Diagnostic significance of hypocomplementemia

Kidney International, Vol. 39 (1991), pp. 811—821

EDITORIAL REVIEW

Diagnostic significance of hypocomplementemia Certain immune complex (IC)-mediated diseases frequently cause hypocomplementemia as assessed by measurement of serum or plasma C3, C4 or CH5O (third component of the complement system, fourth component of the complement system and hemolytic complement activity, respectively).

C5b, which then binds C6 and C7. C5b-7 then releases from its C3b binding site, inserting into a nearby cell membrane. The C5b-7 complex then assembles the membrane attack complex (MAC), C5b-8, poly C9. MAC results in injury to that cell by forming channels and by disrupting the cell membrane.

Thus, the measurement of serum or plasma C3, C4 and CHSO is

Classical pathway activation by IC requires first that the immune system recognize an antigen and mount a specific antibody response to it. By contrast, alternative pathway acti-

used commonly to assess for the presence of an IC-mediated disease or, in patients with a known IC disorder, to assess for disease activity [1). However, there are numerous non-immunologic conditions that also can cause hypocomplementemia. It is important for the clinician to be aware of the non-immunologic conditions that cause hypocomplementemia because some can induce a systemic disorder that can easily be confused with an IC-mediated vasculitis. In addition, a non-immunologic condition that causes hypocomplementemia could develop in a patient with an IC-mediated disease that is inactive. If the physician does not recognize that the hypocomplementemia is the result of the nonimmunologic condition, this could lead to incorrect treatment of the patient. Although the literature on complement is voluminous and includes excellent recent reviews of disorders of complement metabolism [2, 3] and its importance in the diagnosis of renal disease [4], to our knowledge there is no previous publication that attempts to provide a comprehensive listing of the causes of hypocomplementemia and its differential diagnosis. It is the purpose of this work to provide that listing, with annotations regarding differential diagnosis, and to discuss several other issues relevant to the evaluation of the hypocomplementemic patient.

vation can occur in the absence of immune recognition. What is required for alternative pathway activation is that C3 become activated, which is constantly occurring at a low rate in plasma because of hydrolysis of C3 to C3(H20), and bind factor B to permit factor B cleavage by factor D, to form the C3 convertase C3bBb. C3bBb in the fluid phase forms metastable C3b which binds to a surface. Once bound, C3b can bind factor B, permit cleavage of factor B by factor D, thereby forming on the surface

the C3 convertase C3bBb. This convertase results in the formation of more C3b which bind to the surface and permit attachment of CS to the C3b molecule. The CS attached to C3b is now susceptible to cleavage by C3bBb, thereby activating the

terminal pathway, as described above. Alternative pathway activation is enhanced when C3b molecules bind to surfaces that inhibit the inactivation of C3b. Typically such "activator" surfaces are complex polysaccharides (such as, dialysis membranes) or lipopolysaccharides (such as, bacterial cell walls). For simplicity's sake, the control mechanisms of the comple-

ment system, except for factor D, are not shown in Table 1. However, these proteins play an important role in regulating complement activation. The control mechanisms of the comPathophysiology of hypocompiementemia plement system include: Cl inhibitor (a plasma protein that A detailed discussion of the complement system is beyond inhibits Cl by binding to Cir and Cis), C4-binding protein

the scope of this work, which is intended primarily as a clinical

(C4BP, a plasma protein that promotes the dissociation of C4b

guide to the evaluation of the patient who is found to be from C2b and provides co-factor activity for the degradation of hypocomplementemic by measurement of C3, C4 or CH5O. However, to provide a general schema for understanding the pathophysiology of hypocomplementemia, an overview of the complement system is shown in Table 1 and discussed below. As shown in Table 1, hypocomplementemia can occur as a result of activation of the classical or alternative pathways. Classical pathway activation occurs when Clq binds to an immune complex (or to certain tumor cells or viruses) causing the assembly of the Cl complex (Clq + 2Clr + 2Cls), which cleaves and activates C4 to bind C2 which is then cleaved and activated. These steps result in the formation of C4b2b which

C4b by factor I, a plasma protein that inactivates C3b and C4b).

Factor H (a plasma protein that serves as a co-factor which increases the efficiency of factor I to degrade C3b and C4b), complement receptor Type 1 (CR1, a membrane protein that is present on primate erythrocytes, most leukocytes, and glomer-

ular epithelial cells, which acts as a co-factor for factor I degradation of C3b and C4b), decay accelerating factor (DAF, a membrane protein present on all cells which accelerates the

decay of both the classical and alternative pathway C3/C5 convertases, properdin (a plasma protein that binds to C3bBb to

stabilize its C3/C5 convertase activity), factor D (a plasma

cleaves C3 into its active form C3b. C3b bound to a surface can protein that cleaves factor B to its active form Bb), and S bind C5, permitting a nearby C4b2b to cleave and activate C5 to protein (vitronectin, a plasma protein that binds to CSb-7 released from C3b and prevents the CSb-7 from inserting into nearby targets. Received for publication January 11, 1990 and in revised form June 25, 1990 Accepted for publication August 11, 1990

© 1991 by the International Society of Nephrology

As can be seen from Table 1, classical pathway activation results in the consumption of both C3 and C4. By contrast, activation of the alternative pathway results in consumption of C3 but not C4. Thus, a reduction in both C3 and C4 levels is

811

812

Hebert ci a!: Diagnostic significance o.f hypocomplementemia Table 1. Pathways for complement activation

Comment

Classical pathway (CP)

Alternative pathway (AP)

I) Activators: IgG, 1gM containing IC

Most common activators ofCP (left) or AP (right)

I) Activators: IgA containing IC. Damaged tissue. Gram negative bacteria.

2) Clq binds to IC causing

Step that initiates activation of CP (left) or AP (right)

C3 is consumed

2) C3 is hydrolyzed permitting factor B to bind to C3. B is cleaved by factor D. The fluid-phase complex C3bBb is formed which cleaves C3 to C3b which then binds to the activator surface. Surface bound C3b binds factor B. D cleaves B forming C3bBb on the surface.

The anaphylatoxin C3a is released.

3) C3bBb cleaves C3 forming C3a and C3b.

assembly of a complex ' (Clq,C1r2,C1s2) which

cleaves C4 and C2 to form an enzymatically active complex

..,

C4 is consumed

C4b2b.

.-

—

3) C4b2b cleaves C3 forming -.... C3a a

and C3b. C3b binds to --..—-—

surface or to the IC.

——

——

This step leads to

I

4) C3b binds CS permitting a

nearby C4b2b to cleave CS to C5b and C5a. C5b is temporarily bound to C3b.

/

/

"amplification" of complement activation because each C3bBb begets more C3b's, which beget more C3bBb's, etc. The generation of C3b by the CP can also generate C3bBb. The anaphylatoxin C5a is released

4) More C3bBb is formed when the nascent C3b binds to the surface.

5) Some of the surface bound C3bBb binds near another C3b forming (C3b)2Bb, which is a C5 convertase. 6) (C3b)2Bb cleaves C5 to C5a and C5b. The C5b is temporarily bound to C3b.

Terminal pathway

C5b, after binding to C6 and 7, is released from C3b and inserts into the nearest cell surface, causing assembly and activation of the CSb,6,7,8,9 complex (Membrane attack complex (MAC), C5-9 lytic complex). This complex, which is attached to the cell membrane, cause injury to that cell. See text for more detail concerning the mechanisms of complement activation and for discussion of the control proteins of the complement system.

evidence for activation of the classical pathway. By contrast, a reduction in C3 but normal C4 is evidence for activation of the alternative pathway. However, there are numerous exceptions to this "rule," as discussed herein. Low C3, C4, or CH5O levels can also occur as a result of decreased synthesis of complement components, or increased consumption of complement components, because inadequate function of the inhibitors of complement activation permits unregulated complement activation. The role of these mechanisms in causing hypocomplementemia is also discussed. Clinical laboratory assessment of the complement system

Initial assessment of complement Status For the patient being evaluated for the first time the measurement of plasma or serum CH5O, C3 and C4 levels is recom-

mended. The CH5O assay assesses the overall function of the classical and terminal pathway components Cl to C8 (C9 is not required for hemolysis of sheep erythrocytes) [1]. The CH5O assay determines the capacity of the patient's serum to support complement-mediated lysis of sheep erythrocytes that have been coated with anti-sheep erythrocyte antibodies capable of activating the classical pathway. If a complement component is absent, as in congenital C2 deficiency, the patient's serum C3 and C4 levels will be normal but hemolytic activity, assessed by the CH5O assay, will be lacking. Thus, the measurement of CH5O is useful as a screening test for hypocomplementemic

states that would be missed if only C3 and C4 levels are measured. The CH5O assay is less useful as a monitor of serial changes in serum complement levels because, for example, C3 and C4 levels may be abnormally low, yet the capacity of the

Heberi et al: Diagnostic sign(ficance of hypocomplementemia

813

C3 are not due simply to differences in the calibration of the assay. Difference among the C3 assays may be of clinical significance. For example, C3 levels measured by radial immu-

250

nodiffusion, which has a low lower limits of normal, has

200

significantly less diagnostic sensitivity and specificity in active SLE than do C3 levels measured by nephelometry, which has a relatively high lower limits of normal [9]. In the same Centers shown in Figure 1, the range of normal

150

for serum C4 levels also varied widely: from 10 to 28 mg/dl (lower limits) to 35 to 80 mg/dl (upper limits). However, in contrast to the situation with serum C3 levels, the Centers with

100

50 1

2 3 4 5 6 7 8 9 10 11 12 13 14 Center number

Fig. 1. Relationship between the upper and lower limits of normal for serum C3 levels at the 14 Centers involved in the NIH-sponsored Trial of Plasmapheresis in Severe SLE Nephritis. The Centers are arranged in ascending order according to the lower limits of normal for serum C3 levels. Note that values for the upper limits of normal for C3 do not parallel the values for lower limits of normal for C3.

the highest lower limits of normal for C4 also tended to have the highest upper limits of normal for C4 (data not shown).

Serial assessment of complement status Serial measurement of serum complement levels can be used to monitor the activity of IC-mediated diseases. This issue has been particularly well studied in patients with SLE [10—23]. At the present time it is not clear whether measurement of C3, C4, or both is best for monitoring SLE disease activity. However, there is evidence that the greater the decrease in C3 or C4 levels

below normal, the more likely it is that the patient's SLE is patient's serum to hemolyze the sensitized sheep erythrocytes may remain within the normal range [1, unpublished observations]. A CH100 assay (Sigma Diagnostics, St. Louis, Missouri; Kallestad Diagnostics, Austin, Texas, USA) is an alternative to a CHS0 assay and is said to be easier to perform. Serum C3 and C4 levels can be measured functionally, but usually are measured by an antigenic assay using polyclonal antibodies against C3 or C4, and radial immunodiffusion or nephelometry for detection of the antigen-antibody reaction. Storing blood specimens at room temperature, even for several days, has little effect on the antigenic assay for C3 and C4. That is, although degradation of C3 and C4 occurs during storage of blood at room temperature, the degradation products remain reactive with the polyclonal antibodies used to detect C3 and C4. However, for the CH5O assay (a "functional assay") the blood specimens must be drawn into EDTA (to prevent in vitro complement activation) or the serum separated from the blood cells within one hour of obtaining the blood sample, and frozen at —20°C (short term) or —70°C (long term) until assayed. In rare SLE patients, allowing the blood sample to clot at room temperature results in a low CH5O level because of consumption of Cl, C4, and C2. This does not occur when the blood is

active [20, 22].

In SLE patients, serum C3 and C4 levels tend to change in parallel. Figure 2 is an example of the correlation between C3 and C4 levels observed in a single patient. To determine the variability from patient to patient in the correlation of C3 and C4 levels, we calculated the correlation coefficient for serum C3 and C4 levels in each of 20 consecutive SLE patients with GN, who had multiple C3 and C4 measurements by nephelometry 13 SE) over a wide range of C3 and C4 levels. In (mean, 32 these patients the mean correlation coefficient was 0.8 0.1 SD. Thus, in most SLE patients, measuring either serum C3 or C4 levels gives essentially equivalent information regarding the relative change in complement levels. Nevertheless, we recommend the measurement of serum C3 because it has been our experience and the experience of others [9, 18—20, 23] that,

compared to serum C4, serum C3 levels are more likely to change from normal to abnormal as SLE changes from quiescent to active. This recommendation contrasts with a previous, widely quoted report suggesting that low C4 is a more sensitive index of SLE activity than low C3 [18]. However, in that report

part of the reason for the higher percent of patients with abnormally low C4 versus C3 during SLE relapse is that,

allowed to clot at 37°C or is drawn into EDTA [5]. Thus, compared to C3, C4 was more likely to be abnormally low, even allowing blood specimens to clot at room temperature is not recommended if functional assays are to be performed. The method of collecting blood from the patient by venipuncture versus an indwelling venous or arterial catheter appears to be unimportant because assays for blood complement levels have the same result by all three methods of blood collection [6]. The normal range for serum C3 varies considerably from

during SLE remission [18]. Thus, compared to C3, C4 is less likely to change from normal to abnormal as the patient changes

from inactive to active SLE [9, 18—20, 23]. One reason for persistently low C4 levels in SLE patients, even during SLE remission, is that one or more C4 null genes (genes that do not produce C4) [24—26] are more common in SLE (50 to 80%), than

in the general population [271. Thus, some SLE patients have low or borderline low C4 levels, even when there is no disease activity [9, 13, 18, 20, 21]. A further reason for the decreased center trial of plasmapheresis in severe SLE nephritis [7, 8]. diagnostic specificity of C4 in SLE is that, compared to C3, the Note that the lower limit of normal for serum C3 ranges from 70 range of normal for C4 is large. For example, by nephelometry to 116 m%, a difference of 66%. Note also that the upper limits assay for serum C4 there is about a 300% difference between the of normal at these Centers do not parallel the values for the lower and upper limits of normal; by contrast, for serum C3 lower limits of normal. This suggests that the differences among there is only about a 60% difference between the lower and the Centers with respect to the lower limits of normal for serum upper limits of normal. The reason for the relatively wider range laboratory to laboratory. Figure 1 shows the normal range of C3 levels for the 14 Centers involved in the NIH-sponsored multi-

814

Hebert et a!: Diagnostic significance of hypocomp/ementemia

S

50

40

S.. • •5

30

C-)

20

.

S

S

$

S

S

10

Fig. 2. Relationship between serum C3 and C4 levels in a typical patient with SLE and GN. The correlation coefficient of this relationship is 0.93 (P < 0.001). The complement levels in this patient were measured by nephelometry.

0 25

75 C3, mg/dI

of normal for C4 is that 20 to 40% of the normal population Evidence that classical pathway activation is the reason for low includes individuals with one or two C4 "null" genes [271. C3 and C4 levels in a given patient can be strengthened by Thus, an SLE patient who in remission has a relatively high C4 measuring plasma Clq levels. The plasma concentration of this level, can experience a considerable reduction in C4 as a result component should also be low if the low C3 and C4 levels are of SLE activity, and still have the C4 level remain within the the result of classical pathway activation. Discussed below are normal range. This too will decrease the diagnostic specificity the disorders in which IC formation is regarded as the primary of C4 in active SLE. Because of these difficulties in the cause of the hypocomplementemia. interpretation of serum C4 levels, we recommend measurement 1. SLE. In most studies, the majority of SLE patients with of C3 levels to serially monitor complement status. Also, by renal manifestations have hypocomplementemia [9, 10, 14—16, measuring just one of the complement components, the cost of 18, 21—23, 28—34]. However, extra-renal manifestations of SLE monitoring complement status is reduced by one-half. often occur without the development of hypocomplementemia Recent studies suggest that the measurement of the plasma [10, 18, 211. The hypocomplementemia (typically decreased levels of the products of complement activation, such as iC3b, serum C3 and C4 levels) of SLE is thought to be the result of C3a des Arg, C3d, C5a, or C5b-9 complex, may be useful in the clinical evaluation of disorders of complement activation [28— 34]. For example, in patients with low levels of plasma C3, the presence in plasma of increased levels of the stable breakdown product, C3a des Arg is evidence that the low C3 is the result, at least in part, of consumption of C3, not decreased synthesis

activation of the classical pathway by IC [3]. However, in some

plasma levels of the complement components. Although this approach to the evaluation of disorders of complement metabolism is promising, at the present time there is insufficient experience with these methods to determine their role in the clinical evaluation of patients with hypocomplementemia.

limits of normal). However, by eight weeks of therapy, the

patients decreased synthesis of complement may also play a role in causing the hypocomplementemia [351. Falling serum

complement levels can be predictive of impending SLE relapse [10, 14, 16, 18, 21, 22]. Rising serum complement levels during treatment of SLE can be indicative of adequate therapy [15, 16, of C3. In addition, increased plasma levels of the products of 18, 22, 23]. The decreased C3, C4 levels of severe SLE return complement activation may occur before plasma levels of the to normal with effective treatment, but only slowly, For examcomplement components become decreased. Thus, increased ple, four weeks after start of high-dose prednisone therapy and plasma levels of the products of complement activation may be cyclophosphamide therapy, serum C3 and C4 levels remain a more sensitive index of complement activation than decreased near pre-treatment levels (mean values about 60% of the lower mean C3 and C4 levels have returned to normal [81. Occasion-

ally other autoimmune disorders (such as, Sjorgren's syndrome) can be associated with IC formation and hypocomplementemia [361.

2. Idiopathic MPGN. MPGN Type 1 typically has low C3 and C4 levels which appear to be, at least in part, the result of Disorders in which IC formation is regarded as the primary classical pathway activation by IC [3]. In addition, MPGN Type cause of the hypocomplementemia 1 can have NFa (C3 Nef), an antibody that stabilizes the C3 Most patients who are hypocomplementemic as a result of convertase C3bBb, thereby increasing C3 consumption. C3 Nef IC-mediated disease have decreased levels of serum or plasma results mainly in activation of C3 in the fluid phase. Thus, C3 and C4, indicating activation of the classical pathway. membrane bound C3/C5 convertases do not form, and activaHowever, there are exceptions. These are discussed below. tion of the terminal pathway does not occur [3]. C3 Nef is found Causes of hypocomplementemia

Hebert et al: Diagnostic significance of hypocomplementemia

815

most commonly in MPGN Type 2, a disease that is not thought to be an IC-mediated disease. Patients with MPGN type 1 can also have NFt (nephritic factor of the terminal pathway). NFt binds to properdin (P) stabilizing the C3 convertase C3bBbP. This complex can activate the terminal pathway by providing a binding site for C5b [3]. MPGN Type III has low C3 levels and

izine (decreased C3 levels) [541, potassium iodide (decreased C3 and C4 levels) [55], and chiorpropamide (decreased CH5O) [56]. 11. Chemotherapy for malignant disorders. IC formation and decreased C3 and C4 levels have been observed during chemotherapy for Hodgkin's disease [57] and lymphoblastic leukemia

C3 and C4 levels are seen in all three types of cryoglobulinemia [37]. Evidently the immune aggregates that form in cryoglobulinemia are able to activate the classical pathway.

and decreased C3 and C4 levels can occur in these patients [60].

[58].

12. Thyroid disorders. Graves' disease, radiation treatment, sometimes low C4 levels, apparently as a result of classical pathway activation by IC. In addition, NFt has been identified or thyroiditis can result in the formation of thyroglobulin containing IC, IC disease and decreased C3 and C4 levels [59]. in patients with MPGN Type III [3]. 13. Jejunal ileal bypass. IC-mediated arthritis, hemolysis, 3. Cryoglobulinemia Types I, II, and III. Decreased serum The mechanism is thought to be gastrointestinal absorption of antigens resulting in immunization and IC formation. 14. B-cell lymphoproliferative disorders. IC, formed by anti4. Chronic infections causing GN/vasculitis. Bacterial inbodies directed at the idiotype of the monoclonal antibodies fections causing chronic osteomyelitis, endocarditis, infected produced by the B-cell disorder, cause increased activation of ventricular atrial shunts, or various types of visceral (lung, Cl with secondary consumption of Cl inactivator, C4, and C2. intra-abdominal or hepatic) abscess have been associated with C3 levels are not decreased apparently because of defective formation of circulating IC, decreased serum C3 and C4 levels, formation of C3 convertase [61]. Cryoglobulins are not present. and GN [38]. Viral infections have also been associated with IC formation and GN [38]. However, only hepatitis B virus is These patients can present with urticaria [61], panniculitis and known to be associated with a clinically severe form of GN/ hepatitis [62, 631, and recurrent infections [64]. vasculitis. Occasionally, these patients will have decreased C3 Causes of hypocomplementemia

and C4 levels [39]. Parasitic/spirochetal infections that are chronic can also be associated with IC disease and hypocomplementemia. Decreased C3 and C4 levels are observed [40—42].

Diseases/conditions in which factors other than IC formation are regarded as the cause of the hypocomplementemia

5. Post-infectious glomerulonephritis (GN). This category The conditions listed below are thought to cause hypocomdiffers from chronic infections causing GN/vasculitis in that the plementemia by mechanisms that do not involve the immune

former condition requires the presence of a chronic active system, at least not directly. In many instances it is the infection to cause GN. By contrast, in typical post-infectious alternative pathway that is activated. Thus, typically, C3 levels GN the infection occurs acutely and then subsides. However, are decreased but C4 levels are normal. There are, however, days to weeks later the GN appears. Post-infectious GN is seen numerous nonimmune causes of hypocomplementemia that following infections with certain strains of Group A beta cause low C3 and C4 that may or may not be the result of hemolytic streptococcus. However, streptococcus pneumoniae and meningococcus infection have also been associated with post-infectious GN [43—451. Although low serum C3 and C4

levels are seen in some patients early in the course of their

classical pathway activation. Each of these nonimmune conditions causing hypocomplementemia, and the mechanism of the hypocomplementemia is discussed below.

1. Atheromatous embolization. Spontaneous or catheter induced atheromatous embolization can produce a systemic disorder that can be easily confused with IC-mediated vasculi6. Rheumatoid vasculitis. Decreased serum C3 and C4 tis. In addition to hypocomplementemia, patients with atherolevels are seen during active disease, presumably the result of matous embolization may show lower extremity skin rash that high levels of circulating IC [46]. Renal involvement is unusual resembles vasculitis, progressive renal failure, pancreatitis, in this disorder. gastrointestinal bleeding, eosinophilia, and thrombocytopenia. 7. Idiopathic vasculitis. Some patients with vasculitis not Decreased serum complement levels are the result of compleclassifiable by current schema have IC-mediated disease and ment activation by atheromata, apparently mainly by alternate hypocomplementemia. Usually both serum C3 and C4 levels pathway activation [65—67]. The hypocomplementemia of athare decreased [47—49, and unpublished observations]. The eroembolism is usually short lived, occurring mainly during the vasculitis associated with Wegener's granulomatosis, which period of active embolization [65]. probably is not immune complex mediated, is not associated 2. Hemo/ytic-uremic syndrome/thrombolic thrombocy lopewith hypocomplementemia [50]. Temporal arteritis is not asso- nic purpura (HUS/TTP). This multisystem disorder also can ciated with hypocomplementemia, but may be IC mediated easily be confused with an IC-mediated vasculitis. Decreased disease, usually it is the C3 level that is decreased throughout the active phase of the disease [3, 38].

[50].

C3 and sometimes decreased C4 are seen in many patients with

8. Repeated injection offoreign proteins. Serum sickness is

HUS/TTP. The decreased C3 levels may be secondary to

the classic example of the diseases in this category [51].

complement activation by toxins or microorganisms associated with HUS, to plasminogen activation which can activate C3, or to activation of the alternative pathway by damaged endothehum [68—741. Whatever the process activating complement in HUS/TTP, it appears to sometimes include the classical path-

However, repeated vaccine injection can also cause decreased C3 and C4 levels, and IC mediated GN [52]. 9. Drug-induced SLE. Decreased C3 and C4 levels occur, but are unusual in drug induced SLE [53]. 10. Hypersensitivity to drugs. IC disease and hypoconiplementemia have been reported in patients treated with sulfasal-

way because Clq levels are also decreased in some patients with HUS/TTP [70, 72]. Thus, serum complement levels are of

816

Hebert Ct a!: Diagnostic significance of hypocomplementemia

little value in differentiating HUS/TTP from IC-mediated disease. 3. Severe sepsis. Systemic infections tend to increase serum complement levels [75—77]. However, in severe sepsis, especially with shock, serum C3 and C4 levels can be decreased markedly [77]. Shock without sepsis does not cause hypocomplementemia [77]. Severe sepsis with its fever, multisystem involvement, hypocomplementemia and thrombocytopenia can be confused with a severe IC-mediated vasculitis. The mechanism of the hypocomplementemia of severe sepsis appears to involve activation of the classical pathway because both C3 and C4 levels are low, Cl-Cl inhibitor complexes form, and C3a and C4a levels are increased [311. The mechanism by which the classical pathway is activated appears to be proteolytic inactivation of plasma Cl inhibitor [78]. 4. Severe malnutrition. Marked decreases in serum Clq and

abnormal C3 that forms a C3 convertase which is resistant to degradation by Factors H and I [97]. IC-mediated diseases are common in patients with complement pathway abnormalities that result in deficient C3 activation [98, 99]. This tendency to IC-mediated disease may be the result of increased precipitability of IC that form in the plasma of patients who are complement deficient, decreased capacity

of the plasma of patients with complement deficiencies to solubilize already deposited IC, or impaired function of the erythrocyte IC-clearing mechanism in the hypocomplement-

emic patient [98, 99]. Patients with inherited abnormalities that affect terminal complement pathway activation usually have an increased incidence of infection, particularly Neisserial infectiOns [98, 99]. Infections are also more common in patients with abnormalities of classical and alternative pathway activation [98, 99]. Thus, patients with inherited deficiencies of compleC3 have been described in severely malnourished children ment components, or their control proteins, can have decreased [79—811. Serum C4 levels are normal [80, 81]. Of note is that the C3 and C4 levels because of the effects of infection and/or IC relatively severe protein depletion that can accompany ne- formation. phrotic syndrome does not have any important effect on serum 8. Thermal burns. Marked reductions in serum C3 and C4 C3 or C4 levels, although Clq, C2, C8, C9 and Factor B can be have been noted in patients with severe burns involving 25% to low [82]. Moderately poor nutrition in adults does not have 90% of body surface area [100]. obvious effects on serum C3 and C4 levels (unpublished obser9. Acute myocardial infarction. Mild to moderate acute reductions in C3 and C4 have been reported in some but not all vations). 5. Hepatic failure. In fulminant hepatic failure, C3 and C4 patients with acute myocardial infarction [101, 1021. The mechlevels can decrease markedly but Clq levels are normal [83, 84]. anism of the hypocomplementemia is unknown. The hypocomplementemia in fulminant hepatic failure is prob10. Intravascular infusion of agents that can activate comably the result of decreased synthesis of C3 and C4. In addition, plement. Intravascular infusion of radiographic contrast agents in chronic active hepatitis, immune complexes can form and can cause mild complement activation and transient reductions complement activation is present [85]. This also can contribute in serum C3 levels. Alternative pathway complement activation to the hypocomplementemia of liver disease. has been demonstrated with conventional angiographic agents 6. Acute pancreatitis. Pancreatic proteases can convert as well as with the newer hypoosmolar, nonionic radiocontrast complement components into their active forms. This may be agents [103—108]. Complement activation may contribute to the mechanism of the decreased serum C3 and C4 levels seen in acute adverse reactions associated with infusion of radioconacute pancreatitis [86, 87]. Of note is that the decreased C3 trast agents [108]. Inulin is a complex polysaccharide that also levels of acute pancreatitis usually return to normal within days can activate the alternative complement pathway. Inulin levels of the onset of the acute pancreatitis [861. By contrast, in greater than 26 mgldl result in modest acute reductions in C3 patients with SLE, decreased serum C3 and C4 levels usually levels [109]. Heparin-protamine complexes have been shown to require several weeks of effective treatment before they return activate the classical pathway in vitro [1101 and in vivo, in to normal [8]. animals [lii]. However, very large doses of heparin and 7. Inherited abnormalities of complement pathway compo- protamine (beyond the therapeutic range in humans) are needed nents or of certain of the complement control proteins. Inher- to reduce serum complement levels. ited abnormalities of the complement system are rare, however, 11. Hemodialysis. Cuprophane dialysis membranes cause a an inherited deficiency of each of the complement components modest transient reduction in serum C3 levels but usually not Cl to C9, and for most of the complement control proteins have outside the range of normal [112, 113]. Cellulose acetate, been reported [881. Persons who are homozygous deficient for a polycellulose acetate, and polymethylmethacrylate membranes given complement component show virtually no CH5O activity. do not cause reductions in serum C3 or C4 levels with long-term ITeterozygous-deficient individuals generally show 1/2 normal use [113]. The mechanism of the hypocomplementemia is CH5O activity [881. Inherited deficiencies of complement con- alternative pathway activation by the dialysis membrane [114]. trol proteins can also result in hypocomplementemia by permit12. Cardiopulmonary bypass. Mild transient reductions in ting unregulated complement activation. For example, Cl in- C3 and C4 occur during cardiopulmonary bypass. The mechahibitor deficiency causes decreased levels of C2 and C4, but not nism is thought to be complement activation by the nylon mesh Cl and C3 [891. Factor H deficiency causes decreased levels of bubble oxygenators as well as by vigorous oxygenation of

C3, Factor B, and terminal pathway components [90—92]. whole blood [115—1 18]. 13. Hemolytic crisis in malaria. The decrease in serum C3 Factor I deficiency causes decreased levels of C3, Factor B, and the terminal pathway components, and a mild decrease in and C4 levels in this disorder seems to be multifactorial. C4 levels [93, 941. Deficiencies of C4 binding protein [951 and Alternative pathway activation by the malaria organisms, deproperdin [961 are not associated with hypocomplementemia. creased synthesis of C3 and C4 by the liver because of hepatic Inherited structural abnormality of complement factors can also

result in hypocomplementemia. An example is an inherited

infection, and IC formation during acute hemolysis can account for the hypocomplementemia seen in malaria [1191.

817

Hebert et a!: Diagnostic signflcance of hypocomplementemia Table 2. Common nonimmunologic diseases that can mimic an ICmediated vasduhitis because they cause multisystem disorders and hypocomplementemia Disease

1) Atheroembolism

2) Severe sepsis

Complement levels

C3,

. C3,

3) HUS/TTP

C3,

4) Acute pancreatitis

C3,

Table 3. Common non-immunologic conditions that can cause hypocomplementemia and, should they be present in a patient with inactive IC-mediated disease, could spuriously suggest activity of the IC-mediated disease

Comment

C4 The hypocomplementemia is transient, occurring only

during active embolization' The development of shock C4 increases likelihood of sepsis causing hypocomplementemia C4 Hypocomplementemia occurs in about 1/2 of cases C4 The hypocomplementemia clears in a few daysa

a By contrast, the hypocomplementemia of active SLE usually requires weeks to return to normal, with successful therapy.

Condition 1) The conditions listed in Table 2 2) Severe malnutritiona 3) Severe liver disease 4) Inherited C4 deficiency

Complement levels

Comment See Table 2

See Table 2 C3,

NI C4

Clq is also low

C3, .

C4

Clq levels are normal

NI C3,

C4

20—40% of normals have

1 or 2 C4 null genes. SLE patients have a higher incidence of this genetic defect

Abbreviations are: NI, normal; , decreased. a The protein depletion of uncomplicated nephrotic syndrome does not cause low C3 or C4 levels.

14. Systemic viral infection. A large kindred with recurrent I. Membranous nephropathy: Idiopathic[122], drug-induced Herpes simplex infection has been reported in which serum C4 [123—125] or cancer associated [126—129]. If hypocomplementlevels were normal but serum C3 levels and terminal pathway emia is observed in a patient with membranous nephropathy, components were decreased [1201. The mechanism of the SLE or another systemic IC-mediated disease should be sushypocomplementemia is thought to be alternative pathway pected [122]. 2. The GN/vasculitis of Henoch-SchOnlein purpura (HSP). activation by virally infected cells. Despite severe systemic vasculitis and evidence of complement 15. Porphyria. Exposure of patients with erythropoietic protoporphyria to sunlight results in decreased serum C3 levels, as deposition in the IgA deposits, hypocomplementemia is only occasionally observed in this disease [4, 130, unpublished a result of activation of the alternative pathway [1211. Table 2 shows the key nonimmunologic disorders that can observationsi. The reason why even severe systemic vasculitis closely mimic an IC-mediated systemic vasculitis because they can occur with IgA-IC without causing hypocomplementemia is cause a multi-system disorder with hypocomplementemia. Ta- not clear. However, it may relate to the decreased capacity of ble 3 summarizes the non-immunologic conditions that can IgA-IC to activate complement and/or to greater rates of tissue cause hypocomplementemia and, should they complicate the deposition of IgA-IC, compared to IC composed of other Ig course of a patient with inactive IC-mediated disease, could [131—132]. Thus, compared to other IC mediated diseases, HSP spuriously suggest the presence of active IC-mediated disease. may be able to cause a systemic vasculitis at lower rates of IC Thus, these tables summarize the key differential diagnostic formation, rates that do not cause hypocomplementemia. 3. IgA nephropathy (Berger's disease). This disorder, like problems in the evaluation of the hypocomplementemic patient. HSP, is primarily associated with IgA-IC deposition. Despite an increased frequency of C4 null genes [133], hypocomplementemia is absent [4, 38], perhaps for the reasons discussed in Immune complex-mediated renal diseases that are rarely or connection with HSP. never associated with hypocomplementemia 4. Goodpasture's disease. Complement-fixing antibodies are deposited on antigens in lung and kidney; however, hypocomHypocomplementemia is particularly common in IC-mediplementemia is not seen [4, 38]. ated diseases that cause GN. The reason for this association 5. Immunotactoid glomerulopathy. This recently described may be the magnitude of IC formation that is needed before there is sufficient deposition of IC in glomeruli to induce GN. disorder derives its name from immunoglobulin deposits that Nevertheless, in some forms of IC-mediated renal disease, form tactoids in glomeruli. C3 and Clq are present in glomerhypocomplementemia is conspicuously absent. The reasons ular deposits, but usually the patients are not hypocomplementwhy hypocomplementemia is absent in some forms of IC- emic [134, 135]. 6. CIq glomerulopathy. This recently described glomerular mediated GN is not clear. However, it may relate to the fact disorder is not associated with SLE, but contains abundant that, in some IC-mediated renal diseases (such as, idiopathic glomerular deposits of Cl q and immunoglobulins, but hypomembranous nephropathy), the IC are the result of in situ IC complementemia is absent [136]. formation and are present only in the kidney. Thus, the IC may have formed only outside the circulation and, the total mass of Conditions/disorders that can raise serum complement levels IC formed may be relatively small. Therefore relatively little Certain conditions can raise serum complement levels and, complement activation occurs. In any event, it is important to recognize that there are certain types of IC-mediated renal therefore, could offset the effect of IC diseases to lower serum diseases in which hypocomplementemia is not observed. For complement levels. Of particular relevance in this regard are pregnancy, systemic infections, and noninfectious chronic incompleteness sake this list of diseases is provided below.

818

Hebert et at: Diagnostic significance of hypocomplementemia

flammatory states such as rheumatoid arthritis. If these conditions coexist with an active IC-mediated disorder, it is possible that hypocomplementemia might not develop, even though the IC-mediated disease is active. The interaction between a condition that can raise complement levels and a condition that can

6. LANGLOIS PF, GAWRYL MS: Identical complement concentra-

tions in blood obtained from central venous catheters, arterial lines, and antecubital phlebotomy. J Lab Clin Med 110:495—497, 1987

7. Lewis E, LACHIN J, Lupus NEPHRITIS COLLABORATIVE STUDY

GROUP: Primary outcomes in the controlled trial of plasmaphere-

lower serum complement levels has been particularly well described in SLE patients who became pregnant [137, 1381. In normal pregnancy, serum complement levels often increase to above normal values. Thus, a "normal" complement level in a pregnant SLE patient could indicate activity of her SLE.

sis therapy (PPT) in severe lupus nephritis. (abstract) Am Soc Nephrol, Annual Meeting, Washington, DC, 8.

Dee 7—10, 1986

HEBERT L, NIELSEN E, POHL M, LACHIN J, HUN5ICKER L, LEWIS

E: Clinical course of severe lupus nephritis during the controlled clinical trial of plasmapheresis therapy (PPT). (abstract) Am Soc Nephrol, Annual Meeting, Washington, DC, Dec 7—10, 1986 9. RICKER D, HEBERT LA, SEDMAK D, HEBERT MM, ROHDE R,

Summary Hypocomplementemia is an important marker for the presence of IC-mediated disease and can be used to assess disease activity. However, in interpreting the clinical significance of hypocomplementemia, the following must be kept in mind: 1) There are numerous non-immunologic conditions that also can cause hypocomplementemia. Furthermore, some of these conditions can cause a multisystem disease that, along with the hypocomplementemia, can closely resemble an IC-mediated systemic vasculitis. Furthermore, these nonimmunologic conditions that lower serum complement levels can complicate the course of patients with inactive IC-mediated disease, spuriously indicating that the disease is active. The most relevant of these differential diagnostic problems are listed in Table 2. 2) There are a few conditions (for example, pregnancy) that can raise serum complement levels, thereby possibly obscuring the pres-

NEFF J, LEWIS EJ: Mechanisms responsible for the increased diagnostic sensitivity (S) of serum C3 levels measured by nephelometry (N) vs radial immunodiffusion (R) in active SLE. (abstract) Am Soc Nephrol 22nd Annual Meeting, Washington, DC, Dec 3—6, 1989 10. SCHUR PH, SANDSON J: Immunologic factors and clinical activity

in systemic lupus erythematosus. NEnglJMed278:533—538, 1968 11. GOTOFF SP, ISAACS EW, MUEHRCKE RC, SMITH RD: Serum

beta1 globulin in glomerulonephritis and systemic lupus erythematosus. Ann mt Med 71:327—333, 1969 12. PERRIN LH, LAMBERT PH, MIESCHER PA: Complement break-

down products in plasma from patients with systemic lupus erythematosus and patients with membranoproliferative or other glonierulonephritis. J Clin Invest 56:165—176, 1975 13. CAMERON JS, LEssoF MH, OGG CS, WILLIAMS BD, WILLIAMS

DG: Disease activity in the nephritis of systemic lupus erythematosus in relation to serum complement concentrations. Clin Exp Immunol 25:418—427, 1976 14. SINGSEN BH, BERNSTEIN BH, KING KK, HANSON V: Systemic

lupus erythematosus in childhood: Correlations between changes

ence of a disorder (such as, active SLE) that is lowering

complement levels. 3) There are some conditions that might be expected to lower serum complement levels, because of their effect on protein metabolism, but do not. Nephrotic syndrome, and moderately poor nutrition are examples. All of these factors should be considered when interpreting results of serum cornplement levels in a given patient. LEE A. HEBERT, FERNANDO G. Cosio, and JOHN C. NEFF Columbus, Ohio, USA

in disease activity and serum complement levels. J Pediatr 89:358— 365, 1976 15. APPaL AE, SABLAY LB, GOLDEN RA, BARLAND P, GRAYZEL Al,

BANK N: The effect of normalization of serum complement and anti-DNA antibody on the course of lupus nephritis. A two year prospective study. Am J Med 64:274—283, 1978 16. GARIN EH, DONNELLY WH, SHULMAN ST. FERNANDEZ R, FINTON C, WILLIAMS RL, RICHARD GA: The significance of serial measurements of serum complement C3 and C4 components and

DNA binding capacity in patients with lupus nephritis. Clin Nephrol 12:148—155, 1979 17. ABRAsS CK, NIE5 KM. LOUIE JS, BORDER WA, GLASSOCK RI:

Correlation and predictive accuracy of circulating immune complexes with disease activity in patients with systemic lupus erythematosus. Arthr Rheum 23:273—282, 1980

Acknowledgments

This review was supported in part by NIH grant HL-25404, NIH grant DK-39485, NIH grant RR00034, The Mabel Shurtz Research Fund, and a grant from the National Kidney Foundation of Ohio.

18. LLOYD W, SCHUR PH: Immune complexes, complement, and

anti-DNA in exacerbations of systemic lupus erythematosus (SLE). Medicine 60:208—217, 1981 19. MoRRow WJW, ISENBERG DA, TODD-POKROPEK A, PARRY HF,

SNAITH ML: Useful laboratory measurements in the management

Reprint requests to Lee A. Hebert, M.D., The Ohio State University

Medical Center, Room N210 Means Hall, Columbus, Ohio 43210, USA.

of systemic lupus erythematosus. Quart J Med New Series LI:125—138, 1982 20. VALENTIJN RM, VAN OVERHAGEN H, HAZEVOET HM, HERMAN5 J, CATS A, DAHA MR, VAN Es LA: The value of complement and

References 1.

ATKINSON JP, KAINE JL, HOLERS VM: Complement and the

rheumatic diseases, in Immunobiology of the Complement Sys-

tem, edited by GD Ross, Boston, Harcourt Brace Jovanovich, 1986, pp. 197—211

2. DALMASSO AP: Complement in the pathophysiology and diagnosis of human diseases. CRC Crit Rev Clin Lab Sci 24:123—183, 1987 3. WEST CD: The complement profile in clinical medicine. Complement Infiamm 6:49—64, 1989 4. MADAIO MP, HARRINGTON JT: The diagnosis of acute glomerulonephritis. N EnglJ Med 309:1299—1302, 1983 5. NEWELL S, GORMAN JC, BELL E, ATKINSON JP: Hemolytic and antigenic measurements of complement. J Lab Clin Med 100:437— 444, 1982

21.

immune complex determinations in monitoring disease activity in patients with systemic lupus erythematosus. ArthrRheum 28:904— 913, 1985 SWAAK AJG, GROENWOLD J, BRONSVELD W: Predictive value of complement profiles and anti-dsDNA in systemic lupus erythematosus. Ann Rheum Dis 45:359—366, 1986

22. CLOUGH J, P0HL M, HEBERT L, ROHDE R, THE Luus NEPHRITIS COLLABORATIVE STUDY GROUP: The relationship of serological

changes to exacerbations (EX) in SLE. (abstract) Am Soc Nephrol 1987

23. ESDAILE JM, LEVINTON C, FEDERGREEN W, HAYSLETT JP,

KASFIGARIAN M: The clinical and renal biopsy predictors of long-term outcome in lupus nephritis: A study of 87 patients and review of the literature. Quart J Med NS 72:779—833, 1989 24. WELCH TR, BEZSCHEL L, BERRY A, FORRISTAL J, WEST CD: The

819

Hebert et al: Diagnostic sign jflcance of hypocomplementemia

effect of null C4 alleles on complement function. C/in Immunol Immunopathol 34:316—325, 1985 25. UK0 G, CI-1RISTIANSEN FT, DAWKINS RL, MCCANN Vi: Refer-

ence ranges for serum C4 concentrations in subjects with and without C4 null alleles. J Gun Pathol 39:573—576, 1986 26. UK0 G, CHRISTIANSEN FT, DAWKINS RL: Serum C4 concentra-

tion in the monitoring of systemic lupus erythematosus: requirement for C4 allotyping. Rheumatol mt 6:111—114, 1986 27. ATKINSON JP: Complement deficiency. Predisposing factor to autoimmune syndromes. Am J Med 85:45—47, 1988

phamide on

the clinical course and levels of circulating immune

complexes. Ann mt Med 93:407—413, 1980 47. MCDUFFIE FC, SAMS WM, MALDONADO iE, ANDREINI PH, C0NN DL, SAMAYOA EA: Hypocomplementemia with cutaneous

vasculitis and arthritis. Possible immune complex syndrome. Mayo C/in Proc 48:340—348, 1973 48. MATHISON DA, ARROYAVE CM, BHAT KN, HUREWITZ DS,

MARNELL Di: Hypocomplementemia in chronic idiopathic urticaria. Ann mt Med 86:534—538, 1977 49. ZEISS CR, BURCH FX, MARDER Ri, FUREY NL, SCHMID FR,

GEWURZ H: A hypocomplementemic vasculitic urticarial syndrome. Report of four new cases and definition of the disease. Am

28. HoPKINS P, BELMONT HM, BUYON J, PHILIPS M, WEISSMANN G,

ABRAMSON SB: Increased levels of plasma anaphylatoxins in systemic lupus erythematosus predict flares of the disease and

J Med 68:867—875, l980

may elicit vascular injury in lupus cerebritis. Arth Rheum 3 1:632— 636, 1988

50. FAUCI AS, HAYNES BF, KATZ P: The spectrum of vasculitis.

29. SWAAK AJG, VAN ROOYEN A, VOGELAAR C, PILLAY M, HACK E:

ations. Ann mt Med 89:660—676, 1978 51. LAWLEY TJ, BIELORY L, GASCON P, YANCEY KB, YOUNG NS,

Complement (C3) metabolism in systemic lupus erythematosus in relation to the disease course. Rheumatol tnt 6:221—226, 1986 30. GAWRYL MS, CHUDWIN DS, LANGLOIS PF, LINT TF: The terminal complement complex, C5b-9, A marker of disease activity in patients with systemic lupus erythematosus. Arthr Rheum 31:188—199, 1988

31. HACK CE, NUIJENS JH, FELT-BERSMA RJF, SCHREUDER WO, EERENBERG-BELMER AiM, PAARDEKOOPER J, BRONSVELD W,

THIJS LG: Elevated plasma levels of the anaphylatoxins C3a and

C4a are associated with a fatal outcome in sepsis. Am J Med

Clinical, pathologic, immunologic, and therapeutic considerFRANK MM: A prospective clinical and immunologic analysis of patients with serum sickness. N Engi J Med 311:1407—1412, 1984 52. BOULTON-JONES iM, SISSONS JGP, NAISH PF, EVANS Di, PETERS

DK: Self-induced glomerulonephritis. Br Med J 2:387—390, 1974 53. UTSINGER PD, ZVAIFLER Ni, BLUESTEIN HG: Hypocomplement-

emia in Procainamide-associated systemic lupus erythematosus. Ann In! Med 84:293—296, 1976 54. MIHAS AA, GOLDENBERG Di, SLAUGHTER RL: Sulfasalazine

toxic reactions. Hepatitis, fever, and skin rash with hypocomple-

86:20—26, 1989

32. WHALEY K: Measurement of complement activation in clinical practice. Complement Inflamm 6:96-103, 1989

55.

33. NEGORO N, OKAMURA M, TAKEDA T, KODA 5, AMATSU K, INOUE T, CURD JG, KANAYAMA Y: The clinical significance of iC3b neoantigen expression in plasma from patients with systemic lupus erythematosus. (abstract) Arthr Rheum 32:1233, 1989

56.

34. CHARLESWORTH JA, PEAKE PW, GOLDING J, MACKIE JD, PUSSELL BA, TIMMERMANS V, WAKEFIELD D: Hypercatabolism of

syndrome and immune complex glomerulonephritis associated with chlorpropamide therapy. Am J Med 74:337—342, 1983 57. BRANDEIS WE, TAN C, WANG Y, GOOD RA, DAY NK: Circulat-

C3 and C4 in active and inactive systemic lupus erythematosus. Ann Rheum Dis 48: 153—159, 1989

35. SLIWINSKI AJ, ZVAIFLER NJ: Decreased synthesis of the third component of complement (C3) in hypocomplementemic systemic lupus erythematosus. C/in Exp Immunol 22:21—29, 1972 36. MouTsopouLos HM, BALOW JE, LAWLEY Ti, STAHL NI, ANTONOVYCH TT, Cl-fusED TM: Immune complex glomerulonephritis in Sicca syndrome. Am J Med 64:955—960, 1978 37. BROUET JC, CLAUVEL JP, DANON F, KELIN M, SELIGMANN M:

Biologic and clinical significance of cryoglobulins. A report of 86 cases. Am J Med 57:775—788, 1974 38. GLASSOCK RJ, ADLER SG, WARD HG, COHEN AH: Primary glomerular diseases, in The Kidney, edited by BM BRENNER, FC RECTOR, Philadelphia, WB Saunders Company, 1986, pp. 929— 1013

39. DUFFY i, LIDSKY MD, SHARP iT, DAVIS iS, PERSON DA, HOLLINGER FB, MIN KW: Polyarthritis, polyarteritis and hepatitis B. Medicine 55:19—37, 1976 40. PARBTANI A, CAMERON iS: Experimental nephritis associated with plasmodium infection in mice. Kidney mt 16:53—63, 1979 41. DUTRA M, MARTINELLI R, DE CARVALHO EM, RODRIGUE5 LE, BRITO E, ROCHA H: Renal involvement in visceral leishmaniasis. Am J Kid Dis 6:22—27, 1985 42. LOSITO A, BUCCIARELLI E, MASSI-BENEDETTI F, LATO M: Membranous

glomerulonephritis in congenital syphilis. C/in Nephro/

13:32—37, 1979

43. HYMAN LR, iENIS EH, HILL GS, ZIMMERMAN SW, BURKHOLDER PM: Alternate C3 pathway activation in pneumococcal glomerulonephritis. Am J Med 58:810—814, 1975 44. KAEHNY WD, OZAWA T, SCHWARZ MI, STANFORD RE, KOHLER PF, MCINTOSH RM: Acute nephritis and pulmonary alveolitis following pneumococcal pneumonia. Arch Intern Med 138:806— 808, 1978 RAINFORD DJ,

mentemia and immune complexes. JAMA 239:2590—2591, 1978 CURD iG, MILGROM H, STEVENSON DD, MATHISON DA, VAUGHAN JH: Potassium iodide sensitivity in four patients with hypocomplementemic vasculitis. Ann ml Med 91:853—857, 1979 APPEL GB, D'AGATI V, BERGMAN M, PIRANI CL: Nephrotic

58.

ing immune complexes, complement and complement component levels in childhood Hodgkin's disease. C/in Exp Immuno/ 39:55 1— 561, 1980 KALWINSKY DK, URMSON iR, STITZEL AE, SPITZER RE: Acti-

vation of the alternative pathway of complement in childhood acute lymphoblastic leukemia. J Lab C/in Med 88:745—756, 1976 59. JORDAN SC, BUCKINGI-IAM B, SAKAI R, OLSON D: Studies of

immune-complex glomerulonephritis mediated by human thryoglobulin. N Eng/ J Med 304:1212—1215, 1981 60. MOAKE iL, KAGELER WV, CIM0 PL, BLAKELY RW, ROSSEN RD,

HAESSE W: Intravascular hemolysis, thrombocytopenia, leukopenia, and circulating immune complexes after jejunal-ileal bypass surgery. Ann mt Med 86:576-578, 1977 61. GD-IA RS, QUINT! I, AUSTEN KF, CICARDI M, SHEFFER A, ROSEN FS: Acquired Cl-inhibitor deficiency associated with antiidiotypic antibody to monoclonal immunoglobulins. N Eng/ J Med 312:534—540, 1985

62. PASCUAL M, WIDMANN JJ, SCHIFFERLI JA: panniculitis

Recurrent febrile

and hepatitis in two patients with acquired comple-

ment deficiency and paraproteinemia. Am J Med 83:959—962, 1987 63. SCHIFFERLI JA, PASCUAL M, STEIGER G, SCHAPIRA M, RYSER JE,

ESTREICHER J, DASH A: Fast liver catabolism of Clq in patients

with paraproteinaemia and depletion of the classical pathway of complement. C/in Exp immunol 69:188—197, 1987 64. VARGA L, CLAS F, FUST G, PALOCZI K, SZEGEDI G, Loos M, HOLLAN SR: Patients with CLL and hypocomplementaemia have

an impaired serum bactericidal activity against the Salmonella minnesota re mutant. Complement 5:40—45, 1988

65. Coslo FG,

ZAGER RA, SI-IARMA HM: Atheroembolic

renal disease

causes hypocomplementaemia. Lancet 2:118-121, 1985 66. COLT HG, BEGG Ri, SAPORITO JJ, COOPER WM, SHAPIRO

AP: Cholesterol emboli after cardiac catheterization. Medicine 67:389— 400,

1988

WooDRow DF, SLOPER iC, DE WARDENER HE,

67. CAPPIELLO RA, ESPINOZA LR, ADELMAN H, AGUILAR I, VASEY

GRIFFITHS I: Post meningococcal acute glomerular nephritis. C/in Nephrol 9:249—253, 1978 46. ABEL T, ANDREWS BS, CUNNINGHAM PH, BRUNNER CM, DAVIS iS, HORWITZ DA: Rheumatoid vasculitis: Effect of Cyclophos-

68. CARRERAS L, ROMERO R, REQUESENS C, OLIVER AJ, CARRERA

45.

FB, GERMAIN BF: Cholesterol embolism: A pseudovasculitic syndrome. Scm Arthr Rheum 18:240—246, 1989 M, CLAVO M, ALSINA J: Familial hypocomplementemic hemolytic

Hebert et a!: Diagnostic sign,jlcance of hypocomplementemia

820

uremic syndrome with HLA-A3,B7 haplotype. JAMA 245:602— 604, 1981 69. GONZALO A, MAMPASO F, GALLEGO N, BELLAS C, SEGUI J,

ORTUNO J: Hemolytic uremic syndrome with hypocomplementemia and deposits of 1gM and C3 in the involved renal tissue. Clin

92. LEVY M, HALBWACHS-MECARELLI L, GUBLER MC, KonouT A, BENSENOUCI A, NIAUDET P. HAUPTMANN G, LESAVRE P: H deficiency

93. SOLAL-CELIGNY P, LAVIOLETTE M, HEBERT J, ADKINS P, SIROIs

M, BRUN G, LEHNER G, DELAGE J: C3b inactivator deficiency with immune complex manifestations. Clin Exp Immuno! 47:197—

Nephrol 16:193—199, 1981 70. MONNENS L, M0LENAAR J, LAMBERT PH, PROSEMANS W, VAN

MUNSTER P: The complement system in hemolytic-uremic syndrome in childhood. C/in Nephrol 13:168—171, 1980 71. EDITORIAL: Haemolytic-uraemic syndrome in childhood. Lancet 1:26—27, 1978

72. KIM Y, MILLER K, MICHAEL AF: Breakdown products of C3 and Factor B in hemolytic-uremic syndrome. J Lab C/in Med 89:845— 850, 1977 73. BARRE P, KAPLAN BS, DE CHADAREVIAN JP, DRUMMOND KH:

uremic syndrome with hypocomplementemia, serum C3NeF, and glomerular deposits of C3. Arch Patho! Lab Med Hemolytic

101:357—361, 1977 74. CRYER PE, KISSANE J: Hemolytic-uremic

syndrome in a young

woman. Am J Med 60:1005—1013, 1976

75. LEVIN J, POORE E, YOUNG NS, MARGOLIS S. ZAUBER NP, TOWNES AS, BELL WR: Gram-negative sepsis: Detection of endotoxemia with the limulus test. Ann mt Med 76:1—7, 1972 76. POCIDALO MA, GIRERT C, VERROUST P, ADAM M, MADEC Y, GAUDEBOUT C, MOREL-MAROGES L—€irculating immune com-

plexes and severe sepsis: Duration of infection as the main determinant. C/in Exp Immunol 47:513—519, 1982 77. LEON C, RODRIGO MJ, TOMASA A, GALLART M, LATLORE F, Ruts J, BRUGUES J: Complement activation in septic shock due to gram-negative and gram-positive bacteria. Crit Care Med 10:308— 310, 1982 78. NUIJENS JH, EERENBERG-BELMER AJM, HUIJBREGTS CCM, SCHREUDER W, FELT-BERSMA R, ABBINK J, THIJS L, HACK C: Proteolytic

inactivation of plasma Cl

inhibitor

in sepsis. J C/in

Invest 84:443—450, 1989 79.

CHANDRA RK: Serum complement and immunoconglutinin in

malnutrition. Arch Dis Child 50:225—228, 1975 80. SIRISINHA S, SUSKIND R, EDELMAN R, CHARUPATANA CH,

OLSON RE: Complement and C3-proactivator levels in children with protein-calorie malnutrition and effect of dieatry treatment. Lancet May 12:1016, 1973 81. HALLER L, ZUBLER RH, LAMBERT PH: Plasma levels of complement components and complement haemolytic activity in proteinenergy malnutrition C/in Exp Immuno! 34:248—252, 1978 82. STRIFE CF, JACKSON EC, FORRISTAL J, WEST CD: Effect of the

nephrotic syndrome on the concentration of serum complement

in two brothers with atypical dense intramembranous

deposit disease. Kidney mt 30:949—956, 1986

205, 94.

1982

DAY NK: Deficiencies in regulator proteins 2.

Factor I and

H.

Progr Allergy 39:335—338, 1986

95. Ti& RG, FLETCHER M, FORRISTAL J, WEST CD: C4 binding protein deficiency in a patient with atypical Behcet's disease. J Rheumatol 14:135—138, 1987 96. WYATT RJ: Deficiencies in regulator

proteins 3.

Properdin Progr

Allergy 39:339—343, 1986 97. MARDER HK, COLEMANTH, FORRISTAL J, BEISCHEL L, WEST

CD: An inherited defect in the C3 convertase, C3bBb, associated

with glomerulonephritis. Kidney Int 23:749—758, 1983 98. SCHIFFERLI JA, NG YC, PETERS DK: The role of complement and

its receptor in the elimination of immune complexes. N Engl J Med 315:488—495, 1986

99. HEBERT LA, Costo FG: The erythrocyte-immune complex-glomerulonephritis connection in man. Kidney mt 31:877—885, 1987 100. GELFAND JA, DONELAN M, HAWIGER A, BURKE iF: Alternative

complement pathway activation increases mortality in a model of burn injury in mice. J C/in Invest 70:1170—1 176, 1982 101. PINCRARD RN, OLSON MS. GICLAS PC, TERRY R, BOYER JT, O'ROURKE RA: Consumption of classical complement components by heat subcellular membranes in vitro and in patients after acute myocardial infarction. J C/in Invest 56:740—750, 1975 102. MOLLNES TE, TAMBS KE, MYRENG Y, ENGEBRETSEN LF: Acute

phase reactants and complement activation in patients with acute myocardial infarction. Comp/ement 5:33—39, 1988 103. SIEGLE RL, LIEBERMAN P: Measurement of histamine, complement components and immune complexes during patient reactions to iodinated contrast material. Invest Radiol 11:98—101, 1976 104. GONSETTE RE: Complement system activation by contrast media in neuroradiology. Neuroradiol 16:253—255, 1978 105. LASSER EC, LANG JH, LYON SG, HAMBLIN AE: Complement and contrast material reactors. JAI/ergy C/in Immuno/ 64:105—12, 1979 106. SIEGLE RL, LIEBERMAN P, JENNINGS BR, RICE MC: lodinated

contrast material: Studies relating to complement activation, atopy, cellular association, and antigenicity. Invest Radio! 15:S13— S17, 1980 107. LASSER EC, LANG JH, LYON SG, HAMBLIN AE: Changes in complement and coagulation factors in a patient suffering a severe

components. Am J Kid Dis 8:37—42, 1986 83. WYKE RJ, RAJKOVIC IA, WILLIAMS R: Impaired opsonization by

anaphylactoid reaction to injected contrast material: Some con-

serum from patients with chronic liver disease. C/in Exp Immunol

108. SCHNEIDERMAN H, HAMMERSCHMIDT DE, MCCALL AR, JACOB

51:91—98, 1988 84. MUNOZ LE, DE VILLIERS D, MARKHAM D, WHALEY K, THOMAS HC: Complement activation in chronic liver disease. C/in Exp Immuno! 47:548—554, 1982 85. PFUELLER SL, FIRKIN BG, Dn RosBo NK, RIGLAR A, MACKAY

IR: Association of increased immune complexes, platelet IgG and serum IgU in chronic active hepatitis. C/in Exp Immuno! 54:655— 660, 1983 86. GOLDSTEIN IM, CALA D, RADIN AR, KAPLAN HB, HORN J, RANSON J: Evidence of complement catabolism in acute pancreatitis. Am J Med Sd 275:257—263, 1978 87. JACOB HS, GOLDSTEIN IM, SHAPIRO I, CRADDOCK PR, HAMMERSCHMIDT DE, WEISSMANN G: Sudden blindness in acute pancreatitis. Arch Intern Med 141:134—136, 1981 88. SCHUR PH: Inherited complement component abnormalities. Ann Rev Med 37:333—346, 1986

89. FRANK MM: Complement in the pathophysiology of human disease. N EngI J Med 316:1525—1530, 1987 90. THOMPSON RA, WINTERBORN MH: Hypocomplementaemia due

to a genetic deficiency of B1H

globulin.

Clin Exp Immunol

46:110—119, 1981

91. WYATTRJ, JULIAN BA, WEINSTEIN A, ROTHFIELD NF, MCLEAN

RH: Partial H (B IH) deficiency and glomerulonephritis in two families. J C/in Immunol 2:110—117, 1982

siderations of pathogenesis. Invest Radio! 15:S6—S12, 1980

HS: Fatal complement-induced leukostasis after diatrizoate injection. JAMA 250:2340—2342, 1983

109. NISSENSON AR, RICE LE, POTTER EV, QUINTANILLA A, SHAUGHNESSY MA, LEVIN ML: Variations in serum complement following inulin infusion in man. Nephron 23:218—222, 1979 110. RENT R, ERTEL N, EISENSTEIN R, GEWURZ H: Complement

activation by interaction of polyanions and polycations. I. Heparin-protamine induced consumption of complement. J Immunol 114:120-124, 1975

Ill. FEL-IR J, ROHR H: In vivo complement activation by polyanionpolycation complexes: Evidence that C5a is generated intravascularly during heparin-protamine interaction. Clin Immunol immunopatho! 29:7—14, 1983 112. CRADDOCK PR, FEHR J, DALMASSO AP, BRIGHAM KL, JACOB HS:

Hemodialysis leukopenia. Pulmonary vascular leukostasis result-

ing from complement activation by dialyzer cellophane membranes, J Clin Invest 59:879—888, 1977 113. HAKIM RM, FEARON DT, LAZARUS JM: Biocompatibility of dialysis membranes: Effects of chronic complement activation. Kidney ml 26:194—200, 1984 114. CHEUNG AK, PARKER CJ, JANATOVA J: ment

Analysis of the comple-

C3 fragments associated with hemodialysis membranes.

Kidney mt 35:576—588, 1989 115. CHENOWETH DE, COOPER SW, HUGLI TE, STEWART RW,

Hebert et al: Diagnostic significance of hypocomplementemia BLACKSTONE EH, KIRKLIN JW: Complement activation during cardiopulmonary bypass. Evidence for generation of C3a and C5a anaphylatoxins. N Engi J Med 304:497—503, 1981 116. BORALESSA H, SHIFFERLI JA, ZAIMI F, WATTS, WHITWAM JG,

REES AJ: Perioperative changes in complement associated with cardiopulmonary bypass. Br J Anaesth 54:1047—1052, 1982 117. BoscoE Mi, YEWDALL VMA, THOMPSON MA, CAMERON iS: Complement activation during cardiopulmonary bypass: quantita-

tive study of effects of methylprednisolone and pulsatile flow. Br

MedJ 287:1747—1750, 1983 118. COLLETT B, ALI-IAQ A, ABDULLAH NB, KORJTSAO L, WARE R, DODD N, ALIMO B, PONTE J, VERGAMI D: Pathways to comple-

ment activation during cardiopulmonary bypass. Br Med J 289: 1251—1254, 1984 119. PHANUPHAK P, HANVANICH M, SAKULRAMRUNG R, MOOLLAOR

P, SITPRIJA V, PHANTHUMKOSOL D: Complement changes in falciparum malaria infection. Clin Exp Immunol 59:571—576, 1985 120. KAPADIA A, GUPTA S, GOOD RA, DAY NK: Familial herpes

simplex infection associated with activation of the complement system. Am J Med 67:122—126, 1979 121. LIM HW, POHL-FITZPATRICK MB, GIGLI I: Activation of the complement system in patients with porphyrias and irradiation in vivo. J Clin Invest 74:1961—1965, 1984 122. JENNETTE JC, I5KANDAR SS, DALLDORF FG: Pathologic differen-

tiation between lupus and nonlupus membranous glomerulopathy. Kidney mt 24:377—385, 1983

123. HOORNTJE SJ, WEaNING JJ, KALLENBERG CG, PRINS EJ, DONKER AiM: Serum-sickness-like syndrome with membranous glomerulopathy in patient on captopril. Lancet 2:1297, 1979 124. TEXTOR SC, GEPHARDT GN, BL&vo E, TAItzI R, FOUAD F, TUBBS R, MCMAHON J: Membranous glomerulopathy associated with captopril therapy. Am J Med 74:705—712, 1983 125. SENNESAEL J, VAN DEN HOUTE K, VERBEELEN D: Reversible membranous glomerulonephritis associated with ketoprofen. Clin Nephrol 26:213—215, 1986 126. COUSER WG, WAGONFELD JB, SPARGO BH, LEWIS EJ: Glomer-

ular deposition of tumor antigen in membranous nephropathy associated with colonic carcinoma. Am J Med 57:962—966, 1974 127. GAGLIANO RG, C0sTANzI JJ, BEATHARD GA, SARLES HE, BELL

JD: The nephrotic syndrome associated with neoplasia: An un-

821

usual paraneoplastic syndrome. Report of a case and review of the literature. AmJMed6O:1026—103l, 1976 128. WELSENBURGER DD: Membranous nephropathy. Its association with multicentric angiofollicular lymph node hyperpolasia. Arch Paihol Lab Med 103:591—594, 1979 129. WAICASHIN M, WAKASHIN Y, IESATO K, UEDA S, M0RI Y, TSUCHIDA H, SHIGEMATSU H, OKUDA K: Association of gastric

cancer and nephrotic syndrome. An immunologic study in three patients. Gastroenterology 78:749—756, 1980 130. MEADOW SR, GLASGOW EF, WHITE RHR, MONCRIEFF MW, CAMERON JS, OGG CS: Schdnlein-Henoch nephritis. Quart J Med XLI:242—258, 1972 131. WAXMAN FJ, HEBERT LA, Cosio FG, SMEAD W, VAN AMMAN M, TAGUIAM J, BIRMINGHAM J: Differential binding of immunoglobulin A and immunoglobulin G I immune complexed to primate erythrocytes in vivo. J Clin Invest 77:82—89, 1986 132. WALDO FB, COCHRAN AM: Mixed IgA-IgG aggregates as a model of immune complexes in IgA nephropathy. J Immunol 142:3841— 3846, 1989 133. MCLEAN RH, WYATT Ri, JULIAN BA: Complement phenotypes

in glomerulonephritis: Increased frequency of homozygous null C4 phenotypes in IgA nephropathy and Henoch-Schönlein purpura. Kidney mt 26:855—860, 1984 134. STURGILL BC, BOLTON WK, GRIFFITH KM: Congo red-negative amyloidosis-like glomerulopathy. Human Pathol 16:220—224, 1985 135. KORBET SM, SCHWARTZ MM, ROSENBERG BF, SIBLEY RK, LEWIS EJ: Immunotactoid glomerulopathy. Medicine 64:228—243, 1985

136. JENNETTE JC, HIPP CG: C lq nephropathy: A distinct pathologic entity usually causing nephrotic syndrome. Am J Kid Dis 6:103— 10, 1985 137. SCHENA FP, MANNO C, SELVAGGI L, LOVERRO G, BETTOCCHI S,

BoNoMo L: Behaviour of immune complexes and the complement system in normal pregnancy and pre-eclampsia. J Gun Lab immuno! 7:21—26, 1982 138. BUYON JP, CRONSTEIN BN, MoRRIs M, TANNER M, WEISSMANN

G: Serum complement values (C3 and C4) to differentiate between systemic lupus activity and pre-eclampsia. Am J Med 81:194-200, 1986