Intensive Care Med (1990) 16:121-124

Intensive Care Medicine 9 Springer-Verlag1990

Meningococcemia and purpura fulminans in adults: acute deficiencies of proteins C and S and early treatment with antithrombin III concentrates F. Fourrier 1, P. Lestavel 1, C. C h o p i n 1, A. Marey 2, J. G o u d e m a n d 2, A. Rime 1 and J. Mangalaboyi 1 1Service de R6animation Polyvalenteand 2Laboratoire d'H6matologie, H6pital B, Lille, France Received: 3 January 1989;; accepted: 7 August 1989

Abstract. It has been recently suggested that an acquired deficiency o f proteins C and S could contribute to the pathogenesis of meningococcemic purpura fulminans (PF) in children. Our study was designed to measure the levels of antithrombin III (AT III), protein C, and protein S during adult PF and to determine the effects o f an early infusion of high doses of AT III concentrates on clinical and biological alterations of PF. We studied five consecutive adult patients with meningococcemia (type B) and PF. The levels o f AT III, protein C (antigen and activity), and protein S (total and free) were measured at admission and 24 h and 1 month later. The treatment included in each case: amoxycillin, dobutamine and high doses of AT III concentrates. All patients survived and were discharged without any sequelae. At admission, biological data were consistent with severely depressed protein C and protein S levels and moderately decreased AT III levels, without any discrepancy between protein C antigen and activity. After 24 h, AT III and protein S levels were within normal ranges, whereas protein C levels were still depressed. These data are consistent with the theory of a particular imbalance in the anticoagulant systems during meningococcemic PF, contrasting with the usual findings observed during septic disseminated intravascular coagulation. The possibility must be considered that high doses of one anticoagulant (AT III concentrates) could compensate for the acute decrease in the other (protein C system). Key words: Meningococcemia - Purpura fulminans Protein C - Antithrombin III

has been recently suggested that an acquired deficiency o f protein C and protein S activity could contribute to the pathogenesis o f the disease. Powars et al. [1] reported a drop in protein C and free protein S levels in six children with PF and meningococcemia. They suggested that the immaturity of the protein C system might explain the increased risk o f PF in this age group and that the usual failure of heparin therapy might be related to protein C and S deficiency. Antithrombin III (AT III) is another major modulator o f the clotting cascade. It exhibits its activity against thrombin, factors IXa, Xa, XIa, XIIa, plasmin, and kallikrein and controls in part the activation of protein C by the thrombin-thrombomodulin complex [2]. A decrease in AT III levels has been previously documented in severe sepsis and endotoxinemia [3]. Experimental studies have shown that infusion o f AT III concentrates can prevent the endotoxin induced disseminated intravascular coagulation (DIC). [4, 5]. Several reports have suggested that AT III concentrates could be useful in septic DIC in humans. One controlled study showed that AT III supplementation reduced the duration of hematologic disorders [6]. To determine the changes in the three major inhibitors of blood coagulation we measured AT III, protein C, and protein S levels in five adults with meningococcemia and P E We postulated that the infusion of AT III concentrates could stem the course o f the disease. The five patients were treated early with high doses o f AT III concentrates and changes in their clinical and biological variables were noted in sequential measurements. Patients

Purpura fulminans (PF) occurs predominantly in young children. Microvascular thrombosis and hemorragic necrosis leading to distal gangrene are its main features. It Address reprint requests to: F. Fourrier, M.D., Servicede R6animation

Polyvalente. H6pital B, C.H.R.U., F-59037Lille Cedex, France

The five patients were consecutivelystudied from September 1987 to May 1988.They were directlyadmitted from home in our intensivecare unit. Their main clinical featuresare reproducedin Table 1. The diagnosis of PF was assessedon the basis of the followingcriteria: severeinfectious syndrome, diffuse cutaneous purpura, and circulatory shock. Meningococcemicinfection was diagnosed either on the basis of positive blood cultures (one case), cerebrospinalfluid cultures (three cases),

122

F. Fourrier et al.: Meningococcemia and purpura fulminans in adults

Table 1. Patients' clinical data.at admission

Patients

Age (years) SAPS Glasgow score AP: S / D / M (mmHg) Urine output CVP (cmH20) Temperature Purpura Distal vascular occlusion Hemorrhage Meningitis

1

2

3

4

5

22 14 12 60/30/50 0 17 38.9 ~ Diffuse + + +

20 13 11 180/110/130 0 15 39.6 ~ Diffuse +

18 11 11 80/50/65 120 ml/h 15 39.4 ~ Diffuse +

19 11 10 90/40/55 350 ml/h ND 39 ~ Diffuse +

18 16 12 55/20/40 0 15 38.6 ~ Diffuse + + +

SAPS, simplified acute physiological score; AP: S/D/M, arterial pressure: systolic/diastolic/mean; CVP, central venous pressure; ND, not done

or counterimmunoelectrophoresis (one case). Type B meningococcus was isolated in four cases. All patients presented at admission with diffuse purpura, evident signs of septic shock, and acute ventricular failure as reflected by severe hypotension, increase in central venous pressure, cardiac enlargement, and subacute pulmonary edema. No arterial or right heart catheterization was performed, in order to avoid any delay in therapy or risk of hemorrhage due to central venous or arterial punctures. Patient 2 exhibited a transient circulatory arrest in severe sinoatrial dissociation. Patients 1, 2, and 5 were in acute anuric renal failure. Initial laboratory findings are reported in "Pable 2. No patient had biological signs of liver failure as assessed by ASAT (aspartate transaminase), ALAT (alanine transaminase), lactate dehydrogenase and alkaline phosphatase levels. Patients 1, 2, and 4 had mild elevations of serum free bilirubin. Treatment was instituted immediately after admission with dobutamine (20 Ixg/kg/min), amoxycillin (200 mg/kg/24 h) and AT III concentrates (60 IU/kg/3 h, as loading dosage, then 60 IU/kg/24 h for 2 days).

Methods

AT III activity was measured by colorimetry (Stachrom AT III, Diagnostica Stago Labs, Asni~res, France) [7]: after plasma incubation with heparin and thrombin in excess, residual thrombin was measured according to its amidolytic activity on the chromogenic substrate CBS 3447 (release of p-nitroaniline; optical density measured at 405 nm). Normal values were determined with a plasma AT III calibrator with a pool of ten plasmas from healthy subjects (mean 100• range 84-120~ Protein C antigen was measured by the immunoenzymatic "sandwich" method. Protein C activity was measured by colorimetry (Stachrom Protein C, Diagnostica Stago Labs) [8]: plasma protein C was activated with a specific snake venom (Agkistrodon C. Contortrix) and measured according to its amidolytic activity on the chromogenic substrate CBS 6525 (release of p-nitroaniline; optical density measured at 405 nm). Normal values were determined with a protein C calibrator

Table 2. Patients' biological data at admission

Patients t

2

3

4

5

Cerebrospinal fluid Protein (g/l) Glucose (g/l) Cells (/mm 3) PMN (~

0.27 0.82 60 100

3.40 0.60 850 100

2.80 0.50 4000 98

1.10 0.06 13 600 96

0.33 0.69 11 100

Blood BUN (g/l) Creatinine (mg/1) Red cells ( x 106/mm3) Leukocytes ( x 103/mm 3) Platelets ( x 103/mm3) APTT (NR: 27 - 33 s) PT (070) Factor V (070) Factors V I I - X (070) Fibrinogen (g/l) SCFM FDP (mg/1) TT (s)

0.65 32 3.65 22.2 113 40 45 69 32 3.1 + Negative 18

0.42 18 4.90 20.0 149 36 75 83 52 4.6 + Negative 18

0.40 13 4.80 25.2 188 48 45 66 54 4.3 + Negative 19

0.40 11 5.30 24.5 230 31 56 81 43 6.1 + Negative 20

0.60 44 3.20 17.9 142 40 39 50 48 3.4 + Negative 22

PMN, polymorphonuclear neutrophils; BUN, blood urea nitrogen; APTT (NR), activated partial thromboplastin time (normal range); PT, prothrombin time; FDP, Fibrin degradation products; TT, thrombin time

E Fourrier et al.: Meningococcemiaand purpura fulminans in adults

123

with a pool of ten plasmas from healthy subjects (mean 100.9+_14.1; range 70-140070). Protein S antigen was measured accordingto the method of Laurell [9] by electroimmunodiffusion (Assera Plate Protein S, Diagnostica Stago Labs). The dosage of the free protein S, was calculated according to the method of Comp et al. [10]:the C4BP complexedprotein S was precipitated by PEG and the free protein S levelsmeasured by electroimmunodiffusion in the plasma supernatant. Normal values in 10 healthy subjects were 103+_21070(range 70-140070). The reproducibility of the method was +_2.9070at 100070level. Minimal detection was 5070. Plasma fibrinogen was measured accordingto the method of Clauss [11]. Normal values in healthy subjects were 2-4 g/1. The detection of soluble complexes of fibrin monomers (SCFM) was done by hemagglutination (E S. Test, Diagnostica Stago Labs) [12]. The fibrin degradation products (FDP) were measured by latex agglutination (Diagnostica Stago Labs). Normal values are < 10 mg/1.

thrombin time was normal. These points were consistent with lack of fibrinolytic activity. At the same time, the levels of AT III activity were slightly decreased, and there was an important decrease in the levels o f protein C antigen and activity as well as total and free protein S. No discrepancy between protein C antigen and activity could be found. After 24 h (and infusion o f 120 I U / k g o f AT III concentrates) the AT III levels were completely corrected. All patients exhibited slight fibrinolysis with the presence o f fibrin degradation products (20 to 160 mg/l). In each case, thrombocytopenia became worse (mean decrease 28.6_ 18.10a/mm3, range 1 0 - 58). Prothrombin time and fibrinogen levels improved (prothrombin time: mean increase 15.6+6070, range 1 0 - 2 4 ; fibrinogen: mean increase 2.1___0.9g/1, range 1.2-3.5). SCFM were no longer detected. The levels of protein C antigen and activity remained low in four patients. One m o n t h after admission, all patients had normal values for coagulation times, platelet count, AT III and protein C and protein S levels. Unfortunately levels of protein C were not measured daily and it was impossible to know exactly when protein C returned to within the normal range.

Results

Clinical course The five patients survived and were discharged without any sequelae 1 0 - 1 8 days after admission. In all cases, the circulatory shock improved in 10 h or less. No patient required controlled ventilation for respiratory failure or hemodialysis for acute renal failure. Biochemical test results returned to within normal ranges in 48 h. Dobutamine therapy was stopped after 24 h in 4 cases and 36 h in one. After infusion of AT III concentrates, preventive routine heparin prophylaxis was subsequently instituted with subcutaneous low molecular weight heparin (enoxaparin: Lovenox, 20 mg/24 h). Antibiotic therapy was maintained for 3 weeks intravenously or orally at the same dosage. No patient had hemorrhagic or major thrombotic complications and the cutaneous necrotic or purpuric lesions improved without sequelae.

Hematologic data Sequential results are shown in Table 3. At admission, the five patients fulfilled the criteria for incipient disseminated intravascular coagulation: decrease in prothrombin time and factors II, V, and VII-X, slight thrombocytopenia and detection of SCFM. No patient had measurable levels of fibrin degradation products and the

Comments

This report o f an uncontrolled study including only five patients precludes any statistical comparison or conclusion about a possible improvement due to AT III supplementation. However, we would like to emphasize the following points. Firstly, these five patients had the classical features o f PF: rapid onset o f severe infectious syndrome and multiple organ failure with cerebral, cardiac and renal involvement. The rise in central venous pressure, cardiac enlargement and subacute pulmonary edema were consistent with early myocardial depression due to meningococcemia. The cerebral, cardiac, and renal involvement were likely secondary to thrombosis of the microcirculation. Their very rapid improvement was unusual.

Table 3. Results of hematologic tests at admission and after 24 h AT III therapy

Patients 1

Protein C ag. (~ Protein C act. (070) Protein ST (070) Protein SF (~ AT III (act.) (070) PT (070) Platelets (103/mm3) Fibrinogen (g/l)

2

3

4

5

Adm.

24 h

Adm.

24 h

Adm.

24 h

Adm.

24 h

Adm.

24 h

29 32 62 55 60 45 113 3.1

36 48 100 100 84 69 82 4.6

36 30 55 50 82 75 149 4.6

56 50 100 85 103 87 139 6.8

44 44 60 50 80 45 158 4.3

62 48 90 60 110 55 130 7.8

48 47 55 50 70 56 230 6.1

62 62 85 100 110 76 200 7.3

45 45 75 85 75 39 142 3.4

95 90 110 83 51 125 5.8

Adm., admission; ag., antigen; act., activity; T, total; F, free; PT, prothrombin time

124

E Fourrier et al.: Meningococcemiaand purpura fulminans in adults

Secondly, the hematologic data were consistent with the diagnosis of disseminated intravascular coagulation (presence of SCFM) with lack of secondary fibrinolysis. Thirdly, the measurements of protein C and S levels confirm the existence of an acquired acute deficiency in the protein C system during meningococcemic P E The levels we observed are very near those reported by Powars et al. in young children [1], so it seems questionable that an immaturity of the protein C system might explain the frequency of the disease in this age group, as was suggested by Powars. A decrease in protein C and S levels has been previously documented in DIC. Griffin et al. and Mannucci et al. [13, 14] have shown that protein C levels were reduced during either active or compensated DIC. In these situations, in vivo generation o f thrombin might explain the consumption of protein C, after activation by the thrombin-thrombomodulin complex. However, it has been suggested that the short half-life o f protein C (6 h) and its hepatic vitamin K-dependent synthesis must allow its rapid replacement as long as the liver is functioning normally, and indeed, in chronic DIC [15] the decrease in protein C levels correlated with liver dysfunction as assessed by measurements of pseudocholinesterase and albumin. It was therefore suggested that severe depression o f protein C levels would only occur in case of overwhelming defibrination or severe liver failure. That was not the case in our patients, who had neither impairment in hepatic function nor severe decrease in fibrinogen levels. Another mechanism of protein C depression was proposed by Marlar et al. [16]. They performed sequential measurements of protein C levels during acute DIC and showed that the ratio protein C activity:protein C antigen decreased early during the course of the disease, suggesting that protein C might lose its antithrombotic property by forming a complex with a plasma inhibitor. These results were not completely confirmed by Vandermeer et al. [17], who found that the protein C activity: antigen ratio was decreased in only 50~ o f the 13 cases of DIC they studied. In our patients this ratio was always within the normal range. Thus, the mechanism o f protein C depression remains unclear. Whatever it may be, the possibility must be considered that an early depression of protein C activity might account for the thrombotic imbalance and the lack o f secondary fibrinolysis in these patients. A fourth point is that the decrease in AT III levels was less pronounced than the protein C depression. That might be due to the longer half-life of AT III. However, many previously published studies have shown that during septic D I C there is usually an early, quite severe drop in AT III levels [2]. In such cases, it is mandatory to infuse high doses of AT III concentrates for 2 days at least in order to return AT III levels towards the normal range. Moreover, in cases of severe liver failure AT III levels remain low despite the infusion [18]. In our five patients with PF, AT III levels were completely restored in 24 h by infusion. By this time, protein S levels had increased rapidly, whereas protein C levels remained low. These data are consistent with the theory o f a particular imbalance of the anticoagulant systems during P E

Lastly, it is well documented that AT III can exert its antithrombotic property even after thrombin has bound to thrombomodulin and after activation of protein C. That infusion o f high doses of AT III could compensate for the acute decrease in the other anticoagulant system, the protein C system, during meningococcemic purpura is a possibility that must be considered.

References 1. Powars DR, Rogers ZR, Patch MJ, McGehee WG, Francis RB (1987) Purpura fulminans in meningococcemia:association with acquired deficiencies of proteins C and S. N Engl J Med 317:571-572 2. Bick RL (1982) Clinical relevance of antithrombin III. Semin Thromb Hemost 8:276-287 3. Stenbjerg S, Pedersen EG, Laursen H (1983) Coagulation fibrinolytic and antithrombin III profiles in severely infected patients. Thromb Res 31:635-640 4. EmersonTE, Fournel MA, LeachWJ, Redens TB (1987) Protection against disseminated intravascular coagulation and death by antithrombin III in the E. coli endotoxemicrate. Circ Shock 21:1- 13 5. Hauptmann JG, Hassouna HI, Bell TG, Penner JA, Emerson TE (1988) Efficacy of antithrombin III in endotoxin induced disseminated intravascular coagulation. Circ Shock 25:111- 122 6. Blauhut B, Kramar H, Vinazzer H, Bergmann H (1985) Substitution of antithrombin III in shock and disseminated intravascular coagulation. Thromb Res 39:81-89 7. OdegardOR, Lie M, AbildgaardU (1975) Heparin cofactor activity measured with an amydolyticmethod. Thromb Res 6:287-294 8. Nicham F, Ghichaoua JF, Contant G, Martinoli JL (1988) Dosage rapide de l'activit6 de la prot6ine C. Ann Biol Clin (Paris) 46:805- 808 9. LaurellOB (1966) Quantitative estimation of protein by electrophoresis in agarose gel containing antibodies. Ann Biochem 15:45-52 10. Comp PC, Doray D, Patton D, Esmon CT (1986) An abnormal plasma distribution of protein S occurs in functional protein S deficiency. Blood 67:504-608 11. Clauss A (1957) Gerinnungsphysiologische Schnellmethode zur Bestimmung des Fibrinogens. Acta Heamatol 17:237-247 12. Largo R, Heller V, Straub PW (1976) Detection of soluble intermediates of the fibrinogen-fibrin conversionusing erythrocytescoated with fibrin monomers. Blood 47:991-10002 13. Griffin GH, Moscher DF, Zimmermann TS, Kleis AJ (1982) Protein C: an antithrombotic protein is reduced in hospitalized patients with intravascular coagulation. Blood 60:261-264 14. MannucciPM, Vigano S (1982) Deficiencyof protein C, an inhibitor of blood coagulation. Lancet II:463-467 15. Clouse LH, Comp PC (1986) The regulation of homeostasis. The protein C system. N Engl J Med 314.'1298-1304 16. Marlar RA, Endres-BrooksJ, Miller C (1985) Serial studies of protein C and its plasma inhibitor in patients with disseminated intravascular coagulation. Blood 66:59-63 17. Vandermeer FJ, Broekmans AW, Bertina RM, Briet B (1985) Protein C activity and antigen levels in patients with intravascular coagulation. Scand J Clin Lab Invest 45:133 18. Fourrier F, Lestavel P, LeclercL, Chopin C, Mare3,A, Goudemand J (1987) Traitement des coagulations intravasculaires diss6min6es septiques par l'antithrombine III. Reanim Soins Intens Med Urgence 3:281 (abstr) Dr. E Fourrier Service de R6animation Polyvalente, H6pital B C.H.R.U. F-59037 Lille Cedex France