP (CADP). As blood contacts the collagen-coated membrane, the platelet agonist (epinephrine or ADP) is solubilized, resulting in platelet stimulation. The test terminates when the developed primary hemostatic plug obstructs the opening. The instrument provides the Closure Time (CT) parameter, indicating the time duration of the test: results are expressed in seconds, with a maximum test time of 300 seconds. Any value greater than that is reported as non closure. Our assays were performed with CADP cartridges in order to remove the effect of antiplatelets acting on ADP receptors. All PFA-100 measurements were performed within 30 minutes of sample collection.

Patients’ management
Prior to institution of cardiopulmonary bypass (CPB), patients received intravenous porcine heparin (300 IU/kg of body weight) and, during CPB, additional doses (5000 IU), if required, to maintain the activated clotting time (ACT) >480 sec. (ACT II, Medtronic, Minneapolis, MN, USA). ACT measurements were carried out 2 minutes after heparin administration and every 20 minutes thereafter. After termination of CPB and surgical hemostasis, heparin was neutralized with protamine sulphate (1:1 ratio). All patients received an intraoperative infusion of tranexamic acid (1g in 20 minutes before skin incision, followed by a continuous infusion of 400 mg/h until completion of surgery) according to our institutional protocol.
At the end of the surgical procedure, patients were transferred to the Intensive care unit (ICU). Postoperative blood loss was collected in a graduated reservoir connected to a closed evacuation system (Argyle, Aqua-seal, Sherwood Medical, Tullamore, Ireland).

Cardiopulmonary bypass circuit
Extracorporeal circulation was instituted in all patients by draining blood by gravity into an open venous reservoir (Avant Reservoir, Dideco, Mirandola, Italy), and by driving it by means of a roller peristaltic pump (Caps, Stöckert Instruments, Munich, Germany) through a heat exchanger integrated with a hollow fiber membrane oxygenator (Avant D903, or Avant D903 Ph.i.s.i.o., Dideco, Mirandola, Italy) and through an arterial filter (D734 MICRO 40, Dideco, Mirandola, Italy) back into the patient at a mean flowrate of 2.4 l/min/m2. The same custom pack of PVC/silicone tubing (Dideco, Mirandola, Italy) was used for all patients. The circuit was primed with 1500-ml Ringer lactate solution, 100 mL Mannitol 18%, and 5000 IU porcine heparin. Intermittent cold blood cardioplegia (4°C) was infused by means of a heat exchanger (D720 Helios C, Dideco, Mirandola, Italy) and two roller pumps, according to Buckberg’s protocol. Mild-to-moderate hypothermia was employed (mean internal temperature: 31.4°C).
Shed mediastinal blood suction and left heart venting were actively performed with two separated roller pumps. Additionally, blood was aspirated from the operative field with a vacuum suction device (D745, Dideco, Mirandola, Italy), processed in a cell saver (Compact-A, Dideco, Mirandola, Italy), and then reinfused after closure of the chest.

Laboratory assays
Blood samples were collected from a radial arterial catheter by means of a two-syringe technique. The first syringe was used to withdraw 10 ml of blood and then the sample was obtained in the second syringe. Sample time points were as follows:
A) preoperatively (Baseline);
B) before heparin administration (Pre-hep);
C) after removal of the aortic crossclamp (Unclamp);
D) 4 minutes after the administration of protamine (Post-prota);
E) at ICU arrival (ICU-arrival);
F)  4 hours after ICU arrival (ICU-4h);
G) the morning following surgery (ICU-18h).
Blood samples were collected in siliconized Vacutainer tubes (Becton Dickinson, Plymouth, UK) containing tri-sodium citrate (0.19 M); within 1 hour from blood collection, platelet poor plasma was obtained by centrifugation for 20 minutes at 2000 RPM at room temperature.
Plasma aliquots of platelet-poor plasma (0.4 mL) were snap-frozen with methanol and dry ice and finally stored at –80 °C until assayed.


Assay methods
Von Willebrand Factor Antigen (vWF:Ag) was determinated by commercially available ELISA assays (Asserachrom vWF, Diagnostica Stago, Asnier sur Seine, France). Ristocetin cofactor activity (vWF:RiCof) was measured by commercially available lyophilized platelet reagents (Dade-Behring, Marburg, Germany) using an automated coagulometer (ACL 9000; Instrumentation Laboratory, Lexington, Massachusset, USA) [12].
All results are the mean of the two determinations and expressed as IU/dL of plasma. The authors had full access to the data and take responsibility for its integrity. All authors have read and agree with the manuscript as written.

Statistical analyses
Descriptive data are expressed as mean standard deviation or median with interquartile range (IQR) as appropri-ate according to the skewedness of the observed distribution. Statistical significance of the difference between continuous variables was assessed by using a Student t-test or a Mann-Whitney U-test as appropriate. Compari-son of proportions was performed by Chi-Square or Fisher Exact tests as appropriate. All reported p values are two tailed. Differences were considered statistically significant in case p-values were lower than 0.05.
Statistical analyses were performed using the SPSS Statistical Package version 11.5.1 (SPSS Inc, Chicago, USA).



Patients’ demographics were similar in the two groups (mitral valve repair/study group vs aortic valve replace-ment/control group). Platelet count and hematocrit did not differ between the two groups at all time points (Figure 1 and 2).



Figure 1

Time course of platelet count (t-test) mean ┬▒ SD.



Figure 2

Time course of hemostasis: (%) (t-test) mean ┬▒ SD.


There were no differences in Closure Time in the two groups at all time points (Figure 3): the median closure time at baseline was above the normality range (less than 118 sec.) in both groups (146 sec. [106-299, 25th-75th percentile] in the study group vs 183 sec. [128-299, 25th-75th percentile] in the control group, p=0.22); only three patients in the control group exhibited a “normal” CT preoperatively (12%) vs 14 patients (38%) in the study group.



Figure 3

Closure Time (Mann-Whitney) median (25th-75th percentile).


Five patients in the control group had non closure (CT >300 sec.) preoperatively.
Preoperative CT increased following heparin administration and again with initiation of CPB. Fifteen minutes after protamine sulphate administration, CT returned to near baseline values. Median (CT) of all patients significantly decreased from baseline at ICU arrival and 18 hours postoperatively (163sec. [109.5-299, 25th-75th percentile] vs 83sec. [66-112, 25th-75th percentile], p<0.001) (Table 1); similarly the haematocrit and platelet count decreased. These variables were inversely related to CT at all time points (PLT p<0.001, r=-0.424; Hct p<0.003, r=-0.353 at ICU arrival (Spearman test).



Table 1

Time course of Platelet Function Analyzer Closure Time [median (25th - 75th percentile)] in the overall population.


The concentration of plasma vWF:Ag was within normal limits in all patients preoperatively. (VWF:Ag) levels did not differ at all time points in the two groups; after surgery, a significant increase was observed in both groups from baseline (199+/-144 mcg/dL vs 295+/-141 mcg/dL in the study group, p=0.002; 243+/-141 mcg/dL vs 338+/-154 mcg/dL in the control group, p=0.009) (Figures 4, 5 and 6).



Figure 4

Von Willebrand Factor Antigen concentration: time course in the overall study population.



Figure 5

Von Willebrand Factor antigen concentration: time course in the control group.



Figure 6

Von Willebrand Factor Antigen concentration: time course in the study group.


The ratio of vWF:RiCof to vWF:Ag was slightly decreased preoperatively in both groups (ratio=0.91; normal range greater than 0.7) and showed a marked increase in the postoperative period (ratio= 0.22) as, probably, new hemostatically effective large multimeric forms of vWF were released.
The correlations between CT and vWF:Ag and vWF:RiCof could not be calculated because of a number of patients had an infinite CT (>300 sec.). However, our results indicate that, on the whole, the prolongation of the CT was inversely proportional to the level of vWF:Ag and of vWF:RiCof (Figure 3).
Postoperative bleeding did not differ in the two groups (280 ml [220-400, 25th-75th percentile] vs 300 ml [180-400, 25th-75th percentile]) and no significant correlation was detected between nor preoperative or post-protamine CT and postoperative mediastinal drainage. However, among the patients population, five patients had no formation of the platelet plug at baseline (CT>300 or non-closure): this subgroup showed a statistically significant higher postoperative bleeding as compared to the rest of other patients (350 ml [240-400, 25th-75th percentile] vs 220 ml [180-290, 25th-75th percentile]), p=0.05) which was related to CT at baseline (r=0.441, p=0.035).
No patients suffered of any thromboembolic complication.



Patients with high intracardiac shear stress have a certain degree of proteolysis of high molecular weight vWF and such a condition may contribute to the increased risk of postoperative bleeding. PFA-100 has been found to be both sensitive (90-100%) and specific (88-95%) in the detection of von Willebrand disease [13,14,15,16].
Our study shows that patients with mitral valve regurgitation due to prolapse have high intracardiac shear stress which is responsible for the development of defects of primary hemostasis, which can be detected by means of PFA-100.
Despite these biological alterations, however, these patients do not show any pathological mucocutaneous or surgical bleeding, confirming the discrepancy between the low frequency of bleeding symptoms and the high prevalence of hemostatic abnormalities in this setting. This subclinical defect of primary hemostasis is quickly corrected by valve replacement due to the quick restoration of vWF levels; however it is responsible for a higher postoperative bleeding in those patients with the most severe deficiencies.
The PFA-100, however, is a sensitive, specific and reproducible test which might be useful to screen patients preoperatively to identify those who could benefit from correct diagnosis and appropriate therapy to reduce their complications.
High shear rates prevail in stenotic valves: under these conditions, large molecular weight multimers of vWF are changed in shape and exposed to the action of, which results in their proteolysis. High molecular weight multimers of vWF are the most effective in platelet mediated hemostasis. Platelet adhesion to the subendothelial matrix of injured arterial vessels wall requires adhesion of platelets to subendothelially located vWF and of plasma vWF to subendothelial collagen. In addition, vWF enhances the adhesion of platelets to fibrin clots and stabilizes coagulation factor VIII [17,18,19,20].
Although aortic stenosis is a relevant in vivo model of shear stress induced cleavage of vWF with resultant loss of its largest multimers, fluid dynamics of mitral regurgitation are also characterized by high shear stress.
These anomalies measured with PFA-100 were reversed by surgical correction of the regurgitant valve on the first postoperative day, a finding which is consistent with the 12-to-20-hour half-life of vWF in vivo.
The correction of PFA-100 abnormalities after surgical repair or replacement of the diseased valve indicates that the passage of blood through the diseased valve causes a disturbance of flow, thereby generating high shear stress and turbulence, may be responsible, at least in part, for a heightened vWF proteolysis. However, the overall imbalance of primary hemostasis is usually mild and the PFA-100 abnormalities are not an accurate predictor of a bleeding tendency, unless non closure is present [21].
Froom et al. [22] reported their evaluation of vWF in mitral valve prolapse: they conclude that there is a relationship between mitral valve prolapse and low levels of vWF:Ag. However, vWF:Ag levels are normal in those patients with mitral valve prolapse and heart failure, as it may account for an increased release of vWF.
Slaughter et al. [23] have investigated the PFA-100 in CABG patients: they showed that an impaired ADP mediated platelet aggregation is not corrected after surgery in 17 of 19 patients. Moreover, they reported that 80% of patients having CABG with abnormal chest tube output had prolonged PFA-100 measurements 15 minutes after administration of protamine sulphate.
Raman et al. [24] identified 16 bleeders after CABG: in 15 of those whose bleeding was controlled by platelet transfusion, PFA-100 after protamine sulphate was prolonged. They suggested a potential role for PFA-100 measurements to help identify which patient could benefit from platelet transfusion after CABG.
Several limitations of our study should be acknowledged. First, plasma concentration of vWF has large variations, as it is influenced by ABO blood type, Lewis blood type race and age [25].
Moreover, the plasma level of vWF transiently increases in response to a variety of physiological conditions including trauma, surgery, atherosclerosis, inflammatory states and β-adrenergic stimulation.



In conclusion, mitral regurgitation secondary to prolapse is responsible for a certain degree of proteolysis of high molecular weight vWF as shown by abnormal shear induced platelet aggregation. This defect of primary hemostasis is quickly corrected by valve repair due to the prompt restoration of vWF levels and this condition is not associated with an increased risk of bleeding, except for those patients with the most severe deficiencies.
On the basis of these data, postoperative bleeding after mitral valve repair should prompt a search for other causes than von Willebrand disease.
In addition, our observation confirms the discrepancy between the low frequency of bleeding symptoms and the high prevalence of hemostatic abnormalities in the setting heart valve disease. Although aortic stenosis and mitral regurgitation are a relevant in vivo model of shear-stress induced cleavage of vWF with resultant loss of its largest multimers, the diagnosis of von Willebrand syndrome should be restricted to those patients who have bleeding.
Shear induced platelet aggregation is significantly increased after cardiac surgery, but the role that this phe-nomenon might play in the development of postoperative thromboembolic complications remains to be addressed.



  1. Vincentelli A, Susen S, Le Tourneau T, et alii. Acquired von Willebrand syndrome in aortic stenosis. N Engl J Med 2003; 349: 343-349.
  2. Yoshida K, Tobe S, Kawata M. Acquired von Willebrand Disease Type IIA in patients with aortic valve stenosis. Ann Thorac Surg 2006; 81: 1114-1116.
  3. Pareti FI, Lattuada A,, Bressi C, et alii. Proteolysis of von Willebrand factor and shear stress-induced platelet aggregation in patients with aortic valve stenosi. Circulation 2000; 102: 1290-1295.
  4. Sadler JE. Sadler JE. Aortic stenosis, von Willebrand factor, and bleeding. N Engl J Med 2003; 349: 323-325. Sadler JE.
  5. Goldsmith IR, Blann AD, Patel RL, Lip GY. Effect of aortic valve replacement on plasma soluble P-selectin, von Willebrand factor, and fibrinogen. Am J Cardiol 2001; 87: 107-110.
  6. King RM, Pluth JR, Giuliani ER. The association of unexplained gastrointestinal bleeding with calcific aortic stenosis. Ann Thorac Surg 1987; 44: 514-516.
  7. Anderson RP, McGrath K, Street A. Reversal of aortic stenosis, bleeding gastrointestinal angiodysplasia, and von Willebrand syndrome by aortic valve replacement. Lancet 1996; 347: 689-690.
  8. Warkentin TE, Moore JC, Morgan DG. Aortic stenosis and bleeding gastrointestinal angiodysplasia: is acquired von Willebrand disease the link? Lancet 1992; 340: 35-37.
  9. Gill JC, Wilson AD, Endres-Brooks J, Montgomery RR. Loss of the largest von Willebrand factor multimers from the plasma of patients with congenital cardiac defects. Blood 1986; 67: 758-761.
  10. Rauch R, Budde U, Koch A, et alii. Acquired von Willebrand syndrome in children with patent ductus arteriosus. Heart 2002; 88: 87-88.
  11. Martini F, Zuppiroli A, Gori A, et alii. Platelet and blood clotting activation in patients with mitral valve prolapse. Thromb Res 1996; 83: 299-306.
  12. Lattuada A, Preda L, Sacchi E, et alii. A rapid assay for ristocetin cofactor activity using an automated coagulometer (ACL9000). Blood Coagul Fibrinolysis 2004; 15: 505-511.
  13. Kundu SK, Heilmann EJ, Sio R, et alii. Description of an in vitro platelet function analyzer--PFA-100. Semin Thromb Hemost 1995; 21: 106-112.
  14. Kundu SK, Heilmann EJ, Sio R, et alii. Characterization of an in vitro platelet function analyzer, PFA-100. Clin Appl Thromb Hemostas 1996; 2: 241-9.
  15. Mammen EF, Comp PC, Gosselin R, et alii. PFA-100 system: a new method for assessment of platelet dysfunction. Semin Thromb Hemost 1998; 24: 195-202.
  16. Cattaneo M, Federici AB, Lecchi A, et alii. Evaluation of the PFA-100 system in the diagnosis and therapeutic monitoring of patients with von Willebrand disease. Thromb Haemost 1999; 82: 35-39.
  17. Mannucci PM. Treatment of von Willebrand’s Disease. N Engl J Med 2004; 351: 683-694.
  18. Mohri H. Acquired von Willebrand syndrome: its pathophysiology, laboratory features and management. J Thromb Thrombolysis 2003; 15: 141-149.
  19. Reiter RA, Knöbl P, Varadi K, Turecek PL. Changes in von Willebrand factor-cleaving protease (ADAMTS13) activity after infusion of desmopressin. Blood 2003; 101: 946-948.
  20. Matsui H, Sugimoto M, Mizuno T, et alii. Distinct and concerted functions of von Willebrand factor and fibrinogen in mural thrombus growth under high shear flow. Blood 2002; 100: 360-410.
  21. Buyukasik Y, Karakus S, Goker H, et alii. Rational use of the PFA-100 device for screening of platelet function disorders and von Willebrand disease. Blood Coagul Fibrinolysis 2002; 13: 349-353.
  22. Froom P, Margulis T, Grenadier E, et alii. Willebrand factor and mitral valve prolapse. Thromb Haemost 1988; 60: 230-231.
  23. Slaughter TF, Sreeram G, Sharma AD, et alii. Reversible shear-mediated platelet dysfunction during cardiac surgery as assessed by the PFA-100 platelet function analyzer. Blood Coagul Fibrinolysis 2001; 12: 85-93.
  24. Raman S, Silverman NA. Clinical utility of the platelet function analyzer (PFA-100) in cardiothoracic procedures involving extracorporeal circulation. J Thorac Cardiovasc Surg 2001; 122: 190-191.
  25. Nitu-Whalley IC, Lee CA, Brown SA, et alii. The role of the platelet function analyser (PFA-100) in the characterization of patients with von Willebrand’s disease and its relationships with von Willebrand factor and the ABO blood group. Haemophilia 2003; 9: 298-302.

Conflict of interest: No conflict of interest acknowledged by the authors.