Changes of haemostasis in patients undergoing major abdominal surgery—is there a difference between elderly and younger patients?

J. Boldt *,1, I. Hüttner1, St. Suttner1, B. Kumle1, S. N. Piper1 and G. Berchthold2

1Department of Anaesthesiology and Intensive Care Medicine and 2Clinic of Surgery, Klinikum der Stadt Ludwigshafen, Ludwigshafen, Germany*Corresponding author: Department of Anaesthesiology and Intensive Care Medicine, Klinikum der Stadt Ludwigshafen, Bremserstr. 79, D-67063 Ludwigshafen, Germany

Accepted for publication: May 2, 2001


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Many physiological functions are altered in the elderly. It is not clear whether this applies to haemostatic mechanisms in patients undergoing major surgery. Twenty-five consecutive patients <60 yr and 25 consecutive patients >70 yr scheduled for abdominal surgery for malignancies were included in our study. Various standard coagulation variables and specific markers of coagulation were serially measured before surgery (baseline), at arrival on the intensive care unit (ICU), 4 h after arrival on ICU, and on the morning of the first postoperative day. Platelet function was assessed using the Platelet Function Analyser PFA-100TM with adenosine diphosphate (ADP) as an inductor. Anaesthesia and surgery were similar between the elderly (76(3) years) and younger (53(5) years) groups. Baseline plasma levels of prothrombin fragments F1+2, thrombin-antithrombin III (TAT) complex, and D-dimers were significantly different between the two groups, indicating thrombin activation and fibrin formation in the elderly. Postoperatively, only F1+2 plasma levels were significantly higher in the elderly (4.0(0.8) nmol/l) than in the younger patients (2.2(0.9) nmol/l), whereas the course of D-dimer and TAT did not differ significantly between the two groups. Endothelial-derived markers of coagulation (von Willebrand factor, collagen-binding activity of von Willebrand factor) were not different between the groups throughout the study period. Platelet function was impaired in the elderly compared with the younger patients. It is concluded that elderly patients showed more prothrombin activation/thrombin generation and increased fibrinolytic activity prior to surgery than younger patients. However, perioperative changes of coagulation in the elderly were similar to those seen in younger patients.

Br J Anaesth 2001; 87: 435–40

Keywords: surgery, abdominal; age factors; blood, haemostasis; blood, platelet function


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The proportion of elderly people in the population is increasing rapidly.1 By the year 2030, 17% of the population in the USA will be >65 yr.2 It is generally accepted that the function of several organs is altered in the elderly and the margin of reserve for adaptation is reduced.3 A deterioration in pulmonary, renal, and cardiovascular function is well documented.46 Secondary to the high incidence of co-existing diseases elderly patients are at increased risk of developing postoperative complications.1 Age-specific alterations of organ function appear to predispose to the development of postoperative multiorgan failure.7 8 The coagulation system may also be affected by ageing. 9 Imbalances in haemostasis are commonly seen in the surgical patient either due to extensive blood loss, hypothermia or activation of inflammatory pathways with subsequent activation of procoagulatory mechanisms and down-regulation of anticoagulant pathways. Even moderate surgical trauma (e.g. reduction mammoplasty) was shown to activate thrombin generation and fibrinolysis.10 More severe alterations of the haemostatic process are expected in patients undergoing complex, long-lasting surgical procedures. The development of sensitive immunochemical methods enables us to increase our understanding of the coagulation process. Since the influence of advanced age on coagulation in the surgical patient is not well defined, this observational study was designed to assess whether haemostasis differs between elderly and younger patients undergoing major abdominal surgery. For this purpose, various aspects of the coagulation process were serially measured: thrombin generation/thrombin neutralization, fibrin formation/degradation, endothelial-related coagulation, and platelet function.


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The study was performed according to the guidelines of the Institutional Review Board of the hospital. Before the patients were included into the study, written informed consent was obtained from each patient. Twenty-five consecutive elderly (>70 yr) and 25 younger patients (<60 yr) scheduled for major abdominal surgery for malignancies were studied. Patients with preoperative coagulation abnormalities (activated partial thromboplastin time [aPTT] >70 s, fibrinogen <1.5 g dl –1, antithrombin III [AT III] <40%), patients showing perioperative renal insuffiency (plasma creatinine concentration >2 mg dl–1), liver dysfunction (alanine transferase/aspartate aminotransferase >40 U litre–1), diabetes mellitus, or valvular heart disorders were excluded from the study. Patients taking corticosteroids, aspirin, cyclooxygenase inhibitors or other non-steroidal anti-inflammatory agents were not included.

All patients received premedication with lorazepam 1 to 2 mg. Anaesthesia was induced by weight-related doses of thiopental, sufentanil, and atracurium. Anaesthesia was maintained by giving sufentanil, atracurium and isoflurane according to the patients’ need. The lungs of all patients were mechanically ventilated with 60% nitrous oxide in oxygen to keep SaO2 >95% (continuous oximetry). Ventilation patterns were adjusted to keep end-expiratory carbon dioxide between 35 and 40 mm Hg (continuous capnography). After surgery all patients were transferred to the intensive care unit (ICU). On ICU, ventilation was continued if the body temperature was <36°C, gas exchange was insufficient or if there was haemodynamic instability. Regional anaesthesia was not used in the patients. The entire perioperative management was carried out by physicians who were not involved in the study.

Heart rate, mean arterial blood pressure, and central venous pressure (CVP) were measured continuously in all patients. The CVP was maintained between 12 and 14 mm Hg using crystalloid and/or gelatin infusions. Packed red blood cells were administered when the haemoglobin concentration was <9 g dl –1. Crystalloids were administered to compensate for fluid loss by sweating, gastric tubes, and urine output or as a solvent for drugs (e.g. antibiotics). During surgery, 500 ml h–1 of crystalloids were given routinely in all patients. When the mean arterial pressure was <60 mm Hg despite sufficient volume therapy, dopamine was given followed by epinephrine if the mean arterial pressure was still <60 mm Hg.

From arterial blood samples standard coagulation variables (AT III, fibrinogen, platelet count and aPTT) were measured using routine laboratory tests. Additionally, D-dimer (turbidimetric method, Roche Diagnostics, Mannheim, Germany; <0.5 ng litre–1), prothrombin fragments F1+2 (commercially available solid-phase enzyme-linked immonosorbent assay kit [ELISA], Dade-Behring, Marburg, Germany; normal values from healthy volunteers: 0.4–1.0 nmol litre –1), thrombin/antithrombin III complex (TAT; using commercially available ELISA, Dade-Behring, Marburg, Germany; normal values from healthy volunteers: 1.0–4.0 µg litre–1), factor VIII activity (one-stage clotting assay; Roche Diagnostics, Mannheim, Germany; normal values from healthy volunteers: 60–150 U ml–1), von Willebrand factor antigen (vWF-Ag; turbidimetric method, Roche Diagnostics, Mannheim, Germany; normal values from healthy volunteers: 60–160 U dl–1), collagen-binding activity of von Willebrand factor (vWF:CBA; by sandwich ELISA, Immuno-Diagnostika, Heidelberg, Germany; normal values from healthy volunteers: 0.6–1.8 U ml –1) were measured from the same arterial blood samples. Platelet volume was measured using Coulter counter STKS (Coultronics, Margency, France). Platelet function was assessed by the Platelet Function Analyser PFA-100TM (Dade-Behring, Marburg, Germany) using adenosine diphosphate (ADP) as inductor. The time required for the platelet plug to occlude the aperture is the ‘closure time’ and is indicative for platelet function.11 12 The normal range for the PFA-100TM test using ADP in healthy volunteers has been determined to be 77–133 s. 11 All analyses were performed in duplicate by a single operator according to the instructions of the manufacturer. Measurements were performed after induction of surgery (before surgery, defined as ‘baseline’ value; T0), at arrival on ICU (T1), 4 h after arrival on ICU (T3), and on the morning of the first postoperative day (T4).

Statistics
We decided that a 40% change of prothrombin fragment F1+2 from baseline values would reflect abnormal changes in coagulation. The appropriate standard deviation (SD) of F1+2 has been found to be 0.4 nmol litre–1.10 For a study with a power of 0.8, a minimum of 20 patients in each group would be required to detect a difference in F1+2 at the 0.05 level.13 Statistical analysis has been carried out using a SPSS/PC+ software (V4.0 SPSS, Inc., Chicago, USA). Data are presented as mean (SD), unless otherwise indicated. Two-way analysis of variance for repeated measurements (ANOVA) followed by post hoc Scheffé’s test on the matrix of pairwise comparison probabilities was used for all serially measured variables. P values <0.05 were considered significant. Other variables were analysed by one-way analysis of variance (for normally distributed) and Kruskall-Wallis one-way analysis of variance of ranks (when data were not normally distributed data). Chi-square test and Mann-Whitney U test was additionally used when appropriate.


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Duration of anaesthesia and the type of surgery were similar in the two groups (Table 1). The duration of stay on the ICU was significantly longer in the elderly than in the younger patients (3.4 (1.1) vs 5.5 (2.5) days) (Table 1). The blood loss up to the first postoperative day was not significantly different between the groups (Table 2). The volume of blood products and other intravenous fluids (Table 2) as well as haemodynamic variables (Table 3) were similar between the groups throughout the entire study period. No platelet concentrates or fibrinogen were given.


View this table:
[in this window]
[in a new window]
 
Table 1 Patients’ characteristics and data from the perioperative period. *P<0.05 different to the other group
 

View this table:
[in this window]
[in a new window]
 
Table 2 Cumulative use of blood and crystalloids/colloids. PRBC, packed red blood cells; FFP, fresh frozen plasma; +P<0.05 different from the other group
 

View this table:
[in this window]
[in a new window]
 
Table 3 Haemodynamic variables and standard coagulation variables in the two groups. +P <0.05 different to the other group; *P<0.05 different from baseline values
 
At baseline, the haemoglobin concentration was significantly lower in the elderly than in the younger patients (Table 3). From normal baseline values, fibrinogen plasma levels decreased similarly in both groups (Table 3). AT III was lower in the elderly at baseline and remained lower than in the younger patients at T2 and T3 (Table 3 ).

D-dimer levels were higher than normal (>0.5 ng litre–1) at baseline in the elderly and significantly different compared to the younger patients (Fig. 1). D-dimer increased in both groups without the changes showing significant group differences. F1+2 plasma levels at baseline were elevated beyond normal (>2 nmol litre–1) only in the elderly (Fig. 1). F1+2 increased in both groups, being significantly higher in the elderly than in the younger patients at T1 and T2. TAT plasma levels were higher than normal (>4 µg litre–1) at baseline only in the elderly. TAT increased similarly in both groups (elderly: from 8.1 (1.8) to 14.6 (2.9) µg litre–1; younger: from 4.9 (1.7) to 13.2 (4.5) µg litre–1) (Fig. 1). vWF, factor VIII, and vWF:CAB did not differ between the two groups throughout the study period (Fig. 2).



View larger version (15K):
[in this window]
[in a new window]
 
Fig 1 Changes of D-dimer (normal values: <0.5 ng litre–1), F1+2 (normal values: 0.4–1.0 nmol litre–1), and thrombin/antithrombin III (normal values: 1.0–4.0 µg litre–1) in the two groups. Mean ± SD; T0: after induction of surgery; T1: at arrival on ICU; T2: 4 h after arrival on ICU; T3: at the morning of the first postoperative day; +P<0.05 different to the other group; *P<0.05 different from baseline values.

 


View larger version (15K):
[in this window]
[in a new window]
 
Fig 2 Changes of von Willebrand-factor antigen (normal values: 60–160 U dl–1), factor VIII (normal values: 60–150 U ml–1) and collagen-binding activity of von Willebrand factor (normal values: 0.6–1.8 U ml –1). Mean ± SD; T0: after induction of surgery; T1: at arrival on ICU; T2: 4 h after arrival on ICU; T3: at the morning of the first postoperative day; *P<0.05 different from baseline values.

 
The platelet count and platelet volume were without group differences throughout the study (Fig. 3). PFA-100 TM using ADP revealed a significantly longer closing time at baseline in the elderly than in the younger patients (Fig. 3). Immediately after arrival on the ICU (T1), 4 h thereafter (T2), and on the first postoperative day (T3), closing time was significantly higher in the elderly (T2: 148 (18) s) than in the younger patients (T2: 120 (15) s) indicating altered platelet function.



View larger version (15K):
[in this window]
[in a new window]
 
Fig 3 Changes of platelet count (normal range: 150 000 to 350 000 ml–1), platelet volume (normal values: 7.8–10 fl), and closing time of the platelet factor analyser PFA-100TM using ADP as an inductor (normal values: 77–133 s) in the two groups. Mean ± SD; T0: after induction of surgery; T1: at arrival on ICU; T2: 4 h after arrival on ICU; T3: at the morning of the first postoperative day; +P<0.05 different to the other group; *P<0.05 different from baseline values.

 

    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Surgery may cause severe modifications in haemostasis with subsequent development of a hyper- or hypocoagulable state. In addition to the surgical trauma, the neuroendocrine response and hypothermia may also impair haemostatic function resulting in decreased fibrinolysis and altered platelet activity.14 We have included only patients undergoing abdominal surgery because it has been reported that different types of surgery were associated with varying effects on haemostasis. 15 We monitored variables that appear to be more helpful for the diagnosis of coagulation disorders than commonly used global tests. 16 17 Different aspects of haemostasis were assessed: F1+2 (a marker of thrombin generation), TAT (a marker of thrombin neutralization), D-dimer (a marker of fibrin formation and its degradation). Endothelial-related coagulation was assessed by monitoring vWF and vWF:CAB. vWF is a complex that mediates platelet adhesion to a damaged vessel wall. Platelet function was measured using the PFA-100TM that simulates coagulation close to normal haemorheological conditions.12 The PFA-100TM uses whole blood and provides a more realistic environment to the aggregation of platelets than other tests (e.g. aggregometry). 12 When the PFA-ADP test is abnormal, the patient is likely to have a platelet defect that impairs primary haemostasis. 11 12

One major result of our study was that elderly patients showed evidence of activated coagulation at baseline (elevated F1+2 and TAT complex) associated with increased fibrinolytic activation (elevated D-dimer) and increased inhibitor consumption (lower AT III, elevated TAT complex). Thrombin production plays a pivotal role in the development of coagulation dysfunction. Excessive thrombin generation may result in consumption of platelets and fibrinogen. Continuous activation of coagulation will result in a hypocoagulable state which is complicated by further (hyper-)fibrinolysis. At baseline, TAT and F1+2 were elevated only in our elderly patients. Both are regarded as indicators of hypercoagulability. 19 The reasons for this phenomenon can only be speculated upon: intravascular volume deficit, (occult) microcirculatory perfusion deficit, or an age-related general change in the balance between procoagulant and anticoagulant mechanisms may be possible explanations. Since liver function was normal in the elderly throughout the study period, reduced synthesis of antithrombin activity appears to be less likely. Platelet function assessed with the PFA-100TM was altered in the elderly at baseline possibly due to elevated thrombin concentration and an ongoing activation of coagulation. Hypercoagulability may be beneficial in reducing surgical bleeding, but it may also have considerable negative effects, e.g. development of venous thrombosis, pulmonary embolism or coronary artery thrombosis.20 Tuman et al. found that postoperative hypercoagulability increased cardiac morbidity and affected outcome considerably in patients undergoing major vascular surgery. 21 Our study population was much too small to draw conclusions with regard to development of thrombosis, outcome or survival.

With the exception of F1+2, AT III and platelet function all other serially measured coagulation variables did not show significant differences between the elderly and younger patients during and after surgery. Measurement of F1+2 appears to be a sensitive tool for detecting accelerated thrombin generation. Thrombin generation and consecutive activation of fibrinolysis are physiological responses to tissue damage.21 F1+2 in the elderly after surgery was moderately but significantly higher than in our younger patients.

Ongoing activation of the coagulation system may imbalance the haemostatic system and result in increased postoperative bleeding. Reduced platelet function may contribute to a disturbed coagulation process. Our data are in agreement with a study assessing the importance of molecular markers of coagulation in patients undergoing different surgical procedures showing that F1+2 was the best predictor of intraoperative haemostatic disorders, followed by D-dimer.21

There is a tendency to assume that the elderly patient has a worse prognosis than the younger patient because of the presence of pre-existing comorbidity with age.3 In addition to the established age-related changes in lungs, kidneys, and cardiovascular system, there may also be differences with regard to haemostasis. This was confirmed by our data showing a moderate imbalance of the coagulation network prior to surgery. This may be of particular importance in those elderly patients who are scheduled for complex surgery and in whom complications arise postoperatively. In this situation activation of inflammatory pathways may result in additional modifications of the haemostatic network associated with the risk of bleeding or microthrombosis with subsequent development of organ failure.22


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Veering BT. Management of anaesthesia in the elderly patient. Curr Opinion Anaesthesiol 1999; 12: 333–336

2 US Bureau of Census. Statistical Abstracts of the United States. 113th edn. Washington, DC: Department of Commerce, 1993

3 Yates DW. Trauma in the elderly. In: Vincent JL, ed. Yearbook of Intensive Care and Emergency Medicine. Berlin Heidelberg New York: Springer; 1995: 827–833

4 Huchon G. Lung diseases in the elderly. Respiration 1998; 65: 343–344[ISI][Medline]

5 Dodds C, Allison J. Postoperative cognitive deficits in the elderly surical patient. Br J Anaesth 1998; 81: 449–462[ISI][Medline]

6 Piccione W. Cardiac surgery in the elderly: what have we learnt? Crit Care Med 1998; 26: 196–197[ISI][Medline]

7 Gransden WR, Eykyn SJ, Phillips I. Septicaemia in the newborn and elderly. J Antimicrob Chemother 1994; 34 (suppl A): 101–119[Abstract]

8 Fink P. Antibiotic therapy of intra-abdominal sepsis in the elderly: experience with ticarcillin and clavulanic acid. Surg Gynecol Obstet 1991; 172 (suppl): 36–41[ISI][Medline]

9 Boldt J, Schindler E, Knothe Ch, et al. Endothelial-related coagulation in cardiac surgery. Br J Anaesth 1995; 74: 174–179[Abstract/Free Full Text]

10 Payen JF, Baruch M, Horviller E, et al. Changes of specific markers of haemostasis during reduction mammoplasty. Br J Anaesth 1998; 80: 464–466[ISI][Medline]

11 Mammen EF, Alshameeri RS, Comp PC. Preliminary data from a field trial of the PFA-100 system. Sem Thromb Hemost 1995; 21 (suppl 2): 113–121[ISI][Medline]

12 Kundu SK, Heilman E, Sio R, et al. Characterization of an in vitro function analyzer, PFA-100TM. Clin Appl Thromb/Hemost 1996; 2: 241–249

13 Pocock SJ. Clinical Trials. A Practical Approach. Chichester New York Brisbane; Wiley & Sons, 1998; 123–141

14 Niemi TT, Kuitunen AH, Vahtera EM, Rosenberg PH. Haemostatic changes caused by i.v. regional anaesthesia with lignocaine. Br J Anaesth 1996; 76: 822–828[Abstract/Free Full Text]

15 Rosenfeld BA, Faraday N, Dorman T, Herfel B. Comparison of perioperative hemostatic function between vascular and cancer surgery patients. Anesth Analg 1995; 80: SCA45

16 Fareed J, Bick RL, Hoppenstead DA, Bermes EW. Molecular markers of hemostatic activation: applications in the diagnosis of thrombosis and vascular and thrombotic disorders. Clin Appl Thromb/Hemost 1995; 2: 87–102

17 Spiess BD. Maintenance of homeostasis in coagulation during cardiopulmonary bypass. J Cardiothorac Vasc Anesth 1999; 13: 2–5[ISI][Medline]

18 Kambayashi J, Sakon M, Yokota M, et al. Activation of coagulation and fibrinolysis during surgery, analyzed by molecular markers. Thromb Res 1990; 60: 157–167

19 Gibbs NM, Crawford PM, Michalopoulos N. Postoperative changes in coagulant and anticoagulant factors following abdominal aortic surgery. J Cardiothorac Vasc Anesth 1992; 6: 680–685[Medline]

20 Tuman KJ, McCarthy RJ, March RJ, et al. Effects of epidural anesthesia and analgesia on coagulation and outcome after major vascular surgery. Anesth Analg 1991; 73: 696–704[Abstract]

21 Korte W, Truttmann B, Heim C, et al. Preoperative values of molecular coagulation markers identify patients at low risk for perioperative haemostatic disorders and excessive blood loss. Clin Chem Lab Med 1998; 36: 235–240[ISI][Medline]

22 Webb AR, Mythen MG, Jacobson D, Mackie J. Maintaining blood flow in the extracorporeal circuit: haemostasis and anticoagulation. Intensive Care Med 1995; 21: 84–93[ISI][Medline]