Changes in the blood coagulation system associated with septicemia following open-heart surgery- (a review and case report)SV Purandare, AP Chaukar, GB Parulkar, SR Panday
Department of Cardiovascular and Thoracic Surgery, K.E.M. Hospital and Seth G. S. Medical College, Parel, Bombay-400 012., India
Correspondence Address: Source of Support: None, Conflict of Interest: None PMID: 615270
Source of Support: None, Conflict of Interest: None
A rare case of diffuse intravascular clotting (DIC) complicating Gram negative bacterial endocarditis following aortic valve replacement is reported. The clinical and pathological findings are presented. The blood coagulation studies are described in detail and which indicated extensive in vivo clotting. Histological examination of the kidneys however, failed to reveal the presence of fibrin. The literature on the subject of DIC following septicemia is reviewed. The value of the use of heparin in such patients is discussed.
Bacterial endocarditis following open heart surgery is not uncommon. ,, A number of reports over the last decade have emphasized the sudden changes that can occur in the blood clotting mechanism in patients with systemic bacterial infection. The most common observation is probably thrombocytopenia, but a severe alteration in the hemostatic mechanism may result from diffuse intravascular clotting (DIC). ,, Although alterations in the clotting sensitive factors are observed, all the patients may not exhibit a hemorrhagic tendency. Review of the available literature revealed only isolated instances of consumption coagulopathy (CC) in association with either fungal or bacterial endocarditis. , The purpose of this paper is to describe a case of CC associated with bacterial endocarditis.
S.S.M.a male aged 32 years. was admitted to the Cardiovascular & Thoracic Centre of the K.E.M. Hospital for complaints of exertional dyspnea of Grade III and recurrent respiratory tract infection for the past three years. There was no history of jaundice, congestive heart failure or cyanosis in the past. On examination, he was averagely built and nourished and had a barely palpable liver. The spleen was not palpable. A diagnosis of aortic valve incompetence with mitral stenosis was made on the basis of clinical examination and confirmed on cardiac catheterization.
The patient underwent aortic valve replacement using a 10A Starr-Edward prosthesis and a mitral commissurotomy under cardiopulmonary bypass. A disposable Harvey oxygenator was used for the bypass. The extracorporeal circuit was primed with 5% Dextrose. Heparin (3 mg/kg body weight) was administered to the patient and also added to the priming solution (5 mg/100 ml of priming solution). As a routine measure epsilon aminocaproic acid was given prophylactically (150 mg/kg body weight, intravenous) before the chest was opened. The bypass time was 100 minutes. At the end of 60 minutes of bypass, additional heparin (half the initial dose) was given. Heparin was neutralized at the end of perfusion with a calculated dose of protamine. The total drainage was 300 ml when the chest drainage tubes were removed. He received Gentamicin and Ampicillin in the immediate postoperative period. Warfarin was started on the third postoperative day. The patient developed high fever on the fifth postoperative day and it continued for two more days. In addition to Gentamicin and Ampicillin, Cloxacillin (500 mg every six hours. intramuscular) was started. Blood was examined for bacteriological studies. On the tenth day after the operation all the antibiotics were stopped. The fever persisted. The spleen at this stage became just palpable. Trimethoprim and sulphamethoxazole combination (2 tablets twice a day) and injection Kanamycin (200 mg every 6 hours. intramuscular) were started. Warfarin was omitted as the prothrombin time rose to 70 seconds with a control of 15 seconds. On the nineteenth post-operative day, petechial hemorrhages appeared all over the body. The patient was toxic and restless not responding to verbal commands. He stopped passing urine. A coagulogram indicated a diagnosis of DIC secondary to infection. Intravenous heparin was advised but the patient expired before any treatment could be started. Autopsy was performed.
Coagulation studies consisted of (i) Clotting Time (CT), (ii) Bleeding Time (BT), (iii) Platelet count, (iv) Clot Retraction (qualitative), (v) Prothrombin Time (PT). (vi) Partial Thromboplastin Time with Kaolin (PTTK). (vii) Euglobulin lysis time (ELT). (viii) Thrombin Time (TT) and (ix) Fibrinogen level. Standard methods were employed.  A hemogram was also performed.
Coagulation studies done at different intervals are summarized in [Table 1]. A polymorphonuclear leucocytosis was present.
Bacteriological findings, Blood culture and material taken from autopsy showed the presence of Pseudomonas aerugenosa.
Autopsy Findings: The heart showed multiple patches of hemorrhages mainly on the septal surface. A few petechial hemorrhages were seen on the outflow tract. Large polypoid vegetation was seen covering the orifice of aortic valve prosthesis on the ventricular side, causing 30% blocking of the orifice. On the aortic side, small nodular vegetations were seen covering the ball and the intima of the ascending aorta. One of the sutures in the ascending aorta appeared loosened and a small embolus was seen in the abdominal aorta near the origin of both the renal arteries.
Both the lungs, on cut surface, showed multiple septic infarcts. The infarcts were friable, red brown and studded with occasional abscess cavities.
Multiple patches of subarachnoid hemorrhage were present over the cerebrum and cerebellum. They were more extensive on the cerebrum. The kidneys presented a flea beaten appearance with diffuse extensive petechial hemorrhages and multiple tiny pinkish yellow dots of early abscess on the external surface. The vegetation showed a fibrous thrombus with predominantly neutrophilic exudate on periphery. No organism was seen. Special stain for fungus yielded negative result. No active inflammation was seen.
With advances in techniques in the last decade, the operative mortality of patients undergoing open heart surgery is considerably reduced. However, infection following surgery still continues to take its toll. One of the secondary complications that arises following infection is the onset of a bleeding diathesis which in some cases may prove fatal.
The abnormalities which may occur in the hemostatic mechanism during severe bacterial infection extends from isolated thrombocytopenia to DIC. It is now generally recognized that in DIC, the platelets and the clotting sensitive factors V, VIII, II and fibrinogen are characteristically low and that fibrinolytic split products can be found in the serum, thus accounting for the abnormal screening tests.
The best known experimental model of DIC is the generalized Shwartzman reaction (GSR). The lesion which identifies the reaction-bilateral cortical necrosis of the kidneys-is secondary to an occlusion of the glomerular capillaries by fibrin thrombi. Although the renal lesion is easily elicited in the experimental model, it is extremely rare in clinical cases of Gram negative septicemia.  This may be related to the problem of proper dosage and timing of endotoxin which can be controlled in the laboratory animal. We failed to demonstrate fibrin in the kidneys of the patient at autopsy. Preston et al  also failed to detect fibrin in the kidneys of three patients. Clarksor et al  have suggested the use of more refined techniques, other than light microscopy, for the detection of fibrin in acute renal failure.
In the case reported, a combination of thrombocytopenia, a prolonged PT (deficiency of vitamin K dependant factors), raised PTTK (reflecting decreased factors V and VIII) and prolongation of TT (indicative of fibrin split products in the circulation) was observed. A positive blood culture was obtained. The findings suggest that the patient had suffered enough toxemia to trigger DIC. A normal ELT ruled out the presence of hyperfibrinolysis. Rapaport et al  reporting a case of Gram negative septicemia with intravascular clotting also failed to detect increased fibrinolytic activity. Very few cases of bacterial endocarditis associated with CC appear in the literature. Doughten and Pearson  have described a case of Aspergillus endocarditis in association with DIC. Mahvi et al  also reported a case of Aspergillus endocarditis following open heart surgery. Both the patients received heparin and subsequently died of intracranial hemorrhage. However, no coagulation studies have been described by the latter authors.
Thrombocytopenia may result from at least two effects of bacteremia: (i) direct damage to the platelets by the endotoxin or bacteria. This reaction is neither blocked by heparin nor by warfarin.  (ii) Intravascular coagulation with resultant destruction of platelets by thrombin.
A number of investigators have noted that DIC and septic shock occur together, suggesting a causal relationship. , Furthermore, a variety of infectious agents are reported to have this combination. Thus the common denominator appears to be shock. ,,, However, whether the coagulation defect causes the shock or vice versa, is not yet clearly understood.
Experience indicates that anticoagulants are not always necessary.  Often no specific treatment other than supportive measures is needed, when the ,intravascular clotting is so minor and shortlived that it does not contribute to either morbidity or mortality. However, in patients who have a serious bleeding problem and where one cannot wait for complete correction of the primary cause (as in purpura fulminans), heparin therapy should be immediately started. Once the echymotic patches become gangrenous or the patient goes into irreversible shock, therapy of any type is usually ineffective . , An intravenous dose of heparin (100-150 µ/ kg body weight) is usually given every six to eight hours Fresh blood or plasma is simultaneously administered. Beneficial results of the use of heparin in the treatment of DIC following infection appear in the literature. ,,. is We could not ascertain the effect of heparin in the case reported since the patient expired before any treatment could be started.
The usefulness of heparin in the treatment of shock obviously constitutes only one part of the management of the patient. Other measures, as clinically indicated, to restore vascular volume, to correct acidosis and to improve cardiac function must also be undertaken if these desperately ill patients are to survive. Moreover, heparin therapy is continued as long as it is useful in suppressing bleeding. This can be judged by stopping heparin and judging the patient's response. If the hemostatic abnormalities recur, one should seriously think of reinstating heparin regimen.
Finally, the question of using antifibrinolytic agents in a bleeding patient is illunderstood. Primary fibrinolysis is rare and when present is usually secondary to DIC. , Hence one should desist from using these agents in such patients as it could lead to disastrous results. If at all used, it should be in combination with heparin.
Thanks are due to the Dean, K.E.M. Hospital and Seth G. S. Medical College, Bombay 400 012, for permitting us to report the hospital data.