Irving Schulman

Intravenous use of gammaglobulin in the treatment of chronic immune thrombocytopenic purpura as a means to defer splenectomy

Intravenous gammaglobulin was used to treat 12 children with chronic immune thrombocytopenic purpura in order to avoid splenectomy. The average rise in platelet count with initial treatment was 226,000/microliters. Currently, one patient is in remission, four patients maintain platelet counts greater than 40,000/microliters without treatment, four patients maintain platelet counts greater than 40,000/microliters with single maintenance infusions of IV IgG at four- or 10-week intervals; three patients did not respond to treatment. In nine of 12 patients, splenectomy was avoided or at least postponed. In responding patients, we were able to discontinue immunosuppressive medication. Platelet count rises with initial IV IgG therapy were correlated with both platelet antibody levels and with a better long-term outcome. Toxicity was minimal.

Evaluation of a walking-donor blood transfusion program in an intensive care nursery

A prospective study was carried out to identify the immediate and long-range advantages and disadvantages of a walking-donor transfusion program for an intensive care newborn nursery. The effect of heparin on coagulation of blood was evaluated and found to be minimal. There was no evidence of transmission of HBSAg. The prevalence of CMV infection at the time of follow-up was higher in infants who had received blood from donors seropositive for CMV than in infants who had been transfused from seronegative donors. In our experience, a walking-donor program has been a safe and effective method for the provision of small transfusions of blood to sick neonates.

Platelet-associated immunoglobulin G in childhood idiopathic thrombocytopenic purpura

Platelet-associated IgG was studied in children with acute and chronic ITP and in patients with thrombocytopenic SLE, using the microtiter solid-phase radioimmunoassay. Of the children with acute ITP, 85% had elevated PAIgG levels. The degree of elevation of PAIgG at onset of disease did not correlate with the development of chronicity. Of the children with acute ITP, clinically and hematologically indistinguishable from the rest, 15% had normal PAIgG values. All of 22 children with chronic ITP had elevated PAIgG values. Although there was good correlation between the platelet count and the PAIgG value in children with chronic ITP, the association was not as striking in those with acute ITP; thus, factors in addition to the level of PAIgG may contribute to the thrombocytopenia in the latter group. Patients with SLE and thrombocytopenia had higher values of PAIgG than would be predicted from the platelet count; the PAIgG value is probably not the only factor determining the degree of immune thrombocytopenia.

Quantitative Aspects of Blood Coagulation in the Generalized Shwartzman Reaction: 1. Effects of Variation of Preparative and Provocative Doses of E. Coli Endotoxin

Extract: The generalized Shwartzman reaction (GSR) is induced in rabbits by two properly spaced intravenous injections of bacterial endotoxin and is characterized by the occurrence of bilateral renal cortical necrosis. This report describes the quantitative changes in the coagulation mechanism in response to variation of preparative and provocative doses of E. coli endotoxin and of the interval between injections. It also correlates the coagulation changes with the pathologic findings of renal cortical necrosis.Renal cortical necrosis was induced in 34 of 49 (70 %) of rabbits using 0.100 mg/kg of endotoxin as the preparative (first) injection and 0.200 mg/kg 24 hours later as the provocative (second) dose. Four hours after the preparative injection white blood cells and platelets fell significantly while fibrinogen, prothrombin time, and Factors V and VIII fell slightly. At 24 hours all determinants except the platelets had recovered and the white cells, Factor V and fibrinogen had risen to significantly higher levels than baseline. Following the second (provocative) injection of endotoxin, a precipitous fall in all factors occurred. Maximal changes developed two hours after the second injection for white cells and platelets and by four hours for the coagulation factors; at this time kidneys show fibrin. By 48 hours all factors except the platelets had returned to 24 hour levels (table I). Rabbits given 0.100 mg/kg of endotoxin and normal saline 24 hours later did not develop cortical necrosis of kidneys or reduction in levels of any of the coagulation factors or platelets (table II).With 0.1 mg/kg of endotoxin as the preparative dose, the effect of variation in the provocative dose at 24 hours was studied. With 0.2, 0.1, 0.05 mg/kg provocation renal cortical necrosis was observed in 60-100 % of animals. Rabbits given 0.01 mg/kg of endotoxin, or saline, as the second injection did not develop the GSR. Coagulation changes were measured four hours after the provocative injections. With doses of endotoxin of 0.025 mg/kg and greater there was a significant fall in white cells, platelets and each of the coagulation factors measured. With 0.01 mg/kg there was a significant decrease in Factors II, V and VIII but no significant fall in white cells, platelets or fibrinogen. With saline all of the measured factors rose (table V). Changes in fibrinogen following different provocative doses are demonstrated in fig. 5. Renal cortical necrosis did not occur until a fall in fibrinogen of 66 mg/100 ml was produced.The effect of varying the preparative dose was studied. Groups of rabbits were given either saline, 0.001 mg/kg, 0.01 mg/kg or 0.1 mg/kg endotoxin preparation and all animals received 0.1 mg/kg endotoxin 24 hours later. With saline or 0.001 mg/kg endotoxin as preparation no GSR occurred. With a 10 and 100 fold increase in concentration of endotoxin the GSR was regularly produced. With saline preparation significant changes occurred only in white cells and platelets. A significant decline in all factors except Factor VIII was produced by preparation with 0.001 mg/kg endotoxin; the fibrinogen fell 89 mg/100 ml but no animal developed renal cortical necrosis. The inadequately prepared animal appeared able to tolerate an amount of fibrinogen conversion that would, in the properly prepared animal, result in renal cortical necrosis. In rabbits prepared with 0.01 mg/kg and 0.1 mg/kg of endotoxin significant changes occurred in all factors and renal cortical necrosis was produced (table VI). Rabbits were given 0.1 mg/kg endotoxin as preparation and a dose of 0.1 mg/kg was administered at 24 hours to one group of animals and at 48 hours in a second group. In the 24 hour group the GSR occurred in 10 of 13 animals; following endotoxin at 48 hours renal cortical necrosis was not produced in any of 18 rabbits. Changes in the coagulation mechanism four hours after provocation are shown in table VII and demonstrate that the fall was similar in both groups for all factors except fibrinogen. Fibrinogen decreased 178 mg/100 ml in animals given injections 24 hours apart but there was only a 70 mg/100 ml fall in the 48 hour group. The data suggest that by 48 hours after adequate preparation the RES has recovered to the degree that clotting intermediates are removed and significant amounts of fibrin are cleared thus preventing deposition in the kidneys and cortical necrosis.The interdependence of preparative and provocative doses of endotoxin was demonstrated. A dose of 0.01 mg/kg prepared the animal for subsequent provocation with 0.1 mg/kg; 0.1 mg/kg prepared animals for subsequent provocation with as little as 0.025 mg/kg. However, the combination of 0.01 mg/kg as preparation and 0.025 mg/kg as provocation did not result in cortical necrosis of the kidneys or in a fall of any of the coagulation factors (tables VIII and IX).Conclusions: In the presence of adequate preparation provocative doses of endotoxin capable of inducing cortical necrosis of the kidneys were always associated with significant consumption of all coagulation factors measured. With inadequate preparation, produced either by decreasing the preparative dose of endotoxin or by increasing the interval between injections, significant activation of the clotting system occurred but renal cortical necrosis did not develop.Speculation: The development of the generalized Shwartzman reaction, and, by inference, Shwartzman-like human syndromes, appears to require activation of the coagulation mechanism and the development of intravascular clotting. However, the degree of clotting induced by stimuli of known potency and the degree of fibrin deposition in the tissues are critically dependent upon the prior state of preparation of the animals, probably reflecting the functional adequacy of the reticulo-endothelial system.

Quantitative aspects of blood coagulation in the generalized Shwartzman reaction. II. Effect of cortisone.

Extract: In the previous paper quantitative changes in the coagulation mechanism accompanying each of the two injections of endotoxin required for the production of the generalized Shwartzman reaction (GSR) were described. It was demonstrated that the development of cortical necrosis of the kidneys was dependent both on the induction of intravascular clotting and on the state of preparation of the rabbits in response to the first dose of endotoxin. Conversely, the adequacy of preparation could be assessed by the response to a standard dose of endotoxin measured by the degree of intravascular clotting induced and the incidence of cortical necrosis of the kidneys.In 1952 THOMAS and GOOD reported that the GSR could be produced by a single injection of Serratia marcesens or meningococcus endotoxin if rabbits were pre-treated with ACTH or cortisone. It was the purpose of the present investigations to compare the degree of preparation induced by cortisone and by Thorotrast with that induced by E. coli endotoxin employing a provocative dose of endotoxin which had heen shown capable of inducing intravascular clotting and cortical necrosis of the kidneys in adequately prepared rabbits.Rabbits were pre-treated with cortisone 25 mg IM for four days and given a single dose of endotoxin on the third day according to the regimen of THOMAS and GOOD. With provocation using the standard dose of endotoxin (0.100 mg/kg), renal cortical necrosis was not observed in any of 20 animals. In contrast this same provocative dose induced renal cortical necrosis in 74 and 77 % of animals prepared with 0.01 and 0.1 mg/kg of endotoxin. With the same regimen of cortisone preparation provocative doses of endotoxin of 0.2 and 0.5 mg/kg likewise failed to induce cortical necrosis. With a provocative dose of 1.0 mg/kg two of fifteen animals developed cortical necrosis of the kidneys. Five animals receiving 2.0 mg/kg of endotoxin exhibited no GSR. With cortisone preparation of I00 and 250 mg/day for four days provocation with 1.0 mg/kg of endotoxin resulted in cortical necrosis in only 1 of 6 animals in each group (table I). Coagulation studies were carried out in animals prepared with cortisone, 25 mg/day for four days, and provoked with the standard dose of endotoxin of 0.1 mg/kg. Significant decreases were noted in white cells, platelets, Factors II and VIII, but no significant changes occurred in Factor V or fibrinogen concentrations (table II). These changes are almost identical to those found with saline preparation.Thorotrast was given to eleven rabbits in a dose of 3.0 ml/kg intravenously and eighteen hours later 0.1 ml/kg of endotoxin was given. Five of the animals developed cortical necrosis of the kidneys. With Thorotrast preparation the coagulation mechanism was activated in its entirety, with highly significant falls in all of the measured coagulation factors observed (table III).Conclusions: Cortisone preparation, even in doses ten times those employed by THOMAS and GOOD did not lead to development of intravascular clotting or cortical necrosis of the kidneys following a single injection of E. coli endotoxin. This held true even when the provocative dose of endotoxin was raised to ten times the standard dose or forty times the minimum effective dose with endotoxin preparation. Thorotrast preparation, on the other hand, was followed by intravascular clotting and cortical necrosis of the kidneys as response to a single injection of endotoxin in the standard amount.Speculation: The ability to prepare animals for intravascular clotting and the development of the generalized Shwartzman phenomenon does not seem to be an inherent property of cortisone but appears to be dependent upon the type of endotoxin used in provocation. The implications that others have drawn from the data of THOMAS and GOOD that the use of cortisone may be hazardous in treatment of Shwartzman-like clinical conditions in the human appear to be premature in the present state of knowledge.

Increased skin permeability in preterm infants

Localized cutaneous blanching of preterm neonates following the topical application of a 10% solution of Neo-Synephrine attests to the permeability of immature skin. Skin permeability was evaluated in 18 healthy infants between 30 to 40 weeks of gestational age. The response consisted of speckles or islands of blanching which gradually coalesced to form a homogeneous pale area. Infants 30–35 weeks had a very rapid response which persisted from 30 minutes to 8 hours. A response was no longer apparent by 21 days of postnatal life. Infants 36–37 weeks had a longer lag period and responded less intensively. Infants 38–42 weeks failed to exhibit blanching in most cases. The degree of permeability correlated inversely with gestational age. Surface temperatures on strongly blanched and contiguous unblanched skin were identical when measured simultaneously. Toxicologic implications are clear. Other considerations include the intrauterine role of fetal skin as a dialyzing membrane, the relationship of skin structure to barrier function and the potential of further pharmacologic studies on the accessible cutaneous circulation.

Functional Studies of Toung Versus Old Platelets in a Patient with Chronic Thrombocytopenia

The reprdoucible platelet production cycle following infusion of fresh frozen plasma (FFP) in a girl with chronic thrombocytopenia [Blood 16: 943, 1960] provided a unique opportunity to study the functional capacity of young and old platelets. Study cycles were initiated by infusing FFP when the patients platelet count was less than 20,000/mm3. ‘Young’ platelet studies were done 4 days after FFP when the patients platelet count was less than 20,000/mm3); ‘old’ platelet studies were done 21 days after FFP (average platelet count, 200,000/mm3). All studies were repeated during several such cycles. In association with a young platelet population, were normal Ivy bleeding times, normal or increased platelet adhesiveness, normal aggregation to ADP and collagen, and normal platelet factor 3 availability (PF-3a). By contrast, when the patients circulating platelets were old, she was found to have long bleeding times, abnormally low adhesiveness in vivo and in vitro, and decreased PF-3a. Aggregation to ADP and collgen was slightly decreased but remained within normal limits. No abnormalities were seen by electron microscopy on platelet samples obtained throughout the cycle. (Performed by Dr. JAMES WHITE, Univ. of Minn.) It is of interest that on several occasions the patient experienced mild bleeding manifestations at the end of a cycle but while her platelet count was still normal. These episodes correlated with the findings of long bleeding time and decreased adhesiveness.