- Email: [email protected]
Endotoxin, prekallikrein, complement and systemic vascular resistance
Endotoxin, Prekallikrein, Complement and Systemic Vascular Resistance Sequential Measurements in Man
JOHN A. ROBINSON.
M.D.
MARY L. KLODNYCKY,
M.D.
HENRY S. LOEB, M.D. MARY R. RACE,
B.S.
ROLF M. GUNNAR.
M.D.
Hines, Illinois Maywood, Illinois
From the Immunology and Cardiology Research Sections, Veterans Administration Hospital, Hines, Illinois; and Department of Medicine, Loyola University Strltch School of Medicine, Maywood, Illinois. This study was presented in part at the Annual Meeting of the American Federation for Clinical Research, Southern Section, January 24, 1974. Requests for reprints should be addressed to Dr. John A. Robinson, Department of Medicine, Loyola Unlverslty Strltch School of Medicine, Maywood, Illinois 60153. Manuscript accepted August 28, 1974.
Eighteen patients were studied prior to and again within 6 hours after transurethrai resection or cystoscopy. in addition to hemodynamic measurements, detection of endotoxin by iimuius assay and bacterioiogic sampling; prekaiiikrein, C3, C3 proactivator and iysosomai enzyme levels were measured. in five patients iimuius assays were positive, and in one, gram-positive bacteremia developed but iimuius assay remained negative. Ail six had significant decreases in prekaiiikrein, C3 or C3 proactivator. Systemic vascular resistance fell in ail six. Four additional patients who had a decrease in systemic vascular reststance were not endotoxemic or bacteremic; one of these had a decrease in prekaiiikrein only. in the remaining eight patients with neither bacteremia nor endotoxemia, systemk vascular resistance did not change or increase after instrumentation. One had a decrease in C3 proactivator, another in prekaiiikrein. There was no significant difference in age, disease, antibiotk therapy or bacteremia in the two groups of patients. Four of the five resectional procedures were performed in the group that showed decreases in systemk vascular resistance. The data suggest that acute endotoxemia or gram-positive bacteremia in man is associated with depletion of prekaiiikrein, decreased peripheral resistance and, in some instances, activation of the complement system. Although endotoxemia has been implicated as a consistent cause of acute hemodynamic effects in lower primates [ 11, hemodynamic changes during relatively uncomplicated endotoxemia in man are not known. The patient undergoing urologic procedures, especially transurethral resection, is at high risk for the development of endotoxemia [2] with or without bacteremia [3]. This type of patient provides a model for the longitudinal study of hemodynamic changes during endotoxemia and further affords a means to delineate significant changes in or the appearance of vasoactive mediators during endotoxemia in man. Melmon et al. [4] have shown in monkeys that bradykinin, a potent hypotensive peptide, appears early in endotoxemia and is associated with a decrease in peripheral vascular resistance. It has also been shown that infusion of bradykinin into the occluded forearm of man will decrease vascular re-
July 1975
The American Journal of Medfclm
Volume 59
61
ENDOTOXIN AND SYSTEMIC VASCULAR RESISTANCE-ROBINSON
ET AL.
sistance [6]. Similar changes in the kallikrein-kinin system might be implicated in a patient with endotoxemia. However, appearance and elevated levels of bradykinin are short-lived and modulated by several mechanisms. Therefore, it seems apparent that the sequential appearance of other vasoactive mediators, possibly initiated by endotoxin, are necessary to explain continuation of the hemodynamic changes noted in patients with gram-negative infection [6]. A proposed mechanism for continued vasodilatation during endotoxemia is direct or bypass activation of complement, especially the third (C3a) and fifth (C5a) complement components which have anaphylatoxic activity and could explain hemodynamic changes of more than transient character during gram-negative infection. It is hoped that identification of significant early physiologic changes will provide a more rational basis for therapeutic interventions in the bacteremic shock syndrome. However, the need to observe and document changes in vasoactive mediators during the transition from immediate vasodilatation to the more prolonged hemodynamic effects of endotoxemia in man became apparent when comparing hemodynamic studies during early bacteremia with similar studies in the established shock state [7]. The development of the limulus test, a highly sensitive assay for endotoxin [8], has provided a means for detecting endotoxin or endotoxin-like material in the absence of detectable gram-negative bacteremia. Nonbacteremic endotoxemia detected by the limulus assay could explain some of the hemodynamic changes seen in abacteremic patients after genitourinary manipulation and could also increase the yield of patients with endotoxemia in our study. The limulus assay and determinations of prekallikrein, the activated esterase that cleaves kinin from its precursor, kallikrein, and selected complement components were carried out in parallel with hemodynamic assessments preoperatively in 18 patients undergoing cystoscopy or transurethral resection. These data were then compared to subsequent similar determinations in hemodynamics during the immediate 6 hour postoperative period. The data suggest that initial acute endotoxemia or gram-positive bacteremia in man is associated with a depletion of prekallikrein, a decrease in peripheral resistance and, in some instances,
activation
MATERIALS
of the complement
had overt hepatic disease or significant postoperative blood loss. Some patients with gram-negative bacteriuria were being treated with suifathiazides or ampicillin prior to the procedure. in no patient did the septic shock syndrome develop in the postoperative period. Hemodynamic studies were conducted in a shock treatment unit. Central venous pressure was measured with Statham P23 Db transducer through a PE-160 catheter in the superior vena cava. Arterial pressure was measured with a Statham P23 Db transducer through a disposable Cournand needle in the femoral artery. Cardiac output was measured by the indicatordilution method using the Stewart-Hamilton formula. indocyanine green was injected into the superior vena cava and sampled from the femoral artery as blood was drawn through a Waters densitometer. Data for analysis represent the average of values from two to five indicator dilution curves made in rapid succession. Systemic vascular resistance (SVR) was calculated as follows: SVR mean aortic pressure (mm Hg) - central venous pressure (mm Hg) cardiac output (iiter/min) Preoperative and 2 and 6 hour postoperative venous samples were collected in plastic, pyrogen-free syringes for bacteriologic cultures, determination of coagulation factors and assays of endotoxin, prekaliikrein, spontaneous esterase activity and the complement system. Fibrin split products were quantitated by staphyiococcai clumping [9]. Detection of endotoxin was by the method of Levin et al. [a]. One pooled iysate was used for ail tests and consistently detected 0.5 ng endotoxin per O.l/mi heparinized plasma using Salmonella typhosa iipopoiysaccharide as standard. A test was positive when either complete geiation or a definite increase in viscosity was present at 24 hours. Test results were interpreted by two observers and, in one instance, confirmed by a lead sensitized rat bioassay that will detect 0.1 pg endotoxin/mg plasma (Dr. J. Filkins). Attempts to detect circulating antigen-antibody complexes were made by diffusion of Clq (first component of compiement) against whole serum in agar gels. Plasma ribonuclease activity was tested for by an agar diffusion technic sensitive to 4 X 10v5 kg/ml ribonuclease [lo]. Determination of kaiiikrein, prekaliikrein and kailikrein inhibitor was by the calorimetric quantitation of methanol generated by titrated plasma esterase activity with tosyi arginine methyl ester (Sigma Chemical) as substrate before and after the addition of kaolin [ 111. Reproducibility is within 4 pg methanol/hour/sample. Specificity of spontaneous esterase activity was assayed using either benzoyi arginine methyl ester, tosyl arginine methyl ester or lysylarginine methyl ester for determination of optimal esteraselsubstrate ratios and confirmation of kaliikrein esterase activity [ 121. Cathepsin D was determined by a I-nitroso-P-naphthoi fluorescence assay after incubation of plasma with 2 X lob4 M hemoglobin at pH 3.5 [ 131. Serum C3 was determined by quantitative immunodiffusion (Hyland Laboratories). Quantitation of C3 activator (C3PA) was by the Mancini technic in agar plates containing monospecific C3PA and antiserums (Hoechst Laboratories). Clq, C4 and properdin
system.
AND METHODS
Eighteen men, aged 50 to 86 years, were studied within 16 hours of and during the immediate 6 hour interval after cystoscopy or transurethral resection under local or spinal anesthesia. No patient was being treated with vasoactive drugs or was in congestive heart failure, and no patient
62
July 1975
The American
Journal
of Medicine
Volume
59
ENDOTOXIN AND SYSTEMIC
TABLE
I
Characterization
of Patients
by Age, Disease,
Age W
Case No.
Procedure
Disease
I 5 7 9 11 12
78 75 66 54 86 70
Benign prostatic hypertrophy Metastatic carcinoma Benign prostatic hypertrophy Renal mass Carcinoma of prostate Carcinoma of bladder
13
83
Benign
14
73
Neurogenic
15
60
Benign
prostatic
hypertrophy
16
76
Benign
prostatic
hypertrophy
2 3 4 6 8 10 17
60 50 75 53 54 53 59
Benign prostatic hypertrophy Nephrolithiasis Benign prostatic hypertrophy Benign prostatic hypertrophy Benign prostatic hypertrophy Carcinoma of prostate Carcinoma of prostate
18
75
prostatic
* immediately TABLE _--
II --
-__
Benign
Anbhotics+
A
hypertrophy
bladder
prostatic
ET AL.
Prior to Study
None None Sulfathiazide None None None
Cystoscopy Cystoscopy Cystoscopy Cystoscopy Cystoscopy Fulguration and cystoscopy Transurethral resection Cystoscopy
None Ampicillin, sulfathiazide Tetracycline
Transurethral resection Transurethral resection
Group
______
Therapy
RESISTANCE--ROBINSON
Procedure Group
___
and Antibiotic
VASCULAR
Sulfathiazide
B Cystoscopy Cystoscopy cystoscopy Cystoscopy Cystoscopy Cystoscopy Cystoscopy biopsy Cystoscopy
hypertrophy
None Tetracycline None Tetracycline None Erythromycin None
and
Ampicillin
prior to procedure. Preoperative Genitourinary co
(Rest) and Postoperative Procedures
(Postop) Hemodynamic _.~~
in Patients
Undergoing
_..~~
MAP
--
Measurements
~__
Case No.
Rest
Postop
Rest
Postop
1 5 7 9 11 12 13 14 15 16
1.4 2.1 4.6 4.2 4.6 4.6 3.3 5.1 3.7 5.8
2.0 2.1 3.6 4.7 4.5 7.1 4.8 3.8/‘4.9* 4.2 3.5
90 90 110 106 90 80 130 121 122 138
80 70 84 84 68 97 120 116/118* 124 60
2 3 4 6 8 10 17 18
2.8 2.2 3.7 5.5 7.4 6.9 9.7 4.8
2.1 2.6 1.6 3.8 5.6 5.7 5.7 6.3
90 106 120 110 80 78 104 76
80 90 112 84 82 78 78 104
SVR
CVP
Rest
Group A 0
-1 5 3 3 3 12 6 7 2
0 -3 3 -1 5 11 10 4/3* 10 1
29.6 22.0 32.6 24.6 18.9 16.7 35.8 22.5 31.1 25.7
18.4 18 22.5 18.0 14 12.1 22.9 29.5/23.6* 25.7 14.0
Group B 0 1 0 3 3 -7 5 7
14.5 24.0 20.0 14.9 14.0 10.1 9.5 14.8
20.0 32.0 32.0 22.0 18.4 14.2 17.8 15.4
NOTE: CO = cardiac output (mm Hg/liter/min). MAP = mean arterial pressure (mm Hg). CVP = central venous pressure (mm HzO). SVR = systemic vascular resistance (mm Hg/liter/min). * Values to left of slash were obtained 1 hour after the procedure; values to right of slash obtained 2 hours after the procedure.
July 1975
The American Journal of Medicine
Volume 59
63
ENDOTOXIN AND SYSTEMIC VASCULAR
LIMULUS
RESISTANCE-ROBINSON
ET AL.
ASSAY
BACTEREMIA
Figure 1. Directional changes in systemic vascuiaf resistance in patients 2 hours after genitourinary instrumentation. The K within bars indicates a 70 per cent or greater decrease in prekallikrein, a C indicates either a 25 per cent decrease in C3 or C3PA, or both. The immediate postoperative directional change in systemic vascular resistance of patient 8, group A, is indicated by a broken line bar; his 2 hour postoperative value is expressed with similar determinations in patients in group A.
-20
were kindly determined by Dr. H. Gewurz. Hemodynamic and bacteriologic results were not known by the laboratory prior to an assay. Statistical analysis was made using the Chi square analysis and Yates correction factor. Hemodynamic measurements were’ made on the day before or the morning of genitourinary surgery, and venous blood samples for analysis were drawn at the same time. Unless the patient was to have surgery within the next 2 hours, catheters were withdrawn and then reinserted immediately postoperatively. Hemodynamic measurements were repeated twice between 1 and 6 hours postoperatively.
than 10 per cent decrease in systemic vascular resistance when the preoperative values were compared with the 2 hour postoperative measurements (group A) and those who showed no change or an increase in vascular resistance (group 6). The 10 patients in group A (Table I) had an average age of 72 years, and the 8 patients in group B (Table I) had an average age of 60 years. There was no difference in the type of medication, anesthesia or pathology between groups A and B. However, four of the five patients who had a resection, fulguration or biopsy procedure were in group A. There were no significant differences in fluid volume infusions or duration of operation between these patients and others in group A. greater
RESULTS
On the basis of hemodynamic were
divided
TABLE III
into two
groups:
Preoperative Complement
studies, the patients those who showed a
(Preop) and 2 Hour Postoperative (Postop) Clinical Microbiologic, Limulus Assay, Kallikrein and Component Values in Patients Who Had a Postoperative Decrease in Systemic Vascular Resistance
Kallikrein System Bacteriology Case
Urine*
No.
(ww)
1 5 7 9 11 12 13 14 15 16
Proteus Escherichia coli Not determined Not determined Bacteria by urinalysis Escherichia coli Pseudomonas Gram-negative rods (104) Gram-negative rods (lOa) Escherichia coli
Bloodt Preop Postop
July 1975
Spontaneous Esterase Activity
Preop Postop Preop
-
-
-
+ -
7.4 0 0 3.1
-
-
-
+
0
-
-
-
+ + +
-
-
-
Staphylococcus aureus
NOTE: Kallikrein system: normal values = spontaneous and over) rmol methanol/hour. Complement system: * More than 106 colonies/ml unless noted. t A minimum of three cultures.
64
Limulus Assay
The American Journal of Medlclne
0
0 0 3.0 0
Complement System Prekallikrein
c3
C3PA
Postop
Preop
Postop
Preop
Postop
Preop
Postop
21.7 1.8 0 1.2 2.5 0 0 1.8 1.8 0
101 92 94 77 89 74 94 73 72 77
74 101 84 74 69 78 a4 59 47 60
140 a4 110 90 145 150 132 155 145 68
120 94 110 72 127 180 145 110 145 42
410 150 260 105 300 170 380 220 320 140
430 130 300 180 300 200 190 160 160 100
activity 0 to 3 rmol methanol/hour; prekalli krein 70 to 100 (ages 50 years normal values = C3 120 to 180 mg/lOO ml; CSPA 180 to 320 fig/100 ml.
Volume SS
ENDOTOXIN
TABLE IV
_._~._
_
Resistance
in the Postoperative
3 4 6 8 10 17 18
RESISTANCE-ROBINSON
ET AL.
Period
Bacteriology
2
VASCULAR
Preoperative (Preop) and 2 Hour Postoperative (Postop) Clinical Microbiologic, Limulus Assay, Kallikrein and Complement Component Values in Patients with No Change or an Increase in Systemic Vascular
Kallikrein System __._.~__
-..--.--..
Case No.
AND SYSTEMIC
Preop Postop
(preop) Escherichia
Limulus Assay
Bloodt
Urine*
coli
Gram-negative (103) Not determined Escherichia coli Not determined Negative Escherichia coli Not determined ___
Spontaneous Esterase Activity
Preop Postop Preop
Postop
-..__
_ Complement
Prekallikrein Preop
Postop
-
-
-
-
0
1.9
57
63
-
-
-
-
0 0 1.9 0 3.1 0 0
3.7 1.8 0 3.1 3.1 0 0
98 109 70 85 102 94 56
97 112 80 85 112 82 60
-____ NOTE: Kallikrein system: normal values = spontaneous activity 0 to 3 pm methanol/hour; and over) pm methanol/hour. Complement system: normal values = C3 120 to 180 mg/lOO _ * More than lo5 colonies/ml unless noted. t A minimum of three cultures.
Hemodynamic Values. Group A (fable II, Figure 1): When comparing 2 hour postoperative values with preoperative values, systemic vascular resistance fell between 4.0 and 12.9 mm Hg/liter/min. One patient (Case 14) was unique: in contrast to all the other patients in group A, his 1 hour postoperative systemic vascular resistance was increased; at 2 hours it had fallen almost 20 per cent from the 1
hour level. At this time, a limulus test was positive. Cardiac output increased in five patients, was unchanged in two and decreased in three. No patient became symptomatically hypotensive, although in two patients mean arterial pressure fell below 75 mm Hg. Central venous pressure fell in five patients, was unchanged in two and increased in three. Group B (Table II, Figure 1): Two hours after genitourinary instrumentation, systemic vascular resistance increased between 3.6 and 12.0 mm Hg/liter/ min in seven patients and was unchanged in one. At no time did systemic vascular resistance decrease in the postoperative period. Cardiac output increased in one patient and decreased in seven. No patient had a fall in mean arterial pressure below 75 mm Hg. Central venous pressure decreased in four patients, was unchanged in one and increased in three. Bacteriology and Limulus Assay. Group A (Table Ill): Urine cultures were positive for gram-negative rods in 7 of 10 patients; one additional patient had bacteriuria by urinalysis. In two patients culture results were unobtainable. The one patient in whom bacteremia developed during the postoperative study had blood cultures positive for Staphylococcus aureus and multiple negative limulus assays for endotoxin. Five patients, all with bacteremia, had evi-
System
c3
C3PA
Preop
Postop
Preop
Postop
130 140 170 118 142 130 86 90
120 150 170 130 150 142 78 110
265 280 480 370 180 220 180 280
300 300 600 280 180 210 215 320
prekallikrein 70 to 100 (ages ml; C3PA 180 to 320 *g/100 ml.
50 years
dence of endotoxia by limulus assay. In one the less sensitive but very specific rat bioassay was also positive. The plasma of these five patients did not have higher than normal ribonuclease activity and did not precipitate Clq, and no abnormalities were found in coagulation assays, including thrombin time measurement, fibrinogen content and appearance of fibrin split products. Group B (fable IV): No patient who showed no change or an increase in systemic vascular resistance after genitourinary instrumentation had positive blood cultures or evidence of endotoxemia by limulus assay. Four patients had gram-negative bacteriuria; culture results were unobtainable in three others. Prekalllkrein and Spontaneous Esterase Actlvlty. Group A (Table Ill): Seven patients had a 10 per cent or greater decrease in prekallikrein levels after genitourinary instrumentation. In six of these seven patients the decrease was greater than 20 per cent; one patient had an increased plasma spontaneous esterase activity consistent with kallikrein. All five patients with endotoxemia had this evidence of kinin system activation as did the one patient with grampositive bacteremia and one additional patient in this group. In the seventh patient with a prekallikrein decrease, the limulus assay was negative: this patient was not bacteremic and had no complement changes. Group B (Table IV): Only one of the eight patients in this group showed evidence of kallikrein system activation; this patient had an isolated 10 per cent fall in prekallikrein during the postoperative period. Group A (Table Ill): Complement Activation. Three of the patients with endotoxemia and the pa-
July 1975
The American Journal of Medklne
Volume 59
65
ENDOTOXIN AND SYSTEMIC VASCULAR
TABLE V
Tabulation
of Presence
RESISTANCE-ROBINSON
of Bacteremia,
ET AL
Limulus Assays, and Prekallikrein
and Complement
Determinations
in Relation to Hemodynamic Values of Patients Before and After Undergoing Transurethral Resection or Cystoscopy SystemicVascular Resistance (no.) Increasedor No Change
Decreased (>lO%)
Negative
Present
... ... .. . ... . *.
... ... ... ... 1
...
...
...
...
6
...
...
...
1
,.. 5 1 1 3
... ... 5 ... ...
...
with gram-positive
Bacteremia (no.)
Positive
1 ... ... ... ...
tient
LimulusAssay (no.)
bacteremia
had evidence
. . ... . . ... ... .. . ... _____ of
activation of the complement system. Specifically, two patients, one with a positive limulus assay and the other with gram-positive bacteremia, had a greater than 25 per cent decrease in C3 levels and a greater than 25 per cent decrease in CBPA levels. Two patients with positive limulus assays had a greater than 40 per cent decrease in C3PA but no change in C3. No patient with a positive limulus assay or bacteremia had significant decreases in complement components Clq and C4, or properdin. Group B (Table IV): One patient in group B showed evidence of activation of the complement system (decrease in C3PA) without a concomitant change in prekallikrein.
Lysosomal Enzymes and Fibrin Split Products. Groups A and B: There was no significant change in the lysosomal enzyme cathepsin D in any patient. One patient in group B had a transient significant elevation of fibrin split products.
Statistical Correlation.
Groups A and B (Table V):
The incidence of endotoxemia or bacteremia with decreased systemic vascular resistance in patients in group A is significantly different (P <0.05) from that in patients in group 8. COMMENTS The major difficulty in evaluating hemodynamic change in patients with sepsis and septic shock is the lack of base line data relating to the hemodynamic state of the patient prior to investigation. Because of the high incidence of bacteremia and endotoxemia during ‘genitourinary instrumentation, these patients provide a clinical model for the study of the early hemodynamic effects of bacteremia and endotoxemia. Since control values may be obtained just prior to the surgical procedure, each patient may serve as his
66
July 1975
The American Journal Of Medicine
Absent
Volume 59
ElevatedSpontaneous Activity and/or DecreasedPrekallikrein (no.1 Elevatedor No Change
... ... ... ... ... ... ...
Decreased 0 16%)
.. . 1
Complement Activation (no.) Present
... .
1
1 3 1 ...
..
...
5 1
Absent
... ... ... ...
... . . .
own control and the results of the intervention expressed as per cent change in hemodynamic values over control values. Patients in whom evidence of bacteremia or endotoxemia does not develop show little hemodynamic change and therefore serve as a control for the effects of the surgical procedure on the hemodynamic values. The earliest hemodynamic change that we have observed in gram-negative bacteremia is a decrease in systemic vascular resistance, and this is usually accompanied by an increase in cardiac output, so that mean arterial pressure is maintained [ 71. Of the 18 patients we present here, IO showed a decrease in systemic vascular resistance, and 6 of the 10 had evidence of endotoxemia or bacteremia. None of the eight patients who did not vasodilate had evidence of bacteremia or endotoxemia. The mechanism for the fall in vascular resistance at this early stage of bacteremia or endotoxemia could be the activation of the kinin system as has been shown in monkeys by Melmon [4]. In man, Mason et al. [ 141 found that prekallikrein levels are significantly decreased not only during hypotensive, but also during normotensive gram-negative bacteremia. These investigators also were able to show that the kallikrein system was activated in rats and mice by injection of endotoxin. Not only kallikrein, but also the proteolytic enzymes trypsin, thrombin and plasmin are capable of effecting bradykinin release with subsequent vasodilatation [ 151. It is unlikely that any of the latter enzymes were responsible for the elevated esterase activity in patient 1 since his plasma had a substrate-ester ratio distinctive for kallikrein [ 121. Since the prekallikrein-kallikrein system may be depleted or inhibited early in the course of endotoxemia, another mechanism would be needed to explain persistence of the vasodilatation in patients in whom clinical shock develops. There is increasing
ENDOTOXIN AND SYSTEMIC
clinical evidence that the complement system may also be involved in hypotension associated with some bacterial and viral infections. McCabe [16] showed that significant reduction in C3 levels occurred in patients with gram-negative bacteremia and shock. Also, second infections with Dengue virus resulted in Clq precipitins in plasma, and decreases in C3, C3PA and C4 [ 171. These workers proposed that both the classic and alternate pathways were activated. Four of the six patients with endotoxemia or bacteremia in this study also activated components of the complement system. Since properdin levels were normal and did not change, activation was possible either via the alternate pathway at the level of C3PA or by direct activation of C3 with subsequent recruitment of CBPA by C3b. It is less likely that the classic pathway was activated since Clq and C4 levels did not change and Clq reactive material was absent. Anaphylatoxic activity of C3a and C5a generated by activation of either or both pathways could explain the protracted vasodilatation which accompanies the syndrome of septic shock. In one of the patients who had decreases in systemic vascular resistance but neither endotoxemia nor bacteremia, vasodilatation could be explained by activation of the kinin system during undetected endotoxemia. Three patients with evidence of vasodilatation showed no evidence of endotoxemia, bacteremia or activation of vasoactive substances, possibly on the basis of sampling times, plasma endotoxin levels lower than detection limits of the limulus test,
VASCULAR
RESISTANCE--ROBINSON
ET AL.
or activation of an esterase other than those detectable by the methods employed. The eight patients who showed no change or an increase in systemic vascular resistance following the genitourinary procedure serve as a control for the procedure. In none of these patients did endotoxemia or bacteremia develop. However, one did show evidence of C3PA depletion. Although two of the patients in group A became hypotensive, evidence of clinical shock did not develop in any of the patients in this series: this can be accounted for by the self-limited nature of the endotoxemia or bacteremia. It has been noted in previous studies of this type that clinically significant hypotensive episodes associated with genitourinary procedures, if detected early and not associated with persistent infection, generally respond quickly to fluid therapy sufficient to fill the intravascular compartment [7]. From our data, therefore, it is possible to propose that hemodynamic changes associated with endotoxemia and bacteremia come about through the following mechanism. Endotoxemia or bacteremia may activate the prekallikrein-kallikrein system with subsequent release of bradykinin which leads to initial vasodilatation and hypotension. The complement system may also become activated, the activation possibly being dependent on the quality, quantity or duration of endotoxemia or bacteremia, and provide a sequential stimulus for persistent vasodilatation perhaps through generation of substances classed together as anaphylatoxins.
REFERENCES 1.
2.
3.
Hinshaw LB, Emerson TE, Reins DA: Cardiovascular responses of the primate in endotoxin shock. Am J Physiol 210: 335, 1966. Garibaldi RA, Allman GW, Larsen DH, et al.: Detection of endotoxemia by the limulus test in patients with indwelling catheters. J Infect Dis 128: 551, 1973. Sullivan NM, Sutter VL, Mims MM, March VH. Finegold S: Clinical aspects of bacteremia after manipulation of the genitourinary tract. J Infect Dis 127: 49. 1973. Nies AS. Forsyth RP, Williams HE, Melmon KL: Contribution of kinins to endotoxin shock in unanesthetized Rhesus monkeys. Circ Res 22: 155, 1968. Mason DT, Melmon KL: Effects of bradykinin on forearm venous tone and vascular resistance in man. Circ Res 17: 106, 1965. Loeb HS. Cruz A, Teng CY, et al.: Hemodynamic studies in shock associated with infection. Br Heart J 29: 883, 1967. Gunnar RM, Loeb HS, Winslow EJ, et al.: Hemodynamic measurements in man. J Infect Dis 128: s295, 1973. Levin J. Poore TE, Zauber NP, et al.: Detection of endotoxin in the blood of patients with sepsis due to gram-negative bacteremia. N Engl J Med 283: 1313, 1970. Leavelle DE, Martens BF. Bowie FJ, et al.: Staphylococcal clumping on microtiter plates. Am J Clin Pathol 55: 452, 1971.
10.
11.
12.
13.
14.
15.
16.
17.
July 1975
Tan CH: An agar diffusion method for quantitative determination of ribonuclease. Biochim Biophys Acta 247: 463, 1971. Colman RW, Mason JW, Sherry S: Studies on kallikreinogen-kallikrein enzyme system of human plasma. Assay of components and observations in disease states. Ann Intern Med 71: 763, 1969. Colman RW, Mattler L, Sherry S: Studies on the prekallikrein (kallikreinogen)-kailikrein enzyme system of human plasma. II. Evidence relating the kaolin-activated arginine esterase to plasma kallikrein. J Clin Invest 48: 23, 1969. Mednis A, Remold HG: A sensitive fluorometric assay for the determination of cathepsin D. Anal Biochem 49: 134, 1972. Mason JW, Kleeberg U. Dolan P, et al.: Plasma kallikrein and Hageman factor in gram-negative bacteremia. Ann Intern Med 73: 545, 1970. Kellermeyer RW, Graham RC: Kinins-possible physiologic and pathologic roles in man. N Engl J Med 279: 754, 1968. McCabe WR: Serum complement levels in bacteremia due to gram-negative organisms. N Engl J Med 288: 21, 1973. Bokisch VA, Top FH, Russell PK, et al.: Pathogenic role of complement in dengue hemorrhagic shock syndrome. N Engl J Med 289: 966, 1973.
The American Journal of Medicine
Volume 59
67
JOHN A. ROBINSON.
M.D.
MARY L. KLODNYCKY,
M.D.
HENRY S. LOEB, M.D. MARY R. RACE,
B.S.
ROLF M. GUNNAR.
M.D.
Hines, Illinois Maywood, Illinois
From the Immunology and Cardiology Research Sections, Veterans Administration Hospital, Hines, Illinois; and Department of Medicine, Loyola University Strltch School of Medicine, Maywood, Illinois. This study was presented in part at the Annual Meeting of the American Federation for Clinical Research, Southern Section, January 24, 1974. Requests for reprints should be addressed to Dr. John A. Robinson, Department of Medicine, Loyola Unlverslty Strltch School of Medicine, Maywood, Illinois 60153. Manuscript accepted August 28, 1974.
Eighteen patients were studied prior to and again within 6 hours after transurethrai resection or cystoscopy. in addition to hemodynamic measurements, detection of endotoxin by iimuius assay and bacterioiogic sampling; prekaiiikrein, C3, C3 proactivator and iysosomai enzyme levels were measured. in five patients iimuius assays were positive, and in one, gram-positive bacteremia developed but iimuius assay remained negative. Ail six had significant decreases in prekaiiikrein, C3 or C3 proactivator. Systemic vascular resistance fell in ail six. Four additional patients who had a decrease in systemic vascular reststance were not endotoxemic or bacteremic; one of these had a decrease in prekaiiikrein only. in the remaining eight patients with neither bacteremia nor endotoxemia, systemk vascular resistance did not change or increase after instrumentation. One had a decrease in C3 proactivator, another in prekaiiikrein. There was no significant difference in age, disease, antibiotk therapy or bacteremia in the two groups of patients. Four of the five resectional procedures were performed in the group that showed decreases in systemk vascular resistance. The data suggest that acute endotoxemia or gram-positive bacteremia in man is associated with depletion of prekaiiikrein, decreased peripheral resistance and, in some instances, activation of the complement system. Although endotoxemia has been implicated as a consistent cause of acute hemodynamic effects in lower primates [ 11, hemodynamic changes during relatively uncomplicated endotoxemia in man are not known. The patient undergoing urologic procedures, especially transurethral resection, is at high risk for the development of endotoxemia [2] with or without bacteremia [3]. This type of patient provides a model for the longitudinal study of hemodynamic changes during endotoxemia and further affords a means to delineate significant changes in or the appearance of vasoactive mediators during endotoxemia in man. Melmon et al. [4] have shown in monkeys that bradykinin, a potent hypotensive peptide, appears early in endotoxemia and is associated with a decrease in peripheral vascular resistance. It has also been shown that infusion of bradykinin into the occluded forearm of man will decrease vascular re-
July 1975
The American Journal of Medfclm
Volume 59
61
ENDOTOXIN AND SYSTEMIC VASCULAR RESISTANCE-ROBINSON
ET AL.
sistance [6]. Similar changes in the kallikrein-kinin system might be implicated in a patient with endotoxemia. However, appearance and elevated levels of bradykinin are short-lived and modulated by several mechanisms. Therefore, it seems apparent that the sequential appearance of other vasoactive mediators, possibly initiated by endotoxin, are necessary to explain continuation of the hemodynamic changes noted in patients with gram-negative infection [6]. A proposed mechanism for continued vasodilatation during endotoxemia is direct or bypass activation of complement, especially the third (C3a) and fifth (C5a) complement components which have anaphylatoxic activity and could explain hemodynamic changes of more than transient character during gram-negative infection. It is hoped that identification of significant early physiologic changes will provide a more rational basis for therapeutic interventions in the bacteremic shock syndrome. However, the need to observe and document changes in vasoactive mediators during the transition from immediate vasodilatation to the more prolonged hemodynamic effects of endotoxemia in man became apparent when comparing hemodynamic studies during early bacteremia with similar studies in the established shock state [7]. The development of the limulus test, a highly sensitive assay for endotoxin [8], has provided a means for detecting endotoxin or endotoxin-like material in the absence of detectable gram-negative bacteremia. Nonbacteremic endotoxemia detected by the limulus assay could explain some of the hemodynamic changes seen in abacteremic patients after genitourinary manipulation and could also increase the yield of patients with endotoxemia in our study. The limulus assay and determinations of prekallikrein, the activated esterase that cleaves kinin from its precursor, kallikrein, and selected complement components were carried out in parallel with hemodynamic assessments preoperatively in 18 patients undergoing cystoscopy or transurethral resection. These data were then compared to subsequent similar determinations in hemodynamics during the immediate 6 hour postoperative period. The data suggest that initial acute endotoxemia or gram-positive bacteremia in man is associated with a depletion of prekallikrein, a decrease in peripheral resistance and, in some instances,
activation
MATERIALS
of the complement
had overt hepatic disease or significant postoperative blood loss. Some patients with gram-negative bacteriuria were being treated with suifathiazides or ampicillin prior to the procedure. in no patient did the septic shock syndrome develop in the postoperative period. Hemodynamic studies were conducted in a shock treatment unit. Central venous pressure was measured with Statham P23 Db transducer through a PE-160 catheter in the superior vena cava. Arterial pressure was measured with a Statham P23 Db transducer through a disposable Cournand needle in the femoral artery. Cardiac output was measured by the indicatordilution method using the Stewart-Hamilton formula. indocyanine green was injected into the superior vena cava and sampled from the femoral artery as blood was drawn through a Waters densitometer. Data for analysis represent the average of values from two to five indicator dilution curves made in rapid succession. Systemic vascular resistance (SVR) was calculated as follows: SVR mean aortic pressure (mm Hg) - central venous pressure (mm Hg) cardiac output (iiter/min) Preoperative and 2 and 6 hour postoperative venous samples were collected in plastic, pyrogen-free syringes for bacteriologic cultures, determination of coagulation factors and assays of endotoxin, prekaliikrein, spontaneous esterase activity and the complement system. Fibrin split products were quantitated by staphyiococcai clumping [9]. Detection of endotoxin was by the method of Levin et al. [a]. One pooled iysate was used for ail tests and consistently detected 0.5 ng endotoxin per O.l/mi heparinized plasma using Salmonella typhosa iipopoiysaccharide as standard. A test was positive when either complete geiation or a definite increase in viscosity was present at 24 hours. Test results were interpreted by two observers and, in one instance, confirmed by a lead sensitized rat bioassay that will detect 0.1 pg endotoxin/mg plasma (Dr. J. Filkins). Attempts to detect circulating antigen-antibody complexes were made by diffusion of Clq (first component of compiement) against whole serum in agar gels. Plasma ribonuclease activity was tested for by an agar diffusion technic sensitive to 4 X 10v5 kg/ml ribonuclease [lo]. Determination of kaiiikrein, prekaliikrein and kailikrein inhibitor was by the calorimetric quantitation of methanol generated by titrated plasma esterase activity with tosyi arginine methyl ester (Sigma Chemical) as substrate before and after the addition of kaolin [ 111. Reproducibility is within 4 pg methanol/hour/sample. Specificity of spontaneous esterase activity was assayed using either benzoyi arginine methyl ester, tosyl arginine methyl ester or lysylarginine methyl ester for determination of optimal esteraselsubstrate ratios and confirmation of kaliikrein esterase activity [ 121. Cathepsin D was determined by a I-nitroso-P-naphthoi fluorescence assay after incubation of plasma with 2 X lob4 M hemoglobin at pH 3.5 [ 131. Serum C3 was determined by quantitative immunodiffusion (Hyland Laboratories). Quantitation of C3 activator (C3PA) was by the Mancini technic in agar plates containing monospecific C3PA and antiserums (Hoechst Laboratories). Clq, C4 and properdin
system.
AND METHODS
Eighteen men, aged 50 to 86 years, were studied within 16 hours of and during the immediate 6 hour interval after cystoscopy or transurethral resection under local or spinal anesthesia. No patient was being treated with vasoactive drugs or was in congestive heart failure, and no patient
62
July 1975
The American
Journal
of Medicine
Volume
59
ENDOTOXIN AND SYSTEMIC
TABLE
I
Characterization
of Patients
by Age, Disease,
Age W
Case No.
Procedure
Disease
I 5 7 9 11 12
78 75 66 54 86 70
Benign prostatic hypertrophy Metastatic carcinoma Benign prostatic hypertrophy Renal mass Carcinoma of prostate Carcinoma of bladder
13
83
Benign
14
73
Neurogenic
15
60
Benign
prostatic
hypertrophy
16
76
Benign
prostatic
hypertrophy
2 3 4 6 8 10 17
60 50 75 53 54 53 59
Benign prostatic hypertrophy Nephrolithiasis Benign prostatic hypertrophy Benign prostatic hypertrophy Benign prostatic hypertrophy Carcinoma of prostate Carcinoma of prostate
18
75
prostatic
* immediately TABLE _--
II --
-__
Benign
Anbhotics+
A
hypertrophy
bladder
prostatic
ET AL.
Prior to Study
None None Sulfathiazide None None None
Cystoscopy Cystoscopy Cystoscopy Cystoscopy Cystoscopy Fulguration and cystoscopy Transurethral resection Cystoscopy
None Ampicillin, sulfathiazide Tetracycline
Transurethral resection Transurethral resection
Group
______
Therapy
RESISTANCE--ROBINSON
Procedure Group
___
and Antibiotic
VASCULAR
Sulfathiazide
B Cystoscopy Cystoscopy cystoscopy Cystoscopy Cystoscopy Cystoscopy Cystoscopy biopsy Cystoscopy
hypertrophy
None Tetracycline None Tetracycline None Erythromycin None
and
Ampicillin
prior to procedure. Preoperative Genitourinary co
(Rest) and Postoperative Procedures
(Postop) Hemodynamic _.~~
in Patients
Undergoing
_..~~
MAP
--
Measurements
~__
Case No.
Rest
Postop
Rest
Postop
1 5 7 9 11 12 13 14 15 16
1.4 2.1 4.6 4.2 4.6 4.6 3.3 5.1 3.7 5.8
2.0 2.1 3.6 4.7 4.5 7.1 4.8 3.8/‘4.9* 4.2 3.5
90 90 110 106 90 80 130 121 122 138
80 70 84 84 68 97 120 116/118* 124 60
2 3 4 6 8 10 17 18
2.8 2.2 3.7 5.5 7.4 6.9 9.7 4.8
2.1 2.6 1.6 3.8 5.6 5.7 5.7 6.3
90 106 120 110 80 78 104 76
80 90 112 84 82 78 78 104
SVR
CVP
Rest
Group A 0
-1 5 3 3 3 12 6 7 2
0 -3 3 -1 5 11 10 4/3* 10 1
29.6 22.0 32.6 24.6 18.9 16.7 35.8 22.5 31.1 25.7
18.4 18 22.5 18.0 14 12.1 22.9 29.5/23.6* 25.7 14.0
Group B 0 1 0 3 3 -7 5 7
14.5 24.0 20.0 14.9 14.0 10.1 9.5 14.8
20.0 32.0 32.0 22.0 18.4 14.2 17.8 15.4
NOTE: CO = cardiac output (mm Hg/liter/min). MAP = mean arterial pressure (mm Hg). CVP = central venous pressure (mm HzO). SVR = systemic vascular resistance (mm Hg/liter/min). * Values to left of slash were obtained 1 hour after the procedure; values to right of slash obtained 2 hours after the procedure.
July 1975
The American Journal of Medicine
Volume 59
63
ENDOTOXIN AND SYSTEMIC VASCULAR
LIMULUS
RESISTANCE-ROBINSON
ET AL.
ASSAY
BACTEREMIA
Figure 1. Directional changes in systemic vascuiaf resistance in patients 2 hours after genitourinary instrumentation. The K within bars indicates a 70 per cent or greater decrease in prekallikrein, a C indicates either a 25 per cent decrease in C3 or C3PA, or both. The immediate postoperative directional change in systemic vascular resistance of patient 8, group A, is indicated by a broken line bar; his 2 hour postoperative value is expressed with similar determinations in patients in group A.
-20
were kindly determined by Dr. H. Gewurz. Hemodynamic and bacteriologic results were not known by the laboratory prior to an assay. Statistical analysis was made using the Chi square analysis and Yates correction factor. Hemodynamic measurements were’ made on the day before or the morning of genitourinary surgery, and venous blood samples for analysis were drawn at the same time. Unless the patient was to have surgery within the next 2 hours, catheters were withdrawn and then reinserted immediately postoperatively. Hemodynamic measurements were repeated twice between 1 and 6 hours postoperatively.
than 10 per cent decrease in systemic vascular resistance when the preoperative values were compared with the 2 hour postoperative measurements (group A) and those who showed no change or an increase in vascular resistance (group 6). The 10 patients in group A (Table I) had an average age of 72 years, and the 8 patients in group B (Table I) had an average age of 60 years. There was no difference in the type of medication, anesthesia or pathology between groups A and B. However, four of the five patients who had a resection, fulguration or biopsy procedure were in group A. There were no significant differences in fluid volume infusions or duration of operation between these patients and others in group A. greater
RESULTS
On the basis of hemodynamic were
divided
TABLE III
into two
groups:
Preoperative Complement
studies, the patients those who showed a
(Preop) and 2 Hour Postoperative (Postop) Clinical Microbiologic, Limulus Assay, Kallikrein and Component Values in Patients Who Had a Postoperative Decrease in Systemic Vascular Resistance
Kallikrein System Bacteriology Case
Urine*
No.
(ww)
1 5 7 9 11 12 13 14 15 16
Proteus Escherichia coli Not determined Not determined Bacteria by urinalysis Escherichia coli Pseudomonas Gram-negative rods (104) Gram-negative rods (lOa) Escherichia coli
Bloodt Preop Postop
July 1975
Spontaneous Esterase Activity
Preop Postop Preop
-
-
-
+ -
7.4 0 0 3.1
-
-
-
+
0
-
-
-
+ + +
-
-
-
Staphylococcus aureus
NOTE: Kallikrein system: normal values = spontaneous and over) rmol methanol/hour. Complement system: * More than 106 colonies/ml unless noted. t A minimum of three cultures.
64
Limulus Assay
The American Journal of Medlclne
0
0 0 3.0 0
Complement System Prekallikrein
c3
C3PA
Postop
Preop
Postop
Preop
Postop
Preop
Postop
21.7 1.8 0 1.2 2.5 0 0 1.8 1.8 0
101 92 94 77 89 74 94 73 72 77
74 101 84 74 69 78 a4 59 47 60
140 a4 110 90 145 150 132 155 145 68
120 94 110 72 127 180 145 110 145 42
410 150 260 105 300 170 380 220 320 140
430 130 300 180 300 200 190 160 160 100
activity 0 to 3 rmol methanol/hour; prekalli krein 70 to 100 (ages 50 years normal values = C3 120 to 180 mg/lOO ml; CSPA 180 to 320 fig/100 ml.
Volume SS
ENDOTOXIN
TABLE IV
_._~._
_
Resistance
in the Postoperative
3 4 6 8 10 17 18
RESISTANCE-ROBINSON
ET AL.
Period
Bacteriology
2
VASCULAR
Preoperative (Preop) and 2 Hour Postoperative (Postop) Clinical Microbiologic, Limulus Assay, Kallikrein and Complement Component Values in Patients with No Change or an Increase in Systemic Vascular
Kallikrein System __._.~__
-..--.--..
Case No.
AND SYSTEMIC
Preop Postop
(preop) Escherichia
Limulus Assay
Bloodt
Urine*
coli
Gram-negative (103) Not determined Escherichia coli Not determined Negative Escherichia coli Not determined ___
Spontaneous Esterase Activity
Preop Postop Preop
Postop
-..__
_ Complement
Prekallikrein Preop
Postop
-
-
-
-
0
1.9
57
63
-
-
-
-
0 0 1.9 0 3.1 0 0
3.7 1.8 0 3.1 3.1 0 0
98 109 70 85 102 94 56
97 112 80 85 112 82 60
-____ NOTE: Kallikrein system: normal values = spontaneous activity 0 to 3 pm methanol/hour; and over) pm methanol/hour. Complement system: normal values = C3 120 to 180 mg/lOO _ * More than lo5 colonies/ml unless noted. t A minimum of three cultures.
Hemodynamic Values. Group A (fable II, Figure 1): When comparing 2 hour postoperative values with preoperative values, systemic vascular resistance fell between 4.0 and 12.9 mm Hg/liter/min. One patient (Case 14) was unique: in contrast to all the other patients in group A, his 1 hour postoperative systemic vascular resistance was increased; at 2 hours it had fallen almost 20 per cent from the 1
hour level. At this time, a limulus test was positive. Cardiac output increased in five patients, was unchanged in two and decreased in three. No patient became symptomatically hypotensive, although in two patients mean arterial pressure fell below 75 mm Hg. Central venous pressure fell in five patients, was unchanged in two and increased in three. Group B (Table II, Figure 1): Two hours after genitourinary instrumentation, systemic vascular resistance increased between 3.6 and 12.0 mm Hg/liter/ min in seven patients and was unchanged in one. At no time did systemic vascular resistance decrease in the postoperative period. Cardiac output increased in one patient and decreased in seven. No patient had a fall in mean arterial pressure below 75 mm Hg. Central venous pressure decreased in four patients, was unchanged in one and increased in three. Bacteriology and Limulus Assay. Group A (Table Ill): Urine cultures were positive for gram-negative rods in 7 of 10 patients; one additional patient had bacteriuria by urinalysis. In two patients culture results were unobtainable. The one patient in whom bacteremia developed during the postoperative study had blood cultures positive for Staphylococcus aureus and multiple negative limulus assays for endotoxin. Five patients, all with bacteremia, had evi-
System
c3
C3PA
Preop
Postop
Preop
Postop
130 140 170 118 142 130 86 90
120 150 170 130 150 142 78 110
265 280 480 370 180 220 180 280
300 300 600 280 180 210 215 320
prekallikrein 70 to 100 (ages ml; C3PA 180 to 320 *g/100 ml.
50 years
dence of endotoxia by limulus assay. In one the less sensitive but very specific rat bioassay was also positive. The plasma of these five patients did not have higher than normal ribonuclease activity and did not precipitate Clq, and no abnormalities were found in coagulation assays, including thrombin time measurement, fibrinogen content and appearance of fibrin split products. Group B (fable IV): No patient who showed no change or an increase in systemic vascular resistance after genitourinary instrumentation had positive blood cultures or evidence of endotoxemia by limulus assay. Four patients had gram-negative bacteriuria; culture results were unobtainable in three others. Prekalllkrein and Spontaneous Esterase Actlvlty. Group A (Table Ill): Seven patients had a 10 per cent or greater decrease in prekallikrein levels after genitourinary instrumentation. In six of these seven patients the decrease was greater than 20 per cent; one patient had an increased plasma spontaneous esterase activity consistent with kallikrein. All five patients with endotoxemia had this evidence of kinin system activation as did the one patient with grampositive bacteremia and one additional patient in this group. In the seventh patient with a prekallikrein decrease, the limulus assay was negative: this patient was not bacteremic and had no complement changes. Group B (Table IV): Only one of the eight patients in this group showed evidence of kallikrein system activation; this patient had an isolated 10 per cent fall in prekallikrein during the postoperative period. Group A (Table Ill): Complement Activation. Three of the patients with endotoxemia and the pa-
July 1975
The American Journal of Medklne
Volume 59
65
ENDOTOXIN AND SYSTEMIC VASCULAR
TABLE V
Tabulation
of Presence
RESISTANCE-ROBINSON
of Bacteremia,
ET AL
Limulus Assays, and Prekallikrein
and Complement
Determinations
in Relation to Hemodynamic Values of Patients Before and After Undergoing Transurethral Resection or Cystoscopy SystemicVascular Resistance (no.) Increasedor No Change
Decreased (>lO%)
Negative
Present
... ... .. . ... . *.
... ... ... ... 1
...
...
...
...
6
...
...
...
1
,.. 5 1 1 3
... ... 5 ... ...
...
with gram-positive
Bacteremia (no.)
Positive
1 ... ... ... ...
tient
LimulusAssay (no.)
bacteremia
had evidence
. . ... . . ... ... .. . ... _____ of
activation of the complement system. Specifically, two patients, one with a positive limulus assay and the other with gram-positive bacteremia, had a greater than 25 per cent decrease in C3 levels and a greater than 25 per cent decrease in CBPA levels. Two patients with positive limulus assays had a greater than 40 per cent decrease in C3PA but no change in C3. No patient with a positive limulus assay or bacteremia had significant decreases in complement components Clq and C4, or properdin. Group B (Table IV): One patient in group B showed evidence of activation of the complement system (decrease in C3PA) without a concomitant change in prekallikrein.
Lysosomal Enzymes and Fibrin Split Products. Groups A and B: There was no significant change in the lysosomal enzyme cathepsin D in any patient. One patient in group B had a transient significant elevation of fibrin split products.
Statistical Correlation.
Groups A and B (Table V):
The incidence of endotoxemia or bacteremia with decreased systemic vascular resistance in patients in group A is significantly different (P <0.05) from that in patients in group 8. COMMENTS The major difficulty in evaluating hemodynamic change in patients with sepsis and septic shock is the lack of base line data relating to the hemodynamic state of the patient prior to investigation. Because of the high incidence of bacteremia and endotoxemia during ‘genitourinary instrumentation, these patients provide a clinical model for the study of the early hemodynamic effects of bacteremia and endotoxemia. Since control values may be obtained just prior to the surgical procedure, each patient may serve as his
66
July 1975
The American Journal Of Medicine
Absent
Volume 59
ElevatedSpontaneous Activity and/or DecreasedPrekallikrein (no.1 Elevatedor No Change
... ... ... ... ... ... ...
Decreased 0 16%)
.. . 1
Complement Activation (no.) Present
... .
1
1 3 1 ...
..
...
5 1
Absent
... ... ... ...
... . . .
own control and the results of the intervention expressed as per cent change in hemodynamic values over control values. Patients in whom evidence of bacteremia or endotoxemia does not develop show little hemodynamic change and therefore serve as a control for the effects of the surgical procedure on the hemodynamic values. The earliest hemodynamic change that we have observed in gram-negative bacteremia is a decrease in systemic vascular resistance, and this is usually accompanied by an increase in cardiac output, so that mean arterial pressure is maintained [ 71. Of the 18 patients we present here, IO showed a decrease in systemic vascular resistance, and 6 of the 10 had evidence of endotoxemia or bacteremia. None of the eight patients who did not vasodilate had evidence of bacteremia or endotoxemia. The mechanism for the fall in vascular resistance at this early stage of bacteremia or endotoxemia could be the activation of the kinin system as has been shown in monkeys by Melmon [4]. In man, Mason et al. [ 141 found that prekallikrein levels are significantly decreased not only during hypotensive, but also during normotensive gram-negative bacteremia. These investigators also were able to show that the kallikrein system was activated in rats and mice by injection of endotoxin. Not only kallikrein, but also the proteolytic enzymes trypsin, thrombin and plasmin are capable of effecting bradykinin release with subsequent vasodilatation [ 151. It is unlikely that any of the latter enzymes were responsible for the elevated esterase activity in patient 1 since his plasma had a substrate-ester ratio distinctive for kallikrein [ 121. Since the prekallikrein-kallikrein system may be depleted or inhibited early in the course of endotoxemia, another mechanism would be needed to explain persistence of the vasodilatation in patients in whom clinical shock develops. There is increasing
ENDOTOXIN AND SYSTEMIC
clinical evidence that the complement system may also be involved in hypotension associated with some bacterial and viral infections. McCabe [16] showed that significant reduction in C3 levels occurred in patients with gram-negative bacteremia and shock. Also, second infections with Dengue virus resulted in Clq precipitins in plasma, and decreases in C3, C3PA and C4 [ 171. These workers proposed that both the classic and alternate pathways were activated. Four of the six patients with endotoxemia or bacteremia in this study also activated components of the complement system. Since properdin levels were normal and did not change, activation was possible either via the alternate pathway at the level of C3PA or by direct activation of C3 with subsequent recruitment of CBPA by C3b. It is less likely that the classic pathway was activated since Clq and C4 levels did not change and Clq reactive material was absent. Anaphylatoxic activity of C3a and C5a generated by activation of either or both pathways could explain the protracted vasodilatation which accompanies the syndrome of septic shock. In one of the patients who had decreases in systemic vascular resistance but neither endotoxemia nor bacteremia, vasodilatation could be explained by activation of the kinin system during undetected endotoxemia. Three patients with evidence of vasodilatation showed no evidence of endotoxemia, bacteremia or activation of vasoactive substances, possibly on the basis of sampling times, plasma endotoxin levels lower than detection limits of the limulus test,
VASCULAR
RESISTANCE--ROBINSON
ET AL.
or activation of an esterase other than those detectable by the methods employed. The eight patients who showed no change or an increase in systemic vascular resistance following the genitourinary procedure serve as a control for the procedure. In none of these patients did endotoxemia or bacteremia develop. However, one did show evidence of C3PA depletion. Although two of the patients in group A became hypotensive, evidence of clinical shock did not develop in any of the patients in this series: this can be accounted for by the self-limited nature of the endotoxemia or bacteremia. It has been noted in previous studies of this type that clinically significant hypotensive episodes associated with genitourinary procedures, if detected early and not associated with persistent infection, generally respond quickly to fluid therapy sufficient to fill the intravascular compartment [7]. From our data, therefore, it is possible to propose that hemodynamic changes associated with endotoxemia and bacteremia come about through the following mechanism. Endotoxemia or bacteremia may activate the prekallikrein-kallikrein system with subsequent release of bradykinin which leads to initial vasodilatation and hypotension. The complement system may also become activated, the activation possibly being dependent on the quality, quantity or duration of endotoxemia or bacteremia, and provide a sequential stimulus for persistent vasodilatation perhaps through generation of substances classed together as anaphylatoxins.
REFERENCES 1.
2.
3.
Hinshaw LB, Emerson TE, Reins DA: Cardiovascular responses of the primate in endotoxin shock. Am J Physiol 210: 335, 1966. Garibaldi RA, Allman GW, Larsen DH, et al.: Detection of endotoxemia by the limulus test in patients with indwelling catheters. J Infect Dis 128: 551, 1973. Sullivan NM, Sutter VL, Mims MM, March VH. Finegold S: Clinical aspects of bacteremia after manipulation of the genitourinary tract. J Infect Dis 127: 49. 1973. Nies AS. Forsyth RP, Williams HE, Melmon KL: Contribution of kinins to endotoxin shock in unanesthetized Rhesus monkeys. Circ Res 22: 155, 1968. Mason DT, Melmon KL: Effects of bradykinin on forearm venous tone and vascular resistance in man. Circ Res 17: 106, 1965. Loeb HS. Cruz A, Teng CY, et al.: Hemodynamic studies in shock associated with infection. Br Heart J 29: 883, 1967. Gunnar RM, Loeb HS, Winslow EJ, et al.: Hemodynamic measurements in man. J Infect Dis 128: s295, 1973. Levin J. Poore TE, Zauber NP, et al.: Detection of endotoxin in the blood of patients with sepsis due to gram-negative bacteremia. N Engl J Med 283: 1313, 1970. Leavelle DE, Martens BF. Bowie FJ, et al.: Staphylococcal clumping on microtiter plates. Am J Clin Pathol 55: 452, 1971.
10.
11.
12.
13.
14.
15.
16.
17.
July 1975
Tan CH: An agar diffusion method for quantitative determination of ribonuclease. Biochim Biophys Acta 247: 463, 1971. Colman RW, Mason JW, Sherry S: Studies on kallikreinogen-kallikrein enzyme system of human plasma. Assay of components and observations in disease states. Ann Intern Med 71: 763, 1969. Colman RW, Mattler L, Sherry S: Studies on the prekallikrein (kallikreinogen)-kailikrein enzyme system of human plasma. II. Evidence relating the kaolin-activated arginine esterase to plasma kallikrein. J Clin Invest 48: 23, 1969. Mednis A, Remold HG: A sensitive fluorometric assay for the determination of cathepsin D. Anal Biochem 49: 134, 1972. Mason JW, Kleeberg U. Dolan P, et al.: Plasma kallikrein and Hageman factor in gram-negative bacteremia. Ann Intern Med 73: 545, 1970. Kellermeyer RW, Graham RC: Kinins-possible physiologic and pathologic roles in man. N Engl J Med 279: 754, 1968. McCabe WR: Serum complement levels in bacteremia due to gram-negative organisms. N Engl J Med 288: 21, 1973. Bokisch VA, Top FH, Russell PK, et al.: Pathogenic role of complement in dengue hemorrhagic shock syndrome. N Engl J Med 289: 966, 1973.
The American Journal of Medicine
Volume 59
67